void irAddToDecisionTree(Imm_Reward_List node)
{
int i, j, k;
custom_Matrix m;
assert( node != NULL );
/* ensure the decision tree is initialized (ok to call dtInit() more than once) */
dtInit(gNumActions, gNumStates, gNumObservations);
switch( node->type ) {
case ir_value:
if ( gProblemType == POMDP_problem_type ) { /* pomdp */
dtAdd(node->action, node->cur_state, node->next_state, node->obs, node->rep.value);
} else { /* mdp */
dtAdd(node->action, node->cur_state, node->next_state, WILDCARD_SPEC, node->rep.value);
}
break;
case ir_vector:
if ( gProblemType == POMDP_problem_type ) { /* pomdp */
for (i=0; i < gNumObservations; i++) {
dtAdd(node->action, node->cur_state, node->next_state, i, node->rep.vector[i]);
}
} else { /* mdp */
for (i=0; i < gNumStates; i++) {
dtAdd(node->action, node->cur_state, i, WILDCARD_SPEC, node->rep.vector[i]);
}
}
break;
case ir_matrix:
m = node->rep.matrix;
for (i=0; i < m->num_rows; i++) {
for (j=0; j < m->row_length[i]; j++) {
k = m->row_start[i] + j;
if( gProblemType == POMDP_problem_type ) { /* pomdp */
dtAdd(node->action, node->cur_state, i, m->col[k], m->mat_val[k]);
} else { /* mdp */
dtAdd(node->action, i, m->col[k], WILDCARD_SPEC, m->mat_val[k]);
}
}
}
break;
default:
assert(0 /* never reach this point */);
} /* switch */
}
void testOnce(void)
{
double result;
/* set up table */
dtInit(5, 5, 5);
dtAdd(0, 1, 2, 3, 0.5);
dtAdd(0, 1, 2, 3, 0.7);
dtAdd(0, 1, 3, -1, -10.1);
dtAdd(0, 1, 3, -1, -10.3);
dtAdd(1, 0, 0, -1, 6.0);
dtAdd(1, 0, -1, -1, 7.0);
dtAdd(-1, 2, -1, -1, 2.7);
/* do a few queries */
result = dtGet(0, 1, 2, 3);
printf("expecting: result=%lf\n", 0.7);
printf("got: result=%lf\n", result);
result = dtGet(0, 1, 3, 1);
printf("expecting: result=%lf\n", -10.3);
printf("got: result=%lf\n", result);
result = dtGet(1, 0, 0, 0);
printf("expecting: result=%lf\n", 7.0);
printf("got: result=%lf\n", result);
result = dtGet(3, 2, 0, 0);
printf("expecting: result=%lf\n", 2.7);
printf("got: result=%lf\n", result);
/* clean up */
dtDeallocate();
}
