id_translation.mapping.matrix#

Functions and classes used by the Mapper for handling score matrices.

Warning

This module is considered an implementation detail, and may change without notice.

Classes

`Record`(value, candidate, score)	Data concerning a match.
`Reject`(record[, superseding_value, ...])	Data concerning the rejection of a match.
`ScoreHelper`(matrix, min_score[, logger, task_id])	High-level selection operations.
`ScoreMatrix`(values, candidates, *[, grid])	A matrix of match scores.

class Record(value, candidate, score)[source]#

Bases: Generic[ValueType, CandidateType]

Data concerning a match.

candidate#: A hashable candidate.

score#: Likeness score computed by some scoring function.

value#: A hashable value.

class Reject(record, superseding_value=None, superseding_candidate=None)[source]#

Bases: Generic[ValueType, CandidateType]

Data concerning the rejection of a match.

explain(min_score, full=False)[source]#

Create a string which explains the rejection.

Parameters:

min_score – Minimum score to accept a match.
full – If True show full information about superseding matches.

Returns:

An explanatory string.

record#: A Record to describe.

superseding_candidate = None#: A Record that prevents matching of the current candidate.

superseding_value = None#: A Record that prevents matching of the current value.

class ScoreHelper(matrix, min_score, logger=None, *, task_id=None)[source]#

Bases: Generic[ValueType, CandidateType]

High-level selection operations.

Parameters:

matrix – A ScoreMatrix instance.
min_score – Minimum score to make a value -> candidate match.
logger – Explicit Logger instance to use.
task_id – Used for logging.

above()[source]#: Records with scores above the threshold.

below()[source]#: Records with scores below the threshold.

property logger#: Return the Logger that is used by this instance.

to_directional_mapping(cardinality=None)[source]#

Create a DirectionalMapping with a given target Cardinality.

Parameters:: cardinality – Explicit cardinality to set, see cardinality. If None, use the actual cardinality when selecting all matches with scores at or above the minimum.
Returns:: A DirectionalMapping.

class ScoreMatrix(values, candidates, *, grid=None)[source]#

Bases: Generic[ValueType, CandidateType]

A matrix of match scores.

Parameters:

values – Iterable of elements to match to candidates.
candidates – Iterable of candidates to match with value. Duplicate elements will be discarded.
grid – Initial score matrix. Default is to fill with -inf.

Raises:

ValueError – If a bad grid is given.

property candidates#: Unique candidates in order.

get_finite_values()[source]#: Compute all finite values.

property size#: Total number of elements.

to_dict()[source]#: Convert to dict {(value, candidate): score}.

to_native_string(*, decimals=2, lines=True)[source]#: Format score table without pandas.

to_pandas()[source]#: Convert to pandas.DataFrame.

to_string(*, decimals=2)[source]#: Format score table.

property values#: Unique values in order.