Traceback (most recent call last):
File "main.py", line 5, in <module>
from qt5_faceR_window import mywindow
File "/media/luckynote/Tool/face-recognition/10-faceSIS/qt5_faceR_window.py", line 6, in <module>
import face_recognition
File "/home/luckynote/anaconda3/lib/python3.6/site-packages/face_recognition/__init__.py", line 7, in <module>
from .api import load_image_file, face_locations, batch_face_locations, face_landmarks, face_encodings, compare_faces, face_distance
File "/home/luckynote/anaconda3/lib/python3.6/site-packages/face_recognition/api.py", line 23, in <module>
cnn_face_detector = dlib.cnn_face_detection_model_v1(cnn_face_detection_model)
RuntimeError: Error while calling cudaMallocHost(&data, new_size*sizeof(float)) in file /media/luckynote/Tool/face-recognition/dlib-19.15.99/dlib/cuda/gpu_data.cpp:185. code: 30, reason: unknown error
以前都是在CPU上跑face_recognition，這次換了GPU的機器，就報錯了，在官方上https://github.com/ageitgey/face_recognition/issues/358 找到了答案。
原來GPU必須要使用sudo權限才能調用，直接使用python ***.py的命令行方式是不行的，但是加上sudo，即sudo python ***.py再運行就沒問題了。