Skip to content

API reference / Switchback

tsdive.SwitchbackAnalysis dataclass 

SwitchbackAnalysis(
    plan: SwitchbackPlan,
    target: Window,
    covariates: tuple[Window, ...],
    unused: tuple[str, ...],
    direct: SwitchbackEstimate,
    adjusted: SwitchbackEstimate | None,
    *,
    assumptions: str = ASSUMPTIONS,
    covariate_checks: tuple[CovariateCheck, ...] = (),
)

The difference between settings A and B on one target, under a verified plan.

target and covariates are the reads over the schedule. unused names the archives passed in that are neither. adjusted is None when no covariate was declared. covariate_checks holds one CovariateCheck per covariate, in the order they were named; the analysis does not refuse on one that moves, it reports it.

Methods:

Name Description
render

The text tsdive switchback analyze prints.
to_dict

The document tsdive switchback analyze --json prints, ready for json.dumps.

Attributes:

Name Type Description
plan SwitchbackPlan
target Window
covariates tuple[Window, ...]
unused tuple[str, ...]
direct SwitchbackEstimate
adjusted SwitchbackEstimate | None
assumptions str
covariate_checks tuple[CovariateCheck, ...]
moving_covariates tuple[CovariateCheck, ...]

The covariates whose own B - A difference has p below 0.05.
tag str
unit str | None
good_share float | None

GOOD rows over all rows of the target read.
censored bool | None

Whether a target sample sits at its engineering range limit.
frame DataFrame

One row per block: index, start, setting, kept samples of each analysis.

plan instance-attribute 

plan: SwitchbackPlan

target instance-attribute 

target: Window

covariates instance-attribute 

covariates: tuple[Window, ...]

unused instance-attribute 

unused: tuple[str, ...]

direct instance-attribute 

direct: SwitchbackEstimate

adjusted instance-attribute 

adjusted: SwitchbackEstimate | None

_ instance-attribute 

_: KW_ONLY

assumptions class-attribute instance-attribute 

assumptions: str = ASSUMPTIONS

covariate_checks class-attribute instance-attribute 

covariate_checks: tuple[CovariateCheck, ...] = ()

moving_covariates property 

moving_covariates: tuple[CovariateCheck, ...]

The covariates whose own B - A difference has p below 0.05.

tag property 

tag: str

unit property 

unit: str | None

good_share property 

good_share: float | None

GOOD rows over all rows of the target read.

censored property 

censored: bool | None

Whether a target sample sits at its engineering range limit.

None when the target declares no engineering range and no digital range state flags a sample.

frame property 

frame: DataFrame

One row per block: index, start, setting, kept samples of each analysis.

render 

render() -> str

The text tsdive switchback analyze prints.

Examples:

>>> import tsdive
>>> plan = tsdive.switchback_plan("2024-06-03T00:00:00Z", "2024-06-04T00:00:00Z",
...                               block="PT1H", washout="PT15M", seed=7)
>>> archives = ["data/switchback_demo/ti201.parquet",
...             "data/switchback_demo/fi200.parquet",
...             "data/switchback_demo/tt001.parquet"]
>>> result = tsdive.switchback_analyze(archives, plan, target="TI201.PV",
...                                    covariates=["FI200.PV", "TT001.PV"])
>>> print(result.render().splitlines()[0])
demo:TI201.PV   B - A +0.6024 degrees Celsius   p 0.154

to_dict 

to_dict() -> dict[str, object]

The document tsdive switchback analyze --json prints, ready for json.dumps.

The command adds result_kind and tsdive_version in front.