Function main

python/ctranslate2/converters/fairseq.py:304–358

()

def main():
    parser = argparse.ArgumentParser(
        formatter_class=argparse.ArgumentDefaultsHelpFormatter
    )
    parser.add_argument("--model_path", required=True, help="Model path.")
    parser.add_argument(
        "--data_dir",
        required=True,
        help="Data directory containing the source and target vocabularies.",
    )
    parser.add_argument(
        "--user_dir",
        help="Directory containing custom extensions.",
    )
    parser.add_argument(
        "--fixed_dictionary",
        help="Fixed dictionary for multilingual models.",
    )
    parser.add_argument(
        "--source_lang",
        help="Source language. This argument is used to find dictionary file from `data_dir`.",
    )
    parser.add_argument(
        "--target_lang",
        help="Target language. This argument is used to find dictionary file from `data_dir`.",
    )
    parser.add_argument(
        "--no_default_special_tokens",
        action="store_true",
        help=(
            "Require all special tokens to be provided by the user during inference, "
            "including the decoder start token."
        ),
    )
    parser.add_argument(
        "--unsafe_deserialization",
        action="store_true",
        help=(
            "Allow loading legacy checkpoints with unsafe pickle deserialization. "
            "Only enable this option for trusted checkpoints."
        ),
    )
    Converter.declare_arguments(parser)
    args = parser.parse_args()
    converter = FairseqConverter(
        args.model_path,
        args.data_dir,
        source_lang=args.source_lang,
        target_lang=args.target_lang,
        fixed_dictionary=args.fixed_dictionary,
        no_default_special_tokens=args.no_default_special_tokens,
        user_dir=args.user_dir,
        unsafe_deserialization=args.unsafe_deserialization,
    )
    converter.convert_from_args(args)


if __name__ == "__main__":