SHOGUN  4.1.0
 All Classes Namespaces Files Functions Variables Typedefs Enumerations Enumerator Friends Macros Modules Pages
AdaGradUpdater.cpp
Go to the documentation of this file.
1 /*
2  * Copyright (c) The Shogun Machine Learning Toolbox
3  * Written (w) 2015 Wu Lin
4  * All rights reserved.
5  *
6  * Redistribution and use in source and binary forms, with or without
7  * modification, are permitted provided that the following conditions are met:
8  *
9  * 1. Redistributions of source code must retain the above copyright notice, this
10  * list of conditions and the following disclaimer.
11  * 2. Redistributions in binary form must reproduce the above copyright notice,
12  * this list of conditions and the following disclaimer in the documentation
13  * and/or other materials provided with the distribution.
14  *
15  * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
16  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
17  * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
18  * DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR
19  * ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
20  * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
21  * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
22  * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
23  * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
24  * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
25  *
26  * The views and conclusions contained in the software and documentation are those
27  * of the authors and should not be interpreted as representing official policies,
28  * either expressed or implied, of the Shogun Development Team.
29  *
30  */
31 
33 #include <shogun/lib/config.h>
34 using namespace shogun;
35 
38 {
39  init();
40 }
41 
44 {
45  init();
46  set_learning_rate(learning_rate);
47  set_epsilon(epsilon);
48 }
49 
51 {
52  REQUIRE(learning_rate>0,"Learning_rate (%f) must be positive\n",
53  learning_rate);
54  m_build_in_learning_rate=learning_rate;
55 }
56 
58 {
59  REQUIRE(epsilon>=0,"Epsilon (%f) must be non-negative\n",
60  epsilon);
62 }
63 
65 {
66 }
67 
68 void AdaGradUpdater::init()
69 {
70  m_epsilon=1e-6;
73 }
74 
76 {
78  REQUIRE(context, "Context must set\n");
80  std::copy(m_gradient_accuracy.vector,
82  value.vector);
83  std::string key="AdaGradUpdater::m_gradient_accuracy";
84  context->save_data(key, value);
85 }
86 
88 {
90  REQUIRE(context, "Context must set\n");
91  std::string key="AdaGradUpdater::m_gradient_accuracy";
94  std::copy(value.vector, value.vector+value.vlen,
96 }
97 
99  float64_t gradient, index_t idx, float64_t learning_rate)
100 {
101  REQUIRE(idx>=0 && idx<m_gradient_accuracy.vlen, "The index (%d) is invalid\n", idx);
102  float64_t scale=m_gradient_accuracy[idx]+gradient*gradient;
105  return res;
106 }
107 
109  SGVector<float64_t> raw_negative_descend_direction, float64_t learning_rate)
110 {
111  REQUIRE(variable_reference.vlen==raw_negative_descend_direction.vlen,
112  "The length of variable (%d) and the length of negative descend direction (%d) do not match\n",
113  variable_reference.vlen, raw_negative_descend_direction.vlen);
115  {
116  m_gradient_accuracy=SGVector<float64_t>(variable_reference.vlen);
118  }
120  raw_negative_descend_direction, learning_rate);
121 }
virtual void save_data(const std::string &key, SGVector< float64_t > value)
virtual void load_from_context(CMinimizerContext *context)
int32_t index_t
Definition: common.h:62
The class is used to serialize and deserialize variables for the optimization framework.
#define REQUIRE(x,...)
Definition: SGIO.h:206
virtual SGVector< float64_t > get_data_sgvector_float64(const std::string &key)
static const float64_t epsilon
Definition: libbmrm.cpp:25
virtual float64_t get_negative_descend_direction(float64_t variable, float64_t gradient, index_t idx, float64_t learning_rate)
index_t vlen
Definition: SGVector.h:494
double float64_t
Definition: common.h:50
virtual void update_variable(SGVector< float64_t > variable_reference, SGVector< float64_t > raw_negative_descend_direction, float64_t learning_rate)
SGVector< float64_t > m_gradient_accuracy
float64_t m_build_in_learning_rate
virtual void load_from_context(CMinimizerContext *context)
virtual void update_variable(SGVector< float64_t > variable_reference, SGVector< float64_t > raw_negative_descend_direction, float64_t learning_rate)
all of classes and functions are contained in the shogun namespace
Definition: class_list.h:18
This is a base class for descend update with descend based correction.
void scale(Matrix A, Matrix B, typename Matrix::Scalar alpha)
Definition: Core.h:93
static float32_t sqrt(float32_t x)
Definition: Math.h:459
virtual void set_learning_rate(float64_t learning_rate)
void set_const(T const_elem)
Definition: SGVector.cpp:152
virtual void set_epsilon(float64_t epsilon)
virtual void update_context(CMinimizerContext *context)
virtual void update_context(CMinimizerContext *context)

SHOGUN Machine Learning Toolbox - Documentation