Skip to content

GitLab

Unverified Commit 4d4d6bad authored by Cyrus Leung's avatar Cyrus Leung Committed by GitHub
Browse files

[Chore] Separate out `vllm.utils.importlib` (#27022) 


Signed-off-by: default avatarDarkLight1337 <tlleungac@connect.ust.hk>
parent 11ae016b
Hide whitespace changes
Inline Side-by-side
Showing with 346 additions and 331 deletions
vllm/model_executor/model_loader/tensorizer.py
View file @ 4d4d6bad
......@@ -26,7 +26,8 @@ from vllm.config import ModelConfig, ParallelConfig, VllmConfig, set_current_vll
from vllm.logger import init_logger
from vllm.model_executor.layers.vocab_parallel_embedding import VocabParallelEmbedding
from vllm.platforms import current_platform
from vllm.utils import FlexibleArgumentParser, PlaceholderModule
from vllm.utils import FlexibleArgumentParser
from vllm.utils.import_utils import PlaceholderModule
if TYPE_CHECKING:
from vllm.engine.arg_utils import EngineArgs
......
vllm/model_executor/model_loader/weight_utils.py
View file @ 4d4d6bad
......@@ -34,7 +34,7 @@ from vllm.model_executor.layers.quantization import (
get_quantization_config,
)
from vllm.platforms import current_platform
from vllm.utils import PlaceholderModule
from vllm.utils.import_utils import PlaceholderModule
try:
from runai_model_streamer import SafetensorsStreamer
......
vllm/multimodal/audio.py
View file @ 4d4d6bad
......@@ -8,7 +8,7 @@ from typing import Literal
import numpy as np
import numpy.typing as npt
from vllm.utils import PlaceholderModule
from vllm.utils.import_utils import PlaceholderModule
from .base import MediaIO
......
vllm/multimodal/inputs.py
View file @ 4d4d6bad
......@@ -22,8 +22,8 @@ from typing import (
import numpy as np
from typing_extensions import NotRequired, TypeVar, deprecated
from vllm.utils import LazyLoader
from vllm.utils.collections import full_groupby, is_list_of
from vllm.utils.import_utils import LazyLoader
from vllm.utils.jsontree import json_map_leaves
if TYPE_CHECKING:
......
vllm/multimodal/parse.py
View file @ 4d4d6bad
......@@ -19,8 +19,8 @@ import numpy as np
import torch
from typing_extensions import assert_never
from vllm.utils import LazyLoader
from vllm.utils.collections import is_list_of
from vllm.utils.import_utils import LazyLoader
from .audio import AudioResampler
from .inputs import (
......
vllm/platforms/__init__.py
View file @ 4d4d6bad
......@@ -7,7 +7,8 @@ from typing import TYPE_CHECKING
from vllm import envs
from vllm.plugins import PLATFORM_PLUGINS_GROUP, load_plugins_by_group
from vllm.utils import resolve_obj_by_qualname, supports_xccl
from vllm.utils import supports_xccl
from vllm.utils.import_utils import resolve_obj_by_qualname
from .interface import CpuArchEnum, Platform, PlatformEnum
......
vllm/plugins/io_processors/__init__.py
View file @ 4d4d6bad
......@@ -6,7 +6,7 @@ import logging
from vllm.config import VllmConfig
from vllm.plugins import IO_PROCESSOR_PLUGINS_GROUP, load_plugins_by_group
from vllm.plugins.io_processors.interface import IOProcessor
from vllm.utils import resolve_obj_by_qualname
from vllm.utils.import_utils import resolve_obj_by_qualname
logger = logging.getLogger(__name__)
......
vllm/reasoning/abs_reasoning_parsers.py
View file @ 4d4d6bad
......@@ -8,8 +8,8 @@ from functools import cached_property
from typing import TYPE_CHECKING, Any
from vllm.logger import init_logger
from vllm.utils import import_from_path
from vllm.utils.collections import is_list_of
from vllm.utils.import_utils import import_from_path
if TYPE_CHECKING:
from vllm.entrypoints.openai.protocol import (
......
vllm/transformers_utils/runai_utils.py
View file @ 4d4d6bad
......@@ -9,7 +9,7 @@ import signal
from vllm import envs
from vllm.assets.base import get_cache_dir
from vllm.logger import init_logger
from vllm.utils import PlaceholderModule
from vllm.utils.import_utils import PlaceholderModule
logger = init_logger(__name__)
......
vllm/transformers_utils/s3_utils.py
View file @ 4d4d6bad
......@@ -4,7 +4,7 @@
import fnmatch
from typing import TYPE_CHECKING, Optional
from vllm.utils import PlaceholderModule
from vllm.utils.import_utils import PlaceholderModule
if TYPE_CHECKING:
from botocore.client import BaseClient
......
vllm/utils/__init__.py
View file @ 4d4d6bad
......@@ -8,8 +8,6 @@ import gc
import getpass
import hashlib
import importlib
import importlib.metadata
import importlib.util
import inspect
import ipaddress
import json
......@@ -25,7 +23,6 @@ import textwrap
import threading
import time
import traceback
import types
import uuid
import warnings
import weakref
......@@ -68,7 +65,6 @@ import zmq.asyncio
from packaging import version
from packaging.version import Version
from torch.library import Library
from typing_extensions import Never
import vllm.envs as envs
from vllm.logger import enable_trace_function_call, init_logger
......@@ -801,8 +797,6 @@ def find_nccl_include_paths() -> list[str] | None:
paths.append(inc)
try:
import importlib.util
spec = importlib.util.find_spec("nvidia.nccl")
if spec and getattr(spec, "submodule_search_locations", None):
for loc in spec.submodule_search_locations:
......@@ -1560,253 +1554,6 @@ def get_cuda_view_from_cpu_tensor(cpu_tensor: torch.Tensor) -> torch.Tensor:
return torch.ops._C.get_cuda_view_from_cpu_tensor(cpu_tensor)
def import_from_path(module_name: str, file_path: str | os.PathLike):
"""
Import a Python file according to its file path.
Based on the official recipe:
https://docs.python.org/3/library/importlib.html#importing-a-source-file-directly
"""
spec = importlib.util.spec_from_file_location(module_name, file_path)
if spec is None:
raise ModuleNotFoundError(f"No module named '{module_name}'")
assert spec.loader is not None
module = importlib.util.module_from_spec(spec)
sys.modules[module_name] = module
spec.loader.exec_module(module)
return module
@cache
def get_vllm_optional_dependencies():
metadata = importlib.metadata.metadata("vllm")
requirements = metadata.get_all("Requires-Dist", [])
extras = metadata.get_all("Provides-Extra", [])
return {
extra: [
re.split(r";|>=|<=|==", req)[0]
for req in requirements
if req.endswith(f'extra == "{extra}"')
]
for extra in extras
}
class _PlaceholderBase:
"""
Disallows downstream usage of placeholder modules.
We need to explicitly override each dunder method because
[`__getattr__`][vllm.utils._PlaceholderBase.__getattr__]
is not called when they are accessed.
Info:
[Special method lookup](https://docs.python.org/3/reference/datamodel.html#special-lookup)
"""
def __getattr__(self, key: str) -> Never:
"""
The main class should implement this to throw an error
for attribute accesses representing downstream usage.
"""
raise NotImplementedError
# [Basic customization]
def __lt__(self, other: object):
return self.__getattr__("__lt__")
def __le__(self, other: object):
return self.__getattr__("__le__")
def __eq__(self, other: object):
return self.__getattr__("__eq__")
def __ne__(self, other: object):
return self.__getattr__("__ne__")
def __gt__(self, other: object):
return self.__getattr__("__gt__")
def __ge__(self, other: object):
return self.__getattr__("__ge__")
def __hash__(self):
return self.__getattr__("__hash__")
def __bool__(self):
return self.__getattr__("__bool__")
# [Callable objects]
def __call__(self, *args: object, **kwargs: object):
return self.__getattr__("__call__")
# [Container types]
def __len__(self):
return self.__getattr__("__len__")
def __getitem__(self, key: object):
return self.__getattr__("__getitem__")
def __setitem__(self, key: object, value: object):
return self.__getattr__("__setitem__")
def __delitem__(self, key: object):
return self.__getattr__("__delitem__")
