[Bf-blender-cvs] [c1004b1] master: Libmv: Update Ceres to latest upstream
Sergey Sharybin
noreply at git.blender.org
Thu Oct 9 11:40:46 CEST 2014
Commit: c1004b1b77ee6de90afedbccf4ce6f54c056ec5b
Author: Sergey Sharybin
Date: Thu Oct 9 11:29:20 2014 +0200
Branches: master
https://developer.blender.org/rBc1004b1b77ee6de90afedbccf4ce6f54c056ec5b
Libmv: Update Ceres to latest upstream
just to be sure we're dealing with the latest git version when troubleshooting
one annoying OSX issue..
===================================================================
M extern/libmv/third_party/ceres/ChangeLog
M extern/libmv/third_party/ceres/bundle.sh
M extern/libmv/third_party/ceres/include/ceres/gradient_problem_solver.h
M extern/libmv/third_party/ceres/internal/ceres/coordinate_descent_minimizer.cc
===================================================================
diff --git a/extern/libmv/third_party/ceres/ChangeLog b/extern/libmv/third_party/ceres/ChangeLog
index 6f92154..b4f78ee 100644
--- a/extern/libmv/third_party/ceres/ChangeLog
+++ b/extern/libmv/third_party/ceres/ChangeLog
@@ -1,3 +1,78 @@
+commit 0435246de5f45e69b2c97d244ed61bedd340215a
+Author: Sameer Agarwal <sameeragarwal at google.com>
+Date: Wed Oct 8 18:12:53 2014 -0700
+
+ Add seene to users.rst
+
+ Change-Id: If40726775a3d4b234b6e10517fe9943d122a3384
+
+commit fdf32b315f39553639f0becf078ad4eec763a10e
+Author: Sameer Agarwal <sameeragarwal at google.com>
+Date: Wed Oct 8 16:04:32 2014 -0700
+
+ Fix some errant tabs.
+
+ Change-Id: Iaf1906eaade49467ba282656cf0a10879d258b1f
+
+commit 6768b3586a027bb850c0a50e2a27380f5d80142a
+Author: Sameer Agarwal <sameeragarwal at google.com>
+Date: Wed Oct 8 12:48:16 2014 -0700
+
+ Minor cleanups in preparation for a release.
+
+ 1. Fix the release script to ignore the version.h checking.
+ 2. Fix some ceres documentation formatting errors.
+
+ Change-Id: I3fd6b85e771b242f463d6a36c3efd8d691f9242f
+
+commit 7b6bd1cd31aa0b8cb7fb97600c1b9999846e3152
+Author: Sameer Agarwal <sameeragarwal at google.com>
+Date: Thu Oct 2 16:16:26 2014 -0700
+
+ Documentation update.
+
+ 1. Complete restructuring of the documentation to account for
+ GradientProblemSolver.
+ 2. Update the version history to account for changes since 1.9.0.
+ 3. Add links and document the various examples that ship with ceres.
+ 4. Documentation for GradientProblem GradientProblemSolver.
+
+ Change-Id: If3a18f2850cbc98be1bc34435e9ea468785b8b27
+
+commit b7d321f505e936b6c09aeb43ae3f7b1252388a95
+Author: Sameer Agarwal <sameeragarwal at google.com>
+Date: Fri Oct 3 15:47:59 2014 -0700
+
+ Relax the warning/error handing in GCC.
+
+ Thanks to Matthew Woehlke for suggesting this.
+
+ Change-Id: Iae754465c086b0841a7816df1a36781371d0dc9a
+
+commit 94c6e7d27b5d48d81ab54ed9cdcbc55c3c099311
+Author: Sameer Agarwal <sameeragarwal at google.com>
+Date: Wed Oct 1 15:55:13 2014 -0700
+
+ Improve multithreading when using inner iterations.
+
+ Inner iterations by default would use problems where the evaluator
+ was configured to use exactly one thread for doing the evaluation.
+ This is fine when there are multiple inner iteration problems
+ being executed concurrently, but every now and then there are
+ problem decompositions where there is just one parameter block
+ in the current independent set and it touches every single
+ residual block. In such cases it is essential that the evaluator
+ be configured to use multiple threads.
+
+ We now pay attention to the size of the independent set and
+ dynamically configure the number of threads being used by the
+ outer loop and the evaluator loop.
