I posted in http://code.google.com/p/tensorcan/  a C version of the 
canonicalization algorithm for tensors
without free indices. It is roughly 60x faster than `double_coset_can_rep` 
in `tensor_can.py`.

For tensor computations with many contracted indices 
`double_coset_can_rep`  takes most of the time, 
e.g. 95% of the time in `test_riemann_invariants1` in test_tensor.can.py 
(same using 'tensor.py' in PR 1700)

If there are few index contractions `double_coset_can_rep`  takes little 
time, e.g. in the gamma matrix computations
in PR 1699 it takes 14% of the time.

If there is interest in this, I can write a wrapper for the C 
implementation of `double_coset_can_rep`, to speed up
SymPy tensor computations with many contracted indices, if the wrapper is 
installed.
 

-- 
You received this message because you are subscribed to the Google Groups 
"sympy" group.
To view this discussion on the web visit 
https://groups.google.com/d/msg/sympy/-/pkGF-S54Z90J.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/sympy?hl=en.

Reply via email to