CaffeCaffe程序员

Caffe的solver参数介绍

2017-03-01  本文已影响63人  SnailTyan

文章作者:Tyan
博客:noahsnail.com  |  CSDN  |  简书

1. Parameters

solver.prototxt文件是用来告诉caffe如何训练网络的。solver.prototxt的各个参数的解释如下:

2. Demo

# lr_policy为multisetp
base_lr: 0.01
momentum: 0.9
lr_policy: "multistep"
gamma: 0.9
stepvalue: 1000
stepvalue: 2000
stepvalue: 3000
stepvalue: 4000
stepvalue: 5000

# lr_policy为step
base_lr: 0.01
momentum: 0.9
lr_policy: "step"
gamma: 0.9
stepsize: 1000
net: "models/bvlc_alexnet/train_val.prototxt"
# 每次测试时进行1000次迭代
test_iter: 1000
# 每进行1000次训练执行一次测试
test_interval: 1000
base_lr: 0.01
lr_policy: "step"
gamma: 0.1
stepsize: 100000
display: 20
max_iter: 450000
momentum: 0.9
weight_decay: 0.0005
snapshot: 10000
snapshot_prefix: "models/bvlc_alexnet/caffe_alexnet_train"
solver_mode: GPU

参考资料

  1. http://www.cnblogs.com/denny402/p/5074049.html
  2. https://github.com/BVLC/caffe/wiki/Solver-Prototxt
  3. http://stackoverflow.com/questions/30033096/what-is-lr-policy-in-caffe
上一篇 下一篇

猜你喜欢

热点阅读