为什么python app.py后模型看起来在CPU上运行？ #6
Description
按照提供的环境配置方式，搭起了运行环境，torch.cuda.is_available()显示为True
运行python app.py 后的打印信息为
WARNING: All log messages before absl::InitializeLog() is called are written to STDERR
I0000 00:00:1740042889.837939 34520 task_runner.cc:85] GPU suport is not available: INTERNAL: ; RET_CHECK failure (mediapipe/gpu/gl_context_egl.cc:77) display != EGL_NO_DISPLAYeglGetDisplay() returned error 0x300c
W0000 00:00:1740042889.838275 34520 face_landmarker_graph.cc:174] Sets FaceBlendshapesGraph acceleration to xnnpack by default.
INFO: Created TensorFlow Lite XNNPACK delegate for CPU.
W0000 00:00:1740042889.849758 34674 inference_feedback_manager.cc:114] Feedback manager requires a model with a single signature inference. Disabling support for feedback tensors.
W0000 00:00:1740042889.864038 34684 inference_feedback_manager.cc:114] Feedback manager requires a model with a single signature inference. Disabling support for feedback tensors.
I0000 00:00:1740042889.867534 34520 task_runner.cc:85] GPU suport is not available: INTERNAL: ; RET_CHECK failure (mediapipe/gpu/gl_context_egl.cc:77) display != EGL_NO_DISPLAYeglGetDisplay() returned error 0x300c
W0000 00:00:1740042889.871974 34723 inference_feedback_manager.cc:114] Feedback manager requires a model with a single signature inference. Disabling support for feedback tensors.
Applied providers: ['CPUExecutionProvider'], with options: {'CPUExecutionProvider': {}}
model ignore: /root/.insightface/models/buffalo_l/1k3d68.onnx landmark_3d_68
Applied providers: ['CPUExecutionProvider'], with options: {'CPUExecutionProvider': {}}
model ignore: /root/.insightface/models/buffalo_l/2d106det.onnx landmark_2d_106
Applied providers: ['CPUExecutionProvider'], with options: {'CPUExecutionProvider': {}}
find model: /root/.insightface/models/buffalo_l/det_10g.onnx detection [1, 3, '?', '?'] 127.5 128.0
Applied providers: ['CPUExecutionProvider'], with options: {'CPUExecutionProvider': {}}
model ignore: /root/.insightface/models/buffalo_l/genderage.onnx genderage
Applied providers: ['CPUExecutionProvider'], with options: {'CPUExecutionProvider': {}}
model ignore: /root/.insightface/models/buffalo_l/w600k_r50.onnx recognition
set det-size: (640, 640)
Using generic FLAME model
Loading pipeline components...: 0%| | 0/5 [00:00<?, ?it/s]Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
Loading checkpoint shards: 100%|██████████████████████████████████| 2/2 [00:00<00:00, 3.81it/s]
Loading pipeline components...: 100%|█████████████████████████████| 5/5 [00:00<00:00, 7.14it/s]
视频生成极慢。GPU上只有约850MB的显存占用量。模型似乎是在CPU上计算。