Adam Pocock

Given that we don't expose the same set of symbols as the pywrap file does, won't renaming it cause more trouble in the long run when people wonder why there...

You can load in tensorflow-text's native library using https://github.com/tensorflow/java/blob/master/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/TensorFlow.java#L99. I don't recall if it works in the current version on `main`, but we have a test for it in the...

Yeah, that's the right approach, but you're hitting the issue where we don't compile in C++ 11 mode and base TensorFlow does now. If you try and compile the `bazelcism`...

We're releasing TF-Java 1.0.0-rc1 based on TF 2.16.1 this week. It has a test for loading in tensorflow-text ops as part of our CI now so it should work.

1.0.0-rc1 is out on Maven Central - https://central.sonatype.com/artifact/org.tensorflow/tensorflow-core-native

I think we should clone that repo and merge it in. We want the ndarray docs to be in the main docs as otherwise a bunch of the Javadoc will...

Our CI is Github actions which don't provide aarch64-linux as a hosted runner, so we can't easily make binaries for it.

`session.restore(String path)` should do that already - https://github.com/tensorflow/java/blob/master/tensorflow-core/tensorflow-core-api/src/main/java/org/tensorflow/Session.java#L709.

The graph should be defined somewhere as a protobuf, but you can also specify the model yourself. How are you loading in the model architecture at the moment?

Variable checkpoints are stored separately from model structure in TensorFlow, so you need both components to load in a model. The saved model format is a directory containing the model...