
    i                         d Z ddlmZ ddlmZ ddlmZmZmZm	Z	m
Z
 ddlmZmZ ddlmZ  G d de	d	
      Z G d ded	
      Ze G d de             ZdgZy)z 
Processor class for Chameleon.
   )BatchFeature)
ImageInput)MultiModalDataProcessingKwargsProcessorMixin
TextKwargsUnpack)PreTokenizedInput	TextInput)auto_docstringc                       e Zd ZU dZeed<   y)ChameleonTextKwargsa  
    return_for_text_completion (`bool`, *optional*, defaults to `False`):
        Whether the processed text is intended for text completion tasks. When `True`, the processor does not
        append the separator token (`sep_token`) to the end of the prompt, which is typically used for chat
        mode. When `False`, the separator token is appended for proper chat formatting.
    return_for_text_completionN)__name__
__module____qualname____doc__bool__annotations__     /var/www/vps2.regionflexible.com/Desarrollo/venv/lib/python3.12/site-packages/transformers/models/chameleon/processing_chameleon.pyr   r      s     !%$r   r   F)totalc                   .    e Zd ZU eed<   ddddddidZy)ChameleonProcessorKwargstext_kwargsF)paddingr   return_mm_token_type_idsreturn_tensorspt)r   common_kwargsN)r   r   r   r   r   	_defaultsr   r   r   r   r   *   s+    $$ */(-
 d
	Ir   r   c            
            e Zd Zddedef fdZe	 	 ddedz  dee	z  e
e   z  e
e	   z  dz  dee   defd	       Zdd
Z xZS )ChameleonProcessorimage_seq_lengthimage_tokenc                 >   t         |   ||       || _        t        |d      r|j                  n|| _        |j                  | j                        | _        t        |d      r|j                  nd| _        t        |d      r|j                  nd| _
        |j                  | j                        | _        |j                  | j                        | _        |j                  | j                        | _        | j                  | j                  | j                  g| _        y)a  
        image_seq_length (`int`, *optional*, defaults to 1024):
            Sequence length of one image embedding.
        image_token (`str`, *optional*, defaults to `"<image>"`):
            The special token used to indicate image in the text.
        r&   	boi_tokenz<racm3:break>	eoi_tokenz<eoss>N)super__init__r%   hasattrr&   convert_tokens_to_idsimage_token_idr(   image_start_tokenr)   image_end_tokenimage_start_token_idimage_end_token_id	image_ids)selfimage_processor	tokenizerr%   r&   	__class__s        r   r+   zChameleonProcessor.__init__:   s     	)4 04;I}4U900[f'==d>N>NO#*9k#BI 	 7>i6Uy22[c'==d>N>NO$-$C$CDDZDZ$[!"+"A"A$BVBV"W--t/H/H$JaJabr   Nimagestextkwargsreturnc                 j   t        |t              r|g}n.t        |t              st        |d   t              st        d      ||t	        d       | j
                  t        fd| j                  j                  i|}|d   j                  dd      }g }| j                  | j                  | j                  z  z   | j                  z   }|D ]J  }|j                  | j                  |      }|s|| j                  j                  z  }|j!                  |       L i }	| | j"                  |fi |d	   }	|d   j                  d
d      }
|d   j                  dd      } | j                  |fi |d   d
di}| j%                  ||dg       |r| j'                  |d         |d<   t)        i ||	|
      S )aA  
        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`.
            zAInvalid input text. Please provide a string, or a list of stringsNz&You must provide either text or imagestokenizer_init_kwargsr   r   Fimages_kwargsr   r   image)
modalities	input_idsmm_token_type_ids)datatensor_type)
isinstancestrlist	TypeError
ValueError_merge_kwargsr   r6   init_kwargspopr/   r&   r%   r0   replace	sep_tokenappendr5   _check_special_mm_tokenscreate_mm_token_type_idsr   )r4   r8   r9   r:   output_kwargsr   prompt_stringsone_img_tokenssampleimage_inputsr   r   text_inputss                r   __call__zChameleonProcessor.__call__O   s   $ dC 6DD$'
47C0H_``<FNEFF***$
"&.."<"<
 

 &3=%A%E%EFbdi%j" //43C3CdF[F[3[\_c_s_ss 	*F^^D$4$4nEF-$..222!!&)		* /4//Y-:XYL&}599:JDQ#0#?#C#CD^`e#f $dnn^i}]7Sidhi%%nkwi%X#/3/L/L[YdMe/fK+,!@K!@<!@n]]r   c                     i }|?| j                   dz   gt        |      z  }dgt        |      z  }|j                  ||d       t        di |S )a  
        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.

        Returns:
            `MultiModalData`: A `MultiModalData` object holding number of tokens per each of the provided
            input modalities, along with other useful data.
              )num_image_tokensnum_image_patchesr   )r%   lenupdater   )r4   image_sizesr:   vision_datar]   r^   s         r   _get_num_multimodal_tokensz-ChameleonProcessor._get_num_multimodal_tokens   sb     " $ 5 5 9:S=MM!"c+&6 64D[lmn,,,r   )i   z<image>)NN)N)r   r   r   intrG   r+   r   r   r   r
   rH   r	   r   r   rY   rc   __classcell__)r7   s   @r   r$   r$   8   s    cS c^a c*  %)ae3^T!3^ ++d9o=EV@WWZ^^3^ 12	3^
 
3^ 3^j-r   r$   N)r   feature_extraction_utilsr   image_utilsr   processing_utilsr   r   r   r   r	   tokenization_utils_baser
   r   utilsr   r   r   r$   __all__r   r   r   <module>rl      sk    5 %  D #%*E %/u  a- a- a-H  
 r   