Add Mobilenet2SSDLite test
The new test works both with TensorRT and cuDNN. Preprocessing and Postprocessing are missing. Add ClippedReLU (for ReLU6), groups for Conv2d, additional bias for convolution. Other minors: -move the timer in the detector to measure all the processing time for a given frame (both centernet and yolo); -add int8 flag. Signed-off-by: Micaela Verucchi <micaelaverucchi@gmail.com> Davide Sapienza <sapienza.dav@gmail.com>
This commit is contained in:
xavier
+4
-2
@@ -22,7 +22,7 @@ set(CUDA_NVCC_FLAGS ${CUDA_NVCC_FLAGS} --maxrregcount=32)
|
||||
find_package(CUDNN REQUIRED)
|
||||
|
||||
# compile
|
||||
file(GLOB tkdnn_CUSRC "src/kernels/*.cu" "src/*.cu")
|
||||
file(GLOB tkdnn_CUSRC "src/kernels/*.cu" "src/sorting.cu")
|
||||
cuda_include_directories(${CMAKE_CURRENT_SOURCE_DIR}/include ${CUDA_INCLUDE_DIRS} ${CUDNN_INCLUDE_DIRS})
|
||||
cuda_add_library(kernels SHARED ${tkdnn_CUSRC})
|
||||
|
||||
@@ -44,7 +44,7 @@ set(tkdnn_LIBS kernels ${CUDA_LIBRARIES} ${CUDA_CUBLAS_LIBRARIES} ${CUDNN_LIBRAR
|
||||
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -Wall -std=c++11")
|
||||
include_directories(${CMAKE_CURRENT_SOURCE_DIR}/include ${CUDA_INCLUDE_DIRS} ${OPENCV_INCLUDE_DIRS} ${NVINFER_INCLUDES})
|
||||
add_library(tkDNN SHARED ${tkdnn_SRC})
|
||||
target_link_libraries(tkDNN ${tkdnn_LIBS})
|
||||
target_link_libraries(tkDNN ${tkdnn_LIBS} nvinfer_plugin)
|
||||
|
||||
#static
|
||||
#add_library(tkDNN_static STATIC ${tkdnn_SRC})
|
||||
@@ -94,6 +94,8 @@ target_link_libraries(test_yolo3_berkeley tkDNN)
|
||||
add_executable(test_yolo3_flir tests/yolo3_flir/yolo3_flir.cpp)
|
||||
target_link_libraries(test_yolo3_flir tkDNN)
|
||||
|
||||
add_executable(test_mobilenetv2ssd tests/mobilenetv2ssd/mobilenetv2ssd.cpp)
|
||||
target_link_libraries(test_mobilenetv2ssd tkDNN)
|
||||
|
||||
add_executable(test_resnet101 tests/resnet101/resnet101.cpp)
|
||||
target_link_libraries(test_resnet101 tkDNN)
|
||||
|
||||
+1
-1
@@ -3,5 +3,5 @@ map_points : 101 #number of recall points (0 for all, 101 for COCO, 11 Pascal
|
||||
map_levels : 10 #number of IoU step for the AP
|
||||
map_step : 0.05 #step of IoU
|
||||
IoU_thresh : 0.5 #starting IoU threshold
|
||||
conf_thresh : 0.3 #threshold on the condifence of the bbox
|
||||
conf_thresh : 0.0 #threshold on the condifence of the bbox
|
||||
verbose : false #print on screen information
|
||||
+1
-1
@@ -34,7 +34,7 @@ int main(int argc, char *argv[])
|
||||
char * labels_path = "../demo/COCO_val2017/all_labels.txt";
|
||||
bool show = false;
|
||||
bool write_dets = false;
|
||||
int n_images = 1000;
|
||||
int n_images = 5000;
|
||||
|
||||
if(argc > 1)
|
||||
net = argv[1];
|
||||
|
||||
+32
-20
@@ -14,6 +14,8 @@ enum layerType_t {
|
||||
LAYER_DECONV2D,
|
||||
LAYER_DEFORMCONV2D,
|
||||
LAYER_ACTIVATION,
|
||||
LAYER_ACTIVATION_CRELU,
|
||||
LAYER_ACTIVATION_LEAKY,
|
||||
LAYER_FLATTEN,
|
||||
LAYER_MULADD,
|
||||
LAYER_POOLING,
|
||||
@@ -52,22 +54,24 @@ public:
|
||||
std::string getLayerName() {
|
||||
layerType_t type = getLayerType();
|
||||
switch(type) {
|
||||
case LAYER_DENSE: return "Dense";
|
||||
case LAYER_CONV2D: return "Conv2d";
|
||||
case LAYER_DECONV2D: return "DeConv2d";
|
||||
case LAYER_DEFORMCONV2D:return "DeformConv2d";
|
||||
case LAYER_ACTIVATION: return "Activation";
|
||||
case LAYER_FLATTEN: return "Flatten";
|
||||
case LAYER_MULADD: return "MulAdd";
|
||||
case LAYER_POOLING: return "Pooling";
|
||||
case LAYER_SOFTMAX: return "Softmax";
|
||||
case LAYER_ROUTE: return "Route";
|
||||
case LAYER_REORG: return "Reorg";
|
||||
case LAYER_SHORTCUT: return "Shortcut";
|
||||
case LAYER_UPSAMPLE: return "Upsample";
|
||||
case LAYER_REGION: return "Region";
|
||||
case LAYER_YOLO: return "Yolo";
|
||||
default: return "unknown";
|
||||
case LAYER_DENSE: return "Dense";
|
||||
case LAYER_CONV2D: return "Conv2d";
|
||||
case LAYER_DECONV2D: return "DeConv2d";
|
||||
case LAYER_DEFORMCONV2D: return "DeformConv2d";
|
||||
case LAYER_ACTIVATION: return "Activation";
|
||||
case LAYER_ACTIVATION_CRELU: return "ActivationCReLU";
|
||||
case LAYER_ACTIVATION_LEAKY: return "ActivationLeaky";
|
||||
case LAYER_FLATTEN: return "Flatten";
|
||||
case LAYER_MULADD: return "MulAdd";
|
||||
case LAYER_POOLING: return "Pooling";
|
||||
case LAYER_SOFTMAX: return "Softmax";
|
||||
case LAYER_ROUTE: return "Route";
|
||||
case LAYER_REORG: return "Reorg";
|
||||
case LAYER_SHORTCUT: return "Shortcut";
|
||||
case LAYER_UPSAMPLE: return "Upsample";
|
||||
case LAYER_REGION: return "Region";
|
||||
case LAYER_YOLO: return "Yolo";
|
||||
default: return "unknown";
|
||||
}
|
||||
}
|
||||
|
||||
@@ -145,10 +149,18 @@ class Activation : public Layer {
|
||||
|
||||
public:
|
||||
int act_mode;
|
||||
float ceiling;
|
||||
|
||||
Activation(Network *net, int act_mode);
|
||||
Activation(Network *net, int act_mode, const float ceiling=0.0);
|
||||
virtual ~Activation();
|
||||
virtual layerType_t getLayerType() { return LAYER_ACTIVATION; };
|
||||
virtual layerType_t getLayerType() {
|
||||
if(act_mode == CUDNN_ACTIVATION_CLIPPED_RELU)
|
||||
return LAYER_ACTIVATION_CRELU;
|
||||
else if (act_mode == ACTIVATION_LEAKY)
|
||||
return LAYER_ACTIVATION_LEAKY;
|
||||
else
|
||||
return LAYER_ACTIVATION;
|
||||
};
|
||||
|
||||
virtual dnnType* infer(dataDim_t &dim, dnnType* srcData);
|
||||
|
||||
@@ -165,14 +177,14 @@ class Conv2d : public LayerWgs {
|
||||
public:
|
||||
Conv2d( Network *net, int out_ch, int kernelH, int kernelW,
|
||||
int strideH, int strideW, int paddingH, int paddingW,
|
||||
std::string fname_weights, bool batchnorm = false, bool deConv = false, bool final = false, int groups = 1);
|
||||
std::string fname_weights, bool batchnorm = false, bool deConv = false, bool final = false, int groups = 1, bool additional_bias=false);
|
||||
virtual ~Conv2d();
|
||||
virtual layerType_t getLayerType() { return LAYER_CONV2D; };
|
||||
|
||||
virtual dnnType* infer(dataDim_t &dim, dnnType* srcData);
|
||||
|
||||
int kernelH, kernelW, strideH, strideW, paddingH, paddingW;
|
||||
bool deConv;
|
||||
bool deConv, additional_bias;
|
||||
int groups;
|
||||
|
||||
protected:
|
||||
|
||||
@@ -59,7 +59,7 @@ public:
|
||||
dataDim_t input_dim;
|
||||
dataDim_t getOutputDim();
|
||||
|
||||
bool fp16, dla;
|
||||
bool fp16, dla, int8;
|
||||
bool dontLoadWeights;
|
||||
};
|
||||
|
||||
|
||||
@@ -24,6 +24,7 @@ template<typename T> T readBUF(const char*& buffer)
|
||||
|
||||
using namespace nvinfer1;
|
||||
#include "pluginsRT/ActivationLeakyRT.h"
|
||||
#include "pluginsRT/ActivationReLUCeilingRT.h"
|
||||
#include "pluginsRT/ReorgRT.h"
|
||||
#include "pluginsRT/RegionRT.h"
|
||||
//#include "pluginsRT/RouteRT.h"
|
||||
|
||||
@@ -5,6 +5,7 @@
|
||||
|
||||
void activationELUForward(dnnType* srcData, dnnType* dstData, int size, cudaStream_t stream = cudaStream_t(0));
|
||||
void activationLEAKYForward(dnnType* srcData, dnnType* dstData, int size, cudaStream_t stream = cudaStream_t(0));
|
||||
void activationReLUCeilingForward(dnnType* srcData, dnnType* dstData, int size, const float ceiling, cudaStream_t stream= cudaStream_t(0));
|
||||
void activationLOGISTICForward(dnnType* srcData, dnnType* dstData, int size, cudaStream_t stream = cudaStream_t(0));
|
||||
void activationSIGMOIDForward(dnnType* srcData, dnnType* dstData, int size, cudaStream_t stream = cudaStream_t(0));
|
||||
|
||||
|
||||
@@ -0,0 +1,62 @@
|
||||
#include<cassert>
|
||||
#include "../kernels.h"
|
||||
|
||||
class ActivationReLUCeiling : public IPlugin {
|
||||
|
||||
public:
|
||||
ActivationReLUCeiling(const float ceiling) {
|
||||
this->ceiling = ceiling;
|
||||
}
|
||||
|
||||
~ActivationReLUCeiling(){
|
||||
|
||||
}
|
||||
|
||||
int getNbOutputs() const override {
|
||||
return 1;
|
||||
}
|
||||
|
||||
Dims getOutputDimensions(int index, const Dims* inputs, int nbInputDims) override {
|
||||
return inputs[0];
|
||||
}
|
||||
|
||||
void configure(const Dims* inputDims, int nbInputs, const Dims* outputDims, int nbOutputs, int maxBatchSize) override {
|
||||
size = 1;
|
||||
for(int i=0; i<outputDims[0].nbDims; i++)
|
||||
size *= outputDims[0].d[i];
|
||||
}
|
||||
|
||||
int initialize() override {
|
||||
|
||||
return 0;
|
||||
}
|
||||
|
||||
virtual void terminate() override {
|
||||
}
|
||||
|
||||
virtual size_t getWorkspaceSize(int maxBatchSize) const override {
|
||||
return 0;
|
||||
}
|
||||
|
||||
virtual int enqueue(int batchSize, const void*const * inputs, void** outputs, void* workspace, cudaStream_t stream) override {
|
||||
|
||||
activationReLUCeilingForward((dnnType*)reinterpret_cast<const dnnType*>(inputs[0]),
|
||||
reinterpret_cast<dnnType*>(outputs[0]), size, ceiling, stream);
|
||||
return 0;
|
||||
}
|
||||
|
||||
|
||||
virtual size_t getSerializationSize() override {
|
||||
return 1*sizeof(int) + 1*sizeof(float);
|
||||
}
|
||||
|
||||
virtual void serialize(void* buffer) override {
|
||||
char *buf = reinterpret_cast<char*>(buffer);
|
||||
tk::dnn::writeBUF(buf, ceiling);
|
||||
tk::dnn::writeBUF(buf, size);
|
||||
|
||||
}
|
||||
|
||||
int size;
|
||||
float ceiling;
|
||||
};
|
||||
+3
-2
@@ -5,10 +5,11 @@
|
||||
|
||||
namespace tk { namespace dnn {
|
||||
|
||||
Activation::Activation(Network *net, int act_mode) :
|
||||
Activation::Activation(Network *net, int act_mode, const float ceiling) :
|
||||
Layer(net) {
|
||||
|
||||
this->act_mode = act_mode;
|
||||
this->ceiling = ceiling;
|
||||
checkCuda( cudaMalloc(&dstData, input_dim.tot()*sizeof(dnnType)) );
|
||||
|
||||
if(int(act_mode) < 100) {
|
||||
@@ -31,7 +32,7 @@ Activation::Activation(Network *net, int act_mode) :
|
||||
checkCUDNN( cudnnSetActivationDescriptor(activDesc,
|
||||
(cudnnActivationMode_t) act_mode,
|
||||
CUDNN_PROPAGATE_NAN,
|
||||
0.0) );
|
||||
ceiling) );
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
@@ -294,8 +294,6 @@ void CenternetDetection::update(cv::Mat &imageORIG) {
|
||||
netRT->infer(dim2, input_d);
|
||||
TIMER_STOP
|
||||
dim2.print();
|
||||
|
||||
stats.push_back(t_ns);
|
||||
}
|
||||
// checkResult(dim2.tot(), input_h, input);
|
||||
step_t = std::chrono::steady_clock::now();
|
||||
@@ -456,5 +454,6 @@ void CenternetDetection::update(cv::Mat &imageORIG) {
|
||||
|
||||
std::cout<<"TOTAL: \n";
|
||||
TIMER_STOP
|
||||
stats.push_back(t_ns);
|
||||
}
|
||||
}}
|
||||
+23
-11
@@ -100,7 +100,7 @@ void Conv2d::inferCUDNN(dnnType* srcData, bool back) {
|
||||
&beta, dstTensorDesc, dstData));
|
||||
}
|
||||
|
||||
if(!batchnorm) {
|
||||
if(!batchnorm && !additional_bias) { //CHECK WITH IF CORRECT
|
||||
// bias
|
||||
alpha = dnnType(1);
|
||||
beta = dnnType(1);
|
||||
@@ -108,23 +108,34 @@ void Conv2d::inferCUDNN(dnnType* srcData, bool back) {
|
||||
&alpha, biasTensorDesc, bias_d,
|
||||
&beta, dstTensorDesc, dstData) );
|
||||
} else {
|
||||
alpha = dnnType(1);
|
||||
beta = dnnType(0);
|
||||
checkCUDNN( cudnnBatchNormalizationForwardInference(net->cudnnHandle,
|
||||
CUDNN_BATCHNORM_SPATIAL, &alpha, &beta,
|
||||
dstTensorDesc, dstData, dstTensorDesc,
|
||||
dstData, biasTensorDesc, //same tensor descriptor as bias
|
||||
scales_d, bias_d, mean_d, variance_d,
|
||||
TKDNN_BN_MIN_EPSILON) );
|
||||
if(additional_bias)
|
||||
{
|
||||
alpha = dnnType(1);
|
||||
beta = dnnType(1);
|
||||
checkCUDNN( cudnnAddTensor(net->cudnnHandle,
|
||||
&alpha, biasTensorDesc, bias2_d,
|
||||
&beta, dstTensorDesc, dstData) );
|
||||
}
|
||||
if(batchnorm)
|
||||
{
|
||||
alpha = dnnType(1);
|
||||
beta = dnnType(0);
|
||||
checkCUDNN( cudnnBatchNormalizationForwardInference(net->cudnnHandle,
|
||||
CUDNN_BATCHNORM_SPATIAL, &alpha, &beta,
|
||||
dstTensorDesc, dstData, dstTensorDesc,
|
||||
dstData, biasTensorDesc, //same tensor descriptor as bias
|
||||
scales_d, bias_d, mean_d, variance_d,
|
||||
TKDNN_BN_MIN_EPSILON) );
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
Conv2d::Conv2d( Network *net, int out_ch, int kernelH, int kernelW,
|
||||
int strideH, int strideW, int paddingH, int paddingW,
|
||||
std::string fname_weights, bool batchnorm, bool deConv, bool final, int groups) :
|
||||
std::string fname_weights, bool batchnorm, bool deConv, bool final, int groups, bool additional_bias) :
|
||||
|
||||
LayerWgs(net, net->getOutputDim().c, out_ch, kernelH, kernelW, 1,
|
||||
fname_weights, batchnorm, false, final, deConv, groups) {
|
||||
fname_weights, batchnorm, additional_bias, final, deConv, groups) {
|
||||
this->kernelH = kernelH;
|
||||
this->kernelW = kernelW;
|
||||
this->strideH = strideH;
|
||||
@@ -133,6 +144,7 @@ Conv2d::Conv2d( Network *net, int out_ch, int kernelH, int kernelW,
|
||||
this->paddingW = paddingW;
|
||||
this->deConv = deConv;
|
||||
this->groups = groups;
|
||||
this->additional_bias = additional_bias;
|
||||
|
||||
if(!deConv) {
|
||||
output_dim.n = input_dim.n;
|
||||
|
||||
+1
-5
@@ -9,11 +9,7 @@ namespace tk { namespace dnn {
|
||||
LayerWgs::LayerWgs(Network *net, int inputs, int outputs,
|
||||
int kh, int kw, int kl,
|
||||
std::string fname_weights, bool batchnorm, bool additional_bias, bool final, bool deConv, int groups) : Layer(net, final) {
|
||||
|
||||
if(deConv)
|
||||
inputs = inputs/groups;
|
||||
else
|
||||
outputs = outputs/groups;
|
||||
inputs = inputs/groups;
|
||||
|
||||
this->inputs = inputs;
|
||||
this->outputs = outputs;
|
||||
|
||||
+12
-6
@@ -22,19 +22,25 @@ Network::Network(dataDim_t input_dim) {
|
||||
|
||||
fp16 = false;
|
||||
dla = false;
|
||||
int8 = false;
|
||||
if(const char* env_p = std::getenv("TKDNN_MODE")) {
|
||||
if(strcmp(env_p, "FP16") == 0)
|
||||
fp16 = true;
|
||||
else if(strcmp(env_p, "DLA") == 0) {
|
||||
dla = true;
|
||||
fp16 = true;
|
||||
}
|
||||
else if(strcmp(env_p, "DLA") == 0) {
|
||||
dla = true;
|
||||
fp16 = true;
|
||||
}
|
||||
else if(strcmp(env_p, "INT8") == 0) {
|
||||
int8 = true;
|
||||
}
|
||||
}
|
||||
|
||||
if(fp16)
|
||||
std::cout<<COL_REDB<<"!! FP16 INERENCE ENABLED !!"<<COL_END<<"\n";
|
||||
std::cout<<COL_REDB<<"!! FP16 INFERENCE ENABLED !!"<<COL_END<<"\n";
|
||||
if(dla)
|
||||
std::cout<<COL_GREENB<<"!! DLA INERENCE ENABLED !!"<<COL_END<<"\n";
|
||||
std::cout<<COL_GREENB<<"!! DLA INFERENCE ENABLED !!"<<COL_END<<"\n";
|
||||
if(int8)
|
||||
std::cout<<COL_ORANGEB<<"!! INT8 INFERENCE ENABLED !!"<<COL_END<<"\n";
|
||||
|
||||
|
||||
checkCUDNN( cudnnCreate(&cudnnHandle) );
|
||||
|
||||
+22
-3
@@ -9,6 +9,7 @@
|
||||
#include "utils.h"
|
||||
#include "NvInfer.h"
|
||||
#include "NetworkRT.h"
|
||||
// #include "calibrator.h"
|
||||
|
||||
using namespace nvinfer1;
|
||||
|
||||
@@ -58,6 +59,14 @@ NetworkRT::NetworkRT(Network *net, const char *name) {
|
||||
builderRT->setDefaultDeviceType(DeviceType::kDLA);
|
||||
builderRT->setDLACore(0);
|
||||
}
|
||||
// if(net->int8 && builderRT->platformHasFastInt8())
|
||||
// {
|
||||
// dtRT = DataType::kINT8;
|
||||
// builderRT->setInt8Mode(true);
|
||||
// Int8EntropyCalibrator calibrator(1, "../demo/images.txt","../demo/yolov3-calibration.table", 416*416*3, 416, 416);
|
||||
// builderRT->setInt8Calibrator((nvinfer1::IInt8Calibrator * )&calibrator);
|
||||
// // builderRT->setStrictTypeConstraints(true);
|
||||
// }
|
||||
|
||||
//add input layer
|
||||
ITensor *input = networkRT->addInput("data", DataType::kFLOAT,
|
||||
@@ -164,7 +173,7 @@ ILayer* NetworkRT::convert_layer(ITensor *input, Layer *l) {
|
||||
return convert_layer(input, (Conv2d*) l);
|
||||
if(type == LAYER_POOLING)
|
||||
return convert_layer(input, (Pooling*) l);
|
||||
if(type == LAYER_ACTIVATION)
|
||||
if(type == LAYER_ACTIVATION || type == LAYER_ACTIVATION_CRELU || type == LAYER_ACTIVATION_LEAKY)
|
||||
return convert_layer(input, (Activation*) l);
|
||||
if(type == LAYER_SOFTMAX)
|
||||
return convert_layer(input, (Softmax*) l);
|
||||
@@ -337,7 +346,12 @@ ILayer* NetworkRT::convert_layer(ITensor *input, Activation *l) {
|
||||
IActivationLayer *lRT = networkRT->addActivation(*input, ActivationType::kSIGMOID);
|
||||
checkNULL(lRT);
|
||||
return lRT;
|
||||
|
||||
}
|
||||
else if(l->act_mode == CUDNN_ACTIVATION_CLIPPED_RELU) {
|
||||
IPlugin *plugin = new ActivationReLUCeiling(l->ceiling);
|
||||
IPluginLayer *lRT = networkRT->addPlugin(&input, 1, *plugin);
|
||||
checkNULL(lRT);
|
||||
return lRT;
|
||||
} else {
|
||||
FatalError("this Activation mode is not yet implemented");
