class Agent(object):
"""Interacts with and learns from the environment."""
def __init__(self, state_size, action_size, random_seed, config):
"""
Initialize an Agent object.
Params
======
state_size (int): dimension of each state
action_size (int): dimension of each action
random_seed (int): random seed
config (dict) : dictionary of hyper-parameters
"""
self.state_size = state_size
self.action_size = action_size
self.seed = random.seed(random_seed)
self.config = config
# Actor Network (w/ Target Network)
self.actor_local = Actor(state_size,
action_size,
random_seed,
fc_units=[512, 256]).to(device)
self.actor_target = Actor(state_size,
action_size,
random_seed,
fc_units=[512, 256]).to(device)
self.actor_optimizer = optim.Adam(self.actor_local.parameters(),
lr=self.config["LR_ACTOR"])
# Critic Network (w/ Target Network)
self.critic_local = Critic(state_size,
action_size,
random_seed,
fcs_units=[512],
fc_units=[256]).to(device)
self.critic_target = Critic(state_size,
action_size,
random_seed,
fcs_units=[512],
fc_units=[256]).to(device)
self.critic_optimizer = optim.Adam(
self.critic_local.parameters(),
lr=self.config["LR_CRITIC"],
weight_decay=self.config["WEIGHT_DECAY"])
self.hard_copy_weights(self.actor_target, self.actor_local)
self.hard_copy_weights(self.critic_target, self.critic_local)
# Noise process
self.noise = OUNoise(action_size, random_seed)
# Replay memory
self.memory = ReplayBuffer(action_size, self.config["BUFFER_SIZE"],
self.config["BATCH_SIZE"], random_seed)
def hard_copy_weights(self, target, source):
# Copy weights from source to target network (part of initialization)
for target_param, param in zip(target.parameters(),
source.parameters()):
target_param.data.copy_(param.data)
def step(self, state, action, reward, next_state, done):
# Save experience in replay memory, and use random sample from buffer to learn.
self.memory.add(state, action, reward, next_state, done)
# If enough samples are available in memory, train models
if len(self.memory) > self.config["BATCH_SIZE"]:
experiences = self.memory.sample()
self.learn(experiences, self.config["GAMMA"])
def act(self, state, add_noise=True):
# Returns noisy actions for given state as per current policy.
state = torch.from_numpy(state).float().to(device)
self.actor_local.eval()
with torch.no_grad():
action = self.actor_local(state).cpu().data.numpy()
self.actor_local.train()
if add_noise:
action += self.noise.sample()
return np.clip(action, -1, 1)
def reset(self):
# Reset Noise Generator
self.noise.reset()
def learn(self, experiences, gamma, t_step=1, update_every=1):
"""Update policy and value parameters using given batch of experience tuples.
Q_targets = r + γ * critic_target(next_state, actor_target(next_state))
where:
actor_target(state) -> action
critic_target(state, action) -> Q-value
Params
======
experiences (Tuple[torch.Tensor]): tuple of (s, a, r, s', done) tuples
gamma (float): discount factor
t_step (int) : total_step count
update_every (int) : update model once in every n steps
"""
states, actions, rewards, next_states, dones = experiences
# ---------------------------- update critic ---------------------------- #
# Get predicted next-state actions and Q values from target models
actions_next = self.actor_target(next_states)
Q_targets_next = self.critic_target(next_states, actions_next)
# Compute Q targets for current states (y_i)
Q_targets = rewards + (gamma * Q_targets_next * (1 - dones))
# Compute critic loss
Q_expected = self.critic_local(states, actions)
critic_loss = F.mse_loss(Q_expected, Q_targets)
# Minimize the loss
self.critic_optimizer.zero_grad()
critic_loss.backward()
# Clip gradients of critic
# torch.nn.utils.clip_grad_norm_(self.critic_local.parameters(), 2)
self.critic_optimizer.step()
# ---------------------------- update actor ---------------------------- #
# Compute actor loss
actions_pred = self.actor_local(states)
actor_loss = -self.critic_local(states, actions_pred).mean()
# Minimize the loss
self.actor_optimizer.zero_grad()
actor_loss.backward()
# Clip gradients of actor
# torch.nn.utils.clip_grad_norm_(self.actor_local.parameters(), 1)
self.actor_optimizer.step()
if t_step % update_every == 0:
# ----------------------- update target networks ----------------------- #
self.soft_update(self.critic_local, self.critic_target,
self.config["TAU"])
self.soft_update(self.actor_local, self.actor_target,
self.config["TAU"])
def soft_update(self, local_model, target_model, tau):
"""Soft update model parameters.
θ_target = τ*θ_local + (1 - τ)*θ_target
Params
======
local_model: PyTorch model (weights will be copied from)
target_model: PyTorch model (weights will be copied to)
tau (float): interpolation parameter
"""
for target_param, local_param in zip(target_model.parameters(),
local_model.parameters()):
target_param.data.copy_(tau * local_param.data +
(1.0 - tau) * target_param.data)
def save(self, name="agent_1_"):
# Save actor and critic models
torch.save(self.actor_local.state_dict(),
f"{name}_actor_checkpoint_actor.pth")
torch.save(self.critic_local.state_dict(),
f"{name}_critic_checkpoint_actor.pth")
class Agent(AgentABC):
def __init__(self, state_size, action_size, num_agents, random_seed):
"""
Initialize an DDPG Agent object.
:param state_size (int): dimension of each state
:param action_size (int): dimension of each action
:param num_agents (int): number of agents in environment ot use ddpg
:param random_seed (int): random seed
"""
super().__init__(state_size, action_size, num_agents, random_seed)
self.state_size = state_size
self.action_size = action_size
self.num_agents = num_agents
self.seed = random.seed(random_seed)
# Actor Network (w/ Target Network)
self.actor_local = Actor(state_size, action_size,
random_seed).to(device)
self.actor_target = Actor(state_size, action_size,
random_seed).to(device)
self.actor_optimizer = optim.Adam(self.actor_local.parameters(),
lr=LR_ACTOR)
# Critic Network (w/ Target Network)
self.critic_local = Critic(state_size, action_size,
random_seed).to(device)
self.critic_target = Critic(state_size, action_size,
random_seed).to(device)
self.critic_optimizer = optim.Adam(self.critic_local.parameters(),
lr=LR_CRITIC,
weight_decay=WEIGHT_DECAY)
# Noise process for each agent
self.noise = OUNoise((num_agents, action_size), random_seed)
# Replay memory
self.memory = ReplayBuffer(action_size, BUFFER_SIZE, BATCH_SIZE,
random_seed)
# debug of the MSE critic loss
self.mse_error_list = []
def step(self, states, actions, rewards, next_states, dones):
"""Save experience in replay memory, and use random sample from buffer to learn."""
