« first day (1418 days earlier)      last day (1148 days later) » 
00:00 - 20:0020:00 - 23:00

12:18 AM
~ The Wizard, the Prince and the Good Fairy
3 hours later…
3:30 AM
God is everywhere, so am I.
@PrincessLuna your name sounds fishy
1 hour later…
4:50 AM
good morning
@InsaneCat how ya doin' man
Hey @Kaushik
GM bro
i'm fine
wht abt u :)
marriage done
now living with family shifted here from home town :)
5:06 AM
how is my bhabi?
3 hours later…
7:55 AM
@TheLittleNaruto why
@ShadowWizard astonishment (?)
@FOX9000 bewildered
@ShadowWizard rushed (?)
@InsaneCat nice! When the babies are coming? :D
@TheLittleNaruto why?
8:12 AM
@ShadowWizard @PrincessLuna your name sounds like a old user but you have princess as prefix, that is fishy
~ Midnight Shadow
@TheLittleNaruto old user? Lunar Watcher? That's the same person... ;)
!!feed @Princess
poor @Kenny is borked... and @rene isn't pingable... :( :( :(
Just reply to an old message
nah, too lazy
@TheLittleNaruto ok?
8:21 AM
let's kick @Kenny, sometimes it helps
!!come back
!!pee plz
Just wait
@Moosebot revive @KennyBOT
@ShadowWizard Command not recognized.
@rene The land of the living summon thee!
@TheLittleNaruto either way, why is that fishy?
We also had someone called FlamePrincess here long long ago, though she became MysticMagicϡ at some point and left SE after getting married.... maybe @The referring to her? :)
8:35 AM
(dammit, there isn't a single bot here with a shrug command)
@ShadowWizard Command not found.
@ShadowWizard Yep
@TheLittleNaruto bet (?)
>>read 6723188
8:36 AM
@ShadowWizard shrug
@FOX9000 gf
@TheLittleNaruto TheLittleNaruto does not have one (?)
@Princess ^^^
Who did this
8:36 AM
Whoever does it, shall pay for this
I almost forgot that xD
@TheLittleNaruto well, being a princess, @Princess should have enough money to pay. :D
This is not forgiven
8:43 AM
@ShadowWizard :D
@Pseudohuman n = null
And P = Punishment
cc @ShadowWizard
@Moosebot wake @TheLittleNaruto with @ShadowWizard
> BZZZZZZZZZZZZZ! @TheLittleNaruto with @ShadowWizard was shocked to get woken up!
That's gotta hurt
8:51 AM
stabs @PrincessLuna again and again on her neck from behind until she dies
There is no escaping
gets Celestia and Twily to revive me. Gets the Enterprise to beam @TheLittleNaruto into space in Klingon territory
And that's how ya escape :D
Btw, you really should get outta there before the Klingons see you. They're not exactly friendly
@PrincessLuna but they are
Q: Why are the Klingons so friendly in The Undiscovered Country?

Matt JohnsonI recently watched Star Trek: The Undiscovered Country, and it surprised me how un-Klingon-like the Klingons were. When this film was released, we had already seen quite a lot of examples from TNG of Klingon culture and behavior from Worf's experiences dealing with the Klingon High Council. I r...

That was a long time ago though
They're aggressive in TNG, discovery, deep space nine, enterprise and one more I can't remember the name of
You guys watch TV Series ?
9:01 AM
First confirm You're 16+
Because this TV Series I am gonna recommend, it's 16+ rated.
So? Rating doesn't matter
@TheLittleNaruto how do you want me to do that?
Just a "Yes" would be fine
@TheLittleNaruto yes
@PrincessLuna faintly (?)
@FOX9000 bf
9:04 AM
@PrincessLuna zoe does not have one (?)
"Altered Carbon" on Netflix
Thank me later!
It's on my list
Very nice
I finished it in one go
took office leave
Right behind Scorpion, Elementary, and Dr. Who/MLP/Star Trek Discovery if a new season pops up
9:15 AM
@FOX9000 gf
@PrincessLuna That word is already said in the latest 10 words. Please use another. (In case I'm mistaken, run >>rmword gf and then >>reply 6723280)
>>rmword gf
@PrincessLuna Word removed from latest words.
>>reply 6723280
@PrincessLuna TheLittleNaruto does not have one (?)
9:15 AM
Where did the berserk one go?
@FOX9000 wizard
@PrincessLuna hedge (?)
@FOX9000 hog
@PrincessLuna chicken (?)
@FOX9000 @sha :p
@FOX9000 cat
@PrincessLuna cigarette (?)
9:20 AM
@FOX9000 roll
@PrincessLuna thunder (?)
@FOX9000 song
@PrincessLuna ep (?)
@FOX9000 album
@PrincessLuna selling (?)
9:23 AM
@FOX9000 money
@PrincessLuna stealing (?)
:1 robbery
@PrincessLuna bandit (?)
:1 west
@PrincessLuna eastward (?)
9:24 AM
:1 direction
@PrincessLuna aligned (?)
:1 paralell
No associated word found for paralell.
I spelled it wront, didn't I? XD
I hate chatting on my phone
9:40 AM
@PrincessLuna agreed
@ShadowWizard cede (?)
@ShadowWizard rene (?)
@FOX9000 missing
@ShadowWizard kidnapping (?)
9:41 AM
@FOX9000 UFO
@ShadowWizard gerry (?)
@FOX9000 Kristen
@TheLittleNaruto porn (?)
No way!!!
9:52 AM
@TheLittleNaruto lol
@Kaushik lmao (?)
10:11 AM
@FOX9000 addictive
@ShadowWizard synaptic (?)
ha ha ha @ShadowWizard :)
@Kaushik lol tol pol rol dhol
10:34 AM
@DroidDev @berserk : youtube.com/watch?v=2jOeqZBpmyo @Kaushik
@TheLittleNaruto hahahaha ask @InsaneCat how is he.....
@Kaushik i'm fine dude busy in discord ;)
@InsaneCat :P
@Kaushik why are you offline there

