| | |
| | | |
| | | l.biases = calloc(n, sizeof(float)); |
| | | l.bias_updates = calloc(n, sizeof(float)); |
| | | float scale = 1./sqrt(size*size*c); |
| | | //float scale = 1./sqrt(size*size*c); |
| | | float scale = sqrt(2./(size*size*c)); |
| | | for(i = 0; i < c*n*size*size; ++i) l.filters[i] = 2*scale*rand_uniform() - scale; |
| | | for(i = 0; i < n; ++i){ |
| | | l.biases[i] = scale; |
| | |
| | | return l; |
| | | } |
| | | |
| | | void resize_convolutional_layer(convolutional_layer *l, int h, int w) |
| | | void resize_convolutional_layer(convolutional_layer *l, int w, int h) |
| | | { |
| | | l->h = h; |
| | | l->w = w; |
| | | int out_h = convolutional_out_height(*l); |
| | | l->h = h; |
| | | int out_w = convolutional_out_width(*l); |
| | | int out_h = convolutional_out_height(*l); |
| | | |
| | | l->out_w = out_w; |
| | | l->out_h = out_h; |
| | | |
| | | l->outputs = l->out_h * l->out_w * l->out_c; |
| | | l->inputs = l->w * l->h * l->c; |
| | | |
| | | l->col_image = realloc(l->col_image, |
| | | out_h*out_w*l->size*l->size*l->c*sizeof(float)); |
| | |
| | | cuda_free(l->delta_gpu); |
| | | cuda_free(l->output_gpu); |
| | | |
| | | l->col_image_gpu = cuda_make_array(l->col_image, out_h*out_w*l->size*l->size*l->c); |
| | | l->delta_gpu = cuda_make_array(l->delta, l->batch*out_h*out_w*l->n); |
| | | l->output_gpu = cuda_make_array(l->output, l->batch*out_h*out_w*l->n); |
| | | l->col_image_gpu = cuda_make_array(0, out_h*out_w*l->size*l->size*l->c); |
| | | l->delta_gpu = cuda_make_array(0, l->batch*out_h*out_w*l->n); |
| | | l->output_gpu = cuda_make_array(0, l->batch*out_h*out_w*l->n); |
| | | #endif |
| | | } |
| | | |
| | |
| | | gradient_array(l.output, m*k*l.batch, l.activation, l.delta); |
| | | backward_bias(l.bias_updates, l.delta, l.batch, l.n, k); |
| | | |
| | | if(state.delta) memset(state.delta, 0, l.batch*l.h*l.w*l.c*sizeof(float)); |
| | | |
| | | for(i = 0; i < l.batch; ++i){ |
| | | float *a = l.delta + i*m*k; |
| | | float *b = l.col_image; |
| | |
| | | return float_to_image(w,h,c,l.filters+i*h*w*c); |
| | | } |
| | | |
| | | void rgbgr_filters(convolutional_layer l) |
| | | { |
| | | int i; |
| | | for(i = 0; i < l.n; ++i){ |
| | | image im = get_convolutional_filter(l, i); |
| | | if (im.c == 3) { |
| | | rgbgr_image(im); |
| | | } |
| | | } |
| | | } |
| | | |
| | | void rescale_filters(convolutional_layer l, float scale, float trans) |
| | | { |
| | | int i; |
| | | for(i = 0; i < l.n; ++i){ |
| | | image im = get_convolutional_filter(l, i); |
| | | if (im.c == 3) { |
| | | scale_image(im, scale); |
| | | float sum = sum_array(im.data, im.w*im.h*im.c); |
| | | l.biases[i] += sum*trans; |
| | | } |
| | | } |
| | | } |
| | | |
| | | image *get_filters(convolutional_layer l) |
| | | { |
| | | image *filters = calloc(l.n, sizeof(image)); |
| | | int i; |
| | | for(i = 0; i < l.n; ++i){ |
| | | filters[i] = copy_image(get_convolutional_filter(l, i)); |
| | | normalize_image(filters[i]); |
| | | } |
| | | return filters; |
| | | } |