Writing productivity
productivity 4 min read 3 January 2022

Python Libraries for Data Science: A Practical Reference

Curated Python libraries organized by use case - graph analysis, explainable AI, quantitative finance, web frameworks, and MLOps tooling.

Graph Analysis

Explainable AI

LibraryWhat It Does
SHAPShapley values. Theoretically grounded, model-agnostic feature attribution.
LIMELocal surrogate models. Fits a simple model in the neighborhood of a prediction.
ELI5Feature weights, permutation importance, text and image debugging.
InterpretMLExplainable Boosting Machines plus unified dashboard.
ShapashBusiness-facing Shapley dashboard with plain-language labels.
OmniXAISalesforce’s unified XAI library. Multiple methods, one API.
explainerdashboardInteractive Shapley dashboard with classification and regression support.

SHAP is the default choice for production XAI. It has the strongest theoretical backing (Shapley values from cooperative game theory) and works with tree models, deep learning, and linear models.

Quantitative Finance and Time Series

Python-First Web Frameworks

For building data-driven dashboards and apps without JavaScript:

MLOps Tooling

Kedro plus DVC plus MLflow is a coherent MLOps stack that handles data versioning, experiment tracking, and pipeline orchestration in one configuration.

Curated Library Lists

Reference

python libraries tools data-science mlops
← All articles

Have a problem worth solving?

Whether you need a quantitative researcher, a Machine Learning systems builder, or a technical advisor, I take a small number of consulting engagements at a time.

Book a call →