{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "# Vector Databases: Chroma, Pinecone & FAISS\n",
        "Xenors practice notebook. Run each cell and compare the output with the tutorial page.\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "## Compare cosine similarity\n",
        "High cosine similarity means similar direction, not necessarily identical semantics.\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "import numpy as np\n",
        "a=np.array([1.,0.,1.]); b=np.array([.8,.1,.9])\n",
        "cos=a@b/(np.linalg.norm(a)*np.linalg.norm(b))\n",
        "print(round(cos,3))\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "**Expected/checkpoint**\n",
        "```text\n~0.994\n```\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "## Apply metadata filtering\n",
        "Metadata filters can improve relevance before vector ranking.\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "records=[('doc1','policy'),('doc2','manual'),('doc3','policy')]\n",
        "print([r for r in records if r[1]=='policy'])\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "**Expected/checkpoint**\n",
        "```text\n[('doc1', 'policy'), ('doc3', 'policy')]\n```\n"
      ]
    }
  ],
  "metadata": {
    "kernelspec": {
      "display_name": "Python 3",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "name": "python",
      "version": "3.x"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 5
}