如何在Pytorch中使用for循环长度作为优化参数

Question

我正在尝试将循环的长度注册为优化参数，以便优化程序将自动调整循环的长度。

这是一个示例代码（实际问题比较复杂，但是您会明白的）：

   <!-- Bootstrap CSS -->
    <link rel="stylesheet" href="https://stackpath.bootstrapcdn.com/bootstrap/4.5.2/css/bootstrap.min.css" integrity="sha384-JcKb8q3iqJ61gNV9KGb8thSsNjpSL0n8PARn9HuZOnIxN0hoP+VmmDGMN5t9UJ0Z" crossorigin="anonymous">
    
    <!-- Navbar -->
    <nav class="navbar navbar-dark bg-dark">
        <span class="navbar-brand mb-0 h1 mx-auto navbar-brand">Pixel Art Maker</span>
    </nav>

    <!-- Container to put grid in -->
    <div class="container shadow p-3 mb-5 bg-white rounded mt-5 main">
      
      <div class="container canvas d-flex-inline">
        
      </div>

      <form class="d-flex-inline w-50 mx-auto">
        <div class="d-flex">
          <div class="form-group w-25 ml-auto mr-2">
            <label class="d-block text-center" for="gridRows">Rows</label>
            <input type="number" class="form-control row-input" id="gridRows">
          </div>
          <div class="form-group w-25 mr-auto ml-2">
            <label class="d-block text-center" for="gridColumns">Columns</label>
            <input type="number" class="form-control column-input" id="gridColumns">
          </div>
        </div>  
        </form>
        <div class="form-group w-25 mx-auto">
          <button type="submit" class="btn btn-primary w-100">Submit</button>
        </div>
      <form class="d-flex-inline w-50 mx-auto">
        <div class="form-group w-25 mx-auto">
          <label for="color">Color</label>
          <input type="color" class="form-control colorChoice" id="color">
        </div>
      </form>
    </div>

<!-- Optional JavaScript -->
    <!-- jQuery first, then Popper.js, then Bootstrap JS, then local JS -->
    <script src="https://code.jquery.com/jquery-3.5.1.slim.min.js" integrity="sha384-DfXdz2htPH0lsSSs5nCTpuj/zy4C+OGpamoFVy38MVBnE+IbbVYUew+OrCXaRkfj" crossorigin="anonymous"></script>
    <script src="https://cdn.jsdelivr.net/npm/popper.js@1.16.1/dist/umd/popper.min.js" integrity="sha384-9/reFTGAW83EW2RDu2S0VKaIzap3H66lZH81PoYlFhbGU+6BZp6G7niu735Sk7lN" crossorigin="anonymous"></script>
    <script src="https://stackpath.bootstrapcdn.com/bootstrap/4.5.2/js/bootstrap.min.js" integrity="sha384-B4gt1jrGC7Jh4AgTPSdUtOBvfO8shuf57BaghqFfPlYxofvL8/KUEfYiJOMMV+rV" crossorigin="anonymous"></script>

优化似乎不起作用，这是输出：

import torch

positive_iter = torch.tensor([10.0], requires_grad=True)
negative_iter = torch.tensor([20.0], requires_grad=True)

optimizer = torch.optim.Adam([positive_iter, negative_iter], lr=0.02, betas=(0.5, 0.999))

for i in range(100):
    loss = torch.tensor([0.0], requires_grad=True)
    for g in range(int(positive_iter)):
        loss = loss + torch.rand(1)
    for d in range(int(negative_iter)):
        loss = loss - torch.rand(1) * 2

    loss = torch.abs(loss)
    loss.backward()
    optimizer.step()
    print(i, loss.item(), positive_iter.item(), negative_iter.item())

您能否建议如何使for循环长度成为优化参数。

谢谢