mlens

About mlens

mlens is ML-Ensemble, a Python library for high performance ensemble learning. It combines a Scikit-learn high-level API with a low-level computational graph framework to build memory efficient, maximally parallelized ensemble networks. It is available on PyPI and installs with pip install mlens.

Pricing is not clearly stated.

Key features

• Train ensembles by adding layers and a meta estimator, then calling fit
• Remove a layer or replace layer parameters in an existing ensemble
• Pass a mapping to add different preprocessing pipelines for estimator subsets
• Test several estimators and preprocessing pipelines in one run with Evaluator
• Check fit times, prediction times and cv-scores in tabular summaries
• Fall back on memory mapped multiprocessing for process-based concurrency

mlens at a glance