# -*-coding=utf8 -*-
import pandas as pd
import numpy as np
import tensorflow as tf
#定义常量
rnn_unit=128 #10 #hidden layer units
input_size=30
output_size=2
lr=0.001#0.0006 #学习率
f = open('Train.csv')
df = pd.read_csv(f)
data = df.iloc[:, :30].values
#print(data.shape)
# print data.head()
def get_train_data(batch_size=60,time_step=20,train_begin=0,train_end=999):
batch_index=[]
data_train=data[train_begin:train_end]
normalized_train_data=(data_train-np.mean(data_train,axis=0))/np.std(data_train,axis=0) #标准化
train_x,train_y=[],[] #训练集x和y初定义
for i in range(len(normalized_train_data)-time_step):
if i % batch_size==0:
batch_index.append(i)
x=normalized_train_data[i:i+time_step,:30]
y=normalized_train_data[i:i+time_step,-3:-1] #,np.newaxis
train_x.append(x.tolist())
train_y.append(y.tolist())
batch_index.append((len(normalized_train_data)-time_step))
print("(len(normalized_train_data)", len(normalized_train_data))
print("train_x.shape: ", np.array(train_x).shape)
print("train_y.shape is " , np.array(train_y).shape)
print(", batch_index is ", batch_index)
return batch_index,train_x,train_y
#——————————获取测试集——————————
def get_test_data(time_step=20,test_begin=9500):
data_test=data[test_begin:]
print(data_test.shape)
mean=np.mean(data_test,axis=0)
std=np.std(data_test,axis=0)
normalized_test_data=(data_test-mean)/std #标准化
size=(len(normalized_test_data)+time_step-1)//time_step #有size个sample
print("size is",size)
test_x,test_y=[],[]
for i in range(size-1):
x=normalized_test_data[i*time_step:(i+1)*time_step,:30]
y=normalized_test_data[i*time_step:(i+1)*time_step,-3:-1]
test_x.append(x.tolist())
test_y.extend(y.tolist())
test_x.append((normalized_test_data[(i+1)*time_step:,:30]).tolist())
test_y.extend((normalized_test_data[(i+1)*time_step:,-3:-1]).tolist())
#print("(len(normalized_train_data)", len(normalized_train_data))
print("test_x.shape: ", np.array(test_x).shape)
print("test_y.shape is " , np.array(test_y).shape)
print("std is ", std.shape)
print("mean is ", mean.shape)
return mean,std,test_x,test_y
#输入层、输出层权重、偏置
weights={
'in':tf.Variable(tf.random_normal([input_size,rnn_unit])),
'out':tf.Variable(tf.random_normal([rnn_unit,output_size]))## maybe ???
}
biases={
'in':tf.Variable(tf.constant(0.1,shape=[rnn_unit,])),
'out':tf.Variable(tf.constant(0.1,shape=[output_size,]))
}
#——————————————————定义神经网络变量——————————————————
def lstm(X):
batch_size=tf.shape(X)[0]
time_step=tf.shape(X)[1]
w_in=weights['in']
b_in=biases['in']
input=tf.reshape(X,[-1,input_size]) #需要将tensor转成2维进行计算,计算后的结果作为隐藏层的输入
input_rnn=tf.matmul(input,w_in)+b_in
input_rnn=tf.reshape(input_rnn,[-1,time_step,rnn_unit]) #将tensor转成3维,作为lstm cell的输入
cell=tf.nn.rnn_cell.BasicLSTMCell(rnn_unit)
init_state=cell.zero_state(batch_size,dtype=tf.float32)
output_rnn,final_states=tf.nn.dynamic_rnn(cell, input_rnn,initial_state=init_state, dtype=tf.float32) #output_rnn是记录lstm每个输出节点的结果,final_states是最后一个cell的结果
output=tf.reshape(output_rnn,[-1,rnn_unit]) #作为输出层的输入 讲Tensor从三维变成2维
w_out=weights['out']
b_out=biases['out']
pred=tf.matmul(output,w_out)+b_out
return pred,final_states
#——————————————————训练模型——————————————————
def train_lstm(batch_size=80,time_step=15,train_begin=0,train_end=99):
X=tf.placeholder(tf.float32, shape=[None,time_step,input_size])
