RuntimeError: multi-target not supported at /pytorch/aten/src/THCUNN/generic/ClassNLLCriterion.cu:18

1

my input tensor size [[10,3,256,256]) but my output tensor size ([10,1000]) is like this. I chose RGB images and my batch size 10. So my lost function doesn’t work. how can i fix this. i use ResNet-152

["" def conv3x3(in_planes, out_planes, stride=1, groups=1, dilation=1):
“”“3x3 convolution with padding”""
return nn.Conv2d(in_planes, out_planes, kernel_size=3, stride=stride,
padding=dilation, groups=groups, bias=False, dilation=dilation)

def conv1x1(in_planes, out_planes, stride=1):
“”“1x1 convolution”""
return nn.Conv2d(in_planes, out_planes, kernel_size=1, stride=stride, bias=False)

class BasicBlock(nn.Module):
expansion = 1

def __init__(self, inplanes, planes, stride=1, downsample=None, groups=1,
             base_width=64, dilation=1, norm_layer=None):
    super(BasicBlock, self).__init__()
    if norm_layer is None:
        norm_layer = nn.BatchNorm2d
    if groups != 1 or base_width != 64:
        raise ValueError('BasicBlock only supports groups=1 and base_width=64')
    if dilation > 1:
        raise NotImplementedError("Dilation > 1 not supported in BasicBlock")
    # Both self.conv1 and self.downsample layers downsample the input when stride != 1
    self.conv1 = conv3x3(inplanes, planes, stride)
    self.bn1 = norm_layer(planes)
    self.relu = nn.ReLU(inplace=True)
    self.conv2 = conv3x3(planes, planes)
    self.bn2 = norm_layer(planes)
    self.downsample = downsample
    self.stride = stride

def forward(self, x):
    identity = x

    out = self.conv1(x)
    out = self.bn1(out)
    out = self.relu(out)

    out = self.conv2(out)
    out = self.bn2(out)

    if self.downsample is not None:
        identity = self.downsample(x)

    out += identity
    out = self.relu(out)

    return out

class Bottleneck(nn.Module):
# Bottleneck in torchvision places the stride for downsampling at 3x3 convolution(self.conv2)
# while original implementation places the stride at the first 1x1 convolution(self.conv1)
# according to "Deep residual learning for image recognition"https://arxiv.org/abs/1512.03385.
# This variant is also known as ResNet V1.5 and improves accuracy according to
# https://ngc.nvidia.com/catalog/model-scripts/nvidia:resnet_50_v1_5_for_pytorch.

expansion = 4

def __init__(self, inplanes, planes, stride=1, downsample=None, groups=1,
             base_width=64, dilation=1, norm_layer=None):
    super(Bottleneck, self).__init__()
    if norm_layer is None:
        norm_layer = nn.BatchNorm2d
    width = int(planes * (base_width / 64.)) * groups
    # Both self.conv2 and self.downsample layers downsample the input when stride != 1
    self.conv1 = conv1x1(inplanes, width)
    self.bn1 = norm_layer(width)
    self.conv2 = conv3x3(width, width, stride, groups, dilation)
    self.bn2 = norm_layer(width)
    self.conv3 = conv1x1(width, planes * self.expansion)
    self.bn3 = norm_layer(planes * self.expansion)
    self.relu = nn.ReLU(inplace=True)
    self.downsample = downsample
    self.stride = stride

def forward(self, x):
    identity = x

    out = self.conv1(x)
    out = self.bn1(out)
    out = self.relu(out)

    out = self.conv2(out)
    out = self.bn2(out)
    out = self.relu(out)

    out = self.conv3(out)
    out = self.bn3(out)

    if self.downsample is not None:
        identity = self.downsample(x)

    out += identity
    out = self.relu(out)

    return out

class ResNet(nn.Module):

def __init__(self, block, layers, num_classes=1000, zero_init_residual=False,
             groups=1, width_per_group=64, replace_stride_with_dilation=None,
             norm_layer=None):
    super(ResNet, self).__init__()
    if norm_layer is None:
        norm_layer = nn.BatchNorm2d
    self._norm_layer = norm_layer

    self.inplanes = 64
    self.dilation = 1
    if replace_stride_with_dilation is None:
        # each element in the tuple indicates if we should replace
        # the 2x2 stride with a dilated convolution instead
        replace_stride_with_dilation = [False, False, False]
    if len(replace_stride_with_dilation) != 3:
        raise ValueError("replace_stride_with_dilation should be None "
                         "or a 3-element tuple, got {}".format(replace_stride_with_dilation))
    self.groups = groups
    self.base_width = width_per_group
    self.conv1 = nn.Conv2d(3, self.inplanes, kernel_size=7, stride=2, padding=3,
                           bias=False)
    self.bn1 = norm_layer(self.inplanes)
    self.relu = nn.ReLU(inplace=True)
    self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1)
    self.layer1 = self._make_layer(block, 64, layers[0])
    self.layer2 = self._make_layer(block, 128, layers[1], stride=2,
                                   dilate=replace_stride_with_dilation[0])
    self.layer3 = self._make_layer(block, 256, layers[2], stride=2,
                                   dilate=replace_stride_with_dilation[1])
    self.layer4 = self._make_layer(block, 512, layers[3], stride=2,
                                   dilate=replace_stride_with_dilation[2])
    self.avgpool = nn.AdaptiveAvgPool2d((1, 1))
    self.fc = nn.Linear(512 * block.expansion, num_classes)

    for m in self.modules():
        if isinstance(m, nn.Conv2d):
            nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu')
        elif isinstance(m, (nn.BatchNorm2d, nn.GroupNorm)):
            nn.init.constant_(m.weight, 1)
            nn.init.constant_(m.bias, 0)

    # Zero-initialize the last BN in each residual branch,
    # so that the residual branch starts with zeros, and each residual block behaves like an identity.
    # This improves the model by 0.2~0.3% according to https://arxiv.org/abs/1706.02677
    if zero_init_residual:
        for m in self.modules():
            if isinstance(m, Bottleneck):
                nn.init.constant_(m.bn3.weight, 0)
            elif isinstance(m, BasicBlock):
                nn.init.constant_(m.bn2.weight, 0)

def _make_layer(self, block, planes, blocks, stride=1, dilate=False):
    norm_layer = self._norm_layer
    downsample = None
    previous_dilation = self.dilation
    if dilate:
        self.dilation *= stride
        stride = 1
    if stride != 1 or self.inplanes != planes * block.expansion:
        downsample = nn.Sequential(
            conv1x1(self.inplanes, planes * block.expansion, stride),
            norm_layer(planes * block.expansion),
        )

    layers = []
    layers.append(block(self.inplanes, planes, stride, downsample, self.groups,
                        self.base_width, previous_dilation, norm_layer))
    self.inplanes = planes * block.expansion
    for _ in range(1, blocks):
        layers.append(block(self.inplanes, planes, groups=self.groups,
                            base_width=self.base_width, dilation=self.dilation,
                            norm_layer=norm_layer))

    return nn.Sequential(*layers)

def _forward_impl(self, x):
    # See note [TorchScript super()]
    x = self.conv1(x)
    x = self.bn1(x)
    x = self.relu(x)
    x = self.maxpool(x)

    x = self.layer1(x)
    x = self.layer2(x)
    x = self.layer3(x)
    x = self.layer4(x)

    x = self.avgpool(x)
    x = torch.flatten(x, 1)
    x = self.fc(x)

    return x

def forward(self, x):
    return self._forward_impl(x) "]
2

The output shape of [10, 1000] looks valid for a mutli-class classification use case with 1000 classes.
I guess the target shape might be wrong.
nn.CrossEntropyLoss expects the target to have the shape [batch_size=10] containing the class indices in the range [0, nb_classes-1 = 999].
If your target is one-hot encoded, use target = torch.argmax(target, 1) to create the expected target with the class indices.

1 Like
3

I solved the size problem. but I have 6 classes in total. hence, he expects target efficiency to be at [10,6,256,256] dimensions. The output data is [10,2048,256,256] due to the resnet. How can I fix the output data? I want it to be [10,6,256,256].

4

Are you working on a multi-class segmentation use case?
If so, then the target should have the shape [10, 256, 256] and contain values in the range [0, 5].
How did you end up with an output of [10,2048,256,256]?
It seems that the spacial size wasn’t reduced at all, which should be the case for a resnet.

5

Yes, I am working on semantic segmentation on the RGB image.

Since Batch_size = 10 and class_num = 6, my target size is [10,6,256,256].

Yes, I use Resnet architecture, so the output size is [10,2048,256,256].

6

If you are using nn.CrossEntropyLoss or nn.NLLLoss, the target shape is wrong, as explained before.

ResNet would return an output of [batch_size, 1000], so you would have to manipulate it in some way to get this output. What were these modifications?

7

I changed self.avgpool = nn.AdaptiveAvgPool2d ((1, 1)) and self.fc = nn.Linear (512 * block.expansion, num_classes in the last layer and deleted the line torch.flatten (x, 1). I used nn.Softmax2d () and self.avgpool = nn.AdaptiveAvgPool2d ((256, 256)).

8

sorry. class_num = 7

9

In that case you would have to reduce the output channels from 2048 to the number of classes, e.g. by using a conv layer with out_channels=nb_classes. The kernel size might be 1x1, but it depends on your use case, if that’s the best approach.

The target would still be wrong:

the target should have the shape [10, 256, 256] and contain values in the range [0, nb_classes-1]

10

sorry but i don’t think i fully understand.

11

Thanks for the concise reply. This worked for me!