Releases: vortex-exoplanet/VIP
VIP v2.0.0
What's Changed
-
BLD: move static metadata to
pyproject.tomland explicitly specify build backend and requirements by @neutrinoceros in #670 -
DEP: move dependency specs to static metadata by @neutrinoceros in #671
-
DEP: drop support for Python 3.8 and 3.9, require 3.10 or newer by @neutrinoceros in #672
-
TST: setup basic concurrency rules for CI (cancel in progress, obsolete runs) by @neutrinoceros in #676
-
MNT: migrate package layout from flat to src by @neutrinoceros in #673
-
BLD: switch build backend to flit-core by @neutrinoceros in #681
-
DEP/RFC: refactor out dependency on
munchby @neutrinoceros in #682 -
TST: add explicit support for CPython 3.12 and 3.13 by @neutrinoceros in #674
-
TST: use
coveragedirectly, ditchpytest-covby @neutrinoceros in #679 -
ENH: add support for loading Dataset from a
pathlib.Pathby @neutrinoceros in #680 -
upgrades for
cube_inject_fakediskby @IainHammond in #669 -
pca_annularcan now accept delta_sep as a list by @IainHammond in #669 -
Bug fix for max ncomp allowed in PCA RDI or ARDI by @VChristiaens in #666
-
Minor bug fixes and new features for PCA functions by @VChristiaens in #684
-
Major version update for: new flat architecture adopted in VIP + compatibility for Python 3.10 to 3.13 + pyproject.toml installation by @VChristiaens in #685
New Contributors
- @neutrinoceros made their first contribution in #670
Full Changelog: v1.6.6...v2.0.0
VIP v1.6.6
What's Changed
- Add greedy in setup.py by @Sand-jrd in #662
- New features for IPCA and PCA functions and minor bug fix in median-sub annular version by @VChristiaens in #664
- Reorder the principal components outputted when using svd_mode = ARPACK by @JustLatour in #663
Full Changelog: v1.6.5...v1.6.6
VIP v1.6.5
What's Changed
- more options to get_values_optimize, get_mu_and_sigma, speckle_noise_uncertainty. typo & docstring fixes by @IainHammond in #654
- New features for coordinate conversion and cube_planet_free by @VChristiaens in #657
- New functionalities to vip_hci/fm/scattered_light_disk.py by @jmilou in #659
- Modern multiprocessing for MCMC NEGFC, small fix to tutorial 05A by @IainHammond in #658
- Typo README installation by @Robin-Savonet in #660
- New (I)PCA features and a minor bug fix by @VChristiaens in #661
New Contributors
- @Robin-Savonet made their first contribution in #660
Full Changelog: v1.6.4...v1.6.5
VIP v1.6.4
What's Changed
-
New features for NEGFC by @VChristiaens in #652 :
- New option for NEGFC to be compatible with input 4D ASDI cube and single binned flux estimate for the companion
- Changed default number of bins for residual speckle uncertainty deviation histogram (this should minimize the risk of raised error)
- Option added for a better mu and sigma estimation passed to the NEGFC-MCMC log-likelihood expression through passing of
f_guessandpsfnparameters, which trigger subtraction of the estimate of the companion from the cube before calculating mu and sigma at the radius of the companion
-
New features for PCA-SADI by @VChristiaens in #652 :
- Added option to smooth PCA-SDI results before second-pass PCA-ADI in double-PCA SADI mode
- PCA-ADI step of 2-stage PCA-SADI is now compatible with a threshold in rotation (delta_rot, source_xy and min_frames_pca now propagated)
-
Minor bug / deprecation fixes:
- Proper propagation of all pca algorithm options to NEGFC, by popping all potential redundant arguments from dictionary (this was done properly for the default pca_annulus function though)
- Updated outputs of cube_recenter_2dfit to be proper numpy ndarrays
Full Changelog: v1.6.3...v1.6.4
VIP v1.6.3
What's Changed
- New options for the full-frame
pca(new 'ref_strategy' keyword) andsignificancefunctions by @VChristiaens in #646 - Adapted argument in
frame_cropfunction for consistency withcube_crop_framesin #646 - Adapted radial sampling in contrast curve function for enhanced robustness to large FWHM values by @VChristiaens in #650
- Torch FFT Rotation & Juillard23 IPCA by @Sand-jrd in #651
- Some deprecation warning fixes and PEP8 formatting by @VChristiaens in #646, and added myst requirement for docs compilation in #647
- Docs: updated tutorials call to download test data + inclusion of RDI tutorial by @VChristiaens in #648
Full Changelog: v1.6.2...v1.6.3
VIP v1.6.2
What's Changed
- more robust mu sigma calculation, new seeing function, memory improvements, typo fixes by @IainHammond in #644
- Additional features for PCA, IPCA and median-sub functions by @VChristiaens in #645:
- median-RDI now possible, both in full-frame and annular mode in
median_sub; - list of npc matching the number of annuli in
pca_annular(for either ADI, RDI or ASDI) now accepted and documented; - improved convergence behaviour in incremental IPCA.
- median-RDI now possible, both in full-frame and annular mode in
- minor bug fix for contrast curve function (which could occur when max pixel radius was not an integer) by @VChristiaens in #645
- improved behaviour for various utilities including
mask_circle,cube_detect_badfr_pxstatsand fixed warning messages incontrast_curvefunction by @VChristiaens in #645
Full Changelog: v1.6.1...v1.6.2
VIP v1.6.1
What's Changed
- DFT shift fix for new scikit, speed improvement for find_nearest by @IainHammond in #638
- Minor bug fixes:
- In SNR map calculation in the case of integer input fwhm by @VChristiaens in #639
- For PCA-RDI with 4D cubes by @VChristiaens in #642
- For mixed array types when creating bad pixel maps
- A few new features by @VChristiaens in #642
- Added compatibility for NEGFC combined with either PCA-ARDI or PCA RDI with data imputation
- IPCA-ARDI now does PCA-ARDI with the initial number of principal components, instead of RDI with the last number of principal components
- Made frame_by_frame mode of cube_fix_badpix_isolated function compatible with input bad pixel map
- Wider multiprocessing support (now compatible with 'spawn' starting method when 'fork' not available) by @VChristiaens in #643
- Bug fix for IPCA-ARDI with DI initialization by @VChristiaens in #643
- Documentation:
- Updated readthedocs.yml by @VChristiaens in #633, #634 & #635
- Updated dosctrings by @VChristiaens in #640
- Added sphinx rtd theme to requirements by @VChristiaens in #636
- Added info about VIP conventions by @VChristiaens in #637
Full Changelog: v1.6.0...v1.6.1
VIP v1.6.0
What's Changed
- Inclusion of IROLL, IPCA, NEGFD routines and new features to NEGFC routines by @VChristiaens in #628
- Replacement of sci_cube_skysub with DIKL by @Sand-jrd in #629
- Adapted cube_collapse function to also accept an input 4D cube by @VChristiaens in #630
- Updated tests and reverted to original contrast curve definition by @VChristiaens in #631
- Updated readme & fixed links in tutorials shown in docs by @VChristiaens in #632
Full Changelog: v1.5.3...v1.6.0
VIP v1.5.3
What's Changed
- Fix for GPU-based SVD calculation options by @VChristiaens in #619
- cube_recenter_satspots with filter_freq by @ElenaKoko in #621
- fixed VIP installation problems related to scikit-image, general improvements for recenter functions and docstrings , PEP8 imports by @IainHammond in #622
- Correction to how the reference dataset is built in do_pca_patch by @Julat300 in #625
- Made the contrast_curve function more robust to classes with inherited attributes by @Julat300 in #626
New Contributors
- @ElenaKoko made their first contribution in #621
- @Julat300 made their first contribution in #625
Full Changelog: v1.5.2...v1.5.3
VIP v1.5.2
What's Changed
- Bug fix for new behaviour of photutils.detect_sources (photutils version>0.7) in roc.compute_binary_map by @VChristiaens in #616
- Improvements to
pcaandpca_annularfunctions, including possibility to calculate results for an inputncomplist by @VChristiaens in #617
Full Changelog: v1.5.1...v1.5.2
Note: version required to use the latest features in Applefy.