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
120
121
122
|
// Copyright (C) 2003--2004 Ronan Collobert (collober@idiap.ch)
//
// This file is part of Torch 3.1.
//
// All rights reserved.
//
// Redistribution and use in source and binary forms, with or without
// modification, are permitted provided that the following conditions
// are met:
// 1. Redistributions of source code must retain the above copyright
// notice, this list of conditions and the following disclaimer.
// 2. Redistributions in binary form must reproduce the above copyright
// notice, this list of conditions and the following disclaimer in the
// documentation and/or other materials provided with the distribution.
// 3. The name of the author may not be used to endorse or promote products
// derived from this software without specific prior written permission.
//
// THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
// IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
// OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
// IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
// INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
// NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
// THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
#include "MLP.h"
#include "Linear.h"
#include "Tanh.h"
#include "Sigmoid.h"
#include "SoftMax.h"
#include "LogSoftMax.h"
#include "Exp.h"
#include "SoftPlus.h"
namespace Torch {
MLP::MLP(int n_layers_, int n_inputs_, ...)
{
n_layers = n_layers_;
layers = (GradientMachine **)allocator->alloc(sizeof(GradientMachine *)*n_layers);
is_linear = (bool *)allocator->alloc(sizeof(bool)*n_layers);
for(int i = 0; i < n_layers; i++)
is_linear[i] = false;
va_list args;
va_start(args, n_inputs_);
for(int i = 0; i < n_layers; i++)
{
char *layer_type = va_arg(args, char *);
int n_outputs_ = va_arg(args, int);
bool is_valid = false;
if(!strcmp(layer_type, "linear"))
{
layers[i] = new(allocator) Linear(n_inputs_, n_outputs_);
is_linear[i] = true;
is_valid = true;
}
if(!strcmp(layer_type, "tanh"))
{
layers[i] = new(allocator) Tanh(n_outputs_);
is_valid = true;
}
if(!strcmp(layer_type, "sigmoid"))
{
layers[i] = new(allocator) Sigmoid(n_outputs_);
is_valid = true;
}
if(!strcmp(layer_type, "softmax"))
{
layers[i] = new(allocator) SoftMax(n_outputs_);
is_valid = true;
}
if(!strcmp(layer_type, "log-softmax"))
{
layers[i] = new(allocator) LogSoftMax(n_outputs_);
is_valid = true;
}
if(!strcmp(layer_type, "exp"))
{
layers[i] = new(allocator) Exp(n_outputs_);
is_valid = true;
}
if(!strcmp(layer_type, "softplus"))
{
layers[i] = new(allocator) SoftPlus(n_outputs_);
is_valid = true;
}
if(!is_valid)
error("MLP: unknow layer type <%s>", layer_type);
this->addFCL(layers[i]);
n_inputs_ = n_outputs_;
}
build();
va_end(args);
}
void MLP::setWeightDecay(real weight_decay)
{
for(int i = 0; i < n_layers; i++)
{
if(is_linear[i])
layers[i]->setROption("weight decay", weight_decay);
}
}
MLP::~MLP()
{
}
}
|
