9d4d52078b
【Fix】Fix ppl bug about grid sample ( #325 )
...
* fix ppl problems
* fix roialign
* fix grid_sampler bug
* fix grid sampler
* fix config
* fix test
2021-12-23 12:11:07 +08:00
0f9f0a60e4
[Enhancement]: Update configs for OpenVINO ( #306 )
...
* Update config for OpenVINO
* Update test
* Update better test
* Update instance-seg config of OpenVINO
2021-12-20 17:56:25 +08:00
f442dbbe83
[Docs]: MMCLS benchmark ( #300 )
...
* add doc
* Update benchmark
* Update benchmark
* Restore missing words
2021-12-17 14:12:26 +08:00
cde9abd52c
【Fix】Fix mmdet head ( #290 )
...
* fix anchor head and base dense head
* fix base dense head bug
* fix base dense head bug
* fix pad
* add ssd model int8 and fp16 config
* fix a bug about basedensehead
* add config for yolov3 trt fp16 int8
Co-authored-by: maningsheng <mnsheng@yeah.net>
2021-12-17 10:47:17 +08:00
3e8237d8bb
[Feature] Support end2end mmdet2.19 retina mobilessd ( #286 )
...
* support end2end mmdet2.19 retina mobilessd
* fix yapf
* add end2end fsaf
* fix lint
* fix comments
* fix lint
* add static configs
* fix docformatter
* move ssdhead
* add rewrite for l2norm
* fix ncnn ssd
* fix isort
* rename config
* add ssd_head_ut
* fix string
* align ssd
* remove unused bbox rewriter
Co-authored-by: grimoire <yaoqian@sensetime.com>
Co-authored-by: maningsheng <mnsheng@yeah.net>
2021-12-17 10:46:54 +08:00
3eabeae071
Add new configs for ocr ( #295 )
...
* save configs
* add to benchmark
* add datasets to benchmark
* lock mmcor version to 0.3.0
2021-12-15 19:51:38 +08:00
0dea300714
[Fix] Reanme ppl to pplnn for all cases ( #273 )
...
* add speed benchmark for pplnn and replace all ppl in docs with pplnn
* replace ppl with pplnn for all codes
* docs zh
* fix crnn config path
* update docstring
* rename
* update ppl to pplnn in readme_zh
Co-authored-by: maningsheng <mnsheng@yeah.net>
2021-12-14 15:04:39 +08:00
0b23c744fc
remove trt in config ( #280 )
2021-12-13 19:57:15 +08:00
499d06d844
Add mmseg performance ( #254 )
...
* update doc and config
* test with ncnn
* update docs
* rename config
* update test result with openvino
* merge master
* resolve comments
* update dbnet
2021-12-09 20:17:00 +08:00
fa33b89261
[Docs]: Add MMEdit benchmark and supported model list ( #252 )
...
* doc
* update support
* update
* Resolve comments
* Update data
2021-12-09 11:27:20 +08:00
bca8cc0897
[fix] align the names of configs ( #256 )
...
* align names of configs
* set input_shape to None for all ppl
2021-12-07 21:06:44 +08:00
9e82851a73
[Feature] end2end yolov3 with ncnn ( #248 )
...
* support yolov3 ncnn with Yolov3DetectionOutput
* update nms
* fix contiguous in ncnn wrapper
* remove padding to detectionoutput
* format cpp
* Revert "format cpp"
This reverts commit 54050b19cd80d2f8cd851d82a755fd2c8d6c779d.
* fix zero detection
* fix yapf
* onnx2ncnn.cpp
* fix ut
* fix isort
* fix clang-format
* format cpp
* resolve comments
* resolve comments
* fix ut of ncnnend2endmodel
* fix yapf
* fix return list;
Co-authored-by: hanrui1sensetime <hanrui1@sensetime.com>
2021-12-06 19:01:36 +08:00
66d5cddbdc
[Enhancement] Add bicubic resize plugin for tensorrt ( #238 )
...
* save codes
* enable export fake bicubic interpolate op to onnx
* save codes
* enable bicubic interpolate trt plugin
* static export
* enable visualize but need align acc
* use torch bicubic upsample
* add unit tests for bicubic interpolate
* fix unit tests
* change mmedit config
* remove useless comments
* remove useless comments
* resolve comments
* fix lint
* clang-format
Co-authored-by: grimoire <yaoqian@sensetime.com>
2021-12-01 16:31:10 +08:00
bb9b0a98de
[Enhancement]: Added static config and CI tests for OpenVINO. ( #218 )
...
* Add openvino_static.
* Add openvino-dev.
* Fix skipping ORT tests in test_mmocr_models.
* Updated docs.
* Fix print.
* Fix
* Fix
* Fix other backends
* Fix is_available
* fix ncnn
* Add constrict for get rewrite output
* add not
* Fix
* fix
* Fix
* Fix
* Improve tests
* Remove rebundant `cuda`
* Prevent None object and rename variable
* Fix multi-line string
* rename get_backend_checker
* Add Troubleshooting to doc.
* Fix postprocessing_masks with empty masks.
* Fix tests
* lint
* Update docs.
Co-authored-by: SingleZombie <singlezombie@163.com>
2021-12-01 14:03:48 +08:00
cc2993e187
[Refactor] Refactor mmdet config ( #229 )
...
* Refactor mmdet config
* update partition model
* update docs related to mmdet
2021-11-30 15:00:11 +08:00
45a623966f
rm with-mask config and log .2f for timer.py ( #208 )
2021-11-17 17:31:26 +08:00
a4dceb4bb4
[Enhancement] Enhance ppl for all codebases ( #177 )
...
* enhance ppl for all codebases
* fix dump info
* fix md and use not None
* remove redundant codes
* safe convert empty ppl tensor
* add examples and remove useless lines
2021-11-16 19:16:46 +08:00
c05a839cfe
[FIX]fix with_mask to instance_seg ( #181 )
...
* fix with_mask to instance_seg
* fix instance_cfg to instance-cfg
2021-11-09 17:45:05 +08:00
778e3c9db0
[Enhance] Add SSD support for TensorRT ( #182 )
...
* support tensorrt ssd configs
* test ssd/cascade rcnn
* test shufflenetv1
* support trt mmdet end2end
2021-11-09 10:32:11 +08:00
cd51f12f32
[Feature]: Support yolov3 ( #167 )
...
* support yolov3 with ort and trt
* add ncnn compare
* fix yolo_head ncnn rewriter
* align perforance with ort trt for yolov3
* update doc
* add test for compare with equal,less, greater
* change namespace
* reformat cpp
* fix lint
* fix lint
* add unit test for yolov3 head
* remove compare op
* update doc
* update table format in docs
* update comments
* update
Co-authored-by: hanrui1sensetime <hanrui1@sensetime.com>
2021-11-08 16:07:58 +08:00
daee76ec99
[Enhancement] Optional export postprocess of mask ( #161 )
...
* Optional export postprocess of mask
* optional deploy_cfg
* delete print
2021-11-01 12:58:54 +08:00
5fdd75fba1
Add tensorrt int8 and fp16 configs for all the rest codebases ( #152 )
...
* add configs for all codebases except mmocr and fixed some bugs
* add ocr det
* add fp16 and int8 config for text recog
2021-10-27 15:59:02 +08:00
77131ea071
[Feature] Add exporting SARNet to onnxruntime backends. ( #134 )
...
* save codes
* save codes
* support sar exporting to onnxruntime and tensorrt
* use 4G max gpu space
* support sequential sar
* remove trt config for sar
* add sar unit tests
* add sar unit tests
* resolve conflicts
* remove torch.add rewrite in unit tests
2021-10-26 15:42:06 +08:00
c5a87fb1bc
[Enhancement]: OpenVINO deployment support for some models ( #117 )
...
* Fix include and lib paths for onnxruntime.
* Fixes for SSD export test
* Add onnx2openvino and OpenVINODetector. Test models: ssd, retinanet, fcos, fsaf.
* Add support for two-stage models: faster_rcnn, cascade_rcnn
* Add doc
* Add strip_doc_string for openvino.
* Fix openvino preprocess.
* Add OpenVINO to test_wrapper.py.
* Fix
* Add openvino_execute.
* Removed preprocessing.
* Fix onnxruntime cmake.
* Rewrote postprocessing and forward, added docstrings and fixes.
* Added device type change to OpenVINOWrapper.
* Update forward_of_single_roi_extractor_dynamic_openvino and fix doc.
* Update docs.
* Add OpenVINODetector and onn2openvino tests.
* Add input_info to onnx2openvino.
* Add TestOpenVINOExporter and test_single_roi_extractor.
* Moved get_input_shape_from_cfg to openvino_utils.py and added test.
* Added test_cascade_roi_head.
* Add backend.check_env() to tests.
* Add OpenVINO to get_rewrite_outputs and to some tests in test_mmdet_models.
* Moved test_single_roi_extractor to test_mmdet_models.
* Removed TestOpenVINOExporter.
2021-10-21 16:07:35 +08:00
5453f9befa
[Feature] Add tensorrt single stage partition ( #88 )
...
* Add tensorrt single stage partition
* add docstring&typehint
2021-09-23 15:49:36 +08:00
7e8154090a
align mmedit static cfg ( #82 )
2021-09-16 19:55:07 +08:00
4c0b36b7ff
[Refactor] Refactor config v1 ( #80 )
...
* [Refactor] Refactor configs according to new standard (#67 )
* modify cfg and cfg_util
* modify tensorrt config
* fix bug
* lint
* Fix
1. Delete print
2. Modify the return value from "False, None" to "None" and related code
3. Rename 2 get functions
* modify apply_marks
* [Feature] Refactor ocr config (#71 )
* add text detection config refactor
* add text recognition refactor
* add static exporting for mmocr
* fix lint
* set max space in child config
* use Sequence[int] instead
* add assert input_shape
* fix static bug and add ppl ort and trt static (#77 )
* [Feature] Refine setup.py (#61 )
* add setup.py and related files
* lint
* Edit requirements
* modify onnx version
* modify according to comments
* [Refactor] Refactor mmseg configs (#73 )
* refactor mmseg config
* change create_input
* fix lint
* fix lint
* fix lint
* fix yapf
* fix yapf
* update export
* remove Segmentation
* remove tast assert
* add onnx_config
* remove hardcode
* Inherit with static
* Remove blank line
* Add segmentation task enum
* add assert task
* mmocr version 0.3.0 (#79 )
* add dump_info
* [Feature]: Refactor config in mmdet (#75 )
* support onnxruntime
* add two stage
* test two-stage ort and ppl
* update fcos post_params
* fix calib
* test ok with maskrcnn dynamic
* add empty line
* add static into config filename
* add input_shape to create_input in mmdet
* add static to some configs
* remove todo codes
* remove partition config in base
* refactor create_input
* rename task name in mmdet
* return None if input_shape is None
* add size info into mmdet configs filenames
* reorganize mmdet configs
* add object detection task for mmdet
* rename get_mmdet_params
* keep naming style consistent
* update post_params for fcos
* fix typo in ncnn config
* [Refactor] Refactor mmedit static config (#78 )
* add static cfg
* update create_input
* [Refactor]: Refactor mmcls configs (#74 )
* refactor mmcls2.0
* fix classify_tensorrt_dynamic.py
* fix classify_tensorrt_dynmic.py
* classify_tensorrt_dynamic_int8.py
* fix file name
* fix ncnn ppl
* updata prepare_input.py
* update utils.py
* updata constant.py
* add
* fix prepare_input.py
* fix prepare_input.py
* add static config file
* add blank lines
* fix prepare_input.py(wait test)
* fix input_shape(wait test)
* Update prepare_input.py
* fix classification_tensorrt_dynamic(wait test)
* fix classification_tensorrt_dynamic_int8(wait test)
* fix classification_tensorrt_static_int8(wait test)
* Rename classification_tensorrt_dynamic.py to classification_tensorrt_dynamic-224x224-224x224.py
* Rename classification_tensorrt_dynamic_int8.py to classification_tensorrt_dynamic_int8-224x224-224x224.py
* Rename classification_tensorrt_dynamic_int8-224x224-224x224.py to classification_tensorrt_int8_dynamic_224x224-224x224.py
* Rename classification_tensorrt_dynamic-224x224-224x224.py to classification_tensorrt_dynamic_224x224-224x224.py
* Rename classification_tensorrt_static.py to classification_tensorrt_static_224x224.py
* Rename classification_tensorrt_static_int8.py to classification_tensorrt_int8_static_224x224.py
* Update prepare_input.py
* Rename classification_tensorrt_dynamic_224x224-224x224.py to classification_tensorrt_dynamic-224x224-224x224.py
* Rename classification_tensorrt_int8_dynamic_224x224-224x224.py to classification_tensorrt_int8-dynamic_224x224-224x224.py
* Rename classification_tensorrt_int8-dynamic_224x224-224x224.py to classification_tensorrt_int8_dynamic-224x224-224x224.py
* Rename classification_tensorrt_int8_static_224x224.py to classification_tensorrt_int8_static-224x224.py
* Rename classification_tensorrt_static_224x224.py to classification_tensorrt_static-224x224.py
* Update prepare_input.py
* Update prepare_input.py
* Update prepare_input.py
* Update prepare_input.py
* Update prepare_input.py
* Update prepare_input.py
* Update prepare_input.py
* change logging msg
Co-authored-by: maningsheng <mnsheng@yeah.net>
* fix
* fix else branch
* fix bug for trt in mmseg
* enable dump trt info
* fix trt static for mmdet
* remove two-stage_partition_tensorrt_static-800x1344 config
* fix wrong backend in ppl config
* fix partition calibration
Co-authored-by: Yifan Zhou <singlezombie@163.com>
Co-authored-by: AllentDan <41138331+AllentDan@users.noreply.github.com>
Co-authored-by: hanrui1sensetime <83800577+hanrui1sensetime@users.noreply.github.com>
Co-authored-by: RunningLeon <maningsheng@sensetime.com>
Co-authored-by: VVsssssk <88368822+VVsssssk@users.noreply.github.com>
Co-authored-by: maningsheng <mnsheng@yeah.net>
Co-authored-by: AllentDan <AllentDan@yeah.net>
2021-09-16 10:26:09 +08:00
3975d491d5
[Feature] Add ppl backend for ocr edit and seg ( #58 )
...
* add ppl for ocr edit and seg
* remove random crop after update mmedit
2021-09-07 15:40:43 +08:00
56bc1242b2
[Feature] Add NCNN on MMSegmentation ( #55 )
...
* fix custom ops support, fix multiple mark bug, add name mapping
* check if the value_info need to be added
* remove unnecessary print
* add nms implement
* two stage split wip
* add two stage split
* add split retinanet visualize
* add two stage split (wip)
* finish two stage split
* fix lint
* move parse string to mmdeploy.utils
* add calib data generator
* create calib dataset
* finish end2end int8
* add split two stage tensorrt visualize
* first
* fix0
* fix1
* dirty work
* wip
* add allocator
* finally done!
* lint
* fix lint
* better gather
* better onnx2ncnn
* fix tensorslice bugs
* fix lint
* fix clang-format
* remove comments
* fix expand
* int param
* fix lint
* [Fix] NCNN TensorSlice op bugs (#42 )
* fix custom ops support, fix multiple mark bug, add name mapping
* check if the value_info need to be added
* remove unnecessary print
* add nms implement
* two stage split wip
* add two stage split
* add split retinanet visualize
* add two stage split (wip)
* finish two stage split
* fix lint
* move parse string to mmdeploy.utils
* add calib data generator
* create calib dataset
* finish end2end int8
* add split two stage tensorrt visualize
* fix tensorslice bugs
* fix lint
* fix clang-format
* remove comments
* int param
* fix lint
Co-authored-by: grimoire <yaoqian@sensetime.com>
* add two stage ncnn support
* remove unused ops
* git unused config
* remove no_grad, should add in refactor
* add ncnn wrapper
* fix lint
* size return tuple
* Resolve grammar error
* Fix lint
* Trim Trailing Whitespace
* fix trim
* add argmax to topk
* add ArgMax parse
* add ncnn mmseg deploy cfg
* utils add ncnn mmseg
* add ncnn
* fix lint
* fix yapf
* fix clang-format-9
* remove debugging code
Co-authored-by: grimoire <yaoqian@sensetime.com>
Co-authored-by: grimoire <streetyao@live.com>
Co-authored-by: maningsheng <mnsheng@yeah.net>
2021-09-03 15:19:54 +08:00
e73d9fb50b
[Feature] Add mmocr ncnn support ( #53 )
...
