redshred.enrichments package

Submodules

redshred.enrichments.base module

class redshred.enrichments.base.PerspectiveConfiguration(*, name: str, perspective: str, segments: SegmentQuery = None, description: str = '', config: Dict[str, Any] = None, debug: bool = False, **extra_data: Any)

Bases: BaseModel

A model representing a perspective configuration.

This model is used to define a perspective configuration, which includes the name, perspective, segments, description, config, and debug.

Attributes:

name: The canonical enrichment name to run. perspective: A human-readable label which must be unique. segments: A queries to run against the server, as well as any required prerequisites to process. description: A description of the perspective. config: Specific configuration options for the enrichment. debug: Always run if True.

class Config

Bases: _DefaultPydanticConfig

config: Dict[str, Any]

debug: bool

description: str

name: str

perspective: str

segments: SegmentQuery

class redshred.enrichments.base.SegmentQuery(*, queries: List[str] | Dict[str, Any] = None, prerequisites: List[str] = None, **extra_data: Any)

Bases: BaseModel

A model representing a segment query.

This model is used to define a segment query, which includes a list of queries and prerequisites and is used when defining input for a Perspective.

Attributes:

queries: A list of queries. prerequisites: A list of prerequisites in the form of perspective names.

class Config

Bases: _DefaultPydanticConfig

prerequisites: List[str]

queries: List[str] | Dict[str, Any]

redshred.enrichments.defined_acronyms module

class redshred.enrichments.defined_acronyms.DefinedAcronymsPerspective(*, name: Literal['defined_acronyms'] = 'defined_acronyms', perspective: str, segments: SegmentQuery = None, description: str = '', config: DefinedAcronymsPerspectiveConfig = None, debug: bool = False, **extra_data: Any)

Bases: PerspectiveConfiguration

This enrichment can extract acronyms and their abbreviations from a piece of text. Example: Alcoholics Anonymous (AA)

config: DefinedAcronymsPerspectiveConfig

name: Literal['defined_acronyms']

class redshred.enrichments.defined_acronyms.DefinedAcronymsPerspectiveConfig

Bases: BaseModel

This enrichment can extract acronyms and their abbreviations from a piece of text. Example: Alcoholics Anonymous (AA)

class Config

Bases: object

arbitrary_types_allowed = True

use_enum_values = True

redshred.enrichments.defined_acronyms.object_setattr(self, name, value, /)

Implement setattr(self, name, value).

redshred.enrichments.external_api module

class redshred.enrichments.external_api.ExternalAPIPerspective(*, name: Literal['external_api'] = 'external_api', perspective: str, segments: SegmentQuery = None, description: str = '', config: ExternalAPIPerspectiveConfig = None, debug: bool = False, **extra_data: Any)

Bases: PerspectiveConfiguration

Config class for the External API

config: ExternalAPIPerspectiveConfig

name: Literal['external_api']

class redshred.enrichments.external_api.ExternalAPIPerspectiveConfig(*, template: str, jmespath_source: str = '', template_variables: Dict = None, jmespath_transformation: str = '')

Bases: BaseModel

Config class for the External API

class Config

Bases: object

arbitrary_types_allowed = True

use_enum_values = True

jmespath_source: str

jmespath_transformation: str

template: str

template_variables: Dict

redshred.enrichments.external_api.object_setattr(self, name, value, /)

Implement setattr(self, name, value).

redshred.enrichments.grouper module

class redshred.enrichments.grouper.GrouperPerspective(*, name: Literal['grouper'] = 'grouper', perspective: str, segments: SegmentQuery = None, description: str = '', config: GrouperPerspectiveConfig = None, debug: bool = False, **extra_data: Any)

Bases: PerspectiveConfiguration

Sane default BaseModel

config: GrouperPerspectiveConfig

name: Literal['grouper']

class redshred.enrichments.grouper.GrouperPerspectiveConfig(*, operations: str = 'xy', operation_labels: List[str] = None, root_label: str = 'root', x_gap: str = '1%', y_gap: str = '1%', whitespace_calculation_method: str = 'average', whitespace_method_options: Dict = None, hull_method: GrouperPerspectiveHullMethod = 'convex_hull', hull_method_options: Dict = None)

Bases: BaseModel

Sane default BaseModel

class Config

Bases: object

arbitrary_types_allowed = True

use_enum_values = True

hull_method: GrouperPerspectiveHullMethod

hull_method_options: Dict

operation_labels: List[str]

operations: str

root_label: str

whitespace_calculation_method: str

whitespace_method_options: Dict

x_gap: str

y_gap: str

class redshred.enrichments.grouper.GrouperPerspectiveHullMethod(value)

