Here are the stuff I did:
- I found a practice Dataset on Kaggle
- Then imported the .csv file into JupyterLab and dropped incomplete data.
- After which I used the Train-Test split and scaled the data
- Then using correlation heatmap, found important factors affecting the price
- Added a few factors myself using the original data.
- First, using Linear Regression got a score of 0.683
- Second time, using RandomForest Regressor I got a score of 0.822