forked from openturns/openturns
-
Notifications
You must be signed in to change notification settings - Fork 0
/
ChangeLog
2593 lines (2305 loc) · 129 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
= New Features =
== 1.19 release (in-progress) == #release-1.19
=== Library ===
==== Major changes ====
==== New classes ====
==== API changes ====
* Removed deprecated Hanning alias
* Removed deprecated AdaptiveDirectionalSampling alias
* Removed deprecated VisualTest::DrawCobWeb method
* Removed deprecated shims module
* Removed deprecated TBB.SetNumberOfThreads/GetNumberOfThreads
* Removed deprecated MultiFORM.setMaximumNumberOfDesignPoints/getMaximumNumberOfDesignPoints
* Removed deprecated SubsetSampling.getNumberOfSteps
=== Documentation ===
=== Python module ===
=== Miscellaneous ===
=== Bug fixes ===
== 1.18 release (2021-11-10) == #release-1.18
=== Library ===
==== Major changes ====
==== New classes ====
* DistanceToDomainFunction
==== API changes ====
* Removed deprecated MultiStart::setStartingPoints/getStartingPoints
* Removed deprecated KarhunenLoeveResult::getEigenValues
* Removed deprecated coupling_tools.execute is_shell/workdir/shell_exe/check_exit_code arguments
* RandomVector::get/setParameter and getParameterDescription available to PythonRandomVector
* Implemented CovarianceModel::computeCrossCovariance
* Deprecated Hanning in favor of Hann
* Deprecated AdaptiveDirectionalSampling in favor of AdaptiveDirectionalStratification
* Deprecated VisualTest::DrawCobWeb in favor of DrawParallelCoordinates
* IndicatorFunction constructor takes a Domain as input
* Intervals and DomainUnions get new method computeDistanceToDomain(Point or Sample)
* Renamed some ResourceMap keys: cache-max-size>Cache-MaxSize, parallel-threads>TBB-ThreadsNumber
* Deprecated TBB.SetNumberOfThreads/GetNumberOfThreads
* Deprecated MultiFORM.setMaximumNumberOfDesignPoints/getMaximumNumberOfDesignPoints
* Deprecated SubsetSampling.getNumberOfSteps
* Normal and Student distributions: no need to specify the R parameter if it is the identity matrix
* KrigingResult::getConditionalMean and getConditionalMarginalVariance yield Samples instead of Points when applied to Samples
* Deprecated shims module
=== Documentation ===
=== Python module ===
=== Miscellaneous ===
=== Bug fixes ===
* #1840 (HMatrixFactory leaks)
* #1842 (libOT.so.0.0.0 doesn't have a SONAME)
* #1843 (Ali-Mikhail-Haq copula parameter value)
* #1844 (MaximumDistribution::computePDF is wrong when all the marginals are equal and independent)
* #1845 (MemoizeFunction does not propagate to finite difference gradient&hessian)
* #1847 (ParametricFunction require unnecessary function evaluations)
* #1854 (Sample indexing does not work on np.int64 type)
* #1856 (Speed up Normal computeSurvivalFunction)
* #1857 (ProductCovarianceModel fails when constructed with DiracCovarianceModel)
* #1858 (Normal::computeCDF(sample) crash for large dimensions)
* #1861 (Still some instabilities in Kriging with ot.StationaryFunctionalCovarianceModel)
* #1864 (FORM - IMPORTANCE FACTOR)
* #1868 (Not compatible with hmat-oss-1.7.1: no member named 'compressionMethod' in 'hmat_settings_t')
* #1870 (TruncatedDistribution fails to compute quantiles)
* #1874 (ProcessSample.getSampleAtVertex() may be useful as a public method)
* #1877 (How to model singular multivariate distributions?)
* #1878 (Rename Hanning filtering window)
* #1879 (Adaptive Directional Sampling Algorithm: the drawProbabilityConvergence graph may be wrong)
* #1880 (Adaptive Directional Sampling: some enhancements proposed)
* #1882 (Is Distribution::getLinearCorrelation useful ?)
* #1883 (Strange behavior of FORM ?)
* #1884 (setDesign can lead to wrong Sobol' indices)
* #1891 (Correlation of Halton sequence at high dimensions)
* #1911 (DirectionalSampling freezes python if not correctly initialized)
* #1912 (splitter: missing doc)
* #1915 (computePDFGradient(const Sample &) should rely on computePDFGradient(const Point &) in DistributionImplementation)
* #1918 (PythonDistribution does not allow to overload the isDiscrete() method)
== 1.17 release (2021-05-12) == #release-1.17
=== Library ===
==== Major changes ====
==== New classes ====
* VertexValuePointToFieldFunction
* KarhunenLoeveReduction
* KarhunenLoeveValidation
* IsotropicCovarianceModel
* KroneckerCovarianceModel
* VonMisesFactory
* KFoldSplitter, LeaveOneOutSplitter
==== API changes ====
* Removed deprecated Pairs alias
* Removed deprecated SORMResult::getEventProbabilityHohenBichler/getGeneralisedReliabilityIndexHohenBichler
* Removed deprecated OptimizationResult::getLagrangeMultipliers, OptimizationAlgorithm::computeLagrangeMultipliers
* Removed deprecated FittingTest::Kolmogorov, FittingTest::BestModelKolmogorov (DistributionFactory argument only)
* Deprecated Sample::computeStandardDeviationPerComponent, use computeStandardDeviation
* Deprecated KarhunenLoeveResult::getEigenValues, use getEigenvalues
* Removed StationaryCovarianceModel
* CovarianceModel.computeAsScalar allows scalars
* Swapped SimulatedAnnealingLHS constructor SpaceFilling/TemperatureProfile arguments
* Added EfficientGlobalOptimization::getKrigingResult, gets the updated version of the KrigingResult passed to the constructor
* Deprecated MultiStart::setStartingPoints/getStartingPoints, use MultiStart::setStartingSample/getStartingSample instead
* MultiStart::setStartingPoint/getStartingPoint throws (previously did nothing)
=== Documentation ===
* Fixed example and plot of Kolmogorov statistics.
* Added a new example showing how to combine RandomWalkMetropolisHastings and PythonDistribution
=== Python module ===
* Serialize Python wrapper objects using dill (PythonDistribution, PythonFunction ...)
