Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[BugFix] QValue modules and nested action #1351

Merged
merged 13 commits into from
Jul 5, 2023
Prev Previous commit
Next Next commit
typos
Signed-off-by: Matteo Bettini <matbet@meta.com>
  • Loading branch information
matteobettini committed Jul 4, 2023
commit 4b7ce866f17da1ba5166fee3cf13ba76df149e53
8 changes: 4 additions & 4 deletions torchrl/modules/tensordict_module/actors.py
Original file line number Diff line number Diff line change
Expand Up @@ -323,7 +323,7 @@ class QValueModule(TensorDictModuleBase):
Args:
action_space (str, optional): Action space. Must be one of
``"one-hot"``, ``"mult-one-hot"``, ``"binary"`` or ``"categorical"``.
This is argumets is exclusive with ``spec``, since ``spec``
This argument is exclusive with ``spec``, since ``spec``
conditions the action_space.
action_value_key (str or tuple of str, optional): The input key
representing the action value. Defaults to ``"action_value"``.
Expand Down Expand Up @@ -517,7 +517,7 @@ class DistributionalQValueModule(QValueModule):
Args:
action_space (str, optional): Action space. Must be one of
``"one-hot"``, ``"mult-one-hot"``, ``"binary"`` or ``"categorical"``.
This is argumets is exclusive with ``spec``, since ``spec``
This argument is exclusive with ``spec``, since ``spec``
conditions the action_space.
support (torch.Tensor): support of the action values.
action_value_key (str or tuple of str, optional): The input key
Expand Down Expand Up @@ -925,7 +925,7 @@ class QValueActor(SafeSequential):
method. Default is ``False``.
action_space (str, optional): Action space. Must be one of
``"one-hot"``, ``"mult-one-hot"``, ``"binary"`` or ``"categorical"``.
This is argumets is exclusive with ``spec``, since ``spec``
This argument is exclusive with ``spec``, since ``spec``
conditions the action_space.
action_value_key (str or tuple of str, optional): if the input module
is a :class:`tensordict.nn.TensorDictModuleBase` instance, it must
Expand Down Expand Up @@ -1077,7 +1077,7 @@ class DistributionalQValueActor(QValueActor):
support (torch.Tensor): support of the action values.
action_space (str, optional): Action space. Must be one of
``"one-hot"``, ``"mult-one-hot"``, ``"binary"`` or ``"categorical"``.
This is argumets is exclusive with ``spec``, since ``spec``
This argument is exclusive with ``spec``, since ``spec``
conditions the action_space.
make_log_softmax (bool, optional): if ``True`` and if the module is not
of type :class:`torchrl.modules.DistributionalDQNnet`, a log-softmax
Expand Down