Adam Pocock

Our `tf.math.mean` operation can be configured to do `reduce_mean`, depending on how keep_dims and the other arguments are set. There is a euclidean (L2) norm built in, but L1 will...

In general we only expose the ops which exist in the TensorFlow C API. If an op exists in Python but not in the C API then it means the...

If we're doing this mainly for pluggable device support then I think the only extant device is the one Apple released for macOS, so it doesn't matter too much if...

I think #397 was fixed by the name import logic that was added in one of @rnett's PRs.

Looks like the protobuf warnings should go away if we use the latest protobuf release (see https://github.com/protocolbuffers/protobuf/issues/3781).

We might be able to call the appropriate method in the guts of the C++ library to turn off the logger, but really the C++ library should be refactored a...

The IllegalReflectiveAccess warning is because the protobuf jar is messing with fields it shouldn't be. Its emitted on Java 9 and newer. Updating to a newer version will probably fix...

Looks like it's fixed by using protobuf-java 3.7.0 or newer (https://github.com/protocolbuffers/protobuf/releases/tag/v3.7.0). The 1.15.0 proto artifact depends on protobuf-java 3.5.1. @karllessard, @sjamesr, where does the org.tensorflow:proto artifact come from? Is it...

We spent some time looking into exporting TF logging messages into Java so they can be logged using a Java logging framework, but unfortunately that effort stalled as the TF...

We had a brief look into this several months ago, and it's basically a huge pile of Python (https://github.com/tensorflow/tensorflow/tree/master/tensorflow/python/distribute) we'd need to replicate into Java. Doing out-of-band distribution using MPI...