binzzheng / dvc-pytorch Goto Github PK

In the official implementation, the I frame is used by the learning-based image compression algorithm (Ballé et al). Why do you change it to H.265?

can you share your pretrained weights

Can we communicate the DVC code? It's urgent. My wechat is zxl_ Nefelibata

作者您好，我最近在跑您开源的这个代码，目前遇到一个问题：

在subnet/init.py文件中：
from .entropy_params import *
from .context import *
from .feature_convert import *
from .offset_process import *
from .fvc_compression import *
这几个导入没有相应模块呢？请问是怎么回事儿呢？