How to use

import warnings
warnings.simplefilter(action='ignore', category=FutureWarning)

Init Pyspark

import pyspark.sql.functions as F
from pyspark.sql import SparkSession

spark = (
    SparkSession.builder.master("local[*]")
    .config("spark.executor.memory", "6g")
    .config("spark.driver.memory", "6g")
    .getOrCreate()
)

spark.sparkContext.setLogLevel("ERROR")

spark.conf.set("spark.sql.repl.eagerEval.enabled", True)

Setting default log level to "WARN".
To adjust logging level use sc.setLogLevel(newLevel). For SparkR, use setLogLevel(newLevel).

23/02/26 12:48:37 WARN NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable

Create a sample dataset

import numpy as np
import pandas as pd

def make_transactions(n=10000):
    return pd.DataFrame(
        data={
            "transaction_id": range(n),
            "consumer_id": [np.random.choice(list(range(100))) for _ in range(n)],
            "paid_value": np.random.normal(1000, 100, n),
            "discount": [np.random.choice([10, 0, 25]) for _ in range(n)],
            "product_type": [
                str(np.random.choice(["a", "b", "c", "d", "e", "f", "g", "h"])) for _ in range(n)
            ],
            "buy_type": [
                str(
                    np.random.choice(
                        ["f_1", "f_2", "f_3", "f_4", "f_5", "f_6", "f_7", "f_8", "f_9"]
                    )
                )
                for _ in range(n)
            ],
            "paymnent_date": pd.date_range("2022-01-01", "2023-01-01", periods=n),
        }
    )

make_transactions().head()

	transaction_id	consumer_id	paid_value	discount	product_type	buy_type	paymnent_date
0	0	84	938.706334	25	d	f_3	2022-01-01 00:00:00.000000000
1	1	74	978.164629	0	f	f_5	2022-01-01 00:52:33.915391539
2	2	95	1013.975424	25	h	f_1	2022-01-01 01:45:07.830783078
3	3	21	1000.291093	0	g	f_9	2022-01-01 02:37:41.746174617
4	4	32	1115.160083	0	f	f_3	2022-01-01 03:30:15.661566156

Generate public

We will select the monthly active users as our public.

transactions = spark.createDataFrame(make_transactions())

public = transactions.select(F.col("consumer_id").alias("consumer_id_ref"), F.to_date(F.date_trunc("month", "paymnent_date")).alias("date_ref")).distinct()

Generate Dataset

We will defime a 30 days window to build features

In our case it means that we wil select every transaction in the window defined by date_ref and date_ref - 30 days.

from autofeats.types import Dataset

df = Dataset(
    table=transactions,
    primary_key_col="transaction_id",
    table_join_key_col="consumer_id",
    table_join_date_col="paymnent_date",
    numerical_cols=["paid_value", "discount"],
    categorical_cols=["product_type", "buy_type"],
    public=public,
    public_join_key_col="consumer_id_ref",
    public_join_date_col="date_ref",
    subtract_in_end=30
)

from autofeats import make_features

numerical_statistics example

features = make_features.run(
    df=df,
    suites=[
        "numerical_statistics",
    ],
    options={},
)

print(f"Features created: {features.columns}")

print(f"Number of features: {len(features.columns)}")

Features created: ['consumer_id_ref', 'date_ref', 'sum___paid_value', 'sum___discount', 'mean___paid_value', 'mean___discount', 'stddev___paid_value', 'stddev___discount', 'min___paid_value', 'min___discount', 'max___paid_value', 'max___discount', 'kurtosis___paid_value', 'kurtosis___discount', 'skewness___paid_value', 'skewness___discount']
Number of features: 16

features.limit(5)

consumer_id_ref	date_ref	sum___paid_value	sum___discount	mean___paid_value	mean___discount	stddev___paid_value	stddev___discount	min___paid_value	min___discount	max___paid_value	max___discount	kurtosis___paid_value	kurtosis___discount	skewness___paid_value	skewness___discount
29	2022-01-01	null	null	null	null	null	null	null	null	null	null	null	null	null	null
29	2022-02-01	8853.40968120304	135	983.7121868003378	15.0	86.83003813511638	12.24744871391589	861.5299671745906	0	1126.4961447050034	25	-0.9476942065305236	-1.7343749999999998	0.07809048749090707	-0.3788861141556918
26	2022-02-01	11788.166930957694	95	1071.6515391779722	8.636363636363637	90.94906349189161	11.2006493318265	869.5921197423122	0	1165.640254915553	25	0.16024140568439948	-1.263311279143037	-1.1180188780284326	0.6930338494981549
26	2022-01-01	null	null	null	null	null	null	null	null	null	null	null	null	null	null
29	2022-03-01	7154.16997403849	120	1022.0242820054985	17.142857142857142	80.78779032712255	10.350983390135314	913.2907840738209	0	1114.7487402931952	25	-1.5832487921323395	-1.1907407407407404	-0.1951677930825345	-0.6211299937499416

numerical_in_categorical_groups example

features = make_features.run(
    df=df,
    suites=[
        "numerical_in_categorical_groups",
    ],
    options={},
)

print(f"Features created: {features.columns}")

print(f"Number of features: {len(features.columns)}")

Features created: ['consumer_id_ref', 'date_ref', 'product_type=g__sum___paid_value', 'product_type=g__sum___discount', 'product_type=g__mean___paid_value', 'product_type=g__mean___discount', 'product_type=g__stddev___paid_value', 'product_type=g__stddev___discount', 'product_type=g__min___paid_value', 'product_type=g__min___discount', 'product_type=g__max___paid_value', 'product_type=g__max___discount', 'product_type=g__kurtosis___paid_value', 'product_type=g__kurtosis___discount', 'product_type=g__skewness___paid_value', 'product_type=g__skewness___discount', 'product_type=f__sum___paid_value', 'product_type=f__sum___discount', 'product_type=f__mean___paid_value', 'product_type=f__mean___discount', 'product_type=f__stddev___paid_value', 'product_type=f__stddev___discount', 'product_type=f__min___paid_value', 'product_type=f__min___discount', 'product_type=f__max___paid_value', 'product_type=f__max___discount', 'product_type=f__kurtosis___paid_value', 'product_type=f__kurtosis___discount', 'product_type=f__skewness___paid_value', 'product_type=f__skewness___discount', 'product_type=e__sum___paid_value', 'product_type=e__sum___discount', 'product_type=e__mean___paid_value', 'product_type=e__mean___discount', 'product_type=e__stddev___paid_value', 'product_type=e__stddev___discount', 'product_type=e__min___paid_value', 'product_type=e__min___discount', 'product_type=e__max___paid_value', 'product_type=e__max___discount', 'product_type=e__kurtosis___paid_value', 'product_type=e__kurtosis___discount', 'product_type=e__skewness___paid_value', 'product_type=e__skewness___discount', 'product_type=h__sum___paid_value', 'product_type=h__sum___discount', 'product_type=h__mean___paid_value', 'product_type=h__mean___discount', 'product_type=h__stddev___paid_value', 'product_type=h__stddev___discount', 'product_type=h__min___paid_value', 'product_type=h__min___discount', 'product_type=h__max___paid_value', 'product_type=h__max___discount', 'product_type=h__kurtosis___paid_value', 'product_type=h__kurtosis___discount', 'product_type=h__skewness___paid_value', 'product_type=h__skewness___discount', 'product_type=d__sum___paid_value', 'product_type=d__sum___discount', 'product_type=d__mean___paid_value', 'product_type=d__mean___discount', 'product_type=d__stddev___paid_value', 'product_type=d__stddev___discount', 'product_type=d__min___paid_value', 'product_type=d__min___discount', 'product_type=d__max___paid_value', 'product_type=d__max___discount', 'product_type=d__kurtosis___paid_value', 'product_type=d__kurtosis___discount', 'product_type=d__skewness___paid_value', 'product_type=d__skewness___discount', 'product_type=c__sum___paid_value', 'product_type=c__sum___discount', 'product_type=c__mean___paid_value', 'product_type=c__mean___discount', 'product_type=c__stddev___paid_value', 'product_type=c__stddev___discount', 'product_type=c__min___paid_value', 'product_type=c__min___discount', 'product_type=c__max___paid_value', 'product_type=c__max___discount', 'product_type=c__kurtosis___paid_value', 'product_type=c__kurtosis___discount', 'product_type=c__skewness___paid_value', 'product_type=c__skewness___discount', 'product_type=b__sum___paid_value', 'product_type=b__sum___discount', 'product_type=b__mean___paid_value', 'product_type=b__mean___discount', 'product_type=b__stddev___paid_value', 'product_type=b__stddev___discount', 'product_type=b__min___paid_value', 'product_type=b__min___discount', 'product_type=b__max___paid_value', 'product_type=b__max___discount', 'product_type=b__kurtosis___paid_value', 'product_type=b__kurtosis___discount', 'product_type=b__skewness___paid_value', 'product_type=b__skewness___discount', 'product_type=a__sum___paid_value', 'product_type=a__sum___discount', 'product_type=a__mean___paid_value', 'product_type=a__mean___discount', 'product_type=a__stddev___paid_value', 'product_type=a__stddev___discount', 'product_type=a__min___paid_value', 'product_type=a__min___discount', 'product_type=a__max___paid_value', 'product_type=a__max___discount', 'product_type=a__kurtosis___paid_value', 'product_type=a__kurtosis___discount', 'product_type=a__skewness___paid_value', 'product_type=a__skewness___discount', 'buy_type=f_3__sum___paid_value', 'buy_type=f_3__sum___discount', 'buy_type=f_3__mean___paid_value', 'buy_type=f_3__mean___discount', 