Fixed example_imagesegmentation by adding InfAlg::setMaxIter(size_t)
[libdai.git] / include / dai / daialg.h
1 /* This file is part of libDAI - http://www.libdai.org/
2 *
3 * libDAI is licensed under the terms of the GNU General Public License version
4 * 2, or (at your option) any later version. libDAI is distributed without any
5 * warranty. See the file COPYING for more details.
6 *
7 * Copyright (C) 2006-2009 Joris Mooij [joris dot mooij at libdai dot org]
8 * Copyright (C) 2006-2007 Radboud University Nijmegen, The Netherlands
9 */
10
11
12 /// \file
13 /// \brief Defines the general interface for inference methods in libDAI (classes InfAlg, DaiAlg<>, DaiAlgFG and DaiAlgRG).
14
15
16 #ifndef __defined_libdai_daialg_h
17 #define __defined_libdai_daialg_h
18
19
20 #include <string>
21 #include <iostream>
22 #include <vector>
23 #include <dai/factorgraph.h>
24 #include <dai/regiongraph.h>
25 #include <dai/properties.h>
26
27
28 namespace dai {
29
30
31 /// InfAlg is an abstract base class, defining the common interface of all inference algorithms in libDAI.
32 /** \idea General marginalization functions like calcMarginal() now copy a complete InfAlg object. Instead,
33 * it would make more sense that they construct a new object without copying the FactorGraph or RegionGraph.
34 * Or they can simply be made methods of the general InfAlg class.
35 * \idea Use a PropertySet as output of an InfAlg, instead of functions like maxDiff() and Iterations().
36 */
37 class InfAlg {
38 public:
39 /// \name Constructors/destructors
40 //@{
41 /// Virtual destructor (needed because this class contains virtual functions)
42 virtual ~InfAlg() {}
43
44 /// Returns a pointer to a new, cloned copy of \c *this (i.e., virtual copy constructor)
45 virtual InfAlg* clone() const = 0;
46 //@}
47
48 /// \name Queries
49 //@{
50 /// Identifies itself for logging purposes
51 virtual std::string identify() const = 0;
52
53 /// Returns reference to underlying FactorGraph.
54 virtual FactorGraph &fg() = 0;
55
56 /// Returns constant reference to underlying FactorGraph.
57 virtual const FactorGraph &fg() const = 0;
58 //@}
59
60 /// \name Inference interface
61 //@{
62 /// Initializes all data structures of the approximate inference algorithm.
63 /** \note This method should be called at least once before run() is called.
64 */
65 virtual void init() = 0;
66
67 /// Initializes all data structures corresponding to some set of variables.
68 /** This method can be used to do a partial initialization after a part of the factor graph has changed.
69 * Instead of initializing all data structures, it only initializes those involving the variables in \a vs.
70 * \throw NOT_IMPLEMENTED if not implemented/supported
71 */
72 virtual void init( const VarSet &vs ) = 0;
73
74 /// Runs the approximate inference algorithm.
75 /** \note Before run() is called the first time, init() should have been called.
76 */
77 virtual Real run() = 0;
78
79 /// Returns the (approximate) marginal probability distribution of a variable.
80 /** \note Before this method is called, run() should have been called.
81 */
82 virtual Factor belief( const Var &v ) const { return belief( VarSet(v) ); }
83
84 /// Returns the (approximate) marginal probability distribution of a set of variables.
85 /** \note Before this method is called, run() should have been called.
86 * \throw NOT_IMPLEMENTED if not implemented/supported.
87 * \throw BELIEF_NOT_AVAILABLE if the requested belief cannot be calculated with this algorithm.
88 */
89 virtual Factor belief( const VarSet &vs ) const = 0;
90
91 /// Returns the (approximate) marginal probability distribution of the variable with index \a i.
92 /** For some approximate inference algorithms, using beliefV() is preferred to belief() for performance reasons.
93 * \note Before this method is called, run() should have been called.
94 */
95 virtual Factor beliefV( size_t i ) const { return belief( fg().var(i) ); }
96
97 /// Returns the (approximate) marginal probability distribution of the variables on which factor \a I depends.
98 /** For some approximate inference algorithms, using beliefF() is preferred to belief() for performance reasons.
99 * \note Before this method is called, run() should have been called.