Bases: str, Enum

Method for calculating the hull of the group, either convex_hull, concave_hull, bounding_box, minimum_rotated_rectangle, or as None/null to return full multipolygon output

bounding_box = 'bounding_box'

concave_hull = 'concave_hull'

convex_hull = 'convex_hull'

minimum_rotated_rectangle = 'minimum_rotated_rectangle'

redshred.enrichments.grouper.object_setattr(self, name, value, /)

Implement setattr(self, name, value).

redshred.enrichments.huggingface module

class redshred.enrichments.huggingface.HuggingfacePerspective(*, name: Literal['huggingface'] = 'huggingface', perspective: str, segments: SegmentQuery = None, description: str = '', config: HuggingfacePerspectiveConfig = None, debug: bool = False, **extra_data: Any)

Bases: PerspectiveConfiguration

Sane default BaseModel

config: HuggingfacePerspectiveConfig

name: Literal['huggingface']

class redshred.enrichments.huggingface.HuggingfacePerspectiveConfig(*, pipeline_task: str, task_specific_template: str = '', tokenizer: str = '', model: str, model_source: str = 'huggingface', tokenizer_class: str = 'auto', model_class: str = 'auto', task_config: Dict = None, task_config_class: str = 'auto')

Bases: BaseModel

Sane default BaseModel

class Config

Bases: object

arbitrary_types_allowed = True

use_enum_values = True

model: str

model_class: str

model_source: str

pipeline_task: str

task_config: Dict

task_config_class: str

task_specific_template: str

tokenizer: str

tokenizer_class: str

redshred.enrichments.huggingface.object_setattr(self, name, value, /)

Implement setattr(self, name, value).

redshred.enrichments.iris module

class redshred.enrichments.iris.IrisPerspective(*, name: Literal['iris'] = 'iris', perspective: str, segments: SegmentQuery = None, description: str = '', config: IrisPerspectiveConfig = None, debug: bool = False, **extra_data: Any)

Bases: PerspectiveConfiguration

Enrichment configuration documentation Make sure you fill this out fully, since if it is not included here the configuration will not work

config: IrisPerspectiveConfig

name: Literal['iris']

class redshred.enrichments.iris.IrisPerspectiveConfig(*, model: str = 'iris-v0.01', confidence_threshold: float = 0.8, nms_threshold: float = 0.8)

Bases: BaseModel

Enrichment configuration documentation Make sure you fill this out fully, since if it is not included here the configuration will not work

class Config

Bases: object

arbitrary_types_allowed = True

use_enum_values = True

confidence_threshold: float

model: str

nms_threshold: float

redshred.enrichments.iris.object_setattr(self, name, value, /)

Implement setattr(self, name, value).

redshred.enrichments.page_images module

class redshred.enrichments.page_images.BackendOptions(*, cropbox: bool = True)

Bases: BaseModel

cropbox: bool

class redshred.enrichments.page_images.PageImagesPerspective(*, name: Literal['page_images'] = 'page_images', perspective: str, segments: SegmentQuery = None, description: str = '', config: PageImagesPerspectiveConfig = None, debug: bool = False, **extra_data: Any)

Bases: PerspectiveConfiguration

Generate preview images for all pages.

config: PageImagesPerspectiveConfig

name: Literal['page_images']

class redshred.enrichments.page_images.PageImagesPerspectiveBackend(value)

Bases: str, Enum

PDF image generator backend to use

pdfium = 'pdfium'

pdftoppm = 'pdftoppm'

class redshred.enrichments.page_images.PageImagesPerspectiveConfig(*, dpi: int = 150, backend: PageImagesPerspectiveBackend = 'pdftoppm', antialiasing: bool = True, backend_options: dict = None, show_annotations: bool = True, pages: list[str | int] | None = None)

Bases: BaseModel

Generate preview images for all pages.

class Config

Bases: object

arbitrary_types_allowed = True

use_enum_values = True

antialiasing: bool

backend: PageImagesPerspectiveBackend

backend_options: dict

dpi: int

pages: list[str | int] | None

show_annotations: bool

redshred.enrichments.page_images.object_setattr(self, name, value, /)

Implement setattr(self, name, value).

redshred.enrichments.pdftotext module

class redshred.enrichments.pdftotext.PdftotextPerspective(*, name: Literal['pdftotext'] = 'pdftotext', perspective: str, segments: SegmentQuery = None, description: str = '', config: PdftotextPerspectiveConfig = None, debug: bool = False, **extra_data: Any)

Bases: PerspectiveConfiguration

Generate preview images for all pages.

config: PdftotextPerspectiveConfig

name: Literal['pdftotext']

class redshred.enrichments.pdftotext.PdftotextPerspectiveConfig(*, layout: bool = True)

