RNN_net拟合回归曲线

导入函数库

import torch
import torch.nn as nn
import matlibplot.pyplot as plt
from torch.autograd import Variable
import numpy as np

全局参数

TIME_STEP=10
INPUT_SIZE=1
LR=0.02

自变量，因变量的定义

steps=np.linpace(0,np.pi*2,100)
x_np=np.sin(steps)
y_np=np.cos(steps)

展现自变量和因变量之间的关系的图像

plt.plot(steps,x_np,'b*',label='input(sin)')
plt.plot(steps,y_np,'r.',label='target(cos)')
plt.legend(loc='best')
plt.show()

定义RNN网络

class RNN(nn.Module):
	def __init__():
		super(RNN,self).__init__()
		self.rnn=RNN(
			input_size=INPUT_SIZE,
			hidden_size=32,
			num_layers=1,
			batch_first=True,
			)
		self.out=nn.Linear(32,1)
		def forward(self,x):
			r_out,h_state=self.rnn(x,h_state)
			outs=[]
			for time_step in range(r_out.size(1))
				outs.append(self.out(r_out[:,time_step,:]))
			return torch.stack(outs,dim=1),h_state

检验生成的RNN网络

rnn=RNN()
print(rnn)

网络层结构

RNN(
  (rnn): RNN(1, 32, batch_first=True)
  (out): Linear(in_features=32, out_features=1, bias=True)
)

优化神经网络

optimizer=torch.optim.Adam(rnn.parameters(),lr=LR)
loss_func=nn.MSELoss()
h_state=None

训练神经网络

for step in range(100):
	start,end=step*np.pi,（step+1)*np.pi
	steps=np.linspace(start,end,TIME_STEP,dtype=np.float32)
	x_np=np.sin(steps)
	y_np=np.cos(steps)
	x=Variable(torch.from_numpy(x_np[np.newaxis,:,np.newaxis]))
	y=Variable(torch.from_numpy(y_np[np.newaxis,:,np.newaxis]))
	prediction,h_state=rnn(x,h_state)
	h_state=Variable(h_state.data)
	loss=loss_func(prediction,y)
	optimizer.zero_grad()
	loss.backward()
	optimizer.step()
	#画图
	plt.plot(steps, y_np.flatten(), 'r-')
	plt.plot(steps, prediction.data.numpy().flatten(), 'b-')
	plt.draw();
	plt.pause(0.05)
	plt.ioff()
	plt.show()

拟合图像如下