 Kappa&Yokai and Friends

Meet friends and have a good time :)
11:05 AM
dude I am serving notice so got busy
"serving notice"?
can you please remove that message
see this is bad fox sent by hansu right now :D
@TheLittleNaruto @ber is loooong time gone :(
@berserk still got it? :P
@ShadowWizard for his research
11:13 AM
Hi insane shadow
@Kaushik research?
@ShadowWizard yeah his pervy research
@Kaushik ohh
No associated word found for ohh.
>>continue ohh ahh
11:15 AM
@ShadowWizard ahh (?)
oh yeah
lol aauchhh
11:29 AM
:1 break
@PrincessLuna strand (?)
:1 lazy
:89689668968569 foo
11:34 AM
@ShadowWizard nate (?)
:3 InsaneCat
@Pro no way to check if it's actual reply? ^^
@Moosebot wake @sha
@ShadowWizard nope
@PrincessLuna niet (?)
11:40 AM
@ShadowWizard It's not possible, since it would require indexing of messages, which is a real PITA
And there's room check
just checking a random message pops up in a different room, so it would technically be valid
@PrincessLuna yeah, but still...
2 hours later…
1:18 PM
//give @sha beer
gives beer to @sha
3:31 PM
so silent without @Kenny.... :(
4:32 PM
pokes @ShadowWizard
//kill @sha
pings @sha
> Licks @sha. Tastes like a wet dog! shivers
> @sha didn't watch their step and fell off a cliff
//give @Princess something shiny
gives something shiny to @Princess
4:34 PM
5:15 PM
@ShadowWizard Technically yes, by sending an extra web request... but meh, not worth it
@ProgramFOX well, in theory you can use client side hack, checking if it turned into reply
5:31 PM
@ShadowWizard Client side hack? Can you explain how it does not have to trigger an additional HTTP request?
@ProgramFOX sure. Trigger javascript timer that will check after a second if there is element with class "reply-info" and href /transcript/message/[id]#[id]
> JavaScript timer
... in Python
Wait, Python?
How do you do the WS connections to the network?
Thanks! ^^
I'm working on a second chatbot backed with a neural network that I really wanna hook up to the SE network chat :D
5:38 PM
What does the neural network do? Natural language processing?
NLP is not the same as a NN. NLP is how the text is processed (usually backed by a NN btw, at least that's how spiCy works). Neural nets is like the brain of whatever is going on, and in this case it's used to create sensible replies
Wow, that just sounded horribly formulated xD Does it make sense or did my typing not connect with my brain? :D
> NLP is not the same as a NN
I know
> NLP is how the text is processed (usually backed by a NN btw, at least that's how spiCy works)
That's what I meant: does your NN serve as the backing for the NLP?
> in this case it's used to create sensible replies
That's the actual answer to my question :P
I'm not sure I'm using NLP yet
The NN is basically the part that creates the thought vector (which I don't know how works btw :D but it basically creates the replies)
Related: Google Translate has a neural net to help create context-based translations
so the neural network inputs are just the chat messages (possibly pre-processed in some way?) and the outputs are a representation of what you'll reply?
yes. And the input is encoded by comparing it to a vocab with corresponding numbers for keys. I.e. the word yes can be encoded as 12351
Basically, the words are vectorized, inputted into the net, and the result is decoded and outputted as text
5:45 PM
Right. How does the neural network output look like exactly? I guess it contains information about what order the chosen words have?
Yeah, the order is included. I'm not sure exactly how it looks (data-wise), but I can get back to you on that later
And is it possible that a reply will have the same word in it more than once?
Hmm, now I'm really curious about the output representation
Found something online at least: pytorch.org/tutorials/_images/seq2seq.png
5:48 PM
Ah, I think I may have a clue about how the output representation looks like, then
may :P
Let me know when you know how it looks like exactly ^^
Once I figure out how this works ^^
have you written code already for the NN?
but now I gotta go eat dinner. I'll be back in an hour ^^ ANd this will hopefully be working by tonight
@ProgramFOX some of it
I see
anyways, afk
5:51 PM
1 hour later…
6:54 PM
and hey @rah
@PrincessLuna hi!
How's it going?
@PrincessLuna Decent! Finals got over yesterday so I'm finally starting work on some programming projects which I've been thinking about and planning for a looooong time :)
@PrincessLuna Yep! What about you?
6:58 PM
Good! It's winter break this week, so I got tons of time to do stuff :D
@PrincessLuna Awesome :)
@rahuldottech Firefox (?)
I think today was the end of winter here, I went out in a t-shirt and didn't feel cold at all
Winter isn't over for another couple of months here (fortunately :D)
@PrincessLuna Winter >>> Summer
7:02 PM
println(winter > summer)//prints "true"
@PrincessLuna :D
I just love winter :D
feel ya
@ProgramFOX gotten to the encoder now
7:06 PM
The encoder message is created with np.full((maxLen), wList.index("<pad>"), and the items are filled in with encoderMessage[keyId] = wList.indexOf(inputWord)
@rahuldottech ^^
That reminds me, I should probably get around to (re)learning python and PowerShell
(re) because I learnt the basics and then promptly forgot it all
I knew Python in 20(15?) and then forgot it and re-learned it in 2017
memory is annoying :D
(that's why ya need LSTM's, to avoid the vanishing gradient problem :D)
@PrincessLuna You'd never guess which my first programming language was
Heck, you might not even know about it
No :P
7:10 PM
Java? Python? C? C++? C#?
And some (idiots) don't even consider it a real programming language :P
not kotlin, that, didn't exist until last year or something xD
@PrincessLuna Indeed :P
@rahuldottech befall (?)
7:11 PM
That was my second
Actually, I learnt them at more or less the same time
Is it a high-level language?
> A high-level language (HLL) is a programming language such as C, FORTRAN, or Pascal that enables a programmer to write programs that are more or less independent of a particular type of computer.
the "particular type of computer" bit is tricky
But it is pretty modern, yeah, and I'd consider it HLL
Sorry, my wifi keeps dropping out. Had to go cabled
Is it an esolang?
7:14 PM
@PrincessLuna Nah
@rahuldottech mak (?)
Can you even have an esolang as your first programming language?
@rahuldottech You could. It's a horrible idea, but you could
@PrincessLuna Nup
No associated word found for Nup.
7:15 PM
and it'd be pretty impressive to hear someone say "The first programming language I learned was Brainfuck" xD
@PrincessLuna This is sorta close, but just sortaaaaa
Visual Basic?
@PrincessLuna noooooo
No associated word found for noooooo.
I have no clue :D
7:17 PM
A batch file is a kind of script file in DOS, OS/2 and Microsoft Windows. It consists of a series of commands to be executed by the command-line interpreter, stored in a plain text file. A batch file may contain any command the interpreter accepts interactively and use constructs that enable conditional branching and looping within the batch file, such as IF, FOR, and GOTO labels. The term "batch" is from batch processing, meaning "non-interactive execution", though a batch file may not process a batch of multiple data. Similar to Job Control Language (JCL) and other systems on mainframe an...
I didn't think about shell scripts! :D
It seems kinda limited though
7:19 PM
I've used it for crazy stuff like T-H-I-S
def createTrainingMatrices(conversationFile, wList, maxLen):
    convDict = np.load(conversationFile).item()
    examples = len(convDict)
    xTrain = np.zeros((examples, maxLen), dtype = "int32")
    yTrain = np.zeros((examples, maxLen), dtype = "int32")
    for index,(key, value) in enumerate(convDict.iteritems()):
        encoderMessage = np.full((maxLen), wList.index("<pad>"))
        decoderMessage = np.full((maxLen), wList.index("<pad>"))

