Hackernews posts about PyTorch
PyTorch is an open-source machine learning library for Python that provides a seamless interface between dynamic computation graphs and automatic differentiation.
- TorchTPU: Running PyTorch Natively on TPUs at Google Scale (developers.googleblog.com)
- Proximal Policy Optimization with Clojure and PyTorch (clojurecivitas.org)
- TorchWebGPU: Running PyTorch Natively on WebGPU (github.com)
- DeepInverse: PyTorch Library for Inverse Problems (github.com)
- Show HN: SparseLab–real sparse training(CSR+custom kernel) in PyTorch, CPU-first (news.ycombinator.com)
- AutoSP from PyTorch (pytorch.org)
- Proximal Policy Optimization with Clojure and PyTorch (www.wedesoft.de)
- Show HN: Chestnut – The antidote to AI-induced skill atrophy (www.chestnut.so)
- Show HN: Dbg – One CLI debugger for every language (AI-agent ready) (redknightlois.github.io)
- Show HN: Looties – a marketplace for tech swag fans (looties.io)
- Show HN: Coelanox – auditable inference runtime in Rust (BERT runs today) (www.coelanox.com)
- A Primer on LLM Post-Training (pytorch.org)