[deps] goodbye python 3.9 (#9677)

Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
Co-authored-by: hiyouga <16256802+hiyouga@users.noreply.github.com>
Co-authored-by: hiyouga <hiyouga@buaa.edu.cn>

src/llamafactory/v1/plugins/data_plugins/converter.py

@@ -13,9 +13,7 @@
 # limitations under the License.
 
 
-from typing import Any, Literal, TypedDict
-
-from typing_extensions import NotRequired
+from typing import Any, Literal, NotRequired, TypedDict
 
 from ...utils import logging
 from ...utils.plugin import BasePlugin

src/llamafactory/v1/plugins/data_plugins/loader.py

@@ -15,7 +15,7 @@
 
 import os
 import random
-from typing import Any, Literal, Optional, Union
+from typing import Any, Literal
 
 from datasets import load_dataset
 
@@ -70,7 +70,7 @@ class DataIndexPlugin(BasePlugin):
     """Plugin for adjusting dataset index."""
 
     def adjust_data_index(
-        self, data_index: list[tuple[str, int]], size: Optional[int], weight: Optional[float]
+        self, data_index: list[tuple[str, int]], size: int | None, weight: float | None
     ) -> list[tuple[str, int]]:
         """Adjust dataset index by size and weight.
 
@@ -95,8 +95,8 @@ class DataSelectorPlugin(BasePlugin):
     """Plugin for selecting dataset samples."""
 
     def select(
-        self, data_index: list[tuple[str, int]], index: Union[slice, list[int], Any]
-    ) -> Union[tuple[str, int], list[tuple[str, int]]]:
+        self, data_index: list[tuple[str, int]], index: slice | list[int] | Any
+    ) -> tuple[str, int] | list[tuple[str, int]]:
         """Select dataset samples.
 
         Args:

src/llamafactory/v1/plugins/data_plugins/template.py

@@ -14,7 +14,6 @@
 
 
 from dataclasses import dataclass
-from typing import Union
 
 
 @dataclass
@@ -32,7 +31,7 @@ class QwenTemplate:
     message_template: str = "<|im_start|>{role}\n{content}<|im_end|>\n"  # FIXME if role: tool
     thinking_template: str = "<think>\n{content}\n</think>\n\n"
 
-    def _extract_content(self, content_data: Union[str, list[dict[str, str]]]) -> str:
+    def _extract_content(self, content_data: str | list[dict[str, str]]) -> str:
         if isinstance(content_data, str):
             return content_data.strip()
 
@@ -47,7 +46,7 @@ class QwenTemplate:
 
         return ""
 
-    def render_message(self, message: dict[str, Union[str, list[dict[str, str]]]]) -> str:
+    def render_message(self, message: dict[str, str | list[dict[str, str]]]) -> str:
         role = message["role"]
         content = self._extract_content(message.get("content", ""))
 

src/llamafactory/v1/plugins/model_plugins/kernels/registry.py

@@ -13,7 +13,8 @@
 # limitations under the License.
 
 from abc import ABC, ABCMeta, abstractmethod
-from typing import Any, Callable, Optional, Union
+from collections.abc import Callable
+from typing import Any, Optional
 
 from ....accelerator.helper import DeviceType, get_current_accelerator
 from ....utils.types import HFModel
@@ -38,7 +39,7 @@ class KernelRegistry:
         self._initialized = True
 
     def register(
-        self, kernel_type: KernelType, device_type: DeviceType, kernel_impl: Optional[Callable[..., Any]]
+        self, kernel_type: KernelType, device_type: DeviceType, kernel_impl: Callable[..., Any] | None
     ) -> None:
         """Register a kernel implementation.
 
@@ -56,7 +57,7 @@ class KernelRegistry:
         self._registry[kernel_type][device_type] = kernel_impl
         print(f"Registered kernel {kernel_type.name} for device {device_type.name}.")
 
-    def get_kernel(self, kernel_type: KernelType, device_type: DeviceType) -> Optional[Callable[..., Any]]:
+    def get_kernel(self, kernel_type: KernelType, device_type: DeviceType) -> Callable[..., Any] | None:
         return self._registry.get(kernel_type, {}).get(device_type)
 
 
@@ -105,9 +106,9 @@ class MetaKernel(ABC, metaclass=AutoRegisterKernelMeta):
         auto_register: Set to False to disable automatic registration (default: True).
     """
 
-    type: Optional[KernelType] = None
-    device: Optional[DeviceType] = None
-    kernel: Optional[Callable] = None
+    type: KernelType | None = None
+    device: DeviceType | None = None
+    kernel: Callable | None = None
 
     @classmethod
     @abstractmethod
@@ -228,7 +229,7 @@ def discover_kernels(model: HFModel = None) -> list[type[MetaKernel]]:
     return discovered_kernels
 
 
-def apply_kernel(model: HFModel, kernel: Union[type[MetaKernel], Any], /, **kwargs) -> "HFModel":
+def apply_kernel(model: HFModel, kernel: type[MetaKernel] | Any, /, **kwargs) -> "HFModel":
     """Call the MetaKernel's `apply` to perform the replacement.
 
     Corresponding replacement logic is maintained inside each kernel; the only

src/llamafactory/v1/plugins/model_plugins/peft.py

@@ -12,7 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-from typing import Literal, Optional, TypedDict
+from typing import Literal, TypedDict
 
 from peft import LoraConfig, PeftModel, get_peft_model
 
@@ -36,7 +36,7 @@ class FreezeConfigDict(TypedDict, total=False):
     """Plugin name."""
     freeze_trainable_layers: int
     """Freeze trainable layers."""
-    freeze_trainable_modules: Optional[list[str]]
+    freeze_trainable_modules: list[str] | None
     """Freeze trainable modules."""