线性回归
- 线性回归是对n维输入的加权,外加偏差
- 使用平方损失来衡量预测值和真实值得差异
- 线性回归有显示解
- 线性回归可以看做是单层神经网络
基础优化算法
- 梯度下降通过不断沿着反梯度方向更新参数求解
- 小批量梯度下降是深度学习默认的求解算法
- 两个重要的超参数是批量大小和学习率
从0实现梯度下降算法
import random
import torch
def create_data(w, b, nums_example):
X = torch.normal(0, 1, (nums_example, len(w)))
y = torch.matmul(X, w) + b
print("y_shape:", y.shape)
y += torch.normal(0, 0.01, y.shape)
return X, y.reshape(-1, 1)
true_w = torch.tensor([2, -3.4])
true_b = 4.2
features, labels = create_data(true_w, true_b, 1000)
def read_data(batch_size, features, lables):
nums_example = len(features)
indices = list(range(nums_example))
random.shuffle(indices)
for i in range(0, nums_example, batch_size):
index_tensor = torch.tensor(indices[i: min(i + batch_size, nums_example)])
yield features[index_tensor], lables[index_tensor]
batch_size = 10
for X, y in read_data(batch_size, features, labels):
print("X:", X, "\ny", y)
break;
w = torch.normal(0, 0.01, size=(2, 1), requires_grad=True)
b = torch.zeros(1, requires_grad=True)
def net(X, w, b):
return torch.matmul(X, w) + b
def loss(y_hat, y):
return (y_hat - y.reshape(y_hat.shape)) ** 2 / 2
def sgd(params, batch_size, lr):
with torch.no_grad():
for param in params:
param -= lr * param.grad / batch_size
param.grad.zero_()
lr = 0.03
num_epochs = 3
for epoch in range(0, num_epochs):
for X, y in read_data(batch_size, features, labels):
f = loss(net(X, w, b), y)
f.sum().backward()
sgd([w, b], batch_size, lr)
with torch.no_grad():
train_l = loss(net(features, w, b), labels)
print("w {0} \nb {1} \nloss {2:f}".format(w, b, float(train_l.mean())))
print("w误差 ", true_w - w, "\nb误差 ", true_b - b)
线性回归简单实现
import numpy as np
import torch
from torch.utils import data
from d2l import torch as d2l
from torch import nn
def load_array(data_arrays, batch_size, is_train=True):
"""构造一个 Pytorch数据迭代器"""
dataset = data.TensorDataset(*data_arrays)
return data.DataLoader(dataset, batch_size, shuffle=is_train)
true_w = torch.tensor([2, -3.4])
true_b = 4.2
features, labels = d2l.synthetic_data(true_w, true_b, 10000)
batch_size = 10
data_iter = load_array((features, labels), batch_size)
net = nn.Sequential(nn.Linear(2, 1))
net[0].weight.data.normal_(0, 0.01)
net[0].bias.data.fill_(0)
loss = nn.MSELoss()
trainer = torch.optim.SGD(net.parameters(), lr=0.03)
num_epochs = 3
for epoch in range(num_epochs):
for X, y in data_iter:
l = loss(net(X), y)
trainer.zero_grad()
l.backward()
trainer.step()
l = loss(net(features), labels)
print(f'epoch {epoch + 1}, loss {l:f}')
w = net[0].weight.data
b = net[0].bias.data
print('w的估计误差:', true_w - w.reshape(true_w.shape))
print('b的估计误差:', true_b - b)