def main(argv=None): parser = create_parser() args = parser.parse_args(argv) logging.getLogger().setLevel(logging.INFO) client = _utils.get_sagemaker_client(args.region, args.endpoint_url, assume_role_arn=args.assume_role) endpoint_name = None old_endpoint_config_name = None if (args.update_endpoint and _utils.endpoint_name_exists(client, args.endpoint_name)): ## Get the old endpoint config to cleanup later old_endpoint_config_name = _utils.get_endpoint_config(client, args.endpoint_name) logging.info('Submitting Update Endpoint request to SageMaker...') endpoint_name = _utils.update_deployed_model(client, vars(args)) else: logging.info('Submitting Create Endpoint request to SageMaker...') endpoint_name = _utils.deploy_model(client, vars(args)) logging.info('Endpoint creation/update request submitted. Waiting for completion...') _utils.wait_for_endpoint_creation(client, endpoint_name) ## If updating existing endpoint, cleanup old endpoint config if old_endpoint_config_name: logging.info("Deleting old endpoint config: " + old_endpoint_config_name) if _utils.delete_endpoint_config(client, old_endpoint_config_name): logging.info("Deleted old endpoint config: " + old_endpoint_config_name) else: logging.info("Unable to delete old endpoint config: " + old_endpoint_config_name) _utils.write_output(args.endpoint_name_output_path, endpoint_name) logging.info('Endpoint creation/update completed.')
def main(argv=None): parser = create_parser() args = parser.parse_args(argv) logging.getLogger().setLevel(logging.INFO) client = _utils.get_sagemaker_client(args.region) logging.info( 'Submitting HyperParameter Tuning Job request to SageMaker...') hpo_job_name = _utils.create_hyperparameter_tuning_job(client, vars(args)) logging.info( 'HyperParameter Tuning Job request submitted. Waiting for completion...' ) _utils.wait_for_hyperparameter_training_job(client, hpo_job_name) best_job, best_hyperparameters = _utils.get_best_training_job_and_hyperparameters( client, hpo_job_name) model_artifact_url = _utils.get_model_artifacts_from_job(client, best_job) image = _utils.get_image_from_job(client, best_job) logging.info('HyperParameter Tuning Job completed.') _utils.write_output(args.hpo_job_name_output_path, hpo_job_name) _utils.write_output(args.model_artifact_url_output_path, model_artifact_url) _utils.write_output(args.best_job_name_output_path, best_job) _utils.write_output(args.best_hyperparameters_output_path, best_hyperparameters, json_encode=True) _utils.write_output(args.training_image_output_path, image)
def test_write_output_string(self): with patch("common._utils.Path", MagicMock()) as mock_path: _utils.write_output("/tmp/output-path", "output-value") mock_path.assert_called_with("/tmp/output-path") mock_path("/tmp/output-path").parent.mkdir.assert_called() mock_path("/tmp/output-path").write_text.assert_called_with("output-value")
def main(argv=None): parser = create_parser() args = parser.parse_args(argv) logging.getLogger().setLevel(logging.INFO) client = _utils.get_sagemaker_client(args.region, args.endpoint_url, assume_role_arn=args.assume_role) logging.info('Submitting Batch Transformation request to SageMaker...') batch_job_name = _utils.create_transform_job(client, vars(args)) def signal_term_handler(signalNumber, frame): _utils.stop_transform_job(client, batch_job_name) logging.info( f"Transform job: {batch_job_name} request submitted to Stop") signal.signal(signal.SIGTERM, signal_term_handler) logging.info('Batch Job request submitted. Waiting for completion...') try: _utils.wait_for_transform_job(client, batch_job_name) except: raise finally: cw_client = _utils.get_cloudwatch_client( args.region, assume_role_arn=args.assume_role) _utils.print_logs_for_job(cw_client, '/aws/sagemaker/TransformJobs', batch_job_name) _utils.write_output(args.output_location_output_path, args.output_location) logging.info('Batch Transformation creation completed.')
def main(argv=None): parser = create_parser() args = parser.parse_args(argv) logging.getLogger().setLevel(logging.INFO) client = _utils.get_sagemaker_client(args.region, args.endpoint_url) logging.info('Submitting Processing Job to SageMaker...') job_name = _utils.create_processing_job(client, vars(args)) logging.info('Job request submitted. Waiting for completion...') try: _utils.wait_for_processing_job(client, job_name) except: raise finally: cw_client = _utils.get_cloudwatch_client(args.region) _utils.print_logs_for_job(cw_client, '/aws/sagemaker/ProcessingJobs', job_name) outputs = _utils.get_processing_job_outputs(client, job_name) _utils.write_output(args.job_name_output_path, job_name) _utils.write_output(args.output_artifacts_output_path, outputs, json_encode=True) logging.info('Job completed.')
def main(argv=None): parser = create_parser() args = parser.parse_args(argv) logging.getLogger().setLevel(logging.INFO) client = _utils.get_sagemaker_client(args.region, args.endpoint_url) logging.info('Submitting a create workteam request to SageMaker...') workteam_arn = _utils.create_workteam(client, vars(args)) logging.info('Workteam created.') _utils.write_output(args.workteam_arn_output_path, workteam_arn)
def main(argv=None): parser = create_parser() args = parser.parse_args(argv) logging.getLogger().setLevel(logging.INFO) client = _utils.get_sagemaker_client(args.region, args.endpoint_url) logging.info('Submitting model creation request to SageMaker...') _utils.create_model(client, vars(args)) logging.info('Model creation completed.') _utils.write_output(args.model_name_output_path, args.model_name)
def main(argv=None): parser = create_parser() args = parser.parse_args(argv) logging.getLogger().setLevel(logging.INFO) client = _utils.get_sagemaker_client(args.region, args.endpoint_url) logging.info('Submitting Endpoint request to SageMaker...') endpoint_name = _utils.deploy_model(client, vars(args)) logging.info( 'Endpoint creation request submitted. Waiting for completion...') _utils.wait_for_endpoint_creation(client, endpoint_name) _utils.write_output(args.endpoint_name_output_path, endpoint_name) logging.info('Endpoint creation completed.')
def main(argv=None): parser = create_parser() args = parser.parse_args(argv) logging.getLogger().setLevel(logging.INFO) client = _utils.get_sagemaker_client(args.region, args.endpoint_url) logging.info('Submitting Ground Truth Job request to SageMaker...') _utils.create_labeling_job(client, vars(args)) logging.info( 'Ground Truth labeling job request submitted. Waiting for completion...' ) _utils.wait_for_labeling_job(client, args.job_name) output_manifest, active_learning_model_arn = _utils.get_labeling_job_outputs( client, args.job_name, args.enable_auto_labeling) logging.info('Ground Truth Labeling Job completed.') _utils.write_output(args.output_manifest_location_output_path, output_manifest) _utils.write_output(args.active_learning_model_arn_output_path, active_learning_model_arn)
def main(argv=None): parser = create_parser() args = parser.parse_args(argv) logging.getLogger().setLevel(logging.INFO) client = _utils.get_sagemaker_client(args.region, args.endpoint_url) logging.info('Submitting Training Job to SageMaker...') job_name = _utils.create_training_job(client, vars(args)) def signal_term_handler(signalNumber, frame): _utils.stop_training_job(client, job_name) logging.info(f"Training Job: {job_name} request submitted to Stop") signal.signal(signal.SIGTERM, signal_term_handler) logging.info('Job request submitted. Waiting for completion...') try: _utils.wait_for_training_job(client, job_name) _utils.wait_for_debug_rules(client, job_name) except: raise finally: cw_client = _utils.get_cloudwatch_client(args.region) _utils.print_logs_for_job(cw_client, '/aws/sagemaker/TrainingJobs', job_name) image = _utils.get_image_from_job(client, job_name) model_artifact_url = _utils.get_model_artifacts_from_job(client, job_name) logging.info('Get model artifacts %s from training job %s.', model_artifact_url, job_name) _utils.write_output(args.model_artifact_url_output_path, model_artifact_url) _utils.write_output(args.job_name_output_path, job_name) _utils.write_output(args.training_image_output_path, image) logging.info('Job completed.')
