Predicting Mortality of ICU Patients: The PhysioNet/Computing in Cardiology Challenge 2012 1.0.0
(1,345 bytes)
function [ pred ] = pnBRFApply(model, X)
%PNBRFAPPLY Calculate predictions on data using provided model
% [ pred ] = pnBRFApply(model, X) calculates predictions on the
% data using model, which is assumed to be of a certain form compatible
% with this function.
%
% Inputs:
% X - NxD double matrix of data. May contain NaNs for missing
% values. This should be extracted by the associated
% pnBRFFeatures function.
%
% Outputs:
% pred - Predictions for the given data set - vector sized Nx1
%
% Example:
% % Develop and evaluate a model on the training data
% bpath = './set-a/';
% data = pnLoadTextFilesCell(bpath);
% [targets,target_header] = pnLoadTargetFile('Outcomes-a.txt');
% data_target=targets(:,6);
% data_used = pnExtractFeaturesNic(data);
%
% [X,header] = pnBRFFeatures(data);
% model = pnBRFDevelop(X,data_target);
% pred = pnBRFApply(model, X);
%
% See also PNDEVELOPMODELS PNBRFDEVELOP PNBRFFEATURES
% References:
% Physionet Challenge 2012
% Copyright 2012 Nic Johnson
% $LastChangedBy: alistair $
% $LastChangedDate: 2012-08-25 00:44:01 -0400 (Sat, 25 Aug 2012) $
% $Revision: 153 $
% Originally written on GLNXA64 by Alistair Johnson, 15-May-2012 11:25:16
% Contact: alistairewj@gmail.com
%=== Calculate model predictions
pred = NicForest_apply_quick(model , X );
end