Deploy Model

Init Model

Note: TensorFlow only supports numeric versions.

In addition, the version you define here must be LARGER than any other version known to the TensorFlow Serving instance to which you are deploying.



In [ ]:

    
%%bash 

pio init-model \
  --model-server-url http://prediction-tensorflow.community.pipeline.io/ \
  --model-type tensorflow \
  --model-namespace default \
  --model-name tensorflow_linear \
  --model-version 1 \
  --model-path .

Deploy Model (CLI)



In [ ]:

    
%%bash

pio deploy

TODO: Deploy Model (REST)



In [ ]: