Caffe2 - Python API
A deep learning, cross platform ML framework
Public Member Functions | Public Attributes | List of all members
caffe2.python.optimizer.YellowFinOptimizer Class Reference
Inheritance diagram for caffe2.python.optimizer.YellowFinOptimizer:

Public Member Functions

def __init__ (self, alpha=0.1, mu=0.0, beta=0.999, curv_win_width=20, zero_debias=True, epsilon=0.1 **6, policy='fixed', sparse_dedup_aggregator=None, kwargs)
def scale_learning_rate (self, scale)
- Public Member Functions inherited from caffe2.python.optimizer.Optimizer
def __init__ (self)
def __call__ (self, net, param_init_net, param, grad=None)
def get_cpu_blob_name (self, base_str, node_name='')
def get_gpu_blob_name (self, base_str, gpu_id, node_name)
def make_unique_blob_name (self, base_str)
def build_lr (self, net, param_init_net, base_learning_rate, learning_rate_blob=None, policy="fixed", iter_val=0, kwargs)
def add_lr_multiplier (self, lr_multiplier)
def get_auxiliary_parameters (self)
def scale_learning_rate (self, args, kwargs)
def create_lars_inputs (self, param_init_net, weight_decay, trust, lr_max)

Public Attributes


Additional Inherited Members

- Static Public Member Functions inherited from caffe2.python.optimizer.Optimizer
def dedup (net, sparse_dedup_aggregator, grad)

Detailed Description

YellowFin: An automatic tuner for momentum SGD

See for more details. This implementation
has separate learning rate and momentum per each parameter.

Definition at line 1012 of file

The documentation for this class was generated from the following file: