edu.hitsz.c102c.cnn

Class CNN

java.lang.Object

edu.hitsz.c102c.cnn.CNN

All Implemented Interfaces:

java.io.Serializable

public class CNN
extends java.lang.Object
implements java.io.Serializable

Main class to build a comvolutional neural network.

See Also:

Serialized Form

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	CNN.LayerBuilder The constructor mode constructs layers.

Constructor Summary

Constructors

Constructor and Description
CNN(CNN.LayerBuilder layerBuilder, int batchSize) Initialize the convolutional neural network.

Method Summary

Modifier and Type	Method and Description
static CNN	loadModel(java.lang.String fileName) Load the model of this neural network from a file.
void	predict(Dataset testset, java.lang.String fileName) Write a file with predicted lables.
void	saveModel(java.lang.String fileName) Save convolutional network to a file.
void	setup(int batchSize) Set the parameters of each layer of the CNN network.
double	test(Dataset trainset) Test trained neural network.
void	train(Dataset trainset, int repeat) Train the network on the training set

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

CNN

public CNN(CNN.LayerBuilder layerBuilder,
           int batchSize)

Initialize the convolutional neural network.

Parameters:

layerBuilder - Network layer

inputMapSize - Enter the size of the map

classNum - The number of categories, the data set is required to convert the class label to a value of 0-classNum-1

Method Detail

train

public void train(Dataset trainset,
                  int repeat)

Train the network on the training set

Parameters:

trainset -

repeat - The number of iterations

test

public double test(Dataset trainset)

Test trained neural network.

Parameters:

trainset -

Returns:

precison of training.

predict

public void predict(Dataset testset,
                    java.lang.String fileName)

Write a file with predicted lables.

Parameters:

testset - test dataset.

fileName - ouput file with predicted lables.

setup

public void setup(int batchSize)

Set the parameters of each layer of the CNN network.

Parameters:

batchSize - * @param classNum

inputMapSize -

saveModel

public void saveModel(java.lang.String fileName)

Save convolutional network to a file.

Parameters:

fileName - output file name.

loadModel

public static CNN loadModel(java.lang.String fileName)

Load the model of this neural network from a file.

Parameters:

fileName - input file name

Returns: