mirror of
https://github.com/zebrajr/pytorch.git
synced 2025-12-07 00:21:07 +01:00
e76015040a
Summary: Regularization added for caffe2 and dper. This regularization is intended for `dense feature `only. Sparse feature would serve as individual optimizer, see ` D5618405 ` and `D5534579` for details. The implementation of dense regularization is similar to the ones in optimizer. we now support `l1 norm` and ` l2 norm` in regularizer. In dper, we would call different regularization based on regularization type defined in model_definition.thrift. Reviewed By: xianjiec Differential Revision: D5724851 fbshipit-source-id: 0fbee698cfeff1ac477fc9d07785406069f8d9c8
58 lines
1.7 KiB
Python
58 lines
1.7 KiB
Python
# @package optimizer
|
|
# Module caffe2.python.optimizer
|
|
from __future__ import absolute_import
|
|
from __future__ import division
|
|
from __future__ import print_function
|
|
from __future__ import unicode_literals
|
|
|
|
|
|
from caffe2.python import core
|
|
|
|
|
|
class Regularizer(object):
|
|
def __init__(self):
|
|
pass
|
|
|
|
'''
|
|
Adds regularization to train_net for given parameter. Its factor ahead of
|
|
regularization is given when initialization.
|
|
The param should be a BlobReference.
|
|
'''
|
|
|
|
def __call__(self, train_net, param):
|
|
assert isinstance(param, core.BlobReference)
|
|
return self._run(train_net, param)
|
|
|
|
def _run(self, train_net, param):
|
|
raise Exception("Not Impelemented")
|
|
|
|
|
|
class L1Norm(Regularizer):
|
|
def __init__(self, reg_lambda):
|
|
super(L1Norm, self).__init__()
|
|
assert reg_lambda >= 0,\
|
|
'factor ahead of regularization should be 0 or positive'
|
|
|
|
self.reg_lambda = reg_lambda
|
|
|
|
def _run(self, train_net, param):
|
|
output_blob = train_net.NextScopedBlob(param + '_l1_regularization')
|
|
train_net.LpNorm([param], [output_blob], p=1)
|
|
train_net.Scale([output_blob], [output_blob], scale=self.reg_lambda)
|
|
return output_blob
|
|
|
|
|
|
class L2Norm(Regularizer):
|
|
def __init__(self, reg_lambda):
|
|
super(L2Norm, self).__init__()
|
|
assert reg_lambda >= 0,\
|
|
'factor ahead of regularization should be 0 or positive'
|
|
|
|
self.reg_lambda = reg_lambda
|
|
|
|
def _run(self, train_net, param):
|
|
output_blob = train_net.NextScopedBlob(param + '_l2_regularization')
|
|
train_net.LpNorm([param], [output_blob], p=2)
|
|
train_net.Scale([output_blob], [output_blob], scale=self.reg_lambda)
|
|
return output_blob
|