pytorch中的上采样

Upsample

Upsamples a given multi-channel 1D (temporal), 2D (spatial) or 3D (volumetric) data.

torch.nn.Upsample(size=None, scale_factor=None, mode=‘nearest‘, align_corners=None)

UpsamplingNearest2d

torch.nn.UpsamplingNearest2d(size=None, scale_factor=None)

UpsamplingBilinear2d

torch.nn.UpsamplingBilinear2d(size=None, scale_factor=None)

ConvTranspose

torch.nn.ConvTranspose1d(in_channels, out_channels, kernel_size, stride=1, padding=0, output_padding=0, groups=1, bias=True, dilation=1, padding_mode=‘zeros‘)

torch.nn.ConvTranspose2d(in_channels, out_channels, kernel_size, stride=1, padding=0, output_padding=0, groups=1, bias=True, dilation=1, padding_mode=‘zeros‘)

torch.nn.ConvTranspose3d(in_channels, out_channels, kernel_size, stride=1, padding=0, output_padding=0, groups=1, bias=True, dilation=1, padding_mode=‘zeros‘)

pytorch中的上采样

原文：https://www.cnblogs.com/jiangkejie/p/12919155.html