Function-Approximation-Demo

This repo Contains Python code implementing a neural network with the DeepXDE library. The model is used to approximate the Harmonic-Oscilator function

 mẍ + μẋ + kx = 0 

The model is first trained with 200 data points in NN and then 30(randomly selected) data points in NN2

Architecture

A Feedforward Neural network is used. The network has 1 input node, 1 output node and 3 hidden layers with 50 nodes per layer.

Data

A dataset from kaggle was used to train and test the model.

References

DeepXDE docs