# 使用hook查看，请搜register / CPU Simulation即可看到对应位置 import argparse import time from pathlib import Path import cv2 import torch import torch.backends.cudnn as cudnn from numpy import random debug = True # new added import torch.nn as nn from utils.activations import Hardswish, SiLU nn.modules.activation.SiLU = SiLU nn.modules.activation.Hardswish = Hardswish nn.SiLU = SiLU nn.Hardswish = Hardswish # from models.experimental import attempt_load from utils.datasets import LoadStreams, LoadImages from utils.general import check_img_size, check_requirements, check_imshow, non_max_suppression, apply_classifier, \ scale_coords, xyxy2xywh, strip_optimizer, set_logging, increment_path from utils.plots import plot_one_box from utils.torch_utils import select_device, load_classifier, time_synchronized, TracedModel def detect(save_img=False): source, weights, view_img, save_txt, imgsz, trace = opt.source, opt.weights, opt.view_img, opt.save_txt, opt.img_size, not opt.no_trace jit, save, mname, mcore, mlu_det, half_input = opt.jit, opt.save, opt.mname, opt.mcore, opt.mlu_det, opt.half_input save_img = not opt.nosave and not source.endswith('.txt') # save inference images webcam = source.isnumeric() or source.endswith('.txt') or source.lower().startswith( ('rtsp://', 'rtmp://', 'http://', 'https://')) # Directories save_dir = Path(increment_path(Path(opt.project) / opt.name, exist_ok=opt.exist_ok)) # increment run (save_dir / 'labels' if save_txt else save_dir).mkdir(parents=True, exist_ok=True) # make dir # Initialize set_logging() import torch_mlu import torch_mlu.core.mlu_model as ct import torch_mlu.core.mlu_quantize as mlu_quantize from mlu.routine.postprocess import MLU_PostProcessYoloV7, PostProcessPytorchYoloV7, draw_image from models.yolo import get_empty_model model = get_empty_model(opt) another = get_empty_model(opt) print(f'Empty Model is like:\n{model}\n\n\n') stride = 32 # 直接指定成32倍降采样了 imgsz = check_img_size(imgsz, s=stride) # check img_size ct.set_core_number(4) # 配置MLU core类型 ct.set_core_version(opt.mcore) torch.set_grad_enabled(False) device = ct.mlu_device() half = device.type != 'cpu' # half precision only supported on CUDA print("run on %s ..." % device) # 加载量化模型，只允许加载一个 weight = weights[0] quantized_net = torch_mlu.core.mlu_quantize.quantize_dynamic_mlu(model) state_dict = torch.load(weight) quantized_net.load_state_dict(state_dict, strict= False) # hook from torch_mlu.core.utils import sim_quant_utils sim_quant_utils.register_quant_hook(another) another.load_state_dict(torch.load(weight), strict=False) # 设置为推理模式 quantized_net = quantized_net.float().eval() quantized_net.to(device) print(f'\n\nquan.device: {next(quantized_net.parameters()).device}') if debug else None # on the MLU model = quantized_net print(f'model.device: {next(model.parameters()).device}\n\n') if debug else None # on the MLU # print(f'\n\n======inference model has been set, model is:======\n{model}\n\n\n') print(f'\n\n======inference model has been set, model is: why Cant be print======{type(model)}\n\n\n') # Set Dataloader vid_path, vid_writer = None, None if webcam: view_img = check_imshow() cudnn.benchmark = True # set True to speed up constant image size inference dataset = LoadStreams(source, img_size=imgsz, stride=stride) else: dataset = LoadImages(source, img_size=imgsz, stride=stride) # Get names and colors # names = model.module.names if hasattr(model, 'module') else model.names names = ['bottle'] colors = [[random.randint(0, 255) for _ in range(3)] for _ in names] # Run inference print('===first time warmup begins===') if device.type != 'cpu': model(torch.zeros(1, 3, imgsz, imgsz).to(device).type_as(next(model.parameters()))) # run once old_img_w = old_img_h = imgsz old_img_b = 1 print('===first time warmup Ends===') t0 = time.time() img_origin = None for path, img, im0s, vid_cap in dataset: # Front Process # img = torch.from_numpy(img).to(device) # ! float... # uint8 to fp16/32 img_origin = torch.from_numpy(img).float() if img_origin.ndimension() == 3: img_origin = img_origin.unsqueeze(0) # if half: # img = torch.from_numpy(img).half().to(device) # print("Front Processing half now") if debug else None # else: img = torch.from_numpy(img).float().to(device) print("Front Processing half now") if debug else None img /= 255.0 # 0 - 255 to 0.0 - 1.0 print(img.cpu()) if debug else None if img.ndimension() == 3: img = img.unsqueeze(0) # Warmup if device.type != 'cpu' and ( old_img_b != img.shape[0] or old_img_h != img.shape[2] or old_img_w != img.shape[3]): old_img_b = img.shape[0] old_img_h = img.shape[2] old_img_w = img.shape[3] for i in range(3): model(img, augment=opt.augment)[0] print('======WarmUp finished======\n\n\n') # Inference t1 = time_synchronized() print(f'\n\n\n=====time Inference Begins:{t1}=====\n') if debug else None with torch.no_grad(): print(f'model.device: {next(model.parameters()).device}') if debug else None print(f'img.device: {img.device}\n') if debug else None detect_out = model(img)[0] t2 = time_synchronized() print(f'\n=====time Inference Ends: {t2}=====\n') print(f'Time usage:{t2 - t1}') if len(detect_out) == 1: pred = detect_out.cpu().type(torch.FloatTensor) if opt.half_input else detect_out.cpu() else: pred = [out.cpu().type(torch.FloatTensor) for out in detect_out] print("mlu pred:{} {}\n{}".format(type(pred), pred.shape, pred)) if debug else None # Apply NMS pred = non_max_suppression(pred, opt.conf_thres, opt.iou_thres, classes=opt.classes, agnostic=opt.agnostic_nms) t3 = time_synchronized() print("NMS:{} \n{}".format(type(pred), pred)) if debug else None print('\n\n\n ===============CPU Simulation here==================\n') output = another(img_origin)[0] from models.common import printf printf(output) print('\n\n NMS test\n') # Apply NMS output = non_max_suppression(output, opt.conf_thres, opt.iou_thres, classes=opt.classes, agnostic=opt.agnostic_nms) print("NMS:{} \n{}".format(type(output), output)) if debug else None print('===========try this well Come Onnnnnnnnnn========\n\n\n\n\n') if opt.mlu_det: pred = postproc.get_boxes(pred) p, s, im0, frame = path, '', im0s, getattr(dataset, 'frame', 0) p = Path(p) # to Path save_path = str(save_dir / p.name) # img.jpg print(save_path) # todo draw_image还没写 draw_image(pred, img, im0s, path, save_path, names) exit(0) # pred = postproc.yolo_det(pred)[0] # Process detections for i, det in enumerate(pred): # detections per image if webcam: # batch_size >= 1 p, s, im0, frame = path[i], '%g: ' % i, im0s[i].copy(), dataset.count else: p, s, im0, frame = path, '', im0s, getattr(dataset, 'frame', 0) p = Path(p) # to Path save_path = str(save_dir / p.name) # img.jpg # txt_path = str(save_dir / 'labels' / p.stem) + ('' if dataset.mode == 'image' else f'_{frame}') # img.txt txt_path = str(save_dir / 'labels' / p.stem) + ( '' if dataset.mode == 'image' else '_{}'.format(frame)) # img.txt gn = torch.tensor(im0.shape)[[1, 0, 1, 0]] # normalization gain whwh if len(det): # Rescale boxes from img_size to im0 size det[:, :4] = scale_coords(img.shape[2:], det[:, :4], im0.shape).round() # Print results for c in det[:, -1].unique(): n = (det[:, -1] == c).sum().item() # detections per class # print(f"n = {n} \t", type(n)) # k = int(n) # print(f"k = {k} \t", type(k)) # print(type('s')) # print(type(n>1)) s += f"{n} {names[int(c)]}{'s' * (n > 1)}, " # add to string # Write results for *xyxy, conf, cls in reversed(det): if save_txt: # Write to file xywh = (xyxy2xywh(torch.tensor(xyxy).view(1, 4)) / gn).view(-1).tolist() # normalized xywh line = (cls, *xywh, conf) if opt.save_conf else (cls, *xywh) # label format with open(txt_path + '.txt', 'a') as f: f.write(('%g ' * len(line)).rstrip() % line + '\n') if save_img or view_img: # Add bbox to image label = f'{names[int(cls)]} {conf:.2f}' plot_one_box(xyxy, im0, label=label, color=colors[int(cls)], line_thickness=1) # Print time (inference + NMS) print(f'{s}Done. ({(1E3 * (t2 - t1)):.1f}ms) Inference, ({(1E3 * (t3 - t2)):.1f}ms) NMS') # print(f'===wtf s ===Done. ({(1E3 * (t2 - t1)):.1f}ms) Inference, ({(1E3 * (t3 - t2)):.1f}ms) NMS') # Stream results if view_img: cv2.imshow(str(p), im0) cv2.waitKey(1) # 1 millisecond # Save results (image with detections) if save_img: if dataset.mode == 'image': cv2.imwrite(save_path, im0) print(f" The image with the result is saved in: {save_path}") else: # 'video' or 'stream' if vid_path != save_path: # new video vid_path = save_path if isinstance(vid_writer, cv2.VideoWriter): vid_writer.release() # release previous video writer if vid_cap: # video fps = vid_cap.get(cv2.CAP_PROP_FPS) w = int(vid_cap.get(cv2.CAP_PROP_FRAME_WIDTH)) h = int(vid_cap.get(cv2.CAP_PROP_FRAME_HEIGHT)) else: # stream fps, w, h = 30, im0.shape[1], im0.shape[0] save_path += '.mp4' vid_writer = cv2.VideoWriter(save_path, cv2.VideoWriter_fourcc(*'mp4v'), fps, (w, h)) vid_writer.write(im0) if save_txt or save_img: s = f"\n{len(list(save_dir.glob('labels/*.txt')))} labels saved to {save_dir / 'labels'}" if save_txt else '' # print(f"Results saved to {save_dir}{s}") print(f'Done. ({time.time() - t0:.3f}s)') if __name__ == '__main__': parser = argparse.ArgumentParser() parser.add_argument('--weights', nargs='+', type=str, default='yolov7.pt', help='model.pt path(s)') parser.add_argument('--source', type=str, default='inference/images', help='source') # file/folder, 0 for webcam parser.add_argument('--img-size', type=int, default=640, help='inference size (pixels)') parser.add_argument('--conf-thres', type=float, default=0.25, help='object confidence threshold') parser.add_argument('--iou-thres', type=float, default=0.45, help='IOU threshold for NMS') parser.add_argument('--device', default='', help='cuda device, i.e. 0 or 0,1,2,3 or cpu') parser.add_argument('--view-img', action='store_true', help='display results') parser.add_argument('--save-txt', action='store_true', help='save results to *.txt') parser.add_argument('--save-conf', action='store_true', help='save confidences in --save-txt labels') parser.add_argument('--nosave', action='store_true', help='do not save images/videos') parser.add_argument('--classes', nargs='+', type=int, help='filter by class: --class 0, or --class 0 2 3') parser.add_argument('--agnostic-nms', action='store_true', help='class-agnostic NMS') parser.add_argument('--augment', action='store_true', help='augmented inference') parser.add_argument('--update', action='store_true', help='update all models') parser.add_argument('--project', default='runs/detect', help='save results to project/name') parser.add_argument('--name', default='exp', help='save results to project/name') parser.add_argument('--exist-ok', action='store_true', help='existing project/name ok, do not increment') parser.add_argument('--no-trace', action='store_true', help='don`t trace model') # new added parser.add_argument('--cfg', type=str, default='yolov7-flow.yaml', help='model.yaml') parser.add_argument('--batch-size', type=int, default=1, help='I dont know why we need batchsize') # ! 自改 parser.add_argument('--jit', action='store_true', help='declaring model type') parser.add_argument('--save', action='store_true', help='whether saving offline model') parser.add_argument('--mname', type=str, default='offline-model', help='name to save cambricon offline model') parser.add_argument('--mcore', type=str, default='MLU270', help='wdnmd') # ! not certain parser.add_argument('--mlu-det', action='store_true', help='declaring wtf?') parser.add_argument('--half-input', action='store_true', help='name to save cambricon offline model') opt = parser.parse_args() print(opt) # check_requirements(exclude=('pycocotools', 'thop')) with torch.no_grad(): if opt.update: # update all models (to fix SourceChangeWarning) for opt.weights in ['yolov7.pt']: detect() strip_optimizer(opt.weights) else: detect()