Question

我正在为pytorch框架上的多类分割任务训练一个基于unet的模型。使用以下损失函数优化模型，

class MulticlassJaccardLoss(_Loss):
"""Implementation of Jaccard loss for multiclass (semantic) image segmentation task
"""
__name__ = 'mc_jaccard_loss'
def __init__(self, classes: List[int] = None, from_logits=True, weight=None, reduction='elementwise_mean'):
    super(MulticlassJaccardLoss, self).__init__(reduction=reduction)
    self.classes = classes
    self.from_logits = from_logits
    self.weight = weight

def forward(self, y_pred: Tensor, y_true: Tensor) -> Tensor:
    """
    :param y_pred: NxCxHxW
    :param y_true: NxHxW
    :return: scalar
    """
    if self.from_logits:
        y_pred = y_pred.softmax(dim=1)

    n_classes = y_pred.size(1)
    smooth = 1e-3

    if self.classes is None:
        classes = range(n_classes)
    else:
        classes = self.classes
        n_classes = len(classes)

    loss = torch.zeros(n_classes, dtype=torch.float, device=y_pred.device)

    if self.weight is None:
        weights = [1] * n_classes
    else:
        weights = self.weight

    for class_index, weight in zip(classes, weights):

        jaccard_target = (y_true == class_index).float()
        jaccard_output = y_pred[:, class_index, ...]

        num_preds = jaccard_target.long().sum()

        if num_preds == 0:
            loss[class_index-1] = 0 #custom
        else:
            iou = soft_jaccard_score(jaccard_output, jaccard_target, from_logits=False, smooth=smooth)
            loss[class_index-1] = (1.0 - iou) * weight #custom

    if self.reduction == 'elementwise_mean':
        return loss.mean()

    if self.reduction == 'sum':
        return loss.sum()

    return loss

我仅计算两个类的损失（第1类和第2类，而不是背景类）。

MulticlassJaccardLoss(weight=[0.5,10], classes=[1,2], from_logits=False)

当我训练模型时，它会训练前几次迭代，并且出现以下错误，

element 0 of tensors does not require grad and does not have a grad_fn

代码中的错误是什么？

谢谢！

Answer 1

尝试设置：

torch.zeros(..., requires_grad=True)

我相信torch.zeros的默认设置是require_grad = False，所以这可能会有所帮助。

pytorch中的多类加权损失函数

1 个答案: