在运行gen_hard_example的时候出现

[RickeyZhao1]

在运行gen_hard_example时，出现下述报错。
load test data
finish loading
start detecting....
2020-11-17 10:55:51.587938: I tensorflow/stream_executor/dso_loader.cc:152] successfully opened CUDA library cublas64_100.dll locally
100 out of 12880 images done
0.209250 seconds for each image
200 out of 12880 images done
0.150519 seconds for each image
300 out of 12880 images done
0.176074 seconds for each image
400 out of 12880 images done
0.192142 seconds for each image
Traceback (most recent call last):
File "gen_hard_example.py", line 251, in
vis=False)
File "gen_hard_example.py", line 171, in t_net
detections,_ = mtcnn_detector.detect_face(test_data)
File "..\Detection\MtcnnDetector.py", line 458, in detect_face
boxes, boxes_c, landmark = self.detect_rnet(im, boxes_c)
File "..\Detection\MtcnnDetector.py", line 291, in detect_rnet
tmp[dy[i]:edy[i] + 1, dx[i]:edx[i] + 1, :] = im[y[i]:ey[i] + 1, x[i]:ex[i] + 1, :]
ValueError: could not broadcast input array from shape (31,1008,3) into shape (31,0,3)

会不会跟之前的Pnet的检测结果有关，我使用深度可分离卷积更改了Pnet的网络结构。