algos-and-data-structs

[foggy mirage]

Help please

I'm breaking down

    def put(self, num):
        if self.base:
            if num < self.base:
                if self.left is None:
                    self.left = Node(num)
                else:
                    self.left.put(num)

            elif num > self.base:
                if self.right is None:
                    self.right = Node(num)
                else:
                    self.right.put(num)

        else:
            self.base = num

what is put purpose?

It does nothing

[knotty magnet]

if it does nothing, then why is there so much code

[foggy mirage]

That's why I'm asking

[knotty magnet]

why do you think it does nothing?

[foggy mirage]

like when I remove it , it only makes 2 pair of children and stops

[foggy mirage]

Because it is being called from inside itself

like

def put():
    put()

which makes no sense to me

at all

[knotty magnet]

can you write a recursive function to calculate the factorial of a number?

without looking anything up, except maybe what factorial is

[foggy mirage]

um

no

[knotty magnet]

try it

[foggy mirage]

maybe like

def fac(num)
    num * fac(num)
 

[knotty magnet]

you have to return it also

it will also keep recursing forever, so you need to add a base case

essentially, what you want to do when you stop recursing, and when

[grizzled whale]

def f(n):
  if n == 0: return 1
  return n*f(n-1)

[elfin pecan]

There are always some problems which require some mathematical approach or some different approach
like for max subarray sum we use Kadane's Algorithm
Some of the new approaches are mostly learnt by practice but is there any resources(like repositroy,video collection) where such ideas are available at once like the sliding window technique and the non mainstream topics but might help in better approach towards problem?

[last fulcrum]

There was a question from amazon on rotating an image or 2D array 90 degrees.
Per my understanding, column 1 should become row 1. How can I easily translate this to code?

`
a = [[1, 2, 3],
[4, 5, 6],
[7, 8, 9]]
the output should be

rotateImage(a) =
[[7, 4, 1],
[8, 5, 2],
[9, 6, 3]]
`

[runic tinsel]

transpose after reversing

[last fulcrum]

Words I don't understand 😂

[runic tinsel]

[*map(list, zip(*a[::-1]))]

[last fulcrum]

Yea I'm still confused. Maybe this question is too much for me?

[runic tinsel]

just look at where indices come from

the result is ```
2,0 1,0 0,0
2,1 1,1 0,1
2,2 1,2 0,2

to make a nested loop that sets cells in the new array to these values from the original, find a formula that makes 2,1,0 from 0,1,2

[last fulcrum]

Ok thanks

[fiery cosmos]

pinga

[distant mauve]

Hi to all Pythoneers.

[glacial sail]

#help-corn

[fiery cosmos]

I have a sorted array, and want to insert one more item into the array such the resulting array remains sorted.What is an Efficient way to insert a number into a sorted array of numbers?

[shut breach]

iterate over the array until you find the spot, then insert

[humble mauve]

Similar logic to insertion sort. find the right place, insert new element, then right shift all elements to the right, should be linear since the array is already sorted

[fiery cosmos]

that would be O(n), I want the fastest and the most effective way, binary search does it in O(logn) but the code is quite complicated and will take time to write it, I have written it before and it takes me like around 15 mins so i wanted something that is quicker but equally faster

[fiery cosmos]

@humble mauve

[fervent saddle]

It will end up being O(n) no matter what because you have to shift all the further values over by 1

There’s no way to insert into an array faster than O(n)

Doing some kind of binary search to find the right position might make it faster on average, but there’s no way to avoid O(n)

[honest sleet]

Can anybody provide best resources to learn DS Algo?

[rustic glacier]

Well for what if just in general I recommend trying out codehs but like java and python witch ever your preference is java is going to be a lot harder to understand and master while python is easier because python is easier language if you don't like codehs just look up best websites to learn DSA

[keen hearth]

Check out the pinned messages in this channel.

[honest sleet]

How to do that?

[honest sleet]

I'm looking for some free resources.

[keen hearth]

There's a button up at the top right of the Discord client shaped like a push-pin.

[honest sleet]

Got it thanks.

Brilliant sites, thanks to whoever made or shared them.

[stiff mango]

I just made my own function that converts strings to ints without using the int function or anything similar

!e ```py
def str_to_int(string):
n = 0
for i, place in enumerate(string[::-1]):
n += "0123456789".index(place)* 10**i
return n

print((str_to_int("25")))

[halcyon plank]

@stiff mango :white_check_mark: Your eval job has completed with return code 0.

25

[stiff mango]

!e ```py
def str_to_int(string):
return sum(("0123456789".index(place)* 10**i for i,place in enumerate(string[::-1])))

print((str_to_int("25")))

[halcyon plank]

@stiff mango :white_check_mark: Your eval job has completed with return code 0.

25

[stiff mango]

Is this how it works (in theory) under the hood? I know python uses C for stuff like that, but is this how it is implemented there?

[last fulcrum]

#internals-and-peps

[stiff mango]

K

[knotty magnet]

you're using str.index to find the place when you could just use simple math, so there's no way that's how it's implemented

[stiff mango]

!e print(ord("0"))

[halcyon plank]

@stiff mango :white_check_mark: Your eval job has completed with return code 0.

48

[stiff mango]

!e ```py
def str_to_int(string):
n = 0
for i, place in enumerate(string[::-1]):
n += (ord(place)-48)* 10**i
return n

print((str_to_int("25")))

[halcyon plank]

@stiff mango :white_check_mark: Your eval job has completed with return code 0.

25

[stiff mango]

@knotty magnet is that what you mean?

[knotty magnet]

no

[stiff mango]

And what "simple math" do you mean?

[knotty magnet]

division?

[stiff mango]

And where would division come into play? What would you devide to get from the string representation to the number?

[knotty magnet]

you wouldn't use the "0123456789" string in the first place

[stiff mango]

I didn't in this implementation

[knotty magnet]

yeah, that looks ok

[brave oak]

!e

def str_to_int(string):
    n = 0
    for i, place in enumerate(string[::-1]):
        n += (ord(place)-48)* 10**i
    return n

print(str_to_int("-1"))

[halcyon plank]

@brave oak :white_check_mark: Your eval job has completed with return code 0.

-29

[brave oak]

@stiff mango

[cyan verge]

what are simple uses cases for queues and stacks

[brave oak]

task queue

[cyan verge]

I see. just looked it up. what about stacks?

or do they both apply?

[brave oak]

certain data structures can be traversed with stacks

or you can parse with a stack

example

Python's indentation levels

notice how

the last level of indentation is the first to be dedented

[cyan verge]

I haven't even thought about indentation like that. that's an interesting application

[fiery atlas]

Is there any thing I can look at if I wanted to output a csv file for a log file? I search in google but too complicated for me

[last fulcrum]

Python has hacky one liners for DS & A. Are these generally accepted in interviews?

For example rotating an array by 90 degrees. Someone did this.

rotateImage = lambda a: zip(*a[::-1])

Another did this.

def rotateImage(a): return list(zip(*reversed(a)))

[jolly mortar]

assigning a lambda to a variable in general is considered bad practice

[last fulcrum]

I think all these would be frowned upon. I'm thinking the solution should work in every language.

How can I improve this solution I wrote up?

class Solution:
    def twoSum(self, nums: List[int], target: int) -> List[int]:
        """ Returns the index of the two values that gives you the target"""
        
        count = 0
        for value_1 in nums:
            count += 1
            for value_2 in nums:
                if value_1 + value_2 == target:
                   return list((nums.index(value_1), nums.index(value_2)))

This is On2. Very bad lol

[austere sparrow]

Hint: use a set.

[last fulcrum]

Thanks

[fiery cosmos]

Better yet a dict to keep track of the index

[runic spade]

Can someone help me setup matplotlib on visual studio code?

[last fulcrum]

#❓｜how-to-get-help

[runic spade]

Im good now. ty

[compact perch]

Hii I'm nee here could anyone give me best learning material for DATA STRUCTURE AND ALGO

[astral pulsar]

Given 5 audio files, if i want to split them into 3 chunks, 2 2 1 would be a naive split because the 5 audio lengths could be very different. I need to split it into 3 1 1 if the former 3 audio files have about the same combined length as the 4th or 5th. In other words, I need to create chunks that optimize items' combined length instead of item count. What would be the name of the algorithm I should look up if I want an implementation for this?

[fiery cosmos]

https://leetcode.com/problems/minimum-limit-of-balls-in-a-bag/ I tried heap but I failed, it is supposed to involve binary search but I dont get how, anyone?

[proper canyon]

Hi guys, I've got a question, how do I measure the algorythmic complexity of this recursive factorial function?

I mean, I have just one recursive call so it would be 1^n? That does not make sense to me

[fiery cosmos]

For a given n how many total function calls does it make? Say for n = 4?

[woven oar]

Can anyone point me in the direction of some material on search algorithms? In particular I'm trying to figure out how to search a massive CSV file for relevant search terms input by a user. For example, I have a CSV file with thousands of rows of food items from the USDA. Currently I'm using the "if x in y" to perform my search. An example of a current issue is if the person types in blueberry, the CSV file has items listed as "blueberry" and "blueberries" and only returns the one typed. I want both returned, and I'm not sure if this is just an exercise in creating a function to search for both singular and plural items for each item entered, or if there is a better method.

[proper canyon]

Ohhhh I get it

So it'd be n-1

[fiery cosmos]

Yeah, or O(n) time complexity

[proper canyon]

Thank you 😄

[meager slate]

you can use re to match both the singular and plural

but regex could be slow if the file is massive

[loud trail]

• looping in python is slow. solution: use pypy, graalpython, or cython if you just need to loop over lines of a text file quickly, or use pandas to read the csv into a data frame, which does the looping "internally" and therefore can be a lot faster.

• data structures. i think full-text search algorithms often index the text using a trie data structure, to efficiently provide "search-as-you-type" functionality. see https://stackoverflow.com/a/50206518/2954547

• you might need to apply "stemming" or "lemmatizing" to get a "base" word to use for searching. examples include https://www.nltk.org/api/nltk.stem.html (the multi-language SnowballStemmer is often a good default, or WordNetLemmatizer for english) and spacy https://shyambhu20.blogspot.com/2020/11/how-to-do-lemmatization-using-spacy.html

this is also a thing but its quite old, so maybe borrow its ideas but maybe dont rely on it https://pypi.org/project/Whoosh/

you could also load the data into sqlite and use the FTS5 extension

https://www.sqlite.org/fts5.html

oh finally you might want to try several different inflections of a word, see https://pypi.org/project/inflect/

[keen hearth]

I would also recommend loading the data into an SQL database, and storing word-stems for lookup.

In SQL you can 'index' a column of the table, allowing fast lookup of rows by that attribute. You will also be able to access the same data quickly from multiple instances of your application without having to load it into memory multiple times.

[woven oar]

Thank you all for the great suggestions, there's a lot to digest here, but this looks like a good starting point. Thanks again.

[woven oar]

FYI, this added two lines of code and is working flawlessly so far. Why didn't I think of looking for this sooner?!

[loud trail]

next time, you will!

[runic niche]

hi i have a pretty simple question in #help-chocolate if anyone has free time

[last fulcrum]

Does leetcode usually have bad test cases?

My code failed even though it should pass the test case

[fiery cosmos]

I don't think I've ever had problems

[fiery cosmos]

You might be printing instead of returning or something

[last fulcrum]

I'll post the code. Look at it for me

Q

You are given an array prices where prices[i] is the price of a given stock on the ith day.

You want to maximize your profit by choosing a single day to buy one stock and choosing a different day in the future to sell that stock.

Return the maximum profit you can achieve from this transaction. If you cannot achieve any profit, return 0.

Example

Input: prices = [7,1,5,3,6,4]
Output: 5
Explanation: Buy on day 2 (price = 1) and sell on day 5 (price = 6), profit = 6-1 = 5.
Note that buying on day 2 and selling on day 1 is not allowed because you must buy before you sell.

My code

class Solution:
    def maxProfit(self, prices: List[int]) -> int:
        buy = sys.maxsize
        sell = -(sys.maxsize)

        for day in prices:
            if day < buy:
                buy, sell = day, day
            if day > sell:
                sell = day

        if sell > buy:
            return sell - buy
        else:
            return 0

Wrong test case.
[2,4,1]
This test case says it should return 2.
It should rather return 0 because the person would buy on day 2. Day 2 has the value of 1. After 1 the array ends. There is no profit to be made here, thus the answer should be 0. Please redo the test case. Thanks.

[fiery cosmos]

Can't you buy on 2 and sell on 4, thus earning 4-2=2

Iirc this kind of problem requires dynamic programming

But i could be totally wrong there

[last fulcrum]

👍

[simple mist]

hey guys! any recommendation where can i start learn data struct and algo.... also some prerequisites too. Thanks!

[vivid matrix]

https://www.amazon.com/Algorithm-Design-Jon-Kleinberg/dp/0321295358 is the best I have found about algorithms. Of course you can go back to masters i. E. Introduction to algorithms https://en.m.wikipedia.org/wiki/Introduction_to_Algorithms

[humble beacon]

tuple_list = [('234', 3, 2, 1, 3, 5), ('123', 3, 2, 1, 3, 5), ('143', 3, 2, 1, 3, 5), ('2343', 3, 2, 1, 3, 5)]

I want to find the tuple where index 0 == 123... how can i do this the most efficiënt way.. note that i have a tuple list of 12k items.

Is there an optimal way? It has to be fast.

Is there an algorithm for this?

[slim vault]

I don't think you can do better than linear search here

the question is whether you'd want to stop after finding the first one, or find all where the number matches

if you're going to be searching for records often, it may be worth it to build a more appropriate data structure, such as a dictionary mapping from each tuple's index 0 to the whole record

[fiery cosmos]

!e this is a way to construct and lookup with a dict everyone keeps mentioning```py
tuple_list = [('234', 3, 2, 1, 3, 5), ('123', 3, 2, 1, 3, 5), ('143', 3, 2, 1, 3, 5), ('2343', 3, 2, 1, 3, 5)]
d = {t[0]: t[1:] for t in tuple_list}
lookup = '123'

print(d[lookup]) #<- super fast O(1)```

[halcyon plank]

@fiery cosmos :white_check_mark: Your eval job has completed with return code 0.

(3, 2, 1, 3, 5)

[slim vault]

yeah, but keep in mind that this is only worth it for repeated lookups

for a one-off lookup, a linear search should be faster even if you're looking for the very last element (i.e. worst case)

[fiery cosmos]

I think they mentioned in general they need to do it a bunch

[slim vault]

okay, then a dict sounds good 😌

[frail sentinel]

what does the deferent between interators and function generators and when does we use it ??

I've confused in this point and I cant understand them. Pleas help me

[loud trail]

an iterator is more general than a generator function

it can be very enlightening to read https://docs.python.org/3/library/collections.abc

but the short version: an iterator is anything that can be iterated over (kind of)

an iterable is something that can be turned into an iterator

a generator is a specific kind of iterator

[frosty coral]

What do you recommend me to read to improve my logic?

[simple mist]

same here.

[simple mist]

ah yes.. one first book in algorithm. Tbh, my friend use to recommend me this but thanks for the effort man.... appreciated

[dapper sapphire]

can we do string concatenation by using list comprehension for following:

        for key in (dictionary):
            output_string = output_string + dictionary[key]

List comprehension is not faster than a for loop? It just makes it more readable in some cases?

[fiery cosmos]

''.join(dictionary.keys()) would be fastest I expect, and shorter

But yeah, in general, loop vs comprehension speed is similar if they do the same work

[dapper sapphire]

Yeah, I figured I could use an array and then do "".join(array)

[drifting drum]

You're given two arrays of numbers, say array A and array B with the numbers being in the range -10^6 to 10^6
Furthermore, the length of the two arrays is <= 1000. They both have equal lengths, too.

Now you're supposed to pick some indices between 0 and len(A), as many as you like. (Need not pick consecutive indices).
Suppose you pick k of them, a1, a2, ...ak.

Now you find Value V = (A[a1] + A[a2] .... + A[ak])^2 + (B[a1] + B[a2] + .... B[ak])^2.
You're supposed to find the maximum possible value V.

So my thoughts are you could probably do a greedy
where if A[i] and B[i] are both positive, you certainly must pick them
if they're both negative, don't pick them, it's not going to increase our value
In the event of one of them being positive and one of them being negative, get the value before and after adding those numbers and see which's better
So that's the greedy approach that comes to my mind
Is there perhaps a better way of doing this?

[topaz pulsar]

why not pick negative numbers? something like (-10 + -7 + -8) ^ 2 is positive after all

[drifting drum]

Think of it this way

you've picked some indices, the sum(without being squared) of array A is S1, and sum(without being squared) of array B is S2

now suppose you pick index x and A[x] is negative, B[x] is also negative

so your previous value V was = S1^2 + S2^2

Now, A[x] is negative, so S1 +A[x] < S1.
Also B[x] is negative, so S2 + B[x] < S2.

And so (S1 + A[x])^2 < S1^2 and (S2 + B[x])^2 < S2^2

So (S1 + A[x])^2 + (S2 + B[x])^2 < S1^2 + S2^2

And actually

if |A[x]| is larger than S1 itself

you might as well not pick the elements you have picked in S1

But yeah

it boils down to

either pick all positives

either pick all negatives

or you pick a combination of both which is the tricky part

[drifting drum]

So,
What I was saying above was that when you're picking all positives, don't pick indices x such that both arrays have negatives at that index(Because you'll pick them when you're picking all negatives in another iteration perhaps)

[topaz pulsar]

this is what i was getting at, just because S1 +A[x] < S1, that doesn't always mean (S1 +A[x])^2 < (S1)^2, I don't have a full solution but was just trying to make sure you are aware that you can't just disregard indicies that would give you negatives for both numbers

[drifting drum]

yeah you don't disregard them entirely, but you do disregard them when only picking positives

Perhaps I've just expressed everything in a very messy way

[topaz pulsar]

yeah, don't worry I get what you mean, I may of just misunderstton your original comment

[drifting drum]

nah going back and reading what I wrote, I can see what I said was not implying in any way about the separate cycles of picking only numbers of one sign, so that's on me 😅

[last fulcrum]

Should I implement a queue in Python using collections.deque or a circular array?

[austere sparrow]

Should I implement a queue in Python
probably not 😄

well, unless you want to learn how to implement a queue

The advantage of deque is that it can grow and shrink as much as you want, and you won't need to resize the storage

[last fulcrum]

Nice. I was of the opinion you'll be asked to implement a queue or stack yourself.

[austere sparrow]

You can look into CPython's source code if you want to see how deque is implemented.

It's something like a linked list of small (64) arrays.

[last fulcrum]

I saw a linked list implementation of a queue and stack on the web. People are really doing everything themselves😅 .

[wary spear]

Hi I am currently trying to convert a list of objects to a dictionary

I have got an object with two properties object.key and object.value and I want to map that list to a dict like

from
list = [object[0], object[1], etc]
to
dict = {object[0].key : object[0].value, object[1].key : object[1].value, etc}

Hope this is understandable if not please try to explain what was unclear. Would be great if someone could help me

[tame edge]

This the weakness of me .

This channel

[loud trail]

use a for loop?

data_lst = [thing1, thing2, thing3]
data_dct = {thing.key: thing.value for thing in data_lst}

[wary spear]

dicti = dict((x.key, x.value) for x inlist)

already found something (:

[loud trail]

i'd recommend the dict comprehension instead, but that does work too

[wary spear]

What is the difference?

[loud trail]

clearer intent, imo

technically should be a bit faster but that's not really relevant

[lean musk]

Game Crash is problem ?

[hidden rapids]

Where do I get started to learn data structures and algorithms?

I tried leetcode but had real trouble in the very first problem and had to refer to the solution

Any help is much appreciated

[runic tinsel]

that's learning though

you have a problem, you see the solution you won't have this problem later

[hidden rapids]

Yes but it doesn't feel as efficient

Because I spend like 3-4 hours on a problem just stuck

[runic tinsel]

i think that's pretty much optimal

[hidden rapids]

I see

That makes sense then

If that's how you learn I'm down for that

[runic tinsel]

i guess it would be optimal if you know you're trying to do something within your capabilities, if you don;t understand the solution, that's wasted time

some sort of intuition whether something is way too hard to attempt is necessary

[hidden rapids]

@runic tinsel yeah that's what I felt. If it's something I can comprehend and work towards little by little, even if it's 50 percent of the solution, then I'm satisfied

[dapper sapphire]

I'm working on a project that utilizes async routines where index and length is used to create the progress bar. And I see the progress bar constantly fluctuate.

This reminded me of loading screens in games where the loading percentage would fluctuate.
For example: loader would be at 42% and then it would dip down to 20% and then it would jump to 74% and so on. So the async routines could be the reason why the progress percentage would fluctuate?

[coral horizon]

what does this thing I'm looking at mean?

can someone tell me about going on about understanding it

I know the log in that.

[dapper sapphire]

If we have foo, bar, and foobar. How many combinations are there in pairs (of 2 and 3), are there 4 combinations?

1. foo & bar
2. foo & foobar
3. bar & foobar
4. foo & bar & foobar

[fiery cosmos]

yep, if order doesn't matter

[vocal gorge]

C(3,2)+C(3,3) = 3+1 = 4, yes

[dapper sapphire]

What is this formula and how does it work?

[vocal gorge]

C(n,k) being the binomial coefficient (number of ways to take k objects from n). You are interested in 2- and 3-element combinations of 3 elements, so C(3,2) + C(3,3).

C(n,k) = n!/(k!*(n-k)!) in general

[fiery cosmos]

Often called "n choose k" sometimes like nCk

[dapper sapphire]

Interesting how so many things can boil down to a math equation.

[fiery cosmos]

https://leetcode.com/problems/valid-triangle-number/ the only thing I know for now in this question is that we should sort and apply two pointers but I dont understand how to get all combinations using it, anyone ?

[dapper sapphire]

Why are some python methods inconsistent.

We can do llist.reverse(), but we cant do string.reverse(), we have to do "".join(reversed(string))

I guess other programming languages are like this too.

Actually never mind I see what's happening

[dapper sapphire]

How can we create a copy of a list? We cant just simply do the following lol:

list1 = list2

because changing one changes the other.

I have also tried:

list1 = list2.copy()

Which works, but if it's done inside a method then modifying list1 will also modify list2.

[fiery cosmos]

How do you mean

there's also slicing which looks confusing list1 = list2[:]

and copy module .deepcopy if you need to recursively copy the items

[dapper sapphire]

Run the following:

llist = [[1,1,0],[1,0,1],[0,0,0]]
llist2 = llist.copy()

llist.pop()
# llist2.pop()

print(llist)
print(llist2)

And you get

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

---

Now if you do:

from typing import List

class Solution:
    def __flip(self, number: int) -> int:
        if number == 0:
            return 1
        return 0

    def flipAndInvertImage(self, A: List[List[int]]) -> List[List[int]]:
        llist = A.copy()
        for inner_list in llist:
            inner_list.reverse()
            for index in range(len(inner_list)):
                inner_list[index] = self.__flip(inner_list[index])
        return llist

image = [[1,1,0],[1,0,1],[0,0,0]]
image2 = Solution().flipAndInvertImage(image)
print(image2)
print(image)

You get:

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

[lean dome]

you're looking for copy.deepcopy(). list.copy() makes a shallow copy, meaning it doesn't copy things inside the list

[dapper sapphire]

I dont get intellisnse for copy.deepcopy()

[lean dome]

have you done import copy ?

[dapper sapphire]

Oh I cant use it then.

[lean dome]

!e ```py
import copy
llist = [[1,1,0],[1,0,1],[0,0,0]]
llist2 = llist.copy()
llist3 = copy.deepcopy(llist)

print(f"{llist=}")
print(f"{llist2=}")
print(f"{llist3=}")

print(f"{llist[0] is llist2[0]=}")
print(f"{llist[0] is llist3[0]=}")

[halcyon plank]

@lean dome :white_check_mark: Your eval job has completed with return code 0.

001 | llist=[[1, 1, 0], [1, 0, 1], [0, 0, 0]]
002 | llist2=[[1, 1, 0], [1, 0, 1], [0, 0, 0]]
003 | llist3=[[1, 1, 0], [1, 0, 1], [0, 0, 0]]
004 | llist[0] is llist2[0]=True
005 | llist[0] is llist3[0]=False

[lean dome]

both llist2 and llist3 seem like they're identical to llist, but the elements of llist2 are the same lists as are inside llist, while the elements of llist3 are copies of the lists inside llist

[dapper sapphire]

Actually that does work

So then for https://leetcode.com/problems/flipping-an-image/

from typing import List

class Solution:
    def __flip(self, number: int) -> int:
        if number == 0:
            return 1
        return 0

    def flipAndInvertImage(self, A: List[List[int]]) -> List[List[int]]:
        for inner_list in A:
            inner_list.reverse()
            for index in range(len(inner_list)):
                inner_list[index] = self.__flip(inner_list[index])
        return A

image = [[1,1,0],[1,0,1],[0,0,0]]
image2 = Solution().flipAndInvertImage(image)
print(image2)
print(image)

Would this be reasonable? I just dont like modifying in-place unless the question asks for it.

lmao llist = A.copy() doesnt do anything because I'm modifying the same list.

Should I manually create a copy of the list first?

[last fulcrum]

???????. They clearly said use a deep copy

Any copy you create would be shallow. They are referring to the same address in memory.

[dapper sapphire]

Yeah, but I dont think I can import a library in an interview. Otherwise deepcopy is a pretty good solution to create a copy.

[last fulcrum]

You can. It's a built in. How else would we be able to use defaultdict and a deque?

[fiery cosmos]

You may be over-complicating that problem a bit 😅 you can do it in one line with a nested list comprehension

How do you do a code spoiler

||return [[1 - x for x in row[::-1]] for row in image]||

[dapper sapphire]

Oh we can import libraries, thanks for mentioning that. Yeah, I'll import copy so I can use deepycopy

[last fulcrum]

There is a distinction. This is an inbuilt library. I don't know if you can import external libraries.

[dapper sapphire]

What would be external library? Something that you have to pip install?

[last fulcrum]

Yes. from typing import List is an example of a an inbuilt one.

[dapper sapphire]

So if it's in cpython, it's built in and we can use it:?
https://github.com/python/cpython/tree/main/Lib

[last fulcrum]

Yep. Read the documentation to see more

[last fulcrum]

Where is this code from?

[fiery cosmos]

I just wrote it

[last fulcrum]

I was asking about the original you refactored.

[dapper sapphire]

Read the documentation about what?

[last fulcrum]

To see what other builtin modules there are

[fiery cosmos]

I didn't refactor. I did the leetcode Tony posted https://leetcode.com/problems/flipping-an-image/

[last fulcrum]

Oh. That was the question I asked the last time

Is it right to use one line like that in an interview?

I don't think so

[dapper sapphire]

It depends

If one liner is too complicated, then no.

[last fulcrum]

I'll list my reasons why I think it should not be so. But as always, this is a subjective issue.

[dapper sapphire]

I personally dont like list comprehensions.

[last fulcrum]

1. It shows you know how to code, but then you're using what we call hacks and tricks. Most of these things do not work in other languages, and companies are generally looking for solutions that work in other languages, since it's highly likely you might work with other languages.

2. You would solve the problem early, thereby allowing the interviewer to give you even harder problems.

[fiery cosmos]

I really don't think showing you know a language or doing questions fast are bad in an interview 😅

You can always explain and show that there are other ways to write equivalent code

[last fulcrum]

Not necessarily bad, you are just inviting more pressure and hard questions unto yourself. See it as your aim is to do anything to maximise your chances of getting job, not showing that you know how to do this really fast. It's the same thing as the interviewer giving you a question you learnt just a few days before the interview. Of course you can breeze through, but would that really help your chances?

[last fulcrum]

Maybe showing the one liner after using the long way around is better.

[fervent saddle]

I’ve heard that suggestion too lol, that if you already know the answer then you should act like you’re coming up with it on the spot so you’ll look cooler

[dapper sapphire]

I have heard you should give the less optimized solution first, so when they ask you to optimize it, you can show them the optimized way. Is that also true??

A friend's friend had an interview at Google and he did the optimized way, then they asked him to optimize it and he couldn't because it was already optimized.

[last fulcrum]

😂

[dapper sapphire]

No honestly!! I was like wait what you can't be serious

[last fulcrum]

I would look at the time honestly. More time, do unoptimized first. Less time, do optimized first. Anything to help you pass.

Except cheating and lying

[last fulcrum]

Yea Google likes to fail people

They want the best

[dapper sapphire]

How can one cheat and lie on a whiteboard?

[dapper sapphire]

I have to say I like this approach actually

[fervent saddle]

Maybe they meant optimizations not as in time/space complexity

But as in for the computer

idk

[lean junco]

dont wanna type again please see this

why my kadanes algo able to overcome its limitation ......lol

it shouldnt work but it does

??

[lean junco]

i figured it myself

dont bother now

[last fulcrum]

Anyone here with a leetcode premium subscription? If this is wrong to ask here please remove.

[vapid cedar]

Hey I am badly inneed of suggestion in detecting pattern from historical data

can i have a chance to speak with anyone

is there any api or module for detecting pattern , i have looked up talib

except that , i just need to detect upwards and downwards and straight line pattern

🙂

[granite cliff]

hello guys I wanna learn and master data structure and algorithms, i have already queried youtube and I have found something but I wanna find something that is efficient since its a critical subject to learn.

[last fulcrum]

Check the pins

[granite cliff]

thank you that is exactly what i wanted

[frail fjord]

Hello. I have a question that has been boggling me for the past few days. I'm trying to figure out how to remove background noise for a dictation project that I have been working on. What are some ways to integrate background noise reducing libraries into my project. Also, is it possible to use Nvidia's Broadcast for integration by any chance?

well this is an algo chat and my question relates to algorithm libraries for noise reduction

[dark bear]

I keep getting the index out of range error when using pointers to iterate a list or array instead of for loops. Any tips or hints how I can remember to not get this error?

[fiery cosmos]

I doubt that Nvidia has an API for broadcast. Noise reduction is a very large field and there's already many various filters you can apply to your data. Everything from simple low-pass filters to highly complex filters that I don't understand.
Look into scipy they already have a few filters implemented.

[frail fjord]

thanks, ill look into it

[junior wind]

Can anyone help me get started with DSA

I am new to it so idk where to start from. if anyone can give some good recommendations about courses or youtube channels it will be awesome

[ivory yacht]

There's a few pinned here.

[junior wind]

thanks , I just checked them out

btw is there any good youtube channel if you know, I don't wanna get to wrong one cuz there are infinite number of videos on that

[wild river]

i have a doubt with functions

can anyone help me?

#help-cheese message

[leaden magnet]

hello

[wild river]

hi

[leaden magnet]

i have a problem can you help me?

[onyx root]

ask!

[leaden magnet]

from sys import path

path.append('..\\modules')

import module

zeroes = [0 for i in range(5)]
ones = [1 for i in range(5)]
print(module.suml(zeroes))
print(module.prodl(ones))

Traceback (most recent call last):
  File "c:\destop\python\Pcap course\module 5\main.py", line 5, in <module>
    import module
ModuleNotFoundError: No module named 'module'

[knotty magnet]

there doesn't look to be a module with that name

[leaden magnet]

yes but if i put the absolute path works

my problem is i don't understand how works relative and absolute paths

i think i am missing something

[keen hearth]

Hello 🙂 I believe a relative path in sys.path would be relative to the current working directory, which is where you run the code from (not where the file is located). In this case, if your current working directory is "module 5", then I think ..\modules would refer not to the "modules" folder contained within "module 5", but a "modules" folder contained within "Pcap course". Long story short, you want to use the path .\modules instead.

I'm not 100% sure that's the issue though. Try it out and let me know 😄

[leaden magnet]

unfortunately don't work 😦

[snow bridge]

instead of 'import module' try 'import modules'

[keen hearth]

I see the current working directory is your user folder?

You would need to change to the directory containing the file. Or, make the path relative to the file.

[leaden magnet]

oh

[keen hearth]

I think this should work: ```py
from pathlib import Path
p = Path(file)
sys.path.append(p.parent / 'modules')

[leaden magnet]

now it works i open the entire folder and what you say fisrt time worked

thank you

[keen hearth]

Oh right.

[keen hearth]

Did you also try this? 😄

[keen hearth]

It might save you the hassle of having to navigate to the folder each time.

(Assuming it works 😅)

[leaden magnet]

umm

[keen hearth]

Ah, I made a mistake: ```py
from pathlib import Path
p = Path(file)
sys.path.append(str(p.parent / 'modules'))

Items in sys.path currently have to be strings.

There's a pull-request pending in python to allow Path objects in sys.path.

[leaden magnet]

yes is works

thank you so much

i am trying to figure this thing out from a week

XD

[keen hearth]

Ah, sorry to hear that. That's so frustrating!

[leaden magnet]

yes but perseverence is the key

now i know a little more then yesterday and tomorrow i hope i will know little more than today

thank you again

[lyric burrow]

i was solving this problem on leetcode
i wonder how they made it faster

[fiery cosmos]

Some trick with reduce might be possible But your solution looks fine

LeetCode times vary. You hit submit once and you are faster than 99% and click it again and it changes to 20%

[knotty magnet]

itertools.accumulate

[dim snow]

do you have a book you recomend to learn algorithms?

[keen hearth]

Maybe return list(itertools.accumulate(nums))?

[analog ferry]

Anyone aware of a fun to use geometric constraint resolver for python

I want to plan woodworking details and freecad/cad query seem horrendous

[lyric burrow]

how can i optimize this?

[fiery cosmos]

hint: sort it

[knotty magnet]

ya gotta put spoilers on that yo

[runic tinsel]

sorting doesn't help much

[knotty magnet]

why not?

brings it down from n^2 to n log n

[runic tinsel]

you need the answer in the original order

[3,4,2,3]
[2,3,3,4]
like, i sorted them and i saved the original list

[idle marsh]

Which website is that?

[runic tinsel]

now i need that [0,1,1,3] in the order of [1,3,0,1]

https://leetcode.com/problems/how-many-numbers-are-smaller-than-the-current-number/submissions/

i got "faster than 90.39%"

"faster than 99.16%"

sorting is the first step after all

99.83 lol

[knotty magnet]

are you just running the same code over and over until you get a lucky run

[runic tinsel]

yes

it varies a lot

i thought i cheesed it with returning a genexp, but it was just a lucky run

[knotty magnet]

what if you do gc.disable()

that would probably give you a better chance of getting a good run

[runic tinsel]

i don;t really have stuff that would be colledcted

[fiery cosmos]

That way with the Counter is neat. Should be easier (or at least the same) to sort that compared to the original array

my trick was to hang onto the indices by sorting a list of tuples (value, index) ```py
class Solution(object):
def smallerNumbersThanCurrent(self, nums):
ordered = sorted((n, i) for i, n in enumerate(nums))

    smaller_counts = [None] * len(nums)
    last_n, last_i = ordered[0]
    smaller_counts[last_i] = 0
    for index, (n, i) in enumerate(ordered):
        if n == last_n:
            smaller_counts[i] = smaller_counts[last_i]
        else:
            smaller_counts[i] = index
        last_n, last_i = n, i
        
    return smaller_counts```

But it feels bulky

[knotty magnet]

should probably spoiler that

eh, he already has a sol nvm

[fiery cosmos]

how do you spoiler code blocks?

[knotty magnet]

||```py
e

[fiery cosmos]

I was trying the other day and /spoiler nor || seemed to work

[knotty magnet]

take a screenshot like frowny did i guess, lol

[fiery cosmos]

well meh

[knotty magnet]

carbon.now.sh

[dapper sapphire]

How important would it be to know list comprehension??

