[Model] use AutoWeightsLoader for bart (#18299)

Signed-off-by: calvin chen <120380290@qq.com>

vllm/model_executor/models/bart.py

@@ -46,7 +46,7 @@ from vllm.model_executor.sampling_metadata import SamplingMetadata
 from vllm.sequence import IntermediateTensors
 
 from .interfaces import SupportsQuant, SupportsV0Only
-from .utils import maybe_prefix
+from .utils import AutoWeightsLoader, WeightsMapper, maybe_prefix
 
 logger = logging.get_logger(__name__)
 
@@ -700,7 +700,8 @@ class BartDecoder(nn.Module):
 
 class BartModel(nn.Module, SupportsQuant):
     _tied_weights_keys = [
-        "encoder.embed_tokens.weight", "decoder.embed_tokens.weight"
+        "encoder.embed_tokens.weight",
+        "decoder.embed_tokens.weight",
     ]
 
     def __init__(self, *, vllm_config: VllmConfig, prefix: str = ""):
@@ -763,10 +764,54 @@ class BartModel(nn.Module, SupportsQuant):
 
         return decoder_outputs
 
+    def load_weights(self, weights: Iterable[tuple[str,
+                                                   torch.Tensor]]) -> set[str]:
+        stacked_params_mapping = [
+            # (param_name, shard_name, shard_id)
+            ("qkv_proj", "q_proj", "q"),
+            ("qkv_proj", "k_proj", "k"),
+            ("qkv_proj", "v_proj", "v"),
+        ]
+
+        other_weights = []
+        loaded_stacked_params = []
+        model_params_dict = dict(self.named_parameters())
+
+        for name, loaded_weight in weights:
+            for (param_name, weight_name, shard_id) in stacked_params_mapping:
+                if weight_name not in name:
+                    continue
+                name = name.replace(weight_name, param_name)
+                if name not in model_params_dict:
+                    continue
+                param = model_params_dict[name]
+                weight_loader = param.weight_loader
+                weight_loader(param, loaded_weight, shard_id)
+                loaded_stacked_params.append(name)
+                break
+            else:
+                if name in model_params_dict:
+                    other_weights.append((name, loaded_weight))
+
+        loader = AutoWeightsLoader(self)
+        loaded_params = loader.load_weights(other_weights)
+        loaded_params.update(loaded_stacked_params)
+        return loaded_params
+
 
 class BartForConditionalGeneration(nn.Module, SupportsV0Only, SupportsQuant):
-    packed_modules_mapping = {"qkv_proj": ["q_proj", "k_proj", "v_proj"]}
+    hf_to_vllm_mapper = WeightsMapper(
-    base_model_prefix = "model"
+        orig_to_new_prefix={
+            "decoder.": "model.decoder.",
+            "encoder.": "model.encoder.",
+            "shared.": "model.shared."
+        },
+        orig_to_new_substr={
+            "beta": "bias",
+            "gamma": "weight",
+            "LayerNorm": "layernorm",
+        },
+    )
 
     def __init__(self, *, vllm_config: VllmConfig, prefix: str = ""):
 
@@ -789,7 +834,6 @@ class BartForConditionalGeneration(nn.Module, SupportsV0Only, SupportsQuant):
         self.lm_head = BartParallelLMHead(config.vocab_size,
                                           config.d_model,
                                           embed_scale=embed_scale)
-
         self.logits_processor = LogitsProcessor(self.unpadded_vocab_size,
                                                 config.vocab_size)
 
@@ -828,61 +872,12 @@ class BartForConditionalGeneration(nn.Module, SupportsV0Only, SupportsQuant):
                                        sampling_metadata)
         return logits
 
-    stacked_params_mapping = {
+    def load_weights(self, weights: Iterable[tuple[str,
-        "q_proj": {
+                                                   torch.Tensor]]) -> set[str]:
-            "param_name": "qkv_proj",
-            "shard_id": "q",
-        },
-        "k_proj": {
-            "param_name": "qkv_proj",
-            "shard_id": "k",
-        },
-        "v_proj": {
-            "param_name": "qkv_proj",
-            "shard_id": "v",
-        },
-    }
-
-    params_mapping = {
-        "beta": "bias",
-        "gamma": "weight",
-        "LayerNorm": "layernorm",
-    }
-
-    def _rename_key(self, key: str):
-        prefix = f"{self.base_model_prefix}."
-        key = key[len(prefix):] if key.startswith(prefix) else key
-
-        for src, dst in self.params_mapping.items():
-            key = key.replace(src, dst)
-
-        return key
-
-    def _rename_stacked_param(
-        self,
-        name: str,
-    ) -> tuple[str, Optional[str]]:
-        for key, mapping in self.stacked_params_mapping.items():
-            if key in name:
-                name = name.replace(key, mapping["param_name"])
-                return name, mapping["shard_id"]
-        return name, None
-
-    def load_weights(self, weights: Iterable[tuple[str, torch.Tensor]]):
-
-        model_params_dict = dict(self.model.named_parameters())
-        top_params_dict = dict(self.named_parameters())
-
         weights_tuple_list = list(weights)
 
         shared_embedding_weight = None
-        shared_embedding_shard_id = None
-
         for name, loaded_weight in weights_tuple_list:
-
-            name = self._rename_key(name)
-            name, shard_id = self._rename_stacked_param(name)
-
             if ('shared.weight' in name
                     or 'encoder.embed_tokens.weight' in name
                     or 'decoder.embed_tokens.weight' in name
@@ -890,49 +885,24 @@ class BartForConditionalGeneration(nn.Module, SupportsV0Only, SupportsQuant):
                 assert shared_embedding_weight is None, (
                     "Conflicting embedding weights.")
                 shared_embedding_weight = loaded_weight
-                shared_embedding_shard_id = shard_id
-            else:
-                # Skip the specific downstream task weight.
-                if name.startswith('cls.'):
-                    continue
-                # use Pooler instead.
-                if name.startswith('pooler.'):
-                    continue
-                # Skip loading extra bias for GPTQ models.
-                if name.endswith(".bias") and name not in model_params_dict:
-                    continue
 
-                param = model_params_dict[name]
+        loader = AutoWeightsLoader(
-                weight_loader = getattr(param, "weight_loader",
+            self,
-                                        default_weight_loader)
+            skip_prefixes=(["cls.", "pooler."]),
-                if shard_id:
+        )
-                    weight_loader(param, loaded_weight, shard_id)
+        loaded_params = loader.load_weights(weights_tuple_list,
-                else:
+                                            mapper=self.hf_to_vllm_mapper)
-                    weight_loader(param, loaded_weight)
 
-        # Assign shared weight values
+        if shared_embedding_weight is not None:
-        encoder_in_param = model_params_dict['encoder.embed_tokens.weight']
+            weight_loader = getattr(self.lm_head.weight, "weight_loader",
-        encoder_in_weight_loader = getattr(encoder_in_param, "weight_loader",
+                                    default_weight_loader)
-                                           default_weight_loader)
+            weight_loader(self.lm_head.weight, shared_embedding_weight)
 
-        decoder_in_param = model_params_dict['decoder.embed_tokens.weight']
+            self.model.encoder.embed_tokens.weight = self.lm_head.weight
-        decoder_in_weight_loader = getattr(decoder_in_param, "weight_loader",
+            self.model.decoder.embed_tokens.weight = self.lm_head.weight
-                                           default_weight_loader)
+            loaded_params.update({
+                'model.encoder.embed_tokens.weight', 'lm_head.weight',
+                'model.decoder.embed_tokens.weight'
+            })
 
-        lm_head_in_param = top_params_dict['lm_head.weight']
+        return loaded_params
-        lm_head_in_weight_loader = getattr(lm_head_in_param, "weight_loader",
-                                           default_weight_loader)
-
-        assert shared_embedding_weight is not None
-
-        if shared_embedding_shard_id:
-            encoder_in_weight_loader(encoder_in_param, shared_embedding_weight,
-                                     shared_embedding_shard_id)
-            decoder_in_weight_loader(decoder_in_param, shared_embedding_weight,
-                                     shared_embedding_shard_id)
-            lm_head_in_weight_loader(lm_head_in_param, shared_embedding_weight,
-                                     shared_embedding_shard_id)
-        else:
-            encoder_in_weight_loader(encoder_in_param, shared_embedding_weight)
-            decoder_in_weight_loader(decoder_in_param, shared_embedding_weight)
-            lm_head_in_weight_loader(lm_head_in_param, shared_embedding_weight)