Skip to main content

No project description provided

Project description

pycalf: Causal Inference Library

Documentation Status PyPI Latest Release codecov Code style: black License: MIT

A library to make causal inference easier by using Python

Getting Started


pip install pycalf


Inference with IPW

from pycalf import metrics
from pycalf.propensity_score import IPW

# Load Data and Define Variables
df = pd.read_csv('sample/q_data_x.csv')
covariate_cols = [
    'TVwatch_day', 'age', 'sex', 'marry_dummy', 'child_dummy', 'inc', 'pmoney',
    'area_kanto', 'area_tokai', 'area_keihanshin', 'job_dummy1', 'job_dummy2',
    'job_dummy3', 'job_dummy4', 'job_dummy5', 'job_dummy6', 'job_dummy7',
    'fam_str_dummy1', 'fam_str_dummy2', 'fam_str_dummy3', 'fam_str_dummy4'
outcome_cols = ['gamecount', 'gamedummy', 'gamesecond']
treatment_col = 'cm_dummy'

X = df[covariate_cols]
y = df[outcome_cols]
treatment = df[treatment_col].astype(bool).to_numpy()

# Define and Fit IPW Model.
learner = Pipeline([
    ('sclaer', preprocessing.MinMaxScaler()),
    ('clf', LogisticRegression(solver='lbfgs', max_iter=1000, random_state=42))
model = IPW(learner), treatment)

# metrics
print('F1 Score: ', metrics.f1_score(treatment, model.get_score(), threshold='auto'))
metrics.plot_roc_curve(treatment, model.get_score())
metrics.plot_probability_distribution(treatment, model.get_score())

# Estimate ATE
outcome_name = 'gamesecond'
z0, z1, treat_effect = model.estimate_effect(
    X, treatment, y[outcome_name].to_numpy().reshape(-1, 1), mode='ate')
metrics.plot_treatment_effect(outcome_name, z0[0], z1[0], treat_effect[0].round())

Effect size d.

Propensity Score Distribution

Average Treatment Effect

Export requirements.txt

poetry export -f requirements.txt --output docs/requirements.txt


Please read for details on our code of conduct, and the process for submitting pull requests to us.



Project details

Download files

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

Source Distribution

pycalf-0.2.12.tar.gz (225.0 kB view hashes)

Uploaded Source

Built Distribution

pycalf-0.2.12-py3-none-any.whl (88.5 kB view hashes)

Uploaded Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page