neuraloperator
PDEBench
neuraloperator | PDEBench | |
---|---|---|
2 | 2 | |
1,809 | 631 | |
5.1% | 4.9% | |
9.4 | 6.3 | |
7 days ago | 2 days ago | |
Python | Python | |
MIT License | GNU General Public License v3.0 or later |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
neuraloperator
- Learn in Infinite Dimensions
-
Tensorflow version of layer running almost 10x slower
I've been trying to port the method used in this this paper, starting with the 2D version in this file from their github. I succeeded in replicating the results, verifying I get the same results and gradients. However, the resulting layer takes about 10x the time to run with Tensorflow (v2.4) compared to Pytorch (self compiled from git) on a V100. Here's my implementation of the layer using Tensorflow.
PDEBench
-
[P] LagrangeBench: A Lagrangian Fluid Mechanics Benchmarking Suite
LagrangeBench is a machine learning benchmarking library for CFD particle problems based on JAX. It is designed to evaluate and develop learned particle models (e.g. graph neural networks) on challenging physical problems. To our knowledge it's the first benchmark for this specific set of problems. Our work was inspired by the grid-based benchmarks of PDEBench and PDEArena, and we propose it as a Lagrangian alternative.
-
[D] what are the SOTA neural PDE solvers besides FNO?
try https://github.com/pdebench/pdebench
What are some alternatives?
squirrel-core - A Python library that enables ML teams to share, load, and transform data in a collaborative, flexible, and efficient way :chestnut:
transformers - 🤗 Transformers: State-of-the-art Machine Learning for Pytorch, TensorFlow, and JAX.
ivy - The Unified Machine Learning Framework [Moved to: https://github.com/unifyai/ivy]
pdearena
thinc - 🔮 A refreshing functional take on deep learning, compatible with your favorite libraries
d2l-en - Interactive deep learning book with multi-framework code, math, and discussions. Adopted at 500 universities from 70 countries including Stanford, MIT, Harvard, and Cambridge.
best-of-ml-python - 🏆 A ranked list of awesome machine learning Python libraries. Updated weekly.
ivy - The Unified AI Framework