Compute gradients and compile Python functions with JAX transformations.
npx skills add https://github.com/yonesuke/skills --skill jax
Please help me install this Agent Skill. Skill: JAX Source: https://github.com/yonesuke/skills/tree/main/jax Command: npx skills add https://github.com/yonesuke/skills --skill jax
JAX solves the need for high-performance, differentiable computing for ML research by providing automatic differentiation, JIT compilation, and scalable parallelism across devices.
Run a tiny experiment using jax.grad and jax.jit to train a simple linear model on a small dataset.
High-intent search queries and answers about installing and using this skill.
Automatic differentiation in JAX computes gradients for Python code using grad and jacrev, enabling end-to-end differentiation over ML models. You apply grad to your loss function to get gradient computations for training.
Accelerating numerical computing for GPU execution is done using jit compilation, which compiles Python functions just-in-time for XLA-accelerated hardware execution. This significantly speeds up ML workloads and rapid prototyping.
Parallelizing ML workloads across multiple devices is handled by pmap for distributed parallelism and vmap for vectorized mapping. These transformations enable scalable parallelism across hardware for differentiable computing.
Pytrees are container structures for managing model parameters and state through functional APIs. They enable deterministic execution by treating parameters as immutable data, ensuring reproducible ML model prototyping.
JAX works with XLA-accelerated execution and just-in-time compilation for ML research by combining Autograd-style auto-differentiation with XLA. It supports jit, grad, vmap, and pmap transformations for hardware acceleration.
jstzwjRun numerical routines, automatic differentiation, and hardware acceleration with JAX.
GroundMountCompanyAutomate AI image generation via Replicate API with model selection and iterative refinement.
47networkTrain tiny neural networks and visualize backpropagation steps.
tondevrelProvides differentiable quantum circuits for Python ML frameworks like PyTorch, TensorFlow and JAX.
