Skip to content
Unverified Commit 77ce4b43 authored by Jon Lachmann's avatar Jon Lachmann Committed by GitHub
Browse files

Predict (#10)



* Store coefficients for each model. Only gaussian loglik supports it as of now.
* Predict function works but needs further testing and fixes for quantiles.
* Removed redundant sum.
* Added a comment in predict.
* Save pruned results in merge.results.
Predict for multiple populations implemented.
* Add population weights to population-wise predictions in predict fun.
* Update docs.
* Fix mean calculation in predict function.
* Adjust logistic likelihood to take r as a parameter.
* adding two tuning parameters:
1. for bart models
2. for controlling whether we want to check colinearity
* returning an object of interest from summary function
* Rename argument which turns on or off collinearity checks when generating features -> "check.col".
Add an argument which limits the size of projections, can be set to null and is defaulted to be equal to L.
* Initialize simulated annealing with random numbers to avoid all zero features.
Add an option to round alphas when printing features and use 2 decimal places everywhere a feature is just printed for display.

---------

Co-authored-by: Jon <test>
Co-authored-by: default avatarAliaksandr <aliaksah@tjilpa.uio.no>
parent 59c1a00a
Loading
Loading
Loading
Loading
0% Loading or .
You are about to add 0 people to the discussion. Proceed with caution.
Please to comment