hibana2077/hayabusa

It is a simple automl framework

Hayabusa

Introduction

This is a Lightweight and Fast WEB UI AutoML framework for small dataset (Less than 200MB).

Installation

Install from GitHub

Download the source code from GitHub and install dependencies.

git clone https://github.com/hibana2077/hayabusa.git
cd hayabusa
pip install -r requirements.txt

cd src
streamlit run main.py

Install from Docker

Install Docker.

curl -sSL https://get.docker.com | sh

Build the image.

git clone https://github.com/hibana2077/hayabusa.git
cd hayabusa
docker build -t hayabusa .

Run the container.

docker run -p 8501:80 hayabusa

Usage

Upload Data

Upload your data in csv format.

Data Profiling

Click Data Profiling button to get the data profiling report.

Data Preprocessing (TODO)

Choose the columns you want to preprocess.

Modelling

Choose the columns you want to predict.

Download

You can download the model and the data pipeline in pkl format.

License

MIT

Reference

• Nicholas Renotte
• Streamlit