#algos-and-data-structs | Python | Page 82

gusty grove Jul 7, 2020, 2:37 PM

#

to be fair break break break seems weird

cursive cloud Jul 7, 2020, 2:37 PM

#

            
def recursion_starter(rest_number_of_stairs: int) -> int:
    number_of_ways = 0
    def recursive_climb(rest_number_of_stairs: int):
        if rest_number_of_stairs == 0:        
            nonlocal number_of_ways
            number_of_ways += 1
        elif rest_number_of_stairs == 1:
            recursive_climb(rest_number_of_stairs - 1) # one step
        else:
            recursive_climb(rest_number_of_stairs - 1) # one step
            recursive_climb(rest_number_of_stairs - 2) # two steps

    return number_of_ways

class ClimbingStairs:
    @staticmethod
    def climbing_stairs(number_of_stairs: int) -> int:
        if number_of_stairs <= 0:
            return 0
        else:
            return recursion_starter(number_of_stairs)

stable pecan Jul 7, 2020, 2:37 PM

#

def func():
    count = 0
    def new_func():
        nonlocal count
        def new_new_func():
            nonlocal count
...

cursive cloud Jul 7, 2020, 2:37 PM

#

You mean like this?

candid osprey Jul 7, 2020, 2:37 PM

#

Feels like "break gosh darn it"

stable pecan Jul 7, 2020, 2:38 PM

#

this might work, but it's something i've never seen

#

i somehow doubt it will though

cursive cloud Jul 7, 2020, 2:39 PM

#

Because it seems to not change number_of_ways

stable pecan Jul 7, 2020, 2:39 PM

#

you only increment it on == 0

#

you should declare nonlocal outside of the if statements probably

cursive cloud Jul 7, 2020, 2:40 PM

#

Yes I had it befire the if's but same result

#

why is incrementint it on == 0 a problem?

#

It worked before

#

because it should always reach == 0

#

It seems nonlocal does not work as expected

candid osprey Jul 7, 2020, 2:42 PM

#

I finally understand why this function kept bothering me.

#

Why aren't we accumulating and adding number of steps?

#

Like an acc parameter that you keep passing through all recursive calls

cursive cloud Jul 7, 2020, 2:43 PM

#

because I don't want to count the number of steps

candid osprey Jul 7, 2020, 2:43 PM

#

Sadly my brain can't think recursion

cursive cloud Jul 7, 2020, 2:43 PM

#

I want to calculate how many ways their is to climb the stairs

candid osprey Jul 7, 2020, 2:44 PM

#

My brain keeps thinking that shouldn't matter.

#

I'll tell my brain to go quiet. Carry on

cursive cloud Jul 7, 2020, 2:45 PM

#

that is the task, and the first code I posted does work. But I wanted to know if there is a better way

stable pecan Jul 7, 2020, 2:46 PM

#

>>> def stairs(n):
...     if n <= 1:
...             return 1
...     return stairs(n - 1) + stairs(n - 2)
... 
>>> stairs(10)
89
>>>

#

is that right?

candid osprey Jul 7, 2020, 2:48 PM

#

What's the output of stairs 3?

cursive cloud Jul 7, 2020, 2:48 PM

#

Oh yes this seems to work too

stable pecan Jul 7, 2020, 2:48 PM

#

3

candid osprey Jul 7, 2020, 2:49 PM

#

Pretty.

stable pecan Jul 7, 2020, 2:49 PM

#

it's like reverse fibonacci

#

look familiar?:

>>> stairs(3)
3
>>> stairs(4)
5
>>> stairs(5)
8
>>> stairs(6)
13

#

it's not reverse! it is fibonacci

cursive cloud Jul 7, 2020, 2:53 PM

#

nice

#

And I found the problem with my code

#

Forgot to call the recursive function

#

def recursion_starter(number_of_stairs: int) -> int:
    number_of_ways = 0

    def recursive_climb(rest_number_of_stairs: int):
        if rest_number_of_stairs == 0:
            nonlocal number_of_ways
            number_of_ways += 1
        elif rest_number_of_stairs == 1:
            recursive_climb(rest_number_of_stairs - 1) # one step
        else:
            recursive_climb(rest_number_of_stairs - 1) # one step
            recursive_climb(rest_number_of_stairs - 2) # two steps
    
    recursive_climb(number_of_stairs)
    return number_of_ways

#

That worked

#

Thanks guys

quasi oracle Jul 7, 2020, 4:30 PM

#

@cursive cloud you don't need an outer function or nonlocal or anything

def recursive_climb(rest_number_of_stairs: int):
    if rest_number_of_stairs == 0:
        return 1
    elif rest_number_of_stairs == 1:
        return recursive_climb(rest_number_of_stairs - 1)
    else:
        return recursive_climb(rest_number_of_stairs - 1) + recursive_climb(rest_number_of_stairs - 2)

#

This is exactly the same

#

Which you then can turn into what salt-die wrote

tribal pendant Jul 7, 2020, 4:35 PM

#

that complexity...

shrewd rampart Jul 7, 2020, 5:42 PM

#

what state of the art algorithms?
@marble basalt the usual, GA, PSO, CA

marsh scroll Jul 7, 2020, 11:53 PM

#

@cursive cloud you don't need an outer function or nonlocal or anything

def recursive_climb(rest_number_of_stairs: int):
    if rest_number_of_stairs == 0:
        return 1
    elif rest_number_of_stairs == 1:
        return recursive_climb(rest_number_of_stairs - 1)
    else:
        return recursive_climb(rest_number_of_stairs - 1) + recursive_climb(rest_number_of_stairs - 2)

@quasi oracle
How do I post my codes like this?

patent pond Jul 7, 2020, 11:56 PM

#

use the `

#

surround your code in `` and itll do it

#

like this

#

Not quite sure how you get the colors

marble basalt Jul 8, 2020, 12:29 AM

#

you put py after the triple ``` and a newline

urban spear Jul 8, 2020, 12:29 AM

#

the colors are for different languages of code i think

marble basalt Jul 8, 2020, 12:29 AM

#

```py
print("stuff")
```

urban spear Jul 8, 2020, 12:29 AM

#

py for python

marble basalt Jul 8, 2020, 12:29 AM

#

print("stuff")

urban spear Jul 8, 2020, 12:30 AM

#

theres a lot of supported languages like java, c+, html, css, etc

#

wait maybe c+ isnt supported

#

theres a website that lists all of the supported languages

knotty rapids Jul 8, 2020, 4:05 AM

#

hello

#

i am using git

#

and i want to update the changes i made into the branch other than master

#

but when i do
git checkout <branch-name>

#

it tells me i have unsaved changes i need to commit them first

#

but commiting changes would update the changes in master branch

#

how do i solve this

#

?

chrome lava Jul 8, 2020, 4:12 AM

#

Shouldn't it be git checkout -b <branch-name>?

fossil elk Jul 8, 2020, 4:12 AM

#

that happened because you worked in master before @knotty rapids

#

probably you can use git stash first, then do this

Shouldn't it be git checkout -b <branch-name>?
@chrome lava

knotty rapids Jul 8, 2020, 4:14 AM

#

okay

#

@fossil elk isn't -b for a new branch ?

fossil elk Jul 8, 2020, 4:15 AM

#

yea

knotty rapids Jul 8, 2020, 4:15 AM

#

but i have a branch

fossil elk Jul 8, 2020, 4:15 AM

#

if it already exist, you dont have to -b

knotty rapids Jul 8, 2020, 4:15 AM

#

okay

#

thanks

fossil elk Jul 8, 2020, 4:16 AM

#

no problem

knotty rapids Jul 8, 2020, 4:16 AM

#

😄

glossy rose Jul 8, 2020, 8:21 AM

#

how is this

torn scarab Jul 8, 2020, 9:04 AM

#

!warn 555874683086700551 This is not the place for ASCII spam. Do not send irrelevant messages to on-topic channels.

halcyon plankBOT Jul 8, 2020, 9:04 AM

#

:incoming_envelope: :ok_hand: applied warning to @glossy rose.

fervent wagon Jul 8, 2020, 9:25 AM

#

https://www.youtube.com/watch?v=EK32jo7i5LQ&t=693s

Not sure if this is the right channel to ask this question, but i am currently doing an assignment and i need to do this spiral graph by following the youtube link as attached. How do i start?

📎 image-20191010131044-1.png

YouTube

3Blue1Brown

Why do prime numbers make these spirals?

A story of mathematical play.
Home page: https://www.3blue1brown.com
Brought to you by you: http://3b1b.co/spiral-thanks

Based on this Math Stack Exchange post:
https://math.stackexchange.com/questions/885879/meaning-of-rays-in-polar-plot-of-prime-numbers/885894

Want to lear...

▶ Play video

mint jewel Jul 8, 2020, 9:34 AM

#

I would start by figuring you how to draw circles

#

there are quite a few options: Processing.py, pygame, pillow, matplotlib, pyglet, blender, arcade, ...

rain relic Jul 8, 2020, 12:04 PM

#

anyone got a tldr on the chomsky hierarchy?

exotic hatch Jul 8, 2020, 12:11 PM

#

formal languages fall into different categories, depending on the formal automata that generate them

it goes regular < context free < context sensitive < recursive

#

and those correspond to finite acceptors, pushdown automata, queue automata, and turing machines

#

@rain relic

rain relic Jul 8, 2020, 12:13 PM

#

what are the main differences between these 4 levels of the hierarchy though

#

are they equivalent?

exotic hatch Jul 8, 2020, 12:13 PM

#

they contain different languages

#

the smaller ones are included in the bigger ones

rain relic Jul 8, 2020, 12:14 PM

#

Aha

exotic hatch Jul 8, 2020, 12:14 PM

#

regular languages contain every finite language and every infinite language generated by a regular expression

#

context free contain every language generated by context free grammars

#

context sensitive contains every language generated by context sensitive grammar

#

and recursive languages are accepted by turing machines

#

turing machine are a superset of pushdown automata

#

pushdown automata is a superset of finite acceptors

rain relic Jul 8, 2020, 12:16 PM

#

Thanks for the overview, the lack of theory of computation at my undergrad keeps me wanting to learn more of it on my own

exotic hatch Jul 8, 2020, 12:16 PM

#

📎 unknown.png

rain relic Jul 8, 2020, 12:16 PM

#

thanks again

exotic hatch Jul 8, 2020, 12:17 PM

#

good luck with your studies, here's all the facts from automata theory in 2 pictures

📎 unknown.png

fiery cosmos Jul 8, 2020, 12:59 PM

#

Can someone tell me how to fix these rounding issues?
pretty much every calculation ends up with x.9999999999999 or something similar, even when i subtract 0.0499999.. from 0.95 i get 0.899999.. wich makes absolutely no sense to me

mint jewel Jul 8, 2020, 1:00 PM

#

https://0.30000000000000004.com/

#

essentially, floating points are not great for exact math. You can use either decimal.Decimal or fractions.Fraction

fiery cosmos Jul 8, 2020, 1:06 PM

#

My numbers dont have to be very precise, is there any way to influence the output to round to the nearest number with k decimals or something?

mint jewel Jul 8, 2020, 1:07 PM

#

You can use the .3f format specifier to round to that many digits

fiery cosmos Jul 8, 2020, 1:08 PM

#

does that also apply to lists? eg print(x.3f) with x being a list of floats it doesnt appear to work

mint jewel Jul 8, 2020, 1:09 PM

#

no, you will have to do something like

print([format(x, '.3f') for x in some_list])

fiery cosmos Jul 8, 2020, 1:09 PM

#

ah nice tyvm

cyan gull Jul 8, 2020, 4:02 PM

#

Can someone help with c xD

#

Idk how to check if the inputted string has duplicate chars

mint jewel Jul 8, 2020, 4:04 PM

#

if you know you only get trivial ascii (0 - 128), you could make a 128 long array of 0s and put a char there when you find it. If you ever try to place a character twice, it is duplicate

#

essentially a simpler hash table

#

where the hash function is identity

cyan gull Jul 8, 2020, 4:06 PM

#

Sry im new to this and im not rly sure wut u mean

#

📎 Screenshot_20200708-120518.jpg

#

This is what i tried

#

It is 26 characters hence the 13

#

Bc i didnt want the chars to meet in the middle and be equal

#

Although ik theres most definately a better way to go about this and i probably messed up the loops😭

oblique panther Jul 8, 2020, 4:13 PM

#

If you're not required to do it in a C style language and you can use any builtin python functionality, there's a simpler way to do it.

cyan gull Jul 8, 2020, 4:13 PM

#

I wish, unfortuantely this must be done in c..

oblique panther Jul 8, 2020, 4:14 PM

#

This looks like Java though.

mint jewel Jul 8, 2020, 4:14 PM

#

That is c

#

Java has no argv and no printf

oblique panther Jul 8, 2020, 4:14 PM

#

I thought you can't define loop variables in the for loop declaration.

#

In C.

mint jewel Jul 8, 2020, 4:14 PM

#

Since C99 or so you can

oblique panther Jul 8, 2020, 4:14 PM

#

Well big thanks to my C professor.

cyan gull Jul 8, 2020, 4:14 PM

#

xD

mint jewel Jul 8, 2020, 4:15 PM

#

People have outdated knowledge of C and C++, and java, for that matter, quite often

cyan gull Jul 8, 2020, 4:17 PM

#

Python is much more effective to use right?

mint jewel Jul 8, 2020, 4:17 PM

#

Eh, it depends. For most things, python will be much easier than C

cyan gull Jul 8, 2020, 4:17 PM

#

Im assuming they teach c to beginners in this course so that i have to suffer with typing out stuff that other languages would have as a command? XD

#

And understand the logic behind it

#

Idk tho

mint jewel Jul 8, 2020, 4:19 PM

#

C is a solid starter language because it teaches you about the foundations of computing with memory management, pointers and such nonsense, while being pretty high level

oblique panther Jul 8, 2020, 4:19 PM

#

Python is much more effective to use right?
@cyan gull

A lot of people feel that Python more effectively communicates what the code is doing, since that was one of its most important design principles. It's slow compared to C though.

cyan gull Jul 8, 2020, 4:20 PM

#

Ah i see

#

Also i was just wondering

#

They make a big emphasis

#

On memory and stuff when creating variables

#

How often is that knowledge really applied

#

When coding

#

Is it really nessacary to know that much in detail?

oblique panther Jul 8, 2020, 4:21 PM

#

Well, it you're not worried about memory management, it's because the language is handling it for you.

#

C doesn't.

mint jewel Jul 8, 2020, 4:22 PM

#

There are cases for handling your own memory. Most people write C++ or rust, which has nicer memory management than C

cyan gull Jul 8, 2020, 4:23 PM

#

Alright

#

Would C be considered sort of irelevant with all these new languages and c#?

#

Sry for all the questions

oblique panther Jul 8, 2020, 4:24 PM

#

C# isn't meant for the same use cases as C

#

C++ sort of is.

cyan gull Jul 8, 2020, 4:24 PM

#

C++*

oblique panther Jul 8, 2020, 4:25 PM

#

C# is basically Windows Java.

cyan gull Jul 8, 2020, 4:25 PM

#

Ohhh thanks for clearing that up

mint jewel Jul 8, 2020, 4:25 PM

#

These days, there are efforts at replacing C with languages like Zig, V, Go, .... There is no real reason to use C these days beyond short snippets which must go fast

oblique panther Jul 8, 2020, 4:26 PM

#

I would only use C to make python modules

fiery cosmos Jul 8, 2020, 4:26 PM

#

Hello

oblique panther Jul 8, 2020, 4:26 PM

#

And even then that's not the same as C because the Python API for C does a lot of the work.

mint jewel Jul 8, 2020, 4:27 PM

#

Simple take input, do math, write output programs are pretty simple in C and do often benefit from performance. There is a difference in feeling between something taking .5s And .03s

cyan gull Jul 8, 2020, 4:28 PM

#

Alright thx guys for the help

oblique panther Jul 8, 2020, 4:28 PM

#

@fiery cosmos hi

#

This channel is specifically for talking about computer science. python -i in your terminal will open an interactive session.

fiery cosmos Jul 8, 2020, 4:32 PM

#

OK

oblique panther Jul 8, 2020, 4:33 PM

#

Take a look at #❓｜how-to-get-help if you have more questions

nocturne linden Jul 9, 2020, 5:15 AM

#

C# is basically Windows Java.
@oblique panther lmao, so true tho

glossy rose Jul 9, 2020, 5:18 AM

#

aactually yes he was ryte'

#

Csharp is windows java

torn scarab Jul 9, 2020, 5:48 AM

#

!tempmute 555874683086700551 7d You have been told repeatedly to not interrupt on-topic channels, yet you continue to post unproductive messages. Take the time off and review what each of our channels are for by reading the channel description.

halcyon plankBOT Jul 9, 2020, 5:48 AM

#

:incoming_envelope: :ok_hand: applied mute to @glossy rose until 2020-07-16 05:48 (6 days and 23 hours).

nova shard Jul 9, 2020, 7:26 AM

#

def minimax(move, depth, maximizingPlayer):
    if depth == 0:
        move.eval = calculate(move)
        return move

    if maximizingPlayer:
        maximizedMove = Move(-math.inf)
        for m in get_possible_moves(move.board):
            evaluatedMove = minimax(m, depth - 1, False)
            maximizedMove = evaluatedMove if evaluatedMove.eval > maximizedMove.eval else maximizedMove
        return maximizedMove
    else:
        minimizedMove = Move(math.inf)
        for m in get_possible_moves(move.board):
            evaluatedMove = minimax(m, depth - 1, True)
            minimizedMove = evaluatedMove if evaluatedMove.eval < minimizedMove.eval else minimizedMove
        return minimizedMove

Hey guys! I've been trying to make this minimax algorithm work for quite some time, it seems to work well to determine best moves, but the thing is when the function is over it returns the last move that was played, but I want to receive the move that I need to play right now to follow the path determined...

fiery cosmos Jul 9, 2020, 11:04 AM

#

can someone help with RL and time series?

astral path Jul 9, 2020, 11:36 AM

#

Guys, what is the best way to find the index of the lowest integer in a list?

trail anchor Jul 9, 2020, 12:25 PM

#

l = [3, 4, 2, 1, 9]
l.index(min(l))

tribal pendant Jul 9, 2020, 12:32 PM

#

this is inefficient

#

potentially you loop one time to find the min, and then another time to find the index

trail anchor Jul 9, 2020, 12:36 PM

#

maybe a reduce-type of method where u keep the index with the lowest value and return?

mint jewel Jul 9, 2020, 12:36 PM

#

this would work, but it is quite ugly min(map(reverse, enumerate(l)))[1]

#

or, for lists only, min(range(len(l)), key=l.__getitem__)

tribal pendant Jul 9, 2020, 12:38 PM

#

if you want to understand what's your doing, you could create your own function, but I guess it'll slower (?) than using builtins

#

something like that

def imin(l):
  m = l[0]
  m_i = 0
  for i, x in enumerate(l):
    if x < m:
      m, m_i = x, i
  return m_i

trail anchor Jul 9, 2020, 12:44 PM

#

tx

#

took me a second but I get min(range(len(l)), key=l.__getitem__) and thanks for the long-hand explanation too

candid bane Jul 9, 2020, 4:33 PM

#

@astral path You have to iterate over all elements.
Note that @trail anchor will iterate over the list twice, so if performance is a concern, you might want to iterate over the list yourself rather than using min and index.

astral path Jul 9, 2020, 4:36 PM

#

i just found out that in the list i have some integers that are equal to eachother, will that complicate things?

candid bane Jul 9, 2020, 4:39 PM

#

It depends if you only want the first one, last one, any of them or all of them. And either cases, it won't complicate things, just change the code. The usual behavior is to return the lowest index element.

astral path Jul 9, 2020, 4:42 PM

#

okay, the usual behaviour will be enough

#

efficiency will also not be much of a problem

#

this is just practice for a competition so yeah i got time as long as i get it to work

fiery cosmos Jul 9, 2020, 9:49 PM

#

If I want to become better at thinking recursively and writing recursive functions, should I read more python or try some Project Euler exercises in erlang?

cyan gull Jul 9, 2020, 10:52 PM

#

Can anyome help with a basic c question pls

quasi oracle Jul 9, 2020, 10:53 PM

#

What do you mean by reading more python? I don't know if project euler specifically is the best place, but exercising is key @fiery cosmos

#

@cyan gull this is a python server

cyan gull Jul 9, 2020, 10:54 PM

#

Yea ik xD i was just wondering lmao nvm then

fiery cosmos Jul 9, 2020, 10:54 PM

#

Seeking out more recursion examples and recursion challenges.

#

And reading python solutions to them to understand them when I'm stumped

quasi oracle Jul 9, 2020, 10:55 PM

#

There are some classic recursion examples anyone should know. If you're familiar with them, you should try solving problems on your own

fiery cosmos Jul 9, 2020, 10:56 PM

#

That's been my workflow so far. But I remember going to a conference about 8 years ago and attended an Erlang lab. I remember doing well on the exercises because Erlang doesn't have loop constructs. It forced me to think functionally and recursively versus defaulting to loops and iterators.

#

What are some of the examples you mention @quasi oracle ? Like BST traversal?

quasi oracle Jul 9, 2020, 10:59 PM

#

Anything from summing sequences (arithmetic, fibonacci etc) to BST traversal, DFS, and similar problems

#

You can open a help channel if you're struggling with something specific, but you need to try to get through it without looking at a solution (assuming you didn't jump right into super complicated stuff)

fiery cosmos Jul 10, 2020, 2:23 AM

#

Why does this:
print([[list]])
result in this:
[[array(['a', 'b'], dtype='<U15')]]
why wont it just output it without all the nonsense?

fiery cosmos Jul 10, 2020, 2:51 AM

#

I don't have that problem. Prints w/ expected behavior.

#

using numpy?

#

yes

#

def rec_perm(list):
    res = []
    if len(list) <= 2:
        print([[list]])
        return [[list]]
    for i in range(len(list)-1):
        touple = [list[0], list[i+1]]
        tmp = rec_perm(np.delete(np.delete(list, i+1), 0))
        for k in range(len(tmp)):
            tmp[k].append(touple)
        for x in tmp:
            res.append(x)
    return res```
full code

stable pecan Jul 10, 2020, 5:42 AM

#

this is nearly unreadable, what is it supposed to be doing? and if you're doing this much appending/deleting i'd just not use numpy --- it won't be faster, especially with all these python loops

#

numpy creates an new array for each delete

cyan current Jul 10, 2020, 12:30 PM

#

hello everyone, I am new to this server and to python. I graduated in IT majoring in web & s/w dev in 2014 but feel like I dont know even basic coding concepts. I have enrolled in a DS course recently and committed to improving my coding skills.
I am having trouble understanding the use of flags in the code flow below:

#write a program to print prime number b/w 1 to 20

for n in range(2,20):
    flag = True
    for i in range(2,n):
        if n % i == 0:
            flag = False
            break
    if flag :
        print(n)

When I try to run below code without the use of flads, I get even numbers as output:

#write a program to print prime number b/w 1 to 20

for n in range(2,20):
    for i in range(2,n):
        if n % i != 0:
            break
            print(n)```
Any help is appreciated. Thank you.

visual dagger Jul 10, 2020, 12:46 PM

#

you should try using
l = []
x >1
for i in range(2,20):
if x % i != 0:
l.append(i)
print("these are the prime numbers between 2 and 20",l)

topaz pulsar Jul 10, 2020, 12:57 PM

#

@cyan current i don't think you should get any ouput from the bottom code as the print is below a break so its unreachable, instead of using flags, you can use a similar aproach as the top but instead of having a flag, you can have an else or that for loop, that will only be run if there was no break so you can remove the flag variable all together end replace if flag: with else

#

however, a faster method of finding primes upto a limit is the Sieve of Eratosthenes, its not too complicated and should be a faster alternative for larger values if you are interested in that aproach

cyan current Jul 10, 2020, 1:45 PM

#

@visual dagger I tried that code but line x>1 is throwing error TypeError: '>' not supported between instances of 'list' and 'int' I tried declaring it as range(2,20) and list but it did not work. I will try @topaz pulsar 's approach now.
I really appreciate the help from you both. Thank you

fiery cosmos Jul 10, 2020, 2:41 PM

#

hello everyone, I am new to this server and to python. I graduated in IT majoring in web & s/w dev in 2014 but feel like I dont know even basic coding concepts. I have enrolled in a DS course recently and committed to improving my coding skills.
I am having trouble understanding the use of flags in the code flow below:
#write a program to print prime number b/w 1 to 20

for n in range(2,20):
    flag = True
    for i in range(2,n):
        if n % i == 0:
            flag = False
            break
    if flag :
        print(n)
When I try to run below code without the use of flads, I get even numbers as output:
#write a program to print prime number b/w 1 to 20

for n in range(2,20):
    for i in range(2,n):
        if n % i != 0:
            break
            print(n)```
Any help is appreciated. Thank you.

@cyan current If a number n is divisible by any of the numbers 2 up to itself, it is not a prime number, as it is a number only divisible by 1 and itself.
The flag is there such that the non prime number will not be printed by the program (after it breaks from the inner loop because we found one counter example, and thus there is no reason to continue).
It might help if you write out the loops and try to trace the numbers and do the conditional checking for yourself to get more feeling for it

#

although renaming flag to prime or is_prime might have made it more descriptive

visual dagger Jul 10, 2020, 5:23 PM

#

@visual dagger I tried that code but line x>1 is throwing error TypeError: '>' not supported between instances of 'list' and 'int' I tried declaring it as range(2,20) and list but it did not work. I will try @topaz pulsar 's approach now.
I really appreciate the help from you both. Thank you
@cyan current mine loginc was that taking x as greater than one so that I can divide that "i" from the "x" to find if the remainder is zero or not
I'll try to correct my error

#

@cyan current If a number n is divisible by any of the numbers 2 up to itself, it is not a prime number, as it is a number only divisible by 1 and itself.
The flag is there such that the non prime number will not be printed by the program (after it breaks from the inner loop because we found one counter example, and thus there is no reason to continue).
It might help if you write out the loops and try to trace the numbers and do the conditional checking for yourself to get more feeling for it
@fiery cosmos HEY you should really try some short course on python by watching it on youtube or as you like
and for the code you asked about is that
the "Flag" is just a variable that is set to be "True" and when the condition "if n % i ==0" satisfies, then the value of "Flag" is changed to be "False"
and the below code that you tried to run has the wrong indentation and wrong conditions to be fulfilled so it will definately not run

summer reef Jul 10, 2020, 6:33 PM

#

you can have an else or that for loop, that will only be run if there was no break
@topaz pulsar
That's not quite right : an else statement after a loop will always be executed after the last iteation ; in fact, it's the same behaviour as plain code below the loop

#

!e

for x in range(10) :
  print(x)
else :
  print("Else block executed")
print("Out of the loop")

halcyon plankBOT Jul 10, 2020, 6:34 PM

#

You are not allowed to use that command here. Please use the #bot-commands channel instead.

summer reef Jul 10, 2020, 6:35 PM

#

crap
well, it should display

0
1
2
3
4
5
6
7
8
9
Else block executed
Out of the loop

mint jewel Jul 10, 2020, 6:38 PM

#


In [46]: for x in range(10) :
    ...:   print(x)
    ...: else :
    ...:   print("Else block executed")
    ...: print("Out of the loop")
0
1
2
...
7
8
9
Else block executed
Out of the loop

In [47]: for x in range(10) :
    ...:   print(x)
    ...:   break
    ...: else :
    ...:   print("Else block executed")
    ...: print("Out of the loop")
0
Out of the loop
```nah, he is right

summer reef Jul 10, 2020, 11:19 PM

#

All right then, I learned something today ! I thought it was kind of like the finally statement, well, it's not 😉

magic light Jul 11, 2020, 12:11 AM

#

When reading through design pattern literature, I always see the term "concrete class" but, through context, learned it means "a class that uses implements all of its methods"

But then you also find "concrete implementations/methods" that kind of relate (but not really)

Where did this "concrete" term come from and why is it used but never explained?

quasi oracle Jul 11, 2020, 12:51 AM

#

I can't answer why you haven't seen an explicit explanation, but the gist is that you have the "normal" concrete class, which as you said implements all its methods, and then you have an abstract class. An abstract class is any class that has at least one abstract method.
As a typical example, you might have the base class of Animal, and the classes Dog and Duck that inherit from it.
Those two latter classes inherit a function called "make noise", and each overrides it to match the desired behavior. Meaning, a dog will bark, and a duck will quack.
But what noise will a general animal make? there's no meaningful answer to that. You created that method inside the Animal class, to make sure that any class that inherits from it will also have that method.
In that context, the "make_noise" method in the Animal class will be defined as abstract (meaning you can't call it), which will make the Animal class abstract (meaning you can't initialize an object of that class). In order for Dog and Duck to not be abstract (meaning, to be concrete), they have to override the method and implement it properly.

#

A search on abstract classes in python (or not necessarily in python) should yield quite a few results

cyan current Jul 11, 2020, 12:35 PM

#

@visual dagger @fiery cosmos thank you for the explanation guys! I understand it properly now.

magic tendon Jul 11, 2020, 4:13 PM

#

Can someone help me with this program, I'm a complete newbie so i don't know what to do but i realise that the error might lie in the filename

#

def WtoH():

p=open("student1.tst","w")
Rollno=int(input("enter the roll no"))
name=input("Student name")
Mark=float(input("Enter the mark"))
record=str(Rollno)+","+name+","+str(Mark)+"\n"
p.write(record)
p.close()

def RfromH(c):

q=open("student1.tst","r")
str=" "
while str:
    str=q.read()
    print(str)
q.close()

WtoH()
RtoH()

tribal pendant Jul 11, 2020, 4:32 PM

#

not the right channel, see #❓｜how-to-get-help

lavish fox Jul 11, 2020, 5:32 PM

#

@magic tendon
You have made function RfromH() but called RtoH().
Also u have given c in RfromH but didn't passed anything while calling it

fiery cosmos Jul 11, 2020, 10:14 PM

#

Can someone tell me why i get a TypeError wich states that "key" is not a keyword argument for sort() even tho it should be?

np.vstack(([x for x in range(25)],priority)).T.sort(key=lambda x: x[1])```

#

my reference: https://www.programiz.com/python-programming/methods/list/sort

bronze gazelle Jul 11, 2020, 10:19 PM

#

Do you have a function named sort?

fiery cosmos Jul 11, 2020, 10:19 PM

#

i thought its an inbuild function or am i wrong?

bronze gazelle Jul 11, 2020, 10:20 PM

#

Oh, I missed the numpy part

#

Your reference is for a list, not a numpy array

#

https://numpy.org/doc/stable/reference/generated/numpy.ndarray.sort.html

fiery cosmos Jul 11, 2020, 10:21 PM

#

oh can i convert that somehow? this numpy stuff really screws me over all the time :/

#

ah nvm ty for the link 🙂

#

np.vstack(([x for x in range(25)],priority)).T.sort(axis=0)```
gives me `None` as a result. idk why this is so difficult

quasi oracle Jul 11, 2020, 10:36 PM

#

As it should. sort changes your array in-place

#

You can use np.sort instead

fiery cosmos Jul 11, 2020, 10:38 PM

#

oh mb ty

fiery cosmos Jul 11, 2020, 11:03 PM

#

so i found this expression a[a[:,1].argsort()]to sort a 2d-array by its second column, a[:,1] selects the column and a[:,1].argsort() returns the indices wich would sort that, but how does the outer a[] know how to use these indices to sort itself?

#

specifically, the argsort() returns a 1d ndarray, but when i try to change the order of a manually by entering an array into the brackets i get an error

a = [[7,3], [3,1], [4,2]]
a[ [2, 0, 1] ]```

magic light Jul 12, 2020, 3:24 AM

#

@quasi oracle Thanks for taking the time to explain! Your example sounds straight out of the Head First: Design Patterns book

#

Although I don't think I've come across many articles/books that don't use domesticated/farm animals as examples

#

I can understand "animal" is abstract and "duck" or "dog" are so-called "concrete" implementations of animal, but it's mostly the use of the term "concrete."

#

Why not "solid" or "implemented"? My issue is pedantic, but there might be a historical component to this since interface/implementation are originally Java terms that have just become synonymous to general terms

visual dagger Jul 12, 2020, 11:18 AM

#

@visual dagger @fiery cosmos thank you for the explanation guys! I understand it properly now.
@cyan current no problemo bro

brazen tinsel Jul 12, 2020, 11:22 AM

#

Nickname: true

azure owl Jul 12, 2020, 2:53 PM

#

specifically, the argsort() returns a 1d ndarray, but when i try to change the order of a manually by entering an array into the brackets i get an error
a = [[7,3], [3,1], [4,2]]
a[ [2, 0, 1] ]```

@fiery cosmos how are yo manually changing the order

fiery cosmos Jul 12, 2020, 7:25 PM

#

@azure owl sorry i was asleep.
i change it the way i posted in the code:

a[ [2, 0, 1] ]```

#

i gave it a 1d array with the indices as elements in a specific order, like the argsort returns but it doesnt work. (not that i need it to work but i would like to understand why)

#

@fiery cosmos what do you mean?

a[a[:,1].argsort()]

sorts the list a by its second column, while a[:,1].argsort() returns an ndarray with the indices giving the order of the rows.
wouldnt

a = [ [2, 0, 1] ]

just change my array to that order instead of sorting the array?

acoustic hearth Jul 12, 2020, 7:58 PM

#

hey can someone help me

#

I just want to know if the

#

ord()```

#

and py chr()

#

functions are important

#

so I wanna know if I need to learn it or not

marble basalt Jul 12, 2020, 7:59 PM

#

wdym important

#

they have their uses like any function

acoustic hearth Jul 12, 2020, 8:00 PM

#

like

#

Do I need to learn it

marble basalt Jul 12, 2020, 8:00 PM

#

you don't need to learn anything, if you want to deal with character codes then those functions are how you do it

gaunt forge Jul 12, 2020, 11:16 PM

#

The only things you need to learn are the tools to do what you want to do

runic tinsel Jul 13, 2020, 3:13 AM

#

we don;t know if they are important is what we're saying

visual dagger Jul 13, 2020, 5:40 AM

#

so I wanna know if I need to learn it or not
@acoustic hearth there is nothing you have to learn they will come in use if they have to there is nothing that needs to be learn in this
just know how to use them

rough temple Jul 13, 2020, 7:47 AM

#

Hi, I was wondering if anybody knows the difference between PCA and distributed PCA?

📎 Screen_Shot_2020-07-13_at_3.45.26_PM.png

faint sun Jul 13, 2020, 9:31 AM

#

how would you write a tuple to a py file?

#

feel free to @ me

unkempt hatch Jul 13, 2020, 9:32 AM

#

umm tuple = ("value1", "value2") @faint sun

faint sun Jul 13, 2020, 9:33 AM

#

when i try to write a tuple to a file i keep getting ```python
TypeError: write() argument must be str, not tuple

unkempt hatch Jul 13, 2020, 9:34 AM

#

you have to make it into a string

#

if youre writing it from 1 file to another

#

'tuple = ("value1", "value2")'

faint sun Jul 13, 2020, 9:37 AM

#

ok that somehow worked

#

im not sure how because i have been doing exactly that for the past hour

unkempt hatch Jul 13, 2020, 9:44 AM

#

ok great

#

it mightve been written wrong or something because i used 4 " so i edited the code to fix the problem

visual dagger Jul 13, 2020, 10:45 AM

#

when i try to write a tuple to a file i keep getting ```python
TypeError: write() argument must be str, not tuple

@faint sun you can dump it by using pickle module bro if you are writing that tuple to a text or anyfile just for binary you have to convert it to bytes by using

t = bytes(tuple)```
and if you encode it you can but it is optional

faint sun Jul 13, 2020, 10:47 AM

#

i did a string(my_tuple) and then i wrote it and i wasnt able to use it in my python file but then i tried it like 10 more times and somehow i was able to use it

#

ill use the bytes(tuple) next time i bump into the same thing

visual dagger Jul 13, 2020, 10:47 AM

#

wait wait wait

faint sun Jul 13, 2020, 10:48 AM

#

thank you for letting me know I really appreciate it

#

okie dokes

visual dagger Jul 13, 2020, 10:48 AM

#

bytes thing is for binary file bro

#

yeh

#

and you simply can loop it

#

like

#

for correct indentaion into a python file

#

or you can just dump
I prefer the dump method as str(tuple) can be altered easily

#

no problemo bro

thank you for letting me know I really appreciate it
@faint sun

faint sun Jul 13, 2020, 10:51 AM

#

its a huge tuple of over 7000 stuff so now i gotta figure out how to use it

#

thank you though!

visual dagger Jul 13, 2020, 10:54 AM

#

yeh then dump will surely help It is meant for doing that lol

#

you should consider googling it tho pickle module in python

quasi oracle Jul 13, 2020, 10:57 AM

#

@faint sun for future reference this kind of question should go in a help channel #❓｜how-to-get-help

faint sun Jul 13, 2020, 10:59 AM

#

got it! ill put it in the help chats next time! thanks for letting me know!

stray condor Jul 13, 2020, 7:34 PM

#

Hey so recently i just learned about Big O notation and im a little bit curious on why a set takes O(1)/constant time to look up things inside of it instead of O(n)/linear time like a list?

void topaz Jul 13, 2020, 7:35 PM

#

Hello, can anyone help me in understanding the core of Visual Scripting like PyFlow or Unreal Engine Blueprints? I did lot of research but I couldn’t get the gist of this flow-graphs. I tried studying open-source projects from different languages. Some are way too complex to study.

quasi oracle Jul 13, 2020, 7:45 PM

#

Hey so recently i just learned about Big O notation and im a little bit curious on why a set takes O(1)/constant time to look up things inside of it instead of O(n)/linear time like a list?
Because of something called hashing. It's basically a function that takes a value, and turns it into a number. So when you add an element to the set, its corresponding hash value (number) is calculated and using that you know where to store the element. When you want to look it up, you can get the hash value again, and with it know where it's stored.
The hashing takes O(1) time, and is designed to get as few collisions as possible, meaning that the probability of two objects getting the same hash value should be very low (If they do get the same value, there are techniques to deal with that).
So if you want to know more about how sets operate you should have a read about hash functions and hash tables. @stray condor

stray condor Jul 13, 2020, 7:49 PM

#

@quasi oracle Hashing sounds awesome! thank you mbaruh, I'm go and look into that right now!

digital leaf Jul 13, 2020, 9:19 PM

#

any leetcoders here? I am having a hell of a time with this in python

https://leetcode.com/problems/partition-equal-subset-sum/submissions/

Level up your coding skills and quickly land a job. This is the best place to expand your knowledge and get prepared for your next interview.

tropic dawn Jul 14, 2020, 4:36 AM

#

Anyone planning to go to Hack MIT?

swift ingot Jul 14, 2020, 12:57 PM

#

@tropic dawn I went every year while I was in college, it's consistently the best hackathon out there

vital wasp Jul 14, 2020, 1:49 PM

#

@swift ingot how selective are they in choosing the teams and stuff, i have a decent application but regardless, im still wondering

swift ingot Jul 14, 2020, 1:54 PM

#

Hard to say because of covid this year

#

In the past it is pretty selective

#

Reason being that half of the attendees end up being MIT students

#

I always got in through the puzzle

#

The puzzle is legit very fun

#

and very hard

tropic dawn Jul 14, 2020, 4:11 PM

#

Just applied hoping to get in

#

Is it worth it to start making teams now or wait till the status updates?

pallid jay Jul 14, 2020, 4:39 PM

#

Hey,I need urgent help.

#

Ok so I need to write a program that does ctrl f in python.

#

It should extrct values that have a word in them then it should only print thoose words.

#

Please help.

last blaze Jul 14, 2020, 4:42 PM

#

You should ask your question in a normal help channel

pallid jay Jul 14, 2020, 4:42 PM

#

ok

swift ingot Jul 14, 2020, 9:09 PM

#

@tropic dawn in past years, they accepted individuals, not teams

#

So if you start making a team now, theres a chance your whole team would not get in

#

However, if you complete the puzzle, your whole team is guaranteed admission, which is why I always did the puzzle

#

You can try networking with MIT students but the hackathon is virtual anyway...

#

You're not getting the full experience online, I can tell you that for sure

tropic dawn Jul 14, 2020, 9:36 PM

#

When do they give the puzzle

fiery cosmos Jul 15, 2020, 5:23 AM

#

uh i need help not sure where to go

visual dagger Jul 15, 2020, 5:43 AM

#

uh i need help not sure where to go
@fiery cosmos go to any channel in the PYTHON HELP:AVAILABLE and type your query
then the hannel you wrote on will be transfered to PYTHON HELP:OCCUPIED

fiery cosmos Jul 15, 2020, 5:44 AM

#

yes I was able to find help. Thank you!!

golden flume Jul 15, 2020, 7:17 AM

#

i want some image processing help

#

anybody familiar?

gusty grove Jul 15, 2020, 7:26 AM

#

sure

#

what are you trying?

fiery cosmos Jul 15, 2020, 8:08 AM

#

can anyone answer my question please?

#

why do computers use 1 and 0?

#

why not some other pair of digits/

#

Computers have circuits, which have transisitors, now transistors can either be "Off" or "On" these states are reflected by the usage of 0 and 1

#

yeah but why 'on' is 1?

#

why not some other digits?

#

You could use other numbers surely

#

but the thing is, it becomes difficullt

#

binary in 0's and 1's is extremely straight forward

#

And it also follows convenction

fiery cosmos Jul 15, 2020, 8:50 AM

#

ohh yeah

#

thank you so much

fiery cosmos Jul 15, 2020, 9:06 AM

#

now

#

hellow

lavish fox Jul 15, 2020, 9:12 AM

#

What is a bandwidth

summer reef Jul 15, 2020, 9:24 AM

#

that's the amount of data that can pass through a communication line, usually expressed in bauds (symbols/second), or bits/second

#

📎 unknown.png

jovial geyser Jul 15, 2020, 10:20 AM

#

is it like a limit to share data in a connection

visual dagger Jul 15, 2020, 10:35 AM

#

why do computers use 1 and 0?
@fiery cosmos because they are true and false that program the computer to tell what is true and what is false
or you may say that they are the computer as without them it is just useless harware
they just do simplify everything in true and false and that is a program if you see clearly

vague silo Jul 15, 2020, 10:36 AM

#

Does this discussion include only Python language or is there more languages involved as well?

visual dagger Jul 15, 2020, 10:36 AM

#

Does this discussion include only Python language or is there more languages involved as well?
@vague silo someone would help with other language too if they know

fiery cosmos Jul 15, 2020, 11:07 AM

#

Thanks a lot

modest lake Jul 15, 2020, 1:01 PM

#

Hi guys can anyone explain me what is logic gates.

runic tinsel Jul 15, 2020, 1:35 PM

#

same as boolean operations

#

XNOR, XOR

swift ingot Jul 15, 2020, 1:54 PM

#

@tropic dawn just watch for it

#

its usually hidden in the webpage

late harness Jul 15, 2020, 6:47 PM

#

everytime i run with f5 in visual studio this thing comes up

#

📎 unknown.png

#

and i want to skip having to click python file each time

half lotus Jul 15, 2020, 6:48 PM

#

This question belongs in #tools-and-devops

late harness Jul 15, 2020, 6:48 PM

#

ok

regal scroll Jul 15, 2020, 7:15 PM

#

anyone have a recommended python guides on graph algorithms and grasping graph concepts?

chrome jacinth Jul 15, 2020, 11:34 PM

#

hello all, I have a table with two columns and I want to append values of columns to a list of dictionaries. I am trying to do it by [{'key1':i} for i in table['column1], {'key2':j} for j in table['column2']] but its giving me a sysntax error, any ideas?

marble basalt Jul 15, 2020, 11:42 PM

#

you can't have two seperate list comprehensions for the same list

#

I would just use a for loop

#

kinda strange to have a whole bunch of one element dictionaries though

chrome jacinth Jul 15, 2020, 11:50 PM

#

for i in table['column1']: {'key1':i} ? I need it this way to construct a table in plotly dash and the data entry requires a list of dictionaries 😦

marble basalt Jul 16, 2020, 1:12 AM

#

I would assume each dictionary is a row

#

so it should have key1 and key2

#

and just putting a dictionary in the body of the loop doesn't do anything

#

it makes the object and then it goes away as you have no reference to it

chrome jacinth Jul 16, 2020, 1:19 AM

#

I see thank you will try this

#

so it should have key1 and key2
@marble basalt this was it btw. thanks

fiery cosmos Jul 17, 2020, 12:07 PM

#

what is exactly a version control?

#

please help me

shell abyss Jul 17, 2020, 3:54 PM

#

Git is a software for version control. Github is just a service for people using git but github itself is not a version control tool.

shell abyss Jul 17, 2020, 5:13 PM

#

Github is using git for version control so the real version control tool is git, not github. Github is a service built on top.

shell abyss Jul 17, 2020, 5:51 PM

#

It sounded like homework and this is exactly how pedandic some profs can be. If you wanna give half complete answers go ahead, but I will question em.

loud trail Jul 17, 2020, 7:07 PM

#

you can't use github without git, so i think pucch is right here

loud trail Jul 17, 2020, 8:49 PM

#

it's still git though

#

i see your point

last blaze Jul 17, 2020, 9:04 PM

#

You're still using git, you're just doing it on githubs server

#

The fact you're not using it the traditional way doesn't mean you aren't using it

exotic merlin Jul 17, 2020, 10:42 PM

#

Is this a good channel to ask a question about time complexity?

#

or big O notation

fiery cosmos Jul 17, 2020, 10:44 PM

#

I want to output matrix like this

9 8 7 6
10 11 12 5
1 2 3 4

I want to output spiral matrix...

this is my first message in this server... can someone to help me?

!!! i=x, j=x
for more example =>
if i give as arguments function: 3, 5, it mast print like this:

11 10 9 8 7  
12 13 14 15 6  
1 2 3 4 5

quasi oracle Jul 17, 2020, 11:04 PM

#

Is this a good channel to ask a question about time complexity?
@exotic merlin yes

exotic merlin Jul 17, 2020, 11:07 PM

#

cool cool. If I had a multidimensional array with some values in the inner most array and I want to print out all the values from within the inner most arrays. Would that be considered O(n^2) cause it would take at least a nested for loop statement to be able to print out the inner most array values or would it be considered O(n)?

quasi oracle Jul 17, 2020, 11:08 PM

#

If you have n^2 values to print, then it has to take at least O(n^2) time

exotic merlin Jul 17, 2020, 11:08 PM

#

let me create a code example real quick

#

So if I had something like

 [1, 1, 1, 1, 1, 0],
 [0, 0, 0, 0, 0, 0],
 [0, 1, 1, 1, 1, 1],
 [0, 1, 1, 1, 1, 1],
 [0, 0, 0, 0, 0, 0]]

for i in arr:
  for num in i:
    print(num)```

#

would this be considered O(n^2) or O(n)?

quasi oracle Jul 17, 2020, 11:10 PM

#

Let's say the size of the array in each dimension is n
How many values do you have in the array?

exotic merlin Jul 17, 2020, 11:11 PM

#

you have n values

quasi oracle Jul 17, 2020, 11:11 PM

#

I mean, you have n rows, and each row has n values
What is the total number of values across all rows

exotic merlin Jul 17, 2020, 11:12 PM

#

n * n ?

quasi oracle Jul 17, 2020, 11:12 PM

#

Yeah

#

So how many values are you trying to print here?

for i in arr:
  for num in i:
    print(num)

exotic merlin Jul 17, 2020, 11:13 PM

#

n * n values

quasi oracle Jul 17, 2020, 11:13 PM

#

Yep

#

If you have n^2 values to print, then it has to take at least O(n^2) time
So this applies

exotic merlin Jul 17, 2020, 11:13 PM

#

hm

#

Thats what I thought of at first because I just saw that a for loop could just be considered O(n) in itself and since you have a nested for loop, it would be O(n) * O(n) = O(n^2). But I was reading what people were saying about this on a stackoverflow post and some people say that it could be considered O(n)

#

https://stackoverflow.com/questions/30100847/what-is-the-time-complexity-of-traversing-a-2d-array#:~:text=The time complexity will be,array ie%2C the 2nd dimension.&text=The time complexity is O,its time complexity is linear.

Stack Overflow

What is the time complexity of traversing a 2d array

What is the time complexity of traversing (rows ,columns) a two dimensional array?

bool check(int array [9][9])
{
int num=0;
for (int i = 0; i < 9; i++) {
for (int j = 0; j <...

#

And I understand how it could be O(n^2) but not O(n)

quasi oracle Jul 17, 2020, 11:16 PM

#

In this case n is defined as the total number of values, not the size in each dimension

#

The first answer clears some of the confusion

edgy oyster Jul 17, 2020, 11:17 PM

#

hi @exotic merlin this blog has some python examples: https://stackabuse.com/big-o-notation-and-algorithm-analysis-with-python-examples/

Stack Abuse

Big O Notation and Algorithm Analysis with Python Examples

There are multiple ways to solve a problem using a computer program. For
instance, there are several ways to sort items in an array. You can use merge
sort [https://en.wikipedia.org/wiki/Merge_sort], bubble sort
[https://en.wikipedia.org/wiki/Bubble_sort], insertion sort
[http...

quasi oracle Jul 17, 2020, 11:17 PM

#

It's a matter of what you define n to represent

exotic merlin Jul 17, 2020, 11:18 PM

#

So since n is defined as the total number of values, it's considered O(n^2)?

quasi oracle Jul 17, 2020, 11:19 PM

#

If you define n as the total number of values, then you have to print O(n) values, which is how the person asking in the post you sent got to their answer
In what I explained earlier, I referred to n as the size in each dimension. That led to the total number of values being n^2

exotic merlin Jul 17, 2020, 11:23 PM

#

ah, I think I get it

#

If I was given just the snipit of code above I pasted without any reference to what n is, I would automatically think that its just O(n^2) just because of the nested for loops. I can't see anyone look at the code and the first thing that pops up in their head is "yep that's O(n)"

#

But I guess in this case, it all just depends on the context of the problem beforehand

quasi oracle Jul 17, 2020, 11:30 PM

#

Yeah. saying it's O(n^2) would probably be standard

exotic merlin Jul 17, 2020, 11:31 PM

#

cool cool. Thanks for the help.

lost loom Jul 18, 2020, 4:24 AM

#

Just wondering if these notes are correct for calculating the derivative of a cost with respect to a weight or bias

📎 IMG_7893.jpg

#

currently trying to construct my first neural network

#

not sure if this is more suited to #data-science-and-ml, let me know if thats the case

faint vale Jul 18, 2020, 7:04 AM

#

can someone point me to any resource about alogorithms and data structures for beginners?

last blaze Jul 18, 2020, 7:31 AM

#

Your argument is literally the same as saying google isn't a search engine
@fiery cosmos
No. The background api is a search engine, it's searching documents.

#

You're using git through the interface of github

quasi oracle Jul 18, 2020, 10:08 AM

#

can someone point me to any resource about alogorithms and data structures for beginners?
@faint vale this channel's pins have a link

random shale Jul 18, 2020, 12:31 PM

#

@faint vale check out algorithms specialization by stanford, ds & algo specialization by ucsd on Coursera

quasi oracle Jul 18, 2020, 12:35 PM

#

@random shale your second ping isn't right

random shale Jul 18, 2020, 12:48 PM

#

@quasi oracle I'm a beginner taking these courses.

quasi oracle Jul 18, 2020, 12:48 PM

#

No, I mean, in the second message it was a different avi

#

You deleted it so nevermind

faint vale Jul 18, 2020, 1:17 PM

#

@random shale thank you very much. Do u recommend taking these courses after doing basic python?

random shale Jul 18, 2020, 1:26 PM

#

@faint vale i would suggest knowing basics of python and some discrete math (mathematics for CS) before attempting these specializations.

faint vale Jul 18, 2020, 1:34 PM

#

@random shale any resource you would suggest for discrete maths? I am from electrical engineering background. Its a good chance i already have done most of it. But still won't hurt to brush up.

idle crow Jul 18, 2020, 1:37 PM

#

I personally found Concrete Mathematics to be a really useful book

random shale Jul 18, 2020, 1:37 PM

#

@faint vale discrete math and it's applications by Rosen.

#

If you have less time, then try YouTube, lots of CS math videos by reputed institutions on youtube. @faint vale

faint vale Jul 18, 2020, 1:40 PM

#

@idle crow @random shale thank you gentlemen for your valuable advice. Cheers.

random shale Jul 18, 2020, 1:58 PM

#

I personally found Concrete Mathematics to be a really useful book
@idle crow

This is also a great book.

fading spade Jul 18, 2020, 3:26 PM

#

Hi guys, sry, what book I should read? I'm a begginer

last blaze Jul 18, 2020, 3:27 PM

#

!resources has some links for getting started with programming in general - not a huge amount of books though. If you're looking to get started with CS in general, I can recommend CS50 or MIT6.001 video courses (I think both are linked !resources)

halcyon plankBOT Jul 18, 2020, 3:27 PM

#

Resources

The Resources page on our website contains a list of hand-selected learning resources that we regularly recommend to both beginners and experts.

fading spade Jul 18, 2020, 3:28 PM

#

thankss!

orchid finch Jul 18, 2020, 7:23 PM

#

uhh anyone comfortable with regex

#

I am trying to figure out how to capture something that DOES NOT match a certain regex

#

A sentence where there is not duplicate word side by side

#

^(?!.*\b(\w+)\s+\1\b.*)

rare thorn Jul 18, 2020, 7:57 PM

#

https://regex101.com

Regex101 - online regex editor and debugger

Regex101 allows you to create, debug, test and have your expressions explained for PHP, PCRE, Python, Golang and JavaScript. The website also features a community where you can share useful expressions.

#

Might help you

fiery cosmos Jul 18, 2020, 8:43 PM

#

hey all

#

I've a question

lean dome Jul 18, 2020, 8:45 PM

#

!ask

halcyon plankBOT Jul 18, 2020, 8:45 PM

#

Asking good questions will yield a much higher chance of a quick response:

• Don't ask to ask your question, just go ahead and tell us your problem.
• Don't ask if anyone is knowledgeable in some area, filtering serves no purpose.
• Try to solve the problem on your own first, we're not going to write code for you.
• Show us the code you've tried and any errors or unexpected results it's giving.
• Be patient while we're helping you.

You can find a much more detailed explanation on our website.

fiery cosmos Jul 18, 2020, 8:47 PM

#

I want to define a function wich prints spiral matrix like this:

this is printed out by the function when n = 3 and m = 4. n and m are tyhe arguments of the defined function.

#

I need this code as soon as it is possible... ❤️ thanks

lean dome Jul 18, 2020, 8:49 PM

#

what have you tried?

fiery cosmos Jul 18, 2020, 8:49 PM

#

whait

marble basalt Jul 18, 2020, 8:50 PM

#

Try coming up the with how to get the indexes of the next number from a given spot

fiery cosmos Jul 18, 2020, 8:53 PM

#

N=int(input())
A=(N+2)*[(N+2)*[0]]
A[0]=A[N+1]=(N+2)*[-1]
for i in range(1,N+1):
    A[i]=[-1]+N*[0]+[-1]
Di=(0,1,0,-1)
Dj=(1,0,-1,0)
M,i,j,m=0,1,0,0
while M<N*N:
    M+=1
    if A[i+Di[m]][j+Dj[m]]:
        m=(m+1)%4;
    i+=Di[m]
    j+=Dj[m]
    A[i][j]=M
for i in range(1,N+1):
    for j in range(1,N+1):
        print(A[i][j],end=' ')
    print()

#

i tried this but it works only for one argument

lean dome Jul 18, 2020, 8:57 PM

#

well, the first step for making it work for distinct N and M would be to prompt the user for both.

fiery cosmos Jul 18, 2020, 9:04 PM

#

can you help me?

#

pls ❤️ 🙂

#

?????????

#

is here someone?

lean dome Jul 18, 2020, 9:13 PM

#

this seems like homework, and no one here is going to do your homework for you. if you have specific questions about how to do some piece of the work, you'll get some help. if you are just hoping for someone to do the entire thing for you, you're in the wrong spot.

marble basalt Jul 18, 2020, 9:13 PM

#

you should make your variable names more descriptive

#

its essentially completely unreadable rn

fiery cosmos Jul 18, 2020, 9:20 PM

#

this seems like homework, and no one here is going to do your homework for you. if you have specific questions about how to do some piece of the work, you'll get some help. if you are just hoping for someone to do the entire thing for you, you're in the wrong spot.
@lean dome this isn't my home work

#

it's just a problem of e-olymp.com content

lean dome Jul 18, 2020, 9:22 PM

#

that doesn't anything I said. No one will solve the problem for you, and asking for help solving a problem is very different from asking for a solution.

marble basalt Jul 18, 2020, 9:35 PM

#

consider you only have to solve the outside, then you can embed the next layer inside that, like in your example you can get all the outside numbers and then do the same thing for n=1 m=2 to get the inside numbers

icy fulcrum Jul 18, 2020, 10:12 PM

#

Ok so what I've learned about perceptual hashs: they can be used for fingerprinting 8 bit greyscale images. Since every block in minecrsft is an 8bit data value. I should be able to convert the data set into a 8bit bmp and hash it?

onyx root Jul 18, 2020, 10:22 PM

#

@icy fulcrum you were talking earlier about having hash collisions where you thought the blocks were actually different? That's extremely unlikely.

lean dome Jul 18, 2020, 10:23 PM

#

@icy fulcrum I've never played minecraft, so this may be a dumb question, but - aren't there a finite number of types of blocks? Why do you need to hash a block, instead of just storing a unique integer per type of block?

#

can't you just say that lava is 0, basalt is 1, etc?

icy fulcrum Jul 18, 2020, 10:26 PM

#

Not the block itself. But a region of blocks. A region contains 32x32 chunks each chunk is 16x16x256 blocks including air but only an average of 80 in altitude. The goal here is to fingerprint the map at the region file level as im working on 15 gig data set. Out of 4023 regions 451 of them are showing the same sha256 checksum. So I wanted to checksum in a different slightly more accurate way. I dont really care about the data value if the individual blocks but the integrity of the map as a whole

onyx root Jul 18, 2020, 10:27 PM

#

@icy fulcrum you say 451 of them are showing the same checksum. But you think they actually have different contents? That is extremely unlikely.

lean dome Jul 18, 2020, 10:28 PM

#

well, that depends on how you did the checksum

onyx root Jul 18, 2020, 10:28 PM

#

the most likely explanation is a bug in the summing

icy fulcrum Jul 18, 2020, 10:28 PM

#

Its theoretically plausible that each region could contain the exact same dataset it is running a Parlin noise generator

#

I used the Linux cli sha256sum rwgion.name

onyx root Jul 18, 2020, 10:29 PM

#

what is "rwgion.name" there?

lean dome Jul 18, 2020, 10:29 PM

#

yeah. your checksum may have wound up amounting to "number of non-air tiles", in which case many regions would have the same value. It'll be up to you to make sure that the values you're feeding into the hash result in a sane hash value.

icy fulcrum Jul 18, 2020, 10:31 PM

#

📎 unknown-1.png

#

This is after loading the dataset into mariadb

onyx root Jul 18, 2020, 10:31 PM

#

ok, but what data are you feeding into sha256sum?

icy fulcrum Jul 18, 2020, 10:32 PM

#

The minecrsft region files named in the photo. After generating the map with the world generator

onyx root Jul 18, 2020, 10:32 PM

#

can you check two of those files with the same hash? They will have identical contents

icy fulcrum Jul 18, 2020, 10:32 PM

#

I mqy not be understanding your question

#

I should be able to check. A binary diff maybe?

onyx root Jul 18, 2020, 10:33 PM

#

sure.

icy fulcrum Jul 18, 2020, 10:39 PM

#

ok no they are the same byte for byte

#

cmp -b BopGen/DIM-1/region/r.-18.-15.mca BopGen/DIM-1/region/r.-1.-3.mca gave me no return and then i used cmp -b BopGen/DIM-1/region/r.-18.-15.mca BopGen/DIM-1/region/r.-1.3.mca says differ: byte 3, line 1 is 0 ^@ 2 ^B

#

so then all of those map tiles that are identical in the hash values are then identical in the map

#

O_e maybe im missing something here on how Perlin Noise works

onyx root Jul 18, 2020, 10:42 PM

#

yes. sha256 is very good at not randomly colliding

icy fulcrum Jul 18, 2020, 10:43 PM

#

as i thought that it should be since its a cryptographic hashing alg

#

at least i know how to check my work now tho 🙂

#

so if binary data is equal in this case how am i using a Perlin noise generator to generate and save the same data in multiple regions? that would mean every block in the 3d grid was exactly the same

#

is that just a flaw of Perlin noise in this implementation? or... i wonder if i really have cloned chunks O_e

#

thats an interesting thought ^ ill need to dig into that and see where those chunks are to know for sure. Thank you Ned and Godlygeek

marble basalt Jul 19, 2020, 12:13 AM

#

What is a region? Could it just be all air or water blocks or something like that

icy fulcrum Jul 19, 2020, 2:45 AM

#

A region is a structure defining a set of chunks, inturn these are providing information on on all blocks that are not air. Water, lava, stone, dirt, etc. In game these are arranged in a 3d grid to define a terrain map. Usually generated by the game engine (parts of it are perlin noise based) however there are addon programs that can read, write, and generate data from other algorithms. Each file also contains a small set of information pertaining to tile entities (creepers, villagers, etc) if i remember correctly the entities may be a separate file.

#

that being said each region also contains the world coordinates by name of file

#

r.-18.-15.mca and r.-1.-2.mca in my examples are identical regions on the map. these are located at -18 X and -1 X by -15 Z by -2 Z respectively on a square grid these are defiantly not in the same grid location. but all of the data in the region (512×256×512 blocks, 67,108,864 total (although the air is NOT saved) ) are identical..... O_e my brain hurts i cannot fathom how a noise generator (like Perlin Noise) would even accomplish this.

icy fulcrum Jul 19, 2020, 3:28 AM

#

Unless that is the pattern. I wonder if i plotted all the regions on a bigger grid if i could i see the noise patterns

#

Nat you are a genus. Thank you for asking that question I think that's exactly the direction I needed to go.

torn lodge Jul 19, 2020, 4:54 AM

#

Hey guys need some advice
I found this question on stack overflow to find the line that touched most of the rectangles

I have no clue at the moment but we need to find the line which touches the most of the rectanles and does not have to be the corners
any clue?

#

📎 image.png

torn scarab Jul 19, 2020, 5:36 AM

#

the easiest solution I can think of involves starting with a random line, then rotating it 180 degrees, checking the number of rectangles it touches as you rotate

torn lodge Jul 19, 2020, 6:00 AM

#

but thats too much processing and it will not work as height of this line needs to be adjusted

wide prism Jul 19, 2020, 6:12 AM

#

can you give the link for the stack overflow thing

last blaze Jul 19, 2020, 7:18 AM

#

What's the bounds of the input size?

strong ledge Jul 19, 2020, 9:28 AM

#

https://stackoverflow.com/questions/49333960/maximum-possible-number-of-rectangles-that-can-be-crossed-with-a-single-straight

Stack Overflow

Maximum possible number of rectangles that can be crossed with a si...

I found this challenge problem which states the following :
Suppose that there are n rectangles on the XY plane. Write a program to calculate the maximum possible number of rectangles that can be

median topaz Jul 19, 2020, 11:41 AM

#

Hello guys, i am beginner in python and i am working on an OCR thing, this way it works fine. but i want to make it take all the png files in the folder and extract them as text files with the png file name.

📎 Screenshot_2.png

#

what should i do or where should i look for a tutorial for this?

candid osprey Jul 19, 2020, 2:14 PM

#

Well, if you already have a code that can read one file, then this is mostly on you now.

#

Find something online that allows you to take all files in a folder. (for example. Os.listdir) then simply write a loop for this.

lunar hull Jul 19, 2020, 10:25 PM

#

def get_guessed_word(x,y):
li =[]

  for word in x:
      if word in y:
          li.append(word)
  print(li)
   return li

I'm getting error invalid character in identifier in the above function

#

please help

runic tinsel Jul 19, 2020, 10:36 PM

#

type it again

#

it's between word and (

neat thunder Jul 20, 2020, 2:29 AM

#

hi

#

i'm not still at that level of py

#

i'm beginner in python i just know few stuffs about it

#

can anyone tell me how to compile python on mac

#

in terminal

#

i use sublime text

lean dome Jul 20, 2020, 4:05 AM

#

That'd be more on topic in #unix - this channel is about the mathematical theory behind computing.

umbral gulch Jul 20, 2020, 5:46 PM

#

anyone good with algorithms and optimising them. i have a problems and im not even sure where to start

pseudo wave Jul 20, 2020, 5:46 PM

#

!ask

halcyon plankBOT Jul 20, 2020, 5:46 PM

#

Asking good questions will yield a much higher chance of a quick response:

• Don't ask to ask your question, just go ahead and tell us your problem.
• Don't ask if anyone is knowledgeable in some area, filtering serves no purpose.
• Try to solve the problem on your own first, we're not going to write code for you.
• Show us the code you've tried and any errors or unexpected results it's giving.
• Be patient while we're helping you.

You can find a much more detailed explanation on our website.

umbral gulch Jul 20, 2020, 5:54 PM

#

i needed to type it out here because it was too big for discord but here is the problem
https://hastebin.com/cekanaxeri.py

quasi oracle Jul 20, 2020, 6:05 PM

#

Didn't quite get the pen part, but for efficiently handling image data you should look into numpy.
In your case you might benefit from dividing your image to connected components

umbral gulch Jul 20, 2020, 6:08 PM

#

the pen part is just how many times the pen draws something to the screen like this

📎 45KnuG2fMo.mp4

#

@quasi oracle what do you mean by connected components

quasi oracle Jul 20, 2020, 6:12 PM

#

Huh, that's pretty awesome.
So what do you mean by minimize? Because it looks like it deals with one color each time and doesn't go back to it

umbral gulch Jul 20, 2020, 6:17 PM

#

well lets say you have a line of pixels like this
🟦🟥🟦🟩🟦🟥🟦🟦
lets say green has already been drawn
if when drawing the blue line i joined red and blue, thats more efficient as the pen will only need to make 2 movements instead of 4

#

like if i have 20 different colours. trying to join different ones in such a way to make less pen movements. not sure if that makes sense

#

if needed i can put my whole code up somewhere its not that long

#

the icon?

#

is this something youve made?

fiery cosmos Jul 20, 2020, 6:22 PM

#

its iconbitmap or not ?

candid osprey Jul 20, 2020, 6:22 PM

#

Just paste a picture on top of it? Or what.

umbral gulch Jul 20, 2020, 6:23 PM

#

its iconbitmap or not ?
@fiery cosmos its probably just an icon(ico) file

fiery cosmos Jul 20, 2020, 6:23 PM

#

with an other

#

@fiery cosmos its probably just an icon(ico) file
@umbral gulch aight I'll try it

candid osprey Jul 20, 2020, 6:23 PM

#

I think this is where you're making some fundamental assumption about where this is supposed to be changed but you don't realise you didn't tell us

fiery cosmos Jul 20, 2020, 6:24 PM

#

I think this is where you're making some fundamental assumption about where this is supposed to be changed but you don't realise you didn't tell us
@candid osprey wdym

candid osprey Jul 20, 2020, 6:24 PM

#

What are we looking at

fiery cosmos Jul 20, 2020, 6:24 PM

#

This might be more suitable for #❓｜how-to-get-help then theoretical cs?

candid osprey Jul 20, 2020, 6:24 PM

#

Is this a screenshot of a browser tab?

quasi oracle Jul 20, 2020, 6:24 PM

#

well lets say you have a line of pixels like this
🟦🟥🟦🟩🟦🟥🟦🟦
lets say green has already been drawn
if when drawing the blue line i joined red and blue, thats more efficient as the pen will only need to make 2 movements instead of 4
@umbral gulch more efficient compared to what? What are the steps you want versus what you're doing now?

fiery cosmos Jul 20, 2020, 6:25 PM

#

well it is probably the wrong channel for that

#

lol

candid osprey Jul 20, 2020, 6:25 PM

#

I get what they want. Basically imagine if the red boxes were blue instead

fiery cosmos Jul 20, 2020, 6:25 PM

#

it's about tkinter

candid osprey Jul 20, 2020, 6:25 PM

#

Then the pen draws blues in two strokes

#

And then they can always draw the red boxes separately later in 2 strokes

#

That sound about right?

quasi oracle Jul 20, 2020, 6:26 PM

#

Oh, versus the opposite, ok

umbral gulch Jul 20, 2020, 6:27 PM

#

@quasi oracle heres the full code, just now its sorts the colours based on what one will be doing the most pen strokes down to the one that will be doing the least. but instead of wither getting the pen strokes for all of the remaining colours i wanna try find if there is a combination of remaining colours that will lead to less pen strokes than just one colour https://hastebin.com/divodatoxu.py

#

That sound about right?
@candid osprey yeah

candid osprey Jul 20, 2020, 6:29 PM

#

Does drawing more strokes really make it that much slower? That's the only part that I guess must be true.

#

To make this exercise worth it

umbral gulch Jul 20, 2020, 6:30 PM

#

Does drawing more strokes really make it that much slower? That's the only part that I guess must be true.
@candid osprey well the less strokes the quicker it will draw the image so yes, because i need to have a sleep or mspaint doesnt have time to draw the image and it just updates at the end in a big mess

candid osprey Jul 20, 2020, 6:30 PM

#

Ah I see I see. So it's because you're essentially having to interact with Mspaint like almost an automation tool eh?

umbral gulch Jul 20, 2020, 6:32 PM

#

this is it without it sleeping

📎 f8tkOgung0.mp4

quasi oracle Jul 20, 2020, 6:32 PM

#

I imagine the more fragmented a color is, the higher priority it should have. Which seems to be what you're already doing.
Do you have an example of a more efficient order than what this priority sorting provides?

candid osprey Jul 20, 2020, 6:33 PM

#

Oh dear

#

How did you record these videos btw?

umbral gulch Jul 20, 2020, 6:34 PM

#

I imagine the more fragmented a color is, the higher priority it should have. Which seems to be what you're already doing.
Do you have an example of a more efficient order than what this priority sorting provides?
@quasi oracle its hard to show on a small scale

