File importer #4014
File importer #4014
Conversation
|
@tabergma Would be great to get some first feedback from you! |
|
I will fix the remaining tests tomorrow |
| @@ -0,0 +1,188 @@ | |||
| import logging | |||
There was a problem hiding this comment.
I messed up the renaming here, sorry! I removed all the static stuff and adapted it to the TrainingFileImporter interface.
|
This PR is huge. Can we somehow split it up? Maybe just introduce the new importers in the first PR and move the actual usage of them into another PR? |
| async def get_story_data( | ||
| self, | ||
| interpreter: "NaturalLanguageInterpreter" = RegexInterpreter(), | ||
| template_variables: Optional[Dict] = None, |
There was a problem hiding this comment.
What are template_variables?
There was a problem hiding this comment.
@akelad Do you have idea? I did some digging in the code and apparently you can specify some templates in stories which are replaced while reading the files?
@tabergma Sorry, I know it's painful 😞 But just adding the importers would not make them usable at all. Only putting them in the train process makes them usable at all 🤔 |
|
@wochinge I added a couple of more comments. Still need to test it locally. Will do it tomorrow. But in general, I like the idea and the implementation looks good. |
Co-Authored-By: Tanja <tabergma@gmail.com>
|
@tabergma Ready for a final review |
|
Content looks good, thanks for the full example! Will do a quick copy edit and fix the docs warnings while i'm at it (check those next time 😉) |
|
If anybody has input for better names for |
|
Docs be fine 👍 |
Proposed changes:
SkillSelectortorasa/data/skill(I messed up the git rename, I'm sorry!)SkillSelectorto useTrainingFileImporterinterfaceTrainingFileImporterwhich centralizes the file importingStatus (please check what you already did):
black(please check Readme for instructions)