用yolov8训练出的模型转om后基于官方的代码改路径推理，总是报错：Traceback (most recent call last): 

  File "/home/DengYH/sample/main2.py", line 41, in <module> 

    img_dw = draw_bbox(pred_all, img_bgr, (0, 255, 0), 2, labels_dict)  # 画出检测框、类别、概率 

  File "/home/DengYH/sample/det_utils.py", line 223, in draw_bbox 

    img0 = cv2.putText(img0, str(idx) + ' ' + names[int(class_id)], (int(bbox[idx][0]), int(bbox[idx][1] + 16)), cv2.FONT_HERSHEY_SIMPLEX, 0.5, (0, 0, 255), 1) 

KeyError: 8220 

打印了label_dict和pred_all，分别为{0: 'wu', 1: 'qiao_ji', 2: 'chong_wang', 3: 'fan_yue', 4: 'jian_wang', 5: 'pa_wang'} [[ 0. 0. 180.41992 0. 58768.312 8220. ] [ 0. 0. 180.41992 0. 58553.438 8295. ] [ 0. 0. 180.41992 0. 58284.844 8156. ] ... [ 0. 0. 180.41992 0. 50925.375 8383. ] [ 0. 0. 180.41992 0. 50925.375 8254. ] [ 0. 0. 180.41992 0. 50871.656 8338. ]]

它预测的结果很不对。