Name: 面接のためのPythonアルゴリズム (Python Algorithms for Interviews)
Uploaded: 2021-01-14T10:25:04.000Z
Duration: 3 h 47 min 8 s
Description: VoiceTubeの動画で発音を聞きながら英語表現を覚えよう！学べる英語：

程度の副詞

Syria's gonna be for analyzing not just algorithms, but also primarily pieces that are gonna be, you know, let's not just in computer science, but also for interviews.

様態の副詞

And then we're gonna be starting up to something called the Big O so really quickly we're gonna be doing this all in the Jupiter notebook, primarily because it's not that difficult enough that we have to use pi charm.

And it's also a little bit cleaner looking.

So I analyzed algorithms in the first place when algorithm don't get never spoke a word.

It's just procedure or formula for solving a problem.

Some of them are so useful that we've given them names like Merge sort of bubbles sort et cetera.

And then one of the important things were gonna come across you can see very quickly is how can we compare the algorithms, you know, which is better, and by better it's gonna be relative term to the task at hand.

But typically there's gonna be two avenues for analyzing if an algorithm is better than another and that's going to be either by speed, time of completion on, that's gonna go into its scalability.

and also memory, its size, allotment and its requirements and so forth.

So in this 1st 2nd Cody said, imagine keeping dysfunction.

This was just to finding a function called some of one.

It's taking one argument, and so then we have final Sum equal zero.

So we have an object Final sum ever, given the assignment of zero, it says for X and range and plus one.

Range is gonna be whatever define it for in terms of the ends and plus one equals final, some equals fun, awesome plus X And then we're returning the final sum.

So that actually loads into the Jupiter notebook.

And then I have hear someone calling five of the argument.

So what this did was with the object having the argument of five and the formula essentially the algorithm for this was for X and rain.

So for each number in range and plus one and was five, so in six you remember for range started zero.

So we had to add that one of wanted to really get five actual numbers.

So zero is not gonna have any implication for addition on the final sum of starting at zero.

So if we did five times four, we're gonna have 99 plus three is 12.

So, of course we got the answer to be 15th.

So let's see what this is all going towards anyway.

So, again, this is, uh, function that we've created to finding a sum of one.

And we got a result of 15 when Ann Nichols five.

I'm going to go down here and now for this one.

And let's assume that you gave me this function.

You called it some to also taking only one argument of end and you instead of doing a four low Penis returning and times and plus one divided by two.

I'm gonna shift into through that as well and same argument of five, and we also get 15 same exact function.

If you actually did this math by hand, of course, Pem dos is gonna hold place, so parentheses.

You're always gonna be first, so you would have had five plus one, which is going to be 66 times five is 30 35 by two is going to be 15.

We got the same answer with two very different functions that's essential.

So this is my function I created, and this is the one that you created, both getting the same result.

So function one is someone's gonna complain about that, right?

It's using a four loop to interpret it curative Lee at across our range of plus one.

So here's our four lope and we're just iterating over the range that was dictated by the input end.

Number two is simply just taking advantage of a formula.

Both of these are technically algorithms because remember the algorithm is this a procedure or formula for solving a problem?

So we're solving a problem, and here was something problem.

This is just more formula base, and this is an iterated four loop that we're running.

How do we determine whose function is better mine or yours?

And that's gonna come relative to what it is you're trying to determine.

So we have to objectively be able to compare them.

The two avenues I mentioned before that we could do the four is memory, which is space allocation within the computer or the time to run that much time it's gonna take to actually run that software.

So we're going to use built in magic magic commands in the note.

Expect that it was just Python and now it's multiple languages.

So first of this, his is a magic command using the percent side time it and I'm saying, I want you to time it for the sum of one function with an end of 100.

So I'm gonna shift into that and right now what it's doing, it's doing that for loop.

Best of 33.7 for microseconds, poor loop, and then I'm going to do the same magic time at some two with 100 z input.

And then d'oh, it's doing the computation Now.

Took a second there, depending on your computer is going to determine if it happens faster or slower.

So for the same end of 100 which we know that they're gonna give me the same answer.

And you can ignore this for right now that's just talking about cashing best of 344.

So and I put down here just for those who may not know Micro, which is up top here is 10 to the minus six.

So that means Nano is a smaller number by a factor of three.

So in the two different functions, we had function of someone of function of some too for the same end.

My algorithm, your algorithm, your algorithm was a function was much, much faster.

I was on microseconds for time of completion.

But now that's not necessarily the end Albil because that's the time of completion on my computer, your computer, the time of completion might be faster.

Granted, we're talking micro nano seconds.

That difference can get very large, and we have much larger inputs to deal with, and some systems will be faster or slower.

But at the end of the day, regardless, four lives here was gonna be slower, and that's what we try to avoid them, and especially machine and artificial intelligence.

So now that we have that, um, I have your smallest number is obviously faster, we know that.

So we can't rely on just time to run, because again, all computers, a different hardware is gonna be different somewhere fast and others.

We want this to be hardware independent, and that's one of the big takeaways because, as you know, hardware is constantly changing.

So how come we objectively determine if one function is faster or slower, or rather, is better or worse than that of the function?

If we can't use the time parameter per se because of the objective nature of different futures?

So the big O is a way that we can objectively compared the efficiency of those two algorithms.

So we're gonna get the number of assignments of each algorithm.

字幕リスト動画再生

面接のためのPythonアルゴリズム (Python Algorithms for Interviews)

essentially

significant

scenario

constant