yolo alternatives
This commit is contained in:
Francesco Gatti
@@ -7,3 +7,4 @@ build/
|
||||
*.caffemodel
|
||||
*.h5
|
||||
*.tar.gz
|
||||
*.weights
|
||||
|
||||
+11
-1
@@ -57,12 +57,22 @@ target_link_libraries(test_mnist tkDNN)
|
||||
add_executable(test_mnistRT tests/mnist/test_mnistRT.cpp)
|
||||
target_link_libraries(test_mnistRT tkDNN)
|
||||
|
||||
## YOLO NETS
|
||||
add_executable(test_yolo tests/yolo/yolo.cpp)
|
||||
target_link_libraries(test_yolo tkDNN)
|
||||
|
||||
add_executable(test_yolo_tiny tests/yolo-tiny/yolo-tiny.cpp)
|
||||
add_executable(test_yolo_tiny tests/yolo_tiny/yolo_tiny.cpp)
|
||||
target_link_libraries(test_yolo_tiny tkDNN)
|
||||
|
||||
add_executable(test_yolo_relu tests/yolo_relu/yolo_relu.cpp)
|
||||
target_link_libraries(test_yolo_relu tkDNN)
|
||||
|
||||
|
||||
add_executable(test_yolo_224 tests/yolo_224/yolo_224.cpp)
|
||||
target_link_libraries(test_yolo_224 tkDNN)
|
||||
################################################################################
|
||||
|
||||
|
||||
add_executable(test_rtinference tests/test_rtinference/rtinference.cpp)
|
||||
target_link_libraries(test_rtinference tkDNN)
|
||||
|
||||
|
||||
@@ -1,6 +1,7 @@
|
||||
#include<iostream>
|
||||
#include "tkdnn.h"
|
||||
#include <stdlib.h> /* srand, rand */
|
||||
#include <unistd.h>
|
||||
|
||||
#include <opencv2/core/core.hpp>
|
||||
#include <opencv2/highgui/highgui.hpp>
|
||||
@@ -84,35 +85,61 @@ void compute_image( cv::Mat imageORIG,
|
||||
|
||||
}
|
||||
|
||||
int print_usage() {
|
||||
std::cout<<"usage: ./detection net.rt validation_list.txt [-t <thresh>] [-s]\n"
|
||||
<<" -t: set thresh value\n -s: show images as compute\n\n"
|
||||
<<"> validation_list.txt format: \n"
|
||||
<<" path/to/image.jpg path/to/label.txt\n"
|
||||
<<"> label.txt format: \n"
|
||||
<<" <object-class> <x> <y> <width> <height>\n"
|
||||
<<" x and y are the box center, "
|
||||
<<"all values are relative to the image size\n\n";
|
||||
return 1;
|
||||
}
|
||||
|
||||
|
||||
int main(int argc, char *argv[]) {
|
||||
|
||||
if(argc < 3) {
|
||||
std::cout<<"usage: "<<argv[0]<<" net.rt validation_list.txt\n"
|
||||
<<"> validation_list.txt format: \n"
|
||||
<<" path/to/image.jpg path/to/label.txt\n"
|
||||
<<"> label.txt format: \n"
|
||||
<<" <object-class> <x> <y> <width> <height>\n"
|
||||
<<" x and y are the box center, "
|
||||
<<"all values are relative to the image size\n\n";
|
||||
return 1;
|
||||
//params
|
||||
char *tensor_path = NULL;
|
||||
char *imageset_path = NULL;
|
||||
float thresh = 0.3f;
|
||||
bool show = false;
|
||||
|
||||
//parse params
|
||||
int c;
|
||||
while ((c = getopt (argc, argv, "t:s")) != -1) {
|
||||
switch(c) {
|
||||
case 't': thresh = atof(optarg); break;
|
||||
case 's': show = true; break;
|
||||
case '?':
|
||||
return print_usage();
|
||||
default: return print_usage();
|
||||
}
|
||||
}
|
||||
|
||||
if(!fileExist(argv[1]))
|
||||
FatalError("unable to read serialRT file");
|
||||
if(argc - optind == 2) {
|
||||
tensor_path = argv[optind];
|
||||
imageset_path = argv[optind+1];
|
||||
} else {
|
||||
std::cout<<"not enough arguments.\n";
|
||||
return print_usage();
|
||||
}
|
||||
//end parsing
|
||||
|
||||
if(!fileExist(tensor_path))
|
||||
FatalError("unable to read serialRT file");
|
||||
//convert network to tensorRT
|
||||
tkDNN::NetworkRT netRT(NULL, argv[1]);
|
||||
tkDNN::RegionInterpret rI(netRT.input_dim, netRT.output_dim, 80, 4, 5, 0.3f, reg_bias);
|
||||
tkDNN::NetworkRT netRT(NULL, tensor_path);
|
||||
tkDNN::RegionInterpret rI(netRT.input_dim, netRT.output_dim, 80, 4, 5, thresh, reg_bias);
|
||||
|
||||
dnnType *input = new float[netRT.input_dim.tot()];
|
||||
dnnType *output = new float[netRT.output_dim.tot()];
|
||||
|
||||
std::string line;
|
||||
std::ifstream imageset(argv[2]);
|
||||
std::ifstream imageset(imageset_path);
|
||||
if(!imageset.is_open())
|
||||
FatalError("could not read imageset");
|
||||
|
||||
|
||||
float mAP = 0;
|
||||
int processed_images;
|
||||
@@ -193,9 +220,11 @@ int main(int argc, char *argv[]) {
|
||||
<<", mAP: "<<mAP/processed_images<<"\n";
|
||||
|
||||
//show results
|
||||
//cv::namedWindow("result");
|
||||
//cv::imshow("result", img);
|
||||
//cv::waitKey(1);
|
||||
if(show) {
|
||||
cv::namedWindow("result");
|
||||
cv::imshow("result", img);
|
||||
cv::waitKey(1000);
|
||||
}
|
||||
}
|
||||
|
||||
return 0;
|
||||
|
||||
+2
-1
@@ -324,7 +324,8 @@ void RegionInterpret::showImageResult(dnnType *input_h) {
|
||||
|
||||
for(int i=0; i<res_boxes_n; i++) {
|
||||
box bx = res_boxes[i];
|
||||
cv::rectangle(color, cv::Point(bx.x, bx.y), cv::Point(bx.w, bx.h),
|
||||
cv::rectangle(color, cv::Point(bx.x - bx.w/2, bx.y - bx.h/2),
|
||||
cv::Point(bx.x + bx.w/2, bx.y + bx.h/2),
|
||||
cv::Scalar( 0, 0, 255), 2);
|
||||
}
|
||||
cv::namedWindow("result");
|
||||
|
||||
@@ -0,0 +1,258 @@
|
||||
[net]
|
||||
# Testing
|
||||
#batch=1
|
||||
#subdivisions=1
|
||||
# Training
|
||||
batch=32
|
||||
subdivisions=8
|
||||
width=608
|
||||
height=608
|
||||
channels=3
|
||||
momentum=0.9
|
||||
decay=0.0005
|
||||
angle=0
|
||||
saturation = 1.5
|
||||
exposure = 1.5
|
||||
hue=.1
|
||||
|
||||
learning_rate=0.001
|
||||
burn_in=1000
|
||||
max_batches = 500200
|
||||
policy=steps
|
||||
steps=400000,450000
|
||||
scales=.1,.1
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=32
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
[maxpool]
|
||||
size=2
|
||||
stride=2
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=64
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
[maxpool]
|
||||
size=2
|
||||
stride=2
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=128
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=64
|
||||
size=1
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=128
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
[maxpool]
|
||||
size=2
|
||||
stride=2
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=256
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=128
|
||||
size=1
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=256
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
[maxpool]
|
||||
size=2
|
||||
stride=2
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=512
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=256
|
||||
size=1
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=512
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=256
|
||||
size=1
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=512
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
[maxpool]
|
||||
size=2
|
||||
stride=2
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=1024
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=512
|
||||
size=1
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=1024
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=512
|
||||
size=1
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=1024
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
|
||||
#######
|
||||
|
||||
[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]
|
||||
layers=-9
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
size=1
|
||||
stride=1
|
||||
pad=1
|
||||
filters=64
|
||||
activation=leaky
|
||||
|
||||
[reorg]
|
||||
stride=2
|
||||
|
||||
[route]
|
||||
layers=-1,-4
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
filters=1024
|
||||
activation=leaky
|
||||
|
||||
[convolutional]
|
||||
size=1
|
||||
stride=1
|
||||
pad=1
|
||||
filters=425
|
||||
activation=linear
|
||||
|
||||
|
||||
[region]
|
||||
anchors = 0.57273, 0.677385, 1.87446, 2.06253, 3.33843, 5.47434, 7.88282, 3.52778, 9.77052, 9.16828
|
||||
bias_match=1
|
||||
classes=80
|
||||
coords=4
|
||||
num=5
|
||||
softmax=1
|
||||
jitter=.3
|
||||
rescore=1
|
||||
|
||||
object_scale=5
|
||||
noobject_scale=1
|
||||
class_scale=1
|
||||
coord_scale=1
|
||||
|
||||
absolute=1
|
||||
thresh = .6
|
||||
random=1
|
||||
@@ -0,0 +1,258 @@
|
||||
[net]
|
||||
# Testing
|
||||
#batch=1
|
||||
#subdivisions=1
|
||||
# Training
|
||||
batch=64
|
||||
subdivisions=16
|
||||
width=224
|
||||
height=224
|
||||
channels=3
|
||||
momentum=0.9
|
||||
decay=0.0005
|
||||
angle=0
|
||||
saturation = 1.5
|
||||
exposure = 1.5
|
||||
hue=.1
|
||||
|
||||
learning_rate=0.001
|
||||
burn_in=1000
|
||||
max_batches = 500200
|
||||
policy=steps
|
||||
steps=400000,450000
|
||||
scales=.1,.1
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=32
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
[maxpool]
|
||||
size=2
|
||||
stride=2
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=64
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
[maxpool]
|
||||
size=2
|
||||
stride=2
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=128
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=64
|
||||
size=1
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=128
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
[maxpool]
|
||||
size=2
|
||||
stride=2
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=256
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=128
|
||||
size=1
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=256
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
[maxpool]
|
||||
size=2
|
||||
stride=2
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=512
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=256
|
||||
size=1
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=512
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=256
|
||||
size=1
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=512
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
[maxpool]
|
||||
size=2
|
||||
stride=2
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=1024
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=512
|
||||
size=1
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=1024
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=512
|
||||
size=1
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=1024
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
|
||||
#######
|
||||
|
||||
[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]
|
||||
layers=-9
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
size=1
|
||||
stride=1
|
||||
pad=1
|
||||
filters=64
|
||||
activation=leaky
|
||||
|
||||
[reorg]
|
||||
stride=2
|
||||
|
||||
[route]
|
||||
layers=-1,-4
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
filters=1024
|
||||
activation=leaky
|
||||
|
||||
[convolutional]
|
||||
size=1
|
||||
stride=1
|
||||
pad=1
|
||||
filters=425
|
||||
activation=linear
|
||||
|
||||
|
||||
[region]
|
||||
anchors = 0.57273, 0.677385, 1.87446, 2.06253, 3.33843, 5.47434, 7.88282, 3.52778, 9.77052, 9.16828
|
||||
bias_match=1
|
||||
classes=80
|
||||
coords=4
|
||||
num=5
|
||||
softmax=1
|
||||
jitter=.3
|
||||
rescore=1
|
||||
|
||||
object_scale=5
|
||||
noobject_scale=1
|
||||
class_scale=1
|
||||
coord_scale=1
|
||||
|
||||
absolute=1
|
||||
thresh = .6
|
||||
random=1
|
||||
@@ -0,0 +1,150 @@
|
||||
#include<iostream>
|
||||
#include "tkdnn.h"
|
||||
|
||||
const char *input_bin = "../tests/yolo_224/layers/input.bin";
|
||||
const char *c0_bin = "../tests/yolo_224/layers/c0.bin";
|
||||
const char *c2_bin = "../tests/yolo_224/layers/c2.bin";
|
||||
const char *c4_bin = "../tests/yolo_224/layers/c4.bin";
|
||||
const char *c5_bin = "../tests/yolo_224/layers/c5.bin";
|
||||
const char *c6_bin = "../tests/yolo_224/layers/c6.bin";
|
||||
const char *c8_bin = "../tests/yolo_224/layers/c8.bin";
|
||||
const char *c9_bin = "../tests/yolo_224/layers/c9.bin";
|
||||
const char *c10_bin = "../tests/yolo_224/layers/c10.bin";
|
||||
const char *c12_bin = "../tests/yolo_224/layers/c12.bin";
|
||||
const char *c13_bin = "../tests/yolo_224/layers/c13.bin";
|
||||
const char *c14_bin = "../tests/yolo_224/layers/c14.bin";
|
||||
const char *c15_bin = "../tests/yolo_224/layers/c15.bin";
|
||||
const char *c16_bin = "../tests/yolo_224/layers/c16.bin";
|
||||
const char *c18_bin = "../tests/yolo_224/layers/c18.bin";
|
||||
const char *c19_bin = "../tests/yolo_224/layers/c19.bin";
|
||||
const char *c20_bin = "../tests/yolo_224/layers/c20.bin";
|
||||
const char *c21_bin = "../tests/yolo_224/layers/c21.bin";
|
||||
const char *c22_bin = "../tests/yolo_224/layers/c22.bin";
|
||||
const char *c23_bin = "../tests/yolo_224/layers/c23.bin";
|
||||
const char *c24_bin = "../tests/yolo_224/layers/c24.bin";
|
||||
const char *c26_bin = "../tests/yolo_224/layers/c26.bin";
|
||||
const char *c29_bin = "../tests/yolo_224/layers/c29.bin";
|
||||
const char *c30_bin = "../tests/yolo_224/layers/c30.bin";
|
||||
const char *g31_bin = "../tests/yolo_224/layers/g31.bin";