I dont think someone would fail you just because you didnt use list comprehension or dont know it.

But wondering what other people think.

[knotty magnet]

it's useful, but it's just a slightly more expressive way of saying the same thing you can already say

if someone asks you specifically about it, i guess you would need to

[dapper sapphire]

Slightly more expensive way? Do you mean list comprehension is slightly more expensive than list comprehension?

Wondering what people think of this if it's clean and expressive enough leetcode 1816:
https://leetcode.com/problems/truncate-sentence/

class Solution:
    def truncateSentence(self, s: str, k: int) -> str:
        string_list = s.split()
        truncated_string_list = string_list[:k]
        truncated_string = " ".join(truncated_string_list)
        return truncated_string

[runic tinsel]

this can't be done with a list comprehension though

[dapper sapphire]

No this question is apart from list comprehension

I wanted to know what people though it if its clear enough.

[knotty magnet]

i said expressive. list comp is slightly more expressive (when it's simple enough)

[dapper sapphire]

Oh my bad sorry lol. Yeah I think so too, it's sort of like .join

[fiery cosmos]

Splitting it over 4 lines seems excessive. One or two like return ' '.join(s.split()[:k]) feel less daunting to read imo

or use applicable names like words instead of string_list

[dapper sapphire]

actually plus 1 for words. I wholeheartedly agree

[dapper sapphire]

Actually can we compare list comprehension with join. Where list comprehension is done for a list, but join is done for a string.

[knotty magnet]

depends what comparison you're trying to make i guess

[lyric burrow]

it's leetcode

[severe osprey]

https://practice.geeksforgeeks.org/problems/maximum-index-1587115620/1/?track=md-arrays&batchId=

Help me understand the concept of the solution

[lyric bramble]

    out =0
    iter_= 0
    start,end,y_match = False,False,False
    i = 0
    while i <len(s):
        if (s[i] == "x") and (end is False):
            start = True
        if (s[i] == "y") and (start == True):
            y_match = True
        if (s[i] == "x") and (start==True) and (y_match==True):

            out +=1+iter_
            iter_ +=1
            y_match = False
        i+=1
    return out```

is this right?

can someone help me with this?

[jovial sphinx]

Problem:
Given a coordinate on a 199x199 Cartesian plane which marks the starting position of a cube, find the shortest route to the centre (0, 0) which would result in the cube having the same side facing up as it did at its starting position.
Anyone has any idea how to approach & solve this (kind of) problem (which relates to Mahattan distance)??

[keen hearth]

Interesting problem! Well without the constraint (that the same side must be facing up) the length of the shortest path would be the manhattan distance. But it seems you need to be a little more clever than that.

A brute-force approach would be to do a search (e.g. breadth-first search) on the state-space of the problem. That is, a state comprises a coordinate (the location of the cube), and a number representing the orientation of the cube (the side initially facing up could point in six possible directions).

An A* search would probably work reasonably well here, using the manhattan distance as a heuristic (as it always gives an optimistic estimate of the remaining distance).

Constraint relaxation (removing constraints from the problem) is often a good approach to coming up with search heuristics.

[jovial sphinx]

🤔

In the actual task, brute force for any coordinates giving a Manhattan distance of 5 or larger is not allowed. Back-propagation is also not allowed.

I'm pretty sure search algorithms such as A* involve back-propagation(correct me if I'm wrong), so what would be a more "mathematical" approach to this problem?

[jovial sphinx]

Also, if anyone is kind and willing enough to code me a solution to this problem via any method I would really really appreciate that >w<

[small flax]

Should everyone know FizzBuzz?
I personally think the pythonic 1 liner at the end is too cheeky.

https://youtu.be/02RjoAnW2kg

[knotty magnet]

you don't have to memorize it, but if you can't create a solution given the description of the problem..

[small flax]

True. A memorized solution is easy to tell these days unless you’re a great actor

[keen hearth]

I think when people focus on FizzBuzz specifically they're missing the point. It was only ever intended as an example of a test of basic programming competency.

Such tests started to become used in interviews as companies found they were receiving many applications for programming roles from people who had zero prior knowledge or experience in programming.

Here's where it originates: https://blog.codinghorror.com/why-cant-programmers-program/

[keen hearth]

Hmm Well if the source is far enough away from the target, it may be that you can always find a manhattan-distance length path using some kind of simple algorithm. You might want to get a dice and try rolling it around on a grid to help you visualise it better.

[unborn sundial]

!e

from dataclasses import dataclass
from datetime import datetime
import json


def hide_field(field) -> str:
    return "**redacted**"


def format_time(field_timestamp: datetime) -> str:
    return field_timestamp.strftime("%Y-%m-%d %H:%M")


def show_original(event_field):
    return event_field


class EventSerializer:
    def __init__(self, serialization_fields: dict) -> None:
        self.serialization_fields = serialization_fields

    def serialize(self, event) -> dict:
        return {
            field: transformation(getattr(event, field))
            for field, transformation in self.serialization_fields.items()
        }


class Serialization:
    def __init__(self, **transformations):
        self.serializer = EventSerializer(transformations)

    def __call__(self, event_class):
        def serialize_method(event_instance):
            return self.serializer.serialize(event_instance)

        event_class.serialize = serialize_method
        return event_class


@Serialization(
    username=str.lower,
    password=hide_field,
    ip=show_original,
    timestamp=format_time,
)
@dataclass
class LoginEvent:
    username: str
    password: str
    ip: str
    timestamp: datetime


a = LoginEvent('a','b','c',datetime.now())
print(a.serialize())

[halcyon plank]

@unborn sundial :white_check_mark: Your eval job has completed with return code 0.

{'username': 'a', 'password': '**redacted**', 'ip': 'c', 'timestamp': '2021-06-27 16:29'}

[unborn sundial]

huh. classes can be decorated too.

[fiery cosmos]

can someone explain why sets are faster than lists when checking for a key

i know its about the o notation

but why cant lists be searched in the same way as sets

[keen hearth]

Sets are implemented as hash tables. Essentially, a hash table uses a function (called a hash function) to calculate a number (called the hash of the key). The items are stored in an array; the hash determines where (at which index) a given key will be stored in that array.

So, in theory, finding an element in a hash-table doesn't depend on the number of elements in the hash table.

[fiery cosmos]

im intermediate in py but havent touched algos

are algorithms essential

since nothing ive done yet required them

everything is built in

[keen hearth]

I'd say basic knowledge of algorithms is fairly essential. You might want to check out some of the pinned resources.

[fiery cosmos]

ive thought bubble sort

was requested in this year's final exams

but bubble sort is pretty dum

[keen hearth]

Oh right. Sorting algorithms are often used as a way to introduce the concepts behind the analysis of algorithms.

[fiery cosmos]

im hoping to get into a programming uni

there will be algorithms

i guess

[mental parcel]

No cap you could learn it all online quicker and cheaper

[fiery cosmos]

yea ik i just need a degree

[mental parcel]

Fair

[fiery cosmos]

to prove ik something

im only 16 and ive coded for 4 years but who believes me

atleast a github portfolio may help

not sure

[mental parcel]

Yea, I cant say much. I'm at uni

[fiery cosmos]

guess this is #career-advice

[mental parcel]

I think a public portfolio is good

Ah well, im off anyway

[analog ferry]

anyone aware of a recent usable geometric solver for python that allows to declare materials and their properties/relation

[mint jewel]

https://www.freecadweb.org/ has python scripting, it is the closest tool I know of, if I am understading your question right

[analog ferry]

its not cappable of the things i want to do

[mint jewel]

Rhinoceros can also do something along those lines, but it may also be not powerful enough, and also is not free.

but that is all I know of in this category

[dapper sapphire]

Where's the max function in python github repo?

When I do Go to definition it takes me to builtins.pyi

And what's the time complexity of max function in python?

I came across this that says, O(n):
https://stackoverflow.com/questions/5454030/how-efficient-is-pythons-max-function

And the time complexity O(n log(n)) is product of n and log(n) => n * log(n)?
But we just say O n log n or O of n log of n?

[keen hearth]

I think this is it: https://github.com/python/cpython/blob/74d60eab558bffdf5ca8ea2f5305e19b36bdb9a8/Python/bltinmodule.c#L1844

[halcyon plank]

Python/bltinmodule.c line 1844

builtin_max(PyObject *self, PyObject *args, PyObject *kwds)```

[vocal gorge]

it's almost certainly implemented the normal way:

def max(iterable):
    cur = None
    for el in iterable:
        if cur is None or el > cur:
            cur = el
    return cur

the actual code involves using the key and also a lot of checks

[dapper sapphire]

Thanks for the link. But why did you link to that branch instead of main branch?

Yeah I think implementing a max is simple. I wanted to see if the time complexity would be less than O(n). But I dont think less than O(n) would be possible since we would have to look at each element.

Implementing a sort is a real challenge, as long as it's not naiive sort(with nested for loop). Is it called bubble sort?

[analog ferry]

@dapper sapphire what is the context of what yu are doing, im udner the impression you litter lower level details without explaining the intent, which leads to a xy problem

[dapper sapphire]

Litter, as in trash talk, how so? But that wasnt my intent. But if someone felt offended I apologize.

I was just trying to find largest and second largest number in a list. So I thought of using max this would give the largest number, and then I thought of using sort so we can just access the last and second last/largest element.

[analog ferry]

that is a much better context than just the bits you use to implement (with what you wrote it was unclear what you try to do

do you have some more detaills, common solutions are partial sort/selection sort

for example find the largest element + its index, and then find the largest element of the concat of the lements before it and after it using islice

[dapper sapphire]

Initially I thought I'll use max, but that would just give us the largest number.

And then I thought of using sort to get the second largest number. So from there I figured I could skip max and just use sort to sort the list and get largest and second largest number by doing [-1] and [-2] slicing. Other things I mentioned went on a tangent.

[mint jewel]

look into the heapq module

though I am not sure how efficient it is for just 2 numbers

the most efficient way is probably just writing out the sizable for loop yourself, but there is quite a bit of logic there. For simplicity, sorting and indexing is best

[fervent saddle]

That’s wrong, it doesn’t give the same output with the example string

!e ```py
test = "xxyyx"
x_list = test.replace("y", " ").split()
x_count = sum(map(len, x_list))
substrings = 0

for x_section in x_list:
x_count -= len(x_section)
substrings += x_count * len(x_section)

print(substrings)```

[halcyon plank]

@fervent saddle :white_check_mark: Your eval job has completed with return code 0.

2

[fervent saddle]

This is my idea, but it might be wrong and there might be a nicer way to do it

But if it’s right then it should at least pass the time test because it’s O(n)

[analog ferry]

@fervent saddle is the input always well-formed? - if yes just find the count of xy sequences in the string

[fervent saddle]

The description says the string always has only x and y characters

[analog ferry]

is it allowed to use regex?

[fervent saddle]

I’m not really sure, I couldn’t find the webpage for it

[analog ferry]

(as far as i understood you are looking for all occurences of "xy+x"

[fervent saddle]

All occurrences of x + something + x where something has at least one y

[analog ferry]

the regular expression for that is xy+x

[fervent saddle]

Ok yeah, I see what you’re saying now

regex is built-in so I guess you could use it unless it specifically says not to

[analog ferry]

as far sa i understood, the task given requires the regex xy+(?=x) (the lookahead assertion ensures the x wont be consumed so it

so i believe a re.findall("xy+(?=x)", "xxyyx") could do (untested yet

[keen hearth]

I'm not sure...

[dapper sapphire]

oh ok.

[meager jetty]

... max heap ...

[dapper sapphire]

So if we do:

for key in sorted(dictionary):
  ...
  ...

Does usage of sorted in the loop mean dictionary would get sorted each time the loop runs? Or does the python interpreter optimizes this routine, sorts it once and does it sort it again?

[mint jewel]

The thing on the right of in will only ever run exactly once

[dapper sapphire]

Oh wow ok I didnt know that. Thanks a lot!!

