从linear_regression中看detach()的作用

最新推荐文章于 2024-02-13 00:33:55 发布

小饼干超人

最新推荐文章于 2024-02-13 00:33:55 发布

阅读量872

点赞数

分类专栏： pytorch 文章标签： linear regression detach

本文链接：https://blog.csdn.net/m0_37586991/article/details/88373417

版权

pytorch 专栏收录该内容

21 篇文章 5 订阅

订阅专栏

文章目录

linear_regression.py

import torch
import torch.nn as nn
import numpy as np
import matplotlib.pyplot as plt

# Hyper-parameters
input_size = 1
output_size = 1
num_epochs = 60
learning_rate = 0.001

# Toy dataset
x_train = np.array([[3.3], [4.4], [5.5], [6.71], [6.93], [4.168], 
                    [9.779], [6.182], [7.59], [2.167], [7.042], 
                    [10.791], [5.313], [7.997], [3.1]], dtype=np.float32)

y_train = np.array([[1.7], [2.76], [2.09], [3.19], [1.694], [1.573], 
                    [3.366], [2.596], [2.53], [1.221], [2.827], 
                    [3.465], [1.65], [2.904], [1.3]], dtype=np.float32)

# Linear regression model
model = nn.Linear(input_size, output_size)

# Loss and optimizer
criterion = nn.MSELoss()
optimizer = torch.optim.SGD(model.parameters(), lr=learning_rate)  

# Train the model
for epoch in range(num_epochs):
    # Convert numpy arrays to torch tensors
    inputs = torch.from_numpy(x_train)
    targets = torch.from_numpy(y_train)

    # Forward pass
    outputs = model(inputs)
    loss = criterion(outputs, targets)
    
    # Backward and optimize
    optimizer.zero_grad()
    loss.backward()
    optimizer.step()
    
    if (epoch+1) % 5 == 0:
        print ('Epoch [{}/{}], Loss: {:.4f}'.format(epoch+1, num_epochs, loss.item()))

# Plot the graph
predicted = model(torch.from_numpy(x_train)).detach().numpy()
plt.plot(x_train, y_train, 'ro', label='Original data')
plt.plot(x_train, predicted, label='Fitted line')
plt.legend()
plt.show()

# Save the model checkpoint
torch.save(model.state_dict(), 'model.ckpt')

detach()的用法

# 把x_train带入训练好的model，得到结果，相当于是上面for循环中的outputs
predicted = model(torch.from_numpy(x_train)).detach().numpy()

将这行代码拆解开：

predicted=model(torch.from_numpy(x_train))
predicted_2=predicted.detach()
predicted_3=predicted_2.numpy()
print('predicted_type: \n',type(predicted))
print('predicted: \n',predicted)
print('predicted_2_type: \n',type(predicted_2))
print('predicted_2: \n',predicted_2)
print('predicted_3_type: \n',type(predicted_3))
print('predicted_3: \n',predicted_3)

可以看到，detach()把predicted后面的grad_fn=<AddmmBackward>去掉了，也就是起到截断反向传播的梯度流的作用

predicted: 
 <class 'torch.Tensor'>
 tensor([[1.1968],
        [1.6009],
        [2.0050],
        [2.4496],
        [2.5304],
        [1.5156],
        [3.5771],
        [2.2556],
        [2.7729],
        [0.7805],
        [2.5715],
        [3.9489],
        [1.9363],
        [2.9224],
        [1.1233]], grad_fn=<AddmmBackward>)
predicted_2: 
 <class 'torch.Tensor'>
 tensor([[1.1968],
        [1.6009],
        [2.0050],
        [2.4496],
        [2.5304],
        [1.5156],
        [3.5771],
        [2.2556],
        [2.7729],
        [0.7805],
        [2.5715],
        [3.9489],
        [1.9363],
        [2.9224],
        [1.1233]])
predicted_3: 
 <class 'numpy.ndarray'>
 [[ 1.19675481]
 [ 1.60088432]
 [ 2.0050137 ]
 [ 2.44955611]
 [ 2.53038192]
 [ 1.5156498 ]
 [ 3.57707739]
 [ 2.25557399]
 [ 2.77285981]
 [ 0.78050154]
 [ 2.57152963]
 [ 3.94887662]
 [ 1.93631184]
 [ 2.9223876 ]
 [ 1.12327671]]