cuequivariance-jax

Official

Execute equivariant polynomials in JAX with cue.

AuthorNVIDIA
Version1.0.0
Installs0

System Documentation

What problem does it solve?

Execute equivariant polynomials in JAX using cuequivariance primitives, enabling consistent, representation-aware computations for neural networks that respect symmetry groups.

Core Features & Use Cases

  • Multiple representations: Work with RepArray, RepArray-based polynomial evaluation, and ir_dict representations to fit different workflows.
  • Polynomial backends: Utilize segmented_polynomial with naive and uniform_1d backends, plus per-irrep handling for efficient GPU execution.
  • NNX integration: Access Flax NNX layers (IrrepsLinear, SphericalHarmonics) to build symmetry-aware neural networks.

Quick Start

Build a small equivariant polynomial descriptor and evaluate it against sample inputs.

Dependency Matrix

Required Modules

None required

Components

Standard package

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: cuequivariance-jax
Download link: https://github.com/NVIDIA/cuEquivariance/archive/main.zip#cuequivariance-jax

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.