tinyMLjs

This website makes Machine Learning models from microcontroller sensors using Javascript then uploaded to the microcontroller using Gitpod, RocksettaTinyML and The Arduino IDE.

Presently for complex Camera or Sound data use EdgeImpulse.com as the needed compression is at a much better level.

Upload from a raw CSV file or an Arduino style microcontroller using webSerial (Android Pixel phones also work) or your cell phone motion sensor. Keep the raw sensor data then Machine Learning train a tensorflowJS model for export or for live classification all on this single vanilla Javascript webpage!

Show: ------------ Hide: ----------- Load .csv files:
Following is the list of actual labels used in the same order as uploaded (comma-separated) Note: expecting files to be named: "name-lable.csv" or "name-lable (1).csv" etc.

CSV Lables (careful):
Senses Labels (careful):

Machine Learning models often need very specific data.
Count CSV: Count Senses: Count Total:
Number of Samples/count: Number of Senses/sample:

Enter number of epochs: , Learning rate:

...
Convert Model to Arduino: Github or direct load: Gitpod
Once you have made a mode.h file. then install this deprecated but I am working on getting a replacement Arduino Library https://github.com/hpssjellis/RocksettaTinyML based on EloquentArduino to load the code onto your Arduino ...

Select model file (.json):
Select weights file (.bin):
Select labels file (.txt):
Enter Labels (comma seperated):

...

Label:
CSV FileName:

...

Aruino NiclaVision webSerial code that can be adapted for other microcontrollers

The "fancy" Arduino sketch for the Nano33BleSense is on the github here https://github.com/hpssjellis/tinyMLjs/blob/main/public/acceleration/a00-accell-nano33-fancy.txt

/*
 * webSerial for testing javascript connection with an arduino
 * 
 * Latest work at   https://github.com/hpssjellis/webMLserial
 * 
 * Note: On the Arduino Serial monitor make sure no line-ending or if statements will not work
 * 
 * Android https://hpssjellis.github.io/web-serial-polyfill/index.html
 * Laptops Desktops  https://hpssjellis.github.io/my-examples-of-arduino-webUSB-webSerial/public/index.html
 * IOS     not really sure
 * 
 */

#include <Arduino.h> // Only needed for https://platformio.org/
//#include <Arduino_LSM9DS1.h> 
#include <Arduino_LSM6DSOX.h>
#include "VL53L1X.h"

// GLOBAL VAERIABLES
VL53L1X proximity;

#define FREQUENCY_HZ        25     // how many samples per second 
#define COLLECTION_SECONDS  1     // how many seconds to collect samples
#define INTERVAL_MS  (1000 / (FREQUENCY_HZ + 1)) // need for the timer 
#define CONVERT_G_TO_MS2    9.80665f   // accleration conversion

int myMaxData = FREQUENCY_HZ * COLLECTION_SECONDS;
int myCount = 0;
int myDelay = INTERVAL_MS;   // non-block delay in milliseconds
unsigned long myStart;

String readString;
bool mySendData = true;
bool mySendProximity = false;
int myProximityReading = 0;

void setup() {
  Serial.begin(115200);
  pinMode(LED_BUILTIN, OUTPUT);
  digitalWrite(LED_BUILTIN, HIGH);   // onboard LED, HIGH = off
  //while (!Serial) {}                 // do nothing and wait
  
  myStart = millis();   // set delay
  randomSeed(analogRead(A0));  // AO pin on XIAO does not have to be connected to anything
  Serial.println("accX,accY,accZ");  // CSV file heading titles
  
  if (!IMU.begin()) {
     Serial.println("Failed to initialize IMU!");
     while (1);
  }

// Initilize the Proximity sensor and com using wire1
  Wire1.begin();
  Wire1.setClock(400000); // use 400 kHz I2C
  proximity.setBus(&Wire1);

if (!proximity.init()) {
    Serial.println("Failed to detect and initialize sensor!");
    while (1);
  }

proximity.setDistanceMode(VL53L1X::Long);
  proximity.setMeasurementTimingBudget(10000);
  proximity.startContinuous(10);

}

void loop() {
  float x, y, z;
  myProximityReading = proximity.read();
  if ( (millis() - myStart ) >= myDelay) {       
     myStart = millis();      //  reset the delay time
     myCount += 1;
  if (IMU.accelerationAvailable()) {
    IMU.readAcceleration(x, y, z);
    x *= 9.81;
    y *= 9.81;
    z *= 9.81;

}

if (myCount >= myMaxData){
        mySendData = false;   // stop sending data when amount reached
     }

// This should be written better perhaps a case statement etc
     if (mySendData){
        if (mySendProximity){
          Serial.println(String(myProximityReading));  
        } else {
          Serial.println( String(x) + "," + String(y) + "," +  String(z)  );
        }
     } 
   }

while (Serial.available()) {
    delay(3);  
    char myChar = Serial.read();
    readString += myChar; 
  }

if (readString.length() > 0) {
    readString.trim();  // get rid of last weird character
    if (readString == "a"){
      digitalWrite(LED_BUILTIN, LOW); //onboard LED LOW = on
    }
    if (readString == "b"){
      digitalWrite(LED_BUILTIN, HIGH);
    }
    if (readString == "start"){
      mySendProximity = false;
      mySendData = true;
      myStart = millis();      //  reset the delay time
      myCount = 0;
    }
    if (readString == "stop"){
      mySendData = false;
      Serial.println("Stopping at count: "+ String(myCount));  // CSV file heading titles
    }
    if (readString == "firstline"){
      if (mySendProximity){
         Serial.println("proximity"); 
      } else {
        Serial.println("accX,accY,accZ");  // CSV file heading titles
      }
    }

if (readString == "proximity"){
      mySendProximity = true;
      mySendData = true;
      myStart = millis();      //  reset the delay time
      myCount = 0;

}
    readString="";
  } 
  
}

Use at your own risk!
By Jeremy Ellis @rocksetta
Github at https://github.com/hpssjellis/tinyMLjs/tree/main
Demo's Index at https://hpssjellis.github.io/tinyMLjs/public/index.html
This page should be at https://hpssjellis.github.io/tinyMLjs/public/acceleration/a00-best-acceleration.html

A couple of gotchas (as of Jun 22, 2023):

File names have to be in the format "name-lable.csv" or "name-lable (1).csv" or "name-lable (2).csv" etc. Unfortunately Android and iPhone don't auto make the numbering for you.
Android and Apple device have an opposite orientation, so I have made negative all the android motion data so when your phone is on a table z = -9.8 m/s^2 etc. When looking veritacally at your phone y = -9.8 m/s^2. The auto detect of this only works if an Android phone is in mobile format not "desktop site"
Real data has lots of rough data, machine learning models do not like missing data. If your results show "NaN" either your training data or classification data has errors. Note: If the loss is not changing your trained data probably has errors. The "clean,trim,fill" buttons might help.
Presently a CSV label upload bug happens sometimes. Easy to fix by entering the correct labels in the correct order. I will try to fix the issue when I figure out what is causing it.