def test_scenario_1_s(self):
# Smaller version of the corridor collisions scenario above
# to facilitate DRL training
scenario_1_mdp = OvercookedGridworld.from_layout_name(
'scenario1_s', start_order_list=['any'], cook_time=5)
mlp = MediumLevelPlanner.from_pickle_or_compute(
scenario_1_mdp, NO_COUNTERS_PARAMS, force_compute=force_compute)
a0 = GreedyHumanModel(mlp)
a1 = CoupledPlanningAgent(mlp)
agent_pair = AgentPair(a0, a1)
start_state = OvercookedState(
[P((2, 1), s, Obj('onion', (2, 1))),
P((4, 2), s)], {},
order_list=['onion'])
env = OvercookedEnv(scenario_1_mdp, start_state_fn=lambda: start_state)
trajectory, time_taken_hr, _, _ = env.run_agents(
agent_pair, include_final_state=True, display=DISPLAY)
env.reset()
print("\n" * 5)
print("-" * 50)
a0 = CoupledPlanningAgent(mlp)
a1 = CoupledPlanningAgent(mlp)
agent_pair = AgentPair(a0, a1)
trajectory, time_taken_rr, _, _ = env.run_agents(
agent_pair, include_final_state=True, display=DISPLAY)
print("H+R time taken: ", time_taken_hr)
print("R+R time taken: ", time_taken_rr)
self.assertGreater(time_taken_hr, time_taken_rr)
def test_one_player_env(self):
mdp = OvercookedGridworld.from_layout_name("cramped_room_single")
env = OvercookedEnv(mdp, horizon=12)
a0 = FixedPlanAgent([stay, w, w, e, e, n, e, interact, w, n, interact])
ag = AgentGroup(a0)
env.run_agents(ag, display=False)
self.assertEqual(env.state.players_pos_and_or, (((2, 1), (0, -1)), ))
def test_scenario_1(self):
# Myopic corridor collision
#
# X X X X X O X D X X X X X
# X ↓Ho X X
# X X X X X X X X ↓R X
# X X
# X S X X X X X X X X P P X
#
# H on left with onion, further away to the tunnel entrance than R.
# Optimal planner tells R to go first and that H will wait
# for R to pass. H however, starts going through the tunnel
# and they get stuck. The H plan is a bit extreme (it would probably
# realize that it should retrace it's steps at some point)
scenario_1_mdp = OvercookedGridworld.from_layout_name(
'small_corridor', start_order_list=['any'], cook_time=5)
mlp = MediumLevelPlanner.from_pickle_or_compute(
scenario_1_mdp, NO_COUNTERS_PARAMS, force_compute=force_compute)
a0 = GreedyHumanModel(mlp)
a1 = CoupledPlanningAgent(mlp)
agent_pair = AgentPair(a0, a1)
start_state = OvercookedState(
[P((2, 1), s, Obj('onion', (2, 1))),
P((10, 2), s)], {},
order_list=['onion'])
env = OvercookedEnv(scenario_1_mdp, start_state_fn=lambda: start_state)
env.run_agents(agent_pair, include_final_state=True, display=DISPLAY)
def test_four_player_env_fixed(self):
mdp = OvercookedGridworld.from_layout_name("multiplayer_schelling")
assert mdp.num_players == 4
env = OvercookedEnv(mdp, horizon=16)
a0 = FixedPlanAgent([stay, w, w])
a1 = FixedPlanAgent([
stay, stay, e, e, n, n, n, e, interact, n, n, w, w, w, n, interact,
e
])
a2 = FixedPlanAgent(
[stay, w, interact, n, n, e, e, e, n, e, n, interact, w])
a3 = FixedPlanAgent([e, interact, n, n, w, w, w, n, interact, e, s])
ag = AgentGroup(a0, a1, a2, a3)
env.run_agents(ag, display=False)
self.assertEqual(env.state.players_pos_and_or,
(((1, 1), (-1, 0)), ((3, 1), (0, -1)),
((2, 1), (-1, 0)), ((4, 2), (0, 1))))
def test_one_coupled_one_fixed(self):
a0 = CoupledPlanningAgent(self.mlp_large)
a1 = FixedPlanAgent([s, e, n, w])
agent_pair = AgentPair(a0, a1)
env = OvercookedEnv(large_mdp, horizon=10)
trajectory, time_taken, _, _ = env.run_agents(agent_pair,
