java - Tensorflow Lite Android 应用程序崩溃并出现 NullPointerException 'void org.tensorflow.lite.Interpreter.run(java.lang.Object, java.lang.Object)'

标签 java python android tensorflow tensorflow-lite

我正在尝试将一组值传递给我的 Tensorflow Lite 模型。在我的应用程序中输入值后应用程序崩溃，然后将这些值存储在数组中并传递给模型。这是代码 fragment 。

inferButton.setOnClickListener(new View.OnClickListener(){
            @Override
            public void onClick(View view){

                int getText1=Integer.parseInt(inputNumber1.getText().toString());
                int getText2=Integer.parseInt(inputNumber2.getText().toString());
                int getText3=Integer.parseInt(inputNumber3.getText().toString());
                int getText4=Integer.parseInt(inputNumber4.getText().toString());
                int getText5=Integer.parseInt(inputNumber5.getText().toString());
                int getText6=Integer.parseInt(inputNumber6.getText().toString());
                int getText7=Integer.parseInt(inputNumber7.getText().toString());
                int getText8=Integer.parseInt(inputNumber8.getText().toString());
                int getText9=Integer.parseInt(inputNumber9.getText().toString());
                int getText10=Integer.parseInt(inputNumber10.getText().toString());
                int getText11=Integer.parseInt(inputNumber11.getText().toString());
                int getText12=Integer.parseInt(inputNumber12.getText().toString());
                int getText13=Integer.parseInt(inputNumber13.getText().toString());
                int getText14=Integer.parseInt(inputNumber14.getText().toString());

                int attributes[]={getText1,getText2,getText3,getText4,getText5,getText5,getText6,getText7,getText8,getText9,getText10,getText10,getText11,getText12,getText13,getText14};

                int prediction=doInference(attributes);
                //float prediction=doInference(inputNumber.getText().toString());
                outputNumber.setText(prediction);
            }
        });

public int doInference(int [] attributeArray){
        int[] inputVal=new int[14];
        for(int i=0;i<14;i++) {
            inputVal[i] = attributeArray[i];
        }
        //Output shape is [1][1]
        int[][] outputval=new int[1][1];

        //Run inference passing the input shape and getting the output shape
        tflite.run(inputVal, outputval);

        //Inferred value is at [0][0]
        int inferredValue=outputval[0][0];

        return inferredValue;
    }

我的错误是:-

E/AndroidRuntime: FATAL EXCEPTION: main
    Process: com.example.appdoctor, PID: 13022
    java.lang.NullPointerException: Attempt to invoke virtual method 'void org.tensorflow.lite.Interpreter.run(java.lang.Object, java.lang.Object)' on a null object reference
        at com.example.appdoctor.Prediction.doInference(Prediction.java:90)
        at com.example.appdoctor.Prediction$1.onClick(Prediction.java:75)
        at android.view.View.performClick(View.java:7125)
        at android.view.View.performClickInternal(View.java:7102)
        at android.view.View.access$3500(View.java:801)
        at android.view.View$PerformClick.run(View.java:27336)
        at android.os.Handler.handleCallback(Handler.java:883)
        at android.os.Handler.dispatchMessage(Handler.java:100)
        at android.os.Looper.loop(Looper.java:214)
        at android.app.ActivityThread.main(ActivityThread.java:7356)
        at java.lang.reflect.Method.invoke(Native Method)
        at com.android.internal.os.RuntimeInit$MethodAndArgsCaller.run(RuntimeInit.java:492)
        at com.android.internal.os.ZygoteInit.main(ZygoteInit.java:930)

我认为数组可能没有填充值，或者我在 doInference() 方法中犯了错误，因为我不了解我的神经网络结构。

这是 python 代码:-

import glob
import os
from keras.models import Sequential, load_model
import numpy as np
import pandas as pd
from keras.layers import Dense
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import LabelEncoder, MinMaxScaler
import matplotlib.pyplot as plt
import keras as k
import tensorflow as tf
from tensorflow import keras
from tensorflow import lite

df = pd.read_csv("kidney4.csv")
df = df.dropna(axis=0)

for column in df.columns:
        if df[column].dtype == np.number:
            continue
        df[column] = LabelEncoder().fit_transform(df[column])

X = df.drop(["classification"], axis=1)
y = df["classification"]

x_scaler = MinMaxScaler()
x_scaler.fit(X)
column_names = X.columns
X[column_names] = x_scaler.transform(X)

X_train,  X_test, y_train, y_test = train_test_split(X, y, test_size= 0.2, shuffle=True)

model=keras.Sequential([keras.layers.Dense(units=1, input_shape=[14])])
model.compile(optimizer='sgd', loss='mean_squared_error')

model.fit(X_train, y_train, epochs=500)

input_array = np.array([40,8,1,2,0,2,6,10,34,40,16,23,67,25])
input_array_for_prediction = np.expand_dims(input_array,axis=0)

print(model.predict(input_array_for_prediction))

这是我的model.summary()

的截图

enter image description here

这是我的整个 java 代码:-

import androidx.appcompat.app.AppCompatActivity;

import android.content.*;
import android.content.res.*;
import android.view.View;
import android.widget.*;
import org.tensorflow.lite.*;

import java.io.FileInputStream;
import java.io.IOException;
import java.nio.MappedByteBuffer;
import java.nio.ByteBuffer;
import java.nio.ByteOrder;
import java.nio.channels.FileChannel;

import android.os.Bundle;

public class Prediction extends AppCompatActivity {

    EditText inputNumber1,inputNumber2,inputNumber3,inputNumber4,inputNumber5,inputNumber6,inputNumber7,inputNumber8,inputNumber9,inputNumber10,inputNumber11,inputNumber12,inputNumber13,inputNumber14;
    Button inferButton;
    TextView outputNumber;
    Interpreter tflite;

