chroma/examples/basic_functionality/local_persistence.ipynb

{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Local Peristence Demo\n",
    "This notebook demonstrates how to persist the in-memory version of Chroma to disk, then load it back in. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import chromadb\n",
    "from chromadb.config import Settings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running Chroma using direct local API.\n",
      "No existing DB found in db, skipping load\n",
      "No existing DB found in db, skipping load\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/antontroynikov/miniforge3/envs/chroma/lib/python3.9/site-packages/tqdm/auto.py:22: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
      "  from .autonotebook import tqdm as notebook_tqdm\n"
     ]
    }
   ],
   "source": [
    "# Create a new Chroma client with persistence enabled. \n",
    "persist_directory = \"db\"\n",
    "\n",
    "client = chromadb.Client(\n",
    "    Settings(\n",
    "        persist_directory=persist_directory,\n",
    "        chroma_db_impl=\"duckdb+parquet\",\n",
    "    )\n",
    ")\n",
    "\n",
    "# Start from scratch\n",
    "client.reset()\n",
    "\n",
    "# Create a new chroma collection\n",
    "collection_name = \"peristed_collection\"\n",
    "collection = client.create_collection(name=collection_name)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Add some data to the collection\n",
    "collection.add(\n",
    "    embeddings=[\n",
    "        [1.1, 2.3, 3.2],\n",
    "        [4.5, 6.9, 4.4],\n",
    "        [1.1, 2.3, 3.2],\n",
    "        [4.5, 6.9, 4.4],\n",
    "        [1.1, 2.3, 3.2],\n",
    "        [4.5, 6.9, 4.4],\n",
    "        [1.1, 2.3, 3.2],\n",
    "        [4.5, 6.9, 4.4],\n",
    "    ],\n",
    "    metadatas=[\n",
    "        {\"uri\": \"img1.png\", \"style\": \"style1\"},\n",
    "        {\"uri\": \"img2.png\", \"style\": \"style2\"},\n",
    "        {\"uri\": \"img3.png\", \"style\": \"style1\"},\n",
    "        {\"uri\": \"img4.png\", \"style\": \"style1\"},\n",
    "        {\"uri\": \"img5.png\", \"style\": \"style1\"},\n",
    "        {\"uri\": \"img6.png\", \"style\": \"style1\"},\n",
    "        {\"uri\": \"img7.png\", \"style\": \"style1\"},\n",
    "        {\"uri\": \"img8.png\", \"style\": \"style1\"},\n",
    "    ],\n",
    "    documents=[\"doc1\", \"doc2\", \"doc3\", \"doc4\", \"doc5\", \"doc6\", \"doc7\", \"doc8\"],\n",
    "    ids=[\"id1\", \"id2\", \"id3\", \"id4\", \"id5\", \"id6\", \"id7\", \"id8\"],\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Persisting DB to disk, putting it in the save folder db\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Persist the DB. This also happens automatically when the client is garbage collected.\n",
    "# In a notebook, prefer to call persist explicitly.\n",
    "client.persist()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running Chroma using direct local API.\n",
      "loaded in 8 embeddings\n",
      "loaded in 1 collections\n"
     ]
    }
   ],
   "source": [
    "# Create a new client with the same settings\n",
    "client = chromadb.Client(\n",
    "    Settings(\n",
    "        persist_directory=persist_directory,\n",
    "        chroma_db_impl=\"duckdb+parquet\",\n",
    "    )\n",
    ")\n",
    "\n",
    "# Load the collection\n",
    "collection = client.get_collection(collection_name)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'embeddings': [[[1.1, 2.3, 3.2]]], 'documents': [['doc5']], 'ids': [['id5']], 'metadatas': [[{'uri': 'img5.png', 'style': 'style1'}]], 'distances': [[0.0]]}\n"
     ]
    }
   ],
   "source": [
    "# Query the collection\n",
    "results = collection.query(\n",
    "    query_embeddings=[[1.1, 2.3, 3.2]],\n",
    "    n_results=1\n",
    ")\n",
    "\n",
    "print(results)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Persisting DB to disk, putting it in the save folder db\n"
     ]
    }
   ],
   "source": [
    "# Clean up\n",
    "client.reset()\n",
    "client.persist()\n",
    "\n",
    "# You can also just delete the persist directory\n",
    "!rm -rf db/"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "chroma",
   "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": "3.9.16"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "88f09714c9334832bac29166716f9f6a879ee2a4ed4822c1d4120cb2393b58dd"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}