Package co.elastic.clients.elasticsearch.ml

Class StartTrainedModelDeploymentRequest

java.lang.Object
co.elastic.clients.elasticsearch._types.RequestBase
co.elastic.clients.elasticsearch.ml.StartTrainedModelDeploymentRequest
public class StartTrainedModelDeploymentRequest
extends RequestBase
Starts a trained model deployment, which allocates the model to every machine learning node.
See Also:
API specification

  • Field Details

  • Method Details

    • of

      public static StartTrainedModelDeploymentRequest of​(java.util.function.Function<StartTrainedModelDeploymentRequest.Builder,​ObjectBuilder<StartTrainedModelDeploymentRequest>> fn)

    • cacheSize

      @Nullable public final java.lang.String cacheSize()
      The inference cache size (in memory outside the JVM heap) per node for the model. The default value is the same size as the model_size_bytes. To disable the cache, 0b can be provided.

      API name: cache_size

    • modelId

      public final java.lang.String modelId()
      Required - The unique identifier of the trained model. Currently, only PyTorch models are supported.

      API name: model_id

    • numberOfAllocations

      @Nullable public final java.lang.Integer numberOfAllocations()
      The number of model allocations on each node where the model is deployed. All allocations on a node share the same copy of the model in memory but use a separate set of threads to evaluate the model. Increasing this value generally increases the throughput. If this setting is greater than the number of hardware threads it will automatically be changed to a value less than the number of hardware threads.

      API name: number_of_allocations

    • queueCapacity

      @Nullable public final java.lang.Integer queueCapacity()
      Specifies the number of inference requests that are allowed in the queue. After the number of requests exceeds this value, new requests are rejected with a 429 error.

      API name: queue_capacity

    • threadsPerAllocation

      @Nullable public final java.lang.Integer threadsPerAllocation()
      Sets the number of threads used by each model allocation during inference. This generally increases the inference speed. The inference process is a compute-bound process; any number greater than the number of available hardware threads on the machine does not increase the inference speed. If this setting is greater than the number of hardware threads it will automatically be changed to a value less than the number of hardware threads.

      API name: threads_per_allocation

    • timeout

      @Nullable public final Time timeout()
      Specifies the amount of time to wait for the model to deploy.

      API name: timeout

    • waitFor

      @Nullable public final DeploymentAllocationState waitFor()
      Specifies the allocation status to wait for before returning.

      API name: wait_for