%matplotlib inline
import random
import torch
from d2l import torch as d2l
"""训练样本"""
def synthetic_data(w,b,num_examples):
"""生成y=Xw+b+噪声"""
X = torch.normal(0,1,(num_examples,len(w)))
"""torch.normal(0,1)表示生成均值为0 方差为1的随机数 列度是w 行数是样本 """
y = torch.matmul(X,w)+b
"""matmul表示矩阵相乘 """
y += torch.normal(0,0.01,y.shape)
"""表示生成均值为0 方差为0.01的随机数 列度是w 行数是样本 形状与y的形状一样"""
return X,y.reshape((-1,1))
"""reshape -1表示自动计算 后面1表示固定1列 即列向量为1"""
true_w = torch.tensor([2,-3.4])
"""reshape -1表示自动计算 后面1表示固定1列 即列向量为1"""
true_b = 4.2
features,labels = synthetic_data(true_w ,true_b,1000)
"""生成特征以及标注"""
"""对训练样本进行展示"""
print('features:',features[0],'\nlabel:',labels[0])
d2l.set_figsize()
d2l.plt.scatter(features[:,1].detach().numpy(),labels.detach().numpy(),1);
训练样本显示结果
"""data_iter函数接收批量大小、特征矩阵哈标签向量作为输入,生成大小为batch_size的小批量"""
def data_iter(batch_size,features,labels):
num_examples = len(features)
indices = list(range(num_examples))
random.shuffle(indices)
"""打乱下标,以便后面随即访问样本"""
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]
"""yield就是return返回一个值,并记住返回值的位置,下次迭代从此位置后开始-------return返回函数会终止,yield不会"""
batch_size = 10
for X,y in data_iter(batch_size,features,labels):
print(X,'\n',y)
break
此处运行结果如下
"""定义初始化参数模型"""
w = torch.normal(0,0.01,size=(2,1),requires_grad=True)
"""requires_grad=True 需要梯度计算"""
b = torch.zeros(1,requires_grad=True)
"""偏差"""
"""定义模型"""
def linreg(X,w,b):
"""线性回归模型"""
return torch.matmul(X,w)+b
"""定义损失函数"""
def squared_loss(y_hat,y):
"""均方损失"""
return (y_hat-y.reshape(y_hat.shape))**2/2
"""定义优化算法"""
def sgd(params,lr,batch_size):
"""小批量随机梯度下降"""
with torch.no_grad():
for param in params:
param -= lr * param.grad / batch_size
param.grad.zero_()
训练过程
"""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)
"""使用参数的梯度更新参数"""
with torch.no_grad():
train_l = loss(net(features,w,b),labels)
print(f'epoch{epoch+1},loss{float(train_l.mean()):f}')
比较真实参数和通过训练学到的参数来评估训练的成功程度
"""比较真实参数和通过训练学到的参数来评估训练的成功程度"""
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}')
完整代码如下
%matplotlib inline
import random
import torch
from d2l import torch as d2l
"""训练样本"""
def synthetic_data(w,b,num_examples):
"""生成y=Xw+b+噪声"""
X = torch.normal(0,1,(num_examples,len(w)))
"""torch.normal(0,1)表示生成均值为0 方差为1的随机数 列度是w 行数是样本 """
y = torch.matmul(X,w)+b
"""matmul表示矩阵相乘 """
y += torch.normal(0,0.01,y.shape)
"""表示生成均值为0 方差为0.01的随机数 列度是w 行数是样本 形状与y的形状一样"""
return X,y.reshape((-1,1))
"""reshape -1表示自动计算 后面1表示固定1列 即列向量为1"""
true_w = torch.tensor([2,-3.4])
"""reshape -1表示自动计算 后面1表示固定1列 即列向量为1"""
true_b = 4.2
features,labels = synthetic_data(true_w ,true_b,1000)
"""生成特征以及标注"""
print('features:',features[0],'\nlabel:',labels[0])
d2l.set_figsize()
d2l.plt.scatter(features[:,1].detach().numpy(),labels.detach().numpy(),1);
def data_iter(batch_size,features,labels):
num_examples = len(features)
indices = list(range(num_examples))
random.shuffle(indices)
"""打乱下标,以便后面随即访问样本"""
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]
"""yield就是return返回一个值,并记住返回值的位置,下次迭代从此位置后开始-------return返回函数会终止,yield不会"""
batch_size = 10
for X,y in data_iter(batch_size,features,labels):
print(X,'\n',y)
break
"""定义初始化参数模型"""
w = torch.normal(0,0.01,size=(2,1),requires_grad=True)
"""requires_grad=True 需要梯度计算"""
b = torch.zeros(1,requires_grad=True)
"""偏差"""
"""定义模型"""
def linreg(X,w,b):
"""线性回归模型"""
return torch.matmul(X,w)+b
"""定义损失函数"""
def squared_loss(y_hat,y):
"""均方损失"""
return (y_hat-y.reshape(y_hat.shape))**2/2
"""定义优化算法"""
def sgd(params,lr,batch_size):
"""小批量随机梯度下降"""
with torch.no_grad():
for param in params:
param -= lr * param.grad / batch_size
param.grad.zero_()
"""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)
"""使用参数的梯度更新参数"""
with torch.no_grad():
train_l = loss(net(features,w,b),labels)
print(f'epoch{epoch+1},loss{float(train_l.mean()):f}')
"""比较真实参数和通过训练学到的参数来评估训练的成功程度"""
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}')