Software Open Access

opannekoucke/pdenetgen: pde-netgen-GMD

Olivier Pannekoucke

Citation Style Language JSON Export

  "publisher": "Zenodo", 
  "DOI": "10.5281/zenodo.3891101", 
  "title": "opannekoucke/pdenetgen: pde-netgen-GMD", 
  "issued": {
    "date-parts": [
  "abstract": "<p>Bridging physics and deep learning is a topical challenge. While deep learning frameworks open avenues in physical science, the design of physically-consistent deep neural network architectures is an open issue. In the spirit of physics-informed NNs, PDE-NetGen package provides new means to automatically translate physical equations, given as PDEs, into neural network architectures. PDE-NetGen combines symbolic calculus and a neural network generator. The later exploits NN-based implementations of PDE solvers using Keras. With some knowledge of a problem, PDE-NetGen is a plug-and-play tool to generate physics-informed NN architectures. They provide computationally-efficient yet compact representations to address a variety of issues, including among others adjoint derivation, model calibration, forecasting, data assimilation as well as uncertainty quantification.</p>\n\n<ul>\n\t<li>Olivier Pannekoucke and Ronan Fablet. &quot;<a href=\"\">PDE-NetGen 1.0: from symbolic PDE representations of physical processes to trainable neural network representations</a>&quot;, Geoscientific Model Development (2020)</li>\n</ul>\n\n<p><strong>Description of the version</strong></p>\n\n<p>Version of the package based on tensorflow.keras, where neural network can be generated by using <code>TrainableScalar</code> or exogenous network.</p>\n\n<p><strong>Examples</strong></p>\n\n<p>As an illustration, the workflow is first presented for the 2D diffusion equation, then applied to the data-driven and physics-informed identification of uncertainty dynamics for the Burgers equation.</p>", 
  "author": [
      "family": "Olivier Pannekoucke"
  "version": "1.0.1", 
  "type": "article", 
  "id": "3891101"
All versions This version
Views 102102
Downloads 2626
Data volume 47.5 MB47.5 MB
Unique views 8585
Unique downloads 2424


Cite as