GaussianFactorGraph.cpp
Go to the documentation of this file.
1 /* ----------------------------------------------------------------------------
2 
3  * GTSAM Copyright 2010, Georgia Tech Research Corporation,
4  * Atlanta, Georgia 30332-0415
5  * All Rights Reserved
6  * Authors: Frank Dellaert, et al. (see THANKS for the full author list)
7 
8  * See LICENSE for the license information
9 
10  * -------------------------------------------------------------------------- */
11 
28 #include <gtsam/base/debug.h>
29 #include <gtsam/base/timing.h>
30 #include <gtsam/base/cholesky.h>
31 
32 using namespace std;
33 using namespace gtsam;
34 
35 namespace gtsam {
36 
37  // Instantiate base classes
38  template class FactorGraph<GaussianFactor>;
40 
41  /* ************************************************************************* */
42  bool GaussianFactorGraph::equals(const This& fg, double tol) const
43  {
44  return Base::equals(fg, tol);
45  }
46 
47  /* ************************************************************************* */
49  KeySet keys;
50  for (const sharedFactor& factor: *this)
51  if (factor)
52  keys.insert(factor->begin(), factor->end());
53  return keys;
54  }
55 
56  /* ************************************************************************* */
57  std::map<Key, size_t> GaussianFactorGraph::getKeyDimMap() const {
58  map<Key, size_t> spec;
59  for (const GaussianFactor::shared_ptr& gf : *this) {
60  for (GaussianFactor::const_iterator it = gf->begin(); it != gf->end(); it++) {
61  map<Key,size_t>::iterator it2 = spec.find(*it);
62  if ( it2 == spec.end() ) {
63  spec.emplace(*it, gf->getDim(it));
64  }
65  }
66  }
67  return spec;
68  }
69 
70  /* ************************************************************************* */
71  double GaussianFactorGraph::error(const VectorValues& x) const {
72  double total_error = 0.;
73  for(const sharedFactor& factor: *this){
74  if(factor)
75  total_error += factor->error(x);
76  }
77  return total_error;
78  }
79 
80  /* ************************************************************************* */
81  double GaussianFactorGraph::probPrime(const VectorValues& c) const {
82  // NOTE the 0.5 constant is handled by the factor error.
83  return exp(-error(c));
84  }
85 
86  /* ************************************************************************* */
87  GaussianFactorGraph::shared_ptr GaussianFactorGraph::cloneToPtr() const {
89  *result = *this;
90  return result;
91  }
92 
93  /* ************************************************************************* */
94  GaussianFactorGraph GaussianFactorGraph::clone() const {
96  for (const sharedFactor& f : *this) {
97  if (f)
98  result.push_back(f->clone());
99  else
100  result.push_back(sharedFactor()); // Passes on null factors so indices remain valid
101  }
102  return result;
103  }
104 
105  /* ************************************************************************* */
108  for (const sharedFactor& factor: *this) {
109  if (factor)
110  result.push_back(factor->negate());
111  else
112  result.push_back(sharedFactor()); // Passes on null factors so indices remain valid
113  }
114  return result;
115  }
116 
117  /* ************************************************************************* */
118  using SparseTriplets = std::vector<std::tuple<int, int, double> >;
119  SparseTriplets GaussianFactorGraph::sparseJacobian(const Ordering& ordering,
120  size_t& nrows,
121  size_t& ncols) const {
122  gttic_(GaussianFactorGraph_sparseJacobian);
123  // First find dimensions of each variable
124  typedef std::map<Key, size_t> KeySizeMap;
125  KeySizeMap dims;
126  for (const auto& factor : *this) {
127  if (!static_cast<bool>(factor)) continue;
128 
129  for (auto it = factor->begin(); it != factor->end(); ++it) {
130  dims[*it] = factor->getDim(it);
131  }
132  }
133 
