[Docs] Fix typos (#2124)

docs/en/understand_mmcv/cnn.md

@@ -271,7 +271,7 @@ Let us introduce the usage of `initialize` in detail.
           self.cls = nn.Sequential(nn.Conv1d(3, 1, 3), nn.Linear(1,2))
 
   # if we would like to initialize model's weights as 1 and bias as 2
-  # but weight in `cls` as 3 and bias 4, we can use override key
+  # but weight in `reg` as 3 and bias 4, we can use override key
   model = FooNet()
   init_cfg = dict(type='Constant', layer=['Conv1d','Conv2d'], val=1, bias=2,
                   override=dict(type='Constant', name='reg', val=3, bias=4))

docs/zh_cn/understand_mmcv/cnn.md

@@ -259,7 +259,7 @@ conv = ConvModule(
           self.cls = nn.Sequential(nn.Conv1d(3, 1, 3), nn.Linear(1,2))
 
   # 如果我们想将模型的权重初始化为 1，将偏差初始化为 2
-  # 但希望 `cls` 中的权重为 3，偏差为 4，则我们可以使用关键字override
+  # 但希望 `reg` 中的权重为 3，偏差为 4，则我们可以使用关键字override
 
   model = FooNet()
   init_cfg = dict(type='Constant', layer=['Conv1d','Conv2d'], val=1, bias=2,