public void GetTiles2(int [] tiles, int nt, collision_table ct, float f1, float f2, int h1)
{
f_tmp_arr[0] = f1;
f_tmp_arr[1] = f2;
i_tmp_arr[0] = h1;
GetTiles(tiles, nt, ct, f_tmp_arr, 2, i_tmp_arr, 1);
}
public void GetTiles(int [] tiles, int nt, collision_table ct, float [] floats, int nf, int h1, int h2, int h3)
{
i_tmp_arr[0] = h1;
i_tmp_arr[1] = h2;
i_tmp_arr[2] = h3;
GetTiles(tiles, nt, ct, floats, nf, i_tmp_arr, 3);
}
public void GetTiles1(int [] tiles, int nt, collision_table ct, float f1, int h1, int h2)
{
f_tmp_arr[0] = f1;
i_tmp_arr[0] = h1;
i_tmp_arr[1] = h2;
GetTiles(tiles, nt, ct, f_tmp_arr, 1, i_tmp_arr, 2);
}
public static void SaveFile(collision_table ct, string filename)
{
IFormatter bf = new BinaryFormatter();
Stream stream = new FileStream(filename, FileMode.Create, FileAccess.Write, FileShare.None);
bf.Serialize(stream, ct);
stream.Close();
Console.WriteLine("Collision data saves");
}
public void GetTilesWrap(
int [] tiles, // provided array contains returned tiles (tile indices)
int num_tilings, // number of tile indices to be returned in tiles
collision_table ctable, // total number of possible tiles
float [] floats, // array of floating point variables
int num_floats, // number of floating point variables
int [] wrap_widths, // array of widths (length and units as in floats)
int [] ints, // array of integer variables
int num_ints) // number of integer variables
{
int i, j;
int [] qstate = new int[max_num_vars];
int [] baseDim = new int[max_num_vars];
int [] wrap_widths_times_num_tilings = new int[max_num_vars];
int [] coordinates = new int[max_num_vars * 2 + 1]; /* one interval number per relevant dimension */
int num_coordinates = num_floats + num_ints + 1;
for (i = 0; i < num_ints; i++)
{
coordinates[num_floats + 1 + i] = ints[i];
}
/* quantize state to integers (henceforth, tile widths == num_tilings) */
for (i = 0; i < num_floats; i++)
{
qstate[i] = (int)Math.Floor(floats[i] * num_tilings);
baseDim[i] = 0;
wrap_widths_times_num_tilings[i] = wrap_widths[i] * num_tilings;
}
/*compute the tile numbers */
for (j = 0; j < num_tilings; j++)
{
/* loop over each relevant dimension */
for (i = 0; i < num_floats; i++)
{
/* find coordinates of activated tile in tiling space */
coordinates[i] = qstate[i] - mod(qstate[i] - baseDim[i], num_tilings);
if (wrap_widths[i] != 0)
{
coordinates[i] = mod(coordinates[i], wrap_widths_times_num_tilings[i]);
}
/* compute displacement of next tiling in quantized space */
baseDim[i] += 1 + (2 * i);
}
/* add additional indices for tiling and hashing_set so they hash differently */
coordinates[i] = j;
tiles[j] = hash(coordinates, num_coordinates, ctable);
}
return;
}
/* hash
* Takes an array of integers and returns the corresponding tile after hashing
*/
int hash(int [] ints, int num_ints, collision_table ct)
{
int j;
int ccheck;
ct.calls++;
j = hash_UNH(ints, num_ints, ct.m, 449);
ccheck = hash_UNH(ints, num_ints, max_longInt, 457);
if (ccheck == (int)ct.data[j])
{
ct.clearhits++;
}
else if (ct.data[j] == -1)
{
ct.clearhits++;
ct.data[j] = ccheck;
}
else if (ct.safe == 0)
{
ct.collisions++;
}
else
{
int h2 = 1 + 2 * hash_UNH(ints, num_ints, (max_longInt) / 4, 449);
int i = 0;
while (++i > 0)
{
ct.collisions++;
j = (j + h2) % (ct.m);
//printf("(%d)",j);
