SimGus

Hello :) You are absolutely correct about your percentages there. It is not possible at the moment to change the probability of generating each category in a choice, though as...

Hi, This is not an error but rather a preparation: I'd like to make the generation lazy and have the adapters ask for the examples only when they need them....

Hello @tomgun132, thanks a lot for your changes! I'll review them as soon as I can and let you know if any changes are required

Hi @Bohemus307! Since yaml is a superset of json, the json adapter can actually be considered a yaml adapter (which is the reason I never made a "proper" one). However,...

It's been a while I haven't used Rasa, but unless they changed something you should be able to simply add the generated file to the nlu directory

Hi, Indeed, overfitting is a problem that I also encountered. This can actually be expected in any process of creating a dataset for an AI-related task, even though you are...

Hey, I think we are on the same page then: adding "expansion" capabilities to Chatette seems to be a good idea. Transformer-based algorithms such as BERT seem to be the...

Hello! There is currently no way to do that, though it's a feature I found myself needing as well. I worked on implementing such a thing (there might still be...

Hi! I see the kind of use case you have, it's quite similar to what I tried to do a few months back. I still need to think a little...

Hello, Inline entity values in markdown are not currently supported, simply because they didn't used to be documented in [Rasa's documentation](https://rasa.com/docs/rasa/nlu/training-data-format/#markdown-format) when the markdown adapter was implemented. The same is...