dmaivel/cugrad
An automatic differentiation library written in C++ and CUDA
This library is for C++ and CUDA developers who need to calculate derivatives automatically for complex equations on a GPU. It takes C++ code defining mathematical operations on tensors and outputs the numerical results or gradients of those operations, leveraging the power of CUDA for parallel computation. Developers can integrate this into their high-performance computing applications.
No commits in the last 6 months.
Use this if you are a C++ and CUDA developer building applications that require efficient automatic differentiation on a GPU without external dependencies like cuBLAS.
Not ideal if you are a data scientist or machine learning engineer looking for a high-level Python library, or if you require highly optimized kernels for production-level performance right out of the box.
Stars
10
Forks
3
Language
C++
License
MIT
Category
Last pushed
May 28, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/dmaivel/cugrad"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
JonathanRaiman/theano_lstm
:microscope: Nano size Theano LSTM module
google/tangent
Source-to-Source Debuggable Derivatives in Pure Python
ahrefs/ocannl
OCANNL: OCaml Compiles Algorithms for Neural Networks Learning
yoshoku/numo-openblas
Numo::OpenBLAS builds and uses OpenBLAS as a background library for Numo::Linalg
statusfailed/catgrad
a categorical deep learning compiler