What is the true label of -1 in seg_label

[Iranb]

As the TrainDataset output in dataset.py , I got seg_label as [[ 0, 9, 9, ..., 9, 9, -1], [ 9, 9, 9, ..., 9, 9, -1], [ 9, 9, 0, ..., 9, 9, -1], ..., [ 0, 0, 0, ..., 7, 7, -1], [ 0, 0, 0, ..., 7, 7, -1], [ 0, 0, 0, ..., 7, 7, -1]],

    [[ 1,  1,  1,  ...,  1,  1, -1],
     [ 1,  1,  1,  ...,  1,  1, -1],
     [ 1,  1,  1,  ...,  1,  1, -1],
     ...,
     [ 7,  7,  7,  ...,  7,  7, -1],
     [ 7,  7,  7,  ...,  7,  7, -1],
     [ 7,  7,  7,  ...,  7,  7, -1]]])}]`

But the true label range in your paper 0-12 for training and 0-13 for testing. So what is the meaning of "-1" label?

[xksteven]

The -1 label if I recall correctly should be ignored. The docs are from the commit that we based off of in the semantic segmentation repo. So the labels for the model at that version were peculiar in that it didn't use the background if I recall correctly. So we just subtracted all of the labels -1.

Hope that helps.