Welcome to Clairvoyant’s documentation!

Clairvoyant is a machine learning library designed to identify and monitor social/historical cues for short term stock movement.

Many users will be most interested in Clair– the primary machine learning class. The library provides additional classes that aid in training, backtesting, and data munging.

The Backtest class is useful for rapidly testing and calibrating parameters for maximum signal accuracy. The Portfolio class adds trading logic, which helps users ascertain the impact of various trading decisions in response to model predictions. Clair needs to understand your data, so we provide a specific class called History whose primary purpose is to map whatever columns you may have defined into common column names that Clair can identify.

Documentation

Indices and tables