7 projects
leancert
Formal verification for numerical Python code - powered by Lean4
fast-cooc
GPU-first co-occurrence matrix builder for large text corpora
warpdata
Remote-first dataset library for ML training
warpdatasets
Remote-first dataset library for ML training
torchlitex
Tiny DDP training toolkit for quick-launch distributed training loops.
schemas-hf
Dataclass schemas for local Hugging Face datasets
dataspeak-analytics
A library for analytics using Mixpanel and MongoDB