-
Notifications
You must be signed in to change notification settings - Fork 3
/
ChangeLog
7084 lines (7026 loc) · 349 KB
/
ChangeLog
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
Changelog
=========
(unreleased)
------------
Fix
~~~
- Forgot to assign IP to cuIonElement structure fields. [Martin D.
Weinberg]
Other
~~~~~
- Clean up messages in CMakeLists.txt and add SLURM support library
where needed. [mdw]
- Added nvToolsExt library as needed. [Martin D. Weinberg]
- Adjusted CMake config for CUDA extra libraries and removed automake,
autoconf cruft. [Martin D. Weinberg]
- Updated autogen.sh for CMake. [Martin D. Weinberg]
- Added extra CUDA lib to CUDA_LIBRARIES. Added doxygen detection and
configuration. [Martin D. Weinberg]
- Fix linking problem on unity. [Martin D. Weinberg]
- EXP library changed from libexp to libEXP to prevent target conflict
with exp executable. Missing EXP reference added to utils
directories as necessary. [Martin D. Weinberg]
- Added missing EXP globals. [Martin D. Weinberg]
- Configuration updates based on Robert Blackwell's start on the port
from automake/conf to CMake. - Added CMakeLists.txt configuration to
all missing directories. - Added install targets. - Tweaked the
main HDF5 configuration at the top level. - All libraries are now
shared. - Added toggles for user modules (USER_ENABLED and USER_ALL)
[Martin D. Weinberg]
- Change header format to system from local. [Martin D. Weinberg]
- Add config_cmake.h.in (was .gitignore(d)...) [Robert Blackwell]
- Remove FindMKL and do general cleanup of main CMakeLists.txt. [Robert
Blackwell]
- Merge remote-tracking branch 'origin/Eigen' into cmake. [Robert
Blackwell]
- Check in double exponential test. [michael_petersen]
- Fix the mean mass computation. [Martin D. Weinberg]
- Fixed a failed merge. [Martin D. Weinberg]
- Remove debugging line. [Martin D. Weinberg]
- Merge branch 'Eigen' of https://bitbucket.org/mdweinberg/exp into
Eigen. [mdw]
- Merge branch 'Eigen' of https://bitbucket.org/mdweinberg/exp into
Eigen. [Martin D. Weinberg]
- Cleanup in forcetest2. [michael_petersen]
- Update for Trace mean-mass ICs. [Martin D. Weinberg]
- Merge branch 'Eigen' of https://bitbucket.org/mdweinberg/exp into
Eigen. [Martin D. Weinberg]
- Merge of current version, replacing test code. [Martin D. Weinberg]
- Make pair count consistent with detailed balance. [mdw]
- Implemented NoCool option. [mdw]
- Restored the mean-mass option. [mdw]
- Added an explicit energy conservation algorithm for effective particle
masses; this will require changing the equilibrium IC conditions.
[mdw]
- Version update and some cruft removal. [mdw]
- Turn off cell count debugging output. [mdw]
- Additional fixes to algorithm implementation. [Martin D. Weinberg]
- More debugging output for new kinetic algorithm. [Martin D. Weinberg]
- Added main parameter parsing for double exponential conditioning
function. [Martin D. Weinberg]
- Added double exponential conditioning function. [Martin D. Weinberg]
- Fix n_p loops. [Martin D. Weinberg]
- Additional XC_DEEP* reporting. [Martin D. Weinberg]
- Additional XC_DEEP* reporting. [Martin D. Weinberg]
- Additions to XC_DEEP* output. [mdw]
- An additional improvement to Makefile.am for libtool building;
hopefully we will not need this much longer. [Martin D. Weinberg]
- Add restricted range to halocoef. [Martin D. Weinberg]
- Add restricted range to diskcoef. [Martin D. Weinberg]
- Promoted to Eigen; untested. [Martin D. Weinberg]
- Fixed some fencepost errors in DEBUG tests. [Martin D. Weinberg]
- Fixed some fencepost errors in test routine. [Martin D. Weinberg]
- Remove unused ECUT_DF from DiskHalo2.H. [Martin D. Weinberg]
- Add some sane defaults in localmpi.H so one node works without
initialization as a fall back. [Martin D. Weinberg]
- Remove unused ECUT_DF option. [Martin D. Weinberg]
- White-space ormatting changes only. [Martin D. Weinberg]
- Allow parallelization for numproc>NDP in DiskHalo2. Checked with
numproc=32 for NDP=16. [Martin D. Weinberg]
- Fix typo in diagnostic output. [Martin D. Weinberg]
- Removing default std namespace for consistency. [Martin D. Weinberg]
- Promote FDIST and FDISTC are classes to set default num values.
[Martin D. Weinberg]
- Added reporting of start position for >0 in epitable. [Martin D.
Weinberg]
- Improve epitable range checks. [mdw]
- Remove duplicated diagnostic output. [Martin D. Weinberg]
- Update mpi detection logic. [Martin D. Weinberg]
- Remove creation of shared libraries and add DSMC build. [Robert
Blackwell]
- Merge remote-tracking branch 'origin/Eigen' into cmake. [Robert
Blackwell]
- Cruft removal from SLGridMP2.cc; added Luan's latest UserAddMass
version; fixed fence-post errors in slcheck. [Martin D. Weinberg]
- Changed Eigen read/write strategy in SphereSL. [Martin D. Weinberg]
- Fence-post error fix in SphericalOrbit. [Martin D. Weinberg]
- Change Eigen view for testing; add some iostream checks. [Martin D.
Weinberg]
- Merged UserSat.cc. [mdw]
- Merge branch 'Eigen' of https://bitbucket.org/mdweinberg/exp into
Eigen. [mdw]
- Some additional UserSat tests. [Martin D. Weinberg]
- Add satellite mass column to UserSat log file. [mdw]
- Switch MatrixXd to Matrix3d for Euler rotations. [mdw]
- Missing index increment. [Martin D. Weinberg]
- Implementation for a more sane rmin, rmax logic. [Martin D. Weinberg]
- Fix the SLGridSph signature confusion caused by Analysis routines
using the old constructor calls; enforce grid radii in SphereSL.
[Martin D. Weinberg]
- Remove override for modelfile name. [Martin D. Weinberg]
- Change default name from mssaprof to haloprof. [Martin D. Weinberg]
- Remove control cruft. [Martin D. Weinberg]
- Missing dimensions in center variable. [mdw]
- Fix fence-post error in frequency access. [mdw]
- Cuda implementation of UserSat; comments. [Martin D. Weinberg]
- Preliminary implementation of UserSat for cuda. [Martin D. Weinberg]
- Fixed a fence-post error in the 3-vector. [Martin D. Weinberg]
- Fixed fence-post errors and a bunch to list counting error. [Martin D.
Weinberg]
- Comments only. [Martin D. Weinberg]
- Need to clear particle vector before add_particle call. [Martin D.
Weinberg]
- Add radial offsets for fmax table. [Martin D. Weinberg]
- Added buffered block writes to Particle to speed up PSQ. [Martin D.
Weinberg]
- Forgot to reset new counters. [Martin D. Weinberg]
- Added the YAML version of coefficient dumps to SphereSL. [Martin D.
Weinberg]
- Changed default halo particle type to 'dark' [Martin D. Weinberg]
- Code clean-up only; no functional changes. [Martin D. Weinberg]
- Added a reconstruction check on halo fields. [Martin D. Weinberg]
- Fix fence-post error in SphereSL leftover from Eigen conversion; fix
default file constructor for PSPout and PSPspl. [Martin D. Weinberg]
- Fix missing default limit flags for orders. [Martin D. Weinberg]
- Remove init index cruft from haloprof2. [Martin D. Weinberg]
- Add main 'exp' library to cmake. [Robert Blackwell]
- Add CUDA CMake build support for exputil. [Robert Blackwell]
- Add fortran as a cmake project language. [Robert Blackwell]
- Add lots of dependency checks to cmake and generate config.h. [Robert
Blackwell]
- Add cmake skeleton and rudimentary build of exputil. [Robert
Blackwell]
- Add yaml-cpp submodule. [Robert Blackwell]
- Added filePrefix specification to the fileNameCreator instances in
diskprof and haloprof. [Martin D. Weinberg]
- Merge branch 'Eigen' of https://bitbucket.org/mdweinberg/exp into
Eigen. [Martin D. Weinberg]
- Missing removal of PSP.H from pcatest in Analysis. [mdw]
- Problem in Eigen binary writes using .data() for some reason;
reverting to original method. [Martin D. Weinberg]
- Fix YAML difference parser. [Martin D. Weinberg]
- Add some error handing on file failure to diskprof and haloprof.
[Martin D. Weinberg]
- Remove lingering reference to localmpi.cc. [mdw]
- Remove the copy of header.cc from src; this kludge is no longer
necessary. [mdw]
- PSP now needs header.H from src so add -I flag in utils/PhaseSpace.
[mdw]
- Change DSMC libraries to shared for consistency with libexp. [Martin
D. Weinberg]
- Configure DSMC by default for testing. [Martin D. Weinberg]
- More fixes for standalone consistency with main EXP libraries. [Martin
D. Weinberg]
- Try to get Makefile.am order for shared library correct. [Martin D.
Weinberg]
- More clean up for standalone integration. [Martin D. Weinberg]
- MPI synchronization fix. [Martin D. Weinberg]
- Mistakenly used a 6 rather than 5 character index filed for PSP.
Fixed default directories in haloprof2.cc and diskprof2.cc. [Martin D.
Weinberg]
- Preliminary implementation of a general phase-space-reading class.
[Martin D. Weinberg]
Still needs lots of work. PSP is fully implemented. Gadget HDF5
seems to work but does not include all of the essential features, such
as reading multiple files per snapshot, in particular.
- Added Iterable class. [Martin D. Weinberg]
- Just some formatting clean up. [Martin D. Weinberg]
- Reorganize EXP executable into a driver and exp library. [Martin D.
Weinberg]
- Remove all C limits cruft and use C++ limits instead. [Martin D.
Weinberg]
- Deprecate all but one variant of haloprof and diskprof; add optional
radial order (and harmonic order) rangle restrictions on field
evaluation. [Martin D. Weinberg]
- Turn off verbose coefficient diagnostics. [mdw]
- Fence post error in GPU table assignment fixed. [mdw]
- Added toggle for verbose coefficient printing in cuda. [mdw]
- Merge branch 'Eigen' of https://bitbucket.org/mdweinberg/exp into
Eigen. [mdw]
- Added library flags for boost.random, following Mike's sugggestion.
Mike: please check that this helps. [Martin D. Weinberg]
- Pushed coefficient debug checks to their own member function. [Martin
D. Weinberg]
- Merge branch 'Eigen' of https://bitbucket.org/mdweinberg/exp into
Eigen. [mdw]
- Add option for Chebyshev smoothing for Abel transform in DF. [Martin
D. Weinberg]
- Fencepost errors in cuda to cpu coefficient mapping. [Martin D.
Weinberg]
- Additional fixes and debug output for gendisk2. [Martin D. Weinberg]
- Merge branch 'Eigen' of https://bitbucket.org/mdweinberg/exp into
Eigen. [mdw]
- Allow for multiple smoothing types. [Martin D. Weinberg]
- Merge branch 'Eigen' of https://bitbucket.org/mdweinberg/exp into
Eigen. [mdw]
- Change default DF interpolation for testing. [Martin D. Weinberg]
- Merge branch 'Eigen' of https://bitbucket.org/mdweinberg/exp into
Eigen. [mdw]
- Another FP error and additional clean up. [Martin D. Weinberg]
- Merge branch 'Eigen' of https://bitbucket.org/mdweinberg/exp into
Eigen. [mdw]
- Restore global iteration counter. [Martin D. Weinberg]
- Merge branch 'Eigen' of https://bitbucket.org/mdweinberg/exp into
Eigen. [mdw]
- A few more fence post fixes; comment changes and range checks in
distf/dfde/d2fde2. [Martin D. Weinberg]
- Merge branch 'Eigen' of https://bitbucket.org/mdweinberg/exp into
Eigen. [mdw]
- Changes to MPI for Eigen objects. [Martin D. Weinberg]
- Additional spline checks. [Martin D. Weinberg]
- Commented out deep debug. [Martin D. Weinberg]
- Merge branch 'Eigen' of https://bitbucket.org/mdweinberg/exp into
Eigen. [mdw]
- One additional verbose output removal. [Martin D. Weinberg]
- Fence post fixes. [Martin D. Weinberg]
- Additional fence post fixes. [mdw]
- Remove verbose output. [Martin D. Weinberg]
- Fix fencepost errors in cylindrical basis construction. [Martin D.
Weinberg]
- Make the eigtest routine work using Eigen. [Martin D. Weinberg]
- Yet more fencepost error fixes. [Martin D. Weinberg]
- More fencepost error fixes. [Martin D. Weinberg]
- More minor fixes for random_gen, localmpi, etc. [Martin D. Weinberg]
- Still more updates for boost.random and replace PSP2.H by PSP.H. [mdw]
- More updates for boost.random. [mdw]
- Additional fixed for boost.random. [Martin Weinberg]
- Missing commit of separate implemenation for boost.special. [Martin D.
Weinberg]
- More fence-post error fixes. [Martin D. Weinberg]
- Added the global instance of the boost.random generator for stand-
alone drivers. These global variables should go into a header, I
suppose. [Martin D. Weinberg]
- Removed old GNU random classes and replaced them with boost.random.
Removed the expgnu library. [Martin D. Weinberg]
- Some fencepost error fixes. [Martin D. Weinberg]
- Version bump. [Martin D. Weinberg]
- Minor CUDA update for Eigen::Vector3d. [Martin D. Weinberg]
- Added complete set of files for Vector/Matrix --> Eigen translation.
[Martin D. Weinberg]
This code fully compiles without error but almost certainly does NOT work.
- Additional updates. [Martin D. Weinberg]
- Vector/Matrix->Eigen changes that I missed on the first round
- Some additional .h to .H and cruft removal
- Removed old Vector/Matrix classes along with a pile of other bad
cruft. [Martin D. Weinberg]
The code compiles from start to finish but I almost certainly have
introduced a variety of bugs and problems. This will take some time
to test, verify, and sort out.
- Corrected electron scattering mass in accum-trace-scatter algorithm.
[Martin D. Weinberg]
- Added stop signal for off-grid interpolation in playback. [Martin D.
Weinberg]
- Make YAML headers in spherical basis coefficient files the default.
[Martin D. Weinberg]
- Failsafe to prevent an infinite loop. [Martin D. Weinberg]
- Merge branch 'devel' of https://bitbucket.org/mdweinberg/exp into
devel. [Martin D. Weinberg]
- Need strncpy for coefficient file read. [mdw]
- Minor fixes for playback on GPU side. [Martin D. Weinberg]
- Needed <cstring> for strncpy function. [Martin D. Weinberg]
- Allow for both old and new cylindrical coefficient header types in
read for playback. [Martin D. Weinberg]
- Some fixes for playback. [Martin D. Weinberg]
- Check for eof when reading coefficieints. [Martin D. Weinberg]
- Merge branch 'devel' of https://bitbucket.org/mdweinberg/exp into
devel. [Martin D. Weinberg]
- Reference 'comp' should be 'tcomp' [mdw]
- Don't use COV frame for KE for 1 particle components. [Martin D.
Weinberg]
- Fix force references. [Martin D. Weinberg]
- Merge branch 'devel' of https://bitbucket.org/mdweinberg/exp into
devel. [Martin D. Weinberg]
- More commenting. [Martin D. Weinberg]
- More cudaAware overrides to prevent GPU device fetches when there are
not wanted. [Martin D. Weinberg]
- Addition cudaAware restriction for force evaluation. [Martin D.
Weinberg]
- Version bump. [mdw]
- Move particles to cuda for level computation when force is not cuda
aware. [mdw]
- An additional check for whether cuda is used for all interacting
components. [mdw]
- Simplify logic for interactions with cuda. [mdw]
- Revert ParticleToCuda and CudaToParticle restrictions. [mdw]
- Spelling change in comments only. [Martin D. Weinberg]
- Draft suppression of CPU<-->GPU particle exchange for no native
implementation. [Martin D. Weinberg]
- Comments only. [Martin D. Weinberg]
- Implemented the Miyamoto-Nagai particles in the Point class; ifxed a
parameter swap in the UserMNdisk routine (oops) [Martin D. Weinberg]
- Removed uninitialized variables from UserAddMass as a test. Luan may
wish to study and implement the changes. Not that I suspect any
problems here to be honest. [Martin D. Weinberg]
- Bumped up the default DATTRIB_CUDA value from 4 to 12 for H+He DSMC
tests. [mdw]
- Removed duplicated storage allocation check for variance computation;
not hurting anything but unnecessary. [mdw]
- Fixed uninitialized pca counter variables which was leading to
failure. Oops. Not sure how those/when the default initialization got
dropped. [mdw]
- Clear use list. [mdw]
- Merge branch 'devel' of https://bitbucket.org/mdweinberg/exp into
devel. [Martin D. Weinberg]
- Some formatting for clarity and three 'breaks' changed to 'continues'
[mdw]
- Change default output prefix. [Martin D. Weinberg]
- Toggle off some verbose debugging. [Martin D. Weinberg]
- Update for cuda compilation but use_cuda=false. [Martin D. Weinberg]
- Check for sampT allocation. [Martin D. Weinberg]
- More sanity checks for coef2. [mdw]
- Cherry picked from 'massevo' branch. [mdw]
- Sanity check on reduction in coef2 for sampT>0. [mdw]
- Zero use counter before fork. [Martin D. Weinberg]
- Fix flag for pcavar storage initialization. [Martin D. Weinberg]
- Upgraded error messages. [Martin D. Weinberg]
- Merge branch 'devel' of https://bitbucket.org/mdweinberg/exp into
devel. [Martin D. Weinberg]
- Make sampT=1 the default to match defSampT in AxisymmetricBasis and
EmpCylSL. [mdw]
- Merge branch 'devel' of https://bitbucket.org/mdweinberg/exp into
devel. [Martin D. Weinberg]
- Attempt to accommodate the older HDF5 API. [Martin Weinberg]
- Remove unneccessary count field from vtk file names. [Martin D.
Weinberg]
- Add some file exception handling to Coef readers. [Martin D. Weinberg]
- Add new simple routines for producing VTK profiles from coefficient
files. These still need testing. [Martin D. Weinberg]
- Change name from Coefs to CylCoefs for syntatic symmetry. [Martin D.
Weinberg]
- Minor changes for additional diagnostics. [Martin D. Weinberg]
- Remove test variance trim. [Martin D. Weinberg]
- Additional diagnostic fields. [Martin D. Weinberg]
- Fix mutex unlock misordered line. [Martin D. Weinberg]
- Improve center component output formatting. [Martin D. Weinberg]
- Give the user some feedback for a center override. [Martin D.
Weinberg]
- Added expansion and reconstruction center options. I haven't check
this, so le t me know if there are any problems. [Martin D. Weinberg]
- Allow all m values to be summed in trim. [Martin D. Weinberg]
- Change test array normalization. [Martin D. Weinberg]
- Change default on snr reorder. [Martin D. Weinberg]
- Add |evec|^2 to output. [Martin D. Weinberg]
- Merge branch 'devel' of https://bitbucket.org/mdweinberg/exp into
devel. [Martin D. Weinberg]
- Set defSampT unconditionally. [mdw]
- Improve warning message for T>=sampT. [mdw]
- Documentation changes only. [Martin D. Weinberg]
- Set default variance sample size to 1 for Cuda efficiency. [Martin D.
Weinberg]
- Version change. [Martin D. Weinberg]
- Additional comments. [Martin D. Weinberg]
- Made CPU and GPU versions consistent; additional comments. [Martin D.
Weinberg]
- Remove debug chatter; some additional comments. [Martin D. Weinberg]
- Removed T_coef in favor of T_covr for both sample and mean variance
computation; fixed typo kernel for variance computation. [Martin D.
