https://github.com/y-t-g/coffee-bean-defect-anomalib

Coffee bean defect detection using Anomalib
https://github.com/y-t-g/coffee-bean-defect-anomalib

Last synced: 3 months ago
JSON representation

Coffee bean defect detection using Anomalib

• Host: GitHub
• URL: https://github.com/y-t-g/coffee-bean-defect-anomalib
• Owner: Y-T-G
• License: mit
• Created: 2024-01-06T18:44:15.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2024-01-07T10:12:28.000Z (over 1 year ago)
• Last Synced: 2025-02-08T21:25:16.704Z (4 months ago)
• Language: Jupyter Notebook
• Size: 399 KB
• Stars: 1
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# Coffee Bean Defect Detection Using Anomalib

![Prediction](assets/output.png)

## Training

1. Install requirements:

    ```bash

    pip install -r requirements.txt

    ```

2. Set the API key for Roboflow in `train.py`:

`rf = Roboflow(api_key="API")`

3. Run training:

    ```bash

    python train.py

    ```

4. Perform inference using inferencer:

    ```bash

    python gradio_inferencer.py --weights results/custom/run/weights/torch/model.pt

    ```

## Acknowledgement

Dataset: [USK-Coffee](https://universe.roboflow.com/yolo-annotated-dataset/usk-coffe)