Karl Lessard

Ok, it looks like the latency is not happening in TF Java but in TensorFlow itself. I've just done a very quick search and found a few latency issues observed...

@JimClarke5 , we've been discussing Ryan and I about this and would like to revive this proposal, whenever you're ready

Storing your session as a private field of a Spring Bean shouldn't cause any trouble, you'll be able to reuse it for all predictions. For the eager session, since you...

> During performance testing, I monitored the heap memory and found no significant issues. One idea: If you could monitor the native memory (i.e. the memory used by the JVM...

Also @icrecescu in your eager operations (post-processing I think), do you retrieve values of the tensor via `output.asTensor()`? If so, you might want to release these tensors explicitly as well,...

Hi, checking again if we can get some help, since updating the documentation on the website is pretty much the last thing remaining for us before publishing TF Java 1.0.0...

Hi @MarkDaoust and @markmcd , any update on this please? It is a bit frustrating that we can't update the documentation on the website and some of our users think...

Hey @mihaimaruseac , I've marked the PR as non-draft, any chance we can solve this soon? Thank you

Hi @hanfengatonline , it looks like you are still using TensorFlow 1.x and an older version of TF Java. This version is no longer supported, please take a look at...

I also recall trying to enable this experimental API in the past and we were facing compilation issues, maybe it has been cleaned up? But if it now works, that's...