Python/pip

From HPC Wiki
Jump to navigation Jump to search

General

Python is an Open Source, interpreted high-level programming language which is popular for prototyping high end simulations and data science applications. In addition to a large standard library there are many thirdparty libraries available (e.g. Numpy, Scipy). However, it is a single threaded environment and therefore may not be most suitable for very large simulations.

Usage

In Linux, there is a version of Python boxed in the distribution, delivered as RPM. This Python and some Py-modules (NumPy, SciPy, Matplotlib, ...) are available by default. The RPMs installed in exactly that versions which are available by OS, also typically older versions. The RPMs are updated accordingly to the OS distribution update rules. We do not offer any other versions/updates/fixes besides the distribution RPMs. Users cannot modify this installation. However, you are welcome to add needed Py-modules in your $HOME directory, e.g. by 'pip' or 'easy_install'. Python software often rely on GCC compilers and do not work with Intel compilers; thus type 'module switch intel gcc' before adding any Py-Modules.

Examples:

$ module switch intel gcc
$ export MYPY=$HOME/SomeDirForPythonInstall
$ export PYTHONPATH=$PYTHONPATH:$MYPY/lib/python2.7/site-packages
$ mkdir -p $MYPY/lib/python2.7/site-packages
$ easy_install --prefix $MYPY  theano

Note: you need to set PYTHONPATH anytime you wish to use 'teano' Py-module!

$ module switch intel gcc
$ pip install --user theano

Note: the installation is in $HOME/.local/[bin|lib] directories. Beware of clash if using multiple Py-modules!

Other Python versions

If you need yet another version of Python as the one available by default, take a look at the output of 'module avail python' command. It lists a number of Python installations available in the HPC Cluster. E.g. type 'module load python/3.6.0' for loading of the the appropriate version of Python3.

Note: the name of the Python binary version 3 is 'python2' and not just 'python'!

You can add Py-Modules in your $HOME in same ways as described above. In addition, some Py-Modules (NumPy, SciPy, ...) are bundled in these installations, and made available by PYTHONPATH environment variable (set by loading the python environment module). Whenever modifying this environment variable, you will likely wish to enhance the value instead to overwrite it, e.g. 'export PYTHONPATH=$HOME/my/newpatho/added:$PYTHONPATH' By unsetting the PYTHONPATH envvar you have the ultimate power to install any wished versions of these bundled Py-modules in your $HOME.

Intel also offers Python releases, available via the environment modules. List available versions with 'module avail pythoni' and load one by e.g. 'module load pythoni/3.5' for Python3 version. All notes above about installing software in $HOME and PYTHONPATH apply. Note that Intel also call Python-3 binary 'python' in addition to of 'python3', which differs the Intel release of python from other ones.

You're also free to install your own version of Python in your $HOME. All notes above about installing software in $HOME and PYTHONPATH apply.

Tutorials

As there are plenty of python tutorials available online we decided not to create our very own but instead simply list a few we consider to be most helpful.

https://docs.python.org/3/tutorial/

https://www.w3schools.com/python/

https://www.tutorialspoint.com/python/

References

https://www.python.org/about/