Caffe中常用的优化器有哪些
- SGD (Stochastic Gradient Descent)
- Adam (Adaptive Moment Estimation)
- RMSprop (Root Mean Square Propagation)
- Adagrad (Adaptive Gradient Algorithm)
- Adadelta
- Adamax
- Nadam (Nesterov-accelerated Adaptive Moment Estimation)
- RAdam (Rectified Adam)
- Yogi
- AMSGrad (Adaptive Moment Estimation with rectified gradients)