Hugging Face's logo

Transformers documentation

PerceptionLM

You are viewing main version, which requires installation from source. If you'd like regular pip install, checkout the latest stable version (v4.53.2).
Hugging Face's logo
Join the Hugging Face community

and get access to the augmented documentation experience

to get started

PerceptionLM

Overview

The PerceptionLM model was proposed in PerceptionLM: Open-Access Data and Models for Detailed Visual Understanding by Jang Hyun Cho et al. It’s a fully open, reproducible model for transparent research in image and video understanding. PLM consists of a vision encoder with a small scale (<8B parameters) LLM decoder.

The abstract from the paper is the following:

Vision-language models are integral to computer vision research, yet many high-performing models remain closed-source, obscuring their data, design and training recipe. The research community has responded by using distillation from black-box models to label training data, achieving strong benchmark results, at the cost of measurable scientific progress. However, without knowing the details of the teacher model and its data sources, scientific progress remains difficult to measure. In this paper, we study building a Perception Language Model (PLM) in a fully open and reproducible framework for transparent research in image and video understanding. We analyze standard training pipelines without distillation from proprietary models and explore large-scale synthetic data to identify critical data gaps, particularly in detailed video understanding. To bridge these gaps, we release 2.8M human-labeled instances of fine-grained video question-answer pairs and spatio-temporally grounded video captions. Additionally, we introduce PLM–VideoBench, a suite for evaluating challenging video understanding tasks focusing on the ability to reason about “what”, “where”, “when”, and “how” of a video. We make our work fully reproducible by providing data, training recipes, code & models.

This model was contributed by shumingh. The original code can be found here.

PerceptionLMConfig

class transformers.PerceptionLMConfig

< >

( vision_config = None text_config = None vision_use_cls_token = True projector_pooling_ratio = 1 image_token_id = 128002 video_token_id = 128003 **kwargs )

Parameters

  • vision_config (Union[TimmWrapperConfig, dict], optional, defaults to TimmWrapperConfig()) — The config object or dictionary of the vision backbone.
  • text_config (Union[PretrainedConfig, dict], optional, defaults to LlamaConfig()) — The config object or dictionary of the text backbone.
  • vision_use_cls_token (bool, optional, defaults to True) — Whether CLS token is used in the vision backbone. If used, we remove CLS token embedding from vision output.
  • projector_pooling_ratio (int, optional, defaults to 1) — The pooling ratio used in the multimodal projector.
  • image_token_id (int, optional, defaults to 128002) — The image token index to encode the image prompt.
  • video_token_id (int, optional, defaults to 128003) — The video token index to encode the video prompt.

This is the configuration class to store the configuration of a PerceptionLMForConditionalGeneration. It is used to instantiate an PerceptionLM model according to the specified arguments, defining the model architecture.

Example models:

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

PerceptionLMProcessor

class transformers.PerceptionLMProcessor

< >

( video_processor = None image_processor = None tokenizer = None patch_size = None chat_template = None pooling_ratio = 2 **kwargs )

Parameters

  • video_processor (PerceptionLMVideoProcessor, optional) — The video processor to process video inputs.
  • image_processor (PerceptionLMImageProcessorFast, optional) — The image processor to process image inputs.
  • tokenizer (LlamaTokenizerFast or similar, optional) — The tokenizer to process text inputs.
  • patch_size (int, optional) — Patch size from the vision tower.
  • chat_template (str, optional) — A Jinja template which will be used to convert lists of messages in a chat into a tokenizable string.
  • pooling_ratio (int, optional, defaults to 2) — Pooling ratio for vision tokens. If not 1, 2D adaptive pooling is applied over projected vision tokens.

Constructs a PerceptionLM processor which wraps a PerceptionLM image processor, a PerceptionLM video processor, and a tokenizer into a single processor.