    @Override
    protected void onCreate(Bundle savedInstanceState) {
        super.onCreate(savedInstanceState);
        setContentView(R.layout.activity_prediction);

        inputNumber1=(EditText)findViewById(R.id.editText);
        inputNumber2=(EditText)findViewById(R.id.editText2);
        inputNumber3=(EditText)findViewById(R.id.editText3);
        inputNumber4=(EditText)findViewById(R.id.editText4);
        inputNumber5=(EditText)findViewById(R.id.editText5);
        inputNumber6=(EditText)findViewById(R.id.editText6);
        inputNumber7=(EditText)findViewById(R.id.editText7);
        inputNumber8=(EditText)findViewById(R.id.editText8);
        inputNumber9=(EditText)findViewById(R.id.editText9);
        inputNumber10=(EditText)findViewById(R.id.editText10);
        inputNumber11=(EditText)findViewById(R.id.editText11);
        inputNumber12=(EditText)findViewById(R.id.editText12);
        inputNumber13=(EditText)findViewById(R.id.editText13);
        inputNumber14=(EditText)findViewById(R.id.editText14);
        outputNumber=(TextView)findViewById(R.id.outputNumber);
        inferButton=(Button)findViewById(R.id.predictButton);

        try{
            tflite=new Interpreter(loadModelFile());
        }catch(Exception ex){
            ex.printStackTrace();
        }
        inferButton.setOnClickListener(new View.OnClickListener(){
            @Override
            public void onClick(View view){

                int getText1=Integer.parseInt(inputNumber1.getText().toString());
                int getText2=Integer.parseInt(inputNumber2.getText().toString());
                int getText3=Integer.parseInt(inputNumber3.getText().toString());
                int getText4=Integer.parseInt(inputNumber4.getText().toString());
                int getText5=Integer.parseInt(inputNumber5.getText().toString());
                int getText6=Integer.parseInt(inputNumber6.getText().toString());
                int getText7=Integer.parseInt(inputNumber7.getText().toString());
                int getText8=Integer.parseInt(inputNumber8.getText().toString());
                int getText9=Integer.parseInt(inputNumber9.getText().toString());
                int getText10=Integer.parseInt(inputNumber10.getText().toString());
                int getText11=Integer.parseInt(inputNumber11.getText().toString());
                int getText12=Integer.parseInt(inputNumber12.getText().toString());
                int getText13=Integer.parseInt(inputNumber13.getText().toString());
                int getText14=Integer.parseInt(inputNumber14.getText().toString());

                int attributes[]={getText1,getText2,getText3,getText4,getText5,getText5,getText6,getText7,getText8,getText9,getText10,getText10,getText11,getText12,getText13,getText14};

                int prediction=doInference(attributes);
                //float prediction=doInference(inputNumber.getText().toString());
                outputNumber.setText(prediction);
            }
        });
    }
    public int doInference(int [] attributeArray){
        int[] inputVal=new int[14];
        for(int i=0;i<14;i++) {
            inputVal[i] = attributeArray[i];
        }
        int[][] outputval=new int[1][1];

        //Run inference passing the input shape and getting the output shape
        tflite.run(inputVal, outputval);

        //Inferred value is at [0][0]
        int inferredValue=outputval[0][0];

        return inferredValue;
    }
    private MappedByteBuffer loadModelFile() throws IOException {
        AssetFileDescriptor fileDescriptor=this.getAssets().openFd("kidney_test_model.tflite");
        FileInputStream inputStream = new FileInputStream(fileDescriptor.getFileDescriptor());
        FileChannel fileChannel = inputStream.getChannel();
        long startOffset = fileDescriptor.getStartOffset();
        long declaredLength = fileDescriptor.getDeclaredLength();
        return fileChannel.map(FileChannel.MapMode.READ_ONLY, startOffset, declaredLength);
    }

}

感谢任何帮助。谢谢。

最佳答案

在检查我的 build.gradle 文件之前，我遇到了同样的问题。你需要输入

android {
...

    aaptOptions {
        noCompress "tflite"
    }
}

在你的 build.gradle 文件中，然后同步它。

关于java - Tensorflow Lite Android 应用程序崩溃并出现 NullPointerException 'void org.tensorflow.lite.Interpreter.run(java.lang.Object, java.lang.Object)'，我们在Stack Overflow上找到一个类似的问题： https://stackoverflow.com/questions/60580404/

上一篇：logging - 如何将 RabbitMQ 日志记录级别设置为调试，真的？

下一篇：javascript - 如何在 react 中用鼠标选择范围单元格？

相关文章：

java - 如何使用 Spring Boot 开发支持中断(暂停、连接丢失等)的 get 端点？

java - 使用 REST WS 处理附件

python - (fields.E301) 字段定义与模型 'auth.User' 的关系，已被换出

android - 如何从 Samsung Galaxy Phone 禁用 KiesUsbService？

java - Java搜索: "block"

Java 手动客户端证书身份验证

python - 如何用鼠标动态调整QHBoxLayout和QGraphicsView的大小？ ( python ，Pyqt4)

python - 如何将嵌套循环转换为 python 中的列表理解

android - 什么时候需要在应用程序中创建一个单独的进程？

android - 如何创建带圆角的条纹按钮

©2024 IT工具网  联系我们