hadley/r4ds

Apache Superset is a self-hosted, open source platform for exploring data and building dashboards, pairing a no-code chart builder with a full SQL Editor and a semantic layer for reusable metrics.

30 Days Of Python is a free GitHub curriculum by Asabeneh Yetayeh that breaks Python into 30 daily topics, from syntax basics through web scraping, MongoDB, and building an API. The README pitches it at beginners and working professionals, admits the timeline realistically runs past 100 days, and points to a Telegram group and companion YouTube channel for support.

scikit-learn is an open-source Python module for machine learning at scikit-learn/scikit-learn, layered on top of SciPy and covering classification, regression, clustering, and dimensionality reduction. The project traces back to a 2007 Google Summer of Code effort by David Cournapeau, was known early on as scikits.learn, and is now kept up by a volunteer community with backing from several organizations, distributed under the BSD-3-Clause license.

pandas provides a labeled DataFrame for working with tabular data — the labeled DataFrame gives you alignment, indexing, and aggregation that raw NumPy arrays simply can't match without significant boilerplate. The API surface is enormous, covering everything from CSV reading to time-series resampling, and while the learning curve is real, the payoff is a coherent mental model for most data-wrangling tasks. It started at AQR Capital Management in 2008 and has been under active development since, with a mature ecosystem, extensive documentation on PyData.org, and community support through Stack Overflow, mailing lists, and Slack. It's not the fastest option for pure numerical compute (Polars and PyArrow handle large datasets with less overhead), but for exploratory analysis and prototyping in Python, the ecosystem depth and API consistency make pandas the practical default.