torchvision.transforms

1. torchvision.transforms.Compose(transforms)

Composes several transforms together.组合多个转换操作。

Parameters:	transforms (list of Transform objects) – list of transforms to compose.

Example

>>> transforms.Compose([
>>>     transforms.CenterCrop(10),
>>>     transforms.ToTensor(),
>>> ])

2.transforms.ToTensor()

将数据转换成torch的张量格式。

3.torchvision.transforms.Normalize(mean, std, inplace=False)

Normalize a tensor image with mean and standard deviation. Given mean: (M1,...,Mn) and std: (S1,..,Sn) for n channels, this transform will normalize each channel of the input torch.*Tensor i.e. input[channel] =(input[channel] - mean[channel]) / std[channel]

归一化一个张量图像，n通道的mean均值和std标准差。

NOTE

This transform acts out of place, i.e., it does not mutates the input tensor.

Parameters:	mean (sequence) – Sequence of means for each channel. std (sequence) – Sequence of standard deviations for each channel.

__call__(tensor)

Parameters:	tensor (Tensor) – Tensor image of size (C, H, W) to be normalized.
Returns:	Normalized Tensor image.
Return type:	Tensor