RuntimeError: weight tensor should be defined either for all or no classes

今天在添加网络中添加深监督的时候遇到了这个错误，先看出现错误的位置，是在计算交叉熵的时候：

    def forward(self, inputs_scales, targets_scales):losses = []for inputs, targets in zip(inputs_scales, targets_scales):# mask = targets > 0targets_m = targets.clone() #深拷贝targets_m -= 1 #减去loss_all = self.ce_loss(inputs, targets_m.long())

在网上冲浪一个小时，发现了各路大神解决方案：参考方案

1：检查输出预测图与标签的尺寸是否一样

2：没有区分背景及边缘：

3：分割类别与分割标签种类不一样：

我想在深监督中，使用多尺度标签进行监督，在decoder中输出经过self.side_output,得到了一个通道为37的语义分割图。

self.side_output = nn.Conv2d(in_channels//2,out_channels=37,kernel_size=1)

在图像的预处理中，我们对标签进行缩放，使标签的大小和decoder输出的大小一样，这样才可以进行损失计算。

#
class MultiScaleLabel:def __init__(self, downsampling_rates=None):if downsampling_rates is None:self.downsampling_rates = [8, 4, 2]else:self.downsampling_rates = downsampling_ratesdef __call__(self, sample):label = sample['label']h, w = label.shapesample['label_down'] = dict()# Nearest neighbor interpolationfor rate in self.downsampling_rates:label_down = cv2.resize(label.numpy(), (w // rate, h // rate),interpolation=cv2.INTER_NEAREST)sample['label_down'][rate] = torch.from_numpy(label_down)return sample

而在NYUDepth-v2中，标签的类别为40，我的分割类别只有37，因此问题就出现了。即上面的解决办法3。

所以我们将输出的分割类别调整37为标签类别40就可以了。

self.side_output = nn.Conv2d(in_channels//2,out_channels=40,kernel_size=1)

RuntimeError: weight tensor should be defined either for all or no classes相关推荐

torchserve 错误：RuntimeError: Expected tensor for argument #1 ‘input’ to have the same device as tenso
RuntimeError: Expected tensor for argument #1 'input' to have the same device as tensor for argument ...
RuntimeError: Expected tensor for argument #1 ‘indices‘ to have scalar type Long； but got CUDAType
RuntimeError: Expected tensor for argument #1 'indices' to have scalar type Long; but got torch.cuda ...
RuntimeError: Expected tensor for argument #1 ‘indices‘ to have scalar type Long； but got CUDAFloatT
使用torchsummary 中的 summary函数时,模型中有Embedding层的时候,会报错.原因是Embedding 需要整型的Tensor. pip install 的torchsumma ...
Expected tensor for argument #1 'input' to have the same device as tensor for argument #2 'weight';
解决了题目所述问题. 报错完整信息为: RuntimeError: Expected tensor for argument #1 'input' to have the same device as ...
Tensor unsqueeze 以 broadcast
如果对 torch.unsqueeze 了解不多,建议先阅读博主的这篇文章. 1. Tensor 和 Scalar import torchX = torch.tensor([[1,2,3],[4,5 ...
PytorchRuntimeError: inconsistent tensor sizes at /pytorch/torch/lib/TH/generic/THTensorMath.c:2709
1.pytorch在得到dataset时没有问题,得到DataLoader时也没有问题,以batch size = 512遍历数据时,出现了: RuntimeError: inconsistent t ...
Expected tensor for argument #1 ‘indices‘ to have one of the following scalar types: Long, Int； but
RuntimeError: Expected tensor for argument #1 'indices' to have one of the following scalar types: L ...
pytorch 定义torch类型数据_PyTorch官方中文文档：torch.Tensor
torch.Tensor torch.Tensor是一种包含单一数据类型元素的多维矩阵. Torch定义了七种CPU tensor类型和八种GPU tensor类型: Data tyoe CPU te ...
3、TensorFlow 的数据模型-----张量(Tensor)
一.Tensor 类简介 Tensor 定义 A Tensor is a symbolic handle to one of the outputs of an Operation. It does ...
pytorch学习五、深度学习计算
来自于 https://tangshusen.me/Dive-into-DL-PyTorch/#/ 官方文档 https://pytorch.org/docs/stable/tensors.html ...

RuntimeError: weight tensor should be defined either for all or no classes

RuntimeError: weight tensor should be defined either for all or no classes相关推荐

最新文章

热门文章