diff --git a/Scripts/Notebooks/NumbersNb.ipynb b/Scripts/Notebooks/NumbersNb.ipynb
index 2e46690481bf8f57187be8e03aa80ea5b7efefba..fa893f7e652f32965c09ea9962aa82385f0d884f 100644
--- a/Scripts/Notebooks/NumbersNb.ipynb
+++ b/Scripts/Notebooks/NumbersNb.ipynb
@@ -2,7 +2,7 @@
"cells": [
{
"cell_type": "code",
- "execution_count": 18,
+ "execution_count": null,
"metadata": {},
"outputs": [],
"source": [
@@ -20,7 +20,7 @@
},
{
"cell_type": "code",
- "execution_count": 19,
+ "execution_count": null,
"metadata": {},
"outputs": [],
"source": [
@@ -30,7 +30,7 @@
},
{
"cell_type": "code",
- "execution_count": 3,
+ "execution_count": null,
"metadata": {},
"outputs": [],
"source": [
@@ -40,7 +40,7 @@
},
{
"cell_type": "code",
- "execution_count": 4,
+ "execution_count": null,
"metadata": {},
"outputs": [],
"source": [
@@ -52,7 +52,7 @@
},
{
"cell_type": "code",
- "execution_count": 5,
+ "execution_count": null,
"metadata": {},
"outputs": [],
"source": [
@@ -61,7 +61,7 @@
},
{
"cell_type": "code",
- "execution_count": 6,
+ "execution_count": null,
"metadata": {},
"outputs": [],
"source": [
@@ -72,7 +72,7 @@
},
{
"cell_type": "code",
- "execution_count": 7,
+ "execution_count": null,
"metadata": {},
"outputs": [],
"source": [
@@ -81,7 +81,7 @@
},
{
"cell_type": "code",
- "execution_count": 8,
+ "execution_count": null,
"metadata": {},
"outputs": [],
"source": [
@@ -91,130 +91,25 @@
},
{
"cell_type": "code",
- "execution_count": 9,
+ "execution_count": null,
"metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "(60000, 28, 28)"
- ]
- },
- "execution_count": 9,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
+ "outputs": [],
"source": [
"x_train.shape"
]
},
{
"cell_type": "code",
- "execution_count": 10,
+ "execution_count": null,
"metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "array([[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
- " 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
- " 0, 0],\n",
- " [ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
- " 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
- " 0, 0],\n",
- " [ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
- " 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
- " 0, 0],\n",
- " [ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
- " 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
- " 0, 0],\n",
- " [ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
- " 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
- " 0, 0],\n",
- " [ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 3,\n",
- " 18, 18, 18, 126, 136, 175, 26, 166, 255, 247, 127, 0, 0,\n",
- " 0, 0],\n",
- " [ 0, 0, 0, 0, 0, 0, 0, 0, 30, 36, 94, 154, 170,\n",
- " 253, 253, 253, 253, 253, 225, 172, 253, 242, 195, 64, 0, 0,\n",
- " 0, 0],\n",
- " [ 0, 0, 0, 0, 0, 0, 0, 49, 238, 253, 253, 253, 253,\n",
- " 253, 253, 253, 253, 251, 93, 82, 82, 56, 39, 0, 0, 0,\n",
- " 0, 0],\n",
- " [ 0, 0, 0, 0, 0, 0, 0, 18, 219, 253, 253, 253, 253,\n",
- " 253, 198, 182, 247, 241, 0, 0, 0, 0, 0, 0, 0, 0,\n",
- " 0, 0],\n",
- " [ 0, 0, 0, 0, 0, 0, 0, 0, 80, 156, 107, 253, 253,\n",
- " 205, 11, 0, 43, 154, 0, 0, 0, 0, 0, 0, 0, 0,\n",
- " 0, 0],\n",
- " [ 0, 0, 0, 0, 0, 0, 0, 0, 0, 14, 1, 154, 253,\n",
- " 90, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
- " 0, 0],\n",
- " [ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 139, 253,\n",
- " 190, 2, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
- " 0, 0],\n",
- " [ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 11, 190,\n",
- " 253, 70, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
- " 0, 0],\n",
- " [ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 35,\n",
- " 241, 225, 160, 108, 1, 0, 0, 0, 0, 0, 0, 0, 0,\n",
- " 0, 0],\n",
- " [ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
- " 81, 240, 253, 253, 119, 25, 0, 0, 0, 0, 0, 0, 0,\n",
- " 0, 0],\n",
- " [ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
- " 0, 45, 186, 253, 253, 150, 27, 0, 0, 0, 0, 0, 0,\n",
- " 0, 0],\n",
- " [ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
- " 0, 0, 16, 93, 252, 253, 187, 0, 0, 0, 0, 0, 0,\n",
- " 0, 0],\n",
- " [ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
- " 0, 0, 0, 0, 249, 253, 249, 64, 0, 0, 0, 0, 0,\n",
- " 0, 0],\n",
- " [ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
- " 0, 46, 130, 183, 253, 253, 207, 2, 0, 0, 0, 0, 0,\n",
- " 0, 0],\n",
- " [ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 39,\n",
- " 148, 229, 253, 253, 253, 250, 182, 0, 0, 0, 0, 0, 0,\n",
- " 0, 0],\n",
- " [ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 24, 114, 221,\n",
- " 253, 253, 253, 253, 201, 78, 0, 0, 0, 0, 0, 0, 0,\n",
- " 0, 0],\n",
- " [ 0, 0, 0, 0, 0, 0, 0, 0, 23, 66, 213, 253, 253,\n",
- " 253, 253, 198, 81, 2, 0, 0, 0, 0, 0, 0, 0, 0,\n",
- " 0, 0],\n",
- " [ 0, 0, 0, 0, 0, 0, 18, 171, 219, 253, 253, 253, 253,\n",
- " 195, 80, 9, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
- " 0, 0],\n",
- " [ 0, 0, 0, 0, 55, 172, 226, 253, 253, 253, 253, 244, 133,\n",
- " 11, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
- " 0, 0],\n",
- " [ 0, 0, 0, 0, 136, 253, 253, 253, 212, 135, 132, 16, 0,\n",
- " 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
- " 0, 0],\n",
- " [ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
- " 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
- " 0, 0],\n",
- " [ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
- " 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
- " 0, 0],\n",
- " [ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
- " 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
- " 0, 0]], dtype=uint8)"
- ]
- },
- "execution_count": 10,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
+ "outputs": [],
"source": [
"x_train[0]"
]
},
{
"cell_type": "code",
- "execution_count": 11,
+ "execution_count": null,
"metadata": {},
"outputs": [],
"source": [
@@ -224,7 +119,7 @@
},
{
"cell_type": "code",
- "execution_count": 12,
+ "execution_count": null,
"metadata": {},
"outputs": [],
"source": [
@@ -234,15 +129,15 @@
},
{
"cell_type": "code",
- "execution_count": 13,
+ "execution_count": 4,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
- "(60000, 784)\n",
- "(10000, 784)\n"
+ "(60000, 28, 28, 1)\n",
+ "(10000, 28, 28, 1)\n"
]
}
],
@@ -253,22 +148,9 @@
},
{
"cell_type": "code",
- "execution_count": 14,
+ "execution_count": null,
"metadata": {},
- "outputs": [
- {
- "data": {
- "image/png": "iVBORw0KGgoAAAANSUhEUgAAAI4AAACOCAYAAADn/TAIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAHJUlEQVR4nO3dT2gUZxgG8OdRGxFUiEkMMZEoKEVBsCqhpR4KVk0Fycmih1Ik4sEUWsihMb15kB6kt4BEKlYJlkqj5iKSamsp1GoEaf1DjClIItagCAlCsNGvh50u+w1mM767mZ3dfX4QMu9MsvuBj998M7t5l845iLypOYUegBQnBUdMFBwxUXDERMEREwVHTHIKDslmkoMk75PsyNegJPlovY9Dci6AewC2AhgFcB3AHufcnfwNT5JqXg6/2wTgvnPubwAg+T2AFgDTBoek7jYWnyfOuZrwzlxOVfUARjLq0WCflJYHr9uZy4wTCcn9APbP9vNIvHIJzkMAyzPqhmCfxznXDaAb0KmqlORyqroOYDXJlSQrAOwG0JefYUnSmWcc59wUyc8AXAQwF8Bx59ztvI1MEs18OW56Mp2qitEN59ym8E7dORYTBUdMFBwxUXDERMEREwVHTBQcMVFwxETBERMFR0wUHDFRcMREwRETBUdMFBwxUXDEZNbfrF6qli5dmt5ub2/3jh04cMCrFy5c6NW9vb1evWvXLq9+9epVPoY4qzTjiImCIyYKjphojRNRdXW1V586dSq9XVlZ6R3bt2+fV8+Z4///7Onp8ep58/x/hhcvXpjHGRfNOGKi4IiJgiMmWuNMo7Gx0asvXLjg1deuXUtvt7S0eMcmJye9etu2bXkeXeFpxhETBUdMFBwx0RpnGmfOnPHqkZERr25tbU1vv3z5MutjzZ8/36vHxsa8uhhemwrTjCMmMwaH5HGSYyRvZexbQrKf5FDwvTLbY0jpiTLjnADQHNrXAeCSc241gEtBLWVkxjWOc+5XkitCu1sAfBBsfwfgFwBf5nFcsdu7d69Xr1+/3qsbGhq8Otu6hqRXh1+7OnnypFdPTU1FHmdSWNc4tc65R8H2PwBq8zQeKRI5X1U551y2Fm1qV1uarDPOY5J1ABB8H5vuB51z3c65Ta/rIyfFyzrj9AH4FMDXwffzeRtRTGpr/bProUOHvPro0aNeHb73ks3ixYu9eufOnV597NixyI+VVFEux08D+B3A2yRHSbYiFZitJIcAfBjUUkaiXFXtmebQljyPRYqI7hyLSdm+VrVhwwavrq/3P/gmfK8lm4qKCq8+d+5c1p+fmJiI/NhJpRlHTBQcMVFwxKRs1zgzGR4eznp88+bN6e0jR454x5qamrw6/H6b8Ht7ipFmHDFRcMSkbE9V4dPF8+fPvbqvz/+wv6GhIa/ObE0Svvyuq6vz6qdPn3r1s2fP3mywCaQZR0wUHDFRcMREn8kZ2Lhxo1cfPnzYq1etWuXV/f396e3Ozk7v2JUrV7z64UP/U7Wbm8Nv4U40fSan5I+CIyYKjpiU7X2csBs3bnj19u3bI/9uTU2NVy9btsyru7q67ANLKM04YqLgiImCIyZa4+RBW1ubVy9YsMCrz549G+dwYqEZR0wUHDFRcMREa5w8CP9pTbil/uPHj+McTiw044iJgiMmCo6YaI0zCy5fvlzoIcw6zThiEqU/znKSP5O8Q/I2yc+D/WpZW8aizDhTANqdc2sBvAugjeRaqGVtWYvSWOkRgEfB9gTJuwDqUYIta60WLVrk1aXQxmQmb7TGCfodvwPgD6hlbVmLfFVFciGAHwF84Zwbz2wCna1lrdrVlqZIMw7Jt5AKTY9zrjfYHallrdrVlqYZZxymppZvAdx1zn2TcajoW9bmIrMl7ZYtfh/Njo7Sv06Icqp6H8AnAP4ieTPY14lUYH4I2tc+APDx7AxRkijKVdVvADjNYbWsLVO6cywmeq3KaN26dentqqoq79jVq1fjHk7sNOOIiYIjJgqOmGiNY7RmzZr09uTkpHdsfHw87uHETjOOmCg4YqJTlVF1dXV6e2BgwDtWCp3TZ6IZR0wUHDFRcMREaxyjzJccSrGNyUw044iJgiMmCo6YqCW/zEQt+SV/FBwxUXDERMEREwVHTBQcMVFwxCTu16qeIPVXn9XBdhIldWyFGlfj63bGegMw/aTkQFKbECR1bEkbl05VYqLgiEmhgtNdoOeNIqljS9S4CrLGkeKnU5WYxBocks0kB0neJ1nQtlUkj5McI3krY18iejcXQ2/p2IJDci6ALgAfAVgLYE/QL7lQTgBoDu1LSu/m5PeWds7F8gXgPQAXM+qDAA7G9fzTjGkFgFsZ9SCAumC7DsBgIceXMa7zALYmaXxxnqrqAWT+ieNosC9JEte7Oam9pbU4noZL/bcu6CVnuLd05rFCjy/O4DwEsDyjbgj2JUmk3s1xyKW3dBziDM51AKtJriRZAWA3Ur2Sk+T/3s1AAXs3R+gtDRS6t3TMi7wdAO4BGAbwVYEXnKeR+nCTf5Fab7UCqELqamUIwE8AlhRobJuROg39CeBm8LUjKeNzzunOsdhocSwmCo6YKDhiouCIiYIjJgqOmCg4YqLgiMl/XSLB/xBVEsIAAAAASUVORK5CYII=\n",
- "text/plain": [
- "<Figure size 144x144 with 1 Axes>"
- ]
- },
- "metadata": {
- "needs_background": "light"
- },
- "output_type": "display_data"
- }
- ],
+ "outputs": [],
"source": [
"plt.figure(figsize = (2, 2))\n",
"plt.imshow(x_train[900].reshape(28, 28))\n",
@@ -278,7 +160,127 @@
},
{
"cell_type": "code",
- "execution_count": 15,
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "autoencoder.fit(x_train, x_train,\n",
+ " epochs = 50,\n",
+ " batch_size = 256,\n",
+ " shuffle = True,\n",
+ " validation_data = (x_test, x_test))"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "# encode and decode some digits\n",
+ "# note that we take them from the *test* set\n",
+ "encoded_imgs = encoder.predict(x_test)\n",
+ "decoded_imgs = decoder.predict(encoded_imgs)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "n = 10 # how many digits we will display\n",
+ "plt.figure(figsize=(20, 4))\n",
+ "for i in range(n):\n",
+ " # display original\n",
+ " ax = plt.subplot(2, n, i + 1)\n",
+ " plt.imshow(x_test[i].reshape(28, 28))\n",
+ " plt.gray()\n",
+ " ax.get_xaxis().set_visible(False)\n",
+ " ax.get_yaxis().set_visible(False)\n",
+ "\n",
+ " # display reconstruction\n",
+ " ax = plt.subplot(2, n, i + 1 + n)\n",
+ " plt.imshow(decoded_imgs[i].reshape(28, 28))\n",
+ " plt.gray()\n",
+ " ax.get_xaxis().set_visible(False)\n",
+ " ax.get_yaxis().set_visible(False)\n",
+ "plt.show()"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "Using TensorFlow backend.\n",
+ "/home/jon/PycharmProjects/TensorPlay/venv/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:517: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
+ " _np_qint8 = np.dtype([(\"qint8\", np.int8, 1)])\n",
+ "/home/jon/PycharmProjects/TensorPlay/venv/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:518: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
+ " _np_quint8 = np.dtype([(\"quint8\", np.uint8, 1)])\n",
+ "/home/jon/PycharmProjects/TensorPlay/venv/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:519: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
+ " _np_qint16 = np.dtype([(\"qint16\", np.int16, 1)])\n",
+ "/home/jon/PycharmProjects/TensorPlay/venv/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:520: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
+ " _np_quint16 = np.dtype([(\"quint16\", np.uint16, 1)])\n",
+ "/home/jon/PycharmProjects/TensorPlay/venv/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:521: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
+ " _np_qint32 = np.dtype([(\"qint32\", np.int32, 1)])\n",
+ "/home/jon/PycharmProjects/TensorPlay/venv/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:526: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
+ " np_resource = np.dtype([(\"resource\", np.ubyte, 1)])\n"
+ ]
+ }
+ ],
+ "source": [
+ "from keras.layers import Input, Dense, Conv2D, MaxPooling2D, UpSampling2D\n",
+ "from keras.models import Model\n",
+ "from keras import backend as K\n",
+ "\n",
+ "input_img = Input(shape=(28, 28, 1)) # adapt this if using `channels_first` image data format\n",
+ "\n",
+ "x = Conv2D(16, (3, 3), activation='relu', padding='same')(input_img)\n",
+ "x = MaxPooling2D((2, 2), padding='same')(x)\n",
+ "x = Conv2D(8, (3, 3), activation='relu', padding='same')(x)\n",
+ "x = MaxPooling2D((2, 2), padding='same')(x)\n",
+ "x = Conv2D(8, (3, 3), activation='relu', padding='same')(x)\n",
+ "encoded = MaxPooling2D((2, 2), padding='same')(x)\n",
+ "\n",
+ "# at this point the representation is (4, 4, 8) i.e. 128-dimensional\n",
+ "\n",
+ "x = Conv2D(8, (3, 3), activation='relu', padding='same')(encoded)\n",
+ "x = UpSampling2D((2, 2))(x)\n",
+ "x = Conv2D(8, (3, 3), activation='relu', padding='same')(x)\n",
+ "x = UpSampling2D((2, 2))(x)\n",
+ "x = Conv2D(16, (3, 3), activation='relu')(x)\n",
+ "x = UpSampling2D((2, 2))(x)\n",
+ "decoded = Conv2D(1, (3, 3), activation='sigmoid', padding='same')(x)\n",
+ "\n",
+ "autoencoder = Model(input_img, decoded)\n",
+ "autoencoder.compile(optimizer='adadelta', loss='binary_crossentropy')"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 2,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from keras.datasets import mnist\n",
+ "import numpy as np\n",
+ "\n",
+ "(x_train, _), (x_test, _) = mnist.load_data()\n",
+ "\n",
+ "x_train = x_train.astype('float32') / 255.\n",
+ "x_test = x_test.astype('float32') / 255.\n",
+ "x_train = np.reshape(x_train, (len(x_train), 28, 28, 1)) # adapt this if using `channels_first` image data format\n",
+ "x_test = np.reshape(x_test, (len(x_test), 28, 28, 1)) # adapt this if using `channels_first` image data format"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 3,
"metadata": {},
"outputs": [
{
@@ -287,157 +289,145 @@
"text": [
"Train on 60000 samples, validate on 10000 samples\n",
"Epoch 1/50\n",
- "60000/60000 [==============================] - 2s 35us/step - loss: 0.3591 - val_loss: 0.2722\n",
+ "60000/60000 [==============================] - 34s 570us/step - loss: 0.2069 - val_loss: 0.1690\n",
"Epoch 2/50\n",
- "60000/60000 [==============================] - 2s 33us/step - loss: 0.2660 - val_loss: 0.2562\n",
+ "60000/60000 [==============================] - 34s 571us/step - loss: 0.1578 - val_loss: 0.1431\n",
"Epoch 3/50\n",
- "60000/60000 [==============================] - 2s 30us/step - loss: 0.2469 - val_loss: 0.2352\n",
+ "60000/60000 [==============================] - 34s 571us/step - loss: 0.1420 - val_loss: 0.1345\n",
"Epoch 4/50\n",
- "60000/60000 [==============================] - 2s 30us/step - loss: 0.2280 - val_loss: 0.2179\n",
+ "60000/60000 [==============================] - 34s 564us/step - loss: 0.1336 - val_loss: 0.1278\n",
"Epoch 5/50\n",
- "60000/60000 [==============================] - 2s 30us/step - loss: 0.2119 - val_loss: 0.2030\n",
+ "60000/60000 [==============================] - 35s 582us/step - loss: 0.1280 - val_loss: 0.1259\n",
"Epoch 6/50\n",
- "60000/60000 [==============================] - 2s 30us/step - loss: 0.1987 - val_loss: 0.1917\n",
+ "60000/60000 [==============================] - 35s 578us/step - loss: 0.1236 - val_loss: 0.1228\n",
"Epoch 7/50\n",
- "60000/60000 [==============================] - 2s 31us/step - loss: 0.1887 - val_loss: 0.1832\n",
+ "60000/60000 [==============================] - 35s 584us/step - loss: 0.1207 - val_loss: 0.1168\n",
"Epoch 8/50\n",
- "60000/60000 [==============================] - 2s 30us/step - loss: 0.1812 - val_loss: 0.1765\n",
+ "60000/60000 [==============================] - 36s 598us/step - loss: 0.1183 - val_loss: 0.1145\n",
"Epoch 9/50\n",
- "60000/60000 [==============================] - 2s 31us/step - loss: 0.1750 - val_loss: 0.1709\n",
+ "60000/60000 [==============================] - 35s 585us/step - loss: 0.1165 - val_loss: 0.1138\n",
"Epoch 10/50\n",
- "60000/60000 [==============================] - 2s 30us/step - loss: 0.1698 - val_loss: 0.1660\n",
+ "60000/60000 [==============================] - 35s 584us/step - loss: 0.1149 - val_loss: 0.1123\n",
"Epoch 11/50\n",
- "60000/60000 [==============================] - 2s 30us/step - loss: 0.1651 - val_loss: 0.1615\n",
+ "60000/60000 [==============================] - 36s 593us/step - loss: 0.1133 - val_loss: 0.1151\n",
"Epoch 12/50\n",
- "60000/60000 [==============================] - 2s 30us/step - loss: 0.1609 - val_loss: 0.1575\n",
+ "60000/60000 [==============================] - 35s 591us/step - loss: 0.1123 - val_loss: 0.1130\n",
"Epoch 13/50\n",
- "60000/60000 [==============================] - 2s 30us/step - loss: 0.1571 - val_loss: 0.1538\n",
+ "60000/60000 [==============================] - 35s 585us/step - loss: 0.1110 - val_loss: 0.1106\n",
"Epoch 14/50\n",
- "60000/60000 [==============================] - 2s 31us/step - loss: 0.1536 - val_loss: 0.1506\n",
+ "60000/60000 [==============================] - 34s 573us/step - loss: 0.1101 - val_loss: 0.1077\n",
"Epoch 15/50\n",
- "60000/60000 [==============================] - 2s 31us/step - loss: 0.1504 - val_loss: 0.1474\n",
+ "60000/60000 [==============================] - 35s 578us/step - loss: 0.1091 - val_loss: 0.1072\n",
"Epoch 16/50\n",
- "60000/60000 [==============================] - 2s 30us/step - loss: 0.1474 - val_loss: 0.1445\n",
+ "60000/60000 [==============================] - 34s 569us/step - loss: 0.1083 - val_loss: 0.1065\n",
"Epoch 17/50\n",
- "60000/60000 [==============================] - 2s 31us/step - loss: 0.1446 - val_loss: 0.1418\n",
+ "60000/60000 [==============================] - 36s 592us/step - loss: 0.1072 - val_loss: 0.1058\n",
"Epoch 18/50\n",
- "60000/60000 [==============================] - 2s 33us/step - loss: 0.1420 - val_loss: 0.1395\n",
+ "60000/60000 [==============================] - 35s 577us/step - loss: 0.1064 - val_loss: 0.1034\n",
"Epoch 19/50\n",
- "60000/60000 [==============================] - 2s 31us/step - loss: 0.1395 - val_loss: 0.1369\n",
+ "60000/60000 [==============================] - 34s 570us/step - loss: 0.1059 - val_loss: 0.1080\n",
"Epoch 20/50\n",
- "60000/60000 [==============================] - 2s 32us/step - loss: 0.1372 - val_loss: 0.1347\n",
+ "60000/60000 [==============================] - 35s 577us/step - loss: 0.1050 - val_loss: 0.1041\n",
"Epoch 21/50\n",
- "60000/60000 [==============================] - 2s 32us/step - loss: 0.1350 - val_loss: 0.1325\n",
+ "60000/60000 [==============================] - 35s 579us/step - loss: 0.1046 - val_loss: 0.1060\n",
"Epoch 22/50\n",
- "60000/60000 [==============================] - 2s 30us/step - loss: 0.1330 - val_loss: 0.1306\n",
+ "60000/60000 [==============================] - 35s 585us/step - loss: 0.1041 - val_loss: 0.1027\n",
"Epoch 23/50\n",
- "60000/60000 [==============================] - 2s 30us/step - loss: 0.1310 - val_loss: 0.1285\n",
+ "60000/60000 [==============================] - 35s 575us/step - loss: 0.1035 - val_loss: 0.1024\n",
"Epoch 24/50\n",
- "60000/60000 [==============================] - 2s 32us/step - loss: 0.1291 - val_loss: 0.1267\n",
+ "60000/60000 [==============================] - 36s 606us/step - loss: 0.1029 - val_loss: 0.1014\n",
"Epoch 25/50\n",
- "60000/60000 [==============================] - 2s 32us/step - loss: 0.1273 - val_loss: 0.1250\n",
+ "60000/60000 [==============================] - 35s 585us/step - loss: 0.1028 - val_loss: 0.1019\n",
"Epoch 26/50\n",
- "60000/60000 [==============================] - 2s 32us/step - loss: 0.1256 - val_loss: 0.1233\n",
+ "60000/60000 [==============================] - 35s 584us/step - loss: 0.1026 - val_loss: 0.1022\n",
"Epoch 27/50\n",
- "60000/60000 [==============================] - 2s 32us/step - loss: 0.1239 - val_loss: 0.1217\n",
+ "60000/60000 [==============================] - 36s 595us/step - loss: 0.1023 - val_loss: 0.1032\n",
"Epoch 28/50\n",
- "60000/60000 [==============================] - 2s 32us/step - loss: 0.1224 - val_loss: 0.1201\n",
+ "60000/60000 [==============================] - 36s 595us/step - loss: 0.1016 - val_loss: 0.1001\n",
"Epoch 29/50\n",
- "60000/60000 [==============================] - 2s 33us/step - loss: 0.1209 - val_loss: 0.1187\n",
+ "60000/60000 [==============================] - 35s 583us/step - loss: 0.1015 - val_loss: 0.1011\n",
"Epoch 30/50\n",
- "60000/60000 [==============================] - 2s 33us/step - loss: 0.1195 - val_loss: 0.1173\n",
+ "60000/60000 [==============================] - 35s 588us/step - loss: 0.1012 - val_loss: 0.1001\n",
"Epoch 31/50\n",
- "60000/60000 [==============================] - 2s 31us/step - loss: 0.1182 - val_loss: 0.1160\n",
+ "60000/60000 [==============================] - 35s 586us/step - loss: 0.1009 - val_loss: 0.0981\n",
"Epoch 32/50\n",
- "60000/60000 [==============================] - 2s 32us/step - loss: 0.1170 - val_loss: 0.1148\n",
+ "60000/60000 [==============================] - 35s 583us/step - loss: 0.1005 - val_loss: 0.0986\n",
"Epoch 33/50\n",
- "60000/60000 [==============================] - 2s 33us/step - loss: 0.1158 - val_loss: 0.1137\n",
+ "60000/60000 [==============================] - 35s 581us/step - loss: 0.1002 - val_loss: 0.0975\n",
"Epoch 34/50\n",
- "60000/60000 [==============================] - 2s 32us/step - loss: 0.1147 - val_loss: 0.1126\n",
+ "60000/60000 [==============================] - 35s 587us/step - loss: 0.1004 - val_loss: 0.0992\n",
"Epoch 35/50\n",
- "60000/60000 [==============================] - 2s 32us/step - loss: 0.1137 - val_loss: 0.1116\n",
+ "60000/60000 [==============================] - 37s 615us/step - loss: 0.1001 - val_loss: 0.0986\n",
"Epoch 36/50\n",
- "60000/60000 [==============================] - 2s 31us/step - loss: 0.1127 - val_loss: 0.1107\n",
+ "60000/60000 [==============================] - 36s 600us/step - loss: 0.1000 - val_loss: 0.0977\n",
"Epoch 37/50\n",
- "60000/60000 [==============================] - 2s 31us/step - loss: 0.1118 - val_loss: 0.1098\n",
+ "60000/60000 [==============================] - 35s 591us/step - loss: 0.0998 - val_loss: 0.0967\n",
"Epoch 38/50\n",
- "60000/60000 [==============================] - 2s 30us/step - loss: 0.1109 - val_loss: 0.1090\n",
+ "60000/60000 [==============================] - 36s 597us/step - loss: 0.0995 - val_loss: 0.0990\n",
"Epoch 39/50\n",
- "60000/60000 [==============================] - 2s 30us/step - loss: 0.1101 - val_loss: 0.1082\n",
+ "60000/60000 [==============================] - 35s 583us/step - loss: 0.0993 - val_loss: 0.0982\n",
"Epoch 40/50\n",
- "60000/60000 [==============================] - 2s 31us/step - loss: 0.1094 - val_loss: 0.1074\n",
+ "60000/60000 [==============================] - 35s 580us/step - loss: 0.0988 - val_loss: 0.1012\n",
"Epoch 41/50\n",
- "60000/60000 [==============================] - 2s 33us/step - loss: 0.1087 - val_loss: 0.1068\n",
+ "60000/60000 [==============================] - 35s 586us/step - loss: 0.0990 - val_loss: 0.0962\n",
"Epoch 42/50\n",
- "60000/60000 [==============================] - 2s 34us/step - loss: 0.1080 - val_loss: 0.1061\n",
+ "60000/60000 [==============================] - 35s 590us/step - loss: 0.0987 - val_loss: 0.0982\n",
"Epoch 43/50\n",
- "60000/60000 [==============================] - 2s 32us/step - loss: 0.1074 - val_loss: 0.1055\n",
+ "60000/60000 [==============================] - 36s 594us/step - loss: 0.0985 - val_loss: 0.0978\n",
"Epoch 44/50\n",
- "60000/60000 [==============================] - 2s 31us/step - loss: 0.1068 - val_loss: 0.1049\n",
+ "60000/60000 [==============================] - 35s 578us/step - loss: 0.0984 - val_loss: 0.0986\n",
"Epoch 45/50\n",
- "60000/60000 [==============================] - 2s 31us/step - loss: 0.1062 - val_loss: 0.1044\n",
+ "60000/60000 [==============================] - 35s 577us/step - loss: 0.0983 - val_loss: 0.0977\n",
"Epoch 46/50\n",
- "60000/60000 [==============================] - 2s 31us/step - loss: 0.1057 - val_loss: 0.1039\n",
+ "60000/60000 [==============================] - 36s 598us/step - loss: 0.0983 - val_loss: 0.0987\n",
"Epoch 47/50\n",
- "60000/60000 [==============================] - 2s 32us/step - loss: 0.1052 - val_loss: 0.1034\n",
+ "60000/60000 [==============================] - 35s 578us/step - loss: 0.0981 - val_loss: 0.0952\n",
"Epoch 48/50\n",
- "60000/60000 [==============================] - 2s 32us/step - loss: 0.1047 - val_loss: 0.1029\n",
+ "60000/60000 [==============================] - 35s 581us/step - loss: 0.0981 - val_loss: 0.0998\n",
"Epoch 49/50\n",
- "60000/60000 [==============================] - 2s 33us/step - loss: 0.1043 - val_loss: 0.1025\n",
+ "60000/60000 [==============================] - 35s 580us/step - loss: 0.0980 - val_loss: 0.0977\n",
"Epoch 50/50\n",
- "60000/60000 [==============================] - 2s 31us/step - loss: 0.1039 - val_loss: 0.1021\n"
+ "60000/60000 [==============================] - 35s 575us/step - loss: 0.0975 - val_loss: 0.0970\n"
]
},
{
"data": {
"text/plain": [
- "<keras.callbacks.History at 0x7f3f4c0d9e80>"
+ "<keras.callbacks.History at 0x7fb054474a90>"
]
},
- "execution_count": 15,
+ "execution_count": 3,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"autoencoder.fit(x_train, x_train,\n",
- " epochs = 50,\n",
- " batch_size = 256,\n",
- " shuffle = True,\n",
- " validation_data = (x_test, x_test))"
+ " epochs=50,\n",
+ " batch_size=128,\n",
+ " shuffle=True,\n",
+ " validation_data=(x_test, x_test))"
]
},
{
"cell_type": "code",
- "execution_count": 16,
- "metadata": {},
- "outputs": [],
- "source": [
- "# encode and decode some digits\n",
- "# note that we take them from the *test* set\n",
- "encoded_imgs = encoder.predict(x_test)\n",
- "decoded_imgs = decoder.predict(encoded_imgs)"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 17,
+ "execution_count": 6,
"metadata": {},
"outputs": [
{
- "data": {
- "image/png": "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\n",
- "text/plain": [
- "<Figure size 1440x288 with 20 Axes>"
- ]
- },
- "metadata": {
- "needs_background": "light"
- },
- "output_type": "display_data"
+ "ename": "NameError",
+ "evalue": "name 'decoded_imgs' is not defined",
+ "output_type": "error",
+ "traceback": [
+ "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+ "\u001b[0;31mNameError\u001b[0m Traceback (most recent call last)",
+ "\u001b[0;32m<ipython-input-6-d88a015e8277>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m 12\u001b[0m \u001b[0;31m# display reconstruction\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 13\u001b[0m \u001b[0max\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msubplot\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m2\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mn\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mi\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0;36m1\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0mn\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 14\u001b[0;31m \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mimshow\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdecoded_imgs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mreshape\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m28\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m28\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 15\u001b[0m \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mgray\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 16\u001b[0m \u001b[0max\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget_xaxis\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mset_visible\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;32mFalse\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+ "\u001b[0;31mNameError\u001b[0m: name 'decoded_imgs' is not defined"
+ ]
}
],
"source": [
+ "import matplotlib.pyplot as plt\n",
"n = 10 # how many digits we will display\n",
"plt.figure(figsize=(20, 4))\n",
"for i in range(n):\n",
diff --git a/Scripts/Notebooks/SmokeyConvNb.ipynb b/Scripts/Notebooks/SmokeyConvNb.ipynb
new file mode 100644
index 0000000000000000000000000000000000000000..4e0f4567d4f198641de0f22da94fbe3e81c400a4
--- /dev/null
+++ b/Scripts/Notebooks/SmokeyConvNb.ipynb
@@ -0,0 +1,297 @@
+{
+ "cells": [
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "/home/jon/PycharmProjects/TensorPlay/venv/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:517: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
+ " _np_qint8 = np.dtype([(\"qint8\", np.int8, 1)])\n",
+ "/home/jon/PycharmProjects/TensorPlay/venv/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:518: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
+ " _np_quint8 = np.dtype([(\"quint8\", np.uint8, 1)])\n",
+ "/home/jon/PycharmProjects/TensorPlay/venv/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:519: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
+ " _np_qint16 = np.dtype([(\"qint16\", np.int16, 1)])\n",
+ "/home/jon/PycharmProjects/TensorPlay/venv/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:520: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
+ " _np_quint16 = np.dtype([(\"quint16\", np.uint16, 1)])\n",
+ "/home/jon/PycharmProjects/TensorPlay/venv/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:521: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
+ " _np_qint32 = np.dtype([(\"qint32\", np.int32, 1)])\n",
+ "/home/jon/PycharmProjects/TensorPlay/venv/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:526: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
+ " np_resource = np.dtype([(\"resource\", np.ubyte, 1)])\n"
+ ]
+ }
+ ],
+ "source": [
+ "import time\n",
+ "import os\n",
+ "import shutil\n",
+ "import sys\n",
+ "import math\n",
+ "import random\n",
+ "import tensorflow as tf\n",
+ "import numpy as np\n",
+ "import scipy.misc\n",
+ "import matplotlib.pyplot as plt\n",
+ "sys.path.append(\"../tools\")\n",
+ "import uniio"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 2,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "np.random.seed(13) # Set the random seed to the same number\n",
+ "tf.set_random_seed(13)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 3,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "base_path = '../data/' # Path to sim data, trained models and outputs are also saved here"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 4,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "densities = []\n",
+ "\n",
+ "for sim in range(1000, 2000):\n",
+ " if os.path.exists(\"%s/simSimple_%04d\" % (base_path, sim)):\n",
+ " for i in range(0, 100):\n",
+ " filename = \"%s/simSimple_%04d/density_%04d.uni\"\n",
+ " uni_path = filename % (base_path, sim, i) # 100 files per sim\n",
+ " header, content = uniio.readUni(uni_path) # returns [Z,Y,X,C] np array\n",
+ " h = header['dimX']\n",
+ " w = header['dimY']\n",
+ " arr = content[:, ::-1, :, :] # reverse order of Y axis\n",
+ " arr = np.reshape(arr, [w, h, 1]) # discard Z\n",
+ " densities.append(arr)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 5,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "load_num = len(densities)\n",
+ "if load_num < 200:\n",
+ " print(\"Error - use at least two full sims, generate data by running 'manta ./manta_genSimSimple.py' a few times...\")\n",
+ " exit(1)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 6,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "(1100, 64, 64, 1)"
+ ]
+ },
+ "execution_count": 6,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "densities = np.reshape(densities, (len(densities), 64, 64, 1))\n",
+ "densities.shape"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 7,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Read uni files, total data (1100, 64, 64, 1)\n",
+ "Split into 990 training and 110 validation samples\n"
+ ]
+ }
+ ],
+ "source": [
+ "print(\"Read uni files, total data \" + format(densities.shape))\n",
+ "vali_size = max(100, int(load_num * 0.1)) # at least 1 full sim...\n",
+ "vali_data = densities[load_num - vali_size:load_num, :]\n",
+ "densities = densities[0:load_num - vali_size, :]\n",
+ "print(\"Split into %d training and %d validation samples\" % (densities.shape[0], vali_data.shape[0]))\n",
+ "load_num = densities.shape[0]"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 8,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "Using TensorFlow backend.\n"
+ ]
+ }
+ ],
+ "source": [
+ "from keras.layers import Input, Dense, Conv2D, MaxPooling2D, UpSampling2D\n",
+ "from keras.models import Model, Sequential\n",
+ "from keras import backend as K"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 9,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "densities = np.reshape(densities, (len(densities), 64, 64, 1))\n",
+ "vali_data = np.reshape(vali_data, (len(vali_data), 64, 64, 1))"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 10,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "(990, 64, 64, 1)\n",
+ "(110, 64, 64, 1)\n"
+ ]
+ }
+ ],
+ "source": [
+ "print(densities.shape)\n",
+ "print(vali_data.shape)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 21,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "input_img = Input(shape = (64, 64, 1))"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 22,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "l0 = Conv2D(filters = 16, kernel_size = (2, 2), activation = \"relu\", padding = \"same\")(input_img)\n",
+ "l1 = MaxPooling2D((2, 2), padding = \"same\")(l0)\n",
+ "l2 = Conv2D(filters = 8, kernel_size = (2, 2), activation = \"relu\", padding = \"same\")(l1)\n",
+ "l3 = MaxPooling2D((2, 2), padding = \"same\")(l2)\n",
+ "l4 = Conv2D(filters = 8, kernel_size = (2, 2), activation = \"relu\", padding=\"same\")(l3)\n",
+ "encoded = MaxPooling2D((2, 2), padding = \"same\")(l4)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 23,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "l5 = Conv2D(filters = 8, kernel_size = (2, 2), activation = \"relu\", padding = \"same\")(encoded)\n",
+ "l6 = UpSampling2D((2, 2))(l5)\n",
+ "l7 = Conv2D(filters = 8, kernel_size = (2, 2), activation = \"relu\", padding = \"same\")(l6)\n",
+ "l8 = UpSampling2D((2, 2))(l7)\n",
+ "l9 = Conv2D(filters = 16, kernel_size = (2, 2), activation = \"relu\")(l8)\n",
+ "decoded = Conv2D(filters = 1, kernel_size = (2, 2), activation = \"sigmoid\", padding = \"same\")(l9)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 14,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "autoencoder = Model(input_img, decoded)\n",
+ "autoencoder.compile(optimizer = \"adadelta\", loss = \"binary_crossentropy\")"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": []
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 15,
+ "metadata": {},
+ "outputs": [
+ {
+ "ename": "ValueError",
+ "evalue": "Error when checking target: expected conv2d_7 to have shape (31, 31, 1) but got array with shape (64, 64, 1)",
+ "output_type": "error",
+ "traceback": [
+ "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+ "\u001b[0;31mValueError\u001b[0m Traceback (most recent call last)",
+ "\u001b[0;32m<ipython-input-15-8ab6529663e8>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m 3\u001b[0m \u001b[0mbatch_size\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m64\u001b[0m \u001b[0;34m*\u001b[0m \u001b[0;36m64\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 4\u001b[0m \u001b[0mshuffle\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mTrue\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 5\u001b[0;31m validation_data = (vali_data, vali_data))\n\u001b[0m",
+ "\u001b[0;32m~/PycharmProjects/TensorPlay/venv/lib/python3.6/site-packages/keras/engine/training.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, x, y, batch_size, epochs, verbose, callbacks, validation_split, validation_data, shuffle, class_weight, sample_weight, initial_epoch, steps_per_epoch, validation_steps, **kwargs)\u001b[0m\n\u001b[1;32m 1628\u001b[0m \u001b[0msample_weight\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0msample_weight\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1629\u001b[0m \u001b[0mclass_weight\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mclass_weight\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1630\u001b[0;31m batch_size=batch_size)\n\u001b[0m\u001b[1;32m 1631\u001b[0m \u001b[0;31m# Prepare validation data.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1632\u001b[0m \u001b[0mdo_validation\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mFalse\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+ "\u001b[0;32m~/PycharmProjects/TensorPlay/venv/lib/python3.6/site-packages/keras/engine/training.py\u001b[0m in \u001b[0;36m_standardize_user_data\u001b[0;34m(self, x, y, sample_weight, class_weight, check_array_lengths, batch_size)\u001b[0m\n\u001b[1;32m 1478\u001b[0m \u001b[0moutput_shapes\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1479\u001b[0m \u001b[0mcheck_batch_axis\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mFalse\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1480\u001b[0;31m exception_prefix='target')\n\u001b[0m\u001b[1;32m 1481\u001b[0m sample_weights = _standardize_sample_weights(sample_weight,\n\u001b[1;32m 1482\u001b[0m self._feed_output_names)\n",
+ "\u001b[0;32m~/PycharmProjects/TensorPlay/venv/lib/python3.6/site-packages/keras/engine/training.py\u001b[0m in \u001b[0;36m_standardize_input_data\u001b[0;34m(data, names, shapes, check_batch_axis, exception_prefix)\u001b[0m\n\u001b[1;32m 121\u001b[0m \u001b[0;34m': expected '\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0mnames\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0;34m' to have shape '\u001b[0m \u001b[0;34m+\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 122\u001b[0m \u001b[0mstr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mshape\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0;34m' but got array with shape '\u001b[0m \u001b[0;34m+\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 123\u001b[0;31m str(data_shape))\n\u001b[0m\u001b[1;32m 124\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mdata\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 125\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
+ "\u001b[0;31mValueError\u001b[0m: Error when checking target: expected conv2d_7 to have shape (31, 31, 1) but got array with shape (64, 64, 1)"
+ ]
+ }
+ ],
+ "source": [
+ "autoencoder.fit(densities, densities,\n",
+ " epochs = 500,\n",
+ " batch_size = 64 * 64,\n",
+ " shuffle = True,\n",
+ " validation_data = (vali_data, vali_data))"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": []
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "Python 3",
+ "language": "python",
+ "name": "python3"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.6.8"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}
diff --git a/Scripts/Notebooks/SmokeyDeepNb.ipynb b/Scripts/Notebooks/SmokeyDeepNb.ipynb
index 7c69596421bec847af7bbb787b829ee0c3a3bfd5..524ab842cb769ef6b31e4525cc2ff94eb9cd3659 100644
--- a/Scripts/Notebooks/SmokeyDeepNb.ipynb
+++ b/Scripts/Notebooks/SmokeyDeepNb.ipynb
@@ -2,9 +2,28 @@
"cells": [
{
"cell_type": "code",
- "execution_count": 2,
+ "execution_count": 1,
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "/home/jon/PycharmProjects/TensorPlay/venv/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:517: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
+ " _np_qint8 = np.dtype([(\"qint8\", np.int8, 1)])\n",
+ "/home/jon/PycharmProjects/TensorPlay/venv/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:518: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
+ " _np_quint8 = np.dtype([(\"quint8\", np.uint8, 1)])\n",
+ "/home/jon/PycharmProjects/TensorPlay/venv/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:519: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
+ " _np_qint16 = np.dtype([(\"qint16\", np.int16, 1)])\n",
+ "/home/jon/PycharmProjects/TensorPlay/venv/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:520: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
+ " _np_quint16 = np.dtype([(\"quint16\", np.uint16, 1)])\n",
+ "/home/jon/PycharmProjects/TensorPlay/venv/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:521: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
+ " _np_qint32 = np.dtype([(\"qint32\", np.int32, 1)])\n",
+ "/home/jon/PycharmProjects/TensorPlay/venv/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:526: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
+ " np_resource = np.dtype([(\"resource\", np.ubyte, 1)])\n"
+ ]
+ }
+ ],
"source": [
"import time\n",
"import os\n",
@@ -22,7 +41,7 @@
},
{
"cell_type": "code",
- "execution_count": 3,
+ "execution_count": 2,
"metadata": {},
"outputs": [],
"source": [
@@ -32,7 +51,7 @@
},
{
"cell_type": "code",
- "execution_count": 4,
+ "execution_count": 3,
"metadata": {},
"outputs": [],
"source": [
@@ -41,7 +60,7 @@
},
{
"cell_type": "code",
- "execution_count": 5,
+ "execution_count": 4,
"metadata": {},
"outputs": [],
"source": [
@@ -62,7 +81,7 @@
},
{
"cell_type": "code",
- "execution_count": 6,
+ "execution_count": 5,
"metadata": {},
"outputs": [],
"source": [
@@ -74,7 +93,7 @@
},
{
"cell_type": "code",
- "execution_count": 7,
+ "execution_count": 6,
"metadata": {},
"outputs": [],
"source": [
@@ -83,7 +102,7 @@
},
{
"cell_type": "code",
- "execution_count": 8,
+ "execution_count": 7,
"metadata": {},
"outputs": [
{
@@ -92,7 +111,7 @@
"(1100, 64, 64, 1)"
]
},
- "execution_count": 8,
+ "execution_count": 7,
"metadata": {},
"output_type": "execute_result"
}
@@ -103,7 +122,7 @@
},
{
"cell_type": "code",
- "execution_count": 9,
+ "execution_count": 8,
"metadata": {},
"outputs": [
{
@@ -126,9 +145,17 @@
},
{
"cell_type": "code",
- "execution_count": 19,
+ "execution_count": 9,
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "Using TensorFlow backend.\n"
+ ]
+ }
+ ],
"source": [
"from keras.layers import Input, Dense\n",
"from keras.models import Model\n",
@@ -137,7 +164,7 @@
},
{
"cell_type": "code",
- "execution_count": 20,
+ "execution_count": 10,
"metadata": {},
"outputs": [],
"source": [
@@ -147,26 +174,26 @@
},
{
"cell_type": "code",
- "execution_count": 21,
+ "execution_count": 11,
"metadata": {},
"outputs": [],
"source": [
"input_img = Input(shape = (in_size, ))\n",
- "encoded = Dense(256, activation = \"relu\", activity_regularizer=regularizers.l1(10e-5))(input_img)\n",
- "encoded = Dense(128, activation = \"relu\", activity_regularizer=regularizers.l1(10e-5))(encoded)\n",
- "encoded = Dense(64, activation = \"relu\", activity_regularizer=regularizers.l1(10e-5))(encoded)\n",
- "encoded = Dense(32, activation = \"relu\", activity_regularizer=regularizers.l1(10e-5))(encoded)\n",
+ "encoded = Dense(256, activation = \"relu\")(input_img)\n",
+ "encoded = Dense(128, activation = \"relu\")(encoded)\n",
+ "encoded = Dense(64, activation = \"relu\")(encoded)\n",
+ "encoded = Dense(32, activation = \"relu\")(encoded)\n",
"\n",
- "decoded = Dense(64, activation = \"relu\", activity_regularizer=regularizers.l1(10e-5))(encoded)\n",
- "decoded = Dense(128, activation = \"relu\", activity_regularizer=regularizers.l1(10e-5))(decoded)\n",
- "decoded = Dense(256, activation = \"relu\", activity_regularizer=regularizers.l1(10e-5))(decoded)\n",
+ "decoded = Dense(64, activation = \"relu\")(encoded)\n",
+ "decoded = Dense(128, activation = \"relu\")(decoded)\n",
+ "decoded = Dense(256, activation = \"relu\")(decoded)\n",
"decoded = Dense(in_size, activation = \"sigmoid\")(decoded)\n",
"autoencoder = Model(input_img, decoded)"
]
},
{
"cell_type": "code",
- "execution_count": 22,
+ "execution_count": 12,
"metadata": {},
"outputs": [],
"source": [
@@ -175,7 +202,7 @@
},
{
"cell_type": "code",
- "execution_count": 23,
+ "execution_count": 13,
"metadata": {},
"outputs": [],
"source": [
@@ -185,7 +212,7 @@
},
{
"cell_type": "code",
- "execution_count": 24,
+ "execution_count": 14,
"metadata": {},
"outputs": [
{
@@ -204,7 +231,7 @@
},
{
"cell_type": "code",
- "execution_count": 25,
+ "execution_count": 15,
"metadata": {},
"outputs": [
{
@@ -229,16 +256,531 @@
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 16,
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Train on 990 samples, validate on 110 samples\n",
+ "Epoch 1/200\n",
+ "990/990 [==============================] - 2s 2ms/step - loss: 0.6916 - val_loss: 0.6900\n",
+ "Epoch 2/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.6880 - val_loss: 0.6856\n",
+ "Epoch 3/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.6771 - val_loss: 0.6191\n",
+ "Epoch 4/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.1619 - val_loss: 0.1197\n",
+ "Epoch 5/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0824 - val_loss: 0.1151\n",
+ "Epoch 6/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0807 - val_loss: 0.1108\n",
+ "Epoch 7/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0799 - val_loss: 0.1130\n",
+ "Epoch 8/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0794 - val_loss: 0.1093\n",
+ "Epoch 9/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0790 - val_loss: 0.1103\n",
+ "Epoch 10/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0788 - val_loss: 0.1118\n",
+ "Epoch 11/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0787 - val_loss: 0.1113\n",
+ "Epoch 12/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0786 - val_loss: 0.1089\n",
+ "Epoch 13/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0786 - val_loss: 0.1089\n",
+ "Epoch 14/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0785 - val_loss: 0.1102\n",
+ "Epoch 15/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0784 - val_loss: 0.1102\n",
+ "Epoch 16/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0785 - val_loss: 0.1125\n",
+ "Epoch 17/200\n",
+ "990/990 [==============================] - 2s 2ms/step - loss: 0.0783 - val_loss: 0.1104\n",
+ "Epoch 18/200\n",
+ "990/990 [==============================] - 1s 2ms/step - loss: 0.0784 - val_loss: 0.1111\n",
+ "Epoch 19/200\n",
+ "990/990 [==============================] - 2s 2ms/step - loss: 0.0784 - val_loss: 0.1077\n",
+ "Epoch 20/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0783 - val_loss: 0.1090\n",
+ "Epoch 21/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0783 - val_loss: 0.1092\n",
+ "Epoch 22/200\n",
+ "990/990 [==============================] - 2s 2ms/step - loss: 0.0783 - val_loss: 0.1083\n",
+ "Epoch 23/200\n",
+ "990/990 [==============================] - 2s 2ms/step - loss: 0.0783 - val_loss: 0.1099\n",
+ "Epoch 24/200\n",
+ "990/990 [==============================] - 1s 2ms/step - loss: 0.0783 - val_loss: 0.1094\n",
+ "Epoch 25/200\n",
+ "990/990 [==============================] - 1s 2ms/step - loss: 0.0782 - val_loss: 0.1095\n",
+ "Epoch 26/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0783 - val_loss: 0.1091\n",
+ "Epoch 27/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0782 - val_loss: 0.1084\n",
+ "Epoch 28/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0782 - val_loss: 0.1082\n",
+ "Epoch 29/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0783 - val_loss: 0.1076\n",
+ "Epoch 30/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0782 - val_loss: 0.1089\n",
+ "Epoch 31/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0782 - val_loss: 0.1079\n",
+ "Epoch 32/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0782 - val_loss: 0.1093\n",
+ "Epoch 33/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0782 - val_loss: 0.1100\n",
+ "Epoch 34/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0782 - val_loss: 0.1116\n",
+ "Epoch 35/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0782 - val_loss: 0.1108\n",
+ "Epoch 36/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0782 - val_loss: 0.1079\n",
+ "Epoch 37/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0781 - val_loss: 0.1088\n",
+ "Epoch 38/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0782 - val_loss: 0.1089\n",
+ "Epoch 39/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0782 - val_loss: 0.1099\n",
+ "Epoch 40/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0782 - val_loss: 0.1089\n",
+ "Epoch 41/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0782 - val_loss: 0.1084\n",
+ "Epoch 42/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0782 - val_loss: 0.1097\n",
+ "Epoch 43/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0782 - val_loss: 0.1082\n",
+ "Epoch 44/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0782 - val_loss: 0.1083\n",
+ "Epoch 45/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0781 - val_loss: 0.1079\n",
+ "Epoch 46/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0781 - val_loss: 0.1095\n",
+ "Epoch 47/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0781 - val_loss: 0.1090\n",
+ "Epoch 48/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0781 - val_loss: 0.1091\n",
+ "Epoch 49/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0782 - val_loss: 0.1096\n",
+ "Epoch 50/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0782 - val_loss: 0.1081\n",
+ "Epoch 51/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0782 - val_loss: 0.1096\n",
+ "Epoch 52/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0781 - val_loss: 0.1074\n",
+ "Epoch 53/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0781 - val_loss: 0.1075\n",
+ "Epoch 54/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0781 - val_loss: 0.1086\n",
+ "Epoch 55/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0781 - val_loss: 0.1094\n",
+ "Epoch 56/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0781 - val_loss: 0.1109\n",
+ "Epoch 57/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0781 - val_loss: 0.1080\n",
+ "Epoch 58/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0781 - val_loss: 0.1084\n",
+ "Epoch 59/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0781 - val_loss: 0.1085\n",
+ "Epoch 60/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0781 - val_loss: 0.1098\n",
+ "Epoch 61/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0781 - val_loss: 0.1087\n",
+ "Epoch 62/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0781 - val_loss: 0.1076\n",
+ "Epoch 63/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0781 - val_loss: 0.1095\n",
+ "Epoch 64/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0781 - val_loss: 0.1082\n",
+ "Epoch 65/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0781 - val_loss: 0.1095\n",
+ "Epoch 66/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0781 - val_loss: 0.1076\n",
+ "Epoch 67/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0781 - val_loss: 0.1089\n",
+ "Epoch 68/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0781 - val_loss: 0.1090\n",
+ "Epoch 69/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0781 - val_loss: 0.1105\n",
+ "Epoch 70/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0781 - val_loss: 0.1107\n",
+ "Epoch 71/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0781 - val_loss: 0.1090\n",
+ "Epoch 72/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0781 - val_loss: 0.1086\n",
+ "Epoch 73/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0781 - val_loss: 0.1092\n",
+ "Epoch 74/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0781 - val_loss: 0.1089\n",
+ "Epoch 75/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0781 - val_loss: 0.1084\n",
+ "Epoch 76/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1088\n",
+ "Epoch 77/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0781 - val_loss: 0.1083\n",
+ "Epoch 78/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0781 - val_loss: 0.1094\n",
+ "Epoch 79/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0781 - val_loss: 0.1082\n",
+ "Epoch 80/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0781 - val_loss: 0.1093\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Epoch 81/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0781 - val_loss: 0.1089\n",
+ "Epoch 82/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0781 - val_loss: 0.1079\n",
+ "Epoch 83/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1094\n",
+ "Epoch 84/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1098\n",
+ "Epoch 85/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0781 - val_loss: 0.1088\n",
+ "Epoch 86/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1083\n",
+ "Epoch 87/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1084\n",
+ "Epoch 88/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1090\n",
+ "Epoch 89/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1071\n",
+ "Epoch 90/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1077\n",
+ "Epoch 91/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1075\n",
+ "Epoch 92/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1085\n",
+ "Epoch 93/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1084\n",
+ "Epoch 94/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0781 - val_loss: 0.1089\n",
+ "Epoch 95/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1071\n",
+ "Epoch 96/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1076\n",
+ "Epoch 97/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1074\n",
+ "Epoch 98/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1080\n",
+ "Epoch 99/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1083\n",
+ "Epoch 100/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1090\n",
+ "Epoch 101/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1088\n",
+ "Epoch 102/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1079\n",
+ "Epoch 103/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1101\n",
+ "Epoch 104/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1078\n",
+ "Epoch 105/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1085\n",
+ "Epoch 106/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1081\n",
+ "Epoch 107/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1082\n",
+ "Epoch 108/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1081\n",
+ "Epoch 109/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1072\n",
+ "Epoch 110/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1080\n",
+ "Epoch 111/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1073\n",
+ "Epoch 112/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1080\n",
+ "Epoch 113/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1073\n",
+ "Epoch 114/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1080\n",
+ "Epoch 115/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1078\n",
+ "Epoch 116/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1071\n",
+ "Epoch 117/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1089\n",
+ "Epoch 118/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1084\n",
+ "Epoch 119/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1094\n",
+ "Epoch 120/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1082\n",
+ "Epoch 121/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1092\n",
+ "Epoch 122/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1079\n",
+ "Epoch 123/200\n",
+ "990/990 [==============================] - 1s 2ms/step - loss: 0.0780 - val_loss: 0.1083\n",
+ "Epoch 124/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1085\n",
+ "Epoch 125/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1088\n",
+ "Epoch 126/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1076\n",
+ "Epoch 127/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0779 - val_loss: 0.1083\n",
+ "Epoch 128/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0780 - val_loss: 0.1081\n",
+ "Epoch 129/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0779 - val_loss: 0.1082\n",
+ "Epoch 130/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0779 - val_loss: 0.1077\n",
+ "Epoch 131/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0779 - val_loss: 0.1084\n",
+ "Epoch 132/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0779 - val_loss: 0.1079\n",
+ "Epoch 133/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0779 - val_loss: 0.1079\n",
+ "Epoch 134/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0779 - val_loss: 0.1096\n",
+ "Epoch 135/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0779 - val_loss: 0.1090\n",
+ "Epoch 136/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0779 - val_loss: 0.1085\n",
+ "Epoch 137/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0779 - val_loss: 0.1073\n",
+ "Epoch 138/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0779 - val_loss: 0.1077\n",
+ "Epoch 139/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0779 - val_loss: 0.1072\n",
+ "Epoch 140/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0778 - val_loss: 0.1077\n",
+ "Epoch 141/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0778 - val_loss: 0.1087\n",
+ "Epoch 142/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0778 - val_loss: 0.1089\n",
+ "Epoch 143/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0778 - val_loss: 0.1081\n",
+ "Epoch 144/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0777 - val_loss: 0.1089\n",
+ "Epoch 145/200\n",
+ "990/990 [==============================] - 1s 2ms/step - loss: 0.0778 - val_loss: 0.1077\n",
+ "Epoch 146/200\n",
+ "990/990 [==============================] - 2s 2ms/step - loss: 0.0777 - val_loss: 0.1080\n",
+ "Epoch 147/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0777 - val_loss: 0.1086\n",
+ "Epoch 148/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0776 - val_loss: 0.1082\n",
+ "Epoch 149/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0775 - val_loss: 0.1071\n",
+ "Epoch 150/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0775 - val_loss: 0.1085\n",
+ "Epoch 151/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0774 - val_loss: 0.1082\n",
+ "Epoch 152/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0772 - val_loss: 0.1078\n",
+ "Epoch 153/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0771 - val_loss: 0.1081\n",
+ "Epoch 154/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0769 - val_loss: 0.1077\n",
+ "Epoch 155/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0767 - val_loss: 0.1079\n",
+ "Epoch 156/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0764 - val_loss: 0.1085\n",
+ "Epoch 157/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0761 - val_loss: 0.1081\n",
+ "Epoch 158/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0759 - val_loss: 0.1066\n",
+ "Epoch 159/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0756 - val_loss: 0.1077\n",
+ "Epoch 160/200\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0755 - val_loss: 0.1069\n",
+ "Epoch 161/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0753 - val_loss: 0.1095\n",
+ "Epoch 162/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0752 - val_loss: 0.1078\n",
+ "Epoch 163/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0751 - val_loss: 0.1074\n",
+ "Epoch 164/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0750 - val_loss: 0.1085\n",
+ "Epoch 165/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0749 - val_loss: 0.1073\n",
+ "Epoch 166/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0748 - val_loss: 0.1077\n",
+ "Epoch 167/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0748 - val_loss: 0.1078\n",
+ "Epoch 168/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0747 - val_loss: 0.1072\n",
+ "Epoch 169/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0746 - val_loss: 0.1073\n",
+ "Epoch 170/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0746 - val_loss: 0.1080\n",
+ "Epoch 171/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0745 - val_loss: 0.1080\n",
+ "Epoch 172/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0745 - val_loss: 0.1083\n",
+ "Epoch 173/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0745 - val_loss: 0.1063\n",
+ "Epoch 174/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0744 - val_loss: 0.1083\n",
+ "Epoch 175/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0743 - val_loss: 0.1075\n",
+ "Epoch 176/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0743 - val_loss: 0.1077\n",
+ "Epoch 177/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0743 - val_loss: 0.1067\n",
+ "Epoch 178/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0742 - val_loss: 0.1065\n",
+ "Epoch 179/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0742 - val_loss: 0.1067\n",
+ "Epoch 180/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0742 - val_loss: 0.1065\n",
+ "Epoch 181/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0741 - val_loss: 0.1060\n",
+ "Epoch 182/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0741 - val_loss: 0.1072\n",
+ "Epoch 183/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0740 - val_loss: 0.1073\n",
+ "Epoch 184/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0740 - val_loss: 0.1068\n",
+ "Epoch 185/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0740 - val_loss: 0.1057\n",
+ "Epoch 186/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0739 - val_loss: 0.1062\n",
+ "Epoch 187/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0739 - val_loss: 0.1057\n",
+ "Epoch 188/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0739 - val_loss: 0.1059\n",
+ "Epoch 189/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0738 - val_loss: 0.1067\n",
+ "Epoch 190/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0738 - val_loss: 0.1065\n",
+ "Epoch 191/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0737 - val_loss: 0.1066\n",
+ "Epoch 192/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0737 - val_loss: 0.1057\n",
+ "Epoch 193/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0736 - val_loss: 0.1058\n",
+ "Epoch 194/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0736 - val_loss: 0.1067\n",
+ "Epoch 195/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0735 - val_loss: 0.1060\n",
+ "Epoch 196/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0735 - val_loss: 0.1055\n",
+ "Epoch 197/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0735 - val_loss: 0.1054\n",
+ "Epoch 198/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0734 - val_loss: 0.1059\n",
+ "Epoch 199/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0734 - val_loss: 0.1055\n",
+ "Epoch 200/200\n",
+ "990/990 [==============================] - 1s 1ms/step - loss: 0.0733 - val_loss: 0.1058\n"
+ ]
+ },
+ {
+ "data": {
+ "text/plain": [
+ "<keras.callbacks.History at 0x7f6908ac4278>"
+ ]
+ },
+ "execution_count": 16,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
"source": [
"autoencoder.fit(densities, densities,\n",
- " epochs = 500,\n",
+ " epochs = 200,\n",
" batch_size = 20,\n",
" shuffle = True,\n",
" validation_data = (vali_data, vali_data))"
