Hello to my fellow programmers. How was your day 1 of machine learning?
I hope it was great. Here’s how mine went.
Contents
Goals For The Day
My target today was pretty simple:
- What is machine learning?
- Why is machine learning important?
- Some example applications of machine learning
- Setting up Jupyter Notebook. (Cause I have heard it is good for ml)
I know the list is pretty small but I only have 2 hours (learning, blogging, video editing everything.)
And I didn’t want to give up on my first day.🤷♀️
My Learnings
At first, I googled a lot to understand Ipython and Jupyter Notebook. I cleared up a lot of my misunderstandings.
One of which was:
Jupyter Notebook And IPython
Ipython is a python library that provides a facility to execute python code interactively. That’s something python idle does too then, why Ipython?
Well, with all I could understand it provides great features for data analysis.
It is pretty much like Jupyter Notebook offline at least in terms of how it works.
Jupyter Notebook is a web-based application for creating and sharing computational documents.
Interestingly Ipython and Jupyter Notebook are different and Jupyter Project has a lot of other facilities.
I didn’t go in-depth.
Jupyter Notebook uses a lot of IPython to provide python code execution. (It allows you to work with multiple languages)
Jupyter notebook is open source.
I worked bits here and there with the classic Jupyter Notebook. I think I should have gone with Jupyter Labs but that’s ok.
What Is Machine Learning?
After Jupyter Notebook, I went on to the question what is machine learning?
Machine learning is a simple idea of machine learning by itself without being explicitly programmed.
In more sophisticated words: A computer program is said to learn from experience E with respect to some task T and some performance measure P, if its performance on T as measured by P, improves with experience E.
Why ML And Examples
Now that I knew what ml is and had gone through IPython and Jupyter Notebook. I went on to explore why is there even a need for machines to learn.
Cause humans are lazy can be one reason. But, what else can that be? How can they do stuff better than we do?
There are 3 major reasons that I understood:
- Where the traditional approach makes the problem even more complex. Or the problem is so complex that the traditional approach can’t find a solution.
- In a constantly fluctuating environment.
- Where we wish to gain some new insights from the data (data mining)
Then I saw a bunch of different examples where ml can be useful (they qualify the above reasons)
news article classifying, brain tumor detecting, flagging offensive comments, recommendation systems, etc.
That’s it for the day!
Resources
These are some of the resources I referred to:
- Hands-on machine learning book (for an introduction to ml)
- Python for data analysis (ipython and jupyter notebook)
- Ipython Docs
- Jupyter Docs
- Google, GPT, and Bard of course.
And yes if I ever forget to let you know all your suggestions are always welcomed(Badly needed).
See you tomorrow.👋
