InputSpaceScenario
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.
Every possible input of dataset’s image shape: levels values per channel element.
Notes
Equation: Use the dimensional products, ratios, or selection rules documented on the calculated properties below.
Assumptions: Every image channel element independently takes one of the stated discrete levels.
Excludes: The counted input space ignores natural-image structure and does not measure dataset representativeness.
References
Derivation: the calculated properties below state their defining arithmetic and units. They are scenario accounting or input-selection rules, not independently measured relationships.
Examples
>>> from mlsysim import Scenarios
>>> case = Scenarios.ImageNetInputSpace
>>> round(case.elements, 4)
150528Attributes
| Name | Description |
|---|---|
| dataset | |
| digits | Decimal digits required to write the input-space cardinality. Equation: |
| elements | Independent channel values in one image. Equation: image width times height times |
| levels | |
| log10_inputs | Base-ten logarithm of the discrete image input space. Equation: ``elements * |
| test_examples | Registered validation-set size as an integer. Equation: convert |
dataset
Type: DatasetProfile.
Declared default: required.
levels
Type: int.
Declared default: required.
elements
engine.scenario_types.InputSpaceScenario.elements
Independent channel values in one image. Equation: image width times height times channel count. Units: scalar elements; image geometry comes from the dataset entry.
log10_inputs
engine.scenario_types.InputSpaceScenario.log10_inputs
Base-ten logarithm of the discrete image input space. Equation: elements * log10(levels). Units: dimensionless; all element assignments are counted independently.
digits
engine.scenario_types.InputSpaceScenario.digits
Decimal digits required to write the input-space cardinality. Equation: floor(log10_inputs) + 1. Units: digits; this avoids materializing the enormous integer.
test_examples
engine.scenario_types.InputSpaceScenario.test_examples
Registered validation-set size as an integer. Equation: convert dataset.validation_examples to count and then integer. Units: examples; this is sourced dataset metadata, not input-space coverage.