Anton Dmitriev created IGNITE-10234:
---------------------------------------
Summary: ML: Create a skeleton for model inference in Apache Ignite
Key: IGNITE-10234
URL: https://issues.apache.org/jira/browse/IGNITE-10234
Project: Ignite
Issue Type: Sub-task
Components: ml
Affects Versions: 2.8
Reporter: Anton Dmitriev
Assignee: Anton Dmitriev
Fix For: 2.8
To support model inference in Apache Ignite for our models as well as for
loaded foreign models we need a common inference workflow.
This workflow should isolate model _inference/using_ from model
_training/saving_. User should be able to:
* *Load/Unload any possible model (that can be represented as a function).*
_This part assumes that user specifies any underlying model, a bridge that
allows to interact with it and a signature of the model (accepted parameters,
returned value)._
* *Access a list of loaded models.*
_Used should be able to access a list of loaded models, models that can be used
for inference without any additional manipulations. In terms of future Cloud
part it will be a table of models user can use in Web UI and start using it._
* *Start/Stop distributed infrastructure for inference utilizing cluster
resources.*
_Single inference is actually a single function call. We want to utilize all
cluster resources, so we need to replicate the model and start services that
are ready to use the model for inference on every node._
* *Perform inference on top of started infrastructure.*
_There are should be a gateway that allows to use a single entry point to
perform inference utilizing started on the previous step distributed service
infrastructure._
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
