Hugging Face's logo

Transformers documentation

Ernie 4.5 VL MoE

You are viewing main version, which requires installation from source. If you'd like regular pip install, checkout the latest stable version (v5.14.0).
Hugging Face's logo
Join the Hugging Face community

and get access to the augmented documentation experience

to get started

This model was contributed to Hugging Face Transformers on 2025-12-19.

FlashAttention SDPA Tensor parallelism

Ernie 4.5 VL MoE

Overview

The Ernie 4.5 VL MoE model was released in the Ernie 4.5 Model Family release by baidu. This family of models contains multiple different architectures and model sizes. The Vision-Language series in specific is composed of a novel multimodal heterogeneous structure, sharing parameters across modalities and dedicating parameters to specific modalities. This becomes especially apparent in the Mixture of Expert (MoE) which is composed of

  • Dedicated Text Experts
  • Dedicated Vision Experts
  • Shared Experts

This architecture has the advantage to enhance multimodal understanding without compromising, and even improving, performance on text-related tasks. A more detailed breakdown is given in the Technical Report.

Other models from the family can be found at Ernie 4.5 and at Ernie 4.5 MoE.

Usage

The example below demonstrates how to generate text based on an image with Pipeline or the AutoModel class.

Pipeline
AutoModel
from transformers import pipeline


pipe = pipeline(
    task="image-text-to-text",
    model="baidu/ERNIE-4.5-VL-28B-A3B-PT",
    device_map="auto",
    revision="refs/pr/11",
)
message = [
    {
        "role": "user",
        "content": [
            {"type": "text", "text": "What kind of dog is this?"},
            {
                "type": "image",
                "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/pipeline-cat-chonk.jpeg",
            },
        ],
    }
]
print(pipe(text=message, max_new_tokens=20, return_full_text=False))

Using Ernie 4.5 VL MoE with video input is similar to using it with image input. The model can process video data and generate text based on the content of the video.

from transformers import AutoModelForImageTextToText, AutoProcessor


model = AutoModelForImageTextToText.from_pretrained(
    "baidu/ERNIE-4.5-VL-28B-A3B-PT",
    device_map="auto",  # Use tp_plan="auto" instead to enable Tensor Parallelism!
    revision="refs/pr/11",
)
processor = AutoProcessor.from_pretrained("baidu/ERNIE-4.5-VL-28B-A3B-PT", revision="refs/pr/11")
message = [
    {
        "role": "user",
        "content": [
            {"type": "text", "text": "Please describe what you can see during this video."},
            {
                "type": "video",
                "url": "https://huggingface.co/datasets/raushan-testing-hf/videos-test/resolve/main/tiny_video.mp4",
            },
        ],
    }
]

inputs = processor.apply_chat_template(
    message,
    add_generation_prompt=True,
    tokenize=True,
    return_dict=True,
    return_tensors="pt"
).to(model.device)

generated_ids = model.generate(**inputs, max_new_tokens=128)
generated_ids_trimmed = [
    out_ids[len(in_ids) :] for in_ids, out_ids in zip(inputs.input_ids, generated_ids)
]
output_text = processor.batch_decode(
    generated_ids_trimmed, skip_special_tokens=True, clean_up_tokenization_spaces=False
)
print(output_text)

Ernie4_5_VLMoeConfig

class transformers.Ernie4_5_VLMoeConfig

< >

( transformers_version: str | None = Nonearchitectures: list[str] | None = Noneoutput_hidden_states: bool | None = Falsereturn_dict: bool | None = Truedtype: typing.Union[str, ForwardRef('torch.dtype'), NoneType] = Nonechunk_size_feed_forward: int = 0is_encoder_decoder: bool = Falseid2label: dict[int, str] | dict[str, str] | None = Nonelabel2id: dict[str, int] | dict[str, str] | None = Noneproblem_type: typing.Optional[typing.Literal['regression', 'single_label_classification', 'multi_label_classification']] = Nonetext_config: dict | transformers.configuration_utils.PreTrainedConfig | None = Nonevision_config: dict | transformers.configuration_utils.PreTrainedConfig | None = Noneimage_start_token_id: int = 101304image_end_token_id: int = 101305image_token_id: int = 100295video_start_token_id: int = 101306video_end_token_id: int = 101307video_token_id: int = 103367tie_word_embeddings: bool = True )

Parameters

  • text_config (Union[dict, ~configuration_utils.PreTrainedConfig], optional) — The config object or dictionary of the text backbone.
  • vision_config (Union[dict, ~configuration_utils.PreTrainedConfig], optional) — The config object or dictionary of the vision backbone.
  • image_start_token_id (int, optional, defaults to 101304) — The image token index to encode the start of image.
  • image_end_token_id (int, optional, defaults to 101305) — The image token index to encode the end of image.
  • image_token_id (int, optional, defaults to 100295) — The image token index to encode the image prompt.
  • video_start_token_id (int, optional, defaults to 101306) — The video token index to encode the start of video.
  • video_end_token_id (int, optional, defaults to 101307) — The video token index to encode the end of video.
  • video_token_id (int, optional, defaults to 103367) — The video token index to encode the video prompt.
  • tie_word_embeddings (bool, optional, defaults to True) — Whether to tie weight embeddings according to model’s tied_weights_keys mapping.

This is the configuration class to store the configuration of a Ernie4_5_VLMoeModel. It is used to instantiate a Ernie4 5 Vl Moe model according to the specified arguments, defining the model architecture. Instantiating a configuration with the defaults will yield a similar configuration to that of the baidu/ERNIE-4.5-VL-28B-A3B-PT

Configuration objects inherit from PreTrainedConfig and can be used to control the model outputs. Read the documentation from PreTrainedConfig for more information.

Example:

>>> from transformers import Ernie4_5_VLMoeForConditionalGeneration, Ernie4_5_VLMoeConfig

>>> # Initializing a Ernie4_5_VLMoe style configuration
>>> configuration = Ernie4_5_VLMoeConfig()

>>> # Initializing a model from the Ernie 4.5 VL 28B A3B configuration
>>> model = Ernie4_5_VLMoeForConditionalGeneration(configuration)

>>> # Accessing the model configuration
>>> configuration = model.config

Ernie4_5_VLMoeTextConfig

class transformers.Ernie4_5_VLMoeTextConfig

< >

( transformers_version: str | None = Nonearchitectures: list[str] | None = Noneoutput_hidden_states: bool | None = Falsereturn_dict: bool | None = Truedtype: typing.Union[str, ForwardRef('torch.dtype'), NoneType] = Nonechunk_size_feed_forward: int = 0is_encoder_decoder: bool = Falseid2label: dict[int, str] | dict[str, str] | None = Nonelabel2id: dict[str, int] | dict[str, str] | None = Noneproblem_type: typing.Optional[typing.Literal['regression', 'single_label_classification', 'multi_label_classification']] = Nonevocab_size: int = 103424pad_token_id: int | None = Nonebos_token_id: int | None = Noneeos_token_id: int | list[int] | None = Nonehidden_size: int = 2560intermediate_size: int = 12288num_hidden_layers: int = 28num_attention_heads: int = 20num_key_value_heads: int | None = 4hidden_act: str = 'silu'max_position_embeddings: int = 131072initializer_range: float = 0.02rms_norm_eps: float = 1e-05use_cache: bool = Truetie_word_embeddings: bool = Truerope_parameters: transformers.modeling_rope_utils.RopeParameters | dict | None = Noneuse_bias: bool | None = Falsemoe_intermediate_size: list[int] | None = Nonemoe_k: int | None = 6moe_num_experts: int | None = 64moe_num_shared_experts: int | None = 2moe_norm_min: float | None = 1e-12output_router_logits: bool | None = Falserouter_aux_loss_coef: float | None = 0.001mlp_layer_types: list[str] | None = None )

Parameters

  • vocab_size (int, optional, defaults to 103424) — Vocabulary size of the model. Defines the number of different tokens that can be represented by the input_ids.
  • pad_token_id (int, optional) — Token id used for padding in the vocabulary.
  • bos_token_id (int, optional) — Token id used for beginning-of-stream in the vocabulary.
  • eos_token_id (Union[int, list[int]], optional) — Token id used for end-of-stream in the vocabulary.
  • hidden_size (int, optional, defaults to 2560) — Dimension of the hidden representations.
  • intermediate_size (int, optional, defaults to 12288) — Dimension of the MLP representations.
  • num_hidden_layers (int, optional, defaults to 28) — Number of hidden layers in the Transformer decoder.
  • num_attention_heads (int, optional, defaults to 20) — Number of attention heads for each attention layer in the Transformer decoder.
  • num_key_value_heads (int, optional, defaults to 4) — This is the number of key_value heads that should be used to implement Grouped Query Attention. If num_key_value_heads=num_attention_heads, the model will use Multi Head Attention (MHA), if num_key_value_heads=1 the model will use Multi Query Attention (MQA) otherwise GQA is used. When converting a multi-head checkpoint to a GQA checkpoint, each group key and value head should be constructed by meanpooling all the original heads within that group. For more details, check out this paper. If it is not specified, will default to num_attention_heads.
  • hidden_act (str, optional, defaults to silu) — The non-linear activation function (function or string) in the decoder. For example, "gelu", "relu", "silu", etc.
  • max_position_embeddings (int, optional, defaults to 131072) — The maximum sequence length that this model might ever be used with.
  • initializer_range (float, optional, defaults to 0.02) — The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
  • rms_norm_eps (float, optional, defaults to 1e-05) — The epsilon used by the rms normalization layers.
  • use_cache (bool, optional, defaults to True) — Whether or not the model should return the last key/values attentions (not used by all models). Only relevant if config.is_decoder=True or when the model is a decoder-only generative model.
  • tie_word_embeddings (bool, optional, defaults to True) — Whether to tie weight embeddings according to model’s tied_weights_keys mapping.
  • rope_parameters (Union[~modeling_rope_utils.RopeParameters, dict], optional) — Dictionary containing the configuration parameters for the RoPE embeddings. The dictionary should contain a value for rope_theta and optionally parameters used for scaling in case you want to use RoPE with longer max_position_embeddings.
  • use_bias (bool, optional, defaults to False) — Whether to use a bias in any of the projections including mlp and attention for example
  • moe_intermediate_size (list[int], optional) — Intermediate size of the routed expert MLPs.
  • moe_k (int, optional, defaults to 6) — Number of selected experts.
  • moe_num_experts (int, optional, defaults to 64) — Number of routed experts.
  • moe_num_shared_experts (int, optional, defaults to 2) — The number of experts that are shared for all MoE forwards.
  • moe_norm_min (float, optional, defaults to 1e-12) — Minimum division value during routing normalization.
  • output_router_logits (bool, optional, defaults to False) — Whether or not the router logits should be returned by the model. Enabling this will also allow the model to output the auxiliary loss, including load balancing loss and router z-loss.
  • router_aux_loss_coef (float, optional, defaults to 0.001) — Auxiliary load balancing loss coefficient. Used to penalize uneven expert routing in MoE models.
  • mlp_layer_types (list, optional) — MLP (Moe vs Dense) pattern for each layer.

This is the configuration class to store the configuration of a Ernie4_5_VLMoeModel. It is used to instantiate a Ernie4 5 Vl Moe model according to the specified arguments, defining the model architecture. Instantiating a configuration with the defaults will yield a similar configuration to that of the baidu/ERNIE-4.5-VL-28B-A3B-PT

Configuration objects inherit from PreTrainedConfig and can be used to control the model outputs. Read the documentation from PreTrainedConfig for more information.

Ernie4_5_VLMoeVisionConfig

class transformers.Ernie4_5_VLMoeVisionConfig

< >

( transformers_version: str | None = Nonearchitectures: list[str] | None = Noneoutput_hidden_states: bool | None = Falsereturn_dict: bool | None = Truedtype: typing.Union[str, ForwardRef('torch.dtype'), NoneType] = Nonechunk_size_feed_forward: int = 0is_encoder_decoder: bool = Falseid2label: dict[int, str] | dict[str, str] | None = Nonelabel2id: dict[str, int] | dict[str, str] | None = Noneproblem_type: typing.Optional[typing.Literal['regression', 'single_label_classification', 'multi_label_classification']] = Nonedepth: int = 32hidden_size: int = 1280hidden_act: str = 'quick_gelu'num_heads: int = 16in_channels: int = 3patch_size: int | list[int] | tuple[int, int] = 14spatial_merge_size: int = 2initializer_range: float = 0.02intermediate_size: int = 5120temporal_merge_size: int = 2rms_norm_eps: float = 1e-06 )

Parameters

  • depth (int, optional, defaults to 32) — Number of Transformer layers in the vision encoder.
  • hidden_size (int, optional, defaults to 1280) — Dimension of the hidden representations.
  • hidden_act (str, optional, defaults to quick_gelu) — The non-linear activation function (function or string) in the decoder. For example, "gelu", "relu", "silu", etc.
  • num_heads (int, optional, defaults to 16) — Number of attention heads for each attention layer in the Transformer decoder.
  • in_channels (int, optional, defaults to 3) — The number of input channels.
  • patch_size (Union[int, list[int], tuple[int, int]], optional, defaults to 14) — The size (resolution) of each patch.
  • spatial_merge_size (int, optional, defaults to 2) — The size of the spatial merge window used to reduce the number of visual tokens by merging neighboring patches.
  • initializer_range (float, optional, defaults to 0.02) — The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
  • intermediate_size (int, optional, defaults to 5120) — Dimension of the MLP representations.
  • temporal_merge_size (int, optional, defaults to 2) — The size used for merge along the temporal dimension.
  • rms_norm_eps (float, optional, defaults to 1e-06) — The epsilon used by the rms normalization layers.

This is the configuration class to store the configuration of a Ernie4_5_VLMoeModel. It is used to instantiate a Ernie4 5 Vl Moe model according to the specified arguments, defining the model architecture. Instantiating a configuration with the defaults will yield a similar configuration to that of the baidu/ERNIE-4.5-VL-28B-A3B-PT

Configuration objects inherit from PreTrainedConfig and can be used to control the model outputs. Read the documentation from PreTrainedConfig for more information.

Ernie4_5_VLMoeImageProcessor

class transformers.Ernie4_5_VLMoeImageProcessor

< >

( **kwargs: Unpack )

Parameters

  • do_convert_rgb (bool, kwargs, optional, defaults to True) — Whether to convert the image to RGB.
  • do_resize (bool, kwargs, optional, defaults to True) — Whether to resize the image.
  • size (Annotated[int | list[int] | tuple[int, ...] | dict[str, int] | None, None], kwargs, defaults to {'shortest_edge' -- 3136, 'longest_edge': 4842768}): Describes the maximum input dimensions to the model.
  • default_to_square (bool, kwargs, optional, defaults to False) — Whether to default to a square image when resizing, if size is an int.
  • crop_size (Annotated[int | list[int] | tuple[int, ...] | dict[str, int] | None, None], kwargs) — Size of the output image after applying center_crop.
  • resample (Annotated[Union[int, PILImageResampling, NoneType], None], kwargs, defaults to Resampling.BICUBIC) — Resampling filter to use if resizing the image. This can be one of the enum PILImageResampling. Only has an effect if do_resize is set to True.
  • do_rescale (bool, kwargs, optional, defaults to True) — Whether to rescale the image.
  • rescale_factor (float, kwargs, optional, defaults to 0.00392156862745098) — Rescale factor to rescale the image by if do_rescale is set to True.
  • do_normalize (bool, kwargs, optional, defaults to True) — Whether to normalize the image.
  • image_mean (Union[float, list[float], tuple[float, ...]], kwargs, optional, defaults to [0.48145466, 0.4578275, 0.40821073]) — Image mean to use for normalization. Only has an effect if do_normalize is set to True.
  • image_std (Union[float, list[float], tuple[float, ...]], kwargs, optional, defaults to [0.26862954, 0.26130258, 0.27577711]) — Image standard deviation to use for normalization. Only has an effect if do_normalize is set to True.
  • do_pad (bool, kwargs, optional) — Whether to pad the image. Padding is done either to the largest size in the batch or to a fixed square size per image. The exact padding strategy depends on the model.
  • pad_size (Annotated[int | list[int] | tuple[int, ...] | dict[str, int] | None, None], kwargs) — The size in {"height": int, "width" int} to pad the images to. Must be larger than any image size provided for preprocessing. If pad_size is not provided, images will be padded to the largest height and width in the batch. Applied only when do_pad=True.
  • do_center_crop (bool, kwargs, optional) — Whether to center crop the image.
  • data_format (Union[str, ~image_utils.ChannelDimension], kwargs, optional) — Only ChannelDimension.FIRST is supported. Added for compatibility with slow processors.
  • input_data_format (Union[str, ~image_utils.ChannelDimension], kwargs, optional) — The channel dimension format for the input image. If unset, the channel dimension format is inferred from the input image. Can be one of:
    • "channels_first" or ChannelDimension.FIRST: image in (num_channels, height, width) format.
    • "channels_last" or ChannelDimension.LAST: image in (height, width, num_channels) format.
    • "none" or ChannelDimension.NONE: image in (height, width) format.
  • device (Annotated[Union[str, torch.device, NoneType], None], kwargs) — The device to process the videos on. If unset, the device is inferred from the input videos.
  • return_tensors (Annotated[str | ~utils.generic.TensorType | None, None], kwargs) — Returns stacked tensors if set to 'pt', otherwise returns a list of tensors.
  • disable_grouping (bool, kwargs, optional) — Whether to disable grouping of images by size to process them individually and not in batches. If None, will be set to True if the images are on CPU, and False otherwise. This choice is based on empirical observations, as detailed here: https://github.com/huggingface/transformers/pull/38157
  • image_seq_length (int, kwargs, optional) — The number of image tokens to be used for each image in the input. Added for backward compatibility but this should be set as a processor attribute in future models.
  • patch_size (int, kwargs, optional, defaults to 14) — The spatial patch size of the vision encoder.
  • temporal_patch_size (int, kwargs, optional) — The temporal patch size of the vision encoder. Unused in the image processor, only used for videos.
  • merge_size (int, kwargs, optional, defaults to 2) — The merge size of the vision encoder to llm encoder.

Constructs a Ernie4_5_VLMoeImageProcessor image processor.

preprocess

< >

( images: typing.Union[ForwardRef('PIL.Image.Image'), numpy.ndarray, ForwardRef('torch.Tensor'), list['PIL.Image.Image'], list[numpy.ndarray], list['torch.Tensor']]**kwargs: Unpack ) → ~image_processing_base.BatchFeature

Parameters

  • images (Union[PIL.Image.Image, numpy.ndarray, torch.Tensor, list[PIL.Image.Image], list[numpy.ndarray], list[torch.Tensor]]) — Image to preprocess. Expects a single or batch of images with pixel values ranging from 0 to 255. If passing in images with pixel values between 0 and 1, set do_rescale=False.
  • do_convert_rgb (bool, kwargs, optional) — Whether to convert the image to RGB.
  • do_resize (bool, kwargs, optional) — Whether to resize the image.
  • size (Annotated[int | list[int] | tuple[int, ...] | dict[str, int] | None, None], kwargs) — Describes the maximum input dimensions to the model.
  • default_to_square (bool, kwargs, optional) — Whether to default to a square image when resizing, if size is an int.
  • crop_size (Annotated[int | list[int] | tuple[int, ...] | dict[str, int] | None, None], kwargs) — Size of the output image after applying center_crop.
  • resample (Annotated[Union[int, PILImageResampling, NoneType], None], kwargs) — Resampling filter to use if resizing the image. This can be one of the enum PILImageResampling. Only has an effect if do_resize is set to True.
  • do_rescale (bool, kwargs, optional) — Whether to rescale the image.
  • rescale_factor (float, kwargs, optional) — Rescale factor to rescale the image by if do_rescale is set to True.
  • do_normalize (bool, kwargs, optional) — Whether to normalize the image.
  • image_mean (Union[float, list[float], tuple[float, ...]], kwargs, optional) — Image mean to use for normalization. Only has an effect if do_normalize is set to True.
  • image_std (Union[float, list[float], tuple[float, ...]], kwargs, optional) — Image standard deviation to use for normalization. Only has an effect if do_normalize is set to True.
  • do_pad (bool, kwargs, optional) — Whether to pad the image. Padding is done either to the largest size in the batch or to a fixed square size per image. The exact padding strategy depends on the model.
  • pad_size (Annotated[int | list[int] | tuple[int, ...] | dict[str, int] | None, None], kwargs) — The size in {"height": int, "width" int} to pad the images to. Must be larger than any image size provided for preprocessing. If pad_size is not provided, images will be padded to the largest height and width in the batch. Applied only when do_pad=True.
  • do_center_crop (bool, kwargs, optional) — Whether to center crop the image.
  • data_format (Union[str, ~image_utils.ChannelDimension], kwargs, optional) — Only ChannelDimension.FIRST is supported. Added for compatibility with slow processors.
  • input_data_format (Union[str, ~image_utils.ChannelDimension], kwargs, optional) — The channel dimension format for the input image. If unset, the channel dimension format is inferred from the input image. Can be one of:
    • "channels_first" or ChannelDimension.FIRST: image in (num_channels, height, width) format.
    • "channels_last" or ChannelDimension.LAST: image in (height, width, num_channels) format.
    • "none" or ChannelDimension.NONE: image in (height, width) format.
  • device (Annotated[Union[str, torch.device, NoneType], None], kwargs) — The device to process the videos on. If unset, the device is inferred from the input videos.
  • return_tensors (Annotated[str | ~utils.generic.TensorType | None, None], kwargs) — Returns stacked tensors if set to 'pt', otherwise returns a list of tensors.
  • disable_grouping (bool, kwargs, optional) — Whether to disable grouping of images by size to process them individually and not in batches. If None, will be set to True if the images are on CPU, and False otherwise. This choice is based on empirical observations, as detailed here: https://github.com/huggingface/transformers/pull/38157
  • image_seq_length (int, kwargs, optional) — The number of image tokens to be used for each image in the input. Added for backward compatibility but this should be set as a processor attribute in future models.
  • patch_size (int, kwargs, optional, defaults to 14) — The spatial patch size of the vision encoder.
  • temporal_patch_size (int, kwargs, optional) — The temporal patch size of the vision encoder. Unused in the image processor, only used for videos.
  • merge_size (int, kwargs, optional, defaults to 2) — The merge size of the vision encoder to llm encoder.

Returns

~image_processing_base.BatchFeature

  • data (dict) — Dictionary of lists/arrays/tensors returned by the call method (‘pixel_values’, etc.).
  • tensor_type (Union[None, str, TensorType], optional) — You can give a tensor_type here to convert the lists of integers in PyTorch/Numpy Tensors at initialization.

Ernie4_5_VLMoeImageProcessorPil

class transformers.Ernie4_5_VLMoeImageProcessorPil

< >

( **kwargs: Unpack )

Parameters

  • do_convert_rgb (bool, kwargs, optional, defaults to True) — Whether to convert the image to RGB.
  • do_resize (bool, kwargs, optional, defaults to True) — Whether to resize the image.
  • size (Annotated[int | list[int] | tuple[int, ...] | dict[str, int] | None, None], kwargs, defaults to {'shortest_edge' -- 3136, 'longest_edge': 4842768}): Describes the maximum input dimensions to the model.
  • default_to_square (bool, kwargs, optional, defaults to False) — Whether to default to a square image when resizing, if size is an int.
  • crop_size (Annotated[int | list[int] | tuple[int, ...] | dict[str, int] | None, None], kwargs) — Size of the output image after applying center_crop.
  • resample (Annotated[Union[int, PILImageResampling, NoneType], None], kwargs, defaults to Resampling.BICUBIC) — Resampling filter to use if resizing the image. This can be one of the enum PILImageResampling. Only has an effect if do_resize is set to True.
  • do_rescale (bool, kwargs, optional, defaults to True) — Whether to rescale the image.
  • rescale_factor (float, kwargs, optional, defaults to 0.00392156862745098) — Rescale factor to rescale the image by if do_rescale is set to True.
  • do_normalize (bool, kwargs, optional, defaults to True) — Whether to normalize the image.
  • image_mean (Union[float, list[float], tuple[float, ...]], kwargs, optional, defaults to [0.48145466, 0.4578275, 0.40821073]) — Image mean to use for normalization. Only has an effect if do_normalize is set to True.
  • image_std (Union[float, list[float], tuple[float, ...]], kwargs, optional, defaults to [0.26862954, 0.26130258, 0.27577711]) — Image standard deviation to use for normalization. Only has an effect if do_normalize is set to True.
  • do_pad (bool, kwargs, optional) — Whether to pad the image. Padding is done either to the largest size in the batch or to a fixed square size per image. The exact padding strategy depends on the model.
  • pad_size (Annotated[int | list[int] | tuple[int, ...] | dict[str, int] | None, None], kwargs) — The size in {"height": int, "width" int} to pad the images to. Must be larger than any image size provided for preprocessing. If pad_size is not provided, images will be padded to the largest height and width in the batch. Applied only when do_pad=True.
  • do_center_crop (bool, kwargs, optional) — Whether to center crop the image.
  • data_format (Union[str, ~image_utils.ChannelDimension], kwargs, optional) — Only ChannelDimension.FIRST is supported. Added for compatibility with slow processors.
  • input_data_format (Union[str, ~image_utils.ChannelDimension], kwargs, optional) — The channel dimension format for the input image. If unset, the channel dimension format is inferred from the input image. Can be one of:
    • "channels_first" or ChannelDimension.FIRST: image in (num_channels, height, width) format.
    • "channels_last" or ChannelDimension.LAST: image in (height, width, num_channels) format.
    • "none" or ChannelDimension.NONE: image in (height, width) format.
  • device (Annotated[Union[str, torch.device, NoneType], None], kwargs) — The device to process the videos on. If unset, the device is inferred from the input videos.
  • return_tensors (Annotated[str | ~utils.generic.TensorType | None, None], kwargs) — Returns stacked tensors if set to 'pt', otherwise returns a list of tensors.
  • disable_grouping (bool, kwargs, optional) — Whether to disable grouping of images by size to process them individually and not in batches. If None, will be set to True if the images are on CPU, and False otherwise. This choice is based on empirical observations, as detailed here: https://github.com/huggingface/transformers/pull/38157
  • image_seq_length (int, kwargs, optional) — The number of image tokens to be used for each image in the input. Added for backward compatibility but this should be set as a processor attribute in future models.
  • patch_size (int, kwargs, optional, defaults to 14) — The spatial patch size of the vision encoder.
  • temporal_patch_size (int, kwargs, optional) — The temporal patch size of the vision encoder. Unused in the image processor, only used for videos.
  • merge_size (int, kwargs, optional, defaults to 2) — The merge size of the vision encoder to llm encoder.

