SciPy

SciPy builds on NumPy’s array foundation with the actual algorithms — optimization, numerical integration, interpolation, linear algebra beyond NumPy’s basics, statistics, and signal processing — the kind of routines you’d otherwise have to implement from a textbook yourself.

It’s infrastructure more than something you browse directly: most scientific and statistical Python work ends up calling into SciPy somewhere, often without the user needing to think about it explicitly, similar to how NumPy underlies Pandas and scikit-learn.