10_minutes_to_PandasAI.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/jonbaer/googlecolab/blob/master/10_minutes_to_PandasAI.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# PandasAI\n",
        "PandasAI is a library that makes data analysis conversational and fun again. It leverages the power of pandas dataframes combined to the most advanced LLMs to let users to data analysis in a conversational way.\n",
        "\n",
        "Similarly to what `pandas` has done (10 minutes to pandas -> https://pandas.pydata.org/docs/user_guide/10min.html), we wanted to create the most straightforward way to learn how to master PandasAI.\n",
        "\n",
        "Let's start!\n"
      ],
      "metadata": {
        "id": "5m9QRh2ofb0M"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Setup\n",
        "To get started, we need to install the last version of PandasAI."
      ],
      "metadata": {
        "id": "8duxIdoChr2g"
      }
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "L9szBbbJe30_",
        "outputId": "78331162-5b10-4b27-a334-ed4ad491cf0d"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Requirement already satisfied: pandasai in /usr/local/lib/python3.10/dist-packages (2.0.16)\n",
            "Requirement already satisfied: astor<0.9.0,>=0.8.1 in /usr/local/lib/python3.10/dist-packages (from pandasai) (0.8.1)\n",
            "Requirement already satisfied: duckdb<1 in /usr/local/lib/python3.10/dist-packages (from pandasai) (0.9.2)\n",
            "Requirement already satisfied: faker<20.0.0,>=19.12.0 in /usr/local/lib/python3.10/dist-packages (from pandasai) (19.13.0)\n",
            "Requirement already satisfied: jinja2<4.0.0,>=3.1.3 in /usr/local/lib/python3.10/dist-packages (from pandasai) (3.1.3)\n",
            "Requirement already satisfied: matplotlib<4.0.0,>=3.7.1 in /usr/local/lib/python3.10/dist-packages (from pandasai) (3.7.1)\n",
            "Requirement already satisfied: openai<2 in /usr/local/lib/python3.10/dist-packages (from pandasai) (1.14.1)\n",
            "Requirement already satisfied: pandas==1.5.3 in /usr/local/lib/python3.10/dist-packages (from pandasai) (1.5.3)\n",
            "Requirement already satisfied: pillow<11.0.0,>=10.1.0 in /usr/local/lib/python3.10/dist-packages (from pandasai) (10.2.0)\n",
            "Requirement already satisfied: pydantic<3,>=1 in /usr/local/lib/python3.10/dist-packages (from pandasai) (2.6.4)\n",
            "Requirement already satisfied: python-dotenv<2.0.0,>=1.0.0 in /usr/local/lib/python3.10/dist-packages (from pandasai) (1.0.1)\n",
            "Requirement already satisfied: requests<3.0.0,>=2.31.0 in /usr/local/lib/python3.10/dist-packages (from pandasai) (2.31.0)\n",
            "Requirement already satisfied: scipy<2.0.0,>=1.9.0 in /usr/local/lib/python3.10/dist-packages (from pandasai) (1.11.4)\n",
            "Requirement already satisfied: sqlalchemy<3,>=1.4 in /usr/local/lib/python3.10/dist-packages (from pandasai) (2.0.28)\n",
            "Requirement already satisfied: python-dateutil>=2.8.1 in /usr/local/lib/python3.10/dist-packages (from pandas==1.5.3->pandasai) (2.8.2)\n",
            "Requirement already satisfied: pytz>=2020.1 in /usr/local/lib/python3.10/dist-packages (from pandas==1.5.3->pandasai) (2023.4)\n",
            "Requirement already satisfied: numpy>=1.21.0 in /usr/local/lib/python3.10/dist-packages (from pandas==1.5.3->pandasai) (1.25.2)\n",
            "Requirement already satisfied: MarkupSafe>=2.0 in /usr/local/lib/python3.10/dist-packages (from jinja2<4.0.0,>=3.1.3->pandasai) (2.1.5)\n",
            "Requirement already satisfied: contourpy>=1.0.1 in /usr/local/lib/python3.10/dist-packages (from matplotlib<4.0.0,>=3.7.1->pandasai) (1.2.0)\n",
            "Requirement already satisfied: cycler>=0.10 in /usr/local/lib/python3.10/dist-packages (from matplotlib<4.0.0,>=3.7.1->pandasai) (0.12.1)\n",
            "Requirement already satisfied: fonttools>=4.22.0 in /usr/local/lib/python3.10/dist-packages (from matplotlib<4.0.0,>=3.7.1->pandasai) (4.49.0)\n",
            "Requirement already satisfied: kiwisolver>=1.0.1 in /usr/local/lib/python3.10/dist-packages (from matplotlib<4.0.0,>=3.7.1->pandasai) (1.4.5)\n",
            "Requirement already satisfied: packaging>=20.0 in /usr/local/lib/python3.10/dist-packages (from matplotlib<4.0.0,>=3.7.1->pandasai) (24.0)\n",
            "Requirement already satisfied: pyparsing>=2.3.1 in /usr/local/lib/python3.10/dist-packages (from matplotlib<4.0.0,>=3.7.1->pandasai) (3.1.2)\n",
            "Requirement already satisfied: anyio<5,>=3.5.0 in /usr/local/lib/python3.10/dist-packages (from openai<2->pandasai) (3.7.1)\n",
            "Requirement already satisfied: distro<2,>=1.7.0 in /usr/lib/python3/dist-packages (from openai<2->pandasai) (1.7.0)\n",
            "Requirement already satisfied: httpx<1,>=0.23.0 in /usr/local/lib/python3.10/dist-packages (from openai<2->pandasai) (0.27.0)\n",
            "Requirement already satisfied: sniffio in /usr/local/lib/python3.10/dist-packages (from openai<2->pandasai) (1.3.1)\n",
            "Requirement already satisfied: tqdm>4 in /usr/local/lib/python3.10/dist-packages (from openai<2->pandasai) (4.66.2)\n",
            "Requirement already satisfied: typing-extensions<5,>=4.7 in /usr/local/lib/python3.10/dist-packages (from openai<2->pandasai) (4.10.0)\n",
            "Requirement already satisfied: annotated-types>=0.4.0 in /usr/local/lib/python3.10/dist-packages (from pydantic<3,>=1->pandasai) (0.6.0)\n",
            "Requirement already satisfied: pydantic-core==2.16.3 in /usr/local/lib/python3.10/dist-packages (from pydantic<3,>=1->pandasai) (2.16.3)\n",
            "Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.10/dist-packages (from requests<3.0.0,>=2.31.0->pandasai) (3.3.2)\n",
            "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.10/dist-packages (from requests<3.0.0,>=2.31.0->pandasai) (3.6)\n",
            "Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.10/dist-packages (from requests<3.0.0,>=2.31.0->pandasai) (2.0.7)\n",
            "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.10/dist-packages (from requests<3.0.0,>=2.31.0->pandasai) (2024.2.2)\n",
            "Requirement already satisfied: greenlet!=0.4.17 in /usr/local/lib/python3.10/dist-packages (from sqlalchemy<3,>=1.4->pandasai) (3.0.3)\n",
            "Requirement already satisfied: exceptiongroup in /usr/local/lib/python3.10/dist-packages (from anyio<5,>=3.5.0->openai<2->pandasai) (1.2.0)\n",
            "Requirement already satisfied: httpcore==1.* in /usr/local/lib/python3.10/dist-packages (from httpx<1,>=0.23.0->openai<2->pandasai) (1.0.4)\n",
            "Requirement already satisfied: h11<0.15,>=0.13 in /usr/local/lib/python3.10/dist-packages (from httpcore==1.*->httpx<1,>=0.23.0->openai<2->pandasai) (0.14.0)\n",
            "Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.10/dist-packages (from python-dateutil>=2.8.1->pandas==1.5.3->pandasai) (1.16.0)\n"
          ]
        }
      ],
      "source": [
        "!pip install pandasai"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# SmartDataframe\n",
        "\n",
        "A SmartDataframe is a pandas (or polars) dataframe that inherits all the properties and methods from the `pd.DataFrame`, but also adds conversational features to it."
      ],
      "metadata": {
        "id": "EkSwZhHqfIzg"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from pandasai import SmartDataframe"
      ],
      "metadata": {
        "id": "DKTLrefOgGYj"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "You can instantiate a dataframe importing from several different sources (pandas or polars dataframe, csv, xlsx or google sheets)."
      ],
      "metadata": {
        "id": "7NkkotuPf1Tw"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Importing from a pandas dataframe"
      ],
      "metadata": {
        "id": "LJdaiLU9gA4g"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "To import from a pandas dataframe, you need to import the pandas libraries and create a dataframe first."
      ],
      "metadata": {
        "id": "viepQXfkglJY"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import pandas as pd"
      ],
      "metadata": {
        "id": "LOj6QLWGLMlD"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df = pd.DataFrame({\n",
        "    \"country\": [\n",
        "        \"United States\",\n",
        "        \"United Kingdom\",\n",
        "        \"France\",\n",
        "        \"Germany\",\n",
        "        \"Italy\",\n",
        "        \"Spain\",\n",
        "        \"Canada\",\n",
        "        \"Australia\",\n",
        "        \"Japan\",\n",
        "        \"China\",\n",
        "    ],\n",
        "    \"gdp\": [\n",
        "        19294482071552,\n",
        "        2891615567872,\n",
        "        2411255037952,\n",
        "        3435817336832,\n",
        "        1745433788416,\n",
        "        1181205135360,\n",
        "        1607402389504,\n",
        "        1490967855104,\n",
        "        4380756541440,\n",
        "        14631844184064,\n",
        "    ],\n",
        "    \"happiness_index\": [6.94, 7.16, 6.66, 7.07, 6.38, 6.4, 7.23, 7.22, 5.87, 5.12],\n",
        "})"
      ],
      "metadata": {
        "id": "ugogHbQxf0Y6"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Since PandasAI is powered by a LLM, you should import the LLM you'd like to use for your use case. In thic case we're gonna use OpenAI.\n",
        "\n",
        "To use OpenAI, you'll need a API token. Follow below simple steps to generate your API_TOKEN with\n",
        "[openai](https://platform.openai.com/overview).\n",
        "\n",
        "1. Go to  https://openai.com/api/ and signup with your email address or connect your Google Account.\n",
        "2. Go to View API Keys on left side of your Personal Account Settings\n",
        "3. Select Create new Secret key\n",
        "\n",
        "> The API access to openai is a paid service. You have to set up billing.\n",
        ">Read the [Pricing](https://platform.openai.com/docs/quickstart/pricing) information before experimenting."
      ],
      "metadata": {
        "id": "2bD7r0EtgsdD"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from pandasai.llm import OpenAI\n",
