aws / aws/aws-step-functions-data-science-sdk-python

ModelDataUrl is incorrect when result_path is defined in TrainingStep

Open
#104 0 comments 0 reactions 0 assignees View on GitHub
Dominant language
Python
Stars
299
Forks
84
PR merge metrics
No merged PRs in 30d

Description

The ModelDataUrl in the Step Functions' defenition is '“ModelDataUrl.$“: “$[‘ModelArtifacts’][‘S3ModelArtifacts’]”' even though the result_path is defined in the TrainingStep, and it causes error like this.

> “An error occurred while executing the state ‘Save Model’ (entered at the event id #8). The JSONPath ’$[‘ModelArtifacts’][‘S3ModelArtifacts’]' specified for the field ‘ModelDataUrl.$’ could not be found in the input ...

It should be '“ModelDataUrl.$“: “$[‘TrainingStepResults’][‘ModelArtifacts’][‘S3ModelArtifacts’]”'.

```
training_step = steps.TrainingStep(
‘Model Training’,
estimator=tf_estimator,
data={
‘training’: inputs
},
job_name=execution_input[‘TrainingJobName’],
result_path=‘$.TrainingStepResults’,
wait_for_completion=True
)

model_step = steps.ModelStep(
‘Save Model’,
model=training_step.get_expected_model(),
model_name=execution_input[‘ModelName’],
instance_type=‘ml.m5.xlarge’,
result_path=‘$.ModelStepResults’
)
```

Contributor guide

Open the contributing guide

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.