Add WioTerminal_TF_MNIST example

Author: Pillar1989

.travis.yml

@@ -24,8 +24,8 @@ before_install:
       cd "${HOME}/Arduino/libraries/${repositoryName}"
       cd "${TRAVIS_BUILD_DIR}"
     }
-  - buildExampleSketch() { arduino-cli compile  --warnings all --fqbn $BOARD $PWD/examples/$1 --verbose; }
-  - buildExampleUtilitySketch() { arduino-cli compile --warnings all --fqbn $BOARD $PWD/examples/utility/$1 --verbose; }
+  - buildExampleSketch() { arduino-cli compile  --warnings all --fqbn $BOARD $PWD/examples/$1 ; }
+  - buildExampleUtilitySketch() { arduino-cli compile --warnings all --fqbn $BOARD $PWD/examples/utility/$1 ; }
 
 install:
   - mkdir -p $HOME/Arduino/libraries
@@ -139,6 +139,7 @@ script:
     - export BOARD=Seeeduino:samd:seeed_XIAO_m0
     - buildExampleSketch SeeeduinoXIAO_SPISlave;
 
+
 notifications:
   webhooks:
     urls:

examples/WioTerminal_TF_MNIST/TF_MNIST.ipynb

@@ -0,0 +1,157 @@
+#include <TensorFlowLite.h>
+#include "tensorflow/lite/micro/all_ops_resolver.h"
+#include "tensorflow/lite/micro/micro_error_reporter.h"
+#include "tensorflow/lite/micro/micro_interpreter.h"
+#include "tensorflow/lite/schema/schema_generated.h"
+#include "tensorflow/lite/version.h"
+#include "model.h"
+#include "mnist.h"
+#include"TFT_eSPI.h"
+
+TFT_eSPI tft;
+TFT_eSprite spr = TFT_eSprite(&tft);  // Sprite
+
+namespace {
+// Globals
+const tflite::Model* model = nullptr;
+tflite::MicroInterpreter* interpreter = nullptr;
+tflite::ErrorReporter* reporter = nullptr;
+TfLiteTensor* input = nullptr;
+TfLiteTensor* output = nullptr;
+constexpr int kTensorArenaSize = 8000; // Just pick a big enough number
+uint8_t tensor_arena[ kTensorArenaSize ] = { 0 };
+float *input_buffer=nullptr;
+}  // namespace
+
+
+float display_buffer[28224];
+
+void bitmap_to_float_array( float* dest, const unsigned char* bitmap ) { // Populate input_vec with the monochrome 1bpp bitmap
+  int pixel = 0;
+  for( int y = 0; y < 28; y++ ) {
+    for( int x = 0; x < 28; x++ ) {
+      int B = x / 8; // the Byte # of the row
+      int b = x % 8; // the Bit # of the Byte
+      dest[ pixel ] = ( bitmap[ y * 4 + B ] >> ( 7 - b ) ) & 
+                        0x1 ? 1.0f : 0.0f;
+      pixel++;
+    }
+  }
+}
+
+void draw_input_buffer() {
+
+  int pre_i, pre_j, after_i, after_j;//缩放前后对应的像素点坐标
+
+  for (int i = 0; i<168; i++)
+  {
+    for (int j = 0; j<168; j++)
+    {
+        after_i = i;
+        after_j = j;
+        pre_i = (int)(after_i / 6);/////取整，插值方法为：最邻近插值（近邻取样法）
+        pre_j = (int)(after_j / 6);
+        if (pre_i >= 0 && pre_i < 28 && pre_j >= 0 && pre_j < 28)//在原图范围内
+          *(display_buffer + i * 168 + j) = *(input_buffer + pre_i * 28 + pre_j);
+    }
+  }
+
+    for(int cy = 0; cy < 168; cy++)
+    {
+        for(int cx = 0; cx < 168; cx++)
+        {
+
+           tft.drawPixel( cx + 76, cy, display_buffer[ cy * 168  + cx  ] > 0 ? 0xFFFFFFFF : 0xFF000000 );
+
+        }
+    }
+
+    
+}
+
+void print_input_buffer() {
+  char output[ 28 * 29 ]; // Each row should end row newline
+  for( int y = 0; y < 28; y++ ) {
+    for( int x = 0; x < 28; x++ ) {
+      output[ y * 29 + x ] = input_buffer[ y * 28 + x ] > 0 ? ' ' : '#';
+    }
+    output[ y * 29 + 28 ] = '\n';
+  }
+  reporter->Report( output );
+}
+
+
+
+void setup() {
+  // Load Model
+  //Serial.begin(115200);
+  static tflite::MicroErrorReporter error_reporter;
+  reporter = &error_reporter;
+  reporter->Report( "Let's use AI to recognize some numbers!" );
+
+  model = tflite::GetModel( tf_model );
+  if( model->version() != TFLITE_SCHEMA_VERSION ) {
+    reporter->Report( "Model is schema version: %d\nSupported schema version is: %d", model->version(), TFLITE_SCHEMA_VERSION );
+    return;
+  }
+  // Setup our TF runner
+  static tflite::AllOpsResolver resolver;
+  static tflite::MicroInterpreter static_interpreter(
+      model, resolver, tensor_arena, kTensorArenaSize, reporter );
+  interpreter = &static_interpreter;
+  
+  // Allocate memory from the tensor_arena for the model's tensors.
+  TfLiteStatus allocate_status = interpreter->AllocateTensors();
+  if( allocate_status != kTfLiteOk ) {
+    reporter->Report( "AllocateTensors() failed" );
+    return;
+  }
+
+  // Obtain pointers to the model's input and output tensors.
+  input = interpreter->input(0);
+  output = interpreter->output(0);
+
+  // Save the input buffer to put our MNIST images into
+  input_buffer = input->data.f;
+    tft.begin();
+    tft.setRotation(3);
+    tft.fillScreen(TFT_BLACK); // fills entire the screen with colour red
+  tft.setTextColor(TFT_RED);   
+  tft.setTextSize(2);
+  tft.drawString("It looks like the number:",0,200);//prints string at (70,80)
+  TFT_eSprite spr = TFT_eSprite(&tft);  // Sprite
+
+
+}
+
+void loop() {
+  // Pick a random test image for input
+  const int num_test_images = ( sizeof( test_images ) / sizeof( *test_images ) );
+
+
+  bitmap_to_float_array( input_buffer, test_images[ rand() % num_test_images ] );
+
+  draw_input_buffer();
+  //print_input_buffer();
+  // Run our model
+  TfLiteStatus invoke_status = interpreter->Invoke();
+  if( invoke_status != kTfLiteOk ) {
+    reporter->Report( "Invoke failed" );
+    return;
+  }
+  
+  float* result = output->data.f;
+
+  reporter->Report( "It looks like the number: %d", std::distance( result, std::max_element( result, result + 10 ) ) );
+//   tft.drawNumber(std::distance( result, std::max_element( result, result + 10 ) ), 300, 200),
+  spr.createSprite(20, 20);
+//   spr.fillSprite(TFT_RED);
+  spr.setTextColor(TFT_RED);
+  spr.setTextSize(2);
+  spr.drawNumber(std::distance( result, std::max_element( result, result + 10 ) ),0, 0); 
+  spr.pushSprite(300, 200);
+  spr.deleteSprite();
+  // Wait 1-sec til before running again
+  delay( 200 );
+
+}