error of missing key 'anchors' in hyp.scratch.yaml

[sunsdy2018]

hi, thank you for providing yolo v5. I think I found a bug. The code branch was downloaded yesterday.
I run the code just as the link '#12' with the same data coco128.zip by using VS Code with Pytorch 1.7.1 and Conda and python 3.8.6 in Windows 10 professional.

1. At line 20 in file hyp.scratch.yaml , it says '# anchors: 3 # anchors per output layer (0 to ignore)', which means the arg 'anchors' is commented and doesn't exist.
2. However, in train.py, at line 541 "'anchors': (2, 2.0, 10.0), # anchors per output grid (0 to ignore)" , the key 'anchors' is not commented, and at line 591-593, there are 'hyp[k] = max(hyp[k], v[1])', 'hyp[k] = min(hyp[k], v[2])' and 'hyp[k] = round(hyp[k], 5)', hyp[k] is used in the right side.
3. When running at line 191, error of missing key 'anchors' will occur.

[github-actions]

Hello @sunsdy2018, 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://www.ultralytics.com or email Glenn Jocher at glenn.jocher@ultralytics.com.

Requirements

Python 3.8 or later with all requirements.txt dependencies installed, including torch>=1.7. To install run:

$ pip install -r requirements.txt

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 Notebook with free GPU:
• Kaggle Notebook with free GPU: https://www.kaggle.com/ultralytics/yolov5
• Google Cloud Deep Learning VM. See GCP Quickstart Guide
• Docker Image https://hub.docker.com/r/ultralytics/yolov5. 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), testing (test.py), inference (detect.py) and export (export.py) on MacOS, Windows, and Ubuntu every 24 hours and on every commit.

[glenn-jocher]

@sunsdy2018 sorry for the confusion. The anchors can be specified various ways:

• in model.yaml as a list of anchors (i.e. as in yolov5s.yaml)
• in model.yaml as an anchor count per output layer (i.e. anchors: 3)
• in hyp.yaml as an anchor count per output layer (i.e. anchors: 3)

hyp.yaml anchor requests override model.yaml.

[github-actions]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.