def load_nlp_pipeline()

in ludwig/utils/nlp_utils.py [0:0]

• 40 lines of code
• 5 McCabe index (conditional complexity)

def load_nlp_pipeline(language='xx'):
    if language not in language_module_registry:
        logger.error(
            'Language {} is not supported.'
            'Suported languages are: {}'.format(
                language,
                language_module_registry.keys()
            ))
        raise ValueError
    else:
        spacy_module_name = language_module_registry[language]
    global nlp_pipelines
    if nlp_pipelines[language] is None:
        logger.info('Loading NLP pipeline')
        try:
            import spacy
        except ImportError:
            logger.error(
                ' spacy is not installed. '
                'In order to install all text feature dependencies run '
                'pip install ludwig[text]'
            )
            sys.exit(-1)

        try:
            nlp_pipelines[language] = spacy.load(
                spacy_module_name,
                disable=['parser', 'tagger', 'ner']
            )
        except OSError:
            logger.info(
                ' spaCy {} model is missing, downloading it '
                '(this will only happen once)'
            )
            from spacy.cli import download
            download(spacy_module_name)
            nlp_pipelines[language] = spacy.load(
                spacy_module_name,
                disable=['parser', 'tagger', 'ner']
            )

    return nlp_pipelines[language]