SearchAccountsScenario
Agents scenarios · Implementation: mlsysim/engine/scenario_definitions/agents.py.
Construct this scenario with keyword inputs. The fields below show the types and defaults declared by this class.
Flat best-of-N against beam search over the same task, kept as separate accounts.
Notes
Equation: result = SearchAccountsModel.solve(…) with the declared floors, segments, segment_tokens inputs.
Assumptions: The inputs instantiate the assumptions documented by mlsysim.solvers.SearchAccountsModel.solve; accessing result evaluates that solver again.
Excludes: Any extra mechanism beyond SearchAccountsModel; this record supplies configuration and does not collect performance measurements.
References
Derivation: This configuration forwards its stored inputs to mlsysim.solvers.SearchAccountsModel.solve. The solver contract records the equations and their provenance.
Examples
The named scenario supplies illustrative inputs; the output is calculated.
>>> from mlsysim import Scenarios
>>> case = Scenarios.FlatVsBeamSearch
>>> round(float(case.result.token_time.to('s').magnitude), 3)
0.021Attributes
| Name | Description |
|---|---|
| beam_width | |
| extensions | |
| flat_candidates | |
| floors | |
| generation_slots | |
| input_price | |
| output_price | |
| prompt_tokens | |
| result | Modeled result from mlsysim.solvers.SearchAccountsModel.solve using this |
| score_time | |
| scorer_price | |
| segment_tokens | |
| segments | |
| test_slots | |
| test_time |
floors
Type: DecodeScenario.
Declared default: required.
segments
Type: int.
Declared default: required.
segment_tokens
Type: int.
Declared default: required.
prompt_tokens
Type: int.
Declared default: required.
flat_candidates
Type: int.
Declared default: required.
beam_width
Type: int.
Declared default: required.
extensions
Type: int.
Declared default: required.
generation_slots
Type: int.
Declared default: required.
test_slots
Type: int.
Declared default: required.
score_time
Type: Quantity.
Declared default: required.
test_time
Type: Quantity.
Declared default: required.
input_price
Type: Quantity.
Declared default: required.
output_price
Type: Quantity.
Declared default: required.
scorer_price
Type: Quantity.
Declared default: required.
result
engine.scenario_types.SearchAccountsScenario.result
Modeled result from mlsysim.solvers.SearchAccountsModel.solve using this scenario’s declared inputs; see that method for units, limits, and feasibility.
Equation: Delegates the declared fields to mlsysim.solvers.SearchAccountsModel.solve; returned quantities and feasibility follow that contract.