apunkt/ai-privacy-toolkit
1
0
Fork 0
mirror of https://github.com/IBM/ai-privacy-toolkit.git synced 2026-06-14 15:25:12 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions
163 commits 9 branches 4 tags 2.3 MiB
Commit graph

81 commits

Author SHA1 Message Date
abigailt
d58934a4b6 Remove unused import 
Signed-off-by: abigailt <abigailt@il.ibm.com>
 2024-07-03 13:58:17 +03:00
abigailt
bcb7c47cc6 Formatting 
Signed-off-by: abigailt <abigailt@il.ibm.com>
 2024-07-03 13:42:29 +03:00
abigailt
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. 
Signed-off-by: abigailt <abigailt@il.ibm.com>
 2024-07-03 13:29:37 +03:00
abigailt
2895b40f05 Addressing review comments 
Signed-off-by: abigailt <abigailt@il.ibm.com>
 2024-06-19 11:17:17 +03:00
abigailt
846de0f753 Remove check of correct shape of predictions which becomes too complicated with the new output types supported. 
Signed-off-by: abigailt <abigailt@il.ibm.com>
 2024-05-02 18:56:31 +03:00
abigailt
a4816878f9 Formatting 
Signed-off-by: abigailt <abigailt@il.ibm.com>
 2024-05-02 17:04:34 +03:00
abigailt
0f5a1bcaa0 Tests and support for additional model output types 
Signed-off-by: abigailt <abigailt@il.ibm.com>
 2024-05-02 15:17:51 +03:00
abigailt
a8ec87f922 Add tests for single label binary pytorch models 
Signed-off-by: abigailt <abigailt@il.ibm.com>
 2024-05-02 14:46:08 +03:00
abigailt
aa65f0f6f2 Add tests that check the transforemd is identical when no generalizations 
Signed-off-by: abigailt <abigailt@il.ibm.com>
 2024-05-02 12:33:27 +03:00
abigailt
b3f87623b1 Test for sklearn (currently not passing due to ART dependency) 
Signed-off-by: abigailt <abigailt@il.ibm.com>
 2024-04-30 14:51:35 +03:00
abigailt
8b8b461143 Support for multi-label logits/probabilities 
Signed-off-by: abigailt <abigailt@il.ibm.com>
 2024-03-17 11:49:05 +02:00
abigailt
7e34f0d2ff Support for multi-label binary models in minimizer. First test with pytorch model passing. 
Signed-off-by: abigailt <abigailt@il.ibm.com>
 2024-03-13 09:59:02 +02:00
abigailt
076503b248 Working example of anonymization with pytorch multi-output binary model 
Signed-off-by: abigailt <abigailt@il.ibm.com>
 2024-03-12 13:25:31 +02:00
abigailt
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. 
Signed-off-by: abigailt <abigailt@il.ibm.com>
 2024-03-07 19:00:09 +02:00
abigailt
79534b69db Initial support+test for pytorch multi-label binary classifier 
Signed-off-by: abigailt <abigailt@il.ibm.com>
 2024-02-19 14:16:03 +02:00
abigailt
f197199e54 Initial version of general model wrappers and methods supporting multi-label classifiers 
Signed-off-by: abigailt <abigailt@il.ibm.com>
 2024-02-12 09:45:36 +02:00
abigailgold
a8f5326572
Fix issue with computed ranges for one-hot encoded features (#90) 
Signed-off-by: abigailt <abigailt@il.ibm.com>
 2024-01-17 12:45:22 -05:00
abigailgold
6d81cd8ed4
Support for one-hot encoded features in minimization (#87) 
* Initial version with first working test
* Make sure representative values in generalizations for 1-hot encoded features are consistent.
* Updated notebooks for one-hot encoded data
* Review comments

Signed-off-by: abigailt <abigailt@il.ibm.com>
 2023-12-24 18:18:18 -05:00
abigailgold
5dce961092
Support 1-hot encoded features in anonymization + fixes related to encoding in minimization (#86) 
* Support 1-hot encoded features in anonymization (#72)
* Fix anonymization adult notebook + new notebook to demonstrate anonymization on 1-hot encoded data

* Minimizer: 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.

Signed-off-by: abigailt <abigailt@il.ibm.com>
 2023-10-19 11:48:15 +03:00
abigailgold
26addd192f
Support pytorch models in data minimization (#85) 
* Support pytorch models in data minimization

Signed-off-by: abigailt <abigailt@il.ibm.com>
 2023-09-21 17:48:15 +03:00
andersonm-ibm
a40484e0c9
Add column distribution comparison, and a third method for dataset asssessment by membership classification (#84) 
* Add column distribution comparison, and a third method for dataset assessment by membership classification

* Address review comments, add additional distribution comparison tests and make them externally configurable too, in addition to the alpha becoming configurable.

Signed-off-by: Maya Anderson <mayaa@il.ibm.com>
 2023-09-21 16:43:19 +03:00
abigailgold
13a0567183
Make data minimization more consistent and performant (#83) 
* Update requirements

* Update incompatible scipy version

* Reduce runtime of dataset assessment tests

* 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.

* Compute generalizations with test data when possible (for computing better representatives).

* Externalize common test code to methods.
 2023-08-21 18:39:15 +03:00
andersonm-ibm
e9a225501f
Limit scikit-learn version because of API changes (#81) 
* Limit scikit-learn versions between 0.22.2 and 1.1.3, remove deprecated load_boston().

* Set pytest configuration option to show test progress in detail.

* Change np.int to int according to DeprecationWarning

Signed-off-by: Maya Anderson <mayaa@il.ibm.com>
 2023-05-14 08:52:06 +03:00
andersonm-ibm
3885ab9d9d
Change back flake8 warnings to errors. Fix tests not to fail it. (#76) 
Signed-off-by: Maya Anderson <mayaa@il.ibm.com>
 2023-05-11 11:33:50 +03:00
abigailgold
8a9ef80146
Increase version to 0.2.0 (#74) 
* Remove tensorflow dependency if not using keras model
* Remove xgboost dependency if not using xgboost model
* Documentation updates

Signed-off-by: abigailt <abigailt@il.ibm.com>
 2023-05-08 12:50:55 +03:00
Maya Anderson
dbb958f791 Merge pull request #71 from IBM/dataset_assessment 
Add AI privacy Dataset assessment module with two attack implementations.

Signed-off-by: Maya Anderson <mayaa@il.ibm.com>
 2023-03-20 14:21:29 +02:00
abigailgold
d52fcd0041
Formatting (#68) 
Fix most flake/lint errors and ignore a few others

Signed-off-by: abigailt <abigailt@il.ibm.com>
 2022-12-25 15:13:57 +02:00
Maya Anderson
89bdcfc00e Prepare project for CI: cleanup dependencies, fix test data location, cleanup assert. 
Signed-off-by: Maya Anderson <mayaa@il.ibm.com>
 2022-12-20 16:00:36 +02:00
abigailt
64038f76f9 Merge with main 2022-08-01 18:12:03 +03:00
abigailt
dc5cc793ee Merge with main 2022-08-01 18:11:34 +03:00
abigailt
a9e2a35e18 Add support for xgboost XGBClassifier (#53) 2022-07-28 17:21:24 +03:00
olasaadi
74ce92acc4 fix 2022-07-26 18:37:44 +03:00
abigailt
a13415ad67 Externalize BlackboxClassifier dataset (x and predictions) 2022-07-25 16:31:45 +03:00
abigailt
fb534f7a0f BlackboxClassifier based on predictions to work with DatasetWithPredictions 2022-07-25 16:31:45 +03:00
abigailt
77a6e08c8e Keras regression support 2022-07-24 18:45:50 +03:00
Ron Shmelkin
c77e34e373
update pytorch wrapper to use torch loaders 
fix tests
and dataset style
 2022-07-24 14:31:47 +03:00
olasaadi
6f69f5557b fix bug 2022-07-20 18:29:48 +03:00
olasaadi
3bf26b67d2 fix 2022-07-20 17:36:00 +03:00
abigailt
a7d156660e Wrap predict method in BlackBoxClassifierPredictMethod to avoid exception in ART when supplied method returns scalars 2022-07-20 13:33:19 +03:00
abigailt
1cc73b3da1 Check for mismatch between model output type and actual output 2022-07-20 13:33:19 +03:00
abigailt
bc7ab0cc7f Add model type to blackbox classifier (#49) 2022-07-20 13:33:19 +03:00
olasaadi
4973fbebc6 fix 2022-07-19 21:16:39 +03:00
abigailgold
00f9c16863
Support additional use cases for data (#46) 
* Make ART black box classifier not apply preprocessing to data
* Add option to store predictions (in addition to x,y) in Dataset and Data classes
 2022-07-11 14:28:09 +03:00
Shlomit Shachor
e25e58b253
enhance calculation of nb classes + tests (#45) 
* update get_nb_classes method to handle 1-hot and scalar input
 2022-07-05 11:32:17 +03:00
abigailgold
c6eb553a9f
Blackbox predict method (#43) 
* Support output probabilities
* Support black box classifier with predict method
* Update requirements (security alert #1)
 2022-06-30 18:23:53 +03:00
Shlomit Shachor
1c4b963add
Wrappers no train (#40) 
1) Handle train None in Data
2) Update BB Classifier to handle None either for train or test (x or y)
 2022-06-26 14:43:22 +03:00
olasaadi
21cba95a28 fix 2022-06-06 14:32:34 +03:00
olasaadi
c954f53ad7 fix 2022-06-06 14:02:40 +03:00
olasaadi
302d0c4b8c update 2022-06-02 15:25:07 +03:00
olasaadi
a3fb68fb56 update 2022-05-30 12:52:32 +03:00