commit | 8c155d51fab099ee7bf64f4bdbfeda82881925a5 | [log] [tgz] |
---|---|---|
author | Sameer Agarwal <sameeragarwal@google.com> | Fri Nov 08 08:04:44 2013 -0800 |
committer | Sameer Agarwal <sameeragarwal@google.com> | Fri Nov 08 22:18:14 2013 -0800 |
tree | ed1ecafa299dd0e148e1ad6e6ef950a55238bac6 | |
parent | 58792dc8ee0e4b56331f33f753f1b1932c5c2960 [diff] |
Speed up the application of robust loss functions. Since we added special handling for the case for rho[2] < 0, the bulk of CorrectJacobian is pointless in the common case. So add a simple one dimensional loop which rescales the Jacobian. This speeds up this method immensely. The robustification of a Jacobian gets speeded up by > 50%. Change-Id: I97c4e897ccbb5521c053e1fb931c5d0d32f542c7