]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 17,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "decoded_imgs = autoencoder.predict(vali_data)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 18,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "image/png": "iVBORw0KGgoAAAANSUhEUgAAAj8AAAB8CAYAAACG/9HcAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nOx9WWyc53X28818s+/7cIarSIqiRGqxJNuyE1uJaydxYMeJ46RO0ixOUaBAW7Q3bYO2F+lFrloU6EWBALlo/gSIk6ZugniL7Xi3bGsjKVEiRXEfcoaz7/v2XzDnaEhTEkkNKVL6HoDQQnLme+bdznuW5wj1eh0SJEiQIEGCBAl3C2S3+wEkSJAgQYIECRK2E5LxI0GCBAkSJEi4qyAZPxIkSJAgQYKEuwqS8SNBggQJEiRIuKsgGT8SJEiQIEGChLsKkvEjQYIECRIkSLirIG7khwVB2NV18fV6XbjR93c7PwCRer3uuNEP7HaONxtDQOK4GyCtxd3PUZqny7jTOe52frjOWpQ8P3cW5m73A0iQIAGAtBYlSNgpWHMtSsaPBAkSJEiQIOGugmT8SJAgQYIECRLuKkjGjwQJEiRIkCDhroJk/EiQIEGCBAkS7ipIxo8ECRIkSJAg4a6CZPxIkCBBggQJEu4qSMaPBAkSJEiQsIsgCDeVH9q1EASBv7YSGxI53G7I5XL+EGq1Gmq1Gur13a63tAxBECCXyyGTLdufxK9Wq93mJ2seBEGAQqGATCZDvV5HvV5HpVK5ozjKZDIolUrmWKvVmOOdMFdp/anV6hUcy+UyqtXqrucoCAJkMhm0Wi3zq1arKJVKqFQqu54fsMxRFEXodDrmWKlUUCqVUC6X74j1KJPJIIoiDAYD5HI5r8NCocBzdTeD5qlCoYDJZGKO5XIZ+XwexWJxV3Ok81ClUsFisUAul6NSqaBcLiOXy6FQKKBSqTT1PXek8UOL1WQyQRRFXqx3wiAD4ElssVhW8Mvlcsjn800f5NsBuVwOpVIJu90OhULBCzWTySCXy6FcLt/uR7wl0GJVq9Vwu90QRRG1Wg2lUgnpdBqZTAbFYvF2P+YtgYxXjUYDr9cLhUKBarWKQqGAVCqFVCqFQqFwux9z0yDDVafToa2tDUqlEpVKBcViEfF4HIlEArlc7nY/5i2BDhSj0YjOzk4olUqUSiUUCgXE43FEo1FkMpnb/Zi3BFEUodFoYDab0dPTA4VCgWKxiGKxiEgkgnA4jGQyebsf85ZAxqvNZsPevXuhVCpRKBRQKBQQDAYRDAYRj8dv92NuGkqlEgaDAU6nE/39/VAoFGz0BINBLC4uIhKJNPU9d5zx02j97d+/H2q1mm8pwWAQfr8fsVjsdj/mpkEL1Waz4dChQ1CpVCiXyyiVSlhaWsL8/DzC4fDtfsxbgkKhgF6vh8vlwpEjR6BSqVAsFpnjzMwM/H7/7X7MW4JSqYTJZILH48HRo0f5UCkWiwgEArh69Sp8Pt+u9hyo1WpYrVa0tbXh+PHjUCqVKBaLKBQKCAQCGB8fx/T09G3jKAjCLb23VquF3W5HV1cXjh07xgcKjeGlS5dw5cqV28qPsNln0Ov1cLvd2LNnD44dOwaFQoF8Po9yuYxAIICLFy9idHT0tnh/Voc1NsNREASYTCa0trait7cXR44cYY6034yMjOD8+fO3lSPN1c1ytNls6OjoQF9fHwYHB1eMYygUwtDQED788MNt59gYniJuG+Uok8ngdDrR3d2Nvr4+7N+/H6IosiMgHA7j/PnzePfdd5vq+BA28qA36vEhl8s/EcapVqsbeli5XI4DBw7g+PHjGBwchFarhSAIqFQq/PqJRAL//M//jHw+v+7XJdxKDxMK3xC/arXKrtX1QhRFHD16FPfeey/6+/t5QVSrVSgUCsjlchQKBfzDP/wD8vn8ZhbKuXq9fuxGP3A9jnTLpzGk56pWq+vmSK/xwAMP4L777kNHRweHRmq1GtRqNRQKBQDgb/7mb5DL5TbM8VZ67QiCAJVKxRwb5+h6PVH0Gp/5zGdw//33w+FwrPBGarVa5vmXf/mXyGaz28pRJpNBrVZDFEUOF1cqFXYhrwf0Go8//jiOHz8OjUbDG5EgCDAYDNBqtTAajXjuueeQyWSazvF6/ORyObRaLY9hI7eN8NPpdPjyl7+MI0eOQC6XIx6Po1qtQiaTwWw2Q6fTweFw4Nvf/jbS6fS2rUXyKOp0Oh7DRn6lUmldby6Xy2EwGPCnf/qn6OvrAwCEQiFUq1XI5XLYbDYYDAa0tbXh2WefRSqV2rZ5SiEqvV7PHCkEVyqV1s1RoVDAbDbjW9/6Ftra2lCtVuH3+3k/dTqdMBgM6O3txTPPPINkMrmtHJVKJYxGI0Rx2cdAF0C6JK3jdaFQKGC32/Gd73wHVqsV5XIZPp8P1WoVSqUSHo8HBoMBg4OD+PKXv4xEIrHla5HmKF0AKXpB3jb6Wg8/tVoNp9OJ7373u9DpdMhms5ifn0e1WoVKpUJbWxuMRiMOHTqEp59+GvF4vGlrsSmeH7lcDr1eD6VSye5/8tZks9l1WaO04E+cOIEjR46gVqthdHQUqVQKgiDAarXC4XDgyJEj0Gq1KBQK23Ijo4E2Go1QKpWQy+V8WBaLxXXzE0URRqMRJ06cQHd3N4rFIi5cuIBMJgO5XA673Q6n04kTJ05sKz/g2iKzWCycv0J5D4VCAZlMZt0cLRYLTpw4gba2NiSTSYyMjCCbzUIul8PpdMLlcuGhhx6CVqvdrIG3KdBmZLPZ2JClOVooFJBOp29qqAuCAKVSCYfDgRMnTsBmsyEYDGJ4eBi5XA6iKMLtdsPlcuFzn/sctFrtpgy8zYK8pg6Hg41M4pjL5dbNUaVSoaWlBceOHYNarcb8/DwuXLiAQqEAuVwOr9cLh8OBJ554AhqNZlMG3kZBoXC1Wg2Xy8UHCh0kuVwOqVTqpvwov6etrQ0HDx5EpVLBxMQELl26hGKxCIVCwfyefvppaDSaTRl3m+WoUCig0+ngcrkgl8sBgA+TbDaLZDJ5U45k+OzZswddXV3IZDK4evUqrly5gmKxCJVKBY/HA5fLhb1790KtVm/WwNswZDIZVCoVe4ZpP6V1mM1mkUgkbsqR0iL27t0Lt9uNcDiMq1evYnJykt+ntbUVDocD/f39UKlUW86NIJfLodFoYDQa4XA4+JJLXtNMJoNKpXJTjgqFAjabDfv374fRaMTCwgImJiYQCAQgk8kgCALcbjccDgd7g7YDFIIzGo2w2+0QBIHHMJfLIZvNIhaL3ZSfSqWC0+nEwYMHIYoij188Hud83/n5eT73ac03jcctv8AfwzgtLS1QKpX8QVCODm2+N/ogZDIZNBoNOjs70d3djWAwiIsXL2J4eBiiKEIURcjlcrS0tHCI4VZd3usBbbh6vR4tLS384dNtk5Kw1sNPr9ejp6cHHo8H8/PzuHjxIiYnJyGKIh/MdrsdJ06cYEt6O0CHncFggNfr5ZsmeWzItXozjqIowmw2o7+/HyaTCRcvXsTFixcRCoWgUCh4IptMJpw8eRJyuXxbDR+1Wg2LxYKWlhYA4KTkWq3GoY71cLTZbBgYGIBcLsfHH3+M0dFRNnwAIBAIwGQy4fOf//y2zFECXR6sVivcbjfPUfLgabVaPkRvZMgqlUoOV+ZyOXz88ccYGxtb4fWMx+PQ6/V46qmnAGw+JLMRUDKr1WqF3W5nfnSQ6nQ6Hscb8VOr1fB6vTh69Cgbrj6fj9dhoVDA1atXEQwG8dWvfpU9vNsBpVIJs9kMu90Og8HAHh8yiOjCcDOOWq0WHR0dOHbsGGZnZzEyMoJYLMZ7ablcxtzcHHsJtjPpWa1Ww2azwWazQa1W85rTarXQaDR8YbgZR71ej+7ubtxzzz24fPkyRkZGUCgUoFQq+QK3uLjIl9Nisbhta1Gr1cLpdHLibrFYRKVSgVqt5q9sNntTjiaTCX19fTh8+DDOnTuHq1evolwur+AYDAZ5f96uC7PBYIDb7eY52shPoVCsMKZvxM9qtWL//v0YGBjAqVOnsLi4iGq1Co1GA7VaDblcjmg0CplMxo6UZvK7JePHYDDA4/HA4XAgEAggEAigXC6zJ8hqtUKhUCAWiyGTyVzXnel2u/Hggw+ip6cHP/3pTzmkZTAYYDAYYLPZ4HA4YLfbsbS0hKWlpW0ZZLPZjPb2dhgMBszMzCAej6NSqXBOi91uh8fjYX7Xc7u3t7fjoYcegsvlwo9//OMVr280Glfwm5ubQzAY3LaF6nA40NXVBblcjitXriCVSqFWq61IQPN4PJwYuVYITBAEdHd34+GHH4bJZMJPfvITAMtGh9Vq/QTH8fHxbeMoCAK8Xi86OjpQqVRw6dIl9jiRN87lcqGlpYXH8Xoc9+/fj4cffhgKhQI//elPmaPdbofJZILVaoXT6YTNZsO5c+cQCoW2jWNXVxfa2tqQzWYxNDTExqtcLofJZGKONI5rGXkymQxHjhzB/fffj1qthl/+8pd8AbDb7TyOtLG/8847CIfDW85RJpNh7969cDqdiMfjOH/+PD+/TCaDyWRij1ssFkM2m12Tn1wux7333ovDhw8jn8/jd7/7HRvGFAYifiaTCS+//DKi0eiWcmt8tgMHDsBkMiEQCODcuXN8cAiCALPZzByj0ShyudyaHEVRxIMPPoi+vj4kEgmcPXsWgiBAr9fDYrHwvuVyuaDVavE///M/25YoK4oiDh48CLVajdnZWSwsLPDcEQSBLydOp5M5rnV4KhQKnDx5Em1tbQiFQrh06RJEUYTVaoVOp4Ner4fD4WAP4U9+8hMkEolt4ahQKHDPPfegXq9jYmICkUhkTY52ux2xWAz5fP4THMnD/Mgjj8DhcGBychLT09NQq9UwmUwr9man04lKpYL//M//RCqV2nJ+KpUKg4ODKBaLGB8fZ68o5f3Q5auR3+r9gcJdjz76KAwGA86fP49gMAi9Xs8eOrPZDIfDAZvNhkKhgB/+8IfIZrNN5bJp44eSsFpbW1EoFDA7O7tikAuFAkqlElwuFzQaDf97LezZswc9PT3IZrN80yGPkkKhQL1eh0KhQDgcxs9+9rNtO1CcTifcbjei0egKg4sql2q1Gt9gqKRyLezdu5cNCLlczjcwlUqFer3OHqSZmRm8+eab2+oRaWlpgdlsxuLi4op4cblc5g3DarUyx7UMg3q9jn379sFsNiMQCEAURQ6BiqKIYrGIVCoFlUqFxcVFTE1NbStHj8cDtVqNmZmZFWGoarWKZDIJuVwOs9kMlUp1w5JKqkKYm5uDQqGASqViV3MymWRP0uXLl7c1EVgul8Pj8QAAfD4fe3zI85RKpaBQKGA0Gpnj9TxcfX19EAQB09PTUKlUfJsrlUoIhULI5/NIpVKIRCLbxpHCicViEbOzs2zY0YabyWSQSCSg0+k4aflG/ABgenp6RX5WJpNBMplEMpmE3+9HKBTC3Nzcto0hhZ0zmQzm5uZYKoHCG9lsFul0GiqV6oYcBUFAb28vqtUqZmdnodPpoNFoIIoiotEowuEwwuEwJiYmsLS0BL/fv237qUqlgk6nQyQSWbGfUhl3NpvluUoc1zIM5HI5uru7kc1mMTc3B6PRCI1GA7lcznwo92dpaQmhUGjL+dGzaTQayGQy+P3+FYYPjSOFhWjvuF61pCiKaGlpQSqVwsLCAu/BwPLcBYC5uTnkcjkEg8FtMe4oVw4Ae9VoHRK/fD6PXC7H+z+lUKyGUqmExWJBJBJBIpGAw+GASqVCpVLB+Pg4gGXvXiKRQCQS2ZLK2U0bPzKZDG63GxqNBjMzM5/4HrmqarUaT9jVoNj0kSNHUK/XcfXqVVitVk5GBYBwOIxQKISJiQkUCoVtKz2Vy+Vwu90QBAE+n++6/Ojfa/GjuPTg4CAqlQqmp6dht9tXaG0sLS1xfhPFhLcLoijyzWF+fp7/v1GDiAy663GkTbu/vx/FYhEzMzNwOp1cBZDL5RCLxdg9SrlS2wWFQgGr1YpUKrViHCnpkjjSmK6uQGlMOuzq6kI6nUYoFILT6eTficfjiEQi7HreSAL1rYJuiVqtFqFQCLFYjI0e4kiaGcR7LY5KpRJutxtutxvxeBzpdBoul4vnQiAQQCgUWhE62w5JBjo0ZTIZIpEICoUCP79cLl+Rv9W4Ca9+DbVaDY/HA6PRiEgkgnK5zPNUqVRiamoKwWCQL2k3c9k3ExT2FwQBS0tLzIU++0aOdLiu5kiv0dbWBlEUEQ6HuYqGXmNhYQFLS0u8DoHtCVnS82m1WtRqNQQCAQDXKqHoskvaNY36bqtfQ6fToaOjAwDY4DUYDHzRmpyc5D11uzWMKOJRLpcRDAY5n4mkTejiS8nKa3Gk1+jp6QGwfKmyWCwwm838GV2+fBmJRIKTxLfzkmU0GpHP5/kiT95lCsORs6IxTL76NShXC1i+uJhMJthsNnYGXLhwAclkkkN6W6WZtinjhyxcu90On8+HdDoNnU7Hbkqy3JVKJVKpFMcEV0Mmk8FoNEKtVnN5NyVQuVwu2Gw2hEIhjI2N4dKlS5xfs9UgN7Fer8fk5CTHpMnwarz1k9bJan60cZnNZsjlcgSDQT6klEolWlpaYLVa4ff7MTY2htHR0Q1Vc9wq6LMXRRFjY2NQqVSoVqscG1cqleviSO7mer2OcDjMi1Sj0aC9vR1GoxE+nw+XL19mjttl/NDnX61WMTc3B4vFglQqxRu/SqVijslkcs3bdKPxU6vVEI1GYTabudKko6MDGo0Gc3NzuHTpEi5fvrytYpVUuVMoFOD3+2G1WhGLxbisVq1WQ6VSQRRFHsfVHCkJ1eFwoF6vI5lMwmw2Q61Ww+FwoKOjA6IoYnZ2FmNjYxgbG0M+n2dv51ZuvmSgZ7NZZDIZ3hNW85PL5dflR3pMLpcLAJBOp2E2mzmXr7Ozk+fI+Pg4xsfH+Qa7mQqhjUKhUMDlcvHt3eVysWekXq9Do9HwTfp685Sq4NxuNwAgm83CZDLBYrGgvb0dHR0d+MxnPoO5uTlcvXoVExMTyGazyGaziEajWz5fqTIpHo9DrVZDqVRy6LtWq/G+KJPJkEwm18yHofxLr9fLexWFrSnse/LkSczNzWFmZgaTk5PIZDJIp9NsEG3lWJKBnUgkYDQaodPpEAwGuaqUimaAZaOGnAONIA+tx+NhD6fJZILX60V3dzc8Hg8eeugh+Hw++Hw+zMzMIJ1OI5FIrPD6bgU0Gg1HMOx2O8rlMiKRyAp+pAV2vbwt0pxqaWnhKm7aR3t7e2E2m/GpT30Kfr8ffr8f8/PzSKVSiMfjmJqaaqpBu2njh25jlUoFRqORQ1akOFkqldgav16ZrVarRXt7O1QqFXK5HN8OKPHvwIEDGBgYwP79+wEAkUiE3XxbKXRIhwGwfDMym81IJpPML5fLMT+6Va9ltBA/uVzOSbFarRZarRYmkwn79+/Hvn37cODAASiVSoTDYWQyGSwuLm65uizlOlAJOhk/5LWghLxG78haY0g3MXJ50u3MYDDwDWbv3r0YHByEyWRCKBRCMpnEzMwM37C3miMtTEp8JK9FOp1GoVDgctvrcdTr9WhvbwcAFAoFTmDX6/WcXNrb24uDBw/irbfewtLSEpLJJMbGxm6a1HiroEOvXq+zgUbiYOVymcMIjdVRa11EqOyZPiNaixQaam1tRU9PDw4dOoRTp06xl2l4eBj5fH7L1iNVzlDeS61W43GrVCpIJBIrqkyvl7RuMBjQ2trKidL0uVHYq729HV6vF4cOHcLZs2cRjUYRjUbx0Ucf8d62VaB9gZK6a7UaGwDVahXxeJw5UlXN9Ti2tLTwmlIoFPy6MpkMXq8XTqcThw8fxsWLFxGPxxEOh/Huu+/yJXUrOWo0Gj4vaOzIexGLxbia9npFJCSz4Ha7UavV2LNJ6uMAOOfu8OHDmJiYQDKZRDgcxhtvvIFYLLalScE0jrVaDUajEeVymcexXC4jGo2ygUcc1wrrGY1GOJ1OjpooFAr2qFSrVdhsNlgsFuzduxehUAjpdBrhcBivvPIKnyHN5kgeUkpGpjM7lUqxxzsSiTA/KulfK9/HZDLB6XTy+JIXjH7eYrHAaDRiz549LBgbDofx0ksvIRAIcIrBrWLTxg9V8Oh0Ouh0Ov5A0uk0kskkbzKNuj+NoM2W4tPpdBrANZ0UOoQNBgP27duH73//+5iamoLP58Pbb7+NaDS6ZSEiGmgAsFgsKwa+kR+FS1aLV9EEbm9vR09PD6rVKvL5PARB4AOFjCidToeenh5873vfw+zsLBYXF/H73/8e4XC46QleqzmSAUveKQpTpVIpNvZKpdJ1OZpMJnR0dGDPnj08XsSRcoTy+Ty0Wi06OzvxZ3/2Z/D5fFhaWsJLL72EhYWFLS2xbdS8oXEEwLkFtHCLxSK73FdzpFsJeQdo06XbeCaT4QR9r9eLJ554AslkEqFQCIIgYHZ2FvF4fMsMIDLUyRNFIWPKYSEjloy2tcbRbDajra0N7e3tKxKJqeVDIpGAyWSCWq2G3W7Hww8/jGq1yoqrU1NTCIfDWxIGI6+NUqmE1WoFsLyX0Phls1k+0K7Hj7wfXq93RY4CudljsRjvXwaDAcePH4coiiymOjY2hkAgsGVeWeJInkg65IgjeaEoTLLWgWK1WtHe3o7W1lY2YigMQUJxVFWqVCoxMDDAxqIgCBgdHcXc3Nym9NPWA8pxpOo8qrZMJpPMkfYP0qe6HseWlhZWpaZwfKFQwNLSEss8CIKAzs7OFb8/PDyMqakpPmuaCfKCk7YPXaYorzCdTq8IN96IY2trK9xuN3sCKd0gk8kgEAjAarWyY4GSnonr8PAwxsbGNquHc0MQP1qHBoMB9XodiUSCC5pojdyMn9PpRCQS4fAY5SZWKhU+j+girdVqYbPZAAAXLlzAyMhIUwpmNm380CLyer1cIkoPHIvFEAwGEQqFViTRElGVSoVDhw5hcHAQvb29CIfDvOl2dHTg4MGDSKVSkMlkmJmZQSKRgFarRVdXFw4fPowvfvGLGBoawiuvvMJhhmaiMY/C4/FAJpNxKwqZTMYJ0JFIZIVbnBaiSqXCkSNHMDAwgJ6eHiwtLa3gd+jQISSTSSiVSvj9fp7kNpsN7e3t+JM/+ROMjY3h5ZdfxpkzZ7bkVk0bo16v57JCg8HA+QWUlBgOh1dUERBHjUaDY8eOYWBgAHv27OEEQ2C5uu3w4cM8bqFQCFNTU8jn8zCZTOjp6cE//dM/IRgM4tVXX8Wbb765JQcnzTWj0Qi9Xg8AnBRLZaKhUOgTtyUS1NRoNCy42dXVhYWFBfaOtbW14ciRI0gkEjAYDIjFYlhcXEQ8HodWq4XVasVf//Vfo1gs4g9/+AN+85vfbMnhSQea2WzmEAIZZjKZDEtLSwgGg4hEIitKRSn/QKvV4t5778XBgwfR3t6OhYUFvrh4vV7cc889SCQSsFqtiMfjCAQCCAaDEEURCoUCf/7nfw6ZTIYPPvgA/+///b+m36yJH224lPfSWM4cCoW4OqjRuCE9kvvvvx8DAwNoa2vj0EClUmF17kQiAafTyUb/wsIC7ynPPPMMdDodzp8/j//6r/+6bgXSrXKk3DRgWdeHOAqCgMXFRQSDQa6eAa6F1SkUdOLECQwMDMDj8bAIXi6Xg8vlwrFjx5BIJOD1evlyQ/lN5XIZjz76KJ544gksLCzg3/7t39all7QR0HiQYUDyEqQNBwDz8/MIhUKIx+Nr5nXp9Xo88MADGBgYgNvtxvz8PHw+H8rlMux2O+69914kEgl0dHSwJhIZOtlsFvfccw8eeOAB5PN5/OhHP0I0Gm36nkPPajKZ+ALS09PDHKmSN5FIoFgsruBIFVwPPvggBgYGOB9tcXER6XQaBw4cwH333YdkMonu7m6+UF+9epVbsXR2duLQoUOQyWT413/9VywtLTV1z6F5qtVqUalUkEqluNVGvV7H9PQ0txJp5Ed7hdFoxIMPPogDBw7A6XQCAHvJ+/r6cO+99/LfgWUv9eTkJCfpm81mPPXUU3j66afx7//+75ifn7+l/WbTCc+1Wo1zCbLZLCYnJ9mV3NbWxnH6oaGhFYJHpANAFSTlchkej4cnKsWGyTqkMNCpU6cwPj4OuVyOI0eO4L777uNk660QkqtWqxzWSiaTmJiY4EXY2toKl8uFbDaL8+fPI5FIsJeLPERU5VQqleD1ejE7O4t0Og25XA6XywWPxwOPx4N8Po9AIIAPP/wQ8/PzKJVKGBgYQHd3N5555hmMjY0hnU43fcOlGzJtsPF4HFevXoVarYZer0dbWxvcbjfS6TSGhobY20XG3WqOHR0d8Pl8bAza7Xa0tLSgvb0d5XIZS0tL+OijjxAIBDA7O4vu7m643W587Wtfw/nz51lht9kcKbmQ+vzMzMywAJnX64XX60UqlcLw8DASiQQnXKrVami1Wq7IKJVK6OrqYgOnXC7DYrFw64B6vY5QKISPP/4YCwsLmJychNPpRHt7O55++ml88MEHCIVCTQ+f0DjSc1IfnEaOHo8H6XQaw8PD7DKmcEFjhVS5XGZjPRKJIJ/Pc0ltd3c3lEolYrEYTp8+jbm5OSwsLCCRSGDPnj04efIk3njjDSwsLDQ1nEkGnVKpRDabhd/vRzgc5l5OHR0daGtr4zEkqQbaZ0hFmPjt3bsXkUiEw69qtRo9PT3o7u6G0WhEMpnEmTNnMDc3h9nZWVy+fBmdnZ0YHBxEZ2cnpqenmy7OSWNIeVmNeZRmsxmdnZ1ob29HIpHAyMgI7wfUd02v13O7g2q1iv7+fiQSCQQCAUQiEa6OIo7lchnnzp3D/Pw8pqamcOHCBbS1tcHj8aCvrw+XL19et7DpRjjS2kqn05ibm0OxWIROp4PFYkFPTw86OzuZI70/XULpYkbacfv370c2m8Xi4iICgQAqlQo6OzvR29vLocOWlhbMz89jenoaFy5c4PyugwcP8r7dzD2nUZogkUhwIRDJvvT29qKrqwvxeJwFbintQKPRMEeSMejr60M+n4fP58Pc3Bwb8F1dXXyZI44zMzP46KOPYDQa0d3djePHj+PUqVNN9ciSZ7xWqyESiWBubo6NNujHaB0AACAASURBVKvViv7+fnR3dyMajWJ0dJQvW7SXUpl+LpdDpVLBvn37UCqVMD8/j8nJSfT29qKjowNer5dV3Inf/Pw83nzzTQSDQXR3d+OBBx5AsViE3+/ftIG3KeOHKpUoBh2LxTAzM8O3kUQiAbPZDKvVipMnTyKfz+PSpUus61AqlRAIBOB0Orkcvr29HTMzMzh9+jR6e3tx9OhRGI1GWK1WTmicmZnB2NgY3nnnHRw9ehSHDh2C1+vlhdQsED86UKLRKFcKKRQKpNNpHvBHHnkEmUwGly9fht/v51vNwsICbDYbisUia+lMT0/j/Pnz8Hg8ePDBB9lDQCWMi4uLuHz5Mt59912+iXd1dWFiYmJLNlzK6SkUCgiFQvD7/WzdUxa+xWLB5z//ecRiMYyPj7OHhxal2WxGPp+H3W5HT08PpqamMDIyAovFgpMnT3LISa1WcxXElStX8MEHH3Cfmn379q3YDJoFqvgQRRGZTIY1ouiGlsvlWP/k8ccf56pC2kxzuRxmZ2dhMBjgcrlgtVqxb98+XL16FUNDQ1Cr1XjsscdQq9VgsVig0WiQy+XQ0tKC6elpvPXWWwiFQuxZOX369JZsuKRMnEqlVnCksnaDwQCz2YwvfvGLCAaDmJiYQDAYRKVSQSaTwdTUFLuWzWYz9u/fj8nJSYyOjqJer+Nzn/scSqUSJ2L29PSwbtPLL7+MpaUl3H///Th+/DhyuVzTN1zSDsvn82yYyeVyrv7S6/Uwm8144okn4Pf7MTk5yZUi6XSajXqz2QyTyYQDBw5Ao9HgypUrKBQK+MIXvoCWlhbo9XqYTCa0tbXBYDDAbrfj17/+Nfx+P+677z48+OCDyGQyt7ThrgXK45HJZMjn8wiFQkilUlyeTkaC2WzGk08+icXFRUxPT/PNPplM4sqVK+y9NRgMGBgYgF6vx8zMDJ5//nk8+uijcDgcHEJ0uVxcqTkxMYHh4WHWsUokEpidnW1aWgHtpxTWymQyCIVCKBQKUCgUiMfjKJVK0Gg0MJlM+NKXvgSfz4fZ2VksLS1xgvfY2BiA5cRig8GAAwcOwGg0YmJiAj//+c/x8MMPw2w2c8GK1WpFpVKBwWDA6Ogorl69inw+j5MnTyIej2NsbKypqQWNHNPpNHuXEokEh9hXc5ybm2OOlUoFo6OjfPZotVrs378fZrMZo6Oj+PnPf47jx49Dr9fzWqUzUqlUYmRkBH6/H+l0Gg899BDnyDarqSt5TEVRZH6CILBEBKmH2+12PPXUU5ibm4PP50MwGORm1iRGSSkJe/fuhd1ux/nz5/HCCy+gp6cHKpUKVquV91SdTofW1lZYLBb4fD6cPXsWDz30EHvsN9vwdFO9vSiv46tf/SrHxl966SV2OXs8HpZNHx4e5nyQz372s/jCF74AmUyGs2fPcmsHj8eDPXv2QBAEvPvuuxz3+8EPfgCbzYZYLIZXX32VS+t+/etfI51Oo7+/H+3t7fjFL37BFSg3Qn2dPUxo8Xzta19DrVZDOBzGa6+9xvxaW1vR19cHlUqFc+fOcdz8U5/6FL74xS9CFEWcP38e2WwWMpkMDocDe/bsgVarxfvvv8/J3X/3d3+Hjo4OFAoFvPzyy/wcJK7mdrtx33334Re/+AWGh4fX03153f2EKC/pmWeeQbVaxeLiIt5++23m2N7ejv3790OhUODjjz8GAORyOTz88MN48sknoVKpMDQ0xGNoNpvR1dUFq9WK06dPc57LN7/5TRw4cAByuRyvv/46u7TffvttLCwsQKvV4ktf+hKef/55fPTRRzfVq7jZGDZypHDll7/8ZVSrVczMzODUqVPMsaurC/v370etVsPFixc5pHHixAk89dRT0Gg0GB4eRjab5RBKe3s7XC4XLl26hGAwiFKphM985jM4duwYzGYz3nvvPdYO+uijjzA5OYlarYZnn30WL774IgsDNpOj3W7Hk08+iVKphImJCZw/f55zYLq6ujAwMIBqtYqhoSHeoI8dO4avfOUr0Ol0uHjxInOkihWv14v5+Xl2u+/btw/3338/WltbMTw8jHA4DEEQuMItkUjgO9/5Dt577z28/vrrWFxcvCWOxI88pY8//jjy+TxXRhJ30ghTKBQ4e/Ysu9sPHjyIr371qzAajStUuJVKJZxOJ3v8fD4fYrEYzGYzjh8/jr179zLver2O2dlZXLhwAdFoFM8++ywuXLiAF198EbOzszcbonWvRVEU4fV68eijjyKdTrP6O32vq6sLe/fuhSiK+Pjjj1GtVlEoFHDw4EF8/etfh8Vi4YOcOJrNZrS2tnJpORWJUBFJLpeDz+dDqVRCOBzG8PAwgsEgvv71r2NychK//e1vWW9ls2PYyFGhUKCjowMnT55ENBrFhQsXVlwou7q6sG/fPlSrVZw5c4arevr6+vDss8/CbrfjypUryOVynMCv1+vh8Xig0WgQjUbZKLVarRgcHIRer2cvWjabxcjICGZmZvC1r30NS0tL+L//+z+MjIzc8FK5EY4qlYq9EqQeHgwGASwn9e7Zs4c5fvTRR+yZ7urqwje+8Q24XC6uUKMEd6pSNBgM7NmNRCLQarXsKaFm39VqFRcvXsSlS5fw9NNPI5fL4YUXXsCHH354SxyJn1qtxr59+3DkyBH4/X4MDQ3xfq1SqbidiiiKOHv2LJ/7Ho8H3/rWt+D1ejE9PY1sNssV4XK5nFXbyZGysLDACfoHDhxAJBJBJBKBIAgYGxvD0NAQHn/8ccjlcvzmN7/BW2+9dbNLc/N6e5H3JhQKwePxwGazwePxIBAIoFarIRQKwWw2o7u7G62trcjlcshkMnjllVfw29/+FpVKBT/84Q9ZwZIS1kRRxKc//WkAQCwWwwsvvMD5N5lMhsNsWq0W6XQaMzMzOHz4MFpbW5uarEdekXA4DJvNxuq4VH4aCATYvej1epHP55HNZvHGG2/g5ZdfRrlcxj/+4z/ixIkTcDqd7K6VyWT4q7/6KwjCcguQDz/8EO+//z4A8I2BXISRSAQ+nw+f/vSn0draiqtXr67H+Nkwx2QyyeERt9vNi3VpaQkmkwmdnZ1obW1FPp+HSqXCW2+9hd///vcol8v427/9WzzwwANwu92c/V+v1/GNb3yDKzDOnj3Lt1KqNKIEeZlMxlUmXq8XBoOhqWJdVPmUz+eh0WjQ1dXFybkAsLCwAIPBgM7OTk6izGQyeOedd/DGG2+gXC7jL/7iL3D48GEWSiT14Mcee4wVScfHx/HWW2+xR6dQKECn0/F8peTxlpYWGI3Gmxo/G0G9Xl8hDrp3714sLCxwMqHf74der0dnZye8Xi9XT3zwwQd4++23US6X8c1vfhP33HMPu5uz2SxyuRwOHTqEhx56CHq9Hn6/H+Pj4zh79ixEUUQul2NhOcoTq9VqcLlcMJlMNzV+1gvKnaCS7/7+fs5/IVG7Rn6pVAqZTAYff/wx3n//fZTLZXzlK1/B0aNH4fV6odfruUeWy+XCvn37YDKZkM/nMTMzg9dff509n5SrQeW7JGpK3oVmeWIpublarcJgMODgwYNIJBJsRAcCAej1enR0dMDj8fA8HRoawunTp1Eul/H5z38ex44dQ1tbG3MMh8NsCDz88MNQqVTw+Xw4d+4cFzfQXkRpCKVSCWazeUs4ktfTZrPhyJEjLJgpiiIWFxe5qrK1tZUNltHRUfz93/89KpUKPv3pT+P48ePweDzQ6/XI5XL82TidTuzbtw8OhwPBYBDT09Pc7oTGTi6Xc6GJ0WjkIohmcaxUKqyz5XQ6cfToUZw+fRqZTAaiKMLv93MFMIVqc7kcLl26hB/84AeoVCrc+JpUxrPZLBYWFrhRa39/Pzo6OpDL5TA1NYUzZ85w4npjM2MSlW0mRyptJ4eFKIoYGhpigzQQCPAFkTofULuYf/mXf0GlUkF/fz/uvfdedHZ2smbQ/Pw85ubmYLPZ0NLSgqeffhrA8v788ccfQyaToVgscoSkXq8jHo/DaDTyPN0MNp3zQ94Cktzu7OyEWq1GIpHg3AkAKx6MKoqq1Sp++MMfcqWJ3W5n12g2m2W33ne/+11YrdYVyp8Gg4GTTrPZLEqlEpfHNxOVSgULCwsc09yzZw/0ej3i8fiKTHrKxCdvAn02P/rRj1ZUGpnNZg4z0Cb37LPPcpY+HYhGo5HLo+v1ZfVnclU3G+VyGfPz8+jq6oJGo0Fvby/nPUSjUVSrVQ5lUl4CLaRqtYr/+I//YA+SzWbjEs/GMvlHH30Ug4OD0Gg0CIVCXN5Jce1CocDu763iODc3h46ODiiVSuzbtw82mw2JRIJbMzSOGxkwVA7/4x//mBOg6VBozCOqVqvYt28fjh07hvb2dk4sppALVbXk83lO/GsmyIidn59HW1sbt0kg5VRSt20ct8bYfaVSwX//93/jV7/6Fc9VAGzYkUaXVqvFPffcgwMHDvBBptPpuFyZeu80VoI2Y8Ot15f7yy0uLrKo6sGDBxGNRpFIJBAMBlfsMaSxRF+VSgW/+tWv8MILL8BgMKzQfjIYDCvESinB2+FwIBqNQqVScVVpY2+hZs/Ter2OXC6HSCQCp9MJnU6HQ4cOIR6PI5VKIRAIMEcqk6a5Sn0UX3zxRbz22mt8k67X69zzjA5ECksMDg7i0KFDfFBnMhnWZaE5SwdpswwDMmIpJcJsNnMyPeUnkXgjcSyXy8y1XC7jrbfewttvvw2dTgebzcZniVqtXpFILZPJ4HK58Nhjj3EFESXolkolVspudqNMMjroQmm323Hs2DGkUikkEgn4/X7+XIljo/RGuVzGqVOncPbsWW7bYrFYOKevXq+zB4hyvr70pS9xn0kAvB4bq+E2axysxS+RSCCbzUKr1XK/S6osJY8NgZ6rUUJlZGQEFy5cgFarhcVigUwmY05UBahWqzls/sgjj3BPOyo2ooR2lUq1plDkenFLCc/T09PI5XLweDz8Yej1evT19a0QTAuFQnybpMVkt9tht9tXVG40LnBBEPCzn/2MywSp4VljU0OdTseeg2Yny9ZqNVy9ehWpVOoT/ChOSRYoxR4bM8/dbjesVivrIpAR4XA4+Jb85ptvckyWDkcq6aOkVHrdraiGojh6PB5HS0sLV7cZjUb09PQwR5LFJ6OGOJLeBHGkBHEq9QTAHbMpLtworgiAkzxJG6rZqFQquHjxIiKRCFpaWqBWq+F2u7kxIhlt0WiU52ljwi5t1BqNhpO86Ys45vN5vPfee8hkMggGg1wKr9Fo+DkymQy/9lZwHBoaQiAQ4HF0Op3Q6/UciqQkxVAoxPF34mgymWAymT7BkQx/2qxnZmYwMTHBuW1KpZIrBQVBYCHCZuenlctlnD59mntb6XQ6OBwO6HQ6dHZ2sjFDlXuUUEnPoNfrWUyVWuZQ2TVwrYKxUqlwUj55YnU6HVex5nI59kg0G8ViER9++CFz1Gq1sNvtXHxgMplQrVYRDAaZI4ngAcsSIVS5SRypiSYdSCaTCTKZDAsLC3xjB8Brl/L/8vn8lmjFFAoFvP/++9y/S6PRwGKxQKfTcZ5VqVSC3+9HNBrlBG56jsYEfTKyTSYTJ//WajXo9XqWrfjlL3/JvfqUSiUb5SRvkUqlmp5Hmcvl8O6773K3dSo8oObfxHFxcZE5NkpsUMEMiczSOWc2mwGANctoTr766qts1FHBDemypdPpppa8k5f5nXfe4f5dlJND4TlqMEzhZNoLGrWnSJqjWq2y86C1tRUAeN3qdDrIZDKcO3cOiUSCDT4KlZE8zq3IiNyS6Usy16IooqOjg/VdSKuCWgHQ4DQOQqVSQTweZ08PlZGTZ4CUg51OJ6xWK3w+HycS2+325Yf/o3EQDoe3RKArn89zN+SOjg4uR1Uqlfzhk8jUWvwSiQQn1VIMt1FkjppFkgoyJR6TSjLxo7LUrQDdOGUyGTo7OzkJkvRBUqkUwuEw0un0J9R8K5UKG24Gg4EXLOkDNSYCk7ou3f5MJhMbQLlcDktLS1umMUIhAAAckyaVZuphFgwG12xOSzodVKZPBwp5rCqVCvR6PQwGAx9QdECSlpIgCHyD3yrtptUcG6ULSqUSc8xms58wpCnsks/nebxortJhSJIBNJa0IdEmTBUui4uLTQ3PEtLpNF8aiB/tG8ViEfF4nI3X1RehRmHExpYf5Cmg5zUajSyiR9ol5G6nEK3f72/6oUkgSQmqXCJhQEr2Jo5rNTWlg4k8ro192ciYyefznNRNpf2keE5rkSrqGiVKtoJjqVRijnSg0V4UDofXbPhJxkWpVOI+YWTwUWpFPB7nSkCLxQKVSsVqyuR1pXNrq5rW0oFcKBSYI825dDqNWCzGHNf6jGlfIQ09WpOFQgHZbBahUIj3MJvNxgr1jaHaUqmEaDS6or9YM0C6WCSlQOciydhQ3z8qRljrvelsMBqN3BDaYDCwajylz1CBhcViYRVrQq1WY8fDtpe6A8ubJt0QSO2WBokkxSl8shrkXqVbCVnmFOIiYS9qHGo2m3ni02RXKBSYnJxEIBDYEuOAVIDpRkG3KjoslpaWEIvF1rQ8M5kM34pJAoDc6OQFyGQysNvtXI1DXh6KZ6tUKhY+3CrDYC2OarWab0aUX7HWBKODtFHHQavVsms0mUwikUjAbrfDYrGwF4EMQjpU/X4/fD7flvVta+RIjR5VKhXy+Tz3kLleZ2tyr5KXg9oiUCsFqjaw2Wx8UweuJZRTqCEYDGJubm5LDAMAfNMjjhQqJREyMtLXmqsUdqR8M1JYJWOQqvRMJhMcDgdMJhMMBgOAZaEzqsaKx+Mst78V/CiEQZcspVLJrnjit9Y8pWo4KgunNUfhScrjI70fl8sFvV7PZcj0mcViMUxNTW1Z9+xSqcQcaR2uxfF6v0v7jNFohMPhgNlsZnG9en25/Qzp6LS3t7PiMu3BlUoFkUgEU1NTTasQWus5KQeOLsp00aImlteTLiEjjTwhLpcLZrOZjVESyovH4ysuZI2tM+jwnpyc3BIhwEaOZKQ1cqQ+gNfb60hORKPR8OXfYDCwIU5VqNTn0mKxMEcAbMTG43HMzc0hGo02nWOxWGStItL7o04O8XicPVprgbw9FMVxOp0wGo1cEUd5WhSKdTgcnOdLeY3AcnsQ0vjaLL9NVXutBfLc1Ov1dWmZkGteq9Wir68PSqUSoVAIly9fXhHrJeuQDitKOiTjYnZ2dkW/phthvVnt1+NHYY71vJdGo+H4fV9fH9RqNZaWlnDx4kXOA6GeYRQeIzev3W7nuPT8/Dxr7KxjrNZdYdJsjv39/RBFEaFQiCv8gOXJTqWLZMBWKhU4nU5uTUB6MevpE7WR6otmczxw4ABEUcTS0hLOnTvHTQspB4iqFiis5HK5WLZ/cXGR4/E7jWNjiIUq85aWljiZktY15QS53W7epNxuN3tHA4EAYrFYUziuhx/lCW6E3+DgIERRhM/nw5kzZxrfj9XOW1tbOc/H6XQiHo9DEARu6dEY+r0BbmktUigRwLo46nQ6vkEPDg5CJpPB5/OtWIuUH2E0GtHR0YFKpcIyFWTsRCIRzqu4WTjhVufpRjmSLIVOp8ORI0dY/mRoaIjHgzjqdDrs2bOH8wttNhtzjMVi7JnYaRzpIkwcAWBmZgajo6MrxhFYntfUAJX63tEFh4ystTxoG+XYTH50cdLr9Th2bHl5jI+P48qVKyt+n4p/9u7dC5lMhkwmA7PZzJ5dMpTX2VqnedVea2GjOSmUJ6JUKqHT6ZDJZFaI5AHgWxflg9RqNRY+pCQy8iBtNTb6HuQFI48BxUEbY+kUEiE3J1VCUDO+Rn5bcUNZjY1ybPQwkPFG4TtacKTRROElcmcSRworbVd34s1wJM+PUqlEqVTiEuHGcaSbD1W00JymBPedzpHmaeNcbUwcJv6Ux1cul6HRaHgt0oa0HRw3yo/CzhQGogRxSgSmJGbKJSHRUqVSiXw+zyFf0qPZjjHcaA4jCR029l0ipWp63sYQE6UiUNiQOJJxt1XtWBqxGY7k3aKKQyqXb+RIxRPUckEQBC42oR5bO5UjJeXrdDpOe6C1CIAvITTWxImSv8lDRIn6W81xs2NIifipVGqFojold9N6pWgPAE7Ob+R3K7m+zU133wDoUCR3HR0UjVVFjb1nBGG5k7zNZsPMzAzfvrbLMNgoqCyPEiapxBi4VhlGXjLKdVIqlXA4HBgaGuKJu10H5mZAuV0U6kmlUp8IPdAYkZo36bZcuHCBdXV2MkcywPV6Pfe5agw9NFZRUS8fAOju7salS5eQzWZ5Hm/HZrsZqNVq1Ot1XouxWIxDO6srqSi8JggC9uzZg/HxcRan3KkcqXpEq9VCEIQVYQfi1ziOmUwGgiCgo6MDk5OTrBq9U/kBy2uR5imFuIgHee6II1UlyWQytLW1YWZmhjlul1GwGdB+QyHXYDCIfD7P3ofGys1arcaenu7ubvh8Pr6Y3OqhuZWgnB3iSCkdCoWCLyKUY0dhPgCsPt/IcSeOI4WtjEYjgOVy9nK5zOFdCovROCYSCV6LjWk0zWimfNuMH5LUp2aKZBg01vJTJni5XMbx48eh0+kQjUZXhIF26qFJiabUwZgOTKoAqtfrHGsn0Tm9Xo9YLMatHmij2qkgjgaDgeO9VLpIz94Yfjl69CgMBgPi8ThisRiP4U7mSB4dqtKIx+OQy+VcgkoLkP5+8OBBjmFHIpFdwZGkDKjklMq8yShqXGv1eh2Dg4MwGAzcMZuM153KkfIgKJcuHo+zSjCVitP4CYKAgYEBlnwgpeidzA+4xlGr1XLiL5Vbk+FGHGUyGQ4cOACDwcB5fSQquJM50qGo0WjY82+xWLjXVOOXTCZDf38/dDod0uk0/H4/j+NONXyAa9IparUa8XgcmUyGq4yJG+XhCYKA3t5ernpeXFzc8RxpDFUqFSKRCIrFIjo6OliqgC7JVMDU1dXFc/rs2bMr5vGt4rYZP8C1fi/k/aCkYKrEUKlUcLvdrF8SCoUQCoXYlbtTDZ9GVKtVVp2l6gRRFFGr1bhypqWlBQBW8Gt0V+9kUIkwdUanJEMqR9Tr9XC5XHA4HJxQSWqzu4Ej3bIaORqNRuh0Ok5ipP5XFouFe4jtNo5UHkseSCrxJ+OPKksoUTgajXKLjJ3OkTzHjfzsdjtsNhu36CEtLkrYJ367ZQxJSoOSgqk4w+l0QhRF1Ot1Hj+FQsF5L0tLSxxy3+kgSQJah5TnRHozwLL8hk6nAwBueEvtanYDR0EQOBRJFY0GgwEej4erqqgIqF6vs6DpblmLVB1KVZSkZeT1ejlx2m63cyXuzMwMIpEIy080k99tNX5IG4QaSFKViclkgs1mY9dYpVLBlStXWNhspw8wgSqbqHTYYrFwtQxVP9XryyJ14+PjiMViu4ofGQa5XI7H0G63cxUGldWSG/bSpUu7jiMdnDSOKpUKTqcTbW1tbASRd69QKGBkZISFMHcLRzJ+SAmbmiySCivJTpTLZeTzea6U2S0cyTAoFArQarWcj0fCrGq1GjabjdV/Sfuq0QO700GHSqFQYO0VSmymsnCz2cwcR0dHdyVHMmIFQYDBYIDFYmH1da1Wy2Ky+Xwew8PDu44jFQ4RRyoFdzqdPI6Neyo1Wqby853suQOuFSuUy2XW36KzXqVS8ZlRr9e3nJ/s5j+yNSCBMdJ9EUURTqcTra2t0Ol0qFarcLvduP/++/Hkk0/iueeeQ7VavSVRo+0ECRtSQqhKpeLmmIIgIJ/Pw2q14uDBg/jCF76A5557juXRdwM/AKx/QjoiVL6o1Wq5DFyn02Hv3r347Gc/i+9///uccLibOJL4Wzab5RJUKj+mmLTH48F9992H5557jpt77iaOdBsjA4iS0gOBAJLJJOtxDQ4O4rvf/S43p9wNHGkMG/mp1Wq+UcZiMQ4ZdHZ24tvf/jaSySSWlpZ2BT9gJcdGbaLFxUXMz88jEAjg8uXL3PLiG9/4Bnt+dhNHSpcgCYlarYbZ2VlMT09jYWEBp06dwsTEBERRxNe//nWEQiHWjdkNoKR8EmKk5tpXr17F5OQkZmdn8fLLL+PMmTPI5XJ4+umnEQgEsLi4uCs4kqQNaYhR8v2VK1cwOTmJqakpPP/88/jDH/6AYDCIJ598Ej6fb0v43RbPDwmN0Y0sn89Dp9OxaF69XofRaMTw8DD+8Ic/oKWlBX19fezSvJ540k4BuWfpi/iRjgewbAG///77iMViaG9vR29vL0wmExsOO5kfcM1bQPoZdDhSop1MJoNKpcKrr76KRCKBnp4edHd3w2AwsBbUbuDYOI7ZbJbFuEjfSK1W43//93+Ry+Wwb98+dHZ2slryOkuibytoHOmwzGQyMBgMXGWhUCig0+nw/PPPo1gs4sCBA6zGS3pQO5kjjSHtNY38iHelUsEvfvELztmi3lFqtXrbqrtuBeSBpXBzOp3m8KRcLmceL730EkRRRG9vL+urqFSqXcMRADc1JY6UMgEs6+v87ne/4xJpCvOR8Opu4EjVTjKZjItHisUij3GpVMI777wDvV6P8fFxmM3mFRx3MlbzS6VSfGbQOlWr1RgaGoJKpcLU1BSrz1MlaTNx28Je9XqdyzJjsRiA5YREUihVq9X4zW9+g1KphGPHjmFkZATT09NbJhLXbFD5rEKhYGVKMtxICfmDDz5gmfbR0VHMzs5umdDfVoBE4KhFBbDMkW5loihiZGSEXdRjY2OYnZ3dMrXqrUClUoFGo2HDAAALXZL67szMDGQyGRwOByYmJjA3N7cliuNbhetxrFQqnLAeDoe52m1mZgZzc3M7frMFrvU+o9JaOlA0Gg0bRAC4aiiZTMLn82F+fn5bJDSagdUcqfqODFo6WLLZLJdHLy0tYW5ubld4C4BrPd6oCoo4ksHXOJb1PypBx2IxzM7O3san3hjqf1TppnQP0kSjcRQEATabDdVqlT0nCwsLmJmZk+PmCQAAIABJREFU4fD8Tjbw6vU6Xz5oPhI/enbyqlMucDQaxdzc3Jbwa5rI4UZBzUup6oJyJkiTw+Vy4cyZMwiHw6hWqzh27Bg6Ozuh1+u5xPb1119HIBBY93veipjTRkHihaQKTOXepFHhdDoxPDyMUCjEN07Ksejt7cX4+Dhee+01LC0tbeRtb0lYbaNQq9UsIU8918rlMvdlsdlsuHTpEifj9ff3cy+m/fv3Y2xsDK+99lpTxxBoLkeVSgWz2cwyBNSEUKvVsije5OQkV5N0d3ejq6sLNpsNg4ODGBsbw+9///sdzZF6JCmVSjidTgBgeX0A3H6F9I3a2tq4L9o999yDsbExvPrqqzt2Lcrlcm6n4na7eRPVaDSo15f7CIXDYb58eDwedHd3w+l04vjx4xgbG8Mrr7yyo9ciVVmKooiWlhb2hhBH0r2Zmppi0b+enh64XC488MADuHz58oY5bvc8FQSBdWLcbjfnx5AcBelwXblyhVXke3t70dLSgoceemhXcASWzw65XM6J3FQMVKvV2As9OjqKbDYLpVKJ3t5eeL1ePPLIIxgbG8NLL73UVI7N5kfGnNvthkKhYM2iPz4LdDodRkZGkMlkoFKp0N7ejvb2djz22GMYGxvDiy++iGAwuJG33FqRw42CBKgAcKNJqi4pFArw+/3weDyoVCoIh8M4e/YsLly4AI1Gg+9///u31M11O9DYPJKaTFosFsjlcm7e5/F4UCqVEA6HMTIygrGxMWi1Wnzve98DcM3Vu1NBvcjoVkYcqWNxMBhktdxwOIyxsTFMTk5Cp9PB7Xav0OXYqSgWi5y/k8vlYDAYYLVaAVxrB9Da2so95qampuDz+aBWq+H1egFgx3Msl8uIRCIAwKE9umEC4Ma31EvH5/MhGAxCq9Wivb19x49jtVpl7zIdijabjcMJqVQKbreb+y1RuxOdToeuri4AO38tNura5HI57nxOfRJJVd3v9yObzSIWi2FoaAg6nQ69vb0Adj7Her3OkiHE0Wq1chuJWq3GzUSz2Syy2SwuX76M+fl59Pf3A9j5HAGw958KSaxWKwqFAnvZHQ7HCoHDyclJBINBDAwMANj5HMnzPz09ze10GgU6SVS2Xq+zqGwsFsOhQ4cAXJMDuFXc1movMn5IbdVoNHLPFqPRyG71RnGuWq2GU6dOoVwub1n/mWaBnt/v93OpO/VHstlsrC+yFj/Kf9rpoAMyGAwiFApxA0yn08ld4UnvhyqDarUaPvjgg13DkUID0WgUsVgMCwsLzNHpdLI4F3Gkn3/vvfdY3Xm3IJlMct8cKuF3Op0s4EjNRenr3XffRb1eX9F0cCeDxFTJeHM4HHA6nZzH1TiGu5EfsBzCo+ad5LGkRqaUI9Po8X/nnXcAYFdxLJVKrLtF5dHUioQ6nJMcSq1Ww9tvvw1BEK7bw28ngpokk6iozWZjjpQnA1zTGHvrrbe4x95uQS6XQy6X48o2p9OJcDi8gh+txTfffLOp/G5b2Gs9sFgsrDxKaBQH3KjWz3a7926GxqZ8hEZ+m4jHb6urfT0wmUxrqj5vluPtcEPfDFQZRVzImL2TOJJA4HZx3G5+Op1uzYaad9Ja1Gq1a3YSJy/6nTBPr1dMcSdxvF6hwVZx3G5+10vC3yw/7LSw13qQzWbXNHB2S5LezZDNZgHgjuUHXHPh3ukcVxt3wJ3FcfWheadxvF4S/p3CD7g+x90iGLse3A0cr1dMcadwJFHg1Wg2vx1t/FBY7E7FbqiWuVXcDRx3S1XQrWA3qOPeCu50fsCdZchdD3fC4X8zSBybg52dGSVBggQJEiRIkNBkSMaPBAkSJEiQIOGuwkbDXhEAc1vxINuAjnX8zG7mB9z5HNfDD5A47nTc6fMUuPM5SvP0Gu50jruZH3Adjhuq9pIgQYIECRIkSNjtkMJeEiRIkCBBgoS7CpLxI0GCBAkSJEi4qyAZPxIkSJAgQYKEuwqS8SNBggQJEiRIuKsgGT8SJEiQIEGChLsKkvEjQYIECRIkSLirIBk/EiRIkCBBgoS7CpLxI0GCBAkSJEi4qyAZPxIkSJAgQYKEuwqS8SNBggQJEiRIuKuwod5egiA0rReGIAg3/P5WtN2o1+s3fNPdzg9ApF6vO270A7ud483GEJA4bhQ7keNu5wdpLQKQOG4UN+MINJ/nTlqLf3yeZr0dYc21uNHGpk2BIAhQKpUrPgj6e61WY/L1eh31eh21Wg21Wu12POqmsF5+xKmJ/Lat+dxt5LhtkDju/rUok8mgUCiYUyPPLea3bWvxZhyBa/x24xgCdzdHQRB4rq7+qlart/ORNwTiJ5PJUK/X1+QHYAXXJvFbcy1uu/EjCALkcjnkcjlqtRoEQeAv+j5NXEEQUK/X+cPaDU1YN8JPJpPxn7uFHyBxpO/fTRx361qUyWR3LD9gbY4ymWzF9+kwuRM40jPLZLI1D0/6v93KURRFNtoaOQJgY2D1RWU3cKS9hvgR39VjuJ38ttX4EQQBCoWCrVtRFPkDqNVqkMvlqFQqqNfr/Cd9jz6QnTzQ1+MHgDedarW6a/kBEsc7neOdvhbvFH7ANY5KpRIA1uR4p8xT4iiXy9mAvR5HMoQaPUI7Gas5KpXK6+43ZAys5rmTOTbyW2s/bVyL28lv240fumk23sJW31ToA6pUKgCuWfG7YZDphrKa3+qb51r8doMLU+J453C8lbW40zmulx8ZC+VyGcDu2WuAaxyJA3Gjfzd6QURRRLlc5p+7kzg2ztNyubzCE7ubOa7lsWxci43/v5M5rrUWgWv86Ge2m9+2Gj+0CEVRRL1e51tZpVJZsVDpwwKAarX6iQmwUwf6VvnthsW6UY6N8XeJ484BGXd36lzdyBg2GnS7Za8BbswRuOY1uJM5Ap8Mi1Wr1U9cVHYDR4VCgVqtxh4SMlaB3c2R9pmN7KfbwW9bjR+lUgm1Wg2FQsEWHrm9ALDVrlKpIIoiCoUCyuXyCpcYgB276SoUCmg0mg3xK5VKK1yawM7lB0gcgTuDo1Kp3DDH3bQWb7bXVCqVNflVq1UUi8Udzw9YyVGhUKBcLn+Co1wuh1KphEKhQD6f3/UcK5UKh7uAT3JsXIvk/aHf2ekclUolz9VarQZRXD6ey+Uy5HL5mvvNbuC41n7aOIaNew3N08b9lMJ+9O9mYVuNH1EUodPpoFQqeTNqjPmpVComS5nhdKOmRKnVSWw7abAVCsWa/GjwrsevXq/vCn5A8znuNH7A3TGOd/paJH4qlYqNgkZ+SqVyV/MDPsmRnrlxDOmQkcvldwxH8hBUq1UOnTRypEOSODYmSgM7jyPtNyqVasXlgjjSOJLnp5Ej/Z0MCRrPncRRoVBAr9dDqVTyWlzNj+Zt4zylMSSuq79uFdtq/CgUCqjV6k/E+lY80B+ToehPyi2gD6Mxi38nDvJa/ARBYB6N/ADsKn7A1nDcaSWpd/M4NuJOWIu0eQLX8nkaf+ZG/OjQ3In8gGseg/VwpJs1jTX9uVs5AuCwSSNHpVLJFUP082vN1Z2E1XOVnhdYm6NKpVrBsTGHi8JDO8kLRMnOjR7Y1fxEUeQ9lvgBK8vem82v6cYPbaCrH4xI0YdQKBTY8iuVSmylk2uPPijKgG+cyI3xv8Zb63YMdjP5UUyT+BGH28mvkY/EcXdz3Mq1uBM4rpdfPp9nfhRCIH6Uj0CeEuJHr0vGbb1e33FjqFQqN8yxcQxXh1d2M8fGedp4Qfn/7Z1rVxvX1cf/o/sdEAhhY+zEdmLHdtJ2raftuzYfovmubVdfti+by0qzVnNzAAcQICSNNNJopJHmeUH/mzOjkZDQSAg8e60sA1aMftqXs885e+9ZNUYvJxM26snLeJ0v8rrzthkn8SWTSXn/5BsOhxNjjeqLZIjH4/J1EHyBJT/eXaPf9wyuBOfPufvgIjIYDMToe72eZIlq4R6BaSiURSl6GXx+uxN1YV0kX8gYMs7LeBd8ke9bPem5LtaQZVl8QTACEEb+jFec3iJastxVRse5LJBVa5+8elRZeHKwjFZ4L5P3Z2S0LAuxWMzXVrnBWEXGaXRo2zYsyxJdevm8OuT136RYEwRfIMnPdQoGrqq5TdOUD4SBVp1qydfxiAu4zOB5pwnA9SHw+0UuLIvm412nV7H8e+5eFsU3LSOPJW/K6E1cV5HxfdBj0L6o/v1d9kVegd1mrPHj8fvZLIwApGYEGNUh/55C5kUunEEwqle2fJ9eX1RFvTZRE3z+v0FLEDF1HKOmaaJH9b0vk9GPz/vzm/Dxey+f1xfZ8XdTvrmSn0nwVJLaasngr3YlpFIp17/De00K73DT6bS0/rHKncdf/H2c1aFmhHeFT2WZxOcNTHeZkTvM+8x4H/WYSqWWzjgLH08v2N10HZ+a2Gma5oo1PP1adKxZBCP5+P84jiPXtGRQGYFRX1wmI2VaRiZ0LASmqI9QUBdIMnIjRvtfBuO4xO6mjNfZqsqoDn9clC8GzccEPp1O++qQdjoP38Jqfpjx0WFVh2NQ9H5g6hUB7wJ5HJ1IJCRLHA6HiMfj6PV6I62rfv/uqvLxSqHX68mdJ/k4sMuPb9EFwioj/7wJI18/jpHHt7fNOI8er2NcFT0G6YurwjiJj+9tGj7GFXajAFcD2CzLmmini4w18zKyGNaPUY2nvV7PdZXC1yybUeWclZELLeMKAKmjcRzHpcdIJCJt897xFIvmC5JRTdon2SoThkUxBsmnrosUdrRZluW6nlZHH6inz9NK5PqXTCcEVr/2XnPw7s+yLDmSAyCteoTneG86bTqdRiwWk6BLA4jFYrAsS16nBoig5SZ8NMBxfDyaZdfNbfKNY1S7QWZlJIvKqP58OLxsQ71txnn0OA3jKuhxXlud5Is8UVi2L3oTgiD42HWjFmDepp36Ja1BxBsyqlcPZOT/xwVpmToMgjEWi4mdkjUWi4mdcvHk6ciy14ybMiaTSdmMqGzXxVT1xHcV+Pj3g8FATrDIR/1FIhHh43wg+iJ1SHtWf+8scuOTHz9w9e/U41bCqT/PZrOy+ywWiwCAVColC0gkEpHiKBZB8YiLx/aDwQD5fF4UzUKpIAYizcrnVcIsfDRkHsNGIhGp4M/lcsLHa4egBj6tCqOmaRJ07ysj9bgKjPP4IoOU6otkzOfz6Ha7t+6LXr5cLic68fLxteNijWEYsruknQbNtyxGnmrRTtnxpWnaCCPtdJGM3pMlsnARm4eRenQcRxbKXq8nv6/b7d4aI2OMH+M4X0yn07LYe+MNXx+NRmWIpWqrarzhFVGQfKp4Nwfj4qlt2ygUCsJHHfrFU+ByEGK73Zbid+rQ64uz8M197eX9MPgnMz8GT2be/CBs25bjVra1pdNpAHD1+VN5bIEzTROdTsf1ITEYqG10fA9B8/HfJJ+62+BryNfv9335aIzcrah8nU5HONRn8fB73qEGxXebjJqmubpRotGo6Pa+MPrpsdPpLEWPKut1vngTRvoiGYGr4Ke27Abti97EYBIf35sfH4/OvbGGVwemacI0TVlw1cRjEbFmWkbaqR8jk9NxjOrJJHVomqYsoOpCtWhGfu1lJNO8jOo1F8c5WJYFACO+uExGnl54GVVfJKOmaUilUmKbbHMnMxkty3Kti6oeuUlZhC96f+bl48DC6/i8r9U0TU5fbdtGt9uFaZqy2Qoins5d8MzdCOBus1PfBLNfgjHx0TRNjic5AVLTNLkfpGHzGNOyLESjURiGIW1xlmXJ9M9EIuEa2z6v+PGp/7ZqtAB8+ZjR5vN5uYvmboRGzAyX/0+73YZt2yN8yWRSsvmgJGS8v4wq2zS+OC0jfVHTNDkZGeeLQTHelI8Lgx+fN9YwsWARKWONbdtyrbeoWDMNI1/Dr2dhBCCLDe3Uy7joeLpIRtopr4ZUPcZiMZceB4PBiC8uklHlVF/jZZzki7Rr4Oq0iHqk7lRGfhb03UX5opq0+vFNijXq5GevDpnwZjKZER36xVOeAs0ic117MbvzZoHeORGq0riL4v3e1tYWnj17JjUvagV+t9tFPp9HIpFAqVSCYRg4OTlBo9EAAFiWBdu2pRpcPQLjqdAq8XGkvsqXy+WQTCZRKpXQarVwcnKCZrMpf89MmXzM/uflm5WRVfmzMpqmKTrc3t52MTqOIzq8y4x3SY9B+WKr1UKlUhFGP18MinEaPm8QVvm4c75JrNF1Xfg4eI58QcWaaRjVhVK9FhrHyFMBdsgMh0N0u10UCoURX2Q89bPTRTFy8WTSo9rprIysH3Ecx+WLKmO9XhdGrx4X5YveBOimjIlEwtXpxJjqZVT1SEYmKkwUIhH3vLGb8jEBm8TH2iUmnI7juPiePn0qfNShl69cLqPZbKJSqUxtp9PyzXXyo97vqXfSqtLVgq1UKoV2uy1ZXzqdxt7eHorFIpLJJNrtNrLZLBqNBmKxGHZ3d6UlLpfLyfHlTz/95DJW/k7e+QUl0/Dx59PyZTIZNBoNxONx4QOAXC4H4FKxP//8sxzxeo9E1UB4FxgfPnwoVyheRhowGRkE7xrjfdajrusuXyRjJBKBZVn46aef0O/3XQsHfTFIRpVPTRDUHSi/9vLxEQnjYg11qNqpX6xRdbhoO1V1Sb5ZGTudjstOHz165IqnmqaNtVOeGAXNyNM1r40yEVC/Z3G2ypjJZEYYvXr0Y2QNl9cXVT0GxcnrrGl8cRLj5uamS4+qL9JW8/m8y1Z7vZ5rXVT1qH7G81x9+RUaT7JT+pvjOL58pmmOxFOVjzpkQsfC6XG+OC3f3MkPuwXU7ggeWzHQspApGo3KMV40GkWpVEK5XMbm5qa07+VyOTx9+hSJRAKNRgPNZlOOurLZLFKpFHK5HAzDcBlTLBaTTJcZ4byivn8/PgByf06+QqEgmf84Pmb0uq67+Mil8vFYV/0v6ARvVsZZdDgLI3A1v+EuMN51PU5rq7P6ouqDQTHyM4zFrrpb1IJIv1jjx7e1tSU1TNls1mWnuq7LdXoul0Or1UIul5NaCpUx6FgzLSNtclpGrw51XRf9ZrNZpNNpZDIZGIYBTdNcvkhG+mQQwt9B22Si4GVk8uDHuLOzI4zJZFL0mEwmxU7ZSaTqsd1uj5ysBW2nZFD9T9UpcJUcTcPI8QTUo5eRdpxOp4XRm0zyqoms89b8qDpSYw2vxlUd83tvPCWf4zhIJpOyZqRSKTQaDbRaLZkITR3m83kYhjFyqubV4VJrfmi8KixwNXmRR3zxeBxbW1vyut/97neIRCI4PT3Fy5cv8cc//hGtVgv7+/s4Pj6GbdtIp9Oo1+s4Pj6Gpmmo1WrY29tDLBaTD8gwDNl9BlWwRj7VOWOxq2JYVVS+UqkkmfFvf/vba/mo7JOTE2iahouLCxefbduuKvegRWWko07LOE6HzWYTBwcHI4zHx8eIRCITGYM4el4G4yrr0c8Xg2CkL5LR64tk5O4saD51p0w+lWsavvPzc3z88ccuvpOTE/T7fRdfJBJBtVrFkydPlhJrrmPkdYV6PTuJ8cWLF/jDH/7gYhwMBrLDPj4+RjQaFcZ4PC46VBmB0cm684h6+sjnr6mnIRTu7P3WjGg0irOzM7x48ULizeHhIY6OjmQIp67rqFQqosfHjx8jGo0KY6fTQa/XEzsNktEbb8jo54/A5abEj1HVo2EYLl9MJpNoNpsSU8/Pz/H48WNEIhG0Wi0Mh0MYhiF2qtZWBcFHO+WjYfi9Kvxd4/iq1ar4oh+fruvCV61Wsbe3J3xeX+Tvm4XvxskPf5HqkKzyJvBwOEQ6nXadzrDyfGNjA6VSSXYXm5ubcBwH9XodT548QaFQwOnpKSzLwubmpswJKBaLOD09RSqVQqfTEYdut9tSlBiEqHxqWy+zd3LyDlPl0zTNxQdA+Gq12ghfsVgUR1xfX8f5+bnw2baNTCYj7YGGYQTC58fIuRJ+jNwtTcPop8NpGdnSuOqMq6pHP1/kn36+6GWkLwIY64vdbteXUfVFFmAumo8L6bhYo/Lxey/f2toaKpXKSKxZX1/H2dnZCF+n0wk8wfNjdBzHlSwA1+uQ37NNWmU8PT1Ft9sdYVTjqW3byGazksQu0hdZ58GCXtbv8MqD3N41Q9Uj483jx4+Rz+eFcWtry5fRNE3xRZ6ALTKmqt1PjC0AXIzRaHSEkdc41GOtVsPjx49dvlgqlSTebGxsoFKpIJPJwDRN9Ho9YVzUushEw48vk8kAuEoEyVcsFkf4HMeRzSJ1aFkWtra2XPG0UqkgnU7DNE0MBpdPfXCcy9lN3W53ZpbAkh/eKTJrAyBHkjw6ZTtbNpvFw4cPUSwWoWkadnd38fDhQzSbTWmvzGazKBaLSCQSKJfL6PV6aDabODo6QrvdxubmJgaDAS4uLlzHXZGIe8DSvHx0TrKSj87Ke2XyxWKxsXytVgvdbhetVguZTEb4tre30e/3ZUfW6XSwsbEhfGr2zkQriAx+EYzNZnOEMR6Po1wuT8UI4N4zLkqP43yRjF5f9GN89OgRHjx4AF3XfX2RttpsNvHrr7/CNE0Xo+qL0Wg0UF+cFGt4BcfjfT++SCSC3d3dET4/X/TGmuFwKHyqnQZ1YqAyAnAl6uqGUmXkdUM2m8Xu7u5UjLTTXq+HVqsljMViUXTImopFMDIpAK6G+Hk3zdlsFpFIRBhpuyrjo0ePsLOzM5HRtm3Ro2EYoke1XpR2qi7mQTAy3pBRraXidZyqx3g8jkwmg93dXWxubrr02Gg0xjL2+33Ro2EYosdqteoqQF7EughcTTtXa4qoQ/491/1MJoOHDx9ic3MT0Wj0Wjvd2dkRvuPjYxffxcWFi+8mOpzr2ku9C+axLIOfOiAtkUhId0U+n8ebN2/w5z//GZlMBg8ePEAul8N//vMf1Go1UeiXX36Jjz76CB9++CF2d3dRq9VkxkG5XMb5+fklwP92fVzY4vE4TNOcB2uEj22CvEf24+PQMJUvm81iZ2cH+Xwe3377LS4uLiTwfPXVV3j+/PkIXzKZxM7ODqrVqosvkUjIjvomWW4QjH469DJShyrj06dPhZHdRdMwBnUMvWzGWfQYNCOAQGx1Gl9kN4afL3JBC8pWp4k1mqaJDr18jDX5fH7qWOOnQy5q1GFQscbLqGmaNHAMh8MbM3IBUePNo0ePJJHzi6c8nVwWI+0lk8m4GDljiox/+tOfkM1mRxgfPHjgYnz27BkePnyIi4sLadn36pHFt7wxWERMBdy2yk0PT0m8jK9fv3b5YqFQwHfffYd6vY6dnR3Ytu2KN2Qcp0cAUmwcpB55QsxZbUz2mEiqSRD5CoUCXr165eJbW1vDd999JzpU+VQdsn2/XC77xlPDMBCLxWbmmyv5YXseM1sGI7a8MstnQFRb2dRukq2tLRSLRdRqNVdL5v7+PgaDAfb39+UIjG2Me3t76Ha70sLIYrEgpnSO4+NR3Tg+8pumCcMwJvLl8/kRPsuykM1ml8bnZeSObBKjqsObMPZ6PRdjp9MRRtZKkXHergQ/xmn0OK+dzqLH22JUbXUaxl9++WUmRh61L8oXvbGG7dzT6vDi4gKOc9kavba2dq2dmqaJWq0mfIv0RS7O6u7ae+VHRi5CakeQV4emaY6Np14dNhoNSVzJGKT4MapXzeoJGJNNMpqmiWw2O5Uvvn37doTx0aNHwsjrEl7JBK1HLthqkk5b9V7dkrHb7aLb7SKbzaLZbKJUKmFjY8OX0bZtvH37FqVSyeWLu7u7ME1zhHGR6yL5xumQ6wqvy8fxWZaFtbU1HBwcwLZt/Pzzz7586poxjw7nTn64y2QXCJUUiUTQ6XQAXLW7p9NpVCoVlMtlXFxcoNFooFAowDRNrK+vY3t7G4lEAm/fvsVwOJSjdQ6OYxHYxcWFtKpubGyg3W6j2WzK4K6gOhPG8XHS5HV8uq67+Eqlkuwah8MhWq2W8HGAY7VadfGtr6+j0+mg2WzKE26D4vMy8ujSj5G1B0Ez5nK5pTJO0qMfY7VaHbFTlXEwGEylR9VOl8k4jS9ex0g9vnv3TnyRRYjL9kUuJqwpHMeXyWSmijUcmDop1hiGIXydTge6rrueph2UeHWoMqrFpLw+IeP29jaq1Srq9ToKhQIsy0KhUECpVBJGry+S0S/etNttF2OQMhwOZTEm42AwQKfTcXVheRm9eiQj9fjLL7/ImqHaKfVYrVYlCSYj7TToJgs/RsYbcqnXYNls1sVIPfZ6PeTzeWxvbyMej0tip9oqkxAysvV/fX0dhmGg1WrJ6xbNRx+ksDA6nU7j9PTUxZfP59Hv90WH5LNtWxJwla9arYoOvbGGwxxn9cW5kx8GcS6czAQjkYjMLQAAwzCQTqexvb0tgTUWi+Hly5fodrvyofzwww84OzuD41xOe2TGf35+jng8jk6ng4ODA0Qil2O8+YFGIhF5XktQ97eT+LwdGXSkUqmEra0tF59pmhKcfvjhB5yfnwufZVnodruoVqsjfJZloVwuo1arLYRvGka1PXpWRu4QuKPxY6QOV5Uxkbh8+jUZVTtV9TiJ0bIsbG9v3xojfZGMhmEglUqhXC7LAun1xUajMeKL1+lxkYwqH0+uWPejxhpN09BqtZBOp1Eul2VxjEajePXqFTqdzogOh8Ph2Fizv78/Ems0TUOlUlmoDv0Y2VasaZqcupbLZZTLZantImOtVpuK0TTNEcZqtbpQRi7ETHz8GGmn2WwW5XIZOzs70lzix0g7tSzLxZhIJCbaKccYLJrR64scKWAYBvL5PMrlssxE0zQNr169gmmaqNfrqNVqI77Y6XTQ6XRwdnYmjPv7+4hGoxJv+Dm22+3AfdHLx1Nmb6xpt9uSwO3t7SGTyUDTNLx+/VpOU8lHO+WjOvz4WAum+iIT56XW/AAQpapXJmqx3Pr6OjRNw0cffYRUKoVCoYC9vT2sra1hc3MT2WwW//jHP9Dv93F+fi73hbyD1XVdBiMVCgUT9FxOAAAQM0lEQVQYhiGOzqmV3BUxQAd5jHkdHwtFnz17hlQqhfX1dezu7k7FF4lEoOu6FIOtra25+Dh5dJF81zHGYrEbM7LDoNls3knGvb09FAqFqex01RlVW/3oo4+QSCSwvr6OR48eTe2LKqPXF5fBOI6P1yYbGxvCl0wmhY86TKfT+Pvf/45+vy8LPK++ruPzizWcar1IHVIHfox+dupl9Iunfoyc1cYrqWUwst7Hq0fa6ccffzzCWCwWkU6n8be//Q22bYseAUjNh8qYz+fRarVG9MiEh5uzZTNSjy9evEA6nb6WUY2pw+FQGDnfhycg9MVFM6p2ylof4GqOEwvTX758iVQqhWKxOML317/+FbZtSxLDU07yxWIxZDIZZDIZNJtN2cRwmjM3eDxA4ADLaSWQ5Ie7CrWYCwA6nQ5arRaePXuGzz77DAcHB/Ksju+//x6tVksyuFgshu3tbfz666/STcO2Umb8HMjFdrdoNCqTH237anYD7yCDEC8fj9W9fJ9++in29/eRTqcxGAzw/fffwzAMOdWIRqMuvl6vh3Q6LUex5OPAKvKxrW9RfDdltG37WkbeX09ijEQiwtjv91eKkXY6L6Oqx1Vi9PPFVbbVcbHGcRwYhoFms4nnz5/j008/xcHBgTz3iDrkjj8Wi6FUKrn42OrNkQfTxBomBkGKyqh2ejmOI8f8z58/l3iq2mmr1Zqa0RtPuRnjIrmseMM6P9ppu91Gq9XC8+fP8ebNG7x7987FyFoRL6Ou6zPpkadsnL21DD3SVtvtNgzDED0eHh5KTP3vf/8LXdevZWy323JKpPqiqsdl+yLFMIyJfI1GA1tbW6jVajLHSeVLJBIwTdPFp9ppNBqV0RPz8M2d/PCDME1TjhxZ+wBA7idPT09hmqbMmzAMAxsbG/jggw/w5s0bnJ2doV6vS+tqJBLB4eEhbNtGLpfDb37zGyQSCRSLRfT7fRlM1mq1xJgYkLhrCkr8+NSd1Lt371CpVKSwkHzFYhEffPABXr9+PcIHAIeHh+j3+8jn8/jss89cz5ghn2EYwtdutxfCFzQjrzqpw2kZ+V/IuBhGry/m83lUKhW0223xRTI2Gg1hBCBFiPl8Hp9++qnrWTrLtFUvHx+yqvJRh36xZhzfqsQaL2MikXA9SJZ1O2o8rVQqMAwD6+vrwnh+fu6yUy/jZ599Jh1j/X4flUpF2t87nY4wMlFYBqPqi2TsdDpYW1vDyckJWq2W+OLOzs6IL07DSD1ygV4WY6fTuVaPbOfe2NjAhx9+KIy0VSYcBwcHGAwGyGaz4os84alUKrBt22Wr7XZ7IYyO48jnx4J76pDt92dnZ+h0OlhfX5d2fPI9ePAAp6enI3zv3r0THb5+/VrWTNu2hY/6m4cvkOSHwkwQuBpupI4U5zRGZuGlUkm6LjY2NqRDhMOOvv32W5ydneHp06d48OCBTLXMZDJSDMX7z263u5Cpq9fxqZXnPCamEZRKJVSrVV++QqEgfDSERCIxkW/WY71VYTw9Pb23jGy5vQuMfr6YzWZhmia2trZGfLHb7foyPnz48NYZVT4AcvLEjiy+Hw59ow6Hw+FEvmljzTJ1yFMR2mmtVvONp6oO19fXJ9opdchZVSz8ZV3QbTGyZfni4gIAJMHmLYCqx2KxKIyPHz9GLpcbianUIx/Hws/sthi9vghcxRtVj5zTQ8Zut4snT54IoxpvkskkWq2WMLZarRFbXTRfv9+Xa8dYLCa1dcCVDjlkcnNzcyxfNpsdsdNkMin1XyqfqsObSKDJD+AuhlLvAjkCm9mZOoeER4+sZGc9wueff46Liwu8efMG8Xgc7969w8nJCX7++Wecn59LQZSu6+h0OoG2Ds/C5ziO3LdS2HGjadpMfMfHx3j79q1kzMvmCxnvLyPbi8nI+3rWJQDX++Lr16+RSCRweHiIk5OTW2X00+FwOJQaFv6cbcdeHbZaLcTj8RE+NdaM4xsOh0vTobqpY+swGflzNZ5y7D+HNd5FRj6mQh2UqfoiGW3bFsa1tTV8/vnnqNVqePXqFeLxOH799VecnJzgp59+cq0ZjUbDxQgEM3B0GkbHcaQbk80GlEmMqi/WajV88sknwlipVISx2+3i7OwMzWZTujWB+R9oOkl4ukahDtVYw9epfLyu8vpirVbDixcvkEgkcHR05NJht9vF6enp3HyBJz8EZGDKZrOw7ctnH9Fp4/E4PvzwQ+RyOdRqNWn1S6VS+Oabb+QI8IsvvkA6ncYPP/yAL7/8EicnJzg+PkalUpG2Omab/LfVuSaLEi8fq92Hw8v5G4lEAk+fPkUmk5mZj4xePhrSMvimYYzH4/eecRY9plKpO8PIYETG7e3tqXyRjF999RVOTk5wdHS0cr7IWMNFZFysSSaT+Prrr8fqsFKprASfHyMX7MFgIIPkPvjgA+TzeWH88ccfxzJ+//33osNViKdexlwu51osOdTRa6dk/Oabb1AoFMba6TSMHPWxTEZ1jhMZS6WSMGYymRHGo6MjfPHFF0gmk8JYqVSEkZ1dt8HI3+HHl0wmpYtW5Ts+PhY7Jd9f/vIXJJNJ/Pjjj/j6668lno7j44nerHwLSX6Aq4Ioy7IQj8dRr9eRy+Vkrks0GpU5Bjwe63a7Mq0xEongn//8J3q9Hmq1Gvb399FqtSTbI6zjXD0qQN2xL1r8+PL5PPL5vPBxjgGnUvJulHz/+te/YFmWLx+DDvk4M2GZMisjO2dUHd5lxrW1NRmglc/nXXqcltFrp6vCmMvlUCgUpvbFWRiB1fBFNdYwIKt2elNfXDafl5FXPWTkE8Gpw3GMjKcXFxc4ODhYWUYWtaqMa2tr0qqu+qJlWXKFREYO3PQyTrLTRZ9u+TFyyns+n5fuO5WRvjiOsVar4fDwEK1WC41GYyzjIgY5TuIbDAbodrtyhanaqaZpU/FZloWLiwvh03V9hI+ngoyrs/ItLPkBrgqieN83HA7lOJ0j9tPptNw1D4dDyfo5rpyTL1lQaRiGFI3xv+FwKAGOx4rLEFbuk4+G3Wg0kMlkMBhcPgix1+uh3+9jOBzKw+Y4Vp98ahGXWqtxm3yTGHVdH8uo6nAc46rocFpG2uldZvTzRdVWJ/niqjP68Q2HQxkq6cc3zhfVWOPni95JtssS2qllWS4dcpCdbdvSGcVFwsvITqC7wsjFVNf1kTVjHkZeA98mY6/XQzKZlHjDdnUvI/XKq7N///vfUuNGX6Tt+zHyMTrL9EXGBvJRhyofuyW9fNQh+drtti+fGmt4hT+LLDT5ASDtZ6ZpwnEcNBoNNBoNuZOtVCqoVqvSgsi7emb2/ACAyyfb6rouR6KsK2AQCHKc/rSijl93nMsnDJMvFovh5OQEFxcXI3wsTFx1vpDx/jDe1BfvCqOXr16vo16vj+iQi+OssYbTlzmmfxV0WK/Xoes6PvnkE19G7orV06C7xsj36cfIcRleRk7fnsS4Cr7IRbter6PZbOLly5eIRqM4Pj52MbJRgYyWZY0wkmM4dE9DZ7J/m3y1Wg2NRgMvX74UHXJ+UafTcfnisvgWnvwwq02n03Kkt7GxITtD3tfxmJ0PKOWxJz88y7LkzpQ/Z4Ejf8dtGLGXz7IsaZFNpVKSwQ+HQxiGIaPA+dpV5wsZ7y9jr9fD+vr6tb54VxiniTUsdp4m1vDrVeEbx1gsFqXraxpG7rDvCiPHSNAXVcZWqyWF/EzaeZqj6pGvVxmXfeLjx8i5WaqtXseoDp70s1sv47ITHz8+y7Kwubkp9WPkY7Ez37PKxzk+Xj4WU8/Lt/DkhxMtWXjFrLvRaMC2baytraHb7ULXdXlwm1odnkwm0el0XAVbaiHVbRuxl49fk4+7i7vKR0ZvJ819Z7yvevT6ospYKBTulS/OG2v456rwXcc4GAxEh81mU+ZVeRnVReUuMLK4u1arod/vu/Tox5hOp0WPKqPjOCvJmMlkAFyeaPBJ9KovkpEnq15G1VZXhZGDD9UESNUhn7On67o8uNRPhxzWyP8cx3Fdqc/Dt7DkRy3sHA4vC7xyuZzc03JX0m63BaharaLb7WJzcxPtdhu///3vxQCOjo6k4r9Wq8kQp0VX6M/Lx7bfafjYebMKfF5GHifncjlxtPvMeF/1eB9tlXzqiHwvH6fGcnGYJtZw+vxt801iZM1Eu90WRgDX6pBdNtvb2yvJSF/M5/OwbRv9fh/tdhvJZFL0qDJubGzANE1hbDabODo6QiqVctkpH2S7KoymaYoeWdPCeTizMKq2epuM3nhKPs4ychzH1047nY4v38nJycL4Fpr8aJomWWgsFkOr1ZLplltbW7KImqaJRCKBbDaLWCwm2W+j0ZChZQBcj87gn7cls/LF4/Gp+NRn+dy2hIzvJ+M0vhiPx1eGkXwsfBzHZ9u26DCXy43l466cXMvuevKT6xg3NzelOLbT6VzLCFzaKWstVpWx2WwCgDDylMOrx3q9PtEXWZN2m6I2Bnj1CEBsle3ifoy9Xm+iHm+TcRKf6oteHWYyGV8+sizKTheS/PBN836ZRZAMvhwNrx7d5fN5pNNpuf/rdrs4ODiQTFLXdQCQ6vZFjZYP+ULGkPHuME7Dpz7zaTAYIJfLyaRY8h0eHgK4DLSNRmNl+PwYHcfx1eEsjM1mU66EFvX4g0UxcufPFnFd112MvMIm43A4XAlGih/jcDgcYaQvklHTNFiWJYxMLFh/ZxjGrTKqXVh+fPF4fGY+6pAnuEH64sJb3dWvGUw0TYOu6/IEWgCupwoPh1dDvHK5HHRdh2EYACBdCKtgxNPycbd91/iAkDFkvBuM1/FFo1GZYh2NRl2D0chXKBRQr9dXkg+4YuSffozU4STGRqMhpw3vAyOf+r0qjJR5GNfW1kb0eNuM3vqbefl0XZeTv0XwLST5UT8E9RiOmSB3kMwQI5GIVPWrfx+JRGQMOXc0t12oRg7KtHw8yhzHxzHdq8BHFkrIGDKuKuMsfIPBYGysiUajqNVqKxdryEKZl7Fer8vclPvAqBZsq4z3QY8cTjiOkQ9JXhXGoPh4BcZYsyi+pbS6+/2MGZz3g3Gcq4mNPObyPoDutpWsSlB86gTOVeIDQkZ+HTKuNuNN+Dg/pdVqrXysAaZnVItq7yIj37v6Mz9GXvmQkVc/IePtyjx8y7JTbZZ/UNO0hX66aoFhLBYLHNxxnInVYHedD8C/Hcf5v2vew51mvE6H/3sPIeOcctuMd50PoS/yPYSMc8ptM951PozxxYWf/MwiBFXnh9wnue98QMh4X+S+M953PiBkvC9y3xlvi2/W5KcK4GARb2QJ8mSK19xlPuD+M07DB4SMqy733U6B+88Y2umV3HfGu8wHjGGc6dorlFBCCSWUUEIJ5a7L7U+2CiWUUEIJJZRQQlmihMlPKKGEEkoooYTyXkmY/IQSSiihhBJKKO+VhMlPKKGEEkoooYTyXkmY/IQSSiihhBJKKO+VhMlPKKGEEkoooYTyXkmY/IQSSiihhBJKKO+VhMlPKKGEEkoooYTyXkmY/IQSSiihhBJKKO+V/D8nP0q8qgDfOwAAAABJRU5ErkJggg==\n",
+ "text/plain": [
+ "<Figure size 720x144 with 20 Axes>"
+ ]
+ },
+ "metadata": {
+ "needs_background": "light"
+ },
+ "output_type": "display_data"
+ }
+ ],
+ "source": [
+ "n = 10 # Number of frames to display\n",
+ "\n",
+ "plt.figure(figsize = (10, 2))\n",
+ "\n",
+ "for i in range(n):\n",
+ " \n",
+ " ax = plt.subplot(2, n, i + 1)\n",
+ " plt.imshow(vali_data[i].reshape(64, 64))\n",
+ " plt.gray()\n",
+ " ax.get_xaxis().set_visible(False)\n",
+ " ax.get_yaxis().set_visible(False)\n",
+ " \n",
+ " ax = plt.subplot(2, n, i + 1 + n)\n",
+ " plt.imshow(decoded_imgs[i].reshape(64, 64))\n",
+ " plt.gray()\n",
+ " ax.get_xaxis().set_visible(False)\n",
+ " ax.get_yaxis().set_visible(False)\n",
+ "\n",
+ "plt.show()"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 19,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "/home/jon/PycharmProjects/TensorPlay/venv/lib/python3.6/site-packages/ipykernel_launcher.py:5: DeprecationWarning: `toimage` is deprecated!\n",
+ "`toimage` is deprecated in SciPy 1.0.0, and will be removed in 1.2.0.\n",
+ "Use Pillow's ``Image.fromarray`` directly instead.\n",
+ " \"\"\"\n",
+ "/home/jon/PycharmProjects/TensorPlay/venv/lib/python3.6/site-packages/ipykernel_launcher.py:6: DeprecationWarning: `toimage` is deprecated!\n",
+ "`toimage` is deprecated in SciPy 1.0.0, and will be removed in 1.2.0.\n",
+ "Use Pillow's ``Image.fromarray`` directly instead.\n",
+ " \n"
+ ]
+ }
+ ],
+ "source": [
+ "out_dir = \"%s/test_simple\" % base_path\n",
+ "if not os.path.exists(out_dir): os.makedirs(out_dir)\n",
+ "\n",
+ "for i in range(len(vali_data)):\n",
+ " scipy.misc.toimage(np.reshape(vali_data[i], [64, 64])).save(\"%s/in_%d.png\" % (out_dir, i))\n",
+ " scipy.misc.toimage(np.reshape(decoded_imgs[i], [64, 64]),).save(\"%s/out_%d.png\" % (out_dir, i))"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": []
}
],
"metadata": {
diff --git a/Scripts/Notebooks/SmokeyNb.ipynb b/Scripts/Notebooks/SmokeyNb.ipynb
index 6333d7c7bb713130d5707dbed19430d033bf57c6..328e4db95403085ee8a2bfe6a7eb48b2593085fe 100644
--- a/Scripts/Notebooks/SmokeyNb.ipynb
+++ b/Scripts/Notebooks/SmokeyNb.ipynb
@@ -174,7 +174,8 @@
],
"source": [
"from keras.layers import Input, Dense\n",
- "from keras.models import Model"
+ "from keras.models import Model\n",
+ "from keras import regularizers"
]
},
{
@@ -207,366 +208,10 @@
"in_size = 64 * 64"
]
},
- {
- "cell_type": "code",
- "execution_count": null,
- "metadata": {},
- "outputs": [],
- "source": [
- "\n",
- " "
- ]
- },
{
"cell_type": "code",
"execution_count": 12,
"metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "(990, 64, 64, 1)"
- ]
- },
- "execution_count": 12,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "densities.shape"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 13,
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "array([[[[ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " ...,\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00]],\n",
- "\n",
- " [[ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " ...,\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00]],\n",
- "\n",
- " [[ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " ...,\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00]],\n",
- "\n",
- " ...,\n",
- "\n",
- " [[ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " ...,\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00]],\n",
- "\n",
- " [[ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " ...,\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00]],\n",
- "\n",
- " [[ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " ...,\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00]]],\n",
- "\n",
- "\n",
- " [[[ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " ...,\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00]],\n",
- "\n",
- " [[ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " ...,\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00]],\n",
- "\n",
- " [[ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " ...,\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00]],\n",
- "\n",
- " ...,\n",
- "\n",
- " [[ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " ...,\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00]],\n",
- "\n",
- " [[ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " ...,\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00]],\n",
- "\n",
- " [[ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " ...,\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00]]],\n",
- "\n",
- "\n",
- " [[[ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " ...,\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00]],\n",
- "\n",
- " [[ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " ...,\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00]],\n",
- "\n",
- " [[ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " ...,\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00]],\n",
- "\n",
- " ...,\n",
- "\n",
- " [[ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " ...,\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00]],\n",
- "\n",
- " [[ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " ...,\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00]],\n",
- "\n",
- " [[ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " ...,\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00]]],\n",
- "\n",
- "\n",
- " ...,\n",
- "\n",
- "\n",
- " [[[ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " ...,\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00]],\n",
- "\n",
- " [[ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 1.3966976e-31],\n",
- " ...,\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00]],\n",
- "\n",
- " [[ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 2.1248558e-30],\n",
- " ...,\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00]],\n",
- "\n",
- " ...,\n",
- "\n",
- " [[ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [-0.0000000e+00],\n",
- " ...,\n",
- " [ 6.0734843e-24],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00]],\n",
- "\n",
- " [[ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " ...,\n",
- " [ 1.1996959e-24],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00]],\n",
- "\n",
- " [[ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " ...,\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00]]],\n",
- "\n",
- "\n",
- " [[[ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " ...,\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00]],\n",
- "\n",
- " [[ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 3.3538335e-31],\n",
- " ...,\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00]],\n",
- "\n",
- " [[ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 5.0400879e-30],\n",
- " ...,\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00]],\n",
- "\n",
- " ...,\n",
- "\n",
- " [[ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " ...,\n",
- " [ 7.2561730e-24],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00]],\n",
- "\n",
- " [[ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " ...,\n",
- " [ 1.4664798e-24],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00]],\n",
- "\n",
- " [[ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " ...,\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00]]],\n",
- "\n",
- "\n",
- " [[[ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " ...,\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00]],\n",
- "\n",
- " [[ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 8.0198798e-31],\n",
- " ...,\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00]],\n",
- "\n",
- " [[ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 1.1911826e-29],\n",
- " ...,\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00]],\n",
- "\n",
- " ...,\n",
- "\n",
- " [[ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " ...,\n",
- " [ 8.6359404e-24],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00]],\n",
- "\n",
- " [[ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " ...,\n",
- " [ 1.7843059e-24],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00]],\n",
- "\n",
- " [[ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " ...,\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00],\n",
- " [ 0.0000000e+00]]]], dtype=float32)"
- ]
- },
- "execution_count": 13,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "densities"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 14,
- "metadata": {},
"outputs": [],
"source": [
"input_img = Input(shape = (in_size, ))\n",
@@ -591,7 +236,7 @@
},
{
"cell_type": "code",
- "execution_count": 15,
+ "execution_count": 13,
"metadata": {},
"outputs": [],
"source": [
@@ -607,7 +252,7 @@
},
{
"cell_type": "code",
- "execution_count": 16,
+ "execution_count": 14,
"metadata": {},
"outputs": [],
"source": [
@@ -625,7 +270,7 @@
},
{
"cell_type": "code",
- "execution_count": 17,
+ "execution_count": 15,
"metadata": {},
"outputs": [],
"source": [
@@ -634,7 +279,7 @@
},
{
"cell_type": "code",
- "execution_count": 18,
+ "execution_count": 16,
"metadata": {},
"outputs": [],
"source": [
@@ -644,7 +289,7 @@
},
{
"cell_type": "code",
- "execution_count": 19,
+ "execution_count": 17,
"metadata": {},
"outputs": [
{
@@ -663,33 +308,7 @@
},
{
"cell_type": "code",
- "execution_count": 20,
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "array([[0., 0., 0., ..., 0., 0., 0.],\n",
- " [0., 0., 0., ..., 0., 0., 0.],\n",
- " [0., 0., 0., ..., 0., 0., 0.],\n",
- " ...,\n",
- " [0., 0., 0., ..., 0., 0., 0.],\n",
- " [0., 0., 0., ..., 0., 0., 0.],\n",
- " [0., 0., 0., ..., 0., 0., 0.]], dtype=float32)"
- ]
- },
- "execution_count": 20,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "densities"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 21,
+ "execution_count": 18,
"metadata": {},
"outputs": [
{
@@ -721,7 +340,7 @@
},
{
"cell_type": "code",
- "execution_count": 22,
+ "execution_count": 19,
"metadata": {},
"outputs": [
{
@@ -729,4488 +348,2512 @@
"output_type": "stream",
"text": [
"Train on 990 samples, validate on 110 samples\n",
- "Epoch 1/2500\n",
- "990/990 [==============================] - 1s 573us/step - loss: 0.6915 - val_loss: 0.6893\n",
- "Epoch 2/2500\n",
- "990/990 [==============================] - 0s 472us/step - loss: 0.6866 - val_loss: 0.6802\n",
- "Epoch 3/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.6701 - val_loss: 0.6370\n",
- "Epoch 4/2500\n",
- "990/990 [==============================] - 0s 456us/step - loss: 0.5919 - val_loss: 0.4558\n",
- "Epoch 5/2500\n",
- "990/990 [==============================] - 0s 459us/step - loss: 0.3829 - val_loss: 0.2100\n",
- "Epoch 6/2500\n",
- "990/990 [==============================] - 0s 447us/step - loss: 0.2026 - val_loss: 0.1351\n",
- "Epoch 7/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.1364 - val_loss: 0.1228\n",
- "Epoch 8/2500\n",
- "990/990 [==============================] - 0s 447us/step - loss: 0.1135 - val_loss: 0.1210\n",
- "Epoch 9/2500\n",
- "990/990 [==============================] - 0s 441us/step - loss: 0.1035 - val_loss: 0.1205\n",
- "Epoch 10/2500\n",
- "990/990 [==============================] - 0s 441us/step - loss: 0.0981 - val_loss: 0.1203\n",
- "Epoch 11/2500\n",
- "990/990 [==============================] - 0s 445us/step - loss: 0.0948 - val_loss: 0.1201\n",
- "Epoch 12/2500\n",
- "990/990 [==============================] - 0s 445us/step - loss: 0.0926 - val_loss: 0.1199\n",
- "Epoch 13/2500\n",
- "990/990 [==============================] - 0s 444us/step - loss: 0.0910 - val_loss: 0.1198\n",
- "Epoch 14/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0899 - val_loss: 0.1196\n",
- "Epoch 15/2500\n",
- "990/990 [==============================] - 0s 443us/step - loss: 0.0889 - val_loss: 0.1194\n",
- "Epoch 16/2500\n",
- "990/990 [==============================] - 0s 440us/step - loss: 0.0882 - val_loss: 0.1194\n",
- "Epoch 17/2500\n",
- "990/990 [==============================] - 0s 441us/step - loss: 0.0876 - val_loss: 0.1194\n",
- "Epoch 18/2500\n",
- "990/990 [==============================] - 0s 466us/step - loss: 0.0871 - val_loss: 0.1191\n",
- "Epoch 19/2500\n",
- "990/990 [==============================] - 0s 448us/step - loss: 0.0866 - val_loss: 0.1191\n",
- "Epoch 20/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0863 - val_loss: 0.1190\n",
- "Epoch 21/2500\n",
- "990/990 [==============================] - 0s 444us/step - loss: 0.0859 - val_loss: 0.1188\n",
- "Epoch 22/2500\n",
- "990/990 [==============================] - 0s 442us/step - loss: 0.0857 - val_loss: 0.1186\n",
- "Epoch 23/2500\n",
- "990/990 [==============================] - 0s 448us/step - loss: 0.0854 - val_loss: 0.1187\n",
- "Epoch 24/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0852 - val_loss: 0.1187\n",
- "Epoch 25/2500\n",
- "990/990 [==============================] - 0s 458us/step - loss: 0.0850 - val_loss: 0.1186\n",
- "Epoch 26/2500\n",
- "990/990 [==============================] - 0s 446us/step - loss: 0.0848 - val_loss: 0.1184\n",
- "Epoch 27/2500\n",
+ "Epoch 1/1200\n",
+ "990/990 [==============================] - 1s 551us/step - loss: 0.6915 - val_loss: 0.6893\n",
+ "Epoch 2/1200\n",
+ "990/990 [==============================] - 0s 471us/step - loss: 0.6866 - val_loss: 0.6802\n",
+ "Epoch 3/1200\n",
+ "990/990 [==============================] - 0s 452us/step - loss: 0.6701 - val_loss: 0.6370\n",
+ "Epoch 4/1200\n",
+ "990/990 [==============================] - 0s 448us/step - loss: 0.5919 - val_loss: 0.4558\n",
+ "Epoch 5/1200\n",
+ "990/990 [==============================] - 0s 461us/step - loss: 0.3829 - val_loss: 0.2100\n",
+ "Epoch 6/1200\n",
+ "990/990 [==============================] - 0s 465us/step - loss: 0.2026 - val_loss: 0.1351\n",
+ "Epoch 7/1200\n",
+ "990/990 [==============================] - 0s 447us/step - loss: 0.1364 - val_loss: 0.1228\n",
+ "Epoch 8/1200\n",
+ "990/990 [==============================] - 0s 441us/step - loss: 0.1135 - val_loss: 0.1210\n",
+ "Epoch 9/1200\n",
+ "990/990 [==============================] - 0s 447us/step - loss: 0.1035 - val_loss: 0.1205\n",
+ "Epoch 10/1200\n",
+ "990/990 [==============================] - 0s 442us/step - loss: 0.0981 - val_loss: 0.1203\n",
+ "Epoch 11/1200\n",
+ "990/990 [==============================] - 0s 450us/step - loss: 0.0948 - val_loss: 0.1201\n",
+ "Epoch 12/1200\n",
+ "990/990 [==============================] - 0s 478us/step - loss: 0.0926 - val_loss: 0.1199\n",
+ "Epoch 13/1200\n",
+ "990/990 [==============================] - 0s 474us/step - loss: 0.0910 - val_loss: 0.1198\n",
+ "Epoch 14/1200\n",
+ "990/990 [==============================] - 0s 461us/step - loss: 0.0899 - val_loss: 0.1196\n",
+ "Epoch 15/1200\n",
+ "990/990 [==============================] - 0s 436us/step - loss: 0.0889 - val_loss: 0.1194\n",
+ "Epoch 16/1200\n",
+ "990/990 [==============================] - 0s 468us/step - loss: 0.0882 - val_loss: 0.1194\n",
+ "Epoch 17/1200\n",
+ "990/990 [==============================] - 0s 457us/step - loss: 0.0876 - val_loss: 0.1194\n",
+ "Epoch 18/1200\n",
+ "990/990 [==============================] - 0s 445us/step - loss: 0.0871 - val_loss: 0.1191\n",
+ "Epoch 19/1200\n",
+ "990/990 [==============================] - 0s 446us/step - loss: 0.0866 - val_loss: 0.1191\n",
+ "Epoch 20/1200\n",
+ "990/990 [==============================] - 0s 452us/step - loss: 0.0863 - val_loss: 0.1190\n",
+ "Epoch 21/1200\n",
+ "990/990 [==============================] - 0s 448us/step - loss: 0.0859 - val_loss: 0.1188\n",
+ "Epoch 22/1200\n",
+ "990/990 [==============================] - 0s 460us/step - loss: 0.0857 - val_loss: 0.1186\n",
+ "Epoch 23/1200\n",
+ "990/990 [==============================] - 0s 463us/step - loss: 0.0854 - val_loss: 0.1187\n",
+ "Epoch 24/1200\n",
+ "990/990 [==============================] - 0s 466us/step - loss: 0.0852 - val_loss: 0.1187\n",
+ "Epoch 25/1200\n",
+ "990/990 [==============================] - 0s 439us/step - loss: 0.0850 - val_loss: 0.1186\n",
+ "Epoch 26/1200\n",
+ "990/990 [==============================] - 0s 443us/step - loss: 0.0848 - val_loss: 0.1184\n",
+ "Epoch 27/1200\n",
"990/990 [==============================] - 0s 450us/step - loss: 0.0846 - val_loss: 0.1184\n",
- "Epoch 28/2500\n",
- "990/990 [==============================] - 0s 456us/step - loss: 0.0844 - val_loss: 0.1183\n",
- "Epoch 29/2500\n",
- "990/990 [==============================] - 0s 456us/step - loss: 0.0843 - val_loss: 0.1182\n",
- "Epoch 30/2500\n",
- "990/990 [==============================] - 0s 460us/step - loss: 0.0841 - val_loss: 0.1182\n",
- "Epoch 31/2500\n",
- "990/990 [==============================] - 0s 448us/step - loss: 0.0840 - val_loss: 0.1182\n",
- "Epoch 32/2500\n",
- "990/990 [==============================] - 0s 447us/step - loss: 0.0839 - val_loss: 0.1182\n",
- "Epoch 33/2500\n",
- "990/990 [==============================] - 0s 446us/step - loss: 0.0838 - val_loss: 0.1180\n",
- "Epoch 34/2500\n",
- "990/990 [==============================] - 0s 465us/step - loss: 0.0837 - val_loss: 0.1178\n",
- "Epoch 35/2500\n",
- "990/990 [==============================] - 0s 458us/step - loss: 0.0836 - val_loss: 0.1178\n",
- "Epoch 36/2500\n",
- "990/990 [==============================] - 0s 440us/step - loss: 0.0835 - val_loss: 0.1179\n",
- "Epoch 37/2500\n",
- "990/990 [==============================] - 0s 443us/step - loss: 0.0834 - val_loss: 0.1177\n",
- "Epoch 38/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0833 - val_loss: 0.1177\n",
- "Epoch 39/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0832 - val_loss: 0.1178\n",
- "Epoch 40/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0831 - val_loss: 0.1176\n",
- "Epoch 41/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0830 - val_loss: 0.1175\n",
- "Epoch 42/2500\n",
- "990/990 [==============================] - 0s 443us/step - loss: 0.0829 - val_loss: 0.1174\n",
- "Epoch 43/2500\n",
- "990/990 [==============================] - 0s 446us/step - loss: 0.0829 - val_loss: 0.1175\n",
- "Epoch 44/2500\n",
- "990/990 [==============================] - 0s 447us/step - loss: 0.0828 - val_loss: 0.1174\n",
- "Epoch 45/2500\n",
- "990/990 [==============================] - 0s 458us/step - loss: 0.0827 - val_loss: 0.1174\n",
- "Epoch 46/2500\n",
- "990/990 [==============================] - 0s 443us/step - loss: 0.0827 - val_loss: 0.1174\n",
- "Epoch 47/2500\n",
- "990/990 [==============================] - 0s 446us/step - loss: 0.0826 - val_loss: 0.1174\n",
- "Epoch 48/2500\n",
- "990/990 [==============================] - 0s 469us/step - loss: 0.0825 - val_loss: 0.1173\n",
- "Epoch 49/2500\n",
- "990/990 [==============================] - 0s 473us/step - loss: 0.0825 - val_loss: 0.1171\n",
- "Epoch 50/2500\n",
- "990/990 [==============================] - 0s 456us/step - loss: 0.0824 - val_loss: 0.1173\n",
- "Epoch 51/2500\n",
- "990/990 [==============================] - 0s 473us/step - loss: 0.0824 - val_loss: 0.1172\n",
- "Epoch 52/2500\n",
- "990/990 [==============================] - 0s 463us/step - loss: 0.0823 - val_loss: 0.1171\n",
- "Epoch 53/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0822 - val_loss: 0.1170\n",
- "Epoch 54/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0822 - val_loss: 0.1169\n",
- "Epoch 55/2500\n",
- "990/990 [==============================] - 0s 467us/step - loss: 0.0821 - val_loss: 0.1169\n",
- "Epoch 56/2500\n",
- "990/990 [==============================] - 0s 479us/step - loss: 0.0821 - val_loss: 0.1169\n",
- "Epoch 57/2500\n",
- "990/990 [==============================] - 0s 477us/step - loss: 0.0820 - val_loss: 0.1167\n",
- "Epoch 58/2500\n",
- "990/990 [==============================] - 0s 488us/step - loss: 0.0820 - val_loss: 0.1168\n",
- "Epoch 59/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0819 - val_loss: 0.1168\n",
- "Epoch 60/2500\n",
- "990/990 [==============================] - 0s 481us/step - loss: 0.0819 - val_loss: 0.1167\n",
- "Epoch 61/2500\n",
- "990/990 [==============================] - 0s 486us/step - loss: 0.0818 - val_loss: 0.1166\n",
- "Epoch 62/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0818 - val_loss: 0.1167\n",
- "Epoch 63/2500\n",
- "990/990 [==============================] - 0s 442us/step - loss: 0.0817 - val_loss: 0.1166\n",
- "Epoch 64/2500\n",
- "990/990 [==============================] - 0s 446us/step - loss: 0.0817 - val_loss: 0.1166\n",
- "Epoch 65/2500\n",
- "990/990 [==============================] - 0s 443us/step - loss: 0.0816 - val_loss: 0.1166\n",
- "Epoch 66/2500\n",
- "990/990 [==============================] - 0s 479us/step - loss: 0.0816 - val_loss: 0.1165\n",
- "Epoch 67/2500\n",
- "990/990 [==============================] - 0s 460us/step - loss: 0.0816 - val_loss: 0.1165\n",
- "Epoch 68/2500\n",
- "990/990 [==============================] - 0s 459us/step - loss: 0.0815 - val_loss: 0.1163\n",
- "Epoch 69/2500\n",
- "990/990 [==============================] - 1s 554us/step - loss: 0.0815 - val_loss: 0.1165\n",
- "Epoch 70/2500\n",
- "990/990 [==============================] - 1s 527us/step - loss: 0.0814 - val_loss: 0.1164\n",
- "Epoch 71/2500\n",
- "990/990 [==============================] - 0s 485us/step - loss: 0.0814 - val_loss: 0.1164\n",
- "Epoch 72/2500\n",
- "990/990 [==============================] - 1s 537us/step - loss: 0.0813 - val_loss: 0.1163\n",
- "Epoch 73/2500\n",
- "990/990 [==============================] - 1s 522us/step - loss: 0.0813 - val_loss: 0.1162\n",
- "Epoch 74/2500\n",
- "990/990 [==============================] - 0s 447us/step - loss: 0.0812 - val_loss: 0.1163\n",
- "Epoch 75/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0812 - val_loss: 0.1163\n",
- "Epoch 76/2500\n",
- "990/990 [==============================] - 0s 477us/step - loss: 0.0812 - val_loss: 0.1163\n",
- "Epoch 77/2500\n",
- "990/990 [==============================] - 0s 470us/step - loss: 0.0811 - val_loss: 0.1161\n",
- "Epoch 78/2500\n"
+ "Epoch 28/1200\n",
+ "990/990 [==============================] - 0s 439us/step - loss: 0.0844 - val_loss: 0.1183\n",
+ "Epoch 29/1200\n",
+ "990/990 [==============================] - 0s 445us/step - loss: 0.0843 - val_loss: 0.1182\n",
+ "Epoch 30/1200\n",
+ "990/990 [==============================] - 0s 448us/step - loss: 0.0841 - val_loss: 0.1182\n",
+ "Epoch 31/1200\n",
+ "990/990 [==============================] - 0s 445us/step - loss: 0.0840 - val_loss: 0.1182\n",
+ "Epoch 32/1200\n",
+ "990/990 [==============================] - 0s 446us/step - loss: 0.0839 - val_loss: 0.1182\n",
+ "Epoch 33/1200\n",
+ "990/990 [==============================] - 0s 448us/step - loss: 0.0838 - val_loss: 0.1180\n",
+ "Epoch 34/1200\n",
+ "990/990 [==============================] - 0s 449us/step - loss: 0.0837 - val_loss: 0.1178\n",
+ "Epoch 35/1200\n",
+ "990/990 [==============================] - 0s 478us/step - loss: 0.0836 - val_loss: 0.1178\n",
+ "Epoch 36/1200\n",
+ "990/990 [==============================] - 0s 473us/step - loss: 0.0835 - val_loss: 0.1179\n",
+ "Epoch 37/1200\n",
+ "990/990 [==============================] - 0s 448us/step - loss: 0.0834 - val_loss: 0.1177\n",
+ "Epoch 38/1200\n",
+ "990/990 [==============================] - 0s 458us/step - loss: 0.0833 - val_loss: 0.1177\n",
+ "Epoch 39/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0832 - val_loss: 0.1178\n",
+ "Epoch 40/1200\n",
+ "990/990 [==============================] - 0s 429us/step - loss: 0.0831 - val_loss: 0.1176\n",
+ "Epoch 41/1200\n",
+ "990/990 [==============================] - 0s 460us/step - loss: 0.0830 - val_loss: 0.1175\n",
+ "Epoch 42/1200\n",
+ "990/990 [==============================] - 0s 436us/step - loss: 0.0829 - val_loss: 0.1174\n",
+ "Epoch 43/1200\n",
+ "990/990 [==============================] - 0s 441us/step - loss: 0.0829 - val_loss: 0.1175\n",
+ "Epoch 44/1200\n",
+ "990/990 [==============================] - 0s 436us/step - loss: 0.0828 - val_loss: 0.1174\n",
+ "Epoch 45/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0827 - val_loss: 0.1174\n",
+ "Epoch 46/1200\n",
+ "990/990 [==============================] - 0s 430us/step - loss: 0.0827 - val_loss: 0.1174\n",
+ "Epoch 47/1200\n",
+ "990/990 [==============================] - 0s 449us/step - loss: 0.0826 - val_loss: 0.1174\n",
+ "Epoch 48/1200\n",
+ "990/990 [==============================] - 0s 447us/step - loss: 0.0825 - val_loss: 0.1173\n",
+ "Epoch 49/1200\n",
+ "990/990 [==============================] - 0s 444us/step - loss: 0.0825 - val_loss: 0.1171\n",
+ "Epoch 50/1200\n",
+ "990/990 [==============================] - 0s 445us/step - loss: 0.0824 - val_loss: 0.1173\n",
+ "Epoch 51/1200\n",
+ "990/990 [==============================] - 0s 450us/step - loss: 0.0824 - val_loss: 0.1172\n",
+ "Epoch 52/1200\n",
+ "990/990 [==============================] - 0s 466us/step - loss: 0.0823 - val_loss: 0.1171\n",
+ "Epoch 53/1200\n",
+ "990/990 [==============================] - 0s 434us/step - loss: 0.0822 - val_loss: 0.1170\n",
+ "Epoch 54/1200\n",
+ "990/990 [==============================] - 0s 427us/step - loss: 0.0822 - val_loss: 0.1169\n",
+ "Epoch 55/1200\n",
+ "990/990 [==============================] - 0s 430us/step - loss: 0.0821 - val_loss: 0.1169\n",
+ "Epoch 56/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0821 - val_loss: 0.1169\n",
+ "Epoch 57/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0820 - val_loss: 0.1167\n",
+ "Epoch 58/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0820 - val_loss: 0.1168\n",
+ "Epoch 59/1200\n",
+ "990/990 [==============================] - 0s 429us/step - loss: 0.0819 - val_loss: 0.1168\n",
+ "Epoch 60/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0819 - val_loss: 0.1167\n",
+ "Epoch 61/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0818 - val_loss: 0.1166\n",
+ "Epoch 62/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0818 - val_loss: 0.1167\n",
+ "Epoch 63/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0817 - val_loss: 0.1166\n",
+ "Epoch 64/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0817 - val_loss: 0.1166\n",
+ "Epoch 65/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0816 - val_loss: 0.1166\n",
+ "Epoch 66/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0816 - val_loss: 0.1165\n",
+ "Epoch 67/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0816 - val_loss: 0.1165\n",
+ "Epoch 68/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0815 - val_loss: 0.1163\n",
+ "Epoch 69/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0815 - val_loss: 0.1165\n",
+ "Epoch 70/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0814 - val_loss: 0.1164\n",
+ "Epoch 71/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0814 - val_loss: 0.1164\n",
+ "Epoch 72/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0813 - val_loss: 0.1163\n",
+ "Epoch 73/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0813 - val_loss: 0.1162\n",
+ "Epoch 74/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0812 - val_loss: 0.1163\n",
+ "Epoch 75/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0812 - val_loss: 0.1163\n",
+ "Epoch 76/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0812 - val_loss: 0.1163\n",
+ "Epoch 77/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0811 - val_loss: 0.1161\n",
+ "Epoch 78/1200\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
- "990/990 [==============================] - 0s 462us/step - loss: 0.0811 - val_loss: 0.1161\n",
- "Epoch 79/2500\n",
- "990/990 [==============================] - 0s 461us/step - loss: 0.0810 - val_loss: 0.1160\n",
- "Epoch 80/2500\n",
- "990/990 [==============================] - 0s 462us/step - loss: 0.0810 - val_loss: 0.1161\n",
- "Epoch 81/2500\n",
- "990/990 [==============================] - 0s 440us/step - loss: 0.0810 - val_loss: 0.1162\n",
- "Epoch 82/2500\n",
- "990/990 [==============================] - 0s 493us/step - loss: 0.0809 - val_loss: 0.1160\n",
- "Epoch 83/2500\n",
- "990/990 [==============================] - 0s 482us/step - loss: 0.0809 - val_loss: 0.1159\n",
- "Epoch 84/2500\n",
- "990/990 [==============================] - 0s 485us/step - loss: 0.0808 - val_loss: 0.1160\n",
- "Epoch 85/2500\n",
- "990/990 [==============================] - 0s 472us/step - loss: 0.0808 - val_loss: 0.1158\n",
- "Epoch 86/2500\n",
- "990/990 [==============================] - 0s 480us/step - loss: 0.0807 - val_loss: 0.1158\n",
- "Epoch 87/2500\n",
- "990/990 [==============================] - 0s 470us/step - loss: 0.0807 - val_loss: 0.1159\n",
- "Epoch 88/2500\n",
- "990/990 [==============================] - 0s 496us/step - loss: 0.0807 - val_loss: 0.1159\n",
- "Epoch 89/2500\n",
- "990/990 [==============================] - 0s 482us/step - loss: 0.0806 - val_loss: 0.1158\n",
- "Epoch 90/2500\n",
- "990/990 [==============================] - 0s 473us/step - loss: 0.0806 - val_loss: 0.1158\n",
- "Epoch 91/2500\n",
- "990/990 [==============================] - 0s 472us/step - loss: 0.0805 - val_loss: 0.1159\n",
- "Epoch 92/2500\n",
- "990/990 [==============================] - 0s 473us/step - loss: 0.0805 - val_loss: 0.1157\n",
- "Epoch 93/2500\n",
- "990/990 [==============================] - 0s 470us/step - loss: 0.0804 - val_loss: 0.1159\n",
- "Epoch 94/2500\n",
- "990/990 [==============================] - 0s 474us/step - loss: 0.0804 - val_loss: 0.1157\n",
- "Epoch 95/2500\n",
- "990/990 [==============================] - 0s 465us/step - loss: 0.0804 - val_loss: 0.1157\n",
- "Epoch 96/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0803 - val_loss: 0.1156\n",
- "Epoch 97/2500\n",
- "990/990 [==============================] - 0s 456us/step - loss: 0.0803 - val_loss: 0.1156\n",
- "Epoch 98/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0802 - val_loss: 0.1155\n",
- "Epoch 99/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0802 - val_loss: 0.1155\n",
- "Epoch 100/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0801 - val_loss: 0.1154\n",
- "Epoch 101/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0801 - val_loss: 0.1153\n",
- "Epoch 102/2500\n",
- "990/990 [==============================] - 0s 456us/step - loss: 0.0800 - val_loss: 0.1154\n",
- "Epoch 103/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0800 - val_loss: 0.1153\n",
- "Epoch 104/2500\n",
- "990/990 [==============================] - 0s 464us/step - loss: 0.0799 - val_loss: 0.1153\n",
- "Epoch 105/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0799 - val_loss: 0.1153\n",
- "Epoch 106/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0799 - val_loss: 0.1151\n",
- "Epoch 107/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0798 - val_loss: 0.1152\n",
- "Epoch 108/2500\n",
- "990/990 [==============================] - 0s 459us/step - loss: 0.0798 - val_loss: 0.1152\n",
- "Epoch 109/2500\n",
- "990/990 [==============================] - 0s 446us/step - loss: 0.0797 - val_loss: 0.1151\n",
- "Epoch 110/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0797 - val_loss: 0.1151\n",
- "Epoch 111/2500\n",
- "990/990 [==============================] - 0s 443us/step - loss: 0.0796 - val_loss: 0.1150\n",
- "Epoch 112/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0796 - val_loss: 0.1149\n",
- "Epoch 113/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0795 - val_loss: 0.1149\n",
- "Epoch 114/2500\n",
- "990/990 [==============================] - 0s 462us/step - loss: 0.0794 - val_loss: 0.1150\n",
- "Epoch 115/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0794 - val_loss: 0.1148\n",
- "Epoch 116/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0793 - val_loss: 0.1149\n",
- "Epoch 117/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0793 - val_loss: 0.1150\n",
- "Epoch 118/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0792 - val_loss: 0.1149\n",
- "Epoch 119/2500\n",
- "990/990 [==============================] - 0s 448us/step - loss: 0.0792 - val_loss: 0.1149\n",
- "Epoch 120/2500\n",
- "990/990 [==============================] - 0s 466us/step - loss: 0.0791 - val_loss: 0.1148\n",
- "Epoch 121/2500\n",
