Clustering(Unsupervised)

Coded Examples of Different types of Clustering Techniques...

K-means Clustering-

Digit Recognition...Well it do seems to wear off upon convoluted encounters...Hence we got..

Spectral Clustering-open KMeansClustering.py for side by side comparision

Plays pretty good with complex geometry but still, what if we have like- Million DataPoints?

Mini-Batch k-means-

baddest off em' all...Color Quantization..Set k as per ur need..!

K-means elbow method-

Well bruh..do u really know,how many clusters u need?..Go for the elbow!

NOTE !!!

paste this little shite in DigitRecognitionKMeans.py for generating the Heatmap..(A cooler terminology for Confusion Matrix)

from sklearn.metrics import confusion_matrix
mat=confusion_matrix(digits.target,labels)
sns.heatmap(mat.T,square=True,annot=True,fmt='d',cbar='False', xticklabels=digits.target_names,yticklabels=digits.target_names)
plt.xlabel("True Label")
plt.ylabel("False Label")