fix name

ktransformers/optimize/optimize_rules/DeepSeek-V2-Chat-multi-gpu-4.yaml

@@ -190,7 +190,7 @@
 - match:
     name: "^model$"
   replace:
-    class: "ktransformers.operators.layer_wise_prefill.KDeepseekV2Model"
+    class: "ktransformers.operators.models.KDeepseekV2Model"
     kwargs:
       per_layer_prefill_intput_threshold: 0 # 0 is close layer wise prefill
       transfer_map: 

ktransformers/optimize/optimize_rules/DeepSeek-V2-Chat-multi-gpu.yaml

@@ -103,7 +103,7 @@
 - match:
     name: "^model$"
   replace:
-    class: "ktransformers.operators.layer_wise_prefill.KDeepseekV2Model"
+    class: "ktransformers.operators.models.KDeepseekV2Model"
     kwargs:
       per_layer_prefill_intput_threshold: 0 # 0 is close layer wise prefill
       transfer_map: 

ktransformers/optimize/optimize_rules/DeepSeek-V2-Chat.yaml

@@ -55,7 +55,7 @@
 - match:
     name: "^model$"
   replace:
-    class: "ktransformers.operators.layer_wise_prefill.KDeepseekV2Model"
+    class: "ktransformers.operators.models.KDeepseekV2Model"
     kwargs:
       generate_device: "cuda"
       prefill_device: "cuda"

ktransformers/optimize/optimize_rules/DeepSeek-V2-Lite-Chat-multi-gpu.yaml

@@ -103,7 +103,7 @@
 - match:
     name: "^model$"
   replace:
-    class: "ktransformers.operators.layer_wise_prefill.KDeepseekV2Model"
+    class: "ktransformers.operators.models.KDeepseekV2Model"
     kwargs:
       per_layer_prefill_intput_threshold: 0 # 0 is close layer wise prefill
       transfer_map: 

ktransformers/optimize/optimize_rules/Qwen2-57B-A14B-Instruct-multi-gpu.yaml

@@ -89,7 +89,7 @@
 - match:
     name: "^model$"
   replace:
-    class: "ktransformers.operators.layer_wise_prefill.KQwen2MoeModel"
+    class: "ktransformers.operators.models.KQwen2MoeModel"
     kwargs:
       per_layer_prefill_intput_threshold: 0 # 0 is close layer wise prefill
       transfer_map: 

ktransformers/optimize/optimize_rules/Qwen2-57B-A14B-Instruct.yaml

@@ -38,7 +38,7 @@
 - match:
     name: "^model$"
   replace:
-    class: "ktransformers.operators.layer_wise_prefill.KQwen2MoeModel"
+    class: "ktransformers.operators.models.KQwen2MoeModel"
     kwargs:
       per_layer_prefill_intput_threshold: 0 # 0 is close layer wise prefill
 - match: