Both regression decompositions use the checked one-output, one-input
evaluator with a zero device update rate. Their global updates stay explicit
in the learner. Target facts come only from the selected profile.
12def regressionScalarPlan =
13 (lambda unrestricted profile : (family RunPodSM86TargetProfile) .
14 (nvidiaLinearStepPlan
15 1
16 1
17 0
18 runPodSM86QMDArchitecture
19 runPodSM86ComputeClass
20 runPodSM86SPAVersion
21 runPodSM86QMDSlotBytes
22 (runPodSM86ProductIdentityPadded profile)
23 nvidiaPlanHostAmpere580
24 (modelWord64Natural (runPodSM86TotalVideoMemory profile))))The compiler supplied declaration spans and resolved links from this source snapshot. This page does not assert that this file belongs to a checked closure.