JAX
NumPy-like array library with automatic differentiation and XLA compilation.
Resumen
JAX provides composable function transformations: automatic differentiation, vectorisation and just-in-time compilation to XLA, all applicable to ordinary Python functions. It solves the problem that research code needs to be differentiated and accelerated without being rewritten around a framework's graph API. It exists because functional composition turns out to be a cleaner foundation for large-scale training, and it underlies several current model families.
Grafo de tecnologías
Cómo se conecta JAX con el resto del stack.
Usa y depende de
- NVIDIA CUDA platform Hardware JAX uses NVIDIA CUDA High ✓ source
Aplicado a
- Deep Learning concept Algorithms JAX used for Deep Learning Medium ✓ source
Herramientas
Software que implementa o se integra con JAX.