2021-01-15 23:40:54

new/dl-pt-workshop/1.md

@@ -181,7 +181,7 @@ example_2 = torch.randint(low=0, high=2, \
     tensor_1 = torch.tensor([0.1,1,0.9,0.7,0.3])
     tensor_2 = torch.tensor([[0,0.2,0.4,0.6],[1,0.8,0.6,0.4]])
     tensor_3 = torch.tensor([[[0.3,0.6],[1,0]], \
-                             [[0.3,0.6],[0,1]]])
+                             [[0.3,0.6],[0,1]]])
     ```
 
     如果您的计算机具有可用的 GPU，则可以使用 GPU 语法创建等效张量：
@@ -189,9 +189,9 @@ example_2 = torch.randint(low=0, high=2, \
     ```py
     tensor_1 = torch.tensor([0.1,1,0.9,0.7,0.3]).cuda()
     tensor_2 = torch.tensor([[0,0.2,0.4,0.6], \
-                        [1,0.8,0.6,0.4]]).cuda()
+                             [1,0.8,0.6,0.4]]).cuda()
     tensor_3 = torch.tensor([[[0.3,0.6],[1,0]], \
-                         [[0.3,0.6],[0,1]]]).cuda()
+                             [[0.3,0.6],[0,1]]]).cuda()
     ```
 
 3.  Print the shape of each of the tensors using the **shape** property, just as you would do with NumPy arrays: