o Zh)@sddlmZmZmZddlmZddlmZddlm Z m Z m Z m Z m Z ddlmZmZGddde d d ZGd d d e d d ZGd dde d d ZGddde ZdgZdS))ListOptionalUnion) BatchFeature) ImageInput) ImagesKwargsProcessingKwargsProcessorMixin TextKwargsUnpack)PreTokenizedInput TextInputc@seZdZUeed<dS)Emu3TextKwargsreturn_for_image_generationN)__name__ __module__ __qualname__bool__annotations__rrW/var/www/auris/lib/python3.10/site-packages/transformers/models/emu3/processing_emu3.pyrs rF)totalc@seZdZUeed<eed<dS)Emu3ImagesKwargsratio image_areaN)rrrstrrintrrrrrs  rc@s2eZdZUeed<eed<ddiddddZd S) Emu3ProcessorKwargs text_kwargs images_kwargsrFz1:1i)rr)rr N)rrrrrr _defaultsrrrrr"s  rc seZdZdZddgZdgZdZdZ dfdd Z    dd e e d e e e e ee ee fd eed efddZddZd e fddZddZddZeddZZS) 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)Z GPT2TokenizerZGPT2TokenizerFastZEmu3ImageProcessorNc sT|j|_|j|_|j|_|j|_|j|_|j|_|j |_ d|_ t j |||ddS)N)r%) image_tokenZimage_token_idZ boi_tokenimage_start_tokenZ eoi_tokenimage_end_tokenZimage_wrapper_tokenfake_token_around_image eof_token bos_tokendownsample_ratiosuper__init__)selfr#r$r%kwargs __class__rrr/FszEmu3Processor.__init__imagestextr1returnc sVt|tr |g}nt|tst|dtstdjtfdjji|}|ddd}|ddd }|dd d } |rH|d urHt d |sV|d urV|d urVt d i} j } j j } |s|d urӈj |fi|d} t| j} g}|D]J}j|vrt| }|\}}|j}|j}||d }| |d|jd|| }|j|d }j|}j|vs||q~fdd|D}n.|r|| j\}}| |d|jfdd|D}||ggt|| d<|ddd }j|fi|d}j||dgd|jdi| t||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 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 stringsZtokenizer_init_kwargsrrFr rNrzGYou should not provide `images` when `return_for_image_generation=True`zOYou must provide either text or images when `return_for_image_generation=False`* csg|] }|djqS)r9)replacer'.0sample)r0rr sz*Emu3Processor.__call__..csg|] }j|qSr)r,r;Z image_promptr0rrr>s image_sizesreturn_tensorsimage)Z modalities)dataZ tensor_typer) isinstancerlist TypeErrorZ _merge_kwargsrr$Z init_kwargspop ValueErrorr(r+r)r#iterr@r'nextr-r*r:r,appendcalculate_generate_sizelenZ_check_special_mm_tokensupdater)r0r4r5ZaudioZvideosr1Z output_kwargsrrrZimage_featuresZimage_start_tokensZimage_end_tokensr@Zprompt_stringsr=Z image_sizeheightwidthZimage_seq_lengthZimage_placeholderrArCrr?r__call__Ws` *       $  zEmu3Processor.__call__c CsXtt|d\}}||}||d}tt|||}tt|||} || fS)N:g?)maprsplitround) r0rrZspatial_factorrPrOZ current_areaZ target_ratioZ token_heightZ token_widthrrrrLs  z%Emu3Processor.calculate_generate_sizecKs|jj|fi|SN)r# postprocess)r0r4r1rrrrWszEmu3Processor.postprocesscO|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_decoder0argsr1rrrrYzEmu3Processor.batch_decodecOrX)z This method forwards all its arguments to Emu3TokenizerFast's [`~PreTrainedTokenizer.decode`]. Please refer to the docstring of this method for more information. )r$decoderZrrrr]r\zEmu3Processor.decodecCs"|jj}|jj}tt||SrV)r$model_input_namesr#rEdictfromkeys)r0Ztokenizer_input_namesZimage_processor_input_namesrrrr^szEmu3Processor.model_input_namesrV)NNNN)rrr__doc__ attributesZ valid_kwargsZtokenizer_classZimage_processor_classr/rrrrr rr rrrQrLrWrYr]propertyr^ __classcell__rrr2rr"0s6 f r"N)typingrrrZimage_processing_utilsrZ image_utilsrZprocessing_utilsrr r r r Ztokenization_utils_baser rrrrr"__all__rrrrs   /