mirror/darknet
1
0
Fork 0
mirror of https://github.com/pjreddie/darknet.git synced 2023-08-10 21:13:14 +03:00
Code Issues Projects Releases Wiki Activity

renamed sigmoid to logistic

Browse Source
This commit is contained in:
Joseph Redmon 2015-03-08 11:31:12 -07:00
parent 655f636a42
commit f047cfff99
5 changed files with 21 additions and 21 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
src/activation_kernels.cu
View File

@ -4,13 +4,13 @@ extern "C" {
} }
__device__ float linear_activate_kernel(float x){return x;} __device__ float linear_activate_kernel(float x){return x;}
__device__ float sigmoid_activate_kernel(float x){return 1./(1. + exp(-x));} __device__ float logistic_activate_kernel(float x){return 1./(1. + exp(-x));}
__device__ float relu_activate_kernel(float x){return x*(x>0);} __device__ float relu_activate_kernel(float x){return x*(x>0);}
__device__ float ramp_activate_kernel(float x){return x*(x>0)+.1*x;} __device__ float ramp_activate_kernel(float x){return x*(x>0)+.1*x;}
__device__ float tanh_activate_kernel(float x){return (exp(2*x)-1)/(exp(2*x)+1);} __device__ float tanh_activate_kernel(float x){return (exp(2*x)-1)/(exp(2*x)+1);}
__device__ float linear_gradient_kernel(float x){return 1;} __device__ float linear_gradient_kernel(float x){return 1;}
__device__ float sigmoid_gradient_kernel(float x){return (1-x)*x;} __device__ float logistic_gradient_kernel(float x){return (1-x)*x;}
__device__ float relu_gradient_kernel(float x){return (x>0);} __device__ float relu_gradient_kernel(float x){return (x>0);}
__device__ float ramp_gradient_kernel(float x){return (x>0)+.1;} __device__ float ramp_gradient_kernel(float x){return (x>0)+.1;}
__device__ float tanh_gradient_kernel(float x){return 1-x*x;} __device__ float tanh_gradient_kernel(float x){return 1-x*x;}
@ -20,8 +20,8 @@ __device__ float activate_kernel(float x, ACTIVATION a)
switch(a){ switch(a){
case LINEAR: case LINEAR:
return linear_activate_kernel(x); return linear_activate_kernel(x);
case SIGMOID: case LOGISTIC:
return sigmoid_activate_kernel(x); return logistic_activate_kernel(x);
case RELU: case RELU:
return relu_activate_kernel(x); return relu_activate_kernel(x);
case RAMP: case RAMP:
@ -37,8 +37,8 @@ __device__ float gradient_kernel(float x, ACTIVATION a)
switch(a){ switch(a){
case LINEAR: case LINEAR:
return linear_gradient_kernel(x); return linear_gradient_kernel(x);
case SIGMOID: case LOGISTIC:
return sigmoid_gradient_kernel(x); return logistic_gradient_kernel(x);
case RELU: case RELU:
return relu_gradient_kernel(x); return relu_gradient_kernel(x);
case RAMP: case RAMP:

14
src/activations.c
View File

@ -8,8 +8,8 @@
char *get_activation_string(ACTIVATION a) char *get_activation_string(ACTIVATION a)
{ {
switch(a){ switch(a){
case SIGMOID: case LOGISTIC:
return "sigmoid"; return "logistic";
case RELU: case RELU:
return "relu"; return "relu";
case RAMP: case RAMP:
@ -26,7 +26,7 @@ char *get_activation_string(ACTIVATION a)
ACTIVATION get_activation(char *s) ACTIVATION get_activation(char *s)
{ {
if (strcmp(s, "sigmoid")==0) return SIGMOID; if (strcmp(s, "logistic")==0) return LOGISTIC;
if (strcmp(s, "relu")==0) return RELU; if (strcmp(s, "relu")==0) return RELU;
if (strcmp(s, "linear")==0) return LINEAR; if (strcmp(s, "linear")==0) return LINEAR;
if (strcmp(s, "ramp")==0) return RAMP; if (strcmp(s, "ramp")==0) return RAMP;
@ -40,8 +40,8 @@ float activate(float x, ACTIVATION a)
switch(a){ switch(a){
case LINEAR: case LINEAR:
return linear_activate(x); return linear_activate(x);
case SIGMOID: case LOGISTIC:
return sigmoid_activate(x); return logistic_activate(x);
case RELU: case RELU:
return relu_activate(x); return relu_activate(x);
case RAMP: case RAMP:
@ -65,8 +65,8 @@ float gradient(float x, ACTIVATION a)
switch(a){ switch(a){
case LINEAR: case LINEAR:
return linear_gradient(x); return linear_gradient(x);
case SIGMOID: case LOGISTIC:
return sigmoid_gradient(x); return logistic_gradient(x);
case RELU: case RELU:
return relu_gradient(x); return relu_gradient(x);
case RAMP: case RAMP:

6
src/activations.h
View File

@ -3,7 +3,7 @@
#define ACTIVATIONS_H #define ACTIVATIONS_H
typedef enum{ typedef enum{
SIGMOID, RELU, LINEAR, RAMP, TANH LOGISTIC, RELU, LINEAR, RAMP, TANH
}ACTIVATION; }ACTIVATION;
ACTIVATION get_activation(char *s); ACTIVATION get_activation(char *s);
@ -19,13 +19,13 @@ void gradient_array_ongpu(float *x, int n, ACTIVATION a, float *delta);
#endif #endif
static inline float linear_activate(float x){return x;} static inline float linear_activate(float x){return x;}
static inline float sigmoid_activate(float x){return 1./(1. + exp(-x));} static inline float logistic_activate(float x){return 1./(1. + exp(-x));}
static inline float relu_activate(float x){return x*(x>0);} static inline float relu_activate(float x){return x*(x>0);}
static inline float ramp_activate(float x){return x*(x>0)+.1*x;} static inline float ramp_activate(float x){return x*(x>0)+.1*x;}
static inline float tanh_activate(float x){return (exp(2*x)-1)/(exp(2*x)+1);} static inline float tanh_activate(float x){return (exp(2*x)-1)/(exp(2*x)+1);}
static inline float linear_gradient(float x){return 1;} static inline float linear_gradient(float x){return 1;}
static inline float sigmoid_gradient(float x){return (1-x)*x;} static inline float logistic_gradient(float x){return (1-x)*x;}
static inline float relu_gradient(float x){return (x>0);} static inline float relu_gradient(float x){return (x>0);}
static inline float ramp_gradient(float x){return (x>0)+.1;} static inline float ramp_gradient(float x){return (x>0)+.1;}
static inline float tanh_gradient(float x){return 1-x*x;} static inline float tanh_gradient(float x){return 1-x*x;}

4
src/detection_layer.c
View File

@ -53,7 +53,7 @@ void forward_detection_layer(const detection_layer layer, float *in, float *trut
layer.output[out_i++] = scale*in[in_i++]; layer.output[out_i++] = scale*in[in_i++];
} }
softmax_array(layer.output + out_i - layer.classes, layer.classes, layer.output + out_i - layer.classes); softmax_array(layer.output + out_i - layer.classes, layer.classes, layer.output + out_i - layer.classes);
activate_array(in+in_i, layer.coords, SIGMOID); activate_array(in+in_i, layer.coords, LOGISTIC);
for(j = 0; j < layer.coords; ++j){ for(j = 0; j < layer.coords; ++j){
layer.output[out_i++] = mask*in[in_i++]; layer.output[out_i++] = mask*in[in_i++];
} }
@ -75,7 +75,7 @@ void backward_detection_layer(const detection_layer layer, float *in, float *del
delta[in_i++] = scale*layer.delta[out_i++]; delta[in_i++] = scale*layer.delta[out_i++];
} }
gradient_array(layer.output + out_i, layer.coords, SIGMOID, layer.delta + out_i); gradient_array(layer.output + out_i, layer.coords, LOGISTIC, layer.delta + out_i);
for(j = 0; j < layer.coords; ++j){ for(j = 0; j < layer.coords; ++j){
delta[in_i++] = layer.delta[out_i++]; delta[in_i++] = layer.delta[out_i++];
} }

6
src/parser.c
View File

@ -76,7 +76,7 @@ deconvolutional_layer *parse_deconvolutional(list *options, network *net, int co
int n = option_find_int(options, "filters",1); int n = option_find_int(options, "filters",1);
int size = option_find_int(options, "size",1); int size = option_find_int(options, "size",1);
int stride = option_find_int(options, "stride",1); int stride = option_find_int(options, "stride",1);
char *activation_s = option_find_str(options, "activation", "sigmoid"); char *activation_s = option_find_str(options, "activation", "logistic");
ACTIVATION activation = get_activation(activation_s); ACTIVATION activation = get_activation(activation_s);
if(count == 0){ if(count == 0){
learning_rate = option_find_float(options, "learning_rate", .001); learning_rate = option_find_float(options, "learning_rate", .001);
@ -120,7 +120,7 @@ convolutional_layer *parse_convolutional(list *options, network *net, int count)
int size = option_find_int(options, "size",1); int size = option_find_int(options, "size",1);
int stride = option_find_int(options, "stride",1); int stride = option_find_int(options, "stride",1);
int pad = option_find_int(options, "pad",0); int pad = option_find_int(options, "pad",0);
char *activation_s = option_find_str(options, "activation", "sigmoid"); char *activation_s = option_find_str(options, "activation", "logistic");
ACTIVATION activation = get_activation(activation_s); ACTIVATION activation = get_activation(activation_s);
if(count == 0){ if(count == 0){
learning_rate = option_find_float(options, "learning_rate", .001); learning_rate = option_find_float(options, "learning_rate", .001);
@ -161,7 +161,7 @@ connected_layer *parse_connected(list *options, network *net, int count)
int input; int input;
float learning_rate, momentum, decay; float learning_rate, momentum, decay;
int output = option_find_int(options, "output",1); int output = option_find_int(options, "output",1);
char *activation_s = option_find_str(options, "activation", "sigmoid"); char *activation_s = option_find_str(options, "activation", "logistic");
ACTIVATION activation = get_activation(activation_s); ACTIVATION activation = get_activation(activation_s);
if(count == 0){ if(count == 0){
input = option_find_int(options, "input",1); input = option_find_int(options, "input",1);