After converting from .pt to .torchscript, bounding box is off #6947

Activity

1. willsonyee

   added

   questionFurther information is requestedFurther information is requested

   on Mar 11, 2022
2. github-actions commented on Mar 11, 2022

   

   github-actionsBot

   on Mar 11, 2022

   Last edited by UltralyticsAssistant

   Contributor

   More actions

   👋 Hello @willsonyee, 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 support@ultralytics.com.

   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.

   👎React with 👎1Reacted by JeanPSF
3. glenn-jocher commented on Mar 11, 2022

   

   glenn-jocher

   on Mar 11, 2022

   Last edited by glenn-jocher

   Member

   More actions

   @willsonyee 👋 hi, thanks for letting us know about this possible problem with YOLOv5 🚀. Torchscript export and inference works well for me, I'm unable to reproduce any box issues.

   

   We've created a few short guidelines below to help users provide what we need in order to get started investigating a possible problem.

   How to create a Minimal, Reproducible Example

   When asking a question, people will be better able to provide help if you provide code that they can easily understand and use to reproduce the problem. This is referred to by community members as creating a minimum reproducible example. Your code that reproduces the problem should be:

   • ✅ Minimal – Use as little code as possible to produce the problem
   • ✅ Complete – Provide all parts someone else needs to reproduce the problem
   • ✅ Reproducible – Test the code you're about to provide to make sure it reproduces the problem

   For Ultralytics to provide assistance your code should also be:

   • ✅ Current – Verify that your code is up-to-date with GitHub master, and if necessary git pull or git clone a new copy to ensure your problem has not already been solved in master.
   • ✅ Unmodified – Your problem must be reproducible using official YOLOv5 code without changes. Ultralytics does not provide support for custom code ⚠️.

   If you believe your problem meets all the above criteria, please close this issue and raise a new one using the 🐛 Bug Report template with a minimum reproducible example to help us better understand and diagnose your problem.

   Thank you! 😃
4. willsonyee commented on Mar 11, 2022

   

   willsonyee

   on Mar 11, 2022

   Last edited by willsonyee

   Author

   More actions

   @glenn-jocher thanks for your prompt response.

   I am running the code like this:

   import torch
   model = torch.hub.load('/content/drive/MyDrive/yolov5', 'custom', path='/content/drive/MyDrive/yolov5/yolov5s.torchscript', source='local')
   result = model("/content/drive/MyDrive/yolov5/data/images/zidane.jpg")
   result.save() 
   

   The reason I am doing it like this because I am trying to load the model from local and inference without using the detect.py. So I can later on integrate the .torchscript model into my swift IOS app.

   Could you advise me why by running the code I attached will cause the bounding box off ?
5. glenn-jocher commented on Mar 11, 2022

   

   glenn-jocher

   on Mar 11, 2022

   Member

   More actions

   @willsonyee ah yes you are right there is a bug. I'm able to reproduce with this code:

   !python export.py --weights yolov5s.pt --include torchscript
   model = torch.hub.load('ultralytics/yolov5', 'custom', path='yolov5s.torchscript')
   result = model("data/images/zidane.jpg")
   result.save()

   

   I'll take a look!
6. glenn-jocher

   added a commit that references this issue on Mar 11, 2022

   Fix PyTorch Hub export inference shapes

   ...

   6ec0169
7. glenn-jocher

   closed this as completedin #6949on Mar 11, 2022
8. glenn-jocher

   added a commit that references this issue on Mar 11, 2022

   Fix PyTorch Hub export inference shapes (#6949)

   ...

   84efa62
9. bfineran

   added a commit that references this issue on Apr 8, 2022

   Update Neuralmagic-YoloV5 integration to v6.1 (#29)

   ...

   14eb1dc
10. BjarneKuehl

    added a commit that references this issue on Aug 26, 2022

    Fix PyTorch Hub export inference shapes (ultralytics#6949)

    ...

    c6bbbfc
11. SecretStar112

    added a commit that references this issue on May 24, 2023

    Fix PyTorch Hub export inference shapes (#6949)

    ...

    519462f
12. ramdev77

    added a commit that references this issue on Oct 23, 2024

    Fix PyTorch Hub export inference shapes (#6949)

    ...

    13051d9