ml.shifu.guagua.yarn.example.nn
类 NNWorker

java.lang.Object
  ml.shifu.guagua.worker.AbstractWorkerComputable<NNParams,NNParams,GuaguaWritableAdapter<org.apache.hadoop.io.LongWritable>,GuaguaWritableAdapter<org.apache.hadoop.io.Text>>
      ml.shifu.guagua.yarn.example.nn.NNWorker

所有已实现的接口：

ml.shifu.guagua.worker.WorkerComputable<NNParams,NNParams>

public class NNWorker
extends ml.shifu.guagua.worker.AbstractWorkerComputable<NNParams,NNParams,GuaguaWritableAdapter<org.apache.hadoop.io.LongWritable>,GuaguaWritableAdapter<org.apache.hadoop.io.Text>>

NNWorker is used to compute NN model according to splits assigned. The result will be sent to master for accumulation.

Gradients in each worker will be sent to master to update weights of model in worker, which follows Encog's multi-core implementation.

构造方法摘要

NNWorker()

方法摘要

NNParams	doCompute(ml.shifu.guagua.worker.WorkerContext<NNParams,NNParams> workerContext)
org.encog.ml.data.MLDataSet	getTestingData()
org.encog.ml.data.MLDataSet	getTrainingData()
void	init(ml.shifu.guagua.worker.WorkerContext<NNParams,NNParams> workerContext)
void	initRecordReader(ml.shifu.guagua.io.GuaguaFileSplit fileSplit)
void	load(GuaguaWritableAdapter<org.apache.hadoop.io.LongWritable> currentKey, GuaguaWritableAdapter<org.apache.hadoop.io.Text> currentValue, ml.shifu.guagua.worker.WorkerContext<NNParams,NNParams> workerContext)
protected void	postLoad(ml.shifu.guagua.worker.WorkerContext<NNParams,NNParams> workerContext)
void	setTestingData(org.encog.ml.data.MLDataSet testingData)
void	setTrainingData(org.encog.ml.data.MLDataSet trainingData)

从类 ml.shifu.guagua.worker.AbstractWorkerComputable 继承的方法

compute, getRecordReader, preLoad, setRecordReader

从类 java.lang.Object 继承的方法

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

构造方法详细信息

NNWorker

public NNWorker()

方法详细信息

init

public void init(ml.shifu.guagua.worker.WorkerContext<NNParams,NNParams> workerContext)

指定者：

类 ml.shifu.guagua.worker.AbstractWorkerComputable<NNParams,NNParams,GuaguaWritableAdapter<org.apache.hadoop.io.LongWritable>,GuaguaWritableAdapter<org.apache.hadoop.io.Text>> 中的 init

doCompute

public NNParams doCompute(ml.shifu.guagua.worker.WorkerContext<NNParams,NNParams> workerContext)

指定者：

类 ml.shifu.guagua.worker.AbstractWorkerComputable<NNParams,NNParams,GuaguaWritableAdapter<org.apache.hadoop.io.LongWritable>,GuaguaWritableAdapter<org.apache.hadoop.io.Text>> 中的 doCompute

postLoad

protected void postLoad(ml.shifu.guagua.worker.WorkerContext<NNParams,NNParams> workerContext)

覆盖：

类 ml.shifu.guagua.worker.AbstractWorkerComputable<NNParams,NNParams,GuaguaWritableAdapter<org.apache.hadoop.io.LongWritable>,GuaguaWritableAdapter<org.apache.hadoop.io.Text>> 中的 postLoad

load

public void load(GuaguaWritableAdapter<org.apache.hadoop.io.LongWritable> currentKey,
                 GuaguaWritableAdapter<org.apache.hadoop.io.Text> currentValue,
                 ml.shifu.guagua.worker.WorkerContext<NNParams,NNParams> workerContext)

指定者：

类 ml.shifu.guagua.worker.AbstractWorkerComputable<NNParams,NNParams,GuaguaWritableAdapter<org.apache.hadoop.io.LongWritable>,GuaguaWritableAdapter<org.apache.hadoop.io.Text>> 中的 load

initRecordReader

public void initRecordReader(ml.shifu.guagua.io.GuaguaFileSplit fileSplit)
                      throws IOException

指定者：

类 ml.shifu.guagua.worker.AbstractWorkerComputable<NNParams,NNParams,GuaguaWritableAdapter<org.apache.hadoop.io.LongWritable>,GuaguaWritableAdapter<org.apache.hadoop.io.Text>> 中的 initRecordReader

抛出：

IOException

getTrainingData

public org.encog.ml.data.MLDataSet getTrainingData()

setTrainingData

public void setTrainingData(org.encog.ml.data.MLDataSet trainingData)

getTestingData

public org.encog.ml.data.MLDataSet getTestingData()

setTestingData

public void setTestingData(org.encog.ml.data.MLDataSet testingData)