# Save experience / reward
for agent in range(self.num_agents):
self.memory.add(states[agent, :], actions[agent, :],
rewards[agent], next_states[agent, :],
dones[agent])
# Learn, if enough samples are available in memory
if len(self.memory) > BATCH_SIZE:
experiences = self.memory.sample()
self.learn(experiences)
self.debug_loss = np.mean(self.mse_error_list)
def act(self, state, add_noise=True):
"""Returns actions for given state as per current policy."""
state = torch.from_numpy(state).float().to(device)
acts = np.zeros((self.num_agents, self.action_size))
self.actor_local.eval()
with torch.no_grad():
for agent in range(self.num_agents):
acts[agent, :] = self.actor_local(
state[agent, :]).cpu().data.numpy()
self.actor_local.train()
if add_noise:
noise = self.noise.sample()
acts += noise
return np.clip(acts, -1, 1)
def reset(self):
""" see abstract class """
super().reset()
self.noise.reset()
self.mse_error_list = []
def learn(self, experiences):
"""Update policy and value parameters using given batch of experience tuples.
Q_targets = r + γ * critic_target(next_state, actor_target(next_state))
where:
actor_target(state) -> action
critic_target(state, action) -> Q-value
Params
======
experiences (Tuple[torch.Tensor]): tuple of (s, a, r, s', done) tuples
gamma (float): discount factor
"""
states, actions, rewards, next_states, dones = experiences
# ---------------------------- update critic ---------------------------- #
# Get predicted next-state actions and Q values from target models
actions_next = self.actor_target(next_states)
Q_targets_next = self.critic_target(next_states, actions_next)
# Compute Q targets for current states (y_i)
Q_targets = rewards.view(BATCH_SIZE,
-1) + (GAMMA * Q_targets_next *
(1 - dones).view(BATCH_SIZE, -1))
# Compute critic loss
Q_expected = self.critic_local(states, actions)
critic_loss = F.mse_loss(Q_expected, Q_targets)
self.mse_error_list.append(critic_loss.detach().cpu().numpy())
# Minimize the loss
self.critic_optimizer.zero_grad()
critic_loss.backward()
self.critic_optimizer.step()
# ---------------------------- update actor ---------------------------- #
# Compute actor loss
actions_pred = self.actor_local(states)
actor_loss = -self.critic_local(states, actions_pred).mean()
# Minimize the loss
self.actor_optimizer.zero_grad()
actor_loss.backward()
self.actor_optimizer.step()
# ----------------------- update target networks ----------------------- #
self.soft_update(self.critic_local, self.critic_target, TAU)
self.soft_update(self.actor_local, self.actor_target, TAU)
@staticmethod
def soft_update(local_model, target_model, tau):
"""Soft update model parameters.
θ_target = τ*θ_local + (1 - τ)*θ_target
Params
======
local_model: PyTorch model (weights will be copied from)
target_model: PyTorch model (weights will be copied to)
tau (float): interpolation parameter
"""
for target_param, local_param in zip(target_model.parameters(),
local_model.parameters()):
target_param.data.copy_(tau * local_param.data +
(1.0 - tau) * target_param.data)
def load_weights(self, directory_path):
""" see abstract class """
super().load_weights(directory_path)
self.actor_target.load_state_dict(
torch.load(os.path.join(directory_path, an_filename),
map_location=device))
self.critic_target.load_state_dict(
torch.load(os.path.join(directory_path, cn_filename),
map_location=device))
self.actor_local.load_state_dict(
torch.load(os.path.join(directory_path, an_filename),
map_location=device))
self.critic_local.load_state_dict(
torch.load(os.path.join(directory_path, cn_filename),
map_location=device))
def save_weights(self, directory_path):
""" see abstract class """
super().save_weights(directory_path)
torch.save(self.actor_local.state_dict(),
os.path.join(directory_path, an_filename))
torch.save(self.critic_local.state_dict(),
os.path.join(directory_path, cn_filename))
def save_mem(self, directory_path):
""" see abstract class """
super().save_mem(directory_path)
self.memory.save(os.path.join(directory_path, "ddpg_memory"))
def load_mem(self, directory_path):
""" see abstract class """
super().load_mem(directory_path)
self.memory.load(os.path.join(directory_path, "ddpg_memory"))
class Agent(AgentABC):
"""Interacts with and learns from the environment."""
def __init__(self, state_size, action_size, num_agents, random_seed):
"""Initialize an MADDPG Agent object.
Params
======
:param state_size: dimension of each state
:param action_size: dimension of each action
:param num_agents: number of inner agents
:param random_seed: random seed
"""
super().__init__(state_size, action_size, num_agents, random_seed)
self.state_size = state_size
self.action_size = action_size
self.num_agents = num_agents
self.seed = random.seed(random_seed)
self.actors_local = []
self.actors_target = []
self.actor_optimizers = []
self.critics_local = []
self.critics_target = []
self.critic_optimizers = []
for i in range(num_agents):
# Actor Network (w/ Target Network)
self.actors_local.append(
Actor(state_size, action_size, random_seed).to(device))
self.actors_target.append(
Actor(state_size, action_size, random_seed).to(device))
self.actor_optimizers.append(
optim.Adam(self.actors_local[i].parameters(), lr=LR_ACTOR))
# Critic Network (w/ Target Network)
self.critics_local.append(
Critic(num_agents * state_size, num_agents * action_size,
random_seed).to(device))
self.critics_target.append(
Critic(num_agents * state_size, num_agents * action_size,
random_seed).to(device))
self.critic_optimizers.append(
optim.Adam(self.critics_local[i].parameters(),
lr=LR_CRITIC,
weight_decay=WEIGHT_DECAY))
# Noise process for each agent
self.noise = OUNoise((num_agents, action_size), random_seed)
# Replay memory
self.memory = ReplayBuffer(action_size, BUFFER_SIZE, BATCH_SIZE,
random_seed)
# debugging variables
self.step_count = 0
self.mse_error_list = []
def step(self, states, actions, rewards, next_states, dones):
"""Save experience in replay memory, and use random sample from buffer to learn."""
