o W hG@sddlmZddlmZddlZddlZddlmm Z ddl m Z ddl mZmZmZddlmZddlmZmZmZddlmZmZGd d d e ZdS) ) ThreadPool)PathN)DetectionValidator)LOGGER NUM_THREADSops)check_requirements)SegmentMetricsbox_ioumask_iou)output_to_target plot_imagescseZdZdZd"fdd ZfddZfddZd d Zfd d Zfd dZ fddZ ddZ ddZ d#ddZ ddZddZddZddZd d!ZZS)$SegmentationValidatora A class extending the DetectionValidator class for validation based on a segmentation model. This validator handles the evaluation of segmentation models, processing both bounding box and mask predictions to compute metrics such as mAP for both detection and segmentation tasks. Attributes: plot_masks (list): List to store masks for plotting. process (callable): Function to process masks based on save_json and save_txt flags. args (namespace): Arguments for the validator. metrics (SegmentMetrics): Metrics calculator for segmentation tasks. stats (dict): Dictionary to store statistics during validation. Examples: >>> from ultralytics.models.yolo.segment import SegmentationValidator >>> args = dict(model="yolo11n-seg.pt", data="coco8-seg.yaml") >>> validator = SegmentationValidator(args=args) >>> validator() Ncs:t|||||d|_d|_d|j_t|jd|_dS)a Initialize SegmentationValidator and set task to 'segment', metrics to SegmentMetrics. Args: dataloader (torch.utils.data.DataLoader, optional): Dataloader to use for validation. save_dir (Path, optional): Directory to save results. pbar (Any, optional): Progress bar for displaying progress. args (namespace, optional): Arguments for the validator. _callbacks (list, optional): List of callback functions. Nsegment)save_dir) super__init__ plot_masksprocessargstaskr rmetrics)self dataloaderrpbarr _callbacks __class__W/var/www/vscode/kcb/lib/python3.10/site-packages/ultralytics/models/yolo/segment/val.pyr&s  zSegmentationValidator.__init__cs(t|}|d|j|d<|S)zDPreprocess batch by converting masks to float and sending to device.masks)r preprocesstodevicefloat)rbatchrrrr!7s z SegmentationValidator.preprocesscsZt|g|_|jjrtd|jjs|jjrtjntj |_ t ggggggd|_ dS)z Initialize metrics and select mask processing function based on save_json flag. Args: model (torch.nn.Module): Model to validate. pycocotools>=2.0.6)tp_mtpconfpred_cls target_cls target_imgN) r init_metricsrr save_jsonrsave_txtrprocess_mask_native process_maskrdictstats)rmodelrrrr-=s z"SegmentationValidator.init_metricscCsddS)z5Return a formatted description of evaluation metrics.z,%22s%11s%11s%11s%11s%11s%11s%11s%11s%11s%11s) ClassImages InstanceszBox(PRmAP50 mAP50-95)zMask(Pr8r9r:r)rrrrget_descLszSegmentationValidator.get_desccs<t|d}t|ddkr|ddn|d}||fS)a6 Post-process YOLO predictions and return output detections with proto. Args: preds (list): Raw predictions from the model. Returns: p (torch.Tensor): Processed detection predictions. proto (torch.Tensor): Prototype masks for segmentation. r)r postprocesslen)rpredspprotorrrr?\s $z!SegmentationValidator.postprocesscs<t||}|jjr|gn|d|k}|d||d<|S)a+ Prepare a batch for training or inference by processing images and targets. Args: si (int): Batch index. batch (dict): Batch data containing images and targets. Returns: (dict): Prepared batch with processed images and targets. batch_idxr )r_prepare_batchr overlap_mask)rsir%prepared_batchmidxrrrrEks z$SegmentationValidator._prepare_batchcsLt||}|j||ddddf|ddddf|dd}||fS)a Prepare predictions for evaluation by processing bounding boxes and masks. Args: pred (torch.Tensor): Raw predictions from the model. pbatch (dict): Prepared batch data. proto (torch.Tensor): Prototype masks for segmentation. Returns: predn (torch.Tensor): Processed bounding box predictions. pred_masks (torch.Tensor): Processed mask predictions. Nimgsz)shape)r _prepare_predr)rpredpbatchrCpredn pred_masksrrrrN{s 6z#SegmentationValidator._prepare_predc Cstt|d|dD]Y\}\}}|jd7_t|}ttjd|jdtjd|jdtj||jtj |jdtj||jtj |jdd}| ||}| d| d} } t| } | |d<| |d <|dkr| r|j D] } |j | || qo|jjr|jjd | | d q | d } |jjrd|d d d f<||||\}}|d d df|d<|d d d f|d<| r||| | |d<|j|| | || |jjdd|d<|jjr|j|| | |j D] } |j | || qtj|tjd}|jjr|jdkr|j|d d|jddkrtd|jj rD|!