3.0.0 • Published 5 years ago

@datafire/azure_machinelearningservices_execution v3.0.0

Weekly downloads
1
License
MIT
Repository
github
Last release
5 years ago

@datafire/azure_machinelearningservices_execution

Client library for Execution Service

Installation and Usage

npm install --save @datafire/azure_machinelearningservices_execution
let azure_machinelearningservices_execution = require('@datafire/azure_machinelearningservices_execution').create({
  access_token: "",
  refresh_token: "",
  client_id: "",
  client_secret: "",
  redirect_uri: ""
});

.then(data => {
  console.log(data);
});

Description

Actions

Runs_CancelRunWithUri

Cancels a run within an experiment.

azure_machinelearningservices_execution.Runs_CancelRunWithUri({
  "subscriptionId": "",
  "resourceGroupName": "",
  "workspaceName": "",
  "experimentName": "",
  "runId": ""
}, context)

Input

  • input object
    • subscriptionId required string: The Azure Subscription ID.
    • resourceGroupName required string: The Name of the resource group in which the workspace is located.
    • workspaceName required string: The name of the workspace.
    • experimentName required string: The experiment name.
    • runId required string: The id of the run to cancel.

Output

Runs_StartSnapshotRun

Starts an experiment run on the remote compute target using the provided definition.json file to define the run. The code for the run is retrieved using the snapshotId in definition.json.

azure_machinelearningservices_execution.Runs_StartSnapshotRun({
  "subscriptionId": "",
  "resourceGroupName": "",
  "workspaceName": "",
  "experimentName": "",
  "definition": {}
}, context)

Input

  • input object
    • subscriptionId required string: The Azure Subscription ID.
    • resourceGroupName required string: The Name of the resource group in which the workspace is located.
    • workspaceName required string: The name of the workspace.
    • experimentName required string: The experiment name.
    • definition required RunDefinition
    • runId string: A run id. If not supplied a run id will be created automatically.

Output

Runs_StartLocalRun

Starts an experiment run using the provided definition.json file to define the run. The source code and configuration is defined in a zip archive in project.zip.

azure_machinelearningservices_execution.Runs_StartLocalRun({
  "subscriptionId": "",
  "resourceGroupName": "",
  "workspaceName": "",
  "experimentName": "",
  "definition": {}
}, context)

Input

  • input object
    • subscriptionId required string: The Azure Subscription ID.
    • resourceGroupName required string: The Name of the resource group in which the workspace is located.
    • workspaceName required string: The name of the workspace.
    • experimentName required string: The experiment name.
    • definition required RunDefinition
    • runId string: A run id. If not supplied a run id will be created automatically.

Output

  • output file

Runs_StartRun

Starts an experiment run using the provided definition.json file to define the run. The source code and configuration is defined in a zip archive in project.zip.

azure_machinelearningservices_execution.Runs_StartRun({
  "subscriptionId": "",
  "resourceGroupName": "",
  "workspaceName": "",
  "experimentName": "",
  "runDefinitionFile": "",
  "projectZipFile": ""
}, context)

Input

  • input object
    • subscriptionId required string: The Azure Subscription ID.
    • resourceGroupName required string: The Name of the resource group in which the workspace is located.
    • workspaceName required string: The name of the workspace.
    • experimentName required string: The experiment name.
    • runDefinitionFile required string, object: The JSON file containing the RunDefinition
      • content string
      • encoding string (values: ascii, utf8, utf16le, base64, binary, hex)
      • contentType string
      • filename string
    • projectZipFile required string, object: The zip archive of the project folder containing the source code to use for the run.
      • content string
      • encoding string (values: ascii, utf8, utf16le, base64, binary, hex)
      • contentType string
      • filename string
    • runId string: A run id. If not supplied a run id will be created automatically.

Output

Definitions

ContainerRegistry

  • ContainerRegistry object
    • address string
    • password string
    • username string

DataReferenceConfiguration

  • DataReferenceConfiguration object: A class for managing DataReferenceConfiguration.
    • dataStoreName string: The name of the data store.
    • mode string (values: Mount, Download, Upload): Operation on the datastore, mount, download, upload.
    • overwrite boolean: Whether to overwrite the data if existing.
    • pathOnCompute string: The path on the compute target.
    • pathOnDataStore string: Relative path on the datastore.

DockerSection

  • DockerSection object
    • arguments array: Extra arguments to the Docker run command.
      • items string
    • baseDockerfile string: Base Dockerfile used for Docker-based runs. Mutually exclusive with BaseImage.
    • baseImage string: Base image used for Docker-based runs. Mutually exclusive with BaseDockerfile.
    • baseImageRegistry ContainerRegistry
    • enabled boolean: Set true to perform this run inside a Docker container.
    • sharedVolumes boolean: Set false to disable AzureML's usage of the Docker shared volumes feature to work around bugs in certain versions of Docker for Windows.

