xavier
38a1b9dcb2
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>
63 lines
1.3 KiB
C++
63 lines
1.3 KiB
C++
#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;
|
|
};
|