Skip to main content
Join the official 2020 Python Developers SurveyStart the survey!

TSFresh primitives for featuretools

Project description

Featuretools TSFresh Primitives


pip install 'featuretools[tsfresh]'

Calculating Features

In tsfresh, this is how to calculate a feature.

from tsfresh.feature_extraction.feature_calculators import agg_autocorrelation

data = list(range(10))
param = [{'f_agg': 'mean', 'maxlag': 5}]
agg_autocorrelation(data, param=param)
[('f_agg_"mean"__maxlag_5', 0.1717171717171717)]

With tsfresh primtives in featuretools, this is how to calculate the same feature.

from featuretools.tsfresh import AggAutocorrelation

data = list(range(10))
AggAutocorrelation(f_agg='mean', maxlag=5)(data)

Combining Primitives

In featuretools, this is how to combine tsfresh primitives with built-in or other installed primitives.

import featuretools as ft
from featuretools.tsfresh import AggAutocorrelation, Mean

entityset = ft.demo.load_mock_customer(return_entityset=True)
agg_primitives = [Mean, AggAutocorrelation(f_agg='mean', maxlag=5)]
feature_matrix, features = ft.dfs(entityset=entityset, target_entity='sessions', agg_primitives=agg_primitives)

    'AGG_AUTOCORRELATION(transactions.amount, f_agg=mean, maxlag=5)',
            MEAN(transactions.amount)  AGG_AUTOCORRELATION(transactions.amount, f_agg=mean, maxlag=5)
1                           76.813125                                           0.044268
2                           74.696000                                          -0.053110
3                           88.600000                                           0.007520
4                           64.557200                                          -0.034542
5                           70.638182                                          -0.100571

Notice that tsfresh primtives are applied across relationships in an entityset generating many features that are otherwise not possible.

feature_matrix[['customers.AGG_AUTOCORRELATION(transactions.amount, f_agg=mean, maxlag=5)']].head()
            customers.AGG_AUTOCORRELATION(transactions.amount, f_agg=mean, maxlag=5)
1                                                    0.011102
2                                                   -0.001686
3                                                   -0.010679
4                                                    0.011204
5                                                   -0.010679

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Files for featuretools-tsfresh-primitives, version 0.4.0
Filename, size File type Python version Upload date Hashes
Filename, size featuretools_tsfresh_primitives-0.4.0-py3-none-any.whl (53.0 kB) File type Wheel Python version py3 Upload date Hashes View
Filename, size featuretools_tsfresh_primitives-0.4.0.tar.gz (18.9 kB) File type Source Python version None Upload date Hashes View

Supported by

Pingdom Pingdom Monitoring Google Google Object Storage and Download Analytics Sentry Sentry Error logging AWS AWS Cloud computing DataDog DataDog Monitoring Fastly Fastly CDN DigiCert DigiCert EV certificate StatusPage StatusPage Status page