网站首次备案 多久,中国纪检监察报地址,网站设计的原始资料,中国网站建设公司排行简介#xff1a; 主要介绍一下 sin(x)#xff1a; 为 数据 cos(x): 为对应的label 项目包括两个文件 main.py: 模型的训练#xff0c;验证#xff0c;参数保存 lstm.py 模型的构建 目录#xff1a; lstm.py main.py 一 lstm.py # -*- coding: utf-8 -*-主要介绍一下 sin(x) 为 数据 cos(x): 为对应的label 项目包括两个文件 main.py: 模型的训练验证参数保存 lstm.py 模型的构建 目录 lstm.py main.py 一 lstm.py # -*- coding: utf-8 -*-Created on Tue Aug 8 14:01:15 2023author: chengxf2
import torch
import torch.nn as nnclass LSTM(nn.Module):def __init__(self, input_dim, hidden_dim, num_lay, b_first):super(LSTM,self).__init__()self.lstm nn.LSTM(input_size input_dim, hidden_size hidden_dim, num_layers num_lay, batch_firstb_first)self.linear nn.Linear(hidden_dim, 1)def forward(self, X):#X.shape:[batch_size1, seq_num256, input_size1]output, (hidden, cell) self.lstm(X)outs []seq_num output.size(1)#output:[batch_size, seq_num, hidden_dim64]#hidden.shape[num_layer, batch_size, hiden_size]#print(\n output.shape,output.shape)#print(\n hidden.shape,hidden.shape)for time_step in range(seq_num):#h.shape[batch, hidden_dim]h output[:,time_step,:]#print(\n h,h.shape)out self.linear(h)outs.append(out)#沿着一个新维度对输入张量序列进行连接。 #[batch, seq_num, 1]pred torch.stack(outs, dim1)return pred 二 main.py
import numpy as np
from matplotlib import pyplot as plt
import torch
from lstm import LSTM
import torch.nn as nn
from torch.nn import functional as F
from torch import optim
import timedef showDiff(pred, label, steps):plt.figure()plt.rcParams[font.family] SimHei # 正常显示中文plt.title(预测值 and 真实值, fontsize18)plt.plot(steps, pred.cpu().data.numpy().flatten(),colorr,label预测值)plt.plot(steps, label.cpu().data.numpy().flatten(), colorg,label真实值)plt.legend(locbest)plt.show()def get_data(epoch):TIME_STEP 256start, end epoch*np.pi, epoch*np.pi2*np.pisteps np.linspace(start,end,TIME_STEP,dtypenp.float32)sin_x np.sin(steps)cos_x np.cos(steps)sinx_torch torch.from_numpy(sin_x[np.newaxis, :, np.newaxis])if torch.cuda.is_available():sinx_torch torch.from_numpy(sin_x[np.newaxis,:,np.newaxis]).cuda()# print(sinx_input.shape:,sinx_input.shape)cosx_lable torch.from_numpy(cos_x[np.newaxis, :, np.newaxis]).cuda()else:sinx_torch torch.from_numpy(sin_x[np.newaxis, :, np.newaxis])# print(sinx_input.shape:,sinx_input.shape)cosx_lable torch.from_numpy(cos_x[np.newaxis, :, np.newaxis])# [batch,seq_num,input_size] (1,256,1)return sinx_torch,cosx_lable,stepsdef eval(model):#等同于 self.train(False) 就是评估模式。#在评估模式下batchNorm层dropout层等用于优化训练而添加的网络层会被关闭从而使得评估时不会发生偏移model.eval()test_data,test_label,steps get_data(2)with torch.no_grad():y_pred model(test_data)showDiff(y_pred, test_label, steps)def train(model,maxIter,criterion):训练模型----------model : lstm 模型.maxIter : 迭代次数.criterion : 损失函数------#作用是启用 batch normalization 和 dropoutmodel.train()time_stamp time.time()for epoch in range(maxIter):sinx_torch,cosx_lable,steps get_data(epoch)y_pre model(sinx_torch) loss criterion(y_pre,cosx_lable)optimzer.zero_grad()loss.backward()optimzer.step()if epoch%1000:data_time_interval time.time() - time_stampprint(epoch: %d loss: %7.3f interval: %6.2f%(epoch, loss.detach().numpy(),data_time_interval))#torch.save(model.state_dict(), model_params.pth) showDiff(y_pre, cosx_lable,steps)if __name__ __main__:input_dim 1hidden_dim 64num_layers 2batch_first TruemaxIter 3000model LSTM(input_dim, hidden_dim, num_layers, batch_first)DEVICE torch.device(cuda if torch.cuda.is_available() else cpu)optimzer optim.Adam(model.parameters(),lr0.0001,weight_decay0.00001)criterion nn.MSELoss()model.to(DEVICE)criterion.to(DEVICE)train(model,maxIter,criterion)#model.load_state_dict(torch.load(model_params.pth,map_locationcpu))#eval(model) 参考
pytorch利用rnn通过sin预测cos 利用lstm预测手写数字_pytorch lstm cos_薛定谔的智能的博客-CSDN博客
