运行环境为:
Ubuntu16.04 Python2.7 PyTorch-cpu版本
pip install torch1.3.1+cpu torchvision0.4.2+cpu -f https://download.pytorch.org/whl/torch_stable.html
PyTorch 自动求导机制
实例测试
# -*- coding:utf-8 -*-
import torch as t
from torch.autograd import Variable as V
from matplotlib import pyplot as plt
from IPython import display
t.manual_seed(1000)
def get_fake_data(batch_size=8):
x=t.rand(batch_size,1)*20
y=x*2+(1+t.randn(batch_size,1))*3
return x,y
#print(y)
x,y=get_fake_data()
plt.scatter(x.squeeze().numpy(),y.squeeze().numpy())
#plt.show()
#随机初始化参数
w=V(t.rand(1,1),requires_grad=True)
b=V(t.zeros(1,1),requires_grad=True)
lr=0.001
for ii in range(8000):
x,y=get_fake_data()
x,y=V(x),V(y)
#forward:计算loss
y_pred=x.mm(w)+b.expand_as(y)
loss=0.5*(y_pred-y)**2
loss=loss.sum()
#backward: 自动计算梯度
loss.backward()
#更新参数
w.data.sub_(lr*w.grad.data)
b.data.sub_(lr*b.grad.data)
#梯度清零
w.grad.data.zero_()
b.grad.data.zero_()
if ii%1000==0:
#画图
display.clear_output(wait=True)
x=t.arange(0,20).view(-1,1)
y=x.mm(w.data)+b.data.expand_as(x)
plt.plot(x.numpy(),y.numpy())
x2,y2=get_fake_data(batch_size=20)
plt.scatter(x2.numpy(),y2.numpy())
plt.xlim(0,20)
plt.ylim(0,41)
plt.show()
plt.pause(0.5)
print(w.data.squeeze()[0],b.data.squeeze()[0])
源代码有一个报错:
File “variable_linear_regression.py”, line 47, in
y=x.mm(w.data)+b.data.expand_as(x)
RuntimeError: Expected object of scalar type Long but got scalar type Float for argument #2 ‘mat2’ in call to _th_mm
修改更正为:
y=x.mm(w.data.type('torch.LongTensor'))+b.data.expand_as(x)
小试牛刀,亲测可用。理解PyTorch中,autograd模块中,反向传播中的自动求导。