My attempt to explain some python concept after learning them. Currently, the following concepts are explained:

• Python dunder methods
• Class variables

Python Concept: One day at a time

Python dunder methods

Python functions which start with a double underscore are known as dunder – du(double) under(underscore) functions. They are majorly used for overriding the functionality of class in python (think OOPs).

If you define a new class. When you try printing this class instance, python prints a cryptic string which contains the name of the class and its memory address. You can implement __repr__ function to define how it should be printed.

You can also implement __add__ to define the behaviour how adding something to the object should work.

Class Variables

A variable which is defined directly inside the class is called a class variable. You can access them via self.

class NLP:
    taste = "sweet"

class variables should not be mutable objects, but if used should be used with caution. Consider the example below:

In [1]: class NLP: 
   ...:     tools = [] 
   ...: anlp = NLP()                                                                                                                                                    
In [2]: bnlp = NLP()                                                                                                                                                    In [5]: anlp.tools.append('spacy')                                                                                                                                      
In [6]: bnlp.tools.append('nltk')                                                                                                                                       
In [7]: anlp.tools                                                                                                                                                      
Out[7]: ['spacy', 'nltk']

Adding spacy to anlp instance added spacy for bnlp instance as well. This behaviour can be undesirable in cases where another instance removes a tool you were currently using. This might create an issue while debugging as your object is being fiddled by another instance.

Just wondering this could be a desirable behaviour as well. In this case, because this could enable sharing knowledge of available tools across different instances. In all, mutable objects as class variables should be used with caution.

Detailed description: Python tutorial on Classes

Context Managers: with statement

This allow us to write enter and teardown code. It is really helpful to avoid writing explicit try/except blocks and also managing automatic closure of resources.

We need to implement entry and __exit functions. More details about it here

Timing your functions using Timer and with statements

It is possible to time various part of your system using Timer as a context manager, detailed description of it is in this post