vllm.model_executor.models.aria

AriaDummyInputsBuilder

Bases: BaseDummyInputsBuilder[AriaProcessingInfo]

Source code in vllm/model_executor/models/aria.py

class AriaDummyInputsBuilder(BaseDummyInputsBuilder[AriaProcessingInfo]):

    def get_dummy_text(self, mm_counts: Mapping[str, int]) -> str:
        num_images = mm_counts.get("image", 0)

        processor = self.info.get_hf_processor()
        image_token: str = processor.tokenizer.image_token  # type: ignore

        return image_token * num_images

    def get_dummy_mm_data(
        self,
        seq_len: int,
        mm_counts: Mapping[str, int],
    ) -> MultiModalDataDict:
        vision_config = self.info.get_vision_config()

        max_image_size = vision_config.image_size
        num_images = mm_counts.get("image", 0)

        return {
            "image":
            self._get_dummy_images(width=max_image_size,
                                   height=max_image_size,
                                   num_images=num_images)
        }

get_dummy_mm_data

get_dummy_mm_data(
    seq_len: int, mm_counts: Mapping[str, int]
) -> MultiModalDataDict

Source code in vllm/model_executor/models/aria.py

def get_dummy_mm_data(
    self,
    seq_len: int,
    mm_counts: Mapping[str, int],
) -> MultiModalDataDict:
    vision_config = self.info.get_vision_config()

    max_image_size = vision_config.image_size
    num_images = mm_counts.get("image", 0)

    return {
        "image":
        self._get_dummy_images(width=max_image_size,
                               height=max_image_size,
                               num_images=num_images)
    }

get_dummy_text

get_dummy_text(mm_counts: Mapping[str, int]) -> str

Source code in vllm/model_executor/models/aria.py

def get_dummy_text(self, mm_counts: Mapping[str, int]) -> str:
    num_images = mm_counts.get("image", 0)

    processor = self.info.get_hf_processor()
    image_token: str = processor.tokenizer.image_token  # type: ignore

    return image_token * num_images

AriaForConditionalGeneration

Bases: Module, SupportsMultiModal

Aria model for conditional generation tasks.

This model combines a vision tower, a multi-modal projector, and a language model to perform tasks that involve both image and text inputs.

Source code in vllm/model_executor/models/aria.py

@MULTIMODAL_REGISTRY.register_processor(AriaMultiModalProcessor,
                                        info=AriaProcessingInfo,
                                        dummy_inputs=AriaDummyInputsBuilder)
class AriaForConditionalGeneration(nn.Module, SupportsMultiModal):
    """
    Aria model for conditional generation tasks.

    This model combines a vision tower, a multi-modal projector, and a language
    model to perform tasks that involve both image and text inputs.
    """
    hf_to_vllm_mapper = WeightsMapper(
        orig_to_new_prefix={
            # mapping for new names in checkpoint saved after transformers v4.52
            "model.language_model.": "language_model.model.",
            "model.vision_tower.": "vision_tower.",
            "model.multi_modal_projector.": "multi_modal_projector.",
            # mapping for original checkpoint
            "language_model.model": "language_model",
            "language_model.lm_head": "lm_head",
        },
        orig_to_new_suffix={
            "router.weight": "router_weight",
        },
    )

    @classmethod
    def get_placeholder_str(cls, modality: str, i: int) -> Optional[str]:
        if modality.startswith("image"):
            return "<|fim_prefix|><|img|><|fim_suffix|>"

        raise ValueError("Only image modality is supported")

    def __init__(
        self,
        vllm_config: VllmConfig,
        prefix: str = "",
    ):
        super().__init__()
        config = vllm_config.model_config.hf_config
        quant_config = vllm_config.quant_config

        self.config = config
        self.vision_tower = AriaVisionTransformer(
            config.vision_config,
            quant_config=quant_config,
            prefix=f"{prefix}.vision_tower",
        )
        self.multi_modal_projector = AriaProjector(config)
        self.vocab_size = config.text_config.vocab_size
        self.language_model = AriaTextModel(
            vllm_config=vllm_config.with_hf_config(config.text_config),
            prefix=maybe_prefix(prefix, "language_model.model"),
        )
        self.pad_token_id = (self.config.pad_token_id
                             if self.config.pad_token_id is not None else -1)
        self.unpadded_vocab_size = config.text_config.vocab_size
        self.lm_head = ParallelLMHead(
            self.unpadded_vocab_size,
            config.text_config.hidden_size,
            org_num_embeddings=self.language_model.org_vocab_size,
            quant_config=quant_config,
        )
        logit_scale = getattr(config, "logit_scale", 1.0)
        self.logits_processor = LogitsProcessor(self.unpadded_vocab_size,
                                                self.vocab_size, logit_scale)

    def _validate_image_sizes(
            self, images: list[torch.Tensor]) -> list[torch.Tensor]:
        if not all(img.shape == images[0].shape for img in images):
            raise ValueError("All images must be the same size")
        return images

    def _parse_and_validate_image_input(
            self, **kwargs: object) -> Optional[AriaImagePixelInputs]:
        pixel_values = kwargs.pop("pixel_values", None)
        pixel_mask = kwargs.pop("pixel_mask", None)

        if pixel_values is None:
            return None

        if not isinstance(pixel_values, (torch.Tensor, list)):
            raise ValueError("Incorrect type of pixel values. "
                             f"Got type: {type(pixel_values)}")

        pixel_values = self._validate_image_sizes(pixel_values)
        pixel_values = flatten_bn(pixel_values, concat=True)

        if pixel_mask is not None:
            if not isinstance(pixel_mask, (torch.Tensor, list)):
                raise ValueError("Incorrect type of pixel mask. "
                                 f"Got type: {type(pixel_mask)}")

            pixel_mask = flatten_bn(pixel_mask, concat=True)

        return AriaImagePixelInputs(
            pixel_values=pixel_values,
            pixel_mask=pixel_mask,
        )

    def _create_patch_attention_mask(
            self, pixel_mask: Optional[torch.Tensor]) -> torch.Tensor:
        if pixel_mask is None:
            return None

        patches_subgrid = pixel_mask.unfold(
            dimension=1,
            size=self.vision_tower.config.patch_size,
            step=self.vision_tower.config.patch_size,
        ).unfold(
            dimension=2,
            size=self.vision_tower.config.patch_size,
            step=self.vision_tower.config.patch_size,
        )
        return (patches_subgrid.sum(dim=(-1, -2)) > 0).bool()

    def _process_image_input(
        self, image_input: AriaImagePixelInputs
    ) -> tuple[torch.Tensor, torch.Tensor]:
        assert self.vision_tower is not None

        pixel_values = image_input['pixel_values']
        pixel_mask = image_input['pixel_mask']

        patch_attention_mask = self._create_patch_attention_mask(pixel_mask)

        image_outputs = self.vision_tower(
            pixel_values=pixel_values,
            patch_attention_mask=patch_attention_mask,
        )
        image_attn_mask = None
        if patch_attention_mask is not None:
            flattened_mask = patch_attention_mask.flatten(1)
            image_attn_mask = torch.logical_not(flattened_mask)

        return self.multi_modal_projector(image_outputs, image_attn_mask)

    def get_language_model(self) -> torch.nn.Module:
        return self.language_model

    def get_multimodal_embeddings(self,
                                  **kwargs: object) -> MultiModalEmbeddings:
        image_input = self._parse_and_validate_image_input(**kwargs)
        if image_input is None:
            return []
        multimodal_embeddings = self._process_image_input(image_input)
        return multimodal_embeddings

    def get_input_embeddings(
        self,
        input_ids: torch.Tensor,
        multimodal_embeddings: Optional[MultiModalEmbeddings] = None,
    ) -> torch.Tensor:
        inputs_embeds = self.language_model.get_input_embeddings(input_ids)
        if multimodal_embeddings is not None \
            and len(multimodal_embeddings) != 0:
            inputs_embeds = merge_multimodal_embeddings(
                input_ids, inputs_embeds, multimodal_embeddings,
                self.config.image_token_index)
        return inputs_embeds

    def forward(
        self,
        input_ids: torch.Tensor,
        positions: torch.Tensor,
        intermediate_tensors: Optional[IntermediateTensors] = None,
        inputs_embeds: Optional[torch.Tensor] = None,
        **kwargs: object,
    ) -> Union[torch.Tensor, IntermediateTensors]:
        if inputs_embeds is None:
            multimodal_embeddings = self.get_multimodal_embeddings(**kwargs)
            # always pass the input via `inputs_embeds`
            # to make sure the computation graph is consistent
            inputs_embeds = self.get_input_embeddings(input_ids,
                                                      multimodal_embeddings)
            input_ids = None

        hidden_states = self.language_model(
            input_ids,
            positions,
            intermediate_tensors,
            inputs_embeds=inputs_embeds,
        )

        return hidden_states

    def compute_logits(self, hidden_states: torch.Tensor,
                       sampling_metadata: SamplingMetadata) -> torch.Tensor:
        logits = self.logits_processor(self.lm_head, hidden_states,
                                       sampling_metadata)
        return logits

    def load_weights(self, weights: Iterable[tuple[str, torch.Tensor]]):
        loader = AutoWeightsLoader(self)
        loader.load_weights(weights, mapper=self.hf_to_vllm_mapper)

config instance-attribute

config = config

hf_to_vllm_mapper class-attribute instance-attribute

hf_to_vllm_mapper = WeightsMapper(
    orig_to_new_prefix={
        "model.language_model.": "language_model.model.",
        "model.vision_tower.": "vision_tower.",
        "model.multi_modal_projector.": "multi_modal_projector.",
        "language_model.model": "language_model",
        "language_model.lm_head": "lm_head",
    },
    orig_to_new_suffix={"router.weight": "router_weight"},
)

language_model instance-attribute

language_model = AriaTextModel(
    vllm_config=with_hf_config(text_config),
    prefix=maybe_prefix(prefix, "language_model.model"),
)

lm_head instance-attribute

lm_head = ParallelLMHead(
    unpadded_vocab_size,
    hidden_size,
    org_num_embeddings=org_vocab_size,
    quant_config=quant_config,
)

logits_processor instance-attribute

logits_processor = LogitsProcessor(
    unpadded_vocab_size, vocab_size, logit_scale
)

multi_modal_projector instance-attribute

multi_modal_projector = AriaProjector(config)

pad_token_id instance-attribute

pad_token_id = (
    pad_token_id if pad_token_id is not None else -1
)

unpadded_vocab_size instance-attribute

unpadded_vocab_size = vocab_size

vision_tower instance-attribute

vision_tower = AriaVisionTransformer(
    vision_config,
    quant_config=quant_config,
    prefix=f"{prefix}.vision_tower",
)

vocab_size instance-attribute

vocab_size = vocab_size

__init__

__init__(vllm_config: VllmConfig, prefix: str = '')

Source code in vllm/model_executor/models/aria.py

def __init__(
    self,
    vllm_config: VllmConfig,
    prefix: str = "",
):
    super().__init__()
    config = vllm_config.model_config.hf_config
    quant_config = vllm_config.quant_config

    self.config = config
    self.vision_tower = AriaVisionTransformer(
        config.vision_config,
        quant_config=quant_config,
        prefix=f"{prefix}.vision_tower",
    )
    self.multi_modal_projector = AriaProjector(config)
    self.vocab_size = config.text_config.vocab_size
    self.language_model = AriaTextModel(
        vllm_config=vllm_config.with_hf_config(config.text_config),
        prefix=maybe_prefix(prefix, "language_model.model"),
    )
    self.pad_token_id = (self.config.pad_token_id
                         if self.config.pad_token_id is not None else -1)
    self.unpadded_vocab_size = config.text_config.vocab_size
    self.lm_head = ParallelLMHead(
        self.unpadded_vocab_size,
        config.text_config.hidden_size,
        org_num_embeddings=self.language_model.org_vocab_size,
        quant_config=quant_config,
    )
    logit_scale = getattr(config, "logit_scale", 1.0)
    self.logits_processor = LogitsProcessor(self.unpadded_vocab_size,
                                            self.vocab_size, logit_scale)

_create_patch_attention_mask

_create_patch_attention_mask(
    pixel_mask: Optional[Tensor],
) -> Tensor

Source code in vllm/model_executor/models/aria.py

def _create_patch_attention_mask(
        self, pixel_mask: Optional[torch.Tensor]) -> torch.Tensor:
    if pixel_mask is None:
        return None

    patches_subgrid = pixel_mask.unfold(
        dimension=1,
        size=self.vision_tower.config.patch_size,
        step=self.vision_tower.config.patch_size,
    ).unfold(
        dimension=2,
        size=self.vision_tower.config.patch_size,
        step=self.vision_tower.config.patch_size,
    )
    return (patches_subgrid.sum(dim=(-1, -2)) > 0).bool()

_parse_and_validate_image_input

_parse_and_validate_image_input(
    **kwargs: object,
) -> Optional[AriaImagePixelInputs]

Source code in vllm/model_executor/models/aria.py

def _parse_and_validate_image_input(
        self, **kwargs: object) -> Optional[AriaImagePixelInputs]:
    pixel_values = kwargs.pop("pixel_values", None)
    pixel_mask = kwargs.pop("pixel_mask", None)

    if pixel_values is None:
        return None

    if not isinstance(pixel_values, (torch.Tensor, list)):
        raise ValueError("Incorrect type of pixel values. "
                         f"Got type: {type(pixel_values)}")

    pixel_values = self._validate_image_sizes(pixel_values)
    pixel_values = flatten_bn(pixel_values, concat=True)

    if pixel_mask is not None:
        if not isinstance(pixel_mask, (torch.Tensor, list)):
            raise ValueError("Incorrect type of pixel mask. "
                             f"Got type: {type(pixel_mask)}")

        pixel_mask = flatten_bn(pixel_mask, concat=True)

    return AriaImagePixelInputs(
        pixel_values=pixel_values,
        pixel_mask=pixel_mask,
    )

_process_image_input

_process_image_input(
    image_input: AriaImagePixelInputs,
) -> tuple[Tensor, Tensor]

Source code in vllm/model_executor/models/aria.py

def _process_image_input(
    self, image_input: AriaImagePixelInputs
) -> tuple[torch.Tensor, torch.Tensor]:
    assert self.vision_tower is not None

    pixel_values = image_input['pixel_values']
    pixel_mask = image_input['pixel_mask']

    patch_attention_mask = self._create_patch_attention_mask(pixel_mask)

    image_outputs = self.vision_tower(
        pixel_values=pixel_values,
        patch_attention_mask=patch_attention_mask,
    )
    image_attn_mask = None
    if patch_attention_mask is not None:
        flattened_mask = patch_attention_mask.flatten(1)
        image_attn_mask = torch.logical_not(flattened_mask)

    return self.multi_modal_projector(image_outputs, image_attn_mask)

_validate_image_sizes

_validate_image_sizes(images: list[Tensor]) -> list[Tensor]

Source code in vllm/model_executor/models/aria.py

def _validate_image_sizes(
        self, images: list[torch.Tensor]) -> list[torch.Tensor]:
    if not all(img.shape == images[0].shape for img in images):
        raise ValueError("All images must be the same size")
    return images

compute_logits

compute_logits(
    hidden_states: Tensor,
    sampling_metadata: SamplingMetadata,
) -> Tensor

Source code in vllm/model_executor/models/aria.py

def compute_logits(self, hidden_states: torch.Tensor,
                   sampling_metadata: SamplingMetadata) -> torch.Tensor:
    logits = self.logits_processor(self.lm_head, hidden_states,
                                   sampling_metadata)
    return logits

forward

forward(
    input_ids: Tensor,
    positions: Tensor,
    intermediate_tensors: Optional[
        IntermediateTensors
    ] = None,
    inputs_embeds: Optional[Tensor] = None,
    **kwargs: object,
) -> Union[Tensor, IntermediateTensors]

Source code in vllm/model_executor/models/aria.py

def forward(
    self,
    input_ids: torch.Tensor,
    positions: torch.Tensor,
    intermediate_tensors: Optional[IntermediateTensors] = None,
    inputs_embeds: Optional[torch.Tensor] = None,
    **kwargs: object,
) -> Union[torch.Tensor, IntermediateTensors]:
    if inputs_embeds is None:
        multimodal_embeddings = self.get_multimodal_embeddings(**kwargs)
        # always pass the input via `inputs_embeds`
        # to make sure the computation graph is consistent
        inputs_embeds = self.get_input_embeddings(input_ids,
                                                  multimodal_embeddings)
        input_ids = None

    hidden_states = self.language_model(
        input_ids,
        positions,
        intermediate_tensors,
        inputs_embeds=inputs_embeds,
    )

    return hidden_states

get_input_embeddings

get_input_embeddings(
    input_ids: Tensor,
    multimodal_embeddings: Optional[
        MultiModalEmbeddings
    ] = None,
) -> Tensor

Source code in vllm/model_executor/models/aria.py

def get_input_embeddings(
    self,
    input_ids: torch.Tensor,
    multimodal_embeddings: Optional[MultiModalEmbeddings] = None,
) -> torch.Tensor:
    inputs_embeds = self.language_model.get_input_embeddings(input_ids)
    if multimodal_embeddings is not None \
        and len(multimodal_embeddings) != 0:
        inputs_embeds = merge_multimodal_embeddings(
            input_ids, inputs_embeds, multimodal_embeddings,
            self.config.image_token_index)
    return inputs_embeds

get_language_model

get_language_model() -> Module

Source code in vllm/model_executor/models/aria.py

def get_language_model(self) -> torch.nn.Module:
    return self.language_model

get_multimodal_embeddings

get_multimodal_embeddings(
    **kwargs: object,
) -> MultiModalEmbeddings

Source code in vllm/model_executor/models/aria.py

def get_multimodal_embeddings(self,
                              **kwargs: object) -> MultiModalEmbeddings:
    image_input = self._parse_and_validate_image_input(**kwargs)
    if image_input is None:
        return []
    multimodal_embeddings = self._process_image_input(image_input)
    return multimodal_embeddings

get_placeholder_str classmethod

get_placeholder_str(modality: str, i: int) -> Optional[str]

Source code in vllm/model_executor/models/aria.py

@classmethod
def get_placeholder_str(cls, modality: str, i: int) -> Optional[str]:
    if modality.startswith("image"):
        return "<|fim_prefix|><|img|><|fim_suffix|>"

    raise ValueError("Only image modality is supported")

load_weights

load_weights(weights: Iterable[tuple[str, Tensor]])

Source code in vllm/model_executor/models/aria.py

def load_weights(self, weights: Iterable[tuple[str, torch.Tensor]]):
    loader = AutoWeightsLoader(self)
    loader.load_weights(weights, mapper=self.hf_to_vllm_mapper)

