aws / sagemaker-inference-toolkit

Serve machine learning models within a 🐳 Docker container using 🧠 Amazon SageMaker.

Apache License 2.0

379 stars 82 forks source link

Error handling 20191015 #13

Closed ericangelokim closed 4 years ago

ericangelokim commented 4 years ago

Issue #, if available:

Description of changes:

Add error handling to respond appropriately to client
User module is optional

Test

python -m tox python3 -m tox

By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.

ericangelokim commented 4 years ago

Side note: Why does this repository support Python 2 (tox)? We should be deprecating in support of py3 only.

ajaykarpur commented 4 years ago

Side note: Why does this repository support Python 2 (tox)? We should be deprecating in support of py3 only.

We plan to deprecate Python 2 support after Python 2.7 reaches end of life.

ericangelokim commented 4 years ago

I thought there should be a distinction between errors for the server, and error handling that is code used to respond back to the client. However, looking at the code in errors it looks as if that is where the response back to client should be. I can move it and re-push the commit to remote.