75a9ffa611
fix ncnn linear ( #299 )
...
* fix ncnn linear
* lstm rewriter
* fix lint
* add comment
2021-12-17 14:12:37 +08:00
f442dbbe83
[Docs]: MMCLS benchmark ( #300 )
...
* add doc
* Update benchmark
* Update benchmark
* Restore missing words
2021-12-17 14:12:26 +08:00
e54d607f24
[SDK] Add OpenVINO SDK support ( #257 )
...
* Add OpenVINO SDK support
* fix reshape
* remove resize input
2021-12-17 13:45:32 +08:00
f94d7b5abe
[Feature] Dump jsons to SDK ( #279 )
...
* save cods
* save codes
* save codes
* add docstring and fix ut
* use Resize to replace img_scale
* use Dict instead of dict()
* net && weights
* resolve comments
* update pipeline
* update pipeline
* update pipeline
* refactor preprocess
* merge master
* use imagetotensor
* use imagetotensor
* fix typos and improve docstring
* resolve comments
* update lq to img in mmedit
* add mask_thr_binary
* refactor dump info
* resolve comments
* add instance seg to dump info temperarily
* update meta_keys
* update meta_keys
* add meta_keys condition
* update postprocess for mmseg
2021-12-17 11:39:16 +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
31e8aed862
update ppl benchmark ( #293 )
2021-12-17 10:44:49 +08:00
1ed55f86f2
Update README ( #288 )
...
* update README and add three blank docs that are going to present SDK
* check in mmdeploy's logo image
* remove comments in README
* update acknowledgement
* development->deployment
* add Human3D link
* use captical word in citation
2021-12-16 19:22:14 +08:00
de6fc147bd
[Feature]: Add onnx2pplnn tool ( #296 )
...
* Add onnx2pplnn tool
* format
* typo
* update opt-shapes and wrapper
* Update opt-shape
* Update comment
* lint
2021-12-16 17:01:33 +08:00
f807346592
[SDK] update ppl.cv to latest version and use it's cmake package ( #294 )
2021-12-16 16:33:14 +08:00
a494a6f6ff
[SDK] sync changes according to performance benchmarks ( #297 )
...
* sync SDK changes according to performance benchmarks
* fix end-of-file lint
* fix clang-format issue
* fix clang-format by adding 'clang-format off'
* remove useless casts
* remove 'data' argument of 'operator()'
* change 'Tensor2Img' to 'TensorToImg' according to spec
* correct tensor's name according spec
Co-authored-by: lvhan028 <lvhan_028@163.com>
2021-12-16 13:51:22 +08:00
75b5939471
add ncnn custom ops to sdk ( #284 )
2021-12-15 21:34:32 +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
43256977fa
Add onnx2ncnn tool ( #276 )
...
* Add onnx2ncnn tool
* fix docs
* fix lint
2021-12-15 18:43:09 +08:00
b1a202ef04
Add onnx2tensorrt tool ( #275 )
...
* add onnx2tensorrt
* add docs
* fix log
* fix docs
2021-12-15 11:20:10 +08:00
469b6aab4c
Template for issue & pr ( #287 )
...
* add docs
* update
* update
* resolve comments
2021-12-14 16:30:44 +08:00
78b37bbd32
[Fix]: Fix rewriter conflict when processing derived class ( #289 )
...
* Fix rewriter
* lint
* rename function and update docstring
* use is class
* Update docstring
2021-12-14 15:40:29 +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
c8d0d6a285
fix dbnet export ( #291 )
2021-12-13 19:46:21 +08:00
aff568d508
[Fix]: update doc requirement ( #274 )
...
* update requirements
* fix version import
2021-12-13 19:36:30 +08:00
105acc9de9
[Fix] Dbnet performance of trt8 ( #278 )
...
* compatible trt version for dbnet
* judge inside rewrite
2021-12-13 16:44:05 +08:00
bd2867178d
Disable scatternd in TRT8+ ( #277 )
2021-12-13 16:43:22 +08:00
25a5838080
change trt to tensorrt and ort to onnxruntime according to sdk model spec ( #282 )
...
* change trt to tensorrt and ort to onnxruntime according to sdk model spec
* change ort to onnxruntime in config files
2021-12-13 14:58:01 +08:00
f2253a0f29
remove file ( #285 )
2021-12-13 13:01:30 +08:00
a96e5f9d76
[Fix] Fix VFNet test ( #281 )
...
* [Fix] fix bugs for mmcls performance test (#269 )
* fix bugs for mmcls performance test
* fix yapf
* add comments of CLASSES attribute
* Fix test_get_bboxes_of_vfnet_head
* Fix
Co-authored-by: hanrui1sensetime <83800577+hanrui1sensetime@users.noreply.github.com>
2021-12-13 10:43:39 +08:00
0f90a0af0a
[Fix] fix bugs for mmcls performance test ( #269 )
...
* fix bugs for mmcls performance test
* fix yapf
* add comments of CLASSES attribute
2021-12-10 14:15:08 +08:00
54885e5972
【Feature】Support for mmdet2.19 ( #253 )
...
* add one-stage support
* add two-stage support
* fix docstring
* add base_dense_head docstring
* fix docstrign
* fix ncnn code
* fix anchor head
* fix unit test about mmdet model
* fix unit test
* fix base dense head bug:
* fix basehead
* fix basedense head
* fix pad bug
* fix bboxhead
* add rewrite about decode
* fix decode
* rename func
* fix test_mmdet_core and test_object_detection_model
* format test
* fix comments
* remove to(device)
* fix requirements
* fix docs
* fix yolox evaluate butg
2021-12-10 14:03:23 +08:00
4149228716
[Enhancement]: Import codebase only when it is required ( #266 )
...