AriaFusedMoE

Bases: FusedMoE

Source code in vllm/model_executor/models/aria.py

class AriaFusedMoE(FusedMoE):

    def weight_loader(self, param: nn.Parameter, loaded_weight: torch.Tensor,
                      shard_id: str) -> None:
        # Override the weight_loader to handle the expert weights in the Aria
        # model, which are already packed with experts, and merge the gate and
        # up weights for each expert.
        # Note: Loading expert weights with quantization is not supported
        tp_rank = get_tensor_model_parallel_rank()
        if shard_id == 'w13':
            # the shape of loaded_weight is
            # (num_experts, hidden_size, 2 * moe_intermediate_size)
            if self.tp_size > 1:
                up, gate = loaded_weight.chunk(2, dim=-1)
                up_current_rank = up.chunk(self.tp_size, dim=-1)[tp_rank]
                gate_current_rank = gate.chunk(self.tp_size, dim=-1)[tp_rank]
                up_and_gate = torch.cat([up_current_rank, gate_current_rank],
                                        dim=-1).transpose(1, 2)
                param.data.copy_(up_and_gate)
            else:
                param.data.copy_(loaded_weight.transpose(1, 2))
        elif shard_id == 'w2':
            # the shape of loaded_weight is
            # (num_experts, moe_intermediate_size, hidden_size)
            if self.tp_size > 1:
                down_current_rank = loaded_weight.chunk(self.tp_size,
                                                        dim=1)[tp_rank]
                param.data.copy_(down_current_rank.transpose(1, 2))
            else:
                param.data.copy_(loaded_weight.transpose(1, 2))

weight_loader

weight_loader(
    param: Parameter, loaded_weight: Tensor, shard_id: str
) -> None

Source code in vllm/model_executor/models/aria.py

def weight_loader(self, param: nn.Parameter, loaded_weight: torch.Tensor,
                  shard_id: str) -> None:
    # Override the weight_loader to handle the expert weights in the Aria
    # model, which are already packed with experts, and merge the gate and
    # up weights for each expert.
    # Note: Loading expert weights with quantization is not supported
    tp_rank = get_tensor_model_parallel_rank()
    if shard_id == 'w13':
        # the shape of loaded_weight is
        # (num_experts, hidden_size, 2 * moe_intermediate_size)
        if self.tp_size > 1:
            up, gate = loaded_weight.chunk(2, dim=-1)
            up_current_rank = up.chunk(self.tp_size, dim=-1)[tp_rank]
            gate_current_rank = gate.chunk(self.tp_size, dim=-1)[tp_rank]
            up_and_gate = torch.cat([up_current_rank, gate_current_rank],
                                    dim=-1).transpose(1, 2)
            param.data.copy_(up_and_gate)
        else:
            param.data.copy_(loaded_weight.transpose(1, 2))
    elif shard_id == 'w2':
        # the shape of loaded_weight is
        # (num_experts, moe_intermediate_size, hidden_size)
        if self.tp_size > 1:
            down_current_rank = loaded_weight.chunk(self.tp_size,
                                                    dim=1)[tp_rank]
            param.data.copy_(down_current_rank.transpose(1, 2))
        else:
            param.data.copy_(loaded_weight.transpose(1, 2))

AriaImagePixelInputs

Bases: TypedDict

Source code in vllm/model_executor/models/aria.py

class AriaImagePixelInputs(TypedDict):
    pixel_values: torch.Tensor
    pixel_mask: Optional[torch.Tensor]
    """
    Shape:
        pixel_values: `(batch_size * num_images, num_channels, height, width)`
        pixel_mask: `(batch_size * num_images, height, width)`
    """

pixel_mask instance-attribute

pixel_mask: Optional[Tensor]

Shape

pixel_values: (batch_size * num_images, num_channels, height, width) pixel_mask: (batch_size * num_images, height, width)

pixel_values instance-attribute

pixel_values: Tensor

AriaMultiModalProcessor

Bases: BaseMultiModalProcessor[AriaProcessingInfo]

Source code in vllm/model_executor/models/aria.py

class AriaMultiModalProcessor(BaseMultiModalProcessor[AriaProcessingInfo]):

    def _get_mm_fields_config(
        self,
        hf_inputs: BatchFeature,
        hf_processor_mm_kwargs: Mapping[str, object],
    ) -> Mapping[str, MultiModalFieldConfig]:
        return dict(
            pixel_values=MultiModalFieldConfig.batched("image"),
            pixel_mask=MultiModalFieldConfig.batched("image"),
        )

    def _get_prompt_updates(
        self,
        mm_items: MultiModalDataItems,
        hf_processor_mm_kwargs: Mapping[str, object],
        out_mm_kwargs: MultiModalKwargs,
    ) -> Sequence[PromptUpdate]:
        hf_config = self.info.get_hf_config()
        image_token_id = hf_config.image_token_index

        num_image_tokens = self.info.get_num_image_tokens()

        return [
            PromptReplacement(
                modality="image",
                target=[image_token_id],
                replacement=[image_token_id] * num_image_tokens,
            )
        ]

_get_mm_fields_config

_get_mm_fields_config(
    hf_inputs: BatchFeature,
    hf_processor_mm_kwargs: Mapping[str, object],
) -> Mapping[str, MultiModalFieldConfig]

Source code in vllm/model_executor/models/aria.py

def _get_mm_fields_config(
    self,
    hf_inputs: BatchFeature,
    hf_processor_mm_kwargs: Mapping[str, object],
) -> Mapping[str, MultiModalFieldConfig]:
    return dict(
        pixel_values=MultiModalFieldConfig.batched("image"),
        pixel_mask=MultiModalFieldConfig.batched("image"),
    )

_get_prompt_updates

_get_prompt_updates(
    mm_items: MultiModalDataItems,
    hf_processor_mm_kwargs: Mapping[str, object],
    out_mm_kwargs: MultiModalKwargs,
) -> Sequence[PromptUpdate]

Source code in vllm/model_executor/models/aria.py

def _get_prompt_updates(
    self,
    mm_items: MultiModalDataItems,
    hf_processor_mm_kwargs: Mapping[str, object],
    out_mm_kwargs: MultiModalKwargs,
) -> Sequence[PromptUpdate]:
    hf_config = self.info.get_hf_config()
    image_token_id = hf_config.image_token_index

    num_image_tokens = self.info.get_num_image_tokens()

    return [
        PromptReplacement(
            modality="image",
            target=[image_token_id],
            replacement=[image_token_id] * num_image_tokens,
        )
    ]