134  // Compute first scalar column of each variable
135  ncols = 0;
136  KeySizeMap columnIndices = dims;
137  for (const auto key : ordering) {
138  columnIndices[key] = ncols;
139  ncols += dims[key];
140  }
141 
142  // Iterate over all factors, adding sparse scalar entries
143  SparseTriplets entries;
144  entries.reserve(30 * size());
145  nrows = 0;
146  for (const auto& factor : *this) {
147  if (!static_cast<bool>(factor)) continue;
148 
149  // Convert to JacobianFactor if necessary
150  JacobianFactor::shared_ptr jacobianFactor(
151  std::dynamic_pointer_cast<JacobianFactor>(factor));
152  if (!jacobianFactor) {
154  std::dynamic_pointer_cast<HessianFactor>(factor));
155  if (hessian)
156  jacobianFactor.reset(new JacobianFactor(*hessian));
157  else
158  throw std::invalid_argument(
159  "GaussianFactorGraph contains a factor that is neither a "
160  "JacobianFactor nor a HessianFactor.");
161  }
162 
163  // Whiten the factor and add entries for it
164  // iterate over all variables in the factor
165  const JacobianFactor whitened(jacobianFactor->whiten());
166  for (auto key = whitened.begin(); key < whitened.end(); ++key) {
167  JacobianFactor::constABlock whitenedA = whitened.getA(key);
168  // find first column index for this key
169  size_t column_start = columnIndices[*key];
170  for (size_t i = 0; i < (size_t)whitenedA.rows(); i++)
171  for (size_t j = 0; j < (size_t)whitenedA.cols(); j++) {
172  double s = whitenedA(i, j);
173  if (std::abs(s) > 1e-12)
174  entries.emplace_back(nrows + i, column_start + j, s);
175  }
176  }
177 
178  JacobianFactor::constBVector whitenedb(whitened.getb());
179  for (size_t i = 0; i < (size_t)whitenedb.size(); i++) {
180  double s = whitenedb(i);
181  if (std::abs(s) > 1e-12) entries.emplace_back(nrows + i, ncols, s);
182  }
183 
184  // Increment row index
185  nrows += jacobianFactor->rows();
186  }
187 
188  ncols++; // +1 for b-column
189  return entries;
190  }
191 
192  /* ************************************************************************* */
193  SparseTriplets GaussianFactorGraph::sparseJacobian() const {
194  size_t nrows, ncols;
195  return sparseJacobian(Ordering(this->keys()), nrows, ncols);
196  }
197 
198  /* ************************************************************************* */
199  Matrix GaussianFactorGraph::sparseJacobian_() const {
200  gttic_(GaussianFactorGraph_sparseJacobian_matrix);
201  // call sparseJacobian
202  auto result = sparseJacobian();
203 
204  // translate to base 1 matrix
205  size_t nzmax = result.size();
206  Matrix IJS(3, nzmax);
207  for (size_t k = 0; k < result.size(); k++) {
208  const auto& entry = result[k];
209  IJS(0, k) = double(std::get<0>(entry) + 1);
210  IJS(1, k) = double(std::get<1>(entry) + 1);
211  IJS(2, k) = std::get<2>(entry);
212  }
213  return IJS;
214  }
215 
216  /* ************************************************************************* */
217  Matrix GaussianFactorGraph::augmentedJacobian(
218  const Ordering& ordering) const {
219  // combine all factors
220  JacobianFactor combined(*this, ordering);
221  return combined.augmentedJacobian();
222  }
223 
224  /* ************************************************************************* */
225  Matrix GaussianFactorGraph::augmentedJacobian() const {
226  // combine all factors
227  JacobianFactor combined(*this);
228  return combined.augmentedJacobian();
229  }
230 
231  /* ************************************************************************* */
232  pair<Matrix, Vector> GaussianFactorGraph::jacobian(
233  const Ordering& ordering) const {
234  Matrix augmented = augmentedJacobian(ordering);
235  return make_pair(augmented.leftCols(augmented.cols() - 1),
236  augmented.col(augmented.cols() - 1));
237  }
238 
239  /* ************************************************************************* */
240  pair<Matrix, Vector> GaussianFactorGraph::jacobian() const {
241  Matrix augmented = augmentedJacobian();
242  return make_pair(augmented.leftCols(augmented.cols() - 1),
243  augmented.col(augmented.cols() - 1));
244  }
245 
246  /* ************************************************************************* */
247  Matrix GaussianFactorGraph::augmentedHessian(
248  const Ordering& ordering) const {
249  // combine all factors and get upper-triangular part of Hessian
250  Scatter scatter(*this, ordering);
251  HessianFactor combined(*this, scatter);
252  return combined.info().selfadjointView();
253  }
254 
255  /* ************************************************************************* */
256  Matrix GaussianFactorGraph::augmentedHessian() const {
257  // combine all factors and get upper-triangular part of Hessian
258  Scatter scatter(*this);
259  HessianFactor combined(*this, scatter);
260  return combined.info().selfadjointView();
261  }
262 
263  /* ************************************************************************* */
264  pair<Matrix, Vector> GaussianFactorGraph::hessian(
265  const Ordering& ordering) const {
266  Matrix augmented = augmentedHessian(ordering);
267  size_t n = augmented.rows() - 1;
268  return make_pair(augmented.topLeftCorner(n, n), augmented.topRightCorner(n, 1));
269  }
270 
271  /* ************************************************************************* */
272  pair<Matrix, Vector> GaussianFactorGraph::hessian() const {
273  Matrix augmented = augmentedHessian();
274  size_t n = augmented.rows() - 1;
275  return make_pair(augmented.topLeftCorner(n, n), augmented.topRightCorner(n, 1));
276  }
277 
278  /* ************************************************************************* */
279  VectorValues GaussianFactorGraph::hessianDiagonal() const {
280  VectorValues d;
281  for (const sharedFactor& factor : *this) {
282  if(factor){
283  factor->hessianDiagonalAdd(d);
284  }
285  }
286  return d;
287  }
288 
289  /* ************************************************************************* */
290  map<Key,Matrix> GaussianFactorGraph::hessianBlockDiagonal() const {
291  map<Key,Matrix> blocks;
292  for (const sharedFactor& factor : *this) {
293  if (!factor) continue;
294  map<Key,Matrix> BD = factor->hessianBlockDiagonal();
295  map<Key,Matrix>::const_iterator it = BD.begin();
296  for (;it!=BD.end();++it) {
297  Key j = it->first; // variable key for this block
298  const Matrix& Bj = it->second;
299  if (blocks.count(j))
300  blocks[j] += Bj;
301  else
302  blocks.emplace(j,Bj);
303  }
304  }
305  return blocks;
306  }
307 
308  /* ************************************************************************ */
309  VectorValues GaussianFactorGraph::optimize(const Eliminate& function) const {
310  gttic(GaussianFactorGraph_optimize);
311  return BaseEliminateable::eliminateMultifrontal(Ordering::COLAMD, function)
312  ->optimize();
313  }
314 
315  /* ************************************************************************* */
316  VectorValues GaussianFactorGraph::optimize(const Ordering& ordering, const Eliminate& function) const {
317  gttic(GaussianFactorGraph_optimize);
318  return BaseEliminateable::eliminateMultifrontal(ordering, function)->optimize();
319  }
320 
321  /* ************************************************************************* */
322  // TODO(frank): can we cache memory across invocations
323  VectorValues GaussianFactorGraph::optimizeDensely() const {
324  gttic(GaussianFactorGraph_optimizeDensely);
325 
326  // Combine all factors in a single HessianFactor (as done in augmentedHessian)
327  Scatter scatter(*this);
328  HessianFactor combined(*this, scatter);
329 
330  // TODO(frank): cast to large dynamic matrix :-(
331  // NOTE(frank): info only valid (I think) in upper triangle. No problem for LLT...
332  Matrix augmented = combined.info().selfadjointView();
333 
334  // Do Cholesky Factorization
335  size_t n = augmented.rows() - 1;
336  auto AtA = augmented.topLeftCorner(n, n);
337  auto eta = augmented.topRightCorner(n, 1);
339 
340  // Solve and convert, re-using scatter data structure
341  Vector solution = llt.solve(eta);
342  return VectorValues(solution, scatter);
343  }
344 
345  /* ************************************************************************* */
346  namespace {
347  JacobianFactor::shared_ptr convertToJacobianFactorPtr(const GaussianFactor::shared_ptr &gf) {
348  JacobianFactor::shared_ptr result = std::dynamic_pointer_cast<JacobianFactor>(gf);
349  if( !result )
350  // Convert any non-Jacobian factors to Jacobians (e.g. Hessian -> Jacobian with Cholesky)
351  result = std::make_shared<JacobianFactor>(*gf);
352  return result;
353  }
354  }
355 
356  /* ************************************************************************* */
357  VectorValues GaussianFactorGraph::gradient(const VectorValues& x0) const
358  {
359  VectorValues g = VectorValues::Zero(x0);
360  for (const sharedFactor& factor: *this) {
361  JacobianFactor::shared_ptr Ai = convertToJacobianFactorPtr(factor);
362  Vector e = Ai->error_vector(x0);
363  Ai->transposeMultiplyAdd(1.0, e, g);
364  }
365  return g;
366  }
367 
368  /* ************************************************************************* */
369  VectorValues GaussianFactorGraph::gradientAtZero() const {
370  // Zero-out the gradient
371  VectorValues g;
372  for (const sharedFactor& factor: *this) {
373  if (!factor) continue;
374  VectorValues gi = factor->gradientAtZero();
375  g.addInPlace_(gi);
376  }
377  return g;
378  }
379 
380  /* ************************************************************************* */
381  VectorValues GaussianFactorGraph::optimizeGradientSearch() const
382  {
383  gttic(GaussianFactorGraph_optimizeGradientSearch);
384 
385  gttic(Compute_Gradient);
386  // Compute gradient (call gradientAtZero function, which is defined for various linear systems)
387  VectorValues grad = gradientAtZero();
388  double gradientSqNorm = grad.dot(grad);
389  gttoc(Compute_Gradient);
390 
391  gttic(Compute_Rg);
392  // Compute R * g
393  Errors Rg = *this * grad;
394  gttoc(Compute_Rg);
395 
396  gttic(Compute_minimizing_step_size);
397  // Compute minimizing step size
398  double step = -gradientSqNorm / gtsam::dot(Rg, Rg);
399  gttoc(Compute_minimizing_step_size);
400 
401  gttic(Compute_point);
402  // Compute steepest descent point
403  grad *= step;
404  gttoc(Compute_point);
405 
406  return grad;
407  }
408 
409  /* ************************************************************************* */
411  Errors e;
412  for (const GaussianFactor::shared_ptr& factor: *this) {
413  JacobianFactor::shared_ptr Ai = convertToJacobianFactorPtr(factor);
414  e.push_back((*Ai) * x);
415  }
416  return e;
417  }
418 
419  /* ************************************************************************* */
420  void GaussianFactorGraph::multiplyHessianAdd(double alpha,
421  const VectorValues& x, VectorValues& y) const {
422  for (const GaussianFactor::shared_ptr& f: *this)
423  f->multiplyHessianAdd(alpha, x, y);
424  }
425 
426  /* ************************************************************************* */
427  void GaussianFactorGraph::multiplyInPlace(const VectorValues& x, Errors& e) const {
428  multiplyInPlace(x, e.begin());
429  }
430 
431  /* ************************************************************************* */
432  void GaussianFactorGraph::multiplyInPlace(const VectorValues& x, const Errors::iterator& e) const {
433  Errors::iterator ei = e;
434  for (const GaussianFactor::shared_ptr& factor: *this) {
435  JacobianFactor::shared_ptr Ai = convertToJacobianFactorPtr(factor);
436  *ei = (*Ai)*x;
437  ei++;
438  }
439  }
440 
441  /* ************************************************************************* */
443  typedef JacobianFactor J;
