This repository contains three different folders:
-- CNNs: Files used to train the convolutional neural network.
-- PINNs: Files used to train the physics informed convolutional neural network.
-- SimNet: Files used in SimNet.
The architectures presented in the folder /cnns/Models and /pinns/Models were taken from: Mateus Dias Ribeiro, Abdul Rehman, Sheraz Ahmed, and Andreas Dengel. Deepcfd: Efficient steady-state laminar flow approximation with deep convolutional neural networks. 2020. https://github.com/mdribeiro/DeepCFD.
The dataset used for this project can be downloaded from the following link:https://uniandes-my.sharepoint.com/:u:/g/personal/f_giraldo_uniandes_edu_co/Ed8WmAvCFPdDhQBOX-J6LdQBYhcz13zzrn9sx-O8ZHT1fg?e=JsLqqj