#

How did you record these videos btw?
@candid osprey ShareX

quasi oracle Jul 20, 2020, 6:35 PM

#

Well it's hard to help otherwise 😅

umbral gulch Jul 20, 2020, 6:35 PM

#

id say voice but theres people in there

#

@quasi oracle ill try and draw a small image in paint that would benifit from what im thinking about

quasi oracle Jul 20, 2020, 6:37 PM

#

Are strokes lines only? Drawing a region in 2d takes several strokes?

candid osprey Jul 20, 2020, 6:38 PM

#

I'd assume that's the simplest way to do it, changing pen sizes might introduce more steps that make it even slower

umbral gulch Jul 20, 2020, 6:38 PM

#

this is the function thats getting the strokes for each color/set of colours. if 2 pixels are next to each other it joins them

def get_strokes(color_positions):
    strokes = []
    pos_index = 0
    while pos_index < len(color_positions):
        start_pos = color_positions[pos_index][0]+paint_offset[0],color_positions[pos_index][1]+paint_offset[1]
        end_pos = color_positions[pos_index][0]+paint_offset[0],color_positions[pos_index][1]+paint_offset[1]
        while True:
            try:
                if color_positions[pos_index+1][0]+paint_offset[0] == end_pos[0] + 1:
                    pos_index += 1
                    end_pos = color_positions[pos_index][0]+paint_offset[0],color_positions[pos_index][1]+paint_offset[1]
                else:
                    break
            except: break
        strokes.append((start_pos,end_pos))
        pos_index += 1
    return strokes

then this part is it drawing each of the strokes for X colour

for stroke in all_colours:
    pui.moveTo(stroke[0])
    pui.dragTo(stroke[1])

candid osprey Jul 20, 2020, 6:38 PM

#

Since this isn't simply a direct control over the image pixels, more like instructions to a bot

quasi oracle Jul 20, 2020, 6:40 PM

#

So each row is independent of the others. The only reason you draw one color at a time is because switching takes time

umbral gulch Jul 20, 2020, 6:40 PM

#

@quasi oracle yes, i can invite you to my server and talk in voice it might be easier to explain if you want

quasi oracle Jul 20, 2020, 6:41 PM

#

Here is fine for now I think

umbral gulch Jul 20, 2020, 6:41 PM

#

ill try and draw an image that will benifit from what i think the change should be

candid osprey Jul 20, 2020, 6:42 PM

#

Im at the stage where I perfectly understand the problem but can't think of a way to approach the solution directly. I'm wondering if you'd just have* to settle for some kind of compromise

quasi oracle Jul 20, 2020, 6:43 PM

#

ill try and draw an image that will benifit from what i think the change should be
From what we just discussed you really only need to show one row

candid osprey Jul 20, 2020, 6:43 PM

#

I can think of a simple example.

#

Essentially, filling the majority colour isn't always best. Imagine this

#

A row of blue. But one red dot on each end.

#

Red first then blue is two strikes. Blue first then red is 3.

quasi oracle Jul 20, 2020, 6:44 PM

#

Yes, but the red is more fragmented

umbral gulch Jul 20, 2020, 6:44 PM

#

From what we just discussed you really only need to show one row
@quasi oracle not exactly

quasi oracle Jul 20, 2020, 6:44 PM

#

Which makes it work if you order by this priority

candid osprey Jul 20, 2020, 6:44 PM

#

So, the more distinct broken groups first?

quasi oracle Jul 20, 2020, 6:45 PM

#

If we only look at one row, probably

#

But like Pyro said it's not necessarily the case, because in the next line it might be very low priority

#

But if switching colors is expensive enough, the fastest solution would be a priority over the entire image

candid osprey Jul 20, 2020, 6:46 PM

#

I think there simply compute across all lines and just take the overall best, call it a day

#

Exactly

#

Essentially a greedy approach

umbral gulch Jul 20, 2020, 6:48 PM

#

im trying something up 2 secs

quasi oracle Jul 20, 2020, 6:48 PM

#

You can calculate the fragmentation of a row in one pass I believe

candid osprey Jul 20, 2020, 6:52 PM

#

GBRBRBRBG. Blue 4 red 3 green 2. 6 strokes with blue first. 5 strokes if G then B then R. Did I count correctly?

#

Essentially something I'm observing mentally is that eliminating edge pairs is important.

quasi oracle Jul 20, 2020, 6:53 PM

#

Why 6 stokes with blue? It's the most fragmented, so one stroke

candid osprey Jul 20, 2020, 6:53 PM

#

B then R then G

#

Total strokes 6 yah?

quasi oracle Jul 20, 2020, 6:54 PM

#

Yeah

#

Hmm

candid osprey Jul 20, 2020, 6:56 PM

#

I'm not yet convinced if there's an objective answer to this problem or if we'd have to settle for something that's good enough.

umbral gulch Jul 20, 2020, 6:56 PM

#

so like this, https://hastebin.com/oredubarov.shell. so by joining all the different colours in such a way to give the biggest saving for what colours are remaining and still need to be drawn

📎 unknown.png

candid osprey Jul 20, 2020, 6:58 PM

#

You know what, I had a pretty silly idea

umbral gulch Jul 20, 2020, 6:58 PM

#

so for drawing the black colour it would be best to join the positions of the black and green because that gives the biggest saving

candid osprey Jul 20, 2020, 6:58 PM

#

Why don't we simply brute force the calculation

#

Before drawing it out?

umbral gulch Jul 20, 2020, 6:58 PM

#

Why don't we simply brute force the calculation
@candid osprey go through every possible colour combination?, well if you have 16 colours in the colour palette. thats just under 16! or 2.092279e+13. that will take along time to calculates every possible solution and find the shorest

candid osprey Jul 20, 2020, 6:59 PM

#

Just calculate some random sequence of colours. You can even make it calculate only a small subset of all possible combinations and pick the best

#

Almost like a simple genetic algo / heuristic approach to drawing something that's "good enough"

#

Let the machine crunch the numbers first. It's good at those.

topaz pulsar Jul 20, 2020, 6:59 PM

#

would it save time to use the paint bucket on the first (and most common) colour? just fill the screen initially like that

candid osprey Jul 20, 2020, 7:00 PM

#

Not always no

#

Fragmentation creates bigger problems

#

Aye Pyro, I was realising that as I was saying it, so I was only wondering about taking subsets at random

topaz pulsar Jul 20, 2020, 7:01 PM

#

seems like an interesting problem, i just can't think of a better solution

umbral gulch Jul 20, 2020, 7:01 PM

#

ill quickly get a brute force solution up 2 secs

candid osprey Jul 20, 2020, 7:01 PM

#

Keep the current approach too for a "always calculate this" approach

#

Actually once you setup the logic to calculate strokes, visualizing different algos should become faster

#

You'll be able to generate exact stroke counts so even if this approach doesn't do it, the code should help

#

As for an algo based version, I keep thinking of trying to pair off colours from each end.

#

I'm wondering if there's something to that.

umbral gulch Jul 20, 2020, 7:02 PM

#

wait for a brute force how do i make it go through it like get all combinations of 2, then all combos of 3 and so on? ive never done that

candid osprey Jul 20, 2020, 7:03 PM

#

You can simply throw* all colours through itertools permutation.

#

Throw*

umbral gulch Jul 20, 2020, 7:04 PM

#

well first its alot easier to have a small colour palette. ill just use the basic one from paint with 20 colours or is that still too many?

candid osprey Jul 20, 2020, 7:04 PM

#

It's a crapton for sure, but go for it

umbral gulch Jul 20, 2020, 7:04 PM

#

i have like 30 different palettes ive borrows from illustrator

quasi oracle Jul 20, 2020, 7:05 PM

#

So I can formulate the problem as follows:
Let's say each color has its own channel. Joining the channels together creates an image in such a way that the channels in the front take priority. The earlier a color is chosen, the more in the back it is.
We want to create a series of channels such that their joining will create the desired image, and the sum of the number of connected components across all channels is minimal.

#

Something like that

candid osprey Jul 20, 2020, 7:05 PM

#

Actually... It is a crapton

quasi oracle Jul 20, 2020, 7:06 PM

#

Connected-row-wise

candid osprey Jul 20, 2020, 7:06 PM

#

What's 20 factorial? 🤔😅

umbral gulch Jul 20, 2020, 7:06 PM

#

ill try this one. its small and nice. only a max of 720 different itterations

📎 unknown.png

#

What's 20 factorial? 🤔😅
@candid osprey big

#

ill use a small file aswell

candid osprey Jul 20, 2020, 7:06 PM

#

Sounds spot on mbaruh

#

Yeah, I think the brute force idea was dumb

umbral gulch Jul 20, 2020, 7:07 PM

#

with 6 brute force is fine

candid osprey Jul 20, 2020, 7:07 PM

#

Calculating total strokes though seems useful for sure

umbral gulch Jul 20, 2020, 7:08 PM

#

ill do it wil this image, small but lots of varying colours

📎 img.png

#

how do i to the iter thing?

#

for L in range(0, len(remaining_colour_positions)+1):
        for subset in itertools.combinations(remaining_colour_positions, L):
            print(subset)

could it be like this?

quasi oracle Jul 20, 2020, 7:14 PM

#

@candid osprey we could potentially combine the two approaches: what is the number of colors that are "in the way" of every color from being unified

#

Pyro I'm on my phone and code is formatted horribly on it so it's a little hard for me to tell

candid osprey Jul 20, 2020, 7:18 PM

#

hm, im trying to understand why you need the nested loop. you should be able to simply run itertools (permutations. not combinations. order matters) on "all" colour choices at once

#

then pick one sequence, and "walk" through it from start to finish

#

so essentially, itertools happens first. the actual code flow happens later

quasi oracle Jul 20, 2020, 7:19 PM

#

Hmm no, not the number of colors. The number of segments

umbral gulch Jul 20, 2020, 7:21 PM

#

def brute_force_strokes(remaining_colour_positions):
    shortest = 0
    best_combonation = None
    for L in range(0, len(remaining_colour_positions)+1):
        for subset in itertools.combinations(remaining_colour_positions, L):
            if subset != ():
                calculated_strokes = get_strokes(subset[0])
                if len(calculated_strokes) < shortest or shortest == 0:
                    shortest = len(calculated_strokes)
                    best_combonation = calculated_strokes
    return best_combonation

this is the brute force way i came up with. it will just return the shortest combonation

#

ill test it now and see if it gives an improvement to the amount of pen strokes it does

quasi oracle Jul 20, 2020, 7:24 PM

#

After that you should try what I just suggested

#

It should give similar results in polynomial time I think

umbral gulch Jul 20, 2020, 7:26 PM

#

After that you should try what I just suggested
@quasi oracle will do

#

ok so i think i done my brute force wrong
without it it took 5649 pen strokes
with it it took 7860 pen strokes

#

plus it took longer th calculate

umbral gulch Jul 20, 2020, 7:59 PM

#

awe i know why its not working

#

its going through every itteration. i only want it to go through if it contains the current colours its checking

#

how can i check if an array contains a sub array

#

ok my brute force just isnt working. and it takes forever with larger pallettes, @quasi oracle how would i impliment the way you suggested

quasi oracle Jul 20, 2020, 8:18 PM

#

GBRBRBRBG. Blue 4 red 3 green 2. 6 strokes with blue first. 5 strokes if G then B then R. Did I count correctly?
Let's take this example

#

The two Gs which are the farthest apart have 7 segments between them

#

The Bs have 5

#

Sorry not 5, 3

#

Excluding the B

umbral gulch Jul 20, 2020, 8:19 PM

#

remember the total amount of pen strokes will be the sum of all the horizontal parts of the image

quasi oracle Jul 20, 2020, 8:19 PM

#

And Rs have 2

naive cloak Jul 20, 2020, 8:20 PM

#

simple question from a noobie here. How do I create an array [0,1,2,3,4.......N] without using for and if's ?

quasi oracle Jul 20, 2020, 8:20 PM

#

list(range(n+1))

#

So the order is GBR.
From that order you can calculate how many strokes each color takes for that row

#

And you can some that across all rows

#

Or.. Hmm

#

You don't need to get the number of strokes

umbral gulch Jul 20, 2020, 8:22 PM

#

@quasi oracle i think thats what my base one is doing

#

maybe the way im doing it just now is the most optimal thinkmon

#

or trying to best a better one will just slow the code down

quasi oracle Jul 20, 2020, 8:24 PM

#

I didn't really understand your code, but if that's what you're doing I don't see something better

umbral gulch Jul 20, 2020, 8:27 PM

#

i guess if your on a phone itll be confusing to read maybe. like i think there is possibly a better solution but in order to get that to work the code would run slower aka then taking longer to draw the image

#

i guess if i understand of uncompressed png save colours that might help because im sure theyll do something similar, ye?

#

i think ive just though of a different way to do it. im off to walk to dog so ill think about it but i think it might works. its taking youd ideas and changding it slightly

umbral gulch Jul 20, 2020, 10:42 PM

#

@quasi oracle ive found a better method of doing it but its slower. do you know of anyway i could speed it up. i think because its looking in an array most of the time is why its slowing it down

def get_better_strokes(color_positions,current_col):
    strokes = []
    pos_index = 0
    while pos_index < len(color_positions):
        start_pos = color_positions[pos_index][0]+paint_offset[0],color_positions[pos_index][1]+paint_offset[1]
        end_pos = color_positions[pos_index][0]+paint_offset[0],color_positions[pos_index][1]+paint_offset[1]
        contains_current_col = color_positions[pos_index] in current_col[1]
        while True:
            try:
                if color_positions[pos_index+1][0]+paint_offset[0] == end_pos[0] + 1:
                    pos_index += 1
                    end_pos = color_positions[pos_index][0]+paint_offset[0],color_positions[pos_index][1]+paint_offset[1]
                    if not contains_current_col and color_positions[pos_index] in current_col[1]:
                        contains_current_col = True
                else:
                    break
            except: break
        if contains_current_col:
            strokes.append((start_pos,end_pos))
        pos_index += 1
    return strokes

first one is the old method. second if that new method

📎 OXBvwgn3U4.mp4

#

fixed 🙂

📎 DNfOJPGnY4.mp4

#

change the get_strokes function to this

def get_strokes(color_positions,current_col=None,single=False):
    strokes = []
    pos_index = 0
    if not single:
        current_col = set(current_col[1])
    while pos_index < len(color_positions):
        start_pos = color_positions[pos_index][0]+paint_offset[0],color_positions[pos_index][1]+paint_offset[1]
        end_pos = color_positions[pos_index][0]+paint_offset[0],color_positions[pos_index][1]+paint_offset[1]
        if not single: contains_current_col = color_positions[pos_index] in current_col
        while True:
            if pos_index+1 < len(color_positions) and color_positions[pos_index+1][0]+paint_offset[0] == end_pos[0] + 1:
                pos_index += 1
                end_pos = color_positions[pos_index][0]+paint_offset[0],color_positions[pos_index][1]+paint_offset[1]
                if not single and not contains_current_col and color_positions[pos_index] in current_col:
                    contains_current_col = True
            else:
                break
        if single or contains_current_col:
            strokes.append((start_pos,end_pos))
        pos_index += 1
    return strokes

umbral gulch Jul 20, 2020, 11:31 PM

#

https://github.com/DylanMcBean/MsPaintDrawer/tree/master if anyone wants to mess with it

GitHub

DylanMcBean/MsPaintDrawer

draws any image based off a colour palette. Contribute to DylanMcBean/MsPaintDrawer development by creating an account on GitHub.

uneven geyser Jul 21, 2020, 11:41 PM

#

!warn @chrome jacinth do not cross-post a question into multiple channels.

halcyon plankBOT Jul 21, 2020, 11:41 PM

#

:incoming_envelope: :ok_hand: applied warning to @chrome jacinth.

slow raven Jul 22, 2020, 2:48 AM

#

Hi I am new

daring axle Jul 22, 2020, 8:47 AM

#

Hey I'm new to C and am having trouble running my program. My compiler is MinGW and I'm on a Windows. I get access denied in my terminal. Not the end of the world, I might use the python tutor C feature to practice my coding.

last blaze Jul 22, 2020, 8:52 AM

#

This isn't the right channel for that question. And AFAIK there is no right channel. You might have better luck looking for a more general programming Discord

#

@daring axle I can DM you invite links to some maybe relevant servers if you like?, I'd post them here but the filter will nab them

fiery cosmos Jul 22, 2020, 9:37 AM

#

I am trying to create a simple Auto Video Editing tool. I am looking to use, OpenCV, FFMPEG, etc. Can you suggest some features and ideas to add? https://github.com/AP-Atul/Video-Editing-Automation

GitHub

AP-Atul/Video-Editing-Automation

A Python based Video Editing Automation with Motion Detection - AP-Atul/Video-Editing-Automation

last blaze Jul 22, 2020, 9:39 AM

#

this isn't the right channel for that @fiery cosmos

random lotus Jul 22, 2020, 10:45 AM

#

how many module are acceptable for competitive programming in python.
I know that math, random and statistics, itertools and some other module is ok in competition. But when I use numpy or sympy in a coding competition of Codeforces or CodeChef I get runtime error.

last blaze Jul 22, 2020, 11:05 AM

#

normally it's just the standard lib that you have access to. Some let you access numpy specifically, but that's not the default

placid sonnet Jul 22, 2020, 11:56 AM

#

@random lotus Ideally you wouldn't use python in competitive programming, C++ instead.

#

But it's the standard library yeah.

mint jewel Jul 22, 2020, 11:57 AM

#

some problems can be solved with python, but indeed C++/D/Rust/Go/other fast langs tend to be better

placid sonnet Jul 22, 2020, 11:58 AM

#

You don't even need to know much C++, as you won't be doing any allocations or freeing or messing with what people use normally

last blaze Jul 22, 2020, 12:19 PM

#

for competitive programming, python is usually reserved just for the simple problems that people can bash out in 2 lines

mint jewel Jul 22, 2020, 12:20 PM

#

I also hear good things about julia in that regard

rain relic Jul 22, 2020, 12:20 PM

#

julia is definitely an interesting lang

royal whale Jul 22, 2020, 12:37 PM

#

Shop={'Apples':'2', 'Bananas':'5'}
Print(Shop)s

Why do we need a s after the parenthesis

quasi oracle Jul 22, 2020, 12:39 PM

#

@royal whale erm.. you don't?
For further help you should open a help channel #❓｜how-to-get-help

royal whale Jul 22, 2020, 12:40 PM

#

Done thanks

balmy siren Jul 22, 2020, 12:56 PM

#

Hey guys, Recently i'm doing dijkstra algorithm and implemented it using Priority Queues.. i got confused little bit here about it's iteration over different vertex.. I mean, is it okay for an naive dijkstra algorithm to iterate over those nodes who have less weight (or say cost) but do not lead to a correct path ?

📎 IMG_20200721_125553.png

#

Here, the neighbors of vertex(1) are 2 and 5 okay ?
So According to PriorityQueue, it dequeues vertex(5) because it has less weight than vertex(2) and enqueue another vertex(6) in PQ

Now, as i said, vertex having less weight will get Dequeue first.. So, it will dequeue vertex(2)... is it okay here to go to vertex(2) after vertex(5) ??

quasi oracle Jul 22, 2020, 1:07 PM

#

You don't know the path to the target node, otherwise you wouldn't need to run the algorithm

#

So yes, it's fine to consider nodes that eventually don't lead you to your destination

random lotus Jul 22, 2020, 1:11 PM

#

@placid sonnet yeah, actually I use both python and C++. If I see a problem is very easy and can be done by 4 or 5 line python where it is not so small in C++ then I use python. But when It requires less time and complexity I use C++. And also I think C++ is best for competitive programming

placid sonnet Jul 22, 2020, 1:21 PM

#

Good.

balmy siren Jul 22, 2020, 1:46 PM

#

So yes, it's fine to consider nodes that eventually don't lead you to your destination
but don't you think it will increase some extra steps?

#

## Implementation of Dijkstra algorithm

from priority_queue import PriorityQueue

def Dijkstra(Gobj, source, destination):
    PQ = PriorityQueue()
    PQ.enqueue(source, 0)
    visited = {source.id: True}
    distance = {source.id: 0}
    parent = {source.id: None}
    found = False
    
    while PQ.is_empty() is False:
        PQ.printQueue()
        vertex, weight = PQ.dequeue()
        # If duplicate entry arrives and the orignal one got deleted
        # but orignal one exists inside dict(distance). 
        if distance[vertex.id] < weight:
            continue
        for eachNeighbor in vertex.GetNeighbors():
            if eachNeighbor.id not in visited:
                # Calculate their distance
                newDistance = distance[vertex.id] + vertex.GetNeighborWeight(eachNeighbor)
                if eachNeighbor.id in distance.keys():
                    if newDistance < distance[eachNeighbor.id]:
                        distance[eachNeighbor.id] = newDistance
                else:
                    distance[eachNeighbor.id] = newDistance
                # Assign parent of eachNeighbor
                parent[eachNeighbor.id] = vertex.id
                PQ.enqueue(eachNeighbor, newDistance)
                if eachNeighbor == destination:
                    found = True
                    break
            
        # Assign it to visited
        visited[vertex.id] = True
        if found:
            break
    
    print(visited)
    print(distance)
    return parent

@quasi oracle this is my implementation of dijkstra using graph algorithm

