DD-LES

This repository contains codes for different machine learning algorithms for two-dimensional Kraichnan turbulence. We investigate decision tree regreesor, random forest regressor, feedforward neural network, and convolutional neural network in this study.

A top-level directory layout

.
├── data_spectral    # data for training and testing (Fine grid = 1024 x 1024, coarse grid = 64 x 64)
├── ml_codes         # Codes related to decision tree, random forest, and neural network (ANN, CNN) 
├── spectral_solver  # Spectral solver codes for 2D decaying homegenous turbulence problem
├── LICENSE
└── README.md

Required python packages:

• Numpy, SciPy, Matplotlib
• scikit-learn, Keras, Tensorflow