Indoor Localization Project dealing with crowd-path prediction using LSTM

This project deals with predicting the future path of a crowd moving in an indoor location. The current set of crowd path is collected using sensors, which is then fed into an LSTM to predict where the crowd will be after certain time-units. At current state, the model yields a MAE Loss%: 14.34144526720047 and MSE Loss% 3.96946482360363