Skip to content

Instantly share code, notes, and snippets.

@oraoto

oraoto/EffNet-Keras.ipynb

Created March 22, 2018 08:36
Show Gist options
  • Save oraoto/63efd4145d7a875d3581b5b346ae63fb to your computer and use it in GitHub Desktop.
Save oraoto/63efd4145d7a875d3581b5b346ae63fb to your computer and use it in GitHub Desktop.
Download ZIP
Display the source blob
Display the rendered blob
Raw
EffNet-Keras.ipynb
{
"cells": [
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"c:\\users\\oraoto\\.virtualenvs\\ml\\lib\\site-packages\\h5py\\__init__.py:36: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`.\n",
" from ._conv import register_converters as _register_converters\n",
"Using TensorFlow backend.\n"
]
}
],
"source": [
"import keras.layers as layers\n",
"from keras.models import Sequential\n",
"from keras.optimizers import Adam\n",
"from keras.datasets import cifar10\n",
"from keras.utils import to_categorical\n",
"import numpy as np"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {},
"outputs": [],
"source": [
"effNet = Sequential()\n",
"effNet.add(layers.InputLayer((32, 32, 3)))\n",
"\n",
"effNet.add(layers.Conv2D(32, (1,1), use_bias=False))\n",
"effNet.add(layers.BatchNormalization())\n",
"effNet.add(layers.Activation('relu'))\n",
"effNet.add(layers.DepthwiseConv2D((1, 3), padding='same', use_bias=False))\n",
"effNet.add(layers.BatchNormalization())\n",
"effNet.add(layers.Activation('relu'))\n",
"effNet.add(layers.MaxPool2D((1,2), strides=(1, 2)))\n",
"effNet.add(layers.DepthwiseConv2D((3, 1), padding='same', use_bias=False))\n",
"effNet.add(layers.BatchNormalization())\n",
"effNet.add(layers.Activation('relu'))\n",
"effNet.add(layers.Conv2D(64, (2, 1), strides=(2, 1), use_bias=False))\n",
"effNet.add(layers.BatchNormalization())\n",
"effNet.add(layers.Activation('relu'))\n",
"\n",
"effNet.add(layers.Conv2D(64, (1,1), use_bias=False))\n",
"effNet.add(layers.BatchNormalization())\n",
"effNet.add(layers.Activation('relu'))\n",
"effNet.add(layers.DepthwiseConv2D((1, 3), padding='same', use_bias=False))\n",
"effNet.add(layers.BatchNormalization())\n",
"effNet.add(layers.Activation('relu'))\n",
"effNet.add(layers.MaxPool2D((1,2), strides=(1, 2)))\n",
"effNet.add(layers.DepthwiseConv2D((3, 1), padding='same', use_bias=False))\n",
"effNet.add(layers.BatchNormalization())\n",
"effNet.add(layers.Activation('relu'))\n",
"effNet.add(layers.Conv2D(128, (2, 1), strides=(2, 1), use_bias=False))\n",
"effNet.add(layers.BatchNormalization())\n",
"effNet.add(layers.Activation('relu'))\n",
"\n",
"effNet.add(layers.Conv2D(128, (1,1), use_bias=False))\n",
"effNet.add(layers.BatchNormalization())\n",
"effNet.add(layers.Activation('relu'))\n",
"effNet.add(layers.DepthwiseConv2D((1, 3), padding='same', use_bias=False))\n",
"effNet.add(layers.BatchNormalization())\n",
"effNet.add(layers.Activation('relu'))\n",
"effNet.add(layers.MaxPool2D((1,2), strides=(1, 2)))\n",
"effNet.add(layers.DepthwiseConv2D((3, 1), padding='same', use_bias=False))\n",
"effNet.add(layers.BatchNormalization())\n",
"effNet.add(layers.Activation('relu'))\n",
"effNet.add(layers.Conv2D(256, (2, 1), strides=(2, 1), use_bias=False))\n",
"effNet.add(layers.BatchNormalization())\n",
"effNet.add(layers.Activation('relu'))\n",
"\n",
"effNet.add(layers.Flatten())\n",
"effNet.add(layers.Dense(10))\n",
"effNet.add(layers.Activation('softmax'))"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"_________________________________________________________________\n",
"Layer (type) Output Shape Param # \n",
"=================================================================\n",
"input_2 (InputLayer) (None, 32, 32, 3) 0 \n",
"_________________________________________________________________\n",
"conv2d_7 (Conv2D) (None, 32, 32, 32) 96 \n",
"_________________________________________________________________\n",
"batch_normalization_14 (Batc (None, 32, 32, 32) 128 \n",
"_________________________________________________________________\n",
"activation_14 (Activation) (None, 32, 32, 32) 0 \n",
"_________________________________________________________________\n",
"depthwise_conv2d_7 (Depthwis (None, 32, 32, 32) 96 \n",
"_________________________________________________________________\n",
"batch_normalization_15 (Batc (None, 32, 32, 32) 128 \n",
"_________________________________________________________________\n",
"activation_15 (Activation) (None, 32, 32, 32) 0 \n",
"_________________________________________________________________\n",
"max_pooling2d_4 (MaxPooling2 (None, 32, 16, 32) 0 \n",
"_________________________________________________________________\n",
"depthwise_conv2d_8 (Depthwis (None, 32, 16, 32) 96 \n",
"_________________________________________________________________\n",
"batch_normalization_16 (Batc (None, 32, 16, 32) 128 \n",
"_________________________________________________________________\n",
"activation_16 (Activation) (None, 32, 16, 32) 0 \n",
"_________________________________________________________________\n",
"conv2d_8 (Conv2D) (None, 16, 16, 64) 4096 \n",
"_________________________________________________________________\n",
"batch_normalization_17 (Batc (None, 16, 16, 64) 256 \n",
"_________________________________________________________________\n",
"activation_17 (Activation) (None, 16, 16, 64) 0 \n",
"_________________________________________________________________\n",
"conv2d_9 (Conv2D) (None, 16, 16, 64) 4096 \n",
"_________________________________________________________________\n",
"batch_normalization_18 (Batc (None, 16, 16, 64) 256 \n",
"_________________________________________________________________\n",
"activation_18 (Activation) (None, 16, 16, 64) 0 \n",
"_________________________________________________________________\n",
"depthwise_conv2d_9 (Depthwis (None, 16, 16, 64) 192 \n",
"_________________________________________________________________\n",
"batch_normalization_19 (Batc (None, 16, 16, 64) 256 \n",
"_________________________________________________________________\n",
"activation_19 (Activation) (None, 16, 16, 64) 0 \n",
"_________________________________________________________________\n",
"max_pooling2d_5 (MaxPooling2 (None, 16, 8, 64) 0 \n",
"_________________________________________________________________\n",
"depthwise_conv2d_10 (Depthwi (None, 16, 8, 64) 192 \n",
"_________________________________________________________________\n",
"batch_normalization_20 (Batc (None, 16, 8, 64) 256 \n",
"_________________________________________________________________\n",
"activation_20 (Activation) (None, 16, 8, 64) 0 \n",
"_________________________________________________________________\n",
"conv2d_10 (Conv2D) (None, 8, 8, 128) 16384 \n",
"_________________________________________________________________\n",
"batch_normalization_21 (Batc (None, 8, 8, 128) 512 \n",
"_________________________________________________________________\n",
"activation_21 (Activation) (None, 8, 8, 128) 0 \n",
"_________________________________________________________________\n",
"conv2d_11 (Conv2D) (None, 8, 8, 128) 16384 \n",
"_________________________________________________________________\n",
"batch_normalization_22 (Batc (None, 8, 8, 128) 512 \n",
"_________________________________________________________________\n",
"activation_22 (Activation) (None, 8, 8, 128) 0 \n",
"_________________________________________________________________\n",
"depthwise_conv2d_11 (Depthwi (None, 8, 8, 128) 384 \n",
"_________________________________________________________________\n",
"batch_normalization_23 (Batc (None, 8, 8, 128) 512 \n",
"_________________________________________________________________\n",
"activation_23 (Activation) (None, 8, 8, 128) 0 \n",
"_________________________________________________________________\n",
"max_pooling2d_6 (MaxPooling2 (None, 8, 4, 128) 0 \n",
"_________________________________________________________________\n",
"depthwise_conv2d_12 (Depthwi (None, 8, 4, 128) 384 \n",
"_________________________________________________________________\n",
