mask prediction does not not work when keep_ratio=False in data

open-mmlab / mmdetection

OpenMMLab Detection Toolbox and Benchmark

https://mmdetection.readthedocs.io

Apache License 2.0

29.43k stars 9.43k forks source link

Closed PawelFaron closed 4 years ago

PawelFaron commented 4 years ago

I'm training my own data in COCO like format using https://github.com/open-mmlab/mmdetection/blob/master/configs/res2net/htc_r2_101_fpn_20e_coco.py It words good when I have keep_ratio=False set in train_pipeline, test_pipeline, val, test dict but I got all zeros for "segm" when I have this set for data.

ZwwWayne commented 4 years ago

Hi @PawelFaron , Could you provide a reproducible config with the command you use?