Видео

I think there's a bug in your multi_accumulate example. You'd want the first argument of itertools.accumulate to be iterator, not iterable, right? For non exhaustable iterables, like range or sequences, you will end up counting the first element twice. Not an issue for min or max, but you'll definitely see an issue if you use the running sum example. iterator will have that initial value removed, though, so using that instead should solve the problem.

Python 3.14 will be the Pi update.

The combinatorics things are useful to me, but my use case is somewhat mathematical. In physics, you may want to solve an N body problem. That generally requires calculating N² interaction terms, for each pair of bodies. If eagerly calculated, this would correspond to a N×N matrix. However, there is usually some redundancy. First of all, self-interaction terms often can be omitted. So you can skip computing the diagonals. Also, the pairwise interactions are often symmetric. For example, the Earth pulls the moon with the same force as the moon pulls the Earth. So to avoid repeated computations, you can compute just the upper or lower triangles of the matrix. If you want to do the calculation lazily, you can just use itertools.combinations(bodies, 2), and it gives you exactly that triangle - very efficient. If you care about self-interactions, you can use itertools.permutations(bodies, 2), and the triangle including the diagonal will be iterated over. If the interaction is not symmetric, you do the most general case of computing everything in a nested for loop, or like you showed, itertools.product(bodies, bodies).

On types or dictionary question you asked about. In c# we have enums that you r able to cast the value or str. And values could change depending on what the enum is used for. Then you want to use kwargs because you don't know how the structure changed. Like in reflection scenarios. Thanks great update And you explanation about square and override is a nice explanation of polytechnic methods if they were wondering how polymorphism relate back to real life scenarios instead of foo

What is the calling syntax with the colon in it?? I've never seen this before in an argument at the call site except for a lambda. It's almost like type hinting or creating an inline dict without braces. Cheers Eg at video time 16:13 you have itertools.collections(data, r:2)

I was hoping you would do itertools. Please do functools too!

The problem in Calculus is the limit. Words like approaching do not prove something will ever equal. "As close as desired" might mean a number "next to". Then the Archimedean property takes over. It is difficult, using base 10 digits, to show that there are numbers between ".99..." and 1. Each number base has this problem. (".(b-1)(b-1)..." approaches 1). You could then find many numbers between ".99..." and 1, one in each number base higher than 10. (1/16^n, 1/20^n or 1/60^n are closer to 0 than 1/10^n) [What about "at infinity"? What is infinity? Infinity is incomplete, inconsistent and indefinite]

Some of them are quite useful, but as a developer, I need to break my habits first to remember the use case and instead of loops go to the itertools.

In the thumbnail you’re missing a )

I have used itertools.combinations a lot for fuzzy matching of names. If I have a list of cities input by a human along with a lookup dictionary, I can fuzzy-match all the combinations and then lazily pick the top match for each city.

If you need a list with a given length (eg for some sort of buffer/storage), instead of list(itertools.repeat("X", 4)) You could also use ["X"] * 4 Not sure if there are any notable performance differences, but you need one less import, and its shorter

I've created a typed_stream library to use the most important lazy functions from itertools as methods on a Stream class. That's far more readable imho than the functions

Ive found those functions useful while doing simple grid searching where you test combinations of hyperparameters to tune ml models.

The combinatoric ones are useful for Sudoku variant helper tools

I think I used to use minmax from more_itertools to get both in a single pass. combinations is one of the most useful itertools for graphs, and pairwise comparisons, albeit I tend to use numpy hacks more because of performance issues.

I would have used those combinatorics functions (and even zip longest) for advent of code challenges. Probably not any real world code

I use the combinatorial ones a lot, often for testing things. I use all four of them. I have 5 versions of networking device firmware, and I want to make sure they're all compatible with each other, so I iterate over all the pairs of combinations_with_replacement. I have 10 devices networked together, and I want to test throughput for all possible paths. That's combinations, or permutations if I want to test both directions. I have 5 versions of firmware and 3 models of devices, and I want to make sure all firmwares work on all devices, so I iterate over the product of firmwares and models

When she iterate on my object till I throw

If you want more itertools, I recommend the more-itertools library.

filterfalse makes sense when you have a function defined elsewhere that you are using (e.g. filter(my_func, it)). To reverse the conditional would require writing a new function or adding a lambda (e.g. filterfalse(lambda x: not my_func(x), it)), which is ugly.

filterfalse is necessary b/c sometimes your predicate is "bool", and lambda x: not x is blech.

I would love a comprehensive video that goes over the more_itertools library. It has become a massive go-to for me

2:25 a "real" dot product returns this: (abs(u + v)**2 - abs(u - v)**2) / 4

been waiting on this one

5:05 There's now batched and 11:24 pairwise in the standard itertools? That's fantastic, i had to write these so many times🎉 14:40 I feel i'm using zip_longest less often than zip, but still regularily

I used zip_longest recently. I needed to vertically display two lists side by side in a GUI, and there was actually very little chance they'd be the same length. zip_longest with fill="", then ' '.join

"iterable" doesn't sound like a word anymore

Itertools group by is reminiscent of how old fashion Hadoop works with parallel key, value operations

The combinatorics functions are pretty damn useful IMO. Sure, it might "just" be for maths stuff, but the range of computational maths problems that involves them is vast! It's essentially the answer to the question "what are all the ways to put these inputs together?" which is super generic.

Do you use the Rye package manager for Python? I found it very awesome to use. Much better than pip (uses uv under the hood), handles .venv gracefully, and managing dependencies is not a pain anymore. If you haven't checked it out, I strongly recommend it! Love your videos so much

These are mega useful for Leetcode and CodeWars type problems, second only to the king collections.Counter

at 4:41 I'm assuming "min" and "max" are passed to "accumulate" and not the "list" constructor

for the dot product, using my transpiler: dot = 𝚲𝔸⍉ᴍΠ´Σ (which ends up as: (lambda*_1d538,**_1d542:(_03a3(_1d0d(_2349(_1d538),_03a0),NULL)))) lol

11:45 For anyone wanting to use the recipes: They are all implemented in more-itertools so you can import that and access those recipes as functions directly

Nice! But I want MORE! I SAID MORE!!! As in more-itertools, of course.☺

How come all the comments are porn bots? And who's this Scott they're all referring to?

i've been writing `batched` myself or using `chunked` from more_itertools for much too long...

I am working on a graphics applicaiton, and while I only need to render a frame 165 times per second, there is five render passes. So each pass can at most be like 1ms - so I need some better profiling tools

when i use multi processes and build an .exe file from the python file using pyinstaller, I get an infinite loop of proceses and eventually my PC crashes. Do you know why this is? Ive tried Process.pool, ProcessPoolExecutor, ect.. When i use threads instead of processes, the exe works fine.

Pydantic is better in my opinion

0.99999999999999999...... = 1 - 1/infinity

ANYWAY good Job and excellent content. liked and Subscribed ❤

Ok, good and 💪 strong strart. Then 1/10^n smaller than 1/n and smaller than epsilon. hence proof!?😮

r"^\s+?(?!#)(import\s+\w+|from\s+\w+\s+import\s+\w+)" I get what you're saying, but finding local imports with regex might not have been the best example to use, since that one would actually be pretty easy. You would just need to match "(from ...) import ...", preceded by at least one whitespace character. EZPZ

love the full and simple explanaition. i use staticmethods when appling MVC structure to my api's from the user_model UserModel class i need only sertain methods from the UserModel class. so i can call whathever query method i need in my controllers.

Howly cow, 3h of debugging because I wrote "handler" in my json instead of "handlers"

I am not sure why it's a nooby mistake to write your own code when there's an existing algorithm. Sure, it can show inexperience in not knowing that there is an existing algorithm. But right before seeing this video I was watching a video reviewing a Ray Tracing engine. That engine used accumulate. The person doing the review was confused why the debugger showed that ~40% of the compute time of this extremely slow Ray Tracer was happening inside of accumulate, when the math itself was under 1%. So he replaced the function with a for loop and brought the render time from over 7 minutes, down to ~20 seconds.

1. wrong they are as readable or better readable?????

x = 10 10x = 100 10x + 9 = x 9x = -9 x = -1 any number = -1

1/ε → + ∞ Claiming n > 1/ε is saying we have number larger than infinity. Is the reasoning logical? You also mentioned " x approaches 1" , but not showing any proof " x equal one".