helanal cleanup #610
helanal cleanup #610
Conversation
orbeckst
commented
Jan 7, 2016
orbeckst
commented
- fixed broken helanal_trajectory() and helanal_main()
- slightly better tests
- minor code cleanups
orbeckst
force-pushed
the
feature-helanal-tlc
branch
2 times, most recently
from
0b3a486
to
819f4f9
Compare
|
This is done (I am not addressing the QC issues because I actually do not want to add doc strings to |
|
Pending favorable reviews I'd like this to go into 0.13.0 (#600). |
|
|
||
| def vecnorm(a): | ||
| """Return a/|a|""" | ||
| return a / veclength(a) | ||
|
|
||
|
|
||
| def vecangle(a, b): |
There was a problem hiding this comment.
mdamath.angle already covers nan issues, so this function can disappear (find and replace vecangle -> mdamath.angle)
|
I think I'm just being pedantic, if you want to merge just go for it |
|
Pedantic is good. Will fix. Oliver Beckstein Am Jan 7, 2016 um 4:17 schrieb Richard Gowers notifications@github.com:
|
- helanal_trajectory() was broken: fixed by using np.mean for mean - added simple regression test: compare helanal_bending_matrix.dat for alpha helix 8 in AdK in the test PSF/DCD (DIMS) trajectory - cleaned up test_helanal.py
- @quantifiedcode suggested: Consider simplifying bounds checks - removed import of psyco - replaced a "if ... pass else" with "if not (...)"
- return data as dict (instead of just logging) - fixed missing np.mean - use lib.mdamath were possible - added regression test
orbeckst
force-pushed
the
feature-helanal-tlc
branch
from
099cb28
to
3e7d847
Compare
|
Addressed all your comments and rebased against develop. Good to go (assuming travis is happy). |
|
@richardjgowers — all good. |
