我有Python版本的Caffe代码。
那个python代码有一个data layer

name: "hidden_seed"
type: "NumpyData"
top: "hidden_seed"

我喜欢用c++ caffe装箱
layer{
    name: "hidden_seed"
    type: "Data"
    top: "hidden_seed"
}

我的查询是如何创建大小为300 x 250且初始化为0的数据层。

我查看了caffe中的数据层,可用的数据层是
Layers:

    Image Data - read raw images.
    Database - read data from LEVELDB or LMDB.
    HDF5 Input - read HDF5 data, allows data of arbitrary dimensions.
    HDF5 Output - write data as HDF5.
    Input - typically used for networks that are being deployed.
    Window Data - read window data file.
    Memory Data - read data directly from memory.
    Dummy Data - for static data and debugging.

我希望拥有哪一个适合数据层,以及如何为每个单元初始化0值?

最佳答案

看来 "DummyData" 层会为您完成这项工作:

layer {
  type: "DummyData"
  name: "hidden_seed"
  top: "hidden_seed"
  dummy_data_param {
    shape { dim: 300 dim: 250 }  # your desired data shape
    data_filler { type: "constant" value: 0 } # fill with all zeros
  }
}

关于python - 如何在caffe中创建数据层?,我们在Stack Overflow上找到一个类似的问题:https://stackoverflow.com/questions/42500383/

10-12 21:10