braket.jobs.config module

class braket.jobs.config.CheckpointConfig(localPath: str = '/opt/jobs/checkpoints', s3Uri: str | None = None)

Bases: object

Configuration that specifies the location where checkpoint data is stored.

localPath: str = '/opt/jobs/checkpoints'

s3Uri: str | None = None

class braket.jobs.config.InstanceConfig(instanceType: str = 'ml.m5.large', volumeSizeInGb: int = 30, instanceCount: int = 1)

Bases: object

Configuration of the instance(s) used to run the hybrid job.

instanceType: str = 'ml.m5.large'

volumeSizeInGb: int = 30

instanceCount: int = 1

class braket.jobs.config.OutputDataConfig(s3Path: str | None = None, kmsKeyId: str | None = None)

Bases: object

Configuration that specifies the location for the output of the hybrid job.

s3Path: str | None = None

kmsKeyId: str | None = None

class braket.jobs.config.StoppingCondition(maxRuntimeInSeconds: int = 432000)

Bases: object

Conditions that specify when the hybrid job should be forcefully stopped.

maxRuntimeInSeconds: int = 432000

class braket.jobs.config.DeviceConfig(device: 'str')

Bases: object

device: str

class braket.jobs.config.S3DataSourceConfig(s3_data: str, content_type: str | None = None)

Bases: object

Data source for data that lives on S3.

config

config passed to the Braket API

Type:

dict[str, dict]

Create a definition for input data used by a Braket Hybrid job.

Parameters:

• s3_data (str) – Defines the location of s3 data to train on.

• content_type (str | None) – MIME type of the input data (default: None).