Use of Tensorboard - 04 Use addimages to add pictures in CIFAR10 to Tensorboard

Dataset download address:

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

1. Set transforms

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

2. Download the data set, and use DataLoader to load
batch_size, which means grabbing 64 pictures randomly each time

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. Instantiate SummaryWriter and specify the log storage path

writer = SummaryWriter("CIFAR10")

4. Use loop to add pictures in batches

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

5. Close the writer

writer.close()

6. Use in terminal

tensorboard --logdir=CIFAR10

code show as below:

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()

Use in terminal:

tensorboard --logdir=CIFAR10

tensorboard ：

step0-step155 sets of data each set of 64 pictures

step156 is the last set of data only 16