From ae74d0ef31485f84e1856b4733135d2753dbb033 Mon Sep 17 00:00:00 2001
From: AlexeyAB <alexeyab84@gmail.com>
Date: Sat, 21 Oct 2017 20:56:07 +0000
Subject: [PATCH] Fixed fread() for net->seen
---
src/network.c | 184 ++++++++++++++++-----------------------------
1 files changed, 67 insertions(+), 117 deletions(-)
diff --git a/src/network.c b/src/network.c
index 2960d67..2d755ae 100644
--- a/src/network.c
+++ b/src/network.c
@@ -1,5 +1,6 @@
#include <stdio.h>
#include <time.h>
+#include <assert.h>
#include "network.h"
#include "image.h"
#include "data.h"
@@ -14,11 +15,12 @@
#include "local_layer.h"
#include "convolutional_layer.h"
#include "activation_layer.h"
-#include "deconvolutional_layer.h"
#include "detection_layer.h"
+#include "region_layer.h"
#include "normalization_layer.h"
#include "batchnorm_layer.h"
#include "maxpool_layer.h"
+#include "reorg_layer.h"
#include "avgpool_layer.h"
#include "cost_layer.h"
#include "softmax_layer.h"
@@ -39,7 +41,7 @@
net.momentum = 0;
net.decay = 0;
#ifdef GPU
- if(gpu_index >= 0) update_network_gpu(net);
+ //if(net.gpu_index >= 0) update_network_gpu(net);
#endif
}
@@ -58,12 +60,13 @@
for(i = 0; i < net.num_steps; ++i){
if(net.steps[i] > batch_num) return rate;
rate *= net.scales[i];
- if(net.steps[i] > batch_num - 1) reset_momentum(net);
+ //if(net.steps[i] > batch_num - 1 && net.scales[i] > 1) reset_momentum(net);
}
return rate;
case EXP:
return net.learning_rate * pow(net.gamma, batch_num);
case POLY:
+ if (batch_num < net.burn_in) return net.learning_rate * pow((float)batch_num / net.burn_in, net.power);
return net.learning_rate * pow(1 - (float)batch_num / net.max_batches, net.power);
case RANDOM:
return net.learning_rate * pow(rand_uniform(0,1), net.power);
@@ -96,12 +99,16 @@
return "crnn";
case MAXPOOL:
return "maxpool";
+ case REORG:
+ return "reorg";
case AVGPOOL:
return "avgpool";
case SOFTMAX:
return "softmax";
case DETECTION:
return "detection";
+ case REGION:
+ return "region";
case DROPOUT:
return "dropout";
case CROP:
@@ -145,45 +152,7 @@
if(l.delta){
scal_cpu(l.outputs * l.batch, 0, l.delta, 1);
}
- if(l.type == CONVOLUTIONAL){
- forward_convolutional_layer(l, state);
- } else if(l.type == DECONVOLUTIONAL){
- forward_deconvolutional_layer(l, state);
- } else if(l.type == ACTIVE){
- forward_activation_layer(l, state);
- } else if(l.type == LOCAL){
- forward_local_layer(l, state);
- } else if(l.type == NORMALIZATION){
- forward_normalization_layer(l, state);
- } else if(l.type == BATCHNORM){
- forward_batchnorm_layer(l, state);
- } else if(l.type == DETECTION){
- forward_detection_layer(l, state);
- } else if(l.type == CONNECTED){
- forward_connected_layer(l, state);
- } else if(l.type == RNN){
- forward_rnn_layer(l, state);
- } else if(l.type == GRU){
- forward_gru_layer(l, state);
- } else if(l.type == CRNN){
- forward_crnn_layer(l, state);
- } else if(l.type == CROP){
- forward_crop_layer(l, state);
- } else if(l.type == COST){
- forward_cost_layer(l, state);
- } else if(l.type == SOFTMAX){
- forward_softmax_layer(l, state);
- } else if(l.type == MAXPOOL){
- forward_maxpool_layer(l, state);
- } else if(l.type == AVGPOOL){
- forward_avgpool_layer(l, state);
- } else if(l.type == DROPOUT){
