|
||||||||
PREV CLASS NEXT CLASS | FRAMES NO FRAMES | |||||||
SUMMARY: NESTED | ENUM CONSTANTS | FIELD | METHOD | DETAIL: ENUM CONSTANTS | FIELD | METHOD |
ObjectEnum<Algorithm>
Algorithm
public enum Algorithm
Semantically, each "Algorithm" constant represents a distinctive implementation of an inference algorithm.
In practice, an "Algorithm" Object is a factory for manufacturing inference engines.
Some algorithms support partial derivatives
, others do not.
Call partials()
to get the algorithms that do.
Nested Class Summary | |
---|---|
static class |
Algorithm.EliminationOrderHeuristic
elimination order heuristic |
static class |
Algorithm.Key
keys to identify properties of Settings |
static class |
Algorithm.Order2JoinTree
induce a join tree from an elimination order |
static class |
Algorithm.Query
kind of query that an inference engine might support |
static class |
Algorithm.Result
the inference engine itself and other intermediate side-effects |
static class |
Algorithm.Setting
control inference engine compilation |
Enum Constant Summary | |
---|---|
hugin
hugin |
|
recursiveconditioning
recursive conditioning: RCEngine |
|
shenoyshafer
shenoy-shafer |
|
ssnormalized
shenoy-shafer -- normalized |
|
ssnormalizedmax
shenoy-shafer -- normalized |
|
zcnormalized
"zero-conscious" hugin -- normalized |
|
zeroconscioushugin
"zero-conscious" hugin |
Field Summary | |
---|---|
Class<? extends JointEngine> |
clazz
engine class |
Flavor |
flavor
join tree flavor |
boolean |
partial
supports compiling a partial derivative engine ?? |
Method Summary | |
---|---|
Map<Algorithm.Result,Object> |
compile(BayesianNetwork bn,
Map<Algorithm.Setting,?> settings)
create (i.e. |
static Algorithm |
forClass(Class<? extends JointEngine> clazz)
if you've got a JointEngine Class but you need an Algorithm |
static void |
main(String[] args)
test/debug |
static EliminationOrders.Record |
order(BayesianNetwork bn,
Map<Algorithm.Setting,?> ss)
elimination order |
static Collection<Algorithm> |
partials()
the flavors that support compiling a partial derivative engine |
static Algorithm |
valueOf(String name)
Returns the enum constant of this type with the specified name. |
static Algorithm[] |
values()
Returns an array containing the constants of this enum type, in the order they are declared. |
Methods inherited from class Enum |
---|
compareTo, equals, getDeclaringClass, hashCode, name, ordinal, toString, valueOf |
Methods inherited from class Object |
---|
getClass, notify, notifyAll, wait, wait, wait |
Enum Constant Detail |
---|
public static final Algorithm recursiveconditioning
RCEngine
public static final Algorithm ssnormalized
shenoy-shafer -- normalized
public static final Algorithm ssnormalizedmax
shenoy-shafer -- normalized
public static final Algorithm shenoyshafer
shenoy-shafer
public static final Algorithm hugin
hugin
public static final Algorithm zcnormalized
"zero-conscious" hugin -- normalized
public static final Algorithm zeroconscioushugin
"zero-conscious" hugin
Field Detail |
---|
public final Flavor flavor
public final boolean partial
compiling
a partial derivative engine
??
public final Class<? extends JointEngine> clazz
Method Detail |
---|
public static Algorithm[] values()
for (Algorithm c : Algorithm.values()) System.out.println(c);
public static Algorithm valueOf(String name)
name
- the name of the enum constant to be returned.
IllegalArgumentException
- if this enum type has no constant
with the specified name
NullPointerException
- if the argument is nullpublic static Collection<Algorithm> partials()
compiling
a partial derivative engine
public Map<Algorithm.Result,Object> compile(BayesianNetwork bn, Map<Algorithm.Setting,?> settings)
JointEngine
public static EliminationOrders.Record order(BayesianNetwork bn, Map<Algorithm.Setting,?> ss)
public static Algorithm forClass(Class<? extends JointEngine> clazz)
JointEngine
Class
but you need an Algorithm
public static void main(String[] args)
|
||||||||
PREV CLASS NEXT CLASS | FRAMES NO FRAMES | |||||||
SUMMARY: NESTED | ENUM CONSTANTS | FIELD | METHOD | DETAIL: ENUM CONSTANTS | FIELD | METHOD |