Applied review comments

pytorch · Oct 15, 2021 · 27f53b3 · 27f53b3
1 parent 5d645e1
commit 27f53b3
Show file tree

Hide file tree

Showing 4 changed files with 6 additions and 4 deletions.
diff --git a/test/test_nn.py b/test/test_nn.py
@@ -10533,6 +10533,8 @@ def test_upsamplingLinear1d(self):
                     input = torch.randn(1, 1, 2, requires_grad=True)
                     if not recompute_scale_factor:
                         gradcheck(lambda x: F.interpolate(x, out_size, **kwargs), (input,))
+                    else:
+                        gradcheck(lambda x: F.interpolate(x, scale_factor=scale_factor, **kwargs), (input,))
 
     def test_upsamplingLinear1d_spatial_invariance(self):
         m = nn.Upsample(scale_factor=3, mode='linear', align_corners=False)

diff --git a/torch/nn/functional.py b/torch/nn/functional.py
@@ -3627,7 +3627,7 @@ def interpolate(input: Tensor, size: Optional[int] = None, scale_factor: Optiona
             output `size`. The computed output `size` will be used to infer new scales for
             the interpolation. Note that when `scale_factor` is floating-point, it may differ
             from the recomputed `scale_factor` due to rounding and precision issues.
-            If `recomputed_scale_factor` is ``False``, then `size` or `scale_factor` will
+            If `recompute_scale_factor` is ``False``, then `size` or `scale_factor` will
             be used directly for interpolation.
 
     .. note::

diff --git a/torch/nn/functional.pyi.in b/torch/nn/functional.pyi.in
@@ -305,7 +305,7 @@ def upsample(input: Any, size: Optional[Any] = ..., scale_factor: Optional[Any]
 
 
 def interpolate(input: Any, size: Optional[Any] = ..., scale_factor: Optional[Any] = ..., mode: str = ...,
-                align_corners: Optional[Any] = ...): ...
+                align_corners: Optional[Any] = ..., recompute_scale_factor: Optional[Any] = ...): ...
 
 
 def upsample_nearest(input: Any, size: Optional[Any] = ..., scale_factor: Optional[Any] = ...): ...

diff --git a/torch/nn/modules/upsampling.py b/torch/nn/modules/upsampling.py
@@ -38,7 +38,7 @@ class Upsample(Module):
             output `size`. The computed output `size` will be used to infer new scales for
             the interpolation. Note that when `scale_factor` is floating-point, it may differ
             from the recomputed `scale_factor` due to rounding and precision issues.
-            If `recomputed_scale_factor` is ``False``, then `size` or `scale_factor` will
+            If `recompute_scale_factor` is ``False``, then `size` or `scale_factor` will
             be used directly for interpolation.
 
     Shape:
@@ -126,7 +126,7 @@ class Upsample(Module):
                   [ 1.2000,  1.3600,  1.5200,  1.2800,  0.6400,  0.0000],
                   [ 0.0000,  0.0000,  0.0000,  0.0000,  0.0000,  0.0000]]]])
     """
-    __constants__ = ['size', 'scale_factor', 'mode', 'align_corners', 'name']
+    __constants__ = ['size', 'scale_factor', 'mode', 'align_corners', 'name', 'recompute_scale_factor']
     name: str
     size: Optional[_size_any_t]
     scale_factor: Optional[_ratio_any_t]