Python Introduction

 

Python

why python?

different type of python installation setup?

1)Why python?

Why python is getting so much popularity?

---->In the industry we are observed 10 years back or may be 15 years back that point of time java,dot.net this type of

programming languages are popular in that time.

---->Now a days we are using python or maybe java script we are trying to prefer more and more to use scripting based

programming languages as compare to any other programming languages.

1)SO Bye the way what is the python.

2)So each and everything we are trying to explore over here.

3)What all the differnet ways which we will execute python code,we will write python code.

4)we will try to set up our system as well.

5)Not in just one way, we try to set up our system in multiple ways.

6)Then we try undestand how python behave,how python going to work.

Python:-

---->It is also one kind of programming language and again we called as a interpreted language or we can say python

is the High-Level programming language with dynamic syntaxes.

---->Inside python we will able to do object oriented programming as well as like classes or objects we can try to create.

---->So they are many ways by which we can try to define the programming language or may be specially python programming language.

Q)Now ,Why programming is important?

---->As a Data Scientist why do we try to focus on programming or maybe across the tech industry, So why do we try to focus on the Programming part.

Reason:-(Very Simple Over here)

--->Whenever we will try to build some solution weather it is the AI solution or weather it is the Non-AI solution.

--->For sure we will try to serve for some of the client or may be we will try to work for some of the product.

---->So, Now roles and responsbilities of that particular product will be to solve some of the problem which by the client or

which some of the particular person is facing.

Ex:-

---->For example we try to build some of OLA,UBER kind of app.

----->Inside that i have to do prediction and forecasting that how much time is going to take from point-X to Point-Y.

or maybe if I have to find out that So what will be the fare if traveling from one point X to Other point Y,

So i have to predict or i have to forecast what will be the fare before i will start my trip.

---->So I have to find out all of those things for sure we need some interface we have to write some logic depends upon your Business logic, We have to write some KPI(keep performance indicator) and then based on that only based on so we will able to design that application, so we will be able to establish those application and which application will able to solve your problem statement.

----->Now everywhere to write a language let's suppose if I have to communicate with u for sure i try to use some language and am trying to use some combination of words and am trying to use some sort of sentences.

----->So, Similarly if I have to communicate with the system for a sure programming language is required there must be

some language must be available so that I will be able to communicate with the system, In a Similar way am trying to

communicate with all of u Some language may be English, Hindi or some other language I trying to use vocabulary, I am trying to use grammar, I try to use words, I trying to form sentences then based on that u all can understand me.

----->So in the similar way we have to give this kind of understanding to system,so programming language is the way by which we able to interact. This is where any kind of programming language comes into the picture.

----->Now a days we all know that every system weather it is the mechanical system or in any industry whatever system u are

using or u are going to use whatever mobile application u are using as of now so for sure each and every desktop application

each and every Mobile application each and every system that u are using is meant designed for some specific purpose,which

trying to communicate the system which try to solve some of the problems or may be some of the task.

----->So this is the where programming language comes in to the picture. So nothing but it's the way by which we try to the

communicate with the system the way we can give instructions(Ex:- am try to clarify all of u r doubts).

------>For sure python is the one of the language is not like that python is the only language.

------>I can try to use c,c++,java,java script,Go,Ruby,Perl,r,python,Scala.

So, NOW

We have all this languages why u r using python? (or) why only python for data science? (or) why majority of the people across the world always try to go after the python when it comes into a AI implementation? (or) Why we try to use each and every AI kind of system that we have build or that we are building? (or) Python is the only alternate that we have or we can try to build AI system may be on other language as well?

---->So i think we are supposed to know all such kind of alternatives.

---->So python is not the only language that we will be able to find out there are multiple languages.

  1)C
  2)C++
  3)JAVA
  4)JAVA SCRIPT
  5)Go
  6)SCALA
  7)python

---->we are using other languages for AI implementation as well,Is not that only python we are using we can use c,c++,java

etc..

---->Whatever i want i can use it,So i can try to use any of this languages,Each and every languages having AI based libraries

or AI based packages that completely fine that i can try to use it.

----->But the problem over here is that if try to learn c,c++, java ... we know the complexity that these languages are very much

complex in terms of writing syntax, In terms of expressing my views or my command, because anyhow with the help of this

languages is just try to give instruction to my system it's completely fine, I can try to utilize any of this languages to

give instruction to my system, the system will work for sure.

Behind using python for multiple reasons:-

----->But as a practitioner,as a developer,as a Data-Scientist,as a Data analyst, and as a Big-data Engineer i can use other

languages but python is one of the easiest languages that we will be able to find out in terms of implementation.

----->Python is one of the language we will able to findout which provides which hell lot of libraries which u can try to use

to build any kind of system weather it is the system programming that u are trying to do,weather it is the graphical user

interface that u are trying to build,weather it is the managing databases,data mining or any sort of thing u r trying to do.

----->u will be able to do it easily if your trying to use python.

----->The fallowing are the some of points whare u are using python.

https://miro.com/app/board/o9J_lSVkq8I=/?moveToWidget=3074457355061555168&cot=14

------>I can try to use python in each and every places where i can't use other languages an a easy way.

------>Any other language will try to give some sort of restriction.

----->But when it comes python it's actually very very easy utilize the python,the reason behind the python having that kind

of the API or python provides u or supports that kind of the libraries so,we can try to consume those libraries directly

into u r program or directly into u r development and u can try to use python in variety of places.

----->Again in terms of writing syntaxes into python as we are going to learn all those things,as we are going to practice lot of

python code.

----->So u will be able to find out that lerning the python is not at all big deal.

----->To build any kind of app python provides u all sort of libraries.

library:-

--->library is nothing but ita package of functions or methods.

--->For suppose if I have to write something, if I have to print something for sure am going to use print() function.

---->For suppose if I have to perform some sort of summation operation may be i can try to write some function which will try

to take variables and which will do the summation. ex;-.a+b kind of summation again I can try to do it.

---->That to an easiest possible way as compared any other programming languages.

----->Bcz of usage of python with respect to variety or varsity of an application so python is getting huge popularity plus.

----->When it comes to AI so again inside a python or may be whatever AI libraries the people published ex;-TensorFlow, skirt

learn, NumPy, panda, learn...

---->This all AI based Data science libraries is already availbale inbuild inside python.

----->So again as a Data scientist am not supposed to focus more on more on code, am supposed to focus more on more business problems.

----->I am supposed to focus more on more architecture part, data part, algorithm part, and mathematics part.

----->Bcz we are trying to prepare ourselves for a data science job or role.

----->Again as data scientist i have some other task as well am suppoce to focus,so i want some kind of language which will try

to make my life easier interms of implemenation so that if am thinking about data cleaning just an easy base i can able to

do it.

----->If am planing to build some kind of desktop applications for My AI system integration or software integration i must be

able to do it in the easiest possible way.

------>So am just looking for one simple programming language which can provide me flexibility to perform all of this tasks.

----->This is where python comes into the picture,this is where other languages is not having such kind of flexibility and

those languages are going to fail.

---->it's not like u can't create AI system in java,it's possible we can do it there are libraries for AI which availbale in

java,even c,c++ u can do it.

----->As a data scientist my job is to build a system which can do some kind of prediction,detection,forecasting which will be

able to do some kind of classification,which will be able to do tracking,which will be able to do some sort of

segmentation,which will be able to perform some kind of NLP task.

----->Now,If i am going to choose or if i am going to opt for some other languages yes i can do it,but in a hard way,Bcz the syntax

the way these languages have been designed syntaxes are different and very complex in terms of doing the implementation.

----->Where python solves u r issue because the majority of these AI algorithms are available in python programming language in the

easiest possible way.

----> we will see this things for sure we are going experiment all of those things.

----->We try to focus on each and every app building with help of python.

----->So,python is one of the popular languages for data scientists,because it kind of language any one can learn even kids they can

able to learn.

----->one more thing so in any other programming languages, let's suppose I have to do some data structure like implementation so may be linguistic implementation if I have to do in c,c++, java, file reading operation if I have to perform in c,c++,java,if i have to perform some kind of dictionary operation,if i have to perform the tree-based operation in any other programming languages or May in our college life u must have seen that it's very difficult, we have to write 100's line of code to achieve one particular task.

----->But in python just for simplicity those data structures functions or those API's are directly avalibale to u,we are not

supposed write from it down from scratch or we do not suppose to reinvent again to reutilize those things this is the where

python getting popularity.

----->Python is the library support, Python is having a huge amount of libraries support and u can try to use python for a variety

of implementation.

Why use Python:-

https://miro.com/app/board/o9J_lSVkq8I=/?moveToWidget=3074457355061555168&cot=14

Python Versions:-

---->Python comes in multiple versions 1)2.x version 2)3.x version

----->In our classes we are not going to use 2.x version and we are going to use 3.x version,Again in python 3.x version we are

able to find 3.5,3.6...

----->So we will be using some of the stable version according to our requirements.

----->Python supports scripting as well as (OOP)object-oriented programming.

Q)------>Many of us must be confuse b/w scripting and OOP language about this we can discuss later?

----->python supports both scripting and OOP.

----->programming is nothing but writing a syntax

----->Finally python is the easy of use and inbuild libraries,So that we can try to save u r learning time and implementation

time.

----->That's the reason we are going to use python.

Installation:-

System setup:-

----->BCZ System setup is important, unless until we are not doing system setup how u r going to practice.

----->So we try to do system setup in variety of of ways and then u try to write the code.

Why Anaconda:-

---->Even if i don't have anaconda prompt i can write python code,i can install python manually ,then i can install

dependency's it' easy,I will be able to do.

---->So python is nothing but the library,library is nothing but bunch of functions,bunch of classes and methods i will be able

to find out.

----->Anconda always try to give to the complete bundle or anaconda is nothing but it is complete package of the libraries as

well as the user interface(IDE'S) that u r looking for python execution in a easiest possible way this is where anaconda

comes in to the picture.

------>It's not like with out anaconda i can't write python code i can do it.

------>But package management wise GUI or may IDE wise so here it will give the complete bundle again for the simplicity.

------->What this anaconda creator has done so they are just bundle it,They have just package it each and every thing giving

to us.

   (Why again and again install  python s/w and it's dependies something they are given all the things in bundle,let's 
   take this package i have done hard work for u then u guy's are suppoce to use it,that's the reason anconda) 

------->It's the complete bundle of the libraries as well as the some of the IDE'S(Integrated development environment) not

all,we will be able to find out this is where anaconda comes in to the picture.

------->So we will be able to use anaconda in both the ways with the help of GUI or CMD.

------->both are same ,in one place we have to write command(Block screen)(CMD),In other place just we have to do click(GUI).

------->Everthing we can execute python code in (Black screen)CMD itself, may be I don't need some thing or some IDE(like

Jupiter notebook), Maybe I can try to take some text files, I try to write some code and then I save it over there then I can

execute from my block screen.

------->i don't need even a Jupiter notebook to execute python code.

------->But at the time of development, at the time of practice we always look for the kind of tool we are looking for that will make

my life a little bit easier in terms of storing files, in terms of writing some code, in terms of going through the

debugging, in terms of creating packages, in terms of creating API's, in terms of connecting with the databases maybe I look for

some kind of software or some kind of tools which can help me out to write python code in easiest possible way.

------>Now they are huge list of tools we will be able to find out.

       1)jupyter notebook or jupyter lab
       2)spyder
       3)vscode
       4)IntelliJ 
       5)pycharm
       6)atom
       7)txt.py

-------->So, they are multiple ways in which I can write my python code, They are multiple platforms or they are multiple software
where python is available(like all the above 7) or where we try to execute python code.

------->For sure in the block screen is not comfort again if we exit from the screen your code will be gone or unless u have to
save it manually and many other things packaging and many other things we will be not able to do it in easiest possible

way,it's possible do it,its possible but in hard way.

Jupyter notebook:-

---->we are thinking that the Jupiter notebook is the world,if i have going to do some kind of project implementation i have to

do it in the Jupyter notebook.

----->if am creating some AI system i can create in Jupyter notebook and this thinking where we are wrong completely.

----->jupyter notebook we can use as a beginner that is fine at the time of practicing we are using Jupyter notebook.

----->whenever we try to implement the projethenime we are not going to the use, Bcz Jupyter notebook, have some

limitations in terms of managing and maintaining the package in terms of connectivity, in terms of debugging,we will not able to find

out those flexibility in Jupyter notebook.

------>When we start as a begginer implementation of all the code in a Jupyter notebook is fine,but as we will progress even

what project that we are going to give u we won't be execute in jupyter notebook,Bcz those projects is the modular projects

the way we try to develop a S/W AI system or ML system are pipeline in to industry it's the replica of the exact same thing

that we have shared with u as the project.

----->Don't believe in giving u a project just along with some notebook where we can do some simple task,No that is not even

called as the project,that is even not close to a project.

------>So, here Jupiter notebook is having certain sort of limitations,we are going start directly with Jupiter notebook going

forward we are trying to switch Different IDE'S( 1)IntelliJ, 2)pycharm, 3)vs code).

----->Spyder I don't like personally although it's have much more facilities as compared to Jupiter

notebook.

----->But still Spyder(IDE) is the kind of outdated IDE i would say.

----->It's better to use these three always 1)IntelliJ,2)pycharm,3)vs code.

----->Any of this three it will be our choice, I(Sudhanshu sir) am not going to force use that only.

----->Just IDE whatever we can do in one we can do others,All of those things are advanced IDE.

------>Even we can discuss about how to do installation.

------>jupyter notebook is fine as a beginner we can start as a learner but at the end of the day for full scale s/w or AI

system development,We r not supposed to use jupyter notebook.

Note:-

----->Jupyter notebook is the just a web page,web page based Like IDE.

----->Jupyter notebook is not executing your code,Many people are thinks that Jupyter notebook execute the code but NO.

----->It's the python interpreter,It's the python compiler which is trying to execute your code.

Q)------>Now what is the interpreter?,what is the compiler?we will discuss this later.

Imp:-

------>jupyter notebook whatever like code which is trying to write here, whatever syntaxes which am trying to write over here

It's is going in backend to python compiler,python compiler will try to execute it and we will able to get the response.

------>This is how it's happen's.

------>If u r interpreter,if u r compiler will get struct sometime then we need to do restart,restart the complete services.

------>jupyter notebook is just the web page nothing else or nothing more than that.

Assignment:-

---->Whenever we are trying to submit an assignment, Should submit the (.ipynb) format,then only we are able to see the code.

---->Sometime we are doing mistakes we are downloading (.ipynb) file and then we are opening in note pad,That is not good

for open the file.

----->First we should launch jupyter notebook then we need to open that file from the jupyter notebook it's self,BCz it is the

readable one,we can check all the code,we can check all the outputs by ourself and they are assign the marks otherwise we

will get 0 marks.

Multiple Platforms for practice Python In online:-

1)Jupyter notebook online

2)GOOGLE COLAB or GOOGLE JUPYTER NOTEBOOK(Famous platform most people are using this platform)

Change runtime type GPU(Graphical processing unit)

                     TPU(Tensor processing unit)

----->It's hosted on shared system(cluster) it's free.

----->But people think that everything we can do in google colab,But No.

------>we can't do everything in google colab as the begginer this is fine,i can try to utilize it,but let's suppose am trying to

train my NLP model it's dificult.

------>Bcz the resources we can get over here is not a dedicated resource for us, it's the shares resource and time

interval may be 2mins,3mins,3, hrs,11hrs it is trying to decommission it is trying to reallocate the resources(RAM, Disk space and

computational power may be reallocated).

------>This not dedicated resource that u have ,So if u r try to train NLP kind of model which will take may be 1 day,2 day or

5 day of time which will take for sure model train doesn't mean that just with in a minute i will be able to do it,NO.

------> Sometimes it will take like 2 to 3 days consistently, So here we try to build some solutions into for NLP and for

VISION again this will fail.

----->This is the limitation in google colab,many people ask that why i can't to everything into google colab,Bcz it is the

shared resource we are using it is not your dedicate resources from google side.

Google Colab Pro:-

----->But google has released one option called Google Colab Pro, we have that option as well dedicated resource option.

----->Previous it's available only USA Region only,But u can connect that from India with the help of VPN.

----->From(through) VPN to USA Region then only we can use the Google Colab Pro services, It's cost 10$ per month.

----->Now a days it is available in India also.

----->There we can get high-end dedicated resources.

----->Agian we don't need this as of now,we need this kind of system once we can do the computer vision(CV).

Q)----->Some people confuse with the CV and OPEN CV?we will discuss later?

The best version of Python is 3.6 or 3.7 this is a stable version right now

3.8 and 3.9 still they are not releasing Depending

In [ ]:

​