Home | Trees | Indices | Help |
|
---|
|
Multiple example builders might make use of the same features. A feature builder object can be used in different example builders that require the same feature set.
|
|||
|
|||
|
|||
|
|||
|
|||
|
|||
|
|||
|
|||
|
|||
|
|
|
Multiple entities may have the same head token. This returns a list of the types of all entities whose head token this token is. If the FeatureBuilder.maximum flag is set, the list is truncated to a length of two, otherwise to a length of one. This is done because when token features (to which the annotated type belongs to) are combined into other features, a large number of annotated type features can lead to an exponential increase in the number of features. |
Token features are features describing an isolated word token. These subfeatures are often merged into such features like n-grams. This method produces and caches a set of feature names for a token in the sentenceGraph sentence. The various flags can be used to choose which attributes will be included in the feature name list.
|
Some machine learning tasks require feature values to be normalized to range [0,1]. The range is defined as the difference of the largest and smallest feature value in the current feature vector. If this method is used, it should be called as the last step after generating all features. |
Add a feature to the feature vector. If the feature already exists, its current value is replaced with the new value. All features are prefixed with FeatureBuilder.tag.
|
When the feature builder builds features, they are put to this feature vector.
|
Home | Trees | Indices | Help |
|
---|
Generated by Epydoc 3.0.1 on Thu Jul 19 18:20:45 2012 | http://epydoc.sourceforge.net |