One of my readers requested: "Any python exercise projects we will work on for learning you may suggest?"
You wager.
1) A Django Webapp
This is mainly for the ones of you who haven't finished internet development.
(Data scientists: I'm looking at you.)
Being capable of creating web software is a treasured talent for any developer. The motive is that it allows you to take another form of programming you do, and package it in a manner it truly is handy to the masses.
If you haven't carried out net dev earlier than, this wishes to be your #1 precedence, as compared to others on the listing. (If you *have* executed net dev, bypass to the following object... Get from your consolation quarter.)
What framework do you operate? Google will point out a dozen high-quality alternatives for you. It would not rely an excessive amount of which you use. You can select the one you want.
But in case you want a recommendation, I'll give you one:
Use Django.
It's a first-rate full-stack framework, and well documented.. If you locate yourself spending more than a couple of minutes selecting a framework, just use Django and get coding.
So that is one challenge concept. Next one:
2) A Command Line Tool
If you haven't discovered to create command-line applications... You're lacking out.
When you take your program and package it in a scriptable command-line interface...
With configuration controllable with the aid of options and flags...
And inputs and outputs for this system are managed with the aid of command-line args...
This ALWAYS increases the cost of your software. Always. A hundred% of the time.
So if you haven't ever completed it before... You need to learn.
Basically, this indicates gaining knowledge of the "argparse" module. It's built into Python's popular library.
There are other libraries for constructing command-line interfaces obtainable, which are not in Python's standard library. They have their fanatical fans who're already writing indignant emails to me, complete of misspelled phrases, for having the gall to recommend argparse as opposed to their preferred libwhateverz.
Ignore them. Argparse is complete-featured, and difficult to enhance on. And it's battery protected with Python.
So subsequent time you write a Python software, generalize it. Use argparse to make it more automatable, flexible, scriptable, and usual better.
So it truly is the second one undertaking inspiration. And sooner or later:
3) Machine Learning
If you have not ridden this hype teach but, you ought to take at least a brief day experience.
Yes, all of the yapping approximately synthetic system getting to know Intelligenz is over-hyped. But. It has actual substance, too. And you will gain from learning it.
You have options for what to do. I propose you study a library called scikit-research. It consists of tools for each supervised and unsupervised learning, and for building pipelines.
That's one option, and what I advise you to begin with. Another option is to research Tensorflow. I absolutely think you will do better if you visit that one after you have some enjoyment with scikit-analyze, however, if you insist on skipping in advance, at least make certain you study maths for handling "compute graphs" first.
So how do you use your new ML library? Well, it is quality if you can use it on troubles you're going through to your paintings. But that is difficult to do while you're getting to know the ropes.
So there may be an education ground: Kaggle.
Just look for "Kaggle Competitions", and search for the "Getting Started" class. They make it easy for you.