search

Henry-23 / VideoChat

实时语音交互数字人，支持端到端语音方案（GLM-4-Voice - THG）和级联方案（ASR-LLM-TTS-THG）。可自定义形象与音色，无须训练，支持音色克隆，首包延迟低至3s。Real-time voice interactive digital human, supporting end-to-end voice solutions (GLM-4-Voice - THG) and cascaded solutions (ASR-LLM-TTS-THG). Customizable appearance and voice, supporting voice cloning, with initial package delay as low as 3s.
https://www.modelscope.cn/studios/AI-ModelScope/video_chat
MIT License
385 stars 50 forks source link

模型加载问题 #5

Closed Upcreat closed 1 week ago

Upcreat commented 1 week ago

按照项目流程进行部署，但是出现了下面模型加载问题，请问这是什么原因呀 Loads checkpoint by local backend from path: weights/dwpose/dw-ll_ucoco_384.pth cuda start Load from local weight: /home/pod/shared-nvme/VideoChat/video_chat/src/musetalk/utils/face_detection/detection/sfd/s3fd.pth Traceback (most recent call last): File "/home/pod/shared-nvme/VideoChat/video_chat/app.py", line 45, in from src.pipeline_llm import llm_pipeline File "/home/pod/shared-nvme/VideoChat/video_chat/src/pipeline_llm.py", line 25, in from src.thg import Muse_Talk File "/home/pod/shared-nvme/VideoChat/video_chat/src/thg.py", line 25, in from src.musetalk.utils.blending import get_image,get_image_prepare_material,get_image_blending File "/home/pod/shared-nvme/VideoChat/video_chat/src/musetalk/utils/blending.py", line 6, in fp = FaceParsing() File "/home/pod/shared-nvme/VideoChat/video_chat/src/musetalk/utils/face_parsing/init.py", line 12, in init self.net = self.model_init() File "/home/pod/shared-nvme/VideoChat/video_chat/src/musetalk/utils/face_parsing/init.py", line 18, in model_init net = BiSeNet(resnet_path) File "/home/pod/shared-nvme/VideoChat/video_chat/src/musetalk/utils/face_parsing/model.py", line 233, in init self.cp = ContextPath(resnet_path) File "/home/pod/shared-nvme/VideoChat/video_chat/src/musetalk/utils/face_parsing/model.py", line 95, in init self.resnet = Resnet18(resnet_path) File "/home/pod/shared-nvme/VideoChat/video_chat/src/musetalk/utils/face_parsing/resnet.py", line 69, in init self.init_weight(model_path) File "/home/pod/shared-nvme/VideoChat/video_chat/src/musetalk/utils/face_parsing/resnet.py", line 83, in init_weight state_dict = torch.load(model_path) #modelzoo.load_url(resnet18_url) File "/home/pod/miniconda3/envs/metahuman/lib/python3.10/site-packages/torch/serialization.py", line 1028, in load return _legacy_load(opened_file, map_location, pickle_module, pickle_load_args) File "/home/pod/miniconda3/envs/metahuman/lib/python3.10/site-packages/torch/serialization.py", line 1246, in _legacy_load magic_number = pickle_module.load(f, pickle_load_args) _pickle.UnpicklingError: invalid load key, 'v'.

Henry-23 commented 1 week ago

weights文件夹里的权重下载了吗？

Upcreat commented 1 week ago

重新试了一下，原因是模型没有下载完全，下载的模型不完整。