"batch_normalization_24 (Batc (None, 8, 4, 128) 512 \n",
"_________________________________________________________________\n",
"activation_24 (Activation) (None, 8, 4, 128) 0 \n",
"_________________________________________________________________\n",
"conv2d_12 (Conv2D) (None, 4, 4, 256) 65536 \n",
"_________________________________________________________________\n",
"batch_normalization_25 (Batc (None, 4, 4, 256) 1024 \n",
"_________________________________________________________________\n",
"activation_25 (Activation) (None, 4, 4, 256) 0 \n",
"_________________________________________________________________\n",
"flatten_2 (Flatten) (None, 4096) 0 \n",
"_________________________________________________________________\n",
"dense_2 (Dense) (None, 10) 40970 \n",
"_________________________________________________________________\n",
"activation_26 (Activation) (None, 10) 0 \n",
"=================================================================\n",
"Total params: 153,386\n",
"Trainable params: 151,146\n",
"Non-trainable params: 2,240\n",
"_________________________________________________________________\n"
]
}
],
"source": [
"effNet.summary()"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {},
"outputs": [],
"source": [
"effNet.compile(loss='categorical_crossentropy', optimizer=Adam(lr=0.001,beta_1=0.75), metrics=['accuracy'])"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {},
"outputs": [],
"source": [
"(train_x, train_y), (test_x, test_y) = cifar10.load_data()\n",
"train_x = train_x.astype(np.float32) / 255.\n",
"test_x = test_x.astype(np.float32) / 255.\n",
"\n",
"train_y = to_categorical(train_y)\n",
"test_y = to_categorical(test_y)"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {
"scrolled": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Train on 47500 samples, validate on 2500 samples\n",
"Epoch 1/50\n",
"47500/47500 [==============================] - 39s 811us/step - loss: 1.4626 - acc: 0.4844 - val_loss: 1.3037 - val_acc: 0.5504\n",
"Epoch 2/50\n",
"47500/47500 [==============================] - 36s 765us/step - loss: 1.0664 - acc: 0.6258 - val_loss: 1.1230 - val_acc: 0.6192\n",
"Epoch 3/50\n",
"47500/47500 [==============================] - 36s 766us/step - loss: 0.8951 - acc: 0.6857 - val_loss: 1.0444 - val_acc: 0.6528\n",
"Epoch 4/50\n",
"47500/47500 [==============================] - 36s 767us/step - loss: 0.7871 - acc: 0.7233 - val_loss: 1.0146 - val_acc: 0.6688\n",
"Epoch 5/50\n",
"47500/47500 [==============================] - 37s 769us/step - loss: 0.7030 - acc: 0.7506 - val_loss: 0.9509 - val_acc: 0.6860\n",
"Epoch 6/50\n",
"47500/47500 [==============================] - 36s 767us/step - loss: 0.6231 - acc: 0.7800 - val_loss: 0.9570 - val_acc: 0.7024\n",
"Epoch 7/50\n",
"47500/47500 [==============================] - 36s 767us/step - loss: 0.5593 - acc: 0.8020 - val_loss: 0.9414 - val_acc: 0.6904\n",
"Epoch 8/50\n",
"47500/47500 [==============================] - 36s 767us/step - loss: 0.4982 - acc: 0.8263 - val_loss: 0.9735 - val_acc: 0.7064\n",
"Epoch 9/50\n",
"47500/47500 [==============================] - 36s 767us/step - loss: 0.4409 - acc: 0.8449 - val_loss: 1.0570 - val_acc: 0.6812\n",
"Epoch 10/50\n",
"47500/47500 [==============================] - 36s 767us/step - loss: 0.3921 - acc: 0.8606 - val_loss: 1.0248 - val_acc: 0.7120\n",
"Epoch 11/50\n",
"47500/47500 [==============================] - 36s 767us/step - loss: 0.3464 - acc: 0.8781 - val_loss: 1.1533 - val_acc: 0.6884\n",
"Epoch 12/50\n",
"47500/47500 [==============================] - 37s 769us/step - loss: 0.3079 - acc: 0.8909 - val_loss: 1.0387 - val_acc: 0.7160\n",
"Epoch 13/50\n",
"47500/47500 [==============================] - 36s 767us/step - loss: 0.2738 - acc: 0.9024 - val_loss: 1.1322 - val_acc: 0.7088\n",
"Epoch 14/50\n",
"47500/47500 [==============================] - 36s 767us/step - loss: 0.2501 - acc: 0.9116 - val_loss: 1.1958 - val_acc: 0.6992\n",
"Epoch 15/50\n",
