Pyqcm¶

Pyqcm is a Python library developed by David Sénéchal, implementing quantum cluster methods for highly correlated systems.

Pyqcm is available as a precompiled Python package:

module load StdEnv/2023
module load python/3.11.5 scipy-stack/2024b
virtualenv --no-download $HOME/venv/qcm
source $HOME/venv/qcm/bin/activate
pip install --no-index --upgrade pip
pip install --no-index pyqcm==2.4.3

For a custom installation, for instance a development version or to experiment with options:

git clone https://bitbucket.org/dsenechQCM/qcm_wed
cd qcm_wed
module load StdEnv/2023 gcc/12.3
module load python/3.11.5 scipy-stack/2024b
module load flexiblas/3.3.1 eigen/3.4.0 cuba/4.2.2 primme/3.2
virtualenv --no-download $HOME/venv/qcm
source $HOME/venv/qcm/bin/activate
pip install --no-index --upgrade pip
export CMAKE_ARGS="-DEIGEN_HAMILTONIAN=1 -DWITH_PRIMME=1 -DBLA_VENDOR=FlexiBLAS -DPRIMME_DIR=$EBROOTPRIMME -DCUBA_DIR=$EBROOTCUBA -DWITH_GF_OPT_KERNEL=1"
pip install . --no-index

Note

The Pyqcm authors report low performance with BLIS. On the IQ HPC Platform, Intel MKL is used by default. We recommend not to substitute a different BLAS implementation.