confidentz-model

This is a repository for dataset and machine learning model versioning of ConfiDentz Project. Confidentz model is a part of our ConfiDentz project. This model was trained using 500 images of healthy and caries tooth images. This document will guide you to train your own model.

Pre-requisites

• VSCode / any text editor
• Git
• Conda
• DVC (Data Version Control)

Installation

Windows

1. Clone the repository with:

   git clone https://github.com/Confidentz-C23-PS334/confidentz-model

2. Enter the directory

3. Get the dataset by running:

   dvc get https://github.com/Confidentz-C23-PS334/confidentz-model dataset.zip

4. Create conda environment by running:

   conda env create -n ENVNAME --file environment.yml

   Change ENVNAME to your preferred environment name

5. Open train.ipynb with VSCode or other to start training model

How to train your model

1. Connect to Jupyter notebook server and use the environment name you have defined as the kernel
2. Run through every cells in train.ipynb
3. Export the model as .h5 format

Machine Learning Team

Name	StudentID	University	links
Alvian Rahmadani Saputra	M282DKX3756	Universitas Negeri Malang	Github
Fahmi Zulkarnain Habib	M282DSX0429	Universitas Negeri Malang	Github

Links

Link to dataset: Google Drive