8. Object Oriented Programming

MIT 6.0001 Introduction to Computer Science and Programming in Python, Fall 2016
View the complete course: ocw.mit.edu/6-0001F16
Instructor: Dr. Ana Bell
In this lecture, Dr. Bell introduces Object Oriented Programming and discusses its representation in Python.
License: Creative Commons BY-NC-SA
More information at ocw.mit.edu/terms
More courses at ocw.mit.edu

[@ayyoubm]

dude. the internet is crazy, I have been to the best universities in the whole world without leaving my room.

[@tainatcoding4494]

yes

[@anjalisingh-sx5ct]

Any resource for oops c++ please🙏

[@amans6504]

@@anjalisingh-sx5ct oops are same everywhere language doesn't matter

[@josuegisber8150]

Dude, you’re right

[@Jon_Zeku]

for real for real

[@leixun]

*My takeaways:* 1. Object 0:59 2. Object oriented programming (OOP) 2:05 3. Object (continued) 3:30 - objects are data abstractions that capture 1. what attributes can represent an object 2) how can we interact with an object 4. The advantages of OOP 7:33 5. Class 8:55 6. Self 15:35 and an example in 19:05 7. Method 20:55 8. __str__ method 27:22 9. Special operator 32:33

[@SulemanKhan-hm1yc]

ty bhai i was looking for this

[@rafiy7150]

we need more poeple like you

[@leixun]

@@rafiy7150 You are too kind!

[@SharronDenice]

Thank you!

[@leixun]

@@SharronDenice You are welcome!

[@Wander4P]

This explained Python objects so much better than any online tutorial I've read. Will definitely watch more of these lectures.

[@koonsickgreen6272]

As my token of gratitude Id like to give Ana another black shirt as a present to add to her already extensive collection of black shirts.

[@fauzanstster]

Thank you very much MIT, Dr Ana Bell and all of contributors and donors

[@user-fl7vs4ed6l]

0:01:32 Object 0:03:33 0:05:00 0:06:00 0:08:00 0:09:05 Creating and using your own type with classes 0:11:28 Define your own type ​0:13:01 What are attributes ​ 0:14:44​ Defining how to create an instance of a classes 0:16:00 0:18:49 Actually creating an instance of a classes 0:20:38 ​0:22:00 Define a method for the coordinate class 0:24:29 ​How to use a method 0:26:00 0:27:40​ 0:30:52 0:32:36 Special operators 0:33:59 Class "Fraction"​ 0:36:00 ​0:38:08​ Class "Fraction"​ used to compute 0:39:44

[@akshayram6868]

Thank you so very much dr bell! After months of searches and reading books, I’m really happy to have finally understood oop in python ! You are amazing ! Thank you so very much ! And thank you MIT !

[@yukeyang5735]

1. Objects comprise of 2 parts,data members and methods. 2. In effect,methods are a special type of function,function that is exclusive to an object. 3. By default,it would be meaningless to print an object,but we can reimplement our method by defining __str__ function.Likewise,we can implement addition,subtraction of objects and so on. 4. OOP is actually quite prevalent in computer science because it shields the details of a complex object from the user. It is a form of abstraction which simplifies the implementation of an application.

[@johnnyfry2]

I know this is a three year old video, but this is really well produced and presented. This professor represents the MIT brand well. Meticulous in detail but yet, easy to understand. Excellent lecture. The best presentation I have seen on the subject, and I've seen a few. Thank you.

[@MrSrijanb]

this is a very informative and fast paced course. thanks mit.

[@erockromulan9329]

At 20 minutes in I was blown away at how fast it suddenly clicked for me. Wow! Great job, Dr. Bell!

[@danrivera381]

I heard once to think of OOP classes as blueprints. It made so much more sense to me that way.

[@RonSheely]

Thank you Dr Ana Bell. Most instructive. The fraction class gave me a hint into the mysterious world of computer algebra systems and how they might work by symbolically manipulating numerator and denominator.

[@user-tv6dh2jo5h]

Thank you MIT and Professor Ana!!! I really enjoy the lectures and I learned a lot.

[@marsum0971]

Till date , tis is the best way of teaching the class and objects concepts, typical employee class example has done nothing but creating confusion n hence failure in implementing classes when needed to many people.. Ana tis S epic tutorial.. Thanks MIT

[@ucmanhao4386]

Great and very understandable lecture ! Thanks, Dr.Ana Bell!