Y=tf.placeholder(tf.float32, shape=[None,time_step,output_size])
batch_index,train_x,train_y=get_train_data(batch_size,time_step,train_begin,train_end)
pred,_=lstm(X)
#损失函数
loss=tf.reduce_sum(tf.square(tf.reshape(pred,[-1])-tf.reshape(Y, [-1])))/batch_size
train_op=tf.train.AdamOptimizer(lr).minimize(loss)
saver=tf.train.Saver(tf.global_variables(),max_to_keep=15)
# module_file = tf.train.latest_checkpoint()
with tf.Session() as sess:
sess.run(tf.global_variables_initializer())
# saver.restore(sess, module_file)
#重复训练2000次
for i in range(5001):
for step in range(len(batch_index)-1):
_,loss_=sess.run([train_op,loss],
feed_dict={X:train_x[batch_index[step]:batch_index[step+1]],Y:train_y[batch_index[step]:batch_index[step+1]]})
print(i,loss_)
if i % 200==0:
print("保存模型:",saver.save(sess,'saveModel/stock.ckpt',global_step=i))
# #————————————————预测模型————————————————————
# def prediction(time_step=20):
# X=tf.placeholder(tf.float32, shape=[None,time_step,input_size])
# mean,std,test_x,test_y=get_test_data(time_step)
# pred,_=lstm(X)
# saver=tf.train.Saver(tf.global_variables())
# with tf.Session() as sess:
# #参数恢复
# # module_file = tf.train.latest_checkpoint()
#
# #saver.restore(sess, module_file)
# saver.restore(sess, 'save/stock.ckpt-1500')
# test_predict=[]
# for step in range(len(test_x)-1):
# prob=sess.run(pred,feed_dict={X:[test_x[step]]})
# predict=prob.reshape((-1))
# test_predict.extend(predict)
# test_y=np.array(test_y)*std[8]+mean[8]
# test_predict=np.array(test_predict)*std[8]+mean[8]
# acc=np.average(np.abs(test_predict-test_y[:len(test_predict)])/test_y[:len(test_predict)]) #acc为测试集偏差
# print "the acc is: ", acc
import matplotlib.pyplot as plt
def prediction(time_step=20):
X=tf.placeholder(tf.float32, shape=[None,time_step,input_size])
#Y=tf.placeholder(tf.float32, shape=[None,time_step,output_size])
mean,std,test_x,test_y=get_test_data(time_step)
pred,_=lstm(X)
saver=tf.train.Saver(tf.global_variables())
with tf.Session() as sess:
#参数恢复
# module_file = tf.train.latest_checkpoint()
# saver.restore(sess, module_file)
saver.restore(sess, 'saveModel/stock.ckpt-5000')
test_predict=[]
for step in range(len(test_x)-1):
prob=sess.run(pred,feed_dict={X:[test_x[step]]})
predict=prob.reshape((-1))
test_predict.append(predict)
print(np.array(test_predict).shape)
test_y=np.array(test_y)*std[29]+mean[29]
test_predict=np.array(test_predict)*std[29]+mean[29]
test_predict=np.round(test_predict);
#acc=np.average(np.abs(test_predict-test_y[:len(test_predict)])/test_y[:len(test_predict)]) #偏差
#rmse=np.sqrt(np.average(np.square(test_predict-test_y[:len(test_predict)]))) #均方根误差
# error=np.average(test_predict-test_y[:len(test_predict)])
#以折线图表示结果
#print (acc)
print(test_predict[1:100])
print(test_y[1:100])
plt.figure()
plt.plot(list(range(len(test_predict))), test_predict, color='b',label='Predict',markersize=20)
plt.plot(list(range(len(test_y))), test_y, color='r',label='Real',markersize=20)
fig1 = plt.figure(1)
axes = plt.subplot(111)
axes.grid(True) # add grid
plt.legend(loc="lower right") #set legend location
plt.ylabel('Classification') # set ystick label
plt.xlabel('Test sample') # set xstck label
plt.show()
if __name__ == "__main__":
#train_lstm()
prediction()