        keySplit = key.split()
        valSplit = value.split()
Not finished yet, but you get the idea
@rahuldottech cool! ^^
@PrincessLuna Nah, there's workarounds for everything, that's part of the fun of it
@ShadowWizard QUICK! You have to get here now! Stuff is going @berserk in here! :P
@rahuldottech It's still stuck to only Windows though
7:23 PM
That's me
Hiya o/
@PrincessLuna Yeah :/
7:24 PM
@rahuldottech lucas (?)
>>link yeah yes
@PrincessLuna Link added.
@rahuldottech Still cool though ^^
I know a little batch, but it's not a lot
@PrincessLuna It's mostly basics + a ton of different undocumented tricks and workarounds
7:27 PM
sounds easy..
It's programming. Everything sounds easy, and ends up being hard as hell xD
I was being sarcastic lol
I wasn't :D
@berserk haha :P
@rahuldottech shack (?)
7:33 PM
//kill @berserk, for being sarcastic,
My crappy coding skills created a respawn thread that doesn't work :D
threading and IO is a real pain tho
7:50 PM
@ShadowWizard that cake is making me hungry :p
That's not his username @ber
It's @Shadow the Cake Wizard :P
@ShadowWizard Pliz rename to "Shadow the Blizzzzard"
00:00 - 20:0020:00 - 23:00

« first day (1418 days earlier)      last day (1148 days later) »