||d|t"j#|$ddd%&|d|d|d|jj're|(|||jj)|d|j*dt+|d|j,dq d S) z Update metrics with the current batch predictions and targets. Args: preds (list): Predictions from the model. batch (dict): Batch data containing images and targets. rr<r#)dtyper#)r)r*r(r'clsbboxr+r,N) detections gt_bboxesgt_clsr rKr)r*r(T)r r'rTr=2zBLimiting validation plots to first 50 items per image for speed...im_file ori_shape ratio_pad)r`labelsz.txt)- enumeratezipseenr@r2torchzerosr#niouboolrEpopuniquer3keysappendrplotsconfusion_matrix process_batch single_clsrN_process_batchrF as_tensoruint8batch_ircpurMrwarningr. pred_to_jsonr scale_imagepermute contiguousnumpyr/ save_one_txt save_confrrstem)rrAr%rGrOrCnprstatrPrUrVnlkgt_masksrQrRrrrupdate_metricssz$          z$SegmentationValidator.update_metricscOs|j|j_|j|j_dS)a Finalize evaluation metrics by setting the speed attribute in the metrics object. This method is called at the end of validation to set the processing speed for the metrics calculations. It transfers the validator's speed measurement to the metrics object for reporting. Args: *args (Any): Variable length argument list. **kwargs (Any): Arbitrary keyword arguments. N)speedrrn)rrkwargsrrrfinalize_metricss z&SegmentationValidator.finalize_metricsFc Cs|r`|r't|}tj||jd|ddd} ||dd}t|| kdd}|jdd|jddkrLtj |d|jdddddd }| d }t ||jd d ||jd d } n t ||dddd f} | |ddd f|| S)a Compute correct prediction matrix for a batch based on bounding boxes and optional masks. Args: detections (torch.Tensor): Tensor of shape (N, 6) representing detected bounding boxes and associated confidence scores and class indices. Each row is of the format [x1, y1, x2, y2, conf, class]. gt_bboxes (torch.Tensor): Tensor of shape (M, 4) representing ground truth bounding box coordinates. Each row is of the format [x1, y1, x2, y2]. gt_cls (torch.Tensor): Tensor of shape (M,) representing ground truth class indices. pred_masks (torch.Tensor, optional): Tensor representing predicted masks, if available. The shape should match the ground truth masks. gt_masks (torch.Tensor, optional): Tensor of shape (M, H, W) representing ground truth masks, if available. overlap (bool): Flag indicating if overlapping masks should be considered. masks (bool): Flag indicating if the batch contains mask data. Returns: (torch.Tensor): A correct prediction matrix of shape (N, 10), where 10 represents different IoU levels. Note: - If `masks` is True, the function computes IoU between predicted and ground truth masks. - If `overlap` is True and `masks` is True, overlapping masks are taken into account when computing IoU. Examples: >>> detections = torch.tensor([[25, 30, 200, 300, 0.8, 1], [50, 60, 180, 290, 0.75, 0]]) >>> gt_bboxes = torch.tensor([[24, 29, 199, 299], [55, 65, 185, 295]]) >>> gt_cls = torch.tensor([1, 0]) >>> correct_preds = validator._process_batch(detections, gt_bboxes, gt_cls) rSr<g?gNbilinearF)mode align_cornersrg?r>rKrZ)r@rearanger#viewrepeatwhererMF interpolategt_r r match_predictions) rrWrXrYrRroverlapr rindexiourrrrqs$ (z$SegmentationValidator._process_batchc