'buy_type=f_3__stddev___paid_value', 'buy_type=f_3__stddev___discount', 'buy_type=f_3__min___paid_value', 'buy_type=f_3__min___discount', 'buy_type=f_3__max___paid_value', 'buy_type=f_3__max___discount', 'buy_type=f_3__kurtosis___paid_value', 'buy_type=f_3__kurtosis___discount', 'buy_type=f_3__skewness___paid_value', 'buy_type=f_3__skewness___discount', 'buy_type=f_5__sum___paid_value', 'buy_type=f_5__sum___discount', 'buy_type=f_5__mean___paid_value', 'buy_type=f_5__mean___discount', 'buy_type=f_5__stddev___paid_value', 'buy_type=f_5__stddev___discount', 'buy_type=f_5__min___paid_value', 'buy_type=f_5__min___discount', 'buy_type=f_5__max___paid_value', 'buy_type=f_5__max___discount', 'buy_type=f_5__kurtosis___paid_value', 'buy_type=f_5__kurtosis___discount', 'buy_type=f_5__skewness___paid_value', 'buy_type=f_5__skewness___discount', 'buy_type=f_8__sum___paid_value', 'buy_type=f_8__sum___discount', 'buy_type=f_8__mean___paid_value', 'buy_type=f_8__mean___discount', 'buy_type=f_8__stddev___paid_value', 'buy_type=f_8__stddev___discount', 'buy_type=f_8__min___paid_value', 'buy_type=f_8__min___discount', 'buy_type=f_8__max___paid_value', 'buy_type=f_8__max___discount', 'buy_type=f_8__kurtosis___paid_value', 'buy_type=f_8__kurtosis___discount', 'buy_type=f_8__skewness___paid_value', 'buy_type=f_8__skewness___discount', 'buy_type=f_4__sum___paid_value', 'buy_type=f_4__sum___discount', 'buy_type=f_4__mean___paid_value', 'buy_type=f_4__mean___discount', 'buy_type=f_4__stddev___paid_value', 'buy_type=f_4__stddev___discount', 'buy_type=f_4__min___paid_value', 'buy_type=f_4__min___discount', 'buy_type=f_4__max___paid_value', 'buy_type=f_4__max___discount', 'buy_type=f_4__kurtosis___paid_value', 'buy_type=f_4__kurtosis___discount', 'buy_type=f_4__skewness___paid_value', 'buy_type=f_4__skewness___discount', 'buy_type=f_2__sum___paid_value', 'buy_type=f_2__sum___discount', 'buy_type=f_2__mean___paid_value', 'buy_type=f_2__mean___discount', 'buy_type=f_2__stddev___paid_value', 'buy_type=f_2__stddev___discount', 'buy_type=f_2__min___paid_value', 'buy_type=f_2__min___discount', 'buy_type=f_2__max___paid_value', 'buy_type=f_2__max___discount', 'buy_type=f_2__kurtosis___paid_value', 'buy_type=f_2__kurtosis___discount', 'buy_type=f_2__skewness___paid_value', 'buy_type=f_2__skewness___discount', 'buy_type=f_9__sum___paid_value', 'buy_type=f_9__sum___discount', 'buy_type=f_9__mean___paid_value', 'buy_type=f_9__mean___discount', 'buy_type=f_9__stddev___paid_value', 'buy_type=f_9__stddev___discount', 'buy_type=f_9__min___paid_value', 'buy_type=f_9__min___discount', 'buy_type=f_9__max___paid_value', 'buy_type=f_9__max___discount', 'buy_type=f_9__kurtosis___paid_value', 'buy_type=f_9__kurtosis___discount', 'buy_type=f_9__skewness___paid_value', 'buy_type=f_9__skewness___discount', 'buy_type=f_1__sum___paid_value', 'buy_type=f_1__sum___discount', 'buy_type=f_1__mean___paid_value', 'buy_type=f_1__mean___discount', 'buy_type=f_1__stddev___paid_value', 'buy_type=f_1__stddev___discount', 'buy_type=f_1__min___paid_value', 'buy_type=f_1__min___discount', 'buy_type=f_1__max___paid_value', 'buy_type=f_1__max___discount', 'buy_type=f_1__kurtosis___paid_value', 'buy_type=f_1__kurtosis___discount', 'buy_type=f_1__skewness___paid_value', 'buy_type=f_1__skewness___discount', 'buy_type=f_7__sum___paid_value', 'buy_type=f_7__sum___discount', 'buy_type=f_7__mean___paid_value', 'buy_type=f_7__mean___discount', 'buy_type=f_7__stddev___paid_value', 'buy_type=f_7__stddev___discount', 'buy_type=f_7__min___paid_value', 'buy_type=f_7__min___discount', 'buy_type=f_7__max___paid_value', 'buy_type=f_7__max___discount', 'buy_type=f_7__kurtosis___paid_value', 'buy_type=f_7__kurtosis___discount', 'buy_type=f_7__skewness___paid_value', 'buy_type=f_7__skewness___discount', 'buy_type=f_6__sum___paid_value', 'buy_type=f_6__sum___discount', 'buy_type=f_6__mean___paid_value', 'buy_type=f_6__mean___discount', 'buy_type=f_6__stddev___paid_value', 'buy_type=f_6__stddev___discount', 'buy_type=f_6__min___paid_value', 'buy_type=f_6__min___discount', 'buy_type=f_6__max___paid_value', 'buy_type=f_6__max___discount', 'buy_type=f_6__kurtosis___paid_value', 'buy_type=f_6__kurtosis___discount', 'buy_type=f_6__skewness___paid_value', 'buy_type=f_6__skewness___discount']
Number of features: 240

