No detection from example images

[ZilongZhang]

Hi I am new to Yolo and I am trying to run the model on my computer, the model does not detect any object from the two example images (i.e. bus and zidane). I used models of different sizes but none of them succeeded. I wonder what went wrong and how to fix it. Here are some details about my environment:

Windows 10
Python 3.8
Torch 1.10.0

Here are what Yolo prints when detecting objects from bus.jpg:

Using cache found in /.cache\torch\hub\ultralytics_yolov5_master
YOLOv5 2021-12-8 torch 1.10.0 CUDA:0 (NVIDIA GeForce GTX 1650, 4096MiB)
Fusing layers...
Model Summary: 367 layers, 46533693 parameters, 0 gradients
Adding AutoShape...
image 1/1: 1080x810 (no detections)
Speed: 17.0ms pre-process, 307.2ms inference, 0.0ms NMS per image at shape (1, 3, 640, 480)

[github-actions]

👋 Hello @ZilongZhang, 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 Glenn Jocher at glenn.jocher@ultralytics.com.

Requirements

Python>=3.6.0 with all requirements.txt installed including PyTorch>=1.7. To get started:

$ git clone https://github.com/ultralytics/yolov5
$ cd yolov5
$ 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 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]

@ZilongZhang this is probably due to CUDA 11 on your system and downgrading to CUDA 10 will fix this.

There is an open issue ongoing with PyTorch and CUDA 11 unfortunately that is affecting multiple users when running AMP or FP16 inference. If this is the cause then detect.py and val.py will work but PyTorch Hub and train.py will not work for you under CUDA 11.

[glenn-jocher]

@ZilongZhang good news 😃! Your original issue may now be fixed ✅ in PR #5926. This PR default PyTorch Hub models to Automatic Mixed Precision (AMP) disabled status for improved compatibility with CUDA 11 under Windows and/or conda environments. AMP inference can now be toggled with the model.amp attribute:

import torch

# Model
model = torch.hub.load('ultralytics/yolov5', 'yolov5s')  # or yolov5m, yolov5l, yolov5x, custom
model.amp = False  # <--- disable AMP inference (solves Windows/conda/CUDA11 issues)
model.amp = True  # <--- enable AMP inference

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 🚀!