$ show dgasmith/opt_einsum
opt_einsum is a Python library that optimizes the contraction order of tensor expressions (einsum-like) to reduce computational complexity and memory usage. It serves as a drop-in replacement for numpy.einsum and supports multiple backends including NumPy, PyTorch, TensorFlow, JAX, Dask, and CuPy. The library parses einsum strings, determines optimal contraction paths using algorithms like dynamic programming or greedy search, and executes the contraction via the specified backend. Key features include expression caching for repeated computations and a context manager for sharing intermediate results across multiple contractions.
The LLM identified 3 files worth highlighting for reuse:
Implements the core path-finding algorithms (optimal, greedy, branch, dynamic programming) for tensor contraction. Demonstrates sophisticated algorithmic design to solve an NP-hard problem efficiently, with clear separation of cost functions and search strategies.
Contains the main API functions (contract, contract_path, contract_expression) and the ContractExpression class. Implements a pattern for caching optimized contraction paths to avoid recomputation, enabling significant performance gains for repeated tensor operations.
Provides a shared_intermediates context manager that memoizes intermediate results across multiple contractions. This non-obvious optimization pattern allows for significant speedups when computing multiple related tensor expressions by reusing computed sub-results.