pytorch學習筆記---回歸問題1

目錄

• 概要
• 導包
• 導入資料
• • 展示資料
• 搭模組化型
• 定義計算步驟
• 輸出運算結果

概要

本節主要針對MNIST資料集的數字識别問題，寫出一個解決回歸問題的方法。初步體會機器學習的工作流程

導包

import  torch
from    torch import nn
from    torch.nn import functional as F
from    torch import optim

import  torchvision
from    matplotlib import pyplot as plt
#畫圖專用的檔案
from    utils import plot_image, plot_curve, one_hot
           

導入資料

batch_size = 512

# step1. load dataset加載資料集
train_loader = torch.utils.data.DataLoader(
    torchvision.datasets.MNIST('mnist_data', train=True, download=True,
                               transform=torchvision.transforms.Compose([
                                   torchvision.transforms.ToTensor(),
                                   torchvision.transforms.Normalize(
                                       (0.1307,), (0.3081,))
                               ])),
    batch_size=batch_size, shuffle=True)

test_loader = torch.utils.data.DataLoader(
    torchvision.datasets.MNIST('mnist_data/', train=False, download=True,
                               transform=torchvision.transforms.Compose([
                                   torchvision.transforms.ToTensor(),
                                   torchvision.transforms.Normalize(
                                       (0.1307,), (0.3081,))
                               ])),
    batch_size=batch_size, shuffle=False)
           

展示資料

x, y = next(iter(train_loader))
print(x.shape, y.shape, x.min(), x.max())
plot_image(x, y, 'image sample')
           

搭模組化型

class Net(nn.Module):

    def __init__(self):
        super(Net, self).__init__()

        # xw+b
        self.fc1 = nn.Linear(28*28, 256)
        self.fc2 = nn.Linear(256, 64)
        self.fc3 = nn.Linear(64, 10)

    def forward(self, x):
        # x: [b, 1, 28, 28]
        # h1 = relu(xw1+b1) 公式
        x = F.relu(self.fc1(x))
        # h2 = relu(h1w2+b2) 公式
        x = F.relu(self.fc2(x))
        # h3 = h2w3+b3 公式
        x = self.fc3(x)

        return x
           

定義計算步驟

net = Net()
# [w1, b1, w2, b2, w3, b3]
#優化器
optimizer = optim.SGD(net.parameters(), lr=0.01, momentum=0.9)

#記錄loss
train_loss = []

for epoch in range(3):

    for batch_idx, (x, y) in enumerate(train_loader):

        # x: [b, 1, 28, 28], y: [512]
        # [b, 1, 28, 28] => [b, 784] 從四維變換成二維
        x = x.view(x.size(0), 28*28)
        # => [b, 10]
        out = net(x)
        # [b, 10]
        y_onehot = one_hot(y)
        # loss = mse(out, y_onehot)
        loss = F.mse_loss(out, y_onehot)

        # 清零梯度
        optimizer.zero_grad()
        loss.backward()
        # w' = w - lr*grad 梯度更新
        optimizer.step()

        train_loss.append(loss.item())
        # 輸出
        if batch_idx % 10==0:
            print(epoch+1, batch_idx, loss.item())

plot_curve(train_loss)
# we get optimal [w1, b1, w2, b2, w3, b3]
           

輸出運算結果

plot_curve(train_loss)
# we get optimal [w1, b1, w2, b2, w3, b3]


total_correct = 0
for x,y in test_loader:
    x  = x.view(x.size(0), 28*28)
    out = net(x)
    # out: [b, 10] => pred: [b]
    pred = out.argmax(dim=1)
    correct = pred.eq(y).sum().float().item()
    total_correct += correct

total_num = len(test_loader.dataset)
acc = total_correct / total_num
print('test acc:', acc)