Auto format

hypertune.py

@@ -13,7 +13,8 @@ from transformers import Trainer
 def get_optimal_hyperparameters(env: Env) -> dict[str: any]:
     print(f"device = {env.device}")
 
-    train_data, test_data = tools.preprocess_train_data("data/train.csv", env, upscale_factor=7)
+    train_data, test_data = tools.preprocess_train_data(
+        "data/train.csv", env, upscale_factor=7)
     # validate_data = tools.preprocess_test_data("data/dev.csv", env)
 
     with env.tuning_toggle:
@@ -60,7 +61,8 @@ if __name__ == '__main__':
     logger.addHandler(logging.FileHandler(log_path, mode="w+"))
 
     optuna.logging.enable_propagation()  # Propagate logs to the root logger.
-    optuna.logging.disable_default_handler()  # Stop showing logs in sys.stderr.
+    # Stop showing logs in sys.stderr.
+    optuna.logging.disable_default_handler()
 
     # Perform tuning
     configs_optim: dict[str, any] = get_optimal_hyperparameters(env)

tools.py

@@ -19,10 +19,14 @@ def preprocess_train_data(path: str, env: Env, upscale_factor: int = 7) -> tuple
     dataset = Dataset.from_pandas(df[['label', 'text']]).map(lambda d: env.tokeniser(str(d['text']), truncation=True),
                                                              batched=False)
 
-    dataset_0_split = dataset.filter(lambda x: x['label'] == 0).train_test_split(test_size=0.2)
-    dataset_1_split = dataset.filter(lambda x: x['label'] == 1).train_test_split(test_size=0.2)
-    dataset_train = ds.concatenate_datasets([dataset_0_split['train']] + [dataset_1_split['train']] * upscale_factor)
-    dataset_test = ds.concatenate_datasets([dataset_0_split['test'], dataset_1_split['test']])
+    dataset_0_split = dataset.filter(
+        lambda x: x['label'] == 0).train_test_split(test_size=0.2)
+    dataset_1_split = dataset.filter(
+        lambda x: x['label'] == 1).train_test_split(test_size=0.2)
+    dataset_train = ds.concatenate_datasets(
+        [dataset_0_split['train']] + [dataset_1_split['train']] * upscale_factor)
+    dataset_test = ds.concatenate_datasets(
+        [dataset_0_split['test'], dataset_1_split['test']])
 
     return dataset_train, dataset_test
 

train.py

@@ -8,7 +8,8 @@ from transformers import Trainer
 def train(env: Env) -> None:
     print(f"device = {env.device}")
 
-    train_data, _ = tools.preprocess_train_data("data/train.csv", env, upscale_factor=7)
+    train_data, _ = tools.preprocess_train_data(
+        "data/train.csv", env, upscale_factor=7)
     validate_data = tools.preprocess_test_data("data/dev.csv", env)
 
     trainer = Trainer(