h;ddlmZmZddlZddlmZddlmZddl m Z m Z m Z m Z mZmZddlmZmZddlmZdd lmZej.eZGd d ed ZGdde ZGdde d ZGdde ZdgZy))OptionalUnionN) BatchFeature) ImageInput) ImagesKwargsMultiModalDataProcessingKwargsProcessorMixinUnpack VideosKwargs)PreTokenizedInput TextInput)logging) VideoInputc(eZdZUeeeefed<y)Glm4vVideosProcessorKwargsfpsN)__name__ __module__ __qualname__rlistfloat__annotations__h/var/www/html/eduruby.in/venv/lib/python3.12/site-packages/transformers/models/glm4v/processing_glm4v.pyrr$s tE{E! ""rrF)totalc>eZdZUeeed<eeed<eeed<y)Glm4vImagesKwargs patch_sizetemporal_patch_size merge_sizeN)rrrrintrrrrr r (s  !#& rr c6eZdZUeed<dddddidZeed<y) Glm4vProcessorKwargs images_kwargsF)paddingreturn_mm_token_type_idsreturn_metadataT) text_kwargs videos_kwargsr,N)rrrr r _defaultsrrrrr&r&.s.$$(- ,T2 I.-rr&c eZdZdZgdZdZdZdZdfd Z dde de e e e e e e fd ed eed ef d Zdd Z ddZxZS)Glm4vProcessora Constructs a GLM-4V processor which wraps a GLM-4V image processor and a GLM-4 tokenizer into a single processor. [`~Glm4vProcessor.__call__`] and [`~Glm4vProcessor.decode`] for more information. Args: image_processor ([`Glm4vProcessor`], *optional*): The image processor is a required input. tokenizer ([`PreTrainedTokenizerFast`], *optional*): The tokenizer is a required input. video_processor ([`Glm4vVideoProcessor`], *optional*): The video processor 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 tokenizervideo_processorAutoImageProcessorAutoVideoProcessor)PreTrainedTokenizerPreTrainedTokenizerFastc t|||||t|dsdn |j|_t|dsdn |j|_t |ddr |j n|j|j|_t |ddr|j|_y|j|j|_y)N) chat_template image_tokenz <|image|> video_tokenz <|video|>image_token_idvideo_token_id) super__init__hasattrr9r:getattrr;convert_tokens_to_idsr<)selfr0r1r2r8kwargs __class__s rr>zGlm4vProcessor.__init__Os )_Tab.5i.O;U^UjUj.5i.O;U^UjUjy"2D9  $ $001A1AB y"2D9  $ $ 001A1AB rimagestextvideosrCreturnc z|jtfd|jji|}||jdd|i|d}|d}ni}d}|8|j dd|i|d}d|vr|j d } n|d } |d } ni}d} t|ts|g}|j}||jjd z} d } tt|D]} |j|| vrS|| j| z}|| j|jd |zd|| <| dz } |j|| vrS|| jd |j|| <|  |j jd z} d }tt|D]} |j || vr| |d }d} | }|j"t$j'd|j"dn |j"|_|j(ddd }g}td t|D]}|j+|||d|}t||kr'|j+|r|dnd t||kr't|D]'}||}d|jdt-|}||z })|| j|j |d|| <| |j| z| |d z}t|D]:}|j|| vs|| j|jd |zd|| <<|dz }|j || vr|| jd |j|| <|dj dd}|dj dd}|j|fi|d}|j/||ddg|rUt1j2|d}t1j4|d}d|||j6k(<|j9|d<t;i||||S) 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 PreTrainedTokenizerFast's [`~PreTrainedTokenizerFast.__call__`] if `text` is not `None` to encode the text. 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). videos (`np.ndarray`, `torch.Tensor`, `List[np.ndarray]`, `List[torch.Tensor]`): The image or batch of videos to be prepared. Each video can be a 4D NumPy array or PyTorch tensor, or a nested list of 3D frames. Both channels-first and channels-last formats are supported. 