# __missing__ is optional according to __getitem__ specification,
# so it is skipped
# __iter__ and __reversed__ have a default implementation
# based on __len__ and __getitem__, so they are skipped.
# [Numeric Types]
def __add__(self, other: object):
return self.__getattr__("__add__")
def __sub__(self, other: object):
return self.__getattr__("__sub__")
def __mul__(self, other: object):
return self.__getattr__("__mul__")
def __matmul__(self, other: object):
return self.__getattr__("__matmul__")
def __truediv__(self, other: object):
return self.__getattr__("__truediv__")
def __floordiv__(self, other: object):
return self.__getattr__("__floordiv__")
def __mod__(self, other: object):
return self.__getattr__("__mod__")
def __divmod__(self, other: object):
return self.__getattr__("__divmod__")
def __pow__(self, other: object, modulo: object = ...):
return self.__getattr__("__pow__")
def __lshift__(self, other: object):
return self.__getattr__("__lshift__")
def __rshift__(self, other: object):
return self.__getattr__("__rshift__")
def __and__(self, other: object):
return self.__getattr__("__and__")
def __xor__(self, other: object):
return self.__getattr__("__xor__")
def __or__(self, other: object):
return self.__getattr__("__or__")
# r* and i* methods have lower priority than
# the methods for left operand so they are skipped
def __neg__(self):
return self.__getattr__("__neg__")
def __pos__(self):
return self.__getattr__("__pos__")
def __abs__(self):
return self.__getattr__("__abs__")
def __invert__(self):
return self.__getattr__("__invert__")
# __complex__, __int__ and __float__ have a default implementation
# based on __index__, so they are skipped.
def __index__(self):
return self.__getattr__("__index__")
def __round__(self, ndigits: object = ...):
return self.__getattr__("__round__")
def __trunc__(self):
return self.__getattr__("__trunc__")
def __floor__(self):
return self.__getattr__("__floor__")
def __ceil__(self):
return self.__getattr__("__ceil__")
# [Context managers]
def __enter__(self):
return self.__getattr__("__enter__")
def __exit__(self, *args: object, **kwargs: object):
return self.__getattr__("__exit__")
class PlaceholderModule(_PlaceholderBase):
"""
A placeholder object to use when a module does not exist.
This enables more informative errors when trying to access attributes
of a module that does not exist.
"""
def __init__(self, name: str) -> None:
super().__init__()
# Apply name mangling to avoid conflicting with module attributes
self.__name = name
def placeholder_attr(self, attr_path: str):
return _PlaceholderModuleAttr(self, attr_path)
def __getattr__(self, key: str):
name = self.__name
try:
importlib.import_module(name)
except ImportError as exc:
for extra, names in get_vllm_optional_dependencies().items():
if name in names:
msg = f"Please install vllm[{extra}] for {extra} support"
raise ImportError(msg) from exc
raise exc
raise AssertionError(
"PlaceholderModule should not be used "
"when the original module can be imported"
)
class _PlaceholderModuleAttr(_PlaceholderBase):
def __init__(self, module: PlaceholderModule, attr_path: str) -> None:
super().__init__()
# Apply name mangling to avoid conflicting with module attributes
self.__module = module
self.__attr_path = attr_path
def placeholder_attr(self, attr_path: str):
return _PlaceholderModuleAttr(self.__module, f"{self.__attr_path}.{attr_path}")
def __getattr__(self, key: str):
getattr(self.__module, f"{self.__attr_path}.{key}")
raise AssertionError(
"PlaceholderModule should not be used "
"when the original module can be imported"
)
# create a library to hold the custom op
vllm_lib = Library("vllm", "FRAGMENT") # noqa
......@@ -1871,15 +1618,6 @@ def direct_register_custom_op(
my_lib._register_fake(op_name, fake_impl)
def resolve_obj_by_qualname(qualname: str) -> Any:
"""
Resolve an object by its fully-qualified class name.
"""
module_name, obj_name = qualname.rsplit(".", 1)
module = importlib.import_module(module_name)
return getattr(module, obj_name)
def kill_process_tree(pid: int):
"""
Kills all descendant processes of the given pid by sending SIGKILL.
......@@ -2427,57 +2165,6 @@ def warn_for_unimplemented_methods(cls: type[T]) -> type[T]:
return cls
class LazyLoader(types.ModuleType):
"""
LazyLoader module borrowed from Tensorflow
https://github.com/tensorflow/tensorflow/blob/main/tensorflow/python/util/lazy_loader.py
with an addition of "module caching".
Lazily import a module, mainly to avoid pulling in large dependencies.
