Update

[ghstack-poisoned]
pytorch · Nov 18, 2024 · 703bcac · 703bcac
2 parents 6d9a87e + 093fa97
commit 703bcac
Showing 1 changed file with 0 additions and 8 deletions.
diff --git a/torchrl/data/tensor_specs.py b/torchrl/data/tensor_specs.py
@@ -1392,7 +1392,6 @@ def type_check(self, value: torch.Tensor, key: NestedKey | None = None) -> None:
             spec.type_check(val)
 
     def is_in(self, value) -> bool:
-        raise RuntimeError
         if self.dim == 0 and not hasattr(value, "unbind"):
             # We don't use unbind because value could be a tuple or a nested tensor
             return all(
@@ -1821,7 +1820,6 @@ def _project(self, val: torch.Tensor) -> torch.Tensor:
         return val
 
     def is_in(self, val: torch.Tensor) -> bool:
-        raise RuntimeError
         if self.mask is None:
             shape = torch.broadcast_shapes(self._safe_shape, val.shape)
             shape_match = val.shape == shape
@@ -2272,7 +2270,6 @@ def _project(self, val: torch.Tensor) -> torch.Tensor:
         return val
 
     def is_in(self, val: torch.Tensor) -> bool:
-        raise RuntimeError
         val_shape = _remove_neg_shapes(tensordict.utils._shape(val))
         shape = torch.broadcast_shapes(self._safe_shape, val_shape)
         shape = list(shape)
@@ -2470,7 +2467,6 @@ def one(self, shape=None):
         )
 
     def is_in(self, val: torch.Tensor) -> bool:
-        raise RuntimeError
         shape = torch.broadcast_shapes(self._safe_shape, val.shape)
         return (
             isinstance(val, NonTensorData)
@@ -2663,7 +2659,6 @@ def rand(self, shape: torch.Size = None) -> torch.Tensor:
         return torch.empty(shape, device=self.device, dtype=self.dtype).random_()
 
     def is_in(self, val: torch.Tensor) -> bool:
-        raise RuntimeError
         shape = torch.broadcast_shapes(self._safe_shape, val.shape)
         return val.shape == shape and val.dtype == self.dtype
 
@@ -3012,7 +3007,6 @@ def index(self, index: INDEX_TYPING, tensor_to_index: torch.Tensor) -> torch.Ten
         return torch.cat(out, -1)
 
     def is_in(self, val: torch.Tensor) -> bool:
-        raise RuntimeError
         vals = self._split(val)
         if vals is None:
             return False
@@ -3358,7 +3352,6 @@ def _project(self, val: torch.Tensor) -> torch.Tensor:
         return val
 
     def is_in(self, val: torch.Tensor) -> bool:
-        raise RuntimeError
         if self.mask is None:
             shape = torch.broadcast_shapes(self._safe_shape, val.shape)
             shape_match = val.shape == shape
@@ -3984,7 +3977,6 @@ def _project(self, val: torch.Tensor) -> torch.Tensor:
         return val.squeeze(0) if val_is_scalar else val
 
     def is_in(self, val: torch.Tensor) -> bool:
-        raise RuntimeError
         if self.mask is not None:
             vals = val.unbind(-1)
             splits = self._split_self()