[@lakshmin121]

Hi. I am a beginner in python and find this course very helpful. I think the best part (and hence probably the most challenging) is that all the needed information is well packed into a few lectures. Thanks!

[@collinsogunsakin4506]

How's it going so far?

[@sahajthareja9415]

It starts to get real with this video.. Thanks MIT , Thanks Anna

[@CLi-kn6rf]

I was having trouble understanding the concept of object oriented programming and I found this. extremely useful and well explained!

[@obli8984]

did you>

[@Matt-lc6bs]

Thank you Dr. Ana Bell and MIT! The diagram and explanation at 19:30 is when I finally understood why self is needed. (I think)

[@diegoguisasola3858]

It was an "Oh, I get it now" moment for me as well lol. Best 41 minutes I've spent today.

[@ProfitScalpKing1]

amazing articulation. very clear and easy to grasp!! much respect!!

[@WaveShapePlay]

Thank you for this great lesson in OOP!

[@maximus2978]

This is the best explanation of python class I had ever seems!! Thanks for the lecture.

[@kebincui]

Awesome lecture. Thanks Dr Ana Bell

[@sreenimensa]

Amazing teacher!!! Makes the concept super simple and clear!!!

[@zlmsailor]

Anna is an awesome teacher! However the camera should probably focus on the course slides. I know I probably ask for too much for MIT Open Course. Thank you anyway and I will make donation at the end of this course.

[@yvonneachieng7257]

Extremely good explanations...Thank you Anna

[@rohithperumandla4302]

Thank you Dr.Ana for your clear lecture.

[@fishjimmy2007]

oh my god. MIT best. I feel pity didn't study hard in the past

[@uetianeinsteins204]

A bundle of thanks to MIT and especially to instructor : DR. ANA BELL..|||

[@ShermanSitter]

This was the BEST (by far) description of classes I have seen on the internet. Not surprisingly, but there is an obvious benefit to courses from top tier universities.

[@fi4255]

Thank you very much MIT and Dr Ana Bell, really very helpful.

[@abins641]

Finally I understand what is class , objects and attributes. Thank you very much.

[@fckthosealltrades849]

BRILLIANT! Ana is great instructor.

[@4sightanalytics264]

This is the best explanation I have seen. Thank You Thank You.

[@FelipeBatistaOsilva]

Thank you so much for this opportunity to learn!

[@chesq-1557]

Software engineering student here (first year, about to start my second one) I've been taught a bit of POO in my second semester, I've done a shit tone of data types with their own methods and such as is required by my Data Structures subject I'm watching this as a way to prepare for POO, a subject I'll be taught this coming year; this is my refresher and hell it feels nice to know most stuff so far if not all (I'm 9 minutes into the video) I'm really excited to know what my future will look like, so far it's being great (average grade of 7 out of 10, one honorary mention in a CLI orientes subject and good relationships with my classmates and professors) love you all!

[@chesq-1557]

@@summa1749 well hello! Software engineering, for me as for many others that get into it, has been quite a surprise. First of all, most people that get into it think it's like programing but there's a difference: software engineering is about providing quality solutions to a given problem using quality software designed using engineering principals, therefore a software engineer, eventho he/she knows how to program (and trust me, you WILL program) is better suit at designing the system itself, that's why you're taught things as database modeling and all as soon as second year. First and second year, as in most engineering stuff, is kinda Wide, meaning it's a general purpose year where you share all of the subjects with your mates from other engineering paths that might be similar to yours. You'll be taught a lot of mathematical concepts and the basics of programing+the not so basics. You'll also be taught (probably) the fundamentals of computers as in logic gates, secuencial circuits, you'll have to build yours using lab material and all. Overall it's a really cool experience and I love it, as hard as it might be sometimes, you just gotta be organized aaand do exams from past years to know what is required of you. If you have any more questions be sure to ask them! ^-^

[@chesq-1557]

@@summa1749 well if we talk about maths you're gonna have them, but most of it either is reduced to derivates/integrations of functions, which is not really difficult, you just gotta learn a couple of formulas, that and some assing series like Fibonacci and all, and that's only in the first year for the second year, the most maths I've used are limit aproximations at an infinite point to know the complexity of an algorithm; my personal pro-tip, as always, is to practice if anything, the hardest thing to do are physics problems, but that's just because I've had not so good professors, the books themselves are amazing tho aaand no unfortunately I'm not familiar at all with the english system :( , but as far as I know from looking it up A level is like... 16 to 18 sooo just pick any maths and physics related stuff + some programming if you can (HTML courses will NOT be that useful tho, pick C or Java, Python is kinda meh) again, ask whatever you feel the need to :D you can reach me more easily on twitter @memirasytecalo

[@zanakil]

OOP has 4 principles : Abstraction, Encapsulation, Inheritance, Polymorphism. Nice explanation of these here.

