This project utilizes the following technologies:
-
ScimBa: A Python library emphasizing machine learning. ScimBa is used in this project to apply machine learning techniques. -
Feelpp: A library known for its Galerkin methods in PDE solving. Feel++ is used to solve PDEs in this project. -
Docker: A platform used to containerize the application. Docker is used to build a reproducible environment for the project. -
Python: The primary programming language used in this project. -
Git: Version control system used for source code management.
Each of these technologies plays a crucial role in the development and operation of the project.
Follow these steps to get the project up and running on your local machine:
Open the project in Visual Studio Code:
# Clone the repository
git clone https://github.com/master-csmi/2024-stage-feelpp-scimba
# To build a Docker image:
docker buildx build -t feelpp_scimba:latest .
# Run the Docker container
docker run -it feelpp_scimba:latest
#VS Code will detect the .devcontainer configuration and prompt you to reopen the folder in the container.import os
import sys
import feelpp
import feelpp.toolboxes.core as tb
from tools.Poisson import Poisson
# mandatory things
sys.argv = ["feelpp_app"]
e = feelpp.Environment(sys.argv,
opts=tb.toolboxes_options("coefficient-form-pdes", "cfpdes"),
config=feelpp.localRepository('feelpp_cfpde'))
# ------------------------------------------------------------------------- #
# Poisson problem
# - div (diff * grad (u)) = f in Omega
# u = g in Gamma_D
# Omega = domain, either cube or ball
# Approx = lagrange Pk of order order
# mesh of size h
# Example usage of the Poisson class
# Create an instance of the Poisson class for a 2-dimensional problem
P = Poisson(dim=2)
# Solve the Poisson problem with the specified parameters
P( h=0.05, # mesh size
order=1, # polynomial order
name='u', # name of the variable u
rhs='8*pi*pi*sin(2*pi*x)*sin(2*pi*y)', # right hand side
diff='{1,0,0,1}', # diffusion matrix
g='0', # boundary conditions
shape='Rectangle', # domain shape (Rectangle, Disk)
plot=1, # plot the solution
solver='feelpp', # solver to use ('feelpp', 'scimba')
u_exact='sin(2 * pi * x) * sin(2 * pi * y)', # exact solution for comparison
grad_u_exact = '{2*pi*cos(2*pi*x)*sin(2*pi*y),2*pi*sin(2*pi*x)*cos(2*pi*y)}' # gradient of the exact solution for error computation
)