How to deploy naive Bayes model on AWS Sagemaker

Question

I have a naive Bayes model which performs sentiment analysis of movie reviews. I want to deploy this naive Bayes model on AWS Sagemaker. Naive Bayes isn't one of the built in algorithms on Sagemaker.

Can anyone guide me with the steps ? I already have the model built and I'm just left with deploying it on Sagemaker so that I can build a web app using its endpoint instance

Answer 1

If it's based on Sklearn/Numpy/Scipy pandas or if you can install it by pip I recommend you use the SageMaker Sklearn container ; you can follow an example here with a Sklearn random forest and here is the doc explaining you how to format your code and add custom pip dependencies with a requirements.txt