[knotty magnet]

!e

for x in [print("hi"), 2, 3, 5]:
  print(x)

[halcyon plank]

@knotty magnet :white_check_mark: Your eval job has completed with return code 0.

001 | hi
002 | None
003 | 2
004 | 3
005 | 5

[dapper sapphire]

I had a similar idea, but thought of passing in a function after in. But what you did is a lot better and straight forward.

[late fable]

How do I start learning data structures

[covert swallow]

I'm not sure if this is the right channel to ask about classes in python but I think it covers data structs. What's the right way to add value to a class? On init or post_init?

e.g. I have a Person class on on init, its asking for name and email address. If neither is provided, it leaves those fields as empty. If we want to add a value to it, it would be to add it in the params of the class like so: Person("Bob", "[e-mail]")

Another approach is to add the the values in a method in the Person class like so:

p = Person()
p.create("Bob", "[e-mail]")

Which is a better approach? Or at least considered as best practice?

[meager slate]

your p.create is exactly the job of the __init__, so move that code to to __init__

[covert swallow]

Thanks. So I assume you mean that if I have to create a method, it should be to return a different data or mutate any self data if needed?

[meager slate]

__init__ initializes your object,, by setting any attributes needed (by mutating self)

[covert swallow]

I guess for names would be fine in the init but I don't think there's an easier way to handle emails into the Person object since it needs to pass the regex first

So init name, then .addEmail("[e-mail]") after

[meager slate]

you can accept these arguments in the init, so ```py
class Persom:
def init(self, name, email):
self.name = name
self.email = email

bob = Person('Bob', '[e-mail]')

you can add all the validation in the init itself

[covert swallow]

Yeah I'm using dataclasses so:

from dataclasses import dataclass

class Person:
  name: str
  email: str

Person("Bob", "[e-mail]")

However, my problem is validating the data on init

[meager slate]

then don't use dataclasses, just use a normal init

[covert swallow]

Isn't it the same? How do you validate on normal init?

So far, I've found that I can validate post_init

It looks dirty but it works

[meager slate]

def __init__(self, name, email):
       self.name = name
       if not validate(email):
                 raise ...

[meager slate]

i just learnt that you can __post_init__ in dataclasses so, that would work too

[covert swallow]

yeah thats what I did

And that caused me to doubt myself if it was better to just create a separate method specifically for each data validation (for phone numbers, addresses, email addresses, etc)

[meager slate]

you can make validatiors and then validate from init

for example

class Person:
    def _validate_name(self):
      if self.name == 'bob':
          raise ValueError('Sorry Bob.')
    
    def __init__(self, name):
       self.name = name
       self._validate_name()
bob = Person('bob')

[covert swallow]

Hmm... How do I include a _validate_name() function in unittest?

Correct me if I'm wrong, prefix _ to a function name would make it private right?

[meager slate]

im not very familiar with unittest, sorry

[meager slate]

not really, by convention its considered as private but nothing is actually private

no one's gonna stop me from accessing that method from outside

[covert swallow]

True but it won't be able to be called in a separate file easily right? Something like app._validate_name()? I've yet to try this

[meager slate]

!e ```py
class F:
def _am_i_private(self):
print('No')

f = F()
f._am_i_private()

[halcyon plank]

@meager slate :white_check_mark: Your eval job has completed with return code 0.

No

[meager slate]

we all just agree that we wont mess with things starting with underscores

[covert swallow]

Good to know!

OK I think I got it properly designed... I hope. Thanks @meager slate

[meager slate]

👍

[foggy depot]

what would be the best method to find the indexes of the first x occurrence of an element in a list
example
list = [1,2 ,1,1,4,5,6,1,2,1]
here i want to find the indexes of the of the first three "1"
what would be the best way of doing this

[fiery cosmos]

Loop through from the start normally and save the 1 indexes, but stop when you get x of them

[supple coyote]

Hello My dear fello geeks Im new to the Discord Sorry If Im Wrong, I face one problem while practising backtracking algorithm in python I cant find what the problem is please help to find... https://stackoverflow.com/q/68162305/15982124 This is Link Please help to solve this Thank You

[fringe elm]

yes it will be the best way

[dapper sapphire]

There are some good resources that are pinned.

[haughty wave]

i'm creating an application but i don't understand why this code doesn't work, a button should come out with hello kivy world

KV = """
Screen:

MDRectangleFlatButton:
   text: "Hello kivy world!"
   pos_hint: {"center_x": 0.5 "center_y": 0.5}

"""

[lyric burrow]

what things do i need to keep in mind while preparing for programming interviews in python opposed to c++

[dapper sapphire]

You are looking for #user-interfaces channel.

[silk breach]

Could anyone help me with coding the a* pathfinding algorithm

I'm pretty sure I understand how the algorithm works, I just can't figure out how to code it

[keen hearth]

Have you already tried implementing breadth-first and depth-first search, and seen how you can switch from one to the other just by changing out the data structure used for the frontier? (a stack for dfs, and a queue for bfs)

A* search is similar, except that you use a priority queue for the frontier. The priority of an node in the queue should be the sum of the path cost to reach that node and a heuristic estimate of the remaining cost to reach a goal.

[fiery cosmos]

I'd say A* is more similar to best first. while in bfs and dfs there are no heuristics, best first seems more nearer. and as LX said, if you have coded bfs then A* will be easier. and about prio queue. since you're using python, prio queue are built in so they come in handy.

[terse pewter]

why this result returns 3 instead of 4 (main() returns 3 and +1 should equal 4?

def abc():
    return dfe()

def dfe():
    return main()+1

def main(x=2):
    c=3
    return c

print(main(abc()))

OUTPUT : 4

[jolly mortar]

you're printing main(abc()), main always returns 3

[halcyon plank]

Here's how to format Python code on Discord:

```py
print('Hello world!')
```

These are backticks, not quotes. Check this out if you can't find the backtick key.

[terse pewter]

oh thanks got it ))

I want to access main()'s c value in dfe(), how can i achieve this?

    return dfe()

def dfe():
    return c+1

def main(x=2):
    c=3
    return c , x

print(main(abc()))```

[jolly mortar]

you can't access a function's local variable from outside the function

[terse pewter]

somebody offered this solution and worked !

def dfe():
  c,x = main()
  return c+1

[dapper sapphire]

Is this pretty much how we get minimum value in python?:

!e

import sys
minimum_value = -sys.maxsize
print(minimum_value)

[halcyon plank]

@dapper sapphire :white_check_mark: Your eval job has completed with return code 0.

-9223372036854775807

[vocal gorge]

Minimum value of what?

[dapper sapphire]

sort of like int_max, int_min

[vocal gorge]

Python ints are infinite-sized, they can represent any value you have the memory for, and have no min/max value or infinities. Float infinity can be obtained with float("inf").

[dapper sapphire]

I just needed a small enough number to perform comparison with some data. Data given could have -1 upto -1000, so -sys.maxsizeseems reasonable.

[vocal gorge]

You could use negative float infinity, -float("inf").

it compares lower than any int or float.

[jolly mortar]

except itself

[dapper sapphire]

!e

min_inf = -float("inf")
print(float("inf"))
print(-float("inf"))

print(min_inf == (-float("inf")))

[halcyon plank]

@dapper sapphire :white_check_mark: Your eval job has completed with return code 0.

001 | inf
002 | -inf
003 | True

[dapper sapphire]

I am performing an != is not equal to comparison.
So I guess -float("inf") also works and I could use either. Thanks!

[half lotus]

Please stop. If you want to get help see #❓｜how-to-get-help

[lyric burrow]

how do i check if num1 and num2 don't point to the same location in memory?

[knotty magnet]

is keyword

[lyric burrow]

i did say is not , is it not correct?

[knotty magnet]

probably

[lyric burrow]

i also tried ! is but that didn't work either

[fervent saddle]

Are you sure you want to do that?

What does it want you to do?

[lyric burrow]

not so much

[lyric burrow]

it wants me to return the indexes of the numbers, when added up equal to the target value, provided the indexes of those numbers are not the same

[fervent saddle]

Oh yeah, I see what you’re trying

[lyric burrow]

oh wait

[fervent saddle]

What’s happening is python uses the same int objects for values from -5 to 256, and also for literals in the code with the same values

[lyric burrow]

i can check it with nums.index(num1) != nums.index(num2)

but

nah, i don't see how that will work

when num1 = num2

[fervent saddle]

enumerate might help

[lyric burrow]

how would that affect it tho?

[fervent saddle]

If both numbers are the same values then they’ll be the same objects

[lyric burrow]

oh

i'll try enumerate

[fervent saddle]

I should say CPython does this

It might be different with other implementations

But yeah, this seems like a good place to use enumerate

[lyric burrow]

python does work on cpython, right?

[fervent saddle]

Yeah, CPython is regular python

[lyric burrow]

i mean, i've seen how data structures like list and dictionary are implemented in C

and used by python

is that what CPython is?

[fervent saddle]

CPython is written in C, yeah

[glossy zephyr]

Hi guys, I want compare a date in the past and verify it if has 24 hours later or 1 day +.
how is the best approach to do it?

[lyric burrow]

how do i use it tho?

[fervent saddle]

So you know what index you’re on

[lyric burrow]

i can't really figure out how do i do that with enumerate(iterable, start)

[fervent saddle]

The first value in each tuple will be the current index

[vocal gorge]

the common usage pattern is for i, el in enumerate(lst): or something like that.

[spiral fractal]

Anyone know a simple graph dfs algo

[spiral fractal]

plz dm me

[calm spade]

How do i convert a string list ("[data]" / '[data]') to a normal list ([data])

[spiral fractal]

is this right? def switchLst(data1, data2): res = [] for data_1, data_2 in zip(data1, data2): res.append(int(data_1)/int(data_2)) return res

[keen hearth]

If you have the code for a python list contained within a string, then you need to evaluate that string, using either eval (very unsafe if the string is user-provided) or ast.literal_eval (a safer limited alternative). If the string contains a JSON list, e.g. received from an API, then you should parse it using json.loads.

[karmic axle]

hi

my name is Vansh Khera

and i like to code :D

i want to pc professional for my project

but its paid :C

[keen hearth]

No way around that I'm afraid. Although I think they make it available for free for students.

[karmic axle]

i live in india

i have a school email

but it shows that its not valid :(

me

just jking

[fiery cosmos]

can some one help me? With .py to .exe

[dapper sapphire]

So to perform comparison on data type can we use either == or is operator?

!e

value = "This is a test"

print(type(value) is str)
print(type(value) == str)

print(type(value) is "str")
print(type(value) == "str")

[halcyon plank]

@dapper sapphire :white_check_mark: Your eval job has completed with return code 0.

001 | <string>:6: SyntaxWarning: "is" with a literal. Did you mean "=="?
002 | True
003 | True
004 | False
005 | False

[deft scaffold]

hello guys
i have this return:

{'AggregateEvaluationResults': [{'EvaluationResultIdentifier': {'EvaluationResultQualifier': {'ConfigRuleName': 'OrgConfigRule-REQUIRED_TAGS-ruleID', 'ResourceType': 'AWS::CloudFormation::Stack', 'ResourceId': 'arn:aws:cloudformation:sa-east-1:123456789101:stack/stackname/9stack-id'}, 'OrderingTimestamp': datetime.datetime(2021, 6, 3, 17, 10, 15, 66000, tzinfo=tzlocal())}, 'ComplianceType': 'NON_COMPLIANT', 'ResultRecordedTime': datetime.datetime(2021, 6, 10, 17, 42, 48, 860000, tzinfo=tzlocal()), 'ConfigRuleInvokedTime': datetime.datetime(2021, 6, 10, 17, 42, 48, 624000, tzinfo=tzlocal()), 'AccountId': '12345678', 'AwsRegion': 'sa-east-1'},

How can I access the ResourceId?
I've try for some tutos to access** nested dict** but without much successs
Here's the code that i've tried

for key in response:
     for items in response[key]:
         for item in items['EvaluationResultIdentifier']['EvaluationResultQualifier']['ResourceId']:
             print(item, '\n')
 # print(response["AggregateEvalutionResults"]["EvaluationResultIdentifier"]["ResourceID"])
response Example:

[vocal gorge]

sure, types are singletons, so no reason not to use is. That said, comparing types exactly isn't a very common thing to do in general - usually you need isinstance or issubclass instead.

[fiery cosmos]

How can i calculate this?

Jeffrey J. McConnell, Analysis of Algorithms: an Active Learning
Approach , 2001 PAGE 52

[dapper sapphire]

So something like following:

!e

value_string = "This is a string"
value_list = ["This is a list"]

print("Below should be True")
print(isinstance(value_string, str))
print(isinstance(value_list, list))

print("\nBelow should be False")
print(isinstance(value_string, list))
print(isinstance(value_list, str))

[halcyon plank]

@dapper sapphire :white_check_mark: Your eval job has completed with return code 0.

001 | Below should be True
002 | True
003 | True
004 | 
005 | Below should be False
006 | False
007 | False

[dapper sapphire]

We would use isinstance because that's more common then directly comparing data type with type(variable)?

[vocal gorge]

Yeah, since in most cases you care about whether something "acts like a list", for which you can check whether it's an instance of list or its subclass with isinstance. It's rare to care about what something's exact type is.

though for the specific case I mentioned, the nicest way would be to check if isinstance(thing, collections.abc.Sequence) (or MutableSequence, or whatever other "interface" you need)

[deft scaffold]

print(response['AggregateEvaluationResults'][0]['EvaluationResultIdentifier']['EvaluationResultQualifier'].get("ResourceId"))

got it

[dapper sapphire]

I dont understand the nicest way would be to check...:

I mean when I type the if statements, I would have the if. I just left it out in the above. Am I missing something?

So in if isinstance(thing, collections.abc.Sequence) thing would be a variable that for example, could be a "string", [list], or {set} etc.

collections.abc.Sequence would be str, list, set

Oh you literally meant collections.abc.Sequence because that's an actual module! Sorry about that.

!e

import collections

value_string = "This is a string"
value_list = ["This is a list"]
print(isinstance(value_list, collections.abc.Sequence))
print(isinstance(value_string, collections.abc.Sequence))

But the above will return:

True
True

For both list and a string

[half lotus]

Why did you write "but"? How is it different from your expectations or requirements?

[dapper sapphire]

I misunderstood.

ConfusedReptile has a specific and different use case, so they have to use collections.abc.Sequence.

I am trying to identify if something is a list or a string, then in that case, it would be isinstance(variable, list) or isinstance(variable, str)

https://python-reference.readthedocs.io/en/latest/docs/functions/isinstance.html

Thanks for pointing to isinstance!
I think this is one of the good use cases where type and isinstance would work differently:
https://stackoverflow.com/a/1549814

[last fulcrum]

I'm trying to implement a linked list. I get this error. Would like some help

AttributeError: 'NoneType' object has no attribute 'next'

This is the code

if current.next is None:

[brisk saddle]

current itself is None then. @last fulcrum

[last fulcrum]

while current:
            current = current.next

        if current.next is None:
            self.tail = self.Node(element)
            current.next = self.tail
            self.size += 1

There is a None in my list

When I was testing

But still this error

[brisk saddle]

your problem is probably coming from the fact that set current to the next node at the beginning.

at some point down the line, your code reaches the last Node.

from there, it then sets the variable to the last Node's next node, which will always be None.

then you perform the check, which is causing the error.

[last fulcrum]

This line though

while current: current = current.next

That means while current is valid, move to the next value of current

Once it becomes None, quit

Or?

[brisk saddle]

yes, but if current is set to the actual last Node, it will still run, set the next Node to None, then it errors.

[last fulcrum]

So you're saying I should timeout before the last node so that the actual next would be none?

[brisk saddle]

ehm, i said nothing about what you should do. i only pointed out the flaw in your code.

what is the purpose of this?

[last fulcrum]

To implement a linked list

I need to create one for interview questions

[brisk saddle]

should have been more specific, what is the purpose of the code you sent?

[last fulcrum]

To insert at the tail.

Here's the full thing

    def insert_at_tail(self, element: Union[str, int, None]):
        """
        Inserts an element at the back of the linked list, thereby making it become the tail.
        """
        current = self.head

        while current:
            current = current.next

        if current.next is None:
            self.tail = self.Node(element)
            current.next = self.tail
            self.size += 1

[brisk saddle]

so an append operation.

[last fulcrum]

I doubt it's called append, but yes

[brisk saddle]

it is.

i think you misunderstand one of the benefits of a Linked List. Linked Lists have O(1) insertions and deletions on the head and tail of the List.

[last fulcrum]

Thay have O(1) if you know the position also

[brisk saddle]

O(1) operations in this case are achieved by having fields stored in the Linked List object that store the head and tail. you are essentially performing an O(n) search for the last node, when you should have a reference to the head and the tail stored somewhere in the Linked List object.

[last fulcrum]

I wrote the code for that, but I can't even make this work so that's useless

[brisk saddle]

why would it be useless?

[last fulcrum]

Because at the moment I'm struggling to get the tail. That is why i used this

[brisk saddle]

well that is something we can work to fix here. do you have a head and tail field in your Linked List?

[last fulcrum]

Yes. I'll post

[brisk saddle]

so what is the purpose of creating this function when you could just use the tail field?

[last fulcrum]

class SinglyLinkedList:
    """
    Implements a Singly Linked List.
    """

    class Node:
        """
        Initializes the node class for storing an element and it's next value.
        """
        __slots__ = 'element', 'next'

        def __init__(self, element: Union[str, int, None] = None, next=None):
            self.element = element
            self.next = next

    def __init__(self):
        """
        Sets the head of the list and it's size
        """
        self.head = self.Node()
        self.tail = None
        self.size = 0

    def insert_at_head(self, element: Union[str, int, None]):
        """
        Inserts an element at the front of the linked list, thereby making it become the head.
        """
        if self.head is None:
            self.head = self.Node(element)

        self.head = self.Node(element, self.head)
        self.size += 1

    def insert_at_tail(self, element: Union[str, int, None]):
        """
        Inserts an element at the back of the linked list, thereby making it become the tail.
        """
        current = self.head

        while current:
            current = current.next

        if current.next is None:
            self.tail = self.Node(element)
            current.next = self.tail
            self.size += 1

    def print_linked_list(self):
        """
        Prints all the elements of the linked list.
        """
        if self.size == 0:
            return None

        current = self.head
        while current is not None:
            print(current.element)
            current = current.next

[last fulcrum]

I have not implemented it as I said above

I just wrote it in my book

Even then it's not right

[brisk saddle]

oh, well.. implement it. :p
are you having problems implementing the O(1) operations, or something, so you resorted to using the code you have now?

[last fulcrum]

Yes. I'm having problems always getting the tail. So I just decided to iterate through the list.

Far easier

And I got problems with that

So I came for help

[brisk saddle]

i think i misunderstand something here. you already have the tail, it is stored in the tail field in your Linked List..

at the start, when the Linked List only has one Node, head and tail should essentially point to the same Node.

[last fulcrum]

That is wrong. Tail is None. It does not link to the other elements of the linked list.
Take for example me inserting 4,3,2,1
One should become the tail. But that is not possible in my code

[brisk saddle]

oh, then this is a problem with how you are pulling off insertion at the tail in general, not specifically when you already have Nodes in the Linked List.

[brisk saddle]

first off, make sure you do this, so assuming the List is not empty, you can always have access to the tail, and because of this, always keep it consistent with the actual tail during append operations.

[last fulcrum]

Ok

So as you keep on adding to head the tail is still that last value

[brisk saddle]

yup!

[last fulcrum]

Ok. I'll implement it.

[brisk saddle]

second, insertion at the tail should follow these steps:

- make a new node
- if the Linked List is empty (LinkedList.head == None, or you have an internal ``length`` variable that is ``0``), then you
set the head, and tail to the new node
- if the Linked List is *not* empty (LinkedList.head != None, or an internal ``length`` variable is > 0), then you
set the ``next`` property of the *current* tail to the Node you *just made*, and set the new Node as the tail.
- set the value of the Linked List to whatever you want

or, put simpler:

• if the Linked List is empty, set the head and tail to the new Node

• if it is not empty, set the current tail's next node to the new node, and set the tail field to the new node.

[last fulcrum]

Thanks

[brisk saddle]

got it working?

[last fulcrum]

Yet to implement

I'll ping you

[brisk saddle]

alrighty

[last fulcrum]

Done

[brisk saddle]

sick, everything working?

[last fulcrum]

I think so. Just one more question

[brisk saddle]

sup?

[last fulcrum]

This prints the elements and also prints None

It's not supposed to do that

    def print_linked_list(self):
        """
        Prints all the elements of the linked list.
        """
        if self.size == 0:
            return None

        current = self.head
        while current is not None:
            print(current.element)
            current = current.next

Maybe I should do while current

[brisk saddle]

no, that is the same thing more or less. just different way of writing it.

[last fulcrum]

Ah. Should I paste the full code?

[brisk saddle]

no i think we can go off this.

just trying to visualize it in my head.

could you show your output? @last fulcrum

[last fulcrum]

7 9 2 None

`
l = SinglyLinkedList()

l.insert_at_head(9)
l.insert_at_head(7)
l.insert_at_tail(2)

print(l.print_linked_list())
`

[brisk saddle]

ah alright.

oh, you know what, that is not because of your function.

it is because you are printing whatever print_linked_list returns. your function does not return anything, so print will print None.

[last fulcrum]

I assigned a variable to it right now and still got None

I think that is it

[brisk saddle]

send the code.

[last fulcrum]

You're right

I just need to call the function

Without print

[brisk saddle]

yeah

[last fulcrum]

Thank you @brisk saddle

[brisk saddle]

no problem!

[cinder bloom]

Any good courses or books for learning Data Structures and Algorithms in Python?

[dapper sapphire]

I found this helpful for Linked Lists:
https://runestone.academy/runestone/books/published/pythonds3/index.html

There are also some other resources in pinned area.

[crystal prawn]

I recommend this https://www.amazon.com/Computer-Science-Distilled-Computational-Problems/dp/0997316020 @cinder bloom *

[remote carbon]

Hi everyone, not sure I am in the correct room here but I am looking for some tutelage with hand tracing. I am just learning python while taking a cmpsc 101 class and having trouble with a hand trace assignment with if and elif constructs. I know how the program works out but just don't really understand how to put it into a table. is there a good site that anyone can point me to with some examples?

[cinder bloom]

I know about this but the explanations in this book seem a little basic, any other books or resources?

[cinder bloom]

Does this book explain concepts in Python? The reviews on Amazon say it's a really condensed and a book for basics.

[last fulcrum]

Use what I use. Honestly, it's the best book for learning DS & A.

https://www.amazon.com/Structures-Algorithms-Python-Michael-Goodrich/dp/1118290275

[cinder bloom]

Alright, one more thing, I'm learning DSA for interviews so I can use this book for that purpose right?

[last fulcrum]

Yes you can. I'm also learning for interviews so you can DM me if you need help or want to solve together.

Here's the plan I made and use.

1. Go through every chapter in the book. Be detailed, take notes, make sure to really understand the data-structure or algorithm.
   It's big O, runtime, etc.

2. After every chapter in the book, solve questions on Leetcode and this site https://app.codesignal.com/interview-practice
   Solve only questions based on what you've learnt for the chapter.

3. If you don't understand a question, try to solve it using hints. If you still don't understand it, then look at the solution.
   Understand it. Then come and solve the question again in a few days time.

4. Use various other resources.(Ex: Cracking the coding interview.)
   Things are explained better by different people. YouTube is your best friend.

5. After you have finished the book, solve these 75 questions.
   https://www.teamblind.com/post/New-Year-Gift---Curated-List-of-Top-75-LeetCode-Questions-to-Save-Your-Time-OaM1orEU
   Then buy Leetcode premium and solve the questions for the company you want to go to. I don't recommend doing any easy
   questions at this stage. Do all those while going through the book.

6. Use flashcards to help you with your notes while solving the questions at stage 5. You'll be asked those questions at the interview.

7. Good luck lol.

[lyric burrow]

what would be the difference in time and space complexity between

for i in range(len(array)-1):
    array[i]```
and 
```py
for ele in array:
    ele```

[cinder bloom]

Oh, I am using this site https://www.pepcoding.com/resources/ , It has almost 1200 questions, all of them curated and selected personally by top tutors and they have helped many people get placed into FAANG, The teachings and solutions are in Java but almost all of the questions are taken from Leetcode and a simple google search with the question title gets the leetcode link and I can get it's solution in Python, Thanks for the resources and the help!

[fervent saddle]

They’re both the same complexities

[scarlet sleet]

None? I can see no difference.

[last fulcrum]

Awesome

[spice sundial]

does anyone know about the Apriori algorithm

[placid flicker]

Yes, do you have a question about that?

[spice sundial]

yes have a project based on Apriori algorithm

https://github.com/ymoch/apyori
I have tried running this script but the results.txt is coming same as the input tsv

[placid flicker]

I'm not familiar with that repo. Is your input data tab-seperated as stated in the readme?

[spice sundial]

yes .tsv file

[spice sundial]

do you have any project based on this algorithm

[placid flicker]

Yeah but that's written in Haskell though, not python

I can't help you any further without an error message. Maybe try asking the author

[spice sundial]

okay let me try hard enough

[lunar jacinth]

class Node:
        def __init__(self):
                self.data = None
                self.previous = None
                # self.next = None
                # introduce a 'next' node for doubly linked list
        def get(self):
                return self.data
        def getPrev(self):
                return self.previous
        def setPrev(self,previous):
                self.previous = previous
        def set(self,data):
                self.data = data
        node = property(get,set)
        prev = property(getPrev,setPrev)

# Convert Singly Linked List to Doubly Linked List
class SList:
        def __init__(self):
                self.tail = None
                # self.head = None
                # introduce 'head' node for double link
        # appends to tail of list
        def Append(self,data):
                pdata = Node()
                pdata.node = data
                if (self.tail == None):
                        self.tail = pdata         
                else:
                        pdata.prev = self.tail
                        self.tail = pdata
        # prepends to head of list
        def Prepend(self,data):
                newNode = Node(data)
                head = self.tail
                while head.previous is not None:
                        head = head.previous
                head.previous = newNode # this is just a stub
        # inserts data after found data
        def InsertAfter(self,find, data):
                if not find:
                        return
                newNode = Node(data)
                newNode.previous =  find.previous
                find.previous = newNode

Could someone help me with the insertafter method?

[frosty cobalt]

2n, log n, log log n, n2, n, √n , n!, n3, n3/2, n log n, en+1, n2 log n

can anyone sort this in accending order

[knotty magnet]

this seems like homework

[reef remnant]

log log n, log n, n2, n, 2n, n3, n3/2, en+1, n2 log n, √n, n!

[knotty magnet]

how is sqrt n second greatest, when n and n^2 exist

@reef remnant

[reef remnant]

shhhh

[knotty magnet]

half of those are out of order

[reef remnant]

you're right it does seem like homework

[bold panther]

@frosty cobalt go to an online graph site and plug them all in

[frosty cobalt]

i did it,

[frosty cobalt]

thank u

😄

[bold panther]

That isn’t right

Lol

[frosty cobalt]

i know

[bold panther]

Ok

[frosty cobalt]

but still she/he helped

[bold panther]

Sorry dawg you can’t just put homework here and expect people to do it for you

We provide help. Not handouts

[fiery cosmos]

How can i get a certain amount n numbers between two numbers x and y (x<y), but most of them have to be close to y
For example if i want 50 numbers between 1 and 100, i would get: 1 10 70 80 85 90 91 92 93...

[vocal gorge]

why this specific distribution?

[fiery cosmos]

i need it

not exactly like the example tho

just the same way

i can think about log but don't know how to do it

[vocal gorge]

well, if you can write down the formula for the probability distribution, it's possible to sample values from that distribution (by generating a random value from 0 to 1, then plugging that into the (numerically or analytically) inversed cumulative distribution function)

so just make up some probability density function that's:

1. Only nonzero from x to y
2. Mostly concentrated close to y.

Though I guess I need to first ask... why do you want such a distribution?

[tender atlas]

Cracking with interview code books is only helpfull for any tech interview?

[exotic wigeon]

For example you could have a normal distribution with mean of y and dispersion of your choice. For any sampled number t greater than y just use -(t-y)+y = 2y - t (I mirrored t around y) and if the result is less than x you can just sample again.
The resulting distribution will have more numbers around y and if you have enough IRL time you will be able to calculate the average amount of samples needed to get a distribution of n values depending on dispersion you chose.

... it can be done by calculating a probability of getting outside of [x, 2y-x] range with your distribution. Look at 'confidence intervals'

[past escarp]

if i want my function to return 2 int values i do; return int1,int2

but i want to assign those values to 2 variables in 1 line of code

i,j = func(), for some reason doesnt work

[vocal gorge]

this is hardly #algos-and-data-structs. However, tuple unpacking allows doing exactly what you wrote; you simply return a tuple of two ints from the function.

[past escarp]

sry for that, doing a data structure problem just encountered it there, so it should work how i wrote it ?

[vocal gorge]

that's how you can unpack a tuple of two elements into two variables, yes

[past escarp]

arigato sensei

[scenic hawk]

.topic

[grand haven]

**No topics found for this channel.**

Suggest more topics here!

[lunar jacinth]

        def Delete(self,data):
                curr = self.head
                while curr is not None:
                        if curr.data ==  data:
                                if curr.previous is not None:
                                        curr.previous.next = curr.next
                                        curr.next.previous = curr.previous
                                else:
                                        self.head = curr.next
                                        curr.next.prev = None
                        curr = curr.next
``` I'm trying to give it a given node to delete...is anyone experienced enough in doubly linked lists to let me know whats wrong here?

[wheat kiln]

how to answer this?

[keen hearth]

At a glance, it looks ok to me. What happens when you test it?

[lunar jacinth]

When I test it out, nothing seems to delete

[keen hearth]

What does the test-case look like?

[lunar jacinth]

       l = SList()
        l.Append(3)
        l.Append(4)
        l.Append(5)
        l.Append(6)
        l.Prepend(10)
        l.Append(2)
        l.Append(9)
        l.InsertAfter(4, 45)
        l.InsertBefore(2,30)
        l.Delete(2)
        l.Count()
        l.Output()

[lunar jacinth]

my count doesn't seem to work either for some reason. I could paste my code if you want?

[keen hearth]

Sure

Use this:

!paste

[halcyon plank]

Pasting large amounts of code

If your code is too long to fit in a codeblock in discord, you can paste your code here:
https://paste.pydis.com/

After pasting your code, save it by clicking the floppy disk icon in the top right, or by typing ctrl + S. After doing that, the URL should change. Copy the URL and post it here so others can see it.

[lunar jacinth]

https://paste.pythondiscord.com/nayahuvise.rb

[lunar jacinth]

sure, linked it.

never mind, fixed. the count method is still messed up a little though

[keen hearth]

Hmm, well I can't see anything wrong with the function itself, so it would seem that the structure of the linked-list is getting messed up somehow. Do you have a way to visualise this structure?

[lunar jacinth]

i dont

but i got my delete method working

its just this count method now

[valid slate]

def make_maze(draw, start):
    visited = [start]
    stack = [start]
    current_node = start
    while len(stack) > 0:
        nodes = []
        for neighbor in current_node.neighbors:
            if neighbor not in visited:
                nodes.append(neighbor)
        if len(nodes) > 0:
            node_chosen = random.choice(nodes)
            current_node.make_barrier()  # make color black
            node_chosen.make_barrier()
            current_node = node_chosen
            visited.append(node_chosen)
            stack.append(node_chosen)

        else:
            node = stack.pop()
            current_node = stack[-1]

        draw()

I'm trying to make a maze generator
but something is off
can somebody help me?

[wise sierra]

is it important to know the implementation of hash table?

or is it enough just to know about it and to know how to use in

[fervent saddle]

It’s probably good to understand the idea behind it

And maybe the idea behind open and closed addressing

[wise sierra]

thanks bro

[brisk saddle]

are there any situations where you would not want an ordered dictionary?

i am implementing a dictionary and i am thinking about making it ordered, but if there are situations where an unordered idctionary is necessary, i might want to make that behavior toggle-able or something.

[fiery cosmos]

I guess if you don't need it to be ordered and the unordered implementation is faster/ more memory efficient etc.

[fervent saddle]

I think if you want the order to be randomized then you randomize something in the hash formula like they do with strings

Otherwise, the order isn’t really random, it’s just arbitrary

if that’s what kind of thing you’re thinking of

[serene panther]

true

[runic tinsel]

But it's not an option?

If you're using reasonably new python, you can't create an unordered dictionary

[brisk saddle]

i think you misunderstand. i was not talking about Python's built-in dictionary. i was talking about implementing my own.

specifically, using a Hashtable.

[runic tinsel]

Ah ok

[tender atlas]

One of my friend gives me a task

Create a algo to calculate Tendering BOQ deviation

[dawn pebble]

Hi, I had a question regarding Leetcode # 28. Implement strStr().
`
class Solution:
def strStr(self, haystack: str, needle: str) -> int:
if len(needle) == 0:
return 0

    for i in range(len(haystack) - len(needle) + 1):
        char_window = haystack[i:i+len(needle)]
        if char_window == needle:
            return i
    
    return -1

`
this is the algorithm I wrote. Specifically, I'm confused regarding the time complexity of the algorithm. I'm guessing it's O(n) but I'm not sure how the slice function in python3 is implemented under the hood. Can anyone share their insight on this? Thank you.

is the time complexity O(n*k)?

[fiery cosmos]

regardless of how slice works, if char_window == needle: will need to potentially compare every character in needle, so yeah that part is O(k), for O(n*k) overall

but if k is small that's basically O(n)

btw as you may know this task is what ||str.find|| does

[dawn pebble]

ah okay. thank you!

yeah, after coding the solution I saw people use str.find. It sort of defeats the purpose of the problem but I filed it away in my memory

[ivory yacht]

You might want to look up string finding algorithms by the way, there’s a number of different methods which pre-process the needle to allow skipping past sections of the haystack. str.find of course does this.
https://en.wikipedia.org/wiki/String-searching_algorithm

[vocal gorge]

and yes, slicing a string or list is O(k) time like you were worried - it creates a new list/string rather than a "view" into the original one.

[austere sparrow]

@dawn pebble your solution is O((H - N) * N) where len(haystack) = H and len(needle) = N because you're making H - N iterations and comparing two strings of size N

[vocal gorge]

(with an exception of [:] which for strings, since they are immutable, does nothing and just returns the original string)

[fiery cosmos]

same for lists if I'm not wrong however a string is a list of chars......

[vocal gorge]

Nope, for lists [:] creates a shallow copy of the entire list.

a string is a sequence of chars just like a list is a sequence of... anything, but it's not implemented as a list, and doesn't have to behave exactly the same

[fiery cosmos]

ok

It seems that df[:] has also no effect

i.e. give just the df as output

[vocal gorge]

like, df[:] is df?

[fiery cosmos]

pandas DataFrame

[vocal gorge]

I mean, is it exactly the same object (df[:] is df returns true), not a copy?

[fiery cosmos]

for a copy, one need to write df.copy(deep=True), then you would get a copy, otherwise just a pointer to the original df

dfCopy = df

any changes on dfCopy would result in changes of df as well

[vocal gorge]

this is always another reference to df regardless of what object it is, I'm asking about [:]

[vocal gorge]

looks like no, it does create a copy:

!e

import pandas
df = pandas.DataFrame([1,2,3,4,5])
print(df[:] is df)

[halcyon plank]

@vocal gorge :white_check_mark: Your eval job has completed with return code 0.

False

[knotty magnet]

weird, i thought it'd take numpy's stance: create a view

[vocal gorge]

or at least a different object

yeah, it may well be a view

[knotty magnet]

ah, nevermind, my thought makes no sense

[vocal gorge]

!e

a = "hello"
print(a[:] is a)

[halcyon plank]

@vocal gorge :white_check_mark: Your eval job has completed with return code 0.

True

[vocal gorge]

for strings though, [:] is optimized and, since they are immutable, just does nothing and returns the object

[knotty magnet]

how often is that used though, seems like a waste of time to implement

[vocal gorge]

for that matter:

import copy
copy.copy(a) is a # True

[fervent turtle]

Any cp coder here ,I want big help
I can't even solve easy problems .

[fiery cosmos]

you are right, it make a proper copy [:] for DataFrames

interessting

now the next question is, what is quicker copy via [:] or .copy(deep=True)

got it

df = pd.DataFrame(np.random.randint(0,100,size=(10000000, 40)), columns=list('ABCDABCDABCDABCDABCDABCDABCDABCDABCDABCD'))
%timeit af = df[:]
%timeit df2 = df.copy(True)

first method take 119µs
sedond 3.81s

[onyx root]

they do different things, don't they?

[fiery cosmos]

Not sure.
I read somewhere that the safest way to copy a data frame is the latter one. However, the first method seems to create a copy as well. Maybe this “deep =True” thing does stuff which might not be required (necessarily)

[keen hearth]

I believe df[:] is equivalent to df.copy(deep=False). The default value of deep is True.

[keen hearth]

Selecting a slice of a dataframe without doing a deep-copy is sort of like creating a "view" into that dataframe. If you change values in the slice, the original dataframe will be modified also.

[fiery cosmos]

yes, deep=False is even quicker with 96µs

[keen hearth]

Well that's probably random-error. I think it is literally the same operation as [:].

[fiery cosmos]

yeah

[valid slate]

def make_maze(draw, start):
    visited = [start]
    stack = [start]
    current_node = start
    while len(stack) > 0:
        # pygame.time.delay(10)
        # time.sleep(0.1)
        unvisited_nodes = []
        for neighbor in current_node.neighbors:
            if neighbor not in visited:
                unvisited_nodes.append(neighbor)

        if len(unvisited_nodes) > 0:
            node_chosen = random.choice(unvisited_nodes)

            node_chosen.make_barrier()
            current_node.make_barrier()
            # current_node.maze_lines(node_chosen)

            current_node = node_chosen
            visited.append(node_chosen)
            stack.append(node_chosen)

        else:
            node = stack.pop()
            node.make_open() # color: green
        draw()

This doesn't work for some reason it backtracks the whole grid

here is a vid

[queen ruin]

um hope this is the right place, so i heard about reindent.py that auto indents correctly, is there a way to run them on strings?

[tawdry ibex]

Is there a way to make a value in a module that is persistent ex:
import module as m
m.value = 100
then later I can re-import the module in another script and value will be 100

[queen ruin]

i would use a simple pickle or json or a common file..

[tawdry ibex]

simple pickle
explain

because i have a "plugin" module which is required in mutiple places
but i want it to only load plugins/addons once instead of reloading every time it has to be imported

[tawdry ibex]

ok so @queen ruin i tried pickler with a empty dict

folder = f'{__path__[0]}'
with open(f'{folder}/events.pickle',"w") as file:
    pickle.dump({},file,protocol=_pickleLevel_)```
and got ```py
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/home/walksamator/github/McPy/classes/plugin/event/__init__.py", line 16, in _reset_
    pickle.dump({},file,protocol=_pickleLevel_)
TypeError: write() argument must be str, not bytes```

anyone know why it failed to pickle it

[lean dome]

you need to open the file with mode "wb", not "w"

pickle is a binary format, not text.

[runic estuary]

night guys, what's the difference between a 3 node cluster in a red black tree and a four node cluster?

[wintry oracle]

I just asked a data structure algorithm question in #help-avocado if anybody here can lend a hand to help 🙂

[brisk saddle]

evening.

[icy tendon]

Hi I need help with splitting with more than one word in a single go Thanls

Like I have a string having div and span and I want to split by every div and span

[last fulcrum]

Open a help channel. This channel is for Discussion of general computer science, algorithms, mathematical concepts, data structures as they relate to solving problems with Python.

Check #❓｜how-to-get-help for more information.

[halcyon plank]

@fiery cosmos Please don't try to ping @everyone or @here. Your message has been removed. If you believe this was a mistake, please let staff know!

[placid flicker]

!e

url = b"https://i.redd.it/2qy7unjo2j331.png"

url_cleaned = str(url, "utf-8")
print(url_cleaned)

[halcyon plank]

@placid flicker :white_check_mark: Your eval job has completed with return code 0.

https://i.redd.it/2qy7unjo2j331.png

[placid flicker]

@fiery cosmos

[fiery cosmos]

Thanks man!

[placid flicker]

A string with an b in front of it is a bytestring

[fiery cosmos]

right

[fiery cosmos]

I got it

and BTW I have another question

the code above runs successfully, but some memes have a question mark (?) which prevents then from being saved, so how do I strip the question mark from the name?

[last fulcrum]

For a doubly linked list with only one node, would the next value point to None and the previous value also point to None?

[knotty magnet]

yes

[last fulcrum]

Why? Should the previous value not point back to the head? Like a circularly linked list.

[vocal gorge]

so is it a circularly linked list or not?

for a circularly double-linked list, you'd just have one node with both next and prev fields pointing to itself

[last fulcrum]

It is a doubly linked list.

No circular in it.

[vocal gorge]

Well, then no, the first node would have a None prev and the last one would have a None next. If there's only one node, those two are the same node.

[last fulcrum]

Ok. Thank you. Just wanted to be sure because I saw someone say otherwise.

[analog siren]

with regex how can you do 1 or 3 of an element and not {1,3} which is 1, 2, 3

[vocal gorge]

Not a question for #algos-and-data-structs, really, but [13] oh, realised what you're asking

(?:.{1})|(?:.{3}) would be one way, but not sure if it's the right one

[analog siren]

that would make it so that you would 2 right?

[vocal gorge]

this matches either 1 of . or 3 of ., yeah

[analog siren]

it works perfectly once i switched the 3 and 1 around thanks

[sharp gull]

Traceback (most recent call last):

pandas_datareader._utils.RemoteDataError: Unable to read URL: https://finance.yahoo.com/quote/TSLA/history?period1=1356993000&period2=1577917799&interval=1d&frequency=1d&filter=history
Response Text:
b'<!DOCTYPE html>\n <html lang="en-us"><head>\n <meta http-equiv="content-type" content="text/html; charset=UTF-8">\n <meta charset="utf-8">\n <title>Yahoo</title>\n <meta name="viewport" content="width=device-width,initial-scale=1,minimal-ui">\n <meta http-equiv="X-UA-Compatible" content="IE=edge,chrome=1">\n <style>\n html {\n height: 100%;\n }\n body {\n background: #fafafc url(https://s.yimg.com/nn/img/sad-panda-201402200631.png) 50% 50%;\n background-size: cover;\n height: 100%;\n text-align: center;\n font: 300 18px "helvetica neue", helvetica, verdana, tahoma, arial, sans-serif;\n }\n table {\n height: 100%;\n width: 100%;\n table-layout: fixed;\n border-collapse: collapse;\n border-spacing: 0;\n border: none;\n }\n h1 {\n font-size: 42px;\n font-weight: 400;\n color: #400090;\n }\n p {\n color: #1A1A1A;\n }\n #message-1 {\n font-weight: bold;\n margin: 0;\n }\n #message-2 {\n display: inline-block;\n *display: inline;\n zoom: 1;\n max-width: 17em;\n _width: 17em;\n }\n </style>\n <script>\n document.write('<img src="//geo.yahoo.com/b?s=1197757129&t='+new Date().getTime()+'&src=aws&err_url='+encodeURIComponent(document.URL)+'&err=%<pssc>&test='+encodeURIComponent('%<{Bucket}cqh[:200]>')+'" width="0px" height="0px"/>');var beacon = new Image();beacon.src="//bcn.fp.yahoo.com/p?s=1197757129&t="+new Date().getTime()+"&src=aws&err_url="+encodeURIComponent(document.URL)+"&err=%<pssc>&test="+encodeURIComponent('%<{Bucket}cqh[:200]>');\n
</script>\n </head>\n <body>\n <!-- status code : 404 -->\n <!-- Not Found on Server -->\n <table>\n <tbody><tr>\n <td>\n <img src="https://s.yimg.com/rz/p/yahoo_frontpage_en-US_s_f_p_205x58_frontpage.png" alt="Yahoo
Logo">\n <h1 style="margin-top:20px;">Will be right back...</h1>\n <p id="message-1">Thank you for your patience.</p>\n <p id="message-2">Our engineers are working quickly to resolve the issue.</p>\n </td>\n </tr>\n </tbody></table>\n </body></html>'

i am getting this error

pls help

[last fulcrum]

Why are you posting this here? Can't you read the banner or can't you see #❓｜how-to-get-help ?
Your question should not be answered because you can't follow even basic instructions.

[dapper sapphire]

So there isnt a way to make index 0 again with a for loop, we would have to use a while loop for that?

!e

for i in range(10):
    if i == 4:
        i = 0
    print(i)

[halcyon plank]

@dapper sapphire :white_check_mark: Your eval job has completed with return code 0.

001 | 0
002 | 1
003 | 2
004 | 3
005 | 0
006 | 5
007 | 6
008 | 7
009 | 8
010 | 9

[grizzled schooner]

maybe you could use itertools.cycle?

[knotty magnet]

if it's more complex, there's no way to reset it like you said, you'd have to use a while loop

although maybe you could do something like this

sub = 0
for x in range(20):
  if x == 10:
    sub += x
  x -= sub
  # do loop things
```but that seems annoying

[fervent saddle]

x = iter(range(10))
cycles = 0
for n in x:
    if n == 4:
        cycles += 1
        if cycles == 5:
            break
        x.__setstate__(0)
    print(n)
print("done")```
There’s this, but iirc it’s meant for pickling, so it’s probably not a good idea to do it even if it happens to work right now.

[knotty magnet]

interesting, how does __setstate__ work?

[fervent saddle]

It sets the state of something in the iterator, I guess

I’m not really sure what it does internally. Someone showed it to me when I was asking the same question

[knotty magnet]

!e

x = iter(range(5))
cycles = 0
for n in x:
    if n == 4:
        cycles += 1
        if cycles == 2:
            break
        x.__setstate__(0)
    print(n)
print("done")

[halcyon plank]

@knotty magnet :white_check_mark: Your eval job has completed with return code 0.

001 | 0
002 | 1
003 | 2
004 | 3
005 | 4
006 | 0
007 | 1
008 | 2
009 | 3
010 | done

[knotty magnet]

interesting

[keen hearth]

__setstate__ has something to do with pickling, but I've never learned what.

[runic estuary]

Is the black height of the RBT above 2? and the tree height 4? I understand that the height is the longest path from root

[dapper sapphire]

Thanks I think using for loop to set index to 0 would end up becoming more work.

Maybe while loop is a simple clean approach. And __setstate__ is interesting

[fervent saddle]

It’s interesting but it’s probably better to not use it since it wasn’t made to be used like that

[sage coral]

I'm gonna make a snake AI with genetic algorithm (or something like it), and I'm trying to plan out the nn. Ideally I just have binary values for inputs, but I'm struggling to figure out how I can fit data about the snake's body in a binary value. Any suggestions on how to do this?

[last fulcrum]

Apart from BST, what other trees are there and should I learn them? (Are they as important as BST?)

[knotty magnet]

there are a few more specialized trees for searching that you'll only probably see in competitive programming, like segment trees

you could also look at the self balancing BSTs, like red-black or AVL

[vocal gorge]

I've seen (https://www.youtube.com/watch?v=v7EJX38gG-E) this done by feeding the NN the data about the cells in a square around the snake's head, and also stuff like the position of the target.

[austere sparrow]

Database engines use B-trees, and HAMT is also a cool data structure (it's an immutable hashmap with efficient operations)

but I don't know anything in depth about those