eakgun / Time-Series-Prediction-With-RNNs

Generalized Implementation of RNNs for Chaotic Time Series Prediction

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Generalized Implementation of RNNs for Chaotic Time Series Prediction

The complete model has been created and coded without using any library or data pre-processing tool. Main objective is the prediction of the henon chaotic time series in future.

Functions

bptt.m - Back Propagation Through Time

forward.m - Forward Propagation

dsigmoid.m , dtanh - Derivates of activation functions

sigmoid.m - sigmoid activation function

gradDes - Gradient descent