Package | Description |
---|---|
com.structed.dal | |
com.structed.data | |
com.structed.models | |
com.structed.tutorials | |
com.structed.utils |
Modifier and Type | Method and Description |
---|---|
InstancesContainer |
StandardReader.readData(java.lang.String path,
java.lang.String dataSpliter,
java.lang.String valueSpliter)
reads the data and returns an InstanceContainer object
|
InstancesContainer |
Reader.readData(java.lang.String path,
java.lang.String dataSpliter,
java.lang.String valueSpliter) |
InstancesContainer |
RankReader.readData(java.lang.String path,
java.lang.String dataSpliter,
java.lang.String valueSpliter) |
InstancesContainer |
OcrReader.readData(java.lang.String path,
java.lang.String dataSpliter,
java.lang.String valueSpliter)
Reads the data for the OCR task, this reader parse the data that given in the same format as in: http://ai.stanford.edu/~btaskar/ocr/
Each example will contain a matrix where each row is an image of some character and the label will be the word
|
InstancesContainer |
LazyReader.readData(java.lang.String path,
java.lang.String dataSpliter,
java.lang.String valueSpliter)
a Lazy reader, it reads only the paths to the data and not the data itself
this reader will read the actual data on demand
|
InstancesContainer |
OcrReader.readDataMultiClass(java.lang.String path,
java.lang.String dataSpliter,
java.lang.String valueSpliter)
Reads the data for the OCR task, this reader parse the data that given in the same format as in: http://ai.stanford.edu/~btaskar/ocr/
Each example will contain a vector (image) of some character and the label will single character
This reader returns a data set for multi-class classification
|
Modifier and Type | Class and Description |
---|---|
class |
LazyInstancesContainer
Lazy instance container that loads the examples by demand
The standard container loads all the data, this container gets as input all the paths to the raw data and loads it when needed
|
Modifier and Type | Method and Description |
---|---|
static InstancesContainer |
Factory.getInstanceContainer(int type)
Get specific container
This function can be used also for adding new containers types
|
Modifier and Type | Method and Description |
---|---|
java.util.ArrayList<PredictedLabels> |
StructEDModel.predict(InstancesContainer instances,
java.util.ArrayList<java.lang.Double> task_loss_params,
int numPredictions2Return,
boolean verbose)
predict based on the model and return the scores of the best matches
|
void |
StructEDModel.train(InstancesContainer trainInstances,
java.util.ArrayList<java.lang.Double> task_loss_params,
InstancesContainer developInstances,
int epoch,
int isAvg,
boolean verbose)
Train the model on the train instances
|
Vector |
Classifier.train(Vector W,
InstancesContainer trainInstances,
java.util.List<java.lang.Double> params,
InstancesContainer developInstances,
int isAvg,
boolean verbose)
train the algorithm and return the final weights
|
Modifier and Type | Method and Description |
---|---|
static InstancesContainer |
OCRTutorial.getFold(InstancesContainer container,
int fold,
boolean isEqual)
Get a given fold from the whole dataset
|
Modifier and Type | Method and Description |
---|---|
static InstancesContainer |
OCRTutorial.getFold(InstancesContainer container,
int fold,
boolean isEqual)
Get a given fold from the whole dataset
|
Modifier and Type | Method and Description |
---|---|
static InstancesContainer |
ModelHandler.randomShuffle(InstancesContainer instances)
preforming random shuffle on the data instances
|
Modifier and Type | Method and Description |
---|---|
static InstancesContainer |
ModelHandler.randomShuffle(InstancesContainer instances)
preforming random shuffle on the data instances
|