I am trying to deploy a SageMaker model using a serverless endpoint. The model is stored in ECR, and I created the model and endpoint configuration successfully. However, when I try to deploy the serverless endpoint, I encounter the following error in CloudWatch:

I got an email about getting blocked in 30 days due to a violation, but it wasn’t me! I don’t see any contact info in the email to disput it so I am posting it here.

Sagemaker model hosting uses default_input_fn for input transformations. However, in this function I need access to the output of default_model_fn, which is only passed to default_predict_fn.

Sagemaker model hosting uses default_input_fn for input transformations. However, in this function I need access to the output of default_model_fn, which is only passed to default_predict_fn.

I have a trained model in .joblib format and prior to dumping it to an S3 folder, I’m compressing the file using python’s tarfile library. Thereafter, I want to create a SageMaker endpoint using that given file. However, I get errors

I need a solution that can stop my notebook instance when it is idle or not in use

I’ve created a sagemaker pipeline with a register model step show below. The model_metrics is produced by an eval step. It’s based of the sagemaker tutorials but all the examples I can find use an SKLearn estimator, my model is pytorch based (it’s a sentence-transformers model evaluated using their information retrieval evaluator).