Rob Rua

Yeah this was really unfortunate. I've been thinking about what cass' behavior in this sort of situation should be and while I like the idea of having a fallback, I'm...

Hey, thanks for the PR & sorry for the slow response. Unfortunately, for this to work it would also need to be supported in the DataSources (e.g. https://github.com/meraki-analytics/orianna/blob/master/orianna/src/main/java/com/merakianalytics/orianna/datapipeline/riotapi/LeagueAPI.java), which currently...

First thing is I want to figure out if this is a version compat issue or a general bug: Is this a problem with TF 1.9 specifically? Has it worked...

If you can leave some more specifics about how you're obtaining, saving, and loading the model that'd be helpful in figuring out what's wrong

Hey, thanks for the research and the detailed issue. For (1) that's an excellent idea to add here and I'll look into allowing dynamic max sequence length on both the...

Reopening this to remind myself to add this in the future. On (2): Right then, I hadn't read it closely enough and failed to notice you were tracking the start...

Hey, Are you using the token embeddings or the sequence embeddings in this case? In my experience, the BERT sequence embeddings in particular (but sometimes also the token embeddings) don't...

Hey, the Large models currently aren't provided because they exceed the artifact size maximum for Sonatype's Maven Repository (1GB), so I can't distribute them through there. If you can find...

This sounds good to me. I'll add this for both Python and Java next time I do some work on this project, or feel free to send a PR.