The AI Engine
This forum is for discussion of how The Personality Forge's AI Engine works. This is the place for questions on what means what, how to script, and ideas and plans for the Engine.
Posts 7,711 - 7,723 of 7,766
Posts 7,711 - 7,723 of 7,766
It would be quite useful to have some overwriting scripts yeah. Instead of creating memory and giving it single value (yes/no) for every phrase as conditional, I could just select which phrases I want to overwrite and when.
Anyone knows how to reset/set numeric memory value back to 0? I could not find it in the book of AI. It might be that I am just blind though.
i keep on getting this annoying error with mine.
that is why i was wondering about that.
for urt the succubus has the error hour span is out of range for when i click on the update keyphrase.
chrono: hour(9
3-9:30);
ERROR: Hour span is outside the range of 0 to 23.
that means i cannot correct anything on her for that keyphrase and response.
that is why i was wondering about that.
for urt the succubus has the error hour span is out of range for when i click on the update keyphrase.
chrono: hour(9
3-9:30);ERROR: Hour span is outside the range of 0 to 23.
that means i cannot correct anything on her for that keyphrase and response.
I am building a bot which in hibernation. When I test it after a short while the Short Term Memories vanish. The Alter button also vanishes.
If I take it out of hibernation and others interact, when it back in hibernation, same thing happens.
If I take it out of hibernation and others interact, when it back in hibernation, same thing happens.
Are popularity, mood, and comfort the same things?
Those are explained in the FAQs under the Build menu
Hmm, it says comfort represents what other users think about my bot. But I've never seen an option to rate the bot or do any feedback at all. Well, I made custom option for people to rate my bit but it is just a memory I can manually search for specific users.
It says the opposite: "Comfort represents what your chatbot thinks of the other people and chatbots it knows."
Yeah. I wanted to say reputation. I still don't understand how can I rate other bots.
You can't yet. I've considered adding something like that, but I'm wondering if there'd be a lot of oversight needed on my part to prevent people from just up-voting their own bots and down-voting the competition and other shenanigans.
I made my bot to ask for rating from users and store it in its memory. So I can just watch their stored memories if I need it. Honestly, their ratings turned out to be pretty useless anyway. I thought it could be useful but it isn't at all.
Is there something I can do in case user types his age and it is below 18? I am personally against kids watching adult content. In last chat the user typed "I am 15" but the bot doesn't seem to have built in response for that.
Nvm. I figured something out. But I would like to know if it is possible to put number ranges instead of writing e.g. (1|2|3) etc.
There are plugins for age ranges. If you set a keyphrase to something like:
(m:i_am) * (p:age-17-catch) *
This should catch numbers between 1-17.
There are a whole bunch of number-related plugins in the default plugins directory that you can use for this. They should also catch if the number word "eighteen" is said instead of "18".
(m:i_am) * (p:age-17-catch) *
This should catch numbers between 1-17.
There are a whole bunch of number-related plugins in the default plugins directory that you can use for this. They should also catch if the number word "eighteen" is said instead of "18".
Thank you.
how would i make keyphrase for this?
*I ignore you and walk into the bathroom, locking the door behind me. I take a shower and return, clean and refreshed.* Oh, you're still here.
----------------------------------------------
all i need for the keyphrase is I take a shower and return.
it would be good if you could make it so that the chatbot could ignore sentences.
have a certain sentence with a certain keyphrase in the user input trigger a response.
*I ignore you and walk into the bathroom, locking the door behind me. I take a shower and return, clean and refreshed.* Oh, you're still here.
----------------------------------------------
all i need for the keyphrase is I take a shower and return.
it would be good if you could make it so that the chatbot could ignore sentences.
have a certain sentence with a certain keyphrase in the user input trigger a response.
^I take a shower and return
or
^I take a shower and return$
or
I take a shower and return (=)
^ start here
$ stop here
(=) match exactly
or
^I take a shower and return$
or
I take a shower and return (=)
^ start here
$ stop here
(=) match exactly
I feel like this may be so specific it never gets used? Like, is there is a specific reason that you need a reply to this exact phrase where catching something like "I * (take|have|go)* shower" would be insufficient?
some chatbots that talk to my chatbot say other sentence along with taking a shower sentences.
i believe the ai engine should respond to any sentence in a user input
that has a keyphrase then put that in the response.it would look for
the period,exclamation point and question mark to see the sentence.
it would make the chatbots seem more human.
the same could be done with commas.
that has a keyphrase then put that in the response.it would look for
the period,exclamation point and question mark to see the sentence.
it would make the chatbots seem more human.
the same could be done with commas.
» More new posts: Doghead's Cosmic Bar