added FP16/FP32 readout in camera samples

detectNet.cpp

@@ -81,7 +81,7 @@ detectNet* detectNet::Create( const char* prototxt, const char* model, const cha
 
 	for( uint32_t n=0; n < numClasses; n++ )
 	{
-		if( n != 0 )
+		if( n != 1 )
 		{
 			net->mClassColors[0][n*4+0] = 0.0f;	// r
 			net->mClassColors[0][n*4+1] = 200.0f;	// g

detectNet.h

@@ -95,7 +95,7 @@ class detectNet : public tensorNet
 	/**
 	 * Set the visualization color of a particular class of object.
 	 */
-	void SetClassColor( uint32_t classIndex, float r, float g, float b, float a=1.0f );
+	void SetClassColor( uint32_t classIndex, float r, float g, float b, float a=255.0f );
 
 
 protected:

detectnet-camera/detectnet-camera.cpp

@@ -201,7 +201,7 @@ int main( int argc, char** argv )
 			if( display != NULL )
 			{
 				char str[256];
-				sprintf(str, "GIE build %x | %04.1f FPS", NV_GIE_VERSION, display->GetFPS());
+				sprintf(str, "GIE build %x | %s | %04.1f FPS", NV_GIE_VERSION, net->HasFP16() ? "FP16" : "FP32", display->GetFPS());
 				//sprintf(str, "GIE build %x | %s | %04.1f FPS | %05.2f%% %s", NV_GIE_VERSION, net->GetNetworkName(), display->GetFPS(), confidence * 100.0f, net->GetClassDesc(img_class));
 				display->SetTitle(str);	
 			}	

imagenet-camera/imagenet-camera.cpp

@@ -148,15 +148,15 @@ int main( int argc, char** argv )
 			{
 				char str[256];
 				sprintf(str, "%05.2f%% %s", confidence * 100.0f, net->GetClassDesc(img_class));
-				
+
 				font->RenderOverlay((float4*)imgRGBA, (float4*)imgRGBA, camera->GetWidth(), camera->GetHeight(),
 								    str, 0, 0, make_float4(255.0f, 255.0f, 255.0f, 255.0f));
 			}
 
 			if( display != NULL )
 			{
 				char str[256];
-				sprintf(str, "GIE build %x | %s | %04.1f FPS", NV_GIE_VERSION, net->GetNetworkName(), display->GetFPS());
+				sprintf(str, "GIE build %x | %s | %s | %04.1f FPS", NV_GIE_VERSION, net->GetNetworkName(), net->HasFP16() ? "FP16" : "FP32", display->GetFPS());
 				//sprintf(str, "GIE build %x | %s | %04.1f FPS | %05.2f%% %s", NV_GIE_VERSION, net->GetNetworkName(), display->GetFPS(), confidence * 100.0f, net->GetClassDesc(img_class));
 				display->SetTitle(str);	
 			}	

tensorNet.cpp

@@ -20,12 +20,13 @@ tensorNet::tensorNet()
 	mInfer   = NULL;
 	mContext = NULL;
 
-	mWidth     = 0;
-	mHeight    = 0;
-	mInputSize = 0;
-	mInputCPU  = NULL;
-	mInputCUDA = NULL;
-
+	mWidth      = 0;
+	mHeight     = 0;
+	mInputSize  = 0;
+	mInputCPU   = NULL;
+	mInputCUDA  = NULL;
+	mEnableFP16 = false;
+
 	memset(&mInputDims, 0, sizeof(nvinfer1::Dims3));
 }
 
@@ -59,16 +60,16 @@ bool tensorNet::ProfileModel(const std::string& deployFile,			   // name for caf
 	nvinfer1::INetworkDefinition* network = builder->createNetwork();
 
 	builder->setMinFindIterations(3);	// allow time for TX1 GPU to spin up
-    builder->setAverageFindIterations(2);
+     builder->setAverageFindIterations(2);
 
 	// parse the caffe model to populate the network, then set the outputs
 	nvcaffeparser1::ICaffeParser* parser = nvcaffeparser1::createCaffeParser();
 
-	const bool useFp16 = builder->platformHasFastFp16();	// getHalf2Mode();
-	printf(LOG_GIE "platform %s FP16 support.\n", useFp16 ? "has" : "does not have");
+	mEnableFP16 = builder->platformHasFastFp16();
+	printf(LOG_GIE "platform %s FP16 support.\n", mEnableFP16 ? "has" : "does not have");
 	printf(LOG_GIE "loading %s %s\n", deployFile.c_str(), modelFile.c_str());
 
