generate_histories.m

Implementation to linear, quadratic and logistic discriminant analysis, for examples

function [data, labels] = generate_histories(choice_vector, method, analysis_window)

% [DATA, LABELS] = generate_histories(CHOICE_VECTOR, METHOD, ANALYSIS_WINDOW)
%
% Converts an input choice vector into a set of history vectors (into DATA) 
% and switch/not-switch labels (into LABELS)
% 
% Required:
%   CHOICE_VECTOR - The vector of choices given by the user (e.g. c1(:,1)')
%
%   METHOD - Must be either 1, 2, or 3.  Will specify the form of the reward
%   schedule presented under this trial
% 
%   ANALYSIS_WINDOW - The size of the reward history to use.  For example,
%   if we want to include the past three rewards in the history vector,
%   then ANALYSIS_WINDOW will be 3.
%
% The output arguments are [DATA, LABELS].  DATA will contain the history
% vectors and LABELS will contain the labels.
%
% Examples:
%   For the examples described in our project
% 
% >> [train_data, train_labels]= generate_histories(c1(:,1)', 1, 5);

reward_vector = play_game(choice_vector, method);  

for i=1:analysis_window,
    reward_histories(:, i) = [zeros(analysis_window-i+1,1); reward_vector(1:end+i-analysis_window-1)'];    
end;
reward_histories = fliplr(reward_histories);
labels = [0; abs(diff(choice_vector'))];

trial_num = 1:length(choice_vector);
accum_rewards = cumsum(reward_vector);
training_data = [labels reward_histories accum_rewards' trial_num'];
data = training_data(analysis_window+1:end, 2:end);
labels = training_data(analysis_window+1:end, 1);