Analyst sometimes fails to detect min/max

I'm looking at this Kaggle competition: http://www.kaggle.com/c/titanic-gettingStarted/data

This is the training file: http://www.kaggle.com/c/titanic-gettingStarted/download/train.csv (CSV)

I tried to use this simple code (mostly taken from the NormalizeFile example):

public class Titanic {

static final String TRAIN_FILE = "train.csv";
static final String TRAIN_FILE_NORM = "train_norm.csv";

public static void dumpFieldInfo(EncogAnalyst analyst) {
    System.out.println("Fields found in file:");
    for (AnalystField field : analyst.getScript().getNormalize().getNormalizedFields()) {

        StringBuilder line = new StringBuilder();
        line.append(field.getName());
        line.append(",action=");
        line.append(field.getAction());
        line.append(",min=");
        line.append(field.getActualLow());
        line.append(",max=");
        line.append(field.getActualHigh());
        System.out.println(line.toString());
    }
}

public static void main(String[] args) {
    File trainFile = new File(TRAIN_FILE);
    File trainFileNorm = new File(TRAIN_FILE_NORM);

    EncogAnalyst analyst = new EncogAnalyst();
    AnalystWizard wizard = new AnalystWizard(analyst);
    wizard.wizard(trainFile, true, AnalystFileFormat.DECPNT_COMMA);

    dumpFieldInfo(analyst);

    final AnalystNormalizeCSV norm = new AnalystNormalizeCSV();
    norm.analyze(trainFile, true, CSVFormat.ENGLISH, analyst);
    norm.setProduceOutputHeaders(true);
    norm.normalize(trainFileNorm);

    Encog.getInstance().shutdown();
}

}

Output is:

Fields found in file:
passengerid,action=Normalize,min=1.0,max=891.0
survived,action=OneOf,min=0.0,max=0.0
pclass,action=Equilateral,min=0.0,max=0.0
name,action=Ignore,min=0.0,max=0.0
sex,action=OneOf,min=0.0,max=0.0
age,action=Normalize,min=0.42,max=80.0
sibsp,action=Equilateral,min=0.0,max=0.0
parch,action=Equilateral,min=0.0,max=0.0
ticket,action=Ignore,min=0.0,max=0.0
fare,action=Normalize,min=0.0,max=512.3292
cabin,action=Ignore,min=0.0,max=0.0
embarked,action=Equilateral,min=0.0,max=0.0

Look at the "parch" column:

It wants to normalize it as "Equilateral", but it's the "Number of Parents/Children Aboard", so I think that "Normalize" should be a better choice (no problem, I can change this).
It says that "min=0.0,max=0.0", but max should be about 5.

If you look at this line of the training file:

886,0,3,"Rice, Mrs. William (Margaret Norton)",female,39,0,5,382652,29.125,,Q

we have that "parch" is definitely 5.

If I change the normalization method for the "parch" column from Equilateral to Normalize

analyst.getScript().getNormalize().getNormalizedFields().get(7).setAction(NormalizationAction.Normalize);

it still fails to detect the max value.

I also tried

analyst.getScript().getNormalize().setMissingValues(new MeanAndModeMissing());

because I thought it could fail to find the max value because of missing values, but it still fails to detect the max value, I always get "min=0.0,max=0.0".

P.S. It also wants to normalize "survived" and "sex" as "OneOf", but we only have 2 values (0/1, male/female), so I think that "SingleField" normalization could also be a good choice (I can change this, however it uses 0/1 instead of the full -1/1 range for the SingleField values... I don't know if it works this way by design...).

jeffheaton / encog-java-core

Analyst sometimes fails to detect min/max #178