loss function like has a bug

[rrjia]

🐛 Bug

A clear and concise description of what the bug is.

utils/loss.py line108

gain[2:6] = torch.tensor(p[i].shape)[[3, 2, 3, 2]]  # xyxy gain

for yolov5s.yaml
if i=1
p[i].shape = 3,80,80,85
gain[2:6] = [85, 80, 85, 80]

gain[2:6] = torch.tensor(p[i].shape)[[2, 1, 2, 1]]  # xyxy gain

it shoud be [80, 80, 80, 80]

[github-actions]

👋 Hello @rrjia, 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.

[rrjia]

[glenn-jocher]

@rrjia hi thanks for the bug report and the screenshot! We have updated the loss function recently, creating a new ComputeLoss() class to replace it. Can you verify that you see the same bug in the most recent code please?

You can update your code with git pull, or alternatively you can git clone https://github.com/ultralytics/yolov5 again.

[glenn-jocher]

@rrjia if I try to reproduce by training with default settings and placing a breakpoint here, everything looks correct:

[rrjia]

sorry it's my fault, the model get A list of length 3, i use y[0] input compute_loss function, thank you.