GClasses
|
#include <GTransform.h>
Public Member Functions | |
GImputeMissingVals () | |
General-purpose constructor. More... | |
GImputeMissingVals (const GDomNode *pNode, GLearnerLoader &ll) | |
Deserializing constructor. More... | |
virtual | ~GImputeMissingVals () |
virtual GDomNode * | serialize (GDom *pDoc) const |
Marshal this object into a DOM, which can then be converted to a variety of serial formats. More... | |
void | setCollaborativeFilter (GCollaborativeFilter *pCF) |
Sets the collaborative filter used to impute missing values. Takes ownership of pCF. If no collaborative filter is set, the default is to use matrix factorization with some typical parameters. More... | |
void | setLabels (const GMatrix *pLabels) |
Specify a label matrix that should be appended with the training data. This object will not delete pLabels. It is expected that pLabels will remain valid at least until after train is next called. More... | |
virtual void | transform (const GVec &in, GVec &out) |
See the comment for GIncrementalTransform::transform. More... | |
virtual GMatrix * | transformBatch (const GMatrix &in) |
Unlike most other transforms, this one assumes that the matrix passed to this method is the same that was used to train it. (This assumption is necessary in order to utilize the additional label information that may be available at training time, which can be important for imputation.) More... | |
virtual void | untransform (const GVec &in, GVec &out) |
See the comment for GIncrementalTransform::untransform. More... | |
virtual void | untransformToDistribution (const GVec &in, GPrediction *pOut) |
See the comment for GIncrementalTransform::untransformToDistribution. More... | |
Public Member Functions inherited from GClasses::GIncrementalTransform | |
GIncrementalTransform () | |
GIncrementalTransform (const GDomNode *pNode) | |
virtual | ~GIncrementalTransform () |
const GRelation & | after () const |
Returns a relation object describing the data after it is transformed. More... | |
const GRelation & | before () const |
Returns a relation object describing the data before it is transformed. More... | |
GVec & | innerBuf () |
Returns a buffer of sufficient size to store an inner (transformed) vector. The caller should not to delete the buffer. The same buffer will be returned each time. More... | |
virtual GMatrix * | reduce (const GMatrix &in) |
This calls train, then calls transformBatch, and returns the result. More... | |
void | setAfter (GRelation *pRel) |
Sets the after relation. Takes ownership of pRel. More... | |
void | setBefore (GRelation *pRel) |
Sets the before relation. Takes ownership of pRel. More... | |
void | train (const GMatrix &data) |
Trains the transform on the data in pData. (This method may be a no-op for transformations that always behave in the same manner.) More... | |
void | train (const GRelation &pRelation) |
"Trains" the transform without any data. More... | |
virtual std::unique_ptr< GMatrix > | untransformBatch (const GMatrix &in) |
This assumes train was previously called, and untransforms all the rows in pIn and returns the results. More... | |
Public Member Functions inherited from GClasses::GTransform | |
GTransform () | |
GTransform (const GDomNode *pNode) | |
virtual | ~GTransform () |
Protected Member Functions | |
virtual GRelation * | trainInner (const GMatrix &data) |
See the comment for GIncrementalTransform::train. More... | |
virtual GRelation * | trainInner (const GRelation &relation) |
Throws an exception (because this transform cannot be trained without data) More... | |
Protected Member Functions inherited from GClasses::GIncrementalTransform | |
virtual GDomNode * | baseDomNode (GDom *pDoc, const char *szClassName) const |
Child classes should use this in their implementation of serialize. More... | |
Protected Attributes | |
GMatrix * | m_pBatch |
GCollaborativeFilter * | m_pCF |
const GMatrix * | m_pLabels |
GNominalToCat * | m_pNTC |
Additional Inherited Members | |
Static Public Member Functions inherited from GClasses::GIncrementalTransform | |
static void | test () |
Performs unit tests for this class. Throws an exception if there is a failure. More... | |
GClasses::GImputeMissingVals::GImputeMissingVals | ( | ) |
General-purpose constructor.
GClasses::GImputeMissingVals::GImputeMissingVals | ( | const GDomNode * | pNode, |
GLearnerLoader & | ll | ||
) |
Deserializing constructor.
|
virtual |
Marshal this object into a DOM, which can then be converted to a variety of serial formats.
Implements GClasses::GIncrementalTransform.
void GClasses::GImputeMissingVals::setCollaborativeFilter | ( | GCollaborativeFilter * | pCF | ) |
Sets the collaborative filter used to impute missing values. Takes ownership of pCF. If no collaborative filter is set, the default is to use matrix factorization with some typical parameters.
void GClasses::GImputeMissingVals::setLabels | ( | const GMatrix * | pLabels | ) |
Specify a label matrix that should be appended with the training data. This object will not delete pLabels. It is expected that pLabels will remain valid at least until after train is next called.
|
protectedvirtual |
See the comment for GIncrementalTransform::train.
Implements GClasses::GIncrementalTransform.
|
protectedvirtual |
Throws an exception (because this transform cannot be trained without data)
Implements GClasses::GIncrementalTransform.
See the comment for GIncrementalTransform::transform.
Implements GClasses::GIncrementalTransform.
Unlike most other transforms, this one assumes that the matrix passed to this method is the same that was used to train it. (This assumption is necessary in order to utilize the additional label information that may be available at training time, which can be important for imputation.)
Reimplemented from GClasses::GIncrementalTransform.
See the comment for GIncrementalTransform::untransform.
Implements GClasses::GIncrementalTransform.
|
virtual |
See the comment for GIncrementalTransform::untransformToDistribution.
Implements GClasses::GIncrementalTransform.
|
protected |
|
protected |
|
protected |
|
protected |