examples/72_Parsing_the_stars_with_txtai.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Parsing the ⭐'s with txtai\n",
    "\n",
    "[txtai](https://github.com/neuml/txtai) is an all-in-one embeddings database for semantic search, LLM orchestration and language model workflows.\n",
    "\n",
    "In honor of txtai's recent milestone of reaching 10K ⭐'s on GitHub, this notebook will build an astronomical knowledge graph of known stars, planets, galaxies and other astronomical entities. It will show how a compendium of information can be built from a public knowledge source such as Wikipedia.\n",
    "\n",
    "While agents and retrieval augmented generation (RAG) can do great things, they can do even greater things if provided with an assist from the humans in the form of compiled knowledge. It's not only better accuracy, it's also faster as we don't need to wait for an agent to figure out what we already know.\n",
    "\n",
    "Let's dive in."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Install dependencies\n",
    "\n",
    "Install `txtai` and all dependencies."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "%%capture\n",
    "!pip install git+https://github.com/neuml/txtai#egg=txtai[graph,pipeline-text] datasets autoawq autoawq-kernels"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Build the Knowledge Graph\n",
    "\n",
    "The first step is building our knowledge graph or compendium of information. We'll use [txtai-wikipedia](https://hf.co/neuml/txtai-wikipedia) and another source derived from wikipedia with known [stars by constellation](https://huggingface.co/datasets/NeuML/constellations).\n",
    "\n",
    "We'll select articles about planets, constellations, galaxies and stars. We'll then extract entities using a [GLiNER pipeline](https://github.com/urchade/GLiNER). Lastly, we'll join those entries together with the constellation data.\n",
    "\n",
    "To get an idea of what this data will look like, below is the schema for the `constellation` dataset.\n",
    "\n",
    "The following is a description of each of the fields in this dataset. As we can see, there is a lot of identifiers but it also has data such as it's location, brightness and distance from earth.\n",
    "\n",
    "| Field               | Description                                                       |\n",
    "| ------------------- | ----------------------------------------------------------------- | \n",
    "| Constellation       | Name of the constellation the star is a part of                   |\n",
    "| Name                | Proper name                                                       |\n",
    "| Bayer               | Bayer designation                                                 |\n",
    "| Flamsteed           | Flamsteed designation                                             |\n",
    "| Variable Star       | Variable star designation                                         |\n",
    "| Henry Draper Number | Henry Draper Catalogue designation number                         |\n",
    "| Hipparcos Number    | Hipparcos Catalogue designation number                            |\n",
    "| Right Ascension     | Right ascension for the Epoch/Equinox J2000.0                     |\n",
    "| Declination         | Declination for the Epoch/Equinox J2000.0                         |\n",
    "| Visual Magnitude    | Visual Magnitude (m or mv), also known as apparent magnitude      |\n",
    "| Absolute Magnitude  | Absolute Magnitude (Mv)                                           | \n",
    "| Distance            | Distance in light-years from Earth                                |\n",
    "| Spectral class      | Spectral class of the star in the stellar classification system   |\n",
    "| Notes               | Common name(s) or alternate name(s); comments; notable properties |\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from txtai import Embeddings\n",
    "\n",
    "embeddings = Embeddings()\n",
