Scanning all new published packages on PyPI I know that the quality is often quite bad. I try to filter out the worst ones and list here the ones which might be worth a look, being followed or inspire you in some way.

• torch_maml
Gradient checkpointing technique for Model Agnostic Meta Learning

• bayeso-benchmarks
Benchmarks for Bayesian optimization

• plot-utils
A Python library for elegant data visualization

• multiplex-plot
Multiplex: visualizations that tell stories

• clcell
OpenCL-accelerated cellular automata simulator

• emmo
Python reference API for the Europeean Materials & Modelling Ontology

• mlglass
MLglass: A Transparency with models

• keras-adf
Assumed Density Filtering (ADF) Probabilistic Networks

• ImageAug
Image augmentation for PyTorch

• dblue-stores
Dblue stores is an abstraction and a collection of clients to interact with storages.

• py-ml-utils
Machine learning utilities.

• pyevo
Package for applying evolutionary algorithms to various problem models.