[Algorithm] Support Rotated FCOS (ICCV'2019)#223
Conversation
configs/rotated_fcos/README.md
Outdated
|
|
||
| | Backbone | mAP | Angle | Separate Angle | Tricks | lr schd | Mem (GB) | Inf Time (fps) | Aug | Batch Size | Configs | Download | | ||
| |:------------------------:|:-----:|:-----:|:--------------:|:------:|:-------:|:--------:|:--------------:|:---:|:----------:|:-----------------------------------------------------------------------------------------------------:|:--------------------------:| | ||
| | ResNet50 (1024,1024,200) | 70.70 | le90 | Y | Y | 1x | - | - | - | 2 | [rotated_fcos_sep_angle_r50_fpn_1x_dota_le90](./rotated_fcos_sep_angle_r50_fpn_1x_dota_le90.py) | [model](-) | [log](-) | |
There was a problem hiding this comment.
Mem (GB) is recorded in log.json file.
Inf Time (fps) can be test by:
python -m torch.distributed.launch --nproc_per_node=1 --master_port=29500 tools/analysis_tools/benchmark.py YOUR_CFG YOUR_MODEL --launcher pytorch
|
Waiting for |
Codecov Report
@@ Coverage Diff @@
## dev #223 +/- ##
==========================================
- Coverage 31.94% 30.86% -1.08%
==========================================
Files 113 119 +6
Lines 7394 7934 +540
Branches 1135 1220 +85
==========================================
+ Hits 2362 2449 +87
- Misses 4931 5384 +453
Partials 101 101
Flags with carried forward coverage won't be shown. Click here to find out more.
Continue to review full report at Codecov.
|
This reverts commit dca9332.
|
This branch has conflicts that must be resolved. Conflicting files See details below. |
yangxue0827
left a comment
There was a problem hiding this comment.
LGTM
Waiting for models.
|
Need to add metafile.yml, please refer to |
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
#119
#114
https://github.com/jbwang1997/OBBDetection/blob/master/mmdet/models/dense_heads/fcos_head.py
Modification
Note: The result of Rotated IoU Loss is not stable, its still under tuning
BC-breaking (Optional)
Does the modification introduce changes that break the back-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