Exercise6 - Week7 Finished : K-Means & PCA

Author: ProgrammerYuan

machine-learning-ex7/ex7/computeCentroids.m

@@ -1,3 +1,4 @@
+
 function centroids = computeCentroids(X, idx, K)
 %COMPUTECENTROIDS returns the new centroids by computing the means of the 
 %data points assigned to each centroid.
@@ -25,7 +26,10 @@
 %
 % Note: You can use a for-loop over the centroids to compute this.
 %
-
+for i = 1:K
+    idx_sub = idx == i;
+    centroids(i, :) = mean(X(idx_sub, :)); 
+end
 
 
 

machine-learning-ex7/ex7/findClosestCentroids.m

@@ -7,9 +7,11 @@
 
 % Set K
 K = size(centroids, 1);
+% Get sample numbers
+m = size(X, 1);
 
 % You need to return the following variables correctly.
-idx = zeros(size(X,1), 1);
+idx = zeros(m, 1);
 
 % ====================== YOUR CODE HERE ======================
 % Instructions: Go over every example, find its closest centroid, and store
@@ -21,6 +23,18 @@
 % Note: You can use a for-loop over the examples to compute this.
 %
 
+for i = 1:m
+    idx_value = 1;
+    dist_value = sum((centroids(1, :) - X(i, :)).^2);
+    for j = 2:K
+        dist_temp = sum((centroids(j, :) - X(i, :)).^2);
+        if dist_temp < dist_value
+            dist_value = dist_temp;
+            idx_value = j;
+        end
+    end
+    idx(i) = idx_value;
+end
 
 
 

machine-learning-ex7/ex7/kMeansInitCentroids.m

@@ -13,7 +13,11 @@
 %               the dataset X
 %
 
-
+% Initialize the centroids to be random examples
+% Randomly reorder the indices of examples
+randidx = randperm(size(X, 1));
+% Take the first K examples as centroids
+centroids = X(randidx(1:K), :);
 
 
 

machine-learning-ex7/ex7/pca.m

@@ -20,8 +20,8 @@
 %       number of examples).
 %
 
-
-
+sigma = (X' * X) / m;
+[U, S, V] = svd(sigma);
 
 
 

machine-learning-ex7/ex7/projectData.m

@@ -17,7 +17,8 @@
 %                    x = X(i, :)';
 %                    projection_k = x' * U(:, k);
 %
-
+U_reduce = U(:, 1:K);
+Z = X * U_reduce;
 
 
 

machine-learning-ex7/ex7/recoverData.m

@@ -20,8 +20,9 @@
 %
 %               Notice that U(j, 1:K) is a row vector.
 %               
-
-
+U_reduce = U(:, 1:K);
+X_rec = U_reduce * Z';
+X_rec = X_rec';
 
 % =============================================================