Update some stuff

Author: kirklandsign

examples/demo-apps/android/LlamaDemo/app/build.gradle.kts

@@ -60,6 +60,7 @@ dependencies {
   implementation(files("libs/executorch.aar"))
   implementation("com.google.android.material:material:1.12.0")
   implementation("androidx.activity:activity:1.9.0")
+  implementation("org.json:json:20250107")
   testImplementation("junit:junit:4.13.2")
   androidTestImplementation("androidx.test.ext:junit:1.1.5")
   androidTestImplementation("androidx.test.espresso:espresso-core:3.5.1")

examples/demo-apps/android/LlamaDemo/app/src/androidTest/java/com/example/executorchllamademo/PerfTest.java

@@ -18,6 +18,7 @@
 import java.util.ArrayList;
 import java.util.Arrays;
 import java.util.List;
+import org.json.JSONObject;
 import org.junit.Test;
 import org.junit.runner.RunWith;
 import org.pytorch.executorch.extension.llm.LlmCallback;
@@ -64,7 +65,12 @@ public void onResult(String result) {
   }
 
   @Override
-  public void onStats(float tps) {
+  public void onStats(String stats) {
+    JSONObject jsonObject = new JSONObject(stats);
+    int numGeneratedTokens = jsonObject.getInt("num_generated_tokens");
+    int inferenceEndMs = jsonObject.getInt("inference_end_ms");
+    int promptEvalEndMs = jsonObject.getInt("prompt_eval_end_ms");
+    float tps = (float) numGeneratedTokens / (inferenceEndMs - promptEvalEndMs) * 1000;
     tokensPerSecond.add(tps);
   }
 

examples/demo-apps/android/LlamaDemo/app/src/main/java/com/example/executorchllamademo/MainActivity.java

@@ -49,6 +49,7 @@
 import java.util.List;
 import java.util.concurrent.Executor;
 import java.util.concurrent.Executors;
+import org.json.JSONObject;
 import org.pytorch.executorch.extension.llm.LlmCallback;
 import org.pytorch.executorch.extension.llm.LlmModule;
 
@@ -97,10 +98,15 @@ public void onResult(String result) {
   }
 
   @Override
-  public void onStats(float tps) {
+  public void onStats(String result) {
     runOnUiThread(
         () -> {
           if (mResultMessage != null) {
+            JSONObject jsonObject = new JSONObject(stats);
+            int numGeneratedTokens = jsonObject.getInt("num_generated_tokens");
+            int inferenceEndMs = jsonObject.getInt("inference_end_ms");
+            int promptEvalEndMs = jsonObject.getInt("prompt_eval_end_ms");
+            float tps = (float) numGeneratedTokens / (inferenceEndMs - promptEvalEndMs) * 1000;
             mResultMessage.setTokensPerSecond(tps);
             mMessageAdapter.notifyDataSetChanged();
           }

examples/demo-apps/android/LlamaDemo/app/src/main/java/com/example/executorchllamademo/ModelRunner.java

@@ -13,6 +13,7 @@
 import android.os.Looper;
 import android.os.Message;
 import androidx.annotation.NonNull;
+import org.json.JSONObject;
 import org.pytorch.executorch.extension.llm.LlmCallback;
 import org.pytorch.executorch.extension.llm.LlmModule;
 
@@ -69,7 +70,12 @@ public void onResult(String result) {
   }
 
   @Override
-  public void onStats(float tps) {
+  public void onStats(String stats) {
+    JSONObject jsonObject = new JSONObject(stats);
+    int numGeneratedTokens = jsonObject.getInt("num_generated_tokens");
+    int inferenceEndMs = jsonObject.getInt("inference_end_ms");
+    int promptEvalEndMs = jsonObject.getInt("prompt_eval_end_ms");
+    float tps = (float) numGeneratedTokens / (inferenceEndMs - promptEvalEndMs) * 1000;
     mCallback.onStats("tokens/second: " + tps);
   }
 }

examples/demo-apps/android/LlamaDemo/app/src/main/java/com/example/executorchllamademo/ModelRunnerCallback.java

@@ -18,7 +18,7 @@ public interface ModelRunnerCallback {
 
   void onTokenGenerated(String token);
 
-  void onStats(String token);
+  void onStats(String stats);
 
   void onGenerationStopped();
 }

extension/android/executorch_android/src/main/java/org/pytorch/executorch/extension/llm/LlmCallback.java

@@ -37,7 +37,7 @@ public interface LlmCallback {
    */
   @Deprecated
   @DoNotStrip
-  public void onStats(float tps);
+  default public void onStats(float tps) {}
 
   /**
    * Called when the statistics for the generate() is available.