PyTorch -- RNN 快速实践

• RNN Layer torch.nn.RNN(input_size,hidden_size,num_layers,batch_first)

  • input_size: 输入的编码维度
  • hidden_size: 隐含层的维数
  • num_layers: 隐含层的层数
  • batch_first: ·True 指定输入的参数顺序为：
    • x：[batch, seq_len, input_size]
    • h0：[batch, num_layers, hidden_size]

• RNN 的输入：

  • x：[seq_len, batch, input_size]
    • seq_len: 输入的序列长度
    • batch: batch size 批大小
  • h0：[num_layers, batch, hidden_size]

• RNN 的输出：

  • y: [seq_len, batch, hidden_size]

---

---

• 实战之预测 正弦曲线：以下会以此为例，演示 RNN 预测任务的部署
  

  • 步骤一：确定 RNN Layer 相关参数值并基于此创建 Net

    import numpy as np
    from matplotlib import pyplot as plt
    
    import torch
    import torch.nn as nn
    import torch.optim as optim
    
    
    seq_len     = 50
    batch       = 1
    num_time_steps = seq_len
    
    input_size  = 1
    output_size = input_size
    hidden_size = 10  	
    num_layers = 1  	
    batch_first = True 
    
    class Net(nn.Module):  ## model 定义
    	def __init__(self):
    		super(Net, self).__init__()
    		self.rnn = nn.RNN(
    		input_size=input_size,
    		hidden_size=hidden_size,
    		num_layers=num_layers,
    		batch_first=batch_first)
    		# for p in self.rnn.parameters():
    		# 	nn.init.normal_(p, mean=0.0, std=0.001)
    		self.linear = nn.Linear(hidden_size, output_size)
    
    	def forward(self, x, hidden_prev):
    		out, hidden_prev = self.rnn