[@abanerjee3704]

Thanks MIT and Dr. Ana Bell 🤘

[@conceptgraduation694]

00:02 Object-oriented programming is about everything in Python being an object. 02:38 Objects are data abstractions with representation and interface 07:07 Object-oriented programming allows bundling data and interactions into packages for reusable and readable code. 09:20 Defining and using custom classes in Python 13:58 Creating and interacting with coordinate points in object-oriented programming 16:28 Methods and parameters in initializing a class 21:06 Adding methods to interact with objects 23:14 Using dot notation to access data attributes in classes 27:20 Defining a custom method for printing objects of a specific type in Python 29:25 Defining the Str method for the coordinate object. 33:58 Object Oriented Programming: Fraction object representation 36:19 Implementing fraction object using object-oriented programming 41:01 Object-oriented programming allows bundling similar objects with common properties and methods.

[@pengxu8542]

Ana's appeal has been really consistent so far

[@15837pawimomo]

Thank you for uploading

[@surenganesh7609]

thank for the informative presentation.

[@wanderingfido]

I would've been too tempted to try and explain the Iterator pattern as part of this lecture, even though it's supposed to just be a class for beginners. I guess that's why I'm not a [sic] perfessor. LOL. But on a serious note, people should read the Gang Of Four book from 1995 called 'Design Patterns' or one of the newer editions. Although, I wouldn't try to use any of the patterns deliberately. It can get frustrating to try and follow them to the letter. The GoF book should be read just to understand developer lingo during scrum meetings. If you remember to put a role or complicated task into its own class, that'll save you. Then the role or task becomes easier to change because it's already in a separate class file. It can be that simple.

[@marthinusjansevanvuuren2774]

This was very informative and helpful. Thanks!

[@MrDevZero]

Thanks. Great lectures. Love it.

[@SatyamSingh-nq2il]

I love how she measured every word before speaking it.

[@jflopezfernandez]

Awesome lecture, thank you so much

[@youngjordan5619]

guys, tips for you beginners / intermiadtes: function is the same as method except that methods are called methods cos they are in class, but it's the same as function, function is OUTSIDE class and methods are INSIDE class. But aside that it's the same thing. Another tip is that self. You use self ALWAYS (as long it's not static method) in class methods as a first parameter, even if method do not return anything, IT MUST HAVE that 1st parameter called self (stick to self). Another example, lady showed you bad example in 25:50, there is an easy way to do it by just declaring "static method" inside class so then you could use 2 objects like c.x and zero.x. And also I'd like to add this important answet to student who asked "how do we know if it's a float/string or int", we could determine specific data type in that constructor __init__ (or call it whatever you want) by just adding simple notation like def __init__ (self, x: str, y: int): ... so now if you have mypy, it will always watch if X is string and Y int, if not then error. You dont need to add some extra code inside method as lady said, waste of lines. and Lastly important info for you, join official python discord for help/resources here: discord.gg/python and here you have some resources you can learn from / improve: pythondiscord.com/pages/resources/

[@obli8984]

useful thanks for your valuable help

[@soldierkatr]

Do you teach??

[@terutherford4905]

My understanding is that the internal implementation of Python list is an array as it allows O(1) access operation in contrast to the linked list claimed by the lecture.

[@go_learners]

Yeah, suppose you have a list of i elements of same data type and one element takes 4 bit of space in memory. Then python creates a pointer of 4*k to access the kth element. It doesn't matter how long is the list, it will take same time to access any element. Obviously it will take more time for nested lists.

[@souravkumar-ue8uj]

Correct , internal implementation of pythonic lists is array so basically it's dynamic array whereas linked list is altogether a different data structure .

[@bdjeosjfjdskskkdjdnfbdj]

It seems like she’s referring to a custom list impl using singly linked list which is needed to teach any mod on sicp

[@johanv5546]

~36:00. c = a + b. How is a and b assigned to "self" and "other", how do you know which gets assigned to what? Would appreciate an explanation what happens and where a and b values are sent.

[@RicardoHernandez-nd5pp]

MIT, a preview of what came with distance education. Thanks

[@markuswalter3611]

Very clear explanation, great video

[@venkateshaprasads5741]

I understand what classes are.But what are the use of classes? What are the things that can be done in class but not in a normal function?

[@leilab9406]

First thank you for this great tutorial! In 36:00, why do we have inverse function without the leading and trailing underscores? What does it mean those underscores make a function "magical"?

[@FahimKhan-vd8yp]

That is one way to say it I guess, probably a misleading way. Those underscores differentiate built-in "accessories- I might say" from those user-defined. those are like built-in functions. For example, when python interpreter sees a "+" in a script associated with this object instances (like in "c + origin"), it's going to implicitly call "__add__" function(as written on the definition of that instance's object) and act accordingly what's written on that subroutine. You could implement "adding two coordinates/fractions" in your way-like defining a function as " def add(self,other)--the exact subroutine follows--", but then you won't be able to add two instances(coordinates or fractions or whatever) using shorthand "+" notation(like frac1+frac2), you'd have to explicitly tell that frac1.add(frac2), otherwise, you'd get an error. So, __add__, __float__ etc have the traits they have because python implies that they have, not because of "underscores", the underscores just differentiate them from others, marking them special; they do not not make something special. I hope that clarifies.

[@aydinjalilov2328]

Because that is not built in operator but a custom function that she has designed

[@rashidxd]

i really liked her lectures ... God bless You!

[@zigginzag584]

I feel like this could have been better explained in reverse. It wasn't until toward the end of the lecture that it really solidified in my mind that you have to instantiate, create an instance of the class by assigning a variable to an object of the class before you could use the methods on the class objects. That's unfortunate because it seems so pertinent to understanding OOP. However, I'm eternally grateful for the opportunity to be able to write this and actually know wtf I'm talking about! woo hoo for that!

[@albiiist]

I Loved the lecture..

[@waleedktl591]

Thank you internet! ❤️

[@PankajKumar-ji1ig]

Thank you for the lecture professor

[@tissues2441]

This is extremely helpful

[@lucam9806]

it really makes SENSE. thank you

[@anonymous.youtuber]

What a joy it must be to teach at MIT. You get listened to, and you get asked interesting questions. Excellent teaching this is.

[@souvikmakur9695]

improve ur grammer.

[@MichaelGTadesse]

Great introduction to OOP!

[@spiritualitywithkids]

Thank you so much for this awesome lect :)

[@christiananmoh9759]

This is very helpful. Thank you

[@marco.nascimento]

Amazing class heh

[@sandipansarkar9211]

GREAT SESSION

[@theshreyansjain]

Python lists are implemented as overallocated arrays and not linked lists

[@haroldgil7472]

FYI Slide 5 says lists are represented internally as a "linked list" of cells. However, the Python docs and 6.006 state that Python lists are implemented as "arrays", not "linked lists".

[@dinishdavis7329]

array in python ---- : mystr="hello", print mystr[1] ----this gives 'e' :). array is a data structure that stores values of same data type. in python, List store values of different data types ----list=[1,'a',"hello",1.1]

[@aabirbhattacharya3395]

thank u for amazing lecture

[@mahsam6878]

Thank you for the awesome presentation And for those who might be watching this but feel like they're understanding nothing, well I had the same feeling in the first 10 minutes, but things got crystal clear as soon as she started writing her own class. So, take your notes, if you do, and keep watching the video😁

[@linsco7s]

Good video. Well explained.

[@Damoon2543]

Firs of all I wanted to thank you for the awesome course. I also have a suggestion. I find it easier to follow when the class attributes and the function inputs are of different names. for example : def __init__(self, x0, y0): self.x = x0 ...

[@ravivdesai]

Damoon Rastegar what does x represent here vs x0?

[@iBot.]

That would just disrupt/throw me off, but I guess you do you.

[@LF58888]

Guess what? This is better then the 2 hour lecture from my prof

[@RicardoHernandez-nd5pp]

Proffesor Bell. Thanks

[@FahimKhan-vd8yp]

Thank You Mam! #just some random thought, I liked the way you said "oops" in 17:02

[@engineeringmasti4year635]

Thank you dr very nice video

[@19-barathkrishna39]

Will 'return' work instead of '__str__' ? @30:00

[@ZukoNtabo]

Class can be used to contain objects and can be used

[@wanderingfido]

This was a nice minimalistic intro to OOP in Python. But it's kinda weird that you can't 'return self(top, bott)' rather than 'return Fraction(top, bott)' as per Dr Bell's Fraction example. I just tried it in Python 2.6.6. If I recall correctly, Java lets you do that with 'this'. ...But come to think of it, Java probably allows that because you can't do Multiple Inheritance in Java.

[@carlitors]

Interesting, hadn't thought about it.

[@michaeljarmuzek9325]

she is returning a new Fraction, not 'self', if you refer to __add__ and __sub__ in the Fraction class

[@nedbatchelder]

She says that Python lists are linked lists. They are not. Python lists are vectors of pointers to objects.

[@isharthrastogi782]

5:35 python list data isnt actually internally represented as link list. Its internally is an array

[@ispeaknumbers]

Glad you pointed this out! I was confused on this coming from C to Python, and I thought of it more as a dynamic array than a linked list.

[@sreehari563]

I think Arrays in C might actually be represented internally as linked lists

[@hieuvuminh1653]

@@sreehari563 I'm pretty sure that's not the case. Array and linked lists are 2 different data structures and they are used in different applications. With array, the time you need to access any element is the same, whereas the deeper the element is in a linked list, the longer it takes to access that element.

[@michaeljarmuzek9325]

your answer is wrong in itself, "array" does not say anything about its internal representation.

[@davyroger3773]

@@michaeljarmuzek9325 Contiguous blocks of memory vs non contiguous aka linked nodes/blocks of memory

[@abhilasha100]

from where can I download the .py files and slides mentioned early in the video to follow along?

[@mitocw]

The course materials can be found on MIT OpenCourseWare at: ocw.mit.edu/6-0001F16. Best wishes on your studies!

[@notsifat]

great ,thanks

[@mohnishbelani3674]

Can anyone explain how is list.sort() implemented in Python? Which sorting algorithm is used and what is its efficiency? How can I find these details?

[@lepidoptera9337]

You read the source code. How hard is that? Or you just look it up on the web at wikipedia. It's called Timsort.

[@dinhnhobao]

I get the idea but why and when do we should use OOP. Why don't we just use the existent classes such as tuples, lists, dictionaries? Any examples for comparision between normal programming and OOP?

[@shanepython]

This is a question I have had. I "get" OOP but what are real world examples of where OOP is beneficial over Procedural programming? Someone explaining that would be very helpful.

[@quangtung2912]

OOP better than Procedure because of dynamic polymorphism and nothing else. Robert C Martin, the person who write famous books about OOP and SOLID, even said something similar.

[@karthiksk7246]

In 26:00 can we write zero.distance(c) instead of c.distance(zero)??

[@muhammadimranhaider2807]

you may but in function zero would consider self then computation formula became inverse

[@santiagosanchez2704]

Of course. Euclidean distance would be the same doing both things. So to speak... "You are as far to me as I am as far to you".

[@sathviks7442]

As she said before, self refers to any of the objects of a class. Therefore, since zero is also an instance of Coordinate, yes you can!

[@bilbobaggins8877]

Why doesn't the innit have a return statement?

[@stephan3694]

Can you share your code about the other function implementations(sub, add, float etc....)

[@mitocw]

Course materials, code snippets, etc. can be found on MIT OpenCourseWare at: ocw.mit.edu/6-0001F16. Best wishes on your studies!

[@srimurthy7143]

Thank you.

[@annakh9543]

superb

[@akbarrauf2741]

thank you ,mit

[@oftavukcuoglu]

giving "list" as an example is quite confusing but other than that fantastic lecture.

[@bla3029]

Best video 👌👌👌👌

[@ravi1286]

How I can get this presentation

[@jayshankar8086]

Nice ! extremely grateful learn from anna ma'am. Well I have questions( time-----24.11) what is guarantee that use will provide X,Y in distance(self,other) for OTHER parameter.how he/she even know this ? Definitely One way by documentation but any other way .?

[@hansu44]

It is not "guaranteed" in any way, it is the responsibility of the "user" (in this context: a person using this method in her code) to provide the correct number of arguments. The method is defined for two arguments, one of each is self, so the correct usage would be something like 'a.distance(b)'. Trying anything else ( 'a.distance()' or 'a.distance(c, d)') will cause an error and raise TypeError exception.

[@carlitors]

It's a very good question. Basically, the code is not particularly good, it doesn't really consider different use cases or check for types (for example, multiplying a fraction by an integer is a perfectly reasonable use case which the code ignores). There again, it would become a fairly long function if she did a series of if and elif to look at different uses... But I can see it being confusing.

[@DatascienceConcepts]

Nice!

[@blackcat7560]

helpful video

[@thanhnhan840]

great!