YOLO配置檔案理解
阿新 • • 發佈：2018-12-25
http://www.infocool.net/kb/WWW/201703/317548.html
-
[
net]
-
batch=
64 每batch個樣本更新一次引數。
-
subdivisions=
8 如果記憶體不夠大，將batch分割為subdivisions個子batch，每個子batch的大小為batch/subdivisions。
-
在darknet程式碼中，會將batch/subdivisions命名為batch。
-
height=
416 input影象的高
-
width=
416 Input影象的寬
-
channels=
3 Input影象的通道數
-
momentum=
0.9 動量
-
decay=
0.0005 權重衰減正則項，防止過擬合
-
angle=
0
-
saturation =
1.5 通過調整飽和度來生成更多訓練樣本
-
exposure =
1.5 通過調整曝光量來生成更多訓練樣本
-
hue=
.1 通過調整色調來生成更多訓練樣本
-
-
learning_rate=
0.0001 初始學習率
-
max_batches =
45000 訓練達到max_batches後停止學習
-
policy=steps 調整學習率的policy，有如下policy：CONSTANT, STEP, EXP, POLY, STEPS, SIG, RANDOM
-
steps=
100,
25000,
35000 根據batch_num調整學習率
-
scales=
10,
.1,
.1 學習率變化的比例，累計相乘
-
-
[
convolutional]
-
batch_normalize=
1 是否做BN
-
filters=
32 輸出多少個特徵圖
-
size=
3 卷積核的尺寸
-
stride=
1 做卷積運算的步長
-
pad=
1 如果pad為
0,padding由 padding引數指定。如果pad為
1，padding大小為size/
2
-
activation=leaky 啟用函式：
-
logistic，loggy，relu，elu，relie，plse，hardtan，lhtan，linear，ramp，leaky，tanh，stair
-
-
[
maxpool]
-
size=
2 池化層尺寸
-
stride=
2 池化步進
-
-
[
convolutional]
-
batch_normalize=
1
-
filters=
64
-
size=
3
-
stride=
1
-
pad=
1
-
activation=leaky
-
-
[
maxpool]
-
size=
2
-
stride=
2
-
-
......
-
......
-
-
-
#######
-
-
[
convolutional]
-
batch_normalize=
1
-
size=
3
-
stride=
1
-
pad=
1
-
filters=
1024
-
activation=leaky
-
-
[
convolutional]
-
batch_normalize=
1
-
size=
3
-
stride=
1
-
pad=
1
-
filters=
1024
-
activation=leaky
-
-
[
route] the route layer
is to bring finer grained features
in
from earlier
in the network
-
layers=
-9
-
-
[
reorg] the reorg layer
is to make these features match the feature map size at the later layer.
-
The end feature map
is
13x13, the feature map
from earlier
is
26x26x512.
-
The reorg layer maps the
26x26x512 feature map onto a
13x13x2048 feature map
-
so that it can be concatenated with the feature maps at
13x13 resolution.
-
stride=
2
-
-
[
route]
-
layers=
-1,
-3
-
-
[
convolutional]
-
batch_normalize=
1
-
size=
3
-
stride=
1
-
pad=
1
-
filters=
1024
-
activation=leaky
-
-
[
convolutional]
-
size=
1
-
stride=
1
-
pad=
1
-
filters=
125 region前最後一個卷積層的filters數是特定的，計算公式為filter=num*(classes+
5)
-
5的意義是
5個座標，論文中的tx,ty,tw,th,to
-
activation=linear
-
-
[
region]
-
anchors =
1.08,
1.19,
3.42,
4.41,
6.63,
11.38,
9.42,
5.11,
16.62,
10.52 預選框，可以手工挑選，
-
也可以通過k means 從訓練樣本中學出
-
bias_match=
1
-
classes=
20 網路需要識別的物體種類數
-
coords=
4 每個box的
4個座標tx,ty,tw,th
-
num=
5 每個grid cell預測幾個box
-
softmax=
1 使用softmax做啟用函式
-
jitter=
.2 通過抖動增加噪聲來抑制過擬合
-
rescore=
1 暫理解為一個開關，非
0時通過重打分來調整l.delta（預測值與真實值的差）
-
-
object_scale=
5 暫理解為計算損失時預測框中有物體時的權重
-
noobject_scale=
1 暫理解為計算損失時預測框中無物體時的權重
-
class_scale=
1 暫理解為計算類別損失時的權重
-
coord_scale=
1 暫理解為計算損失時座標偏差的權重
-
-
absolute=
1
-
thresh =
.6
-
random=
0 是否隨機確定最後一個預測框
darknet對應程式碼
找到cfg檔案解析的程式碼，選擇detector demo 作為入口
darknet.c檔案 main 函式開始
-
}
else
if (
0 ==
strcmp(argv[
1],
"detector")){
-
run_detector(argc, argv);
Detector.c檔案 run_detector函式
-
char *prefix = find_char_arg(argc, argv,
"-prefix",
0);
-
float thresh = find_float_arg(argc, argv,
"-thresh",
.24);
-
float hier_thresh = find_float_arg(argc, argv,
"-hier",
.5);
-
int cam_index = find_int_arg(argc, argv,
"-c",
0);
-
int frame_skip = find_int_arg(argc, argv,
"-s",
0);
-
if(argc <
4){
-
fprintf(
stderr,
"usage: %s %s [train/test/valid] [cfg] [weights (optional)]\n", argv[
0], argv[
1]);
-
return;
-
}
-
char *gpu_list = find_char_arg(argc, argv,
"-gpus",
0);
-
char *outfile = find_char_arg(argc, argv,
"-out",
0);
-
-
......
-
......
-
-
else
if(
0==
strcmp(argv[
2],
"demo")) {
-
list *options = read_data_cfg(datacfg);
-
int classes = option_find_int(options,
"classes",
20);
-
char *name_list = option_find_str(options,
"names",
"data/names.list");
-
char **names = get_labels(name_list);
-
demo(cfg, weights, thresh, cam_index, filename, names, classes, frame_skip, prefix, hier_thresh);
-
}
read_data_cfg函式解析配置檔案，儲存到options指標。
class
int classes = option_find_int(options, "classes", 20);
classes為YOLO可識別的種類數
batch、learning_rate、momentum、decay和 subdivisions
demo.c檔案demo函式
net = parse_network_cfg(cfgfile);
Parser.c檔案 parse_network_cfg函式
-
list *sections = read_cfg(filename);
-
node *n = sections->front;
-
if(!n) error(
"Config file has no sections");
-
network net = make_network(sections->size -
1);
-
net.gpu_index = gpu_index;
-
size_params params;
-
-
section *s = (section *)n->val;
-
list *options = s->options;
-
if(!is_network(s)) error(
"First section must be [net] or [network]");
-
parse_net_options(options, &net);
parse_net_options函式
-
net->batch = option_find_int(options,
"batch",
1);
-
net->learning_rate = option_find_float(options,
"learning_rate",
.001);
-
net->momentum = option_find_float(options,
"momentum",
.9);
-
net->decay = option_find_float(options,
"decay",
.0001);
-
int subdivs = option_find_int(options,
"subdivisions",
1);
-
net->time_steps = option_find_int_quiet(options,
"time_steps",
1);
-
net->batch /= subdivs;
-
net->batch *= net->time_steps;
-
net->subdivisions = subdivs;
learning_rate為初始學習率，訓練時的真正學習率和學習率的策略及初始學習率有關。
momentum為動量，在訓練時加入動量可以幫助走出local minima 以及saddle point。
decay是權重衰減正則項，用來防止過擬合。
batch的值等於cfg檔案中的batch/subdivisions 再乘以time_steps。
time_steps在yolo預設的cfg中是沒有配置的，所以是預設值1。
因此batch可以認為就是cfg檔案中的batch/subdivisions。
前面有提到batch的意義是每batch個樣本更新一次引數。
而subdivisions的意義在於降低對GPU memory的要求。
darknet將batch分割為subdivisions個子batch，每個子batch的大小為batch/subdivisions，並將子batch命名為batch。
我們看下訓練時和batch有關的程式碼
Detector.c檔案的train_detector函式
-
#ifdef GPU
-
if(ngpus ==
1){
-
loss = train_network(net, train);
-
}
else {
-
loss = train_networks(nets, ngpus, train,
4);
-
}
-
#else
-
loss = train_network(net, train);
-
#endif
Network.c檔案的train_network函式
-
int batch = net.batch;
-
int n = d.X.rows / batch;
-
float *X =
calloc(batch*d.X.cols,
sizeof(
float));
-
float *y =
calloc(batch*d.y.cols,
sizeof(
float));
-
-
int i;
-
float sum =
0;
-
for(i =
0; i < n; ++i){
-
get_next_batch(d, batch, i*batch, X, y);
-
float err = train_network_datum(net, X, y);
-
sum += err;
-
}
train_network_datum函式
-
*net.seen += net.batch;
-
......
-
......
-
forward_network(net,
state);
-
backward_network(net,
state);
-
float error = get_network_cost(net);
-
if(((*net.seen)/net.batch)%net.subdivisions ==
0) update_network(net);
我們看到，只有((*net.seen)/net.batch)%net.subdivisions == 0時才會更新網路引數。
*net.seen是已經訓練過的子batch數，((*net.seen)/net.batch)%net.subdivisions的意義正是已經訓練過了多少個真正的batch。
policy、steps和scales
Parser.c檔案 parse_network_cfg函式
-
char *policy_s = option_find_str(options,
"policy",
"constant");
-
net->policy = get_policy(policy_s);
-
net->burn_in = option_find_int_quiet(options,
"burn_in",
0);
-
if(net->policy == STEP){
-
net->step = option_find_int(options,
"step",
1);
-
net->scale = option_find_float(options,
"scale",
1);
-
}
else
if (net->policy == STEPS){
-
char *l = option_find(options,
"steps");
-
char *p = option_find(options,
"scales");
-
if(!l || !p) error(
"STEPS policy must have steps and scales in cfg file");
-
-
int len =
strlen(l);
-
int n =
1;
-
int i;
-
for(i =
0; i < len; ++i){
-
if (l[i] ==
',') ++n;
-
}
-
int *steps =
calloc(n,
sizeof(
int));
-
float *scales =
calloc(n,
sizeof(
float));
-
for(i =
0; i < n; ++i){
-
int step = atoi(l);
-
float scale = atof(p);
-
l =
strchr(l,
',')+
1;
-
p =
strchr(p,
',')+
1;
-
steps[i] = step;
-
scales[i] = scale;
-
}
-
net->scales = scales;
-
net->steps = steps;
-
net->num_steps = n;
-
}
else
if (net->policy == EXP){
-
net->gamma = option_find_float(options,
"gamma",
1);
-
}
else
if (net->policy == SIG){
-
net->gamma = option_find_floa