Skip to content
This repository has been archived by the owner on Aug 5, 2022. It is now read-only.

Solver Prototxt

Jacek Czaja edited this page Nov 25, 2016 · 23 revisions

The solver.prototxt is a configuration file used to tell caffe how you want the network trained.

Parameters

base_lr

This parameter indicates the base (beginning) learning rate of the network. The value is a real number (floating point).

lr_policy

This parameter indicates how the learning rate should change over time. This value is a quoted string.

engine

This parameter sets engine to be used for model to be trained. Possible values are: "MKL2017", "MKLDNN", "CAFFE".

Example of lenet solver using MKL2017 engine:

net: "examples/mnist/lenet_train_test.prototxt" test_iter: 100 test_interval: 500 base_lr: 0.01 momentum: 0.9 weight_decay: 0.0005 lr_policy: "inv" gamma: 0.0001 power: 0.75 display: 100 max_iter: 10000 snapshot: 5000 snapshot_prefix: "examples/mnist/lenet" solver_mode: CPU engine: "MKL2017"

Clone this wiki locally