The Teradata Package for Python (teradataml), which is Teradata's Python package for client-side scripts, is extended with support for Amazon SageMaker. This support is included in the tdapiclient library.
Many of the Amazon SageMaker APIs are callable through tdapiclient. Specifically, tdapiclient integration with Amazon SageMaker works with their estimator class and derived classes.
The Amazon SageMaker Python SDK provides the following interfaces SageMaker APIs.
| Amazon SageMaker Interface | Supported through tdapiclient? |
|---|---|
| Estimators | Supported, through the same API as the Estimators class. You provide a teradataml DataFrame for training through the fit Method. Supported estimators: "sagemaker.mxnet.estimator", "sagemaker.sklearn.estimator", "sagemaker.chainer", "sagemaker.huggingface", "sagemaker.pytorch", "sagemaker.rl.estimator", "sagemaker.tensorflow", "sagemaker.estimator", "sagemaker.xgboost.estimator" |
| Predictors | Supported |
| Model | Supported for Teradata BYOM use case. You can also specify model path. |
The teradataml SageMaker extension library (tdapiclient) includes the following functions and interfaces.