cCi!@SSKrSSKJr SSKJrJrJr SSKJrJ r J r J r J r J r SSKJrJr \ "5(a SSKJr SS KJr \ "5(a SSKrSS KJr \ "5(a SS KJr SS KJr \ R8"\5r\"\"S S95"SS\55rg)N)UserDict)AnyUnionoverload)add_end_docstringsis_tf_availableis_torch_availableis_vision_availableloggingrequires_backends)Pipelinebuild_pipeline_init_args)Image) load_image)6MODEL_FOR_ZERO_SHOT_IMAGE_CLASSIFICATION_MAPPING_NAMES)9TF_MODEL_FOR_ZERO_SHOT_IMAGE_CLASSIFICATION_MAPPING_NAMES)stable_softmaxT)has_image_processorcn^\rSrSrSrSrSrSrSrU4Sjr \ S\ \ S4S\ \ S \S \ \\ \44S j5r\ S\ \ \ \ S4S\ \ S \S \ \ \\ \44S j5rS\ \ \ \ S\ S4S\ \ S \S \ \ \\ \4\ \ \\ \444U4S jjrSSjrSSjrSrSrSrU=r$)#ZeroShotImageClassificationPipeline!a Zero shot image classification pipeline using `CLIPModel`. This pipeline predicts the class of an image when you provide an image and a set of `candidate_labels`. Example: ```python >>> from transformers import pipeline >>> classifier = pipeline(model="google/siglip-so400m-patch14-384") >>> classifier( ... "https://huggingface.co/datasets/Narsil/image_dummy/raw/main/parrots.png", ... candidate_labels=["animals", "humans", "landscape"], ... ) [{'score': 0.965, 'label': 'animals'}, {'score': 0.03, 'label': 'humans'}, {'score': 0.005, 'label': 'landscape'}] >>> classifier( ... "https://huggingface.co/datasets/Narsil/image_dummy/raw/main/parrots.png", ... candidate_labels=["black and white", "photorealist", "painting"], ... ) [{'score': 0.996, 'label': 'black and white'}, {'score': 0.003, 'label': 'photorealist'}, {'score': 0.0, 'label': 'painting'}] ``` Learn more about the basics of using a pipeline in the [pipeline tutorial](../pipeline_tutorial) This image classification pipeline can currently be loaded from [`pipeline`] using the following task identifier: `"zero-shot-image-classification"`. See the list of available models on [huggingface.co/models](https://huggingface.co/models?filter=zero-shot-image-classification). FTc >[TU]"S0UD6 [US5 URURS:Xa [ 5 g[ 5 g)Nvisiontf)super__init__r check_model_type frameworkrr)selfkwargs __class__s o/home/james-whalen/.local/lib/python3.13/site-packages/transformers/pipelines/zero_shot_image_classification.pyr,ZeroShotImageClassificationPipeline.__init__HsH "6"$) ~~% F H imagez Image.Imagecandidate_labelsr#returnc gNrr"r(r)r#s r%__call__,ZeroShotImageClassificationPipeline.__call__Rs #r'c gr,rr-s r%r.r/Ws&)r'c t>SU;aURS5nUc [S5e[TU] "U4SU0UD6$)a; Assign labels to the image(s) passed as inputs. Args: image (`str`, `list[str]`, `PIL.Image` or `list[PIL.Image]`): The pipeline handles three types of images: - A string containing a http link pointing to an image - A string containing a local path to an image - An image loaded in PIL directly candidate_labels (`list[str]`): The candidate labels for this image. They will be formatted using *hypothesis_template*. hypothesis_template (`str`, *optional*, defaults to `"This is a photo of {}"`): The format used in conjunction with *candidate_labels* to attempt the image classification by replacing the placeholder with the candidate_labels. Pass "{}" if *candidate_labels* are already formatted. timeout (`float`, *optional*, defaults to None): The maximum time in seconds to wait for fetching images from the web. If None, no timeout is set and the call may block forever. Return: A list of dictionaries containing one entry per proposed label. Each dictionary contains the following keys: - **label** (`str`) -- One of the suggested *candidate_labels*. - **score** (`float`) -- The score attributed by the model to that label. It is a value between 0 and 1, computed as the `softmax` of `logits_per_image`. imageszSCannot call the zero-shot-image-classification pipeline without an images argument!r))pop ValueErrorrr.)r"r(r)r#r$s r%r.r/\sHJ v JJx(E =rs swS8HSFSSr'c 0nSU;aUSUS'SU;aUSUS'SU;aUSUS'Ub[R"S[5 XS'U004$)Nr)timeouthypothesis_templatez^The `tokenizer_kwargs` argument is deprecated and will be removed in version 5 of Transformerstokenizer_kwargs)warningswarn FutureWarning)r"r8r#preprocess_paramss r%_sanitize_parameters8ZeroShotImageClassificationPipeline._sanitize_parameterss  '4:;M4N 0 1  +1)+< i ( F *7=>S7T 3 4  ' MMp 5E0 1 "b((r'cUc0n[XS9nURU/URS9nURS:XaURUR5nX&S'UVs/sHosR U5PM nnSS0n SUR RR;aU RSS SS 9 U RU5 UR"U4S UR0U D6n U /US 'U$s snf) N)r6)r2return_tensorsptr)paddingTsiglip max_length@)rBrD truncationr@ text_inputs) rimage_processorr!todtypeformatmodelconfig model_typeupdate tokenizer) r"r(r)r7r6r8inputsx sequencestokenizer_default_kwargsrGs r% preprocess.ZeroShotImageClassificationPipeline.preprocesss  #! 52%%eWT^^%T >>T !YYtzz*F%5!"?"&&}5 k!nh / /%a.K&a.+K**;{;l;!1.. r'cURS5nUSSnURS:XarSURRR;aN[ R "U5RS5nUR5n[U[5(dU/nOURS:XaGURSS9RS5nUR5n[U[5(dU/nOQURS:Xa)[USS 9nUR5R5nO[S UR35e[[!XR5S S 9VVs/sH upgXgS .PM nnnU$s snnf)Nr)rXrrArC)dimr)axiszUnsupported framework: cUS*$)Nrr)rRs r%AZeroShotImageClassificationPipeline.postprocess..s_`ab_c^cr')key)scorelabel)r3r!rLrMrNtorchsigmoidsqueezetolistrYlistsoftmaxrnumpyr4sortedzip) r"r]r)rXprobsscoresrhcandidate_labelresults r% postprocess/ZeroShotImageClassificationPipeline.postprocesssK(,,-?@x(+ >>T !h$**2C2C2N2N&NMM&)11"5E\\^Ffd++  ^^t #NNrN*2226E\\^Ffd++  ^^t #"63E[[]))+F6t~~6FGH H+1V1NTc*d *d& 6*d    sE0rr,)NzThis is a photo of {}.NN)__name__ __module__ __qualname____firstlineno____doc___load_processor_load_image_processor_load_feature_extractor_load_tokenizerrrrstrrnrdictr.r=rUr^rw__static_attributes__ __classcell__)r$s@r%rr!so@O #O #3 -.#BFs)#WZ# d38n ##)49d=&99:)NRSVi)cf) d4S>" #)))TS$s)]D4GGH)Ts))T )T tDcN#T$tCH~*>%?? @ )TV)(4 0"r'r) r9 collectionsrtypingrrrutilsrr r r r r baserrPILr image_utilsrrjmodels.auto.modeling_autormodels.auto.modeling_tf_autortf_utilsr get_loggerryloggerrrr'r%rs ''5(bh)   H %,FGv(vHvr'