AriaProcessingInfo

Bases: BaseProcessingInfo

Source code in vllm/model_executor/models/aria.py

class AriaProcessingInfo(BaseProcessingInfo):

    def get_hf_config(self):
        return self.ctx.get_hf_config(AriaConfig)

    def get_vision_config(self):
        return self.get_hf_config().vision_config

    def get_hf_processor(self, **kwargs: object):
        return self.ctx.get_hf_processor(AriaProcessor, **kwargs)

    def get_supported_mm_limits(self) -> Mapping[str, Optional[int]]:
        return {"image": None}

    def get_num_image_tokens(self) -> int:
        hf_config = self.get_hf_config()
        return max(hf_config.projector_patch_to_query_dict.values())

get_hf_config

get_hf_config()

Source code in vllm/model_executor/models/aria.py

def get_hf_config(self):
    return self.ctx.get_hf_config(AriaConfig)

get_hf_processor

get_hf_processor(**kwargs: object)

Source code in vllm/model_executor/models/aria.py

def get_hf_processor(self, **kwargs: object):
    return self.ctx.get_hf_processor(AriaProcessor, **kwargs)

get_num_image_tokens

get_num_image_tokens() -> int

Source code in vllm/model_executor/models/aria.py

def get_num_image_tokens(self) -> int:
    hf_config = self.get_hf_config()
    return max(hf_config.projector_patch_to_query_dict.values())

get_supported_mm_limits

get_supported_mm_limits() -> Mapping[str, Optional[int]]

Source code in vllm/model_executor/models/aria.py

def get_supported_mm_limits(self) -> Mapping[str, Optional[int]]:
    return {"image": None}

get_vision_config

get_vision_config()

Source code in vllm/model_executor/models/aria.py

def get_vision_config(self):
    return self.get_hf_config().vision_config

AriaProjector

Bases: Module

A projection module with one cross attention layer and one FFN layer, which projects ViT's outputs into MoE's inputs.

Parameters:

Name	Type	Description	Default
patch_to_query_dict	dict	Maps patch numbers to their corresponding	required
embed_dim	int	Embedding dimension.	required
num_heads	int	Number of attention heads.	required
kv_dim	int	Dimension of key and value.	required
ff_dim	int	Hidden dimension of the feed-forward network.	required
output_dim	int	Output dimension.	required
norm_layer	Module	Normalization layer. Default is nn.LayerNorm.	required

Outputs

A tensor with the shape of (batch_size, query_number, output_dim)

Source code in vllm/model_executor/models/aria.py

class AriaProjector(nn.Module):
    """
    A projection module with one cross attention layer and one FFN layer, which
    projects ViT's outputs into MoE's inputs.

    Args:
        patch_to_query_dict (dict): Maps patch numbers to their corresponding
        query numbers,
            e.g., {1225: 128, 4900: 256}. This allows for different query sizes
            based on image resolution.
        embed_dim (int): Embedding dimension.
        num_heads (int): Number of attention heads.
        kv_dim (int): Dimension of key and value.
        ff_dim (int): Hidden dimension of the feed-forward network.
        output_dim (int): Output dimension.
        norm_layer (nn.Module): Normalization layer. Default is nn.LayerNorm.

    Outputs:
        A tensor with the shape of (batch_size, query_number, output_dim)
    """

    def __init__(self, config: AriaConfig) -> None:
        super().__init__()

        self.patch_to_query_dict = config.projector_patch_to_query_dict
        self.in_features = config.vision_config.hidden_size
        self.num_heads = config.vision_config.num_attention_heads
        self.kv_dim = config.vision_config.hidden_size
        self.hidden_features = config.text_config.hidden_size
        self.output_dim = config.text_config.hidden_size

        self.query = nn.Parameter(
            torch.empty(config.max_value_projector_patch_to_query_dict,
                        self.in_features))

        self.cross_attn = AriaCrossAttention(config)

        self.layer_norm = nn.LayerNorm(self.in_features)
        self.feed_forward = AriaProjectorMLP(self.in_features,
                                             self.hidden_features,
                                             self.output_dim)

    def forward(
        self,
        x: torch.Tensor,
        attn_mask: Optional[torch.Tensor] = None,
    ) -> torch.Tensor:
        batch_size, num_patches = x.shape[0], x.shape[1]

        if num_patches not in self.patch_to_query_dict:
            raise KeyError(f"Number of patches {num_patches} not found in "
                           "patch_to_query_dict amongst possible values "
                           f"{self.patch_to_query_dict.keys()}.")

        query_num = self.patch_to_query_dict[num_patches]

        queries = self.query[:query_num].unsqueeze(0).repeat(batch_size, 1, 1)

        if attn_mask is not None:
            attn_mask = attn_mask.repeat_interleave(self.num_heads, 0)
            attn_mask = attn_mask.unsqueeze(1).expand(-1, queries.size(1), -1)

        attention_out = self.cross_attn(x, queries, attn_mask=attn_mask)

        out = self.feed_forward(self.layer_norm(attention_out))

        return out

cross_attn instance-attribute

cross_attn = AriaCrossAttention(config)

feed_forward instance-attribute

feed_forward = AriaProjectorMLP(
    in_features, hidden_features, output_dim
)

hidden_features instance-attribute

hidden_features = hidden_size

in_features instance-attribute

in_features = hidden_size

kv_dim instance-attribute

kv_dim = hidden_size

layer_norm instance-attribute

layer_norm = LayerNorm(in_features)

num_heads instance-attribute

num_heads = num_attention_heads

output_dim instance-attribute

output_dim = hidden_size

patch_to_query_dict instance-attribute

patch_to_query_dict = projector_patch_to_query_dict

query instance-attribute

query = Parameter(
    empty(
        max_value_projector_patch_to_query_dict, in_features
    )
)

