remove simple .data from torch/nn

Summary: Pull Request resolved: pytorch#31482

Test Plan: Imported from OSS

Differential Revision: D19303185

Pulled By: albanD

fbshipit-source-id: 610eae096bab24a7b9f651b9af2e3ecd19df55b0

torch/autograd/gradcheck.py

@@ -4,12 +4,11 @@
 from itertools import product
 import warnings
 
-
 def zero_gradients(x):
     if isinstance(x, torch.Tensor):
         if x.grad is not None:
             x.grad.detach_()
-            x.grad.data.zero_()
+            x.grad.zero_()
     elif isinstance(x, container_abcs.Iterable):
         for elem in x:
             zero_gradients(elem)
@@ -63,8 +62,6 @@ def get_numerical_jacobian(fn, input, target=None, eps=1e-3):
 
     # TODO: compare structure
     for x_tensor, d_tensor in zip(x_tensors, j_tensors):
-        # need data here to get around the version check because without .data,
-        # the following code updates version but doesn't change content
         if x_tensor.is_sparse:
             def get_stride(size):
                 dim = len(size)
@@ -78,9 +75,12 @@ def get_stride(size):
             x_nnz = x_tensor._nnz()
             x_size = list(x_tensor.size())
             x_indices = x_tensor._indices().t()
-            x_values = x_tensor._values().data
+            x_values = x_tensor._values()
             x_stride = get_stride(x_size)
 
+            # Use .data here to get around the version check
+            x_values = x_values.data
+
             for i in range(x_nnz):
                 x_value = x_values[i]
                 for x_idx in product(*[range(m) for m in x_values.size()[1:]]):
@@ -95,10 +95,11 @@ def get_stride(size):
                     r = (outb - outa) / (2 * eps)
                     d_tensor[d_idx] = r.detach().reshape(-1)
         elif x_tensor.layout == torch._mkldnn:
+            # Use .data here to get around the version check
+            x_tensor = x_tensor.data
             if len(input) != 1:
                 raise ValueError('gradcheck currently only supports functions with 1 input, but got: ',
                                  len(input))
-            x_tensor = x_tensor.data
             for d_idx, x_idx in enumerate(product(*[range(m) for m in x_tensor.size()])):
                 # this is really inefficient, but without indexing implemented, there's
                 # not really a better way than converting back and forth
@@ -116,6 +117,7 @@ def get_stride(size):
                 r = (outb - outa) / (2 * eps)
                 d_tensor[d_idx] = r.detach().reshape(-1)
         else:
+            # Use .data here to get around the version check
             x_tensor = x_tensor.data
             for d_idx, x_idx in enumerate(product(*[range(m) for m in x_tensor.size()])):
                 orig = x_tensor[x_idx].item()

torch/testing/__init__.py

@@ -58,7 +58,6 @@ def _unravel_index(index, shape):
         rtol, atol, list(index), actual[index].item(), expected[index].item(),
         count - 1, 100 * count / actual.numel()))
 
-
 def make_non_contiguous(tensor):
     if tensor.numel() <= 1:  # can't make non-contiguous
         return tensor.clone()
@@ -83,7 +82,9 @@ def make_non_contiguous(tensor):
             input = input.narrow(i, bounds, tensor.size(i))
 
     input.copy_(tensor)
-    return input
+
+    # Use .data here to hide the view relation between input and other temporary Tensors
+    return input.data
 
 
 def get_all_dtypes():