Heap - Max Heapify

[@Alicia0Cramel]

Thank u god all mighty i couldn't find a SINGLE video to explain this concept quickly and concisely omg!!

[@dariuszd6207]

It was hard to find simple and clear explanation of this problem. Thank you.

[@kimberlyjohnson2580]

This made so much sense! Thank you for explaining it so clearly!

[@MarsTheProgrammer]

Thanks buddy, you explained it and made it clearer than my professor who has 2 Phds...

[@heatxtm]

well, having 2 Phds doesn't mean anything one thing is to know something (2 Phds says that) and other thing is to teach or explain

[@dghthh2163]

A phd is not a job training for teachers.

[@kadentaylor8503]

@@dghthh2163 LOUDER FOR THE FOLKS IN THE BACK!

[@AlexChambersXYZ]

I call you "Zeus" god of the algorithms ruler of data structures thanks man.

[@HaloElite]

I love you so much for creating this video! Saved my life!

[@bobbymcgrath8025]

Very well explained video and the visual representation was very well illustrated and broken down into a step-by-step formula that made it even easier to understand, thank you sir! :)

[@Shobu]

much love man. my teacher is a 70 years old who it does not care about students. Thanks a lot !

[@botqqq]

Perfect and simple explanation. Thank you!

[@KnownUser139]

I think this guy was sick too..Hats off to you it was a great video

[@Burak-pl1jl]

Brief and understandable explanation! Thank you. :)

[@MorleyDev]

Is there a particular reason you traverse the tree starting with the right subtree of 16 first on your tree at 2:00? I thought tree traversals (pre-order, post-order, in-order) put left before right in all cases?

[@Z00CE]

Nathan Morley, good question. A heap is not like a binary search tree, where pre/post/in-order traversals are useful. When we begin to heapify, we have to start from the very last subtree which starts at the floor(N/2) "root". Remember that when we build a heap, we insert from top to bottom, left to right, that way the tree stays nearly complete, so if we didn't have the last two nodes, we would have started from that left subtree at the bottom. I hope this makes sense...if not, let me know and I can explain further. Thanks for watching!

[@ExploreWithNam]

Could you please explain how to compute time complexity for this?

[@MyVanir]

Great explanation, helped me a good bit.

[@DrakeLuce]

Basically a Khan Academy video. Great job :)

[@RafaelNascimento-qo1jp]

6 minutes of "visualization" is worth more than 6 hours of explanation

[@minkihairoil]

Thank you so much this FINALLY makes sense :))

[@devinj374]

fantastic video!

[@向前走五十五步]

Super clear. Thank you for the posting.

[@vijaydulam5886]

Awesome Sir ! clear This Topics

[@CloudStrife0896]

Simple as 1-2-3, good job on explaining another thing not done right in Greek universities.

[@Simoky99]

or in brazilian universities

[@teeheeee7807]

Thank u so much it helps a lot 💗

[@אלעדשגיב]

It was a perfect explanation , thank you!

[@jontran3194]

I dont get it. If this is big O(logN) Everytime you max heap something. You mess up the sub tree below. So you have to go back and max heap it again. Isn't it O(n^2)?

[@kendellferguson5021]

Who knew Ben Shapiro would be teaching me Data Structures lol Great Video btw

[@diana-hus]

Thank you! It was extremely helpful

[@saxophone12player]

Thanks man !

[@coyotulnegru]

thank you mister

[@chagantirbbsrao5930]

great work!! thanks

[@ericknicolasnunesdahora4860]

Thank you sir!

[@elyartaker1139]

wow thank you for this

[@minhhuynh2646]

Thank you man you are great

[@transam351]

How does the algorithm know when it needs to re-heapify subtrees?

[@Z00CE]

It always checks the subtree which it swapped with. If that subtree is a max heap (i.e. the child nodes of that subtree are less than the new parent) then it doesn't need to heapify that subtree, otherwise it will re-heapify it, and so on.

[@prof.dalmar6989]

Grt video thanks sir now i understand clearly thnx

[@yongkailiu1448]

Really helpful!!!