Modules such as `xgrammar` might do additional side effects, so we
only want to use this when it is needed, delaying all eager effects
"""
def __init__(
self,
local_name: str,
parent_module_globals: dict[str, Any],
name: str,
):
self._local_name = local_name
self._parent_module_globals = parent_module_globals
self._module: types.ModuleType | None = None
super().__init__(str(name))
def _load(self) -> types.ModuleType:
# Import the target module and insert it into the parent's namespace
try:
module = importlib.import_module(self.__name__)
self._parent_module_globals[self._local_name] = module
# The additional add to sys.modules
# ensures library is actually loaded.
sys.modules[self._local_name] = module
except ModuleNotFoundError as err:
raise err from None
# Update this object's dict so that if someone keeps a
# reference to the LazyLoader, lookups are efficient
# (__getattr__ is only called on lookups that fail).
self.__dict__.update(module.__dict__)
return module
def __getattr__(self, item: Any) -> Any:
if self._module is None:
self._module = self._load()
return getattr(self._module, item)
def __dir__(self) -> list[str]:
if self._module is None:
self._module = self._load()
return dir(self._module)
@contextlib.contextmanager
def cprofile_context(save_file: str | None = None):
"""Run a cprofile
......
vllm/utils/import_utils.py 0 → 100644
View file @ 4d4d6bad
# SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
"""
Contains helpers related to importing modules.
This is similar in concept to the `importlib` module.
"""
import importlib.metadata
import importlib.util
import os
import sys
from functools import cache
from types import ModuleType
from typing import Any
import regex as re
from typing_extensions import Never
def import_from_path(module_name: str, file_path: str | os.PathLike):
"""
Import a Python file according to its file path.
Based on the official recipe:
https://docs.python.org/3/library/importlib.html#importing-a-source-file-directly
"""
spec = importlib.util.spec_from_file_location(module_name, file_path)
if spec is None:
raise ModuleNotFoundError(f"No module named {module_name!r}")
assert spec.loader is not None
module = importlib.util.module_from_spec(spec)
sys.modules[module_name] = module
spec.loader.exec_module(module)
return module
def resolve_obj_by_qualname(qualname: str) -> Any:
"""
Resolve an object by its fully-qualified class name.
"""
module_name, obj_name = qualname.rsplit(".", 1)
module = importlib.import_module(module_name)
return getattr(module, obj_name)
@cache
def get_vllm_optional_dependencies():
metadata = importlib.metadata.metadata("vllm")
requirements = metadata.get_all("Requires-Dist", [])
extras = metadata.get_all("Provides-Extra", [])
return {
extra: [
re.split(r";|>=|<=|==", req)[0]
for req in requirements
if req.endswith(f'extra == "{extra}"')
]
for extra in extras
}
class _PlaceholderBase:
"""
Disallows downstream usage of placeholder modules.
We need to explicitly override each dunder method because
[`__getattr__`][vllm.utils.import_utils._PlaceholderBase.__getattr__]
is not called when they are accessed.
Info:
[Special method lookup](https://docs.python.org/3/reference/datamodel.html#special-lookup)
"""
def __getattr__(self, key: str) -> Never:
"""
The main class should implement this to throw an error
for attribute accesses representing downstream usage.
"""
raise NotImplementedError
# [Basic customization]
def __lt__(self, other: object):
return self.__getattr__("__lt__")
def __le__(self, other: object):
return self.__getattr__("__le__")
def __eq__(self, other: object):
return self.__getattr__("__eq__")
def __ne__(self, other: object):
return self.__getattr__("__ne__")
def __gt__(self, other: object):
return self.__getattr__("__gt__")
def __ge__(self, other: object):
return self.__getattr__("__ge__")
def __hash__(self):
return self.__getattr__("__hash__")
def __bool__(self):
return self.__getattr__("__bool__")
# [Callable objects]
def __call__(self, *args: object, **kwargs: object):
return self.__getattr__("__call__")
# [Container types]
def __len__(self):
return self.__getattr__("__len__")
def __getitem__(self, key: object):
return self.__getattr__("__getitem__")
def __setitem__(self, key: object, value: object):
return self.__getattr__("__setitem__")
def __delitem__(self, key: object):
return self.__getattr__("__delitem__")
