Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Preprocess speed up and some fix #54

Merged
merged 22 commits into from
Sep 16, 2024
Merged

Preprocess speed up and some fix #54

merged 22 commits into from
Sep 16, 2024

Conversation

fateshelled
Copy link
Contributor

Changes

  • Speed up blobFromImage
    • Inference speed improved from about 5ms/frame to 3.5ms/frame with TensorRT tiny_model.
  • update for ONNXRuntime 1.19.0
    • fix header include.
  • Replace TensorRT enqueueV3 to executeV2
    • error occurs with enqueueV3 if release build, so replace to executeV2.
  • fix CMakeLists.txt
    • fix an issue cannot build with tflite
    • fix an issue where multiple backends could not be turned on at the same time.
  • fix launch files
    • fix launch argument name
  • update README.md
  • and some fix.

Test Environments

  • Ubuntu22.04
  • CPU: Intel Core i5 11500
  • GPU: RTX3060
  • CUDA 12.4
  • TensorRT 10.1
  • ONNXRuntime 1.19.0
  • tflite v2.17.0

Ar-Ray-code
Ar-Ray-code approved these changes Sep 16, 2024
View reviewed changes
Copy link
Owner

@Ar-Ray-code Ar-Ray-code left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

All cases have been checked and passed!
Thanks a lot 🐢 !

@Ar-Ray-code Ar-Ray-code merged commit 612f82f into Ar-Ray-code:iron Sep 16, 2024
6 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@Ar-Ray-code Ar-Ray-code Ar-Ray-code approved these changes

Assignees
No one assigned
Labels
TESTING
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
2 participants
@fateshelled @Ar-Ray-code