444  for (const GaussianFactor::shared_ptr& factor: factors) {
445  J::shared_ptr jacobian(std::dynamic_pointer_cast<J>(factor));
446  if (jacobian && jacobian->get_model() && jacobian->get_model()->isConstrained()) {
447  return true;
448  }
449  }
450  return false;
451  }
452 
453  /* ************************************************************************* */
454  // x += alpha*A'*e
455  void GaussianFactorGraph::transposeMultiplyAdd(double alpha, const Errors& e,
456  VectorValues& x) const {
457  // For each factor add the gradient contribution
458  Errors::const_iterator ei = e.begin();
459  for (const sharedFactor& factor: *this) {
460  JacobianFactor::shared_ptr Ai = convertToJacobianFactorPtr(factor);
461  Ai->transposeMultiplyAdd(alpha, *(ei++), x);
462  }
463  }
464 
466  //void residual(const GaussianFactorGraph& fg, const VectorValues &x, VectorValues &r) {
467  // Key i = 0 ;
468  // for (const GaussianFactor::shared_ptr& factor, fg) {
469  // JacobianFactor::shared_ptr Ai = convertToJacobianFactorPtr(factor);
470  // r[i] = Ai->getb();
471  // i++;
472  // }
473  // VectorValues Ax = VectorValues::SameStructure(r);
474  // multiply(fg,x,Ax);
475  // axpy(-1.0,Ax,r);
476  //}
477 
479  //void multiply(const GaussianFactorGraph& fg, const VectorValues &x, VectorValues &r) {
480  // r.setZero();
481  // Key i = 0;
482  // for (const GaussianFactor::shared_ptr& factor, fg) {
483  // JacobianFactor::shared_ptr Ai = convertToJacobianFactorPtr(factor);
484  // Vector &y = r[i];
485  // for (JacobianFactor::const_iterator j = Ai->begin(); j != Ai->end(); ++j) {
486  // y += Ai->getA(j) * x[*j];
487  // }
488  // ++i;
489  // }
490  //}
491 
492  /* ************************************************************************* */
493  VectorValues GaussianFactorGraph::transposeMultiply(const Errors& e) const
494  {
495  VectorValues x;
496  Errors::const_iterator ei = e.begin();
497  for (const sharedFactor& factor: *this) {
498  JacobianFactor::shared_ptr Ai = convertToJacobianFactorPtr(factor);
499  for (JacobianFactor::const_iterator j = Ai->begin(); j != Ai->end(); ++j) {
500  // Create the value as a zero vector if it does not exist.
501  pair<VectorValues::iterator, bool> xi = x.tryInsert(*j, Vector());
502  if(xi.second)
503  xi.first->second = Vector::Zero(Ai->getDim(j));
504  xi.first->second += Ai->getA(j).transpose() * *ei;
505  }
506  ++ ei;
507  }
508  return x;
509  }
510 
511  /* ************************************************************************* */
512  Errors GaussianFactorGraph::gaussianErrors(const VectorValues& x) const
513  {
514  Errors e;
515  for (const sharedFactor& factor: *this) {
516  JacobianFactor::shared_ptr Ai = convertToJacobianFactorPtr(factor);
517  e.push_back(Ai->error_vector(x));
518  }
519  return e;
520  }
521 
522  /* ************************************************************************* */
523  void GaussianFactorGraph::printErrors(
524  const VectorValues& values, const std::string& str,
525  const KeyFormatter& keyFormatter,
526  const std::function<bool(const Factor* /*factor*/,
527  double /*whitenedError*/, size_t /*index*/)>&
528  printCondition) const {
529  cout << str << "size: " << size() << endl << endl;
530  for (size_t i = 0; i < (*this).size(); i++) {
531  const sharedFactor& factor = (*this)[i];
532  const double errorValue =
533  (factor != nullptr ? (*this)[i]->error(values) : .0);
534  if (!printCondition(factor.get(), errorValue, i))
535  continue; // User-provided filter did not pass
536 
537  stringstream ss;
538  ss << "Factor " << i << ": ";
539  if (factor == nullptr) {
540  cout << "nullptr"
541  << "\n";
542  } else {
543  factor->print(ss.str(), keyFormatter);
544  cout << "error = " << errorValue << "\n";
545  }
546  cout << endl; // only one "endl" at end might be faster, \n for each
547  // factor
548  }
549  }
550 
551 } // namespace gtsam
gttoc
#define gttoc(label)
Definition: timing.h:296
timing.h
Timing utilities.
