Integrating MAPIE to have prediction intervals of classifiers and regressors to Galaxy-ML tools
Supervisor: @anuprulez @bgruening
For degree: Bachelor/Project
Status: Open (assigned to ...)
Keywords: MAPIE - Scikit-learn - Classifier - Regressor - Prediction intervals
Global Biological/Research context
Integrating MAPIE (Model Agnostic Prediction Interval Estimation) into Galaxy-ML tools represents a significant advancement in predictive modeling capabilities. By incorporating prediction intervals for classifiers and regressors, Galaxy-ML enhances its predictive accuracy and provides users with a more comprehensive understanding of model uncertainty. MAPIE's model-agnostic approach ensures flexibility across various machine learning algorithms, allowing for robust interval estimation regardless of the specific model employed. This integration empowers users to make more informed decisions by quantifying the range of possible outcomes, thereby improving the reliability and interpretability of predictions generated by Galaxy-ML tools. Ultimately, the inclusion of prediction intervals enhances the utility of Galaxy-ML for several predictive tasks/analyses in Bioinformatics, by offering a better understanding of predictive uncertainty.
Project context
The integration of project is a proof-of-concept to know how MAPIE can be integrated, first to a few ML tools in Galaxy
Apply MAPIE on a few classification and regression datasets to do hands-on using Penn Machine learning Benchmark datasets (https://github.com/EpistasisLab/pmlb).
Integrate MAPIE to a few Galaxy ML tools (classifiers and regressors) as a feature.
Integrating MAPIE to have prediction intervals of classifiers and regressors to Galaxy-ML tools
Supervisor: @anuprulez @bgruening For degree: Bachelor/Project Status: Open (assigned to ...) Keywords: MAPIE - Scikit-learn - Classifier - Regressor - Prediction intervals
Global Biological/Research context
Integrating MAPIE (Model Agnostic Prediction Interval Estimation) into Galaxy-ML tools represents a significant advancement in predictive modeling capabilities. By incorporating prediction intervals for classifiers and regressors, Galaxy-ML enhances its predictive accuracy and provides users with a more comprehensive understanding of model uncertainty. MAPIE's model-agnostic approach ensures flexibility across various machine learning algorithms, allowing for robust interval estimation regardless of the specific model employed. This integration empowers users to make more informed decisions by quantifying the range of possible outcomes, thereby improving the reliability and interpretability of predictions generated by Galaxy-ML tools. Ultimately, the inclusion of prediction intervals enhances the utility of Galaxy-ML for several predictive tasks/analyses in Bioinformatics, by offering a better understanding of predictive uncertainty.
Project context
The integration of project is a proof-of-concept to know how MAPIE can be integrated, first to a few ML tools in Galaxy
Objectives of the project
General objectives of the project
Proposed agenda for the project
Prerequisites
Further reading and useful links