Training form the rasa-x UI fails with TrainingError
See original GitHub issueRasa version:
rasa --version Rasa 1.0.0
Rasa X version (if used & relevant):
rasa/rasa-x:stable (docker)
Python version:
python3 --version Python 3.6.8
Operating system (windows, osx, …): Linux 143c6802c949 4.4.0-1083-aws #93-Ubuntu SMP Wed May 8 16:08:41 UTC 2019 x86_64 GNU/Linux
Issue:
The training button does nothing. Error in the logs is:
rasa-x_1 | Model training failed with status '500' and message '{"version":"1.0.0","status":"failure","message":"An unexpected error occurred during training. Error: expected str, bytes or os.PathLike object, not NoneType","reason":"TrainingError","details":{},"help":null,"code":500}'.
No idea what path it’s looking for. I followed the server deploy instructions at http://rasa.com/docs/rasa-x/deploy/
After fixing the issue in https://github.com/RasaHQ/rasa/issues/3726, I was able to talk to existing models as trained in my local instance. Now, in the public deployment, I cannot train.
For that matter, none of the training data is present (no stories, no NLU data), and I have no idea how to import it into the postgres database.
Issue Analytics
- State:
- Created 4 years ago
- Comments:5 (3 by maintainers)
Top GitHub Comments
hi @ddombrowsky - there’s a new release of rasa x that brings in first-class support for the domain file. It’ll be ready very soon, will post back here when it is.
This is released and part of the 0.19.1 release. Please let us know if there are still issues with your workflow.