fix sine example

Author: agrimagsrl

README.md

@@ -28,11 +28,12 @@ git clone https://github.com/eloquentarduino/EloquentTinyML.git
 Eloquent::TinyML::TinyML<
     NUMBER_OF_INPUTS,
     NUMBER_OF_OUTPUTS,
-    TENSOR_ARENA_SIZE> ml(sine_model_quantized_tflite);
+    TENSOR_ARENA_SIZE> ml;
 
 
 void setup() {
     Serial.begin(115200);
+    ml.begin(sine_model_quantized_tflite);
 }
 
 void loop() {

examples/ColorClassificationTrainingExample/ColorClassificationTrainingExample.ino

@@ -7,11 +7,12 @@
 // in future projects you may need to tweek this value: it's a trial and error process
 #define TENSOR_ARENA_SIZE 2*1024
 
-Eloquent::TinyML::TfLite<NUMBER_OF_INPUTS, NUMBER_OF_OUTPUTS, TENSOR_ARENA_SIZE> ml(sine_model);
+Eloquent::TinyML::TfLite<NUMBER_OF_INPUTS, NUMBER_OF_OUTPUTS, TENSOR_ARENA_SIZE> ml;
 
 
 void setup() {
     Serial.begin(115200);
+    ml.begin(sine_model);
 }
 
 void loop() {

examples/ColorClassificationTrainingExample/RGB.h

@@ -1,6 +1,6 @@
 #pragma once
 
-unsigned char sine_model[] = {
+const unsigned char sine_model[] = {
         0x18, 0x00, 0x00, 0x00, 0x54, 0x46, 0x4c, 0x33, 0x00, 0x00, 0x0e, 0x00,
         0x18, 0x00, 0x04, 0x00, 0x08, 0x00, 0x0c, 0x00, 0x10, 0x00, 0x14, 0x00,
         0x0e, 0x00, 0x00, 0x00, 0x03, 0x00, 0x00, 0x00, 0x10, 0x0a, 0x00, 0x00,

examples/SineExample/SineExample.ino

@@ -0,0 +1,88 @@
+#pragma once
+
+#include "EloquentTinyML.h"
+#include "persondetection/model.h"
+
+// Keeping these as constant expressions allow us to allocate fixed-sized arrays
+// on the stack for our working memory.
+
+// All of these values are derived from the values used during model training,
+// if you change your model you'll need to update these constants.
+const int kNumCols = 96;
+const int kNumRows = 96;
+const int kNumChannels = 1;
+const int kMaxImageSize = kNumCols * kNumRows * kNumChannels;
+const int kCategoryCount = 3;
+const int kPersonIndex = 1;
+const int kNotAPersonIndex = 2;
+const char* kCategoryLabels[kCategoryCount] = {
+        "unused",
+        "person",
+        "notperson",
+};
+
+
+namespace Eloquent {
+    namespace TinyML {
+
+        /**
+         *
+         * @tparam tensorArenaSize
+         */
+        template<size_t tensorArenaSize>
+        class PersonDetector {
+        public:
+            PersonDetector() {
+
+            }
+
+            /**
+             *
+             * @return
+             */
+            bool begin() {
+                return tf.begin(g_person_detect_model_data);
+            }
+
+            /**
+             *
+             * @return
+             */
+            bool initialized() {
+                return tf.initialized();
+            }
+
+            void detect(uint8_t *grayscaleImage) {
+                tf.predict(grayscaleImage, scores);
+            }
+
+            /**
+             * 
+             * @return 
+             */
+            uint8_t getPersonScore() {
+                return scores[1];
+            }
+
+            /**
+             * 
+             * @return 
+             */
+            uint8_t getNonPersonScore() {
+                return scores[2];
+            }
+            
+            /**
+             * 
+             * @return 
+             */
+            bool isPerson() {
+                return getPersonScore() > getNonPersonScore();
+            }
+
+        protected:
+            uint8_t scores[3];
+            TfLite<kNumRows * kNumCols, 3, tensorArenaSize> tf;
+        };
+    }
+}