TrainingStateScenario
Training scenarios · Implementation: mlsysim/engine/scenario_definitions/training.py.
Construct this scenario with keyword inputs. The fields below show the types and defaults declared by this class.
Persistent mixed-precision Adam state of workload and its shards.
Notes
Equation: Compose mlsysim.solvers.TrainingStateModel.
Assumptions: The stated fields are inputs to mlsysim.solvers.TrainingStateModel; its documented validity regime, units, and modeling choices apply unchanged.
Excludes: This scenario adds no measurement, empirical calibration, or mechanisms beyond mlsysim.solvers.TrainingStateModel.
References
Derivation: this scenario binds its input fields to the referenced implementation contracts: mlsysim.solvers.TrainingStateModel. The methods below identify any additional arithmetic.
Examples
>>> from mlsysim import Scenarios
>>> case = Scenarios.Gpt3TrainingStateTp8
>>> type(case.result).__name__
'TrainingStateResult'Attributes
| Name | Description |
|---|---|
| activations | |
| dp_size | |
| result | Evaluate this scenario through mlsysim.solvers.TrainingStateModel.solve. Equation: |
| tp_size | |
| workload |
workload
Type: Workload.
Declared default: required.
tp_size
Type: int.
Declared default: required.
dp_size
Type: int.
Declared default: required.
activations
Type: Quantity.
Declared default: required.
result
engine.scenario_types.TrainingStateScenario.result
Evaluate this scenario through mlsysim.solvers.TrainingStateModel.solve. Equation: pass the stated fields to that implementation; its contract defines the calculation and feasibility tests. Units: each returned field follows the referenced result schema; no independent equation is evaluated by this wrapper.