1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
|
# Copyright 2015 The TensorFlow Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ==============================================================================
"""Adagrad optimizer for TensorFlow."""
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
from tensorflow.python.keras.optimizer_v2 import optimizer_v2
from tensorflow.python.ops import array_ops
from tensorflow.python.ops import gen_array_ops
from tensorflow.python.ops import init_ops
from tensorflow.python.ops import math_ops
from tensorflow.python.training import training_ops
class Adagrad(optimizer_v2.OptimizerV2):
"""Adagrad optimizer.
It is recommended to leave the parameters of this optimizer at their default
values.
See this [paper](http://www.jmlr.org/papers/volume12/duchi11a/duchi11a.pdf)
or this
[intro](https://ppasupat.github.io/a9online/uploads/proximal_notes.pdf).
The learning_rate arg below is a hyperparameter, where a hyperparameter is
defined as a scalar Tensor, a regular Python value, or a callable (which
will be evaluated when `apply_gradients` is called) returning a scalar
Tensor or a Python value.
Arguments:
learning_rate: float hyperparameter >= 0. Learning rate.
initial_accumulator_value: A floating point value. Starting value for the
accumulators, must be positive.
name: Optional name prefix for the operations created when applying
gradients. Defaults to 'Adagrad'.
Raises:
ValueError: If the `initial_accumulator_value` is invalid.
"""
def __init__(self,
learning_rate=0.001,
initial_accumulator_value=0.1,
name="Adagrad"):
if initial_accumulator_value <= 0.0:
raise ValueError("initial_accumulator_value must be positive: %s" %
initial_accumulator_value)
super(Adagrad, self).__init__(name)
self._set_hyper("learning_rate", learning_rate)
self._initial_accumulator_value = initial_accumulator_value
def _create_vars(self, var_list, state):
for v in var_list:
dtype = v.dtype.base_dtype
if v.get_shape().is_fully_defined():
init = init_ops.constant_initializer(self._initial_accumulator_value,
dtype=dtype)
else:
def init(v=v, dtype=dtype):
# Use a Tensor instead of initializer if variable does not have
# static shape.
init_constant = gen_array_ops.fill(array_ops.shape(v),
self._initial_accumulator_value)
return math_ops.cast(init_constant, dtype)
state.create_slot_with_initializer(v, init, v.get_shape(), dtype,
"accumulator")
def _apply_dense(self, grad, var, state):
acc = state.get_slot(var, "accumulator")
return training_ops.apply_adagrad(
var,
acc,
state.get_hyper("learning_rate", var.dtype.base_dtype),
grad,
use_locking=self._use_locking)
def _resource_apply_dense(self, grad, var, state):
acc = state.get_slot(var, "accumulator")
return training_ops.resource_apply_adagrad(
var.handle,
acc.handle,
state.get_hyper("learning_rate", var.dtype.base_dtype),
grad,
use_locking=self._use_locking)
def _apply_sparse(self, grad, var, state):
acc = state.get_slot(var, "accumulator")
return training_ops.sparse_apply_adagrad(
var,
acc,
state.get_hyper("learning_rate", var.dtype.base_dtype),
grad.values,
grad.indices,
use_locking=self._use_locking)
def _resource_apply_sparse(self, grad, var, indices, state):
acc = state.get_slot(var, "accumulator")
return training_ops.resource_sparse_apply_adagrad(
var.handle,
acc.handle,
state.get_hyper("learning_rate", var.dtype.base_dtype),
grad,
indices,
use_locking=self._use_locking)
|
