ee3b5fe | Paul Koch | 16 March 2024, 07:36:21 UTC | version 0.6.0 release | 16 March 2024, 08:56:38 UTC |
7c2284a | Paul Koch | 16 March 2024, 06:50:10 UTC | update release process and hyperparameters | 16 March 2024, 06:50:10 UTC |
139fddc | Paul Koch | 16 March 2024, 00:33:32 UTC | add warning regarding using monotonic constraints during fitting | 16 March 2024, 00:33:32 UTC |
86e60fe | Paul Koch | 15 March 2024, 20:55:18 UTC | handle monotonicity when using random boosting | 15 March 2024, 22:30:43 UTC |
e660298 | Paul Koch | 15 March 2024, 18:44:05 UTC | change smoothing rounds back to 200 and 50 for interaction smoothing rounds | 15 March 2024, 18:44:05 UTC |
274379c | Paul Koch | 14 March 2024, 23:07:22 UTC | restore tests for python 3.12 | 14 March 2024, 23:07:22 UTC |
fc0c81c | Paul Koch | 14 March 2024, 21:59:46 UTC | change cyclic_progress to accept bools and make default True | 14 March 2024, 21:59:46 UTC |
bf3189d | Paul Koch | 14 March 2024, 21:20:04 UTC | remove DecisionListClassifier from documentation since skope-rules is no longer maintained and the documentation no longer builds | 14 March 2024, 21:55:42 UTC |
7a30293 | Paul Koch | 14 March 2024, 20:59:43 UTC | rename breakpoint_iteration to best_iteration to align with XGBoost naming | 14 March 2024, 21:03:48 UTC |
019c713 | Paul Koch | 14 March 2024, 18:57:23 UTC | update documentation to use cyclic_progress variable instead of greediness (which has been deprecated) | 14 March 2024, 18:57:23 UTC |
a6f3120 | Paul Koch | 14 March 2024, 18:48:11 UTC | rename greediness to greedy_ratio to better match it's new behavior | 14 March 2024, 18:49:45 UTC |
aeec521 | Paul Koch | 14 March 2024, 05:57:03 UTC | reformatting | 14 March 2024, 05:57:03 UTC |
008e63b | Paul Koch | 14 March 2024, 05:21:23 UTC | expose boosting based monotonicity | 14 March 2024, 05:52:59 UTC |
e1ead2e | Paul Koch | 14 March 2024, 00:31:20 UTC | add monotonicity to test api | 14 March 2024, 00:31:20 UTC |
3328880 | Paul Koch | 13 March 2024, 23:54:48 UTC | update to use python 3.9 for building | 13 March 2024, 23:54:48 UTC |
7a8c0c5 | Paul Koch | 13 March 2024, 02:44:36 UTC | fix the shap issue in version 0.45 where tree shap has changed the axis where it expresses the class index | 13 March 2024, 02:44:36 UTC |
66e04fa | Paul Koch | 13 March 2024, 02:06:10 UTC | C++ reformat | 13 March 2024, 02:06:10 UTC |
9498ba6 | Paul Koch | 12 March 2024, 23:15:52 UTC | ability to constrain by monotonicity | 12 March 2024, 23:15:52 UTC |
6e01242 | Paul Koch | 12 March 2024, 21:48:34 UTC | use same version of clang-format as VS and shaping for future monotonicity | 12 March 2024, 21:48:34 UTC |
d5ec3fa | Paul Koch | 09 March 2024, 06:09:53 UTC | update URLs to https | 09 March 2024, 06:09:53 UTC |
9631291 | Paul Koch | 05 March 2024, 15:47:08 UTC | update hyperparameter values | 05 March 2024, 15:47:49 UTC |
5813304 | DerWeh | 05 March 2024, 00:12:51 UTC | TST: loosen requirements for tests (#520) Signed-off-by: DerWeh <andreas.weh@web.de> | 05 March 2024, 00:12:51 UTC |
1877154 | Paul Koch | 04 March 2024, 19:48:27 UTC | modify boost loop to handle fractional cyclic_progress values | 04 March 2024, 19:48:27 UTC |
cc75f87 | Paul Koch | 04 March 2024, 19:25:45 UTC | change cyclic_progress to accept a float percentage | 04 March 2024, 19:28:26 UTC |
3b9012b | Paul Koch | 04 March 2024, 07:39:06 UTC | update greediness parameter to accept numbers greater than 1.0 in accordance to the newer interpretation of that parameter and remove the refresh_rate parameter and expose the cyclic_progress parameter | 04 March 2024, 07:39:06 UTC |
620a496 | Paul Koch | 04 March 2024, 02:03:25 UTC | replace refresh_period with cyclic_progress bool which optionally turns off boosting during the cyclic rounds in order to update the gain values instead | 04 March 2024, 04:14:08 UTC |
44d3c69 | Paul Koch | 04 March 2024, 00:51:37 UTC | change the greediness algorithm to allow for better control of the amount of greedy steps | 04 March 2024, 01:36:54 UTC |
ec870b2 | dependabot[bot] | 01 March 2024, 17:37:36 UTC | Bump es5-ext from 0.10.61 to 0.10.63 in /python/stitch (#516) Bumps [es5-ext](https://github.com/medikoo/es5-ext) from 0.10.61 to 0.10.63. - [Release notes](https://github.com/medikoo/es5-ext/releases) - [Changelog](https://github.com/medikoo/es5-ext/blob/main/CHANGELOG.md) - [Commits](https://github.com/medikoo/es5-ext/compare/v0.10.61...v0.10.63) --- updated-dependencies: - dependency-name: es5-ext dependency-type: indirect ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> | 01 March 2024, 17:37:36 UTC |
0e8dad8 | Paul Koch | 29 February 2024, 09:05:29 UTC | remove experimental tag on ML2SQL, update readme with new links, and update hyperparameter tuning recommendations | 01 March 2024, 01:20:36 UTC |
bcf86e6 | DerWeh | 29 February 2024, 23:08:54 UTC | Increase compatibility of EBM with scikit-learn (#518) * MAINT: sort imports Signed-off-by: DerWeh <andreas.weh@web.de> * ENH: sklearn compatibility: warn about 2d y Signed-off-by: DerWeh <andreas.weh@web.de> * ENH: sklearn compatibility: use None instead of [] Signed-off-by: DerWeh <andreas.weh@web.de> * ENH: sklearn: add tag that EBM supports NaN inputs Signed-off-by: DerWeh <andreas.weh@web.de> * ENH: raise ValueError for complex data (sklearn) Raising a ValueError is compatible with sklearn and therefore the expected behavior. TypeError is inappropriate in the Python world, as `type(X)` has the correct type (an np.ndarray). Just the X.dtype is wrong. This is a NumPy concept and not a native Python concept. Signed-off-by: DerWeh <andreas.weh@web.de> * ENH: sklearn: check also for inf in labels Signed-off-by: DerWeh <andreas.weh@web.de> * BLD: bump sklearn dependency Signed-off-by: DerWeh <andreas.weh@web.de> * ENH: test sklearn compatibility Signed-off-by: DerWeh <andreas.weh@web.de> * Revert "ENH: sklearn compatibility: warn about 2d y" Support of 2d y we `y.shape[-1] = 1` is desired This reverts commit 230cddd2ec4c7b7f60c684b0c8ffa83a6e2b4d92. Signed-off-by: DerWeh <andreas.weh@web.de> * ENH: update tags Signed-off-by: DerWeh <andreas.weh@web.de> * TST: update selection of scikit-learn checks Signed-off-by: DerWeh <andreas.weh@web.de> * MAINT: drop old skipped scikit-learn checks Signed-off-by: DerWeh <andreas.weh@web.de> --------- Signed-off-by: DerWeh <andreas.weh@web.de> | 29 February 2024, 23:08:54 UTC |
0cf3f2f | Paul Koch | 29 February 2024, 07:00:22 UTC | fix bug in convert_categorical_to_continuous that did not handle scenarios where there is only one continuous section | 29 February 2024, 07:00:22 UTC |
a7e345a | DerWeh | 29 February 2024, 06:28:29 UTC | MAINT: Rebase EBM utils (#517) * MAINT: avoid code duplications Signed-off-by: DerWeh <andreas.weh@web.de> * MAINT: early error to reduce complexity Signed-off-by: DerWeh <andreas.weh@web.de> * MAINT: avoid unnecessary allocations Signed-off-by: DerWeh <andreas.weh@web.de> * MAINT: simplify function conversion function Signed-off-by: DerWeh <andreas.weh@web.de> * MAINT: refactor midpoint into own function Signed-off-by: DerWeh <andreas.weh@web.de> * MAINT: sort imports Signed-off-by: DerWeh <andreas.weh@web.de> * MAINT: check directly for finite numbers Signed-off-by: DerWeh <andreas.weh@web.de> * ENH: handle special case of empty cuts Signed-off-by: DerWeh <andreas.weh@web.de> * TST: minimal test for conversion between cuts and intervals Signed-off-by: DerWeh <andreas.weh@web.de> --------- Signed-off-by: DerWeh <andreas.weh@web.de> | 29 February 2024, 06:28:29 UTC |
9cbb353 | Paul Koch | 27 February 2024, 10:51:41 UTC | increase the default max_rounds value to 25000 | 27 February 2024, 10:51:41 UTC |
14f140c | Paul Koch | 25 February 2024, 22:12:08 UTC | add Hyperparameters to the documentation | 27 February 2024, 10:49:21 UTC |
abad985 | Paul Koch | 24 February 2024, 16:47:49 UTC | add RMSE to synthetic notebook output | 24 February 2024, 20:53:33 UTC |
24f4f11 | Paul Koch | 15 February 2024, 00:46:27 UTC | add refresh_rate option to public parameters | 24 February 2024, 20:53:32 UTC |
f2ffdd1 | alvanli | 20 February 2024, 19:53:23 UTC | Return self in sweep() (#511) Signed-off-by: alvanli <51011489+alvanli@users.noreply.github.com> | 20 February 2024, 19:53:23 UTC |
fe334e8 | Paul Koch | 08 February 2024, 21:09:06 UTC | update release process notes and readme | 13 February 2024, 21:07:25 UTC |
979d93f | Paul Koch | 08 February 2024, 05:43:02 UTC | update interpret to version 0.5.1 | 08 February 2024, 05:58:56 UTC |
5d56f07 | Paul Koch | 07 February 2024, 10:20:25 UTC | add internal option to disable smoothing for nominals | 07 February 2024, 22:08:44 UTC |
7b0c331 | Paul Koch | 07 February 2024, 06:04:29 UTC | add interface to access the nominal feature definitions from the dataset object | 07 February 2024, 06:04:29 UTC |
3a6abe8 | Paul Koch | 07 February 2024, 03:49:01 UTC | small changes to EBM defaults | 07 February 2024, 05:16:17 UTC |
f8cb826 | Paul Koch | 06 February 2024, 23:36:38 UTC | re-expose min_samples_leaf | 06 February 2024, 23:47:43 UTC |
ed3501b | Paul Koch | 06 February 2024, 20:36:22 UTC | change EBM parameters -> set tolerance to zero, set regression min_hessian to 1.01 | 06 February 2024, 20:38:36 UTC |
f1115eb | Paul Koch | 06 February 2024, 12:44:34 UTC | go back to hessian boosting during smoothing because it's faster | 06 February 2024, 12:44:34 UTC |
240a620 | Paul Koch | 06 February 2024, 09:08:12 UTC | update default EBM parameters | 06 February 2024, 09:08:12 UTC |
edab834 | Paul Koch | 06 February 2024, 08:29:42 UTC | update docs to remove references to min_samples_leaf | 06 February 2024, 08:29:42 UTC |
52b8966 | Paul Koch | 06 February 2024, 08:22:11 UTC | change early stopping tolerance to be a percentage | 06 February 2024, 08:22:11 UTC |
b65426f | Paul Koch | 06 February 2024, 01:03:35 UTC | early stop on no improvement (if there is a validation set) | 06 February 2024, 07:45:22 UTC |
4b5cef6 | Paul Koch | 05 February 2024, 22:28:00 UTC | resolve python tests that needed updating for hessians, and update R to use the min_hessian | 05 February 2024, 22:28:00 UTC |
b798ed5 | Paul Koch | 05 February 2024, 13:30:48 UTC | change split limiting metric to the hessian | 05 February 2024, 13:58:42 UTC |
81e4ef9 | Paul Koch | 05 February 2024, 12:13:29 UTC | add the ability to limit cuts via hessians | 05 February 2024, 12:19:48 UTC |
eb40978 | Paul Koch | 05 February 2024, 11:28:17 UTC | restructure some code for future change to use minHessian | 05 February 2024, 11:28:17 UTC |
0294197 | Paul Koch | 05 February 2024, 07:42:30 UTC | enable Newton during interaction detection | 05 February 2024, 07:42:30 UTC |
a52a803 | Paul Koch | 05 February 2024, 07:19:18 UTC | simplify flag comparison operations and flip the flag for CalcInteractions from enable to disable for Newton | 05 February 2024, 07:19:18 UTC |
b2fbba5 | Paul Koch | 04 February 2024, 04:14:14 UTC | honor the use of the CalcInteractionFlags_EnableNewton flag | 04 February 2024, 04:14:14 UTC |
b9b27b2 | Paul Koch | 03 February 2024, 11:57:40 UTC | honor the TermBoostFlags_DisableNewtonGain flag | 03 February 2024, 12:07:23 UTC |
61fa960 | Paul Koch | 03 February 2024, 10:58:29 UTC | honor the use of the TermBoostFlags_DisableNewtonUpdate flag | 03 February 2024, 10:58:29 UTC |
adf71ce | Paul Koch | 03 February 2024, 04:32:19 UTC | add interaction_smoothing_rounds parameter | 03 February 2024, 04:46:43 UTC |
d05ced7 | Paul Koch | 03 February 2024, 03:45:22 UTC | make the early stopping more consistent | 03 February 2024, 03:45:22 UTC |
4a85b1e | Paul Koch | 03 February 2024, 02:23:18 UTC | change GetBestTermScore to handle scenarios when the validation set has zero samples | 03 February 2024, 02:27:38 UTC |
5eff4df | DerWeh | 29 January 2024, 19:34:09 UTC | MAINT: pull if clauses out of loop (#503) Signed-off-by: Weh Andreas <andreas.weh@student.uni-augsburg.de> | 29 January 2024, 19:34:09 UTC |
b877fb1 | Paul Koch | 28 January 2024, 05:59:39 UTC | reformat C++ with clang-format | 28 January 2024, 06:00:17 UTC |
e6b74cd | Paul Koch | 27 January 2024, 20:27:13 UTC | add .clang-format file for C++ project | 28 January 2024, 05:57:49 UTC |
7bd487d | Paul Koch | 27 January 2024, 01:26:21 UTC | reformat | 27 January 2024, 01:28:44 UTC |
8e30f7e | dependabot[bot] | 26 January 2024, 00:13:51 UTC | Bump jinja2 from 3.1.2 to 3.1.3 in /docs (#501) Bumps [jinja2](https://github.com/pallets/jinja) from 3.1.2 to 3.1.3. - [Release notes](https://github.com/pallets/jinja/releases) - [Changelog](https://github.com/pallets/jinja/blob/main/CHANGES.rst) - [Commits](https://github.com/pallets/jinja/compare/3.1.2...3.1.3) --- updated-dependencies: - dependency-name: jinja2 dependency-type: direct:production ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> | 26 January 2024, 00:13:51 UTC |
25a5be6 | Kshitij Dahal | 26 January 2024, 00:13:03 UTC | Update README.md (#502) Signed-off-by: Kshitij Dahal <63812052+geokshitij@users.noreply.github.com> | 26 January 2024, 00:13:03 UTC |
3c3c19b | Paul Koch | 25 January 2024, 06:45:13 UTC | add links to README | 25 January 2024, 06:45:13 UTC |
a8f0ae1 | Paul Koch | 21 January 2024, 07:16:18 UTC | fix paper link location in READMe | 21 January 2024, 07:16:18 UTC |
dd1d01c | Paul Koch | 21 January 2024, 07:06:09 UTC | add link to SynapseML on README | 21 January 2024, 07:06:09 UTC |
48f0e61 | Paul Koch | 20 January 2024, 01:32:01 UTC | add option to disable synthetic target inputs | 20 January 2024, 01:32:01 UTC |
bbc20af | Paul Koch | 19 January 2024, 23:50:54 UTC | change interactions to also use smoothing rounds when smoothing is specified | 19 January 2024, 23:50:54 UTC |
279af11 | Paul Koch | 19 January 2024, 11:19:05 UTC | add interaction with the categorical to synthetic dataset | 19 January 2024, 11:26:38 UTC |
12d8f03 | Paul Koch | 18 January 2024, 17:25:44 UTC | add ability to output pandas, scipy and str data from synthetic generator | 19 January 2024, 07:38:24 UTC |
9be98cd | Paul Koch | 15 January 2024, 00:32:35 UTC | add option to vary the proportions of higher classes chosen in the synthetic dataset generator | 15 January 2024, 08:01:09 UTC |
3281393 | Paul Koch | 14 January 2024, 01:22:48 UTC | add links to README | 14 January 2024, 01:22:48 UTC |
98894b4 | Paul Koch | 13 January 2024, 09:57:46 UTC | change monoclassification format to include a single -inf score per sample/bin | 13 January 2024, 10:55:55 UTC |
4d996d5 | Paul Koch | 11 January 2024, 23:54:30 UTC | fix bug that occurs when there are duplicate columns in pandas | 11 January 2024, 23:54:30 UTC |
b4a6196 | Paul Koch | 11 January 2024, 22:58:30 UTC | remove dependency on distutils due to its depreciation | 11 January 2024, 23:26:36 UTC |
b57f814 | DerWeh | 09 January 2024, 23:14:44 UTC | BUG: define missing logger objects (#496) Signed-off-by: DerWeh <andreas.weh@web.de> | 09 January 2024, 23:14:44 UTC |
d576a62 | Paul Koch | 09 January 2024, 08:20:51 UTC | make the category counts and distributions configurable in the synthetic dataset generator and rename to make_synthetic | 09 January 2024, 09:07:39 UTC |
0c1e444 | Paul Koch | 09 January 2024, 04:35:38 UTC | add sawtooth wave to synthetic dataset, make the categorical categories non-uniform, and other tweaks to improve synthetic dataset | 09 January 2024, 07:35:30 UTC |
5eb9306 | Paul Koch | 08 January 2024, 01:28:38 UTC | change x ** 3 into square wave in the synthetic dataset | 08 January 2024, 12:11:38 UTC |
45ee7a8 | Paul Koch | 06 January 2024, 10:26:01 UTC | special case the 1 dimension option for interaction detection when hessian is unused because it is a scenario that would have been a bug in the future | 07 January 2024, 10:25:58 UTC |
22f71a2 | Paul Koch | 06 January 2024, 04:40:32 UTC | reformatting | 06 January 2024, 04:40:32 UTC |
d68b6ec | Paul Koch | 06 January 2024, 03:20:46 UTC | update readme links | 06 January 2024, 03:55:18 UTC |
b9d577e | Paul Koch | 06 January 2024, 00:00:57 UTC | added ability to specify percentage of missing values | 06 January 2024, 00:16:09 UTC |
e604c32 | Paul Koch | 04 January 2024, 23:04:29 UTC | add list of urls | 04 January 2024, 23:04:29 UTC |
94d36bc | Paul Koch | 04 January 2024, 22:31:15 UTC | reduce the number of categories to the point all of them show up in the docs UI and remove the dashboard examples that don't work in the inline docs | 04 January 2024, 22:31:15 UTC |
a3aed93 | Paul Koch | 04 January 2024, 06:54:07 UTC | add redirects to the docs | 04 January 2024, 06:54:07 UTC |
ea82256 | Paul Koch | 04 January 2024, 05:40:38 UTC | remove unnecessary copies of images in the docs (since jupyter book already copies them) | 04 January 2024, 06:37:26 UTC |
6b31669 | Paul Koch | 04 January 2024, 03:45:13 UTC | change partial dependence and morris sensitivity to use Fortran ordered float data for speed | 04 January 2024, 03:45:13 UTC |
2e27101 | Paul Koch | 04 January 2024, 03:09:18 UTC | disable the help text in the EBM explanations, except for in Synapse. The additional help text was adding scroll bars to the notebooks due to the added height | 04 January 2024, 03:09:18 UTC |
48aa026 | Paul Koch | 04 January 2024, 02:17:07 UTC | add padding to inline iframe to remove jupyter notebook scroll bars | 04 January 2024, 02:17:07 UTC |
86fa037 | Paul Koch | 04 January 2024, 02:15:19 UTC | increase sample size on the DP test because DP needs sufficient samples to maintain privacy | 04 January 2024, 02:15:19 UTC |
b83604a | Paul Koch | 04 January 2024, 00:05:15 UTC | replace adult dataset with synthetic dataset for the tests since the adult dataset causes URL load errors occasionally | 04 January 2024, 01:31:39 UTC |
b35987d | Paul Koch | 03 January 2024, 22:35:11 UTC | add support for named features and negative indexes in the interactions specification | 03 January 2024, 22:35:11 UTC |
1fb233a | Paul Koch | 03 January 2024, 08:37:11 UTC | add missing group importances notebook pngs and fix DecisionRuleClassifier link | 03 January 2024, 08:56:58 UTC |
9218da6 | Paul Koch | 03 January 2024, 08:07:14 UTC | change interpretable regression notebook to use the California housing dataset from scikit-learn | 03 January 2024, 08:07:14 UTC |
6d5efb3 | Paul Koch | 02 January 2024, 22:03:06 UTC | add links to the examples folder in the docs and fixup relative links between the API references and the algorithm descriptions | 02 January 2024, 22:55:39 UTC |