aws / aws/sagemaker-python-sdk

Receiving configuration error when loading SageMaker Session

Open
#4,403 2 comments 2 reactions 1 assignee Claimed by @mollyheamazon View on GitHub
component: pysdk-team type: bug
Dominant language
Python
Stars
2.3k
Forks
1.3k
Avg merge
2d 2h
Merged PRs (30d)
32

Description

**Describe the bug**
In SageMaker Studio within a SageMaker Notebook instance, we use the SageMaker Python SDK. When using the sagemaker library, we are receiving errors in the SageMaker session and other methods.

It looks like the session tries to read a default config file that doesn't exists.

Not sure how to mitigate the error, and prevent SageMaker to look at default config file.

**To reproduce**
We execute:
`from sagemaker.session import Session`

And receive the following message:
```
sagemaker.config INFO - Not applying SDK defaults from location: /etc/xdg/sagemaker/config.yaml
Unable to create default JumpStart SageMaker Session due to the following error: None is not of type 'string'

Failed validating 'type' in schema['properties']['SageMaker']['properties']['FeatureGroup']['properties']['Tags']['items']['properties']['Value']:
{'maxLength': 256,
'minLength': 0,
'pattern': '^[\\w\\s\\d_.:/=+\\-@]*$',
'type': 'string'}

On instance['SageMaker']['FeatureGroup']['Tags'][0]['Value']:
None.
```

Then later we execute:
`sagemaker_session = Session()`

And receive this error:
```
File /opt/conda/lib/python3.8/site-packages/jsonschema/validators.py:1308, in validate(instance, schema, cls, *args, **kwargs)
1306 error = exceptions.best_match(validator.iter_errors(instance))
1307 if error is not None:
-> 1308 raise error

ValidationError: None is not of type 'string'

Failed validating 'type' in schema['properties']['SageMaker']['properties']['FeatureGroup']['properties']['Tags']['items']['properties']['Value']:
{'maxLength': 256,
'minLength': 0,
'pattern': '^[\\w\\s\\d_.:/=+\\-@]*$',
'type': 'string'}

On instance['SageMaker']['FeatureGroup']['Tags'][0]['Value']:
None
```

**System information**
Using a SageMaker ml.t3.medium instance with a Data Science 3.0 kernel on SageMaker Studio within a SageMaker Notebook Instance. Using sagemaker version: 2.206.0

**Additional context**
Add any other context about the problem here.

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.