97 lines
4.1 KiB
Java
97 lines
4.1 KiB
Java
package org.dromara.easyai.nerveEntity;
|
|
|
|
import org.dromara.easyai.entity.ThreeChannelMatrix;
|
|
import org.dromara.easyai.i.CustomEncoding;
|
|
import org.dromara.easyai.matrixTools.Matrix;
|
|
import org.dromara.easyai.matrixTools.MatrixList;
|
|
import org.dromara.easyai.matrixTools.MatrixOperation;
|
|
import org.dromara.easyai.i.ActiveFunction;
|
|
import org.dromara.easyai.i.OutBack;
|
|
|
|
import java.util.ArrayList;
|
|
import java.util.List;
|
|
import java.util.Map;
|
|
|
|
/**
|
|
* @author lidapeng
|
|
* 隐层神经元
|
|
* @date 9:30 上午 2019/12/21
|
|
*/
|
|
public class HiddenNerve extends Nerve {
|
|
private final boolean isConvFinish;//卷积最后一层
|
|
private final MatrixOperation matrixOperation = new MatrixOperation();
|
|
|
|
public HiddenNerve(int id, int depth, int upNub, int downNub, float studyPoint,
|
|
boolean init, ActiveFunction activeFunction, boolean isDynamic, int rzType, float lParam
|
|
, int kernLen, int matrixX, int matrixY, boolean isConvFinish, int coreNumber, int channelNo, float oneConvStudy, boolean norm
|
|
, CustomEncoding customEncoding, float gaMa, float gMaxTh, boolean auTo, float GRate) throws Exception {//隐层神经元
|
|
super(id, upNub, "HiddenNerve", downNub, studyPoint,
|
|
init, activeFunction, isDynamic, rzType, lParam, kernLen, depth, matrixX, matrixY
|
|
, coreNumber, channelNo, oneConvStudy, norm, customEncoding, gaMa, gMaxTh, auTo, GRate);
|
|
this.isConvFinish = isConvFinish;
|
|
}
|
|
|
|
@Override
|
|
public void input(long eventId, float parameter, boolean isKernelStudy, Map<Integer, Float> E
|
|
, OutBack outBack) throws Exception {//接收上一层的输入
|
|
boolean allReady = insertParameter(eventId, parameter);
|
|
if (allReady) {//参数齐了，开始计算 sigma - threshold
|
|
float sigma = calculation(eventId);
|
|
float out = activeFunction.function(sigma);//激活函数输出数值
|
|
if (isKernelStudy) {
|
|
outNub = out;
|
|
} else {
|
|
destoryParameter(eventId);
|
|
}
|
|
sendMessage(eventId, out, isKernelStudy, E, outBack);
|
|
}
|
|
}
|
|
|
|
@Override
|
|
protected void inputMatrixFeature(long eventId, List<Float> parameters, boolean isStudy, Map<Integer, Float> E, OutBack imageBack) throws Exception {
|
|
insertParameters(eventId, parameters);
|
|
float sigma = calculation(eventId);
|
|
float out = activeFunction.function(sigma);//激活函数输出数值
|
|
if (isStudy) {
|
|
outNub = out;
|
|
} else {
|
|
destoryParameter(eventId);
|
|
}
|
|
sendMessage(eventId, out, isStudy, E, imageBack);
|
|
}
|
|
|
|
@Override
|
|
protected void inputMatrix(long eventId, List<Matrix> matrix, boolean isStudy
|
|
, Map<Integer, Float> E, OutBack outBack, boolean needMatrix) throws Exception {
|
|
List<Matrix> myMatrix = conv(matrix);//处理过的矩阵
|
|
if (isConvFinish) {
|
|
Matrix ourMatrix;
|
|
if (myMatrix.size() == 1) {
|
|
ourMatrix = myMatrix.get(0);
|
|
} else {
|
|
MatrixList matrixList = new MatrixList(myMatrix.get(0), true, 100);
|
|
for (int i = 1; i < myMatrix.size(); i++) {
|
|
matrixList.add(myMatrix.get(i));
|
|
}
|
|
ourMatrix = matrixList.getMatrix();
|
|
}
|
|
if (!isStudy && needMatrix) {
|
|
outBack.getBackMatrix(ourMatrix, getId(), eventId);
|
|
}
|
|
sendMatrixList(eventId, matrixOperation.matrixToList(ourMatrix), isStudy, E, outBack);
|
|
} else {
|
|
sendMatrix(eventId, myMatrix, isStudy, E, outBack, needMatrix);
|
|
}
|
|
}
|
|
|
|
@Override
|
|
protected void inputThreeChannelMatrix(long eventId, ThreeChannelMatrix picture, boolean isKernelStudy, Map<Integer, Float> E, OutBack outBack, boolean needMatrix) throws Exception {
|
|
//接收三通道矩阵
|
|
List<Matrix> matrixList = new ArrayList<>();
|
|
matrixList.add(picture.getMatrixR());
|
|
matrixList.add(picture.getMatrixG());
|
|
matrixList.add(picture.getMatrixB());
|
|
demRedByMatrixList(eventId, matrixList, isKernelStudy, E, outBack, needMatrix);
|
|
}
|
|
}
|