Examples of Backpropagation

com.greentea.relaxation.algorithms.backprop.BackPropagation
org.encog.neural.networks.training.propagation.back.Backpropagation
This class implements a backpropagation training algorithm for feed forward neural networks. It is used in the same manner as any other training class that implements the Train interface. Backpropagation is a common neural network training algorithm. It works by analyzing the error of the output of the neural network. Each neuron in the output layer's contribution, according to weight, to this error is determined. These weights are then adjusted to minimize this error. This process continues working its way backwards through the layers of the neural network. This implementation of the backpropagation algorithm uses both momentum and a learning rate. The learning rate specifies the degree to which the weight matrixes will be modified through each iteration. The momentum specifies how much the previous learning iteration affects the current. To use no momentum at all specify zero. One primary problem with backpropagation is that the magnitude of the partial derivative is often detrimental to the training of the neural network. The other propagation methods of Manhatten and Resilient address this issue in different ways. In general, it is suggested that you use the resilient propagation technique for most Encog training tasks over back propagation.
org.neuroph.nnet.learning.BackPropagation
Back Propagation learning rule for Multi Layer Perceptron neural networks. @author Zoran Sevarac
zdenekdrahos.AI.BackPropagation.BackPropagation

Examples of com.greentea.relaxation.algorithms.backprop.BackPropagation

   {
      super(new NetworkBuilder());


      casCorMath = new CasCorMathematics(builder.getNetwork());


      backPropagation = new BackPropagation(builder.getNetwork());


      quickPropagation = new QuickPropagation(builder.getNetwork());
   }

View Full Code Here

Examples of org.encog.neural.networks.training.propagation.back.Backpropagation

    // train the neural network
    CalculateScore score = new TrainingSetScore(trainingSet);
    final MLTrain trainAlt = new NeuralSimulatedAnnealing(
        network, score, 10, 2, 100);


    final MLTrain trainMain = new Backpropagation(network, trainingSet,0.000001, 0.0);


    ((Propagation)trainMain).setNumThreads(1);
    final StopTrainingStrategy stop = new StopTrainingStrategy();
    trainMain.addStrategy(new Greedy());
    trainMain.addStrategy(new HybridStrategy(trainAlt));
    trainMain.addStrategy(stop);


    int epoch = 0;
    while (!stop.shouldStop()) {
      trainMain.iteration();
      System.out.println("Training " + what + ", Epoch #" + epoch
          + " Error:" + trainMain.getError());
      epoch++;
    }
    return trainMain.getError();
  }

View Full Code Here

Examples of org.encog.neural.networks.training.propagation.back.Backpropagation

    for(int i=0;i<TRIES;i++) {
    
      MLMethod method = EncogUtility.simpleFeedForward(INPUT_OUTPUT_COUNT,
          HIDDEN_COUNT, 0, INPUT_OUTPUT_COUNT, false);
      
      Propagation train = new Backpropagation((ContainsFlat)method, trainingData,1.7,0);
      //Propagation train = new ResilientPropagation((ContainsFlat)method, trainingData);
      ((Propagation)train).fixFlatSpot(true);
      
      int iteration = 0;
      do {
        train.iteration();
        
        iteration++;
      } while( train.getError()>0.01 );
      count[i] = iteration;
      System.out.println("Begin Try #" + (i+1) + ", took " + iteration + " iterations.");      
    }
    
    System.out.println("Tries: " + TRIES);

View Full Code Here

Examples of org.encog.neural.networks.training.propagation.back.Backpropagation

    network.reset();


    MLDataSet trainingSet = new BasicMLDataSet(input, output);


    // train the neural network
    MLTrain train = new Backpropagation(network, trainingSet, 0.7, 0.7);


    Stopwatch sw = new Stopwatch();
    sw.start();
    // run epoch of learning procedure
    for (int i = 0; i < ITERATIONS; i++) {
      train.iteration();
    }
    sw.stop();


    return sw.getElapsedMilliseconds();
  }

View Full Code Here

Examples of org.encog.neural.networks.training.propagation.back.Backpropagation

    // train the neural network
    CalculateScore score = new TrainingSetScore(trainingSet);
    final MLTrain trainAlt = new NeuralSimulatedAnnealing(
        network, score, 10, 2, 100);


    final MLTrain trainMain = new Backpropagation(network, trainingSet,0.000001, 0.0);


