TraceRetentionScenario
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.
One cohort period of trials in disjoint verdict classes, each with a trace_size body, retained over window by keeping everything, head sampling, or verdict-aware selection.
Notes
Equation: result evaluates the configured solver; its contract owns the governing equations.
Assumptions: Outcome cohorts, trace sizes, and retention probabilities are supplied for a fixed period.
Excludes: Compression variation, correlation between sampling decisions, and downstream storage overhead.
References
Derivation: The scenario binds declared inputs to TraceRetentionModel. Those implementations document their derivations and sources.
Examples
Execute the registered worked inputs; this is a modeled calculation.
>>> from mlsysim import Scenarios
>>> case = Scenarios.VerdictAwareTraceRetention
>>> value = case.result
>>> print(f"{value.tail_volume:.3f~P}")
23.184 GBAttributes
| Name | Description |
|---|---|
| cohort_period | |
| control_probability | |
| control_step | |
| failed | |
| head_probability | |
| ordinary | |
| result | Evaluate the configured inputs with mlsysim.solvers.TraceRetentionModel.solve. |
| trace_size | |
| trials | Total trial count across failed, unresolved, and ordinary outcomes. |
| unresolved | |
| window |
failed
Type: int.
Declared default: required.
unresolved
Type: int.
Declared default: required.
ordinary
Type: int.
Declared default: required.
cohort_period
Type: Quantity.
Declared default: required.
trace_size
Type: Quantity.
Declared default: required.
window
Type: Quantity.
Declared default: required.
head_probability
Type: float.
Declared default: required.
control_probability
Type: float.
Declared default: required.
control_step
Type: float.
Declared default: 0.01.
trials
engine.scenario_types.TraceRetentionScenario.trials
Total trial count across failed, unresolved, and ordinary outcomes.
Equation: return self.failed + self.unresolved + self.ordinary
result
engine.scenario_types.TraceRetentionScenario.result
Evaluate the configured inputs with mlsysim.solvers.TraceRetentionModel.solve.
The returned typed result retains the solver’s documented units, assumptions, and feasibility semantics; this property adds no measurement or safety claim. Equation: result = TraceRetentionModel().solve(...) using the fields shown below.