Adding support for Multiple endpoints

bentoml / aws-sagemaker-deploy

Fast model deployment on AWS Sagemaker

Apache License 2.0

15 stars 15 forks source link

Adding support for Multiple endpoints #20

Closed jjmachan closed 3 years ago

jjmachan commented 3 years ago

This PR uses a HttpAPI proxy + Lambda function to call the Sagemaker endpoint. This allows us to add support for multiple endpoints and better integration for the different input handlers and better error messages.

jjmachan commented 3 years ago

/test-e2e-deploy sha=d254f3b

jjmachan commented 3 years ago

/test-e2e-deploy sha=b80db2b

jjmachan commented 3 years ago

/test-e2e-deploy sha=0b26b61

jjmachan commented 3 years ago

/test-e2e-deploy sha=bec16e8

jjmachan commented 3 years ago

also fixes #18

jjmachan commented 3 years ago

/test-e2e-deploy sha=7e39966

jjmachan commented 3 years ago

Added a timeout option to lambda function too. So the timeout option in the config will provide the timeout for lambda as well as the Gunicorn_worker. The other option was to separate out both timeouts into 2 different ones but I'm not sure there is a use-case for that?

jjmachan commented 3 years ago

/test-e2e-deploy sha=a92cc1b