__init__

__init__(config: AriaConfig) -> None

Source code in vllm/model_executor/models/aria.py

def __init__(self, config: AriaConfig) -> None:
    super().__init__()

    self.patch_to_query_dict = config.projector_patch_to_query_dict
    self.in_features = config.vision_config.hidden_size
    self.num_heads = config.vision_config.num_attention_heads
    self.kv_dim = config.vision_config.hidden_size
    self.hidden_features = config.text_config.hidden_size
    self.output_dim = config.text_config.hidden_size

    self.query = nn.Parameter(
        torch.empty(config.max_value_projector_patch_to_query_dict,
                    self.in_features))

    self.cross_attn = AriaCrossAttention(config)

    self.layer_norm = nn.LayerNorm(self.in_features)
    self.feed_forward = AriaProjectorMLP(self.in_features,
                                         self.hidden_features,
                                         self.output_dim)

forward

forward(
    x: Tensor, attn_mask: Optional[Tensor] = None
) -> Tensor

Source code in vllm/model_executor/models/aria.py

def forward(
    self,
    x: torch.Tensor,
    attn_mask: Optional[torch.Tensor] = None,
) -> torch.Tensor:
    batch_size, num_patches = x.shape[0], x.shape[1]

    if num_patches not in self.patch_to_query_dict:
        raise KeyError(f"Number of patches {num_patches} not found in "
                       "patch_to_query_dict amongst possible values "
                       f"{self.patch_to_query_dict.keys()}.")

    query_num = self.patch_to_query_dict[num_patches]

    queries = self.query[:query_num].unsqueeze(0).repeat(batch_size, 1, 1)

    if attn_mask is not None:
        attn_mask = attn_mask.repeat_interleave(self.num_heads, 0)
        attn_mask = attn_mask.unsqueeze(1).expand(-1, queries.size(1), -1)

    attention_out = self.cross_attn(x, queries, attn_mask=attn_mask)

    out = self.feed_forward(self.layer_norm(attention_out))

    return out

AriaProjectorMLP

Bases: Module

Source code in vllm/model_executor/models/aria.py

class AriaProjectorMLP(nn.Module):

    def __init__(
        self,
        in_features: int,
        hidden_features: int,
        output_dim: int,
    ) -> None:
        super().__init__()

        self.linear_in = ColumnParallelLinear(in_features,
                                              hidden_features,
                                              bias=False)
        self.linear_out = RowParallelLinear(hidden_features,
                                            output_dim,
                                            bias=False)
        self.act = get_act_fn("gelu_new")

    def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
        hidden_states, _ = self.linear_in(hidden_states)
        hidden_states = self.act(hidden_states)
        hidden_states, _ = self.linear_out(hidden_states)
        return hidden_states

act instance-attribute

act = get_act_fn('gelu_new')

linear_in instance-attribute

linear_in = ColumnParallelLinear(
    in_features, hidden_features, bias=False
)

linear_out instance-attribute

linear_out = RowParallelLinear(
    hidden_features, output_dim, bias=False
)

__init__

__init__(
    in_features: int, hidden_features: int, output_dim: int
) -> None

Source code in vllm/model_executor/models/aria.py

def __init__(
    self,
    in_features: int,
    hidden_features: int,
    output_dim: int,
) -> None:
    super().__init__()

    self.linear_in = ColumnParallelLinear(in_features,
                                          hidden_features,
                                          bias=False)
    self.linear_out = RowParallelLinear(hidden_features,
                                        output_dim,
                                        bias=False)
    self.act = get_act_fn("gelu_new")

forward

forward(hidden_states: Tensor) -> Tensor

Source code in vllm/model_executor/models/aria.py

def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
    hidden_states, _ = self.linear_in(hidden_states)
    hidden_states = self.act(hidden_states)
    hidden_states, _ = self.linear_out(hidden_states)
    return hidden_states

AriaTextDecoderLayer

Bases: LlamaDecoderLayer

Custom Decoder Layer for the AriaMoE model which modifies the standard LlamaDecoderLayer by replacing the traditional MLP with a Mixture of Experts (MoE) Layer.

Source code in vllm/model_executor/models/aria.py

class AriaTextDecoderLayer(LlamaDecoderLayer):
    """
    Custom Decoder Layer for the AriaMoE model which modifies the standard
    `LlamaDecoderLayer` by replacing the traditional MLP with a Mixture of
    Experts (MoE) Layer.
    """

    def __init__(
        self,
        config: AriaTextConfig,
        cache_config: Optional[CacheConfig] = None,
        quant_config: Optional[QuantizationConfig] = None,
        prefix: str = "",
    ) -> None:
        super().__init__(config, cache_config, quant_config, prefix)
        self.mlp = AriaTextMoELayer(config,
                                    quant_config=quant_config,
                                    prefix=f"{prefix}.mlp")

mlp instance-attribute

mlp = AriaTextMoELayer(
    config,
    quant_config=quant_config,
    prefix=f"{prefix}.mlp",
)

__init__

__init__(
    config: AriaTextConfig,
    cache_config: Optional[CacheConfig] = None,
    quant_config: Optional[QuantizationConfig] = None,
    prefix: str = "",
) -> None

Source code in vllm/model_executor/models/aria.py

def __init__(
    self,
    config: AriaTextConfig,
    cache_config: Optional[CacheConfig] = None,
    quant_config: Optional[QuantizationConfig] = None,
    prefix: str = "",
) -> None:
    super().__init__(config, cache_config, quant_config, prefix)
    self.mlp = AriaTextMoELayer(config,
                                quant_config=quant_config,
                                prefix=f"{prefix}.mlp")

AriaTextMoELayer

Bases: Module

Mixture of Experts (MoE) Layer for the AriaMoE model.

This layer implements the MoE mechanism, which routes input tokens to different experts based on a routing algorithm, processes them through the experts, and then combines the outputs.

Source code in vllm/model_executor/models/aria.py

class AriaTextMoELayer(nn.Module):
    """
    Mixture of Experts (MoE) Layer for the AriaMoE model.

    This layer implements the MoE mechanism, which routes input tokens to
    different experts based on a routing algorithm, processes them through the
    experts, and then combines the outputs.
    """

    def __init__(
        self,
        config: AriaTextConfig,
        quant_config: Optional[QuantizationConfig],
        prefix: str = "",
    ) -> None:
        super().__init__()
        self.config = config

        self.router_weight = nn.Parameter(
            torch.empty(
                (self.config.moe_num_experts, self.config.hidden_size)))

        self.experts = AriaFusedMoE(
            num_experts=config.moe_num_experts,
            top_k=config.moe_topk,
            hidden_size=config.hidden_size,
            intermediate_size=config.intermediate_size,
            quant_config=quant_config,
            reduce_results=True,
            prefix=f"{prefix}.experts",
        )
        self.shared_experts = LlamaMLP(
            config.hidden_size,
            config.intermediate_size * config.moe_num_shared_experts,
            "silu",
            quant_config=quant_config,
            bias=config.mlp_bias,
        )

    def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
        """
        Forward pass of the MoE Layer.

