Caffe2 - Python API
A deep learning, cross platform ML framework
Public Member Functions | Static Public Member Functions | List of all members
caffe2.python.optimizer.Optimizer Class Reference
Inheritance diagram for caffe2.python.optimizer.Optimizer:
caffe2.python.optimizer.AdadeltaOptimizer caffe2.python.optimizer.AdagradOptimizer caffe2.python.optimizer.AdamOptimizer caffe2.python.optimizer.FtrlOptimizer caffe2.python.optimizer.GFtrlOptimizer caffe2.python.optimizer.RmsPropOptimizer caffe2.python.optimizer.SgdOptimizer caffe2.python.optimizer.WeightDecayBuilder caffe2.python.optimizer.WngradOptimizer caffe2.python.optimizer.YellowFinOptimizer

Public Member Functions

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)

Static Public Member Functions

def dedup (net, sparse_dedup_aggregator, grad)

Detailed Description

Definition at line 30 of file

Member Function Documentation

def caffe2.python.optimizer.Optimizer.add_lr_multiplier (   self,
Set the global learning rate multiplier. If a multiplier already
existed, this will overwrite the existing multiplier. The multiplier is
used for all future calls to _run(), unless it is overwritten.

Definition at line 147 of file

def caffe2.python.optimizer.Optimizer.get_auxiliary_parameters (   self)
Returns a list of auxiliary parameters.

    aux_params: A namedtuple, AuxParams.

    aux_params.local stores a list of blobs. Each blob is a local
    auxiliary parameter. A local auxiliary parameter is a parameter in
    parallel to a learning rate parameter. Take adagrad as an example,
    the local auxiliary parameter is the squared sum parameter, because
    every learning rate has a squared sum associated with it.

    aux_params.shared also stores a list of blobs. Each blob is a shared
    auxiliary parameter. A shared auxiliary parameter is a parameter
    that is shared across all the learning rate parameters. Take adam as
    an example, the iteration parameter is a shared parameter, because
    all the learning rates share the same iteration parameter.

Definition at line 181 of file

def caffe2.python.optimizer.Optimizer.make_unique_blob_name (   self,
Returns a blob name that will be unique to the current device
and optimizer instance.

Definition at line 75 of file

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