Clustering On The Fly - A Pursuit for the Optimal Algorithm
Our pipeline helps to select the most optimal clustering at runtime based on the cost model we have developed (Includes computation cost and Memory cost). 

Data-sets to run:
1. make_blobs_data.txt
2. make_moons_data.txt
3. make_circles_data.txt

Corresponding labels:
1. make_blobs_labels.txt
2. make_moons_labels.txt
3. make_circles_labels.txt

Commands to run:
python3 <Our Pipeline> <Fraction of dataset> <data-set> <labels>
python3 Library.py 0.14 make_circles_data.txt make_circles_labels.txt

Installations Required:
1. Valgrind
2. Sklearn
3. CACTI Tool

Files Prosent:
KMeans.py runs only KMeans Clustering Algorithm
DBSCAN.py runs only DBSCAN Clustering Algorithm
Library.py runs our pipeline
Results folder has the results of our runs