def test_kh_physical_diffusion():
search_space = search.SearchSpace(n_agents=5,
n_variables=2,
lower_bound=[0, 0],
upper_bound=[10, 10])
new_kh = kh.KH()
new_kh.compile(search_space)
new_physical = new_kh._physical_diffusion(1, 1, 1, 20)
assert new_physical.shape == (1, 1)
def test_kh_mutation():
search_space = search.SearchSpace(n_agents=5,
n_variables=2,
lower_bound=[0, 0],
upper_bound=[10, 10])
new_kh = kh.KH()
new_kh.compile(search_space)
mutation = new_kh._mutation(search_space.agents, 0)
assert mutation.position.shape == (2, 1)
def test_kh_crossover():
search_space = search.SearchSpace(n_agents=5,
n_variables=2,
lower_bound=[0, 0],
upper_bound=[10, 10])
new_kh = kh.KH()
new_kh.compile(search_space)
crossover = new_kh._crossover(search_space.agents, 0)
assert crossover.position.shape == (2, 1)
def test_kh_update():
def square(x):
return np.sum(x**2)
search_space = search.SearchSpace(n_agents=5,
n_variables=2,
lower_bound=[0, 0],
upper_bound=[10, 10])
new_kh = kh.KH()
new_kh.compile(search_space)
new_kh.update(search_space, square, 1, 10)
def test_kh_food_beta():
new_kh = kh.KH()
search_space = search.SearchSpace(n_agents=5,
n_iterations=20,
n_variables=2,
lower_bound=[0, 0],
upper_bound=[10, 10])
beta = new_kh._food_beta(search_space.agents[0], search_space.agents[-1],
search_space.agents[0], search_space.agents[0], 1)
assert beta.shape == (2, 1)
def test_kh_best_beta():
search_space = search.SearchSpace(n_agents=5,
n_variables=2,
lower_bound=[0, 0],
upper_bound=[10, 10])
new_kh = kh.KH()
new_kh.compile(search_space)
beta = new_kh._best_beta(search_space.agents[0], search_space.agents[-1],
search_space.agents[0])
assert beta.shape == (2, 1)
def test_kh_sensing_distance():
search_space = search.SearchSpace(n_agents=5,
n_variables=2,
lower_bound=[0, 0],
upper_bound=[10, 10])
new_kh = kh.KH()
new_kh.compile(search_space)
distance, eucl_distance = new_kh._sensing_distance(search_space.agents, 0)
assert distance >= 0
assert len(eucl_distance) >= 0
def test_kh_target_alpha():
search_space = search.SearchSpace(n_agents=5,
n_variables=2,
lower_bound=[0, 0],
upper_bound=[10, 10])
new_kh = kh.KH()
new_kh.compile(search_space)
alpha = new_kh._target_alpha(search_space.agents[0],
search_space.agents[-1],
search_space.agents[0], 1)
assert alpha.shape == (2, 1) or alpha == 0
def test_kh_get_neighbours():
new_kh = kh.KH()
search_space = search.SearchSpace(n_agents=5,
n_iterations=20,
n_variables=2,
lower_bound=[0, 0],
upper_bound=[10, 10])
distance, eucl_distance = new_kh._sensing_distance(search_space.agents, 0)
neighbours = new_kh._get_neighbours(search_space.agents, 0, distance,
eucl_distance)
assert len(neighbours) >= 0
def test_kh_food_location():
def square(x):
return np.sum(x**2)
search_space = search.SearchSpace(n_agents=5,
n_variables=2,
lower_bound=[0, 0],
upper_bound=[10, 10])
new_kh = kh.KH()
new_kh.compile(search_space)
food = new_kh._food_location(search_space.agents, square)
assert food.fit >= 0
def test_kh_foraging_motion():
search_space = search.SearchSpace(n_agents=5,
n_variables=2,
lower_bound=[0, 0],
upper_bound=[10, 10])
new_kh = kh.KH()
new_kh.compile(search_space)
foraging = np.zeros((5, 2, 1))
new_foraging = new_kh._foraging_motion(search_space.agents, 0, 1, 20,
search_space.agents[0], foraging)
assert new_foraging.shape == (5, 2, 1)
def test_kh_neighbour_motion():
search_space = search.SearchSpace(n_agents=5,
n_variables=2,
lower_bound=[0, 0],
upper_bound=[10, 10])
new_kh = kh.KH()
new_kh.compile(search_space)
motion = np.zeros((5, 2, 1))
new_motion = new_kh._neighbour_motion(search_space.agents, 0, 1, 20,
motion)
assert new_motion.shape == (5, 2, 1)
def test_kh_food_location():
def square(x):
return np.sum(x**2)
new_function = function.Function(pointer=square)
new_kh = kh.KH()
search_space = search.SearchSpace(n_agents=5,
n_iterations=20,
n_variables=2,
lower_bound=[0, 0],
upper_bound=[10, 10])
food = new_kh._food_location(search_space.agents, new_function)
assert food.fit >= 0
def test_kh_update_position():
search_space = search.SearchSpace(n_agents=5,
n_variables=2,
lower_bound=[0, 0],
upper_bound=[10, 10])
new_kh = kh.KH()
new_kh.compile(search_space)
motion = np.zeros((2, 1))
foraging = np.zeros((2, 1))
new_position = new_kh._update_position(search_space.agents, 0, 1, 20,
search_space.agents[0], motion,
foraging)
assert new_position.shape == (2, 1)
def test_kh_local_alpha():
search_space = search.SearchSpace(n_agents=5,
n_variables=2,
lower_bound=[0, 0],
upper_bound=[10, 10])
new_kh = kh.KH()
new_kh.compile(search_space)
distance, eucl_distance = new_kh._sensing_distance(search_space.agents, 0)
neighbours = new_kh._get_neighbours(search_space.agents, 0, distance,
eucl_distance)
alpha = new_kh._local_alpha(search_space.agents[0],
search_space.agents[-1],
search_space.agents[0], neighbours)
assert alpha.shape == (2, 1) or alpha == 0
def test_kh_compile():
search_space = search.SearchSpace(n_agents=5,
n_variables=2,
lower_bound=[0, 0],
upper_bound=[10, 10])
new_kh = kh.KH()
new_kh.compile(search_space)
try:
new_kh.motion = 1
except:
new_kh.motion = np.array([1])
assert new_kh.motion == np.array([1])
try:
new_kh.foraging = 1
except:
new_kh.foraging = np.array([1])
assert new_kh.foraging == np.array([1])
def test_kh_run():
def square(x):
return np.sum(x**2)
def hook(optimizer, space, function):
return
new_function = function.Function(pointer=square)
new_kh = kh.KH()
search_space = search.SearchSpace(n_agents=5,
n_iterations=20,
n_variables=2,
lower_bound=[0, 0],
upper_bound=[10, 10])
history = new_kh.run(search_space, new_function, pre_evaluation=hook)
assert len(history.agents) > 0
assert len(history.best_agent) > 0
best_fitness = history.best_agent[-1][1]
assert best_fitness <= constants.TEST_EPSILON, 'The algorithm kh failed to converge.'
