colonel.base_rich_sentence_element module

Module providing the BaseRichSentenceElement class.

class colonel.base_rich_sentence_element.BaseRichSentenceElement(lemma=None, upos=None, xpos=None, feats=None, deps=None, **kwargs)

Bases: colonel.base_sentence_element.BaseSentenceElement

Abstract class containing basic information in common with some specific elements being part of a sentence.

It is compliant with the CoNLL-U format, in the sense that it provides a common foundation for elements of type word and empty nodes, which can be made up of a richer set of fields in comparison to other elements, such as the (multiword) tokens.

deps

Enhanced dependency graph, usually in the form of a list of head-deprel pairs.

It is compatible with CoNLL-U DEPS field.

You are free to assign to it any kind of value suitable for your project.

feats

List of morphological features from the universal feature inventory or from a defined language-specific extension.

It is compatible with CoNLL-U FEATS field.

You are free to assign to it any kind of value suitable for your project.

is_valid()

Returns whether or not the object can be considered valid, however ignoring the context of the sentence in which the word itself is possibly inserted.

An instance of type BaseRichSentenceElement is always considered valid, independently from any value of its attributes (it doesn’t provide any additional check to the overridden superclass method).

lemma

Lemma of the element.

It is compatible with CoNLL-U LEMMA field.

to_conllu()

Returns a CoNLL-U formatted representation of the element.

This method is expected to be overridden by each specific element.

upos

Universal part-of-speech tag.

It is compatible with CoNLL-U UPOS field.

xpos

Language-specific part-of-speech tag.

It is compatible with CoNLL-U XPOS field.