include_final_state=True,
display=DISPLAY)
self.assertEqual(time_taken, 10)
def test_fixed_plan_agents(self):
a0 = FixedPlanAgent([s, e, n, w])
a1 = FixedPlanAgent([s, w, n, e])
agent_pair = AgentPair(a0, a1)
env = OvercookedEnv(large_mdp, horizon=10)
trajectory, time_taken, _, _ = env.run_agents(agent_pair,
include_final_state=True,
display=DISPLAY)
end_state = trajectory[-1][0]
self.assertEqual(time_taken, 10)
self.assertEqual(env.mdp.get_standard_start_state().player_positions,
end_state.player_positions)
def test_two_coupled_agents(self):
a0 = CoupledPlanningAgent(self.mlp_large)
a1 = CoupledPlanningAgent(self.mlp_large)
agent_pair = AgentPair(a0, a1)
start_state = OvercookedState([P(
(2, 2), n), P((2, 1), n)], {},
order_list=['any'])
env = OvercookedEnv(large_mdp, start_state_fn=lambda: start_state)
trajectory, time_taken, _, _ = env.run_agents(agent_pair,
include_final_state=True,
display=DISPLAY)
end_state = trajectory[-1][0]
self.assertEqual(end_state.order_list, [])
def test_two_coupled_agents_coupled_pair(self):
mlp_simple = MediumLevelPlanner.from_pickle_or_compute(
simple_mdp, NO_COUNTERS_PARAMS, force_compute=force_compute)
cp_agent = CoupledPlanningAgent(mlp_simple)
agent_pair = CoupledPlanningPair(cp_agent)
start_state = OvercookedState([P(
(2, 2), n), P((2, 1), n)], {},
order_list=['any'])
env = OvercookedEnv(simple_mdp, start_state_fn=lambda: start_state)
trajectory, time_taken, _, _ = env.run_agents(agent_pair,
include_final_state=True,
display=DISPLAY)
end_state = trajectory[-1][0]
self.assertEqual(end_state.order_list, [])
def test_one_coupled_one_greedy_human(self):
# Even though in the first ~10 timesteps it seems like agent 1 is wasting time
# it turns out that this is actually not suboptimal as the true bottleneck is
# going to be agent 0 later on (when it goes to get the 3rd onion)
a0 = GreedyHumanModel(self.mlp_large)
a1 = CoupledPlanningAgent(self.mlp_large)
agent_pair = AgentPair(a0, a1)
start_state = OvercookedState([P(
(2, 1), s), P((1, 1), s)], {},
order_list=['onion'])
env = OvercookedEnv(large_mdp, start_state_fn=lambda: start_state)
trajectory, time_taken, _, _ = env.run_agents(agent_pair,
include_final_state=True,
display=DISPLAY)
end_state = trajectory[-1][0]
self.assertEqual(end_state.order_list, [])
def test_two_greedy_human_open_map(self):
scenario_2_mdp = OvercookedGridworld.from_layout_name(
'scenario2', start_order_list=['any'], cook_time=5)
mlp = MediumLevelPlanner.from_pickle_or_compute(
scenario_2_mdp, NO_COUNTERS_PARAMS, force_compute=force_compute)
a0 = GreedyHumanModel(mlp)
a1 = GreedyHumanModel(mlp)
agent_pair = AgentPair(a0, a1)
start_state = OvercookedState([P(
(8, 1), s), P((1, 1), s)], {},
order_list=['onion'])
env = OvercookedEnv(scenario_2_mdp,
start_state_fn=lambda: start_state,
horizon=100)
trajectory, time_taken, _, _ = env.run_agents(agent_pair,
include_final_state=True,
display=DISPLAY)
end_state = trajectory[-1][0]
self.assertEqual(len(end_state.order_list), 0)
class TestOvercookedEnvironment(unittest.TestCase):
def setUp(self):
self.base_mdp = OvercookedGridworld.from_layout_name("cramped_room")
self.env = OvercookedEnv(self.base_mdp, **DEFAULT_ENV_PARAMS)
self.rnd_agent_pair = AgentPair(FixedPlanAgent([stay, w, w]),
FixedPlanAgent([stay, e, e]))
np.random.seed(0)
def test_constructor(self):
try:
OvercookedEnv(self.base_mdp, horizon=10)
