【深度学习】torchvision.utils.make_grid() 拼接图片

最新推荐文章于 2024-07-17 16:39:53 发布

梅津太郎

最新推荐文章于 2024-07-17 16:39:53 发布

阅读量1.3k

点赞数

分类专栏：动手学深度学习 python 文章标签：深度学习 pytorch python

本文链接：https://blog.csdn.net/gaocui883/article/details/121028170

版权

动手学深度学习同时被 2 个专栏收录

37 篇文章 2 订阅

订阅专栏

python

30 篇文章 0 订阅

订阅专栏

import torch
from torchvision import datasets, transforms
from torch.utils.data import DataLoader
import torchvision
import matplotlib.pyplot as plt
 
def image_show(images):
    images = images.numpy()
    images = images.transpose((1, 2, 0))
    print(images.shape)
    plt.imshow(images)
    plt.show()
 
def main():
    train_dataset = datasets.MNIST(root='./datasets', train=False, download=False,
                                   transform=transforms.ToTensor())
    train_loader = DataLoader(train_dataset, batch_size=32, shuffle=False)
 
    device = torch.device('cuda:0')
    # for batch_idx, (inputs, targets) in enumerate(train_loader):
    #     inputs = inputs.to(device)
    #     print(inputs.shape)
    inputs, targets = next(iter(train_loader))
    print(inputs.shape)
    print(targets.shape)
 
    images = torchvision.utils.make_grid(inputs)
    print(f'images.shape:{images.shape}')
    image_show(images)
 
 
 
 
if __name__=='__main__':
    main()