mmr/tkDNN
Archived
1
0
Fork 0
Code Issues 46 Pull Requests 12 Releases 4 Activity
567 Commits 11 Branches 5 Tags
7298dcfb2f4db8c98dbc47a0282d6b99256d8f49
Commit Graph

567 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
perseusdg 7298dcfb2f Added monodepth2.cpp 2022-01-06 16:30:11 +00:00
perseusdg b75cecb105 Added constant and zero padding with ISliceLayer,need to add them for tensorrt versions less than 8.2 using IPluginV2Ext instead of ISliceLayer since they dont seem to support reflect and zero 2022-01-04 09:03:04 +05:30
perseusdg ba022663f1 completed padding migrations from github 2022-01-03 19:49:32 +05:30
perseusdg 6837644eb2 add reflection padding from github 2022-01-03 18:58:13 +05:30
perseusdg dcf4054bc6 performance improvements for yolo based networks,significant reduction in inference time can be seen yolo4tiny ,yolo4 and minor reduction in inference time can be seen in yolo4_berkeley_f1 and yolo4_berkeley - tested with a batchsize of 1 and 2 and on gtx 1070,it is possible that the performance improvement is more signficant in newer hardware 2022-01-03 13:52:16 +05:30
Micaela Verucchi a4781244f4 Fix merge problem 
Signed-off-by: Micaela Verucchi <micaelaverucchi@gmail.com>
 2021-11-24 18:16:15 +01:00
Micaela Verucchi bbae618118 Update README and add print in demo 
Signed-off-by: Micaela Verucchi <micaelaverucchi@gmail.com>
 2021-11-24 18:08:41 +01:00
Micaela Verucchi 0707c26bbd Merge of perseusdg-tensorrt8 inside tkDNN 
Signed-off-by: Micaela Verucchi <micaelaverucchi@gmail.com>
 2021-11-24 17:59:27 +01:00
perseusdg 9cecc5051a -Modified Dockerfile.base to cudagl 
-Changed demo to take in input from demoConfig.yaml file
-Readme changes for demo.md
 2021-11-24 03:39:57 +05:30
perseusdg a8c98e3c31 Signed-off-by: perseusdg <43143075+perseusdg@users.noreply.github.com> 
removed all TRT8_DEPRACTED functions
 2021-11-19 22:40:52 +05:30
perseusdg 6f936096ae Added num of layers in net and netRT,suggestions from pull request 270 on main repo 2021-11-14 07:13:42 +05:30
perseusdg 367061fea2 fps fix 2021-11-13 18:11:46 +05:30
perseusdg d488c3bd17 remove duplicated lines 2021-11-13 02:12:50 +05:30
perseusdg ee5000ccca bug fixes for dla networks and ported optimization from different pull request 2021-11-13 02:00:49 +05:30
perseusdg 9e328c0daa Migrate tkDNN max pooling plugin creation to pluginRegistry from the default method 2021-11-10 20:09:57 +05:30
perseusdg 7dd33cd118 added assert for supportsFormat in plugins 2021-11-10 14:37:16 +05:30
perseusdg 936b680f2f Remove debug prints 2021-11-10 13:45:44 +05:30
perseusdg c0e2097397 - CMakeLists.txt opencv cuda contrib autodetect 
- Updated Docker to cuda-11.3+cudnn-8.2.1+TensorRT-8.0.34,Ubuntu to 20.04 and OpenCV to 4.5.4
- Updated OpenCV4 to 4.5.4 in install_OpenCV4.sh
- Updated README.md
 2021-11-10 12:30:57 +05:30
perseusdg 744396fb0e Update README.md,windows.md and demo.cpp 
Small fixes in DeformableConvRT.cpp
 2021-11-09 13:37:26 +05:30
Micaela Verucchi d2d44e9e92 Fix max elem (remove thrust) for segmentation 
Signed-off-by: Micaela Verucchi <micaelaverucchi@gmail.com>
 2021-11-05 19:54:02 +05:30
perseusdg 802c01bd8f windows debug fix 2021-10-30 19:29:21 +05:30
perseusdg 7a89a4a573 ReshapeRT.cpp fix 2021-10-28 23:43:12 +05:30
perseusdg c5e66c6bf6 TRT8 works with almost every nerual network now!!!!(including demo3d) 2021-10-28 23:35:37 +05:30
perseusdg 8c36dd0431 mobilenet works with trt7(IPluginV2IOExt) ,need to test it with trt8 2021-10-27 18:00:33 +05:30
perseusdg 54e7af11ed tensorrt7 support for ipluginv2 2021-10-18 18:07:57 +05:30
perseusdg 18cc6abbb6 DeformableConvRT.h fixes ,cmake cuda arch auto detection 2021-09-07 23:57:52 +05:30
perseusdg 03473743c4 tkDNN works with trt8!!,need to test int8 and mobilenet,dla_cnet (fps seems to be a bit low 350 on trt8 compared to 396 on trt7) 2021-09-07 22:00:28 +05:30
perseusdg 9fa116ce4a tkdnn(trt8) runs now minus the detections ,[executionContext.cpp::enqueueInternal::312] Error Code 3: Internal Error (Parameter check failed at: runtime/api/executionContext.cpp::enqueueInternal::312, condition: mDeviceMemorySize == 0 || mExecutionResources->getDeviceMemory() != nullptr - is the error that causes it 2021-09-07 07:20:46 +05:30
perseusdg 504dee5016 test_yolo4 netRT->destroy() 2021-09-06 15:35:57 +05:30
Harshvardhan Chandirasekar 65ba5c9844 tkDNN can now deserialize tensorrt-8 engine (both through test_* and trtexec) 
but demo has issues in yolo::computeDetections
 2021-09-05 01:06:56 +05:30
perseusdg de83ae5d25 update tensorrt8 branch 2021-08-30 19:04:26 +05:30
perseusdg 2ffe07057e Mnist works at the moment with trt8,others like yolo4tiny and mobilenet generate the engine files but crash after throwing nvifer1::CudaRuntimeError and when demo is being run ,it doesnt deserialize properly and crashes 2021-08-29 03:18:58 +05:30
perseusdg 3d8b1ac494 Yolo3Detection.cpp doesn't build yet,issues with dependecies of preprocessing and postprocessing on IPluginFactory 2021-08-21 22:29:36 +05:30
perseusdg eba78e7e78 Updates to build libkernel.so under TensorRT 8 2021-08-16 13:59:19 +05:30
perseusdg d2ce966c46 fix return type for bool functions 2021-08-04 12:37:58 +05:30
Harshvardhan Chandirasekar bb902f5b65 fix for downloading weights(utils.cpp) when using docker 2021-07-25 23:53:30 +05:30
perseusdg 1412aa66c4 minor fixes for windows ,for release v0.6 2021-07-24 08:51:44 -07:00
Harshvardhan Chandirasekar e58ddadb7d Merge branch 'ceccocats:master' into master 2021-07-24 08:18:17 -07:00
Micaela Verucchi a992c9feb5 Update README.md 2021-07-23 16:50:15 +02:00
Micaela Verucchi 09080709a9 Remove Issues.md 
Signed-off-by: Micaela Verucchi <micaelaverucchi@gmail.com>
 2021-07-23 15:23:47 +02:00
Micaela Verucchi 7521d10ba7 Update README with gifs 
Signed-off-by: Micaela Verucchi <micaelaverucchi@gmail.com>
 2021-07-23 15:07:52 +02:00
Micaela Verucchi ab349083bf Merge with cnet works, all tests passed. 
Signed-off-by: Micaela Verucchi <micaelaverucchi@gmail.com>
v0.6 		2021-07-23 14:37:04 +02:00
Micaela Verucchi 4f9f27152a Update READMEs 
Signed-off-by: Micaela Verucchi <micaelaverucchi@gmail.com>
 2021-07-23 10:31:34 +02:00
Davide Sapienza 0292e21c33 Merge remote-tracking branch 'origin/master' into cnet 2021-07-22 17:03:23 +02:00
Davide Sapienza 2bb70859da Add FPS results for CenterTrack 2D/3D 
Signed-off-by: Davide Sapienza <sapienza.dav@gmail.com>
 2021-07-22 16:54:25 +02:00
Davide Sapienza b9f82510d3 Add demo_utils. 
This commit splits the utils file into two different files:
utils and demo_utils, to overcome some opencv problems caused
by demo_utils includes.

Signed-off-by: Davide Sapienza <sapienza.dav@gmail.com>
 2021-07-22 10:45:18 +02:00
Davide Sapienza 1216e8bb74 Add camera calibration file reading for demo3d and demoTracker. 
Some minor fixes.

Signed-off-by: Davide Sapienza <sapienza.dav@gmail.com>
 2021-07-21 16:14:53 +02:00
Davide Sapienza be2d361ac6 Update 3d detection 
This commit updates the CenternetDetection3d class and fixes
some bugs.

It also removes the resnet101 network for CenterNet 3D.
This network doesn't exist.

Signed-off-by: Davide Sapienza <sapienza.dav@gmail.com>
 2021-07-21 15:51:13 +02:00
Davide Sapienza f78f7bfddc Fix link in readme files 
Signed-off-by: Davide Sapienza <sapienza.dav@gmail.com>
 2021-07-21 15:35:17 +02:00
Micaela Verucchi f4e71a5a28 Add yolo4_berkeley_f1 
Signed-off-by: Micaela Verucchi <micaelaverucchi@gmail.com>
 2021-07-21 12:56:00 +02:00