In 'evolve' mode, If the original hyp is 0, It will never update

[GuozhenLee]

            for i, k in enumerate(hyp.keys()):  # plt.hist(v.ravel(), 300)
                hyp[k] = float(x[i + 7] * v[i])  # mutate

Because of this line, if x[ele] == 0，new hyp[ele] will never > 0

I don't whether this feature is what you designed or just a bug.

[github-actions]

👋 Hello @GuozhenLee, 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]

@GuozhenLee yes that's a good observation! In the hyp evolution tutorial you can see this in some hyps, such as iou_t and fl_gamma. I suppose it's a desired trait in some cases (i.e. for COCO we probably don't want any up-down flips, so leaving this value at 0.0 would make sense), though the hyp constraints (i.e. limits) are also settable directly in train.py meta dictionary here:

yolov5/train.py

Lines 526 to 529 in 73a0669

	# Hyperparameter evolution metadata (mutation scale 0-1, lower_limit, upper_limit)
	meta = {'lr0': (1, 1e-5, 1e-1), # initial learning rate (SGD=1E-2, Adam=1E-3)
	'lrf': (1, 0.01, 1.0), # final OneCycleLR learning rate (lr0 * lrf)
	'momentum': (0.3, 0.6, 0.98), # SGD momentum/Adam beta1

[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.