In the Streamlit app, we perform EDA on a dataset containing Indian flight info. such as: source, destination, arrival, departure, duration time, intermediate stoppage, price of ticket, etc. We have trained a RandomForestRegressor using this data and the model predicts the price of a ticket after providing information about the flight. After taking user input, a dedicated page to predict ticket prices is added to the streamlit app. The Streamlit app is deployed in Streamlit Cloud.
In the flight_prediction.ipynb notebook, we explore and process the data for training, perform feature importance analysis, and train a RandomForestRegressor with a RandomizedSearchCV hyperparameter tuning.
Reference: Flight Price Prediction with Flask