model_config.proto¶

enum DataType¶

Data types supported for input and output tensors.

enumerator DataType::INVALID = 0¶

enumerator DataType::BOOL = 1¶

enumerator DataType::UINT8 = 2¶

enumerator DataType::UINT16 = 3¶

enumerator DataType::UINT32 = 4¶

enumerator DataType::UINT64 = 5¶

enumerator DataType::INT8 = 6¶

enumerator DataType::INT16 = 7¶

enumerator DataType::INT32 = 8¶

enumerator DataType::INT64 = 9¶

enumerator DataType::FP16 = 10¶

enumerator DataType::FP32 = 11¶

enumerator DataType::FP64 = 12¶

enumerator DataType::STRING = 13¶

message ModelInstanceGroup¶

A group of one or more instances of a model and resources made available for those instances.

enum Kind¶

Kind of this instance group.

enumerator Kind::KIND_AUTO = 0¶: This instance group represents instances that can run on either CPU or GPU. If all GPUs listed in ‘gpus’ are available then instances will be created on GPU(s), otherwise instances will be created on CPU.

enumerator Kind::KIND_GPU = 1¶: This instance group represents instances that must run on the GPU.

enumerator Kind::KIND_CPU = 2¶: This instance group represents instances that must run on the CPU.

enumerator Kind::KIND_MODEL = 3¶: This instance group represents instances that should run on the CPU and/or GPU(s) as specified by the model or backend itself. The inference server will not override the model/backend settings. Currently, this option is supported only for Tensorflow models.

string name¶: Optional name of this group of instances. If not specified the name will be formed as <model name>_<group number>. The name of individual instances will be further formed by a unique instance number and GPU index:

Kind kind¶: The kind of this instance group. Default is KIND_AUTO. If KIND_AUTO or KIND_GPU then both ‘count’ and ‘gpu’ are valid and may be specified. If KIND_CPU or KIND_MODEL only ‘count’ is valid and ‘gpu’ cannot be specified.

int32 count¶: For a group assigned to GPU, the number of instances created for each GPU listed in ‘gpus’. For a group assigned to CPU the number of instances created. Default is 1.

int32 gpus(repeated)¶: GPU(s) where instances should be available. For each GPU listed, ‘count’ instances of the model will be available. Setting ‘gpus’ to empty (or not specifying at all) is eqivalent to listing all available GPUs.

string profile(repeated)¶: For TensorRT models, using inputs with dynamic shape, this parameter specifies a set of optimization profiles available to this instance group. The inference server will choose the optimal profile based on the shapes of the input tensors. This field should lie between 0 and <TotalNumberOfOptimizationProfilesInPlanModel> - 1 and be specified only for TensorRT backend, otherwise an error will be generated.

message ModelTensorReshape¶

Reshape specification for input and output tensors.

int64 shape(repeated)¶: The shape to use for reshaping.

message ModelInput¶

An input required by the model.

enum Format¶

The format for the input.

enumerator Format::FORMAT_NONE = 0¶: The input has no specific format. This is the default.

enumerator Format::FORMAT_NHWC = 1¶: HWC image format. Tensors with this format require 3 dimensions if the model does not support batching (max_batch_size = 0) or 4 dimensions if the model does support batching (max_batch_size >= 1). In either case the ‘dims’ below should only specify the 3 non-batch dimensions (i.e. HWC or CHW).

enumerator Format::FORMAT_NCHW = 2¶: CHW image format. Tensors with this format require 3 dimensions if the model does not support batching (max_batch_size = 0) or 4 dimensions if the model does support batching (max_batch_size >= 1). In either case the ‘dims’ below should only specify the 3 non-batch dimensions (i.e. HWC or CHW).

string name¶: The name of the input.

DataType data_type¶: The data-type of the input.

Format format¶: The format of the input. Optional.

int64 dims(repeated)¶: The dimensions/shape of the input tensor that must be provided when invoking the inference API for this model.

ModelTensorReshape reshape¶: The shape expected for this input by the backend. The input will be reshaped to this before being presented to the backend. The reshape must have the same number of elements as the input shape specified by ‘dims’. Optional.

bool is_shape_tensor¶: Whether or not the input is a shape tensor to the model. This field is currently supported only for the TensorRT model. An error will be generated if this specification does not comply with underlying model.

bool allow_ragged_batch¶: Whether or not the input is allowed to be “ragged” in a dynamically created batch. Default is false indicating that two requests will only be batched if this tensor has the same shape in both requests. True indicates that two requests can be batched even if this tensor has a different shape in each request. A true value is currently supported only for custom models.

message ModelOutput¶

An output produced by the model.

string name¶: The name of the output.

DataType data_type¶: The data-type of the output.

int64 dims(repeated)¶: The dimensions/shape of the output tensor.

ModelTensorReshape reshape¶: The shape produced for this output by the backend. The output will be reshaped from this to the shape specifed in ‘dims’ before being returned in the inference response. The reshape must have the same number of elements as the output shape specified by ‘dims’. Optional.

string label_filename¶: The label file associated with this output. Should be specified only for outputs that represent classifications. Optional.

bool is_shape_tensor¶: Whether or not the output is a shape tensor to the model. This field is currently supported only for the TensorRT model. An error will be generated if this specification does not comply with underlying model.

message ModelVersionPolicy¶

Policy indicating which versions of a model should be made available by the inference server.

message Latest¶

Serve only the latest version(s) of a model. This is the default policy.

uint32 num_versions¶: Serve only the ‘num_versions’ highest-numbered versions. T The default value of ‘num_versions’ is 1, indicating that by default only the single highest-number version of a model will be served.

message All¶: Serve all versions of the model.

message Specific¶

Serve only specific versions of the model.

int64 versions(repeated)¶: The specific versions of the model that will be served.

oneof policy_choice¶

Each model must implement only a single version policy. The default policy is ‘Latest’.

Latest latest¶: Serve only latest version(s) of the model.

All all¶: Serve all versions of the model.

Specific specific¶: Serve only specific version(s) of the model.

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.

message Graph¶

Enable generic graph optimization of the model. If not specified the framework’s default level of optimization is used. Currently only supported for TensorFlow graphdef and savedmodel models and causes XLA to be enabled/disabled for the model.

int32 level¶

The optimization level. Defaults to 0 (zero) if not specified.

-1: Disabled

0: Framework default

1+: Enable optimization level (greater values indicate higher optimization levels)

enum ModelPriority¶

Model priorities. A model will be given scheduling and execution preference over models at lower priorities. Current model priorities only work for TensorRT models.

enumerator ModelPriority::PRIORITY_DEFAULT = 0¶: The default model priority.

enumerator ModelPriority::PRIORITY_MAX = 1¶: The maximum model priority.

enumerator ModelPriority::PRIORITY_MIN = 2¶: The minimum model priority.

message Cuda¶

CUDA-specific optimization settings.

bool graphs¶: Use CUDA graphs API to capture model operations and execute them more efficiently. Currently only recognized by TensorRT backend.

message ExecutionAccelerators¶

Specify the preferred execution accelerators to be used to execute the model. Currently only recognized by ONNX Runtime backend and TensorFlow backend.

For ONNX Runtime backend, it will deploy the model with the execution accelerators by priority, the priority is determined based on the order that they are set, i.e. the provider at the front has highest priority. Overall, the priority will be in the following order:

<gpu_execution_accelerator> (if instance is on GPU) CUDA Execution Provider (if instance is on GPU) <cpu_execution_accelerator> Default CPU Execution Provider

message Accelerator¶

Specify the accelerator to be used to execute the model. Accelerator with the same name may accept different parameters depending on the backends.

string name¶: The name of the execution accelerator.

map<string, string> parameters¶: Additional paremeters used to configure the accelerator.

Accelerator gpu_execution_accelerator(repeated)¶

The preferred execution provider to be used if the model instance is deployed on GPU.

For ONNX Runtime backend, possible value is “tensorrt” as name, and no parameters are required.

For TensorFlow backend, possible values are “tensorrt”, “gpu_io”.

For “tensorrt”, the following parameters can be specified:

“precision_mode”: The precision used for optimization. Allowed values are “FP32” and “FP16”. Default value is “FP32”.

