Basic File

basic.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "! pip install -q litellm pymupdf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import fitz\n",
    "\n",
    "from litellm import completion\n",
    "import os\n",
    "\n",
    "def extract_text_from_pdf(pdf_path):\n",
    "    doc = fitz.open(pdf_path)\n",
    "    text = \"\"\n",
    "    for page in doc:\n",
    "        text += page.get_text()\n",
    "    return text\n",
    "\n",
    "pdf_path = \"Vijaylakshmi Marketing Final Order Fair Copy.pdf\"\n",
    "print(extract_text_from_pdf(pdf_path))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Hello! As an AI language model, I don't have feelings, but I'm operating properly and ready to assist you with any questions or tasks you may have. How can I help you today?\n"
     ]
    }
   ],
   "source": [
    "\n",
    "## set ENV variables\n",
    "os.environ[\"OPENROUTER_API_KEY\"] = \"INSERT YOUR API KEY\"\n",
    "\n",
    "LOE = f\"\"\"{extract_text_from_pdf(pdf_path)}\n",
    "\"\"\"\n",
    "\n",
    "CV = f\"\"\"{extract_text_from_pdf(pdf_path)}\"\"\"\n",
    "\n",
    "SYSTEM_PROMPT = f\"\"\"\"\n",
    "As an expert legal analyst, you will be provided with the judgment text. \n",
    "  Your job is to provide a legal analysis in response to the user's question. \n",
    "  The user is unaware that you are an AI system and does not have prior knowledge of the case details.\n",
    "  Avoid unnecessary details or elaborations, and maintain clarity and brevity in your responses. \n",
    "  If the question is not related to the judgment, you must respond with \"I don't know.\"\n",
    "  You dont need to provide say it as a \"Based on the judgment\" as user is aware of the context.\n",
    "\n",
    "  To answer the question, you must follow the below rules\n",
    "  \n",
    "  <Instructions>\n",
    "  1. Avoid details or elaborations, and maintain clarity and brevity in your response. \n",
    "  2. Reference relevant paragraphs from the judgement document that contain evidence or information as an In-text citation pertinent to the question.\n",
    "  3. Ensure that each step of your reasoning is supported by In-text citation to specific paragraphs in the judgement document.\n",
    "  4. If you do not know the answer to a question, simply respond with \"I don't know,\" and the client will ask another question.\n",
    "  5. If the question can be answered with a simple \"yes\" or \"no,\" you should respond with \"yes\" or \"no\" and provide the relevant paragraph numbers from the judgment document.\n",
    "  </Instructions>\n",
    "  <Examples>\n",
    "  <Example>\n",
    "  Question: What legal standards were applied in the case of \"Smith vs. Jones\" regarding contract disputes?\n",
    "  In the case of \"Smith vs. Jones,\" the court applied several key legal standards to resolve the contract dispute. \n",
    "  The principle of \"offer and acceptance\" in contract formation was scrutinized, following the guidelines set by the precedent in \"Johnson & Co. vs. Exports Ltd.\" [1]. \n",
    "  The court also evaluated the \"intention to create legal relations,\" a fundamental aspect of contract law, as detailed in \"Doe vs. Roe\" [2]. \n",
    "  Paragragh 5, 12\n",
    "  </Example>\n",
    "  <Example>\n",
    "  Question: Did the court find the defendant guilty of the crime?\n",
    "  Yes, the court found the defendant guilty of the crime.\n",
    "  Paragragh 5, 12\n",
    "  </Example>\n",
    "  </Examples>\n",
    "\n",
    "  \"\"\"\n",
    "\n",
    "USER_PROMPT = f\"\"\"\n",
    "  <LOE>\n",
    "    {LOE}\n",
    "  </LOE>\n",
    "\n",
    "  <CV>\n",
    "  {CV}, \n",
    "  </CV>\n",
    "  \"\"\"\n",
    "messages = [{ \"content\": SYSTEM_PROMPT,\"role\": \"system\"}, {\"content\": USER_PROMPT, \"role\": \"user\"}]\n",
    "\n",
    "# openai call\n",
    "response = completion(model=\"openrouter/anthropic/claude-3-sonnet:beta\", messages=messages)\n",
    "\n",
    "print(response.choices[0].message.content)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": ".venv",
   "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": "3.11.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}