|
||||
return NULL;
|
||||
@@ -535,11 +549,16 @@ IPlugin* PluginFactory::createPlugin(const char* layerName, const void* serialDa
|
||||
std::string name(layerName);
|
||||
std::cout<<name<<std::endl;
|
||||
|
||||
if(name.find("Activation") == 0) {
|
||||
if(name.find("ActivationLeaky") == 0) {
|
||||
ActivationLeakyRT *a = new ActivationLeakyRT();
|
||||
a->size = readBUF<int>(buf);
|
||||
return a;
|
||||
}
|
||||
if(name.find("ActivationCReLU") == 0) {
|
||||
ActivationReLUCeiling *a = new ActivationReLUCeiling(readBUF<float>(buf));
|
||||
a->size = readBUF<int>(buf);
|
||||
return a;
|
||||
}
|
||||
|
||||
if(name.find("Region") == 0) {
|
||||
RegionRT *r = new RegionRT(readBUF<int>(buf), //classes
|
||||
|
||||
@@ -63,6 +63,7 @@ bool Yolo3Detection::init(std::string tensor_path) {
|
||||
|
||||
void Yolo3Detection::update(cv::Mat &imageORIG) {
|
||||
|
||||
TIMER_START
|
||||
if(!imageORIG.data) {
|
||||
std::cout<<"YOLO: NO IMAGE DATA\n";
|
||||
return;
|
||||
@@ -100,7 +101,6 @@ void Yolo3Detection::update(cv::Mat &imageORIG) {
|
||||
stats.push_back(t_ns);
|
||||
}
|
||||
|
||||
TIMER_START
|
||||
// compute dets
|
||||
ndets = 0;
|
||||
for(int i=0; i<netRT->pluginFactory->n_yolos; i++) {
|
||||
@@ -109,7 +109,6 @@ void Yolo3Detection::update(cv::Mat &imageORIG) {
|
||||
yolo[i]->computeDetections(dets, ndets, netRT->input_dim.w, netRT->input_dim.h, thresh);
|
||||
}
|
||||
tk::dnn::Yolo::mergeDetections(dets, ndets, classes);
|
||||
TIMER_STOP
|
||||
|
||||
// fill detected
|
||||
detected.clear();
|
||||
@@ -148,6 +147,8 @@ void Yolo3Detection::update(cv::Mat &imageORIG) {
|
||||
detected.push_back(res);
|
||||
}
|
||||
}
|
||||
TIMER_STOP
|
||||
stats.push_back(t_ns);
|
||||
|
||||
}
|
||||
|
||||
|
||||
@@ -0,0 +1,33 @@
|
||||
#include "kernels.h"
|
||||
|
||||
__global__
|
||||
void activation_relu_ceiling(dnnType *input, dnnType *output, int size, const float ceiling) {
|
||||
|
||||
int i = blockDim.x*blockIdx.x + threadIdx.x;
|
||||
|
||||
if(i<size) {
|
||||
if (input[i]>0)
|
||||
{
|
||||
if (input[i]>ceiling)
|
||||
output[i] = ceiling;
|
||||
else
|
||||
output[i] = input[i];
|
||||
}
|
||||
else
|
||||
output[i] = 0.0f;
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
/**
|
||||
Relu ceiling activation function
|
||||
*/
|
||||
void activationReLUCeilingForward(dnnType* srcData, dnnType* dstData, int size, const float ceiling, cudaStream_t stream)
|
||||
{
|
||||
int blocks = (size+255)/256;
|
||||
int threads = 256;
|
||||
|
||||
activation_relu_ceiling<<<blocks, threads, 0, stream>>>(srcData, dstData, size, ceiling);
|
||||
}
|
||||
|
||||
|
||||
@@ -0,0 +1,448 @@
|
||||
#include <iostream>
|
||||
#include "tkdnn.h"
|
||||
|
||||
const char *output_bin = "../tests/mobilenetv2ssd/debug/regression_headers-5.bin";
|
||||
const char *input_bin = "../tests/mobilenetv2ssd/debug/input.bin";
|
||||
|
||||
const char *conv0_bin = "../tests/mobilenetv2ssd/layers/base_net-0-0.bin";
|
||||
const char *inverted_residual1[]={
|
||||
"../tests/mobilenetv2ssd/layers/base_net-1-conv-0.bin",
|
||||
"../tests/mobilenetv2ssd/layers/base_net-1-conv-3.bin"};
|
||||
const char *inverted_residual2[]={
|
||||
"../tests/mobilenetv2ssd/layers/base_net-2-conv-0.bin",
|
||||
"../tests/mobilenetv2ssd/layers/base_net-2-conv-3.bin",
|
||||
"../tests/mobilenetv2ssd/layers/base_net-2-conv-6.bin"};
|
||||
const char *inverted_residual3[]={
|
||||
"../tests/mobilenetv2ssd/layers/base_net-3-conv-0.bin",
|
||||
"../tests/mobilenetv2ssd/layers/base_net-3-conv-3.bin",
|
||||
"../tests/mobilenetv2ssd/layers/base_net-3-conv-6.bin"};
|
||||
const char *inverted_residual4[]={
|
||||
"../tests/mobilenetv2ssd/layers/base_net-4-conv-0.bin",
|
||||
"../tests/mobilenetv2ssd/layers/base_net-4-conv-3.bin",
|
||||
"../tests/mobilenetv2ssd/layers/base_net-4-conv-6.bin"};
|
||||
const char *inverted_residual5[]={
|
||||
"../tests/mobilenetv2ssd/layers/base_net-5-conv-0.bin",
|
||||
"../tests/mobilenetv2ssd/layers/base_net-5-conv-3.bin",
|
||||
"../tests/mobilenetv2ssd/layers/base_net-5-conv-6.bin"};
|
||||
const char *inverted_residual6[]={
|
||||
"../tests/mobilenetv2ssd/layers/base_net-6-conv-0.bin",
|
||||
"../tests/mobilenetv2ssd/layers/base_net-6-conv-3.bin",
|
||||
"../tests/mobilenetv2ssd/layers/base_net-6-conv-6.bin"};
|
||||
const char *inverted_residual7[]={
|
||||
"../tests/mobilenetv2ssd/layers/base_net-7-conv-0.bin",
|
||||
"../tests/mobilenetv2ssd/layers/base_net-7-conv-3.bin",
|
||||
"../tests/mobilenetv2ssd/layers/base_net-7-conv-6.bin"};
|
||||
const char *inverted_residual8[]={
|
||||
"../tests/mobilenetv2ssd/layers/base_net-8-conv-0.bin",
|
||||
"../tests/mobilenetv2ssd/layers/base_net-8-conv-3.bin",
|
||||
"../tests/mobilenetv2ssd/layers/base_net-8-conv-6.bin"};
|
||||
const char *inverted_residual9[]={
|
||||
"../tests/mobilenetv2ssd/layers/base_net-9-conv-0.bin",
|
||||
"../tests/mobilenetv2ssd/layers/base_net-9-conv-3.bin",
|
||||
"../tests/mobilenetv2ssd/layers/base_net-9-conv-6.bin"};
|
||||
const char *inverted_residual10[]={
|
||||
"../tests/mobilenetv2ssd/layers/base_net-10-conv-0.bin",
|
||||
"../tests/mobilenetv2ssd/layers/base_net-10-conv-3.bin",
|
||||
"../tests/mobilenetv2ssd/layers/base_net-10-conv-6.bin"};
|
||||
const char *inverted_residual11[]={
|
||||
"../tests/mobilenetv2ssd/layers/base_net-11-conv-0.bin",
|
||||
"../tests/mobilenetv2ssd/layers/base_net-11-conv-3.bin",
|
||||
