Marcel Keller
Marcel Keller
I'm afraid there are still a few issues: 1. decision_tree_optimized still exists, did you mean to delete this? 2. The changes to decision_tree break all applications thereof, adult, easy_adult, bench-dt,...
I'm afraid `breast_tree` still breaks in the sense that the accuracy is considerably less: ``` $ Scripts/compile-run.py ring breast_tree (...) train for height 1: 159/426 (159/159, 0/267) test for height...
I'm not sure that's the whole explanation. Even a "bad" index choice can be corrected at a further level, so I would still expect the training set accuracy to approach...
Thank you for your latest commit. However, there are still issues regarding accuracy. When I run `Scripts/compile-emulate.py easy_adult` on the master branch, I get: ``` train for height 1: 27799/36631...
Thank you for raising this. You should find that 7e576fab0d5274961ca64d0c4dd8c6082b572565 fixes it.
You should find that 383c18f1cb fixes that.
I don't think MP-SPDZ offers much for the share conversion you're looking for. Shamir secret sharing is implemented in `Protocols/ShamirInput.hpp` where `init` sets up reconstruction factors (a matrix such that...
If by servers you mean the MP-SPDZ computing parties, they cannot switch the protocol/sharing scheme, and you would need to run another set for another secret sharing scheme. Furthermore, an...
Which file do you mean? The client interface works without files.
I cannot reproduce this. What is the output if you add the following line? ``` print_ln("%s %s %s", a.v.reveal(), b.v.reveal(), result.v.reveal()) ```