PerceptionLMProcessor offers all the functionalities of PerceptionLMImageProcessorFast, PerceptionLMVideoProcessor, and the tokenizer (e.g. LlamaTokenizerFast). See the __call__() and decode() for more information.

batch_decode

< >

( *args **kwargs )

This method forwards all its arguments to PerceptionLMTokenizerFast’s batch_decode(). Please refer to the docstring of this method for more information.

decode

< >

( *args **kwargs )

This method forwards all its arguments to PerceptionLMTokenizerFast’s decode(). Please refer to the docstring of this method for more information.

PerceptionLMImageProcessorFast

class transformers.PerceptionLMImageProcessorFast

< >

( **kwargs: typing_extensions.Unpack[transformers.models.perception_lm.image_processing_perception_lm_fast.PerceptionLMFastImageProcessorKwargs] )

Constructs a fast PerceptionLM image processor.

PerceptionLMVideoProcessor

class transformers.PerceptionLMVideoProcessor

< >

( **kwargs: typing_extensions.Unpack[transformers.models.perception_lm.video_processing_perception_lm.PerceptionLMFastVideoProcessorInitKwargs] )

PerceptionLMModel

class transformers.PerceptionLMModel

< >

( config: PerceptionLMConfig )

Parameters

  • config (PerceptionLMConfig) — 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 Perception Lm 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] = None pixel_values: typing.Optional[torch.FloatTensor] = None pixel_values_videos: typing.Optional[torch.FloatTensor] = None attention_mask: typing.Optional[torch.Tensor] = None position_ids: typing.Optional[torch.LongTensor] = None past_key_values: typing.Optional[list[torch.FloatTensor]] = None inputs_embeds: typing.Optional[torch.FloatTensor] = None use_cache: typing.Optional[bool] = None output_attentions: typing.Optional[bool] = None output_hidden_states: typing.Optional[bool] = None cache_position: typing.Optional[torch.LongTensor] = None logits_to_keep: typing.Union[int, torch.Tensor] = 0 **lm_kwargs ) → transformers.models.perception_lm.modeling_perception_lm.PerceptionLMModelOutputWithPast 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?

  • pixel_values (torch.FloatTensor 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).
  • 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?

  • past_key_values (list[torch.FloatTensor], 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).
  • output_attentions (bool, optional) — Whether or not to return the attentions tensors of all attention layers. See attentions under returned tensors for more detail.
  • output_hidden_states (bool, optional) — Whether or not to return the hidden states of all layers. See hidden_states under returned tensors for more detail.
  • cache_position (torch.LongTensor of shape (sequence_length), optional) — Indices depicting the position of the input sequence tokens in the sequence. Contrarily to position_ids, this tensor is not affected by padding. It is used to update the cache in the correct position and to infer the complete sequence length.
  • logits_to_keep (Union[int, torch.Tensor], 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

transformers.models.perception_lm.modeling_perception_lm.PerceptionLMModelOutputWithPast or tuple(torch.FloatTensor)

A transformers.models.perception_lm.modeling_perception_lm.PerceptionLMModelOutputWithPast 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 (PerceptionLMConfig) and inputs.

  • last_hidden_state (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size), optional) — 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) — Tuple of tuple(torch.FloatTensor) of length config.n_layers, with each tuple having 2 tensors of shape (batch_size, num_heads, sequence_length, embed_size_per_head))

    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.

  • image_hidden_states (torch.FloatTensor, optional) — A torch.FloatTensor of size (batch_size, num_images, sequence_length, hidden_size). Image hidden_states of the model produced by the vision encoder and after projecting the last hidden state.

  • video_hidden_states (torch.FloatTensor, optional) — A torch.FloatTensor of size (batch_size, num_videos, sequence_length, hidden_size). Video hidden_states of the model produced by the vision encoder and after projecting the last hidden state.

The PerceptionLMModel 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.

get_image_features

< >

( pixel_values: FloatTensor **kwargs ) → image_features (torch.Tensor)

Parameters

  • pixel_values (torch.FloatTensor] of shape (batch_size, num_tiles, channels, height, width)) — The tensors corresponding to the input images.

Returns

image_features (torch.Tensor)

Image feature tensor of shape (num_tiles, num_patches, embed_dim)).

Obtains image last hidden states from the vision tower and apply multimodal projection.

PerceptionLMForConditionalGeneration

class transformers.PerceptionLMForConditionalGeneration

< >

( config: PerceptionLMConfig )

Parameters

  • config (PerceptionLMConfig) — 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 Perception Lm Model for token generation conditioned on other modalities (e.g. image-text-to-text generation).

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] = None pixel_values: typing.Optional[torch.FloatTensor] = None pixel_values_videos: typing.Optional[torch.FloatTensor] = None attention_mask: typing.Optional[torch.Tensor] = None position_ids: typing.Optional[torch.LongTensor] = None past_key_values: typing.Optional[list[torch.FloatTensor]] = None inputs_embeds: typing.Optional[torch.FloatTensor] = None labels: typing.Optional[torch.LongTensor] = None use_cache: typing.Optional[bool] = None output_attentions: typing.Optional[bool] = None output_hidden_states: typing.Optional[bool] = None cache_position: typing.Optional[torch.LongTensor] = None logits_to_keep: typing.Union[int, torch.Tensor] = 0 **lm_kwargs ) → transformers.models.perception_lm.modeling_perception_lm.PerceptionLMCausalLMOutputWithPast 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?

  • pixel_values (torch.FloatTensor 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).
  • 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?

  • past_key_values (list[torch.FloatTensor], 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_attentions (bool, optional) — Whether or not to return the attentions tensors of all attention layers. See attentions under returned tensors for more detail.
  • output_hidden_states (bool, optional) — Whether or not to return the hidden states of all layers. See hidden_states under returned tensors for more detail.
  • cache_position (torch.LongTensor of shape (sequence_length), optional) — Indices depicting the position of the input sequence tokens in the sequence. Contrarily to position_ids, this tensor is not affected by padding. It is used to update the cache in the correct position and to infer the complete sequence length.
  • logits_to_keep (Union[int, torch.Tensor], 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

transformers.models.perception_lm.modeling_perception_lm.PerceptionLMCausalLMOutputWithPast or tuple(torch.FloatTensor)

A transformers.models.perception_lm.modeling_perception_lm.PerceptionLMCausalLMOutputWithPast 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 (PerceptionLMConfig) and inputs.

  • 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).

  • past_key_values (Cache, optional, returned when use_cache=True is passed or when config.use_cache=True) — Tuple of tuple(torch.FloatTensor) of length config.n_layers, with each tuple having 2 tensors of shape (batch_size, num_heads, sequence_length, embed_size_per_head))

    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.

  • image_hidden_states (torch.FloatTensor, optional) — A torch.FloatTensor of size (batch_size, num_images, sequence_length, hidden_size). Image hidden_states of the model produced by the vision encoder and after projecting the last hidden state.

  • video_hidden_states (torch.FloatTensor, optional) — A torch.FloatTensor of size (batch_size, num_videos, sequence_length, hidden_size). Video hidden_states of the model produced by the vision encoder and after projecting the last hidden state.

The PerceptionLMForConditionalGeneration 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.

Example:

>>> from PIL import Image
>>> import requests
>>> from transformers import AutoProcessor, PerceptionLMForConditionalGeneration

>>> model = PerceptionLMForConditionalGeneration.from_pretrained("perception_lm-hf/perception_lm-1.5-7b-hf")
>>> processor = AutoProcessor.from_pretrained("perception_lm-hf/perception_lm-1.5-7b-hf")

>>> prompt = "USER: <image>\nWhat's the content of the image? ASSISTANT:"
>>> url = "https://www.ilankelman.org/stopsigns/australia.jpg"
>>> image = Image.open(requests.get(url, stream=True).raw)

>>> inputs = processor(images=image, text=prompt, return_tensors="pt")

>>> # Generate
>>> generate_ids = model.generate(**inputs, max_new_tokens=15)
>>> processor.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
"USER:  \nWhat's the content of the image? ASSISTANT: The image features a busy city street with a stop sign prominently displayed"
< > Update on GitHub

←Perceiver Phi4 Multimodal→