|
||||
const char *output_bin = "../tests/yolo_224/layers/output.bin";
|
||||
|
||||
int main() {
|
||||
|
||||
// Network layout
|
||||
tkDNN::dataDim_t dim(1, 3, 224, 224, 1);
|
||||
tkDNN::Network net(dim);
|
||||
|
||||
tkDNN::Conv2d c0 (&net, 32, 3, 3, 1, 1, 1, 1, c0_bin, true);
|
||||
tkDNN::Activation a0 (&net, tkDNN::ACTIVATION_LEAKY);
|
||||
tkDNN::Pooling p1 (&net, 2, 2, 2, 2, tkDNN::POOLING_MAX);
|
||||
|
||||
tkDNN::Conv2d c2 (&net, 64, 3, 3, 1, 1, 1, 1, c2_bin, true);
|
||||
tkDNN::Activation a2 (&net, tkDNN::ACTIVATION_LEAKY);
|
||||
tkDNN::Pooling p3 (&net, 2, 2, 2, 2, tkDNN::POOLING_MAX);
|
||||
|
||||
tkDNN::Conv2d c4 (&net, 128, 3, 3, 1, 1, 1, 1, c4_bin, true);
|
||||
tkDNN::Activation a4 (&net, tkDNN::ACTIVATION_LEAKY);
|
||||
tkDNN::Conv2d c5 (&net, 64, 1, 1, 1, 1, 0, 0, c5_bin, true);
|
||||
tkDNN::Activation a5 (&net, tkDNN::ACTIVATION_LEAKY);
|
||||
tkDNN::Conv2d c6 (&net, 128, 3, 3, 1, 1, 1, 1, c6_bin, true);
|
||||
tkDNN::Activation a6 (&net, tkDNN::ACTIVATION_LEAKY);
|
||||
tkDNN::Pooling p7 (&net, 2, 2, 2, 2, tkDNN::POOLING_MAX);
|
||||
|
||||
tkDNN::Conv2d c8 (&net, 256, 3, 3, 1, 1, 1, 1, c8_bin, true);
|
||||
tkDNN::Activation a8 (&net, tkDNN::ACTIVATION_LEAKY);
|
||||
tkDNN::Conv2d c9 (&net, 128, 1, 1, 1, 1, 0, 0, c9_bin, true);
|
||||
tkDNN::Activation a9 (&net, tkDNN::ACTIVATION_LEAKY);
|
||||
tkDNN::Conv2d c10(&net, 256, 3, 3, 1, 1, 1, 1, c10_bin, true);
|
||||
tkDNN::Activation a10(&net, tkDNN::ACTIVATION_LEAKY);
|
||||
tkDNN::Pooling p11(&net, 2, 2, 2, 2, tkDNN::POOLING_MAX);
|
||||
|
||||
tkDNN::Conv2d c12(&net, 512, 3, 3, 1, 1, 1, 1, c12_bin, true);
|
||||
tkDNN::Activation a12(&net, tkDNN::ACTIVATION_LEAKY);
|
||||
tkDNN::Conv2d c13(&net, 256, 1, 1, 1, 1, 0, 0, c13_bin, true);
|
||||
tkDNN::Activation a13(&net, tkDNN::ACTIVATION_LEAKY);
|
||||
tkDNN::Conv2d c14(&net, 512, 3, 3, 1, 1, 1, 1, c14_bin, true);
|
||||
tkDNN::Activation a14(&net, tkDNN::ACTIVATION_LEAKY);
|
||||
tkDNN::Conv2d c15(&net, 256, 1, 1, 1, 1, 0, 0, c15_bin, true);
|
||||
tkDNN::Activation a15(&net, tkDNN::ACTIVATION_LEAKY);
|
||||
tkDNN::Conv2d c16(&net, 512, 3, 3, 1, 1, 1, 1, c16_bin, true);
|
||||
tkDNN::Activation a16(&net, tkDNN::ACTIVATION_LEAKY);
|
||||
tkDNN::Pooling p17(&net, 2, 2, 2, 2, tkDNN::POOLING_MAX);
|
||||
|
||||
tkDNN::Conv2d c18(&net, 1024, 3, 3, 1, 1, 1, 1, c18_bin, true);
|
||||
tkDNN::Activation a18(&net, tkDNN::ACTIVATION_LEAKY);
|
||||
tkDNN::Conv2d c19(&net, 512, 1, 1, 1, 1, 0, 0, c19_bin, true);
|
||||
tkDNN::Activation a19(&net, tkDNN::ACTIVATION_LEAKY);
|
||||
tkDNN::Conv2d c20(&net, 1024, 3, 3, 1, 1, 1, 1, c20_bin, true);
|
||||
tkDNN::Activation a20(&net, tkDNN::ACTIVATION_LEAKY);
|
||||
tkDNN::Conv2d c21(&net, 512, 1, 1, 1, 1, 0, 0, c21_bin, true);
|
||||
tkDNN::Activation a21(&net, tkDNN::ACTIVATION_LEAKY);
|
||||
tkDNN::Conv2d c22(&net, 1024, 3, 3, 1, 1, 1, 1, c22_bin, true);
|
||||
tkDNN::Activation a22(&net, tkDNN::ACTIVATION_LEAKY);
|
||||
tkDNN::Conv2d c23(&net, 1024, 3, 3, 1, 1, 1, 1, c23_bin, true);
|
||||
tkDNN::Activation a23(&net, tkDNN::ACTIVATION_LEAKY);
|
||||
tkDNN::Conv2d c24(&net, 1024, 3, 3, 1, 1, 1, 1, c24_bin, true);
|
||||
tkDNN::Activation a24(&net, tkDNN::ACTIVATION_LEAKY);
|
||||
|
||||
tkDNN::Layer *m25_layers[1] = { &a16 };
|
||||
tkDNN::Route m25(&net, m25_layers, 1);
|
||||
tkDNN::Conv2d c26(&net, 64, 1, 1, 1, 1, 0, 0, c26_bin, true);
|
||||
tkDNN::Activation a26(&net, tkDNN::ACTIVATION_LEAKY);
|
||||
tkDNN::Reorg r27(&net, 2);
|
||||
|
||||
tkDNN::Layer *m28_layers[2] = { &r27, &a24 };
|
||||
tkDNN::Route m28(&net, m28_layers, 2);
|
||||
|
||||
tkDNN::Conv2d c29(&net, 1024, 3, 3, 1, 1, 1, 1, c29_bin, true);
|
||||
tkDNN::Activation a29(&net, tkDNN::ACTIVATION_LEAKY);
|
||||
tkDNN::Conv2d c30(&net, 425, 1, 1, 1, 1, 0, 0, c30_bin, false);
|
||||
tkDNN::Region g31(&net, 80, 4, 5);
|
||||
|
||||
tkDNN::RegionInterpret rI(dim, g31.output_dim, 80, 4, 5, 0.6f, g31_bin);
|
||||
|
||||
// Load input
|
||||
dnnType *data;
|
||||
dnnType *input_h;
|
||||
readBinaryFile(input_bin, dim.tot(), &input_h, &data);
|
||||
|
||||
//print network model
|
||||
net.print();
|
||||
|
||||
//convert network to tensorRT
|
||||
tkDNN::NetworkRT netRT(&net, "yolo_224.rt");
