fTh  SSKJrJrJrJr SSKJrJrJrJ r J r SSK J r J r \"5(aSSKJr \"5(a SSKrSSKJrJr \ R("\5r\\\4r\\r\"\ "S S 95"S S \ 55rg) )AnyDictListUnion)add_end_docstringsis_torch_availableis_vision_availableloggingrequires_backends)Pipelinebuild_pipeline_init_args) load_imageN)(MODEL_FOR_OBJECT_DETECTION_MAPPING_NAMES,MODEL_FOR_TOKEN_CLASSIFICATION_MAPPING_NAMEST)has_image_processorc^\rSrSrSrU4SjrSrS\\\ \ 44U4Sjjr SSjr Sr SS jrS S S\\\44S jrS rU=r$)ObjectDetectionPipelinea Object detection pipeline using any `AutoModelForObjectDetection`. This pipeline predicts bounding boxes of objects and their classes. Example: ```python >>> from transformers import pipeline >>> detector = pipeline(model="facebook/detr-resnet-50") >>> detector("https://huggingface.co/datasets/Narsil/image_dummy/raw/main/parrots.png") [{'score': 0.997, 'label': 'bird', 'box': {'xmin': 69, 'ymin': 171, 'xmax': 396, 'ymax': 507}}, {'score': 0.999, 'label': 'bird', 'box': {'xmin': 398, 'ymin': 105, 'xmax': 767, 'ymax': 507}}] >>> # x, y are expressed relative to the top left hand corner. ``` Learn more about the basics of using a pipeline in the [pipeline tutorial](../pipeline_tutorial) This object detection pipeline can currently be loaded from [`pipeline`] using the following task identifier: `"object-detection"`. See the list of available models on [huggingface.co/models](https://huggingface.co/models?filter=object-detection). c>[TU]"U0UD6 URS:Xa[SURS35e[ US5 [ R"5nUR[5 URU5 g)NtfzThe z is only available in PyTorch.vision) super__init__ framework ValueError __class__r rcopyupdatercheck_model_type)selfargskwargsmappingrs _/var/www/auris/envauris/lib/python3.13/site-packages/transformers/pipelines/object_detection.pyr ObjectDetectionPipeline.__init__4sn $)&) >>T !tDNN#33QRS S$):??ACD g&c L0nSU;aUSUS'0nSU;aUSUS'U0U4$)Ntimeout threshold)r"r$preprocess_paramspostprocess_kwargss r&_sanitize_parameters,ObjectDetectionPipeline._sanitize_parameters?sL  +1)+< i ( & .4[.A { + "&888r(returncd>SU;aSU;aURS5US'[TU]"U0UD6$)a Detect objects (bounding boxes & classes) in the image(s) passed as inputs. Args: inputs (`str`, `List[str]`, `PIL.Image` or `List[PIL.Image]`): The pipeline handles three types of images: - A string containing an HTTP(S) link pointing to an image - A string containing a local path to an image - An image loaded in PIL directly The pipeline accepts either a single image or a batch of images. Images in a batch must all be in the same format: all as HTTP(S) links, all as local paths, or all as PIL images. threshold (`float`, *optional*, defaults to 0.5): The probability necessary to make a prediction. timeout (`float`, *optional*, defaults to None): The maximum time in seconds to wait for fetching images from the web. If None, no timeout is set and the call may block forever. Return: A list of dictionaries or a list of list of dictionaries containing the result. If the input is a single image, will return a list of dictionaries, if the input is a list of several images, will return a list of list of dictionaries corresponding to each image. The dictionaries contain the following keys: - **label** (`str`) -- The class label identified by the model. - **score** (`float`) -- The score attributed by the model for that label. - **box** (`List[Dict[str, int]]`) -- The bounding box of detected object in image's original size. imagesinputs)popr__call__)r"r#r$rs r&r6 ObjectDetectionPipeline.__call__Hs=@ v (&"8%zz(3F8 w000r(c:[XS9n[R"URUR//5nUR U/SS9nUR S:XaURUR5nURbURUSUSSS9nX4S'U$)N)r*pt)r3return_tensorswordsboxes)textr<r: target_size) rtorch IntTensorheightwidthimage_processorrto torch_dtype tokenizer)r"imager*r>r4s r& preprocess"ObjectDetectionPipeline.preprocessls52oo ekk'B&CD %%eWT%J >>T !YYt//0F >> %^^w`d^eF +} r(cURS5nUR"S0UD6nURSU0UE5nURbUSUS'U$)Nr>bboxr,)r5modelrrF)r" model_inputsr>outputs model_outputss r&_forward ObjectDetectionPipeline._forwardws\"&&}5 **,|,))=+*Q*QR >> %$0$8M& !r(c H^^^USnTRGbUSR5ummUUU4SjnUSRS5RSS9R SS9upVUR5Vs/sH&nTR R RUPM( nnUSRS5V s/sH o"U 5PM n n /SQn [UR5X5V s/sH!oSU:dM [[X55PM# n n U $TRRXU5nUSnUS nUS nUS n UR5US 'UVs/sH4nTR R RUR5PM6 snUS 'U Vs/sHnTRU5PM snUS '/SQn [US US US 5V s/sHn [[X55PM n n U $s snfs sn fs sn fs snfs snfs sn f) Nr>rc >TR[R"TUS-S- TUS-S- TUS-S- TUS-S- /55$)Nrir r)_get_bounding_boxr?Tensor)rKrAr"rBs r& unnormalize8ObjectDetectionPipeline.postprocess..unnormalizesi--LL"T!W_t3#d1g-4"T!W_t3#d1g-4   r(logits)dimrK)scorelabelboxscoreslabelsr<)rFtolistsqueezesoftmaxmaxrLconfigid2labelzipdictrCpost_process_object_detectionitemrU)r"rOr+r>rWr_classes predictionr`rKr<keysvals annotationraw_annotationsraw_annotationr]r^rArBs` @@r& postprocess#ObjectDetectionPipeline.postprocesss.#M2 >> %(N113MFE ,H5==a@HHRHPTTY[T\OFOV~~O_`O_djj''00>T !XY Y!$!1!1!3D    r(r,)N)g?)__name__ __module__ __qualname____firstlineno____doc__rr/r Predictionsr Predictionr6rHrPrrrstrryrU__static_attributes__ __classcell__)rs@r&rrs^0 '9"15d:>N1N+O"1H +Z^S#Xr(r)typingrrrrutilsrr r r r baserr image_utilsrr?models.auto.modeling_autorr get_loggerr{loggerrrrrr,r(r&rs))kk4(   H %#s(^ : ,FGdhdHdr(