修改\ultralytics-main\ultralytics\engine\predictor.py，即可

# Ultralytics YOLO 🚀, AGPL-3.0 license
"""
Run prediction on images, videos, directories, globs, YouTube, webcam, streams, etc.Usage - sources:$ yolo mode=predict model=yolov8n.pt source=0                               # webcamimg.jpg                         # imagevid.mp4                         # videoscreen                          # screenshotpath/                           # directorylist.txt                        # list of imageslist.streams                    # list of streams'path/*.jpg'                    # glob'https://youtu.be/LNwODJXcvt4'  # YouTube'rtsp://example.com/media.mp4'  # RTSP, RTMP, HTTP, TCP streamUsage - formats:$ yolo mode=predict model=yolov8n.pt                 # PyTorchyolov8n.torchscript        # TorchScriptyolov8n.onnx               # ONNX Runtime or OpenCV DNN with dnn=Trueyolov8n_openvino_model     # OpenVINOyolov8n.engine             # TensorRTyolov8n.mlpackage          # CoreML (macOS-only)yolov8n_saved_model        # TensorFlow SavedModelyolov8n.pb                 # TensorFlow GraphDefyolov8n.tflite             # TensorFlow Liteyolov8n_edgetpu.tflite     # TensorFlow Edge TPUyolov8n_paddle_model       # PaddlePaddleyolov8n_ncnn_model         # NCNN
"""import platform
import re
import threading
from pathlib import Pathimport cv2
import numpy as np
import torchfrom ultralytics.cfg import get_cfg, get_save_dir
from ultralytics.data import load_inference_source
from ultralytics.data.augment import LetterBox, classify_transforms
from ultralytics.nn.autobackend import AutoBackend
from ultralytics.utils import DEFAULT_CFG, LOGGER, MACOS, WINDOWS, callbacks, colorstr, ops
from ultralytics.utils.checks import check_imgsz, check_imshow
from ultralytics.utils.files import increment_path
from ultralytics.utils.torch_utils import select_device, smart_inference_modeSTREAM_WARNING = """
WARNING ?? 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 objectsfor r in results:boxes = r.boxes  # Boxes object for bbox outputsmasks = r.masks  # Masks object for segment masks outputsprobs = r.probs  # Class probabilities for classification outputs
"""class BasePredictor:"""BasePredictor.A base class for creating predictors.Attributes:args (SimpleNamespace): Configuration for the predictor.save_dir (Path): Directory to save results.done_warmup (bool): Whether the predictor has finished setup.model (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): Dictionary of {save_path: video_writer, ...} writer for saving video output."""def __init__(self, cfg=DEFAULT_CFG, overrides=None, _callbacks=None):"""Initializes the BasePredictor class.Args:cfg (str, optional): Path to a configuration file. Defaults to DEFAULT_CFG.overrides (dict, optional): Configuration overrides. Defaults to None."""self.args = get_cfg(cfg, overrides)self.save_dir = get_save_dir(self.args)if self.args.conf is None:self.args.conf = 0.25  # default conf=0.25self.done_warmup = Falseif self.args.show:self.args.show = check_imshow(warn=True)# Usable if setup is doneself.model = Noneself.data = self.args.data  # data_dictself.imgsz = Noneself.device = Noneself.dataset = Noneself.vid_writer = {}  # dict of {save_path: video_writer, ...}self.plotted_img = Noneself.source_type = Noneself.seen = 0self.windows = []self.batch = Noneself.results = Noneself.transforms = Noneself.callbacks = _callbacks or callbacks.get_default_callbacks()self.txt_path = Noneself._lock = threading.Lock()  # for automatic thread-safe inferencecallbacks.add_integration_callbacks(self)def preprocess(self, im):"""Prepares input image before inference.Args:im (torch.Tensor | List(np.ndarray)): BCHW for tensor, [(HWC) x B] for list."""not_tensor = not isinstance(im, torch.Tensor)if not_tensor:im = np.stack(self.pre_transform(im))im = im[..., ::-1].transpose((0, 3, 1, 2))  # BGR to RGB, BHWC to BCHW, (n, 3, h, w)im = np.ascontiguousarray(im)  # contiguousim = torch.from_numpy(im)im = im.to(self.device)im = im.half() if self.model.fp16 else im.float()  # uint8 to fp16/32if not_tensor:im /= 255  # 0 - 255 to 0.0 - 1.0return imdef