microsoft / Swin-Transformer

This is an official implementation for "Swin Transformer: Hierarchical Vision Transformer using Shifted Windows".
https://arxiv.org/abs/2103.14030
MIT License
13.72k stars 2.04k forks source link

coco data [if isinstance(segms[i]['counts'], bytes):IndexError: list index out of range ] #81

Closed huangshouyou closed 2 years ago

huangshouyou commented 3 years ago

File "/home/atm/Object-Swins/mmdet/core/evaluation/eval_hooks.py", line 279, in after_train_epoch key_score = self.evaluate(runner, results) File "/home/atm/Object-Swins/mmdet/core/evaluation/eval_hooks.py", line 176, in evaluate eval_res = self.dataloader.dataset.evaluate( File "/home/atm/Object-Swins/mmdet/datasets/coco.py", line 419, in evaluate result_files, tmp_dir = self.format_results(results, jsonfile_prefix) File "/home/atm/Object-Swins/mmdet/datasets/coco.py", line 364, in format_results result_files = self.results2json(results, jsonfile_prefix) File "/home/atm/Object-Swins/mmdet/datasets/coco.py", line 301, in results2json json_results = self._segm2json(results) File "/home/atm/Object-Swins/mmdet/datasets/coco.py", line 269, in _segm2json if isinstance(segms[i]['counts'], bytes): IndexError: list index out of range

huangshouyou commented 3 years ago

and the same coco datas are applied to mask-rcnn(mmdetection) without problems.

ancientmooner commented 2 years ago

please go to https://github.com/SwinTransformer/Swin-Transformer-Object-Detection for an example