Bases: BaseModel

Generate preview images for all pages.

class Config

Bases: object

arbitrary_types_allowed = True

use_enum_values = True

layout: bool

redshred.enrichments.pdftotext.object_setattr(self, name, value, /)

Implement setattr(self, name, value).

redshred.enrichments.preprocess module

class redshred.enrichments.preprocess.LemmatizingStep(*, operation: str = 'lemmatize', comment: str = '', method: LemmatizingStepMethod = 'wordnet')

Bases: BaseModel

Sane default BaseModel

comment: str

method: LemmatizingStepMethod

operation: str

class redshred.enrichments.preprocess.LemmatizingStepMethod(value)

Bases: str, Enum

An enumeration.

wordnet = 'wordnet'

class redshred.enrichments.preprocess.PreprocessPerspective(*, name: Literal['preprocess'] = 'preprocess', perspective: str, segments: SegmentQuery = None, description: str = '', config: PreprocessPerspectiveConfig = None, debug: bool = False, **extra_data: Any)

Bases: PerspectiveConfiguration

Configuration for the RegEx enrichment

config: PreprocessPerspectiveConfig

name: Literal['preprocess']

class redshred.enrichments.preprocess.PreprocessPerspectiveConfig(*, steps: List[StringMethodStep | RegexStep | StopwordStep | StemmingStep | LemmatizingStep])

Bases: BaseModel

Configuration for the RegEx enrichment

class Config

Bases: object

arbitrary_types_allowed = True

use_enum_values = True

steps: List[StringMethodStep | RegexStep | StopwordStep | StemmingStep | LemmatizingStep]

class redshred.enrichments.preprocess.RegexStep(*, operation: str = 'preprocessing-pipeline_replace', comment: str = '', pattern: str, replacement: str, flags: int = 0)

Bases: BaseModel

A single step in the preprocessing pipeline

comment: str

flags: int

operation: str

pattern: str

replacement: str

class redshred.enrichments.preprocess.StemmingStep(*, operation: str = 'stem', comment: str = '', method: StemmingStepMethod = 'porter')

Bases: BaseModel

Sane default BaseModel

comment: str

method: StemmingStepMethod

operation: str

class redshred.enrichments.preprocess.StemmingStepMethod(value)

Bases: str, Enum

An enumeration.

porter = 'porter'

class redshred.enrichments.preprocess.StopwordStep(*, operation: str = 'stopword', comment: str = '', stopwords: StopwordStepStopwords = 'nltk', case_sensitive: bool = False, normalize_whitespace: bool = True)

Bases: BaseModel

A single step in the preprocessing pipeline

case_sensitive: bool

comment: str

normalize_whitespace: bool

operation: str

stopwords: StopwordStepStopwords

class redshred.enrichments.preprocess.StopwordStepStopwords(value)

Bases: str, Enum

An enumeration.

nltk = 'nltk'

class redshred.enrichments.preprocess.StringMethodStep(*, operation: str = 'string_method', comment: str = '', method: StringMethodStepMethod)

Bases: BaseModel

A single step in the preprocessing pipeline

comment: str

method: StringMethodStepMethod

operation: str

class redshred.enrichments.preprocess.StringMethodStepMethod(value)

Bases: str, Enum

An enumeration.

capitalize = 'capitalize'

lower = 'lower'

title = 'title'

upper = 'upper'

redshred.enrichments.preprocess.object_setattr(self, name, value, /)

Implement setattr(self, name, value).

redshred.enrichments.regex module

class redshred.enrichments.regex.RegexPattern(*, label: str, flags: str = '', pattern: str)

Bases: BaseModel

Each Regex Pattern defined here will be queried against the input text(s). This allows you to define multiple patterns that you want to run across a single group of inputs in a single batch.

flags: str

label: str

pattern: str

class redshred.enrichments.regex.RegexPerspective(*, name: Literal['regex'] = 'regex', perspective: str, segments: SegmentQuery = None, description: str = '', config: RegexPerspectiveConfig = None, debug: bool = False, **extra_data: Any)

Bases: PerspectiveConfiguration

Configuration settings for the RegEx Enrichment

config: RegexPerspectiveConfig

name: Literal['regex']

class redshred.enrichments.regex.RegexPerspectiveConfig(*, search_patterns: List[RegexPattern])

Bases: BaseModel

Configuration settings for the RegEx Enrichment

class Config

Bases: object

arbitrary_types_allowed = True

use_enum_values = True

search_patterns: List[RegexPattern]

redshred.enrichments.regex.object_setattr(self, name, value, /)

Implement setattr(self, name, value).

