nvm.aux_spacy.factories package

Submodules

nvm.aux_spacy.factories.get_doc_basic_metrics module

nvm.aux_spacy.factories.get_doc_basic_metrics.get_doc_basic_metrics_component(nlp, name, log0)

Get Doc basic metrics.

Examples

>>> import spacy
>>> from dframcy import DframCy
>>>
>>> from nvm import disp_df
>>> from nvm.aux_spacy import get_doc_basic_metrics_component
>>>
>>> nlp = spacy.load("en_core_web_sm")
>>> nlp.add_pipe("get_doc_basic_metrics", "BASIC")
>>>
>>> dframcy = DframCy(nlp)
>>>
>>> doc = dframcy.nlp(
>>>     "This sentence contains two verbs and this is how many verbs should be found."
>>> )
>>>
>>> df0 = dframcy.to_dataframe(
>>>     doc,
>>>     columns=["text", "lemma_", "is_alpha", "pos_", "tag_", "is_sent_start"],
>>>     custom_attributes=tok_exts[:12],
>>> )
>>> disp_df(df0)

class nvm.aux_spacy.factories.get_doc_basic_metrics.DocBasicMetricsComponent(nlp, log0=<Logger dummy (WARNING)>)

Bases: object

DocBasicMetricsComponent.

__call__:

some description

DocBasicMetricsComponent.

nvm.aux_spacy.factories.get_doc_count_of_dict_items module

nvm.aux_spacy.factories.get_doc_count_of_dict_items.get_doc_count_of_dict_items_component(nlp, name, dict0, prefix, suffix, exclude, pos, tag, log0)

class nvm.aux_spacy.factories.get_doc_count_of_dict_items.CountDictItemsComponent(nlp, dict0, name, prefix=None, suffix=None, exclude=None, pos=None, tag=None, log0=<Logger dummy (WARNING)>)

Bases: object

Get counts of items from arbitrary LIWC-like dictionary.

Examples

>>> from nvm import disp_df
>>> from nvm import Log0
>>> logZ = Log0()
>>> log0 = logZ.logger
>>>
>>> import textwrap
>>> import srsly
>>> import spacy
>>> from spacy.tokens.underscore import Underscore
>>>
>>> from dframcy import DframCy
>>>
>>> from nvm import jsonable
>>> from nvm.aux_spacy import get_doc_count_of_dict_items_component
>>> from nvm.aux_spacy import get_doc_summary_dict_component
>>>
>>> dict0 = {"pos": ["good", "marvel*"], "neg": ["bad", "awful*"]}
>>>
>>> config0 = dict(
>>>     dict0=dict0,
>>> )
>>> config1 = dict(
>>>     dict0=dict0,
>>>     pos = ["PROPN"],
>>> )
>>> nlp = spacy.load("en_core_web_sm")
>>>
>>> nlp.add_pipe("get_doc_count_of_dict_items", "LEX0", config=config0)
>>> nlp.add_pipe("get_doc_count_of_dict_items", "LEX1", config=config1)
>>> nlp.add_pipe("get_doc_summary_dict", "SUMMARY")
>>>
>>> dframcy = DframCy(nlp)
>>>
>>> doc = dframcy.nlp(
>>>     "GoOd. Bad Good WhatEver Awful Marvelous."
>>>     "toobad not-marvelous unmarvel goodyear badZ bAD."
>>>     "Bad Bad WhatEver Awful Marvelous."
>>> )
>>>
>>> tok_exts = list(Underscore.token_extensions.keys())
>>> doc_exts = list(Underscore.doc_extensions.keys())
>>>
>>> df0 = dframcy.to_dataframe(
>>>     doc,
>>>     columns=["text", "lemma_", "pos_", "tag_"],
>>>     custom_attributes=tok_exts[:12],
>>> )
>>> disp_df(df0)
>>>
>>> print(nlp.pipe_names)
>>> print(tok_exts)
>>> print(doc_exts)
>>>
>>> print(textwrap.indent(srsly.yaml_dumps(jsonable(dict(doc._.SUMMARY))), '   '))

nvm.aux_spacy.factories.get_doc_sentences module

nvm.aux_spacy.factories.get_doc_sentences.get_doc_sentences_as_list_component(nlp, name, log0)

Get document sentences as a list.

Examples

>>> import spacy
>>> from nvm.aux_spacy import get_doc_sentences_as_list_component
>>> # nlp = spacy.blank("en")
>>> nlp = spacy.load("en_core_web_sm")
>>> nlp.add_pipe("get_doc_sentences_as_list", "SENTS")
>>> doc = nlp("This is the first sentence. This is the second sentence.")
>>> assert len(doc._.sents) == 2
>>> doc._.sents
['This is the first sentence.', 'This is the second sentence.']

class nvm.aux_spacy.factories.get_doc_sentences.DocSentsAsListComponent(nlp, log0=<Logger dummy (WARNING)>)

Bases: object

nvm.aux_spacy.factories.get_doc_summary_dict module

nvm.aux_spacy.factories.get_doc_summary_dict.get_doc_summary_dict_component(nlp, name, exclude, add_text, log0)

Get underscore attributes as dictionary.

Important

CAUTION: Add this to nlp.pipe after elements that need to be in the summary dictionary.

Examples

>>> import textwrap
>>> import srsly
>>> import spacy
>>> from nvm import jsonable
>>> from nvm.aux_spacy import get_doc_summary_dict_component
>>> from nvm.aux_spacy import get_doc_basic_metrics_component
>>>
>>> nlp = spacy.load("en_core_web_sm")
>>> nlp.add_pipe("get_doc_basic_metrics", "BASIC")
>>> nlp.add_pipe("get_doc_summary_dict", "SUMMARY", last=True)  # Add AFTER other elements
>>>
>>> doc = nlp("This is the first sentence. This is the second sentence.")
>>>
>>> print(textwrap.indent(srsly.yaml_dumps(jsonable(dict(doc._.SUMMARY))), '   '))

class nvm.aux_spacy.factories.get_doc_summary_dict.DocSummaryDictComponent(nlp, exclude=None, add_text=False, log0=<Logger dummy (WARNING)>)

Bases: object

nvm.aux_spacy.factories.get_doc_word_count module

nvm.aux_spacy.factories.get_doc_word_count.get_doc_word_count_component(nlp, name, log0)

Get Doc word count.

Examples

>>> import spacy
>>> from nvm.aux_spacy import get_doc_word_count_component
>>> nlp = spacy.load("en_core_web_sm")
>>> nlp.add_pipe("get_doc_word_count", "WC")
>>> doc = nlp("One two three four five.")
>>> assert doc._.word_count == 5
>>> doc._.word_count

class nvm.aux_spacy.factories.get_doc_word_count.DocWordCountComponent(nlp, log0=<Logger dummy (WARNING)>)

Bases: object

Module contents