Toolkit for allowing inference and serving with MXNet in SageMaker. Dockerfiles used for building SageMaker MXNet Containers are at https://github.com/aws/deep-learning-containers.
Describe the feature you'd like
As an AWS service, we may call other AWS services from SageMaker endpoint. Pre-installing boto3 in docker image is helpful. Currently we cannot use it by default in 763104351884.dkr.ecr.us-west-2.amazonaws.com/mxnet-inference:1.6.0-gpu-py3
ModelError: An error occurred (ModelError) when calling the InvokeEndpoint operation: Received server error (500) from model with message "Object of type 'ModuleNotFoundError' is not JSON serializable
Traceback (most recent call last):
File "/opt/ml/model/code/deploy.py", line 24, in transform_fn
import boto3
ModuleNotFoundError: No module named 'boto3'
How would this feature be used? Please describe.
Downloading S3 file, connecting DynamoDB, etc. via boto3.
Describe the feature you'd like As an AWS service, we may call other AWS services from SageMaker endpoint. Pre-installing boto3 in docker image is helpful. Currently we cannot use it by default in
763104351884.dkr.ecr.us-west-2.amazonaws.com/mxnet-inference:1.6.0-gpu-py3How would this feature be used? Please describe. Downloading S3 file, connecting DynamoDB, etc. via boto3.
Describe alternatives you've considered N/A
Additional context N/A