|
||||||||||
PREV NEXT | FRAMES NO FRAMES |
Storable
contract.
Accuracy
statistics and the underlying raw counts.Storable
to this container, by
calling its Storable.storeFields()
method and adding the
resulting field map.
Extraction.addToken(TokenDetails, Probability, boolean)
with a probability
of -1 ("confirmed").
TokenCounter.isWhitespaceAfterLast()
to true
.
<All>
.
ExtractionContainer
of answer
keys from an annotated text (in XML format).Util.asBoolean(Object)
on each element.
Util.asByte(Object)
on each element.
Util.asChar(Object)
on each element.
Util.asDouble(Object)
on each element.
Util.asFloat(Object)
on each element.
Util.asInt(Object)
on each element.
Util.asLong(Object)
on each element.
Util.asShort(Object)
on each element.
Util.asString(Object)
on each element.
DOMUtils.name(Attribute)
.
FieldContainer.add(FieldMap)
operation, any key/value pairs from this map are
added to field map prior to storing it.
Sensor
interface that stores a
configuration and provides a factory method to initialize a set of sensors.Storable
interface that implements the BaseStorable.toString()
method based on the field
map created by Storable.storeFields()
.ExtractionContainer
of
answer keys from from an annotated XML document.
featureList
.
XMLAdjuster.logEvent(String, String)
methods whenever
an event occurred to ensure the event is acceptable.
TargetClass
object for a given class name, if
defined in this target structure.
standard configuration
.
standard configuration
.
Processor
that collects
all the input arguments and processes the collected arguments when shutting
down.Collection
s and arrays.CombinationState.isDiscardPreceding()
to
false
.
Prediction
s.
VelocityService.TEMPLATE_DIR
and appending the VelocityService.TEMPLATE_EXT
.
TiesConfiguration.TiesConfiguration(String)
using "ties" as base name.
PropertiesConfiguration
format.
gzip
format.
control characters
(which are not allowed in
XML 1.0 and discouraged in XML 1.1).
true
, a fully incremental setup is
used where the trainer is trained on each document after the extractor
processed it.
Util.CONFIG_LOGGER_LOG
).
EvaluatedExtractionContainer.isMatchingPosition()
.
TrainEval.CONFIG_TUNE
is used.
Trainer.isTrainingOnlyErrors()
).
TrainEval.CONFIG_TUNE_EACH
is enabled.
[+|-]key[=value]
pairs in a string array.
logger
from the Util.CONFIG_LOGGER_LOG
and Util.CONFIG_LOGGER_SHOW
values in the provided configuration.
true
if this map contains a mapping for the
specified key.
TagVariety.TENTATIVE
.
TokenDetails
class by also storing
the context of a token.TokenDetails
instance.
ClassTrain.KEY_CLASSIFICATION
field for correct predictions:
"+".
TrainableClassifier.createClassifier(Set, TiesConfiguration)
using the
standard configuration.
TrainableClassifier.createClassifier(Set, File, TiesConfiguration)
without
specifying an run directory.
TrainableClassifier.createClassifier(Set, File, FeatureTransformer, String[],
TiesConfiguration)
.
TreeSet
.
DelimSepValues
format.
DelimSepValues
format.
DelimSepValues
format.
DelimSepValues
format.
FMetrics
instance of the
required type.
FMetrics
instance of the
required type from a field map.
Util.createObject(Class, String[])
, reading the
class name from the first element in the array.
Util.createObject(Class, Object[], Class)
, setting
the paramType
to the String
class.
FieldMap.createObject(Class)
for each of the field maps contained
in this container.
Recognition
s and passed
as argument to the Representation.buildContext(Element, String, String, String,
PriorRecognitions, Map, String)
method.
CombinationStrategy.createStrategy(Set, TiesConfiguration)
using the
standard configuration.
CombinationStrategy.CONFIG_COMBINATION
key in the provided configuration.
FeatureTransformer.createTransformer()
using the
standard configuration.
FeatureTransformer.CONFIG_TRANSFORMERS
key in the provided configuration.
Feature
class.StorableContainer
.
PropertiesConfiguration
format).
DefaultRepresentation.getHeadElement()
.
TextProcessor.KEY_OUT_DIRECTORY
configuration key in a given
configuration.
DefaultRepresentation.calculatePositionalValues(String, ElementPosition, List)
to
collapse a position in to one of five values.
Processor
that operates
on the contents of directories.Processor
that operates
on XML documents.Processor
that read
XML documents.NodeFilter
over to an
ElementProcessor
.ClassTrain.classifyAndTrain(FieldContainer, File, String)
.
input
text with the output of the TreeTagger.
TrainableClassifier.trainOnError(FeatureVector, String, Set)
method
instead of this one.
DOMUtils.name(Element)
.
FieldMap
s in this container in the
order they were added.
Object.equals(java.lang.Object)
contract.
Object.equals(java.lang.Object)
contract.
Object.equals(java.lang.Object)
contract.
Object.equals(java.lang.Object)
contract.
Object.equals(java.lang.Object)
contract.
Object.equals(java.lang.Object)
contract.
embeddingElements
.
