[doc] Fix typo under colossalai and doc(#3618)

* Fixed several spelling errors under colossalai

* Fix the spelling error in colossalai and docs directory

* Cautious Changed the spelling error under the example folder

* Update runtime_preparation_pass.py

revert autograft to autograd

* Update search_chunk.py

utile to until

* Update check_installation.py

change misteach to mismatch in line 91

* Update 1D_tensor_parallel.md

revert to perceptron

* Update 2D_tensor_parallel.md

revert to perceptron in line 73

* Update 2p5D_tensor_parallel.md

revert to perceptron in line 71

* Update 3D_tensor_parallel.md

revert to perceptron in line 80

* Update README.md

revert to resnet in line 42

* Update reorder_graph.py

revert to indice in line 7

* Update p2p.py

revert to megatron in line 94

* Update initialize.py

revert to torchrun in line 198

* Update routers.py

change to detailed in line 63

* Update routers.py

change to detailed in line 146

* Update README.md

revert  random number in line 402

colossalai/auto_parallel/offload/region.py

@@ -111,7 +111,7 @@ class Region:
         Copy data slice to the memory space indexed by the input tensor in the region.
 
         Args:
-            param (torch.nn.Parameter): the param used to retrive meta information
+            param (torch.nn.Parameter): the param used to retrieve meta information
             data_slice (torch.Tensor): the tensor to be copied to the region
         """
 

colossalai/auto_parallel/offload/training_simulator.py

@@ -22,7 +22,7 @@ class TrainingSimulator(ABC):
 
     Args:
         region_list (List[Region]): represents the linearized DNN computing graph.
-        comp_power (float): the NVIDIA GPU FP16 compuing power.
+        comp_power (float): the NVIDIA GPU FP16 computing power.
         link_to_bw (Dict[str, Dict[float, float]]): communication links and the corresponding bandwidth.
     """
 

colossalai/auto_parallel/passes/runtime_preparation_pass.py

@@ -149,7 +149,7 @@ def size_value_converting_pass(gm: torch.fx.GraphModule, device_mesh: DeviceMesh
 
     def _extract_target_dim(node):
         '''
-        A helper function to etract the target dimension from size node.
+        A helper function to extract the target dimension from size node.
         There are two usages of torch.Tensor.size:
         1. tensor.size()
         2. tensor.size(dim)
@@ -427,7 +427,7 @@ def module_params_sharding_pass(gm: torch.fx.GraphModule, device_mesh: DeviceMes
         if target_sharding_spec.dim_partition_dict != {}:
             origin_sharding_spec = ShardingSpec(device_mesh, param.shape, {})
             setattr(param, 'sharding_spec', origin_sharding_spec)
-            # TODO: build a ColoParamter class to manager the distributed parameters
+            # TODO: build a ColoParameter class to manager the distributed parameters
             # we could use .data here, because all the operations just happen before the real training
             # loop, so we don't need to track these operations in the autograd graph.
             param = torch.nn.Parameter(