Hi Everyone!
Sharing my first machine learning guided project. Using K numbers to predict the market price of cars.
Lot’s of new concepts and loads of numbers generated throughout so I’m not really sure how I did or how accurate my project is! Any tips and advice would be very much appreciated. Thanks a lot.
Predicting Car Prices.ipynb (221.6 KB)
Click here to view the jupyter notebook file in a new tab
Hey, this was excellent and very well presented. thanks for sharing!
Hey!
Well done! After reviewing your work, I did have one thing to mention:
When you split Train/Test, I would avoid having overlap between the two arrays, otherwise you are evaluating the accuracy of your prediction on data that has already been seen.
So, best practice would be to break-up things such that you shuffle the data (well done there!) but also you avoid any overlap between Train/Test datasets. This way you get a more objective measure of performance.
Looks really nice. Thanks for sharing! However, I think you made a mistake when using the Mean Squared Error. According to the documentation (sklearn.metrics.mean_squared_error — scikit-learn 0.24.1 documentation), the inputs are y_true then y_predict, and from what I saw you did it the other way around. Not sure whether this affects the outcome though.