"../tests/mobilenetv2ssd/layers/base_net-11-conv-6.bin"};
|
||||
const char *inverted_residual12[]={
|
||||
"../tests/mobilenetv2ssd/layers/base_net-12-conv-0.bin",
|
||||
"../tests/mobilenetv2ssd/layers/base_net-12-conv-3.bin",
|
||||
"../tests/mobilenetv2ssd/layers/base_net-12-conv-6.bin"};
|
||||
const char *inverted_residual13[]={
|
||||
"../tests/mobilenetv2ssd/layers/base_net-13-conv-0.bin",
|
||||
"../tests/mobilenetv2ssd/layers/base_net-13-conv-3.bin",
|
||||
"../tests/mobilenetv2ssd/layers/base_net-13-conv-6.bin"};
|
||||
const char *inverted_residual14[]={
|
||||
"../tests/mobilenetv2ssd/layers/base_net-14-conv-0.bin",
|
||||
"../tests/mobilenetv2ssd/layers/base_net-14-conv-3.bin",
|
||||
"../tests/mobilenetv2ssd/layers/base_net-14-conv-6.bin"};
|
||||
const char *inverted_residual15[]={
|
||||
"../tests/mobilenetv2ssd/layers/base_net-15-conv-0.bin",
|
||||
"../tests/mobilenetv2ssd/layers/base_net-15-conv-3.bin",
|
||||
"../tests/mobilenetv2ssd/layers/base_net-15-conv-6.bin"};
|
||||
const char *inverted_residual16[]={
|
||||
"../tests/mobilenetv2ssd/layers/base_net-16-conv-0.bin",
|
||||
"../tests/mobilenetv2ssd/layers/base_net-16-conv-3.bin",
|
||||
"../tests/mobilenetv2ssd/layers/base_net-16-conv-6.bin"};
|
||||
const char *inverted_residual17[]={
|
||||
"../tests/mobilenetv2ssd/layers/base_net-17-conv-0.bin",
|
||||
"../tests/mobilenetv2ssd/layers/base_net-17-conv-3.bin",
|
||||
"../tests/mobilenetv2ssd/layers/base_net-17-conv-6.bin"};
|
||||
|
||||
const char *conv18 = "../tests/mobilenetv2ssd/layers/base_net-18-0.bin";
|
||||
|
||||
const char *extras0[]={
|
||||
"../tests/mobilenetv2ssd/layers/extras-0-conv-0.bin",
|
||||
"../tests/mobilenetv2ssd/layers/extras-0-conv-3.bin",
|
||||
"../tests/mobilenetv2ssd/layers/extras-0-conv-6.bin"};
|
||||
const char *extras1[]={
|
||||
"../tests/mobilenetv2ssd/layers/extras-1-conv-0.bin",
|
||||
"../tests/mobilenetv2ssd/layers/extras-1-conv-3.bin",
|
||||
"../tests/mobilenetv2ssd/layers/extras-1-conv-6.bin"};
|
||||
const char *extras2[]={
|
||||
"../tests/mobilenetv2ssd/layers/extras-2-conv-0.bin",
|
||||
"../tests/mobilenetv2ssd/layers/extras-2-conv-3.bin",
|
||||
"../tests/mobilenetv2ssd/layers/extras-2-conv-6.bin"};
|
||||
const char *extras3[]={
|
||||
"../tests/mobilenetv2ssd/layers/extras-3-conv-0.bin",
|
||||
"../tests/mobilenetv2ssd/layers/extras-3-conv-3.bin",
|
||||
"../tests/mobilenetv2ssd/layers/extras-3-conv-6.bin"};
|
||||
|
||||
const char *classification_header0[]={
|
||||
"../tests/mobilenetv2ssd/layers/classification_headers-0-0.bin",
|
||||
"../tests/mobilenetv2ssd/layers/classification_headers-0-3.bin"};
|
||||
const char *classification_header1[]={
|
||||
"../tests/mobilenetv2ssd/layers/classification_headers-1-0.bin",
|
||||
"../tests/mobilenetv2ssd/layers/classification_headers-1-3.bin"};
|
||||
const char *classification_header2[]={
|
||||
"../tests/mobilenetv2ssd/layers/classification_headers-2-0.bin",
|
||||
"../tests/mobilenetv2ssd/layers/classification_headers-2-3.bin"};
|
||||
const char *classification_header3[]={
|
||||
"../tests/mobilenetv2ssd/layers/classification_headers-3-0.bin",
|
||||
"../tests/mobilenetv2ssd/layers/classification_headers-3-3.bin"};
|
||||
const char *classification_header4[]={
|
||||
"../tests/mobilenetv2ssd/layers/classification_headers-4-0.bin",
|
||||
"../tests/mobilenetv2ssd/layers/classification_headers-4-3.bin"};
|
||||
|
||||
const char *classification_header5 = "../tests/mobilenetv2ssd/layers/classification_headers-5.bin";
|
||||
|
||||
const char *regression_header0[]={
|
||||
"../tests/mobilenetv2ssd/layers/regression_headers-0-0.bin",
|
||||
"../tests/mobilenetv2ssd/layers/regression_headers-0-3.bin"};
|
||||
const char *regression_header1[]={
|
||||
"../tests/mobilenetv2ssd/layers/regression_headers-1-0.bin",
|
||||
"../tests/mobilenetv2ssd/layers/regression_headers-1-3.bin"};
|
||||
const char *regression_header2[]={
|
||||
"../tests/mobilenetv2ssd/layers/regression_headers-2-0.bin",
|
||||
"../tests/mobilenetv2ssd/layers/regression_headers-2-3.bin"};
|
||||
const char *regression_header3[]={
|
||||
"../tests/mobilenetv2ssd/layers/regression_headers-3-0.bin",
|
||||
"../tests/mobilenetv2ssd/layers/regression_headers-3-3.bin"};
|
||||
const char *regression_header4[]={
|
||||
"../tests/mobilenetv2ssd/layers/regression_headers-4-0.bin",
|
||||
"../tests/mobilenetv2ssd/layers/regression_headers-4-3.bin"};
|
||||
|
||||
const char *regression_header5 = "../tests/mobilenetv2ssd/layers/regression_headers-5.bin";
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
int main()
|
||||
{
|
||||
|
||||
// Network layout
|
||||
tk::dnn::dataDim_t dim(1, 3, 300, 300, 1);
|
||||
tk::dnn::Network net(dim);
|
||||
|
||||
tk::dnn::Conv2d conv1(&net, 32, 3, 3, 2, 2, 1, 1, conv0_bin, true);
|
||||
tk::dnn::Activation relu3(&net, CUDNN_ACTIVATION_RELU);
|
||||
|
||||
//Inverted Residual 1
|
||||
|
||||
tk::dnn::Conv2d conv2(&net, 32, 3, 3, 1, 1, 1, 1, inverted_residual1[0], true,false, false,32);
|
||||
tk::dnn::Activation relu5(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d conv3(&net, 16, 1, 1, 1, 1, 0, 0, inverted_residual1[1], true);
|
||||
|
||||
|
||||
//Inverted Residual 2
|
||||
tk::dnn::Conv2d ir_2_conv1(&net, 96, 1, 1, 1, 1, 0, 0, inverted_residual2[0], true);
|
||||
tk::dnn::Activation relu_2_1(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d ir_2_conv2(&net, 96, 3, 3, 2, 2, 1, 1, inverted_residual2[1], true, false, false, 96);
|
||||
tk::dnn::Activation relu_2_2(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d ir_2_conv3(&net, 24, 1, 1, 1, 1, 0, 0, inverted_residual2[2], true);
|
||||
|
||||
//Inverted Residual 3
|
||||
tk::dnn::Layer *last = &ir_2_conv3;
|
||||
tk::dnn::Conv2d ir_3_conv1(&net, 144, 1, 1, 1, 1, 0, 0, inverted_residual3[0], true);
|
||||
tk::dnn::Activation relu_3_1(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d ir_3_conv2(&net, 144, 3, 3, 1, 1, 1, 1, inverted_residual3[1], true, false, false, 144);
|
||||
tk::dnn::Activation relu_3_2(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d ir_3_conv3(&net, 24, 1, 1, 1, 1, 0, 0, inverted_residual3[2], true);
|
||||
|
||||
tk::dnn::Shortcut s3_0 (&net, last);
|
||||
// //Inverted Residual 4
|
||||
tk::dnn::Conv2d ir_4_conv1(&net, 144, 1, 1, 1, 1, 0, 0, inverted_residual4[0], true);
|
||||
tk::dnn::Activation relu_4_1(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d ir_4_conv2(&net, 144, 3, 3, 2, 2, 1, 1, inverted_residual4[1], true, false, false, 144);
|
||||
tk::dnn::Activation relu_4_2(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d ir_4_conv3(&net, 32, 1, 1, 1, 1, 0, 0, inverted_residual4[2], true);
|
||||
|
||||
// // //Inverted Residual 5
|
||||
last = &ir_4_conv3;
|
||||
tk::dnn::Conv2d ir_5_conv1(&net, 192, 1, 1, 1, 1, 0, 0, inverted_residual5[0], true);
|
||||
tk::dnn::Activation relu_5_1(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d ir_5_conv2(&net, 192, 3, 3, 1, 1, 1, 1, inverted_residual5[1], true, false, false, 192);
|
||||
tk::dnn::Activation relu_5_2(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d ir_5_conv3(&net, 32, 1, 1, 1, 1, 0, 0, inverted_residual5[2], true);
|
||||
|
||||
tk::dnn::Shortcut s5_0 (&net, last);
|
||||
// // // //Inverted Residual 6
|
||||
last = &s5_0;
|
||||
tk::dnn::Conv2d ir_6_conv1(&net, 192, 1, 1, 1, 1, 0, 0, inverted_residual6[0], true);
|
||||
tk::dnn::Activation relu_6_1(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d ir_6_conv2(&net, 192, 3, 3, 1, 1, 1, 1, inverted_residual6[1], true, false, false, 192);
|
||||
tk::dnn::Activation relu_6_2(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d ir_6_conv3(&net, 32, 1, 1, 1, 1, 0, 0, inverted_residual6[2], true);
|
||||
|
||||
tk::dnn::Shortcut s6_0 (&net, last);
|
||||
//Inverted Residual 7
|
||||
tk::dnn::Conv2d ir_7_conv1(&net, 192, 1, 1, 1, 1, 0, 0, inverted_residual7[0], true);
|
||||
tk::dnn::Activation relu_7_1(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d ir_7_conv2(&net, 192, 3, 3, 2, 2, 1, 1, inverted_residual7[1], true, false, false, 192);
|
||||
tk::dnn::Activation relu_7_2(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d ir_7_conv3(&net, 64, 1, 1, 1, 1, 0, 0, inverted_residual7[2], true);
|
||||
|
||||
// //Inverted Residual 8
|
||||
last = &ir_7_conv3;
|
||||
tk::dnn::Conv2d ir_8_conv1(&net, 384, 1, 1, 1, 1, 0, 0, inverted_residual8[0], true);
|
||||
tk::dnn::Activation relu_8_1(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d ir_8_conv2(&net, 384, 3, 3, 1, 1, 1, 1, inverted_residual8[1], true, false, false, 384);
|
||||
tk::dnn::Activation relu_8_2(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d ir_8_conv3(&net, 64, 1, 1, 1, 1, 0, 0, inverted_residual8[2], true);
|
||||
|
||||
tk::dnn::Shortcut s8_0 (&net, last);
|
||||
//Inverted Residual 9
|
||||
last = &s8_0;
|
||||
tk::dnn::Conv2d ir_9_conv1(&net, 384, 1, 1, 1, 1, 0, 0, inverted_residual9[0], true);
|
||||
tk::dnn::Activation relu_9_1(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d ir_9_conv2(&net, 384, 3, 3, 1, 1, 1, 1, inverted_residual9[1], true, false, false, 384);
|
||||
tk::dnn::Activation relu_9_2(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d ir_9_conv3(&net, 64, 1, 1, 1, 1, 0, 0, inverted_residual9[2], true);
|
||||
|
||||
tk::dnn::Shortcut s9_0 (&net, last);
|
||||
//Inverted Residual 10
|
||||
last = &s9_0;
|
||||
tk::dnn::Conv2d ir_10_conv1(&net, 384, 1, 1, 1, 1, 0, 0, inverted_residual10[0], true);
|
||||
tk::dnn::Activation relu_10_1(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d ir_10_conv2(&net, 384, 3, 3, 1, 1, 1, 1, inverted_residual10[1], true, false, false, 384);
|
||||
tk::dnn::Activation relu_10_2(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d ir_10_conv3(&net, 64, 1, 1, 1, 1, 0, 0, inverted_residual10[2], true);
|
||||
|
||||
tk::dnn::Shortcut s10_0 (&net, last);
|
||||
//Inverted Residual 11
|
||||
tk::dnn::Conv2d ir_11_conv1(&net, 384, 1, 1, 1, 1, 0, 0, inverted_residual11[0], true);
|
||||
tk::dnn::Activation relu_11_1(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d ir_11_conv2(&net, 384, 3, 3, 1, 1, 1, 1, inverted_residual11[1], true, false, false, 384);
|
||||
tk::dnn::Activation relu_11_2(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d ir_11_conv3(&net, 96, 1, 1, 1, 1, 0, 0, inverted_residual11[2], true);
|
||||
|
||||
last = &ir_11_conv3;
|
||||
//Inverted Residual 12
|
||||
tk::dnn::Conv2d ir_12_conv1(&net, 576, 1, 1, 1, 1, 0, 0, inverted_residual12[0], true);
|
||||
tk::dnn::Activation relu_12_1(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d ir_12_conv2(&net, 576, 3, 3, 1, 1, 1, 1, inverted_residual12[1], true, false, false, 576);
|
||||
tk::dnn::Activation relu_12_2(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d ir_12_conv3(&net, 96, 1, 1, 1, 1, 0, 0, inverted_residual12[2], true);
|
||||
|
||||