"47500/47500 [==============================] - 36s 767us/step - loss: 0.2138 - acc: 0.9248 - val_loss: 1.2758 - val_acc: 0.6788\n",
"Epoch 16/50\n",
"47500/47500 [==============================] - 36s 767us/step - loss: 0.2091 - acc: 0.9251 - val_loss: 1.2653 - val_acc: 0.6968\n",
"Epoch 17/50\n",
"47500/47500 [==============================] - 36s 766us/step - loss: 0.1868 - acc: 0.9348 - val_loss: 1.3214 - val_acc: 0.6980\n",
"Epoch 18/50\n",
"47500/47500 [==============================] - 36s 766us/step - loss: 0.1654 - acc: 0.9420 - val_loss: 1.3099 - val_acc: 0.7016\n",
"Epoch 19/50\n",
"47500/47500 [==============================] - 37s 773us/step - loss: 0.1559 - acc: 0.9453 - val_loss: 1.3823 - val_acc: 0.6972\n",
"Epoch 20/50\n",
"47500/47500 [==============================] - 37s 770us/step - loss: 0.1566 - acc: 0.9430 - val_loss: 1.3759 - val_acc: 0.7032\n",
"Epoch 21/50\n",
"47500/47500 [==============================] - 37s 769us/step - loss: 0.1418 - acc: 0.9496 - val_loss: 1.4486 - val_acc: 0.6924\n",
"Epoch 22/50\n",
"47500/47500 [==============================] - 37s 772us/step - loss: 0.1377 - acc: 0.9502 - val_loss: 1.4630 - val_acc: 0.6980\n",
"Epoch 23/50\n",
"47500/47500 [==============================] - 37s 770us/step - loss: 0.1169 - acc: 0.9584 - val_loss: 1.4023 - val_acc: 0.7048\n",
"Epoch 24/50\n",
"47500/47500 [==============================] - 36s 767us/step - loss: 0.1224 - acc: 0.9563 - val_loss: 1.5270 - val_acc: 0.7048\n",
"Epoch 25/50\n",
"47500/47500 [==============================] - 36s 768us/step - loss: 0.1120 - acc: 0.9596 - val_loss: 1.4748 - val_acc: 0.6996\n",
"Epoch 26/50\n",
"47500/47500 [==============================] - 37s 769us/step - loss: 0.1138 - acc: 0.9594 - val_loss: 1.5078 - val_acc: 0.7016\n",
"Epoch 27/50\n",
" 6016/47500 [==>...........................] - ETA: 31s - loss: 0.1028 - acc: 0.9621"
]
},
{
"ename": "KeyboardInterrupt",
"evalue": "",
"output_type": "error",
"traceback": [
"\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
"\u001b[1;31mKeyboardInterrupt\u001b[0m Traceback (most recent call last)",
"\u001b[1;32m<ipython-input-10-2c75c6906c37>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m()\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[0mhistory\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0meffNet\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mfit\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mtrain_x\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mtrain_y\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mbatch_size\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;36m64\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mepochs\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;36m50\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mvalidation_split\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;36m0.05\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m",
"\u001b[1;32mc:\\users\\oraoto\\.virtualenvs\\ml\\lib\\site-packages\\keras\\models.py\u001b[0m in \u001b[0;36mfit\u001b[1;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[0;32m 961\u001b[0m \u001b[0minitial_epoch\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0minitial_epoch\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m 962\u001b[0m \u001b[0msteps_per_epoch\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0msteps_per_epoch\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 963\u001b[1;33m validation_steps=validation_steps)\n\u001b[0m\u001b[0;32m 964\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m 965\u001b[0m def evaluate(self, x=None, y=None,\n",
"\u001b[1;32mc:\\users\\oraoto\\.virtualenvs\\ml\\lib\\site-packages\\keras\\engine\\training.py\u001b[0m in \u001b[0;36mfit\u001b[1;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[0;32m 1703\u001b[0m \u001b[0minitial_epoch\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0minitial_epoch\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m 1704\u001b[0m \u001b[0msteps_per_epoch\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0msteps_per_epoch\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 1705\u001b[1;33m validation_steps=validation_steps)\n\u001b[0m\u001b[0;32m 1706\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m 1707\u001b[0m def evaluate(self, x=None, y=None,\n",
