How to use the depccg.tools.reader.read_trees_guess_extension function in depccg

To help you get started, we’ve selected a few depccg examples, based on popular ways it is used in public projects.

Secure your code as it's written. Use Snyk Code to scan source code in minutes - no build needed - and fix issues immediately.

github masashi-y / depccg / depccg / tools / diff.py View on Github external
if __name__ == '__main__':
    parser = argparse.ArgumentParser('show diffs between the structures of two CCG trees')
    parser.add_argument('file1')
    parser.add_argument('file2', nargs='?', default=None)
    parser.add_argument('--max-output-num', default=50, type=int)
    parser.add_argument('--sampling', default='head', choices=['head', 'tail', 'random'])
    args = parser.parse_args()
    if args.file2:
        file1_trees = [(args.file1, tree) for _, _, tree in read_trees_guess_extension(args.file1)]
        file2_trees = [(args.file2, tree) for _, _, tree in read_trees_guess_extension(args.file2)]
        assert len(file1_trees) == len(file2_trees)
        tree_pairs = list(zip(file1_trees, file2_trees))
    else:
        file1_trees = ((args.file1, tree) for _, tree in read_trees_guess_extension(args.file1))
        tree_pairs = list(zip(file1_trees, file1_trees))

    print(to_diffs(tree_pairs, args.max_output_num, args.sampling))
github masashi-y / depccg / depccg / tools / visualize.py View on Github external
mathml_html = to_mathml([nbest])
        with (out_directory / filename).open('w') as f:
            print(mathml_html, file=f)
        words = nbest[0][0].word if isinstance(nbest[0], tuple) else nbest[0].word
        trs.append(f'{i}<a href="{filename}">{filename}</a>{words}')
    with (out_directory / 'index.html').open('w') as f:
        print(index_html.format('\n'.join(trs)), file=f)


if __name__ == '__main__':
    parser = argparse.ArgumentParser()
    parser.add_argument('PATH', help='path to either of *.auto, *.xml, *.jigg.xml, *.ptb')
    parser.add_argument('--mkdir', default=None, help='if specified, creates a directory')
    args = parser.parse_args()

    trees = [[tree] for _, _, tree in read_trees_guess_extension(args.PATH)]
    if not args.mkdir:
        print(to_mathml(trees))
    else:
        to_mathml_separate_files(trees, args.mkdir)
github masashi-y / depccg / depccg / tools / tagger.py View on Github external
parser.add_argument('--annotator',
                        default='spacy',
                        choices=english_annotator.keys(),
                        help='annotate POS, named entity, and lemmas using this library')
    parser.add_argument('-f',
                        '--format',
                        default='xml',
                        choices=['auto', 'xml', 'prolog', 'jigg_xml', 'jigg_xml_ccg2lambda', 'json'],
                        help='output format')
    parser.add_argument('--semantic-templates',
                        help='semantic templates used in "ccg2lambda" format output')
    args = parser.parse_args()

    annotate_fun = english_annotator[args.annotator]
    doc, trees = [], []
    for _, tokens, tree in read_trees_guess_extension(args.PATH):
        doc.append([token.word for token in tokens])
        trees.append([(tree, 0)])
    tagged_doc = annotate_fun(doc, tokenize=False)

    semantic_templates = args.semantic_templates or SEMANTIC_TEMPLATES[LANG]
    print_(trees,
           tagged_doc,
           format=args.format,
           lang=LANG,
           semantic_templates=semantic_templates)
github masashi-y / depccg / depccg / tools / ja / tagger.py View on Github external
parser.add_argument('--annotator',
                        default='janome',
                        choices=japanese_annotator.keys(),
                        help='annotate POS, named entity, and lemmas using this library')
    parser.add_argument('-f',
                        '--format',
                        default='jigg_xml',
                        choices=['prolog', 'jigg_xml', 'jigg_xml_ccg2lambda', 'json', 'deriv'],
                        help='output format')
    parser.add_argument('--semantic-templates',
                        help='semantic templates used in "ccg2lambda" format output')
    args = parser.parse_args()

    annotate_fun = japanese_annotator[args.annotator]
    doc, trees = [], []
    for _, tokens, tree in read_trees_guess_extension(args.PATH, lang='ja'):
        doc.append(tokens)
        trees.append([(tree, 0)])

    semantic_templates = args.semantic_templates or SEMANTIC_TEMPLATES[LANG]
    print_(trees,
           doc,
           format=args.format,
           lang=LANG,
           semantic_templates=semantic_templates)
github masashi-y / depccg / depccg / tools / diff.py View on Github external
result += f'<p><strong>{name1}</strong> ID={i}: {tree1.word}</p>'
        result += f'<math xmlns="http://www.w3.org/1998/Math/MathML">{tree1_str}</math>'
        result += f'<p><strong>{name2}</strong> ID={i}: {tree2.word}</p>'
        result += f'<math xmlns="http://www.w3.org/1998/Math/MathML">{tree2_str}</math>'
    return MATHML_MAIN.format(result)


if __name__ == '__main__':
    parser = argparse.ArgumentParser('show diffs between the structures of two CCG trees')
    parser.add_argument('file1')
    parser.add_argument('file2', nargs='?', default=None)
    parser.add_argument('--max-output-num', default=50, type=int)
    parser.add_argument('--sampling', default='head', choices=['head', 'tail', 'random'])
    args = parser.parse_args()
    if args.file2:
        file1_trees = [(args.file1, tree) for _, _, tree in read_trees_guess_extension(args.file1)]
        file2_trees = [(args.file2, tree) for _, _, tree in read_trees_guess_extension(args.file2)]
        assert len(file1_trees) == len(file2_trees)
        tree_pairs = list(zip(file1_trees, file2_trees))
    else:
        file1_trees = ((args.file1, tree) for _, tree in read_trees_guess_extension(args.file1))
        tree_pairs = list(zip(file1_trees, file1_trees))

    print(to_diffs(tree_pairs, args.max_output_num, args.sampling))
github masashi-y / depccg / depccg / tools / diff.py View on Github external
result += f'<math xmlns="http://www.w3.org/1998/Math/MathML">{tree1_str}</math>'
        result += f'<p><strong>{name2}</strong> ID={i}: {tree2.word}</p>'
        result += f'<math xmlns="http://www.w3.org/1998/Math/MathML">{tree2_str}</math>'
    return MATHML_MAIN.format(result)


if __name__ == '__main__':
    parser = argparse.ArgumentParser('show diffs between the structures of two CCG trees')
    parser.add_argument('file1')
    parser.add_argument('file2', nargs='?', default=None)
    parser.add_argument('--max-output-num', default=50, type=int)
    parser.add_argument('--sampling', default='head', choices=['head', 'tail', 'random'])
    args = parser.parse_args()
    if args.file2:
        file1_trees = [(args.file1, tree) for _, _, tree in read_trees_guess_extension(args.file1)]
        file2_trees = [(args.file2, tree) for _, _, tree in read_trees_guess_extension(args.file2)]
        assert len(file1_trees) == len(file2_trees)
        tree_pairs = list(zip(file1_trees, file2_trees))
    else:
        file1_trees = ((args.file1, tree) for _, tree in read_trees_guess_extension(args.file1))
        tree_pairs = list(zip(file1_trees, file1_trees))

    print(to_diffs(tree_pairs, args.max_output_num, args.sampling))