muSpectre/6464caaabf74feat/muLib_outputs
feat/muLib_outputs vs master
Commit | Author | Details | Committed | ||||
---|---|---|---|---|---|---|---|
763a0c1627a3 | MartinDoskar | Forgot changes | Jan 13 2019 | ||||
889754cd3652 | MartinDoskar | Fixing output format | Jan 13 2019 | ||||
ad453e3c5c8d | MartinDoskar | Correct CMAKE setting and doxygenized classes | Nov 19 2018 | ||||
2cf1ca4e9e9b | MartinDoskar | Change indices to match muLib indexing | Nov 18 2018 | ||||
68eff119b9f3 | MartinDoskar | Finished result outputting | Nov 18 2018 | ||||
7ab093260a67 | MartinDoskar | Dependecies, initiall version of mulib_output | Nov 5 2018 | ||||
a0e879bacc0b | MartinDoskar | Initial commit to feat/muLib_outputs | Nov 2 2018 | ||||
cfe3c6cf107e | MartinDoskar | Testing access | Nov 2 2018 | ||||
2a851c25b4fd | junge | functional muLib runner in python3 | Oct 4 2018 | ||||
fd8d67d65cf0 | junge | small CMake fix | Sep 25 2018 |
/
README
µSpectre
Copyright © 2018 Till Junge
µSpectre is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3, or (at your option) any later version.
µSpectre is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.
You should have received a copy of the GNU General Public License along with GNU Emacs; see the file COPYING. If not, write to the Free Software Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
Building µSpectre
µSpectre is a CMake project that uses C++14. It depends on the Boost unit test framework for testing and uses uses python3 as secondary API. You will need a modern C++ compiler (µSpectre was tested with gcc-6 , gcc-7, clang-4 and clang5) and CMake version 3.7.0 or higher.
Compilation
$ git clone https://c4science.ch/source/muSpectre.git $ cd build $ cmake -DCMAKE_BUILD_TYPE=Release .. $ make
µSpectre makes use of expression templates, as a result, production code must be compiled with the CMAKE_BUILD_TYPE flag set to Release in order to get performance (under gcc, non-optimised code is about 50 times slower than the release). Be careful with parallel compilation: compiling µSpectre is quite memory-hungry because of the use of expression templates, a parallel make -j requires currently about 10 GB of RAM under GCC.
Simple usage example
The following is a simple example for using µSpectre through its convenient Python interface
#!/usr/bin/env python3 import numpy as np import pyMuSpectre as µ # setting the geometry resolution = [51, 51] center = np.array([r//2 for r in resolution]) incl = resolution[0]//5 lengths = [7., 5.] formulation = µ.Formulation.small_strain # creating the periodic cell rve = µ.SystemFactory(resolution, lengths, formulation) hard = µ.material.MaterialHooke2d.make( rve, "hard", 10e9, .33) soft = µ.material.MaterialHooke2d.make( rve, "soft", 70e9, .33) # assign a material to each pixel for i, pixel in enumerate(rve): if np.linalg.norm(center - np.array(pixel),2)<incl: hard.add_pixel(pixel) else: soft.add_pixel(pixel) tol = 1e-5 cg_tol = 1e-8 # set macroscopic strain Del0 = np.array([[.0, .0], [0, .03]]) if formulation == µ.Formulation.small_strain: Del0 = .5*(Del0 + Del0.T) maxiter = 401 verbose = 2 solver = SolverCG(rve, cg_tol, maxiter, verbose=False) r = µ.solvers.newton_cg(rve, Del0, solver, tol, verbose) print("nb of {} iterations: {}".format(solver.name(), r.nb_fev))
You can find more examples using both the python and the c++ interface in the bin/ and tests folder.