=== Miscellaneous ===
=== Bug fixes ===
* #1010 (The doc for ExpectationSimulationAlgorithm is confusing)
* #1052 (The Dirichlet and Normal distributions only have 1D graphics in the doc)
* #1224 (The examples in the doc of the DomainEvent class are unclear)
* #1229 (Better encapsulation of optional 3rd-party headers)
* #1240 (There is no theory documentation for ExpectationSimulationAlgorithm)
* #1257 (Cannot create a SimulatedAnnealingLHS without specifying the temperature profile)
* #1287 (The constructors of KernelSmoothing have no doc)
* #1418 (The Contour example does not present the second constructor)
* #1425 (There is no method to create a train / test pair)
* #1431 (The figure for LHSExperiment is not accurate)
* #1459 (There is no example which shows how to set a column of a Sample)
* #1497 (There is no example to create a multivariate Normal distribution)
* #1506 (The Brent class has no example)
* #1570 (Wrong formula for MauntzKucherenkoSensitivityAlgorithm)
* #1650 (There is no example of a parametric StationaryFunctionalCovarianceModel)
* #1656 (The help page of ExpectationSimulationAlgorithm is unclear)
* #1661 (XMLH5StorageManager does not store IndicesCollection into HDF5 files)
* #1662 (Operator() of a CovarianceModel with multidimensional output should yield object of type SquareMatrix)
* #1680 (QuadraticBasisFactory multiplies cross products by 2)
* #1710 (Misleading y labels in VisualTest.DrawPairsMarginals())
* #1713 (The doc of the ResourceMap does not match the content of the ResourceMap)
* #1714 (The log-PDF of the Pareto is wrong)
* #1721 (The `draw` method of `SobolSimulationAlgorithm` does not reuse the descriptions)
* #1723 (There is no interesting example of the ExprTk feature of SymbolicFunction)
* #1725 (The BetaFactory help page has wrong equations)
* #1729 (LinearModelAnalysis sometimes fail)
* #1731 (The Extreme value example may be improved)
* #1737 (Low rank tensor doc issues)
* #1742 (The example which shows how to set the figure size is wrong.)
* #1751 (DrawCorrelationCoefficients has a wrong Text height)
* #1752 (Error while estimating the reduced log-likelihood when using a StationaryFunctionalCovarianceModel)
* #1758 (The LinearModelStepwiseAlgorithm has no example)
* #1759 (Kriging model with StationaryFunctionalCovarianceModel might provide bad results)
* #1768 (Bug in ExponentiallyDampedCosineModel & SphericalModel)
* #1771 (GridLayout hides the titles in the graph)
* #1772 (MinimumVolumeClassifier cannot draw 1D samples)
* #1774 (ExponentialModel::partialGradient is wrong)
* #1775 (ExponentialModel does not account correlation with Covariance ctor)
* #1776 (Typo in the Branin use case implementation)
* #1781 (The link_to_an_external_code example has small bugs)
* #1784 (The drawPDF method of Histogram sometimes fail)
* #1787 (CovarianceMatrix from SymmetricMatrix raises InvalidArgumentException)
* #1790 (LinearModelResult.getFormula() method is not updated by Stepwise regression)
* #1794 (Some doc examples seem to behave with new infrastructure)
* #1796 (VonMisesFactory is missing)
* #1803 (The API example of Experiment has a format issue)
* #1805 (Brent's implementation has a stability issue)
* #1807 (Beta::computeCharacteristicFunction is wrong if the lower bound is not zero)
* #1815 (1.16 fails with dlib-cpp-19.22)
* #1818 (Problem with wilksNumber)
* #1820 (Incoherent results in LinearModelAnalysis)
* #1835 (RegularizedIncompleteBeta returns nan)
* #1836 (Hypergeometric results differ without boost)
* #1837 (Poor performance when using pickle on OT objects)
== 1.16 release (2020-11-16) == #release-1.16
=== Library ===
==== Major changes ====
* Drop normalization in KrigingAlgorithm
* Drop normalization & transformation handling in GeneralLinearModelAlgorithm
* XML/H5 storage (hdf5 library)
* C++11 requirement
==== New classes ====
* BlockIndependentDistribution
* FejerAlgorithm
* GridLayout
* MinimumVolumeClassifier
* StationaryFunctionalCovarianceModel
* XMLH5StorageManager
==== API changes ====
* Removed deprecated Weibull, WeibullFactory, WeibullMuSigma classes
* Removed deprecated GumbelAB class
* Removed deprecated Event class
* Removed deprecated EnumerateFunction constructors
* Removed various deprecated distribution accessors
* Deprecated Pairs class, see VisualTest.DrawPairs
* Deprecated SORMResult::getEventProbabilityHohenBichler, use SORMResult::getEventProbabilityHohenbichler instead
* Deprecated SORMResult::getGeneralisedReliabilityIndexHohenBichler, use SORMResult::getGeneralisedReliabilityIndexHohenbichler instead
* Renamed SobolSequence::MaximumNumberOfDimension as SobolSequence::MaximumDimension
* Added VisualTest::DrawLinearModel(linearModelResult), useful if the test is performed on the training samples
* Added VisualTest::DrawLinearModelResidual(linearModelResult), useful if the test is performed on the training samples
* Deprecated OptimizationResult::getLagrangeMultipliers
* Moved OptimizationAlgorithm::computeLagrangeMultipliers to OptimizationResult
* Added AIC & BestModelAIC static methods in FittingTest
* Added AICC & BestModelAICC static methods in FittingTest
* Moved BuildDistribution from FunctionalChaosAlgorithm to MetaModelAlgorithm
* Added Drawable::BuildRainbowPalette(size)
* Added Drawable::BuildTableauPalette(size), which is now the default palette.
* Added Drawable::ConvertFromRGBIntoHSV
* Added FittingTest::Lilliefors, BestModelLilliefors
* Deprecated FittingTest::BestModelKolmogorov(Sample, DistributionFactoryCollection, TestResult), use BestModelLilliefors
* Deprecated FittingTest::Kolmogorov(Sample, DistributionFactory, TestResult, level), use Lilliefors
* MetamodelValidation: now computePredictivityFactor returns Point, drawValidation return GridLayout
* Deprecated coupling_tools.execute is_shell/workdir/shell_exe/check_exit_code arguments
=== Documentation ===
* Sphinx-gallery used to render examples
==== API documentation ====
* Clarified SobolIndicesExperiment page, notations now consistent with SobolIndicesAlgorithm page
* Clarified SobolIndicesAlgorithm and (Saltellli|Martinez|MauntzKucherenko|Jansen)SensitivityAlgorithm pages, corrected formulas
* Documented how to turn warnings off or write them on a file
=== Python module ===
* Renamed Viewer *_kwargs arguments to *_kw (matplotlib convention)
* Add ProcessSample Field accessors
=== Miscellaneous ===
* Do not compute Lagrange multipliers by default during an optimization
* Add ResourceMap::FindKeys
* Allow computeLogPDF methods to output values lower than SpecFunc::LogMinScalar
=== Bug fixes ===
* #1001 (Add method SobolSimulationResult::draw)
* #1259 (The diagonal of a scatter plot matrix should have the histograms)
* #1267 (Some CSV files cannot be imported)
* #1377 (The `setKnownParameter` method is not compatible with `buildEstimator`)
* #1407 (GeneralLinearModelAlgorithm mishandles user-specified scale parameter when normalize is True)
* #1415 (The BuildDistribution static method should not use the KS-test)
* #1421 (UserDefinedStationaryCovarianceModel doc suggests input dimension can be >1)
* #1436 (The style of the curves is unpleasing to my eyes)
* #1447 (Highly inaccurate result in reliability model when using subset of RandomVector)
* #1465 (The Sample constructor based on a list and an integer should not exist)
* #1470 (setNbModes is sometimes ignored)
* #1474 (optimization defaults)
* #1507 (Leak in Collection typemaps)
* #1510 (ot.Ceres('LEVENBERG_MARQUARDT') and ot.Ceres('DOGLEG') do not handle bound constraints)
* #1515 (KernelSmoothing build failure)
* #1520 (The NLopt test is dubious)
* #1521 (Basis of MonomialFunction)
* #1529 (The error of the NonLinearLeastSquaresCalibration and GaussianNonLinearCalibration are different)
* #1540 (SubsetSampling: incorrect event sample)
* #1547 (Mesh does not check the simplices indices)
* #1549 (Doc of evaluation operator of KrigingResult)
* #1553 (Optimization algorithms ignore MaxEvaluationNumber parameter in SORM)
* #1556 (WeibullMin::computePDFGradient yields the partial derivatives in the wrong order)
* #1558 (Example estimate_multivariate_normal: FittingTest::BestModelBIC fails to compute the BIC)
* #1564 (Set a Point makes ot crash)
* #1567 (The API doc of SobolIndicesExperiment has a format issue)
* #1573 (LinearModelAnalysis::drawQQPlot line is not the first bisector)
* #1578 (Option to suppressing and/or save warnings?)
* #1581 (LinearModelAlgorithm run() fails to parse Sample description)
* #1586 (Documentation: description error in the API for the FittingTest_BestModelKolmogorov and FittingTest_BestModelChiSquaredclasses)
* #1590 (The equation of the Fejer quadrature rule is triplicated)
* #1592 (SubsetSampling returns an error if Pf=1)
* #1594 (LinearLeastSquaresCalibration and CalibrationResult)
* #1599 (FieldToPointConnection-BlockSize is missing)
* #1603 (FieldToPointConnection generates an invalid exception)
* #1605 (MaximumLikelihoodFactory cannot be used with FittingTest.Kolmogorov)
* #1624 (The graphs_loglikelihood_contour example has a bug)
* #1642 (Big white space at the beginning of examples)
* #1643 (Problem in MaximumDistribution PDF)
* #1647 (MCMC::computeLogLikelihood does not compute the log-likelihood)
* #1651 (Cobyla freezes in 0T1.16rc1)
* #1658 (TimeSeries accessor)
* #1660 (Cannot extract continuous modes from KLResult when dimension>1)
* #1668 (LevelSetMesher does not take into account the comparison operator)
== 1.15 release (2020-05-25) == #release-1.15
=== Library ===
==== Major changes ====
* New EV solver for KarhunenLoeveP1Algorithm (Spectra), with sparse matrix and HMatrix support
* Enable HMat AcaRandom compression method
==== New classes ====
* Ipopt optimization solver
==== Documentation ====
==== API changes ====
* Removed deprecated OptimizationAlgorithm::GetLeastSquaresAlgorithmNames
* Removed deprecated GaussianNonLinearCalibration,NonLinearLeastSquaresCalibration::set,getAlgorithm
* Removed deprecated MethodOfMomentsFactory::set,getOptimizationProblem
* ResourceMap::Set* methods no longer add new keys, the new Add* methods must be used instead
* Removed OPTpp
* Renamed HistogramFactory::computeSilvermanBandwidth into HistogramFactory::computeBandwidth.
=== Python module ===
* ProcessSample __getitem__ returns Sample instead of Field
* Implement list indexing
=== Miscellaneous ===
* Add Sample::getMarginal(Description)
* Fixed TBB performance when used together with OpenBLAS
=== Bug fixes ===
* #1124 (DistributionFactory::buildAsXXX methods not documented)
* #1213 (The legend of the graphics in MetaModelValidation is wrong)
* #1222 (There is no kriging example based on HMAT)
* #1331 (FittingTest_BestModelBIC sometimes fail)
* #1335 (The return of the unsafe ResourceMap)
* #1337 (The rDiscrete function has no help page)
* #1349 (Problem in the graph of Histogram)
* #1351 (Text has a zero size)
* #1354 (The doc of GeneralizedParetoFactory does not reflect the implementation)
* #1371 (Memory leak in ot.TruncatedDistribution)
* #1372 (wrap MultiStart::OptimizationResultCollection)
* #1374 (The setKnownParameter of the factories are not documented enough)
* #1376 (The View class has no example)
* #1378 (Kriging-related covariance model weirdness)
* #1383 (The ExprTk engine for SymbolicFunction does not document the "var" keyword)
* #1384 (The ExprTk engine is not case-sensitive)
* #1388 (Kolmogorov fails on a PythonDistribution)
* #1390 (The doc for the `computeQuantile` method does not describe the optional `tail` argument)
* #1393 (SobolSimulationAlgorithm should be simpler)
* #1395 (Indexing Sample improvement)
* #1403 (Python import otagrum throws an error)
* #1404 (Bug in KrigingAlgorithm+hmat-oss)
* #1405 (Most of the simulation algorithms for rare event fail on a coronavirus example)
* #1416 (MethodOfMomentsFactory has no setOptimizationBounds method)
* #1419 (BoundingVolumeHierarchy segaults/hangs)
* #1423 (The computeSilvermanBandwidth of the HistogramFactory has no help)
* #1432 (Expected improvement-based EfficientGlobalOptimization stopping criterion could be improved)
* #1437 (OrderStatisticsMarginalChecker bound message)
* #1438 (Normal distribution: computeComplementaryCDF)
* #1443 (Not all distributions have a getRoughness() method)
* #1448 (Memory consumption leads to crash)
* #1449 (P1LagrangeInterpolation sometimes fails)
* #1455 (GLM::setCovarianceModel could lead to unexpected behavior of parameter optimization in KrigingAlgorithm)
* #1456 (truncation of distribution)
* #1461 (ComparisonOperator().getImplementation().getClassName() segfault)
* #1471 (The graphics of KarhunenLoeveQuadratureAlgorithm has no axes)
* #1485 (The Normal().getRoughness() method is wrong)
* #1495 (Wrong formula for Expected Improvement evaluation)
== 1.14 release (2019-11-13) == #release-1.14
=== Library ===
==== IMPORTANT: Distributions parametrization changes ====
* New argument ordering in Frechet ctor: scale(beta), shape(alpha), location(gamma) (swapped alpha and beta)
* New parametrization in Gumbel: scale(beta), position(gamma) (beta=1/alpha for first argument)
* New parametrization in Beta: shape(alpha), shape(beta), location(a), location(b) (beta=t-r for second argument)
* New parametrization in InverseGamma: rate(lambda), shape(k) (swapped arguments)
* New parametrization in InverseNormal: location(mu), rate(lambda) (swapped arguments)
* New parametrization in Laplace: mean(mu), rate(lambda) (swapped arguments)
=> One can use "import openturns.shims as ot" to maintain compatibility with older scripts
==== Major changes ====
* New optimization solvers (Dlib, Bonmin for mixed integer optimization problems)
* New distributions (SquaredNormal, WeibullMax, Pareto, DiscreteCompoundDistribution, MixedHistogramUserDefined)
* New estimators (ParetoFactory, GeneralizedExtremeValueFactory, LeastSquaresDistributionFactory, PlackettCopulaFactory)
* New copulas (JoeCopula, MarshallOlkinCopula, PlackettCopula)
* Linear model learner (LinearModelStepwiseAlgorithm)
* System events (IntersectionEvent, UnionEvent, SystemFORM, MultiFORM)
==== New classes ====
* Dlib
* SquaredNormal
* NullHessian
* GumbelLambdaGamma
* LogNormalMuErrorFactor
* WeibullMax
* WeibullMaxFactory
* WeibullMaxMuSigma
* GeneralizedExtremeValueFactory
* Pareto
* LeastSquaresDistributionFactory
* ParetoFactory
* DiscreteCompoundDistribution
* Bonmin
* IntersectionEvent
* UnionEvent
* SystemFORM
* MultiFORM
* JoeCopula
* MarginalEvaluation/Gradient/Hessian
* MarshallOlkinCopula
* LinearModelStepwiseAlgorithm
* MixedHistogramUserDefined
* PlackettCopula
* PlackettCopulaFactory
==== API changes ====
* FittingTest methods to return fitted distributions with factory as argument
* Removed deprecated specific RandomVector constructors
* Removed deprecated HypothesisTest::Smirnov
* Removed deprecated FittingTest::TwoSamplesKolmogorov
* Removed deprecated LinearModel, LinearModelFactory
* Removed deprecated HypothesisTest::(Partial|Full)Regression
* Removed deprecated OptimizationProblem(levelFunction, levelValue) ctor
* Removed deprecated (Linear|Quadratic)(LeastSquares|Taylor)::getResponseSurface
* Removed deprecated VisualTest::DrawEmpiricalCDF,DrawHistogram,DrawClouds
* Moved dot to Point::dot
* Deprecated Weibull in favor of WeibullMin
* Deprecated WeibullMuSigma in favor of WeibullMinMuSigma
* Deprecated WeibullFactory in favor of WeibullMinFactory, buildAsWeibull
* Deprecated GumbelAB
* Deprecated GaussianNonLinearCalibration,NonLinearLeastSquaresCalibration::set,getAlgorithm
* Added getConditionalMarginalCovariance method to KrigingResult
* Added getConditionalMarginalVariance method to KrigingResult
* Deprecated OptimizationAlgorithm::GetLeastSquaresAlgorithmNames
* Added linearity features to Function
* Added 'removeKey' method to ResourceMap
* Removed Copula class
* Deprecated Event class, use ThresholdEvent/ProcessEvent/DomainEvent classes
* Deprecated EnumerateFunction constructors
* Added a minimum probability accessor to SubsetSampling
* Added a UniVariateFunction interface to solvers and integration algorithms
=== Python module ===
* Add Domain.__contains__ operator
=== Miscellaneous ===
* Add GeneralizedPareto location parameter
* Add getSobolGroupedTotalIndex method for FunctionalChaosSobolIndices for the indice of a group of variables
=== Bug fixes ===
* #997 (Adding minimum volume set examples)
* #1004 (The doc for SobolSimulationAlgorithm has issues)
* #1006 (Text drawable does not handle size)
* #1130 (Inconsistency in FittingTest)
* #1160 (2-d GaussianProcess realization graph regression)
* #1169 (Missing key in ResourceMap)
* #1173 (There is no dot product example)
* #1185 (Bug with Normal.computeMinimumLevelSet method)
* #1190 (computeProbability clamped by Domain-SmallVolume)
* #1197 (Doc error: TrendTransform)
* #1198 (Doc error: ValueFunction)
* #1202 (Sample::sort & Sample::sortAccordingToAComponent only return new Samples)
* #1204 (sortAccordingToAComponent does not check its inputs arguments)
* #1209 (LinearModelStepwiseAlgorithm from otlm does not exist in OT)
* #1216 (The CalibrationResult doc does not match the code)
* #1247 (KrigingAlgorithm could not compute amplitude analytically with ProductCovarianceModel )
* #1264 (ProductCovarianceModel ignore active parameters of its 1d marginals)
* #1282 (The dlib example fails)
* #1283 (LogNormalFactory::buildMethodOfLeastSquares has no doc)
* #1289 (The help page of PythonFunction has formatting issues)
* #1303 (Rosenblatt transformation segfault)
== 1.13 release (2019-06-06) == #release-1.13
=== Library ===
==== Major changes ====
* Added OPT++ solvers
* Improved a lot the performace of all the Rosenblatt related computations
* Added elementary calibration capabilities
* Added CMinpack, Ceres Solver least-squares solvers
==== New classes ====
* ParametricPointToFieldFunction
* LinearModelResult, LinearModelAlgorithm, LinearModelAnalysis
* OPTpp
* NearestPointProblem, LeastSquaresProblem
* CMinpack
* Ceres
* DiscreteMarkovChain
* CalibrationAlgorithm, CalibrationResult, GaussianLinearCalibration, LinearLeastSquaresCalibration
* NonLinearLeastSquaresCalibration, GaussianNonLinearCalibration
* Hypergeometric
==== API changes ====
* Removed deprecated FunctionalChaosRandomVector::getSobol* methods
* Removed deprecated UserDefinedCovarianceModel constructor based on a Collection<CovarianceMatrix>
* Removed deprecated FieldFunction,FieldToPointFunction::getSpatialDimension
* Removed deprecated LinearModelRSquared, LinearModelAdjustedRSquared
* Deprecated LinearModel, LinearModelFactory
* Moved HypothesisTest::(Partial|Full)Regression to LinearModelTest
* Deprecated OptimizationProblem(levelFunction, levelValue)
* Deprecated (Linear|Quadratic)(LeastSquares|Taylor)::getResponseSurface
* FittingTest::BestModelBIC returns the Distribution and the BIC value
* Deprecated VisualTest::DrawEmpiricalCDF,DrawHistogram,DrawClouds
* Add statistic attribute and accessor to the TestResult class
* Deprecated Point::getDescription,setDescription
=== Python module ===
=== Miscellaneous ===
* Changed bugtracker to GitHub issues (https://github.com/openturns/openturns/issues)
* Dropped rot dependency
=== Bug fixes ===
* #289 (Move the LinearModelFactory class from the Base namespace to the Uncertainty namespace)
* #579 (There is no example of a typical "Central Tendency" study)
* #839 (more optim solvers details)
* #931 (The PostAnalyticalImportanceSampling and PostAnalyticalControlledImportanceSampling are too briefly documented)
* #979 (Viewer does not handle BarPlot's fillStyle)
* #977 (HypothesisTest::ChiSquared is bogus)
* #980 (CorrelationAnalysis_SRC scales the coefficients so that they sum to 1)
* #981 (HypothesisTest_{Full, Partial} regression should be moved to LinearModelTest)
* #982 (LinearModelTest::LinearModelDurbinWatson with several factors)
* #983 (Bogus Normal parameter distribution)
* #989 (Fixed the documentation of BIC according to the code)
* #995 (Pip does not recognize conda install of openturns)
* #1000 (SaltelliSensitivityAlgorithm with LowDiscrepancyExperiment produces wrong results)
* #1005 (The stopping criteria of SobolSimulationAlgorithm is weird)
* #1024 (The unary operator is undefined for Normal distribution)
* #1025 (Empty throw in PythonWrappingFunctions)
* #1028 (Build fails with clang-60: token is not a valid binary operator in a preprocessor subexpression)
* #1031 (The doc for LogNormal and LogNormalMuSigma is confusing )
* #1035 (ProbabilitySimulationResult does not provide the distribution of the probability)
* #1036 (Doc error for KrigingAlgorithm method getReducedLogLikelihoodFunction())
* #1043 (The DrawSobolIndices doc is wrong)
* #1045 (LowDiscrepancyExperiment/ComposedCopula correlation across blocks ?)
