The plain idea
Python’s real power is not just the language but its vast ecosystem of libraries, ready-made toolkits you import to do heavy work in a line or two. This page is my running list of the ones worth knowing, especially for numerical and scientific work: NumPy for arrays, SciPy for maths, iPython/Jupyter for interactive exploration, PyMC for Bayesian modelling, and more, plus tutorials and learning resources.
Why it matters
You rarely write serious Python from scratch; you assemble it from libraries, and knowing which library solves which problem is most of the skill. NumPy turns slow Python loops into fast array operations; pandas makes spreadsheets programmable; the whole data-science and AI boom rests on this stack. A good mental map of the ecosystem saves you from reinventing wheels, badly.
An everyday analogy
The Python language is a well-equipped workshop, but the libraries are the power tools hanging on the wall. You could cut the wood with a hand saw (plain Python), but reaching for the right power tool (NumPy, pandas) does the job in a fraction of the time and effort. This page is my labelled pegboard of which tool is which.
A 2026 note
The core scientific stack below, NumPy, SciPy, Jupyter, is more central than ever and all actively developed. Worth adding to any modern list: pandas and the faster Polars for data tables, scikit-learn for classic machine learning, PyTorch for deep learning, matplotlib/Plotly for charts, and uv or poetry for managing packages and environments (a big improvement over the old pip-and-virtualenv dance). Note: Python 2 is long dead, everything is Python 3 now, so ignore any Python 2 syntax in older tutorials.
How it connects
- Python: the language itself.
- Open Data Science and Machine Learning: where these libraries shine.
- SciPy, NumPy, iPython: deeper notes on the key ones.
The truth revealed: fluency in Python is really fluency in its libraries, knowing which tool to lift off the wall, and keeping a good map of them is worth more than memorising syntax.
Libraries, Tools & Links
- SciPy | numPy | iPython
- PyMC
- easygui
- Welcome to the Little Pythoner! — Little Pythoner 0.1 documentation
- Your first Python program - Dive Into Python 3
- make art with python
- [Programming and Bayesian Methods for Hackers](http://nbviewer.jupyter.org/github/CamDavidsonPilon/Probabilistic-Programming-and-Bayesian-Methods-for-Hackers/blob/master/Prologue/Prologue.ipynb Probabilistic)
- Python Freelancing
- Corey Schafer Python Tutorials
- GitHub - TheAlgorithms/Python: All Algorithms implemented in Python