So I'm trying something new. I'm going to try and spend 100 days learning machine learning. I've been using Python for about a year and also have some theoretical knowledge of machine learning after completing this course. I now need to learn how to actually do machine learning.
Why am I doing this? What else is there to do right now? I'm bored, stuck in the house. The main reason though is being able to do machine learning would help with my PhD research immensely!
Why are you telling me all this? I'm hoping by posting on DEV it'll help me to stay motivated and keep practising. I'm also hoping that maybe I can find other people who want to learn machine learning (or are super bored) and we could figure some of this stuff out together.
Anyway...
Day 1
So today is the first day. I'm starting using data camp to learn scikit-learn. I'm hoping if I can pick up some of the coding knowledge here. Then once I understand it a bit, move over to Kaggle and try it out on some real datasets. Today I spent about an hour working on a K-nearest neighbor model. I was hoping to start with linear regression, but this was actually quite simple to pick up (always nice). The examples showed how to use it with the iris dataset and then I got to work with a dataset from american politics.
My favourite thing I learnt today is how to use a for loop to change the number of neighbors to be used, and evaluate the score of the KNN each time. This is so simple but blew my mind at the same time. It's such a clever way to work out the best number of neighbors each time.
Top comments (0)