Hi,
I’m just wondering if the ‘Weak Supervision’ (Labs) is still actively supported/maintained. I reported a nasty bug in the ‘Weak Label Model Predictor’ node in 1/2020 (see: Weak Label Model Predictor - 'Execute failed: The probabilities do not sum up to 1. Consider setting a proper epsilon.' - #2 by nemad) which is still not fixed.
Thx!
Erich
Hi Erich,
Weak Supervision is currently not far up on our priority list. That being said, we will still have a look at issues and fix them if they are reasonably easy to fix. Unfortunately, the issue you reported is anything but easy, so it’s very unlikely that we will get around to fixing it in any of the upcoming releases.
Kind regards,
Adrian
Thx Adrian for letting me know!
Best
Erich
This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.