RuntimeError: GET was unable to find an engine to execute this computation

[208wuqi]

Hello, I encountered the following problems when testing the demo：
“RuntimeError: GET was unable to find an engine to execute this computation”
how to solve it？thank you!

[hujiecpp]

Can you provide the runtime information more detailly?

[208wuqi]

python demo/demo.py --config-file projects/YOSO/configs/YOSO-R50.yaml --video-input 2.mkv --output output_video.mkv --opts MODEL.WEIGHTS ./model_zoo/yoso_res50_coco.pth
[05/23 10:05:25 detectron2]: Arguments: Namespace(confidence_threshold=0.8, config_file='projects/YOSO/configs/YOSO-R50.yaml', input=None, opts=['MODEL.WEIGHTS', './model_zoo/yoso_res50_coco.pth'], output='output_video.mkv', overlap_threshold=0.98, video_input='2.mkv', webcam=False)
[05/23 10:05:26 fvcore.common.checkpoint]: [Checkpointer] Loading from ./model_zoo/yoso_res50_coco.pth ...
512 256
[ERROR:0@2.038] global cap_ffmpeg_impl.hpp:2991 open Could not find encoder for codec_id=27, error: Encoder not found
[ERROR:0@2.038] global cap_ffmpeg_impl.hpp:3066 open VIDEOIO/FFMPEG: Failed to initialize VideoWriter
[ERROR:0@2.038] global cap.cpp:595 open VIDEOIO(CV_IMAGES): raised OpenCV exception:

OpenCV(4.7.0) /io/opencv/modules/videoio/src/cap_images.cpp:253: error: (-5:Bad argument) CAP_IMAGES: can't find starting number (in the name of file): /tmp/video_format_testjaz2izyw/test_file.mkv in function 'icvExtractPattern'

0%| | 0/7501 [00:00<?, ?it/s]
Traceback (most recent call last):
File "demo/demo.py", line 189, in
for vis_frame in tqdm.tqdm(demo.run_on_video(video), total=num_frames):
File "/home/wuqi/.local/lib/python3.8/site-packages/tqdm/std.py", line 1178, in iter
for obj in iterable:
File "/home/wuqi/Code/YOSO/demo/predictor.py", line 130, in run_on_video
yield process_predictions(frame, self.predictor(frame))
File "/home/wuqi/Code/YOSO/detectron2/engine/defaults.py", line 223, in call
predictions = self.model([inputs])[0]
File "/home/wuqi/.local/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1501, in _call_impl
return forward_call(*args, **kwargs)
File "/home/wuqi/Code/YOSO/projects/YOSO/yoso/segmentator.py", line 78, in forward
backbone_feats = self.backbone(images.tensor)
File "/home/wuqi/.local/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1501, in _call_impl
return forward_call(*args, **kwargs)
File "/home/wuqi/Code/YOSO/detectron2/modeling/backbone/resnet.py", line 434, in forward
x = self.stem(x)
File "/home/wuqi/.local/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1501, in _call_impl
return forward_call(*args, **kwargs)
File "/home/wuqi/Code/YOSO/detectron2/modeling/backbone/resnet.py", line 356, in forward
x = self.conv1(x)
File "/home/wuqi/.local/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1501, in _call_impl
return forward_call(*args, **kwargs)
File "/home/wuqi/Code/YOSO/detectron2/layers/wrappers.py", line 76, in forward
x = F.conv2d(
RuntimeError: GET was unable to find an engine to execute this computation

[hujiecpp]

It seems that the VideoWriter does not support the .mkv format. Can you try other video format such as .mp4?

[208wuqi]

What I input is a .mp4 video file.

[hujiecpp]

--video-input 2.mkv --output output_video.mkv

[208wuqi]

python demo/demo.py --config-file projects/YOSO/configs/YOSO-R50.yaml --video-input 1.mp4 --output output_video.mp4 --opts MODEL.WEIGHTS ./model_zoo/yoso_res50_coco.pth
[05/23 10:54:00 detectron2]: Arguments: Namespace(confidence_threshold=0.8, config_file='projects/YOSO/configs/YOSO-R50.yaml', input=None, opts=['MODEL.WEIGHTS', './model_zoo/yoso_res50_coco.pth'], output='output_video.mp4', overlap_threshold=0.98, video_input='1.mp4', webcam=False)
[05/23 10:54:01 fvcore.common.checkpoint]: [Checkpointer] Loading from ./model_zoo/yoso_res50_coco.pth ...
512 256
[ERROR:0@2.862] global cap_ffmpeg_impl.hpp:2991 open Could not find encoder for codec_id=27, error: Encoder not found
[ERROR:0@2.862] global cap_ffmpeg_impl.hpp:3066 open VIDEOIO/FFMPEG: Failed to initialize VideoWriter
[ERROR:0@2.863] global cap.cpp:595 open VIDEOIO(CV_IMAGES): raised OpenCV exception:

OpenCV(4.7.0) /io/opencv/modules/videoio/src/cap_images.cpp:253: error: (-5:Bad argument) CAP_IMAGES: can't find starting number (in the name of file): /tmp/video_format_testkiva3ir6/test_file.mkv in function 'icvExtractPattern'

0%| | 0/24800 [00:01<?, ?it/s]
Traceback (most recent call last):
File "demo/demo.py", line 189, in
for vis_frame in tqdm.tqdm(demo.run_on_video(video), total=num_frames):
File "/home/wuqi/.local/lib/python3.8/site-packages/tqdm/std.py", line 1178, in iter
for obj in iterable:
File "/home/wuqi/Code/YOSO/demo/predictor.py", line 130, in run_on_video
yield process_predictions(frame, self.predictor(frame))
File "/home/wuqi/Code/YOSO/detectron2/engine/defaults.py", line 223, in call
predictions = self.model([inputs])[0]
File "/home/wuqi/.local/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1501, in _call_impl
return forward_call(*args, **kwargs)
File "/home/wuqi/Code/YOSO/projects/YOSO/yoso/segmentator.py", line 78, in forward
backbone_feats = self.backbone(images.tensor)
File "/home/wuqi/.local/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1501, in _call_impl
return forward_call(*args, **kwargs)
File "/home/wuqi/Code/YOSO/detectron2/modeling/backbone/resnet.py", line 434, in forward
x = self.stem(x)
File "/home/wuqi/.local/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1501, in _call_impl
return forward_call(*args, **kwargs)
File "/home/wuqi/Code/YOSO/detectron2/modeling/backbone/resnet.py", line 356, in forward
x = self.conv1(x)
File "/home/wuqi/.local/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1501, in _call_impl
return forward_call(*args, **kwargs)
File "/home/wuqi/Code/YOSO/detectron2/layers/wrappers.py", line 76, in forward
x = F.conv2d(
RuntimeError: GET was unable to find an engine to execute this computation