        "\n",
        "llm = OpenAI(api_token=\"YOUR TOKEN\")"
      ],
      "metadata": {
        "id": "rB64X01YgrmH"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Now that we have instantiated the LLM, we can finally instantiate the `SmartDataframe`"
      ],
      "metadata": {
        "id": "TlhiAynHiQ0y"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "sdf = SmartDataframe(df, config={\"llm\": llm})"
      ],
      "metadata": {
        "id": "hXFeO02JiYRC"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "You can now query it in natural language"
      ],
      "metadata": {
        "id": "SEsxIYyTjHm1"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "sdf.chat(\"Return the top 5 countries by GDP\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "0pVAthBSjEpx",
        "outputId": "a0e4b6a7-9abd-45b9-f88f-47c935a27fef"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "          country             gdp  happiness_index\n",
              "0   United States  19294482071552             6.94\n",
              "9           China  14631844184064             5.12\n",
              "8           Japan   4380756541440             5.87\n",
              "3         Germany   3435817336832             7.07\n",
              "1  United Kingdom   2891615567872             7.16"
            ],
            "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>country</th>\n",
              "      <th>gdp</th>\n",
              "      <th>happiness_index</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>United States</td>\n",
              "      <td>19294482071552</td>\n",
              "      <td>6.94</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>China</td>\n",
              "      <td>14631844184064</td>\n",
              "      <td>5.12</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>Japan</td>\n",
              "      <td>4380756541440</td>\n",
              "      <td>5.87</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Germany</td>\n",
              "      <td>3435817336832</td>\n",
              "      <td>7.07</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>United Kingdom</td>\n",
              "      <td>2891615567872</td>\n",
              "      <td>7.16</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ]
          },
          "metadata": {},
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "sdf.chat(\"What's the sum of the gdp of the 2 unhappiest countries?\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "us1r15KQjZlu",
        "outputId": "efa8a617-e197-4f7f-9fed-366ad5948479"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "19012600725504"
            ]
          },
          "metadata": {},
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(sdf.last_code_generated)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "uzOEjV3LjqDF",
        "outputId": "78c4fe2d-7bd8-4569-b058-34ac5ff2d29b"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "def analyze_data(dfs: list[pd.DataFrame]) ->dict:\n",
            "    df_combined = pd.concat(dfs)\n",
            "    df_sorted = df_combined.sort_values('happiness_index')\n",
            "    sum_gdp = df_sorted.head(2)['gdp'].sum()\n",
            "    return {'type': 'number', 'value': sum_gdp}\n",
            "\n",
            "\n",
            "result = analyze_data(dfs)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Plotting a chart\n",
        "\n",
        "You can also use PandasAI to easily plot a chart"
      ],
      "metadata": {
        "id": "Pl_PQflF4_1u"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "sdf.chat(\"Plot a chart of the gdp by country\")"
      ],
      "metadata": {
        "id": "OeAXD-9pzlSP",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 406
        },
        "outputId": "a2d2bf83-6ee6-4dc9-940c-bedc0982e770"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAgAAAAGFCAYAAACL7UsMAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAABp60lEQVR4nO3dd3gUVd/G8e/sbpJNryQkoQtIEZCOdJAqIMVHRJqCCIqCgiBIkd4RBB5UEFRAkSJFioJIE1CaYug9IZQQSCAhfdu8f/Bmn0SKAZLNbvb3uS4ukuzszpnZmTn3OXNmRlFVVUUIIYQQTkWT3wUQQgghhO1JABBCCCGckAQAIYQQwglJABBCCCGckAQAIYQQwglJABBCCCGckAQAIYQQwglJABBCCCGckAQAIYQQwglJABBCCCGckAQAIYQQwglJABBCCCGckAQAIYQQwglJABBCCCGckAQAIYQQwglJABBCCCGckAQAIYQQwglJABBCCCGckAQAIYQQwglJABBCCCGckAQAIYQQwglJABBCCCGckAQAIYQQwglJABBCCCGckAQAIYQQwglJABBCCCGckAQAIYQQwglJABBCCCGckAQAIYQQwglJABBCCCGckAQAIYQQwglJABBCCCGckAQAIYQQwglJABBCCCGckAQAIQoQk8lEeno6Foslv4sihLBzuvwugBAFUWpqKpcuXSIxMRGz2YynpyeFCxemUKFCaLVakpOTOXv2LCkpKSiKgk6nw9PTk7CwMAICAlAUBYDExETOnj1Leno6iqLg6upKcHAwRYsWRavV3jPfVatW8dlnn7F69WpCQ0NzZVlMJhMxMTHExsaSkZGBXq8nJCSE0NDQ+5YhN1ksFuLi4nBzc8PX1zdP5yWEs5EAIEQui4mJ4bPPPmPv3r14eHjg4uKC0WjEx8eHIUOGUK1aNc6cOUOvXr0IDAzE19cXVVVJS0sjLCyMXr16UbduXVxcXDh69ChvvPEGxYsXx8PDg6SkJFxcXBg8eDAtW7a8Z94WiwWz2YyqqrmyLGazmXXr1vHdd9+RmpqKu7s7JpMJvV7PsGHDqFmzpjWs5AWj0cisWbN45pln6N69e57NRwhnJAFAiFyUkpLCrFmz+PXXX5k2bRqlS5fG1dWVhIQE9u7dS0pKSrbKeeTIkVSoUAGj0ciNGzdYsGABH374IYsWLeKZZ54BQKfTMWHCBIoWLUpCQgJDhw5lxowZ1K9fH09PzweWJfM0QNYKWlVV6++Z/6uqiqqqaDT3nhE8cOAA48aNo1OnTrzyyiv4+vqSnJzMH3/8wfXr163vz/p/1s/P/GxFUay//7McmT9nXS+Zr5nNZiIiIvDy8sq2PFnfnxl6dDqddT5Zl+ufyyuEuEsCgBC56NChQ/z444/MnTuX5s2bWyud8PBwypUrl62CAggKCiI0NBRFUShWrBj+/v706tWLtWvX8vTTTwN3K67g4GDCw8MJCwujUaNGfPvtt8THx983AKSlpbFs2TJ+/vlnkpKS6NatG3369MFisdCvXz+aNWtG79690Wq1WCwWhg4dyu3bt/nqq6+yfY7JZGLq1KmUKVOGYcOG4eHhYS17mTJlrBX2lStXmDVrFrt27UKj0dC8eXM++OADChUqxNmzZ+natStLly6lYsWKAPz000988803fPrpp7i5ufHaa69Rv359Ll26xO+//07x4sWZMGECFSpUYMGCBezZs4fDhw/z7bff0qRJE0aPHs3AgQMpWrQoFouFw4cP06hRI3777TfmzZtH1apVAUhPT6dnz540bNiQd955RwKAEP8ggwCFyCWqqnLu3Dm0Wi1Vq1a1VjgGg4GMjAxMJtNDB+cpikJYWBg1a9bkwIEDGI1G62sZGRnWcQWHDh0iICAAPz+/+37OxYsXOX78OIMGDeK1115j6dKlbNy4Eb1eT506ddi5cycJCQnA3cr7t99+o127dvd8ztWrVzl37hzNmjXD09MzWwWq1WrR6XQkJiYyefJkjhw5wsiRIxk0aBA7duxg4sSJ1uVNTk7GbDZb32symUhJScFisaCqKklJSaxevZoqVaowefJkdDods2bNAqBdu3ZUqlSJl156iW+++Yb33nsPLy8vkpOT2bx5M+Hh4UyePJnu3btTvHhxVqxYYZ3PpUuXOH36NFWrVr1v74YQzk56AITIJWazmYSEBPz9/dHp/rdrzZs3jz/++AOA5s2b07t37wd+hl6vJzAwkJs3b1rDgslkYvDgwbi7uxMXF0d8fDyTJk3Cx8fnvp8RFBTEa6+9xvPPP09aWhpXrlxh8+bNtG/fnlq1arFq1SpiY2MJCAiwlqthw4b3fE5cXBwmk4mwsLAHlvfChQscPHiQSZMmWcckWCwWpkyZwuXLl/9ljWGdvnXr1rz++uu4u7uTlJTEtGnTUFWV8PBw/Pz8KFq0KHXq1AHuDrBUVZX69evTt29faxBq2bIlc+fOJTExEV9fX9auXUvRokV59tlnc1QOIZyNxOIcOHDgAN27d6dKlSosWLAgW2vmfkwmE3PnzqV169Y0bdqUP//80/rapUuXGDBgAA0aNKBp06aMGzeO5OTkvF4EYQOKouDi4oLBYMh2Prtu3bp06tQJg8HAX3/99dBeAIvFgsFgwM3Nzfo3jUZDmzZt6Nq1K2PHjuXnn3+mbdu2D/yMwMBASpUqhaIouLu7U6ZMGaKjozEajZQvX55ixYqxdetWUlJS2Lt3Lw0bNrxvmHB1dUVRFDIyMh44r9u3bwNQqlQp6zqoXbs2BoOBS5cuPXhlZaHVailTpgx6vd5a/qSkpIcOZFQUhVKlSmUrd9u2bTEYDOzYsYPk5GQ2btxIu3btHjpOQghnJgEgB/R6PZ06daJKlSrcuXPH+neLxcLx48f5+eef+fXXX7l27Zr1oFW0aFG6d++Ot7d3tgOoh4cH3bp1Y/ny5cydO5cDBw6wdOlSmy+TyH0ajYYiRYoQHx9PfHy8dVuoU6cOnTp1sp7Tf5iEhATOnTtHuXLlrL0IGo2GF154gZdeeokmTZpQtGjRbD0M/5Q5KC6T2WxGq9WiKAr+/v40btyYH3/8kXPnznH06FHatm17388rWrQovr6+/Pnnnw8MLRqNBlVVs71uMBislzZmDszLLI+qqqSmpmb7DEVRrOXL/P3frmLQaDTWgJLJ39+fFi1asHnzZn755ReSk5Pp0KGDnPsX4gEkAORAlSpV6NSp0z1doXv37mXGjBns2bOHjRs38vHHH5OQkIBOp6Njx440aNAADw+PbO8pVKgQderUoWjRohQrVozg4OBs53qF41IUhSpVqlCkSBEWLFhAenq69e85YbFY2LVrF3/99RctWrTA1dX1ns//5yDC+4mNjeXUqVPW8+tHjx6ldOnSuLi4oCgKzZo148aNGyxbtgwvLy9rb8E/+fn50b59e3755Rf++uuvbJXytWvXOHv2LEFBQbi4uHDs2DHrMuzevRu9Xk+pUqWs9zSIioqyvr5nz54crQ+4W9FrNJoH9kL8s9wvvfQSR48eZdGiRdSoUYOQkJAcz0sIZyNjAB5TcnIy8+fPp2fPnjz33HNkZGQwcOBAtm3bRufOnR/63rS0NDp37syNGzcoU6bMv04vHEfJkiUZPnw4H3zwAZcvX6Z58+b4+Phw+vRpNm3aRKdOnbJVWrt37yYqKoo7d+5w8OBBtm/fzgsvvECzZs0e+yY7SUlJfPnll1y5coXIyEh27drF9OnTcXd3B+5211erVo1vvvmGd99996E3DOrXrx8RERG89dZbdOjQgbJlyxIZGcmWLVt444036NSpk/VU1tWrV0lNTWXZsmX06NGDwoULA1CzZk1mz55NXFwcFy9e5MCBAzmumF1dXSlWrBjbtm2jUqVKlChRgjJlyjxw+rJly+Lt7c2uXbtYvXq1tP6FeAgJAI/pxo0bXLp0iQkTJlhbaiaTicTExH99r5ubG3PnzuXy5ct88sknbN68mT59+uR1kYUN6HQ6WrduTdmyZVm8eDHr168nIyODEiVKMHHiRFq0aIGLiwuenp5UrlyZVatWodFo8Pb2pnTp0syfP5/69evj4uICgLe3N5UrV842JuBhgoKC6NSpEy1btmTt2rWkp6db55tZGWq1Wl566SX27dtHo0aNrPN60OctXLiQ9evX8/PPP7N9+3bCw8Otlb+7uzsjRowgLCyMzZs34+LiYr36IHPk/dSpUxk3bhyrVq2iTp06TJgwgV9//RU3Nzd0Oh0VKlQgKCjIOk9fX1+effZZa2/H0KFDmTJlCvPmzeO5555j8ODBPP300/f0yCmKgre3N02bNuXOnTtUqVLlkb47IZyNoubWLcOcwPDhwwkMDGTw4MFcu3aNrl27MnHiROv1zZmDrjK7/aOjoxk2bBgDBgygbt269/3MJUuW8MMPP7Bx40abLYcQn376KT///DMrVqzA398/v4uTa1JSUhg2bBje3t6MHj36nlNwQoj/kTEAOZCamsqFCxdISEggPj6eqKgovLy8qF+/PosWLeLq1askJSXx+++/Ex0dDcD169eJjo4mOTmZq1evWkdEHzt2jN9//50rV65w5MgRNmzYYL3jmxB57caNG6xZs4alS5fy4osvPvBeAo7o7NmzrFq1iv3799OoUSPrKQ8hxP3JKYAcOHr0KP379yctLQ1FUdi/fz/Dhg1j5MiRzJ8/n0GDBmE2m6lUqRKVKlUC4JNPPmHbtm2kpaUxYcIEQkNDrZdeffHFF1y4cAEfHx8aN27M22+/nc9LKJzFn3/+yaRJk6hfvz69evUqUOfIV65cyU8//UT37t1p1KhRgVo2IfKCnAIQQgghnJCcAhBCCCGckJwCeID7PbVMCCFEweZMT5B0iACQmJjIhQsXuHXrFq6urpQqVYqwsLD7PuAjISGBkydPkpqaStGiRSlVqhQuLi6oqmq9eYmiKJQpU4bQ0NAHPiREVVXOnDnD0aNHH3rXNSGEEAWH2WymbNmyVKpU6bHvxeEoHKJm++uvv/jhhx/w9PQkOTkZg8HAmDFjKFKkSLaElpSUxH//+1/OnDmDr68vcXFxDB06lGrVqhEdHc2kSZPQaDQYjUZ0Oh0TJ04kKCjovilPVVW2b9/Ozz//TM2aNQt8EhRCCGenqionT56kfPnyVKxYUQKAPahSpQqlS5fG39+fxMREhgwZwsGDBylSpEi26Y4ePcrBgweZPHkyRYoU4ZNPPmH58uU8++yzbNu2DYPBwJQpU7BYLLz33nts376dV155xfr+f46H1Ol0tGrVin79+snjRIUQooBTVZW1a9dae4oLOocIAAEBAQQEBGR7kMj9nvAVGRlJWFgYxYsXx9vbm3r16jFjxgxSU1M5deoUderUITg4GI1GQ/Xq1Tlw4EC2AGAwGDh79iwJCQlYLBbOnTtHqVKl0Ol0EgCEEKKAU1XVqY71DhEAMhkMBr7//ns8PT2pVq3aPQktLS0NV1dX6xfo5eVFcnIyRqORlJQUfHx8rLcX9fb25vz58/e8f+fOnZw+fdraFZT5mFMhhBCiIHGIAJDZNb906VL279/PiBEjst07PJObmxsmk8k6fVpaGh4eHuh0OvR6PSkpKaiqau1J+Gcvgq+vL/3797c+3nTRokV5v3BCCCFEPnCIvo6MjAwWLlzI2rVrGT9+POXLlwf+92zx27dvo6oqRYsWJSYmhri4ONLT0zly5AjlypXDw8OD0qVLc+TIEZKSkkhMTOTo0aM8++yz2eaT+QxzFxcXXFxcCvwAECGEEM7LIXoAtm7dyqRJk+jcuTO//vorALVr16Zq1aps3ryZHTt2MHv2bKpWrUrp0qWZMmUKoaGhREREMGTIELRaLS1atOD3339n+vTppKenY7FYaN26dT4vmRBCCJE/HCIAlC5dmpEjR2b7W+b5/ypVqhAQEIBOp8PPz48hQ4awd+9ekpKSeOGFF6yPFS1Tpgxjx47l4MGDKIpCzZo1CQkJcYqRnkIIIcQ/OUQAqFixovWRu/9UtmxZypYta/09JCSEl1566Z7pFEW5Z1ohhBDCWTnEGAAhhBBC5C4JAEIIIYQTkgAghBBCOCGHGAPg6K4mpJGcbrLZ/EoHe6HVyOBGIYQQDyYBwAYmbDrJluPXbTa/iDEt8HV3sdn8hBBCOB45BSCEEEI4IQkAQgghhBOSAJCP1H+f5BEnFEIIIXJGAkA+ytEwPfV/E0oOEEIIkVskANg75b4/CiGEEE9EAoC9kua+EEKIPCQBwF5Jc18IIUQekgBgL6TFL4QQwoYkANgLafELIYSwIQkAQgghhBOSACCEEEI4IQkAQgghhBOSACCEEEI4IQkAQgghhBOSAOAA5ApBIYQQuU0CgAOQKwSFEELkNgkA9k6a/0IIIfKABAB7J81/IYQQeUACgBBCCOGEJAAIIYQQTkiX3wXIidu3b/PLL79w8OBBAgICePPNNwkODs42TUJCAvPnz+f8+fPWv/n6+vLhhx+i1+tZvHgxJ0+eBMDDw4M33niDatWq2XQ5hBBCCHvhEAEgJSWFGzduEBoaytq1a+nSpcs9AcDDw4MOHTqQnJyMqqrs3r2b3377zfr+gwcP0qJFCypXroxWq6VkyZL5sShCCCGEXXCIUwDh4eG8++67dO3a9YHTuLi4UKFCBWrVqkXlypW5fPkyTZo0oVChQgCoqkp6ejoGg4HAwEC8vb1tVXwhhBDC7jhED4CiKNn+f9g0ADdv3uT48eO8+uqruLi44OrqyjPPPMOlS5c4c+YMCQkJDBw4kJo1a2Z7n6qqGI1GLBYLFosFo9GIRmN/GUlFLg4QQgjxZBwiADwKVVXZunUrhQoV4tlnnwUgMDCQgQMH4uLiQlpaGp9//jk//PADFStWxNPT0/reuLg4Jk2axOHDh1FVlevXrzN48OB8WpIHk8pfCCHEkypwASA9PZ3vv/+efv364eHhAYBOpyMgIAAALy8vKlasyKZNmzAYDNkCQEBAAOPHj7f2AixZsiRflkEIIYTIaw4RAMxmM0lJScTHx2M0Grl16xaFChVCVVX+/PNPqlSpQmBgIAB79uwhPT2dFi1aWLv379y5Q1RUFEWKFCEhIYHdu3dTpEgR3Nzcss1Hq9Xi4+MDgMViwcvLC7PZbNuFFUIIIWzAIQLA7du3mTx5MgcOHODs2bMMHDiQVq1a0aVLF7788kuGDRtGYGAgRqORjRs30qxZM/z9/a3vv3HjBpMnTyY5ORkXFxeKFStGt27dcHd3z8elEkIIIfKPQwQAf39/Ro8ejcFgsP5Nr9fj6enJ559/bu3G1+l0TJw4ERcXl2zvL1GiBJ999hkmkwmNRoO7uzvu7u4PHVQohBBCFGQOEQC0Wm22Fn1Wfn5+1p8VRcHX1/eeabKOASjI5OoAIYQQOWV/17iJe+T0gYBS+QshhMgpCQAOQCp2IYQQuU0CgJ3KaatfCCGEeBwSAGzo3yr1rK9Lq18IIURekgBgQ/9WqT9ppS+9BkIIIXJKAkABIr0GQgghckoCgBBCCOGEJAAIIYQQTkgCgBBCCOGEJADkoxwP2pPRfUIIIXKZBIB8lKNBe1nu7ys5QAghRG6RAGDvlPv+KIQQQjwRCQD2Spr7Qggh8pAEAHslzX0hhBB5SAKAvZAWvxBCCBuSAGAvpMUvhBDChiQACCGEEE5IAoAQQgjhhCQACCGEEE5IAoAQQgjhhCQACCGEEE5IAoADkCsEhRBC5DYJAA5ArhAUQgiR2yQA2Dtp/gshhMgDEgDsnTT/hRBC5AFdfhcgJ0wmE4mJiaSkpKDT6ShUqBAuLi7ZprFYLNy+fZvExETr3/z9/fHz80NRFDIyMoiPj8dgMODt7Y2fnx9ardbWiyKEEELYBYcIANevX2fhwoUcPnyYhIQEli1bxlNPPZVtmrS0NCZPnsxvv/1GyZIlAejcuTMdOnTAZDKxatUqfvzxR1RVxcfHhwEDBlC1alU0GukEEUII4XwcIgD4+fnxyiuv0KZNGwYNGvTQaXv16sVrr72Goii4uLig1WqJiopiw4YN9OzZk9q1a7NgwQJWrlxJuXLl8PT0tNFSCCGEEPbDIZq/Xl5eVKxYkWLFij1wmswKf9GiRXTq1ImpU6cSExMDQExMDIqiUKtWLUJCQmjQoAGRkZGkpKRk+wxVVe/5J4QQQhREDtEDkBMuLi68+uqrvPzyy2i1WhYsWMC0adOYOHEiCQkJeHh44O7ujqIoeHl5kZGRgclkyvYZ6enpHD58mBs3bqCqKn///TeVK1fOpyUSQggh8k6BCgBVqlSx/v7GG28wadIk4uPj0ev1GI1Ga4VvNBrRarUoSvYh9kajkaioKC5evIiqqly9etUuA4CKXBwghBDiyThEAMjsis/aJZ/5s8VisQ7ky/q6wWDAbDajKApBQUGkp6cTHx9PQEAAV69exdvbG71en20+3t7edO3a1dr9/+WXX9rlaQCp/IUQQjwphwgAaWlp7N+/n7NnzxIfH8/mzZupXr06oaGhfPLJJ/Tv35/Q0FCWL19OoUKFAFizZg01atQgJCSEgIAAihcvzvz586levTrr1q3jlVdewcfHJ9t8FEWxXhqYGSzMZrPNl1cIIYTIaw4RAAwGA6dOneLcuXO0bt2aixcv4uPjw1NPPUW1atXw8fHBzc2NgIAAIiIi0Gg0/Oc//6F58+Z4eXkBMHjwYNavX8+pU6fo2bMnLVq0kPsACCGEcFoOEQD8/Px455137vvaG2+8Yf25a9euD/yMokWLMmDAgFwvmxBCiPubtuU0F24k22ReLloNc1+tilYjJ0lzyiECgBBCCMdzMPIWf166bZN56XWa/x+zJQEgpxziPgAiZ+xvuKIQQgh7JQHAAeS0YpfcK4QQIqckADgAqdiFEELkNgkAdkq684UQQuQlCQA29G+VetbXpdUvhBAPJo2kJycBwIb+rVJ/0kpfdgghhLOQRtKTkwBQgMgOIYQQIqckAAghhBBOSAKAEEKIfCOnLvOPBAAhhBD5Rk5d5h8JAPkox8lXIrIQwsnJYTD3SQDIRzlKvllubS07gBCioHjU45n0FOQ+CQD2Trnvj0II4dDkeJb/JADYK2nuCyGEyEMSAOyVxGMhRAGVm+0baSs9PgkA9kK2YiGEk8hx+yYHx0VpKz0+CQD2QrZiIYTITo6LeUoCgBBCCMcgPaW5SgKAEEIIxyA9ArlKAoAQQgjhhCQACCGEEE5IAoAQQgiHIcMAco8EAAcgG7wQQtwlwwByjwQAByAbvBBCiNwmAcDeSfNfCCFEHtDldwFyIj4+nvXr1/Pbb79RqFAhPvjgA0JDQ7NNk5qayoYNG9ixYwd37tyhdu3adO3aleDgYOLi4pg9ezZHjx4FwMPDg8GDB1OnTp38WJxHI81/IYQQecAhegDS09NRVZV69erx+++/k5qaes80aWlpXLlyhXbt2jFo0CBOnDjB3LlzSU1NJT09nfPnz9O9e3dmzZrF5MmTeeaZZ/JhSYQQQgj74BA9AOHh4fTp04eYmBi++eab+04TGBjIkCFDAFBVlTt37rB06VKSk5MBMJvNnD9/Hr1eT9myZXFzc7vnM1RVxWw2Y7FYsFgsmM3mPFsmIYQQIj85RAB4VAaDgQMHDlC0aFG8vLxISUmhVq1aGAwG9u/fz7Jly3j//fepV68eGs3/OkFu377NwoULOX78OKqqcv78eXr27JmPSyKEEELkjQIXAEwmE2vWrOHw4cNMnToVDw8P9Ho97777LlqtloyMDBYuXMiaNWuoWrUqXl5e1vd6e3vTuXNnWrdujaqq/PDDD/m4JEIIIUTeKVABICUlhdWrV7N27VpmzpxJmTJlUBQFrVaLp6cnAG5ubpQoUYITJ05gNBqzvV+n01GyZEng7umA/fv3y2kAIYQQBZJDBACz2czt27eJjY3FYDBw48YNAgICUFWVPXv2UKdOHXx9fVmxYgWbN29m6NChhIaGkpqairu7O4mJiZw9e5aSJUsSHx/Pli1bKFmyJO7u7tnmoyj/G3KvqvZ7/Z2KXBwghBDiyThEAIiPj+fjjz9m//79REVF0a9fP1544QV69erFypUrKV68OGazmUWLFnH58mWGDBmCVqulfPnyTJo0ieTkZKZPn87Nmzfx8PCgatWq9OzZ874DAR2BVP5CCCGelEMEgEKFCjFnzhwsFov1b1qtFp1OxzfffINOp0NRFHbt2pVtGo1Gg6urK8HBwSxfvtzaqtfpdGi12mwtfiGEEMKZOEQAUBTlga11V1dX688PmuZh7xdCCCGckUPcCEgIIYQQuUsCgBBCCOGEJAAUIPZ73YIQQgh7IwHAAeS0YpchjUIIIXJKAoADkIpdCCFEbpMAYKekO18IIURekgBgQ/9WqWd9XVr9QgjxYNJIenISAGzo3yr1J630ZYcQQjgLaSQ9OQkABYjsEEIIIXJKAoAQQgjhhCQACCGEyDdy6jL/SAAQQgiRb+TUZf6RAJCPcpx8JSILIZycHAZznwSAfJSj5Kv+b0LZAYQQBcWjHs+kpyD3SQCwd8p9fxRCCIcmx7P8JwHAXklzXwghRB6yWQBQVZU7d+5w4cIFbt++japKDfdQEo+FEAVUbh79pSZ5fDYLAGfPnqVbt2506dKFLl26EBERYatZOwbZioUQTiLH7ZscHBelrfT4bBYApk+fTqlSpZg7dy41a9Zk3LhxWCwWW83e/slWLIQQ2clxMU/pbDWjs2fP8t1331GsWDEqVapE8+bNSUlJwdvb21ZFEEII4ciyXBUlnpzNAkBaWhoJCQkAmEwmzGYzFy9exN/fH61WS3h4uK2KIoQQwhFJ5Z+rbBYAdDodXbt2RVEUVFXl1q1bdO3aFY1GQ+HChdm2bZutiiKEEEI4PZsFgM2bN2M2m+/7mkYjVyMKIYQQtmSzABAYGIjBYODs2bNcvHgRgFKlSlGmTBnc3NxsVQwhhBAOTIYB5B6bBYD09HSmTJnCmjVrrC1+i8VCx44dGT58OJ6eng98r6qqmEwmLBYLiqKg0+nu22uQOZ3ZbEaj0WSbzmKxWD9Dq9Wi0+lQFMfYjGSDF0KIu+RYmHts1vf+008/sXPnTr744gsOHDjAwYMH+fLLL/njjz/YuHHjQ9977do1xo8fz/PPP0/z5s2Jioq6ZxqLxcKxY8fo168fzz//PN26dWP37t2YzWaMRiM///wznTt3plmzZrz33nucO3fOYW5GJBu8EEKI3GazAPDjjz8yaNAg6tevj7u7O3q9nueee47333//XwOAh4cHLVq0YNy4cRiNxvtW3CkpKSxZsoQiRYrw3Xff0bhxY7766itu3rzJ1atXWbJkCZ06dWLp0qW4ubnx3XffkZ6enleLm3scI6MIIYRwMDYLAJcuXaJGjRr3/L1atWpcunTpoe/19/enQYMGVKhQ4YHTJCYmcunSJVq1akWJEiVo0aIFiYmJxMXFERMTg8VioWnTppQsWZJWrVpx5swZkpOTs32GqqrWf3ZDmv9CCCHygM3GAKSkpGA2m0lMTMz2d5PJRGpq6hN/vsFgICMjAx8fHwC8vb3RaDQkJSURFxeHh4cHnp6eKIqCn58fqampGI3Ge8py5coVkpOTsVgsXL16lcKFCz9x2YQQQgh7Y7MAoCgK48aNu2fEv8FgyJVbAiuKgqIo1s+yWCyoqopWq8XFxQWz2ZztNY1Gc88gwKSkJFasWMGxY8dQVZXz58/z+uuvP3HZhBBCCHtjswAwbty4e7rcM3Xq1OmJP1+v16PX64mLiwPg1q1bAPj4+KDRaKx3IgwICCA2NhZPT09cXV2zfYa/vz9Dhw7FYrFgsVhYvHjxE5dLCCGEsEc2CwB169bl7Nmz1KxZkyVLlhAVFWVtgffr1++h7zUYDJw/f57IyEiSk5OJiIjAYrHg5ubGokWL6N69O8WKFaN69eosWbIEnU7Hhg0bKFasGIULF8ZgMODn58eyZcto1KgR3333HU2bNrWeLsiUeYkhYL1c8EE3LxJCCCEcmc0GAX799dfs378fgOXLl3P16lVSUlL4+++/+fzzzx/63jt37rB06VLmz59PeHg4ixYtYvXq1QDW6/nd3Nzo06cPzz77LHPmzMHd3Z0hQ4bg6+tLoUKFGD58OKmpqcyfP5/nn3+erl27Wit7IYQQwtnYrAbcs2cPkyZNAsDFxYVhw4bx1FNPceLECYYMGfLQ9wYFBTF16tT7vvbxxx9bfw4MDGTQoEEMGjTonunKli3L9OnTn2AJ7IfcGEgIIcSTslkPwI0bN/Dw8ABg5syZ1qf/+fr6Ws/Xi5yRyl8IIcSTslkACAsL46+//kJVVcqVK4der0dVVSIiIggLC7NVMYQQQgiBDU8BvPTSS8ybNw+tVkvZsmVRFIVz584xb948unbtaqtiCCGEEAIbBoAXXniBc+fOMW7cOOsDesxmM23btqV9+/a2KoYQQgghsGEA8PT0ZMiQIbz++utERUVhsVgoWbIkQUFB8jhgIYQQwsZseidANzc3wsPDrQMARe6SqwOEEELklM0GAYrHl9NHE0nlL4QQIqckADgAqdiFEELkNgkAdsqOHkgshBCiAJIAYEP/VqlnfV1a/UII8WDSSHpyEgBs6N8q9Set9GWHEEI4C2kkPTkJAAWI7BBCCCFySgKAEEII4YQkAAghhMg3cuoy/0gAEEIIkW/k1GX+kQCQj3KcfCUiCyGcnBwGc58EgHyUo+Sb5f6+sgMIIQqKRz2eSU9B7pMAYO+U+/4ohBAOTY5n+U8CgL2S5r4QQog8JAHAXkk8FkIUULnZvpG20uOTAGAvZCsWQjiJHLdvcnBclLbS45MAYC9kKxZCiOzkuJinJAAIIYRwDNJTmqskAAghhHAM0iOQqyQACCGEEE5IAoAQQgjhhHT5XYBHsXfvXrZt24arqysdOnSgQoUKKMr/+oROnDjBqlWrSE5Otv6tdOnS9OvXj1OnTvHDDz9w584dAAoXLsxbb72Ft7e3zZdDCCHE48lyc1TxhBymB+DIkSOMGDGCQoUKYTKZeP/997l161a2aYKDg2nSpAktW7akSZMmnD59mujoaFRV5eLFi+zdu5dGjRrRsmVL6tWrh6uraz4tzaORcS9CCHGXVP65xyF6AFRV5bvvvqNRo0b0798fi8VC586dWbduHX369LFOFxQURKNGjQA4deoUZrOZl156CY3mbs5xdXUlICAAf39/wsPD7wkAqqpisVhQVdX6sz2QDV4IIURuc4gAYDAYiIyMpFu3biiKgk6n49lnn+XEiRPZpss8HWA2m/n7778JDg6mePHiKIqCn58fYWFhfP/99yQmJlKiRAmGDh2Kj4+P9X1JSUls2LCBixcvoqoqhw8fplWrVjZf3mykv0sIIUQecIgAYDKZMBqNuLu7W//m7u7OtWvX7jt9eno6O3bsoHbt2vj7+wPw7LPPUqZMGfR6PVFRUQwfPpwDBw7QvHlz6/t0Oh3FihXDzc0NVVW5cuVK3i5YTkjlL4QQIg84RADQ6/V4e3sTGxsL3O2qv3HjBsHBwfed/urVq5w5c4YBAwag091dRG9vb+uAv0qVKlGsWDHOnj2bLQC4u7vToEED6ymA+Ph4uzkNIIQQQuQmhxgEqNVqqV+/Plu2bOHKlSucOXOGP/74g8aNGxMdHc2hQ4cwm83W6VesWEH58uUpX748ABaLhatXrxIdHU1iYiJHjhwhMjKS0qVLZ5uPoigoioJGo7H+LIQQQhREDtEDAPDSSy9x7Ngx3nvvPUwmE02aNOG5555j/fr1bNu2jfnz5+Pu7k5CQgJbtmxh1KhR1kF+qqqyb98+1q1bZ/29QYMGPPfcc/m5SEIIIUS+cZgAEBwczJQpU7h16xYajYagoCD0ej1t2rShSZMmuLm5AeDl5cWKFSsICQmxvlej0dC6dWtq166N2Wy2Xg3g7u4urXwhhBBOyWECAICvry++vr7Z/ubj44OPj4/198yBfFkpipJtDIAQQgjh7BxiDIDITm4MJIQQ4klJAHBActJCCCHEk5IAIIQQQjghCQBCCCGEE5IAIIQQQjghCQBCCCGEE5IAUIDI1QFCCCFySgKAA8hpxS5XBwghhMgpCQAOQCp2IYQQuU0CgJ2S7nwhhBB5SQKADf1bpZ71dWn1CyHEg0kj6clJALChf6vUn7TSlx1CCOEspJH05CQAFCCyQwghhMgpCQBCCCGEE5IAIIQQIt/Iqcv8IwFACCFEvpFTl/lHAkA+ynHylYgshHBychjMfRIA8lGOkq/6vwllBxBCFBSPejyTnoLcJwHA3in3/VEIIRyaHM/ynwQAeyXNfSGEEHlIAoC9kngshCigcrN9I22lxycBwF7IViyEcBI5bt/k4LgobaXHJwHAXshWLIQQ2clxMU9JABBCCOEYpKc0V0kAEEII4RikRyBXOUwAMBqNfPHFFzRu3JgWLVqwfv16zGZztmkuXrxIx44defbZZ6lWrRo1a9bkzJkzABgMBpYtW8bzzz9P8+bN+e6778jIyEBVJVIKIYRwPrr8LkBObdq0ieXLlzNjxgzi4+MZNWoUZcqUoWLFitZpjEYjiYmJfPfdd4SFhQHg7e2Nqqr8+uuvLFmyhLFjx5Kens7EiRN56qmnqF27dn4tkhBCCJFvHKIHwGKxsGXLFjp16kStWrVo3bo1VatW5eeff75nWpPJxDfffMOiRYs4ffo0qqpiMpk4dOgQjRo1ok6dOjRt2pQaNWqwffv2fFgaIYQQj0v6bHOPQwSA9PR0YmNjKV26NACKolCmTBmioqKyTeft7U337t2pXr06FouFUaNGcfDgQev7S5QogVarRavVUqxYMS5dupTt/RaLhTt37hAfH098fDwpKSm2WsSHkg1eCCHukmEAucdhTgHkRGhoKH369AHuVuZms5mVK1fyzDPP3Hd6Rcm+KcXFxTFp0iQOHToEQGxsLIMHD87bQueAbPBCCCFym0MEAL1eT0hICOfOnQNAVVXOnTtHhQoVsk2nKIq1UlcUBW9vb86fP4+bmxshISFERUVhNptRVZXo6GiKFSuW7f2FChVixowZWCwWLBYLixcvts0CPkyWhwEJIYQQucUhTgFoNBpat27NunXrOHDgAJs3b+bIkSO0bt2aP/74g0WLFmEwGDh//jy7du3i8uXL7Nmzh40bN/Lcc8/h6upKrVq12L17N3/88Qfbt2/n8OHDNGvWLNt8FEXB1dUVvV6PXq/HxcUln5Y4a6HyuwBCCCEKIofoAQBo06YNN27cYPjw4ej1esaMGcPTTz/Nhg0bOH78uPX8/eLFi7lw4QKBgYG8/PLLvPLKKyiKQrNmzbh16xbjx49Ho9HQr18/nn322fxeLCGEECJfOEwAcHFxoW/fvvTt2zfb3zt27EjHjh0BqFatGsuWLXvg+7t160a3bt3yvKxCCCGEvXOIUwBCCCGEyF0SAIQQQggnJAFACCGEcEIOMwZA/I9cGSiEuJ+bSRks3htps/nVLhlAk3LBNpufyF0SAByQVP5CiPu5nWrgi90XbDY/i0WVAODA5BSAEEII4YQkAAghhBBOSAKAEEII4YQkAAghhBBOSAJAASKPDRZCCJFTEgAcQE4rdrk6QAjxMNJIEFlJAHAAUrELIXKDHEtEVhIA7JQkdSFEbsj1Y4kcnAoMCQA29G/7TdbXJakLIR5Xnh5L7OTgJDnkyUkAsKF/22+edL+SHUIIAXZTR+cpZ1jGvCYBoACRHUIIIUROSQAQQgghnJA8DEiI/3c46hZf74t6vDcrPPI5mNfqlqBWyYDHm58QBYQ83TT/SAAQ4v9dTUhj87GYJ/uQRziaNasQ8mTzEqIAkMo//8gpgHyU4wajjO5zHHI0E/bMgY8lDlx0uyUBIB/lqK7I0qKUHcB2ZF2LAsfOjiWPWgbJ1rlPAoC9U+77o8hjsq5FgWNnxxJ7KIOzkwBgr+whoovHIl+dsCuyQYoHkABgryQe54uHHitzeCDN/OrkuCvsgh0eS3Jz35D97PHJVQD2wkbXwly4mcyYH0/k/Yz+X6dq4XSqVsRm83tSD/0KHvH7scPjrnAGDnBdXY6Ll4NlsfNFtWsSAOyFjbbi5HQTe8/H2WZmQPXi/jablxCCglUjFqRlsUMOFQBOnDjB33//jU6no06dOhQrVgxF+d8WkpCQwN9//82VK1fQ6XRUrFiRcuXK4eLiQnR0NHv37iUlJQWAgIAAWrdujYeHR34tjihIHKDVJYTDk/0sVznMGICzZ88yfPhwTpw4wd69exk+fDhJSUnZpjl58iSbNm3i2rVrnDp1inHjxnHy5ElUVSUiIoLPP/+c5ORkjEYjRqMxn5ZEFEhyUBIFiN2eV5f9LFc5RA+Aqqp8//33lCpVilGjRmEymejcuTM//fQTXbp0sU5XrVo1KlSogLu7OyaTiY8++oidO3dSqVIlAAIDA+nYsSPe3t54enri5uaWX4skhBB2S+pZ5+AQPQBGo5Fjx45Rv3593N3d8fb2pnbt2hw+fDjbdHq9Hj8/P9zc3MjIyCApKYmQkLu3W/X09ATgvffeo2/fvixYsIDU1FRU9X9ZV1VVTCaTtYfAbDbbbiGFEEIIG3KIHgCj0YjBYMDb2xsARVHw8vIiNjb2vtMbDAaWL1+ORqOhYcOGKIpClSpVmDFjBt7e3pw4cYLJkydTrVo16tevb31fYmIi3377LadPn0ZVVU6cOMHLL79sk2UUjk9OTwpH5ijbr6OU0xE4RABwdXVFr9eTkJAA3G2pJyYm4uvre8+0ZrOZ5cuXs2vXLsaOHUtoaCiKohAYGEhgYCAAhQoVYsWKFRw9ejRbAHB3d6dhw4ZUrlwZi8WCXq+3yfL9G3vc4HOrTAO+P0JkXEoufNK/83TVsqJvnWwDR3OTvX1HQvzTw/ZbR9l+HaWcjsAhAoBOp6NGjRps376dli1bkpGRwe+//87777/PrVu3SEpKomjRoqSlpfHtt9+ybt065syZQ8mSJbFYLCiKQmJiIhaLBXd3d65du0ZMTAytW7fONh9XV1freAGLxcLp06ft4jSAPW7wuVWm8zeSOBWT9O8T/tNjJBBvvc4uw5QQtiLbvsjKIQKAoii88sorDB8+nJEjR5KamkrZsmVp2rQpmzdvZvv27cydO5c//viDqVOnUqNGDZYtWwZA7dq1ad26NTt27GD37t14eHhw8+ZNSpQoQcOGDe+Zz/1+zlfOUGM9zjLmwzpxhq9CFGCyAYt/cIgAAFC8eHGmTJnCuXPn0Ol0VKhQAU9PT+rVq0fZsmVxcXGhfPnyfP7559la7cWLF0ej0VCrVi38/f3JyMjAx8eHMmXKEBgYaD8V/YPkZfHs5YBgD2XIAQcpphD396QbsL0cL0SucZgAAFCiRAlKlCiR7W9FihShSJG7t5oNDw8nPDz8vu/NOp34fwVpZ5aDkxB5S0H2swLGIS4DFOJf2clByW5voCJEbrCT/UzkDgkAQuQiOT4KIRyFBABRsEgTXAghckQCgChYpAkuhBA54lCDAMVdzjAOxxmW8WFUVWXu9nOkGS02mV8hbzfeqF/SJvMS9sPZ9zNnJwHAATnDDusMy/hvvtoXRWKabZ5aWTbESwKAE5L9zLlJABBCOBSzReVUzB2bDffw0esoHuhpo7kJYTsSAIQQDiUlw0S7eXttFgBaVAhhYc8aNpqbELYjAUAIIZ7AV3sj2X8x3mbzm/FyFXzdXWw2P1FwSQAQQogncPxqIr+cvP+jyfPCRJMZkAAgnpxcBliA5EaXqFxGL8Tjud++I/uTsGcSABxATg8iuTGiN6efIQc2IbK7376TG8/fyU2y34qs5BSAA7DHS3XssUwi7xhMFgwm29yTAMDdVYtWI1tZbq8BWaMiKwkAdipPb9CRyx8uNxMp+Bb8doFZ287abH5r3q5LtWL+eT4fR9p2H7esub6MjrTSxENJALChf9tvsr6ep/tXLn+4HAsKPlW9+892M3yyt+Z0m3SkbfdRypqnxxI7WWmSQ56cjAGwoX/bWO3hfGF+nSNU//G/eDLOsB4ftIwFoVJ40u+vIKyDf+MMy5jXJAAUILYcBJjbbNLz4UScYT0W5GUsyMsm7IcEACGEEMIJSQAQ4gGcoRtd/I983/lD1nv+kQAgxAM8VjesIxzNnrSMjrCMj0G63fOHrPf8IwEgH+X4OGqPB1x7LJM9yOOjWa6s9ictYy4so2w+OSR3ArJy4KLbLQkA+ShHx9Es17rYww5gLUNuVnT2sGD/YIdFAvK3tZSb68SeW3129d0r5F6B7PVYkkP2vM04KgkA9k6574/5Jk/KYA8L9g92WKR85yzrxO6WM7cK5AzHEvFIJADYK3uI6DnlSGW1gTxdHU/44XbzVdlLQeylHDnxuGV1pGUUNiUBwF45Ujy2QVltdQx76HxyWIjM1ZEnZX7CdW03m1UeFOSR1/dDbiVnl3Xm464zu/nS/yc3169dflcOwqFuBayqKur/349UUZRs///bNJl/+7f355uCdF/LPFgWW62ah87nEQtRUL5OR/HI6/shb3Do784BjiU5Ll4OlsXOF9WuOUwPQFpaGtOnT6dhw4Y8//zzLF26FLPZnG0ak8nE1q1badeuHfXr12f48OHExsaiqippaWl89tlnNGnShKZNm/LFF1+QlpZmDQT5riBtxQVpWZyAnewBIrcUpP2vIC2LHXKYAPDjjz/y66+/8t///pfRo0fz3//+l2PHjmWb5ty5c8ybN4/XX3+dL7/8kitXrrBixQpUVeXXX39l06ZNzJgxg/Hjx/PDDz9w+PDhfFoaUeA4Yi36/2WWY6xwGI64n9kxhzgFYLFY2LZtG+3bt6dKlSoAVK9ena1bt1K1alXrdCdPniQoKIjmzZvj5+fHiy++yPfff0+/fv04fPgwDRs2pGrVqmg0GmrUqMHOnTtp0KCB9f2qqmIwGDCbzdl+TktLe6JTBWZDBhZjxuOvgEegKJCeloaLarzv6+npaTYrC4AxI53U1NQHvm7LdaNqLXe/ywe8bkhPt+m6MaSnPXTdWIwZWIz3/x5zm8Xo+tCyGDNsu24y0tNITXW772vpGSZUkwGLjXrvTIaMh64bkw23YYD0tHRStZYHvGbj/dtgP/u3BS1paWloNY9/rM487ttNz3Aec4gAkJ6eTlxcHCVLlgTunrcvWbIkly5dyjbd7du38fHxwcXFBYDQ0FBiY2NJT0/n5s2b1KtXD61Wi0ajITw8/J4ehBs3bjB27Fj2798PQHJyMqqqsnTp0icq/+1UAwbj/XfYB1FVlcTERHx9fVCUR+uoafnrBB6UV4xmC4ZkwyN9Htz9DlRVRa/XP1IYWviTju/1D97MbiZnYDI/2s72v3Xj+0hliVeg/k/6B76eZjRjSH20CldVVVJTU3FxccHV1fWR3jtmqwtTXbQPfD3uTvojP4LXYrGQlJSEj4/PI62b0xqFet/fv8IFSM4wYUg3PVJZVFUlOTkZd3d3dLpHO9S8uc0VF+39t3tVhfQ76Y/0eQBGo5GMjAw8PT0fad3s2KSh3vwHf7cJqUYMRvMDX78fVVW5c+cO3t7eaDSPtn+32+6G5gHlN1lUDEmPXuFmZGRgNptxd3d/pHWz9Ccd68c/+LuNTzZgMD/use/R9m8j0HDrg/fvnEpNTeW1117DYnm0cjsihwgAiqKg0WiynfM3m81otdkPnhqNJltys1gsaDSaHL+/UKFCzJ49G4vFgqqqGP+/9eXi4mLzwYLp6ek0b96cTZs24evra9N5/5PFYmHZsmWkpKTQr1+/e9abraWkpNC8eXO2b9+Ou7t7vpbFZDIxceJEatWqRevWrfN9UOm1a9fo27cva9aswc3twRW6LRgMBvr06cPw4cOpUKFCvpZFVVUOHz7MN998w6effmptJOSX9PR02rRpw6pVqwgMDMzXslgsFn744QcuXbrE4MGD833/Tk1N5fnnn2f79u14eHjYfP4mkwmdTpfv24gtOEQAcHNzIzQ0lDNnztCuXTtUVeX06dNUq1Yt23RBQUHcvn0bg8GAp6cn0dHRhIeHo9frCQ0N5eLFi5hMJjQaDVFRUdYehUwajQa9/skTZG5QFAWtVouHh0e+7ARZWSwWXF1dMRqNeHh45PsBIjPYeXh42EUAcHFxwc3NDQ8Pj3wPAO7u7tZ1k98BQKfTodPpcHd3f+SWZW5TVdXaE+Hh4ZHvB3eNRmNX+7ebmxuurq52sX8D1m04v9dNQecQgwA1Gg3t27dn3bp17Nixg5UrV3LixAnatWvHrl27mDVrFgaDgcqVK5OUlMT333/P3r17+eGHH2jTpg2urq7UrVuXPXv2sH37djZs2MCRI0do2bJlfi/aA2UGAHuR2YtiD+xt3WT2MtkDe1s39lQWwG62YbCvstjT/g32t90UVA7RAwDQvHlzbt++zaeffoper2fmzJmULFmSM2fOcPv2bVRVpXjx4kyYMIFPP/2UTZs20aZNGzp16oSiKDRu3JiUlBQWLFiAVqvlww8/pGLFivm9WA+k0+kYOnRovrdw4e7BoW7duhiNRrs4SLi6ujJs2LB8b8XB3YN4u3btCAkJye+iAODr68uAAQMe+Zx7XtBqtfTq1YvQ0ND8LgoAJUqUoGvXrnZRueh0OgYPHoynp2d+FwVFUahRowZPP/20XQRZV1dXhg8f/shjasSjU1RnGe4ohBBCCKv8b84JIYQQwuYkAAghhBBOSAKAEEII4YTyf6SQsJmsD0Syu4chOZF/PpCqoMxLFCyy7RR80gPgZOLj4/nmm2/4888/rU9OtIdxoKqqZrsBk9lstv7+qJ/zz39Z/24vVFUlIiKCW7du5Xm5LBYLCQkJ1ltc56fM79lsNmMwGDCZHu3ugnktczsxm83W9ZXf207m/DP3B1vcoS5zeZOSkjhz5gzXr1+33iI3v9ZFfn8PBZH0ADiRpKQkpk2bRkJCAuXLl7ceSOzhsii429KIiIhg165d+Pn50axZM8LDwx/5c0wmE1euXOH69esEBQURGhpqF5dbZZWamsrcuXOpVasWffr0ybPvQFEUjh8/zsKFC5kwYQJ+fn7Wv9ta5gE8IyOD7du3ExERwVNPPUXjxo0JCQnJ95ZmZuVy+fJl64PCKlasSNmyZa2v27qMmWWKjo7m5MmTaLVaqlSpQqFChfLkktysFWxCQgLDhg3j0qVLeHl50aFDBzp27Iinp6fN10VmuZKTk4mLi0On0xESEpIvd2ktSCQAOJGIiAjOnz/Pp59+itFoZNmyZRgMBl588UUKFy6cb5VCpsz7PAQHB3PgwAF27tzJmDFjKFWq1CN93m+//cZ3332Hh4cHsbGxtG3blpdffjlf7yr2z+5UT09PWrVqxcqVK+natSteXl7ZXs+NeWXy8/MjNjaWAwcO0KpVqyf+/CcpT2bI27BhA8HBwWzevJm///6b/v37U7Ro0Xw/mF++fJmxY8fi7e2N2Wzmyy+/ZNy4cdSsWROwXQjIus6ioqIYP348vr6+xMTEEBQUxIABAyhTpkyeBcdbt27x008/4eXlxYwZM9i/fz/r16/HYDDw8ssv4+3tbfN1cfPmTb744gsuX76M0WikZcuWtG7d+pGfGSD+R04BFGD36w6Pjo5myZIlTJs2jT/++IOtW7eyZs0a63MP8qOMcPe+8XFxcZQtW5bRo0cza9YsvL29+eijj7h48WKOuv4yHyKyZMkSmjZtytChQ4mMjMRgMODi4mIXXYgXLlzgxo0bqKpK8+bNycjIYMOGDdbXn6R8FovF2quT2Y2tqirh4eE0btyYzZs3YzabbXqwtFgsxMXFWX//448/WLZsGQ0aNGDEiBEMGjSImzdvMnv2bJucDsmUuS3ExsZy584dVFUlLS2Nb7/9lqeeeoqhQ4fi5uZGXFwco0eP5tChQzYrV3R0NBkZdx/ok5qaytdff03p0qUZNGgQgYGB7NixgzFjxnD27NlcP62jKAoZGRmsW7eOOXPmUL58eSpVqkS3bt3o2LEjmzdvZvXq1Q99AmBeSElJ4auvvuLWrVv079+f8PBwJk6cyA8//EBqamq+79eOSgJAAZW5QyQmJhIREcHff/9NmTJleO2114iLi+OVV15h1qxZtGvXLlvFYcvyZbYgjhw5Qv/+/Xn99dc5cOAASUlJFC5cmClTphASEsKYMWOIjY3918+CuxWOwWCgbNmyfPTRR3Tu3JnXXnuNI0eOcOHCBVst3j1lUxSFmJgYPv30U3r27MmXX35JdHQ0L7/8Mr/99lu2SvJxmEwmVq9ezbfffktaWhqHDh2ib9++bNq0iejoaFq1asWxY8fYuXOnTb/rTZs2sWTJEiwWC+np6fz555/8/vvvxMTEoNFoqFKlCgMHDiQlJYXRo0fbrFxwt5U7evRoJk2aRFxcHG5ublStWpX//Oc/bNy4kaSkJJYsWYK7uzvvv/++Tbafq1evsnTpUq5cuYLJZMJisVC9enV69erFunXrMJvNbNy4kfj4eMaNG8eVK1eeeJ7/DMZubm40bNiQZ599lg0bNnDy5En0ej0vv/wyHTt25OTJk9y5c8emQTIlJYWAgADeffddLly4wLlz5+jevTvTp0/n22+/JSEhQULAY5A7ARZQqqpy8+ZNZs2aRUJCAklJSXh5eTFhwgQCAgK4desW+/btY/Xq1fTt25eGDRva9Da/Wbv15s2bh7u7O6VKlWLRokXUr1+ft956i5CQEGJiYtiyZQtt27YlODj4vp9jsViIjIzE1dUVLy8v+vTpQ3p6OvXr12fw4MHcvHmTGTNm0L59e5o2bWqzA1fWyj+zgklPT2f//v1s376dmzdv8tRTT7Fy5Upmz55Nq1atHrtLNzU1lZUrV7Jp0ybat29P7dq12b17N0eOHCE1NZW6devy559/EhQUxIgRI6xjIvJqXWT9fr28vDhy5AiKolCuXDm+/fZbduzYQffu3enQoQOKovDXX39x+fJl6+95/R1lfjcRERGMGTOGSpUq0b9/fwoXLkx8fDwjRoygU6dONGjQgPHjx2M0GhkwYAClSpXKk7KpqsqlS5fw9/fHaDRiNBrZtWsXTZo0wdPTk4SEBMaMGcPrr79O5cqVGT16NKqqMnToUIoVK/ZEZcr6XUVGRuLu7k6xYsWIjY1lzpw5aDQa+vbtS8WKFcnIyMBoNOLj4wPk/faTuf8YjUbrLd/feust+vfvT7169fjPf/5DaGgoo0aNomTJknIq4BHJGIACJuuOs3XrVhITExk0aBDLly/nwIED1vvnHz16lC1btvDaa69Rt25dm9/jX1EUrl+/zvDhw7lx4wZTp06lUqVKFC9enPHjx6PT6ejfvz+hoaH06tXLukz328H379/PnDlzePnll3nhhRfo2LEjc+fOxc3NjR07drBjxw5cXFyoVKmSzZYv6zn/9evX89133+Hl5UXnzp1p2bIl9erV4+zZs+zfv5/AwEB27txJo0aN8PDweOTvIvNJd//5z39QFIUff/wRi8VC165dadOmDX/99RdHjhzhwIEDuLq60r59e2rVqpUXi32P4OBgMjIy2LdvH3/++SdDhw7l1VdfBWDlypW4urrSrl07qlevTo0aNWxSpqzCwsLo3Lkzw4YNQ6fT8fbbb+Pq6opWq2XTpk0cO3aMmJgYJk2a9MQV7cPEx8czYcIEmjRpwiuvvMKZM2dYtWoVly9f5s0330RRFOLj49mxYwd//PEHGRkZTJo0icDAwFyp/C9fvswnn3zCjRs30Gq1VK5cmd69e/POO+/w2WefMWfOHEaMGEGpUqXw8PDI09Z25mcnJCSwefNmbt26RZ06dahZsyanTp3i9u3b+Pr6sm/fPgoXLsywYcPsYvyII5JTAAXAPy8Tyry86tixY9SvX58jR45w6tQpZs+ejcFgYOfOnVStWpWJEyfSrFkzmz10459djYUKFaJx48ZcuXKFgwcPkpycTO3atRk7diy7du1i165dwMNbGZcvX2bcuHF07dqVtm3b4u7uTvv27Rk6dCj79u3j66+/plChQgwZMoRChQrZtPVvsViIiIjg+++/p2fPnpQvX57PPvuM9evX4+npSZ06dXj77beZNWsWFy5cIDo6+pHLl3V9enl58corr9CmTRs2btzI6tWrCQgI4IUXXmDo0KGsWLGCmjVr8vfff2MwGGzSest8gNSrr75K2bJlmTBhAlevXqVr1640btyYL7/8kqNHj2YLPbb6jq5evcqQIUOIjIyke/fu7N69mzlz5mA2m3n99dfRarVERkYyYsSIPK1gVFXFz8+Phg0bsnr1alavXk2xYsXo06cPf/31F/Pnz8fLy4t33nmHM2fOcPr0aQYPHkxgYGCuBPe0tDQWLlxIYGCg9UqRL7/8kmnTphEaGspbb71FyZIlra1+IE96abL2mKWlpfHVV1+xZs0a/vrrLz788EN+/PFHSpQoQalSpRg4cCDjxo2jW7dulC1b1i4eDOaIpAegAMgczHT58mWCg4PZsGEDPXv2pHLlyvzwww8YDAYmTZrE008/zbp16zhw4ACVKlWy6aVXWSuqyMhI4uLicHd3p3PnzqSnp7NhwwY8PT158cUXqV27NvPmzaN48eLW9zyonJcvX8bNzY1mzZrh7u6OxWLB3d2dFi1a0KZNG/R6Paqq2vzpeIqicOLECebNm8dTTz1FmzZtaNWqFYULF2b58uWYzWbatm2Lp6cnYWFh+Pj4kJSU9EjzyHrN+rlz50hJSSEoKMjaE7Bx40Y0Gg2dOnXCw8ODsmXL0qxZM3766Sf+85//4OrqmmcHcaPRyL59+zh37hx+fn40aNCAfv368eWXXzJp0iQGDRpEly5dqFy5MpUqVcqXS+xOnTqFoij06dOHkJAQIiIiGDx4MPPnz+ett95izpw5mEwmm1xqptVq6dy5M4qisHLlSlRVpV27diiKwtKlS/n888/p3bs3X3/9NRqNJtdCu6IoWCwWSpUqRfXq1dm6dSs3b95k0KBBLFq0CIBhw4YxcuTIXJnfg1y6dInQ0FBcXFxIT0/n+++/JzU1lQkTJlCiRAlWrVrF3LlzMRqNzJs3j+PHjxMSEpKnvTLOQAJAAaAoCtHR0UyZMgWz2cxTTz2F2WymQYMGrF+/HldXV06ePMmRI0dYt24d7777LkFBQfmy4xw6dIhZs2ah1+tJSEigbt26vPXWW6iqyvLly9FoNLRv357y5ctnaxE8iIuLC9evXycyMpJnnnkGjUZDbGwsP/30E23bts3XS/9cXV1JTEzk0qVLREVF8dRTT9GlSxcsFgtr166lYsWKVK5cmcTERIoXL07hwoUfeR6KorBmzRrWrVuHp6cnGo2G3r1707FjRwA2bNiA2WymS5cuaLVarl69SlBQUJ72+iiKwtq1a9m4cSPPPPMMO3bsYNu2bYwaNYo333yT+fPns2jRIj7++GPq16+fZ+X4N3q9njt37pCRkYGqqlSpUoWWLVvyxRdf4OfnxzvvvGOT3jFFUVBVFb1eT4cOHbBYLKxcuRKAtm3boigKixYtIjw8nK5duz5xmL3fJaldunTh4sWL/PLLL4wfPx5vb2/27duHyWTCZDJZy5gX0tPT+frrr3njjTcoWrQoycnJ7Nixg6ioKNq0aYOnpyf/+c9/sFgszJ07F09PT1544YU8KYuzkUGABUDmpVbDhg3jl19+Yc6cObRr1w5XV1cuX77MZ599RlRUFAEBAbz66qvUqVPHJl1mmXdT0+l01nOYw4YNo0GDBjz//POcOXOGqVOn0rp1a/r168eSJUu4fPkykyZNQqPR3FPxZ3arZ3aXFy5cmJIlSzJ8+HAuX77MkCFDCA8PZ8GCBVgsFkaOHImPj4/NrlXO/Jc5elun03Hp0iWGDx9OkSJFGDhwICVLliQ9PZ0rV64QHh6Ou7s76enpZGRk4OPjk6Mu3awH8H379jF9+nSGDBnCrVu3GDZsGCVKlGDYsGHUrFmTlStX4uPjQ8eOHdFoNNbTDEWLFs2zcR+RkZGMGTOGfv364eLiwoABA3Bzc6Ns2bKMGzcOgCtXrlClShXc3Nxs+v2YTCaMRiMuLi7cunWLyZMnYzKZmDZtGjqdji+//JLg4GBq1aqV54PKMst0/fp1Dh8+jNlstvbMrV27ltWrV9O1a1fatWvHyZMnCQ0NfaJTEVkP9Tdv3mT58uVcvnyZJk2aUKtWLa5fv87AgQP56KOPrIOER48ebT11llfrwmKxcOXKFby8vIiJiaFEiRKcP3+e6dOn4+npyYQJEwgODiY5OZnNmzdTrlw5nn322Twpi7ORAODAsn51aWlpnD59ml27dvHLL7/Qt29fWrZsaW1pZ1aEWq3WZgfcY8eOsXr1at544w2KFSvG1atXGThwICNHjqR69epYLBZ2797NjBkzmD9/PiVLlszW4s9azsxlPXToEFOnTsVkMpGWlkaXLl146aWXrDcr8fHxoVixYowaNcqmvRwWiwWTycSWLVvYuXMnsbGxtGvXjoYNG2KxWPjwww8JDAzkvffeo3Tp0vf9jJyU1WQyodFosFgsZGRkcPjwYaKjo6lSpQqTJk2iYcOGnDp1ir/++ovp06dTt27dbK23vHgGxD8/OyUlhe3bt1vP+WdewTFgwACee+45FixYgFartX7XttoeL1y4wNKlS7lw4QIlSpSgR48emEwm5s+fT2RkJKGhoVy9epUFCxZQokQJm1yJkFnRpaenYzKZiIuLY9SoUdSsWZPVq1ezatUqPvzwQxo0aPDE6yrzezIajXz00UfExcURHh5OREQEVapU4fXXX2fz5s1s3LiRgIAARo0aReXKlW3y3BBVVTl37hz9+/fnhRdeoF+/fly8eJHp06fj7u7OmDFjCAsLs04v3f65Q04BOKjMnTk+Pp7jx4+TkJBA9erV6d+/P56enixdupTk5GT8/f05dOgQH330Ee7u7jYto4eHB5cvX2by5MlMnDgRX19fwsPD+emnnyhRogRBQUGULl0as9lMenp6tp36fq3/48ePs3HjRl566SWaNm3K3r17WbhwISaTiVGjRnHt2jXMZjNhYWHWO+vZiqIo/Pbbb6xatYoWLVpw+fJlxo8fT48ePXj99dcZPXo08+bNs54KeJzWd2pqKrt37yY8PJzjx4+TkZFBt27