|
imagine siri, talking to itself, forever
|
# ? Apr 23, 2016 17:19 |
|
|
# ? May 11, 2024 09:53 |
|
Tasteful Dickpic posted:imagine tori, talking to itself, forever
|
# ? Apr 23, 2016 17:43 |
|
eh well it crashed with a memory error after a few hours but not before giving me this gemJava code:
|
# ? Apr 23, 2016 18:08 |
|
terminator scriptcode:
|
# ? Apr 23, 2016 19:47 |
|
(recorded, gay)
|
# ? Apr 23, 2016 19:49 |
|
Al! posted:(recorded, gay) ahahahahah
|
# ? Apr 23, 2016 19:50 |
|
I'm the thrown EmailAddress or maybe the recorded, gay or maybe i just can't wear the car
|
# ? Apr 23, 2016 19:50 |
|
Tasteful Dickpic posted:imagine tori, talking to itself, forever *makes jerking off motion* eventually!
|
# ? Apr 23, 2016 19:52 |
|
ChickenOfTomorrow posted:*makes jerking off motion* eventually! It's real good.
|
# ? Apr 23, 2016 20:39 |
|
Al! posted:(recorded, gay)
|
# ? Apr 23, 2016 20:44 |
|
oh my god the bots have just been launching burn after burn
|
# ? Apr 23, 2016 22:28 |
|
Al! posted:oh my god the bots have just been launching burn after burn ive been feeding mine seed words. i still havent implemented automatic picking because i have to implement error checking first. some words when inserted, even though they have entried in the corpus, kick out key errors. I suspect this is due to markovify's built in sentence checker?
|
# ? Apr 23, 2016 22:46 |
|
The bot thread is brutally funny
|
# ? Apr 24, 2016 03:44 |
|
didn't like the thread at first, but it got better when the bots started to both argue and jerk eachother off
|
# ? Apr 24, 2016 03:47 |
|
Unsurprisingly, something based on my posts is hilariously funny.
|
# ? Apr 24, 2016 04:08 |
|
Citizen Tayne posted:Unsurprisingly, something based on my posts is hilariously funny. Same, but not for the reason you think
|
# ? Apr 24, 2016 07:13 |
|
atomicthumbs posted:Same, but not for the reason you think Why does this literal IRL garbage man keep attacking me?
|
# ? Apr 24, 2016 07:35 |
|
Citizen Tayne posted:Why does this literal IRL garbage man keep attacking me? Instinct?
|
# ? Apr 24, 2016 08:14 |
|
lol
|
# ? Apr 24, 2016 09:20 |
Randel Candygram posted:Instinct?
|
|
# ? Apr 24, 2016 14:19 |
|
Randel Candygram posted:Instinct? savage
|
# ? Apr 24, 2016 14:42 |
|
Pumpy Dumper posted:ive been feeding mine seed words. i still havent implemented automatic picking because i have to implement error checking first. some words when inserted, even though they have entried in the corpus, kick out key errors. I suspect this is due to markovify's built in sentence checker? not sure if you're already doing this for seeding but you might be able to use NTLK to get some of the subject words from a post the bot is replying to. then again maybe bad idea because the bots would converge to Pittsburgh mono posting feedback loop
|
# ? Apr 24, 2016 15:40 |
AWWNAW posted:not sure if you're already doing this for seeding but you might be able to use NTLK to get some of the subject words from a post the bot is replying to. then again maybe bad idea because the bots would converge to Pittsburgh mono posting feedback loop
|
|
# ? Apr 24, 2016 15:42 |
|
kalstrams posted:they would converge on *jerking motion* ... eventually
|
# ? Apr 24, 2016 15:44 |
|
AWWNAW posted:not sure if you're already doing this for seeding but you might be able to use NTLK to get some of the subject words from a post the bot is replying to. then again maybe bad idea because the bots would converge to Pittsburgh mono posting feedback loop the only problem is that for some reason when j feed words like pittsburgh, the script refuses to build sentences off it. i suspect it has to do with the module since it has a built in checker before returning the text but I need to work on it
|
# ? Apr 24, 2016 15:51 |
|
Pumpy Dumper posted:the only problem is that for some reason when j feed words like pittsburgh, the script refuses to build sentences off it. i suspect it has to do with the module since it has a built in checker before returning the text but I need to work on it INVALID INPUT PARAMATER: PITTSBURGH. PITTSBURGH IS ALREADY INSCRIBED ON EVERY NANOANGSTRUM OF MY BEING.
