support to read eos_id from config.json (#569)

Co-authored-by: wangzaijun <[email protected]>

lightllm/server/api_server.py

@@ -365,7 +365,9 @@ def make_argument_parser() -> argparse.ArgumentParser:
         default=None,
         help="max tokens num for new cat batch, it control prefill batch size to Preventing OOM",
     )
-    parser.add_argument("--eos_id", nargs="+", type=int, default=[2], help="eos stop token id")
+    parser.add_argument(
+        "--eos_id", nargs="+", type=int, default=None, help="eos stop token id, if None, will load from config.json"
+    )
     parser.add_argument(
         "--running_max_req_size", type=int, default=1000, help="the max size for forward requests in the same time"
     )
@@ -505,7 +507,7 @@ def main():
     assert args.max_req_input_len < args.max_req_total_len
     assert args.max_req_total_len <= args.max_total_token_num
     assert not (args.beam_mode and args.use_dynamic_prompt_cache), "Beam mode incompatible with dynamic prompt cache"
-    
+
     # splitfuse_mode 和 cuda_graph 不能同时开启
     if args.splitfuse_mode:
         assert args.disable_cudagraph
@@ -537,6 +539,18 @@ def main():
             batch_max_tokens = max(batch_max_tokens, args.splitfuse_block_size)
             args.batch_max_tokens = batch_max_tokens
 
+    # help to manage data stored on Ceph
+    if "s3://" in args.model_dir:
+        from lightllm.utils.petrel_helper import s3_model_prepare
+
+        s3_model_prepare(args.model_dir)
+
+    # 如果args.eos_id 是 None, 从 config.json 中读取 eos_token_id 相关的信息，赋值给 args
+    if args.eos_id is None:
+        from lightllm.utils.config_utils import get_eos_token_ids
+
+        args.eos_id = get_eos_token_ids(args.model_dir)
+
     logger.info(f"all start args:{args}")
 
     can_use_ports = alloc_can_use_network_port(num=6 + args.tp, used_nccl_port=args.nccl_port)
@@ -560,12 +574,6 @@ def main():
     global metric_client
     metric_client = MetricClient(metric_port)
 
-    # help to manage data stored on Ceph
-    if "s3://" in args.model_dir:
-        from lightllm.utils.petrel_helper import s3_model_prepare
-
-        s3_model_prepare(args.model_dir)
-
     global httpserver_manager
     httpserver_manager = HttpServerManager(
         args,

lightllm/utils/config_utils.py

@@ -0,0 +1,18 @@
+import json
+import os
+
+
+def get_config_json(model_path: str):
+    with open(os.path.join(model_path, "config.json"), "r") as file:
+        json_obj = json.load(file)
+    return json_obj
+
+
+def get_eos_token_ids(model_path: str):
+    config_json = get_config_json(model_path)
+    eos_token_id = config_json["eos_token_id"]
+    if isinstance(eos_token_id, int):
+        return [eos_token_id]
+    if isinstance(eos_token_id, list):
+        return eos_token_id
+    assert False, "error eos_token_id format in config.json"