add SGDR policy

This commit is contained in:
Josh Veitch-Michaelis 2019-03-18 23:26:04 +00:00
parent 8bcba6c105
commit d64693eb77
3 changed files with 14 additions and 1 deletions

View File

@ -518,7 +518,7 @@ struct layer {
// network.h
typedef enum {
CONSTANT, STEP, EXP, POLY, STEPS, SIG, RANDOM
CONSTANT, STEP, EXP, POLY, STEPS, SIG, RANDOM, SGDR
} learning_rate_policy;
// network.h
@ -534,6 +534,9 @@ typedef struct network {
learning_rate_policy policy;
float learning_rate;
float learning_rate_min;
float learning_rate_max;
int batches_per_cycle;
float momentum;
float decay;
float gamma;

View File

@ -117,6 +117,12 @@ float get_current_rate(network net)
return net.learning_rate * pow(rand_uniform(0,1), net.power);
case SIG:
return net.learning_rate * (1./(1.+exp(net.gamma*(batch_num - net.step))));
case SGDR:
rate = net.learning_rate_min +
0.5*(net.learning_rate_max-net.learning_rate_min)
* (1. + cos( (float) (batch_num % net.batches_per_cycle)*3.14159265 / net.batches_per_cycle));
return rate;
default:
fprintf(stderr, "Policy is weird!\n");
return net.learning_rate;

View File

@ -629,6 +629,7 @@ learning_rate_policy get_policy(char *s)
if (strcmp(s, "exp")==0) return EXP;
if (strcmp(s, "sigmoid")==0) return SIG;
if (strcmp(s, "steps")==0) return STEPS;
if (strcmp(s, "sgdr")==0) return SGDR;
fprintf(stderr, "Couldn't find policy %s, going with constant\n", s);
return CONSTANT;
}
@ -637,6 +638,9 @@ void parse_net_options(list *options, network *net)
{
net->batch = option_find_int(options, "batch",1);
net->learning_rate = option_find_float(options, "learning_rate", .001);
net->learning_rate_min = option_find_float_quiet(options, "learning_rate_min", .00001);
net->learning_rate_max = option_find_float_quiet(options, "learning_rate_max", .001);
net->batches_per_cycle = option_find_int(options, "sgdr_cycle", 500);
net->momentum = option_find_float(options, "momentum", .9);
net->decay = option_find_float(options, "decay", .0001);
int subdivs = option_find_int(options, "subdivisions",1);