fit_unregularized_stacked_model: Estimate "optimal" unregularized weights for a linear...

Description Usage Arguments Value

Description

This function uses either a grid search or the "degenerate EM" algorithm outlined at http://www.cs.cmu.edu/~roni/11761-f16/Presentations/degenerateEM.pdf

Usage

1
2
3
fit_unregularized_stacked_model(component_model_log_scores,
  method = "em", tol = .Machine$double.eps, grid_size = 101,
  return_type = "optimal")

Arguments

component_model_log_scores

a data frame or matrix of log scores. Each column gives log scores for a particular predictive model. Each row corresponds to one observation

method

character, either "em" for degenerate EM or "grid-search" for grid search

tol

numeric, if method was "em", stopping tolerance

grid_size

number of grid points for each model to evaluate in a grid search

return_type

character, if method was "grid-search", either "optimal" to return only the best weights examined or "all" to return the full grid

Value

model weights


reichlab/2018-2019-cdc-flu-contest documentation built on May 24, 2019, 7:36 a.m.