Closed DanielTatarkin closed 4 years ago
Manually updating SageMaker SDK helped solve the issue.
!pip install sagemaker --upgrade
Hello, yes sorry about that issue -- what is happening is SageMaker notebook instances have an older version of the SageMaker Python SDK with that bug, which has been fixed in the latest version. I have informed the SageMaker service team, they are working on getting the new version into notebook instances.
Manually updating SageMaker SDK helped solve the issue.
!pip install sagemaker --upgrade
Can you confirm this fixed this for you? I have tried doing this and it's not working for me ... I upgraded boto3 too but to no avail.
Manually updating SageMaker SDK helped solve the issue.
!pip install sagemaker --upgrade
Can you confirm this fixed this for you? I have tried doing this and it's not working for me ... I upgraded boto3 too but to no avail.
Nevermind my comment - restarted the notebook and upgraded at the beginning and it is working indeed. Thanks! :-)
When going through Introduction to Amazon SageMaker module Sentiment Analysis, I'm getting:
TypeError: type object got multiple values for keyword argument 'enable_network_isolation'
This happens in cell 13 when creating a transformer:
transformer = estimator.transformer(instance_count=1, instance_type='ml.c5.xlarge')
Similar problem outlined here: https://forums.aws.amazon.com/thread.jspa?threadID=321194
I tried using different Kernels, creating a brand new Notebook instance and restarting it. Other modules have been working without issues. Anything that could be suggested?
Thanks, Daniel Tatarkin
Screenshot (Same error using conda_tensorflow_p36 default kernel, but this screenshot was taken when I tried conda_amazonei_tensorflow_p36)