features.limit(5)

consumer_id_ref	date_ref	product_type=g__sum___paid_value	product_type=g__sum___discount	product_type=g__mean___paid_value	product_type=g__mean___discount	product_type=g__stddev___paid_value	product_type=g__stddev___discount	product_type=g__min___paid_value	product_type=g__min___discount	product_type=g__max___paid_value	product_type=g__max___discount	product_type=g__kurtosis___paid_value	product_type=g__kurtosis___discount	product_type=g__skewness___paid_value	product_type=g__skewness___discount	product_type=f__sum___paid_value	product_type=f__sum___discount	product_type=f__mean___paid_value	product_type=f__mean___discount	product_type=f__stddev___paid_value	product_type=f__stddev___discount	product_type=f__min___paid_value	product_type=f__min___discount	product_type=f__max___paid_value	product_type=f__max___discount	product_type=f__kurtosis___paid_value	product_type=f__kurtosis___discount	product_type=f__skewness___paid_value	product_type=f__skewness___discount	product_type=e__sum___paid_value	product_type=e__sum___discount	product_type=e__mean___paid_value	product_type=e__mean___discount	product_type=e__stddev___paid_value	product_type=e__stddev___discount	product_type=e__min___paid_value	product_type=e__min___discount	product_type=e__max___paid_value	product_type=e__max___discount	product_type=e__kurtosis___paid_value	product_type=e__kurtosis___discount	product_type=e__skewness___paid_value	product_type=e__skewness___discount	product_type=h__sum___paid_value	product_type=h__sum___discount	product_type=h__mean___paid_value	product_type=h__mean___discount	product_type=h__stddev___paid_value	product_type=h__stddev___discount	product_type=h__min___paid_value	product_type=h__min___discount	product_type=h__max___paid_value	product_type=h__max___discount	product_type=h__kurtosis___paid_value	product_type=h__kurtosis___discount	product_type=h__skewness___paid_value	product_type=h__skewness___discount	product_type=d__sum___paid_value	product_type=d__sum___discount	product_type=d__mean___paid_value	product_type=d__mean___discount	product_type=d__stddev___paid_value	product_type=d__stddev___discount	product_type=d__min___paid_value	product_type=d__min___discount	product_type=d__max___paid_value	product_type=d__max___discount	product_type=d__kurtosis___paid_value	product_type=d__kurtosis___discount	product_type=d__skewness___paid_value	product_type=d__skewness___discount	product_type=c__sum___paid_value	product_type=c__sum___discount	product_type=c__mean___paid_value	product_type=c__mean___discount	product_type=c__stddev___paid_value	product_type=c__stddev___discount	product_type=c__min___paid_value	product_type=c__min___discount	product_type=c__max___paid_value	product_type=c__max___discount	product_type=c__kurtosis___paid_value	product_type=c__kurtosis___discount	product_type=c__skewness___paid_value	product_type=c__skewness___discount	product_type=b__sum___paid_value	product_type=b__sum___discount	product_type=b__mean___paid_value	product_type=b__mean___discount	product_type=b__stddev___paid_value	product_type=b__stddev___discount	product_type=b__min___paid_value	product_type=b__min___discount	product_type=b__max___paid_value	product_type=b__max___discount	product_type=b__kurtosis___paid_value	product_type=b__kurtosis___discount	product_type=b__skewness___paid_value	product_type=b__skewness___discount	product_type=a__sum___paid_value	product_type=a__sum___discount	product_type=a__mean___paid_value	product_type=a__mean___discount	product_type=a__stddev___paid_value	product_type=a__stddev___discount	product_type=a__min___paid_value	product_type=a__min___discount	product_type=a__max___paid_value	product_type=a__max___discount	product_type=a__kurtosis___paid_value	product_type=a__kurtosis___discount	product_type=a__skewness___paid_value	product_type=a__skewness___discount	buy_type=f_3__sum___paid_value	buy_type=f_3__sum___discount	buy_type=f_3__mean___paid_value	buy_type=f_3__mean___discount	buy_type=f_3__stddev___paid_value	buy_type=f_3__stddev___discount	buy_type=f_3__min___paid_value	buy_type=f_3__min___discount	buy_type=f_3__max___paid_value	buy_type=f_3__max___discount	buy_type=f_3__kurtosis___paid_value	buy_type=f_3__kurtosis___discount	buy_type=f_3__skewness___paid_value	buy_type=f_3__skewness___discount	buy_type=f_5__sum___paid_value	buy_type=f_5__sum___discount	buy_type=f_5__mean___paid_value	buy_type=f_5__mean___discount	buy_type=f_5__stddev___paid_value	buy_type=f_5__stddev___discount	buy_type=f_5__min___paid_value	buy_type=f_5__min___discount	buy_type=f_5__max___paid_value	buy_type=f_5__max___discount	buy_type=f_5__kurtosis___paid_value	buy_type=f_5__kurtosis___discount	buy_type=f_5__skewness___paid_value	buy_type=f_5__skewness___discount	buy_type=f_8__sum___paid_value	buy_type=f_8__sum___discount	buy_type=f_8__mean___paid_value	buy_type=f_8__mean___discount	buy_type=f_8__stddev___paid_value	buy_type=f_8__stddev___discount	buy_type=f_8__min___paid_value	buy_type=f_8__min___discount	buy_type=f_8__max___paid_value	buy_type=f_8__max___discount	buy_type=f_8__kurtosis___paid_value	buy_type=f_8__kurtosis___discount	buy_type=f_8__skewness___paid_value	buy_type=f_8__skewness___discount	buy_type=f_4__sum___paid_value	buy_type=f_4__sum___discount	buy_type=f_4__mean___paid_value	buy_type=f_4__mean___discount	buy_type=f_4__stddev___paid_value	buy_type=f_4__stddev___discount	buy_type=f_4__min___paid_value	buy_type=f_4__min___discount	buy_type=f_4__max___paid_value	buy_type=f_4__max___discount	buy_type=f_4__kurtosis___paid_value	buy_type=f_4__kurtosis___discount	buy_type=f_4__skewness___paid_value	buy_type=f_4__skewness___discount	buy_type=f_2__sum___paid_value	buy_type=f_2__sum___discount	buy_type=f_2__mean___paid_value	buy_type=f_2__mean___discount	buy_type=f_2__stddev___paid_value	buy_type=f_2__stddev___discount	buy_type=f_2__min___paid_value	buy_type=f_2__min___discount	buy_type=f_2__max___paid_value	buy_type=f_2__max___discount	buy_type=f_2__kurtosis___paid_value	buy_type=f_2__kurtosis___discount	buy_type=f_2__skewness___paid_value	buy_type=f_2__skewness___discount	buy_type=f_9__sum___paid_value	buy_type=f_9__sum___discount	buy_type=f_9__mean___paid_value	buy_type=f_9__mean___discount	buy_type=f_9__stddev___paid_value	buy_type=f_9__stddev___discount	buy_type=f_9__min___paid_value	buy_type=f_9__min___discount	buy_type=f_9__max___paid_value	buy_type=f_9__max___discount	buy_type=f_9__kurtosis___paid_value	buy_type=f_9__kurtosis___discount	buy_type=f_9__skewness___paid_value	buy_type=f_9__skewness___discount	buy_type=f_1__sum___paid_value	buy_type=f_1__sum___discount	buy_type=f_1__mean___paid_value	buy_type=f_1__mean___discount	buy_type=f_1__stddev___paid_value	buy_type=f_1__stddev___discount	buy_type=f_1__min___paid_value	buy_type=f_1__min___discount	buy_type=f_1__max___paid_value	buy_type=f_1__max___discount	buy_type=f_1__kurtosis___paid_value	buy_type=f_1__kurtosis___discount	buy_type=f_1__skewness___paid_value	buy_type=f_1__skewness___discount	buy_type=f_7__sum___paid_value	buy_type=f_7__sum___discount	buy_type=f_7__mean___paid_value	buy_type=f_7__mean___discount	buy_type=f_7__stddev___paid_value	buy_type=f_7__stddev___discount	buy_type=f_7__min___paid_value	buy_type=f_7__min___discount	buy_type=f_7__max___paid_value	buy_type=f_7__max___discount	buy_type=f_7__kurtosis___paid_value	buy_type=f_7__kurtosis___discount	buy_type=f_7__skewness___paid_value	buy_type=f_7__skewness___discount	buy_type=f_6__sum___paid_value	buy_type=f_6__sum___discount	buy_type=f_6__mean___paid_value	buy_type=f_6__mean___discount	buy_type=f_6__stddev___paid_value	buy_type=f_6__stddev___discount	buy_type=f_6__min___paid_value	buy_type=f_6__min___discount	buy_type=f_6__max___paid_value	buy_type=f_6__max___discount	buy_type=f_6__kurtosis___paid_value	buy_type=f_6__kurtosis___discount	buy_type=f_6__skewness___paid_value	buy_type=f_6__skewness___discount
29	2022-01-01	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null
29	2022-02-01	978.1700967289302	25	978.1700967289302	25.0	null	null	978.1700967289302	25	978.1700967289302	25	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	2005.3660480401836	50	1002.6830240200918	25.0	175.09819447233872	0.0	878.8699033351803	25	1126.4961447050034	25	-2.0000000000000004	null	0.0	null	1908.3430273105937	10	954.1715136552968	5.0	48.71265890243976	7.0710678118654755	919.7264622157545	0	988.6165650948393	10	-1.9999999999999996	-2.0	0.0	0.0	null	null	null	null	null	null	null	null	null	null	null	null	null	null	1926.6506829967452	25	963.3253414983726	12.5	143.96039895553838	17.67766952966369	861.5299671745906	0	1065.1207158221546	25	-1.9999999999999991	-2.0	0.0	0.0	null	null	null	null	null	null	null	null	null	null	null	null	null	null	2034.879826126586	25	1017.439913063293	12.5	30.92859641042705	17.67766952966369	995.5700928088982	0	1039.309733317688	25	-2.000000000000001	-2.0	0.0	0.0	1065.1207158221546	25	1065.1207158221546	25.0	null	null	1065.1207158221546	25	1065.1207158221546	25	null	null	null	null	4811.646288406451	75	962.3292576812903	15.0	63.36864552021739	13.693063937629153	878.8699033351803	0	1039.309733317688	25	-1.2904340099690859	-1.8333333333333335	-0.1776892000819694	-0.4082482904638631	null	null	null	null	null	null	null	null	null	null	null	null	null	null	988.6165650948393	10	988.6165650948393	10.0	null	null	988.6165650948393	10	988.6165650948393	10	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	1988.026111879594	25	994.013055939797	12.5	187.35938091683352	17.67766952966369	861.5299671745906	0	1126.4961447050034	25	-2.0	-2.0	0.0	0.0	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null
26	2022-02-01	4361.685585232964	35	1090.421396308241	8.75	89.23134611680102	11.814539065631521	961.9595311462253	0	1165.640254915553	25	-0.8498265554253606	-1.1419024281577181	-0.8841013532743225	0.6892544771146777	1148.7468889856796	25	1148.7468889856796	25.0	null	null	1148.7468889856796	25	1148.7468889856796	25	null	null	null	null	869.5921197423122	25	869.5921197423122	25.0	null	null	869.5921197423122	25	869.5921197423122	25	null	null	null	null	2168.860067298752	0	1084.430033649376	0.0	52.41536136399558	0.0	1047.3667761905513	0	1121.4932911082005	0	-2.000000000000001	null	0.0	null	1009.6418154519044	10	1009.6418154519044	10.0	null	null	1009.6418154519044	10	1009.6418154519044	10	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	1113.5043630627235	0	1113.5043630627235	0.0	null	null	1113.5043630627235	0	1113.5043630627235	0	null	null	null	null	1116.136091183358	0	1116.136091183358	0.0	null	null	1116.136091183358	0	1116.136091183358	0	null	null	null	null	869.5921197423122	25	869.5921197423122	25.0	null	null	869.5921197423122	25	869.5921197423122	25	null	null	null	null	1113.5043630627235	0	1113.5043630627235	0.0	null	null	1113.5043630627235	0	1113.5043630627235	0	null	null	null	null	961.9595311462253	0	961.9595311462253	0.0	null	null	961.9595311462253	0	961.9595311462253	0	null	null	null	null	1104.6138403921893	10	1104.6138403921893	10.0	null	null	1104.6138403921893	10	1104.6138403921893	10	null	null	null	null	3342.4789898851004	25	1114.1596632950334	8.333333333333334	60.605311130044484	14.433756729740644	1047.3667761905513	0	1165.640254915553	25	-1.5	-1.5000000000000004	-0.4345291137240896	0.7071067811865478	2264.8829801690376	25	1132.4414900845188	12.5	23.059316265924885	17.67766952966369	1116.136091183358	0	1148.7468889856796	25	-1.999999999999999	-2.0	0.0	0.0	null	null	null	null	null	null	null	null	null	null	null	null	null	null	1009.6418154519044	10	1009.6418154519044	10.0	null	null	1009.6418154519044	10	1009.6418154519044	10	null	null	null	null	1121.4932911082005	0	1121.4932911082005	0.0	null	null	1121.4932911082005	0	1121.4932911082005	0	null	null	null	null
26	2022-01-01	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null
29	2022-03-01	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	1114.7487402931952	25	1114.7487402931952	25.0	null	null	1114.7487402931952	25	1114.7487402931952	25	null	null	null	null	2017.6059546732924	35	1008.8029773366462	17.5	60.73613126509213	10.606601717798213	965.8560470560633	10	1051.7499076172292	25	-2.0000000000000036	-2.0	3.674123667735640...	0.0	null	null	null	null	null	null	null	null	null	null	null	null	null	null	1065.1207158221546	25	1065.1207158221546	25.0	null	null	1065.1207158221546	25	1065.1207158221546	25	null	null	null	null	2956.6945632498478	35	985.564854416616	11.666666666666666	101.68215577844472	12.583057392117917	913.2907840738209	0	1101.8368062649242	25	-1.4999999999999991	-1.4999999999999998	0.6460883288015307	0.23906314692954492	null	null	null	null	null	null	null	null	null	null	null	null	null	null	2116.870623439384	50	1058.435311719692	25.0	9.454589151647497	0.0	1051.7499076172292	25	1065.1207158221546	25	-1.9999999999999976	null	-3.80475937533656...	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	1101.8368062649242	0	1101.8368062649242	0.0	null	null	1101.8368062649242	0	1101.8368062649242	0	null	null	null	null	1879.146831129884	35	939.573415564942	17.5	37.16925390959781	10.606601717798213	913.2907840738209	10	965.8560470560633	25	-1.9999999999999996	-2.0	0.0	0.0	null	null	null	null	null	null	null	null	null	null	null	null	null	null	1114.7487402931952	25	1114.7487402931952	25.0	null	null	1114.7487402931952	25	1114.7487402931952	25	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	941.5669729111025	10	941.5669729111025	10.0	null	null	941.5669729111025	10	941.5669729111025	10	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null

correlation example

features = make_features.run(
    df=df,
    suites=[
        "correlation",
    ],
    options={},
)

print(f"Features created: {features.columns}")

print(f"Number of features: {len(features.columns)}")

Features created: ['consumer_id_ref', 'date_ref', 'corr_between___paid_value_discount']
Number of features: 3

features.limit(5)

consumer_id_ref	date_ref	corr_between___paid_value_discount
29	2022-01-01	null
29	2022-02-01	0.5035732318016096
26	2022-02-01	-0.10453224926952913
26	2022-01-01	null
29	2022-03-01	0.010776511903292301

features = make_features.run(
    df=df,
    suites=[
        "categorical_statistics",
    ],
    options={},
)

print(f"Features created: {features.columns}")

print(f"Number of features: {len(features.columns)}")

Features created: ['consumer_id_ref', 'date_ref', 'product_type=g__count___product_type', 'product_type=f__count___product_type', 'product_type=e__count___product_type', 'product_type=h__count___product_type', 'product_type=d__count___product_type', 'product_type=c__count___product_type', 'product_type=b__count___product_type', 'product_type=a__count___product_type', 'buy_type=f_3__count___buy_type', 'buy_type=f_5__count___buy_type', 'buy_type=f_8__count___buy_type', 'buy_type=f_4__count___buy_type', 'buy_type=f_2__count___buy_type', 'buy_type=f_9__count___buy_type', 'buy_type=f_1__count___buy_type', 'buy_type=f_7__count___buy_type', 'buy_type=f_6__count___buy_type', 'count___product_type', 'countDistinct___product_type', 'count___buy_type', 'countDistinct___buy_type']
Number of features: 23

features.limit(5)

consumer_id_ref	date_ref	product_type=g__count___product_type	product_type=f__count___product_type	product_type=e__count___product_type	product_type=h__count___product_type	product_type=d__count___product_type	product_type=c__count___product_type	product_type=b__count___product_type	product_type=a__count___product_type	buy_type=f_3__count___buy_type	buy_type=f_5__count___buy_type	buy_type=f_8__count___buy_type	buy_type=f_4__count___buy_type	buy_type=f_2__count___buy_type	buy_type=f_9__count___buy_type	buy_type=f_1__count___buy_type	buy_type=f_7__count___buy_type	buy_type=f_6__count___buy_type	count___product_type	countDistinct___product_type	count___buy_type	countDistinct___buy_type
70	2022-02-01	3	0	3	1	1	1	1	1	3	1	1	3	1	1	0	0	1	11	7	11	7
46	2022-03-01	3	1	0	0	0	1	2	0	2	2	0	0	0	1	1	1	0	7	4	7	5
55	2022-03-01	2	0	1	3	2	1	0	1	2	0	1	1	1	0	0	4	1	10	6	10	6
1	2022-04-01	1	1	2	2	1	1	1	1	2	0	1	2	1	0	1	1	2	10	8	10	7
80	2022-11-01	0	3	2	0	1	1	0	0	2	0	1	1	0	1	0	1	1	7	4	7	6

first_observation_features example

features = make_features.run(
    df=df,
    suites=[
        "first_observation_features",
    ],
    options={},
)

print(f"Features created: {features.columns}")

print(f"Number of features: {len(features.columns)}")

Features created: ['consumer_id_ref', 'date_ref', 'first___paid_value', 'first___discount']
Number of features: 4

features.limit(5)

consumer_id_ref	date_ref	first___paid_value	first___discount
0	2022-01-01	null	null
0	2022-02-01	1160.6002074014891	0
0	2022-03-01	907.0417846729197	25
0	2022-04-01	1146.7341607305582	10
0	2022-05-01	1112.5181517372503	10

last_observation_features example

features = make_features.run(
    df=df,
    suites=[
        "last_observation_features",
    ],
    options={},
)

print(f"Features created: {features.columns}")

print(f"Number of features: {len(features.columns)}")

