Package | Description |
---|---|
org.apache.mahout.cf.taste.impl.eval | |
org.apache.mahout.cf.taste.impl.model | |
org.apache.mahout.cf.taste.impl.recommender.svd | |
org.apache.mahout.cf.taste.model |
Modifier and Type | Method and Description |
---|---|
protected void |
AverageAbsoluteDifferenceRecommenderEvaluator.processOneEstimate(float estimatedPreference,
Preference realPref) |
protected abstract void |
AbstractDifferenceRecommenderEvaluator.processOneEstimate(float estimatedPreference,
Preference realPref) |
protected void |
RMSRecommenderEvaluator.processOneEstimate(float estimatedPreference,
Preference realPref) |
Modifier and Type | Class and Description |
---|---|
class |
BooleanPreference
Encapsulates a simple boolean "preference" for an item whose value does not matter (is fixed at 1.0).
|
class |
GenericPreference
A simple
Preference encapsulating an item and preference value. |
Modifier and Type | Method and Description |
---|---|
Preference |
BooleanItemPreferenceArray.get(int i) |
Preference |
GenericUserPreferenceArray.get(int i) |
Preference |
BooleanUserPreferenceArray.get(int i) |
Preference |
GenericItemPreferenceArray.get(int i) |
Modifier and Type | Method and Description |
---|---|
Iterator<Preference> |
BooleanItemPreferenceArray.iterator() |
Iterator<Preference> |
GenericUserPreferenceArray.iterator() |
Iterator<Preference> |
BooleanUserPreferenceArray.iterator() |
Iterator<Preference> |
GenericItemPreferenceArray.iterator() |
Modifier and Type | Method and Description |
---|---|
void |
BooleanItemPreferenceArray.set(int i,
Preference pref) |
void |
GenericUserPreferenceArray.set(int i,
Preference pref) |
void |
BooleanUserPreferenceArray.set(int i,
Preference pref) |
void |
GenericItemPreferenceArray.set(int i,
Preference pref) |
Modifier and Type | Method and Description |
---|---|
static FastByIDMap<PreferenceArray> |
GenericDataModel.toDataMap(FastByIDMap<Collection<Preference>> data,
boolean byUser)
|
Constructor and Description |
---|
BooleanItemPreferenceArray(List<? extends Preference> prefs,
boolean forOneUser) |
BooleanUserPreferenceArray(List<? extends Preference> prefs) |
GenericItemPreferenceArray(List<? extends Preference> prefs) |
GenericUserPreferenceArray(List<? extends Preference> prefs) |
Modifier and Type | Method and Description |
---|---|
Preference |
ParallelSGDFactorizer.PreferenceShuffler.get(int i) |
Modifier and Type | Method and Description |
---|---|
protected void |
ParallelSGDFactorizer.update(Preference preference,
double mu)
TODO: this is the vanilla sgd by Tacaks 2009, I speculate that using scaling technique proposed in:
Towards Optimal One Pass Large Scale Learning with Averaged Stochastic Gradient Descent section 5, page 6
can be beneficial in term s of both speed and accuracy.
|
Modifier and Type | Method and Description |
---|---|
Preference |
PreferenceArray.get(int i) |
Modifier and Type | Method and Description |
---|---|
void |
PreferenceArray.set(int i,
Preference pref)
Sets preference at i from information in the given
Preference |
Copyright © 2008–2017 The Apache Software Foundation. All rights reserved.