Update quality tooling for formatting (#21480)

* Result of black 23.1

* Update target to Python 3.7

* Switch flake8 to ruff

* Configure isort

* Configure isort

* Apply isort with line limit

* Put the right black version

* adapt black in check copies

* Fix copies

examples/legacy/seq2seq/finetune_trainer.py

@@ -19,9 +19,10 @@ import sys
 from dataclasses import dataclass, field
 from typing import Optional
 
-import transformers
 from seq2seq_trainer import Seq2SeqTrainer
 from seq2seq_training_args import Seq2SeqTrainingArguments
+
+import transformers
 from transformers import (
     AutoConfig,
     AutoModelForSeq2SeqLM,
@@ -337,7 +338,6 @@ def main():
         metrics["val_loss"] = round(metrics["val_loss"], 4)
 
         if trainer.is_world_process_zero():
-
             handle_metrics("val", metrics, training_args.output_dir)
             all_metrics.update(metrics)
 

examples/legacy/seq2seq/old_test_calculate_rouge.py

@@ -16,8 +16,8 @@ from collections import defaultdict
 from pathlib import Path
 
 import pandas as pd
-
 from rouge_cli import calculate_rouge_path
+
 from utils import calculate_rouge
 
 
@@ -87,7 +87,6 @@ def test_single_sent_scores_dont_depend_on_newline_sep():
 
 
 def test_pegasus_newline():
-
     pred = [
         """" "a person who has such a video needs to immediately give it to the investigators," prosecutor says .<n> "it is a very disturbing scene," editor-in-chief of bild online tells "erin burnett: outfront" """
     ]

examples/legacy/seq2seq/old_test_datasets.py

@@ -17,11 +17,11 @@ from pathlib import Path
 
 import numpy as np
 import pytest
-from torch.utils.data import DataLoader
-
 from pack_dataset import pack_data_dir
 from parameterized import parameterized
 from save_len_file import save_len_file
+from torch.utils.data import DataLoader
+
 from transformers import AutoTokenizer
 from transformers.models.mbart.modeling_mbart import shift_tokens_right
 from transformers.testing_utils import TestCasePlus, slow

examples/legacy/seq2seq/old_test_fsmt_bleu_score.py

@@ -18,6 +18,7 @@ import json
 import unittest
 
 from parameterized import parameterized
+
 from transformers import FSMTForConditionalGeneration, FSMTTokenizer
 from transformers.testing_utils import get_tests_dir, require_torch, slow, torch_device
 from utils import calculate_bleu

examples/legacy/seq2seq/old_test_seq2seq_examples.py

@@ -21,6 +21,7 @@ from unittest.mock import patch
 from parameterized import parameterized
 from run_eval import run_generate
 from run_eval_search import run_search
+
 from transformers.testing_utils import CaptureStdout, TestCasePlus, slow
 from utils import ROUGE_KEYS
 

examples/legacy/seq2seq/pack_dataset.py

@@ -29,7 +29,6 @@ from transformers import AutoTokenizer
 
 
 def pack_examples(tok, src_examples, tgt_examples, max_tokens=1024):
-
     finished_src, finished_tgt = [], []
 
     sorted_examples = list(zip(src_examples, tgt_examples))

examples/legacy/seq2seq/run_eval_search.py

@@ -20,6 +20,7 @@ import sys
 from collections import OrderedDict
 
 from run_eval import datetime_now, run_generate
+
 from utils import ROUGE_KEYS
 
 

examples/legacy/seq2seq/seq2seq_training_args.py

@@ -17,6 +17,7 @@ from dataclasses import dataclass, field
 from typing import Optional
 
 from seq2seq_trainer import arg_to_scheduler
+
 from transformers import TrainingArguments
 
 

examples/legacy/seq2seq/utils.py

@@ -29,10 +29,10 @@ import torch
 import torch.distributed as dist
 from rouge_score import rouge_scorer, scoring
 from sacrebleu import corpus_bleu
+from sentence_splitter import add_newline_to_end_of_each_sentence
 from torch import nn
 from torch.utils.data import Dataset, Sampler
 
-from sentence_splitter import add_newline_to_end_of_each_sentence
 from transformers import BartTokenizer, EvalPrediction, PreTrainedTokenizer, T5Tokenizer
 from transformers.models.bart.modeling_bart import shift_tokens_right
 from transformers.utils import cached_property
@@ -132,7 +132,7 @@ class AbstractSeq2SeqDataset(Dataset):
         type_path="train",
         n_obs=None,
         prefix="",
-        **dataset_kwargs
+        **dataset_kwargs,
     ):
         super().__init__()
         self.src_file = Path(data_dir).joinpath(type_path + ".source")