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`. - **pixel_values_videos** -- Pixel values of videos to be fed to a model. Returned when `videos` is not `None`. - **image_grid_thw** -- List of image 3D grid in LLM. Returned when `images` is not `None`. - **video_grid_thw** -- List of video 3D grid in LLM. Returned when `videos` is not `None`. tokenizer_init_kwargsNrEr'image_grid_thwrGr,r*video_metadatavideo_grid_thwrz<|placeholder|>aSmolVLM requires frame timestamps to construct prompts, but the `fps` of the input video could not be inferred. Probably `video_metadata` was missing from inputs and you passed pre-sampled frames. Defaulting to `fps=24`. Please provide `video_metadata` for more accurate results.z<|begin_of_image|>z<|end_of_image|>r+return_tensorsr)Fimagevideo) modalities input_idsmm_token_type_ids)data tensor_typer) _merge_kwargsr&r1 init_kwargsr0r2pop isinstancercopyr#rangelenr9prodreplacer:rlogger warning_once timestampsappendr$_check_special_mm_tokensnparray zeros_liker;tolistr)rBrErFrGrC output_kwargs image_inputsrK videos_inputsrLrM merge_lengthindexinum_image_tokens video_index num_framesvideo_structuremetadatarfunique_timestampsidxselected_timestamps frame_idx timestamp_secframe_structurerSr) text_inputs array_idsrXs r__call__zGlm4vProcessor.__call__^sT+**  "&.."<"<   /4//`v`A_`L)*:;NL!N  0D00aa-P_B`aM .!.!2!23C!D!./?!@*+;.>@QTd@dfghDGQJE&&$q'1q'//*;T=M=MNQ  O  %//::A=LK3t9%& O&&$q'1!/ !t?O?O>PP`aderas`t*u'?:; #1good.>.>QRSDG&{388:lJn]hNijkNll%&+:%6q ++tAw6&*1good6F6FHY\lHlno&pDGq 1$KG&&$q'1Jq'//*;T=M=MNQM& ON'}599:JDQ#0#?#C#CD^`e#f $dnnTJ]=-IJ  %%dKWgDV%W #[!9:I " k+.F G BC i4+>+>> ?/@/G/G/IK+ ,!QK!Q>PRSM   (&**<>a$BVBVBaBaJ#.!A$$@@\*\m\! !Sdd; A !=d d   4D[lm n  "0::>>PRSM   (#.!A$$@@\*\m\! !Sdd; A !=d d.>K* +, ,,#! e ! es$%D7D<+%EEc B|jj|f||d|S)a Post-process the output of the model to decode the text. Args: generated_outputs (`torch.Tensor` or `np.ndarray`): The output of the model `generate` function. The output is expected to be a tensor of shape `(batch_size, sequence_length)` or `(sequence_length,)`. skip_special_tokens (`bool`, *optional*, defaults to `True`): Whether or not to remove special tokens in the output. Argument passed to the tokenizer's `batch_decode` method. clean_up_tokenization_spaces (`bool`, *optional*, defaults to `False`): Whether or not to clean up the tokenization spaces. Argument passed to the tokenizer's `batch_decode` method. **kwargs: Additional arguments to be passed to the tokenizer's `batch_decode method`. Returns: `list[str]`: The decoded text. )skip_special_tokensclean_up_tokenization_spaces)r1 batch_decode)rBgenerated_outputsrrrCs rpost_process_image_text_to_textz.Glm4vProcessor.post_process_image_text_to_text s5(+t~~**  3)E    r)NNNN)NNN)NN)TF)rrr__doc__ attributesimage_processor_classvideo_processor_classtokenizer_classr>rrrrrrr r&rrrr __classcell__)rDs@rr/r/:s EJ00HO  ""^b! DoDoI0$y/4HYCZZ[Do Do -. Do  DoL$-NY^ rr/)typingrrnumpyrifeature_extraction_utilsr image_utilsrprocessing_utilsrr r r r r tokenization_utils_baserrutilsr video_utilsr get_loggerrrdrr r&r/__all__rrrrsx*#4%ttC%   H %#U#  .+5 .i ^i X  r