    "embeddings.load(provider=\"huggingface-hub\", container=\"neuml/txtai-wikipedia\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from txtai.pipeline import Entity\n",
    "\n",
    "entity = Entity(\"gliner-community/gliner_medium-v2.5\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "from datasets import load_dataset\n",
    "\n",
    "def parsefloat(value):\n",
    "    try:\n",
    "        return float(value.replace(\"−\", \"-\")) if value else value\n",
    "    except ValueError:\n",
    "        return value\n",
    "\n",
    "def parse(star):\n",
    "    text = star[\"Name\"] if star[\"Name\"] else f'HD-{star[\"Henry Draper Number\"]}'\n",
    "\n",
    "    output = {\"id\": star[\"Name\"], \"type\": \"star\", \"text\": text}\n",
    "    for field, value in star.items():\n",
    "        field = field.lower().replace(\" \", \"_\")\n",
    "        output[field] = parsefloat(value)\n",
    "\n",
    "    return output\n",
    "\n",
    "# Star by constellation dataset\n",
    "stars = load_dataset(\"neuml/constellations\", split=\"train\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The sections above loaded `txtai-wikipedia`, the `GLiNER` pipeline, and the `constellations` dataset. Now it's time to build the knowledge graph!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tqdm.auto import tqdm\n",
    "\n",
    "labels = {\n",
    "    \"name of constellation\": \"constellation\",\n",
    "    \"name of galaxy\": \"galaxy\",\n",
    "    \"name of star in astronomy\": \"star\",\n",
    "    \"name of planet\": \"planet\"\n",
    "}\n",
    "\n",
    "def stream():\n",
    "    # Seed with stars dataset\n",
    "    rows = {star[\"Name\"]: parse(star) for star in stars}\n",
    "\n",
    "    # Get wikipedia results\n",
    "    results = embeddings.search(\n",
    "        \"SELECT id, text, percentile FROM txtai WHERE similar('scientific article defining a constellation, galaxy, star or planet') AND score >= 0.8\",\n",
    "        embeddings.count()\n",
    "    )\n",
    "\n",
    "    # Aggregate into entity rows\n",
    "    for row in tqdm(results):\n",
    "        entities = entity(row[\"text\"], labels=list(labels.keys()))\n",
    "        entities = [(entity, labels[tag]) for entity, tag, score in entities if score >= 0.95 and entity != row[\"id\"]]\n",
    "        if entities or row[\"id\"] in rows:\n",
    "            # Collect relationships    \n",
    "            relationships = set()\n",
    "            for uid, tag in entities:\n",
    "                if uid not in rows:\n",
    "                    rows[uid] = {\"id\": uid, \"type\": tag, \"text\": tag}\n",
    "\n",
    "                relationships.add(uid)\n",
    "\n",
    "            # Add relationships and yield main record\n",
    "            row[\"relationships\"] = list(relationships)\n",
    "\n",
    "            # Merge existing record (if any) and save\n",
    "            rows[row[\"id\"]] = {**rows[row[\"id\"]], **row} if row[\"id\"] in rows else row\n",
    "\n",
    "    yield from rows.values()\n",
    "\n",
    "astronomy = Embeddings(\n",
    "    path = \"intfloat/e5-base\",\n",
    "    instructions={\"query\": \"query: \", \"data\": \"passage: \"},\n",
    "    faiss = {\"quantize\": True, \"sample\": 0.05},\n",
    "    content = True,\n",
    "    graph = {\"approximate\": True, \"topics\": {}, \"copyattributes\": True}\n",
    ")\n",
    "astronomy.index(stream())\n",
    "astronomy.save(\"txtai-astronomy\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Explore the Knowledge Graph\n",
    "\n",
    "The next section creates a function to plot a knowledge graph. This function is specific to our astronomical data. It color codes galaxies, planets and stars by [spectral class](https://en.wikipedia.org/wiki/Stellar_classification).\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import networkx as nx\n",
    "\n",
    "def plot(graph):\n",