except Exception as e:
self.fail("Failed to instantiate OvercookedEnv:\n{}".format(e))
with self.assertRaises(TypeError):
OvercookedEnv(self.base_mdp, **{"invalid_env_param": None})
def test_step_fn(self):
for _ in range(10):
joint_action = random_joint_action()
self.env.step(joint_action)
def test_execute_plan(self):
action_plan = [random_joint_action() for _ in range(10)]
self.env.execute_plan(self.base_mdp.get_standard_start_state(),
action_plan)
def test_run_agents(self):
start_state = self.env.state
self.env.run_agents(self.rnd_agent_pair)
self.assertNotEqual(self.env.state, start_state)
def test_rollouts(self):
try:
self.env.get_rollouts(self.rnd_agent_pair, 3)
except Exception as e:
self.fail("Failed to get rollouts from environment:\n{}".format(e))
def test_one_player_env(self):
mdp = OvercookedGridworld.from_layout_name("cramped_room_single")
env = OvercookedEnv(mdp, horizon=12)
a0 = FixedPlanAgent([stay, w, w, e, e, n, e, interact, w, n, interact])
ag = AgentGroup(a0)
env.run_agents(ag, display=False)
self.assertEqual(env.state.players_pos_and_or, (((2, 1), (0, -1)), ))
def test_four_player_env_fixed(self):
mdp = OvercookedGridworld.from_layout_name("multiplayer_schelling")
assert mdp.num_players == 4
env = OvercookedEnv(mdp, horizon=16)
a0 = FixedPlanAgent([stay, w, w])
a1 = FixedPlanAgent([
stay, stay, e, e, n, n, n, e, interact, n, n, w, w, w, n, interact,
e
])
a2 = FixedPlanAgent(
[stay, w, interact, n, n, e, e, e, n, e, n, interact, w])
a3 = FixedPlanAgent([e, interact, n, n, w, w, w, n, interact, e, s])
ag = AgentGroup(a0, a1, a2, a3)
env.run_agents(ag, display=False)
self.assertEqual(env.state.players_pos_and_or,
(((1, 1), (-1, 0)), ((3, 1), (0, -1)),
((2, 1), (-1, 0)), ((4, 2), (0, 1))))
def test_multiple_mdp_env(self):
mdp0 = OvercookedGridworld.from_layout_name("cramped_room")
mdp1 = OvercookedGridworld.from_layout_name("counter_circuit")
mdp_fn = lambda: np.random.choice([mdp0, mdp1])
# Default env
env = OvercookedEnv(mdp_fn, horizon=100)
env.get_rollouts(self.rnd_agent_pair, 5)
def test_starting_position_randomization(self):
self.base_mdp = OvercookedGridworld.from_layout_name("cramped_room")
start_state_fn = self.base_mdp.get_random_start_state_fn(
random_start_pos=True, rnd_obj_prob_thresh=0.0)
env = OvercookedEnv(self.base_mdp, start_state_fn)
start_state = env.state.players_pos_and_or
for _ in range(3):
env.reset()
print(env)
curr_terrain = env.state.players_pos_and_or
self.assertFalse(np.array_equal(start_state, curr_terrain))
def test_starting_obj_randomization(self):
self.base_mdp = OvercookedGridworld.from_layout_name("cramped_room")
start_state_fn = self.base_mdp.get_random_start_state_fn(
random_start_pos=False, rnd_obj_prob_thresh=0.8)
env = OvercookedEnv(self.base_mdp, start_state_fn)
start_state = env.state.all_objects_list
for _ in range(3):
env.reset()
print(env)
curr_terrain = env.state.all_objects_list
self.assertFalse(np.array_equal(start_state, curr_terrain))
def test_failing_rnd_layout(self):
with self.assertRaises(TypeError):
mdp_gen_params = {"None": None}
mdp_fn = LayoutGenerator.mdp_gen_fn_from_dict(**mdp_gen_params)
OvercookedEnv(mdp=mdp_fn, **DEFAULT_ENV_PARAMS)
def test_random_layout(self):
mdp_gen_params = {"prop_feats": (1, 1)}
mdp_fn = LayoutGenerator.mdp_gen_fn_from_dict(**mdp_gen_params)
env = OvercookedEnv(mdp=mdp_fn, **DEFAULT_ENV_PARAMS)
start_terrain = env.mdp.terrain_mtx
for _ in range(3):
env.reset()
print(env)
curr_terrain = env.mdp.terrain_mtx
self.assertFalse(np.array_equal(start_terrain, curr_terrain))
mdp_gen_params = {
"mdp_choices": ['cramped_room', 'asymmetric_advantages']
}
mdp_fn = LayoutGenerator.mdp_gen_fn_from_dict(**mdp_gen_params)
env = OvercookedEnv(mdp=mdp_fn, **DEFAULT_ENV_PARAMS)
layouts_seen = []
for _ in range(10):
layouts_seen.append(env.mdp.terrain_mtx)
env.reset()
all_same_layout = all([
np.array_equal(env.mdp.terrain_mtx, terrain)
for terrain in layouts_seen
])
self.assertFalse(all_same_layout)
def learn(*,
network,
env,
total_timesteps,
early_stopping=False,
eval_env=None,
seed=None,
nsteps=2048,
ent_coef=0.0,
lr=3e-4,
vf_coef=0.5,
max_grad_norm=0.5,
gamma=0.99,
lam=0.95,
log_interval=10,
nminibatches=4,
noptepochs=4,
cliprange=0.2,
save_interval=0,
load_path=None,
model_fn=None,
scope='',
**network_kwargs):
'''
Learn policy using PPO algorithm (https://arxiv.org/abs/1707.06347)
Parameters:
----------
network: policy network architecture. Either string (mlp, lstm, lnlstm, cnn_lstm, cnn, cnn_small, conv_only - see baselines.common/models.py for full list)
specifying the standard network architecture, or a function that takes tensorflow tensor as input and returns
tuple (output_tensor, extra_feed) where output tensor is the last network layer output, extra_feed is None for feed-forward
neural nets, and extra_feed is a dictionary describing how to feed state into the network for recurrent neural nets.
See common/models.py/lstm for more details on using recurrent nets in policies
env: baselines.common.vec_env.VecEnv environment. Needs to be vectorized for parallel environment simulation.
The environments produced by gym.make can be wrapped using baselines.common.vec_env.DummyVecEnv class.
nsteps: int number of steps of the vectorized environment per update (i.e. batch size is nsteps * nenv where
nenv is number of environment copies simulated in parallel)
total_timesteps: int number of timesteps (i.e. number of actions taken in the environment)
ent_coef: float policy entropy coefficient in the optimization objective
lr: float or function learning rate, constant or a schedule function [0,1] -> R+ where 1 is beginning of the
training and 0 is the end of the training.
vf_coef: float value function loss coefficient in the optimization objective
max_grad_norm: float or None gradient norm clipping coefficient
gamma: float discounting factor
lam: float advantage estimation discounting factor (lambda in the paper)
log_interval: int number of timesteps between logging events
nminibatches: int number of training minibatches per update. For recurrent policies,
should be smaller or equal than number of environments run in parallel.
noptepochs: int number of training epochs per update
cliprange: float or function clipping range, constant or schedule function [0,1] -> R+ where 1 is beginning of the training
and 0 is the end of the training
save_interval: int number of timesteps between saving events
load_path: str path to load the model from
**network_kwargs: keyword arguments to the policy / network builder. See baselines.common/policies.py/build_policy and arguments to a particular type of network
For instance, 'mlp' network architecture has arguments num_hidden and num_layers.
