Add embeddings for AwaEmbedding (#8353)

- Description: Adds AwaEmbeddings class for embeddings, which provides
users with a convenient way to do fine-tuning, as well as the potential
need for multimodality

  - Tag maintainer: @baskaryan

Create `Awa.ipynb`: an example notebook for AwaEmbeddings class
Modify `embeddings/__init__.py`: Import the class
Create `embeddings/awa.py`: The embedding class
Create `embeddings/test_awa.py`: The test file.

---------

Co-authored-by: taozhiwang <taozhiwa@gmail.com>

docs/extras/integrations/text_embedding/Awa.ipynb

@@ -0,0 +1,109 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "b14a24db",
+   "metadata": {},
+   "source": [
+    "# AwaEmbedding\n",
+    "\n",
+    "This notebook explains how to use AwaEmbedding, which is included in [awadb](https://github.com/awa-ai/awadb), to embedding texts in langchain."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "0ab948fc",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# pip install awadb"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "67c637ca",
+   "metadata": {},
+   "source": [
+    "## import the library"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "5709b030",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from langchain.embeddings import AwaEmbeddings"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "1756b1ba",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "Embedding = AwaEmbeddings()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "4a2a098d",
+   "metadata": {},
+   "source": [
+    "# Set embedding model\n",
+    "Users can use `Embedding.set_model()` to specify the embedding model. \\\n",
+    "The input of this function is a string which represents the model's name. \\\n",
+    "The list of currently supported models can be obtained [here](https://github.com/awa-ai/awadb) \\ \\ \n",
+    "\n",
+    "The **default model** is `all-mpnet-base-v2`, it can be used without setting."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "584b9af5",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "text = \"our embedding test\"\n",
+    "\n",
+    "Embedding.set_model(\"all-mpnet-base-v2\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "id": "be18b873",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "res_query = Embedding.embed_query(\"The test information\")\n",
+    "res_document = Embedding.embed_documents([\"test1\", \"another test\"])"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.11.4"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

libs/langchain/langchain/embeddings/__init__.py

@@ -6,6 +6,7 @@ from langchain.embeddings.aleph_alpha import (
     AlephAlphaAsymmetricSemanticEmbedding,
     AlephAlphaSymmetricSemanticEmbedding,
 )
+from langchain.embeddings.awa import AwaEmbeddings
 from langchain.embeddings.bedrock import BedrockEmbeddings
 from langchain.embeddings.clarifai import ClarifaiEmbeddings
 from langchain.embeddings.cohere import CohereEmbeddings
@@ -78,6 +79,7 @@ __all__ = [
     "NLPCloudEmbeddings",
     "GPT4AllEmbeddings",
     "LocalAIEmbeddings",
+    "AwaEmbeddings",
 ]
 
 

libs/langchain/langchain/embeddings/awa.py

@@ -0,0 +1,56 @@
+from typing import Any, Dict, List
+
+from pydantic import BaseModel, root_validator
+
+from langchain.embeddings.base import Embeddings
+
+
+class AwaEmbeddings(BaseModel, Embeddings):
+    client: Any  #: :meta private:
+    model: str = "all-mpnet-base-v2"
+
+    @root_validator()
+    def validate_environment(cls, values: Dict) -> Dict:
+        """Validate that awadb library is installed."""
+
+        try:
+            from awadb import AwaEmbedding
+        except ImportError as exc:
+            raise ImportError(
+                "Could not import awadb library. "
+                "Please install it with `pip install awadb`"
+            ) from exc
+        values["client"] = AwaEmbedding()
+        return values
+
+    def set_model(self, model_name: str) -> None:
+        """Set the model used for embedding.
+        The default model used is all-mpnet-base-v2
+
+        Args:
+            model_name: A string which represents the name of model.
+        """
+        self.model = model_name
+        self.client.model_name = model_name
+
+    def embed_documents(self, texts: List[str]) -> List[List[float]]:
+        """Embed a list of documents using AwaEmbedding.
+
+        Args:
+            texts: The list of texts need to be embedded
+
+        Returns:
+            List of embeddings, one for each text.
+        """
+        return self.client.EmbeddingBatch(texts)
+
+    def embed_query(self, text: str) -> List[float]:
+        """Compute query embeddings using AwaEmbedding.
+
+        Args:
+            text: The text to embed.
+
+        Returns:
+            Embeddings for the text.
+        """
+        return self.client.Embedding(text)

libs/langchain/tests/integration_tests/embeddings/test_awa.py

@@ -0,0 +1,19 @@
+"""Test Awa Embedding"""
+from langchain.embeddings.awa import AwaEmbeddings
+
+
+def test_awa_embedding_documents() -> None:
+    """Test Awa embeddings for documents."""
+    documents = ["foo bar", "test document"]
+    embedding = AwaEmbeddings()
+    output = embedding.embed_documents(documents)
+    assert len(output) == 2
+    assert len(output[0]) == 768
+
+
+def test_awa_embedding_query() -> None:
+    """Test Awa embeddings for query."""
+    document = "foo bar"
+    embedding = AwaEmbeddings()
+    output = embedding.embed_query(document)
+    assert len(output) == 768