Functions | |
void | backward_batchnorm_layer (Darknet::Layer &l, Darknet::NetworkState state) |
void | backward_batchnorm_layer_gpu (Darknet::Layer &l, Darknet::NetworkState state) |
void | forward_batchnorm_layer (Darknet::Layer &l, Darknet::NetworkState state) |
void | forward_batchnorm_layer_gpu (Darknet::Layer &l, Darknet::NetworkState state) |
Darknet::Layer | make_batchnorm_layer (int batch, int w, int h, int c, int train) |
void | pull_batchnorm_layer (Darknet::Layer &l) |
void | push_batchnorm_layer (Darknet::Layer &l) |
void | resize_batchnorm_layer (Darknet::Layer *l, int w, int h) |
void | update_batchnorm_layer (Darknet::Layer &l, int batch, float learning_rate, float momentum, float decay) |
void | update_batchnorm_layer_gpu (Darknet::Layer &l, int batch, float learning_rate_init, float momentum, float decay, float loss_scale) |
void backward_batchnorm_layer | ( | Darknet::Layer & | l, |
Darknet::NetworkState | state | ||
) |
void backward_batchnorm_layer_gpu | ( | Darknet::Layer & | l, |
Darknet::NetworkState | state | ||
) |
void forward_batchnorm_layer | ( | Darknet::Layer & | l, |
Darknet::NetworkState | state | ||
) |
void forward_batchnorm_layer_gpu | ( | Darknet::Layer & | l, |
Darknet::NetworkState | state | ||
) |
Darknet::Layer make_batchnorm_layer | ( | int | batch, |
int | w, | ||
int | h, | ||
int | c, | ||
int | train | ||
) |
void pull_batchnorm_layer | ( | Darknet::Layer & | l | ) |
void push_batchnorm_layer | ( | Darknet::Layer & | l | ) |
void resize_batchnorm_layer | ( | Darknet::Layer * | l, |
int | w, | ||
int | h | ||
) |
void update_batchnorm_layer | ( | Darknet::Layer & | l, |
int | batch, | ||
float | learning_rate, | ||
float | momentum, | ||
float | decay | ||
) |
void update_batchnorm_layer_gpu | ( | Darknet::Layer & | l, |
int | batch, | ||
float | learning_rate_init, | ||
float | momentum, | ||
float | decay, | ||
float | loss_scale | ||
) |