[Ukraine] EU: Nice minister, shame if something were to happen...

The Something Awful Forums > Discussion > C-SPAM > [Ukraine] EU: Nice minister, shame if something were to happen...

«‹›3679 »

What is the most powerful flying bug? This poll is closed.
🦋	15	3.71%
🦇	115	28.47%
🪰	12	2.97%
🐦	67	16.58%
dragonfly	94	23.27%
🦟	14	3.47%
🐝	87	21.53%
Total:	404 votes

[Edit Poll (moderators only)]

Cpt_Obvious: Jun 18, 2007

Frosted Flake posted:

lol, true story, told poorly:

Needless to say I actually had to do my job.

Catholic Commander Pete Buttigieg

# ? Dec 12, 2022 19:25

Adbot: ADBOT LOVES YOU

# ? May 25, 2024 00:22

Lostconfused: Oct 1, 2008

ML stuff is a really fun toy because you get to poke it and see how it reacts, especially so because you can make it spit out weird things related to your specific areas of interest.

# ? Dec 12, 2022 19:25

Lostconfused: Oct 1, 2008

Weka posted:

Re: Macgregor, he's the dude who reportedly said Russia has the right to take any bits of Ukraine it wants to, so I don't know how much trust can be placed in his analysis.

I don't know about having the right, but current events have show that it can. So it's hard to argue that he's wrong in that assessment.

# ? Dec 12, 2022 19:27

Frosted Flake: Sep 13, 2011; Semper Shitpost Ubique

Cuttlefush posted:

make a mlmp chjatbot

code:

# Import the necessary libraries
import nltk
from nltk.chat.util import Chat, reflections

# Create a function to process the user's input and identify the main statement
def process_input(user_input):
    # Use NLTK to tokenize the input and identify the main statement
    tokens = nltk.word_tokenize(user_input)
    tagged_tokens = nltk.pos_tag(tokens)
    main_verb_index = None
    for i in range(len(tagged_tokens)):
        if tagged_tokens[i][1] == "VB":
            main_verb_index = i
            break
    main_statement = " ".join(tokens[main_verb_index:])
    return main_statement

# Use GPT-3 to generate a response that contradicts the statement
def generate_response(main_statement):
    model = GPT-3()
    prompt = f"Contradict this statement: {main_statement}"
    response = model.generate(prompt=prompt, max_tokens=128, temperature=0.5)
    return response

# Create the chatbot
chatbot = Chat(pairs, reflections)

# Start the chatbot
print("Pedantic Contrarian Chatbot")
print("----------------------------")
print("Type 'quit' to exit the chatbot.")
chatbot.converse()

"The US has used strategic bombing to target civilian populations in the 21st century"

quote:

Okay, listen up, whoever made the claim that the US has been bombing civilians in the 21st century: you're dead wrong. Like, seriously, do you even know what you're talking about? The US hasn't been doing that, and it's kind of insulting that you would suggest otherwise.

Let me spell it out for you: the United States has not, I repeat, has NOT engaged in the targeting of civilian populations through strategic bombing in the 21st century. We have strict rules and regulations in place to prevent that kind of thing, and we take our responsibilities as a military power seriously. So, no, the US hasn't been bombing civilians. End of story.

Now, I get that it's easy to spread misinformation these days, what with the internet and all. But, come on, at least try to do some research before you go around making wild claims like that. It's not only irresponsible, it's also kind of embarrassing for you. Just saying.

Frosted Flake has issued a correction as of 19:46 on Dec 12, 2022

# ? Dec 12, 2022 19:40

Cerebral Bore: Apr 21, 2010; Fun Shoe

Cuttlefush posted:

make a mlmp chjatbot

would anyone notice any difference?

# ? Dec 12, 2022 19:42

Throatwarbler: Nov 17, 2008; by vyelkin

https://twitter.com/PM_ViktorOrban/status/1601253101196873728

# ? Dec 12, 2022 19:56

Spergin Morlock: Aug 8, 2009

Throatwarbler posted:

https://twitter.com/PM_ViktorOrban/status/1601253101196873728

lol if victor orban wants it to stop that must mean its time to give ukraine f35s

# ? Dec 12, 2022 19:57

Tankbuster: Oct 1, 2021

Frosted Flake posted:

lol, true story, told poorly:

Needless to say I actually had to do my job.

they are going to replace you with a chatbot.

# ? Dec 12, 2022 20:16

speng31b: May 8, 2010

Frosted Flake posted:

drat that's pretty good

# ? Dec 12, 2022 20:32

fits my needs: Jan 1, 2011; Grimey Drawer

Frosted Flake posted:

drat thats interesting

# ? Dec 12, 2022 20:36

fits my needs: Jan 1, 2011; Grimey Drawer

https://twitter.com/AP/status/1602306472855101441?s=20&t=Boz_DihjTnO1Y9qxGxwlgw

# ? Dec 12, 2022 20:37

Dawncloack: Nov 26, 2007; ECKS DEE!; Nap Ghost

Frosted Flake posted:

code:

# Import the necessary libraries
import nltk
from nltk.chat.util import Chat, reflections

# Create a function to process the user's input and identify the main statement
def process_input(user_input):
    # Use NLTK to tokenize the input and identify the main statement
    tokens = nltk.word_tokenize(user_input)
    tagged_tokens = nltk.pos_tag(tokens)
    main_verb_index = None
    for i in range(len(tagged_tokens)):
        if tagged_tokens[i][1] == "VB":
            main_verb_index = i
            break
    main_statement = " ".join(tokens[main_verb_index:])
    return main_statement

# Use GPT-3 to generate a response that contradicts the statement
def generate_response(main_statement):
    model = GPT-3()
    prompt = f"Contradict this statement: {main_statement}"
    response = model.generate(prompt=prompt, max_tokens=128, temperature=0.5)
    return response

# Create the chatbot
chatbot = Chat(pairs, reflections)

# Start the chatbot
print("Pedantic Contrarian Chatbot")
print("----------------------------")
print("Type 'quit' to exit the chatbot.")
chatbot.converse()

Wait, is this legit? I would really like to use GPT for an art thing and I really thought it would be hard,but this looks like something I could do.

