It seems there is a problem with one of the unit tests that explicitly rely on timeouts. Removing this test to avoid these issues in builds.
Signed-off-by: Patrick Bloebaum <bloebp@amazon.com>
- Simplify the sampling process of permutations
- Using Halton sequence now instead of Sobol, since Halton generalizes beyond a basis of 2
This should also fix the current build error.
Signed-off-by: Patrick Bloebaum <bloebp@amazon.com>
- Shapley config parameter now clearly distinguishes between num_subset_sampling and num_permutations
- Using a quasi-random sequence generator now to improve uniform sampling of permutation. This should improve convergence (i.e., more accurate estimations)
- Fix bug in early stopping where one can get negative percentages and, thus, the estimator would falsely stop too early.
- Early stopping criteria is now applied to the change of each Shapley value individually instead of the average change. This should ensure less variance in the results between two runs and more accurate results.
- Change minimum change in percentage threshold to 0.05 by default for early stopping. This balances out the potential slower runtime from the change above.
Signed-off-by: Patrick Bloebaum <bloebp@amazon.com>
* update workflow to include formatting
Signed-off-by: Chris Trevino <chtrevin@microsoft.com>
* execute formatter
Signed-off-by: Chris Trevino <chtrevin@microsoft.com>
Signed-off-by: Chris Trevino <chtrevin@microsoft.com>
Patrick Bloebaum
Chris Trevino
Peter Goetz