Seating Arrangement.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "gpuType": "T4"
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "code",
      "source": [
        "import random\n",
        "import torch\n",
        "import torch.nn as nn\n",
        "import torch.optim as optim"
      ],
      "metadata": {
        "id": "08ufANNC7GP7"
      },
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def generate_full_names(num_names):\n",
        "    \"\"\"\n",
        "    Generates a list of random full names.\n",
        "\n",
        "    Args:\n",
        "        num_names (int): The number of full names to generate.\n",
        "\n",
        "    Returns:\n",
        "        list: A list of randomly generated full names.\n",
        "    \"\"\"\n",
        "    first_names = [ \"Ali\", \"Zahra\",\"Reza\",\"Sara\", \"Mohammad\",\"Fatemeh\",\"Hossein\", \"Maryam\",\"Mehdi\",\"Narges\",\"Hamed\",\"Roya\",\"Bahram\",\"Roze\",\"Samin\",\"Sogol\",\"sahar\",\"Ahmad\",\"Raha\",\"Hakan\" ]\n",
        "    last_names = [\"Ahmadi\",\"Hosseini\", \"Karimi\", \"Rahimi\",\"Hashemi\",\"Ebrahimi\", \"Moradi\", \"Mohammadi\",\"Rostami\",\"Fazeli\",\"Hosseinzadeh\", \"Niknam\",\"Lotfi\",\"Rajabi\",\"Rasoli\",\"Safari\",\"Sadeghi\",\"Safavi\",\"Mehrzad\",\"Ahmadzadeh\"  ]\n",
        "\n",
        "    random.seed(0)\n",
        "    full_names = set()\n",
        "    while len(full_names) < num_names:\n",
        "        full_name = random.choice(first_names) + \" \" + random.choice(last_names)\n",
        "        full_names.add(full_name)\n",
        "    return list(full_names)\n",
        "\n",
        "full_names = generate_full_names(400)\n",
        "print(full_names)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "RUN0QBeGIFn4",
        "outputId": "790c71de-e6d7-4b94-b3c1-072e7bb25774"
      },
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "['Bahram Mehrzad', 'Roya Sadeghi', 'Maryam Safavi', 'Mehdi Lotfi', 'sahar Ebrahimi', 'Hakan Rostami', 'Sara Ahmadi', 'Maryam Hosseinzadeh', 'Roze Rasoli', 'Roya Hosseinzadeh', 'Roya Safavi', 'Ahmad Niknam', 'Zahra Ebrahimi', 'Raha Rahimi', 'Maryam Fazeli', 'Reza Niknam', 'Reza Moradi', 'Raha Mohammadi', 'Hakan Hosseini', 'Roze Hosseini', 'Sogol Fazeli', 'Hamed Moradi', 'Hossein Ahmadzadeh', 'sahar Mohammadi', 'Narges Hosseini', 'Reza Fazeli', 'Bahram Fazeli', 'Samin Ahmadzadeh', 'sahar Moradi', 'Roze Safari', 'Bahram Hosseini', 'Hamed Hosseini', 'Ahmad Rajabi', 'Mehdi Ahmadi', 'Mehdi Mehrzad', 'Samin Rostami', 'Ahmad Mohammadi', 'Sogol Mehrzad', 'Bahram Rahimi', 'sahar Rostami', 'Maryam Karimi', 'Hossein Karimi', 'Sara Safavi', 'Sogol Karimi', 'Hossein Safavi', 'Zahra Mehrzad', 'Zahra Lotfi', 'Hamed Lotfi', 'Fatemeh Hashemi', 'Roze Rajabi', 'sahar Ahmadi', 'Roya Rasoli', 'Maryam Sadeghi', 'Mohammad Mehrzad', 'Mehdi Moradi', 'Reza Hashemi', 'Hossein Mehrzad', 'Roze Ahmadzadeh', 'Sogol Hosseinzadeh', 'Hakan Mehrzad', 'Roya Karimi', 'Mohammad Safavi', 'Ali Rasoli', 'Ahmad Karimi', 'Sogol Rasoli', 'Zahra Hosseinzadeh', 'Zahra Rajabi', 'Reza Safavi', 'Ali Karimi', 'Sara Moradi', 'Bahram Rajabi', 'Samin Fazeli', 'Hossein Fazeli', 'sahar Fazeli', 'Reza Rostami', 'Samin Ebrahimi', 'Raha Fazeli', 'Mohammad Rasoli', 'Sara Rasoli', 'Hakan Lotfi', 'Hamed Mehrzad', 'Sara Hosseinzadeh', 'Sogol Sadeghi', 'Fatemeh Karimi', 'Narges Ahmadi', 'Zahra Hashemi', 'Sara Lotfi', 'Hossein Moradi', 'Samin Hosseini', 'Maryam Rasoli', 'Bahram Hashemi', 'Hamed Rostami', 'Zahra Karimi', 'Hakan Niknam', 'Maryam Safari', 'Ali Ahmadzadeh', 'Narges Lotfi', 'Mehdi Fazeli', 'Mehdi Hosseinzadeh', 'Raha Safavi', 'Zahra Ahmadzadeh', 'Roze Sadeghi', 'Roze Karimi', 'Hamed Rahimi', 'Bahram Niknam', 'Sogol Safavi', 'Samin Safari', 'Raha Hosseinzadeh', 'Sogol Ahmadi', 'sahar Safari', 'Roya Ahmadi', 'Sara Safari', 'Hamed Ahmadi', 'Sara Ahmadzadeh', 'Mehdi Niknam', 'Zahra Rahimi', 'Mehdi Safavi', 'Hakan Rahimi', 'Samin Niknam', 'Ali Lotfi', 'Roze Rahimi', 'Narges Fazeli', 'Hossein Rajabi', 'Zahra Rasoli', 'Ali Moradi', 'Mohammad Ahmadzadeh', 'Maryam Rostami', 'Mohammad Hosseini', 'Roya Rahimi', 'Ahmad Ahmadi', 'Ali Sadeghi', 'Hossein Ahmadi', 'Maryam Mohammadi', 'Samin Rasoli', 'Mehdi Rajabi', 'Hamed Niknam', 'Roya Fazeli', 'Narges Safavi', 'Roze Mehrzad', 'Hamed Safari', 'Narges Ahmadzadeh', 'Hakan Rajabi', 'Ahmad Ahmadzadeh', 'Narges Rostami', 'Raha Moradi', 'Narges Ebrahimi', 'Narges Rahimi', 'Roya Ebrahimi', 'Hossein Hosseini', 'Ahmad Sadeghi', 'Raha Ebrahimi', 'Hossein Ebrahimi', 'Fatemeh Ebrahimi', 'Hakan Ahmadi', 'Hamed Ahmadzadeh', 'Mohammad Ahmadi', 'Samin Moradi', 'Zahra Niknam', 'Fatemeh Mohammadi', 'Mehdi Ebrahimi', 'Hakan Karimi', 'Hakan Ebrahimi', 'Roze Safavi', 'Mehdi Hashemi', 'Hakan Hashemi', 'Sogol Hashemi', 'Bahram Sadeghi', 'Bahram Rasoli', 'Hakan Mohammadi', 'Mohammad Rahimi', 'Narges Rajabi', 'Ahmad Mehrzad', 'Sara Fazeli', 'Ali Hashemi', 'Sogol Ebrahimi', 'Samin Hosseinzadeh', 'Ahmad Ebrahimi', 'Bahram Mohammadi', 'Roya Safari', 'Ali Ebrahimi', 'sahar Safavi', 'Maryam Ebrahimi', 'Sogol Rahimi', 'Narges Rasoli', 'Fatemeh Fazeli', 'Ahmad Safari', 'Hossein Rahimi', 'Raha Ahmadi', 'Reza Mehrzad', 'Roya Rajabi', 'Zahra Safari', 'Sara Rostami', 'Hossein Lotfi', 'sahar Mehrzad', 'Fatemeh Rostami', 'Fatemeh Ahmadzadeh', 'Reza Sadeghi', 'Ali Hosseini', 'Roze Hashemi', 'Hakan Fazeli', 'Ali Niknam', 'Bahram Ahmadzadeh', 'Ali Rostami', 'Roya Rostami', 'Fatemeh Safavi', 'Zahra Rostami', 'Fatemeh Lotfi', 'Reza Rahimi', 'Raha Safari', 'Fatemeh Hosseini', 'sahar Hosseinzadeh', 'Sara Ebrahimi', 'Ahmad Moradi', 'sahar Ahmadzadeh', 'Hakan Safavi', 'Hamed Ebrahimi', 'Hakan Hosseinzadeh', 'Hossein Sadeghi', 'Reza Mohammadi', 'Roze Lotfi', 'Mohammad Ebrahimi', 'Ahmad Hosseinzadeh', 'Mohammad Hosseinzadeh', 'Ahmad Lotfi', 'Ali Rahimi', 'Ali Mehrzad', 'Mohammad Karimi', 'sahar Rajabi', 'Hamed Safavi', 'Sara Rahimi', 'Zahra Fazeli', 'Maryam Rajabi', 'Reza Rasoli', 'sahar Hosseini', 'Reza Ahmadi', 'Roya Niknam', 'Raha Karimi', 'Hamed Mohammadi', 'Maryam Moradi', 'sahar Sadeghi', 'Hakan Moradi', 'Narges Hosseinzadeh', 'Mehdi Sadeghi', 'Sara Hosseini', 'Sogol Mohammadi', 'Hakan Rasoli', 'Fatemeh Hosseinzadeh', 'Mehdi Rahimi', 'Reza Ahmadzadeh', 'Raha Hashemi', 'Fatemeh Niknam', 'Reza Hosseini', 'Samin Ahmadi', 'Sara Sadeghi', 'Hakan Sadeghi', 'Mohammad Lotfi', 'Bahram Lotfi', 'Ali Safari', 'Sogol Rajabi', 'Ahmad Fazeli', 'Maryam Ahmadi', 'Ahmad Hashemi', 'Mehdi Mohammadi', 'sahar Niknam', 'Ahmad Hosseini', 'Bahram Ebrahimi', 'Sogol Safari', 'Sogol Lotfi', 'sahar Rahimi', 'Mohammad Niknam', 'Reza Rajabi', 'Raha Mehrzad', 'Hossein Rostami', 'Mehdi Karimi', 'Maryam Ahmadzadeh', 'Hamed Rajabi', 'Samin Karimi', 'Mohammad Sadeghi', 'Roya Hashemi', 'Hakan Ahmadzadeh', 'Samin Hashemi', 'Mohammad Moradi', 'Mohammad Rajabi', 'Ahmad Rasoli', 'Bahram Rostami', 'Hossein Rasoli', 'Ali Rajabi', 'Mohammad Hashemi', 'Zahra Ahmadi', 'Roze Hosseinzadeh', 'Maryam Niknam', 'sahar Rasoli', 'Roya Mohammadi', 'Roze Niknam', 'Zahra Sadeghi', 'Reza Ebrahimi', 'Raha Rasoli', 'Mehdi Safari', 'Raha Hosseini', 'sahar Hashemi', 'Fatemeh Sadeghi', 'Roya Mehrzad', 'Narges Mohammadi', 'Raha Sadeghi', 'Raha Ahmadzadeh', 'Mehdi Rostami', 'Zahra Safavi', 'Roze Moradi', 'Fatemeh Moradi', 'Sogol Hosseini', 'Reza Lotfi', 'Fatemeh Rahimi', 'Sara Niknam', 'Reza Hosseinzadeh', 'Maryam Mehrzad', 'Roze Rostami', 'Zahra Mohammadi', 'Bahram Hosseinzadeh', 'Hossein Hosseinzadeh', 'Mohammad Rostami', 'Sogol Moradi', 'Raha Niknam', 'Samin Lotfi', 'Roze Ahmadi', 'Ali Safavi', 'Samin Safavi', 'Ali Fazeli', 'Roya Hosseini', 'Raha Rajabi', 'Hossein Mohammadi', 'Narges Sadeghi', 'Bahram Safari', 'Bahram Karimi', 'Sogol Niknam', 'Narges Niknam', 'Bahram Safavi', 'Sara Hashemi', 'Narges Moradi', 'Sara Rajabi', 'Hamed Hosseinzadeh', 'Fatemeh Rajabi', 'Sara Karimi', 'Narges Mehrzad', 'Maryam Hosseini', 'Sogol Rostami', 'Narges Safari', 'Roze Fazeli', 'Raha Lotfi', 'Reza Karimi', 'Ahmad Rahimi', 'Bahram Moradi', 'Fatemeh Ahmadi', 'Samin Rajabi', 'Mehdi Rasoli', 'Ahmad Rostami', 'Sara Mehrzad', 'Ahmad Safavi', 'Mohammad Fazeli', 'Ali Ahmadi', 'Hakan Safari', 'sahar Lotfi', 'Zahra Moradi', 'Ali Mohammadi', 'Maryam Hashemi', 'Narges Karimi', 'Bahram Ahmadi', 'Hamed Fazeli', 'Sara Mohammadi', 'Roya Ahmadzadeh', 'Fatemeh Mehrzad', 'Sogol Ahmadzadeh', 'Maryam Rahimi', 'Mohammad Mohammadi', 'Hossein Niknam', 'Hamed Sadeghi', 'Hamed Hashemi', 'Mohammad Safari', 'Mehdi Hosseini', 'Ali Hosseinzadeh', 'Maryam Lotfi', 'Narges Hashemi', 'Samin Rahimi', 'Fatemeh Rasoli', 'Reza Safari', 'Mehdi Ahmadzadeh', 'Raha Rostami', 'Fatemeh Safari', 'Roze Ebrahimi', 'Hamed Karimi', 'Roze Mohammadi', 'Hossein Safari', 'Hossein Hashemi', 'sahar Karimi', 'Zahra Hosseini', 'Samin Mohammadi', 'Roya Lotfi', 'Samin Mehrzad', 'Roya Moradi', 'Samin Sadeghi', 'Hamed Rasoli']\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def group_create(names, group_size):\n",
        "    \"\"\"\n",
        "    Groups the names into groups of specified size.\n",
        "\n",
        "    Args:\n",
        "        names (list): List of names to be grouped.\n",
        "        group_size (int): Size of each group.\n",
        "\n",
        "    Returns:\n",
        "        list: List of groups, each containing a list of names.\n",
        "        list: List of groups, each containing a list of indices.\n",
        "    \"\"\"\n",
        "    indices = list(range(len(names)))\n",
        "    combined = list(zip(names, indices))\n",
        "\n",
        "    random.shuffle(combined)\n",
        "    shuffled_names, shuffled_indices = zip(*combined)\n",
        "\n",
        "    groups_name = [list(shuffled_names[i:i + group_size]) for i in range(0, len(names), group_size)]\n",
        "    groups_number = [list(shuffled_indices[i:i + group_size]) for i in range(0, len(names), group_size)]\n",
        "\n",
        "    return groups_name, groups_number\n",
        "\n",
        "# Generate full names\n",
        "full_names = generate_full_names(400)\n",
        "\n",
        "# Group names into groups of 5\n",
        "groups_name, groups_number = group_create(full_names, 5)\n",
        "\n",
        "# Print groups\n",
        "for i, (group_names, group_numbers) in enumerate(zip(groups_name, groups_number)):\n",
        "    print(f\"Group {i + 1}: Names = {group_names}, Numbers = {group_numbers}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "iGlTl6dt-Lib",
        "outputId": "1ac35636-cd5a-4ba6-9b17-154f87ac5a43"
      },
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Group 1: Names = ['Sogol Hashemi', 'Samin Hashemi', 'Ahmad Niknam', 'Samin Safari', 'Mohammad Ahmadi'], Numbers = [165, 280, 11, 106, 155]\n",
            "Group 2: Names = ['Ahmad Karimi', 'Narges Rajabi', 'Hakan Fazeli', 'Hakan Ahmadzadeh', 'Hossein Rasoli'], Numbers = [63, 170, 199, 279, 285]\n",
            "Group 3: Names = ['Ali Rostami', 'Ali Niknam', 'Sara Mehrzad', 'Ahmad Mehrzad', 'Samin Moradi'], Numbers = [202, 200, 355, 171, 156]\n",
            "Group 4: Names = ['Sogol Rostami', 'sahar Lotfi', 'Samin Ebrahimi', 'Mohammad Ebrahimi', 'Ahmad Ebrahimi'], Numbers = [344, 360, 75, 220, 176]\n",
            "Group 5: Names = ['Mohammad Mohammadi', 'Zahra Fazeli', 'Hakan Rahimi', 'Sogol Rasoli', 'Maryam Ahmadzadeh'], Numbers = [372, 230, 117, 64, 274]\n",
            "Group 6: Names = ['Ahmad Hosseinzadeh', 'Hakan Hashemi', 'Maryam Rajabi', 'Zahra Safari', 'Raha Sadeghi'], Numbers = [221, 164, 231, 190, 303]\n",
            "Group 7: Names = ['Hossein Hosseini', 'Fatemeh Ebrahimi', 'Roya Hosseini', 'Sogol Sadeghi', 'Zahra Rasoli'], Numbers = [148, 152, 327, 82, 123]\n",
            "Group 8: Names = ['Ahmad Safari', 'Ali Rasoli', 'Roze Safavi', 'Sogol Fazeli', 'Roze Hashemi'], Numbers = [185, 62, 162, 20, 198]\n",
            "Group 9: Names = ['sahar Mehrzad', 'Reza Mohammadi', 'Maryam Lotfi', 'Reza Rahimi', 'Narges Safari'], Numbers = [193, 218, 379, 207, 345]\n",
            "Group 10: Names = ['Narges Ebrahimi', 'Sara Safavi', 'Mohammad Lotfi', 'Roze Niknam', 'Samin Rajabi'], Numbers = [145, 42, 255, 293, 352]\n",
            "Group 11: Names = ['Roya Safavi', 'Bahram Rasoli', 'Sogol Rahimi', 'sahar Rostami', 'Narges Rasoli'], Numbers = [10, 167, 182, 39, 183]\n",
            "Group 12: Names = ['Narges Ahmadzadeh', 'Sogol Safavi', 'Mohammad Moradi', 'Reza Ahmadi', 'Hossein Safavi'], Numbers = [140, 105, 281, 234, 44]\n",
            "Group 13: Names = ['Roya Ahmadi', 'Roya Ahmadzadeh', 'Ali Safari', 'Narges Fazeli', 'Reza Fazeli'], Numbers = [110, 368, 257, 121, 25]\n",
            "Group 14: Names = ['Zahra Rajabi', 'Raha Rasoli', 'sahar Rasoli', 'Reza Sadeghi', 'Roze Rahimi'], Numbers = [66, 296, 291, 196, 120]\n",
            "Group 15: Names = ['Samin Mehrzad', 'Sogol Hosseinzadeh', 'Roze Ahmadi', 'Roze Safari', 'Mehdi Niknam'], Numbers = [396, 58, 323, 29, 114]\n",
            "Group 16: Names = ['Mehdi Mohammadi', 'Roze Rostami', 'Roya Ebrahimi', 'Reza Ebrahimi', 'Narges Moradi'], Numbers = [262, 315, 147, 295, 337]\n",
            "Group 17: Names = ['Maryam Hosseini', 'Ali Mehrzad', 'Reza Hosseini', 'Bahram Mohammadi', 'Bahram Ahmadzadeh'], Numbers = [343, 225, 251, 177, 201]\n",
            "Group 18: Names = ['Mehdi Lotfi', 'Reza Niknam', 'Mehdi Karimi', 'Mehdi Ebrahimi', 'Ahmad Rostami'], Numbers = [3, 15, 273, 159, 354]\n",
            "Group 19: Names = ['Fatemeh Lotfi', 'Raha Niknam', 'Roze Mehrzad', 'Fatemeh Moradi', 'Narges Ahmadi'], Numbers = [206, 321, 138, 308, 84]\n",
            "Group 20: Names = ['Maryam Mohammadi', 'Sara Lotfi', 'Roze Ahmadzadeh', 'Ali Moradi', 'Mehdi Safari'], Numbers = [132, 86, 57, 124, 297]\n",
            "Group 21: Names = ['Mehdi Moradi', 'Ali Hashemi', 'Roya Lotfi', 'Fatemeh Rasoli', 'Bahram Rahimi'], Numbers = [54, 173, 395, 382, 38]\n",
            "Group 22: Names = ['Bahram Ebrahimi', 'Bahram Lotfi', 'Samin Hosseinzadeh', 'Sara Fazeli', 'Hamed Moradi'], Numbers = [265, 256, 175, 172, 21]\n",
            "Group 23: Names = ['Hossein Sadeghi', 'Zahra Hosseini', 'sahar Rajabi', 'Hamed Rasoli', 'Bahram Rajabi'], Numbers = [217, 393, 227, 399, 70]\n",
            "Group 24: Names = ['Maryam Rahimi', 'Zahra Ahmadzadeh', 'Raha Karimi', 'Samin Safavi', 'Zahra Rahimi'], Numbers = [371, 100, 236, 325, 115]\n",
            "Group 25: Names = ['Hamed Safavi', 'Mohammad Rostami', 'Mehdi Rajabi', 'Zahra Sadeghi', 'sahar Safari'], Numbers = [228, 319, 134, 294, 109]\n",
            "Group 26: Names = ['Hamed Rahimi', 'Sogol Karimi', 'Sogol Mohammadi', 'Roze Rajabi', 'Fatemeh Safari'], Numbers = [103, 43, 244, 49, 386]\n",
            "Group 27: Names = ['Roya Mohammadi', 'Narges Safavi', 'Raha Mohammadi', 'Mohammad Mehrzad', 'Fatemeh Hashemi'], Numbers = [292, 137, 17, 53, 48]\n",
            "Group 28: Names = ['Hamed Ebrahimi', 'Narges Karimi', 'Narges Lotfi', 'Fatemeh Ahmadi', 'Mehdi Ahmadi'], Numbers = [215, 364, 96, 351, 33]\n",
            "Group 29: Names = ['Mohammad Safavi', 'Mohammad Sadeghi', 'Reza Ahmadzadeh', 'Raha Safavi', 'Maryam Hashemi'], Numbers = [61, 277, 248, 99, 363]\n",
            "Group 30: Names = ['sahar Hosseinzadeh', 'Roya Safari', 'Sara Ahmadi', 'Hossein Moradi', 'Raha Rajabi'], Numbers = [210, 178, 6, 87, 328]\n",
            "Group 31: Names = ['Maryam Rasoli', 'sahar Ahmadi', 'Sara Ahmadzadeh', 'Mehdi Hosseini', 'Ahmad Ahmadi'], Numbers = [89, 50, 113, 377, 129]\n",
            "Group 32: Names = ['sahar Ahmadzadeh', 'Maryam Ahmadi', 'Ali Mohammadi', 'Fatemeh Safavi', 'Samin Niknam'], Numbers = [213, 260, 362, 204, 118]\n",
            "Group 33: Names = ['Reza Hashemi', 'Roze Mohammadi', 'Hakan Karimi', 'Sogol Rajabi', 'Mehdi Fazeli'], Numbers = [55, 389, 160, 258, 97]\n",
            "Group 34: Names = ['Reza Karimi', 'sahar Fazeli', 'Hossein Ahmadzadeh', 'Hakan Rajabi', 'Roze Lotfi'], Numbers = [348, 73, 22, 141, 219]\n",
            "Group 35: Names = ['Fatemeh Mohammadi', 'Raha Mehrzad', 'Zahra Mehrzad', 'Narges Niknam', 'Hakan Hosseini'], Numbers = [158, 271, 45, 334, 18]\n",
            "Group 36: Names = ['Mohammad Hashemi', 'Zahra Safavi', 'Raha Ahmadzadeh', 'Raha Hashemi', 'Roze Hosseini'], Numbers = [287, 306, 304, 249, 19]\n",
            "Group 37: Names = ['Maryam Sadeghi', 'Hossein Mehrzad', 'Zahra Hashemi', 'Mohammad Ahmadzadeh', 'sahar Karimi'], Numbers = [52, 56, 85, 125, 392]\n",
            "Group 38: Names = ['Roya Sadeghi', 'Samin Fazeli', 'Fatemeh Rostami', 'Raha Hosseini', 'Sara Hosseini'], Numbers = [1, 71, 194, 298, 243]\n",
            "Group 39: Names = ['Mohammad Fazeli', 'Raha Rahimi', 'Hakan Niknam', 'Ali Karimi', 'Ali Safavi'], Numbers = [357, 13, 93, 68, 324]\n",
            "Group 40: Names = ['Reza Rajabi', 'Zahra Ahmadi', 'Hamed Safari', 'Ahmad Hashemi', 'Samin Lotfi'], Numbers = [270, 288, 139, 261, 322]\n",
            "Group 41: Names = ['Hossein Rostami', 'Raha Rostami', 'Hossein Rahimi', 'Sogol Safari', 'Hossein Ebrahimi'], Numbers = [272, 385, 186, 266, 151]\n",
            "Group 42: Names = ['Ali Hosseini', 'Maryam Moradi', 'Reza Safavi', 'Ahmad Rahimi', 'Roya Moradi'], Numbers = [197, 238, 67, 349, 397]\n",
            "Group 43: Names = ['Samin Ahmadzadeh', 'Fatemeh Ahmadzadeh', 'Sara Karimi', 'Ali Lotfi', 'Zahra Hosseinzadeh'], Numbers = [27, 195, 341, 119, 65]\n",
            "Group 44: Names = ['Hossein Fazeli', 'Sara Niknam', 'Ali Ebrahimi', 'Bahram Sadeghi', 'Hamed Fazeli'], Numbers = [72, 312, 179, 166, 366]\n",
            "Group 45: Names = ['Reza Mehrzad', 'Hakan Ebrahimi', 'Hakan Sadeghi', 'Fatemeh Karimi', 'Samin Mohammadi'], Numbers = [188, 161, 254, 83, 394]\n",
            "Group 46: Names = ['Roya Rostami', 'Roze Moradi', 'Hakan Rostami', 'Roze Rasoli', 'Maryam Fazeli'], Numbers = [203, 307, 5, 8, 14]\n",
            "Group 47: Names = ['Roya Hashemi', 'Fatemeh Rahimi', 'Roya Karimi', 'Ahmad Hosseini', 'Hakan Safari'], Numbers = [278, 311, 60, 264, 359]\n",
            "Group 48: Names = ['Ahmad Fazeli', 'Sara Rostami', 'Hakan Mohammadi', 'Sara Rajabi', 'Raha Moradi'], Numbers = [259, 191, 168, 338, 144]\n",
            "Group 49: Names = ['Maryam Hosseinzadeh', 'Narges Rostami', 'Reza Lotfi', 'Bahram Niknam', 'Zahra Ebrahimi'], Numbers = [7, 143, 310, 104, 12]\n",
            "Group 50: Names = ['Raha Safari', 'Hamed Lotfi', 'Roya Niknam', 'Ali Fazeli', 'Fatemeh Hosseini'], Numbers = [208, 47, 235, 326, 209]\n",
            "Group 51: Names = ['Bahram Hosseini', 'Roze Sadeghi', 'Ali Ahmadi', 'Ali Ahmadzadeh', 'Mohammad Rajabi'], Numbers = [30, 101, 358, 95, 282]\n",
            "Group 52: Names = ['Sara Rasoli', 'Ali Sadeghi', 'Maryam Mehrzad', 'Ahmad Moradi', 'Sara Hashemi'], Numbers = [78, 130, 314, 212, 336]\n",
            "Group 53: Names = ['Hossein Mohammadi', 'Roya Mehrzad', 'Maryam Ebrahimi', 'Maryam Safari', 'Mohammad Safari'], Numbers = [329, 301, 181, 94, 376]\n",
            "Group 54: Names = ['Maryam Rostami', 'Hakan Ahmadi', 'sahar Niknam', 'Sogol Ahmadi', 'Roze Karimi'], Numbers = [126, 153, 263, 108, 102]\n",
            "Group 55: Names = ['Sogol Lotfi', 'Hossein Hashemi', 'Roya Rajabi', 'Maryam Karimi', 'Raha Fazeli'], Numbers = [267, 391, 189, 40, 76]\n",
            "Group 56: Names = ['Roya Rasoli', 'Samin Rasoli', 'Sara Sadeghi', 'sahar Ebrahimi', 'Hamed Niknam'], Numbers = [51, 133, 253, 4, 135]\n",
            "Group 57: Names = ['Zahra Rostami', 'Mehdi Safavi', 'sahar Moradi', 'Fatemeh Mehrzad', 'Roze Ebrahimi'], Numbers = [205, 116, 28, 369, 387]\n",
            "Group 58: Names = ['Ahmad Mohammadi', 'Mohammad Rasoli', 'Mohammad Rahimi', 'Narges Hashemi', 'Sara Rahimi'], Numbers = [36, 77, 169, 380, 229]\n",
            "Group 59: Names = ['Hamed Hosseini', 'Hamed Hosseinzadeh', 'Samin Karimi', 'Hamed Ahmadi', 'Hossein Rajabi'], Numbers = [31, 339, 276, 112, 122]\n",
            "Group 60: Names = ['Sara Safari', 'Sara Hosseinzadeh', 'Ali Rahimi', 'Raha Lotfi', 'Hossein Safari'], Numbers = [111, 81, 224, 347, 390]\n",
            "Group 61: Names = ['Sogol Hosseini', 'Mehdi Mehrzad', 'Hossein Hosseinzadeh', 'Mehdi Sadeghi', 'Samin Hosseini'], Numbers = [309, 34, 318, 242, 88]\n",
            "Group 62: Names = ['Roya Hosseinzadeh', 'Roya Fazeli', 'Hakan Mehrzad', 'Fatemeh Rajabi', 'Mehdi Rostami'], Numbers = [9, 136, 59, 340, 305]\n",
            "Group 63: Names = ['Hakan Moradi', 'Hossein Lotfi', 'Zahra Mohammadi', 'Zahra Lotfi', 'Hakan Hosseinzadeh'], Numbers = [240, 192, 316, 46, 216]\n",
            "Group 64: Names = ['Mohammad Hosseini', 'sahar Mohammadi', 'Reza Moradi', 'Narges Hosseini', 'Sogol Ahmadzadeh'], Numbers = [127, 23, 16, 24, 370]\n",
            "Group 65: Names = ['Hamed Ahmadzadeh', 'Zahra Moradi', 'Mohammad Hosseinzadeh', 'Bahram Karimi', 'Zahra Karimi'], Numbers = [154, 361, 222, 332, 92]\n",
            "Group 66: Names = ['Ahmad Ahmadzadeh', 'Mehdi Rahimi', 'Bahram Rostami', 'Mehdi Ahmadzadeh', 'Hossein Niknam'], Numbers = [142, 247, 284, 384, 373]\n",
            "Group 67: Names = ['sahar Hashemi', 'Ali Rajabi', 'Samin Sadeghi', 'Bahram Hashemi', 'Hamed Karimi'], Numbers = [299, 286, 398, 90, 388]\n",
            "Group 68: Names = ['Sara Mohammadi', 'Narges Rahimi', 'Mohammad Niknam', 'Sara Moradi', 'Fatemeh Sadeghi'], Numbers = [367, 146, 269, 69, 300]\n",
            "Group 69: Names = ['Mehdi Rasoli', 'Roze Hosseinzadeh', 'Bahram Safari', 'Hossein Ahmadi', 'Ahmad Rajabi'], Numbers = [353, 289, 331, 131, 32]\n",
            "Group 70: Names = ['Fatemeh Fazeli', 'Fatemeh Hosseinzadeh', 'Reza Rostami', 'Narges Sadeghi', 'Mohammad Karimi'], Numbers = [184, 246, 74, 330, 226]\n",
            "Group 71: Names = ['sahar Safavi', 'Hamed Mohammadi', 'Ali Hosseinzadeh', 'Raha Hosseinzadeh', 'Zahra Niknam'], Numbers = [180, 237, 378, 107, 157]\n",
            "Group 72: Names = ['Sogol Niknam', 'Samin Rahimi', 'Reza Safari', 'Hakan Rasoli', 'Raha Ahmadi'], Numbers = [333, 381, 383, 245, 187]\n",
            "Group 73: Names = ['Ahmad Rasoli', 'Bahram Moradi', 'sahar Sadeghi', 'Hamed Sadeghi', 'Narges Hosseinzadeh'], Numbers = [283, 350, 239, 374, 241]\n",
            "Group 74: Names = ['Samin Rostami', 'Samin Ahmadi', 'Bahram Safavi', 'Bahram Mehrzad', 'Narges Mohammadi'], Numbers = [35, 252, 335, 0, 302]\n",
            "Group 75: Names = ['Ahmad Lotfi', 'Narges Mehrzad', 'Raha Ebrahimi', 'Hamed Rajabi', 'Ahmad Sadeghi'], Numbers = [223, 342, 150, 275, 149]\n",
            "Group 76: Names = ['Roya Rahimi', 'Hamed Mehrzad', 'Bahram Fazeli', 'Hakan Safavi', 'Roze Fazeli'], Numbers = [128, 80, 26, 214, 346]\n",
            "Group 77: Names = ['Fatemeh Niknam', 'Bahram Ahmadi', 'Mehdi Hosseinzadeh', 'Bahram Hosseinzadeh', 'sahar Hosseini'], Numbers = [250, 365, 98, 317, 233]\n",
            "Group 78: Names = ['Sogol Ebrahimi', 'Reza Hosseinzadeh', 'Maryam Niknam', 'Hossein Karimi', 'Hamed Rostami'], Numbers = [174, 313, 290, 41, 91]\n",
            "Group 79: Names = ['Hamed Hashemi', 'Mehdi Hashemi', 'Ahmad Safavi', 'Sogol Moradi', 'Sara Ebrahimi'], Numbers = [375, 163, 356, 320, 211]\n",
            "Group 80: Names = ['Sogol Mehrzad', 'Reza Rasoli', 'sahar Rahimi', 'Maryam Safavi', 'Hakan Lotfi'], Numbers = [37, 232, 268, 2, 79]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "groups_number"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "GlDjj3-nUo4m",
        "outputId": "85b76207-76bf-4b80-8a87-c021b63d5b0e"
      },
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[[165, 280, 11, 106, 155],\n",
              " [63, 170, 199, 279, 285],\n",
              " [202, 200, 355, 171, 156],\n",
              " [344, 360, 75, 220, 176],\n",
              " [372, 230, 117, 64, 274],\n",
              " [221, 164, 231, 190, 303],\n",
              " [148, 152, 327, 82, 123],\n",
              " [185, 62, 162, 20, 198],\n",
              " [193, 218, 379, 207, 345],\n",
              " [145, 42, 255, 293, 352],\n",
              " [10, 167, 182, 39, 183],\n",
              " [140, 105, 281, 234, 44],\n",
              " [110, 368, 257, 121, 25],\n",
              " [66, 296, 291, 196, 120],\n",
              " [396, 58, 323, 29, 114],\n",
              " [262, 315, 147, 295, 337],\n",
              " [343, 225, 251, 177, 201],\n",
              " [3, 15, 273, 159, 354],\n",
              " [206, 321, 138, 308, 84],\n",
              " [132, 86, 57, 124, 297],\n",
              " [54, 173, 395, 382, 38],\n",
              " [265, 256, 175, 172, 21],\n",
              " [217, 393, 227, 399, 70],\n",
              " [371, 100, 236, 325, 115],\n",
              " [228, 319, 134, 294, 109],\n",
              " [103, 43, 244, 49, 386],\n",
              " [292, 137, 17, 53, 48],\n",
              " [215, 364, 96, 351, 33],\n",
              " [61, 277, 248, 99, 363],\n",
              " [210, 178, 6, 87, 328],\n",
              " [89, 50, 113, 377, 129],\n",
              " [213, 260, 362, 204, 118],\n",
              " [55, 389, 160, 258, 97],\n",
              " [348, 73, 22, 141, 219],\n",
              " [158, 271, 45, 334, 18],\n",
              " [287, 306, 304, 249, 19],\n",
              " [52, 56, 85, 125, 392],\n",
              " [1, 71, 194, 298, 243],\n",
              " [357, 13, 93, 68, 324],\n",
              " [270, 288, 139, 261, 322],\n",
              " [272, 385, 186, 266, 151],\n",
              " [197, 238, 67, 349, 397],\n",
              " [27, 195, 341, 119, 65],\n",
              " [72, 312, 179, 166, 366],\n",
              " [188, 161, 254, 83, 394],\n",
              " [203, 307, 5, 8, 14],\n",
              " [278, 311, 60, 264, 359],\n",
              " [259, 191, 168, 338, 144],\n",
              " [7, 143, 310, 104, 12],\n",
              " [208, 47, 235, 326, 209],\n",
              " [30, 101, 358, 95, 282],\n",
              " [78, 130, 314, 212, 336],\n",
              " [329, 301, 181, 94, 376],\n",
              " [126, 153, 263, 108, 102],\n",
              " [267, 391, 189, 40, 76],\n",
              " [51, 133, 253, 4, 135],\n",
              " [205, 116, 28, 369, 387],\n",
              " [36, 77, 169, 380, 229],\n",
              " [31, 339, 276, 112, 122],\n",
              " [111, 81, 224, 347, 390],\n",
              " [309, 34, 318, 242, 88],\n",
              " [9, 136, 59, 340, 305],\n",
              " [240, 192, 316, 46, 216],\n",
              " [127, 23, 16, 24, 370],\n",
              " [154, 361, 222, 332, 92],\n",
              " [142, 247, 284, 384, 373],\n",
              " [299, 286, 398, 90, 388],\n",
              " [367, 146, 269, 69, 300],\n",
              " [353, 289, 331, 131, 32],\n",
              " [184, 246, 74, 330, 226],\n",
              " [180, 237, 378, 107, 157],\n",
              " [333, 381, 383, 245, 187],\n",
              " [283, 350, 239, 374, 241],\n",
              " [35, 252, 335, 0, 302],\n",
              " [223, 342, 150, 275, 149],\n",
              " [128, 80, 26, 214, 346],\n",
              " [250, 365, 98, 317, 233],\n",
              " [174, 313, 290, 41, 91],\n",
              " [375, 163, 356, 320, 211],\n",
              " [37, 232, 268, 2, 79]]"
            ]
          },
          "metadata": {},
          "execution_count": 26
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def get_matrix_conflicts(num_people: int, max_conflicts: int, group_numbers):\n",
        "    \"\"\"\n",
        "    Generates a matrix representing conflicts between people.\n",
        "\n",
        "    Args:\n",
        "        num_people (int): The number of people.\n",
        "        max_conflicts (int): The maximum number of conflicts to generate.\n",
        "        group_numbers (list): List of groups with their member indices.\n",
        "\n",
        "    Returns:\n",
        "        torch.Tensor: A matrix representing conflicts between people.\n",
        "    \"\"\"\n",
        "    matrix_conflicts = torch.zeros(num_people, num_people)\n",
        "\n",
        "    count_one = 0\n",
        "    while count_one < max_conflicts:\n",
        "        i, j = random.sample(range(num_people), 2)\n",
        "        same_group = False\n",
        "        for group in group_numbers:\n",
        "            if i in group and j in group:\n",
        "                same_group = True\n",
        "                break\n",
        "\n",
        "        if not same_group and matrix_conflicts[i][j] == 0:\n",
        "            matrix_conflicts[i][j] = 1\n",
        "            count_one += 1\n",
        "            print(f\"i: {i}, j: {j}\")\n",
        "\n",
        "    return matrix_conflicts.float()\n",
        "\n",
        "# Set print options for better output display\n",
        "torch.set_printoptions(precision=0, linewidth=150)\n",
        "\n",
        "# Call the function to generate the matrix\n",
        "incompatible_pairs = get_matrix_conflicts(num_people=400, max_conflicts=600, group_numbers=groups_number)\n",
        "\n",
        "print(incompatible_pairs)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "nvV1POTV_IPf",
        "outputId": "e1caaad7-d4d9-4958-8575-16aa2573d944"
      },
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "i: 260, j: 240\n",
            "i: 248, j: 198\n",
            "i: 137, j: 334\n",
            "i: 384, j: 232\n",
            "i: 173, j: 314\n",
            "i: 312, j: 278\n",
            "i: 334, j: 324\n",
            "i: 50, j: 389\n",
            "i: 87, j: 294\n",
            "i: 344, j: 333\n",
            "i: 369, j: 335\n",
            "i: 20, j: 229\n",
            "i: 248, j: 245\n",
            "i: 136, j: 217\n",
            "i: 115, j: 390\n",
            "i: 375, j: 377\n",
            "i: 167, j: 195\n",
            "i: 248, j: 120\n",
            "i: 310, j: 300\n",
            "i: 11, j: 273\n",
            "i: 321, j: 102\n",
            "i: 290, j: 398\n",
            "i: 171, j: 99\n",
            "i: 354, j: 79\n",
            "i: 150, j: 273\n",
            "i: 274, j: 358\n",
            "i: 93, j: 223\n",
            "i: 318, j: 54\n",
            "i: 358, j: 131\n",
            "i: 196, j: 111\n",
            "i: 32, j: 315\n",
            "i: 189, j: 203\n",
            "i: 76, j: 395\n",
            "i: 346, j: 210\n",
            "i: 274, j: 250\n",
            "i: 53, j: 135\n",
            "i: 216, j: 182\n",
            "i: 187, j: 393\n",
            "i: 204, j: 182\n",
            "i: 388, j: 210\n",
            "i: 343, j: 133\n",
            "i: 3, j: 106\n",
            "i: 234, j: 187\n",
            "i: 109, j: 47\n",
            "i: 162, j: 6\n",
            "i: 192, j: 214\n",
            "i: 104, j: 300\n",
            "i: 217, j: 127\n",
            "i: 333, j: 322\n",
            "i: 374, j: 356\n",
            "i: 329, j: 341\n",
            "i: 214, j: 162\n",
            "i: 59, j: 8\n",
            "i: 106, j: 325\n",
            "i: 152, j: 38\n",
            "i: 109, j: 154\n",
            "i: 43, j: 104\n",
            "i: 81, j: 120\n",
            "i: 294, j: 356\n",
            "i: 396, j: 222\n",
            "i: 326, j: 166\n",
            "i: 150, j: 80\n",
            "i: 123, j: 332\n",
            "i: 266, j: 252\n",
            "i: 3, j: 384\n",
            "i: 67, j: 248\n",
            "i: 328, j: 318\n",
            "i: 33, j: 1\n",
            "i: 157, j: 226\n",
            "i: 371, j: 137\n",
            "i: 101, j: 263\n",
            "i: 278, j: 71\n",
            "i: 321, j: 351\n",
            "i: 174, j: 156\n",
            "i: 69, j: 94\n",
            "i: 396, j: 165\n",
            "i: 206, j: 362\n",
            "i: 151, j: 179\n",
            "i: 230, j: 246\n",
            "i: 212, j: 90\n",
            "i: 304, j: 232\n",
            "i: 107, j: 135\n",
            "i: 391, j: 214\n",
            "i: 333, j: 223\n",
            "i: 251, j: 31\n",
            "i: 175, j: 251\n",
            "i: 175, j: 320\n",
            "i: 25, j: 165\n",
            "i: 253, j: 186\n",
            "i: 328, j: 323\n",
            "i: 224, j: 34\n",
            "i: 47, j: 383\n",
            "i: 273, j: 145\n",
            "i: 332, j: 151\n",
            "i: 319, j: 223\n",
            "i: 174, j: 335\n",
            "i: 331, j: 233\n",
            "i: 57, j: 177\n",
            "i: 347, j: 228\n",
            "i: 330, j: 211\n",
            "i: 279, j: 16\n",
            "i: 203, j: 277\n",
            "i: 179, j: 326\n",
            "i: 3, j: 181\n",
            "i: 396, j: 11\n",
            "i: 303, j: 240\n",
            "i: 258, j: 313\n",
            "i: 217, j: 112\n",
            "i: 350, j: 367\n",
            "i: 30, j: 156\n",
            "i: 263, j: 274\n",
            "i: 99, j: 309\n",
            "i: 275, j: 111\n",
            "i: 350, j: 58\n",
            "i: 258, j: 207\n",
            "i: 60, j: 281\n",
            "i: 27, j: 134\n",
            "i: 163, j: 173\n",
            "i: 26, j: 231\n",
            "i: 350, j: 373\n",
            "i: 395, j: 76\n",
            "i: 260, j: 301\n",
            "i: 9, j: 306\n",
            "i: 333, j: 310\n",
            "i: 312, j: 208\n",
            "i: 128, j: 246\n",
            "i: 139, j: 314\n",
            "i: 376, j: 272\n",
            "i: 345, j: 114\n",
            "i: 150, j: 83\n",
            "i: 251, j: 378\n",
            "i: 124, j: 392\n",
            "i: 138, j: 176\n",
            "i: 96, j: 292\n",
            "i: 76, j: 32\n",
            "i: 196, j: 99\n",
            "i: 389, j: 105\n",
            "i: 314, j: 309\n",
            "i: 106, j: 195\n",
            "i: 83, j: 248\n",
            "i: 353, j: 154\n",
            "i: 314, j: 40\n",
            "i: 380, j: 48\n",
            "i: 392, j: 365\n",
            "i: 157, j: 111\n",
            "i: 14, j: 27\n",
            "i: 178, j: 44\n",
            "i: 230, j: 148\n",
            "i: 159, j: 252\n",
            "i: 192, j: 141\n",
            "i: 113, j: 346\n",
            "i: 286, j: 190\n",
            "i: 271, j: 311\n",
            "i: 55, j: 320\n",
            "i: 166, j: 201\n",
            "i: 189, j: 60\n",
            "i: 273, j: 396\n",
            "i: 74, j: 317\n",
            "i: 243, j: 312\n",
            "i: 300, j: 196\n",
            "i: 269, j: 179\n",
            "i: 136, j: 273\n",
            "i: 76, j: 348\n",
            "i: 69, j: 38\n",
            "i: 78, j: 47\n",
            "i: 131, j: 22\n",
            "i: 354, j: 330\n",
            "i: 112, j: 335\n",
            "i: 258, j: 182\n",
            "i: 372, j: 199\n",
            "i: 286, j: 178\n",
            "i: 379, j: 54\n",
            "i: 357, j: 379\n",
            "i: 257, j: 361\n",
            "i: 299, j: 372\n",
            "i: 240, j: 12\n",
            "i: 396, j: 103\n",
            "i: 278, j: 210\n",
            "i: 10, j: 211\n",
            "i: 340, j: 217\n",
            "i: 301, j: 282\n",
            "i: 286, j: 124\n",
            "i: 53, j: 41\n",
            "i: 172, j: 212\n",
            "i: 272, j: 156\n",
            "i: 332, j: 109\n",
            "i: 252, j: 236\n",
            "i: 88, j: 323\n",
            "i: 298, j: 300\n",
            "i: 146, j: 266\n",
            "i: 276, j: 240\n",
            "i: 186, j: 50\n",
            "i: 370, j: 0\n",
            "i: 174, j: 388\n",
            "i: 177, j: 235\n",
            "i: 77, j: 272\n",
            "i: 140, j: 26\n",
            "i: 32, j: 231\n",
            "i: 92, j: 252\n",
            "i: 335, j: 154\n",
            "i: 325, j: 261\n",
            "i: 77, j: 81\n",
            "i: 303, j: 36\n",
            "i: 346, j: 123\n",
            "i: 256, j: 351\n",
            "i: 363, j: 82\n",
            "i: 235, j: 19\n",
            "i: 61, j: 264\n",
            "i: 180, j: 384\n",
            "i: 171, j: 218\n",
            "i: 179, j: 116\n",
            "i: 57, j: 210\n",
            "i: 380, j: 248\n",
            "i: 368, j: 309\n",
            "i: 296, j: 262\n",
            "i: 45, j: 293\n",
            "i: 308, j: 364\n",
            "i: 303, j: 239\n",
            "i: 155, j: 15\n",
            "i: 174, j: 275\n",
            "i: 245, j: 60\n",
            "i: 337, j: 343\n",
            "i: 287, j: 368\n",
            "i: 153, j: 52\n",
            "i: 314, j: 190\n",
            "i: 21, j: 397\n",
            "i: 299, j: 390\n",
            "i: 92, j: 159\n",
            "i: 339, j: 93\n",
            "i: 174, j: 244\n",
            "i: 114, j: 194\n",
            "i: 172, j: 320\n",
            "i: 362, j: 275\n",
            "i: 6, j: 190\n",
            "i: 173, j: 290\n",
            "i: 228, j: 104\n",
            "i: 234, j: 213\n",
            "i: 289, j: 379\n",
            "i: 294, j: 161\n",
            "i: 220, j: 219\n",
            "i: 244, j: 97\n",
            "i: 81, j: 245\n",
            "i: 26, j: 397\n",
            "i: 98, j: 21\n",
            "i: 253, j: 0\n",
            "i: 81, j: 53\n",
            "i: 83, j: 265\n",
            "i: 108, j: 348\n",
            "i: 379, j: 57\n",
            "i: 351, j: 62\n",
            "i: 219, j: 96\n",
            "i: 270, j: 10\n",
            "i: 82, j: 193\n",
            "i: 82, j: 60\n",
            "i: 215, j: 171\n",
            "i: 262, j: 389\n",
            "i: 115, j: 345\n",
            "i: 234, j: 30\n",
            "i: 56, j: 101\n",
            "i: 268, j: 53\n",
            "i: 53, j: 287\n",
            "i: 367, j: 43\n",
            "i: 316, j: 40\n",
            "i: 4, j: 92\n",
            "i: 130, j: 234\n",
            "i: 297, j: 268\n",
            "i: 74, j: 158\n",
            "i: 336, j: 150\n",
            "i: 292, j: 289\n",
            "i: 313, j: 114\n",
            "i: 164, j: 142\n",
            "i: 390, j: 66\n",
            "i: 374, j: 326\n",
            "i: 253, j: 85\n",
            "i: 168, j: 341\n",
            "i: 102, j: 122\n",
            "i: 64, j: 381\n",
            "i: 317, j: 201\n",
            "i: 242, j: 194\n",
            "i: 30, j: 301\n",
            "i: 93, j: 311\n",
            "i: 9, j: 144\n",
            "i: 50, j: 24\n",
            "i: 219, j: 17\n",
            "i: 76, j: 108\n",
            "i: 156, j: 39\n",
            "i: 340, j: 98\n",
            "i: 10, j: 295\n",
            "i: 230, j: 193\n",
            "i: 223, j: 340\n",
            "i: 242, j: 106\n",
            "i: 140, j: 352\n",
            "i: 344, j: 38\n",
            "i: 365, j: 180\n",
            "i: 251, j: 218\n",
            "i: 169, j: 314\n",
            "i: 8, j: 336\n",
            "i: 16, j: 58\n",
            "i: 303, j: 307\n",
            "i: 240, j: 225\n",
            "i: 100, j: 354\n",
            "i: 229, j: 257\n",
            "i: 3, j: 129\n",
            "i: 125, j: 21\n",
            "i: 116, j: 384\n",
            "i: 161, j: 369\n",
            "i: 259, j: 141\n",
            "i: 385, j: 16\n",
            "i: 200, j: 187\n",
            "i: 354, j: 106\n",
            "i: 124, j: 96\n",
            "i: 161, j: 9\n",
            "i: 319, j: 364\n",
            "i: 53, j: 25\n",
            "i: 45, j: 112\n",
            "i: 181, j: 306\n",
            "i: 369, j: 54\n",
            "i: 274, j: 200\n",
            "i: 139, j: 321\n",
            "i: 102, j: 95\n",
            "i: 224, j: 231\n",
            "i: 234, j: 112\n",
            "i: 301, j: 47\n",
            "i: 394, j: 256\n",
            "i: 280, j: 93\n",
            "i: 354, j: 63\n",
            "i: 174, j: 299\n",
            "i: 94, j: 37\n",
            "i: 360, j: 132\n",
            "i: 365, j: 384\n",
            "i: 221, j: 155\n",
            "i: 105, j: 68\n",
            "i: 55, j: 351\n",
            "i: 124, j: 327\n",
            "i: 107, j: 158\n",
            "i: 129, j: 24\n",
            "i: 54, j: 75\n",
            "i: 278, j: 255\n",
            "i: 134, j: 361\n",
            "i: 68, j: 42\n",
            "i: 57, j: 388\n",
            "i: 102, j: 316\n",
            "i: 258, j: 127\n",
            "i: 113, j: 67\n",
            "i: 231, j: 101\n",
            "i: 254, j: 319\n",
            "i: 253, j: 52\n",
            "i: 185, j: 217\n",
            "i: 278, j: 90\n",
            "i: 0, j: 209\n",
            "i: 330, j: 292\n",
            "i: 312, j: 65\n",
            "i: 303, j: 363\n",
            "i: 67, j: 356\n",
            "i: 199, j: 188\n",
            "i: 368, j: 271\n",
            "i: 278, j: 101\n",
            "i: 366, j: 104\n",
            "i: 183, j: 115\n",
            "i: 316, j: 255\n",
            "i: 326, j: 25\n",
            "i: 396, j: 211\n",
            "i: 309, j: 349\n",
            "i: 237, j: 21\n",
            "i: 276, j: 164\n",
            "i: 103, j: 161\n",
            "i: 83, j: 14\n",
            "i: 166, j: 251\n",
            "i: 390, j: 37\n",
            "i: 285, j: 312\n",
            "i: 393, j: 206\n",
            "i: 157, j: 293\n",
            "i: 275, j: 91\n",
            "i: 115, j: 340\n",
            "i: 8, j: 243\n",
            "i: 59, j: 357\n",
            "i: 364, j: 105\n",
            "i: 228, j: 148\n",
            "i: 302, j: 155\n",
            "i: 148, j: 3\n",
            "i: 360, j: 313\n",
            "i: 179, j: 138\n",
            "i: 340, j: 46\n",
            "i: 7, j: 304\n",
            "i: 12, j: 239\n",
            "i: 24, j: 76\n",
            "i: 43, j: 28\n",
            "i: 131, j: 159\n",
            "i: 373, j: 37\n",
            "i: 32, j: 41\n",
            "i: 114, j: 301\n",
            "i: 312, j: 17\n",
            "i: 321, j: 160\n",
            "i: 170, j: 309\n",
            "i: 311, j: 237\n",
            "i: 12, j: 272\n",
            "i: 344, j: 311\n",
            "i: 375, j: 249\n",
            "i: 372, j: 206\n",
            "i: 148, j: 313\n",
            "i: 13, j: 122\n",
            "i: 182, j: 179\n",
            "i: 164, j: 87\n",
            "i: 214, j: 97\n",
            "i: 202, j: 26\n",
            "i: 14, j: 137\n",
            "i: 394, j: 309\n",
            "i: 278, j: 109\n",
            "i: 153, j: 77\n",
            "i: 305, j: 6\n",
            "i: 131, j: 126\n",
            "i: 9, j: 20\n",
            "i: 383, j: 231\n",
            "i: 304, j: 76\n",
            "i: 114, j: 313\n",
            "i: 373, j: 168\n",
            "i: 152, j: 335\n",
            "i: 200, j: 256\n",
            "i: 356, j: 395\n",
            "i: 39, j: 262\n",
            "i: 243, j: 112\n",
            "i: 137, j: 135\n",
            "i: 17, j: 187\n",
            "i: 295, j: 152\n",
            "i: 289, j: 130\n",
            "i: 119, j: 358\n",
            "i: 157, j: 170\n",
            "i: 268, j: 149\n",
            "i: 140, j: 193\n",
            "i: 248, j: 57\n",
            "i: 358, j: 39\n",
            "i: 213, j: 100\n",
            "i: 332, j: 89\n",
            "i: 350, j: 347\n",
            "i: 106, j: 140\n",
            "i: 190, j: 333\n",
            "i: 268, j: 226\n",
            "i: 330, j: 225\n",
            "i: 359, j: 316\n",
            "i: 398, j: 209\n",
            "i: 107, j: 396\n",
            "i: 277, j: 227\n",
            "i: 175, j: 48\n",
            "i: 377, j: 111\n",
            "i: 314, j: 244\n",
            "i: 158, j: 300\n",
            "i: 385, j: 53\n",
            "i: 11, j: 16\n",
            "i: 36, j: 347\n",
            "i: 356, j: 358\n",
            "i: 119, j: 283\n",
            "i: 216, j: 332\n",
            "i: 164, j: 257\n",
            "i: 48, j: 46\n",
            "i: 247, j: 151\n",
            "i: 368, j: 253\n",
            "i: 239, j: 30\n",
            "i: 158, j: 277\n",
            "i: 335, j: 281\n",
            "i: 242, j: 189\n",
            "i: 317, j: 242\n",
            "i: 65, j: 340\n",
            "i: 276, j: 115\n",
            "i: 325, j: 25\n",
            "i: 230, j: 290\n",
            "i: 123, j: 127\n",
            "i: 303, j: 116\n",
            "i: 347, j: 17\n",
            "i: 145, j: 56\n",
            "i: 197, j: 172\n",
            "i: 213, j: 176\n",
            "i: 72, j: 302\n",
            "i: 196, j: 233\n",
            "i: 65, j: 212\n",
            "i: 80, j: 118\n",
            "i: 19, j: 191\n",
            "i: 236, j: 336\n",
            "i: 232, j: 58\n",
            "i: 350, j: 168\n",
            "i: 132, j: 305\n",
            "i: 66, j: 374\n",
            "i: 361, j: 152\n",
            "i: 18, j: 100\n",
            "i: 133, j: 121\n",
            "i: 41, j: 234\n",
            "i: 58, j: 272\n",
            "i: 249, j: 382\n",
            "i: 164, j: 92\n",
            "i: 156, j: 257\n",
            "i: 53, j: 1\n",
            "i: 104, j: 113\n",
            "i: 185, j: 275\n",
            "i: 91, j: 190\n",
            "i: 170, j: 118\n",
            "i: 158, j: 99\n",
            "i: 327, j: 263\n",
            "i: 388, j: 28\n",
            "i: 131, j: 148\n",
            "i: 198, j: 137\n",
            "i: 43, j: 320\n",
            "i: 29, j: 0\n",
            "i: 299, j: 67\n",
            "i: 116, j: 286\n",
            "i: 0, j: 276\n",
            "i: 370, j: 65\n",
            "i: 301, j: 319\n",
            "i: 147, j: 338\n",
            "i: 88, j: 82\n",
            "i: 299, j: 249\n",
            "i: 19, j: 335\n",
            "i: 254, j: 305\n",
            "i: 159, j: 283\n",
            "i: 122, j: 347\n",
            "i: 366, j: 348\n",
            "i: 183, j: 61\n",
            "i: 107, j: 112\n",
            "i: 78, j: 79\n",
            "i: 143, j: 264\n",
            "i: 244, j: 156\n",
            "i: 228, j: 151\n",
            "i: 98, j: 175\n",
            "i: 105, j: 69\n",
            "i: 236, j: 244\n",
            "i: 83, j: 262\n",
            "i: 113, j: 390\n",
            "i: 220, j: 199\n",
            "i: 141, j: 111\n",
            "i: 178, j: 332\n",
            "i: 157, j: 236\n",
            "i: 347, j: 67\n",
            "i: 358, j: 221\n",
            "i: 192, j: 229\n",
            "i: 209, j: 372\n",
            "i: 108, j: 314\n",
            "i: 26, j: 130\n",
            "i: 291, j: 267\n",
            "i: 381, j: 25\n",
            "i: 336, j: 246\n",
            "i: 295, j: 24\n",
            "i: 211, j: 318\n",
            "i: 305, j: 89\n",
            "i: 14, j: 7\n",
            "i: 305, j: 129\n",
            "i: 90, j: 35\n",
            "i: 16, j: 139\n",
            "i: 170, j: 11\n",
            "i: 362, j: 153\n",
            "i: 200, j: 52\n",
            "i: 194, j: 208\n",
            "i: 377, j: 169\n",
            "i: 384, j: 105\n",
            "i: 355, j: 127\n",
            "i: 227, j: 196\n",
            "i: 160, j: 231\n",
            "i: 273, j: 377\n",
            "i: 75, j: 291\n",
            "i: 49, j: 376\n",
            "i: 398, j: 144\n",
            "i: 231, j: 317\n",
            "i: 59, j: 69\n",
            "i: 155, j: 273\n",
            "i: 127, j: 334\n",
            "i: 256, j: 174\n",
            "i: 393, j: 347\n",
            "i: 210, j: 293\n",
            "i: 171, j: 393\n",
            "i: 34, j: 185\n",
            "i: 252, j: 7\n",
            "i: 57, j: 75\n",
            "i: 69, j: 340\n",
            "i: 131, j: 141\n",
            "i: 65, j: 122\n",
            "i: 82, j: 178\n",
            "i: 174, j: 349\n",
            "i: 256, j: 113\n",
            "i: 386, j: 120\n",
            "i: 302, j: 23\n",
            "i: 263, j: 341\n",
            "i: 99, j: 278\n",
            "i: 250, j: 190\n",
            "i: 89, j: 202\n",
            "i: 240, j: 253\n",
            "i: 104, j: 229\n",
            "i: 279, j: 196\n",
            "i: 359, j: 154\n",
            "i: 116, j: 61\n",
            "i: 264, j: 372\n",
            "i: 236, j: 212\n",
            "i: 318, j: 374\n",
            "i: 139, j: 161\n",
            "i: 44, j: 49\n",
            "i: 392, j: 77\n",
            "i: 329, j: 115\n",
            "i: 253, j: 273\n",
            "i: 92, j: 30\n",
            "i: 235, j: 95\n",
            "i: 269, j: 382\n",
            "i: 246, j: 66\n",
            "i: 13, j: 121\n",
            "i: 95, j: 100\n",
            "tensor([[0., 0., 0.,  ..., 0., 0., 0.],\n",
            "        [0., 0., 0.,  ..., 0., 0., 0.],\n",
            "        [0., 0., 0.,  ..., 0., 0., 0.],\n",
            "        ...,\n",
            "        [0., 0., 0.,  ..., 0., 0., 0.],\n",
            "        [0., 0., 0.,  ..., 0., 0., 0.],\n",
            "        [0., 0., 0.,  ..., 0., 0., 0.]])\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "incompatible_pairs.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0UNpVdXGRjm3",
        "outputId": "425f0669-6548-4401-fc45-e3851e3bd62b"
      },
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "torch.Size([400, 400])"
            ]
          },
          "metadata": {},
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def group_conflict_matrix(groups_number, conflict_matrix):\n",
        "    \"\"\"\n",
        "    Generates a matrix representing conflicts between groups.\n",
        "\n",
        "    Args:\n",
        "        groups_number (list): List of groups with their member indices.\n",
        "        conflict_matrix (torch.Tensor): A matrix representing conflicts between people.\n",
        "\n",
        "    Returns:\n",
        "        torch.Tensor: A matrix representing conflicts between groups.\n",
        "    \"\"\"\n",
        "    num_groups = len(groups_number)\n",
        "    group_conflicts = torch.zeros(num_groups, num_groups)\n",
        "    xs, ys = [], []\n",
        "\n",
        "    for i, group1 in enumerate(groups_number):\n",
        "        for j, group2 in enumerate(groups_number):\n",
        "            if i != j:\n",
        "                conflicts = 0\n",
        "                for member1 in group1:\n",
        "                    for member2 in group2:\n",
        "                        conflicts += conflict_matrix[member1, member2]\n",
        "                group_conflicts[i, j] = conflicts\n",
        "                if conflicts > 0:\n",
        "                  print(f\"i: {i+1}, j: {j+1}\")\n",
        "                else:\n",
        "                  xs.append(i)\n",
        "                  ys.append(j)\n",
        "\n",
        "    return group_conflicts, xs, ys\n",
        "\n",
        "# Generate the group-to-group conflict matrix\n",
        "group_conflicts, xs, ys = group_conflict_matrix(groups_number, incompatible_pairs)\n",
        "\n",
        "print(group_conflicts)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4KT0GtrbCQXe",
        "outputId": "140d8700-af19-4e71-cc94-947294f05a4c"
      },
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "i: 1, j: 12\n",
            "i: 1, j: 18\n",
            "i: 1, j: 24\n",
            "i: 1, j: 39\n",
            "i: 1, j: 43\n",
            "i: 1, j: 64\n",
            "i: 2, j: 1\n",
            "i: 2, j: 14\n",
            "i: 2, j: 32\n",
            "i: 2, j: 44\n",
            "i: 2, j: 45\n",
            "i: 2, j: 61\n",
            "i: 2, j: 64\n",
            "i: 3, j: 9\n",
            "i: 3, j: 11\n",
            "i: 3, j: 13\n",
            "i: 3, j: 22\n",
            "i: 3, j: 23\n",
            "i: 3, j: 29\n",
            "i: 3, j: 37\n",
            "i: 3, j: 64\n",
            "i: 3, j: 72\n",
            "i: 3, j: 76\n",
            "i: 4, j: 2\n",
            "i: 4, j: 14\n",
            "i: 4, j: 20\n",
            "i: 4, j: 21\n",
            "i: 4, j: 34\n",
            "i: 4, j: 47\n",
            "i: 4, j: 72\n",
            "i: 4, j: 78\n",
            "i: 5, j: 2\n",
            "i: 5, j: 3\n",
            "i: 5, j: 7\n",
            "i: 5, j: 9\n",
            "i: 5, j: 19\n",
            "i: 5, j: 51\n",
            "i: 5, j: 70\n",
            "i: 5, j: 72\n",
            "i: 5, j: 77\n",
            "i: 5, j: 78\n",
            "i: 6, j: 1\n",
            "i: 6, j: 13\n",
            "i: 6, j: 29\n",
            "i: 6, j: 30\n",
            "i: 6, j: 46\n",
            "i: 6, j: 51\n",
            "i: 6, j: 57\n",
            "i: 6, j: 58\n",
            "i: 6, j: 63\n",
            "i: 6, j: 65\n",
            "i: 6, j: 66\n",
            "i: 6, j: 72\n",
            "i: 6, j: 73\n",
            "i: 6, j: 77\n",
            "i: 7, j: 9\n",
            "i: 7, j: 18\n",
            "i: 7, j: 21\n",
            "i: 7, j: 30\n",
            "i: 7, j: 47\n",
            "i: 7, j: 54\n",
            "i: 7, j: 64\n",
            "i: 7, j: 65\n",
            "i: 7, j: 74\n",
            "i: 7, j: 78\n",
            "i: 8, j: 23\n",
            "i: 8, j: 27\n",
            "i: 8, j: 30\n",
            "i: 8, j: 58\n",
            "i: 8, j: 75\n",
            "i: 9, j: 15\n",
            "i: 9, j: 20\n",
            "i: 9, j: 21\n",
            "i: 10, j: 37\n",
            "i: 11, j: 16\n",
            "i: 11, j: 24\n",
            "i: 11, j: 29\n",
            "i: 11, j: 43\n",
            "i: 11, j: 44\n",
            "i: 11, j: 79\n",
            "i: 12, j: 9\n",
            "i: 12, j: 10\n",
            "i: 12, j: 26\n",
            "i: 12, j: 32\n",
            "i: 12, j: 39\n",
            "i: 12, j: 51\n",
            "i: 12, j: 59\n",
            "i: 12, j: 68\n",
            "i: 12, j: 72\n",
            "i: 12, j: 76\n",
            "i: 13, j: 1\n",
            "i: 13, j: 35\n",
            "i: 13, j: 56\n",
            "i: 13, j: 61\n",
            "i: 13, j: 65\n",
            "i: 14, j: 16\n",
            "i: 14, j: 29\n",
            "i: 14, j: 55\n",
            "i: 14, j: 60\n",
            "i: 14, j: 73\n",
            "i: 14, j: 77\n",
            "i: 15, j: 1\n",
            "i: 15, j: 26\n",
            "i: 15, j: 38\n",
            "i: 15, j: 41\n",
            "i: 15, j: 53\n",
            "i: 15, j: 65\n",
            "i: 15, j: 74\n",
            "i: 15, j: 78\n",
            "i: 15, j: 79\n",
            "i: 16, j: 7\n",
            "i: 16, j: 17\n",
            "i: 16, j: 33\n",
            "i: 16, j: 48\n",
            "i: 16, j: 64\n",
            "i: 17, j: 9\n",
            "i: 17, j: 50\n",
            "i: 17, j: 56\n",
            "i: 17, j: 59\n",
            "i: 17, j: 71\n",
            "i: 18, j: 1\n",
            "i: 18, j: 2\n",
            "i: 18, j: 10\n",
            "i: 18, j: 15\n",
            "i: 18, j: 31\n",
            "i: 18, j: 53\n",
            "i: 18, j: 66\n",
            "i: 18, j: 70\n",
            "i: 18, j: 73\n",
            "i: 18, j: 74\n",
            "i: 18, j: 80\n",
            "i: 19, j: 4\n",
            "i: 19, j: 28\n",
            "i: 19, j: 32\n",
            "i: 19, j: 33\n",
            "i: 19, j: 54\n",
            "i: 20, j: 4\n",
            "i: 20, j: 7\n",
            "i: 20, j: 17\n",
            "i: 20, j: 28\n",
            "i: 20, j: 30\n",
            "i: 20, j: 37\n",
            "i: 20, j: 62\n",
            "i: 20, j: 67\n",
            "i: 20, j: 80\n",
            "i: 21, j: 4\n",
            "i: 21, j: 52\n",
            "i: 21, j: 55\n",
            "i: 21, j: 78\n",
            "i: 22, j: 17\n",
            "i: 22, j: 27\n",
            "i: 22, j: 28\n",
            "i: 22, j: 31\n",
            "i: 22, j: 42\n",
            "i: 22, j: 52\n",
            "i: 22, j: 78\n",
            "i: 22, j: 79\n",
            "i: 23, j: 14\n",
            "i: 23, j: 19\n",
            "i: 23, j: 59\n",
            "i: 23, j: 60\n",
            "i: 23, j: 64\n",
            "i: 24, j: 9\n",
            "i: 24, j: 13\n",
            "i: 24, j: 18\n",
            "i: 24, j: 26\n",
            "i: 24, j: 27\n",
            "i: 24, j: 40\n",
            "i: 24, j: 52\n",
            "i: 24, j: 60\n",
            "i: 24, j: 62\n",
            "i: 25, j: 7\n",
            "i: 25, j: 28\n",
            "i: 25, j: 41\n",
            "i: 25, j: 45\n",
            "i: 25, j: 49\n",
            "i: 25, j: 50\n",
            "i: 25, j: 65\n",
            "i: 25, j: 75\n",
            "i: 25, j: 79\n",
            "i: 26, j: 3\n",
            "i: 26, j: 14\n",
            "i: 26, j: 33\n",
            "i: 26, j: 45\n",
            "i: 26, j: 49\n",
            "i: 26, j: 53\n",
            "i: 26, j: 57\n",
            "i: 26, j: 79\n",
            "i: 27, j: 13\n",
            "i: 27, j: 35\n",
            "i: 27, j: 36\n",
            "i: 27, j: 38\n",
            "i: 27, j: 56\n",
            "i: 27, j: 63\n",
            "i: 27, j: 69\n",
            "i: 27, j: 72\n",
            "i: 27, j: 78\n",
            "i: 28, j: 3\n",
            "i: 28, j: 8\n",
            "i: 28, j: 12\n",
            "i: 28, j: 27\n",
            "i: 28, j: 38\n",
            "i: 29, j: 7\n",
            "i: 29, j: 8\n",
            "i: 29, j: 14\n",
            "i: 29, j: 20\n",
            "i: 29, j: 23\n",
            "i: 29, j: 47\n",
            "i: 29, j: 61\n",
            "i: 29, j: 72\n",
            "i: 30, j: 6\n",
            "i: 30, j: 10\n",
            "i: 30, j: 12\n",
            "i: 30, j: 15\n",
            "i: 30, j: 25\n",
            "i: 30, j: 61\n",
            "i: 30, j: 65\n",
            "i: 31, j: 3\n",
            "i: 31, j: 33\n",
            "i: 31, j: 42\n",
            "i: 31, j: 58\n",
            "i: 31, j: 60\n",
            "i: 31, j: 64\n",
            "i: 31, j: 76\n",
            "i: 32, j: 4\n",
            "i: 32, j: 11\n",
            "i: 32, j: 24\n",
            "i: 32, j: 53\n",
            "i: 32, j: 54\n",
            "i: 32, j: 63\n",
            "i: 32, j: 75\n",
            "i: 33, j: 6\n",
            "i: 33, j: 9\n",
            "i: 33, j: 11\n",
            "i: 33, j: 12\n",
            "i: 33, j: 28\n",
            "i: 33, j: 64\n",
            "i: 33, j: 78\n",
            "i: 33, j: 79\n",
            "i: 34, j: 27\n",
            "i: 34, j: 28\n",
            "i: 34, j: 60\n",
            "i: 35, j: 10\n",
            "i: 35, j: 24\n",
            "i: 35, j: 29\n",
            "i: 35, j: 39\n",
            "i: 35, j: 47\n",
            "i: 35, j: 59\n",
            "i: 35, j: 68\n",
            "i: 36, j: 13\n",
            "i: 36, j: 21\n",
            "i: 36, j: 48\n",
            "i: 36, j: 55\n",
            "i: 36, j: 74\n",
            "i: 36, j: 80\n",
            "i: 37, j: 22\n",
            "i: 37, j: 51\n",
            "i: 37, j: 58\n",
            "i: 37, j: 77\n",
            "i: 38, j: 44\n",
            "i: 38, j: 50\n",
            "i: 38, j: 59\n",
            "i: 38, j: 68\n",
            "i: 39, j: 9\n",
            "i: 39, j: 10\n",
            "i: 39, j: 13\n",
            "i: 39, j: 47\n",
            "i: 39, j: 59\n",
            "i: 39, j: 75\n",
            "i: 40, j: 11\n",
            "i: 40, j: 19\n",
            "i: 40, j: 45\n",
            "i: 40, j: 52\n",
            "i: 41, j: 3\n",
            "i: 41, j: 27\n",
            "i: 41, j: 31\n",
            "i: 41, j: 44\n",
            "i: 41, j: 64\n",
            "i: 41, j: 74\n",
            "i: 42, j: 22\n",
            "i: 42, j: 29\n",
            "i: 42, j: 79\n",
            "i: 43, j: 25\n",
            "i: 43, j: 51\n",
            "i: 43, j: 52\n",
            "i: 43, j: 59\n",
            "i: 43, j: 62\n",
            "i: 43, j: 73\n",
            "i: 44, j: 17\n",
            "i: 44, j: 19\n",
            "i: 44, j: 27\n",
            "i: 44, j: 34\n",
            "i: 44, j: 43\n",
            "i: 44, j: 47\n",
            "i: 44, j: 49\n",
            "i: 44, j: 50\n",
            "i: 44, j: 57\n",
            "i: 44, j: 74\n",
            "i: 45, j: 16\n",
            "i: 45, j: 22\n",
            "i: 45, j: 25\n",
            "i: 45, j: 29\n",
            "i: 45, j: 46\n",
            "i: 45, j: 57\n",
            "i: 45, j: 61\n",
            "i: 45, j: 62\n",
            "i: 46, j: 27\n",
            "i: 46, j: 29\n",
            "i: 46, j: 38\n",
            "i: 46, j: 43\n",
            "i: 46, j: 49\n",
            "i: 46, j: 52\n",
            "i: 47, j: 5\n",
            "i: 47, j: 10\n",
            "i: 47, j: 12\n",
            "i: 47, j: 25\n",
            "i: 47, j: 30\n",
            "i: 47, j: 38\n",
            "i: 47, j: 51\n",
            "i: 47, j: 63\n",
            "i: 47, j: 65\n",
            "i: 47, j: 67\n",
            "i: 47, j: 71\n",
            "i: 48, j: 34\n",
            "i: 48, j: 43\n",
            "i: 49, j: 31\n",
            "i: 49, j: 36\n",
            "i: 49, j: 41\n",
            "i: 49, j: 47\n",
            "i: 49, j: 58\n",
            "i: 49, j: 68\n",
            "i: 49, j: 73\n",
            "i: 50, j: 5\n",
            "i: 50, j: 13\n",
            "i: 50, j: 36\n",
            "i: 50, j: 44\n",
            "i: 50, j: 51\n",
            "i: 50, j: 72\n",
            "i: 51, j: 3\n",
            "i: 51, j: 6\n",
            "i: 51, j: 11\n",
            "i: 51, j: 24\n",
            "i: 51, j: 53\n",
            "i: 51, j: 54\n",
            "i: 51, j: 69\n",
            "i: 52, j: 6\n",
            "i: 52, j: 12\n",
            "i: 52, j: 26\n",
            "i: 52, j: 50\n",
            "i: 52, j: 55\n",
            "i: 52, j: 61\n",
            "i: 52, j: 67\n",
            "i: 52, j: 70\n",
            "i: 52, j: 75\n",
            "i: 52, j: 80\n",
            "i: 53, j: 24\n",
            "i: 53, j: 25\n",
            "i: 53, j: 36\n",
            "i: 53, j: 41\n",
            "i: 53, j: 43\n",
            "i: 53, j: 50\n",
            "i: 53, j: 51\n",
            "i: 53, j: 80\n",
            "i: 54, j: 5\n",
            "i: 54, j: 34\n",
            "i: 54, j: 37\n",
            "i: 54, j: 43\n",
            "i: 54, j: 51\n",
            "i: 54, j: 52\n",
            "i: 54, j: 58\n",
            "i: 54, j: 59\n",
            "i: 54, j: 63\n",
            "i: 55, j: 21\n",
            "i: 55, j: 34\n",
            "i: 55, j: 46\n",
            "i: 55, j: 47\n",
            "i: 55, j: 54\n",
            "i: 55, j: 69\n",
            "i: 55, j: 76\n",
            "i: 56, j: 13\n",
            "i: 56, j: 18\n",
            "i: 56, j: 37\n",
            "i: 56, j: 41\n",
            "i: 56, j: 65\n",
            "i: 56, j: 74\n",
            "i: 57, j: 21\n",
            "i: 57, j: 29\n",
            "i: 57, j: 66\n",
            "i: 57, j: 67\n",
            "i: 57, j: 74\n",
            "i: 58, j: 13\n",
            "i: 58, j: 27\n",
            "i: 58, j: 29\n",
            "i: 58, j: 41\n",
            "i: 58, j: 52\n",
            "i: 58, j: 60\n",
            "i: 59, j: 6\n",
            "i: 59, j: 24\n",
            "i: 59, j: 39\n",
            "i: 59, j: 60\n",
            "i: 59, j: 63\n",
            "i: 59, j: 74\n",
            "i: 60, j: 6\n",
            "i: 60, j: 14\n",
            "i: 60, j: 25\n",
            "i: 60, j: 27\n",
            "i: 60, j: 42\n",
            "i: 60, j: 61\n",
            "i: 60, j: 72\n",
            "i: 60, j: 80\n",
            "i: 61, j: 1\n",
            "i: 61, j: 7\n",
            "i: 61, j: 8\n",
            "i: 61, j: 15\n",
            "i: 61, j: 21\n",
            "i: 61, j: 38\n",
            "i: 61, j: 42\n",
            "i: 61, j: 55\n",
            "i: 61, j: 73\n",
            "i: 62, j: 8\n",
            "i: 62, j: 18\n",
            "i: 62, j: 23\n",
            "i: 62, j: 30\n",
            "i: 62, j: 31\n",
            "i: 62, j: 36\n",
            "i: 62, j: 39\n",
            "i: 62, j: 46\n",
            "i: 62, j: 48\n",
            "i: 62, j: 63\n",
            "i: 62, j: 68\n",
            "i: 62, j: 77\n",
            "i: 63, j: 10\n",
            "i: 63, j: 11\n",
            "i: 63, j: 17\n",
            "i: 63, j: 34\n",
            "i: 63, j: 49\n",
            "i: 63, j: 55\n",
            "i: 63, j: 56\n",
            "i: 63, j: 58\n",
            "i: 63, j: 65\n",
            "i: 63, j: 76\n",
            "i: 64, j: 15\n",
            "i: 64, j: 35\n",
            "i: 64, j: 40\n",
            "i: 64, j: 43\n",
            "i: 64, j: 55\n",
            "i: 64, j: 74\n",
            "i: 65, j: 7\n",
            "i: 65, j: 18\n",
            "i: 65, j: 25\n",
            "i: 65, j: 31\n",
            "i: 65, j: 41\n",
            "i: 65, j: 51\n",
            "i: 65, j: 74\n",
            "i: 66, j: 12\n",
            "i: 66, j: 41\n",
            "i: 66, j: 48\n",
            "i: 66, j: 80\n",
            "i: 67, j: 5\n",
            "i: 67, j: 6\n",
            "i: 67, j: 20\n",
            "i: 67, j: 30\n",
            "i: 67, j: 36\n",
            "i: 67, j: 42\n",
            "i: 67, j: 48\n",
            "i: 67, j: 50\n",
            "i: 67, j: 57\n",
            "i: 67, j: 60\n",
            "i: 67, j: 74\n",
            "i: 68, j: 14\n",
            "i: 68, j: 21\n",
            "i: 68, j: 26\n",
            "i: 68, j: 41\n",
            "i: 68, j: 44\n",
            "i: 68, j: 53\n",
            "i: 68, j: 62\n",
            "i: 69, j: 6\n",
            "i: 69, j: 7\n",
            "i: 69, j: 9\n",
            "i: 69, j: 16\n",
            "i: 69, j: 18\n",
            "i: 69, j: 34\n",
            "i: 69, j: 52\n",
            "i: 69, j: 54\n",
            "i: 69, j: 65\n",
            "i: 69, j: 77\n",
            "i: 69, j: 78\n",
            "i: 70, j: 14\n",
            "i: 70, j: 17\n",
            "i: 70, j: 27\n",
            "i: 70, j: 35\n",
            "i: 70, j: 77\n",
            "i: 70, j: 79\n",
            "i: 71, j: 2\n",
            "i: 71, j: 10\n",
            "i: 71, j: 15\n",
            "i: 71, j: 22\n",
            "i: 71, j: 24\n",
            "i: 71, j: 35\n",
            "i: 71, j: 56\n",
            "i: 71, j: 59\n",
            "i: 71, j: 60\n",
            "i: 71, j: 66\n",
            "i: 71, j: 70\n",
            "i: 72, j: 6\n",
            "i: 72, j: 13\n",
            "i: 72, j: 23\n",
            "i: 72, j: 40\n",
            "i: 72, j: 47\n",