deOaZZ1i+fDnly5enR48ebNiwgfPnzxMbG4uqqvftTcktWQf8Xb16FZ1Oh5+fH61bt2br1q24urrSrFkzoqKiaNiwIbVq1bKWxZZXnsTHxzN79mz8/Pzo0aMH7777LlFRUQwcOJCPP/6YP/74g7S0NKpVq0bx4sVtUsFYLBY2bdpEkSJFePPNN5k5cybnz59n2rRpjBo1ig4dOhAUFESRIkVyJShl3otixYoVXL16lalTp1KiRAkOHjzI3LlzOXjwID179qRdu3Z4enoSEhJifV9eyHrvDovFQsmSJRkxYgQzZ85Eq9XSt29fPvjgA2bOnMm8efMYP3683Po3l0kAcGBxcXFMnz6d2NhY0tLS+Pnnn+nRowfdunXD1dWVH3/8kYyMDN59912bD4LL3KE/+OADRo8ezejRo5k4cSIvv/wy06ZNIz09nXr16vHHH38QFhZGsWLF7nu+P/MgYTab2bZtG2vWrOG///0vhQsXpm3btuj1eqZOnYrZbObtt9/Ol/uTZ7awo6OjeeGFF3jmmWdYt24dLVu25Ndff8VsNtO3b1/GjBnz2MHEYrGQlJTE4cOHrYPTJk+ejEajwdPTk9TUVG7fvs2VK1fYu3cvr7/+unUQWV7J2gO1cuVKFi9ejE6no3Llyrz11lv4+/tz7do15s2bx8WLF2nUqBE9evSw+SWnADExMQQHB9O3b1+WLFlCvXr1cHNzY8aMGXzwwQfWMROQ9y1duHsFQkpKCrVq1cLDw4NVq1aRnJzM3LlzGTduHP379+err76iVatWTxTg/tk74+LiwvHjx4mIiCA5Odl6o6GWLVuyb98+2rRpY+2hyovOYVVVSUhIIDU1lcKFC5OYmMi6deuIiIiwXp46YsQIRo0ahaqq9OvXj48++ojU1FS7eWZJQSKXATqYrOcyM28hO2nSJDp06MBvv/3GzJkziYiIoEuXLnzxxRd89dVXtGjRwuaXyWg0GiIiIpg/fz7u7u7s2rWL4cOHU7x4cUaPHk1ycjKLFi0iISGBefPm3XewXtbWpVarpVOnTjRt2pTZs2cTFRWFXq+nVatWjBkzhitXrtj09qRZD44WiwVFUWjdujWNGjXiq6++okmTJowfP57q1auzceNG66hlT0/PxzqYHzhwgIkTJ1KsWDESExPx9fW1hjoXFxfq169PbGws77zzDmFhYXTq1ClPW0tZu5MPHTrEDz/8wLhx4xgxYgRpaWmMGzeOQoUK0b9/f27dukXTpk3p2bOn9SBuqxv9ZN4qt1ixYrz++uv89ddfnDlzhsmTJ/Pyyy9z+vRp1q9fj9FotNnpiOvXrzNlyhT27NlD7dq1CQ4O5s8//6R379489dRT1KhRg1dffZXChQs/0Sm7rNtoRkYG6enp+Pv7M23aNCpXrsynn37KmTNnuH37Nqmpqfj4+GSrZHN7fWReHbJy5UoGDhzIyZMnWbFiBVu3bsXDw4Phw4fzzTffUK5cOaZPn86GDRtYsWIFTz/9NDVq1LDZ6UtnImMAHEDmV5R5/jchIQGdTkdERAQBAQFcvHiRb7/9lp49e/Lf//4XvV7PRx99lK271dYMBgMfffQRAQEBvPnmm0RHRzN79my8vLwYO3YsQUFBGI1G3Nzc0Gg01q7qrMtssVg4ceIEFy5csN6eNDY2lnnz5hETE8O4ceMoV66c9dHBtgw5meXbt28fMTExlCxZkurVqxMbG8vQoUNp3749Tz/9NN9//z116tShXbt2T9SSu3r1KjNnzuTatWt069aNPXv2kJCQwBtvvEHNmjVJTEzk9u3bmM1mawsuryq05ORkIiMjCQoK4saNG0ybNg1vb2/mz5+PTqfjwoULjBgxghYtWtC7d2/r+2x9zj86Opo9e/bg5+fHc889h6+vL9988w1///037733Hr/99hu3b9/mlVdesXaz51VZ4O73cfPmTT7//HPu3LnD4MGDCQ0N5ebNmwwePJiQkBBCQkK4evUqw4YNIzQ0NFfO+V+6dInNmzeTlJREuXLlaNSoEUlJSXz88cdcunSJBg0acOvWLbp160bt2rXz7Bp/uHsa6/Lly3z55ZdERkYSFhZGjx49qFWrFr/88gtz586lefPmdOvWjfPnz2MymXjuueds3oPpLKQHwEHcuXOHadOmcezYMSZMmMChQ4eoXbs2xYoVY+3atbRv354WLVpQo0YNwsLCbP7428wKMevviYmJBAcHExwcTNWqVenTpw/79+/nvffeIy0tLdtd7+7XLXzw4EHGjh3L3r17+fLLL5k1a5Z1IF3hwoUZPXo058+fR6vV2ryHI3ME/tSpU9myZQvDhg1j06ZN+Pn58fTTT/P1118zZswYAgMDad68+RN3e4eFhTFgwACKFSvGzp076dy5M4GBgSxYsIC1a9cyceJEzp49S5kyZawH8Lyo0FJSUvj666+ZP38+Fy5cICQkhICAAPbt22d9OE2pUqWoVq0akZGR1kvIIO8CSVaZLf/4+HhGjRrFrl27WLx4MSNHjiQtLY1SpUpx/vx5hg8fzq+//srzzz+fZ5X/P298lZycjMlkYseOHfz999/cuXMHgICAAN555x3g7rM7evTokWv36Lh69SozZszg2rVr+Pv7M2fOHL755hv8/PwYNWoUhQsX5tChQ7z55pvUrl37ied3P1m/k+nTp7NkyRJat27NU089xbZt24iLi8NisdCkSRP69+/Pjz/+yLp166hVqxb169eXyj8PSQ+AHcvaBW6xWHj11Vc5cOAALVu2ZMqUKQQGBpKQkMC7775L5cqVCQ8PZ+PGjUydOpXixYvb7Fxr1k3o2rVrJCcnU7JkSZYuXcrevXsZMmQI5cuX5/z58yxcuBB/f3+GDx9+T5de1s+5ceMGQ4YMoWPHjlSsWJH333+fyMhI2rdvz4cffsidO3c4evQoDRs2xN/f32bLmfnPYDCwZs0afHx8qFevHtu3b2fGjBl8+OGHNG/enGPHjuHm5kbFihXR6/W5MhDPYrFw7do1ZsyYgaqq9OjRg61bt/Lnn39St25d3n777Ty9x7/ZbGbhwoX89ttvDB06lAoVKqDVaklJSbFeivnee+9ZB3M9//zz9O3b12atfrj7Hd24cYODBw9y/Phx3n77bWJiYpgyZQomk4k5c+ZYt9EiRYpQpEiRPDu3rKoqMTExfP3117z44ossXryYN998k/T0dMaOHcuzzz7LwIEDCQoKsl7VAVi3l8ed55UrV4iJiaFChQps27aNkydP0qNHD1atWsXGjRuJi4ujb9++dO/eHYPBwPvvv0+hQoUYNWpUrt8cLHN/iY2NZfjw4ZQpU4YuXboQFhaG0Whk7NixREVFWZ/HoKoqe/bsITAwkEqVKkmXf15ThV2yWCyq2WxWY2Ji1P3796vx8fHqli1b1Bo1aqh9+vRRz5w5o5rNZtVisahbt25Vu3Xrpnbp0kXdsmWLajKZbFrOzH+nT59WR44cqdatW1fdu3evGhkZqX7wwQdq9+7d1UWLFqkjR45UR44cqSYnJ1vf88/PMhgM6smTJ9Vz586py5cvV2NiYtSBAweqEydOVLdt26Y+/fTT6ocffqimpqbe835bLKfBYFA3bNigDh8+XK1bt6767bffqmazWTUajerKlSvVpk2bqj/88INqNBqt32FultNisaixsbHq+++/rw4cOFA9deqUeufOHdVsNuf6vP7p1q1baqdOndS9e/eqFotFvXDhgrp69Wp11qxZ6tatW9X3339fDQ8PV7t3767Onj1bNRqNeV6mfzKbzeovv/yiPvPMM2r37t2t8798+bLas2dPtXfv3uq1a9fuu/3lNovFop46dUqtW7euWqpUKXXMmDFqamqqajab1QMHDqgdOnRQx48fr0ZHR+faejIajeqaNWvUF198UV29erUaGRmpRkVFqevWrVM7deqknj59Wp06dapaoUIF9ZtvvlGNRqMaExOjvv322+r27dvzZJ1kZGSon3/+uTpw4EDrvp+5vDdv3lQHDRqkdurUSf3rr79Uk8lkk+9G3CWnAOyQ+v8t4czz3D///DNJSUnUqlWLRYsWoaoqkyZNIjo6GlVVCQsLY+LEicybN4/nn38+X0bLHjlyhHHjxuHr60vhwoX5+OOPiYmJsV7Xe+TIEfz8/HjjjTfQ6/XA/1qpmdfQA5w/f57PPvuMlJQU2rZty/nz50lJSaFHjx6UKFGCevXq4e/vn6cPI3mY6OhoNmzYQFBQENWqVWP58uUcO3YMjUZD27ZtefPNN63LpChKrl+CpygKhQoVYuDAgfj4+HDt2jXrwMK8bmmr/9+6jo6OZuPGjUycOJEtW7Zw4cIFpk6dSps2bXjppZdISkrihRdesEmZ/klRFGrWrMnQoUOJjIxk/fr1KIpCeHg4I0eOpHz58qSkpNhkXQGULVuW1q1bExQURFJSEkaj0TryfsSIERw+fJgtW7ZgNpufuDzq/w+Wbdy4Ma1bt2bFihUcOXKEQoUKsXv3burUqUOZMmUA6NOnDxUrVkSj0RASEsLMmTOpV69enqyT9PR0Tp8+TfXq1XFzcwP+d8ov8xG/JUuW5NNPPyUjI8Pm24wzk1MAdkhVVW7fvs2gQYMoV64cvXv3JiAgwHrf7oSEBN5//33MZjN169Zl7dq1LF26NNu9823JaDQyceJEVFXlgw8+wGAw8PXXX7Nx40ZmzJhB9erVSUtLw9XVNdv95xVFwWQysWnTJuLj42ncuDHLly/H09OTfv364eHhwf79+xk9ejRvvPEGf/75J97e3gwaNAhvb2+b3kFOURR2797N4sWLKV26NEOGDCEjI4MZM2Zw5swZJk2aRLly5awP2dFqtXl6/b3FYuHOnTu4ublZB1LmNZPJxOeff86SJUvw9vamZ8+eNG7cGE9PT95++21atWrFSy+9xOjRo7l06RJffPEFRYoUydMyZX4/N2/e5PDhwxgMBmrUqIGvry8bNmxg8eLFvPfee9ZLIg0GAy4uLnkakjPLdO3aNaKioihcuDCxsbHMnj2b4sWLM2XKFDQaDdeuXcNoNKLVanPl3gNZD+UpKSksWbKEnTt30qtXLy5dusTixYupVasW586dY8WKFQQGBlqnz8t9KSkpidGjR+Pv78+wYcOsIQDuNnIWL15M69atcXV1td7OW9iGrGk7pCgKR44cwWg08t577+Hl5UVERARffPEF48eP58SJE8yfP59SpUpx5MgRxo8fT9GiRW1WvswDXNbfExISMJvN6PV6AgMDreejP/roIyIiIvDy8kKv12er/DMyMvjxxx/573//S7ly5di1axfffvstsbGx1rDzzDPP0LlzZ3744QcUReHtt9+2aeUPdyu+ffv2odfruXLlCgcOHCA2NhY/Pz8+/PBDypUrx8iRI0lJScHNzQ1XV9c8rfzhbgvKz88v2zrNa1qtlrffftt6t7jXXnuN4sWLk5ycjJubG2XKlMHX15c5c+bQsGHDPBuI+s/tLyoqirFjx7Js2TJWrVpFz549OXPmDB06dOCNN95g7ty5HDp0CEVR0Ov1Nukhu3DhAh9//DG//PILHh4e1K5dm9GjRxMZGckHH3zAZ599xpgxYwgKCsq1Gw9l/QxPT0969+5NkyZN+PrrrylcuDADBgwgPDyczz//nICAgHvuEZBX9Ho9lSpVYvv27ezbtw+DwQDcHVNy+vRpkpOTCQkJoVKlSlL525j0ANipbdu2MWHCBIYNG8axY8c4ffo03t7eWCwWoqKimDNnDiVKlMBgMDzRoKFHlXVzuX37NgaDgeDgYLZv387s2bN5++23ad68OWazmaFDhxIfH4+npyeffvpptorbaDSyfv16Pv/8c6ZMmULt2rU5d+4cY8eOJTExkZ49e9KhQwdcXFxQVZU7d+6g1+utpw9suaynTp1i/vz59OzZE19fXz766CNKly7N+++/T1hYGDdu3GDPnj28+OKLTnWnsqSkJH7//XfWr19PyZIlGTx4sDX45OUldaqqkpSUREpKCgEBASxatIj4+Hjrg4Z++uknihYtypgxYyhdujS7du2iXLlylC1b1ibfjdlsZubMmSQmJjJo0CCSk5M5ffq09bK+b7/9FqPRSI8ePahWrVqeXYEAkJCQwPfff8+OHTvo3bs3TZs2xc3N7Z7LbvNafHw8M2bM4Pfff+fll1+mVKlSxMfHs2nTJrp168YLL7wgj/TNBxIA7MD9knhCQgIzZszg77//pmjRorz00kvUqFGDK1euMHLkSCZPnkzFihXz5c53AIcPH+azzz7j2rVrvPjii7z44ots3LiRjRs3UrRoUeu0HTp0YP78+Xz//ffZHsyTmprKpEmTOH78OJ988ol1ZHTXrl2JiYlh586dvPbaa7Rt2/aedWNLERERfPzxx1SqVImhQ4fi4eHBsWPHGDt2LBUqVGDo0KHWrlRbtajsRUREBJ9++in16tXjlVdesV7WmZcVmqqqREZG8umnn6LT6RgwYADnz58nODiYffv2ceTIEXr37s24ceO4desWy5Yto0yZMnneI5N13zAYDAwcOJCwsDBCQkI4cuQIly9fJjAwkI8//pjQ0FDMZjPe3t55WglnlikxMZFly5Zx8OBBRowYQbly5QDbbqfq/18GuHz5cjZt2gRAmTJlaN++PY0aNcp2WkDYjlxgmc/U/79eXq/XZ3symq+vLx9//DHJycno9Xrc3d0xmUzWA4mfn59Ny2ixWKwHq+joaFauXEn9+vUpXLgwixcvJjY2lr59+1KtWjX27duHi4sLXbp0YevWrYSFhd1zsHF3d2fAgAF88sknjBw5ksTERNq1a0ebNm2s59G/+OILQkJCqFWrlk2XFbIPUMx8nHJmb0blypWZMWMGAwcO5I8//sj3kJJfKlSowMyZM613kLPFwLqzZ88ycOBAunXrRqtWrQgMDKR48eLExsZy8OBB3nrrLcqUKcNzzz2Hqqp4enrabFDs9evX+f3336lTpw4dO3Zk5syZlCxZkldeeYXChQszbdo0jEbjY98N8nH5+vrSq1cvdu3axd9//20NALakKAqBgYH079+fXr16WcfJ2Gr8irg/CQD5TFVVpkyZwjPPPEOnTp3w8PCwHhxcXV2trcszZ85w+PBha5dZ4cKFbXYefP/+/QQGBlKmTBlSUlKYOXMmycnJNG3alOLFixMcHMzs2bNZtGgRvXv3ZvDgwaSnp/Pjjz+ydetW+vbti5eX1z0P+ylcuDBDhgxhwoQJJCYmUr16dVxcXNDpdLz22muEhITY9GCV9dxy5tUHYWFhDBs2jDlz5jBv3jwGDx5MWFgYpUuXZunSpQQFBd33GQbOwMXFhcDAQJv0fCiKQlpaGkuWLKFp06bWZwpkztvFxQWj0cjPP//M/v37iYuLY8SIEYSGhuZZmTJlliE5OZmVK1eyfft2Ro8ezapVq/D09CQuLo6ff/4ZT09PfH1987w8mTK/j8ztOjAw0PpEvfzYXjMr/cxjgXQ+5z+JXvkos+KoUqUKX3/9tfXhPZl/z7qTnjp1ioiICPr06UPz5s1tenesEydOoNfrrS3zQoUKcezYMY4cOYKqqlSrVo0hQ4bw119/sWfPHuv7goODGTx4MM8999wDU35wcDBDhw6latWqLFmyhHPnzqHRaPDw8KBTp054e3vbajGBuwepDRs2MHLkSD755BM++ugj0tLSePfdd0lISGDWrFlcv34dRVEIDg6W1gu2ucMf3B1zcubMGZo2bQqQLXj5+/vTpUsXUlNTiYuLy9W76f2bzEtYS5YsyahRo4iPj2fcuHFYLBYMBgNffPEF+/fvp1evXjYL7lkpioJOp6Nnz575envwzLLY8s6Q4l888Z0ExGOzWCyq0WhUk5KS1M8//1x97rnn1LVr16rp6en33AgjNTVVjY+Pt+lNfkwmkxodHa3evn1bvXjxojp58mT10KFD6q1bt9QpU6aobdu2VXfu3Gm94cvly5fVxMRE600+TCaTajab/3U+me/98MMP1ddff12Njo62wdL9T9Ybjxw5ckRt1aqV+vvvv6u//vqrWqNGDXXLli2qwWBQ//zzT7VLly7q8uXLbVo+cVdsbKzaqVMndeHChdluGGOxWNQTJ06oAwYMUK9du6YmJyfnaLt7VFm3k8yfb926pb7//vvqr7/+qppMJtVkMqmnT59W27Rpo/bv31+NiYlRT548qUZHR9t0331Y2YXIJM2XfJaSksKUKVM4ffo0RYoUYcqUKaxbt47U1NRsXWTu7u74+/vb9CY/586dY/jw4aSlpaEod58l/vnnn3PhwgXefvtt6tevz+zZs/n1118xmUyEh4dnG+mfeS38v9FoNISHhzNkyBDq1atn0/ENaWlpREdHExUVhcFgsD7YR6/Xs3DhQvr06UP9+vU5ceIExYoVY8aMGbzyyivSfZkPfH19qVmzJt9//z0HDx60Xk6WmprKwYMHCQoKwsfHB09PzzzrmTGbzVy7do1bt25hsVjQarWULl2a8ePH88svv2Aymay3u920aRMbNmygTJkyeXrL4ZyQ1ra4HxkDkI8UReHo0aPs37+fL774goCAALZs2cKCBQtQVZX27dvj7u6ercvMlu7cuYNWq8Xd3Z2QkBAGDRrE3LlzrRVjv3790Gg0nDhxgkaNGuHq6vrY81IUhaCgIHr16mWzA2VKSgpr1qzh5MmTlClThm7duhEQEMClS5eYNWsWzz//PF27diU+Pp4lS5bw+uuvU7lyZZuUTdzLzc3NelObESNG0L59e8LDw4mMjOTYsWMMHDgQd3f3PC3DlStXGD16NB4eHrRt25YqVarQo0cPXF1dmT17Nunp6TRu3Jjk5GR69+7N888/n+dXIAjxuCQA5JPMFuTNmzdxdXXlqaeewmw206lTJ44dO8bHH3+Mp6cn7dq1s3nZEhIS8PX1JSUlxfqAEkVRKFGihHXQ3qJFi3jrrbd4++23ycjIsF5b/CQHusxBQrZgsViszyJ///33KVu2LNevXycqKopKlSrx22+/0bt3b65du8Ynn3xCcHCw9TG7mWUVthccHMy4ceNYsmQJGzduRFEUatWqxQcffGCTu8gpikJoaCjh4eHs3r2bhQsX0r59e1588UUURWHmzJksWLAAs9nM3LlzKVWqlGwrwm7JfQCEEEIIJyRjAIQQQggnJAFACCGEcEISAIQQQggnJAFACCGEcEISAIQQQggnJAFACCGEcEISAIQQQggnJAFACCGEcEISAIQQQggn9H9KLEr6rSO+/wAAAABJRU5ErkJggg==\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "You can also provide additional instructions. For example, imagine you want to use different colors for each bar. You just need to ask to PandasAI:"
      ],
      "metadata": {
        "id": "J0lqXjcq5QyN"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "sdf.chat(\"Plot a histogram of the gdp by country, using a different color for each bar\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 334
        },
        "id": "U5pJRgyY5QR2",
        "outputId": "ba021037-b019-4c31-8ea0-e00db13c299c"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# SmartDatalake\n",
        "\n",
        "Sometimes, you might want to work with multiple dataframes at a time, letting the LLM orchestrate which one(s) to use to answer your queries. In such cases, instead of using a `SmartDataframe` you should rather use a `SmartDatalake`.\n",
        "\n",
        "The concept is very similar to the `SmartDataframe`, but instead of accepting only 1 df as input, it can accept multiple ones."
