Introduction

This repo is regarding time series in LLM.

We will begin by experimenting with the following three approaches using numeric and univariate time series data.

• prompts only
• function calling
• code generation

Setup

1. Install required packages: pip install -e .
2. [Optional] Install dev packages (pre-commit hook and ruff): pip install -e .[dev]

Folder structure

├── data
├── docs # for all documents like markdown files, experimentation results
├── notebooks # for jupyter notebooks
├── src # for .py code
├── pyproject.toml # for required python packages and tool settings
├── .pre-commit-config.yaml # configuration for pre-commit hook
├── .gitignore
└── etc.