Skip to content

Instantly share code, notes, and snippets.

@ishikota

ishikota/KerasTutorial.ipynb

Created June 29, 2016 07:41
Show Gist options
  • Save ishikota/4f965436efb6b566cd51d04fcd9c3c12 to your computer and use it in GitHub Desktop.
Save ishikota/4f965436efb6b566cd51d04fcd9c3c12 to your computer and use it in GitHub Desktop.
Download ZIP
Simple sin fitting by Keras
Display the source blob
Display the rendered blob
Raw
KerasTutorial.ipynb
{
"cells": [
{
"cell_type": "code",
"execution_count": 20,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"%matplotlib inline\n",
"\n",
"import seaborn\n",
"import numpy as np\n",
"import matplotlib.pyplot as plt\n",
"\n",
"X = np.random.random(1000) * 10 - 5\n",
"Y = np.sin(X)\n",
"\n",
"training_num = 800\n",
"idx = np.arange(1000)\n",
"np.random.shuffle(idx)\n",
"training_idx = idx[:training_num]\n",
"test_idx = idx[training_num:]\n",
"\n",
"trainingX, testX = X[training_idx], X[test_idx]\n",
"trainingY, testY = np.sin(trainingX), np.sin(testX)"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"____________________________________________________________________________________________________\n",
"Layer (type) Output Shape Param # Connected to \n",
"====================================================================================================\n",
"dense_1 (Dense) (None, 30) 60 dense_input_1[0][0] \n",
"____________________________________________________________________________________________________\n",
"activation_1 (Activation) (None, 30) 0 dense_1[0][0] \n",
"____________________________________________________________________________________________________\n",
"dense_2 (Dense) (None, 30) 930 activation_1[0][0] \n",
"____________________________________________________________________________________________________\n",
"activation_2 (Activation) (None, 30) 0 dense_2[0][0] \n",
"____________________________________________________________________________________________________\n",
"dense_3 (Dense) (None, 1) 31 activation_2[0][0] \n",
"====================================================================================================\n",
"Total params: 1021\n",
"____________________________________________________________________________________________________\n"
]
}
],
"source": [
"from keras.models import Sequential\n",
"from keras.layers.core import Dense, Activation\n",
"\n",
"model = Sequential()\n",
"model.add(Dense(30, input_shape=(1,)))\n",
"model.add(Activation(\"sigmoid\"))\n",
"model.add(Dense(30))\n",
"model.add(Activation(\"sigmoid\"))\n",
"model.add(Dense(1))\n",
"model.compile(loss=\"mse\", optimizer=\"sgd\")\n",
"model.summary()"
]
},
{
"cell_type": "code",
"execution_count": 24,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Train on 720 samples, validate on 80 samples\n",
"Epoch 1/1000\n",
"0s - loss: 0.0104 - val_loss: 0.0109\n",
"Epoch 2/1000\n",
"0s - loss: 0.0104 - val_loss: 0.0108\n",
"Epoch 3/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 4/1000\n",
"0s - loss: 0.0104 - val_loss: 0.0108\n",
"Epoch 5/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 6/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 7/1000\n",
"0s - loss: 0.0104 - val_loss: 0.0108\n",
"Epoch 8/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 9/1000\n",
"0s - loss: 0.0104 - val_loss: 0.0108\n",
"Epoch 10/1000\n",
"0s - loss: 0.0104 - val_loss: 0.0108\n",
"Epoch 11/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 12/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 13/1000\n",
"0s - loss: 0.0104 - val_loss: 0.0108\n",
"Epoch 14/1000\n",
"0s - loss: 0.0104 - val_loss: 0.0108\n",
"Epoch 15/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 16/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 17/1000\n",
"0s - loss: 0.0104 - val_loss: 0.0108\n",
"Epoch 18/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 19/1000\n",
"0s - loss: 0.0104 - val_loss: 0.0108\n",
"Epoch 20/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 21/1000\n",
"0s - loss: 0.0104 - val_loss: 0.0108\n",
"Epoch 22/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 23/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 24/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 25/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 26/1000\n",
"0s - loss: 0.0104 - val_loss: 0.0108\n",
"Epoch 27/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 28/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 29/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0107\n",
"Epoch 30/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 31/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 32/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 33/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 34/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 35/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 36/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 37/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 38/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 39/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 40/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 41/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 42/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 43/1000\n",
"0s - loss: 0.0104 - val_loss: 0.0108\n",
"Epoch 44/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 45/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 46/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 47/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 48/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 49/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 50/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 51/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 52/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 53/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 54/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 55/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0109\n",
"Epoch 56/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 57/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0107\n",
"Epoch 58/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 59/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 60/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 61/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 62/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 63/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 64/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 65/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 66/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 67/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 68/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 69/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0107\n",
"Epoch 70/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0107\n",
"Epoch 71/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0110\n",
"Epoch 72/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 73/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 74/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 75/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 76/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 77/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 78/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 79/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 80/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0107\n",
"Epoch 81/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 82/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 83/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 84/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0107\n",
"Epoch 85/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 86/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 87/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 88/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 89/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 90/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 91/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0107\n",
"Epoch 92/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0108\n",
"Epoch 93/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 94/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 95/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0107\n",
"Epoch 96/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0107\n",
"Epoch 97/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0107\n",
"Epoch 98/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 99/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 100/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0107\n",
"Epoch 101/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 102/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 103/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 104/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0108\n",
"Epoch 105/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 106/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 107/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0108\n",
"Epoch 108/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 109/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0107\n",
"Epoch 110/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0107\n",
"Epoch 111/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0108\n",
"Epoch 112/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0108\n",
"Epoch 113/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 114/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0108\n",
"Epoch 115/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 116/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0107\n",
"Epoch 117/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0108\n",
"Epoch 118/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0107\n",
"Epoch 119/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0108\n",
"Epoch 120/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0107\n",
"Epoch 121/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0107\n",
"Epoch 122/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0107\n",
"Epoch 123/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0107\n",
"Epoch 124/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0108\n",
"Epoch 125/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0107\n",
"Epoch 126/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 127/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0108\n",
"Epoch 128/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 129/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0107\n",
"Epoch 130/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 131/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 132/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0107\n",
"Epoch 133/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 134/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 135/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 136/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 137/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 138/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0108\n",
"Epoch 139/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0109\n",
"Epoch 140/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 141/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 142/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0107\n",
"Epoch 143/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 144/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0108\n",
"Epoch 145/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 146/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 147/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 148/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 149/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 150/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0108\n",
"Epoch 151/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 152/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 153/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 154/1000\n",
"0s - loss: 0.0104 - val_loss: 0.0108\n",
"Epoch 155/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 156/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0108\n",
"Epoch 157/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 158/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 159/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 160/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 161/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 162/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0108\n",
"Epoch 163/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0107\n",
"Epoch 164/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0108\n",
"Epoch 165/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 166/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 167/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0108\n",
"Epoch 168/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 169/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0108\n",
"Epoch 170/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 171/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 172/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 173/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 174/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 175/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 176/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 177/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 178/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 179/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 180/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0108\n",
"Epoch 181/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 182/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 183/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 184/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 185/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 186/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 187/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 188/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 189/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 190/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 191/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 192/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 193/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 194/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0108\n",
"Epoch 195/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 196/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 197/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 198/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 199/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 200/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 201/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 202/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 203/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 204/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 205/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 206/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 207/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 208/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 209/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 210/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 211/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 212/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 213/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 214/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0108\n",
"Epoch 215/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 216/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 217/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0108\n",
"Epoch 218/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 219/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 220/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 221/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 222/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 223/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 224/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 225/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0108\n",
"Epoch 226/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 227/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 228/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 229/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 230/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 231/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 232/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 233/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 234/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 235/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 236/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 237/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 238/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 239/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 240/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 241/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 242/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 243/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 244/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 245/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 246/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 247/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0108\n",
"Epoch 248/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 249/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 250/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 251/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 252/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 253/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 254/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 255/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 256/1000\n",
"0s - loss: 0.0103 - val_loss: 0.0107\n",
"Epoch 257/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 258/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 259/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0108\n",
"Epoch 260/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 261/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 262/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0108\n",
"Epoch 263/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 264/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 265/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 266/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 267/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 268/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 269/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 270/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 271/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 272/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 273/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0108\n",
"Epoch 274/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 275/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0108\n",
"Epoch 276/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 277/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 278/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 279/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 280/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 281/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 282/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 283/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 284/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 285/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 286/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 287/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 288/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 289/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 290/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 291/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 292/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 293/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 294/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 295/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 296/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 297/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 298/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 299/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 300/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 301/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 302/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 303/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 304/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 305/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 306/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0108\n",
"Epoch 307/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0107\n",
"Epoch 308/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 309/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 310/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 311/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 312/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 313/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 314/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 315/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 316/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 317/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 318/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 319/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 320/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 321/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 322/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 323/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 324/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 325/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 326/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 327/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 328/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 329/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 330/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 331/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 332/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 333/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 334/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 335/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 336/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 337/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 338/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 339/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 340/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 341/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 342/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 343/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 344/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 345/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 346/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 347/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 348/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 349/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 350/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 351/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 352/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 353/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 354/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 355/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 356/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 357/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 358/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 359/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 360/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 361/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 362/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 363/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 364/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 365/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 366/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 367/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0107\n",
"Epoch 368/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 369/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0110\n",
"Epoch 370/1000\n",
"0s - loss: 0.0102 - val_loss: 0.0106\n",
"Epoch 371/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 372/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 373/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 374/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 375/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 376/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 377/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 378/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 379/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 380/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 381/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 382/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 383/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 384/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 385/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 386/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 387/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 388/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 389/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 390/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 391/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 392/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 393/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 394/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 395/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 396/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 397/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 398/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 399/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0107\n",
"Epoch 400/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 401/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 402/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 403/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 404/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 405/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0107\n",
"Epoch 406/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 407/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0107\n",
"Epoch 408/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 409/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 410/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 411/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 412/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 413/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 414/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 415/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 416/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0107\n",
"Epoch 417/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 418/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 419/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 420/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 421/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 422/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 423/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 424/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 425/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 426/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 427/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 428/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0107\n",
"Epoch 429/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 430/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 431/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 432/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 433/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 434/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0107\n",
"Epoch 435/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 436/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 437/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 438/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 439/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 440/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 441/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 442/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 443/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 444/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 445/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 446/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 447/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 448/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 449/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 450/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 451/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 452/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 453/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 454/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 455/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 456/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 457/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 458/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 459/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 460/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 461/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 462/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 463/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 464/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 465/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 466/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 467/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 468/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 469/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 470/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 471/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 472/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 473/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 474/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 475/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 476/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0107\n",
"Epoch 477/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 478/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 479/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 480/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 481/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 482/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 483/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 484/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 485/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 486/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 487/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 488/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 489/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 490/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 491/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0108\n",
"Epoch 492/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 493/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 494/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 495/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 496/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 497/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 498/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 499/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 500/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 501/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 502/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 503/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0105\n",
"Epoch 504/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 505/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0105\n",
"Epoch 506/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 507/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 508/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 509/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 510/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0105\n",
"Epoch 511/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 512/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0105\n",
"Epoch 513/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 514/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 515/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 516/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 517/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 518/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0105\n",
"Epoch 519/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 520/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 521/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 522/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 523/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0105\n",
"Epoch 524/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 525/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 526/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 527/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 528/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 529/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 530/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 531/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 532/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 533/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 534/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 535/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 536/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 537/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0105\n",
"Epoch 538/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 539/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0105\n",
"Epoch 540/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 541/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0105\n",
"Epoch 542/1000\n",
"0s - loss: 0.0101 - val_loss: 0.0106\n",
"Epoch 543/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 544/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0107\n",
"Epoch 545/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 546/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 547/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 548/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 549/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 550/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 551/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 552/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0106\n",
"Epoch 553/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 554/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0105\n",
"Epoch 555/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 556/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 557/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0105\n",
"Epoch 558/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 559/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0105\n",
"Epoch 560/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 561/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 562/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 563/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0105\n",
"Epoch 564/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 565/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 566/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0106\n",
"Epoch 567/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0105\n",
"Epoch 568/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 569/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0105\n",
"Epoch 570/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0106\n",
"Epoch 571/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 572/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0107\n",
"Epoch 573/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0106\n",
"Epoch 574/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 575/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0106\n",
"Epoch 576/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 577/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0106\n",
"Epoch 578/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0105\n",
"Epoch 579/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0106\n",
"Epoch 580/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0106\n",
"Epoch 581/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0105\n",
"Epoch 582/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0105\n",
"Epoch 583/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0106\n",
"Epoch 584/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0105\n",
"Epoch 585/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 586/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0105\n",
"Epoch 587/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 588/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 589/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 590/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 591/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0105\n",
"Epoch 592/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 593/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0105\n",
"Epoch 594/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 595/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 596/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0105\n",
"Epoch 597/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 598/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0106\n",
"Epoch 599/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 600/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 601/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 602/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 603/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 604/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 605/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 606/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 607/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0105\n",
"Epoch 608/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0105\n",
"Epoch 609/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0105\n",
"Epoch 610/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0106\n",
"Epoch 611/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 612/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 613/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0106\n",
"Epoch 614/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 615/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 616/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0106\n",
"Epoch 617/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 618/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 619/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 620/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 621/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 622/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0105\n",
"Epoch 623/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 624/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 625/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 626/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 627/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 628/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0107\n",
"Epoch 629/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 630/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 631/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 632/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 633/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 634/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 635/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0106\n",
"Epoch 636/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 637/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 638/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0106\n",
"Epoch 639/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 640/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 641/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0106\n",
"Epoch 642/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 643/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 644/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 645/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 646/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 647/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 648/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0106\n",
"Epoch 649/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 650/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0106\n",
"Epoch 651/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 652/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 653/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 654/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0106\n",
"Epoch 655/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0105\n",
"Epoch 656/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 657/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0106\n",
"Epoch 658/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0105\n",
"Epoch 659/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 660/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 661/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 662/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 663/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 664/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 665/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 666/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 667/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 668/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0106\n",
"Epoch 669/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 670/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 671/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 672/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 673/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 674/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 675/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0105\n",
"Epoch 676/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 677/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 678/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 679/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 680/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 681/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 682/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 683/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0106\n",
"Epoch 684/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 685/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 686/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 687/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 688/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 689/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 690/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0105\n",
"Epoch 691/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 692/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 693/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 694/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 695/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 696/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 697/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 698/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 699/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 700/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 701/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0106\n",
"Epoch 702/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 703/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 704/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 705/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 706/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 707/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 708/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 709/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 710/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 711/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 712/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 713/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0105\n",
"Epoch 714/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 715/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 716/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 717/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 718/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 719/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 720/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 721/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 722/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 723/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 724/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 725/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 726/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 727/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 728/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 729/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 730/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 731/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 732/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 733/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 734/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 735/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 736/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 737/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 738/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 739/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 740/1000\n",
"0s - loss: 0.0100 - val_loss: 0.0105\n",
"Epoch 741/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 742/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 743/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 744/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 745/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 746/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 747/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 748/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 749/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 750/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 751/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 752/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0106\n",
"Epoch 753/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 754/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 755/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 756/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 757/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 758/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 759/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 760/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 761/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 762/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 763/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0104\n",
"Epoch 764/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 765/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 766/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 767/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 768/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0104\n",
"Epoch 769/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 770/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 771/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 772/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 773/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 774/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 775/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 776/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 777/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 778/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 779/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 780/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 781/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 782/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0104\n",
"Epoch 783/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 784/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 785/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 786/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 787/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 788/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 789/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 790/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 791/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 792/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 793/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0104\n",
"Epoch 794/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 795/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 796/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 797/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 798/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 799/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 800/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 801/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0104\n",
"Epoch 802/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 803/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 804/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 805/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 806/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0104\n",
"Epoch 807/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 808/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 809/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 810/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 811/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 812/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 813/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 814/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 815/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 816/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 817/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 818/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 819/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 820/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 821/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 822/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 823/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 824/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 825/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 826/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 827/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 828/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 829/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 830/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 831/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 832/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 833/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 834/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 835/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0105\n",
"Epoch 836/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 837/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 838/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 839/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 840/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 841/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0104\n",
"Epoch 842/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 843/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 844/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 845/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 846/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 847/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 848/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 849/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 850/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 851/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 852/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 853/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 854/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 855/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 856/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 857/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 858/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 859/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 860/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 861/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 862/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 863/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 864/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 865/1000\n",
"0s - loss: 0.0099 - val_loss: 0.0104\n",
"Epoch 866/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 867/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 868/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 869/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 870/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 871/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 872/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 873/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 874/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 875/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 876/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 877/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 878/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 879/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 880/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 881/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 882/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 883/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 884/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 885/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 886/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 887/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 888/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 889/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 890/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 891/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 892/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 893/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 894/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 895/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 896/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 897/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 898/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 899/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 900/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 901/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 902/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 903/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 904/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 905/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 906/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 907/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 908/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 909/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 910/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 911/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 912/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 913/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 914/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 915/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 916/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 917/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 918/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 919/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 920/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 921/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 922/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 923/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 924/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 925/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 926/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 927/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 928/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 929/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 930/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 931/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 932/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 933/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 934/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 935/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 936/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 937/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 938/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 939/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 940/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 941/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 942/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 943/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 944/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 945/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 946/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 947/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 948/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 949/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 950/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 951/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 952/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 953/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 954/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 955/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 956/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 957/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 958/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 959/1000\n",
"0s - loss: 0.0097 - val_loss: 0.0104\n",
"Epoch 960/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 961/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 962/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 963/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 964/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0105\n",
"Epoch 965/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 966/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 967/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 968/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 969/1000\n",
"0s - loss: 0.0097 - val_loss: 0.0104\n",
"Epoch 970/1000\n",
"0s - loss: 0.0097 - val_loss: 0.0104\n",
"Epoch 971/1000\n",
"0s - loss: 0.0097 - val_loss: 0.0104\n",
"Epoch 972/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 973/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 974/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 975/1000\n",
"0s - loss: 0.0097 - val_loss: 0.0104\n",
"Epoch 976/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 977/1000\n",
"0s - loss: 0.0097 - val_loss: 0.0104\n",
"Epoch 978/1000\n",
"0s - loss: 0.0097 - val_loss: 0.0104\n",
"Epoch 979/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 980/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 981/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 982/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 983/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 984/1000\n",
"0s - loss: 0.0097 - val_loss: 0.0104\n",
"Epoch 985/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 986/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 987/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 988/1000\n",
"0s - loss: 0.0097 - val_loss: 0.0105\n",
"Epoch 989/1000\n",
"0s - loss: 0.0097 - val_loss: 0.0104\n",
"Epoch 990/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 991/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 992/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 993/1000\n",
"0s - loss: 0.0097 - val_loss: 0.0104\n",
"Epoch 994/1000\n",
"0s - loss: 0.0097 - val_loss: 0.0104\n",
"Epoch 995/1000\n",
"0s - loss: 0.0097 - val_loss: 0.0104\n",
"Epoch 996/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 997/1000\n",
"0s - loss: 0.0097 - val_loss: 0.0104\n",
"Epoch 998/1000\n",
"0s - loss: 0.0097 - val_loss: 0.0104\n",
"Epoch 999/1000\n",
"0s - loss: 0.0098 - val_loss: 0.0104\n",
"Epoch 1000/1000\n",
"0s - loss: 0.0097 - val_loss: 0.0104\n"
]
}
],
"source": [
"history = model.fit(trainingX, trainingY, batch_size=128, nb_epoch=1000, validation_split=0.1, verbose=2)"
]
},
{
"cell_type": "code",
"execution_count": 25,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"[<matplotlib.lines.Line2D at 0x116747210>]"
]
},
"execution_count": 25,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAfoAAAFXCAYAAABKl4x5AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nIeax/wFZZeYggFtXUwAAAABJRU5ErkJggg==\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x114501510>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"predictY = model.predict(testX)\n",
"plt.subplot(121)\n",
"plt.plot(np.arange(1000), history.history[\"loss\"], color=\"r\", alpha=0.3, label=\"loss\")\n",
"plt.plot(np.arange(1000), history.history[\"val_loss\"], color=\"b\", alpha=0.3, label=\"val_loss\")\n",
"plt.subplot(122)\n",
"plt.plot(testX, predictY, \"bo\", alpha=0.3)\n",
"plt.plot(testX, testY, \"ro\", alpha=0.3)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": []
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 2",
"language": "python",
"name": "python2"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 2
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython2",
"version": "2.7.11"
}
},
"nbformat": 4,
"nbformat_minor": 0
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment