hVSSKJr SSKrSSKJr SSKJr SSKrSSK r SSK J r SSK JrJrJr SSKJr SSKJrJrJrJr SSKJr SS KJrJrJr SS KJr "S S \5r g) ) annotationsN)Path)Any)build_dataloaderbuild_yolo_dataset converter) BaseValidator)LOGGERRANKnmsops)check_requirements)ConfusionMatrix DetMetricsbox_iou) plot_imagesc@^\rSrSrSrSSU4SjjjrSSjrSSjrSSjrSSjr S Sjr S!S jr S"S jr SS jr SS jrS#S jrSSjrS$SjrS%S&SjjrS'SjrS(SjrS)S*SjjrS+SjrS,SjrS-SjrS.SjrS/S0SjjrSrU=r$)1DetectionValidatora2 A class extending the BaseValidator class for validation based on a detection model. This class implements validation functionality specific to object detection tasks, including metrics calculation, prediction processing, and visualization of results. Attributes: is_coco (bool): Whether the dataset is COCO. is_lvis (bool): Whether the dataset is LVIS. class_map (list[int]): Mapping from model class indices to dataset class indices. metrics (DetMetrics): Object detection metrics calculator. iouv (torch.Tensor): IoU thresholds for mAP calculation. niou (int): Number of IoU thresholds. lb (list[Any]): List for storing ground truth labels for hybrid saving. jdict (list[dict[str, Any]]): List for storing JSON detection results. stats (dict[str, list[torch.Tensor]]): Dictionary for storing statistics during validation. Examples: >>> from ultralytics.models.yolo.detect import DetectionValidator >>> args = dict(model="yolo11n.pt", data="coco8.yaml") >>> validator = DetectionValidator(args=args) >>> validator() c>[TU]XX45 SUlSUlSUlSUR l[R"SSS5Ul URR5Ul [5Ul g)af Initialize detection validator with necessary variables and settings. Args: dataloader (torch.utils.data.DataLoader, optional): Dataloader to use for validation. save_dir (Path, optional): Directory to save results. args (dict[str, Any], optional): Arguments for the validator. _callbacks (list[Any], optional): List of callback functions. FNdetectg?gffffff? )super__init__is_cocois_lvis class_mapargstasktorchlinspaceiouvnumelniourmetrics)self dataloadersave_dirr _callbacks __class__s \/home/james-whalen/.local/lib/python3.13/site-packages/ultralytics/models/yolo/detect/val.pyrDetectionValidator.__init__.sc t@  ! NN3b1 IIOO% !| cnUR5HYup#[U[R5(dM&UR UR UR R S:HS9X'M[ URR(aUSR5OUSR5S- US'U$)z Preprocess batch of images for YOLO validation. Args: batch (dict[str, Any]): Batch containing images and annotations. Returns: (dict[str, Any]): Preprocessed batch. cuda) non_blockingimg) items isinstancer Tensortodevicetyperhalffloat)r&batchkvs r+ preprocessDetectionValidator.preprocessAsKKMDA!U\\**44 $++:J:Jf:T4U"04yy~~e ))+5 yy""!!,7yy!!!!! II'';'MMH,$vi7H2I2N2N1Ot/TT ?)r-cJURR(a7SH1nURRURXR S9 M3 UR URlURURlURURlg)z8Set final values for metrics speed and confusion matrix.)TF)r( normalizeon_plotN)rrZr\plotr(rspeedr%)r&rs r+finalize_metrics#DetectionValidator.finalize_metricssl 99??( %%**DMMY`l`l*m)!ZZ (,(=(= % $  r-c[S:XaS/[R"5-n[R"URR USS9 URR R 5Vs0sHo"/_M nnUH/nUR 5HnX2RXB5 M M1 X0Rl[URR5Ul g[S:aE[R"URR SSS9 URR5 ggs snf)zGather stats from all GPUs.rN)dst) r distget_world_size gather_objectr%statskeysextendrRr'datasetrX clear_stats)r&gathered_statskey merged_stats stats_dicts r+ gather_statsDetectionValidator.gather_statss 19"Vd&9&9&;;N   t||11>q I/3||/A/A/F/F/HI/HG/HLI, ',,.C %,,Z_=/-".LL DOO334DI AX   t||114Q ? LL $ $ & Js4 D<cURRURURRUR S9 URR 5 URR$)zq Calculate and return metrics statistics. Returns: (dict[str, Any]): Dictionary containing metrics results. )r(rr)r%processr(rrZrr results_dictrgs r+ get_statsDetectionValidator.get_statssP dmm$))//SWS_S_`   "||(((r-cSS[URR5--n[R"USUR URR R5/URR5Q7-5 URR R5S:Xa.[R"SURRS35 URR(aUR(dURS:a[URR5(a[!URR"5Hxup#[R"UUR$UURR&UURR U/URR)U5Q7-5 Mz ggggg) z0Print training/validation set metrics per class.z %22s%11i%11iz%11.3gallrzno labels found in z, set, can not compute metrics without labelsrDN)rRr%rr inforX nt_per_classsum mean_resultswarningrrverboserUrVrrap_class_indexrF nt_per_image class_result)r&pfics r+ print_results DetectionValidator.print_resultssY 8c$,,2C2C.D#D D B%DLL,E,E,I,I,KjdllNgNgNijjk << $ $ ( ( *a / NN00@@lm n 99  T]]tww{s4<  1  11!4 11!42215 ?H_{] r-cZUSRSS:XdUSRSS:Xa7S[R"USRSUR4[S90$[ USUS5nSUR USUSU5R5R50$)a Return correct prediction matrix. Args: preds (dict[str, torch.Tensor]): Dictionary containing prediction data with 'bboxes' and 'cls' keys. batch (dict[str, Any]): Batch dictionary containing ground truth data with 'bboxes' and 'cls' keys. Returns: (dict[str, np.ndarray]): Dictionary containing 'tp' key with correct prediction matrix of shape (N, 10) for 10 IoU levels. rurtpdtypers) rrrr$boolrmatch_predictionsrr)r&r{r;rxs r+r!DetectionValidator._process_batchs <  a A %u););A)>!)C"((E%L$6$6q$9499#ETRS SeHouX7d,,U5\5<MQQSYY[\\r-c V[URXURX RS9$)a, Build YOLO Dataset. Args: img_path (str): Path to the folder containing images. mode (str): `train` mode or `val` mode, users are able to customize different augmentations for each mode. batch (int, optional): Size of batches, this is for `rect`. Returns: (Dataset): YOLO dataset. )modestride)rrrHr)r&img_pathrr;s r+ build_dataset DetectionValidator.build_dataset$s#"$))Xdiid[f[fggr-c URXSS9n[UUURRSSURRUR S9$)z Construct and return dataloader. Args: dataset_path (str): Path to the dataset. batch_size (int): Size of each batch. Returns: (torch.utils.data.DataLoader): Dataloader for validation. rT)r;rFr)shufflerank drop_last pin_memory)rrrworkerscompilerU)r& dataset_path batch_sizers r+get_dataloader!DetectionValidator.get_dataloader2sS$$\%$P   II  ii''}}  r-cn[UUSURSUS3- URURS9 g)z Plot validation image samples. Args: batch (dict[str, Any]): Batch containing images and annotations. ni (int): Batch index. r val_batchz _labels.jpg)rpathsfnamerFrN)rr(rFr)r&r;nis r+plot_val_samples#DetectionValidator.plot_val_samplesHs:  "--IbT"==**LL  r-c ,[U5H$upV[R"US5U-US'M& USR5nU=(d URR nUVV s0sH.o[R "UV s/sH oUSUPM sn SS9_M0 n nn [R"U SSS2SS245U SSS2SS24'[USU US URS US 3- URURS 9 gs sn fs sn nf) a= Plot predicted bounding boxes on input images and save the result. Args: batch (dict[str, Any]): Batch containing images and annotations. preds (list[dict[str, torch.Tensor]]): List of predictions from the model. ni (int): Batch index. max_det (Optional[int]): Maximum number of detections to plot. rtrrNdimrsrpr1rrz _pred.jpg)imagesrrrrFr) rr ones_likerrrncatr xyxy2xywhrr(rFr) r&r;r{rrnrrrr<r} batched_predss r+plot_predictions#DetectionValidator.plot_predictionsXs !'GA %V = AD (Qx}}.TYY..W[\W[RSEIIu&Eu!tHW~u&E1MMW[ \),}X7NqRTSTRTu7U)V h2A2&<  "--IbT";;**LL  'F\s+DD  D Dc SSKJn U"[R"USUS4[RS9SUR [ R"USUSRS5US RS5/SS 9S 9RXBS 9 g) a Save YOLO detections to a txt file in normalized coordinates in a specific format. Args: predn (dict[str, torch.Tensor]): Dictionary containing predictions with keys 'bboxes', 'conf', and 'cls'. save_conf (bool): Whether to save confidence scores. shape (tuple[int, int]): Shape of the original image (height, width). file (Path): File path to save the detections. r)ResultsrDrNrsrtrrur)pathrFboxes)r) ultralytics.engine.resultsrrruint8rFr r unsqueezer)r&rrrfilers r+rDetectionValidator.save_one_txtus 7 HHeAha) :**))U8_eFm.E.Eb.I5QV>> result = { ... "image_id": 42, ... "file_name": "42.jpg", ... "category_id": 18, ... "bbox": [258.15, 41.29, 348.26, 243.78], ... "score": 0.236, ... } rrsNrrtrurq)image_id file_name category_idrscore) rr isnumericintr rziptolistrYappendnamerround) r&rrrrrboxbsrr}s r+rDetectionValidator.pred_to_jsons$F9%&yy $ 0 03t9dmmE(O, ArrE c!QR%j1n$ 3::<v)=)=)?uATATAVWGA! JJ   (!%#'>>#aΛ4!QU1a[!4"1a[  X 5s4D#c p0UES[R"USUSR5USUSS90E$)z.Scales predictions to the original image size.rsrrr)r)r scale_boxesclone)r&rrs r+rDetectionValidator.scale_predssN  coowh%%'{# -   r-cURS- nURSS- UR(aSOSURRS3- nUR XU5$)z Evaluate YOLO output in JSON format and return performance statistics. Args: stats (dict[str, Any]): Current statistics dictionary. Returns: (dict[str, Any]): Updated statistics dictionary with COCO/LVIS evaluation results. zpredictions.jsonrrzinstances_val2017.jsonlvis_v1_z.json)r(rHrrrJ coco_evaluate)r&r pred_json anno_jsons r+ eval_jsonDetectionValidator.eval_jsonsfMM$66 IIf  +/<<'x GXX]=^ `  !!%I>>r-c bURR(Ga\UR(dUR(Ga9[ UR 5(Ga[ R"SUSUS35 X#4H"nUR5(aMUS35e [U[5(aU/OUn[U[5(aU/OUn[S5 SSK J nJn U"U5n U RU5n [!U5GHLupU"XXR[ RS9n UR"R$R&Vs/sH!n[)[+U5R,5PM# snU R.lU R35 U R55 U R75 U R8S US X[SS 3'U R8S US X[SS 3'UR(dMU R8SUSX[SS 3'U R8SUSX[SS 3'U R8SUSX[SS 3'GMO UR(aUSUS'U$U$s snf![:a$n[ R<"SU35 SnAU$SnAff=f)a Evaluate COCO/LVIS metrics using faster-coco-eval library. Performs evaluation using the faster-coco-eval library to compute mAP metrics for object detection. Updates the provided stats dictionary with computed metrics including mAP50, mAP50-95, and LVIS-specific metrics if applicable. Args: stats (dict[str, Any]): Dictionary to store computed metrics and statistics. pred_json (str | Path]): Path to JSON file containing predictions in COCO format. anno_json (str | Path]): Path to JSON file containing ground truth annotations in COCO format. iou_types (str | list[str]]): IoU type(s) for evaluation. Can be single string or list of strings. Common values include "bbox", "segm", "keypoints". Defaults to "bbox". suffix (str | list[str]]): Suffix to append to metric names in stats dictionary. Should correspond to iou_types if multiple types provided. Defaults to "Box". Returns: (dict[str, Any]): Updated stats dictionary containing the computed COCO/LVIS evaluation metrics. z' Evaluating faster-coco-eval mAP using z and z...z file not foundzfaster-coco-eval>=1.6.7r)COCOCOCOeval_faster)iouType lvis_styleprint_functionAP_50zmetrics/mAP50()AP_allzmetrics/mAP50-95(APrz metrics/APr(APcz metrics/APc(APfz metrics/APf(zmetrics/mAP50-95(B)fitnessz faster-coco-eval unable to run: N)rrSrrrRrYr ris_filer4rKrfaster_coco_evalr+r,loadResrr'rim_filesrrrparamsimgIdsevaluate accumulate summarize stats_as_dict Exceptionr)r&rr&r' iou_typessuffixr}r+r,annorriou_typerTes r+r% DetectionValidator.coco_evaluatesp6 99   DLLDLLLc$**oo KKB9+US\R]]`a b G"-A99;;=1#_(==;.+5i+E+EYK9 %/%<%<&&"#<=BI||I.#,Y#7KA)H^d^i^iCEIOOD[D[DdDd(eDdqT!W\\):Dd(eCJJ%LLNNN$MMO?B>O>OPW>XEN69Q<.:;ADARARS[A\E-fil^1=>|||@C@Q@QRW@X VYq\N!<=@C@Q@QRW@X VYq\N!<=@C@Q@QRW@X VYq\N!<=!$8$<<',-B'CE)$ u %)f  G!A!EFF  Gs99JB?J(I;=BJA/J;J J. J))J.) rr\rEr"rrrYr%rFrVr$rX)NNNN)returnNone)r;dict[str, Any]rHrJ)r]ztorch.nn.ModulerHrI)rHrK)r{z torch.TensorrHlist[dict[str, torch.Tensor]])rrr;rJrHrJ)rdict[str, torch.Tensor]rHrL)r{rKr;rJrHrI)rHrJ)r{rLr;rJrHzdict[str, np.ndarray])rTN)rrKrrKr; int | NonerHztorch.utils.data.Dataset)rrKrrrHztorch.utils.data.DataLoader)r;rJrrrHrI)N) r;rJr{rKrrrnrMrHrI) rrLrrrztuple[int, int]r rrHrI)rrLrrJrHrI)rrLrrJrHrL)rrJrHrJ)rBox) rrJr&rKr'rKrBstr | list[str]rCrOrHrJ)__name__ __module__ __qualname____firstlineno____doc__rr>r^rhr~rrrrrrrrrrrrrrrr(r%__static_attributes__ __classcell__)r*s@r+rrs0$$& y0lk0 8 ,\. ' )(] h , "ko # ,I OR ]g  :.& D  ?.&,"' === = # =  = ==r-r)! __future__rrMpathlibrtypingrrrr torch.distributed distributedrultralytics.datarrrultralytics.engine.validatorr ultralytics.utilsr r r r ultralytics.utils.checksrultralytics.utils.metricsrrrultralytics.utils.plottingrrrfr-r+rbsE#  LL6447JJ2ppr-