# Save experience / reward
self.memory.add(states, actions, rewards, next_states, dones)
# Learn, if enough samples are available in memory
# in order to add some stability to the learning, we don't modify weights every turn.
self.step_count += 1
if (self.step_count %
UPDATE_EVERY) == 0: # learn every #UPDATE_EVERY steps
for i in range(NUM_UPDATES): # update #NUM_UPDATES times
if len(self.memory) > 1000:
experiences = self.memory.sample()
self.learn(experiences)
self.debug_loss = np.mean(self.mse_error_list)
self.update_target_networks()
def act(self, state, add_noise=True):
"""Returns actions for given state as per current policy."""
state = torch.from_numpy(state).float().to(device)
acts = np.zeros((self.num_agents, self.action_size))
for agent in range(self.num_agents):
self.actors_local[agent].eval()
with torch.no_grad():
acts[agent, :] = self.actors_local[agent](
state[agent, :]).cpu().data.numpy()
self.actors_local[agent].train()
if add_noise:
acts += self.noise.sample()
return np.clip(acts, -1, 1)
def reset(self):
""" see abstract class """
super().reset()
self.noise.reset()
self.mse_error_list = []
def learn(self, experiences):
"""Update policy and value parameters using given batch of experience tuples.
Q_targets = r + γ * critic_target(next_full_state, actors_target(next_partial_state) )
where:
actor_target(state) -> action
critic_target(state, action) -> Q-value
Params
======
experiences (Tuple[torch.Tensor]): tuple of (s, a, r, s', done) tuples
gamma (float): discount factor
"""
states_batched, actions_batched, rewards, next_states_batched, dones = experiences
states_concated = states_batched.view(
[BATCH_SIZE, self.num_agents * self.state_size])
next_states_concated = next_states_batched.view(
[BATCH_SIZE, self.num_agents * self.state_size])
actions_concated = actions_batched.view(
[BATCH_SIZE, self.num_agents * self.action_size])
for agent in range(self.num_agents):
actions_next_batched = [
self.actors_target[i](next_states_batched[:, i, :])
for i in range(self.num_agents)
]
actions_next_whole = torch.cat(actions_next_batched, 1)
# ---------------------------- update critic ---------------------------- #
# Get predicted next-state actions and Q values from target models
q_targets_next = self.critics_target[agent](next_states_concated,
actions_next_whole)
# Compute Q targets for current states (y_i)
q_targets = rewards[:, agent].view(
BATCH_SIZE, -1) + (GAMMA * q_targets_next *
(1 - dones[:, agent].view(BATCH_SIZE, -1)))
# Compute critic loss
q_expected = self.critics_local[agent](states_concated,
actions_concated)
critic_loss = F.mse_loss(q_expected, q_targets)
# Minimize the loss
self.critic_optimizers[agent].zero_grad()
critic_loss.backward()
self.critic_optimizers[agent].step()
# save the error for statistics
self.mse_error_list.append(critic_loss.detach().cpu().numpy())
# ---------------------------- update actor ---------------------------- #
action_i = self.actors_local[agent](states_batched[:, agent, :])
actions_pred = actions_batched.clone()
actions_pred[:, agent, :] = action_i
actions_pred_whole = actions_pred.view(BATCH_SIZE, -1)
# Compute actor loss
actor_loss = -self.critics_local[agent](states_concated,
actions_pred_whole).mean()
# Minimize the loss
self.actor_optimizers[agent].zero_grad()
actor_loss.backward()
self.actor_optimizers[agent].step()
def update_target_networks(self):
# ----------------------- update target networks ----------------------- #
for agent in range(self.num_agents):
self.soft_update(self.critics_local[agent],
self.critics_target[agent], TAU)
self.soft_update(self.actors_local[agent],
self.actors_target[agent], TAU)
@staticmethod
def soft_update(local_model, target_model, tau):
"""Soft update model parameters.
θ_target = τ*θ_local + (1 - τ)*θ_target
Params
======
local_model: PyTorch model (weights will be copied from)
target_model: PyTorch model (weights will be copied to)
tau (float): interpolation parameter
"""
for target_param, local_param in zip(target_model.parameters(),
local_model.parameters()):
target_param.data.copy_(tau * local_param.data +
(1.0 - tau) * target_param.data)
def load_weights(self, directory_path):
""" see abstract class """
super().load_weights(directory_path)
actor_weights = os.path.join(directory_path, an_filename)
critic_weights = os.path.join(directory_path, cn_filename)
for agent in range(self.num_agents):
self.actors_target[agent].load_state_dict(
torch.load(actor_weights + "_" + str(agent),
map_location=device))
self.critics_target[agent].load_state_dict(
torch.load(critic_weights + "_" + str(agent),
map_location=device))
self.actors_local[agent].load_state_dict(
torch.load(actor_weights + "_" + str(agent),
map_location=device))
self.critics_local[agent].load_state_dict(
torch.load(critic_weights + "_" + str(agent),
map_location=device))
def save_weights(self, directory_path):
""" see abstract class """
super().save_weights(directory_path)
actor_weights = os.path.join(directory_path, an_filename)
critic_weights = os.path.join(directory_path, cn_filename)
for agent in range(self.num_agents):
torch.save(self.actors_local[agent].state_dict(),
actor_weights + "_" + str(agent))
torch.save(self.critics_local[agent].state_dict(),
critic_weights + "_" + str(agent))
def save_mem(self, directory_path):
""" see abstract class """
super().save_mem(directory_path)
self.memory.save(os.path.join(directory_path, memory_filename))
def load_mem(self, directory_path):
""" see abstract class """
super().load_mem(directory_path)
self.memory.load(os.path.join(directory_path, memory_filename))
class Agent():
"""Interacts with and learns from the environment."""
def __init__(self, state_size, action_size, num_agents, random_seed):
"""Initialize an Agent object.
Params
======
state_size (int): dimension of each state
action_size (int): dimension of each action
random_seed (int): random seed
"""
self.state_size = state_size
self.action_size = action_size
self.num_agents = num_agents
self.seed = random.seed(random_seed)
# Actor Network (w/ Target Network)
self.actor_local = Actor(state_size, action_size, random_seed).to(device)
self.actor_target = Actor(state_size, action_size, random_seed).to(device)
self.actor_optimizer = optim.Adam(self.actor_local.parameters(), lr=LR_ACTOR)
for target_param, local_param in zip(self.actor_target.parameters(), self.actor_local.parameters()):
target_param.data.copy_(local_param.data)
# Critic Network (w/ Target Network)
self.critic_local = Critic(state_size, action_size, random_seed).to(device)
self.critic_target = Critic(state_size, action_size, random_seed).to(device)
self.critic_optimizer = optim.Adam(self.critic_local.parameters(), lr=LR_CRITIC, weight_decay=WEIGHT_DECAY)
for target_param, local_param in zip(self.critic_target.parameters(), self.critic_local.parameters()):
target_param.data.copy_(local_param.data)
# Noise process
self.noise = OUNoise((num_agents, action_size), random_seed)
# Replay memory
self.memory = ReplayBuffer(action_size, BUFFER_SIZE, BATCH_SIZE, random_seed)
# Initialize time step (for updating every UPDATE_EVERY steps)
self.t_step = 0
def step(self, states, actions, rewards, next_states, dones):
"""Save experience in replay memory, and use random sample from buffer to learn."""
