predictionball

A library for pulling in and normalising prediction market data.

Dependencies 🌐

Python 3.11.6:

• pandas
• py-clob-client
• python-dotenv
• pydantic
• requests
• pyarrow
• sentence_transformers
• scikit-learn

Raison D'être 💭

predictionball aims to be a library that pulls in both historical and live prediction market data. It currently handles the following markets:

• Polymarket
• Kalshi
• PredictIt

Architecture 📐

predictionball is a functional library that does the following steps:

1. Pulls down different prediction markets.
2. Cluster the markets by event type.

Installation 📥

This is a python package hosted on pypi, so to install simply run the following command:

pip install predictionball

or install using this local repository:

python setup.py install --old-and-unmanageable

Usage example 👀

The only way to access the data from predictionball is to use it as a python library.

Python

To pull the models containing all the information, the following example can be used:

from predictionball.pull import pull

models = pull()

This results in a dataframe where each game is represented by all its features.

License 📝

The project is available under the MIT License.