Golaboratory的tensorflow的练习Demo

1.准备环境

win10、python3.7、tensorflow1.9、keras2.2.0

具体安装请参考

环境安装

使用tensorflow进行数据分析和预测

from keras.models import Sequential
from keras.layers import LSTM, Dense

import pandas as pd
from random import shuffle
import numpy as np
import codecs ,csv

注意：出现下面Using TensorFlow backend. 才算正确
若果出现报错：module ‘tensorflow.python.keras.backend’ has no attribute ‘get_graph’
请参考前面的keras和tensorflow的配置

##定义生成数据参数

data_dim=16
timesteps=8
nb_classes=10
filename="./code/LSTM-Outputs.txt"

###生成训练集
x_train = np.random.random((1000,timesteps,data_dim))
y_train = np.random.random((1000,nb_classes))


###生成验证数据
x_val = np.random.random((100,timesteps,data_dim))
y_val = np.random.random((100,nb_classes))


###生成测试数据
x_test = np.random.random((100,timesteps,data_dim))
y_test = np.random.random((100,nb_classes))


####简单搭建一个LSTM

model=Sequential()  ##模型初始化

#####
model.add(LSTM(32,return_sequences=True, input_shape=(timesteps,data_dim)))

##return a sequence of vectors of dimension

model.add(LSTM(32,return_sequences=True))

model.add(LSTM(32))

model.add(Dense(10,activation='softmax'))


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


#######训练模型

model.fit(x_train,y_train,batch_size=64,epochs=10,validation_data=(x_val,y_val))

##预测结果
results=model.predict(x_test,batch_size=32,verbose=0)

##保存预测结果

with codecs.open(filename,'w',encoding='utf-8') as f:
    f_csv=csv.writer(f)
    f_csv.writerows(results)
score = model.evaluate(x_test,y_test,batch_size=16) ##在test数据上评估模型
print("\n Evaluation Metrics: \n",model.metrics_names)
print(score)

数据下载：