* first
* fix0
* fix1
* dirty work
* wip
* add allocator
* finally done!
* lint
* fix lint
* better gather
* better onnx2ncnn
* fix expand
* [Fix] NCNN TensorSlice op bugs (#42 )
* fix custom ops support, fix multiple mark bug, add name mapping
* check if the value_info need to be added
* remove unnecessary print
* add nms implement
* two stage split wip
* add two stage split
* add split retinanet visualize
* add two stage split (wip)
* finish two stage split
* fix lint
* move parse string to mmdeploy.utils
* add calib data generator
* create calib dataset
* finish end2end int8
* add split two stage tensorrt visualize
* fix tensorslice bugs
* fix lint
* fix clang-format
* remove comments
* int param
* fix lint
Co-authored-by: grimoire <yaoqian@sensetime.com>
* add two stage ncnn support
* remove unused ops
* git unused config
* remove no_grad, should add in refactor
* add ncnn wrapper
* fix lint
* size return tuple
* Resolve grammar error
* Fix lint
* Trim Trailing Whitespace
* fix trim
* update wrapper
* remove logs
* remove
* csrc optimize
* add ncnn dbnet support
* finish crnn support
* add comment
Co-authored-by: hanrui1sensetime <83800577+hanrui1sensetime@users.noreply.github.com>
2021-09-03 15:16:20 +08:00
2b98040b26
[Feature] Refactor v1 ( #56 )
...
* [Refactor] add enum class and use functions to get configuration (#40 )
* add task and codebase enum class
* use funcitons to get config
* Refactor wrappers of mmcls and mmseg (#41 )
* move wrappers of cls & det to apis
* remove get_classes_from_config
* rename onnx_helper to onnx_utils
* move import to outside of class
* refactor ortwrappers
* Refactor build dataset and dataloader for mmseg (#44 )
* refactor build_dataset and build_dataloader for mmcls and mmseg
* remove repeated classes
* set build_dataloader with shuffle=False
* [Refactor] pplwrapper and mmocr refactor (#46 )
* add
* add pplwrapper and refactor mmocr
* fix lint
* remove unused arguments
* apply dict input for pplwrapper and ortwrapper
* add condition before import ppl and ort stuff
* update ppl (#51 )
* Refactor return value and extract_model (#54 )
* remove ret_value
* refactor extract_model
* fix typo
* resolve comments
* [Refactor] Refactor model inference pipeline (#52 )
* move attribute_to_dict to extract_model
* simplify the inference and visualization
* remove unused import
* [Feature] Support SRCNN in mmedit with ONNXRuntime and TensorRT (#45 )
* finish mmedit-ort
* edit __init__ files
* add noqa
* add tensorrt support
* 1. Rename "base.py"
2. Move srcnn.py to correct directory
* fix bugs
* remove figures
* align to refactor-v1
* update comment in srcnn
* fix lint
* newfunc -> new_func
* Add visualize.py
split visualize() in each codebase
* fix lint
* fix lint
* remove unnecessary code in ORTRestorer
* remove .api
* edit super(), remove dataset
* [Refactor]: Change name of split to partition (#57 )
* refactor mmcls configs
* refactor mmdet configs and split params
* rename rest split to partition from master
* remove base.py
* fix init of inference class
* fix mmocr init, add show_result alias
Co-authored-by: AllentDan <41138331+AllentDan@users.noreply.github.com>
Co-authored-by: RunningLeon <maningsheng@sensetime.com>
Co-authored-by: Yifan Zhou <singlezombie@163.com>
2021-09-01 15:43:49 +08:00
823ca38646
[Feature] Add NCNN mmdetection support ( #49 )
...
* first
* fix0
* fix1
* dirty work
* wip
* add allocator
* finally done!
