Codecademy export

script.py

import codecademylib
import pandas as pd

ad_clicks = pd.read_csv('ad_clicks.csv')
# reviewing the table
print(ad_clicks.head())

ad_clicks_count = ad_clicks.groupby('utm_source').user_id.count().reset_index()
print(ad_clicks_count)

ad_clicks['is_click'] = ~ad_clicks.ad_click_timestamp.isnull() 
print(ad_clicks.head())

clicks_by_source = ad_clicks.groupby(['utm_source', 'is_click']).user_id.count().reset_index()
clicks_pivot = clicks_by_source.pivot(
  columns = 'is_click',
  index = 'utm_source',
  values = 'user_id'
).reset_index()
print(clicks_pivot)
clicks_pivot['percent_clicked'] = clicks_pivot[True]\
/ (clicks_pivot[True] + clicks_pivot[False])
print(clicks_pivot)

print(ad_clicks.groupby('experimental_group').user_id.count())

is_click_group = ad_clicks.groupby(['experimental_group','is_click']).user_id.count().reset_index()
is_click_group_pivot = is_click_group.pivot(
  columns = 'is_click',
  index = 'experimental_group',
  values = 'user_id'
).reset_index()
print(is_click_group_pivot) 

percentage = is_click_group_pivot[True]/(is_click_group_pivot[True] + is_click_group_pivot[False])
print(percentage)

a_clicks = ad_clicks[ad_clicks.experimental_group == 'A']
b_clicks = ad_clicks[ad_clicks.experimental_group == 'B']
print(b_clicks.head())


a_b_clicks = a_clicks.groupby(['is_click', 'day']).user_id.count().reset_index()
print(a_b_clicks.head())

a_b_clicks_pivot = a_b_clicks.pivot(
  columns = 'is_click',
  index = 'day',
  values = 'user_id'
).reset_index()
print(a_b_clicks_pivot.head()) 

a_percent = a_b_clicks_pivot[True] / (a_b_clicks_pivot[True] + a_b_clicks_pivot[False])
print(a_percent)

ab_clicks = b_clicks.groupby(['is_click', 'day']).user_id.count().reset_index()
print(ab_clicks.head())

ab_clicks_pivot = ab_clicks.pivot(
  columns = 'is_click',
  index = 'day',
  values = 'user_id'
).reset_index()
print(ab_clicks_pivot.head()) 

b_percent = ab_clicks_pivot[True] / (ab_clicks_pivot[True] + ab_clicks_pivot[False])
print(b_percent)