redshred.enrichments.sentences module

class redshred.enrichments.sentences.SentencesPerspective(*, name: Literal['sentences'] = 'sentences', perspective: str, segments: SegmentQuery = None, description: str = '', config: SentencesPerspectiveConfig = None, debug: bool = False, **extra_data: Any)

Bases: PerspectiveConfiguration

Generate preview images for all pages.

config: SentencesPerspectiveConfig

name: Literal['sentences']

class redshred.enrichments.sentences.SentencesPerspectiveConfig(*, method: SentencesPerspectiveMethod = 'nltk')

Bases: BaseModel

Generate preview images for all pages.

class Config

Bases: object

arbitrary_types_allowed = True

use_enum_values = True

method: SentencesPerspectiveMethod

class redshred.enrichments.sentences.SentencesPerspectiveMethod(value)

Bases: str, Enum

An enumeration.

nltk = 'nltk'

pysbd = 'pysbd'

spacy = 'spacy'

redshred.enrichments.sentences.object_setattr(self, name, value, /)

Implement setattr(self, name, value).

redshred.enrichments.spacy module

class redshred.enrichments.spacy.SpacyPerspective(*, name: Literal['spacy'] = 'spacy', perspective: str, segments: SegmentQuery = None, description: str = '', config: SpacyPerspectiveConfig = None, debug: bool = False, **extra_data: Any)

Bases: PerspectiveConfiguration

Configuration for the RegEx enrichment

config: SpacyPerspectiveConfig

name: Literal['spacy']

class redshred.enrichments.spacy.SpacyPerspectiveConfig(*, text_field: str = '$.text', model: str = 'en_core_web_sm', task: SpacyPerspectiveTask = 'noun_chunks')

Bases: BaseModel

Configuration for the RegEx enrichment

class Config

Bases: object

arbitrary_types_allowed = True

use_enum_values = True

model: str

task: SpacyPerspectiveTask

text_field: str

class redshred.enrichments.spacy.SpacyPerspectiveTask(value)

Bases: str, Enum

An enumeration.

dep = 'dep'

morphology = 'morphology'

ner = 'ner'

noun_chunks = 'noun_chunks'

pos = 'pos'

spans = 'spans'

vectors = 'vectors'

redshred.enrichments.spacy.object_setattr(self, name, value, /)

Implement setattr(self, name, value).

redshred.enrichments.tfidf module

class redshred.enrichments.tfidf.TFIDFPerspective(*, name: Literal['tfidf'] = 'tfidf', perspective: str, segments: SegmentQuery = None, description: str = '', config: TFIDFPerspectiveConfig = None, debug: bool = False, **extra_data: Any)

Bases: PerspectiveConfiguration

Sane default BaseModel

config: TFIDFPerspectiveConfig

name: Literal['tfidf']

class redshred.enrichments.tfidf.TFIDFPerspectiveConfig(*, stop_words: None = None, norm: TFIDFPerspectiveNorm = 'l2', ngram_range: List[int] = None, lowercase: bool = True, max_df: float = 1.0, min_df: None = None, max_features: int = 5000)

Bases: BaseModel

Sane default BaseModel

class Config

Bases: object

arbitrary_types_allowed = True

use_enum_values = True

lowercase: bool

max_df: float

max_features: int

min_df: None

ngram_range: List[int]

norm: TFIDFPerspectiveNorm

stop_words: None

class redshred.enrichments.tfidf.TFIDFPerspectiveNorm(value)

Bases: str, Enum

See ‘https://scikit-learn.org/stable/modules/generated/sklearn.feature_extraction.text.TfidfVectorizer.html’ for documentation

l1 = 'l1'

l2 = 'l2'

redshred.enrichments.tfidf.object_setattr(self, name, value, /)

Implement setattr(self, name, value).

redshred.enrichments.typography module

class redshred.enrichments.typography.TypographyPerspective(*, name: Literal['typography'] = 'typography', perspective: str, segments: SegmentQuery = None, description: str = '', config: TypographyPerspectiveConfig = None, debug: bool = False, **extra_data: Any)

Bases: PerspectiveConfiguration

config: TypographyPerspectiveConfig

name: Literal['typography']

class redshred.enrichments.typography.TypographyPerspectiveConfig(*, options: Dict = None, segmenter: TypographyPerspectiveSegmenter = 'pdftotext')

Bases: BaseModel

class Config

Bases: object

arbitrary_types_allowed = True

use_enum_values = True

options: Dict

segmenter: TypographyPerspectiveSegmenter

class redshred.enrichments.typography.TypographyPerspectiveSegmenter(value)

Bases: str, Enum

An enumeration.

pdfminer = 'pdfminer'

pdftotext = 'pdftotext'

tesseract = 'tesseract'

redshred.enrichments.typography.object_setattr(self, name, value, /)

Implement setattr(self, name, value).

Module contents