multimodal_RAG_unified_text/multi_modal_demo.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Multimodal Retrieval Augmented Generation (RAG)\n",
    "\n",
    "##### Note: This example requires a KDB.AI endpoint and API key. Sign up for a free [KDB.AI account](https://kdb.ai/get-started).\n",
    "\n",
    "This example explores preparing, embedding, and storing both text and image data within a KDB.AI vector database. Our goal is to store both images descriptions and text within the same vector space so we can search over both modalities at the same time. KDB.AI can then be used as a retrieval tool within a RAG pipeline for both text and images data, making it a multimodal retriever. \n",
    "\n",
    "The dataset we are working with contains images of several different animals, and text descriptions of those animals. After implementation, a user should be able to write a query regarding any of these animals and get returned both images and text related to the animal of interest.\n",
    "\n",
    "The method in this notebook uses a multimodal LLM to describe images within our dataset. Since the images will be described in text, the descriptions can be treated similary to our raw text data. Therefore all data can be embedded by the same model, and stored in the same vector space!\n",
    "\n",
    "### Agenda:\n",
    "0. Setup\n",
    "1. Helper Functions\n",
    "2. Data Import\n",
    "3. Embed Texts and Images\n",
    "4. Set Up KDB.AI Vector Database table\n",
    "5. Insert data into KDB.AI Vector Database\n",
    "6. Query vector database to retrieve most relevant data\n",
    "7. Retrieval Augmented Generation\n",
    "8. Drop the table\n",
    "\n",
    "Animal Images Source: https://www.kaggle.com/datasets/iamsouravbanerjee/animal-image-dataset-90-different-animals"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 0. Setup\n",
    "Dependencies, Imports, and API Keys\n",
    "\n",
    "In order to successfully run this sample, note the following steps depending on where you are running this notebook:\n",
    "\n",
    "-***Run Locally / Private Environment:*** The [Setup](https://github.com/KxSystems/kdbai-samples/blob/main/README.md#setup) steps in the repository's `README.md` will guide you on prerequisites and how to run this with Jupyter.\n",
    "\n",
    "\n",
    "-***Colab / Hosted Environment:*** Open this notebook in Colab and run through the cells."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "!pip install kdbai_client langchain openai"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import os\n",
    "import io\n",
    "from getpass import getpass\n",
    "import langchain\n",
    "import openai\n",
    "from openai import OpenAI\n",
    "import numpy as np\n",
    "from PIL import Image\n",
    "from langchain.chat_models import ChatOpenAI\n",
    "from langchain.schema.messages import HumanMessage, SystemMessage\n",
    "from IPython.display import display\n",
    "import time\n",
    "import warnings\n",
    "import requests\n",
    "warnings.filterwarnings('ignore')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "### !!! Only run this cell if you need to download the data into your environment, for example in Colab\n",
    "### This downloads image and text data\n",
    "\n",
    "!mkdir -p ./data/images\n",
    "!mkdir -p ./data/text\n",
    "\n",
    "def get_github_repo_contents(repo_owner, repo_name, branch, folder_path):\n",
    "    # Construct the API URL\n",
    "    api_url = f\"https://api.github.com/repos/{repo_owner}/{repo_name}/contents/{folder_path}?ref={branch}\"\n",
    "    \n",
    "    # Send the request and process the response\n",
    "    contents = requests.get(api_url).json()\n",
    "\n",
    "    # Create the local directory if it doesn't exist\n",
    "    fPath = f\"./data/{folder_path.split('/')[-1]}\"\n",
    "\n",
    "    for item in contents:\n",
    "        # Recursively list contents of subfolder\n",
    "        if item['type'] == 'dir':\n",
    "            get_github_repo_contents(repo_owner, repo_name, branch, f\"{folder_path}/{item['name']}\")\n",
    "        # Download and save file\n",
    "        elif item['type'] == 'file':\n",
    "            file_url = f\"https://raw.githubusercontent.com/{repo_owner}/{repo_name}/{branch}/{folder_path}/{item['name']}\"\n",
    "            print(file_url)\n",
    "            r = requests.get(file_url, timeout=4.0)\n",
    "            r.raise_for_status()  # Raises an exception for HTTP errors\n",
    "            file_path = f\"{fPath}/{item['name']}\"\n",
    "\n",
    "            if item['name'].lower().endswith(('.png', '.jpg', '.jpeg', '.gif', '.bmp')):\n",
    "                # Save image file\n",
    "                with Image.open(io.BytesIO(r.content)) as im:\n",
    "                    im.save(file_path)\n",
    "            else:\n",
    "                # Save text file\n",
    "                with open(file_path, 'wb') as f:\n",
    "                    f.write(r.content)\n",
    "\n",
    "# Get images and texts\n",
    "get_github_repo_contents(\n",
    "    repo_owner='KxSystems',\n",
    "    repo_name='kdbai-samples',\n",
    "    branch='main',\n",
    "    folder_path='multimodal_RAG_unified_text/data'\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "OpenAI API Key:  ········\n"
     ]
    }
   ],
   "source": [
    "os.environ[\"OPENAI_API_KEY\"] = (\n",
    "    os.environ[\"OPENAI_API_KEY\"]\n",
    "    if \"OPENAI_API_KEY\" in os.environ\n",
    "    else getpass(\"OpenAI API Key: \")\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "openai.api_key = os.environ[\"OPENAI_API_KEY\"]\n",
    "client = OpenAI()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1. Helper functions\n",
    "1. encode_image: this function takes in an image file's path, and converts the image to base64 encoding which can be used by OpenAI's LLM \"gpt-4-vision-preview\"\n",
    "2. image_summarize: takes in a base64 encoded image and a prompt, returns a text description of the image\n",
    "3. read_text_from_file: takes in a file path, returns the text within that file\n",
    "4. text_to_embedding: takes in text, outputs a vector embedding of that text that can be stored in a vector database"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "import base64\n",
    "\n",
    "# Helper function to convert a file to base64 representation\n",
    "def encode_image(image_path):\n",
    "    with open(image_path, \"rb\") as image_file:\n",
    "        return base64.b64encode(image_file.read()).decode('utf-8')\n",
    "\n",
    "# Takes in a base64 encoded image and a prompt, returns a text description of the image\n",
    "def image_summarize(img_base64,prompt):\n",
    "    ''' Image summary '''\n",
    "    for _ in range(3):\n",
    "        response = openai.chat.completions.create(\n",
    "            model=\"gpt-4-vision-preview\",\n",
    "            messages=[\n",
    "                {\n",
    "                \"role\": \"user\",\n",
    "                \"content\": [\n",
    "                    {\"type\": \"text\", \"text\": prompt},\n",
    "                    {\n",
    "                    \"type\": \"image_url\",\n",
    "                    \"image_url\": {\n",
    "                        \"url\": f\"data:image/jpeg;base64,{img_base64}\",\n",
    "                    },\n",
    "                    },\n",
    "                ],\n",
    "                }\n",
    "            ],\n",
    "            max_tokens=150,\n",
    "        )\n",
    "        content = response.choices[0].message.content\n",
    "\n",
    "        if \"I'm sorry\" not in content:\n",
    "            return content\n",
    "        time.sleep(3)\n",
    "    return content\n",
    "\n",
    "#Return the text from a file\n",
    "def read_text_from_file(filename):\n",
    "    try:\n",
    "        # Open the file in read mode ('r')\n",
    "        with open(filename, 'r') as file:\n",
    "            # Read the contents of the file into a string\n",
    "            text = file.read()\n",
    "        return text\n",
    "    except IOError as e:\n",
    "        # Handle any I/O errors\n",
    "        print(f\"An error occurred: {e}\")\n",
    "        return None\n",
    "\n",
    "#model=\"text-embedding-3-small\"\n",
    "#Takes in text and returns a vector embedding using text-embedding-3-small\n",
    "def text_to_embedding(text):\n",
    "    text = text.replace(\"\\n\", \" \")\n",
    "    return client.embeddings.create(input = [text], model=\"text-embedding-3-small\").data[0].embedding\n",
    "   "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2. Data Import\n",
    "Get the filepaths for all text and image files\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "images = os.listdir(\"./data/images\")\n",
    "texts = os.listdir(\"./data/text\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Prepare a dataframe to store file path, media_type, text, and embeddings in\n",
    "columns = ['path','media_type','text','embeddings']\n",
    "df = pd.DataFrame(columns=columns)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3. Embed Texts and Images\n",
    "Loop through all text files and embed them, store relevant data in dataframe.\n",
    "Loop through all image files, get image summaries, embed the summaries, store relevant data in dataframe."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Loop through all text files and embed them, store relevant data in dataframe\n",
    "for text in texts:\n",
    "    path = \"./data/text/\" + text\n",
    "    media_type = \"text\"\n",
    "    text1 = read_text_from_file(path)\n",
    "    embedding = text_to_embedding(text1)\n",
    "    new_row = {'path': path,\n",
    "               'media_type':'text',\n",
    "               'text' : text1,\n",
    "               'embeddings': embedding}\n",
    "    df = pd.concat([df, pd.DataFrame([new_row])], ignore_index=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>path</th>\n",
       "      <th>media_type</th>\n",
       "      <th>text</th>\n",
       "      <th>embeddings</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>./data/text/deer.txt</td>\n",
       "      <td>text</td>\n",
       "      <td>Deer are hoofed mammals known for their gracef...</td>\n",
       "      <td>[0.07531881332397461, 0.03122134692966938, 0.0...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>./data/text/caterpillar.txt</td>\n",
       "      <td>text</td>\n",
       "      <td>Caterpillars are the larval stage of butterfli...</td>\n",
       "      <td>[0.05927688628435135, -0.017723916098475456, 0...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>./data/text/bat.txt</td>\n",
       "      <td>text</td>\n",
       "      <td>Bats are the only mammals capable of sustained...</td>\n",
       "      <td>[0.043051157146692276, 0.05017940700054169, 0....</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>./data/text/bear.txt</td>\n",
       "      <td>text</td>\n",
       "      <td>Bears are large mammals with a stocky body, po...</td>\n",
       "      <td>[0.052179791033267975, 0.007267165929079056, 0...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>./data/text/hedgehog.txt</td>\n",
       "      <td>text</td>\n",
       "      <td>Hedgehogs are small, nocturnal mammals known f...</td>\n",
       "      <td>[0.0144755644723773, 0.00036793306935578585, 0...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                          path media_type  \\\n",
       "0         ./data/text/deer.txt       text   \n",
       "1  ./data/text/caterpillar.txt       text   \n",
       "2          ./data/text/bat.txt       text   \n",
       "3         ./data/text/bear.txt       text   \n",
       "4     ./data/text/hedgehog.txt       text   \n",
       "\n",
       "                                                text  \\\n",
       "0  Deer are hoofed mammals known for their gracef...   \n",
       "1  Caterpillars are the larval stage of butterfli...   \n",
       "2  Bats are the only mammals capable of sustained...   \n",
       "3  Bears are large mammals with a stocky body, po...   \n",
       "4  Hedgehogs are small, nocturnal mammals known f...   \n",
       "\n",
       "                                          embeddings  \n",
       "0  [0.07531881332397461, 0.03122134692966938, 0.0...  \n",
       "1  [0.05927688628435135, -0.017723916098475456, 0...  \n",
       "2  [0.043051157146692276, 0.05017940700054169, 0....  \n",
       "3  [0.052179791033267975, 0.007267165929079056, 0...  \n",
       "4  [0.0144755644723773, 0.00036793306935578585, 0...  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Let's take a look at the dataframe so far\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Loop through all image files, get image summaries, embed the summaries, store relevant data in dataframe. Takes a couple of minutes\n",
    "for image in images:\n",
    "    path = \"./data/images/\" + image\n",
    "    media_type = \"images\"\n",
    "    base64_image = encode_image(path)\n",
    "    prompt = \"Describe the image in detail.\"\n",
    "    summarization = image_summarize(base64_image,prompt)\n",
    "    embedding = text_to_embedding(summarization)\n",
    "    new_row = {'path': path,\n",
    "               'media_type':'image',\n",
    "               'text' : summarization,\n",
    "               'embeddings': embedding}\n",
    "    df = pd.concat([df, pd.DataFrame([new_row])], ignore_index=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>path</th>\n",
       "      <th>media_type</th>\n",
       "      <th>text</th>\n",
       "      <th>embeddings</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>./data/text/deer.txt</td>\n",
       "      <td>text</td>\n",
       "      <td>Deer are hoofed mammals known for their gracef...</td>\n",
       "      <td>[0.07531881332397461, 0.03122134692966938, 0.0...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>./data/text/caterpillar.txt</td>\n",
       "      <td>text</td>\n",
       "      <td>Caterpillars are the larval stage of butterfli...</td>\n",
       "      <td>[0.05927688628435135, -0.017723916098475456, 0...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>./data/text/bat.txt</td>\n",
       "      <td>text</td>\n",
       "      <td>Bats are the only mammals capable of sustained...</td>\n",
       "      <td>[0.043051157146692276, 0.05017940700054169, 0....</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>./data/text/bear.txt</td>\n",
       "      <td>text</td>\n",
       "      <td>Bears are large mammals with a stocky body, po...</td>\n",
       "      <td>[0.052179791033267975, 0.007267165929079056, 0...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>./data/text/hedgehog.txt</td>\n",
       "      <td>text</td>\n",
       "      <td>Hedgehogs are small, nocturnal mammals known f...</td>\n",
       "      <td>[0.0144755644723773, 0.00036793306935578585, 0...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>./data/text/fox.txt</td>\n",
       "      <td>text</td>\n",
       "      <td>Foxes are small to medium-sized, omnivorous ma...</td>\n",
       "      <td>[0.01701943203806877, 0.017458384856581688, 0....</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>./data/images/caterpillar2.jpg</td>\n",
       "      <td>image</td>\n",
       "      <td>The image shows a hairy caterpillar perched up...</td>\n",
       "      <td>[0.07125178724527359, -0.013986187987029552, 0...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>./data/images/fox1.jpg</td>\n",
       "      <td>image</td>\n",
       "      <td>The image shows a fox seated on a snow-covered...</td>\n",
       "      <td>[0.008546297438442707, -0.029755879193544388, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>./data/images/deer1.jpg</td>\n",
       "      <td>image</td>\n",
       "      <td>The image features a solitary deer standing in...</td>\n",
       "      <td>[0.0711188018321991, -0.029708458110690117, -0...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>./data/images/caterpillar1.jpg</td>\n",
       "      <td>image</td>\n",
       "      <td>The image shows a green caterpillar on a woode...</td>\n",
       "      <td>[0.07201803475618362, -0.004797806032001972, 0...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>./data/images/bat1.jpg</td>\n",
       "      <td>image</td>\n",
       "      <td>The image shows a bat in mid-flight, captured ...</td>\n",
       "      <td>[0.020450666546821594, 0.00680321641266346, 0....</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>./data/images/bat2.jpg</td>\n",
       "      <td>image</td>\n",
       "      <td>The image shows a bat hanging upside down from...</td>\n",
       "      <td>[0.044002678245306015, -0.003725415328517556, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>./data/images/hedgehog1.jpg</td>\n",
       "      <td>image</td>\n",
       "      <td>The image shows a small, adorable hedgehog bei...</td>\n",
       "      <td>[0.06011241674423218, -0.03956378251314163, -0...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>./data/images/deer2.jpg</td>\n",
       "      <td>image</td>\n",
       "      <td>You've provided an image of a white-tailed dee...</td>\n",
       "      <td>[0.07073069363832474, 0.009371200576424599, -0...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>./data/images/hedgehog2.jpg</td>\n",
       "      <td>image</td>\n",
       "      <td>The image shows a hedgehog in a natural enviro...</td>\n",
       "      <td>[0.028540847823023796, -0.02835400030016899, -...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>./data/images/bear1.jpg</td>\n",
       "      <td>image</td>\n",
       "      <td>The image shows a bear walking toward the came...</td>\n",
       "      <td>[0.014313085936009884, -0.018470464274287224, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>./data/images/fox2.jpg</td>\n",
       "      <td>image</td>\n",
       "      <td>The image shows a red fox positioned in front ...</td>\n",
       "      <td>[0.008771030232310295, -0.025080697610974312, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>./data/images/bear2.jpg</td>\n",
       "      <td>image</td>\n",
       "      <td>The image shows a close-up portrait of a brown...</td>\n",
       "      <td>[0.023404834792017937, -0.016726182773709297, ...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                              path media_type  \\\n",
       "0             ./data/text/deer.txt       text   \n",
       "1      ./data/text/caterpillar.txt       text   \n",
       "2              ./data/text/bat.txt       text   \n",
       "3             ./data/text/bear.txt       text   \n",
       "4         ./data/text/hedgehog.txt       text   \n",
       "5              ./data/text/fox.txt       text   \n",
       "6   ./data/images/caterpillar2.jpg      image   \n",
       "7           ./data/images/fox1.jpg      image   \n",
       "8          ./data/images/deer1.jpg      image   \n",
       "9   ./data/images/caterpillar1.jpg      image   \n",
       "10          ./data/images/bat1.jpg      image   \n",
       "11          ./data/images/bat2.jpg      image   \n",
       "12     ./data/images/hedgehog1.jpg      image   \n",
       "13         ./data/images/deer2.jpg      image   \n",
       "14     ./data/images/hedgehog2.jpg      image   \n",
       "15         ./data/images/bear1.jpg      image   \n",
       "16          ./data/images/fox2.jpg      image   \n",
       "17         ./data/images/bear2.jpg      image   \n",
       "\n",
       "                                                 text  \\\n",
       "0   Deer are hoofed mammals known for their gracef...   \n",
       "1   Caterpillars are the larval stage of butterfli...   \n",
       "2   Bats are the only mammals capable of sustained...   \n",
       "3   Bears are large mammals with a stocky body, po...   \n",
       "4   Hedgehogs are small, nocturnal mammals known f...   \n",
       "5   Foxes are small to medium-sized, omnivorous ma...   \n",
       "6   The image shows a hairy caterpillar perched up...   \n",
       "7   The image shows a fox seated on a snow-covered...   \n",
       "8   The image features a solitary deer standing in...   \n",
       "9   The image shows a green caterpillar on a woode...   \n",
       "10  The image shows a bat in mid-flight, captured ...   \n",
       "11  The image shows a bat hanging upside down from...   \n",
       "12  The image shows a small, adorable hedgehog bei...   \n",
       "13  You've provided an image of a white-tailed dee...   \n",
       "14  The image shows a hedgehog in a natural enviro...   \n",
       "15  The image shows a bear walking toward the came...   \n",
       "16  The image shows a red fox positioned in front ...   \n",
       "17  The image shows a close-up portrait of a brown...   \n",
       "\n",
       "                                           embeddings  \n",
       "0   [0.07531881332397461, 0.03122134692966938, 0.0...  \n",
       "1   [0.05927688628435135, -0.017723916098475456, 0...  \n",
       "2   [0.043051157146692276, 0.05017940700054169, 0....  \n",
       "3   [0.052179791033267975, 0.007267165929079056, 0...  \n",
       "4   [0.0144755644723773, 0.00036793306935578585, 0...  \n",
       "5   [0.01701943203806877, 0.017458384856581688, 0....  \n",
       "6   [0.07125178724527359, -0.013986187987029552, 0...  \n",
       "7   [0.008546297438442707, -0.029755879193544388, ...  \n",
       "8   [0.0711188018321991, -0.029708458110690117, -0...  \n",
       "9   [0.07201803475618362, -0.004797806032001972, 0...  \n",
       "10  [0.020450666546821594, 0.00680321641266346, 0....  \n",
       "11  [0.044002678245306015, -0.003725415328517556, ...  \n",
       "12  [0.06011241674423218, -0.03956378251314163, -0...  \n",
       "13  [0.07073069363832474, 0.009371200576424599, -0...  \n",
       "14  [0.028540847823023796, -0.02835400030016899, -...  \n",
       "15  [0.014313085936009884, -0.018470464274287224, ...  \n",
       "16  [0.008771030232310295, -0.025080697610974312, ...  \n",
       "17  [0.023404834792017937, -0.016726182773709297, ...  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Print the entire dataframe\n",
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4. Set Up KDB.AI Vector Database table\n",
    "Time to get KDB.AI vector database set up. If you do not have an account, go to KDB.AI to sign-up free!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# vector DB imports\n",
    "import os\n",
    "from getpass import getpass\n",
    "import kdbai_client as kdbai\n",
    "import time"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "KDB.AI endpoint: your_kdbai_endpoint\n",
      "KDB.AI API key: ········\n"
     ]
    }
   ],
   "source": [
    "#Set up KDB.AI endpoint and API key\n",
    "KDBAI_ENDPOINT = (\n",
    "    os.environ[\"KDBAI_ENDPOINT\"]\n",
    "    if \"KDBAI_ENDPOINT\" in os.environ\n",
    "    else input(\"KDB.AI endpoint: \")\n",
    ")\n",
    "KDBAI_API_KEY = (\n",
    "    os.environ[\"KDBAI_API_KEY\"]\n",
    "    if \"KDBAI_API_KEY\" in os.environ\n",
    "    else getpass(\"KDB.AI API key: \")\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Connect to KDB.AI\n",
    "session = kdbai.Session(api_key=KDBAI_API_KEY, endpoint=KDBAI_ENDPOINT)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define table schema for our table with columns for path, media_type, text, and embeddings\n",
    "table_schema = {\n",
    "    \"columns\": [\n",
    "        {\"name\": \"path\", \"pytype\": \"str\"},\n",
    "        {\"name\": \"media_type\", \"pytype\": \"str\"},\n",
    "        {\"name\": \"text\", \"pytype\": \"str\"},\n",
    "        {\n",
    "            \"name\": \"embeddings\",\n",
    "            \"pytype\": \"float32\",\n",
    "            \"vectorIndex\": {\"dims\": 1536, \"metric\": \"CS\", \"type\": \"flat\"},\n",
    "        },\n",
    "    ]\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "# First ensure the table does not already exist\n",
    "try:\n",
    "    session.table(\"multi_modal_demo\").drop()\n",
    "    time.sleep(5)\n",
    "except kdbai.KDBAIException:\n",
    "    pass"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create the table called \"metadata_demo\"\n",
    "table = session.create_table(\"multi_modal_demo\", table_schema)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 5. Insert data into KDB.AI Vector Database"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 1/1 [00:01<00:00,  1.76s/it]\n"
     ]
    }
   ],
   "source": [
    "#Insert the data into the table, split into 2000 row batches\n",
    "from tqdm import tqdm \n",
    "n = 2000  # chunk row size\n",
    "\n",
    "for i in tqdm(range(0, df.shape[0], n)):\n",
    "    table.insert(df[i:i+n].reset_index(drop=True))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>path</th>\n",
       "      <th>media_type</th>\n",
       "      <th>text</th>\n",
       "      <th>embeddings</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>./data/text/deer.txt</td>\n",
       "      <td>text</td>\n",
       "      <td>Deer are hoofed mammals known for their gracef...</td>\n",
       "      <td>[0.07531881332397461, 0.03122134692966938, 0.0...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>./data/text/caterpillar.txt</td>\n",
       "      <td>text</td>\n",
       "      <td>Caterpillars are the larval stage of butterfli...</td>\n",
       "      <td>[0.05927688628435135, -0.017723916098475456, 0...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>./data/text/bat.txt</td>\n",
       "      <td>text</td>\n",
       "      <td>Bats are the only mammals capable of sustained...</td>\n",
       "      <td>[0.043051157146692276, 0.05017940700054169, 0....</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>./data/text/bear.txt</td>\n",
       "      <td>text</td>\n",
       "      <td>Bears are large mammals with a stocky body, po...</td>\n",
       "      <td>[0.052179791033267975, 0.007267165929079056, 0...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>./data/text/hedgehog.txt</td>\n",
       "      <td>text</td>\n",
       "      <td>Hedgehogs are small, nocturnal mammals known f...</td>\n",
       "      <td>[0.0144755644723773, 0.00036793306935578585, 0...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>./data/text/fox.txt</td>\n",
       "      <td>text</td>\n",
       "      <td>Foxes are small to medium-sized, omnivorous ma...</td>\n",
       "      <td>[0.01701943203806877, 0.017458384856581688, 0....</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>./data/images/caterpillar2.jpg</td>\n",
       "      <td>image</td>\n",
       "      <td>The image shows a hairy caterpillar perched up...</td>\n",
       "      <td>[0.07125178724527359, -0.013986187987029552, 0...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>./data/images/fox1.jpg</td>\n",
       "      <td>image</td>\n",
       "      <td>The image shows a fox seated on a snow-covered...</td>\n",
       "      <td>[0.008546297438442707, -0.029755879193544388, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>./data/images/deer1.jpg</td>\n",
       "      <td>image</td>\n",
       "      <td>The image features a solitary deer standing in...</td>\n",
       "      <td>[0.0711188018321991, -0.029708458110690117, -0...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>./data/images/caterpillar1.jpg</td>\n",
       "      <td>image</td>\n",
       "      <td>The image shows a green caterpillar on a woode...</td>\n",
       "      <td>[0.07201803475618362, -0.004797806032001972, 0...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>./data/images/bat1.jpg</td>\n",
       "      <td>image</td>\n",
       "      <td>The image shows a bat in mid-flight, captured ...</td>\n",
       "      <td>[0.020450666546821594, 0.00680321641266346, 0....</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>./data/images/bat2.jpg</td>\n",
       "      <td>image</td>\n",
       "      <td>The image shows a bat hanging upside down from...</td>\n",
       "      <td>[0.044002678245306015, -0.003725415328517556, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>./data/images/hedgehog1.jpg</td>\n",
       "      <td>image</td>\n",
       "      <td>The image shows a small, adorable hedgehog bei...</td>\n",
       "      <td>[0.06011241674423218, -0.03956378251314163, -0...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>./data/images/deer2.jpg</td>\n",
       "      <td>image</td>\n",
       "      <td>You've provided an image of a white-tailed dee...</td>\n",
       "      <td>[0.07073069363832474, 0.009371200576424599, -0...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>./data/images/hedgehog2.jpg</td>\n",
       "      <td>image</td>\n",
       "      <td>The image shows a hedgehog in a natural enviro...</td>\n",
       "      <td>[0.028540847823023796, -0.02835400030016899, -...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>./data/images/bear1.jpg</td>\n",
       "      <td>image</td>\n",
       "      <td>The image shows a bear walking toward the came...</td>\n",
       "      <td>[0.014313085936009884, -0.018470464274287224, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>./data/images/fox2.jpg</td>\n",
       "      <td>image</td>\n",
       "      <td>The image shows a red fox positioned in front ...</td>\n",
       "      <td>[0.008771030232310295, -0.025080697610974312, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>./data/images/bear2.jpg</td>\n",
       "      <td>image</td>\n",
       "      <td>The image shows a close-up portrait of a brown...</td>\n",
       "      <td>[0.023404834792017937, -0.016726182773709297, ...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                              path media_type  \\\n",
       "0             ./data/text/deer.txt       text   \n",
       "1      ./data/text/caterpillar.txt       text   \n",
       "2              ./data/text/bat.txt       text   \n",
       "3             ./data/text/bear.txt       text   \n",
       "4         ./data/text/hedgehog.txt       text   \n",
       "5              ./data/text/fox.txt       text   \n",
       "6   ./data/images/caterpillar2.jpg      image   \n",
       "7           ./data/images/fox1.jpg      image   \n",
       "8          ./data/images/deer1.jpg      image   \n",
       "9   ./data/images/caterpillar1.jpg      image   \n",
       "10          ./data/images/bat1.jpg      image   \n",
       "11          ./data/images/bat2.jpg      image   \n",
       "12     ./data/images/hedgehog1.jpg      image   \n",
       "13         ./data/images/deer2.jpg      image   \n",
       "14     ./data/images/hedgehog2.jpg      image   \n",
       "15         ./data/images/bear1.jpg      image   \n",
       "16          ./data/images/fox2.jpg      image   \n",
       "17         ./data/images/bear2.jpg      image   \n",
       "\n",
       "                                                 text  \\\n",
       "0   Deer are hoofed mammals known for their gracef...   \n",
       "1   Caterpillars are the larval stage of butterfli...   \n",
       "2   Bats are the only mammals capable of sustained...   \n",
       "3   Bears are large mammals with a stocky body, po...   \n",
       "4   Hedgehogs are small, nocturnal mammals known f...   \n",
       "5   Foxes are small to medium-sized, omnivorous ma...   \n",
       "6   The image shows a hairy caterpillar perched up...   \n",
       "7   The image shows a fox seated on a snow-covered...   \n",
       "8   The image features a solitary deer standing in...   \n",
       "9   The image shows a green caterpillar on a woode...   \n",
       "10  The image shows a bat in mid-flight, captured ...   \n",
       "11  The image shows a bat hanging upside down from...   \n",
       "12  The image shows a small, adorable hedgehog bei...   \n",
       "13  You've provided an image of a white-tailed dee...   \n",
       "14  The image shows a hedgehog in a natural enviro...   \n",
       "15  The image shows a bear walking toward the came...   \n",
       "16  The image shows a red fox positioned in front ...   \n",
       "17  The image shows a close-up portrait of a brown...   \n",
       "\n",
       "                                           embeddings  \n",
       "0   [0.07531881332397461, 0.03122134692966938, 0.0...  \n",
       "1   [0.05927688628435135, -0.017723916098475456, 0...  \n",
       "2   [0.043051157146692276, 0.05017940700054169, 0....  \n",
       "3   [0.052179791033267975, 0.007267165929079056, 0...  \n",
       "4   [0.0144755644723773, 0.00036793306935578585, 0...  \n",
       "5   [0.01701943203806877, 0.017458384856581688, 0....  \n",
       "6   [0.07125178724527359, -0.013986187987029552, 0...  \n",
       "7   [0.008546297438442707, -0.029755879193544388, ...  \n",
       "8   [0.0711188018321991, -0.029708458110690117, -0...  \n",
       "9   [0.07201803475618362, -0.004797806032001972, 0...  \n",
       "10  [0.020450666546821594, 0.00680321641266346, 0....  \n",
       "11  [0.044002678245306015, -0.003725415328517556, ...  \n",
       "12  [0.06011241674423218, -0.03956378251314163, -0...  \n",
       "13  [0.07073069363832474, 0.009371200576424599, -0...  \n",
       "14  [0.028540847823023796, -0.02835400030016899, -...  \n",
       "15  [0.014313085936009884, -0.018470464274287224, ...  \n",
       "16  [0.008771030232310295, -0.025080697610974312, ...  \n",
       "17  [0.023404834792017937, -0.016726182773709297, ...  "
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# See what is in our table - confirm the data was inserted\n",
    "table.query()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 6. Query vector database to retrieve most relevant data \n",
    "Our vector database now contains embeddings representing both text files and images! We can query the vector database to find the most relevant animal based on our query. \n",
    "\n",
    "Retrieved results can then be sent to an LLM for RAG"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "query_vector = [text_to_embedding(\"what is the purpose of having antlers?\")]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[                      path media_type  \\\n",
      "0     ./data/text/deer.txt       text   \n",
      "1  ./data/images/deer2.jpg      image   \n",
      "2  ./data/images/deer1.jpg      image   \n",
      "\n",
      "                                                text  \\\n",
      "0  Deer are hoofed mammals known for their gracef...   \n",
      "1  You've provided an image of a white-tailed dee...   \n",
      "2  The image features a solitary deer standing in...   \n",
      "\n",
      "                                          embeddings  __nn_distance  \n",
      "0  [0.07531881332397461, 0.03122134692966938, 0.0...       0.438880  \n",
      "1  [0.07073069363832474, 0.009371200576424599, -0...       0.418537  \n",
      "2  [0.0711188018321991, -0.029708458110690117, -0...       0.385440  ]\n"
     ]
    }
   ],
   "source": [
    "results = table.search(query_vector, n=3)\n",
    "print(results)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### View the data that was retrieved:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Deer are hoofed mammals known for their graceful bodies and long legs. Most male deer have antlers, which they shed and regrow annually. They are herbivores, generally feeding on a mix of grasses, plants, and leaves. Deer are found in various habitats across the world, including forests, grasslands, and wetlands.\n"
     ]
    },
    {
     "data": {
      "image/jpeg": "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",
      "image/png": "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",
      "text/plain": [
       "<PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=259x194>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/jpeg": "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",
      "image/png": "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",
      "text/plain": [
       "<PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=290x174>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "retrieved_data_for_RAG = []\n",
    "for index, row in results[0].iterrows():\n",
    "    if row[1] == 'image':\n",
    "        retrieved_data_for_RAG.append(row[2])\n",
    "        image = Image.open(row[0])\n",
    "        display(image)\n",
    "    elif row[1] == 'text':\n",
    "        retrieved_data_for_RAG.append(row[2])\n",
    "        print(row[2])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 7. RAG Time!\n",
    "Finally, we can pass our retrieved data and prompt into an LLM -  this is the generation step in RAG."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Helper function to execute RAG, we pass in a list of retrieved data, and the prompt\n",
    "# The function returns the LLM's response\n",
    "\n",
    "def RAG(retrieved_data,prompt):\n",
    "    messages = \"\"\n",
    "    messages += prompt + \"\\n\"\n",
    "    if retrieved_data:\n",
    "        for data in retrieved_data:\n",
    "            messages += data + \"\\n\"\n",
    "    response = openai.chat.completions.create(\n",
    "        model=\"gpt-4-turbo-preview\",\n",
    "        messages=[           \n",
    "            {\n",
    "            \"role\": \"user\",\n",
    "            \"content\": [\n",
    "                {\"type\": \"text\", \"text\": messages},\n",
    "            ],\n",
    "            }\n",
    "        ],\n",
    "        max_tokens=300, \n",
    "    )\n",
    "    content = response.choices[0].message.content\n",
    "    return content\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "prompt = \"what is the purpose of having antlers?\"\n",
    "query = \"You will answer the given prompt using attached content: \"+prompt\n",
    "response = RAG(retrieved_data_for_RAG, query)\n",
    "print(response)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 8. Delete the KDB.AI Table\n",
    "\n",
    "Once finished with the table, it is best practice to drop it."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "table.drop()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Take Our Survey\n",
    "We hope you found this sample helpful! Your feedback is important to us, and we would appreciate it if you could take a moment to fill out our brief survey. Your input helps us improve our content.\n",
    "\n",
    "Take the [Survey](https://delighted.com/t/dfAHRG9R)\n"
   ]
  }
 ],
 "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": "3.10.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}