    ((Propagation)trainMain).setNumThreads(1);
    final StopTrainingStrategy stop = new StopTrainingStrategy();
    trainMain.addStrategy(new Greedy());
    trainMain.addStrategy(new HybridStrategy(trainAlt));
    trainMain.addStrategy(stop);


    int epoch = 0;
    while (!stop.shouldStop()) {
      trainMain.iteration();
      System.out.println("Training " + what + ", Epoch #" + epoch
          + " Error:" + trainMain.getError());
      epoch++;
    }
    return trainMain.getError();
  }

View Full Code Here

Examples of org.encog.neural.networks.training.propagation.back.Backpropagation

  /**
   * {@inheritDoc}
   */
  @Override
  public final void createTrainer(final boolean singleThreaded) {
    final Propagation train = new Backpropagation(getNetwork(),
        getTraining(), getLearningRate(), getMomentum());


    if (singleThreaded) {
      train.setNumThreads(1);
    } else {
      train.setNumThreads(0);
    }


    for (final Strategy strategy : getStrategies()) {
      train.addStrategy(strategy);
    }


    setTrain(train);
  }

View Full Code Here

Examples of org.encog.neural.networks.training.propagation.back.Backpropagation

    BasicNetwork network1 = NetworkUtil.createXORNetworkUntrained();
    BasicNetwork network2 = NetworkUtil.createXORNetworkUntrained();
    MLDataSet trainingData = new BasicMLDataSet(XOR.XOR_INPUT,XOR.XOR_IDEAL);
    
    // train network 1, no continue
    Backpropagation rprop1 = new Backpropagation(network1,trainingData,0.4,0.4);
    rprop1.iteration();
    rprop1.iteration();
    rprop1.iteration();
    rprop1.iteration();
    
    // train network 2, continue
    Backpropagation rprop2 = new Backpropagation(network2,trainingData,0.4,0.4);
    rprop2.iteration();
    rprop2.iteration();
    TrainingContinuation state = rprop2.pause();
    rprop2 = new Backpropagation(network2,trainingData,0.4,0.4);
    rprop2.resume(state);
    rprop2.iteration();
    rprop2.iteration();
    
    // verify weights are the same
    double[] weights1 = NetworkCODEC.networkToArray(network1);
    double[] weights2 = NetworkCODEC.networkToArray(network2);
    
    Assert.assertEquals(rprop1.getError(), rprop2.getError(), 0.01);
    Assert.assertEquals(weights1.length, weights2.length);
    Assert.assertArrayEquals(weights1, weights2, 0.01);
    
  }

View Full Code Here

Examples of org.encog.neural.networks.training.propagation.back.Backpropagation

           


           
           BasicNetwork network3 = NetworkUtil.createXORNetworknNguyenWidrowUntrained();
           
           MLTrain bpropNguyen = new Backpropagation( network3, trainingData3, 0.9, 0.8 );     
           train(i, bpropNguyen, "NguyenWidrowRandomizer" );
           
           BasicNetwork network2 = NetworkUtil.createXORNetworkUntrained();
           
           MLTrain bpropRange = new Backpropagation( network2, trainingData2, 0.9, 0.8 );     
           train(i, bpropRange,  "RangeRandomizer       ");
 
       }
   }

View Full Code Here

Examples of org.encog.neural.networks.training.propagation.back.Backpropagation

  {
    MLDataSet trainingData = new BasicMLDataSet(XOR.XOR_INPUT,XOR.XOR_IDEAL);
    
    BasicNetwork network = NetworkUtil.createXORNetworkUntrained();


    MLTrain bprop = new Backpropagation(network, trainingData, 0.7, 0.9);
    NetworkUtil.testTraining(bprop,0.01);
  }

View Full Code Here

Examples of org.encog.neural.networks.training.propagation.back.Backpropagation

      
      if( kFold>0 ) {
        trainingData = this.wrapTrainingData(trainingData);
      }


      MLTrain train = new Backpropagation((BasicNetwork) file.getObject(),
          trainingData, learningRate, momentum);
      
      if( kFold>0 ) {
        train = this.wrapTrainer(trainingData,train,kFold);
      }

View Full Code Here

0 1

TOP

All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.