try pyupgrade-up py38 (#1999)

Author: Borda

extensions/xla/scripts/prepare_alpaca.py

@@ -33,15 +33,15 @@ def prepare(
     which stores the preprocessed and tokenized prompts and labels.
     """
     if max_seq_length is None:
-        with open(checkpoint_dir / "model_config.yaml", "r", encoding="utf-8") as file:
+        with open(checkpoint_dir / "model_config.yaml", encoding="utf-8") as file:
             config = yaml.safe_load(file)
             max_seq_length = config["block_size"]
 
     destination_path.mkdir(parents=True, exist_ok=True)
     data_file_path = destination_path / data_file_name
     print("Loading data file...")
     download_if_missing(data_file_path, data_file_url)
-    with open(data_file_path, "r", encoding="utf-8") as file:
+    with open(data_file_path, encoding="utf-8") as file:
         data = json.load(file)
 
     print("Loading tokenizer...")

litgpt/data/alpaca.py

@@ -63,7 +63,7 @@ def prepare_data(self) -> None:
         download_if_missing(self.download_dir / self.file_name, self.file_url)
 
     def setup(self, stage: str = "") -> None:
-        with open(self.download_dir / self.file_name, "r", encoding="utf-8") as file:
+        with open(self.download_dir / self.file_name, encoding="utf-8") as file:
             data = json.load(file)
 
         # Partition the dataset into train and test

litgpt/data/flan.py

@@ -107,7 +107,7 @@ def _dataloader(self, split: str) -> DataLoader:
 
 def load_jsonl(filename: Path) -> List[Dict[str, str]]:
     data = []
-    with open(filename, "r", encoding="utf-8") as f:
+    with open(filename, encoding="utf-8") as f:
         for line in f:
             data.append(json.loads(line))
     return data

litgpt/data/json_data.py

@@ -139,10 +139,10 @@ def find_split(self, split_name: str) -> Optional[Path]:
 
 def load_split(json_path: Path) -> Any:
     if json_path.suffix == ".json":
-        with open(json_path, "r", encoding="utf-8") as file:
+        with open(json_path, encoding="utf-8") as file:
             return json.load(file)
     if json_path.suffix == ".jsonl":
-        with open(json_path, "r", encoding="utf-8") as file:
+        with open(json_path, encoding="utf-8") as file:
             return [json.loads(line) for line in file]
     else:
         raise ValueError(f"Unsupported file format: {json_path.suffix}. Expected `.json` or `.jsonl`.")

litgpt/data/longform.py

@@ -63,7 +63,7 @@ def val_dataloader(self):
         return self._dataloader("val")
 
     def _dataloader(self, split: str) -> DataLoader:
-        with open(self.download_dir / f"{split}.json", "r", encoding="utf-8") as file:
+        with open(self.download_dir / f"{split}.json", encoding="utf-8") as file:
             data = json.load(file)
 
         dataset = SFTDataset(

litgpt/data/text_files.py

@@ -131,7 +131,7 @@ def val_dataloader(self) -> DataLoader:
 
 
 def tokenize(filename: str, tokenizer: Tokenizer):
-    with open(filename, "r", encoding="utf-8") as file:
+    with open(filename, encoding="utf-8") as file:
         text = file.read()
     text = text.strip()
     yield tokenizer.encode(text, bos=True, eos=False)

litgpt/data/tinystories.py

@@ -104,7 +104,7 @@ def val_dataloader(self) -> DataLoader:
 
 
 def tokenize(filename: str, tokenizer: Tokenizer):
-    with open(filename, "r", encoding="utf-8") as f:
+    with open(filename, encoding="utf-8") as f:
         data = json.load(f)
     global_rank = int(os.environ["DATA_OPTIMIZER_GLOBAL_RANK"])
     num_workers = int(os.environ["DATA_OPTIMIZER_NUM_WORKERS"])

litgpt/generate/sequentially.py

@@ -11,7 +11,7 @@
 from functools import partial
 from pathlib import Path
 from pprint import pprint
-from typing import Literal, Optional
+from typing import Literal, Optional, Type
 
 import lightning as L
 import torch
@@ -20,7 +20,6 @@
 from lightning.fabric.utilities.init import _materialize_meta_tensors
 from lightning_utilities.core.imports import RequirementCache
 from tqdm import tqdm
-from typing_extensions import Type
 
 import litgpt.generate.base as generate_base
 from litgpt.config import Config

litgpt/prompts.py

@@ -479,7 +479,7 @@ def save_prompt_style(style: Union[str, PromptStyle], checkpoint_dir: Path) -> N
 
 
 def load_prompt_style(checkpoint_dir: Path) -> PromptStyle:
-    with open(checkpoint_dir / "prompt_style.yaml", "r", encoding="utf-8") as file:
+    with open(checkpoint_dir / "prompt_style.yaml", encoding="utf-8") as file:
         config = yaml.safe_load(file)
     # Support loading the full module path for user-defined prompt classes
     full_module_path, cls_name = config["class_path"].rsplit(".", 1)

litgpt/scripts/merge_lora.py

@@ -92,7 +92,7 @@ def load_lora_metadata(checkpoint_dir: Path) -> Tuple[Dict[str, Any], Path, Opti
             f" the `litgpt/finetune/lora.py` script."
         )
 
-    with open(hparams_file, "r", encoding="utf-8") as file:
+    with open(hparams_file, encoding="utf-8") as file:
         hparams = yaml.safe_load(file)
 
     lora_params = {k: v for k, v in hparams.items() if k.startswith("lora_")}