Update 2021-10-26-accelerating-pytorch-with-cuda-graphs.md

Author: pytorchsam

_posts/2021-10-26-accelerating-pytorch-with-cuda-graphs.md

@@ -62,7 +62,7 @@ You should try CUDA graphs if all or part of your network is graph-safe (usually
 
 ### API example
 
-PyTorch exposes graphs via a raw [`torch.cuda.CUDAGraph`](https://pytorch.org/docs/master/generated/torch.cuda.graph.html#torch.cuda.graph)class and two convenience wrappers, [`torch.cuda.graph`](https://pytorch.org/docs/master/generated/torch.cuda.CUDAGraph.html#torch.cuda.CUDAGraph) and [`torch.cuda.make_graphed_callables`](https://pytorch.org/docs/master/generated/torch.cuda.make_graphed_callables.html#torch.cuda.make_graphed_callables).
+PyTorch exposes graphs via a raw [`torch.cuda.CUDAGraph`](https://pytorch.org/docs/master/generated/torch.cuda.graph.html#torch.cuda.graph)class and two convenience wrappers, [`torch.cuda.graph`](https://pytorch.org/docs/master/generated/torch.cuda.graph.html#torch.cuda.graph) and [`torch.cuda.make_graphed_callables`](https://pytorch.org/docs/master/generated/torch.cuda.make_graphed_callables.html#torch.cuda.make_graphed_callables).
 
 ```python
 N, D_in, H, D_out = 640, 4096, 2048, 1024
@@ -349,4 +349,4 @@ Alban studied engineering and did a PhD in Machine Learning and Optimization, du
 [**Edward Yang**](mailto:ezyang@fb.com)
 *Research Engineer, Facebook*
 
-Edward studied CS at MIT and then Stanford before starting at Facebook. He is a part of the PyTorch core team and is one of the leading contributors to PyTorch. 
+Edward studied CS at MIT and then Stanford before starting at Facebook. He is a part of the PyTorch core team and is one of the leading contributors to PyTorch.