Megvii-BaseDetection / Megvii-BaseDetection/YOLOX

请问为什么我训练出来的Yolox nano 模型parameter和Gflops与论文中的不匹配

Open
#764 11 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

之前训练yolox_s的时候还好,训练出来的parameter是8.95M,和论文里面的9M正好匹配。可惜觉得速度还是慢了,在cpu下只有7帧的效果,所以想试下最小的nano,可是训练完后发现parameter有2.25M,Gflops2.91。和论文里面的0.91M差距很大,我就把yolox_voc_s里面的width和depth改成了nano的0.33和0.25,然后换了yolox_nano.pth,请问为什么效果不理想

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 by comparing the yolox_voc_s configuration changes for width 0.33 and depth 0.25 with the yolox_nano.pth model used for training. Then inspect how YOLOX reports parameter count and GFLOPs and compare those results with the paper's nano values. Done means explaining the 2.25M parameters and 2.91 GFLOPs, or identifying the configuration or checkpoint mismatch.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, pytorch
Domain
machine-learning
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.