lint fix: raise from None

Summary: New linter warning is complaining about `raise` inside `except`.

Reviewed By: kjchalup

Differential Revision: D37819264

fbshipit-source-id: 56ad5d0558ea39e1125f3c76b43b7376aea2bc7c

pytorch3d/datasets/r2n2/utils.py

@@ -276,7 +276,7 @@ def _read_binvox_header(f):  # pragma: no cover
     try:
         dims = [int(d) for d in dims[1:]]
     except ValueError:
-        raise ValueError("Invalid header (line 2)")
+        raise ValueError("Invalid header (line 2)") from None
     if len(dims) != 3 or dims[0] != dims[1] or dims[0] != dims[2]:
         raise ValueError("Invalid header (line 2)")
     size = dims[0]
@@ -291,7 +291,7 @@ def _read_binvox_header(f):  # pragma: no cover
     try:
         translation = tuple(float(t) for t in translation[1:])
     except ValueError:
-        raise ValueError("Invalid header (line 3)")
+        raise ValueError("Invalid header (line 3)") from None
 
     # Fourth line of the header should be "scale [float]"
     line = f.readline().strip()

pytorch3d/implicitron/models/global_encoder/autodecoder.py

@@ -107,7 +107,7 @@ class Autodecoder(Configurable, torch.nn.Module):
                     device=next(self.parameters()).device,
                 )
             except StopIteration:
-                raise ValueError("Not enough n_instances in the autodecoder")
+                raise ValueError("Not enough n_instances in the autodecoder") from None
 
         # pyre-fixme[29]: `Union[torch.Tensor, torch.nn.Module]` is not a function.
         return self._autodecoder_codes(x)

pytorch3d/implicitron/tools/stats.py

@@ -238,7 +238,7 @@ class Stats(object):
                             "could not extract prediction %s\
                                           from the prediction dictionary"
                             % stat
-                        )
+                        ) from None
                 else:
                     val = None
 

pytorch3d/io/off_io.py

@@ -85,7 +85,7 @@ def _read_faces_lump(
         if n_faces > 1 and "Wrong number of columns" in e.args[0]:
             file.seek(old_offset)
             return None
-        raise ValueError("Not enough face data.")
+        raise ValueError("Not enough face data.") from None
 
     if len(data) != n_faces:
         raise ValueError("Not enough face data.")
@@ -247,11 +247,11 @@ def _load_off_stream(file) -> dict:
     try:
         n_verts = int(items[0])
     except ValueError:
-        raise ValueError("Invalid counts line: %s" % header)
+        raise ValueError("Invalid counts line: %s" % header) from None
     try:
         n_faces = int(items[1])
     except ValueError:
-        raise ValueError("Invalid counts line: %s" % header)
+        raise ValueError("Invalid counts line: %s" % header) from None
 
     if (len(items) > 3 and not items[3].startswith(b"#")) or n_verts < 0 or n_faces < 0:
         raise ValueError("Invalid counts line: %s" % header)

pytorch3d/io/ply_io.py

@@ -236,7 +236,7 @@ class _PlyHeader:
             count = int(items[2])
         except ValueError:
             msg = "Number of items for %s was not a number."
-            raise ValueError(msg % items[1])
+            raise ValueError(msg % items[1]) from None
         self.elements.append(_PlyElementType(items[1], count))
 
 
@@ -409,12 +409,12 @@ def _parse_heterogeneous_property_ascii(datum, line_iter, property: _Property):
             else:
                 datum.append(int(value))
         except ValueError:
-            raise ValueError("Bad numerical data.")
+            raise ValueError("Bad numerical data.") from None
     else:
         try:
             length = int(value)
         except ValueError:
-            raise ValueError("A list length was not a number.")
+            raise ValueError("A list length was not a number.") from None
         list_value = np.zeros(length, dtype=_PLY_TYPES[property.data_type].np_type)
         for i in range(length):
             inner_value = next(line_iter, None)
@@ -423,7 +423,7 @@ def _parse_heterogeneous_property_ascii(datum, line_iter, property: _Property):
             try:
                 list_value[i] = float(inner_value)
             except ValueError:
-                raise ValueError("Bad numerical data.")
+                raise ValueError("Bad numerical data.") from None
         datum.append(list_value)
 
 

pytorch3d/ops/points_alignment.py

@@ -141,7 +141,7 @@ def iterative_closest_point(
                 "(minibatch, dim, dim), T is a batch of dim-dimensional "
                 "translations of shape (minibatch, dim) and s is a batch "
                 "of scalars of shape (minibatch,)."
-            )
+            ) from None
         # apply the init transform to the input point cloud
         Xt = _apply_similarity_transform(Xt, R, T, s)
     else: