TensorFlow

TensorFlow Estimator

class sagemaker.tensorflow.estimator.TensorFlow(py_version: Optional[str] = None, framework_version: Optional[str] = None, model_dir: Optional[Union[str, sagemaker.workflow.entities.PipelineVariable]] = None, image_uri: Optional[Union[str, sagemaker.workflow.entities.PipelineVariable]] = None, distribution: Optional[Dict[str, str]] = None, compiler_config: Optional[sagemaker.tensorflow.training_compiler.config.TrainingCompilerConfig] = None, **kwargs)

Bases: sagemaker.estimator.Framework

Handle end-to-end training and deployment of user-provided TensorFlow code.

Initialize a TensorFlow estimator.

Parameters
  • py_version (str) – Python version you want to use for executing your model training code. Defaults to None. Required unless image_uri is provided.

  • framework_version (str) – TensorFlow version you want to use for executing your model training code. Defaults to None. Required unless image_uri is provided. List of supported versions: https://github.com/aws/sagemaker-python-sdk#tensorflow-sagemaker-estimators.

  • model_dir (str) –

    S3 location where the checkpoint data and models can be exported to during training (default: None). It will be passed in the training script as one of the command line arguments. If not specified, one is provided based on your training configuration:

    • distributed training with SMDistributed or MPI with Horovod - /opt/ml/model

    • single-machine training or distributed training without MPI - s3://{output_path}/model

    • Local Mode with local sources (file:// instead of s3://) - /opt/ml/shared/model

    To disable having model_dir passed to your training script, set model_dir=False.

  • image_uri (str) –

    If specified, the estimator will use this image for training and hosting, instead of selecting the appropriate SageMaker official image based on framework_version and py_version. It can be an ECR url or dockerhub image and tag.

    Examples

    123.dkr.ecr.us-west-2.amazonaws.com/my-custom-image:1.0 custom-image:latest.

    If framework_version or py_version are None, then image_uri is required. If also None, then a ValueError will be raised.

  • distribution (dict) –

    A dictionary with information on how to run distributed training (default: None). Currently, the following are supported: distributed training with parameter servers, SageMaker Distributed (SMD) Data and Model Parallelism, and MPI. SMD Model Parallelism can only be used with MPI.

    To enable the SageMaker distributed data parallelism:

    { "smdistributed": { "dataparallel": { "enabled": True } } }
    

    To enable the SageMaker distributed model parallelism:

    {
        "smdistributed": {
            "modelparallel": {
                "enabled":True,
                "parameters": {
                    "partitions": 2,
                    "microbatches": 4,
                    "placement_strategy": "spread",
                    "pipeline": "interleaved",
                    "optimize": "speed",
                    "ddp": True,
                }
        },
        "mpi": {
            "enabled" : True,
            "processes_per_host" : 8,
        }
    }
    

    Note

    The SageMaker distributed model parallel library internally uses MPI. In order to use model parallelism, MPI also must be enabled.

    See also

    To find a complete list of parameters for SageMaker model parallelism, see Run a Distributed Training Job Using the SageMaker Python SDK.

    To enable MPI:

    {
        "mpi": {
            "enabled": True
        }
    }
    

    To learn more, see Training with Horovod.

    To enable parameter server:

    {
        "parameter_server": {
            "enabled": True
        }
    }
    

    To learn more, see Training with parameter servers.

  • compiler_config (TrainingCompilerConfig) – Configures SageMaker Training Compiler to accelerate training.

  • **kwargs – Additional kwargs passed to the Framework constructor.

Tip

You can find additional parameters for initializing this class at Framework and EstimatorBase.

create_model(role=None, vpc_config_override='VPC_CONFIG_DEFAULT', entry_point=None, source_dir=None, dependencies=None, **kwargs)

Creates TensorFlowModel object to be used for creating SageMaker model entities.

This can be done by deploying it to a SageMaker endpoint, or starting SageMaker Batch Transform jobs.

Parameters
  • role (str) – The TensorFlowModel, which is also used during transform jobs. If not specified, the role from the Estimator is used.

  • vpc_config_override (dict[str, list[str]]) –

    Optional override for VpcConfig set on the model. Default: use subnets and security groups from this Estimator.

    • ’Subnets’ (list[str]): List of subnet ids.

    • ’SecurityGroupIds’ (list[str]): List of security group ids.

  • entry_point (str) – Path (absolute or relative) to the local Python source file which should be executed as the entry point to training. If source_dir is specified, then entry_point must point to a file located at the root of source_dir. If not specified and endpoint_type is ‘tensorflow-serving’, no entry point is used. If endpoint_type is also None, then the training entry point is used.

  • source_dir (str) – Path (absolute or relative or an S3 URI) to a directory with any other serving source code dependencies aside from the entry point file (default: None).

  • dependencies (list[str]) – A list of paths to directories (absolute or relative) with any additional libraries that will be exported to the container (default: None).

  • **kwargs – Additional kwargs passed to TensorFlowModel.

Returns

A TensorFlowModel object.

See TensorFlowModel for full details.

Return type

sagemaker.tensorflow.model.TensorFlowModel

hyperparameters()

Return hyperparameters used by your custom TensorFlow code during model training.

transformer(instance_count, instance_type, strategy=None, assemble_with=None, output_path=None, output_kms_key=None, accept=None, env=None, max_concurrent_transforms=None, max_payload=None, tags=None, role=None, volume_kms_key=None, entry_point=None, vpc_config_override='VPC_CONFIG_DEFAULT', enable_network_isolation=None, model_name=None)

Return a Transformer that uses a SageMaker Model based on the training job.

It reuses the SageMaker Session and base job name used by the Estimator.

Parameters
  • instance_count (int) – Number of EC2 instances to use.

  • instance_type (str) – Type of EC2 instance to use, for example, ‘ml.c4.xlarge’.

  • strategy (str) – The strategy used to decide how to batch records in a single request (default: None). Valid values: ‘MultiRecord’ and ‘SingleRecord’.

  • assemble_with (str) – How the output is assembled (default: None). Valid values: ‘Line’ or ‘None’.

  • output_path (str) – S3 location for saving the transform result. If not specified, results are stored to a default bucket.

  • output_kms_key (str) – Optional. KMS key ID for encrypting the transform output (default: None).

  • accept (str) – The accept header passed by the client to the inference endpoint. If it is supported by the endpoint, it will be the format of the batch transform output.

  • env (dict) – Environment variables to be set for use during the transform job (default: None).

  • max_concurrent_transforms (int) – The maximum number of HTTP requests to be made to each individual transform container at one time.

  • max_payload (int) – Maximum size of the payload in a single HTTP request to the container in MB.

  • tags (list[dict]) – List of tags for labeling a transform job. If none specified, then the tags used for the training job are used for the transform job.

  • role (str) – The IAM Role ARN for the TensorFlowModel, which is also used during transform jobs. If not specified, the role from the Estimator is used.

  • volume_kms_key (str) – Optional. KMS key ID for encrypting the volume attached to the ML compute instance (default: None).

  • entry_point (str) – Path (absolute or relative) to the local Python source file which should be executed as the entry point to training. If source_dir is specified, then entry_point must point to a file located at the root of source_dir. If not specified and endpoint_type is ‘tensorflow-serving’, no entry point is used. If endpoint_type is also None, then the training entry point is used.

  • vpc_config_override (dict[str, list[str]]) –

    Optional override for the VpcConfig set on the model. Default: use subnets and security groups from this Estimator.

    • ’Subnets’ (list[str]): List of subnet ids.

    • ’SecurityGroupIds’ (list[str]): List of security group ids.

  • enable_network_isolation (bool) – Specifies whether container will run in network isolation mode. Network isolation mode restricts the container access to outside networks (such as the internet). The container does not make any inbound or outbound network calls. If True, a channel named “code” will be created for any user entry script for inference. Also known as Internet-free mode. If not specified, this setting is taken from the estimator’s current configuration.

  • model_name (str) – Name to use for creating an Amazon SageMaker model. If not specified, the estimator generates a default job name based on the training image name and current timestamp.

TensorFlow Training Compiler Configuration

class sagemaker.tensorflow.TrainingCompilerConfig(enabled=True, debug=False)

Bases: sagemaker.training_compiler.config.TrainingCompilerConfig

The SageMaker Training Compiler configuration class.

This class initializes a TrainingCompilerConfig instance.

Amazon SageMaker Training Compiler is a feature of SageMaker Training and speeds up training jobs by optimizing model execution graphs.

You can compile TensorFlow models by passing the object of this configuration class to the compiler_config parameter of the TensorFlow estimator.

Parameters
  • enabled (bool) – Optional. Switch to enable SageMaker Training Compiler. The default is True.

  • debug (bool) – Optional. Whether to dump detailed logs for debugging. This comes with a potential performance slowdown. The default is False.

Example: The following code shows the basic usage of the sagemaker.tensorflow.TrainingCompilerConfig() class to run a TensorFlow training job with the compiler.

from sagemaker.tensorflow import TensorFlow, TrainingCompilerConfig

tensorflow_estimator=TensorFlow(
    ...
    compiler_config=TrainingCompilerConfig()
)

See also

For more information about how to enable SageMaker Training Compiler for various training settings such as using TensorFlow-based models, PyTorch-based models, and distributed training, see Enable SageMaker Training Compiler in the Amazon SageMaker Training Compiler developer guide.

SUPPORTED_INSTANCE_CLASS_PREFIXES = ['p3', 'g4dn', 'p4', 'g5']
MIN_SUPPORTED_VERSION = '2.9'
classmethod validate(estimator)

Checks if SageMaker Training Compiler is configured correctly.

Parameters

estimator (str) – A estimator object If SageMaker Training Compiler is enabled, it will validate whether the estimator is configured to be compatible with Training Compiler.

Raises

ValueError – Raised if the requested configuration is not compatible with SageMaker Training Compiler.

TensorFlow Serving Model

class sagemaker.tensorflow.model.TensorFlowModel(model_data: Union[str, sagemaker.workflow.entities.PipelineVariable], role: str, entry_point: Optional[str] = None, image_uri: Optional[Union[str, sagemaker.workflow.entities.PipelineVariable]] = None, framework_version: Optional[str] = None, container_log_level: Optional[int] = None, predictor_cls: callable = <class 'sagemaker.tensorflow.model.TensorFlowPredictor'>, **kwargs)

Bases: sagemaker.model.FrameworkModel

A FrameworkModel implementation for inference with TensorFlow Serving.

Initialize a Model.

Parameters
  • model_data (str) – The S3 location of a SageMaker model data .tar.gz file.

  • role (str) – An AWS IAM role (either name or full ARN). The Amazon SageMaker training jobs and APIs that create Amazon SageMaker endpoints use this role to access training data and model artifacts. After the endpoint is created, the inference code might use the IAM role, if it needs to access an AWS resource.

  • entry_point (str) – Path (absolute or relative) to the Python source file which should be executed as the entry point to model hosting. If source_dir is specified, then entry_point must point to a file located at the root of source_dir.

  • image_uri (str) – A Docker image URI (default: None). If not specified, a default image for TensorFlow Serving will be used. If framework_version is None, then image_uri is required. If image_uri is also None, then a ValueError will be raised.

  • framework_version (str) – Optional. TensorFlow Serving version you want to use. Defaults to None. Required unless image_uri is provided.

  • container_log_level (int) – Log level to use within the container (default: logging.ERROR). Valid values are defined in the Python logging module.

  • predictor_cls (callable[str, sagemaker.session.Session]) – A function to call to create a predictor with an endpoint name and SageMaker Session. If specified, deploy() returns the result of invoking this function on the created endpoint name.

  • **kwargs – Keyword arguments passed to the superclass FrameworkModel and, subsequently, its superclass Model.

Tip

You can find additional parameters for initializing this class at FrameworkModel and Model.

LOG_LEVEL_PARAM_NAME = 'SAGEMAKER_TFS_NGINX_LOGLEVEL'
LOG_LEVEL_MAP = {10: 'debug', 20: 'info', 30: 'warn', 40: 'error', 50: 'crit'}
LATEST_EIA_VERSION = [2, 3]
register(content_types: List[Union[str, sagemaker.workflow.entities.PipelineVariable]], response_types: List[Union[str, sagemaker.workflow.entities.PipelineVariable]], inference_instances: Optional[List[Union[str, sagemaker.workflow.entities.PipelineVariable]]] = None, transform_instances: Optional[List[Union[str, sagemaker.workflow.entities.PipelineVariable]]] = None, model_package_name: Optional[Union[str, sagemaker.workflow.entities.PipelineVariable]] = None, model_package_group_name: Optional[Union[str, sagemaker.workflow.entities.PipelineVariable]] = None, image_uri: Optional[Union[str, sagemaker.workflow.entities.PipelineVariable]] = None, model_metrics: Optional[sagemaker.model_metrics.ModelMetrics] = None, metadata_properties: Optional[sagemaker.metadata_properties.MetadataProperties] = None, marketplace_cert: bool = False, approval_status: Optional[Union[str, sagemaker.workflow.entities.PipelineVariable]] = None, description: Optional[str] = None, drift_check_baselines: Optional[sagemaker.drift_check_baselines.DriftCheckBaselines] = None, customer_metadata_properties: Optional[Dict[str, Union[str, sagemaker.workflow.entities.PipelineVariable]]] = None, domain: Optional[Union[str, sagemaker.workflow.entities.PipelineVariable]] = None, sample_payload_url: Optional[Union[str, sagemaker.workflow.entities.PipelineVariable]] = None, task: Optional[Union[str, sagemaker.workflow.entities.PipelineVariable]] = None, framework: Optional[Union[str, sagemaker.workflow.entities.PipelineVariable]] = None, framework_version: Optional[Union[str, sagemaker.workflow.entities.PipelineVariable]] = None, nearest_model_name: Optional[Union[str, sagemaker.workflow.entities.PipelineVariable]] = None, data_input_configuration: Optional[Union[str, sagemaker.workflow.entities.PipelineVariable]] = None)

Creates a model package for creating SageMaker models or listing on Marketplace.

Parameters
  • content_types (list) – The supported MIME types for the input data.

  • response_types (list) – The supported MIME types for the output data.

  • inference_instances (list) – A list of the instance types that are used to generate inferences in real-time (default: None).

  • transform_instances (list) – A list of the instance types on which a transformation job can be run or on which an endpoint can be deployed (default: None).

  • model_package_name (str) – Model Package name, exclusive to model_package_group_name, using model_package_name makes the Model Package un-versioned (default: None).

  • model_package_group_name (str) – Model Package Group name, exclusive to model_package_name, using model_package_group_name makes the Model Package versioned (default: None).

  • image_uri (str) – Inference image uri for the container. Model class’ self.image will be used if it is None (default: None).

  • model_metrics (ModelMetrics) – ModelMetrics object (default: None).

  • metadata_properties (MetadataProperties) – MetadataProperties object (default: None).

  • marketplace_cert (bool) – A boolean value indicating if the Model Package is certified for AWS Marketplace (default: False).

  • approval_status (str) – Model Approval Status, values can be “Approved”, “Rejected”, or “PendingManualApproval” (default: “PendingManualApproval”).

  • description (str) – Model Package description (default: None).

  • drift_check_baselines (DriftCheckBaselines) – DriftCheckBaselines object (default: None).

  • customer_metadata_properties (dict[str, str]) – A dictionary of key-value paired metadata properties (default: None).

