Package inference

Interface ModelConfigOuterClass.ModelSequenceBatching.StateOrBuilder

All Superinterfaces:
com.google.protobuf.MessageLiteOrBuilder, com.google.protobuf.MessageOrBuilder
All Known Implementing Classes:
ModelConfigOuterClass.ModelSequenceBatching.State, ModelConfigOuterClass.ModelSequenceBatching.State.Builder
Enclosing class:
ModelConfigOuterClass.ModelSequenceBatching

public static interface ModelConfigOuterClass.ModelSequenceBatching.StateOrBuilder extends com.google.protobuf.MessageOrBuilder
  • Method Details

    • getInputName

      String getInputName()
      @@    .. cpp:var:: string input_name
      @@
      @@       The name of the model state input.
      @@
       
      string input_name = 1;
      Returns:
      The inputName.
    • getInputNameBytes

      com.google.protobuf.ByteString getInputNameBytes()
      @@    .. cpp:var:: string input_name
      @@
      @@       The name of the model state input.
      @@
       
      string input_name = 1;
      Returns:
      The bytes for inputName.
    • getOutputName

      String getOutputName()
      @@    .. cpp:var:: string output_name
      @@
      @@       The name of the model state output.
      @@
       
      string output_name = 2;
      Returns:
      The outputName.
    • getOutputNameBytes

      com.google.protobuf.ByteString getOutputNameBytes()
      @@    .. cpp:var:: string output_name
      @@
      @@       The name of the model state output.
      @@
       
      string output_name = 2;
      Returns:
      The bytes for outputName.
    • getDataTypeValue

      int getDataTypeValue()
      @@    .. cpp:var:: DataType data_type
      @@
      @@       The data-type of the state.
      @@
       
      .inference.DataType data_type = 3;
      Returns:
      The enum numeric value on the wire for dataType.
    • getDataType

      @@    .. cpp:var:: DataType data_type
      @@
      @@       The data-type of the state.
      @@
       
      .inference.DataType data_type = 3;
      Returns:
      The dataType.
    • getDimsList

      List<Long> getDimsList()
      @@    .. cpp:var:: int64 dim (repeated)
      @@
      @@       The dimension.
      @@
       
      repeated int64 dims = 4;
      Returns:
      A list containing the dims.
    • getDimsCount

      int getDimsCount()
      @@    .. cpp:var:: int64 dim (repeated)
      @@
      @@       The dimension.
      @@
       
      repeated int64 dims = 4;
      Returns:
      The count of dims.
    • getDims

      long getDims(int index)
      @@    .. cpp:var:: int64 dim (repeated)
      @@
      @@       The dimension.
      @@
       
      repeated int64 dims = 4;
      Parameters:
      index - The index of the element to return.
      Returns:
      The dims at the given index.
    • getInitialStateList

      @@  .. cpp:var:: InitialState initial_state (repeated)
      @@
      @@     The optional field to specify the initial state for the model.
      @@
       
      repeated .inference.ModelSequenceBatching.InitialState initial_state = 5;
    • getInitialState

      @@  .. cpp:var:: InitialState initial_state (repeated)
      @@
      @@     The optional field to specify the initial state for the model.
      @@
       
      repeated .inference.ModelSequenceBatching.InitialState initial_state = 5;
    • getInitialStateCount

      int getInitialStateCount()
      @@  .. cpp:var:: InitialState initial_state (repeated)
      @@
      @@     The optional field to specify the initial state for the model.
      @@
       
      repeated .inference.ModelSequenceBatching.InitialState initial_state = 5;
    • getInitialStateOrBuilderList

      @@  .. cpp:var:: InitialState initial_state (repeated)
      @@
      @@     The optional field to specify the initial state for the model.
      @@
       
      repeated .inference.ModelSequenceBatching.InitialState initial_state = 5;
    • getInitialStateOrBuilder

      @@  .. cpp:var:: InitialState initial_state (repeated)
      @@
      @@     The optional field to specify the initial state for the model.
      @@
       
      repeated .inference.ModelSequenceBatching.InitialState initial_state = 5;
    • getUseSameBufferForInputOutput

      boolean getUseSameBufferForInputOutput()
      @@  .. cpp:var:: bool use_same_buffer_for_input_output
      @@
      @@     The optional field to use a single buffer for both input and output
      @@     state. Without this option, Triton allocates separate buffers
      @@     for input and output state
      @@     which can be problematic if the state size is
      @@     large. This option reduces the memory usage by allocating a single
      @@     buffer. Enabling this option is recommended whenever
      @@     the input state is processed before the output state is written.
      @@     When enabled the state
      @@     will always be updated independent of whether
      @@     TRITONBACKEND_StateUpdate is called
      @@     (however TRITONBACKEND_StateUpdate should still be called for
      @@     completeness).
      @@
      @@     The default value is false.
      @@
       
      bool use_same_buffer_for_input_output = 6;
      Returns:
      The useSameBufferForInputOutput.
    • getUseGrowableMemory

      boolean getUseGrowableMemory()
      @@  .. cpp:var:: bool use_growable_memory
      @@
      @@     The optional field to enable an implicit state buffer to grow
      @@     without reallocating or copying existing memory.
      @@     Additional memory will be appended to the end of the buffer and
      @@     existing data will be preserved.
      @@     This option is only available for CUDA memory and requires enabling
      @@     use_same_buffer_for_input_output. When using this option,
      @@     StateBuffer call will always return CUDA memory even if CPU memory
      @@     is requested.
      @@
      @@     The default value is false.
      @@
       
      bool use_growable_memory = 7;
      Returns:
      The useGrowableMemory.