Border-waiting-times---Time-Series-Forecasting-with-LSTM

Mexico-USA border-waiting-time series forecasting using LSTM neural networks. The dataset contains the waiting time and an UTC information format of almost all the sentry lines across Mexico-USA border.

The data was collected from July to October 2017.

Requirements

To run Tensorflow under Linux, here are the steps to achieve "Hello World" in Tensorflow on Jupyter via Anaconda.

Python 3.6
TensorFlow 1.3.0
Numpy 1.13.3
Keras 2.0.6
Matplotlib 2.1.0
Pandas 0.21.0
Sci-kit learn 0.19.1
Anaconda 4.3.30
Jupyter notebook

Output for waiting-times prediction of San Ysidro 'vehicle-ready' line:

Questions and contributions to the project are accepted. You can contact me at: agrija9@gmail.com