a yf6@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 Zd d Zfd dZ fddZ ddZ ddZ d#ddZ ddZddZddZddZd d!ZZS)$SegmentationValidatorai A class extending the DetectionValidator class for validation based on a segmentation model. Example: ```python from ultralytics.models.yolo.segment import SegmentationValidator args = dict(model="yolov8n-seg.pt", data="coco8-seg.yaml") validator = SegmentationValidator(args=args) validator() ``` Ncs>t|||||d|_d|_d|j_t|j|jd|_ dS)zVInitialize SegmentationValidator and set task to 'segment', metrics to SegmentMetrics.Nsegment)save_diron_plot) super__init__ plot_masksprocessargstaskr rrmetrics)self dataloaderrZpbarr _callbacks __class___/var/www/html/django/DPS/env/lib/python3.9/site-packages/ultralytics/models/yolo/segment/val.pyrs zSegmentationValidator.__init__cs(t|}|d|j|d<|S)zFPreprocesses batch by converting masks to float and sending to device.masks)r preprocesstodevicefloat)rbatchrrrr!'s z SegmentationValidator.preprocesscsZt|g|_|jjr"td|jjs2|jjr8tjntj |_ t ggggggd|_ dS)zOInitialize metrics and select mask processing function based on save_json flag.pycocotools>=2.0.6)tp_mtpconfpred_cls target_cls target_imgN) r init_metricsrr save_jsonrsave_txtrZprocess_mask_nativeZ 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) ClassZImagesZ InstanceszBox(PRmAP50 mAP50-95)zMask(Pr4r5r6r)rrrrget_desc7szSegmentationValidator.get_descc Csftj|d|jj|jj|jd|jjp*|jj|jj|j d}t |ddkrV|ddn|d}||fS)zIPost-processes YOLO predictions and returns output detections with proto.rT)labelsZ multi_labelZagnosticmax_detnc) rZnon_max_suppressionrr)iouZlb single_clsZ agnostic_nmsr9r:len)rpredspprotorrr postprocessGs $z!SegmentationValidator.postprocesscs<t||}|jjr|gn |d|k}|d||d<|S)LPrepares a batch for training or inference by processing images and targets. batch_idxr )r_prepare_batchr overlap_mask)rsir%Zprepared_batchZmidxrrrrGVsz$SegmentationValidator._prepare_batchcsLt||}|j||ddddf|ddddf|dd}||fS)rENZimgsz)shape)r _prepare_predr)rpredpbatchrCpredn pred_masksrrrrM]s6z#SegmentationValidator._prepare_predc Cstt|d|dD]\}\}}|jd7_t|}ttjd|jdtjd|jdtj||jtj |jdtj||jtj |jdd}| ||}| d| d} } t| } | |d<| |d <|dkr| r|j D]} |j | || q|jjr|jjd | | d q| d } |jjr:d|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|jjrx|||d|tj |!ddd"#|d|d|d|jj$r|%|||jj&|d|j'dt(|d|j)dqd S)zMetrics.rr;r#)dtyper#)r)r*r(r'clsbboxr+r,N) detections gt_bboxesgt_clsr rKr)r*r(T)r r'rSr<im_fileZ ori_shape ratio_pad)r^r8z.txt)* enumeratezipseenr@r0torchzerosr#ZniouboolrGpopuniquer1keysappendrZplotsconfusion_matrixZ process_batchr?rM_process_batchrHZ as_tensoruint8Zbatch_ircpur. pred_to_jsonrZ scale_imageZpermute contiguousnumpyr/ save_one_txt save_confrrstem)rrAr%rIrNrCZnprstatrOrTrUnlkgt_masksrPrQrrrupdate_metricscsr$            z$SegmentationValidator.update_metricscOs|j|j_|j|j_dS)z7Sets speed and confusion matrix for evaluation metrics.N)speedrri)rrkwargsrrrfinalize_metricss z&SegmentationValidator.finalize_metricsFc Cs|r|rNt|}tj||jd|ddd} ||dd}t|| kdd}|jdd|jddkrtj |d|jdddddd }| d }t ||jd d ||jd d } nt ||dddd f} | |ddd f|| S)a Compute correct prediction matrix for a batch based on bounding boxes and optional masks. Args: detections (torch.Tensor): Tensor of shape (N, 6) representing detected bounding boxes and associated confidence scores and class indices. Each row is of the format [x1, y1, x2, y2, conf, class]. gt_bboxes (torch.Tensor): Tensor of shape (M, 4) representing ground truth bounding box coordinates. Each row is of the format [x1, y1, x2, y2]. gt_cls (torch.Tensor): Tensor of shape (M,) representing ground truth class indices. pred_masks (torch.Tensor | None): Tensor representing predicted masks, if available. The shape should match the ground truth masks. gt_masks (torch.Tensor | None): Tensor of shape (M, H, W) representing ground truth masks, if available. overlap (bool): Flag indicating if overlapping masks should be considered. masks (bool): Flag indicating if the batch contains mask data. Returns: (torch.Tensor): A correct prediction matrix of shape (N, 10), where 10 represents different IoU levels. Note: - If `masks` is True, the function computes IoU between predicted and ground truth masks. - If `overlap` is True and `masks` is True, overlapping masks are taken into account when computing IoU. Example: ```python detections = torch.tensor([[25, 30, 200, 300, 0.8, 1], [50, 60, 180, 290, 0.75, 0]]) gt_bboxes = torch.tensor([[24, 29, 199, 299], [55, 65, 185, 295]]) gt_cls = torch.tensor([1, 0]) correct_preds = validator._process_batch(detections, gt_bboxes, gt_cls) ``` rRr;g?gNZbilinearF)modeZ align_cornersrg?r=rKrY)r@rbZaranger#viewrepeatwhererLFZ interpolateZgt_r r Zmatch_predictions) rrVrWrXrQrvoverlapr rtindexr>rrrrjs$ (z$SegmentationValidator._process_batchc CsNt|d|d|dd|d|d|d|jd|d |j|jd d S) z2Plots validation samples with bounding box labels.imgrFrTr=Zbboxesr r\ val_batchz _labels.jpg)r pathsfnamenamesrN)r Zsqueezerrr)rr%nirrrplot_val_sampless z&SegmentationValidator.plot_val_samplescCspt|dgt|dddt|jr4tj|jddn|jR|d|jd|d|j|jd |j d S) z6Plots batch predictions with masks and bounding boxes.rrr[)r9)dimr\rz _pred.jpg)rrrrN) r r r@rrbcatrrrclear)rr%rArrrrplot_predictionss z&SegmentationValidator.plot_predictionscCsVddlm}|tj|d|dftjdd|j|ddddf|dj||ddS) zRSave YOLO detections to a txt file in normalized coordinates in a specific format.r)Resultsr;rZNrJ)pathrZboxesr )rq)Zultralytics.engine.resultsrnprcrkrr/)rrPrQrqrLfilerrrrrps z"SegmentationValidator.save_one_txtc s&ddlmfdd}t|j}|r2t|n|}t|ddddf}|ddddf|ddddfd8<t |d}t t }| ||} Wdn1s0Yt t||D]J\} \} } |j||jt| d d d | Dt| dd | | d qdS) z Save one JSON result. Examples: >>> result = {"image_id": 42, "category_id": 18, "bbox": [258.15, 41.29, 348.26, 243.78], "score": 0.236} rencodecs@tj|dddddfdddd}|dd|d<|S)z:Encode predicted masks as RLE and append results to jdict.Nrrk)orderrSrcountszutf-8)rZasarraydecode)xZrlerrr single_encode s*z9SegmentationValidator.pred_to_json..single_encodeNrKr])r]rr;rYcSsg|]}t|dqS)r<)round.0rrrr z6SegmentationValidator.pred_to_json..)image_idZ category_idrUZscoreZ segmentation)Zpycocotools.maskrrrr isnumericintrZ xyxy2xywhrZ transposerrmapr_r`tolistjdictrhZ class_mapr) rrPfilenamerQrrrrboxpoolZrlesirBbrrrrms$   0  *"  z"SegmentationValidator.pred_to_jsonc Cs|jjr~|jr~t|jr~|jdd}|jd}td|d|dzt ddd l m }dd l m }||fD]}|sxJ|d qx|t|}|t|}t|||d |||d gD]x\} } |jrdd|jjjD| j_| | | | dd} | jdd\||jj| d<||jj| <qWn6ty|} ztd| WYd} ~ n d} ~ 00|S)z6Return COCO-style object detection evaluation metrics.rz"annotations/instances_val2017.jsonzpredictions.jsonz" Evaluating pycocotools mAP using z and z...r&r)COCO)COCOevalz file not foundrUZsegmcSsg|]}tt|jqSr)rrrrrrrrr4rz3SegmentationValidator.eval_json..rKr]Nr;zpycocotools unable to run: )rr.Zis_cocor@rdatarrinforZpycocotools.cocorZpycocotools.cocoevalris_filestrZloadResr_rZdatasetZim_filesparamsZimgIdsevaluate accumulateZ summarizer1rrg Exceptionwarning) rr1Z anno_jsonZ pred_jsonrrrannorNrevalidxerrr eval_json#s2     $ *&zSegmentationValidator.eval_json)NNNNN)NNFF)__name__ __module__ __qualname____doc__rr!r-r7rDrGrMrwrzrjrrrprmr __classcell__rrrrrs      E .  !r)Zmultiprocessing.poolrpathlibrrorrbZtorch.nn.functionalnnZ functionalrZultralytics.models.yolo.detectrZultralytics.utilsrrrZultralytics.utils.checksrZultralytics.utils.metricsr r r Zultralytics.utils.plottingr r rrrrrs