AWS Sagemaker - PCA Model 未部署

Question

I've created a PCA Model in SageMaker..something along those lines:

pca = PCA(role=role,
             train_instance_count=1,
             train_instance_type='ml.c4.xlarge',
             output_path=output_path, 
             num_components=N_COMPONENTS, 
             sagemaker_session=session)

Then I've fitted it against my data

pca.fit(data)

And finally I would like to deploy the model, but when I run:

pca_predictor = pca.deploy(initial_instance_count=1, 
                              instance_type='ml.t2.medium')

I see this error message:

TypeError: create_model() got an unexpected keyword argument 'model_kms_key'

I have no idea how to understand that error and what it refers to...any pointers?

Answer 1

Upgrade sagemaker, I was getting this error with version 1.42.6. It's fixed in 1.42.9

Run the command:

    ! pip install --upgrade sagemaker

If using a jupyter notebook instance you may need to restart the kernel to pick up the change.

To check the current sagemaker version use:

    print(sagemaker.__version__)