用RNN做MNIST分类

来源：互联网发布：js iframe 跳转编辑：程序博客网时间：2024/05/11 16:33

1.前言

RNN常用作NLP中，像图片生成文字、自动生成古诗词等。这篇文章用RNN做MNIST手写数字识别，分类效果虽然没有CNN效果好，但准确率也能够达到96%。

2.环境

Mac os系统，python：3.5，Keras

3.代码实现

import numpy as npnp.random.seed(1337)  from keras.datasets import mnistfrom keras.utils import np_utilsfrom keras.models import Sequentialfrom keras.layers import SimpleRNN, Activation, Densefrom keras.optimizers import AdamTIME_STEPS = 28     INPUT_SIZE = 28    BATCH_SIZE = 50BATCH_INDEX = 0OUTPUT_SIZE = 10CELL_SIZE = 50LR = 0.001(X_train, y_train), (X_test, y_test) = mnist.load_data()# data pre-processingX_train = X_train.reshape(-1, 28, 28) / 255.      # normalizeX_test = X_test.reshape(-1, 28, 28) / 255.        # normalizey_train = np_utils.to_categorical(y_train, num_classes=10)y_test = np_utils.to_categorical(y_test, num_classes=10)# build RNN modelmodel = Sequential()# RNN cellmodel.add(SimpleRNN(    batch_input_shape=(None, TIME_STEPS, INPUT_SIZE),           output_dim=CELL_SIZE,    unroll=True,))# output layermodel.add(Dense(OUTPUT_SIZE))model.add(Activation('softmax'))# optimizeradam = Adam(LR)model.compile(optimizer=adam,              loss='categorical_crossentropy',              metrics=['accuracy'])# trainingfor step in range(40001):    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)

4.结果

阅读全文

2 0