aws / aws/amazon-sagemaker-examples
[Bug Report] `KeyError` in SgaeMakerPipelines local-mode notebook
- Dominant language
- Jupyter Notebook
- Stars
- 11k
- Forks
- 7k
- Avg merge
- 8h 29m
- Merged PRs (30d)
- 8
Description
**Link to the notebook**
Small bug in [[local-mode-notebook]](https://github.com/aws/amazon-sagemaker-examples/tree/main/sagemaker-pipelines/tabular/local-mode) at
`sagemaker-pipelines/tabular/local-mode/sagemaker-pipelines-local-mode.ipynb`
**Describe the bug**
The the following cell should use index [-2] instead of [-1] for accessing the step outputs returned from local pipeline execution.
```
# Get output of ModelStep
import json
model_name = steps["PipelineExecutionSteps"][-1]["Metadata"]["Model"]["Arn"]
outputs = local_pipeline_session.sagemaker_client.describe_model(ModelName=model_name)
```
**To reproduce**
- Running the cell with [-2] correctly accesses the dict. in reference to the ListSteps response.
- The currently annotated [-1] gives a `KeyError` on "Model" as that references the last step in the list, a transform job, which has a `TransformJob` as it's key, instead of `Model`. For referenece, the generated step output is [[here]](https://paste.amazon.com/show/svia/1698698004)
Contributor guide
Assessment
This issue has not been assessed yet.