100 */
101 virtual Factor beliefF( size_t I ) const { return belief( fg().factor(I).vars() ); }
102
103 /// Returns all beliefs (approximate marginal probability distributions) calculated by the algorithm.
104 /** \note Before this method is called, run() should have been called.
105 */
106 virtual std::vector<Factor> beliefs() const = 0;
107
108 /// Returns the logarithm of the (approximated) partition sum (normalizing constant of the factor graph).
109 /** \note Before this method is called, run() should have been called.
110 * \throw NOT_IMPLEMENTED if not implemented/supported
111 */
112 virtual Real logZ() const = 0;
113
114 /// Calculates the joint state of all variables that has maximum probability
115 /** \note Before this method is called, run() should have been called.
116 * \throw NOT_IMPLEMENTED if not implemented/supported
117 */
118 virtual std::vector<std::size_t> findMaximum() const { DAI_THROW(NOT_IMPLEMENTED); }
119
120 /// Returns maximum difference between single variable beliefs in the last iteration.
121 /** \throw NOT_IMPLEMENTED if not implemented/supported
122 */
123 virtual Real maxDiff() const { DAI_THROW(NOT_IMPLEMENTED); };
124
125 /// Returns number of iterations done (one iteration passes over the complete factorgraph).
126 /** \throw NOT_IMPLEMENTED if not implemented/supported
127 */
128 virtual size_t Iterations() const { DAI_THROW(NOT_IMPLEMENTED); };
129
130 /// Sets maximum number of iterations (one iteration passes over the complete factorgraph).
131 /** \throw NOT_IMPLEMENTED if not implemented/supported
132 */
133 virtual void setMaxIter( size_t /*maxiter*/ ) { DAI_THROW(NOT_IMPLEMENTED); }
134 //@}
135
136 /// \name Changing the factor graph
137 //@{
138 /// Clamp variable with index \a i to value \a x (i.e. multiply with a Kronecker delta \f$\delta_{x_i, x}\f$)
139 /** If \a backup == \c true, make a backup of all factors that are changed.
140 */
141 virtual void clamp( size_t i, size_t x, bool backup = false ) = 0;
142
143 /// Sets all factors interacting with variable with index \a i to one.
144 /** If \a backup == \c true, make a backup of all factors that are changed.
145 */
146 virtual void makeCavity( size_t i, bool backup = false ) = 0;
147 //@}
148
149 /// \name Backup/restore mechanism for factors
150 //@{
151 /// Make a backup copy of factor \a I
152 /** \throw MULTIPLE_UNDO if a backup already exists
153 */
154 virtual void backupFactor( size_t I ) = 0;
155 /// Make backup copies of all factors involving the variables in \a vs
156 /** \throw MULTIPLE_UNDO if a backup already exists
157 */
158 virtual void backupFactors( const VarSet &vs ) = 0;
159
160 /// Restore factor \a I from its backup copy
161 virtual void restoreFactor( size_t I ) = 0;
162 /// Restore the factors involving the variables in \a vs from their backup copies
163 virtual void restoreFactors( const VarSet &vs ) = 0;
164 //@}
165
166 /// \name Managing parameters
167 //@{
168 /// Set parameters of this inference algorithm.
169 /** The parameters are set according to the PropertySet \a opts.
170 * The values can be stored either as std::string or as the type of the corresponding MF::props member.
171 */
172 virtual void setProperties( const PropertySet &opts ) = 0;
173 /// Returns parameters of this inference algorithm converted into a PropertySet.
174 virtual PropertySet getProperties() const = 0;
175 /// Returns parameters of this inference algorithm formatted as a string in the format "[key1=val1,key2=val2,...,keyn=valn]".
176 virtual std::string printProperties() const = 0;
177 //@}
178 };
179
180
181 /// Combines the abstract base class InfAlg with a graphical model (e.g., a FactorGraph or RegionGraph).
182 /** Inference algorithms in libDAI directly inherit from a DAIAlg, currently either
183 * from a DAIAlg<FactorGraph> or from a DAIAlg<RegionGraph>.
