Bike-sharing-prediction-rides

In this model, I build a neural network and use it to predict the demand of daily bike rental rider. I implement a basic Multi-Layer-Perceptron network in order to build our model. Using concepts of feedforward, backward propogation and gradient descent, this model is trained and used to predict the patterns in bike sharing i.e- at what day how many number of bikes would be required. In this model we alos take in the consideration of weather, days of week, timmings etc. to to make my model more accurate and realistic.