🎨 fmt

Signed-off-by: Prashant Gupta <prashantgupta@us.ibm.com>

Author: prashantgupta24

vllm_spyre/v1/worker/spyre_model_runner.py

@@ -879,9 +879,9 @@ def _prepare_decode(
             slot = [start_slot + offset]
             slot_mapping.append(slot)
             output_token_ids = req_state.output_token_ids
-            generation_token = torch.tensor(
-                output_token_ids[-1], dtype=torch.long, device=self.device
-            )
+            generation_token = torch.tensor(output_token_ids[-1],
+                                            dtype=torch.long,
+                                            device=self.device)
             input_tokens.append([generation_token])
             seq_len = cached_request_data.num_computed_tokens[
                 cached_reqs_map[req_id]]

vllm_spyre/v1/worker/spyre_worker.py

@@ -358,8 +358,8 @@ def _warmup_spyre_dynamic_size(self, special_token_ids):
                 self.execute_model(scheduler_output)
 
             # one decode iteration across all sequences
-            req_ids = []
-            new_block_ids = []
+            req_ids: list[str] = []
+            new_block_ids: list[tuple[list[int], ...]] = []
             num_computed_tokens = []
             for req in dummy_requests:
                 req_ids.append(req.req_id)