Inheritance diagram for polygraphy.backend.trt.loader.CreateConfig:

Collaboration diagram for polygraphy.backend.trt.loader.CreateConfig:

Public Member Functions
def	__init__ (self, max_workspace_size=None, tf32=None, fp16=None, int8=None, profiles=None, calibrator=None, strict_types=None)

def	__call__ (self, builder, network)

Public Attributes
	max_workspace_size

	tf32

	fp16

	int8

	profiles

	calibrator

	strict_types

Constructor & Destructor Documentation

◆ init()

def polygraphy.backend.trt.loader.CreateConfig.__init__	(	self,
		max_workspace_size = `None`,
		tf32 = `None`,
		fp16 = `None`,
		int8 = `None`,
		profiles = `None`,
		calibrator = `None`,
		strict_types = `None`
	)

Functor that creates a TensorRT IBuilderConfig.

Args:
    max_workspace_size (int): The maximum workspace size, in bytes, when building the engine.
    tf32 (bool): Whether to build the engine with TF32 precision enabled. Defaults to False.
    fp16 (bool): Whether to build the engine with FP16 precision enabled. Defaults to False.
    int8 (bool): Whether to build the engine with INT8 precision enabled. Defaults to False.
    profiles (List[Profile]):
    A list of optimization profiles to add to the configuration. Only needed for
    networks with dynamic input shapes. If this is omitted for a network with
    dynamic shapes, a default profile is created, where dynamic dimensions are
    replaced with Polygraphy's DEFAULT_SHAPE_VALUE  (defined in util/constants.py).
    See `Profile` for details.
    calibrator (trt.IInt8Calibrator):
    An int8 calibrator. Only required in int8 mode when
    the network does not have explicit precision. For networks with
    dynamic shapes, the last profile provided (or default profile if
    no profiles are provided) is used during calibration.

Member Function Documentation

◆ call()

def polygraphy.backend.trt.loader.CreateConfig.__call__	(	self,
		builder,
		network
	)

Creates a TensorRT IBuilderConfig that can be used by the EngineFromNetwork.

Args:
    builder (trt.Builder):
    The TensorRT builder to use to create the configuration.
    network (trt.INetworkDefinition):
    The TensorRT network for which to create the config. The network is used to
    automatically create a default optimization profile if none are provided.

Returns:
    trt.IBuilderConfig: The TensorRT builder configuration.

Member Data Documentation

◆ max_workspace_size

polygraphy.backend.trt.loader.CreateConfig.max_workspace_size

◆ tf32

polygraphy.backend.trt.loader.CreateConfig.tf32

◆ fp16

polygraphy.backend.trt.loader.CreateConfig.fp16

◆ int8

polygraphy.backend.trt.loader.CreateConfig.int8

◆ profiles

polygraphy.backend.trt.loader.CreateConfig.profiles

◆ calibrator

polygraphy.backend.trt.loader.CreateConfig.calibrator

◆ strict_types

polygraphy.backend.trt.loader.CreateConfig.strict_types

The documentation for this class was generated from the following file:

backend/trt/loader.py

Public Member Functions

Public Attributes