python - RuntimeError : Given groups=1, 大小 [64, 3, 3, 3] 的权重，预期输入 [4, 5000, 5000, 3] 有 3 个 channel ，但有 5000 个 channel

Question

所以，我有一个 U-Net 模型，我将 5000x5000x3 的图像输入到模型中，我得到了上面的错误。

这是我的模型。

import torch
import torch.nn as nn


def double_conv(in_channels, out_channels):
    return nn.Sequential(
        nn.Conv2d(in_channels, out_channels, 3, padding=1),
        nn.ReLU(inplace=True),
        nn.Conv2d(out_channels, out_channels, 3, padding=1),
        nn.ReLU(inplace=True)
    )


class UNeT(nn.Module):
    def __init__(self, n_class):
        super().__init__()
        self.dconv_down1 = double_conv(3, 64)
        self.dconv_down2 = double_conv(64, 128)
        self.dconv_down3 = double_conv(128, 256)
        self.dconv_down4 = double_conv(256, 512)
        self.maxpool = nn.MaxPool2d(2)
        self.upsample = nn.Upsample(scale_factor=2, mode='bilinear',
                                    align_corners=True)
        self.dconv_up3 = double_conv(256 + 512, 256)
        self.dconv_up2 = double_conv(128 + 256, 128)
        self.dconv_up1 = double_conv(128 + 64, 64)
        self.conv_last = nn.Conv2d(64, n_class, 1)

    def forward(self, x):
        conv1 = self.dconv_down1(x)
        x = self.maxpool(conv1)
        conv2 = self.dconv_down2(x)
        x = self.maxpool(conv2)
        conv3 = self.dconv_down3(x)
        x = self.maxpool(conv3)
        x = self.dconv_down4(x)
        x = self.upsample(x)
        x = torch.cat([x, conv3], dim=1)
        x = self.dconv_up3(x)
        x = self.upsample(x)
        x = torch.cat([x, conv2], dim=1)
        x = self.dconv_up2(x)
        x = self.upsample(x)
        x = torch.cat([x, conv1], dim=1)
        x = self.dconv_up1(x)
        out = self.conv_last(x)
        return out

我尝试做 model(inputs.unsqueeze_(0)) 但我得到了一个不同的错误。

Answer 1

pytorch 中的维度顺序与您期望的不同。您的输入张量具有 4x5000x5000x3 的 shape，您将其解释为一批大小 4，图像为 5000x5000 像素，每个像素有3个 channel 。也就是说，您的尺寸是 batch-height-width-channel。

但是，pytorch 期望张量维度的顺序不同:batch-channel-height-width。也就是说，channel 维度应该在宽度和高度空间维度之前。

您需要 permute输入张量的尺寸以解决您的问题:

model(inputs.permute(0, 3, 1, 2))

有关详细信息，请参阅 nn.Conv2d 的文档。 .