import pandas
df=pandas.read_csv('studentsperformance.csv')
print(df)

import sklearn

empties=df.isnull().sum()
print(empties)

df['race_ethnicity'].replace({'group A':0,'group B':1, 'group C':2,'group D':3,'group E':4}, inplace=True)
df['gender'].replace({'male':0, 'female':1},inplace=True)
df['lunch'].replace({'standard':0, 'free/reduced':1},inplace=True)
df['test_preparation_course'].replace({'none':0, 'completed':1},inplace=True)
df['parental_level_of_education'].replace({'bachelors_degree':0,'some_college':1,'masters_degree':2,'associates_degree':3,
                                           'high_school':4,'some_high_school':5},inplace=True)
print(df)

subset=df[['math_score','reading_score','writing_score']]
array= subset.values
X= array[:,0:3]

from sklearn.cluster import KMeans
model = KMeans(n_clusters=3, random_state=42)
model.fit_predict(X)
print('Learning// Done')

centronoids = model.cluster_centers_
dataframe = pandas.DataFrame(centronoids,columns=['math_score','reading_score','writing_score'])
print(dataframe)

subset['label']=model.labels_
subset = subset[subset['label']== 1]
print(subset)

import xlwt
subset.to_excel('cluster1.xls', columns=['math_score','reading_score','writing_score'])