Constructs a Ernie4_5_VLMoeImageProcessor image processor.

preprocess

< >

( images: typing.Union[ForwardRef('PIL.Image.Image'), numpy.ndarray, ForwardRef('torch.Tensor'), list['PIL.Image.Image'], list[numpy.ndarray], list['torch.Tensor']]**kwargs: Unpack ) → ~image_processing_base.BatchFeature

Parameters

  • images (Union[PIL.Image.Image, numpy.ndarray, torch.Tensor, list[PIL.Image.Image], list[numpy.ndarray], list[torch.Tensor]]) — Image to preprocess. Expects a single or batch of images with pixel values ranging from 0 to 255. If passing in images with pixel values between 0 and 1, set do_rescale=False.
  • do_convert_rgb (bool, kwargs, optional) — Whether to convert the image to RGB.
  • do_resize (bool, kwargs, optional) — Whether to resize the image.
  • size (Annotated[int | list[int] | tuple[int, ...] | dict[str, int] | None, None], kwargs) — Describes the maximum input dimensions to the model.
  • default_to_square (bool, kwargs, optional) — Whether to default to a square image when resizing, if size is an int.
  • crop_size (Annotated[int | list[int] | tuple[int, ...] | dict[str, int] | None, None], kwargs) — Size of the output image after applying center_crop.
  • resample (Annotated[Union[int, PILImageResampling, NoneType], None], kwargs) — Resampling filter to use if resizing the image. This can be one of the enum PILImageResampling. Only has an effect if do_resize is set to True.
  • do_rescale (bool, kwargs, optional) — Whether to rescale the image.
  • rescale_factor (float, kwargs, optional) — Rescale factor to rescale the image by if do_rescale is set to True.
  • do_normalize (bool, kwargs, optional) — Whether to normalize the image.
  • image_mean (Union[float, list[float], tuple[float, ...]], kwargs, optional) — Image mean to use for normalization. Only has an effect if do_normalize is set to True.
  • image_std (Union[float, list[float], tuple[float, ...]], kwargs, optional) — Image standard deviation to use for normalization. Only has an effect if do_normalize is set to True.
  • do_pad (bool, kwargs, optional) — Whether to pad the image. Padding is done either to the largest size in the batch or to a fixed square size per image. The exact padding strategy depends on the model.
  • pad_size (Annotated[int | list[int] | tuple[int, ...] | dict[str, int] | None, None], kwargs) — The size in {"height": int, "width" int} to pad the images to. Must be larger than any image size provided for preprocessing. If pad_size is not provided, images will be padded to the largest height and width in the batch. Applied only when do_pad=True.
  • do_center_crop (bool, kwargs, optional) — Whether to center crop the image.
  • data_format (Union[str, ~image_utils.ChannelDimension], kwargs, optional) — Only ChannelDimension.FIRST is supported. Added for compatibility with slow processors.
  • input_data_format (Union[str, ~image_utils.ChannelDimension], kwargs, optional) — The channel dimension format for the input image. If unset, the channel dimension format is inferred from the input image. Can be one of:
    • "channels_first" or ChannelDimension.FIRST: image in (num_channels, height, width) format.
    • "channels_last" or ChannelDimension.LAST: image in (height, width, num_channels) format.
    • "none" or ChannelDimension.NONE: image in (height, width) format.
  • device (Annotated[Union[str, torch.device, NoneType], None], kwargs) — The device to process the videos on. If unset, the device is inferred from the input videos.
  • return_tensors (Annotated[str | ~utils.generic.TensorType | None, None], kwargs) — Returns stacked tensors if set to 'pt', otherwise returns a list of tensors.
  • disable_grouping (bool, kwargs, optional) — Whether to disable grouping of images by size to process them individually and not in batches. If None, will be set to True if the images are on CPU, and False otherwise. This choice is based on empirical observations, as detailed here: https://github.com/huggingface/transformers/pull/38157
  • image_seq_length (int, kwargs, optional) — The number of image tokens to be used for each image in the input. Added for backward compatibility but this should be set as a processor attribute in future models.
  • patch_size (int, kwargs, optional, defaults to 14) — The spatial patch size of the vision encoder.
  • temporal_patch_size (int, kwargs, optional) — The temporal patch size of the vision encoder. Unused in the image processor, only used for videos.
  • merge_size (int, kwargs, optional, defaults to 2) — The merge size of the vision encoder to llm encoder.

Returns

~image_processing_base.BatchFeature

  • data (dict) — Dictionary of lists/arrays/tensors returned by the call method (‘pixel_values’, etc.).
  • tensor_type (Union[None, str, TensorType], optional) — You can give a tensor_type here to convert the lists of integers in PyTorch/Numpy Tensors at initialization.

Ernie4_5_VLMoeVideoProcessor

class transformers.Ernie4_5_VLMoeVideoProcessor

< >

( **kwargs: Unpack )

Parameters

  • **kwargs (Ernie4_5_VLMoeVideoProcessorInitKwargs, optional) — Additional image preprocessing options. Model-specific kwargs are listed above; see the TypedDict class for the complete list of supported arguments.

Constructs a Ernie4_5_VLMoeVideoProcessor video processor.

preprocess

< >

( videos: typing.Union[list['PIL.Image.Image'], numpy.ndarray, ForwardRef('torch.Tensor'), list[numpy.ndarray], list['torch.Tensor'], list[list['PIL.Image.Image']], list[list[numpy.ndarray]], list[list['torch.Tensor']], transformers.video_utils.URL, list[transformers.video_utils.URL], list[list[transformers.video_utils.URL]], transformers.video_utils.Path, list[transformers.video_utils.Path], list[list[transformers.video_utils.Path]]]**kwargs: Unpack ) → ~image_processing_base.BatchFeature

