devitocodes/devito
DSL and compiler framework for automated finite-differences and stencil computation
Builds on SymPy to translate high-level symbolic PDE specifications into optimized C/Fortran code via automated code generation and JIT compilation, targeting CPUs, GPUs, and distributed clusters. Applies aggressive compiler optimizations including SIMD vectorization, loop blocking, symbolic transformations for FLOP reduction, and multi-level parallelism (OpenMP, OpenACC, MPI). Integrates seamlessly with NumPy, Dask, SciPy, and machine learning frameworks like TensorFlow and PyTorch for hybrid scientific computing workflows.
679 stars and 5,103 monthly downloads. Actively maintained with 57 commits in the last 30 days. Available on PyPI.
Stars
679
Forks
253
Language
Python
License
MIT
Category
Last pushed
Mar 18, 2026
Monthly downloads
5,103
Commits (30d)
57
Dependencies
10
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/devitocodes/devito"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.