- BMTI deltaFs error improvements by @charliematteo in #133
- typo fixed by @wildromi in #134
- updated docs and minor fixes to density_advanced by @charliematteo in #136
- Fix DII unittests for newer Python versions by @FelixWodaczek in #137
- Dii optimization cosine decay by @wildromi in #143
- Bmti error dev by @charliematteo in #141
- codecov: update to v4 by @AldoGl in #144
- BID by @acevedo-s in #140
- Add jax implementation of DII by @vdeltatto in #146
- @acevedo-s made their first contribution in #140
Full Changelog: https://github.com/sissa-data-science/DADApy/compare/v0.3.1...v0.3.2
- bugfix: kstar_gride by @AldoGl in #96
- added cython version for data_overlap by @diegodoimo in #97
- Add basic jackknife for information imbalance (WIP) by @AldoGl in #98
- Fix docs by @AldoGl in #99
- Fixed cross NN with periodic box and added tests. by @ollyfutur in #102
- add @ollyfutur to list of contributors by @AldoGl in #103
- Causality dev by @vdeltatto in #105
- fix warnings related to float64-32 conversion by @AldoGl in #106
- add n_jobs option to compute_nn_distances by @diegodoimo in #109
- fixes issue 108 by @diegodoimo in #110
- Refactor overlap functions and add new test envs by @diegodoimo in #111
- fix Gride range_max + return ranks/datasets sizes in ID scaling functions by @diegodoimo in #112
- add python 12 compatibility by @diegodoimo in #113
- Class imbalance support for return_label_overlap method by @alexserra98 in #114
- renaming decimation and fraction in ID + cleanupoverlap unreachable code by @diegodoimo in #115
- Avoid warning about data type when using float64 by @alexdepremia in #116
- Differentiable Imbalance by @FelixWodaczek in #117
- fix feature_selection unittest by @FelixWodaczek in #118
- Add the BMTI method for density estimation by @AldoGl in #119
- kstar computation is mandatory when calling PAk by @diegodoimo in #120
- fix black linting by @diegodoimo in #121
- DII dev by @vdeltatto in #122
- Differentiable imbalance ex fix2 by @wildromi in #125
- Added instructions for readthedocs Differentiable Information Imbalance by @wildromi in #126
- information imbalance with automatic subsampling by @diegodoimo in #127
- Causality bugfix by @vdeltatto in #129
- Fix error generated in tutorial "notebook_on_intrinsicdim_densityest_clustering.ipynb" by @vdeltatto in #130
- Pvalues by @imacocco in #128
- @ollyfutur made their first contribution in #102
- @alexserra98 made their first contribution in #114
- @FelixWodaczek made their first contribution in #117
Full Changelog: https://github.com/sissa-data-science/DADApy/compare/v0.2.0...v0.3.0
- PaK: solution for singular likelihood hessian and infinite volume shells by @diegodoimo in #77
- Addition clustering adpy v2 and other improvements by @diegodoimo in #78
- adding naive outlier detection to mus gride by @diegodoimo in #80
- Update reference article from ArXiv to published paper by @AldoGl in #81
- optimizing overlaps with labels by @diegodoimo in #82
- fix flake8 errors by @AldoGl in #86
- add initial pyproject.toml by @AldoGl in #88
- update copyright 2021-2022 -> 2021-2023 by @AldoGl in #89
- add kstar-gride id estimator and basic test by @AldoGl in #87
- Develop causality by @vdeltatto in #90
- added Kstar binomial estimator by @imacocco in #93
- 'Return_id_scaling_2NN' now works when the class is initalized with only distance matrices by @diegodoimo in #92
- I3D by @imacocco in #83
- @vdeltatto made their first contribution in #90
Full Changelog: https://github.com/sissa-data-science/DADApy/compare/v0.1.1...v0.2.0
- added a reference to the arxiv package description by @AldoGl in #59
- add warning for datasets with overlapping datapoints by @diegodoimo in #60
- add option to save mus in compute_id_2nn by @diegodoimo in #62
- docs/usage: duly -> dadapy typo found by @alexdepremia by @AldoGl in #64
- docs: added a new documentation page describing the hands-on tutorial given at SISSA by @AldoGl in #65
- New ID tutorial jupyter notebbok/ small fixed to compute_id_twoNN by @diegodoimo in #66
- Updating the examples files by @alexdepremia in #63
- Save mu attribute in GRIDE by @diegodoimo in #68
- solution issue #70 in ADP cython and pure_python version by @diegodoimo in #74
- added vectorized pak version by @diegodoimo in #72
- Changed dendrogram representation by @alexdepremia in #73
- examples: add jupyter Mobius strip Jupyter notebook by @imacocco and @AldoGl in #75
- examples: add Jupyter notebook with beta-hairpin data analysis by @wildromi and @AldoGl in #76
- First release of DADApy