Showing numbers based on movement patterns with TinyML and m

#include "M5Atom.h"
#include <faces_m5stack_inferencing.h>
#include "numbers.h"

#define CONVERT_G_TO_MS2    9.80665f
#define FREQUENCY_HZ        100
#define INTERVAL_MS         (1000 / (FREQUENCY_HZ + 1))

static unsigned long last_interval_ms = 0;

float gyroX = 0.0F;
float gyroY = 0.0F;
float gyroZ = 0.0F;

float accX = 0.0F;
float accY = 0.0F;
float accZ = 0.0F;

float features[EI_CLASSIFIER_DSP_INPUT_FRAME_SIZE];

//std::map<String, float> inferencia;
struct Inferencia {
  String label;
  float accuracy;
};

Inferencia resp;
// keep track of where we are in the feature array
size_t feature_ix = 0;
float maximo=0;

void setup(){
  M5.begin(true, false, true);
  M5.IMU.Init();

  //Serial.printf("accX accY accZ gyroX gyroY gyroZ\n");
}

void loop(){
  //M5.update(); //Add M5.update() to read the button state, details please see System
  //if (M5.Btn.isPressed()) { //If the button is pressed
  if (millis() > last_interval_ms + INTERVAL_MS) {
    last_interval_ms = millis();

    M5.IMU.getAccelData(&accX,&accY,&accZ);
    M5.IMU.getGyroData(&gyroX,&gyroY,&gyroZ);

    // fill the features buffer
    features[feature_ix++] = accX;
    features[feature_ix++] = accY;
    features[feature_ix++] = accZ;
    features[feature_ix++] = gyroX;
    features[feature_ix++] = gyroY;
    features[feature_ix++] = gyroZ;    

    // features buffer full? then classify!
        if (feature_ix == EI_CLASSIFIER_DSP_INPUT_FRAME_SIZE) {
            ei_impulse_result_t result;

            // create signal from features frame
            signal_t signal;
            numpy::signal_from_buffer(features, EI_CLASSIFIER_DSP_INPUT_FRAME_SIZE, &signal);

            // run classifier
            EI_IMPULSE_ERROR res = run_classifier(&signal, &result, false);
            ei_printf("run_classifier returned: %d\n", res);
            if (res != 0) return;

            // print predictions
            ei_printf("Predictions (DSP: %d ms., Classification: %d ms., Anomaly: %d ms.): \n",
                result.timing.dsp, result.timing.classification, result.timing.anomaly);

            maximo=0;
            // print the predictions
            for (size_t ix = 0; ix < EI_CLASSIFIER_LABEL_COUNT; ix++) {
                ei_printf("%s:\t%.5f\n", result.classification[ix].label, result.classification[ix].value); 
                
                if(result.classification[ix].value>maximo){
                    maximo = result.classification[ix].value;
                    resp.label = result.classification[ix].label;
                    resp.accuracy = result.classification[ix].value;
                    //inferencia[result.classification[ix].label] =  result.classification[ix].value;
                }
                      
            }

            if(resp.accuracy>=0.8){
              ei_printf("Prediccion Final: %s:\t%.5f\n", resp.label, resp.accuracy); 
              if(resp.label=="1"){
                show_face(NUMBER_ONE);
              }else if(resp.label=="2"){
                show_face(NUMBER_TWO);
              }else if(resp.label=="idle"){
                show_face(FACE_SAD);
              }
            }else{
              show_face(FACE_SAD);
            }
        #if EI_CLASSIFIER_HAS_ANOMALY == 1
            ei_printf("anomaly:\t%.3f\n", result.anomaly);
        #endif

            // reset features frame
            feature_ix = 0;
        }
    //Serial.printf("%5.2f  %5.2f  %5.2f %6.2f %6.2f %6.2f\n", accX, accY, accZ, gyroX, gyroY, gyroZ);
    //delay(10);
  }
  //}
}

void ei_printf(const char *format, ...) {
    static char print_buf[1024] = { 0 };

    va_list args;
    va_start(args, format);
    int r = vsnprintf(print_buf, sizeof(print_buf), format, args);
    va_end(args);

    if (r > 0) {
        Serial.write(print_buf);
    }
}