- forward_dropout_layer(l, state);
- } else if(l.type == ROUTE){
- forward_route_layer(l, net);
- } else if(l.type == SHORTCUT){
- forward_shortcut_layer(l, state);
- }
+ l.forward(l, state);
state.input = l.output;
}
}
@@ -195,29 +164,17 @@
float rate = get_current_rate(net);
for(i = 0; i < net.n; ++i){
layer l = net.layers[i];
- if(l.type == CONVOLUTIONAL){
- update_convolutional_layer(l, update_batch, rate, net.momentum, net.decay);
- } else if(l.type == DECONVOLUTIONAL){
- update_deconvolutional_layer(l, rate, net.momentum, net.decay);
- } else if(l.type == CONNECTED){
- update_connected_layer(l, update_batch, rate, net.momentum, net.decay);
- } else if(l.type == RNN){
- update_rnn_layer(l, update_batch, rate, net.momentum, net.decay);
- } else if(l.type == GRU){
- update_gru_layer(l, update_batch, rate, net.momentum, net.decay);
- } else if(l.type == CRNN){
- update_crnn_layer(l, update_batch, rate, net.momentum, net.decay);
- } else if(l.type == LOCAL){
- update_local_layer(l, update_batch, rate, net.momentum, net.decay);
+ if(l.update){
+ l.update(l, update_batch, rate, net.momentum, net.decay);
}
}
}
float *get_network_output(network net)
{
- #ifdef GPU
- return get_network_output_gpu(net);
- #endif
+#ifdef GPU
+ if (gpu_index >= 0) return get_network_output_gpu(net);
+#endif
int i;
for(i = net.n-1; i > 0; --i) if(net.layers[i].type != COST) break;
return net.layers[i].output;
@@ -229,11 +186,7 @@
float sum = 0;
int count = 0;
for(i = 0; i < net.n; ++i){
- if(net.layers[i].type == COST){
- sum += net.layers[i].cost[0];
- ++count;
- }
- if(net.layers[i].type == DETECTION){
+ if(net.layers[i].cost){
sum += net.layers[i].cost[0];
++count;
}
@@ -265,53 +218,17 @@
state.delta = prev.delta;
}
layer l = net.layers[i];
- if(l.type == CONVOLUTIONAL){
- backward_convolutional_layer(l, state);
- } else if(l.type == DECONVOLUTIONAL){
- backward_deconvolutional_layer(l, state);
- } else if(l.type == ACTIVE){
- backward_activation_layer(l, state);
- } else if(l.type == NORMALIZATION){
- backward_normalization_layer(l, state);
- } else if(l.type == BATCHNORM){
- backward_batchnorm_layer(l, state);
- } else if(l.type == MAXPOOL){
- if(i != 0) backward_maxpool_layer(l, state);
- } else if(l.type == AVGPOOL){
- backward_avgpool_layer(l, state);
- } else if(l.type == DROPOUT){
- backward_dropout_layer(l, state);
- } else if(l.type == DETECTION){
- backward_detection_layer(l, state);
- } else if(l.type == SOFTMAX){
- if(i != 0) backward_softmax_layer(l, state);
- } else if(l.type == CONNECTED){
- backward_connected_layer(l, state);
- } else if(l.type == RNN){
- backward_rnn_layer(l, state);
- } else if(l.type == GRU){
- backward_gru_layer(l, state);
- } else if(l.type == CRNN){
- backward_crnn_layer(l, state);
- } else if(l.type == LOCAL){
- backward_local_layer(l, state);
- } else if(l.type == COST){
- backward_cost_layer(l, state);
- } else if(l.type == ROUTE){
- backward_route_layer(l, net);
- } else if(l.type == SHORTCUT){
- backward_shortcut_layer(l, state);
- }
+ l.backward(l, state);
}
}
float train_network_datum(network net, float *x, float *y)
{
- *net.seen += net.batch;
#ifdef GPU
if(gpu_index >= 0) return train_network_datum_gpu(net, x, y);
