TrappedResonance - #2257

Open

ejpaul wants to merge 260 commits into

PlasmaControl:masterfrom

jlabbate15:ejp-merge-master

This pull request is big! We're only showing the most recent 250 commits

Commits on Jul 31, 2025

Updated notebook
unalmis
committed
7dc7916
Dummy wrapper to avoid circular import
unalmis
committed
1bef805
Update _fast_ion.py
unalmis
authored
f22e7d4
Cast to array first
unalmis
committed
39d1912
Remove deprecated code
unalmis
committed
8558e89

Commits on Aug 2, 2025

Merge branch 'master' into ku/partialsum
unalmis
authored
b3a872c

Commits on Aug 4, 2025

Add tests for NUFFTS
unalmis
committed
c95b6ab
Add nuft2 vectorized
unalmis
committed
0062d6a
Merge branch 'master' into ku/partialsum
unalmis
authored
b961570

Commits on Aug 5, 2025

Add tests for rfft2 to nufft transform
unalmis
committed
48fe226
Merge branch 'ku/partialsum' into ku/nufft
unalmis
committed
4344e6f
Update changelog
unalmis
committed
66a48c2
Add failing test to show nufft developers
unalmis
committed
b52ad96

Commits on Aug 6, 2025

Resolve some todos
unalmis
committed
e9cad5d
Fix typo in comment
unalmis
committed
f2853be
Update root-finding
unalmis
committed
546aee6
Pull changes from ku/nufft
unalmis
committed
c985023
Merge branch 'ku/partialsum' into ku/nufft
unalmis
committed
e9e4cd3
Remove comments
unalmis
committed
64511a7

Commits on Aug 7, 2025

use partial summation approach in bounce2d.compute_theta
unalmis
committed
5543fe9
Merge branch 'master' into ku/partialsum
unalmis
committed
bdd174a
Merge branch 'ku/partialsum' into ku/nufft
unalmis
committed
ad505ad
small optimization
unalmis
committed
39611e4
small edits
unalmis
committed
a97b38e

Commits on Aug 8, 2025

working attempt at nuffts
unalmis
committed
8e3702a
clean up
unalmis
committed
11a3bf3
Add comment
unalmis
committed
7c59869
clean up logic
unalmis
committed
7ad892a
update comment
unalmis
committed
5d743f9
upload notebook to see if diff
unalmis
committed
270f4e3

Commits on Aug 9, 2025

Adding tests
unalmis
committed
32583cc
Updating eps setting
unalmis
committed
4ff85e8

Commits on Aug 10, 2025

updating tests
unalmis
committed
996b58d
adding warnings for AD bug
unalmis
committed
90dd6ea
running notebook with nufft compute
unalmis
committed
d70a06b
add back important comment
unalmis
committed
7348dff
clarify code
unalmis
committed
15e8fc6
clean up coordinate map
unalmis
committed
9c44d07
progress
unalmis
committed
b4e9b9a

Commits on Aug 11, 2025

Updating
unalmis
committed
6d3246e
Merge branch 'master' into ku/partialsum
unalmis
committed
74f2b12
update plots
unalmis
committed
a7ba43a
add assert statement
unalmis
committed
a83de07
change comment
unalmis
authored
c6f4523

Commits on Aug 12, 2025

Reduce tolerance
unalmis
committed
9065461
Add links to issues
unalmis
committed
31a399d
Update comment
unalmis
committed
44d68b6

Commits on Aug 13, 2025

Resolves #1574
unalmis
committed
8a9a482
Merge branch 'master' into ku/partialsum
unalmis
committed
03174ba
Merge branch 'ku/partialsum' into ku/nufft
unalmis
committed
4c5fd6b
last commit
unalmis
committed
dc1d21c
Remove temp variables to help garbarge collection
unalmis
committed
9341e97
dario comment suggestion

unalmis
and
dpanici
authored
7be33eb
Pulling changes down from #1834 which are necessary to address @f0uriest review comment
unalmis
committed
775bdf1
Merge branch 'ku/partialsum' into ku/nufft
unalmis
committed
b919f01
Add comment to address Rory comment
unalmis
committed
d5ff809

Commits on Aug 14, 2025

Merge branch 'ku/partialsum' into ku/nufft
unalmis
committed
29dc77e
Add more warnings
unalmis
committed
2dc4bda
clean up warnings
unalmis
committed
4d9962e
fixing doc build
unalmis
committed
16dff27
Changing variable name for Rahul
unalmis
committed
2b20f8d
Merge branch 'ku/partialsum' into ku/nufft
unalmis
authored
4156e12
Add comments
unalmis
committed
fd2bc47
Merge branch 'master' into ku/nufft
unalmis
committed
b58e34e

Commits on Aug 15, 2025

Fixing stuff from 1529
unalmis
committed
bd12d2e
Fixing stuff from 1529
unalmis
committed
01ee90e
Removing NUFFTs from previous commit
unalmis
committed
41e1f7e
Ensure gamma_c in [-1, 1]
unalmis
committed
eae25c9
Ku/real nufft (#1857)
unalmis
authored
0f46db3
Reducing tolerance
unalmis
committed
fcba0a4
clean conditional
unalmis
committed
79b6d3d
Add important fftfreq test
unalmis
committed
86a38c0

Commits on Aug 17, 2025

Change code to increase code coverage
unalmis
committed
dbc22bf
Preparing for merge
unalmis
committed
afb552b
Merge branch 'ku/fix_1529' into ku/nufft
unalmis
committed
15f4d96
Clean up verbosity
unalmis
committed
51cb599
Update _interp_utils.py
unalmis
authored
314f852

Commits on Aug 18, 2025

Making requested change to assume nufft2 is always used on real input
unalmis
committed
f628399
fix docstring
unalmis
committed
2921171
fix benign bug
unalmis
committed
7d88e81
Update default resolutions
unalmis
committed
aafc2f4
Merge branch 'ku/fix_1529' into ku/nufft
unalmis
committed
30fb2a9
Update plot
unalmis
committed
340bde2
fix comment
unalmis
committed
d67269c
update label
unalmis
committed
cf1dba1
Review requests
unalmis
committed
8fcf3cd

Commits on Aug 20, 2025

Merge branch 'master' into ku/nufft
unalmis
authored
48bbc27
Merge branch 'master' into ku/nufft
unalmis
committed
9e1b9e7
Add derivative tests
unalmis
committed
6a5dece
Make names consistent with recently officialized finufft tutorial
unalmis
committed
4e7b964
Reduce memory in test
unalmis
committed
4c02276
fix typo
unalmis
committed
17a4ff7

Commits on Aug 22, 2025

Merge branch 'master' into ku/nufft
unalmis
committed
bde9479
Merge
unalmis
committed
b6596f9

Commits on Aug 23, 2025

Add comment to docstring
unalmis
committed
5044d04
patching jax-finufft bug
unalmis
committed
01de804
Upload nufft optimization notebook
unalmis
committed
3b182a3

Commits on Aug 24, 2025

Reduce number of reshape/transpose to prepare for any unjitting required by #1303
unalmis
committed
c052d4e
Add another test
unalmis
committed
aed8897
Update benchmarks
unalmis
committed
946dcac
Fix benchmarks
unalmis
committed
741e044
Fixing benchmarks
unalmis
committed
154ac4e
STill trying to fix these dumb and useless benchmarks
unalmis
committed
935d747

Commits on Aug 25, 2025

Merge branch 'master' into ku/nufft
unalmis
authored
8fd0106
Merge branch 'master' into ku/nufft
unalmis
committed
6f770b0

Commits on Aug 27, 2025

Ku/finufft gpu (#1881)
unalmis
authored
14dca06
Update CHANGELOG.md
unalmis
authored
1090f0b

Commits on Aug 28, 2025

Merge branch 'master' into ku/nufft
unalmis
authored
7347a90
Dancing to appease other developers
unalmis
committed
d4d5d24
Update benchmarks
unalmis
committed
66fc063
fix link in docs
unalmis
committed
b0b4678
More dancing to appease developers
unalmis
committed
eaa6e4f

Commits on Aug 29, 2025

Increasing codecov
unalmis
committed
99a1342
Adding TrappedResonance Objective, although is not currently running

paullab
authored and
paullab
committed
16fe1c8

Commits on Sep 4, 2025

f_tr1 function update, optimizer not working and objective function running but outputting slightly incorrect numerical values

paullab
authored and
paullab
committed
8209995
Add files via upload
jlabbate15
authored
07bd9f8

Commits on Sep 7, 2025

objective function now working and outputting correct frequencies, have not yet checked actual value of objective function, but optimizer is running although not finishing correctly
paullab
committed
ce61025
some retouches to previous commit
paullab
committed
7d3dba1
middle of testing objective function, have to go to Italy dindin
paullab
committed
b25761c
post-checks of TrappedResonance objective function, slight changes
paullab
committed
752d40e
successful termination of optimization for both QS Triple Product and TrappedResonance objectives for QH_aScaling equilibrium
paullab
committed
e321e89

Commits on Sep 17, 2025

barely-trapped and ripple-trapped particle filtering flags added (tested to confirm omega_arr output is correct, but need to test with equilibria that have barely and ripple-trapped particles)
paullab
committed
f31e652

Commits on Sep 20, 2025

Updated objective function form to be more intuitive. See DESMOS

paullab
authored and
paullab
committed
2d58c21

Commits on Sep 29, 2025

Adding energy dimension into objective function

paullab
authored and
paullab
committed
9c1dc58
added radial drift (grad-psi) term to objective function

paullab
authored and
paullab
committed
909a2f6
fix to use correct iotas

paullab
authored and
paullab
committed
84793a8

Commits on Oct 7, 2025

updates to TrappedResonance objective function, lots of them. is not perfect yet

paullab
authored and
paullab
committed
67e2dd4

Commits on Oct 9, 2025

new plotting for TrappedResonance testing, minor updates to TrappedResonance objective

paullab
authored and
paullab
committed
403a539

Commits on Oct 11, 2025

normalization updates to TrappedResonance

paullab
authored and
paullab
committed
760eeda

Commits on Oct 12, 2025

bug fixes in TrappedResonance, without psidrift term, objective function now works

paullab
authored and
paullab
committed
f954694
oh my gosh it works TrappedResonance is working

paullab
authored and
paullab
committed
334e1bc

Commits on Oct 17, 2025

uploading progress for NERSC switch

paullab
authored and
paullab
committed
bea4014

Commits on Oct 18, 2025

save my computer please

paullab
authored and
paullab
committed
0a12954
updates

paullab
authored and
paullab
committed
2a723f3
connection with poincare plotting in firm3d is working

paullab
authored and
paullab
committed
a94d686

Commits on Oct 23, 2025

small compute changes

paullab
authored and
paullab
committed
1716505

Commits on Nov 1, 2025

beginning of new objective function form

paullab
authored and
paullab
committed
4dcb8e1
progress on new objective function

paullab
authored and
paullab
committed
5614866
completed new TRObj writeup based on Eq. 50 of Bounce Resonance Objective doc, debug time

paullab
authored and
paullab
committed
339a122

Commits on Nov 2, 2025

TRObj is running without errors, now can check values

paullab
authored and
paullab
committed
eb1c844
initial testing of f_tr2 complete

paullab
authored and
paullab
committed
ac0ac77
same as previous just something finished running

paullab
authored and
paullab
committed
06fe5b3

Commits on Nov 4, 2025

new objective function testing

paullab
authored and
paullab
committed
af267fc
omega_eta checks

paullab
authored and
paullab
committed
20d1c89

Commits on Nov 8, 2025

moving to nersc

paullab
authored and
paullab
committed
3a276b6

Commits on Dec 16, 2025

changes over the past few weeks, trying to get organized

paullab
authored and
paullab
committed
5818f15

Commits on Dec 17, 2025

trying to create an optimization example

paullab
authored and
paullab
committed
41c1deb
optimization not erroring out anymore

paullab
authored and
paullab
committed
149601f

Commits on Jan 5, 2026

fixed bug, n^-2 not n^-1 in island width

paullab
authored and
paullab
committed
f3a082c

Commits on Jan 19, 2026

updates pt 1 to f_tr2 to include 01-15-2026 meeting editions

paullab
authored and
paullab
committed
33f7d74

Commits on Jan 22, 2026

DESC dev meeting editions (edits to TRObj to increase speed and optimization ability

paullab
authored and
paullab
committed
011307c

Commits on Jan 26, 2026

fixing my mistakes

paullab
authored and
paullab
committed
58f16fb

Commits on Jan 27, 2026

pre-testing code updates

paullab
authored and
paullab
committed
5b7c3b5
barely-trapped filter update

paullab
authored and
paullab
committed
41f52e5
new updates to code now work with optimization

paullab
authored and
paullab
committed
284b6da

Commits on Jan 28, 2026

cleaning up comments

paullab
authored and
paullab
committed
ece01b2
cleaning up comments

paullab
authored and
paullab
committed
5688bd8

Commits on Jan 29, 2026

minor obj updates, fixing alpha averaging and some f bugs

paullab
authored and
paullab
committed
5ac4991

Commits on Jan 30, 2026

add surface-driven pitch_inv setting with user only specifying num_pitch

paullab
authored and
paullab
committed
28439a4
TR obj and opt example updated

paullab
authored and
paullab
committed
e24691f
fixed non-scalar output of TRObj

paullab
authored and
paullab
committed
ef2b564

Commits on Feb 5, 2026

resolution bug fixes and testing updates
John Anthony Labbate
committed
9e040e8

Commits on Feb 7, 2026

Removing restriction on duplicate resonances.
ejpaul
committed
6984d2c
Redefining drifts and initial testing.
ejpaul
committed
98c8c6f

Commits on Feb 10, 2026

Reverting some changes.
ejpaul
committed
e47db96

Commits on Feb 17, 2026

Update after more testing.
ejpaul
committed
5b812ae

Commits on Feb 19, 2026

Island width now working.
ejpaul
committed
223be95

Commits on Feb 20, 2026

Adding more debug diagnostics.
ejpaul
committed
45cd0a2
More diagnostics.
ejpaul
committed
cbc80b8

Commits on Feb 21, 2026

Update to eta grid.
ejpaul
committed
cba36a3

Commits on Feb 26, 2026

Fixing error with zero propagation. Adding in bump function capabilities.
ejpaul
committed
2e4ee61

Commits on Feb 28, 2026

Renaming integers.
ejpaul
committed
e368c88
more updating.
ejpaul
committed
e56b88d
Adding QA island width benchmarks.
ejpaul
committed
67a0542
Adding QH benchmarks.
ejpaul
committed
8670533
Update to QA benchmark.
ejpaul
committed
a1f5344

Commits on Mar 3, 2026

intermediate commit with phase-space averaging.
ejpaul
committed
16c8943

Commits on Mar 4, 2026

Improving debug data.
ejpaul
committed
bb2110d

Commits on Mar 6, 2026

Adding QI example.
ejpaul
committed
00d4ef1

Commits on Mar 9, 2026

Commits on Mar 10, 2026

Fix to wd_blur.
ejpaul
committed
6dd4cf3
updates to optimization example script
John Anthony Labbate
committed
6b3ffa6

Commits on Mar 12, 2026

Fix to DEBUG output.
ejpaul
committed
c12e56e
Merge branch 'ejp' of https://github.com/jlabbate15/DESC_TrappedRes into ejp
ejpaul
committed
6386653
able to optimize TRObj, albeit slow
John Anthony Labbate
committed
4fe4570
Merge branch 'ejp' of https://github.com/jlabbate15/DESC_TrappedRes into ejp
ejpaul
committed
b6d2dc4

Commits on Mar 15, 2026

Fixing radial weighting.
ejpaul
committed
7f47c50

Commits on Mar 20, 2026

small fix to TRObj build method, num_quad sustanance
John Anthony Labbate
committed
4d893be

Commits on Mar 26, 2026

updates to optimization function call, further updates to come soon
John Anthony Labbate
committed
d5ae5f5
removed f_q_conservative, added wd_blur. optimization issues persist
John Anthony Labbate
committed
d3f6c81

Commits on Apr 1, 2026

documentation updates and some minor fixes
John Anthony Labbate
committed
2579e1a

Commits on Jul 3, 2026

Restructuring and redefining FSA.
ejpaul
committed
71f2517
Merge resolution.
ejpaul
committed
9a670ee

Commits on Jul 4, 2026

Update to example notebooks.
ejpaul
committed
e2638d7
Restore QuasisymmetryTwoTermNormalized dropped by merge, notebook API fixes
ejpaul
committed
aa3a623
Merge remote-tracking branch 'upstream/master' into ejp-merge-master
ejpaul
committed
a736681
Apply black/isort/flake8 style fixes per CONTRIBUTING.rst
ejpaul
committed
d345164
Remove scratch/debug artifacts from ResonanceOpt before PR
ejpaul
committed
8dd8ab8
Add unit test coverage for TrappedResonance
ejpaul
committed
dd29fa5
Add value-correctness test for TrappedResonance
ejpaul
committed
da3076a
Add 4-point stencil for linear resonance-weighting, enable resolution test
ejpaul
committed
bad8c2e

Commits on Jul 6, 2026

Revert "Add 4-point stencil for linear resonance-weighting, enable resolution test"
ejpaul
committed
58ee02e
Remove QuasisymmetryTwoTermNormalized before PR
ejpaul
committed
2ecb981
Remove scratch research artifacts before PR
ejpaul
committed
6fb110b
Remove orphaned pytest-mpl baseline images
ejpaul
committed
b81e927
Remove duplicate CHANGELOG Performance Improvements section
ejpaul
committed
4b77719

Commits on Jul 21, 2026

Remove duplicate <L|r,a> compute quantity

ejpaul
and
claude
committed
dbca84e
Reuse _radial_drift for _s_drift_integrand

ejpaul
and
claude
committed
cd03d3a
Stop threading Equilibrium through trapped-resonance compute function

ejpaul
and
claude
committed
966aa34
Document the 3.5 MeV alpha birth energy and KE_frac in _resonance_physics

ejpaul
and
claude
committed
6f781d1
Fix stale 2.5*pi in barely-trapped filter comment/docstring

ejpaul
and
claude
committed
05f62fa
Document each key of _resonance_physics's returned dict

ejpaul
and
claude
committed
697775b
Clarify eta_drift/eta_drift_avg is omega_eta, not omega_zeta

ejpaul
and
claude
committed
5f25609
Remove Hermite-stencil resonance-weight refinement

ejpaul
and
claude
committed
db2bd7d
Call _radial_drift directly instead of wrapping it in _s_drift_integrand

ejpaul
and
claude
committed
f839743
Drop redundant safediv in _v_tau and _alpha_drift_integrand

ejpaul
and
claude
committed
fd59670
Drop redundant jnp.maximum floor in _v_tau, matching _fast_ion.py

ejpaul
and
claude
committed
d9ed862
Drop dead floors in f_q_abs/Delta_s_profile; remove unused num_well option

ejpaul
and
claude
committed
e32abea
Declare all kwargs that "trapped EP resonance" reads

ejpaul
and
claude
committed
d058321
Enable resolution-convergence test for TrappedResonance

ejpaul
and
claude
committed
025c11a
Remove redundant local simpson import

ejpaul
and
claude
committed
27e9c75

Commits on Jul 23, 2026

Merge branch 'master' into ejp-merge-master
dpanici
authored
f202ffa

Commits on Jul 24, 2026

Avoid double-squaring TrappedResonance's island-width penalty

authored and
Elizabeth Paul
committed
b36f56b

Commits on Jul 28, 2026

Drop the 1e-30 radicand clamp in _alpha_drift_integrand

ejpaul
and
claude
committed
3e39305
Allow per-flux-surface field line labels in Bounce2D

ejpaul
and
claude
committed
5e442fe
Add a Bounce2D backend for TrappedResonance

ejpaul
and
claude
committed
b87593b
State the nufft_eps evidence accurately

ejpaul
and
claude
committed
8dd6e3b
Pin Bounce2D against Bounce1D for TrappedResonance

ejpaul
and
claude
committed
34ae7d7

Commits on Jul 29, 2026

Simplify docstring for starting field line labels
ejpaul
authored
81f95b9
Clean up comments in _trapped_resonance.py
ejpaul
authored
f2b419a
Clean up comments in test_objective_funs.py
ejpaul
authored
23b6e85
Revert .gitignore and the CPU benchmark to master

ejpaul
and
claude
committed
1f57b6c
Merge branch 'master' into ejp-merge-master
unalmis
authored
fef9138
Return pointwise residuals from TrappedResonance

ejpaul
and
claude
committed
18f533e

Commits on Aug 1, 2026

Merge branch 'master' into ejp-merge-master
unalmis
authored
b3d8a6a

Commits on Aug 3, 2026

Merge branch 'master' into ejp-merge-master
unalmis
authored
9b001f6

Commits on Aug 5, 2026

Revert "Return pointwise residuals from TrappedResonance"

ejpaul
and
claude
committed
70dd8f0
Clean up comments in _trapped_resonance.py
ejpaul
authored
88f0c2f

Commits on Aug 12, 2026

Let TrappedResonance take a custom rho array

ejpaul
and
claude
committed
e1922f4
Fix isort import ordering in _trapped_resonance.py

ejpaul
and
claude
committed
f95567f
Add analytic Omega'(s) option to TrappedResonance

ejpaul
and
claude
committed
dc3273a

Commits on Aug 14, 2026

Make _root_cubic NaN-safe under reverse-mode AD

ejpaul
and
claude
committed
c4cdda1
Compute Omega'(s) analytically and drop the finite-difference option

ejpaul
and
claude
committed
03eb5ca
Remove duplicated code in the trapped resonance path

ejpaul
and
claude
committed
0674f7d
Simplify the resonance weighting and drop an unread dependency

ejpaul
and
claude
committed
b008331
Scope the TrappedResonance tests that gate CI

ejpaul
and
claude
committed
9347298

Commits on Aug 15, 2026

Restore the finite-difference Omega'(s) as an option, and inline three wrappers

ejpaul
and
claude
committed
b999dd5