Dynamic batchsize export support for ONNX and TensorRT #280
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
This was referenced Jul 22, 2022
|
#329 contains the same changes for upstream changes to ONNX and TensorRT export Closing on favor of #329 |
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Outdated suggestions cannot be applied.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 enables the "--dynamic-batch" option for ONNX export. This will set the input dimension of the network to [batch,3,640,640] and allows arbitrary batch sizes to be run with the model.
Working Colab Demo:
https://colab.research.google.com/drive/1vH5wwUvWoGrJ989EzNcESs1tFxW6i1K7?usp=sharing
Netron Info:
Works with NMS plugin:
TensorRT now enables optimization profiles for arbitrary batch sizes:
1900 FPS on RTX 3090 for yolov7-tiny (batch size 16 in this screenshot)
Also a few additional refactoring changes - e.g. to set score_thresh and detections_per_image for NMS plugin