Parameters

  • videos (Union[list[PIL.Image.Image], numpy.ndarray, torch.Tensor, list[numpy.ndarray], list[torch.Tensor], list[list[PIL.Image.Image]], list[list[numpy.ndarray]], list[list[torch.Tensor]], ~video_utils.URL, list[~video_utils.URL], list[list[~video_utils.URL]], ~video_utils.Path, list[~video_utils.Path], list[list[~video_utils.Path]]]) — Video to preprocess. Expects a single or batch of videos with pixel values ranging from 0 to 255. If passing in videos with pixel values between 0 and 1, set do_rescale=False.
  • do_convert_rgb (bool, kwargs, optional) — Whether to convert the image to RGB.
  • do_resize (bool, kwargs, optional) — Whether to resize the image.
  • size (Annotated[int | list[int] | tuple[int, ...] | dict[str, int] | None, None], kwargs) — Describes the maximum input dimensions to the model.
  • default_to_square (bool, kwargs, optional) — Whether to default to a square image when resizing, if size is an int.
  • resample (Annotated[Union[int, PILImageResampling, NoneType], None], kwargs) — Resampling filter to use if resizing the image. This can be one of the enum PILImageResampling. Only has an effect if do_resize is set to True.
  • do_rescale (bool, kwargs, optional) — Whether to rescale the image.
  • rescale_factor (float, kwargs, optional) — Rescale factor to rescale the image by if do_rescale is set to True.
  • do_normalize (bool, kwargs, optional) — Whether to normalize the image.
  • image_mean (Union[float, list[float], tuple[float, ...]], kwargs, optional) — Image mean to use for normalization. Only has an effect if do_normalize is set to True.
  • image_std (Union[float, list[float], tuple[float, ...]], kwargs, optional) — Image standard deviation to use for normalization. Only has an effect if do_normalize is set to True.
  • do_center_crop (bool, kwargs, optional) — Whether to center crop the image.
  • do_pad (bool, kwargs, optional) — Whether to pad the image. Padding is done either to the largest size in the batch or to a fixed square size per image. The exact padding strategy depends on the model.
  • crop_size (Annotated[int | list[int] | tuple[int, ...] | dict[str, int] | None, None], kwargs) — Size of the output image after applying center_crop.
  • data_format (Union[str, ~image_utils.ChannelDimension], kwargs, optional) — Only ChannelDimension.FIRST is supported. Added for compatibility with slow processors.
  • input_data_format (Union[str, ~image_utils.ChannelDimension], kwargs, optional) — The channel dimension format for the input image. If unset, the channel dimension format is inferred from the input image. Can be one of:
    • "channels_first" or ChannelDimension.FIRST: image in (num_channels, height, width) format.
    • "channels_last" or ChannelDimension.LAST: image in (height, width, num_channels) format.
    • "none" or ChannelDimension.NONE: image in (height, width) format.
  • device (Annotated[Union[str, torch.device, NoneType], None], kwargs) — The device to process the videos on. If unset, the device is inferred from the input videos.
  • do_sample_frames (bool, kwargs, optional) — Whether to sample frames from the video before processing or to process the whole video.
  • video_metadata (Annotated[~video_utils.VideoMetadata | dict | list[dict | ~video_utils.VideoMetadata] | list[list[dict | ~video_utils.VideoMetadata]] | None, None], kwargs) — Metadata of the video containing information about total duration, fps and total number of frames. It will be used to sample frames from video or compute timestamps. Don’t pass any metadata unless you are trying to decode the video manually before processing
  • fps (Annotated[int | float | None, None], kwargs) — Target frames to sample per second when do_sample_frames=True.
  • num_frames (Annotated[int | None, None], kwargs) — Maximum number of frames to sample when do_sample_frames=True.
  • return_metadata (bool, kwargs, optional) — Whether to return video metadata or not. Video metadats is an object containing info about video duration, fps, decoding backend, etc.
  • return_tensors (Annotated[str | ~utils.generic.TensorType | None, None], kwargs) — Returns stacked tensors if set to 'pt', otherwise returns a list of tensors.

Returns

~image_processing_base.BatchFeature

  • data (dict) — Dictionary of lists/arrays/tensors returned by the call method (‘pixel_values’, etc.).
  • tensor_type (Union[None, str, TensorType], optional) — You can give a tensor_type here to convert the lists of integers in PyTorch/Numpy Tensors at initialization.

Ernie4_5_VLMoeProcessor

class transformers.Ernie4_5_VLMoeProcessor

< >

( image_processor = Nonetokenizer = Nonevideo_processor = Nonechat_template = None**kwargs )

Parameters

  • image_processor (Ernie4_5_VLMoeImageProcessor, optional) — The image processor is a required input.
  • tokenizer (LlamaTokenizerFast, optional) — The tokenizer is a required input.
  • video_processor (Ernie4_5_VLMoeVideoProcessor, optional) — The video processor is a required input.
  • chat_template (str, optional) — A Jinja template which will be used to convert lists of messages in a chat into a tokenizable string.

Constructs a Ernie 4.5 VL processor which wraps a Ernie 4.5 VL image processor and a Llama tokenizer into a single processor. Ernie4_5_VLMoeProcessor offers all the functionalities of Ernie4_5_VLMoeImageProcessor and LlamaTokenizerFast. See the call() and decode() for more information.

__call__

< >

( images: typing.Union[ForwardRef('PIL.Image.Image'), numpy.ndarray, ForwardRef('torch.Tensor'), list['PIL.Image.Image'], list[numpy.ndarray], list['torch.Tensor'], NoneType] = Nonetext: str | list[str] | list[list[str]] = Nonevideos: typing.Union[list['PIL.Image.Image'], numpy.ndarray, ForwardRef('torch.Tensor'), list[numpy.ndarray], list['torch.Tensor'], list[list['PIL.Image.Image']], list[list[numpy.ndarray]], list[list['torch.Tensor']], transformers.video_utils.URL, list[transformers.video_utils.URL], list[list[transformers.video_utils.URL]], transformers.video_utils.Path, list[transformers.video_utils.Path], list[list[transformers.video_utils.Path]], NoneType] = None**kwargs: Unpack ) → BatchFeature

Parameters

  • images (PIL.Image.Image, np.ndarray, torch.Tensor, list[PIL.Image.Image], list[np.ndarray], list[torch.Tensor]) — The image or batch of images to be prepared. Each image can be a PIL image, NumPy array or PyTorch tensor. Both channels-first and channels-last formats are supported.
  • text (str, list[str], list[list[str]]) — The sequence or batch of sequences to be encoded. Each sequence can be a string or a list of strings (pretokenized string). If the sequences are provided as list of strings (pretokenized), you must set is_split_into_words=True (to lift the ambiguity with a batch of sequences).
  • videos (np.ndarray, torch.Tensor, list[np.ndarray], list[torch.Tensor]) — The image or batch of videos to be prepared. Each video can be a 4D NumPy array or PyTorch tensor, or a nested list of 3D frames. Both channels-first and channels-last formats are supported.
  • return_tensors (str or TensorType, optional) — If set, will return tensors of a particular framework. Acceptable values are:
    • 'pt': Return PyTorch torch.Tensor objects.
    • 'np': Return NumPy np.ndarray objects.

Returns

BatchFeature

A BatchFeature with the following fields:

  • input_ids — List of token ids to be fed to a model. Returned when text is not None.
  • attention_mask — List of indices specifying which tokens should be attended to by the model (when return_attention_mask=True or if “attention_mask” is in self.model_input_names and if text is not None).
  • pixel_values — Pixel values to be fed to a model. Returned when images is not None.
  • pixel_values_videos — Pixel values of videos to be fed to a model. Returned when videos is not None.
  • image_grid_thw — List of image 3D grid in LLM. Returned when images is not None.
  • video_grid_thw — List of video 3D grid in LLM. Returned when videos is not None.
  • mm_token_type_ids — List of token type ids differentiating between image, video and text input. Returned when text is not None.

Main method to prepare for the model one or several sequences(s) and image(s). This method forwards the text and kwargs arguments to Qwen2TokenizerFast’s call() if text is not None to encode the text. To prepare the vision inputs, this method forwards the vision_infos and kwargs arguments to Ernie4_5_VLMoeImageProcessor’s __call__() if vision_infos is not None.

Ernie4_5_VLMoeTextModel

class transformers.Ernie4_5_VLMoeTextModel

< >

( config: Ernie4_5_VLMoeTextConfig )

Parameters

  • config (Ernie4_5_VLMoeTextConfig) — Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the from_pretrained() method to load the model weights.

The bare Ernie4 5 Vl Moe Text Model outputting raw hidden-states without any specific head on top.

This model inherits from PreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.)

This model is also a PyTorch torch.nn.Module subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.

forward

< >

( input_ids: typing.Optional[torch.LongTensor] = Noneattention_mask: typing.Optional[torch.Tensor] = Noneposition_ids: typing.Optional[torch.LongTensor] = Nonemoe_mm_token_type_ids: typing.Optional[torch.IntTensor] = Nonepast_key_values: transformers.cache_utils.Cache | None = Noneinputs_embeds: typing.Optional[torch.FloatTensor] = Noneuse_cache: bool | None = None**kwargs: Unpack ) → MoeModelOutputWithPast or tuple(torch.FloatTensor)

Parameters

  • input_ids (torch.LongTensor of shape (batch_size, sequence_length), optional) — Indices of input sequence tokens in the vocabulary. Padding will be ignored by default.

    Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.

    What are input IDs?

  • attention_mask (torch.Tensor of shape (batch_size, sequence_length), optional) — Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]:

    • 1 for tokens that are not masked,
    • 0 for tokens that are masked.

    What are attention masks?

  • position_ids (torch.LongTensor of shape (batch_size, sequence_length), optional) — Indices of positions of each input sequence tokens in the position embeddings. Selected in the range [0, config.n_positions - 1].

    What are position IDs?

  • moe_mm_token_type_ids (torch.IntTensor of shape (batch_size, sequence_length), optional) — The same as mm_token_type_ids while additionally considering start/end image/video tokens as respective vision tokens.
  • past_key_values (~cache_utils.Cache, optional) — Pre-computed hidden-states (key and values in the self-attention blocks and in the cross-attention blocks) that can be used to speed up sequential decoding. This typically consists in the past_key_values returned by the model at a previous stage of decoding, when use_cache=True or config.use_cache=True.

    Only Cache instance is allowed as input, see our kv cache guide. If no past_key_values are passed, DynamicCache will be initialized by default.

    The model will output the same cache format that is fed as input.

    If past_key_values are used, the user is expected to input only unprocessed input_ids (those that don’t have their past key value states given to this model) of shape (batch_size, unprocessed_length) instead of all input_ids of shape (batch_size, sequence_length).

  • inputs_embeds (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size), optional) — Optionally, instead of passing input_ids you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert input_ids indices into associated vectors than the model’s internal embedding lookup matrix.
  • use_cache (bool, optional) — If set to True, past_key_values key value states are returned and can be used to speed up decoding (see past_key_values).

Returns

MoeModelOutputWithPast or tuple(torch.FloatTensor)

A MoeModelOutputWithPast or a tuple of torch.FloatTensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (Ernie4_5_VLMoeConfig) and inputs.

The Ernie4_5_VLMoeTextModel forward method, overrides the __call__ special method.

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

  • last_hidden_state (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size)) — Sequence of hidden-states at the output of the last layer of the model.

  • past_key_values (Cache, optional, returned when use_cache=True is passed or when config.use_cache=True) — It is a Cache instance. For more details, see our kv cache guide.

    Contains pre-computed hidden-states (key and values in the self-attention blocks and optionally if config.is_encoder_decoder=True in the cross-attention blocks) that can be used (see past_key_values input) to speed up sequential decoding.

  • hidden_states (tuple(torch.FloatTensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of torch.FloatTensor (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).

    Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.

  • attentions (tuple(torch.FloatTensor), optional, returned when output_attentions=True is passed or when config.output_attentions=True) — Tuple of torch.FloatTensor (one for each layer) of shape (batch_size, num_heads, sequence_length, sequence_length).

    Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.

  • router_logits (tuple(torch.FloatTensor), optional, returned when output_router_probs=True and config.add_router_probs=True is passed or when config.output_router_probs=True) — Tuple of torch.FloatTensor (one for each layer) of shape (batch_size, sequence_length, num_experts).

    Raw router logits (post-softmax) that are computed by MoE routers, these terms are used to compute the auxiliary loss for Mixture of Experts models.

Ernie4_5_VLMoeVisionTransformerPretrainedModel

class transformers.Ernie4_5_VLMoeVisionTransformerPretrainedModel

< >

( config )

Parameters

The bare Ernie4 5 Vl Moe Model outputting raw hidden-states without any specific head on top.

This model inherits from PreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.)

This model is also a PyTorch torch.nn.Module subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.

forward

< >

( hidden_states: Tensorgrid_thw: Tensor**kwargs: Unpack )

grid_thw (torch.LongTensor of shape (num_images, 3)): The temporal, height and width dimensions of feature shape for each image. Each row contains [t, h, w] values.

Ernie4_5_VLMoeVariableResolutionResamplerModel

class transformers.Ernie4_5_VLMoeVariableResolutionResamplerModel

< >

( config: Ernie4_5_VLMoeConfig )

forward

< >

( hidden_statesgrid_thw )

Ernie4_5_VLMoeModel

class transformers.Ernie4_5_VLMoeModel

< >

( config: Ernie4_5_VLMoeConfig )

Parameters

  • config (Ernie4_5_VLMoeConfig) — Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the from_pretrained() method to load the model weights.

The bare Ernie4 5 Vl Moe Model outputting raw hidden-states without any specific head on top.

This model inherits from PreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.)

This model is also a PyTorch torch.nn.Module subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.

forward

< >

( input_ids: LongTensor = Noneattention_mask: typing.Optional[torch.Tensor] = Noneposition_ids: typing.Optional[torch.LongTensor] = Nonemm_token_type_ids: typing.Optional[torch.IntTensor] = Nonemoe_mm_token_type_ids: typing.Optional[torch.IntTensor] = Nonepast_key_values: transformers.cache_utils.Cache | None = Noneinputs_embeds: typing.Optional[torch.FloatTensor] = Noneuse_cache: bool | None = Nonepixel_values: typing.Optional[torch.Tensor] = Nonepixel_values_videos: typing.Optional[torch.FloatTensor] = Noneimage_grid_thw: typing.Optional[torch.LongTensor] = Nonevideo_grid_thw: typing.Optional[torch.LongTensor] = None**kwargs: Unpack ) → MoeModelOutputWithPast or tuple(torch.FloatTensor)

Parameters

  • input_ids (torch.LongTensor of shape (batch_size, sequence_length), optional) — Indices of input sequence tokens in the vocabulary. Padding will be ignored by default.

    Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.

    What are input IDs?

  • attention_mask (torch.Tensor of shape (batch_size, sequence_length), optional) — Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]:

    • 1 for tokens that are not masked,
    • 0 for tokens that are masked.

    What are attention masks?

  • position_ids (torch.LongTensor of shape (batch_size, sequence_length), optional) — Indices of positions of each input sequence tokens in the position embeddings. Selected in the range [0, config.n_positions - 1].

    What are position IDs?

  • mm_token_type_ids (torch.IntTensor of shape (batch_size, sequence_length), optional) — Token type ids matching each modality to a different value in the input sequence, i.e. text (0), image (1), video (2).
  • moe_mm_token_type_ids (torch.IntTensor of shape (batch_size, sequence_length), optional) — The same as mm_token_type_ids while additionally considering start/end image/video tokens as respective vision tokens.
  • past_key_values (~cache_utils.Cache, optional) — Pre-computed hidden-states (key and values in the self-attention blocks and in the cross-attention blocks) that can be used to speed up sequential decoding. This typically consists in the past_key_values returned by the model at a previous stage of decoding, when use_cache=True or config.use_cache=True.

    Only Cache instance is allowed as input, see our kv cache guide. If no past_key_values are passed, DynamicCache will be initialized by default.

    The model will output the same cache format that is fed as input.

    If past_key_values are used, the user is expected to input only unprocessed input_ids (those that don’t have their past key value states given to this model) of shape (batch_size, unprocessed_length) instead of all input_ids of shape (batch_size, sequence_length).

  • inputs_embeds (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size), optional) — Optionally, instead of passing input_ids you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert input_ids indices into associated vectors than the model’s internal embedding lookup matrix.
  • use_cache (bool, optional) — If set to True, past_key_values key value states are returned and can be used to speed up decoding (see past_key_values).
  • pixel_values (torch.Tensor of shape (batch_size, num_channels, image_size, image_size), optional) — The tensors corresponding to the input images. Pixel values can be obtained using image_processor_class. See image_processor_class.__call__ for details (processor_class uses image_processor_class for processing images).
  • pixel_values_videos (torch.FloatTensor of shape (batch_size, num_frames, num_channels, frame_size, frame_size), optional) — The tensors corresponding to the input video. Pixel values for videos can be obtained using video_processor_class. See video_processor_class.__call__ for details (processor_class uses video_processor_class for processing videos).
  • image_grid_thw (torch.LongTensor of shape (num_images, 3), optional) — The temporal, height and width of feature shape of each image in LLM.
  • video_grid_thw (torch.LongTensor of shape (num_videos, 3), optional) — The temporal, height and width of feature shape of each video in LLM.

