-
This is a sample webapp I built which gives real-time predictions of how many bikes a bike renting company will rent out depending on the weather. It depend mainly on 3 environmental variables:
- Temperature
- Humidity
- Windspeed.
-
All 3 variables above are my features, and the target/label is the number of bikes that will be rented out.
-
The I trained my data on the Machine Learning Algorithm, XGBClassifier, and after which I saved the trained model in a pickle format. The Jupyter notebook containing the data exploration and visualization, and then training of the model has been included in this repository
-
I built my webapp using the Flask framework, where I created a simple HTML file to take in the input of the user, and return the prediction.
-
The webapp was deployed on Heroku
-
All the files used in creating the webapp has been included in this repository
The deployed webapp can be found here: https://bike-usage.herokuapp.com/
- Please click the link and see for yourself.