* #1051 (The default value of computeSecondOrder in SobolIndicesExperiment should be False)
* #1054 (Optimization algorithms ignore MaxEvaluationNumber parameter)
* #1057 (FittingTest.BestModelBIC to return bic value)
* #1058 (GEV problem)
* #1064 (The summary of a FunctionalChaosSobolIndices fails with more than 14 dimensions)
* #1071 (Event cannot be created from a RandomVector)
* #1078 (KrigingAlgorithm documentation still references "inputTransformation")
* #1080 (Brent, Secant, Bissection documentation)
* #1085 (GPD documentation)
* #1090 (PythonFunction can make OT crash)
* #1092 (The parameterGradient of a ParametricFunction can loss accuracy)
* #1099 (PythonDistribution does not work with ConditionalDistribution)
* #1111 (Function::draw() does not take the scale parameter properly into account)
* #1112 (Several bugs in Multinomial)
* #1119 (The parameter description is not taken into account in CalibrationResult)
* #1129 (Segmentation fault with RandomMixture)
* #1139 (CopulaImplementation should derive from DistributionImplementation)
* #1143 (The ResourceMap is not correctly formatted in help pages)
* #1148 (Fails to find cminpack)
* #1155 (The description of Points can be set, but get is empty)
== 1.12 release (2018-11-08) == #release-1.12
=== Library ===
==== Major changes ====
* FieldFunction and co knows its input/output meshes before evaluation
* SobolSequence has been extended from maximum dimension 40 to 1111
* Parametrized statistical tests by first kind risk instead of 1-risk
* FittingTest now compute a correct p-value in Kolmogorov tests even if the parameters are estimated from the tested sample
* Completed documentation migration with process content
==== New classes ====
* ProbabilitySimulationResult
* ExpectationSimulationAlgorithm, ExpectationSimulationResult
* ExtremeValueCopula
* ChaospyDistribution
* SobolSimulationAlgorithm, SobolSimulationResult
* FractionalBrownianMotionModel
==== API changes ====
* Removed deprecated MonteCarlo, ImportanceSampling, QuasiMonteCarlo, RandomizedQuasiMonteCarlo, RandomizedLHS classes
* Removed deprecated Field::getDimension, getSpatialDimension, getSpatialMean, getTemporalMean methods
* Remove deprecated Process::getDimension, getSpatialDimension methods
* Removed deprecated CovarianceModel::getDimension, getSpatialDimension, getSpatialCorrelation, setSpatialCorrelation methods
* Removed deprecated SpectralModel::getDimension, getSpatialDimension, getSpatialCorrelation methods
* Removed deprecated TruncatedDistribution single bound accessors
* Removed deprecated Function constructors
* Removed deprecated Sample,SampleImplementation::operator*(SquareMatrix) and operator/(SquareMatrix) (and in-place operators)
* Removed deprecated SampleImplementation::scale(SquareMatrix)
* Removed deprecated Domain(a, b) constructor
* Removed deprecated Domain,DomainImplementation::numericallyContains, isEmpty, isNumericallyEmpty, getVolume, getNumericalVolume, computeVolume
* Removed deprecated Domain,DomainImplementation::getLowerBound, getUpperBound methods
* Removed deprecated Interval,Mesh::computeVolume
* Removed deprecated SobolIndicesAlgorithm::[sg]etBootstrapConfidenceLevel
* Removed deprecated Mesh::getVerticesToSimplicesMap,computeSimplexVolume
* Removed deprecated Evaluation,EvaluationImplementation,EvaluationProxy,Function,FunctionImplementation,ParametricEvaluation::operator(inP,parameter)
* Removed deprecated Gradient,GradientImplementation,ParametricGradient::gradient(inP,parameter)
* Removed deprecated Hessian,HessianImplementation,ParametricHessian::operator(inP,parameter)
* Removed ExponentialCauchy, SecondOrderModel, SecondOrderImplementation
* LowDiscrepancyExperiment to work with dependent distributions
* Deprecated UserDefinedCovarianceModel constructor based on a Collection<CovarianceMatrix>
* Deprecated FieldFunction,FieldToPointFunction::getSpatialDimension
* Removed VertexFunction class
* Deprecated LinearModelRSquared, LinearModelAdjustedRSquared
* Added ResourceMap::GetType to access the type of a given key.
* Extended OrthogonalBasis::build() to multi-indices (see ticket #967)
* Deprecated specific RandomVector constructors
=== Python module ===
=== Miscellaneous ===
* CMake >=2.8.8 is required to build OpenTURNS
* The format of openturns.conf has changed to make a distinction between types of keys
=== Bug fixes ===
* #660 (Strange behavior of KernelSmoothing)
* #684 (Strange behavior of Student distribution in multidimensional case)
* #778 (DirectionalSampling is unstable)
* #915 (Distribution.computeQuantile properties)
* #949 (LowDiscrepancyExperiment for distributions having dependent copula)
* #952 (Limitation on distribution type for polynomial chaos?)
* #953 (RandomMixture can fail with truncated distribution)
* #954 (CompositeProcess::getFuture() broken)
* #955 (StudentFactory does not estimate the standard deviation)
* #957 (RandomMixture::getDispersionIndicator() takes a long time)
* #958 (The Sobol' indices plot is wrong for chaos)
* #959 (The PolygonArray and Polygon classes poorly manage the colors)
* #960 (Polygon sometimes make OT crash)
* #961 (The PolygonArray class seem to ignore the legends)
* #962 (The doc of MetaModelValidation.computePredictivityFactor is wrong)
* #963 (VisualTest_DrawHistogram sometimes has overlapping X labels)
* #964 (Study does not load LogNormalMuSigma variables from XML)
* #965 (A RandomVector from a RandomVector can make OT crash)
* #968 (Empty legend make OT crash)
* #970 (Composing gaussian copulas can crash the chaos)
* #972 (FittingTest::ChiSquared slow and buggy)
* #974 (Default constructor of the TruncatedDistribution)
* #975 (EmpiricalBernsteinCopula is a copula as sample is truncated)
== 1.11 release (2018-05-11) == #release-1.11
=== Library ===
==== Major changes ====
==== New classes ====
* FrechetFactory
* Fehlberg
* TranslationFunction
* DomainComplement, DomainIntersection, DomainUnion, DomainDisjunctiveUnion, DomainDifference
* SmoothedUniform
* NearestNeighbourAlgorithm, RegularGridNearestNeighbour, NaiveNearestNeighbour, NearestNeighbour1D
* EnclosingSimplex, EnclosingSimplexImplementation, NaiveEnclosingSimplex, RegularGridEnclosingSimplex, EnclosingSimplexMonotonic1D, BoundingVolumeHierarchy
* IndicesCollection
* SymbolicParserExprTk
* EvaluationProxy
* MemoizeFunction, MemoizeEvaluation
* Evaluation, Gradient, Hessian
* MeshDomain
* NormInfEnumerateFunction
* FunctionalChaosSobolIndices
* P1LagrangeInterpolation
* FilonQuadrature
==== API changes ====
* Removed deprecated NumericalMathFunction class
* Removed deprecated QuadraticNumericalMathFunction class
* Removed deprecated LinearNumericalMathFunction class
* Removed deprecated NumericalSample class
* Removed deprecated NumericalPoint[WithDescription] class
* Removed deprecated NumericalScalarCollection class
* Removed deprecated NumericalComplexCollection class
* Removed deprecated PosteriorRandomVector class
* Removed deprecated ConditionedNormalProcess class
* Removed deprecated ResourceMap::[SG]AsNumericalScalar methods
* Removed deprecated SpecFunc::*NumericalScalar* constants
* Removed deprecated PlatformInfo::GetConfigureCommandLine method
* Removed deprecated Field::getSample method
* Removed deprecated SobolIndicesAlgorithm::Generate method
* Removed deprecated NumericalScalar, NumericalComplex types
* Removed deprecated Function constructors
* CorrelationAnalysis::(Pearson|Spearman)Correlation accepts a multivariate sample and returns a Point
* Deprecated Field::getDimension, getSpatialDimension, getSpatialMean, getTemporalMean
* Deprecated Process::getDimension, getSpatialDimension
* Deprecated CovarianceModel::getDimension, getSpatialDimension, getSpatialCorrelation, setSpatialCorrelation
* Deprecated SecondOrderModel::getDimension, getSpatialDimension
* Deprecated SpectralModel::getDimension, getSpatialDimension, getSpatialCorrelation
* Deprecated TruncatedDistribution single bound accessors in favor of setBounds/getBounds
* Deprecated remaining analytical & database Function ctors
* Mesh constructor no longer builds a KDTree, new method Mesh::setNearestNeighbourAlgorithm must be called explicitly
* Deprecated Domain::numericallyContains, isEmpty, isNumericallyEmpty, getVolume, getNumericalVolume, computeVolume, getLowerBound, getUpperBound
* Add an argument to KarhunenLoeveQuadratureAlgorithm to pass domain bounds
* Add an argument to Mesh::streamToVTKFile() and Mesh::exportToVTKFile() to pass custom simplices
* All KDTree methods are modified
* IndicesCollection is no more an alias to Collection<Indices>, it has its own class and has a fixed size
* Removed BipartiteGraph::add method
* Deprecated usage of _e and _pi constants when defining analytical functions, e_ and pi_ must be used instead.
* Removed Function::enableHistory,disableHistory,isHistoryEnabled,clearHistory,getHistoryInput,getHistoryOutput,getInputPointHistory,getInputParameterHistory
* Removed Function::enableCache,disableCache,isCacheEnabled,getCacheHits,addCacheContent,getCacheInput,getCacheOutput,clearCache
* Added Drawable::getPaletteAsNormalizedRGBA() to get the palette into a native matplotlib format.
* Changed DistributionImplementation::getCopula,getMarginal,getStandardRepresentative,getStandardDistribution to return a Distribution instead of Pointer<DistributionImplementation>
* Changed DistributionFactoryImplementation::build to return a Distribution instead of Pointer<DistributionImplementation>
* Changed ProcessImplementation::getMarginal to return a Process instead of Pointer<ProcessImplementation>
* Changed RandomVector::getAntecedent,getMarginal to return a RandomVector instead of Pointer<RandomVectorImplementation>
* Changed all {get,set}{Evaluation,Gradient,Hessian} methods to work on Evaluation/Gradient/Hessian instead of Pointer
* Deprecated SobolIndicesAlgorithm::setBootstrapConfidenceLevel, getBootstrapConfidenceLevel
* TNC, Cobyla, EGO are parametrized by setMaximumEvaluationNumber instead of setMaximumIterationNumber
* Removed all KDTree services from Mesh,RegularGrid,Field::getNearestVertex, getNearestVertexIndex, etc, client code must use NearestNeighbourAlgorithm if needed
* Changed Mesh to not derive from DomainImplementation
* Deprecated Mesh::computeSimplexVolume, use Mesh::computeSimplicesVolume instead
* Deprecated Mesh,Interval::computeVolume
* Deprecated FunctionalChaosRandomVector::getSobol* methods in favor of FunctionalChaosSensitivity ones
* DeprecatedEvaluation,EvaluationImplementation,EvaluationProxy,Function,FunctionImplementation,ParametricEvaluation::operator()(inP,parameter)
* Deprecated Gradient,GradientImplementation,ParametricGradient::gradient(inP,parameter)
* Deprecated Hessian,HessianImplementation,ParametricHessian::hessian(inP,parameter)
=== Python module ===
* Depend on swig>=2.0.9
=== Miscellaneous ===
* Multivariate TruncatedDistribution
* Asymptotic Sobol' variance estimators
=== Bug fixes ===
* #870 (Problem with TensorizedCovarianceModel with spatial dim > 1)
* #926 (Covariance model active parameter set behavior)
* #928 (Covariance model/ Field/Process properties naming)
* #932 (Optim algo iteration/evaluation number)
* #937 (Small bugs in FunctionalChaosResult::get{Residuals, RelativeErrors})
* #938 (DistributionFactory::buildEstimator to handle Exception)
* #939 (The help of the getMaximumDegreeStrataIndex method is wrong.)
* #941 (Duplicate with SobolIndicesExperiment)
* #944 (Inline plots crash in notebooks if too many points)
* #945 (build fails with cmake 3.11)
* #946 (OT via anaconda, usage of ot.HypothesisTest error: R_EXECUTABLE-NOTFOUND)
* #947 (EmpiricalBernsteinCopula::setCopulaSample() too restrictive)
* #948 (Bug with BayesDistribution)
* #950 (Circular call to getShape() in Copula)
== 1.10 release (2017-11-13) == #release-1.10
=== Library ===
==== Major changes ====
* It is now possible to define numerical model acting on either Point or Field
to produce either Point or Field.
The Python binding has been extended to allow the user to define such
functions based on either a Python function or a Python class.
All the possible compositions have been implemented.
==== New classes ====
* ProbabilitySimulation
* SobolIndicesExperiment
* VertexFunction
* FieldToPointFunction
* FieldToPointFunctionImplementation
* PythonFieldToPointFunction
* OpenTURNSPythonFieldToPointFunction
* PointToFieldFunction
* PointToFieldFunctionImplementation
* PythonPointToFieldFunction
* OpenTURNSPythonPointToFieldFunction
* KarhunenLoeveLifting
* KarhunenLoeveProjection
* PointToPointEvaluation
* PointToPointConnection
* PointToFieldConnection
* FieldToFieldConnection
* FieldToPointConnection
==== API changes ====
* Removed deprecated LAR class
* Remove deprecated Function::GetValidConstants|GetValidFunctions|GetValidOperators
* Removed deprecated TemporalNormalProcess, SpectralNormalProcess classes
* Removed deprecated GeneralizedLinearModelAlgorithm, GeneralizedLinearModelResult classes
* Removed deprecated DynamicalFunction, SpatialFunction, TemporalFunction classes
* Removed deprecated KarhunenLoeveP1Factory, KarhunenLoeveQuadratureFactory classes
* Removed deprecated GramSchmidtAlgorithm, ChebychevAlgorithm classes
* Removed [gs]etOptimizationSolver methods
* Removed CovarianceModel::compute{AsScalar,StandardRepresentative} overloads
* Deprecated PosteriorRandomVector
* Deprecated MonteCarlo, ImportanceSampling, QuasiMonteCarlo, RandomizedQuasiMonteCarlo, RandomizedLHS classes
* Made MatrixImplementation::isPositiveDefinite const and removed its argument
* Renamed EfficientGlobalOptimization::setAIETradeoff to setAEITradeoff
* Deprecated PlatformInfo::GetConfigureCommandLine.
* Renamed ConditionedNormalProcess to ConditionedGaussianProcess
* Deprecated Field::getSample in favor of getValues
* Deprecated SobolIndicesAlgorithm::Generate in favor of SobolIndicesExperiment.
=== Python module ===
=== Miscellaneous ===
* Changed bounds evaluation in UniformFactory, BetaFactory
* Worked around bug #864 (parallel segfault in BernsteinCopulaFactory)
=== Bug fixes ===
* #890 (Cannot build triangular distribution)
* #891 (Viewer issue with Pairs drawables)
* #895 (Trouble reading CSV files with separators in description)
* #896 (Python iteration in ProcessSample leads to capacity overflow)
* #897 (Bug in Graph::draw with small data)
* #898 (Could not save/load some persistent classes)
* #899 (PythonDistribution copula crash when parallelism is active)
* #902 (NormalGamma constructor builds wrong link function)
* #905 (Bogus MaternModel::setParameter)
* #906 (t_LevelSetMesher_std fails on most non-Intel based chips)
* #907 (Notation)
* #908 (Documentation: change titles)
* #909 (Wrong argument type in the API doc)
* #910 (Graph of a d-dimensionnal distribution)
* #911 (In the Field class, the getSample and getValues methods are duplicate)
* #912 (Wrong description of Histogram constructor parameters)
* #914 (KarhunenLoeveQuadratureAlgorithm crashes for covariance models of dimension>1)
* #917 (Bug in RandomMixture::computeCDF())
* #918 (The class SobolIndicesAlgorithm has a draw method which has no example)
* #919 (Wrong simplification mechanism in MarginalTransformationEvaluation for Exponential distribution)
* #921 (Cannot print a FixedExperiment when built from sample and weight)
* #923 (Fix ExponentialModel::getParameter for diagonal correlation)
* #924 (Probleme with the factory of a Generalized Pareto distribution)
* #927 (Functional chaos is memory hungry)
* #929 (The labels of the sensitivity analysis graphics are poor)
* #930 (The getMean method has a weird behavior on parametrized distribution)
== 1.9 release (2017-04-18) == #release-1.9
=== Library ===
==== Major changes ====
* Integrate otlhs module
* New function API
* Canonical format low-rank tensor approximation
* EGO global optimization algorithm
==== New classes ====
* SpaceFillingPhiP, SpaceFillingMinDist, SpaceFillingC2
* LinearProfile, GeometricProfile
* MonteCarloLHS, SimulatedAnnealingLHS
* LHSResult
* MultiStart
* MethodOfMomentsFactory
* SymbolicFunction, AggregatedFunction, ComposedFunction, DatabaseFunction, DualLinearCombinationFunction
* LinearCombinationFunction, LinearFunction, QuadraticFunction, ParametricFunction, IndicatorFunction
* DistributionTransformation
* GeneralizedExtremeValue
* UniVariateFunctionFamily, UniVariateFunctionFactory, TensorizedUniVariateFunctionFactory
* MonomialFunction, MonomialFunctionFactory
* KarhunenLoeveSVDAlgorithm
* RankMCovarianceModel
* SparseMethod
* CanonicalTensorEvaluation|Gradient, TensorApproximationAlgorithm|Result
* GaussLegendre
* EfficientGlobalOptimization
==== API changes ====
* Removed deprecated SLSQP, LBFGS and NelderMead classes
* Removed deprecated QuadraticCumul class
* Removed classes UserDefinedPair, HistogramPair
* Removed deprecated method WeightedExperiment::getWeight
* Removed deprecated method DistributionFactory::build(NumericalSample, CovarianceMatrix&)
* Removed deprecated distributions alternative parameters constructors, accessors
* Added a generic implementation of the computeLogPDFGradient() method in the DistributionImplementation class.
* Allow Box to support bounds
* Deprecated LinearNumericalMathFunction in favor of LinearFunction
* Deprecated QuadraticNumericalMathFunction in favor of QuadraticFunction
* Deprecated NumericalMathFunction::GetValidConstants|GetValidFunctions|GetValidOperators
* Renamed ComposedNumericalMathFunction to ComposedFunction
* Renamed LinearNumericalMathFunction to LinearFunction
* Swap covModel and basis arguments in KrigingAlgorithm constructors
* Removed useless keepCholesky argument in KrigingAlgorithm constructors
* Renamed OptimizationSolver to OptimizationAlgorithm
* Renamed TemporalNormalProcess to GaussianProcess
* Renamed SpectralNormalProcess to SpectralGaussianProcess
* Renamed GeneralizedLinearModelAlgorithm to GeneralLinearModelAlgorithm
* Renamed GeneralizedLinearModelResult to GeneralLinearModelResult
* Renamed DynamicalFunction to FieldFunction
* Renamed SpatialFunction to ValueFunction
* Renamed TemporalFunction to VertexValueFunction
* Deprecated [gs]etOptimizationSolver methods
* Renamed ProductNumericalMathFunction to ProductFunction
* Deprecated KarhunenLoeveP1Factory, KarhunenLoeveQuadratureFactory
* Deprecated GramSchmidtAlgorithm, ChebychevAlgorithm
* Added getSobolAggregatedIndices() to FunctionalChaosRandomVector
* Added computeWeight() to Mesh
* Deprecated NumericalMathFunction ctors
* Deprecated NumericalMathFunction for Function
* Deprecated NumericalSample for Sample
* Deprecated NumericalPoint[WithDescription] for Point[WithDescription]
* Deprecated ResourceMap::[SG]AsNumericalScalar for [SG]AsScalar
* Deprecated SpecFunc::*NumericalScalar*
* Deprecated NumericalScalar for Scalar
* Deprecated NumericalComplex for Complex
* Deprecated DistributionImplementation::getGaussNodesAndWeights
=== Python module ===
=== Miscellaneous ===
=== Bug fixes ===
* #351 (FORM is it possible to hav a ".setMaximumNumberOfEvaluations")
* #729 (KDTree & save)
* #774 (Exact limits of a normal distribution with unknown mean and variance)
* #866 (Check the parameter estimate for the kriging model)
* #869 (The ProductNumericalMathFunction class has no example)
* #871 (GeneralizedExponential P parameter : int or float ?)
* #872 (Cannot draw a Text drawable using R)
* #874 (Compatibility between a distribution factory and an alternate parametrization not checked)
* #875 (TruncatedNormalFactory randomly crashes)
* #876 (Bad time grid in StationaryCovarianceModelFactory::build)
* #877 (centered whitenoise limitation)
* #878 (Viewer does not take into account labels in Contour)
* #879 (Incomplete arguments in FunctionalChaosRandomVector docstrings)
* #882 (RandomMixture segfaults with Dirac)
* #883 (VisualTest.DrawHistogram should rely on Histogram.drawPDF)
* #886 (Bogus RandomMixture::getSupport)
* #887 (Bogus PDF evaluation in RandomMixture with mix of continuous/discrete variables)
* #888 (Bogus RandomMixture::getSample)
== 1.8 release (2016-11-18) == #release-1.8
=== Library ===
==== Major changes ====
* Changed the default orthonormalization algorithm of StandardDistributionPolynomialFactory from GramSchmidtAlgorithm to AdaptiveStieltjesAlgorithm
* New api for sensitivity analysis
* New methods to compute confidence regions in Distribution
==== New classes ====
* SubsetSampling
* AdaptiveDirectionalSampling
* KarhunenLoeveQuadratureFactory
* SobolIndicesAlgorithm
* SaltelliSensitivityAlgorithm
* MartinezSensitivityAlgorithm
* JansenSensitivityAlgorithm
* MauntzKucherenkoSensitivityAlgorithm
* SoizeGhanemFactory
* LevelSetMesher
* HistogramPolynomialFactory
* ChebychevFactory
* FourierSeriesFactory, HaarWaveletFactory
* OrthogonalProductFunctionFactory
==== API changes ====
* Removed deprecated (AbdoRackwitz|Cobyla|SQP|TNC)SpecificParameters classes
* Removed AbdoRackwitz|Cobyla|SQP::[gs]etLevelFunction|[gs]etLevelValue
* Removed deprecated OptimizationSolver::setMaximumIterationsNumber
* Removed deprecated method Distribution::setParametersCollection(NP)
* Removed deprecated PersistentFactory string constructor
* Deprecated QuadraticCumul class in favor of TaylorExpansionMoments
* Renamed __contains__ to contains
* Modified NumericalMathFunction::[sg]etParameter to operate on NumericalPoint instead NumericalPointWithDescription
* Add NumericalMathFunction::[sg]etParameterDescription to access the parameter description
* Deprecated classes UserDefinedPair, HistogramPair
* Removed SensitivityAnalysis class
* Deprecated SLSQP, LBFGS and NelderMead classes in favor of NLopt class
* Deprecated LAR in favor of LARS
* Deprecated DistributionFactory::build(NumericalSample, CovarianceMatrix&)
* Deprecated distributions alternative parameters constructors, accessors
* Swap SpectralModel scale & amplitude parameters: CauchyModel, ExponentialCauchy
=== Python module ===
* Added the possibility to distribute PythonFunction calls with multiprocessing
=== Miscellaneous ===
* Improved the computeCDF() method of Normal
* Added the computeMinimumVolumeInterval(), computeBilateralConfidenceInterval(), computeUnilateralConfidenceInterval() and computeMinimumVolumeLevelSet() methods to compute several kind of confidence regions in Distribution
* Added HarrisonMcCabe, BreuschPagan and DurbinWatson tests to test homoskedasticity, autocorrelation of linear regression residuals
* Added two samples Kolmogorov test
* Improved the speed of many algorithms based on method binding
* Added more options to control LHSExperiment and LowDiscrepancyExperiment
* Improved the IntervalMesher class: now it takes into account the diamond flag
* Shortened ResourceMap keys to not contain 'Implementation'
* Improved the performance of Classifier/MixtureClassifier/ExpertMixture
=== Bug fixes ===
* #535 (parallel-threads option cannot be changed at runtime with TBB)
* #565 (The SensitivityAnalysis class manages only one single output.)
* #604 (Bug concerning the NonCentralStudent distribution)
* #698 (KernelSmoothing() as a factory)
* #786 (Bug in sensitivity analysis)
* #802 (Python issue with ComplexMatrix::solveLinearSystem)
* #803 (prefix openturns includes)
* #813 (Error when multiplying a Matrix by a SymmetricMatrix)
* #815 (ConditionedNormalProcess test fails randomly)
* #820 (Python distribution fails randomly when computing the PDF over a sample)
* #822 (Incorect Matrix / point operations with cast)
* #824 (Confusing behavior of NumericalSample::sort)
* #828 (ImportFromCSVFile fails on a file created by exportToCSVFile)
* #830 (more optim algos examples)
* #831 (Missing get/setParameter in OpenTURNSPythonFunction)
* #833 (Homogeneity in Covariance Models)
* #837 (TruncatedDistribution::setParameter segfaults)
* #838 (Symmetry of SymmetricMatrix not always enforced)
* #840 (Remove WeightedExperiment::getWeight)
* #841 (Better CovarianceModelCollection in Python)
* #842 (Better ProcessCollection in Python)
* #843 (Remove all the specific isCopula() methods)
* #848 (Inverse Wishart sampling)
* #849 (Ambiguous NumericalSample::computeQuantile)
* #853 (Switch the default for normalize boolean from TRUE to FALSE in ot.GeneralizedLinearModelAlgorithm)
* #854 (InverseWishart.computeLogPDF)
* #861 (document HMatrix classes)
== 1.7 release (2016-01-27) == #release-1.7