


FORMAT prt_plot_prediction_reg_bar(PRT, model, axes_handle)
This function plots the bar plot that appears on prt_ui_results
Inputs:
PRT - data/design/model structure (it needs to contain
at least one estimated model).
model - the number of the model that will be ploted
axes_handle - (Optional) axes where the plot will be displayed
Output:
None
__________________________________________________________________________
Copyright (C) 2011 Machine Learning & Neuroimaging Laboratory

0001 function prt_plot_prediction_reg_bar(PRT, model, axes_handle) 0002 % FORMAT prt_plot_prediction_reg_bar(PRT, model, axes_handle) 0003 % 0004 % This function plots the bar plot that appears on prt_ui_results 0005 % Inputs: 0006 % PRT - data/design/model structure (it needs to contain 0007 % at least one estimated model). 0008 % model - the number of the model that will be ploted 0009 % axes_handle - (Optional) axes where the plot will be displayed 0010 % 0011 % Output: 0012 % None 0013 %__________________________________________________________________________ 0014 % Copyright (C) 2011 Machine Learning & Neuroimaging Laboratory 0015 0016 % Written by M. J. Rosa 0017 % $Id: prt_plot_prediction_reg_bar.m 706 2013-06-07 14:33:34Z cphillip $ 0018 0019 nfold = length(PRT.model(model).output.fold); 0020 0021 %If no axes_handle is given, create a new window 0022 if ~exist('axes_handle', 'var') 0023 figure; 0024 axes_handle = axes; 0025 else 0026 set(axes_handle, 'XScale','linear'); 0027 end 0028 0029 cla(axes_handle, 'reset'); 0030 preds1 = []; 0031 preds2 = []; 0032 for f = 1:nfold 0033 preds1 = [preds1; PRT.model(model).output.fold(f).targets]; 0034 preds2 = [preds2; PRT.model(model).output.fold(f).predictions]; 0035 end 0036 bar(axes_handle,[preds1 preds2]); 0037 xlabel(axes_handle,'subjects','FontWeight','bold'); 0038 ylabel(axes_handle,'targets and predictions','FontWeight','bold'); 0039 legend(axes_handle,{'Target', 'Predicted'});