a problem with Interpreter

[luanle-padua]

Environment (please complete the following information):

• OS/OS Version: window
• Source Version: 2.10.0-p1
• Unity Version: 2021.3.14f1

Describe the bug

Hi, My name is Luan, and I am very new to Unity. I am trying to make predictions by a deep-learning model in Unity. I train a model in GG colab using TF. I follow this tutorial on Youtube:

https://www.youtube.com/watch?v=s42RTgKSXis&list=PLlPnLTVCBE5VX_-1yze2ustaJwkAGcZo1&index=36&t=547s&ab_channel=Alkr

The code for prediction is as below:

`using System.Collections; using System.Collections.Generic; using UnityEngine; using TensorFlowLite; using UnityEngine.UI;

public class predict : MonoBehaviour { [SerializeField] string fileName = "model.tflite"; [SerializeField] Text outputTextWiew = null; [SerializeField] Button button = null; [SerializeField] InputField inputField = null;

Interpreter interpreter;

// Start is called before the first frame update
void Start()
{
    var options = new InterpreterOptions()
    {
        threads = 2,
    };

    interpreter = new Interpreter(FileUtil.LoadFile(fileName), options);

    interpreter.AllocateTensors();

    button.onClick.AddListener(OnbuttonClick);

}

void OnDestroy()
{
    interpreter?.Dispose();
}

public void OnbuttonClick()
{
    // doInference(inputField.text);
    Debug.Log(inputField.text);
}

// public void doInference(string val)
// {
//     var input = new float[1]; ////// dua input vao day
//     input[0] = float.Parse(val);
//     var output = new float[1];

//     interpreter.SetInputTensorData(0, input);
//     interpreter.Invoke();
//     interpreter.GetOutputTensorData(0, output);
//     for (int i = 0; i < output.Length; i++)
//     {
//         Debug.Log(output[i]);
//         outputTextWiew.text = output[i].ToString();

//     }
// }

// Update is called once per frame
// void Update()
// {

// }

} `

when I play in Unity, it comes with this error. Could you please help me to solve this problem?

NullReferenceException: Object reference not set to an instance of an object TensorFlowLite.Interpreter..ctor (System.Byte[] modelData, TensorFlowLite.InterpreterOptions options) (at Library/PackageCache/com.github.asus4.tflite@2.10.0-p1/Runtime/Interpreter.cs:66) predict.Start () (at Assets/predict.cs:26)

[Ale1]

Hi Luan. I am not the developer, but I have seen this error show up before when I move the location of the model file and forget to update the path in the options. Check that your line FileUtil.LoadFile(filename) is pointing to the correct place.

[luanle-padua]

Thank you, I will try.

[Ale1]

@luanle-padua . is your issue fixed?

[luanle-padua]

hi Ale1,

Thank you so much. I could not fix it properly, then I changed my workflow to ONNX. And it is okay with me now.

Best regards Thanh Luan

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.