read_data.asv

function [X, yc, pc, ynan, p, text_vars] = read_data(beh_type, lag,...
    beh, dirstring, pats)
    % This function was made to read in the data.  The data function
    % reads in the data as a table which reads in blanks as NaNs.  It
    % renames the variables as necessary and selects a subset of variables
    % to analyze.
    % It runs the cleaning and lagging data to provide an X and y dataset.
    % Parameters: 
    % beh_type: 'AGG','SIB','BOTH' - the type of behavior
    % lag: number of lagged days to examine; positive integer
    % beh: whether or not to include prior behavior as a lagged variable
    % (true or false)
    % dirstring: a string to save out the directory
    % pats: a string array of variable patterns, can be prefix to i

    T                           = readtable(strcat(beh_type,'.csv'),...
        'PreserveVariableNames',true);
    T                           = renamevars(T,["BM_Menses"],["Menses"]);
    if ~exist('beh','var')
       beh                      = false;
    end

    A                           = {'Weather_TMAX',...
                                    'Weather_TMIN',...
                                    'Weather_PRCP',...
                                    'Weather_SNOW',...
                                    'Weather_SNWD',...
                                    'Moon Rise',...
                                    'Moon Culmination',...
                                    'Moon Set',...
                                    'Moon Altitude',...
                                    'Moon Azimuth',...
                                    'Moon Shadow Length',...
                                    'Moon Distance',...
                                    'Moon Phase',...
                                    'Moon Age',...
                                    'Moon Next New',...
                                    'Moon Next Full',...
                                    'Allergen_Trees',...
                                    'Allergen_Weeds',...
                                    'Allergen_Grasses',...
                                    'Allergen_Total',...
                                    'Sleep_Hours',...
                                    'Sleep_Interruptions',...
                                    'BM_Count',...
                                    'BM_Bristol_Extreme',...
                                    'BM_Bristol_Max',...
                                    'BM_Bristol_Min',...
                                    'BM_Bristol',...
                                    'BM_Intervention',...
                                    'BM_Time',...
                                    'BM_Abnormal',...
                                    'BM_Constipation',...
                                    'BM_Diarrhea',...
                                    'Menses'};

    if beh
        A{end+1}                = 'Behavior_Factor';
        pats                    = [pats, "Behavior_Factor"];
    end


    B                           = string(A);

    T_X                         = table2array(T(:,B));
    T_y                         = table2array(T(:,'Behavior_Factor'));
    T_person                    = table2array(T(:,'Person_ID'));

%      females = load('females.csv');
%      non_females = setdiff(unique(T_person), females);
%     %females = non_females;
%     for i=1:length(females)
%         subplot(round(length(females)/2),2,i)
%         menses = T_X(T_person==females(i),end);
%         plot(menses);
%         hold on;
%         unique(menses(~isnan(menses)))
%     end
%     error()
    
    dataset                     = 'complete';
    [Xc,yc,pc, ynan,p]        =  CleanUpData_lag(T_X,...
        T_y,dataset,T_person,lag);
    %X                           =  zscore(Xc);
    X = Xc;
    vars                        = zeros(length(B),1);
    for k=1:length(pats)
        for i=1:length(B)
            if startsWith(B(i), pats(k))
                vars(i)         = 1;
            end
        end
    end
    vars                        = logical(vars);
    lagged_vars                 = [];
    for i=1:lag
        lagged_vars             = [lagged_vars; vars];
    end
    lagged_vars                 = logical(lagged_vars);
    text_vars                   = B(lagged_vars);

    log_output(dirstring,string(text_vars));
    size(lagged_vars)
    writematrix(lagged_vars, strcat(dirstring,'/variables.txt'))
    pause(3)
    size(X)
    X                           = X(:,lagged_vars);
    size(X)
    pause(2)
end