Added BAAI/bge-small-en-v1.5 models (#17)

Author: dliubarskyi

langchain4j-embeddings-bge-small-en-v15-q/pom.xml

@@ -0,0 +1,55 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project xmlns="http://maven.apache.org/POM/4.0.0"
+         xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
+         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>
+
+    <parent>
+        <groupId>dev.langchain4j</groupId>
+        <artifactId>langchain4j-embeddings-parent</artifactId>
+        <version>0.29.0-SNAPSHOT</version>
+        <relativePath>../pom.xml</relativePath>
+    </parent>
+
+    <artifactId>langchain4j-embeddings-bge-small-en-v15-q</artifactId>
+    <name>langchain4j-embeddings-bge-small-en-v15-q</name>
+    <description>In-process bge-small-en-v1.5 (quantized) embedding model</description>
+
+    <dependencies>
+
+        <dependency>
+            <groupId>dev.langchain4j</groupId>
+            <artifactId>langchain4j-embeddings</artifactId>
+            <version>${project.version}</version>
+        </dependency>
+
+        <dependency>
+            <groupId>org.junit.jupiter</groupId>
+            <artifactId>junit-jupiter-engine</artifactId>
+            <scope>test</scope>
+        </dependency>
+
+        <dependency>
+            <groupId>org.junit.jupiter</groupId>
+            <artifactId>junit-jupiter-params</artifactId>
+            <scope>test</scope>
+        </dependency>
+
+        <dependency>
+            <groupId>org.assertj</groupId>
+            <artifactId>assertj-core</artifactId>
+            <scope>test</scope>
+        </dependency>
+
+    </dependencies>
+
+    <licenses>
+        <license>
+            <name>Apache-2.0</name>
+            <url>https://www.apache.org/licenses/LICENSE-2.0.txt</url>
+            <distribution>repo</distribution>
+            <comments>A business-friendly OSS license</comments>
+        </license>
+    </licenses>
+
+</project>

langchain4j-embeddings-bge-small-en-v15-q/src/main/java/dev/langchain4j/model/embedding/bge/small/en/v15/BgeSmallEnV15QuantizedEmbeddingModel.java

@@ -0,0 +1,32 @@
+package dev.langchain4j.model.embedding.bge.small.en.v15;
+
+import dev.langchain4j.model.embedding.AbstractInProcessEmbeddingModel;
+import dev.langchain4j.model.embedding.OnnxBertBiEncoder;
+import dev.langchain4j.model.embedding.PoolingMode;
+
+/**
+ * Quantized BAAI bge-small-en-v1.5 embedding model that runs within your Java application's process.
+ * <p>
+ * Maximum length of text (in tokens) that can be embedded at once: unlimited.
+ * However, while you can embed very long texts, the quality of the embedding degrades as the text lengthens.
+ * It is recommended to embed segments of no more than 512 tokens long.
+ * <p>
+ * Embedding dimensions: 384
+ * <p>
+ * It is recommended to add "Represent this sentence for searching relevant passages:" prefix to a query.
+ * <p>
+ * More details <a href="https://huggingface.co/BAAI/bge-small-en-v1.5">here</a>
+ */
+public class BgeSmallEnV15QuantizedEmbeddingModel extends AbstractInProcessEmbeddingModel {
+
+    private static final OnnxBertBiEncoder MODEL = loadFromJar(
+            "bge-small-en-v1.5-q.onnx",
+            "bge-small-en-v1.5-tokenizer.json",
+            PoolingMode.CLS
+    );
+
+    @Override
+    protected OnnxBertBiEncoder model() {
+        return MODEL;
+    }
+}

langchain4j-embeddings-bge-small-en-v15-q/src/main/resources/README

@@ -0,0 +1,11 @@
+Original model card:
+https://huggingface.co/BAAI/bge-small-en-v1.5
+
+ONNX model card:
+https://huggingface.co/Xenova/bge-small-en-v1.5
+
+ONNX model file:
+https://huggingface.co/Xenova/bge-small-en-v1.5/resolve/main/onnx/model_quantized.onnx
+
+Tokenizer:
+https://huggingface.co/Xenova/bge-small-en-v1.5/resolve/main/tokenizer.json

langchain4j-embeddings-bge-small-en-v15-q/src/main/resources/bge-small-en-v1.5-q.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6c9c6101a956d62dfb5e7190c538226c0c5bb9cb27b651234b6df063ee7dbfe4
+size 34014426