{"categories":[{"label":"Scientific/Engineering","url":"https://skillfed.io/packages/category/scientific-engineering/6"}],"enrichment":{"capability":"Builds arbitrary contrasts for statistical models defined using the formulaic formula system, enabling flexible comparison coding for categorical variables in model specifications.","skillfed_tags":["statistical-modeling","contrasts","formula-system"],"use_cases":["Specify custom contrast schemes (e.g., treatment, sum, Helmert) for categorical variables in regression or generalized linear models.","Build multi-condition comparisons in biological or scientific models where standard contrast coding is insufficient.","Integrate contrast specification into formulaic-based model definitions for reproducible statistical workflows.","Develop custom statistical modeling libraries that need flexible contrast matrix generation.","Encode categorical predictors with domain-specific comparison logic in machine learning pipelines."],"what_it_does":"formulaic-contrasts extends the formulaic formula system with tools to build custom contrast matrices for categorical variables in statistical models. It provides functions like cond() to specify arbitrary comparison schemes, enabling researchers to define exactly how categorical predictors are encoded when fitting models. The package is designed for use within the scverse ecosystem and statistical modeling workflows where fine-grained control over contrast coding is needed.\n\nThe package depends on formulaic for formula parsing, pandas for data handling, and session-info for environment tracking. It targets Python 3.10+ and is actively maintained. Installation is straightforward via pip, and the library is intended to be used both directly by developers building custom models and indirectly as a dependency of higher-level statistical packages.","worth_installing":"Yes, if you use formulaic-based models and need custom contrast coding. The package is actively maintained, has no known vulnerabilities, low install friction, and a permissive license. It is niche\u2014primarily useful for statistical modeling and research workflows\u2014but well-suited to that purpose. Not necessary for general-purpose Python development."},"id":"formulaic-contrasts","links":{"html":"https://skillfed.io/packages/formulaic-contrasts","md":"https://skillfed.io/packages/formulaic-contrasts.md","pypi":"https://pypi.org/project/formulaic-contrasts/"},"maintenance":{"status":"active"},"meta":{"latest_release":"2024-12-15","license_spdx":null,"license_treatment":"permissive","name":"formulaic-contrasts","python_support":"supports_current","summary":"Build contrasts for models defined with formulaic"},"popularity":{"monthly_downloads":159637,"position":10692,"tier":"top_15000"},"security":{"n_vulnerabilities":0},"version":"1.0.0"}