Updated Text Classification Android to use Task Library (BertNLClassifier + NLClassifier)

lite/examples/object_detection/android/app/src/main/res/layout/tfe_od_layout_bottom_sheet.xml

@@ -163,7 +163,7 @@
         android:layout_width="match_parent"
         android:layout_height="wrap_content"
         android:orientation="horizontal"
-        android:visibility="gone">
+       >
 
         <TextView
             android:id="@+id/api"

lite/examples/text_classification/android/README.md

@@ -1,9 +1,11 @@
 # TensorFlow Lite text classification sample
 
+<img src="https://www.tensorflow.org/lite/examples/text_classification/images/screenshot.gif" width="400px" alt="Video">
+
 ## Overview
 
 This is an end-to-end example of movie review sentiment classification built
-with TensorFlow 2.0 (Keras API), and trained on IMDB dataset. The demo app
+with TensorFlow 2.0 (Keras API), and trained on [IMDB dataset](http://ai.stanford.edu/%7Eamaas/data/sentiment/) version 1.0. The demo app
 processes input movie review texts, and classifies its sentiment into negative
 (0) or positive (1).
 
@@ -14,19 +16,23 @@ mobile app.
 ## Model
 
 See
-[Text Classification with Movie Reviews](https://www.tensorflow.org/tutorials/keras/basic_text_classification)
+[Text Classification with Movie Reviews](https://www.tensorflow.org/tutorials/keras/text_classification)
 for a step-by-step instruction of building a simple text classification model.
 
-## Android app
-
-Follow the steps below to build and run the sample Android app.
+## Build the demo using Android Studio
+                                                                                                                                          
+### Prerequisites
 
-### Requirements
+*   If you don't have already, install
+    [Android Studio](https://developer.android.com/studio/index.html), following
+    the instructions on the website.
 
-*   Android Studio 3.2 or later. Install instructions can be found on
-    [Android Studio](https://developer.android.com/studio/index.html) website.
+*   Android Studio 3.2 or later.
+    - Gradle 4.6 or higher.
+    - SDK Build Tools 29.0.2 or higher.
 
-*   An Android device or an Android emulator and with API level higher than 15.
+*   You need an Android device or Android emulator and Android development
+    environment with minimum API 21.
 
 ### Building
 
@@ -53,34 +59,26 @@ Follow the steps below to build and run the sample Android app.
 
 *   Click `Run` to run the demo app on your Android device.
 
-#### Switch between inference solutions (Task library vs TFLite Interpreter)
+## Build the demo using gradle (command line)
+
+### Building and Installing
 
-This Text Classification Android reference app demonstrates two implementation
-solutions:
+*   Use the following command to build a demo apk:
 
-(1)
-[`lib_task_api`](https://github.com/tensorflow/examples/tree/master/lite/examples/nl_classification/android/lib_task_api)
-that leverages the out-of-box API from the
-[TensorFlow Lite Task Library](https://www.tensorflow.org/lite/inference_with_metadata/task_library/text_classifier);
+```
+cd lite/examples/bert_qa/android   # Folder for Android app.
 
-(2)
-[`lib_interpreter`](https://github.com/tensorflow/examples/tree/master/lite/examples/text_classification/android/lib_interpreter)
-that creates the custom inference pipleline using the
-[TensorFlow Lite Interpreter Java API](https://www.tensorflow.org/lite/guide/inference#load_and_run_a_model_in_java).
+./gradlew build
+```
 
-The [`build.gradle`](app/build.gradle) inside `app` folder shows how to change
-`flavorDimensions "tfliteInference"` to switch between the two solutions.
+*   Use the following command to install the apk onto your connected device:
 
-Inside **Android Studio**, you can change the build variant to whichever one you
-want to build and run—just go to `Build > Select Build Variant` and select one
-from the drop-down menu. See
-[configure product flavors in Android Studio](https://developer.android.com/studio/build/build-variants#product-flavors)
-for more details.
+```
+adb install app/build/outputs/apk/debug/app-debug.apk
+```
 
-For gradle CLI, running `./gradlew build` can create APKs for both solutions
-under `app/build/outputs/apk`.
+## Assets folder
 