    "    labels = {x: f\"{graph.attribute(x, 'id')}\" for x in graph.scan()}\n",
    "    mapping = {\n",
    "        \"constellation\": \"#f44336\", \"galaxy\": \"#9575cd\", \"planet\": \"#9e9e9e\", \"star\": \"#ff9800\",\n",
    "        \"O\": \"#92b5ff\", \"B\": \"#a2c0ff\", \"A\": \"#d5e0ff\", \"F\": \"#f9f5ff\", \"G\": \"#ffede3\",\n",
    "        \"K\": \"#ffdab5\", \"M\": \"#ffb56c\"\n",
    "    }\n",
    "\n",
    "    colors = []\n",
    "    for x in graph.scan():\n",
    "        etype, spectral  = graph.attribute(x, \"type\"), graph.attribute(x, \"spectral_class\")\n",
    "        if etype == \"star\":\n",
    "            etype = spectral[0] if spectral else etype\n",
    "        \n",
    "        colors.append(mapping.get(etype, \"#03a9f4\"))\n",
    "\n",
    "    options = {\n",
    "        \"node_size\": 700,\n",
    "        \"node_color\": colors,\n",
    "        \"edge_color\": \"#454545\",\n",
    "        \"alpha\": 1.0,\n",
    "    }\n",
    "\n",
    "    # Draw graph\n",
    "    fig, ax = plt.subplots(figsize=(20, 9))\n",
    "    pos = nx.spring_layout(graph.backend, seed=0, k=0.9, iterations=50)\n",
    "    nx.draw_networkx(graph.backend, pos=pos, with_labels=False, **options)\n",
    "\n",
    "    # Calculate label positions\n",
    "    pos = {node: (x, y - 0.1) for node, (x, y) in pos.items()}\n",
    "\n",
    "    nx.draw_networkx_labels(\n",
    "        graph.backend, pos, labels=labels, font_color=\"#efefef\", font_size=10\n",
    "    )\n",
    "\n",
    "    # Disable axes and draw margins\n",
    "    ax.axis(\"off\")\n",
    "    plt.margins(x=0.15)\n",
    "\n",
    "    # Set background color\n",
    "    ax.set_facecolor(\"#303030\")\n",
    "    fig.set_facecolor(\"#303030\")\n",
    "\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we're ready to run a query. `txtai` supports vector queries, SQL queries and Graph queries (using [openCypher](https://github.com/aplbrain/grand-cypher)). As of txtai 8.3, queries are automatically routed to the correct index based on type. \n",
    "\n",
    "The path traversal query below takes a starting node and plots the 10 relationships that are closest to earth."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x900 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "subgraph = astronomy.search(\"\"\"\n",
    "MATCH P=({id: \"Andromeda\"})-[]->(A)\n",
    "WHERE A.distance > 0\n",
    "RETURN P \n",
    "ORDER BY A.distance\n",
    "LIMIT 10\n",
    "\"\"\", graph=True)\n",
    "\n",
    "plot(subgraph)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can also find stars with planets."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x900 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WASP-21b planet\n",
      "WASP-21 WASP-21 is a G-type star (spectral type G3V) that is reaching the end of its main sequence lifetime  approximately 850 light years from Earth in the constellation of Pegasus. The star is relatively metal-poor, having 40% of heavy elements compared to the Sun. Kinematically, WASP-21 belongs to the thick disk of the Milky Way. It has an exoplanet named WASP-21b.\n",
      "WASP-13 WASP-13, also named Gloas, is a star in the Lynx constellation. The star is similar, in terms of metallicity and mass, to the Sun, although it is hotter and most likely older. The star was first observed in 1997, according to the SIMBAD database, and was targeted by SuperWASP after the star was observed by one of the SuperWASP telescopes beginning in 2006. Follow-up observations on the star led to the discovery of planet Cruinlagh in 2008; the discovery paper was published in 2009.\n",
      "Cruinlagh planet\n",
      "Gamma Cephei Gamma Cephei (γ Cephei, abbreviated Gamma Cep, γ Cep) is a binary star system approximately 45 light-years away in the northern constellation of Cepheus. The primary (designated Gamma Cephei A, officially named Errai , the traditional name of the system) is a stellar class K1 orange giant or subgiant star; it has a red dwarf companion (Gamma Cephei B). An exoplanet (designated Gamma Cephei Ab, later named Tadmor) has been confirmed to be orbiting the primary.\n",
      "Tadmor planet\n"
     ]
    }
   ],
   "source": [
    "subgraph = astronomy.search(\"\"\"\n",
    "MATCH P=(A {type: \"star\"})-[]->(B {type: \"planet\"})\n",
    "WHERE B.id <> \"Jupiter\" AND B.id <> \"Saturn\"\n",
    "RETURN P\n",
    "\"\"\", graph=True)\n",
    "\n",
    "plot(subgraph)\n",
    "\n",
    "for x in subgraph.scan():\n",
    "    print(subgraph.attribute(x, \"id\"), subgraph.attribute(x, \"text\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Standard SQL queries are also supported. Let's show the stars that are brightest from Earth."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[{'id': 'Sirius',\n",
       "  'text': \"Sirius is the brightest star in the night sky. Its name is derived from the Greek word  (Latin script: ), meaning  'glowing' or 'scorching'. The star is designated \\xa0Canis Majoris, Latinized to Alpha Canis Majoris, and abbreviated \\xa0CMa or Alpha\\xa0CMa. With a visual apparent magnitude of −1.46, Sirius is almost twice as bright as Canopus, the next brightest star. Sirius is a binary star consisting of a main-sequence star of spectral type A0 or A1, termed Sirius\\xa0A, and a faint white dwarf companion of spectral type\\xa0DA2, termed Sirius\\xa0B. The distance between the two varies between 8.2 and 31.5\\xa0astronomical units as they orbit every 50\\xa0years.\",\n",
       "  'visual_magnitude': -1.46},\n",
       " {'id': 'Canopus',\n",
       "  'text': 'Canopus is the brightest star in the southern constellation of Carina and the second-brightest star in the night sky. It is also designated α\\xa0Carinae, which is romanized (transliterated) to Alpha\\xa0Carinae. With a visual apparent magnitude of −0.74, it is outshone only by Sirius.',\n",
       "  'visual_magnitude': -0.72},\n",
       " {'id': 'Arcturus',\n",
       "  'text': 'Arcturus is the brightest star in the northern constellation of Boötes. With an apparent visual magnitude of −0.05, it is the fourth-brightest  star in the night sky, and the brightest in the northern celestial hemisphere. The name Arcturus originated from ancient Greece; it was then cataloged as α\\xa0Boötis by Johann Bayer in 1603, which is Latinized to Alpha\\xa0Boötis. Arcturus forms one corner of the Spring Triangle asterism.',\n",
       "  'visual_magnitude': -0.05},\n",
       " {'id': 'Vega',\n",
       "  'text': \"Vega is the brightest star in the northern constellation of Lyra. It has the Bayer designation α Lyrae, which is Latinised to Alpha Lyrae and abbreviated Alpha Lyr or α Lyr. This star is relatively close at only  from the Sun, and one of the most luminous stars in the Sun's neighborhood. It is the fifth-brightest star in the night sky, and the second-brightest star in the northern celestial hemisphere, after Arcturus.\",\n",
       "  'visual_magnitude': 0.03},\n",
       " {'id': 'Capella', 'text': 'Capella', 'visual_magnitude': 0.08}]"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "astronomy.search(\"\"\"\n",
    "SELECT id, text, visual_magnitude FROM txtai\n",
    "WHERE visual_magnitude IS NOT NULL\n",
    "ORDER BY visual_magnitude\n",
    "LIMIT 5\n",
    "\"\"\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Retrieval Augmented Generation (RAG)\n",
    "\n",
    "Next, we'll run a series of RAG queries using the knowledge graph."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from txtai import LLM, RAG\n",
    "\n",
    "# Define prompt template\n",
    "template = \"\"\"\n",
    "Answer the following question using only the context below. Only include information\n",
    "specifically discussed. Answer the question without explaining how you found the answer.\n",
    "\n",
    "question: {question}\n",
    "context: {context}\"\"\"\n",
    "\n",
    "# Load LLM\n",
    "llm = LLM(\"hugging-quants/Meta-Llama-3.1-8B-Instruct-AWQ-INT4\")\n",
    "\n",
    "# Create RAG pipeline\n",
    "rag = RAG(\n",
    "    astronomy,\n",
    "    llm, \n",
    "    system=\"You are a friendly assistant. You answer questions from users.\",\n",
    "    template=template,\n",
    "    context=10\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The stars in the Andromeda constellation mentioned in the context include:\n",
      "\n",
      "1. Omicron Andromedae (ο And), a blue-white B-type giant approximately 692 light years from Earth.\n",
      "2. RT Andromedae, a variable star approximately 322 light years from Earth.\n",
      "3. 8 Andromedae, a probable triple star system approximately 570 light years from Earth.\n",
      "4. 64 Andromedae, a deep-yellow coloured G-type giant approximately 419 light years from Earth.\n",
      "5. 3 Andromedae, a single star approximately 181 light years from Earth.\n",
      "6. Epsilon Andromedae, a star approximately 155 light years from the Sun.\n",
      "7. HD 166 or V439 Andromedae, a variable star approximately 45 light years away from Earth.\n",
      "8. 7 Andromedae, a single, yellow-white hued star approximately 79.6 light years from Earth.\n",
      "9. 18 Andromedae, a single star approximately 413 light years from Earth.\n",
      "10. 12 Andromedae, a single star approximately 137 light years from Earth.\n",
      "\n",
      "These stars vary in their distance from Earth, ranging from 45 to 692 light years.\n"
     ]
    }
   ],
   "source": [
    "print(rag(\"Tell me about the stars in the Andromeda constellation and how far they are from Earth\", maxlength=4096)[\"answer\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now let's try a Graph RAG query. Instead of vector search, we'll run a graph traversal query to generate the RAG context."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x900 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Here's a summary of the stars discussed along with a short description about them, sorted by distance to Earth:\n",
      "\n",
      "1. **Proxima Centauri**: The closest star to Earth after the Sun, located 4.25 light-years away. It's a small, low-mass star too faint to be seen with the naked eye.\n",
      "\n",
      "2. **Alpha Centauri (Rigil Kentaurus, Toliman, and Proxima Centauri)**: A triple star system located 4.25 light-years away. It consists of three stars: Rigil Kentaurus, Toliman, and Proxima Centauri.\n",
      "\n",
      "3. **Iota Centauri**: A star located approximately 51.5 light-years away. It has an apparent visual magnitude of +2.73, making it easily visible to the naked eye.\n",
      "\n",
      "4. **HD 113538 (Gliese 496.1)**: A star with two planetary companions located 53 light-years away. It is much too faint to be viewed with the naked eye.\n",
      "\n",
      "5. **HD 101930 (Gliese 3683)**: An orange-hued star with an orbiting exoplanet located 98 light-years away. It has a relatively large proper motion and is receding with a radial velocity of.\n",
      "\n",
      "6. **HD 114386**: A star with a pair of orbiting exoplanets located 91 light-years away. It has an apparent visual magnitude of 8.73, which means it cannot be viewed with the naked eye but can be seen with a telescope or good binoculars.\n",
      "\n",
      "7. **HD 125595**: A star with a close Neptunian companion located 92 light-years away. It is too faint to be viewed with the naked eye.\n",
      "\n",
      "8. **BPM 37093 (V886 Centauri)**: A variable white dwarf star of the DAV, or ZZ Ceti, type, located approximately  light-years away. It has a hydrogen atmosphere and an unusually high mass of approximately 1.1 times the Sun's.\n"
     ]
    }
   ],
   "source": [
    "subgraph = astronomy.search(\"\"\"\n",
    "MATCH P=({id: \"Centaurus\"})-[]->(B)\n",
    "WHERE B.distance > 0\n",
    "RETURN P\n",
    "ORDER BY B.distance\n",
    "LIMIT 10\n",
    "\"\"\", graph=True)\n",
    "\n",
    "plot(subgraph)\n",
    "\n",
    "command = \"\"\"\n",
    "Write a summary of the stars discussed along with a short description about them.\n",
    "Sort by distance to Earth.\n",
    "\"\"\"\n",
    "\n",
    "context = [subgraph.attribute(x, \"text\") for x in subgraph.scan()]\n",
    "print(rag(command, context, maxlength=4096)[\"answer\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Agents\n",
    "\n",
    "Our last example is an Agent that searches the astronomy knowledge graph."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from txtai import Agent\n",
    "\n",
    "def search(query):\n",
    "    \"\"\"\n",
    "    Searches a database of astronomy data.\n",
    "\n",
    "    Make sure to call this tool only with a string input, never use JSON.    \n",
    "\n",
    "    Args:\n",
    "        query: concepts to search for using similarity search\n",
    " \n",
    "    Returns:\n",
    "        list of search results with for each match\n",
    "    \"\"\"\n",
    "\n",
    "    return embeddings.search(\n",
    "        \"SELECT id, text, distance FROM txtai WHERE similar(:query)\",\n",
    "        10, parameters={\"query\": query}\n",
    "    )\n",
    "\n",
    "agent = Agent(\n",
    "    tools=[search],\n",
    "    llm=llm,\n",
    "    max_iterations=10,\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\u001b[32;20;1m======== New task ========\u001b[0m\n",
      "\u001b[37;1m\n",
      "\n",
      "Write a detailed list with explanations of 10 candidate stars that could potentially be habitable to life.\n",
      "\n",
      "\n",
      "Do the following.\n",
      " - Search for results related to the topic.\n",
      " - Analyze the results\n",
      " - Continue querying until conclusive answers are found\n",
      " - Write a Markdown report\n",
      "\u001b[0m\n",
      "\u001b[33;1m=== Agent thoughts:\u001b[0m\n",
      "\u001b[0mThought: I will need to find the list of candidate stars that could potentially be habitable to life. To do this, I will use the `search` tool to look for related results.\u001b[0m\n",
      "\u001b[33;1m>>> Calling tool: 'search' with arguments: candidate stars for habitable planets\u001b[0m\n",
      "\u001b[33;1m=== Agent thoughts:\u001b[0m\n",
      "\u001b[0mThought: The results from the `search` tool provide a list of candidate stars for habitable planets. However, they are not organized in a clear and concise manner. To create a detailed list with explanations, I will need to analyze the results and identify the key points.\u001b[0m\n",
      "\u001b[33;1m>>> Calling tool: 'final_answer' with arguments: {'answer': \"Here is a list of 10 candidate stars that could potentially be habitable to life:\\n\\n1. Luyten's Star (red dwarf star) - has a confirmed exoplanet, Luyten b, which is a super-Earth and receives only 6% more starlight than Earth.\\n\\n2. Gliese 273 (red dwarf star) - has a confirmed exoplanet, Luyten b, which is a super-Earth and receives only 6% more starlight than Earth.\\n\\n3. Ross 128 (red dwarf star) - has a confirmed exoplanet, Ross 128 b, which is a super-Earth and receives only 6% more starlight than Earth.\\n\\n4. Proxima Centauri (red dwarf star) - has a confirmed exoplanet, Proxima Centauri b, which is a super-Earth and receives only 6% more starlight than Earth.\\n\\n5. Kepler-560 (binary star system) - has a confirmed exoplanet, Kepler-560b, which is a super-Earth and orbits within the habitable zone of the binary star system.\\n\\n6. K2-332 (M4V star) - has a confirmed exoplanet, K2-332 b, which is a potentially habitable Super-Earth or Mini-Neptune exoplanet and receives 1.17 times the light that Earth gets from the sun.\\n\\n7. Alpha Centauri A (G-type main-sequence star) - has a candidate exoplanet, Alpha Centauri Ab, which is a giant planet and orbits at approximately 1.1 AU away from the star.\\n\\n8. Kepler-298 (orange dwarf star) - has a confirmed exoplanet, Kepler-298d, which orbits within the habitable zone of the star and may be an ocean planet with a thick gas atmosphere.\\n\\n9. LHS 1140 (M-dwarf star) - has a confirmed exoplanet, LHS 1140 b, which is a super-Earth and orbits within the habitable zone of the star.\\n\\n10. Teegarden's Star (M-dwarf star) - has a confirmed exoplanet, Teegarden's Star b, which is a super-Earth and orbits within the habitable zone of the star.\\n\\nNote: The habitability of these stars and their exoplanets is still a topic of debate and research, and more studies are needed to confirm their potential for supporting life.\"}\u001b[0m\n"
     ]
    },
    {
     "data": {
      "text/markdown": [
       "Here is a list of 10 candidate stars that could potentially be habitable to life:\n",
       "\n",
       "1. Luyten's Star (red dwarf star) - has a confirmed exoplanet, Luyten b, which is a super-Earth and receives only 6% more starlight than Earth.\n",
       "\n",
       "2. Gliese 273 (red dwarf star) - has a confirmed exoplanet, Luyten b, which is a super-Earth and receives only 6% more starlight than Earth.\n",
       "\n",
       "3. Ross 128 (red dwarf star) - has a confirmed exoplanet, Ross 128 b, which is a super-Earth and receives only 6% more starlight than Earth.\n",
       "\n",
       "4. Proxima Centauri (red dwarf star) - has a confirmed exoplanet, Proxima Centauri b, which is a super-Earth and receives only 6% more starlight than Earth.\n",
       "\n",
       "5. Kepler-560 (binary star system) - has a confirmed exoplanet, Kepler-560b, which is a super-Earth and orbits within the habitable zone of the binary star system.\n",
       "\n",
       "6. K2-332 (M4V star) - has a confirmed exoplanet, K2-332 b, which is a potentially habitable Super-Earth or Mini-Neptune exoplanet and receives 1.17 times the light that Earth gets from the sun.\n",
       "\n",
       "7. Alpha Centauri A (G-type main-sequence star) - has a candidate exoplanet, Alpha Centauri Ab, which is a giant planet and orbits at approximately 1.1 AU away from the star.\n",
       "\n",
       "8. Kepler-298 (orange dwarf star) - has a confirmed exoplanet, Kepler-298d, which orbits within the habitable zone of the star and may be an ocean planet with a thick gas atmosphere.\n",
       "\n",
       "9. LHS 1140 (M-dwarf star) - has a confirmed exoplanet, LHS 1140 b, which is a super-Earth and orbits within the habitable zone of the star.\n",
       "\n",
       "10. Teegarden's Star (M-dwarf star) - has a confirmed exoplanet, Teegarden's Star b, which is a super-Earth and orbits within the habitable zone of the star.\n",
       "\n",
       "Note: The habitability of these stars and their exoplanets is still a topic of debate and research, and more studies are needed to confirm their potential for supporting life."
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from IPython.display import display, Markdown\n",
    "\n",
    "def md(output):\n",
    "    display(Markdown(output))\n",
    "\n",
    "researcher = \"\"\"\n",
    "{command}\n",
    "\n",
    "Do the following.\n",
    " - Search for results related to the topic.\n",
    " - Analyze the results\n",
    " - Continue querying until conclusive answers are found\n",
    " - Write a Markdown report\n",
    "\"\"\"\n",
    "\n",
    "md(agent(researcher.format(command=\"\"\"\n",
    "Write a detailed list with explanations of 10 candidate stars that could potentially be habitable to life.\n",
    "\"\"\"), maxlength=16000))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Wrapping up\n",
    "\n",
    "This notebook parsed the ⭐'s with txtai. It showed how knowledge graphs can be built to drive RAG. We also covered how to use agents to build an \"agentic\" or \"self-querying\" RAG system.\n",
    "\n",
    "Are we alone in the universe? No one knows for sure. Perhaps AI will figure it out for us 😀\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "local",
   "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": "3.9.21"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}