      ],
      "metadata": {
        "id": "JvFf3OQPJ2T5"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from pandasai import SmartDatalake"
      ],
      "metadata": {
        "id": "Xf-q8r35KYL1"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "For example, in this example, we are provided with 2 different dfs.\n",
        "In the first one, it's reported for each employee, an employee id, a name and a department.\n",
        "In the second one, instad, it's provided the employee id and the salary for each employee.\n",
        "\n",
        "Asking PandasAI, it will join the 2 different dataframes by id and figure out the name of the one that is paid the most."
      ],
      "metadata": {
        "id": "yX3Ph1yYKfst"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "employees_df = pd.DataFrame(\n",
        "    {\n",
        "        \"EmployeeID\": [1, 2, 3, 4, 5],\n",
        "        \"Name\": [\"John\", \"Emma\", \"Liam\", \"Olivia\", \"William\"],\n",
        "        \"Department\": [\"HR\", \"Sales\", \"IT\", \"Marketing\", \"Finance\"],\n",
        "    }\n",
        ")\n",
        "\n",
        "salaries_df = pd.DataFrame(\n",
        "    {\n",
        "        \"EmployeeID\": [1, 2, 3, 4, 5],\n",
        "        \"Salary\": [5000, 6000, 4500, 7000, 5500],\n",
        "    }\n",
        ")\n",
        "\n",
        "lake = SmartDatalake(\n",
        "    [employees_df, salaries_df],\n",
        "    config={\"llm\": llm}\n",
        ")\n",
        "lake.chat(\"Who gets paid the most?\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        },
        "id": "3GIglds_KbHo",
        "outputId": "01bdbac2-a5a7-4488-f3dc-4aa179b42cd4"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'The employee who gets paid the most is Olivia.'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Here's an example of the code that is generated:"
      ],
      "metadata": {
        "id": "FVElosbbK-_i"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "print(lake.last_code_executed)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "jyaxD4N4LEl2",
        "outputId": "ed1dc2d9-126a-49e5-aaba-a84fcae3ae63"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "def analyze_data(dfs: list[pd.DataFrame]) ->dict:\n",
            "    \"\"\"\n",
            "    Analyze the data\n",
            "    1. Prepare: Preprocessing and cleaning data if necessary\n",
            "    2. Process: Manipulating data for analysis (grouping, filtering, aggregating, etc.)\n",
            "    3. Analyze: Conducting the actual analysis (if the user asks to plot a chart save it to an image in exports/charts/temp_chart.png and do not show the chart.)\n",
            "    4. Output: return a dictionary of:\n",
            "    - type (possible values \"text\", \"number\", \"dataframe\", \"plot\")\n",
            "    - value (can be a string, a dataframe or the path of the plot, NOT a dictionary)\n",
            "    Example output: { \"type\": \"text\", \"value\": \"The average loan amount is $15,000.\" }\n",
            "    \"\"\"\n",
            "    merged_df = pd.merge(dfs[0], dfs[1], on='EmployeeID')\n",
            "    max_salary_employee = merged_df.loc[merged_df['Salary'].idxmax()]\n",
            "    employee_name = max_salary_employee['Name']\n",
            "    return {'type': 'text', 'value': f'The employee who gets paid the most is {employee_name}.'}\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Ok in this case it was easy: both the tables are share a common value called `EmployeeID`, right?\n",
        "\n",
        "Let's try with something more complex"
      ],
      "metadata": {
        "id": "wr-xjtHvMf_q"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "users_df = pd.DataFrame(\n",
        "    {\n",
        "        \"id\": [1, 2, 3, 4, 5],\n",
        "        \"name\": [\"John\", \"Emma\", \"Liam\", \"Olivia\", \"William\"]\n",
        "    }\n",
        ")\n",
        "users = SmartDataframe(users_df, name=\"users\")\n",
        "\n",
        "photos_df = pd.DataFrame(\n",
        "    {\n",
        "        \"id\": [31, 32, 33, 34, 35],\n",
        "        \"user_id\": [1, 1, 2, 4, 5]\n",
        "    }\n",
        ")\n",
        "photos = SmartDataframe(photos_df, name=\"photos\")\n",
        "\n",
        "lake = SmartDatalake([users, photos], config={\"llm\": llm})\n",
        "lake.chat(\"How many photos has been uploaded by John?\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "VjJc8HH9MfNH",
        "outputId": "90712d31-8fe0-4f3d-b550-0daa093cf840"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "2"
            ]
          },
          "metadata": {},
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "In this case we provided for each df a table name, so that the LLM has some context and can better perform the joins. As you can see on the example below, it succeeded at figuring out the right join to do. In fact, the user \"John\" has actually 2 photos."
      ],
      "metadata": {
        "id": "pstO0KLvNjK9"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "print(lake.last_code_executed)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "7KNnAKWyNenA",
        "outputId": "5056fd38-1438-420f-8ca7-602d200a0d86"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "def analyze_data(dfs: list[pd.DataFrame]) ->dict:\n",
            "    users = dfs[0]\n",
            "    photos = dfs[1]\n",
            "    merged_df = pd.merge(users, photos, left_on='id', right_on='user_id')\n",
            "    john_photos = merged_df[merged_df['name'] == 'John']\n",
            "    num_photos = john_photos.shape[0]\n",
            "    return {'type': 'number', 'value': num_photos}\n",
            "\n",
            "\n",
            "result = analyze_data(dfs)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Different LLMs\n",
        "\n",
        "Although at the moment OpenAI GPT3.5 and GPT4 are the recommended models, we also support other models, like AzureOpenAI.\n",
        "\n",
        "You can use them as if follows:"
      ],
      "metadata": {
        "id": "AKaXv9VApKF1"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from pandasai import SmartDataframe\n",
        "from pandasai.llm import Starcoder, Falcon\n",
        "from pandasai.llm import AzureOpenAI\n",
        "from pandasai.llm import GoogleVertexAI\n",
        "\n",
        "azure_llm = AzureOpenAI(\n",
        "    api_token=\"my-azure-openai-api-key\",\n",
        "    azure_endpoint=\"my-azure-openai-api-endpoint\",\n",
        "    api_version=\"2023-05-15\",\n",
        "    deployment_name=\"my-deployment-name\"\n",
        ")\n",
        "\n",
        "vertexai_llm = GoogleVertexAI(\n",
        "  project_id=\"generative-ai-training\",\n",
        "  location=\"us-central1\",\n",
        "  model=\"text-bison@001\"\n",
        ")\n",
        "\n",
        "df1 = SmartDataframe(df, config={\"llm\": azure_llm})\n",
        "df2 = SmartDataframe(df, config={\"llm\": vertexai_llm})\n",
        "\n",
        "print(df1.chat(\"Which country has the highest GDP?\"))\n",
        "print(df2.chat(\"Which one is the unhappiest country?\"))"
      ],
      "metadata": {
        "id": "qMYya2bZpJCp"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## LangChain LLMs"
      ],
      "metadata": {
        "id": "cPweiIeKqrXA"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "In some cases, you might want to use LangChain LLMs instead."
      ],
      "metadata": {
        "id": "VC8obpOBq1eG"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "!pip install pandasai[langchain]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "kqWtvULErC73",
        "outputId": "b0a984cf-258b-4c98-ea85-698f96557683"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Requirement already satisfied: pandasai[langchain] in /usr/local/lib/python3.10/dist-packages (1.1.1)\n",
            "Requirement already satisfied: astor<0.9.0,>=0.8.1 in /usr/local/lib/python3.10/dist-packages (from pandasai[langchain]) (0.8.1)\n",
            "Requirement already satisfied: ipython<9.0.0,>=8.13.1 in /usr/local/lib/python3.10/dist-packages (from pandasai[langchain]) (8.15.0)\n",
            "Requirement already satisfied: matplotlib<4.0.0,>=3.7.1 in /usr/local/lib/python3.10/dist-packages (from pandasai[langchain]) (3.7.1)\n",
            "Requirement already satisfied: openai<0.28.0,>=0.27.5 in /usr/local/lib/python3.10/dist-packages (from pandasai[langchain]) (0.27.10)\n",
            "Requirement already satisfied: pandas==1.5.3 in /usr/local/lib/python3.10/dist-packages (from pandasai[langchain]) (1.5.3)\n",
            "Requirement already satisfied: pydantic<2,>=1 in /usr/local/lib/python3.10/dist-packages (from pandasai[langchain]) (1.10.12)\n",
            "Requirement already satisfied: python-dotenv<2.0.0,>=1.0.0 in /usr/local/lib/python3.10/dist-packages (from pandasai[langchain]) (1.0.0)\n",
            "Requirement already satisfied: scipy<2.0.0,>=1.9.0 in /usr/local/lib/python3.10/dist-packages (from pandasai[langchain]) (1.10.1)\n",
            "Collecting langchain<0.0.200,>=0.0.199 (from pandasai[langchain])\n",
            "  Downloading langchain-0.0.199-py3-none-any.whl (1.0 MB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m1.0/1.0 MB\u001b[0m \u001b[31m6.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: python-dateutil>=2.8.1 in /usr/local/lib/python3.10/dist-packages (from pandas==1.5.3->pandasai[langchain]) (2.8.2)\n",
            "Requirement already satisfied: pytz>=2020.1 in /usr/local/lib/python3.10/dist-packages (from pandas==1.5.3->pandasai[langchain]) (2023.3)\n",
            "Requirement already satisfied: numpy>=1.21.0 in /usr/local/lib/python3.10/dist-packages (from pandas==1.5.3->pandasai[langchain]) (1.23.5)\n",
            "Requirement already satisfied: backcall in /usr/local/lib/python3.10/dist-packages (from ipython<9.0.0,>=8.13.1->pandasai[langchain]) (0.2.0)\n",
            "Requirement already satisfied: decorator in /usr/local/lib/python3.10/dist-packages (from ipython<9.0.0,>=8.13.1->pandasai[langchain]) (4.4.2)\n",
            "Requirement already satisfied: jedi>=0.16 in /usr/local/lib/python3.10/dist-packages (from ipython<9.0.0,>=8.13.1->pandasai[langchain]) (0.19.0)\n",
            "Requirement already satisfied: matplotlib-inline in /usr/local/lib/python3.10/dist-packages (from ipython<9.0.0,>=8.13.1->pandasai[langchain]) (0.1.6)\n",
            "Requirement already satisfied: pickleshare in /usr/local/lib/python3.10/dist-packages (from ipython<9.0.0,>=8.13.1->pandasai[langchain]) (0.7.5)\n",
            "Requirement already satisfied: prompt-toolkit!=3.0.37,<3.1.0,>=3.0.30 in /usr/local/lib/python3.10/dist-packages (from ipython<9.0.0,>=8.13.1->pandasai[langchain]) (3.0.39)\n",
            "Requirement already satisfied: pygments>=2.4.0 in /usr/local/lib/python3.10/dist-packages (from ipython<9.0.0,>=8.13.1->pandasai[langchain]) (2.16.1)\n",
            "Requirement already satisfied: stack-data in /usr/local/lib/python3.10/dist-packages (from ipython<9.0.0,>=8.13.1->pandasai[langchain]) (0.6.2)\n",
            "Requirement already satisfied: traitlets>=5 in /usr/local/lib/python3.10/dist-packages (from ipython<9.0.0,>=8.13.1->pandasai[langchain]) (5.7.1)\n",
            "Requirement already satisfied: exceptiongroup in /usr/local/lib/python3.10/dist-packages (from ipython<9.0.0,>=8.13.1->pandasai[langchain]) (1.1.3)\n",
            "Requirement already satisfied: pexpect>4.3 in /usr/local/lib/python3.10/dist-packages (from ipython<9.0.0,>=8.13.1->pandasai[langchain]) (4.8.0)\n",
            "Requirement already satisfied: PyYAML>=5.4.1 in /usr/local/lib/python3.10/dist-packages (from langchain<0.0.200,>=0.0.199->pandasai[langchain]) (6.0.1)\n",
            "Requirement already satisfied: SQLAlchemy<3,>=1.4 in /usr/local/lib/python3.10/dist-packages (from langchain<0.0.200,>=0.0.199->pandasai[langchain]) (2.0.20)\n",
            "Requirement already satisfied: aiohttp<4.0.0,>=3.8.3 in /usr/local/lib/python3.10/dist-packages (from langchain<0.0.200,>=0.0.199->pandasai[langchain]) (3.8.5)\n",
            "Requirement already satisfied: async-timeout<5.0.0,>=4.0.0 in /usr/local/lib/python3.10/dist-packages (from langchain<0.0.200,>=0.0.199->pandasai[langchain]) (4.0.3)\n",
            "Collecting dataclasses-json<0.6.0,>=0.5.7 (from langchain<0.0.200,>=0.0.199->pandasai[langchain])\n",
            "  Downloading dataclasses_json-0.5.14-py3-none-any.whl (26 kB)\n",
            "Collecting langchainplus-sdk>=0.0.9 (from langchain<0.0.200,>=0.0.199->pandasai[langchain])\n",
            "  Downloading langchainplus_sdk-0.0.20-py3-none-any.whl (25 kB)\n",
            "Requirement already satisfied: numexpr<3.0.0,>=2.8.4 in /usr/local/lib/python3.10/dist-packages (from langchain<0.0.200,>=0.0.199->pandasai[langchain]) (2.8.5)\n",
            "Collecting openapi-schema-pydantic<2.0,>=1.2 (from langchain<0.0.200,>=0.0.199->pandasai[langchain])\n",
            "  Downloading openapi_schema_pydantic-1.2.4-py3-none-any.whl (90 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m90.0/90.0 kB\u001b[0m \u001b[31m11.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: requests<3,>=2 in /usr/local/lib/python3.10/dist-packages (from langchain<0.0.200,>=0.0.199->pandasai[langchain]) (2.31.0)\n",
            "Requirement already satisfied: tenacity<9.0.0,>=8.1.0 in /usr/local/lib/python3.10/dist-packages (from langchain<0.0.200,>=0.0.199->pandasai[langchain]) (8.2.3)\n",
            "Requirement already satisfied: contourpy>=1.0.1 in /usr/local/lib/python3.10/dist-packages (from matplotlib<4.0.0,>=3.7.1->pandasai[langchain]) (1.1.0)\n",
            "Requirement already satisfied: cycler>=0.10 in /usr/local/lib/python3.10/dist-packages (from matplotlib<4.0.0,>=3.7.1->pandasai[langchain]) (0.11.0)\n",
            "Requirement already satisfied: fonttools>=4.22.0 in /usr/local/lib/python3.10/dist-packages (from matplotlib<4.0.0,>=3.7.1->pandasai[langchain]) (4.42.1)\n",
            "Requirement already satisfied: kiwisolver>=1.0.1 in /usr/local/lib/python3.10/dist-packages (from matplotlib<4.0.0,>=3.7.1->pandasai[langchain]) (1.4.4)\n",
            "Requirement already satisfied: packaging>=20.0 in /usr/local/lib/python3.10/dist-packages (from matplotlib<4.0.0,>=3.7.1->pandasai[langchain]) (23.1)\n",
            "Requirement already satisfied: pillow>=6.2.0 in /usr/local/lib/python3.10/dist-packages (from matplotlib<4.0.0,>=3.7.1->pandasai[langchain]) (9.4.0)\n",
            "Requirement already satisfied: pyparsing>=2.3.1 in /usr/local/lib/python3.10/dist-packages (from matplotlib<4.0.0,>=3.7.1->pandasai[langchain]) (3.1.1)\n",
            "Requirement already satisfied: tqdm in /usr/local/lib/python3.10/dist-packages (from openai<0.28.0,>=0.27.5->pandasai[langchain]) (4.66.1)\n",
            "Requirement already satisfied: typing-extensions>=4.2.0 in /usr/local/lib/python3.10/dist-packages (from pydantic<2,>=1->pandasai[langchain]) (4.7.1)\n",
            "Requirement already satisfied: attrs>=17.3.0 in /usr/local/lib/python3.10/dist-packages (from aiohttp<4.0.0,>=3.8.3->langchain<0.0.200,>=0.0.199->pandasai[langchain]) (23.1.0)\n",
            "Requirement already satisfied: charset-normalizer<4.0,>=2.0 in /usr/local/lib/python3.10/dist-packages (from aiohttp<4.0.0,>=3.8.3->langchain<0.0.200,>=0.0.199->pandasai[langchain]) (3.2.0)\n",
            "Requirement already satisfied: multidict<7.0,>=4.5 in /usr/local/lib/python3.10/dist-packages (from aiohttp<4.0.0,>=3.8.3->langchain<0.0.200,>=0.0.199->pandasai[langchain]) (6.0.4)\n",
            "Requirement already satisfied: yarl<2.0,>=1.0 in /usr/local/lib/python3.10/dist-packages (from aiohttp<4.0.0,>=3.8.3->langchain<0.0.200,>=0.0.199->pandasai[langchain]) (1.9.2)\n",
            "Requirement already satisfied: frozenlist>=1.1.1 in /usr/local/lib/python3.10/dist-packages (from aiohttp<4.0.0,>=3.8.3->langchain<0.0.200,>=0.0.199->pandasai[langchain]) (1.4.0)\n",
            "Requirement already satisfied: aiosignal>=1.1.2 in /usr/local/lib/python3.10/dist-packages (from aiohttp<4.0.0,>=3.8.3->langchain<0.0.200,>=0.0.199->pandasai[langchain]) (1.3.1)\n",
            "Collecting marshmallow<4.0.0,>=3.18.0 (from dataclasses-json<0.6.0,>=0.5.7->langchain<0.0.200,>=0.0.199->pandasai[langchain])\n",
            "  Downloading marshmallow-3.20.1-py3-none-any.whl (49 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m49.4/49.4 kB\u001b[0m \u001b[31m6.8 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hCollecting typing-inspect<1,>=0.4.0 (from dataclasses-json<0.6.0,>=0.5.7->langchain<0.0.200,>=0.0.199->pandasai[langchain])\n",
