DummyTraining

Training step that uses a pretrained model

SageMaker step type	Processing
Node type	internal (consumes upstream, produces downstream)
Container entry point	dummy_training.py
Interface file	steps/interfaces/dummy_training.step.yaml

Compute

Compute kind	framework
SDK class	SKLearn (SageMaker DLC via image_uris.retrieve)

Functionality

Dummy training step with flexible input modes. Adds hyperparameters.json to model.tar.gz for downstream packaging. Supports INTERNAL mode (accepts inputs) or SOURCE mode (reads from source directory).

Inputs (dependencies)

Input	Type	Required	Compatible producers
hyperparameters_s3_uri	hyperparameters	no	HyperparameterPrep, ProcessingStep
model_artifacts_input	model_artifacts	no	PyTorchTraining, XGBoostTraining, LightGBMTraining, ProcessingStep

Outputs

Output	Type
model_output	model_artifacts

Consumers (downstream steps)

Steps that declare this step as a compatible input source:

• LightGBMMTModelEval

• LightGBMMTModelInference

• LightGBMModelEval

• LightGBMModelInference

• Payload

• PyTorchModelEval

• PyTorchModelInference

• TSAModelEval

• XGBoostModelEval

• XGBoostModelInference

• XgboostMtModelEval

Framework requirements

Package	Version
python	>=3.7

---

← Back to the Step Catalog