Returns

MoeModelOutputWithPast or tuple(torch.FloatTensor)

A MoeModelOutputWithPast or a tuple of torch.FloatTensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (None) and inputs.

The Ernie4_5_VLMoeModel forward method, overrides the __call__ special method.

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

  • last_hidden_state (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size)) — Sequence of hidden-states at the output of the last layer of the model.

  • past_key_values (Cache, optional, returned when use_cache=True is passed or when config.use_cache=True) — It is a Cache instance. For more details, see our kv cache guide.

    Contains pre-computed hidden-states (key and values in the self-attention blocks and optionally if config.is_encoder_decoder=True in the cross-attention blocks) that can be used (see past_key_values input) to speed up sequential decoding.

  • hidden_states (tuple(torch.FloatTensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of torch.FloatTensor (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).

    Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.

  • attentions (tuple(torch.FloatTensor), optional, returned when output_attentions=True is passed or when config.output_attentions=True) — Tuple of torch.FloatTensor (one for each layer) of shape (batch_size, num_heads, sequence_length, sequence_length).

    Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.

  • router_logits (tuple(torch.FloatTensor), optional, returned when output_router_probs=True and config.add_router_probs=True is passed or when config.output_router_probs=True) — Tuple of torch.FloatTensor (one for each layer) of shape (batch_size, sequence_length, num_experts).

    Raw router logits (post-softmax) that are computed by MoE routers, these terms are used to compute the auxiliary loss for Mixture of Experts models.

get_video_features

< >

( pixel_values_videos: FloatTensorvideo_grid_thw: typing.Optional[torch.LongTensor] = None**kwargs: Unpack ) → BaseModelOutputWithPooling or tuple(torch.FloatTensor)

Parameters

  • pixel_values_videos (torch.FloatTensor of shape (batch_size, num_frames, num_channels, frame_size, frame_size)) — The tensors corresponding to the input video. Pixel values for videos can be obtained using Ernie4_5_VLMoeVideoProcessor. See Ernie4_5_VLMoeVideoProcessor.__call__() for details (Ernie4_5_VLMoeProcessor uses Ernie4_5_VLMoeVideoProcessor for processing videos).
  • video_grid_thw (torch.LongTensor of shape (num_videos, 3), optional) — The temporal, height and width of feature shape of each video in LLM.

Returns

BaseModelOutputWithPooling or tuple(torch.FloatTensor)

A BaseModelOutputWithPooling or a tuple of torch.FloatTensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (Ernie4_5_VLMoeConfig) and inputs.

  • last_hidden_state (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size)) — Sequence of hidden-states at the output of the last layer of the model.

  • pooler_output (torch.FloatTensor of shape (batch_size, hidden_size)) — Last layer hidden-state of the first token of the sequence (classification token) after further processing through the layers used for the auxiliary pretraining task. E.g. for BERT-family of models, this returns the classification token after processing through a linear layer and a tanh activation function. The linear layer weights are trained from the next sentence prediction (classification) objective during pretraining.

  • hidden_states (tuple(torch.FloatTensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of torch.FloatTensor (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).

    Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.

  • attentions (tuple(torch.FloatTensor), optional, returned when output_attentions=True is passed or when config.output_attentions=True) — Tuple of torch.FloatTensor (one for each layer) of shape (batch_size, num_heads, sequence_length, sequence_length).

    Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.

get_image_features

< >

( pixel_values: FloatTensorimage_grid_thw: typing.Optional[torch.LongTensor] = None**kwargs: Unpack ) → BaseModelOutputWithPooling or tuple(torch.FloatTensor)

Parameters

  • pixel_values (torch.FloatTensor of shape (batch_size, num_channels, image_size, image_size)) — The tensors corresponding to the input images. Pixel values can be obtained using Ernie4_5_VLMoeImageProcessor. See Ernie4_5_VLMoeImageProcessor.__call__() for details (Ernie4_5_VLMoeProcessor uses Ernie4_5_VLMoeImageProcessor for processing images).
  • image_grid_thw (torch.LongTensor of shape (num_images, 3), optional) — The temporal, height and width of feature shape of each image in LLM.

Returns

BaseModelOutputWithPooling or tuple(torch.FloatTensor)

A BaseModelOutputWithPooling or a tuple of torch.FloatTensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (Ernie4_5_VLMoeConfig) and inputs.

  • last_hidden_state (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size)) — Sequence of hidden-states at the output of the last layer of the model.

  • pooler_output (torch.FloatTensor of shape (batch_size, hidden_size)) — Last layer hidden-state of the first token of the sequence (classification token) after further processing through the layers used for the auxiliary pretraining task. E.g. for BERT-family of models, this returns the classification token after processing through a linear layer and a tanh activation function. The linear layer weights are trained from the next sentence prediction (classification) objective during pretraining.

  • hidden_states (tuple(torch.FloatTensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of torch.FloatTensor (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).

    Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.

  • attentions (tuple(torch.FloatTensor), optional, returned when output_attentions=True is passed or when config.output_attentions=True) — Tuple of torch.FloatTensor (one for each layer) of shape (batch_size, num_heads, sequence_length, sequence_length).

    Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.

Ernie4_5_VLMoeForConditionalGeneration

class transformers.Ernie4_5_VLMoeForConditionalGeneration

< >

( config )

forward

< >

( input_ids: LongTensor = Noneattention_mask: typing.Optional[torch.Tensor] = Noneposition_ids: typing.Optional[torch.LongTensor] = Nonemm_token_type_ids: typing.Optional[torch.IntTensor] = Nonemoe_mm_token_type_ids: typing.Optional[torch.IntTensor] = Nonepast_key_values: transformers.cache_utils.Cache | None = Noneinputs_embeds: typing.Optional[torch.FloatTensor] = Nonelabels: typing.Optional[torch.LongTensor] = Noneuse_cache: bool | None = Noneoutput_router_logits: bool | None = Nonepixel_values: typing.Optional[torch.Tensor] = Nonepixel_values_videos: typing.Optional[torch.FloatTensor] = Noneimage_grid_thw: typing.Optional[torch.LongTensor] = Nonevideo_grid_thw: typing.Optional[torch.LongTensor] = Nonelogits_to_keep: typing.Union[int, torch.Tensor] = 0**kwargs: Unpack ) → MoeCausalLMOutputWithPast or tuple(torch.FloatTensor)

Parameters

  • input_ids (torch.LongTensor of shape (batch_size, sequence_length), optional) — Indices of input sequence tokens in the vocabulary. Padding will be ignored by default.

    Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.

    What are input IDs?

  • attention_mask (torch.Tensor of shape (batch_size, sequence_length), optional) — Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]:

    • 1 for tokens that are not masked,
    • 0 for tokens that are masked.

    What are attention masks?

  • position_ids (torch.LongTensor of shape (batch_size, sequence_length), optional) — Indices of positions of each input sequence tokens in the position embeddings. Selected in the range [0, config.n_positions - 1].

    What are position IDs?

  • mm_token_type_ids (torch.IntTensor of shape (batch_size, sequence_length), optional) — Token type ids matching each modality to a different value in the input sequence, i.e. text (0), image (1), video (2).
  • moe_mm_token_type_ids (torch.IntTensor of shape (batch_size, sequence_length), optional) — The same as mm_token_type_ids while additionally considering start/end image/video tokens as respective vision tokens.
  • past_key_values (~cache_utils.Cache, optional) — Pre-computed hidden-states (key and values in the self-attention blocks and in the cross-attention blocks) that can be used to speed up sequential decoding. This typically consists in the past_key_values returned by the model at a previous stage of decoding, when use_cache=True or config.use_cache=True.

    Only Cache instance is allowed as input, see our kv cache guide. If no past_key_values are passed, DynamicCache will be initialized by default.

    The model will output the same cache format that is fed as input.

    If past_key_values are used, the user is expected to input only unprocessed input_ids (those that don’t have their past key value states given to this model) of shape (batch_size, unprocessed_length) instead of all input_ids of shape (batch_size, sequence_length).

  • inputs_embeds (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size), optional) — Optionally, instead of passing input_ids you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert input_ids indices into associated vectors than the model’s internal embedding lookup matrix.
  • labels (torch.LongTensor of shape (batch_size, sequence_length), optional) — Labels for computing the masked language modeling loss. Indices should either be in [0, ..., config.vocab_size] or -100 (see input_ids docstring). Tokens with indices set to -100 are ignored (masked), the loss is only computed for the tokens with labels in [0, ..., config.vocab_size].
  • use_cache (bool, optional) — If set to True, past_key_values key value states are returned and can be used to speed up decoding (see past_key_values).
  • output_router_logits (bool, optional) — Whether or not to return the logits of all the routers. They are useful for computing the router loss, and should not be returned during inference.
  • pixel_values (torch.Tensor of shape (batch_size, num_channels, image_size, image_size), optional) — The tensors corresponding to the input images. Pixel values can be obtained using image_processor_class. See image_processor_class.__call__ for details (processor_class uses image_processor_class for processing images).
  • pixel_values_videos (torch.FloatTensor of shape (batch_size, num_frames, num_channels, frame_size, frame_size), optional) — The tensors corresponding to the input video. Pixel values for videos can be obtained using video_processor_class. See video_processor_class.__call__ for details (processor_class uses video_processor_class for processing videos).
  • image_grid_thw (torch.LongTensor of shape (num_images, 3), optional) — The temporal, height and width of feature shape of each image in LLM.
  • video_grid_thw (torch.LongTensor of shape (num_videos, 3), optional) — The temporal, height and width of feature shape of each video in LLM.
  • logits_to_keep (Union[int, torch.Tensor], optional, defaults to 0) — If an int, compute logits for the last logits_to_keep tokens. If 0, calculate logits for all input_ids (special case). Only last token logits are needed for generation, and calculating them only for that token can save memory, which becomes pretty significant for long sequences or large vocabulary size. If a torch.Tensor, must be 1D corresponding to the indices to keep in the sequence length dimension. This is useful when using packed tensor format (single dimension for batch and sequence length).

Returns

MoeCausalLMOutputWithPast or tuple(torch.FloatTensor)

A MoeCausalLMOutputWithPast or a tuple of torch.FloatTensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (None) and inputs.

The Ernie4_5_VLMoeForConditionalGeneration forward method, overrides the __call__ special method.

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

  • loss (torch.FloatTensor of shape (1,), optional, returned when labels is provided) — Language modeling loss (for next-token prediction).

  • logits (torch.FloatTensor of shape (batch_size, sequence_length, config.vocab_size)) — Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax).

  • aux_loss (torch.FloatTensor, optional, returned when labels is provided) — aux_loss for the sparse modules.

  • router_logits (tuple(torch.FloatTensor), optional, returned when output_router_probs=True and config.add_router_probs=True is passed or when config.output_router_probs=True) — Tuple of torch.FloatTensor (one for each layer) of shape (batch_size, sequence_length, num_experts).

    Raw router logits (post-softmax) that are computed by MoE routers, these terms are used to compute the auxiliary loss for Mixture of Experts models.

  • past_key_values (Cache, optional, returned when use_cache=True is passed or when config.use_cache=True) — It is a Cache instance. For more details, see our kv cache guide.

    Contains pre-computed hidden-states (key and values in the self-attention blocks) that can be used (see past_key_values input) to speed up sequential decoding.

  • hidden_states (tuple(torch.FloatTensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of torch.FloatTensor (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).

    Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.

  • attentions (tuple(torch.FloatTensor), optional, returned when output_attentions=True is passed or when config.output_attentions=True) — Tuple of torch.FloatTensor (one for each layer) of shape (batch_size, num_heads, sequence_length, sequence_length).

    Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.

get_video_features

< >

( pixel_values_videos: FloatTensorvideo_grid_thw: typing.Optional[torch.LongTensor] = None**kwargs: Unpack ) → BaseModelOutputWithPooling or tuple(torch.FloatTensor)

Parameters

  • pixel_values_videos (torch.FloatTensor of shape (batch_size, num_channels, image_size, image_size)) — The tensors corresponding to the input videos.
  • video_grid_thw (torch.LongTensor of shape (num_videos, 3), optional) — The temporal, height and width of feature shape of each video in LLM.

Returns

BaseModelOutputWithPooling or tuple(torch.FloatTensor)

A BaseModelOutputWithPooling or a tuple of torch.FloatTensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (Ernie4_5_VLMoeConfig) and inputs.

  • last_hidden_state (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size)) — Sequence of hidden-states at the output of the last layer of the model.

  • pooler_output (torch.FloatTensor of shape (batch_size, hidden_size)) — Last layer hidden-state of the first token of the sequence (classification token) after further processing through the layers used for the auxiliary pretraining task. E.g. for BERT-family of models, this returns the classification token after processing through a linear layer and a tanh activation function. The linear layer weights are trained from the next sentence prediction (classification) objective during pretraining.

  • hidden_states (tuple(torch.FloatTensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of torch.FloatTensor (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).

    Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.

  • attentions (tuple(torch.FloatTensor), optional, returned when output_attentions=True is passed or when config.output_attentions=True) — Tuple of torch.FloatTensor (one for each layer) of shape (batch_size, num_heads, sequence_length, sequence_length).

    Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.

Example:

>>> from PIL import Image
>>> from transformers import AutoProcessor, Ernie4_5_VLMoeForConditionalGeneration

>>> model = Ernie4_5_VLMoeForConditionalGeneration.from_pretrained("baidu/ERNIE-4.5-VL-28B-A3B-PT")
>>> processor = AutoProcessor.from_pretrained("baidu/ERNIE-4.5-VL-28B-A3B-PT")

>>> messages = [
...     {
...         "role": "user", "content": [
...             {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/pipeline-cat-chonk.jpeg"},
...             {"type": "text", "text": "Where is the cat standing?"},
...         ]
...     },
... ]

>>> inputs = processor.apply_chat_template(
...     messages,
...     tokenize=True,
...     return_dict=True,
...     return_tensors="pt",
...     add_generation_prompt=True
... )
>>> # Generate
>>> generate_ids = model.generate(**inputs)
>>> processor.batch_decode(generate_ids, skip_special_tokens=True)[0]

get_image_features

< >

( pixel_values: FloatTensorimage_grid_thw: typing.Optional[torch.LongTensor] = None**kwargs: Unpack ) → BaseModelOutputWithPooling or tuple(torch.FloatTensor)

Parameters

  • pixel_values (torch.FloatTensor of shape (batch_size, num_channels, image_size, image_size)) — The tensors corresponding to the input images.
  • image_grid_thw (torch.LongTensor of shape (num_images, 3), optional) — The temporal, height and width of feature shape of each image in LLM.

Returns

BaseModelOutputWithPooling or tuple(torch.FloatTensor)

A BaseModelOutputWithPooling or a tuple of torch.FloatTensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (Ernie4_5_VLMoeConfig) and inputs.

  • last_hidden_state (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size)) — Sequence of hidden-states at the output of the last layer of the model.

  • pooler_output (torch.FloatTensor of shape (batch_size, hidden_size)) — Last layer hidden-state of the first token of the sequence (classification token) after further processing through the layers used for the auxiliary pretraining task. E.g. for BERT-family of models, this returns the classification token after processing through a linear layer and a tanh activation function. The linear layer weights are trained from the next sentence prediction (classification) objective during pretraining.

  • hidden_states (tuple(torch.FloatTensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of torch.FloatTensor (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).

    Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.

  • attentions (tuple(torch.FloatTensor), optional, returned when output_attentions=True is passed or when config.output_attentions=True) — Tuple of torch.FloatTensor (one for each layer) of shape (batch_size, num_heads, sequence_length, sequence_length).

    Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.

Example:

>>> from PIL import Image
>>> from transformers import AutoProcessor, Ernie4_5_VLMoeForConditionalGeneration

>>> model = Ernie4_5_VLMoeForConditionalGeneration.from_pretrained("baidu/ERNIE-4.5-VL-28B-A3B-PT")
>>> processor = AutoProcessor.from_pretrained("baidu/ERNIE-4.5-VL-28B-A3B-PT")

>>> messages = [
...     {
...         "role": "user", "content": [
...             {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/pipeline-cat-chonk.jpeg"},
...             {"type": "text", "text": "Where is the cat standing?"},
...         ]
...     },
... ]

>>> inputs = processor.apply_chat_template(
...     messages,
...     tokenize=True,
...     return_dict=True,
...     return_tensors="pt",
...     add_generation_prompt=True
... )
>>> # Generate
>>> generate_ids = model.generate(**inputs)
>>> processor.batch_decode(generate_ids, skip_special_tokens=True)[0]
Update on GitHub

←Emu3 Evolla→