"\u001b[1;32mc:\\users\\oraoto\\.virtualenvs\\ml\\lib\\site-packages\\keras\\engine\\training.py\u001b[0m in \u001b[0;36m_fit_loop\u001b[1;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[0;32m 1233\u001b[0m \u001b[0mins_batch\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mi\u001b[0m\u001b[1;33m]\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mins_batch\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mi\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mtoarray\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m 1234\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 1235\u001b[1;33m \u001b[0mouts\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mf\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mins_batch\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m 1236\u001b[0m \u001b[1;32mif\u001b[0m \u001b[1;32mnot\u001b[0m \u001b[0misinstance\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mouts\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mlist\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m 1237\u001b[0m \u001b[0mouts\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;33m[\u001b[0m\u001b[0mouts\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
"\u001b[1;32mc:\\users\\oraoto\\.virtualenvs\\ml\\lib\\site-packages\\keras\\backend\\tensorflow_backend.py\u001b[0m in \u001b[0;36m__call__\u001b[1;34m(self, inputs)\u001b[0m\n\u001b[0;32m 2476\u001b[0m \u001b[0msession\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mget_session\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m 2477\u001b[0m updated = session.run(fetches=fetches, feed_dict=feed_dict,\n\u001b[1;32m-> 2478\u001b[1;33m **self.session_kwargs)\n\u001b[0m\u001b[0;32m 2479\u001b[0m \u001b[1;32mreturn\u001b[0m \u001b[0mupdated\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;33m:\u001b[0m\u001b[0mlen\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0moutputs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m 2480\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
"\u001b[1;32mc:\\users\\oraoto\\.virtualenvs\\ml\\lib\\site-packages\\tensorflow\\python\\client\\session.py\u001b[0m in \u001b[0;36mrun\u001b[1;34m(self, fetches, feed_dict, options, run_metadata)\u001b[0m\n\u001b[0;32m 903\u001b[0m \u001b[1;32mtry\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m 904\u001b[0m result = self._run(None, fetches, feed_dict, options_ptr,\n\u001b[1;32m--> 905\u001b[1;33m run_metadata_ptr)\n\u001b[0m\u001b[0;32m 906\u001b[0m \u001b[1;32mif\u001b[0m \u001b[0mrun_metadata\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m 907\u001b[0m \u001b[0mproto_data\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mtf_session\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mTF_GetBuffer\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mrun_metadata_ptr\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
"\u001b[1;32mc:\\users\\oraoto\\.virtualenvs\\ml\\lib\\site-packages\\tensorflow\\python\\client\\session.py\u001b[0m in \u001b[0;36m_run\u001b[1;34m(self, handle, fetches, feed_dict, options, run_metadata)\u001b[0m\n\u001b[0;32m 1135\u001b[0m \u001b[1;32mif\u001b[0m \u001b[0mfinal_fetches\u001b[0m \u001b[1;32mor\u001b[0m \u001b[0mfinal_targets\u001b[0m \u001b[1;32mor\u001b[0m \u001b[1;33m(\u001b[0m\u001b[0mhandle\u001b[0m \u001b[1;32mand\u001b[0m \u001b[0mfeed_dict_tensor\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m 1136\u001b[0m results = self._do_run(handle, final_targets, final_fetches,\n\u001b[1;32m-> 1137\u001b[1;33m feed_dict_tensor, options, run_metadata)\n\u001b[0m\u001b[0;32m 1138\u001b[0m \u001b[1;32melse\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m 1139\u001b[0m \u001b[0mresults\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;33m[\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
