Python Generators Explained

Welcome back to another KZbin video! In this video, I will be talking about generators in Python. Generators are similar to iterators but a more advanced concept in the Python language. However, they are not too complicated and anyone should be able to follow along.
💻 Thanks to HiCounselor for sponsoring this video! Apply to their next cohort to land your dream job in tech here: hicounselor.com/ HiCounsoler not only provides great learning resources but also has great interview prep material!
⭐️ Timestamps ⭐️
00:00 | Overview
01:43 | Generators vs Iterators
02:40 | Iterators Explained
08:33 | next()
13:58 | iter()
15:24 | Creating Legacy Iterators
19:22 | Creating Generators
23:15 | Generators Use Case
26:58 | Generator Comprehensions
◼️◼️◼️◼️◼️◼️◼️◼️◼️◼️◼️◼️◼️◼️
💰 Courses & Merch 💰
💻 The Fundamentals of Programming w/ Python: tech-with-tim.teachable.com/p...
👕 Merchandise: teespring.com/stores/tech-wit...
🔗 Social Medias 🔗
📸 Instagram: / tech_with_tim
📱 Twitter: / techwithtimm
⭐ Discord: / discord
📝 LinkedIn: / tim-ruscica-82631b179
🌎 Website: techwithtim.net
📂 GitHub: github.com/techwithtim
🔊 Podcast: anchor.fm/tech-with-tim
🎬 My KZbin Gear 🎬
🎥 Main Camera (EOS Canon 90D): amzn.to/3cY23y9
🎥 Secondary Camera (Panasonic Lumix G7): amzn.to/3fl2iEV
📹 Main Lens (EFS 24mm f/2.8): amzn.to/2Yuol5r
🕹 Tripod: amzn.to/3hpSprv
🎤 Main Microphone (Rode NT1): amzn.to/2HrZxXc
🎤 Secondary Microphone (Synco Wireless Lapel System): amzn.to/3e07Swl
🎤 Third Microphone (Rode NTG4+): amzn.to/3oi0v8Z
☀️ Lights: amzn.to/2ApeiXr
⌨ Keyboard (Daskeyboard 4Q): amzn.to/2YpN5vm
🖱 Mouse (Logitech MX Master): amzn.to/2HsmRDN
📸 Webcam (Logitech 1080p Pro): amzn.to/2B2IXcQ
📢 Speaker (Beats Pill): amzn.to/2XYc5ef
🎧 Headphones (Bose Quiet Comfort 35): amzn.to/2MWbl3e
🌞 Lamp (BenQ E-reading Lamp): amzn.to/3e0UCr8
🌞 Secondary Lamp (BenQ Screenbar Plus): amzn.to/30Dtafi
💻 Monitor (BenQ EX2780Q): amzn.to/2HsmUPZ
💻 Monitor (LG Ultrawide 34WN750): amzn.to/3dSD7tS
🎙 Mic Boom Arm (Rode PSA 1): amzn.to/30EZw9m
🎚 Audio Interface (Focusrite Scarlet 4i4): amzn.to/2TjXsih
💸 Donations 💸
💵 One-Time Donations: www.paypal.com/donate?hosted_...
💰 Patreon: / techwithtim
◼️◼️◼️◼️◼️◼️◼️◼️◼️◼️◼️◼️◼️◼️
⭐️ Tags ⭐️
- Tech With Tim
- Generators
- Generators Explained
- Iterators
- Python
- HiCounselor
⭐️ Hashtags ⭐️
#TechWithTim #Generators1

[@goboy6882]

I started programming 60 years ago. Thank you Tim for teaching an old dog new tricks.

[@Ian-bb7vv]

wow, life long learning!!!

[@piadas804]

Wow

[@JasonJorgensonPhD]

What was it you were learning 60 years ago? Was it COBOL? Fortran? I started programming in 1980/81 but it was BASIC by then, and I graduated to COBOL and Fortran, eventually settled on C++ as my go-to language.

[@me-zu3gm]

Lies!🥱🥱🥱

[@thelavagod]

​@@me-zu3gm how u know😂

[@williamselinder9138]

I very rarely leave comments on tutorials, but this was so incredibly well and robustly explained that I seriously need to leave my comment of appreciation, it was so clear to understand!

[@gustavocinci4044]

Thanks Tim. You have become the go-to guy when I learn stuff and get stuck. your explanations are very practical. as a new programmer who's attempting to beef up my resume, I have been teaching myself python. you have been a great resource, keep it up. THANK YOU!

[@Factory400]

It is nice and refreshing to see a presentation of the back-end logic to the magic functions we have in Python. As much as I love the built-in functions - I really love knowing how they do what they do.

[@tyrelllaszlo6024]

Great video and explanation! Was going to look for “how to process large amounts of data”, but clicked on this instead. Turned out to be the core of the answer👍

[@harshraj22_]

could have added example of "yield from" as well. A great coding exercise would be, print inorder traversal of binary tree using generators.

[@TechWithTim]

Great idea!

[@xx-jk1iq]

ok please explain what "inorder traversal of binary tree" is

[@gregfrost8399]

@@xx-jk1iq Traversing a tree is just the order in which you access it's nodes. In-order traversal of a binary tree means you visit the left branch, then the current node, and then the right branch. Basically put it visits the nodes in ascending order.

[@xx-jk1iq]

@@gregfrost8399 but from what I thought, a binary tree isn't a linear sequence. So how would that work

[@gregfrost8399]

@@xx-jk1iq Say we have the following tree: A / \ B C / \ D E In order traversal does the following: 1. searches the left branch until it finds a leaf node 2. visits (e.g. prints) the current node 3. searches the right branch until it finds a leaf node Once it has completed the 3 steps it will return upwards to it's parent node The result = (D, B, E, A, C)

[@vinni113]

i watched this video just to learn something new about python, turns out the generator is really applicable in my code and has helped memory usage heaps. Thanks!

[@AleVZero]

This channel is just so incredible! i can't tell how much i learned from this channel, thank you

[@NabeelButt]

Good thing about this video is that it walks you through other concepts too such as class and dunder methods which is kind of a good refresher to what I have studied earlier. You've earned a subscriber Tim!

[@michapodlaszuk9025]

good stuff Tim, there's never enough of knowledge about python

[@LNMLucasMasiero]

Thanks you. This was really interesting. It's crazy how the most someone knows about how something works, then for intuition you can understand other stuff or libraries created from others. One of the usefully methods I use as a fast document of a library is "dir"

[@TheRealKitWalker]

Wow! This was so refreshing and highly insightful. Thanks so much Tim for this really helpful video on generators. It helped me a lot understand this better!

[@CaratacusAD]

This was an awesome video Tim. I'm learning about Deep Learning and generators have been particularly useful for chunking up huge datasets for feature extraction as it would be impossible to read them all into memory on my laptop. Now I understand how and why this works as a Python newbie. The explanation on Iterators was great too. I love the nuts-and-bolts detail on what’s going on under the hood. Many thanks.

[@fredericoamigo]

Great video! Best explanation I’ve seen so far on this topic!

[@saishyam8583]

I would love to see a video on how you record and edit your videos Man This is some real good quality content!!

[@Mr_Spock512]

Indeed it is, I've learned a ton of stuff from him about Python.

[@STFU665]

Really appreciate your videos. Thank's for sharing your knowledge. I'm just at the start with Python deep-diving, but already it keeps getting me staying awake at night. So many possibilities :D Keep on the good work

[@bipolarchemist]

For a deeper dive into generators, you can go into how you can use .send to update process parameters that may not have been relevant when initially creating the object. This was an area that originally stumped me when I was looking at some more advanced examples of generators as the function took in one set of parameters, but the subsequent calls with yield accepted different parameters. A very neat trick and one of the reasons why I find myself growing more enamored with the language.

[@lepidoptera9337]

One can update objects at any time with a call to an object method just like one can update the value in a loop variable with a simple assignment. Writing a loop and an assignment makes the code readable for absolutely everybody with even the slightest amount of knowledge about procedural programming languages. Using send() will make 99% of programmers scratch their heads (including the ones who know what it does) and ask what kind of idiot will use such a construct. Guess who they mean by "idiot". They mean you. And that, by the way, includes yourself. You will call yourself an idiot for using this construct the first time you have to read your own code six months later. Simple is beautiful. Keep it simple.

[@enzovalendino606]

Hi Tim, you keep providing us useful contents: I'm glad for everything you've been doing! I'm truly interested in CS, and curious about how much of these information you have learned at university or on your own. Could you please answer me?

[@ChromaDotNova]

I appreciate the video, Tim! I have a much better understanding of Python generators now.

[@ddbarenco]

Amazing explanation! Learn a lot from it. Thank you soo much!!

[@MissingTricks]

This video provides a clear image of how it works. Thank you.

[@anirbanc88]

i love your videos man, and your clarity of explanations, carry on!

[@abdullahmohammed6115]

Love your vids! Great as always!

[@gustavojuantorena]

This video has so much value! Thank you

[@rodrigo2112-]

You're a great teacher, congrats!

[@manishsingh900]

It really is a great video. Thank you!!!!. One of the best generators video generated🙂.

[@network_noob]

Now i can finally understand generators! Thanks Tim!!

[@vammy55]

Great stuff mate,thank you and cheers!

[@omarelsebaey9589]

the example for the generator is perfect , thanks!!!

[@onlyforscience8255]

I learned new things which don't know anything about that. Thanks to make such a nice video for us. ☺️☺️👏👏

[@elitearmedforce]

Wow, really well explained, good that stumbled upon this. If u can do the same with other python(decorators...) and c++ concepts like this, pls. Subbed and liked,

[@NearLWatson]

Great job breaking it down 👍🏼

[@eddywang5035]

Useful tutorial for me. Thanks.

[@ChandanSingh-bh7wu]

******BEST GENERATOR EXPLAINATION********** read 3 books + 4 online tutorials, still confused about Generators******* Finally, got it ............!!!! Awesome Brother !!!!!

[@amalkrishnas1696]

Thanks for the video, well explained

[@parvishrao6414]

Tons thanks, only video which clearly said what is iterator and generator - CRYSTAL CLEAR.

[@davidring5071]

Amazing video Tim. Thank you

[@berkaybb5733]

Could you make a video about Web scraping by using requests and beautifulsoup modules?And maybe some decorators would be great : )

[@danielmarx3106]

You can use ctrl + d to copy a line to the next line without copy pasting in pycharm, not sure if something like that in sublime.

[@Barejs]

awesome content, keep up with these videos Tim

[@haneulkim4902]

Thanks Tim. One question. So creating generator don't even take up disk space? Say I have a for-loop that downloads a file, yield data, then delete the file where end result is a generator. This won't take up disk space and only memory upon calling __next__()?

[@estebangarro7254]

Excellent Video, thanks!

[@Mdroudian]

Great video. You have a good format. Don't change it =)

[@boooringlearning]

thats brilliant video!

[@johnnytoobad7785]

Looking forward to "Learning Cobol with Tim" course !

[@yoyokoko5153]

Thanks for the good stuff

[@JaBoss397]

Thank you !!

[@servantofourlordjesuschris6456]

Thanks!

[@korkunctheterrible4302]

09:46 and on: Hi I'm a total noob and I don't get why for loop is able to pick the iteration up from where the last next call left it. why does the for loop know what happens before it? Is where next stops also stored and the loop preprogrammed to check that address? I'm sorry if I've misused words, I just don't know how else to ask about it. Any tip would be much appreciated. Also thanks to the maker of this video.

[@alexandruteodor3585]

Thank you!

[@aryankathawale9269]

lets goo new tech with tim

[@nicholas_obert]

Could you make a video on C extensions for Python? It took a while for me to learn as there aren't many resoruces out there and the documentation is partially outdated, but the speed increase using such extensions is exponential with regards to input size. I think a tutorial on this topic would prove highly useful to newcomers. I've already written an article about them on Medium, but I think you, Tim, can do a much better job at explaining them.

[@hossumquat]

Yes, this would be nice. Does Tim know C though? If so, I've love to learn how to integrate C and Python

[@nicholas_obert]

@@hossumquat he has made a series about C++, plus C is very easy at a beginner level.

[@hossumquat]

@@nicholas_obert Ah, ok, didn't know that. I'd guess he already knows this then and could teach it. Awesome!

[@antmitchell9930]

You should put the 'Pause 1' before the 'yield 1', I suspect, otherwise the pause text appears on the next call, not the one it refers to. Just learning Python - this is great.

[@whatdoiputhere5089]

yeah, he should've done that so everyone can know when is the 'Pause 1' getting printed

[@user-zb8tq5pr4x]

but the execution pauses after the yield, so pause 1 before makes no sense.

[@Kema109]

@14:40, if I would print(next(iter(x))) it would give me 1 thats true, but if I would do that again, I would get another 1 and so on. So it does not go to the next Element. That was kind of confusing to me. So I guess, you have to make the variable "x" an iter first, and then you can use the next(x) function to "jump" to the next element. Or did I not understand that right?

[@bradyb2415]

First time i understood what you are saying....☺

[@thaimeuu]

thank you, Tim

[@UgochukwuEzeani-ut6vd]

Awesome 🎉

[@Tussu17]

Thank you genius!

[@barbaraulitsky8292]

This is a super tutorial! Super helpful and clear explanation! Tim, you are awesome! Thank you so much!!!

[@robinferizi9073]

I like how the first generator example is literally obscolete because you used for i in range: to say for i in range:

[@grantpeterson2524]

How does a for loop work on an object that is already an iterator, like map() function you showed? Does Python implicitly know if the .__iter__() method needs to be called, or if it doesn't?

[@alexramirez5104]

Damn I love this channel

[@wubangzheng]

hi i wonder any python seniors could answer my question , i just started learning python and the things i learned included all the basics (i assumed) (exp:file handling,oop,the basic commands,if,loop,tuple,dict.....).Besides,i also learned some algorithms like linked list,binary tree,stack&push.For now ,what should i do in the next step?

[@wubangzheng]

i did some python projects as well like creating a simple alien invasion game

[@Plepple]

I'm not a senior, but I think working on projects is the best way. Do webscraping with selenium, API calls with requests, build a small website with django, use pandas to work with relational data, then improve all of the IO bound tasks with asyncio

[@elpython3471]

Hiyo tim! Good tutorial as always! Could you think about making more c++ content, and teach us about c++ libraries?

[@jphvnet]

If for loop stops by the exception, it's not clear where gen() raise the exception. Or there is a different logic inside for() .

[@rekhasurya3536]

Love your Vids Tim! keep up the good work(hope u heart this!)

[@lukaszno8988]

Hi Tim can you explain interface in python?

[@SkyFly19853]

Thanks for the explanation. Btw, are you interested in making a video about Cython ( C amd C++ extension on Python ) ? Thank you in advance!

[@jonathan3488]

Yep. I have tried using cython, but encountered IDE errors and performence issues.

[@SkyFly19853]

@@jonathan3488 I see... Even so, I think it has the potential.

[@mufasaruleroftheanimalking1026]

It’s better to say next() consumes the list. If you printed the list y after invoking next() four times the console showed you the iterated elements don’t exist anymore in y.

[@innocentharelimana4449]

we, we had teachers who didn't know these stuff😂😂😂, and graduated without this knowledge😊😊😊

[@robinbreed2439]

super good

[@terryr5713]

Tim what can I do with a computer science degree except programming!? 👍👉💪

[@amrsalaheldinabdallahhammo663]

Thanks genius :)

[@georgianabhishek3561]

Please make a pygame video explaining sprites

[@mohamedelzend2337]

what is that code editor tim using ?plzzzzzzz

[@UgochukwuEzeani-ut6vd]

Awesome

[@pabloshi4863]

Could anyone help me of how to show the [Finished in X ms] in the console?

[@ncmathsadist]

the object returned by open("foo.txt", "r") has an iterator that walks through the file a line at a time.

[@lepidoptera9337]

That's too easy for the man. He needs to make things complicated. ;-)

[@mgmartin51]

when i put the code at 7:29 into IDLE, (for i in y: print(i)), nothing happened. You got the list output. Why is that?

[@youssefelamrani7905]

can you do a video about testing Tim, i heard spaceX is using Python for testing

[@xx-jk1iq]

what do you mean by testing?

[@64imma]

What text editor are you using?

[@entity5678]

sublime text

[@Jschmuck8987]

After every explanation: “HOPEFULLY that KINDA makes sense”. Yes bro it makes sense ffs.

[@neerajvaradi8505]

How your code showing the run time..?

[@tcgvsocg1458]

Can you create a "screen recorder"app on python please

[@stele339]

Хвала!

[@belowasmelashgebremariam]

Nice

[@belowasmelashgebremariam]

Asme Tchaikovsky climb to uucp

[@belowasmelashgebremariam]

Razed St Ed Sry egg re zed zarathustra ink

[@HarryBGamer2570]

what's that font (name?)

