When I run the Face_verify code in linux
I got the error as below
Process Process-1:
Traceback (most recent call last):
File "/home/super/anaconda3/envs/test/lib/python3.8/multiprocessing/process.py", line 315, in _bootstrap
self.run()
File "/home/super/anaconda3/envs/test/lib/python3.8/multiprocessing/process.py", line 108, in run
self._target(*self._args, **self._kwargs)
File "face_rec1.py", line 147, in func_detect
bboxes, faces = mtcnn.align_multi(image, conf.face_limit, 16)
File "/home/super/Documents/Mahesh/mtcnn.py", line 33, in align_multi
boxes, landmarks = self.detect_faces(img, min_face_size)
File "/home/super/Documents/Mahesh/mtcnn.py", line 112, in detect_faces
boxes = run_first_stage(image, self.pnet, scale=s, threshold=thresholds[0])
File "/home/super/Documents/Mahesh/mtcnn_pytorch/src/first_stage.py", line 34, in run_first_stage
img = torch.FloatTensor(_preprocess(img)).to(device)
File "/home/super/anaconda3/envs/test/lib/python3.8/site-packages/torch/cuda/init.py", line 184, in _lazy_init
raise RuntimeError(
RuntimeError: Cannot re-initialize CUDA in forked subprocess. To use CUDA with multiprocessing, you must use the 'spawn' start method
we tried torch.multiprocesing.set_start_method('spawn', force=True)
But giving another error.
Please provide the solution for this.
When I run the Face_verify code in linux I got the error as below
Process Process-1: Traceback (most recent call last): File "/home/super/anaconda3/envs/test/lib/python3.8/multiprocessing/process.py", line 315, in _bootstrap self.run() File "/home/super/anaconda3/envs/test/lib/python3.8/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "face_rec1.py", line 147, in func_detect bboxes, faces = mtcnn.align_multi(image, conf.face_limit, 16) File "/home/super/Documents/Mahesh/mtcnn.py", line 33, in align_multi boxes, landmarks = self.detect_faces(img, min_face_size) File "/home/super/Documents/Mahesh/mtcnn.py", line 112, in detect_faces boxes = run_first_stage(image, self.pnet, scale=s, threshold=thresholds[0]) File "/home/super/Documents/Mahesh/mtcnn_pytorch/src/first_stage.py", line 34, in run_first_stage img = torch.FloatTensor(_preprocess(img)).to(device) File "/home/super/anaconda3/envs/test/lib/python3.8/site-packages/torch/cuda/init.py", line 184, in _lazy_init raise RuntimeError( RuntimeError: Cannot re-initialize CUDA in forked subprocess. To use CUDA with multiprocessing, you must use the 'spawn' start method
we tried torch.multiprocesing.set_start_method('spawn', force=True) But giving another error. Please provide the solution for this.