184 *
185 * \tparam GRM Should be castable to FactorGraph
186 * \idea A DAIAlg should not inherit from a FactorGraph or RegionGraph, but should
187 * store a reference to the graphical model object. This prevents needless copying
188 * of (possibly large) data structures. Disadvantage: the caller must not change
189 * the graphical model between calls to the inference algorithm (maybe a smart_ptr
190 * or some locking mechanism would help here?).
191 */
192 template <class GRM>
193 class DAIAlg : public InfAlg, public GRM {
194 public:
195 /// \name Constructors/destructors
196 //@{
197 /// Default constructor
198 DAIAlg() : InfAlg(), GRM() {}
199
200 /// Construct from GRM
201 DAIAlg( const GRM &grm ) : InfAlg(), GRM(grm) {}
202 //@}
203
204 /// \name Queries
205 //@{
206 /// Returns reference to underlying FactorGraph.
207 FactorGraph &fg() { return (FactorGraph &)(*this); }
208
209 /// Returns constant reference to underlying FactorGraph.
210 const FactorGraph &fg() const { return (const FactorGraph &)(*this); }
211 //@}
212
213 /// \name Changing the factor graph
214 //@{
215 /// Clamp variable with index \a i to value \a x (i.e. multiply with a Kronecker delta \f$\delta_{x_i, x}\f$)
216 /** If \a backup == \c true, make a backup of all factors that are changed.
217 */
218 void clamp( size_t i, size_t x, bool backup = false ) { GRM::clamp( i, x, backup ); }
219
220 /// Sets all factors interacting with variable with index \a i to one.
221 /** If \a backup == \c true, make a backup of all factors that are changed.
222 */
223 void makeCavity( size_t i, bool backup = false ) { GRM::makeCavity( i, backup ); }
224 //@}
225
226 /// \name Backup/restore mechanism for factors
227 //@{
228 /// Make a backup copy of factor \a I
229 void backupFactor( size_t I ) { GRM::backupFactor( I ); }
230 /// Make backup copies of all factors involving the variables in \a vs
231 void backupFactors( const VarSet &vs ) { GRM::backupFactors( vs ); }
232
233 /// Restore factor \a I from its backup copy
234 void restoreFactor( size_t I ) { GRM::restoreFactor( I ); }
235 /// Restore the factors involving the variables in \a vs from their backup copies
236 void restoreFactors( const VarSet &vs ) { GRM::restoreFactors( vs ); }
237 //@}
238 };
239
240
241 /// Base class for inference algorithms that operate on a FactorGraph
242 typedef DAIAlg<FactorGraph> DAIAlgFG;
243
244 /// Base class for inference algorithms that operate on a RegionGraph
245 typedef DAIAlg<RegionGraph> DAIAlgRG;
246
247
248 /// Calculates the marginal probability distribution for \a vs using inference algorithm \a obj.
249 /** calcMarginal() works by clamping all variables in \a vs and calculating the partition sum for each clamped state.
250 * Therefore, it can be used in combination with any inference algorithm that can calculate/approximate partition sums.
251 * \param obj instance of inference algorithm to be used
252 * \param vs variables for which the marginal should be calculated
253 * \param reInit should be set to \c true if at least one of the possible clamped states would be invalid (leading to a factor graph with zero partition sum).
254 */
255 Factor calcMarginal( const InfAlg& obj, const VarSet& vs, bool reInit );
256
257 /// Calculates beliefs for all pairs of variables in \a vs using inference algorithm \a obj.
258 /** calcPairBeliefs() works by
259 * - clamping single variables in \a vs and calculating the partition sum and the single variable beliefs for each clamped state, if \a accurate == \c false;
260 * - clamping pairs of variables in \a vs and calculating the partition sum for each clamped state, if \a accurate == \c true.
261 *
262 * Therefore, it can be used in combination with any inference algorithm that can calculate/approximate partition sums (and single variable beliefs, if
263 * \a accurate == \c true).
264 * \param obj instance of inference algorithm to be used
265 * \param vs variables for which the pair beliefs should be calculated
266 * \param reInit should be set to \c true if at least one of the possible clamped states would be invalid (leading to a factor graph with zero partition sum).
267 * \param accurate if \c true, uses a slower but more accurate approximation algorithm
268 */
269 std::vector<Factor> calcPairBeliefs( const InfAlg& obj, const VarSet& vs, bool reInit, bool accurate=false );
270
271
272 } // end of namespace dai
273
274
275 #endif