Bynaryman/OSFNTC
Open-Source Framework for Numerically-Tailored Computations This project is "An Open-Source Framework for Efficient Numerically-Tailored Computations", hence the acronym OSFNTC. The title of the corresponding paper is enclosed in the quotation marks.
This framework helps scientists and engineers achieve higher accuracy and energy efficiency in complex computational tasks involving Matrix-Matrix Multiplications (MMMs). It takes your existing data and computation requirements, then generates specialized hardware configurations for FPGAs to deliver optimized results. This is for researchers and practitioners in fields like AI inference, climate modeling, and high-performance computing who need precise numerical computations.
No commits in the last 6 months.
Use this if you need to significantly improve the accuracy and energy efficiency of Matrix-Matrix Multiplications (MMMs) in high-performance computing applications, especially on FPGA platforms.
Not ideal if you do not work with FPGAs, are not concerned with highly optimized numerical precision and energy consumption, or do not perform Matrix-Matrix Multiplications.
Stars
7
Forks
—
Language
Fortran
License
—
Category
Last pushed
Jun 15, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/Bynaryman/OSFNTC"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
mlverse/torch
R Interface to Torch
modern-fortran/neural-fortran
A parallel framework for deep learning
Beliavsky/Fortran-code-on-GitHub
Directory of Fortran codes on GitHub, arranged by topic
Cambridge-ICCS/FTorch
A library for directly calling PyTorch ML models from Fortran.
NVIDIA/TorchFort
An Online Deep Learning Interface for HPC programs on NVIDIA GPUs