jwwangchn / bonai Goto Github PK

Excellent work,
I want to ask a question. Which tools you are used it for image annotations? Thank you!!

How can I get the test dataset? Google driver only include train and val

can you share your evaluation codes?

When I want to run train.py in the tools directory, it prompts me that mmcv=1.0.5 is required. Does the training of this program only work in this environment?

Can you provide me with the config file used for training and testing?

I appreciate your work on this project and I have some feedback for you.
I noticed that the data and code you provided do not match with each other. The code seems to expect a annotation file named:
'BONAI/coco/bonai_shanghai_trainval.json'.
However, there is no such file in the given dataset. This makes it hard to reproduce your results and understand your methods. Could you please check your data and code files and make sure they are consistent? Thank you for your attention and cooperation.

Some errors when run:

1. from .post_processing import * # noqa: F401, F403
   AttributeError: module 'mmdet.core.post_processing' has no attribute 'multiclass_rnms'
   This may be related to mmdet version?
2. ModuleNotFoundError: No module named 'bstool'
   The bstool code is missing
3. KeyError: 'OffsetHeadExpandFeature is not in the head registry'
   After i delete codes about bstool, it reported this error