  • domain (str) – Domain values can be “COMPUTER_VISION”, “NATURAL_LANGUAGE_PROCESSING”, “MACHINE_LEARNING” (default: None).

  • sample_payload_url (str) – The S3 path where the sample payload is stored (default: None).

  • task (str) – Task values which are supported by Inference Recommender are “FILL_MASK”, “IMAGE_CLASSIFICATION”, “OBJECT_DETECTION”, “TEXT_GENERATION”, “IMAGE_SEGMENTATION”, “CLASSIFICATION”, “REGRESSION”, “OTHER” (default: None).

  • framework (str) – Machine learning framework of the model package container image (default: None).

  • framework_version (str) – Framework version of the Model Package Container Image (default: None).

  • nearest_model_name (str) – Name of a pre-trained machine learning benchmarked by Amazon SageMaker Inference Recommender (default: None).

  • data_input_configuration (str) – Input object for the model (default: None).

Returns

A sagemaker.model.ModelPackage instance.

deploy(initial_instance_count=None, instance_type=None, serializer=None, deserializer=None, accelerator_type=None, endpoint_name=None, tags=None, kms_key=None, wait=True, data_capture_config=None, update_endpoint=None, async_inference_config=None, serverless_inference_config=None)

Deploy a Tensorflow Model to a SageMaker Endpoint.

prepare_container_def(instance_type=None, accelerator_type=None, serverless_inference_config=None)

Prepare the container definition.

Parameters
  • instance_type – Instance type of the container.

  • accelerator_type – Accelerator type, if applicable.

  • serverless_inference_config (sagemaker.serverless.ServerlessInferenceConfig) – Specifies configuration related to serverless endpoint. Instance type is not provided in serverless inference. So this is used to find image URIs.

Returns

A container definition for deploying a Model to an Endpoint.

serving_image_uri(region_name, instance_type, accelerator_type=None, serverless_inference_config=None)

Create a URI for the serving image.

Parameters
  • region_name (str) – AWS region where the image is uploaded.

  • instance_type (str) – SageMaker instance type. Used to determine device type (cpu/gpu/family-specific optimized).

  • accelerator_type (str) – The Elastic Inference accelerator type to deploy to the instance for loading and making inferences to the model (default: None). For example, ‘ml.eia1.medium’.

  • serverless_inference_config (sagemaker.serverless.ServerlessInferenceConfig) – Specifies configuration related to serverless endpoint. Instance type is not provided in serverless inference. So this is used to determine device type.

Returns

The appropriate image URI based on the given parameters.

Return type

str

TensorFlow Serving Predictor

class sagemaker.tensorflow.model.TensorFlowPredictor(endpoint_name, sagemaker_session=None, serializer=<sagemaker.serializers.JSONSerializer object>, deserializer=<sagemaker.deserializers.JSONDeserializer object>, model_name=None, model_version=None, **kwargs)

Bases: sagemaker.predictor.Predictor

A Predictor implementation for inference against TensorFlow Serving endpoints.

Initialize a TensorFlowPredictor.

See Predictor for more info about parameters.

Parameters
  • endpoint_name (str) – The name of the endpoint to perform inference on.

  • sagemaker_session (sagemaker.session.Session) – Session object which manages interactions with Amazon SageMaker APIs and any other AWS services needed. If not specified, the estimator creates one using the default AWS configuration chain.

  • serializer (callable) – Optional. Default serializes input data to json. Handles dicts, lists, and numpy arrays.

  • deserializer (callable) – Optional. Default parses the response using json.load(...).

  • model_name (str) – Optional. The name of the SavedModel model that should handle the request. If not specified, the endpoint’s default model will handle the request.

  • model_version (str) – Optional. The version of the SavedModel model that should handle the request. If not specified, the latest version of the model will be used.

classify(data)

Placeholder docstring.

regress(data)

Placeholder docstring.

predict(data, initial_args=None)

Placeholder docstring.