EnvironmentDefinition

  • EnvironmentDefinition object
    • docker DockerSection
    • environmentVariables object: Definition of environment variables to be defined in the environment.
    • inferencingStackVersion string: The inferencing stack version added to the image. To avoid adding an inferencing stack, do not set this value. Valid values: "latest".
    • name string: The name of the environment.
    • python PythonSection
    • spark SparkSection
    • version string: The environment version.

ErrorDetails

  • ErrorDetails object: The error details.
    • code string: The error code.
    • message string: The error message.
    • target string: The target of the error (e.g., the name of the property in error).

ErrorResponse

  • ErrorResponse object: The error response.
    • correlation object: Dictionary containing correlation details for the error.
    • environment string: The hosting environment.
    • error RootError
    • location string: The Azure region.
    • time string: The time in UTC.

HdiConfiguration

  • HdiConfiguration object
    • yarnDeployMode string (values: None, Client, Cluster)

HistoryConfiguration

  • HistoryConfiguration object
    • directoriesToWatch array: The list of directories to monitor and upload files from.
      • items string
    • outputCollection boolean: Set to true to collect outputs and store in run history.

InnerErrorResponse

  • InnerErrorResponse object: A nested structure of errors.

MpiConfiguration

  • MpiConfiguration object
    • processCountPerNode integer: Number of processes per node.

PythonSection

  • PythonSection object
    • baseCondaEnvironment string
    • condaDependencies object
    • interpreterPath string: The python interpreter path. This is only used when user_managed_dependencies=True.
    • userManagedDependencies boolean: True means that AzureML reuses an existing python environment; False means that AzureML will create a python environment based on the Conda dependencies specification.

RootError

  • RootError object: The root error.
    • code string: The service-defined error code. Supported error codes: ServiceError, UserError, ValidationError, AzureStorageError, TransientError, RequestThrottled.
    • details array: The related errors that occurred during the request.
    • innerError InnerErrorResponse
    • message string: A human-readable representation of the error.
    • target string: The target of the error (e.g., the name of the property in error).

RunConfiguration

  • RunConfiguration object
    • arguments array: Command line arguments for the python script file.
      • items string
    • communicator string (values: None, ParameterServer, Gloo, Mpi, Nccl): The supported communicators are None, ParameterServer, OpenMpi, and IntelMpi Keep in mind that OpenMpi requires a custom image with OpenMpi installed.
    • dataReferences object: All the data sources are made available to the run during execution based on each configuration.
    • environment EnvironmentDefinition
    • framework string (values: Python, PySpark, Cntk, TensorFlow, PyTorch): The supported frameworks are Python, PySpark, CNTK, TensorFlow, and PyTorch. Use Tensorflow for AmlCompute clusters, and Python for distributed training jobs.
    • hdi HdiConfiguration
    • history HistoryConfiguration
    • jobName string: This is primarily intended for notebooks to override the default job name.
    • maxRunDurationSeconds integer: Maximum allowed time for the run. The system will attempt to automatically cancel the run if it took longer than this value.
    • mpi MpiConfiguration
    • nodeCount integer: Number of compute nodes to run the job on. Only applies to AMLCompute.
    • script string: The relative path to the python script file. The file path is relative to the source_directory passed to submit run.
    • spark SparkConfiguration
    • target string: Target refers to compute where the job is scheduled for execution. The default target is "local" referring to the local machine.
    • tensorflow TensorflowConfiguration

RunDefinition

  • RunDefinition object
    • configuration RunConfiguration
    • parentRunId string: Specifies that the run history entry for this execution should be scoped within
    • runType string: Specifies the runsource property for this run. The default value is "experiment" if not specified.
    • snapshotId string: Snapshots are user project folders that have been uploaded to the cloud for subsequent

SparkConfiguration

  • SparkConfiguration object
    • configuration object

SparkMavenPackage

  • SparkMavenPackage object
    • artifact string
    • group string
    • version string

SparkSection

  • SparkSection object
    • packages array: The Spark packages to use.
    • precachePackages boolean: Whether to precache the packages.
    • repositories array: The list of spark repositories.
      • items string

StartRunResult

  • StartRunResult object: Contains the details of a run.
    • runId required string: The identifier for a run.

TensorflowConfiguration

  • TensorflowConfiguration object
    • parameterServerCount integer: Number of parameter servers.
    • workerCount integer: The number of workers.
datafire
@infinitebrahmanuniverse/nolb-_datafire_az@everything-registry/sub-chunk-221
3.0.0

5 years ago