Example usage for weka.classifiers.functions.neural NeuralConnection PURE_INPUT

Introduction

In this page you can find the example usage for weka.classifiers.functions.neural NeuralConnection PURE_INPUT.

Prototype

int PURE_INPUT

To view the source code for weka.classifiers.functions.neural NeuralConnection PURE_INPUT.

Click Source Link

Document

This unit is a pure input unit.

Usage

From source file:classifiers.mlp.MultilayerPerceptronCustom.java

License:Open Source License

/**
 * @return string describing the model.//from   www.  j a v  a  2s . c o  m
 */
public String toString() {
    // only ZeroR model?
    if (m_useDefaultModel) {
        StringBuffer buf = new StringBuffer();
        buf.append(this.getClass().getName().replaceAll(".*\\.", "") + "\n");
        buf.append(this.getClass().getName().replaceAll(".*\\.", "").replaceAll(".", "=") + "\n\n");
        buf.append("Warning: No model could be built, hence ZeroR model is used:\n\n");
        buf.append(m_ZeroR.toString());
        return buf.toString();
    }

    StringBuffer model = new StringBuffer(m_neuralNodes.length * 100);
    //just a rough size guess
    NeuralNode con;
    double[] weights;
    NeuralConnection[] inputs;
    for (int noa = 0; noa < m_neuralNodes.length; noa++) {
        con = (NeuralNode) m_neuralNodes[noa]; //this would need a change
                                               //for items other than nodes!!!
        weights = con.getWeights();
        inputs = con.getInputs();
        if (con.getMethod() instanceof SigmoidUnit) {
            model.append("Sigmoid ");
        } else if (con.getMethod() instanceof LinearUnit) {
            model.append("Linear ");
        }
        model.append("Node " + con.getId() + "\n    Inputs    Weights\n");
        model.append("    Threshold    " + weights[0] + "\n");
        for (int nob = 1; nob < con.getNumInputs() + 1; nob++) {
            if ((inputs[nob - 1].getType() & NeuralConnection.PURE_INPUT) == NeuralConnection.PURE_INPUT) {
                model.append(
                        "    Attrib " + m_instances.attribute(((NeuralEnd) inputs[nob - 1]).getLink()).name()
                                + "    " + weights[nob] + "\n");
            } else {
                model.append("    Node " + inputs[nob - 1].getId() + "    " + weights[nob] + "\n");
            }
        }
    }
    //now put in the ends
    for (int noa = 0; noa < m_outputs.length; noa++) {
        inputs = m_outputs[noa].getInputs();
        model.append("Class " + m_instances.classAttribute().value(m_outputs[noa].getLink()) + "\n    Input\n");
        for (int nob = 0; nob < m_outputs[noa].getNumInputs(); nob++) {
            if ((inputs[nob].getType() & NeuralConnection.PURE_INPUT) == NeuralConnection.PURE_INPUT) {
                model.append("    Attrib " + m_instances.attribute(((NeuralEnd) inputs[nob]).getLink()).name()
                        + "\n");
            } else {
                model.append("    Node " + inputs[nob].getId() + "\n");
            }
        }
    }
    return model.toString();
}

---

From source file:com.ifmo.recommendersystem.metafeatures.classifierbased.internal.extractors.MultilayerPerceptron.java

License:Open Source License

/**
 * @return string describing the model.//from w w w  .  j a va2 s  .c om
 */
@Override
public String toString() {
    // only ZeroR model?
    if (m_useDefaultModel) {
        StringBuffer buf = new StringBuffer();
        buf.append(this.getClass().getName().replaceAll(".*\\.", "") + "\n");
        buf.append(this.getClass().getName().replaceAll(".*\\.", "").replaceAll(".", "=") + "\n\n");
        buf.append("Warning: No model could be built, hence ZeroR model is used:\n\n");
        buf.append(m_ZeroR.toString());
        return buf.toString();
    }

    StringBuffer model = new StringBuffer(m_neuralNodes.length * 100);
    // just a rough size guess
    NeuralNode con;
    double[] weights;
    NeuralConnection[] inputs;
    for (NeuralConnection m_neuralNode : m_neuralNodes) {
        con = (NeuralNode) m_neuralNode; // this would need a change
        // for items other than nodes!!!
        weights = con.getWeights();
        inputs = con.getInputs();
        if (con.getMethod() instanceof SigmoidUnit) {
            model.append("Sigmoid ");
        } else if (con.getMethod() instanceof LinearUnit) {
            model.append("Linear ");
        }
        model.append("Node " + con.getId() + "\n    Inputs    Weights\n");
        model.append("    Threshold    " + weights[0] + "\n");
        for (int nob = 1; nob < con.getNumInputs() + 1; nob++) {
            if ((inputs[nob - 1].getType() & NeuralConnection.PURE_INPUT) == NeuralConnection.PURE_INPUT) {
                model.append(
                        "    Attrib " + m_instances.attribute(((NeuralEnd) inputs[nob - 1]).getLink()).name()
                                + "    " + weights[nob] + "\n");
            } else {
                model.append("    Node " + inputs[nob - 1].getId() + "    " + weights[nob] + "\n");
            }
        }
    }
    // now put in the ends
    for (NeuralEnd m_output : m_outputs) {
        inputs = m_output.getInputs();
        model.append("Class " + m_instances.classAttribute().value(m_output.getLink()) + "\n    Input\n");
        for (int nob = 0; nob < m_output.getNumInputs(); nob++) {
            if ((inputs[nob].getType() & NeuralConnection.PURE_INPUT) == NeuralConnection.PURE_INPUT) {
                model.append("    Attrib " + m_instances.attribute(((NeuralEnd) inputs[nob]).getLink()).name()
                        + "\n");
            } else {
                model.append("    Node " + inputs[nob].getId() + "\n");
            }
        }
    }
    return model.toString();
}

---

From source file:ffnn.MultilayerPerceptron.java

License:Open Source License

/**
 * @return string describing the model./*from w  w  w .  j  a  v  a  2s .co m*/
 */
@Override
public String toString() {
    // only ZeroR model?
    if (m_useDefaultModel) {
        StringBuffer buf = new StringBuffer();
        buf.append(this.getClass().getName().replaceAll(".*\\.", "") + "\n");
        buf.append(this.getClass().getName().replaceAll(".*\\.", "").replaceAll(".", "=") + "\n\n");
        buf.append("Warning: No model could be built, hence ZeroR model is used:\n\n");
        buf.append(m_ZeroR.toString());
        return buf.toString();
    }

    StringBuffer model = new StringBuffer(m_neuralNodes.length * 100);
    // just a rough size guess
    NeuralNode con;
    double[] weights;
    NeuralConnection[] inputs;
    for (NeuralConnection m_neuralNode : m_neuralNodes) {
        con = (NeuralNode) m_neuralNode; // this would need a change
                                         // for items other than nodes!!!
        weights = con.getWeights();
        inputs = con.getInputs();
        if (con.getMethod() instanceof SigmoidUnit) {
            model.append("Sigmoid ");
        } else if (con.getMethod() instanceof LinearUnit) {
            model.append("Linear ");
        }
        model.append("Node " + con.getId() + "\n    Inputs    Weights\n");
        model.append("    Threshold    " + weights[0] + "\n");
        for (int nob = 1; nob < con.getNumInputs() + 1; nob++) {
            if ((inputs[nob - 1].getType() & NeuralConnection.PURE_INPUT) == NeuralConnection.PURE_INPUT) {
                model.append(
                        "    Attrib " + m_instances.attribute(((NeuralEnd) inputs[nob - 1]).getLink()).name()
                                + "    " + weights[nob] + "\n");
            } else {
                model.append("    Node " + inputs[nob - 1].getId() + "    " + weights[nob] + "\n");
            }
        }
    }
    // now put in the ends
    for (NeuralEnd m_output : m_outputs) {
        inputs = m_output.getInputs();
        model.append("Class " + m_instances.classAttribute().value(m_output.getLink()) + "\n    Input\n");
        for (int nob = 0; nob < m_output.getNumInputs(); nob++) {
            if ((inputs[nob].getType() & NeuralConnection.PURE_INPUT) == NeuralConnection.PURE_INPUT) {
                model.append("    Attrib " + m_instances.attribute(((NeuralEnd) inputs[nob]).getLink()).name()
                        + "\n");
            } else {
                model.append("    Node " + inputs[nob].getId() + "\n");
            }
        }
    }
    return model.toString();
}

---

From source file:org.ssase.debt.classification.OnlineMultilayerPerceptron.java

License:Open Source License

/**
 * @return string describing the model./*w  w w.j a  va  2s . com*/
 */
public String toString() {
    // only ZeroR model?
    if (m_useDefaultModel) {
        StringBuffer buf = new StringBuffer();
        buf.append(this.getClass().getName().replaceAll(".*\\.", "") + "\n");
        buf.append(this.getClass().getName().replaceAll(".*\\.", "").replaceAll(".", "=") + "\n\n");
        buf.append("Warning: No model could be built, hence ZeroR model is used:\n\n");
        buf.append(m_ZeroR.toString());
        return buf.toString();
    }

