S^(h(ddlmZmZmZddlmZddlmZddlm Z m Z m Z m Z m Z ddlmZmZGdde d ZGd d e d ZGd de d ZGdde ZdgZy))ListOptionalUnion) BatchFeature) ImageInput) ImagesKwargsProcessingKwargsProcessorMixin TextKwargsUnpack)PreTokenizedInput TextInputceZdZUeed<y)Emu3TextKwargsreturn_for_image_generationN)__name__ __module__ __qualname__bool__annotations__f/var/www/skyplay_api_hub/venv/lib/python3.12/site-packages/transformers/models/emu3/processing_emu3.pyrrs!%%rrF)totalc"eZdZUeed<eed<y)Emu3ImagesKwargsratio image_areaN)rrrstrrintrrrrrs JOrrc6eZdZUeed<eed<ddiddddZy ) Emu3ProcessorKwargs text_kwargs images_kwargsrFz1:1i)rr)r$r%N)rrrrrr _defaultsrrrr#r#"s.## *5   Irr#c eZdZdZddgZdgZdZdZ dfd Z dde e d e e e e ee ee fd eed efd Zd Zde fdZdZdZedZxZS) Emu3Processora Constructs a Emu3 processor which wraps a Emu3 image processor and a GPT2 tokenizer into a single processor. [`Emu3Processor`] offers all the functionalities of [`Emu3ImageProcessor`] and [`GPT2TokenizerFast`]. See the [`~Emu3Processor.__call__`] and [`~Emu3Processor.decode`] for more information. Args: image_processor ([`Emu3ImageProcessor`]): The image processor is a required input. tokenizer ([`Emu3TokenizerFast`]): The tokenizer is a required input. chat_template (`str`, *optional*): A Jinja template which will be used to convert lists of messages in a chat into a tokenizable string. image_processor tokenizer chat_template) GPT2TokenizerGPT2TokenizerFastEmu3ImageProcessorc |j|_|j|_|j|_|j |_|j|_|j|_d|_ t|-|||y)N)r+) image_token boi_tokenimage_start_token eoi_tokenimage_end_tokenimage_wrapper_tokenfake_token_around_image eof_token bos_tokendownsample_ratiosuper__init__)selfr)r*r+kwargs __class__s rr<zEmu3Processor.__init__Fss%00!*!4!4(22'0'D'D$",,",, ! )=Qrimagestextr>returnc t|tr|g}n.t|tst|dts td|jt fd|j ji|}|djdd}|djdd }|djd d } |r | td |s| | td i} |j} |j|j} |s| |j|fi|d} t| j} g}|D]}|j |vrt#| }|\}}||j$z}||j$z}||d zz}| |d||j&d|z| }|j)|j |d }|j*|}|j |vr|j-||Dcgc]}|j)d|j  }}nj|rh|j/|| |j$\}}| |d||j&}|Dcgc]}|j*||}}||ggt1|z| d<|j |fi|d}|j2di| t5||djdd Scc}wcc}w)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 Emu3TokenizerFast's [`~Emu3TokenizerFast.__call__`] if `text` is not `None` to encode the text. To prepare the image(s), this method forwards the `images` and `kwrags` arguments to CLIPImageProcessor's [`~CLIPImageProcessor.__call__`] if `images` is not `None`. Please refer to the docstring of the above two methods for more information. 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). 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`. rzAInvalid input text. Please provide a string, or a list of stringstokenizer_init_kwargsr$rFr%rNrzGYou should not provide `images` when `return_for_image_generation=True`zOYou must provide either text or images when `return_for_image_generation=False`*z image_sizes common_kwargsreturn_tensors)data tensor_typer) isinstancer list TypeError _merge_kwargsr#r* init_kwargspop ValueErrorr3r8r5r)iterrGr1nextr:r7replacer9appendcalculate_generate_sizelenupdater)r=r@rAaudiovideosr> output_kwargsrrrimage_featuresimage_start_tokensimage_end_tokensrGprompt_stringssample image_sizeheightwidthimage_seq_lengthimage_placeholder image_promptrJs r__call__zEmu3Processor.__call__VsCT dC 6DD$' 47C0H_` `***  "&.."<"<  '4M&B&F&FGdfk&l#o.227DA"?377 dK &6+=fg g*t|no o $ 6 67"nn-d.B.B-CD+v/A1T11&[M/vhawtOkOkNlm|@PnPmQRbQc)d%#^^D,<,<>OQRSF $/x8F&&&0%%f- .Ucc&FNN?D4D4DEcDc) 88 DLaLabMFE01&5'$B^B^A_`LLPQ&t~~&vh|n=QDQ.4e_,=D ,IN= )t~~dCmM&BC %n%=3Q3U3UVfhl3mnnd Rs ##J;Kctt|jd\}}||z}||z dz}tt||z|z }tt||z|z } || fS)N:g?)mapr!splitround) r=rrspatial_factorrdrc current_area target_ratio token_height token_widths rrWz%Emu3Processor.calculate_generate_sizespCS!12 vv~ "\1c9 5,!6!GHI % 4~ EFG [((rc <|jj|fi|SN)r) postprocess)r=r@r>s rruzEmu3Processor.postprocesss /t##//A&AArc:|jj|i|S)z This method forwards all its arguments to Emu3TokenizerFast's [`~PreTrainedTokenizer.batch_decode`]. Please refer to the docstring of this method for more information. )r* batch_decoder=argsr>s rrwzEmu3Processor.batch_decodes +t~~**D;F;;rc:|jj|i|S)z This method forwards all its arguments to Emu3TokenizerFast's [`~PreTrainedTokenizer.decode`]. Please refer to the docstring of this method for more information. )r*decoderxs rr{zEmu3Processor.decodes %t~~$$d5f55rc|jj}|jj}ttj ||zSrt)r*model_input_namesr)rMdictfromkeys)r=tokenizer_input_namesimage_processor_input_namess rr}zEmu3Processor.model_input_namess? $ @ @&*&:&:&L&L#DMM"7:U"UVWWrrt)NNNN)rrr__doc__ attributes valid_kwargstokenizer_classimage_processor_classr<rrrrrrr r#rrhrWrurwr{propertyr} __classcell__)r?s@rr(r(0s $[1J#$Lrsb")(2%bbC&Zu&|5 *% gXNgXT  r