def test_best_pos_search_recursion(self):
schedules = [[t1] * 100 + [t2] * 100 + [t3] * 100]
expectations = [(
0, 300, 120
) # 120 seems to be the result returned by the code and appears to be correct, but I'm not convinced...
]
for i, schedule in enumerate(schedules):
a = Agent(parameters={
"name": "1",
"skill": 0.95,
"task_time": 1,
"move_speed": 1
})
a.schedule = schedule
p_i = a.best_pos_search(tI, range(len(a.schedule) + 1))
p_j = a.best_pos_search(tJ, range(len(a.schedule) + 1))
p_k = a.best_pos_search(tK, range(len(a.schedule) + 1))
p_ijk = p_i[0], p_j[0], p_k[0]
#print(p_ijk)
assert p_ijk == expectations[i], "Expected {} but got {}".format(
expectations[i], p_ijk)
def main():
agents, tasks = generate_painter(X, Y, 3)
# We give back m tasks in a giveback phase, and each iteration we assign one task,
# so for m agents we should have a giveback probability < 1/m
auctioneer = Auctioneer(agents, tasks, 0.25) #1 / ( * len(agents)))
schedule = auctioneer.optimize_schedule()
Agent.print_schedule(X, Y, auctioneer, schedule)
def test_worst_task_idx(self):
a = Agent(parameters={
"name": "1",
"skill": 0.95,
"task_time": 1,
"move_speed": 1
})
a.schedule = [t1] * 10 + [tJ] + [t1] * 10
wt = a.worst_task_idx()
assert wt == 10, "Should have been {} but was {}".format(10, wt)
def main():
agents, tasks = generate_painter(X, Y, 3)
# nFeas = The number of feasible 1-moves
# Given a schedule, we can move tasks equal to the number of agents.
# Using the result by Osman, we set alpha = n * Nfeas, gamma = n
alpha = math.factorial(X * Y)
# The number of iterations is n
annealer = Annealer(agents, tasks, 2500, 100, 0.1, 1)
schedule = annealer.optimize_schedule()
Agent.print_schedule(X, Y, annealer, schedule)
def test_best_pos_search_base_case(self):
schedules = [[], [tX], [t1, t2, t3]]
expectations = [(0, 0, 0), (1, 1, 1), (0, 3, 1)]
for i, schedule in enumerate(schedules):
a = Agent(parameters={
"name": "1",
"skill": 0.95,
"task_time": 1,
"move_speed": 1
})
a.schedule = schedule
p_i = a.best_pos_search(tI, range(len(a.schedule) + 1))
p_j = a.best_pos_search(tJ, range(len(a.schedule) + 1))
p_k = a.best_pos_search(tK, range(len(a.schedule) + 1))
p_ijk = p_i[0], p_j[0], p_k[0]
assert p_ijk == expectations[i], "Expected {} but got {}".format(
expectations[i], p_ijk)
def generate_painter(X, Y, num_agents):
agents = []
skill_diff = 1.0 / (num_agents + 1)
time_diff = 1.0 / (2 * num_agents + 1)
for i in range(num_agents):
agents += [
Agent(
parameters={
"name": str(i),
"skill": 1.0 - (i * skill_diff),
"task_time": 1.0 - (i * time_diff),
"move_speed": (i + 1)
})
]
#print(agents[-1])
total = 0
tasks = []
half_diagonal = ((X**2 + Y**2)**0.5) / 2
for i in range(X):
for j in range(Y):
# p = (X^2 + Y^2) / 4 - euclidean distance from center to (i, j)
# = ( (i - (X/2))^2 + (j - (X/2)^2) ^ 0.5
center_distance = ((i - X / 2)**2 + (j - Y / 2)**2)**0.5
p = half_diagonal - center_distance
total += p
t = Task(
parameters={
"name": "({},{})".format(i, j),
"i": i,
"j": j,
"chosen_probability": p
})
tasks += [t]
for task in tasks:
task.parameters["chosen_probability"] /= total
#for i in range(X):
# for j in range(Y):
# print(round(tasks[i * Y + j].chosen_probability(), 2), end=" ")
# print("\n")
return agents, tasks
# print(generate_painter(5, 5, 3))