#voice-chat-text-0
1 messages Β· Page 85 of 1
it also doesn't matter
Draft dodger lol
I asked during manager interview what interview was and languages acceptable, they said it was some leetcode and whatever language I wanted
Nah π But it's a thing people say.
Told the manager "Nope, this interview process is over now"
Fair enough
I'd just want practical code interview questions.
Like related to what I'd actually be doing
servername:{OS: 'Windows', SoftwareInstalled: ['Adobe', Office']}```
Gotta love that
Wait, Lance is still doing that?
Thought the event was over
The heart thingy
It's configured by month
Goooootcha
So always the entire month
Forgot
We need a month of hate, to balance things.
What should the reaction be?
π ?
I think we get that enough during gay pride
Yeah fair enough
@rich dockπ
Ah, yes. The delicious, chocolaty taste of hazelnoots.
Noothing like it.
Love me some hazelnuts
There's an Italian product. Nuttino. It's also very good.
Same idea?
Yes.
Hi
What???????
Nutella. It's canonically pronounced "Noo-tella"
Even though pretty much everyone says "Nut-ella"
I'm from Russia, but I love Nutella>)
So what're folks up to
Well shit, now I'm second guessing myself
There's a part of me that really REALLY wants to do TypeScript using Sublime
But I know the tooling for it is so good in VSC
VSC and devcontainers. Thought over
Yeah still need to try those
Reading up on how to make one now
@molten pewter What up home slice
I don't think we've really asked ourselves.... What IS "up dog"?
Is it a state of mind, a state of home boyedness...
The philosophical possibilities are endless
Nice
wait the bot can rea the gifs?
thts amazing
!
It'll be the url.
As we do
@amber raptor Do I set it up inside an existing project folder or do I make the container and then clone/add the stuff there
yeh i thought the same but i was mistaken becz when i clicked on "speak msg"
Or does it matter
@molten pewter Whatcha watching?
in order to find out
yuss:}
have you guys seen, "Platform"?
It just sounds like someone talking far away in a different room but also muffled
And it's occasional
Haven't heard of it, no
hmm
Huh, also didn't know this was a thing
also the fact that netflix servers and so are its contents dofferent
help
pleez
ERROR: Could not find a version that satisfies the requirement discord_components (from versions: none)
ERROR: No matching distribution found for discord_components
What version of Python are you using?
You might need to downgrade to 3.10 or something
3.10.0
Do py It says you're on 3.10.0, right?
yes
Although honestly.... I think discord.py's latest has component support
I think?
I'd have to look
@willow light Hey buuuuuuuuuddy
I'm increasingly convinced that the UDeploy team at IBM comprises entirely of sadists.
Sounds right
Possibly one of the slowest and buggiest build systems I have ever had the displeasure of using.
Makes me wish I was using Jenkins.
"You said IBM? Oh dude, me too!"
You wish for Jenkins?
Dear Dog
@regal boltπ
@zenith radishπ
Hemalloc
czxfsdfsd
hemset
You're welcome
Nerdy
Learning about dev containers
Yeah
That's my brain
Keeps everything isolated and consistent
Embrace containers!!
Rabbit would be proud
nah man, run everything on the same machine in the same virtual environment. screw namespaces, import everything as `*
Through VSC
That's not the same thing
Don't know exactly what it does, but it's different
Right
It was fun back then
It's tedious now
Was that Uru?
There's a reason it failed
Eh... They had that in some of the very very old shooters
Think it was an option in Doom
Unless your midi driver was bad
In which case it'd be the only thing saving your ears
I have those from time to time
Kinky
What're you up to LP?
Are you currently playing RS or...
@zenith radish Notice me LPai
Scratches an itch
It's not bad
No no
Scratching an itch of like... seeing numbers go up and slightly mindless
Also yes
As a former incremental gameaholic
I get it
I have an incremental game in mind
That I've been wanting to make
I just never knew how to do the cost scaling
I'm not
I legit have had an idea kicking around my head for about 3 or 4 years
It's dumb and simple
Overall concept is a dog digging in a backyard to find bones. Then he gets a shovel. Then a bigger shovel. Then a backhoe. Etc.
Then hires other dogs to get more bones in other yards
@zenith radish What would be nostalgia scratching for you?
What were your childhood gaming jams?
Strawberry?
I'm not sorry
I take that as a compliment
@sweet lodge Do I have to make this in a new project folder or can I plop it into an existing one?
Poor guy
?
VSC creates a container and puts your code inside it
Code has to go inside the container
Okies
π
Hey Mosa
@rain moon π
How much of it?
very little
What is the question?
Oh I can open folder in container
@somber heath Wait i didnt hear you if you were talking i just put my heasphones and apperently its broken
He was saying hi
can I send it here ?
Local folder?
Huh, TIL
But the defeats the entire purpose
To some extent at least, you still get to segregate deps and stuffs
i did a post about it but i dont remember how to send you the channel
Should be able to right click and then Copy Link I think
On the channel list on the right
Left
Left Hemlock
Get your shit together
Oh , thanks
they dont want uppercase letters
no the Falses and Trues after the commas in the function call are representing the answer I should have
- dont mind me if i dont understand you my English suck a little but
No worries
I would think the main benefit is the environment itself. Code can go anywhere, but your specific build chain and stuff are where they are
So I don't think it defeats the purpose at all

Welcome back
You could just... turn off the monitor by hand
How far away is it
Skill issue
I say it with love
Hey finesse
What's new with you
We had a pipe burst near the end of December, and my regular office got thrashed
It's what we call the small conference room
And it's in the middle of the business services area
Eh
It's fine
I'm patient
Nice
Jesus how
I mean how do you even manage that
That's cray cray
!ot
Off-topic channel: #ot2-never-nesterβs-nightmare
Please read our off-topic etiquette before participating in conversations.
Yeah, we've got a category of 3
You can do it here too. What's the link? Server appropriate right?
yeh
i just put together a new dnb mix you might enjoy when coding / hacking ... https://soundcloud.com/tripaddict/trip-addict-2023-fresh-heavy-dark-dnb
fresh heavy dark dnb
Oooo nifty
its very heavy though lol
not for the feint hearted lololol
ill stick it in the ot channel too
but i hope you enjoy it if you have chance to give it a listen π
For sure
wicked π
@rugged root is voicefullock back yet?
Not for a little while still
I wrote an answer, remove the duplicate print()
Because q is a constant, it's only n that is going higher
Because the assignment says so π
quonstant.
print()
for x in nums:
print(x)
``` so this is O(n+1) which is simplified to O(n) bc constant doesnt really do that much , i get this
for x in fruits:
print(x)
for x in nums:
print(x)
``` this is O(n+m) which is simplified to O(n)
why
@slim edgethank you !
ye but this is two linear loops
So the assumption is that there's only one variable that will change?
This all breaks my head
@slim edgecan i massage you in dms ?
Sure, I love thai-massage
but then O(2n) is O(n) this is literally the exact opposite
O(1+n+n^2+5^n)
its O(5^n)?
lim(f1 -f2) -> infinity
O(n) is just a way to say the function is constant, they are not trying to say 0(1n)
which linux distro should I use for WSL?
just the ubuntu or the lts
ya literally understood nothing lol
I think that's preference
alright immaa go with the lts
With these, the implication is that there's only one input that changes?
MHM
but what if n is an array and m is a array list
m should be larger
even a queue
Then m becomes n and vice versa?
The question is by convention: "What happens if the value N goes up"
If the assignment asks for M instead then sure
@rugged root :x: Your 3.11 eval job has completed with return code 1.
001 | Traceback (most recent call last):
002 | File "<string>", line 1, in <module>
003 | File "/snekbox/user_base/lib/python3.11/site-packages/matplotlib/__init__.py", line 900, in <module>
004 | dict.update(rcParams, _rc_params_in_file(matplotlib_fname()))
005 | ^^^^^^^^^^^^^^^^^^
006 | File "/snekbox/user_base/lib/python3.11/site-packages/matplotlib/__init__.py", line 583, in matplotlib_fname
007 | for fname in gen_candidates():
008 | File "/snekbox/user_base/lib/python3.11/site-packages/matplotlib/__init__.py", line 580, in gen_candidates
009 | yield os.path.join(get_configdir(), 'matplotlibrc')
010 | ^^^^^^^^^^^^^^^
011 | File "/snekbox/user_base/lib/python3.11/site-packages/matplotlib/__init__.py", line 279, in wrapper
... (truncated - too many lines)
Full output: https://paste.pythondiscord.com/unebiwezud.txt?noredirect
But M has the same time complexity of N, So its O(2n) => O(n)?
what happens if its async then
as in m is async and n is sync
It'd still be the number of times the operation happens
The time is based on the operations taken or needed (I think that's right anyway....)
So the delays due to async wouldn't matter
so O(n) each way
Unless it has to do more things based on increased inputs
So the dishes take a little bit longer to clean, so what?
Real math has curves.
Right
You'd be consistent at the amount of time you take per plate
You don't end up getting slower per plate or faster per plate
y=xΒ² is the sexiest
thicc maths
If you have a wide variety of herbs in your garden, would that be thyme complexity?
x^2 and 2^x are the same? wtf
Like Superman in a hurry.
"I left the oven on!" -bam-
Daww
ah makes sense now
We do our best to keep it a friendly place
Oh yeah? Well I hate you
β @tepid edge can now stream until <t:1676661454:f>.
@glass vigil π
ye exponential is the opposite of logarithmic
CSE Code With Engineering Playbook
@whole bear π
hello
hello
what's happening exactly in stream?
and whats the usecase of exponential
How much?
time complexity -topic?
kill @slim edge
Oddly and instantly harsh
The usecase is mostly just for exams, but when you deal with big user-bases this theory will tell you if your function will not work
Hey why me
You are for loop
lmao
Yeah if you could not randomly threaten our users that'd be great, @whole bear
so also O(2n + m) space should be O(n) space i presume?
yes
It's okey, I will kill him first
yes yes
As long as O(n) does not reach O(nΒ²) , it will always be O(n)
@slim edge You are foor loop or while loop?
How tf can O(n^2) space exist....
While for sure
ok while loop
@strong arch Sup
robb
They are just asking if the space required will multiply exponentially or linearly
By a rising value of N
you are not funny
Good thing your presence is adding so much value
Certainly more than yours
What
What'd I do now
I'm not following
@quaint oyster I mean the polite thing is to not bullrush in
ok thnx moritz still have no clue how O(2n) is O(n) but the rest got more clear atleast
They are not the same, but these assignments are asking if both are linear, and they are
@somber heath
About time I figured this one out lol
https://imgur.com/a/TOqCO6z
I think I don't have the permission to open mic.
Yeah, I understand
Cool
Where are you from?
I guess, I am gonna talk like this for now π
I'm from India.
What do you do?
Backend
Webdev
Any framework in specific?
That you have used
Okay
Golang is cool
Do you like python?
That's cool
Me
I worked on PHP earlier
Mainly on a framework named Drupal
Yup
Currently, I'm working on django
Looking like you're on the right track.
Hello how can I enable my voice chat
!voice
Voice verification
Canβt talk in voice chat? Check out #voice-verification to get access. The criteria for verifying are specified there.
bruh
I'm not able to hear anything while using discord on browser, how can I fix my browser settings?

there may be per-site permissions for playing audio
some browsers (e.g. tor) may have it blocked by default
shift+enter?
hello
world
this one looks suspiciously symmetric
a lot of them seem symmetric along the diagonal
I haven't made anything new since this
these three were quite early (even before bug fixes)
any1 familiar with abstract factory pattern?
I have a project with too many of it
You'd be great help than @vocal basin
what's the question?
ordering instead of shuffling the direction
its a whole prompt
self.offsets = [
(-1, 0),
(-1, 1),
(0, 1),
(1, 1),
(1, 0),
(1, -1),
(0, -1),
(-1, -1),
]
# in another method
a = self.angle + np.random.uniform(-np.pi / 8, np.pi / 8)
self.offsets.sort(
key=lambda t: (np.cos(a) - t[0] / np.hypot(*t)) ** 2 + (np.sin(a) - t[1] / np.hypot(*t)) ** 2
)
should have probably used np.sort instead
it's not very plausible to use numpy for stacks
not much gain
writing C extensions can help
(or Rust)
or anything that can interact with Python
fork numpy
or Fortran
some python libraries use it inside
very old language
Does any1 in this discord offer tutoring by any chance, I would actually pay because these university professors are not the best at teaching sometimes
!rule 9
read documentations
and see yt tutorials:}
i agree uni prof are not the way
at least for me too
depends
yea that's true
i have seen michigan unis
didnt look good
and then i thought to not waste time anymore
less circular
still it's not working
π
thing on the bottom left looks like electric circuit or something
can these ever be same for two people?
or they are randomly generated?
just curious:-
this uses angle deviation/randomness narrowed to 0
by default it's pi/8
although I'm not sure if it should be pi/8
what should I do now?
I've changed the ordering algorithm from distance to dot product
(although, I think, it shouldn't change the result)
512x512 takes around 36s to generate
(not very fast)
haven't really tried to optimise the generation itself yet
colour generation: 49s
merging (video): 41s
four colour arrays generated separately
then merged based on when the pixel was filled
just # in dockerfile
there are things like RUN where # can be a part of the command
like
(shell thingy)
I'll turn off all logic and measure the time
I know I can't optimise merging reasonably
like
black video: 40s
merging: 50s
The first one of these is almost bang on.
(cv2 on itself takes most of the time)
36s -> 6s
-> 4s
it's more true for C
whereas python with its dictionary lookup on each variable access... idk
hash tables do help though for branch-less (compared to trees)
-> 2.5s
Hellooooo!
I am a beginner at python and I was having trouble downloading a ai software. Is there anyone here I can speak to about it?
what library?
what do you use to download/install it?
Do you mind if we hop in a call? I have been asking chat gpt to help me this whole time and now my fingers are tired of typing lol. If not I can type it out here.
I am very much a beginner so I don't know how to put it all in words
I don't do DM calls but I can attempt helping in text form
I stored cos/sin in a variable instead of calling it over and over
it reduced the execution complexity in 3~5 times
@verbal zenith
how to do angles? or something else?
I am trying to download AI stable fusion. I followed all the steos correctly putting everything into GIThub but when I went to download python I got the wrong version which was the newest one. After I uninstalled the old one I deleted the old path in my environment and downloaded the version I needed. But now the directory is still going to the newest version in python and its giving me a error now
what dimensions do you have now?
[[1, 2, 3],[4, 5, 6]]
reshape: performs in place
resize: returns a new array
[1, 2, 3, 4, 5, 6]
resize doesn't take a lot of time
or...
I don't actually remember
I think it only changes the reference structure
Ill be here whenever youre ready, thank u
do you have a GPU powerful enough to bother installing locally?
I think using their colab notebook would be easier
use venv to avoid breaking everything
if you build something on top of it, then fine
for playing around there's colab
~18s
cut in half
The pathway looks correct on the environment variable. But whenever I hit the webui file it opens up python with the wrong one. If that makes sense
The previous version still shows up on the ui which I dont understand how to change
I apologize I am very new trying to figure it out
it might be having it saved in config/venv
venv "C:\AI-Stable Diffusion\stable-diffusion-webui\venv\Scripts\Python.exe"
No Python at '"C:\Users\Owner\AppData\Local\Programs\Python\Python311\python.exe'
Press any key to continue . . .
is what it says when I open the web ui. My variable says C:\Users\Owner\AppData\Local\Programs\Python\Python310\Scripts\
I should probably add "palette" settings
you can't change python version in an existing venv
how do I change that?
you can't change it
you need to repeat the process that created the venv
to some extent
Do you think it was a problem with how I did the git hub code?
venv gets created with the default python
activate, then install
@verbal zenith
pip install from an activated venv
PyCharm activates by default
there are activation scripts
im so lost lol
venv\Scripts\Activate.ps1
for windows
I didn't know .\venv\Scripts\activate_this.py existed
PyCharm does it best to make you use venvs
Owner@DESKTOP-55AVG4I MINGW64 /c/AI-Stable Diffusion
$ cd C:\AI-Stable Diffusion\stable-diffusion-webui/webui-user
bash: cd: too many arguments
I put this into github when I had the wrong version then I put two files into folders I had downloaded. I believe I cloned it already in github
mixed slashes
also, in mingw it's /c/ not C:
how do create a new venv for that path
I think im going to restart completely on the git hub side of it. $ git clone https://github.com/AUTOMATIC1111/stable-diffusion-webui.git
fatal: destination path 'stable-diffusion-webui' already exists and is not an empty directory.
if I cloned this already what should I do to delete the clone?
docs show python -m venv
maybe pipenv has different
or something
just use WolframScript instead
Mathematica for people without money
civ, yes
I'm applying some OOP abstractions to my code
!e py import numpy as np arr = np.array([0, 255], dtype = np.uint8) arr[0] -= 1 arr[1] += 1 print(arr)
@somber heath :white_check_mark: Your 3.11 eval job has completed with return code 0.
[255 0]
np.uint8
(dtype=np.uint8 or .astype(np.uint8) specifically)
conceptually, bytearray is not an array of uint8
you can get segfault
by breaking things in sys or something
stack overflow is easier to achieve
raise recursion limit
!e
from __future__ import braces
@verbal zenith :x: Your 3.11 eval job has completed with return code 1.
001 | File "<string>", line 1
002 | SyntaxError: not a chance
!e py def func(): try: func() except RecursionError: func() func()
@somber heath :warning: Your 3.11 eval job timed out or ran out of memory.
[No output]
there are different contexts
for StopIteration
in a generator, raising StopIteration causes RuntimeError
to cause StopIteration in the outside code, you need to return not raise
just don't do it in Ruby
(it suffers if you use raise-try-except too much because it actually generates traceback and everything)
!e
def test():
try:
return True
except StopIteration:
return False
print(test())
@verbal zenith :white_check_mark: Your 3.11 eval job has completed with return code 0.
True
try-except on their own don't generate full traceback, iirc
to execute code after return, there is finally
!e
def test():
try:
return True
except StopIteration:
return False
finally:
return 0
print(test())
@verbal zenith :white_check_mark: Your 3.11 eval job has completed with return code 0.
0
return doesn't raise StopIteration inside the function
it raises it when next is called on the generator
!e
def test():
yield
return
gen = test()
next(gen)
try:
next(gen)
except StopIteration:
print('caught')
@vocal basin :white_check_mark: Your 3.11 eval job has completed with return code 0.
caught
!e
# with less next() calls
def test():
return
yield
gen = test()
try:
next(gen)
except StopIteration:
print('caught')
@vocal basin :white_check_mark: Your 3.11 eval job has completed with return code 0.
caught
(swapped return and yield)
!e
def test():
yield
return
def annoy():
try:
next(gen)
except StopIteration:
print('caught')
gen = test()
while True:
annoy()
@verbal zenith :x: Your 3.10 eval job has completed with return code 143 (SIGTERM).
001 | caught
002 | caught
003 | caught
004 | caught
005 | caught
006 | caught
007 | caught
008 | caught
009 | caught
010 | caught
011 | caught
... (truncated - too many lines)
Full output: too long to upload
!e
import ctypes
i = ctypes.c_char('a')
j = ctypes.pointer(i)
c = 0
while True:
j[c] = 'a'
c += 1
j
@verbal zenith :x: Your 3.11 eval job has completed with return code 1.
001 | Traceback (most recent call last):
002 | File "<string>", line 3, in <module>
003 | TypeError: one character bytes, bytearray or integer expected
I've changed how colours mutate
now there's a (global) parameter COPY: bool
I can make it non-global, tbf
copying makes it darker
@eternal anvil π
how to speak
Voice verification
Canβt talk in voice chat? Check out #voice-verification to get access. The criteria for verifying are specified there.
!voice
I moved it to Params class
seems like for shuffle, it mutates twice slower
π @slow ruin
Hello π
I know why it has this colour bias (generating close to r, g, and b separately)
I incorrectly increase the contrast
hello
copying does help to deal with noise somewhat
these abstract pictures have taken over the whole voice text
it is
i meant it in a good way
π.
this uses square root
and this uses square
i have seen this b4...
I have like 0 idea what's going on, does anyone mind telling me?
using square root of brightness as a value passed to PIL is the correct way
(linear is the naive approach which sometimes causes problems)
copying combined with directional
Voice verification
Canβt talk in voice chat? Check out #voice-verification to get access. The criteria for verifying are specified there.
