Skip to content

Latest commit

 

History

History
37 lines (24 loc) · 867 Bytes

File metadata and controls

37 lines (24 loc) · 867 Bytes

Image classification

The purpose of this project to used CNN network to solve a classification problem. The dedicated application is identifying plant disease from leaf image.

Usage

For cloud trainning, this project include Colab.ipynb file which is used to train the model with Colab.

For normal tranning, this project need python tensorflow or tensorflow-gpu to be pre-installed:

  • Install cuda toolkit 10.0
  • Install cuda cudnn7
  • Install tensorfow-gpu

You need to retrain the model yourself.

To test the model:

python3 gui.py

Result:

[test loss, test acc] = [0.38438466123360043, 0.90295553]

Authors

  • Ngo Van Tuan - Initial work - tuannv

License

This project is licensed under the MIT License.