o ۷i;@sddlmZmZddlZddlmZddlmZddl m Z m Z m Z m Z mZmZddlmZmZddlmZdd lmZeeZGd d d ed d ZGddde ZGddde d d ZGddde ZdgZdS))OptionalUnionN) BatchFeature) ImageInput) ImagesKwargsMultiModalDataProcessingKwargsProcessorMixinUnpack VideosKwargs)PreTokenizedInput TextInput)logging) VideoInputc@s"eZdZUeeeefed<dS)Glm4vVideosProcessorKwargsfpsN)__name__ __module__ __qualname__rlistfloat__annotations__rr`/home/ubuntu/vllm_env/lib/python3.10/site-packages/transformers/models/glm4v/processing_glm4v.pyr$s rF)totalc@s2eZdZUeeed<eeed<eeed<dS)Glm4vImagesKwargs patch_sizetemporal_patch_size merge_sizeN)rrrrintrrrrrr(s  rc@s4eZdZUeed<ddddddidZeed<dS) Glm4vProcessorKwargs images_kwargsF)paddingreturn_token_type_idsreturn_mm_token_type_idsreturn_metadataT) text_kwargs videos_kwargsr(N)rrrrr _defaultsrrrrrr!.s  r!c seZdZdZgdZdZdZdZdfdd Z   dd e e d e e e ee ee fd e ed eed ef ddZdddZ dddZZS)Glm4vProcessora Constructs a GLM-4V processor which wraps a GLM-4V image processor and a GLM-4 tokenizer into a single processor. [`~Glm4vProcessor.__call__`] and [`~Glm4vProcessor.decode`] for more information. Args: image_processor ([`Glm4vProcessor`], *optional*): The image processor is a required input. tokenizer ([`PreTrainedTokenizerFast`], *optional*): The tokenizer is a required input. video_processor ([`Glm4vVideoProcessor`], *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. )image_processor tokenizervideo_processorAutoImageProcessorAutoVideoProcessor)PreTrainedTokenizerPreTrainedTokenizerFastNc stj||||dt|dsdn|j|_t|dsdn|j|_t|ddr)|jn||j|_t|ddr<|j|_dS||j|_dS)N) chat_template image_tokenz <|image|> video_tokenz <|video|>image_token_idvideo_token_id) super__init__hasattrr3r4getattrr5convert_tokens_to_idsr6)selfr+r,r-r2kwargs __class__rrr8Ps    zGlm4vProcessor.__init__imagestextvideosr=returncKs|jtfd|jji|}|dur"|jdd|i|d}|d}ni}d}|durI|jdd|i|d}d|vr@|d } n|d } |d } ni}d} t|tsU|g}| }|dur|jj d } d } t t |D]5} |j || vr|| | }|| |j d |d|| <| d7} |j || vst|| d |j || <qk| dur|jj d } d }t t |D]} |j|| vrx| |d }d}| |}|jdurtd|jdurdn|j|_|jddd }g}t d t |D] }|||q|d|}t ||kr||r|dnd t ||kst |D]}||}d|j dt|}||7}q|| |j|d|| <| || | |d }t |D]}|j || vrk|| |j d |d|| <qS|d7}|j|| vs|| d |j || <q|ddd}|ddd}|j|fi|d}|j||ddgd|rt|d}t|d}d|||jk<||d<ti||||dS) a^ Main method to prepare for the model one or several sequences(s) and image(s). This method forwards the `text` and `kwargs` arguments to PreTrainedTokenizerFast's [`~PreTrainedTokenizerFast.__call__`] if `text` is not `None` to encode the text. Args: 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 [`~utils.TensorType`], *optional*): If set, will return tensors of a particular framework. Acceptable values are: - `'tf'`: Return TensorFlow `tf.constant` objects. - `'pt'`: Return PyTorch `torch.Tensor` objects. - `'np'`: Return NumPy `np.ndarray` objects. - `'jax'`: Return JAX `jnp.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`. tokenizer_init_kwargsNr@r"image_grid_thwrBr(r&video_metadatavideo_grid_thwrz<|placeholder|>aSmolVLM requires frame timestamps to construct prompts, but the `fps` of the input video could not be inferred. Probably `video_metadata` was missing from inputs and you passed pre-sampled frames. Defaulting to `fps=24`. Please provide `video_metadata` for more accurate results.z<|begin_of_image|>z<|end_of_image|>r'return_tensorsr%Fimagevideo) modalities input_idsmm_token_type_ids)data tensor_typer) _merge_kwargsr!r, init_kwargsr+r-pop isinstancercopyrrangelenr3prodreplacer4rlogger warning_once timestampsappendr _check_special_mm_tokensnparray zeros_liker5tolistr)r<r@rArBr= output_kwargs image_inputsrE videos_inputsrFrG merge_lengthindexinum_image_tokens video_index num_framesvideo_structuremetadatar`unique_timestampsidxselected_timestamps frame_idx timestamp_secframe_structurerMr% text_inputs array_idsrRrrr__call___s*             % zGlm4vProcessor.__call__c  si}|dur7tjdi|ddpjjfdd|D}fdd|D}|||d|dur^tjdi|fd d|D}fd d|D}||d <td i|S) aK Computes the number of placeholder tokens needed for multimodal inputs with the given sizes. Args: image_sizes (`list[list[int]]`, *optional*): The input sizes formatted as (height, width) per each image. video_sizes (`list[list[int]]`, *optional*): The input sizes formatted as (num_frames, height, width) per each video. Returns: `MultiModalData`: A `MultiModalData` object holding number of tokens per each of the provided input modalities, along with other useful data. Nr"rcs"g|] }jjg|RqSr)r+get_number_of_image_patches).0 image_size)r"r<rr z=Glm4vProcessor._get_num_multimodal_tokens..cg|]}|dqSrHrr| num_patchesrrrr~)rmnum_image_patchesr(cs"g|] }jjg|RqSr)r-get_number_of_video_patches)r| video_size)r<r(rrr~rcrrrrrrrr~rnum_video_tokensr)r!r)getupdater+rr) r< image_sizes video_sizesr= vision_datarrmnum_video_patchesrr)r"rr<r(r_get_num_multimodal_tokenss&     z)Glm4vProcessor._get_num_multimodal_tokensTFcKs|jj|f||d|S)a Post-process the output of the model to decode the text. Args: generated_outputs (`torch.Tensor` or `np.ndarray`): The output of the model `generate` function. The output is expected to be a tensor of shape `(batch_size, sequence_length)` or `(sequence_length,)`. skip_special_tokens (`bool`, *optional*, defaults to `True`): Whether or not to remove special tokens in the output. Argument passed to the tokenizer's `batch_decode` method. clean_up_tokenization_spaces (`bool`, *optional*, defaults to `False`): Whether or not to clean up the tokenization spaces. Argument passed to the tokenizer's `batch_decode` method. **kwargs: Additional arguments to be passed to the tokenizer's `batch_decode method`. Returns: `list[str]`: The decoded text. )skip_special_tokensclean_up_tokenization_spaces)r, batch_decode)r<generated_outputsrrr=rrrpost_process_image_text_to_text sz.Glm4vProcessor.post_process_image_text_to_text)NNNN)NNN)NN)TF)rrr__doc__ attributesimage_processor_classvideo_processor_classtokenizer_classr8rrrrr rrr r!rrzrr __classcell__rrr>rr*;s2  'r*)typingrrnumpyrcfeature_extraction_utilsr image_utilsrprocessing_utilsrrr r r r tokenization_utils_baser rutilsr video_utilsr get_loggerrr^rrr!r*__all__rrrrs        m