-	nvinfer1::DataType modelDataType = useFp16 ? nvinfer1::DataType::kHALF : nvinfer1::DataType::kFLOAT; // create a 16-bit model if it's natively supported
+	nvinfer1::DataType modelDataType = mEnableFP16 ? nvinfer1::DataType::kHALF : nvinfer1::DataType::kFLOAT; // create a 16-bit model if it's natively supported
 	const nvcaffeparser1::IBlobNameToTensor *blobNameToTensor =
 		parser->parse(deployFile.c_str(),		// caffe deploy file
 					  modelFile.c_str(),		// caffe model file
@@ -95,7 +96,7 @@ bool tensorNet::ProfileModel(const std::string& deployFile,			   // name for caf
 	builder->setMaxWorkspaceSize(16 << 20);
 
 	// set up the network for paired-fp16 format
-	if(useFp16)
+	if(mEnableFP16)
 		builder->setHalf2Mode(true);
 
 	printf(LOG_GIE "building CUDA engine\n");
@@ -137,7 +138,7 @@ bool tensorNet::LoadNetwork( const char* prototxt_path, const char* model_path,
 		return false;
 
 	/*
-	 * load and parse network definition and model file
+	 * attempt to load network from cache before profiling with tensorRT
 	 */
 	std::stringstream gieModelStream;
 	gieModelStream.seekg(0, gieModelStream.beg);
@@ -148,16 +149,16 @@ bool tensorNet::LoadNetwork( const char* prototxt_path, const char* model_path,
 
 	std::ifstream cache( cache_path );
 
-    if( !cache )
-    {
+	if( !cache )
+	{
 		printf(LOG_GIE "cache file not found, profiling network model\n");
-		
+
 		if( !ProfileModel(prototxt_path, model_path, output_blobs, MAX_BATCH_SIZE, gieModelStream) )
 		{
 			printf("failed to load %s\n", model_path);
 			return 0;
 		}
-		
+
 		printf(LOG_GIE "network profiling complete, writing cache to %s\n", cache_path);
 		std::ofstream outFile;
 		outFile.open(cache_path);
@@ -169,9 +170,19 @@ bool tensorNet::LoadNetwork( const char* prototxt_path, const char* model_path,
 	else
 	{
 		printf(LOG_GIE "loading network profile from cache... %s\n", cache_path);
-        gieModelStream << cache.rdbuf();
-        cache.close();
-    }
+		gieModelStream << cache.rdbuf();
+		cache.close();
+
+		// test for half FP16 support
+		nvinfer1::IBuilder* builder = createInferBuilder(gLogger);
+
+		if( builder != NULL )
+		{
+			mEnableFP16 = builder->platformHasFastFp16();
+			printf(LOG_GIE "platform %s FP16 support.\n", mEnableFP16 ? "has" : "does not have");
+			builder->destroy();	
+		}
+	}
 
 	printf(LOG_GIE "%s loaded\n", model_path);
 

tensorNet.h

@@ -24,21 +24,14 @@ class tensorNet
 	 */
 	virtual ~tensorNet();
 
-protected:
-
-	/**
-	 * Constructor.
-	 */
-	tensorNet();
-
 	/**
 	 * Load a new network instance
 	 * @param prototxt File path to the deployable network prototxt
 	 * @param model File path to the caffemodel 
 	 * @param mean File path to the mean value binary proto (NULL if none)
 	 */
 	bool LoadNetwork( const char* prototxt, const char* model, const char* mean=NULL,
-					  const char* input_blob="data", const char* output_blob="prob");
+				   const char* input_blob="data", const char* output_blob="prob");
 
 	/**
 	 * Load a new network instance with multiple output layers
@@ -47,8 +40,20 @@ class tensorNet
 	 * @param mean File path to the mean value binary proto (NULL if none)
 	 */
 	bool LoadNetwork( const char* prototxt, const char* model, const char* mean,
-					  const char* input_blob, const std::vector<std::string>& output_blobs);
-
+				   const char* input_blob, const std::vector<std::string>& output_blobs);
+
+	/**
+ 	 * Query for half-precision FP16 support.
+	 */
+	inline bool HasFP16() const		{ return mEnableFP16; }
+
+protected:
+
+	/**
+	 * Constructor.
+	 */
+	tensorNet();
+
 	/**
 	 * Create and output an optimized network model
 	 * @note this function is automatically used by LoadNetwork, but also can 
@@ -60,9 +65,9 @@ class tensorNet
 	 * @param modelStream output model stream
 	 */
 	bool ProfileModel( const std::string& deployFile, const std::string& modelFile,
-					   const std::vector<std::string>& outputs,
-					   uint32_t maxBatchSize, std::ostream& modelStream);
-							
+				    const std::vector<std::string>& outputs,
+				    uint32_t maxBatchSize, std::ostream& modelStream);
+
 	/**
 	 * Prefix used for tagging printed log output
 	 */
@@ -96,6 +101,7 @@ class tensorNet
 	uint32_t mInputSize;
 	float*   mInputCPU;
 	float*   mInputCUDA;
+	bool	    mEnableFP16;
 
 	nvinfer1::Dims3 mInputDims;