SpeedupCompositionScenario
Optimization scenarios · Implementation: mlsysim/engine/scenario_definitions/optimization.py.
Construct this scenario with keyword inputs. The fields below show the types and defaults declared by this class.
Independent paper speedups multiplied together, against the measured speedup.
Notes
Equation: paper is the worked quantity; each property below states its equation and units.
Assumptions: The product assumes independent component speedups over compatible baselines.
Excludes: Establishing independence or predicting end-to-end measured performance.
References
Derivation: The equations below follow the stated counts, geometric relations, or accounting definitions; scenario inputs are supplied assumptions or observations, not fitted predictions.
Examples
Execute the registered worked inputs; this is a modeled calculation.
>>> from mlsysim import Scenarios
>>> case = Scenarios.PrunedInt8PaperVsMeasured
>>> value = case.paper
>>> round(value, 3)
8.0Attributes
| Name | Description |
|---|---|
| factors | |
| measured | |
| paper | Product of supplied dimensionless speedup factors under independence. |
| realized_share | Dimensionless observed speedup divided by the product of component factors. |
factors
Type: Dict[str, float].
Declared default: required.
measured
Type: float.
Declared default: required.
paper
engine.scenario_types.SpeedupCompositionScenario.paper
Product of supplied dimensionless speedup factors under independence.
Equation: out = 1.0 for f in self.factors.values(): out *= f return out