def __init__(self, *args, td3_kwargs, her_kwargs, base_kwargs, **kwargs):
HERTrainer.__init__(
self,
**her_kwargs,
)
TD3.__init__(self, *args, **kwargs, **td3_kwargs, **base_kwargs)
assert isinstance(self.replay_buffer, ObsDictRelabelingBuffer)
def __init__(self,
env,
qf1,
qf2,
exploration_policy,
td3_kwargs,
tdm_kwargs,
base_kwargs,
policy=None,
eval_policy=None,
replay_buffer=None,
optimizer_class=optim.Adam,
use_policy_saturation_cost=False,
**kwargs):
TD3.__init__(self,
env=env,
qf1=qf1,
qf2=qf2,
policy=policy,
exploration_policy=exploration_policy,
replay_buffer=replay_buffer,
eval_policy=eval_policy,
optimizer_class=optimizer_class,
**td3_kwargs,
**base_kwargs)
super().__init__(**tdm_kwargs)
self.use_policy_saturation_cost = use_policy_saturation_cost
def __init__(
self,
*args,
td3_kwargs,
her_kwargs,
base_kwargs,
**kwargs
):
HER.__init__(
self,
**her_kwargs,
)
TD3.__init__(self, *args, **kwargs, **td3_kwargs, **base_kwargs)
assert isinstance(
self.replay_buffer, SimpleHerReplayBuffer
) or isinstance(
self.replay_buffer, RelabelingReplayBuffer
) or isinstance(
self.replay_buffer, ObsDictRelabelingBuffer
)