apunkt/ai-privacy-toolkit
1
0
Fork 0
mirror of https://github.com/IBM/ai-privacy-toolkit.git synced 2026-06-08 15:05:13 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions

Commit graph

  • 07aa3969c9
         Merge 3a82a01314 into 57e38ea4fa ibm-mend-app[bot] 2025-09-17 22:22:47 +00:00
  • 3a82a01314
         Add .whitesource configuration file whitesource/configure ibm-mend-app[bot] 2025-09-17 22:22:44 +00:00
  • 734d8ea1cc
         Merge f5acdde8c6 into 57e38ea4fa luxingzhe 2025-08-21 01:06:11 +00:00
  • f5acdde8c6 improve the error message such that it is more specific luxingzhe 2025-08-20 21:02:55 -04:00
  • b550092d39
         Add renovate.json renovate/configure renovate[bot] 2025-05-08 03:09:09 +00:00
  • a7dc7b14ee v.1.0 cmalvr 2025-02-24 00:07:10 +01:00
  • 57e38ea4fa
         Support for many new model output types (#93) main abigailgold 2024-07-03 09:04:59 -04:00
  • d58934a4b6 Remove unused import abigailt 2024-07-03 13:58:17 +03:00
  • bcb7c47cc6 Formatting abigailt 2024-07-03 13:42:29 +03:00
  • 367cae679b ModelOutputType is now a Flag instead of regular enum. Combinations of the base flags are provided for all of the previous output types for convenience. All checks in the code now use the basic flags and not the complex types. abigailt 2024-07-03 13:29:37 +03:00
  • 2895b40f05 Addressing review comments abigailt 2024-06-19 11:17:17 +03:00
  • 846de0f753 Remove check of correct shape of predictions which becomes too complicated with the new output types supported. abigailt 2024-05-02 18:56:31 +03:00
  • a4816878f9 Formatting abigailt 2024-05-02 17:04:34 +03:00
  • 0f5a1bcaa0 Tests and support for additional model output types abigailt 2024-05-02 15:17:51 +03:00
  • a8ec87f922 Add tests for single label binary pytorch models abigailt 2024-05-02 14:46:08 +03:00
  • aa65f0f6f2 Add tests that check the transforemd is identical when no generalizations abigailt 2024-05-02 12:33:27 +03:00
  • b3f87623b1 Test for sklearn (currently not passing due to ART dependency) abigailt 2024-04-30 14:51:35 +03:00
  • 8b8b461143 Support for multi-label logits/probabilities abigailt 2024-03-17 11:49:05 +02:00
  • 7e34f0d2ff Support for multi-label binary models in minimizer. First test with pytorch model passing. abigailt 2024-03-13 09:59:02 +02:00
  • 076503b248 Working example of anonymization with pytorch multi-output binary model abigailt 2024-03-12 13:25:31 +02:00
  • 5e19d4ae27 New model output types + single implementation of score method that supports multiple output types. Existing tests pass. Still need more tests for new types. abigailt 2024-03-07 19:00:09 +02:00
  • 79534b69db Initial support+test for pytorch multi-label binary classifier abigailt 2024-02-19 14:16:03 +02:00
  • e00535d120
         Fix error with pandas dataframes (#92) abigailgold 2024-02-13 08:56:12 -05:00
  • 70c827154d Update documentation of classes to reflect that all data should be encoded and scaled. abigailt 2024-02-13 11:22:18 +02:00
  • 132812f772 Remove unused import abigailt 2024-02-13 11:06:27 +02:00
  • 90980de234 Fix error with pandas dataframes in _columns_different_distributions + add appropriate test abigailt 2024-02-13 10:48:56 +02:00
  • f197199e54 Initial version of general model wrappers and methods supporting multi-label classifiers abigailt 2024-02-12 09:45:36 +02:00
  • cb70ca10e6
         Fix calculation of score for pytorch models when a single column of probabilities is used for binary classification (#91) abigailgold 2024-01-21 06:56:57 -05:00
  • 1e3dd399d0 Fix calculation of score for pytorch models when a single column of probabilities is used for binary classification abigailt 2024-01-18 17:50:20 +02:00
  • a8f5326572
         Fix issue with computed ranges for one-hot encoded features (#90) abigailgold 2024-01-17 12:45:22 -05:00
  • 4282cf3208 Fix issue with computed ranges for one-hot encoded features abigailt 2024-01-17 15:55:06 +02:00
  • d8de062d43
         bump version (#88) v0.2.1 abigailgold 2024-01-01 06:55:00 -05:00
  • 38413a6f8e bump version abigailt 2024-01-01 13:24:56 +02:00
  • 6d81cd8ed4
         Support for one-hot encoded features in minimization (#87) abigailgold 2023-12-24 18:18:18 -05:00
  • e7a0a6ad85 Formatting abigailt 2023-12-24 17:26:09 -05:00
  • 686969eb86 Review comments abigailt 2023-12-24 17:17:59 -05:00
  • a3d294af2d Fix test abigailt 2023-12-24 17:11:38 -05:00
  • f646109e84 Replace values in multi-column 1-hot encoded features instead of appending so that options are narrowed down abigailt 2023-12-24 13:13:52 -05:00
  • 0e01e19e0c Updated notebooks for one-hot encoded data abigailt 2023-12-04 08:52:31 -05:00
  • f602806133 Indication when fitting failed abigailt 2023-12-04 08:52:15 -05:00
  • 8c1a186ef7 Formatting abigailt 2023-11-19 14:18:59 -05:00
  • c122fc7387 More tests and fixes. Make sure representative values in generalizations for 1-hot encoded features are consistent. abigailt 2023-11-19 11:02:21 -05:00
  • 904462a6a8 One more test + fixes abigailt 2023-11-15 08:54:27 -05:00
  • e7ee42fdc8 Second test (pandas) abigailt 2023-11-15 08:27:17 -05:00
  • 2a657388af Initial version with first working test abigailt 2023-11-15 08:21:40 -05:00
  • 5dce961092
         Support 1-hot encoded features in anonymization + fixes related to encoding in minimization (#86) abigailgold 2023-10-19 11:48:15 +03:00
  • ea8564bc4b Extract common code to methods abigailt 2023-10-18 12:22:26 +03:00
  • 570c6f8966 Fix anonymization adult notebook + new notebook to demonstrate anonymization on 1-hot encoded data abigailt 2023-10-17 11:25:03 +03:00
  • 88f4258493 Remove unused imports abigailt 2023-10-05 14:07:12 +03:00
  • 30cb705062 No default encoder, if none provided data is supplied to the model as is. Fix data type of representative values. Fix and add more tests. abigailt 2023-10-05 13:58:40 +03:00
  • ef406cea62 Remove reshaping abigailt 2023-10-05 13:57:04 +03:00
  • fe9eae45fc Improve tests abigailt 2023-09-24 13:44:02 +03:00
  • 5cffc7e347 Fix typing abigailt 2023-09-22 13:57:16 +03:00
  • 31cbc6ad5f Formatting abigailt 2023-09-22 11:24:37 +03:00
  • 364ebf68eb Add test for pandas abigailt 2023-09-21 19:10:05 +03:00
  • a814404534 Support 1-hot encoded features in anonymization (#72) abigailt 2023-09-21 19:00:27 +03:00
  • 26addd192f
         Support pytorch models in data minimization (#85) abigailgold 2023-09-21 17:48:15 +03:00
  • a40484e0c9
         Add column distribution comparison, and a third method for dataset asssessment by membership classification (#84) andersonm-ibm 2023-09-21 16:43:19 +03:00
  • 82ce575b97 Bring back comments Maya Anderson 2023-09-21 13:03:44 +03:00
  • eeab625c3e Formatting abigailt 2023-09-21 12:48:47 +03:00
  • a46c4cad9e Support pytorch models in data minimization abigailt 2023-09-20 20:40:27 +03:00
  • c852e2bd37 Fix imports Maya Anderson 2023-09-20 20:11:10 +03:00
  • 34de3ff93b Address review comments, add additional distribution comparison tests and make them externally configurable too, in addition to the alpha becoming configurable. Maya Anderson 2023-09-20 19:44:54 +03:00
  • 0ee0bf05d6 Flake code cleanups Maya Anderson 2023-09-20 09:23:22 +03:00
  • ad65f6f993 Add column distribution comparison, and a third method for dataset assessment by membership classification Maya Anderson 2023-09-19 15:52:34 +03:00
  • 13a0567183
         Make data minimization more consistent and performant (#83) abigailgold 2023-08-21 18:39:15 +03:00
  • 5e84f3fac4 Revert to having generalize_using_transform as an instance param (passed at init) and throwing an exception when used incorrectly. abigailt 2023-08-21 18:09:06 +03:00
  • 256dfbbc71 Remove unused code, renaming and additional review comments abigailt 2023-08-21 16:40:52 +03:00
  • 69e45d99e5 formatting abigailt 2023-08-10 13:20:28 +03:00
  • f85fc87bdd Externalize common test code to methods. Support for sparse matrix. abigailt 2023-08-10 12:57:59 +03:00
  • 3de93a87f1 Update comments, use lowercase variables, mode data in tests to fixtures abigailt 2023-08-08 12:34:41 +03:00
  • c2e0fced03 Compute generalizations with test data when possible (for computing better representatives). abigailt 2023-08-07 15:59:22 +03:00
  • b48b829a01 generalize_using_transform is now passed separately to the fit and calculate_ncp methods. ncp is now a class that contains 3 values: fit_score, transform_score and generalizations_score so that it doesn't matter in what order the different methods are called, all calculated ncp scores are stored. Generalizations can now be applied either from tree cells or from global generalizations struct depending on the value of generalize_using_transform. Representative values can also be computed from global generalizations. Removing a feature from the generalization can also be applied in either mode. abigailt 2023-08-07 15:31:49 +03:00
  • 7d22423ac3 Reduce runtime of dataset assessment tests abigailt 2023-05-31 14:54:16 +03:00
  • 7f0fef9ed5 Update incompatible scipy version abigailt 2023-05-31 14:53:54 +03:00
  • 7403a0e93a Formatting abigailt 2023-05-31 14:09:41 +03:00
  • 30725b33b7 Fix test abigailt 2023-05-31 14:07:52 +03:00
  • 87b9a8c2ea Removed redundant transforming of data abigailt 2023-05-29 21:34:25 +03:00
  • ec6fd56a6b Fix computing generalizations from transformed data + add some tests abigailt 2023-05-29 21:27:01 +03:00
  • 90f6b56ecc All tests pass, still need to review TODOs abigailt 2023-05-29 19:15:16 +03:00
  • 1279f1d7a5 Many fixes, some tests pass abigailt 2023-05-29 19:13:35 +03:00
  • 66f1ef08a7 generalize_using_transform=False supported abigailt 2023-05-18 10:32:54 +03:00
  • fdfeaf2f03 Initial commit. Tests not yet passing. abigailt 2023-05-16 13:03:01 +03:00
  • 49de29dc1f Update requirements abigailt 2023-02-23 10:59:51 +02:00
  • eead75406d Merge remote-tracking branch 'origin/ncp_fixes' into ncp_fixes ncp_fixes abigailt 2023-05-29 21:45:14 +03:00
  • 87c3e3a5d0 Removed redundant transforming of data abigailt 2023-05-29 21:34:25 +03:00
  • c8b8058006 Fix computing generalizations from transformed data + add some tests abigailt 2023-05-29 21:27:01 +03:00
  • 0f39834e1d All tests pass, still need to review TODOs abigailt 2023-05-29 19:15:16 +03:00
  • 2c01187f1f Many fixes, some tests pass abigailt 2023-05-29 19:13:35 +03:00
  • 4a252e65fe generalize_using_transform=False supported abigailt 2023-05-18 10:32:54 +03:00
  • 51340fa554 Initial commit. Tests not yet passing. abigailt 2023-05-16 13:03:01 +03:00
  • c3d2e9c7d0 Update requirements abigailt 2023-02-23 10:59:51 +02:00
  • f1995ea6f9 Removed redundant transforming of data abigailt 2023-05-29 21:34:25 +03:00
  • aa38a1d716 Fix computing generalizations from transformed data + add some tests abigailt 2023-05-29 21:27:01 +03:00
  • 26adcf3528 All tests pass, still need to review TODOs abigailt 2023-05-29 19:15:16 +03:00
  • cc4cba0d8e Many fixes, some tests pass abigailt 2023-05-29 19:13:35 +03:00
  • 4541ee60a2 generalize_using_transform=False supported abigailt 2023-05-18 10:32:54 +03:00
  • 710aae4083 Initial commit. Tests not yet passing. abigailt 2023-05-16 13:03:01 +03:00
  • e9a225501f
         Limit scikit-learn version because of API changes (#81) andersonm-ibm 2023-05-14 08:52:06 +03:00
  • 4d81ec9ab0 Change np.int to int according to DeprecationWarning Maya Anderson 2023-05-11 14:29:36 +03:00