涨跌的例题用C语言编辑,基于LSTM的股票涨跌分析-pytorch

通过输入是个指标对每天的涨跌进行相关预测，实现的准确率达到93%，加入交叉熵进行相关损失函数，尽量减小过拟合现象，但是在参数的最有参数选择的时候，并没有加入最优适应，需要后期进行相关的模型优化，代码如下

import numpy as np

import pandas as pd

import matplotlib.pyplot as plt

import torch

from torch import nn

from torch.autograd import Variable

data_csv = pd.read_csv('D:/Python/gs/data.csv')

data_csv = data_csv.dropna(axis=0, how='any')

data_csv = data_csv.values

data_x = data_csv[:, :10].astype('float32')

data_y = data_csv[:, 10].astype('int32')

data_x_normed = (data_x - np.min(data_x, axis=0)) / (np.max(data_x, axis=0)-np.min(data_x, axis=0))

#data_x_normed=(data_x-np.mean(data_x,axis=0))/np.std(data_x,axis=0)

data_x = np.array(data_x_normed)

data_y = np.array(data_y)

train_size = int(len(data_x) * 0.9)

test_size = len(data_x) - train_size

train_x = data_x[:train_size]

train_y = data_y[:train_size]

test_x = data_x[train_size:]

test_y = data_y[train_size:]

train_x = train_x.reshape(-1, 1, 10)

#train_y = train_y.reshape(-1, 1, 1)

train_x = torch.from_numpy(train_x)

train_y = torch.from_numpy(train_y)

class NET(nn.Module):

def __init__(self,input_size=10,hidden_size=40,output_size=2,num_layer=2):

super(NET,self).__init__()

self.rnn=nn.LSTM(input_size,hidden_size,num_layer)

self.out=nn.Linear(hidden_size,output_size)

def forward(self,x):

out,_=self.rnn(x)

out=self.out(out[:,-1,:])

return out

net = NET()

optimizer = torch.optim.Adam(net.parameters(), lr=0.08, betas=(0.9, 0.999), eps=1e-08, weight_decay=0)

loss_func = torch.nn.CrossEntropyLoss()

for epoch in range(1000):

var_x=Variable(train_x).type(torch.FloatTensor)

var_y=Variable(train_y).type(torch.LongTensor)

out = net(var_x)

loss = loss_func(out, var_y)

optimizer.zero_grad()

loss.backward()

optimizer.step()

if (epoch + 1) % 100 == 0:

print('Epoch: {}, Loss: {:.5f}'.format(epoch + 1, loss.data.numpy()))

test_x = test_x.reshape(-1, 1, 10)

test_x = torch.from_numpy(test_x)

var_data = Variable(test_x)

pred_test = net(var_data)

#pred_test = pred_test.view(-1).data.numpy()

pred_test=torch.max(pred_test,1)[1].data.numpy().squeeze()

plt.plot(pred_test, 'r', label='prediction')

plt.plot(test_y, 'b', label='real')

plt.legend(loc='best')

plt.show()

print(pred_test,'prediction number')

print(test_y,'real number')

j=0

for i in range(test_size):

if(pred_test[i] == test_y[i]):

j=j+1;

j=j/test_size

print('Identification:',j)