Features created: ['consumer_id_ref', 'date_ref', 'last___paid_value', 'last___discount']
Number of features: 4

features.limit(5)

consumer_id_ref	date_ref	last___paid_value	last___discount
0	2022-01-01	null	null
0	2022-02-01	1017.1085418841382	25
0	2022-03-01	1014.1861518669832	10
0	2022-04-01	1232.613528271461	25
0	2022-05-01	1014.6397369752315	25

lags example

features = make_features.run(
    df=df,
    suites=[
        "numerical_statistics",
        "lags",
    ],
    options={"n_lags": [1, 3, 4, 5]},
)

print(f"Features created: {features.columns}")

print(f"Number of features: {len(features.columns)}")

Features created: ['consumer_id_ref', 'date_ref', 'sum___paid_value', 'sum___discount', 'mean___paid_value', 'mean___discount', 'stddev___paid_value', 'stddev___discount', 'min___paid_value', 'min___discount', 'max___paid_value', 'max___discount', 'kurtosis___paid_value', 'kurtosis___discount', 'skewness___paid_value', 'skewness___discount', 'lag=1_sum___paid_value', 'lag=1_sum___discount', 'lag=1_mean___paid_value', 'lag=1_mean___discount', 'lag=1_stddev___paid_value', 'lag=1_stddev___discount', 'lag=1_min___paid_value', 'lag=1_min___discount', 'lag=1_max___paid_value', 'lag=1_max___discount', 'lag=1_kurtosis___paid_value', 'lag=1_kurtosis___discount', 'lag=1_skewness___paid_value', 'lag=1_skewness___discount', 'lag=3_sum___paid_value', 'lag=3_sum___discount', 'lag=3_mean___paid_value', 'lag=3_mean___discount', 'lag=3_stddev___paid_value', 'lag=3_stddev___discount', 'lag=3_min___paid_value', 'lag=3_min___discount', 'lag=3_max___paid_value', 'lag=3_max___discount', 'lag=3_kurtosis___paid_value', 'lag=3_kurtosis___discount', 'lag=3_skewness___paid_value', 'lag=3_skewness___discount', 'lag=4_sum___paid_value', 'lag=4_sum___discount', 'lag=4_mean___paid_value', 'lag=4_mean___discount', 'lag=4_stddev___paid_value', 'lag=4_stddev___discount', 'lag=4_min___paid_value', 'lag=4_min___discount', 'lag=4_max___paid_value', 'lag=4_max___discount', 'lag=4_kurtosis___paid_value', 'lag=4_kurtosis___discount', 'lag=4_skewness___paid_value', 'lag=4_skewness___discount', 'lag=5_sum___paid_value', 'lag=5_sum___discount', 'lag=5_mean___paid_value', 'lag=5_mean___discount', 'lag=5_stddev___paid_value', 'lag=5_stddev___discount', 'lag=5_min___paid_value', 'lag=5_min___discount', 'lag=5_max___paid_value', 'lag=5_max___discount', 'lag=5_kurtosis___paid_value', 'lag=5_kurtosis___discount', 'lag=5_skewness___paid_value', 'lag=5_skewness___discount']
Number of features: 72

features.limit(5)

consumer_id_ref	date_ref	sum___paid_value	sum___discount	mean___paid_value	mean___discount	stddev___paid_value	stddev___discount	min___paid_value	min___discount	max___paid_value	max___discount	kurtosis___paid_value	kurtosis___discount	skewness___paid_value	skewness___discount	lag=1_sum___paid_value	lag=1_sum___discount	lag=1_mean___paid_value	lag=1_mean___discount	lag=1_stddev___paid_value	lag=1_stddev___discount	lag=1_min___paid_value	lag=1_min___discount	lag=1_max___paid_value	lag=1_max___discount	lag=1_kurtosis___paid_value	lag=1_kurtosis___discount	lag=1_skewness___paid_value	lag=1_skewness___discount	lag=3_sum___paid_value	lag=3_sum___discount	lag=3_mean___paid_value	lag=3_mean___discount	lag=3_stddev___paid_value	lag=3_stddev___discount	lag=3_min___paid_value	lag=3_min___discount	lag=3_max___paid_value	lag=3_max___discount	lag=3_kurtosis___paid_value	lag=3_kurtosis___discount	lag=3_skewness___paid_value	lag=3_skewness___discount	lag=4_sum___paid_value	lag=4_sum___discount	lag=4_mean___paid_value	lag=4_mean___discount	lag=4_stddev___paid_value	lag=4_stddev___discount	lag=4_min___paid_value	lag=4_min___discount	lag=4_max___paid_value	lag=4_max___discount	lag=4_kurtosis___paid_value	lag=4_kurtosis___discount	lag=4_skewness___paid_value	lag=4_skewness___discount	lag=5_sum___paid_value	lag=5_sum___discount	lag=5_mean___paid_value	lag=5_mean___discount	lag=5_stddev___paid_value	lag=5_stddev___discount	lag=5_min___paid_value	lag=5_min___discount	lag=5_max___paid_value	lag=5_max___discount	lag=5_kurtosis___paid_value	lag=5_kurtosis___discount	lag=5_skewness___paid_value	lag=5_skewness___discount
0	2022-01-01	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null
0	2022-02-01	6425.02303262522	45	1070.8371721042033	7.5	70.58560769477131	9.874208829065749	991.1823185540896	0	1160.6002074014891	25	-1.4862128264472743	-0.3905325443786989	0.33803836307444013	0.9387234089965415	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null
0	2022-03-01	1921.2279365399029	35	960.6139682699514	17.5	75.76250860886378	10.606601717798213	907.0417846729197	10	1014.1861518669832	25	-2.000000000000001	-2.0	-3.78583866001275...	0.0	6425.02303262522	45	1070.8371721042033	7.5	70.58560769477131	9.874208829065749	991.1823185540896	0	1160.6002074014891	25	-1.4862128264472743	-0.3905325443786989	0.33803836307444013	0.9387234089965415	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null
0	2022-04-01	7241.335698029857	90	1034.4765282899796	12.857142857142858	121.03820134423728	9.063269671749657	913.4362107269889	0	1232.613528271461	25	-1.0770315605844292	-0.9517958412098295	0.5482194738419417	0.3293992662501247	1921.2279365399029	35	960.6139682699514	17.5	75.76250860886378	10.606601717798213	907.0417846729197	10	1014.1861518669832	25	-2.000000000000001	-2.0	-3.78583866001275...	0.0	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null
0	2022-05-01	13005.825538275483	155	1000.4481183288833	11.923076923076923	78.80920113421799	11.46343126687016	848.2943132152986	0	1112.5181517372503	25	-0.6377760480965216	-1.7071386079714457	-0.639341122500975	0.1545915837587542	7241.335698029857	90	1034.4765282899796	12.857142857142858	121.03820134423728	9.063269671749657	913.4362107269889	0	1232.613528271461	25	-1.0770315605844292	-0.9517958412098295	0.5482194738419417	0.3293992662501247	6425.02303262522	45	1070.8371721042033	7.5	70.58560769477131	9.874208829065749	991.1823185540896	0	1160.6002074014891	25	-1.4862128264472743	-0.3905325443786989	0.33803836307444013	0.9387234089965415	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null

