Markus Moll 47d26bcd3b Do not implicitly negate the step in the TrustRegionMinimizer.
In the TrustRegionMinimizer, the step is currently implicitly negated.
This is done so that the linearized residual is |r - J*step|^2, which
corresponds to J*step = r, so neither J nor r have to be modified.
However, it leads to the rather unintuitive situation that the strategy
returns a step in positive gradient direction, which you would expect to
increase the function value. One way is to rename the "step" parameter in
the strategy to "negative_step" and document it.
This patch instead moves the negation inside the strategy, just around
the linear solver call, so that it is done in a local context and easier
to document.

Change-Id: Idb258149a01f61c64e22128ea221c5a30cd89c89
2012-08-17 13:19:14 -07:00
2012-08-02 18:35:16 -07:00
2012-04-30 23:09:08 -07:00
2012-07-19 12:11:32 -07:00
2012-08-16 16:31:23 -07:00
2012-08-15 19:05:20 -07:00
2012-04-30 23:09:08 -07:00
2012-08-14 14:47:13 -07:00

Ceres Solver - A non-linear least squares minimizer
==================================================
Please see ceres-solver.pdf in docs/ for a tutorial and reference.
S
Description
No description provided
Readme Multiple Licenses 42 MiB
Languages
C++ 94.4%
CMake 3.8%
Python 0.8%
C 0.5%
Starlark 0.4%