mmdetection
mmdetection copied to clipboard
[Fix] Fix mmdet flops error
Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand some items, don't worry, just make the pull request and seek help from maintainers.
Motivation
Now, mmdet don't support compute flops for all configs. I add some functions and modify many configs to fix mmdet flops error.
Modification
- I mainly modify some configs and some metafile
- I add forward function to compute flops in cascade_roi_head, scnet_roi_head, htc_roi_head
- I add a function parameter in most of dense heads
BC-breaking (Optional)
Does the modification introduce changes that break the backward-compatibility of the downstream repos? If so, please describe how it breaks the compatibility and how the downstream projects should modify their code to keep compatibility with this PR.
Use cases (Optional)
If this PR introduces a new feature, it is better to list some use cases here, and update the documentation.
Checklist
- Pre-commit or other linting tools are used to fix the potential lint issues.
- The modification is covered by complete unit tests. If not, please add more unit test to ensure the correctness.
- If the modification has potential influence on downstream projects, this PR should be tested with downstream projects, like MMDet or MMCls.
- The documentation has been modified accordingly, like docstring or example tutorials.
@zwhus I have no doubts about the code modification of this pr. The core question is how to calculate flops in two-stage is the most reasonable? I'm not very sure.
@zwhus I have no doubts about the code modification of this pr. The core question is how to calculate flops in two-stage is the most reasonable? I'm not very sure.
We can do it in the later PR. In this PR, we mainly fix the issue of metafile.
Dear zwhus, First of all, we want to express our gratitude for your significant PR in the MMDet project. Your contribution is highly appreciated, and we are grateful for your efforts in helping improve this open-source project during your personal time. We believe that many developers will benefit from your PR. As a valued contributor, we would also like to invite you to join our Special Interest Group (SIG) private channel, where you can share your experiences, ideas, and build connections with like-minded peers. Join us :https://discord.gg/raweFPmdzG To join the SIG channel, simply message moderator— OpenMMLab on Discord or briefly share your open-source contributions in the #introductions channel and we will assist you. We look forward to seeing you there! Also,if you have WeChat,welcome to join our community on WeChat. You can add our assistant :openmmlabwx,or scan the following QR code to join us. Please add "mmsig + Github ID" as a remark when adding friends:) Best regards! @zwhus