Algorithm: Applewatch dataset with KMeans / research

applewatchKMeans.py

# Author Remon Hasan, University of Asia Pacific
# Applewatch dataset solve with unsupervised way
# using KMeans
# implementation with google Colab
import pandas as pd
import numpy as np

data = pd.read_csv('AppleWatch.csv')

# mapping gender
data['Gender'].replace(to_replace ="M",  value ="2", inplace = True) 
data['Gender'].replace(to_replace ="F",  value ="1", inplace = True) 
#print(data['Gender'].head())
#data['Gender'].astype(int)

#mapping Activity
data['Activity'].replace(to_replace ="0.Sleep",  value ="0", inplace = True)
data['Activity'].replace(to_replace ="1.Sedentary",  value ="1", inplace = True)
data['Activity'].replace(to_replace ="2.Light",  value ="2", inplace = True)
data['Activity'].replace(to_replace ="3.Moderate",  value ="3", inplace = True)
data['Activity'].replace(to_replace ="4.Vigorous",  value ="4", inplace = True)
#print(data['Activity'].head())

# Number of clusters
kmeans = KMeans(n_clusters=2)
# Fitting the input data
kmeans = kmeans.fit(data)
# Getting the cluster labels
labels = kmeans.predict(data)
print(labels)
# Centroid values
# centroids = kmeans.cluster_centers_