ce3e47daa9
comment out some eme notes
2024-03-18 10:54:51 -07:00
99461dc129
alternate connect_s
2024-03-18 10:54:36 -07:00
0ad78271ca
more wip
2024-03-14 22:38:17 -07:00
2c5c40c1e7
wip junk
2024-03-14 22:17:09 -07:00
1b3d322fc6
some work on FDFD derivation
2024-01-18 00:54:03 -08:00
107c0fcd7e
set fdfd sources in a way that catches incorrect j.e calculation
2024-01-17 23:30:08 -08:00
c3b6fd94a6
should be e dot j.conj()
2024-01-17 22:20:59 -08:00
5c3b3532a9
M should be same timestep as E
2024-01-16 23:17:05 -08:00
24da3f673b
eigness is positive
2023-05-23 13:07:26 -07:00
b1a5cdcda9
bloch example updates
2023-05-23 12:56:48 -07:00
d8ec46674d
sqrtm increases precision, so cast back to double
2023-05-23 12:56:38 -07:00
be620f7137
comment updates
2023-05-23 12:54:18 -07:00
2c16c3c9ab
Fixup in-place operators
2023-05-23 12:54:07 -07:00
1ec9375359
loosen default tolerance
2023-05-23 12:52:25 -07:00
98c973743f
use if False instead of commenting out code
2023-05-23 12:52:17 -07:00
2a9e482e44
Z is y0 transposed
2023-05-23 12:51:12 -07:00
01e7aae41e
comment updates
2023-05-23 12:50:17 -07:00
c7e823b0b3
allow initial value
2023-05-23 12:49:55 -07:00
44465f1bc9
modernize type annotations
2023-05-22 10:53:13 -07:00
7009e505e7
fix accidental creation of array(None)
2023-05-22 10:52:55 -07:00
3bf56c16c1
More in-place ops
...
Z.copy() is needed since op is in-place now
2023-05-22 09:55:47 -07:00
5c7deedb70
do more in-place
2023-05-22 09:53:56 -07:00
fd1a83b5b9
cleaner way to ravel
2023-05-22 09:53:30 -07:00
7cf90fe0de
use keepdims instead of readding dims
2023-05-22 09:52:52 -07:00
09aa9761c6
Use same variable names as in code
2022-11-27 12:36:53 -08:00
697770ce97
improve top level bloch comment
2022-11-24 23:16:25 -08:00
b7bd825bce
make saving previous GtKG direction more obvious
2022-11-22 22:23:00 -08:00
ff395277b0
add another comment about minmization
2022-11-22 22:21:51 -08:00
dfbb845bee
add some comments
2022-11-22 22:21:31 -08:00
a64afcbe4d
Add more info to log message
2022-11-22 14:44:54 -08:00
03c15c8486
store ZtAZ instead of AZU
...
U is small (~number of modes)^2
2022-11-22 14:44:43 -08:00
bec0137c99
update type hints and formatting
2022-11-20 19:56:45 -08:00
68a9818388
use moveaxis instead of deprecated rollaxis
2022-11-20 19:55:51 -08:00
23d5a160c8
use f-string
2022-11-20 19:55:11 -08:00
4144e6dc37
update comment
2022-11-19 19:33:50 -08:00
8f8c130c2f
use previous result as next guess
2022-11-14 12:39:41 -08:00
4a9198ade7
allow setting initial guess
2022-11-14 12:39:29 -08:00
c7c71a3a82
add per-iteration callback
2022-11-14 12:20:50 -08:00
a82d8dfc7e
pass k_bounds and k_guess instad of just k_min and k_max
2022-11-14 12:20:33 -08:00
86feb5461c
don't use patient planner by default
2022-11-14 12:19:55 -08:00
edfd9a49c6
formatting
2022-10-06 13:45:04 -07:00
ad7c27eee3
bump version to v0.8
2022-10-04 17:29:13 -07:00
640b4d1ef7
Make find-k also return eigenvalues
2022-10-04 17:27:58 -07:00
4e240988c9
use "is None" to check for default args
...
numpy.any(numpy.equal(x, None)) is more general, because
`numpy.array(None) is not None`, but checking for that doesn't make much
sense if you're already type-checking
2022-10-04 17:27:44 -07:00
52df24ad98
Major typing updates
...
Split field types to differentiate between complex and purely-real
Fix lots of numpy-related stuff
2022-10-04 17:17:44 -07:00
31e6e0ec60
just check if file exists rather than trying to open
2022-10-04 17:09:14 -07:00
ad1ec6acfb
Back off on FFTW args
...
multithreading in particular seems pretty detrimental
2022-10-04 14:35:30 -07:00
d4fcfa1e07
fix var name
2022-10-04 14:33:47 -07:00
ec9a6edc85
be explicit about stack axis
2022-10-04 14:33:36 -07:00
71d6ceec1e
update to new fdtd approach
2022-10-04 14:32:54 -07:00
