How to load a custom model?

[int3r5t377ar]

Search before asking

• I have searched the YOLOv5 issues and discussions and found no similar questions.

Question

I use the code
model = torch.hub.load('ultralytics/yolov5', 'custom', path='C:/Users/Documents/yolov5/runs/train/exp/weights/best.bt', force_reload=True)
to load the custom model I have trained.
The code is found in the internet and it works for the examples I referenced at.
But for me, it keeps generating errors:

PS C:\Users\Documents\yolov5> & C:/Users/anaconda3/python.exe "c:/Users/Documents/yolov5/myfile.py"
Downloading: "https://github.com/ultralytics/yolov5/zipball/master" to C:\Users/.cache\torch\hub\master.zip
YOLOv5  2022-10-13 Python-3.9.13 torch-1.12.1 CUDA:0 (NVIDIA GeForce RTX 3060, 12288MiB)

Traceback (most recent call last):
  File "C:\Users/.cache\torch\hub\ultralytics_yolov5_master\hubconf.py", line 47, in _create
    model = DetectMultiBackend(path, device=device, fuse=autoshape)  # detection model
  File "C:\Users/.cache\torch\hub\ultralytics_yolov5_master\models\common.py", line 331, in __init__
    pt, jit, onnx, xml, engine, coreml, saved_model, pb, tflite, edgetpu, tfjs, paddle, triton = self._model_type(w)
  File "C:\Users/.cache\torch\hub\ultralytics_yolov5_master\models\common.py", line 596, in _model_type
    check_suffix(p, sf)  # checks
  File "C:\Users/.cache\torch\hub\ultralytics_yolov5_master\utils\general.py", line 409, in check_suffix

Traceback (most recent call last):  File "c:\Users\Documents\yolov5\myfile.py", line 11, in <module>
    model = torch.hub.load('ultralytics/yolov5', 'custom', path='C:/Users/Documents/yolov5/runs/train/exp/weights/best.bt', force_reload=True)   
  File "C:\Users\anaconda3\lib\site-packages\torch\hub.py", line 540, in load
    model = _load_local(repo_or_dir, model, *args, **kwargs)
  File "C:\Users\anaconda3\lib\site-packages\torch\hub.py", line 569, in _load_local
    model = entry(*args, **kwargs)
  File "C:\Users/.cache\torch\hub\ultralytics_yolov5_master\hubconf.py", line 81, in custom
    return _create(path, autoshape=autoshape, verbose=_verbose, device=device)
  File "C:\Users/.cache\torch\hub\ultralytics_yolov5_master\hubconf.py", line 76, in _create
    raise Exception(s) from e
Exception: [Errno 2] No such file or directory: 'C:\\Users\\Documents\\yolov5\\runs\\train\\exp\\weights\\best.bt'. Cache may be out of date, try `force_reload=True` or see https://github.com/ultralytics/yolov5/issues/36 for help.

Also, I have no idea why the path becomes 'C:\\Users\\Documents\\yolov5\\runs\\train\\exp\\weights\\best.bt'.

Additional

No response

[github-actions]

👋 Hello @int3r5t377ar, 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):

• 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, validation, inference, export and benchmarks on MacOS, Windows, and Ubuntu every 24 hours and on every commit.

[glenn-jocher]

@int3r5t377ar the error is telling you that there is no file in the location you specified:

Exception: [Errno 2] No such file or directory: 'C:\\Users\\Documents\\yolov5\\runs\\train\\exp\\weights\\[best.bt](http://best.bt/)'

Windows loading of YOLOv5 PyTorch Hub models is run every 24 hours and operates correctly:
https://github.com/ultralytics/yolov5/actions/runs/3238824126/jobs/5307473149

yolov5/.github/workflows/ci-testing.yml

Lines 133 to 135 in 16f87bb

	for path in '$m', '$b':
	model = torch.hub.load('.', 'custom', path=path, source='local')
	print(model('data/images/bus.jpg'))

[int3r5t377ar]

@glenn-jocher

@int3r5t377ar the error is telling you that there is no file in the location you specified:

Exception: [Errno 2] No such file or directory: 'C:\\Users\\Documents\\yolov5\\runs\\train\\exp\\weights\\[best.bt](http://best.bt/)'

What should I do to use the best.bt file? The file is in the location there.

[glenn-jocher]

Looks like user error due a mispelling. You've got a *.bt suffix which you probably meant to write as *.pt

[int3r5t377ar]

Looks like user error due a mispelling. You've got a *.bt suffix which you probably meant to write as *.pt

Oh thank you so much. I overlooked such a simple mistake over long time. And it works fine now. I am gonna close this question.

[bukobeer01]

@int3r5t377ar the error is telling you that there is no file in the location you specified:

Exception: [Errno 2] No such file or directory: 'C:\\Users\\Documents\\yolov5\\runs\\train\\exp\\weights\\[best.bt](http://best.bt/)'

Windows loading of YOLOv5 PyTorch Hub models is run every 24 hours and operates correctly: https://github.com/ultralytics/yolov5/actions/runs/3238824126/jobs/5307473149

yolov5/.github/workflows/ci-testing.yml

Lines 133 to 135 in 16f87bb

	for path in '$m', '$b':
	model = torch.hub.load('.', 'custom', path=path, source='local')
	print(model('data/images/bus.jpg'))

---

can you help me I'm using torch 1.11.0+cu113

import torch
path = 'yolov5s.pt'
model = torch.hub.load('.', 'custom', path=path, source='local')
print(model('images\1.jpg'))

I got

File ".\load3.py", line 3, in
model = torch.hub.load('.', 'custom', path=path, source='local')
File "C:\Users\bukob\AppData\Local\Programs\Python\Python38\lib\site-packages\torch\hub.py", line 404, in load
model = _load_local(repo_or_dir, model, *args, **kwargs)
File "C:\Users\bukob\AppData\Local\Programs\Python\Python38\lib\site-packages\torch\hub.py", line 430, in _load_local
hub_module = _import_module(MODULE_HUBCONF, hubconf_path)
File "C:\Users\bukob\AppData\Local\Programs\Python\Python38\lib\site-packages\torch\hub.py", line 76, in _import_module
spec.loader.exec_module(module)
File "", line 779, in exec_module
File "", line 915, in get_code
File "", line 972, in get_data
FileNotFoundError: [Errno 2] No such file or directory: '.\hubconf.py'

[Sha312003]

In the first argument of torch.hub.load write the path of yolo directory.
For example, instead of model = torch.hub.load('.', 'custom', path=path, source='local') write
model = torch.hub.load('/content/yolov5', 'custom', path=path, source='local').
Hope this helps

[glenn-jocher]

@Sha312003 your error message suggests that your current directory does not contain a file called hubconf.py. You can specify the path to the directory that contains yolov5s.pt file using torch.hub.load() function like this:

import torch

path = 'yolov5s.pt'
model = torch.hub.load('ultralytics/yolov5', 'custom', path=path, source='local') 
print(model(torch.randn(1,3,640,640))))

This code loads the yolov5s.pt file from the Ultralytics YOLOv5 hub repository and returns a detection model that can be used for inference on images. You can then call the model with an image tensor to get the predicted detections.

[dark292929]

tengo un error

model = torch.hub.load('ultralytics/yolov5','custom',
path = 'D:/python/pythonProject/model/carros.pt', force_reload=True)

para subir mi modelo personalizable no reconoce,no se si alguien le paso

[glenn-jocher]

Hola! Parece que estás enfrentado un problema al cargar tu modelo personalizado. Asegúrate que la ruta al archivo .pt es correcta y accesible.

Tu código parece estar mayormente correcto, solo verifica la ruta. Aquí tienes un ejemplo ligeramente ajustado para asegurarnos que se usa correctamente:

import torch
model = torch.hub.load('ultralytics/yolov5', 'custom', path='D:/python/pythonProject/model/carros.pt', force_reload=True)

Si el modelo sigue sin reconocer, verifica que 'carros.pt' exista en la ruta especificada y que no haya errores en el nombre del archivo. ¡Espero que esto te ayude! 😊