10a2385d01
Support exporting to ONNX for Faster R-CNN ( #20 )
...
* update rewriter for rpn head
* new nms implement
* update bbox_head to support fasterrcnn with tensorrt
* rewrite SingleRoIExtractor
* remove unnecessary import
* fix fcos and bbox_nms
* resolve comments
* resolve comments
Co-authored-by: grimoire <yaoqian@sensetime.com>
2021-07-21 19:46:23 +08:00
b9e64f9e1c
Add trt instance norm plugin ( #16 )
...
* add trt instance norm plugin
* last line empty
* fix clang format
* fix grid_sample clang format
* remove redundant
* fix lint
* refine codes
* fix clang format
* clang format
* clang format
* clang format
2021-07-16 16:05:09 +08:00
1718501b4f
[Feature] Support fsaf fcos ( #14 )
...
* add fcos and fsaf
* remove is_in_onnx_export
* fix lint
* fix lint
* remove forwards
2021-07-13 20:13:39 +08:00
c4b7dad2ec
add visualization for mmdet ( #13 )
...
* add visualization for mmdet
* resolve comments
* update code, enable visualize on nvidia driver>11
Co-authored-by: grimoire <yaoqian@sensetime.com>
2021-07-13 17:21:02 +08:00
0eca9ebcbf
add docker ( #15 )
...
Need tests...
2021-07-13 08:45:42 +00:00
7227de907d
use new rewriter ( #12 )
2021-07-12 16:31:37 +08:00
9f0e2d4339
changed codebase structure ( #10 )
...
* add delta2bbox rewriter
* rename onnx2trt
* add rewriter: anchor_generator_single_level_grid_priors
* Revert "add rewriter: anchor_generator_single_level_grid_priors"
This reverts commit ac7cf272942c4787bf143c0d67e414b0d2603b75.
* update comments
* remove clamp rewriter
* remove unused func
* reconstruct codes
* remomve mark
* create export package for codebases
* remove grid_sampler symbolic
* reorganize nms parts
* fix conflicts
* update unittest
2021-07-12 16:26:44 +08:00
3fa94f4424
[Feature] better tensorrt cpp code ( #11 )
...
* better tensorrt cpp code
* fix end of file
2021-07-12 10:01:52 +08:00
342e195a1b
Update README.md
2021-07-10 16:16:11 +08:00
8408ae2bcf
[Feature] Multi level roi align ( #9 )
...
* add multi level roi align
* better trt plugin code
* fix lint
* optimizer trt_multi_level_roi_align.cpp
* fix lint
2021-07-09 13:56:50 +08:00
dae6a8ccf9
Rewrite delta2bbox ( #8 )
...
* add delta2bbox rewriter
* rename onnx2trt
* add rewriter: anchor_generator_single_level_grid_priors
* Revert "add rewriter: anchor_generator_single_level_grid_priors"
This reverts commit ac7cf272942c4787bf143c0d67e414b0d2603b75.
* update comments
* remove clamp rewriter
* remove unused func
2021-07-07 19:11:54 +08:00
66300c0c74
Supports setting input/output names explicitly via mark decorator ( #4 )
...
* setting names of inputs and outputs via mark decorator
* fix bug in mark, move extract_model to mmdeploy/apis and add test cases
* remove fsaf rewrite, which is not needed any more after recent changes
* replace print with log and add fixture to clear test tmp file
2021-07-07 14:15:26 +08:00
66a099faf9
[Feature] Apis unit test ( #7 )
...
* add apis test
* split torch2onnx impl, prepare for codebase test
* add is_available to backend
* lint
2021-07-05 12:51:43 +08:00
dff06eeacc
fix lint
2021-07-01 18:50:16 +08:00
52fd08febd
[Feature] much better retinanet ( #6 )
...
* better retinanet support
* prepare split export tensorrt
* optimizer cfg
* free anchor when static shape
* fix docstring
* use function rewriter instead of module rewriter on retinanet
* fix bug of mmcls tensorrt config
* add single stage mark, static shape support
2021-07-01 17:32:33 +08:00
27880afdcd
[Feature] better retinanet support ( #5 )
...
* better retinanet support
* prepare split export tensorrt
* optimizer cfg
* free anchor when static shape
* fix docstring
2021-07-01 11:42:07 +08:00
8fd4ef4d47
remove unused flag, fix lint
2021-06-29 10:23:52 +08:00
e08dbf26b3
add CI
2021-06-29 10:10:15 +08:00
ef41f69553
add model splitting support ( #1 )
...
* add function marker and model extractor
* add fsaf split & partial mask rcnn split, import extract.py
* 1. add value renaming 2. add apply_marks in config to turn on/off marks
* rewind changes on pytorch2onnx.py
Co-authored-by: q.yao <streetyao@live.com>
2021-06-28 20:35:15 +08:00
5998d24766
[Feature] Add TensorRT batched NMS support ( #3 )
...
* add trt batched_nms plugin
* update trt batched nms
2021-06-25 19:31:16 +08:00
6c47ee3d2a
add tensorrt support ( #2 )
2021-06-23 13:14:28 +08:00
6eb2e89016
update cmakelists
2021-06-21 10:34:48 +08:00
2bd96243e2
better mmdet support
2021-06-18 15:16:21 +08:00
2317ee659a
support retinanet pytorch2onnx
2021-06-17 17:37:08 +08:00
b777d27bcf
add return value in deploy.py
2021-06-17 17:26:32 +08:00
34caea27c1
use function rewriter convert classifier
2021-06-17 16:04:45 +08:00
3bf3b9cdda
fix lint
2021-06-17 15:29:12 +08:00
12792727e7
add configs, add convert pipeline
2021-06-17 15:28:23 +08:00
5b2459570a
better register
2021-06-16 19:25:09 +08:00
100ba694ce
add init
2021-06-16 17:56:17 +08:00
93408c7cd8
add type annotation
...
add onnxruntime plugin support, add more custom ops
update function_rewriter, enable use origin function
add setup.cfg
update setup.cfg
2021-06-16 15:36:58 +08:00
1e47821e49
add backend plugin build system
2021-06-15 20:44:45 +08:00
f90ebf8c2c
add register
2021-06-11 13:26:05 +08:00
b6d8d48d7c
Initial commit
2021-06-11 13:24:18 +08:00
RunningLeon
AllentDan
q.yao
lzhangzz
grimoire
maningsheng