hi i have archived conversation/chat in file.txt like
me : ..
friend : ..
me : ..
me : ..
friend : ..
me : ..
friend : ..
friend : ..
me : ..
and so on..
can i use that as dataset ?
thanks
Yes, you can store this conversation in proper corpus format (.JSON or .YML) as you can see in _site-packages/chatterbot_corpus/data/_.
You can add your conversation in _site-packages/chatterbot_corpus/data/custom/_.
I prefer .YML, as it is more human readable, as compared to .JSON
Yes using existing export feature http://chatterbot.readthedocs.io/en/stable/corpus.html?highlight=export#exporting-your-chat-bot-s-database-as-a-training-corpus
its possible to have multiple statement in one section ?
as i see in documentation :
Satement
Response
Satement
Response
Satement
Response
etc ...
and here what would i do :
Satement
Response
Statement
Statement
Response
statement
statement
statement
Response
@guangrei It won't possible with current training algorithm. But i didn't understand your usecase, if possible can with some examples with your usecase?
the case look like a multiple statement with one response e.g:
when statement match one of statement 1 or statement 2 or statement 3 then will get same response
@guangrei It currently isn't possible to have more than one statement as a response. I agree this might be useful, and I've had a number of requests for this functionality. I'll look into the possibility of adding this feature (https://github.com/gunthercox/chatterbot-corpus/issues/58).
This thread has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs.
Most helpful comment
Yes using existing export feature http://chatterbot.readthedocs.io/en/stable/corpus.html?highlight=export#exporting-your-chat-bot-s-database-as-a-training-corpus