* lint
* fix lint
* better gather
* better onnx2ncnn
* fix expand
* [Fix] NCNN TensorSlice op bugs (#42 )
* fix custom ops support, fix multiple mark bug, add name mapping
* check if the value_info need to be added
* remove unnecessary print
* add nms implement
* two stage split wip
* add two stage split
* add split retinanet visualize
* add two stage split (wip)
* finish two stage split
* fix lint
* move parse string to mmdeploy.utils
* add calib data generator
* create calib dataset
* finish end2end int8
* add split two stage tensorrt visualize
* fix tensorslice bugs
* fix lint
* fix clang-format
* remove comments
* int param
* fix lint
Co-authored-by: grimoire <yaoqian@sensetime.com>
* add two stage ncnn support
* remove unused ops
* git unused config
* remove no_grad, should add in refactor
* add ncnn wrapper
* fix lint
* size return tuple
* Resolve grammar error
* Fix lint
* Trim Trailing Whitespace
* fix trim
* update wrapper
* remove logs
* remove
* csrc optimize
Co-authored-by: hanrui1sensetime <83800577+hanrui1sensetime@users.noreply.github.com>
2021-08-26 18:40:14 +08:00
8fe8056080
[Feature] Add trt ort mdcn plugins ( #43 )
...
* add trt mdcn plugin
* add onnxruntime mdcn op
* add mdcn trt ort plugins
* fix lint
* remove comment
* remove plugin condition lines
* apply new form
* use serialized_size
2021-08-25 10:17:21 +08:00
e5dc959276
[Feature]: Support PSPNet with static shape in ORT and TRT ( #39 )
...
* support pspnet with static shape
* set default as None to --test-img
2021-08-20 13:39:03 +08:00
fd40119694
[Feature] TensorRT int8 support ( #35 )
...
* fix custom ops support, fix multiple mark bug, add name mapping
* check if the value_info need to be added
* remove unnecessary print
* add nms implement
* two stage split wip
* add two stage split
* add split retinanet visualize
* add two stage split (wip)
* finish two stage split
* fix lint
* move parse string to mmdeploy.utils
* add func mark count dict
* use assert_cfg_valid
* update func count before add Mark
* fix dynamic shape support
* add calib data generator
* create calib dataset
* finish end2end int8
* add split two stage tensorrt visualize
2021-08-19 12:56:00 +08:00
7c465654b9
[Feature] Add dbnet18 and crnn support ( #36 )
...
* add mmocr
re-orgnize codes
save codes
save codes
save codes
add dbnet18
* fix lint
* support crnn
* fix lint
* resolve comments
* resolve conflicts
* resolve module registry error
* test
* resolve comments
* fix lint
* use algrithm_type and add dataset evaluation
* suit newest mmocr
* fix lint and remove trt version check
2021-08-19 11:53:00 +08:00
4c8c34d354
[Feature] Enable faster rcnn split onnxruntime ( #27 )
...
* fix custom ops support, fix multiple mark bug, add name mapping
* check if the value_info need to be added
* remove unnecessary print
* add nms implement
* two stage split wip
* add two stage split
* add split retinanet visualize
* add two stage split (wip)
* finish two stage split
* fix lint
* move parse string to mmdeploy.utils
* add func mark count dict
* use assert_cfg_valid
* update func count before add Mark
* fix dynamic shape support
2021-08-13 10:06:28 +08:00
7dbc12d23f
[Feature]: Support FCN,DeeepLabV3, DeepLabV3Plus in mmseg with ONNXRuntime and TensorRT ( #31 )
...
* fix mask empty result
* support fcn exporting to ONNX for ort and trt in whole mode
* resolve comments
* remove unnecessary code
* update prepare_input
* rewrite psp_head & aspp_head
* test fcn deeplabv3 deeplabv3plus with trt
2021-08-12 16:44:16 +08:00
dcb88e4439
[Feature] Add ppl backend. ( #32 )
...
* add ppl support
* save
* add ppl speed test
* drop "add ppl speed test"
This reverts commit 467504a23074a2defdadeff9819c8d19bfefda71.
