Machine Learning

The terms sounds very daunting at first, when I started with ML I was like why god why do I have to study this subject, the reason? NUMERICALS!! I don’t like solving numerical problems, until I discovered their power and still don’t like them, but I know their value know.

Our syllabus is more focused towards solving the elementary algorithms in ML by hand and I have no issues with that because to build a solid foundation you must know the math behind that.

What I had issues with are I prefer learning things after knowing the ins and outs of it. I mean like our study started with basic intro to ML and then zooop, we’ve got our exams. This didn’t help me to solidify my understanding of ML at all. Instead it instilled a fear of ML inside me. I started to think what’s this subject, why do I need to study this subject and all.

But, ML is beautiful. Every since I started to dive deep in DSA and Algorithms I’ve started to see the real computer science not just building some fancy crud applications with event driven architectures.

I was also frustrated because anyone whom I used to ask what do you mean by ML model won’t give me a good answer, their answers seemed pretty vague. I couldn’t get a grasp of what they were trying to make me understand.

Finally, I started to search about ML after all these exams and assignments and projects. Though ML is not my interest of study, but it’s changing slowly.

I found that ML model is nothing but a function which is learned on a given dataset, in simple terms:

• You have a basic function y = mx + c, and you need a dataset, now you run the whole data set through this function to find the best value of the slope(c) and intercept(b) for which we will now test the model against some test data data. and then we measure the accuracy and just say the model is this % correct.

That’s all fella!