Nanodet jit and c fixes #490
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…t_plus_fast model
fix optimization procedure and add dynamic and channel last implementations for faster inference
fix optimization procedure and add dynamic and channel last implementations for faster inference
…to Real-Time-OD-In-AGRICULTURAL
fix nanodet_RepVGG_A0_416.yml and nanodet_g.yml, simplify vgg backbone transfer tensorRT dependencies into gpu installation small fixes into prints - docks
check all backbone, fpns, heads, modules for jit cripting compatible implementations. update docks. add export compatible with C api. fix C api, outputs and preprocessing. add new demo for c export - update docks.
ManosMpampis
added
test sources
ManosMpampis
requested review from
passalis,
omichel and
stefaniapedrazzi
as code owners
omichel
requested changes
Nov 24, 2023
projects/python/perception/object_detection_2d/nanodet/README.md
Outdated
Show resolved
Hide resolved
projects/python/perception/object_detection_2d/nanodet/README.md
Outdated
Show resolved
Hide resolved
Co-authored-by: Olivier Michel <Olivier.Michel@cyberbotics.com>
Co-authored-by: Olivier Michel <Olivier.Michel@cyberbotics.com>
tsampazk
requested changes
Nov 24, 2023
projects/python/perception/object_detection_2d/nanodet/export_c_compatible_network.py
Outdated
Show resolved
Hide resolved
Co-authored-by: Kostas Tsampazis <27914645+tsampazk@users.noreply.github.com>
|
I suggest not merging this until #488 is reviewed and ready to be merged as well, as i think it would complicate #488's review. @omichel @ManosMpampis @passalis. |
Co-authored-by: Kostas Tsampazis <27914645+tsampazk@users.noreply.github.com>
update model heads to have unified initializations and update config_file_detail.md . delete unused parameter of yml files.
…ndr-eu/opendr into nanodet-jit-and-c-fixes
added 2 commits
November 29, 2023 13:30
…nodet-jit-and-c-fixes # Conflicts: # docs/reference/object-detection-2d-nanodet.md # src/opendr/perception/object_detection_2d/nanodet/algorithm/nanodet/data/transform/warp.py # src/opendr/perception/object_detection_2d/nanodet/algorithm/nanodet/inferencer/utilities.py # src/opendr/perception/object_detection_2d/nanodet/algorithm/nanodet/model/arch/one_stage_detector.py # src/opendr/perception/object_detection_2d/nanodet/algorithm/nanodet/model/backbone/custom_csp.py # src/opendr/perception/object_detection_2d/nanodet/algorithm/nanodet/model/backbone/efficientnet_lite.py # src/opendr/perception/object_detection_2d/nanodet/algorithm/nanodet/model/fpn/ghost_pan.py # src/opendr/perception/object_detection_2d/nanodet/algorithm/nanodet/model/head/gfl_head.py # src/opendr/perception/object_detection_2d/nanodet/algorithm/nanodet/model/head/nanodet_plus_head.py # src/opendr/perception/object_detection_2d/nanodet/algorithm/nanodet/model/module/conv.py # src/opendr/perception/object_detection_2d/nanodet/nanodet_learner.py
tsampazk
approved these changes
Nov 29, 2023
lucamarchionni
pushed a commit
to lucamarchionni/opendr
that referenced
this pull request
Jun 10, 2024
* add new augmentation strategies * add VGG backbone * easier code readability to add custom losses * better code readability * better code readability and small bug fixes * better and more stable training implementation and logging * add cache in dataset * add faster post processing in nanodet_plus_head.py and add new nanodet_plus_fast model * add TensorRT optimizations and fix embedded device inference, fix optimization procedure and add dynamic and channel last implementations for faster inference * add TensorRT optimizations and fix embedded device inference, fix optimization procedure and add dynamic and channel last implementations for faster inference * update docks and add warning ignores in test_nanodet * delete unused code, fix nanodet_RepVGG_A0_416.yml and nanodet_g.yml, simplify vgg backbone transfer tensorRT dependencies into gpu installation small fixes into prints - docks * optimize some memory usage during optimizations. check all backbone, fpns, heads, modules for jit cripting compatible implementations. update docks. add export compatible with C api. fix C api, outputs and preprocessing. add new demo for c export - update docks. * styletest fix * fix a bug that still uses gpu nodes even self.device was set to "cpu" * update c test * Apply suggestions from code review Co-authored-by: Olivier Michel <Olivier.Michel@cyberbotics.com> * Apply suggestions from code review Co-authored-by: Olivier Michel <Olivier.Michel@cyberbotics.com> * Apply suggestions from code review Co-authored-by: Kostas Tsampazis <27914645+tsampazk@users.noreply.github.com> * update optimize_c_model dock as noted in PR suggestions * update tool dock as noted in PR suggestions * Apply suggestions from code review Co-authored-by: Kostas Tsampazis <27914645+tsampazk@users.noreply.github.com> * update dynamic input explanation accross optimization and inference. update model heads to have unified initializations and update config_file_detail.md . delete unused parameter of yml files. * fix TensorRT loading message and docs as suggested * update dock about dynamic input * delete unused code --------- Co-authored-by: ManosMpampis <gmp.manos@gmail.com> Co-authored-by: Nikolaos Passalis <passalis@users.noreply.github.com> Co-authored-by: Olivier Michel <Olivier.Michel@cyberbotics.com> Co-authored-by: Kostas Tsampazis <27914645+tsampazk@users.noreply.github.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR aims to translate some changes of #488 into C API and simplify some optimizations for smaller memory footprint during inference and optimizations.
Adds dynamic shape capabilities into JIT optimizations for Python and C APIs.
Fix some buggs with C APIs during model export and when the output had zero bounding boxes.