Python modules for quantum chemistry applications
qctoolkit is quantum chemistry tool kit. It meant to provide universal interface to ab initio code to test ideas or to produce data reliably. The code includes Abinit, QuantumESPRESSO, Gaussian, NwChem, CPMD, BigDFT, ... etc. It also provide some basic molecule operations, including rotation, stretching, alignment, bond identification, ... etc, and data formatting, including xyz files, Gaussian CUBE files, V_SIM ascii files, pdb files, ... etc.
It seems worthwile to put effort to rewrite my bash/perl/python/C tools in to an integrated module or package. It should boosts the reusability, productivity, and reproducibility of my results generated during my PhD in Basel. More importantly, every results should be easily reproduced, examined, and especially furthre developed. This package starts as collections of modules of format I/O, analysis, plots. Hopefully, these modules can one day become a package for general purpose chemistry tool kit.
Installation on Ubuntu 32/64 systems:
-
To install:
cd /path/to/qctoolkit && python setup.py install --user
or install by pip usingpip install qctoolkit --user
. - Install on Amazon Ec2: It is tested and working on amazon Ec2 ubuntu instances. For a fresh install, all dependencies must be installed
sudo apt-get update && sudo apt-get install -y gcc g++ gfortran liblapack-dev liblapack-doc-man liblapack-doc liblapack-pic liblapack3 liblapack-test liblapacke liblapacke-dev libgsl0-dev libatlas-base-dev build-essential libffi6 libffi-dev libssl-dev libyaml-dev libpython2.7-dev python-pip python-dev freetype* python-matplotlib cython
Note that matplotlib and cython are installed through ubuntu repository for convenience. It might be necessary to create temperary swap if the memory run out:
sudo /bin/dd if=/dev/zero of=/var/swap.1 bs=1M count=1024
sudo /sbin/mkswap /var/swap.1
sudo /sbin/swapon /var/swap.1
Then do pip install pip install qctoolkit --user
If temerary swap is added, remove after installation:
sudo swapoff /var/swap.1
sudo rm /var/swap.1
-
To remove: Manually remove all created files. List of files can
be obtained by the --record flag during install
python setup.py install --user --record fileList.txt
-
Note that the
setup.py
script depends on python setuptools package. This can be installed bywget https://bootstrap.pypa.io/ez_setup.py --no-check-certificate -O - | python - --user
with superuser priviledge - The package depends on NumPy > 1.11.0, SciPy > 0.16.0, pandas > 0.17.1, and matplotlib > 1.5.1.
-
Note that newer version for many python modules are required. They must NOT
be installed via ubuntu repository. When a module is installed
through ubuntu repository as python-modeul (e.g. python-numpy),
import path of such module WILL GET highest priority.
In other words, stable but out-dated versions will always get loaded.
To circumvent this,
the best solution is to use virtual enviroment and setup dependancy.
However, it is also possible to modify the system behaviour
by edditing the easy_install path
/usr/local/lib/python2.7/dist-packages/easy-install.pth
Simply comment out the second line/usr/lib/python2.7/dist-packages
supresses the system to insert this path before PYTHONPATH. However, for some systems this fix is still not working. In that case, one has to modify the python default behaviour of constructingsys.path
variable. It can be done via modifying/usr/lib/python2.7/site.py
and add the following two lines to the end of the file:
sys.path.remove('/usr/lib/python2.7/dist-packages')
sys.path.append('/usr/lib/python2.7/dist-packages')
-
Note that all code are writen with 2-space indentation.
To change it according to pep8 standard, use the following command:
cd /path/to/qctoolkit && find . -name "*.py"|xargs -n 1 autopep8 --in-place
whereautopep8
can be installed simply viapip install autopep8 --user
Dependent Python packages:
- numpy 1.11.0
- scipy 0.16.0
- pandas 0.17.1
- matplotlib 1.5.1
- matplotlib.pyplot
- PyYAML 3.11
- cython
- psutil
- networkx
- periodictable
- mdtraj
- paramiko (newest version might be problematic, 1.17 works fine)
- And standard libraries: sys, re, os, glob, math, subprocess, multiprocessing, copy, collections, compiler.ast, shutil, fileinput, operator, inspect, xml.etree.ElementTree
- pymol is also used for visualization
- pyscf and horton is optional
Implemented interfaces to QM codes:
- Gaussian basis:
- Plane wave basis:
- Wavelet basis:
Required libraries:
- OpenMP
- openmpi
- gsl (GNU Scientific Library)
- LAPACK
- libxc-3.0.0
20150702 KYSC