Weinberg]
- Removed T_coef in favor of T_covr for both sample and mean variance
computation. [Martin D. Weinberg]
- Added a positive variance sanity check; simplifed normalization of
variance. [Martin D. Weinberg]
- A few more minor updates to variance tapering; checked cuda code
successfully with both versions 11.1 and 11.3. [Martin D. Weinberg]
- Updates for internal variance smoothing computation. [Martin D.
Weinberg]
- Move exp_version() back into parse so that timestamp is kept up to
date. [Martin D. Weinberg]
- Some clean up for variance code; change variance to full population
rather than 1-particle variance for consistency. [Martin D. Weinberg]
- Fix mistaken restriction introduced in multistep update. [Martin D.
Weinberg]
- Undefined default value for subsampling. [Martin D. Weinberg]
- Merge branch 'devel' of https://bitbucket.org/mdweinberg/exp into
devel. [mdw]
- Some updates for variance trimming. [Martin D. Weinberg]
- Convert bootstrap to report 1-particle covariance using CLT scaling.
[Martin D. Weinberg]
- Add conversion to 1-particle covariance. [mdw]
- Merge branch 'devel' of https://bitbucket.org/mdweinberg/exp into
devel. [Martin D. Weinberg]
- Fix norm constants. [mdw]
- Added missing new line. [Martin D. Weinberg]
- Added density-only cross validation for Sphere. [Martin D. Weinberg]
- Corrected 4pi factors. [Martin D. Weinberg]
- Propagate user SEED to AxiSymModel. [Martin D. Weinberg]
- Merge branch 'devel' of https://bitbucket.org/mdweinberg/exp into
devel. [mdw]
- Added orthgonal slice options for field output. [Martin D. Weinberg]
- Remove Lmax>=1 restriction. [Martin D. Weinberg]
- Allow mmax=0 evaluation. [Martin D. Weinberg]
- Allow rank=1 vectors and matricies. [Martin D. Weinberg]
- Updated cross-validation code for weighted trunction. [mdw]
- Add version of halo coefficient reader that prints out cosine and sine
coefficients together. [mdw]
- Change for mass norm in get_trimmed. [mdw]
- Change default number of eigenvalues; add total power diagnostics.
[mdw]
- Merge branch 'devel' of https://bitbucket.org/mdweinberg/exp into
devel. [mdw]
- Restriction so that lmax=0 does not fail. [Martin D. Weinberg]
- Merge branch 'devel' of https://bitbucket.org/mdweinberg/exp into
devel. [Martin D. Weinberg]
- Merge branch 'devel' of https://bitbucket.org/mdweinberg/exp into
devel. [Martin D. Weinberg]
- Documentation only. [Martin D. Weinberg]
- Remove variance of mean S/N scaling and leave pure sample variance S/N
instead. [mdw]
- Add radial order cut experiment; leaving the code but this does not
seem promising. [mdw]
- Scale from standard covariance to match bootstrap covariance for
testing. [mdw]
- Allow user to modify sampT from the sqrt(N) default. [mdw]
- Add power diagnostic in trim. [Martin D. Weinberg]
- Updates for mass and variance scaling. [mdw]
- Generalize bootstrap variance scaling for arbitrary partitioning, not
only sqrt. [mdw]
- Clean up norms for SNR smoothing. [mdw]
- Clean up norms for SNR smoothing. [mdw]
- Added toggle for verbose covariance output. [mdw]
- Remove debug and add L1 & L2 restriction to surface reconstruction.
[mdw]
- Add differenced images for SNR trim. [Martin D. Weinberg]
- Removed unused variable. [Martin D. Weinberg]
- Checks okay; remove debugging output. [mdw]
- Add phi averaging and parallelization for evaluation loop. [mdw]
- Set nodd and neven to zero when not set; doesn't matter for anything.
[mdw]
- New style YAML header caches are now default. [mdw]
- Merge branch 'devel' of https://bitbucket.org/mdweinberg/exp into
devel. [mdw]
- Remove covariance computation. [Martin D. Weinberg]
- Merge branch 'devel' of https://bitbucket.org/mdweinberg/exp into
devel. [mdw]
- Added EmpCylSL expansion for testing. [Martin D. Weinberg]
- Merge branch 'devel' of https://bitbucket.org/mdweinberg/exp into
devel. [mdw]
- Test k-d nn density map. [Martin D. Weinberg]
- Updates for k-d NN density estimate. [Martin D. Weinberg]
- Add mass/weightt accumulation to nearestN in k-d interface. [Martin D.
Weinberg]
- Minor updates and testing code. [mdw]
- Use ball density estimate by default. [Martin D. Weinberg]
- Added local density estimation using a k-d nearest N neighbor search.
[Martin D. Weinberg]
- Updates for KL bootstrap amplitude. [Martin D. Weinberg]
- Updated versions of KL analysis. [Martin D. Weinberg]
- Fix index overwrite detected by nvcc 11.3. [Martin D. Weinberg]
- Include the spherical harmonic prefactor in dump_coefs output to
simply power computation. [Martin D. Weinberg]
The YAML stanza will now contain the {normed: true} tag to indicate this to readers
- Name change in default files. [Martin D. Weinberg]
- Missing check for SPL in init phase space. [Martin D. Weinberg]
- Use com0 for local centering in the cuda Orient implementation for
consistency with the cpu implementation. [mdw]
- Noticed typo in nsiz computation which ended up dividing bunch size by
numprocs twice; it should not matter in practice. [mdw]
- Merge branch 'CUDA_DEVEL' into devel. [mdw]
Devel is currently deprecated owing the many fixes to both timestepping and
cuda implementation. After the merge, CUDA_DEVEL will be discarded or
repurposed.
- Fix restriction of grid indices in the coordinate kernel which was
truncating at NUMX-1 and NUMY-1 rather than NUMX and NUMY. [mdw]
- Get true grid radius directly from EmpCylSL; clean up mixture factor
application for future use; move interpolation constants of loop and
designate them constexpr. [mdw]
- Add member function to get Rtable size. [mdw]
- Add debug off-grid warning. [mdw]
- Get true grid radius directly from EmpCylSL; clean up mixture factor
application for future use; move interpolation constants of loop and
designate them constexpr. [mdw]
- Just cosmetic white space. [mdw]
- Floor timestep as done in CPU version; no evidence that this has
effected anything so far . . . [mdw]
- Fixed to centering algorithm on GPU. [mdw]
- Use regime interpolation and potential value. [mdw]
- Version bump for ID. [mdw]
- Make vertical quantile histogram consistently oriented with density
and surface histograms. [mdw]
- Left justify header. [mdw]
- -v output is now the default for -h, so consolidate. [mdw]
- Put GPU device constant reporting at VERBOSE>4; I'm tired of seeing
the clutter in the stdout. [mdw]
- Also change PID info stanza printed at top of log for cosmetic
consistency. [mdw]
- Minor typo fix. [mdw]
- Cosmetic changes to stdout EXP intialization for consistency; while
I'm waiting . . . [mdw]
- Added an additional debugging field for checking; checks okay and
disabled; updated for mdrft. [mdw]
- Added an additional debugging field for checking; checks okay and
disabled. [mdw]
- Comment the change from diagnostic output at mstep==0 to
mstep+1==Mstep. [Martin D. Weinberg]
- Simply adjust call logic and put reporting at mstep=Mstep rather than
mstep=0. [Martin D. Weinberg]
- Made a new variable for mstep+1 to help with notation. [Martin D.
Weinberg]
- Fix multistep for this method. [Martin D. Weinberg]
- Swap option parsing with EXP welcome header. [Martin D. Weinberg]
- Version bump. [Martin D. Weinberg]
- Make percentages in timer output more readable using boost::format
(until C++20 std::format is available) [Martin D. Weinberg]
- Cleaned up notation for computing coefficients on drifted step.
[Martin D. Weinberg]
- Merge branch 'CUDA_DEVEL' of https://bitbucket.org/mdweinberg/exp into
CUDA_DEVEL. [mdw]
- Minor correction to mainpage document. [Martin D. Weinberg]
- Fix +1 fencepost error; this explains why my interpolation had a
pattern with mstep. [mdw]
- Added some additional debug output only. [Martin D. Weinberg]
- Removed 'tstp'; added automatic branch, commit, and compile time to
stdout. [mdw]
- Minor changes for clarity: use component rather than cC for orient and
center vector; use mfirst[mstep] rather than passed mlevel value.
[mdw]
- Additional comment. [mdw]
- More comments; adjusted mstep for final kick. [mdw]
- After a bunch more algorithm checking, I believe that this is now
implemented as intended. The major changes are fixing the drifted
step coefficient interpolation. The original step.cc implementation
was correct, just a bit confusing in retrospect since the positions
have drifted past the velocities so one has to be careful to keep
track of that in the multstep coefficient interpolation. In
particular: the position in the interpolation is "mstep+1" not
"mstep". Added a bit more clarification in the comments. The
evolution still need to be testing and compared. [mdw]
- Return to the original step strategy but put the coefficient
interpolation offset into the coefficient interpolation code, rather
than step.cc. [mdw]
- Clean up the comments only. [mdw]
- Proposed fencepost error fix for multistep algorithm. Also added some
additional comments about what's going on in this implementation.
[Martin D. Weinberg]
- Fixed the weird multistep issue; downward level changes need to be
limited to the minimum active level or coefficient updates will be
discarded and only included at the next computation on the target
particle's new level. This is broken in both the CUDA and CPU code.
I doubt it matter much except under usual circumstances, such as very
large values of multistep. [mdw]
- Attempt to remove deprecated PhaseSpace class utilities. [Martin D.
Weinberg]
- Update diskprof2 for PSP2 (OUT & SPL) input. [Martin D. Weinberg]
- Formatting fix to timer output. [mdw]
- Add system time to step timer output. [mdw]
- Changed 'samplesize' to 'samplesz' for compactness; added some header
documentation. [Martin D. Weinberg]
- Override sqrt(N) subsample size. [Martin D. Weinberg]
- Fixes to variance computation. [Martin D. Weinberg]
- Restored 'printf' lines for deep debugging but unused by default.
[Martin D. Weinberg]
- Added a flattened covariance matrix to Cuda for noise computation.
[Martin D. Weinberg]
- Add helper member to get variance matrix from Cuda. [Martin D.
Weinberg]
- Enforce nintsub >=0 to prevent user from causing divide_by_zero.
[Martin D. Weinberg]
- Merge branch 'CUDA_DEVEL' of https://bitbucket.org/mdweinberg/exp into
CUDA_DEVEL. [Martin D. Weinberg]
- Okay, found the source of the CUDA multilevel implementation at long
last! [mdw]
Sigh. That took a really long time.
o Embarassingly, it was a typo: the multistep short range needed to be
[mlevel, mlevel] rather than [mlevel, multistep]. I think I had that
last range in my head from the coefficient implementation.
o This deserves a version bump. And the whole thing needs to be
tested now.
- Remove the cuda override debugging which is cumbersome and not much
help anyhow. [mdw]
- Some extra comments only. [mdw]
- Merge branch 'CUDA_DEVEL' of https://bitbucket.org/mdweinberg/exp into
CUDA_DEVEL. [Martin D. Weinberg]
- Override coefficient and accelration for testing based on a boolean
toggle. [mdw]
- Comment changes only. [mdw]
- Remove commented cruft for clarity. [mdw]
- Merge branch 'CUDA_DEVEL' of https://bitbucket.org/mdweinberg/exp into
CUDA_DEVEL. [Martin D. Weinberg]
- Replace the 'compute' argument with 'false' in the coefficient
adjustment kernel call. [Martin D. Weinberg]
- Replace the 'compute' argument with 'false' in the coefficient
adjustment kernel call. [Martin D. Weinberg]
- Use the native operator< rather than custom pairLess functor. [Martin
D. Weinberg]
- Merge branch 'CUDA_DEVEL' of https://bitbucket.org/mdweinberg/exp into
CUDA_DEVEL. [Martin D. Weinberg]
- Change routine name to cuda_zero_coefs() [Martin D. Weinberg]
- Change routine name to cuda_zero_coefs() [Martin D. Weinberg]
- Removed old comment. [Martin D. Weinberg]
- A few more error checks for insurance. [Martin D. Weinberg]
- Fixes for compilation without cuda enviroment. [Martin D. Weinberg]
- Check that npart < I._s not P._s, although I._s should equal P._s, so
it should not matter. [Martin D. Weinberg]
- Remove blank source file; some minor clean ups. [Martin D. Weinberg]
- Routine name changes to prevent me from getting confused. [mdw]
- Test coefficient zeroing logic. [Martin D. Weinberg]
- Some additional comments only. [Martin D. Weinberg]
- Comments only. [Martin D. Weinberg]
- Use /usr/bin/env to find python executable in a module environment.
[mdw]
- Forget to commit this source file. [Martin D. Weinberg]
- Merge branch 'CUDA_DEVEL' of https://bitbucket.org/mdweinberg/exp into
CUDA_DEVEL. [Martin D. Weinberg]
- Merge branch 'CUDA_DEVEL' of https://bitbucket.org/mdweinberg/exp into
CUDA_DEVEL. [mdw]
- Attempt to fix compilation problems on Unity by adding cuda support to
libexputil. [mdw]
- Merge branch 'CUDA_DEVEL' of https://bitbucket.org/mdweinberg/exp into
CUDA_DEVEL. [Martin D. Weinberg]
- Put Emacs mode tag at top of file for detection. [Martin D. Weinberg]
- Implemented Orient on the GPU device. [Martin D. Weinberg]
- Try to incorporate CPU<-->GPU fetch logic. [Martin D. Weinberg]
- Change preprocessor config flag for consistency. [Martin D. Weinberg]
- Range fix for level indicies. [Martin D. Weinberg]
- Name change for routine only. [Martin D. Weinberg]
- Fix for move of sledge.f. [Martin D. Weinberg]
- Reorganize names of fix_position routines. [Martin D. Weinberg]
- Move sledge.f to exputil. [Martin D. Weinberg]
- Clean fix_positions organization. [Martin D. Weinberg]
- Minor clean up; still need to cudaize orient. [Martin D. Weinberg]
- Reverted; there was no gaff in device copy; try kernel updates with
pointer rather than reference. [Martin D. Weinberg]
- Comments only. [Martin D. Weinberg]
- Only update particle levels at or above current level; won't save much
time but . . . [Martin D. Weinberg]
- Gaff in host to dev copy sequence fixed. [Martin D. Weinberg]
- Only need to compute difference matrices for mfirst[mstep] since other
changes are discarded until multistepping hits the lower particle
level. [Martin D. Weinberg]
- Don't call CudaGetLevelRange for multistep=0; no big deal; should save
a tiny chuck of time. [Martin D. Weinberg]
- Remove commented lines. [Martin D. Weinberg]
- Additional comment only. [Martin D. Weinberg]
- Additional tweaks to UserTestCuda. [Martin D. Weinberg]
- Restore the mlevel restrictions in coefficient updates to test the old
code. [Martin D. Weinberg]
- Get a little fancier with output control and parameters. [Martin D.
Weinberg]
- User test routine that implements a CUDA kernel. [Martin D. Weinberg]
- Documentation version bump. [Martin D. Weinberg]
- Change default double attributes from 12 to 4. [Martin D. Weinberg]
- Use local header reads to prevent confusion. [Martin D. Weinberg]
- Clean up fractional time for easier to read output. [Martin D.
Weinberg]
- New test implementation using various CUDA constructs to allow or
coefficient update on the device. [Martin D. Weinberg]
- Version bump
- The adjust step is no longer the bottle neck
- Version should be algorithmically identical to the original version
- Still needs work to compare with original version
- Speed up is a factor 20x over the original version with a single node test
- Yet more comments. [Martin D. Weinberg]
- Additional comments only. [Martin D. Weinberg]
- Allow host and device calls in structure for future use. [Martin D.
Weinberg]
- Variable name change for clarity. [Martin D. Weinberg]
- Update for index indirection. [Martin D. Weinberg]
- Need to use kernel rather than thrust for indirection. [Martin D.
Weinberg]
- Variable name change for clarity. [Martin D. Weinberg]
- Replace multiple thrust::reduce with a single cuda kernel in
cuda_fix_positions. [Martin D. Weinberg]
- Removed cruft. [Martin D. Weinberg]
- Implementation of indirect indexing. [Martin D. Weinberg]
- Added relative fractions to step timing. [Martin D. Weinberg]
- Fix typo in dynA computation. [Martin D. Weinberg]
- Additional timing diagnostics; use thrust rather than kernel for level
update (doesn't matter much) [Martin D. Weinberg]
- Code clean-up: use reference to cudaParticle and not pointer
(equilvalent code but easier to read) [Martin D. Weinberg]
- Remove restriction in differ list to >=mfirst[mstep] [Martin D.
Weinberg]
- Remove restriction in differ list to >=mfirst[mstep] [Martin D.
Weinberg]
- Do not check levlist when using cuda. [Martin D. Weinberg]
- Remove redundant particle copies. [Martin D. Weinberg]
- Something strange about limiting coefficient difference matrix to
mfirst[mstep] and above; this commit changes that which seems to be
what I originally intended (???) [Martin D. Weinberg]
- Go back to per component stream selection. [Martin D. Weinberg]
- Remove test timer. [Martin D. Weinberg]
- Missing check for use_cuda in ParticlesToCuda. [Martin D. Weinberg]
- Use differ matrices. [Martin D. Weinberg]
- A few additional time stepping tests; implemented level output from
GPU; something is still broken in acceleration or updates. [Martin D.
Weinberg]
- Move CUDA device context assignment to the top level; it doesn't make
sense for Component to assign context. Allow component to query and
cache device #. [Martin D. Weinberg]
- Updates for level sort and range split. [Martin D. Weinberg]
- Merge branch 'CUDA_DEVEL' of https://bitbucket.org/mdweinberg/exp into
CUDA_DEVEL. [Martin D. Weinberg]
- Split sort and range finding for levels. [Martin D. Weinberg]
- New routines for computing multistep time steps on the CUDA device.
[Martin D. Weinberg]
- Constant printing change only. [Martin D. Weinberg]
- Fixed mistake in coordinate addressing; only affected diagnostic
output. [Martin D. Weinberg]
- Fix cuda-free compilation. [Martin D. Weinberg]
- Save some updates including a preliminary implementation of the update
step; needs more debugging. [Martin D. Weinberg]
- Preliminary start on adjust step, but this version is broken. [Martin
D. Weinberg]
- Partially working implementation with multistep=0 only. [Martin D.
Weinberg]
- Add particle use counter correctly for cudaCylinder. [Martin D.
Weinberg]
- Partial implementation of a CUDA-only EXP. [Martin D. Weinberg]
- Begin implemented an ondevice EXP version. [Martin D. Weinberg]
- Use /usr/bin/env to find python executable in a module environment.
[mdw]
- Test for CUDA implementation of DSMC with CPU implementation. [Martin
D. Weinberg]
- Compliing implementation of CUDA Trace algorithm; still needs testing.
[Martin D. Weinberg]
- Merge branch 'devel' of https://bitbucket.org/mdweinberg/exp into
devel. [Martin D. Weinberg]
- Add a cuda mem copy timer diagnostic. [Martin D. Weinberg]
- Updates to CUDA trace DSMC implementation. Note to self: still will
not work so do not use it. [Martin D. Weinberg]
- Documentation changes only. [Martin D. Weinberg]
- Prevent overflow and underflow in tau computation by checking for
large tau value first, and adding intermediate check for minimum tau
value. [mdw]
- Fixes for energy conservation in Coulombic scattering routine. [Martin
D. Weinberg]
- Comments only. [Martin D. Weinberg]
- Add default python path back into libtool script. [Martin D. Weinberg]
- Merge branch 'devel' of https://bitbucket.org/mdweinberg/exp into
devel. [Martin D. Weinberg]
- Improve pair selection in accumTraceScatter. [mdw]
- Zero charge check for mean charge computation. [mdw]
- Comments and minor updates. [Martin D. Weinberg]
- Energy bookkeeping update for recombination using newRecombAlg.
[Martin D. Weinberg]
- Fix GauntFF database initialization error. [Martin D. Weinberg]
- Fix diagnostic exit message. [Martin D. Weinberg]
- Merge branch 'devel' of https://bitbucket.org/mdweinberg/exp into
devel. [Martin D. Weinberg]
- Added rejected fraction to output file. [mdw]
- Add sanity check for dimension of PNG diagnostic output. [mdw]
- Fix output formating and a few mistakes in KL bunch assignment. [mdw]
- Fix minSNR lower bound check and trimmed coefficient cacheing. [Martin
D. Weinberg]
- Fix minSNR lower bound check. [Martin D. Weinberg]
- Fix minSNR lower bound check. [Martin D. Weinberg]
- Changes to progress labeling only. [Martin D. Weinberg]
- Draft implementation for KL analysis. [Martin D. Weinberg]
- Preliminary KL anaysis routine for Cylinder. [Martin D. Weinberg]
- Merge branch 'devel' of https://bitbucket.org/mdweinberg/exp into
devel. [Martin D. Weinberg]
- Remove a debug print statement. [mdw]
- Changes to allow SNR cuts for diskprof2 visualization. [mdw]
- Changes for external force time in multistepping. [Martin D. Weinberg]
- Minor fixes. [Martin D. Weinberg]
- Minor correction to coefficient trimming logic in EmpCylSL. [Martin D.
Weinberg]
- Used preprocessor to remove nintsub option by default. [Martin D.
Weinberg]
- Merge branch 'devel' of https://bitbucket.org/mdweinberg/exp into
devel. [Martin D. Weinberg]
- White space correction only. [Martin D. Weinberg]
- Version bump. [Martin D. Weinberg]
- Need to call output->Run(n) at end of multistep ladder. [Martin D.
Weinberg]
- Put output->Run at the beginning of the step. [Martin D. Weinberg]
- Fix ordering of time step increment and Output->Run call to get output
at the desired time step. [Martin D. Weinberg]
- Fix ordering of time step increment and Output->Run call to get output
at the desired time step. [Martin D. Weinberg]
- Whitespace and comments only. [Martin D. Weinberg]
- Whitespace and comments only. [Martin D. Weinberg]
- Merge branch 'devel' of https://bitbucket.org/mdweinberg/exp into
devel. [Martin D. Weinberg]
- Cleaned up MPI exits; added Hall shape exponent to the spherical
method. [Martin D. Weinberg]
- A few fixes for pca implementation. Still needs work so should not
yet be used. [Martin D. Weinberg]
- CuStream ring is causing trouble in the CUDA pca reduction. I
probably need some additional synchronization or perhaps I'm not using
streams correctly. Set to 1 as a work around (which seems to work).
Anyway, the pca cuda implementation is still unfinished and should not
be used for production. [Martin D. Weinberg]
- Formatting clean up only. [Martin D. Weinberg]
- Fix broken reporting of Hall method; doesn't change results, just log
file. [Martin D. Weinberg]
- Toggle on verbose output for spherical grid parameters in cuda.
[Martin D. Weinberg]
- Tag debug output with DEBUG as a reminder. [Martin D. Weinberg]
- Minor clean up to cross validation stuff. [Martin D. Weinberg]
- Merge branch 'devel' of https://bitbucket.org/mdweinberg/exp into
devel. [Martin D. Weinberg]
- Merge branch 'devel' of https://bitbucket.org/mdweinberg/exp into
devel. [mdw]
- Not sure what happened, but this did not stick the first time. [mdw]
- Include eigen3 headers in nvcc compile flags. [Martin D. Weinberg]
- Remove thread number output test. [Martin D. Weinberg]
- Merge branch 'devel' of https://bitbucket.org/mdweinberg/exp into
devel. [Martin D. Weinberg]
- The intended Thrust API works as of CUDA 11.1; adjusted workaround
logic accordingly. Fixed up a gaff in coefficient sizing introduced
as I started to implement EOF on the CUDA side. Currently testing but
good so far. [Martin D. Weinberg]
- Update draft implementation of EOF covariance for consistency to fix
indexing; additional output simplification and comments. [Martin D.
Weinberg]
- Comments, white space, name changes only. [Martin D. Weinberg]
- Changed class name chdata to new name atomicData in cuda members.
[Martin D. Weinberg]
- Clean up to output signaling for OutPS* classes and DSMC updates.
[Martin D. Weinberg]
- Test weighted scheme for Coulombic interactions. [Martin D. Weinberg]
- Fix weighting scheme for low ionization fraction. [Martin D. Weinberg]
- Fixed to charge and rate computation. [Martin D. Weinberg]
- Merge branch 'devel' of https://bitbucket.org/mdweinberg/exp into
devel. [Martin D. Weinberg]
- Comments only. [Martin D. Weinberg]
- Restore level computation at substeps; fix for doc dir scan in
configure.ac. [Martin D. Weinberg]
- Fixed typo in Ei/Ef ratio. [Martin D. Weinberg]
- A bit more clean up. [Martin D. Weinberg]
- Fixes to new Gaunt factor interpolation. [Martin D. Weinberg]
- Add van Hoof et al. Gaunt factor toggle. [Martin D. Weinberg]
- Unit fix in free-free cross secton. [Martin D. Weinberg]
- Implemented Karzas-Latter free-free cross section with modern Gaunt
factor approximation. [Martin D. Weinberg]
- Comment changes only. [Martin D. Weinberg]
- Added user settable NUMDF. [mdw]
- Add verbose output to VFLAG & 8. [mdw]
- Allow selection of gridded or ungridded cross sections. [Martin D.
Weinberg]
- Added quadrature to recombintation cross section to resolve
dielelectronic resonances. [Martin D. Weinberg]
- Added more comments. [Martin D. Weinberg]
- Remove ancient rcs id string. [Martin D. Weinberg]
- Fix MPI_Reduce misplacement! [Martin D. Weinberg]
- Formatting diffs only. [Martin D. Weinberg]
- Formatting diffs only. [Martin D. Weinberg]
- More chat suppression. [Martin D. Weinberg]
- Remove some of the insane verbosity in DiskHalo2 and initial2. [Martin
D. Weinberg]
- Version bump. [Martin D. Weinberg]
- More clean up for fudge factors and python calls. [Martin D. Weinberg]
- Add some exception handling for directory scan. [Martin D. Weinberg]
- Added some comments and corrections for energy units. Ratios now
check, but so far we are only exact for H and He. [Martin D. Weinberg]
- Added energy reweighting to bin center. [Martin D. Weinberg]
- Updates for ratio testing. [Martin D. Weinberg]
- Corrected file name convention. [Martin D. Weinberg]
- Merge branch 'devel' of https://bitbucket.org/mdweinberg/exp into
devel. [Martin D. Weinberg]
- Add runtag and prefix to coefficient output. [mdw]
- Allow a range of norders for reconstruction (e.g. even and odd) [mdw]
- Added preliminary test Badnell data reader. [Martin D. Weinberg]
- Allow colormap choice. [mdw]
- Added VTK output per rotated coefficient for diskeof. [Martin D.
Weinberg]
- Fixes for EOF conditioning tests. [mdw]
- Missing MPI reduction. [Martin D. Weinberg]
- Merge branch 'devel' of https://bitbucket.org/mdweinberg/exp into
devel. [Martin D. Weinberg]
- Fix restart logic. [mdw]
- Added EOF basis test for Cylinder. [Martin D. Weinberg]
- Added EOF basis test for Cylinder. [Martin D. Weinberg]
- Fix object count for Vector zero in expcoefT,T1,M,M1. [mdw]
- Missing reference make_shared creattion. [Martin D. Weinberg]
- Add member functions to Component to convert from local to global
coordinate system; convert particle positions and velocities in
UserAddMass to global coordinates. [Martin D. Weinberg]
- Version bump to mark changes (albeit very minor) to add
determine_fields_at_point to some basis classes. [Martin D. Weinberg]
- Merge branch 'devel' of https://bitbucket.org/mdweinberg/exp into
devel. [Martin D. Weinberg]
- Merge branch 'devel' of https://bitbucket.org/mdweinberg/exp into
devel. [mdw]
- Accel is turned off if no basis forces are available; other minor
formatting changes. [mdw]
- Some header documentation only. [Martin D. Weinberg]
- Added toggle for initial acceleration from components. [Martin D.
Weinberg]
- Another typo fix: acceleration from each component needs to be
accumulated. [mdw]
- Typo fix. [mdw]
- Report accel components. [mdw]
- Minor changes to Mike's fix. [Martin D. Weinberg]
- Merge branch 'devel' of https://bitbucket.org/mdweinberg/exp into
devel. [Martin D. Weinberg]
- Draft zero acceleration multistep bypass. [Michael]
- Updates for accelration adding for UserAddMass. [Martin D. Weinberg]
- Added monopole initialization routine. [mdw]
- Return to mass-weighted variance for error analysis. [Martin D.
Weinberg]
- Revert covar computation. [mdw]
- White space change only. [Martin D. Weinberg]
- Error propagation for coefficients. [mdw]
- Merged Luan's UserAddMass code. [Martin D. Weinberg]
- Updates for cross validation. [mdw]
- Added varialble exponent to Hall smoothing function. [mdw]
- Fixed cache read error: all nodes read so do not reduce table entries
on reread. [mdw]
- White space change only. [mdw]
- Cleaned up expcoef interface; made algorithmic changes in error
analysis; fixed wrong Ylm coefficient in analysis routines. [Martin D.
Weinberg]
- Remove some default debugging output. [Martin D. Weinberg]
- Make SNR computation consistent between Spherical and Cylinder.
[Martin D. Weinberg]
- Updates for particle number tracking in noise computation. [Martin D.
Weinberg]
- Put priorty queue stuff in separate header. [Martin D. Weinberg]
- Sign change in Term 3. [Martin D. Weinberg]
- Fixed progress bar, 2nd try. [Martin D. Weinberg]
- Fixed progress bar. [Martin D. Weinberg]
- More updates for Cylinder cross validation; still needs more work.
[Martin D. Weinberg]
- Minor fixes and some additional debugging output. [Martin D. Weinberg]
- Added multipole version for Cylinder. [Martin D. Weinberg]
- Fixed typo in cross_validation_cyl.cc; added EVEN_M flag to Cylinder.
[Martin D. Weinberg]
- More debugging. [Martin D. Weinberg]
- Some additional debugging. [Martin D. Weinberg]
- More fixes for cross_validation_cyl speed up; added Progress.H to shut
up boost deprecation warnings. [Martin D. Weinberg]
- More updates for cross validation. [Martin D. Weinberg]
- Merge branch 'devel' of https://bitbucket.org/mdweinberg/exp into
devel. [Martin D. Weinberg]
- Fix/restore definition of global in exputil lib. [Martin D. Weinberg]
- Merge branch 'devel' of https://bitbucket.org/mdweinberg/exp into
devel. [Martin D. Weinberg]
- Added gdb_trace option. [Martin D. Weinberg]
- Fix header packing order for recon_bin. [Martin D. Weinberg]
- Implementation of cross validation algorithm for Cylinder; not tested
but it does compile. [Martin D. Weinberg]
- Reorder division by mass per bin. [Martin D. Weinberg]
- Added interpolation to ang mom grid. [Martin D. Weinberg]
- This script will be autogenerated by the application that needs it.
[Martin D. Weinberg]
- Added electron energy cutoff option. [Martin D. Weinberg]
- Merge branch 'devel' of https://bitbucket.org/mdweinberg/exp into
devel. [Martin D. Weinberg]
- Fixes for SNR log values in loop. [Martin D. Weinberg]
- Header for PNG image output. [Martin D. Weinberg]
- Rename of cross_validation for cylinder. [Martin D. Weinberg]
- Rename of cross_validation. [Martin D. Weinberg]
- Included Hall smoothing. [Martin D. Weinberg]
- Turn off debug output by default. [Martin D. Weinberg]
- Change norm for covariance, add resampling option for
checking/debugging, zero accumulators in snr loop in main, add PNG
eigenfunction output. [Martin D. Weinberg]
- Fixed total mass computation in the SphereSL analysis class. [Martin
D. Weinberg]
- Missing verbose switch. [Martin D. Weinberg]
- Fixed typo in truncation for S/N. [Martin D. Weinberg]
- Remove per-node output from main. [Martin D. Weinberg]
- Implement S/N cuts for total cross-validation. [Martin D. Weinberg]
- Added rscl option to SphereSL in utils/Analysis. [Martin D. Weinberg]
- Parallelized coefficient cross-validation. [Martin D. Weinberg]
- Added cross validation test code. [Martin D. Weinberg]
- Minor fix to collisional excitation line selection: wrong line energy
for selected cross section with grid version. [Martin D. Weinberg]
- Fix typo in error string only. [Martin D. Weinberg]
- Added logic to preserve the workaround for the Thrust binary search
bug when (if) it is fixed; this is all preprocessor so the code will
be indentical. [Martin D. Weinberg]
- Sanity check: don't call scatterPhotoTrace if weight=0. [Martin D.
Weinberg]
- Fix a few typos: workaround flag misspelled. [Martin D. Weinberg]
- Work around for bug in thrust::binary-search. It looks like this will
be fixed eventually; see https://github.com/NVIDIA/thrust/pull/1104.
[Martin D. Weinberg]
- Version bump. [Martin D. Weinberg]
- Merge branch 'devel' of https://bitbucket.org/mdweinberg/exp into
devel. [Martin D. Weinberg]
- Fix thrust calls from device, which is unsupported usage. [Martin D.
Weinberg]
- Updated configuration for C++-14 required for Cuda 11.x. [Martin D.
Weinberg]