SubgroupSampleScenario

Robotics scenarios · Implementation: mlsysim/engine/scenario_definitions/robotics/supervision.py.

Construct this scenario with keyword inputs. The fields below show the types and defaults declared by this class.

An evaluation set of total people with a group of small_share, both detected at recall: exact intervals for each group, and the people per group a two-sided test at significance with power needs to tell rate_a from rate_b.

Notes

Equation: Group counts are rounded shares of total; recall intervals use exact binomial tails, while group-size planning uses a pooled normal approximation.

Assumptions: People are independent Bernoulli observations; assumed recall is converted to an integer success count by rounding.

Excludes: A study-design correction for correlated people, repeated measurements, or multiple subgroup comparisons.

References

Derivation: the scenario composes mlsysim.physics.statistics.calc_clopper_pearson_interval, mlsysim.physics.statistics.calc_two_proportion_sample_size. Their contracts supply the underlying equations; property equations show the remaining unit conversions and accounting.

Examples

>>> from mlsysim import Scenarios
>>> scenario = Scenarios.PedestrianGroupEvaluationSample
>>> value = scenario.small
>>> round(float(value), 3)
200.0

Attributes

Name	Description
confidence
large	Remaining people after subtracting the rounded smaller subgroup count.
large_interval	Dimensionless exact binomial recall interval for the larger group.
per_group_needed	Ceiling of the pooled two-proportion planning approximation at the declared significance
power
rate_a
rate_b
recall
share_needed	Required per-group sample count as a fraction of the current total evaluation count.
significance
small	Rounded people count in the smaller subgroup, from total count times subgroup share.
small_interval	Dimensionless exact binomial recall interval for the smaller group.
small_share
total
width_ratio	Small-group interval width divided by large-group width, dimensionless.

Methods

Name	Description
interval	Clopper-Pearson recall interval for `n` observations and a rounded assumed success

interval

engine.scenario_types.SubgroupSampleScenario.interval(n)

Clopper-Pearson recall interval for n observations and a rounded assumed success count.

Equation: calc_clopper_pearson_interval(round(recall * n), n, confidence).