b900fdf6f2
Closes #947 This removes all kwargs from all policy constructors. While doing that, I also improved several names and added a whole lot of TODOs. ## Functional changes: 1. Added possibility to pass None as `critic2` and `critic2_optim`. In fact, the default behavior then should cover the absolute majority of cases 2. Added a function called `clone_optimizer` as a temporary measure to support passing `critic2_optim=None` ## Breaking changes: 1. `action_space` is no longer optional. In fact, it already was non-optional, as there was a ValueError in BasePolicy.init. So now several examples were fixed to reflect that 2. `reward_normalization` removed from DDPG and children. It was never allowed to pass it as `True` there, an error would have been raised in `compute_n_step_reward`. Now I removed it from the interface 3. renamed `critic1` and similar to `critic`, in order to have uniform interfaces. Note that the `critic` in DDPG was optional for the sole reason that child classes used `critic1`. I removed this optionality (DDPG can't do anything with `critic=None`) 4. Several renamings of fields (mostly private to public, so backwards compatible) ## Additional changes: 1. Removed type and default declaration from docstring. This kind of duplication is really not necessary 2. Policy constructors are now only called using named arguments, not a fragile mixture of positional and named as before 5. Minor beautifications in typing and code 6. Generally shortened docstrings and made them uniform across all policies (hopefully) ## Comment: With these changes, several problems in tianshou's inheritance hierarchy become more apparent. I tried highlighting them for future work. --------- Co-authored-by: Dominik Jain <d.jain@appliedai.de>
233 lines
9.3 KiB
Python
233 lines
9.3 KiB
Python
from typing import Any, Literal
|
|
|
|
import numpy as np
|
|
|
|
from tianshou.data import Batch, ReplayBuffer
|
|
from tianshou.data.batch import BatchProtocol
|
|
from tianshou.data.types import RolloutBatchProtocol
|
|
from tianshou.policy import BasePolicy
|
|
from tianshou.policy.base import TLearningRateScheduler
|
|
|
|
try:
|
|
from tianshou.env.pettingzoo_env import PettingZooEnv
|
|
except ImportError:
|
|
PettingZooEnv = None # type: ignore
|
|
|
|
|
|
class MultiAgentPolicyManager(BasePolicy):
|
|
"""Multi-agent policy manager for MARL.
|
|
|
|
This multi-agent policy manager accepts a list of
|
|
:class:`~tianshou.policy.BasePolicy`. It dispatches the batch data to each
|
|
of these policies when the "forward" is called. The same as "process_fn"
|
|
and "learn": it splits the data and feeds them to each policy. A figure in
|
|
:ref:`marl_example` can help you better understand this procedure.
|
|
|
|
:param policies: a list of policies.
|
|
:param env: a PettingZooEnv.
|
|
:param action_scaling: if True, scale the action from [-1, 1] to the range
|
|
of action_space. Only used if the action_space is continuous.
|
|
:param action_bound_method: method to bound action to range [-1, 1].
|
|
Only used if the action_space is continuous.
|
|
:param lr_scheduler: if not None, will be called in `policy.update()`.
|
|
"""
|
|
|
|
def __init__(
|
|
self,
|
|
*,
|
|
policies: list[BasePolicy],
|
|
# TODO: 1 why restrict to PettingZooEnv?
|
|
# TODO: 2 This is the only policy that takes an env in init, is it really needed?
|
|
env: PettingZooEnv,
|
|
action_scaling: bool = False,
|
|
action_bound_method: Literal["clip", "tanh"] | None = "clip",
|
|
lr_scheduler: TLearningRateScheduler | None = None,
|
|
) -> None:
|
|
super().__init__(
|
|
action_space=env.action_space,
|
|
observation_space=env.observation_space,
|
|
action_scaling=action_scaling,
|
|
action_bound_method=action_bound_method,
|
|
lr_scheduler=lr_scheduler,
|
|
)
|
|
assert len(policies) == len(env.agents), "One policy must be assigned for each agent."
|
|
|
|
self.agent_idx = env.agent_idx
|
|
for i, policy in enumerate(policies):
|
|
# agent_id 0 is reserved for the environment proxy
|
|
# (this MultiAgentPolicyManager)
|
|
policy.set_agent_id(env.agents[i])
|
|
|
|
self.policies = dict(zip(env.agents, policies, strict=True))
|
|
|
|
def replace_policy(self, policy: BasePolicy, agent_id: int) -> None:
|
|
"""Replace the "agent_id"th policy in this manager."""
|
|
policy.set_agent_id(agent_id)
|
|
self.policies[agent_id] = policy
|
|
|
|
# TODO: violates Liskov substitution principle
|
|
def process_fn( # type: ignore
|
|
self,
|
|
batch: RolloutBatchProtocol,
|
|
buffer: ReplayBuffer,
|
|
indice: np.ndarray,
|
|
) -> BatchProtocol:
|
|
"""Dispatch batch data from obs.agent_id to every policy's process_fn.
|
|
|
|
Save original multi-dimensional rew in "save_rew", set rew to the
|
|
reward of each agent during their "process_fn", and restore the
|
|
original reward afterwards.
|
|
"""
|
|
results = {}
|
|
assert isinstance(
|
|
batch.obs,
|
|
BatchProtocol,
|
|
), f"here only observations of type Batch are permitted, but got {type(batch.obs)}"
|
|
# reward can be empty Batch (after initial reset) or nparray.
|
|
has_rew = isinstance(buffer.rew, np.ndarray)
|
|
if has_rew: # save the original reward in save_rew
|
|
# Since we do not override buffer.__setattr__, here we use _meta to
|
|
# change buffer.rew, otherwise buffer.rew = Batch() has no effect.
|
|
save_rew, buffer._meta.rew = buffer.rew, Batch() # type: ignore
|
|
for agent, policy in self.policies.items():
|
|
agent_index = np.nonzero(batch.obs.agent_id == agent)[0]
|
|
if len(agent_index) == 0:
|
|
results[agent] = Batch()
|
|
continue
|
|
tmp_batch, tmp_indice = batch[agent_index], indice[agent_index]
|
|
if has_rew:
|
|
tmp_batch.rew = tmp_batch.rew[:, self.agent_idx[agent]]
|
|
buffer._meta.rew = save_rew[:, self.agent_idx[agent]]
|
|
if not hasattr(tmp_batch.obs, "mask"):
|
|
if hasattr(tmp_batch.obs, "obs"):
|
|
tmp_batch.obs = tmp_batch.obs.obs
|
|
if hasattr(tmp_batch.obs_next, "obs"):
|
|
tmp_batch.obs_next = tmp_batch.obs_next.obs
|
|
results[agent] = policy.process_fn(tmp_batch, buffer, tmp_indice)
|
|
if has_rew: # restore from save_rew
|
|
buffer._meta.rew = save_rew
|
|
return Batch(results)
|
|
|
|
def exploration_noise(
|
|
self,
|
|
act: np.ndarray | BatchProtocol,
|
|
batch: RolloutBatchProtocol,
|
|
) -> np.ndarray | BatchProtocol:
|
|
"""Add exploration noise from sub-policy onto act."""
|
|
assert isinstance(
|
|
batch.obs,
|
|
BatchProtocol,
|
|
), f"here only observations of type Batch are permitted, but got {type(batch.obs)}"
|
|
for agent_id, policy in self.policies.items():
|
|
agent_index = np.nonzero(batch.obs.agent_id == agent_id)[0]
|
|
if len(agent_index) == 0:
|
|
continue
|
|
act[agent_index] = policy.exploration_noise(act[agent_index], batch[agent_index])
|
|
return act
|
|
|
|
def forward( # type: ignore
|
|
self,
|
|
batch: Batch,
|
|
state: dict | Batch | None = None,
|
|
**kwargs: Any,
|
|
) -> Batch:
|
|
"""Dispatch batch data from obs.agent_id to every policy's forward.
|
|
|
|
:param state: if None, it means all agents have no state. If not
|
|
None, it should contain keys of "agent_1", "agent_2", ...
|
|
|
|
:return: a Batch with the following contents:
|
|
|
|
::
|
|
|
|
{
|
|
"act": actions corresponding to the input
|
|
"state": {
|
|
"agent_1": output state of agent_1's policy for the state
|
|
"agent_2": xxx
|
|
...
|
|
"agent_n": xxx}
|
|
"out": {
|
|
"agent_1": output of agent_1's policy for the input
|
|
"agent_2": xxx
|
|
...
|
|
"agent_n": xxx}
|
|
}
|
|
"""
|
|
results: list[tuple[bool, np.ndarray, Batch, np.ndarray | Batch, Batch]] = []
|
|
for agent_id, policy in self.policies.items():
|
|
# This part of code is difficult to understand.
|
|
# Let's follow an example with two agents
|
|
# batch.obs.agent_id is [1, 2, 1, 2, 1, 2] (with batch_size == 6)
|
|
# each agent plays for three transitions
|
|
# agent_index for agent 1 is [0, 2, 4]
|
|
# agent_index for agent 2 is [1, 3, 5]
|
|
# we separate the transition of each agent according to agent_id
|
|
agent_index = np.nonzero(batch.obs.agent_id == agent_id)[0]
|
|
if len(agent_index) == 0:
|
|
# (has_data, agent_index, out, act, state)
|
|
results.append((False, np.array([-1]), Batch(), Batch(), Batch()))
|
|
continue
|
|
tmp_batch = batch[agent_index]
|
|
if isinstance(tmp_batch.rew, np.ndarray):
|
|
# reward can be empty Batch (after initial reset) or nparray.
|
|
tmp_batch.rew = tmp_batch.rew[:, self.agent_idx[agent_id]]
|
|
if not hasattr(tmp_batch.obs, "mask"):
|
|
if hasattr(tmp_batch.obs, "obs"):
|
|
tmp_batch.obs = tmp_batch.obs.obs
|
|
if hasattr(tmp_batch.obs_next, "obs"):
|
|
tmp_batch.obs_next = tmp_batch.obs_next.obs
|
|
out = policy(
|
|
batch=tmp_batch,
|
|
state=None if state is None else state[agent_id],
|
|
**kwargs,
|
|
)
|
|
act = out.act
|
|
each_state = out.state if (hasattr(out, "state") and out.state is not None) else Batch()
|
|
results.append((True, agent_index, out, act, each_state))
|
|
holder: Batch = Batch.cat(
|
|
[{"act": act} for (has_data, agent_index, out, act, each_state) in results if has_data],
|
|
)
|
|
state_dict, out_dict = {}, {}
|
|
for (agent_id, _), (has_data, agent_index, out, act, state) in zip(
|
|
self.policies.items(),
|
|
results,
|
|
strict=True,
|
|
):
|
|
if has_data:
|
|
holder.act[agent_index] = act
|
|
state_dict[agent_id] = state
|
|
out_dict[agent_id] = out
|
|
holder["out"] = out_dict
|
|
holder["state"] = state_dict
|
|
return holder
|
|
|
|
def learn(
|
|
self,
|
|
batch: RolloutBatchProtocol,
|
|
*args: Any,
|
|
**kwargs: Any,
|
|
) -> dict[str, float | list[float]]:
|
|
"""Dispatch the data to all policies for learning.
|
|
|
|
:return: a dict with the following contents:
|
|
|
|
::
|
|
|
|
{
|
|
"agent_1/item1": item 1 of agent_1's policy.learn output
|
|
"agent_1/item2": item 2 of agent_1's policy.learn output
|
|
"agent_2/xxx": xxx
|
|
...
|
|
"agent_n/xxx": xxx
|
|
}
|
|
"""
|
|
results = {}
|
|
for agent_id, policy in self.policies.items():
|
|
data = batch[agent_id]
|
|
if not data.is_empty():
|
|
out = policy.learn(batch=data, **kwargs)
|
|
for k, v in out.items():
|
|
results[agent_id + "/" + k] = v
|
|
return results
|