Reorganize package structure and update imports to use `org.beehive.g…

README.md

@@ -209,7 +209,7 @@ llama-tornado --gpu --model beehive-llama-3.2-1b-instruct-fp16.gguf --prompt "te
     @/home/mikepapadim/manchester/TornadoVM/bin/sdk/etc/exportLists/opencl-exports \
     --add-modules ALL-SYSTEM,tornado.runtime,tornado.annotation,tornado.drivers.common,tornado.drivers.opencl \
     -cp /home/mikepapadim/repos/gpu-llama3.java/target/gpu-llama3-1.0-SNAPSHOT.jar \
-    com.example.LlamaApp \
+    org.beehive.gpullama3.LlamaApp \
     -m beehive-llama-3.2-1b-instruct-fp16.gguf \
     --temperature 0.1 \
     --top-p 0.95 \

llama-tornado

@@ -172,7 +172,7 @@ class LlamaRunner:
             [
                 "-cp",
                 f"{self.llama_root}/target/gpu-llama3-1.0-SNAPSHOT.jar",
-                "com.example.LlamaApp",
+                "org.beehive.gpullama3.LlamaApp",
             ]
         )
         cmd.extend(module_config)

pom.xml

@@ -4,7 +4,7 @@
          xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
     <modelVersion>4.0.0</modelVersion>
 
-    <groupId>com.example</groupId>
+    <groupId>org.beehive.gpullama3</groupId>
     <artifactId>gpu-llama3</artifactId>
     <version>1.0-SNAPSHOT</version>
 
@@ -61,7 +61,7 @@
                         <configuration>
                             <transformers>
                                 <transformer implementation="org.apache.maven.plugins.shade.resource.ManifestResourceTransformer">
-                                    <mainClass>com.example.LlamaApp</mainClass>
+                                    <mainClass>org.beehive.gpullama3.LlamaApp</mainClass>
                                 </transformer>
                             </transformers>
                         </configuration>

src/main/java/com/example/LlamaApp.java → src/main/java/org/beehive/gpullama3/LlamaApp.java

@@ -1,13 +1,13 @@
-package com.example;
+package org.beehive.gpullama3;
 
-import com.example.aot.AOT;
-import com.example.core.model.tensor.FloatTensor;
-import com.example.inference.sampler.CategoricalSampler;
-import com.example.inference.sampler.Sampler;
-import com.example.inference.sampler.ToppSampler;
-import com.example.model.loader.ModelLoader;
-import com.example.model.Model;
-import com.example.tornadovm.FloatArrayUtils;
+import org.beehive.gpullama3.aot.AOT;
+import org.beehive.gpullama3.core.model.tensor.FloatTensor;
+import org.beehive.gpullama3.inference.sampler.CategoricalSampler;
+import org.beehive.gpullama3.inference.sampler.Sampler;
+import org.beehive.gpullama3.inference.sampler.ToppSampler;
+import org.beehive.gpullama3.model.loader.ModelLoader;
+import org.beehive.gpullama3.model.Model;
+import org.beehive.gpullama3.tornadovm.FloatArrayUtils;
 import uk.ac.manchester.tornado.api.types.arrays.FloatArray;
 
 import java.io.IOException;

src/main/java/com/example/Options.java → src/main/java/org/beehive/gpullama3/Options.java

@@ -1,4 +1,4 @@
-package com.example;
+package org.beehive.gpullama3;
 
 import java.io.PrintStream;
 import java.nio.file.Path;

src/main/java/com/example/aot/AOT.java → src/main/java/org/beehive/gpullama3/aot/AOT.java

@@ -1,15 +1,15 @@
-package com.example.aot;
+package org.beehive.gpullama3.aot;
 
-import com.example.auxiliary.Timer;
-import com.example.core.model.GGUF;
-import com.example.core.model.tensor.GGMLTensorEntry;
-import com.example.model.loader.LlamaModelLoader;
-import com.example.model.Model;
-import com.example.Options;
-import com.example.model.format.LlamaChatFormat;
-import com.example.model.llama.Llama;
-import com.example.inference.weights.Weights;
-import com.example.tokenizer.impl.LlamaTokenizer;
+import org.beehive.gpullama3.auxiliary.Timer;
+import org.beehive.gpullama3.core.model.GGUF;
+import org.beehive.gpullama3.core.model.tensor.GGMLTensorEntry;
+import org.beehive.gpullama3.model.loader.LlamaModelLoader;
+import org.beehive.gpullama3.model.Model;
+import org.beehive.gpullama3.Options;
+import org.beehive.gpullama3.model.format.LlamaChatFormat;
+import org.beehive.gpullama3.model.llama.Llama;
+import org.beehive.gpullama3.inference.weights.Weights;
+import org.beehive.gpullama3.tokenizer.impl.LlamaTokenizer;
 
 import java.io.IOException;
 import java.nio.channels.FileChannel;

src/main/java/com/example/auxiliary/LastRunMetrics.java → src/main/java/org/beehive/gpullama3/auxiliary/LastRunMetrics.java

@@ -1,4 +1,4 @@
-package com.example.auxiliary;
+package org.beehive.gpullama3.auxiliary;
 
 /**
  * Record to store metrics from the last model run.

src/main/java/com/example/auxiliary/Parallel.java → src/main/java/org/beehive/gpullama3/auxiliary/Parallel.java

@@ -1,4 +1,4 @@
-package com.example.auxiliary;
+package org.beehive.gpullama3.auxiliary;
 
 import java.util.function.IntConsumer;
 import java.util.function.LongConsumer;

src/main/java/com/example/auxiliary/Timer.java → src/main/java/org/beehive/gpullama3/auxiliary/Timer.java

@@ -1,4 +1,4 @@
-package com.example.auxiliary;
+package org.beehive.gpullama3.auxiliary;
 
 import java.util.concurrent.TimeUnit;
 

src/main/java/com/example/auxiliary/Tuple2.java → src/main/java/org/beehive/gpullama3/auxiliary/Tuple2.java

@@ -1,4 +1,4 @@
-package com.example.auxiliary;
+package org.beehive.gpullama3.auxiliary;
 
 public class Tuple2<T, U> {
     private final T first;

src/main/java/com/example/auxiliary/Utf8Mask.java → src/main/java/org/beehive/gpullama3/auxiliary/Utf8Mask.java

@@ -1,4 +1,4 @@
-package com.example.auxiliary;
+package org.beehive.gpullama3.auxiliary;
 
 /** mask of a byte-sequence in UTF-8 encoding */
 public record Utf8Mask(int mask, int pattern, int len) {

src/main/java/com/example/core/model/GGMLType.java → src/main/java/org/beehive/gpullama3/core/model/GGMLType.java

@@ -1,6 +1,4 @@
-package com.example.core.model;
-
-import com.example.core.types.Float16;
+package org.beehive.gpullama3.core.model;
 
 public enum GGMLType {
     // Floating point types

src/main/java/com/example/core/model/GGUF.java → src/main/java/org/beehive/gpullama3/core/model/GGUF.java

@@ -1,11 +1,10 @@
-package com.example.core.model;
+package org.beehive.gpullama3.core.model;
 
-import com.example.auxiliary.Timer;
-import com.example.core.model.GGUF.GGUFTensorInfo;
-import com.example.core.model.tensor.FloatTensor;
-import com.example.core.model.tensor.GGMLTensorEntry;
-import com.example.core.types.MetadataValueType;
-import com.example.core.types.Pair;
+import org.beehive.gpullama3.auxiliary.Timer;
+import org.beehive.gpullama3.core.model.tensor.FloatTensor;
+import org.beehive.gpullama3.core.model.tensor.GGMLTensorEntry;
+import org.beehive.gpullama3.core.types.MetadataValueType;
+import org.beehive.gpullama3.core.types.Pair;
 
 import java.io.FileNotFoundException;
 import java.io.IOException;

src/main/java/com/example/core/model/tensor/ArrayFloatTensor.java → src/main/java/org/beehive/gpullama3/core/model/tensor/ArrayFloatTensor.java

@@ -1,14 +1,12 @@
-package com.example.core.model.tensor;
+package org.beehive.gpullama3.core.model.tensor;
 
-import com.example.core.model.GGMLType;
+import org.beehive.gpullama3.core.model.GGMLType;
 import jdk.incubator.vector.FloatVector;
 import jdk.incubator.vector.VectorSpecies;
 
 import java.lang.foreign.MemorySegment;
 import java.util.Arrays;
 
-import static com.example.LlamaApp.USE_VECTOR_API;
-
 public final class ArrayFloatTensor extends FloatTensor {
 
     final float[] values;

src/main/java/com/example/core/model/tensor/F16FloatTensor.java → src/main/java/org/beehive/gpullama3/core/model/tensor/F16FloatTensor.java

@@ -1,6 +1,6 @@
-package com.example.core.model.tensor;
+package org.beehive.gpullama3.core.model.tensor;
 
-import com.example.core.model.GGMLType;
+import org.beehive.gpullama3.core.model.GGMLType;
 import jdk.incubator.vector.FloatVector;
 import jdk.incubator.vector.ShortVector;
 import jdk.incubator.vector.VectorOperators;

src/main/java/com/example/core/model/tensor/F32FloatTensor.java → src/main/java/org/beehive/gpullama3/core/model/tensor/F32FloatTensor.java

@@ -1,6 +1,6 @@
-package com.example.core.model.tensor;
+package org.beehive.gpullama3.core.model.tensor;
 
-import com.example.core.model.GGMLType;
+import org.beehive.gpullama3.core.model.GGMLType;
 import jdk.incubator.vector.FloatVector;
 import jdk.incubator.vector.VectorSpecies;
 

src/main/java/com/example/core/model/tensor/FloatTensor.java → src/main/java/org/beehive/gpullama3/core/model/tensor/FloatTensor.java

@@ -1,14 +1,13 @@
-package com.example.core.model.tensor;
+package org.beehive.gpullama3.core.model.tensor;
 
-import com.example.auxiliary.Parallel;
-import com.example.core.model.GGMLType;
+import org.beehive.gpullama3.auxiliary.Parallel;
+import org.beehive.gpullama3.core.model.GGMLType;
 import jdk.incubator.vector.FloatVector;
 import jdk.incubator.vector.VectorShape;
 import jdk.incubator.vector.VectorSpecies;
 import sun.misc.Unsafe;
 
 import java.lang.foreign.MemorySegment;
-import java.lang.foreign.ValueLayout;
 import java.lang.reflect.Field;
 import java.util.Arrays;
 

src/main/java/com/example/core/model/tensor/GGMLTensorEntry.java → src/main/java/org/beehive/gpullama3/core/model/tensor/GGMLTensorEntry.java

@@ -1,6 +1,6 @@
-package com.example.core.model.tensor;
+package org.beehive.gpullama3.core.model.tensor;
 
-import com.example.core.model.GGMLType;
+import org.beehive.gpullama3.core.model.GGMLType;
 
 import java.lang.foreign.MemorySegment;
 

src/main/java/com/example/core/model/tensor/Q4_0FloatTensor.java → src/main/java/org/beehive/gpullama3/core/model/tensor/Q4_0FloatTensor.java

@@ -1,8 +1,8 @@
-package com.example.core.model.tensor;
+package org.beehive.gpullama3.core.model.tensor;
 
-import com.example.LlamaApp;
-import com.example.core.model.GGMLType;
-import com.example.core.types.Float16;
+import org.beehive.gpullama3.LlamaApp;
+import org.beehive.gpullama3.core.model.GGMLType;
+import org.beehive.gpullama3.core.types.Float16;
 import jdk.incubator.vector.ByteVector;
 import jdk.incubator.vector.FloatVector;
 import jdk.incubator.vector.VectorOperators;

src/main/java/com/example/core/model/tensor/Q8_0FloatTensor.java → src/main/java/org/beehive/gpullama3/core/model/tensor/Q8_0FloatTensor.java

@@ -1,8 +1,8 @@
-package com.example.core.model.tensor;
+package org.beehive.gpullama3.core.model.tensor;
 
 
-import com.example.core.model.GGMLType;
-import com.example.core.types.Float16;
+import org.beehive.gpullama3.core.model.GGMLType;
+import org.beehive.gpullama3.core.types.Float16;
 import jdk.incubator.vector.ByteVector;
 import jdk.incubator.vector.FloatVector;
 import jdk.incubator.vector.VectorOperators;
@@ -12,8 +12,6 @@
 import java.lang.foreign.ValueLayout;
 import java.nio.ByteOrder;
 
-import static com.example.LlamaApp.USE_VECTOR_API;
-
 public final class Q8_0FloatTensor extends FloatTensor {
 
     final int size;

src/main/java/com/example/core/types/Float16.java → src/main/java/org/beehive/gpullama3/core/types/Float16.java

@@ -1,4 +1,4 @@
-package com.example.core.types;
+package org.beehive.gpullama3.core.types;
 
 public final class Float16 {
     public static final int BYTES = 2;

src/main/java/com/example/core/types/MetadataValueType.java → src/main/java/org/beehive/gpullama3/core/types/MetadataValueType.java

@@ -1,4 +1,4 @@
-package com.example.core.types;
+package org.beehive.gpullama3.core.types;
 
 public enum MetadataValueType {
     // The value is a 8-bit unsigned integer.

src/main/java/com/example/core/types/Pair.java → src/main/java/org/beehive/gpullama3/core/types/Pair.java

@@ -1,4 +1,4 @@
-package com.example.core.types;
+package org.beehive.gpullama3.core.types;
 
 public record Pair<First, Second>(First first, Second second) {
 }

src/main/java/com/example/inference/InferenceCore.java → src/main/java/org/beehive/gpullama3/inference/InferenceCore.java

@@ -1,18 +1,18 @@
-package com.example.inference;
-
-import com.example.auxiliary.Parallel;
-import com.example.core.model.tensor.FloatTensor;
-import com.example.inference.state.Phi3State;
-import com.example.inference.state.State;
-import com.example.inference.weights.standard.Phi3StandardWeights;
-import com.example.inference.weights.standard.Qwen3StandardWeights;
-import com.example.inference.weights.standard.StandardWeights;
-import com.example.inference.weights.tornado.TornadoWeights;
-import com.example.model.Configuration;
-import com.example.model.Model;
-import com.example.model.phi3.Phi3Configuration;
-import com.example.model.qwen3.Qwen3Configuration;
-import com.example.tornadovm.TornadoVMMasterPlan;
+package org.beehive.gpullama3.inference;
+
+import org.beehive.gpullama3.auxiliary.Parallel;
+import org.beehive.gpullama3.core.model.tensor.FloatTensor;
+import org.beehive.gpullama3.inference.state.Phi3State;
+import org.beehive.gpullama3.inference.state.State;
+import org.beehive.gpullama3.inference.weights.standard.Phi3StandardWeights;
+import org.beehive.gpullama3.inference.weights.standard.Qwen3StandardWeights;
+import org.beehive.gpullama3.inference.weights.standard.StandardWeights;
+import org.beehive.gpullama3.inference.weights.tornado.TornadoWeights;
+import org.beehive.gpullama3.model.Configuration;
+import org.beehive.gpullama3.model.Model;
+import org.beehive.gpullama3.model.phi3.Phi3Configuration;
+import org.beehive.gpullama3.model.qwen3.Qwen3Configuration;
+import org.beehive.gpullama3.tornadovm.TornadoVMMasterPlan;
 import uk.ac.manchester.tornado.api.types.arrays.FloatArray;
 
 import java.lang.foreign.MemorySegment;

src/main/java/com/example/inference/InferenceEngine.java → src/main/java/org/beehive/gpullama3/inference/InferenceEngine.java

@@ -1,12 +1,12 @@
-package com.example.inference;
-
-import com.example.auxiliary.LastRunMetrics;
-import com.example.inference.sampler.Sampler;
-import com.example.inference.state.State;
-import com.example.model.Configuration;
-import com.example.model.Model;
-import com.example.tokenizer.impl.Tokenizer;
-import com.example.tornadovm.TornadoVMMasterPlan;
+package org.beehive.gpullama3.inference;
+
+import org.beehive.gpullama3.auxiliary.LastRunMetrics;
+import org.beehive.gpullama3.inference.sampler.Sampler;
+import org.beehive.gpullama3.inference.state.State;
+import org.beehive.gpullama3.model.Configuration;
+import org.beehive.gpullama3.model.Model;
+import org.beehive.gpullama3.tokenizer.impl.Tokenizer;
+import org.beehive.gpullama3.tornadovm.TornadoVMMasterPlan;
 import uk.ac.manchester.tornado.api.types.arrays.FloatArray;
 
 import java.io.ByteArrayOutputStream;

src/main/java/com/example/inference/operation/RoPE.java → src/main/java/org/beehive/gpullama3/inference/operation/RoPE.java

@@ -1,6 +1,6 @@
-package com.example.inference.operation;
+package org.beehive.gpullama3.inference.operation;
 
-import com.example.core.types.Pair;
+import org.beehive.gpullama3.core.types.Pair;
 
 public final class RoPE {
     public static Pair<float[], float[]> precomputeFreqsCis(int contextLength, int headSize, double theta,

src/main/java/com/example/inference/sampler/CategoricalSampler.java → src/main/java/org/beehive/gpullama3/inference/sampler/CategoricalSampler.java

@@ -1,6 +1,6 @@
-package com.example.inference.sampler;
+package org.beehive.gpullama3.inference.sampler;
 
-import com.example.core.model.tensor.FloatTensor;
+import org.beehive.gpullama3.core.model.tensor.FloatTensor;
 import uk.ac.manchester.tornado.api.types.arrays.FloatArray;
 
 import java.util.random.RandomGenerator;

src/main/java/com/example/inference/sampler/Sampler.java → src/main/java/org/beehive/gpullama3/inference/sampler/Sampler.java

@@ -1,6 +1,6 @@
-package com.example.inference.sampler;
+package org.beehive.gpullama3.inference.sampler;
 
-import com.example.core.model.tensor.FloatTensor;
+import org.beehive.gpullama3.core.model.tensor.FloatTensor;
 import uk.ac.manchester.tornado.api.types.arrays.FloatArray;
 
 /**

src/main/java/com/example/inference/sampler/ToppSampler.java → src/main/java/org/beehive/gpullama3/inference/sampler/ToppSampler.java

@@ -1,6 +1,6 @@
-package com.example.inference.sampler;
+package org.beehive.gpullama3.inference.sampler;
 
-import com.example.core.model.tensor.FloatTensor;
+import org.beehive.gpullama3.core.model.tensor.FloatTensor;
 import uk.ac.manchester.tornado.api.types.arrays.FloatArray;
 
 import java.util.Comparator;

src/main/java/com/example/inference/state/LlamaState.java → src/main/java/org/beehive/gpullama3/inference/state/LlamaState.java

@@ -1,8 +1,8 @@
-package com.example.inference.state;
+package org.beehive.gpullama3.inference.state;
 
-import com.example.core.model.tensor.ArrayFloatTensor;
-import com.example.core.model.tensor.FloatTensor;
-import com.example.model.Configuration;
+import org.beehive.gpullama3.core.model.tensor.ArrayFloatTensor;
+import org.beehive.gpullama3.core.model.tensor.FloatTensor;
+import org.beehive.gpullama3.model.Configuration;
 import uk.ac.manchester.tornado.api.types.arrays.FloatArray;
 import uk.ac.manchester.tornado.api.types.arrays.IntArray;
 

src/main/java/com/example/inference/state/Phi3State.java → src/main/java/org/beehive/gpullama3/inference/state/Phi3State.java

@@ -1,9 +1,9 @@
-package com.example.inference.state;
+package org.beehive.gpullama3.inference.state;
 
-import com.example.core.model.tensor.ArrayFloatTensor;
-import com.example.core.model.tensor.FloatTensor;
-import com.example.model.Configuration;
-import com.example.model.phi3.Phi3Configuration;
+import org.beehive.gpullama3.core.model.tensor.ArrayFloatTensor;
+import org.beehive.gpullama3.core.model.tensor.FloatTensor;
+import org.beehive.gpullama3.model.Configuration;
+import org.beehive.gpullama3.model.phi3.Phi3Configuration;
 import uk.ac.manchester.tornado.api.types.arrays.FloatArray;
 import uk.ac.manchester.tornado.api.types.arrays.IntArray;
 

src/main/java/com/example/inference/state/Qwen3State.java → src/main/java/org/beehive/gpullama3/inference/state/Qwen3State.java

@@ -1,9 +1,9 @@
-package com.example.inference.state;
+package org.beehive.gpullama3.inference.state;
 
-import com.example.core.model.tensor.ArrayFloatTensor;
-import com.example.core.model.tensor.FloatTensor;
-import com.example.model.Configuration;
-import com.example.model.qwen3.Qwen3Configuration;
+import org.beehive.gpullama3.core.model.tensor.ArrayFloatTensor;
+import org.beehive.gpullama3.core.model.tensor.FloatTensor;
+import org.beehive.gpullama3.model.Configuration;
+import org.beehive.gpullama3.model.qwen3.Qwen3Configuration;
 import uk.ac.manchester.tornado.api.types.arrays.FloatArray;
 import uk.ac.manchester.tornado.api.types.arrays.IntArray;