# __missing__ is optional according to __getitem__ specification,
# so it is skipped
# __iter__ and __reversed__ have a default implementation
# based on __len__ and __getitem__, so they are skipped.
# [Numeric Types]
def __add__(self, other: object):
return self.__getattr__("__add__")
def __sub__(self, other: object):
return self.__getattr__("__sub__")
def __mul__(self, other: object):
return self.__getattr__("__mul__")
def __matmul__(self, other: object):
return self.__getattr__("__matmul__")
def __truediv__(self, other: object):
return self.__getattr__("__truediv__")
def __floordiv__(self, other: object):
return self.__getattr__("__floordiv__")
def __mod__(self, other: object):
return self.__getattr__("__mod__")
def __divmod__(self, other: object):
return self.__getattr__("__divmod__")
def __pow__(self, other: object, modulo: object = ...):
return self.__getattr__("__pow__")
def __lshift__(self, other: object):
return self.__getattr__("__lshift__")
def __rshift__(self, other: object):
return self.__getattr__("__rshift__")
def __and__(self, other: object):
return self.__getattr__("__and__")
def __xor__(self, other: object):
return self.__getattr__("__xor__")
def __or__(self, other: object):
return self.__getattr__("__or__")
# r* and i* methods have lower priority than
# the methods for left operand so they are skipped
def __neg__(self):
return self.__getattr__("__neg__")
def __pos__(self):
return self.__getattr__("__pos__")
def __abs__(self):
return self.__getattr__("__abs__")
def __invert__(self):
return self.__getattr__("__invert__")
# __complex__, __int__ and __float__ have a default implementation
# based on __index__, so they are skipped.
def __index__(self):
return self.__getattr__("__index__")
def __round__(self, ndigits: object = ...):
return self.__getattr__("__round__")
def __trunc__(self):
return self.__getattr__("__trunc__")
def __floor__(self):
return self.__getattr__("__floor__")
def __ceil__(self):
return self.__getattr__("__ceil__")
# [Context managers]
def __enter__(self):
return self.__getattr__("__enter__")
def __exit__(self, *args: object, **kwargs: object):
return self.__getattr__("__exit__")
class PlaceholderModule(_PlaceholderBase):
"""
A placeholder object to use when a module does not exist.
This enables more informative errors when trying to access attributes
of a module that does not exist.
"""
def __init__(self, name: str) -> None:
super().__init__()
# Apply name mangling to avoid conflicting with module attributes
self.__name = name
def placeholder_attr(self, attr_path: str):
return _PlaceholderModuleAttr(self, attr_path)
def __getattr__(self, key: str) -> Never:
name = self.__name
try:
importlib.import_module(name)
except ImportError as exc:
for extra, names in get_vllm_optional_dependencies().items():
if name in names:
msg = f"Please install vllm[{extra}] for {extra} support"
raise ImportError(msg) from exc
raise exc
raise AssertionError(
"PlaceholderModule should not be used "
"when the original module can be imported"
)
class _PlaceholderModuleAttr(_PlaceholderBase):
def __init__(self, module: PlaceholderModule, attr_path: str) -> None:
super().__init__()
# Apply name mangling to avoid conflicting with module attributes
self.__module = module
self.__attr_path = attr_path
def placeholder_attr(self, attr_path: str):
return _PlaceholderModuleAttr(self.__module, f"{self.__attr_path}.{attr_path}")
def __getattr__(self, key: str) -> Never:
getattr(self.__module, f"{self.__attr_path}.{key}")
raise AssertionError(
"PlaceholderModule should not be used "
"when the original module can be imported"
)
class LazyLoader(ModuleType):
"""
`LazyLoader` module borrowed from [Tensorflow]
(https://github.com/tensorflow/tensorflow/blob/main/tensorflow/python/util/lazy_loader.py)
with an addition of "module caching".
Lazily import a module, mainly to avoid pulling in large dependencies.
Modules such as `xgrammar` might do additional side effects, so we
only want to use this when it is needed, delaying all eager effects.