- "990/990 [==============================] - 0s 498us/step - loss: 0.0791 - val_loss: 0.1148\n",
- "Epoch 122/2500\n",
- "990/990 [==============================] - 0s 473us/step - loss: 0.0790 - val_loss: 0.1148\n",
- "Epoch 123/2500\n",
- "990/990 [==============================] - 0s 471us/step - loss: 0.0789 - val_loss: 0.1148\n",
- "Epoch 124/2500\n",
- "990/990 [==============================] - 0s 471us/step - loss: 0.0789 - val_loss: 0.1147\n",
- "Epoch 125/2500\n",
- "990/990 [==============================] - 0s 494us/step - loss: 0.0788 - val_loss: 0.1146\n",
- "Epoch 126/2500\n",
- "990/990 [==============================] - 0s 487us/step - loss: 0.0788 - val_loss: 0.1146\n",
- "Epoch 127/2500\n",
- "990/990 [==============================] - 1s 510us/step - loss: 0.0787 - val_loss: 0.1145\n",
- "Epoch 128/2500\n",
- "990/990 [==============================] - 0s 464us/step - loss: 0.0786 - val_loss: 0.1145\n",
- "Epoch 129/2500\n",
- "990/990 [==============================] - 0s 459us/step - loss: 0.0786 - val_loss: 0.1145\n",
- "Epoch 130/2500\n",
- "990/990 [==============================] - 0s 472us/step - loss: 0.0785 - val_loss: 0.1145\n",
- "Epoch 131/2500\n",
- "990/990 [==============================] - 0s 463us/step - loss: 0.0784 - val_loss: 0.1144\n",
- "Epoch 132/2500\n",
- "990/990 [==============================] - 0s 464us/step - loss: 0.0784 - val_loss: 0.1146\n",
- "Epoch 133/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0783 - val_loss: 0.1145\n",
- "Epoch 134/2500\n",
- "990/990 [==============================] - 0s 461us/step - loss: 0.0782 - val_loss: 0.1145\n",
- "Epoch 135/2500\n",
- "990/990 [==============================] - 0s 460us/step - loss: 0.0782 - val_loss: 0.1144\n",
- "Epoch 136/2500\n",
- "990/990 [==============================] - 0s 446us/step - loss: 0.0781 - val_loss: 0.1142\n",
- "Epoch 137/2500\n",
- "990/990 [==============================] - 0s 463us/step - loss: 0.0780 - val_loss: 0.1143\n",
- "Epoch 138/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0780 - val_loss: 0.1142\n",
- "Epoch 139/2500\n",
- "990/990 [==============================] - 0s 466us/step - loss: 0.0779 - val_loss: 0.1142\n",
- "Epoch 140/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0778 - val_loss: 0.1143\n",
- "Epoch 141/2500\n",
- "990/990 [==============================] - 0s 465us/step - loss: 0.0777 - val_loss: 0.1143\n",
- "Epoch 142/2500\n",
- "990/990 [==============================] - 0s 461us/step - loss: 0.0777 - val_loss: 0.1142\n",
- "Epoch 143/2500\n",
- "990/990 [==============================] - 0s 468us/step - loss: 0.0776 - val_loss: 0.1141\n",
- "Epoch 144/2500\n",
- "990/990 [==============================] - 0s 470us/step - loss: 0.0775 - val_loss: 0.1142\n",
- "Epoch 145/2500\n",
- "990/990 [==============================] - 0s 468us/step - loss: 0.0775 - val_loss: 0.1142\n",
- "Epoch 146/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0774 - val_loss: 0.1141\n",
- "Epoch 147/2500\n",
- "990/990 [==============================] - 0s 471us/step - loss: 0.0773 - val_loss: 0.1140\n",
- "Epoch 148/2500\n",
- "990/990 [==============================] - 0s 469us/step - loss: 0.0772 - val_loss: 0.1140\n",
- "Epoch 149/2500\n",
- "990/990 [==============================] - 0s 448us/step - loss: 0.0772 - val_loss: 0.1140\n",
- "Epoch 150/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0771 - val_loss: 0.1139\n",
- "Epoch 151/2500\n",
- "990/990 [==============================] - 0s 460us/step - loss: 0.0770 - val_loss: 0.1139\n",
- "Epoch 152/2500\n",
- "990/990 [==============================] - 0s 465us/step - loss: 0.0769 - val_loss: 0.1139\n",
- "Epoch 153/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0769 - val_loss: 0.1138\n",
- "Epoch 154/2500\n",
- "990/990 [==============================] - 0s 456us/step - loss: 0.0768 - val_loss: 0.1139\n",
- "Epoch 155/2500\n"
+ "990/990 [==============================] - 0s 416us/step - loss: 0.0811 - val_loss: 0.1161\n",
+ "Epoch 79/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0810 - val_loss: 0.1160\n",
+ "Epoch 80/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0810 - val_loss: 0.1161\n",
+ "Epoch 81/1200\n",
+ "990/990 [==============================] - 0s 417us/step - loss: 0.0810 - val_loss: 0.1162\n",
+ "Epoch 82/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0809 - val_loss: 0.1160\n",
+ "Epoch 83/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0809 - val_loss: 0.1159\n",
+ "Epoch 84/1200\n",
+ "990/990 [==============================] - 0s 427us/step - loss: 0.0808 - val_loss: 0.1160\n",
+ "Epoch 85/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0808 - val_loss: 0.1158\n",
+ "Epoch 86/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0807 - val_loss: 0.1158\n",
+ "Epoch 87/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0807 - val_loss: 0.1159\n",
+ "Epoch 88/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0807 - val_loss: 0.1159\n",
+ "Epoch 89/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0806 - val_loss: 0.1158\n",
+ "Epoch 90/1200\n",
+ "990/990 [==============================] - 0s 427us/step - loss: 0.0806 - val_loss: 0.1158\n",
+ "Epoch 91/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0805 - val_loss: 0.1159\n",
+ "Epoch 92/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0805 - val_loss: 0.1157\n",
+ "Epoch 93/1200\n",
+ "990/990 [==============================] - 0s 427us/step - loss: 0.0804 - val_loss: 0.1159\n",
+ "Epoch 94/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0804 - val_loss: 0.1157\n",
+ "Epoch 95/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0804 - val_loss: 0.1157\n",
+ "Epoch 96/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0803 - val_loss: 0.1156\n",
+ "Epoch 97/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0803 - val_loss: 0.1156\n",
+ "Epoch 98/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0802 - val_loss: 0.1155\n",
+ "Epoch 99/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0802 - val_loss: 0.1155\n",
+ "Epoch 100/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0801 - val_loss: 0.1154\n",
+ "Epoch 101/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0801 - val_loss: 0.1153\n",
+ "Epoch 102/1200\n",
+ "990/990 [==============================] - 0s 432us/step - loss: 0.0800 - val_loss: 0.1154\n",
+ "Epoch 103/1200\n",
+ "990/990 [==============================] - 0s 432us/step - loss: 0.0800 - val_loss: 0.1153\n",
+ "Epoch 104/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0799 - val_loss: 0.1153\n",
+ "Epoch 105/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0799 - val_loss: 0.1153\n",
+ "Epoch 106/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0799 - val_loss: 0.1151\n",
+ "Epoch 107/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0798 - val_loss: 0.1152\n",
+ "Epoch 108/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0798 - val_loss: 0.1152\n",
+ "Epoch 109/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0797 - val_loss: 0.1151\n",
+ "Epoch 110/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0797 - val_loss: 0.1151\n",
+ "Epoch 111/1200\n",
+ "990/990 [==============================] - 0s 433us/step - loss: 0.0796 - val_loss: 0.1150\n",
+ "Epoch 112/1200\n",
+ "990/990 [==============================] - 0s 434us/step - loss: 0.0796 - val_loss: 0.1149\n",
+ "Epoch 113/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0795 - val_loss: 0.1149\n",
+ "Epoch 114/1200\n",
+ "990/990 [==============================] - 0s 438us/step - loss: 0.0794 - val_loss: 0.1150\n",
+ "Epoch 115/1200\n",
+ "990/990 [==============================] - 0s 430us/step - loss: 0.0794 - val_loss: 0.1148\n",
+ "Epoch 116/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0793 - val_loss: 0.1149\n",
+ "Epoch 117/1200\n",
+ "990/990 [==============================] - 0s 438us/step - loss: 0.0793 - val_loss: 0.1150\n",
+ "Epoch 118/1200\n",
+ "990/990 [==============================] - 0s 442us/step - loss: 0.0792 - val_loss: 0.1149\n",
+ "Epoch 119/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0792 - val_loss: 0.1149\n",
+ "Epoch 120/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0791 - val_loss: 0.1148\n",
+ "Epoch 121/1200\n",
+ "990/990 [==============================] - 0s 430us/step - loss: 0.0791 - val_loss: 0.1148\n",
+ "Epoch 122/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0790 - val_loss: 0.1148\n",
+ "Epoch 123/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0789 - val_loss: 0.1148\n",
+ "Epoch 124/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0789 - val_loss: 0.1147\n",
+ "Epoch 125/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0788 - val_loss: 0.1146\n",
+ "Epoch 126/1200\n",
+ "990/990 [==============================] - 0s 427us/step - loss: 0.0788 - val_loss: 0.1146\n",
+ "Epoch 127/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0787 - val_loss: 0.1145\n",
+ "Epoch 128/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0786 - val_loss: 0.1145\n",
+ "Epoch 129/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0786 - val_loss: 0.1145\n",
+ "Epoch 130/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0785 - val_loss: 0.1145\n",
+ "Epoch 131/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0784 - val_loss: 0.1144\n",
+ "Epoch 132/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0784 - val_loss: 0.1146\n",
+ "Epoch 133/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0783 - val_loss: 0.1145\n",
+ "Epoch 134/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0782 - val_loss: 0.1145\n",
+ "Epoch 135/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0782 - val_loss: 0.1144\n",
+ "Epoch 136/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0781 - val_loss: 0.1142\n",
+ "Epoch 137/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0780 - val_loss: 0.1143\n",
+ "Epoch 138/1200\n",
+ "990/990 [==============================] - 0s 427us/step - loss: 0.0780 - val_loss: 0.1142\n",
+ "Epoch 139/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0779 - val_loss: 0.1142\n",
+ "Epoch 140/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0778 - val_loss: 0.1143\n",
+ "Epoch 141/1200\n",
+ "990/990 [==============================] - 0s 427us/step - loss: 0.0777 - val_loss: 0.1143\n",
+ "Epoch 142/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0777 - val_loss: 0.1142\n",
+ "Epoch 143/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0776 - val_loss: 0.1141\n",
+ "Epoch 144/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0775 - val_loss: 0.1142\n",
+ "Epoch 145/1200\n",
+ "990/990 [==============================] - 0s 429us/step - loss: 0.0775 - val_loss: 0.1142\n",
+ "Epoch 146/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0774 - val_loss: 0.1141\n",
+ "Epoch 147/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0773 - val_loss: 0.1140\n",
+ "Epoch 148/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0772 - val_loss: 0.1140\n",
+ "Epoch 149/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0772 - val_loss: 0.1140\n",
+ "Epoch 150/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0771 - val_loss: 0.1139\n",
+ "Epoch 151/1200\n",
+ "990/990 [==============================] - 0s 427us/step - loss: 0.0770 - val_loss: 0.1139\n",
+ "Epoch 152/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0769 - val_loss: 0.1139\n",
+ "Epoch 153/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0769 - val_loss: 0.1138\n",
+ "Epoch 154/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0768 - val_loss: 0.1139\n",
+ "Epoch 155/1200\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
- "990/990 [==============================] - 0s 459us/step - loss: 0.0767 - val_loss: 0.1138\n",
- "Epoch 156/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0766 - val_loss: 0.1138\n",
- "Epoch 157/2500\n",
- "990/990 [==============================] - 0s 439us/step - loss: 0.0766 - val_loss: 0.1138\n",
- "Epoch 158/2500\n",
- "990/990 [==============================] - 0s 444us/step - loss: 0.0765 - val_loss: 0.1138\n",
- "Epoch 159/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0764 - val_loss: 0.1138\n",
- "Epoch 160/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0763 - val_loss: 0.1139\n",
- "Epoch 161/2500\n",
- "990/990 [==============================] - 0s 458us/step - loss: 0.0762 - val_loss: 0.1137\n",
- "Epoch 162/2500\n",
- "990/990 [==============================] - 0s 460us/step - loss: 0.0762 - val_loss: 0.1137\n",
- "Epoch 163/2500\n",
- "990/990 [==============================] - 0s 456us/step - loss: 0.0761 - val_loss: 0.1138\n",
- "Epoch 164/2500\n",
- "990/990 [==============================] - 0s 470us/step - loss: 0.0760 - val_loss: 0.1137\n",
- "Epoch 165/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0759 - val_loss: 0.1137\n",
- "Epoch 166/2500\n",
- "990/990 [==============================] - 0s 465us/step - loss: 0.0759 - val_loss: 0.1136\n",
- "Epoch 167/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0758 - val_loss: 0.1135\n",
- "Epoch 168/2500\n",
- "990/990 [==============================] - 0s 467us/step - loss: 0.0757 - val_loss: 0.1136\n",
- "Epoch 169/2500\n",
- "990/990 [==============================] - 0s 485us/step - loss: 0.0756 - val_loss: 0.1136\n",
- "Epoch 170/2500\n",
- "990/990 [==============================] - 0s 464us/step - loss: 0.0755 - val_loss: 0.1136\n",
- "Epoch 171/2500\n",
- "990/990 [==============================] - 0s 468us/step - loss: 0.0755 - val_loss: 0.1136\n",
- "Epoch 172/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0754 - val_loss: 0.1135\n",
- "Epoch 173/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0753 - val_loss: 0.1134\n",
- "Epoch 174/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0752 - val_loss: 0.1134\n",
- "Epoch 175/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0751 - val_loss: 0.1134\n",
- "Epoch 176/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0751 - val_loss: 0.1134\n",
- "Epoch 177/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0750 - val_loss: 0.1134\n",
- "Epoch 178/2500\n",
- "990/990 [==============================] - 0s 463us/step - loss: 0.0749 - val_loss: 0.1134\n",
- "Epoch 179/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0748 - val_loss: 0.1133\n",
- "Epoch 180/2500\n",
- "990/990 [==============================] - 0s 464us/step - loss: 0.0748 - val_loss: 0.1132\n",
- "Epoch 181/2500\n",
- "990/990 [==============================] - 0s 461us/step - loss: 0.0747 - val_loss: 0.1132\n",
- "Epoch 182/2500\n",
- "990/990 [==============================] - 0s 471us/step - loss: 0.0746 - val_loss: 0.1133\n",
- "Epoch 183/2500\n",
- "990/990 [==============================] - 0s 475us/step - loss: 0.0745 - val_loss: 0.1133\n",
- "Epoch 184/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0744 - val_loss: 0.1133\n",
- "Epoch 185/2500\n",
- "990/990 [==============================] - 0s 459us/step - loss: 0.0744 - val_loss: 0.1132\n",
- "Epoch 186/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0743 - val_loss: 0.1132\n",
- "Epoch 187/2500\n",
- "990/990 [==============================] - 0s 458us/step - loss: 0.0742 - val_loss: 0.1132\n",
- "Epoch 188/2500\n",
- "990/990 [==============================] - 0s 459us/step - loss: 0.0741 - val_loss: 0.1131\n",
- "Epoch 189/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0740 - val_loss: 0.1131\n",
- "Epoch 190/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0740 - val_loss: 0.1131\n",
- "Epoch 191/2500\n",
- "990/990 [==============================] - 0s 443us/step - loss: 0.0739 - val_loss: 0.1132\n",
- "Epoch 192/2500\n",
- "990/990 [==============================] - 0s 448us/step - loss: 0.0738 - val_loss: 0.1131\n",
- "Epoch 193/2500\n",
- "990/990 [==============================] - 0s 448us/step - loss: 0.0737 - val_loss: 0.1131\n",
- "Epoch 194/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0736 - val_loss: 0.1131\n",
- "Epoch 195/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0736 - val_loss: 0.1130\n",
- "Epoch 196/2500\n",
- "990/990 [==============================] - 0s 458us/step - loss: 0.0735 - val_loss: 0.1131\n",
- "Epoch 197/2500\n",
- "990/990 [==============================] - 0s 456us/step - loss: 0.0734 - val_loss: 0.1131\n",
- "Epoch 198/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0733 - val_loss: 0.1130\n",
- "Epoch 199/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0732 - val_loss: 0.1130\n",
- "Epoch 200/2500\n",
- "990/990 [==============================] - 0s 461us/step - loss: 0.0732 - val_loss: 0.1130\n",
- "Epoch 201/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0731 - val_loss: 0.1129\n",
- "Epoch 202/2500\n",
- "990/990 [==============================] - 0s 446us/step - loss: 0.0730 - val_loss: 0.1129\n",
- "Epoch 203/2500\n",
- "990/990 [==============================] - 0s 448us/step - loss: 0.0729 - val_loss: 0.1129\n",
- "Epoch 204/2500\n",
- "990/990 [==============================] - 0s 456us/step - loss: 0.0728 - val_loss: 0.1128\n",
- "Epoch 205/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0728 - val_loss: 0.1128\n",
- "Epoch 206/2500\n",
- "990/990 [==============================] - 0s 464us/step - loss: 0.0727 - val_loss: 0.1127\n",
- "Epoch 207/2500\n",
- "990/990 [==============================] - 0s 459us/step - loss: 0.0726 - val_loss: 0.1128\n",
- "Epoch 208/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0725 - val_loss: 0.1128\n",
- "Epoch 209/2500\n",
- "990/990 [==============================] - 0s 447us/step - loss: 0.0724 - val_loss: 0.1127\n",
- "Epoch 210/2500\n",
- "990/990 [==============================] - 0s 459us/step - loss: 0.0724 - val_loss: 0.1127\n",
- "Epoch 211/2500\n",
- "990/990 [==============================] - 0s 456us/step - loss: 0.0723 - val_loss: 0.1127\n",
- "Epoch 212/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0722 - val_loss: 0.1127\n",
- "Epoch 213/2500\n",
- "990/990 [==============================] - 0s 469us/step - loss: 0.0721 - val_loss: 0.1127\n",
- "Epoch 214/2500\n",
- "990/990 [==============================] - 0s 481us/step - loss: 0.0720 - val_loss: 0.1127\n",
- "Epoch 215/2500\n",
- "990/990 [==============================] - 0s 470us/step - loss: 0.0720 - val_loss: 0.1127\n",
- "Epoch 216/2500\n",
- "990/990 [==============================] - 0s 466us/step - loss: 0.0719 - val_loss: 0.1126\n",
- "Epoch 217/2500\n",
- "990/990 [==============================] - 0s 464us/step - loss: 0.0718 - val_loss: 0.1126\n",
- "Epoch 218/2500\n",
- "990/990 [==============================] - 0s 468us/step - loss: 0.0717 - val_loss: 0.1126\n",
- "Epoch 219/2500\n",
- "990/990 [==============================] - 0s 463us/step - loss: 0.0717 - val_loss: 0.1124\n",
- "Epoch 220/2500\n",
- "990/990 [==============================] - 0s 461us/step - loss: 0.0716 - val_loss: 0.1125\n",
- "Epoch 221/2500\n",
- "990/990 [==============================] - 0s 461us/step - loss: 0.0715 - val_loss: 0.1125\n",
- "Epoch 222/2500\n",
- "990/990 [==============================] - 0s 459us/step - loss: 0.0714 - val_loss: 0.1124\n",
- "Epoch 223/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0713 - val_loss: 0.1124\n",
- "Epoch 224/2500\n",
- "990/990 [==============================] - 0s 463us/step - loss: 0.0713 - val_loss: 0.1124\n",
- "Epoch 225/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0712 - val_loss: 0.1124\n",
- "Epoch 226/2500\n",
- "990/990 [==============================] - 0s 458us/step - loss: 0.0711 - val_loss: 0.1123\n",
- "Epoch 227/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0710 - val_loss: 0.1123\n",
- "Epoch 228/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0710 - val_loss: 0.1123\n",
- "Epoch 229/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0709 - val_loss: 0.1122\n",
- "Epoch 230/2500\n",
- "990/990 [==============================] - 0s 459us/step - loss: 0.0708 - val_loss: 0.1122\n",
- "Epoch 231/2500\n",
- "990/990 [==============================] - 0s 480us/step - loss: 0.0708 - val_loss: 0.1122\n",
- "Epoch 232/2500\n"
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0767 - val_loss: 0.1138\n",
+ "Epoch 156/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0766 - val_loss: 0.1138\n",
+ "Epoch 157/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0766 - val_loss: 0.1138\n",
+ "Epoch 158/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0765 - val_loss: 0.1138\n",
+ "Epoch 159/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0764 - val_loss: 0.1138\n",
+ "Epoch 160/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0763 - val_loss: 0.1139\n",
+ "Epoch 161/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0762 - val_loss: 0.1137\n",
+ "Epoch 162/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0762 - val_loss: 0.1137\n",
+ "Epoch 163/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0761 - val_loss: 0.1138\n",
+ "Epoch 164/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0760 - val_loss: 0.1137\n",
+ "Epoch 165/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0759 - val_loss: 0.1137\n",
+ "Epoch 166/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0759 - val_loss: 0.1136\n",
+ "Epoch 167/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0758 - val_loss: 0.1135\n",
+ "Epoch 168/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0757 - val_loss: 0.1136\n",
+ "Epoch 169/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0756 - val_loss: 0.1136\n",
+ "Epoch 170/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0755 - val_loss: 0.1136\n",
+ "Epoch 171/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0755 - val_loss: 0.1136\n",
+ "Epoch 172/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0754 - val_loss: 0.1135\n",
+ "Epoch 173/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0753 - val_loss: 0.1134\n",
+ "Epoch 174/1200\n",
+ "990/990 [==============================] - 0s 427us/step - loss: 0.0752 - val_loss: 0.1134\n",
+ "Epoch 175/1200\n",
+ "990/990 [==============================] - 0s 430us/step - loss: 0.0751 - val_loss: 0.1134\n",
+ "Epoch 176/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0751 - val_loss: 0.1134\n",
+ "Epoch 177/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0750 - val_loss: 0.1134\n",
+ "Epoch 178/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0749 - val_loss: 0.1134\n",
+ "Epoch 179/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0748 - val_loss: 0.1133\n",
+ "Epoch 180/1200\n",
+ "990/990 [==============================] - 0s 434us/step - loss: 0.0748 - val_loss: 0.1132\n",
+ "Epoch 181/1200\n",
+ "990/990 [==============================] - 0s 439us/step - loss: 0.0747 - val_loss: 0.1132\n",
+ "Epoch 182/1200\n",
+ "990/990 [==============================] - 0s 440us/step - loss: 0.0746 - val_loss: 0.1133\n",
+ "Epoch 183/1200\n",
+ "990/990 [==============================] - 0s 427us/step - loss: 0.0745 - val_loss: 0.1133\n",
+ "Epoch 184/1200\n",
+ "990/990 [==============================] - 0s 437us/step - loss: 0.0744 - val_loss: 0.1133\n",
+ "Epoch 185/1200\n",
+ "990/990 [==============================] - 0s 430us/step - loss: 0.0744 - val_loss: 0.1132\n",
+ "Epoch 186/1200\n",
+ "990/990 [==============================] - 0s 434us/step - loss: 0.0743 - val_loss: 0.1132\n",
+ "Epoch 187/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0742 - val_loss: 0.1132\n",
+ "Epoch 188/1200\n",
+ "990/990 [==============================] - 0s 451us/step - loss: 0.0741 - val_loss: 0.1131\n",
+ "Epoch 189/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0740 - val_loss: 0.1131\n",
+ "Epoch 190/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0740 - val_loss: 0.1131\n",
+ "Epoch 191/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0739 - val_loss: 0.1132\n",
+ "Epoch 192/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0738 - val_loss: 0.1131\n",
+ "Epoch 193/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0737 - val_loss: 0.1131\n",
+ "Epoch 194/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0736 - val_loss: 0.1131\n",
+ "Epoch 195/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0736 - val_loss: 0.1130\n",
+ "Epoch 196/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0735 - val_loss: 0.1131\n",
+ "Epoch 197/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0734 - val_loss: 0.1131\n",
+ "Epoch 198/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0733 - val_loss: 0.1130\n",
+ "Epoch 199/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0732 - val_loss: 0.1130\n",
+ "Epoch 200/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0732 - val_loss: 0.1130\n",
+ "Epoch 201/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0731 - val_loss: 0.1129\n",
+ "Epoch 202/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0730 - val_loss: 0.1129\n",
+ "Epoch 203/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0729 - val_loss: 0.1129\n",
+ "Epoch 204/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0728 - val_loss: 0.1128\n",
+ "Epoch 205/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0728 - val_loss: 0.1128\n",
+ "Epoch 206/1200\n",
+ "990/990 [==============================] - 0s 427us/step - loss: 0.0727 - val_loss: 0.1127\n",
+ "Epoch 207/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0726 - val_loss: 0.1128\n",
+ "Epoch 208/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0725 - val_loss: 0.1128\n",
+ "Epoch 209/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0724 - val_loss: 0.1127\n",
+ "Epoch 210/1200\n",
+ "990/990 [==============================] - 0s 432us/step - loss: 0.0724 - val_loss: 0.1127\n",
+ "Epoch 211/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0723 - val_loss: 0.1127\n",
+ "Epoch 212/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0722 - val_loss: 0.1127\n",
+ "Epoch 213/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0721 - val_loss: 0.1127\n",
+ "Epoch 214/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0720 - val_loss: 0.1127\n",
+ "Epoch 215/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0720 - val_loss: 0.1127\n",
+ "Epoch 216/1200\n",
+ "990/990 [==============================] - 0s 430us/step - loss: 0.0719 - val_loss: 0.1126\n",
+ "Epoch 217/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0718 - val_loss: 0.1126\n",
+ "Epoch 218/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0717 - val_loss: 0.1126\n",
+ "Epoch 219/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0717 - val_loss: 0.1124\n",
+ "Epoch 220/1200\n",
+ "990/990 [==============================] - 0s 417us/step - loss: 0.0716 - val_loss: 0.1125\n",
+ "Epoch 221/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0715 - val_loss: 0.1125\n",
+ "Epoch 222/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0714 - val_loss: 0.1124\n",
+ "Epoch 223/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0713 - val_loss: 0.1124\n",
+ "Epoch 224/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0713 - val_loss: 0.1124\n",
+ "Epoch 225/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0712 - val_loss: 0.1124\n",
+ "Epoch 226/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0711 - val_loss: 0.1123\n",
+ "Epoch 227/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0710 - val_loss: 0.1123\n",
+ "Epoch 228/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0710 - val_loss: 0.1123\n",
+ "Epoch 229/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0709 - val_loss: 0.1122\n",
+ "Epoch 230/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0708 - val_loss: 0.1122\n",
+ "Epoch 231/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0708 - val_loss: 0.1122\n",
+ "Epoch 232/1200\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
- "990/990 [==============================] - 0s 462us/step - loss: 0.0707 - val_loss: 0.1122\n",
- "Epoch 233/2500\n",
- "990/990 [==============================] - 0s 477us/step - loss: 0.0706 - val_loss: 0.1121\n",
- "Epoch 234/2500\n",
- "990/990 [==============================] - 0s 497us/step - loss: 0.0705 - val_loss: 0.1122\n",
- "Epoch 235/2500\n",
- "990/990 [==============================] - 0s 484us/step - loss: 0.0705 - val_loss: 0.1121\n",
- "Epoch 236/2500\n",
- "990/990 [==============================] - 0s 465us/step - loss: 0.0704 - val_loss: 0.1121\n",
- "Epoch 237/2500\n",
- "990/990 [==============================] - 0s 456us/step - loss: 0.0703 - val_loss: 0.1120\n",
- "Epoch 238/2500\n",
- "990/990 [==============================] - 0s 436us/step - loss: 0.0703 - val_loss: 0.1120\n",
- "Epoch 239/2500\n",
- "990/990 [==============================] - 0s 464us/step - loss: 0.0702 - val_loss: 0.1120\n",
- "Epoch 240/2500\n",
- "990/990 [==============================] - 0s 470us/step - loss: 0.0701 - val_loss: 0.1119\n",
- "Epoch 241/2500\n",
- "990/990 [==============================] - 0s 437us/step - loss: 0.0700 - val_loss: 0.1119\n",
- "Epoch 242/2500\n",
- "990/990 [==============================] - 0s 474us/step - loss: 0.0700 - val_loss: 0.1120\n",
- "Epoch 243/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0699 - val_loss: 0.1119\n",
- "Epoch 244/2500\n",
- "990/990 [==============================] - 0s 464us/step - loss: 0.0698 - val_loss: 0.1119\n",
- "Epoch 245/2500\n",
- "990/990 [==============================] - 0s 472us/step - loss: 0.0698 - val_loss: 0.1119\n",
- "Epoch 246/2500\n",
- "990/990 [==============================] - 0s 439us/step - loss: 0.0697 - val_loss: 0.1118\n",
- "Epoch 247/2500\n",
- "990/990 [==============================] - 0s 437us/step - loss: 0.0696 - val_loss: 0.1117\n",
- "Epoch 248/2500\n",
- "990/990 [==============================] - 0s 437us/step - loss: 0.0696 - val_loss: 0.1117\n",
- "Epoch 249/2500\n",
- "990/990 [==============================] - 0s 444us/step - loss: 0.0695 - val_loss: 0.1117\n",
- "Epoch 250/2500\n",
- "990/990 [==============================] - 0s 438us/step - loss: 0.0695 - val_loss: 0.1117\n",
- "Epoch 251/2500\n",
- "990/990 [==============================] - 0s 441us/step - loss: 0.0694 - val_loss: 0.1117\n",
- "Epoch 252/2500\n",
- "990/990 [==============================] - 0s 444us/step - loss: 0.0693 - val_loss: 0.1117\n",
- "Epoch 253/2500\n",
- "990/990 [==============================] - 0s 438us/step - loss: 0.0693 - val_loss: 0.1116\n",
- "Epoch 254/2500\n",
- "990/990 [==============================] - 0s 445us/step - loss: 0.0692 - val_loss: 0.1116\n",
- "Epoch 255/2500\n",
- "990/990 [==============================] - 0s 439us/step - loss: 0.0691 - val_loss: 0.1116\n",
- "Epoch 256/2500\n",
- "990/990 [==============================] - 0s 439us/step - loss: 0.0691 - val_loss: 0.1115\n",
- "Epoch 257/2500\n",
- "990/990 [==============================] - 0s 446us/step - loss: 0.0690 - val_loss: 0.1114\n",
- "Epoch 258/2500\n",
- "990/990 [==============================] - 0s 438us/step - loss: 0.0690 - val_loss: 0.1114\n",
- "Epoch 259/2500\n",
- "990/990 [==============================] - 0s 448us/step - loss: 0.0689 - val_loss: 0.1114\n",
- "Epoch 260/2500\n",
- "990/990 [==============================] - 0s 440us/step - loss: 0.0689 - val_loss: 0.1114\n",
- "Epoch 261/2500\n",
- "990/990 [==============================] - 0s 444us/step - loss: 0.0688 - val_loss: 0.1114\n",
- "Epoch 262/2500\n",
- "990/990 [==============================] - 0s 485us/step - loss: 0.0687 - val_loss: 0.1114\n",
- "Epoch 263/2500\n",
- "990/990 [==============================] - 0s 444us/step - loss: 0.0687 - val_loss: 0.1113\n",
- "Epoch 264/2500\n",
- "990/990 [==============================] - 0s 442us/step - loss: 0.0686 - val_loss: 0.1113\n",
- "Epoch 265/2500\n",
- "990/990 [==============================] - 0s 439us/step - loss: 0.0686 - val_loss: 0.1113\n",
- "Epoch 266/2500\n",
- "990/990 [==============================] - 0s 443us/step - loss: 0.0685 - val_loss: 0.1113\n",
- "Epoch 267/2500\n",
- "990/990 [==============================] - 0s 441us/step - loss: 0.0685 - val_loss: 0.1112\n",
- "Epoch 268/2500\n",
- "990/990 [==============================] - 0s 443us/step - loss: 0.0684 - val_loss: 0.1112\n",
- "Epoch 269/2500\n",
- "990/990 [==============================] - 0s 441us/step - loss: 0.0684 - val_loss: 0.1111\n",
- "Epoch 270/2500\n",
- "990/990 [==============================] - 0s 443us/step - loss: 0.0683 - val_loss: 0.1111\n",
- "Epoch 271/2500\n",
- "990/990 [==============================] - 0s 438us/step - loss: 0.0682 - val_loss: 0.1110\n",
- "Epoch 272/2500\n",
- "990/990 [==============================] - 0s 440us/step - loss: 0.0682 - val_loss: 0.1110\n",
- "Epoch 273/2500\n",
- "990/990 [==============================] - 0s 445us/step - loss: 0.0681 - val_loss: 0.1110\n",
- "Epoch 274/2500\n",
- "990/990 [==============================] - 0s 440us/step - loss: 0.0681 - val_loss: 0.1111\n",
- "Epoch 275/2500\n",
- "990/990 [==============================] - 0s 438us/step - loss: 0.0680 - val_loss: 0.1110\n",
- "Epoch 276/2500\n",
- "990/990 [==============================] - 0s 445us/step - loss: 0.0680 - val_loss: 0.1110\n",
- "Epoch 277/2500\n",
- "990/990 [==============================] - 0s 439us/step - loss: 0.0679 - val_loss: 0.1110\n",
- "Epoch 278/2500\n",
- "990/990 [==============================] - 0s 441us/step - loss: 0.0679 - val_loss: 0.1109\n",
- "Epoch 279/2500\n",
- "990/990 [==============================] - 0s 440us/step - loss: 0.0678 - val_loss: 0.1109\n",
- "Epoch 280/2500\n",
- "990/990 [==============================] - 0s 447us/step - loss: 0.0678 - val_loss: 0.1108\n",
- "Epoch 281/2500\n",
- "990/990 [==============================] - 0s 445us/step - loss: 0.0677 - val_loss: 0.1108\n",
- "Epoch 282/2500\n",
- "990/990 [==============================] - 0s 443us/step - loss: 0.0677 - val_loss: 0.1108\n",
- "Epoch 283/2500\n",
- "990/990 [==============================] - 0s 443us/step - loss: 0.0677 - val_loss: 0.1107\n",
- "Epoch 284/2500\n",
- "990/990 [==============================] - 0s 443us/step - loss: 0.0676 - val_loss: 0.1107\n",
- "Epoch 285/2500\n",
- "990/990 [==============================] - 0s 439us/step - loss: 0.0676 - val_loss: 0.1107\n",
- "Epoch 286/2500\n",
- "990/990 [==============================] - 0s 439us/step - loss: 0.0675 - val_loss: 0.1106\n",
- "Epoch 287/2500\n",
- "990/990 [==============================] - 0s 436us/step - loss: 0.0675 - val_loss: 0.1107\n",
- "Epoch 288/2500\n",
- "990/990 [==============================] - 0s 439us/step - loss: 0.0674 - val_loss: 0.1106\n",
- "Epoch 289/2500\n",
- "990/990 [==============================] - 0s 442us/step - loss: 0.0674 - val_loss: 0.1106\n",
- "Epoch 290/2500\n",
- "990/990 [==============================] - 0s 441us/step - loss: 0.0673 - val_loss: 0.1105\n",
- "Epoch 291/2500\n",
- "990/990 [==============================] - 0s 441us/step - loss: 0.0673 - val_loss: 0.1105\n",
- "Epoch 292/2500\n",
- "990/990 [==============================] - 0s 442us/step - loss: 0.0672 - val_loss: 0.1104\n",
- "Epoch 293/2500\n",
- "990/990 [==============================] - 0s 436us/step - loss: 0.0672 - val_loss: 0.1104\n",
- "Epoch 294/2500\n",
- "990/990 [==============================] - 0s 447us/step - loss: 0.0672 - val_loss: 0.1104\n",
- "Epoch 295/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0671 - val_loss: 0.1104\n",
- "Epoch 296/2500\n",
- "990/990 [==============================] - 0s 447us/step - loss: 0.0671 - val_loss: 0.1104\n",
- "Epoch 297/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0670 - val_loss: 0.1104\n",
- "Epoch 298/2500\n",
- "990/990 [==============================] - 0s 460us/step - loss: 0.0670 - val_loss: 0.1103\n",
- "Epoch 299/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0670 - val_loss: 0.1103\n",
- "Epoch 300/2500\n",
- "990/990 [==============================] - 0s 423us/step - loss: 0.0669 - val_loss: 0.1102\n",
- "Epoch 301/2500\n",
- "990/990 [==============================] - 0s 432us/step - loss: 0.0669 - val_loss: 0.1102\n",
- "Epoch 302/2500\n",
- "990/990 [==============================] - 0s 441us/step - loss: 0.0668 - val_loss: 0.1102\n",
- "Epoch 303/2500\n",
- "990/990 [==============================] - 0s 437us/step - loss: 0.0668 - val_loss: 0.1102\n",
- "Epoch 304/2500\n",
- "990/990 [==============================] - 0s 436us/step - loss: 0.0667 - val_loss: 0.1101\n",
- "Epoch 305/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0667 - val_loss: 0.1101\n",
- "Epoch 306/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0667 - val_loss: 0.1101\n",
- "Epoch 307/2500\n",
- "990/990 [==============================] - 0s 443us/step - loss: 0.0666 - val_loss: 0.1102\n",
- "Epoch 308/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0666 - val_loss: 0.1101\n",
- "Epoch 309/2500\n"
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0707 - val_loss: 0.1122\n",
+ "Epoch 233/1200\n",
+ "990/990 [==============================] - 0s 427us/step - loss: 0.0706 - val_loss: 0.1121\n",
+ "Epoch 234/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0705 - val_loss: 0.1122\n",
+ "Epoch 235/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0705 - val_loss: 0.1121\n",
+ "Epoch 236/1200\n",
+ "990/990 [==============================] - 0s 430us/step - loss: 0.0704 - val_loss: 0.1121\n",
+ "Epoch 237/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0703 - val_loss: 0.1120\n",
+ "Epoch 238/1200\n",
+ "990/990 [==============================] - 0s 441us/step - loss: 0.0703 - val_loss: 0.1120\n",
+ "Epoch 239/1200\n",
+ "990/990 [==============================] - 0s 430us/step - loss: 0.0702 - val_loss: 0.1120\n",
+ "Epoch 240/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0701 - val_loss: 0.1119\n",
+ "Epoch 241/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0700 - val_loss: 0.1119\n",
+ "Epoch 242/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0700 - val_loss: 0.1120\n",
+ "Epoch 243/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0699 - val_loss: 0.1119\n",
+ "Epoch 244/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0698 - val_loss: 0.1119\n",
+ "Epoch 245/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0698 - val_loss: 0.1119\n",
+ "Epoch 246/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0697 - val_loss: 0.1118\n",
+ "Epoch 247/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0696 - val_loss: 0.1117\n",
+ "Epoch 248/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0696 - val_loss: 0.1117\n",
+ "Epoch 249/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0695 - val_loss: 0.1117\n",
+ "Epoch 250/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0695 - val_loss: 0.1117\n",
+ "Epoch 251/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0694 - val_loss: 0.1117\n",
+ "Epoch 252/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0693 - val_loss: 0.1117\n",
+ "Epoch 253/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0693 - val_loss: 0.1116\n",
+ "Epoch 254/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0692 - val_loss: 0.1116\n",
+ "Epoch 255/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0691 - val_loss: 0.1116\n",
+ "Epoch 256/1200\n",
+ "990/990 [==============================] - 0s 430us/step - loss: 0.0691 - val_loss: 0.1115\n",
+ "Epoch 257/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0690 - val_loss: 0.1114\n",
+ "Epoch 258/1200\n",
+ "990/990 [==============================] - 0s 433us/step - loss: 0.0690 - val_loss: 0.1114\n",
+ "Epoch 259/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0689 - val_loss: 0.1114\n",
+ "Epoch 260/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0689 - val_loss: 0.1114\n",
+ "Epoch 261/1200\n",
+ "990/990 [==============================] - 0s 430us/step - loss: 0.0688 - val_loss: 0.1114\n",
+ "Epoch 262/1200\n",
+ "990/990 [==============================] - 0s 427us/step - loss: 0.0687 - val_loss: 0.1114\n",
+ "Epoch 263/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0687 - val_loss: 0.1113\n",
+ "Epoch 264/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0686 - val_loss: 0.1113\n",
+ "Epoch 265/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0686 - val_loss: 0.1113\n",
+ "Epoch 266/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0685 - val_loss: 0.1113\n",
+ "Epoch 267/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0685 - val_loss: 0.1112\n",
+ "Epoch 268/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0684 - val_loss: 0.1112\n",
+ "Epoch 269/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0684 - val_loss: 0.1111\n",
+ "Epoch 270/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0683 - val_loss: 0.1111\n",
+ "Epoch 271/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0682 - val_loss: 0.1110\n",
+ "Epoch 272/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0682 - val_loss: 0.1110\n",
+ "Epoch 273/1200\n",
+ "990/990 [==============================] - 0s 427us/step - loss: 0.0681 - val_loss: 0.1110\n",
+ "Epoch 274/1200\n",
+ "990/990 [==============================] - 0s 429us/step - loss: 0.0681 - val_loss: 0.1111\n",
+ "Epoch 275/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0680 - val_loss: 0.1110\n",
+ "Epoch 276/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0680 - val_loss: 0.1110\n",
+ "Epoch 277/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0679 - val_loss: 0.1110\n",
+ "Epoch 278/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0679 - val_loss: 0.1109\n",
+ "Epoch 279/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0678 - val_loss: 0.1109\n",
+ "Epoch 280/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0678 - val_loss: 0.1108\n",
+ "Epoch 281/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0677 - val_loss: 0.1108\n",
+ "Epoch 282/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0677 - val_loss: 0.1108\n",
+ "Epoch 283/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0677 - val_loss: 0.1107\n",
+ "Epoch 284/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0676 - val_loss: 0.1107\n",
+ "Epoch 285/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0676 - val_loss: 0.1107\n",
+ "Epoch 286/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0675 - val_loss: 0.1106\n",
+ "Epoch 287/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0675 - val_loss: 0.1107\n",
+ "Epoch 288/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0674 - val_loss: 0.1106\n",
+ "Epoch 289/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0674 - val_loss: 0.1106\n",
+ "Epoch 290/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0673 - val_loss: 0.1105\n",
+ "Epoch 291/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0673 - val_loss: 0.1105\n",
+ "Epoch 292/1200\n",
+ "990/990 [==============================] - 0s 427us/step - loss: 0.0672 - val_loss: 0.1104\n",
+ "Epoch 293/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0672 - val_loss: 0.1104\n",
+ "Epoch 294/1200\n",
+ "990/990 [==============================] - 0s 417us/step - loss: 0.0672 - val_loss: 0.1104\n",
+ "Epoch 295/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0671 - val_loss: 0.1104\n",
+ "Epoch 296/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0671 - val_loss: 0.1104\n",
+ "Epoch 297/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0670 - val_loss: 0.1104\n",
+ "Epoch 298/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0670 - val_loss: 0.1103\n",
+ "Epoch 299/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0670 - val_loss: 0.1103\n",
+ "Epoch 300/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0669 - val_loss: 0.1102\n",
+ "Epoch 301/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0669 - val_loss: 0.1102\n",
+ "Epoch 302/1200\n",
+ "990/990 [==============================] - 0s 430us/step - loss: 0.0668 - val_loss: 0.1102\n",
+ "Epoch 303/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0668 - val_loss: 0.1102\n",
+ "Epoch 304/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0667 - val_loss: 0.1101\n",
+ "Epoch 305/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0667 - val_loss: 0.1101\n",
+ "Epoch 306/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0667 - val_loss: 0.1101\n",
+ "Epoch 307/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0666 - val_loss: 0.1102\n",
+ "Epoch 308/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0666 - val_loss: 0.1101\n",
+ "Epoch 309/1200\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
- "990/990 [==============================] - 0s 430us/step - loss: 0.0666 - val_loss: 0.1100\n",
- "Epoch 310/2500\n",
- "990/990 [==============================] - 0s 447us/step - loss: 0.0665 - val_loss: 0.1100\n",
- "Epoch 311/2500\n",
- "990/990 [==============================] - 0s 426us/step - loss: 0.0665 - val_loss: 0.1099\n",
- "Epoch 312/2500\n",
- "990/990 [==============================] - 0s 432us/step - loss: 0.0664 - val_loss: 0.1099\n",
- "Epoch 313/2500\n",
- "990/990 [==============================] - 0s 448us/step - loss: 0.0664 - val_loss: 0.1098\n",
- "Epoch 314/2500\n",
- "990/990 [==============================] - 0s 432us/step - loss: 0.0664 - val_loss: 0.1099\n",
- "Epoch 315/2500\n",
- "990/990 [==============================] - 0s 433us/step - loss: 0.0663 - val_loss: 0.1098\n",
- "Epoch 316/2500\n",
- "990/990 [==============================] - 0s 430us/step - loss: 0.0663 - val_loss: 0.1098\n",
- "Epoch 317/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0663 - val_loss: 0.1097\n",
- "Epoch 318/2500\n",
- "990/990 [==============================] - 0s 434us/step - loss: 0.0662 - val_loss: 0.1098\n",
- "Epoch 319/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0662 - val_loss: 0.1097\n",
- "Epoch 320/2500\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0666 - val_loss: 0.1100\n",
+ "Epoch 310/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0665 - val_loss: 0.1100\n",
+ "Epoch 311/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0665 - val_loss: 0.1099\n",
+ "Epoch 312/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0664 - val_loss: 0.1099\n",
+ "Epoch 313/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0664 - val_loss: 0.1098\n",
+ "Epoch 314/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0664 - val_loss: 0.1099\n",
+ "Epoch 315/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0663 - val_loss: 0.1098\n",
+ "Epoch 316/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0663 - val_loss: 0.1098\n",
+ "Epoch 317/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0663 - val_loss: 0.1097\n",
+ "Epoch 318/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0662 - val_loss: 0.1098\n",
+ "Epoch 319/1200\n",
+ "990/990 [==============================] - 0s 427us/step - loss: 0.0662 - val_loss: 0.1097\n",
+ "Epoch 320/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0661 - val_loss: 0.1097\n",
+ "Epoch 321/1200\n",
"990/990 [==============================] - 0s 429us/step - loss: 0.0661 - val_loss: 0.1097\n",
- "Epoch 321/2500\n",
- "990/990 [==============================] - 0s 440us/step - loss: 0.0661 - val_loss: 0.1097\n",
- "Epoch 322/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0661 - val_loss: 0.1097\n",
- "Epoch 323/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0660 - val_loss: 0.1096\n",
- "Epoch 324/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0660 - val_loss: 0.1096\n",
- "Epoch 325/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0660 - val_loss: 0.1095\n",
- "Epoch 326/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0659 - val_loss: 0.1095\n",
- "Epoch 327/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0659 - val_loss: 0.1094\n",
- "Epoch 328/2500\n",
- "990/990 [==============================] - 0s 433us/step - loss: 0.0659 - val_loss: 0.1095\n",
- "Epoch 329/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0658 - val_loss: 0.1095\n",
- "Epoch 330/2500\n",
- "990/990 [==============================] - 0s 430us/step - loss: 0.0658 - val_loss: 0.1094\n",
- "Epoch 331/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0658 - val_loss: 0.1093\n",
- "Epoch 332/2500\n",
- "990/990 [==============================] - 0s 426us/step - loss: 0.0657 - val_loss: 0.1093\n",
- "Epoch 333/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0657 - val_loss: 0.1093\n",
- "Epoch 334/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0657 - val_loss: 0.1093\n",
- "Epoch 335/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0656 - val_loss: 0.1093\n",
- "Epoch 336/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0656 - val_loss: 0.1092\n",
- "Epoch 337/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0656 - val_loss: 0.1092\n",
- "Epoch 338/2500\n",
- "990/990 [==============================] - 0s 426us/step - loss: 0.0655 - val_loss: 0.1091\n",
- "Epoch 339/2500\n",
- "990/990 [==============================] - 0s 426us/step - loss: 0.0655 - val_loss: 0.1091\n",
- "Epoch 340/2500\n",
- "990/990 [==============================] - 0s 426us/step - loss: 0.0655 - val_loss: 0.1091\n",
- "Epoch 341/2500\n",
- "990/990 [==============================] - 0s 426us/step - loss: 0.0654 - val_loss: 0.1091\n",
- "Epoch 342/2500\n",
- "990/990 [==============================] - 0s 434us/step - loss: 0.0654 - val_loss: 0.1091\n",
- "Epoch 343/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0654 - val_loss: 0.1090\n",
- "Epoch 344/2500\n",
- "990/990 [==============================] - 0s 434us/step - loss: 0.0653 - val_loss: 0.1090\n",
- "Epoch 345/2500\n",
- "990/990 [==============================] - 0s 424us/step - loss: 0.0653 - val_loss: 0.1089\n",
- "Epoch 346/2500\n",
- "990/990 [==============================] - 0s 423us/step - loss: 0.0653 - val_loss: 0.1089\n",
- "Epoch 347/2500\n",
- "990/990 [==============================] - 0s 421us/step - loss: 0.0652 - val_loss: 0.1090\n",
- "Epoch 348/2500\n",
- "990/990 [==============================] - 0s 433us/step - loss: 0.0652 - val_loss: 0.1089\n",
- "Epoch 349/2500\n",
- "990/990 [==============================] - 0s 426us/step - loss: 0.0652 - val_loss: 0.1089\n",
- "Epoch 350/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0651 - val_loss: 0.1089\n",
- "Epoch 351/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0651 - val_loss: 0.1088\n",
- "Epoch 352/2500\n",
- "990/990 [==============================] - 0s 432us/step - loss: 0.0651 - val_loss: 0.1088\n",
- "Epoch 353/2500\n",
- "990/990 [==============================] - 0s 430us/step - loss: 0.0651 - val_loss: 0.1088\n",
- "Epoch 354/2500\n",
- "990/990 [==============================] - 0s 430us/step - loss: 0.0650 - val_loss: 0.1088\n",
- "Epoch 355/2500\n",
- "990/990 [==============================] - 0s 423us/step - loss: 0.0650 - val_loss: 0.1087\n",
- "Epoch 356/2500\n",
- "990/990 [==============================] - 0s 423us/step - loss: 0.0650 - val_loss: 0.1087\n",
- "Epoch 357/2500\n",
+ "Epoch 322/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0661 - val_loss: 0.1097\n",
+ "Epoch 323/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0660 - val_loss: 0.1096\n",
+ "Epoch 324/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0660 - val_loss: 0.1096\n",
+ "Epoch 325/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0660 - val_loss: 0.1095\n",
+ "Epoch 326/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0659 - val_loss: 0.1095\n",
+ "Epoch 327/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0659 - val_loss: 0.1094\n",
+ "Epoch 328/1200\n",
+ "990/990 [==============================] - 0s 429us/step - loss: 0.0659 - val_loss: 0.1095\n",
+ "Epoch 329/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0658 - val_loss: 0.1095\n",
+ "Epoch 330/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0658 - val_loss: 0.1094\n",
+ "Epoch 331/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0658 - val_loss: 0.1093\n",
+ "Epoch 332/1200\n",
+ "990/990 [==============================] - 0s 427us/step - loss: 0.0657 - val_loss: 0.1093\n",
+ "Epoch 333/1200\n",
+ "990/990 [==============================] - 0s 427us/step - loss: 0.0657 - val_loss: 0.1093\n",
+ "Epoch 334/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0657 - val_loss: 0.1093\n",
+ "Epoch 335/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0656 - val_loss: 0.1093\n",
+ "Epoch 336/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0656 - val_loss: 0.1092\n",
+ "Epoch 337/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0656 - val_loss: 0.1092\n",
+ "Epoch 338/1200\n",
+ "990/990 [==============================] - 0s 427us/step - loss: 0.0655 - val_loss: 0.1091\n",
+ "Epoch 339/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0655 - val_loss: 0.1091\n",
+ "Epoch 340/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0655 - val_loss: 0.1091\n",
+ "Epoch 341/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0654 - val_loss: 0.1091\n",
+ "Epoch 342/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0654 - val_loss: 0.1091\n",
+ "Epoch 343/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0654 - val_loss: 0.1090\n",
+ "Epoch 344/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0653 - val_loss: 0.1090\n",
+ "Epoch 345/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0653 - val_loss: 0.1089\n",
+ "Epoch 346/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0653 - val_loss: 0.1089\n",
+ "Epoch 347/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0652 - val_loss: 0.1090\n",
+ "Epoch 348/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0652 - val_loss: 0.1089\n",
+ "Epoch 349/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0652 - val_loss: 0.1089\n",
+ "Epoch 350/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0651 - val_loss: 0.1089\n",
+ "Epoch 351/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0651 - val_loss: 0.1088\n",
+ "Epoch 352/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0651 - val_loss: 0.1088\n",
+ "Epoch 353/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0651 - val_loss: 0.1088\n",
+ "Epoch 354/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0650 - val_loss: 0.1088\n",
+ "Epoch 355/1200\n",
+ "990/990 [==============================] - 0s 417us/step - loss: 0.0650 - val_loss: 0.1087\n",
+ "Epoch 356/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0650 - val_loss: 0.1087\n",
+ "Epoch 357/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0649 - val_loss: 0.1086\n",
+ "Epoch 358/1200\n",
"990/990 [==============================] - 0s 425us/step - loss: 0.0649 - val_loss: 0.1086\n",
- "Epoch 358/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0649 - val_loss: 0.1086\n",
- "Epoch 359/2500\n",
- "990/990 [==============================] - 0s 432us/step - loss: 0.0649 - val_loss: 0.1086\n",
- "Epoch 360/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0648 - val_loss: 0.1086\n",
- "Epoch 361/2500\n",
- "990/990 [==============================] - 0s 432us/step - loss: 0.0648 - val_loss: 0.1085\n",
- "Epoch 362/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0648 - val_loss: 0.1086\n",
- "Epoch 363/2500\n",
- "990/990 [==============================] - 0s 435us/step - loss: 0.0648 - val_loss: 0.1085\n",
- "Epoch 364/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0647 - val_loss: 0.1085\n",
- "Epoch 365/2500\n",
+ "Epoch 359/1200\n",
+ "990/990 [==============================] - 0s 427us/step - loss: 0.0649 - val_loss: 0.1086\n",
+ "Epoch 360/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0648 - val_loss: 0.1086\n",
+ "Epoch 361/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0648 - val_loss: 0.1085\n",
+ "Epoch 362/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0648 - val_loss: 0.1086\n",
+ "Epoch 363/1200\n",
+ "990/990 [==============================] - 0s 417us/step - loss: 0.0648 - val_loss: 0.1085\n",
+ "Epoch 364/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0647 - val_loss: 0.1085\n",
+ "Epoch 365/1200\n",
"990/990 [==============================] - 0s 425us/step - loss: 0.0647 - val_loss: 0.1084\n",
- "Epoch 366/2500\n",
- "990/990 [==============================] - 0s 431us/step - loss: 0.0647 - val_loss: 0.1084\n",
- "Epoch 367/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0646 - val_loss: 0.1084\n",
- "Epoch 368/2500\n",
- "990/990 [==============================] - 0s 432us/step - loss: 0.0646 - val_loss: 0.1084\n",
- "Epoch 369/2500\n",
- "990/990 [==============================] - 0s 430us/step - loss: 0.0646 - val_loss: 0.1084\n",
- "Epoch 370/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0646 - val_loss: 0.1084\n",
- "Epoch 371/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0645 - val_loss: 0.1083\n",
- "Epoch 372/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0645 - val_loss: 0.1083\n",
- "Epoch 373/2500\n",
- "990/990 [==============================] - 0s 433us/step - loss: 0.0645 - val_loss: 0.1083\n",
- "Epoch 374/2500\n",
+ "Epoch 366/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0647 - val_loss: 0.1084\n",
+ "Epoch 367/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0646 - val_loss: 0.1084\n",
+ "Epoch 368/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0646 - val_loss: 0.1084\n",
+ "Epoch 369/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0646 - val_loss: 0.1084\n",
+ "Epoch 370/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0646 - val_loss: 0.1084\n",
+ "Epoch 371/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0645 - val_loss: 0.1083\n",
+ "Epoch 372/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0645 - val_loss: 0.1083\n",
+ "Epoch 373/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0645 - val_loss: 0.1083\n",
+ "Epoch 374/1200\n",
"990/990 [==============================] - 0s 425us/step - loss: 0.0644 - val_loss: 0.1083\n",
- "Epoch 375/2500\n",
+ "Epoch 375/1200\n",
"990/990 [==============================] - 0s 421us/step - loss: 0.0644 - val_loss: 0.1082\n",
- "Epoch 376/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0644 - val_loss: 0.1082\n",
- "Epoch 377/2500\n",
- "990/990 [==============================] - 0s 434us/step - loss: 0.0644 - val_loss: 0.1082\n",
- "Epoch 378/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0643 - val_loss: 0.1082\n",
- "Epoch 379/2500\n",
- "990/990 [==============================] - 0s 434us/step - loss: 0.0643 - val_loss: 0.1081\n",
- "Epoch 380/2500\n",
- "990/990 [==============================] - 0s 424us/step - loss: 0.0643 - val_loss: 0.1081\n",
- "Epoch 381/2500\n",
- "990/990 [==============================] - 0s 426us/step - loss: 0.0643 - val_loss: 0.1080\n",
- "Epoch 382/2500\n",
- "990/990 [==============================] - 0s 431us/step - loss: 0.0642 - val_loss: 0.1080\n",
- "Epoch 383/2500\n",
- "990/990 [==============================] - 0s 425us/step - loss: 0.0642 - val_loss: 0.1080\n",
- "Epoch 384/2500\n",
- "990/990 [==============================] - 0s 430us/step - loss: 0.0642 - val_loss: 0.1080\n",
- "Epoch 385/2500\n",
- "990/990 [==============================] - 0s 431us/step - loss: 0.0641 - val_loss: 0.1079\n",
- "Epoch 386/2500\n"
+ "Epoch 376/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0644 - val_loss: 0.1082\n",
+ "Epoch 377/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0644 - val_loss: 0.1082\n",
+ "Epoch 378/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0643 - val_loss: 0.1082\n",
+ "Epoch 379/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0643 - val_loss: 0.1081\n",
+ "Epoch 380/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0643 - val_loss: 0.1081\n",
+ "Epoch 381/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0643 - val_loss: 0.1080\n",
+ "Epoch 382/1200\n",
+ "990/990 [==============================] - 0s 427us/step - loss: 0.0642 - val_loss: 0.1080\n",
+ "Epoch 383/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0642 - val_loss: 0.1080\n",
+ "Epoch 384/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0642 - val_loss: 0.1080\n",
+ "Epoch 385/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0641 - val_loss: 0.1079\n",
+ "Epoch 386/1200\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
- "990/990 [==============================] - 0s 425us/step - loss: 0.0641 - val_loss: 0.1079\n",
- "Epoch 387/2500\n",
- "990/990 [==============================] - 0s 423us/step - loss: 0.0641 - val_loss: 0.1078\n",
- "Epoch 388/2500\n",
- "990/990 [==============================] - 0s 421us/step - loss: 0.0641 - val_loss: 0.1078\n",
- "Epoch 389/2500\n",
- "990/990 [==============================] - 0s 424us/step - loss: 0.0640 - val_loss: 0.1078\n",
- "Epoch 390/2500\n",
- "990/990 [==============================] - 0s 423us/step - loss: 0.0640 - val_loss: 0.1078\n",
- "Epoch 391/2500\n",
- "990/990 [==============================] - 0s 425us/step - loss: 0.0640 - val_loss: 0.1078\n",
- "Epoch 392/2500\n",
- "990/990 [==============================] - 0s 433us/step - loss: 0.0640 - val_loss: 0.1078\n",
- "Epoch 393/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0639 - val_loss: 0.1077\n",
- "Epoch 394/2500\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0641 - val_loss: 0.1079\n",
+ "Epoch 387/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0641 - val_loss: 0.1078\n",
+ "Epoch 388/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0641 - val_loss: 0.1078\n",
+ "Epoch 389/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0640 - val_loss: 0.1078\n",
+ "Epoch 390/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0640 - val_loss: 0.1078\n",
+ "Epoch 391/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0640 - val_loss: 0.1078\n",
+ "Epoch 392/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0640 - val_loss: 0.1078\n",
+ "Epoch 393/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0639 - val_loss: 0.1077\n",
+ "Epoch 394/1200\n",
"990/990 [==============================] - 0s 422us/step - loss: 0.0639 - val_loss: 0.1077\n",
- "Epoch 395/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0639 - val_loss: 0.1077\n",
- "Epoch 396/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0639 - val_loss: 0.1077\n",
- "Epoch 397/2500\n",
- "990/990 [==============================] - 0s 430us/step - loss: 0.0638 - val_loss: 0.1076\n",
- "Epoch 398/2500\n",
- "990/990 [==============================] - 0s 436us/step - loss: 0.0638 - val_loss: 0.1076\n",
- "Epoch 399/2500\n",
- "990/990 [==============================] - 0s 445us/step - loss: 0.0638 - val_loss: 0.1076\n",
- "Epoch 400/2500\n",
- "990/990 [==============================] - 0s 430us/step - loss: 0.0638 - val_loss: 0.1076\n",
- "Epoch 401/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0637 - val_loss: 0.1076\n",
- "Epoch 402/2500\n",
- "990/990 [==============================] - 0s 482us/step - loss: 0.0637 - val_loss: 0.1076\n",
- "Epoch 403/2500\n",
- "990/990 [==============================] - 0s 464us/step - loss: 0.0637 - val_loss: 0.1075\n",
- "Epoch 404/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0637 - val_loss: 0.1075\n",
- "Epoch 405/2500\n",
- "990/990 [==============================] - 0s 469us/step - loss: 0.0636 - val_loss: 0.1074\n",
- "Epoch 406/2500\n",
- "990/990 [==============================] - 0s 469us/step - loss: 0.0636 - val_loss: 0.1074\n",
- "Epoch 407/2500\n",
- "990/990 [==============================] - 0s 474us/step - loss: 0.0636 - val_loss: 0.1074\n",
- "Epoch 408/2500\n",
- "990/990 [==============================] - 0s 465us/step - loss: 0.0636 - val_loss: 0.1073\n",
- "Epoch 409/2500\n",
- "990/990 [==============================] - 0s 464us/step - loss: 0.0636 - val_loss: 0.1073\n",
- "Epoch 410/2500\n",
- "990/990 [==============================] - 0s 466us/step - loss: 0.0635 - val_loss: 0.1073\n",
- "Epoch 411/2500\n",
- "990/990 [==============================] - 0s 470us/step - loss: 0.0635 - val_loss: 0.1073\n",
- "Epoch 412/2500\n",
- "990/990 [==============================] - 0s 474us/step - loss: 0.0635 - val_loss: 0.1073\n",
- "Epoch 413/2500\n",
- "990/990 [==============================] - 0s 460us/step - loss: 0.0635 - val_loss: 0.1072\n",
- "Epoch 414/2500\n",
- "990/990 [==============================] - 0s 482us/step - loss: 0.0634 - val_loss: 0.1071\n",
- "Epoch 415/2500\n",
- "990/990 [==============================] - 0s 461us/step - loss: 0.0634 - val_loss: 0.1072\n",
- "Epoch 416/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0634 - val_loss: 0.1071\n",
- "Epoch 417/2500\n",
- "990/990 [==============================] - 0s 459us/step - loss: 0.0634 - val_loss: 0.1072\n",
- "Epoch 418/2500\n",
- "990/990 [==============================] - 0s 475us/step - loss: 0.0633 - val_loss: 0.1071\n",
- "Epoch 419/2500\n",
- "990/990 [==============================] - 0s 458us/step - loss: 0.0633 - val_loss: 0.1071\n",
- "Epoch 420/2500\n",
- "990/990 [==============================] - 0s 471us/step - loss: 0.0633 - val_loss: 0.1071\n",
- "Epoch 421/2500\n",
- "990/990 [==============================] - 0s 463us/step - loss: 0.0633 - val_loss: 0.1071\n",
- "Epoch 422/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0633 - val_loss: 0.1070\n",
- "Epoch 423/2500\n",
- "990/990 [==============================] - 0s 476us/step - loss: 0.0632 - val_loss: 0.1070\n",
- "Epoch 424/2500\n",
- "990/990 [==============================] - 0s 483us/step - loss: 0.0632 - val_loss: 0.1070\n",
- "Epoch 425/2500\n",
- "990/990 [==============================] - 0s 465us/step - loss: 0.0632 - val_loss: 0.1070\n",
- "Epoch 426/2500\n",
- "990/990 [==============================] - 0s 482us/step - loss: 0.0632 - val_loss: 0.1070\n",
- "Epoch 427/2500\n",
- "990/990 [==============================] - 0s 470us/step - loss: 0.0631 - val_loss: 0.1070\n",
- "Epoch 428/2500\n",
- "990/990 [==============================] - 0s 477us/step - loss: 0.0631 - val_loss: 0.1069\n",
- "Epoch 429/2500\n",
- "990/990 [==============================] - 0s 465us/step - loss: 0.0631 - val_loss: 0.1069\n",
- "Epoch 430/2500\n",
- "990/990 [==============================] - 0s 469us/step - loss: 0.0631 - val_loss: 0.1069\n",
- "Epoch 431/2500\n",
- "990/990 [==============================] - 0s 474us/step - loss: 0.0631 - val_loss: 0.1068\n",
- "Epoch 432/2500\n",
- "990/990 [==============================] - 0s 475us/step - loss: 0.0630 - val_loss: 0.1068\n",
- "Epoch 433/2500\n",
- "990/990 [==============================] - 0s 472us/step - loss: 0.0630 - val_loss: 0.1068\n",
- "Epoch 434/2500\n",
- "990/990 [==============================] - 0s 471us/step - loss: 0.0630 - val_loss: 0.1068\n",
- "Epoch 435/2500\n",
- "990/990 [==============================] - 0s 465us/step - loss: 0.0630 - val_loss: 0.1068\n",
- "Epoch 436/2500\n",
- "990/990 [==============================] - 0s 479us/step - loss: 0.0629 - val_loss: 0.1068\n",
