node9909 / neuropredict

Easy and comprehensive assessment of predictive power, with support for neuroimaging features

Documentation: https://raamana.github.io/neuropredict/

News

• As of v0.6, neuropredict now supports regression applications i.e. predicting continuous targets (in addition to categorical classes), as well as allow you to regress out covariates / confounds within the nested-CV (following all the best practices). Utilizing this feature requires the input datasets be specified in the pyradigm data structures: code @ https://github.com/raamana/pyradigm, docs @ https://raamana.github.io/pyradigm/. Check the changelog below for more details.

Older news

• neuropredict can handle missing data now (that are encoded with numpy.NaN). This is done respecting the cross-validation splits without any data leakage.

Overview

On a high level,

On a more detailed level,

• Docs: https://raamana.github.io/neuropredict/
• Contributors most welcome: check ideas and the following guidelines. Thanks.

Long term goals

neuropredict, the tool, is part of a broader initiative described below to develop easy, comprehensive and standardized predictive analysis:

Citation

If neuropredict helped you in your research in one way or another, please consider citing one or more of the following, which were essential building blocks of neuropredict: - Pradeep Reddy Raamana. (2017, November 18). neuropredict: easy machine learning and standardized predictive analysis of biomarkers (Version 0.4.5). Zenodo. http://doi.org/10.5281/zenodo.1058993 - Raamana et al, (2017), Python class defining a machine learning dataset ensuring key-based correspondence and maintaining integrity, Journal of Open Source Software, 2(17), 382, doi:10.21105/joss.00382

Change Log - version 0.6

• Major feature: Ability to predict continuous variables (regression)
• Major feature: Ability to handle confounds (regress them out, augmenting etc)
• Redesigned the internal structure for easier extensibility
• New CVResults class for easier management of a wealth of outputs generated in the Classification and Regression workflows
• API access is refreshed and easier

Change Log - version 0.5.2

• Imputation of missing values
• Additional classifiers such as XGBoost, Decision Trees
• Better internal code structure
• Lot more tests
• More precise tests, as we vary number of classes wildly in test suites
• several bug fixes and enhancements
• More cmd line options such as --print_options from a previous run