        Args:
            hidden_states (torch.Tensor): Input tensor of shape (batch_size,
            sequence_length, hidden_size).

        Returns:
            torch.Tensor: Output tensor after passing through the MoE layer.
        """

        router_output = torch.nn.functional.linear(hidden_states,
                                                   self.router_weight)

        hidden_states_copy = hidden_states.clone()
        # NOTE: hidden_states will be modified inplace by `FusedMoE`
        sparse_expert_output = self.experts(hidden_states, router_output)
        shared_expert_output = self.shared_experts(hidden_states_copy)

        return sparse_expert_output + shared_expert_output

config instance-attribute

config = config

experts instance-attribute

experts = AriaFusedMoE(
    num_experts=moe_num_experts,
    top_k=moe_topk,
    hidden_size=hidden_size,
    intermediate_size=intermediate_size,
    quant_config=quant_config,
    reduce_results=True,
    prefix=f"{prefix}.experts",
)

router_weight instance-attribute

router_weight = Parameter(
    empty((moe_num_experts, hidden_size))
)

shared_experts instance-attribute

shared_experts = LlamaMLP(
    hidden_size,
    intermediate_size * moe_num_shared_experts,
    "silu",
    quant_config=quant_config,
    bias=mlp_bias,
)

__init__

__init__(
    config: AriaTextConfig,
    quant_config: Optional[QuantizationConfig],
    prefix: str = "",
) -> None

Source code in vllm/model_executor/models/aria.py

def __init__(
    self,
    config: AriaTextConfig,
    quant_config: Optional[QuantizationConfig],
    prefix: str = "",
) -> None:
    super().__init__()
    self.config = config

    self.router_weight = nn.Parameter(
        torch.empty(
            (self.config.moe_num_experts, self.config.hidden_size)))

    self.experts = AriaFusedMoE(
        num_experts=config.moe_num_experts,
        top_k=config.moe_topk,
        hidden_size=config.hidden_size,
        intermediate_size=config.intermediate_size,
        quant_config=quant_config,
        reduce_results=True,
        prefix=f"{prefix}.experts",
    )
    self.shared_experts = LlamaMLP(
        config.hidden_size,
        config.intermediate_size * config.moe_num_shared_experts,
        "silu",
        quant_config=quant_config,
        bias=config.mlp_bias,
    )

forward

forward(hidden_states: Tensor) -> Tensor

Forward pass of the MoE Layer.

Parameters:

Name	Type	Description	Default
hidden_states	Tensor	Input tensor of shape (batch_size,	required

Returns:

Type	Description
Tensor	torch.Tensor: Output tensor after passing through the MoE layer.

Source code in vllm/model_executor/models/aria.py

def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
    """
    Forward pass of the MoE Layer.

    Args:
        hidden_states (torch.Tensor): Input tensor of shape (batch_size,
        sequence_length, hidden_size).

    Returns:
        torch.Tensor: Output tensor after passing through the MoE layer.
    """

    router_output = torch.nn.functional.linear(hidden_states,
                                               self.router_weight)

    hidden_states_copy = hidden_states.clone()
    # NOTE: hidden_states will be modified inplace by `FusedMoE`
    sparse_expert_output = self.experts(hidden_states, router_output)
    shared_expert_output = self.shared_experts(hidden_states_copy)

    return sparse_expert_output + shared_expert_output

AriaTextModel

Bases: LlamaModel, SupportsQuant

Custom LlamaModel for the AriaMoE model which modifies the standard LlamaModel by replacing the LlamaDecoderLayer with MoEDecoderLayer.

Source code in vllm/model_executor/models/aria.py

class AriaTextModel(LlamaModel, SupportsQuant):
    """
    Custom LlamaModel for the AriaMoE model which modifies the standard
    LlamaModel by replacing the `LlamaDecoderLayer` with `MoEDecoderLayer`.
    """
    packed_modules_mapping = {
        "qkv_proj": ["q_proj", "k_proj", "v_proj"],
        "gate_up_proj": ["gate_proj", "up_proj"],
        "experts.w13_weight": ["experts.fc1.weight"],
        "experts.w2_weight": ["experts.fc2.weight"],
    }

    def __init__(self, *, vllm_config: VllmConfig, prefix: str = ""):
        super().__init__(vllm_config=vllm_config,
                         prefix=prefix,
                         layer_type=AriaTextDecoderLayer)

    # Adapted from LlamaModel.load_weights with the modification of adding
    # the expert weights mapping to `stacked_params_mapping`
    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"),
            (".gate_up_proj", ".gate_proj", 0),
            (".gate_up_proj", ".up_proj", 1),
            ("experts.w13_weight", "experts.fc1.weight", 'w13'),
            ("experts.w2_weight", "experts.fc2.weight", 'w2'),
        ]
        params_dict = dict(self.named_parameters())
        loaded_params: set[str] = set()
        for name, loaded_weight in weights:
            if "rotary_emb.inv_freq" in name:
                continue
            if ("rotary_emb.cos_cached" in name
                    or "rotary_emb.sin_cached" in name):
                # Models trained using ColossalAI may include these tensors in
                # the checkpoint. Skip them.
                continue
            if (self.quant_config is not None and
                (scale_name := self.quant_config.get_cache_scale(name))):
                # Loading kv cache quantization scales
                param = params_dict[scale_name]
                weight_loader = getattr(param, "weight_loader",
                                        default_weight_loader)
                loaded_weight = (loaded_weight if loaded_weight.dim() == 0 else
                                 loaded_weight[0])
                weight_loader(param, loaded_weight)
                loaded_params.add(scale_name)
                continue
            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)
                # Skip loading extra bias for GPTQ models.
                if name.endswith(".bias") and name not in params_dict:
                    continue

                if is_pp_missing_parameter(name, self):
                    continue

                param = params_dict[name]
                weight_loader = param.weight_loader
                weight_loader(param, loaded_weight, shard_id)
                break
            else:
                # Skip loading extra bias for GPTQ models.
                if name.endswith(".bias") and name not in params_dict:
                    continue
                # Remapping the name of FP8 kv-scale.
                name = maybe_remap_kv_scale_name(name, params_dict)
                if name is None:
                    continue

                if is_pp_missing_parameter(name, self):
                    continue

                param = params_dict[name]
                weight_loader = getattr(param, "weight_loader",
                                        default_weight_loader)
                weight_loader(param, loaded_weight)
            loaded_params.add(name)
        return loaded_params

packed_modules_mapping class-attribute instance-attribute

packed_modules_mapping = {
    "qkv_proj": ["q_proj", "k_proj", "v_proj"],
    "gate_up_proj": ["gate_proj", "up_proj"],
    "experts.w13_weight": ["experts.fc1.weight"],
    "experts.w2_weight": ["experts.fc2.weight"],
}

