Can I automatically determine the depthmutiple [dm] and widthmutiple [wm] to get a model for my dataset（by accuracy and speed balance）？

AllenZYJ commented 2 years ago

Search before asking

[X] I have searched the YOLOv5 issues and discussions and found no similar questions.

Question

hi ,thankyou for your helpful work,it is great~ but i have some questions for ask...like title， I found the larger model parameter [dm] and [wm] is not always better , often the model is small, but the ap dont have big disparity between small model and larger model, I don't know if it's a coincidence. and ,the dm and wm determine the model size and speed, so, when i can train the model from 0.1 to 1.0, i can determine the model parameter to balance speed and map. I don't know if I'm right or not , please give me some advice, thank you very much~.

Additional

No response

github-actions[bot] commented 2 years ago

👋 Hello @AllenZYJ, thank you for your interest in YOLOv5 🚀! Please visit our ⭐️ Tutorials to get started, where you can find quickstart guides for simple tasks like Custom Data Training all the way to advanced concepts like Hyperparameter Evolution.

If this is a 🐛 Bug Report, please provide screenshots and minimum viable code to reproduce your issue, otherwise we can not help you.

If this is a custom training ❓ Question, please provide as much information as possible, including dataset images, training logs, screenshots, and a public link to online W&B logging if available.

For business inquiries or professional support requests please visit https://ultralytics.com or email support@ultralytics.com.

Requirements

Python>=3.7.0 with all requirements.txt installed including PyTorch>=1.7. To get started:

git clone https://github.com/ultralytics/yolov5  # clone
cd yolov5
pip install -r requirements.txt  # install

Environments

YOLOv5 may be run in any of the following up-to-date verified environments (with all dependencies including CUDA/CUDNN, Python and PyTorch preinstalled):

Google Colab and Kaggle notebooks with free GPU:
Google Cloud Deep Learning VM. See GCP Quickstart Guide
Amazon Deep Learning AMI. See AWS Quickstart Guide
Docker Image. See Docker Quickstart Guide

Status

If this badge is green, all YOLOv5 GitHub Actions Continuous Integration (CI) tests are currently passing. CI tests verify correct operation of YOLOv5 training (train.py), validation (val.py), inference (detect.py) and export (export.py) on macOS, Windows, and Ubuntu every 24 hours and on every commit.

glenn-jocher commented 2 years ago

@AllenZYJ yes modifying the depth and width multiples is the easiest way to scale YOLOv5 models from nano to xlarge. This technique was first popularized by EfficientDet, and this is what we do for the official models.

You can use your own values to further customize a model to your particular use case.

AllenZYJ commented 2 years ago

@AllenZYJ yes modifying the depth and width multiples is the easiest way to scale YOLOv5 models from nano to xlarge. This technique was first popularized by EfficientDet, and this is what we do for the official models.

You can use your own values to further customize a model to your particular use case. thank you very much ~ now I try to run a task automatic，design the dm and wm from 0 to 1.5 to scale yolov5 model ，It work

ultralytics / yolov5