dropout_layer.c
1.57 KB
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
#include "dropout_layer.h"
#include "utils.h"
#include "cuda.h"
#include <stdlib.h>
#include <stdio.h>
dropout_layer make_dropout_layer(int batch, int inputs, float probability)
{
dropout_layer l = {0};
l.type = DROPOUT;
l.probability = probability;
l.inputs = inputs;
l.outputs = inputs;
l.batch = batch;
l.rand = calloc(inputs*batch, sizeof(float));
l.scale = 1./(1.-probability);
l.forward = forward_dropout_layer;
l.backward = backward_dropout_layer;
#ifdef GPU
l.forward_gpu = forward_dropout_layer_gpu;
l.backward_gpu = backward_dropout_layer_gpu;
l.rand_gpu = cuda_make_array(l.rand, inputs*batch);
#endif
fprintf(stderr, "dropout p = %.2f %4d -> %4d\n", probability, inputs, inputs);
return l;
}
void resize_dropout_layer(dropout_layer *l, int inputs)
{
l->rand = realloc(l->rand, l->inputs*l->batch*sizeof(float));
#ifdef GPU
cuda_free(l->rand_gpu);
l->rand_gpu = cuda_make_array(l->rand, inputs*l->batch);
#endif
}
void forward_dropout_layer(dropout_layer l, network net)
{
int i;
if (!net.train) return;
for(i = 0; i < l.batch * l.inputs; ++i){
float r = rand_uniform(0, 1);
l.rand[i] = r;
if(r < l.probability) net.input[i] = 0;
else net.input[i] *= l.scale;
}
}
void backward_dropout_layer(dropout_layer l, network net)
{
int i;
if(!net.delta) return;
for(i = 0; i < l.batch * l.inputs; ++i){
float r = l.rand[i];
if(r < l.probability) net.delta[i] = 0;
else net.delta[i] *= l.scale;
}
}