public final class ItemClassifierSettings extends Object implements Serializable
| Modifier and Type | Field and Description |
|---|---|
edu.columbia.tjw.item.ItemSettings |
_settings |
| Constructor and Description |
|---|
ItemClassifierSettings(ItemClassifierSettings base_,
edu.columbia.tjw.item.ItemSettings settings_) |
ItemClassifierSettings(edu.columbia.tjw.item.ItemSettings settings_,
String intercept_,
edu.columbia.tjw.item.base.SimpleStatus status_,
int maxParamCount_,
List<String> regressors_,
Set<String> curveRegressors_)
Use this to construct settings unless you really know what you're doing.
|
| Modifier and Type | Method and Description |
|---|---|
SortedSet<edu.columbia.tjw.item.base.SimpleRegressor> |
getCurveRegressors() |
edu.columbia.tjw.item.ItemCurveFactory<edu.columbia.tjw.item.base.SimpleRegressor,edu.columbia.tjw.item.base.StandardCurveType> |
getFactory() |
edu.columbia.tjw.item.base.SimpleStatus |
getFromStatus() |
edu.columbia.tjw.item.base.SimpleRegressor |
getIntercept() |
int |
getMaxParamCount() |
Set<edu.columbia.tjw.item.base.SimpleRegressor> |
getNonCurveRegressors() |
List<edu.columbia.tjw.item.base.SimpleRegressor> |
getRegressors() |
edu.columbia.tjw.item.ItemSettings |
getSettings() |
static ItemClassifierSettings |
load(String filename_) |
void |
save(String fileName_) |
public ItemClassifierSettings(ItemClassifierSettings base_, edu.columbia.tjw.item.ItemSettings settings_)
public ItemClassifierSettings(edu.columbia.tjw.item.ItemSettings settings_,
String intercept_,
edu.columbia.tjw.item.base.SimpleStatus status_,
int maxParamCount_,
List<String> regressors_,
Set<String> curveRegressors_)
settings_ - Item settings to use, can be null.intercept_ - Which regressor will be the intercept. You don't need
to specify this in the data, it will be assumed to be 1.0 always.status_ - Which status are we projecting from (only interesting for
markov chains, just make a simple 2 status family if in doubt).maxParamCount_ - How many parameters is the model allowed to use.
Each curve will use approximately 3 params.regressors_ - The list (in order) of the regressors to use. This
must be in the same order that they appear in the features column.curveRegressors_ - The regressors that can support curves (i.e. they
are not binary flags), order doesn't matter for these.public edu.columbia.tjw.item.ItemSettings getSettings()
public edu.columbia.tjw.item.base.SimpleStatus getFromStatus()
public edu.columbia.tjw.item.base.SimpleRegressor getIntercept()
public edu.columbia.tjw.item.ItemCurveFactory<edu.columbia.tjw.item.base.SimpleRegressor,edu.columbia.tjw.item.base.StandardCurveType> getFactory()
public int getMaxParamCount()
public List<edu.columbia.tjw.item.base.SimpleRegressor> getRegressors()
public SortedSet<edu.columbia.tjw.item.base.SimpleRegressor> getCurveRegressors()
public Set<edu.columbia.tjw.item.base.SimpleRegressor> getNonCurveRegressors()
public void save(String fileName_) throws IOException
IOExceptionpublic static ItemClassifierSettings load(String filename_) throws IOException
IOExceptionCopyright © 2020. All rights reserved.