Skip to content
No description, website, or topics provided.
Jupyter Notebook TeX Python Makefile
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Type Name Latest commit message Commit time
Failed to load latest commit information.
nnpde Update log info Jan 3, 2019
.mailmap update emails Jan 4, 2019

EPFL CS-433 Machine Learning: Project 2

ICLR 2019 Reproducibility Challenge: Learning Neural PDE Solvers with Convergence Guarantees

This is the repository for the second project in the CS-433 class at EPFL.

We tried to reproduce the results of a paper handed for the ICLR conference:

The link to our issue is:



conda env create -f environment.yml --name <your chosen name>

Followed by:

source activate <your chosen name>


# Note: not everything is listed here, use this as a guidance.
├── environment.yml                          # environment file
├── nnpde
│   ├── main.ipynb                           # main notebook, entry point
│   └── nnpde
│       ├──
│       ├──                    # geometries: shapes and boundaries
│       ├──                       # more project based helpers
│       ├──             # definition of iterative solver
│       ├──
│       ├──                         # model definition
│       ├──
│       ├──                      # definition problems
│       └── utils                            # various helpers
│           ├──
│           ├── jupyter.ipynb
│           ├──
│           ├──
│           ├──
│           └──
├──                                # this file
├── report                                   # latex script, plots, etc.
└── references
    └── paper.pdf                            # paper on which this is based

The notebook files were converted using this script, but should be viewed as a notebook.

General comments about the code

The deep learning part is implemented in PyTorch, therefore when in doubt it's a PyTorch tensor.

Authors (in alphabetical order)

Francesco Bardi, Samuel Edler von Baussnern, Emiljano Gjiriti,,

You can’t perform that action at this time.