/// <summary>
/// Initializes a new instance of the <see cref="T:com.kiranpatel.crimecluster.framework.ModelEvaluation"/> class.
/// </summary>
/// <param name="mixedMarkovModel">Mixed markov model.</param>
/// <param name="incidentService">Incident Service.</param>
/// <param name="logger">Logger.</param>
public ModelEvaluation(IMixedMarkovModel mixedMarkovModel, IIncidentService incidentService, IDistanceMeasure distanceMeasure, ILogger logger)
{
this.mixedMarkovModel = mixedMarkovModel;
this.incidentService = incidentService;
this.distanceMeasure = distanceMeasure;
this.logger = logger;
}
public void setDistanceMeasure(IDistanceMeasure distanceMeasure)
{
if (distanceMeasure != null)
{
_distanceMeasure = distanceMeasure;
}
}
/// <summary>
/// Initializes a new instance of the <see cref="T:com.kiranpatel.crimecluster.framework.LocationService"/> class.
/// </summary>
/// <param name="repository">Repository.</param>
/// <param name="logger">Logger.</param>
public LocationService(
IRepository repository,
ILogger logger,
IDistanceMeasure distanceMeasure)
: base(repository, logger)
{
this.distanceMeasure = distanceMeasure;
}
public SimpleKnnClassifier(
IDistanceMeasure distanceMeasure,
IQuantitativeDataNormalizer dataNormalizer,
Func <double, double> weightingFunc = null,
IDistanceMeasure similarityMeasure = null,
bool normalizeNumericValues = false)
: base(distanceMeasure, dataNormalizer, VoteForBestCategoricalValue, weightingFunc, similarityMeasure, normalizeNumericValues)
{
}
public BackwardsEliminationKnnRegressor(
IDistanceMeasure distanceMeasure,
IQuantitativeDataNormalizer dataNormalizer,
Func <double, double> weightingFunc = null,
IDistanceMeasure similarityMeasure = null,
bool normalizeNumericValues = false)
: base(distanceMeasure, dataNormalizer, FindBestRegressionValue, weightingFunc, similarityMeasure, normalizeNumericValues)
{
}
public AbstractLazyClassifier(IDistanceMeasure distanceMeasure, Dataset database, double[] weights)
{
this._distanceMeasure = distanceMeasure;
this._database = database;
this._classBasedWeights = new double[this._database.Metadata.Target.Values.Length][];
for (int i = 0; i < this._classBasedWeights.GetLength(0); i++)
{
this._classBasedWeights[i] = weights;
}
}
protected BackwardsEliminationPredictor(
IDistanceMeasure distanceMeasure,
IQuantitativeDataNormalizer dataNormalizer,
KnnResultHandler <TPredictionResult> resultHandlingFunc,
Func <double, double> weightingFunc = null,
IDistanceMeasure similarityMeasure = null,
bool normalizeNumericValues = false)
: base(distanceMeasure, dataNormalizer, resultHandlingFunc, weightingFunc, similarityMeasure, normalizeNumericValues)
{
}
/// <summary>
/// Initializes a new instance of the <see cref="T:com.kiranpatel.crimecluster.framework.DJClusterAlgorithm`1"/> class.