            "  Downloading typing_inspect-0.9.0-py3-none-any.whl (8.8 kB)\n",
            "Requirement already satisfied: parso<0.9.0,>=0.8.3 in /usr/local/lib/python3.10/dist-packages (from jedi>=0.16->ipython<9.0.0,>=8.13.1->pandasai[langchain]) (0.8.3)\n",
            "Requirement already satisfied: ptyprocess>=0.5 in /usr/local/lib/python3.10/dist-packages (from pexpect>4.3->ipython<9.0.0,>=8.13.1->pandasai[langchain]) (0.7.0)\n",
            "Requirement already satisfied: wcwidth in /usr/local/lib/python3.10/dist-packages (from prompt-toolkit!=3.0.37,<3.1.0,>=3.0.30->ipython<9.0.0,>=8.13.1->pandasai[langchain]) (0.2.6)\n",
            "Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.10/dist-packages (from python-dateutil>=2.8.1->pandas==1.5.3->pandasai[langchain]) (1.16.0)\n",
            "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.10/dist-packages (from requests<3,>=2->langchain<0.0.200,>=0.0.199->pandasai[langchain]) (3.4)\n",
            "Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.10/dist-packages (from requests<3,>=2->langchain<0.0.200,>=0.0.199->pandasai[langchain]) (2.0.4)\n",
            "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.10/dist-packages (from requests<3,>=2->langchain<0.0.200,>=0.0.199->pandasai[langchain]) (2023.7.22)\n",
            "Requirement already satisfied: greenlet!=0.4.17 in /usr/local/lib/python3.10/dist-packages (from SQLAlchemy<3,>=1.4->langchain<0.0.200,>=0.0.199->pandasai[langchain]) (2.0.2)\n",
            "Requirement already satisfied: executing>=1.2.0 in /usr/local/lib/python3.10/dist-packages (from stack-data->ipython<9.0.0,>=8.13.1->pandasai[langchain]) (1.2.0)\n",
            "Requirement already satisfied: asttokens>=2.1.0 in /usr/local/lib/python3.10/dist-packages (from stack-data->ipython<9.0.0,>=8.13.1->pandasai[langchain]) (2.3.0)\n",
            "Requirement already satisfied: pure-eval in /usr/local/lib/python3.10/dist-packages (from stack-data->ipython<9.0.0,>=8.13.1->pandasai[langchain]) (0.2.2)\n",
            "Collecting mypy-extensions>=0.3.0 (from typing-inspect<1,>=0.4.0->dataclasses-json<0.6.0,>=0.5.7->langchain<0.0.200,>=0.0.199->pandasai[langchain])\n",
            "  Downloading mypy_extensions-1.0.0-py3-none-any.whl (4.7 kB)\n",
            "Installing collected packages: mypy-extensions, marshmallow, typing-inspect, openapi-schema-pydantic, langchainplus-sdk, dataclasses-json, langchain\n",
            "Successfully installed dataclasses-json-0.5.14 langchain-0.0.199 langchainplus-sdk-0.0.20 marshmallow-3.20.1 mypy-extensions-1.0.0 openapi-schema-pydantic-1.2.4 typing-inspect-0.9.0\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Then you can use them as if they were PandasAI LLMs."
      ],
      "metadata": {
        "id": "_gMX50dXrFET"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from pandasai import SmartDataframe\n",
        "from langchain.llms import OpenAI\n",
        "# from langchain.llms import Anthropic\n",
        "# from langchain.llms import LlamaCpp\n",
        "\n",
        "langchain_llm = OpenAI(openai_api_key=\"YOUR TOKEN\", max_tokens=1000)\n",
        "langchain_sdf = SmartDataframe(df, config={\"llm\": langchain_llm})\n",
        "langchain_sdf.chat(\"Which are the top 5 countries by GPD?\")"
      ],
      "metadata": {
        "id": "kJzSEArCqtkZ",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "outputId": "cab00d4e-f260-49b9-bf86-4b7f6815800a"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "          country             gdp  happiness_index\n",
              "0   United States  19294482071552             6.94\n",
              "9           China  14631844184064             5.12\n",
              "8           Japan   4380756541440             5.87\n",
              "3         Germany   3435817336832             7.07\n",
              "1  United Kingdom   2891615567872             7.16"
            ],
            "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>country</th>\n",
              "      <th>gdp</th>\n",
              "      <th>happiness_index</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>United States</td>\n",
              "      <td>19294482071552</td>\n",
              "      <td>6.94</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>China</td>\n",
              "      <td>14631844184064</td>\n",
              "      <td>5.12</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>Japan</td>\n",
              "      <td>4380756541440</td>\n",
              "      <td>5.87</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Germany</td>\n",
              "      <td>3435817336832</td>\n",
              "      <td>7.07</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>United Kingdom</td>\n",
              "      <td>2891615567872</td>\n",
              "      <td>7.16</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ]
          },
          "metadata": {},
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Connectors\n",
        "\n",
        "PandasAI provides a number of connectors that allow you to connect to different data sources. These connectors are designed to be easy to use, even if you are not familiar with the data source or with PandasAI.\n",
        "\n",
        "To use a connector, you first need to install the required dependencies. You can do this by running the following command:"
      ],
      "metadata": {
        "id": "ehM7uOydUB_7"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "!pip install pandasai[connectors]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "rj_w3RylUdXh",
        "outputId": "e877244b-a739-474b-db35-3d8c364e9f05"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Requirement already satisfied: pandasai[connectors] in /usr/local/lib/python3.10/dist-packages (1.2.7)\n",
            "Requirement already satisfied: astor<0.9.0,>=0.8.1 in /usr/local/lib/python3.10/dist-packages (from pandasai[connectors]) (0.8.1)\n",
            "Requirement already satisfied: duckdb<0.9.0,>=0.8.1 in /usr/local/lib/python3.10/dist-packages (from pandasai[connectors]) (0.8.1)\n",
            "Requirement already satisfied: ipython<9.0.0,>=8.13.1 in /usr/local/lib/python3.10/dist-packages (from pandasai[connectors]) (8.15.0)\n",
            "Requirement already satisfied: matplotlib<4.0.0,>=3.7.1 in /usr/local/lib/python3.10/dist-packages (from pandasai[connectors]) (3.7.1)\n",
            "Requirement already satisfied: openai<0.28.0,>=0.27.5 in /usr/local/lib/python3.10/dist-packages (from pandasai[connectors]) (0.27.10)\n",
            "Requirement already satisfied: pandas==1.5.3 in /usr/local/lib/python3.10/dist-packages (from pandasai[connectors]) (1.5.3)\n",
            "Requirement already satisfied: pydantic<2,>=1 in /usr/local/lib/python3.10/dist-packages (from pandasai[connectors]) (1.10.12)\n",
            "Requirement already satisfied: python-dotenv<2.0.0,>=1.0.0 in /usr/local/lib/python3.10/dist-packages (from pandasai[connectors]) (1.0.0)\n",
            "Requirement already satisfied: scipy<2.0.0,>=1.9.0 in /usr/local/lib/python3.10/dist-packages (from pandasai[connectors]) (1.11.2)\n",
            "Requirement already satisfied: sqlalchemy<2.0.0,>=1.4.49 in /usr/local/lib/python3.10/dist-packages (from pandasai[connectors]) (1.4.49)\n",
            "Requirement already satisfied: psycopg2<3.0.0,>=2.9.7 in /usr/local/lib/python3.10/dist-packages (from pandasai[connectors]) (2.9.7)\n",
            "Collecting pymysql<2.0.0,>=1.1.0 (from pandasai[connectors])\n",
            "  Downloading PyMySQL-1.1.0-py3-none-any.whl (44 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m44.8/44.8 kB\u001b[0m \u001b[31m709.3 kB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hCollecting snowflake-sqlalchemy<2.0.0,>=1.5.0 (from pandasai[connectors])\n",
            "  Downloading snowflake_sqlalchemy-1.5.0-py2.py3-none-any.whl (33 kB)\n",
            "Collecting sqlalchemy-databricks<0.3.0,>=0.2.0 (from pandasai[connectors])\n",
            "  Downloading sqlalchemy_databricks-0.2.0-py3-none-any.whl (4.3 kB)\n",
            "Requirement already satisfied: python-dateutil>=2.8.1 in /usr/local/lib/python3.10/dist-packages (from pandas==1.5.3->pandasai[connectors]) (2.8.2)\n",
            "Requirement already satisfied: pytz>=2020.1 in /usr/local/lib/python3.10/dist-packages (from pandas==1.5.3->pandasai[connectors]) (2023.3.post1)\n",
            "Requirement already satisfied: numpy>=1.21.0 in /usr/local/lib/python3.10/dist-packages (from pandas==1.5.3->pandasai[connectors]) (1.23.5)\n",
            "Requirement already satisfied: backcall in /usr/local/lib/python3.10/dist-packages (from ipython<9.0.0,>=8.13.1->pandasai[connectors]) (0.2.0)\n",
            "Requirement already satisfied: decorator in /usr/local/lib/python3.10/dist-packages (from ipython<9.0.0,>=8.13.1->pandasai[connectors]) (4.4.2)\n",
            "Requirement already satisfied: jedi>=0.16 in /usr/local/lib/python3.10/dist-packages (from ipython<9.0.0,>=8.13.1->pandasai[connectors]) (0.19.0)\n",
            "Requirement already satisfied: matplotlib-inline in /usr/local/lib/python3.10/dist-packages (from ipython<9.0.0,>=8.13.1->pandasai[connectors]) (0.1.6)\n",
            "Requirement already satisfied: pickleshare in /usr/local/lib/python3.10/dist-packages (from ipython<9.0.0,>=8.13.1->pandasai[connectors]) (0.7.5)\n",
            "Requirement already satisfied: prompt-toolkit!=3.0.37,<3.1.0,>=3.0.30 in /usr/local/lib/python3.10/dist-packages (from ipython<9.0.0,>=8.13.1->pandasai[connectors]) (3.0.39)\n",
            "Requirement already satisfied: pygments>=2.4.0 in /usr/local/lib/python3.10/dist-packages (from ipython<9.0.0,>=8.13.1->pandasai[connectors]) (2.16.1)\n",
            "Requirement already satisfied: stack-data in /usr/local/lib/python3.10/dist-packages (from ipython<9.0.0,>=8.13.1->pandasai[connectors]) (0.6.2)\n",
            "Requirement already satisfied: traitlets>=5 in /usr/local/lib/python3.10/dist-packages (from ipython<9.0.0,>=8.13.1->pandasai[connectors]) (5.7.1)\n",
            "Requirement already satisfied: exceptiongroup in /usr/local/lib/python3.10/dist-packages (from ipython<9.0.0,>=8.13.1->pandasai[connectors]) (1.1.3)\n",
            "Requirement already satisfied: pexpect>4.3 in /usr/local/lib/python3.10/dist-packages (from ipython<9.0.0,>=8.13.1->pandasai[connectors]) (4.8.0)\n",
            "Requirement already satisfied: contourpy>=1.0.1 in /usr/local/lib/python3.10/dist-packages (from matplotlib<4.0.0,>=3.7.1->pandasai[connectors]) (1.1.0)\n",
            "Requirement already satisfied: cycler>=0.10 in /usr/local/lib/python3.10/dist-packages (from matplotlib<4.0.0,>=3.7.1->pandasai[connectors]) (0.11.0)\n",
            "Requirement already satisfied: fonttools>=4.22.0 in /usr/local/lib/python3.10/dist-packages (from matplotlib<4.0.0,>=3.7.1->pandasai[connectors]) (4.42.1)\n",
            "Requirement already satisfied: kiwisolver>=1.0.1 in /usr/local/lib/python3.10/dist-packages (from matplotlib<4.0.0,>=3.7.1->pandasai[connectors]) (1.4.5)\n",
            "Requirement already satisfied: packaging>=20.0 in /usr/local/lib/python3.10/dist-packages (from matplotlib<4.0.0,>=3.7.1->pandasai[connectors]) (23.1)\n",
            "Requirement already satisfied: pillow>=6.2.0 in /usr/local/lib/python3.10/dist-packages (from matplotlib<4.0.0,>=3.7.1->pandasai[connectors]) (9.4.0)\n",
            "Requirement already satisfied: pyparsing>=2.3.1 in /usr/local/lib/python3.10/dist-packages (from matplotlib<4.0.0,>=3.7.1->pandasai[connectors]) (3.1.1)\n",
            "Requirement already satisfied: requests>=2.20 in /usr/local/lib/python3.10/dist-packages (from openai<0.28.0,>=0.27.5->pandasai[connectors]) (2.31.0)\n",
            "Requirement already satisfied: tqdm in /usr/local/lib/python3.10/dist-packages (from openai<0.28.0,>=0.27.5->pandasai[connectors]) (4.66.1)\n",
            "Requirement already satisfied: aiohttp in /usr/local/lib/python3.10/dist-packages (from openai<0.28.0,>=0.27.5->pandasai[connectors]) (3.8.5)\n",
            "Requirement already satisfied: typing-extensions>=4.2.0 in /usr/local/lib/python3.10/dist-packages (from pydantic<2,>=1->pandasai[connectors]) (4.5.0)\n",
            "Collecting snowflake-connector-python<4.0.0 (from snowflake-sqlalchemy<2.0.0,>=1.5.0->pandasai[connectors])\n",
            "  Downloading snowflake_connector_python-3.2.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (24.6 MB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m24.6/24.6 MB\u001b[0m \u001b[31m34.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: greenlet!=0.4.17 in /usr/local/lib/python3.10/dist-packages (from sqlalchemy<2.0.0,>=1.4.49->pandasai[connectors]) (2.0.2)\n",
            "Collecting PyHive<1,>=0 (from sqlalchemy-databricks<0.3.0,>=0.2.0->pandasai[connectors])\n",
            "  Downloading PyHive-0.7.0.tar.gz (46 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m46.5/46.5 kB\u001b[0m \u001b[31m4.8 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25h  Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "Collecting databricks-sql-connector<3,>=2 (from sqlalchemy-databricks<0.3.0,>=0.2.0->pandasai[connectors])\n",
            "  Downloading databricks_sql_connector-2.9.3-py3-none-any.whl (297 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m297.3/297.3 kB\u001b[0m \u001b[31m25.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hCollecting alembic<2.0.0,>=1.0.11 (from databricks-sql-connector<3,>=2->sqlalchemy-databricks<0.3.0,>=0.2.0->pandasai[connectors])\n",
            "  Downloading alembic-1.12.0-py3-none-any.whl (226 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m226.0/226.0 kB\u001b[0m \u001b[31m22.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hCollecting lz4<5.0.0,>=4.0.2 (from databricks-sql-connector<3,>=2->sqlalchemy-databricks<0.3.0,>=0.2.0->pandasai[connectors])\n",
            "  Downloading lz4-4.3.2-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (1.3 MB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m1.3/1.3 MB\u001b[0m \u001b[31m56.0 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: oauthlib<4.0.0,>=3.1.0 in /usr/local/lib/python3.10/dist-packages (from databricks-sql-connector<3,>=2->sqlalchemy-databricks<0.3.0,>=0.2.0->pandasai[connectors]) (3.2.2)\n",
            "Requirement already satisfied: openpyxl<4.0.0,>=3.0.10 in /usr/local/lib/python3.10/dist-packages (from databricks-sql-connector<3,>=2->sqlalchemy-databricks<0.3.0,>=0.2.0->pandasai[connectors]) (3.1.2)\n",
            "Requirement already satisfied: pyarrow>=6.0.0 in /usr/local/lib/python3.10/dist-packages (from databricks-sql-connector<3,>=2->sqlalchemy-databricks<0.3.0,>=0.2.0->pandasai[connectors]) (9.0.0)\n",
            "Collecting thrift<0.17.0,>=0.16.0 (from databricks-sql-connector<3,>=2->sqlalchemy-databricks<0.3.0,>=0.2.0->pandasai[connectors])\n",
            "  Downloading thrift-0.16.0.tar.gz (59 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m59.6/59.6 kB\u001b[0m \u001b[31m6.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25h  Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "Requirement already satisfied: urllib3>=1.0 in /usr/local/lib/python3.10/dist-packages (from databricks-sql-connector<3,>=2->sqlalchemy-databricks<0.3.0,>=0.2.0->pandasai[connectors]) (2.0.4)\n",
            "Requirement already satisfied: parso<0.9.0,>=0.8.3 in /usr/local/lib/python3.10/dist-packages (from jedi>=0.16->ipython<9.0.0,>=8.13.1->pandasai[connectors]) (0.8.3)\n",
            "Requirement already satisfied: ptyprocess>=0.5 in /usr/local/lib/python3.10/dist-packages (from pexpect>4.3->ipython<9.0.0,>=8.13.1->pandasai[connectors]) (0.7.0)\n",
            "Requirement already satisfied: wcwidth in /usr/local/lib/python3.10/dist-packages (from prompt-toolkit!=3.0.37,<3.1.0,>=3.0.30->ipython<9.0.0,>=8.13.1->pandasai[connectors]) (0.2.6)\n",
            "Requirement already satisfied: future in /usr/local/lib/python3.10/dist-packages (from PyHive<1,>=0->sqlalchemy-databricks<0.3.0,>=0.2.0->pandasai[connectors]) (0.18.3)\n",
            "Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.10/dist-packages (from python-dateutil>=2.8.1->pandas==1.5.3->pandasai[connectors]) (1.16.0)\n",
            "Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.10/dist-packages (from requests>=2.20->openai<0.28.0,>=0.27.5->pandasai[connectors]) (3.2.0)\n",
            "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.10/dist-packages (from requests>=2.20->openai<0.28.0,>=0.27.5->pandasai[connectors]) (3.4)\n",
            "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.10/dist-packages (from requests>=2.20->openai<0.28.0,>=0.27.5->pandasai[connectors]) (2023.7.22)\n",
            "Collecting asn1crypto<2.0.0,>0.24.0 (from snowflake-connector-python<4.0.0->snowflake-sqlalchemy<2.0.0,>=1.5.0->pandasai[connectors])\n",
            "  Downloading asn1crypto-1.5.1-py2.py3-none-any.whl (105 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m105.0/105.0 kB\u001b[0m \u001b[31m11.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: cffi<2.0.0,>=1.9 in /usr/local/lib/python3.10/dist-packages (from snowflake-connector-python<4.0.0->snowflake-sqlalchemy<2.0.0,>=1.5.0->pandasai[connectors]) (1.15.1)\n",
            "Requirement already satisfied: cryptography<42.0.0,>=3.1.0 in /usr/local/lib/python3.10/dist-packages (from snowflake-connector-python<4.0.0->snowflake-sqlalchemy<2.0.0,>=1.5.0->pandasai[connectors]) (41.0.3)\n",
            "Collecting oscrypto<2.0.0 (from snowflake-connector-python<4.0.0->snowflake-sqlalchemy<2.0.0,>=1.5.0->pandasai[connectors])\n",
            "  Downloading oscrypto-1.3.0-py2.py3-none-any.whl (194 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m194.6/194.6 kB\u001b[0m \u001b[31m18.8 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: pyOpenSSL<24.0.0,>=16.2.0 in /usr/local/lib/python3.10/dist-packages (from snowflake-connector-python<4.0.0->snowflake-sqlalchemy<2.0.0,>=1.5.0->pandasai[connectors]) (23.2.0)\n",
            "Collecting pycryptodomex!=3.5.0,<4.0.0,>=3.2 (from snowflake-connector-python<4.0.0->snowflake-sqlalchemy<2.0.0,>=1.5.0->pandasai[connectors])\n",
            "  Downloading pycryptodomex-3.19.0-cp35-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (2.1 MB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2.1/2.1 MB\u001b[0m \u001b[31m43.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: pyjwt<3.0.0 in /usr/lib/python3/dist-packages (from snowflake-connector-python<4.0.0->snowflake-sqlalchemy<2.0.0,>=1.5.0->pandasai[connectors]) (2.3.0)\n",
            "Collecting urllib3>=1.0 (from databricks-sql-connector<3,>=2->sqlalchemy-databricks<0.3.0,>=0.2.0->pandasai[connectors])\n",
            "  Downloading urllib3-1.26.16-py2.py3-none-any.whl (143 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m143.1/143.1 kB\u001b[0m \u001b[31m14.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: filelock<4,>=3.5 in /usr/local/lib/python3.10/dist-packages (from snowflake-connector-python<4.0.0->snowflake-sqlalchemy<2.0.0,>=1.5.0->pandasai[connectors]) (3.12.2)\n",
            "Requirement already satisfied: sortedcontainers>=2.4.0 in /usr/local/lib/python3.10/dist-packages (from snowflake-connector-python<4.0.0->snowflake-sqlalchemy<2.0.0,>=1.5.0->pandasai[connectors]) (2.4.0)\n",
            "Collecting platformdirs<3.9.0,>=2.6.0 (from snowflake-connector-python<4.0.0->snowflake-sqlalchemy<2.0.0,>=1.5.0->pandasai[connectors])\n",
            "  Downloading platformdirs-3.8.1-py3-none-any.whl (16 kB)\n",
            "Collecting tomlkit (from snowflake-connector-python<4.0.0->snowflake-sqlalchemy<2.0.0,>=1.5.0->pandasai[connectors])\n",
            "  Downloading tomlkit-0.12.1-py3-none-any.whl (37 kB)\n",
            "Requirement already satisfied: attrs>=17.3.0 in /usr/local/lib/python3.10/dist-packages (from aiohttp->openai<0.28.0,>=0.27.5->pandasai[connectors]) (23.1.0)\n",
            "Requirement already satisfied: multidict<7.0,>=4.5 in /usr/local/lib/python3.10/dist-packages (from aiohttp->openai<0.28.0,>=0.27.5->pandasai[connectors]) (6.0.4)\n",
            "Requirement already satisfied: async-timeout<5.0,>=4.0.0a3 in /usr/local/lib/python3.10/dist-packages (from aiohttp->openai<0.28.0,>=0.27.5->pandasai[connectors]) (4.0.3)\n",
            "Requirement already satisfied: yarl<2.0,>=1.0 in /usr/local/lib/python3.10/dist-packages (from aiohttp->openai<0.28.0,>=0.27.5->pandasai[connectors]) (1.9.2)\n",
            "Requirement already satisfied: frozenlist>=1.1.1 in /usr/local/lib/python3.10/dist-packages (from aiohttp->openai<0.28.0,>=0.27.5->pandasai[connectors]) (1.4.0)\n",
            "Requirement already satisfied: aiosignal>=1.1.2 in /usr/local/lib/python3.10/dist-packages (from aiohttp->openai<0.28.0,>=0.27.5->pandasai[connectors]) (1.3.1)\n",
            "Requirement already satisfied: executing>=1.2.0 in /usr/local/lib/python3.10/dist-packages (from stack-data->ipython<9.0.0,>=8.13.1->pandasai[connectors]) (1.2.0)\n",
            "Requirement already satisfied: asttokens>=2.1.0 in /usr/local/lib/python3.10/dist-packages (from stack-data->ipython<9.0.0,>=8.13.1->pandasai[connectors]) (2.4.0)\n",
            "Requirement already satisfied: pure-eval in /usr/local/lib/python3.10/dist-packages (from stack-data->ipython<9.0.0,>=8.13.1->pandasai[connectors]) (0.2.2)\n",
            "Collecting Mako (from alembic<2.0.0,>=1.0.11->databricks-sql-connector<3,>=2->sqlalchemy-databricks<0.3.0,>=0.2.0->pandasai[connectors])\n",
            "  Downloading Mako-1.2.4-py3-none-any.whl (78 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m78.7/78.7 kB\u001b[0m \u001b[31m8.9 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: pycparser in /usr/local/lib/python3.10/dist-packages (from cffi<2.0.0,>=1.9->snowflake-connector-python<4.0.0->snowflake-sqlalchemy<2.0.0,>=1.5.0->pandasai[connectors]) (2.21)\n",
            "Requirement already satisfied: et-xmlfile in /usr/local/lib/python3.10/dist-packages (from openpyxl<4.0.0,>=3.0.10->databricks-sql-connector<3,>=2->sqlalchemy-databricks<0.3.0,>=0.2.0->pandasai[connectors]) (1.1.0)\n",
            "Requirement already satisfied: MarkupSafe>=0.9.2 in /usr/local/lib/python3.10/dist-packages (from Mako->alembic<2.0.0,>=1.0.11->databricks-sql-connector<3,>=2->sqlalchemy-databricks<0.3.0,>=0.2.0->pandasai[connectors]) (2.1.3)\n",
            "Building wheels for collected packages: PyHive, thrift\n",
            "  Building wheel for PyHive (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for PyHive: filename=PyHive-0.7.0-py3-none-any.whl size=53872 sha256=1d2a90767825eb44f25f15a386a3191b47df6b0fe2ee1c8b3718ad3c9e9c3592\n",
            "  Stored in directory: /root/.cache/pip/wheels/d3/fc/31/6974270c69ccc5bf8f848e2e41b527d0e8f5b9b973696a29a9\n",
            "  Building wheel for thrift (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for thrift: filename=thrift-0.16.0-cp310-cp310-linux_x86_64.whl size=373871 sha256=9b6ad9cfb506732a6582e3d6f87721c28d3255de8eeeadf47ed54445c360ad89\n",
            "  Stored in directory: /root/.cache/pip/wheels/52/f8/d2/acfd995e8247eb0cad372fa6a640a5fcf279ab2ed7c5c4490e\n",
            "Successfully built PyHive thrift\n",
            "Installing collected packages: asn1crypto, urllib3, tomlkit, thrift, pymysql, pycryptodomex, platformdirs, oscrypto, Mako, lz4, PyHive, alembic, databricks-sql-connector, sqlalchemy-databricks, snowflake-connector-python, snowflake-sqlalchemy\n",
            "  Attempting uninstall: urllib3\n",
            "    Found existing installation: urllib3 2.0.4\n",
            "    Uninstalling urllib3-2.0.4:\n",
            "      Successfully uninstalled urllib3-2.0.4\n",
            "  Attempting uninstall: platformdirs\n",
            "    Found existing installation: platformdirs 3.10.0\n",
            "    Uninstalling platformdirs-3.10.0:\n",
            "      Successfully uninstalled platformdirs-3.10.0\n",
            "Successfully installed Mako-1.2.4 PyHive-0.7.0 alembic-1.12.0 asn1crypto-1.5.1 databricks-sql-connector-2.9.3 lz4-4.3.2 oscrypto-1.3.0 platformdirs-3.8.1 pycryptodomex-3.19.0 pymysql-1.1.0 snowflake-connector-python-3.2.0 snowflake-sqlalchemy-1.5.0 sqlalchemy-databricks-0.2.0 thrift-0.16.0 tomlkit-0.12.1 urllib3-1.26.16\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.colab-display-data+json": {
              "pip_warning": {
                "packages": [
                  "urllib3"
                ]
              }
            }
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from pandasai.connectors import MySQLConnector, PostgreSQLConnector\n",
        "\n",
        "# With a MySQL database\n",
        "loan_connector = MySQLConnector(\n",
        "    config={\n",
        "        \"host\": \"localhost\",\n",
        "        \"port\": 3306,\n",
        "        \"database\": \"mydb\",\n",
        "        \"username\": \"root\",\n",
        "        \"password\": \"root\",\n",
        "        \"table\": \"loans\",\n",
        "        \"where\": [\n",
        "            # this is optional and filters the data to\n",
        "            # reduce the size of the dataframe\n",
        "            [\"loan_status\", \"=\", \"PAIDOFF\"],\n",
        "        ],\n",
        "    }\n",
        ")\n",
        "\n",
        "# With a PostgreSQL database\n",
        "payment_connector = PostgreSQLConnector(\n",
        "    config={\n",
        "        \"host\": \"localhost\",\n",
        "        \"port\": 5432,\n",
        "        \"database\": \"mydb\",\n",
        "        \"username\": \"root\",\n",
        "        \"password\": \"root\",\n",
        "        \"table\": \"payments\",\n",
        "        \"where\": [\n",
        "            # this is optional and filters the data to\n",
        "            # reduce the size of the dataframe\n",
        "            [\"payment_status\", \"=\", \"PAIDOFF\"],\n",
        "        ],\n",
        "    }\n",
        ")\n",
        "\n",
        "df_connector = SmartDatalake([loan_connector, payment_connector], config={\"llm\": llm})\n",
        "response = df_connector.chat(\"How many loans from the United states?\")\n",
        "print(response)"
      ],
      "metadata": {
        "id": "KF23V1fdVeXO"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from pandasai.connectors.yahoo_finance import YahooFinanceConnector\n",
        "\n",
        "yahoo_connector = YahooFinanceConnector(\"MSFT\")\n",
        "df = SmartDataframe(yahoo_connector, config={\"llm\": llm})\n",
        "\n",
        "response = df.chat(\"What is the closing price for yesterday?\")\n",
        "print(response)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "MrTYJHxDUBhS",
        "outputId": "37022ba3-083c-4f8c-e019-b54add277bf0"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The closing price for yesterday was $319.53.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "yahoo_connector = YahooFinanceConnector(\"TSLA\")\n",
        "\n",
        "df_connector = SmartDataframe(yahoo_connector, config={\"llm\": llm})\n",
        "response = df_connector.chat(\"Plot the chart of tesla over time\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 406
        },
        "id": "Svy1neJBVx86",
        "outputId": "218b46b0-5d63-47c3-d3d2-47db8789211b"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "You can find more info about the connectors (and many more connectors) here: https://docs.pandas-ai.com/en/latest/connectors/"
      ],
      "metadata": {
        "id": "_ijEXc8OVsPO"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Train PandasAI\n",
        "You can train PandasAI to understand your data better and to improve its performance. Training is as easy as calling the `train` method on the `SmartDataframe`, `SmartDatalake` or `Agent`.\n",
        "\n",
        "There are two kinds of training:\n",
        "\n",
        "- instructions training\n",
        "- q/a training"
      ],
      "metadata": {
        "id": "6XZa2UyS-XHF"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Instructions training\n",
        "\n",
        "Instructions training is used to teach PandasAI how you expect it to respond to certain queries. You can provide generic instructions about how you expect the model to approach certain types of queries, and PandasAI will use these instructions to generate responses to similar queries.\n",
        "\n",
        "For example, you might want the LLM to be aware that your company's fiscal year starts in April, or about specific ways you want to handle missing data. Or you might want to teach it about specific business rules or data analysis best practices that are specific to your organization.\n",
        "\n",
        "To train PandasAI with instructions, you can use the `train` method on the `Agent`, `SmartDataframe` or `SmartDatalake`, as it follows.\n",
        "\n",
        "** NOTE **: You can get your API KEY from https://PandaBI.ai"
      ],
      "metadata": {
        "id": "NckmeoK4-gXz"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import os\n",
        "\n",
        "os.environ[\"PANDASAI_API_KEY\"] = \"YOUR_PANDASAI_API_KEY\""
      ],
      "metadata": {
        "id": "fV7ojfdl_fXQ"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from pandasai import Agent\n",
        "from pandasai.llm import OpenAI\n",
        "\n",
        "llm = OpenAI()\n",
        "\n",
        "agent = Agent(df, config={\"llm\": llm})\n",
        "agent.train(docs=\"Exclude the US from the highest GDP\")"
      ],
      "metadata": {
        "id": "vTE1fXyo-WXP"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "response = agent.chat(\"Which country has the highest GDP?\")\n",
        "print(response)"
      ],
      "metadata": {
        "id": "tLl0vEz1BIBa",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "677417af-966b-49c8-9f7a-ef95872f9060"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The country with the highest GDP (excluding the US) is China.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Q/A training\n",
        "\n",
        "Q/A training is used to teach PandasAI the desired process to answer specific questions, enhancing the model's performance and determinism. One of the biggest challenges with LLMs is that they are not deterministic, meaning that the same question can produce different answers at different times. Q/A training can help to mitigate this issue.\n",
        "\n",
        "To train PandasAI with Q/A, you can use the `train` method on the `Agent`, `SmartDataframe` or `SmartDatalake`, as it follows:"
      ],
      "metadata": {
        "id": "Mc-1_aEB-mKK"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Train the model\n",
        "query = \"What are the top 5 countries by GDP?\"\n",
        "response = \"\"\"\n",
        "import pandas as pd\n",
        "\n",
        "df = dfs[0]\n",
        "\n",
        "# Convert GDP column to numeric values\n",
        "df['gdp'] = pd.to_numeric(df['gdp'])\n",
        "\n",
        "# Sort the DataFrame by GDP in descending order\n",
        "sorted_df = df.sort_values(by='gdp', ascending=False)\n",
        "\n",
        "# Get the top 5 countries by GDP\n",
        "top_5_countries = sorted_df.head(5)\n",
        "\n",
        "# Declare result variable\n",
        "result = {\n",
        "    \"type\": \"dataframe\",\n",
        "    \"value\": top_5_countries\n",
        "}\n",
        "\"\"\"\n",
        "agent.train(queries=[query], codes=[response])"
      ],
      "metadata": {
        "id": "Cm5C9-lI-mrw"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "response = agent.chat(\"List the top 10 countries by GDP\")\n",
        "print(response)\n",
        "print(agent.last_code_generated)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "flWks4vpF38X",
        "outputId": "edaa7ad3-4431-4b5d-f4c1-6e5b077e5112"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "          country             gdp  happiness_index\n",
            "9           China  14631844184064             5.12\n",
            "8           Japan   4380756541440             5.87\n",
            "3         Germany   3435817336832             7.07\n",
            "1  United Kingdom   2891615567872             7.16\n",
            "2          France   2411255037952             6.66\n",
            "4           Italy   1745433788416             6.38\n",
            "6          Canada   1607402389504             7.23\n",
            "7       Australia   1490967855104             7.22\n",
            "5           Spain   1181205135360             6.40\n",
            "# TODO: import the required dependencies\n",
            "import pandas as pd\n",
            "\n",
            "df = dfs[0]\n",
            "\n",
            "# Convert GDP column to numeric values\n",
            "df['gdp'] = pd.to_numeric(df['gdp'])\n",
            "\n",
            "# Sort the DataFrame by GDP in descending order\n",
            "sorted_df = df.sort_values(by='gdp', ascending=False)\n",
            "\n",
            "# Get the top 10 countries by GDP (excluding the US)\n",
            "top_10_countries = sorted_df[sorted_df['country'] != 'United States'].head(10)\n",
            "\n",
            "# Declare result variable\n",
            "result = {\n",
            "    \"type\": \"dataframe\",\n",
            "    \"value\": top_10_countries\n",
            "}\n"
          ]
        }
      ]
    }
  ]
}