increase_rate example

features = make_features.run(
    df=df,
    suites=[
        "numerical_statistics",
        "increase_rate",
    ],
    options={},
)

print(f"Features created: {features.columns}")

print(f"Number of features: {len(features.columns)}")

Features created: ['consumer_id_ref', 'date_ref', 'sum___paid_value', 'sum___discount', 'mean___paid_value', 'mean___discount', 'stddev___paid_value', 'stddev___discount', 'min___paid_value', 'min___discount', 'max___paid_value', 'max___discount', 'kurtosis___paid_value', 'kurtosis___discount', 'skewness___paid_value', 'skewness___discount', 'increase_rate_sum___paid_value', 'increase_rate_sum___discount', 'increase_rate_mean___paid_value', 'increase_rate_mean___discount', 'increase_rate_stddev___paid_value', 'increase_rate_stddev___discount', 'increase_rate_min___paid_value', 'increase_rate_min___discount', 'increase_rate_max___paid_value', 'increase_rate_max___discount', 'increase_rate_kurtosis___paid_value', 'increase_rate_kurtosis___discount', 'increase_rate_skewness___paid_value', 'increase_rate_skewness___discount']
Number of features: 30

features.limit(5)

consumer_id_ref	date_ref	sum___paid_value	sum___discount	mean___paid_value	mean___discount	stddev___paid_value	stddev___discount	min___paid_value	min___discount	max___paid_value	max___discount	kurtosis___paid_value	kurtosis___discount	skewness___paid_value	skewness___discount	increase_rate_sum___paid_value	increase_rate_sum___discount	increase_rate_mean___paid_value	increase_rate_mean___discount	increase_rate_stddev___paid_value	increase_rate_stddev___discount	increase_rate_min___paid_value	increase_rate_min___discount	increase_rate_max___paid_value	increase_rate_max___discount	increase_rate_kurtosis___paid_value	increase_rate_kurtosis___discount	increase_rate_skewness___paid_value	increase_rate_skewness___discount
0	2022-01-01	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null	null
0	2022-02-01	6425.02303262522	45	1070.8371721042033	7.5	70.58560769477131	9.874208829065749	991.1823185540896	0	1160.6002074014891	25	-1.4862128264472743	-0.3905325443786989	0.33803836307444013	0.9387234089965415	null	null	null	null	null	null	null	null	null	null	null	null	null	null
0	2022-03-01	1921.2279365399029	35	960.6139682699514	17.5	75.76250860886378	10.606601717798213	907.0417846729197	10	1014.1861518669832	25	-2.000000000000001	-2.0	-3.78583866001275...	0.0	-0.7009772685974478	-0.2222222222222222	-0.10293180579234323	1.3333333333333333	0.07334215972863196	0.07417231105914944	-0.08488905855777566	null	-0.12615373890232	0.0	0.3457022873237557	4.121212121212112	-1.000000000000001	-1.0
0	2022-04-01	7241.335698029857	90	1034.4765282899796	12.857142857142858	121.03820134423728	9.063269671749657	913.4362107269889	0	1232.613528271461	25	-1.0770315605844292	-0.9517958412098295	0.5482194738419417	0.3293992662501247	2.7691184686140744	1.5714285714285714	0.0768909910325927	-0.26530612244897955	0.5976002321823396	-0.14550674071779243	0.007049759076286682	-1.0	0.21537207543445727	0.0	-0.46148421970778564	-0.5241020793950852	-1.44807933743297...	null
0	2022-05-01	13005.825538275483	155	1000.4481183288833	11.923076923076923	78.80920113421799	11.46343126687016	848.2943132152986	0	1112.5181517372503	25	-0.6377760480965216	-1.7071386079714457	-0.639341122500975	0.1545915837587542	0.796053391339662	0.7222222222222222	-0.03289432774018...	-0.07264957264957267	-0.3488898524682997	0.264822926167787	-0.07131521254214879	null	-0.09743149314824164	0.0	-0.40783903514355196	0.7935974649789376	-2.1662138121808217	-0.5306863141541813