for cohort analysis
Project description
fabcohort
A small demo library for a fab_cohort about cohort analysis
Installation
pip install fabcohort
Get started
How to do cohort analysis with this lib:
FUNCTION1:
Vanilla cohort analysis
Pandas df.head(5) should look like -
user_id | date | count |
---|---|---|
5fb507360cd5c0 | 2023-04-01 | 1 |
weg507360cwfw3 | 2023-03-01 | 1 |
6001ef966c13w3 | 2023-02-01 | 1 |
weg507360cwfw3 | 2023-04-01 | 1 |
6001ef966c13w3 | 2023-03-01 | 1 |
from fab_cohort import Cohort
# Instantiate a Cohort object
cohort = Cohort()
# Call the count_cohort method, e.g., MS for month start, W-MON for week start
result = cohort.count_cohort(df, frequency)
FUNCTION2:
Cohort analysis by segments
Pandas df.head(5) should look like -
user_id | date | segment | count |
---|---|---|---|
5fb507360cd5c0 | 2023-04-01 | A,B | 1 |
weg507360cwfw3 | 2023-03-01 | A, | 1 |
6001ef966c13w3 | 2023-02-01 | C,D | 1 |
weg507360cwfw3 | 2023-04-01 | B,D | 1 |
6001ef966c13w3 | 2023-03-01 | A,B | 1 |
from fab_cohort import Cohort
# Instantiate a Cohort object
cohort = Cohort()
# Call the count_cohort_segments method, e.g., MS for month start, W-MON for week start
result = cohort.count_cohort_segments(df, frequency)
# (Optional) if you have multiple segments just parse it
result[['segment1', 'segment2']] = result['segment'].str.split(',', expand=True)
result.drop('segment', axis=1, inplace=True)
FUNCTION3:
Convert the count to percentage
# once the above result is obtained
# Call the count_cohort method
result_pct = cohort.to_pct(result)
# (Optional) if you have multiple segments just parse it
result[['segment1', 'segment2']] = result['segment'].str.split(',', expand=True)
result.drop('segment', axis=1, inplace=True)
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
fabcohort-0.3.0.tar.gz
(3.1 kB
view hashes)
Built Distribution
Close
Hashes for fabcohort-0.3.0-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 57b60abe31aa72a42bf3c4df2f9cd99bf1fee92ee81602842acbf59ea5ef8a81 |
|
MD5 | 6d5f2d34ac1517df587e8a4bd4717071 |
|
BLAKE2b-256 | 737cdc496d422c3e9c8dd69cad8a3875ab0a237bc582debc05f7795c6a5d2c30 |