(Deleted) AttributeError: module 'fenics' has no attribute 'HDF5File'


72
views
0
7 weeks ago by
Hey

I installed fenics from source as suggested on https://fenicsproject.org/download/, including support for HDF5.

However, when i try:  outFile = HDF5File(mesh.mpi_comm(), 'filename.h5')
it gives: AttributeError: module 'fenics' has no attribute 'HDF5File'

I rechecked the cmake configuration for dolfin and it is able to detect HDF5 package on my system. Pasting a snippet of the 'cmake .." execution below.
I am new to using fenics. Primarily, I want to be able to write and read function outputs from a solver. From the information I have gathered, one can do this by working with hdf5 format.
In case there better ways to achieve that, please let me know. Thanks.

...
HDF5: Using hdf5 compiler wrapper to determine C configuration
-- Checking for Trilinos
-- Unable to find Trilinos (>= 12.4.0)
-- Trilinos could not be found
-- Checking for package 'Sphinx'
Traceback (most recent call last):
File "<string>", line 1, in <module>
ModuleNotFoundError: No module named 'sphinx'
-- Could NOT find Sphinx (missing: SPHINX_VERSION_OK) (Required is at least version "1.1.0")
-- The following features have been enabled:

* BUILD_SHARED_LIBS, Build DOLFIN with shared libraries.
* DOLFIN_AUTO_DETECT_MPI, Detect MPI automatically (turn this off to use the MPI compiler wrappers directly via setting CXX, CXX, FC).
* DOLFIN_WITH_LIBRARY_VERSION, Build with library version information.
* DOLFIN_ENABLE_GTEST, Enable C++ unit tests with Google Test if DOLFIN_ENABLE_TESTING is true (requires Internet connection to download Google Test when first configured).
* DOLFIN_ENABLE_DOCS, Enable generation of documentation.
* CMAKE_INSTALL_RPATH_USE_LINK_PATH, Add paths to linker search and installed rpath.
* DOLFIN_ENABLE_MPI, Compile with support for MPI.
* DOLFIN_ENABLE_PETSC, Compile with support for PETSc.
* DOLFIN_ENABLE_PETSC4PY, Compile with support for PETSc4py.
* DOLFIN_ENABLE_SLEPC, Compile with support for SLEPc.
* DOLFIN_ENABLE_SLEPC4PY, Compile with support for SLEPc4py.
* DOLFIN_ENABLE_TRILINOS, Compile with support for Trilinos.
* DOLFIN_ENABLE_UMFPACK, Compile with support for UMFPACK.
* DOLFIN_ENABLE_CHOLMOD, Compile with support for CHOLMOD.
* DOLFIN_ENABLE_SCOTCH, Compile with support for SCOTCH.
* DOLFIN_ENABLE_PARMETIS, Compile with support for ParMETIS.
* DOLFIN_ENABLE_ZLIB, Compile with support for zlib.
* DOLFIN_ENABLE_PYTHON, Compile with support for Python.
* DOLFIN_ENABLE_SPHINX, Compile with support for Sphinx.
* DOLFIN_ENABLE_HDF5, Compile with support for HDF5.


The following OPTIONAL packages have been found:

* MPI, Message Passing Interface (MPI)
Enables DOLFIN to run in parallel with MPI
* UMFPACK, Sparse LU factorization library, <http://faculty.cse.tamu.edu/davis/suitesparse.html>
* BLAS, Basic Linear Algebra Subprograms, <http://netlib.org/blas/>
* Threads
* CHOLMOD, Sparse Cholesky factorization library for sparse matrices, <http://faculty.cse.tamu.edu/davis/suitesparse.html>
* HDF5, Hierarchical Data Format 5 (HDF5), <https://www.hdfgroup.org/HDF5>
* ZLIB, Compression library, <http://www.zlib.net>
Community: FEniCS Project

1 Answer


0
7 weeks ago by
What do you get with has_hdf5() and has_hdf5_parallel() in dolfin?
Both give me 'False'. Although I am able to store and retrieve data using XDMF format now which solves my purpose, I would still like to know how to fix this issue with HDF5.
written 7 weeks ago by Pratyuksh Bansal  
Could anyone please suggest what would be the best way to store reference solutions and then be able to load them later during post-processing? I tried the checkpoint writing with XDMF but it does not help, I am unable to load the information that I saved. Thank you for your time.
written 5 weeks ago by Pratyuksh Bansal  
Please login to add an answer/comment or follow this question.
The thread is closed. No new answer/comment may be added.

Similar posts:
Search »
  • Nothing matches yet.