quasi oracle Jul 22, 2020, 1:56 PM

#

You have no way of knowing in which direction you need to go. The idea of Dijkstra is to search for the destination in the vicinity of the source node, with increasing radius.
If there was some assumption you could make about the graph, then sure, it might be wasteful, but if you have no knowledge of the topology of your graph ahead of time, do you see any way to make it more efficient?
You're saying it's a waste to go in the vertex(2) route because you're looking at the whole graph from above, but in your algorithm, all you know is the vertices you already visited, and their neighbors

balmy siren Jul 22, 2020, 2:07 PM

#

The idea of Dijkstra is to search for the destination in the vicinity of the source node, with increasing radius.
you meant, go through every possible nodes and then find the shortest distance from destination to source.. Just like i've done in this code snippet :

parent = {1: None, 2: 1, 5: 1, 6: 5, 3: 2, 4: 3, 7: 6, 8: 6, 9: 8, 10: 9} # obtained from Dijkstra function.
def printPath(parent, source, destination):
    if source == destination:
        return
    print(parent[destination])
    printPath(parent, source, parent[destination])

And yeah, it returns 9, 8, 6, 5, 1 (of that graph pasted above)

You're saying it's a waste to go in the vertex(2) route because you're looking at the whole graph from above, but in your algorithm, all you know is the vertices you already visited, and their neighbors
Hmm that's right i guess because telling a shortest path by just looking at a graph and obtaining a shortest path via code, both have lot of dfiference.,

if you have no knowledge of the topology of your graph ahead of time, do you see any way to make it more efficient?
Can you pleease tell me when you said "topology of your graph" what did you mean here? Like is it something about which node is pointing to other and in which structure (acyclic, cyclic, directed and other)
I just understood it recently and implemented by myself but i guess the problem is, it doesn't work in Dense graph because it has to go those nodes which are unnecessary .. what about you think, if i'm using a dense graph then what should i use?

#

@quasi oracle

hollow valley Jul 22, 2020, 5:28 PM

#

does anyone know how to create a dictionary given user input

quasi oracle Jul 22, 2020, 5:39 PM

#

@balmy siren I'm not sure I understand your algorithm. You assume your graph is a tree?

balmy siren Jul 22, 2020, 5:41 PM

#

@quasi oracle nope.. it's not a tree, it is a graph tho (directed graph)

quasi oracle Jul 22, 2020, 5:41 PM

#

I don't see the 5->4 edge

#

Is this the shortest distances graph?

balmy siren Jul 22, 2020, 5:46 PM

#

ohh well i mean, i just took this graph as an example to implement dijkstra on it..
so you thought it isn't that correct ?

#

and does it make any difference or dijkstra ones are different ?

quasi oracle Jul 22, 2020, 5:53 PM

#

It just doesn't seem like you could represent a general graph that way

#

Also, in order to properly direct all edges, you sort of have to run a variation of Dijkstra anyway

quasi oracle Jul 22, 2020, 10:03 PM

#

@grim hedge I think you'll get more informative answers in #career-advice

grim hedge Jul 22, 2020, 10:04 PM

#

@quasi oracle Thank you, didnt even notice that section 🙂

jolly sigil Jul 23, 2020, 1:38 AM

#

Starting here:
pd.DataFrame.reset_index(df, level=None, drop=False, inplace=True, col_level=0, col_fill='')
df.rename(columns={'index':'pcount'}, inplace=True)
returns a warning, "SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame"
How should I do this to avoid the warning?

oblique panther Jul 23, 2020, 4:18 AM

#

@viscid mountain this is a better place to ask about algorithms and data structures in general.

#

When I took that course I looked up the algorithms on Geeks for Geeks and reimplemented them my way.

placid sonnet Jul 23, 2020, 6:16 AM

#

GeeksforGeeks saved all of us, even if the explanations were a bit crap at times.

torn scaffold Jul 23, 2020, 6:26 AM

#

🙂

sterile hinge Jul 23, 2020, 6:18 PM

#

how to transistor logic?

marble basalt Jul 24, 2020, 2:27 AM

#

You mean like boolean algebra?

balmy siren Jul 24, 2020, 12:39 PM

#

@magic locust hey are you doing graphs ?

#

oh yeah you're 😄
actually i want to ask something if you've done upto shortest path algo

magic locust Jul 24, 2020, 12:41 PM

#

yes

fiery cosmos Jul 24, 2020, 12:47 PM

#

I love fortnite

#

yes
@magic locust Hey

balmy siren Jul 24, 2020, 12:52 PM

#

@magic locust can you tell me how did you store the edges of undirected Graphs ??
One method which i'm thinking right now is to, when user input the two vertex then first it creates (if it doesn't exist) and then connect them in this way :
adjacencyList[A] = B
adjacencyList[B] = A
am i right here? or can you show me your implementation of undirected graphs???

I used to insert both (A and B) two times if i've to make it as undirected graph but might be there's best way to do so

magic locust Jul 24, 2020, 1:09 PM

#

@magic locust Hey
@fiery cosmos Cool

#

@magic locust can you tell me how did you store the edges of undirected Graphs ??
One method which i'm thinking right now is to, when user input the two vertex then first it creates (if it doesn't exist) and then connect them in this way :
adjacencyList[A] = B
adjacencyList[B] = A
am i right here? or can you show me your implementation of undirected graphs???

I used to insert both (A and B) two times if i've to make it as undirected graph but might be there's best way to do so
@balmy siren you can use adjacency matrix

balmy siren Jul 24, 2020, 1:10 PM

#

any guess implementing it using adjacency list?

magic locust Jul 24, 2020, 1:16 PM

#

implementing what exactly ?

balmy siren Jul 24, 2020, 1:33 PM

#

@magic locust recently i'm doing dijkstra but i got a problem because the graph which i'm using was an undirected graph and i'm using an adjancency list to represent the connection between vertices
so i want to know how do i exactly represent undirected graph..
basically, we do something like this if we're using an adjacency list ->

A -> B, C, D
B -> C 
D -> A

you familiar with this?

daring axle Jul 24, 2020, 1:57 PM

#

@daring axle I can DM you invite links to some maybe relevant servers if you like?, I'd post them here but the filter will nab them
@last blaze Please send them. I tried switching over from VS Code to Notepad++ and that still won't let me access my file.

fiery cosmos Jul 24, 2020, 2:20 PM

#

I hope not

dark egret Jul 24, 2020, 3:24 PM

#

@balmy siren you can use a dictionary of lists or sets. Using defaultdict would make implementation simple

import collections
graph = collections.defaultdict(list) # graph = collections.defaultdict(set)

balmy siren Jul 24, 2020, 4:53 PM

#

@dark egret i already implemented it using arrays what i want to know i.e., if i'm going to add an undirected edge then is it right to store both the vertices like this (below) in the array :

adjacencyList[A] = B
adjacencyList[B] = A

self.vertList 
[(A, B), (B, A)]

quasi oracle Jul 24, 2020, 5:17 PM

#

vertlist implies it's a list of vertices though, not edges

#

The adjacency lists can give you all the information you need about the structure of the graph, I don't think there's a point in having another list

dark egret Jul 24, 2020, 8:29 PM

#

Although it seems kind of redundant to store the undirected edge A-B
as

adjacencyList[A] = [B]
adjacencyList[B] = [A]

I think it is really helpful(in terms of time complexity) especially if you are running an algorithm like Djikstras. I agree with @quasi oracle that storing an edge once would capture all the information in the graph but it would be fast.

If you just store

adjacencyList[A] = [B]
adjacencyList[B] = []

then while running Djikstras if you encounter B before A then you wont be able to reach A because you haven't stored A as a neighbor of B

Also, for problems where you need to check if a node is a neighbor of another, I would prefer sets to store neighbors as you can do this in O(1) time

adjacencyList[A] = {B, ...}
adjacencyList[B] = {A, ...}

#

@balmy siren

#

Hope it makes sense. Also correct me if I am wrong

quasi oracle Jul 24, 2020, 8:31 PM

#

Ah, I should clarify.

adjacencyList[A] = [B]
adjacencyList[B] = [A]

Is good.

I was referring to

self.vertList 
[(A, B), (B, A)]

The case of:

adjacencyList[A] = [B]
adjacencyList[B] = []

Implies a directed graph

dark egret Jul 24, 2020, 8:36 PM

#

oh alright
I misunderstood

icy fulcrum Jul 25, 2020, 12:29 AM

#

This is a minecraft region (60 gig in data) under active pregeneration. each color is mapped to to the sha256 checksums of the regions using a bit mask. the data is rendered in Povray and then animated using moviepy. as you can see from the chunk gen there are a number of identical map tiles on the map in a noise pattern that i think aligns with Perlin Noise Generator. Due to the nature of sha256 checksums a single change in a block inside the game will cause an avalanche effect on the 256bit hash. slicing this hash into a 16 bit value gives a color. each frame in the animation provided is a 15 minute snapshot of sha256 sums of the regions if you look closely at the map you can see how much terrain is being effected by the map pregen. thank you to all the people who helped with this project 🙂

📎 region.mp4

radiant gate Jul 25, 2020, 1:24 AM

#

@austere sparrow @candid osprey hii

candid osprey Jul 25, 2020, 1:39 AM

#

Hi. I stepped away for a bit. I got the gist of it already though, not a lot to add the the convo

naive tinsel Jul 25, 2020, 8:31 AM

#

Hey everyone, I have script that I cannot get to work and need some guidance. It iterates through a directory tree searching the containing files for a phone number. On the first iteration it breaks and returns a FileNotFound error, the problem is that not only does the file exist but the script does identify the file and puts its path into a viariable, which the traceback points to, so I know that it is being accessed, I am not sure what I am doing wrong so hopefully someone will see something in the code that I am overlooking.

# Search through files for a phone  number
# Imports

import os, re

# Set search pattern
pattern = r'\d{3}-\d{3}-\d{4}'
path = '/home/dschuster/Udemy_Devs/Udemy_Python/Homework/extracted_content'

# Search function
def search(file, pattern=r'\d{3}-\d{3}-\d{4}'):
    f = open(file, mode='r')
    text = f.read()

    if re.search(pattern, text):
        return re.search(pattern,text)

results = []

# Walk through files
for root, dirs, files in os.walk(path):
    for f in files:
        full_path = path + '/' + f
        results.append(search(full_path))

# Examine for pattern match
for number in results:
    if number == '':
        continue
    else:
        print(number.group())

#

Here is the Traceback:

Traceback (most recent call last):
  File "/home/dschuster/Udemy_Devs/Udemy_Python/Homework/unpack.py", line 24, in <module>
    results.append(search(full_path))
  File "/home/dschuster/Udemy_Devs/Udemy_Python/Homework/unpack.py", line 12, in search
    f = open(file, mode='r')
FileNotFoundError: [Errno 2] No such file or directory: '/home/dschuster/Udemy_Devs/Udemy_Python/Homework/extracted_content/XHZPVUQTXIO.txt'

#

As you can see in the above path, the file is identified! I can't get my head around this one.

last blaze Jul 25, 2020, 8:44 AM

#

@naive tinsel you're probably gonna have better luck in a normal help channel. See #❓｜how-to-get-help

worthy flicker Jul 25, 2020, 12:27 PM

#

!rank

halcyon plankBOT Jul 25, 2020, 12:27 PM

#

Iterating over range(len(...)) is a common approach to accessing each item in an ordered collection.

for i in range(len(my_list)):
    do_something(my_list[i])

The pythonic syntax is much simpler, and is guaranteed to produce elements in the same order:

for item in my_list:
    do_something(item)

Python has other solutions for cases when the index itself might be needed. To get the element at the same index from two or more lists, use zip. To get both the index and the element at that index, use enumerate.

naive tinsel Jul 25, 2020, 4:59 PM

#

@naive tinsel you're probably gonna have better luck in a normal help channel. See #❓｜how-to-get-help
@last blaze Ah... lol didn't realize... haha thank you!

serene mango Jul 26, 2020, 7:37 AM

#

@magic locust Regarding our previous discussion, del for a slice won't create a copy. So it is truly in-place. It interprets the slice a bit differently, all it does is, lst[i:] tells del to delete items from some index. Since del is a special syntax, it inteprets lst[i:] a bit differently.

river jolt Jul 26, 2020, 7:46 AM

#

hey I have a subdomain[example.demonx.ex] I want to open it in python and it should return a protocol and the subdomain for me eg.[http://example.demonx.ex] how do I do this? Do I need sockets or the urllib module can handle this for me

magic locust Jul 26, 2020, 7:54 AM

#

@magic locust Regarding our previous discussion, del for a slice won't create a copy. So it is truly in-place. It interprets the slice a bit differently, all it does is, lst[i:] tells del to delete items from some index. Since del is a special syntax, it inteprets lst[i:] a bit differently.
@serene mango Thanks for the clarification

serene mango Jul 26, 2020, 9:13 AM

#

pystrong

mystic osprey Jul 26, 2020, 9:28 AM

#

hey ,Can I get a help with C?

last blaze Jul 26, 2020, 9:31 AM

#

@mystic osprey you might be able to in the offtopic channels, but it's possible you might have better luck in another discord. Or if it's specifically writing a python extension, maybe #c-extensions

mystic osprey Jul 26, 2020, 9:31 AM

#

umm thanks!

west pawn Jul 26, 2020, 2:57 PM

#

Hey guys I need help with visualizing an exponential function inside a for loop. I understand that if we have n number of bits that gives us 2^n different combinations that these bits can be ordered in. What I’m trying to understand is the doubling operations part. I just can’t visualize in my head what it means by doubling the operations. Is there a real life example of this somewhere?

📎 image0.png

fiery cosmos Jul 26, 2020, 3:06 PM

#

Is this what you mean?

2^1 = 2
2^2 = 4
2^3 = 8
2^4 = 16
2^5 = 32
...

west pawn Jul 26, 2020, 3:08 PM

#

Yeah but I’m trying to understand how the number of bits has anything to do with the number of operations that we perform inside the for loop, for example I’m trying to understand why the number of operations inside the for loop would double.

quasi oracle Jul 26, 2020, 3:10 PM

#

I guess one example could be to simply print all numbers of up to n bits
In the first iteration it would be 0 and 1
In the second iteration: 00, 01, 10, 11
In each iteration the number of values you need to print doubles

west pawn Jul 26, 2020, 3:11 PM

#

Ah that makes much more sense, thank you!

quasi oracle Jul 26, 2020, 3:12 PM

#

There is another example that can be made with the analysis of merge sort, but to not confuse you it's better to study it as a whole

#

A breadth-first-search on a full binary tree can also be an example

west pawn Jul 26, 2020, 3:15 PM

#

I see, I’m sure the book I’m studying will introduce me to those later on, meanwhile I’ll take a quick look on yt to get a gist of how they work

quasi oracle Jul 26, 2020, 3:16 PM

#

If you want to delve into it I'd start with the tree example, the merge-sort example is a little less trivial, because it doesn't end with an exponential function

west pawn Jul 26, 2020, 3:19 PM

#

Okay I’ll do that, thank you

agile sundial Jul 27, 2020, 12:30 AM

#

try to draw a binary tree

#

the amount of nodes increases exponentially as you go farther from the root node

eternal bronze Jul 27, 2020, 6:54 AM

#

I need to wait for a process to finish before executing the next command in my py script. Is there a better way to do it than using time.sleep()??

findStaticFolder(fileToImport)  #imports a file inside the game engine
os.remove(fileToImport)  #should delete the file from it's location after it is imported inside the game engine but is getting executed before the previous task has finished.

Since the importing time can differ, I don't wanna use sleep().

mortal arch Jul 27, 2020, 8:51 AM

#

Make the script remove itself?

#

I mean, the script that loads the file

#

After its done, make it remove the file

loud trail Jul 27, 2020, 3:29 PM

#

@eternal bronze how are you running the background process

#

Oh i see. I recommend using a ThreadPoolExecutor to run this

#

you can do other work after the job has been submitted

#

Then when you're ready you can go back and wait for the job to finish

#

Ill post an example when i get to a computer

oblique panther Jul 27, 2020, 4:09 PM

#

Suppose you have a number of numerical ranges

#

do you think there's an O(1) algorithm to figure out which of those ranges a given integer belongs to?

#

(if any?)

quasi oracle Jul 27, 2020, 4:19 PM

#

If you can assume the list of ranges is ordered, you can make it in O(logn). Otherwise, the mere act of finding out what ranges you even have will take O(n)

#

I don't see much of a difference from an ordinary search in a list

mint jewel Jul 27, 2020, 4:22 PM

#

there is an O(1) algorithm assuming all ranges are finite and you are allowed to precompute things by just exploding all the ranges into a list of booleans and indexing into it (checking bounds naturally)

range(1, 3), range(4, 6) -> [0, 1, 1, 0, 1, 1, 0]

gusty grove Jul 27, 2020, 4:23 PM

#

are they non-overlapping?

quasi oracle Jul 27, 2020, 4:28 PM

#

Yeah, it crossed my mind, but didn't seem like a general algorithm for any set of ranges considering the space requirements

oblique panther Jul 27, 2020, 4:48 PM

#

Otherwise, the mere act of finding out what ranges you even have will take O(n)
@quasi oracle

#

Excellent point

#

I was thinking there might be a way to do it with hashing but I'm never going to have a high enough value of n for it to matter.

quasi oracle Jul 27, 2020, 4:53 PM

#

Like lak said you can explode the ranges

#

But you can place them in a set, and then not waste space on values that aren't in a range

oblique panther Jul 27, 2020, 4:56 PM

#

I have a bunch of spacy Span objects, which know their character spans in the document they came from, so I just need a mapping of character spans to Spans.

#

def find_sentence(start: int, sentences: t.Dict[range, Span]):
    for r, s in sentences.items():
        if start in r:
            return s
    raise RuntimeError(f'start value {start} not in any range')

quasi oracle Jul 27, 2020, 4:57 PM

#

The preprocessing still has to be O(n) though

oblique panther Jul 27, 2020, 4:57 PM

#

right

eternal bronze Jul 27, 2020, 7:50 PM

#

@loud trail @mortal arch hey thanks guys! I found another way to do it inside the game engine.

gleaming ridge Jul 28, 2020, 7:43 AM

#

is it possible to write pure functions without doing functional programming stuff like recursion, map, reduce, filter, higher-order functions and still reap the benefits of FP? Those FP stuff are not easy to master.

runic tinsel Jul 28, 2020, 7:51 AM

#

what benefits then?

#

the vaguest question I've heard today

mint jewel Jul 28, 2020, 8:47 AM

#

you can write entirely pure functions and without using HOFs and still get all the benefits in easy to reason about code, but without HOFs it is extremely difficult to actually get something done

gleaming ridge Jul 28, 2020, 8:50 AM

#

you can write entirely pure functions and without using HOFs and still get all the benefits in easy to reason about code, but without HOFs it is extremely difficult to actually get something done
@mint jewel good to hear that. HOFs need some mental adjustment. Not easy. I think pure functions shouldn't be too hard. Avoid global variables, avoid call by reference, use immutable input data etc Those are easy guidelines to follow

mint jewel Jul 28, 2020, 8:52 AM

#

I would suggest not trying to do fully pure FP in python. Keeping most function pure is a good thing, but you cannot really do effects with pure functions in python like you can in e.g. haskell

gleaming ridge Jul 28, 2020, 8:53 AM

#

A good practice would be to use pure functions as much as possible unless the particular application doesn't allow that. Maybe one can mix FP and non-FP code with python since python supports some FP features

pliant crane Jul 28, 2020, 1:25 PM

#

Has anyone used a virtual machine before

#

I'm planning on setting it up

#

Even tho I don't have the CPU resources

summer reef Jul 28, 2020, 2:11 PM

#

I did, with Virtualbox on Ubuntu.
It's pretty straightforward, in fact, you install your OS like on a genuine machine.
Depending on your needs, you may want to add the Guest Additions, though (typically, to share a folder). It's like a driver to install on the guest machine.
As for your CPU, it really depends on what you have, and what you plan to do with your VM. If it works on the physical machine, and you just want to virtualize it (move a task from physical machine to virtual one), it shouldn't be a problem. I would be more concerned with the RAM space, as you will have the equivalent on two OS on the same space.

frail valve Jul 28, 2020, 7:31 PM

#

can - W10 based , 4 core @ 2Ghz , 16 Gig mem ok for virtual machine?

shut osprey Jul 28, 2020, 7:38 PM

#

You can run a VM on pretty bare hardware, it will just be slow

fiery cosmos Jul 29, 2020, 11:23 AM

#

hey guys

#

does anyone know a good way to learn the basics of computer science?

runic tinsel Jul 29, 2020, 11:25 AM

#

they are all roughly equally good

marble basalt Jul 29, 2020, 4:57 PM

#

@fiery cosmos computer science is a huge subject with a fairly unclear progression

#

What specifically are you interested in?

chrome jacinth Jul 29, 2020, 8:02 PM

#

Hello guys, I have a dictionary that looks like this:

          'Mammal': [('Mammal','Bear'),('Mammal','Tiger'), ('Mammal','Dolphin')]```


Assuming that this is a network graph, I am trying to get connected components from a network graph with the following algorithm 
```def get_all_connected_groups(d):