inference(self, im, *args, **kwargs):"""Runs inference on a given image using the specified model and arguments."""visualize = (increment_path(self.save_dir / Path(self.batch[0][0]).stem, mkdir=True)if self.args.visualize and (not self.source_type.tensor)else False)return self.model(im, augment=self.args.augment, visualize=visualize, embed=self.args.embed, *args, **kwargs)def pre_transform(self, im):"""Pre-transform input image before inference.Args:im (List(np.ndarray)): (N, 3, h, w) for tensor, [(h, w, 3) x N] for list.Returns:(list): A list of transformed images."""same_shapes = len({x.shape for x in im}) == 1letterbox = LetterBox(self.imgsz, auto=same_shapes and self.model.pt, stride=self.model.stride)return [letterbox(image=x) for x in im]def postprocess(self, preds, img, orig_imgs):"""Post-processes predictions for an image and returns them."""return predsdef __call__(self, source=None, model=None, stream=False, *args, **kwargs):"""Performs inference on an image or stream."""self.stream = streamif stream:return self.stream_inference(source, model, *args, **kwargs)else:return list(self.stream_inference(source, model, *args, **kwargs))  # merge list of Result into onedef predict_cli(self, source=None, model=None):"""Method used for CLI prediction.It uses always generator as outputs as not required by CLI mode."""gen = self.stream_inference(source, model)for _ in gen:  # noqa, running CLI inference without accumulating any outputs (do not modify)passdef setup_source(self, source):"""Sets up source and inference mode."""self.imgsz = check_imgsz(self.args.imgsz, stride=self.model.stride, min_dim=2)  # check image sizeself.transforms = (getattr(self.model.model,"transforms",classify_transforms(self.imgsz[0], crop_fraction=self.args.crop_fraction),)if self.args.task == "classify"else None)self.dataset = load_inference_source(source=source,batch=self.args.batch,vid_stride=self.args.vid_stride,buffer=self.args.stream_buffer,)self.source_type = self.dataset.source_typeif not getattr(self, "stream", True) and (self.source_type.streamor self.source_type.screenshotor len(self.dataset) > 1000  # many imagesor any(getattr(self.dataset, "video_flag", [False]))):  # videosLOGGER.warning(STREAM_WARNING)self.vid_writer = {}@smart_inference_mode()def stream_inference(self, source=None, model=None, *args, **kwargs):"""Streams real-time inference on camera feed and saves results to file."""if self.args.verbose:LOGGER.info("")# Setup modelif not self.model:self.setup_model(model)with self._lock:  # for thread-safe inference# Setup source every time predict is calledself.setup_source(source if source is not None else self.args.source)# Check if save_dir/ label file existsif self.args.save or self.args.save_txt:(self.save_dir / "labels" if self.args.save_txt else self.save_dir).mkdir(parents=True, exist_ok=True)# Warmup modelif not self.done_warmup:self.model.warmup(imgsz=(1 if self.model.pt or self.model.triton else self.dataset.bs, 3, *self.imgsz))self.done_warmup = Trueself.seen, self.windows, self.batch = 0, [], Noneprofilers = (ops.Profile(device=self.device),ops.Profile(device=self.device),ops.Profile(device=self.device),)self.run_callbacks("on_predict_start")for self.batch in self.dataset:self.run_callbacks("on_predict_batch_start")paths, im0s, s = self.batch# Preprocesswith profilers[0]:im = self.preprocess(im0s)# Inferencewith profilers[1]:preds = self.inference(im, *args, **kwargs)if self.args.embed:yield from [preds] if isinstance(preds, torch.Tensor) else preds  # yield embedding tensorscontinue# Postprocesswith profilers[2]:self.results = self.postprocess(preds, im, im0s)self.run_callbacks("on_predict_postprocess_end")# Visualize, save, write resultsn = len(im0s)for i in range(n):self.seen += 1self.results[i].speed = {"preprocess": profilers[0].dt * 1e3 / n,"inference": profilers[1].dt * 1e3 / n,"postprocess": profilers[2].dt * 1e3 / n,}if self.args.verbose or self.args.save or self.args.save_txt or self.args.show:s[i] += self.write_results(i, Path(paths[i]), im, s)# Print batch resultsif self.args.verbose:LOGGER.info("\n".join(s))self.run_callbacks("on_predict_batch_end")yield