how can I just use yolov5s.pt

DarkLeaves commented 2 months ago

Search before asking

[X] I have searched the YOLOv5 issues and discussions and found no similar questions.

Question

I want to use yolov5s.pt to do some experiments. but when I run my code: model = YOLO('yolov5s.pt')

the ultralytics will Replace 'model=pics/yolov5s.pt' with new 'model=pics/yolov5su.pt'. how to cancel this replacement?

Additional

No response

UltralyticsAssistant commented 2 months ago

👋 Hello @DarkLeaves, thank you for reaching out about YOLOv5 🚀!

For using yolov5s.pt, please ensure you're following the correct setup and process. You can find guidance in our ⭐️ Tutorials to help you get started with different tasks, including the use of YOLOv5 models.

If this is a 🐛 Bug Report, please provide a minimum reproducible example to help us debug it. This will allow our team to better understand the problem.

If this is a ❓ Question related to custom use, please share detailed information, ensuring you're using our Tips for Best Training Results.

Requirements

Please ensure your environment meets these requirements:

Python>=3.8.0 with necessary requirements.txt installed, including PyTorch>=1.8. Start with:

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

Environments

Consider running YOLOv5 in one of our verified environments:

Notebooks:
Google Cloud: See GCP Quickstart Guide
Amazon: See AWS Quickstart Guide
Docker Image: See Docker Quickstart Guide

Status

Check our current status with this badge:

Rest assured, an Ultralytics engineer will assist you soon. Meanwhile, explore using our latest YOLOv8 for cutting-edge results!

Thanks again for your patience and interest 🌟

pderrenger commented 3 weeks ago

@DarkLeaves to prevent the automatic replacement, ensure you are using the correct version of YOLOv5 that supports your intended usage. You can specify the exact model path without triggering updates by using the torch.hub.load method like this:

import torch
model = torch.hub.load('ultralytics/yolov5', 'custom', 'yolov5s.pt')

This should allow you to use yolov5s.pt without any automatic changes.

ultralytics / yolov5