Monkey patching cv2 methods crashes own scripts

[janzumsteg]

Search before asking

• I have searched the YOLOv5 issues and found no similar bug report.

YOLOv5 Component

PyTorch Hub

Bug

The current master branch contains monkey patches of cv2.imread, cv2.imwrite and cv2.imshow (see

yolov5/utils/general.py

Line 944 in ea72b84

cv2.imread, cv2.imwrite, cv2.imshow = imread, imwrite, imshow # redefine

). When importing yolov5 (e.g. via torch.hub.load), the respective methods in my own scripts will also be overwritten without any hints. Moreover, the signature of cv2.imread changed with these monkey patches resulting in TypeError: imread() takes 1 positional argument but 2 were given.

Related PRs:

• Fix cv2.imwrite on non-ASCII paths #7139
• np.fromfile() Chinese image paths fix #6979

Environment

No response

Minimal Reproducible Example

import cv2
import torch

cv2.imread("path/to/image.png", cv2.IMREAD_COLOR)  # works
torch.hub.load("ultralytics/yolov5", "yolov5m", force_reload=True)
cv2.imread("path/to/image.png", cv2.IMREAD_COLOR)  # fails

Additional

No response

Are you willing to submit a PR?

• Yes I'd like to help by submitting a PR!

[github-actions]

👋 Hello @janzumsteg, 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 [email protected].

Requirements

Python>=3.7.0 with all requirements.txt installed including PyTorch>=1.7. To get started:

git clone https://github.com/ultralytics/yolov5  # clone
cd yolov5
pip install -r requirements.txt  # install

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), validation (val.py), inference (detect.py) and export (export.py) on MacOS, Windows, and Ubuntu every 24 hours and on every commit.

[glenn-jocher]

@janzumsteg good news 😃! Your original issue may now be fixed ✅ in PR #7287. This PR updates the patched function to accept a second argument just like the original function, and the code to reproduce the error now runs without issue:

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 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 🚀!

[grzegorzk]

Hi @glenn-jocher, I have been playing with yolov5==7.0.11 and found that monkey-patched imread has different signature than the one provided by opencv. Your argument is called path while in opencv it's called filename. This is resulting in my other scripts that are executed after loading yolov5 to fail, even though they are working fine if I don't use yolov5.

>>> import cv2
>>> help(cv2.imread)
Help on built-in function imread:

imread(...)
    imread(filename[, flags]) -> retval
(...)

>>> import yolov5
>>> help(cv2.imread)
Help on function imread in module yolov5.utils.general:

imread(path, flags=1)

I proposed fix here

I would personally prefer if these methods are not monkey-patched each time I import yolov5 though, I can prepare a PR if this is something you would consider.

[glenn-jocher]

Hi @grzegorzk 👋, it's great to hear your feedback on the monkey-patched imread method. Your proposed fix in PR #11209 looks like a step in the right direction. I'll make sure to discuss this with the team to see if we can merge your changes.

Regarding the monkey-patching, I totally understand your concern. We are currently evaluating different options to reduce the impact of monkey-patching on user-defined scripts. Your willingness to submit a PR to help address this is much appreciated! We'd certainly be open to reviewing your solution. Thank you for your contribution and for helping make YOLOv5 even better 🙌!