A cost effective way of deploying models on AWS is to use 'zappa' (autotemplating for lambda + API gateway). This just works, and costs my team less for a year

A cost effective way of deploying models on AWS is to use 'zappa' (autotemplating for lambda + API gateway). This just works, and costs my team less for a year than one week of Sagemaker or kubernetes architectures. https://www.corbettanalytics.com/deploy-machine-learning-model-aws-lambda-zappa/amp/ https://www.corbettanalytics.com/deploy-machine-learning-mod...