Tao

Results 14 comments of Tao

I also met this issue. I appreciate any comments on this issue.

> Hi, > Just a reminder, the paper reports performance of 69.8 mIOU with COCO+ImageNet pre-training. We report a performance of 65.4 in the paper when the network is pretrained...

The bug still exists, I got the same error during validation.

You were trying to train the pre-trained model on your custom dataset. Due to the difference in the number of categories, there would be a size mismatch in the model...