{ "cells": [ { "cell_type": "markdown", "metadata": { "id": "cb7RLf9gpEoN" }, "source": [ "# RAG pipeline using FastEmbed for embeddings generation\n", "\n", "[FastEmbed](https://qdrant.github.io/fastembed/) is a lightweight, fast, Python library built for embedding generation, maintained by Qdrant. \n", "It is suitable for generating embeddings efficiently and fast on CPU-only machines.\n", "\n", "In this notebook, we will use FastEmbed-Haystack integration to generate embeddings for indexing and RAG.\n", "\n", "**Haystack Useful Sources**\n", "\n", "* [Docs](https://docs.haystack.deepset.ai/docs/intro)\n", "* [Tutorials](https://haystack.deepset.ai/tutorials)\n", "* [Other Cookbooks](https://github.com/deepset-ai/haystack-cookbook)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Install dependencies" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "id": "tnSq1XK_ovZV" }, "outputs": [], "source": [ "!pip install fastembed-haystack qdrant-haystack wikipedia transformers" ] }, { "cell_type": "markdown", "metadata": { "id": "x8Bpy1ri_Ipx" }, "source": [ "## Download contents and create docs" ] }, { "cell_type": "code", "execution_count": 1, "metadata": { "id": "2NIxMNQTzcfV" }, "outputs": [], "source": [ "favourite_bands=\"\"\"Audioslave\n", "Green Day\n", "Muse (band)\n", "Foo Fighters (band)\n", "Nirvana (band)\"\"\".split(\"\\n\")" ] }, { "cell_type": "code", "execution_count": 3, "metadata": { "id": "7FpCSSnUzuuP" }, "outputs": [], "source": [ "import wikipedia\n", "from haystack.dataclasses import Document\n", "\n", "raw_docs=[]\n", "for title in favourite_bands:\n", " page = wikipedia.page(title=title, auto_suggest=False)\n", " doc = Document(content=page.content, meta={\"title\": page.title, \"url\":page.url})\n", " raw_docs.append(doc)" ] }, { "cell_type": "markdown", "metadata": { "id": "DLiNhYKV_g8u" }, "source": [ "## Clean, split and index documents on Qdrant" ] }, { "cell_type": "code", "execution_count": 4, "metadata": { "id": "a1taDmfx1HCM" }, "outputs": [], "source": [ "from haystack_integrations.document_stores.qdrant import QdrantDocumentStore\n", "from haystack.components.preprocessors import DocumentCleaner, DocumentSplitter\n", "from haystack_integrations.components.embedders.fastembed import FastembedDocumentEmbedder\n", "from haystack.document_stores.types import DuplicatePolicy" ] }, { "cell_type": "code", "execution_count": 5, "metadata": { "id": "eAKP9icf1Inj" }, "outputs": [], "source": [ "document_store = QdrantDocumentStore(\n", " \":memory:\",\n", " embedding_dim =384,\n", " recreate_index=True,\n", " return_embedding=True,\n", " wait_result_from_api=True,\n", ")" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "id": "6rbZVTxR907n" }, "outputs": [], "source": [ "cleaner = DocumentCleaner()\n", "splitter = DocumentSplitter(split_by='period', split_length=3)\n", "splitted_docs = splitter.run(cleaner.run(raw_docs)[\"documents\"])" ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "493" ] }, "execution_count": 7, "metadata": {}, "output_type": "execute_result" } ], "source": [ "len(splitted_docs[\"documents\"])" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### FastEmbed Document Embedder\n", "\n", "Here we are initializing the FastEmbed Document Embedder and using it to generate embeddings for the documents.\n", "We are using a small and good model, `BAAI/bge-small-en-v1.5` and specifying the `parallel` parameter to 0 to use all available CPU cores for embedding generation.\n", "\n", "⚠️ If you are running this notebook on Google Colab, please note that Google Colab only provides 2 CPU cores, so the embedding generation could be not as fast as it can be on a standard machine.\n", "\n", "For more information on FastEmbed-Haystack integration, please refer to the [documentation](https://docs.haystack.deepset.ai/docs/fastembeddocumentembedder) and [API reference](https://docs.haystack.deepset.ai/reference/fastembed-embedders)." ] }, { "cell_type": "code", "execution_count": null, "metadata": { "id": "U3s_68uw9FSW" }, "outputs": [ { "data": { "application/vnd.jupyter.widget-view+json": { "model_id": "a0d7dbcf196047dfa767e6e6b78374b2", "version_major": 2, "version_minor": 0 }, "text/plain": [ "Fetching 9 files: 0%| | 0/9 [00:00