Just a bunch of stuff that I keep here instead of as bookmarks so that I don't forget - libraries that are not that common, e.g., Pytorch...
-
torchlayers: Shape and dimension inference (Keras-like) for PyTorch layers and neural networks
-
torchtyping: Type annotations and dynamic checking for a tensor's shape, dtype, names, etc.
-
DeepSpeed: A deep learning optimization library that makes distributed training easy, efficient, and effective.
-
accelerate: A simple way to train and use PyTorch models with multi-GPU, TPU, mixed-precision
-
python-patterns: A collection of design patterns/idioms in Python.
-
pytorchvideo: A deep learning library for video understanding research.
-
AlphaPose: Real-Time and Accurate Full-Body Multi-Person Pose Estimation&Tracking System.
-
Click: Python composable command line interface toolkit.
-
einops: Deep learning operations reinvented (for pytorch, tensorflow, jax and others)