KerasによるMNISTの手書き文字の分類

#KerasによるMNISTの手書き文字の分類

##概要
今回はkerasを用いてmnistという手書き文字がたくさん入ったimageデータセットから1~9の文字を分類します

###必要なライブラリのimport

import os,re
import matplotlib.pyplot as plt
import numpy as np
from sklearn.model_selection import train_test_split
import keras
from keras.datasets import mnist

###mnistダウンロード
データを学習用とテスト用に分割
次に学習用のデータを学習用と検証用に8:2の割合で分割

from keras.datasets import mnist
from sklearn.model_selection import train_test_split

(train_data, train_labels), (test_data, test_labels) = mnist.load_data()
x_train, x_valid, y_train, y_valid = train_test_split(train_data, train_labels, test_size=0.2)

###画像データと正解ラベルのリサイズ
①x_train,x_validは現在(4800,28,28),(1200,28,28)になっているのでkerasのモデルの学習に使える形にするために(4800,28,28,1),(1200,28,28,1)の形に変換
②画像データのデータ型をfloat型に変換
③画像の画素値は0~255までで数値が大きいので0~1の間に変換
④正解ラベルをone-hot-encodhingにする

from keras.utils import to_categorical

#①
x_train = x_train.reshape(x_train.shape[0], 28, 28, 1)
x_valid = x_valid.reshape(x_valid.shape[0], 28, 28, 1)
#②
x_train = x_train.astype('float32')
x_valid = x_valid.astype('float32')
#③
x_train /= 255
x_valid /= 255
#④
y_train = keras.utils.to_categorical(y_train, 10)
y_valid = keras.utils.to_categorical(y_valid, 10)

###モデルの作成

from keras.models import Sequential
from keras.layers import Dense, Dropout, Flatten
from keras.layers import Conv2D, MaxPooling2D

model = Sequential()
model.add(Conv2D(32, kernel_size=(3, 3), activation='relu', input_shape=(28, 28, 1)))
model.add(Conv2D(64, (3, 3), activation='relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.25))
model.add(Flatten())
model.add(Dense(128, activation='relu'))
model.add(Dropout(0.5))
model.add(Dense(10, activation='softmax'))
model.summary()

Model: "sequential_1"
_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
conv2d_2 (Conv2D)            (None, 26, 26, 32)        320       
_________________________________________________________________
conv2d_3 (Conv2D)            (None, 24, 24, 64)        18496     
_________________________________________________________________
max_pooling2d_1 (MaxPooling2 (None, 12, 12, 64)        0         
_________________________________________________________________
dropout_2 (Dropout)          (None, 12, 12, 64)        0         
_________________________________________________________________
flatten_1 (Flatten)          (None, 9216)              0         
_________________________________________________________________
dense_2 (Dense)              (None, 128)               1179776   
_________________________________________________________________
dropout_3 (Dropout)          (None, 128)               0         
_________________________________________________________________
dense_3 (Dense)              (None, 10)                1290      
=================================================================
Total params: 1,199,882
Trainable params: 1,199,882
Non-trainable params: 0
_________________________________________________________________

###モデルのコンパイル

from keras.optimizers import RMSprop

model.compile(loss='categorical_crossentropy',
              optimizer=RMSprop(),
              metrics=['accuracy'])

###画像の水増しと学習

datagen = ImageDataGenerator(
    featurewise_center=False,               # データセット全体で,入力の平均を0に調整
    samplewise_center=False,                # 各サンプルの平均を0に調整
    featurewise_std_normalization=False,    # 入力をデータセットの標準偏差で正規化
    samplewise_std_normalization=False,     # 各入力をその標準偏差で正規化
    zca_whitening=False,                    # ZCA白色化のイプシロン
    rotation_range=50,                      # 回転角度(-50～50度)
    width_shift_range=0.3,                  # 左右のスライド幅
    height_shift_range=0.2,                 # 上下のスライド幅
    zoom_range=[1.0,1.5],                   # 拡大・縮小率
    horizontal_flip=False,                  # 水平反転しない
    vertical_flip=False)    
 
hist = model.fit_generator(datagen.flow(x_train, y_train, batch_size=32),
                               steps_per_epoch=len(x_train)/32, epochs=10,
                               validation_data=(x_valid, y_valid)).history

###精度の表示

# 精度のplot
plt.plot(hist['accuracy'], marker='.', label='acc')
plt.plot(hist['val_accuracy'], marker='.', label='val_acc')
plt.title('model accuracy')
plt.grid()
plt.xlabel('epoch')
plt.ylabel('accuracy')
plt.legend(loc='best')
plt.show()

# 損失のplot
plt.plot(hist['loss'], marker='.', label='loss')
plt.plot(hist['val_loss'], marker='.', label='val_loss')
plt.title('model loss')
plt.grid()
plt.xlabel('epoch')
plt.ylabel('loss')
plt.legend(loc='best')
plt.show()

###テストデータで精度の評価

test_data = test_data.reshape(test_data.shape[0], 28, 28, 1)
test_data = test_data.astype('float32')
test_data /= 255
test_labels = keras.utils.to_categorical(test_labels, 10)

score = model.evaluate(test_data, test_labels, verbose=0)
print('Test loss:', score[0])
print('Test accuracy:', score[1])