# Generated by API Evangelist (build-phrasing.py). Our phrasing, not observed demand.
overlay: 1.0.0
info:
  title: API Evangelist conversational phrasing for Amazon SageMaker Training Jobs API
  version: 1.0.0
extends: openapi/amazon-sagemaker-training-jobs-api-openapi.yml
actions:
- target: $.info
  update:
    x-apievangelist-phrasing:
      method: generated
      generated: '2026-10-01'
      generator: build-phrasing.py
      label: Generated by API Evangelist
      operations: 3
- target: $.paths['/#CreateTrainingJob'].post
  update:
    x-apievangelist-phrasing:
      intent: Start a model training job
      effect: write
      questions:
      - How do I kick off training a model on my own data in SageMaker?
      - Can I pass hyperparameters and cap how long a training run is allowed to take?
      - Where do the trained model artifacts end up after a training job finishes?
      instructions:
      - text: Start training job {job_name} with algorithm {algorithm}, role {role_arn}, compute {resources}, output to {output}, stopping at {stopping}.
        slots:
          job_name: requestBody.TrainingJobName
          algorithm: requestBody.AlgorithmSpecification
          role_arn: requestBody.RoleArn
          resources: requestBody.ResourceConfig
          output: requestBody.OutputDataConfig
          stopping: requestBody.StoppingCondition
      - text: Train {job_name} on input data {input} using hyperparameters {hyperparameters}.
        slots:
          job_name: requestBody.TrainingJobName
          input: requestBody.InputDataConfig
          hyperparameters: requestBody.HyperParameters
      method: generated
      generated: '2026-10-01'
- target: $.paths['/#DescribeTrainingJob'].post
  update:
    x-apievangelist-phrasing:
      intent: Check the progress of one training job
      effect: read
      questions:
      - Has my training job finished, and did it succeed or fail?
      - What hyperparameters and compute did a past training run use?
      instructions:
      - text: Describe training job {job_name}.
        slots:
          job_name: requestBody.TrainingJobName
      - text: Tell me the current status of training job {job_name}.
        slots:
          job_name: requestBody.TrainingJobName
      method: generated
      generated: '2026-10-01'
- target: $.paths['/#ListTrainingJobs'].post
  update:
    x-apievangelist-phrasing:
      intent: List training jobs
      effect: read
      questions:
      - Which training jobs have I run recently?
      - Can I filter my training jobs down to only the ones still in progress?
      instructions:
      - text: List all of my training jobs.
      - text: List training jobs with status {status}.
        slots:
          status: requestBody.StatusEquals
      - text: Show {max_results} training jobs sorted by {sort_by} in {sort_order} order.
        slots:
          max_results: requestBody.MaxResults
          sort_by: requestBody.SortBy
          sort_order: requestBody.SortOrder
      method: generated
      generated: '2026-10-01'