GaussianFactorGraph.h
Linear Factor Graph where all factors are Gaussians.
gtsam::GaussianFactorGraph::shared_ptr
std::shared_ptr< This > shared_ptr
shared_ptr to this class
Definition: GaussianFactorGraph.h:82
llt
EIGEN_DONT_INLINE void llt(const Mat &A, const Mat &B, Mat &C)
Definition: llt.cpp:5
Eigen::Block
Expression of a fixed-size or dynamic-size block.
Definition: Block.h:103
gtsam::Values::size
size_t size() const
Definition: Values.h:178
alpha
RealScalar alpha
Definition: level1_cplx_impl.h:147
gtsam::HessianFactor
A Gaussian factor using the canonical parameters (information form)
Definition: HessianFactor.h:100
s
RealScalar s
Definition: level1_cplx_impl.h:126
e
Array< double, 1, 3 > e(1./3., 0.5, 2.)
d
static const double d[K][N]
Definition: igam.h:11
screwPose2::xi
Vector xi
Definition: testPose2.cpp:148
optimize
int optimize(const SfmData &db, const NonlinearFactorGraph &graph, const Values &initial, bool separateCalibration=false)
Definition: timeSFMBAL.h:64
keys
const KeyVector keys
Definition: testRegularImplicitSchurFactor.cpp:40
c
Scalar Scalar * c
Definition: benchVecAdd.cpp:17
simple_graph::factors
const GaussianFactorGraph factors
Definition: testJacobianFactor.cpp:213
x
set noclip points set clip one set noclip two set bar set border lt lw set xdata set ydata set zdata set x2data set y2data set boxwidth set dummy x
Definition: gnuplot_common_settings.hh:12
EliminateableFactorGraph-inst.h
gtsam::JacobianFactor
Definition: JacobianFactor.h:91
cholesky.h
Efficient incomplete Cholesky on rank-deficient matrices, todo: constrained Cholesky.
gtsam::VectorValues::dot
double dot(const VectorValues &v) const
Definition: VectorValues.cpp:235
HessianFactor.h
Contains the HessianFactor class, a general quadratic factor.
gtsam::SymmetricBlockMatrix::selfadjointView
Eigen::SelfAdjointView< constBlock, Eigen::Upper > selfadjointView(DenseIndex I, DenseIndex J) const
Return the square sub-matrix that contains blocks(i:j, i:j).
Definition: SymmetricBlockMatrix.h:158
gtsam::Matrix
Eigen::MatrixXd Matrix
Definition: base/Matrix.h:39
gtsam::HessianFactor::shared_ptr
std::shared_ptr< This > shared_ptr
A shared_ptr to this class.
Definition: HessianFactor.h:109
gtsam::Factor
Definition: Factor.h:69
gtsam::FastSet< Key >
GaussianJunctionTree.h
exp
const EIGEN_DEVICE_FUNC ExpReturnType exp() const
Definition: ArrayCwiseUnaryOps.h:97
gtsam::HessianFactor::info
const SymmetricBlockMatrix & info() const
Return underlying information matrix.
Definition: HessianFactor.h:264
gtsam::Factor::begin
const_iterator begin() const
Definition: Factor.h:145
gtsam::Vector
Eigen::VectorXd Vector
Definition: Vector.h:38
result
Values result
Definition: OdometryOptimize.cpp:8
gtsam::JacobianFactor::getb
const constBVector getb() const
Definition: JacobianFactor.h:298
FactorGraph-inst.h
Factor Graph Base Class.
size
Scalar Scalar int size
Definition: benchVecAdd.cpp:17
ss
static std::stringstream ss
Definition: testBTree.cpp:31
gtsam::FactorGraph< GaussianFactor >
n
int n
Definition: BiCGSTAB_simple.cpp:1
gtsam::GaussianFactorGraph
Definition: GaussianFactorGraph.h:73
GaussianEliminationTree.h
J
JacobiRotation< float > J
Definition: Jacobi_makeJacobi.cpp:3
gtsam::VectorValues
Definition: VectorValues.h:74
Eigen::internal::negate
T negate(const T &x)
Definition: packetmath_test_shared.h:24
j
std::ptrdiff_t j
Definition: tut_arithmetic_redux_minmax.cpp:2
gttic_
#define gttic_(label)
Definition: timing.h:245
gtsam::KeyFormatter
std::function< std::string(Key)> KeyFormatter
Typedef for a function to format a key, i.e. to convert it to a string.
Definition: Key.h:35
gtsam::GaussianFactor::shared_ptr
std::shared_ptr< This > shared_ptr
shared_ptr to this class
Definition: GaussianFactor.h:42
x0
static Symbol x0('x', 0)
gtsam::dot
double dot(const V1 &a, const V2 &b)
Definition: Vector.h:195
gtsam::utils.visual_isam.step
def step(data, isam, result, truth, currPoseIndex, isamArgs=())
Definition: visual_isam.py:82
gtsam::JacobianFactor::augmentedJacobian
Matrix augmentedJacobian() const override
Definition: JacobianFactor.cpp:730
gtsam::Scatter
Definition: Scatter.h:49
BD
static double BD[10]
Definition: dawsn.c:93
gtsam::Factor::end
const_iterator end() const
Definition: Factor.h:148
gtsam::FastList< Vector >
size_t
std::size_t size_t
Definition: wrap/pybind11/include/pybind11/detail/common.h:476
g
void g(const string &key, int i)
Definition: testBTree.cpp:41
ordering
static enum @1096 ordering
y
Scalar * y
Definition: level1_cplx_impl.h:124
str
Definition: pytypes.h:1524
key
const gtsam::Symbol key('X', 0)
tree::f
Point2(* f)(const Point3 &, OptionalJacobian< 2, 3 >)
Definition: testExpression.cpp:218
Eigen::LLT
Standard Cholesky decomposition (LL^T) of a matrix and associated features.
Definition: LLT.h:66
gtsam::Factor::const_iterator
KeyVector::const_iterator const_iterator
Const iterator over keys.
Definition: Factor.h:82
gtsam::JacobianFactor::constBVector
constABlock::ConstColXpr constBVector
Definition: JacobianFactor.h:102
gtsam
traits
Definition: chartTesting.h:28
error
static double error
Definition: testRot3.cpp:37
gtsam::JacobianFactor::getA
constABlock getA(const_iterator variable) const
Definition: JacobianFactor.h:301
gtsam::SparseTriplets
std::vector< std::tuple< int, int, double > > SparseTriplets
Definition: GaussianFactorGraph.cpp:118
leaf::values
leaf::MyValues values
std
Definition: BFloat16.h:88
GaussianBayesTree.h
Gaussian Bayes Tree, the result of eliminating a GaussianJunctionTree.
gtsam::tol
const G double tol
Definition: Group.h:79
gtsam::JacobianFactor::shared_ptr
std::shared_ptr< This > shared_ptr
shared_ptr to this class
Definition: JacobianFactor.h:97
abs
#define abs(x)
Definition: datatypes.h:17
gtsam::hasConstraints
bool hasConstraints(const GaussianFactorGraph &factors)
Definition: GaussianFactorGraph.cpp:442
gtsam::EliminateableFactorGraph< GaussianFactorGraph >
gtsam::FactorGraph< GaussianFactor >::sharedFactor
std::shared_ptr< GaussianFactor > sharedFactor
Shared pointer to a factor.
Definition: FactorGraph.h:62
gtsam::Key
std::uint64_t Key
Integer nonlinear key type.
Definition: types.h:97
gtsam::Ordering
Definition: inference/Ordering.h:33
ceres::Vector
Eigen::Matrix< double, Eigen::Dynamic, 1 > Vector
Definition: gtsam/3rdparty/ceres/eigen.h:38
i
int i
Definition: BiCGSTAB_step_by_step.cpp:9
operator*
Vector3_ operator*(const Double_ &s, const Vector3_ &v)
Definition: InverseKinematicsExampleExpressions.cpp:42
gttic
#define gttic(label)
Definition: timing.h:295
debug.h
Global debugging flags.


gtsam
Author(s):
autogenerated on Thu Jun 13 2024 03:02:23