- "Epoch 437/2500\n",
- "990/990 [==============================] - 0s 464us/step - loss: 0.0629 - val_loss: 0.1067\n",
- "Epoch 438/2500\n",
- "990/990 [==============================] - 0s 477us/step - loss: 0.0629 - val_loss: 0.1067\n",
- "Epoch 439/2500\n",
- "990/990 [==============================] - 0s 483us/step - loss: 0.0629 - val_loss: 0.1067\n",
- "Epoch 440/2500\n",
- "990/990 [==============================] - 0s 477us/step - loss: 0.0629 - val_loss: 0.1066\n",
- "Epoch 441/2500\n",
- "990/990 [==============================] - 0s 469us/step - loss: 0.0628 - val_loss: 0.1066\n",
- "Epoch 442/2500\n",
- "990/990 [==============================] - 0s 486us/step - loss: 0.0628 - val_loss: 0.1066\n",
- "Epoch 443/2500\n",
- "990/990 [==============================] - 0s 486us/step - loss: 0.0628 - val_loss: 0.1066\n",
- "Epoch 444/2500\n",
- "990/990 [==============================] - 0s 482us/step - loss: 0.0628 - val_loss: 0.1065\n",
- "Epoch 445/2500\n",
- "990/990 [==============================] - 0s 479us/step - loss: 0.0628 - val_loss: 0.1065\n",
- "Epoch 446/2500\n",
- "990/990 [==============================] - 0s 482us/step - loss: 0.0627 - val_loss: 0.1065\n",
- "Epoch 447/2500\n",
- "990/990 [==============================] - 0s 486us/step - loss: 0.0627 - val_loss: 0.1065\n",
- "Epoch 448/2500\n",
- "990/990 [==============================] - 0s 483us/step - loss: 0.0627 - val_loss: 0.1065\n",
- "Epoch 449/2500\n",
- "990/990 [==============================] - 0s 471us/step - loss: 0.0627 - val_loss: 0.1065\n",
- "Epoch 450/2500\n",
- "990/990 [==============================] - 0s 472us/step - loss: 0.0627 - val_loss: 0.1064\n",
- "Epoch 451/2500\n",
- "990/990 [==============================] - 0s 471us/step - loss: 0.0626 - val_loss: 0.1064\n",
- "Epoch 452/2500\n",
- "990/990 [==============================] - 0s 465us/step - loss: 0.0626 - val_loss: 0.1064\n",
- "Epoch 453/2500\n",
- "990/990 [==============================] - 0s 465us/step - loss: 0.0626 - val_loss: 0.1064\n",
- "Epoch 454/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0626 - val_loss: 0.1064\n",
- "Epoch 455/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0626 - val_loss: 0.1064\n",
- "Epoch 456/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0625 - val_loss: 0.1063\n",
- "Epoch 457/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0625 - val_loss: 0.1063\n",
- "Epoch 458/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0625 - val_loss: 0.1063\n",
- "Epoch 459/2500\n",
- "990/990 [==============================] - 0s 490us/step - loss: 0.0625 - val_loss: 0.1062\n",
- "Epoch 460/2500\n",
- "990/990 [==============================] - 0s 490us/step - loss: 0.0625 - val_loss: 0.1062\n",
- "Epoch 461/2500\n",
- "990/990 [==============================] - 0s 479us/step - loss: 0.0624 - val_loss: 0.1062\n",
- "Epoch 462/2500\n",
- "990/990 [==============================] - 0s 472us/step - loss: 0.0624 - val_loss: 0.1062\n",
- "Epoch 463/2500\n"
+ "Epoch 395/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0639 - val_loss: 0.1077\n",
+ "Epoch 396/1200\n",
+ "990/990 [==============================] - 0s 416us/step - loss: 0.0639 - val_loss: 0.1077\n",
+ "Epoch 397/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0638 - val_loss: 0.1076\n",
+ "Epoch 398/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0638 - val_loss: 0.1076\n",
+ "Epoch 399/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0638 - val_loss: 0.1076\n",
+ "Epoch 400/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0638 - val_loss: 0.1076\n",
+ "Epoch 401/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0637 - val_loss: 0.1076\n",
+ "Epoch 402/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0637 - val_loss: 0.1076\n",
+ "Epoch 403/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0637 - val_loss: 0.1075\n",
+ "Epoch 404/1200\n",
+ "990/990 [==============================] - 0s 417us/step - loss: 0.0637 - val_loss: 0.1075\n",
+ "Epoch 405/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0636 - val_loss: 0.1074\n",
+ "Epoch 406/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0636 - val_loss: 0.1074\n",
+ "Epoch 407/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0636 - val_loss: 0.1074\n",
+ "Epoch 408/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0636 - val_loss: 0.1073\n",
+ "Epoch 409/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0636 - val_loss: 0.1073\n",
+ "Epoch 410/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0635 - val_loss: 0.1073\n",
+ "Epoch 411/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0635 - val_loss: 0.1073\n",
+ "Epoch 412/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0635 - val_loss: 0.1073\n",
+ "Epoch 413/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0635 - val_loss: 0.1072\n",
+ "Epoch 414/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0634 - val_loss: 0.1071\n",
+ "Epoch 415/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0634 - val_loss: 0.1072\n",
+ "Epoch 416/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0634 - val_loss: 0.1071\n",
+ "Epoch 417/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0634 - val_loss: 0.1072\n",
+ "Epoch 418/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0633 - val_loss: 0.1071\n",
+ "Epoch 419/1200\n",
+ "990/990 [==============================] - 0s 416us/step - loss: 0.0633 - val_loss: 0.1071\n",
+ "Epoch 420/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0633 - val_loss: 0.1071\n",
+ "Epoch 421/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0633 - val_loss: 0.1071\n",
+ "Epoch 422/1200\n",
+ "990/990 [==============================] - 0s 427us/step - loss: 0.0633 - val_loss: 0.1070\n",
+ "Epoch 423/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0632 - val_loss: 0.1070\n",
+ "Epoch 424/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0632 - val_loss: 0.1070\n",
+ "Epoch 425/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0632 - val_loss: 0.1070\n",
+ "Epoch 426/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0632 - val_loss: 0.1070\n",
+ "Epoch 427/1200\n",
+ "990/990 [==============================] - 0s 427us/step - loss: 0.0631 - val_loss: 0.1070\n",
+ "Epoch 428/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0631 - val_loss: 0.1069\n",
+ "Epoch 429/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0631 - val_loss: 0.1069\n",
+ "Epoch 430/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0631 - val_loss: 0.1069\n",
+ "Epoch 431/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0631 - val_loss: 0.1068\n",
+ "Epoch 432/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0630 - val_loss: 0.1068\n",
+ "Epoch 433/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0630 - val_loss: 0.1068\n",
+ "Epoch 434/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0630 - val_loss: 0.1068\n",
+ "Epoch 435/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0630 - val_loss: 0.1068\n",
+ "Epoch 436/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0629 - val_loss: 0.1068\n",
+ "Epoch 437/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0629 - val_loss: 0.1067\n",
+ "Epoch 438/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0629 - val_loss: 0.1067\n",
+ "Epoch 439/1200\n",
+ "990/990 [==============================] - 0s 414us/step - loss: 0.0629 - val_loss: 0.1067\n",
+ "Epoch 440/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0629 - val_loss: 0.1066\n",
+ "Epoch 441/1200\n",
+ "990/990 [==============================] - 0s 416us/step - loss: 0.0628 - val_loss: 0.1066\n",
+ "Epoch 442/1200\n",
+ "990/990 [==============================] - 0s 417us/step - loss: 0.0628 - val_loss: 0.1066\n",
+ "Epoch 443/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0628 - val_loss: 0.1066\n",
+ "Epoch 444/1200\n",
+ "990/990 [==============================] - 0s 417us/step - loss: 0.0628 - val_loss: 0.1065\n",
+ "Epoch 445/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0628 - val_loss: 0.1065\n",
+ "Epoch 446/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0627 - val_loss: 0.1065\n",
+ "Epoch 447/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0627 - val_loss: 0.1065\n",
+ "Epoch 448/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0627 - val_loss: 0.1065\n",
+ "Epoch 449/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0627 - val_loss: 0.1065\n",
+ "Epoch 450/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0627 - val_loss: 0.1064\n",
+ "Epoch 451/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0626 - val_loss: 0.1064\n",
+ "Epoch 452/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0626 - val_loss: 0.1064\n",
+ "Epoch 453/1200\n",
+ "990/990 [==============================] - 0s 417us/step - loss: 0.0626 - val_loss: 0.1064\n",
+ "Epoch 454/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0626 - val_loss: 0.1064\n",
+ "Epoch 455/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0626 - val_loss: 0.1064\n",
+ "Epoch 456/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0625 - val_loss: 0.1063\n",
+ "Epoch 457/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0625 - val_loss: 0.1063\n",
+ "Epoch 458/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0625 - val_loss: 0.1063\n",
+ "Epoch 459/1200\n",
+ "990/990 [==============================] - 0s 415us/step - loss: 0.0625 - val_loss: 0.1062\n",
+ "Epoch 460/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0625 - val_loss: 0.1062\n",
+ "Epoch 461/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0624 - val_loss: 0.1062\n",
+ "Epoch 462/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0624 - val_loss: 0.1062\n",
+ "Epoch 463/1200\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
- "990/990 [==============================] - 0s 474us/step - loss: 0.0624 - val_loss: 0.1062\n",
- "Epoch 464/2500\n",
- "990/990 [==============================] - 0s 477us/step - loss: 0.0624 - val_loss: 0.1062\n",
- "Epoch 465/2500\n",
- "990/990 [==============================] - 0s 444us/step - loss: 0.0624 - val_loss: 0.1062\n",
- "Epoch 466/2500\n",
- "990/990 [==============================] - 0s 461us/step - loss: 0.0623 - val_loss: 0.1062\n",
- "Epoch 467/2500\n",
- "990/990 [==============================] - 0s 461us/step - loss: 0.0623 - val_loss: 0.1061\n",
- "Epoch 468/2500\n",
- "990/990 [==============================] - 0s 461us/step - loss: 0.0623 - val_loss: 0.1060\n",
- "Epoch 469/2500\n",
- "990/990 [==============================] - 0s 448us/step - loss: 0.0623 - val_loss: 0.1060\n",
- "Epoch 470/2500\n",
- "990/990 [==============================] - 0s 470us/step - loss: 0.0623 - val_loss: 0.1061\n",
- "Epoch 471/2500\n",
- "990/990 [==============================] - 0s 483us/step - loss: 0.0622 - val_loss: 0.1060\n",
- "Epoch 472/2500\n",
- "990/990 [==============================] - 0s 481us/step - loss: 0.0622 - val_loss: 0.1060\n",
- "Epoch 473/2500\n",
- "990/990 [==============================] - 0s 482us/step - loss: 0.0622 - val_loss: 0.1060\n",
- "Epoch 474/2500\n",
- "990/990 [==============================] - 0s 484us/step - loss: 0.0622 - val_loss: 0.1060\n",
- "Epoch 475/2500\n",
- "990/990 [==============================] - 0s 467us/step - loss: 0.0622 - val_loss: 0.1060\n",
- "Epoch 476/2500\n",
- "990/990 [==============================] - 0s 488us/step - loss: 0.0622 - val_loss: 0.1059\n",
- "Epoch 477/2500\n",
- "990/990 [==============================] - 0s 471us/step - loss: 0.0621 - val_loss: 0.1058\n",
- "Epoch 478/2500\n",
- "990/990 [==============================] - 0s 476us/step - loss: 0.0621 - val_loss: 0.1058\n",
- "Epoch 479/2500\n",
- "990/990 [==============================] - 0s 476us/step - loss: 0.0621 - val_loss: 0.1058\n",
- "Epoch 480/2500\n",
- "990/990 [==============================] - 0s 470us/step - loss: 0.0621 - val_loss: 0.1059\n",
- "Epoch 481/2500\n",
- "990/990 [==============================] - 0s 456us/step - loss: 0.0621 - val_loss: 0.1058\n",
- "Epoch 482/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0620 - val_loss: 0.1058\n",
- "Epoch 483/2500\n",
- "990/990 [==============================] - 0s 456us/step - loss: 0.0620 - val_loss: 0.1058\n",
- "Epoch 484/2500\n",
- "990/990 [==============================] - 0s 466us/step - loss: 0.0620 - val_loss: 0.1058\n",
- "Epoch 485/2500\n",
- "990/990 [==============================] - 0s 458us/step - loss: 0.0620 - val_loss: 0.1058\n",
- "Epoch 486/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0620 - val_loss: 0.1058\n",
- "Epoch 487/2500\n",
- "990/990 [==============================] - 0s 460us/step - loss: 0.0620 - val_loss: 0.1057\n",
- "Epoch 488/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0619 - val_loss: 0.1057\n",
- "Epoch 489/2500\n",
- "990/990 [==============================] - 0s 460us/step - loss: 0.0619 - val_loss: 0.1057\n",
- "Epoch 490/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0619 - val_loss: 0.1057\n",
- "Epoch 491/2500\n",
- "990/990 [==============================] - 0s 462us/step - loss: 0.0619 - val_loss: 0.1056\n",
- "Epoch 492/2500\n",
- "990/990 [==============================] - 0s 462us/step - loss: 0.0619 - val_loss: 0.1057\n",
- "Epoch 493/2500\n",
- "990/990 [==============================] - 0s 466us/step - loss: 0.0619 - val_loss: 0.1056\n",
- "Epoch 494/2500\n",
- "990/990 [==============================] - 0s 485us/step - loss: 0.0618 - val_loss: 0.1056\n",
- "Epoch 495/2500\n",
- "990/990 [==============================] - 0s 470us/step - loss: 0.0618 - val_loss: 0.1056\n",
- "Epoch 496/2500\n",
- "990/990 [==============================] - 0s 488us/step - loss: 0.0618 - val_loss: 0.1055\n",
- "Epoch 497/2500\n",
- "990/990 [==============================] - 0s 493us/step - loss: 0.0618 - val_loss: 0.1055\n",
- "Epoch 498/2500\n",
- "990/990 [==============================] - 0s 478us/step - loss: 0.0618 - val_loss: 0.1055\n",
- "Epoch 499/2500\n",
- "990/990 [==============================] - 0s 476us/step - loss: 0.0618 - val_loss: 0.1055\n",
- "Epoch 500/2500\n",
- "990/990 [==============================] - 0s 476us/step - loss: 0.0617 - val_loss: 0.1055\n",
- "Epoch 501/2500\n",
- "990/990 [==============================] - 0s 489us/step - loss: 0.0617 - val_loss: 0.1055\n",
- "Epoch 502/2500\n",
- "990/990 [==============================] - 0s 475us/step - loss: 0.0617 - val_loss: 0.1054\n",
- "Epoch 503/2500\n",
- "990/990 [==============================] - 0s 483us/step - loss: 0.0617 - val_loss: 0.1054\n",
- "Epoch 504/2500\n",
- "990/990 [==============================] - 0s 471us/step - loss: 0.0617 - val_loss: 0.1054\n",
- "Epoch 505/2500\n",
- "990/990 [==============================] - 0s 477us/step - loss: 0.0617 - val_loss: 0.1054\n",
- "Epoch 506/2500\n",
- "990/990 [==============================] - 0s 463us/step - loss: 0.0616 - val_loss: 0.1053\n",
- "Epoch 507/2500\n",
- "990/990 [==============================] - 0s 465us/step - loss: 0.0616 - val_loss: 0.1053\n",
- "Epoch 508/2500\n",
- "990/990 [==============================] - 0s 465us/step - loss: 0.0616 - val_loss: 0.1053\n",
- "Epoch 509/2500\n",
- "990/990 [==============================] - 0s 461us/step - loss: 0.0616 - val_loss: 0.1053\n",
- "Epoch 510/2500\n",
- "990/990 [==============================] - 0s 468us/step - loss: 0.0616 - val_loss: 0.1053\n",
- "Epoch 511/2500\n",
- "990/990 [==============================] - 0s 458us/step - loss: 0.0616 - val_loss: 0.1053\n",
- "Epoch 512/2500\n",
- "990/990 [==============================] - 0s 466us/step - loss: 0.0615 - val_loss: 0.1053\n",
- "Epoch 513/2500\n",
- "990/990 [==============================] - 0s 464us/step - loss: 0.0615 - val_loss: 0.1052\n",
- "Epoch 514/2500\n",
- "990/990 [==============================] - 0s 458us/step - loss: 0.0615 - val_loss: 0.1052\n",
- "Epoch 515/2500\n",
- "990/990 [==============================] - 0s 461us/step - loss: 0.0615 - val_loss: 0.1052\n",
- "Epoch 516/2500\n",
- "990/990 [==============================] - 0s 458us/step - loss: 0.0615 - val_loss: 0.1052\n",
- "Epoch 517/2500\n",
- "990/990 [==============================] - 0s 471us/step - loss: 0.0615 - val_loss: 0.1052\n",
- "Epoch 518/2500\n",
- "990/990 [==============================] - 0s 480us/step - loss: 0.0614 - val_loss: 0.1052\n",
- "Epoch 519/2500\n",
- "990/990 [==============================] - 0s 481us/step - loss: 0.0614 - val_loss: 0.1051\n",
- "Epoch 520/2500\n",
- "990/990 [==============================] - 0s 476us/step - loss: 0.0614 - val_loss: 0.1051\n",
- "Epoch 521/2500\n",
- "990/990 [==============================] - 0s 482us/step - loss: 0.0614 - val_loss: 0.1051\n",
- "Epoch 522/2500\n",
- "990/990 [==============================] - 0s 468us/step - loss: 0.0614 - val_loss: 0.1051\n",
- "Epoch 523/2500\n",
- "990/990 [==============================] - 0s 474us/step - loss: 0.0614 - val_loss: 0.1050\n",
- "Epoch 524/2500\n",
- "990/990 [==============================] - 0s 474us/step - loss: 0.0613 - val_loss: 0.1050\n",
- "Epoch 525/2500\n",
- "990/990 [==============================] - 0s 488us/step - loss: 0.0613 - val_loss: 0.1051\n",
- "Epoch 526/2500\n",
- "990/990 [==============================] - 0s 481us/step - loss: 0.0613 - val_loss: 0.1050\n",
- "Epoch 527/2500\n",
- "990/990 [==============================] - 0s 480us/step - loss: 0.0613 - val_loss: 0.1050\n",
- "Epoch 528/2500\n",
- "990/990 [==============================] - 0s 482us/step - loss: 0.0613 - val_loss: 0.1050\n",
- "Epoch 529/2500\n",
- "990/990 [==============================] - 0s 493us/step - loss: 0.0613 - val_loss: 0.1050\n",
- "Epoch 530/2500\n",
- "990/990 [==============================] - 0s 488us/step - loss: 0.0613 - val_loss: 0.1050\n",
- "Epoch 531/2500\n",
- "990/990 [==============================] - 0s 488us/step - loss: 0.0612 - val_loss: 0.1049\n",
- "Epoch 532/2500\n",
- "990/990 [==============================] - 0s 470us/step - loss: 0.0612 - val_loss: 0.1049\n",
- "Epoch 533/2500\n",
- "990/990 [==============================] - 0s 458us/step - loss: 0.0612 - val_loss: 0.1049\n",
- "Epoch 534/2500\n",
- "990/990 [==============================] - 0s 461us/step - loss: 0.0612 - val_loss: 0.1048\n",
- "Epoch 535/2500\n",
- "990/990 [==============================] - 0s 469us/step - loss: 0.0612 - val_loss: 0.1048\n",
- "Epoch 536/2500\n",
- "990/990 [==============================] - 0s 464us/step - loss: 0.0612 - val_loss: 0.1048\n",
- "Epoch 537/2500\n",
- "990/990 [==============================] - 0s 479us/step - loss: 0.0611 - val_loss: 0.1048\n",
- "Epoch 538/2500\n",
- "990/990 [==============================] - 0s 473us/step - loss: 0.0611 - val_loss: 0.1048\n",
- "Epoch 539/2500\n",
- "990/990 [==============================] - 0s 473us/step - loss: 0.0611 - val_loss: 0.1047\n",
- "Epoch 540/2500\n"
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0624 - val_loss: 0.1062\n",
+ "Epoch 464/1200\n",
+ "990/990 [==============================] - 0s 417us/step - loss: 0.0624 - val_loss: 0.1062\n",
+ "Epoch 465/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0624 - val_loss: 0.1062\n",
+ "Epoch 466/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0623 - val_loss: 0.1062\n",
+ "Epoch 467/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0623 - val_loss: 0.1061\n",
+ "Epoch 468/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0623 - val_loss: 0.1060\n",
+ "Epoch 469/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0623 - val_loss: 0.1060\n",
+ "Epoch 470/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0623 - val_loss: 0.1061\n",
+ "Epoch 471/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0622 - val_loss: 0.1060\n",
+ "Epoch 472/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0622 - val_loss: 0.1060\n",
+ "Epoch 473/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0622 - val_loss: 0.1060\n",
+ "Epoch 474/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0622 - val_loss: 0.1060\n",
+ "Epoch 475/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0622 - val_loss: 0.1060\n",
+ "Epoch 476/1200\n",
+ "990/990 [==============================] - 0s 417us/step - loss: 0.0622 - val_loss: 0.1059\n",
+ "Epoch 477/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0621 - val_loss: 0.1058\n",
+ "Epoch 478/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0621 - val_loss: 0.1058\n",
+ "Epoch 479/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0621 - val_loss: 0.1058\n",
+ "Epoch 480/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0621 - val_loss: 0.1059\n",
+ "Epoch 481/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0621 - val_loss: 0.1058\n",
+ "Epoch 482/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0620 - val_loss: 0.1058\n",
+ "Epoch 483/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0620 - val_loss: 0.1058\n",
+ "Epoch 484/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0620 - val_loss: 0.1058\n",
+ "Epoch 485/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0620 - val_loss: 0.1058\n",
+ "Epoch 486/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0620 - val_loss: 0.1058\n",
+ "Epoch 487/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0620 - val_loss: 0.1057\n",
+ "Epoch 488/1200\n",
+ "990/990 [==============================] - 0s 416us/step - loss: 0.0619 - val_loss: 0.1057\n",
+ "Epoch 489/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0619 - val_loss: 0.1057\n",
+ "Epoch 490/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0619 - val_loss: 0.1057\n",
+ "Epoch 491/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0619 - val_loss: 0.1056\n",
+ "Epoch 492/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0619 - val_loss: 0.1057\n",
+ "Epoch 493/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0619 - val_loss: 0.1056\n",
+ "Epoch 494/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0618 - val_loss: 0.1056\n",
+ "Epoch 495/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0618 - val_loss: 0.1056\n",
+ "Epoch 496/1200\n",
+ "990/990 [==============================] - 0s 413us/step - loss: 0.0618 - val_loss: 0.1055\n",
+ "Epoch 497/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0618 - val_loss: 0.1055\n",
+ "Epoch 498/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0618 - val_loss: 0.1055\n",
+ "Epoch 499/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0618 - val_loss: 0.1055\n",
+ "Epoch 500/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0617 - val_loss: 0.1055\n",
+ "Epoch 501/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0617 - val_loss: 0.1055\n",
+ "Epoch 502/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0617 - val_loss: 0.1054\n",
+ "Epoch 503/1200\n",
+ "990/990 [==============================] - 0s 427us/step - loss: 0.0617 - val_loss: 0.1054\n",
+ "Epoch 504/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0617 - val_loss: 0.1054\n",
+ "Epoch 505/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0617 - val_loss: 0.1054\n",
+ "Epoch 506/1200\n",
+ "990/990 [==============================] - 0s 415us/step - loss: 0.0616 - val_loss: 0.1053\n",
+ "Epoch 507/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0616 - val_loss: 0.1053\n",
+ "Epoch 508/1200\n",
+ "990/990 [==============================] - 0s 415us/step - loss: 0.0616 - val_loss: 0.1053\n",
+ "Epoch 509/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0616 - val_loss: 0.1053\n",
+ "Epoch 510/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0616 - val_loss: 0.1053\n",
+ "Epoch 511/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0616 - val_loss: 0.1053\n",
+ "Epoch 512/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0615 - val_loss: 0.1053\n",
+ "Epoch 513/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0615 - val_loss: 0.1052\n",
+ "Epoch 514/1200\n",
+ "990/990 [==============================] - 0s 431us/step - loss: 0.0615 - val_loss: 0.1052\n",
+ "Epoch 515/1200\n",
+ "990/990 [==============================] - 0s 427us/step - loss: 0.0615 - val_loss: 0.1052\n",
+ "Epoch 516/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0615 - val_loss: 0.1052\n",
+ "Epoch 517/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0615 - val_loss: 0.1052\n",
+ "Epoch 518/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0614 - val_loss: 0.1052\n",
+ "Epoch 519/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0614 - val_loss: 0.1051\n",
+ "Epoch 520/1200\n",
+ "990/990 [==============================] - 0s 429us/step - loss: 0.0614 - val_loss: 0.1051\n",
+ "Epoch 521/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0614 - val_loss: 0.1051\n",
+ "Epoch 522/1200\n",
+ "990/990 [==============================] - 0s 432us/step - loss: 0.0614 - val_loss: 0.1051\n",
+ "Epoch 523/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0614 - val_loss: 0.1050\n",
+ "Epoch 524/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0613 - val_loss: 0.1050\n",
+ "Epoch 525/1200\n",
+ "990/990 [==============================] - 0s 434us/step - loss: 0.0613 - val_loss: 0.1051\n",
+ "Epoch 526/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0613 - val_loss: 0.1050\n",
+ "Epoch 527/1200\n",
+ "990/990 [==============================] - 0s 417us/step - loss: 0.0613 - val_loss: 0.1050\n",
+ "Epoch 528/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0613 - val_loss: 0.1050\n",
+ "Epoch 529/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0613 - val_loss: 0.1050\n",
+ "Epoch 530/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0613 - val_loss: 0.1050\n",
+ "Epoch 531/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0612 - val_loss: 0.1049\n",
+ "Epoch 532/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0612 - val_loss: 0.1049\n",
+ "Epoch 533/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0612 - val_loss: 0.1049\n",
+ "Epoch 534/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0612 - val_loss: 0.1048\n",
+ "Epoch 535/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0612 - val_loss: 0.1048\n",
+ "Epoch 536/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0612 - val_loss: 0.1048\n",
+ "Epoch 537/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0611 - val_loss: 0.1048\n",
+ "Epoch 538/1200\n",
+ "990/990 [==============================] - 0s 427us/step - loss: 0.0611 - val_loss: 0.1048\n",
+ "Epoch 539/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0611 - val_loss: 0.1047\n",
+ "Epoch 540/1200\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
- "990/990 [==============================] - 0s 474us/step - loss: 0.0611 - val_loss: 0.1048\n",
- "Epoch 541/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0611 - val_loss: 0.1047\n",
- "Epoch 542/2500\n",
- "990/990 [==============================] - 0s 466us/step - loss: 0.0611 - val_loss: 0.1047\n",
- "Epoch 543/2500\n",
- "990/990 [==============================] - 0s 466us/step - loss: 0.0611 - val_loss: 0.1047\n",
- "Epoch 544/2500\n",
- "990/990 [==============================] - 0s 472us/step - loss: 0.0610 - val_loss: 0.1046\n",
- "Epoch 545/2500\n",
- "990/990 [==============================] - 0s 466us/step - loss: 0.0610 - val_loss: 0.1047\n",
- "Epoch 546/2500\n",
- "990/990 [==============================] - 0s 468us/step - loss: 0.0610 - val_loss: 0.1046\n",
- "Epoch 547/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0610 - val_loss: 0.1046\n",
- "Epoch 548/2500\n",
- "990/990 [==============================] - 0s 456us/step - loss: 0.0610 - val_loss: 0.1046\n",
- "Epoch 549/2500\n",
- "990/990 [==============================] - 0s 461us/step - loss: 0.0610 - val_loss: 0.1046\n",
- "Epoch 550/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0610 - val_loss: 0.1046\n",
- "Epoch 551/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0609 - val_loss: 0.1046\n",
- "Epoch 552/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0609 - val_loss: 0.1045\n",
- "Epoch 553/2500\n",
- "990/990 [==============================] - 0s 459us/step - loss: 0.0609 - val_loss: 0.1046\n",
- "Epoch 554/2500\n",
- "990/990 [==============================] - 0s 463us/step - loss: 0.0609 - val_loss: 0.1045\n",
- "Epoch 555/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0609 - val_loss: 0.1045\n",
- "Epoch 556/2500\n",
- "990/990 [==============================] - 0s 464us/step - loss: 0.0609 - val_loss: 0.1045\n",
- "Epoch 557/2500\n",
- "990/990 [==============================] - 0s 466us/step - loss: 0.0609 - val_loss: 0.1045\n",
- "Epoch 558/2500\n",
- "990/990 [==============================] - 0s 463us/step - loss: 0.0608 - val_loss: 0.1045\n",
- "Epoch 559/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0608 - val_loss: 0.1045\n",
- "Epoch 560/2500\n",
- "990/990 [==============================] - 0s 461us/step - loss: 0.0608 - val_loss: 0.1044\n",
- "Epoch 561/2500\n",
- "990/990 [==============================] - 0s 465us/step - loss: 0.0608 - val_loss: 0.1044\n",
- "Epoch 562/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0608 - val_loss: 0.1044\n",
- "Epoch 563/2500\n",
- "990/990 [==============================] - 0s 460us/step - loss: 0.0608 - val_loss: 0.1044\n",
- "Epoch 564/2500\n",
- "990/990 [==============================] - 0s 465us/step - loss: 0.0608 - val_loss: 0.1044\n",
- "Epoch 565/2500\n",
- "990/990 [==============================] - 0s 459us/step - loss: 0.0607 - val_loss: 0.1043\n",
- "Epoch 566/2500\n",
- "990/990 [==============================] - 0s 461us/step - loss: 0.0607 - val_loss: 0.1043\n",
- "Epoch 567/2500\n",
- "990/990 [==============================] - 0s 465us/step - loss: 0.0607 - val_loss: 0.1043\n",
- "Epoch 568/2500\n",
- "990/990 [==============================] - 0s 462us/step - loss: 0.0607 - val_loss: 0.1043\n",
- "Epoch 569/2500\n",
- "990/990 [==============================] - 0s 463us/step - loss: 0.0607 - val_loss: 0.1043\n",
- "Epoch 570/2500\n",
- "990/990 [==============================] - 0s 463us/step - loss: 0.0607 - val_loss: 0.1043\n",
- "Epoch 571/2500\n",
- "990/990 [==============================] - 0s 464us/step - loss: 0.0607 - val_loss: 0.1042\n",
- "Epoch 572/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0607 - val_loss: 0.1042\n",
- "Epoch 573/2500\n",
- "990/990 [==============================] - 0s 462us/step - loss: 0.0606 - val_loss: 0.1042\n",
- "Epoch 574/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0606 - val_loss: 0.1042\n",
- "Epoch 575/2500\n",
- "990/990 [==============================] - 0s 464us/step - loss: 0.0606 - val_loss: 0.1042\n",
- "Epoch 576/2500\n",
- "990/990 [==============================] - 0s 467us/step - loss: 0.0606 - val_loss: 0.1042\n",
- "Epoch 577/2500\n",
- "990/990 [==============================] - 0s 469us/step - loss: 0.0606 - val_loss: 0.1042\n",
- "Epoch 578/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0606 - val_loss: 0.1041\n",
- "Epoch 579/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0606 - val_loss: 0.1041\n",
- "Epoch 580/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0605 - val_loss: 0.1041\n",
- "Epoch 581/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0605 - val_loss: 0.1041\n",
- "Epoch 582/2500\n",
- "990/990 [==============================] - 0s 458us/step - loss: 0.0605 - val_loss: 0.1041\n",
- "Epoch 583/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0605 - val_loss: 0.1041\n",
- "Epoch 584/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0605 - val_loss: 0.1040\n",
- "Epoch 585/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0605 - val_loss: 0.1040\n",
- "Epoch 586/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0605 - val_loss: 0.1040\n",
- "Epoch 587/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0605 - val_loss: 0.1040\n",
- "Epoch 588/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0604 - val_loss: 0.1040\n",
- "Epoch 589/2500\n",
- "990/990 [==============================] - 0s 462us/step - loss: 0.0604 - val_loss: 0.1040\n",
- "Epoch 590/2500\n",
- "990/990 [==============================] - 0s 458us/step - loss: 0.0604 - val_loss: 0.1040\n",
- "Epoch 591/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0604 - val_loss: 0.1039\n",
- "Epoch 592/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0604 - val_loss: 0.1039\n",
- "Epoch 593/2500\n",
- "990/990 [==============================] - 0s 459us/step - loss: 0.0604 - val_loss: 0.1039\n",
- "Epoch 594/2500\n",
- "990/990 [==============================] - 0s 463us/step - loss: 0.0604 - val_loss: 0.1039\n",
- "Epoch 595/2500\n",
- "990/990 [==============================] - 0s 456us/step - loss: 0.0604 - val_loss: 0.1039\n",
- "Epoch 596/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0603 - val_loss: 0.1039\n",
- "Epoch 597/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0603 - val_loss: 0.1039\n",
- "Epoch 598/2500\n",
- "990/990 [==============================] - 0s 462us/step - loss: 0.0603 - val_loss: 0.1038\n",
- "Epoch 599/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0603 - val_loss: 0.1038\n",
- "Epoch 600/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0603 - val_loss: 0.1038\n",
- "Epoch 601/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0603 - val_loss: 0.1038\n",
- "Epoch 602/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0603 - val_loss: 0.1038\n",
- "Epoch 603/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0603 - val_loss: 0.1038\n",
- "Epoch 604/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0602 - val_loss: 0.1038\n",
- "Epoch 605/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0602 - val_loss: 0.1037\n",
- "Epoch 606/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0602 - val_loss: 0.1037\n",
- "Epoch 607/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0602 - val_loss: 0.1038\n",
- "Epoch 608/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0602 - val_loss: 0.1037\n",
- "Epoch 609/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0602 - val_loss: 0.1037\n",
- "Epoch 610/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0602 - val_loss: 0.1036\n",
- "Epoch 611/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0602 - val_loss: 0.1036\n",
- "Epoch 612/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0601 - val_loss: 0.1036\n",
- "Epoch 613/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0601 - val_loss: 0.1036\n",
- "Epoch 614/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0601 - val_loss: 0.1036\n",
- "Epoch 615/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0601 - val_loss: 0.1036\n",
- "Epoch 616/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0601 - val_loss: 0.1036\n",
- "Epoch 617/2500\n"
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0611 - val_loss: 0.1048\n",
+ "Epoch 541/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0611 - val_loss: 0.1047\n",
+ "Epoch 542/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0611 - val_loss: 0.1047\n",
+ "Epoch 543/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0611 - val_loss: 0.1047\n",
+ "Epoch 544/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0610 - val_loss: 0.1046\n",
+ "Epoch 545/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0610 - val_loss: 0.1047\n",
+ "Epoch 546/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0610 - val_loss: 0.1046\n",
+ "Epoch 547/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0610 - val_loss: 0.1046\n",
+ "Epoch 548/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0610 - val_loss: 0.1046\n",
+ "Epoch 549/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0610 - val_loss: 0.1046\n",
+ "Epoch 550/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0610 - val_loss: 0.1046\n",
+ "Epoch 551/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0609 - val_loss: 0.1046\n",
+ "Epoch 552/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0609 - val_loss: 0.1045\n",
+ "Epoch 553/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0609 - val_loss: 0.1046\n",
+ "Epoch 554/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0609 - val_loss: 0.1045\n",
+ "Epoch 555/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0609 - val_loss: 0.1045\n",
+ "Epoch 556/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0609 - val_loss: 0.1045\n",
+ "Epoch 557/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0609 - val_loss: 0.1045\n",
+ "Epoch 558/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0608 - val_loss: 0.1045\n",
+ "Epoch 559/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0608 - val_loss: 0.1045\n",
+ "Epoch 560/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0608 - val_loss: 0.1044\n",
+ "Epoch 561/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0608 - val_loss: 0.1044\n",
+ "Epoch 562/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0608 - val_loss: 0.1044\n",
+ "Epoch 563/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0608 - val_loss: 0.1044\n",
+ "Epoch 564/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0608 - val_loss: 0.1044\n",
+ "Epoch 565/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0607 - val_loss: 0.1043\n",
+ "Epoch 566/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0607 - val_loss: 0.1043\n",
+ "Epoch 567/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0607 - val_loss: 0.1043\n",
+ "Epoch 568/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0607 - val_loss: 0.1043\n",
+ "Epoch 569/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0607 - val_loss: 0.1043\n",
+ "Epoch 570/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0607 - val_loss: 0.1043\n",
+ "Epoch 571/1200\n",
+ "990/990 [==============================] - 0s 427us/step - loss: 0.0607 - val_loss: 0.1042\n",
+ "Epoch 572/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0607 - val_loss: 0.1042\n",
+ "Epoch 573/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0606 - val_loss: 0.1042\n",
+ "Epoch 574/1200\n",
+ "990/990 [==============================] - 0s 413us/step - loss: 0.0606 - val_loss: 0.1042\n",
+ "Epoch 575/1200\n",
+ "990/990 [==============================] - 0s 417us/step - loss: 0.0606 - val_loss: 0.1042\n",
+ "Epoch 576/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0606 - val_loss: 0.1042\n",
+ "Epoch 577/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0606 - val_loss: 0.1042\n",
+ "Epoch 578/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0606 - val_loss: 0.1041\n",
+ "Epoch 579/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0606 - val_loss: 0.1041\n",
+ "Epoch 580/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0605 - val_loss: 0.1041\n",
+ "Epoch 581/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0605 - val_loss: 0.1041\n",
+ "Epoch 582/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0605 - val_loss: 0.1041\n",
+ "Epoch 583/1200\n",
+ "990/990 [==============================] - 0s 430us/step - loss: 0.0605 - val_loss: 0.1041\n",
+ "Epoch 584/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0605 - val_loss: 0.1040\n",
+ "Epoch 585/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0605 - val_loss: 0.1040\n",
+ "Epoch 586/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0605 - val_loss: 0.1040\n",
+ "Epoch 587/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0605 - val_loss: 0.1040\n",
+ "Epoch 588/1200\n",
+ "990/990 [==============================] - 0s 427us/step - loss: 0.0604 - val_loss: 0.1040\n",
+ "Epoch 589/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0604 - val_loss: 0.1040\n",
+ "Epoch 590/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0604 - val_loss: 0.1040\n",
+ "Epoch 591/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0604 - val_loss: 0.1039\n",
+ "Epoch 592/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0604 - val_loss: 0.1039\n",
+ "Epoch 593/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0604 - val_loss: 0.1039\n",
+ "Epoch 594/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0604 - val_loss: 0.1039\n",
+ "Epoch 595/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0604 - val_loss: 0.1039\n",
+ "Epoch 596/1200\n",
+ "990/990 [==============================] - 0s 432us/step - loss: 0.0603 - val_loss: 0.1039\n",
+ "Epoch 597/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0603 - val_loss: 0.1039\n",
+ "Epoch 598/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0603 - val_loss: 0.1038\n",
+ "Epoch 599/1200\n",
+ "990/990 [==============================] - 0s 427us/step - loss: 0.0603 - val_loss: 0.1038\n",
+ "Epoch 600/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0603 - val_loss: 0.1038\n",
+ "Epoch 601/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0603 - val_loss: 0.1038\n",
+ "Epoch 602/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0603 - val_loss: 0.1038\n",
+ "Epoch 603/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0603 - val_loss: 0.1038\n",
+ "Epoch 604/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0602 - val_loss: 0.1038\n",
+ "Epoch 605/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0602 - val_loss: 0.1037\n",
+ "Epoch 606/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0602 - val_loss: 0.1037\n",
+ "Epoch 607/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0602 - val_loss: 0.1038\n",
+ "Epoch 608/1200\n",
+ "990/990 [==============================] - 0s 430us/step - loss: 0.0602 - val_loss: 0.1037\n",
+ "Epoch 609/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0602 - val_loss: 0.1037\n",
+ "Epoch 610/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0602 - val_loss: 0.1036\n",
+ "Epoch 611/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0602 - val_loss: 0.1036\n",
+ "Epoch 612/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0601 - val_loss: 0.1036\n",
+ "Epoch 613/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0601 - val_loss: 0.1036\n",
+ "Epoch 614/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0601 - val_loss: 0.1036\n",
+ "Epoch 615/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0601 - val_loss: 0.1036\n",
+ "Epoch 616/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0601 - val_loss: 0.1036\n",
+ "Epoch 617/1200\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
- "990/990 [==============================] - 0s 454us/step - loss: 0.0601 - val_loss: 0.1036\n",
- "Epoch 618/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0601 - val_loss: 0.1035\n",
- "Epoch 619/2500\n",
- "990/990 [==============================] - 0s 444us/step - loss: 0.0601 - val_loss: 0.1035\n",
- "Epoch 620/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0601 - val_loss: 0.1035\n",
- "Epoch 621/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0600 - val_loss: 0.1035\n",
- "Epoch 622/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0600 - val_loss: 0.1035\n",
- "Epoch 623/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0600 - val_loss: 0.1035\n",
- "Epoch 624/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0600 - val_loss: 0.1035\n",
- "Epoch 625/2500\n",
- "990/990 [==============================] - 0s 461us/step - loss: 0.0600 - val_loss: 0.1035\n",
- "Epoch 626/2500\n",
- "990/990 [==============================] - 0s 441us/step - loss: 0.0600 - val_loss: 0.1035\n",
- "Epoch 627/2500\n",
- "990/990 [==============================] - 0s 446us/step - loss: 0.0600 - val_loss: 0.1034\n",
- "Epoch 628/2500\n",
- "990/990 [==============================] - 0s 443us/step - loss: 0.0600 - val_loss: 0.1034\n",
- "Epoch 629/2500\n",
- "990/990 [==============================] - 0s 448us/step - loss: 0.0599 - val_loss: 0.1034\n",
- "Epoch 630/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0599 - val_loss: 0.1034\n",
- "Epoch 631/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0599 - val_loss: 0.1033\n",
- "Epoch 632/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0599 - val_loss: 0.1034\n",
- "Epoch 633/2500\n",
- "990/990 [==============================] - 0s 445us/step - loss: 0.0599 - val_loss: 0.1033\n",
- "Epoch 634/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0599 - val_loss: 0.1033\n",
- "Epoch 635/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0599 - val_loss: 0.1033\n",
- "Epoch 636/2500\n",
- "990/990 [==============================] - 0s 445us/step - loss: 0.0599 - val_loss: 0.1033\n",
- "Epoch 637/2500\n",
- "990/990 [==============================] - 0s 446us/step - loss: 0.0599 - val_loss: 0.1033\n",
- "Epoch 638/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0598 - val_loss: 0.1033\n",
- "Epoch 639/2500\n",
- "990/990 [==============================] - 0s 443us/step - loss: 0.0598 - val_loss: 0.1032\n",
- "Epoch 640/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0598 - val_loss: 0.1032\n",
- "Epoch 641/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0598 - val_loss: 0.1032\n",
- "Epoch 642/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0598 - val_loss: 0.1032\n",
- "Epoch 643/2500\n",
- "990/990 [==============================] - 0s 443us/step - loss: 0.0598 - val_loss: 0.1032\n",
- "Epoch 644/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0598 - val_loss: 0.1032\n",
- "Epoch 645/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0598 - val_loss: 0.1031\n",
- "Epoch 646/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0598 - val_loss: 0.1032\n",
- "Epoch 647/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0597 - val_loss: 0.1032\n",
- "Epoch 648/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0597 - val_loss: 0.1031\n",
- "Epoch 649/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0597 - val_loss: 0.1031\n",
- "Epoch 650/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0597 - val_loss: 0.1031\n",
- "Epoch 651/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0597 - val_loss: 0.1031\n",
- "Epoch 652/2500\n",
- "990/990 [==============================] - 0s 446us/step - loss: 0.0597 - val_loss: 0.1031\n",
- "Epoch 653/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0597 - val_loss: 0.1031\n",
- "Epoch 654/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0597 - val_loss: 0.1030\n",
- "Epoch 655/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0597 - val_loss: 0.1030\n",
- "Epoch 656/2500\n",
+ "990/990 [==============================] - 0s 417us/step - loss: 0.0601 - val_loss: 0.1036\n",
+ "Epoch 618/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0601 - val_loss: 0.1035\n",
+ "Epoch 619/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0601 - val_loss: 0.1035\n",
+ "Epoch 620/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0601 - val_loss: 0.1035\n",
+ "Epoch 621/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0600 - val_loss: 0.1035\n",
+ "Epoch 622/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0600 - val_loss: 0.1035\n",
+ "Epoch 623/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0600 - val_loss: 0.1035\n",
+ "Epoch 624/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0600 - val_loss: 0.1035\n",
+ "Epoch 625/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0600 - val_loss: 0.1035\n",
+ "Epoch 626/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0600 - val_loss: 0.1035\n",
+ "Epoch 627/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0600 - val_loss: 0.1034\n",
+ "Epoch 628/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0600 - val_loss: 0.1034\n",
+ "Epoch 629/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0599 - val_loss: 0.1034\n",
+ "Epoch 630/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0599 - val_loss: 0.1034\n",
+ "Epoch 631/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0599 - val_loss: 0.1033\n",
+ "Epoch 632/1200\n",
+ "990/990 [==============================] - 0s 432us/step - loss: 0.0599 - val_loss: 0.1034\n",
+ "Epoch 633/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0599 - val_loss: 0.1033\n",
+ "Epoch 634/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0599 - val_loss: 0.1033\n",
+ "Epoch 635/1200\n",
+ "990/990 [==============================] - 0s 427us/step - loss: 0.0599 - val_loss: 0.1033\n",
+ "Epoch 636/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0599 - val_loss: 0.1033\n",
+ "Epoch 637/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0599 - val_loss: 0.1033\n",
+ "Epoch 638/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0598 - val_loss: 0.1033\n",
+ "Epoch 639/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0598 - val_loss: 0.1032\n",
+ "Epoch 640/1200\n",
+ "990/990 [==============================] - 0s 427us/step - loss: 0.0598 - val_loss: 0.1032\n",
+ "Epoch 641/1200\n",
+ "990/990 [==============================] - 0s 433us/step - loss: 0.0598 - val_loss: 0.1032\n",
+ "Epoch 642/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0598 - val_loss: 0.1032\n",
+ "Epoch 643/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0598 - val_loss: 0.1032\n",
+ "Epoch 644/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0598 - val_loss: 0.1032\n",
+ "Epoch 645/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0598 - val_loss: 0.1031\n",
+ "Epoch 646/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0598 - val_loss: 0.1032\n",
+ "Epoch 647/1200\n",
+ "990/990 [==============================] - 0s 460us/step - loss: 0.0597 - val_loss: 0.1032\n",
+ "Epoch 648/1200\n",
"990/990 [==============================] - 0s 451us/step - loss: 0.0597 - val_loss: 0.1031\n",
- "Epoch 657/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0596 - val_loss: 0.1030\n",
- "Epoch 658/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0596 - val_loss: 0.1030\n",
- "Epoch 659/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0596 - val_loss: 0.1030\n",
- "Epoch 660/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0596 - val_loss: 0.1030\n",
- "Epoch 661/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0596 - val_loss: 0.1030\n",
- "Epoch 662/2500\n",
- "990/990 [==============================] - 0s 458us/step - loss: 0.0596 - val_loss: 0.1029\n",
- "Epoch 663/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0596 - val_loss: 0.1029\n",
- "Epoch 664/2500\n",
- "990/990 [==============================] - 0s 448us/step - loss: 0.0596 - val_loss: 0.1029\n",
- "Epoch 665/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0596 - val_loss: 0.1029\n",
- "Epoch 666/2500\n",
- "990/990 [==============================] - 0s 456us/step - loss: 0.0595 - val_loss: 0.1029\n",
- "Epoch 667/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0595 - val_loss: 0.1029\n",
- "Epoch 668/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0595 - val_loss: 0.1029\n",
- "Epoch 669/2500\n",
- "990/990 [==============================] - 0s 447us/step - loss: 0.0595 - val_loss: 0.1029\n",
- "Epoch 670/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0595 - val_loss: 0.1029\n",
- "Epoch 671/2500\n",
- "990/990 [==============================] - 0s 446us/step - loss: 0.0595 - val_loss: 0.1028\n",
- "Epoch 672/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0595 - val_loss: 0.1028\n",
- "Epoch 673/2500\n",
- "990/990 [==============================] - 0s 459us/step - loss: 0.0595 - val_loss: 0.1028\n",
- "Epoch 674/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0595 - val_loss: 0.1029\n",
- "Epoch 675/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0595 - val_loss: 0.1028\n",
- "Epoch 676/2500\n",
- "990/990 [==============================] - 0s 448us/step - loss: 0.0594 - val_loss: 0.1027\n",
- "Epoch 677/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0594 - val_loss: 0.1027\n",
- "Epoch 678/2500\n",
- "990/990 [==============================] - 0s 456us/step - loss: 0.0594 - val_loss: 0.1028\n",
- "Epoch 679/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0594 - val_loss: 0.1027\n",
- "Epoch 680/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0594 - val_loss: 0.1027\n",
- "Epoch 681/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0594 - val_loss: 0.1027\n",
- "Epoch 682/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0594 - val_loss: 0.1027\n",
- "Epoch 683/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0594 - val_loss: 0.1027\n",
- "Epoch 684/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0594 - val_loss: 0.1027\n",
- "Epoch 685/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0594 - val_loss: 0.1027\n",
- "Epoch 686/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0594 - val_loss: 0.1026\n",
- "Epoch 687/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0593 - val_loss: 0.1026\n",
- "Epoch 688/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0593 - val_loss: 0.1026\n",
- "Epoch 689/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0593 - val_loss: 0.1026\n",
- "Epoch 690/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0593 - val_loss: 0.1026\n",
- "Epoch 691/2500\n",
- "990/990 [==============================] - 0s 458us/step - loss: 0.0593 - val_loss: 0.1026\n",
- "Epoch 692/2500\n",
- "990/990 [==============================] - 0s 456us/step - loss: 0.0593 - val_loss: 0.1026\n",
- "Epoch 693/2500\n",
- "990/990 [==============================] - 0s 458us/step - loss: 0.0593 - val_loss: 0.1026\n",
- "Epoch 694/2500\n"
+ "Epoch 649/1200\n",
+ "990/990 [==============================] - 0s 503us/step - loss: 0.0597 - val_loss: 0.1031\n",
+ "Epoch 650/1200\n",
+ "990/990 [==============================] - 0s 471us/step - loss: 0.0597 - val_loss: 0.1031\n",
+ "Epoch 651/1200\n",
+ "990/990 [==============================] - 0s 433us/step - loss: 0.0597 - val_loss: 0.1031\n",
+ "Epoch 652/1200\n",
+ "990/990 [==============================] - 0s 449us/step - loss: 0.0597 - val_loss: 0.1031\n",
+ "Epoch 653/1200\n",
+ "990/990 [==============================] - 0s 465us/step - loss: 0.0597 - val_loss: 0.1031\n",
+ "Epoch 654/1200\n",
+ "990/990 [==============================] - 0s 438us/step - loss: 0.0597 - val_loss: 0.1030\n",
+ "Epoch 655/1200\n",
+ "990/990 [==============================] - 0s 434us/step - loss: 0.0597 - val_loss: 0.1030\n",
+ "Epoch 656/1200\n",
+ "990/990 [==============================] - 0s 429us/step - loss: 0.0597 - val_loss: 0.1031\n",
+ "Epoch 657/1200\n",
+ "990/990 [==============================] - 0s 448us/step - loss: 0.0596 - val_loss: 0.1030\n",
+ "Epoch 658/1200\n",
+ "990/990 [==============================] - 0s 427us/step - loss: 0.0596 - val_loss: 0.1030\n",
+ "Epoch 659/1200\n",
+ "990/990 [==============================] - 0s 431us/step - loss: 0.0596 - val_loss: 0.1030\n",
+ "Epoch 660/1200\n",
+ "990/990 [==============================] - 0s 443us/step - loss: 0.0596 - val_loss: 0.1030\n",
+ "Epoch 661/1200\n",
+ "990/990 [==============================] - 0s 438us/step - loss: 0.0596 - val_loss: 0.1030\n",
+ "Epoch 662/1200\n",
+ "990/990 [==============================] - 0s 436us/step - loss: 0.0596 - val_loss: 0.1029\n",
+ "Epoch 663/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0596 - val_loss: 0.1029\n",
+ "Epoch 664/1200\n",
+ "990/990 [==============================] - 0s 441us/step - loss: 0.0596 - val_loss: 0.1029\n",
+ "Epoch 665/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0596 - val_loss: 0.1029\n",
+ "Epoch 666/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0595 - val_loss: 0.1029\n",
+ "Epoch 667/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0595 - val_loss: 0.1029\n",
+ "Epoch 668/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0595 - val_loss: 0.1029\n",
+ "Epoch 669/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0595 - val_loss: 0.1029\n",
+ "Epoch 670/1200\n",
+ "990/990 [==============================] - 0s 437us/step - loss: 0.0595 - val_loss: 0.1029\n",
+ "Epoch 671/1200\n",
+ "990/990 [==============================] - 0s 448us/step - loss: 0.0595 - val_loss: 0.1028\n",
+ "Epoch 672/1200\n",
+ "990/990 [==============================] - 0s 432us/step - loss: 0.0595 - val_loss: 0.1028\n",
+ "Epoch 673/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0595 - val_loss: 0.1028\n",
+ "Epoch 674/1200\n",
+ "990/990 [==============================] - 0s 430us/step - loss: 0.0595 - val_loss: 0.1029\n",
+ "Epoch 675/1200\n",
+ "990/990 [==============================] - 0s 454us/step - loss: 0.0595 - val_loss: 0.1028\n",
+ "Epoch 676/1200\n",
+ "990/990 [==============================] - 0s 447us/step - loss: 0.0594 - val_loss: 0.1027\n",
+ "Epoch 677/1200\n",
+ "990/990 [==============================] - 0s 443us/step - loss: 0.0594 - val_loss: 0.1027\n",
+ "Epoch 678/1200\n",
+ "990/990 [==============================] - 0s 446us/step - loss: 0.0594 - val_loss: 0.1028\n",
+ "Epoch 679/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0594 - val_loss: 0.1027\n",
+ "Epoch 680/1200\n",
+ "990/990 [==============================] - 0s 443us/step - loss: 0.0594 - val_loss: 0.1027\n",
+ "Epoch 681/1200\n",
+ "990/990 [==============================] - 0s 435us/step - loss: 0.0594 - val_loss: 0.1027\n",
+ "Epoch 682/1200\n",
+ "990/990 [==============================] - 0s 434us/step - loss: 0.0594 - val_loss: 0.1027\n",
+ "Epoch 683/1200\n",
+ "990/990 [==============================] - 0s 434us/step - loss: 0.0594 - val_loss: 0.1027\n",
+ "Epoch 684/1200\n",
+ "990/990 [==============================] - 0s 457us/step - loss: 0.0594 - val_loss: 0.1027\n",
+ "Epoch 685/1200\n",
+ "990/990 [==============================] - 0s 446us/step - loss: 0.0594 - val_loss: 0.1027\n",
+ "Epoch 686/1200\n",
+ "990/990 [==============================] - 0s 464us/step - loss: 0.0594 - val_loss: 0.1026\n",
+ "Epoch 687/1200\n",
+ "990/990 [==============================] - 0s 427us/step - loss: 0.0593 - val_loss: 0.1026\n",
+ "Epoch 688/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0593 - val_loss: 0.1026\n",
+ "Epoch 689/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0593 - val_loss: 0.1026\n",
+ "Epoch 690/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0593 - val_loss: 0.1026\n",
+ "Epoch 691/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0593 - val_loss: 0.1026\n",
+ "Epoch 692/1200\n",
+ "990/990 [==============================] - 0s 427us/step - loss: 0.0593 - val_loss: 0.1026\n",
+ "Epoch 693/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0593 - val_loss: 0.1026\n",
+ "Epoch 694/1200\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
- "990/990 [==============================] - 0s 454us/step - loss: 0.0593 - val_loss: 0.1026\n",
- "Epoch 695/2500\n",
- "990/990 [==============================] - 0s 458us/step - loss: 0.0593 - val_loss: 0.1026\n",
- "Epoch 696/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0593 - val_loss: 0.1025\n",
- "Epoch 697/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0592 - val_loss: 0.1025\n",
- "Epoch 698/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0592 - val_loss: 0.1025\n",
- "Epoch 699/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0592 - val_loss: 0.1025\n",
- "Epoch 700/2500\n",
- "990/990 [==============================] - 0s 445us/step - loss: 0.0592 - val_loss: 0.1025\n",
- "Epoch 701/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0592 - val_loss: 0.1025\n",
- "Epoch 702/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0592 - val_loss: 0.1025\n",
- "Epoch 703/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0592 - val_loss: 0.1024\n",
- "Epoch 704/2500\n",
- "990/990 [==============================] - 0s 458us/step - loss: 0.0592 - val_loss: 0.1024\n",
- "Epoch 705/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0592 - val_loss: 0.1024\n",
- "Epoch 706/2500\n",
- "990/990 [==============================] - 0s 448us/step - loss: 0.0592 - val_loss: 0.1024\n",
- "Epoch 707/2500\n",
- "990/990 [==============================] - 0s 459us/step - loss: 0.0592 - val_loss: 0.1024\n",
- "Epoch 708/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0591 - val_loss: 0.1024\n",
- "Epoch 709/2500\n",
- "990/990 [==============================] - 0s 445us/step - loss: 0.0591 - val_loss: 0.1024\n",
- "Epoch 710/2500\n",
- "990/990 [==============================] - 0s 445us/step - loss: 0.0591 - val_loss: 0.1024\n",
- "Epoch 711/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0591 - val_loss: 0.1024\n",
- "Epoch 712/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0591 - val_loss: 0.1023\n",
- "Epoch 713/2500\n",
- "990/990 [==============================] - 0s 446us/step - loss: 0.0591 - val_loss: 0.1023\n",
- "Epoch 714/2500\n",
- "990/990 [==============================] - 0s 456us/step - loss: 0.0591 - val_loss: 0.1023\n",
- "Epoch 715/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0591 - val_loss: 0.1023\n",
- "Epoch 716/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0591 - val_loss: 0.1024\n",
- "Epoch 717/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0591 - val_loss: 0.1023\n",
- "Epoch 718/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0591 - val_loss: 0.1023\n",
- "Epoch 719/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0590 - val_loss: 0.1023\n",
- "Epoch 720/2500\n",
- "990/990 [==============================] - 0s 446us/step - loss: 0.0590 - val_loss: 0.1022\n",
- "Epoch 721/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0590 - val_loss: 0.1022\n",
- "Epoch 722/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0590 - val_loss: 0.1022\n",
- "Epoch 723/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0590 - val_loss: 0.1023\n",
- "Epoch 724/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0590 - val_loss: 0.1022\n",
- "Epoch 725/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0590 - val_loss: 0.1022\n",
- "Epoch 726/2500\n",
- "990/990 [==============================] - 0s 458us/step - loss: 0.0590 - val_loss: 0.1022\n",
- "Epoch 727/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0590 - val_loss: 0.1022\n",
- "Epoch 728/2500\n",
- "990/990 [==============================] - 0s 448us/step - loss: 0.0590 - val_loss: 0.1023\n",
- "Epoch 729/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0590 - val_loss: 0.1022\n",
- "Epoch 730/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0589 - val_loss: 0.1022\n",
- "Epoch 731/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0589 - val_loss: 0.1022\n",
- "Epoch 732/2500\n",
- "990/990 [==============================] - 0s 463us/step - loss: 0.0589 - val_loss: 0.1022\n",
- "Epoch 733/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0589 - val_loss: 0.1022\n",
- "Epoch 734/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0589 - val_loss: 0.1021\n",
- "Epoch 735/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0589 - val_loss: 0.1021\n",
- "Epoch 736/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0589 - val_loss: 0.1021\n",
- "Epoch 737/2500\n",
- "990/990 [==============================] - 0s 460us/step - loss: 0.0589 - val_loss: 0.1021\n",
- "Epoch 738/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0589 - val_loss: 0.1021\n",
- "Epoch 739/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0589 - val_loss: 0.1021\n",
- "Epoch 740/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0589 - val_loss: 0.1021\n",
- "Epoch 741/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0589 - val_loss: 0.1021\n",
- "Epoch 742/2500\n",
- "990/990 [==============================] - 0s 461us/step - loss: 0.0588 - val_loss: 0.1020\n",
- "Epoch 743/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0588 - val_loss: 0.1020\n",
- "Epoch 744/2500\n",
- "990/990 [==============================] - 0s 446us/step - loss: 0.0588 - val_loss: 0.1021\n",
- "Epoch 745/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0588 - val_loss: 0.1021\n",
- "Epoch 746/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0588 - val_loss: 0.1021\n",
- "Epoch 747/2500\n",
- "990/990 [==============================] - 0s 463us/step - loss: 0.0588 - val_loss: 0.1020\n",
- "Epoch 748/2500\n",
- "990/990 [==============================] - 0s 458us/step - loss: 0.0588 - val_loss: 0.1020\n",
- "Epoch 749/2500\n",
- "990/990 [==============================] - 0s 483us/step - loss: 0.0588 - val_loss: 0.1020\n",
- "Epoch 750/2500\n",
- "990/990 [==============================] - 0s 474us/step - loss: 0.0588 - val_loss: 0.1020\n",
- "Epoch 751/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0588 - val_loss: 0.1020\n",
- "Epoch 752/2500\n",
- "990/990 [==============================] - 0s 463us/step - loss: 0.0588 - val_loss: 0.1020\n",
- "Epoch 753/2500\n",
- "990/990 [==============================] - 0s 470us/step - loss: 0.0587 - val_loss: 0.1020\n",
- "Epoch 754/2500\n",
- "990/990 [==============================] - 0s 464us/step - loss: 0.0587 - val_loss: 0.1020\n",
- "Epoch 755/2500\n",
- "990/990 [==============================] - 0s 471us/step - loss: 0.0587 - val_loss: 0.1019\n",
- "Epoch 756/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0587 - val_loss: 0.1019\n",
- "Epoch 757/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0587 - val_loss: 0.1019\n",
- "Epoch 758/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0587 - val_loss: 0.1019\n",
- "Epoch 759/2500\n",
- "990/990 [==============================] - 0s 467us/step - loss: 0.0587 - val_loss: 0.1019\n",
- "Epoch 760/2500\n",
- "990/990 [==============================] - 0s 470us/step - loss: 0.0587 - val_loss: 0.1019\n",
- "Epoch 761/2500\n",
- "990/990 [==============================] - 0s 475us/step - loss: 0.0587 - val_loss: 0.1019\n",
- "Epoch 762/2500\n",
- "990/990 [==============================] - 0s 487us/step - loss: 0.0587 - val_loss: 0.1019\n",
- "Epoch 763/2500\n",
- "990/990 [==============================] - 0s 492us/step - loss: 0.0587 - val_loss: 0.1019\n",
- "Epoch 764/2500\n",
- "990/990 [==============================] - 0s 491us/step - loss: 0.0587 - val_loss: 0.1019\n",
- "Epoch 765/2500\n",
- "990/990 [==============================] - 0s 501us/step - loss: 0.0587 - val_loss: 0.1019\n",
- "Epoch 766/2500\n",
- "990/990 [==============================] - 0s 502us/step - loss: 0.0586 - val_loss: 0.1019\n",
- "Epoch 767/2500\n",
- "990/990 [==============================] - 0s 498us/step - loss: 0.0586 - val_loss: 0.1018\n",
- "Epoch 768/2500\n",
- "990/990 [==============================] - 0s 487us/step - loss: 0.0586 - val_loss: 0.1018\n",
- "Epoch 769/2500\n",
- "990/990 [==============================] - 0s 482us/step - loss: 0.0586 - val_loss: 0.1018\n",
- "Epoch 770/2500\n",
- "990/990 [==============================] - 0s 478us/step - loss: 0.0586 - val_loss: 0.1018\n",
- "Epoch 771/2500\n"
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0593 - val_loss: 0.1026\n",
+ "Epoch 695/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0593 - val_loss: 0.1026\n",
+ "Epoch 696/1200\n",
+ "990/990 [==============================] - 0s 430us/step - loss: 0.0593 - val_loss: 0.1025\n",
