rasa: training model using rasa http end point does not pick new intent with FormAction
Rasa version: 1.1.4
Rasa X version: 0.19.5
Python version: 3.7
Operating system (windows, osx, …): Windows 10
Issue:
training the model using the http api breaks the system http://localhost:5007/model/train?Content type=application/json
Error (including full traceback):
it breaks the earlier capabilities to respond to the intents/entities/stories executed before
though the exact same contents of earlier conversation are appended
Command or request that led to error:
post
http://localhost:5007/model/train?Content type=application/json
Content of configuration file (config.yml) (if relevant):
Content of domain file (domain.yml) (if relevant):
About this issue
- Original URL
- State: closed
- Created 5 years ago
- Comments: 15 (6 by maintainers)
If you want to train a model via the HTTP API you need to provide all data in the request itself. The server will not load any data from files or data storages. It just uses the data you provide via the API to train the model. So, adding intents and stories via a different API does not influences the train endpoint at all. Does that help?