punytorch is a simple and feature poor deep learning library implemented in Python. It is inspired by PyTorch, and is designed to be a simple and easy-to-understand implementation of a deep learning library (once it's working)
The goal is to have something very simple that will let me implement and experiment with new concepts quickly in my deep learning journey. I'm not trying to build a production-ready library, but rather a learning tool.
-
If poetry is not installed, install it via:
curl -sSL https://install.python-poetry.org | python - -
Run
poetry installto install the dependencies -
Run
poetry shellto activate the virtual environment -
Run
python examples/mnist.pyto train a simple model on the MNIST dataset
can it run an LLM?
nope
can it do classification jobs?
only 1, and barely
what about computer vision?
punytorch is blind