Find your next starting point.

Multidimensional array operations and numerical linear algebra for scientific Python.

Property-based testing tools that generate and simplify test inputs.

A replication about reinforcement-learning-based program debloating and its relationship to unit tests and reduction procedures.

Training and inference orchestration for studying device placement, mixed precision and distributed execution without changing the scientific objective.

Experiment tracking and comparison tooling for inspecting metric histories and run metadata.

Bioinformatics tools for sequence data, biological file formats, and related analyses.

Low-precision tensor and optimizer components for investigating memory and numerical-error tradeoffs.

Model interpretability algorithms for checking attribution behavior against simple functions with known contributions.

Assertions and testing utilities for JAX programs, useful for expressing shape, type and transformation expectations.

Serialization support for Python constructs used in distributed and interactive computation.

Task graphs and parallel collections for array, table, and general Python computation.

Deep-learning optimization system for investigating distributed training memory, communication and state partitioning.