caffe的python接口生成solver文件详解学习

# -*- coding: utf-8 -*-
"""
Created on Sun Jul 17 18:20:57 2016
@author: root
"""
path='/home/xxx/data/'
solver_file=path+'solver.prototxt'     #solver文件保存位置
sp={}
sp['train_net']=‘“'+path+'train.prototxt”'  # 训练配置文件
sp['test_net']=‘“'+path+'val.prototxt”'     # 测试配置文件
sp['test_iter']='313'                  # 测试迭代次数
sp['test_interval']='782'              # 测试间隔
sp['base_lr']='0.001'                  # 基础学习率
sp['display']='782'                    # 屏幕日志显示间隔
sp['max_iter']='78200'                 # 最大迭代次数
sp['lr_policy']='“step”'                 # 学习率变化规律
sp['gamma']='0.1'                      # 学习率变化指数
sp['momentum']='0.9'                   # 动量
sp['weight_decay']='0.0005'            # 权值衰减
sp['stepsize']='26067'                 # 学习率变化频率
sp['snapshot']='7820'                   # 保存model间隔
sp['snapshot_prefix']=‘"snapshot"'       # 保存的model前缀
sp['solver_mode']='GPU'                # 是否使用gpu
sp['solver_type']='SGD'                # 优化算法
def write_solver():
    #写入文件
    with open(solver_file, 'w') as f:
        for key, value in sorted(sp.items()):
            if not(type(value) is str):
                raise TypeError('All solver parameters must be strings')
            f.write('%s: %s\n' % (key, value))
if __name__ == '__main__':
    write_solver()

# -*- coding: utf-8 -*-
from caffe.proto import caffe_pb2
s = caffe_pb2.SolverParameter()
path='/home/xxx/data/'
solver_file=path+'solver1.prototxt'
s.train_net = path+'train.prototxt'
s.test_net.append(path+'val.prototxt')
s.test_interval = 782  
s.test_iter.append(313) 
s.max_iter = 78200 
s.base_lr = 0.001 
s.momentum = 0.9
s.weight_decay = 5e-4
s.lr_policy = 'step'
s.stepsize=26067
s.gamma = 0.1
s.display = 782
s.snapshot = 7820
s.snapshot_prefix = 'shapshot'
s.type = “SGD”
s.solver_mode = caffe_pb2.SolverParameter.GPU
with open(solver_file, 'w') as f:
    f.write(str(s))

import caffe
caffe.set_device(0)
caffe.set_mode_gpu()
solver = caffe.SGDSolver('/home/xxx/data/solver.prototxt')
solver.solve()