This file was written in python and performs KNN and Random Forest regression on a bike sharing dataset. The dataset contains hourly data of rented bikes along with weather conditions and other relevant features. The goal is to predict the number of rented bikes based on a given hour and rain level.
It utilizes stklearn for model implementation and evaluation, pandas for data manipulation, and matplotlib for visualization.
This was the model used and analyzed in the analysis portion of the assignment.
This file is an enhanced version of models.py, created with the assistance of ChatGPT. Unlike the original models.py which only used "Hour" and "Rainfall" as input features, this version incorporates all available features from the dataset to potentially improve model performance.
It did greatly improve the R^2 scores for both KNN and Random Forest models compared to the original models.py.
The average +- of bikes went from roughly 525 to under 200 for random forest, and from roughly 550 to about 300 for KNN.
This file was just for fun to see how much the model could be imporved by using all features instead of just two.