# Save experience / reward
for i in range(self.num_agents):
self.memory.add(states[i], actions[i], rewards[i], next_states[i], dones[i])
self.t_step = (self.t_step + 1) % UPDATE_EVERY
if self.t_step == 0:
# Learn, if enough samples are available in memory
if len(self.memory) > BATCH_SIZE:
for _ in range(NUM_LEARN):
experiences = self.memory.sample()
self.learn(experiences, GAMMA)
def act(self, states, eps=0, add_noise=True):
"""Returns actions for given state as per current policy."""
states = torch.from_numpy(states).float().to(device)
self.actor_local.eval()
with torch.no_grad():
actions = self.actor_local(states).cpu().data.numpy()
self.actor_local.train()
if add_noise and random.random() < eps:
actions += self.noise.sample()
return np.clip(actions, -1, 1)
def reset(self):
self.noise.reset()
def learn(self, experiences, gamma):
"""Update policy and value parameters using given batch of experience tuples.
Q_targets = r + γ * critic_target(next_state, actor_target(next_state))
where:
actor_target(state) -> action
critic_target(state, action) -> Q-value
Params
======
experiences (Tuple[torch.Tensor]): tuple of (s, a, r, s', done) tuples
gamma (float): discount factor
"""
states, actions, rewards, next_states, dones = experiences
# ---------------------------- update critic ---------------------------- #
# Get predicted next-state actions and Q values from target models
actions_next = self.actor_target(next_states)
Q_targets_next = self.critic_target(next_states, actions_next)
# Compute Q targets for current states (y_i)
Q_targets = rewards + (gamma * Q_targets_next * (1 - dones))
# Compute critic loss
Q_expected = self.critic_local(states, actions)
critic_loss = F.mse_loss(Q_expected, Q_targets)
# Minimize the loss
self.critic_optimizer.zero_grad()
critic_loss.backward()
# Gradient clipping prevents the gradient update moving the parameters extermly far,
# when there are cliff-like structures in the strongly non-linear function
# that a neural network wants to approximate. {Goodfellow, Section 10.11.1, Deep Learning}
# {Mikolov, 2012; Pascanu et al., 2013}
torch.nn.utils.clip_grad_norm_(self.critic_local.parameters(), 1.0)
self.critic_optimizer.step()
# ---------------------------- update actor ---------------------------- #
# Compute actor loss
actions_pred = self.actor_local(states)
actor_loss = -self.critic_local(states, actions_pred).mean()
# Minimize the loss
self.actor_optimizer.zero_grad()
actor_loss.backward()
self.actor_optimizer.step()
# ----------------------- update target networks ----------------------- #
self.soft_update(self.critic_local, self.critic_target, TAU)
self.soft_update(self.actor_local, self.actor_target, TAU)
def soft_update(self, local_model, target_model, tau):
"""Soft update model parameters.
θ_target = τ*θ_local + (1 - τ)*θ_target
Params
======
local_model: PyTorch model (weights will be copied from)
target_model: PyTorch model (weights will be copied to)
tau (float): interpolation parameter
"""
for target_param, local_param in zip(target_model.parameters(), local_model.parameters()):
target_param.data.copy_(tau*local_param.data + (1.0-tau)*target_param.data)
def __init__(self,
algo_type="MADDPG",
act_space=None,
obs_space=None,
rnn_policy=False,
rnn_critic=False,
hidden_dim=64,
lr=0.01,
env_obs_space=None,
env_act_space=None):
"""
Inputs:
act_space: single agent action space (single space or Dict)
obs_space: single agent observation space (single space Dict)
"""
self.algo_type = algo_type
self.act_space = act_space
self.obs_space = obs_space
# continuous or discrete action (only look at `move` action, assume
# move and comm space both discrete or continuous)
if isinstance(act_space, Box) or isinstance(act_space["move"], Box):
discrete_action = False
elif isinstance(act_space, Discrete) or isinstance(
act_space["move"], Discrete):
discrete_action = True
self.discrete_action = discrete_action
# Exploration noise
if not discrete_action:
# `move`, `comm` share same continuous noise source
self.exploration = OUNoise(self.get_shape(act_space))
else:
self.exploration = 0.3 # epsilon for eps-greedy
# Policy (supports multiple outputs)
self.rnn_policy = rnn_policy
self.policy_hidden_states = None
num_in_pol = obs_space.shape[0]
if isinstance(act_space, Dict):
# hard specify now, could generalize later
num_out_pol = {
"move": self.get_shape(act_space, "move"),
"comm": self.get_shape(act_space, "comm")
}
else:
num_out_pol = self.get_shape(act_space)
self.policy = Policy(num_in_pol,
num_out_pol,
hidden_dim=hidden_dim,
constrain_out=True,
discrete_action=discrete_action,
rnn_policy=rnn_policy)
self.target_policy = Policy(num_in_pol,
num_out_pol,
hidden_dim=hidden_dim,
constrain_out=True,
discrete_action=discrete_action,
rnn_policy=rnn_policy)
hard_update(self.target_policy, self.policy)
# Critic
self.rnn_critic = rnn_critic
self.critic_hidden_states = None
if algo_type == "MADDPG":
num_in_critic = 0
for oobsp in env_observation_space:
num_in_critic += oobsp.shape[0]
for oacsp in env_action_space:
# feed all acts to centralized critic
num_in_critic += self.get_shape(oacsp)
else: # only DDPG, local critic
num_in_critic = obs_space.shape[0] + self.get_shape(act_space)
critic_net_fn = RecurrentNetwork if rnn_critic else MLPNetwork
self.critic = critic_net_fn(num_in_critic,
1,
hidden_dim=hidden_dim,
constrain_out=False)
self.target_critic = critic_net_fn(num_in_critic,
1,
hidden_dim=hidden_dim,
constrain_out=False)
hard_update(self.target_critic, self.critic)
# Optimizers
self.policy_optimizer = Adam(self.policy.parameters(), lr=lr)
self.critic_optimizer = Adam(self.critic.parameters(), lr=lr)
class Agent():
def __init__(self, num_agents, state_size, action_size, opts):
