TensorFlow+Keras深度学习人工智能实践应用/19 Tensorflow实现卷积网络做Mnist手写体识别.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "import tensorflow.examples.tutorials.mnist.input_data as input_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From <ipython-input-2-20514739e6d7>:1: read_data_sets (from tensorflow.contrib.learn.python.learn.datasets.mnist) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Please use alternatives such as official/mnist/dataset.py from tensorflow/models.\n",
      "WARNING:tensorflow:From c:\\users\\sha\\anaconda3\\envs\\tensorflow-gpu\\lib\\site-packages\\tensorflow\\contrib\\learn\\python\\learn\\datasets\\mnist.py:260: maybe_download (from tensorflow.contrib.learn.python.learn.datasets.base) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Please write your own downloading logic.\n",
      "WARNING:tensorflow:From c:\\users\\sha\\anaconda3\\envs\\tensorflow-gpu\\lib\\site-packages\\tensorflow\\contrib\\learn\\python\\learn\\datasets\\mnist.py:262: extract_images (from tensorflow.contrib.learn.python.learn.datasets.mnist) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Please use tf.data to implement this functionality.\n",
      "Extracting D:\\data\\train-images-idx3-ubyte.gz\n",
      "WARNING:tensorflow:From c:\\users\\sha\\anaconda3\\envs\\tensorflow-gpu\\lib\\site-packages\\tensorflow\\contrib\\learn\\python\\learn\\datasets\\mnist.py:267: extract_labels (from tensorflow.contrib.learn.python.learn.datasets.mnist) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Please use tf.data to implement this functionality.\n",
      "Extracting D:\\data\\train-labels-idx1-ubyte.gz\n",
      "WARNING:tensorflow:From c:\\users\\sha\\anaconda3\\envs\\tensorflow-gpu\\lib\\site-packages\\tensorflow\\contrib\\learn\\python\\learn\\datasets\\mnist.py:110: dense_to_one_hot (from tensorflow.contrib.learn.python.learn.datasets.mnist) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Please use tf.one_hot on tensors.\n",
      "Extracting D:\\data\\t10k-images-idx3-ubyte.gz\n",
      "Extracting D:\\data\\t10k-labels-idx1-ubyte.gz\n",
      "WARNING:tensorflow:From c:\\users\\sha\\anaconda3\\envs\\tensorflow-gpu\\lib\\site-packages\\tensorflow\\contrib\\learn\\python\\learn\\datasets\\mnist.py:290: DataSet.__init__ (from tensorflow.contrib.learn.python.learn.datasets.mnist) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Please use alternatives such as official/mnist/dataset.py from tensorflow/models.\n"
     ]
    }
   ],
   "source": [
    "mnist = input_data.read_data_sets(r\"D:\\data\",one_hot=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def weight(shape):\n",
    "    W=tf.Variable(tf.truncated_normal(shape,stddev=0.1),name='W')\n",
    "    tf.summary.histogram('W',W)\n",
    "    return W"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "def bias(shape):\n",
    "    b=tf.Variable(tf.constant(0.1,shape=shape),name='b')\n",
    "    tf.summary.histogram('b',b)\n",
    "    return b"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "def conv2d(x,W):\n",
    "    return tf.nn.conv2d(x,W,strides=[1,1,1,1],padding='SAME')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "def max_pool_2x2(x):\n",
    "    return tf.nn.max_pool(x,ksize=[1,2,2,1],strides=[1,2,2,1],padding='SAME')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "with tf.name_scope('Input_Layer'):\n",
    "    x=tf.placeholder(\"float\",shape=[None,784])\n",
    "    x_image=tf.reshape(x,[-1,28,28,1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "with tf.name_scope('C1_Conv'):\n",
    "    W1=weight([5,5,1,16])\n",
    "    b1=bias([16])\n",
    "    Conv1=conv2d(x_image,W1)+b1\n",
    "    C1_Conv=tf.nn.relu(Conv1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "with tf.name_scope('C1_Pool'):\n",
    "    C1_Pool=max_pool_2x2(C1_Conv,)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "with tf.name_scope('C2_Conv'):\n",
    "    W2=weight([5,5,16,36],)\n",
    "    b2=bias([36],)\n",
    "    Conv2=conv2d(C1_Pool,W2)+b2\n",
    "    C2_Conv=tf.nn.relu(Conv2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "with tf.name_scope('C2_Pool'):\n",
    "    C2_Pool=max_pool_2x2(C2_Conv)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "with tf.name_scope('D_Flat'):\n",
    "    D_Flat=tf.reshape(C2_Pool,[-1,1764])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "with tf.name_scope('D_Hidden_Layer'):\n",
    "    W3=weight([1764,128])\n",
    "    b3=bias([128])\n",
    "    D_Hidden=tf.nn.relu(tf.matmul(D_Flat,W3)+b3)\n",
    "    D_Hidden_Dropout=tf.nn.dropout(D_Hidden,keep_prob=0.8)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "with tf.name_scope('Output_Layer'):\n",
    "    W4=weight([128,10])\n",
    "    b4=bias([10])\n",
    "    y_predict=tf.nn.softmax(tf.matmul(D_Hidden_Dropout,W4)+b4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "with tf.name_scope(\"optimizer\"):\n",
    "    y_label=tf.placeholder(\"float\",shape=[None,10],name=\"y_label\")\n",
    "    loss_function=tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits_v2(logits=y_predict,labels=y_label))\n",
    "    acc_summary = tf.summary.scalar('loss',loss_function) \n",
    "    optimizer=tf.train.AdamOptimizer(learning_rate=0.0001).minimize(loss_function)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "with tf.name_scope(\"evaluate_model\"):\n",
    "    correct_prediction=tf.equal(tf.argmax(y_predict,1),tf.argmax(y_label,1))\n",
    "    accuracy=tf.reduce_mean(tf.cast(correct_prediction,\"float\"))\n",
    "    acc_summary = tf.summary.scalar('accuracy',accuracy) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "merged=tf.summary.merge_all()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "trainEpochs=30\n",
    "batchSize=100\n",
    "totalBatchs=int(mnist.train.num_examples/batchSize)\n",
    "loss_list=[]\n",
    "epoch_list=[]\n",
    "accuracy_list=[]\n",
    "from time import time"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "sess=tf.Session()\n",
    "sess.run(tf.global_variables_initializer())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_writer=tf.summary.FileWriter('log/CNN',sess.graph)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Eposh: 01 Loss= 1.673885107 Accuracy= 0.8056\n",
      "Train Eposh: 02 Loss= 1.633459091 Accuracy= 0.8362\n",
      "Train Eposh: 03 Loss= 1.532889485 Accuracy= 0.9384\n",
      "Train Eposh: 04 Loss= 1.514334202 Accuracy= 0.9536\n",
      "Train Eposh: 05 Loss= 1.506441593 Accuracy= 0.9576\n",
      "Train Eposh: 06 Loss= 1.498090863 Accuracy= 0.9682\n",
      "Train Eposh: 07 Loss= 1.494894624 Accuracy= 0.9702\n",
      "Train Eposh: 08 Loss= 1.491240382 Accuracy= 0.9728\n",
      "Train Eposh: 09 Loss= 1.491295457 Accuracy= 0.9726\n",
      "Train Eposh: 10 Loss= 1.487052560 Accuracy= 0.9766\n",
      "Train Eposh: 11 Loss= 1.486057639 Accuracy= 0.976\n",
      "Train Eposh: 12 Loss= 1.483321309 Accuracy= 0.98\n",
      "Train Eposh: 13 Loss= 1.483799458 Accuracy= 0.9792\n",
      "Train Eposh: 14 Loss= 1.482974648 Accuracy= 0.98\n",
      "Train Eposh: 15 Loss= 1.480804086 Accuracy= 0.9824\n",
      "Train Eposh: 16 Loss= 1.479769111 Accuracy= 0.9832\n",
      "Train Eposh: 17 Loss= 1.478971720 Accuracy= 0.984\n",
      "Train Eposh: 18 Loss= 1.479239106 Accuracy= 0.983\n",
      "Train Eposh: 19 Loss= 1.479031682 Accuracy= 0.9838\n",
      "Train Eposh: 20 Loss= 1.477454066 Accuracy= 0.9866\n",
      "Train Eposh: 21 Loss= 1.478296041 Accuracy= 0.985\n",
      "Train Eposh: 22 Loss= 1.476940989 Accuracy= 0.986\n",
      "Train Eposh: 23 Loss= 1.476793647 Accuracy= 0.986\n",
      "Train Eposh: 24 Loss= 1.476213813 Accuracy= 0.9858\n",
      "Train Eposh: 25 Loss= 1.476969957 Accuracy= 0.9854\n",
      "Train Eposh: 26 Loss= 1.476826191 Accuracy= 0.9868\n",
      "Train Eposh: 27 Loss= 1.475919962 Accuracy= 0.986\n",
      "Train Eposh: 28 Loss= 1.475146651 Accuracy= 0.9868\n",
      "Train Eposh: 29 Loss= 1.475298285 Accuracy= 0.987\n",
      "Train Eposh: 30 Loss= 1.474952102 Accuracy= 0.987\n",
      "Train Finished takes: 84.44034099578857\n"
     ]
    }
   ],
   "source": [
    "startTime=time()\n",
    "for epoch in range(trainEpochs):\n",
    "    for i in range(totalBatchs):\n",
    "        batch_x,batch_y=mnist.train.next_batch(batchSize)\n",
    "        sess.run(optimizer,feed_dict={x:batch_x,y_label:batch_y})\n",
    "    loss,acc,s=sess.run([loss_function,accuracy,merged],feed_dict={x:mnist.validation.images,y_label:mnist.validation.labels})\n",
    "    train_writer.add_summary(s,epoch)\n",
    "    epoch_list.append(epoch)\n",
    "    loss_list.append(loss)\n",
    "    accuracy_list.append(acc)\n",
    "    print(\"Train Eposh:\",\"%02d\"%(epoch+1),\"Loss=\",\"{:.9f}\".format(loss),\"Accuracy=\",acc)\n",
    "duration=time()-startTime\n",
    "print(\"Train Finished takes:\",duration)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "def show_train_history(epoch_list,loss_list,title):\n",
    "    fig=plt.gcf()\n",
    "    fig.set_size_inches(4,2)\n",
    "    plt.plot(epoch_list,loss_list,label='loss')\n",
    "    plt.title(title)\n",
    "    plt.ylabel(\"loss\")\n",
    "    plt.xlabel('Epoch')\n",
    "    plt.legend(['loss'],loc='upper left')\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 288x144 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "show_train_history(epoch_list,loss_list,\"Loss\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 288x144 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "show_train_history(epoch_list,accuracy_list,\"Acc\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.9869\n"
     ]
    }
   ],
   "source": [
    "print(\"Accuracy:\",sess.run(accuracy,feed_dict={x:mnist.test.images,y_label:mnist.test.labels}))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "prediction=sess.run(tf.argmax(y_predict,1),feed_dict={x:mnist.test.images})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([7, 2, 1, 0, 4, 1, 4, 9, 5, 9], dtype=int64)"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prediction[:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "def plot_images_labels_prediction(images,labels,prediction,idx,num=10):\n",
    "    fig=plt.gcf()\n",
    "    fig.set_size_inches(12,14)\n",
    "    if num>25:num=25\n",
    "    for i in range(0,num):\n",
    "        ax=plt.subplot(5,5,1+i)\n",
    "        ax.imshow(np.reshape(images[idx],(28,28)),cmap='binary')\n",
    "        title= \"label=\"+str(np.argmax(labels[idx]))\n",
    "        if len(prediction)>0:\n",
    "            title += \",predict=\"+str(prediction[idx])\n",
    "        ax.set_title(title,fontsize=10)\n",
    "        ax.set_xticks([]);ax.set_yticks([])\n",
    "        idx+=1\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x1008 with 10 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_images_labels_prediction(mnist.test.images,mnist.test.labels,prediction,0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "sess.close()"
   ]
  },
  {
   "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": "3.6.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}