__init__

__init__(*, vllm_config: VllmConfig, prefix: str = '')

Source code in vllm/model_executor/models/aria.py

def __init__(self, *, vllm_config: VllmConfig, prefix: str = ""):
    super().__init__(vllm_config=vllm_config,
                     prefix=prefix,
                     layer_type=AriaTextDecoderLayer)

load_weights

load_weights(
    weights: Iterable[tuple[str, Tensor]],
) -> set[str]

Source code in vllm/model_executor/models/aria.py

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"),
        (".gate_up_proj", ".gate_proj", 0),
        (".gate_up_proj", ".up_proj", 1),
        ("experts.w13_weight", "experts.fc1.weight", 'w13'),
        ("experts.w2_weight", "experts.fc2.weight", 'w2'),
    ]
    params_dict = dict(self.named_parameters())
    loaded_params: set[str] = set()
    for name, loaded_weight in weights:
        if "rotary_emb.inv_freq" in name:
            continue
        if ("rotary_emb.cos_cached" in name
                or "rotary_emb.sin_cached" in name):
            # Models trained using ColossalAI may include these tensors in
            # the checkpoint. Skip them.
            continue
        if (self.quant_config is not None and
            (scale_name := self.quant_config.get_cache_scale(name))):
            # Loading kv cache quantization scales
            param = params_dict[scale_name]
            weight_loader = getattr(param, "weight_loader",
                                    default_weight_loader)
            loaded_weight = (loaded_weight if loaded_weight.dim() == 0 else
                             loaded_weight[0])
            weight_loader(param, loaded_weight)
            loaded_params.add(scale_name)
            continue
        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)
            # Skip loading extra bias for GPTQ models.
            if name.endswith(".bias") and name not in params_dict:
                continue

            if is_pp_missing_parameter(name, self):
                continue

            param = params_dict[name]
            weight_loader = param.weight_loader
            weight_loader(param, loaded_weight, shard_id)
            break
        else:
            # Skip loading extra bias for GPTQ models.
            if name.endswith(".bias") and name not in params_dict:
                continue
            # Remapping the name of FP8 kv-scale.
            name = maybe_remap_kv_scale_name(name, params_dict)
            if name is None:
                continue

            if is_pp_missing_parameter(name, self):
                continue

            param = params_dict[name]
            weight_loader = getattr(param, "weight_loader",
                                    default_weight_loader)
            weight_loader(param, loaded_weight)
        loaded_params.add(name)
    return loaded_params

AriaVisionTransformer

Bases: Idefics2VisionTransformer, SupportsQuant

Source code in vllm/model_executor/models/aria.py

class AriaVisionTransformer(Idefics3VisionTransformer, SupportsQuant):
    packed_modules_mapping = {"qkv_proj": ["q_proj", "k_proj", "v_proj"]}

    def __init__(
        self,
        config: Idefics2VisionConfig,
        quant_config: Optional[QuantizationConfig] = None,
        prefix: str = "",
    ) -> None:
        super().__init__(config, quant_config=quant_config, prefix=prefix)
        # Unlike Idefics3VisionTransformer which uses LayerNorm after the
        # final layer, Aria omits this normalization, so we replace it with an
        # Identity layer
        self.post_layernorm = nn.Identity()

    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"),
        ]
        params_dict = dict(self.named_parameters())
        loaded_params: set[str] = set()
        for name, loaded_weight in weights:

            # NOTE: post_layernorm is not used in Aria
            if "post_layernorm" in name:
                continue

            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)
                param = params_dict[name]
                weight_loader = param.weight_loader
                weight_loader(param, loaded_weight, shard_id)
                break
            else:
                param = params_dict[name]
                weight_loader = getattr(param, "weight_loader",
                                        default_weight_loader)
                weight_loader(param, loaded_weight)
            loaded_params.add(name)
        return loaded_params

packed_modules_mapping class-attribute instance-attribute

packed_modules_mapping = {
    "qkv_proj": ["q_proj", "k_proj", "v_proj"]
}

post_layernorm instance-attribute

post_layernorm = Identity()

__init__

__init__(
    config: Idefics2VisionConfig,
    quant_config: Optional[QuantizationConfig] = None,
    prefix: str = "",
) -> None

Source code in vllm/model_executor/models/aria.py

def __init__(
    self,
    config: Idefics2VisionConfig,
    quant_config: Optional[QuantizationConfig] = None,
    prefix: str = "",
) -> None:
    super().__init__(config, quant_config=quant_config, prefix=prefix)
    # Unlike Idefics3VisionTransformer which uses LayerNorm after the
    # final layer, Aria omits this normalization, so we replace it with an
    # Identity layer
    self.post_layernorm = nn.Identity()

load_weights

load_weights(
    weights: Iterable[tuple[str, Tensor]],
) -> set[str]

Source code in vllm/model_executor/models/aria.py

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"),
    ]
    params_dict = dict(self.named_parameters())
    loaded_params: set[str] = set()
    for name, loaded_weight in weights:

        # NOTE: post_layernorm is not used in Aria
        if "post_layernorm" in name:
            continue

        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)
            param = params_dict[name]
            weight_loader = param.weight_loader
            weight_loader(param, loaded_weight, shard_id)
            break
        else:
            param = params_dict[name]
            weight_loader = getattr(param, "weight_loader",
                                    default_weight_loader)
            weight_loader(param, loaded_weight)
        loaded_params.add(name)
    return loaded_params