/// </summary>
/// <param name="configService">configuration service.</param>
/// <param name="logger">logger service.</param>
/// <param name="distanceMeasure">distance measure.</param>
public DJClusterAlgorithm(
IConfigurationService configService,
ILogger logger,
IDistanceMeasure distanceMeasure)
{
this.configService = configService;
this.logger = logger;
this.measure = distanceMeasure;
this.raduisEps = Convert.ToDouble(this.configService.Get(ConfigurationKey.DJClusterRadiusEps, "0.05"));
this.minPoints = Convert.ToInt32(this.configService.Get(ConfigurationKey.DJClusterMinPts, "10"));
}
public void setParameters(IDistanceMeasure distanceMeasure,
int k)
{
if (distanceMeasure != null)
{
_distanceMeasure = distanceMeasure;
}
if (k > 0)
{
_k = k;
}
}
public ModelBase(double missingValue,
int indexTargetAttribute,
int countAttributes,
double[][] data) :
base(missingValue,
indexTargetAttribute,
countAttributes)
{
DistanceMeasureEuclidean dme =
new DistanceMeasureEuclidean();
dme.setUseSqrt(false);
_distanceMeasure = dme;
}
public AbstractLazyClassifier(IDistanceMeasure distanceMeasure, Dataset database)
{
this._distanceMeasure = distanceMeasure;
this._database = database;
this._classBasedWeights = new double[this._database.Metadata.Target.Values.Length][];
for (int i = 0; i < this._classBasedWeights.GetLength(0); i++)
{
this._classBasedWeights[i] = new double[this._database.Metadata.Attributes.Length];
for (int j = 0; j < this._database.Metadata.Attributes.Length; j++)
{
this._classBasedWeights[i][j] = 1;
}
}
}
public SimpleKnnPredictor(
IDistanceMeasure distanceMeasure,
IQuantitativeDataNormalizer dataNormalizer,
KnnResultHandler <TPredictionResult> resultHandlingFunc,
Func <double, double> weightingFunc = null,
IDistanceMeasure similarityMeasure = null,
bool normalizeNumericValues = false)
{
_resultHandler = resultHandlingFunc;
DistanceMeasure = distanceMeasure;
SimilarityMeasure = similarityMeasure ?? distanceMeasure;
DataNormalizer = dataNormalizer;
WeightingFunction = weightingFunc;
NormalizeNumericValues = normalizeNumericValues;
}
public DBScanClusterer(double epsilon, int minPts, IDistanceMeasure <T> measure)
: base(measure)
{
if (epsilon < 0.0d)
{
throw new ArgumentOutOfRangeException("epsilon", epsilon, "Argument must be greather than 0.0");
}
if (minPts < 0)
{
throw new ArgumentOutOfRangeException("minPts", minPts, "Argument must be greather than 0.0");
}
this.Epsilon = epsilon;
this.MinPts = minPts;
}
public GaussianKernelEstimator(double kernelParameter, IDistanceMeasure distanceMeasure, Dataset database)
: base(distanceMeasure, database)
{
this.KernelParameter = kernelParameter;
}
public NearestClassClassifier(IDistanceMeasure distanceMeasure, Dataset database)
: base(distanceMeasure, database)
{
}
public NearestClassClassifier(IDistanceMeasure distanceMeasure, Dataset database, double[] weights)
: base(distanceMeasure, database, weights)
{
}
public NearestClassClassifier(IDistanceMeasure distanceMeasure, Dataset database, double similarityThreshold)
: base(distanceMeasure, database)
{
this.SimilarityThreshold = similarityThreshold;
}
protected Clusterer(IDistanceMeasure <T> measure) => this.measure = measure;
public KMeansClustering()
{
rand = new RandomNumberGenerator();
distanceMeasure = new EuclideanDistance();
}
public KNearestNeighbours(IDistanceMeasure distanceMeasure, Dataset database, double[] weights, bool useWeightedVote)
: base(distanceMeasure, database, weights)
{
this._useWeightedVote = useWeightedVote;
}
public AbstractLazyClassifier(IDistanceMeasure distanceMeasuer, Dataset database, double[][] weights)
{
this._distanceMeasure = distanceMeasuer;
this._database = database;
this._classBasedWeights = weights;
}
public KNearestNeighbours(int k, IDistanceMeasure distanceMeasure, Dataset database, bool useWeightedVote)
: base(distanceMeasure, database)
{
this._k = k;
this._useWeightedVote = useWeightedVote;
}
public void SetUp()
{
this.configService = new Mock <IConfigurationService>();
this.logger = new Mock <ILogger>();
this.distanceMeasure = new EuclideanDistance();
}