+ "Epoch 697/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0592 - val_loss: 0.1025\n",
+ "Epoch 698/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0592 - val_loss: 0.1025\n",
+ "Epoch 699/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0592 - val_loss: 0.1025\n",
+ "Epoch 700/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0592 - val_loss: 0.1025\n",
+ "Epoch 701/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0592 - val_loss: 0.1025\n",
+ "Epoch 702/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0592 - val_loss: 0.1025\n",
+ "Epoch 703/1200\n",
+ "990/990 [==============================] - 0s 429us/step - loss: 0.0592 - val_loss: 0.1024\n",
+ "Epoch 704/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0592 - val_loss: 0.1024\n",
+ "Epoch 705/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0592 - val_loss: 0.1024\n",
+ "Epoch 706/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0592 - val_loss: 0.1024\n",
+ "Epoch 707/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0592 - val_loss: 0.1024\n",
+ "Epoch 708/1200\n",
+ "990/990 [==============================] - 0s 416us/step - loss: 0.0591 - val_loss: 0.1024\n",
+ "Epoch 709/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0591 - val_loss: 0.1024\n",
+ "Epoch 710/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0591 - val_loss: 0.1024\n",
+ "Epoch 711/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0591 - val_loss: 0.1024\n",
+ "Epoch 712/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0591 - val_loss: 0.1023\n",
+ "Epoch 713/1200\n",
+ "990/990 [==============================] - 0s 417us/step - loss: 0.0591 - val_loss: 0.1023\n",
+ "Epoch 714/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0591 - val_loss: 0.1023\n",
+ "Epoch 715/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0591 - val_loss: 0.1023\n",
+ "Epoch 716/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0591 - val_loss: 0.1024\n",
+ "Epoch 717/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0591 - val_loss: 0.1023\n",
+ "Epoch 718/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0591 - val_loss: 0.1023\n",
+ "Epoch 719/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0590 - val_loss: 0.1023\n",
+ "Epoch 720/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0590 - val_loss: 0.1022\n",
+ "Epoch 721/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0590 - val_loss: 0.1022\n",
+ "Epoch 722/1200\n",
+ "990/990 [==============================] - 0s 415us/step - loss: 0.0590 - val_loss: 0.1022\n",
+ "Epoch 723/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0590 - val_loss: 0.1023\n",
+ "Epoch 724/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0590 - val_loss: 0.1022\n",
+ "Epoch 725/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0590 - val_loss: 0.1022\n",
+ "Epoch 726/1200\n",
+ "990/990 [==============================] - 0s 417us/step - loss: 0.0590 - val_loss: 0.1022\n",
+ "Epoch 727/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0590 - val_loss: 0.1022\n",
+ "Epoch 728/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0590 - val_loss: 0.1023\n",
+ "Epoch 729/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0590 - val_loss: 0.1022\n",
+ "Epoch 730/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0589 - val_loss: 0.1022\n",
+ "Epoch 731/1200\n",
+ "990/990 [==============================] - 0s 427us/step - loss: 0.0589 - val_loss: 0.1022\n",
+ "Epoch 732/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0589 - val_loss: 0.1022\n",
+ "Epoch 733/1200\n",
+ "990/990 [==============================] - 0s 427us/step - loss: 0.0589 - val_loss: 0.1022\n",
+ "Epoch 734/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0589 - val_loss: 0.1021\n",
+ "Epoch 735/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0589 - val_loss: 0.1021\n",
+ "Epoch 736/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0589 - val_loss: 0.1021\n",
+ "Epoch 737/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0589 - val_loss: 0.1021\n",
+ "Epoch 738/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0589 - val_loss: 0.1021\n",
+ "Epoch 739/1200\n",
+ "990/990 [==============================] - 0s 416us/step - loss: 0.0589 - val_loss: 0.1021\n",
+ "Epoch 740/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0589 - val_loss: 0.1021\n",
+ "Epoch 741/1200\n",
+ "990/990 [==============================] - 0s 427us/step - loss: 0.0589 - val_loss: 0.1021\n",
+ "Epoch 742/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0588 - val_loss: 0.1020\n",
+ "Epoch 743/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0588 - val_loss: 0.1020\n",
+ "Epoch 744/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0588 - val_loss: 0.1021\n",
+ "Epoch 745/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0588 - val_loss: 0.1021\n",
+ "Epoch 746/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0588 - val_loss: 0.1021\n",
+ "Epoch 747/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0588 - val_loss: 0.1020\n",
+ "Epoch 748/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0588 - val_loss: 0.1020\n",
+ "Epoch 749/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0588 - val_loss: 0.1020\n",
+ "Epoch 750/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0588 - val_loss: 0.1020\n",
+ "Epoch 751/1200\n",
+ "990/990 [==============================] - 0s 417us/step - loss: 0.0588 - val_loss: 0.1020\n",
+ "Epoch 752/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0588 - val_loss: 0.1020\n",
+ "Epoch 753/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0587 - val_loss: 0.1020\n",
+ "Epoch 754/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0587 - val_loss: 0.1020\n",
+ "Epoch 755/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0587 - val_loss: 0.1019\n",
+ "Epoch 756/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0587 - val_loss: 0.1019\n",
+ "Epoch 757/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0587 - val_loss: 0.1019\n",
+ "Epoch 758/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0587 - val_loss: 0.1019\n",
+ "Epoch 759/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0587 - val_loss: 0.1019\n",
+ "Epoch 760/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0587 - val_loss: 0.1019\n",
+ "Epoch 761/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0587 - val_loss: 0.1019\n",
+ "Epoch 762/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0587 - val_loss: 0.1019\n",
+ "Epoch 763/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0587 - val_loss: 0.1019\n",
+ "Epoch 764/1200\n",
+ "990/990 [==============================] - 0s 429us/step - loss: 0.0587 - val_loss: 0.1019\n",
+ "Epoch 765/1200\n",
+ "990/990 [==============================] - 0s 436us/step - loss: 0.0587 - val_loss: 0.1019\n",
+ "Epoch 766/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0586 - val_loss: 0.1019\n",
+ "Epoch 767/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0586 - val_loss: 0.1018\n",
+ "Epoch 768/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0586 - val_loss: 0.1018\n",
+ "Epoch 769/1200\n",
+ "990/990 [==============================] - 0s 416us/step - loss: 0.0586 - val_loss: 0.1018\n",
+ "Epoch 770/1200\n",
+ "990/990 [==============================] - 0s 417us/step - loss: 0.0586 - val_loss: 0.1018\n",
+ "Epoch 771/1200\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
- "990/990 [==============================] - 0s 478us/step - loss: 0.0586 - val_loss: 0.1018\n",
- "Epoch 772/2500\n",
- "990/990 [==============================] - 0s 477us/step - loss: 0.0586 - val_loss: 0.1018\n",
- "Epoch 773/2500\n",
- "990/990 [==============================] - 0s 481us/step - loss: 0.0586 - val_loss: 0.1018\n",
- "Epoch 774/2500\n",
- "990/990 [==============================] - 0s 459us/step - loss: 0.0586 - val_loss: 0.1018\n",
- "Epoch 775/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0586 - val_loss: 0.1018\n",
- "Epoch 776/2500\n",
- "990/990 [==============================] - 0s 468us/step - loss: 0.0586 - val_loss: 0.1017\n",
- "Epoch 777/2500\n",
- "990/990 [==============================] - 0s 480us/step - loss: 0.0586 - val_loss: 0.1018\n",
- "Epoch 778/2500\n",
- "990/990 [==============================] - 0s 471us/step - loss: 0.0585 - val_loss: 0.1018\n",
- "Epoch 779/2500\n",
- "990/990 [==============================] - 0s 478us/step - loss: 0.0585 - val_loss: 0.1018\n",
- "Epoch 780/2500\n",
- "990/990 [==============================] - 0s 480us/step - loss: 0.0585 - val_loss: 0.1017\n",
- "Epoch 781/2500\n",
- "990/990 [==============================] - 0s 472us/step - loss: 0.0585 - val_loss: 0.1017\n",
- "Epoch 782/2500\n",
- "990/990 [==============================] - 0s 474us/step - loss: 0.0585 - val_loss: 0.1017\n",
- "Epoch 783/2500\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0586 - val_loss: 0.1018\n",
+ "Epoch 772/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0586 - val_loss: 0.1018\n",
+ "Epoch 773/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0586 - val_loss: 0.1018\n",
+ "Epoch 774/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0586 - val_loss: 0.1018\n",
+ "Epoch 775/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0586 - val_loss: 0.1018\n",
+ "Epoch 776/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0586 - val_loss: 0.1017\n",
+ "Epoch 777/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0586 - val_loss: 0.1018\n",
+ "Epoch 778/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0585 - val_loss: 0.1018\n",
+ "Epoch 779/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0585 - val_loss: 0.1018\n",
+ "Epoch 780/1200\n",
+ "990/990 [==============================] - 0s 431us/step - loss: 0.0585 - val_loss: 0.1017\n",
+ "Epoch 781/1200\n",
+ "990/990 [==============================] - 0s 444us/step - loss: 0.0585 - val_loss: 0.1017\n",
+ "Epoch 782/1200\n",
+ "990/990 [==============================] - 0s 445us/step - loss: 0.0585 - val_loss: 0.1017\n",
+ "Epoch 783/1200\n",
+ "990/990 [==============================] - 0s 440us/step - loss: 0.0585 - val_loss: 0.1017\n",
+ "Epoch 784/1200\n",
+ "990/990 [==============================] - 0s 442us/step - loss: 0.0585 - val_loss: 0.1017\n",
+ "Epoch 785/1200\n",
+ "990/990 [==============================] - 1s 538us/step - loss: 0.0585 - val_loss: 0.1017\n",
+ "Epoch 786/1200\n",
+ "990/990 [==============================] - 0s 503us/step - loss: 0.0585 - val_loss: 0.1017\n",
+ "Epoch 787/1200\n",
"990/990 [==============================] - 0s 468us/step - loss: 0.0585 - val_loss: 0.1017\n",
- "Epoch 784/2500\n",
- "990/990 [==============================] - 0s 484us/step - loss: 0.0585 - val_loss: 0.1017\n",
- "Epoch 785/2500\n",
- "990/990 [==============================] - 0s 488us/step - loss: 0.0585 - val_loss: 0.1017\n",
- "Epoch 786/2500\n",
- "990/990 [==============================] - 0s 479us/step - loss: 0.0585 - val_loss: 0.1017\n",
- "Epoch 787/2500\n",
- "990/990 [==============================] - 0s 471us/step - loss: 0.0585 - val_loss: 0.1017\n",
- "Epoch 788/2500\n",
- "990/990 [==============================] - 0s 480us/step - loss: 0.0585 - val_loss: 0.1017\n",
- "Epoch 789/2500\n",
- "990/990 [==============================] - 0s 489us/step - loss: 0.0585 - val_loss: 0.1017\n",
- "Epoch 790/2500\n",
- "990/990 [==============================] - 0s 481us/step - loss: 0.0585 - val_loss: 0.1017\n",
- "Epoch 791/2500\n",
- "990/990 [==============================] - 0s 479us/step - loss: 0.0584 - val_loss: 0.1017\n",
- "Epoch 792/2500\n",
- "990/990 [==============================] - 0s 465us/step - loss: 0.0584 - val_loss: 0.1016\n",
- "Epoch 793/2500\n",
- "990/990 [==============================] - 0s 471us/step - loss: 0.0584 - val_loss: 0.1016\n",
- "Epoch 794/2500\n",
+ "Epoch 788/1200\n",
+ "990/990 [==============================] - 0s 448us/step - loss: 0.0585 - val_loss: 0.1017\n",
+ "Epoch 789/1200\n",
+ "990/990 [==============================] - 0s 441us/step - loss: 0.0585 - val_loss: 0.1017\n",
+ "Epoch 790/1200\n",
+ "990/990 [==============================] - 0s 463us/step - loss: 0.0585 - val_loss: 0.1017\n",
+ "Epoch 791/1200\n",
+ "990/990 [==============================] - 0s 498us/step - loss: 0.0584 - val_loss: 0.1017\n",
+ "Epoch 792/1200\n",
+ "990/990 [==============================] - 0s 430us/step - loss: 0.0584 - val_loss: 0.1016\n",
+ "Epoch 793/1200\n",
+ "990/990 [==============================] - 0s 438us/step - loss: 0.0584 - val_loss: 0.1016\n",
+ "Epoch 794/1200\n",
+ "990/990 [==============================] - 0s 452us/step - loss: 0.0584 - val_loss: 0.1016\n",
+ "Epoch 795/1200\n",
+ "990/990 [==============================] - 1s 548us/step - loss: 0.0584 - val_loss: 0.1016\n",
+ "Epoch 796/1200\n",
+ "990/990 [==============================] - 0s 484us/step - loss: 0.0584 - val_loss: 0.1016\n",
+ "Epoch 797/1200\n",
+ "990/990 [==============================] - 0s 489us/step - loss: 0.0584 - val_loss: 0.1016\n",
+ "Epoch 798/1200\n",
+ "990/990 [==============================] - 0s 468us/step - loss: 0.0584 - val_loss: 0.1016\n",
+ "Epoch 799/1200\n",
+ "990/990 [==============================] - 0s 484us/step - loss: 0.0584 - val_loss: 0.1016\n",
+ "Epoch 800/1200\n",
"990/990 [==============================] - 0s 470us/step - loss: 0.0584 - val_loss: 0.1016\n",
- "Epoch 795/2500\n",
- "990/990 [==============================] - 0s 467us/step - loss: 0.0584 - val_loss: 0.1016\n",
- "Epoch 796/2500\n",
- "990/990 [==============================] - 0s 481us/step - loss: 0.0584 - val_loss: 0.1016\n",
- "Epoch 797/2500\n",
- "990/990 [==============================] - 0s 456us/step - loss: 0.0584 - val_loss: 0.1016\n",
- "Epoch 798/2500\n",
- "990/990 [==============================] - 0s 475us/step - loss: 0.0584 - val_loss: 0.1016\n",
- "Epoch 799/2500\n",
- "990/990 [==============================] - 0s 471us/step - loss: 0.0584 - val_loss: 0.1016\n",
- "Epoch 800/2500\n",
- "990/990 [==============================] - 0s 472us/step - loss: 0.0584 - val_loss: 0.1016\n",
- "Epoch 801/2500\n",
+ "Epoch 801/1200\n",
+ "990/990 [==============================] - 0s 458us/step - loss: 0.0584 - val_loss: 0.1016\n",
+ "Epoch 802/1200\n",
+ "990/990 [==============================] - 0s 491us/step - loss: 0.0584 - val_loss: 0.1016\n",
+ "Epoch 803/1200\n",
+ "990/990 [==============================] - 1s 520us/step - loss: 0.0584 - val_loss: 0.1016\n",
+ "Epoch 804/1200\n",
"990/990 [==============================] - 0s 469us/step - loss: 0.0584 - val_loss: 0.1016\n",
- "Epoch 802/2500\n",
- "990/990 [==============================] - 0s 490us/step - loss: 0.0584 - val_loss: 0.1016\n",
- "Epoch 803/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0584 - val_loss: 0.1016\n",
- "Epoch 804/2500\n",
- "990/990 [==============================] - 0s 487us/step - loss: 0.0584 - val_loss: 0.1016\n",
- "Epoch 805/2500\n",
- "990/990 [==============================] - 0s 479us/step - loss: 0.0583 - val_loss: 0.1016\n",
- "Epoch 806/2500\n",
- "990/990 [==============================] - 0s 460us/step - loss: 0.0583 - val_loss: 0.1015\n",
- "Epoch 807/2500\n",
- "990/990 [==============================] - 0s 489us/step - loss: 0.0583 - val_loss: 0.1015\n",
- "Epoch 808/2500\n",
- "990/990 [==============================] - 0s 487us/step - loss: 0.0583 - val_loss: 0.1015\n",
- "Epoch 809/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0583 - val_loss: 0.1015\n",
- "Epoch 810/2500\n",
- "990/990 [==============================] - 0s 485us/step - loss: 0.0583 - val_loss: 0.1015\n",
- "Epoch 811/2500\n",
- "990/990 [==============================] - 0s 500us/step - loss: 0.0583 - val_loss: 0.1015\n",
- "Epoch 812/2500\n",
- "990/990 [==============================] - 0s 463us/step - loss: 0.0583 - val_loss: 0.1015\n",
- "Epoch 813/2500\n",
- "990/990 [==============================] - 0s 461us/step - loss: 0.0583 - val_loss: 0.1015\n",
- "Epoch 814/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0583 - val_loss: 0.1015\n",
- "Epoch 815/2500\n",
- "990/990 [==============================] - 0s 465us/step - loss: 0.0583 - val_loss: 0.1015\n",
- "Epoch 816/2500\n",
+ "Epoch 805/1200\n",
+ "990/990 [==============================] - 1s 530us/step - loss: 0.0583 - val_loss: 0.1016\n",
+ "Epoch 806/1200\n",
+ "990/990 [==============================] - 0s 492us/step - loss: 0.0583 - val_loss: 0.1015\n",
+ "Epoch 807/1200\n",
+ "990/990 [==============================] - 0s 497us/step - loss: 0.0583 - val_loss: 0.1015\n",
+ "Epoch 808/1200\n",
+ "990/990 [==============================] - 0s 488us/step - loss: 0.0583 - val_loss: 0.1015\n",
+ "Epoch 809/1200\n",
+ "990/990 [==============================] - 0s 479us/step - loss: 0.0583 - val_loss: 0.1015\n",
+ "Epoch 810/1200\n",
+ "990/990 [==============================] - 0s 458us/step - loss: 0.0583 - val_loss: 0.1015\n",
+ "Epoch 811/1200\n",
"990/990 [==============================] - 0s 467us/step - loss: 0.0583 - val_loss: 0.1015\n",
- "Epoch 817/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0583 - val_loss: 0.1014\n",
- "Epoch 818/2500\n",
- "990/990 [==============================] - 0s 442us/step - loss: 0.0582 - val_loss: 0.1014\n",
- "Epoch 819/2500\n",
- "990/990 [==============================] - 0s 434us/step - loss: 0.0582 - val_loss: 0.1015\n",
- "Epoch 820/2500\n",
- "990/990 [==============================] - 0s 447us/step - loss: 0.0582 - val_loss: 0.1015\n",
- "Epoch 821/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0582 - val_loss: 0.1015\n",
- "Epoch 822/2500\n",
- "990/990 [==============================] - 0s 465us/step - loss: 0.0582 - val_loss: 0.1014\n",
- "Epoch 823/2500\n",
- "990/990 [==============================] - 0s 439us/step - loss: 0.0582 - val_loss: 0.1014\n",
- "Epoch 824/2500\n",
- "990/990 [==============================] - 0s 442us/step - loss: 0.0582 - val_loss: 0.1014\n",
- "Epoch 825/2500\n",
- "990/990 [==============================] - 0s 431us/step - loss: 0.0582 - val_loss: 0.1014\n",
- "Epoch 826/2500\n",
- "990/990 [==============================] - 0s 430us/step - loss: 0.0582 - val_loss: 0.1014\n",
- "Epoch 827/2500\n",
- "990/990 [==============================] - 0s 461us/step - loss: 0.0582 - val_loss: 0.1014\n",
- "Epoch 828/2500\n",
- "990/990 [==============================] - 0s 465us/step - loss: 0.0582 - val_loss: 0.1014\n",
- "Epoch 829/2500\n",
- "990/990 [==============================] - 1s 511us/step - loss: 0.0582 - val_loss: 0.1013\n",
- "Epoch 830/2500\n",
- "990/990 [==============================] - 1s 505us/step - loss: 0.0582 - val_loss: 0.1014\n",
- "Epoch 831/2500\n",
- "990/990 [==============================] - 0s 475us/step - loss: 0.0582 - val_loss: 0.1013\n",
- "Epoch 832/2500\n",
+ "Epoch 812/1200\n",
+ "990/990 [==============================] - 0s 460us/step - loss: 0.0583 - val_loss: 0.1015\n",
+ "Epoch 813/1200\n",
+ "990/990 [==============================] - 0s 505us/step - loss: 0.0583 - val_loss: 0.1015\n",
+ "Epoch 814/1200\n",
+ "990/990 [==============================] - 0s 481us/step - loss: 0.0583 - val_loss: 0.1015\n",
+ "Epoch 815/1200\n",
+ "990/990 [==============================] - 0s 464us/step - loss: 0.0583 - val_loss: 0.1015\n",
+ "Epoch 816/1200\n",
+ "990/990 [==============================] - 0s 468us/step - loss: 0.0583 - val_loss: 0.1015\n",
+ "Epoch 817/1200\n",
+ "990/990 [==============================] - 0s 456us/step - loss: 0.0583 - val_loss: 0.1014\n",
+ "Epoch 818/1200\n",
+ "990/990 [==============================] - 0s 469us/step - loss: 0.0582 - val_loss: 0.1014\n",
+ "Epoch 819/1200\n",
+ "990/990 [==============================] - 1s 559us/step - loss: 0.0582 - val_loss: 0.1015\n",
+ "Epoch 820/1200\n",
+ "990/990 [==============================] - 0s 478us/step - loss: 0.0582 - val_loss: 0.1015\n",
+ "Epoch 821/1200\n",
+ "990/990 [==============================] - 0s 444us/step - loss: 0.0582 - val_loss: 0.1015\n",
+ "Epoch 822/1200\n",
+ "990/990 [==============================] - 0s 444us/step - loss: 0.0582 - val_loss: 0.1014\n",
+ "Epoch 823/1200\n",
+ "990/990 [==============================] - 0s 434us/step - loss: 0.0582 - val_loss: 0.1014\n",
+ "Epoch 824/1200\n",
+ "990/990 [==============================] - 0s 438us/step - loss: 0.0582 - val_loss: 0.1014\n",
+ "Epoch 825/1200\n",
+ "990/990 [==============================] - 0s 435us/step - loss: 0.0582 - val_loss: 0.1014\n",
+ "Epoch 826/1200\n",
"990/990 [==============================] - 0s 446us/step - loss: 0.0582 - val_loss: 0.1014\n",
- "Epoch 833/2500\n",
- "990/990 [==============================] - 0s 436us/step - loss: 0.0581 - val_loss: 0.1014\n",
- "Epoch 834/2500\n",
- "990/990 [==============================] - 0s 437us/step - loss: 0.0581 - val_loss: 0.1013\n",
- "Epoch 835/2500\n",
- "990/990 [==============================] - 0s 422us/step - loss: 0.0581 - val_loss: 0.1014\n",
- "Epoch 836/2500\n",
- "990/990 [==============================] - 0s 434us/step - loss: 0.0581 - val_loss: 0.1013\n",
- "Epoch 837/2500\n",
- "990/990 [==============================] - 0s 459us/step - loss: 0.0581 - val_loss: 0.1013\n",
- "Epoch 838/2500\n",
- "990/990 [==============================] - 0s 466us/step - loss: 0.0581 - val_loss: 0.1013\n",
- "Epoch 839/2500\n",
- "990/990 [==============================] - 0s 468us/step - loss: 0.0581 - val_loss: 0.1013\n",
- "Epoch 840/2500\n",
- "990/990 [==============================] - 0s 471us/step - loss: 0.0581 - val_loss: 0.1013\n",
- "Epoch 841/2500\n",
- "990/990 [==============================] - 0s 480us/step - loss: 0.0581 - val_loss: 0.1013\n",
- "Epoch 842/2500\n",
- "990/990 [==============================] - 0s 462us/step - loss: 0.0581 - val_loss: 0.1013\n",
- "Epoch 843/2500\n",
- "990/990 [==============================] - 0s 461us/step - loss: 0.0581 - val_loss: 0.1013\n",
- "Epoch 844/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0581 - val_loss: 0.1013\n",
- "Epoch 845/2500\n",
- "990/990 [==============================] - 0s 446us/step - loss: 0.0581 - val_loss: 0.1013\n",
- "Epoch 846/2500\n",
+ "Epoch 827/1200\n",
+ "990/990 [==============================] - 0s 436us/step - loss: 0.0582 - val_loss: 0.1014\n",
+ "Epoch 828/1200\n",
+ "990/990 [==============================] - 0s 457us/step - loss: 0.0582 - val_loss: 0.1014\n",
+ "Epoch 829/1200\n",
+ "990/990 [==============================] - 0s 440us/step - loss: 0.0582 - val_loss: 0.1013\n",
+ "Epoch 830/1200\n",
+ "990/990 [==============================] - 0s 450us/step - loss: 0.0582 - val_loss: 0.1014\n",
+ "Epoch 831/1200\n",
+ "990/990 [==============================] - 0s 429us/step - loss: 0.0582 - val_loss: 0.1013\n",
+ "Epoch 832/1200\n",
+ "990/990 [==============================] - 0s 437us/step - loss: 0.0582 - val_loss: 0.1014\n",
+ "Epoch 833/1200\n",
+ "990/990 [==============================] - 0s 430us/step - loss: 0.0581 - val_loss: 0.1014\n",
+ "Epoch 834/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0581 - val_loss: 0.1013\n",
+ "Epoch 835/1200\n",
+ "990/990 [==============================] - 0s 430us/step - loss: 0.0581 - val_loss: 0.1014\n",
+ "Epoch 836/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0581 - val_loss: 0.1013\n",
+ "Epoch 837/1200\n",
"990/990 [==============================] - 0s 455us/step - loss: 0.0581 - val_loss: 0.1013\n",
- "Epoch 847/2500\n",
- "990/990 [==============================] - 0s 458us/step - loss: 0.0580 - val_loss: 0.1013\n",
- "Epoch 848/2500\n"
+ "Epoch 838/1200\n",
+ "990/990 [==============================] - 0s 478us/step - loss: 0.0581 - val_loss: 0.1013\n",
+ "Epoch 839/1200\n",
+ "990/990 [==============================] - 0s 448us/step - loss: 0.0581 - val_loss: 0.1013\n",
+ "Epoch 840/1200\n",
+ "990/990 [==============================] - 0s 478us/step - loss: 0.0581 - val_loss: 0.1013\n",
+ "Epoch 841/1200\n",
+ "990/990 [==============================] - 0s 435us/step - loss: 0.0581 - val_loss: 0.1013\n",
+ "Epoch 842/1200\n",
+ "990/990 [==============================] - 0s 448us/step - loss: 0.0581 - val_loss: 0.1013\n",
+ "Epoch 843/1200\n",
+ "990/990 [==============================] - 0s 438us/step - loss: 0.0581 - val_loss: 0.1013\n",
+ "Epoch 844/1200\n",
+ "990/990 [==============================] - 0s 488us/step - loss: 0.0581 - val_loss: 0.1013\n",
+ "Epoch 845/1200\n",
+ "990/990 [==============================] - 0s 445us/step - loss: 0.0581 - val_loss: 0.1013\n",
+ "Epoch 846/1200\n",
+ "990/990 [==============================] - 0s 433us/step - loss: 0.0581 - val_loss: 0.1013\n",
+ "Epoch 847/1200\n",
+ "990/990 [==============================] - 0s 436us/step - loss: 0.0580 - val_loss: 0.1013\n",
+ "Epoch 848/1200\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
- "990/990 [==============================] - 0s 472us/step - loss: 0.0580 - val_loss: 0.1013\n",
- "Epoch 849/2500\n",
- "990/990 [==============================] - 0s 441us/step - loss: 0.0580 - val_loss: 0.1012\n",
- "Epoch 850/2500\n",
- "990/990 [==============================] - 0s 438us/step - loss: 0.0580 - val_loss: 0.1013\n",
- "Epoch 851/2500\n",
- "990/990 [==============================] - 0s 432us/step - loss: 0.0580 - val_loss: 0.1013\n",
- "Epoch 852/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0580 - val_loss: 0.1012\n",
- "Epoch 853/2500\n",
- "990/990 [==============================] - 0s 443us/step - loss: 0.0580 - val_loss: 0.1012\n",
- "Epoch 854/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0580 - val_loss: 0.1012\n",
- "Epoch 855/2500\n",
- "990/990 [==============================] - 0s 442us/step - loss: 0.0580 - val_loss: 0.1012\n",
- "Epoch 856/2500\n",
- "990/990 [==============================] - 0s 482us/step - loss: 0.0580 - val_loss: 0.1012\n",
- "Epoch 857/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0580 - val_loss: 0.1012\n",
- "Epoch 858/2500\n",
- "990/990 [==============================] - 0s 482us/step - loss: 0.0580 - val_loss: 0.1012\n",
- "Epoch 859/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0580 - val_loss: 0.1012\n",
- "Epoch 860/2500\n",
- "990/990 [==============================] - 0s 443us/step - loss: 0.0580 - val_loss: 0.1012\n",
- "Epoch 861/2500\n",
- "990/990 [==============================] - 0s 471us/step - loss: 0.0580 - val_loss: 0.1012\n",
- "Epoch 862/2500\n",
- "990/990 [==============================] - 0s 445us/step - loss: 0.0579 - val_loss: 0.1012\n",
- "Epoch 863/2500\n",
- "990/990 [==============================] - 0s 447us/step - loss: 0.0579 - val_loss: 0.1012\n",
- "Epoch 864/2500\n",
+ "990/990 [==============================] - 0s 433us/step - loss: 0.0580 - val_loss: 0.1012\n",
+ "Epoch 849/1200\n",
+ "990/990 [==============================] - 0s 432us/step - loss: 0.0580 - val_loss: 0.1012\n",
+ "Epoch 850/1200\n",
+ "990/990 [==============================] - 0s 435us/step - loss: 0.0580 - val_loss: 0.1013\n",
+ "Epoch 851/1200\n",
+ "990/990 [==============================] - 0s 441us/step - loss: 0.0580 - val_loss: 0.1013\n",
+ "Epoch 852/1200\n",
+ "990/990 [==============================] - 0s 456us/step - loss: 0.0580 - val_loss: 0.1012\n",
+ "Epoch 853/1200\n",
+ "990/990 [==============================] - 0s 450us/step - loss: 0.0580 - val_loss: 0.1012\n",
+ "Epoch 854/1200\n",
+ "990/990 [==============================] - 0s 431us/step - loss: 0.0580 - val_loss: 0.1012\n",
+ "Epoch 855/1200\n",
+ "990/990 [==============================] - 0s 427us/step - loss: 0.0580 - val_loss: 0.1012\n",
+ "Epoch 856/1200\n",
+ "990/990 [==============================] - 0s 436us/step - loss: 0.0580 - val_loss: 0.1012\n",
+ "Epoch 857/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0580 - val_loss: 0.1012\n",
+ "Epoch 858/1200\n",
+ "990/990 [==============================] - 0s 493us/step - loss: 0.0580 - val_loss: 0.1012\n",
+ "Epoch 859/1200\n",
+ "990/990 [==============================] - 0s 474us/step - loss: 0.0580 - val_loss: 0.1012\n",
+ "Epoch 860/1200\n",
+ "990/990 [==============================] - 0s 466us/step - loss: 0.0580 - val_loss: 0.1012\n",
+ "Epoch 861/1200\n",
+ "990/990 [==============================] - 0s 465us/step - loss: 0.0580 - val_loss: 0.1012\n",
+ "Epoch 862/1200\n",
+ "990/990 [==============================] - 0s 488us/step - loss: 0.0579 - val_loss: 0.1012\n",
+ "Epoch 863/1200\n",
"990/990 [==============================] - 0s 456us/step - loss: 0.0579 - val_loss: 0.1012\n",
- "Epoch 865/2500\n",
- "990/990 [==============================] - 0s 441us/step - loss: 0.0579 - val_loss: 0.1012\n",
- "Epoch 866/2500\n",
- "990/990 [==============================] - 0s 445us/step - loss: 0.0579 - val_loss: 0.1012\n",
- "Epoch 867/2500\n",
- "990/990 [==============================] - 0s 445us/step - loss: 0.0579 - val_loss: 0.1011\n",
- "Epoch 868/2500\n",
- "990/990 [==============================] - 0s 458us/step - loss: 0.0579 - val_loss: 0.1011\n",
- "Epoch 869/2500\n",
- "990/990 [==============================] - 0s 459us/step - loss: 0.0579 - val_loss: 0.1011\n",
- "Epoch 870/2500\n",
- "990/990 [==============================] - 0s 476us/step - loss: 0.0579 - val_loss: 0.1011\n",
- "Epoch 871/2500\n",
- "990/990 [==============================] - 0s 463us/step - loss: 0.0579 - val_loss: 0.1011\n",
- "Epoch 872/2500\n",
- "990/990 [==============================] - 0s 448us/step - loss: 0.0579 - val_loss: 0.1011\n",
- "Epoch 873/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0579 - val_loss: 0.1011\n",
- "Epoch 874/2500\n",
- "990/990 [==============================] - 0s 437us/step - loss: 0.0579 - val_loss: 0.1011\n",
- "Epoch 875/2500\n",
- "990/990 [==============================] - 0s 437us/step - loss: 0.0579 - val_loss: 0.1011\n",
- "Epoch 876/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0579 - val_loss: 0.1011\n",
- "Epoch 877/2500\n",
+ "Epoch 864/1200\n",
+ "990/990 [==============================] - 0s 497us/step - loss: 0.0579 - val_loss: 0.1012\n",
+ "Epoch 865/1200\n",
+ "990/990 [==============================] - 0s 471us/step - loss: 0.0579 - val_loss: 0.1012\n",
+ "Epoch 866/1200\n",
+ "990/990 [==============================] - 0s 450us/step - loss: 0.0579 - val_loss: 0.1012\n",
+ "Epoch 867/1200\n",
+ "990/990 [==============================] - 0s 488us/step - loss: 0.0579 - val_loss: 0.1011\n",
+ "Epoch 868/1200\n",
+ "990/990 [==============================] - 0s 451us/step - loss: 0.0579 - val_loss: 0.1011\n",
+ "Epoch 869/1200\n",
+ "990/990 [==============================] - 0s 434us/step - loss: 0.0579 - val_loss: 0.1011\n",
+ "Epoch 870/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0579 - val_loss: 0.1011\n",
+ "Epoch 871/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0579 - val_loss: 0.1011\n",
+ "Epoch 872/1200\n",
+ "990/990 [==============================] - 0s 430us/step - loss: 0.0579 - val_loss: 0.1011\n",
+ "Epoch 873/1200\n",
+ "990/990 [==============================] - 0s 431us/step - loss: 0.0579 - val_loss: 0.1011\n",
+ "Epoch 874/1200\n",
"990/990 [==============================] - 0s 428us/step - loss: 0.0579 - val_loss: 0.1011\n",
- "Epoch 878/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0578 - val_loss: 0.1011\n",
- "Epoch 879/2500\n",
- "990/990 [==============================] - 0s 463us/step - loss: 0.0578 - val_loss: 0.1011\n",
- "Epoch 880/2500\n",
- "990/990 [==============================] - 0s 473us/step - loss: 0.0578 - val_loss: 0.1011\n",
- "Epoch 881/2500\n",
- "990/990 [==============================] - 0s 477us/step - loss: 0.0578 - val_loss: 0.1011\n",
- "Epoch 882/2500\n",
- "990/990 [==============================] - 0s 465us/step - loss: 0.0578 - val_loss: 0.1011\n",
- "Epoch 883/2500\n",
- "990/990 [==============================] - 0s 487us/step - loss: 0.0578 - val_loss: 0.1011\n",
- "Epoch 884/2500\n",
- "990/990 [==============================] - 0s 472us/step - loss: 0.0578 - val_loss: 0.1011\n",
- "Epoch 885/2500\n",
- "990/990 [==============================] - 0s 447us/step - loss: 0.0578 - val_loss: 0.1010\n",
- "Epoch 886/2500\n",
- "990/990 [==============================] - 0s 446us/step - loss: 0.0578 - val_loss: 0.1010\n",
- "Epoch 887/2500\n",
- "990/990 [==============================] - 0s 443us/step - loss: 0.0578 - val_loss: 0.1010\n",
- "Epoch 888/2500\n",
- "990/990 [==============================] - 0s 441us/step - loss: 0.0578 - val_loss: 0.1010\n",
- "Epoch 889/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0578 - val_loss: 0.1011\n",
- "Epoch 890/2500\n",
- "990/990 [==============================] - 0s 468us/step - loss: 0.0578 - val_loss: 0.1011\n",
- "Epoch 891/2500\n",
- "990/990 [==============================] - 0s 469us/step - loss: 0.0578 - val_loss: 0.1011\n",
- "Epoch 892/2500\n",
- "990/990 [==============================] - 0s 440us/step - loss: 0.0578 - val_loss: 0.1011\n",
- "Epoch 893/2500\n",
- "990/990 [==============================] - 0s 431us/step - loss: 0.0578 - val_loss: 0.1010\n",
- "Epoch 894/2500\n",
- "990/990 [==============================] - 0s 444us/step - loss: 0.0577 - val_loss: 0.1010\n",
- "Epoch 895/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0577 - val_loss: 0.1010\n",
- "Epoch 896/2500\n",
- "990/990 [==============================] - 0s 426us/step - loss: 0.0577 - val_loss: 0.1011\n",
- "Epoch 897/2500\n",
- "990/990 [==============================] - 0s 474us/step - loss: 0.0577 - val_loss: 0.1010\n",
- "Epoch 898/2500\n",
- "990/990 [==============================] - 0s 462us/step - loss: 0.0577 - val_loss: 0.1010\n",
- "Epoch 899/2500\n",
- "990/990 [==============================] - 0s 473us/step - loss: 0.0577 - val_loss: 0.1010\n",
- "Epoch 900/2500\n",
- "990/990 [==============================] - 0s 440us/step - loss: 0.0577 - val_loss: 0.1010\n",
- "Epoch 901/2500\n",
- "990/990 [==============================] - 0s 432us/step - loss: 0.0577 - val_loss: 0.1010\n",
- "Epoch 902/2500\n",
- "990/990 [==============================] - 0s 437us/step - loss: 0.0577 - val_loss: 0.1010\n",
- "Epoch 903/2500\n",
- "990/990 [==============================] - 0s 425us/step - loss: 0.0577 - val_loss: 0.1010\n",
- "Epoch 904/2500\n",
- "990/990 [==============================] - 0s 430us/step - loss: 0.0577 - val_loss: 0.1010\n",
- "Epoch 905/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0577 - val_loss: 0.1010\n",
- "Epoch 906/2500\n",
- "990/990 [==============================] - 0s 422us/step - loss: 0.0577 - val_loss: 0.1010\n",
- "Epoch 907/2500\n",
+ "Epoch 875/1200\n",
+ "990/990 [==============================] - 0s 434us/step - loss: 0.0579 - val_loss: 0.1011\n",
+ "Epoch 876/1200\n",
+ "990/990 [==============================] - 0s 467us/step - loss: 0.0579 - val_loss: 0.1011\n",
+ "Epoch 877/1200\n",
+ "990/990 [==============================] - 0s 450us/step - loss: 0.0579 - val_loss: 0.1011\n",
+ "Epoch 878/1200\n",
+ "990/990 [==============================] - 0s 429us/step - loss: 0.0578 - val_loss: 0.1011\n",
+ "Epoch 879/1200\n",
+ "990/990 [==============================] - 0s 433us/step - loss: 0.0578 - val_loss: 0.1011\n",
+ "Epoch 880/1200\n",
+ "990/990 [==============================] - 0s 432us/step - loss: 0.0578 - val_loss: 0.1011\n",
+ "Epoch 881/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0578 - val_loss: 0.1011\n",
+ "Epoch 882/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0578 - val_loss: 0.1011\n",
+ "Epoch 883/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0578 - val_loss: 0.1011\n",
+ "Epoch 884/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0578 - val_loss: 0.1011\n",
+ "Epoch 885/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0578 - val_loss: 0.1010\n",
+ "Epoch 886/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0578 - val_loss: 0.1010\n",
+ "Epoch 887/1200\n",
+ "990/990 [==============================] - 0s 450us/step - loss: 0.0578 - val_loss: 0.1010\n",
+ "Epoch 888/1200\n",
+ "990/990 [==============================] - 0s 432us/step - loss: 0.0578 - val_loss: 0.1010\n",
+ "Epoch 889/1200\n",
+ "990/990 [==============================] - 0s 433us/step - loss: 0.0578 - val_loss: 0.1011\n",
+ "Epoch 890/1200\n",
+ "990/990 [==============================] - 0s 429us/step - loss: 0.0578 - val_loss: 0.1011\n",
+ "Epoch 891/1200\n",
+ "990/990 [==============================] - 0s 438us/step - loss: 0.0578 - val_loss: 0.1011\n",
+ "Epoch 892/1200\n",
+ "990/990 [==============================] - 0s 443us/step - loss: 0.0578 - val_loss: 0.1011\n",
+ "Epoch 893/1200\n",
+ "990/990 [==============================] - 0s 439us/step - loss: 0.0578 - val_loss: 0.1010\n",
+ "Epoch 894/1200\n",
+ "990/990 [==============================] - 1s 505us/step - loss: 0.0577 - val_loss: 0.1010\n",
+ "Epoch 895/1200\n",
+ "990/990 [==============================] - 0s 501us/step - loss: 0.0577 - val_loss: 0.1010\n",
+ "Epoch 896/1200\n",
+ "990/990 [==============================] - 0s 454us/step - loss: 0.0577 - val_loss: 0.1011\n",
+ "Epoch 897/1200\n",
+ "990/990 [==============================] - 0s 456us/step - loss: 0.0577 - val_loss: 0.1010\n",
+ "Epoch 898/1200\n",
+ "990/990 [==============================] - 0s 472us/step - loss: 0.0577 - val_loss: 0.1010\n",
+ "Epoch 899/1200\n",
"990/990 [==============================] - 0s 442us/step - loss: 0.0577 - val_loss: 0.1010\n",
- "Epoch 908/2500\n",
- "990/990 [==============================] - 0s 423us/step - loss: 0.0577 - val_loss: 0.1010\n",
- "Epoch 909/2500\n",
- "990/990 [==============================] - 0s 430us/step - loss: 0.0577 - val_loss: 0.1010\n",
- "Epoch 910/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0576 - val_loss: 0.1009\n",
- "Epoch 911/2500\n",
- "990/990 [==============================] - 0s 430us/step - loss: 0.0576 - val_loss: 0.1009\n",
- "Epoch 912/2500\n",
- "990/990 [==============================] - 0s 440us/step - loss: 0.0576 - val_loss: 0.1009\n",
- "Epoch 913/2500\n",
- "990/990 [==============================] - 0s 475us/step - loss: 0.0576 - val_loss: 0.1010\n",
- "Epoch 914/2500\n",
- "990/990 [==============================] - 0s 467us/step - loss: 0.0576 - val_loss: 0.1009\n",
- "Epoch 915/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0576 - val_loss: 0.1009\n",
- "Epoch 916/2500\n",
- "990/990 [==============================] - 0s 492us/step - loss: 0.0576 - val_loss: 0.1009\n",
- "Epoch 917/2500\n",
- "990/990 [==============================] - 0s 465us/step - loss: 0.0576 - val_loss: 0.1009\n",
- "Epoch 918/2500\n",
- "990/990 [==============================] - 0s 474us/step - loss: 0.0576 - val_loss: 0.1009\n",
- "Epoch 919/2500\n",
- "990/990 [==============================] - 0s 445us/step - loss: 0.0576 - val_loss: 0.1009\n",
- "Epoch 920/2500\n",
- "990/990 [==============================] - 0s 441us/step - loss: 0.0576 - val_loss: 0.1009\n",
- "Epoch 921/2500\n",
- "990/990 [==============================] - 0s 435us/step - loss: 0.0576 - val_loss: 0.1009\n",
- "Epoch 922/2500\n",
+ "Epoch 900/1200\n",
+ "990/990 [==============================] - 0s 479us/step - loss: 0.0577 - val_loss: 0.1010\n",
+ "Epoch 901/1200\n",
+ "990/990 [==============================] - 0s 456us/step - loss: 0.0577 - val_loss: 0.1010\n",
+ "Epoch 902/1200\n",
+ "990/990 [==============================] - 0s 487us/step - loss: 0.0577 - val_loss: 0.1010\n",
+ "Epoch 903/1200\n",
+ "990/990 [==============================] - 0s 485us/step - loss: 0.0577 - val_loss: 0.1010\n",
+ "Epoch 904/1200\n",
+ "990/990 [==============================] - 0s 486us/step - loss: 0.0577 - val_loss: 0.1010\n",
+ "Epoch 905/1200\n",
+ "990/990 [==============================] - 0s 463us/step - loss: 0.0577 - val_loss: 0.1010\n",
+ "Epoch 906/1200\n",
+ "990/990 [==============================] - 0s 464us/step - loss: 0.0577 - val_loss: 0.1010\n",
+ "Epoch 907/1200\n",
+ "990/990 [==============================] - 0s 476us/step - loss: 0.0577 - val_loss: 0.1010\n",
+ "Epoch 908/1200\n",
+ "990/990 [==============================] - 1s 557us/step - loss: 0.0577 - val_loss: 0.1010\n",
+ "Epoch 909/1200\n",
+ "990/990 [==============================] - 0s 480us/step - loss: 0.0577 - val_loss: 0.1010\n",
+ "Epoch 910/1200\n",
+ "990/990 [==============================] - 0s 482us/step - loss: 0.0576 - val_loss: 0.1009\n",
+ "Epoch 911/1200\n",
+ "990/990 [==============================] - 0s 449us/step - loss: 0.0576 - val_loss: 0.1009\n",
+ "Epoch 912/1200\n",
"990/990 [==============================] - 0s 440us/step - loss: 0.0576 - val_loss: 0.1009\n",
- "Epoch 923/2500\n",
- "990/990 [==============================] - 0s 443us/step - loss: 0.0576 - val_loss: 0.1009\n",
- "Epoch 924/2500\n",
- "990/990 [==============================] - 0s 435us/step - loss: 0.0576 - val_loss: 0.1009\n",
- "Epoch 925/2500\n"
+ "Epoch 913/1200\n",
+ "990/990 [==============================] - 0s 449us/step - loss: 0.0576 - val_loss: 0.1010\n",
+ "Epoch 914/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0576 - val_loss: 0.1009\n",
+ "Epoch 915/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0576 - val_loss: 0.1009\n",
+ "Epoch 916/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0576 - val_loss: 0.1009\n",
+ "Epoch 917/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0576 - val_loss: 0.1009\n",
+ "Epoch 918/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0576 - val_loss: 0.1009\n",
+ "Epoch 919/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0576 - val_loss: 0.1009\n",
+ "Epoch 920/1200\n",
+ "990/990 [==============================] - 0s 446us/step - loss: 0.0576 - val_loss: 0.1009\n",
+ "Epoch 921/1200\n",
+ "990/990 [==============================] - 0s 427us/step - loss: 0.0576 - val_loss: 0.1009\n",
+ "Epoch 922/1200\n",
+ "990/990 [==============================] - 0s 431us/step - loss: 0.0576 - val_loss: 0.1009\n",
+ "Epoch 923/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0576 - val_loss: 0.1009\n",
+ "Epoch 924/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0576 - val_loss: 0.1009\n",
+ "Epoch 925/1200\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
- "990/990 [==============================] - 0s 440us/step - loss: 0.0576 - val_loss: 0.1009\n",
- "Epoch 926/2500\n",
- "990/990 [==============================] - 0s 435us/step - loss: 0.0576 - val_loss: 0.1009\n",
- "Epoch 927/2500\n",
- "990/990 [==============================] - 0s 443us/step - loss: 0.0575 - val_loss: 0.1009\n",
- "Epoch 928/2500\n",
- "990/990 [==============================] - 0s 431us/step - loss: 0.0575 - val_loss: 0.1009\n",
- "Epoch 929/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0575 - val_loss: 0.1009\n",
- "Epoch 930/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0575 - val_loss: 0.1009\n",
- "Epoch 931/2500\n",
- "990/990 [==============================] - 0s 485us/step - loss: 0.0575 - val_loss: 0.1009\n",
- "Epoch 932/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0575 - val_loss: 0.1009\n",
- "Epoch 933/2500\n",
- "990/990 [==============================] - 0s 471us/step - loss: 0.0575 - val_loss: 0.1008\n",
- "Epoch 934/2500\n",
- "990/990 [==============================] - 0s 448us/step - loss: 0.0575 - val_loss: 0.1009\n",
- "Epoch 935/2500\n",
- "990/990 [==============================] - 0s 447us/step - loss: 0.0575 - val_loss: 0.1009\n",
- "Epoch 936/2500\n",
- "990/990 [==============================] - 0s 436us/step - loss: 0.0575 - val_loss: 0.1009\n",
- "Epoch 937/2500\n",
+ "990/990 [==============================] - 0s 436us/step - loss: 0.0576 - val_loss: 0.1009\n",
+ "Epoch 926/1200\n",
+ "990/990 [==============================] - 0s 431us/step - loss: 0.0576 - val_loss: 0.1009\n",
+ "Epoch 927/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0575 - val_loss: 0.1009\n",
+ "Epoch 928/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0575 - val_loss: 0.1009\n",
+ "Epoch 929/1200\n",
"990/990 [==============================] - 0s 444us/step - loss: 0.0575 - val_loss: 0.1009\n",
- "Epoch 938/2500\n",
- "990/990 [==============================] - 0s 472us/step - loss: 0.0575 - val_loss: 0.1009\n",
- "Epoch 939/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0575 - val_loss: 0.1009\n",
- "Epoch 940/2500\n",
- "990/990 [==============================] - 0s 442us/step - loss: 0.0575 - val_loss: 0.1008\n",
- "Epoch 941/2500\n",
- "990/990 [==============================] - 0s 439us/step - loss: 0.0575 - val_loss: 0.1009\n",
- "Epoch 942/2500\n",
- "990/990 [==============================] - 0s 438us/step - loss: 0.0575 - val_loss: 0.1008\n",
- "Epoch 943/2500\n",
- "990/990 [==============================] - 0s 438us/step - loss: 0.0575 - val_loss: 0.1008\n",
- "Epoch 944/2500\n",
- "990/990 [==============================] - 0s 436us/step - loss: 0.0574 - val_loss: 0.1009\n",
- "Epoch 945/2500\n",
+ "Epoch 930/1200\n",
+ "990/990 [==============================] - 0s 479us/step - loss: 0.0575 - val_loss: 0.1009\n",
+ "Epoch 931/1200\n",
+ "990/990 [==============================] - 0s 481us/step - loss: 0.0575 - val_loss: 0.1009\n",
+ "Epoch 932/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0575 - val_loss: 0.1009\n",
+ "Epoch 933/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0575 - val_loss: 0.1008\n",
+ "Epoch 934/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0575 - val_loss: 0.1009\n",
+ "Epoch 935/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0575 - val_loss: 0.1009\n",
+ "Epoch 936/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0575 - val_loss: 0.1009\n",
+ "Epoch 937/1200\n",
+ "990/990 [==============================] - 0s 415us/step - loss: 0.0575 - val_loss: 0.1009\n",
+ "Epoch 938/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0575 - val_loss: 0.1009\n",
+ "Epoch 939/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0575 - val_loss: 0.1009\n",
+ "Epoch 940/1200\n",
+ "990/990 [==============================] - 0s 439us/step - loss: 0.0575 - val_loss: 0.1008\n",
+ "Epoch 941/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0575 - val_loss: 0.1009\n",
+ "Epoch 942/1200\n",
+ "990/990 [==============================] - 0s 429us/step - loss: 0.0575 - val_loss: 0.1008\n",
+ "Epoch 943/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0575 - val_loss: 0.1008\n",
+ "Epoch 944/1200\n",
+ "990/990 [==============================] - 0s 429us/step - loss: 0.0574 - val_loss: 0.1009\n",
+ "Epoch 945/1200\n",
+ "990/990 [==============================] - 0s 439us/step - loss: 0.0574 - val_loss: 0.1009\n",
+ "Epoch 946/1200\n",
+ "990/990 [==============================] - 0s 470us/step - loss: 0.0574 - val_loss: 0.1008\n",
+ "Epoch 947/1200\n",
+ "990/990 [==============================] - 0s 451us/step - loss: 0.0574 - val_loss: 0.1009\n",
+ "Epoch 948/1200\n",
+ "990/990 [==============================] - 0s 459us/step - loss: 0.0574 - val_loss: 0.1008\n",
+ "Epoch 949/1200\n",
"990/990 [==============================] - 0s 443us/step - loss: 0.0574 - val_loss: 0.1009\n",
- "Epoch 946/2500\n",
- "990/990 [==============================] - 0s 436us/step - loss: 0.0574 - val_loss: 0.1008\n",
- "Epoch 947/2500\n",
+ "Epoch 950/1200\n",
+ "990/990 [==============================] - 0s 455us/step - loss: 0.0574 - val_loss: 0.1009\n",
+ "Epoch 951/1200\n",
+ "990/990 [==============================] - 0s 490us/step - loss: 0.0574 - val_loss: 0.1008\n",
+ "Epoch 952/1200\n",
+ "990/990 [==============================] - 0s 487us/step - loss: 0.0574 - val_loss: 0.1008\n",
+ "Epoch 953/1200\n",
"990/990 [==============================] - 0s 442us/step - loss: 0.0574 - val_loss: 0.1009\n",
- "Epoch 948/2500\n",
- "990/990 [==============================] - 0s 440us/step - loss: 0.0574 - val_loss: 0.1008\n",
- "Epoch 949/2500\n",
- "990/990 [==============================] - 0s 434us/step - loss: 0.0574 - val_loss: 0.1009\n",
- "Epoch 950/2500\n",
- "990/990 [==============================] - 0s 443us/step - loss: 0.0574 - val_loss: 0.1009\n",
- "Epoch 951/2500\n",
- "990/990 [==============================] - 0s 441us/step - loss: 0.0574 - val_loss: 0.1008\n",
- "Epoch 952/2500\n",
- "990/990 [==============================] - 0s 438us/step - loss: 0.0574 - val_loss: 0.1008\n",
- "Epoch 953/2500\n",
- "990/990 [==============================] - 0s 438us/step - loss: 0.0574 - val_loss: 0.1009\n",
- "Epoch 954/2500\n",
- "990/990 [==============================] - 0s 439us/step - loss: 0.0574 - val_loss: 0.1009\n",
- "Epoch 955/2500\n",
- "990/990 [==============================] - 0s 456us/step - loss: 0.0574 - val_loss: 0.1008\n",
- "Epoch 956/2500\n",
- "990/990 [==============================] - 0s 420us/step - loss: 0.0574 - val_loss: 0.1008\n",
- "Epoch 957/2500\n",
- "990/990 [==============================] - 0s 425us/step - loss: 0.0574 - val_loss: 0.1008\n",
- "Epoch 958/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0574 - val_loss: 0.1008\n",
- "Epoch 959/2500\n",
- "990/990 [==============================] - 0s 435us/step - loss: 0.0574 - val_loss: 0.1008\n",
- "Epoch 960/2500\n",
- "990/990 [==============================] - 0s 465us/step - loss: 0.0574 - val_loss: 0.1008\n",
- "Epoch 961/2500\n",
- "990/990 [==============================] - 0s 481us/step - loss: 0.0574 - val_loss: 0.1008\n",
- "Epoch 962/2500\n",
- "990/990 [==============================] - 0s 458us/step - loss: 0.0573 - val_loss: 0.1008\n",
- "Epoch 963/2500\n",
- "990/990 [==============================] - 0s 477us/step - loss: 0.0573 - val_loss: 0.1008\n",
- "Epoch 964/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0573 - val_loss: 0.1008\n",
- "Epoch 965/2500\n",
- "990/990 [==============================] - 0s 462us/step - loss: 0.0573 - val_loss: 0.1008\n",
- "Epoch 966/2500\n",
+ "Epoch 954/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0574 - val_loss: 0.1009\n",
+ "Epoch 955/1200\n",
+ "990/990 [==============================] - 0s 430us/step - loss: 0.0574 - val_loss: 0.1008\n",
+ "Epoch 956/1200\n",
+ "990/990 [==============================] - 0s 433us/step - loss: 0.0574 - val_loss: 0.1008\n",
+ "Epoch 957/1200\n",
+ "990/990 [==============================] - 0s 437us/step - loss: 0.0574 - val_loss: 0.1008\n",
+ "Epoch 958/1200\n",
+ "990/990 [==============================] - 0s 431us/step - loss: 0.0574 - val_loss: 0.1008\n",
+ "Epoch 959/1200\n",
+ "990/990 [==============================] - 0s 437us/step - loss: 0.0574 - val_loss: 0.1008\n",
+ "Epoch 960/1200\n",
+ "990/990 [==============================] - 0s 437us/step - loss: 0.0574 - val_loss: 0.1008\n",
+ "Epoch 961/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0574 - val_loss: 0.1008\n",
+ "Epoch 962/1200\n",
+ "990/990 [==============================] - 0s 469us/step - loss: 0.0573 - val_loss: 0.1008\n",
+ "Epoch 963/1200\n",
"990/990 [==============================] - 0s 459us/step - loss: 0.0573 - val_loss: 0.1008\n",
- "Epoch 967/2500\n",
- "990/990 [==============================] - 0s 436us/step - loss: 0.0573 - val_loss: 0.1007\n",
- "Epoch 968/2500\n",
- "990/990 [==============================] - 0s 448us/step - loss: 0.0573 - val_loss: 0.1008\n",
- "Epoch 969/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0573 - val_loss: 0.1008\n",
- "Epoch 970/2500\n",
- "990/990 [==============================] - 0s 447us/step - loss: 0.0573 - val_loss: 0.1008\n",
- "Epoch 971/2500\n",
- "990/990 [==============================] - 0s 456us/step - loss: 0.0573 - val_loss: 0.1008\n",
- "Epoch 972/2500\n",
- "990/990 [==============================] - 0s 460us/step - loss: 0.0573 - val_loss: 0.1007\n",
- "Epoch 973/2500\n",
+ "Epoch 964/1200\n",
+ "990/990 [==============================] - 0s 475us/step - loss: 0.0573 - val_loss: 0.1008\n",
+ "Epoch 965/1200\n",
+ "990/990 [==============================] - 0s 468us/step - loss: 0.0573 - val_loss: 0.1008\n",
+ "Epoch 966/1200\n",
+ "990/990 [==============================] - 0s 485us/step - loss: 0.0573 - val_loss: 0.1008\n",
+ "Epoch 967/1200\n",
+ "990/990 [==============================] - 0s 459us/step - loss: 0.0573 - val_loss: 0.1007\n",
+ "Epoch 968/1200\n",
+ "990/990 [==============================] - 0s 467us/step - loss: 0.0573 - val_loss: 0.1008\n",
+ "Epoch 969/1200\n",
+ "990/990 [==============================] - 0s 465us/step - loss: 0.0573 - val_loss: 0.1008\n",
+ "Epoch 970/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0573 - val_loss: 0.1008\n",
+ "Epoch 971/1200\n",
+ "990/990 [==============================] - 0s 443us/step - loss: 0.0573 - val_loss: 0.1008\n",
+ "Epoch 972/1200\n",
+ "990/990 [==============================] - 0s 491us/step - loss: 0.0573 - val_loss: 0.1007\n",
+ "Epoch 973/1200\n",
+ "990/990 [==============================] - 0s 483us/step - loss: 0.0573 - val_loss: 0.1008\n",
+ "Epoch 974/1200\n",
+ "990/990 [==============================] - 0s 478us/step - loss: 0.0573 - val_loss: 0.1008\n",
+ "Epoch 975/1200\n",
"990/990 [==============================] - 0s 445us/step - loss: 0.0573 - val_loss: 0.1008\n",
- "Epoch 974/2500\n",
- "990/990 [==============================] - 0s 464us/step - loss: 0.0573 - val_loss: 0.1008\n",
- "Epoch 975/2500\n",
- "990/990 [==============================] - 0s 448us/step - loss: 0.0573 - val_loss: 0.1008\n",
- "Epoch 976/2500\n",
+ "Epoch 976/1200\n",
+ "990/990 [==============================] - 0s 481us/step - loss: 0.0573 - val_loss: 0.1008\n",
+ "Epoch 977/1200\n",
+ "990/990 [==============================] - 0s 440us/step - loss: 0.0573 - val_loss: 0.1007\n",
+ "Epoch 978/1200\n",
+ "990/990 [==============================] - 0s 441us/step - loss: 0.0573 - val_loss: 0.1008\n",
+ "Epoch 979/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0573 - val_loss: 0.1008\n",
+ "Epoch 980/1200\n",
"990/990 [==============================] - 0s 430us/step - loss: 0.0573 - val_loss: 0.1008\n",
- "Epoch 977/2500\n",
- "990/990 [==============================] - 0s 430us/step - loss: 0.0573 - val_loss: 0.1007\n",
- "Epoch 978/2500\n",
- "990/990 [==============================] - 0s 431us/step - loss: 0.0573 - val_loss: 0.1008\n",
- "Epoch 979/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0573 - val_loss: 0.1008\n",
- "Epoch 980/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0573 - val_loss: 0.1008\n",
- "Epoch 981/2500\n",
- "990/990 [==============================] - 0s 426us/step - loss: 0.0572 - val_loss: 0.1008\n",
- "Epoch 982/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0572 - val_loss: 0.1007\n",
- "Epoch 983/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0572 - val_loss: 0.1008\n",
- "Epoch 984/2500\n",
+ "Epoch 981/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0572 - val_loss: 0.1008\n",
+ "Epoch 982/1200\n",
"990/990 [==============================] - 0s 429us/step - loss: 0.0572 - val_loss: 0.1007\n",
- "Epoch 985/2500\n",
+ "Epoch 983/1200\n",
"990/990 [==============================] - 0s 429us/step - loss: 0.0572 - val_loss: 0.1008\n",
- "Epoch 986/2500\n",
+ "Epoch 984/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0572 - val_loss: 0.1007\n",
+ "Epoch 985/1200\n",
+ "990/990 [==============================] - 0s 434us/step - loss: 0.0572 - val_loss: 0.1008\n",
+ "Epoch 986/1200\n",
"990/990 [==============================] - 0s 424us/step - loss: 0.0572 - val_loss: 0.1008\n",
- "Epoch 987/2500\n",
- "990/990 [==============================] - 0s 431us/step - loss: 0.0572 - val_loss: 0.1007\n",
- "Epoch 988/2500\n",
- "990/990 [==============================] - 0s 431us/step - loss: 0.0572 - val_loss: 0.1007\n",
- "Epoch 989/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0572 - val_loss: 0.1007\n",
- "Epoch 990/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0572 - val_loss: 0.1007\n",
- "Epoch 991/2500\n",
- "990/990 [==============================] - 0s 430us/step - loss: 0.0572 - val_loss: 0.1007\n",
- "Epoch 992/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0572 - val_loss: 0.1007\n",
- "Epoch 993/2500\n",
+ "Epoch 987/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0572 - val_loss: 0.1007\n",
+ "Epoch 988/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0572 - val_loss: 0.1007\n",
+ "Epoch 989/1200\n",
"990/990 [==============================] - 0s 430us/step - loss: 0.0572 - val_loss: 0.1007\n",
- "Epoch 994/2500\n",
- "990/990 [==============================] - 0s 431us/step - loss: 0.0572 - val_loss: 0.1007\n",
- "Epoch 995/2500\n",
- "990/990 [==============================] - 0s 432us/step - loss: 0.0572 - val_loss: 0.1007\n",
- "Epoch 996/2500\n",
+ "Epoch 990/1200\n",
"990/990 [==============================] - 0s 427us/step - loss: 0.0572 - val_loss: 0.1007\n",
- "Epoch 997/2500\n",
- "990/990 [==============================] - 0s 435us/step - loss: 0.0572 - val_loss: 0.1007\n",
- "Epoch 998/2500\n",
- "990/990 [==============================] - 0s 425us/step - loss: 0.0572 - val_loss: 0.1007\n",
- "Epoch 999/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0572 - val_loss: 0.1007\n",
- "Epoch 1000/2500\n",
- "990/990 [==============================] - 0s 424us/step - loss: 0.0571 - val_loss: 0.1007\n",
- "Epoch 1001/2500\n",
- "990/990 [==============================] - 0s 424us/step - loss: 0.0571 - val_loss: 0.1007\n",
- "Epoch 1002/2500\n"
+ "Epoch 991/1200\n",
+ "990/990 [==============================] - 0s 475us/step - loss: 0.0572 - val_loss: 0.1007\n",
+ "Epoch 992/1200\n",
+ "990/990 [==============================] - 0s 467us/step - loss: 0.0572 - val_loss: 0.1007\n",
+ "Epoch 993/1200\n",
+ "990/990 [==============================] - 0s 452us/step - loss: 0.0572 - val_loss: 0.1007\n",
+ "Epoch 994/1200\n",
+ "990/990 [==============================] - 0s 434us/step - loss: 0.0572 - val_loss: 0.1007\n",
+ "Epoch 995/1200\n",
+ "990/990 [==============================] - 0s 450us/step - loss: 0.0572 - val_loss: 0.1007\n",
+ "Epoch 996/1200\n",
+ "990/990 [==============================] - 0s 432us/step - loss: 0.0572 - val_loss: 0.1007\n",
+ "Epoch 997/1200\n",
+ "990/990 [==============================] - 0s 437us/step - loss: 0.0572 - val_loss: 0.1007\n",
+ "Epoch 998/1200\n",
+ "990/990 [==============================] - 0s 468us/step - loss: 0.0572 - val_loss: 0.1007\n",
+ "Epoch 999/1200\n",
+ "990/990 [==============================] - 0s 451us/step - loss: 0.0572 - val_loss: 0.1007\n",
+ "Epoch 1000/1200\n",
+ "990/990 [==============================] - 0s 438us/step - loss: 0.0571 - val_loss: 0.1007\n",
+ "Epoch 1001/1200\n",
+ "990/990 [==============================] - 0s 459us/step - loss: 0.0571 - val_loss: 0.1007\n",
+ "Epoch 1002/1200\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
+ "990/990 [==============================] - 0s 441us/step - loss: 0.0571 - val_loss: 0.1007\n",
+ "Epoch 1003/1200\n",
+ "990/990 [==============================] - 0s 458us/step - loss: 0.0571 - val_loss: 0.1007\n",
+ "Epoch 1004/1200\n",
+ "990/990 [==============================] - 0s 498us/step - loss: 0.0571 - val_loss: 0.1007\n",
+ "Epoch 1005/1200\n",
+ "990/990 [==============================] - 1s 505us/step - loss: 0.0571 - val_loss: 0.1007\n",
+ "Epoch 1006/1200\n",
+ "990/990 [==============================] - 0s 503us/step - loss: 0.0571 - val_loss: 0.1007\n",
+ "Epoch 1007/1200\n",
+ "990/990 [==============================] - 0s 477us/step - loss: 0.0571 - val_loss: 0.1007\n",
+ "Epoch 1008/1200\n",
+ "990/990 [==============================] - 0s 465us/step - loss: 0.0571 - val_loss: 0.1007\n",
+ "Epoch 1009/1200\n",
+ "990/990 [==============================] - 0s 455us/step - loss: 0.0571 - val_loss: 0.1007\n",
+ "Epoch 1010/1200\n",
+ "990/990 [==============================] - 0s 458us/step - loss: 0.0571 - val_loss: 0.1007\n",
+ "Epoch 1011/1200\n",
+ "990/990 [==============================] - 0s 451us/step - loss: 0.0571 - val_loss: 0.1007\n",
+ "Epoch 1012/1200\n",
+ "990/990 [==============================] - 0s 457us/step - loss: 0.0571 - val_loss: 0.1006\n",
+ "Epoch 1013/1200\n",
+ "990/990 [==============================] - 0s 443us/step - loss: 0.0571 - val_loss: 0.1007\n",
+ "Epoch 1014/1200\n",
"990/990 [==============================] - 0s 447us/step - loss: 0.0571 - val_loss: 0.1007\n",
- "Epoch 1003/2500\n",
- "990/990 [==============================] - 0s 423us/step - loss: 0.0571 - val_loss: 0.1007\n",
- "Epoch 1004/2500\n",
- "990/990 [==============================] - 0s 439us/step - loss: 0.0571 - val_loss: 0.1007\n",
- "Epoch 1005/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0571 - val_loss: 0.1007\n",
- "Epoch 1006/2500\n",
- "990/990 [==============================] - 0s 474us/step - loss: 0.0571 - val_loss: 0.1007\n",
- "Epoch 1007/2500\n",
- "990/990 [==============================] - 0s 460us/step - loss: 0.0571 - val_loss: 0.1007\n",
- "Epoch 1008/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0571 - val_loss: 0.1007\n",
- "Epoch 1009/2500\n",
- "990/990 [==============================] - 0s 448us/step - loss: 0.0571 - val_loss: 0.1007\n",
- "Epoch 1010/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0571 - val_loss: 0.1007\n",
- "Epoch 1011/2500\n",
- "990/990 [==============================] - 0s 438us/step - loss: 0.0571 - val_loss: 0.1007\n",
- "Epoch 1012/2500\n",
- "990/990 [==============================] - 0s 447us/step - loss: 0.0571 - val_loss: 0.1006\n",
- "Epoch 1013/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0571 - val_loss: 0.1007\n",
- "Epoch 1014/2500\n",
- "990/990 [==============================] - 0s 480us/step - loss: 0.0571 - val_loss: 0.1007\n",
- "Epoch 1015/2500\n",
+ "Epoch 1015/1200\n",
"990/990 [==============================] - 0s 441us/step - loss: 0.0571 - val_loss: 0.1007\n",
- "Epoch 1016/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0571 - val_loss: 0.1007\n",
- "Epoch 1017/2500\n",
- "990/990 [==============================] - 0s 437us/step - loss: 0.0571 - val_loss: 0.1007\n",
- "Epoch 1018/2500\n",
- "990/990 [==============================] - 0s 432us/step - loss: 0.0571 - val_loss: 0.1007\n",
- "Epoch 1019/2500\n",
- "990/990 [==============================] - 0s 426us/step - loss: 0.0571 - val_loss: 0.1007\n",
- "Epoch 1020/2500\n",
- "990/990 [==============================] - 0s 442us/step - loss: 0.0570 - val_loss: 0.1006\n",
- "Epoch 1021/2500\n",
- "990/990 [==============================] - 0s 432us/step - loss: 0.0570 - val_loss: 0.1007\n",
- "Epoch 1022/2500\n",
- "990/990 [==============================] - 0s 432us/step - loss: 0.0570 - val_loss: 0.1006\n",
- "Epoch 1023/2500\n",
- "990/990 [==============================] - 0s 459us/step - loss: 0.0570 - val_loss: 0.1006\n",
- "Epoch 1024/2500\n",
+ "Epoch 1016/1200\n",
+ "990/990 [==============================] - 0s 439us/step - loss: 0.0571 - val_loss: 0.1007\n",
+ "Epoch 1017/1200\n",
+ "990/990 [==============================] - 0s 470us/step - loss: 0.0571 - val_loss: 0.1007\n",
+ "Epoch 1018/1200\n",
+ "990/990 [==============================] - 0s 499us/step - loss: 0.0571 - val_loss: 0.1007\n",
+ "Epoch 1019/1200\n",
+ "990/990 [==============================] - 0s 468us/step - loss: 0.0571 - val_loss: 0.1007\n",
+ "Epoch 1020/1200\n",
+ "990/990 [==============================] - 0s 492us/step - loss: 0.0570 - val_loss: 0.1006\n",
+ "Epoch 1021/1200\n",
+ "990/990 [==============================] - 0s 453us/step - loss: 0.0570 - val_loss: 0.1007\n",
+ "Epoch 1022/1200\n",
+ "990/990 [==============================] - 0s 465us/step - loss: 0.0570 - val_loss: 0.1006\n",
+ "Epoch 1023/1200\n",
"990/990 [==============================] - 0s 437us/step - loss: 0.0570 - val_loss: 0.1006\n",
- "Epoch 1025/2500\n",
- "990/990 [==============================] - 0s 440us/step - loss: 0.0570 - val_loss: 0.1006\n",
- "Epoch 1026/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0570 - val_loss: 0.1007\n",
- "Epoch 1027/2500\n",
- "990/990 [==============================] - 0s 485us/step - loss: 0.0570 - val_loss: 0.1006\n",
- "Epoch 1028/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0570 - val_loss: 0.1007\n",
- "Epoch 1029/2500\n",
- "990/990 [==============================] - 0s 459us/step - loss: 0.0570 - val_loss: 0.1007\n",
- "Epoch 1030/2500\n",
- "990/990 [==============================] - 0s 444us/step - loss: 0.0570 - val_loss: 0.1007\n",
- "Epoch 1031/2500\n",
- "990/990 [==============================] - 0s 488us/step - loss: 0.0570 - val_loss: 0.1006\n",
- "Epoch 1032/2500\n",
- "990/990 [==============================] - 0s 466us/step - loss: 0.0570 - val_loss: 0.1007\n",
- "Epoch 1033/2500\n",
- "990/990 [==============================] - 0s 440us/step - loss: 0.0570 - val_loss: 0.1007\n",
- "Epoch 1034/2500\n",
- "990/990 [==============================] - 0s 446us/step - loss: 0.0570 - val_loss: 0.1006\n",
- "Epoch 1035/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0570 - val_loss: 0.1006\n",
- "Epoch 1036/2500\n",
- "990/990 [==============================] - 0s 458us/step - loss: 0.0570 - val_loss: 0.1007\n",
- "Epoch 1037/2500\n",
- "990/990 [==============================] - 0s 468us/step - loss: 0.0570 - val_loss: 0.1007\n",
- "Epoch 1038/2500\n",
- "990/990 [==============================] - 0s 469us/step - loss: 0.0570 - val_loss: 0.1006\n",
- "Epoch 1039/2500\n",
- "990/990 [==============================] - 0s 434us/step - loss: 0.0570 - val_loss: 0.1007\n",
- "Epoch 1040/2500\n",
- "990/990 [==============================] - 0s 436us/step - loss: 0.0569 - val_loss: 0.1007\n",
- "Epoch 1041/2500\n",
- "990/990 [==============================] - 0s 433us/step - loss: 0.0569 - val_loss: 0.1007\n",
- "Epoch 1042/2500\n",
- "990/990 [==============================] - 0s 420us/step - loss: 0.0569 - val_loss: 0.1007\n",
- "Epoch 1043/2500\n",
- "990/990 [==============================] - 0s 446us/step - loss: 0.0569 - val_loss: 0.1007\n",
- "Epoch 1044/2500\n",
- "990/990 [==============================] - 0s 430us/step - loss: 0.0569 - val_loss: 0.1006\n",
- "Epoch 1045/2500\n",
- "990/990 [==============================] - 0s 441us/step - loss: 0.0569 - val_loss: 0.1006\n",
- "Epoch 1046/2500\n",
- "990/990 [==============================] - 0s 463us/step - loss: 0.0569 - val_loss: 0.1006\n",
- "Epoch 1047/2500\n",
- "990/990 [==============================] - 0s 466us/step - loss: 0.0569 - val_loss: 0.1006\n",
- "Epoch 1048/2500\n",
- "990/990 [==============================] - 0s 492us/step - loss: 0.0569 - val_loss: 0.1006\n",
- "Epoch 1049/2500\n",
- "990/990 [==============================] - 0s 468us/step - loss: 0.0569 - val_loss: 0.1006\n",
- "Epoch 1050/2500\n",
- "990/990 [==============================] - 0s 462us/step - loss: 0.0569 - val_loss: 0.1006\n",
- "Epoch 1051/2500\n",
- "990/990 [==============================] - 0s 483us/step - loss: 0.0569 - val_loss: 0.1006\n",
- "Epoch 1052/2500\n",
- "990/990 [==============================] - 0s 469us/step - loss: 0.0569 - val_loss: 0.1006\n",
- "Epoch 1053/2500\n",
- "990/990 [==============================] - 0s 468us/step - loss: 0.0569 - val_loss: 0.1006\n",
- "Epoch 1054/2500\n",
- "990/990 [==============================] - 0s 497us/step - loss: 0.0569 - val_loss: 0.1006\n",
- "Epoch 1055/2500\n",
- "990/990 [==============================] - 0s 461us/step - loss: 0.0569 - val_loss: 0.1006\n",
- "Epoch 1056/2500\n",
- "990/990 [==============================] - 0s 483us/step - loss: 0.0569 - val_loss: 0.1006\n",
- "Epoch 1057/2500\n",
- "990/990 [==============================] - 0s 467us/step - loss: 0.0569 - val_loss: 0.1006\n",
- "Epoch 1058/2500\n",
- "990/990 [==============================] - 0s 460us/step - loss: 0.0569 - val_loss: 0.1006\n",
- "Epoch 1059/2500\n",
- "990/990 [==============================] - 0s 467us/step - loss: 0.0569 - val_loss: 0.1006\n",
- "Epoch 1060/2500\n",
- "990/990 [==============================] - 0s 470us/step - loss: 0.0569 - val_loss: 0.1006\n",
- "Epoch 1061/2500\n",
- "990/990 [==============================] - 0s 491us/step - loss: 0.0569 - val_loss: 0.1006\n",
- "Epoch 1062/2500\n",
- "990/990 [==============================] - 0s 487us/step - loss: 0.0568 - val_loss: 0.1006\n",
- "Epoch 1063/2500\n",
- "990/990 [==============================] - 0s 489us/step - loss: 0.0568 - val_loss: 0.1006\n",
- "Epoch 1064/2500\n",
- "990/990 [==============================] - 0s 487us/step - loss: 0.0568 - val_loss: 0.1006\n",
- "Epoch 1065/2500\n",
- "990/990 [==============================] - 0s 492us/step - loss: 0.0568 - val_loss: 0.1006\n",
- "Epoch 1066/2500\n",
- "990/990 [==============================] - 0s 477us/step - loss: 0.0568 - val_loss: 0.1006\n",
- "Epoch 1067/2500\n",
- "990/990 [==============================] - 0s 469us/step - loss: 0.0568 - val_loss: 0.1006\n",
- "Epoch 1068/2500\n",
- "990/990 [==============================] - 0s 456us/step - loss: 0.0568 - val_loss: 0.1006\n",
- "Epoch 1069/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0568 - val_loss: 0.1006\n",
- "Epoch 1070/2500\n",
- "990/990 [==============================] - 0s 494us/step - loss: 0.0568 - val_loss: 0.1006\n",
- "Epoch 1071/2500\n",
- "990/990 [==============================] - 0s 497us/step - loss: 0.0568 - val_loss: 0.1006\n",
- "Epoch 1072/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0568 - val_loss: 0.1006\n",
- "Epoch 1073/2500\n",
- "990/990 [==============================] - 0s 448us/step - loss: 0.0568 - val_loss: 0.1006\n",
- "Epoch 1074/2500\n",
+ "Epoch 1024/1200\n",
+ "990/990 [==============================] - 0s 436us/step - loss: 0.0570 - val_loss: 0.1006\n",
+ "Epoch 1025/1200\n",
+ "990/990 [==============================] - 0s 441us/step - loss: 0.0570 - val_loss: 0.1006\n",
+ "Epoch 1026/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0570 - val_loss: 0.1007\n",
+ "Epoch 1027/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0570 - val_loss: 0.1006\n",
+ "Epoch 1028/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0570 - val_loss: 0.1007\n",
+ "Epoch 1029/1200\n",
+ "990/990 [==============================] - 0s 433us/step - loss: 0.0570 - val_loss: 0.1007\n",
+ "Epoch 1030/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0570 - val_loss: 0.1007\n",
+ "Epoch 1031/1200\n",
+ "990/990 [==============================] - 0s 430us/step - loss: 0.0570 - val_loss: 0.1006\n",
+ "Epoch 1032/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0570 - val_loss: 0.1007\n",
+ "Epoch 1033/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0570 - val_loss: 0.1007\n",
+ "Epoch 1034/1200\n",
+ "990/990 [==============================] - 0s 449us/step - loss: 0.0570 - val_loss: 0.1006\n",
+ "Epoch 1035/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0570 - val_loss: 0.1006\n",
+ "Epoch 1036/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0570 - val_loss: 0.1007\n",
+ "Epoch 1037/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0570 - val_loss: 0.1007\n",
+ "Epoch 1038/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0570 - val_loss: 0.1006\n",
+ "Epoch 1039/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0570 - val_loss: 0.1007\n",
+ "Epoch 1040/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0569 - val_loss: 0.1007\n",
+ "Epoch 1041/1200\n",
+ "990/990 [==============================] - 0s 430us/step - loss: 0.0569 - val_loss: 0.1007\n",
+ "Epoch 1042/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0569 - val_loss: 0.1007\n",
+ "Epoch 1043/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0569 - val_loss: 0.1007\n",
+ "Epoch 1044/1200\n",
+ "990/990 [==============================] - 0s 438us/step - loss: 0.0569 - val_loss: 0.1006\n",
+ "Epoch 1045/1200\n",
+ "990/990 [==============================] - 0s 434us/step - loss: 0.0569 - val_loss: 0.1006\n",
+ "Epoch 1046/1200\n",
+ "990/990 [==============================] - 0s 444us/step - loss: 0.0569 - val_loss: 0.1006\n",
+ "Epoch 1047/1200\n",
+ "990/990 [==============================] - 0s 448us/step - loss: 0.0569 - val_loss: 0.1006\n",
+ "Epoch 1048/1200\n",
+ "990/990 [==============================] - 0s 428us/step - loss: 0.0569 - val_loss: 0.1006\n",
+ "Epoch 1049/1200\n",
+ "990/990 [==============================] - 0s 452us/step - loss: 0.0569 - val_loss: 0.1006\n",
+ "Epoch 1050/1200\n",
+ "990/990 [==============================] - 0s 450us/step - loss: 0.0569 - val_loss: 0.1006\n",
+ "Epoch 1051/1200\n",
+ "990/990 [==============================] - 0s 433us/step - loss: 0.0569 - val_loss: 0.1006\n",
+ "Epoch 1052/1200\n",
+ "990/990 [==============================] - 0s 444us/step - loss: 0.0569 - val_loss: 0.1006\n",
+ "Epoch 1053/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0569 - val_loss: 0.1006\n",
+ "Epoch 1054/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0569 - val_loss: 0.1006\n",
+ "Epoch 1055/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0569 - val_loss: 0.1006\n",
+ "Epoch 1056/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0569 - val_loss: 0.1006\n",
+ "Epoch 1057/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0569 - val_loss: 0.1006\n",
+ "Epoch 1058/1200\n",
+ "990/990 [==============================] - 0s 443us/step - loss: 0.0569 - val_loss: 0.1006\n",
+ "Epoch 1059/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0569 - val_loss: 0.1006\n",
+ "Epoch 1060/1200\n",
+ "990/990 [==============================] - 0s 447us/step - loss: 0.0569 - val_loss: 0.1006\n",
+ "Epoch 1061/1200\n",
+ "990/990 [==============================] - 0s 438us/step - loss: 0.0569 - val_loss: 0.1006\n",
+ "Epoch 1062/1200\n",
+ "990/990 [==============================] - 0s 417us/step - loss: 0.0568 - val_loss: 0.1006\n",
+ "Epoch 1063/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0568 - val_loss: 0.1006\n",
+ "Epoch 1064/1200\n",
+ "990/990 [==============================] - 0s 414us/step - loss: 0.0568 - val_loss: 0.1006\n",
+ "Epoch 1065/1200\n",
+ "990/990 [==============================] - 0s 446us/step - loss: 0.0568 - val_loss: 0.1006\n",
+ "Epoch 1066/1200\n",
+ "990/990 [==============================] - 0s 417us/step - loss: 0.0568 - val_loss: 0.1006\n",
+ "Epoch 1067/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0568 - val_loss: 0.1006\n",
+ "Epoch 1068/1200\n",
+ "990/990 [==============================] - 0s 451us/step - loss: 0.0568 - val_loss: 0.1006\n",
+ "Epoch 1069/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0568 - val_loss: 0.1006\n",
+ "Epoch 1070/1200\n",
+ "990/990 [==============================] - 0s 435us/step - loss: 0.0568 - val_loss: 0.1006\n",
+ "Epoch 1071/1200\n",
+ "990/990 [==============================] - 0s 443us/step - loss: 0.0568 - val_loss: 0.1006\n",
+ "Epoch 1072/1200\n",
+ "990/990 [==============================] - 0s 449us/step - loss: 0.0568 - val_loss: 0.1006\n",
+ "Epoch 1073/1200\n",
"990/990 [==============================] - 0s 450us/step - loss: 0.0568 - val_loss: 0.1006\n",
- "Epoch 1075/2500\n",
- "990/990 [==============================] - 0s 467us/step - loss: 0.0568 - val_loss: 0.1006\n",
- "Epoch 1076/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0568 - val_loss: 0.1006\n",
- "Epoch 1077/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0568 - val_loss: 0.1006\n",
- "Epoch 1078/2500\n"
+ "Epoch 1074/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0568 - val_loss: 0.1006\n",
+ "Epoch 1075/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0568 - val_loss: 0.1006\n",
+ "Epoch 1076/1200\n",
+ "990/990 [==============================] - 0s 415us/step - loss: 0.0568 - val_loss: 0.1006\n",
+ "Epoch 1077/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0568 - val_loss: 0.1006\n",
+ "Epoch 1078/1200\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
- "990/990 [==============================] - 0s 458us/step - loss: 0.0568 - val_loss: 0.1006\n",
- "Epoch 1079/2500\n",
- "990/990 [==============================] - 0s 458us/step - loss: 0.0568 - val_loss: 0.1006\n",
- "Epoch 1080/2500\n",
- "990/990 [==============================] - 0s 458us/step - loss: 0.0568 - val_loss: 0.1006\n",
- "Epoch 1081/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0568 - val_loss: 0.1006\n",
- "Epoch 1082/2500\n",
- "990/990 [==============================] - 0s 460us/step - loss: 0.0568 - val_loss: 0.1006\n",
- "Epoch 1083/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0568 - val_loss: 0.1006\n",
- "Epoch 1084/2500\n",
- "990/990 [==============================] - 0s 466us/step - loss: 0.0567 - val_loss: 0.1006\n",
- "Epoch 1085/2500\n",
- "990/990 [==============================] - 0s 461us/step - loss: 0.0567 - val_loss: 0.1006\n",
- "Epoch 1086/2500\n",
- "990/990 [==============================] - 0s 467us/step - loss: 0.0567 - val_loss: 0.1006\n",
- "Epoch 1087/2500\n",
- "990/990 [==============================] - 0s 445us/step - loss: 0.0567 - val_loss: 0.1006\n",
- "Epoch 1088/2500\n",
- "990/990 [==============================] - 0s 462us/step - loss: 0.0567 - val_loss: 0.1006\n",
- "Epoch 1089/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0567 - val_loss: 0.1006\n",
- "Epoch 1090/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0567 - val_loss: 0.1006\n",
- "Epoch 1091/2500\n",
- "990/990 [==============================] - 0s 469us/step - loss: 0.0567 - val_loss: 0.1006\n",
- "Epoch 1092/2500\n",
- "990/990 [==============================] - 0s 465us/step - loss: 0.0567 - val_loss: 0.1006\n",
- "Epoch 1093/2500\n",
- "990/990 [==============================] - 0s 464us/step - loss: 0.0567 - val_loss: 0.1006\n",
- "Epoch 1094/2500\n",
- "990/990 [==============================] - 0s 499us/step - loss: 0.0567 - val_loss: 0.1006\n",
- "Epoch 1095/2500\n",
- "990/990 [==============================] - 0s 474us/step - loss: 0.0567 - val_loss: 0.1006\n",
- "Epoch 1096/2500\n",
- "990/990 [==============================] - 0s 499us/step - loss: 0.0567 - val_loss: 0.1006\n",
- "Epoch 1097/2500\n",
- "990/990 [==============================] - 0s 483us/step - loss: 0.0567 - val_loss: 0.1006\n",
- "Epoch 1098/2500\n",
- "990/990 [==============================] - 0s 477us/step - loss: 0.0567 - val_loss: 0.1006\n",
- "Epoch 1099/2500\n",
- "990/990 [==============================] - 0s 471us/step - loss: 0.0567 - val_loss: 0.1006\n",
- "Epoch 1100/2500\n",
- "990/990 [==============================] - 0s 486us/step - loss: 0.0567 - val_loss: 0.1006\n",
- "Epoch 1101/2500\n",
- "990/990 [==============================] - 0s 475us/step - loss: 0.0567 - val_loss: 0.1006\n",
- "Epoch 1102/2500\n",
- "990/990 [==============================] - 0s 478us/step - loss: 0.0567 - val_loss: 0.1006\n",
- "Epoch 1103/2500\n",
- "990/990 [==============================] - 0s 471us/step - loss: 0.0567 - val_loss: 0.1006\n",
- "Epoch 1104/2500\n",
- "990/990 [==============================] - 0s 464us/step - loss: 0.0567 - val_loss: 0.1006\n",
- "Epoch 1105/2500\n",
- "990/990 [==============================] - 0s 471us/step - loss: 0.0567 - val_loss: 0.1006\n",
- "Epoch 1106/2500\n",
- "990/990 [==============================] - 0s 478us/step - loss: 0.0567 - val_loss: 0.1006\n",
- "Epoch 1107/2500\n",
- "990/990 [==============================] - 0s 480us/step - loss: 0.0566 - val_loss: 0.1006\n",
- "Epoch 1108/2500\n",
- "990/990 [==============================] - 0s 480us/step - loss: 0.0566 - val_loss: 0.1006\n",
- "Epoch 1109/2500\n",
- "990/990 [==============================] - 0s 480us/step - loss: 0.0566 - val_loss: 0.1006\n",
- "Epoch 1110/2500\n",
- "990/990 [==============================] - 0s 470us/step - loss: 0.0566 - val_loss: 0.1005\n",
- "Epoch 1111/2500\n",
- "990/990 [==============================] - 0s 477us/step - loss: 0.0566 - val_loss: 0.1006\n",
- "Epoch 1112/2500\n",
- "990/990 [==============================] - 0s 462us/step - loss: 0.0566 - val_loss: 0.1006\n",
- "Epoch 1113/2500\n",
- "990/990 [==============================] - 0s 481us/step - loss: 0.0566 - val_loss: 0.1006\n",
- "Epoch 1114/2500\n",
- "990/990 [==============================] - 0s 471us/step - loss: 0.0566 - val_loss: 0.1006\n",
- "Epoch 1115/2500\n",
- "990/990 [==============================] - 0s 466us/step - loss: 0.0566 - val_loss: 0.1006\n",
- "Epoch 1116/2500\n",
- "990/990 [==============================] - 0s 479us/step - loss: 0.0566 - val_loss: 0.1006\n",
- "Epoch 1117/2500\n",
- "990/990 [==============================] - 0s 477us/step - loss: 0.0566 - val_loss: 0.1006\n",
- "Epoch 1118/2500\n",
- "990/990 [==============================] - 0s 482us/step - loss: 0.0566 - val_loss: 0.1006\n",
- "Epoch 1119/2500\n",
- "990/990 [==============================] - 0s 476us/step - loss: 0.0566 - val_loss: 0.1006\n",
- "Epoch 1120/2500\n",
- "990/990 [==============================] - 0s 480us/step - loss: 0.0566 - val_loss: 0.1006\n",
- "Epoch 1121/2500\n",
- "990/990 [==============================] - 0s 465us/step - loss: 0.0566 - val_loss: 0.1006\n",
- "Epoch 1122/2500\n",
- "990/990 [==============================] - 0s 479us/step - loss: 0.0566 - val_loss: 0.1006\n",
- "Epoch 1123/2500\n",
- "990/990 [==============================] - 0s 469us/step - loss: 0.0566 - val_loss: 0.1005\n",
- "Epoch 1124/2500\n",
- "990/990 [==============================] - 0s 467us/step - loss: 0.0566 - val_loss: 0.1006\n",
- "Epoch 1125/2500\n",
- "990/990 [==============================] - 0s 486us/step - loss: 0.0566 - val_loss: 0.1006\n",
- "Epoch 1126/2500\n",
- "990/990 [==============================] - 0s 472us/step - loss: 0.0566 - val_loss: 0.1006\n",
- "Epoch 1127/2500\n",
- "990/990 [==============================] - 0s 468us/step - loss: 0.0566 - val_loss: 0.1006\n",
- "Epoch 1128/2500\n",
- "990/990 [==============================] - 0s 464us/step - loss: 0.0566 - val_loss: 0.1006\n",
- "Epoch 1129/2500\n",
- "990/990 [==============================] - 0s 475us/step - loss: 0.0566 - val_loss: 0.1006\n",
- "Epoch 1130/2500\n",
- "990/990 [==============================] - 0s 463us/step - loss: 0.0565 - val_loss: 0.1006\n",
- "Epoch 1131/2500\n",
- "990/990 [==============================] - 0s 483us/step - loss: 0.0565 - val_loss: 0.1006\n",
- "Epoch 1132/2500\n",
- "990/990 [==============================] - 0s 473us/step - loss: 0.0565 - val_loss: 0.1006\n",
- "Epoch 1133/2500\n",
- "990/990 [==============================] - 0s 481us/step - loss: 0.0565 - val_loss: 0.1006\n",
- "Epoch 1134/2500\n",
- "990/990 [==============================] - 0s 471us/step - loss: 0.0565 - val_loss: 0.1006\n",
- "Epoch 1135/2500\n",
- "990/990 [==============================] - 0s 481us/step - loss: 0.0565 - val_loss: 0.1006\n",
- "Epoch 1136/2500\n",
- "990/990 [==============================] - 0s 469us/step - loss: 0.0565 - val_loss: 0.1006\n",
- "Epoch 1137/2500\n",
- "990/990 [==============================] - 0s 481us/step - loss: 0.0565 - val_loss: 0.1006\n",
- "Epoch 1138/2500\n",
- "990/990 [==============================] - 0s 476us/step - loss: 0.0565 - val_loss: 0.1006\n",
- "Epoch 1139/2500\n",
- "990/990 [==============================] - 0s 463us/step - loss: 0.0565 - val_loss: 0.1006\n",
- "Epoch 1140/2500\n",
- "990/990 [==============================] - 0s 466us/step - loss: 0.0565 - val_loss: 0.1005\n",
- "Epoch 1141/2500\n",
- "990/990 [==============================] - 0s 463us/step - loss: 0.0565 - val_loss: 0.1006\n",
- "Epoch 1142/2500\n",
- "990/990 [==============================] - 0s 481us/step - loss: 0.0565 - val_loss: 0.1006\n",
- "Epoch 1143/2500\n",
- "990/990 [==============================] - 0s 472us/step - loss: 0.0565 - val_loss: 0.1006\n",
- "Epoch 1144/2500\n",
- "990/990 [==============================] - 0s 472us/step - loss: 0.0565 - val_loss: 0.1006\n",
- "Epoch 1145/2500\n",
- "990/990 [==============================] - 0s 471us/step - loss: 0.0565 - val_loss: 0.1006\n",
- "Epoch 1146/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0565 - val_loss: 0.1006\n",
- "Epoch 1147/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0565 - val_loss: 0.1006\n",
- "Epoch 1148/2500\n",
- "990/990 [==============================] - 0s 444us/step - loss: 0.0565 - val_loss: 0.1006\n",
- "Epoch 1149/2500\n",
- "990/990 [==============================] - 0s 441us/step - loss: 0.0565 - val_loss: 0.1006\n",
- "Epoch 1150/2500\n",
- "990/990 [==============================] - 0s 447us/step - loss: 0.0565 - val_loss: 0.1006\n",
- "Epoch 1151/2500\n",
- "990/990 [==============================] - 0s 439us/step - loss: 0.0565 - val_loss: 0.1006\n",
- "Epoch 1152/2500\n",
- "990/990 [==============================] - 0s 441us/step - loss: 0.0565 - val_loss: 0.1006\n",
- "Epoch 1153/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0565 - val_loss: 0.1006\n",
- "Epoch 1154/2500\n"
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0568 - val_loss: 0.1006\n",
+ "Epoch 1079/1200\n",
+ "990/990 [==============================] - 0s 415us/step - loss: 0.0568 - val_loss: 0.1006\n",
+ "Epoch 1080/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0568 - val_loss: 0.1006\n",
+ "Epoch 1081/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0568 - val_loss: 0.1006\n",
+ "Epoch 1082/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0568 - val_loss: 0.1006\n",
+ "Epoch 1083/1200\n",
+ "990/990 [==============================] - 0s 431us/step - loss: 0.0568 - val_loss: 0.1006\n",
+ "Epoch 1084/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0567 - val_loss: 0.1006\n",
+ "Epoch 1085/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0567 - val_loss: 0.1006\n",
+ "Epoch 1086/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0567 - val_loss: 0.1006\n",
+ "Epoch 1087/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0567 - val_loss: 0.1006\n",
+ "Epoch 1088/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0567 - val_loss: 0.1006\n",
+ "Epoch 1089/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0567 - val_loss: 0.1006\n",
+ "Epoch 1090/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0567 - val_loss: 0.1006\n",
+ "Epoch 1091/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0567 - val_loss: 0.1006\n",
+ "Epoch 1092/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0567 - val_loss: 0.1006\n",
+ "Epoch 1093/1200\n",
+ "990/990 [==============================] - 0s 417us/step - loss: 0.0567 - val_loss: 0.1006\n",
+ "Epoch 1094/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0567 - val_loss: 0.1006\n",
+ "Epoch 1095/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0567 - val_loss: 0.1006\n",
+ "Epoch 1096/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0567 - val_loss: 0.1006\n",
+ "Epoch 1097/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0567 - val_loss: 0.1006\n",
+ "Epoch 1098/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0567 - val_loss: 0.1006\n",
+ "Epoch 1099/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0567 - val_loss: 0.1006\n",
+ "Epoch 1100/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0567 - val_loss: 0.1006\n",
+ "Epoch 1101/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0567 - val_loss: 0.1006\n",
+ "Epoch 1102/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0567 - val_loss: 0.1006\n",
+ "Epoch 1103/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0567 - val_loss: 0.1006\n",
+ "Epoch 1104/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0567 - val_loss: 0.1006\n",
+ "Epoch 1105/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0567 - val_loss: 0.1006\n",
+ "Epoch 1106/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0567 - val_loss: 0.1006\n",
+ "Epoch 1107/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0566 - val_loss: 0.1006\n",
+ "Epoch 1108/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0566 - val_loss: 0.1006\n",
+ "Epoch 1109/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0566 - val_loss: 0.1006\n",
+ "Epoch 1110/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0566 - val_loss: 0.1005\n",
+ "Epoch 1111/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0566 - val_loss: 0.1006\n",
+ "Epoch 1112/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0566 - val_loss: 0.1006\n",
+ "Epoch 1113/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0566 - val_loss: 0.1006\n",
+ "Epoch 1114/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0566 - val_loss: 0.1006\n",
+ "Epoch 1115/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0566 - val_loss: 0.1006\n",
+ "Epoch 1116/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0566 - val_loss: 0.1006\n",
+ "Epoch 1117/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0566 - val_loss: 0.1006\n",
+ "Epoch 1118/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0566 - val_loss: 0.1006\n",
+ "Epoch 1119/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0566 - val_loss: 0.1006\n",
+ "Epoch 1120/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0566 - val_loss: 0.1006\n",
+ "Epoch 1121/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0566 - val_loss: 0.1006\n",
+ "Epoch 1122/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0566 - val_loss: 0.1006\n",
+ "Epoch 1123/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0566 - val_loss: 0.1005\n",
+ "Epoch 1124/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0566 - val_loss: 0.1006\n",
+ "Epoch 1125/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0566 - val_loss: 0.1006\n",
+ "Epoch 1126/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0566 - val_loss: 0.1006\n",
+ "Epoch 1127/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0566 - val_loss: 0.1006\n",
+ "Epoch 1128/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0566 - val_loss: 0.1006\n",
+ "Epoch 1129/1200\n",
+ "990/990 [==============================] - 0s 414us/step - loss: 0.0566 - val_loss: 0.1006\n",
+ "Epoch 1130/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0565 - val_loss: 0.1006\n",
+ "Epoch 1131/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0565 - val_loss: 0.1006\n",
+ "Epoch 1132/1200\n",
+ "990/990 [==============================] - 0s 415us/step - loss: 0.0565 - val_loss: 0.1006\n",
+ "Epoch 1133/1200\n",
+ "990/990 [==============================] - 0s 416us/step - loss: 0.0565 - val_loss: 0.1006\n",
+ "Epoch 1134/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0565 - val_loss: 0.1006\n",
+ "Epoch 1135/1200\n",
+ "990/990 [==============================] - 0s 434us/step - loss: 0.0565 - val_loss: 0.1006\n",
+ "Epoch 1136/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0565 - val_loss: 0.1006\n",
+ "Epoch 1137/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0565 - val_loss: 0.1006\n",
+ "Epoch 1138/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0565 - val_loss: 0.1006\n",
+ "Epoch 1139/1200\n",
+ "990/990 [==============================] - 0s 426us/step - loss: 0.0565 - val_loss: 0.1006\n",
+ "Epoch 1140/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0565 - val_loss: 0.1005\n",
+ "Epoch 1141/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0565 - val_loss: 0.1006\n",
+ "Epoch 1142/1200\n",
+ "990/990 [==============================] - 0s 417us/step - loss: 0.0565 - val_loss: 0.1006\n",
+ "Epoch 1143/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0565 - val_loss: 0.1006\n",
+ "Epoch 1144/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0565 - val_loss: 0.1006\n",
+ "Epoch 1145/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0565 - val_loss: 0.1006\n",
+ "Epoch 1146/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0565 - val_loss: 0.1006\n",
+ "Epoch 1147/1200\n",
+ "990/990 [==============================] - 0s 429us/step - loss: 0.0565 - val_loss: 0.1006\n",
+ "Epoch 1148/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0565 - val_loss: 0.1006\n",
+ "Epoch 1149/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0565 - val_loss: 0.1006\n",
+ "Epoch 1150/1200\n",
+ "990/990 [==============================] - 0s 417us/step - loss: 0.0565 - val_loss: 0.1006\n",
+ "Epoch 1151/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0565 - val_loss: 0.1006\n",
+ "Epoch 1152/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0565 - val_loss: 0.1006\n",
+ "Epoch 1153/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0565 - val_loss: 0.1006\n",
+ "Epoch 1154/1200\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
- "990/990 [==============================] - 0s 442us/step - loss: 0.0565 - val_loss: 0.1006\n",
- "Epoch 1155/2500\n",
- "990/990 [==============================] - 0s 441us/step - loss: 0.0564 - val_loss: 0.1006\n",
- "Epoch 1156/2500\n",
- "990/990 [==============================] - 0s 437us/step - loss: 0.0564 - val_loss: 0.1006\n",
- "Epoch 1157/2500\n",
- "990/990 [==============================] - 0s 443us/step - loss: 0.0564 - val_loss: 0.1006\n",
- "Epoch 1158/2500\n",
- "990/990 [==============================] - 0s 437us/step - loss: 0.0564 - val_loss: 0.1006\n",
- "Epoch 1159/2500\n",
- "990/990 [==============================] - 0s 437us/step - loss: 0.0564 - val_loss: 0.1006\n",
- "Epoch 1160/2500\n",
- "990/990 [==============================] - 0s 438us/step - loss: 0.0564 - val_loss: 0.1006\n",
- "Epoch 1161/2500\n",
- "990/990 [==============================] - 0s 436us/step - loss: 0.0564 - val_loss: 0.1005\n",
- "Epoch 1162/2500\n",
- "990/990 [==============================] - 0s 437us/step - loss: 0.0564 - val_loss: 0.1005\n",
- "Epoch 1163/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0564 - val_loss: 0.1006\n",
- "Epoch 1164/2500\n",
- "990/990 [==============================] - 0s 442us/step - loss: 0.0564 - val_loss: 0.1005\n",
- "Epoch 1165/2500\n",
- "990/990 [==============================] - 0s 440us/step - loss: 0.0564 - val_loss: 0.1006\n",
- "Epoch 1166/2500\n",
- "990/990 [==============================] - 0s 440us/step - loss: 0.0564 - val_loss: 0.1006\n",
- "Epoch 1167/2500\n",
- "990/990 [==============================] - 0s 440us/step - loss: 0.0564 - val_loss: 0.1006\n",
- "Epoch 1168/2500\n",
- "990/990 [==============================] - 0s 440us/step - loss: 0.0564 - val_loss: 0.1006\n",
- "Epoch 1169/2500\n",
- "990/990 [==============================] - 0s 439us/step - loss: 0.0564 - val_loss: 0.1006\n",
- "Epoch 1170/2500\n",
- "990/990 [==============================] - 0s 437us/step - loss: 0.0564 - val_loss: 0.1006\n",
- "Epoch 1171/2500\n",
- "990/990 [==============================] - 0s 441us/step - loss: 0.0564 - val_loss: 0.1006\n",
- "Epoch 1172/2500\n",
- "990/990 [==============================] - 0s 442us/step - loss: 0.0564 - val_loss: 0.1006\n",
- "Epoch 1173/2500\n",
- "990/990 [==============================] - 0s 440us/step - loss: 0.0564 - val_loss: 0.1006\n",
- "Epoch 1174/2500\n",
- "990/990 [==============================] - 0s 433us/step - loss: 0.0564 - val_loss: 0.1006\n",
- "Epoch 1175/2500\n",
- "990/990 [==============================] - 0s 444us/step - loss: 0.0564 - val_loss: 0.1006\n",
- "Epoch 1176/2500\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0565 - val_loss: 0.1006\n",
+ "Epoch 1155/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0564 - val_loss: 0.1006\n",
+ "Epoch 1156/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0564 - val_loss: 0.1006\n",
+ "Epoch 1157/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0564 - val_loss: 0.1006\n",
+ "Epoch 1158/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0564 - val_loss: 0.1006\n",
+ "Epoch 1159/1200\n",
+ "990/990 [==============================] - 0s 417us/step - loss: 0.0564 - val_loss: 0.1006\n",
+ "Epoch 1160/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0564 - val_loss: 0.1006\n",
+ "Epoch 1161/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0564 - val_loss: 0.1005\n",
+ "Epoch 1162/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0564 - val_loss: 0.1005\n",
+ "Epoch 1163/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0564 - val_loss: 0.1006\n",
+ "Epoch 1164/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0564 - val_loss: 0.1005\n",
+ "Epoch 1165/1200\n",
+ "990/990 [==============================] - 0s 417us/step - loss: 0.0564 - val_loss: 0.1006\n",
+ "Epoch 1166/1200\n",
+ "990/990 [==============================] - 0s 417us/step - loss: 0.0564 - val_loss: 0.1006\n",
+ "Epoch 1167/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0564 - val_loss: 0.1006\n",
+ "Epoch 1168/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0564 - val_loss: 0.1006\n",
+ "Epoch 1169/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0564 - val_loss: 0.1006\n",
+ "Epoch 1170/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0564 - val_loss: 0.1006\n",
+ "Epoch 1171/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0564 - val_loss: 0.1006\n",
+ "Epoch 1172/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0564 - val_loss: 0.1006\n",
+ "Epoch 1173/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0564 - val_loss: 0.1006\n",
+ "Epoch 1174/1200\n",
+ "990/990 [==============================] - 0s 424us/step - loss: 0.0564 - val_loss: 0.1006\n",
+ "Epoch 1175/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0564 - val_loss: 0.1006\n",
+ "Epoch 1176/1200\n",
+ "990/990 [==============================] - 0s 450us/step - loss: 0.0564 - val_loss: 0.1006\n",
+ "Epoch 1177/1200\n",
+ "990/990 [==============================] - 0s 450us/step - loss: 0.0564 - val_loss: 0.1006\n",
+ "Epoch 1178/1200\n",
"990/990 [==============================] - 0s 435us/step - loss: 0.0564 - val_loss: 0.1006\n",
- "Epoch 1177/2500\n",
- "990/990 [==============================] - 0s 437us/step - loss: 0.0564 - val_loss: 0.1006\n",
- "Epoch 1178/2500\n",
- "990/990 [==============================] - 0s 444us/step - loss: 0.0564 - val_loss: 0.1006\n",
- "Epoch 1179/2500\n",
- "990/990 [==============================] - 0s 438us/step - loss: 0.0564 - val_loss: 0.1006\n",
- "Epoch 1180/2500\n",
- "990/990 [==============================] - 0s 446us/step - loss: 0.0563 - val_loss: 0.1006\n",
- "Epoch 1181/2500\n",
- "990/990 [==============================] - 0s 437us/step - loss: 0.0563 - val_loss: 0.1006\n",
- "Epoch 1182/2500\n",
- "990/990 [==============================] - 0s 443us/step - loss: 0.0563 - val_loss: 0.1006\n",
- "Epoch 1183/2500\n",
- "990/990 [==============================] - 0s 439us/step - loss: 0.0563 - val_loss: 0.1006\n",
- "Epoch 1184/2500\n",
- "990/990 [==============================] - 0s 437us/step - loss: 0.0563 - val_loss: 0.1006\n",
- "Epoch 1185/2500\n",
+ "Epoch 1179/1200\n",
+ "990/990 [==============================] - 0s 434us/step - loss: 0.0564 - val_loss: 0.1006\n",
+ "Epoch 1180/1200\n",
"990/990 [==============================] - 0s 435us/step - loss: 0.0563 - val_loss: 0.1006\n",
- "Epoch 1186/2500\n",
- "990/990 [==============================] - 0s 431us/step - loss: 0.0563 - val_loss: 0.1006\n",
- "Epoch 1187/2500\n",
- "990/990 [==============================] - 0s 434us/step - loss: 0.0563 - val_loss: 0.1006\n",
- "Epoch 1188/2500\n",
+ "Epoch 1181/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0563 - val_loss: 0.1006\n",
+ "Epoch 1182/1200\n",
+ "990/990 [==============================] - 0s 453us/step - loss: 0.0563 - val_loss: 0.1006\n",
+ "Epoch 1183/1200\n",
"990/990 [==============================] - 0s 439us/step - loss: 0.0563 - val_loss: 0.1006\n",
- "Epoch 1189/2500\n",
- "990/990 [==============================] - 0s 440us/step - loss: 0.0563 - val_loss: 0.1006\n",
- "Epoch 1190/2500\n",
- "990/990 [==============================] - 0s 438us/step - loss: 0.0563 - val_loss: 0.1006\n",
- "Epoch 1191/2500\n",
- "990/990 [==============================] - 0s 435us/step - loss: 0.0563 - val_loss: 0.1006\n",
- "Epoch 1192/2500\n",
- "990/990 [==============================] - 0s 443us/step - loss: 0.0563 - val_loss: 0.1006\n",
- "Epoch 1193/2500\n",
+ "Epoch 1184/1200\n",
"990/990 [==============================] - 0s 436us/step - loss: 0.0563 - val_loss: 0.1006\n",
- "Epoch 1194/2500\n",
- "990/990 [==============================] - 0s 441us/step - loss: 0.0563 - val_loss: 0.1006\n",
- "Epoch 1195/2500\n",
- "990/990 [==============================] - 0s 439us/step - loss: 0.0563 - val_loss: 0.1006\n",
- "Epoch 1196/2500\n",
+ "Epoch 1185/1200\n",
+ "990/990 [==============================] - 0s 435us/step - loss: 0.0563 - val_loss: 0.1006\n",
+ "Epoch 1186/1200\n",
"990/990 [==============================] - 0s 447us/step - loss: 0.0563 - val_loss: 0.1006\n",
- "Epoch 1197/2500\n",
- "990/990 [==============================] - 0s 444us/step - loss: 0.0563 - val_loss: 0.1006\n",
- "Epoch 1198/2500\n",
- "990/990 [==============================] - 0s 440us/step - loss: 0.0563 - val_loss: 0.1006\n",
- "Epoch 1199/2500\n",
- "990/990 [==============================] - 0s 437us/step - loss: 0.0563 - val_loss: 0.1006\n",
- "Epoch 1200/2500\n",
- "990/990 [==============================] - 0s 443us/step - loss: 0.0563 - val_loss: 0.1006\n",
- "Epoch 1201/2500\n",
- "990/990 [==============================] - 0s 441us/step - loss: 0.0563 - val_loss: 0.1006\n",
- "Epoch 1202/2500\n",
- "990/990 [==============================] - 0s 440us/step - loss: 0.0563 - val_loss: 0.1006\n",
- "Epoch 1203/2500\n",
- "990/990 [==============================] - 0s 437us/step - loss: 0.0563 - val_loss: 0.1006\n",
- "Epoch 1204/2500\n",
- "990/990 [==============================] - 0s 439us/step - loss: 0.0563 - val_loss: 0.1006\n",
- "Epoch 1205/2500\n",
- "990/990 [==============================] - 0s 446us/step - loss: 0.0563 - val_loss: 0.1006\n",
- "Epoch 1206/2500\n",
- "990/990 [==============================] - 0s 440us/step - loss: 0.0563 - val_loss: 0.1006\n",
- "Epoch 1207/2500\n",
- "990/990 [==============================] - 0s 443us/step - loss: 0.0562 - val_loss: 0.1006\n",
- "Epoch 1208/2500\n",
- "990/990 [==============================] - 0s 447us/step - loss: 0.0562 - val_loss: 0.1006\n",
- "Epoch 1209/2500\n",
- "990/990 [==============================] - 0s 464us/step - loss: 0.0562 - val_loss: 0.1006\n",
- "Epoch 1210/2500\n",
- "990/990 [==============================] - 0s 444us/step - loss: 0.0562 - val_loss: 0.1006\n",
- "Epoch 1211/2500\n",
- "990/990 [==============================] - 0s 456us/step - loss: 0.0562 - val_loss: 0.1006\n",
- "Epoch 1212/2500\n",
- "990/990 [==============================] - 0s 443us/step - loss: 0.0562 - val_loss: 0.1006\n",
- "Epoch 1213/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0562 - val_loss: 0.1006\n",
- "Epoch 1214/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0562 - val_loss: 0.1006\n",
- "Epoch 1215/2500\n",
- "990/990 [==============================] - 0s 435us/step - loss: 0.0562 - val_loss: 0.1006\n",
- "Epoch 1216/2500\n",
- "990/990 [==============================] - 0s 485us/step - loss: 0.0562 - val_loss: 0.1006\n",
- "Epoch 1217/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0562 - val_loss: 0.1006\n",
- "Epoch 1218/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0562 - val_loss: 0.1006\n",
- "Epoch 1219/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0562 - val_loss: 0.1006\n",
- "Epoch 1220/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0562 - val_loss: 0.1006\n",
- "Epoch 1221/2500\n",
- "990/990 [==============================] - 0s 461us/step - loss: 0.0562 - val_loss: 0.1006\n",
- "Epoch 1222/2500\n",
- "990/990 [==============================] - 0s 447us/step - loss: 0.0562 - val_loss: 0.1006\n",
- "Epoch 1223/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0562 - val_loss: 0.1006\n",
- "Epoch 1224/2500\n",
- "990/990 [==============================] - 0s 469us/step - loss: 0.0562 - val_loss: 0.1006\n",
- "Epoch 1225/2500\n",
- "990/990 [==============================] - 0s 456us/step - loss: 0.0562 - val_loss: 0.1006\n",
- "Epoch 1226/2500\n",
- "990/990 [==============================] - 0s 444us/step - loss: 0.0562 - val_loss: 0.1006\n",
- "Epoch 1227/2500\n",
- "990/990 [==============================] - 0s 479us/step - loss: 0.0562 - val_loss: 0.1006\n",
- "Epoch 1228/2500\n",
- "990/990 [==============================] - 0s 486us/step - loss: 0.0562 - val_loss: 0.1006\n",
- "Epoch 1229/2500\n",
- "990/990 [==============================] - 0s 477us/step - loss: 0.0562 - val_loss: 0.1006\n",
- "Epoch 1230/2500\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "990/990 [==============================] - 0s 483us/step - loss: 0.0562 - val_loss: 0.1006\n",
- "Epoch 1231/2500\n",
- "990/990 [==============================] - 0s 469us/step - loss: 0.0562 - val_loss: 0.1006\n",
- "Epoch 1232/2500\n",
- "990/990 [==============================] - 0s 479us/step - loss: 0.0562 - val_loss: 0.1006\n",
- "Epoch 1233/2500\n",
- "990/990 [==============================] - 0s 479us/step - loss: 0.0562 - val_loss: 0.1006\n",
- "Epoch 1234/2500\n",
- "990/990 [==============================] - 0s 446us/step - loss: 0.0561 - val_loss: 0.1006\n",
- "Epoch 1235/2500\n",
- "990/990 [==============================] - 0s 459us/step - loss: 0.0561 - val_loss: 0.1006\n",
- "Epoch 1236/2500\n",
- "990/990 [==============================] - 0s 470us/step - loss: 0.0561 - val_loss: 0.1006\n",
- "Epoch 1237/2500\n",
- "990/990 [==============================] - 0s 481us/step - loss: 0.0561 - val_loss: 0.1006\n",
- "Epoch 1238/2500\n",
- "990/990 [==============================] - 0s 479us/step - loss: 0.0561 - val_loss: 0.1006\n",
- "Epoch 1239/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0561 - val_loss: 0.1006\n",
- "Epoch 1240/2500\n",
- "990/990 [==============================] - 0s 463us/step - loss: 0.0561 - val_loss: 0.1006\n",
- "Epoch 1241/2500\n",
- "990/990 [==============================] - 1s 518us/step - loss: 0.0561 - val_loss: 0.1006\n",
- "Epoch 1242/2500\n",
- "990/990 [==============================] - 0s 482us/step - loss: 0.0561 - val_loss: 0.1006\n",
- "Epoch 1243/2500\n",
- "990/990 [==============================] - 0s 483us/step - loss: 0.0561 - val_loss: 0.1006\n",
- "Epoch 1244/2500\n",
- "990/990 [==============================] - 0s 482us/step - loss: 0.0561 - val_loss: 0.1006\n",
- "Epoch 1245/2500\n",
- "990/990 [==============================] - 0s 479us/step - loss: 0.0561 - val_loss: 0.1006\n",
- "Epoch 1246/2500\n",
- "990/990 [==============================] - 0s 488us/step - loss: 0.0561 - val_loss: 0.1006\n",
- "Epoch 1247/2500\n",
- "990/990 [==============================] - 0s 459us/step - loss: 0.0561 - val_loss: 0.1006\n",
- "Epoch 1248/2500\n",
- "990/990 [==============================] - 0s 445us/step - loss: 0.0561 - val_loss: 0.1006\n",
- "Epoch 1249/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0561 - val_loss: 0.1006\n",
- "Epoch 1250/2500\n",
- "990/990 [==============================] - 0s 470us/step - loss: 0.0561 - val_loss: 0.1006\n",
- "Epoch 1251/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0561 - val_loss: 0.1006\n",
- "Epoch 1252/2500\n",
- "990/990 [==============================] - 0s 447us/step - loss: 0.0561 - val_loss: 0.1006\n",
- "Epoch 1253/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0561 - val_loss: 0.1006\n",
- "Epoch 1254/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0561 - val_loss: 0.1006\n",
- "Epoch 1255/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0561 - val_loss: 0.1006\n",
- "Epoch 1256/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0561 - val_loss: 0.1006\n",
- "Epoch 1257/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0561 - val_loss: 0.1006\n",
- "Epoch 1258/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0561 - val_loss: 0.1006\n",
- "Epoch 1259/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0561 - val_loss: 0.1006\n",
- "Epoch 1260/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0561 - val_loss: 0.1006\n",
- "Epoch 1261/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0561 - val_loss: 0.1006\n",
- "Epoch 1262/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0561 - val_loss: 0.1006\n",
- "Epoch 1263/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0560 - val_loss: 0.1006\n",
- "Epoch 1264/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0560 - val_loss: 0.1006\n",
- "Epoch 1265/2500\n",
- "990/990 [==============================] - 0s 456us/step - loss: 0.0560 - val_loss: 0.1006\n",
- "Epoch 1266/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0560 - val_loss: 0.1006\n",
- "Epoch 1267/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0560 - val_loss: 0.1006\n",
- "Epoch 1268/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0560 - val_loss: 0.1006\n",
- "Epoch 1269/2500\n",
- "990/990 [==============================] - 0s 459us/step - loss: 0.0560 - val_loss: 0.1007\n",
- "Epoch 1270/2500\n",
- "990/990 [==============================] - 0s 456us/step - loss: 0.0560 - val_loss: 0.1006\n",
- "Epoch 1271/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0560 - val_loss: 0.1006\n",
- "Epoch 1272/2500\n",
- "990/990 [==============================] - 0s 466us/step - loss: 0.0560 - val_loss: 0.1006\n",
- "Epoch 1273/2500\n",
- "990/990 [==============================] - 0s 475us/step - loss: 0.0560 - val_loss: 0.1007\n",
- "Epoch 1274/2500\n",
- "990/990 [==============================] - 0s 480us/step - loss: 0.0560 - val_loss: 0.1006\n",
- "Epoch 1275/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0560 - val_loss: 0.1006\n",
- "Epoch 1276/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0560 - val_loss: 0.1006\n",
- "Epoch 1277/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0560 - val_loss: 0.1006\n",
- "Epoch 1278/2500\n",
- "990/990 [==============================] - 0s 462us/step - loss: 0.0560 - val_loss: 0.1006\n",
- "Epoch 1279/2500\n",
- "990/990 [==============================] - 0s 460us/step - loss: 0.0560 - val_loss: 0.1006\n",
- "Epoch 1280/2500\n",
- "990/990 [==============================] - 0s 473us/step - loss: 0.0560 - val_loss: 0.1006\n",
- "Epoch 1281/2500\n",
- "990/990 [==============================] - 0s 475us/step - loss: 0.0560 - val_loss: 0.1006\n",
- "Epoch 1282/2500\n",
- "990/990 [==============================] - 0s 475us/step - loss: 0.0560 - val_loss: 0.1007\n",
- "Epoch 1283/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0560 - val_loss: 0.1006\n",
- "Epoch 1284/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0560 - val_loss: 0.1006\n",
- "Epoch 1285/2500\n",
- "990/990 [==============================] - 0s 445us/step - loss: 0.0560 - val_loss: 0.1007\n",
- "Epoch 1286/2500\n",
- "990/990 [==============================] - 0s 461us/step - loss: 0.0560 - val_loss: 0.1007\n",
- "Epoch 1287/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0560 - val_loss: 0.1006\n",
- "Epoch 1288/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0560 - val_loss: 0.1007\n",
- "Epoch 1289/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0560 - val_loss: 0.1006\n",
- "Epoch 1290/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0560 - val_loss: 0.1006\n",
- "Epoch 1291/2500\n",
- "990/990 [==============================] - 0s 448us/step - loss: 0.0560 - val_loss: 0.1006\n",
- "Epoch 1292/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0560 - val_loss: 0.1007\n",
- "Epoch 1293/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0559 - val_loss: 0.1007\n",
- "Epoch 1294/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0559 - val_loss: 0.1006\n",
- "Epoch 1295/2500\n",
- "990/990 [==============================] - 0s 459us/step - loss: 0.0559 - val_loss: 0.1007\n",
- "Epoch 1296/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0559 - val_loss: 0.1007\n",
- "Epoch 1297/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0559 - val_loss: 0.1007\n",
- "Epoch 1298/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0559 - val_loss: 0.1007\n",
- "Epoch 1299/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0559 - val_loss: 0.1007\n",
- "Epoch 1300/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0559 - val_loss: 0.1007\n",
- "Epoch 1301/2500\n",
- "990/990 [==============================] - 0s 447us/step - loss: 0.0559 - val_loss: 0.1007\n",
- "Epoch 1302/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0559 - val_loss: 0.1007\n",
- "Epoch 1303/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0559 - val_loss: 0.1007\n",
- "Epoch 1304/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0559 - val_loss: 0.1007\n",
- "Epoch 1305/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0559 - val_loss: 0.1007\n",
- "Epoch 1306/2500\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "990/990 [==============================] - 0s 446us/step - loss: 0.0559 - val_loss: 0.1007\n",
- "Epoch 1307/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0559 - val_loss: 0.1007\n",
- "Epoch 1308/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0559 - val_loss: 0.1007\n",
- "Epoch 1309/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0559 - val_loss: 0.1007\n",
- "Epoch 1310/2500\n",
- "990/990 [==============================] - 0s 459us/step - loss: 0.0559 - val_loss: 0.1007\n",
- "Epoch 1311/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0559 - val_loss: 0.1007\n",
- "Epoch 1312/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0559 - val_loss: 0.1007\n",
- "Epoch 1313/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0559 - val_loss: 0.1007\n",
- "Epoch 1314/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0559 - val_loss: 0.1007\n",
- "Epoch 1315/2500\n",
- "990/990 [==============================] - 0s 445us/step - loss: 0.0559 - val_loss: 0.1007\n",
- "Epoch 1316/2500\n",
- "990/990 [==============================] - 0s 448us/step - loss: 0.0559 - val_loss: 0.1007\n",
- "Epoch 1317/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0559 - val_loss: 0.1007\n",
- "Epoch 1318/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0559 - val_loss: 0.1007\n",
- "Epoch 1319/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0559 - val_loss: 0.1007\n",
- "Epoch 1320/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0559 - val_loss: 0.1007\n",
- "Epoch 1321/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0559 - val_loss: 0.1007\n",
- "Epoch 1322/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0559 - val_loss: 0.1007\n",
- "Epoch 1323/2500\n",
- "990/990 [==============================] - 0s 445us/step - loss: 0.0559 - val_loss: 0.1007\n",
- "Epoch 1324/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0558 - val_loss: 0.1007\n",
- "Epoch 1325/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0558 - val_loss: 0.1007\n",
- "Epoch 1326/2500\n",
- "990/990 [==============================] - 0s 444us/step - loss: 0.0558 - val_loss: 0.1007\n",
- "Epoch 1327/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0558 - val_loss: 0.1007\n",
- "Epoch 1328/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0558 - val_loss: 0.1007\n",
- "Epoch 1329/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0558 - val_loss: 0.1007\n",
- "Epoch 1330/2500\n",
- "990/990 [==============================] - 0s 448us/step - loss: 0.0558 - val_loss: 0.1007\n",
- "Epoch 1331/2500\n",
- "990/990 [==============================] - 0s 448us/step - loss: 0.0558 - val_loss: 0.1007\n",
- "Epoch 1332/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0558 - val_loss: 0.1007\n",
- "Epoch 1333/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0558 - val_loss: 0.1007\n",
- "Epoch 1334/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0558 - val_loss: 0.1007\n",
- "Epoch 1335/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0558 - val_loss: 0.1007\n",
- "Epoch 1336/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0558 - val_loss: 0.1007\n",
- "Epoch 1337/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0558 - val_loss: 0.1007\n",
- "Epoch 1338/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0558 - val_loss: 0.1007\n",
- "Epoch 1339/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0558 - val_loss: 0.1007\n",
- "Epoch 1340/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0558 - val_loss: 0.1007\n",
- "Epoch 1341/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0558 - val_loss: 0.1007\n",
- "Epoch 1342/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0558 - val_loss: 0.1007\n",
- "Epoch 1343/2500\n",
- "990/990 [==============================] - 0s 459us/step - loss: 0.0558 - val_loss: 0.1007\n",
- "Epoch 1344/2500\n",
- "990/990 [==============================] - 0s 447us/step - loss: 0.0558 - val_loss: 0.1007\n",
- "Epoch 1345/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0558 - val_loss: 0.1007\n",
- "Epoch 1346/2500\n",
- "990/990 [==============================] - 0s 463us/step - loss: 0.0558 - val_loss: 0.1007\n",
- "Epoch 1347/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0558 - val_loss: 0.1007\n",
- "Epoch 1348/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0558 - val_loss: 0.1007\n",
- "Epoch 1349/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0558 - val_loss: 0.1007\n",
- "Epoch 1350/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0558 - val_loss: 0.1007\n",
- "Epoch 1351/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0558 - val_loss: 0.1007\n",
- "Epoch 1352/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0558 - val_loss: 0.1008\n",
- "Epoch 1353/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0558 - val_loss: 0.1007\n",
- "Epoch 1354/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0558 - val_loss: 0.1007\n",
- "Epoch 1355/2500\n",
- "990/990 [==============================] - 0s 459us/step - loss: 0.0558 - val_loss: 0.1007\n",
- "Epoch 1356/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0558 - val_loss: 0.1007\n",
- "Epoch 1357/2500\n",
- "990/990 [==============================] - 0s 459us/step - loss: 0.0557 - val_loss: 0.1007\n",
- "Epoch 1358/2500\n",
- "990/990 [==============================] - 0s 459us/step - loss: 0.0557 - val_loss: 0.1007\n",
- "Epoch 1359/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0557 - val_loss: 0.1008\n",
- "Epoch 1360/2500\n",
- "990/990 [==============================] - 0s 446us/step - loss: 0.0557 - val_loss: 0.1007\n",
- "Epoch 1361/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0557 - val_loss: 0.1007\n",
- "Epoch 1362/2500\n",
- "990/990 [==============================] - 0s 462us/step - loss: 0.0557 - val_loss: 0.1007\n",
- "Epoch 1363/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0557 - val_loss: 0.1008\n",
- "Epoch 1364/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0557 - val_loss: 0.1007\n",
- "Epoch 1365/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0557 - val_loss: 0.1007\n",
- "Epoch 1366/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0557 - val_loss: 0.1007\n",
- "Epoch 1367/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0557 - val_loss: 0.1007\n",
- "Epoch 1368/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0557 - val_loss: 0.1007\n",
- "Epoch 1369/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0557 - val_loss: 0.1007\n",
- "Epoch 1370/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0557 - val_loss: 0.1007\n",
- "Epoch 1371/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0557 - val_loss: 0.1008\n",
- "Epoch 1372/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0557 - val_loss: 0.1008\n",
- "Epoch 1373/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0557 - val_loss: 0.1008\n",
- "Epoch 1374/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0557 - val_loss: 0.1007\n",
- "Epoch 1375/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0557 - val_loss: 0.1008\n",
- "Epoch 1376/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0557 - val_loss: 0.1007\n",
- "Epoch 1377/2500\n",
- "990/990 [==============================] - 0s 460us/step - loss: 0.0557 - val_loss: 0.1008\n",
- "Epoch 1378/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0557 - val_loss: 0.1008\n",
- "Epoch 1379/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0557 - val_loss: 0.1008\n",
- "Epoch 1380/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0557 - val_loss: 0.1008\n",
- "Epoch 1381/2500\n",
- "990/990 [==============================] - 0s 460us/step - loss: 0.0557 - val_loss: 0.1008\n",
- "Epoch 1382/2500\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "990/990 [==============================] - 0s 442us/step - loss: 0.0557 - val_loss: 0.1008\n",
- "Epoch 1383/2500\n",
- "990/990 [==============================] - 0s 448us/step - loss: 0.0557 - val_loss: 0.1008\n",
- "Epoch 1384/2500\n",
- "990/990 [==============================] - 0s 447us/step - loss: 0.0557 - val_loss: 0.1008\n",
- "Epoch 1385/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0557 - val_loss: 0.1008\n",
- "Epoch 1386/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0557 - val_loss: 0.1007\n",
- "Epoch 1387/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0557 - val_loss: 0.1008\n",
- "Epoch 1388/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0557 - val_loss: 0.1008\n",
- "Epoch 1389/2500\n",
- "990/990 [==============================] - 0s 460us/step - loss: 0.0557 - val_loss: 0.1008\n",
- "Epoch 1390/2500\n",
- "990/990 [==============================] - 0s 466us/step - loss: 0.0557 - val_loss: 0.1008\n",
- "Epoch 1391/2500\n",
- "990/990 [==============================] - 0s 471us/step - loss: 0.0556 - val_loss: 0.1008\n",
- "Epoch 1392/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0556 - val_loss: 0.1008\n",
- "Epoch 1393/2500\n",
- "990/990 [==============================] - 0s 472us/step - loss: 0.0556 - val_loss: 0.1008\n",
- "Epoch 1394/2500\n",
- "990/990 [==============================] - 0s 471us/step - loss: 0.0556 - val_loss: 0.1008\n",
- "Epoch 1395/2500\n",
- "990/990 [==============================] - 0s 471us/step - loss: 0.0556 - val_loss: 0.1008\n",
- "Epoch 1396/2500\n",
- "990/990 [==============================] - 0s 485us/step - loss: 0.0556 - val_loss: 0.1008\n",
- "Epoch 1397/2500\n",
- "990/990 [==============================] - 0s 493us/step - loss: 0.0556 - val_loss: 0.1008\n",
- "Epoch 1398/2500\n",
- "990/990 [==============================] - 0s 488us/step - loss: 0.0556 - val_loss: 0.1008\n",
- "Epoch 1399/2500\n",
- "990/990 [==============================] - 0s 465us/step - loss: 0.0556 - val_loss: 0.1008\n",
- "Epoch 1400/2500\n",
- "990/990 [==============================] - 0s 484us/step - loss: 0.0556 - val_loss: 0.1008\n",
- "Epoch 1401/2500\n",
- "990/990 [==============================] - 0s 486us/step - loss: 0.0556 - val_loss: 0.1008\n",
- "Epoch 1402/2500\n",
- "990/990 [==============================] - 0s 458us/step - loss: 0.0556 - val_loss: 0.1008\n",
- "Epoch 1403/2500\n",
- "990/990 [==============================] - 0s 478us/step - loss: 0.0556 - val_loss: 0.1008\n",
- "Epoch 1404/2500\n",
- "990/990 [==============================] - 0s 488us/step - loss: 0.0556 - val_loss: 0.1008\n",
- "Epoch 1405/2500\n",
- "990/990 [==============================] - 0s 462us/step - loss: 0.0556 - val_loss: 0.1008\n",
- "Epoch 1406/2500\n",
- "990/990 [==============================] - 0s 471us/step - loss: 0.0556 - val_loss: 0.1008\n",
- "Epoch 1407/2500\n",
- "990/990 [==============================] - 0s 477us/step - loss: 0.0556 - val_loss: 0.1008\n",
- "Epoch 1408/2500\n",
- "990/990 [==============================] - 0s 472us/step - loss: 0.0556 - val_loss: 0.1008\n",
- "Epoch 1409/2500\n",
- "990/990 [==============================] - 0s 468us/step - loss: 0.0556 - val_loss: 0.1008\n",
- "Epoch 1410/2500\n",
- "990/990 [==============================] - 0s 480us/step - loss: 0.0556 - val_loss: 0.1008\n",
- "Epoch 1411/2500\n",
- "990/990 [==============================] - 0s 476us/step - loss: 0.0556 - val_loss: 0.1008\n",
- "Epoch 1412/2500\n",
- "990/990 [==============================] - 0s 494us/step - loss: 0.0556 - val_loss: 0.1008\n",
- "Epoch 1413/2500\n",
- "990/990 [==============================] - 0s 481us/step - loss: 0.0556 - val_loss: 0.1008\n",
- "Epoch 1414/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0556 - val_loss: 0.1008\n",
- "Epoch 1415/2500\n",
- "990/990 [==============================] - 0s 466us/step - loss: 0.0556 - val_loss: 0.1008\n",
- "Epoch 1416/2500\n",
- "990/990 [==============================] - 0s 478us/step - loss: 0.0556 - val_loss: 0.1008\n",
- "Epoch 1417/2500\n",
- "990/990 [==============================] - 0s 477us/step - loss: 0.0556 - val_loss: 0.1008\n",
- "Epoch 1418/2500\n",
- "990/990 [==============================] - 0s 471us/step - loss: 0.0556 - val_loss: 0.1008\n",
- "Epoch 1419/2500\n",
- "990/990 [==============================] - 0s 474us/step - loss: 0.0556 - val_loss: 0.1008\n",
- "Epoch 1420/2500\n",
- "990/990 [==============================] - 0s 470us/step - loss: 0.0556 - val_loss: 0.1008\n",
- "Epoch 1421/2500\n",
- "990/990 [==============================] - 0s 472us/step - loss: 0.0556 - val_loss: 0.1008\n",
- "Epoch 1422/2500\n",
- "990/990 [==============================] - 0s 478us/step - loss: 0.0556 - val_loss: 0.1008\n",
- "Epoch 1423/2500\n",
- "990/990 [==============================] - 0s 463us/step - loss: 0.0556 - val_loss: 0.1009\n",
- "Epoch 1424/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0556 - val_loss: 0.1009\n",
- "Epoch 1425/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0556 - val_loss: 0.1008\n",
- "Epoch 1426/2500\n",
- "990/990 [==============================] - 0s 456us/step - loss: 0.0556 - val_loss: 0.1009\n",
- "Epoch 1427/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0555 - val_loss: 0.1008\n",
- "Epoch 1428/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0555 - val_loss: 0.1008\n",
- "Epoch 1429/2500\n",
- "990/990 [==============================] - 0s 465us/step - loss: 0.0555 - val_loss: 0.1008\n",
- "Epoch 1430/2500\n",
- "990/990 [==============================] - 0s 448us/step - loss: 0.0555 - val_loss: 0.1008\n",
- "Epoch 1431/2500\n",
- "990/990 [==============================] - 0s 489us/step - loss: 0.0555 - val_loss: 0.1008\n",
- "Epoch 1432/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0555 - val_loss: 0.1009\n",
- "Epoch 1433/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0555 - val_loss: 0.1008\n",
- "Epoch 1434/2500\n",
- "990/990 [==============================] - 0s 465us/step - loss: 0.0555 - val_loss: 0.1009\n",
- "Epoch 1435/2500\n",
- "990/990 [==============================] - 0s 473us/step - loss: 0.0555 - val_loss: 0.1008\n",
- "Epoch 1436/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0555 - val_loss: 0.1008\n",
- "Epoch 1437/2500\n",
- "990/990 [==============================] - 0s 460us/step - loss: 0.0555 - val_loss: 0.1009\n",
- "Epoch 1438/2500\n",
- "990/990 [==============================] - 0s 462us/step - loss: 0.0555 - val_loss: 0.1009\n",
- "Epoch 1439/2500\n",
- "990/990 [==============================] - 0s 461us/step - loss: 0.0555 - val_loss: 0.1009\n",
- "Epoch 1440/2500\n",
- "990/990 [==============================] - 0s 471us/step - loss: 0.0555 - val_loss: 0.1009\n",
- "Epoch 1441/2500\n",
- "990/990 [==============================] - 0s 463us/step - loss: 0.0555 - val_loss: 0.1009\n",
- "Epoch 1442/2500\n",
- "990/990 [==============================] - 0s 463us/step - loss: 0.0555 - val_loss: 0.1009\n",
- "Epoch 1443/2500\n",
- "990/990 [==============================] - 0s 472us/step - loss: 0.0555 - val_loss: 0.1009\n",
- "Epoch 1444/2500\n",
- "990/990 [==============================] - 0s 471us/step - loss: 0.0555 - val_loss: 0.1009\n",
- "Epoch 1445/2500\n",
- "990/990 [==============================] - 0s 462us/step - loss: 0.0555 - val_loss: 0.1009\n",
- "Epoch 1446/2500\n",
- "990/990 [==============================] - 0s 463us/step - loss: 0.0555 - val_loss: 0.1009\n",
- "Epoch 1447/2500\n",
- "990/990 [==============================] - 0s 463us/step - loss: 0.0555 - val_loss: 0.1009\n",
- "Epoch 1448/2500\n",
- "990/990 [==============================] - 0s 462us/step - loss: 0.0555 - val_loss: 0.1009\n",
- "Epoch 1449/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0555 - val_loss: 0.1009\n",
- "Epoch 1450/2500\n",
- "990/990 [==============================] - 0s 464us/step - loss: 0.0555 - val_loss: 0.1009\n",
- "Epoch 1451/2500\n",
- "990/990 [==============================] - 0s 468us/step - loss: 0.0555 - val_loss: 0.1009\n",
- "Epoch 1452/2500\n",
- "990/990 [==============================] - 0s 460us/step - loss: 0.0555 - val_loss: 0.1009\n",
- "Epoch 1453/2500\n",
- "990/990 [==============================] - 0s 464us/step - loss: 0.0555 - val_loss: 0.1009\n",
- "Epoch 1454/2500\n",
- "990/990 [==============================] - 0s 469us/step - loss: 0.0555 - val_loss: 0.1009\n",
- "Epoch 1455/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0555 - val_loss: 0.1009\n",
- "Epoch 1456/2500\n",
- "990/990 [==============================] - 0s 464us/step - loss: 0.0555 - val_loss: 0.1009\n",
- "Epoch 1457/2500\n",
- "990/990 [==============================] - 0s 456us/step - loss: 0.0555 - val_loss: 0.1009\n",
- "Epoch 1458/2500\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "990/990 [==============================] - 0s 451us/step - loss: 0.0555 - val_loss: 0.1009\n",
- "Epoch 1459/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0555 - val_loss: 0.1009\n",
- "Epoch 1460/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0555 - val_loss: 0.1009\n",
- "Epoch 1461/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0555 - val_loss: 0.1009\n",
- "Epoch 1462/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0555 - val_loss: 0.1009\n",
- "Epoch 1463/2500\n",
- "990/990 [==============================] - 0s 464us/step - loss: 0.0555 - val_loss: 0.1009\n",
- "Epoch 1464/2500\n",
- "990/990 [==============================] - 0s 447us/step - loss: 0.0555 - val_loss: 0.1009\n",
- "Epoch 1465/2500\n",
- "990/990 [==============================] - 0s 444us/step - loss: 0.0554 - val_loss: 0.1009\n",
- "Epoch 1466/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0554 - val_loss: 0.1009\n",
- "Epoch 1467/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0554 - val_loss: 0.1009\n",
- "Epoch 1468/2500\n",
- "990/990 [==============================] - 0s 459us/step - loss: 0.0554 - val_loss: 0.1009\n",
- "Epoch 1469/2500\n",
- "990/990 [==============================] - 0s 461us/step - loss: 0.0554 - val_loss: 0.1009\n",
- "Epoch 1470/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0554 - val_loss: 0.1009\n",
- "Epoch 1471/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0554 - val_loss: 0.1009\n",
- "Epoch 1472/2500\n",
- "990/990 [==============================] - 0s 443us/step - loss: 0.0554 - val_loss: 0.1009\n",
- "Epoch 1473/2500\n",
- "990/990 [==============================] - 0s 447us/step - loss: 0.0554 - val_loss: 0.1009\n",
- "Epoch 1474/2500\n",
- "990/990 [==============================] - 0s 444us/step - loss: 0.0554 - val_loss: 0.1009\n",
- "Epoch 1475/2500\n",
- "990/990 [==============================] - 0s 469us/step - loss: 0.0554 - val_loss: 0.1009\n",
- "Epoch 1476/2500\n",
- "990/990 [==============================] - 0s 470us/step - loss: 0.0554 - val_loss: 0.1009\n",
- "Epoch 1477/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0554 - val_loss: 0.1009\n",
- "Epoch 1478/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0554 - val_loss: 0.1009\n",
- "Epoch 1479/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0554 - val_loss: 0.1009\n",
- "Epoch 1480/2500\n",
- "990/990 [==============================] - 0s 464us/step - loss: 0.0554 - val_loss: 0.1009\n",
- "Epoch 1481/2500\n",
- "990/990 [==============================] - 0s 443us/step - loss: 0.0554 - val_loss: 0.1009\n",
- "Epoch 1482/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0554 - val_loss: 0.1009\n",
- "Epoch 1483/2500\n",
- "990/990 [==============================] - 0s 465us/step - loss: 0.0554 - val_loss: 0.1009\n",
- "Epoch 1484/2500\n",
- "990/990 [==============================] - 0s 465us/step - loss: 0.0554 - val_loss: 0.1009\n",
- "Epoch 1485/2500\n",
- "990/990 [==============================] - 0s 473us/step - loss: 0.0554 - val_loss: 0.1009\n",
- "Epoch 1486/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0554 - val_loss: 0.1009\n",
- "Epoch 1487/2500\n",
- "990/990 [==============================] - 0s 462us/step - loss: 0.0554 - val_loss: 0.1009\n",
- "Epoch 1488/2500\n",
- "990/990 [==============================] - 0s 460us/step - loss: 0.0554 - val_loss: 0.1009\n",
- "Epoch 1489/2500\n",
- "990/990 [==============================] - 0s 448us/step - loss: 0.0554 - val_loss: 0.1009\n",
- "Epoch 1490/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0554 - val_loss: 0.1009\n",
- "Epoch 1491/2500\n",
- "990/990 [==============================] - 0s 473us/step - loss: 0.0554 - val_loss: 0.1009\n",
- "Epoch 1492/2500\n",
- "990/990 [==============================] - 0s 447us/step - loss: 0.0554 - val_loss: 0.1010\n",
- "Epoch 1493/2500\n",
- "990/990 [==============================] - 0s 446us/step - loss: 0.0554 - val_loss: 0.1009\n",
- "Epoch 1494/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0554 - val_loss: 0.1010\n",
- "Epoch 1495/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0554 - val_loss: 0.1010\n",
- "Epoch 1496/2500\n",
- "990/990 [==============================] - 0s 447us/step - loss: 0.0554 - val_loss: 0.1010\n",
- "Epoch 1497/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0554 - val_loss: 0.1010\n",