'''
additional_params = network_kwargs["network_kwargs"]
from baselines import logger
# set_global_seeds(seed) We deal with seeds upstream
if "LR_ANNEALING" in additional_params.keys():
lr_reduction_factor = additional_params["LR_ANNEALING"]
start_lr = lr
lr = lambda prop: (start_lr / lr_reduction_factor) + (
start_lr - (start_lr / lr_reduction_factor
)) * prop # Anneals linearly from lr to lr/red factor
if isinstance(lr, float): lr = constfn(lr)
else: assert callable(lr)
if isinstance(cliprange, float): cliprange = constfn(cliprange)
else: assert callable(cliprange)
total_timesteps = int(total_timesteps)
policy = build_policy(env, network, **network_kwargs)
bestrew = 0
# Get the nb of env
nenvs = env.num_envs
# Get state_space and action_space
ob_space = env.observation_space
ac_space = env.action_space
# Calculate the batch_size
nbatch = nenvs * nsteps
nbatch_train = nbatch // nminibatches
# Instantiate the model object (that creates act_model and train_model)
if model_fn is None:
from baselines.ppo2.model import Model
model_fn = Model
model = model_fn(policy=policy,
ob_space=ob_space,
ac_space=ac_space,
nbatch_act=nenvs,
nbatch_train=nbatch_train,
nsteps=nsteps,
ent_coef=ent_coef,
vf_coef=vf_coef,
max_grad_norm=max_grad_norm,
scope=scope)
if load_path is not None:
model.load(load_path)
# Instantiate the runner object
runner = Runner(env=env, model=model, nsteps=nsteps, gamma=gamma, lam=lam)
if eval_env is not None:
eval_runner = Runner(env=eval_env,
model=model,
nsteps=nsteps,
gamma=gamma,
lam=lam)
epinfobuf = deque(maxlen=100)
if eval_env is not None:
eval_epinfobuf = deque(maxlen=100)
# Start total timer
tfirststart = time.perf_counter()
best_rew_per_step = 0
run_info = defaultdict(list)
nupdates = total_timesteps // nbatch
print("TOT NUM UPDATES", nupdates)
for update in range(1, nupdates + 1):
assert nbatch % nminibatches == 0, "Have {} total batch size and want {} minibatches, can't split evenly".format(
nbatch, nminibatches)
# Start timer
tstart = time.perf_counter()
frac = 1.0 - (update - 1.0) / nupdates
# Calculate the learning rate
lrnow = lr(frac)
# Calculate the cliprange
cliprangenow = cliprange(frac)
# Get minibatch
obs, returns, masks, actions, values, neglogpacs, states, epinfos = runner.run(
) #pylint: disable=E0632
if eval_env is not None:
eval_obs, eval_returns, eval_masks, eval_actions, eval_values, eval_neglogpacs, eval_states, eval_epinfos = eval_runner.run(
) #pylint: disable=E0632
eplenmean = safemean([epinfo['ep_length'] for epinfo in epinfos])
eprewmean = safemean([epinfo['r'] for epinfo in epinfos])
rew_per_step = eprewmean / eplenmean
print("Curr learning rate {} \t Curr reward per step {}".format(
lrnow, rew_per_step))
if rew_per_step > best_rew_per_step and early_stopping:
# Avoid updating best model at first iteration because the means might be a bit off because
# of how the multithreaded batch simulation works
best_rew_per_step = eprewmean / eplenmean
checkdir = osp.join(logger.get_dir(), 'checkpoints')
model.save(checkdir + ".temp_best_model")
print("Saved model as best", best_rew_per_step, "avg rew/step")
epinfobuf.extend(epinfos)
if eval_env is not None:
eval_epinfobuf.extend(eval_epinfos)
# Here what we're going to do is for each minibatch calculate the loss and append it.
mblossvals = []
if states is None: # nonrecurrent version
# Index of each element of batch_size
# Create the indices array
inds = np.arange(nbatch)
for _ in range(noptepochs):
# Randomize the indexes
np.random.shuffle(inds)
# 0 to batch_size with batch_train_size step
for start in tqdm.trange(0,
nbatch,
nbatch_train,
desc="{}/{}".format(_, noptepochs)):
end = start + nbatch_train
mbinds = inds[start:end]
slices = (arr[mbinds]
for arr in (obs, returns, masks, actions, values,
neglogpacs))
mblossvals.append(model.train(lrnow, cliprangenow,
*slices))
else: # recurrent version
assert nenvs % nminibatches == 0
envsperbatch = nenvs // nminibatches
envinds = np.arange(nenvs)
flatinds = np.arange(nenvs * nsteps).reshape(nenvs, nsteps)