def test_kh_params():
params = {
'N_max': 0.01,
'w_n': 0.42,
'NN': 5,
'V_f': 0.02,
'w_f': 0.38,
'D_max': 0.002,
'C_t': 0.5,
'Cr': 0.2,
'Mu': 0.05
}
new_kh = kh.KH(params=params)
assert new_kh.N_max == 0.01
assert new_kh.w_n == 0.42
assert new_kh.NN == 5
assert new_kh.V_f == 0.02
assert new_kh.w_f == 0.38
assert new_kh.D_max == 0.002
assert new_kh.C_t == 0.5
assert new_kh.Cr == 0.2
assert new_kh.Mu == 0.05
def test_kh_params():
params = {
"N_max": 0.01,
"w_n": 0.42,
"NN": 5,
"V_f": 0.02,
"w_f": 0.38,
"D_max": 0.002,
"C_t": 0.5,
"Cr": 0.2,
"Mu": 0.05,
}
new_kh = kh.KH(params=params)
assert new_kh.N_max == 0.01
assert new_kh.w_n == 0.42
assert new_kh.NN == 5
assert new_kh.V_f == 0.02
assert new_kh.w_f == 0.38
assert new_kh.D_max == 0.002
assert new_kh.C_t == 0.5
assert new_kh.Cr == 0.2
assert new_kh.Mu == 0.05
def test_kh_build():
new_kh = kh.KH()
assert new_kh.built == True
def test_kh_physical_diffusion():
new_kh = kh.KH()
new_physical = new_kh._physical_diffusion(1, 1, 1, 20)
assert new_physical.shape == (1, 1)
def test_kh_params_setter():
new_kh = kh.KH()
try:
new_kh.N_max = 'a'
except:
new_kh.N_max = 0.01
assert new_kh.N_max == 0.01
try:
new_kh.N_max = -1
except:
new_kh.N_max = 0.01
assert new_kh.N_max == 0.01
try:
new_kh.w_n = 'a'
except:
new_kh.w_n = 0.42
assert new_kh.w_n == 0.42
try:
new_kh.w_n = 1.01
except:
new_kh.w_n = 0.42
assert new_kh.w_n == 0.42
try:
new_kh.NN = 0.5
except:
new_kh.NN = 5
assert new_kh.NN == 5
try:
new_kh.NN = -1
except:
new_kh.NN = 5
assert new_kh.NN == 5
try:
new_kh.V_f = 'a'
except:
new_kh.V_f = 0.02
assert new_kh.V_f == 0.02
try:
new_kh.V_f = -1
except:
new_kh.V_f = 0.02
assert new_kh.V_f == 0.02
try:
new_kh.w_f = 'a'
except:
new_kh.w_f = 0.38
assert new_kh.w_f == 0.38
try:
new_kh.w_f = 1.01
except:
new_kh.w_f = 0.38
assert new_kh.w_f == 0.38
try:
new_kh.D_max = 'a'
except:
new_kh.D_max = 0.02
assert new_kh.D_max == 0.02
try:
new_kh.D_max = -1
except:
new_kh.D_max = 0.02
assert new_kh.D_max == 0.02
try:
new_kh.C_t = 'a'
except:
new_kh.C_t = 0.5
assert new_kh.C_t == 0.5
try:
new_kh.C_t = 2.01
except:
new_kh.C_t = 0.5
assert new_kh.C_t == 0.5
try:
new_kh.Cr = 'a'
except:
new_kh.Cr = 0.2
assert new_kh.Cr == 0.2
try:
new_kh.Cr = 1.1
except:
new_kh.Cr = 0.2
assert new_kh.Cr == 0.2
try:
new_kh.Mu = 'a'
except:
new_kh.Mu = 0.05
assert new_kh.Mu == 0.05
try:
new_kh.Mu = 1.1
except:
new_kh.Mu = 0.05
assert new_kh.Mu == 0.05