if (i > ct.m)
{
Console.WriteLine("\nOut of Memory");
Environment.Exit(0);
}
if (ccheck == ct.data[j])
{
break;
}
if (ct.data[j] == -1)
{
ct.data[j] = ccheck;
break;
}
}
}
return(j);
}
public static collision_table LoadFile(string fileToLoad)
{
if (File.Exists(fileToLoad))
{
IFormatter bf = new BinaryFormatter();
Stream stream = new FileStream(fileToLoad, FileMode.Open, FileAccess.Read, FileShare.Read);
collision_table ct = (collision_table)bf.Deserialize(stream);
stream.Close();
Console.WriteLine("Collision data saves");
Console.WriteLine("example data retrieved is {0}", ct.m);
return(ct);
}
else
{
return(null);
}
}
public void GetTiles1(int [] tiles, int nt, collision_table ct, float f1)
{
f_tmp_arr[0] = f1;
GetTiles(tiles, nt, ct, f_tmp_arr, 1, i_tmp_arr, 0);
}
public void GetTiles(int [] tiles, int nt, collision_table ct, float [] floats, int nf, int h1)
{
i_tmp_arr[0] = h1;
GetTiles(tiles, nt, ct, floats, nf, i_tmp_arr, 1);
}
public void GetTiles(int [] tiles, int nt, collision_table ct, float [] floats, int nf)
{
GetTiles(tiles, nt, ct, floats, nf, i_tmp_arr, 0);
}
public Q_LearningAgent(int numFeatures, int numActions, bool bLearn, double rewards,
double [] widths,
string loadWeightsFile, string saveWeightsFile) : base(numFeatures, numActions)
{
this.numActions = numActions;
this.numFeatures = numFeatures;
numTiles = numFeatures * numDimTiles;
tileWidths = new double[numFeatures];
Q = new double[numActions];
//weights = new double[numActions,numTiles];
//weightsRaw=new double[numActions,numTiles];
//weights = weightsRaw;
traces = new double[numActions, numTiles];
traceAll = new double[numTiles];
// number of tiles in each tiling
tile = new int[numFeatures, numDimTiles];
actionTiles = new int[numActions, rl_memory_Size];
tiless = new Tile[numFeatures, numDimTiles];
nonzeroTraces = new int[rl_max_nonzero_Traces];
nonzeroTracesInverse = new int[rl_memory_Size];
currentState = new int[numActions, numDimTiles];
previousState = new int[numActions, numDimTiles];
weightsFile = saveWeightsFile;
bLearning = bLearn;
lastReward = rewards;
for (int i = 0; i < getNumFeatures(); i++)
{
tileWidths[i] = widths[i];
}
config = RLConfig.Load("RLConfig.xml");
alpha = config.alpha;
gamma = config.gamma;
lambda = config.lambda;
epsilon = config.epsilon;
minimumTrace = config.traceability;
//for (int i=0; i < rl_memory_Size;i++)
for (int j = 0; j < numActions; j++) //yet to do ... each Q(s,a) should represent
{
Q[j] = 0;
}
epochNum = 0;
lastAction = -1;
numNonzeroTraces = 0;
//Sabre changed this to initial weights for all tiles per action
//need to check if this yields to weights being always zero when ever you call a SarsaAgent object
//if it is remove this initialisation and put it in the start-episode method
if (config.transfer == 1)
{
weights = loadWeights(loadWeightsFile);
//load traces too
//at the moment initial the traces but need to explore saving them too
for (int j = 0; j < numActions; j++)
{
for (int i = 0; i < numTiles; i++)
{
traces[j, i] = 0;
}
}
}
else
{
for (int j = 0; j < numActions; j++)
{
for (int i = 0; i < numTiles; i++)
{
weights[j, i] = 0;
traces[j, i] = 0;
}
}
}
int [] tmp = new int[2];
float [] tmpf = new float[2];
colTab = new collision_table(rl_memory_Size, 1);
tiles obj_tiles = new tiles();
}