• I did this project just to perform some next level feature engineerings to elevate the performane of this model.
• I have also created a gui using HTML,Bootstrap,CSS and implemnted the backend in Flask for this project and will be uploading that on Heroku soon.
• Actually this was a Kaggle problem.
• Initially I used RandomForestRegressor but that was not giving very satisfying results so I used RandomSearchCV to find best hyerparameters and came up witha a good model with r2 score of around 0.82.
• Also used ExtraTreeRegressor to visualize the importances of all the features.
• Just run app.py to see the gui on local host.

Do visit my blog for better explanations: https://machinelearningprojects.net/flight-price-prediction/