Welcome to IMPROVE’s documentation!
Version: latest
The IMPROVE (Innovative Methodologies and New Data for Predictive Oncology Model Evaluation) project is dedicated to standardizing the execution of prediction models, enabling rigorous benchmarking and fair comparison. The IMPROVE software stack offers Python-based tools for standardizing model code structure, along with workflows for large-scale model evaluation and comparison.
We have been using anti-cancer drug response prediction (DRP) as a use case to drive the development of the IMPROVE software. As part of this effort, we have curated DRP community models to conform to a standardized structure and compiled a benchmark dataset for monotherapy DRP.
Do you have a model you would like to compare with state-of-the-art models? Try IMPROVE!
Contents:
Access the code base https://github.com/JDACS4C-IMPROVE/