"""
def __init__(
self,
local_name: str,
parent_module_globals: dict[str, Any],
name: str,
):
self._local_name = local_name
self._parent_module_globals = parent_module_globals
self._module: ModuleType | None = None
super().__init__(str(name))
def _load(self) -> ModuleType:
# Import the target module and insert it into the parent's namespace
try:
module = importlib.import_module(self.__name__)
self._parent_module_globals[self._local_name] = module
# The additional add to sys.modules
# ensures library is actually loaded.
sys.modules[self._local_name] = module
except ModuleNotFoundError as err:
raise err from None
# Update this object's dict so that if someone keeps a
# reference to the LazyLoader, lookups are efficient
# (__getattr__ is only called on lookups that fail).
self.__dict__.update(module.__dict__)
return module
def __getattr__(self, item: Any) -> Any:
if self._module is None:
self._module = self._load()
return getattr(self._module, item)
def __dir__(self) -> list[str]:
if self._module is None:
self._module = self._load()
return dir(self._module)
vllm/v1/engine/core.py
View file @ 4d4d6bad
......@@ -32,10 +32,10 @@ from vllm.utils import (
decorate_logs,
get_hash_fn_by_name,
make_zmq_socket,
resolve_obj_by_qualname,
set_process_title,
)
from vllm.utils.gc_utils import maybe_attach_gc_debug_callback
from vllm.utils.import_utils import resolve_obj_by_qualname
from vllm.v1.core.kv_cache_utils import (
BlockHash,
generate_scheduler_kv_cache_config,
......
vllm/v1/executor/abstract.py
View file @ 4d4d6bad
......@@ -14,7 +14,7 @@ from vllm.executor.uniproc_executor import ( # noqa
ExecutorWithExternalLauncher as ExecutorWithExternalLauncherV0,
)
from vllm.executor.uniproc_executor import UniProcExecutor as UniProcExecutorV0 # noqa
from vllm.utils import resolve_obj_by_qualname
from vllm.utils.import_utils import resolve_obj_by_qualname
from vllm.v1.core.sched.output import SchedulerOutput
from vllm.v1.kv_cache_interface import KVCacheConfig, KVCacheSpec
from vllm.v1.outputs import DraftTokenIds, ModelRunnerOutput
......
vllm/v1/structured_output/__init__.py
View file @ 4d4d6bad
......@@ -8,7 +8,7 @@ from vllm.config import VllmConfig
from vllm.logger import init_logger
from vllm.reasoning import ReasoningParserManager
from vllm.transformers_utils.tokenizer import init_tokenizer_from_configs
from vllm.utils import LazyLoader
from vllm.utils.import_utils import LazyLoader
from vllm.v1.structured_output.backend_guidance import GuidanceBackend
from vllm.v1.structured_output.backend_types import (
StructuredOutputBackend,
......
vllm/v1/structured_output/backend_guidance.py
View file @ 4d4d6bad
......@@ -11,7 +11,7 @@ import torch
from vllm.logger import init_logger
from vllm.sampling_params import SamplingParams
from vllm.utils import LazyLoader
from vllm.utils.import_utils import LazyLoader
from vllm.v1.structured_output.backend_types import (
StructuredOutputBackend,
StructuredOutputGrammar,
......
vllm/v1/structured_output/backend_lm_format_enforcer.py
View file @ 4d4d6bad
......@@ -10,7 +10,7 @@ import torch
from transformers import PreTrainedTokenizerBase
from vllm.sampling_params import SamplingParams
from vllm.utils import LazyLoader
from vllm.utils.import_utils import LazyLoader
from vllm.v1.structured_output.backend_types import (
StructuredOutputBackend,
StructuredOutputGrammar,
......
vllm/v1/structured_output/backend_outlines.py
View file @ 4d4d6bad
......@@ -12,7 +12,7 @@ import torch
from regex import escape as regex_escape
from vllm.sampling_params import SamplingParams
from vllm.utils import LazyLoader
from vllm.utils.import_utils import LazyLoader
from vllm.v1.structured_output.backend_types import (
StructuredOutputBackend,
StructuredOutputGrammar,
......
vllm/v1/structured_output/backend_xgrammar.py
View file @ 4d4d6bad
......@@ -11,7 +11,7 @@ import vllm.envs
from vllm.logger import init_logger
from vllm.sampling_params import SamplingParams
from vllm.transformers_utils.tokenizers.mistral import MistralTokenizer
from vllm.utils import LazyLoader
from vllm.utils.import_utils import LazyLoader
from vllm.v1.structured_output.backend_types import (
StructuredOutputBackend,
StructuredOutputGrammar,
......
vllm/v1/structured_output/utils.py
View file @ 4d4d6bad
......@@ -13,7 +13,7 @@ from diskcache import Cache
import vllm.envs as envs
from vllm.logger import init_logger
from vllm.utils import LazyLoader
from vllm.utils.import_utils import LazyLoader
if TYPE_CHECKING:
import outlines_core as oc
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment