tjAddlmZmZddlmZddlmZmZddlm Z m Z m Z m Z ddl mZmZddlmZmZerddlZGd d e d ZeGd de ZdgZdS))OptionalUnion) BatchFeature) ImageInputis_valid_image)MultiModalDataProcessingKwargsProcessorMixinUnpack)PreTokenizedInput TextInput)auto_docstringis_torch_availableNc(eZdZddidddddidZd S) ColQwen2ProcessorKwargspaddinglongestchannels_firstT) data_formatdo_convert_rgbreturn_tensorspt) text_kwargs images_kwargs common_kwargsN)__name__ __module__ __qualname__ _defaultsn/var/www/html/CA-Chatbot/venv/lib/python3.11/site-packages/transformers/models/colqwen2/processing_colqwen2.pyrr"sB y ,"  +D1  IIIr"rF)totalc eZdZ ddedzdedzffd Ze ddedzdeeze eze ezde e de fd Z dd Zed Zedefd Z ddedzde e de fd Zdee ezde e de fdZ ddede dfdede dfdededdedefddf dZxZS)ColQwen2ProcessorNvisual_prompt_prefix query_prefixc t|||t|dsdn|j|_t|dsdn|j|_|pd|_|pd|_dS) ar visual_prompt_prefix (`str`, *optional*, defaults to `"<|im_start|>user\n<|vision_start|><|image_pad|><|vision_end|>Describe the image.<|im_end|><|endoftext|>"`): A string that gets tokenized and prepended to the image tokens. query_prefix (`str`, *optional*, defaults to `"Query: "`): A prefix to be used for the query. ) chat_template image_tokenz <|image_pad|> video_tokenz <|video_pad|>zf<|im_start|>user <|vision_start|><|image_pad|><|vision_end|>Describe the image.<|im_end|><|endoftext|>zQuery: N)super__init__hasattrr+r,r'r()selfimage_processor tokenizerr*r'r(kwargs __class__s r#r.zColQwen2Processor.__init__1s )=QQQ29)]2S2Sn??YbYn29)]2S2Sn??YbYn$8% u !)5Ir"imagestextr3returnc &|jtfd|jji|}|ddd}|du}||t d||t d|t |r|g}nt|trt |drnZt|tr6t|dtrt |ddst d|j gt|z}|j dd |i|d }|d } | |j j d z} d} tt|D]} |j|| vrW|| |jd | | | zzd|| <| dz } |j|| vW|| d |j|| <|j|fddi|d} t#i| |}|d dddf|d ddd fz}tt%j|d|}t$jjj|d|d<|r=|d|ddkd}|d|i|S|t|t6r|g}n?t|trt|dt6st d| |jdz}g}|D]$}|j|z|z}||%|j|fddi|d}|SdS)a Returns: [`BatchFeature`]: A [`BatchFeature`] with the following fields: - **input_ids** -- List of token ids to be fed to a model. - **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`. tokenizer_init_kwargsrsuffixNz&Either text or images must be providedz5Only one of text or images can be processed at a timerzAimages must be an image, list of images or list of list of imagesr5rimage_grid_thwz<|placeholder|>return_token_type_idsF)data pixel_valuesT) batch_first input_idstoken_type_idsilabelsz*Text must be a string or a list of strings r!) _merge_kwargsrr2 init_kwargspop ValueErrorr isinstancelistr'lenr1 merge_sizeranger+replaceprodrtorchsplittolistnnutilsrnn pad_sequence masked_fillupdatestrquery_augmentation_tokenr(append)r0r5r6r3 output_kwargsr:r> texts_doc image_inputsr; merge_lengthindexi text_inputs return_dataoffsetsr@rD texts_queryqueryaugmented_query batch_querys r#__call__zColQwen2Processor.__call__Isg"+* #  "&."<   }-11(DAA &d 2 A s9~~..]]A*il::'0|';'; ,.?>RWCXC]C]C_C_coCo.prs(( !   *il:: $-Q<#7#78I4K[#\#\IaLL($.&+ .K ',K{,Kl,KLLLK""23AAAqD9KHX[_`>`bfgg""Hf#5666   $$$ Ov t,, ODGS1I1I O !MNNN~6;%'K 4 4"&"3e";f"D""?3333($.&+ .K  + r"c @i}|tjdi|ddp jjfd|D}fd|D}|||dt di|S)a Computes the number of placeholder tokens needed for multimodal inputs with the given sizes. Args: image_sizes (`list[list[int]]`, *optional*): The input sizes formatted as (height, width) per each image. Returns: `MultiModalData`: A `MultiModalData` object holding number of tokens per each of the provided input modalities, along with other useful data. NrrMc8g|]}jjg|RSr!)r1get_number_of_image_patches).0 image_sizerr0s r# z@ColQwen2Processor._get_num_multimodal_tokens..sE!!!A$@\*\m\\\!!!r"c g|] }|dzz S)r<r!)rn num_patchesrMs r#rpz@ColQwen2Processor._get_num_multimodal_tokens..s"ddd; A !=dddr")num_image_tokensnum_image_patchesr!)rr getrYr1rMr )r0 image_sizesr3 vision_datartrsrrMs` @@r#_get_num_multimodal_tokensz,ColQwen2Processor._get_num_multimodal_tokenss  "3=AA/SUVVM   ( ( (&**<>>a$BVBaJ!!!!!"-!!!  edddRcddd    4D[lmm n n n,, ,,,r"cT|jj}|jj}d|D}||zS)Ncg|]}|dv| S))pixel_values_videosvideo_grid_thwr!)rnnames r#rpz7ColQwen2Processor.model_input_names..s*' ' ' DHqt}55F5f555r"c |jdd|i|S)ai Prepare for the model one or several texts. This method is a wrapper around the `__call__` method of the ColQwen2Processor's [`ColQwen2Processor.__call__`]. This method forwards the `text` and `kwargs` arguments to the tokenizer. Args: 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). return_tensors (`str` or [`~utils.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. - **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`). r6r!r)r0r6r3s r#process_queriesz!ColQwen2Processor.process_queriess!<t}11$1&111r"cpuquery_embeddingsz torch.Tensorpassage_embeddings batch_size output_dtypez torch.dtype output_devicez torch.devicec 8t|dkrtdt|dkrtd|dj|djkrtd|dj|djkrtd| |dj}g}t dt||D]:}g}t jjj ||||zdd} t dt||D]} t jjj || | |zdd} | t j d | |  d d d | t j|d || usually obtained by padding the list of tensors. Args: query_embeddings (`Union[torch.Tensor, list[torch.Tensor]`): Query embeddings. passage_embeddings (`Union[torch.Tensor, list[torch.Tensor]`): Passage embeddings. batch_size (`int`, *optional*, defaults to 128): Batch size for computing scores. output_dtype (`torch.dtype`, *optional*, defaults to `torch.float32`): The dtype of the output tensor. If `None`, the dtype of the input embeddings is used. output_device (`torch.device` or `str`, *optional*, defaults to "cpu"): The device of the output tensor. Returns: `torch.Tensor`: A tensor of shape `(n_queries, n_passages)` containing the scores. The score tensor is saved on the "cpu" device. rzNo queries providedzNo passages providedz/Queries and passages must be on the same devicez-Queries and passages must have the same dtypeNT)rA padding_valuez bnd,csd->bcnsr)dimr<r=)rLrIdevicedtyperNrQrTrUrVrWr\einsummaxsumcatto) r0rrrrrscoresrb batch_scores batch_queriesjbatch_passagess r#score_retrievalz!ColQwen2Processor.score_retrieval"s*@  A % %233 3 ! " "a ' '344 4 A  %);A)>)E E ENOO O A  $(:1(=(C C CLMM M  +A.4L%'q#.//<< ] ]A/1L!HN.;; Q^!34$VW<M1c"455zBB  !&!3!@!@&q1z>'9:\]"A""##L-PPTTYZT[[\]^bbghbii MM%)La888;;LIILL][[ \ \ \ \yQ''''r")NNNNN)NN)N)rNr)rrrrZr.rrrr rKr rrrjrxpropertyr~r[rrrintrr __classcell__)r4s@r#r&r&/s[+/#' 66 "Dj 6 Dj 6666660%)Z^ffT!f++d9o=EV@WWf01 f  fff^fP----4 C CX C(#(((X(%)66T!6016  6666B2$y/)2012  2222H0449 >(>(^0D DE>(".$~2F"FG>( >( }- >( ^S01 >( >(>(>(>(>(>(>(>(r"r&)typingrrfeature_extraction_utilsr image_utilsrrprocessing_utilsr r r r tokenization_utils_baser rrUrrrQrr&__all__r!r"r#rs8*#"""""""44444455555555XXXXXXXXXXXXCCCCCCCC77777777LLL     .e    p(p(p(p(p(p(p(p(f  r"