|
||||
|
||||
dnnType *out_data, *out_data2; // cudnn output, tensorRT output
|
||||
|
||||
tkDNN::dataDim_t dim1 = dim; //input dim
|
||||
printCenteredTitle(" CUDNN inference ", '=', 30); {
|
||||
dim1.print();
|
||||
TIMER_START
|
||||
out_data = net.infer(dim1, data);
|
||||
TIMER_STOP
|
||||
dim1.print();
|
||||
}
|
||||
|
||||
tkDNN::dataDim_t dim2 = dim;
|
||||
printCenteredTitle(" TENSORRT inference ", '=', 30); {
|
||||
dim2.print();
|
||||
TIMER_START
|
||||
out_data2 = netRT.infer(dim2, data);
|
||||
TIMER_STOP
|
||||
dim2.print();
|
||||
}
|
||||
|
||||
printCenteredTitle(" CHECK RESULTS ", '=', 30);
|
||||
dnnType *out, *out_h;
|
||||
int out_dim = net.getOutputDim().tot();
|
||||
readBinaryFile(output_bin, out_dim, &out_h, &out);
|
||||
std::cout<<"CUDNN vs correct"; checkResult(out_dim, out_data, out);
|
||||
std::cout<<"TRT vs correct"; checkResult(out_dim, out_data2, out);
|
||||
std::cout<<"CUDNN vs TRT "; checkResult(out_dim, out_data, out_data2);
|
||||
|
||||
std::cout<<"\n\nDetected objects: \n";
|
||||
dnnType *output_h = new dnnType[rI.output_dim.tot()];
|
||||
checkCuda(cudaMemcpy(output_h, out_data2,
|
||||
rI.output_dim.tot()*sizeof(dnnType), cudaMemcpyDeviceToHost));
|
||||
rI.interpretData(output_h);
|
||||
rI.showImageResult(input_h);
|
||||
return 0;
|
||||
}
|
||||
@@ -0,0 +1,258 @@
|
||||
[net]
|
||||
# Testing
|
||||
#batch=1
|
||||
#subdivisions=1
|
||||
# Training
|
||||
batch=64
|
||||
subdivisions=16
|
||||
width=608
|
||||
height=608
|
||||
channels=3
|
||||
momentum=0.9
|
||||
decay=0.0005
|
||||
angle=0
|
||||
saturation = 1.5
|
||||
exposure = 1.5
|
||||
hue=.1
|
||||
|
||||
learning_rate=0.001
|
||||
burn_in=1000
|
||||
max_batches = 500200
|
||||
policy=steps
|
||||
steps=400000,450000
|
||||
scales=.1,.1
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=32
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=relu
|
||||
|
||||
[maxpool]
|
||||
size=2
|
||||
stride=2
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=64
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=relu
|
||||
|
||||
[maxpool]
|
||||
size=2
|
||||
stride=2
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=128
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=relu
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=64
|
||||
size=1
|
||||
stride=1
|
||||
pad=1
|
||||
activation=relu
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=128
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=relu
|
||||
|
||||
[maxpool]
|
||||
size=2
|
||||
stride=2
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=256
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=relu
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=128
|
||||
size=1
|
||||
stride=1
|
||||
pad=1
|
||||
activation=relu
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=256
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=relu
|
||||
|
||||
[maxpool]
|
||||
size=2
|
||||
stride=2
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=512
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=relu
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=256
|
||||
size=1
|
||||
stride=1
|
||||
pad=1
|
||||
activation=relu
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=512
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=relu
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=256
|
||||
size=1
|
||||
stride=1
|
||||
pad=1
|
||||
activation=relu
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=512
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=relu
|
||||
|
||||
[maxpool]
|
||||
size=2
|
||||
stride=2
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=1024
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=relu
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=512
|
||||
size=1
|
||||
stride=1
|
||||
pad=1
|
||||
activation=relu
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=1024
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=relu
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=512
|
||||
size=1
|
||||
stride=1
|
||||
pad=1
|
||||
activation=relu
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=1024
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=relu
|
||||
|
||||
|
||||
#######
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
filters=1024
|
||||
activation=relu
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
filters=1024
|
||||
activation=relu
|
||||
|
||||
[route]
|
||||
layers=-9
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
size=1
|
||||
stride=1
|
||||
pad=1
|
||||
filters=64
|
||||
activation=relu
|
||||
|
||||
[reorg]
|
||||
stride=2
|
||||
|
||||
[route]
|
||||
layers=-1,-4
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
filters=1024
|
||||
activation=relu
|
||||
|
||||
[convolutional]
|
||||
size=1
|
||||
stride=1
|
||||
pad=1
|
||||
filters=425
|
||||
activation=linear
|
||||
|
||||
|
||||
[region]
|
||||
anchors = 0.57273, 0.677385, 1.87446, 2.06253, 3.33843, 5.47434, 7.88282, 3.52778, 9.77052, 9.16828
|
||||
bias_match=1
|
||||
classes=80
|
||||
coords=4
|
||||
num=5
|
||||
softmax=1
|
||||
jitter=.3
|
||||
rescore=1
|
||||
|
||||
object_scale=5
|
||||
noobject_scale=1
|
||||
class_scale=1
|
||||
coord_scale=1
|
||||
|
||||
absolute=1
|
||||
thresh = .6
|
||||
random=1
|
||||
@@ -0,0 +1,150 @@
|
||||
#include<iostream>
|
||||
#include "tkdnn.h"
|
||||
|
||||
const char *input_bin = "../tests/yolo_relu/layers/input.bin";
|
||||
const char *c0_bin = "../tests/yolo_relu/layers/c0.bin";
|
||||
const char *c2_bin = "../tests/yolo_relu/layers/c2.bin";
|
||||
const char *c4_bin = "../tests/yolo_relu/layers/c4.bin";
|
||||
const char *c5_bin = "../tests/yolo_relu/layers/c5.bin";
|
||||
const char *c6_bin = "../tests/yolo_relu/layers/c6.bin";
|
||||
const char *c8_bin = "../tests/yolo_relu/layers/c8.bin";
|
||||
const char *c9_bin = "../tests/yolo_relu/layers/c9.bin";
|
||||
const char *c10_bin = "../tests/yolo_relu/layers/c10.bin";
|
||||
const char *c12_bin = "../tests/yolo_relu/layers/c12.bin";
|
||||
const char *c13_bin = "../tests/yolo_relu/layers/c13.bin";
|
||||
const char *c14_bin = "../tests/yolo_relu/layers/c14.bin";
|
||||
const char *c15_bin = "../tests/yolo_relu/layers/c15.bin";
|
||||
const char *c16_bin = "../tests/yolo_relu/layers/c16.bin";
|
||||
const char *c18_bin = "../tests/yolo_relu/layers/c18.bin";
|
||||
const char *c19_bin = "../tests/yolo_relu/layers/c19.bin";
|
||||
const char *c20_bin = "../tests/yolo_relu/layers/c20.bin";
|
||||
const char *c21_bin = "../tests/yolo_relu/layers/c21.bin";
|
||||
const char *c22_bin = "../tests/yolo_relu/layers/c22.bin";
|
||||
const char *c23_bin = "../tests/yolo_relu/layers/c23.bin";
|
||||
const char *c24_bin = "../tests/yolo_relu/layers/c24.bin";
|
||||
const char *c26_bin = "../tests/yolo_relu/layers/c26.bin";
|
||||
const char *c29_bin = "../tests/yolo_relu/layers/c29.bin";
|
||||
const char *c30_bin = "../tests/yolo_relu/layers/c30.bin";
|
||||
const char *g31_bin = "../tests/yolo_relu/layers/g31.bin";
|
||||
const char *output_bin = "../tests/yolo_relu/layers/output.bin";
|
||||
|
||||
int main() {
|
||||
|
||||
// Network layout
|
||||
tkDNN::dataDim_t dim(1, 3, 608, 608, 1);
|
||||
tkDNN::Network net(dim);
|
||||
|
||||
tkDNN::Conv2d c0 (&net, 32, 3, 3, 1, 1, 1, 1, c0_bin, true);
|
||||
tkDNN::Activation a0 (&net, CUDNN_ACTIVATION_RELU);
|
||||
tkDNN::Pooling p1 (&net, 2, 2, 2, 2, tkDNN::POOLING_MAX);
|
||||
|
||||
tkDNN::Conv2d c2 (&net, 64, 3, 3, 1, 1, 1, 1, c2_bin, true);
|
||||
tkDNN::Activation a2 (&net, CUDNN_ACTIVATION_RELU);
|
||||
tkDNN::Pooling p3 (&net, 2, 2, 2, 2, tkDNN::POOLING_MAX);
|
||||
|
||||
tkDNN::Conv2d c4 (&net, 128, 3, 3, 1, 1, 1, 1, c4_bin, true);
|
||||
tkDNN::Activation a4 (&net, CUDNN_ACTIVATION_RELU);
|
||||
tkDNN::Conv2d c5 (&net, 64, 1, 1, 1, 1, 0, 0, c5_bin, true);
|
||||
tkDNN::Activation a5 (&net, CUDNN_ACTIVATION_RELU);
|
||||
tkDNN::Conv2d c6 (&net, 128, 3, 3, 1, 1, 1, 1, c6_bin, true);
|
||||
tkDNN::Activation a6 (&net, CUDNN_ACTIVATION_RELU);
|
||||
tkDNN::Pooling p7 (&net, 2, 2, 2, 2, tkDNN::POOLING_MAX);
|
||||
|
||||
tkDNN::Conv2d c8 (&net, 256, 3, 3, 1, 1, 1, 1, c8_bin, true);
|
||||
tkDNN::Activation a8 (&net, CUDNN_ACTIVATION_RELU);
|
||||
tkDNN::Conv2d c9 (&net, 128, 1, 1, 1, 1, 0, 0, c9_bin, true);
|
||||
tkDNN::Activation a9 (&net, CUDNN_ACTIVATION_RELU);
|
||||
tkDNN::Conv2d c10(&net, 256, 3, 3, 1, 1, 1, 1, c10_bin, true);
|
||||
tkDNN::Activation a10(&net, CUDNN_ACTIVATION_RELU);
|
||||
tkDNN::Pooling p11(&net, 2, 2, 2, 2, tkDNN::POOLING_MAX);
|
||||
|
||||
tkDNN::Conv2d c12(&net, 512, 3, 3, 1, 1, 1, 1, c12_bin, true);
|
||||
tkDNN::Activation a12(&net, CUDNN_ACTIVATION_RELU);
|
||||
tkDNN::Conv2d c13(&net, 256, 1, 1, 1, 1, 0, 0, c13_bin, true);
|
||||
tkDNN::Activation a13(&net, CUDNN_ACTIVATION_RELU);
|
||||
tkDNN::Conv2d c14(&net, 512, 3, 3, 1, 1, 1, 1, c14_bin, true);
|
||||
tkDNN::Activation a14(&net, CUDNN_ACTIVATION_RELU);
|
||||
tkDNN::Conv2d c15(&net, 256, 1, 1, 1, 1, 0, 0, c15_bin, true);
|
||||
tkDNN::Activation a15(&net, CUDNN_ACTIVATION_RELU);
|
||||
tkDNN::Conv2d c16(&net, 512, 3, 3, 1, 1, 1, 1, c16_bin, true);
|
||||
tkDNN::Activation a16(&net, CUDNN_ACTIVATION_RELU);
|
||||
tkDNN::Pooling p17(&net, 2, 2, 2, 2, tkDNN::POOLING_MAX);
|
||||
|
||||
tkDNN::Conv2d c18(&net, 1024, 3, 3, 1, 1, 1, 1, c18_bin, true);
|
||||
tkDNN::Activation a18(&net, CUDNN_ACTIVATION_RELU);
|
||||
tkDNN::Conv2d c19(&net, 512, 1, 1, 1, 1, 0, 0, c19_bin, true);
|
||||
tkDNN::Activation a19(&net, CUDNN_ACTIVATION_RELU);
|
||||
tkDNN::Conv2d c20(&net, 1024, 3, 3, 1, 1, 1, 1, c20_bin, true);
|
||||
tkDNN::Activation a20(&net, CUDNN_ACTIVATION_RELU);
|
||||
tkDNN::Conv2d c21(&net, 512, 1, 1, 1, 1, 0, 0, c21_bin, true);
|
||||
tkDNN::Activation a21(&net, CUDNN_ACTIVATION_RELU);
|
||||
tkDNN::Conv2d c22(&net, 1024, 3, 3, 1, 1, 1, 1, c22_bin, true);
|
||||
tkDNN::Activation a22(&net, CUDNN_ACTIVATION_RELU);
|
||||
tkDNN::Conv2d c23(&net, 1024, 3, 3, 1, 1, 1, 1, c23_bin, true);
|
||||
tkDNN::Activation a23(&net, CUDNN_ACTIVATION_RELU);
|
||||
tkDNN::Conv2d c24(&net, 1024, 3, 3, 1, 1, 1, 1, c24_bin, true);
|
||||
tkDNN::Activation a24(&net, CUDNN_ACTIVATION_RELU);
|
||||
|
||||
tkDNN::Layer *m25_layers[1] = { &a16 };
|
||||
tkDNN::Route m25(&net, m25_layers, 1);
|
||||
tkDNN::Conv2d c26(&net, 64, 1, 1, 1, 1, 0, 0, c26_bin, true);
|
||||
tkDNN::Activation a26(&net, CUDNN_ACTIVATION_RELU);
|
||||
tkDNN::Reorg r27(&net, 2);
|
||||
|
||||
tkDNN::Layer *m28_layers[2] = { &r27, &a24 };
|
||||
tkDNN::Route m28(&net, m28_layers, 2);
|
||||
|
||||
tkDNN::Conv2d c29(&net, 1024, 3, 3, 1, 1, 1, 1, c29_bin, true);
|
||||
tkDNN::Activation a29(&net, CUDNN_ACTIVATION_RELU);
|
||||
tkDNN::Conv2d c30(&net, 425, 1, 1, 1, 1, 0, 0, c30_bin, false);
|
||||
tkDNN::Region g31(&net, 80, 4, 5);
|
||||
|
||||
tkDNN::RegionInterpret rI(dim, g31.output_dim, 80, 4, 5, 0.3f, g31_bin);
|
||||
|
||||
// Load input
|
||||
dnnType *data;
|
||||
dnnType *input_h;
|
||||
readBinaryFile(input_bin, dim.tot(), &input_h, &data);
|
||||
|
||||
//print network model
|
||||
net.print();
|
||||
|
||||
//convert network to tensorRT
|
||||
tkDNN::NetworkRT netRT(&net, "yolo_relu.rt");
|
||||
|
||||
dnnType *out_data, *out_data2; // cudnn output, tensorRT output
|
||||
|
||||
tkDNN::dataDim_t dim1 = dim; //input dim
|
||||
printCenteredTitle(" CUDNN inference ", '=', 30); {
|
||||
dim1.print();
|
||||
TIMER_START
|
||||
out_data = net.infer(dim1, data);
|
||||
TIMER_STOP
|
||||
dim1.print();
|
||||
}
|
||||
|
||||
tkDNN::dataDim_t dim2 = dim;
|
||||
printCenteredTitle(" TENSORRT inference ", '=', 30); {
|
||||
dim2.print();
|
||||
TIMER_START
|
||||
out_data2 = netRT.infer(dim2, data);
|
||||
TIMER_STOP
|
||||
dim2.print();
|
||||
}
|
||||
|
||||
printCenteredTitle(" CHECK RESULTS ", '=', 30);
|
||||
dnnType *out, *out_h;
|
||||
int out_dim = net.getOutputDim().tot();
|
||||
readBinaryFile(output_bin, out_dim, &out_h, &out);
|
||||
std::cout<<"CUDNN vs correct"; checkResult(out_dim, out_data, out);
|
||||
std::cout<<"TRT vs correct"; checkResult(out_dim, out_data2, out);
|
||||
std::cout<<"CUDNN vs TRT "; checkResult(out_dim, out_data, out_data2);
|
||||
|
||||
std::cout<<"\n\nDetected objects: \n";
|
||||
dnnType *output_h = new dnnType[rI.output_dim.tot()];
|
||||
checkCuda(cudaMemcpy(output_h, out_data2,
|
||||
rI.output_dim.tot()*sizeof(dnnType), cudaMemcpyDeviceToHost));
|
||||
rI.interpretData(output_h, 608, 608);
|
||||
rI.showImageResult(input_h);
|
||||
return 0;
|
||||
}
|
||||
@@ -0,0 +1,139 @@
|
||||
[net]
|
||||
Training
|
||||
batch=64
|
||||
subdivisions=8
|
||||
# Testing
|
||||
# batch=1
|
||||
# subdivisions=1
|
||||
width=416
|
||||
height=416
|
||||
channels=3
|
||||
momentum=0.9
|
||||
decay=0.0005
|
||||
angle=0
|
||||
saturation = 1.5
|
||||
exposure = 1.5
|
||||
hue=.1
|
||||
|
||||
learning_rate=0.001
|
||||
burn_in=1000
|
||||
max_batches = 500200
|
||||
policy=steps
|
||||
steps=400000,450000
|
||||
scales=.1,.1
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=16
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
[maxpool]
|
||||
size=2
|
||||
stride=2
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=32
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
[maxpool]
|
||||
size=2
|
||||
stride=2
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=64
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
[maxpool]
|
||||
size=2
|
||||
stride=2
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=128
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
[maxpool]
|
||||
size=2
|
||||
stride=2
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=256
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
[maxpool]
|
||||
size=2
|
||||
stride=2
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=512
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
#[maxpool]
|
||||
#size=2
|
||||
#stride=1
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
filters=1024
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
activation=leaky
|
||||
|
||||
###########
|
||||
|
||||
[convolutional]
|
||||
batch_normalize=1
|
||||
size=3
|
||||
stride=1
|
||||
pad=1
|
||||
filters=512
|
||||
activation=leaky
|
||||
|
||||
[convolutional]
|
||||
size=1
|
||||
stride=1
|
||||
pad=1
|
||||
filters=425
|
||||
activation=linear
|
||||
|
||||
[region]
|
||||
anchors = 0.57273, 0.677385, 1.87446, 2.06253, 3.33843, 5.47434, 7.88282, 3.52778, 9.77052, 9.16828
|
||||
bias_match=1
|
||||
classes=80
|
||||
coords=4
|
||||
num=5
|
||||
softmax=1
|
||||
jitter=.2
|
||||
rescore=0
|
||||
|
||||
object_scale=5
|
||||
noobject_scale=1
|
||||
class_scale=1
|
||||
coord_scale=1
|
||||
|
||||
absolute=1
|
||||
thresh = .6
|
||||
random=1
|
||||
@@ -1,19 +1,19 @@
|
||||
#include<iostream>
|
||||
#include "tkdnn.h"
|
||||
|
||||
const char *input_bin = "../tests/yolo-tiny/layers/input.bin";
|
||||
const char *c0_bin = "../tests/yolo-tiny/layers/c0.bin";
|
||||
const char *c2_bin = "../tests/yolo-tiny/layers/c2.bin";
|
||||
const char *c4_bin = "../tests/yolo-tiny/layers/c4.bin";
|
||||
const char *c5_bin = "../tests/yolo-tiny/layers/c5.bin";
|
||||
const char *c6_bin = "../tests/yolo-tiny/layers/c6.bin";
|
||||
const char *c8_bin = "../tests/yolo-tiny/layers/c8.bin";
|
||||
const char *c10_bin = "../tests/yolo-tiny/layers/c10.bin";
|
||||
const char *c11_bin = "../tests/yolo-tiny/layers/c11.bin";
|
||||
const char *c12_bin = "../tests/yolo-tiny/layers/c12.bin";
|
||||
const char *c13_bin = "../tests/yolo-tiny/layers/c13.bin";
|
||||
const char *g14_bin = "../tests/yolo-tiny/layers/g14.bin";
|
||||
const char *output_bin = "../tests/yolo-tiny/layers/output.bin";
|
||||
const char *input_bin = "../tests/yolo_tiny/layers/input.bin";
|
||||
const char *c0_bin = "../tests/yolo_tiny/layers/c0.bin";
|
||||
const char *c2_bin = "../tests/yolo_tiny/layers/c2.bin";
|
||||
const char *c4_bin = "../tests/yolo_tiny/layers/c4.bin";
|
||||
const char *c5_bin = "../tests/yolo_tiny/layers/c5.bin";
|
||||
const char *c6_bin = "../tests/yolo_tiny/layers/c6.bin";
|
||||
const char *c8_bin = "../tests/yolo_tiny/layers/c8.bin";
|
||||
const char *c10_bin = "../tests/yolo_tiny/layers/c10.bin";
|
||||
const char *c11_bin = "../tests/yolo_tiny/layers/c11.bin";
|
||||
const char *c12_bin = "../tests/yolo_tiny/layers/c12.bin";
|
||||
const char *c13_bin = "../tests/yolo_tiny/layers/c13.bin";
|
||||
const char *g14_bin = "../tests/yolo_tiny/layers/g14.bin";
|
||||
const char *output_bin = "../tests/yolo_tiny/layers/output.bin";
|
||||
|
||||
int main() {
|
||||
|
||||
@@ -60,7 +60,7 @@ int main() {
|
||||
net.print();
|
||||
|
||||
//convert network to tensorRT
|
||||
tkDNN::NetworkRT netRT(&net, "yolo-tiny.rt");
|
||||
tkDNN::NetworkRT netRT(&net, "yolo_tiny.rt");
|
||||
|
||||
dnnType *out_data, *out_data2; // cudnn output, tensorRT output
|
||||
|
||||
Reference in New Issue
Block a user