pycaffe中SGDSolver和get_solver方法有什么区别？
What is the difference of SGDSolver and get_solver method in pycaffe?
我正在尝试在 pycaffe 上实现 DNN (mnist)。
pretrained_solver = caffe.SGDSolver('mnist/lenet_solver.prototxt')
pretrained_solver = caffe.get_solver('mnist/lenet_solver.prototxt')
上面两行有什么区别？
似乎我在任何一行中都得到了相同的结果...
根据https://github.com/BVLC/caffe/issues/3033,
The first (get_solver) allows you to specify the solver type in the
prototxt.
The second (SGDSolver) is less reliable because if you specify a
different solver type in the prototxt, then it will still use
SGDSolver and not tell you that you have a mismatch.
我正在尝试在 pycaffe 上实现 DNN (mnist)。
pretrained_solver = caffe.SGDSolver('mnist/lenet_solver.prototxt')
pretrained_solver = caffe.get_solver('mnist/lenet_solver.prototxt')
上面两行有什么区别？ 似乎我在任何一行中都得到了相同的结果...
根据https://github.com/BVLC/caffe/issues/3033,
The first (get_solver) allows you to specify the solver type in the prototxt.
The second (SGDSolver) is less reliable because if you specify a different solver type in the prototxt, then it will still use SGDSolver and not tell you that you have a mismatch.