Interfaces for CIT implementations

class ITestCI

Interface specifying how to expose (custom) CIT-implementations. Supports the efficient dispatch of large collections of blocks.

See also

For use with the current mCIT implementation, also the interfaces for IProvideVarianceForCIT and potentially IProvideAnalyticQuantilesForCIT should be exposed. For details, see also Interfaces for Custom Stages.

class Result

Output format for (custom) CIT-implementations.

global_score: float

Score for the entire dataset.

dependent: bool

Dependency on the entire dataset.

block_scores: ndarray | None = None

Score for each block individually (if applicable).

__init__(global_score: float, dependent: bool, block_scores: ndarray | None = None) → None
run_single(data: CIT_DataPatterned) → Result

Run a single CIT on all data.

Parameters:

data (CIT_DataPatterned) – The data-set to use.

Returns:

Structured test-output.

Return type:

ITestCI.Result

run_many(data: BlockView) → Result

Run CITs on many blocks (efficiently).

Parameters:

data (BlockView) – The data-set to use.

Returns:

Structured test-output.

Return type:

ITestCI.Result

static ITestCI._extract_cache_id(fname: str, data: CIT_DataPatterned | BlockView) → tuple

Extract a cache-id from a given query. A fallback returning data.cache_id is provided; it is possible, but not typically necessary to overwrite this fallback.

See also

For a discussion of cache-IDs, see Cache IDs.

Parameters:
Returns:

extracted cache-id, here simply data.cache_id.

Return type:

tuple

class IProvideAnalyticQuantilesForCIT

Interface to expose for (custom) CIT-implementations if the homogeneity-test Homogeneity_Binomial is used. Quantiles can also be bootstrapped, if no implementation (None) of this interface is provided.

cit_quantile_estimate(data: BlockView, cit_result: Result, beta: float, cit_obj: ITestCI) → float

Provide an estimate of the \(\beta\)-quantile for the test implemented by cit_obj.

Parameters:

  • data (BlockView) – The data-blocks to operate on.

  • cit_result (ITestCI.Result) – The CIT result for the data (currently always run previously anyway).

  • beta (float) – The quantile \(\beta\) to estimate.

  • cit_obj (ITestCI) – The underlying cit-instance for which the quantile should be computed. (The present interface IProvideAnalyticQuantilesForCIT can, but does not have to, be exposed on the CIT-type itself.)

Returns:

Estimate of the dependence-value at the quantile \(\beta\)

Return type:

float

class IProvideVarianceForCIT

Interface to expose for (custom) CIT-implementations if one of the acceptance-interval tests (IndicatorImplication_AcceptanceInterval or IndicatorImplication_AcceptanceInterval) is used. These tests require variance-estimates for the block-wise dependence-scores provided by the CIT.

get_variance_estimate(N: int, dim_Z: int, cit_obj: ITestCI) → float

Get an estimate of the block-wise variance of the dependence score implemented by cit_obj.

Parameters:

  • N (int) – The sample count N.

  • dim_Z (int) – The dimension of (number of variables in) the condition conditioning set Z.

  • cit_obj (ITestCI) – The underlying cit-instance for which the variance should be computed. (The present interface IProvideVarianceForCIT can, but does not have to, be exposed on the CIT-type itself.)

Returns:

The estimated value of the variance.

Return type:

float

get_std_estimate(N: int, dim_Z: int, cit_obj: ITestCI) → float

Get an estimate of the block-wise standard-deviation of the dependence score implemented by cit_obj. Implementation is optional, if this method is not overridden, the square-root of the variance is used.

Parameters:

  • N (int) – The sample count N.

  • dim_Z (int) – The dimension of (number of variables in) the condition conditioning set Z.

  • cit_obj (ITestCI) – The underlying cit-instance for which the standard-deviation should be computed. (The present interface IProvideVarianceForCIT can, but does not have to, be exposed on the CIT-type itself.)

Returns:

The estimated value of the standard-deviation.

Return type:

float