sym 0.1.7

Unified wrapper over symbolic manipulation libraries in Python.

sym provides a unified wrapper to some symbolic manipulation libraries in Python. It makes it easy for library authors to test their packages against several symbolic manipulation libraries.

Currently the following Python pacakges are available as “backends”:

• SymPy

• SymEngine

• PySym

• SymCXX

The capabilities exposed here are those needed by

• pyodesys

• pyneqsys

and include:

• Differentiation

• Numerical evaluation (including “lambdify” support)

see tests for examples. Note that pyodesys and pyneqsys also act as test suits for this package.

Documentation

Autogenerated API documentation is found here: https://pythonhosted.org/sym (and dev docs are here: http://hera.physchem.kth.se/~sym/branches/master/html).

Installation

Simplest way to install sym and its (optional) dependencies is to use pip:

$ pip install --user sym pytest
$ python -m pytest --pyargs sym

or the conda package manager:

$ conda install -c bjodah sym pytest
$ python -m pytest --pyargs sym

Source distribution is available here: https://pypi.python.org/pypi/sym

Example

Differentiation

>>> from sym import Backend
>>> be = Backend('pysym')  # just an example, use SymPy rather than pysym
>>> x, y = map(be.Symbol, 'x y'.split())
>>> expr = x*y**2 - be.tan(2*x)
>>> print(expr.diff(x))
((y**2) - ((1 + (tan((2*x))**2))*2))

for more examples, see examples/, and rendered jupyter notebooks here: http://hera.physchem.kth.se/~sym/master/examples

License

The source code is Open Source and is released under the simplified 2-clause BSD license. See LICENSE for further details. Contributors are welcome to suggest improvements at https://github.com/bjodah/sym

Author

Björn I. Dahlgren, contact:

• gmail address: bjodah

• kth.se address: bda