whitespace fixes

ghstack-source-id: 5432d9b11af3a2227d31f155e64e548f7142dfde
Pull Request resolved: fairinternal/xformers#1038

__original_commit__ = fairinternal/xformers@8efc885

CONTRIBUTING.md

@@ -60,14 +60,14 @@ flake8 --config .flake8
 mypy --ignore-missing-imports --scripts-are-modules --pretty --exclude build/ --exclude stubs/ .
 ```
 
-* or you can just install [pre-commit](https://pre-commit.com/), which will make sure that all of the above is run automatically anytime you commit 
-in that case, you would need to 
+* or you can just install [pre-commit](https://pre-commit.com/), which will make sure that all of the above is run automatically anytime you commit
+in that case, you would need to
 ```bash
-pip install pre-commit 
+pip install pre-commit
 ```
 then (in the xformers repository, just once)
 ```bash
-pre-commit install 
+pre-commit install
 ```
 
 After these steps each of your commits will run the same linting and formatting routines as the xformers continuous integration, which greatly helps getting your PRs all green !

examples/README.md

@@ -47,4 +47,3 @@ This is meant to be an easy introduction to using xformers in practice, mirrorin
 This is very close to the MicroViT example above, but illustrating the use of a hierarchical Transformer ([Metaformer](https://arxiv.org/pdf/2111.11418.pdf)) this time, through a helper function which generates the required configuration given the pooling parameters. The suggested configuration is about 6.6M parameters big (half of a ResNet18) and trains to about 86% top-1 Cifar10 within minutes.
 
 ![Example curves](../docs/assets/metaformer.png)
-

examples/llama_inference/README.md

@@ -7,7 +7,7 @@ Example runs:
 $ python -m generate --ckpt_dir models/CodeLlama-7b-Instruct/
 loaded SentencePiece model: #words: 32016 - bos id: 1 - eos id: 2
 loaded model in 12.36 seconds
-> [INST]abc[/INST] 
+> [INST]abc[/INST]
 I'm not sure I understand what you are saying with "abc". Could you explain?
 ---------------
 > [INST]can you write a hello world program in C#[/INST]

xformers/benchmarks/LRA/code/config.json

@@ -319,4 +319,4 @@
             }
         }
     }
-}
+}

xformers/benchmarks/LRA/code/config_nystrom.json

@@ -349,4 +349,4 @@
             }
         }
     }
-}
+}

xformers/benchmarks/LRA/code/config_orig_lra.json

@@ -328,4 +328,4 @@
             }
         }
     }
-}
+}

xformers/benchmarks/LRA/code/config_orig_lra_paper.json

@@ -320,4 +320,4 @@
             }
         }
     }
-}
+}

xformers/ops/fmha/__init__.py

@@ -144,14 +144,14 @@ def memory_efficient_attention(
 
         scale = 1.0 / query.shape[-1] ** 0.5
         query = query * scale
-        query = query.transpose(1, 2) 
+        query = query.transpose(1, 2)
         key = key.transpose(1, 2)
         value = value.transpose(1, 2)
         attn = query @ key.transpose(-2, -1)
         if attn_bias is not None:
             attn = attn + attn_bias
         attn = attn.softmax(-1)
-        attn = F.dropout(attn, p) 
+        attn = F.dropout(attn, p)
         attn = attn @ value
         return attn.transpose(1, 2)