337 lines
10 KiB
Python
337 lines
10 KiB
Python
# type: ignore
|
|
from __future__ import annotations
|
|
import collections.abc
|
|
import openvino._pyopenvino
|
|
import typing
|
|
"""
|
|
openvino.properties.hint submodule that simulates ov::hint
|
|
"""
|
|
__all__ = ['ExecutionMode', 'ModelDistributionPolicy', 'PerformanceMode', 'Priority', 'SchedulingCoreType', 'activations_scale_factor', 'allow_auto_batching', 'compiled_blob', 'dynamic_quantization_group_size', 'enable_cpu_pinning', 'enable_hyper_threading', 'execution_mode', 'inference_precision', 'kv_cache_precision', 'model', 'model_distribution_policy', 'model_priority', 'num_requests', 'performance_mode', 'scheduling_core_type']
|
|
class ExecutionMode:
|
|
"""
|
|
Members:
|
|
|
|
PERFORMANCE
|
|
|
|
ACCURACY
|
|
"""
|
|
ACCURACY: typing.ClassVar[ExecutionMode] # value = <ExecutionMode.ACCURACY: 2>
|
|
PERFORMANCE: typing.ClassVar[ExecutionMode] # value = <ExecutionMode.PERFORMANCE: 1>
|
|
__members__: typing.ClassVar[dict[str, ExecutionMode]] # value = {'PERFORMANCE': <ExecutionMode.PERFORMANCE: 1>, 'ACCURACY': <ExecutionMode.ACCURACY: 2>}
|
|
def __eq__(self, other: typing.Any) -> bool:
|
|
...
|
|
def __ge__(self, other: typing.Any) -> bool:
|
|
...
|
|
def __getstate__(self) -> int:
|
|
...
|
|
def __gt__(self, other: typing.Any) -> bool:
|
|
...
|
|
def __hash__(self) -> int:
|
|
...
|
|
def __index__(self) -> int:
|
|
...
|
|
def __init__(self, value: typing.SupportsInt) -> None:
|
|
...
|
|
def __int__(self) -> int:
|
|
...
|
|
def __le__(self, other: typing.Any) -> bool:
|
|
...
|
|
def __lt__(self, other: typing.Any) -> bool:
|
|
...
|
|
def __ne__(self, other: typing.Any) -> bool:
|
|
...
|
|
def __repr__(self) -> str:
|
|
...
|
|
def __setstate__(self, state: typing.SupportsInt) -> None:
|
|
...
|
|
def __str__(self) -> str:
|
|
...
|
|
@property
|
|
def name(self) -> str:
|
|
...
|
|
@property
|
|
def value(self) -> int:
|
|
...
|
|
class ModelDistributionPolicy:
|
|
"""
|
|
Members:
|
|
|
|
TENSOR_PARALLEL
|
|
|
|
PIPELINE_PARALLEL
|
|
"""
|
|
PIPELINE_PARALLEL: typing.ClassVar[ModelDistributionPolicy] # value = <ModelDistributionPolicy.PIPELINE_PARALLEL: 1>
|
|
TENSOR_PARALLEL: typing.ClassVar[ModelDistributionPolicy] # value = <ModelDistributionPolicy.TENSOR_PARALLEL: 0>
|
|
__members__: typing.ClassVar[dict[str, ModelDistributionPolicy]] # value = {'TENSOR_PARALLEL': <ModelDistributionPolicy.TENSOR_PARALLEL: 0>, 'PIPELINE_PARALLEL': <ModelDistributionPolicy.PIPELINE_PARALLEL: 1>}
|
|
def __eq__(self, other: typing.Any) -> bool:
|
|
...
|
|
def __ge__(self, other: typing.Any) -> bool:
|
|
...
|
|
def __getstate__(self) -> int:
|
|
...
|
|
def __gt__(self, other: typing.Any) -> bool:
|
|
...
|
|
def __hash__(self) -> int:
|
|
...
|
|
def __index__(self) -> int:
|
|
...
|
|
def __init__(self, value: typing.SupportsInt) -> None:
|
|
...
|
|
def __int__(self) -> int:
|
|
...
|
|
def __le__(self, other: typing.Any) -> bool:
|
|
...
|
|
def __lt__(self, other: typing.Any) -> bool:
|
|
...
|
|
def __ne__(self, other: typing.Any) -> bool:
|
|
...
|
|
def __repr__(self) -> str:
|
|
...
|
|
def __setstate__(self, state: typing.SupportsInt) -> None:
|
|
...
|
|
def __str__(self) -> str:
|
|
...
|
|
@property
|
|
def name(self) -> str:
|
|
...
|
|
@property
|
|
def value(self) -> int:
|
|
...
|
|
class PerformanceMode:
|
|
"""
|
|
Members:
|
|
|
|
LATENCY
|
|
|
|
THROUGHPUT
|
|
|
|
CUMULATIVE_THROUGHPUT
|
|
"""
|
|
CUMULATIVE_THROUGHPUT: typing.ClassVar[PerformanceMode] # value = <PerformanceMode.CUMULATIVE_THROUGHPUT: 3>
|
|
LATENCY: typing.ClassVar[PerformanceMode] # value = <PerformanceMode.LATENCY: 1>
|
|
THROUGHPUT: typing.ClassVar[PerformanceMode] # value = <PerformanceMode.THROUGHPUT: 2>
|
|
__members__: typing.ClassVar[dict[str, PerformanceMode]] # value = {'LATENCY': <PerformanceMode.LATENCY: 1>, 'THROUGHPUT': <PerformanceMode.THROUGHPUT: 2>, 'CUMULATIVE_THROUGHPUT': <PerformanceMode.CUMULATIVE_THROUGHPUT: 3>}
|
|
def __eq__(self, other: typing.Any) -> bool:
|
|
...
|
|
def __ge__(self, other: typing.Any) -> bool:
|
|
...
|
|
def __getstate__(self) -> int:
|
|
...
|
|
def __gt__(self, other: typing.Any) -> bool:
|
|
...
|
|
def __hash__(self) -> int:
|
|
...
|
|
def __index__(self) -> int:
|
|
...
|
|
def __init__(self, value: typing.SupportsInt) -> None:
|
|
...
|
|
def __int__(self) -> int:
|
|
...
|
|
def __le__(self, other: typing.Any) -> bool:
|
|
...
|
|
def __lt__(self, other: typing.Any) -> bool:
|
|
...
|
|
def __ne__(self, other: typing.Any) -> bool:
|
|
...
|
|
def __repr__(self) -> str:
|
|
...
|
|
def __setstate__(self, state: typing.SupportsInt) -> None:
|
|
...
|
|
def __str__(self) -> str:
|
|
...
|
|
@property
|
|
def name(self) -> str:
|
|
...
|
|
@property
|
|
def value(self) -> int:
|
|
...
|
|
class Priority:
|
|
"""
|
|
Members:
|
|
|
|
LOW
|
|
|
|
MEDIUM
|
|
|
|
HIGH
|
|
|
|
DEFAULT
|
|
"""
|
|
DEFAULT: typing.ClassVar[Priority] # value = <Priority.MEDIUM: 1>
|
|
HIGH: typing.ClassVar[Priority] # value = <Priority.HIGH: 2>
|
|
LOW: typing.ClassVar[Priority] # value = <Priority.LOW: 0>
|
|
MEDIUM: typing.ClassVar[Priority] # value = <Priority.MEDIUM: 1>
|
|
__members__: typing.ClassVar[dict[str, Priority]] # value = {'LOW': <Priority.LOW: 0>, 'MEDIUM': <Priority.MEDIUM: 1>, 'HIGH': <Priority.HIGH: 2>, 'DEFAULT': <Priority.MEDIUM: 1>}
|
|
def __eq__(self, other: typing.Any) -> bool:
|
|
...
|
|
def __ge__(self, other: typing.Any) -> bool:
|
|
...
|
|
def __getstate__(self) -> int:
|
|
...
|
|
def __gt__(self, other: typing.Any) -> bool:
|
|
...
|
|
def __hash__(self) -> int:
|
|
...
|
|
def __index__(self) -> int:
|
|
...
|
|
def __init__(self, value: typing.SupportsInt) -> None:
|
|
...
|
|
def __int__(self) -> int:
|
|
...
|
|
def __le__(self, other: typing.Any) -> bool:
|
|
...
|
|
def __lt__(self, other: typing.Any) -> bool:
|
|
...
|
|
def __ne__(self, other: typing.Any) -> bool:
|
|
...
|
|
def __repr__(self) -> str:
|
|
...
|
|
def __setstate__(self, state: typing.SupportsInt) -> None:
|
|
...
|
|
def __str__(self) -> str:
|
|
...
|
|
@property
|
|
def name(self) -> str:
|
|
...
|
|
@property
|
|
def value(self) -> int:
|
|
...
|
|
class SchedulingCoreType:
|
|
"""
|
|
Members:
|
|
|
|
ANY_CORE
|
|
|
|
PCORE_ONLY
|
|
|
|
ECORE_ONLY
|
|
"""
|
|
ANY_CORE: typing.ClassVar[SchedulingCoreType] # value = <SchedulingCoreType.ANY_CORE: 0>
|
|
ECORE_ONLY: typing.ClassVar[SchedulingCoreType] # value = <SchedulingCoreType.ECORE_ONLY: 2>
|
|
PCORE_ONLY: typing.ClassVar[SchedulingCoreType] # value = <SchedulingCoreType.PCORE_ONLY: 1>
|
|
__members__: typing.ClassVar[dict[str, SchedulingCoreType]] # value = {'ANY_CORE': <SchedulingCoreType.ANY_CORE: 0>, 'PCORE_ONLY': <SchedulingCoreType.PCORE_ONLY: 1>, 'ECORE_ONLY': <SchedulingCoreType.ECORE_ONLY: 2>}
|
|
def __eq__(self, other: typing.Any) -> bool:
|
|
...
|
|
def __ge__(self, other: typing.Any) -> bool:
|
|
...
|
|
def __getstate__(self) -> int:
|
|
...
|
|
def __gt__(self, other: typing.Any) -> bool:
|
|
...
|
|
def __hash__(self) -> int:
|
|
...
|
|
def __index__(self) -> int:
|
|
...
|
|
def __init__(self, value: typing.SupportsInt) -> None:
|
|
...
|
|
def __int__(self) -> int:
|
|
...
|
|
def __le__(self, other: typing.Any) -> bool:
|
|
...
|
|
def __lt__(self, other: typing.Any) -> bool:
|
|
...
|
|
def __ne__(self, other: typing.Any) -> bool:
|
|
...
|
|
def __repr__(self) -> str:
|
|
...
|
|
def __setstate__(self, state: typing.SupportsInt) -> None:
|
|
...
|
|
def __str__(self) -> str:
|
|
...
|
|
@property
|
|
def name(self) -> str:
|
|
...
|
|
@property
|
|
def value(self) -> int:
|
|
...
|
|
@typing.overload
|
|
def activations_scale_factor() -> str:
|
|
...
|
|
@typing.overload
|
|
def activations_scale_factor(arg0: typing.SupportsFloat) -> tuple[str, openvino._pyopenvino.OVAny]:
|
|
...
|
|
@typing.overload
|
|
def allow_auto_batching() -> str:
|
|
...
|
|
@typing.overload
|
|
def allow_auto_batching(arg0: bool) -> tuple[str, openvino._pyopenvino.OVAny]:
|
|
...
|
|
@typing.overload
|
|
def compiled_blob() -> str:
|
|
...
|
|
@typing.overload
|
|
def compiled_blob(arg0: openvino._pyopenvino.Tensor) -> tuple[str, openvino._pyopenvino.OVAny]:
|
|
...
|
|
@typing.overload
|
|
def dynamic_quantization_group_size() -> str:
|
|
...
|
|
@typing.overload
|
|
def dynamic_quantization_group_size(arg0: typing.SupportsInt) -> tuple[str, openvino._pyopenvino.OVAny]:
|
|
...
|
|
@typing.overload
|
|
def enable_cpu_pinning() -> str:
|
|
...
|
|
@typing.overload
|
|
def enable_cpu_pinning(arg0: bool) -> tuple[str, openvino._pyopenvino.OVAny]:
|
|
...
|
|
@typing.overload
|
|
def enable_hyper_threading() -> str:
|
|
...
|
|
@typing.overload
|
|
def enable_hyper_threading(arg0: bool) -> tuple[str, openvino._pyopenvino.OVAny]:
|
|
...
|
|
@typing.overload
|
|
def execution_mode() -> str:
|
|
...
|
|
@typing.overload
|
|
def execution_mode(arg0: ExecutionMode) -> tuple[str, openvino._pyopenvino.OVAny]:
|
|
...
|
|
@typing.overload
|
|
def inference_precision() -> str:
|
|
...
|
|
@typing.overload
|
|
def inference_precision(arg0: openvino._pyopenvino.Type) -> tuple[str, openvino._pyopenvino.OVAny]:
|
|
...
|
|
@typing.overload
|
|
def kv_cache_precision() -> str:
|
|
...
|
|
@typing.overload
|
|
def kv_cache_precision(arg0: openvino._pyopenvino.Type) -> tuple[str, openvino._pyopenvino.OVAny]:
|
|
...
|
|
@typing.overload
|
|
def model() -> str:
|
|
...
|
|
@typing.overload
|
|
def model(arg0: openvino._pyopenvino.Model) -> tuple[str, openvino._pyopenvino.OVAny]:
|
|
...
|
|
@typing.overload
|
|
def model_distribution_policy() -> str:
|
|
...
|
|
@typing.overload
|
|
def model_distribution_policy(arg0: collections.abc.Set[ModelDistributionPolicy]) -> tuple[str, openvino._pyopenvino.OVAny]:
|
|
...
|
|
@typing.overload
|
|
def model_priority() -> str:
|
|
...
|
|
@typing.overload
|
|
def model_priority(arg0: Priority) -> tuple[str, openvino._pyopenvino.OVAny]:
|
|
...
|
|
@typing.overload
|
|
def num_requests() -> str:
|
|
...
|
|
@typing.overload
|
|
def num_requests(arg0: typing.SupportsInt) -> tuple[str, openvino._pyopenvino.OVAny]:
|
|
...
|
|
@typing.overload
|
|
def performance_mode() -> str:
|
|
...
|
|
@typing.overload
|
|
def performance_mode(arg0: PerformanceMode) -> tuple[str, openvino._pyopenvino.OVAny]:
|
|
...
|
|
@typing.overload
|
|
def scheduling_core_type() -> str:
|
|
...
|
|
@typing.overload
|
|
def scheduling_core_type(arg0: SchedulingCoreType) -> tuple[str, openvino._pyopenvino.OVAny]:
|
|
...
|