hXSrSSKJr SSKrSSKrSSKrSSKJr SSKJ r SSK r SSK r SSK r SSKJrJr SSKJr SSKJr SS KJr SS KJrJrJrJrJrJrJr SS KJ r J!r! SS K"J#r# SS K$J%r%J&r&J'r' Sr("SS5r)g)ar Run prediction on images, videos, directories, globs, YouTube, webcam, streams, etc. Usage - sources: $ yolo mode=predict model=yolo11n.pt source=0 # webcam img.jpg # image vid.mp4 # video screen # screenshot path/ # directory list.txt # list of images list.streams # list of streams 'path/*.jpg' # glob 'https://youtu.be/LNwODJXcvt4' # YouTube 'rtsp://example.com/media.mp4' # RTSP, RTMP, HTTP, TCP stream Usage - formats: $ yolo mode=predict model=yolo11n.pt # PyTorch yolo11n.torchscript # TorchScript yolo11n.onnx # ONNX Runtime or OpenCV DNN with dnn=True yolo11n_openvino_model # OpenVINO yolo11n.engine # TensorRT yolo11n.mlpackage # CoreML (macOS-only) yolo11n_saved_model # TensorFlow SavedModel yolo11n.pb # TensorFlow GraphDef yolo11n.tflite # TensorFlow Lite yolo11n_edgetpu.tflite # TensorFlow Edge TPU yolo11n_paddle_model # PaddlePaddle yolo11n.mnn # MNN yolo11n_ncnn_model # NCNN yolo11n_imx_model # Sony IMX yolo11n_rknn_model # Rockchip RKNN ) annotationsN)Path)Any)get_cfg get_save_dir)load_inference_source) LetterBox) AutoBackend) DEFAULT_CFGLOGGERMACOSWINDOWS callbackscolorstrops) check_imgsz check_imshow)increment_path)attempt_compile select_devicesmart_inference_modea inference results will accumulate in RAM unless `stream=True` is passed, causing potential out-of-memory errors for large sources or long-running streams and videos. See https://docs.ultralytics.com/modes/predict/ for help. Example: results = model(source=..., stream=True) # generator of Results objects for r in results: boxes = r.boxes # Boxes object for bbox outputs masks = r.masks # Masks object for segment masks outputs probs = r.probs # Class probabilities for classification outputs c\rSrSrSr\SS4SSjjrSSjrSSjrSSjr Sr SSS jjr SS jr S r \"5SS j5rSSS jjrSSjrSSSjjrS S!SjjrS"SjrS#SjrSrg)$ BasePredictorEa| A base class for creating predictors. This class provides the foundation for prediction functionality, handling model setup, inference, and result processing across various input sources. Attributes: args (SimpleNamespace): Configuration for the predictor. save_dir (Path): Directory to save results. done_warmup (bool): Whether the predictor has finished setup. model (torch.nn.Module): Model used for prediction. data (dict): Data configuration. device (torch.device): Device used for prediction. dataset (Dataset): Dataset used for prediction. vid_writer (dict[str, cv2.VideoWriter]): Dictionary of {save_path: video_writer} for saving video output. plotted_img (np.ndarray): Last plotted image. source_type (SimpleNamespace): Type of input source. seen (int): Number of images processed. windows (list[str]): List of window names for visualization. batch (tuple): Current batch data. results (list[Any]): Current batch results. transforms (callable): Image transforms for classification. callbacks (dict[str, list[callable]]): Callback functions for different events. txt_path (Path): Path to save text results. _lock (threading.Lock): Lock for thread-safe inference. Methods: preprocess: Prepare input image before inference. inference: Run inference on a given image. postprocess: Process raw predictions into structured results. predict_cli: Run prediction for command line interface. setup_source: Set up input source and inference mode. stream_inference: Stream inference on input source. setup_model: Initialize and configure the model. write_results: Write inference results to files. save_predicted_images: Save prediction visualizations. show: Display results in a window. run_callbacks: Execute registered callbacks for an event. add_callback: Register a new callback function. Nc[X5Ul[UR5UlURRcSURlSUlURR (a[SS9URlSUlURRUl SUl SUl SUl 0Ul SUlSUlSUl/UlSUlSUlSUlU=(d [*R,"5UlSUl[0R2"5Ul[*R6"U5 g)z Initialize the BasePredictor class. Args: cfg (str | dict): Path to a configuration file or a configuration dictionary. overrides (dict, optional): Configuration overrides. _callbacks (dict, optional): Dictionary of callback functions. Ng?FT)warnr)rargsrsave_dirconf done_warmupshowrmodeldataimgszdevicedataset vid_writer plotted_img source_typeseenwindowsbatchresults transformsrget_default_callbackstxt_path threadingLock_lockadd_integration_callbacks)selfcfg overrides _callbackss V/home/james-whalen/.local/lib/python3.13/site-packages/ultralytics/engine/predictor.py__init__BasePredictor.__init__osC+ $TYY/ 99>> !!DIIN  99>>)t4DIIN IINN        #Hy'F'F'H ^^% ++D1c[U[R5(+nU(a[R"UR U55nUR SS:Xa USSSS24nURS5n[R"U5n[R"U5nURUR5nURR(aUR5OUR5nU(aUS-nU$)z Prepare input image before inference. Args: im (torch.Tensor | list[np.ndarray]): Images of shape (N, 3, H, W) for tensor, [(H, W, 3) x N] for list. Returns: (torch.Tensor): Preprocessed image tensor of shape (N, 3, H, W). .N)rr?) isinstancetorchTensornpstack pre_transformshape transposeascontiguousarray from_numpytor%r"fp16halffloat)r5im not_tensors r9 preprocessBasePredictor.preprocesss$B 55 $,,R01Bxx|q TrT ]l+B%%b)B!!"%B UU4;; **//RWWYrxxz  #IB r<chURR(aUURR(d:[ UR [ URSS5R- SS9OSnUR"U/UQ7URRX@RRS.UD6$)zGRun inference on a given image using the specified model and arguments.rT)mkdirF)augment visualizeembed) rrXr)tensorrrrr,stemr"rWrY)r5rQrkwargsrXs r9 inferenceBasePredictor.inferencesyy""D,<,<,C,C 4==4 1 a0@+A+F+FFd S  zz"ufjudii&7&79T]T]TcTcuntuur<c[UVs1sHo"RiM sn5S:Hn[URU=(as URR =(aV UR R=(d9 [UR SS5=(a UR R(+UR RS9nUVs/sHo$"US9PM sn$s snfs snf)z Pre-transform input image before inference. Args: im (list[np.ndarray]): List of images with shape [(H, W, 3) x N]. Returns: (list[np.ndarray]): List of transformed images. r@dynamicF)autostride)image) lenrIr r$rrectr"ptgetattrimxrb)r5rQx same_shapes letterboxs r9rHBasePredictor.pre_transformsB/Bq77B/0A5  JJb b`74::y%#H#_QUQ[Q[Q_Q_M_::$$  -//Bq "B//00s C"C'cU$)z6Post-process predictions for an image and return them.)r5predsimg orig_imgss r9 postprocessBasePredictor.postprocesss r<cX0lU(aUR"X/UQ70UD6$[UR"X/UQ70UD65$)a_ Perform inference on an image or stream. Args: source (str | Path | list[str] | list[Path] | list[np.ndarray] | np.ndarray | torch.Tensor, optional): Source for inference. model (str | Path | torch.nn.Module, optional): Model for inference. stream (bool): Whether to stream the inference results. If True, returns a generator. *args (Any): Additional arguments for the inference method. **kwargs (Any): Additional keyword arguments for the inference method. Returns: (list[ultralytics.engine.results.Results] | generator): Results objects or generator of Results objects. )streamstream_inferencelist)r5sourcer"rurr\s r9__call__BasePredictor.__call__sG ((HHH H--fMdMfMN Nr<c8URX5nUHnM g)a Method used for Command Line Interface (CLI) prediction. This function is designed to run predictions using the CLI. It sets up the source and model, then processes the inputs in a streaming manner. This method ensures that no outputs accumulate in memory by consuming the generator without storing results. Args: source (str | Path | list[str] | list[Path] | list[np.ndarray] | np.ndarray | torch.Tensor, optional): Source for inference. model (str | Path | torch.nn.Module, optional): Model for inference. Note: Do not modify this function or remove the generator. The generator ensures that no outputs are accumulated in memory, which is critical for preventing memory issues during long-running predictions. N)rv)r5rxr"gen_s r9 predict_cliBasePredictor.predict_clis!"##F2A r<c [URRURRSS9Ul[ UURR URRURR[URSS5S9Ul URRUl URR=(d] URR=(d@ [UR5S:=(d! [[URSS/55nU(a0S S Kn[US S 5(d["R$"[&5 0Ulg ) z Set up source and inference mode. Args: source (str | Path | list[str] | list[Path] | list[np.ndarray] | np.ndarray | torch.Tensor): Source for inference. rA)rbmin_dimchr?)rxr, vid_stridebufferchannelsi video_flagFrNruT)rrr$r"rbrr,r stream_bufferrgr&r)ru screenshotrdany torchvisionr warningSTREAM_WARNINGr')r5rx long_sequencers r9 setup_sourceBasePredictor.setup_sources!9J9JTUV ,))//yy++99**TZZq1    <<33    # # A** A4<< 4' A74<<w?@   4400~.r<c /^# TRR(a[R"S5 TR(dTR U5 TR  TRUbUOTRR5 TRR(dTRR(aDTRR(aTRS- O TRRSSS9 TR(dTRRTRR(dTRR (aSOTR"R$TRR&/TR(Q7S9 STl S/SsTlTlTl[0R2"TR4S 9[0R2"TR4S 9[0R2"TR4S 94nTR7S 5 TR"GHxTlTR7S 5 TR.upgnUS TR9U5n SSS5 US TR:"W /UQ70UD6n TRR<(a7[?U [@RB5(aU /OU ShvN SSS5 MSSS5 US  TREW W U5Tl#SSS5 TR7S 5 [IU5n [KU 5GHn T=R*S- slUSRLS-U - USRLS-U - US RLS-U - S.TRFU l'TRR(dSTRR(d8TRR(dTRRP(dMX==TRSU [UXl5W U5- ss'GM TRR(a%[R"SRYU55 TR7S5 TRFShvN GM{ SSS5 TRZR]5H4n [?U [^R`5(dM$U Rc5 M6 TRRP(a[^Rd"5 TRR(aTR*(a[gU4SjW55n[R"S[iTRR.TR*5[kTRSS5/W RlS SQ73U-5 TRR(d6TRR(dTRRn(a[I[qTRRsS555nTRR(aSUSSUS:-STRS- 3OSn[R"S[uSTR5U35 TR7S5 g!,(df  GNt=fGN!,(df  GN=f!,(df  GN=f![Va  GMxf=fGN!,(df  GN=f7f)a Stream real-time inference on camera feed and save results to file. Args: source (str | Path | list[str] | list[Path] | list[np.ndarray] | np.ndarray | torch.Tensor, optional): Source for inference. model (str | Path | torch.nn.Module, optional): Model for inference. *args (Any): Additional arguments for the inference method. **kwargs (Any): Additional keyword arguments for the inference method. Yields: (ultralytics.engine.results.Results): Results objects. NlabelsTparentsexist_okr@)r$r)r%on_predict_starton_predict_batch_startrAon_predict_postprocess_end@@)rSr]rr on_predict_batch_endc3Z># UH oRTR- S-v M" g7f)rN)tr*).0rir5s r9 1BasePredictor.stream_inference..zs ?YccDIIo+Ys(+zRSpeed: %.1fms preprocess, %.1fms inference, %.1fms postprocess per image at shape rr?z labels/*.txtz labelsz saved to zResults saved to boldon_predict_end);rverboser infor" setup_modelr3rrxsavesave_txtrrVr warmuprftritonr&bsrr$r*r+r,rProfiler% run_callbacksrSr]rYrCrDrErrr-rdrangedtspeedr! write_resultsr StopIterationjoinr'valuescv2 VideoWriterreleasedestroyAllWindowstupleminrgrI save_croprwglobr)r5rxr"rr\ profilerspathsim0srrQronivrnls` r9rvBasePredictor.stream_inferences$ 99   KKOzz   U # ZZ   (:f @P@P Qyy~~!3!3-1YY-?-?)T]]YYbfquYv## !! $ 1B1B1 Y]YcYcYfYftimisist"$( 23R /DIt|TZ 4;;/ 4;;/ 4;;/I   1 2"ll ""#;)B+4Q>  ! ! # 99  ?Y??A KKd 3WTZZq5QaTVT\T\]^]_T`abdfgh  99>>TYY//4993F3FT$--,,^<=>BW[W`W`WiWi"RDsb1f~.j9Q8RSoqA KK+HVT]],K+LQCP Q +,q"\Y"\"\ %(}ZsA ]#G?\0"[!4 \0A[6[3 [6 \0+ \08\$\06C\.\A \0'\-( \01A]7G*]! [0 +\03[66 \ \0 \ \0 \*$\0)\**\00 \?:]c [U=(d URR[URRUS9URR URR URRSUS9UlURRUlURRURl[URS5(aA[URSS5(d%URRURl URR5 [URURURRS9Ulg) z Initialize YOLO model with given parameters and set it to evaluation mode. Args: model (str | Path | torch.nn.Module, optional): Model to load or use. verbose (bool): Whether to print verbose output. )rT)r"r%dnnr#rNfuserr$r`F)r%modeN)r rr"rr%rr#rOrNhasattrrgr$evalrcompile)r5r"rs r9rBasePredictor.setup_models!*499?? !1!17C   jj''   4::w ' ' Iu0U0U"jj..DIIO $TZZ $))J[J[\ r<cSn[UR5S:XaUSnURR(d6URR(dURR (aXQS3- nUR RnO0[R"SXA5nU(a[US5OSnURS- URUR RS:XaSOS U3-- UlUS R"URS S6- nUR UnURR#5Ul XXR%5UR&S S S3- nUR(R*(dUR(R,(aUR/UR(R0UR(R2UR(R4UR(R6UR(R8(aSOX1S9UlUR(R<(a1UR=URS3UR(R>S9 UR(R@(a1URAURS- URRS9 UR(R,(aUR-[CU55 UR(R*(a)UREURURF- U5 U$)a) Write inference results to a file or directory. Args: i (int): Index of the current image in the batch. p (Path): Path to the current image. im (torch.Tensor): Preprocessed image tensor. s (list[str]): List of result strings. Returns: (str): String with result information. rr?Nz: z frame (\d+)/r@rrcr}z {:g}x{:g} rAr]z.1fms) line_widthboxesrrim_gpuz.txt) save_confcrops)r file_name)$rdrIr)rufrom_imgrZr&countresearchintrr[rr0formatr-__str__rrrrr!plotr show_boxes show_conf show_labels retina_masksr(rrrstrsave_predicted_imagesname) r5rprQrstringframematchresults r9rBasePredictor.write_resultssS rxx=A DB    " "d&6&6&?&?4CSCSCZCZ 2h FLL&&EIIoqt4E%*CaME 0AFFDLLDUDUY`D`bhijoipfq4rs ,%%rxx|44a--//1^^%&v||K'@&ERHH 99>>TYY^^%{{99//ii**YY((yy,,#yy55t25 + D  99   OOt}}oT2dii>Q>QO R 99     dmmg&=I[I[  \ 99>> IIc!f  99>>  & &t}}qvv'=u E r<cURnURRS;GaURRS:XaURROSnURUR S3- nXR ;aURR(a[U5RSSS9 [(aSO[(aSOS upg[R"[[U5R!U55[R""U6UUR$S UR$S 4S 9UR U'UR UR'U5 URR(a+[R("US UR SUS3U5 gg[R("[UR!S55U5 g)z Save video predictions as mp4 or images as jpg at specified path. Args: save_path (Path): Path to save the results. frame (int): Frame number for video mode. >videorur_framesTr)z.mp4avc1).aviWMV2)rMJPGr@r)filenamefourccfps frameSize/r}z.jpgN)r(r&rrrr[r'r save_framesrrVr rrrr with_suffixVideoWriter_fourccrIwriteimwrite)r5 save_pathrrQr frames_pathsuffixrs r9r#BasePredictor.save_predicted_imagesso   <<   3 3&*ll&7&77&B$,,""C--Y^^,r.s\B#   12.2[[[>2^^ @+@+r<