Yehor-Mishchyriak/PureML
Transparent, NumPy-only deep learning framework for teaching, small-scale projects, prototyping, and reproducible experiments. No CUDA, no giant dependency tree. Batteries included: VJP autograd, layers, activations, losses, optimizers, Zarr checkpoints, and more!
PureML is a deep learning framework designed for educational purposes, small-scale projects, and experimenting with new ideas. It allows users to build and train neural networks using familiar NumPy arrays as input and produce trained models and predictions. This tool is ideal for data scientists, machine learning engineers, and students who want to understand the inner workings of deep learning models without complex dependencies.
Use this if you are a deep learning practitioner or student who wants a transparent, easy-to-understand framework for learning, prototyping, or running small experiments, especially if you prefer working directly with NumPy.
Not ideal if you are building large-scale, production-grade deep learning models that require GPU acceleration or advanced distributed training features.
Stars
10
Forks
1
Language
Python
License
Apache-2.0
Category
Last pushed
Feb 14, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/Yehor-Mishchyriak/PureML"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
pymc-devs/pytensor
PyTensor allows you to define, optimize, and efficiently evaluate mathematical expressions...
arogozhnikov/einops
Flexible and powerful tensor operations for readable and reliable code (for pytorch, jax, TF and others)
lava-nc/lava-dl
Deep Learning library for Lava
tensorly/tensorly
TensorLy: Tensor Learning in Python.
tensorpack/tensorpack
A Neural Net Training Interface on TensorFlow, with focus on speed + flexibility