What is the IMPROVE project?
============================
IMPROVE aims to establish methodology to systematically and rigorously compare supervised learning models. To this end, IMPROVE has four components:

1) IMPROVE Framework: standardized scaffolding for machine learning models, written in Python

2) IMPROVE :doc:`Evaluation Schemes <BenchmarkEvaluationSchemes>`: workflows to evaluated and compare models

3) IMPROVE Benchmark Datasets: a collection of standardized data for each application to allow for benchmarking of models

4) IMPROVE Community Curated Models: models for each application from the scientific communitity that have been curated into the IMPROVE framework


IMPROVE can be applied to a variety of supervised learning models. We have begun with the :doc:`Drug Response Prediction <app_drp_intro>` problem, standardizing :doc:`community models <Models>` that predict pan-cancer, pan-drug, single drug response for precision oncology with a standardized :doc:`benchmark dataset <CSA_Benchmark_Datasets>`.




For more information
----------------------

.. toctree::
   :titlesonly:

   Background
   Future Directions <FutureDirections>
   Acknowledgments <acknowledgment>


Access the code base https://github.com/JDACS4C-IMPROVE/

References
---------------------------------
`1. <https://www.frontiersin.org/articles/10.3389/fmed.2023.1086097/full>`_ A. Partin et al. "Deep learning methods for drug response prediction in cancer: Predominant and emerging trends", Frontiers in Medicine, Section Prediction Oncology, 2023