Benchmark tensorflow model in Android.

build.gradle

apply plugin: 'com.android.application'
apply plugin: 'kotlin-android'
apply plugin: 'kotlin-android-extensions'

project.ext.ASSET_DIR = projectDir.toString() + '/assets'

android {
    compileSdkVersion 26
    defaultConfig {
        applicationId "tfexample.myapp.com.myapplication"
        minSdkVersion 21
        targetSdkVersion 26
        versionCode 1
        versionName "1.0"
        testInstrumentationRunner "android.support.test.runner.AndroidJUnitRunner"
    }
    buildTypes {
        release {
            minifyEnabled false
            proguardFiles getDefaultProguardFile('proguard-android.txt'), 'proguard-rules.pro'
        }
    }
    sourceSets {
        main {
            assets.srcDirs = [project.ext.ASSET_DIR]
        }
    }
}

dependencies {
    implementation fileTree(dir: 'libs', include: ['*.jar'])
    implementation"org.jetbrains.kotlin:kotlin-stdlib-jre7:$kotlin_version"
    implementation 'com.android.support:appcompat-v7:26.1.0'
    implementation 'com.android.support.constraint:constraint-layout:1.0.2'
    implementation 'org.tensorflow:tensorflow-android:1.4.0'
    testImplementation 'junit:junit:4.12'
    androidTestImplementation 'com.android.support.test:runner:1.0.1'
    androidTestImplementation 'com.android.support.test.espresso:espresso-core:3.0.1'
}

MainActivity.kt

package tfexample.myapp.com.myapplication

import android.os.Bundle
import android.support.v7.app.AppCompatActivity
import android.util.Log
import org.tensorflow.contrib.android.TensorFlowInferenceInterface
import kotlin.system.measureTimeMillis


val modelName = "file:///android_asset/mobile_unet_160_100_100.pb"

class MainActivity : AppCompatActivity() {

    override fun onCreate(savedInstanceState: Bundle?) {
        super.onCreate(savedInstanceState)
        setContentView(R.layout.activity_main)

        measureInference(modelName, 160, 160)
    }

    private fun measureInference(modelName: String, height: Long, width: Long) {
        val inferenceInterface = TensorFlowInferenceInterface(assets, modelName)
        val inputs = FloatArray((width * height * 3).toInt())
        val outputs = FloatArray((width * height).toInt())
        
        // It's for excluding 1st measurement.
        inferenceInterface.feed("input_1", inputs, 1, height, width, 3)
        inferenceInterface.run(arrayOf("output_0"))
        inferenceInterface.fetch("output_0", outputs)
        
        val times = (1..10).map {
            measureTimeMillis {
                inferenceInterface.feed("input_1", inputs, 1, height, width, 3)
                inferenceInterface.run(arrayOf("output_0"))
                inferenceInterface.fetch("output_0", outputs)
            }
        }

        Log.d("MainActivity", times.toString())
        Log.d("MainActivity", times.average().toString())
    }
}

Hi
Where I can download the pretrained model