Megvii-BaseDetection / Megvii-BaseDetection/YOLOX

YOLOX Inference Speed on RTX 3060 Ti

Open
#730 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
10.6k
Forks
2.5k
PR merge metrics
No merged PRs in 30d

Description

Has anyone tested inferencing with YOLOX on pytorch using a RTX 3060 Ti. I only achieved an inference speed of 0.8-0.9 seconds (around 1 fps) . Just wanna double confirm cause I expeceted at least 0.1 seconds (10 fps). Already checked that pytorch is installed correctly and CUDA driver is enabled.

Further information:
CUDA 11.2
Pytorch 1.91

Running inference with default settings shown in demo:
yolox_s - 0.2243s
yolox_m - 0.51s
yolox_l - 0.8635

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start with the YOLOX demo's default inference settings and compare the reported timings for yolox-s, yolox-m, and yolox-l on the RTX 3060 Ti. Check whether the CUDA 11.2 and PyTorch 1.91 environment reproduces the observed speeds and identify why they differ from the expected 0.1 seconds; done means an explained discrepancy or a confirmed performance issue.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, pytorch
Domain
machine-learning, performance
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
20/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.