pranftw/neograd
A deep learning framework created from scratch with Python and NumPy
This is a specialized deep learning framework built in Python and NumPy that simplifies the inner workings of neural networks. It takes numerical datasets and model architectures as input to train and evaluate neural networks, outputting trained models and performance metrics. It's for machine learning practitioners and students who want to deeply understand how modern deep learning frameworks function without the complexity of large, production-grade tools.
238 stars. No commits in the last 6 months. Available on PyPI.
Use this if you are an aspiring machine learning engineer or student who wants to learn the fundamental concepts of automatic differentiation and neural network training by examining clean, readable code.
Not ideal if you need a high-performance deep learning solution for large-scale, real-world applications or require GPU acceleration.
Stars
238
Forks
9
Language
Python
License
GPL-3.0
Category
Last pushed
Dec 26, 2022
Commits (30d)
0
Dependencies
2
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/pranftw/neograd"
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