[@Aidanbaro]

Hi Tim

[@laponiec]

Why would we want to implement our won iterators? Edit: So, we don't need to create iterators, because they have been replaced with generators?

[@ilyasseidkhebbach3812]

good

[@avigupta6064]

Tell me your theme and color of vs code pleaseeee🥺

[@error2839]

He’s not using vs

[@avigupta6064]

@error i know, i am just talking in general like that spotify api tutorial he did.

[@memk]

Ahh, you also forgot the send method of the generator. That might be it's own video tho as it's the base of coroutine

[@comedyclub333]

I think your comparison between the generator and "legacy" syntax is a bit unclear. The class based iterator seems way more inconvenient than it acutally is since you used "range" in the generator but implemented a range-like behavior from scratch in your class. When returning a range object in the _ _ iter _ _ method the class would not be that more bloated and actually useful depending on your use case. I think it could've been made more clear that a generator relies on existing iterators while the iterator class, well, does not since it is made to provide its own iterator.

[@gagansai8880]

generators need not rely on existing iterators... def upto(x): current = 0 while current

[@comedyclub333]

@@gagansai8880 Yeah, you are right. Totally forgot about the fact that for loops and while loops are interchangable. I wrote bullshit about generators relying on existing iterators. My criticism still remains that the way custom iterator classes and generators are compared is not really fair.

[@korkunctheterrible4302]

@@comedyclub333 I don't understand. Wasn't it said or implied in the vid somewhere that yield uses next() as a built in? (I am not a geek or anything, I truly don't understand and I am a 5 year old when it comes to this)

[@shawnbeans7389]

tim finally decided to change from monokai

[@aryankathawale9269]

here before 8 comments

[@ScOrPiOnE905]

I'd like to make a note regarding the execution of generators, maybe somehow it'll be useful to someone. A generator will not re-iterate if it has already finished the range it was given in a variable for example. However, it will re-iterate if you call directly to the generator itself, not the variable you might have assigned it to. You can try this out with a range, a for loop and the print function (I've posted some code below to give an example) Code to try it out for yourselves below (I hope the formatting won't be broken): def generator_example(n: int): print(" Generator has started. ") start = 0 while start < n: print("Generator is returning {}".format(start)) yield start start += 1 range_example = generator_example(5) list_example = [] print(next(range_example)) # Exhausts 1 of the total of 5 values in the range_example variable. input("Press Enter to proceed.") for values in range_example: input() list_example.append(values) # Only 4 values remain to be appended. print() print("Generator has already finished and will not execute the for loop in the code in the same variable (range_example) below.") print("The print(next(range_example)) function will also exhaust one execution of the generator.") for i in range_example: list_example.append(i) print() print("Calling the function itself, however, will execute the for loop however many times you want.") print() for j in generator_example(5): list_example.append(j) for k in generator_example(5): # print(k) list_example.append(k)

[@korkunctheterrible4302]

So where is the information for the first next() yield stored, in range_example ? and if it is stored, why does that not take up memory or is it that it just takes smaller memory in comparison? when you call the generator_example function with the argument 5, does the function def block create 5 spots to fill for in the range_example first, and one gets filled by the first "next", and a mark is put in there? does the info for where the mark will go get overwritten in the same space of memory with each execution? lol sorry for my dreadful misuse of any term or anything and I hope you still understand my question (I actually asked about it before I bumped into your explanation, and YT doesn't let me copy any comment from other users, although I will try it out myself later when I'm able but there's a chance that I'll misinterpret somethings) Thank you for your comment btw.

[@korkunctheterrible4302]

Also what'd happen if there were no range_example to which to assign generator_example()? it would start from the start like at every call, like you said later in your post right? but I don't remember Tim using another alias like thing, still it picked up where next() left off (4 nexts and then for loop picks up the fifth, at 09:46)

[@malanbandara]

This guy is almost like Mark Z.

[@abcdxcxd8280]

The range function is actually xrange. Xrange was renamed to range and the old range function is gone. The xrange function stores the result in a xrange object, not an array, which is why the new range function is smaller than the old one. Also, you can just use “for x in [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]”

[@jammincoder]

Those 2 thumbs-downers had their computers upside down.

[@wongkingshun]

When u say hopefully that is clear, and i think nope, i stlll dont get it...