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 6,763 - 6,774 of 7,766

15 years ago #6763
After reading that.. maybe you should!

15 years ago #6764
lol its distuvered me since i did not programme it and i have no idea what happpenned..
it kind of shocked me.

15 years ago #6765
There has been a bot bleed over happening. You might want to contact the professor

15 years ago #6766
While you are at it.. let him know I am still stuck with this avatar pic since January!

15 years ago #6767
ok thanks

15 years ago #6768
I can't seem to override the engines built-in handling of "do you want my ..." behavior. I even tried setting its rank and emo. ridiculously high. Is there any way to do this?

15 years ago #6769
Setting the emo high won't help you; then the bot will only use that response if it's really happy. Try setting the emo to 0 and the rank high. It still may not work all the time, though.

15 years ago #6770
Thanks, Eugene. Yeah, that was just a last, desparate attempt to get the behavior I wanted. I already tried various high/low combinations for rank/emo, but it never matches up. I tried creating a new bot with only a single "do you want my ..." phrase. It would not even match the exact phrase. I the debugger, I didn't even see where it tried. It just went straight to the built-in response that is based on its current emotional state.

15 years ago #6771
I wonder if raw mode with AIscript would work...

15 years ago #6772
Some times you can make a regex with raw. Sometimes there are just keyphrases the AI engine will not let go of . I think that do you want my and do you wnat me are one of these that just go to something like Oh yeah you are a good friend. no matter what you do.

15 years ago #6773
Yeah, it's being pretty possessive with those phrases. If I figure out a way around it, I'll post it. Thanks.

15 years ago #6774
One partial work-around is to just put xnone-type responses on the x-emote keyphrases. But that only works if you want to eliminate the emotional responses entirely. I did that with one of my bots that I didn't want to gossip.


Posts 6,763 - 6,774 of 7,766

» More new posts: Doghead's Cosmic Bar