def test_population_controller_mutate():
from aw_nas.controller import EvoController, ParetoEvoController
device = "cuda"
for ss, rollout_type in zip(("cnn", "ofa"), ("discrete", "ofa")):
search_space = get_search_space(cls=ss)
controller = EvoController(search_space,
device,
rollout_type=rollout_type,
mode="train",
population_size=1,
parent_pool_size=1,
mutate_kwargs={},
avoid_mutate_repeat=True,
avoid_mutate_repeat_worst_threshold=1,
eval_sample_strategy="population",
elimination_strategy="regularized")
controller.mode = "train"
rollouts = controller.sample(1)
for rollout in rollouts:
rollout.set_perf(np.random.random(), "reward")
controller.step(rollouts)
new_rollout = controller.sample(1)[0]
assert str(rollouts[0].genotype) != str(new_rollout.genotype)
controller = ParetoEvoController(search_space,
device,
rollout_type=rollout_type,
mode="train",
init_population_size=1,
perf_names=["reward"],
eval_sample_strategy="all")
controller.mode = "train"
rollouts = controller.sample(1)
for rollout in rollouts:
rollout.set_perf(np.random.random(), "reward")
controller.step(rollouts)
new_rollout = controller.sample(1)[0]
assert str(rollouts[0].genotype) != str(new_rollout.genotype)
def test_population_controller_avoid_repeat():
from aw_nas.controller import EvoController
ss_cfgs = {
"cell_layout": [0, 1, 0, 1, 0],
"num_layers": 5,
"shared_primitives": ["skip_connect", "sep_conv_3x3"],
"num_init_nodes": 1,
"num_steps": 3,
"num_cell_groups": 2,
"reduce_cell_groups": [1],
}
ss = get_search_space("cnn", **ss_cfgs)
controller = EvoController(ss,
"cuda",
rollout_type="discrete",
avoid_mutate_repeat=True,
avoid_mutate_repeat_worst_threshold=3,
avoid_repeat_fallback="raise",
population_size=100,
parent_pool_size=1)
rollout = controller.sample(n=1)[0]
rollout.set_perf(1.0)
controller.set_mode("train")
# make it the highest reward
rollouts = []
for i in range(99):
rollouts.append(ss.mutate(rollout).set_perf(0.3))
controller.step([rollout] + rollouts)
controller.population_size = len(controller.population)
with pytest.raises(Exception):
for _ in range(3):
controller.sample(n=1)
controller.avoid_repeat_fallback = "return" # let's fallback to return, not rais
for _ in range(3):
controller.sample(n=1)