from self.results# Release assetsfor v in self.vid_writer.values():if isinstance(v, cv2.VideoWriter):v.release()# Print final resultsif self.args.verbose and self.seen:t = tuple(x.t / self.seen * 1e3 for x in profilers)  # speeds per imageLOGGER.info(f"Speed: %.1fms preprocess, %.1fms inference, %.1fms postprocess per image at shape "f"{(min(self.args.batch, self.seen), 3, *im.shape[2:])}" % t)if self.args.save or self.args.save_txt or self.args.save_crop:nl = len(list(self.save_dir.glob("labels/*.txt")))  # number of labelss = f"\n{nl} label{'s' * (nl > 1)} saved to {self.save_dir / 'labels'}" if self.args.save_txt else ""LOGGER.info(f"Results saved to {colorstr('bold', self.save_dir)}{s}")self.run_callbacks("on_predict_end")def setup_model(self, model, verbose=True):"""Initialize YOLO model with given parameters and set it to evaluation mode."""self.model = AutoBackend(weights=model or self.args.model,device=select_device(self.args.device, verbose=verbose),dnn=self.args.dnn,data=self.args.data,fp16=self.args.half,batch=self.args.batch,fuse=True,verbose=verbose,)self.device = self.model.device  # update deviceself.args.half = self.model.fp16  # update halfself.model.eval()def write_results(self, i, p, im, s):"""Write inference results to a file or directory."""string = ""  # print stringif len(im.shape) == 3:im = im[None]  # expand for batch dimif self.source_type.stream or self.source_type.from_img or self.source_type.tensor:  # batch_size >= 1string += f"{i}: "frame = self.dataset.countelse:match = re.search(r"frame (\d+)/", s[i])frame = int(match.group(1)) if match else None  # 0 if frame undeterminedself.txt_path = self.save_dir / "labels" / (p.stem + ("" if self.dataset.mode == "image" else f"_{frame}"))string += "%gx%g " % im.shape[2:]result = self.results[i]result.save_dir = self.save_dir.__str__()  # used in other locationsstring += result.verbose() + f"{result.speed['inference']:.1f}ms"# Add predictions to imageif self.args.save or self.args.show:self.plotted_img = result.plot(line_width=self.args.line_width,boxes=self.args.show_boxes,conf=self.args.show_conf,labels=self.args.show_labels,im_gpu=None if self.args.retina_masks else im[i],)# Save resultsif self.args.save_txt:result.save_txt(f"{self.txt_path}.txt", save_conf=self.args.save_conf)if self.args.save_crop:result.save_crop(save_dir=self.save_dir / "crops", file_name=self.txt_path.stem)if self.args.show:self.show(str(p))if self.args.save:self.save_predicted_images(str(self.save_dir / p.name), frame)return stringdef save_predicted_images(self, save_path="", frame=0):"""Save video predictions as mp4 at specified path."""im = self.plotted_img# Save videos and streamsif self.dataset.mode in {"stream", "video"}:fps = self.dataset.fps if self.dataset.mode == "video" else 30frames_path = f'{save_path.split(".", 1)[0]}_frames/'if save_path not in self.vid_writer:  # new videoif self.args.save_frames:Path(frames_path).mkdir(parents=True, exist_ok=True)# Always save as MP4 regardless of OSsuffix, fourcc = (".mp4", "avc1")self.vid_writer[save_path] = cv2.VideoWriter(filename=str(Path(save_path).with_suffix(suffix)),fourcc=cv2.VideoWriter_fourcc(*fourcc),fps=fps,  # integer required, floats produce error in MP4 codecframeSize=(im.shape[1], im.shape[0]),  # (width, height))# Save videoself.vid_writer[save_path].write(im)if self.args.save_frames:cv2.imwrite(f"{frames_path}{frame}.jpg", im)# Save imageselse:cv2.imwrite(save_path, im)def show(self, p=""):"""Display an image in a window using OpenCV imshow()."""im = self.plotted_imgif platform.system() == "Linux" and p not in self.windows:self.windows.append(p)cv2.namedWindow(p, cv2.WINDOW_NORMAL | cv2.WINDOW_KEEPRATIO)  # allow window resize (Linux)cv2.resizeWindow(p, im.shape[1], im.shape[0])  # (width, height)cv2.imshow(p, im)cv2.waitKey(300 if self.dataset.mode == "image" else 1)  # 1 milliseconddef run_callbacks(self, event: str):"""Runs all registered callbacks for a specific event."""for callback in self.callbacks.get(event, []):callback(self)def add_callback(self, event: str, func):"""Add callback."""self.callbacks[event].append(func)