From d9f1b0b16edeb59281355a855e18a8be343fc33c Mon Sep 17 00:00:00 2001
From: Joseph Redmon <pjreddie@gmail.com>
Date: Fri, 08 Aug 2014 19:04:15 +0000
Subject: [PATCH] probably how maxpool layers should be

---
 src/maxpool_layer.c |  134 ++++++++++++++++++++++++++++++++++----------
 1 files changed, 102 insertions(+), 32 deletions(-)

diff --git a/src/maxpool_layer.c b/src/maxpool_layer.c
index 5a82e0b..08c9f2f 100644
--- a/src/maxpool_layer.c
+++ b/src/maxpool_layer.c
@@ -6,7 +6,7 @@
     int h = (layer.h-1)/layer.stride + 1;
     int w = (layer.w-1)/layer.stride + 1;
     int c = layer.c;
-    return double_to_image(h,w,c,layer.output);
+    return float_to_image(h,w,c,layer.output);
 }
 
 image get_maxpool_delta(maxpool_layer layer)
@@ -14,56 +14,126 @@
     int h = (layer.h-1)/layer.stride + 1;
     int w = (layer.w-1)/layer.stride + 1;
     int c = layer.c;
-    return double_to_image(h,w,c,layer.delta);
+    return float_to_image(h,w,c,layer.delta);
 }
 
-maxpool_layer *make_maxpool_layer(int h, int w, int c, int stride)
+maxpool_layer *make_maxpool_layer(int batch, int h, int w, int c, int size, int stride)
 {
-    printf("Maxpool Layer: %d x %d x %d image, %d stride\n", h,w,c,stride);
+    fprintf(stderr, "Maxpool Layer: %d x %d x %d image, %d size, %d stride\n", h,w,c,size,stride);
     maxpool_layer *layer = calloc(1, sizeof(maxpool_layer));
+    layer->batch = batch;
     layer->h = h;
     layer->w = w;
     layer->c = c;
+    layer->size = size;
     layer->stride = stride;
-    layer->output = calloc(((h-1)/stride+1) * ((w-1)/stride+1) * c, sizeof(double));
-    layer->delta = calloc(((h-1)/stride+1) * ((w-1)/stride+1) * c, sizeof(double));
+    layer->output = calloc(((h-1)/stride+1) * ((w-1)/stride+1) * c*batch, sizeof(float));
+    layer->delta = calloc(((h-1)/stride+1) * ((w-1)/stride+1) * c*batch, sizeof(float));
     return layer;
 }
 
-void forward_maxpool_layer(const maxpool_layer layer, double *in)
+void resize_maxpool_layer(maxpool_layer *layer, int h, int w, int c)
 {
-    image input = double_to_image(layer.h, layer.w, layer.c, in);
-    image output = get_maxpool_image(layer);
-    int i,j,k;
-    for(i = 0; i < output.h*output.w*output.c; ++i) output.data[i] = -DBL_MAX;
-    for(k = 0; k < input.c; ++k){
-        for(i = 0; i < input.h; ++i){
-            for(j = 0; j < input.w; ++j){
-                double val = get_pixel(input, i, j, k);
-                double cur = get_pixel(output, i/layer.stride, j/layer.stride, k);
-                if(val > cur) set_pixel(output, i/layer.stride, j/layer.stride, k, val);
+    layer->h = h;
+    layer->w = w;
+    layer->c = c;
+    layer->output = realloc(layer->output, ((h-1)/layer->stride+1) * ((w-1)/layer->stride+1) * c * layer->batch* sizeof(float));
+    layer->delta = realloc(layer->delta, ((h-1)/layer->stride+1) * ((w-1)/layer->stride+1) * c * layer->batch*sizeof(float));
+}
+
+float get_max_region(image im, int h, int w, int c, int size)
+{
+    int i,j;
+    int lower = (-size-1)/2 + 1;
+    int upper = size/2 + 1;
+    
+    int lh = (h-lower < 0)      ? 0 : h-lower;
+    int uh = (h+upper > im.h)   ? im.h : h+upper;
+
+    int lw = (w-lower < 0)      ? 0 : w-lower;
+    int uw = (w+upper > im.w)   ? im.w : w+upper;
+    
+    //printf("%d\n", -3/2);
+    //printf("%d %d\n", lower, upper);
+    //printf("%d %d %d %d\n", lh, uh, lw, uw);
+    
+    float max = -FLT_MAX;
+    for(i = lh; i < uh; ++i){
+        for(j = lw; j < uw; ++j){
+            float val = get_pixel(im, i, j, c);
+            if (val > max) max = val;
+        }
+    }
+    return max;
+}
+
+void forward_maxpool_layer(const maxpool_layer layer, float *in)
+{
+    int b;
+    for(b = 0; b < layer.batch; ++b){
+        image input = float_to_image(layer.h, layer.w, layer.c, in+b*layer.h*layer.w*layer.c);
+
+        int h = (layer.h-1)/layer.stride + 1;
+        int w = (layer.w-1)/layer.stride + 1;
+        int c = layer.c;
+        image output = float_to_image(h,w,c,layer.output+b*h*w*c);
+
+        int i,j,k;
+        for(k = 0; k < input.c; ++k){
+            for(i = 0; i < input.h; i += layer.stride){
+                for(j = 0; j < input.w; j += layer.stride){
+                    float max = get_max_region(input, i, j, k, layer.size);
+                    set_pixel(output, i/layer.stride, j/layer.stride, k, max);
+                }
             }
         }
     }
 }
 
-void backward_maxpool_layer(const maxpool_layer layer, double *in, double *delta)
+float set_max_region_delta(image im, image delta, int h, int w, int c, int size, float max, float error)
 {
-    image input = double_to_image(layer.h, layer.w, layer.c, in);
-    image input_delta = double_to_image(layer.h, layer.w, layer.c, delta);
-    image output_delta = get_maxpool_delta(layer);
-    image output = get_maxpool_image(layer);
-    int i,j,k;
-    for(k = 0; k < input.c; ++k){
-        for(i = 0; i < input.h; ++i){
-            for(j = 0; j < input.w; ++j){
-                double val = get_pixel(input, i, j, k);
-                double cur = get_pixel(output, i/layer.stride, j/layer.stride, k);
-                double d = get_pixel(output_delta, i/layer.stride, j/layer.stride, k);
-                if(val == cur) {
-                    set_pixel(input_delta, i, j, k, d);
+    int i,j;
+    int lower = (-size-1)/2 + 1;
+    int upper = size/2 + 1;
+    
+    int lh = (h-lower < 0)      ? 0 : h-lower;
+    int uh = (h+upper > im.h)   ? im.h : h+upper;
+
+    int lw = (w-lower < 0)      ? 0 : w-lower;
+    int uw = (w+upper > im.w)   ? im.w : w+upper;
+    
+    for(i = lh; i < uh; ++i){
+        for(j = lw; j < uw; ++j){
+            float val = get_pixel(im, i, j, c);
+            if (val == max){
+               add_pixel(delta, i, j, c, error);
+            }
+        }
+    }
+    return max;
+}
+
+void backward_maxpool_layer(const maxpool_layer layer, float *in, float *delta)
+{
+    int b;
+    for(b = 0; b < layer.batch; ++b){
+        image input = float_to_image(layer.h, layer.w, layer.c, in+b*layer.h*layer.w*layer.c);
+        image input_delta = float_to_image(layer.h, layer.w, layer.c, delta+b*layer.h*layer.w*layer.c);
+        int h = (layer.h-1)/layer.stride + 1;
+        int w = (layer.w-1)/layer.stride + 1;
+        int c = layer.c;
+        image output = float_to_image(h,w,c,layer.output+b*h*w*c);
+        image output_delta = float_to_image(h,w,c,layer.delta+b*h*w*c);
+        zero_image(input_delta);
+
+        int i,j,k;
+        for(k = 0; k < input.c; ++k){
+            for(i = 0; i < input.h; i += layer.stride){
+                for(j = 0; j < input.w; j += layer.stride){
+                    float max = get_pixel(output, i/layer.stride, j/layer.stride, k);
+                    float error = get_pixel(output_delta, i/layer.stride, j/layer.stride, k);
+                    set_max_region_delta(input, input_delta, i, j, k, layer.size, max, error);
                 }
-                else set_pixel(input_delta, i, j, k, 0);
             }
         }
     }

--
Gitblit v1.10.0