日本免费高清视频-国产福利视频导航-黄色在线播放国产-天天操天天操天天操天天操|www.shdianci.com

學(xué)無先后，達(dá)者為師

網(wǎng)站首頁 編程語言 正文

python神經(jīng)網(wǎng)絡(luò)使用Keras構(gòu)建RNN訓(xùn)練_python

作者：Bubbliiiing ? 更新時(shí)間： 2022-06-28 編程語言

Keras中構(gòu)建RNN的重要函數(shù)

1、SimpleRNN

SimpleRNN用于在Keras中構(gòu)建普通的簡單RNN層，在使用前需要import。

from keras.layers import SimpleRNN

在實(shí)際使用時(shí)，需要用到幾個(gè)參數(shù)。

model.add(
    SimpleRNN(
        batch_input_shape = (BATCH_SIZE,TIME_STEPS,INPUT_SIZE),
        output_dim = CELL_SIZE,
    )
)

其中，batch_input_shape代表RNN輸入數(shù)據(jù)的shape，shape的內(nèi)容分別是每一次訓(xùn)練使用的BATCH，TIME_STEPS表示這個(gè)RNN按順序輸入的時(shí)間點(diǎn)的數(shù)量，INPUT_SIZE表示每一個(gè)時(shí)間點(diǎn)的輸入數(shù)據(jù)大小。
CELL_SIZE代表訓(xùn)練每一個(gè)時(shí)間點(diǎn)的神經(jīng)元數(shù)量。

2、model.train_on_batch

與之前的訓(xùn)練CNN網(wǎng)絡(luò)和普通分類網(wǎng)絡(luò)不同，RNN網(wǎng)絡(luò)在建立時(shí)就規(guī)定了batch_input_shape，所以訓(xùn)練的時(shí)候也需要一定量一定量的傳入訓(xùn)練數(shù)據(jù)。
model.train_on_batch在使用前需要對數(shù)據(jù)進(jìn)行處理。獲取指定BATCH大小的訓(xùn)練集。

X_batch = X_train[index_start:index_start + BATCH_SIZE,:,:]
Y_batch = Y_train[index_start:index_start + BATCH_SIZE,:]
index_start += BATCH_SIZE

具體訓(xùn)練過程如下：

for i in range(500):
    X_batch = X_train[index_start:index_start + BATCH_SIZE,:,:]
    Y_batch = Y_train[index_start:index_start + BATCH_SIZE,:]
    index_start += BATCH_SIZE
    cost = model.train_on_batch(X_batch,Y_batch)
    if index_start >= X_train.shape[0]:
        index_start = 0
    if i%100 == 0:
        ## acc
        cost,accuracy = model.evaluate(X_test,Y_test,batch_size=50)
        ## W,b = model.layers[0].get_weights()
        print("accuracy:",accuracy)
        x = X_test[1].reshape(1,28,28)

全部代碼

這是一個(gè)RNN神經(jīng)網(wǎng)絡(luò)的例子，用于識別手寫體。

import numpy as np
from keras.models import Sequential
from keras.layers import SimpleRNN,Activation,Dense ## 全連接層
from keras.datasets import mnist
from keras.utils import np_utils
from keras.optimizers import Adam
TIME_STEPS = 28
INPUT_SIZE = 28
BATCH_SIZE = 50
index_start = 0
OUTPUT_SIZE = 10
CELL_SIZE = 75
LR = 1e-3
(X_train,Y_train),(X_test,Y_test) = mnist.load_data()
X_train = X_train.reshape(-1,28,28)/255
X_test = X_test.reshape(-1,28,28)/255
Y_train = np_utils.to_categorical(Y_train,num_classes= 10)
Y_test = np_utils.to_categorical(Y_test,num_classes= 10)
model = Sequential()
# conv1
model.add(
    SimpleRNN(
        batch_input_shape = (BATCH_SIZE,TIME_STEPS,INPUT_SIZE),
        output_dim = CELL_SIZE,
    )
)
model.add(Dense(OUTPUT_SIZE))
model.add(Activation("softmax"))
adam = Adam(LR)
## compile
model.compile(loss = 'categorical_crossentropy',optimizer = adam,metrics = ['accuracy'])
## tarin
for i in range(500):
    X_batch = X_train[index_start:index_start + BATCH_SIZE,:,:]
    Y_batch = Y_train[index_start:index_start + BATCH_SIZE,:]
    index_start += BATCH_SIZE
    cost = model.train_on_batch(X_batch,Y_batch)
    if index_start >= X_train.shape[0]:
        index_start = 0
    if i%100 == 0:
        ## acc
        cost,accuracy = model.evaluate(X_test,Y_test,batch_size=50)
        ## W,b = model.layers[0].get_weights()
        print("accuracy:",accuracy)

實(shí)驗(yàn)結(jié)果為：

10000/10000 [==============================] - 1s 147us/step
accuracy: 0.09329999938607215
…………………………
10000/10000 [==============================] - 1s 112us/step
accuracy: 0.9395000022649765
10000/10000 [==============================] - 1s 109us/step
accuracy: 0.9422999995946885
10000/10000 [==============================] - 1s 114us/step
accuracy: 0.9534000000357628
10000/10000 [==============================] - 1s 112us/step
accuracy: 0.9566000008583069
10000/10000 [==============================] - 1s 113us/step
accuracy: 0.950799999833107
10000/10000 [==============================] - 1s 116us/step
10000/10000 [==============================] - 1s 112us/step
accuracy: 0.9474999988079071
10000/10000 [==============================] - 1s 111us/step
accuracy: 0.9515000003576278
10000/10000 [==============================] - 1s 114us/step
accuracy: 0.9288999977707862
10000/10000 [==============================] - 1s 115us/step
accuracy: 0.9487999993562698

原文鏈接：https://blog.csdn.net/weixin_44791964/article/details/101609556

欄目分類
最近更新