用一类coco格式数据集训练swim-transformer-objetc-detection时报错

Question

raceback (most recent call last):
  File "tools/train.py", line 187, in <module>
    main()
  File "tools/train.py", line 183, in main
    meta=meta)
  File "/home/zzk/project/Swin-Transformer-Object-Detection-fire/mmdet/apis/trai                                                  n.py", line 185, in train_detector
    runner.run(data_loaders, cfg.workflow)
  File "/opt/conda/lib/python3.7/site-packages/mmcv/runner/epoch_based_runner.py                                                  ", line 125, in run
    epoch_runner(data_loaders[i], **kwargs)
  File "/opt/conda/lib/python3.7/site-packages/mmcv/runner/epoch_based_runner.py                                                  ", line 54, in train
    self.call_hook('after_train_epoch')
  File "/opt/conda/lib/python3.7/site-packages/mmcv/runner/base_runner.py", line                                                   307, in call_hook
    getattr(hook, fn_name)(self)
  File "/home/zzk/project/Swin-Transformer-Object-Detection-fire/mmdet/core/eval                                                  uation/eval_hooks.py", line 147, in after_train_epoch
    key_score = self.evaluate(runner, results)
  File "/home/zzk/project/Swin-Transformer-Object-Detection-fire/mmdet/core/eval                                                  uation/eval_hooks.py", line 177, in evaluate
    results, logger=runner.logger, **self.eval_kwargs)
  File "/home/zzk/project/Swin-Transformer-Object-Detection-fire/mmdet/datasets/                                                  coco.py", line 404, in evaluate
    result_files, tmp_dir = self.format_results(results, jsonfile_prefix)
  File "/home/zzk/project/Swin-Transformer-Object-Detection-fire/mmdet/datasets/                                                  coco.py", line 349, in format_results
    result_files = self.results2json(results, jsonfile_prefix)
  File "/home/zzk/project/Swin-Transformer-Object-Detection-fire/mmdet/datasets/                                                  coco.py", line 286, in results2json
    json_results = self._segm2json(results)
  File "/home/zzk/project/Swin-Transformer-Object-Detection-fire/mmdet/datasets/                                                  coco.py", line 254, in _segm2json
    if isinstance(segms[i]['counts'], bytes):
IndexError: list index out of range

Answer 1

这个github是回答https://github.com/SwinTransformer/Swin-Transformer-Object-Detection/issues/29

https://app.codecov.io/gh/open-mmlab/mmdetection/compare/5065/diff?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr%20comments&utm_term=open-mmlab#diff-bW1kZXQvbW9kZWxzL2RldGVjdG9ycy9iYXNlLnB5

按照这个上面说的把mmdet/model/roi_heads/test_mixins.py文件进行相应位置的修改