Andreas Lauser
ee5ee215bd
explicitly include all Evaluations in the file for each specialization
...
indirectly this was already the case: Math.hpp includes
Evaluation.hpp, but this change should make it more explicit.
2017-06-07 15:20:01 +02:00
Andreas Lauser
1a1b17a5c8
revert optimizing the division code for evaluations
2017-06-07 15:20:01 +02:00
Andreas Lauser
19fb18dc20
clean up the code generator script for Evaluations a bit
...
this mainly removes superfluous newlines in the generated files, but
also makes them bit more consistent.
2017-06-07 15:20:01 +02:00
Andreas Lauser
2f86c78dd8
Revert "Revert "[WIP] Evaluation: specialize evaluations for used numbers in Blackoil setting""
...
This reverts commit cfc79fd6d5 .
2017-06-07 15:20:01 +02:00
Atgeirr Flø Rasmussen
cfc79fd6d5
Revert "[WIP] Evaluation: specialize evaluations for used numbers in Blackoil setting"
2017-03-28 13:26:10 +02:00
Robert Kloefkorn
49ff83d140
[cleanup] Avoid creation of empty lines and make work with python3.
2017-03-20 14:51:15 +01:00
Andreas Lauser
2f44918a2b
dense AD: fix a few stupid bugs
...
e.g., looping over the wrong range or an infinite loop. also, the
dense-AD unit test is shortend to test one specialization and the
unspecialized class.
2017-03-17 20:48:39 +01:00
Andreas Lauser
bf771b09b4
auto-generate and specialize the whole Evaluation class
...
older compilers (-> GCC < 5) seem to have trouble with inlining here
which leads to sub-optimal performance. since it is not a big problem
to also generate the unspecialized Evaluation class, let's do that
instead. (for hand-written code, this would be a huge
consistency/maintainance problem, though.)
2017-03-17 17:25:49 +01:00
Andreas Lauser
dff851951e
add description to the code generator for the Evaluation specializations
...
and make the maximum number of variables for which specializations
should be generated settable via a command line argument.
2017-03-16 11:51:24 +01:00
Andreas Lauser
72b9830b6c
move the script that generates the specializations for Evaluation to bin/
2017-03-16 11:17:33 +01:00