[@VersatileAnthem]

yeah i understood but how to make that understandable to computer

[@romansmirnov3351]

thanks man, that was good

[@batuhanbayr7613]

KZbin AWARDS GOES TO..

[@tusharpandey6584]

very nice explanation thanks

[@Simoky99]

I finally understood this crap. thanks

[@deepakrout6795]

Which drawing app are you using for lecture?

[@43_ayonroy4]

I have to submit an assignment tomorrow but I also have an exam on the same day. I need help

[@ashitpanda2905]

Ahh thanks for this video. But i must say it could have great if you had described with the algorithm description.

[@vikasyadav-og2cg]

sir if value is 23 in the place of 5 what should i do???, which value can be swapped???

[@lamyagawdat9203]

god bless you buddy

[@natsuvatsan6108]

What about min heap?

[@haisulii]

Simple to understand.

[@persianshawn92]

so when we go through each subtree, we always go from right to left?

[@jonasgrnbek7113]

What software is used to create this?

[@Aevatheone]

Awesome! :D

[@karolkarol1430]

Thank you a lot ;)

[@akashaggarwal3382]

grt video

[@JananiAnbarasan]

what tool/software do u use to write?

[@leadguitar53]

Isnt subtree 2 still unfinished at the end of the video?

[@darshfify]

thank you :)

[@elie3423]

not indian: check clear voice: check good video quality: check knows what he is doing: check Thanks bro for the perfect video

[@KyleBridenstine]

On the third sub-tree where both child nodes are greater than the parent, for some reason I always thought you start with the left child and swap those then you evaluate the right child and if the rights larger you swap those, I didn't know you just pick the one that's the largest :0

[@Z00CE]

If you took that approach (starting by swapping with the left child) you'd be adding quite a bit of extra, unnecessary computation. Heapify is recursive, so when you swap, you Heapify the subtree that you just swapped with...thus you'd have to come back up to the initial subtree after who knows how many Heapify's of that left subtree, then swap with the right child, and recursively Heapify the right subtree. (Sorry you might have to read that a couple of times...lots of "subtrees" and "Heapify's") I realize that it's difficult to visualize this with the example in the video, but if your Max-Heap was very large, you could see just how much extra work you cut out by simply swapping with the larger of the two. Hope that helps clear things up. Thanks for watching!

[@KyleBridenstine]

Yeah I thought about that as soon as I saw this video it definitely makes sense this way! Wish me luck on my final tomorrow :(

[@Z00CE]

Awesome! Good luck!

[@fog1257]

Thanks!

[@mohammadishaqnoori9728]

thank you Sir

[@Alex-ii3me]

Thanks!!!

[@cav3man82]

didn't you just go through the whole build heap function (and not just max-heapify on a particular node)?

[@Z00CE]

Yep

[@elliotgehin]

Cheers m'dears

[@abdulmalikjahar-al-buhairi9754]

HEEELP I UNDERSTAND THIS BUT I HAVE PSEUDO CODE THAT DOES THIS TOP DOWN! Why would you do that to me prof?! Isnt it stupid to go top down? You have to make redundant checks? Or doesnt it matter?

[@qwertek8413]

thanks ;)

[@Lukas-bb4sg]

Thanks

[@nishantingle1438]

Someone please give me iterative code for max heapify.

[@allenyin7630]

A binary heap is not nearly a complete binary tree. It is a complete binary tree.

[@MaleeshaHuththo]

Ben Shapiro before he was famous. jokes aside, this was very informative.

[@zilinli187]

save my life for only 6min

[@ORagnar]

So, you are pushing the maximum value to the top by this method.

[@Z00CE]

Yep, that's correct. And to be even more correct, every "parent" value is greater than (or equal to) it's "child" values.

[@pranupranav5563]

NIce

[@jortvincenti7831]

t'es un bro

[@Bbehemothh]

At 1.5x you sound like Ben Shapiro.

[@dinhomhm]

thank you sir