Skip to content

CreateMLTransformRequest

Source code in docs_src/capo_glue/types/create_ml_transform_request.py
class CreateMLTransformRequest(TypedDict, closed=True):
    name: "capo_glue.types.name_string.NameString"
    """<p>The unique name that you give the transform when you create it.</p>"""
    description: NotRequired["capo_glue.types.description_string.DescriptionString"]
    """<p>A description of the machine learning transform that is being defined. The default is an empty string.</p>"""
    input_record_tables: "capo_glue.types.glue_tables.GlueTables"
    """<p>A list of Glue table definitions used by the transform.</p>"""
    parameters: "capo_glue.types.transform_parameters.TransformParameters"
    """<p>The algorithmic parameters that are specific to the transform type used. Conditionally dependent on the transform type.</p>"""
    role: "capo_glue.types.role_string.RoleString"
    """<p>The name or Amazon Resource Name (ARN) of the IAM role with the required permissions. The required permissions include both Glue service role permissions to Glue resources, and Amazon S3 permissions required by the transform. </p> <ul> <li> <p>This role needs Glue service role permissions to allow access to resources in Glue. See <a href="https://docs.aws.amazon.com/glue/latest/dg/attach-policy-iam-user.html">Attach a Policy to IAM Users That Access Glue</a>.</p> </li> <li> <p>This role needs permission to your Amazon Simple Storage Service (Amazon S3) sources, targets, temporary directory, scripts, and any libraries used by the task run for this transform.</p> </li> </ul>"""
    glue_version: NotRequired["capo_glue.types.glue_version_string.GlueVersionString"]
    """<p>This value determines which version of Glue this machine learning transform is compatible with. Glue 1.0 is recommended for most customers. If the value is not set, the Glue compatibility defaults to Glue 0.9. For more information, see <a href="https://docs.aws.amazon.com/glue/latest/dg/release-notes.html#release-notes-versions">Glue Versions</a> in the developer guide.</p>"""
    max_capacity: NotRequired["capo_glue.types.nullable_double.NullableDouble"]
    """<p>The number of Glue data processing units (DPUs) that are allocated to task runs for this transform. You can allocate from 2 to 100 DPUs; the default is 10. A DPU is a relative measure of processing power that consists of 4 vCPUs of compute capacity and 16 GB of memory. For more information, see the <a href="https://aws.amazon.com/glue/pricing/">Glue pricing page</a>. </p> <p> <code>MaxCapacity</code> is a mutually exclusive option with <code>NumberOfWorkers</code> and <code>WorkerType</code>.</p> <ul> <li> <p>If either <code>NumberOfWorkers</code> or <code>WorkerType</code> is set, then <code>MaxCapacity</code> cannot be set.</p> </li> <li> <p>If <code>MaxCapacity</code> is set then neither <code>NumberOfWorkers</code> or <code>WorkerType</code> can be set.</p> </li> <li> <p>If <code>WorkerType</code> is set, then <code>NumberOfWorkers</code> is required (and vice versa).</p> </li> <li> <p> <code>MaxCapacity</code> and <code>NumberOfWorkers</code> must both be at least 1.</p> </li> </ul> <p>When the <code>WorkerType</code> field is set to a value other than <code>Standard</code>, the <code>MaxCapacity</code> field is set automatically and becomes read-only.</p> <p>When the <code>WorkerType</code> field is set to a value other than <code>Standard</code>, the <code>MaxCapacity</code> field is set automatically and becomes read-only.</p>"""
    worker_type: NotRequired["capo_glue.types.worker_type.WorkerType"]
    """<p>The type of predefined worker that is allocated when this task runs. Accepts a value of Standard, G.1X, or G.2X.</p> <ul> <li> <p>For the <code>Standard</code> worker type, each worker provides 4 vCPU, 16 GB of memory and a 50GB disk, and 2 executors per worker.</p> </li> <li> <p>For the <code>G.1X</code> worker type, each worker provides 4 vCPU, 16 GB of memory and a 64GB disk, and 1 executor per worker.</p> </li> <li> <p>For the <code>G.2X</code> worker type, each worker provides 8 vCPU, 32 GB of memory and a 128GB disk, and 1 executor per worker.</p> </li> </ul> <p> <code>MaxCapacity</code> is a mutually exclusive option with <code>NumberOfWorkers</code> and <code>WorkerType</code>.</p> <ul> <li> <p>If either <code>NumberOfWorkers</code> or <code>WorkerType</code> is set, then <code>MaxCapacity</code> cannot be set.</p> </li> <li> <p>If <code>MaxCapacity</code> is set then neither <code>NumberOfWorkers</code> or <code>WorkerType</code> can be set.</p> </li> <li> <p>If <code>WorkerType</code> is set, then <code>NumberOfWorkers</code> is required (and vice versa).</p> </li> <li> <p> <code>MaxCapacity</code> and <code>NumberOfWorkers</code> must both be at least 1.</p> </li> </ul>"""
    number_of_workers: NotRequired["capo_glue.types.nullable_integer.NullableInteger"]
    """<p>The number of workers of a defined <code>workerType</code> that are allocated when this task runs.</p> <p>If <code>WorkerType</code> is set, then <code>NumberOfWorkers</code> is required (and vice versa).</p>"""
    timeout: NotRequired["capo_glue.types.timeout.Timeout"]
    """<p>The timeout of the task run for this transform in minutes. This is the maximum time that a task run for this transform can consume resources before it is terminated and enters <code>TIMEOUT</code> status. The default is 2,880 minutes (48 hours).</p>"""
    max_retries: NotRequired["capo_glue.types.nullable_integer.NullableInteger"]
    """<p>The maximum number of times to retry a task for this transform after a task run fails.</p>"""
    tags: NotRequired["capo_glue.types.tags_map.TagsMap"]
    """<p>The tags to use with this machine learning transform. You may use tags to limit access to the machine learning transform. For more information about tags in Glue, see <a href="https://docs.aws.amazon.com/glue/latest/dg/monitor-tags.html">Amazon Web Services Tags in Glue</a> in the developer guide.</p>"""
    transform_encryption: NotRequired[
        "capo_glue.types.transform_encryption.TransformEncryption"
    ]
    """<p>The encryption-at-rest settings of the transform that apply to accessing user data. Machine learning transforms can access user data encrypted in Amazon S3 using KMS.</p>"""

description instance-attribute

description: NotRequired[DescriptionString]

A description of the machine learning transform that is being defined. The default is an empty string.

glue_version instance-attribute

glue_version: NotRequired[GlueVersionString]

This value determines which version of Glue this machine learning transform is compatible with. Glue 1.0 is recommended for most customers. If the value is not set, the Glue compatibility defaults to Glue 0.9. For more information, see Glue Versions in the developer guide.

input_record_tables instance-attribute

input_record_tables: GlueTables

A list of Glue table definitions used by the transform.

max_capacity instance-attribute

max_capacity: NotRequired[NullableDouble]

The number of Glue data processing units (DPUs) that are allocated to task runs for this transform. You can allocate from 2 to 100 DPUs; the default is 10. A DPU is a relative measure of processing power that consists of 4 vCPUs of compute capacity and 16 GB of memory. For more information, see the Glue pricing page.

MaxCapacity is a mutually exclusive option with NumberOfWorkers and WorkerType.

  • If either NumberOfWorkers or WorkerType is set, then MaxCapacity cannot be set.

  • If MaxCapacity is set then neither NumberOfWorkers or WorkerType can be set.

  • If WorkerType is set, then NumberOfWorkers is required (and vice versa).

  • MaxCapacity and NumberOfWorkers must both be at least 1.

When the WorkerType field is set to a value other than Standard, the MaxCapacity field is set automatically and becomes read-only.

When the WorkerType field is set to a value other than Standard, the MaxCapacity field is set automatically and becomes read-only.

max_retries instance-attribute

max_retries: NotRequired[NullableInteger]

The maximum number of times to retry a task for this transform after a task run fails.

name instance-attribute

name: NameString

The unique name that you give the transform when you create it.

number_of_workers instance-attribute

number_of_workers: NotRequired[NullableInteger]

The number of workers of a defined workerType that are allocated when this task runs.

If WorkerType is set, then NumberOfWorkers is required (and vice versa).

parameters instance-attribute

parameters: TransformParameters

The algorithmic parameters that are specific to the transform type used. Conditionally dependent on the transform type.

role instance-attribute

role: RoleString

The name or Amazon Resource Name (ARN) of the IAM role with the required permissions. The required permissions include both Glue service role permissions to Glue resources, and Amazon S3 permissions required by the transform.

  • This role needs Glue service role permissions to allow access to resources in Glue. See Attach a Policy to IAM Users That Access Glue.

  • This role needs permission to your Amazon Simple Storage Service (Amazon S3) sources, targets, temporary directory, scripts, and any libraries used by the task run for this transform.

tags instance-attribute

tags: NotRequired[TagsMap]

The tags to use with this machine learning transform. You may use tags to limit access to the machine learning transform. For more information about tags in Glue, see Amazon Web Services Tags in Glue in the developer guide.

timeout instance-attribute

timeout: NotRequired[Timeout]

The timeout of the task run for this transform in minutes. This is the maximum time that a task run for this transform can consume resources before it is terminated and enters TIMEOUT status. The default is 2,880 minutes (48 hours).

transform_encryption instance-attribute

transform_encryption: NotRequired[TransformEncryption]

The encryption-at-rest settings of the transform that apply to accessing user data. Machine learning transforms can access user data encrypted in Amazon S3 using KMS.

worker_type instance-attribute

worker_type: NotRequired[WorkerType]

The type of predefined worker that is allocated when this task runs. Accepts a value of Standard, G.1X, or G.2X.

  • For the Standard worker type, each worker provides 4 vCPU, 16 GB of memory and a 50GB disk, and 2 executors per worker.

  • For the G.1X worker type, each worker provides 4 vCPU, 16 GB of memory and a 64GB disk, and 1 executor per worker.

  • For the G.2X worker type, each worker provides 8 vCPU, 32 GB of memory and a 128GB disk, and 1 executor per worker.

MaxCapacity is a mutually exclusive option with NumberOfWorkers and WorkerType.

  • If either NumberOfWorkers or WorkerType is set, then MaxCapacity cannot be set.

  • If MaxCapacity is set then neither NumberOfWorkers or WorkerType can be set.

  • If WorkerType is set, then NumberOfWorkers is required (and vice versa).

  • MaxCapacity and NumberOfWorkers must both be at least 1.