    already_seen=set()

    result=[]

    for node in d:

        if node not in already_seen:

            connected_group,already_seen=get_connected_group(node,already_seen)

            result.append(connected_group)

    return result

def get_connected_group(node,already_seen):

    result=[]

    nodes=set([node])

    while nodes:

        node=nodes.pop()

        already_seen.add(node)

        nodes.update(n for n in d[node] if n not in already_seen)

        result.append(node)

    return result,already_seen

components=get_all_connected_groups(d) ```
 
However it stops with the second element of the dictionary 
```('Bird','Falcon')```
 and gives me an error  highlighting this line 
```nodes.update(n for n in d[node] if n not in already_seen)```

I think is because the example uses numbers as the nodes and I am using strings but I am not sure, any help works!

quasi oracle Jul 29, 2020, 8:10 PM

#

@chrome jacinth this is better suited for a help channel. See #❓｜how-to-get-help

chrome jacinth Jul 29, 2020, 8:10 PM

#

Thanks!

serene mango Jul 29, 2020, 8:21 PM

#

Suppose I have a function which can return a random integer within a given range of integers then how can I make it such that it return a unique integer at every call.
I know about Fisher-Yates shuffling, but are there any other alternatives? (I am not looking from an optimization perspective, I just want to know an alternative method, But it should converge accordingly), but of course we have choosing a random integer and storing the visited elements in a set and then we can keep picking a random integer but it will converge slowly or may juggle within the set for the final element for a large set of numbers (for an unacceptable amount of time, I know the term unacceptable is vague here, but the point is, it should converge effectively). So if there is any feasible alternative then please recommend.

runic tinsel Jul 29, 2020, 8:30 PM

#

shuffle the range, give numbers from it, then shuffle a new one when it runs out

#

@serene mango

#

I guess you;re asking about low level implementation, I was just thinking python where shuffle is built in

serene mango Jul 29, 2020, 8:39 PM

#

Yes, using inbuilt shuffle kills the purpose, how to shuffle it?, that's one of the simplification of the problem statement(given we are relating shuffling and a random unique integer like that). I want to work on it from scratch with the only condition that I have a function which can give me random integer within a given range. So for python that would be random.randrange(low, high +1)
@runic tinsel

runic tinsel Jul 29, 2020, 8:40 PM

#

instead of building a set of already rolled numbers, remove them from the range?

