DolbyUUU / clustering_algorithm_implementation_python Public

Notifications You must be signed in to change notification settings
Fork 0
Star 4

Clustering algorithm implementaions from scratch with python (k-means, EM-GMM, mean-shift, agglomerative)

Notifications

Name		Name	Last commit message	Last commit date
Latest commit History 15 Commits
data_2d_test		data_2d_test
test_2d_visualization_results		test_2d_visualization_results
GaussianMM.py		GaussianMM.py
GaussianMM_Ver0.py		GaussianMM_Ver0.py
KMeans.py		KMeans.py
MeanShift.py		MeanShift.py
MeanShift_Ver0.py		MeanShift_Ver0.py
README.md		README.md
test_2d_visualization.py		test_2d_visualization.py

Repository files navigation

Clustering algorithm implementaion and visualization from scratch with python

Three popular clustering algorithms (for d >= 2 dimensions, k >= 2 clusters): (1) k-means (2) EM-GMM (Gaussian mixture model - expectation maximization) (3) mean-shift

Python implementations: KMeans.py: k-means GaussianMM.py: EM-GMM (Gaussian mixture model - expectation maximization) GaussianMM_Ver0.py: another version for EM-GMM (Gaussian mixture model - expectation maximization) MeanShift.py: mean-shift MeanShift_Ver0.py: another version for mean-shift

Test on 2D datasets for visualization, compared with sklearn implementation: (1) blobs pattern (2) sticks pattern (3) moon and stars pattern