for _ in range(noptepochs):
np.random.shuffle(envinds)
for start in range(0, nenvs, envsperbatch):
end = start + envsperbatch
mbenvinds = envinds[start:end]
mbflatinds = flatinds[mbenvinds].ravel()
slices = (arr[mbflatinds]
for arr in (obs, returns, masks, actions, values,
neglogpacs))
mbstates = states[mbenvinds]
mblossvals.append(
model.train(lrnow, cliprangenow, *slices, mbstates))
# Feedforward --> get losses --> update
lossvals = np.mean(mblossvals, axis=0)
# End timer
tnow = time.perf_counter()
# Calculate the fps (frame per second)
fps = int(nbatch / (tnow - tstart))
if update % log_interval == 0 or update == 1:
# Calculates if value function is a good predicator of the returns (ev > 1)
# or if it's just worse than predicting nothing (ev =< 0)
ev = explained_variance(values, returns)
logger.logkv("serial_timesteps", update * nsteps)
logger.logkv("nupdates", update)
logger.logkv("total_timesteps", update * nbatch)
logger.logkv("fps", fps)
logger.logkv("explained_variance", float(ev))
eprewmean = safemean([epinfo['r'] for epinfo in epinfobuf])
ep_dense_rew_mean = safemean(
[epinfo['ep_shaped_r'] for epinfo in epinfobuf])
ep_sparse_rew_mean = safemean(
[epinfo['ep_sparse_r'] for epinfo in epinfobuf])
eplenmean = safemean([epinfo['ep_length'] for epinfo in epinfobuf])
run_info['eprewmean'].append(eprewmean)
run_info['ep_dense_rew_mean'].append(ep_dense_rew_mean)
run_info['ep_sparse_rew_mean'].append(ep_sparse_rew_mean)
run_info['eplenmean'].append(eplenmean)
run_info['explained_variance'].append(float(ev))
logger.logkv(
'true_eprew',
safemean([epinfo['ep_sparse_r'] for epinfo in epinfobuf]))
logger.logkv('eprewmean', eprewmean)
logger.logkv('eplenmean', eplenmean)
if eval_env is not None:
logger.logkv(
'eval_eprewmean',
safemean([epinfo['r'] for epinfo in eval_epinfobuf]))
logger.logkv(
'eval_eplenmean',
safemean([epinfo['l'] for epinfo in eval_epinfobuf]))
time_elapsed = tnow - tfirststart
logger.logkv('time_elapsed', time_elapsed)
time_per_update = time_elapsed / update
time_remaining = (nupdates - update) * time_per_update
logger.logkv('time_remaining', time_remaining / 60)
for (lossval, lossname) in zip(lossvals, model.loss_names):
run_info[lossname].append(lossval)
logger.logkv(lossname, lossval)
if MPI is None or MPI.COMM_WORLD.Get_rank() == 0:
logger.dumpkvs()
# Update current logs
if additional_params["RUN_TYPE"] in ["ppo", "joint_ppo"]:
from overcooked_ai_py.utils import save_dict_to_file
save_dict_to_file(run_info,
additional_params["SAVE_DIR"] + "logs")
# Linear annealing of reward shaping
if additional_params["REW_SHAPING_HORIZON"] != 0:
# Piecewise linear annealing schedule
# annealing_thresh: until when we should stop doing 100% reward shaping
# annealing_horizon: when we should reach doing 0% reward shaping
annealing_horizon = additional_params[
"REW_SHAPING_HORIZON"]
annealing_thresh = 0
def fn(x):
if annealing_thresh != 0 and annealing_thresh - (
annealing_horizon / annealing_thresh) * x > 1:
return 1
else:
fn = lambda x: -1 * (x - annealing_thresh) * 1 / (
annealing_horizon - annealing_thresh) + 1
return max(fn(x), 0)
curr_timestep = update * nbatch
curr_reward_shaping = fn(curr_timestep)
env.update_reward_shaping_param(curr_reward_shaping)
print("Current reward shaping", curr_reward_shaping)
sp_horizon = additional_params["SELF_PLAY_HORIZON"]
# Save/overwrite best model if past a certain threshold
if ep_sparse_rew_mean > bestrew and ep_sparse_rew_mean > additional_params[
"SAVE_BEST_THRESH"]:
# Don't save best model if still doing some self play and it's supposed to be a BC model
if additional_params[
"OTHER_AGENT_TYPE"][:
2] == "bc" and sp_horizon != 0 and env.self_play_randomization > 0:
pass
else:
from human_aware_rl.ppo.ppo import save_ppo_model
print("BEST REW", ep_sparse_rew_mean,
"overwriting previous model with", bestrew)
save_ppo_model(
model, "{}seed{}/best".format(
additional_params["SAVE_DIR"],
additional_params["CURR_SEED"]))
bestrew = max(ep_sparse_rew_mean, bestrew)
# If not sp run, and horizon is not None,
# vary amount of self play over time, either with a sigmoidal feedback loop
# or with a fixed piecewise linear schedule.
if additional_params[
"OTHER_AGENT_TYPE"] != "sp" and sp_horizon is not None:
if type(sp_horizon) is not list:
# Sigmoid self-play schedule based on current performance (not recommended)
curr_reward = ep_sparse_rew_mean
rew_target = sp_horizon
shift = rew_target / 2
t = (1 / rew_target) * 10
fn = lambda x: -1 * (np.exp(t * (x - shift)) /
(1 + np.exp(t * (x - shift)))) + 1
env.self_play_randomization = fn(curr_reward)
print("Current self-play randomization",
env.self_play_randomization)
else:
assert len(sp_horizon) == 2
# Piecewise linear self-play schedule
# self_play_thresh: when we should stop doing 100% self-play
# self_play_timeline: when we should reach doing 0% self-play
self_play_thresh, self_play_timeline = sp_horizon
def fn(x):
if self_play_thresh != 0 and self_play_timeline - (
self_play_timeline /
self_play_thresh) * x > 1:
return 1
else:
fn = lambda x: -1 * (
x - self_play_thresh) * 1 / (
self_play_timeline - self_play_thresh
) + 1
return max(fn(x), 0)
curr_timestep = update * nbatch
env.self_play_randomization = fn(curr_timestep)
print("Current self-play randomization",
env.self_play_randomization)
if save_interval and (update % save_interval == 0
or update == 1) and logger.get_dir() and (
MPI is None
or MPI.COMM_WORLD.Get_rank() == 0):
checkdir = osp.join(logger.get_dir(), 'checkpoints')
os.makedirs(checkdir, exist_ok=True)
savepath = osp.join(checkdir, '%.5i' % update)
print('Saving to', savepath)
model.save(savepath)
# Visualization of rollouts with actual other agent
run_type = additional_params["RUN_TYPE"]
if run_type in ["ppo", "joint_ppo"
] and update % additional_params["VIZ_FREQUENCY"] == 0:
from overcooked_ai_py.mdp.overcooked_env import OvercookedEnv
from overcooked_ai_py.mdp.overcooked_mdp import OvercookedGridworld
from overcooked_ai_py.agents.agent import AgentPair
from overcooked_ai_py.agents.benchmarking import AgentEvaluator
from human_aware_rl.baselines_utils import get_agent_from_model
print(additional_params["SAVE_DIR"])
mdp = OvercookedGridworld.from_layout_name(
**additional_params["mdp_params"])
overcooked_env = OvercookedEnv(mdp,
**additional_params["env_params"])
agent = get_agent_from_model(
model,
additional_params["sim_threads"],
is_joint_action=(run_type == "joint_ppo"))
agent.set_mdp(mdp)
if run_type == "ppo":
if additional_params["OTHER_AGENT_TYPE"] == 'sp':
agent_pair = AgentPair(agent,
agent,
allow_duplicate_agents=True)
else:
print("PPO agent on index 0:")
env.other_agent.set_mdp(mdp)
agent_pair = AgentPair(agent, env.other_agent)
trajectory, time_taken, tot_rewards, tot_shaped_rewards = overcooked_env.run_agents(
agent_pair, display=True, display_until=100)
overcooked_env.reset()
agent_pair.reset()
print("tot rew", tot_rewards, "tot rew shaped",
tot_shaped_rewards)
print("PPO agent on index 1:")
agent_pair = AgentPair(env.other_agent, agent)
else:
agent_pair = AgentPair(agent)
trajectory, time_taken, tot_rewards, tot_shaped_rewards = overcooked_env.run_agents(
agent_pair, display=True, display_until=100)
overcooked_env.reset()
agent_pair.reset()
print("tot rew", tot_rewards, "tot rew shaped", tot_shaped_rewards)
print(additional_params["SAVE_DIR"])
if nupdates > 0 and early_stopping:
checkdir = osp.join(logger.get_dir(), 'checkpoints')
print("Loaded best model", best_rew_per_step)
model.load(checkdir + ".temp_best_model")
return model, run_info
