examples: measure performance of inference

Signed-off-by: Inki Dae <inki.dae@samsung.com>

2 files changed, 34 insertions, 12 deletions

diff --git a/examples/cpp_classification/classification_profiling.cpp b/examples/cpp_classification/classification_profiling.cpp
index f5d5eaed..50afe181 100644
--- a/examples/cpp_classification/classification_profiling.cpp
+++ b/examples/cpp_classification/classification_profiling.cpp
@@ -530,13 +530,24 @@ int main(int argc, char** argv) {
 
   cv::Mat img = cv::imread(file, -1);
   CHECK(!img.empty()) << "Unable to decode image " << file;
-  std::vector<Prediction> predictions = classifier.Classify(img);
 
-  /* Print the top N predictions. */
-  for (size_t i = 0; i < predictions.size(); ++i) {
-    Prediction p = predictions[i];
-    std::cout << std::fixed << std::setprecision(4) << p.second << " - \""
-              << p.first << "\"" << std::endl;
+  struct timeval t1, t2;
+
+  for (int i = 0; i < 5; i++) {
+    gettimeofday(&t1, NULL);
+    std::vector<Prediction> predictions = classifier.Classify(img);
+    gettimeofday(&t2, NULL);
+
+    /* Print the top N predictions. */
+    for (size_t i = 0; i < predictions.size(); ++i) {
+      Prediction p = predictions[i];
+      std::cout << std::fixed << std::setprecision(4) << p.second << " - \""
+                << p.first << "\"" << std::endl;
+    }
+
+    double elapsedTime = (t2.tv_sec - t1.tv_sec) * 1000.0;
+    elapsedTime += (t2.tv_usec - t1.tv_usec) / 1000.0;   // us to ms
+    std::cout << elapsedTime << " ms.\n";
   }
 }
 #else
diff --git a/examples/cpp_classification/classification_profiling_gpu.cpp b/examples/cpp_classification/classification_profiling_gpu.cpp
index 3c5e04ad..2854415d 100644
--- a/examples/cpp_classification/classification_profiling_gpu.cpp
+++ b/examples/cpp_classification/classification_profiling_gpu.cpp
@@ -530,13 +530,24 @@ int main(int argc, char** argv) {
 
   cv::Mat img = cv::imread(file, -1);
   CHECK(!img.empty()) << "Unable to decode image " << file;
-  std::vector<Prediction> predictions = classifier.Classify(img);
 
-  /* Print the top N predictions. */
-  for (size_t i = 0; i < predictions.size(); ++i) {
-    Prediction p = predictions[i];
-    std::cout << std::fixed << std::setprecision(4) << p.second << " - \""
-              << p.first << "\"" << std::endl;
+  struct timeval t1, t2;
+
+  for (int i = 0; i < 5; i++) {
+    gettimeofday(&t1, NULL);
+    std::vector<Prediction> predictions = classifier.Classify(img);
+    gettimeofday(&t2, NULL);
+
+    /* Print the top N predictions. */
+    for (size_t i = 0; i < predictions.size(); ++i) {
+      Prediction p = predictions[i];
+      std::cout << std::fixed << std::setprecision(4) << p.second << " - \""
+                << p.first << "\"" << std::endl;
+    }
+
+    double elapsedTime = (t2.tv_sec - t1.tv_sec) * 1000.0;
+    elapsedTime += (t2.tv_usec - t1.tv_usec) / 1000.0;   // us to ms
+    std::cout << elapsedTime << " ms.\n";
   }
 }
 #else