DidaDidaDidaD
DidaDidaDidaD
By the way ,the whole paragragh is below: (venv) D:\学习资料\语音口型资料\voca-master>python visualize_sequence.py --sequence_path FLAME_eye_blink/meshes --audio_fname audio/test_sentence.wav --out_path FLAME_eye_blink --uv_ template_fname template/texture_mesh.obj --texture_img_fname template/texture_mesh.png 2021-04-27 17:22:58.211253: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library...
The route is changed into English, the error is: (venv) D:\xuexiziliao\yuyinkouxingziliao\voca-master>python visualize_sequence.py --sequence_path FLAME_eye_blink/meshes --audio_fname audio/test_sentence.wav --out_path FLAME_eye_bl ink --uv_template_fname template/texture_mesh.obj --texture_img_fname template/texture_mesh.png 2021-04-27 22:00:53.352324: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic...
> windows is not supported > […](#) > On Mon, Apr 11, 2022 at 8:29 PM helloxuejuxi ***@***.***> wrote: win10 is anyone else succeed in running the examples? — Reply...
hugging face 里用模型Helsinki-NLP/opus-mt-zh-en实现的中译英,pipeline("translation", model="Helsinki-NLP/opus-mt-zh-en"),能否使用lightseq来转换翻译模型?多谢
这个对小白确实不友好啊,列出来那么多翻译的对比数据,却没有给翻译的转换代码
我用的GTX 1660Ti显卡,转换后的模型时间也是比Hugging Face 的慢,按说算力达到了,win10上以前运行程序显示的算力是7.5
作者人呢?平时都不来这啊
我也遇到了这个问题,什么情况? python36._pth这里是空的,还有解压后的python36文件夹里是有encodings文件夹的,bug啊
> RAM = 8GB GPU = Nvidia GTX 1050 (4GB) I am using demo video. 请问该问题解决了吗?这个demo我没遇到这个问题,但运行第二个demo我遇到了bus问题
cuda torch 版本要匹配,太扯淡了,试探了好几个版本,还有v1.1 sourcecode,上面报告的错误就是torch版本不匹配导致的