TypeError: forward() got an unexpected keyword argument 'ignore_index'

[JinxiaoWang]

I add a new loss 'my_loss' to mmseg/models/losses/ as the tutorial doc did.

my_loss.py :

`import torch
import torch.nn as nn

from ..builder import LOSSES
from .utils import weighted_loss

@weighted_loss
def my_loss(pred, target):
assert pred.size() == target.size() and target.numel() > 0
loss = torch.abs(pred - target)
return loss

@LOSSES.register_module
class MyLoss(nn.Module):

def __init__(self, reduction='mean', loss_weight=1.0):
    super(MyLoss, self).__init__()
    self.reduction = reduction
    self.loss_weight = loss_weight

def forward(self,
            pred,
            target,
            weight=None,
            avg_factor=None,
            reduction_override=None):
    assert reduction_override in (None, 'none', 'mean', 'sum')
    reduction = (
        reduction_override if reduction_override else self.reduction)
    loss = self.loss_weight * my_loss(
        pred, target, weight, reduction=reduction, avg_factor=avg_factor)
    return loss`

Then , when I use this loss to train the model, this TypeError appeared.

[xvjiarui]

Hi @DreamerWangDr
Try this

def my_loss(pred, target, **kwargs):

We will update tutorial to fix this.

[JinxiaoWang]

Hi @DreamerWangDr
Try this

def my_loss(pred, target, **kwargs):

We will update tutorial to fix this.

Thanks， I tried this, still TypeError: forward() got an unexpected keyword argument 'ignore_index'.
Hope that Dice Loss and Focal Loss can be added to the losses module.

[xvjiarui]

Hi @DreamerWangDr
Sorry. My mistake. You may try this

def forward(self,
            pred,
            target,
            weight=None,
            avg_factor=None,
            reduction_override=None,
            **kwargs):

[JinxiaoWang]

Hi @DreamerWangDr
Sorry. My mistake. You may try this

def forward(self,
            pred,
            target,
            weight=None,
            avg_factor=None,
            reduction_override=None,
            **kwargs):

Hi, Thanks. I tried this, TypeError: forward() got an unexpected keyword argument 'ignore_index' disppeard.
Then , I meet an assert error at line : assert pred.size() == target.size() and target.numel() > 0
I don't know is pred.size different from target.size or target.numel() <=0 ?