import random
import torch
from d2l import torch as d2l
def synthtic_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)#给y加上一个随机噪音
return X,y.reshape(-1,1)
true_w=torch.tensor([2,-3.4])
true_b=4.2
features,labels=synthtic_data(true_w,true_b,1000)
print('features:',features[0],'\nlabels:',labels[0])
d2l.set_figsize()
d2l.plt.scatter(features[:,0].detach().numpy(),labels.detach().numpy(),1)#最后一个1是绘制点直径的大小
d2l.plt.show()
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]#torch的tensor中,一维张量可以做二维张量的下标
batch_size=10
for X,y in data_iter(batch_size,features,labels):
print(X,'\n',y)
break
w=torch.normal(0,0.01,size=(2,1),requires_grad=True)
b=torch.zeros(1,requires_grad=True)
def linreg(X,w,b):
return torch.matmul(X,w)+b
def squared_loss(y_hat,y):#定义损失函数
return (y_hat-y.reshape(y_hat.shape))**2/2
def sgd(params,lr,batch_size):
with torch.no_grad():
for param in params:
param-=lr*param.grad/batch_size
param.grad.zero_()#下一次计算梯度就不会和上一次相关
lr=0.01
num_epoches=4
net=linreg
loss=squared_loss
for epoch in range(num_epoches):
for X,y in data_iter(batch_size,features,labels):
l=loss(net(X,w,b),y)#X,y的小批量损失,l的形状不是一个标量,而是一个长度为batchsize的向量
l.sum().backward()
sgd(([w,b]),lr,batch_size)#每次只用batch_size个样本来更新参数
with torch.no_grad():#避免进行不必要的计算浪费显存
train_l=loss(net(features,w,b),labels)
print(f'epoch{epoch+1},loss {float(train_l.mean()):f}')
线性回归的全部代码见上图。本次采用的数据集为人造数据集,总样本个数为1000,batch_size为10。shuffle用来打乱数据集,实现随机梯度下降。