apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-10-23 16:44:22 +08:00
Code Issues Actions 5 Packages Projects Releases Wiki Activity

Fix Speculative Config bug (#3049)

Browse Source 
* fix speculative bug

* fix rl
This commit is contained in:
YuanRisheng
2025-07-29 10:50:48 +08:00
committed by GitHub
parent f2a528f9ae
commit 1a815b7a2a
6 changed files with 21 additions and 58 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
fastdeploy/engine/engine.py
View File

@@ -1081,11 +1081,7 @@ class LLMEngine:
f" --expert_parallel_size {self.cfg.parallel_config.expert_parallel_size}"
f" --quantization {self.cfg.model_config.quantization}"
f" --ori_vocab_size {ori_vocab_size}"
f" --speculative_method {self.cfg.speculative_config.method}"
f" --speculative_max_draft_token_num {self.cfg.speculative_config.num_speculative_tokens}"
f" --speculative_model_name_or_path {self.cfg.speculative_config.model_name_or_path}"
f" --speculative_model_quantization {self.cfg.speculative_config.quantization}"
f" --speculative_benchmark_mode {self.cfg.speculative_config.benchmark_mode}"
f" --speculative_config '{self.cfg.speculative_config.to_json_string()}'"
f" --graph_optimization_config '{self.cfg.graph_optimization_config.to_json_string()}'"
f" --guided_decoding_backend {self.cfg.guided_decoding_backend}"
f" --load_strategy {self.cfg.model_config.load_strategy}"