Day015_Cifar_HW.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 『本次練習內容』\n",
    "#### 運用這幾天所學觀念搭建一個CNN分類器"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 『本次練習目的』\n",
    "  #### 熟悉CNN分類器搭建步驟與原理\n",
    "  #### 學員們可以嘗試不同搭法，如使用不同的Maxpooling層，用GlobalAveragePooling取代Flatten等等"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "from keras.models import Sequential\n",
    "from keras.layers import Convolution2D\n",
    "from keras.layers import MaxPooling2D\n",
    "from keras.layers import Flatten\n",
    "from keras.layers import Dense\n",
    "from keras.layers import Dropout\n",
    "from keras.layers import BatchNormalization\n",
    "from keras.datasets import cifar10\n",
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "from sklearn.preprocessing import OneHotEncoder\n",
    "from keras import regularizers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(50000, 32, 32, 3)\n"
     ]
    }
   ],
   "source": [
    "(x_train, y_train), (x_test, y_test) = cifar10.load_data()\n",
    "\n",
    "print(x_train.shape) #(50000, 32, 32, 3)\n",
    "\n",
    "## Normalize Data\n",
    "#您好，我們常見的前處理包含Normalization 和 Standardization，\n",
    "#狹義的Normalization 指的是將所有資料rescale到0-1之間，\n",
    "#在影像中/255就是其中一種，上方程式部分比較正確的名稱是Standardization，\n",
    "#也就是將所有的資料-平均值/STD，之所以會加上一個很小的值是為了避免分母為0，\n",
    "#算是廣義的Normalization ，這種方式結果有正有負\n",
    "def normalize(X_train,X_test):\n",
    "        mean = np.mean(X_train,axis=(0,1,2,3))\n",
    "        std = np.std(X_train, axis=(0, 1, 2, 3))\n",
    "        X_train = (X_train-mean)/(std+1e-7)\n",
    "        X_test = (X_test-mean)/(std+1e-7) \n",
    "        return X_train, X_test,mean,std\n",
    "    \n",
    "    \n",
    "## Normalize Training and Testset    \n",
    "x_train, x_test,mean_train,std_train = normalize(x_train, x_test) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\preprocessing\\_encoders.py:368: FutureWarning: The handling of integer data will change in version 0.22. Currently, the categories are determined based on the range [0, max(values)], while in the future they will be determined based on the unique values.\n",
      "If you want the future behaviour and silence this warning, you can specify \"categories='auto'\".\n",
      "In case you used a LabelEncoder before this OneHotEncoder to convert the categories to integers, then you can now use the OneHotEncoder directly.\n",
      "  warnings.warn(msg, FutureWarning)\n"
     ]
    }
   ],
   "source": [
    "## OneHot Label 由(None, 1)-(None, 10)\n",
    "## ex. label=2,變成[0,0,1,0,0,0,0,0,0,0]\n",
    "one_hot=OneHotEncoder()\n",
    "y_train=one_hot.fit_transform(y_train).toarray()\n",
    "y_test=one_hot.transform(y_test).toarray()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:20: UserWarning: Update your `Dense` call to the Keras 2 API: `Dense(activation=\"relu\", units=100)`\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:23: UserWarning: Update your `Dense` call to the Keras 2 API: `Dense(activation=\"softmax\", units=10)`\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/10\n",
      "50000/50000 [==============================] - 7716s 154ms/step - loss: 1.3998 - acc: 0.5035\n",
      "Epoch 2/10\n",
      "50000/50000 [==============================] - 739s 15ms/step - loss: 1.0144 - acc: 0.6440\n",
      "Epoch 3/10\n",
      "50000/50000 [==============================] - 762s 15ms/step - loss: 0.8829 - acc: 0.6895\n",
      "Epoch 4/10\n",
      "18000/50000 [=========>....................] - ETA: 8:04 - loss: 0.7806 - acc: 0.7286"
     ]
    },
    {
     "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-11-aca6b26dc7a2>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[0;32m     25\u001b[0m \u001b[1;31m#超過兩個就要選categorical_crossentrophy\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     26\u001b[0m \u001b[0mclassifier\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mcompile\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0moptimizer\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;34m'adam'\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mloss\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;34m'categorical_crossentropy'\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mmetrics\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;33m[\u001b[0m\u001b[1;34m'accuracy'\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;32m---> 27\u001b[1;33m \u001b[0mclassifier\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mfit\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mx_train\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0my_train\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0mbatch_size\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;36m100\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0mepochs\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;36m10\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[1;32mC:\\ProgramData\\Anaconda3\\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   1037\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[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1038\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[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 1039\u001b[1;33m                                         validation_steps=validation_steps)\n\u001b[0m\u001b[0;32m   1040\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1041\u001b[0m     def evaluate(self, x=None, y=None,\n",
