{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "c1f56ae3-a056-4b31-bcab-27c2c97c00f1",
   "metadata": {},
   "outputs": [],
   "source": [
    "from pathlib import Path\n",
    "\n",
    "from importlib.util import find_spec\n",
    "if find_spec(\"rag\") is None:\n",
    "    import sys\n",
    "    sys.path.append('..')\n",
    "from rag.rag import RAG"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "6b5cb12e-df7e-4532-b78b-216e11ed6161",
   "metadata": {},
   "outputs": [],
   "source": [
    "path = Path(\"/home/aktersnurra/projects/library/quant/math/a-signal-processing-perspective-on-financial-engineering.pdf\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "b8382795-9610-4b24-80b7-31397b2faf90",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\u001b[32m2024-04-06 01:37:11.913\u001b[0m | \u001b[34m\u001b[1mDEBUG   \u001b[0m | \u001b[36mrag.db.document\u001b[0m:\u001b[36m__configure\u001b[0m:\u001b[36m26\u001b[0m - \u001b[34m\u001b[1mCreating documents table if it does not exist...\u001b[0m\n",
      "\u001b[32m2024-04-06 01:37:11.926\u001b[0m | \u001b[34m\u001b[1mDEBUG   \u001b[0m | \u001b[36mrag.db.vector\u001b[0m:\u001b[36m__configure\u001b[0m:\u001b[36m36\u001b[0m - \u001b[34m\u001b[1mCollection knowledge-base already exists...\u001b[0m\n"
     ]
    }
   ],
   "source": [
    "rag = RAG()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "ac57e50d-1fc3-4fc9-90e5-5bdb97bd2f5e",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\u001b[32m2024-04-06 01:37:17.243\u001b[0m | \u001b[34m\u001b[1mDEBUG   \u001b[0m | \u001b[36mrag.db.document\u001b[0m:\u001b[36madd_document\u001b[0m:\u001b[36m37\u001b[0m - \u001b[34m\u001b[1mInserting document hash into documents db...\u001b[0m\n",
      "\u001b[32m2024-04-06 01:37:17.244\u001b[0m | \u001b[34m\u001b[1mDEBUG   \u001b[0m | \u001b[36mrag.db.document\u001b[0m:\u001b[36m__hash\u001b[0m:\u001b[36m32\u001b[0m - \u001b[34m\u001b[1mGenerating sha256 hash for pdf document\u001b[0m\n",
      "\u001b[32m2024-04-06 01:37:17.247\u001b[0m | \u001b[34m\u001b[1mDEBUG   \u001b[0m | \u001b[36mrag.rag\u001b[0m:\u001b[36madd_pdf\u001b[0m:\u001b[36m31\u001b[0m - \u001b[34m\u001b[1mDocument already exists!\u001b[0m\n"
     ]
    }
   ],
   "source": [
    "rag.add_pdf(path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "1c6b48d2-eb04-4a7c-8224-78aabfc7c887",
   "metadata": {},
   "outputs": [],
   "source": [
    "query = \"What is a factor model?\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "a95c8250-00b2-4cbc-a9c6-a76d14ef2da5",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\u001b[32m2024-04-06 01:37:17.265\u001b[0m | \u001b[34m\u001b[1mDEBUG   \u001b[0m | \u001b[36mrag.llm.encoder\u001b[0m:\u001b[36mencode_query\u001b[0m:\u001b[36m33\u001b[0m - \u001b[34m\u001b[1mEncoding query: What is a factor model?\u001b[0m\n",
      "\u001b[32m2024-04-06 01:37:17.858\u001b[0m | \u001b[34m\u001b[1mDEBUG   \u001b[0m | \u001b[36mrag.db.vector\u001b[0m:\u001b[36msearch\u001b[0m:\u001b[36m51\u001b[0m - \u001b[34m\u001b[1mSearching for vectors...\u001b[0m\n",
      "\u001b[32m2024-04-06 01:37:17.864\u001b[0m | \u001b[34m\u001b[1mDEBUG   \u001b[0m | \u001b[36mrag.rag\u001b[0m:\u001b[36m__context\u001b[0m:\u001b[36m35\u001b[0m - \u001b[34m\u001b[1mGot 5 hits in the vector db with limit=5\u001b[0m\n",
      "\u001b[32m2024-04-06 01:37:17.865\u001b[0m | \u001b[34m\u001b[1mDEBUG   \u001b[0m | \u001b[36mrag.llm.generator\u001b[0m:\u001b[36mgenerate\u001b[0m:\u001b[36m32\u001b[0m - \u001b[34m\u001b[1mGenerating answer...\u001b[0m\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "'A factor model is a type of model used to explain the returns or movements of financial assets by decomposing them into two parts: systematic risk, which is driven by a small number of factors affecting many securities; and idiosyncratic risk, which is specific to individual stocks. The general factor model is rt = φ0 + h(ft) + wt, where rt denotes the return of an asset at time t, φ0 represents a constant vector, ft is a vector of factors responsible for most of the randomness in the market, and h is a function that summarizes how these low-dimensional factors affect higher-dimensional markets. The residual wt accounts for any remaining uncorrelated perturbations with only a marginal effect on returns.'"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rag.rag(query, \"quant researcher\", limit=5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "11ac0d46-8589-4700-abf7-7959afbf611c",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.11.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}