diff --git a/Assignment_2/Image2Graph.m~ b/Assignment_2/Image2Graph.m~
deleted file mode 100644
index aed0513..0000000
--- a/Assignment_2/Image2Graph.m~
+++ /dev/null
@@ -1,36 +0,0 @@
-function myAffinityMat = Image2Graph (imIn)
-%Image2Graph produces a graph, in the form of a matrix, from an image
-%   Usage myAffinityMat = Image2Graph (imIn), where:
-%       Inputs
-%       - imIn is the input image
-%       Output
-%       - myAffinityMat is the graph produced
-
-    imageWidth = size(imIn, 2);
-    imageHeight = size(imIn, 1);
-    imageChannels = size(imIn, 3);
-
-    %myAffinityMat = 1 / exp(norm(imIn(:, :)));
-    myAffinityMat = zeros(imageWidth * imageHeight, imageWidth * imageHeight);
-    
-    for row = 1:imageHeight
-        for column = 1:imageWidth
-            diff = imIn - imIn(row, column, :);
-            %diffT = permute(diff, [2 1 3]);
-            
-            dist = sum(diff .^ 2, 3) .^ 0.5;
-            
-%             dist = zeros(imageWidth, imageHeight);
-%             for channel = 1:imageChannels
-%                 dist = dist + sqrt(sum (diff(:, :, channel) * diffT(:, :, channel)));
-%                 dist = dist + sqrt(diff(:, :, channel) * diffT(:, :, channel));
-%             end
-            
-            som = 1 ./ exp(dist);
-            som2 = reshape(som, 1, []);
-            som2 = reshape(som.', 1, []);
-            
-            myAffinityMat((row - 1) * imageWidth + column, :) = reshape((1 ./ exp(dist), 1, []);
-        end
-    end
-end
diff --git a/Assignment_2/mySpectralClustering.m b/Assignment_2/mySpectralClustering.m
new file mode 100644
index 0000000..2e5834e
--- /dev/null
+++ b/Assignment_2/mySpectralClustering.m
@@ -0,0 +1,15 @@
+function clusterIdx = mySpectralClustering (anAffinityMat, k)
+%Implementation of spectral clustering
+%   Usage clusterIdx = mySpectralClustering (anAffinityMat, k), where:
+%       Inputs
+%       - anAffinityMat is a rectagular, symmetrical affinity matrix
+%           representation of an image
+%       - k is the desired number of clusters
+%       Output
+%       - clusterIdx is a vector storing the cluster Id of each node
+
+    L = diag(sum(anAffinityMat, 2)) - anAffinityMat;
+    [eigenvectorsMatrix, ~] = eigs(L, k, 'sm');
+    clusterIdx = kmeans(eigenvectorsMatrix, k);
+end
+