def main(argv=None): parser = create_parser() args = parser.parse_args(argv) logging.getLogger().setLevel(logging.INFO) client = _utils.get_sagemaker_client(args.region, args.endpoint_url) logging.info('Submitting Batch Transformation request to SageMaker...') batch_job_name = _utils.create_transform_job(client, vars(args)) logging.info('Batch Job request submitted. Waiting for completion...') try: _utils.wait_for_transform_job(client, batch_job_name) except: raise finally: cw_client = _utils.get_cloudwatch_client(args.region) _utils.print_logs_for_job(cw_client, '/aws/sagemaker/TransformJobs', batch_job_name) _utils.write_output(args.output_location_output_path, args.output_location) logging.info('Batch Transformation creation completed.')
def main(argv=None): parser = create_parser() args = parser.parse_args(argv) logging.getLogger().setLevel(logging.INFO) client = _utils.get_sagemaker_client(args.region, args.endpoint_url, assume_role_arn=args.assume_role) logging.info('Submitting Processing Job to SageMaker...') job_name = _utils.create_processing_job(client, vars(args)) def signal_term_handler(signalNumber, frame): logging.info(f"Stopping Processing Job: {job_name}") _utils.stop_processing_job(client, job_name) logging.info(f"Processing Job: {job_name} request submitted to Stop") signal.signal(signal.SIGTERM, signal_term_handler) logging.info('Job request submitted. Waiting for completion...') try: _utils.wait_for_processing_job(client, job_name) except: raise finally: cw_client = _utils.get_cloudwatch_client( args.region, assume_role_arn=args.assume_role) _utils.print_logs_for_job(cw_client, '/aws/sagemaker/ProcessingJobs', job_name) outputs = _utils.get_processing_job_outputs(client, job_name) _utils.write_output(args.job_name_output_path, job_name) _utils.write_output(args.output_artifacts_output_path, outputs, json_encode=True) logging.info('Job completed.')
def main(argv=None): parser = create_parser() args = parser.parse_args(argv) logging.getLogger().setLevel(logging.INFO) client = _utils.get_sagemaker_client(args.region, args.endpoint_url, assume_role_arn=args.assume_role) logging.info('Submitting Ground Truth Job request to SageMaker...') _utils.create_labeling_job(client, vars(args)) def signal_term_handler(signalNumber, frame): _utils.stop_labeling_job(client, args.job_name) logging.info(f"Ground Truth labeling job: {args.job_name} request submitted to Stop") signal.signal(signal.SIGTERM, signal_term_handler) logging.info('Ground Truth labeling job request submitted. Waiting for completion...') _utils.wait_for_labeling_job(client, args.job_name) output_manifest, active_learning_model_arn = _utils.get_labeling_job_outputs(client, args.job_name, args.enable_auto_labeling) logging.info('Ground Truth Labeling Job completed.') _utils.write_output(args.output_manifest_location_output_path, output_manifest) _utils.write_output(args.active_learning_model_arn_output_path, active_learning_model_arn)
def main(argv=None): parser = create_parser() args = parser.parse_args(argv) logging.getLogger().setLevel(logging.INFO) client = _utils.get_sagemaker_client(args.region, args.endpoint_url, assume_role_arn=args.assume_role) logging.info( 'Submitting HyperParameter Tuning Job request to SageMaker...') hpo_job_name = _utils.create_hyperparameter_tuning_job(client, vars(args)) def signal_term_handler(signalNumber, frame): _utils.stop_hyperparameter_tuning_job(client, hpo_job_name) logging.info( f"HyperParameter Tuning Job: {hpo_job_name} request submitted to Stop" ) signal.signal(signal.SIGTERM, signal_term_handler) logging.info( 'HyperParameter Tuning Job request submitted. Waiting for completion...' ) _utils.wait_for_hyperparameter_training_job(client, hpo_job_name) best_job, best_hyperparameters = _utils.get_best_training_job_and_hyperparameters( client, hpo_job_name) model_artifact_url = _utils.get_model_artifacts_from_job(client, best_job) image = _utils.get_image_from_job(client, best_job) logging.info('HyperParameter Tuning Job completed.') _utils.write_output(args.hpo_job_name_output_path, hpo_job_name) _utils.write_output(args.model_artifact_url_output_path, model_artifact_url) _utils.write_output(args.best_job_name_output_path, best_job) _utils.write_output(args.best_hyperparameters_output_path, best_hyperparameters, json_encode=True) _utils.write_output(args.training_image_output_path, image)