sufiyandilawarkhan / ga1909 Goto Github PK

Python Editiors for Pushing Code to Production

spyder pycharm Vscode Atom Not jupyter

We should use function -- makes code Reusable

Never use keywords as function name, variable name dont use single letter variable name (example, a, b, x, y, etc.) Variable names and function names which convey the meaning of it

This makes our code more Readable

technical debt

Code should always contain documentation

Docstring to each function, class, modules Avoid inline comments as much as possible

Maintainable

Explainability -- how easy it is to understand the code in terms of logic -- how easy it is to understand the code in terms of machine learning

Scalability -- vertical scalability -- add more ram and harddisk to the same laptop or add GPU -- horizontal scalability -- add a new laptop

Assume 4GB RAM, 1TB HD, 4 Core CPU laptop ML model with 10000 datapoints ML model with 1Million datapoints We need to scale our infrastructure

Modular Structure Use of modules(multiple code files), classes and functions

path, config should be written in a seperate config file

Assignment - Take any past project check it for pep8 compliance check for issues and remove all those issues

Empirical logic to find summary Split the sentence and use first sentence and find the top 5 sentences with maximum number of words among the rest Document summarization ga1909 /bin - preprocessor.py Class ProcessDoc functions - special char removal - tokenization - stopword removal - summarizer.py Class SummarizeDoc functions - readDocs - loadConfig - splitter - groupSentence - findNumWords - findTop3Sent - sentenceCombiner

relative paths

git init git add . git commit -m "first commit" git remote add origin "https://github.com/SufiyanDilawarkhan/ga1909.git" git push origin master

Set your user name: git config --global user.name "FIRST_NAME LAST_NAME" Set your email address: git config --global user.email "[email protected]"

git add . git commit -m "first commit" git push origin master

APIs SOAP - little outdated, used where DotNet is used REST - very popular, most important reason

• fundamental priciple of databases (db)

CRUD C - Create R - Read U - Update D - Delete

REST GET - Read PUT - Update POST - Create DELETE - Delete HEAD

Framework Python

Flask - Most popular Django - popular Bottle - One page design FastAPI - New