github-actions[bot]
commented
2 years ago 👋 Hello @yangtiming, 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 Glenn Jocher at glenn.jocher@ultralytics.com.
Requirements
Python>=3.6.0 with all requirements.txt installed including PyTorch>=1.7. To get started:
$ git clone https://github.com/ultralytics/yolov5
$ cd yolov5
$ pip install -r requirements.txtEnvironments
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
❔Question
Background: I take part in a Underwater Object Detection competition ，there are the number of 8200 pictures in training dataset , 350 in val dataset. I use yolov5x6. GPU : 2080ti.
when training:
python train.py --weights weights/yolov5x6.pt --cfg models/hub/yolov5x6.yaml --data data/underwater.yaml --hyp data/hyps/hyp.scratch-p6.yaml --epochs 25 --batch-size 2 --img 1280
There are some questions about parameter adjustment of YOLOV5：
Yolov5x6 is the largest model in the yolov5, it costs lots of time to train. Can I train in the smaller model, like yolov5x , YOLOv5m etc? And then , adjust parameters in the smaller model. After performing well in the smaller model, and then I change model back to Yolov5x6. Can it perform well in Yolov5x6 ,too?
In the "Tips for Best Training Results", We need start with 300 epochs. However , it costs too much time for me to start with 300 epochs to judge if some of the parameters are helpful or not. So, what is the minimum epoch to determine whether the parameter Settings are helpful or not？
In the file named "data/hyps/hyp.scratch-p6.yaml", there are lots of hyperparameter,like hsv_(h,s,v) , mosaic etc. Taking “ mosaic: 1.0” as an example, I don't think "mosaic: 1.0" is good for mAP in my task. So in my view, I want to setting it to "mosaic: 0.0". If I set it to "mosaic: 0.0", will it do harm to mAP? (I mean, whether the hyperparameter “mosaic: 1.0” is an optimization parameter specific to the YOLOv5-P6, if I adjust it, then the mAP score will go down？ )