-*Note: If you simply want the out-of-box API to run the app, we recommend
-`lib_task_api`for inference. If you want to customize your own models and
-control the detail of inputs and outputs, it might be easier to adapt your model
-inputs and outputs by using `lib_interpreter`.*
+_Do not delete the assets folder content_. If you explicitly deleted the files,
+choose `Build -> Rebuild` to re-download the deleted model files into the assets
+folder.

lite/examples/text_classification/android/app/build.gradle

@@ -1,12 +1,13 @@
 apply plugin: 'com.android.application'
+apply plugin: 'de.undercouch.download'
 
 android {
-    compileSdkVersion 28
-    buildToolsVersion "29.0.0"
+    compileSdkVersion 29
+    buildToolsVersion "29.0.2"
     defaultConfig {
         applicationId "org.tensorflow.lite.examples.textclassification"
         minSdkVersion 21
-        targetSdkVersion 28
+        targetSdkVersion 29
         versionCode 1
         versionName "1.0"
         testInstrumentationRunner "android.support.test.runner.AndroidJUnitRunner"
@@ -29,27 +30,22 @@ android {
             includeAndroidResources = true
         }
     }
-
-    flavorDimensions "tfliteInference"
-    productFlavors {
-        // The TFLite inference is built using the TFLite Java interpreter.
-        interpreter {
-            dimension "tfliteInference"
-        }
-        // Default: The TFLite inference is built using the TFLite Task library (high-level API).
-        taskApi {
-            getIsDefault().set(true)
-            dimension "tfliteInference"
-        }
-    }
 }
 
+// Download the pre-trained model from the internet
+project.ext.ASSET_DIR = projectDir.toString() + '/src/main/assets'
+apply from:'download_model.gradle'
+
 dependencies {
-    interpreterImplementation project(":lib_interpreter")
-    taskApiImplementation project(":lib_task_api")
+    implementation fileTree(dir: 'libs', include: ['*.jar'])
+
     implementation 'androidx.appcompat:appcompat:1.1.0'
     implementation 'androidx.constraintlayout:constraintlayout:1.1.3'
     implementation 'org.jetbrains:annotations:15.0'
+    implementation 'com.google.android.material:material:1.0.0'
+
+    //Task Text Library dependency
+    implementation 'org.tensorflow:tensorflow-lite-task-text:0.2.0'
 
     testImplementation 'androidx.test:core:1.2.0'
     testImplementation 'junit:junit:4.12'

lite/examples/text_classification/android/app/src/androidTest/java/org/tensorflow/lite/examples/textclassification/UnitTest.java

@@ -0,0 +1,49 @@
+package org.tensorflow.lite.examples.textclassification;
+
+
+import org.junit.Before;
+import org.junit.Test;
+import org.junit.runner.RunWith;
+import org.tensorflow.lite.examples.textclassification.ml.Result;
+import org.tensorflow.lite.examples.textclassification.ml.TextClassificationClient;
+
+import static org.junit.Assert.*;
+
+import android.content.Context;
+import android.support.test.InstrumentationRegistry;
+import android.support.test.runner.AndroidJUnit4;
+
+/** Tests of {@link TextClassificationClient} */
+@RunWith(AndroidJUnit4.class)
+public final class UnitTest {
+    private TextClassificationClient client;
+    private String api = "NLCLASSIFIER";
+
+    @Before
+    public void setUp() {
+        // Context of the app under test.
+        Context appContext = InstrumentationRegistry.getInstrumentation().getTargetContext();
+
+        client = new TextClassificationClient(appContext, api);
+        client.load();
+    }
+//
+//    @Test
+//    public void loadModelTest() {
+//        assertNotNull(client.classifier);
+//    }
+
+    @Test
+    public void predictTest() {
+        Result positiveText =
+                     client
+                        .classify("This is an interesting film. My family and I all liked it very much.")
+                        .get(0);
+        assertEquals("Positive", positiveText.getTitle());
+        assertTrue(positiveText.getConfidence() > 0.55);
+        Result negativeText =
+                client.classify("This film cannot be worse. It is way too boring.").get(0);
+        assertEquals("Negative", negativeText.getTitle());
+        assertTrue(negativeText.getConfidence() > 0.6);
+    }
+}