InfodeslibΒΆ

Late Fusion based Dynamic Ensemble Learning library.

Infodeslib is open source library focusing the implementation of the state-of-the-art techniques for dynamic classifier and ensemble selection with late fusion setting. This project is under active development. Contributions are welcomed through its GitHub page: https://github.com/adv-panda/infodeslib

_images/logo_small.png

Install

Quick Start

Example

Here we present an example of the KNORA-U techniques using a random forest to generate the pool of classifiers:

from infodeslib.des.knorau import KNORAU

pool_classifiers = [classifier1, ..., classifierN]
# feature_set1 is a list of columns
feature_sets = [feature_set1, ..., feature_setN]

# Initialize the DS model
knorau = KNORAU(pool_classifiers, feature_sets)

# Fit the dynamic selection model
knorau.fit(X_dsel, y_dsel)

# Predict new examples
knorau.predict(X_test, plot=True)

# Check performance (based on accuracy)
knorau.score(X_test, y_test)