TensorFlowLite_ESP32 version.h

Projects Programming
1

I am using TensorFlowLite_ESP32 library and i am following intructions on the link of the authors. it says that i must include these:

#include "tensorflow/lite/micro/micro_mutable_op_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"

But version.h is nowhere to be found. The problem results like this:

C:\University\Do an co dien tu\testcode\testcode11\testcode11.ino:8:10: fatal error: tensorflow/lite/version.h: No such file or directory
    8 | #include "tensorflow/lite/version.h"
      |          ^~~~~~~~~~~~~~~~~~~~~~~~~~~
compilation terminated.
exit status 1

Compilation error: tensorflow/lite/version.h: No such file or directory

Can someone help me on this? Here's my full code

#include <Arduino.h>
#include <TensorFlowLite_ESP32.h>
#include "pid_ann_model.h"
#include "tensorflow/lite/micro/all_ops_resolver.h"
#include "tensorflow/lite/micro/micro_interpreter.h"
#include "tensorflow/lite/micro/micro_error_reporter.h"
#include "tensorflow/lite/schema/schema_generated.h"
#include "tensorflow/lite/version.h"
#include "PID_v1.h"

#define MOTOR_PIN 5 

const int tensor_arena_size = 2 * 1024;
uint8_t tensor_arena[tensor_arena_size];

double Setpoint, Input, Output;
double Kp, Ki, Kd;
PID myPID(&Input, &Output, &Setpoint, Kp, Ki, Kd, DIRECT);

// TensorFlow Lite
tflite::MicroErrorReporter micro_error_reporter;
tflite::AllOpsResolver resolver; 
const tflite::Model* model;
tflite::MicroInterpreter* interpreter;
TfLiteTensor* input_tensor;
TfLiteTensor* output_tensor;

void setup() {
  Serial.begin(115200);
  pinMode(MOTOR_PIN, OUTPUT);

  model = tflite::GetModel(pid_ann_model);
  if (model->version() != TFLITE_SCHEMA_VERSION) {
    Serial.println("Phiên bản model không khớp!");
    return;
  }

  static tflite::MicroInterpreter static_interpreter(model, resolver, tensor_arena, tensor_arena_size, &micro_error_reporter);
  interpreter = &static_interpreter;

  if (interpreter->AllocateTensors() != kTfLiteOk) {
    Serial.println("Lỗi cấp phát bộ nhớ!");
    return;
  }

  input_tensor = interpreter->input(0);
  output_tensor = interpreter->output(0);

  Setpoint = 50.0;
  myPID.SetMode(AUTOMATIC);

  Serial.println("TensorFlowLite_ESP32 đã khởi động thành công!");
}

void loop() {
  Input = analogRead(A0) * (100.0 / 1023.0);
    
  double error = Setpoint - Input;
  static double last_error = 0;
  double integral = error * 0.1;
  double derivative = (error - last_error) / 0.1;
  last_error = error;

  input_tensor->data.f[0] = error;
  input_tensor->data.f[1] = integral;
  input_tensor->data.f[2] = derivative;

  if (interpreter->Invoke() != kTfLiteOk) {
    Serial.println("Lỗi khi chạy mô hình!");
    return;
  }

  Kp = output_tensor->data.f[0];
  Ki = output_tensor->data.f[1];
  Kd = output_tensor->data.f[2];

  myPID.SetTunings(Kp, Ki, Kd);

  myPID.Compute();

  analogWrite(MOTOR_PIN, Output);

  Serial.print("Kp: "); 
  Serial.print(Kp);
  Serial.print(" Ki: "); 
  Serial.print(Ki);
  Serial.print(" Kd: ");
  Serial.print(Kd);
  Serial.print(" Output: "); 
  Serial.println(Output);

  delay(100);
}
2

Hi @anhkhoi041.

Please provide a the link to those instructions in a reply here on the forum topic. That might provide the forum helpers with some valuable context that will allow us to more effectively assist you.

3

Here's the link I'm following. Hope this help

Get started with microcontrollers | Google AI Edge | Google AI for Developers

4

Those instructions are for using the "LiteRT for Microcontrollers" C++ library, not for the "TensorFlowLite_ESP32" Arduino library. Although the two codebases are related, they are not the same. So the "LiteRT for Microcontrollers" C++ library documentation won't necessarily be perfectly applicable to the "TensorFlowLite_ESP32" Arduino library

You should use the examples provided by the "TensorFlowLite_ESP32" library developer as a reference to learn how to use that library. You will find them listed under the File > Examples > TensorFlowLite_ESP32 menu in Arduino IDE.

6

This topic was automatically closed 180 days after the last reply. New replies are no longer allowed.