"\u001b[1;32mc:\\users\\oraoto\\.virtualenvs\\ml\\lib\\site-packages\\tensorflow\\python\\client\\session.py\u001b[0m in \u001b[0;36m_do_run\u001b[1;34m(self, handle, target_list, fetch_list, feed_dict, options, run_metadata)\u001b[0m\n\u001b[0;32m 1353\u001b[0m \u001b[1;32mif\u001b[0m \u001b[0mhandle\u001b[0m \u001b[1;32mis\u001b[0m \u001b[1;32mNone\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m 1354\u001b[0m return self._do_call(_run_fn, self._session, feeds, fetches, targets,\n\u001b[1;32m-> 1355\u001b[1;33m options, run_metadata)\n\u001b[0m\u001b[0;32m 1356\u001b[0m \u001b[1;32melse\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m 1357\u001b[0m \u001b[1;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_do_call\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0m_prun_fn\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_session\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mhandle\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mfeeds\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mfetches\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
"\u001b[1;32mc:\\users\\oraoto\\.virtualenvs\\ml\\lib\\site-packages\\tensorflow\\python\\client\\session.py\u001b[0m in \u001b[0;36m_do_call\u001b[1;34m(self, fn, *args)\u001b[0m\n\u001b[0;32m 1359\u001b[0m \u001b[1;32mdef\u001b[0m \u001b[0m_do_call\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mfn\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m*\u001b[0m\u001b[0margs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m 1360\u001b[0m \u001b[1;32mtry\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 1361\u001b[1;33m \u001b[1;32mreturn\u001b[0m \u001b[0mfn\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m*\u001b[0m\u001b[0margs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m 1362\u001b[0m \u001b[1;32mexcept\u001b[0m \u001b[0merrors\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mOpError\u001b[0m \u001b[1;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m 1363\u001b[0m \u001b[0mmessage\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mcompat\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mas_text\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0me\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mmessage\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
"\u001b[1;32mc:\\users\\oraoto\\.virtualenvs\\ml\\lib\\site-packages\\tensorflow\\python\\client\\session.py\u001b[0m in \u001b[0;36m_run_fn\u001b[1;34m(session, feed_dict, fetch_list, target_list, options, run_metadata)\u001b[0m\n\u001b[0;32m 1338\u001b[0m \u001b[1;32melse\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m 1339\u001b[0m return tf_session.TF_Run(session, options, feed_dict, fetch_list,\n\u001b[1;32m-> 1340\u001b[1;33m target_list, status, run_metadata)\n\u001b[0m\u001b[0;32m 1341\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m 1342\u001b[0m \u001b[1;32mdef\u001b[0m \u001b[0m_prun_fn\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0msession\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mhandle\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mfeed_dict\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mfetch_list\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
"\u001b[1;31mKeyboardInterrupt\u001b[0m: "
]
}
],
"source": [
"history = effNet.fit(train_x, train_y, batch_size=64, epochs=50, validation_split=0.05)"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"10000/10000 [==============================] - 4s 352us/step\n"
]
},
{
"data": {
"text/plain": [
"[1.5003245077133178, 0.7005]"
]
},
"execution_count": 11,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"effNet.evaluate(test_x, test_y)"
]
},
{
"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": "3.6.2"
}
},
"nbformat": 4,
"nbformat_minor": 2
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment