Fabian Hoppe

Current problem: everything fails due to circular imports

Done so far: * a new module `communication_backends` has been introduced * `communication.py` from `heat/core` has been divided into `communications.py` (containing the basis `Communication`-class) and `mpi4py4torch.py` (containing the `MPICommunication`-class and...

Merged with #1288

Is this related to checking compatibility with PyTorch 2.1 and CUDA 12.2 (=CUDA-version on Stage2024 on HDFML)?

For some of the issues with QR that lead to the need for extensive refactoring as proposed in this issue see, e.g., #1324 #1317

@mtar @ClaudiaComito @JuanPedroGHM @Markus-Goetz @krajsek do you agree with aiming for a next release in April and thus dropping support of previous versions as suggested above?

Actually, we should go through the whole list of dependencies in `setup.py` and the `ci.yaml` files.

| Dependency | currently | suggested for next release | |--------|--------|--------| | Python | >= 3.8 | >= 3.10 | | mpi4py | >= 3.0.0| | | numpy | >=...

Both the tests on AMD- and Nvidia-Hardware failed at `meshgrid` (see above under "ci/codebase — The build failed! " -> "Details") Feel free to ask in case you need help...

Hi @samadpls, did you run the tests serial or in parallel? You can run them in parallel using, e.g., ``` cd heat mpirun -n 2 python -m unittest -vf /heat/core/tests/test_manipulations.py...