* add ppl backend
* resolve comments
* fix baseDeployClassifier
2021-08-12 14:44:10 +08:00
e05521c933
[Feature] Merge NCNN deployment to grimoire based on mmcls - revert [ #25 ]( https://github.com/grimoire/deploy_prototype/pull/25 ) ( #30 )
...
* add
* change VulkanSDK to 1.2.176.1
* add ncnn cmakelist
* add ncnn source code as third party
* add all ncnn
* ncnn compile passed
* onnx2ncnn correctly
* fix code style
* merge_as_grimoire_design, only backend_ops, manually register.
* remove data and test sh
* remove build example
* remove config ncnn
* remove onnx2ncnn intermediate files
* remove other files auto-generated
* remove vulkan tools
* remove Vulkan, gitignore new rules, __init__ new lines
* rollback __init__ to grimoire
* remove pytorch version pending
* grimoire comments reply 1, 3, 4
* reply comment 5,6,7
* add auto definer, add python register
* fix lint
* add ncnn deploy support
* add model_wrapper, fix a typo bug, and add code comment for onnx2ncnn(WIP)
* add model wrapper ncnn
* fix lint
* fix pep8
* fix pre-commit-config.yaml paths
* fix import
* fix lint
* remove sys.path.append
* remove sys
* isort fix
* fix double quoted
* fix trailing space
* try fix isort
* fix clang-format-9
* fix requests
* fix all comments
* Fix typo
* test code for grimoire
* fix ops register
* new definere
* fix visualization of mmcls
* remove temp
* fix flake8
* fix seed-isort-config
* fix thirdparty
* fix thirdparty
* fix yapf
* fix third_party_sort
* fix third party
* fix clang-format
* try fix clang-format
* try to fix clang format 9 customreshape
* try fix clang-format-9
* try fix clang-format-9
* try fix clang-format-9
* try fix ext
* fix onnx2ncnn
* Fix comments
* Fix Comments
* Fix Comments
* Fix Comments
* Fix conflict
* Fix flake8
* Update .isort.cfg
* Update ncnn_ext.cpp
* Update ncnn_ext.cpp
* fix missing ncnn backend code
* delete out of date comments of gather.cpp
* add DeployBaseClassifier
* add return -100 error
* clear out-of-date to do comments
Co-authored-by: 韩睿 <SENSETIME\hanrui1@cn0614008774l.domain.sensetime.com>
Co-authored-by: grimoire <yaoqian@sensetime.com>
Co-authored-by: grimoire <streetyao@live.com>
2021-08-05 14:06:47 +08:00
f607f1965b
[Feature] Align datasets ( #29 )
...
* add test tool and re-orgnize apis.utils
* handle topk and refine codes
* add cls export and test support
* fix lint
* move ort into wrapper
* resolve conflicts
* resolve comments
* resolve conflicts
* resolve comments and padding mrcnn
* resolve comments
2021-08-03 17:12:44 +08:00
90ce7207da
[Feature] Support Mask R-CNN with ONNX Runtime and TensorRT ( #24 )
...
* add grid_sampler plugin
* update test_mixins to support maskrcnn
* add config for maskrcnn exporting to onnx
* fix lints
* fix lint
* resolve comments
* fix lint
* fix lint
* fix lint
* support trt8
* move mark to single_stage
* fix lint
* remove mark in single stage
2021-07-28 19:02:43 +08:00
ff39377833
[Feature] Retinanet split support ( #19 )
...
* prepare for split
* mark multiclass nms
* redesign mark decorator to handle input/output names better (#17 )
* redesign mark decorator to handle input/output names better
* fixed a bug in rename_value; robustify extract_model
* fix lint
* fix log in deploy pipeline
* fix onnxruntime config
* fix lint
* update connect init
* rename is_inspecting
* simplify function_mark
Co-authored-by: Li Zhang <lzhang329@gmail.com>
2021-07-23 13:18:32 +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
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
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
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
6c47ee3d2a
add tensorrt support ( #2 )
2021-06-23 13:14:28 +08:00
VVsssssk
Yifan Zhou
hanrui1sensetime
AllentDan
RunningLeon
Semyon Bevzyuk
q.yao
lzhangzz
grimoire