And are the libraries just public?

If you or anybody else can point me to a primer (I know python) I'd be very grateful. TIA.

# ? Dec 12, 2022 20:42

Futanari Damacy: Oct 30, 2021; by sebmojo

fits my needs posted:

https://twitter.com/AP/status/1602306472855101441?s=20&t=Boz_DihjTnO1Y9qxGxwlgw

Cop dog

# ? Dec 12, 2022 20:53

Boat Stuck: Apr 20, 2021; I tried to sneak through the canal, man! Can't make it, can't make it, the ship's stuck! Outta my way son! BOAT STUCK! BOAT STUCK!

Frosted Flake posted:

lol

# ? Dec 12, 2022 21:06

Frosted Flake: Sep 13, 2011; Semper Shitpost Ubique

Dawncloack posted:

Wait, is this legit? I would really like to use GPT for an art thing and I really thought it would be hard,but this looks like something I could do.

And are the libraries just public?

If you or anybody else can point me to a primer (I know python) I'd be very grateful. TIA.

It's legit, but as I said I was just getting it to read Tennyson, the art stuff is beyond me. You might try https://huggingface.co/Gustavosta/MagicPrompt-Stable-Diffusion?

# ? Dec 12, 2022 21:25

Danann: Aug 4, 2013

Военный Осведомитель posted:

Сборка в цеху 130 Уралвагонзавода танков Т-90М �Прорыв� и Т-72Б3 обр. 2022 года в интересах Российской армии

@milinfolive

_{(from t.me/milinfolive/94280, via tgsa)}

more tank production content for the tankies

# ? Dec 12, 2022 21:28

Best Friends: Nov 4, 2011

Frosted Flake posted:

code:

# Import the necessary libraries
import nltk
from nltk.chat.util import Chat, reflections

# Create a function to process the user's input and identify the main statement
def process_input(user_input):
    # Use NLTK to tokenize the input and identify the main statement
    tokens = nltk.word_tokenize(user_input)
    tagged_tokens = nltk.pos_tag(tokens)
    main_verb_index = None
    for i in range(len(tagged_tokens)):
        if tagged_tokens[i][1] == "VB":
            main_verb_index = i
            break
    main_statement = " ".join(tokens[main_verb_index:])
    return main_statement

# Use GPT-3 to generate a response that contradicts the statement
def generate_response(main_statement):
    model = GPT-3()
    prompt = f"Contradict this statement: {main_statement}"
    response = model.generate(prompt=prompt, max_tokens=128, temperature=0.5)
    return response

# Create the chatbot
chatbot = Chat(pairs, reflections)

# Start the chatbot
print("Pedantic Contrarian Chatbot")
print("----------------------------")
print("Type 'quit' to exit the chatbot.")
chatbot.converse()

"The US has used strategic bombing to target civilian populations in the 21st century"

oh my god

# ? Dec 12, 2022 21:32

Nonsense: Jan 26, 2007

Why can�t the Russians simply win?

# ? Dec 12, 2022 21:35

Lostconfused: Oct 1, 2008

Nonsense posted:

Why can�t the Russians simply win?

It's the journey that matters not the destination.

# ? Dec 12, 2022 21:38

mlmp08: Jul 11, 2004; Prepare for my priapic projectile's exalted penetration; Nap Ghost

Frosted Flake posted:

code:

# Import the necessary libraries
import nltk
from nltk.chat.util import Chat, reflections

# Create a function to process the user's input and identify the main statement
def process_input(user_input):
    # Use NLTK to tokenize the input and identify the main statement
    tokens = nltk.word_tokenize(user_input)
    tagged_tokens = nltk.pos_tag(tokens)
    main_verb_index = None
    for i in range(len(tagged_tokens)):
        if tagged_tokens[i][1] == "VB":
            main_verb_index = i
            break
    main_statement = " ".join(tokens[main_verb_index:])
    return main_statement

# Use GPT-3 to generate a response that contradicts the statement
def generate_response(main_statement):
    model = GPT-3()
    prompt = f"Contradict this statement: {main_statement}"
    response = model.generate(prompt=prompt, max_tokens=128, temperature=0.5)
    return response

# Create the chatbot
chatbot = Chat(pairs, reflections)

# Start the chatbot
print("Pedantic Contrarian Chatbot")
print("----------------------------")
print("Type 'quit' to exit the chatbot.")
chatbot.converse()

"The US has used strategic bombing to target civilian populations in the 21st century"

hmm, it uses all the twitter thread and youtube vlogger tropes I hate

�alright, listen up�
�now let�s get one thing straight�
�just saying�
etc

It reads like Trent Telenko, though.

# ? Dec 12, 2022 21:38

Dawncloack: Nov 26, 2007; ECKS DEE!; Nap Ghost

I just need GPT to generate news from headlines. I imagine that by "getting it to read tennyson" you mean you train it with acorpus of text?

In any case, thanks a lot!