#

it kinda still has performance issues

serene mango Jul 29, 2020, 8:44 PM

#

I am not worried about the performance issues now as in O(x) thing, the only condition is it should converge effectively because I already have Fisher-Yates with the best performance.
How do you suggest removing from the range will work? I mean, I would have to tell the algorithm that such numbers are removed and then they would have to be stored somewhere for me to tell the algo, then in the end, wouldn't it be same as storing in a set?

runic tinsel Jul 29, 2020, 8:46 PM

#

like you have an array of them [1,2,3,4,5], you roll a 3 replace it with the next number: [1,2,4,4,5], then you roll a 4, replace all 4: [1,2,5,5,5]. I'm just making this up nevermind

#

oops biased too

serene mango Jul 29, 2020, 8:50 PM

#

I was thinking what can I do on top of this to somehow make this work, but I got nothing xD'
xD' A decent attempt though.

runic tinsel Jul 29, 2020, 8:56 PM

#

i'm reading this link http://okmij.org/ftp/Haskell/perfect-shuffle.txt from wikipedia article on fisher yates and I don't understand it lmao

#

ok they're saying the random number list has a zero appended for some reason

serene mango Jul 29, 2020, 9:02 PM

#

Oh fisher-yates is very simple and kinda really not that smart but its interesting.
so u simply roll a number and swap it with the number at first index, and then you roll from the numbers from second index to the end, and whatever u get, u swap it with second index, and u keep on doing this till the end. lol

#

u not only get random, but u get uniform random .. O(n) time and O(1) space xD;

runic tinsel Jul 29, 2020, 9:03 PM

#

they aren't doing fisher yates, it's completely different they just start with a naive one and it has weird phrasing

serene mango Jul 29, 2020, 9:07 PM

#

the whole article is hard to read 😐

#

The article is labelled 'Perfect Shuffle', isn't perfect shuffle more like a permutation, rather than a random process?

runic tinsel Jul 29, 2020, 9:10 PM

#

yeah they didn;t know that they correct it the footnotes

#

how about, you add rolled numbers into a sorted list, you roll a smaller range and you find how many numbers you skipped and add that number to the rolled one

#

so it's O(nlogn) time?

serene mango Jul 29, 2020, 9:16 PM

#

Can you please explain on what do you mean by add that number to the rolled one, you mean, these numbers to be in the list to be rolled?

runic tinsel Jul 29, 2020, 9:17 PM

#

say you alrady rolled [5,7,3,2] previously. You don't have anything else in memory.

#

oh it's sorted, [2,3,5,7]

#

you roll a random number in range (0, n-4)

#

say it's 5

#

you search [2,3,5,7] and find out [2,3,5] are "on the left side" of 5, so you add 3

#

5+3 is 8

#

each roll takes (log n) time to update the list and (log n) again to find out the correction

#

guess it's the same operation, so just once

#

no it's not the same you find the position of 5 but you update with 8

#

i'm confused too

serene mango Jul 29, 2020, 9:24 PM

#

But this won't guarantee uniqueness, it might happen that when you find 5 then 5 + 3 = 8 is already in the list, and it may happen that all elements 5 +2 = 7 and all elements before 5 are in the list

#

then how do we guarantee uniqueness

runic tinsel Jul 29, 2020, 9:25 PM

#

yeah it doesn't work out

#

imagine our range [ . X X . . ] we already rolled Xs, our list is [1, 2] and we roll 0-2 because that's how many dots are left

#

we get 1, there's one number <= 1 in the list so we correct it to 2, and we repeat this because it changed the position, when it stablizes we got our number

#

we already know we rolled the second dot, we just need to know the actual position of that second dot

#

we simply need to correct it without doing the binary search again

#

Surely there's a complete algorithm in Knuth he like did all the shuffle methods on earth. Although he wasn't doing python so maybe not this one exactly

shut osprey Jul 29, 2020, 9:38 PM

#

You already implemented fisher-yates?

runic tinsel Jul 29, 2020, 9:38 PM

#

no

shut osprey Jul 29, 2020, 9:38 PM

#

Can't you just use like a queue and pop elements off it

serene mango Jul 29, 2020, 9:40 PM

#

import random

def rnd(low, high):
    return random.randrange(low, high + 1)


def shuffle(the_list):
    if len(the_list) <= 1:
        return the_list
    
    last_ind = len(the_list) - 1
    for i in range(0,last_ind):
        ran = rnd(i, last_ind)
        if ran != i:
            the_list[i], the_list[ran] = the_list[ran], the_list[i]

This is my fisher-yates implementation

#

we get 1, there's one number <= 1 in the list so we correct it to 2, and we repeat this because it changed the position, when it stablizes we got our number
@runic tinsel Can you please, explain this bit, or give a short example of your approach, like with 5 elements or so? I am having trouble understanding the stabilization part.

#

Can't you just use like a queue and pop elements off it
@shut osprey Will that be uniform though?

shut osprey Jul 29, 2020, 9:42 PM

#

wdym uniform

#

The shuffling algorithm is what makes it random

#

you are just getting one element at a time from the shuffled list

#

like if you shuffle a deck of cards well, the probability of drawing any given card is the same

runic tinsel Jul 29, 2020, 9:44 PM

#

we're trying to avoid fisher yates, so any other shuffling or no shuffling

serene mango Jul 29, 2020, 9:44 PM

#

you are just getting one element at a time from the shuffled list
@shut osprey not all shuffling is uniform .. for example: simple swapping

#

we're trying to avoid fisher yates, so any other shuffling or no shuffling
@runic tinsel yeah and this.

runic tinsel Jul 29, 2020, 9:47 PM

#

you rolled 4 out of a smaller range and previously rolled [2,3,5,7]

shut osprey Jul 29, 2020, 9:48 PM

#

import random
def random_walk(list):
  while list:
    index = random.randint(len(list))
    yield list.pop(index)

#

Can't you just do this

runic tinsel Jul 29, 2020, 9:48 PM

#

you see that you're like off by 2, immediately you change 4 to 6, but that means you actually past 5 too, so you add 1 more, you get 7, 7 collides again, 8

#

8 is great

#

intuitively this stutter stepping could be done without binary search but it's probably wrong, and you should just search again on the right side

#

[ . . X X . X . X . . X] we can see that fifth dot is indeed in position 9

serene mango Jul 29, 2020, 9:54 PM

#

intuitively this stutter stepping could be done without binary search but it's probably wrong, and you should just search again on the right side
@runic tinsel
Yeah and then we would have to search the right side, and for an overflow in range we ignore it and roll again or we use a modulo on the range peak. I will try evaluating this one mathematically and see what could the edge case be.

runic tinsel Jul 29, 2020, 9:55 PM

#

we don't overflow the range

#

the range decreases by one after each roll

#

what do you mean

serene mango Jul 29, 2020, 9:57 PM

#

Oh, that was just for the edge, I mean when we add 1 more. But I get the idea.

#

Can't you just do this
@shut osprey yeah that works too. Uniqueness will be guaranteed by popping a visited element.

#

But that is kinda an implementation of fisher yates, I mean instead of swapping with an initial index(and move on with it), we pop this index and move on

#

@runic tinsel thanks for the detailed approach, I will work it out and see if something comes up in it, will mention you if that happens.

runic tinsel Jul 29, 2020, 10:17 PM

#

nice

formal rock Jul 30, 2020, 1:54 AM

#

What's a good api for image recognition

keen estuary Jul 30, 2020, 5:44 AM

#

Vision AI???

#

umm I need a small help

#

hey I cant unmute on the general voice channel

plain heath Jul 30, 2020, 7:26 AM

#

ok.

somber wind Jul 30, 2020, 1:56 PM

#

I'm looking for a package suitable for spatial indexing and spatial queries (2d and 3d). It would be nice if someone at least somewhat familiar with one of those could recommend one.

Suitable in my case means:

thread-safe (which rules out https://toblerity.org/rtree/)
no unnecessary data structure baggage such as Pandas data frames. Bare minimum please (such as tuple, list or numpy array based) (which rules out https://geopandas.org/)
sufficiently well-performing for what it does. Indexing and querying a set of objects amounting to a magnitude of about 100 to 1000 should be worthwhile - in comparison to naive iteration over all those objects. Indexing does not need to be particularly fast. All objects will be indexed once at the beginning of the run, whereafter there will be about a million or several million queries into the same index.
maintained at least in the last 12 months and usable from modern Python (i.e. 3.6 or higher)

loud trail Jul 30, 2020, 6:26 PM

#

@somber wind have you considered postgres + postgis?

somber wind Jul 30, 2020, 6:37 PM

#

@loud trail I don't want a database. No storage. I just want a lightweight python module/package to integrate with my application. The objects are just needed for the run and are not that many. They don't need long-term storage. A full-blown database is not justified for my use-case.

#

Just an R-tree module (or equivalent approach) that is also thread-safe (in contrast to the above mentioned module).

loud trail Jul 30, 2020, 6:42 PM

#

i see. thread-safety might be a bit tricky

#

you need concurrent inserts/updates?

somber wind Jul 30, 2020, 6:42 PM

#

No, just concurrent queries.

loud trail Jul 30, 2020, 6:42 PM

#

is an exclusive lock for inserts ok?

#

rtree doesnt support concurrent queries?

somber wind Jul 30, 2020, 6:42 PM

#

it does not

loud trail Jul 30, 2020, 6:43 PM

#

is there a C library that works for your use case? you could write a python lib to wrap it

#

evidently rtree wraps libspatialindex

somber wind Jul 30, 2020, 6:43 PM

#

yes

#

which isn't thread-safe

loud trail Jul 30, 2020, 6:47 PM

#

i see

#

how about spatialite?

#

after all, sqlite is just a file on disk and fully in-memory databases are supported

#

you might have to build your own version of the sqlite3 python module to support it

somber wind Jul 30, 2020, 6:48 PM

#

Still, I'd like to avoid the added complexity of databases.

loud trail Jul 30, 2020, 6:49 PM

#

sql syntax seems like a small price to pay for a fully in-memory postgis equivalent

somber wind Jul 30, 2020, 6:49 PM

#

I'm really just looking for a recommendation of someone who has used a spatial query module. I'm going without spatial queries for now.

loud trail Jul 30, 2020, 6:52 PM

#

maybe re-ask your question in #databases where someone with experience is more likely to see it

#

and i really dont think sqlite is that much more complexity - you have very specific needs and it appears to meet them

#

@somber wind you can also look into PyGEOS which uses numpy but not pandas. not sure about its thread-safety though

red solstice Jul 30, 2020, 6:56 PM

#

You want to try spatial queries in 3d without a database? If you can find it, I'd be interested to hear about it!

somber wind Jul 30, 2020, 6:57 PM

#

Well, the rtree module mentioned above fits the bill, apart from thread-safety.

red solstice Jul 30, 2020, 6:59 PM

#

what's the trade off between thread-safety and the data integrity that a database would provide?

somber wind Jul 30, 2020, 7:02 PM

#

data integrity? not sure how that's relevant for my use-case

loud trail Jul 30, 2020, 7:02 PM

#

@red solstice the underlying library appears to just segfault if you use multiple threads even for read only queries

#

So its not a matter of data integrity in this particular case

quasi oracle Jul 30, 2020, 9:35 PM

#

You could look at scipy's cKDTrees. I suppose you can achieve parallelism by copying the data structure created, but I don't know if it suits your needs

frail valve Jul 30, 2020, 10:52 PM

#

i want to understand how other coders do stuff -- but i feel i have learned - BAD HABITS -- i want to have better more standard code -- is ther a good tutorial that can step me through - i want to do this so i can read what you do
and also i can show stuff

loud trail Jul 31, 2020, 1:16 AM

#

@frail valve maybe post a code sample that you are concerned about. I don't know of a good single resource. Reading code for big libraries can be enlightening. And read the PEP8 and Google style guides.

frail valve Jul 31, 2020, 1:17 AM

#

mmm ok thanks

#

is to get overload - so i ask first

#

easy to get overloaded - i ask first - time saver

#

my code is too crappy to show -- maybe one day

serene mango Jul 31, 2020, 4:58 AM

#

@runic tinsel Hey that worked well, with a slight modification!
Just letting you know!

runic tinsel Jul 31, 2020, 7:53 AM

#

@serene mango cool so it's O(n²)?

unique cape Jul 31, 2020, 11:01 AM

#

WELP

serene mango Jul 31, 2020, 11:59 AM

#

@serene mango cool so it's O(n²)?
@runic tinsel I haven't done the asymptotic analysis, but if we use Dynamic Programming then it could be done in less than O(n²) I guess. I will try!

fiery cosmos Jul 31, 2020, 12:56 PM

#

what are the things we can do in python

#

?

ionic zenith Jul 31, 2020, 2:20 PM

#

what are the things we can do in python
@fiery cosmos Not just python you have to use other modules and frameworks to take the most out of it

gaunt forge Aug 1, 2020, 12:43 AM

#

There's a lot of things you can do in Python that there really isn't much accessible outside of it

#

The public library support is just so robust

sudden lance Aug 1, 2020, 2:22 AM

#

Hi, I was wondering if there is like any free software to know if code is plagiarize or similar to an online source .

marble basalt Aug 1, 2020, 2:57 AM

#

I doubt it, good code is usually pretty darn similar to other good code doing the same thing

sudden lance Aug 1, 2020, 3:54 AM

#

Just want to look at something to gain an idea but don’t want to copy it fully

cunning flame Aug 1, 2020, 4:38 AM

#

a quick google search for "code plagiarism checker" yields some interesting results. but yeah i think generally its agreed that you cant really plagiarize code lol, a lot of CS is abstraction and building upon other people's published work

tacit sleet Aug 1, 2020, 8:43 AM

#

can anyone take a look at my code and tell me if its good or not? if for a minigame that me and my friends have made

potent pier Aug 1, 2020, 10:24 AM

#

anyone have any advice to start learning phyton?

arctic dome Aug 1, 2020, 3:34 PM

#

@potent pier don't look at it as if learning python, but programming in python.

potent pier Aug 1, 2020, 3:40 PM

#

so, anyone have any advice to start programming phyton?

solemn topaz Aug 1, 2020, 5:41 PM

#

I would start on udemy that is how I started

quick granite Aug 1, 2020, 8:23 PM

#

thenewboston has good Python Tutorials but like @arctic dome learn concepts to problem solve, make algorithms and think like a programmer

marble basalt Aug 2, 2020, 12:22 AM

#

Regardless of where you learn, make sure your always making things

#

Showing it to other people and asking if it's good can also be super helpful

fiery cosmos Aug 2, 2020, 12:43 AM

#

Hi I am trying to use a raspberry pi, I don't know what the hell i am supposed to do. Anybody know what I should start with to learn how to make a raspberry pi work?

marble basalt Aug 2, 2020, 2:48 AM

#

It's just a computer without a case

#

https://www.raspberrypi.org/help/noobs-setup/2/

Raspberry Pi

Raspberry Pi NOOBS Setup

To get started with Raspberry Pi you need an operating system. Raspberry Pi NOOBS (New Out Of the Box Software) is an easy operating system install manager.

#

Heres a guide on first time setup

fallen field Aug 2, 2020, 7:47 AM

#

Hey, can anybody here help me open a .res sound file?

#

I can't open it as it requires c++ or other visual studio to be opened.

gentle spoke Aug 2, 2020, 12:09 PM

#

@fiery cosmos congrats on the pi! It's really easy

#

I'd strongly encourage you to learn about ssh and headless setup

#

I use my raspberry pi over the terminal 99% of the time

fiery cosmos Aug 2, 2020, 4:15 PM

#

@gentle spoke what is ssh and headless setup and where should i learn about it

gentle spoke Aug 2, 2020, 4:15 PM

#

basically, ssh is a protocol of communication

#

"Secure Shell is a cryptographic network protocol for operating network services securely over an unsecured network."

#

it's the standard way for easy, secure access over computers over the web

fiery cosmos Aug 2, 2020, 4:16 PM

#

oh I THINK I get that

gentle spoke Aug 2, 2020, 4:16 PM

#

do you have a terminal?

#

what's your main os?

#

(headless means no display, btw)

sharp yarrow Aug 2, 2020, 5:16 PM

#

os.system("hostname")

f= open("name" , "w+")
f.write(????)
f.close()

question: how do i save command output into "name" file?
/
|
os.system("hostname")

#

someone can help

shut osprey Aug 2, 2020, 5:24 PM

#

Don't use os.system

sharp yarrow Aug 2, 2020, 5:24 PM

#

why