Getting the following error while training my rasa 2.0.2 version bot, my rasa-sdk is 2.0.0 and I haven’t added ‘nlu_fallback’ in my domain file as an intent. Please help. @akelad
UserWarning: Intent ‘nlu_fallback’ has only 1 training examples! Minimum is 2, training may fail.
kaiogu
(Kaio Giurizatto Utsch)
December 1, 2020, 1:06pm
2
Hey, I found the same problem it seems to be a bug about counting NLU examples, but it doesn’t seem to actually affect the model working.
I opened an issue tracking this:
opened 09:53AM - 26 Nov 20 UTC
closed 09:16AM - 19 Apr 21 UTC
type:bug
type:enhancement
area:rasa-oss
effort:atom-squad/2
area:rasa-oss/cli
feature:ux-cli+training-data
**Description of Problem**:
<!-- Short overview of the current situation.
Why … is this feature needed? Please link any relevant
[forum](https://forum.rasa.com) threads here. -->
The message about too few **core** training examples doesn't inform of what kind of training example is needed (i.e. more intent phrasings or more stories containing the intent). It also doesn't help that the message appears right after `Training NLU model...` giving the impression that this is about nlu data.
```
UserWarning: Intent 'out_of_scope' has only 1 training examples! Minimum is 2, training may fail.
```
**Overview of the Solution**:
<!-- What would a possible solution look like?
Describe, without going too low into technical details,
what changes need to happen during implementation of this feature. -->
One way to make this better would be to explicit what kind of training example is meant:
```
UserWarning: Intent '{intent}' is present in only one story! Minimum is 2, training may fail.
```
Alternatively, raise the warning only after `Training Core model...` which would also help put the warning in context.
<!-- What needs to be there to consider this feature as done?
- [ ] Tests are added
- [ ] Feature described the docs
- [ ] Feature mentioned in the changlog
- [ ] ... -->
Hi, have you found a solution to this? Do I need to set an intent named as “nlu_fallback” ?
No, you are not supposed to do that. We shouldn’t be creating training examples for nlu_fallback since it’s a default behavior.
1 Like