self.num_agents = num_agents
self.state_size = state_size
self.action_size = action_size
self.opts = opts
# Actor Network
self.actor_local = ActorNet(state_size,
action_size,
fc1_units=opts.a_fc1,
fc2_units=opts.a_fc2).to(opts.device)
self.actor_target = ActorNet(state_size,
action_size,
fc1_units=opts.a_fc1,
fc2_units=opts.a_fc2).to(opts.device)
self.actor_optimizer = torch.optim.Adam(self.actor_local.parameters(),
lr=opts.actor_lr)
# Critic Network
self.critic_local = CriticNet(state_size,
action_size,
fc1_units=opts.c_fc1,
fc2_units=opts.c_fc2).to(opts.device)
self.critic_target = CriticNet(state_size,
action_size,
fc1_units=opts.c_fc1,
fc2_units=opts.c_fc2).to(opts.device)
self.critic_optimizer = torch.optim.Adam(
self.critic_local.parameters(),
lr=opts.critic_lr,
weight_decay=opts.critic_weight_decay)
# Noise process
self.noise = OUNoise((num_agents, action_size), opts.random_seed)
self.step_idx = 0
# Replay memory
self.memory = ReplayBuffer(action_size, opts.buffer_size,
opts.batch_size, opts.random_seed,
opts.device)
def step(self, state, action, reward, next_state, done):
for i in range(self.num_agents):
self.memory.add(state[i, :], action[i, :], reward[i],
next_state[i, :], done[i])
self.step_idx += 1
is_learn_iteration = (self.step_idx % self.opts.learn_every) == 0
is_update_iteration = (self.step_idx % self.opts.update_every) == 0
if len(self.memory) > self.opts.batch_size:
if is_learn_iteration:
experiences = self.memory.sample()
self.learn(experiences, self.opts.gamma)
if is_update_iteration:
soft_update(self.critic_local, self.critic_target,
self.opts.tau)
soft_update(self.actor_local, self.actor_target, self.opts.tau)
def act(self, state):
state = torch.from_numpy(state).float().to(self.opts.device)
self.actor_local.eval()
with torch.no_grad():
action = self.actor_local(state).cpu().data.numpy()
self.actor_local.train()
action += self.noise.sample()
return np.clip(action, self.opts.minimum_action_value,
self.opts.maximum_action_value)
def save(self):
torch.save(self.critic_local.state_dict(),
self.opts.output_data_path + "critic_local.pth")
torch.save(self.critic_target.state_dict(),
self.opts.output_data_path + "critic_target.pth")
torch.save(self.actor_local.state_dict(),
self.opts.output_data_path + "actor_local.pth")
torch.save(self.actor_target.state_dict(),
self.opts.output_data_path + "actor_target.pth")
def learn(self, experiences, gamma):
states, actions, rewards, next_states, dones = experiences
states = tensor(states, self.opts.device)
actions = tensor(actions, self.opts.device)
rewards = tensor(rewards, self.opts.device)
next_states = tensor(next_states, self.opts.device)
mask = tensor(1 - dones, self.opts.device)
# Update critic
actions_next = self.actor_target(next_states)
Q_targets_next = self.critic_target(next_states, actions_next)
Q_targets = rewards + (gamma * Q_targets_next * mask)
# Compute & minimize critic loss
Q_expected = self.critic_local(states, actions)
critic_loss = F.mse_loss(Q_expected, Q_targets)
self.critic_optimizer.zero_grad()
critic_loss.backward()
self.critic_optimizer.step()
# Update actor
actions_pred = self.actor_local(states)
# Compute & minimize critic loss
actor_loss = -self.critic_local(states, actions_pred).mean()
self.actor_optimizer.zero_grad()
actor_loss.backward()
self.actor_optimizer.step()
class DDPGAgent(object):
"""
General class for DDPG agents (policy, critic, target policy, target
critic, exploration noise)
"""
def __init__(self,
algo_type="MADDPG",
act_space=None,
obs_space=None,
rnn_policy=False,
rnn_critic=False,
hidden_dim=64,
lr=0.01,
env_obs_space=None,
env_act_space=None):
"""
Inputs:
act_space: single agent action space (single space or Dict)
obs_space: single agent observation space (single space Dict)
"""
self.algo_type = algo_type
self.act_space = act_space
self.obs_space = obs_space
# continuous or discrete action (only look at `move` action, assume
# move and comm space both discrete or continuous)
if isinstance(act_space, Box) or isinstance(act_space["move"], Box):
discrete_action = False
elif isinstance(act_space, Discrete) or isinstance(
act_space["move"], Discrete):
discrete_action = True
self.discrete_action = discrete_action
# Exploration noise
if not discrete_action:
# `move`, `comm` share same continuous noise source
self.exploration = OUNoise(self.get_shape(act_space))
else:
self.exploration = 0.3 # epsilon for eps-greedy
# Policy (supports multiple outputs)
self.rnn_policy = rnn_policy
self.policy_hidden_states = None
num_in_pol = obs_space.shape[0]
if isinstance(act_space, Dict):
# hard specify now, could generalize later
num_out_pol = {
"move": self.get_shape(act_space, "move"),
"comm": self.get_shape(act_space, "comm")
}
else:
num_out_pol = self.get_shape(act_space)
self.policy = Policy(num_in_pol,
num_out_pol,
hidden_dim=hidden_dim,
constrain_out=True,
discrete_action=discrete_action,
rnn_policy=rnn_policy)
self.target_policy = Policy(num_in_pol,
num_out_pol,
hidden_dim=hidden_dim,
constrain_out=True,
discrete_action=discrete_action,
rnn_policy=rnn_policy)
hard_update(self.target_policy, self.policy)
# Critic
self.rnn_critic = rnn_critic
self.critic_hidden_states = None
if algo_type == "MADDPG":
num_in_critic = 0
for oobsp in env_observation_space:
num_in_critic += oobsp.shape[0]
for oacsp in env_action_space:
# feed all acts to centralized critic
num_in_critic += self.get_shape(oacsp)
else: # only DDPG, local critic
num_in_critic = obs_space.shape[0] + self.get_shape(act_space)
critic_net_fn = RecurrentNetwork if rnn_critic else MLPNetwork
self.critic = critic_net_fn(num_in_critic,
1,
hidden_dim=hidden_dim,
constrain_out=False)
self.target_critic = critic_net_fn(num_in_critic,
