+ j/f^RIHtHt^RIHt^RIHtHt^RIH t H t H t H t ^RI HtHtHt^RIHtHt]!4'd^RIt!RR ] R R 7tR t]!R 4Uu.uF pRVR R2NK up]!^4Uu.uF pRVR R2NK up,t]!RR] 44tR.tR#uupiuupi))OptionalUnion) BatchFeature) ImageInputmake_flat_list_of_images)MultiModalDataProcessingKwargsProcessorMixinUnpack) AddedTokenPreTokenizedInput TextInput)auto_docstringis_torch_availableNc>]tRt^"tRRRRRRR/RRR R R /R R R//tRtR#)ColPaliProcessorKwargs text_kwargspaddinglongestreturn_mm_token_type_idsFreturn_text_replacement_offsets images_kwargs data_formatchannels_firstdo_convert_rgbT common_kwargsreturn_tensorsptN)__name__ __module__ __qualname____firstlineno__ _defaults__static_attributes__ro/home/luca/screen-checking/.venv/lib/python3.14/site-packages/transformers/models/colpali/processing_colpali.pyrr"sB y & -u  + d  *D1 Ir&rF)totalziz4>z3cTaa]tRt^5to]tRV3RlV3Rlllt]RV3RlV3Rlll4tRV3Rllt RV3RlV3Rlllt V3RlR lt RR lt ] V3R l4t] V3R lR l4t]!RR7RV3RlRll4t]!RR7V3RlRl4tRV3RlRlltRtVtV;t#)ColPaliProcessorc&<V^8dQhRS[RS[/#)visual_prompt_prefix query_prefixstr)format __classdict__s"r' __annotate__ColPaliProcessor.__annotate__9s%#R#R " #R  #Rr&c <W@nWPn\VR4'g \R4hVPVn\VR4'gO\ \ RRR7pRV./pVPV4VP\ 4Vn \ Vn M"VPVn VPVn VP\4RVn RVn\SV`AWVR7R #) z visual_prompt_prefix (`str`, *optional*, defaults to `"Describe the image."`): A string that gets tokenized and prepended to the image tokens. query_prefix (`str`, *optional*, defaults to `"Question: "`): A prefix to be used for the query. image_seq_lengthz;Image processor is missing an `image_seq_length` attribute. image_tokenFT) normalizedspecialadditional_special_tokens) chat_templateN)r.r/hasattr ValueErrorr7r IMAGE_TOKENadd_special_tokensconvert_tokens_to_idsimage_token_idr8 add_tokens EXTRA_TOKENS add_bos_token add_eos_tokensuper__init__) selfimage_processor tokenizerr<r.r/r8 tokens_to_add __class__s &&&&&& r'rHColPaliProcessor.__init__9s%9!((:;;Z[ [ / @ @y-00$[UDQK8;-HM  ( ( 7"+"A"A+"ND *D "+":":D (44D \*"' "'  =Qr&c<V^8dQhRS[R,RS[S[,S[S[,,S[S[,,RS[S[,RS[/#)r-imagesNtextkwargsreturn)rrr listr rr)r2r3s"r'r4r5_sX%%T!%++d9o=EV@WW%/0 %  %r&c p<VeVe \R4hRVR&VP!VP3RVPP/VBpVR,P RVP ^ ,4pVeyVUu.uF+q`PP VP V V R2NK- ppVR,PR R4e'VR,R ;;,VP, uu&\SV`0!RR VR V/VBpVe'VR ,PVR ,^8HR4VR&V#uupi)a Returns: [`BatchFeature`]: A [`BatchFeature`] with the following fields: - **input_ids** -- List of token ids to be fed to a model. Returned when `text` is not `None`. If `suffix` is provided, the `input_ids` will also contain the suffix input ids. - **attention_mask** -- List of indices specifying which tokens should be attended to by the model (when `return_attention_mask=True` or if *"attention_mask"* is in `self.model_input_names` and if `text` is not `None`). - **pixel_values** -- Pixel values to be fed to a model. Returned when `images` is not `None`. - **labels** -- Labels compatible with training if `suffix` is not None Nz5Only one of text or images can be processed at a timeTreturn_token_type_idstokenizer_init_kwargsrsuffix max_lengthrPrQ input_idstoken_type_idslabelsri)r> _merge_kwargsvalid_processor_kwargsrK init_kwargspopquery_augmentation_token bos_tokenr/getr7rG__call__ masked_fill) rIrPrQrR output_kwargsrXsample model_inputsrMs &&&, r'reColPaliProcessor.__call__^sL&   2TU U*.&'**  ' ' "&.."<"<  }-11(DS>SS:w'RvRDRMR  %1+%>%J%Jt?X?X>YY[\+ T4'' s:A>c<V^8dQhRS[R,RS[S[,S[S[,,S[S[,,R,RS[S[,/#)r-rPNrQrR)rrr rTr r )r2r3s"r'r4r5sYGGT!G++d9o=EV@WWZ^^G)* Gr&c R<\SV`WR7VfVf \R4hR#R#))rPrQNz&Either text or images must be provided)rGvalidate_inputsr>)rIrPrQrRrMs&&&,r'rs ColPaliProcessor.validate_inputss1 v9 (>(^0D DE>(".$~2F"FG>( >( }- >( ^S01 >( >(r&c  8\V4^8Xd \R4h\V4^8Xd \R4hV^,PV^,P8wd \R4hV^,PV^,P8wd \R4hVfV^,Pp.p\ ^\V4V4EF1p.p\ P PPPWWs,R^R7p \ ^\V4V4Fp \ P PPPW*W,R^R7p VP\ P!RW4P^R7^,P^R74K VP\ P!V^R7PV4PV44EK4 \ P!V^R7#) a Compute the late-interaction/MaxSim score (ColBERT-like) for the given multi-vector query embeddings (`qs`) and passage embeddings (`ps`). For ColPali, a passage is the image of a document page. Because the embedding tensors are multi-vector and can thus have different shapes, they should be fed as: (1) a list of tensors, where the i-th tensor is of shape (sequence_length_i, embedding_dim) (2) a single tensor of shape (n_passages, max_sequence_length, embedding_dim) -> 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. zNo queries providedzNo passages providedz/Queries and passages must be on the same devicez-Queries and passages must have the same dtypeT) batch_first padding_valuez bnd,csd->bcns)dim)rnr>devicedtypermtorchnnutilsrnn pad_sequenceappendeinsummaxsumcatto) rIrrrrrscoresi batch_scores batch_queriesjbatch_passagess &&&&&& r'score_retrieval ColPaliProcessor.score_retrievals@  A %23 3 ! "a '34 4 A  % %);A)>)E)E ENO O A  $ $(:1(=(C(C CLM M  +A.44L%'q#./:\]"A"##LL-PTTYZT[\]^bbghbi C MM%))La8;;LILL][ \=yyQ''r&)r7r8rBr/r.)NNNzDescribe the image.z Question: )NNr{)Ncpu)r r!r"r#rr_rHrrerlrsr|rpropertyrrbrrrr%__classdictcell__ __classcell__)rMr3s@@r'r+r+5s3#R#RJ%%%N(GG88-$HH((F666D22>(>(>(r&r+)typingrrfeature_extraction_utilsr image_utilsrrprocessing_utilsrr r r tokenization_utils_baser r rrrrrrr?rmrDr+__all__)rs0r'rs*#4?XXOO7 -U  ).t5A$qgQ5RWX[R\8]R\Q4#waR\8]] X(~X(X(v  68]s B);B.