Tensorboard的使用——04使用addimages将CIFAR10中的图片添加到Tensorboard中

数据集下载地址：

http://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz

1、设置transforms

dataset_transform = torchvision.transforms.Compose([
    torchvision.transforms.ToTensor()
])

2、下载数据集，使用DataLoader加载
batch_size指每次随机抓取64个图片

test_data = torchvision.datasets.CIFAR10(root="./test10_dataset", train=False, transform=dataset_transform)
test_loader = DataLoader(dataset=test_data, batch_size=64, shuffle=True, num_workers=0, drop_last=False)

3、实例化SummaryWriter，并指明日志存放路径

writer = SummaryWriter("CIFAR10")

4、使用循环批量添加图片

step = 0
for data in test_loader:
    img, target = data
    writer.add_images("test_set", img, step)
    step = step + 1

5、关闭writer

writer.close()

6、在terminal中使用

tensorboard --logdir=CIFAR10

代码如下：

import torchvision
from torch.utils.data import DataLoader
from torch.utils.tensorboard import SummaryWriter

dataset_transform = torchvision.transforms.Compose([
    torchvision.transforms.ToTensor()
])
test_data = torchvision.datasets.CIFAR10(root="./test10_dataset", train=False, transform=dataset_transform)
test_loader = DataLoader(dataset=test_data, batch_size=64, shuffle=True, num_workers=0, drop_last=False)
writer = SummaryWriter("p11")
step = 0
for data in test_loader:
    img, target = data
    writer.add_images("test_set", img, step)
    step = step + 1
writer.close()

在terminal中使用：

tensorboard --logdir=CIFAR10

tensorboard ：

step0-step155组数据每组64张图片

step156为最后一组数据只有16个