如何使用F.interpolate调整PyTorch中所有4个尺寸（NCHW）的大小？

Question

我一直在试图找出如何在张量中调整批次、通道、高度和宽度维度的大小。目前，我可以调整通道、高度和宽度尺寸，但批次尺寸保持不变

x = torch.ones(3,4,64,64)

x = F.interpolate(x.unsqueeze(0), size=(3,4,4), mode="trilinear").squeeze(0)

x.size() # (3,3,4,4) # batch dimension has not been resized.
# I need x to be resized so that it has a size of: (1,3,4,4)


# Is this a good idea?
x = x.permute(1,0,2,3)
x = F.interpolate(x.unsqueeze(0), size=(1, x.size(2), x.size(3)), mode="trilinear").squeeze(0)
x = x.permute(1,0,2,3)

x.size() # (1,3,4,4)

我应该排列张量来调整批次维度的大小吗？或者以某种方式对其进行迭代

Answer 1

这似乎让我可以调整批次维度的大小：

x = x.permute(1,0,2,3)
x = F.interpolate(x.unsqueeze(0), size=(1, x.size(2), x.size(3)), mode="trilinear").squeeze(0)
x = x.permute(1,0,2,3)