[mypy] Forward pass function type hints in lora

vllm/lora/layers.py

@@ -405,7 +405,9 @@ def __init__(self, base_layer: ReplicatedLinear) -> None:
         self.output_size = self.base_layer.output_size
         self.n_slices = 1
 
-    def forward(self, input_):
+    def forward(
+            self,
+            input_) -> Tuple[Optional[torch.Tensor], Optional[torch.Tensor]]:
         """Forward of ReplicatedLinearWithLoRA
 
         Args:
@@ -496,7 +498,9 @@ def slice_bias(self, bias: torch.Tensor) -> torch.Tensor:
         bias = bias[start_idx:end_idx]
         return bias
 
-    def forward(self, input_):
+    def forward(
+            self,
+            input_) -> Tuple[Optional[torch.Tensor], Optional[torch.Tensor]]:
         """Forward of ColumnParallelLinear
 
         Args:
@@ -833,7 +837,9 @@ def slice_lora_b(self, lora_b: torch.Tensor) -> torch.Tensor:
     def slice_bias(self, bias: torch.Tensor) -> torch.Tensor:
         return bias
 
-    def forward(self, input_):
+    def forward(
+            self,
+            input_) -> Tuple[Optional[torch.Tensor], Optional[torch.Tensor]]:
         """Forward of RowParallelLinear
 
         Args:

vllm/lora/models.py

@@ -219,6 +219,7 @@ def from_local_checkpoint(
 
         config["vllm_max_position_embeddings"] = max_position_embeddings
         peft_helper = PEFTHelper.from_dict(config)
+        unexpected_modules: List[str]
         if os.path.isfile(lora_tensor_path):
             tensors: Dict[str, torch.Tensor] = {}
             # Find unexpected modules.

vllm/model_executor/layers/linear.py

@@ -238,7 +238,9 @@ def weight_loader(self, param: Parameter, loaded_weight: torch.Tensor):
         assert param.size() == loaded_weight.size()
         param.data.copy_(loaded_weight)
 
-    def forward(self, x: torch.Tensor) -> torch.Tensor:
+    def forward(
+        self, x: torch.Tensor
+    ) -> Tuple[Optional[torch.Tensor], Optional[torch.Tensor]]:
         bias = self.bias if not self.skip_bias_add else None
         assert self.quant_method is not None
         output = self.quant_method.apply(self, x, bias)