0.9.1

What's Changed

• Remove useless requirement files. by @vnmabus in #584
• Add standard deviation as a function by @pcuestas in #566
• Change example from Basemap to Cartopy. by @vnmabus in #586
• Feature/concurrent regression by @rafa9811 in #531
• Functional PLS for dimensionality reduction and regression by @Ddelval in #548

Full Changelog: 0.9...0.9.1

Version 0.9

What's Changed

• Document conda installation. by @vnmabus in #511
• Deprecate 3 as the default number of components in FPCA by @Ddelval in #510
• Add an example showing how to create new bases for scikit-fda. by @vnmabus in #517
• New example for creating interpolation/extrapolation methods. by @vnmabus in #518
• FPCA Regression by @Ddelval in #466
• Add CSV loading example to the tutorial. by @vnmabus in #520
• Update THANKS.txt by @jltorrecilla in #523
• Fix inner product integrate by @m5signorini in #522
• Simplification example make_gaussian_process by @jltorrecilla in #526
• Fix code coverage. by @vnmabus in #532
• Add missing data imputation via interpolation by @vnmabus in #533
• Make CustomBasis documentation accesible by @Ddelval in #534
• SRSF formula doc fix by @ego-thales in #542
• Speed up the calculation of the penalization matrix for FDataGrid by @Ddelval in #519
• Change type of axis in FData's mean method to accept zero. by @Ddelval in #546
• Covariance function as tensor product by @m5signorini in #505
• Polish fpca regression example by @Ddelval in #552
• Fix IndexError in FPCAPlot by @Quentin62 in #554
• ddof parameter in fdatacov by @pcuestas in #556
• Improve list of contributors by @vnmabus in #557
• FDataGrid.restrict option with_bounds by @ego-thales in #561
• Deleted forgotten __doc__ bits by @eliegoudout in #565
• Add a version switcher for the docs in the web. by @vnmabus in #571
• Update citation and See also in DTMClassifier class by @E105D104U125 in #573
• Change ddof argument to correction in var and cov functions by @pcuestas in #578
• BSplineBasis initialization incorrect error message by @pcuestas in #577
• Smoothing in several dimensions by @ElenaPetrunina in #436
• Set correction=0 by default in cov and var functions by @pcuestas in #582

New Contributors

• @jltorrecilla made their first contribution in #523
• @ego-thales made their first contribution in #542
• @Quentin62 made their first contribution in #554
• @pcuestas made their first contribution in #556
• @allcontributors made their first contribution in #564
• @eliegoudout made their first contribution in #565
• @E105D104U125 made their first contribution in #573

Full Changelog: 0.8.1...0.9

Version 0.8.1

What's Changed

• Fix packaging.
• Fix FPCA grid implementation with regularization by @Ddelval in #485
• Hotfix: Correct the order of the parameters for the FPCA in Mahalanobis by @m5signorini in #506

Full Changelog: 0.8...0.8.1

Version 0.8

What's Changed

• Local averages construction by @alvaro-castillo in #418
• Kernel Regression and Hat Matrix by @ElenaPetrunina in #417
• Implement unary minus for FData by @jiduque in #425
• Remove alpha parameter of Magnitude-Shape plot and outlier detector. by @vnmabus in #424
• Add deprecated classes and modules for Kernel Smoothing. by @vnmabus in #423
• Update _fourier.py by @ElenaPetrunina in #431
• Change example of per class transformer by @alvaro-castillo in #422
• LLR for non orthonormal basis by @ElenaPetrunina in #430
• Occupation measure construction by @alvaro-castillo in #396
• Fix HatMatrix documentation by @ElenaPetrunina in #432
• Add "solver" and "max_iter" parameters to class LogisticRegression by @dSerna4 in #435
• Number of up crossings construction by @alvaro-castillo in #419
• Fixing a typo on Predictors sklearn vs skfda by @alvaro-castillo in #437
• Organization of preprocessing module by @alvaro-castillo in #439
• Paremeterized Functional QDA by @alvaro-castillo in #427
• Style fixes on classification example by @alvaro-castillo in #444
• Update classification rst by @alvaro-castillo in #445
• Fix docstrings for read the docs style by @alvaro-castillo in #446
• Comment fix by @alvaro-castillo in #447
• Add rst by @alvaro-castillo in #449
• Mahalanobis by @pedrorponga in #397
• Function transformers by @alvaro-castillo in #442
• Integration with binder. by @vnmabus in #450
• Add return_X_y parameter for UCR datasets. by @vnmabus in #453
• fix examples by @alvaro-castillo in #451
• Adding a comparison table to the example by @alvaro-castillo in #452
• Feature Construction Moments by @alvaro-castillo in #438
• Add MCO dataset. by @vnmabus in #455
• Add NOx dataset. by @vnmabus in #456
• Move tests inside package. by @vnmabus in #459
• Ignore the entire .vscode folder by @Ddelval in #463
• Improve k_neighbors example by @Ddelval in #462
• Feature/refactor neighbors by @vnmabus in #465
• Speed up evaluation of FDataGrid by @vnmabus in #474
• Documentation fixes, weights deprecation and linter warnings by @Ddelval in #467
• Update plot_kernel_smoothing.py by @ElenaPetrunina in #473
• Support for dataframes by @rafa9811 in #390
• Error sphinx windows by @pedrog99 in #477
• Fix wrong variable name in fpca by @Ddelval in #478
• Score functions for FData by @ElenaPetrunina in #429
• Make weights parameter private by @Ddelval in #483
• Change quadrature in fpca test by @Ddelval in #489
• Issue/classifier classes/test by @m5signorini in #481
• Issue/classifier_classes/kneighbours by @m5signorini in #491
• Issue/classifier classes/centroid by @m5signorini in #492
• Issue/classifier_classes/ddc by @m5signorini in #493
• Issue/classifier classes/qda by @m5signorini in #494
• Feature/rename basis by @Ddelval in #490
• Issue/classifier_classes/logreg by @m5signorini in #496
• Changed Gramian matrix to Gram matrix for better consistency. Issue #501. by @Saumya-ranjan in #502
• Custom Basis by @Ddelval in #495
• Move metadata to pyproject.toml by @vnmabus in #503

New Contributors

• @jiduque made their first contribution in #425
• @Ddelval made their first contribution in #463
• @rafa9811 made their first contribution in #390
• @pedrog99 made their first contribution in #477
• @m5signorini made their first contribution in #481
• @Saumya-ranjan made their first contribution in #502

Full Changelog: 0.7.1...0.8

Version 0.7.1

Fix license in Zenodo citation file.

Version 0.7

• TikhonovRegularization renamed to L2Regularization. The old name is deprecated. See #409 for details.
• Add the option for MagnitudeShapePlot to plot the inlier ellipse.
• Additional bug fixes.

Version 0.6.1

This version only adds small bug fixes.

Version 0.6

Some highlights of this version:

• Add typing support for most functionalities of the package. Now you can use Mypy to check for errors.
• New depth-based classifiers
• New visualization functions
  • Parametric plot
  • Outliergram
  • Multiple interactive plots
• New historical functional regression model
• Other minor additions and fixes

Version 0.5

Some highlights of the new functionality:

• Depth functions refactored
• Removed C code and added fdasrsf as a dependency
• Added new variable selection methods
• Added Maximum Depth classifier
• Optimized FDataGrid inner product

Version 0.4

Adds functional component analysis, ANOVA and Hotelling tests, basis representation for multivariate and vector valued functions and Tikhonov regularization.