What are the inputs of fitcsvm to train a clasifier model?

Can anyone tell me what will be the X and Y..i have read theory on Matab website but i do not understand..

SVMModel = fitcsvm(X,Y)

i am using bag of features approach...my algo so far is

1.Surf feature Extraction of dataset of images

2. clutering using k means

3. histogram representation of images

pls suggest according to my approach what should be inputs X and Y..i have to do binary classification my code is:

if true
  % clc;
clear;
close all;
features = {};  
folder = 'D:\Dataset\data1\data';
filePattern = fullfile(folder, '*.jpg');
f=dir(filePattern);
files={f.name}; 
imds = imageDatastore(folder ,'IncludeSubfolders',true,'LabelSource',...
  'foldernames');
num_features = zeros(numel(files), 1); % New - for keeping track of # of features per image
for k=1:numel(files)
  fullFileName = fullfile(folder, files{k});
  H = fspecial('log');
  image=imfilter(imread(fullFileName),H);
  image=rgb2gray(image);
  temp =  detectSURFFeatures(image);
  [im_features, temp] = extractFeatures(image, temp);
num_features(k) = size(im_features, 1); % New - # of features per image
  features{k}= im_features;    
end

features = vertcat(features{:});

num_clusters = 200;

[assignments,centers] = kmeans(double(features), num_clusters);

counter = 1;

features_hist = zeros(numel(files), num_clusters);

for k = 1 : numel(files)

    a = assignments(counter : counter + num_features(k) - 1);

    h = histcounts(a, 1 : num_clusters + 1);

   features_hist(k, :) = h;

    % Increment counter 

    counter = counter + num_features(k);
end

end