* Add import codebase
* lint
* Fix import order
* typo
* Fix partition
* docstring
* lint
2021-12-10 11:34:22 +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
1948af6d06
eliminate inline template when build using release mode ( #268 )
2021-12-09 17:35:41 +08:00
0897139744
[Fix] Fix onnx multiple domain registry ( #270 )
...
* Fix onnx multiple domain registry
* recover test args
* remove wrong status
* replace map with unordered_map
* add symbolic rewriter
2021-12-09 17:35:28 +08:00
fd4297a2a3
add speed benchmark for pplnn and replace all ppl in docs with pplnn ( #264 )
2021-12-09 16:37:36 +08:00
3659b515eb
[Enhancement]: The ability to sort the dataset in tools/test.py. ( #244 )
...
* Add sort-data arg to test.py
* Set is_sort_dataset to True.
* Add a check for the possibility of sorting.
* lint
* Added mmdeploy.utils.dataset.
* Add unit test
Co-authored-by: SingleZombie <singlezombie@163.com>
2021-12-09 16:21:50 +08:00
ad72c19482
[Feature] Add torch2onnx tools ( #260 )
...
* Add torch2onnx tools
* fix lint
* update docs
* fix docs2
* fix log
2021-12-09 14:38:54 +08:00
d5c3be9739
[Fix] Apply kernel loop for bicubic interpolate ( #271 )
2021-12-09 12:56:32 +08:00
c213879103
fix yolov3 head test with ncnn ( #267 )
...
* fix yolov3 head test with ncnn
* add ncnn test
* remove ncnn test
2021-12-09 11:56:12 +08:00
30af457b5e
[Feature] Support mobilenetv2ssdlite ncnn end2end deployment. ( #259 )
...
* support ssdlitemobilenetv2
* fix yapf
* fix ut try
* fix ut try
* try to fix ut
* fix ut general 320->128
* move onnx ops to mmcv
* fix lint
* fix docformatter
* increase doc coverage
* refact self defined ops
* fix yapf
* add description
* fix isort
* fix comments
* fix comments
* fix lint
* sigmoid after concat
2021-12-09 11:55:38 +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
d96ee9e9f3
Change op domain ( #262 )
...
* change domain to mmdeploy
* update tests
* resolve comments
2021-12-08 15:06:41 +08:00
03c95a1149
[Enhancement]: Optimize config utils ( #263 )
...
* Optimize config utils
* Update `get_backend`
* Add assert
2021-12-08 14:55:59 +08:00
f424fca509
[Doc] add mmocr performance benchmark ( #240 )
...
* add mmocr performance benchmark
* add notes
* fix typo
* add openvino for mmocr and use T4 results for dbnet18
2021-12-08 10:31:57 +08:00
4f70434070
[Fix] correct ncnn_DIR ( #261 )
...
* correct ncnn_DIR
* change NCNN_DIR to ncnn_DIR
2021-12-08 10:29:07 +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
ec1c15dc07
Fix load tensorrt plugin anyway ( #255 )
2021-12-07 21:06:32 +08:00
8f769e2355
Fix ncnn_ext build error ( #258 )
2021-12-07 17:00:39 +08:00
36124f6205
Merge sdk ( #251 )
...
* check in cmake
* move backend_ops to csrc/backend_ops
* check in preprocess, model, some codebase and their c-apis
* check in CMakeLists.txt
* check in parts of test_csrc
* commit everything else
* add readme
* update core's BUILD_INTERFACE directory
* skip codespell on third_party
* update trt_net and ort_net's CMakeLists
* ignore clion's build directory
* check in pybind11
* add onnx.proto. Remove MMDeploy's dependency on ncnn's source code
* export MMDeployTargets only when MMDEPLOY_BUILD_SDK is ON
* remove useless message
* target include directory is wrong
* change target name from mmdeploy_ppl_net to mmdeploy_pplnn_net
* skip install directory
* update project's cmake
* remove useless code
* set CMAKE_BUILD_TYPE to Release by force if it isn't set by user
* update custom ops CMakeLists
* pass object target's source lists
* fix lint end-of-file
* fix lint: trailing whitespace
* fix codespell hook
* remove bicubic_interpolate to csrc/backend_ops/
* set MMDEPLOY_BUILD_SDK OFF
* change custom ops build command
* add spdlog installation command
* update docs on how to checkout pybind11
* move bicubic_interpolate to backend_ops/tensorrt directory
* remove useless code
* correct cmake
* fix typo
* fix typo
* fix install directory
* correct sdk's readme
* set cub dir when cuda version < 11.0
* change directory where clang-format will apply to
* fix build command
* add .clang-format
* change clang-format style from google to file
* reformat csrc/backend_ops
* format sdk's code
* turn off clang-format for some files
* add -Xcompiler=-fno-gnu-unique
* fix trt topk initialize
* check in config for sdk demo
* update cmake script and csrc's readme
* correct config's path
* add cuda include directory, otherwise compile failed in case of tensorrt8.2
* clang-format onnx2ncnn.cpp
Co-authored-by: zhangli <lzhang329@gmail.com>
Co-authored-by: grimoire <yaoqian@sensetime.com>
2021-12-07 10:57:55 +08:00
eceeaf0159
[Docs] fix ncnn docs ( #249 )
...
* fix ncnn docs
* add prepare for protobuf
* move install protob position
2021-12-06 19:03:18 +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
2f6f6f87a1
Add get_mo_command. ( #247 )
2021-12-06 10:31:32 +08:00
q.yao
Yifan Zhou
AllentDan
VVsssssk
hanrui1sensetime
RunningLeon
lvhan028
lzhangzz
Semyon Bevzyuk