Skip to content

Deep feature representation h是一个向量吗 #1

@BigPig-LittleTail

Description

@BigPig-LittleTail

我在论文中看到:The network defines a trasnformation
image which transforms an raw input text x to r-dimensional deep representation h.
这是不是就意味着最后的h是一个r维的向量呢?
因为注意到文章在超参数一节中写道CNN的两层中共使用了两次Folding,我疑惑的是,48行的矩阵输入经过两次folding是不是就应该变成12行矩阵呢,想请问最后是怎么成为h向量的呢?

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions