azhangmn
azhangmn
When annotating an image, labelme will sometimes crash if I delete a point on a polygon/linestrip/etc. It doesn't always happen but happens enough that it is noticeable.
``` def loss(self, _, y_pred): dy = torch.abs(y_pred[:, :, 1:, :, :] - y_pred[:, :, :-1, :, :]) dx = torch.abs(y_pred[:, :, :, 1:, :] - y_pred[:, :, :, :-1,...