From efbf99801d5dc94addd3db2d58eeb30633a7ce74 Mon Sep 17 00:00:00 2001 From: ro-jun Date: Tue, 31 Dec 2024 06:15:05 +0900 Subject: [PATCH] [Title] OpenAI Embeddings [Version] initial [Language] ENG [Packages] langchain_openai, scikit-learn, matplotlib - Added available embedding dimension - Enhanced PCA visualization (documents and query) --- 08-EMBEDDING/01-OpenAIEmbeddings.ipynb | 433 +++++++++++++++++++++++++ 1 file changed, 433 insertions(+) create mode 100644 08-EMBEDDING/01-OpenAIEmbeddings.ipynb diff --git a/08-EMBEDDING/01-OpenAIEmbeddings.ipynb b/08-EMBEDDING/01-OpenAIEmbeddings.ipynb new file mode 100644 index 000000000..c7395c07c --- /dev/null +++ b/08-EMBEDDING/01-OpenAIEmbeddings.ipynb @@ -0,0 +1,433 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# OpenAI Embeddings\n", + "\n", + "- Author: [ro__o_jun](https://github.com/ro-jun)\n", + "- Design: []()\n", + "- Peer Review: []() []()\n", + "- This is a part of [LangChain Open Tutorial](https://github.com/LangChain-OpenTutorial/LangChain-OpenTutorial)\n", + "\n", + "[![Open in Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/langchain-ai/langchain-academy/blob/main/module-4/sub-graph.ipynb) [![Open in LangChain Academy](https://cdn.prod.website-files.com/65b8cd72835ceeacd4449a53/66e9eba12c7b7688aa3dbb5e_LCA-badge-green.svg)](https://academy.langchain.com/courses/take/intro-to-langgraph/lessons/58239937-lesson-2-sub-graphs)\n", + "\n", + "## Overview\n", + "\n", + "This tutorial explores the use of `OpenAI Text embedding` models within the `LangChain` framework. \n", + "\n", + "It showcases how to generate embeddings for text queries and documents, reduce their dimensionality using **PCA** , and visualize them in 2D for better interpretability. \n", + "\n", + "By analyzing relationships between the query and documents through `cosine similarity`, it provides insights into how embeddings can enhance workflows, including **text analysis** and **data visualization**. \n", + "\n", + "### Table of Contents\n", + "\n", + "1. [Environment Setup](#environment-setup)\n", + "2. [Load model and set dimension](#Load-model-and-set-dimension)\n", + "3. [Similarity Calculation (Cosine Similarity)](#similarity-calculation-cosine-similarity)\n", + "4. [Embeddings Visualization(PCA)](#Embeddings-Visualization(PCA))\n", + "\n", + "### References\n", + "\n", + "- [OpenAI](https://openai.com/index/new-embedding-models-and-api-updates/)\n", + "- [LangChain OpenAI Embeddings](https://python.langchain.com/api_reference/openai/embeddings/langchain_openai.embeddings.base.OpenAIEmbeddings.html)\n", + "- [Cosine Similarity](https://en.wikipedia.org/wiki/Cosine_similarity)\n", + "- [Principal component analysis](https://en.wikipedia.org/wiki/Principal_component_analysis)\n", + "----" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Environment Setup\n", + "\n", + "Set up the environment. You may refer to [Environment Setup](https://wikidocs.net/257836) for more details.\n", + "\n", + "**[Note]**\n", + "- `langchain-opentutorial` is a package that provides a set of easy-to-use environment setup, useful functions and utilities for tutorials. \n", + "- You can checkout the [`langchain-opentutorial`](https://github.com/LangChain-OpenTutorial/langchain-opentutorial-pypi) for more details." + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "metadata": {}, + "outputs": [], + "source": [ + "%%capture --no-stderr\n", + "%pip install langchain-opentutorial" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "metadata": {}, + "outputs": [], + "source": [ + "# Install required packages\n", + "from langchain_opentutorial import package\n", + "\n", + "package.install(\n", + " [\n", + " \"langchain_openai\",\n", + " \"scikit-learn\",\n", + " \"matplotlib\",\n", + " ],\n", + " verbose=False,\n", + " upgrade=False,\n", + ")" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Environment variables have been set successfully.\n" + ] + } + ], + "source": [ + "# Set environment variables\n", + "from langchain_opentutorial import set_env\n", + "\n", + "set_env(\n", + " {\n", + " \"OPENAI_API_KEY\": \"\",\n", + " \"LANGCHAIN_API_KEY\": \"\",\n", + " \"LANGCHAIN_TRACING_V2\": \"true\",\n", + " \"LANGCHAIN_ENDPOINT\": \"https://api.smith.langchain.com\",\n", + " \"LANGCHAIN_PROJECT\": \"OpenAI-Embeddings\",\n", + " },\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "[Note] If you are using a `.env` file, proceed as follows." + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "True" + ] + }, + "execution_count": 4, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "from dotenv import load_dotenv\n", + "\n", + "load_dotenv(override=True)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Load model and set dimension\n", + "\n", + "Describes the `Embedding` model and dimension settings supported by OpenAI.\n", + "\n", + "### Why Adjust Embedding Dimensions?\n", + "- **Optimize Resources** : Shortened embeddings use less memory and compute.\n", + "- **Flexible Usage** : Models like text-embedding-3-large allow size reduction with the dimensions API.\n", + "- **Key Insight** : Even at 256 dimensions, performance can surpass larger models like text-embedding-ada-002." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "This is a **description of the models** supported by OpenAI\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
Model~ Pages per DollarPerformance on MTEB EvalMax InputAvailable dimension
text-embedding-3-small62,50062.3%8191512, 1536
text-embedding-3-large9,61564.6%8191256, 1024, 3072
text-embedding-ada-00212,50061.0%81911536
" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "langchain_openai를 사용하여 OpenAI 임베딩 모델 초기화 및 활용" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "metadata": {}, + "outputs": [], + "source": [ + "from langchain_openai import OpenAIEmbeddings\n", + "\n", + "# Set desired model\n", + "openai_embedding = OpenAIEmbeddings(model=\"text-embedding-3-large\")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**[note]** If dimension reduction is necessary, please set as below." + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "metadata": {}, + "outputs": [], + "source": [ + "from langchain_openai import OpenAIEmbeddings\n", + "\n", + "# Set desired model and dimension\n", + "openai_embedding = OpenAIEmbeddings(model=\"text-embedding-3-large\", dimensions=1024)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "`Query` and `document` definitions" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "metadata": {}, + "outputs": [], + "source": [ + "query = \"What is the Open AI's gpt embedding model?\"\n", + "\n", + "# Various embedding models\n", + "documents = [\n", + " \"all-mpnet-base-v2\", \"bert-base-nli-mean-tokens\", \"bert-large-nli-mean-tokens\", \"distilbert-base-nli-mean-tokens\", \"roberta-base-nli-stsb-mean-tokens\", \"roberta-large-nli-stsb-mean-tokens\", \"sentence-transformers/all-MiniLM-L12-v2\", \"sentence-transformers/all-distilroberta-v1\", \"sentence-transformers/paraphrase-MiniLM-L3-v2\", \"sentence-transformers/paraphrase-mpnet-base-v2\", \"sentence-transformers/msmarco-distilbert-base-v3\", \"sentence-transformers/msmarco-MiniLM-L6-cos-v5\", \"sentence-transformers/msmarco-roberta-base-v2\", \"xlnet-base-cased\", \"facebook/bart-large\", \"facebook/dpr-question_encoder-single-nq-base\", \"google/electra-small-discriminator\", \"google/electra-base-discriminator\", \"google/electra-large-discriminator\", \"deepset/sentence_bert\", \"deepset/roberta-base-squad2\", \"gpt-neo-125M\", \"gpt-neo-1.3B\", \"gpt-neo-2.7B\", \"gpt-j-6B\", \"text-embedding-ada-002\", \"text-embedding-3-small\", \"text-embedding-3-large\", \"all-MiniLM-L6-v2\", \"multilingual-e5-base\",\n", + "]" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Now we embed the query and document using the set embedding model." + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "number of documents: 30\n", + "dimension: 3072\n", + "query: [-0.00288043892942369, 0.020187586545944214, -0.011613684706389904, -0.02147459052503109, 0.02403634414076805]\n", + "documents[0]: [-0.03383158519864082, 0.004126258660107851, -0.025896472856402397, -0.013592381030321121, -0.0021926583722233772]\n", + "documents[1]: [0.0051429239101707935, -0.015500376932322979, -0.019089050590991974, -0.027715347707271576, -0.00695410929620266]\n" + ] + } + ], + "source": [ + "query_vector = openai_embedding.embed_query(query)\n", + "docs_vector = openai_embedding.embed_documents(documents)\n", + "\n", + "print(\"number of documents: \" + str(len(docs_vector)))\n", + "print(\"dimension: \" + str(len(docs_vector[0])))\n", + "\n", + "# Part of the sliced ​​vector\n", + "print(\"query: \" + str(query_vector[:5]))\n", + "print(\"documents[0]: \" + str(docs_vector[0][:5]))\n", + "print(\"documents[1]: \" + str(docs_vector[1][:5]))" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Similarity Calculation (Cosine Similarity)\n", + "\n", + "This code calculates the similarity between the query and the document through `Cosine Similarity` . \n", + "Find the documents **similar (top 3) and (bottom 3)** ." + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Top 3 most similar document:\n", + "[1] similarity: 0.514 | text-embedding-3-large\n", + "[2] similarity: 0.467 | text-embedding-ada-002\n", + "[3] similarity: 0.457 | text-embedding-3-small\n", + "\n", + "Bottom 3 least similar documents:\n", + "[1] similarity: 0.050 | facebook/bart-large\n", + "[2] similarity: 0.143 | multilingual-e5-base\n", + "[3] similarity: 0.171 | all-mpnet-base-v2\n" + ] + } + ], + "source": [ + "from sklearn.metrics.pairwise import cosine_similarity\n", + "\n", + "# Calculate Cosine Similarity\n", + "similarity = cosine_similarity([query_vector], docs_vector)\n", + "\n", + "#Sorting by in descending order\n", + "sorted_idx = similarity.argsort()[0][::-1]\n", + "\n", + "# Display top 3 and bottom 3 documents based on similarity\n", + "print(\"Top 3 most similar document:\")\n", + "for i in range(0, 3):\n", + " print(\n", + " f\"[{i+1}] similarity: {similarity[0][sorted_idx[i]]:.3f} | {documents[sorted_idx[i]]}\"\n", + " )\n", + "\n", + "print(\"\\nBottom 3 least similar documents:\")\n", + "for i in range(1, 4):\n", + " print(\n", + " f\"[{i}] similarity: {similarity[0][sorted_idx[-i]]:.3f} | {documents[sorted_idx[-i]]}\"\n", + " )" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Embeddings visualization(PCA)\n", + "\n", + "Reduce the dimensionality of the embeddings for visualization purposes. \n", + "This code uses **principal component analysis (PCA)** to reduce high-dimensional embedding vectors to **two dimensions**. \n", + "The resulting **2D points** are displayed in a scatterplot, with each point labeled for its corresponding document.\n", + "\n", + "### Why Dimension Reduction?\n", + "High-dimensional embedding vectors are challenging to interpret and analyze directly.\n", + "By reducing them to 2D, we can:\n", + "\n", + "- **Visually explore relationships** between embeddings (e.g., clustering, grouping).\n", + "- **Identify patterns or anomalies** in the data that may not be obvious in high dimensions.\n", + "- **Improve interpretability** , making the data more accessible for human analysis and decision-making.\n" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "import matplotlib.pyplot as plt\n", + "from sklearn.decomposition import PCA\n", + "import numpy as np\n", + "\n", + "# Combine documents and query for PCA\n", + "all_vectors = np.vstack([docs_vector, query_vector]) # Stack query vector with docs\n", + "pca = PCA(n_components=2)\n", + "reduced_vectors = pca.fit_transform(all_vectors)\n", + "\n", + "# Separate reduced vectors for documents and query\n", + "doc_vectors_2d = reduced_vectors[:-1] # All but the last point (documents)\n", + "query_vector_2d = reduced_vectors[-1] # Last point (query)\n", + "\n", + "# Plot the reduced vectors\n", + "plt.scatter(doc_vectors_2d[:, 0], doc_vectors_2d[:, 1], color=\"blue\", label=\"Documents\")\n", + "plt.scatter(query_vector_2d[0], query_vector_2d[1], color=\"red\", label=\"Query\", marker='x', s=300)\n", + "\n", + "# Annotate document points\n", + "for i, doc in enumerate(documents):\n", + " plt.text(doc_vectors_2d[i, 0], doc_vectors_2d[i, 1], doc, fontsize=8)\n", + "\n", + "# Add plot details\n", + "plt.title(\"2D Visualization of Embedding Vectors with Query\")\n", + "plt.xlabel(\"PCA Dimension 1\")\n", + "plt.ylabel(\"PCA Dimension 2\")\n", + "plt.legend()\n", + "plt.show()\n" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "langchain-opentutorial-XrZComUd-py3.11", + "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": "3.11.9" + } + }, + "nbformat": 4, + "nbformat_minor": 2 +}