Are you intending to say something or just speaking to yourself, cuz all I can hear is noise :D
I realised why it loops correctly
overflow happens
or no...
no, not that
weird anyway
can't overflow
because addition instead of subtraction
(and uint32 should be enough to fit everything)
although there are negative values
just remembered:
if I generate 5D space, I can make an image with uniform luminosity
oh, not with uniform
with limited
for, uniform 3D is enough
(just normalisation)
hey i need help
What's up?
so i want the turtle to half fill the circle
normalise in 5D -> get 5D shpere
project to 3D -> get filled 3D sphere
@somber heath
they aren't here
here is code
import turtle
t = turtle.Turtle()
t.circle(100)
t.color("green")
t.circle(-100)
t.
thanks @somber heath
(a preview for something)
something that may take up all my RAM
bots aren't magic
ML isn't magic
stocks aren't magic
there also exist people banned from trading
reduced size
6D space
each colour rotates through its own 2D space
idk if anything noticeably changed
I tried to do the 5D->3D projection
(with normalisation somewhere in between)
it took 32 dimensions to make me realise I can't raise contrast by adding dimensions
(only can make the image weirdly darker)
I know that it should get darker
but I forgot how the distribution is expected to behave
@analog stoneπ
I raised the contrast
this: 10D -> 5D -> 3D
ffmpeg -y -i .\out.mp4 reencoded.mp4
out.mp4 is always broken
-vf scale=1920:1080
for rescaling
usually it will change pixels to non-square to preserve the aspect ratio
If one and only one of the values is -n with n >= 1, the scale filter will use a value that maintains the aspect ratio of the input image, calculated from the other specified dimension. After that it will, however, make sure that the calculated dimension is divisible by n and adjust the value if necessary.
-vf scale=-2:1080
idk what to pick for -N
π§€ <- Handbrakes.
this picks at random
this less so
@weary gustπ
hello
something might have went wrong with the colours
I'm trying to make it work with another algorithm
copy + no random swaps
less random version seems fine
@gritty sapphire π
@slow ventureπ
hi
I'm trying to bring back the lightning pattern from this video
#voice-chat-text-0 message
or not "lightning"
idk how to call it
that algorithm is inherently noisy, I think
@ebon willowπ
ποΈ
noise noise noise
if you want some actual server-side rendering, use JS not Python
Python may be okay for APIs
there are better ways
((I think Rabbit can explain how even if you don't want to))
Django and Flask are built on some faulty basis
WSGI, in particular
iirc, Flask uses global variables for request parameters
not even context variables
just globals
if you don't know the law well enough, don't risk it
if you threaten YouTube's profits in any way, then don't do it
another party that gets angry is Spotify which has an official integration with Discord
as in, wherever you play the music from, Spotify will be unhappy
the pixel weirdness is what I'm trying to achieve right now
like this
"The Warp is a strange and terrible place. You might as well throw a traveller into a sea of sharks and tell him to swim home as send him through the Warp unprotected. Better it is not to let common man travel through the stars. Better still, let him not know such a thing is feasible." βFra Safrane, 5th Aide to Navigator De'el The Immaterium, al...
seems like no such pattern
although some remotely similar things do appear
ffmpeg relatively struggled with this one
(twice slower conversion)
testing random parameters (no video)
oh, I forgot, there is video, but no merging
easier to see here that I forgot about that full rotation fill result in colours repeating
half-rotation is enough because (-r, -g, -b) is equivalent to (r, g, b) there
seems like, those patterns are just rare
(there is not guarantee they will occur; it may depend on the per-image parameters like angle deviation)
there isn't a goal or the goal but there are multiple small goals like:
implementing colour rotation
implementing colour projection
implementing support for multiprocessing
optimisation
I had a visualisation thing for projections but I kind of lost it
I might have found images of that
so this is matplotlib.pyplot.scatter of 3D sphere projected into 2D space
4D -> 2D vs 3D -> 2D
@whole bearπ
.
projection from two dimensions up makes the distribution uniform
this is a possible solve to set it to the right path
but It s in the default one
usr/bin
that one
do you have this enabled?
did you restart the integrated terminal?
this is the defualt one
most images that I have of it are just this in different themes (4D -> 2D coloured by the original distance from origin)
does it work outside of VS Code?
enabling this and restarting VS Code should make it work if it works outside of it
it was already applied but still doesn t work
it went bsod
@pseudo crescentπ
Hello there
whats up
could you help me with python im kinda stuck
basically its just a simple input so if somebody wants a translation of a word
so for example from english to dutch
teacher told me i just needed a few words only and this is what i got so far
i just realized i deleted almost everything after i saw almost everything i did was wrong π€£
sorry ill elaborate
this is what i did
OH
aha ok i get that
so instead i should put n = str because its a variable? or did i misunderstood
!e py dictionary = {"apple": "red", "banana": "yellow"} key = "apple" #key = input("Type a fruit") result = dictionary[key] print(result)
@somber heath :white_check_mark: Your 3.11 eval job has completed with return code 0.
red
okok wait let me try it out
so what if i were to do
input and then a direct translation to the word
!e py d = {"apple": "red"} print(d) d["banana"] = "yellow" print(d)
@somber heath :white_check_mark: Your 3.11 eval job has completed with return code 0.
001 | {'apple': 'red'}
002 | {'apple': 'red', 'banana': 'yellow'}
!e py d = {"apple": "red"} d["banana"]
@somber heath :x: Your 3.11 eval job has completed with return code 1.
001 | Traceback (most recent call last):
002 | File "<string>", line 2, in <module>
003 | KeyError: 'banana'
!e py d = {"apple": "red"} a = d.get("apple", "!@#$") b = d.get("banana", "!@#$") print(a) print(b)
@somber heath :white_check_mark: Your 3.11 eval job has completed with return code 0.
001 | red
002 | !@#$
haha you was losing me on this part
oh yh makes sense
oh really?
!e py d = {"apple": "red", "apple": "green"} print(d)
@somber heath :white_check_mark: Your 3.11 eval job has completed with return code 0.
{'apple': 'green'}
Voice verification
Canβt talk in voice chat? Check out #voice-verification to get access. The criteria for verifying are specified there.
@whole bearπ
!rule 5
5. Do not provide or request help on projects that may break laws, breach terms of services, or are malicious or inappropriate.
sorry
hi
ππΌ
@oblique gardenπ
hey π
am new to the server how do i gain permission to speak
Voice verification
Canβt talk in voice chat? Check out #voice-verification to get access. The criteria for verifying are specified there.
ahhh ok rightio thank you π
@somber heath why?
If someone with a neutral to favourable display picture of Hitler came in, I would behave exactly the same way. I choose to avoid you because you have a photograph of a mass murderer in your display picture. Do not speak to me further.
okay lol it's sad you don't get the satire
π maro
hey
eat it
whoah yall left me here
hi
hi
i cant speak
I guess you are new
correct
read pinned text, you basically have to spend some time b4 you can
yes i can
i know, but i still dont have 50 msgs
@lofty crater whats your occupation btw
sfdc dev
salesforce?
yup
cool, webdev? front?
both
react?
lwc a lesser known framework
interesting, i learning data science btw
no, im new at DS, now, im basically doin math
ive got good mentor so i decided to learn
through tg i think
I was asking if you'd found your way to the #voice-verification channel.
yes, ive found but i Still dont have 50 msgs
So I don't know what you meant by this.
be careful not to spam or you'll get banned π
Oh, i thought you said "i hope you got voice verification"
yeah, i will try
@somber heath i think ive seen you there before
Hello maro π
@tawdry trench π
davinki?
marolokio?
oh ok
π
mr banaccio
banani
bananini
maronaΓ±o
thats sick
rip py launcher
try checking in control panel if you still have python
u need to be voice verified
ok i pull up
i speek german to
@midnight agate woah your too smart
π
sick
i would buy these 100%
yes?
shutup n take my money
help me
no clue
@midnight agate so casa is house in italian?
casinha
casa nostra?
casΓ£o
@wispy forge π
P(no German-speaking person) = (1 - P(any individual speaks German)) ^ number_of_people
Assuming independence.
Now what's the probability that any two people in voice-chat share a birthday? π
Fair enough
Err, sorry I didn't catch the question?
Oh right yeah π
My favourite question like that is: You flip a fair coin 10 times, and the first 9 times it comes up heads. Is it more likely to come up heads or tails on the 10th flip?
In reality, yes lol
But given the assumption of "fair coin".
That was a very Bayesian answer Maro.
You're treating the probability of heads as a random variable.
Yeah. I mean eventually you have to make assumptions. Even assuming the flips are completely independent might not be entirely valid!
Randomness, coins and dental floss!
Featuring Professor Persi Diaconis from Stanford University.
More links & stuff in full description below βββ
To catch or not to catch: http://youtu.be/Obg7JPd6cmw
Extra footage (Numberphile2): http://youtu.be/9RKKoXw7wJw
Superbowl coin tosses: http://youtu.be/bFfSfzjhfC8
Paper shown in this video (Dynamical...
