thesis-ml

ML Development and Experiment for Thesis.

Features

• Dataset with annotation result from Penyuluh Bahasa Bali: dataset
• Numpy-based implementation: notebook.py
  • TF-IDF
  • Multinomial Naive Bayes
  • Label Binarizer
  • Chi-square
  • SMOTE
  • split data, accuracy score, cross validation
  • Pipeline
• ML development and experiment: model.ipynb

Prerequisites

This project mainly using uv and Python, so you need:

• Python 3.10 or higher
• uv

Installation

• Clone the repository:

git clone https://github.com/putuwaw/thesis-ml.git

• Install dependencies:

uv sync --dev

• Start building your own model!