tk::dnn::Shortcut s12_0 (&net, last);
|
||||
last = &s12_0;
|
||||
//Inverted Residual 13
|
||||
tk::dnn::Conv2d ir_13_conv1(&net, 576, 1, 1, 1, 1, 0, 0, inverted_residual13[0], true);
|
||||
tk::dnn::Activation relu_13_1(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d ir_13_conv2(&net, 576, 3, 3, 1, 1, 1, 1, inverted_residual13[1], true, false, false, 576);
|
||||
tk::dnn::Activation relu_13_2(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d ir_13_conv3(&net, 96, 1, 1, 1, 1, 0, 0, inverted_residual13[2], true);
|
||||
|
||||
tk::dnn::Shortcut s13_0 (&net, last);
|
||||
// //Inverted Residual 14
|
||||
tk::dnn::Conv2d ir_14_conv1(&net, 576, 1, 1, 1, 1, 0, 0, inverted_residual14[0], true);
|
||||
tk::dnn::Activation relu_14_1(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d ir_14_conv2(&net, 576, 3, 3, 2, 2, 1, 1, inverted_residual14[1], true, false, false, 576);
|
||||
tk::dnn::Activation relu_14_2(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d ir_14_conv3(&net, 160, 1, 1, 1, 1, 0, 0, inverted_residual14[2], true);
|
||||
|
||||
// //Inverted Residual 15
|
||||
last = &ir_14_conv3;
|
||||
tk::dnn::Conv2d ir_15_conv1(&net, 960, 1, 1, 1, 1, 0, 0, inverted_residual15[0], true);
|
||||
tk::dnn::Activation relu_15_1(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d ir_15_conv2(&net, 960, 3, 3, 1, 1, 1, 1, inverted_residual15[1], true, false, false, 960);
|
||||
tk::dnn::Activation relu_15_2(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d ir_15_conv3(&net, 160, 1, 1, 1, 1, 0, 0, inverted_residual15[2], true);
|
||||
|
||||
tk::dnn::Shortcut s15_0 (&net, last);
|
||||
//Inverted Residual 16
|
||||
last = &s15_0;
|
||||
tk::dnn::Conv2d ir_16_conv1(&net, 960, 1, 1, 1, 1, 0, 0, inverted_residual16[0], true);
|
||||
tk::dnn::Activation relu_16_1(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d ir_16_conv2(&net, 960, 3, 3, 1, 1, 1, 1, inverted_residual16[1], true, false, false, 960);
|
||||
tk::dnn::Activation relu_16_2(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d ir_16_conv3(&net, 160, 1, 1, 1, 1, 0, 0, inverted_residual16[2], true);
|
||||
|
||||
tk::dnn::Shortcut s16_0 (&net, last);
|
||||
//Inverted Residual 17
|
||||
tk::dnn::Conv2d ir_17_conv1(&net, 960, 1, 1, 1, 1, 0, 0, inverted_residual17[0], true);
|
||||
tk::dnn::Activation relu_17_1(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d ir_17_conv2(&net, 960, 3, 3, 1, 1, 1, 1, inverted_residual17[1], true, false, false, 960);
|
||||
tk::dnn::Activation relu_17_2(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d ir_17_conv3(&net, 320, 1, 1, 1, 1, 0, 0, inverted_residual17[2], true);
|
||||
|
||||
//Conv 18
|
||||
tk::dnn::Conv2d ir_18_conv1(&net, 1280, 1, 1, 1, 1, 0, 0, conv18, true);
|
||||
tk::dnn::Activation relu_18_1(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Layer * header_1[1] = {&relu_18_1};
|
||||
|
||||
// //extras Inverted Residual 0
|
||||
tk::dnn::Conv2d e_0_conv1(&net, 256, 1, 1, 1, 1, 0, 0, extras0[0], true);
|
||||
tk::dnn::Activation e_relu_0_1(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d e_0_conv2(&net, 256, 3, 3, 2, 2, 1, 1, extras0[1], true, false, false, 256);
|
||||
tk::dnn::Activation e_relu_0_2(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d e_0_conv3(&net, 512, 1, 1, 1, 1, 0, 0, extras0[2], true);
|
||||
tk::dnn::Layer * header_2[1] = {&e_0_conv3};
|
||||
|
||||
// //extras Inverted Residual 1
|
||||
tk::dnn::Conv2d e_1_conv1(&net, 128, 1, 1, 1, 1, 0, 0, extras1[0], true);
|
||||
tk::dnn::Activation e_relu_1_1(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d e_1_conv2(&net, 128, 3, 3, 2, 2, 1, 1, extras1[1], true, false, false, 128);
|
||||
tk::dnn::Activation e_relu_1_2(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d e_1_conv3(&net, 256, 1, 1, 1, 1, 0, 0, extras1[2], true);
|
||||
tk::dnn::Layer * header_3[1] = {&e_1_conv3};
|
||||
|
||||
//extras Inverted Residual 2
|
||||
tk::dnn::Conv2d e_2_conv1(&net, 128, 1, 1, 1, 1, 0, 0, extras2[0], true);
|
||||
tk::dnn::Activation e_relu_2_1(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d e_2_conv2(&net, 128, 3, 3, 2, 2, 1, 1, extras2[1], true, false, false, 128);
|
||||
tk::dnn::Activation e_relu_2_2(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d e_2_conv3(&net, 256, 1, 1, 1, 1, 0, 0, extras2[2], true);
|
||||
tk::dnn::Layer * header_4[1] = {&e_2_conv3};
|
||||
|
||||
//extras Inverted Residual 3
|
||||
tk::dnn::Conv2d e_3_conv1(&net, 64, 1, 1, 1, 1, 0, 0, extras3[0], true);
|
||||
tk::dnn::Activation e_relu_3_1(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d e_3_conv2(&net, 64, 3, 3, 2, 2, 1, 1, extras3[1], true, false, false, 64);
|
||||
tk::dnn::Activation e_relu_3_2(&net, CUDNN_ACTIVATION_RELU);
|
||||
tk::dnn::Conv2d e_3_conv3(&net, 64, 1, 1, 1, 1, 0, 0, extras3[2], true);
|
||||
tk::dnn::Layer * header_5[1] = {&e_3_conv3};
|
||||
|
||||
// classification header 0
|
||||
tk::dnn::Layer * header_0[1] = {&relu_14_1};
|
||||
tk::dnn::Route rout_ch_0(&net, header_0, 1);
|
||||
tk::dnn::Conv2d ch_0_conv1(&net, 576, 3, 3, 1, 1, 1, 1, classification_header0[0], true, false, false, 576,true);
|
||||
tk::dnn::Activation ch_relu_0_1(&net, CUDNN_ACTIVATION_CLIPPED_RELU,6);
|
||||
tk::dnn::Conv2d ch_0_conv2(&net, 126, 1, 1, 1, 1, 0, 0, classification_header0[1], false);
|
||||
|
||||
// // classification header 1
|
||||
tk::dnn::Route rout_ch_1(&net, header_1, 1);
|
||||
