[1]中的程序可以改成如下對應的Tensor形式：

import torch


dtype = torch.FloatTensor
# dtype = torch.cuda.FloatTensor # Uncomment this to run on GPU

# N is batch size; D_in is input dimension;
# H is hidden dimension; D_out is output dimension.
N, D_in, H, D_out = 64, 1000, 100, 10

# Create random input and output data
x = torch.randn(N, D_in).type(dtype)
y 
 
= torch.randn(N, D_out).type(dtype)

# Randomly initialize weights
w1 = torch.randn(D_in, H).type(dtype)
w2 = torch.randn(H, D_out).type(dtype)

learning_rate = 1e-6
for t in range(500):
    # Forward pass: compute predicted y
    h = x.mm(w1)
    h_relu = h.clamp(min=0)
    y_pred = h_relu.mm(w2)

    
 
# Compute and print loss 
    loss = (y_pred - y).pow(2).sum()
    print(t, loss) 

    # Backprop to compute gradients of w1 and w2 with respect to loss
    grad_y_pred = 2.0 * (y_pred - y)
    grad_w2 = h_relu.t().mm(grad_y_pred)
    grad_h_relu = grad_y_pred.mm(w2.t())
    grad_h = grad_h_relu.clone() # copy一份，硬拷貝 可以用這樣的代碼測試 a=torch.Tensor(3) b=a.clone() b[2]=100 b[2] b[2] 
    grad_h[h 
 
< 0] = 0
    grad_w1 = x.t().mm(grad_h) #x.t()表示x的轉置，x沒變；如果想改變x,x.t_() _表示原地操作

    # Update weights using gradient descent
    w1 -= learning_rate * grad_w1
    w2 -= learning_rate * grad_w2

有兩個函數需要說明 h.clamp(min=0)

clamp表示夾緊，夾住的意思，torch.clamp(input,min,max,out=None)-> Tensor

將input中的元素限制在[min,max]範圍內並返回一個Tensor

用法：

下面的doc有錯誤： 應為

torch.clamp(input,min,*,out=None)->Tensor

pytorch教程[2] Tensor的使用