hubconf.py

# Copyright (c) Facebook, Inc. and its affiliates.
#
# Source code slightly modified to work with GreynirSeq.
#
# This source code is licensed under the MIT license found in the
# LICENSE file in the root directory of this source tree.

"""isort:skip_file"""

import functools
import importlib

dependencies = ["dataclasses", "hydra", "numpy", "omegaconf", "regex", "requests", "torch", "fairseq"]


# Check for required dependencies and raise a RuntimeError if any are missing.
missing_deps = []
for dep in dependencies:
    try:
        importlib.import_module(dep)
    except ImportError:
        # Hack: the hydra package is provided under the "hydra-core" name in
        # pypi. We don't want the user mistakenly calling `pip install hydra`
        # since that will install an unrelated package.
        if dep == "hydra":
            dep = "hydra-core"
        missing_deps.append(dep)
if len(missing_deps) > 0:
    raise RuntimeError("Missing dependencies: {}".format(", ".join(missing_deps)))


# only do fairseq imports after checking for dependencies
from fairseq.hub_utils import (  # noqa; noqa
    BPEHubInterface as bpe,
    TokenizerHubInterface as tokenizer,
)

# We import the greynirseq.nicenlp.models to ensure that all those models are added to the fairseq MODEL_REGISTRY
# This is done via the @register_model decorator.
from greynirseq.nicenlp import models  # noqa
from fairseq.models import MODEL_REGISTRY  # noqa


# torch.hub doesn't build Cython components, so if they are not found then try
# to build them here
try:
    import fairseq.data.token_block_utils_fast  # noqa
except ImportError:
    try:
        import cython  # noqa
        import os
        from setuptools import sandbox

        sandbox.run_setup(
            os.path.join(os.path.dirname(__file__), "setup.py"),
            ["build_ext", "--inplace"],
        )
    except ImportError:
        print(
            "Unable to build Cython components. Please make sure Cython is "
            "installed if the torch.hub model you are loading depends on it."
        )

# automatically expose models defined in FairseqModel::hub_models
for _model_type, _cls in MODEL_REGISTRY.items():
    for model_name in _cls.hub_models().keys():
        globals()[model_name] = functools.partial(
            _cls.from_pretrained,
            model_name,
        )