privacy
privacy copied to clipboard
Library for training machine learning models with privacy for training data
Bump version for release.
Add DPEvent to the state in DifferentiallyPrivateFactory's next_fn. This requires implementing initial_sample_state in TreeRangeSumQuery for tests to pass.
Add functions to derive epsilon lower bounds.
Automated rollback of commit cff47686f6cd9b6967f95b6eb366f3a4bf6f192f
Internal test.
I receive the below error when running the privacy report (privacy_report.ipynb) from tutorials: ModuleNotFoundError Traceback (most recent call last) [](https://localhost:8080/#) in () ----> 1 from tensorflow_privacy.privacy.privacy_tests.membership_inference_attack import membership_inference_attack as mia...
Dear developers of the privacy framework. I was checking the implementation of [quantile_adaptive_clip_tree_query.py](https://github.com/tensorflow/privacy/blob/master/tensorflow_privacy/privacy/dp_query/quantile_adaptive_clip_tree_query.py) I can see that a new tree is required to inject noise after aggregating the norm bit...
Fixed by calling `compute_dp_sgd_privacy_statement`.
Internal Code Change
Problem: Fresh install of tensorflow_privacy doesn't work Steps to recreate: Create a new environment, run `pip install tensorflow-privacy` Now from within python, run `import tensorflow_privacy` I get this horrendous-looking error...