test_mlgp.m~

clear;clc;

ranks = [3,3,3];
s2 = 1e-3;
D = 45;
T1 = 138;
T2 = 3;
dims = [D,,3];


[X_obv, Y_obv]  = mlgp_load_data('RestaurantDataset.mat');


N = size(X_obv,1);
data_ind = randperm(N);
train_ratio = 0.8;
train_size = ceil(N*train_ratio);
train_ind = data_ind(1:train_size);
test_ind = data_ind(train_size+1:end);



%% training
[model, L_list ] = mlgp_train( X_obv, Y_obv, dims,ranks);

[ Y_pred, V_pred, MSE ] = mlgp_predict(X_obv, Y_obv, train_ind, test_ind, dims, model );
fprintf('mse %d', MSE);
plot(L_list(10:end));