Linear Classifiers
Today I studied linear classifiers and will break-down the concept in easy words.
So linear classifiers is a fundamental topic in Machine Learning , which is applied widely in field of ML and have lot of practical applications.
In our daily life , lets suppose if I've to design a model to determine whether a person is healthy or not , email classification as spam or not and so on ! it’s widely used in these!
Well, you may be wondering of why linear , why not simply just classifiers!?
So , the thing is that linear classifier means whatever dimensions ( easy words: features ) our dataset has , it tends to draw straight line in between and classify the prediction of those points or features of dataset as positive or negative!
The algorithm focuses on choosing the hypothesis which fits the most to our desired case!
Get Muhammad Talha Fayyaz’s stories in your inbox
Join Medium for free to get updates from this writer.
learned about arg_min indexing and mathematical notation of linear classifiers!
also learnt about hyperplanes , vectors , normal , theta ( vector ) .
Learnt about finding vectors if hyperplane is through origin.
hyperplane is that line — that hypothesis which we chose!
Also learnt about arrays in NumPy and some of functions associated with it which were practically important for ML.
That's all for today.
You can connect to me on LinkedIn.
