最近在看Morvan老師的視訊,入門RNN,在這裡也貼上自己根據老師的課程修改過的RNN程式碼,作為學習~

用到的是RNN 神經網路,mnist資料集

#  -*- coding: utf-8 -*-

import numpy as np
np.random.seed(1337)

from keras.datasets import mnist
from keras.utils import np_utils
from keras.models import Sequential
from keras.layers import SimpleRNN,Activation,Dense
from keras.optimizers import Adam

#每個圖片就是28行,一個時間段就讀取一行
TIME_STEPS = 28
#輸入就是一個時間段輸入28個列
INPUT_SIZE = 28
#一次迴圈放入多少張圖片
BATCH_SIZE = 50
#
BATCH_INDEX = 0
#輸出的標籤大小是多少
OUTPUT_SIZE = 10
#RNN中間神經元的數量
CELL_SIZE = 50
#學習率
LR = 0.001

(X_train,y_train),(X_test,y_test) = mnist.load_data()

#data process---------------------------
X_train = X_train.reshape(-1, 28, 28) / 255.      # normalize
X_test = X_test.reshape(-1, 28, 28) / 255.        # normalize

#對標籤進行one-hot 編碼
y_train = np_utils.to_categorical(y_train, num_classes=10)
y_test = np_utils.to_categorical(y_test, num_classes=10)

#start to build RNN model----------------
model = Sequential()
model.add(SimpleRNN(input_dim=INPUT_SIZE, input_length=TIME_STEPS,output_dim=CELL_SIZE,
unroll=True))

#輸出層
model.add(Dense(OUTPUT_SIZE))
model.add(Activation('softmax'))

#優化器設定
adam = Adam(LR)
model.compile(optimizer=adam,loss='categorical_crossentropy',metrics=['accuracy'])

#訓練
#迭代次數
train_num = 5000

for step in range(train_num):
    # data shape = (batch_num, steps, inputs/outputs)
    X_batch = X_train[BATCH_INDEX: BATCH_INDEX + BATCH_SIZE, :, :]
    Y_batch = y_train[BATCH_INDEX: BATCH_INDEX + BATCH_SIZE, :]
    cost = model.train_on_batch(X_batch, Y_batch)
    BATCH_INDEX += BATCH_SIZE
    BATCH_INDEX = 0 if BATCH_INDEX >= X_train.shape[0] else BATCH_INDEX

    if step % 500 == 0:
        cost, accuracy = model.evaluate(X_test, y_test, batch_size=y_test.shape[0], verbose=False)
        print('test cost: ', cost, 'test accuracy: ', accuracy)