pytorch反归一化

pytorch反归一化

pytorch在进行图像数据加载时，使用torchvision.transform中的Normalize进行归一化操作，但有时候我们需要在加载之后用到归一化前的数据。
torchvision中，Normalize使用torchvision.transforms.functional.normalize函数归一化，normalize函数如下

def normalize(tensor: Tensor, mean: List[float], std: List[float], inplace: bool = False) -> Tensor:
    """Normalize a tensor image with mean and standard deviation.

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

    See :class:`~torchvision.transforms.Normalize` for more details.

    Args:
        tensor (Tensor): Tensor image of size (C, H, W) or (B, C, H, W) to be normalized.
        mean (sequence): Sequence of means for each channel.
        std (sequence): Sequence of standard deviations for each channel.
        inplace(bool,optional): Bool to make this operation inplace.

    Returns:
        Tensor: Normalized Tensor image.
    """
    if not isinstance(tensor, torch.Tensor):
        raise TypeError('Input tensor should be a torch tensor. Got {}.'.format(type(tensor)))

    if tensor.ndim < 3:
        raise ValueError('Expected tensor to be a tensor image of size (..., C, H, W). Got tensor.size() = '
                         '{}.'.format(tensor.size()))

    if not inplace:
        tensor = tensor.clone()

    dtype = tensor.dtype
    mean = torch.as_tensor(mean, dtype=dtype, device=tensor.device)
    std = torch.as_tensor(std, dtype=dtype, device=tensor.device)
    if (std == 0).any():
        raise ValueError('std evaluated to zero after conversion to {}, leading to division by zero.'.format(dtype))
    if mean.ndim == 1:
        mean = mean.view(-1, 1, 1)
    if std.ndim == 1:
        std = std.view(-1, 1, 1)
    tensor.sub_(mean).div_(std)
    return tensor

源码地址
其中 tensor.sub_(mean).div_(std)就是归一化的操作：减去均值除以方差。反归一化时，只需将这行代码改为乘以方差加上均值即可，代码如下

def unnormalize(tensor: Tensor, mean: List[float], std: List[float], inplace: bool = False) -> Tensor:
    """Unnormalize a tensor image with mean and standard deviation.

    Args:
        tensor (Tensor): Tensor image of size (C, H, W) or (B, C, H, W) to be normalized.
        mean (sequence): Sequence of means for each channel.
        std (sequence): Sequence of standard deviations for each channel.
        inplace(bool,optional): Bool to make this operation inplace.

    Returns:
        Tensor: Normalized Tensor image.
    """
    if not isinstance(tensor, torch.Tensor):
        raise TypeError('Input tensor should be a torch tensor. Got {}.'.format(type(tensor)))

    if tensor.ndim < 3:
        raise ValueError('Expected tensor to be a tensor image of size (..., C, H, W). Got tensor.size() = '
                         '{}.'.format(tensor.size()))

    if not inplace:
        tensor = tensor.clone()

    dtype = tensor.dtype
    mean = torch.as_tensor(mean, dtype=dtype, device=tensor.device)
    std = torch.as_tensor(std, dtype=dtype, device=tensor.device)
    if (std == 0).any():
        raise ValueError('std evaluated to zero after conversion to {}, leading to division by zero.'.format(dtype))
    if mean.ndim == 1:
        mean = mean.view(-1, 1, 1)
    if std.ndim == 1:
        std = std.view(-1, 1, 1)
    tensor.mul_(std).add_(mean)
    return tensor
    

验证：

img_cv2为归一化前的tensor，x为归一化后再反归一化的tensor，可以看出两个tensor的内容相同。
(注意函数也支持批量输入，即B×C×H×W大小的tensor)