- 工具:jypter
- 语言:python
%matplotlib inline
import random
import torch
from d2l import torch as d2l
-
%matplotlib inline
将图表内嵌到notebook中,省略plt.show() ipython环境中魔法命令以%或者%%开头 -
d2l包 《动手学深度学习》提供的包
def synthetic_data(w,b,num_examples):
X=torch.normal(0,1,(num_examples,len(w)))
y=torch.matmul(X,w)+b
y+=torch.normal(0,0.01,y.shape)
return X,y.reshape(-1,1)
- 函数功能:用来产生数据集,其中y=Wx+b+e 噪声
- X是normal 产生正态分布的数据
true_w=torch.tensor([2,-3.4])
true_b=4.2
features,labels=synthetic_data(true_w,true_b,1000)
- 调用函数,给定W,b产生带噪声干扰的数据,其中features,labels分别为特征数据和标签数据
#训练样本
d2l.set_figsize()
d2l.plt.scatter(features[:,1].detach().numpy(),
labels.detach().numpy(),1)
- 绘制图像,调用了d2l包中的函数,实质上还是matplotlib绘图
- detach使requirse_grad=false
#读取小批量
def data_iter(batch_size,features,labels):
num_examples=len(features)
indices=list(range(num_examples))
random.shuffle(indices)
for i in range(0,num_examples,batch_size):
batch_indices=torch.tensor(indices[i:min(i+batch_size,num_examples)])
yield features[batch_indices],labels[batch_indices]
- 产生一个索引值的随机列表,依次取n个索引值,就可以获得n个随机的数据
batch_size=10
for X,y in data_iter(batch_size,features,labels):
print(X,'\n',y)
break
#定义初始化模型参数
w=torch.normal(0,0.01,size=(2,1),requires_grad=True)
b=torch.zeros(1,requires_grad=True)
#线性回归模型
def linreg(X,w,b):
return torch.matmul(X,w)+b
#定义损失函数
def squared_loss(y_hat,y):
return (y_hat-y.reshape(y_hat.shape))**2/2
- 损失函数没有进行平均,在优化算法中进行了平均
#优化算法,梯度下降
def sgd(params,lr,batch_size):
with torch.no_grad():
for param in params:
param-=lr*param.grad/batch_size
param.grad.zero_()
L ( w , b ) = ∑ ( y ^ − W b X ) 2 / 2 L_{(w,b)}=\sum(\hat{y}-W_bX) ^2/2 L(w,b)=∑(y^−WbX)2/2
w = w − η ∗ ∂ L ( w , b ) ∂ W b / s i z e w=w-\eta*\frac{\partial{L_{(w,b)}}}{\partial{W_b}}/size w=w−η∗∂Wb∂L(w,b)/size
#训练过程
lr=0.03
num_epochs=3
net=linreg
loss=squared_loss
for epoch in range(num_epochs):
for X,y in data_iter(batch_size,features,labels):
l=loss(net(X,w,b),y)
l.sum().backward()
sgd([w,b],lr,batch_size)
with torch.no_grad():
train_1=loss(net(features,w,b),labels)
print(f'epoch{epoch+1},loss{float(train_1.mean()):f}')
print(f'w的估计误差:{true_w-w.reshape(true_w.shape)}')
print(f'b的估计误差:{true_b-b}')