Merge pull request #594 from whiteswordLI/fix/lora-load-ddp-weights

Fix: support loading DDP-saved LoRA weights for inference

Author: jingyaogong

model/model_lora.py

@@ -34,6 +34,16 @@ def forward_with_lora(x, layer1=original_forward, layer2=lora):
 
 def load_lora(model, path):
     state_dict = torch.load(path, map_location=model.device)
+    
+    # 兼容DDP训练保存的权重（带有module.前缀），去除前缀以匹配当前模型
+    new_state_dict = {}
+    for k, v in state_dict.items():
+        if k.startswith('module.'):
+            new_state_dict[k[7:]] = v
+        else:
+            new_state_dict[k] = v
+    state_dict = new_state_dict
+
     for name, module in model.named_modules():
         if hasattr(module, 'lora'):
             lora_state = {k.replace(f'{name}.lora.', ''): v for k, v in state_dict.items() if f'{name}.lora.' in k}