Struct ModelOptimizationPolicy

pub struct ModelOptimizationPolicy {
    pub graph: Option<Graph>,
    pub priority: i32,
    pub cuda: Option<Cuda>,
    pub execution_accelerators: Option<ExecutionAccelerators>,
    pub input_pinned_memory: Option<PinnedMemoryBuffer>,
    pub output_pinned_memory: Option<PinnedMemoryBuffer>,
    pub gather_kernel_buffer_threshold: u32,
    pub eager_batching: bool,
}

@@ @@.. cpp:var:: message ModelOptimizationPolicy @@ @@ Optimization settings for a model. These settings control if/how a @@ model is optimized and prioritized by the backend framework when @@ it is loaded. @@

Fields

graph: Option<Graph>

@@ .. cpp:var:: Graph graph @@ @@ The graph optimization setting for the model. Optional. @@

priority: i32

@@ .. cpp:var:: ModelPriority priority @@ @@ The priority setting for the model. Optional. @@

cuda: Option<Cuda>

@@ .. cpp:var:: Cuda cuda @@ @@ CUDA-specific optimization settings. Optional. @@

execution_accelerators: Option<ExecutionAccelerators>

@@ .. cpp:var:: ExecutionAccelerators execution_accelerators @@ @@ The accelerators used for the model. Optional. @@

input_pinned_memory: Option<PinnedMemoryBuffer>

@@ .. cpp:var:: PinnedMemoryBuffer input_pinned_memory @@ @@ Use pinned memory buffer when the data transfer for inputs @@ is between GPU memory and non-pinned system memory. @@ Default is true. @@

output_pinned_memory: Option<PinnedMemoryBuffer>

@@ .. cpp:var:: PinnedMemoryBuffer output_pinned_memory @@ @@ Use pinned memory buffer when the data transfer for outputs @@ is between GPU memory and non-pinned system memory. @@ Default is true. @@

gather_kernel_buffer_threshold: u32

@@ .. cpp:var:: uint32 gather_kernel_buffer_threshold @@ @@ The backend may use a gather kernel to gather input data if the @@ device has direct access to the source buffer and the destination @@ buffer. In such case, the gather kernel will be used only if the @@ number of buffers to be gathered is greater or equal to @@ the specifed value. If 0, the gather kernel will be disabled. @@ Default value is 0. @@ Currently only recognized by TensorRT backend. @@

eager_batching: bool

@@ .. cpp:var:: bool eager_batching @@ @@ Start preparing the next batch before the model instance is ready @@ for the next inference. This option can be used to overlap the @@ batch preparation with model execution, with the trade-off that @@ the next batch might be smaller than what it could have been. @@ Default value is false. @@ Currently only recognized by TensorRT backend. @@

Implementations

impl ModelOptimizationPolicy

pub fn priority(&self) -> ModelPriority

Returns the enum value of priority, or the default if the field is set to an invalid enum value.

pub fn set_priority(&mut self, value: ModelPriority)

Sets priority to the provided enum value.

Trait Implementations

impl Clone for ModelOptimizationPolicy

fn clone(&self) -> ModelOptimizationPolicy

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for ModelOptimizationPolicy

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Default for ModelOptimizationPolicy

fn default() -> Self

Returns the “default value” for a type.

impl Message for ModelOptimizationPolicy

fn encoded_len(&self) -> usize

Returns the encoded length of the message without a length delimiter.

fn clear(&mut self)

Clears the message, resetting all fields to their default.

fn encode<B>(&self, buf: &mut B) -> Result<(), EncodeError>

where B: BufMut, Self: Sized,

Encodes the message to a buffer.

fn encode_to_vec(&self) -> Vec<u8>

where Self: Sized,

Encodes the message to a newly allocated buffer.

fn encode_length_delimited<B>(&self, buf: &mut B) -> Result<(), EncodeError>

where B: BufMut, Self: Sized,

Encodes the message with a length-delimiter to a buffer.

fn encode_length_delimited_to_vec(&self) -> Vec<u8>

where Self: Sized,

Encodes the message with a length-delimiter to a newly allocated buffer.

fn decode<B>(buf: B) -> Result<Self, DecodeError>

where B: Buf, Self: Default,

Decodes an instance of the message from a buffer.

fn decode_length_delimited<B>(buf: B) -> Result<Self, DecodeError>

where B: Buf, Self: Default,

Decodes a length-delimited instance of the message from the buffer.

fn merge<B>(&mut self, buf: B) -> Result<(), DecodeError>

where B: Buf, Self: Sized,

Decodes an instance of the message from a buffer, and merges it into self.

fn merge_length_delimited<B>(&mut self, buf: B) -> Result<(), DecodeError>

where B: Buf, Self: Sized,

Decodes a length-delimited instance of the message from buffer, and merges it into self.

impl PartialEq for ModelOptimizationPolicy

fn eq(&self, other: &ModelOptimizationPolicy) -> bool

Tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl StructuralPartialEq for ModelOptimizationPolicy