cjacoby / ml-experiment

Python framework for managing and executing Machine Learning Experiments.

experiment

Module for managing machine learning experiments.

Modules

experiment (in init) - parent class which handles executing the segments of the experiment based on the input parameters.

path_mgr - things related to managing the experiment paths

feature_extractor - utility for managing the feature extraction process. No feature extraction should actually be implemented here; this utility essentially just executes the feature extraction given the input data.

sample_factory - module for managing generation of sample data. Handles train/test splits, cross validation, "training mode", "validation mode", "test mode"

model - module for executing model code, and getting result outputs.