OrtTensorRTProviderOptions class final

\brief TensorRT Provider Options

\see OrtApi::SessionOptionsAppendExecutionProvider_TensorRT

Inheritance
Implemented types
Available extensions

Properties

address → Pointer<T>

Available on T, provided by the StructAddress extension

The memory address of the underlying data.
no setter
device_id ↔ int
< CUDA device id (0 = default device)
getter/setter pair
has_user_compute_stream ↔ int
indicator of user specified CUDA compute stream.
getter/setter pair
hashCode → int
The hash code for this object.
no setterinherited
runtimeType → Type
A representation of the runtime type of the object.
no setterinherited
trt_dla_core ↔ int
DLA core number. Default 0
getter/setter pair
trt_dla_enable ↔ int
enable DLA. Default 0 = false, nonzero = true
getter/setter pair
trt_dump_subgraphs ↔ int
dump TRT subgraph. Default 0 = false, nonzero = true
getter/setter pair
trt_engine_cache_enable ↔ int
enable engine caching. Default 0 = false, nonzero = true
getter/setter pair
trt_engine_cache_path ↔ Pointer<Char>
specify engine cache path
getter/setter pair
trt_engine_decryption_enable ↔ int
enable engine decryption. Default 0 = false, nonzero = true
getter/setter pair
trt_engine_decryption_lib_path ↔ Pointer<Char>
specify engine decryption library path
getter/setter pair
trt_force_sequential_engine_build ↔ int
force building TensorRT engine sequentially. Default 0 = false, nonzero = true
getter/setter pair
trt_fp16_enable ↔ int
enable TensorRT FP16 precision. Default 0 = false, nonzero = true
getter/setter pair
trt_int8_calibration_table_name ↔ Pointer<Char>
TensorRT INT8 calibration table name.
getter/setter pair
trt_int8_enable ↔ int
enable TensorRT INT8 precision. Default 0 = false, nonzero = true
getter/setter pair
trt_int8_use_native_calibration_table ↔ int
use native TensorRT generated calibration table. Default 0 = false, nonzero = true
getter/setter pair
trt_max_partition_iterations ↔ int
maximum iterations for TensorRT parser to get capability
getter/setter pair
trt_max_workspace_size ↔ int
maximum workspace size for TensorRT.
getter/setter pair
trt_min_subgraph_size ↔ int
minimum size of TensorRT subgraphs
getter/setter pair
user_compute_stream ↔ Pointer<Void>
user specified CUDA compute stream.
getter/setter pair

Methods

noSuchMethod(Invocation invocation) → dynamic
Invoked when a nonexistent method or property is accessed.
inherited
toString() → String
A string representation of this object.
inherited

Operators

operator ==(Object other) → bool
The equality operator.
inherited