+
+ Thanks to William Rucklidge for reporting this issue and providing
+ a test problem to debug.
+
+ Change-Id: Iaff9a4ab6d2658cf7b61ea213575d23aab604e3b
+
commit 9e11cd16d09403b9270e621e839d5948b6a74b8d
Author: Sameer Agarwal <sameeragarwal at google.com>
Date: Mon Sep 29 14:27:58 2014 -0700
@@ -596,76 +671,3 @@ Date: Thu Jun 5 21:30:13 2014 -0700
solve.
Change-Id: I80f35cfc9f2cbf78f1df4aceace27075779d8a3a
-
-commit bd90384226a7f8629467f72fc410a9e8086a2dff
-Author: Sameer Agarwal <sameeragarwal at google.com>
-Date: Mon Aug 18 11:27:06 2014 -0700
-
- Lint comments from William Rucklidge.
-
- Also some minor refactoring of the trust_region_preprocessor_test.cc
-
- Change-Id: Ica28002254c95722faf93a7ef35bf3deab557f0b
-
-commit 3150321db4a0cb1bb4894961a030d95dacae3591
-Author: Sameer Agarwal <sameeragarwal at google.com>
-Date: Tue Aug 12 22:46:51 2014 -0700
-
- Preprocessor for the LineSearchMinimizer.
-
- Change-Id: Ieb5dfe1c0b96ef323c1130edd0c3a8a8b2c644cc
-
-commit f7da411ef0d0067e269629887d64cdb769368800
-Author: Sameer Agarwal <sameeragarwal at google.com>
-Date: Thu Aug 7 14:30:33 2014 -0700
-
- Preprocessor for the TrustRegionMinimizer.
-
- 1. Base class for preprocessors.
- 2. A preprocessor for problems that will be solved using
- the trust region minimizer.
- 3. Added sanity tests to the program reordering options
- for Schur type linear solvers.
- 4. Tests for the TrustRegionPreprocessor.
-
- Change-Id: I88cd926f0053bbbf2bd6b11e03ec55b8bf473cf1
-
-commit 54893ba523106e38ab06eb72fb5d8748685c7797
-Author: Alex Stewart <alexs.mac at gmail.com>
-Date: Mon Aug 11 19:04:18 2014 +0100
-
- Add missing #include of <limits> for loss functions.
-
- Change-Id: Id632451429e03031a1533a9be795270debc70706
-
-commit 4a2a888905fd1ce7203e45df15762d52740bb240
-Author: Sameer Agarwal <sameeragarwal at google.com>
-Date: Thu Aug 7 11:48:03 2014 -0700
-
- Change ownership of pointers in Minimizer::Options.
-
- This is a intermediate change to clean things up
- in preparation for a broader refactoring of the SolverImpl.
-
- Essentially we are replacing raw pointers in Minimizer::Options
- with shared_ptr objects. For now this only makes things a bit
- more complicated looking inside solver_impl.cc, but going
- forward this will lead to considerable simplifications in
- tracking ownership of various pointers.
-
- Change-Id: I21db8fc6763c29b0d15e834d7c968a0f514042a0
-
-commit 0d4e3bd664d442b700fee2895c7a8ac37717dc08
-Author: Sameer Agarwal <sameeragarwal at google.com>
-Date: Thu Aug 7 12:19:10 2014 -0700
-
- GradientCheckingProblem's parameter blocks are initialized correctly.
-
- Ensure that when a new problem object is constructed for validing
- gradients, the parameter blocks have their data pointers point to
- the user's parameter blocks.
-
- We used to do this inside solver_impl.cc, but doing this at
- construction is the right thing to do.
-
- Change-Id: I3bfdc89bb0027c8d67cde937e8f2fa385d89c30c
diff --git a/extern/libmv/third_party/ceres/bundle.sh b/extern/libmv/third_party/ceres/bundle.sh
index 51a463a..14ce2c5 100755
--- a/extern/libmv/third_party/ceres/bundle.sh
+++ b/extern/libmv/third_party/ceres/bundle.sh
@@ -143,7 +143,7 @@ if(WITH_LIBMV_SCHUR_SPECIALIZATIONS)
${generated_sources}
)
else()
- add_definitions-DCERES_RESTRICT_SCHUR_SPECIALIZATION)
+ add_definitions(-DCERES_RESTRICT_SCHUR_SPECIALIZATION)
endif()
if(WIN32)
diff --git a/extern/libmv/third_party/ceres/include/ceres/gradient_problem_solver.h b/extern/libmv/third_party/ceres/include/ceres/gradient_problem_solver.h
index 484d88e..db706f7 100644
--- a/extern/libmv/third_party/ceres/include/ceres/gradient_problem_solver.h
+++ b/extern/libmv/third_party/ceres/include/ceres/gradient_problem_solver.h
@@ -72,7 +72,6 @@ class CERES_EXPORT GradientProblemSolver {
max_line_search_step_expansion = 10.0;
max_num_iterations = 50;
max_solver_time_in_seconds = 1e9;
- num_threads = 1;
function_tolerance = 1e-6;
gradient_tolerance = 1e-10;
logging_type = PER_MINIMIZER_ITERATION;
@@ -224,10 +223,6 @@ class CERES_EXPORT GradientProblemSolver {
// Maximum time for which the minimizer should run for.
double max_solver_time_in_seconds;
- // Number of threads used by Ceres for evaluating the cost and
- // jacobians.
- int num_threads;
-
// Minimizer terminates when
//
// (new_cost - old_cost) < function_tolerance * old_cost;
@@ -251,13 +246,6 @@ class CERES_EXPORT GradientProblemSolver {
// is sent to STDOUT.
bool minimizer_progress_to_stdout;
- // If true, the user's parameter blocks are updated at the end of
- // every Minimizer iteration, otherwise they are updated when the
- // Minimizer terminates. This is useful if, for example, the user
- // wishes to visualize the state of the optimization every
- // iteration.
- bool update_state_every_iteration;
-
// Callbacks that are executed at the end of each iteration of the
// Minimizer. An iteration may terminate midway, either due to
// numerical failures or because one of the convergence tests has
@@ -309,10 +297,10 @@ class CERES_EXPORT GradientProblemSolver {
// Sum total of all time spent inside Ceres when Solve is called.
double total_time_in_seconds;
- // Time (in seconds) spent evaluating the residual vector.
+ // Time (in seconds) spent evaluating the cost.
double cost_evaluation_time_in_seconds;
- // Time (in seconds) spent evaluating the jacobian matrix.
+ // Time (in seconds) spent evaluating the gradient.
double gradient_evaluation_time_in_seconds;
// Number of parameters in the probem.
diff --git a/extern/libmv/third_party/ceres/internal/ceres/coordinate_descent_minimizer.cc b/extern/libmv/third_party/ceres/internal/ceres/coordinate_descent_minimizer.cc
index 1d55458..0d8adee 100644
--- a/extern/libmv/third_party/ceres/internal/ceres/coordinate_descent_minimizer.cc
+++ b/extern/libmv/third_party/ceres/internal/ceres/coordinate_descent_minimizer.cc
@@ -1,5 +1,5 @@
// Ceres Solver - A fast non-linear least squares minimizer
-// Copyright 2012 Google Inc. All rights reserved.
+// Copyright 2014 Google Inc. All rights reserved.
// http://code.google.com/p/ceres-solver/
//
// Redistribution and use in source and binary forms, with or without
@@ -140,15 +140,24 @@ void CoordinateDescentMinimizer::Minimize(
}
for (int i = 0; i < independent_set_offsets_.size() - 1; ++i) {
- // No point paying the price for an OpemMP call if the set if of
+ const int num_problems =
+ independent_set_offsets_[i + 1] - independent_set_offsets_[i];
+ // No point paying the price for an OpemMP call if the set is of
// size zero.
- if (independent_set_offsets_[i] == independent_set_offsets_[i + 1]) {
+ if (num_problems == 0) {
continue;
}
+#ifdef CERES_USE_OPENMP
+ const int num_inner_iteration_threads =
+ min(options.num_threads, num_problems);
+ evaluator_options_.num_threads =
+ max(1, options.num_threads / num_inner_iteration_threads);
+
// The parameter blocks in each independent set can be optimized
// in parallel, since they do not
@@ Diff output truncated at 10240 characters. @@
More information about the Bf-blender-cvs
mailing list