Just think of all the questions that arise when a seemingly-healthy baby dies. Was there an illness that went undetected? Was there a rare genetic defect that couldnβt possibly have been known? Are any signs of physical trauma related to the resuscitation efforts of a fragile infant, or something more sinister? Every time a baby dies, police inv...
Here's an interesting one: I just flipped a coin. What's the probability that it's heads?
For you, but not for me, right?
Because I can see what it is.
It must be either 100% or 0% for me.
But for you it's 50%
So probability is subjective.

I didn't have any coffee today, which might be why I feel pretty tired now that I think about it lol 
sad
@whole bear π
Err, I didn't follow sorry.
Sorry, I'm just not sure what you're asking.
I'm completely blind when it comes to this kind of thing lol.
Oh you keep adding to PATH?
Ah right.
Errm, I probably missed that sorry.
Where above?
Pinging modmail won't actually ping the mods sorry.
You need to either DM @rapid crown or ping @moderators.
Yeah sorry about that, I don't think any mod would have seen that.
Erm, I'll make a note or something.
Doesn't affect me much seeing as I don't have a job lol
Working the remote
I have a thought about this actually.
Imagine we put last week's lottery winner on trial, accused of fraud (cheating).
As the only evidence, we present the fact that they won the lottery.
The probability of winning the lottery is 1 in 1 million (or whatever).
So there's a 999,999 in a million chance they cheated, right?
I think the issue is essentially double-spending information. We're using the same information to both identify a suspect and verify their guilt.
Anyway, gtg, cya π
that was an interesting story
@cursive sundial if you want help, see #βο½how-to-get-help. don't ask people to DM you. and don't spam a bunch of channels with the same message
!unmute 1040377694673711155
:incoming_envelope: :ok_hand: pardoned infraction mute for @cursive sundial.
ok
Whoever it was that just joined and left, I was occupied in that moment by something else. Sorry I didn't say hi. π
From his new album Solid State, out April 28.
The album has a companion graphic novel written by Matt Fraction and drawn by Albert Monteys. It's a science fiction story about the internet, the future, artificial intelligence, and how probably only love will save us.
The book, CD, Vinyl, Tshirts, and deluxe bundles of fantastic things can be pr...
important thing that tutor is expected to provide is oversight
ChatGPT can't
it itself needs a lot of oversight
@lucid blade
steganography
https://en.wikipedia.org/wiki/Shorthand Stenography
https://en.wikipedia.org/wiki/Steganography Steganography
and about injecting things to detect text being AI-generated
https://www.youtube.com/watch?v=XZJc1p6RE78
Mike explains a paper from the University of Maryland, proposing a neat trick to 'watermark' the output of large language models such as ChatGPT. Dr Mike Pound is an image analyst at the University of Nottingham.
The University of Maryland paper: https://arxiv.org/pdf/2301.10226.pdf
https://www.facebook.com/computerphile
https://twitter.com...
you there @lofty crater ?
I am here now
Here's a program in Python that finds an element in an array that is smaller than all elements to its right and greater than all elements to its left:
L = [3,1,2,5,7,6,9]
output all such elements or one/zero?
for the time being only one is needed
write the naive solution first
L = [2,1,3,5,8,7,9]
for i in range(1 ,len(L)-1):
if max(L[:i]) < L[i] < min(L[i+1:]):
print(L[i])```
now find a way to compute max and min faster
hint: given max(L[:i]) and L[i+1] you can get max(L[:i+1])
L = [3,1,2,5,8,7,9]
max_arr = []
min_arr = []
max_val = L[0]
min_val = L[-1]
for i in L:
if i>max_val:
max_val = i
max_arr.append(max_val)
for i in range(len(L)-1,-1,-1):
if L[i] < min_val:
min_val = L[i]
min_arr.insert(0,min_val)
for i in range(1,len(L)-1):
if max_arr[i-1] < L[i] < min_arr[i+1]:
print(L[i]) ```
all i need is an alternative way
if any
otherwise i will take it :]
@devout mica lil bit of disturbance from your end is audible
if you could fix it ...:}ty
Hey
I'm sure dynamic programming is the only viable solution
depends
for competitive/industrial usecases you'd usually need O(N) not O(N^2)
if you can do O(N) and it doesn't produce too much issues, use O(N)
for this problem O(N) solution uses three times more memory
or two times if you write it differently
depends
depends on the requirements
ohk understood