#endif
network_state state;
+ *net.seen += net.batch;
state.index = 0;
state.net = net;
state.input = x;
@@ -345,6 +262,7 @@
float train_network(network net, data d)
{
+ assert(d.X.rows % net.batch == 0);
int batch = net.batch;
int n = d.X.rows / batch;
float *X = calloc(batch*d.X.cols, sizeof(float));
@@ -362,6 +280,7 @@
return (float)sum/(n*batch);
}
+
float train_network_batch(network net, data d, int n)
{
int i,j;
@@ -392,18 +311,29 @@
int i;
for(i = 0; i < net->n; ++i){
net->layers[i].batch = b;
+#ifdef CUDNN
+ if(net->layers[i].type == CONVOLUTIONAL){
+ cudnn_convolutional_setup(net->layers + i);
+ }
+#endif
}
}
int resize_network(network *net, int w, int h)
{
+#ifdef GPU
+ cuda_set_device(net->gpu_index);
+ if(gpu_index >= 0){
+ cuda_free(net->workspace);
+ }
+#endif
int i;
//if(w == net->w && h == net->h) return 0;
net->w = w;
net->h = h;
int inputs = 0;
size_t workspace_size = 0;
- //fprintf(stderr, "Resizing to %d x %d...", w, h);
+ //fprintf(stderr, "Resizing to %d x %d...\n", w, h);
//fflush(stderr);
for (i = 0; i < net->n; ++i){
layer l = net->layers[i];
@@ -413,6 +343,12 @@
resize_crop_layer(&l, w, h);
}else if(l.type == MAXPOOL){
resize_maxpool_layer(&l, w, h);
+ }else if(l.type == REGION){
+ resize_region_layer(&l, w, h);
+ }else if(l.type == ROUTE){
+ resize_route_layer(&l, net);
+ }else if(l.type == REORG){
+ resize_reorg_layer(&l, w, h);
}else if(l.type == AVGPOOL){
resize_avgpool_layer(&l, w, h);
}else if(l.type == NORMALIZATION){
@@ -420,6 +356,7 @@
}else if(l.type == COST){
resize_cost_layer(&l, inputs);
}else{
+ fprintf(stderr, "Resizing type %d \n", (int)l.type);
error("Cannot resize this type of layer");
}
if(l.workspace_size > workspace_size) workspace_size = l.workspace_size;
@@ -430,11 +367,21 @@
if(l.type == AVGPOOL) break;
}
#ifdef GPU
- cuda_free(net->workspace);
+ if(gpu_index >= 0){
+ if(net->input_gpu) {
+ cuda_free(*net->input_gpu);
+ *net->input_gpu = 0;
+ cuda_free(*net->truth_gpu);
+ *net->truth_gpu = 0;
+ }
net->workspace = cuda_make_array(0, (workspace_size-1)/sizeof(float)+1);
-#else
+ }else {
free(net->workspace);
net->workspace = calloc(1, workspace_size);
+ }
+#else
+ free(net->workspace);
+ net->workspace = calloc(1, workspace_size);
#endif
//fprintf(stderr, " Done!\n");
return 0;
@@ -634,7 +581,6 @@
return acc;
}
-
float network_accuracy_multi(network net, data d, int n)
{
matrix guess = network_predict_data_multi(net, d, n);
@@ -645,15 +591,19 @@
void free_network(network net)
{
- int i;
- for(i = 0; i < net.n; ++i){
- free_layer(net.layers[i]);
- }
- free(net.layers);
- #ifdef GPU
- if(*net.input_gpu) cuda_free(*net.input_gpu);
- if(*net.truth_gpu) cuda_free(*net.truth_gpu);
- if(net.input_gpu) free(net.input_gpu);
- if(net.truth_gpu) free(net.truth_gpu);
- #endif
+ int i;
+ for (i = 0; i < net.n; ++i) {
+ free_layer(net.layers[i]);
+ }
+ free(net.layers);
+#ifdef GPU
+ if (gpu_index >= 0) cuda_free(net.workspace);
+ else free(net.workspace);
+ if (*net.input_gpu) cuda_free(*net.input_gpu);
+ if (*net.truth_gpu) cuda_free(*net.truth_gpu);
+ if (net.input_gpu) free(net.input_gpu);
+ if (net.truth_gpu) free(net.truth_gpu);
+#else
+ free(net.workspace);
+#endif
}
--
Gitblit v1.10.0