ServicesShow__hero__img

Creating API calls to get predictions endpoint set up

Getting prediction of typical model on physical server can be cumbersome, with the help of serverless and api gateway, we could cut down the need of having manual server spinup host the model .

Business Challenges

Supposed we want to get predictor to set up, we would need container to host the applications.

Solution

  • API to get the model listed in AWS catalog hosted in sage-maker

Deliverables

  • Create API gateway that would then tell lambda to perform a cli call to get sage-maker model
  • host them in cluster/lambda functions

Expected
Outcome

  • Models can be served in cloud marketplace and sell as model as a service in sage-maker marketplace

Have a question?

Drop us a line and we will get back to you