Open pansk opened 8 years ago
(Comes from https://github.com/NVIDIA/DIGITS/pull/728)
Good point. I never got around to setting the backend for Generic datasets because LMDB is the only backend supported. And I'm actually finding it convenient at the moment because it gives me a way to distinguish Classification datasets from Generic ones. But you're right - we should display them as LMDB and find some other way to distinguish between dataset types.
@jmancewicz you were working on adding data extensions to the home page for sorting, right? Any update?
@lukeyeager, I just created a PR for it #1246
I have one classification DB and several DBs of type "Other". The backend column reports "lmdb" for the classification DB, but is empty for all the other DBs.
I also tried to create a new one (all those DBs were created before upgrading), and even the new one don't report anything as backend.