Don't you want to at least leave the naming open to allow for two different types of 16-bit float? Get that right NOW, before you lock in a name that's suboptimal when you switch from "GPU interop" to "Swift for AI learning"...
-
-
-
What would your preferred nomenclature be (keeping in mind that bikeshedding over nomenclature and—to a much lesser extent, formats—is what kept us from having a half-precision type in Swift for this long)?
- Još 2 druga odgovora
Novi razgovor -
-
-
that’s great! but on the other hand I’m still hoping to see Float64 on the GPU.
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
Čini se da učitavanje traje već neko vrijeme.
Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.
