[mypy] Add mypy type annotation part 1 (#4006)

vllm/executor/ray_gpu_executor.py

@@ -3,7 +3,7 @@ import copy
 import os
 import pickle
 from collections import defaultdict
-from typing import TYPE_CHECKING, Any, Dict, List, Optional
+from typing import TYPE_CHECKING, Any, Dict, List, Optional, Tuple
 
 from vllm.config import (CacheConfig, DeviceConfig, LoRAConfig, ModelConfig,
                          ParallelConfig, SchedulerConfig, SpeculativeConfig,
@@ -197,7 +197,7 @@ class RayGPUExecutor(ExecutorBase):
             max_parallel_loading_workers,
         )
 
-    def determine_num_available_blocks(self) -> tuple[int, int]:
+    def determine_num_available_blocks(self) -> Tuple[int, int]:
         """Determine the number of available KV blocks.
 
         This invokes `determine_num_available_blocks` on each worker and takes
@@ -205,7 +205,7 @@ class RayGPUExecutor(ExecutorBase):
         compatible with all workers.
 
         Returns:
-            - tuple[num_gpu_blocks, num_cpu_blocks]
+            - Tuple[num_gpu_blocks, num_cpu_blocks]
         """
         # Get the maximum number of blocks that can be allocated on GPU and CPU.
         num_blocks = self._run_workers("determine_num_available_blocks", )
@@ -276,7 +276,7 @@ class RayGPUExecutor(ExecutorBase):
         self,
         method: str,
         *args,
-        driver_args: Optional[List[Any]] = None,
+        driver_args: Optional[Tuple[Any, ...]] = None,
         driver_kwargs: Optional[Dict[str, Any]] = None,
         max_concurrent_workers: Optional[int] = None,
         use_ray_compiled_dag: bool = False,
@@ -291,6 +291,7 @@ class RayGPUExecutor(ExecutorBase):
         if use_ray_compiled_dag:
             # Right now, compiled DAG can only accept a single
             # input. TODO(sang): Fix it.
+            assert self.forward_dag is not None
             output_channels = self.forward_dag.execute(1)
         else:
             # Start the ray workers first.
@@ -369,7 +370,7 @@ class RayGPUExecutorAsync(RayGPUExecutor, ExecutorAsyncBase):
         self,
         method: str,
         *args,
-        driver_args: Optional[List[Any]] = None,
+        driver_args: Optional[Tuple[Any, ...]] = None,
         driver_kwargs: Optional[Dict[str, Any]] = None,
         **kwargs,
     ) -> Any: