EB 0.1.4

Ensemble Builder

EnsembleBuilder uses compound activity data to train structure based ensembles to prospectively classify active and inactive compounds.

• Free software: BSD license

• Documentation: https://EB.readthedocs.org.

Features

• Works with Python 2.6, 2.7, 3.3, 3.4

Installation

Installation prerequisites

• NumPy: www.numpy.org

• SciPy: www.scipy.org

Install EnsembleBuilder with pip

pip install EB

Install EnsembleBuilder by cloning the GitHub repo

git clone https://github.com/rvswift/EB
cd EB
make install

Usage

To run, type ensemblebuilder on the command line

usage:        ensemblebuilder <mode> <args>

              ensemblebuilder <mode> to show help for that mode.

modes:
  exhaustive     Determine the best performer by considering all possible ensembles, O(2^N).
  fastheuristic  Determine the best ensemble using a O(N) heuristic.
  slowheuristic  Determine the best ensemble using an O(N^2) heuristic.
  postanalysis   Plot and analyze the performance of one or more ensembles.
  splitter       Split csv input into training and test sets.

History

0.1.0 (ls)

• First release on PyPI.