dowhy 0.11.1

DoWhy is a Python library for causal inference that supports explicit modeling and testing of causal assumptions

Stars: 7057, Watchers: 7057, Forks: 924, Open Issues: 135

The py-why/dowhy repo was created 6 years ago and the last code push was 4 days ago.
The project is extremely popular with a mindblowing 7057 github stars!

How to Install dowhy

You can install dowhy using pip

pip install dowhy

or add it to a project with poetry

poetry add dowhy

Package Details

Author

PyWhy Community

License

MIT

Homepage

https://github.com/py-why/dowhy

PyPi:

https://pypi.org/project/dowhy/

Documentation:

https://py-why.github.io/dowhy

GitHub Repo:

https://github.com/microsoft/dowhy

Classifiers

Related Packages

Errors

A list of common dowhy errors.

Code Examples

Here are some dowhy code examples and snippets.

GitHub Issues

The dowhy package has 135 open issues on GitHub

• Member Rewards Program example breaks at estimate step
• CausalModel.estimate_effect - UnboundLocalError: local variable 'identifier_name' referenced before assignment
• method_params["init_params"] for dowhy core estimators
• Support use of arbitrary external estimators
• Plugging in my own estimators into CausalModel.estimate_effect()
• AttributeError: 'CausalEstimate' object has no attribute '_estimator_object'
• Visualization of causal graphs
• Is the translation from causal graph to EconML Double Machine Learning notation incorrect?
• Is there possible that estimate_conditional_effects() generate output dataframe more than 1000 rows?
• STOP: TOTAL NO. of ITERATIONS REACHED LIMIT. in backdoor.propensity_score_stratification
• What is beta parameter in datasets?
• Model creation fails when there are no common causes.
• Conditional Average Treatment Effects (CATE) with DoWhy and EconML
• Include conditional effects, p-value and confidence intervals for DML
• Pin networkx upper bound to avoid installtion of matplotlib

See more issues on GitHub