- "Epoch 1498/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0554 - val_loss: 0.1009\n",
- "Epoch 1499/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0554 - val_loss: 0.1010\n",
- "Epoch 1500/2500\n",
- "990/990 [==============================] - 0s 472us/step - loss: 0.0554 - val_loss: 0.1010\n",
- "Epoch 1501/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0554 - val_loss: 0.1010\n",
- "Epoch 1502/2500\n",
- "990/990 [==============================] - 0s 466us/step - loss: 0.0554 - val_loss: 0.1010\n",
- "Epoch 1503/2500\n",
- "990/990 [==============================] - 0s 462us/step - loss: 0.0554 - val_loss: 0.1010\n",
- "Epoch 1504/2500\n",
- "990/990 [==============================] - 0s 489us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1505/2500\n",
- "990/990 [==============================] - 0s 489us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1506/2500\n",
- "990/990 [==============================] - 0s 487us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1507/2500\n",
- "990/990 [==============================] - 0s 472us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1508/2500\n",
- "990/990 [==============================] - 0s 479us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1509/2500\n",
- "990/990 [==============================] - 0s 466us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1510/2500\n",
- "990/990 [==============================] - 0s 480us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1511/2500\n",
- "990/990 [==============================] - 0s 477us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1512/2500\n",
- "990/990 [==============================] - 0s 473us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1513/2500\n",
- "990/990 [==============================] - 0s 462us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1514/2500\n",
- "990/990 [==============================] - 0s 473us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1515/2500\n",
- "990/990 [==============================] - 0s 478us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1516/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1517/2500\n",
- "990/990 [==============================] - 0s 477us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1518/2500\n",
- "990/990 [==============================] - 0s 481us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1519/2500\n",
- "990/990 [==============================] - 0s 465us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1520/2500\n",
- "990/990 [==============================] - 0s 458us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1521/2500\n",
- "990/990 [==============================] - 0s 475us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1522/2500\n",
- "990/990 [==============================] - 0s 461us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1523/2500\n",
- "990/990 [==============================] - 0s 443us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1524/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1525/2500\n",
- "990/990 [==============================] - 0s 465us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1526/2500\n",
- "990/990 [==============================] - 0s 460us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1527/2500\n",
- "990/990 [==============================] - 0s 443us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1528/2500\n",
- "990/990 [==============================] - 0s 463us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1529/2500\n",
- "990/990 [==============================] - 0s 443us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1530/2500\n",
- "990/990 [==============================] - 0s 437us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1531/2500\n",
- "990/990 [==============================] - 0s 432us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1532/2500\n",
- "990/990 [==============================] - 0s 432us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1533/2500\n",
- "990/990 [==============================] - 0s 430us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1534/2500\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "990/990 [==============================] - 0s 433us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1535/2500\n",
- "990/990 [==============================] - 0s 422us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1536/2500\n",
- "990/990 [==============================] - 0s 434us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1537/2500\n",
- "990/990 [==============================] - 0s 439us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1538/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1539/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1540/2500\n",
- "990/990 [==============================] - 0s 470us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1541/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1542/2500\n",
- "990/990 [==============================] - 0s 445us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1543/2500\n",
- "990/990 [==============================] - 0s 464us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1544/2500\n",
- "990/990 [==============================] - 0s 436us/step - loss: 0.0553 - val_loss: 0.1010\n",
- "Epoch 1545/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0553 - val_loss: 0.1011\n",
- "Epoch 1546/2500\n",
- "990/990 [==============================] - 0s 463us/step - loss: 0.0553 - val_loss: 0.1011\n",
- "Epoch 1547/2500\n",
- "990/990 [==============================] - 0s 463us/step - loss: 0.0552 - val_loss: 0.1010\n",
- "Epoch 1548/2500\n",
- "990/990 [==============================] - 0s 448us/step - loss: 0.0552 - val_loss: 0.1010\n",
- "Epoch 1549/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0552 - val_loss: 0.1011\n",
- "Epoch 1550/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0552 - val_loss: 0.1010\n",
- "Epoch 1551/2500\n",
- "990/990 [==============================] - 0s 466us/step - loss: 0.0552 - val_loss: 0.1010\n",
- "Epoch 1552/2500\n",
- "990/990 [==============================] - 0s 479us/step - loss: 0.0552 - val_loss: 0.1011\n",
- "Epoch 1553/2500\n",
- "990/990 [==============================] - 0s 458us/step - loss: 0.0552 - val_loss: 0.1010\n",
- "Epoch 1554/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0552 - val_loss: 0.1011\n",
- "Epoch 1555/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0552 - val_loss: 0.1011\n",
- "Epoch 1556/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0552 - val_loss: 0.1011\n",
- "Epoch 1557/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0552 - val_loss: 0.1011\n",
- "Epoch 1558/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0552 - val_loss: 0.1010\n",
- "Epoch 1559/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0552 - val_loss: 0.1011\n",
- "Epoch 1560/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0552 - val_loss: 0.1011\n",
- "Epoch 1561/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0552 - val_loss: 0.1011\n",
- "Epoch 1562/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0552 - val_loss: 0.1011\n",
- "Epoch 1563/2500\n",
- "990/990 [==============================] - 0s 448us/step - loss: 0.0552 - val_loss: 0.1011\n",
- "Epoch 1564/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0552 - val_loss: 0.1011\n",
- "Epoch 1565/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0552 - val_loss: 0.1011\n",
- "Epoch 1566/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0552 - val_loss: 0.1011\n",
- "Epoch 1567/2500\n",
- "990/990 [==============================] - 0s 448us/step - loss: 0.0552 - val_loss: 0.1011\n",
- "Epoch 1568/2500\n",
- "990/990 [==============================] - 0s 446us/step - loss: 0.0552 - val_loss: 0.1011\n",
- "Epoch 1569/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0552 - val_loss: 0.1011\n",
- "Epoch 1570/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0552 - val_loss: 0.1011\n",
- "Epoch 1571/2500\n",
- "990/990 [==============================] - 0s 447us/step - loss: 0.0552 - val_loss: 0.1011\n",
- "Epoch 1572/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0552 - val_loss: 0.1011\n",
- "Epoch 1573/2500\n",
- "990/990 [==============================] - 0s 445us/step - loss: 0.0552 - val_loss: 0.1011\n",
- "Epoch 1574/2500\n",
- "990/990 [==============================] - 0s 458us/step - loss: 0.0552 - val_loss: 0.1011\n",
- "Epoch 1575/2500\n",
- "990/990 [==============================] - 0s 466us/step - loss: 0.0552 - val_loss: 0.1011\n",
- "Epoch 1576/2500\n",
- "990/990 [==============================] - 0s 465us/step - loss: 0.0552 - val_loss: 0.1011\n",
- "Epoch 1577/2500\n",
- "990/990 [==============================] - 0s 433us/step - loss: 0.0552 - val_loss: 0.1011\n",
- "Epoch 1578/2500\n",
- "990/990 [==============================] - 0s 442us/step - loss: 0.0552 - val_loss: 0.1011\n",
- "Epoch 1579/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0552 - val_loss: 0.1011\n",
- "Epoch 1580/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0552 - val_loss: 0.1011\n",
- "Epoch 1581/2500\n",
- "990/990 [==============================] - 0s 465us/step - loss: 0.0552 - val_loss: 0.1011\n",
- "Epoch 1582/2500\n",
- "990/990 [==============================] - 0s 459us/step - loss: 0.0552 - val_loss: 0.1011\n",
- "Epoch 1583/2500\n",
- "990/990 [==============================] - 0s 468us/step - loss: 0.0552 - val_loss: 0.1011\n",
- "Epoch 1584/2500\n",
- "990/990 [==============================] - 0s 475us/step - loss: 0.0552 - val_loss: 0.1011\n",
- "Epoch 1585/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0552 - val_loss: 0.1011\n",
- "Epoch 1586/2500\n",
- "990/990 [==============================] - 0s 446us/step - loss: 0.0552 - val_loss: 0.1011\n",
- "Epoch 1587/2500\n",
- "990/990 [==============================] - 0s 481us/step - loss: 0.0552 - val_loss: 0.1011\n",
- "Epoch 1588/2500\n",
- "990/990 [==============================] - 0s 459us/step - loss: 0.0552 - val_loss: 0.1011\n",
- "Epoch 1589/2500\n",
- "990/990 [==============================] - 0s 459us/step - loss: 0.0552 - val_loss: 0.1011\n",
- "Epoch 1590/2500\n",
- "990/990 [==============================] - 0s 446us/step - loss: 0.0552 - val_loss: 0.1011\n",
- "Epoch 1591/2500\n",
- "990/990 [==============================] - 0s 439us/step - loss: 0.0551 - val_loss: 0.1011\n",
- "Epoch 1592/2500\n",
- "990/990 [==============================] - 0s 444us/step - loss: 0.0551 - val_loss: 0.1011\n",
- "Epoch 1593/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0551 - val_loss: 0.1012\n",
- "Epoch 1594/2500\n",
- "990/990 [==============================] - 0s 470us/step - loss: 0.0551 - val_loss: 0.1011\n",
- "Epoch 1595/2500\n",
- "990/990 [==============================] - 0s 478us/step - loss: 0.0551 - val_loss: 0.1012\n",
- "Epoch 1596/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0551 - val_loss: 0.1012\n",
- "Epoch 1597/2500\n",
- "990/990 [==============================] - 0s 446us/step - loss: 0.0551 - val_loss: 0.1012\n",
- "Epoch 1598/2500\n",
- "990/990 [==============================] - 0s 442us/step - loss: 0.0551 - val_loss: 0.1011\n",
- "Epoch 1599/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0551 - val_loss: 0.1011\n",
- "Epoch 1600/2500\n",
- "990/990 [==============================] - 0s 440us/step - loss: 0.0551 - val_loss: 0.1011\n",
- "Epoch 1601/2500\n",
- "990/990 [==============================] - 0s 440us/step - loss: 0.0551 - val_loss: 0.1011\n",
- "Epoch 1602/2500\n",
- "990/990 [==============================] - 0s 466us/step - loss: 0.0551 - val_loss: 0.1012\n",
- "Epoch 1603/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0551 - val_loss: 0.1012\n",
- "Epoch 1604/2500\n",
- "990/990 [==============================] - 0s 473us/step - loss: 0.0551 - val_loss: 0.1012\n",
- "Epoch 1605/2500\n",
- "990/990 [==============================] - 0s 434us/step - loss: 0.0551 - val_loss: 0.1012\n",
- "Epoch 1606/2500\n",
- "990/990 [==============================] - 0s 435us/step - loss: 0.0551 - val_loss: 0.1012\n",
- "Epoch 1607/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0551 - val_loss: 0.1012\n",
- "Epoch 1608/2500\n",
- "990/990 [==============================] - 0s 432us/step - loss: 0.0551 - val_loss: 0.1012\n",
- "Epoch 1609/2500\n",
- "990/990 [==============================] - 0s 437us/step - loss: 0.0551 - val_loss: 0.1012\n",
- "Epoch 1610/2500\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "990/990 [==============================] - 0s 422us/step - loss: 0.0551 - val_loss: 0.1012\n",
- "Epoch 1611/2500\n",
- "990/990 [==============================] - 0s 431us/step - loss: 0.0551 - val_loss: 0.1012\n",
- "Epoch 1612/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0551 - val_loss: 0.1012\n",
- "Epoch 1613/2500\n",
- "990/990 [==============================] - 0s 418us/step - loss: 0.0551 - val_loss: 0.1012\n",
- "Epoch 1614/2500\n",
- "990/990 [==============================] - 0s 440us/step - loss: 0.0551 - val_loss: 0.1012\n",
- "Epoch 1615/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0551 - val_loss: 0.1012\n",
- "Epoch 1616/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0551 - val_loss: 0.1012\n",
- "Epoch 1617/2500\n",
- "990/990 [==============================] - 0s 440us/step - loss: 0.0551 - val_loss: 0.1012\n",
- "Epoch 1618/2500\n",
- "990/990 [==============================] - 0s 491us/step - loss: 0.0551 - val_loss: 0.1012\n",
- "Epoch 1619/2500\n",
- "990/990 [==============================] - 0s 469us/step - loss: 0.0551 - val_loss: 0.1012\n",
- "Epoch 1620/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0551 - val_loss: 0.1012\n",
- "Epoch 1621/2500\n",
- "990/990 [==============================] - 0s 460us/step - loss: 0.0551 - val_loss: 0.1012\n",
- "Epoch 1622/2500\n",
- "990/990 [==============================] - 0s 486us/step - loss: 0.0551 - val_loss: 0.1012\n",
- "Epoch 1623/2500\n",
- "990/990 [==============================] - 0s 467us/step - loss: 0.0551 - val_loss: 0.1012\n",
- "Epoch 1624/2500\n",
- "990/990 [==============================] - 0s 468us/step - loss: 0.0551 - val_loss: 0.1012\n",
- "Epoch 1625/2500\n",
- "990/990 [==============================] - 0s 462us/step - loss: 0.0551 - val_loss: 0.1012\n",
- "Epoch 1626/2500\n",
- "990/990 [==============================] - 0s 472us/step - loss: 0.0551 - val_loss: 0.1012\n",
- "Epoch 1627/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0551 - val_loss: 0.1012\n",
- "Epoch 1628/2500\n",
- "990/990 [==============================] - 0s 459us/step - loss: 0.0551 - val_loss: 0.1012\n",
- "Epoch 1629/2500\n",
- "990/990 [==============================] - 0s 467us/step - loss: 0.0551 - val_loss: 0.1012\n",
- "Epoch 1630/2500\n",
- "990/990 [==============================] - 0s 465us/step - loss: 0.0551 - val_loss: 0.1012\n",
- "Epoch 1631/2500\n",
- "990/990 [==============================] - 0s 454us/step - loss: 0.0551 - val_loss: 0.1012\n",
- "Epoch 1632/2500\n",
- "990/990 [==============================] - 0s 444us/step - loss: 0.0551 - val_loss: 0.1012\n",
- "Epoch 1633/2500\n",
- "990/990 [==============================] - 0s 436us/step - loss: 0.0551 - val_loss: 0.1012\n",
- "Epoch 1634/2500\n",
- "990/990 [==============================] - 0s 442us/step - loss: 0.0551 - val_loss: 0.1012\n",
- "Epoch 1635/2500\n",
- "990/990 [==============================] - 0s 434us/step - loss: 0.0551 - val_loss: 0.1012\n",
- "Epoch 1636/2500\n",
- "990/990 [==============================] - 0s 443us/step - loss: 0.0551 - val_loss: 0.1012\n",
- "Epoch 1637/2500\n",
- "990/990 [==============================] - 0s 478us/step - loss: 0.0551 - val_loss: 0.1012\n",
- "Epoch 1638/2500\n",
- "990/990 [==============================] - 0s 463us/step - loss: 0.0550 - val_loss: 0.1012\n",
- "Epoch 1639/2500\n",
- "990/990 [==============================] - 0s 486us/step - loss: 0.0550 - val_loss: 0.1012\n",
- "Epoch 1640/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0550 - val_loss: 0.1012\n",
- "Epoch 1641/2500\n",
- "990/990 [==============================] - 0s 435us/step - loss: 0.0550 - val_loss: 0.1012\n",
- "Epoch 1642/2500\n",
- "990/990 [==============================] - 0s 436us/step - loss: 0.0550 - val_loss: 0.1012\n",
- "Epoch 1643/2500\n",
- "990/990 [==============================] - 0s 434us/step - loss: 0.0550 - val_loss: 0.1013\n",
- "Epoch 1644/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0550 - val_loss: 0.1012\n",
- "Epoch 1645/2500\n",
- "990/990 [==============================] - 0s 437us/step - loss: 0.0550 - val_loss: 0.1012\n",
- "Epoch 1646/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0550 - val_loss: 0.1012\n",
- "Epoch 1647/2500\n",
- "990/990 [==============================] - 0s 422us/step - loss: 0.0550 - val_loss: 0.1013\n",
- "Epoch 1648/2500\n",
- "990/990 [==============================] - 0s 425us/step - loss: 0.0550 - val_loss: 0.1013\n",
- "Epoch 1649/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0550 - val_loss: 0.1013\n",
- "Epoch 1650/2500\n",
- "990/990 [==============================] - 0s 422us/step - loss: 0.0550 - val_loss: 0.1012\n",
- "Epoch 1651/2500\n",
- "990/990 [==============================] - 0s 431us/step - loss: 0.0550 - val_loss: 0.1012\n",
- "Epoch 1652/2500\n",
- "990/990 [==============================] - 0s 431us/step - loss: 0.0550 - val_loss: 0.1012\n",
- "Epoch 1653/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0550 - val_loss: 0.1013\n",
- "Epoch 1654/2500\n",
- "990/990 [==============================] - 0s 426us/step - loss: 0.0550 - val_loss: 0.1013\n",
- "Epoch 1655/2500\n",
- "990/990 [==============================] - 0s 424us/step - loss: 0.0550 - val_loss: 0.1013\n",
- "Epoch 1656/2500\n",
- "990/990 [==============================] - 0s 435us/step - loss: 0.0550 - val_loss: 0.1012\n",
- "Epoch 1657/2500\n",
- "990/990 [==============================] - 0s 432us/step - loss: 0.0550 - val_loss: 0.1013\n",
- "Epoch 1658/2500\n",
- "990/990 [==============================] - 0s 430us/step - loss: 0.0550 - val_loss: 0.1013\n",
- "Epoch 1659/2500\n",
- "990/990 [==============================] - 0s 434us/step - loss: 0.0550 - val_loss: 0.1013\n",
- "Epoch 1660/2500\n",
- "990/990 [==============================] - 0s 448us/step - loss: 0.0550 - val_loss: 0.1013\n",
- "Epoch 1661/2500\n",
- "990/990 [==============================] - 0s 448us/step - loss: 0.0550 - val_loss: 0.1013\n",
- "Epoch 1662/2500\n",
- "990/990 [==============================] - 0s 426us/step - loss: 0.0550 - val_loss: 0.1013\n",
- "Epoch 1663/2500\n",
- "990/990 [==============================] - 0s 431us/step - loss: 0.0550 - val_loss: 0.1013\n",
- "Epoch 1664/2500\n",
- "990/990 [==============================] - 0s 432us/step - loss: 0.0550 - val_loss: 0.1013\n",
- "Epoch 1665/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0550 - val_loss: 0.1013\n",
- "Epoch 1666/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0550 - val_loss: 0.1013\n",
- "Epoch 1667/2500\n",
- "990/990 [==============================] - 0s 426us/step - loss: 0.0550 - val_loss: 0.1013\n",
- "Epoch 1668/2500\n",
- "990/990 [==============================] - 0s 419us/step - loss: 0.0550 - val_loss: 0.1013\n",
- "Epoch 1669/2500\n",
- "990/990 [==============================] - 0s 423us/step - loss: 0.0550 - val_loss: 0.1013\n",
- "Epoch 1670/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0550 - val_loss: 0.1013\n",
- "Epoch 1671/2500\n",
- "990/990 [==============================] - 0s 422us/step - loss: 0.0550 - val_loss: 0.1013\n",
- "Epoch 1672/2500\n",
- "990/990 [==============================] - 0s 426us/step - loss: 0.0550 - val_loss: 0.1013\n",
- "Epoch 1673/2500\n",
- "990/990 [==============================] - 0s 420us/step - loss: 0.0550 - val_loss: 0.1013\n",
- "Epoch 1674/2500\n",
- "990/990 [==============================] - 0s 425us/step - loss: 0.0550 - val_loss: 0.1013\n",
- "Epoch 1675/2500\n",
- "990/990 [==============================] - 0s 422us/step - loss: 0.0550 - val_loss: 0.1013\n",
- "Epoch 1676/2500\n",
- "990/990 [==============================] - 0s 430us/step - loss: 0.0550 - val_loss: 0.1013\n",
- "Epoch 1677/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0550 - val_loss: 0.1013\n",
- "Epoch 1678/2500\n",
- "990/990 [==============================] - 0s 420us/step - loss: 0.0550 - val_loss: 0.1013\n",
- "Epoch 1679/2500\n",
- "990/990 [==============================] - 0s 426us/step - loss: 0.0550 - val_loss: 0.1013\n",
- "Epoch 1680/2500\n",
- "990/990 [==============================] - 0s 431us/step - loss: 0.0550 - val_loss: 0.1013\n",
- "Epoch 1681/2500\n",
- "990/990 [==============================] - 0s 421us/step - loss: 0.0550 - val_loss: 0.1013\n",
- "Epoch 1682/2500\n",
- "990/990 [==============================] - 0s 421us/step - loss: 0.0550 - val_loss: 0.1013\n",
- "Epoch 1683/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0550 - val_loss: 0.1013\n",
- "Epoch 1684/2500\n",
- "990/990 [==============================] - 0s 423us/step - loss: 0.0550 - val_loss: 0.1013\n",
- "Epoch 1685/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0550 - val_loss: 0.1013\n",
- "Epoch 1686/2500\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "990/990 [==============================] - 0s 430us/step - loss: 0.0550 - val_loss: 0.1013\n",
- "Epoch 1687/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0550 - val_loss: 0.1013\n",
- "Epoch 1688/2500\n",
- "990/990 [==============================] - 0s 424us/step - loss: 0.0550 - val_loss: 0.1013\n",
- "Epoch 1689/2500\n",
- "990/990 [==============================] - 0s 419us/step - loss: 0.0549 - val_loss: 0.1013\n",
- "Epoch 1690/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0549 - val_loss: 0.1013\n",
- "Epoch 1691/2500\n",
- "990/990 [==============================] - 0s 422us/step - loss: 0.0549 - val_loss: 0.1013\n",
- "Epoch 1692/2500\n",
- "990/990 [==============================] - 0s 426us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1693/2500\n",
- "990/990 [==============================] - 0s 425us/step - loss: 0.0549 - val_loss: 0.1013\n",
- "Epoch 1694/2500\n",
- "990/990 [==============================] - 0s 418us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1695/2500\n",
- "990/990 [==============================] - 0s 424us/step - loss: 0.0549 - val_loss: 0.1013\n",
- "Epoch 1696/2500\n",
- "990/990 [==============================] - 0s 425us/step - loss: 0.0549 - val_loss: 0.1013\n",
- "Epoch 1697/2500\n",
- "990/990 [==============================] - 0s 426us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1698/2500\n",
- "990/990 [==============================] - 0s 420us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1699/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1700/2500\n",
- "990/990 [==============================] - 0s 424us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1701/2500\n",
- "990/990 [==============================] - 0s 422us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1702/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1703/2500\n",
- "990/990 [==============================] - 0s 422us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1704/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1705/2500\n",
- "990/990 [==============================] - 0s 425us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1706/2500\n",
- "990/990 [==============================] - 0s 425us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1707/2500\n",
- "990/990 [==============================] - 0s 426us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1708/2500\n",
- "990/990 [==============================] - 0s 432us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1709/2500\n",
- "990/990 [==============================] - 0s 425us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1710/2500\n",
- "990/990 [==============================] - 0s 420us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1711/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1712/2500\n",
- "990/990 [==============================] - 0s 433us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1713/2500\n",
- "990/990 [==============================] - 0s 433us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1714/2500\n",
- "990/990 [==============================] - 0s 423us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1715/2500\n",
- "990/990 [==============================] - 0s 430us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1716/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1717/2500\n",
- "990/990 [==============================] - 0s 423us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1718/2500\n",
- "990/990 [==============================] - 0s 426us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1719/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1720/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1721/2500\n",
- "990/990 [==============================] - 0s 424us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1722/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1723/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1724/2500\n",
- "990/990 [==============================] - 0s 425us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1725/2500\n",
- "990/990 [==============================] - 0s 426us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1726/2500\n",
- "990/990 [==============================] - 0s 423us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1727/2500\n",
- "990/990 [==============================] - 0s 425us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1728/2500\n",
- "990/990 [==============================] - 0s 426us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1729/2500\n",
- "990/990 [==============================] - 0s 426us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1730/2500\n",
- "990/990 [==============================] - 0s 431us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1731/2500\n",
- "990/990 [==============================] - 0s 421us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1732/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1733/2500\n",
- "990/990 [==============================] - 0s 425us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1734/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1735/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0549 - val_loss: 0.1015\n",
- "Epoch 1736/2500\n",
- "990/990 [==============================] - 0s 426us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1737/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1738/2500\n",
- "990/990 [==============================] - 0s 425us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1739/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0549 - val_loss: 0.1015\n",
- "Epoch 1740/2500\n",
- "990/990 [==============================] - 0s 421us/step - loss: 0.0549 - val_loss: 0.1015\n",
- "Epoch 1741/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0549 - val_loss: 0.1014\n",
- "Epoch 1742/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1743/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1744/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1745/2500\n",
- "990/990 [==============================] - 0s 424us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1746/2500\n",
- "990/990 [==============================] - 0s 423us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1747/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1748/2500\n",
- "990/990 [==============================] - 0s 423us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1749/2500\n",
- "990/990 [==============================] - 0s 426us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1750/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1751/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1752/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1753/2500\n",
- "990/990 [==============================] - 0s 424us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1754/2500\n",
- "990/990 [==============================] - 0s 422us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1755/2500\n",
- "990/990 [==============================] - 0s 425us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1756/2500\n",
- "990/990 [==============================] - 0s 424us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1757/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1758/2500\n",
- "990/990 [==============================] - 0s 430us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1759/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1760/2500\n",
- "990/990 [==============================] - 0s 426us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1761/2500\n",
- "990/990 [==============================] - 0s 421us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1762/2500\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "990/990 [==============================] - 0s 430us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1763/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1764/2500\n",
- "990/990 [==============================] - 0s 447us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1765/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1766/2500\n",
- "990/990 [==============================] - 0s 435us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1767/2500\n",
- "990/990 [==============================] - 0s 433us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1768/2500\n",
- "990/990 [==============================] - 0s 435us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1769/2500\n",
- "990/990 [==============================] - 0s 443us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1770/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1771/2500\n",
- "990/990 [==============================] - 0s 434us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1772/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1773/2500\n",
- "990/990 [==============================] - 0s 425us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1774/2500\n",
- "990/990 [==============================] - 0s 426us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1775/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1776/2500\n",
- "990/990 [==============================] - 0s 424us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1777/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1778/2500\n",
- "990/990 [==============================] - 0s 426us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1779/2500\n",
- "990/990 [==============================] - 0s 441us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1780/2500\n",
- "990/990 [==============================] - 0s 436us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1781/2500\n",
- "990/990 [==============================] - 0s 431us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1782/2500\n",
- "990/990 [==============================] - 0s 463us/step - loss: 0.0548 - val_loss: 0.1016\n",
- "Epoch 1783/2500\n",
- "990/990 [==============================] - 0s 441us/step - loss: 0.0548 - val_loss: 0.1016\n",
- "Epoch 1784/2500\n",
- "990/990 [==============================] - 0s 410us/step - loss: 0.0548 - val_loss: 0.1016\n",
- "Epoch 1785/2500\n",
- "990/990 [==============================] - 0s 423us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1786/2500\n",
- "990/990 [==============================] - 0s 438us/step - loss: 0.0548 - val_loss: 0.1016\n",
- "Epoch 1787/2500\n",
- "990/990 [==============================] - 0s 442us/step - loss: 0.0548 - val_loss: 0.1016\n",
- "Epoch 1788/2500\n",
- "990/990 [==============================] - 0s 433us/step - loss: 0.0548 - val_loss: 0.1015\n",
- "Epoch 1789/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0548 - val_loss: 0.1016\n",
- "Epoch 1790/2500\n",
- "990/990 [==============================] - 0s 432us/step - loss: 0.0548 - val_loss: 0.1016\n",
- "Epoch 1791/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0548 - val_loss: 0.1016\n",
- "Epoch 1792/2500\n",
- "990/990 [==============================] - 0s 435us/step - loss: 0.0548 - val_loss: 0.1016\n",
- "Epoch 1793/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0548 - val_loss: 0.1016\n",
- "Epoch 1794/2500\n",
- "990/990 [==============================] - 0s 442us/step - loss: 0.0548 - val_loss: 0.1016\n",
- "Epoch 1795/2500\n",
- "990/990 [==============================] - 0s 439us/step - loss: 0.0548 - val_loss: 0.1016\n",
- "Epoch 1796/2500\n",
- "990/990 [==============================] - 0s 433us/step - loss: 0.0548 - val_loss: 0.1016\n",
- "Epoch 1797/2500\n",
- "990/990 [==============================] - 0s 430us/step - loss: 0.0548 - val_loss: 0.1016\n",
- "Epoch 1798/2500\n",
- "990/990 [==============================] - 0s 432us/step - loss: 0.0548 - val_loss: 0.1016\n",
- "Epoch 1799/2500\n",
- "990/990 [==============================] - 0s 432us/step - loss: 0.0547 - val_loss: 0.1016\n",
- "Epoch 1800/2500\n",
- "990/990 [==============================] - 0s 433us/step - loss: 0.0547 - val_loss: 0.1016\n",
- "Epoch 1801/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0547 - val_loss: 0.1016\n",
- "Epoch 1802/2500\n",
- "990/990 [==============================] - 0s 432us/step - loss: 0.0547 - val_loss: 0.1016\n",
- "Epoch 1803/2500\n",
- "990/990 [==============================] - 0s 434us/step - loss: 0.0547 - val_loss: 0.1016\n",
- "Epoch 1804/2500\n",
- "990/990 [==============================] - 0s 439us/step - loss: 0.0547 - val_loss: 0.1016\n",
- "Epoch 1805/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0547 - val_loss: 0.1016\n",
- "Epoch 1806/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0547 - val_loss: 0.1016\n",
- "Epoch 1807/2500\n",
- "990/990 [==============================] - 0s 436us/step - loss: 0.0547 - val_loss: 0.1016\n",
- "Epoch 1808/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0547 - val_loss: 0.1016\n",
- "Epoch 1809/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0547 - val_loss: 0.1016\n",
- "Epoch 1810/2500\n",
- "990/990 [==============================] - 0s 434us/step - loss: 0.0547 - val_loss: 0.1016\n",
- "Epoch 1811/2500\n",
- "990/990 [==============================] - 0s 432us/step - loss: 0.0547 - val_loss: 0.1016\n",
- "Epoch 1812/2500\n",
- "990/990 [==============================] - 0s 439us/step - loss: 0.0547 - val_loss: 0.1016\n",
- "Epoch 1813/2500\n",
- "990/990 [==============================] - 0s 425us/step - loss: 0.0547 - val_loss: 0.1016\n",
- "Epoch 1814/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0547 - val_loss: 0.1016\n",
- "Epoch 1815/2500\n",
- "990/990 [==============================] - 0s 430us/step - loss: 0.0547 - val_loss: 0.1016\n",
- "Epoch 1816/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0547 - val_loss: 0.1016\n",
- "Epoch 1817/2500\n",
- "990/990 [==============================] - 0s 432us/step - loss: 0.0547 - val_loss: 0.1016\n",
- "Epoch 1818/2500\n",
- "990/990 [==============================] - 0s 432us/step - loss: 0.0547 - val_loss: 0.1016\n",
- "Epoch 1819/2500\n",
- "990/990 [==============================] - 0s 425us/step - loss: 0.0547 - val_loss: 0.1016\n",
- "Epoch 1820/2500\n",
- "990/990 [==============================] - 0s 425us/step - loss: 0.0547 - val_loss: 0.1016\n",
- "Epoch 1821/2500\n",
- "990/990 [==============================] - 0s 433us/step - loss: 0.0547 - val_loss: 0.1016\n",
- "Epoch 1822/2500\n",
- "990/990 [==============================] - 0s 432us/step - loss: 0.0547 - val_loss: 0.1016\n",
- "Epoch 1823/2500\n",
- "990/990 [==============================] - 0s 430us/step - loss: 0.0547 - val_loss: 0.1016\n",
- "Epoch 1824/2500\n",
- "990/990 [==============================] - 0s 438us/step - loss: 0.0547 - val_loss: 0.1017\n",
- "Epoch 1825/2500\n",
- "990/990 [==============================] - 0s 430us/step - loss: 0.0547 - val_loss: 0.1017\n",
- "Epoch 1826/2500\n",
- "990/990 [==============================] - 0s 432us/step - loss: 0.0547 - val_loss: 0.1017\n",
- "Epoch 1827/2500\n",
- "990/990 [==============================] - 0s 433us/step - loss: 0.0547 - val_loss: 0.1017\n",
- "Epoch 1828/2500\n",
- "990/990 [==============================] - 0s 422us/step - loss: 0.0547 - val_loss: 0.1016\n",
- "Epoch 1829/2500\n",
- "990/990 [==============================] - 0s 436us/step - loss: 0.0547 - val_loss: 0.1017\n",
- "Epoch 1830/2500\n",
- "990/990 [==============================] - 0s 432us/step - loss: 0.0547 - val_loss: 0.1016\n",
- "Epoch 1831/2500\n",
- "990/990 [==============================] - 0s 425us/step - loss: 0.0547 - val_loss: 0.1017\n",
- "Epoch 1832/2500\n",
- "990/990 [==============================] - 0s 437us/step - loss: 0.0547 - val_loss: 0.1017\n",
- "Epoch 1833/2500\n",
- "990/990 [==============================] - 0s 439us/step - loss: 0.0547 - val_loss: 0.1017\n",
- "Epoch 1834/2500\n",
- "990/990 [==============================] - 0s 440us/step - loss: 0.0547 - val_loss: 0.1017\n",
- "Epoch 1835/2500\n",
- "990/990 [==============================] - 0s 431us/step - loss: 0.0547 - val_loss: 0.1017\n",
- "Epoch 1836/2500\n",
- "990/990 [==============================] - 0s 425us/step - loss: 0.0547 - val_loss: 0.1017\n",
- "Epoch 1837/2500\n",
- "990/990 [==============================] - 0s 434us/step - loss: 0.0547 - val_loss: 0.1017\n",
- "Epoch 1838/2500\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "990/990 [==============================] - 0s 436us/step - loss: 0.0547 - val_loss: 0.1017\n",
- "Epoch 1839/2500\n",
- "990/990 [==============================] - 0s 436us/step - loss: 0.0547 - val_loss: 0.1017\n",
- "Epoch 1840/2500\n",
- "990/990 [==============================] - 0s 422us/step - loss: 0.0547 - val_loss: 0.1017\n",
- "Epoch 1841/2500\n",
- "990/990 [==============================] - 0s 431us/step - loss: 0.0547 - val_loss: 0.1017\n",
- "Epoch 1842/2500\n",
- "990/990 [==============================] - 0s 436us/step - loss: 0.0547 - val_loss: 0.1017\n",
- "Epoch 1843/2500\n",
- "990/990 [==============================] - 0s 431us/step - loss: 0.0547 - val_loss: 0.1017\n",
- "Epoch 1844/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0547 - val_loss: 0.1017\n",
- "Epoch 1845/2500\n",
- "990/990 [==============================] - 0s 441us/step - loss: 0.0547 - val_loss: 0.1017\n",
- "Epoch 1846/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0547 - val_loss: 0.1017\n",
- "Epoch 1847/2500\n",
- "990/990 [==============================] - 0s 431us/step - loss: 0.0547 - val_loss: 0.1017\n",
- "Epoch 1848/2500\n",
- "990/990 [==============================] - 0s 430us/step - loss: 0.0547 - val_loss: 0.1017\n",
- "Epoch 1849/2500\n",
- "990/990 [==============================] - 0s 445us/step - loss: 0.0547 - val_loss: 0.1017\n",
- "Epoch 1850/2500\n",
- "990/990 [==============================] - 0s 436us/step - loss: 0.0547 - val_loss: 0.1017\n",
- "Epoch 1851/2500\n",
- "990/990 [==============================] - 0s 438us/step - loss: 0.0547 - val_loss: 0.1017\n",
- "Epoch 1852/2500\n",
- "990/990 [==============================] - 0s 423us/step - loss: 0.0547 - val_loss: 0.1017\n",
- "Epoch 1853/2500\n",
- "990/990 [==============================] - 0s 433us/step - loss: 0.0547 - val_loss: 0.1017\n",
- "Epoch 1854/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0547 - val_loss: 0.1017\n",
- "Epoch 1855/2500\n",
- "990/990 [==============================] - 0s 443us/step - loss: 0.0547 - val_loss: 0.1017\n",
- "Epoch 1856/2500\n",
- "990/990 [==============================] - 0s 435us/step - loss: 0.0547 - val_loss: 0.1017\n",
- "Epoch 1857/2500\n",
- "990/990 [==============================] - 0s 430us/step - loss: 0.0547 - val_loss: 0.1017\n",
- "Epoch 1858/2500\n",
- "990/990 [==============================] - 0s 430us/step - loss: 0.0547 - val_loss: 0.1017\n",
- "Epoch 1859/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0547 - val_loss: 0.1017\n",
- "Epoch 1860/2500\n",
- "990/990 [==============================] - 0s 434us/step - loss: 0.0547 - val_loss: 0.1017\n",
- "Epoch 1861/2500\n",
- "990/990 [==============================] - 0s 432us/step - loss: 0.0546 - val_loss: 0.1017\n",
- "Epoch 1862/2500\n",
- "990/990 [==============================] - 0s 432us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1863/2500\n",
- "990/990 [==============================] - 0s 437us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1864/2500\n",
- "990/990 [==============================] - 0s 437us/step - loss: 0.0546 - val_loss: 0.1017\n",
- "Epoch 1865/2500\n",
- "990/990 [==============================] - 0s 441us/step - loss: 0.0546 - val_loss: 0.1017\n",
- "Epoch 1866/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0546 - val_loss: 0.1017\n",
- "Epoch 1867/2500\n",
- "990/990 [==============================] - 0s 431us/step - loss: 0.0546 - val_loss: 0.1017\n",
- "Epoch 1868/2500\n",
- "990/990 [==============================] - 0s 437us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1869/2500\n",
- "990/990 [==============================] - 0s 435us/step - loss: 0.0546 - val_loss: 0.1017\n",
- "Epoch 1870/2500\n",
- "990/990 [==============================] - 0s 437us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1871/2500\n",
- "990/990 [==============================] - 0s 425us/step - loss: 0.0546 - val_loss: 0.1017\n",
- "Epoch 1872/2500\n",
- "990/990 [==============================] - 0s 438us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1873/2500\n",
- "990/990 [==============================] - 0s 446us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1874/2500\n",
- "990/990 [==============================] - 0s 433us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1875/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1876/2500\n",
- "990/990 [==============================] - 0s 431us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1877/2500\n",
- "990/990 [==============================] - 0s 423us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1878/2500\n",
- "990/990 [==============================] - 0s 424us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1879/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1880/2500\n",
- "990/990 [==============================] - 0s 431us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1881/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1882/2500\n",
- "990/990 [==============================] - 0s 425us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1883/2500\n",
- "990/990 [==============================] - 0s 426us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1884/2500\n",
- "990/990 [==============================] - 0s 423us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1885/2500\n",
- "990/990 [==============================] - 0s 422us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1886/2500\n",
- "990/990 [==============================] - 0s 431us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1887/2500\n",
- "990/990 [==============================] - 0s 431us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1888/2500\n",
- "990/990 [==============================] - 0s 426us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1889/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1890/2500\n",
- "990/990 [==============================] - 0s 436us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1891/2500\n",
- "990/990 [==============================] - 0s 425us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1892/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1893/2500\n",
- "990/990 [==============================] - 0s 433us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1894/2500\n",
- "990/990 [==============================] - 0s 432us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1895/2500\n",
- "990/990 [==============================] - 0s 433us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1896/2500\n",
- "990/990 [==============================] - 0s 433us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1897/2500\n",
- "990/990 [==============================] - 0s 445us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1898/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1899/2500\n",
- "990/990 [==============================] - 0s 437us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1900/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1901/2500\n",
- "990/990 [==============================] - 0s 437us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1902/2500\n",
- "990/990 [==============================] - 0s 435us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1903/2500\n",
- "990/990 [==============================] - 0s 434us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1904/2500\n",
- "990/990 [==============================] - 0s 430us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1905/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1906/2500\n",
- "990/990 [==============================] - 0s 431us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1907/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0546 - val_loss: 0.1019\n",
- "Epoch 1908/2500\n",
- "990/990 [==============================] - 0s 425us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1909/2500\n",
- "990/990 [==============================] - 0s 425us/step - loss: 0.0546 - val_loss: 0.1019\n",
- "Epoch 1910/2500\n",
- "990/990 [==============================] - 0s 439us/step - loss: 0.0546 - val_loss: 0.1019\n",
- "Epoch 1911/2500\n",
- "990/990 [==============================] - 0s 438us/step - loss: 0.0546 - val_loss: 0.1019\n",
- "Epoch 1912/2500\n",
- "990/990 [==============================] - 0s 440us/step - loss: 0.0546 - val_loss: 0.1019\n",
- "Epoch 1913/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0546 - val_loss: 0.1019\n",
- "Epoch 1914/2500\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "990/990 [==============================] - 0s 426us/step - loss: 0.0546 - val_loss: 0.1019\n",
- "Epoch 1915/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0546 - val_loss: 0.1019\n",
- "Epoch 1916/2500\n",
- "990/990 [==============================] - 0s 431us/step - loss: 0.0546 - val_loss: 0.1019\n",
- "Epoch 1917/2500\n",
- "990/990 [==============================] - 0s 435us/step - loss: 0.0546 - val_loss: 0.1018\n",
- "Epoch 1918/2500\n",
- "990/990 [==============================] - 0s 431us/step - loss: 0.0546 - val_loss: 0.1019\n",
- "Epoch 1919/2500\n",
- "990/990 [==============================] - 0s 439us/step - loss: 0.0546 - val_loss: 0.1019\n",
- "Epoch 1920/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0546 - val_loss: 0.1019\n",
- "Epoch 1921/2500\n",
- "990/990 [==============================] - 0s 434us/step - loss: 0.0546 - val_loss: 0.1019\n",
- "Epoch 1922/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0546 - val_loss: 0.1019\n",
- "Epoch 1923/2500\n",
- "990/990 [==============================] - 0s 434us/step - loss: 0.0546 - val_loss: 0.1019\n",
- "Epoch 1924/2500\n",
- "990/990 [==============================] - 0s 432us/step - loss: 0.0546 - val_loss: 0.1019\n",
- "Epoch 1925/2500\n",
- "990/990 [==============================] - 0s 442us/step - loss: 0.0546 - val_loss: 0.1019\n",
- "Epoch 1926/2500\n",
- "990/990 [==============================] - 0s 425us/step - loss: 0.0545 - val_loss: 0.1019\n",
- "Epoch 1927/2500\n",
- "990/990 [==============================] - 0s 430us/step - loss: 0.0545 - val_loss: 0.1019\n",
- "Epoch 1928/2500\n",
- "990/990 [==============================] - 0s 441us/step - loss: 0.0545 - val_loss: 0.1019\n",
- "Epoch 1929/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0545 - val_loss: 0.1019\n",
- "Epoch 1930/2500\n",
- "990/990 [==============================] - 0s 445us/step - loss: 0.0545 - val_loss: 0.1019\n",
- "Epoch 1931/2500\n",
- "990/990 [==============================] - 0s 425us/step - loss: 0.0545 - val_loss: 0.1019\n",
- "Epoch 1932/2500\n",
- "990/990 [==============================] - 0s 431us/step - loss: 0.0545 - val_loss: 0.1019\n",
- "Epoch 1933/2500\n",
- "990/990 [==============================] - 0s 431us/step - loss: 0.0545 - val_loss: 0.1019\n",
- "Epoch 1934/2500\n",
- "990/990 [==============================] - 0s 425us/step - loss: 0.0545 - val_loss: 0.1019\n",
- "Epoch 1935/2500\n",
- "990/990 [==============================] - 0s 432us/step - loss: 0.0545 - val_loss: 0.1019\n",
- "Epoch 1936/2500\n",
- "990/990 [==============================] - 0s 432us/step - loss: 0.0545 - val_loss: 0.1019\n",
- "Epoch 1937/2500\n",
- "990/990 [==============================] - 0s 432us/step - loss: 0.0545 - val_loss: 0.1019\n",
- "Epoch 1938/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0545 - val_loss: 0.1019\n",
- "Epoch 1939/2500\n",
- "990/990 [==============================] - 0s 421us/step - loss: 0.0545 - val_loss: 0.1019\n",
- "Epoch 1940/2500\n",
- "990/990 [==============================] - 0s 425us/step - loss: 0.0545 - val_loss: 0.1019\n",
- "Epoch 1941/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0545 - val_loss: 0.1019\n",
- "Epoch 1942/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0545 - val_loss: 0.1019\n",
- "Epoch 1943/2500\n",
- "990/990 [==============================] - 0s 436us/step - loss: 0.0545 - val_loss: 0.1019\n",
- "Epoch 1944/2500\n",
- "990/990 [==============================] - 0s 432us/step - loss: 0.0545 - val_loss: 0.1019\n",
- "Epoch 1945/2500\n",
- "990/990 [==============================] - 0s 433us/step - loss: 0.0545 - val_loss: 0.1019\n",
- "Epoch 1946/2500\n",
- "990/990 [==============================] - 0s 431us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1947/2500\n",
- "990/990 [==============================] - 0s 426us/step - loss: 0.0545 - val_loss: 0.1019\n",
- "Epoch 1948/2500\n",
- "990/990 [==============================] - 0s 434us/step - loss: 0.0545 - val_loss: 0.1019\n",
- "Epoch 1949/2500\n",
- "990/990 [==============================] - 0s 434us/step - loss: 0.0545 - val_loss: 0.1019\n",
- "Epoch 1950/2500\n",
- "990/990 [==============================] - 0s 431us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1951/2500\n",
- "990/990 [==============================] - 0s 426us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1952/2500\n",
- "990/990 [==============================] - 0s 434us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1953/2500\n",
- "990/990 [==============================] - 0s 430us/step - loss: 0.0545 - val_loss: 0.1019\n",
- "Epoch 1954/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1955/2500\n",
- "990/990 [==============================] - 0s 426us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1956/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1957/2500\n",
- "990/990 [==============================] - 0s 436us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1958/2500\n",
- "990/990 [==============================] - 0s 431us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1959/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1960/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1961/2500\n",
- "990/990 [==============================] - 0s 426us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1962/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1963/2500\n",
- "990/990 [==============================] - 0s 433us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1964/2500\n",
- "990/990 [==============================] - 0s 426us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1965/2500\n",
- "990/990 [==============================] - 0s 423us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1966/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1967/2500\n",
- "990/990 [==============================] - 0s 426us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1968/2500\n",
- "990/990 [==============================] - 0s 430us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1969/2500\n",
- "990/990 [==============================] - 0s 430us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1970/2500\n",
- "990/990 [==============================] - 0s 431us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1971/2500\n",
- "990/990 [==============================] - 0s 430us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1972/2500\n",
- "990/990 [==============================] - 0s 423us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1973/2500\n",
- "990/990 [==============================] - 0s 430us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1974/2500\n",
- "990/990 [==============================] - 0s 426us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1975/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1976/2500\n",
- "990/990 [==============================] - 0s 449us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1977/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1978/2500\n",
- "990/990 [==============================] - 0s 432us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1979/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1980/2500\n",
- "990/990 [==============================] - 0s 437us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1981/2500\n",
- "990/990 [==============================] - 0s 433us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1982/2500\n",
- "990/990 [==============================] - 0s 433us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1983/2500\n",
- "990/990 [==============================] - 0s 431us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1984/2500\n",
- "990/990 [==============================] - 0s 437us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1985/2500\n",
- "990/990 [==============================] - 0s 437us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1986/2500\n",
- "990/990 [==============================] - 0s 430us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1987/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1988/2500\n",
- "990/990 [==============================] - 0s 446us/step - loss: 0.0545 - val_loss: 0.1021\n",
- "Epoch 1989/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1990/2500\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "990/990 [==============================] - 0s 440us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1991/2500\n",
- "990/990 [==============================] - 0s 434us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1992/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0545 - val_loss: 0.1020\n",
- "Epoch 1993/2500\n",
- "990/990 [==============================] - 0s 442us/step - loss: 0.0545 - val_loss: 0.1021\n",
- "Epoch 1994/2500\n",
- "990/990 [==============================] - 0s 436us/step - loss: 0.0545 - val_loss: 0.1021\n",
- "Epoch 1995/2500\n",
- "990/990 [==============================] - 0s 444us/step - loss: 0.0545 - val_loss: 0.1021\n",
- "Epoch 1996/2500\n",
- "990/990 [==============================] - 0s 433us/step - loss: 0.0545 - val_loss: 0.1021\n",
- "Epoch 1997/2500\n",
- "990/990 [==============================] - 0s 430us/step - loss: 0.0544 - val_loss: 0.1021\n",
- "Epoch 1998/2500\n",
- "990/990 [==============================] - 0s 434us/step - loss: 0.0544 - val_loss: 0.1021\n",
- "Epoch 1999/2500\n",
- "990/990 [==============================] - 0s 436us/step - loss: 0.0544 - val_loss: 0.1021\n",
- "Epoch 2000/2500\n",
- "990/990 [==============================] - 0s 433us/step - loss: 0.0544 - val_loss: 0.1021\n",
- "Epoch 2001/2500\n",
- "990/990 [==============================] - 0s 432us/step - loss: 0.0544 - val_loss: 0.1021\n",
- "Epoch 2002/2500\n",
- "990/990 [==============================] - 0s 435us/step - loss: 0.0544 - val_loss: 0.1021\n",
- "Epoch 2003/2500\n",
- "990/990 [==============================] - 0s 446us/step - loss: 0.0544 - val_loss: 0.1021\n",
- "Epoch 2004/2500\n",
- "990/990 [==============================] - 0s 436us/step - loss: 0.0544 - val_loss: 0.1021\n",
- "Epoch 2005/2500\n",
- "990/990 [==============================] - 0s 431us/step - loss: 0.0544 - val_loss: 0.1021\n",
- "Epoch 2006/2500\n",
- "990/990 [==============================] - 0s 433us/step - loss: 0.0544 - val_loss: 0.1021\n",
- "Epoch 2007/2500\n",
- "990/990 [==============================] - 0s 447us/step - loss: 0.0544 - val_loss: 0.1021\n",
- "Epoch 2008/2500\n",
- "990/990 [==============================] - 0s 436us/step - loss: 0.0544 - val_loss: 0.1021\n",
- "Epoch 2009/2500\n",
- "990/990 [==============================] - 0s 435us/step - loss: 0.0544 - val_loss: 0.1021\n",
- "Epoch 2010/2500\n",
- "990/990 [==============================] - 0s 433us/step - loss: 0.0544 - val_loss: 0.1021\n",
- "Epoch 2011/2500\n",
- "990/990 [==============================] - 0s 435us/step - loss: 0.0544 - val_loss: 0.1021\n",
- "Epoch 2012/2500\n",
- "990/990 [==============================] - 0s 431us/step - loss: 0.0544 - val_loss: 0.1021\n",
- "Epoch 2013/2500\n",
- "990/990 [==============================] - 0s 442us/step - loss: 0.0544 - val_loss: 0.1021\n",
- "Epoch 2014/2500\n",
- "990/990 [==============================] - 0s 444us/step - loss: 0.0544 - val_loss: 0.1021\n",
- "Epoch 2015/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0544 - val_loss: 0.1021\n",
- "Epoch 2016/2500\n",
- "990/990 [==============================] - 0s 438us/step - loss: 0.0544 - val_loss: 0.1021\n",
- "Epoch 2017/2500\n",
- "990/990 [==============================] - 0s 447us/step - loss: 0.0544 - val_loss: 0.1021\n",
- "Epoch 2018/2500\n",
- "990/990 [==============================] - 0s 448us/step - loss: 0.0544 - val_loss: 0.1021\n",
- "Epoch 2019/2500\n",
- "990/990 [==============================] - 0s 444us/step - loss: 0.0544 - val_loss: 0.1021\n",
- "Epoch 2020/2500\n",
- "990/990 [==============================] - 0s 444us/step - loss: 0.0544 - val_loss: 0.1021\n",
- "Epoch 2021/2500\n",
- "990/990 [==============================] - 0s 470us/step - loss: 0.0544 - val_loss: 0.1021\n",
- "Epoch 2022/2500\n",
- "990/990 [==============================] - 0s 446us/step - loss: 0.0544 - val_loss: 0.1021\n",
- "Epoch 2023/2500\n",
- "990/990 [==============================] - 0s 439us/step - loss: 0.0544 - val_loss: 0.1021\n",
- "Epoch 2024/2500\n",
- "990/990 [==============================] - 0s 441us/step - loss: 0.0544 - val_loss: 0.1021\n",
- "Epoch 2025/2500\n",
- "990/990 [==============================] - 0s 439us/step - loss: 0.0544 - val_loss: 0.1021\n",
- "Epoch 2026/2500\n",
- "990/990 [==============================] - 0s 441us/step - loss: 0.0544 - val_loss: 0.1021\n",
- "Epoch 2027/2500\n",
- "990/990 [==============================] - 0s 446us/step - loss: 0.0544 - val_loss: 0.1021\n",
- "Epoch 2028/2500\n",
- "990/990 [==============================] - 0s 439us/step - loss: 0.0544 - val_loss: 0.1021\n",
- "Epoch 2029/2500\n",
- "990/990 [==============================] - 0s 441us/step - loss: 0.0544 - val_loss: 0.1022\n",
- "Epoch 2030/2500\n",
- "990/990 [==============================] - 0s 445us/step - loss: 0.0544 - val_loss: 0.1021\n",
- "Epoch 2031/2500\n",
- "990/990 [==============================] - 0s 441us/step - loss: 0.0544 - val_loss: 0.1022\n",
- "Epoch 2032/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0544 - val_loss: 0.1021\n",
- "Epoch 2033/2500\n",
- "990/990 [==============================] - 0s 466us/step - loss: 0.0544 - val_loss: 0.1021\n",
- "Epoch 2034/2500\n",
- "990/990 [==============================] - 0s 467us/step - loss: 0.0544 - val_loss: 0.1021\n",
- "Epoch 2035/2500\n",
- "990/990 [==============================] - 0s 483us/step - loss: 0.0544 - val_loss: 0.1022\n",
- "Epoch 2036/2500\n",
- "990/990 [==============================] - 0s 443us/step - loss: 0.0544 - val_loss: 0.1022\n",
- "Epoch 2037/2500\n",
- "990/990 [==============================] - 0s 457us/step - loss: 0.0544 - val_loss: 0.1021\n",
- "Epoch 2038/2500\n",
- "990/990 [==============================] - 0s 489us/step - loss: 0.0544 - val_loss: 0.1022\n",
- "Epoch 2039/2500\n",
- "990/990 [==============================] - 0s 442us/step - loss: 0.0544 - val_loss: 0.1022\n",
- "Epoch 2040/2500\n",
- "990/990 [==============================] - 0s 437us/step - loss: 0.0544 - val_loss: 0.1022\n",
- "Epoch 2041/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0544 - val_loss: 0.1022\n",
- "Epoch 2042/2500\n",
- "990/990 [==============================] - 0s 417us/step - loss: 0.0544 - val_loss: 0.1022\n",
- "Epoch 2043/2500\n",
- "990/990 [==============================] - 0s 418us/step - loss: 0.0544 - val_loss: 0.1022\n",
- "Epoch 2044/2500\n",
- "990/990 [==============================] - 0s 435us/step - loss: 0.0544 - val_loss: 0.1022\n",
- "Epoch 2045/2500\n",
- "990/990 [==============================] - 0s 441us/step - loss: 0.0544 - val_loss: 0.1022\n",
- "Epoch 2046/2500\n",
- "990/990 [==============================] - 0s 430us/step - loss: 0.0544 - val_loss: 0.1022\n",
- "Epoch 2047/2500\n",
- "990/990 [==============================] - 0s 421us/step - loss: 0.0544 - val_loss: 0.1022\n",
- "Epoch 2048/2500\n",
- "990/990 [==============================] - 0s 414us/step - loss: 0.0544 - val_loss: 0.1022\n",
- "Epoch 2049/2500\n",
- "990/990 [==============================] - 0s 419us/step - loss: 0.0544 - val_loss: 0.1022\n",
- "Epoch 2050/2500\n",
- "990/990 [==============================] - 0s 418us/step - loss: 0.0544 - val_loss: 0.1022\n",
- "Epoch 2051/2500\n",
- "990/990 [==============================] - 0s 418us/step - loss: 0.0544 - val_loss: 0.1022\n",
- "Epoch 2052/2500\n",
- "990/990 [==============================] - 0s 420us/step - loss: 0.0544 - val_loss: 0.1022\n",
- "Epoch 2053/2500\n",
- "990/990 [==============================] - 0s 417us/step - loss: 0.0544 - val_loss: 0.1022\n",
- "Epoch 2054/2500\n",
- "990/990 [==============================] - 0s 420us/step - loss: 0.0544 - val_loss: 0.1022\n",
- "Epoch 2055/2500\n",
- "990/990 [==============================] - 0s 417us/step - loss: 0.0544 - val_loss: 0.1022\n",
- "Epoch 2056/2500\n",
- "990/990 [==============================] - 0s 423us/step - loss: 0.0544 - val_loss: 0.1022\n",
- "Epoch 2057/2500\n",
- "990/990 [==============================] - 0s 419us/step - loss: 0.0544 - val_loss: 0.1022\n",
- "Epoch 2058/2500\n",
- "990/990 [==============================] - 0s 422us/step - loss: 0.0544 - val_loss: 0.1022\n",
- "Epoch 2059/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0544 - val_loss: 0.1022\n",
- "Epoch 2060/2500\n",
- "990/990 [==============================] - 0s 419us/step - loss: 0.0544 - val_loss: 0.1022\n",
- "Epoch 2061/2500\n",
- "990/990 [==============================] - 0s 426us/step - loss: 0.0544 - val_loss: 0.1022\n",
- "Epoch 2062/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0544 - val_loss: 0.1022\n",
- "Epoch 2063/2500\n",
- "990/990 [==============================] - 0s 451us/step - loss: 0.0544 - val_loss: 0.1022\n",
- "Epoch 2064/2500\n",
- "990/990 [==============================] - 0s 423us/step - loss: 0.0544 - val_loss: 0.1022\n",
- "Epoch 2065/2500\n",
- "990/990 [==============================] - 0s 419us/step - loss: 0.0544 - val_loss: 0.1022\n",
- "Epoch 2066/2500\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "990/990 [==============================] - 0s 421us/step - loss: 0.0544 - val_loss: 0.1022\n",
- "Epoch 2067/2500\n",
- "990/990 [==============================] - 0s 418us/step - loss: 0.0544 - val_loss: 0.1022\n",
- "Epoch 2068/2500\n",
- "990/990 [==============================] - 0s 425us/step - loss: 0.0544 - val_loss: 0.1022\n",
- "Epoch 2069/2500\n",
- "990/990 [==============================] - 0s 425us/step - loss: 0.0544 - val_loss: 0.1023\n",
- "Epoch 2070/2500\n",
- "990/990 [==============================] - 0s 418us/step - loss: 0.0544 - val_loss: 0.1022\n",
- "Epoch 2071/2500\n",
- "990/990 [==============================] - 0s 419us/step - loss: 0.0544 - val_loss: 0.1022\n",
- "Epoch 2072/2500\n",
- "990/990 [==============================] - 0s 420us/step - loss: 0.0544 - val_loss: 0.1023\n",
- "Epoch 2073/2500\n",
- "990/990 [==============================] - 0s 420us/step - loss: 0.0544 - val_loss: 0.1022\n",
- "Epoch 2074/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0543 - val_loss: 0.1022\n",
- "Epoch 2075/2500\n",
- "990/990 [==============================] - 0s 429us/step - loss: 0.0543 - val_loss: 0.1022\n",
- "Epoch 2076/2500\n",
- "990/990 [==============================] - 0s 417us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2077/2500\n",
- "990/990 [==============================] - 0s 414us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2078/2500\n",
- "990/990 [==============================] - 0s 417us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2079/2500\n",
- "990/990 [==============================] - 0s 417us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2080/2500\n",
- "990/990 [==============================] - 0s 421us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2081/2500\n",
- "990/990 [==============================] - 0s 419us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2082/2500\n",
- "990/990 [==============================] - 0s 433us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2083/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2084/2500\n",
- "990/990 [==============================] - 0s 428us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2085/2500\n",
- "990/990 [==============================] - 0s 424us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2086/2500\n",
- "990/990 [==============================] - 0s 433us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2087/2500\n",
- "990/990 [==============================] - 0s 421us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2088/2500\n",
- "990/990 [==============================] - 0s 422us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2089/2500\n",
- "990/990 [==============================] - 0s 426us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2090/2500\n",
- "990/990 [==============================] - 0s 422us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2091/2500\n",
- "990/990 [==============================] - 0s 418us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2092/2500\n",
