12345qiupeng
/
gtsam
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
gtsam/cpp/SymbolicBayesNet.h

45 lines
1.1 KiB
C
Raw Normal View History

Prototype framework for Bayes chain (new name for Chordal Bayes net) and Bayes tree class hierarchy
2009-10-28 04:23:19 +08:00
/**
BIG CHANGE: I got rid of the BayesChain/ChordalBayesNet classes and we now simply have a BayesNet class. It will just happen to be chordal when it is the result of an elimination. This will simplify a lot of things. The main renaming that happened is BayesChain -> BayesNet ChordalBayesNet -> GaussianBayesNet == BayesNet<ConditionalGaussian> SymbolicBayesChain -> SymbolicBayesNet == BayesNet<SymbolicConditional>
2009-11-01 03:53:20 +08:00
* @file SymbolicBayesNet.h
Prototype framework for Bayes chain (new name for Chordal Bayes net) and Bayes tree class hierarchy
2009-10-28 04:23:19 +08:00
* @brief Symbolic Chordal Bayes Net, the result of eliminating a factor graph
* @author Frank Dellaert
*/
// \callgraph
#pragma once
#include <list>
#include <boost/serialization/map.hpp>
#include <boost/serialization/list.hpp>
#include "Testable.h"
BIG CHANGE: I got rid of the BayesChain/ChordalBayesNet classes and we now simply have a BayesNet class. It will just happen to be chordal when it is the result of an elimination. This will simplify a lot of things. The main renaming that happened is BayesChain -> BayesNet ChordalBayesNet -> GaussianBayesNet == BayesNet<ConditionalGaussian> SymbolicBayesChain -> SymbolicBayesNet == BayesNet<SymbolicConditional>
2009-11-01 03:53:20 +08:00
#include "BayesNet.h"
Made some progress on symbolic analysis
2009-10-28 10:57:38 +08:00
#include "SymbolicConditional.h"
Refactoring to use a new Symbol key instead of strings in Bayes*, Gaussian*, Ordering, Symbolic*, VectorConfig. Renamed existing type-checking key Symbol<C,T> to TypedSymbol<C,T>
2010-01-18 03:34:57 +08:00
#include "Key.h"
Prototype framework for Bayes chain (new name for Chordal Bayes net) and Bayes tree class hierarchy
2009-10-28 04:23:19 +08:00
namespace gtsam {
/**
* Symbolic Bayes Chain, the (symbolic) result of eliminating a factor graph
*/
BIG CHANGE: 1) eliminate methods no longer return a shared pointer. Shared pointers are good for Factors and Conditionals (which are also non-copyable), because these are often passed around under the hood. However, a BayesNet is simple a list of shared pointers and hence does not cost a lot to return as an object (which is compiler-optimized anyway: there is no copy). So, the signature of all eliminate methods changed to simply return a BayesNet<> object (not a shared pointer). 2) GaussianBayesNet::optimize is now replaced by optimize(GaussianBayesNet) and returns a VectorConfig and not a shared pointer 3) GaussianBayesNet and SymbolicBayesNet are now simply typedefs, not derived classes. This is desirable because the BayesTree class uses templated methods that return BayesNet<Conditional>, not a specific BayesNet derived class.
2009-11-09 15:04:26 +08:00
typedef BayesNet<SymbolicConditional> SymbolicBayesNet;
Construct symbolic version of any Bayes net
2009-11-08 12:39:54 +08:00
BIG CHANGE: 1) eliminate methods no longer return a shared pointer. Shared pointers are good for Factors and Conditionals (which are also non-copyable), because these are often passed around under the hood. However, a BayesNet is simple a list of shared pointers and hence does not cost a lot to return as an object (which is compiler-optimized anyway: there is no copy). So, the signature of all eliminate methods changed to simply return a BayesNet<> object (not a shared pointer). 2) GaussianBayesNet::optimize is now replaced by optimize(GaussianBayesNet) and returns a VectorConfig and not a shared pointer 3) GaussianBayesNet and SymbolicBayesNet are now simply typedefs, not derived classes. This is desirable because the BayesTree class uses templated methods that return BayesNet<Conditional>, not a specific BayesNet derived class.
2009-11-09 15:04:26 +08:00
/**
* Construct from a Bayes net of any type
*/
template<class Conditional>
SymbolicBayesNet symbolic(const BayesNet<Conditional>& bn) {
SymbolicBayesNet result;
typename BayesNet<Conditional>::const_iterator it = bn.begin();
for (; it != bn.end(); it++) {
boost::shared_ptr<Conditional> conditional = *it;
Refactoring to use a new Symbol key instead of strings in Bayes*, Gaussian*, Ordering, Symbolic*, VectorConfig. Renamed existing type-checking key Symbol<C,T> to TypedSymbol<C,T>
2010-01-18 03:34:57 +08:00
Symbol key = conditional->key();
std::list<Symbol> parents = conditional->parents();
BIG CHANGE: 1) eliminate methods no longer return a shared pointer. Shared pointers are good for Factors and Conditionals (which are also non-copyable), because these are often passed around under the hood. However, a BayesNet is simple a list of shared pointers and hence does not cost a lot to return as an object (which is compiler-optimized anyway: there is no copy). So, the signature of all eliminate methods changed to simply return a BayesNet<> object (not a shared pointer). 2) GaussianBayesNet::optimize is now replaced by optimize(GaussianBayesNet) and returns a VectorConfig and not a shared pointer 3) GaussianBayesNet and SymbolicBayesNet are now simply typedefs, not derived classes. This is desirable because the BayesTree class uses templated methods that return BayesNet<Conditional>, not a specific BayesNet derived class.
2009-11-09 15:04:26 +08:00
SymbolicConditional::shared_ptr c(new SymbolicConditional(key, parents));
result.push_back(c);
Prototype framework for Bayes chain (new name for Chordal Bayes net) and Bayes tree class hierarchy
2009-10-28 04:23:19 +08:00
}
BIG CHANGE: 1) eliminate methods no longer return a shared pointer. Shared pointers are good for Factors and Conditionals (which are also non-copyable), because these are often passed around under the hood. However, a BayesNet is simple a list of shared pointers and hence does not cost a lot to return as an object (which is compiler-optimized anyway: there is no copy). So, the signature of all eliminate methods changed to simply return a BayesNet<> object (not a shared pointer). 2) GaussianBayesNet::optimize is now replaced by optimize(GaussianBayesNet) and returns a VectorConfig and not a shared pointer 3) GaussianBayesNet and SymbolicBayesNet are now simply typedefs, not derived classes. This is desirable because the BayesTree class uses templated methods that return BayesNet<Conditional>, not a specific BayesNet derived class.
2009-11-09 15:04:26 +08:00
return result;
}
Prototype framework for Bayes chain (new name for Chordal Bayes net) and Bayes tree class hierarchy
2009-10-28 04:23:19 +08:00
} /// namespace gtsam