Machine learning models programmed from scratch with numpy and pandas
This project is one that is still on-going for me. Recently I have been studying Machine Learning from lectures recorded from schools like Princeston, and Cornell. And I wanted to make sure I understand it all and completely grasp the subjects before I move on to more and more complicated models and tactics. So I started coding my own versions of the models, without allowing myself to copy paste, or even refer to others' versions of the models. There were multiple times that I got stuck and looked up how a certain equation worked, or what algorithms are used to solve things like numerical instability.
See the file called "notes.txt" for more details on what I learned on different steps, and where I added things in. That is still being updated so feel free to reach out to me with any questions or suggestions.
Here are a few charts from the models.
This first one is the prediction line of the Binary Regression model on a simple test dataset
This one is the prediction line from the Simple Linear Regression model on a dataset where a day's low temperature is being predicted
given the high for that same day.
And this one is a bar chart of the error distribution of the Linear Regression model as it models the same set