1,
hidden_dim=hidden_dim,
constrain_out=False)
hard_update(self.target_critic, self.critic)
# Optimizers
self.policy_optimizer = Adam(self.policy.parameters(), lr=lr)
self.critic_optimizer = Adam(self.critic.parameters(), lr=lr)
def get_shape(self, x, key=None):
""" func to infer action output shape """
if isinstance(x, Dict):
if key is None: # sum of action space dims
return sum([
x[k].n if self.discrete_action else x[k].shape[0]
for k in x.spaces
])
elif key in x.spaces:
return x[key].n if self.discrete_action else x[key].shape[0]
else: # key not in action spaces
return 0
else:
return x.n if self.discrete_action else x.shape[0]
def reset_noise(self):
if not self.discrete_action:
self.exploration.reset()
def scale_noise(self, scale):
if self.discrete_action:
self.exploration = scale
else:
self.exploration.scale = scale
def init_hidden(self, batch_size):
# (1,H) -> (B,H)
# policy.init_hidden().unsqueeze(0).expand(batch_size, self.n_agents, -1)
if self.rnn_policy:
self.policy_hidden_states = self.policy.init_hidden().expand(
batch_size, -1)
if self.rnn_critic:
self.critic_hidden_states = self.policy.init_hidden().expand(
batch_size, -1)
def compute_q_val(self, vf_in, h=None, target=False):
""" training critic forward with specified policy
Arguments:
agent_i: index to agent; critic: critic network to agent; vf_in: (B,T,K);
bs: batch size; ts: length of episode; target: if use target network
Returns:
q: (B*T,1)
"""
bs, ts, _ = vf_in.shape
critic = self.target_critic if target else self.critic
if self.rnn_critic:
q = [] # (B,1)*T
h_t = self.critic_hidden_states.clone() # (B,H)
for t in range(ts):
q_t, h_t = critic(vf_in[:, t], h_t)
q.append(q_t)
q = torch.stack(q, 0).permute(1, 0, 2) # (T,B,1) -> (B,T,1)
q = q.reshape(bs * ts, -1) # (B*T,1)
else:
# (B,T,D) -> (B*T,1)
q = critic(vf_in.reshape(bs * ts, -1))
return q
def compute_action(self, obs, h=None, target=False, requires_grad=True):
""" traininsg actor forward with specified policy
concat all actions to be fed in critics
Arguments:
agent_i: index to agent; pi: policy to agent; obs: (B,T,O);
bs: batch size; ts: length of episode; target: if use target network
Returns:
act: dict of (B,T,A)
"""
def _soft_act(x): # x: (B,A)
if not self.discrete_action:
return x
if requires_grad:
return gumbel_softmax(x, hard=True)
else:
return onehot_from_logits(x)
bs, ts, _ = obs.shape
pi = self.target_policy if target else self.policy
if self.rnn_policy:
act = [] # [(B,sum(A_k))]*T
h_t = self.policy_hidden_states.clone() # (B,H)
for t in range(ts):
act_t, h_t = pi(obs[:, t], h_t) # act_t is dict!!
cat_act = torch.concat(
[_soft_act(a) for k, a in act_t.items()],
-1) # (B,sum(A_k))
act.append(cat_act)
act = torch.stack(act, 0).permute(1, 0, 2) # (B,T,sum(A_k))
else:
stacked_obs = obs.reshape(bs * ts, -1) # (B*T,O)
act, _ = pi(stacked_obs) # act is dict of (B*T,A)
act = torch.concat(
[_soft_act(a).reshape(bs, ts, -1) for k, a in act.items()],
-1) # (B,T,sum(A_k))
return act
def step(self, obs, explore=False):
"""
Take a step forward in environment for a minibatch of observations
equivalent to `act` or `compute_actions`
Arguments:
obs: (B,O)
explore: Whether or not to add exploration noise
Returns:
action: dict of actions for this agent, (B,A)
"""
with torch.no_grad():
action, hidden_states = self.policy(obs, self.policy_hidden_states)
self.policy_hidden_states = hidden_states # if mlp, still defafult None
if self.discrete_action:
for k in action:
if explore:
action[k] = gumbel_softmax(action[k], hard=True)
else:
action[k] = onehot_from_logits(action[k])
else: # continuous action
idx = 0
noise = Variable(Tensor(self.exploration.noise()),
requires_grad=False)
for k in action:
if explore:
dim = action[k].shape[-1]
action[k] += noise[idx:idx + dim]
idx += dim
action[k] = action[k].clamp(-1, 1)
return action
def get_params(self):
return {
'policy': self.policy.state_dict(),
'critic': self.critic.state_dict(),
'target_policy': self.target_policy.state_dict(),
'target_critic': self.target_critic.state_dict(),
'policy_optimizer': self.policy_optimizer.state_dict(),
'critic_optimizer': self.critic_optimizer.state_dict()
}
def load_params(self, params):
self.policy.load_state_dict(params['policy'])
self.critic.load_state_dict(params['critic'])
self.target_policy.load_state_dict(params['target_policy'])
self.target_critic.load_state_dict(params['target_critic'])
self.policy_optimizer.load_state_dict(params['policy_optimizer'])
self.critic_optimizer.load_state_dict(params['critic_optimizer'])
if __name__ == '__main__':
with tf.Session() as sess:
env = gym.make('LunarLanderContinuous-v2')
env.seed(0)
np.random.seed(0)
tf.set_random_seed(0)
ep = 2000
tau = 0.001
gamma = 0.99
min_batch = 65
actor_lr = 0.00005
critic_lr = 0.0005
buffer_size = 1000000
state_dim = env.observation_space.shape[0]
action_dim = env.action_space.shape[0]
action_bound = env.action_space.high
actor_noise = OUNoise(mu=np.zeros(action_dim))
actor = ActorNetwork(sess, state_dim, action_dim, action_bound, actor_lr, tau, min_batch)
critic = CriticNetwork(sess, state_dim, action_dim, critic_lr, tau, gamma, actor.get_num_trainable_vars())
scores = train(sess, env, actor, critic, actor_noise, buffer_size, min_batch, ep)
plt.plot([i + 1 for i in range(0, len(scores), 4)], scores[::4])
plt.show()
def __init__(self, algo_type="MADDPG", act_space=None, obs_space=None,
rnn_policy=False, rnn_critic=False, hidden_dim=64, lr=0.01,
norm_in=False, constrain_out=False, thought_dim=64,
env_obs_space=None, env_act_space=None, **kwargs):
"""
Inputs:
act_space: single agent action space (single space or Dict)
obs_space: single agent observation space (single space Dict)
"""
self.algo_type = algo_type
self.act_space = act_space
self.obs_space = obs_space
# continuous or discrete action (only look at `move` action, assume
# move and comm space both discrete or continuous)
tmp = act_space.spaces["move"] if isinstance(act_space, Dict) else act_space
self.discrete_action = False if isinstance(tmp, Box) else True
# Exploration noise
if not self.discrete_action:
# `move`, `comm` share same continuous noise source
self.exploration = OUNoise(self.get_shape(act_space))
else:
self.exploration = 0.3 # epsilon for eps-greedy
# Policy (supports multiple outputs)
self.rnn_policy = rnn_policy
self.policy_hidden_states = None
num_in_pol = obs_space.shape[0]
if isinstance(act_space, Dict):
# hard specify now, could generalize later
num_out_pol = {
"move": self.get_shape(act_space, "move"),
"comm": self.get_shape(act_space, "comm")
}
else:
num_out_pol = self.get_shape(act_space)
# atoc policy
policy_kwargs = dict(
hidden_dim=hidden_dim,
norm_in=norm_in,
constrain_out=constrain_out,
discrete_action=self.discrete_action,
rnn_policy=rnn_policy,
thought_dim=thought_dim
)
self.policy = ATOCPolicy(num_in_pol, num_out_pol, **policy_kwargs)
self.target_policy = ATOCPolicy(num_in_pol, num_out_pol, **policy_kwargs)
hard_update(self.target_policy, self.policy)
# Critic
self.rnn_critic = rnn_critic
self.critic_hidden_states = None
if algo_type == "MADDPG":
num_in_critic = 0
for oobsp in env_obs_space:
num_in_critic += oobsp.shape[0]
for oacsp in env_act_space:
# feed all acts to centralized critic
num_in_critic += self.get_shape(oacsp)
else: # only DDPG, local critic
num_in_critic = obs_space.shape[0] + self.get_shape(act_space)
critic_net_fn = RecurrentNetwork if rnn_critic else MLPNetwork
critic_kwargs = dict(
hidden_dim=hidden_dim,
norm_in=norm_in,
constrain_out=constrain_out
)
self.critic = critic_net_fn(num_in_critic, 1, **critic_kwargs)
self.target_critic = critic_net_fn(num_in_critic, 1, **critic_kwargs)
hard_update(self.target_critic, self.critic)
# NOTE: atoc modules
# attention unit, MLP (used here) or RNN, output comm probability
self.thought_dim = thought_dim
self.attention_unit = nn.Sequential(
MLPNetwork(thought_dim, 1, hidden_dim=hidden_dim,
norm_in=norm_in, constrain_out=False),
nn.Sigmoid()
)
# communication channel, bi-LSTM (used here) or graph
self.comm_channel = nn.LSTM(thought_dim, thought_dim, 1,
batch_first=False, bidirectional=True)
# Optimizers
self.policy_optimizer = Adam(self.policy.parameters(), lr=lr)
self.critic_optimizer = Adam(self.critic.parameters(), lr=lr)
self.attention_unit_optimizer = Adam(self.attention_unit.parameters(), lr=lr)
self.comm_channel_optimizer = Adam(self.comm_channel.parameters(), lr=lr)
class ATOCAgent(object):
"""
General class for DDPG agents (policy, critic, target policy, target
critic, exploration noise)
"""
def __init__(self, algo_type="MADDPG", act_space=None, obs_space=None,
rnn_policy=False, rnn_critic=False, hidden_dim=64, lr=0.01,
norm_in=False, constrain_out=False, thought_dim=64,
env_obs_space=None, env_act_space=None, **kwargs):
"""
Inputs:
act_space: single agent action space (single space or Dict)
obs_space: single agent observation space (single space Dict)
"""
self.algo_type = algo_type
self.act_space = act_space
self.obs_space = obs_space
# continuous or discrete action (only look at `move` action, assume
# move and comm space both discrete or continuous)
tmp = act_space.spaces["move"] if isinstance(act_space, Dict) else act_space
self.discrete_action = False if isinstance(tmp, Box) else True
# Exploration noise
if not self.discrete_action:
# `move`, `comm` share same continuous noise source
self.exploration = OUNoise(self.get_shape(act_space))
else:
self.exploration = 0.3 # epsilon for eps-greedy
# Policy (supports multiple outputs)
self.rnn_policy = rnn_policy
self.policy_hidden_states = None
num_in_pol = obs_space.shape[0]
if isinstance(act_space, Dict):
# hard specify now, could generalize later
num_out_pol = {
"move": self.get_shape(act_space, "move"),
"comm": self.get_shape(act_space, "comm")
}
else:
num_out_pol = self.get_shape(act_space)
# atoc policy
policy_kwargs = dict(
hidden_dim=hidden_dim,
norm_in=norm_in,
constrain_out=constrain_out,
discrete_action=self.discrete_action,
rnn_policy=rnn_policy,
thought_dim=thought_dim
)
self.policy = ATOCPolicy(num_in_pol, num_out_pol, **policy_kwargs)
self.target_policy = ATOCPolicy(num_in_pol, num_out_pol, **policy_kwargs)
hard_update(self.target_policy, self.policy)
# Critic
self.rnn_critic = rnn_critic
self.critic_hidden_states = None
if algo_type == "MADDPG":
num_in_critic = 0
for oobsp in env_obs_space:
num_in_critic += oobsp.shape[0]
for oacsp in env_act_space:
# feed all acts to centralized critic
num_in_critic += self.get_shape(oacsp)
else: # only DDPG, local critic
num_in_critic = obs_space.shape[0] + self.get_shape(act_space)
critic_net_fn = RecurrentNetwork if rnn_critic else MLPNetwork
critic_kwargs = dict(
hidden_dim=hidden_dim,
norm_in=norm_in,
constrain_out=constrain_out
)
self.critic = critic_net_fn(num_in_critic, 1, **critic_kwargs)
self.target_critic = critic_net_fn(num_in_critic, 1, **critic_kwargs)
hard_update(self.target_critic, self.critic)
# NOTE: atoc modules
# attention unit, MLP (used here) or RNN, output comm probability
self.thought_dim = thought_dim
self.attention_unit = nn.Sequential(
MLPNetwork(thought_dim, 1, hidden_dim=hidden_dim,
norm_in=norm_in, constrain_out=False),
nn.Sigmoid()
)
# communication channel, bi-LSTM (used here) or graph
self.comm_channel = nn.LSTM(thought_dim, thought_dim, 1,
batch_first=False, bidirectional=True)
# Optimizers
self.policy_optimizer = Adam(self.policy.parameters(), lr=lr)
self.critic_optimizer = Adam(self.critic.parameters(), lr=lr)
self.attention_unit_optimizer = Adam(self.attention_unit.parameters(), lr=lr)
self.comm_channel_optimizer = Adam(self.comm_channel.parameters(), lr=lr)
def get_shape(self, x, key=None):
""" func to infer action output shape """
if isinstance(x, Dict):
if key is None: # sum of action space dims
return sum([
x.spaces[k].n if self.discrete_action else x.spaces[k].shape[0]
for k in x.spaces
])
elif key in x.spaces:
return x.spaces[key].n if self.discrete_action else x.spaces[key].shape[0]
else: # key not in action spaces
return 0
else:
return x.n if self.discrete_action else x.shape[0]
def reset_noise(self):
if not self.discrete_action:
self.exploration.reset()
def scale_noise(self, scale):
if self.discrete_action:
self.exploration = scale
else:
self.exploration.scale = scale
def init_hidden(self, batch_size):
# (1,H) -> (B,H)
# policy.init_hidden().unsqueeze(0).expand(batch_size, self.n_agents, -1)
if self.rnn_policy:
self.policy_hidden_states = self.policy.init_hidden().expand(batch_size, -1)
if self.rnn_critic:
self.critic_hidden_states = self.critic.init_hidden().expand(batch_size, -1)
def compute_value(self, vf_in, h_critic=None, target=False, truncate_steps=-1):
""" training critic forward with specified policy
Arguments:
vf_in: (B,T,K)
target: if use target network
truncate_steps: number of BPTT steps to truncate if used
Returns:
q: (B*T,1)
"""
bs, ts, _ = vf_in.shape
critic = self.target_critic if target else self.critic
if self.rnn_critic:
if h_critic is None:
h_t = self.critic_hidden_states.clone() # (B,H)
else:
h_t = h_critic #.clone()
# rollout
q = rnn_forward_sequence(
critic, vf_in, h_t, truncate_steps=truncate_steps)
# q = [] # (B,1)*T
# for t in range(ts):
# q_t, h_t = critic(vf_in[:,t], h_t)
# q.append(q_t)
q = torch.stack(q, 0).permute(1,0,2) # (T,B,1) -> (B,T,1)
q = q.reshape(bs*ts, -1) # (B*T,1)
else:
# (B,T,D) -> (B*T,1)
q, _ = critic(vf_in.reshape(bs*ts, -1))
return q
def _soft_act(self, x, requires_grad=True):
""" soften action if discrete, x: (B,A) """
if not self.discrete_action:
return x
if requires_grad:
return gumbel_softmax(x, hard=True)
else:
return onehot_from_logits(x)
def compute_action(self, obs, h_actor=None, target=False, requires_grad=True, truncate_steps=-1):
""" traininsg actor forward with specified policy
concat all actions to be fed in critics
Arguments:
obs: (B,T,O)
target: if use target network
requires_grad: if use _soft_act to differentiate discrete action
Returns:
act: dict of (B,T,A)
"""
bs, ts, _ = obs.shape
pi = self.target_policy if target else self.policy
if self.rnn_policy:
if h_actor is None:
h_t = self.policy_hidden_states.clone() # (B,H)
else:
h_t = h_actor #.clone()
# rollout
seq_logits = rnn_forward_sequence(
pi, obs, h_t, truncate_steps=truncate_steps)
# seq_logits = []
# for t in range(ts):
# act_t, h_t = pi(obs[:,t], h_t) # act_t is dict (B,A)
# seq_logits.append(act_t)
# soften deterministic output for backprop
act = defaultdict(list)
for act_t in seq_logits:
for k, a in act_t.items():
act[k].append(self._soft_act(a, requires_grad))
act = {
k: torch.stack(ac, 0).permute(1,0,2)
for k, ac in act.items()
} # dict [(B,A)]*T -> dict (B,T,A)
else:
stacked_obs = obs.reshape(bs*ts, -1) # (B*T,O)
act, _ = pi(stacked_obs) # act is dict of (B*T,A)
act = {
k: self._soft_act(ac, requires_grad).reshape(bs, ts, -1)
for k, ac in act.items()
} # dict of (B,T,A)
return act
def step_thought(self, obs):
""" run the first part of policy to generate thought
Arguments:
obs: (B,O)
Returns:
thought: (B,H)
"""
with torch.no_grad():
thought, hidden_states = self.policy.get_thought(obs, self.policy_hidden_states)
self.policy_hidden_states = hidden_states # if mlp, still defafult None
return thought
def step_action(self, h_i, h_j=None, explore=False):
"""
Take a step forward in environment for a minibatch of observations
equivalent to `act` or `compute_actions`
Arguments:
h_i, h_j: (B,H)
explore: Whether or not to add exploration noise
Returns:
action: dict of actions for this agent, (B,A)
"""
with torch.no_grad():
action = self.policy.get_action(h_i, h_j)
if self.discrete_action:
for k in action:
if explore:
action[k] = gumbel_softmax(action[k], hard=True)
else:
action[k] = onehot_from_logits(action[k])
else: # continuous action
idx = 0
noise = Variable(Tensor(self.exploration.noise()),
requires_grad=False)
for k in action:
if explore:
dim = action[k].shape[-1]
action[k] += noise[idx : idx+dim]
idx += dim
action[k] = action[k].clamp(-1, 1)
return action
def initiate_comm(self, thought_vec, binary=False):
""" output probability if to initiate communication
if binary, return 0 or 1s (used in sampling)
Arguments:
- thought_vec: (B,H)
Returns:
- is_comm: (B,1)
"""
is_comm = self.attention_unit(thought_vec)
if binary:
is_comm = (is_comm > 0.5).long()
return is_comm
def integrate_thoughts(self, thought_vecs)
writer = SummaryWriter()
else:
writer = None
env = gym.make(args.env_name)
action_dim = env.action_space.shape[0]
state_dim = env.observation_space.shape[0]
state_rms = RunningMeanStd(state_dim)
if args.algo == 'ppo':
agent = PPO(writer, device, state_dim, action_dim, agent_args)
elif args.algo == 'sac':
agent = SAC(writer, device, state_dim, action_dim, agent_args)
elif args.algo == 'ddpg':
from utils.noise import OUNoise
noise = OUNoise(action_dim, 0)
agent = DDPG(writer, device, state_dim, action_dim, agent_args, noise)
if (torch.cuda.is_available()) and (args.use_cuda):
agent = agent.cuda()
if args.load != 'no':
agent.load_state_dict(torch.load("./model_weights/" + args.load))
score_lst = []
state_lst = []
if agent_args.on_policy == True:
score = 0.0
state_ = (env.reset())
state = np.clip((state_ - state_rms.mean) / (state_rms.var**0.5 + 1e-8),