官方Github上给的例子是CPU版本:https://github.com/pytorch/examples/tree/master/time_sequence_prediction
下面的代码将其改为了GPU版本。除了将网络模型和输入迁移到显卡上之外,还有注意两点:
一是计算隐藏层状态和细胞状态时也需要把h和c的值迁移到显卡上;
二是在画图之前需要把预测值迁移回到CPU上才能将其转化为numpy数组。
from __future__ import print_function
import torch
import torch.nn as nn
import torch.optim as optim
import numpy as np
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt
class Sequence(nn.Module):
def __init__(self):
super(Sequence, self).__init__()
self.lstm1 = nn.LSTMCell(1, 51)
self.lstm2 = nn.LSTMCell(51, 51)
self.linear = nn.Linear(51, 1)
def forward(self, input, future = 0):
outputs = []
#h_t = torch.zeros(input.size(0), 51, dtype=torch.double)
#c_t = torch.zeros(input.size(0), 51, dtype=torch.double)
#h_t2 = torch.zeros(input.size(0), 51, dtype=torch.double)
#c_t2 = torch.zeros(input.size(0), 51, dtype=torch.double)
h_t = torch.zeros(input.size(0), 51, dtype=input.dtype, device=input.device)
c_t = torch.zeros(input.size(0), 51, dtype=input.dtype, device=input.device)
h_t2 = torch.zeros(input.size(0), 51, dtype=input.dtype, device=input.device)
c_t2 = torch.zeros(input.size(0), 51, dtype=input.dtype, device=input.device)
for i, input_t in enumerate(input.chunk(input.size(1), dim=1)):
h_t, c_t = self.lstm1(input_t, (h_t, c_t))
h_t2, c_t2 = self.lstm2(h_t, (h_t2, c_t2))
output = self.linear(h_t2)
outputs += [output]
for i in range(future):# if we should predict the future
h_t, c_t = self.lstm1(output, (h_t, c_t))
h_t2, c_t2 = self.lstm2(h_t, (h_t2, c_t2))
output = self.linear(h_t2)
outputs += [output]
outputs = torch.stack(outputs, 1).squeeze(2)
return outputs
if __name__ == '__main__':
# set random seed to 0
np.random.seed(0)
torch.manual_seed(0)
# load data and make training set
data = torch.load('traindata.pt')
input = torch.from_numpy(data[3:, :-1])
target = torch.from_numpy(data[3:, 1:])
test_input = torch.from_numpy(data[:3, :-1])
test_target = torch.from_numpy(data[:3, 1:])
# move model to GPU
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
print(device)
input, target = input.to(device), target.to(device)
test_input, test_target = test_input.to(device), test_target.to(device)
# build the model
seq = Sequence()
seq.to(device)
seq.double()
criterion = nn.MSELoss()
# use LBFGS as optimizer since we can load the whole data to train
optimizer = optim.LBFGS(seq.parameters(), lr=0.8)
epoch = 15
#begin to train
for i in range(epoch):
print('STEP: ', i)
def closure():
optimizer.zero_grad()
out = seq(input)
#print('input size: ', input.size())
loss = criterion(out, target)
print('loss:', loss.item())
loss.backward()
return loss
optimizer.step(closure)
# begin to predict, no need to track gradient here
with torch.no_grad():
future = 1000
pred = seq(test_input, future=future)
loss = criterion(pred[:, :-future], test_target)
print('test loss:', loss.item())
#y = pred.detach().numpy()
y = pred.detach().cpu().numpy()
# draw the result
plt.figure(figsize=(30,10))
plt.title('Predict future values for time sequences\n(Dashlines are predicted values)', fontsize=30)
plt.xlabel('x', fontsize=20)
plt.ylabel('y', fontsize=20)
plt.xticks(fontsize=20)
plt.yticks(fontsize=20)
def draw(yi, color):
plt.plot(np.arange(input.size(1)), yi[:input.size(1)], color, linewidth = 2.0)
plt.plot(np.arange(input.size(1), input.size(1) + future), yi[input.size(1):], color + ':', linewidth = 2.0)
draw(y[0], 'r')
draw(y[1], 'g')
draw(y[2], 'b')
plt.savefig('predict%d.pdf'%i)
plt.close()