/jtddlZddlZddlmZddlmZddlmZGddeZ Gdde eZ dS) N)LoadVisualPrompt)DetectionPredictor)SegmentationPredictorcVeZdZdZd deffd ZdZfdZd fd Zfd Z d Z xZ S) YOLOEVPDetectPredictoraaA class extending DetectionPredictor for YOLO-EVP (Enhanced Visual Prompting) predictions. This class provides common functionality for YOLO models that use visual prompting, including model setup, prompt handling, and preprocessing transformations. Attributes: model (torch.nn.Module): The YOLO model for inference. device (torch.device): Device to run the model on (CPU or CUDA). prompts (dict | torch.Tensor): Visual prompts containing class indices and bounding boxes or masks. Methods: setup_model: Initialize the YOLO model and set it to evaluation mode. set_prompts: Set the visual prompts for the model. pre_transform: Preprocess images and prompts before inference. inference: Run inference with visual prompts. get_vpe: Process source to get visual prompt embeddings. Tverbosec\t||d|_dS)zSet up the model for prediction. Args: model (torch.nn.Module): Model to load or use. verbose (bool, optional): If True, provides detailed logging. )rTN)super setup_model done_warmup)selfmodelr __class__s j/home/longshao/multi-rider-rag/.venv/lib/python3.11/site-packages/ultralytics/models/yolo/yoloe/predict.pyr z"YOLOEVPDetectPredictor.setup_models0 E7333c||_dS)zSet the visual prompts for the model. Args: prompts (dict): Dictionary containing class indices and bounding boxes or masks. Must include a 'cls' key with class indices. N)prompts)r rs r set_promptsz"YOLOEVPDetectPredictor.set_prompts(s rc tjddjdd}jdt dkrndjdddjdd|}|dj }nhJdd ttrtd DsJd d ttrtd DsJd d t t cxkrt ks=nJd t dt dt d fdtt D}tjjj|dj }jjr|n|_S)aPreprocess images and prompts before inference. This method applies letterboxing to the input image and transforms the visual prompts (bounding boxes or masks) accordingly. Args: im (list): List of input images. Returns: (list): Preprocessed images ready for model inference. Raises: ValueError: If neither valid bounding boxes nor masks are provided in the prompts. bboxesNmasksclsrzExpected bboxes, but got !c3JK|]}t|tjVdSN isinstancenpndarray.0bs r z7YOLOEVPDetectPredictor.pre_transform..Ks/3^3^RSJq"*4M4M3^3^3^3^3^3^rz#Expected list[np.ndarray], but got c3JK|]}t|tjVdSrrr"s rr%z7YOLOEVPDetectPredictor.pre_transform..Ns/5b5bTUjBJ6O6O5b5b5b5b5b5brz-Expected same length for all inputs, but got vsc g|]L}|jdd|jdd||MS)Nr)_process_single_imageshape)r#ircategoryimimgr s r z8YOLOEVPDetectPredictor.pre_transform..Tsf**3q6<+;RU[!_hWXk[abc[deerT) batch_first)r pre_transformrpoplenr)r* unsqueezetodevicerlistallrangetorchnnutilsrnn pad_sequencerfp16halffloat) r r-rvisualsrrr,r.rs `` @@@rr1z$YOLOEVPDetectPredictor.pre_transform1sgg##B''!!(D11   $//<& s88q==00Qbqb1A2a5;rPQr?T\^dfkllG''**--dk::GG%%'L6'L'L'L%%%fd++ 3^3^W]3^3^3^0^0^  ?f???  ^h-- #5b5bYa5b5b5b2b2b  AhAAA  br77c(mm::::s6{{:::::jBjj3x==jj\_`f\g\gjjj;::s3xxGhn(55g45PPSSTXT_``G)-Mw||~~~gmmoo  rNc |t|rtj|tj}|jdkr |dddf}t |d|dz |d|dz }||z}|ddddfxxt |dt |d|zz dz dz z cc<|ddddfxxt |dt |d|zz dz dz z cc<nP|?t|}tj |}d||dk<ntd t ||||S) aProcess a single image by resizing bounding boxes or masks and generating visuals. Args: dst_shape (tuple): The target shape (height, width) of the image. src_shape (tuple): The original shape (height, width) of the image. category (list | np.ndarray): The category indices for visual prompts. bboxes (list | np.ndarray, optional): A list of bounding boxes in the format [x1, y1, x2, y2]. masks (np.ndarray, optional): A list of masks corresponding to the image. Returns: (torch.Tensor): The processed visuals for the image. Raises: ValueError: If neither `bboxes` nor `masks` are provided. N)dtyperr.rg?rz$Please provide valid bboxes or masks) r3r arrayfloat32ndimminroundr r1stack ValueErrorr get_visuals) r dst_shape src_shaper,rrgain resized_masksrs rr)z,YOLOEVPDetectPredictor._process_single_image\s  #f++ XfBJ777F{aaaay|il2IaL9Q<4OPPD dNF 319    ! uYq\D=P7Q7Q(QUV'VY\'\!]!] ]    319    ! uYq\D=P7Q7Q(QUV'VY\'\!]!] ]      !GG11%88MH]++E"#E%3,  CDD D !!--h 65QQQrcLtj|g|Rd|ji|S)a&Run inference with visual prompts. Args: im (torch.Tensor): Input image tensor. *args (Any): Variable length argument list. **kwargs (Any): Arbitrary keyword arguments. Returns: (torch.Tensor): Model prediction results. vpe)r inferencer)r r-argskwargsrs rrTz YOLOEVPDetectPredictor.inferences3!uww GGGGGGGGrc||t|jdks Jd|jD]9\}}}||}|||jdcSdS)aProcess the source to get the visual prompt embeddings (VPE). Args: source (str | Path | int | PIL.Image | np.ndarray | torch.Tensor | list | tuple): The source of the image to make predictions on. Accepts various types including file paths, URLs, PIL images, numpy arrays, and torch tensors. Returns: (torch.Tensor): The visual prompt embeddings (VPE) from the model. rz get_vpe only supports one image!T)rS return_vpeN) setup_sourcer3dataset preprocessrr)r source_im0sr-s rget_vpezYOLOEVPDetectPredictor.get_vpes &!!!4<  A%%%'I%%%, E EJAtQ&&B::bdlt:DD D D D E Er)T)NN) __name__ __module__ __qualname____doc__boolr rr1r)rTr_ __classcell__)rs@rrr s$  $      )))))V"R"R"R"R"R"RH H H H H HEEEEEEErrceZdZdZdS)YOLOEVPSegPredictorz\Predictor for YOLO-EVP segmentation tasks combining detection and segmentation capabilities.N)r`rarbrcrrrgrgsffDrrg) numpyr r:ultralytics.data.augmentrultralytics.models.yolo.detectrultralytics.models.yolo.segmentrrrgrhrrrms 555555======AAAAAAQEQEQEQEQE/QEQEQEh     02G     r