close all;

% Normalisation et visualisation des données
Xn=(X-repmat(mean(X),size(X,1),1))*inv(diag(std(X,1)'));
% classr(X,L,pays)

% Normalisation et projection de Xc
Xnc=(Xc-repmat(mean(X),size(Xc,1),1))*inv(diag(std(X,1)'));
% figure
% classr2(X,Xnc,L,pays,paysc)

% Options pour la fonction knnclassify
% 'nearest'   Majority rule with nearest point tie-break
% 'random'    Majority rule with random point tie-break
% 'consensus' Consensus rule
% The default behavior is to use majority rule. That is, a sample point
% is assigned to the class from which the majority of the K nearest
% neighbors are from. Use 'consensus' to require a consensus, as opposed
% to majority rule. When using the consensus option, points where not all
% of the K nearest neighbors are from the same class are not assigned
% to one of the classes. Instead the output CLASS for these points is NaN
% for numerical groups or '' for string named groups. When classifying to
% more than two groups or when using an even value for K, it might be
% necessary to break a tie in the number of nearest neighbors. Options
% are 'random', which selects a random tiebreaker, and 'nearest', which
% uses the nearest neighbor among the tied groups to break the tie. The
% default behavior is majority rule, nearest tie-break.

% Application de la fonction knnclassify
class1n = knnclassify(Xnc,Xn,L,1,'euclidean','nearest')
class2n = knnclassify(Xnc,Xn,L,2,'euclidean','nearest')
class3n = knnclassify(Xnc,Xn,L,3,'euclidean','nearest')
class4n = knnclassify(Xnc,Xn,L,4,'euclidean','nearest')
class1r = knnclassify(Xnc,Xn,L,1,'euclidean','random')
class2r = knnclassify(Xnc,Xn,L,2,'euclidean','random')
class3r = knnclassify(Xnc,Xn,L,3,'euclidean','random')
class4r = knnclassify(Xnc,Xn,L,4,'euclidean','random')
class1c = knnclassify(Xnc,Xn,L,1,'euclidean','consensus')
class2c = knnclassify(Xnc,Xn,L,2,'euclidean','consensus')
class3c = knnclassify(Xnc,Xn,L,3,'euclidean','consensus')
class4c = knnclassify(Xnc,Xn,L,4,'euclidean','consensus')

% Comment les fonctions discriminantes sont calculées
% Linear discrimination fits a multivariate normal density to each group, 
% with a pooled estimate of covariance. 
% Quadratic discrimination fits MVN densities with covariance estimates 
% stratified by group.
% Mahalanobis discrimination uses Mahalanobis distances with stratified 
% covariance estimates.

% Application de la fonction classify
[classl,Pl,errl] = classify(Xnc,Xn,L,'linear')
[classq,Pq,errq] = classify(Xnc,Xn,L,'quadratic')
[classm,Pm,errm] = classify(Xnc,Xn,L,'mahalanobis')