/// <summary>
/// This function will initalise the cluster list to index number of locations.
/// </summary>
/// <param name="index">the number of elements to initalise</param>
private void InitaliseClusters(int index)
{
int increment = (index - Clusters.Count) + 1;
for (int i = 0; i < increment; i++)
{
Clusters.Insert(Clusters.Count, new EventCollection());
}
}
public static void ClusterGeneration()
{
Key k = new Key();
//generate the clusters
for (int i = 0; i < grid_num; i++)
{
List <Cell> collection = new List <Cell>();
Cell c = Cell_No[i];
if (c.visited == 0 && c.added == 0)
{
c.added = 1;
Cell_No[i] = c;
collection.Add(c);
ClusterFunctions.Recursive_neighbour(c, collection, Cell_No);
Clusters.Insert(cluster_number, collection);
cluster_number++;
collection = new List <Cell>();
}
}
k.cluster_num = cluster_number;
int[] k_cluster_size = new int[cluster_number];
int[] k_surface_area = new int[cluster_number];
float[] k_compactness = new float[cluster_number];
for (int i = 0; i < cluster_number; i++)
{
k_cluster_size[i] = Clusters[i].Count;
}
k.cluster_size_min = ClusterFunctions.Get_Min(k_cluster_size);
k.cluster_size_max = ClusterFunctions.Get_Max(k_cluster_size);
k.cluster_size_median = ClusterFunctions.Get_Median(k_cluster_size);
//calculate the surface area (how many grid sides around the surface)
surface_area = new int[cluster_number];
for (int i = 0; i < cluster_number; i++)
{
List <Cell> Collection = Clusters[i];
List <Cell> Dup_Collection = new List <Cell>(Collection);
ClusterFunctions.Calculate_Surface_Area(Dup_Collection, Cell_No, i);
}
//print the info
for (int j = 0; j < surface_area.Length; j++)
{
//Console.WriteLine("cluster " + j + " surface area: " + surface_area[j]);
k_surface_area[j] = surface_area[j];
}
k.surface_area_min = ClusterFunctions.Get_Min(k_surface_area);
k.surface_area_max = ClusterFunctions.Get_Max(k_surface_area);
k.surface_area_median = ClusterFunctions.Get_Median(k_surface_area);
//decide the compactness of clusters
compactness = new float[cluster_number];
for (int i = 0; i < cluster_number; i++)
{
List <Cell> collection = Clusters[i];
ClusterFunctions.Compactness(collection, grid, i);
}
//print the info
for (int j = 0; j < compactness.Length; j++)
{
//Console.WriteLine("cluster "+j+" compact: "+compactness[j]);
k_compactness[j] = compactness[j];
}
k.compactness_min = (int)Math.Round(ClusterFunctions.Get_Min(k_compactness) * 1000);
//Console.WriteLine("compact min: " + k.compactness_min);
k.compactness_max = (int)Math.Round(ClusterFunctions.Get_Max(k_compactness) * 1000);
if (k.compactness_max != 1)
{
//Console.WriteLine("compact max: " + k.compactness_max);
}
k.compactness_median = (int)Math.Round(ClusterFunctions.Get_Median(k_compactness) * 1000);
//Console.WriteLine("compact median: " + k.compactness_max);
//decide the sides of each cluster
for (int i = 0; i < cluster_number; i++)
{
List <Cell> collection = Clusters[i];
ClusterFunctions.SideCheck(collection, Cell_No);
}
k.path_length = path_vec.Count;
keyTable.Add(path_vec, k);
if (!Table.ContainsKey(k))
{
Table.Add(k, 1);
}
else
{
Table[k]++;
}
}
