DO-Conv: Depthwise Over-parameterized Convolutional Layer
作者引入了over-parameterized的卷积层,这种卷积层的好处是在training的时候容易训练,在inference的时候可以将宽参数融合,得到单个卷积层进行推理。
From here you can search these documents. Enter your search terms below.
Keys | Action |
---|---|
? | Open this help |
n | Next page |
p | Previous page |
s | Search |