WG211/M23Singh

From WG 2.11
Revision as of 19:18, 7 March 2024 by Jeremy-y (talk | contribs) (Created page with " == Haste: A Haskell DSL for Programming the Groq TSP/LPU Machine Learning Chips (Satnam Singh) == I'll describe a Haskell DSL that uses Lava-like combinators to describe lin...")
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

Haste: A Haskell DSL for Programming the Groq TSP/LPU Machine Learning Chips (Satnam Singh)

I'll describe a Haskell DSL that uses Lava-like combinators to describe linear algebra-based programs that are compiled to run on Groq's TSP/LPU machine learning chips. The approach will be demonstrated with some examples including a high speed parallel sorter and an FFT implementation.

Retrieved from "https://wiki.hh.se/wg211/index.php?title=WG211/M23Singh&oldid=2536"