      "\u001b[1;32mC:\\ProgramData\\Anaconda3\\lib\\site-packages\\keras\\engine\\training_arrays.py\u001b[0m in \u001b[0;36mfit_loop\u001b[1;34m(model, 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    197\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[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    198\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 199\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[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    200\u001b[0m                 \u001b[0mouts\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mto_list\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mouts\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    201\u001b[0m                 \u001b[1;32mfor\u001b[0m \u001b[0ml\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mo\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mzip\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mout_labels\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mouts\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:\\ProgramData\\Anaconda3\\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   2713\u001b[0m                 \u001b[1;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_legacy_call\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0minputs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   2714\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 2715\u001b[1;33m             \u001b[1;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_call\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0minputs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   2716\u001b[0m         \u001b[1;32melse\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   2717\u001b[0m             \u001b[1;32mif\u001b[0m \u001b[0mpy_any\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mis_tensor\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mx\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;32mfor\u001b[0m \u001b[0mx\u001b[0m \u001b[1;32min\u001b[0m \u001b[0minputs\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:\\ProgramData\\Anaconda3\\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   2673\u001b[0m             \u001b[0mfetched\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_callable_fn\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m*\u001b[0m\u001b[0marray_vals\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mrun_metadata\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mrun_metadata\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   2674\u001b[0m         \u001b[1;32melse\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 2675\u001b[1;33m             \u001b[0mfetched\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_callable_fn\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m*\u001b[0m\u001b[0marray_vals\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   2676\u001b[0m         \u001b[1;32mreturn\u001b[0m \u001b[0mfetched\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[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   2677\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mC:\\ProgramData\\Anaconda3\\lib\\site-packages\\tensorflow\\python\\client\\session.py\u001b[0m in \u001b[0;36m__call__\u001b[1;34m(self, *args, **kwargs)\u001b[0m\n\u001b[0;32m   1437\u001b[0m           ret = tf_session.TF_SessionRunCallable(\n\u001b[0;32m   1438\u001b[0m               \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_session\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_session\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_handle\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0margs\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mstatus\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 1439\u001b[1;33m               run_metadata_ptr)\n\u001b[0m\u001b[0;32m   1440\u001b[0m         \u001b[1;32mif\u001b[0m \u001b[0mrun_metadata\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1441\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[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "\n",
    "classifier=Sequential()\n",
    "\n",
    "#卷積組合 #32,3,3,input_shape=(32,32,3),activation='relu''\n",
    "classifier.add(Convolution2D(32, kernel_size=(3, 3), padding='same',input_shape=(32,32,3),activation='relu'))\n",
    "classifier.add(BatchNormalization(axis=-1, momentum=0.99, epsilon=0.001))\n",
    "classifier.add(MaxPooling2D(pool_size=(2,2)))\n",
    "\n",
    "#卷積組合\n",
    "classifier.add(Convolution2D(32, kernel_size=(3, 3),activation='relu')))\n",
    "classifier.add(BatchNormalization(axis=-1, momentum=0.99, epsilon=0.001))\n",
    "classifier.add(MaxPooling2D(pool_size=(2,2)))\n",
    "\n",
    "#flatten\n",
    "classifier.add(Flatten())\n",
    "\n",
    "#FC\n",
    "classifier.add(Dense(output_dim=100,activation='relu',kernel_regularizer=regularizers.l2(0.001))) #output_dim=100,activation=relu\n",
    "classifier.add(BatchNormalization()) ##BatchNormalization\n",
    "classifier.add(Dropout(p=0.5)) ##Dropout\n",
    "\n",
    "classifier.add(Dense(output_dim=10,activation='relu',kernel_regularizer=regularizers.l2(0.001))) ##regularizers\n",
    "classifier.add(BatchNormalization()) ##BatchNormalization\n",
    "classifier.add(Dropout(p=0.3))##Dropout\n",
    "\n",
    "#輸出\n",
    "classifier.add(Dense(output_dim=10,activation='softmax'))\n",
    "\n",
    "#超過兩個就要選categorical_crossentrophy\n",
    "classifier.compile(optimizer = 'adam', loss = 'categorical_crossentropy', metrics = ['accuracy'])\n",
    "#classifier.fit(x_train,y_train,batch_size=100,epochs=10)\n",
    "\n",
    "\n",
    "\n",
    "from keras.preprocessing.image import ImageDataGenerator ##Augmentation\n",
    "img_gen = ImageDataGenerator( featurewise_center=True,featurewise_std_normalization=True,\n",
    "                             rotation_range=10,width_shift_range=0.1,\n",
    "                             height_shift_range=0.1,shear_range=0.1,zoom_range=0.1,\n",
    "                             horizontal_flip=True,vertical_flip=False,dtype=np.float32)\n",
    "img_gen.fit(x_train)\n",
    "\n",
    "\n",
    "from keras.callbacks import EarlyStopping\n",
    "earlystop = EarlyStopping(monitor='test_loss', patience=8, verbose=1) ##earlystop\n",
    "\n",
    "##開始訓練\n",
    "classifier.fit_generator(img_gen.flow(x_train, y_train, batch_size=100) ,steps_per_epoch=500,\n",
    "                               epochs=100, validation_data = (x_test, y_test),callbacks = [earlystop])\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 預測新圖片，輸入影像前處理要與訓練時相同\n",
    "#### ((X-mean)/(std+1e-7) ):這裡的mean跟std是訓練集的\n",
    "## 維度如下方示範"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "input_example=(np.zeros(shape=(1,32,32,3))-mean_train)/(std_train+1e-7) \n",
    "classifier.predict(input_example)"
   ]
  }
 ],
 "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": "3.7.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}