Top 20 Open‑Source Python Machine‑Learning Projects on GitHub

Open source drives rapid innovation in artificial intelligence. Below is a curated list of the 20 most active Python machine‑learning projects on GitHub, each with a brief description of its main features and a link to the repository.

Scikit‑learn : A comprehensive library built on SciPy offering classification, regression, clustering (including SVM, logistic regression, Naïve Bayes, random forest, gradient boosting, DBSCAN) and integration with NumPy and SciPy. https://github.com/scikit-learn/scikit-learn Pylearn2 : A Theano‑based library that simplifies machine‑learning research. https://github.com/lisa-lab/pylearn2 NuPIC : Implements hierarchical temporal memory (HTM) algorithms for anomaly detection and streaming data prediction. https://github.com/numenta/nupic Nilearn : Facilitates fast statistical learning on neuro‑imaging data by leveraging scikit‑learn tools for prediction, classification, decoding, and connectivity analysis. https://github.com/nilearn/nilearn PyBrain : Provides flexible reinforcement‑learning, AI, and neural‑network algorithms, aiming for easy use and extensive testing environments. https://github.com/pybrain/pybrain Pattern : A web‑mining toolkit offering data‑mining, NLP, network analysis, and machine‑learning utilities such as vector‑space models, clustering, SVM, perceptron, and K‑NN classification. https://github.com/clips/pattern Fuel : Supplies ready‑to‑use datasets (e.g., MNIST, CIFAR‑10, Google’s One Billion Words) for training machine‑learning models. http://www.github.com/mila-udem/fuel Bob : A free signal‑processing and machine‑learning toolbox written in Python and C++, covering image, audio, video processing, and pattern recognition. www.github.com/idiap/bob Skdata : Offers a collection of datasets for machine‑learning and statistics, useful for toy problems, computer‑vision, and natural‑language tasks. www.github.com/jaberg/skdata MILK : Implements supervised classifiers (SVM, K‑NN, random forest, decision trees) and feature selection, supporting hybrid pipelines such as unsupervised learning and K‑means clustering. www.github.com/luispedro/milk IEPY : An open‑source information‑extraction framework focused on relation extraction for large datasets. www.github.com/machinalis/iepy Quepy : Transforms natural‑language questions into database queries (supports SPARQL and MQL) via a Python framework, enabling NL‑driven database access. www.github.com/machinalis/quepy Hebel : A deep‑learning library for neural networks that uses PyCUDA for GPU acceleration, offering various activation functions and model utilities. www.github.com/hannes-brt/hebel mlxtend : Provides a collection of useful extensions for everyday data‑science tasks. www.github.com/rasbt/mlxtend nolearn : Supplies a suite of utilities that complement scikit‑learn for machine‑learning workflows. www.github.com/dnouri/nolearn Ramp : A lightweight, pandas‑based framework that accelerates prototype development by offering a declarative syntax on top of existing tools like scikit‑learn and rpy2. www.github.com/kvh/ramp Feature Forge : A set of tools compatible with scikit‑learn’s API for creating and testing machine‑learning features. www.github.com/machinalis/featureforge REP : Provides a unified classifier wrapper supporting TMVA, scikit‑learn, XGBoost, uBoost, and enables parallel training and interactive visualisation. www.github.com/yandex/rep Python Machine Learning Samples : A collection of simple software samples built with Amazon’s machine‑learning services. www.github.com/awslabs/machine-learning-samples Python‑ELM : Implements Extreme Learning Machine algorithms on top of scikit‑learn.

www.github.com/dclambert/Python-ELM