LinearOperator::FooMultiply -> LinearOperator::FooMultiplyAndAccumulate
These methods were historically poorly named and every time I read code
I get confused whether they are just multiplying or multiplying and
adding. Clarifying them also gives us the changce to introduce
RightMultiply and LeftMultiply methods in the base class which will
simplify a number call sites in a subsequent CL.
Fixes https://github.com/ceres-solver/ceres-solver/issues/855
Change-Id: Ice4fb483f1acd02527a6dd753ef0c5a66037f4b0
diff --git a/internal/ceres/low_rank_inverse_hessian.h b/internal/ceres/low_rank_inverse_hessian.h
index de30f54..878db81 100644
--- a/internal/ceres/low_rank_inverse_hessian.h
+++ b/internal/ceres/low_rank_inverse_hessian.h
@@ -64,7 +64,7 @@
// num_parameters is the row/column size of the Hessian.
// max_num_corrections is the rank of the Hessian approximation.
// use_approximate_eigenvalue_scaling controls whether the initial
- // inverse Hessian used during Right/LeftMultiply() is scaled by
+ // inverse Hessian used during Right/LeftMultiplyAndAccumulate() is scaled by
// the approximate eigenvalue of the true inverse Hessian at the
// current operating point.
// The approximation uses:
@@ -83,9 +83,9 @@
bool Update(const Vector& delta_x, const Vector& delta_gradient);
// LinearOperator interface
- void RightMultiply(const double* x, double* y) const final;
- void LeftMultiply(const double* x, double* y) const final {
- RightMultiply(x, y);
+ void RightMultiplyAndAccumulate(const double* x, double* y) const final;
+ void LeftMultiplyAndAccumulate(const double* x, double* y) const final {
+ RightMultiplyAndAccumulate(x, y);
}
int num_rows() const final { return num_parameters_; }
int num_cols() const final { return num_parameters_; }