- "990/990 [==============================] - 0s 417us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2093/2500\n",
- "990/990 [==============================] - 0s 425us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2094/2500\n",
- "990/990 [==============================] - 0s 413us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2095/2500\n",
- "990/990 [==============================] - 0s 424us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2096/2500\n",
- "990/990 [==============================] - 0s 416us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2097/2500\n",
- "990/990 [==============================] - 0s 414us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2098/2500\n",
- "990/990 [==============================] - 0s 416us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2099/2500\n",
- "990/990 [==============================] - 0s 420us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2100/2500\n",
- "990/990 [==============================] - 0s 419us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2101/2500\n",
- "990/990 [==============================] - 0s 417us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2102/2500\n",
- "990/990 [==============================] - 0s 416us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2103/2500\n",
- "990/990 [==============================] - 0s 414us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2104/2500\n",
- "990/990 [==============================] - 0s 422us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2105/2500\n",
- "990/990 [==============================] - 0s 421us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2106/2500\n",
- "990/990 [==============================] - 0s 416us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2107/2500\n",
- "990/990 [==============================] - 0s 425us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2108/2500\n",
- "990/990 [==============================] - 0s 427us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2109/2500\n",
- "990/990 [==============================] - 0s 469us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2110/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2111/2500\n",
- "990/990 [==============================] - 0s 465us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2112/2500\n",
- "990/990 [==============================] - 0s 460us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2113/2500\n",
- "990/990 [==============================] - 0s 459us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2114/2500\n",
- "990/990 [==============================] - 0s 462us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2115/2500\n",
- "990/990 [==============================] - 0s 455us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2116/2500\n",
- "990/990 [==============================] - 0s 446us/step - loss: 0.0543 - val_loss: 0.1024\n",
- "Epoch 2117/2500\n",
- "990/990 [==============================] - 0s 466us/step - loss: 0.0543 - val_loss: 0.1023\n",
- "Epoch 2118/2500\n",
- "990/990 [==============================] - 0s 474us/step - loss: 0.0543 - val_loss: 0.1024\n",
- "Epoch 2119/2500\n",
- "990/990 [==============================] - 0s 474us/step - loss: 0.0543 - val_loss: 0.1024\n",
- "Epoch 2120/2500\n",
- "990/990 [==============================] - 0s 458us/step - loss: 0.0543 - val_loss: 0.1024\n",
- "Epoch 2121/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0543 - val_loss: 0.1024\n",
- "Epoch 2122/2500\n",
- "990/990 [==============================] - 0s 444us/step - loss: 0.0543 - val_loss: 0.1024\n",
- "Epoch 2123/2500\n",
- "990/990 [==============================] - 0s 448us/step - loss: 0.0543 - val_loss: 0.1024\n",
- "Epoch 2124/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0543 - val_loss: 0.1024\n",
- "Epoch 2125/2500\n",
- "990/990 [==============================] - 0s 465us/step - loss: 0.0543 - val_loss: 0.1024\n",
- "Epoch 2126/2500\n",
- "990/990 [==============================] - 0s 456us/step - loss: 0.0543 - val_loss: 0.1024\n",
- "Epoch 2127/2500\n",
- "990/990 [==============================] - 0s 460us/step - loss: 0.0543 - val_loss: 0.1024\n",
- "Epoch 2128/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0543 - val_loss: 0.1024\n",
- "Epoch 2129/2500\n",
- "990/990 [==============================] - 0s 463us/step - loss: 0.0543 - val_loss: 0.1024\n",
- "Epoch 2130/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0543 - val_loss: 0.1024\n",
- "Epoch 2131/2500\n",
- "990/990 [==============================] - 0s 445us/step - loss: 0.0543 - val_loss: 0.1024\n",
- "Epoch 2132/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0543 - val_loss: 0.1024\n",
- "Epoch 2133/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0543 - val_loss: 0.1024\n",
- "Epoch 2134/2500\n",
- "990/990 [==============================] - 0s 453us/step - loss: 0.0543 - val_loss: 0.1024\n",
- "Epoch 2135/2500\n",
- "990/990 [==============================] - 0s 452us/step - loss: 0.0543 - val_loss: 0.1024\n",
- "Epoch 2136/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0543 - val_loss: 0.1024\n",
- "Epoch 2137/2500\n",
- "990/990 [==============================] - 0s 440us/step - loss: 0.0543 - val_loss: 0.1024\n",
- "Epoch 2138/2500\n",
- "990/990 [==============================] - 0s 477us/step - loss: 0.0543 - val_loss: 0.1024\n",
- "Epoch 2139/2500\n",
- "990/990 [==============================] - 0s 447us/step - loss: 0.0543 - val_loss: 0.1024\n",
- "Epoch 2140/2500\n",
- "990/990 [==============================] - 0s 443us/step - loss: 0.0543 - val_loss: 0.1024\n",
- "Epoch 2141/2500\n",
- "990/990 [==============================] - 0s 471us/step - loss: 0.0543 - val_loss: 0.1024\n",
- "Epoch 2142/2500\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "990/990 [==============================] - 0s 452us/step - loss: 0.0543 - val_loss: 0.1024\n",
- "Epoch 2143/2500\n",
- "990/990 [==============================] - 0s 450us/step - loss: 0.0543 - val_loss: 0.1024\n",
- "Epoch 2144/2500\n",
- " 20/990 [..............................] - ETA: 0s - loss: 0.0545"
+ "Epoch 1187/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0563 - val_loss: 0.1006\n",
+ "Epoch 1188/1200\n",
+ "990/990 [==============================] - 0s 416us/step - loss: 0.0563 - val_loss: 0.1006\n",
+ "Epoch 1189/1200\n",
+ "990/990 [==============================] - 0s 417us/step - loss: 0.0563 - val_loss: 0.1006\n",
+ "Epoch 1190/1200\n",
+ "990/990 [==============================] - 0s 423us/step - loss: 0.0563 - val_loss: 0.1006\n",
+ "Epoch 1191/1200\n",
+ "990/990 [==============================] - 0s 435us/step - loss: 0.0563 - val_loss: 0.1006\n",
+ "Epoch 1192/1200\n",
+ "990/990 [==============================] - 0s 415us/step - loss: 0.0563 - val_loss: 0.1006\n",
+ "Epoch 1193/1200\n",
+ "990/990 [==============================] - 0s 425us/step - loss: 0.0563 - val_loss: 0.1006\n",
+ "Epoch 1194/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0563 - val_loss: 0.1006\n",
+ "Epoch 1195/1200\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0563 - val_loss: 0.1006\n",
+ "Epoch 1196/1200\n",
+ "990/990 [==============================] - 0s 420us/step - loss: 0.0563 - val_loss: 0.1006\n",
+ "Epoch 1197/1200\n",
+ "990/990 [==============================] - 0s 417us/step - loss: 0.0563 - val_loss: 0.1006\n",
+ "Epoch 1198/1200\n",
+ "990/990 [==============================] - 0s 419us/step - loss: 0.0563 - val_loss: 0.1006\n",
+ "Epoch 1199/1200\n",
+ "990/990 [==============================] - 0s 418us/step - loss: 0.0563 - val_loss: 0.1006\n",
+ "Epoch 1200/1200\n",
+ "990/990 [==============================] - 0s 421us/step - loss: 0.0563 - val_loss: 0.1006\n"
]
},
{
- "ename": "KeyboardInterrupt",
- "evalue": "",
- "output_type": "error",
- "traceback": [
- "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
- "\u001b[0;31mKeyboardInterrupt\u001b[0m Traceback (most recent call last)",
- "\u001b[0;32m<ipython-input-22-91cd69abf217>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m 3\u001b[0m \u001b[0mbatch_size\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m20\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 4\u001b[0m \u001b[0mshuffle\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mTrue\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 5\u001b[0;31m validation_data = (vali_data, vali_data))\n\u001b[0m",
- "\u001b[0;32m~/PycharmProjects/TensorPlay/venv/lib/python3.6/site-packages/keras/engine/training.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, x, y, batch_size, epochs, verbose, callbacks, validation_split, validation_data, shuffle, class_weight, sample_weight, initial_epoch, steps_per_epoch, validation_steps, **kwargs)\u001b[0m\n\u001b[1;32m 1703\u001b[0m \u001b[0minitial_epoch\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0minitial_epoch\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1704\u001b[0m \u001b[0msteps_per_epoch\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0msteps_per_epoch\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1705\u001b[0;31m validation_steps=validation_steps)\n\u001b[0m\u001b[1;32m 1706\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1707\u001b[0m def evaluate(self, x=None, y=None,\n",
- "\u001b[0;32m~/PycharmProjects/TensorPlay/venv/lib/python3.6/site-packages/keras/engine/training.py\u001b[0m in \u001b[0;36m_fit_loop\u001b[0;34m(self, f, ins, out_labels, batch_size, epochs, verbose, callbacks, val_f, val_ins, shuffle, callback_metrics, initial_epoch, steps_per_epoch, validation_steps)\u001b[0m\n\u001b[1;32m 1234\u001b[0m \u001b[0mins_batch\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mins_batch\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtoarray\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1235\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1236\u001b[0;31m \u001b[0mouts\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mins_batch\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1237\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mouts\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlist\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1238\u001b[0m \u001b[0mouts\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mouts\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
- "\u001b[0;32m~/PycharmProjects/TensorPlay/venv/lib/python3.6/site-packages/keras/backend/tensorflow_backend.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, inputs)\u001b[0m\n\u001b[1;32m 2480\u001b[0m \u001b[0msession\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mget_session\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2481\u001b[0m updated = session.run(fetches=fetches, feed_dict=feed_dict,\n\u001b[0;32m-> 2482\u001b[0;31m **self.session_kwargs)\n\u001b[0m\u001b[1;32m 2483\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mupdated\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moutputs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2484\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
- "\u001b[0;32m~/PycharmProjects/TensorPlay/venv/lib/python3.6/site-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36mrun\u001b[0;34m(self, fetches, feed_dict, options, run_metadata)\u001b[0m\n\u001b[1;32m 903\u001b[0m \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 904\u001b[0m result = self._run(None, fetches, feed_dict, options_ptr,\n\u001b[0;32m--> 905\u001b[0;31m run_metadata_ptr)\n\u001b[0m\u001b[1;32m 906\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mrun_metadata\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 907\u001b[0m \u001b[0mproto_data\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtf_session\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mTF_GetBuffer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mrun_metadata_ptr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
- "\u001b[0;32m~/PycharmProjects/TensorPlay/venv/lib/python3.6/site-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36m_run\u001b[0;34m(self, handle, fetches, feed_dict, options, run_metadata)\u001b[0m\n\u001b[1;32m 1138\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mfinal_fetches\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0mfinal_targets\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mhandle\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0mfeed_dict_tensor\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1139\u001b[0m results = self._do_run(handle, final_targets, final_fetches,\n\u001b[0;32m-> 1140\u001b[0;31m feed_dict_tensor, options, run_metadata)\n\u001b[0m\u001b[1;32m 1141\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1142\u001b[0m \u001b[0mresults\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
- "\u001b[0;32m~/PycharmProjects/TensorPlay/venv/lib/python3.6/site-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36m_do_run\u001b[0;34m(self, handle, target_list, fetch_list, feed_dict, options, run_metadata)\u001b[0m\n\u001b[1;32m 1319\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mhandle\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1320\u001b[0m return self._do_call(_run_fn, feeds, fetches, targets, options,\n\u001b[0;32m-> 1321\u001b[0;31m run_metadata)\n\u001b[0m\u001b[1;32m 1322\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1323\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_do_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0m_prun_fn\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mhandle\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfeeds\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfetches\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
- "\u001b[0;32m~/PycharmProjects/TensorPlay/venv/lib/python3.6/site-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36m_do_call\u001b[0;34m(self, fn, *args)\u001b[0m\n\u001b[1;32m 1325\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0m_do_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfn\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1326\u001b[0m \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1327\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mfn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1328\u001b[0m \u001b[0;32mexcept\u001b[0m \u001b[0merrors\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mOpError\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1329\u001b[0m \u001b[0mmessage\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mcompat\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mas_text\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0me\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmessage\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
- "\u001b[0;32m~/PycharmProjects/TensorPlay/venv/lib/python3.6/site-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36m_run_fn\u001b[0;34m(feed_dict, fetch_list, target_list, options, run_metadata)\u001b[0m\n\u001b[1;32m 1310\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_extend_graph\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1311\u001b[0m return self._call_tf_sessionrun(\n\u001b[0;32m-> 1312\u001b[0;31m options, feed_dict, fetch_list, target_list, run_metadata)\n\u001b[0m\u001b[1;32m 1313\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1314\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0m_prun_fn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mhandle\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfeed_dict\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfetch_list\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
- "\u001b[0;32m~/PycharmProjects/TensorPlay/venv/lib/python3.6/site-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36m_call_tf_sessionrun\u001b[0;34m(self, options, feed_dict, fetch_list, target_list, run_metadata)\u001b[0m\n\u001b[1;32m 1418\u001b[0m return tf_session.TF_Run(\n\u001b[1;32m 1419\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_session\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0moptions\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfeed_dict\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfetch_list\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtarget_list\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1420\u001b[0;31m status, run_metadata)\n\u001b[0m\u001b[1;32m 1421\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1422\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0m_call_tf_sessionprun\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mhandle\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfeed_dict\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfetch_list\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
- "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
- ]
+ "data": {
+ "text/plain": [
+ "<keras.callbacks.History at 0x7faf39bafd68>"
+ ]
+ },
+ "execution_count": 19,
+ "metadata": {},
+ "output_type": "execute_result"
}
],
"source": [
"autoencoder.fit(densities, densities,\n",
- " epochs = 2500,\n",
+ " epochs = 1200,\n",
" batch_size = 20,\n",
" shuffle = True,\n",
" validation_data = (vali_data, vali_data))"
@@ -5225,7 +2868,7 @@
},
{
"cell_type": "code",
- "execution_count": 23,
+ "execution_count": 20,
"metadata": {},
"outputs": [],
"source": [
@@ -5235,12 +2878,12 @@
},
{
"cell_type": "code",
- "execution_count": 24,
+ "execution_count": 21,
"metadata": {},
"outputs": [
{
"data": {
- "image/png": "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\n",
+ "image/png": "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\n",
"text/plain": [
"<Figure size 360x360 with 1 Axes>"
]
@@ -5259,12 +2902,12 @@
},
{
"cell_type": "code",
- "execution_count": 25,
+ "execution_count": 22,
"metadata": {},
"outputs": [
{
"data": {
- "image/png": "iVBORw0KGgoAAAANSUhEUgAAAj8AAAB8CAYAAACG/9HcAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nOx9WWyc53X28818s+/7cIaruIkSqcWSbMtObCWuncSBHSeOkzpJszhFgQJt0d60DdpepBe5alGgFwUC5KL5EyBOmroJ4i22492yLUsiKVEiRXEfcvZ93+e/YM7RUCYtiiIpkvoeYGBZpGa+Z97tvGd5jtBoNCBBggQJEiRIkHCnQHa7H0CCBAkSJEiQIGE7IRk/EiRIkCBBgoQ7CpLxI0GCBAkSJEi4oyAZPxIkSJAgQYKEOwqS8SNBggQJEiRIuKMgGT8SJEiQIEGChDsK4s38siAIu7ouvtFoCJ/0893OD0C00Wg4PukXdjvHG40hIHHcDZDW4u7nKM3TZex1jrudH9ZYi5LnZ29h/nY/gAQJEgBIa1GChJ2CVdeiZPxIkCBBggQJEu4oSMaPBAkSJEiQIOGOgmT8SJAgQYIECRLuKEjGjwQJEiRIkCDhjoJk/EiQIEGCBAkS7ihIxo8ECRIkSJAg4Y6CZPxIkCBBggQJuwiCcEP5oV0LQRD4tZW4KZHD7YZcLucvoV6vo16vo9HY7XpLyxAEAXK5HDLZsv1J/Or1+m1+ss2DIAhQKBSQyWRoNBpoNBqoVqt7iqNMJoNSqWSO9XqdOe6FuUrrT61Wr+BYqVRQq9V2PUdBECCTyaDVaplfrVZDuVxGtVrd9fyAZY6iKEKn0zHHarWKcrmMSqWyJ9ajTCaDKIowGAyQy+W8DovFIs/V3QyapwqFAiaTiTlWKhUUCgWUSqVdzZHOQ5VKBYvFArlcjmq1ikqlgnw+j2KxiGq1uqmfuSONH1qsJpMJoijyYt0LgwyAJ7HFYlnBL5/Po1AobPog3w7I5XIolUrY7XYoFApeqNlsFvl8HpVK5XY/4i2BFqtarYbb7YYoiqjX6yiXy8hkMshmsyiVSrf7MW8JZLxqNBp4vV4oFArUajUUi0Wk02mk02kUi8Xb/ZgbBhmuOp0ObW1tUCqVqFarKJVKSCQSSCaTyOfzt/sxbwl0oBiNRnR2dkKpVKJcLqNYLCKRSCAWiyGbzd7ux7wliKIIjUYDs9mMnp4eKBQKlEollEolRKNRRCIRpFKp2/2YtwQyXm02G/r6+qBUKlEsFlEsFhEKhRAKhZBIJG73Y24YSqUSBoMBTqcTAwMDUCgUbPSEQiEsLS0hGo1u6mfuOOOn2fo7cOAA1Go131JCoRD8fj/i8fjtfswNgxaqzWbD4cOHoVKpUKlUUC6XEQwGsbCwgEgkcrsf85agUCig1+vhcrlw9OhRqFQqlEol5jg7Owu/33+7H/OWoFQqYTKZ4PF4cOzYMT5USqUSAoEArl69Cp/Pt6s9B2q1GlarFW1tbThx4gSUSiVKpRKKxSICgQAmJiYwMzNz2zgKgnBLn63VamG329HV1YXjx4/zgUJjeOnSJVy5cuW28iNs9Bn0ej3cbjf27duH48ePQ6FQoFAooFKpIBAI4OLFixgbG7st3p/rwxob4SgIAkwmE1pbW9Hb24ujR48yR9pvRkdHcf78+dvKkebqRjnabDZ0dHSgv78fQ0NDK8YxHA5jeHgY77///rZzbA5PEbeb5SiTyeB0OtHd3Y3+/n4cOHAAoiiyIyASieD8+fN4++23N9XxIdzMg35Sjw+5XP6xME6tVruph5XL5Th48CBOnDiBoaEhaLVaCIKAarXK759MJvHP//zPKBQK635fwq30MKHwDfGr1WrsWl0vRFHEsWPHcPfdd2NgYIAXRK1Wg0KhgFwuR7FYxD/8wz+gUChsZKGcazQaxz/pF9biSLd8GkN6rlqttm6O9B733Xcf7rnnHnR0dHBopF6vQ61WQ6FQAAD+5m/+Bvl8/qY53kqvHUEQoFKpmGPzHF2vJ4re4zOf+QzuvfdeOByOFd5IrVbLPP/yL/8SuVxuWznKZDKo1WqIosjh4mq1yi7k9YDe49FHH8WJEyeg0Wh4IxIEAQaDAVqtFkajEc888wyy2eymc1yLn1wuh1ar5TFs5nYz/HQ6Hb785S/j6NGjkMvlSCQSqNVqkMlkMJvN0Ol0cDgc+Pa3v41MJrNta5E8ijqdjsewmV+5XF7Xh8vlchgMBvzpn/4p+vv7AQDhcBi1Wg1yuRw2mw0GgwFtbW14+umnkU6nt22eUohKr9czRwrBlcvldXNUKBQwm8341re+hba2NtRqNfj9ft5PnU4nDAYDent78dRTTyGVSm0rR6VSCaPRCFFc9jHQBZAuSet4XygUCtjtdnznO9+B1WpFpVKBz+dDrVaDUqmEx+OBwWDA0NAQvvzlLyOZTG75WqQ5ShdAil6Qt41e6+GnVqvhdDrx3e9+FzqdDrlcDgsLC6jValCpVGhra4PRaMThw4fx5JNPIpFIbNpa3BTPj1wuh16vh1KpZPc/eWtyudy6rFFa8CdPnsTRo0dRr9cxNjaGdDoNQRBgtVrhcDhw9OhRaLVaFIvFbbmR0UAbjUYolUrI5XI+LEul0rr5iaIIo9GIkydPoru7G6VSCRcuXEA2m4VcLofdbofT6cTJkye3lR9wbZFZLBbOX6G8h2KxiGw2u26OFosFJ0+eRFtbG1KpFEZHR5HL5SCXy+F0OuFyufDAAw9Aq9Vu1MDbEGgzstlsbMjSHC0Wi8hkMjc01AVBgFKphMPhwMmTJ2Gz2RAKhTAyMoJ8Pg9RFOF2u+FyufC5z30OWq12QwbeRkFeU4fDwUYmcczn8+vmqFKp0NLSguPHj0OtVmNhYQEXLlxAsViEXC6H1+uFw+HAY489Bo1GsyED72ZBoXC1Wg2Xy8UHCh0k+Xwe6XT6hvwov6etrQ2HDh1CtVrF5OQkLl26hFKpBIVCwfyefPJJaDSaDRl3G+WoUCig0+ngcrkgl8sBgA+TXC6HVCp1Q45k+Ozbtw9dXV3IZrO4evUqrly5glKpBJVKBY/HA5fLhb6+PqjV6o0aeDcNmUwGlUrFnmHaT2kd5nI5JJPJG3KktIi+vj643W5EIhFcvXoVU1NT/Dmtra1wOBwYGBiASqXacm4EuVwOjUYDo9EIh8PBl1zymmazWVSr1RtyVCgUsNlsOHDgAIxGIxYXFzE5OYlAIACZTAZBEOB2u+FwONgbtB2gEJzRaITdbocgCDyG+XweuVwO8Xj8hvxUKhWcTicOHToEURR5/BKJBOf7Liws8LlPa37TeNzyG/wxjNPS0gKlUslfBOXo0Ob7SV+ETCaDRqNBZ2cnuru7EQqFcPHiRYyMjEAURYiiCLlcjpaWFg4x3KrLez2gDVev16OlpYW/fLptUhLWevjp9Xr09PTA4/FgYWEBFy9exNTUFERR5IPZbrfj5MmTbElvB+iwMxgM8Hq9fNMkjw25Vm/EURRFmM1mDAwMwGQy4eLFi7h48SLC4TAUCgVPZJPJhFOnTkEul2+r4aNWq2GxWNDS0gIAnJRcr9c51LEejjabDYODg5DL5fjwww8xNjbGhg8ABAIBmEwmfP7zn9+WOUqgy4PVaoXb7eY5Sh48rVbLh+gnGbJKpZLDlfl8Hh9++CHGx8dXeD0TiQT0ej2eeOIJABsPydwMKJnVarXCbrczPzpIdTodj+Mn8VOr1fB6vTh27Bgbrj6fj9dhsVjE1atXEQqF8NWvfpU9vNsBpVIJs9kMu90Og8HAHh8yiOjCcCOOWq0WHR0dOH78OObm5jA6Oop4PM57aaVSwfz8PHsJtjPpWa1Ww2azwWazQa1W85rTarXQaDR8YbgRR71ej+7ubtx11124fPkyRkdHUSwWoVQq+QK3tLTEl9NSqbRta1Gr1cLpdHLibqlUQrVahVqt5lcul7shR5PJhP7+fhw5cgTnzp3D1atXUalUVnAMhUK8P2/XhdlgMMDtdvMcbeanUChWGNOfxM9qteLAgQMYHBzE6dOnsbS0hFqtBo1GA7VaDblcjlgsBplMxo6UzeR3S8aPwWCAx+OBw+FAIBBAIBBApVJhT5DVaoVCoUA8Hkc2m13Tnel2u3H//fejp6cHP/3pTzmkZTAYYDAYYLPZ4HA4YLfbEQwGEQwGt2WQzWYz2tvbYTAYMDs7i0QigWq1yjktdrsdHo+H+a3ldm9vb8cDDzwAl8uFH//4xyve32g0ruA3Pz+PUCi0bQvV4XCgq6sLcrkcV65cQTqdRr1eX5GA5vF4ODFytRCYIAjo7u7Ggw8+CJPJhJ/85CcAlo0Oq9X6MY4TExPbxlEQBHi9XnR0dKBareLSpUvscSJvnMvlQktLC4/jWhwPHDiABx98EAqFAj/96U+Zo91uh8lkgtVqhdPphM1mw7lz5xAOh7eNY1dXF9ra2pDL5TA8PMzGq1wuh8lkYo40jqsZeTKZDEePHsW9996Ler2OX/7yl3wBsNvtPI60sb/11luIRCJbzlEmk6Gvrw9OpxOJRALnz5/n55fJZDCZTOxxi8fjyOVyq/KTy+W4++67ceTIERQKBfzud79jw5jCQMTPZDLhxRdfRCwW21Juzc928OBBmEwmBAIBnDt3jg8OQRBgNpuZYywWQz6fX5WjKIq4//770d/fj2QyibNnz0IQBOj1elgsFt63XC4XtFot/ud//mfbEmVFUcShQ4egVqsxNzeHxcVFnjuCIPDlxOl0MsfVDk+FQoFTp06hra0N4XAYly5dgiiKsFqt0Ol00Ov1cDgc7CH8yU9+gmQyuS0cFQoF7rrrLjQaDUxOTiIaja7K0W63Ix6Po1AofIwjeZgfeughOBwOTE1NYWZmBmq1GiaTacXe7HQ6Ua1W8Z//+Z9Ip9Nbzk+lUmFoaAilUgkTExPsFaW8H7p8NfO7fn+gcNfDDz8Mg8GA8+fPIxQKQa/Xs4fObDbD4XDAZrOhWCzihz/8IXK53KZy2bDxQ0lYra2tKBaLmJubWzHIxWIR5XIZLpcLGo2G/3817Nu3Dz09PcjlcnzTIY+SQqFAo9GAQqFAJBLBz372s207UJxOJ9xuN2Kx2AqDiyqX6vU632CopHI19PX1sQEhl8v5BqZSqdBoNNiDNDs7i9dff31bPSItLS0wm81YWlpaES+uVCq8YVitVua4mmHQaDSwf/9+mM1mBAIBiKLIIVBRFFEqlZBOp6FSqbC0tITp6elt5ejxeKBWqzE7O7siDFWr1ZBKpSCXy2E2m6FSqT6xpJKqEObn56FQKKBSqdjVnEql2JN0+fLlbU0Elsvl8Hg8AACfz8ceH/I8pdNpKBQKGI1G5riWh6u/vx+CIGBmZgYqlYpvc+VyGeFwGIVCAel0GtFodNs4UjixVCphbm6ODTvacLPZLJLJJHQ6HSctfxI/AJiZmVmRn5XNZpFKpZBKpeD3+xEOhzE/P79tY0hh52w2i/n5eZZKoPBGLpdDJpOBSqX6RI6CIKC3txe1Wg1zc3PQ6XTQaDQQRRGxWAyRSASRSASTk5MIBoPw+/3btp+qVCrodDpEo9EV+ymVcedyOZ6rxHE1w0Aul6O7uxu5XA7z8/MwGo3QaDSQy+XMh3J/gsEgwuHwlvOjZ9NoNJDJZPD7/SsMHxpHCgvR3rFWtaQoimhpaUE6ncbi4iLvwcDy3AWA+fl55PN5hEKhbTHuKFcOAHvVaB0Sv0KhgHw+z/s/pVBcD6VSCYvFgmg0imQyCYfDAZVKhWq1iomJCQDL3r1kMoloNLollbMbNn5kMhncbjc0Gg1mZ2c/9jNyVdXrdZ6w14Ni00ePHkWj0cDVq1dhtVo5GRUAIpEIwuEwJicnUSwWt630VC6Xw+12QxAE+Hy+NfnR/6/Gj+LSQ0NDqFarmJmZgd1uX6G1EQwGOb+JYsLbBVEU+eawsLDAf9+sQUQG3VocadMeGBhAqVTC7OwsnE4nVwHk83nE43F2j1Ku1HZBoVDAarUinU6vGEdKuiSONKbXV6A0Jx12dXUhk8kgHA7D6XTyv0kkEohGo+x6vpkE6lsF3RK1Wi3C4TDi8TgbPcSRNDOI92oclUol3G433G43EokEMpkMXC4Xz4VAIIBwOLwidLYdkgx0aMpkMkSjURSLRX5+uVy+In+reRO+/j3UajU8Hg+MRiOi0SgqlQrPU6VSienpaYRCIb6k3chlv5mgsL8gCAgGg8yFvvtmjnS4Xs+R3qOtrQ2iKCISiXAVDb3H4uIigsEgr0Nge0KW9HxarRb1eh2BQADAtUoouuySdk2zvtv176HT6dDR0QEAbPAaDAa+aE1NTfGeut0aRhTxqFQqCIVCnM9E0iZ08aVk5dU40nv09PQAWL5UWSwWmM1m/o4uX76MZDLJSeLbeckyGo0oFAp8kSfvMoXhyFnRHCa//j0oVwtYvriYTCbYbDZ2Bly4cAGpVIpDelulmbYh44csXLvdDp/Ph0wmA51Ox25KstyVSiXS6TTHBK+HTCaD0WiEWq3m8m5KoHK5XLDZbAiHwxgfH8elS5c4v2arQW5ivV6PqakpjkmT4dV86yetk+v50cZlNpshl8sRCoX4kFIqlWhpaYHVaoXf78f4+DjGxsZuqprjVkHfvSiKGB8fh0qlQq1W49i4UqlcF0dyNzcaDUQiEV6kGo0G7e3tMBqN8Pl8uHz5MnPcLuOHvv9arYb5+XlYLBak02ne+FUqFXNMpVKr3qabjZ96vY5YLAaz2cyVJh0dHdBoNJifn8elS5dw+fLlbRWrpMqdYrEIv98Pq9WKeDzOZbVqtRoqlQqiKPI4Xs+RklAdDgcajQZSqRTMZjPUajUcDgc6OjogiiLm5uYwPj6O8fFxFAoF9nZu5eZLBnoul0M2m+U94Xp+crl8TX6kx+RyuQAAmUwGZrOZc/k6Ozt5jkxMTGBiYoJvsBupELpZKBQKuFwuvr27XC72jDQaDWg0Gr5JrzVPqQrO7XYDAHK5HEwmEywWC9rb29HR0YHPfOYzmJ+fx9WrVzE5OYlcLodcLodYLLbl85UqkxKJBNRqNZRKJYe+6/U674symQypVGrVfBjKv/R6vbxXUdiawr6nTp3C/Pw8ZmdnMTU1hWw2i0wmwwbRVo4lGdjJZBJGoxE6nQ6hUIirSqloBlg2asg50Azy0Ho8HvZwmkwmeL1edHd3w+Px4IEHHoDP54PP58Ps7CwymQySyeQKr+9WQKPRcATDbrejUqkgGo2u4EdaYGvlbZHmVEtLC1dx0z7a29sLs9mMT33qU/D7/fD7/VhYWEA6nUYikcD09PSmGrQbNn7oNlatVmE0GjlkRYqT5XKZrfG1ymy1Wi3a29uhUqmQz+f5dkCJfwcPHsTg4CAOHDgAAIhGo+zm20qhQzoMgOWbkdlsRiqVYn75fJ750a16NaOF+Mnlck6K1Wq10Gq1MJlMOHDgAPbv34+DBw9CqVQiEokgm81iaWlpy9VlKdeBStDJ+CGvBSXkNXtHVhtDuomRy5NuZwaDgW8wfX19GBoagslkQjgcRiqVwuzsLN+wt5ojLUxKfCSvRSaTQbFY5HLbtTjq9Xq0t7cDAIrFIiew6/V6Ti7t7e3FoUOH8MYbbyAYDCKVSmF8fPyGSY23Cjr0Go0GG2gkDlapVDiM0FwdtdpFhMqe6TuitUihodbWVvT09ODw4cM4ffo0e5lGRkZQKBS2bD1S5QzlvdTrdR63arWKZDK5osp0raR1g8GA1tZWTpSm743CXu3t7fB6vTh8+DDOnj2LWCyGWCyGDz74gPe2rQLtC5TUXa/X2QCo1WpIJBLMkapq1uLY0tLCa0qhUPD7ymQyeL1eOJ1OHDlyBBcvXkQikUAkEsHbb7/Nl9St5KjRaPi8oLEj70U8Hudq2rWKSEhmwe12o16vs2eT1McBcM7dkSNHMDk5iVQqhUgkgtdeew3xeHxLk4JpHOv1OoxGIyqVCo9jpVJBLBZjA484rhbWMxqNcDqdHDVRKBTsUanVarDZbLBYLOjr60M4HEYmk0EkEsFLL73EZ8hmcyQPKSUj05mdTqfZ4x2NRpkflfSvlu9jMpngdDp5fMkLRr9vsVhgNBqxb98+FoyNRCJ44YUXEAgEOMXgVrFh44cqeHQ6HXQ6HX8hmUwGqVSKN5lm3Z9m0GZL8elMJgPgmk4KHcIGgwH79+/H97//fUxPT8Pn8+HNN99ELBbbshARDTQAWCyWFQPfzI/CJdeLV9EEbm9vR09PD2q1GgqFAgRB4AOFjCidToeenh5873vfw9zcHJaWlvD73/8ekUhk0xO8rudIBix5pyhMlU6n2dgrl8trcjSZTOjo6MC+fft4vIgj5QgVCgVotVp0dnbiz/7sz+Dz+RAMBvHCCy9gcXFxS0tsmzVvaBwBcG4BLdxSqcQu9+s50q2EvAO06dJtPJvNcoK+1+vFY489hlQqhXA4DEEQMDc3h0QisWUGEBnq5ImikDHlsJARS0bbauNoNpvR1taG9vb2FYnE1PIhmUzCZDJBrVbDbrfjwQcfRK1WY8XV6elpRCKRLQmDkddGqVTCarUCWN5LaPxyuRwfaGvxI++H1+tdkaNAbvZ4PM77l8FgwIkTJyCKIoupjo+PIxAIbJlXljiSJ5IOOeJIXigKk6x2oFitVrS3t6O1tZWNGApDkFAcVZUqlUoMDg6ysSgIAsbGxjA/P78h/bT1gHIcqTqPqi1TqRRzpP2D9KnW4tjS0sKq1BSOLxaLCAaDLPMgCAI6OztX/PuRkRFMT0/zWbOZIC84afvQZYryCjOZzIpw4ydxbG1thdvtZk8gpRtks1kEAgFYrVZ2LFDSM3EdGRnB+Pj4RvVwPhHEj9ahwWBAo9FAMpnkgiZaIzfi53Q6EY1GOTxGuYnVapXPI7pIa7Va2Gw2AMCFCxcwOjq6KQUzGzZ+aBF5vV4uEaUHjsfjCIVCCIfDK5JoiahKpcLhw4cxNDSE3t5eRCIR3nQ7Ojpw6NAhpNNpyGQyzM7OIplMQqvVoqurC0eOHMEXv/hFDA8P46WXXuIww2aiOY/C4/FAJpNxKwqZTMYJ0NFodIVbnBaiSqXC0aNHMTg4iJ6eHgSDwRX8Dh8+jFQqBaVSCb/fz5PcZrOhvb0df/Inf4Lx8XG8+OKL+Oijj7bkVk0bo16v57JCg8HA+QWUlBiJRFZUERBHjUaD48ePY3BwEPv27eMEQ2C5uu3IkSM8buFwGNPT0ygUCjCZTOjp6cE//dM/IRQK4eWXX8brr7++JQcnzTWj0Qi9Xg8AnBRLZaLhcPhjtyUS1NRoNCy42dXVhcXFRfaOtbW14ejRo0gmkzAYDIjH41haWkIikYBWq4XVasVf//Vfo1Qq4Q9/+AN+85vfbMnhSQea2WzmEAIZZjKZDMFgEKFQCNFodEWpKOUfaLVa3H333Th06BDa29uxuLjIFxev14u77roLyWQSVqsViUQCgUAAoVAIoihCoVDgz//8zyGTyfDee+/h//2//7fpN2viRxsu5b00lzOHw2GuDmo2bkiP5N5778Xg4CDa2to4NFCtVlmdO5lMwul0stG/uLjIe8pTTz0FnU6H8+fP47/+67/WrEC6VY6UmwYs6/oQR0EQsLS0hFAoxNUzwLWwOoWCTp48icHBQXg8HhbBy+fzcLlcOH78OJLJJLxeL19uKL+pUqng4YcfxmOPPYbFxUX827/927r0km4GNB5kGJC8BGnDAcDCwgLC4TASicSqeV16vR733XcfBgcH4Xa7sbCwAJ/Ph0qlArvdjrvvvhvJZBIdHR2siUSGTi6Xw1133YX77rsPhUIBP/rRjxCLxTZ9z6FnNZlMfAHp6elhjlTJm0wmUSqVVnCkCq77778fg4ODnI+2tLSETCaDgwcP4p577kEqlUJ3dzdfqK9evcqtWDo7O3H48GHIZDL867/+K4LB4KbuOTRPtVotqtUq0uk0t9poNBqYmZnhViLN/GivMBqNuP/++3Hw4EE4nU4AYC95f38/7r77bv4zsOylnpqa4iR9s9mMJ554Ak8++ST+/d//HQsLC7e032w44bler3MuQS6Xw9TUFLuS29raOE4/PDy8QvCIdACogqRSqcDj8fBEpdgwWYcUBjp9+jQmJiYgl8tx9OhR3HPPPZxsvRVCcrVajcNaqVQKk5OTvAhbW1vhcrmQy+Vw/vx5JJNJ9nKRh4iqnMrlMrxeL+bm5pDJZCCXy+FyueDxeODxeFAoFBAIBPD+++9jYWEB5XIZg4OD6O7uxlNPPYXx8XFkMplN33DphkwbbCKRwNWrV6FWq6HX69HW1ga3241MJoPh4WH2dpFxdz3Hjo4O+Hw+NgbtdjtaWlrQ3t6OSqWCYDCIDz74AIFAAHNzc+ju7obb7cbXvvY1nD9/nhV2N5sjJRdSn5/Z2VkWIPN6vfB6vUin0xgZGUEymeSES7VaDa1WyxUZ5XIZXV1dbOBUKhVYLBZuHdBoNBAOh/Hhhx9icXERU1NTcDqdaG9vx5NPPon33nsP4XB408MnNI70nNQHp5mjx+NBJpPByMgIu4wpXNBcIVWpVNhYj0ajKBQKXFLb3d0NpVKJeDyOM2fOYH5+HouLi0gmk9i3bx9OnTqF1157DYuLi5saziSDTqlUIpfLwe/3IxKJcC+njo4OtLW18RiSVAPtM6QiTPz6+voQjUY5/KpWq9HT04Pu7m4YjUakUil89NFHmJ+fx9zcHC5fvozOzk4MDQ2hs7MTMzMzmy7OSWNIeVnNeZRmsxmdnZ1ob29HMpnE6Ogo7wfUd02v13O7g1qthoGBASSTSQQCAUSjUa6OIo6VSgXnzp3DwsICpqenceHCBbS1tcHj8aC/vx+XL19et7DpzXCktZXJZDA/P49SqQSdTgeLxYKenh50dnYyR/p8uoTSxYy04w4cOIBcLoelpSUEAgFUq1V0dnait7eXQ4ctLS1YWFjAzMwMLly4wPldhw4d4n17M/ecZmmCZDLJhUAk+9Lb24uuri4kEgkWuKW0A41GwxxJxqC/vx+FQgE+nw/z8/NswHd1dfFljjjOzs7igw8+gNFoRHd3N06cOIHTp09vqkeWPOP1eh3RaBTz8/NstFmtVlnUZJ4AACAASURBVAwMDKC7uxuxWAxjY2N82aK9lMr08/k8qtUq9u/fj3K5jIWFBUxNTaG3txcdHR3wer2s4k78FhYW8PrrryMUCqG7uxv33XcfSqUS/H7/hg28DRk/VKlEMeh4PI7Z2Vm+jSSTSZjNZlitVpw6dQqFQgGXLl1iXYdyuYxAIACn08nl8O3t7ZidncWZM2fQ29uLY8eOwWg0wmq1ckLj7OwsxsfH8dZbb+HYsWM4fPgwvF4vL6TNAvGjAyUWi3GlkEKhQCaT4QF/6KGHkM1mcfnyZfj9fr7VLC4uwmazoVQqsZbOzMwMzp8/D4/Hg/vvv589BFTCuLS0hMuXL+Ptt9/mm3hXVxcmJye3ZMOlnJ5isYhwOAy/38/WPWXhWywWfP7zn0c8HsfExAR7eGhRms1mFAoF2O129PT0YHp6GqOjo7BYLDh16hSHnNRqNVdBXLlyBe+99x73qdm/f/+KzWCzQBUfoigim82yRhTd0PL5POufPProo1xVSJtpPp/H3NwcDAYDXC4XrFYr9u/fj6tXr2J4eBhqtRqPPPII6vU6LBYLNBoN8vk8WlpaMDMzgzfeeAPhcJg9K2fOnNmSDZeUidPp9AqOVNZuMBhgNpvxxS9+EaFQCJOTkwiFQqhWq8hms5ienmbXstlsxoEDBzA1NYWxsTE0Gg187nOfQ7lc5kTMnp4e1m168cUXEQwGce+99+LEiRPI5/ObvuGSdlihUGDDTC6Xc/WXXq+H2WzGY489Br/fj6mpKa4UyWQybNSbzWaYTCYcPHgQGo0GV65cQbFYxBe+8AW0tLRAr9fDZDKhra0NBoMBdrsdv/71r+H3+3HPPffg/vvvRzabvaUNdzVQHo9MJkOhUEA4HEY6nebydDISzGYzHn/8cSwtLWFmZoZv9qlUCleuXGHvrcFgwODgIPR6PWZnZ/Hss8/i4YcfhsPh4BCiy+XiSs3JyUmMjIywjlUymcTc3NympRXQfkphrWw2i3A4jGKxCIVCgUQigXK5DI1GA5PJhC996Uvw+XyYm5tDMBjkBO/x8XEAy4nFBoMBBw8ehNFoxOTkJH7+85/jwQcfhNls5oIVq9WKarUKg8GAsbExXL16FYVCAadOnUIikcD4+PimphY0c8xkMuxdSiaTHGK/nuP8/DxzrFarGBsb47NHq9XiwIEDMJvNGBsbw89//nOcOHECer2e1yqdkUqlEqOjo/D7/chkMnjggQc4R3azmrqSx1QUReYnCAJLRJB6uN1uxxNPPIH5+Xn4fD6EQiFuZk1ilJSS0NfXB7vdjvPnz+O5555DT08PVCoVrFYr76k6nQ6tra2wWCzw+Xw4e/YsHnjgAfbYb7Th6YZ6e1Fex1e/+lWOjb/wwgvscvZ4PCybPjIywvkgn/3sZ/GFL3wBMpkMZ8+e5dYOHo8H+/btgyAIePvttznu94Mf/AA2mw3xeBwvv/wyl9b9+te/RiaTwcDAANrb2/GLX/yCK1A+CY119jChxfO1r30N9XodkUgEr7zyCvNrbW1Ff38/VCoVzp07x3HzT33qU/jiF78IURRx/vx55HI5yGQyOBwO7Nu3D1qtFu+++y4nd//d3/0dOjo6UCwW8eKLL/JzkLia2+3GPffcg1/84hcYGRlZT/fldfcTorykp556CrVaDUtLS3jzzTeZY3t7Ow4cOACFQoEPP/wQAJDP5/Hggw/i8ccfh0qlwvDwMI+h2WxGV1cXrFYrzpw5w3ku3/zmN3Hw4EHI5XK8+uqr7NJ+8803sbi4CK1Wiy996Ut49tln8cEHH9xQr+JGY9jMkcKVX/7yl1Gr1TA7O4vTp08zx66uLhw4cAD1eh0XL17kkMbJkyfxxBNPQKPRYGRkBLlcjkMo7e3tcLlcuHTpEkKhEMrlMj7zmc/g+PHjMJvNeOedd1g76IMPPsDU1BTq9TqefvppPP/88ywMuJkc7XY7Hn/8cZTLZUxOTuL8+fOcA9PV1YXBwUHUajUMDw/zBn38+HF85StfgU6nw8WLF5kjVax4vV4sLCyw233//v2499570draipGREUQiEQiCwBVuyWQS3/nOd/DOO+/g1VdfxdLS0i1xJH7kKX300UdRKBS4MpK4k0aYQqHA2bNn2d1+6NAhfPWrX4XRaFyhwq1UKuF0Otnj5/P5EI/HYTabceLECfT19THvRqOBubk5XLhwAbFYDE8//TQuXLiA559/HnNzczcaonWvRVEU4fV68fDDDyOTybD6O/2sq6sLfX19EEURH374IWq1GorFIg4dOoSvf/3rsFgsfJATR7PZjNbWVi4tpyIRKiLJ5/Pw+Xwol8uIRCIYGRlBKBTC17/+dUxNTeG3v/0t661sdAybOSoUCnR0dODUqVOIxWK4cOHCigtlV1cX9u/fj1qtho8++oirevr7+/H000/DbrfjypUryOfznMCv1+vh8Xig0WgQi8XYKLVarRgaGoJer2cvWi6Xw+joKGZnZ/G1r30NwWAQ//d//4fR0dFPvFTeDEeVSsVeCVIPD4VCAJaTevft28ccP/jgA/ZMd3V14Rvf+AZcLhdXqFGCO1UpGgwG9uxGo1FotVr2lFCz71qthosXL+LSpUt48sknkc/n8dxzz+H999+/JY7ET61WY//+/Th69Cj8fj+Gh4d5v1apVNxORRRFnD17ls99j8eDb33rW/B6vZiZmUEul+OKcLlczqrt5EhZXFzkBP2DBw8iGo0iGo1CEASMj49jeHgYjz76KORyOX7zm9/gjTfeuNGlefN6e5H3JhwOw+PxwGazwePxIBAIoF6vIxwOw2w2o7u7G62trcjn88hms3jppZfw29/+FtVqFT/84Q9ZwZIS1kRRxKc//WkAQDwex3PPPcf5N9lslsNsWq0WmUwGs7OzOHLkCFpbWzc1WY+8IpFIBDabjdVxqfw0EAiwe9Hr9aJQKCCXy+G1117Diy++iEqlgn/8x3/EyZMn4XQ62V0rk8nwV3/1VxCE5RYg77//Pt59910A4BsDuQij0Sh8Ph8+/elPo7W1FVevXl2P8XPTHFOpFIdH3G43L9ZgMAiTyYTOzk60traiUChApVLhjTfewO9//3tUKhX87d/+Le677z643W7O/m80GvjGN77BFRhnz57lWylVGlGCvEwm4yoTr9cLg8GwqWJdVPlUKBSg0WjQ1dXFybkAsLi4CIPBgM7OTk6izGazeOutt/Daa6+hUqngL/7iL3DkyBEWSiT14EceeYQVSScmJvDGG2+wR6dYLEKn0/F8peTxlpYWGI3GGxo/N4NGo7FCHLSvrw+Li4ucTOj3+6HX69HZ2Qmv18vVE++99x7efPNNVCoVfPOb38Rdd93F7uZcLod8Po/Dhw/jgQcegF6vh9/vx8TEBM6ePQtRFJHP51lYjvLE6vU6XC4XTCbTDY2f9YJyJ6jke2BggPNfSNSumV86nUY2m8WHH36Id999F5VKBV/5yldw7NgxeL1e6PV67pHlcrmwf/9+mEwmFAoFzM7O4tVXX2XPJ+VqUPkuiZqSd2GzPLGU3Fyr1WAwGHDo0CEkk0k2ogOBAPR6PTo6OuDxeHieDg8P48yZM6hUKvj85z+P48ePo62tjTlGIhE2BB588EGoVCr4fD6cO3eOixtoL6I0hHK5DLPZvCUcyetps9lw9OhRFswURRFLS0tcVdna2soGy9jYGP7+7/8e1WoVn/70p3HixAl4PB7o9Xrk83n+bpxOJ/bv3w+Hw4FQKISZmRlud0JjJ5fLudDEaDRyEcRmcaxWq6yz5XQ6cezYMZw5cwbZbBaiKMLv93MFMIVq8/k8Ll26hB/84AeoVqvc+JpUxnO5HBYXF7lR68DAADo6OpDP5zE9PY2PPvqIE9ebmxmTqOxmcqTSdnJYiKKI4eFhNkgDgQBfEKnzAbWL+Zd/+RdUq1UMDAzg7rvvRmdnJ2sGLSwsYH5+HjabDS0tLXjyyScBLO/PH374IWQyGUqlEkdIGo0GEokEjEYjz9ONYMM5P+QtIMntzs5OqNVqJJNJzp0AsOLBqKKoVqvhhz/8IVea2O12do3mcjl26333u9+F1WpdofxpMBg46TSXy6FcLnN5/GaiWq1icXGRY5r79u2DXq9HIpFYkUlPmfjkTaDv5kc/+tGKSiOz2cxhBtrknn76ac7SpwPRaDRyeXSjsaz+TK7qzUalUsHCwgK6urqg0WjQ29vLeQ+xWAy1Wo1DmZSXQAupVqvhP/7jP9iDZLPZuMSzuUz+4YcfxtDQEDQaDcLhMJd3Uly7WCyy+3urOM7Pz6OjowNKpRL79++HzWZDMpnk1gzN40YGDJXD//jHP+YEaDoUmvOIarUa9u/fj+PHj6O9vZ0TiynkQlUthUKBE/82E2TELiwsoK2tjdskkHIqqds2j1tz7L5areK///u/8atf/YrnKgA27EijS6vV4q677sLBgwf5INPpdFyuTL13mitBN2PDbTSW+8stLS2xqOqhQ4cQi8WQTCYRCoVW7DGksUSvarWKX/3qV3juuedgMBhWaD8ZDIYVYqWU4O1wOBCLxaBSqbiqtLm30GbP00ajgXw+j2g0CqfTCZ1Oh8OHDyORSCCdTiMQCDBHKpOmuUp9FJ9//nm88sorfJNuNBrc84wORApLDA0N4fDhw3xQZ7NZ1mWhOUsH6WYZBmTEUkqE2WzmZHrKTyLxRuJYqVSYa6VSwRtvvIE333wTOp0ONpuNzxK1Wr0ikVomk8HlcuGRRx7hCiJK0C2Xy6yUvdmNMsnooAul3W7H8ePHkU6nkUwm4ff7+Xsljs3SG5VKBadPn8bZs2e5bYvFYuGcvkajwR4gyvn60pe+xH0mAfB6bK6G26hxsBq/ZDKJXC4HrVbL/S6pspQ8NgR6rmYJldHRUVy4cAFarRYWiwUymYw5URWgWq3msPlDDz3EPe2o2IgS2lUq1apCkevFLSU8z8zMIJ/Pw+Px8Jeh1+vR39+/QjAtHA7zbZIWk91uh91uX1G50bzABUHAz372My4TpIZnzU0NdTodew42O1m2Xq/j6tWrSKfTH+NHcUqyQCn22Jx57na7YbVaWReBjAiHw8G35Ndff51jsnQ4UkkfJaXS+25FNRTF0ROJBFpaWri6zWg0oqenhzmSLD4ZNcSR9CaIIyWIU6knAO6YTXHhZnFFAJzkSdpQm41qtYqLFy8iGo2ipaUFarUabrebGyOS0RaLxXieNifs0kat0Wg4yZtexLFQKOCdd95BNptFKBTiUniNRsPPkc1m+b23guPw8DACgQCPo9PphF6v51AkJSmGw2GOvxNHk8kEk8n0MY5k+NNmPTs7i8nJSc5tUyqVXCkoCAILEW52flqlUsGZM2e4t5VOp4PD4YBOp0NnZycbM1S5RwmV9Ax6vZ7FVKllDpVdA9cqGKvVKiflkydWp9NxFWs+n2ePxGajVCrh/fffZ45arRZ2u52LD0wmE2q1GkKhEHMkETxgWSKEKjeJIzXRpAPJZDJBJpNhcXGRb+wAeO1S/l+hUNgSrZhisYh3332X+3dpNBpYLBbodDrOsyqXy/D7/YjFYpzATc/RnKBPRrbJZOLk33q9Dr1ez7IVv/zlL7lXn1KpZKOc5C3S6fSm51Hm83m8/fbb3G2dCg+o+TdxXFpaYo7NEhtUMEMis3TOmc1mAGDNMpqTL7/8Mht1VHBDumyZTGZTS97Jy/zWW29x/y7KyaHwHDUYpnAy7QXN2lMkzVGr1dh50NraCgC8bnU6HWQyGc6dO4dkMskGH4XKSB7nVmREbsn0JZlrURTR0dHB+i6kVUGtAGhwmgehWq0ikUiwp4fKyMkzQMrBTqcTVqsVPp+PE4ntdvvyw//ROIhEIlsi0FUoFLgbckdHB5ejKpVK/vJJZGo1fslkkpNqKYbbLDJHzSJJBZkSj0klmfhRWepWgG6cMpkMnZ2dnARJ+iDpdBqRSASZTOZjar7VapUNN4PBwAuW9IGaE4FJXZdufyaTiQ2gfD6PYDC4ZRojFAIAwDFpUmmmHmahUGjV5rSk00Fl+nSgkMeqWq1Cr9fDYDDwAUUHJGkpCYLAN/it0m66nmOzdEG5XGaOuVzuY4Y0hV0KhQKPF81VOgxJMoDGkjYk2oSpwmVpaWlTw7OETCbDlwbiR/tGqVRCIpFg4/X6i1CzMGJzyw/yFNDzGo1GFtEj7RJyt1OI1u/3b/qhSSBJCapcImFASvYmjqs1NaWDiTyuzX3ZyJgpFAqc1E2l/aR4TmuRKuqaJUq2gmO5XGaOdKDRXhSJRFZt+EnGRblc5j5hZPBRakUikeBKQIvFApVKxWrK5HWlc2urmtbSgVwsFpkjzblMJoN4PM4cV/uOaV8hDT1ak8ViEblcDuFwmPcwm83GCvXNodpyuYxYLLaiv9hmgHSxSEqBzkWSsaG+f1SMsNpn09lgNBq5IbTBYGDVeEqfoQILi8XCKtaEer3OjodtL3UHljdNuiGQ2i0NEkmKU/jkepB7lW4lZJlTiIuEvahxqNls5olPk12hUGBqagqBQGBLjANSAaYbBd2q6LAIBoOIx+OrWp7ZbJZvxSQBQG508gJks1nY7XauxiEvD8WzVSoVCx9ulWGwGke1Ws03I8qvWG2C0UHarOOg1WrZNZpKpZBMJmG322GxWNiLQAYhHap+vx8+n2/L+rY1c6RGjyqVCoVCgXvIrNXZmtyr5OWgtgjUSoGqDWw2G9/UgWsJ5RRqCIVCmJ+f3xLDAADf9IgjhUpJhIyM9NXmKoUdKd+MFFbJGKQqPZPJBIfDAZPJBIPBAGBZ6IyqsRKJBMvtbwU/CmHQJUupVLIrnvitNk+pGo7KwmnNUXiS8vhI78flckGv13MZMn1n8Xgc09PTW9Y9u1wuM0dah6txXOvf0j5jNBrhcDhgNptZXK/RWG4/Qzo67e3trLhMe3C1WkU0GsX09PSmVQit9pyUA0cXZbpoURPLtaRLyEgjT4jL5YLZbGZjlITyEonEigtZc+sMOrynpqa2RAiwmSMZac0cqQ/gWnsdyYloNBq+/BsMBjbEqQqV+lxaLBbmCICN2EQigfn5ecRisU3nWCqVWKuI9P6ok0MikWCP1mogbw9FcZxOJ4xGI1fEUZ4WhWIdDgfn+VJeI7DcHoQ0vjbKb0PVXquBPDeNRmNdWibkmtdqtejv74dSqUQ4HMbly5dXxHrJOqTDipIOybiYm5tb0a/pk7DerPa1+FGYYz2fpdFoOH7f398PtVqNYDCIixcvch4I9Qyj8Bi5ee12O8elFxYWWGNnHWO17gqTzeY4MDAAURQRDoe5wg9YnuxUukgGbLVahdPp5NYEpBeznj5RN1N9sdkcDx48CFEUEQwGce7cOW5aSDlAVLVAYSWXy8Wy/UtLSxyP32kcm0MsVJkXDAY5mZLWNeUEud1u3qTcbjd7RwOBAOLx+KZwXA8/yhO8GX5DQ0MQRRE+nw8fffRR8+ex2nlrayvn+TidTiQSCQiCwC09mkO/n4BbWosUSgSwLo46nY5v0ENDQ5DJZPD5fCvWIuVHGI1GdHR0oFqtskwFGTvRaJTzKm4UTrjVeXqzHEmWQqfT4ejRoyx/Mjw8zONBHHU6Hfbt28f5hTabjTnG43H2TOw0jnQRJo4AMDs7i7GxsRXjCCzPa2qASn3v6IJDRtZqHrSb5biZ/OjipNfrcfz48vKYmJjAlStXVvx7Kv7p6+uDTCZDNpuF2Wxmzy4ZyutsrbN51V6r4WZzUihPRKlUQqfTIZvNrhDJA8C3LsoHqdfrLHxISWTkQdpq3OxnkBeMPAYUB22OpVNIhNycVAlBzfia+W3FDeV63CzHZg8DGW8UvqMFRxpNFF4idyZxpLDSdnUn3ghH8vwolUqUy2UuEW4eR7r5UEULzWlKcN/pHGmeNs/V5sRh4k95fJVKBRqNhtcibUjbwfFm+VHYmcJAlCBOicCUxEy5JCRaqlQqUSgUOORLejTbMYY3m8NIQofNfZdIqZqetznERKkIFDYkjmTcbVU7lmZshCN5t6jikMrlmzlS8QS1XBAEgYtNqMfWTuVISfk6nY7THmgtAuBLCI01caLkb/IQUaL+VnPc6BhSIn46nV6hqE7J3bReKdoDgJPzm/ndSq7v5qa73wToUCR3HR0UzVVFzb1nBGG5k7zNZsPs7CzfvrbLMLhZUFkeJUxSiTFwrTKMvGSU66RUKuFwODA8PMwTd7sOzI2Acrso1JNOpz8WeqAxIjVv0m25cOEC6+rsZI5kgOv1eu5z1Rx6aK6iol4+ANDd3Y1Lly4hl8vxPN6OzXYjUKvVaDQavBbj8TiHdq6vpKLwmiAI2LdvHyYmJliccqdypOoRrVYLQRBWhB2IX/M4ZrNZCIKAjo4OTE1NsWr0TuUHLK9FmqcU4iIe5LkjjlSVJJPJ0NbWhtnZWea4XUbBRkD7DYVcQ6EQCoUCex+aKzfr9Tp7erq7u+Hz+fhicquH5laCcnaII6V0KBQKvohQjh2F+QCw+nwzx504jhS2MhqNAJbL2SuVCod3KSxG45hMJnktNqfRbEYz5dtm/JCkPjVTJMOguZafMsErlQpOnDgBnU6HWCy2Igy0Uw9NSjSlDsZ0YFIFUKPR4Fg7ic7p9XrE43Fu9UAb1U4FcTQYDBzvpdJFevbm8MuxY8dgMBiQSCQQj8d5DHcyR/LoUJVGIpGAXC7nElRagPTnQ4cOcQw7Go3uCo4kZUAlp1TmTUZR81prNBoYGhqCwWDgjtlkvO5UjpQHQbl0iUSCVYKpVJzGTxAEDA4OsuQDKUXvZH7ANY5arZYTf6ncmgw34iiTyXDw4EEYDAbO6yNRwZ3MkQ5FjUbDnn+LxcK9pppfMpkMAwMD0Ol0yGQy8Pv9PI471fABrkmnqNVqJBIJZLNZrjImbpSHJwgCent7uep5aWlpx3OkMVSpVIhGoyiVSujo6GCpArokUwFTV1cXz+mzZ8+umMe3ittm/ADX+r2Q94OSgqkSQ6VSwe12s35JOBxGOBxmV+5ONXyaUavVWHWWqhNEUUS9XufKmZaWFgBYwa/ZXb2TQSXC1BmdkgypHFGv18PlcsHhcHBCJanN7gaOdMtq5mg0GqHT6TiJkfpfWSwW7iG22zhSeSx5IKnEn4w/qiyhROFYLMYtMnY6R/IcN/Oz2+2w2Wzcooe0uChhn/jtljEkKQ1KCqbiDKfTCVEU0Wg0ePwUCgXnvQSDQQ6573SQJAGtQ8pzIr0ZYFl+Q6fTAQA3vKV2NbuBoyAIHIqkikaDwQCPx8NVVVQE1Gg0WNB0t6xFqg6lKkrSMvJ6vZw4bbfbuRJ3dnYW0WiU5Sc2k99tNX5IG4QaSFKViclkgs1mY9dYtVrFlStXWNhspw8wgSqbqHTYYrFwtQxVPzUayyJ1ExMTiMfju4ofGQb5fJ7H0G63cxUGldWSG/bSpUu7jiMdnDSOKpUKTqcTbW1tbASRd69YLGJ0dJSFMHcLRzJ+SAmbmiySCivJTlQqFRQKBa6U2S0cyTAoFovQarWcj0fCrGq1GjabjdV/Sfuq2QO700GHSrFYZO0VSmymsnCz2cwcx8bGdiVHMmIFQYDBYIDFYmH1da1Wy2KyhUIBIyMju44jFQ4RRyoFdzqdPI7Neyo1Wqby853suQOuFStUKhXW36KzXqVS8ZnRaDS2nJ/sxr+yNSCBMdJ9EUURTqcTra2t0Ol0qNVqcLvduPfee/H444/jmWeeQa1WuyVRo+0ECRtSQqhKpeLmmIIgoFAowGq14tChQ/jCF76AZ555huXRdwM/AKx/QjoiVL6o1Wq5DFyn06Gvrw+f/exn8f3vf58TDncTRxJ/y+VyXIJK5ccUk/Z4PLjnnnvwzDPPcHPP3cSRbmNkAFFSeiAQQCqVYj2uoaEhfPe73+XmlLuBI41hMz+1Ws03yng8ziGDzs5OfPvb30YqlUIwGNwV/ICVHJu1iZaWlrCwsIBAIIDLly9zy4tvfOMb7PnZTRwpXYIkJOr1Oubm5jAzM4PFxUWcPn0ak5OTEEURX//61xEOh1k3ZjeAkvJJiJGaa1+9ehVTU1OYm5vDiy++iI8++gj5fB5PPvkkAoEAlpaWdgVHkrQhDTFKvr9y5QqmpqYwPT2NZ599Fn/4wx8QCoXw+OOPw+fzbQm/2+L5IaExupEVCgXodDoWzWs0GjAajRgZGcEf/vAHtLS0oL+/n12aa4kn7RSQe5ZexI90PIBlC/jdd99FPB5He3s7ent7YTKZ2HDYyfyAa94C0s+gw5ES7WQyGVQqFV5++WUkk0n09PSgu7sbBoOBtaB2A8fmcczlcizGRfpGarUa//u//4t8Po/9+/ejs7OT1ZLXWRJ9W0HjSIdlNpuFwWDgKguFQgGdTodnn30WpVIJBw8eZDVe0oPayRxpDGmvaeZHvKvVKn7xi19wzhb1jlKr1dtW3XUrIA8shZszmQyHJ+VyOfN44YUXIIoient7WV9FpVLtGo4AuKkpcaSUCWBZX+d3v/sdl0hTmI+EV3cDR6p2kslkXDxSKpV4jMvlMt566y3o9XpMTEzAbDav4LiTcT2/dDrNZwatU7VajeHhYahUKkxPT7P6PFWSbiZuW9ir0WhwWWY8HgewnJBICqVqtRq/+c1vUC6Xcfz4cYyOjmJmZmbLROI2G1Q+q1AoWJmSDDdSQn7vvfdYpn1sbAxzc3NbJvS3FSAROGpRASxzpFuZKIoYHR1lF/X4+Djm5ua2TK16K1CtVqHRaNgwAMBCl6S+Ozs7C5lMBofDgcnJSczPz2+J4vhWYS2O1WqVE9YjkQhXu83OzmJ+fn7Hb7bAtd5nVFpLB4pGo2GDCABXDaVSKfh8PiwsLGyLhMZm4HqOVH1HBi0dLLlcjsujg8Eg5ufnd4W3ALjW442qoIgjGXzNY9n4oxJ0PB7H3NzcbXzqm0PjjyrdlO5Bmmg0joIgwGazoVarsedkcXERs7OzBQjXCQAAIABJREFUHJ7fyQZeo9HgywfNR+JHz05edcoFjsVimJ+f3xJ+myZyeLOg5qVUdUE5E6TJ4XK58NFHHyESiaBWq+H48ePo7OyEXq/nEttXX30VgUBg3Z95K2JONwsSLyRVYCr3Jo0Kp9OJkZERhMNhvnFSjkVvby8mJibwyiuvIBgM3szH3pKw2s1CrVazhDz1XKtUKtyXxWaz4dKlS5yMNzAwwL2YDhw4gPHxcbzyyiubOobA5nJUqVQwm80sQ0BNCLVaLYviTU1NcTVJd3c3urq6YLPZMDQ0hPHxcfz+97/f0RypR5JSqYTT6QQAltcHwO1XSN+ora2N+6LdddddGB8fx8svv7xj16JcLud2Km63mzdRjUaDRmO5j1AkEuHLh8fjQXd3N5xOJ06cOIHx8XG89NJLO3otUpWlKIpoaWlhbwhxJN2b6elpFv3r6emBy+XCfffdh8uXL980x+2ep4IgsE6M2+3m/BiSoyAdritXrrCKfG9vL1paWvDAAw/sCo7A8tkhl8s5kZuKger1Onuhx8bGkMvloFQq0dvbC6/Xi4ceegjj4+N44YUXNpXjZvMjY87tdkOhULBm0R+fBTqdDqOjo8hms1CpVGhvb0d7ezseeeQRjI+P4/nnn0coFLqZj9xakcObBQlQAeBGk1RdUiwW4ff74fF4UK1WEYlEcPbsWVy4cAEajQbf//73b6mb63aguXkkNZm0WCyQy+XcvM/j8aBcLiMSiWB0dBTj4+PQarX43ve+B+Caq3engnqR0a2MOFLH4lAoxGq5kUgE4+PjmJqagk6ng9vtXqHLsVNRKpU4fyefz8NgMMBqtQK41g6gtbWVe8xNT0/D5/NBrVbD6/UCwI7nWKlUEI1GAYBDe3TDBMCNb6mXjs/nQygUglarRXt7+44fx1qtxt5lOhRtNhuHE9LpNNxuN/dbonYnOp0OXV1dAHb+WmzWtcnn89z5nPokkqq63+9HLpdDPB7H8PAwdDodent7Aex8jo1GgyVDiKPVauU2EvV6nZuJ5nI55HI5XL58GQsLCxgYGACw8zkCYO8/FZJYrVYUi0X2sjscjhUCh1NTUwiFQhgcHASw8zmS539mZobb6TQLdJKobKPRYFHZeDyOw4cPA7gmB3CruK3VXmT8kNqq0Wjkni1Go5Hd6s3iXPV6HadPn0alUtmy/jObBXp+v9/Ppe7UH8lms7G+yGr8KP9pp4MOyFAohHA4zA0wnU4nd4UnvR+qDKrX63jvvfd2DUcKDcRiMcTjcSwuLjJHp9PJ4lzEkX7/nXfeYXXn3YJUKsV9c6iE3+l0soAjNRel19tvv41Go7Gi6eBOBompkvHmcDjgdDo5j6t5DHcjP2A5hEfNO8ljSY1MKUem2eP/1ltvAcCu4lgul1l3i8qjqRUJdTgnOZR6vY4333wTgiCs2cNvJ4KaJJOoqM1mY46UJwNc0xh74403uMfebkE+n0c+n+fKNqfTiUgksoIfrcXXX399U/ndtrDXemCxWFh5lNAsDnizWj/b7d67EZqb8hGa+W0gHr+trvb1wGQyrar6vFGOt8MNfSNQZRRxIWN2L3EkgcDt4rjd/HQ63aoNNffSWtRqtat2Eicv+l6Yp2sVU+wljmsVGmwVx+3mt1YS/kb5YaeFvdaDXC63qoGzW5L0boRcLgcAe5YfcM2Fu9c5Xm/cAXuL4/WH5l7juFYS/l7hB6zNcbcIxq4HdwLHtYop9gpHEgW+HpvNb0cbPxQW26vYDdUyt4o7geNuqQq6FewGddxbwV7nB+wtQ24t7IXD/0aQOG4OdnZmlAQJEiRIkCBBwiZDMn4kSJAgQYIECXcUbjbsFQUwvxUPsg3oWMfv7GZ+wN7nuB5+gMRxp2Ovz1Ng73OU5uk17HWOu5kfsAbHm6r2kiBBggQJEiRI2O2Qwl4SJEiQIEGChDsKkvEjQYIECRIkSLijIBk/EiRIkCBBgoQ7CpLxI0GCBAkSJEi4oyAZPxIkSJAgQYKEOwqS8SNBggQJEiRIuKMgGT8SJEiQIEGChDsKkvEjQYIECRIkSLijIBk/EiRIkCBBgoQ7CpLxI0GCBAkSJEi4o3BTvb0EQdjVvTAajYbwST/f7fwARBuNhuOTfmG3c7zRGAISx90AaS3ufo7SPF3GXue42/lhjbUoeX72FnZz8zkJEvYSpLUoQcLOwKprUTJ+JEiQIEGCBAl3FCTjR4IECRIkSJBwR0EyfiRIkCBBggQJdxQk40eCBAkSJEiQcEdhRxo/giCs+v+CIHzsZ7sRe50fIHHcKxyvx53AEfj42O5FSBz3BvY6x63ityONH2D1zbXRaKDRaOyJwd7r/IA7h+P12Esc1zJy9hrH6//caOz26t6VkDjuDex1jtvJ76Z0fm4Hmm+aexF7nR+wdznS4b/agt0ruBM4AsucaJNdzWDfC7ieY7PxKnHcPdjrHLeL347z/DRvPDKZjDfetUIMuw1r8fuk39ttuFM4Xh8C2mscgY+HufYax2Z+Mtnq2+Fu5geszXEvGbQSx4//ebdhu/ntKM/PavkE9CXU63XIZDLU63UAu9PCvRE/4Bqv3cgPuPM4Nv+3GRLHnY21+DXfOnczP0DiKHHc+VjNsNkufjvC+FltgxUEAaIo8t/V63XUarUVX8huwY34Uf5ErVZDvV7fdfyAO5ejTCaDXC5fMU/ptdc4EhqNxq7luNf5AWtzJC8srUXiuVc4CoLwsXGUOO5MXM+pOby1Xfy23PghYuS1aSbQ/LNmN5dSqYRKpUJfXx9UKhXMZjMSiQSi0SiWlpaQTCa3+rHXjVvh193dDZVKBZPJhEQigUQigaWlJaRSqdtBZU1czxG4ZoGvxlEQBCgUCiiVSnR1dUGtVsNgMCCVSiEejyMYDO56jsDyOCoUCrS3t0OlUkGv1yOVSiGdTiMUCiGdTt8WLmthI3NVoVBAoVDA6/VCqVRCq9Uim80ik8kgEokgk8ncDiqrojk0R5vmaj8jIwBY5ieKItxuN5RKJTQaDbLZLHK5HGKxGLLZ7O2isyo2wlEURYiiCIfDAaVSCbVajXw+j3w+j3g8jlwud7vorIqb4Uh/J4oi5HI5HA4HRFGEWq1GLpdDqVRCIpHYUxytVitzLBQKzDGfz98uOh/DRudpMz+VSsX8kskk8vn8pjo/NtX4ud41vloIpPln178AQK1Ww2QywWKxoL29HQaDATKZDGazGRaLBUajEe+///5mPva6sZrr//rnX+3vr+dnNBphsVjQ1tYGnU4HQRCYn16vx5kzZ7aFz2pYD8fVjILVOBqNRrS1tUGtVkMmk8FkMsFkMkGv1+PcuXPbR2oVPqv93c1wVKlUMBgMMJlM8Hq9UKvVAACj0YhkMgmtVovR0dFtYvRxbGSuNnNsNBpQq9XQ6XQwGo1wu91QqVT/v70z723rOtr4c7nzXu6rJEuyJTvxGmRBgQBJi6JBgL4pkO/XBSjQfpMCRdomTdDGjh0HcWztK/d9ve8f7DMaXlOyZIkU5XAAwZZIUfxxzjJntgMACIfDCIVCCAaD+O677ybC4xR9UhzF0ev1Rv7cqUPLsmBZFjKZDLxeLwDANE3UajWYpoknT55MFuwYca6lJzHq59NgJSMfpxEbDAbxww8/XOim8rpyHsZgMAjTNJFMJuXxYDCIer0Ov9+PZ8+eXUlG/oyMwWAQqVRKHqcBFAgEpoLRyed8P6P4AAzxJZPJob2EfM+fP5/OhGe+WZ4c3W73SyEB/Ty32w232y3WHj+UUCiEdDqNeDyOfr+PZrOJRqOBZrOJQCCA27dvDz1/UjKKj1+vwxeNRtHr9dBqtdBqtdBsNuHz+fD2229fCp9+7xyc52FMpVKIxWLodrvodDrC6fP5cOvWralh1Jyjnkc+MvIxy7KQTCYRDofR7XbRbrfRbrfR6XTg9/uxsrIytYzOqqZRjC6XC6ZpIpFIIBQKiR47nQ663S68Xi+WlpamYi7qr1fx6Y0zGAyKMd7tduWr3+/D5/Nhfn7+UnWo9XcWRj1OyWiaJtrtNvr9vjD6/X5ks9mh518VRr158rBlmqawdbtd2LYNr9eLTCZzZRm1ERAOhxEMBtHpdCQNBBh4LxOJxKUwvopPP0/zeb3eoefQc+73+4Wv2+0CGPDF4/GXjN7zyrk9P4T3eAYvxTdIBTJ23ul0ABwlLmvL0OPxwOfzwePxIJPJAAAKhQL29vZgGAYCgQAMw0A0GsV7770Hv9+PZrM5ZAWOy9p18vF7KpHhAyefZtR8qVQKAFAsFrG/vw8A4jWIRqN48ODBRPmOYwSO3JB0W3Y6HRiGIWEhzeh2u+Hz+cRt2e/3USwWcXBwAMMw4Pf7xTC6e/fupTHqzR3AkFHACUdGPt/JSE9kr9dDuVzG4eEhXC4XfD4fAIjxN2lGcjkXQY5V6rHb7UroaxSj1+uFYRgIh8Po9/uoVCrI5/MwDEO8I6FQCMlkcuKMzrnFn3HsclM4DV8oFEK/30e1WpVQut/vh23bME0T8XgcPp8PrVZr4jrUC70+jPDvc5zqPAnn2mQYBkzTRK/XQ71eR7lcHtIhvezTwAi8PBd1jqeTkfsL9wcarpVK5SXGSCQyNYxOPZ6GERiMy16vh263i1qtJoz00F6GHs/C1+/3XzpI60Oz3+8XHTYaDeHv9/ti+F30WnMu40dPNpfLJYlJhmGIVUpL1eVyodlsjky6o3s9FAoNeUO63a4s2j6fD91uF2tra0NJtOMUzUeFOeOXwWBQNhPNRwOBpy+6nrvdLprNppyiOUk5uDc3NyfGdxIjgCEdkrHVah2rQ9M0ZbFttVriJeAE5mDe3t6+FEZ9kqJ+KGR0u91DjFqPgUAAwWAQgUAA/X4f7XZbJix5yLi/v3/pjPy73Eg4xlwuF9rt9khGv9+PQCAgjPV6XZLUNaNt2zg8PJRxM6m56NQh37+Tj2uOk4/5dnozoZfA4/HIGLBtG/l8Hh6PB+12e2I6BPCSDrV3wOfzyQGr0+nIpqnHs2ma8Pv98Pl8sG0b7XYbvV5PGPn7AFAqlaaCETiak049UvfcX4DBfPX7/bJ+0lsAQMYnf69SqcDj8aDT6QyFmCbBSNGMo/R4HCPzCqlH/j4Z+feq1erEGE/Lp3Wo1wnNx5xCHq6pNx5QaGDV63Xxhl1U8vO5PT80fCgE5ILJU5lt2xLf5JunWzIcDiMWi8Hn86FYLA5tKAx98VRerVaHlD4p6117A3QlSLvdPhVfNBqFx+NBqVRCr9dDp9OR36M1exl8mhEYLinkgtlqtWTj0/olo8/nQzgcRjgchsfjQbFYlMHMxYlhvWKxeKmMWo+6ioB65ISlHvm4z+dDKBRCOByWBZVjkh6fVquFXq+HXq+HWq02NXokA+ci5+VxjJZlAQBqtZoYFj6fD4ZhiKFQqVSEUY+LSfI5DXDNp/Wr+Zjj0+/30Wg05HHNx8+l0WhMlI+M2ihwjlOerJ2PaUYeQmzbRr1el+f4/f4hvTOdYBoYNQM3Sz7XWUHp9XphmiaCwaDsD2QMBoPCaBjGECMw2bnoNO60IUsj+yRGHrS4RwAYYqQR0Ww25fFJMZ6WT7eoOY6POuRrmKYJt9stRhydIWS7KDmX8UP3Kr0+ekPkoKYHR5/cNCDdzy6XC9VqFd1uF6FQSFyZ/GBoDHHj1Mofl6K1i9yp4FF8ZCMfPR7BYBCGYaBWq6HX68GyLAQCAVmEms0mer0ePB4P6vX6RHsaaR1qo01b8jw5akaXywXLskSHnJCsOKCbkow0DAzDkAV50oyj9Mj/04vDSU3OUCgki5RpmjAMA81mE4ZhIBKJwO/3y6SmHgGIwT5JRn26curQMAw5FY5i5M+CwaC8f7fbjXA4LB6EXq8nnxM/B71xjpNRu9PJ5Px7PP2TjY9znAJHIWYePvQCrA9a9I5Nik8zOnWox6nWoZOR32uPDw0FhvM0I8cE56L+O+NmpDj5eAjU6QOjGJl83+12JXGd3hQAsmeQkX/7shmBo/C6TgHhe2MBDL2wZAwEAgiFQnKYpEHEscBwlw4bTmIuOtmcfM59n3wAhtYV0zQRCoVG6tDj8chaoyNM5+V7beOHrjfG7fRpixsen6cnNRdPboB0fTGsxVyJSCSCQCCAdruNer2O9fV1NJtN2YjpomZS1EULQ0E6pHfc5slJypAJY+w0gui67PV6YhSM4mu1WuLa5OAaF59m1MaP8+skRp78yciBG4vF4PV6xZtHxo2NDQmH9Xq9iTLqTcXpvaKQ7zhG5m6RUeuRp6+NjQ1JDp6kHp0bp9PFrL0j9JA4GWnActHlYhSNRqXstNlsYnNzc8hrNG5G6lCHW8mkDXb9WTCUwxwJYHAIYbIlMKjM83g8Mk4bjQYajQZ2dnYmyqfft2Z0en2czx3FGAgEZJwaxqCKlIxer1d0uLu7O/QZco0bJyPwsgd2lEGiDU4y9vv9oXHKg6nL5RL90btOPe7t7cnfpZHAeT9uRu1J04fiUYxer/dERo/Hg0QiIWuq2+1GvV5Hs9nEwcEBAIz8PCfFN+rAxZ/rcerkc7lcsjemUil4PB7E43G43W7UajU0Gg3k8/khTzW9nhch5/L80PBhfI8uZ278AMRNDkCMJE5mt9stfXzoNQgEAkin0/KhsdS2VquhXq/L4kzrcpxCPq/XK4ugbdsv8QEY8hBxU+VnwwHbbDalHJObbCgUQiQSQb1eR61WkwFh2zYajcZY+cjocrnEOOMA1n9bMzq9fNQhs/GbzSaCwSDS6bQMfjI2Gg2ZtJNk5MapE9QZznGOU31KcTJyA2m327AsC+l0GsBRZRRDCRynNJQmkWegGblI0ONGRl0dQj1rxkAgIGMzn8/DsixJ0AeAeDwO2x6EQenh42uNk1EbbeTiv+12W543is+51sTjcXg8HhQKBZmLmo8Vio1GA7VabSJ8ZKQOyQYcNc10MuoN1cnIPinFYlHWU75WLBYTI6fZbEquiH7tcQo3TmeOofaycW2nkcr3znGg9VipVGCaJtLptHxWbKnBsFelUhnS4ySMA37pVAlnjpZm1HrUjF6vV1oSZDIZ0V00GpV9pdFooFwuD+lx3J4t6lAfro7jc45T8iWTSeELBALIZrPyOuRjqXuhUBA+bVOcR4yzvIhhGLb6P2KxGEzTFECdCNXpdFCtVlGtVsVK1Kc0n8+HdDqNpaUlLC8vY39/H4Zh4J133sHS0hJs28bOzs5Qk7Xvv/8eP/zwAzweD7a2trC1tYXd3d1TfxC2bZ8YMHTyRaNR8Uxpq9rlckl+DhsvOePWPp8P2WwWi4uLWFxcRC6Xg2EYePDgAZaXl4WPiYnLy8t48uQJHj9+LHybm5vY2dk5i6K/tm37F2dhpOeCjFqH9NCdhvHatWsoFovCqHXIROHV1VU8evQIDx8+hNfrxfr6OjY2Ns7E+CodjmKkt03ng9DI4UbHCoNRjJlMBouLi5ifn0epVILL5cLdu3exuLgIANjd3UUwGEQoFMJbb72Fhw8f4r///S88Hg82Njawvr4+VkYAEp7S+SAAJHY+ipFzkeXACwsLyGazqFQqcLvduH37NhYWFgAA+/v7UjZ97949fPvtt/jmm2/Gxujko8v/OD563o7jS6fTmJubQzqdls/h1q1bQ3yWZSEej+Odd97Bo0eP8PXXX8PtdmN9ff3MfDjjXAQAy7KGDhh6nDJc1Wq1XhqnAKSSNJPJCKPH48HKygrm5uYAAIeHh9Ki4YMPPsCjR4/w73//G4ZhyFzc3t4e6zjlfkH9ORmZ36ENI82YTCaRyWSQSCQkN+j69evIZrMwDAMHBwdyMPnwww/x+PFjfPnll7BtG2traxNj1LcTkJHGAiMYfNw5VpPJpHwxd2Z5eVmcAmTMZDL45S9/ie+//x5ffPEFAGBtbQ3r6+vY2toa21ykB1XzaWOPEYxRfNQhW6H0+4NCimvXriGbzQIADg4OYJomstksPvnkE/z444/4+9//Dtu28fz5c6ytrWFzc/Pcc/G1PT/axcXQBuH44YTDYezv70vVDy1SuvqYC8N4psvlwo0bN/D+++8jHo9jd3cX+Xwe7XYb9+7dk1AXe3N0Oh3s7e2NxcrVfNotyc6a9ErlcjnhA44y4NlJ1fhfTlO73Ybb7R7i29vbQ6FQgG3beO+99+D1etFqtRCLxWBZFlqt1pmMu/MwslKAJ0jmt1iWhXw+P1RFMIoxEomIgbi8vDzEyIqSjz76CD6fD81mE5FIBMFg8FIYgaMqPGAwmYvFoiQr83E9ToFBmIQTfXFxEe+++y7i8bh0Og4EAvjkk08khKJDYuNkJKeTkScsl8sljSY1I+ciK6AAiNvZ7XZjYWEB9+/fRzQaRT6fR7VaRTgcxv/93/9Jl2D25mg0GmPX43F8XEPK5fJLfMBg/jLXJ5FIiNdqbm4Ob7/9tjSmrNfriMVi+N3vfjfU6JCn60nokPOP6wlzP/hvpVIZSivgc8jocrmQTCbFwMtkMrh58yZCoRDK5TIajQYSiQQ+++wzaXQYCASE8YwG3mszer3eIc89xx/DdfSYck0BMLRnpNNpSYJNJpNYXl6GZVm4du0ams0m0uk0fvvb38KyLJTLZWlRcVmMnGfkYP6nM3fN6/XKnM1kMrKvJhIJLC4uIhAIIJPJoNVqIZ1O49NPP0UoFEKxWBQvUr1eP5OB9zp8AMTLTCNH67BarY5MndBzcW5uTvacRCKBubk56enTbDaRSqXwm9/8BuFwGPl8Xg4+1Wr1TMbdcXKusBe9IO12W5KVAQzlgFy/fh3AoKRyb29P8nr6/b50jV1eXpawF118Dx48QDgcxk8//SQdHzOZDFKplLh16/U6Hj9+PLQQXLQwYY6DFRguQVxcXIRhDKp/9vb20Ov1xJWu+RinrdVqwheLxfDixQtkMhnxMNAqZtxz3HzAsJeHjEyW9Hg8cjquVqs4PDxEr9dDIpGAYRiYn58XPTOnoF6vC2MymcTGxgYWFhbg9/sxNzcnjHRbf/fdd2NnNAxDvDycsKxG83q9mJubk6T7XC6HXq8nhgAn5crKCnZ2diRvIhgM4t69e9jf38fu7q4sTvPz84jH40ilUjCMQRnqJBl1cjkZPR6PnKzq9ToKhQK63a7khJDx+vXr2N/fl9OpaZq4c+cODg8PkcvlcP36dRmrsVhMxnq1Wp3IXCQfPXhMiKTnwzAGCfWsqoxEIuLZ8ng8WF5eRi6Xk2RK0zRx69YtFItFFItF3LhxA263G6lUCpFIBMlkUnIVJjEXDcOQxHIdrmQInp83Qx0AhhjpJSgUChLyNE0TKysrKJfLqFQqWFlZgWEY0sAykUjg3r17E9Mh9abL03mwdLvdiMfjAAYJr6yqZP4ZuxsvLS2hVCpJywbLsrC8vIxKpYJ6vY7V1VXYto1IJALLshCLxSbGSE4yUo8M17rdbgk/NptN2Vs0o8fjweLiIiqVinhRgsEgFhYWUKlU0Gw2sbKygk6nI5VvmnGc6w2NHd0mgnxMo4hGoxIiZ9WoZVnw+/2yvy0uLqJarUpbCoZoy+Uy2u02VlZW0Gg0JCeYXudyuXwhfK8d9gIGbtpwOIxAIDDkHSAIT090Xd6+fRu/+tWv4PP58Le//Q3NZhMejwcPHjyQzSYSicDtHjTKu3//Pvx+vzToevLkCYrFIkKhEGKxGDY3N/HHP/5xKAfnJDmre48Z6H6/H8ViUSYrT5yaz7Zt3Lp1Cx9//DF8Ph+++OIL4bt//z5s20YymZSKhVgshjt37sA0TbRaLTQaDTx69AjFYhHhcBiRSAQbGxtn4sNruNoZrvH5fCiVSuKhY/4AT4fAYHDfvHkTH330EbxeL7766ithvHfvHgAMMYbDYdy6dQvhcFhO5N9++y1yuRxisRjC4TDW1tbw+9///sJ0eBwj3e2VSkUqEZiTFQwGJWHUtm1cv34dH374ITweDx4+fCj9m+7evQvDMJBKpcQbFAqFsLS0hEgkAmAQhnn06BEODw+RSCRgWRbW1tbwhz/8QUqnx8HI0CIrInnq4lilp4ZjdWlpCR988AE8Hg+ePn0qoQZ2GE+lUlJmy3g88yhCoRAeP36Mg4MDJJNJhEIhvHjxAn/6058ujNHJ5/f7JQmUiykPX/q+KvJdu3YN7777LgzDwIsXL6R6bXV1VUILzB3kwsocing8jqdPn+Lw8BCZTAamaeLFixf485//fGo+vMZcZP8hemF0wjW9c/Ra2baNbDaLd955B7ZtY3t7W/Kfbt68KSdoNo/zer2SP+nxDO75evbsGXK5HObm5hAIBPDixQv85S9/Ges4JQfzA7k3OBmpx2w2izt37qDf7+Pg4EB64ty4cUM6G9PDwkTvcDgsnsvnz58jn8/LFTRkdDbLu2hGeppogGpGn883pMd0Oo1bt27Btm3ZZwDInYFM5GY1Kqsw6YFeX19HoVDA8vIygsEgXrx4gb/+9a8Xxngcn2EYYqSfxJdKpbC6uop+vy/GDgAsLS1J0Yjb7ZYwpm0fVYUtLy9je3sbhUIBKysrMhfPwoeLDnsBg3JR9kdhsi9wVFrLxZMVCT/99BNarRbm5uaQSCSknJ2xam6o3KS0+5BfpmnC5XKhUCjA5XIhHo/LSfei3Xy03J2N3MjHxZNhr7W1NfT7fbmeIxAIoNPpiEvP7/fjyy+/FOOOXhYu3jrZOZ/PAwBisdjY+E7LyL5EvV4PGxsb+Mc//oFEIoFoNIpMJoNOp4NYLIaFhQUEg0H885//hMvlQiKREG8gy1FpWHW7XeRyOUlQHDcjFw4tNA5Y6kzGnZ0dfP3114jH45Ij0e12EYlEsLCwANM08a9//Uvy3uh+50Tm9zRqmcBHj8U4GNmwb1RF1CjG/f19PHz4UPL24vE4ut0uwuEw5ufnYZomvvrqKwCDMcg+HEz8pqe32WxKEjvZx8FIPnoOKJqPz+n1ejg8PMTTp0+lrQS7VVuHbUC1AAAWQElEQVSWJXzffPMNgIEB22g0sL29LQnDpVIJrVZLLnGt1WoIhUJj1yE9dqN0qBsz9vuDDuo//fSTeKJ5yAgGg8hmszBNE//5z39g27YcqHZ3d+H3+1EoFFAul1Gr1aS7da1Wk3D7uBgZztJ5S05Gfc1IqVTCxsaGrJHsyk2DPBAI4NtvvwVwdBg/ODiAz+dDpVJBpVJBtVqV9bRUKsGyLEljGBejzi08jpFhr0qlgp2dHWHkOklPSCAQwMOHDwEcHcgLhQK8Xq9ES+ixBga3B4yTkTrU5e4cr6P4qtUq9vf34Xa75SDK3mG0Ax4/fixeSsuyUCqVxCDiRcO8LYDFGOflO5fxo/N4dEk4PyBWu+gKjWKxiEKhIHB+vx/5fB5vvfUWstks6vU6TNNELpeTU55lWVhZWUE4HEan08Hh4aGEHpjIOo5KBc1Hi1vnE7B3BhNpWUWSz+dRLBbFEi4Wi3jrrbcwNzcnYS/GMMm3urqKcDiMdruNw8ND1Ot1cfnpv3vRokt6dV8NMmod0hAtFArI5XKiQ5/Ph3K5jHK5jPn5eWHktQ/cOFZWVmBZFprNJvb392XiMlF3nIxaj1yUOGEZZ+c4pacvn88Lo9frRalUQqVSwcLCAmq1GgKBgBjhHLcrKysIBoOSI1Kv16XahCWf42LU3hBdacIxytMox265XEaxWJQkblZeVKtVLCwsoF6vw+fzCSPHqva0bG9vo9FoSF6XrgC5SNH5A9oA0onP5AMgOUDc5FkJV6vVUKvVsLCwgGq1KrrmXNTdkQ3DwNbWlniex8nnZASOKrrIqHVIxkqlgkKhgEqlIowMPWezWalYKxaLQxW1zFskY7PZlM9h3Iw6D+9VegwGg+ItqFarwsg7H9PptHT/ZXIw56ZOzN3Y2EC73RZGzo9xiDNXSReRAEd65Hpnmibq9TqKxaJU/LI6uF6vI5VKodFoiDHAcUxvKKugyFgoFIaSyMfB5+wwTT4mdrN1jWEYwscqUc1Xq9WQSqVQq9VEh5qPHlsA2NzcRLvdlqt2zst3rrAXMFAkb3jmoNYDL5fLSZ4FPyieRDi5eZs5K0d41UAoFIJtD3oC8ETCSZpKpRCPx/H48WPs7OwMxVaPk7O69wBIOTq9WACGBl0+nx/KIyEfE7iZzEW+zc1N8XiRz7IsBINBmdClUgnxeBzRaBQ//PAD9vb2TsWH13C1k5HxWApDJWTUjdE42OmaZEiPIcK1tbUhRj6XZYuc6LFYDJFIBM+ePROX9nl1eBIjW+JrRt65xpO+k5GJ32RkVdX6+rp49kKhkCSH8zTGDZM3oK+vr+Pw8PBUZZrnYQwEApJYySRLbibMH+DCTC7TNOU1OBcDgQA2NzdlEaPHlZUsZKxUKrAsC6ZpYmdnB/l8/kIYR/HRw8PETvKxQqparYpxqY14hijJx7G4tbUlIROOZTZErFaraDabKJfLklOxt7cnifHjmos8OeuO6jwc0ADl58tNAYCsT4ZhIB6PwzRNBAIBqZjtdDoSNiQrGZmszwML86XGNU7p4R7FyJwteoxHMQIQj2UgEMDe3p4kwutxCkAYmdjNwzYTx8fNSCYeLjjvGo2GMNJ4AY7uljOMQbUx9cLrcsjIsdrr9WTc12o1me+lUgnVavVCGI/j48GcTE4+7vmaj5EN8rHr+sHBgfCxBxA9RFy3NB+9lueZi+e+3oJltPquFVq27XYbjUYDrVZLNkMOeHpx+EEywVI3h2Poq1arYWtrCwDk591uV5rnjbOvQb8/6N2iN0p+MYG21WqJwab59ELMBEudN8RcGr4GT+1M5KPlP+6+DSy/5MDioGZ4irphaIFJbZZlSQyeie9cuGjhk5GlyBz4jG3T4JsEI2+VZwiVJ5BRjCzHpXuVi/QoRuYLkZGnEeqR4a9xMzJRlnx6c2GiMBl1ciLv3+Oprdvtolwui+eTHgPmMDBECUCSEzlexz0XGWbXzQoBCB87/vb7fdExqwq191ZXhhnGoPu62+0WfZGDGwt/Pq5QiWZkGJqc/Htc805i1CdihiYZDmXuGscBGWu1mqy9k2Kkl0b3UaMxrhk5F2kUUo+6p5wOibLKzanHer0uaRrjDHk5GTnmdHSCjAz9jGLk7xiGIWsK9ddoNKSdDD8rMrbbbfj9/rEzMvSlDR/dJ+44PtoF5HO73TLmRvFxTtu2LXdi6n30PHznNn4Iyomn73Jil0ana1PfGs2FmRsKBy43Vy7AlmWJwVEqleRUViqVxpq5b9u2DDJOOi4gDL2N4nO5XFIa7eTTg5vWP5OEu92uVGow32DclQnUIbPuqR962nRvEe2CZ9m/3lBo2PHzoueg2WzKqZoTk3eGTYqRl8lqRo5dbeDocBgZGbunUXASI3VOPdIVP87FlsLFQY9Vveg73cUch/TykYMLG3+H92FpL1Cn00GhUJDFbBKMOiGfhwUyc345+RjW5IbJ8ab5uEgzZ4aP0UjnfJ+EDpnvosN85NOMOpTLAyeNpXK5LN4hGgw0MsjI1+V9ipzvk2LUDfCM/1W58UsfTICjcao7yeurjnjQpJeBRgA/J+qR69qkGKkPzajTDLh2Ulc8uNADpBk5LnQuLZPkmQPG/MxJMGo+zjcnH9cUMjCpnd5p8lHXOp2EfHSc6HsxL4Lv3MYPAEnQpTK5+PCDMAxDYpXaiwIMPkC6tTwej1zIp0+fzG5PJpPodDrI5XLodrunDnddBB8b+HFDOQtftVqVjd6yLOGjB8vtdktZbbvdxsHBgSSlToKPUiqV5PTPuL9m5HslI5OY6VYmYzgcHmLkSTYajSKVSqHVamF/f18qOE4TJrkoYQKkzlFz6pHxaIY3ya4ZWd2l9ehyuYYY2YPqtKGgixI2FtVzUSdhUo+tVksq+phHwbnIclw+n+EWGvUcq2Q8Qzjo3MJu7zpJfxQfF0+GuXigoA7ZWb7T6QwdTkKhEFKplOSm2bYtOp6UsLEo87ecjDQAyciKzU6nM6RDdl6nl5qMXE/1OKWOJyU8POocNXJqxk6nM8TIvUEzMp+OngJg0PgzHo9LPzj+zUky8vB4GsZ+vz90r2W5XJaxym7IoxjZ8JFjVd8xOAk+jlPORc3HeUrDjaG8drstfG63G+l0eiQfu8xzX7xIvgsxfii0/pxJbcDRyZtuPCal6ZMcf5eLNt22/F0mgwFHN71OakMhg04sPS0fF2M+j+4+Gga0cmu1mnge6I2YJB/fn+bTf18vvjR4dGY/NyPq0O12o1wuCyN1CGDI/T5NjDTayFipVGTBdVY40BCmkc6QhZNxkgas5tBu6Fcx8oRG3XEho+tdM3IjBY76YE3SuNN89ESOmosM9ZCPXoVROqT3g2NcX6Som5hOUvQ4PYmRVziQUesQOLrCplQqySZLDx8AMaKmlZHhLOc45YZLRh1O58FNN26dZj0ex+jUo86Joq4vm1EbO6/iY2GEk4+OBQBihNOeYHoMjaiLMuwu1PgBjkqk+UEARx8GF1T+rNfrSUY3Xcp0+7ndgy7DBKeFyecydDJpOS0fFxgnH3NO2Ambrl+egJg3cll8Z2HkJspTNRmZ+8VcBA5q5o2QUTc5m7ScRY88XWtGuu2pRz1Oyagvqr0McS5K/Blw8ljlxq8bJNKQB4YZGUq8DEbNMupnTj4moPLQRLc9E6m5SOtxepl8mmfUz8hIDwI9JJqROmTIgYdH9ufSeRSTNtCdPKN+pjdUJyM3QurRWSmmGXkInUZG/T2NGRaHkLHdbsv4nUbGs/Ax75B5kNQr1xqGK4FhPl21exFy4cYPMPqae+CoeyldgKykooKZvc2EJ/4OKzx2d3dlU7msQQycjo8xaicfT826GoX5Ftvb21PBB5yOkTlempFlqXTbU8i4ubk50dyCk+RVjHTn+v3+oUaNurpCb7LMDbpqjMx5CYfDcuI6jpH5BdPCeBo+AC/x1Wo16T4LHHmtmW/AsuHL5gNGbyzAyzr0er0vMVKHwNHFpdThVWMEjnpp6VvCtR513ki328X6+rqEUqaVkXlJnGdcb5gITD2ycSA9PQxzkvEyD1rAyTrUCe2jdMhIAg12OkLa7TbW1tYkJ+0i+VyvfsrFyKhFikqnp4fP42RmUyteIKldfNMmo0ILjNVrPuDoNHOV+IDROmTiKU/QeqPUOrx27RqAI0/KtIpm1CdrzcjncZwGg0FpDghcLUZ+z3wEutb5c+dY5SWZPI1Oo4wap84wHp9HYafzTCYji/W08gGjdcjkWR0O0iECMqZSKQBXc5wCR9coUHRIieOUVyJxTZ1WOY6RY5WiQ9hshMgrhqZ53xjl/QGOGszqcap1yMarnLfjSI8Yi+fHKaMWGwBDTY0YDmIirdfrxerqqlReTDKh8qxCPqdVOopPl5PfvHlTXJmFQmFq+YDjGZkToatNtA5v3rwpVQxah9PIeRwjY+msmCErAGFkgjFLqK8Ko/bOsfKJxjp7cng8Hty8eVNKq1nVN42Mo3RII87JR13y3ja63nUPmGnjA44fpwzlAUd5Pqy+9Xg8uH79uoRPdA+Yq8SoDTYyMuTl8QzubmPbDl7zcpUYGRpil3iGLFlV5fEM7vxiOGia9XgcX6/Xk3wz6pB5vm63G9euXUOj0ZBmh+Pim5jnBxg+SdLdSquV7deXlpZw9+5dLC4uYmlpCYlEAnfv3sXnn3+OdDo9ybf7WuLk47+hUAjhcBjXrl3D7du3sbCwIBdgvv322/j888/l4slpl5MYQ6EQ5ufnpWN3MplELBbD6uoqPv/8c/EcTLtoRuaYsZTdsixks1msrq4ilUohGo0iHA5jeXkZn3322ZXSI93MWo9sWsjbsmOxmDQanZubw6effnol5iKl0+lIZRF7kvEiyGvXrkkD0mAwiEQigV//+tfIZDKX/bbPJByn7BnGxozhcBjZbFYaPTKk8vHHH18pHQKQcVqtVtFoNKTJoWVZSKfTUphAXX744Yfi5boqwrFKRvbP49hkVRtzuH7xi19cKUamRFSrVbkAm4flSCQiuXZMgH7//fflMt+LlokaPxTngsukQlYd1Ot15HI5uRwyFovh/v37cnP4tAv5eK8Tm5ORr9Fo4PDwEAcHB3LB5507d7C4uHgl+IAjRjbW0ve8MQafy+WQz+cRjUZhWRbefvttzM/PXylG5mjp5FfdXj+fzyOXy0mX85s3b8oN6VdFtHGgGTkXeSWNaZrw+Xy4ceMGstns1DPSgKUO9XzUfKVSCYVCQVpULC0tSWnxtMtxjNQh52ilUkG5XBZv18LCgpSIT7s4GUeNUxoMrJZyu92Ym5uTi1yvimg9cl09iTGdTl8pRvKx6aTmY5U0W5G43W65OWAcfBP9xLSLnV1j6RJj3xTd5bHZbGJ7exvlchm7u7vSh2ZaRb839sPRiZRszqQrnra2tlCpVHBwcIByuXxZb/3UchIj+zfQ0KOBsL29jUqlIm3lp12c41S3auf1GNQlGXd3d1Gr1VAqla6cHlmSTz3SS8ATJhNG9/b20Gg0UKvVpl6Po/iow1F87fbgTj3qdFJNKc8jTkad8MpGeLpxbLs9uBep3W7jyZMn0n19mkXn3TkZ2ROOFV/cTNm24OnTp9KSYZpF50kyjYCMlmUBgORyaUbLsvDjjz9OPeMoPn6vdch+VGxoaFkWnj17JiGyC39fZ5ngxog7Pk7xO/Kv/luBQADAIKbZbDaRTqelNwNLGa9fv47nz58jk8lgfn5eWrhvbW0hn89ja2tLYqOnEfs17jC5KL5UKiVKZm7M8vIyXrx4gUwmI7f3ttttbG5uolgsYmdn56wJia91n9DrMvr9fsmhcDLyhLm0tISNjQ2k02kkk0kpi9/Y2ECpVML+/v6ZGF+lw3Ex0qOVSCQAYKhl++LiIra3t+UURj1ubGygWq3KvWXTysjSWXq0mEipy6QXFhawu7uLVColjdg4Vmu1GvL5/IUyXjSf1iETKXnqdLlcmJubw8HBwdDdX61WC9vb2+L9OmP/lInORScjG1SS0TAMZDIZ5HI5xONxuTOq0+lga2sLjUZDejmdViY9TnX+S7PZfIkRALLZLHK5HGKxmISLOp2OVNKetVnlZTNGo1Ex+siYTqdRKBQQDoclObjb7coFtcwDuijG1+XTRT+n4WNRUDqdRrFYlLsiNV+73T4zH8Z1t9dJoq12Z78Rup7ZW4PZ3yyzZRKwbds4PDzE7u7u4A3/r9ulLhW/LHHyaUVrPgBDzdXIl8/nAQD5fB7b29sABoMjGo0OVYddppzEyHu+NCMfZ0PKYrEI27ZRKBSGGHnVxTSIXoCcE5Z6YwmtvlpBMwJAoVCQO+iYvD8N4xR4NaNhGLIpUI/AUafhcrkMwxg0kdve3hajV1fAXaY4+fRa49Qhm47ati3X71QqFRjG4LLPvb094WNbjmmQ8zLyVvRarYa9vb2hhOirwKg7eAM4kbFer2N/f3/oWo9piRq8LiM9IOya3Gq1kMvlJHlfV6Bepmg+5+eu+Zg6wWRuzUdvbT6fH+K7SBmb8aOBR71pfS9Nr9dDPp+XW7G5oLL/RKlUEteexzO4UVq3wL5s0QOZ4uQrFAoj+XgnC/tUMMOdF8NNi7yKsdvtSl4Iq2kMwxAdFgqFIUb2kZkmRsooRuN//UaKxeJLjCzDdOqRl/hNOyObAJKxVCpJby2Wo5KxXC5LuISVUp1O59IZR5XUchFmxQ8PV+VyeYgPgPCVSqUhHdK4v8z+KZSzMFYqFam45IbIQwpvxLZtWz4D3VH3MuW0jOy3xTvbnIyVSkUYud6yI/Jly1kYeZO5btFAhwEZgaOqqWnQ40l8es9gjx9GejQfb3jQfFyfLsoIGnvY64TXAvCy5aubG0ajUTmRsUyV8joKHod774TX4t88li8Sicg9TBfBhzG52k94LQAnM4bDYWmUpxlft3RxXG7oE16Lf3eIkYutYQzugqKH5E1g1CdIfvEONwATYZwUHx8n30XpEJc8F3WOBR83TVOMPGeI8iqM0/+93omMDHO9yYzstj4pxqvOh8sIe50kzs1E/5yPsU+DUy7bsj2NHMen+6vUarUrywecjlEntTt/9yrIq8YpgDeK0fl/MujEdv28q8Co36OTj//SC+v8navAB7zMCLwcXjmuk/NVYQRe1ouT8bhOzm8S43EN/64K47TwXWp9nAZ2LqS6M6d+3lVRMDCajwvvm8AHzBht+6ix3JvA6PyevG8Coxbnex/Fd9XFyeAMGbzpjG8CH/DmMx43F0c9dpEyNc0BRilT3wdy1eVN5wN+3oxvkozaIN8kxlFj8U3iO05mjG+GvOmMk9orpiPF/wR5UzbN4+RN5wN+HowzmclMZjKTqyNn9fwcAlgbxxuZgFw/xXOuMh/w5jOehg+YMU67vOnjFHjzGWfj9EjedMarzAccw3imaq+ZzGQmM5nJTGYyk6suUx/2mslMZjKTmcxkJjO5SJkZPzOZyUxmMpOZzORnJTPjZyYzmclMZjKTmfysZGb8zGQmM5nJTGYyk5+VzIyfmcxkJjOZyUxm8rOSmfEzk5nMZCYzmclMflYyM35mMpOZzGQmM5nJz0pmxs9MZjKTmcxkJjP5WcnM+JnJTGYyk5nMZCY/K/l/RaShV4cXEAYAAAAASUVORK5CYII=\n",
+ "image/png": "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\n",
"text/plain": [
"<Figure size 720x144 with 20 Axes>"
]
@@ -5297,7 +2940,7 @@
},
{
"cell_type": "code",
- "execution_count": 26,
+ "execution_count": 23,
"metadata": {},
"outputs": [
{
diff --git a/Scripts/Notebooks/Untitled.ipynb b/Scripts/Notebooks/Untitled.ipynb
new file mode 100644
index 0000000000000000000000000000000000000000..9d6af69c9ab2941c9bf98bac7a7f2b2a93ec5bd5
--- /dev/null
+++ b/Scripts/Notebooks/Untitled.ipynb
@@ -0,0 +1,386 @@
+{
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "# Autoencoder"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Importamos las bibliotecas necesarias"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 3,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "import time\n",
+ "import os\n",
+ "import shutil\n",
+ "import sys\n",
+ "import math\n",
+ "import random\n",
+ "import tensorflow as tf\n",
+ "import numpy as np\n",
+ "import scipy.misc\n",
+ "import matplotlib.pyplot as plt\n",
+ "sys.path.append(\"../tools\") # Herramientas propias de MantaFlow\n",
+ "import uniio # Biblioteca para la lectura de ficheros .uni"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Inicializamos las seed para funciones random. Al ser inicializadas al mismo número, el resultado no cambiará en cada ejecución."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 5,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "np.random.seed(13)\n",
+ "tf.set_random_seed(13)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Ruta a los datos de simulación, donde también se guardan los resultados."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 6,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "base_path = \"../data\""
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Carga de datos de simulación"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Cargamos los datos desde los ficheros .uni en arrays de numpy."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 7,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "densities = []"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 8,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "for sim in range(1000, 2000):\n",
+ " if os.path.exists(\"%s/simSimple_%04d\" % (base_path, sim)): # Comprueba la existencia de las carpetas (cada una 100 frames de datos)\n",
+ " for i in range(0, 100):\n",
+ " filename = \"%s/simSimple_%04d/density_%04d.uni\" # Nombre de cada frame (densidad)\n",
+ " uni_path = filename % (base_path, sim, i) # 100 frames por sim, rellena parametros de la ruta\n",
+ " header, content = uniio.readUni(uni_path) # Devuelve una array np [Z, Y, X, C]\n",
+ " h = header[\"dimX\"]\n",
+ " w = header[\"dimY\"]\n",
+ " arr = content[:, ::-1, :, :] # Cambia el orden de Y\n",
+ " arr = np.reshape(arr, [w, h, 1]) # Deshecha Z\n",
+ " densities.append(arr)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Necesitamos al menos 2 simulaciones para trabajar de manera adecuada."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 9,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "load_num = len(densities)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 10,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "if load_num <200:\n",
+ " print(\"Error - usa al menos dos simulaciones completas\")\n",
+ " exit(1)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Convertimos la lista \"densities\" en una array de Numpy."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 13,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "(1100, 64, 64, 1)"
+ ]
+ },
+ "execution_count": 13,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "densities = np.reshape(densities, (len(densities), 64, 64, 1))\n",
+ "densities.shape"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Creación de set de validación"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Creamos el set de validación de entre los datos de simulación generados, al menos una simulación completa o el 10% de los datos (el que sea mayor de los dos)."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 14,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Read uni files, total data (1100, 64, 64, 1)\n",
+ "Split into 990 training and 110 validation samples\n"
+ ]
+ }
+ ],
+ "source": [
+ "print(\"Read uni files, total data \" + format(densities.shape))\n",
+ "\n",
+ "vali_size = max(100, int(load_num * 0.1)) # Al menos una simu completa\n",
+ "vali_data = densities[load_num - vali_size : load_num, :]\n",
+ "densities = densities[0 : load_num - vali_size, :]\n",
+ "\n",
+ "print(\"Split into %d training and %d validation samples\" % (densities.shape[0], vali_data.shape[0]))\n",
+ "\n",
+ "load_num = densities.shape[0]"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 16,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "(990, 64, 64, 1)\n",
+ "(110, 64, 64, 1)\n"
+ ]
+ }
+ ],
+ "source": [
+ "densities = np.reshape(densities, (len(densities), 64, 64, 1))\n",
+ "vali_data = np.reshape(vali_data, (len(vali_data), 64, 64, 1))\n",
+ "\n",
+ "print(densities.shape)\n",
+ "print(vali_data.shape)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Creación del modelo Autoencoder mediante Keras (Sequential)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Importamos las bibliotecas de Keras"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 17,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "Using TensorFlow backend.\n"
+ ]
+ }
+ ],
+ "source": [
+ "from keras.models import Sequential\n",
+ "from keras.layers import Conv2D, MaxPooling2D, UpSampling2D"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Creacion de las capas del modelo"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "En la primera capa debemos definir las dimensiones del input esperado."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 22,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "l0 = Conv2D(filters = 1, \n",
+ " kernel_size = (3, 3), \n",
+ " activation = \"relu\", \n",
+ " padding = \"same\", \n",
+ " input_shape = (densities.shape[1], densities.shape[2], densities.shape[3]))"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 23,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "autoencoder = Sequential([l0])"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 24,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "autoencoder.compile(optimizer = \"adadelta\", loss = \"binary_crossentropy\")"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 25,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Train on 990 samples, validate on 110 samples\n",
+ "Epoch 1/10\n",
+ "990/990 [==============================] - 1s 508us/step - loss: 0.1760 - val_loss: 0.1559\n",
+ "Epoch 2/10\n",
+ "990/990 [==============================] - 0s 406us/step - loss: 0.1219 - val_loss: 0.1284\n",
+ "Epoch 3/10\n",
+ "990/990 [==============================] - 0s 407us/step - loss: 0.1087 - val_loss: 0.1143\n",
+ "Epoch 4/10\n",
+ "990/990 [==============================] - 0s 405us/step - loss: 0.1011 - val_loss: 0.1074\n",
+ "Epoch 5/10\n",
+ "990/990 [==============================] - 0s 409us/step - loss: 0.0943 - val_loss: 0.0977\n",
+ "Epoch 6/10\n",
+ "990/990 [==============================] - 0s 408us/step - loss: 0.0870 - val_loss: 0.0896\n",
+ "Epoch 7/10\n",
+ "990/990 [==============================] - 0s 416us/step - loss: 0.0790 - val_loss: 0.0824\n",
+ "Epoch 8/10\n",
+ "990/990 [==============================] - 0s 409us/step - loss: 0.0731 - val_loss: 0.0789\n",
+ "Epoch 9/10\n",
+ "990/990 [==============================] - 0s 422us/step - loss: 0.0688 - val_loss: 0.0766\n",
+ "Epoch 10/10\n",
+ "990/990 [==============================] - 0s 414us/step - loss: 0.0658 - val_loss: 0.0754\n"
+ ]
+ },
+ {
+ "data": {
+ "text/plain": [
+ "<keras.callbacks.History at 0x7fcd0db7bf98>"
+ ]
+ },
+ "execution_count": 25,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "autoencoder.fit(densities, densities, \n",
+ " epochs = 10,\n",
+ " verbose = 1,\n",
+ " validation_data = (vali_data, vali_data),\n",
+ " shuffle = True)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": []
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "Python 3",
+ "language": "python",
+ "name": "python3"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.6.8"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}