ActiveLearningRoundsScenario
Data scenarios · Implementation: mlsysim/engine/scenario_definitions/data.py.
Construct this scenario with keyword inputs. The fields below show the types and defaults declared by this class.
Active learning from initial_labels adding labels_per_round for rounds rounds, each round paying round_scoring_cost to score the pool, against random labeling that needs random_labels at the same label_price.
Notes
Equation: active_labels is the worked quantity; each property below states its equation and units.
Assumptions: Each round buys fixed labels and pays a fixed scoring bill; baseline label demand is supplied.
Excludes: Quality prediction, turnaround time in the returned costs, and changing unit prices.
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.ActiveLearningTenRounds
>>> value = case.active_labels
>>> value
2000Attributes
| Name | Description |
|---|---|
| active_cost | Combined active-learning labeling and scoring cost in USD. |
| active_label_cost | Active-learning labeling cost in USD, excluding scoring. |
| active_labels | Total labels acquired across the initial pool and subsequent rounds. |
| initial_labels | |
| label_price | |
| label_turnaround | |
| labels_per_round | |
| random_cost | Random-baseline labeling cost in USD. |
| random_labels | |
| roi | Dimensionless net saving divided by active-learning cost. |
| round_scoring_cost | |
| rounds | |
| scoring_cost | Total pool-scoring cost in USD across all rounds. |
label_price
Type: Quantity.
Declared default: required.
initial_labels
Type: int.
Declared default: required.
labels_per_round
Type: int.
Declared default: required.
rounds
Type: int.
Declared default: required.
round_scoring_cost
Type: Quantity.
Declared default: required.
random_labels
Type: int.
Declared default: required.
label_turnaround
Type: Optional[Quantity].
Declared default: None.
active_labels
engine.scenario_types.ActiveLearningRoundsScenario.active_labels
Total labels acquired across the initial pool and subsequent rounds.
Equation: return self.initial_labels + self.rounds * self.labels_per_round
random_cost
engine.scenario_types.ActiveLearningRoundsScenario.random_cost
Random-baseline labeling cost in USD.
Equation: return (self.random_labels * self.label_price).to(‘USD’)
active_label_cost
engine.scenario_types.ActiveLearningRoundsScenario.active_label_cost
Active-learning labeling cost in USD, excluding scoring.
Equation: return (self.active_labels * self.label_price).to(‘USD’)
scoring_cost
engine.scenario_types.ActiveLearningRoundsScenario.scoring_cost
Total pool-scoring cost in USD across all rounds.
Equation: return (self.rounds * self.round_scoring_cost).to(‘USD’)
active_cost
engine.scenario_types.ActiveLearningRoundsScenario.active_cost
Combined active-learning labeling and scoring cost in USD.
Equation: return self.active_label_cost + self.scoring_cost
roi
engine.scenario_types.ActiveLearningRoundsScenario.roi
Dimensionless net saving divided by active-learning cost.
Saving over the active-learning cost.
Equation: return ((self.random_cost - self.active_cost) / self.active_cost).to(’’).magnitude