Machine learning end to end flow

data extraction and cleaning

important_features_index=[9,14,17,20,692,699,700,701,702,703,704,705,1368,1370,1371,1410,1411]
important_features_index=[x-2 for x in important_features_index]
demographics_data=demographics_data.iloc[important_features_index,:]
demographics_data=demographics_data.set_index('Attribute')
demographics_data=demographics_data.transpose()
new_columns_name=['Total population' ,'age 15-24','age 25-34','age 35-44','people with income','20-30 thousand','30-40 thousand','40-50 thousand','50-60 thousand','60-80 thousand','80-90 thousand','100 thousand and more','Total population over 15 for education','High school diploma or equivalent','Postsecondary certificate, diploma or degree','In the labour force','Employed']
demographics_data.columns=new_columns_name
demographics_data=demographics_data.reset_index(drop=False)
demographics_data.rename(columns={'index':'Neighborhood'},inplace=True)
demographics_data.drop(index=[0,1,2],axis=1,inplace=True)
demographics_data.reset_index(drop=True,inplace=True)
# calcaulating the number of 15-45 population and change the name of the columns
demographics_data['number of educated people']=(demographics_data['High school diploma or equivalent']+demographics_data['Postsecondary certificate, diploma or degree'])
demographics_data['number of 15-45']=(demographics_data['age 15-24']+demographics_data['age 25-34']+demographics_data['age 35-44'])
demographics_data['number of employers']=demographics_data['Employed']
demographics_data.drop(columns=new_columns_name[1:],axis=1,inplace=True)
demographics_data.sort_values(by='Neighborhood')
demographics_data.head()