JackIRose
JackIRose
我也是这个问题,如果用tool/deploy.py一步转换的话获得engine文件,无法直接推理,不知道怎么加载插件库
> > 我也是这个问题,如果用tool/deploy.py一步转换的话获得engine文件,无法直接推理,不知道怎么加载插件库 > > 插件可以通过ctypes.cdll.LoadLibrary加载 我是使用python生成engine文件,然后打算用C++推理,该去哪找插件,C++推理又怎么加载。谢谢
> 感谢您的尝试,在inplace前先clone()就可以解决这个问题, 可以同步下huggingface上最新的模型代码,可以参考这个https://huggingface.co/Qwen/Qwen-VL-Chat/commit/e71876524333aa79e65b19c557c639175552e689 但是我修改后,运行 hugginface会冲掉我的修改,这个怎么解决?感谢
> > 感谢您的尝试,在inplace前先clone()就可以解决这个问题, 可以同步下huggingface上最新的模型代码,可以参考这个https://huggingface.co/Qwen/Qwen-VL-Chat/commit/e71876524333aa79e65b19c557c639175552e689 > > 但是我修改后,运行 hugginface会冲掉我的修改,这个怎么解决?感谢 这个问题已经解决了,要修改原来下载的文件内容而不是在.cache/transformer的文件内容,因为前者会覆盖后者的内容,因此在后面的文件改没有用
> Problem was in fact ffmpeg, I downloaded its files and placed them in the deepfakes folder and it ran, thanks everyone you dowload by youself? and put it in...
I would like to know what the format of your training set is, and how I can convert my own video into the same format as your training set through...
> I would like to know what the format of your training set is, and how I can convert my own video into the same format as your training set...