Master Thesis Project: Time Series Forecasting of KANs

This repository contains the full codebase, experiments, and datasets for my master's thesis research project focused on time series forecasting using machine learning models - with the focus on KANs.

Repository Structure

• src/ — Main source code (data processing, training, evaluation)
• models/ — Saved models and checkpoints (~1.2 GB)
• data/ — Preprocessed datasets used in experiments (~1.2 GB)
• mlruns/ — MLflow experiment logs and artifacts (~8.5 GB)
• results/ — Plots, metrics, and output files
• requirements.txt — Python dependencies
• requirements_colab.txt — Minimal dependencies for running in Google Colab

How to Run

1. Install dependencies:

   pip install -r requirements.txt