KeyError: 'anchors' when training with --evolve

[sinhau]

🐛 Bug

When I run training with the --evolve parameter, I get a KeyError: 'anchors' error.

Digging into the train.py script, it looks like the meta variable defined in line 536 contains the anchor param but that param is commented out in the default hyper parameter file located at data/hyp.scratch.yaml. Is this intentional or a bug?

To Reproduce (REQUIRED)

Input:

python train.py --epochs 10 --evolve

Output:

github: up to date with https://github.com/ultralytics/yolov5 ✅
YOLOv5 🚀 v4.0-138-ged2c742 torch 1.7.1 CUDA:0 (Tesla V100-SXM2-16GB, 16160.5MB)
Namespace(adam=False, batch_size=16, bucket='', cache_images=False, cfg='', data='data/coco128.yaml', device='', entity=None, epochs=10, evolve=True, exist_ok=False, global_rank=-1, hyp='data/hyp.scratch.yaml', image_weights=False, img_size=[640,
 640], linear_lr=False, local_rank=-1, log_artifacts=False, log_imgs=16, multi_scale=False, name='evolve', noautoanchor=False, nosave=False, notest=False, project='runs/train', quad=False, rect=False, resume=False, save_dir='runs/train/evolve', s
ingle_cls=False, sync_bn=False, total_batch_size=16, weights='yolov5s.pt', workers=8, world_size=1)
Traceback (most recent call last):
  File "train.py", line 600, in <module>
    hyp[k] = max(hyp[k], v[1])  # lower limit
KeyError: 'anchors'

Environment

If applicable, add screenshots to help explain your problem.

• OS: Debian 10
• GPU: Tesla V100

[github-actions]

👋 Hello @sinhau, 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 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), testing (test.py), inference (detect.py) and export (export.py) on MacOS, Windows, and Ubuntu every 24 hours and on every commit.

[glenn-jocher]

@sinhau yes you are right! hyp anchors will override yaml anchors if present (uncommented), but for hyp evolution this line should be uncommented, and then your anchor anchor count will evolve along with the other hyps.

I know this workflow isn't the best, I'm not quite sure how to fix this, but I'm open to ideas.

yolov5/data/hyp.scratch.yaml

Line 20 in ed2c742

# anchors: 3 # anchors per output layer (0 to ignore)

[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.

[glenn-jocher]

@sinhau good news 😃! Your original issue may now be fixed ✅ in PR #3887. This PR adds hyp['anchors']=3 if the anchors key is missing/commented in your hyperparameter file, fixing a long-running problem when starting hyp evolution using the default command:

python train.py --evolve

To receive this update:

• Git – git pull from within your yolov5/ directory or git clone https://github.com/ultralytics/yolov5 again
• PyTorch Hub – Force-reload with model = torch.hub.load('ultralytics/yolov5', 'yolov5s', force_reload=True)
• Notebooks – View updated notebooks
• Docker – sudo docker pull ultralytics/yolov5:latest to update your image

Thank you for spotting this issue and informing us of the problem. Please let us know if this update resolves the issue for you, and feel free to inform us of any other issues you discover or feature requests that come to mind. Happy trainings with YOLOv5 🚀!