noqa

.github/workflows/python.yml

@@ -25,7 +25,7 @@ jobs:
       - name: Lint with ruff
         run: |
           # stop the build if there are Python syntax errors or undefined names
-          ruff --output-format=github --select=E9,F63,F7,F82 --ignore=E722 --target-version=py37 .
+          ruff --output-format=github --select=E9,F63,F7,F82 --target-version=py37 .
           # default set of ruff rules with GitHub Annotations
           ruff --output-format=github --target-version=py37 .
       - name: Test with pytest

adapters/tests/test_encoder_decoder.py

@@ -4,7 +4,7 @@
 import adapters
 from hf_transformers.tests.models.encoder_decoder.test_modeling_encoder_decoder import *  # Imported to execute model tests
 from transformers import AutoModelForSeq2SeqLM, AutoTokenizer, BertConfig
-from transformers.testing_utils import require_torch, torch_device
+from transformers.testing_utils import require_torch
 
 from .methods import (
     BottleneckAdapterTestMixin,

wtpsplit/__init__.py

@@ -512,7 +512,7 @@ def __init__(
                     # merge lora weights into transformer for 0 efficiency overhead
                     self.model.model.merge_adapter("sat-lora")
                     self.use_lora = True
-                except:
+                except:  # noqa
                     if lora_path:
                         print(f"LoRA at {lora_path} not found, using base model...")
                     else:

wtpsplit/data_acquisition/extract_all_data.py

@@ -93,7 +93,7 @@ def corrupt_asr(sentences, lang):
 
     try:
         tokenizer = MosesTokenizer(lang)
-    except:
+    except:  # noqa
         corrupted_sentences = [
             preprocess_sentence("".join([char for char in sentence if char not in punct_chars]).lower())
             for sentence in sentences

wtpsplit/evaluation/intrinsic_ted.py

@@ -261,7 +261,7 @@ def main(args):
     ):
         try:
             model_path = os.path.join(args.model_path, os.listdir(args.model_path)[0], "en")
-        except:
+        except:  # noqa
             model_path = args.model_path
         print(model_path)
     else:

wtpsplit/evaluation/llm_sentence.py

@@ -425,7 +425,7 @@ def align_llm_output(row):
         )
         # same as aligned_in, aligned_llm, but with additional formatting. Latter used to debug only.
         formatted_alignment = alignment._format_alignment(aligned_in, aligned_llm).split("\n")
-    except:    # ruff: ignore=E722
+    except:  # noqa
         print("Alignment failed: ", row.name)
         formatted_alignment = [row["test_chunks"], "", " " * len(row["test_chunks"])]
     return pd.Series(

wtpsplit/utils.py

@@ -229,7 +229,7 @@ def corrupt_asr(text: str, lang):
         corrupted_sentences = [
             tokenizer.detokenize(corrupted_tokens).lower() for corrupted_tokens in corrupted_tokenized_sentences
         ]
-    except:
+    except:  # noqa
         corrupted_sentences = [
             "".join([char for char in sentence if char not in Constants.PUNCTUATION_CHARS]).lower()
             for sentence in sentences