“max_cached_engines”: The maximum number of cached TensorRT engines in dynamic TensorRT ops. Default value is 100.

“minimum_segment_size”: The smallest model subgraph that will be considered for optimization by TensorRT. Default value is 3.

“max_workspace_size_bytes”: The maximum GPU memory the model can use temporarily during execution. Default value is 1GB.

For “gpu_io”, no parameters are required. If set, the model will be executed using TensorFlow Callable API to set input and output tensors in GPU memory if possible, which can reduce data transfer overhead if the model is used in ensemble. However, the Callable object will be created on model creation and it will request all outputs for every model execution, which may impact the performance if a request does not require all outputs. This optimization will only take affect if the model instance is created with KIND_GPU.

Accelerator cpu_execution_accelerator(repeated)¶

The preferred execution provider to be used if the model instance is deployed on CPU.

For ONNX Runtime backend, possible value is “openvino” as name, and no parameters are required.

message PinnedMemoryBuffer¶

Specify whether to use a pinned memory buffer when transferring data between non-pinned system memory and GPU memory. Using a pinned memory buffer for system from/to GPU transfers will typically provide increased performance. For example, in the common use case where the request provides inputs and delivers outputs via non-pinned system memory, if the model instance accepts GPU IOs, the inputs will be processed by two copies: from non-pinned system memory to pinned memory, and from pinned memory to GPU memory. Similarly, pinned memory will be used for delivering the outputs.

bool enable¶: Use pinned memory buffer. Default is true.

Graph graph¶: The graph optimization setting for the model. Optional.

ModelPriority priority¶: The priority setting for the model. Optional.

Cuda cuda¶: CUDA-specific optimization settings. Optional.

ExecutionAccelerators execution_accelerators¶: The accelerators used for the model. Optional.

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.

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.

message ModelDynamicBatching¶

Dynamic batching configuration. These settings control how dynamic batching operates for the model.

int32 preferred_batch_size(repeated)¶: Preferred batch sizes for dynamic batching. If a batch of one of these sizes can be formed it will be executed immediately. If not specified a preferred batch size will be chosen automatically based on model and GPU characteristics.

uint64 max_queue_delay_microseconds¶: The maximum time, in microseconds, a request will be delayed in the scheduling queue to wait for additional requests for batching. Default is 0.

bool preserve_ordering¶: Should the dynamic batcher preserve the ordering of responses to match the order of requests received by the scheduler. Default is false. If true, the responses will be returned in the same order as the order of requests sent to the scheduler. If false, the responses may be returned in arbitrary order. This option is specifically needed when a sequence of related inference requests (i.e. inference requests with the same correlation ID) are sent to the dynamic batcher to ensure that the sequence responses are in the correct order.

message ModelSequenceBatching¶

Sequence batching configuration. These settings control how sequence batching operates for the model.

message Control¶

A control is a signal that the sequence batcher uses to communicate with a backend.

enum Kind¶

The kind of the control.

enumerator Kind::CONTROL_SEQUENCE_START = 0¶: A new sequence is/is-not starting. If true a sequence is starting, if false a sequence is continuing. Must specify either int32_false_true or fp32_false_true for this control. This control is optional.

enumerator Kind::CONTROL_SEQUENCE_READY = 1¶: A sequence is/is-not ready for inference. If true the input tensor data is valid and should be used. If false the input tensor data is invalid and inferencing should be “skipped”. Must specify either int32_false_true or fp32_false_true for this control. This control is optional.

enumerator Kind::CONTROL_SEQUENCE_END = 2¶: A sequence is/is-not ending. If true a sequence is ending, if false a sequence is continuing. Must specify either int32_false_true or fp32_false_true for this control. This control is optional.

enumerator Kind::CONTROL_SEQUENCE_CORRID = 3¶: The correlation ID of the sequence. The correlation ID is an uint64_t value that is communicated in whole or in part by the tensor. The tensor’s datatype must be specified by data_type and must be TYPE_UINT64, TYPE_INT64, TYPE_UINT32 or TYPE_INT32. If a 32-bit datatype is specified the correlation ID will be truncated to the low-order 32 bits. This control is optional.

Kind kind¶: The kind of this control.

int32 int32_false_true(repeated)¶: The control’s true and false setting is indicated by setting a value in an int32 tensor. The tensor must be a 1-dimensional tensor with size equal to the batch size of the request. ‘int32_false_true’ must have two entries: the first the false value and the second the true value.

float fp32_false_true(repeated)¶: The control’s true and false setting is indicated by setting a value in a fp32 tensor. The tensor must be a 1-dimensional tensor with size equal to the batch size of the request. ‘fp32_false_true’ must have two entries: the first the false value and the second the true value.

DataType data_type¶: The control’s datatype.

message ControlInput¶

The sequence control values to communicate by a model input.

string name¶: The name of the model input.

Control control(repeated)¶: The control value(s) that should be communicated to the model using this model input.

message StrategyDirect¶: The sequence batcher uses a specific, unique batch slot for each sequence. All inference requests in a sequence are directed to the same batch slot in the same model instance over the lifetime of the sequence. This is the default strategy.

message StrategyOldest¶

The sequence batcher maintains up to ‘max_candidate_sequences’ candidate sequences. ‘max_candidate_sequences’ can be greater than the model’s ‘max_batch_size’. For inferencing the batcher chooses from the candidate sequences up to ‘max_batch_size’ inference requests. Requests are chosen in an oldest-first manner across all candidate sequences. A given sequence is not guaranteed to be assigned to the same batch slot for all inference requests of that sequence.

int32 max_candidate_sequences¶: Maximum number of candidate sequences that the batcher maintains. Excess seqences are kept in an ordered backlog and become candidates when existing candidate sequences complete.

int32 preferred_batch_size(repeated)¶: Preferred batch sizes for dynamic batching of candidate sequences. If a batch of one of these sizes can be formed it will be executed immediately. If not specified a preferred batch size will be chosen automatically based on model and GPU characteristics.

uint64 max_queue_delay_microseconds¶: The maximum time, in microseconds, a candidate request will be delayed in the dynamic batch scheduling queue to wait for additional requests for batching. Default is 0.

oneof strategy_choice¶

The strategy used by the sequence batcher. Default strategy is ‘direct’.

StrategyDirect direct¶: StrategyDirect scheduling strategy.

StrategyOldest oldest¶: StrategyOldest scheduling strategy.

uint64 max_sequence_idle_microseconds¶: The maximum time, in microseconds, that a sequence is allowed to be idle before it is aborted. The inference server considers a sequence idle when it does not have any inference request queued for the sequence. If this limit is exceeded, the inference server will free the sequence slot allocated by the sequence and make it available for another sequence. If not specified (or specified as zero) a default value of 1000000 (1 second) is used.

ControlInput control_input(repeated)¶: The model input(s) that the server should use to communicate sequence start, stop, ready and similar control values to the model.

message ModelEnsembling¶

Model ensembling configuration. These settings specify the models that compose the ensemble and how data flows between the models.

message Step¶: Each step specifies a model included in the ensemble, maps ensemble tensor names to the model input tensors, and maps model output tensors to ensemble tensor names

string model_name¶: The name of the model to execute for this step of the ensemble.

int64 model_version¶: The version of the model to use for inference. If -1 the latest/most-recent version of the model is used.

map<string, string> input_map¶: Map from name of an input tensor on this step’s model to ensemble tensor name. The ensemble tensor must have the same data type and shape as the model input. Each model input must be assigned to one ensemble tensor, but the same ensemble tensor can be assigned to multiple model inputs.

map<string, string> output_map¶: Map from name of an output tensor on this step’s model to ensemble tensor name. The data type and shape of the ensemble tensor will be inferred from the model output. It is optional to assign all model outputs to ensemble tensors. One ensemble tensor name can appear in an output map only once.

Step step(repeated)¶: The models and the input / output mappings used within the ensemble.

message ModelParameter¶

A model parameter.

string string_value¶: The string value of the parameter.

message ModelWarmup¶

Settings used to construct the request sample for model warmup.

message Input¶

Meta data associated with an input.

DataType data_type¶: The data-type of the input.

int64 dims(repeated)¶: The shape of the input tensor, not including the batch dimension.

oneof input_data_type¶: Specify how the input data is generated. If the input has STRING data type and ‘random_data’ is set, the data generation will fall back to ‘zero_data’.

bool zero_data¶: The identifier for using zeros as input data. Note that the value of ‘zero_data’ will not be checked, instead, zero data will be used as long as the field is set.

bool random_data¶: The identifier for using random data as input data. Note that the value of ‘random_data’ will not be checked, instead, random data will be used as long as the field is set.

string input_data_file¶: The file whose content will be used as raw input data in row-major order. The file must be provided in a sub-directory ‘warmup’ under the model directory.

string name¶: The name of the request sample.

uint32 batch_size¶: The batch size of the inference request. This must be >= 1. For models that don’t support batching, batch_size must be 1. If batch_size > 1, the ‘inputs’ specified below will be duplicated to match the batch size requested.

map<string, Input> inputs¶: The warmup meta data associated with every model input, including control tensors.

message ModelConfig¶

A model configuration.

string name¶: The name of the model.

string platform¶: The framework for the model. Possible values are “tensorrt_plan”, “tensorflow_graphdef”, “tensorflow_savedmodel”, “caffe2_netdef”, “onnxruntime_onnx”, “pytorch_libtorch” and “custom”.

ModelVersionPolicy version_policy¶: Policy indicating which version(s) of the model will be served.

int32 max_batch_size¶: Maximum batch size allowed for inference. This can only decrease what is allowed by the model itself. A max_batch_size value of 0 indicates that batching is not allowed for the model and the dimension/shape of the input and output tensors must exactly match what is specified in the input and output configuration. A max_batch_size value > 0 indicates that batching is allowed and so the model expects the input tensors to have an additional initial dimension for the batching that is not specified in the input (for example, if the model supports batched inputs of 2-dimensional tensors then the model configuration will specify the input shape as [ X, Y ] but the model will expect the actual input tensors to have shape [ N, X, Y ]). For max_batch_size > 0 returned outputs will also have an additional initial dimension for the batch.

ModelInput input(repeated)¶: The inputs request by the model.

ModelOutput output(repeated)¶: The outputs produced by the model.

ModelOptimizationPolicy optimization¶: Optimization configuration for the model. If not specified then default optimization policy is used.

oneof scheduling_choice¶

The scheduling policy for the model. If not specified the default scheduling policy is used for the model. The default policy is to execute each inference request independently.

ModelDynamicBatching dynamic_batching¶: If specified, enables the dynamic-batching scheduling policy. With dynamic-batching the scheduler may group together independent requests into a single batch to improve inference throughput.

ModelSequenceBatching sequence_batching¶: If specified, enables the sequence-batching scheduling policy. With sequence-batching, inference requests with the same correlation ID are routed to the same model instance. Multiple sequences of inference requests may be batched together into a single batch to improve inference throughput.

ModelEnsembling ensemble_scheduling¶: If specified, enables the model-ensembling scheduling policy. With model-ensembling, inference requests will be processed according to the specification, such as an execution sequence of models. The input specified in this model config will be the input for the ensemble, and the output specified will be the output of the ensemble.

ModelInstanceGroup instance_group(repeated)¶: Instances of this model. If not specified, one instance of the model will be instantiated on each available GPU.

string default_model_filename¶: Optional filename of the model file to use if a compute-capability specific model is not specified in cc_model_filenames. If not specified the default name is ‘model.graphdef’, ‘model.savedmodel’, ‘model.plan’ or ‘model.netdef’ depending on the model type.

map<string, string> cc_model_filenames¶: Optional map from CUDA compute capability to the filename of the model that supports that compute capability. The filename refers to a file within the model version directory.

map<string, string> metric_tags¶: Optional metric tags. User-specific key-value pairs for metrics reported for this model. These tags are applied to the metrics reported on the HTTP metrics port.

map<string, ModelParameter> parameters¶: Optional model parameters. User-specified parameter values that are made available to custom backends.

ModelWarmup model_warmup(repeated)¶: Warmup setting of this model. If specified, all instances will be run with the request samples in sequence before serving the model. This field can only be specified if the model is not an ensemble model.