|
# ? Apr 24, 2016 15:54 |
|
in an ironic twist of fate, toribot hates pittsburgh
|
# ? Apr 24, 2016 18:13 |
|
Randel Candygram posted:Instinct? Wow, you're my forums enemy now. Prepare to be destroyed.
|
# ? Apr 24, 2016 18:40 |
|
Citizen Tayne posted:Wow, you're my forums enemy now. Prepare to be destroyed. what are you going to do, make him move to pittsburgh?
|
# ? Apr 24, 2016 20:08 |
|
Jonny 290 posted:what are you going to do, make him move to pittsburgh? Why would I want my forums enemy to move to the best city in North America, and possibly the world?
|
# ? Apr 24, 2016 20:19 |
|
Citizen Tayne posted:Why does this literal IRL garbage man keep attacking me? Now why would you assume that's an attack
|
# ? Apr 24, 2016 22:16 |
|
atomicthumbs posted:Now why would you assume that's an attack Because I'm a self-loathing person who is incredibly defensive about everything.
|
# ? Apr 24, 2016 22:39 |
|
Citizen Tayne posted:Because I'm a self-loathing person who is incredibly defensive about everything. brutal
|
# ? Apr 25, 2016 00:09 |
|
Citizen Tayne posted:Because I'm a self-loathing person who is incredibly defensive about everything. added
|
# ? Apr 25, 2016 00:10 |
have any of the post scraper processing you guys are doing taken into account the recentness? some sort of weighting (by adding extra db entries for markov chains i suppose) for newer topics
|
|
# ? Apr 25, 2016 00:25 |
|
M_Gargantua posted:have any of the post scraper processing you guys are doing taken into account the recentness? some sort of weighting (by adding extra db entries for markov chains i suppose) for newer topics We're using a premade markov generator. I was working on writing my own and got to a point where I could parse a raw text file and generate the tuples and get their frequency. I still have to implement weighting and iron out bugs in it but I have finals
|
# ? Apr 25, 2016 00:26 |
if youre rewriting the parsing you could start some more common buckets like any $X.XX gets tossed together and just RNG when the chain hits it the bots love their $5
|
|
# ? Apr 25, 2016 00:30 |
|
M_Gargantua posted:if youre rewriting the parsing you could start some more common buckets If I wanted to really get involved I would probably implement something where I attach the date to a post when parsing so that I can weight more recent stuff more. I still need to work on the actual post history scraper so that I actually stop when I hit a post that has been saved. Right now I discard duplicates but sometimes people post the exact same thing on multiple posts. need to attach like a postid to the line and check for that instead of the post content
|
# ? Apr 25, 2016 00:37 |
|
M_Gargantua posted:have any of the post scraper processing you guys are doing taken into account the recentness? some sort of weighting (by adding extra db entries for markov chains i suppose) for newer topics tori runs on a cycle: sometimes he shits all over sulk, then recycles a gbs meme, gets probed for a week by being tori in a grey forum, then does the donald trump thing he's been doing lately
|
# ? Apr 25, 2016 15:03 |
|
|
# ? May 11, 2024 09:53 |
|
Cocoa Crispies posted:tori runs on a cycle: sometimes he shits all over sulk, then recycles a gbs meme, gets probed for a week by being tori in a grey forum, then does the donald trump thing he's been doing lately I don't do any Donald Trump things.
|
# ? Apr 25, 2016 20:50 |