After exporting your model to a Google Cloud Storage bucket you can use your exported model to make predictions in a Docker image. You may refer to this documentation on how to deploy to a container.
I just didn’t understand that I need to select the “edge” option to have the “export model” available.
Thank you for the useful documentation .
As a additionnal question : can I know the version of Tensorflow used for these export model? (I have some incompatibiliy to use these in my software).