So I have this new programming paradigm.......

Joseph Redmon

2016-09-25 481b57a96a9ef29b112caec1bb3e17ffb043ceae

 src/batchnorm_layer.c

@@ -28,7 +28,13 @@

    layer.rolling_mean = calloc(c, sizeof(float));
    layer.rolling_variance = calloc(c, sizeof(float));

    layer.forward = forward_batchnorm_layer;
    layer.backward = backward_batchnorm_layer;
#ifdef GPU
    layer.forward_gpu = forward_batchnorm_layer_gpu;
    layer.backward_gpu = backward_batchnorm_layer_gpu;

    layer.output_gpu =  cuda_make_array(layer.output, h * w * c * batch);
    layer.delta_gpu =   cuda_make_array(layer.delta, h * w * c * batch);

@@ -135,6 +141,20 @@
}

#ifdef GPU

void pull_batchnorm_layer(layer l)
{
    cuda_pull_array(l.scales_gpu, l.scales, l.c);
    cuda_pull_array(l.rolling_mean_gpu, l.rolling_mean, l.c);
    cuda_pull_array(l.rolling_variance_gpu, l.rolling_variance, l.c);
}
void push_batchnorm_layer(layer l)
{
    cuda_push_array(l.scales_gpu, l.scales, l.c);
    cuda_push_array(l.rolling_mean_gpu, l.rolling_mean, l.c);
    cuda_push_array(l.rolling_variance_gpu, l.rolling_variance, l.c);
}

void forward_batchnorm_layer_gpu(layer l, network_state state)
{
    if(l.type == BATCHNORM) copy_ongpu(l.outputs*l.batch, state.input, 1, l.output_gpu, 1);