Prediction
by also storing the
extracted text and location data.Storable
contract.
EvalStatus.TRUTH
.
EvalStatus.TRUTH
.
EvalStatus.UNKNOWN
.
Extraction
s of different
classes.ExtractionContainer.restoreEntries(FieldContainer)
.
Classifier
on a list of items/nodes and combines their results using a
CombinationStrategy
.Extractor
and
Trainer
.DefaultRepresentation
, node filter and combination strategy from
the provided configuration.
DefaultRepresentation
, node filter, combination strategy and
tokenizer factory from the provided configuration.
Storable
contract.
FeatureCount
class and the underlying raw
counts.FeatureSet
(a multi-set of
features).
FieldMap
s.StorableContainer
.
TokenProcessor
on the subset of tokens
that are children of an element accepted by a provided
ElementFilter
.Storable
contract.
FMetrics
class and the underlying raw counts.IOUtils.getExtension(File)
and preceding dot
).
IOUtils.getExtension(File)
and preceding dot
).
TargetClass
es at the top of the inheritance
hierarchy.
null
.
CombinationState
for this
instance.
last
added
string in the original text (indexing starts
with 0).
last
added
string in the original text (counting starts
with 0, as the first occurrence is the "0th repetition").
Extraction.isFirstTokenRepIgnored()
is true
.
TokenContainer.add(String)
operation.
TokenContainer.add(String)
operation.
last
added token in the
original text (counting starts with 0, as the first occurrence is the
"0th repetition").
TextTokenizer.isNormalizedWhitespacePrepended()
is true
.
Double.NaN
if not known/not relevant.
CombinationState
predicted by the current
classification model for this token if there was an classification error;
or null
if the expected state was correctly predicted.
Recognition
s from the current
document.
sigmoid score
(activation value) of this prediction.
TargetClass
es.
null
if we're outside of any instance (CombinationState.OUTSIDE
).
int
value.
TagVariety
of this tag.
LocalFeature
s into global features,
adding the created global features to a linked list.
decision
and the correct
decision via OR.
Object.hashCode()
contract.
Object.hashCode()
contract.
Object.hashCode()
contract.
Object.hashCode()
contract.
Object.hashCode()
contract.
Object.hashCode()
contract.
true
if there is a next element.
TextTokenizer.nextToken()
is preceded by whitespace (i.e., text not matched by any token).
true
if there is a previous element.
ElementFilter.matches(Element)
or ElementFilter.prefers(Element)
on elements of
this document.
stored
configuration
.
InsideOutsideStrategy.isBStartingAll()
to false
).
control characters
are deleted (these
characters are not allowed in XML 1.0 and discouraged in XML 1.1).
true
if this map contains no key-value mappings.
true
, the positions of extraction and answer keys must
match; otherwise only their contents must match (string compare).
TextTokenizer.getNormalizedWhitespace()
) to those tokens where TextTokenizer.hasPrecedingWhitespace()
would return true
.
true
the trainer only ensures that all answer keys exist
and can be located in the document instead of doing any training.
true
if training the embedded filter is enabled
(default).
last
added
string.
last
added
token.
last
added
string.
last
added
token.
Feature
s stored in this vector.
Feature
s stored in this vector.
Class
of the stored object.
Class
of the stored object.
Pattern
by
combining several alternatives.
ClassTrain.CORRECT_CLASS
if the correct class was predicted or the
wrongly predicted class in case of an error.
TextProcessor.KEY_DIRECTORY
is used instead.
TokenContainer.add(String)
operation
contains the specified token.
TokenContainer.add(String)
operation.
TextTokenizer.nextToken()
.
PropertiesConfiguration
or
XML
format.
TiesConfiguration.CONFIG_LANG
} key (if this key doesn't exist, the language of
default locale used by the Java Virtual Machine is used).
null
, but null
are not
allowed).
null
, but null
are not
allowed).
standard configuration
to specified
files (or standard out).
MetaClassifier
.
key[=value]
pair.
MultiBinaryClassifier
.
FMetrics
for different types.FMetrics
and the sums and averages calculated over them.MultiValueMap
allows storing multiple values for each key.HashMap
as storage.
int
whose value can be changed.TextUtils.NEWLINE_ALTERNATIVES
in a non-capturing group).
null
if there are no
more tokens left in the provided text.
OneAgainstTheRestClassifier
.
gzip
format.
gzip
format).
gzip
format).
ElementFilter
s
should match elements.null
).
cause
.
push
ed into this container.
push
ed into this container.
push
ed into this container.
push
ed into this container.
TextTokenizer.nextToken()
.
TextTokenizer.nextToken()
matches the defined whitespace pattern.
Storable
contract.
EvalStatus.UNKNOWN
.
Prediction
s based on their
probabilities.Recognition
s that should
be considered in the context representation.Double.NaN
(unknown).
ExtractionContainer
of
answer keys from from an annotated XML document.
TextProcessor.doProcess(Reader, Writer, ContextMap)
method and invokes a post-processor, if configured.
TextProcessor.process(File, Writer, ContextMap)
method.
TextProcessor.process(Reader, Writer, ContextMap)
method.
process
method.
TextProcessor.process(URLConnection, Writer, ContextMap)
method.
TextProcessor.process(Reader, Writer, ContextMap)
method.
cause
.
TokenProcessor
.
TokenProcessor
.
Configuration.getProperty(String)
is empty.
XMLAdjuster.isEscapingPseudoEntities()
is true
.
LocalFeature.OPEN
and
LocalFeature.CLOSE
character).
AnswerBuilder.process(Document, Writer, ContextMap)
method of an instance of
this class.
text/uri-list
) into an array of strings.
text/uri-list
) into an array of strings.
FieldContainer
s.
EvaluatedExtractionContainer
.ReEvaluator.reEvalulate(ExtractionContainer, EvaluatedExtractionContainer)
.
GlobalFeature
s to remove extraneous
FeatureType.MARKER
features.
added
to this
container.
input
matched by the
given pattern matcher with the given
replacement.
input
that matches the given
Pattern
with the given replacement.
Representation
to convert elements into
feature vectors.CombinationStrategy.state()
of this instance to the initial value
CombinationState.OUTSIDE
.
CombinationStrategy.reset()
method to query whether the last
extraction should be discarded, analogously to
CombinationState.isDiscardPreceding()
.
CombinationStrategy.reset()
method to query whether the last
extraction should be discarded, analogously to
CombinationState.isDiscardPreceding()
.
Storable
objects and support serialization
and deserialization of these objects in a human-readable format.TextTokenizer.nextToken()
.
PropertiesConfiguration
format.
PropertiesConfiguration
format.
TextTokenizer.isNormalizedWhitespacePrepended()
is true
.
TextTokenizer.getNormalizedWhitespace()
) to those tokens where TextTokenizer.hasPrecedingWhitespace()
would return true
.
int
value.
TagVariety
of this tag.
TextUtils.shorten(String, int, int)
, using the same number
of characters at the start and the end of the shortened string.
TextUtils.shorten(String, int, int)
, showing up to
24 characters at the start and the end of the shortened
string.
TrainableClassifier.trainOnError(FeatureVector, String, Set)
to decide
whether to train an instance.
FilteringTokenWalker
whenever some
tokens are skipped.
FilteringTokenWalker
whenever some
tokens are skipped.
FilteringTokenWalker
whenever some tokens
are skipped.MultiValueMap
that sorts the values
stored for each key, discarding duplicates.TreeMap
.
FieldMap
.Storable
objects and support serialization of
these objects in a human-readable format, by storing them in a
FieldContainer
.FieldContainer.store(Writer)
.
Storable
items in this object to a field
container for serialization.
Storable
items in this object to a field
container for serialization.
FMetrics
extension that additionally
calculates a StatisticalSummary
of the
intermediate precision, recall, and F1 metrics resulting from different
update
operations.Storable
contract.
Runnable
tasks.Thread.NORM_PRIORITY
) for threads.
Processor
that operates
on text documents.TiesConfiguration.addConfiguration(Configuration, Configuration)
.
TiesConfiguration.load(String)
.
TrainableClassifier
should extend this method and
the corresponding constructor from Element
to
ensure (de)serialization works as expected.
DefaultRepresentation.calculateValuesFromText(String, String, List)
to determine the
"tokenType" value.
TextTokenizer
s of
different types.TokenizerFactory.CONFIG_TOKEN_PATTERNS
and
TokenizerFactory.CONFIG_WHITESPACE_PATTERN
keys of the provided configuration.
TokenizerFactory.CONFIG_TOKEN_PATTERNS
and
TokenizerFactory.CONFIG_WHITESPACE_PATTERN
keys of the provided configuration,
adapted by
appending the suffix
.
TokenProcessor
.Storable
object,
printing all field name/value pairs in the order used to insert them
into the FieldMap
.
FieldMap
.
TrainableClassifier.doTrain(FeatureVector, String, ContextMap)
method.trainable classifier
for training.Classifier
to be used for extraction.standard configuration
to configure the
training mode and the superclasses.
TrainableClassifier.trainOnError(FeatureVector, String,
java.util.Set)
method on the stored trainable classifier.
ContextDetails
class by
also storing the expected (true) and predicted state on of a token.TokenDetails
instance.
Class
of the stored object.
Util
instances should NOT be constructed in standard
programming.
FMetrics
of the specified type.
FMetrics
of the specified type.
FMetrics
containing the sums and
averages over all types.
FMetrics
containing the sums and
averages over all types.
statistical
summaries of precision, recall, and F1 metrics
over all types,
if calculated.
statistical
summaries of precision, recall, and F1 metrics
over all types,
if calculated by the used implementation.
Trainer.resetGlobalAccuracy()
) by
each classifier.
statistical
summaries of precision, recall, and F1 metrics
of the specified type,
if calculated.
statistical
summaries of precision, recall, and F1 metrics
of the specified type,
if calculated by the used implementation.
Winnow classifier
.Winnow
algorithm.Storable
contract.
EvalStatus.UNKNOWN
.
Winnow
.TextTokenizer
s
for XML-like input.
|
||||||||||
PREV NEXT | FRAMES NO FRAMES |