multiclass beta

v1.1.0-beta1

* version for relesae

1.0

We are stoked to release this new version. Lots of code has been re-written, with a much simpler API, better quality code, 10x more test coverage, more database support, and many other great things. Please see the new example files and docs for details, as the API has entirely changed.

0.1.8

[0.1.8] - 2017-02-16

Added

• Added changelog
• Changed all docs to markdown
• added setup.cfg and some functions in setup.py for PyPI
• mkdocs.yml for readthedocs hosting
• Conda environment files

Changed

• Lots of documentation, especially around installation on the various platforms.
• Removed doc templates and css now that docs will live on readthedocs

Fixed

• example jupyter notebook

healthcareai v0.1.7-beta

This release encompasses basic healthcare ML functionality:

• Model comparison between random forest and logistic regression algorithms
• Model deployment to SQL Server, providing top-three most important features
• Imputation (column mean for numeric and column mode for categorical)
• Hyperparameter tuning, using mtry and number of trees for random forest
• Plots
  • ROC
  • Random forest feature ranking
• Model performance evaluated via AU_ROC and AU_PR

First release after setting up the following infrastructure:

• AppveyorCI
• Sphinx
• Nose unit testing
• Docker