手写数字识别pytorch 附loss图和结果验证图

最新推荐文章于 2024-08-09 10:27:20 发布

接近真理的真

最新推荐文章于 2024-08-09 10:27:20 发布

阅读量2k

点赞数

分类专栏： torch 文章标签： pytorch

本文链接：https://blog.csdn.net/qq_52740539/article/details/121412667

版权

import torch
import torchvision
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim
import numpy as np
import os
import matplotlib.pyplot as plt

from torchvision import datasets, transforms
n_epochs = 3
random_seed = 1
BATCH_SIZE=512
torch.manual_seed(random_seed)
train_losses = []
train_counter = []
train_loader = torch.utils.data.DataLoader(
    datasets.MNIST('data', train=True, download=True,
                   transform=transforms.Compose([
                       transforms.ToTensor(),
                       transforms.Normalize((0.1307,), (0.3081,))
                   ])),
    batch_size=BATCH_SIZE, shuffle=True)

test_loader = torch.utils.data.DataLoader(
    datasets.MNIST('data', train=False, transform=transforms.Compose([
        transforms.ToTensor(),
        transforms.Normalize((0.1307,), (0.3081,))
    ])),
    batch_size=BATCH_SIZE