Installation and quick start#
Installation#
taupy is distributed on PyPi, the Python Package Index. To install the latest
release version, run:
pip install taupy
To install taupy from source, clone the
GitHub repository
and run pip install ./ in the root folder containing setup.py.
Quick start#
After installation taupy can be imported to Python either by importing
the module to the current namespace:
import taupy
or by importing every public object from taupy to the current
namespace.
from taupy import *
Note
All objects in the user guide can be publicly accessed even though they are
often referenced with their full path.
For example, the class taupy.basic.core.Argument can
be accessed simply as Argument if all taupy objects have
been imported to the current namespace using the *-notation, or as
taupy.Argument if the module has been imported to the current
namespace. It would also be possible to use the pedestrian reference
taupy.basic.core.Argument when the module has been imported.
Known installation issues#
taupy is written in pure Python and no installation problems should
occur for taupy itself, provided that Pyton 3.9 or newer is used.
There are some known problems installing dependencies that taupy relies
on. These can usually be solved by installing these packages before initiating
the taupy installation in pip.
On Windows#
Prior to version 0.5,
taupyrelied oniteration-utilities. This package implements functions in pure C, and its source code must be compiled before it can be used in Python. Unfortunately the package became unmaintained around Python 3.10 and no pre-compiled wheel packages were available. Windows users who use Python 3.10 or newer andtaupy0.4 or older are particularly affected by this, as they need to install Microsoft’s Visual C++ Build Tools in order to compile the C code.As of version 0.5,
taupyno longer depends oniteration-utilitiesbut usesmore-itertoolsfor advanced combinatorial tasks.taupyversion 0.4 or older can be used with Python 3.9 on Windows without installing additional build tools.
On Mac OS#
No wheels are provided for
scipyon the ARM version of Mac OS 11 via pip. On ARM Macs (“Apple Silicon M1/M2”) that run Mac OS 11,scipyneeds to be pre-installed via conda beforetaupycan be installed via pip. This issue can also be resolved by upgrading to a newer version of Mac OS.