Tramac
Tramac
You can put the code [here](https://github.com/Tramac/awesome-semantic-segmentation-pytorch/blob/c77d34929b43dcc0da13e2452fdc803ddc0b4334/core/utils/score.py#L47).
Please check the input of this [line](https://github.com/Tramac/awesome-semantic-segmentation-pytorch/blob/5843f75215dadc5d734155a238b425a753a665d9/core/data/dataloader/segbase.py#L66).
Hi, maybe the hyperparameters are not optimal or due to the reason of the dataset.
It is best to modify on `cityscapes.py`. Do you have WeChat？I have not had enough time to follow github recently, maybe solve the problem by WeChat.
Maybe there is something wrong with the label.
> Hi,if i want to use my own dataset,which dataloader.py should i use,please tell me when you saw the question. It depends on your dataset directory structure.
You can refer to the script cityscapes.py, it is similar to your file structure.
Please check your dataloader.
Please change shape [4, 480, 480, 3] to [4, 3, 480, 480]. Just like: `image = image.transpose(0, 3, 1, 2)`.
Please post a more detailed log.