GPU Practice

Makefile

run_jupyter:
	docker run --runtime=nvidia --rm -d -p 80:8888 -v /notebooks:/notebooks \
		tensorflow/tensorflow:latest-gpu-py3 jupyter notebook \
			--notebook-dir=/notebooks --allow-root --NotebookApp.token=''

mnist_cnn.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.5/dist-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"
     ]
    }
   ],
   "source": [
    "from tensorflow.contrib import keras"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "num_classes = 10"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading data from https://s3.amazonaws.com/img-datasets/mnist.npz\n",
      "11493376/11490434 [==============================]11493376/11490434 [==============================] - 6s 1us/step\n",
      "\n"
     ]
    }
   ],
   "source": [
    "(x_train, y_train), (x_test, y_test) = keras.datasets.mnist.load_data()\n",
    "\n",
    "input_shape = x_train.shape[1:] + (1,)\n",
    "x_train = x_train.reshape(x_train.shape[0], *input_shape).astype('float32') / 0xff\n",
    "x_test = x_test.reshape(x_test.shape[0], *input_shape).astype('float32') / 0xff\n",
    "y_train = keras.utils.to_categorical(y_train, num_classes)\n",
    "y_test = keras.utils.to_categorical(y_test, num_classes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = keras.models.Sequential()\n",
    "model.add(keras.layers.Conv2D(32, kernel_size=(3, 3), activation='relu', input_shape=input_shape))\n",
    "model.add(keras.layers.Conv2D(64, (3, 3), activation='relu'))\n",
    "model.add(keras.layers.MaxPooling2D(pool_size=(2, 2)))\n",
    "model.add(keras.layers.Dropout(0.25))\n",
    "model.add(keras.layers.Flatten())\n",
    "model.add(keras.layers.Dense(128, activation='relu'))\n",
    "model.add(keras.layers.Dropout(0.5))\n",
    "model.add(keras.layers.Dense(num_classes, activation='softmax'))\n",
    "\n",
    "model.compile(loss=keras.losses.categorical_crossentropy,\n",
    "              optimizer=keras.optimizers.Adam(),\n",
    "              metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 60000 samples, validate on 10000 samples\n",
      "Epoch 1/12\n",
      "60000/60000 [==============================]60000/60000 [==============================] - 8s 130us/step - loss: 0.2440 - acc: 0.9263 - val_loss: 0.0503 - val_acc: 0.9843\n",
      "\n",
      "Epoch 2/12\n",
      "60000/60000 [==============================]60000/60000 [==============================] - 5s 81us/step - loss: 0.0862 - acc: 0.9744 - val_loss: 0.0363 - val_acc: 0.9870\n",
      "\n",
      "Epoch 3/12\n",
      "60000/60000 [==============================]60000/60000 [==============================] - 5s 81us/step - loss: 0.0647 - acc: 0.9801 - val_loss: 0.0349 - val_acc: 0.9878\n",
      "\n",
      "Epoch 4/12\n",
      "60000/60000 [==============================]60000/60000 [==============================] - 4s 74us/step - loss: 0.0510 - acc: 0.9846 - val_loss: 0.0304 - val_acc: 0.9901\n",
      "\n",
      "Epoch 5/12\n",
      "60000/60000 [==============================]60000/60000 [==============================] - 5s 76us/step - loss: 0.0434 - acc: 0.9869 - val_loss: 0.0287 - val_acc: 0.9901\n",
      "\n",
      "Epoch 6/12\n",
      "60000/60000 [==============================]60000/60000 [==============================] - 4s 71us/step - loss: 0.0381 - acc: 0.9881 - val_loss: 0.0287 - val_acc: 0.9900\n",
      "\n",
      "Epoch 7/12\n",
      "60000/60000 [==============================]60000/60000 [==============================] - 5s 77us/step - loss: 0.0341 - acc: 0.9893 - val_loss: 0.0298 - val_acc: 0.9900\n",
      "\n",
      "Epoch 8/12\n",
      "60000/60000 [==============================]60000/60000 [==============================] - 4s 71us/step - loss: 0.0292 - acc: 0.9907 - val_loss: 0.0281 - val_acc: 0.9913\n",
      "\n",
      "Epoch 9/12\n",
      "60000/60000 [==============================]60000/60000 [==============================] - 4s 72us/step - loss: 0.0260 - acc: 0.9914 - val_loss: 0.0310 - val_acc: 0.9909\n",
      "\n",
      "Epoch 10/12\n",
      "60000/60000 [==============================]60000/60000 [==============================] - 5s 81us/step - loss: 0.0247 - acc: 0.9919 - val_loss: 0.0283 - val_acc: 0.9915\n",
      "\n",
      "Epoch 11/12\n",
      "60000/60000 [==============================]60000/60000 [==============================] - 5s 77us/step - loss: 0.0224 - acc: 0.9927 - val_loss: 0.0322 - val_acc: 0.9912\n",
      "\n",
      "Epoch 12/12\n",
      "60000/60000 [==============================]60000/60000 [==============================] - 5s 76us/step - loss: 0.0222 - acc: 0.9925 - val_loss: 0.0293 - val_acc: 0.9919\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.keras._impl.keras.callbacks.History at 0x7f3ab55df320>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit(x_train, y_train,\n",
    "          batch_size=128,\n",
    "          epochs=12,\n",
    "          verbose=1,\n",
    "          validation_data=(x_test, y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test loss: 0.02925125805952248\n",
      "Test accuracy: 0.9919\n"
     ]
    }
   ],
   "source": [
    "score = model.evaluate(x_test, y_test, verbose=0)\n",
    "print('Test loss: {}\\nTest accuracy: {}'.format(*score))"
   ]
  }
 ],
 "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": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}