CsNt|d|d|dd|d|d|d|jd|d |j|jd d S) z Plot validation samples with bounding box labels and masks. Args: batch (dict): Batch data containing images and targets. ni (int): Batch index. imgrDrUr>bboxesr r] val_batchz _labels.jpg)r pathsfnamenameson_plotN)r squeezerrr)rr%nirrrplot_val_sampless  z&SegmentationValidator.plot_val_samplescCspt|dgt|dddt|jrtj|jddn|jR|d|jd|d|j|jd |j d S) z Plot batch predictions with masks and bounding boxes. Args: batch (dict): Batch data containing images. preds (list): Predictions from the model. ni (int): Batch index. rrr\)max_det)dimr]rz _pred.jpg)rrrrN) r r r@rrecatrrrclear)rr%rArrrrplot_predictions's  z&SegmentationValidator.plot_predictionscCsVddlm}|tj|d|dftjdd|j|ddddf|dj||ddS) a Save YOLO detections to a txt file in normalized coordinates in a specific format. Args: predn (torch.Tensor): Predictions in the format [x1, y1, x2, y2, conf, cls]. pred_masks (torch.Tensor): Predicted masks. save_conf (bool): Whether to save confidence scores. shape (tuple): Original image shape. file (Path): File path to save the detections. r)Resultsr<r[NrJ)pathrboxesr )r})ultralytics.engine.resultsrnprfrsrr/)rrQrRr}rMfilerrrrr|;s  z"SegmentationValidator.save_one_txtc s&ddlmfdd}t|j}|rt|n|}t|ddddf}|ddddf|ddddfd8<t |d}t t }| ||} Wdn1s[wYt t||D]%\} \} } |j||jt| d d d | Dt| dd | | d qkdS) a Save one JSON result for COCO evaluation. Args: predn (torch.Tensor): Predictions in the format [x1, y1, x2, y2, conf, cls]. filename (str): Image filename. pred_masks (numpy.ndarray): Predicted masks. Examples: >>> result = {"image_id": 42, "category_id": 18, "bbox": [258.15, 41.29, 348.26, 243.78], "score": 0.236} rencodecs@tj|dddddfdddd}|dd|d<|S)z:Encode predicted masks as RLE and append results to jdict.Nrrs)orderrTrcountszutf-8)rasarraydecode)xrlerrr single_encode^s*z9SegmentationValidator.pred_to_json..single_encodeNrKr^)r^rr<rZcSsg|]}t|dqS)r=)round.0rrrr psz6SegmentationValidator.pred_to_json..)image_id category_idrVscore segmentation)pycocotools.maskrrr~ isnumericintr xyxy2xywhr transposerrmaprbrctolistjdictrl class_mapr) rrQfilenamerRrr~rboxpoolrlesirBbrrrrwPs(  0  "  z"SegmentationValidator.pred_to_jsonc Cs|jjrU|js |jrUt|jrU|jd}|jdd|jr"dnd|jjd}|jr0dnd}t d |d |d |d z||fD] }| sSJ|d qFt |jrZdnd|jrddl m}ddlm}|t|}|t|} ||| d||| dg} n!ddlm} m} | t|}|t|} | || d| || dg} t| D]{\} }dd|jjjD|j_||||jr| | dd}|jr|j!ddn |j"d|j"dg\||j#j$|d<||j#j$|<|jr&| dkrdnd}|j"d |d!|d"<|j"d#|d$|d"<|j"d%|d&|d"<q|jr4|d'|d(<W|SW|St%yT}zt &|d)|WYd}~|Sd}~ww|S)*z6Return COCO-style object detection evaluation metrics.zpredictions.jsonr annotationszinstances_val2017.jsonlvis_v1_z.json pycocotoolslvisz Evaluating z mAP using z and z...z file not foundr&z lvis>=0.5.3r)COCO)COCOevalrVsegm)LVISLVISEvalcSsg|] }tt|jqSr)rrr~rrrrrsz3SegmentationValidator.eval_json..rKr^NAPAP50r<BMAPrz metrics/APr()APcz metrics/APc(APfz metrics/APf(zmetrics/mAP50-95(B)fitnessz unable to run: )'rr.is_lvisis_cocor@rrdatasplitrinfois_filerpycocotools.cocorpycocotools.cocoevalrstrloadResrrr _load_jsonrbrdatasetim_filesparamsimgIdsevaluate accumulate summarize print_resultsr3resultsrrk Exceptionrv)rr3 pred_json anno_jsonpkgrrrannorOvalsrrrevalidxtagerrr eval_jsonvsf$       &"zSegmentationValidator.eval_json)NNNNN)NNFF)__name__ __module__ __qualname____doc__rr!r-r;r?rErNrrrqrrr|rwr __classcell__rrrrrs"     M ,&r)multiprocessing.poolrpathlibrr{rretorch.nn.functionalnn functionalrultralytics.models.yolo.detectrultralytics.utilsrrrultralytics.utils.checksrultralytics.utils.metricsr r r ultralytics.utils.plottingr r rrrrrs