    StringBuffer model = new StringBuffer(m_neuralNodes.length * 100);
    // just a rough size guess
    NeuralNode con;
    double[] weights;
    NeuralConnection[] inputs;
    for (int noa = 0; noa < m_neuralNodes.length; noa++) {
        con = (NeuralNode) m_neuralNodes[noa]; // this would need a change
        // for items other than
        // nodes!!!
        weights = con.getWeights();
        inputs = con.getInputs();
        if (con.getMethod() instanceof SigmoidUnit) {
            model.append("Sigmoid ");
        } else if (con.getMethod() instanceof LinearUnit) {
            model.append("Linear ");
        }
        model.append("Node " + con.getId() + "\n    Inputs    Weights\n");
        model.append("    Threshold    " + weights[0] + "\n");
        for (int nob = 1; nob < con.getNumInputs() + 1; nob++) {
            if ((inputs[nob - 1].getType() & NeuralConnection.PURE_INPUT) == NeuralConnection.PURE_INPUT) {
                model.append(
                        "    Attrib " + m_instances.attribute(((NeuralEnd) inputs[nob - 1]).getLink()).name()
                                + "    " + weights[nob] + "\n");
            } else {
                model.append("    Node " + inputs[nob - 1].getId() + "    " + weights[nob] + "\n");
            }
        }
    }
    // now put in the ends
    for (int noa = 0; noa < m_outputs.length; noa++) {
        inputs = m_outputs[noa].getInputs();
        model.append("Class " + m_instances.classAttribute().value(m_outputs[noa].getLink()) + "\n    Input\n");
        for (int nob = 0; nob < m_outputs[noa].getNumInputs(); nob++) {
            if ((inputs[nob].getType() & NeuralConnection.PURE_INPUT) == NeuralConnection.PURE_INPUT) {
                model.append("    Attrib " + m_instances.attribute(((NeuralEnd) inputs[nob]).getLink()).name()
                        + "\n");
            } else {
                model.append("    Node " + inputs[nob].getId() + "\n");
            }
        }
    }
    return model.toString();
}

---

From source file:uzholdem.classifier.OnlineMultilayerPerceptron.java

License:Open Source License

/**
 * @return string describing the model./*from  w  w  w  . j a v  a  2  s  . co m*/
 */
public String toString() {
    // only ZeroR model?
    if (m_ZeroR != null) {
        StringBuffer buf = new StringBuffer();
        buf.append(this.getClass().getName().replaceAll(".*\\.", "") + "\n");
        buf.append(this.getClass().getName().replaceAll(".*\\.", "").replaceAll(".", "=") + "\n\n");
        buf.append("Warning: No model could be built, hence ZeroR model is used:\n\n");
        buf.append(m_ZeroR.toString());
        return buf.toString();
    }

    StringBuffer model = new StringBuffer(m_neuralNodes.length * 100);
    //just a rough size guess
    NeuralNode con;
    double[] weights;
    NeuralConnection[] inputs;
    for (int noa = 0; noa < m_neuralNodes.length; noa++) {
        con = (NeuralNode) m_neuralNodes[noa]; //this would need a change
                                               //for items other than nodes!!!
        weights = con.getWeights();
        inputs = con.getInputs();
        if (con.getMethod() instanceof SigmoidUnit) {
            model.append("Sigmoid ");
        } else if (con.getMethod() instanceof LinearUnit) {
            model.append("Linear ");
        }
        model.append("Node " + con.getId() + "\n    Inputs    Weights\n");
        model.append("    Threshold    " + weights[0] + "\n");
        for (int nob = 1; nob < con.getNumInputs() + 1; nob++) {
            if ((inputs[nob - 1].getType() & NeuralConnection.PURE_INPUT) == NeuralConnection.PURE_INPUT) {
                model.append(
                        "    Attrib " + m_instances.attribute(((NeuralEnd) inputs[nob - 1]).getLink()).name()
                                + "    " + weights[nob] + "\n");
            } else {
                model.append("    Node " + inputs[nob - 1].getId() + "    " + weights[nob] + "\n");
            }
        }
    }
    //now put in the ends
    for (int noa = 0; noa < m_outputs.length; noa++) {
        inputs = m_outputs[noa].getInputs();
        model.append("Class " + m_instances.classAttribute().value(m_outputs[noa].getLink()) + "\n    Input\n");
        for (int nob = 0; nob < m_outputs[noa].getNumInputs(); nob++) {
            if ((inputs[nob].getType() & NeuralConnection.PURE_INPUT) == NeuralConnection.PURE_INPUT) {
                model.append("    Attrib " + m_instances.attribute(((NeuralEnd) inputs[nob]).getLink()).name()
                        + "\n");
            } else {
                model.append("    Node " + inputs[nob].getId() + "\n");
            }
        }
    }
    return model.toString();
}

---