1 . blob
看blob.hpp
Blob(const int num, const int channels, const int height, const int width);
2 . conv layer参数
blobs_lr: 1 # learning rate multiplier for the filters
blobs_lr: 2 # learning rate multiplier for the biases
weight_decay: 1 # weight decay multiplier for the filters [Regularization]
weight_decay: 0 # weight decay multiplier for the biases
weight_filler [default type: 'constant' value: 0]:参数的初始化方法
bias_filler:偏置的初始化方法
通过卷积后的大小变化
输入:
n∗ci∗hi∗wi
输出:
n∗co∗ho∗wo
,
其中
ho=(hi+2∗P−K)S+1
,
wo
通过同样的方法计算
P: pad, K: kernel size, S: stride
- inner_product_param
num_output: 1000