zhengyu-huang / operator-learning Goto Github PK

View Code? Open in Web Editor NEW

32.0 32.0 6.0 366.91 MB

Julia 1.56% Jupyter Notebook 97.37% Python 1.04% Shell 0.02% GLSL 0.01%

operator-learning's People

Contributors

Stargazers

Watchers

Forkers

zhb0318 matthieudarcy zxycxj take-koshizuka bravedrxutf z-l1u

operator-learning's Issues

What discretization is the Helmholtz problem used in the paper?

In the paper, it just says "Equation (4.2) is solved using a finite element method on a 100 × 100 grid". I am not an expert in Julia, but is the porder chosen as 2 here for the Helmholtz data generated (i.e., piecewise bi-quadratic conforming element)? Thanks.

Operator-Learning/Helmholtz/src/Domain.jl

Lines 70 to 75 in 982f5eb

 function Domain(nodes::Array{Float64, 2}, elnodes::Array{Int64, 2}, 

 bc_nodes::Array{Bool, 2}, bc_types::Array{String, 1}, bc_funcs::Array{Function, 1}, 

 porder::Int64 = 2, ngp::Int64 = 3; 

 ω::Float64 = 0.0, 

 c_func::Function = (x,y)->1.0, 

 s_func::Function = (x,y)->0.0)

Availability of data for Navier-Stokes

Hello, in the Navier-Stokes DeepONet model:

Operator-Learning/Navier-Stokes/nn/DeepONet/learnDeepONet.py

Lines 31 to 33 in b7c4841

 theta = np.load(prefix+"Random_NS_theta_" + str(N_theta) + ".npy") 

 cs = np.load(prefix+"Random_NS_curl_f_" + str(N_theta) + ".npy") 

 K = np.load(prefix+"Random_NS_omega_" + str(N_theta) + ".npy")

where are the three .npy files located? In the data set link, the NS data set contains the input output pair of shapes (40000, *mesh_size), is this the same file to get the vorticity the variable names suggested above?

@Zhengyu-Huang

zhengyu-huang / operator-learning Goto Github PK

operator-learning's People

Contributors

Stargazers

Watchers

Forkers

operator-learning's Issues

What discretization is the Helmholtz problem used in the paper?

Availability of data for Navier-Stokes

Recommend Projects

React

Vue.js

Typescript

TensorFlow

Django

Laravel

D3

Recommend Topics

javascript

web

server

Machine learning

Visualization

Game

Recommend Org

Facebook

Microsoft

Google

Alibaba

D3

Tencent

	function Domain(nodes::Array{Float64, 2}, elnodes::Array{Int64, 2},
	bc_nodes::Array{Bool, 2}, bc_types::Array{String, 1}, bc_funcs::Array{Function, 1},
	porder::Int64 = 2, ngp::Int64 = 3;
	ω::Float64 = 0.0,
	c_func::Function = (x,y)->1.0,
	s_func::Function = (x,y)->0.0)

	theta = np.load(prefix+"Random_NS_theta_" + str(N_theta) + ".npy")
	cs = np.load(prefix+"Random_NS_curl_f_" + str(N_theta) + ".npy")
	K = np.load(prefix+"Random_NS_omega_" + str(N_theta) + ".npy")