extracted all demos

chapter13/LDS/TODO.txt

@@ -0,0 +1,2 @@
+(1) test against matlab implementation of kalman filter
+(2) simplify ldsEm with less parameters (G=diag(g), S=I) 

demo/ch10/mixGaussVb_demo.m

@@ -0,0 +1,22 @@
+
+%% Variational Bayesian for Gaussian Mixture Model
+close all; clear;
+d = 2;
+k = 3;
+n = 2000;
+[X,z] = mixGaussRnd(d,k,n);
+plotClass(X,z);
+m = floor(n/2);
+X1 = X(:,1:m);
+X2 = X(:,(m+1):end);
+% VB fitting
+[y1, model, L] = mixGaussVb(X1,10);
+figure;
+plotClass(X1,y1);
+figure;
+plot(L)
+% Predict testing data
+[y2, R] = mixGaussVbPred(model,X2);
+figure;
+plotClass(X2,y2);
+

demo/ch10/rvmRegVb_demo.m

@@ -0,0 +1,19 @@
+clear; close all;
+
+d = 100;
+beta = 1e-1;
+X = rand(1,d);
+w = randn;
+b = randn;
+t = w'*X+b+beta*randn(1,d);
+x = linspace(min(X),max(X),d);   % test data
+
+[model,llh] = linRegVb(X,t);
+% [model,llh] = rvmRegVb(X,t);
+plot(llh);
+[y, sigma] = linRegPred(model,x,t);
+figure
+plotCurveBar(x,y,sigma);
+hold on;
+plot(X,t,'o');
+hold off

chapter11/demo.m → demo/ch11/gauss_demo.m

@@ -1,17 +1,4 @@
 
-% demos for ch11
-% 
-%% Collapse Gibbs sampling for Dirichelt process gaussian mixture model
-close all; clear;
-d = 2;
-k = 3;
-n = 1000;
-[X,label] = mixGaussRnd(d,k,n);
-plotClass(X,label);
-
-[y,model] = mixGaussGb(X);
-figure
-plotClass(X,y);
 %% Sequential update for Gaussian 
 close all; clear;
 d = 2;

demo/ch11/mixGaussGb_demo.m

@@ -0,0 +1,11 @@
+%% Collapse Gibbs sampling for Dirichelt process gaussian mixture model
+close all; clear;
+d = 2;
+k = 3;
+n = 500;
+[X,label] = mixGaussRnd(d,k,n);
+plotClass(X,label);
+
+[y,model] = mixGaussGb(X);
+figure
+plotClass(X,y);

demo/ch12/fa_demo.m

@@ -0,0 +1,13 @@
+% demos for ch12
+
+clear; close all;
+d = 3;
+m = 2;
+n = 1000;
+
+X = ppcaRnd(m,d,n);
+plotClass(X);
+
+%% Factor analysis
+[W, mu, psi, llh] = fa(X, m);
+plot(llh);

chapter12/demo.m → demo/ch12/pca_demo.m

@@ -27,14 +27,3 @@
 figure;
 plotClass(Z1);
 mse3
-%% EM probabilistic PCA
-[W,mu,beta,llh] = ppcaEm(X,m);
-plot(llh)
-
-%% Variational Bayesian probabilistic PCA
-[model, energy] = ppcaVb(X);
-plot(energy);
-% 
-%% Factor analysis
-[W, mu, psi, llh] = fa(X, m);
-plot(llh);

demo/ch12/ppcaEm_demo.m

@@ -0,0 +1,13 @@
+% demos for ch12
+
+clear; close all;
+d = 3;
+m = 2;
+n = 1000;
+
+X = ppcaRnd(m,d,n);
+plotClass(X);
+
+%% EM probabilistic PCA
+[W,mu,beta,llh] = ppcaEm(X,m);
+plot(llh)

demo/ch12/ppcaVb_demo.m

@@ -0,0 +1,13 @@
+% demos for ch12
+
+clear; close all;
+d = 3;
+m = 2;
+n = 1000;
+
+X = ppcaRnd(m,d,n);
+plotClass(X);
+
+%% Variational Bayesian probabilistic PCA
+[model, energy] = ppcaVb(X);
+plot(energy);

chapter13/LDS/demo.m → demo/ch13/lds_demo.m

@@ -1,7 +1,3 @@
-% TODO: 
-% 1) test against matlab implementation of kalman filter
-% 2) simplify ldsEm with less parameters (G=diag(g), S=I) 
-
 % demos for LDS in ch13
 
 clear; close all;

demo/ch14/adaboostBin_demo.m

@@ -0,0 +1,10 @@
+
+%% adaboost
+d = 2;
+k = 2;
+n = 500;
+[X,t] = kmeansRnd(d,k,n);
+model = adaboostBin(X,t-1);
+y = adaboostBinPred(model,X);
+plotClass(X,y+1)
+

demo/ch14/mixLinReg_demo.m

@@ -0,0 +1,14 @@
+%% Mixture of linear regression
+close all; clear
+d = 1;
+k = 2;
+n = 500;
+[X,y] = mixLinRnd(d,k,n);
+plot(X,y,'.');
+[label,model,llh] = mixLinReg(X, y, k);
+plotClass([X;y],label);
+figure
+plot(llh);
+[y_,z,p] = mixLinPred(model,X,y);
+figure;
+plotClass([X;y],label);

demo/ch14/mixLogitBin_demo.m

@@ -0,0 +1,10 @@
+%% Mixture of logistic regression
+d = 2;
+c = 4;
+k = 4;
+n = 500;
+[X,t] = kmeansRnd(d,c,n);
+
+model = mixLogitBin(X,t-1,k);
+y = mixLogitBinPred(model,X);
+plotClass(X,y+1)