pytorch搭建卷积神经网络【第十二课_自动梯度】

---

前言

案例代码https://github.com/2012Netsky/pytorch_cnn/blob/main/4_time_series_bikes.ipynb

一、requires_grad=True自动梯度

二、计算损失前为none

三、 计算损失后不为none

四、梯度清零

五、不计算梯度 更新参数

#!/usr/bin/env python
# coding: utf-8

get_ipython().run_line_magic('matplotlib', 'inline')
import numpy as np
import torch
torch.set_printoptions(edgeitems=2)


t_c = torch.tensor([0.5, 14.0, 15.0, 28.0, 11.0, 8.0,
                    3.0, -4.0, 6.0, 13.0, 21.0])
t_u = torch.tensor([35.7, 55.9, 58.2, 81.9, 56.3, 48.9,
                    33.9, 21.8, 48.4, 60.4, 68.4])
t_un = 0.1 * t_u



def model(t_u, w, b):
    return w * t_u + b



def loss_fn(t_p, t_c):
    squared_diffs = (t_p - t_c)**2
    return squared_diffs.mean()


# requires_grad=True自动梯度 计算损失后会自动梯度
params = torch.tensor([1.0, 0.0], requires_grad=True)


# 为计算损失前为none
params.grad is None


# 损失计算
loss = loss_fn(model(t_u, *params), t_c)
# 梯度计算
loss.backward()
# 梯度值（w b）
params.grad



# 计算损失后不为none
params.grad is None


# 梯度清零
if params.grad is not None:
    params.grad.zero_()

def training_loop(n_epochs, learning_rate, params, t_u, t_c):
    for epoch in range(1, n_epochs + 1):
        if params.grad is not None:  # <1>
            params.grad.zero_()
        
        t_p = model(t_u, *params) 
        loss = loss_fn(t_p, t_c)
        loss.backward()
        
        # 不计算梯度 更新参数
        with torch.no_grad():  # <2>
            params -= learning_rate * params.grad

        if epoch % 500 == 0:
            print('Epoch %d, Loss %f' % (epoch, float(loss)))
            
    return params

training_loop(
    n_epochs = 5000, 
    learning_rate = 1e-2, 
    params = torch.tensor([1.0, 0.0], requires_grad=True), # <1> 
    t_u = t_un, # <2> 
    t_c = t_c)

总结