tk::dnn::Conv2d ch_1_conv1(&net, 1280, 3, 3, 1, 1, 1, 1, classification_header1[0], true, false, false, 1280,true);
|
||||
tk::dnn::Activation ch_relu_1_1(&net, CUDNN_ACTIVATION_CLIPPED_RELU,6);
|
||||
tk::dnn::Conv2d ch_1_conv2(&net, 126, 1, 1, 1, 1, 0, 0, classification_header1[1], false);
|
||||
|
||||
// //classification header 2
|
||||
tk::dnn::Route rout_ch_2(&net, header_2, 1);
|
||||
tk::dnn::Conv2d ch_2_conv1(&net, 512, 3, 3, 1, 1, 1, 1, classification_header2[0], true, false, false, 512, true);
|
||||
tk::dnn::Activation ch_relu_2_1(&net, CUDNN_ACTIVATION_CLIPPED_RELU,6);
|
||||
tk::dnn::Conv2d ch_2_conv2(&net, 126, 1, 1, 1, 1, 0, 0, classification_header2[1], false);
|
||||
|
||||
// //classification header 3
|
||||
tk::dnn::Route rout_ch_3(&net, header_3, 1);
|
||||
tk::dnn::Conv2d ch_3_conv1(&net, 256, 3, 3, 1, 1, 1, 1, classification_header3[0], true, false, false, 256, true);
|
||||
tk::dnn::Activation ch_relu_3_1(&net, CUDNN_ACTIVATION_CLIPPED_RELU,6);
|
||||
tk::dnn::Conv2d ch_3_conv2(&net, 126, 1, 1, 1, 1, 0, 0, classification_header3[1], false);
|
||||
|
||||
// //classification header 4
|
||||
tk::dnn::Route rout_ch_4(&net, header_4, 1);
|
||||
tk::dnn::Conv2d ch_4_conv1(&net, 256, 3, 3, 1, 1, 1, 1, classification_header4[0], true, false, false, 256, true);
|
||||
tk::dnn::Activation ch_relu_4_1(&net, CUDNN_ACTIVATION_CLIPPED_RELU,6);
|
||||
tk::dnn::Conv2d ch_4_conv2(&net, 126, 1, 1, 1, 1, 0, 0, classification_header4[1], false);
|
||||
|
||||
// //classification header 5
|
||||
tk::dnn::Route rout_ch_5(&net, header_5, 1);
|
||||
tk::dnn::Conv2d ch_5_conv(&net, 126, 1, 1, 1, 1, 0, 0, classification_header5, false);
|
||||
|
||||
//regression header 0
|
||||
tk::dnn::Route rout_rh_0(&net, header_0, 1);
|
||||
tk::dnn::Conv2d rh_0_conv1(&net, 576, 3, 3, 1, 1, 1, 1, regression_header0[0], true, false, false, 576, true);
|
||||
tk::dnn::Activation rh_relu_0_1(&net, CUDNN_ACTIVATION_CLIPPED_RELU,6);
|
||||
tk::dnn::Conv2d rh_0_conv2(&net, 24, 1, 1, 1, 1, 0, 0, regression_header0[1], false);
|
||||
|
||||
// //regression header 1
|
||||
tk::dnn::Route rout_rh_1(&net, header_1, 1);
|
||||
tk::dnn::Conv2d rh_1_conv1(&net, 1280, 3, 3, 1, 1, 1, 1, regression_header1[0], true, false, false, 1280, true);
|
||||
tk::dnn::Activation rh_relu_1_1(&net, CUDNN_ACTIVATION_CLIPPED_RELU,6);
|
||||
tk::dnn::Conv2d rh_1_conv2(&net, 24, 1, 1, 1, 1, 0, 0, regression_header1[1], false);
|
||||
|
||||
//regression header 2
|
||||
tk::dnn::Route rout_rh_2(&net, header_2, 1);
|
||||
tk::dnn::Conv2d rh_2_conv1(&net, 512, 3, 3, 1, 1, 1, 1, regression_header2[0], true, false, false, 512, true);
|
||||
tk::dnn::Activation rh_relu_2_1(&net, CUDNN_ACTIVATION_CLIPPED_RELU,6);
|
||||
tk::dnn::Conv2d rh_2_conv2(&net, 24, 1, 1, 1, 1, 0, 0, regression_header2[1], false);
|
||||
|
||||
//regression header 3
|
||||
tk::dnn::Route rout_rh_3(&net, header_3, 1);
|
||||
tk::dnn::Conv2d rh_3_conv1(&net, 256, 3, 3, 1, 1, 1, 1, regression_header3[0], true, false, false, 256, true);
|
||||
tk::dnn::Activation rh_relu_3_1(&net, CUDNN_ACTIVATION_CLIPPED_RELU,6);
|
||||
tk::dnn::Conv2d rh_3_conv2(&net, 24, 1, 1, 1, 1, 0, 0, regression_header3[1], false);
|
||||
|
||||
//regression header 4
|
||||
|
||||
tk::dnn::Route rout_rh_4(&net, header_4, 1);
|
||||
tk::dnn::Conv2d rh_4_conv1(&net, 256, 3, 3, 1, 1, 1, 1, regression_header4[0], true, false, false, 256, true);
|
||||
tk::dnn::Activation rh_relu_4_1(&net, CUDNN_ACTIVATION_CLIPPED_RELU,6);
|
||||
tk::dnn::Conv2d rh_4_conv2(&net, 24, 1, 1, 1, 1, 0, 0, regression_header4[1], false);
|
||||
|
||||
//regression header 5
|
||||
tk::dnn::Route rout_rh_5(&net, header_5, 1);
|
||||
tk::dnn::Conv2d rh_5_conv(&net, 24, 1, 1, 1, 1, 0, 0, regression_header5, false);
|
||||
|
||||
//flatten confidence and flatten locations
|
||||
|
||||
// Load input
|
||||
dnnType *data;
|
||||
dnnType *input_h;
|
||||
readBinaryFile(input_bin, dim.tot(), &input_h, &data);
|
||||
//printDeviceVector(64, data, true);
|
||||
|
||||
//print network model
|
||||
net.print();
|
||||
|
||||
//convert network to tensorRT
|
||||
tk::dnn::NetworkRT netRT(&net, "mobilenetv2ssd.rt");
|
||||
|
||||
|
||||
tk::dnn::dataDim_t out_dim;
|
||||
out_dim = net.layers[net.num_layers-1]->output_dim;
|
||||
dnnType *cudnn_out, *rt_out;
|
||||
|
||||
tk::dnn::dataDim_t dim1 = dim; //input dim
|
||||
printCenteredTitle(" CUDNN inference ", '=', 30);
|
||||
{
|
||||
dim1.print();
|
||||
TIMER_START
|
||||
net.infer(dim1, data);
|
||||
TIMER_STOP
|
||||
dim1.print();
|
||||
}
|
||||
cudnn_out = net.layers[net.num_layers-1]->dstData;
|
||||
|
||||
printDeviceVector(64, cudnn_out, true);
|
||||
|
||||
tk::dnn::dataDim_t dim2 = dim;
|
||||
printCenteredTitle(" TENSORRT inference ", '=', 30);
|
||||
{
|
||||
dim2.print();
|
||||
TIMER_START
|
||||
netRT.infer(dim2, data);
|
||||
TIMER_STOP
|
||||
dim2.print();
|
||||
}
|
||||
rt_out = (dnnType *)netRT.buffersRT[1];
|
||||
|
||||
|
||||
printCenteredTitle(std::string(" RESNET CHECK RESULTS ").c_str(), '=', 30);
|
||||
dnnType *out, *out_h;
|
||||
int odim = out_dim.tot();
|
||||
readBinaryFile(output_bin, odim, &out_h, &out);
|
||||
std::cout << "CUDNN vs correct";
|
||||
checkResult(odim, cudnn_out, out);
|
||||
|
||||
std::cout << "TRT vs correct";
|
||||
checkResult(odim, rt_out, out);
|
||||
std::cout << "CUDNN vs TRT ";
|
||||
checkResult(odim, cudnn_out, rt_out);
|
||||
|
||||
return 0;
|
||||
}
|
||||
Reference in New Issue
Block a user