            "i: 72, j: 49\n",
            "i: 72, j: 75\n",
            "i: 73, j: 15\n",
            "i: 73, j: 48\n",
            "i: 73, j: 50\n",
            "i: 73, j: 51\n",
            "i: 73, j: 60\n",
            "i: 73, j: 66\n",
            "i: 73, j: 68\n",
            "i: 73, j: 79\n",
            "i: 74, j: 1\n",
            "i: 74, j: 12\n",
            "i: 74, j: 24\n",
            "i: 74, j: 49\n",
            "i: 74, j: 50\n",
            "i: 74, j: 59\n",
            "i: 74, j: 64\n",
            "i: 74, j: 65\n",
            "i: 75, j: 18\n",
            "i: 75, j: 45\n",
            "i: 75, j: 60\n",
            "i: 75, j: 62\n",
            "i: 75, j: 76\n",
            "i: 75, j: 78\n",
            "i: 76, j: 6\n",
            "i: 76, j: 7\n",
            "i: 76, j: 8\n",
            "i: 76, j: 30\n",
            "i: 76, j: 32\n",
            "i: 76, j: 33\n",
            "i: 76, j: 42\n",
            "i: 76, j: 52\n",
            "i: 76, j: 70\n",
            "i: 77, j: 6\n",
            "i: 77, j: 17\n",
            "i: 77, j: 22\n",
            "i: 77, j: 61\n",
            "i: 77, j: 66\n",
            "i: 77, j: 71\n",
            "i: 78, j: 3\n",
            "i: 78, j: 6\n",
            "i: 78, j: 12\n",
            "i: 78, j: 15\n",
            "i: 78, j: 26\n",
            "i: 78, j: 42\n",
            "i: 78, j: 67\n",
            "i: 78, j: 74\n",
            "i: 78, j: 75\n",
            "i: 79, j: 21\n",
            "i: 79, j: 31\n",
            "i: 79, j: 36\n",
            "i: 79, j: 51\n",
            "i: 79, j: 61\n",
            "i: 80, j: 15\n",
            "i: 80, j: 27\n",
            "i: 80, j: 70\n",
            "i: 80, j: 75\n",
            "tensor([[0., 0., 0.,  ..., 0., 0., 0.],\n",
            "        [1., 0., 0.,  ..., 0., 0., 0.],\n",
            "        [0., 0., 0.,  ..., 0., 0., 0.],\n",
            "        ...,\n",
            "        [0., 0., 1.,  ..., 0., 0., 0.],\n",
            "        [0., 0., 0.,  ..., 0., 0., 0.],\n",
            "        [0., 0., 0.,  ..., 0., 0., 0.]])\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "group_conflicts.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "nlhq5dOcRgLo",
        "outputId": "f2e3bba4-cfa2-4db9-e0ac-e884266b9c54"
      },
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "torch.Size([80, 80])"
            ]
          },
          "metadata": {},
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(len(xs),len(ys))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "sNziqhPg24D5",
        "outputId": "29952ce7-07db-42a3-d92e-d11a0a9e7e7f"
      },
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "5754 5754\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "group_conflicts[0]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "SzJL-i4vCzX2",
        "outputId": "a48d4593-efaa-40cf-bfcf-672cc4db72d1"
      },
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "tensor([0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 3., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
              "        0., 0., 0., 1., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0.,\n",
              "        0., 0., 0., 0., 0., 0., 0., 0., 0., 0.])"
            ]
          },
          "metadata": {},
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import matplotlib.pyplot as plt\n",
        "%matplotlib inline\n",
        "\n",
        "# Plot the group conflict matrix with row and column numbers\n",
        "fig, ax = plt.subplots(figsize=(16, 16))\n",
        "cax = ax.matshow(group_conflicts, cmap='Blues')\n",
        "\n",
        "# Add colorbar\n",
        "fig.colorbar(cax)\n",
        "\n",
        "# Set axis labels\n",
        "ax.set_xticks(range(len(group_conflicts)))\n",
        "ax.set_yticks(range(len(group_conflicts)))\n",
        "ax.set_xticklabels(range(1, len(group_conflicts) + 1))\n",
        "ax.set_yticklabels(range(1, len(group_conflicts) + 1))\n",
        "\n",
        "# Rotate the x-axis labels\n",
        "plt.xticks(rotation=90)\n",
        "\n",
        "# Add text annotations\n",
        "for i in range(len(group_conflicts)):\n",
        "    for j in range(len(group_conflicts)):\n",
        "          plt.text(j, i, int(group_conflicts[i, j].item()), ha=\"center\", va=\"center\", color='gray')\n",
        "\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "_f3MIKrAH5Bb",
        "outputId": "f2e9926a-f0c9-4a16-bb25-25065ba81b43"
      },
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1600x1600 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "group_conflicts[0]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5qMV-6I7RtaY",
        "outputId": "5916c952-9563-4ea7-8ae8-c9501470c368"
      },
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "tensor([0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 3., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
              "        0., 0., 0., 1., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0.,\n",
              "        0., 0., 0., 0., 0., 0., 0., 0., 0., 0.])"
            ]
          },
          "metadata": {},
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "p = group_conflicts[0].float()\n",
        "p = p / p.sum()\n",
        "p.numpy()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "gioNO-vDJTIj",
        "outputId": "d9c5674d-fef9-4a3d-e8dd-74a2143fee3a"
      },
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([0.   , 0.   , 0.   , 0.   , 0.   , 0.   , 0.   , 0.   , 0.   ,\n",
              "       0.   , 0.   , 0.125, 0.   , 0.   , 0.   , 0.   , 0.   , 0.375,\n",
              "       0.   , 0.   , 0.   , 0.   , 0.   , 0.125, 0.   , 0.   , 0.   ,\n",
              "       0.   , 0.   , 0.   , 0.   , 0.   , 0.   , 0.   , 0.   , 0.   ,\n",
              "       0.   , 0.   , 0.125, 0.   , 0.   , 0.   , 0.125, 0.   , 0.   ,\n",
              "       0.   , 0.   , 0.   , 0.   , 0.   , 0.   , 0.   , 0.   , 0.   ,\n",
              "       0.   , 0.   , 0.   , 0.   , 0.   , 0.   , 0.   , 0.   , 0.   ,\n",
              "       0.125, 0.   , 0.   , 0.   , 0.   , 0.   , 0.   , 0.   , 0.   ,\n",
              "       0.   , 0.   , 0.   , 0.   , 0.   , 0.   , 0.   , 0.   ],\n",
              "      dtype=float32)"
            ]
          },
          "metadata": {},
          "execution_count": 11
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "g = torch.Generator().manual_seed(2147483647)\n",
        "ix = torch.multinomial(p, num_samples=1, replacement=True, generator=g).item()\n",
        "ix"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "M-3KqrsKKKjo",
        "outputId": "380908ed-2707-4c11-a5c8-1f597bd21dc1"
      },
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "42"
            ]
          },
          "metadata": {},
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "P = group_conflicts.float()\n",
        "P /= P.sum(1, keepdims=True)\n",
        "P[10].numpy()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "D6TXD6X4K3I5",
        "outputId": "632b01d2-7f51-4371-dc30-f51bae228a68"
      },
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([0.        , 0.        , 0.        , 0.        , 0.        ,\n",
              "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
              "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
              "       0.2857143 , 0.        , 0.        , 0.        , 0.        ,\n",
              "       0.        , 0.        , 0.        , 0.14285715, 0.        ,\n",
              "       0.        , 0.        , 0.        , 0.14285715, 0.        ,\n",
              "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
              "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
              "       0.        , 0.        , 0.14285715, 0.14285715, 0.        ,\n",
              "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
              "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
              "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
              "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
              "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
              "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
              "       0.        , 0.        , 0.        , 0.14285715, 0.        ],\n",
              "      dtype=float32)"
            ]
          },
          "metadata": {},
          "execution_count": 14
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "P.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "_StD6GFIR10T",
        "outputId": "62d7eb16-d018-4dfa-e51f-a94ac675c10d"
      },
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "torch.Size([80, 80])"
            ]
          },
          "metadata": {},
          "execution_count": 17
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "g = torch.Generator().manual_seed(2147483647)\n",
        "\n",
        "for i in range(5):\n",
        "\n",
        "  out = []\n",
        "  ix = 0\n",
        "  while True:\n",
        "    p = P[ix]\n",
        "    ix = torch.multinomial(p, num_samples=1, replacement=True, generator=g).item()\n",
        "    out.append(str(ix)+\",\")\n",
        "    if ix == 0:\n",
        "      break\n",
        "  print(''.join(out))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ZMmrjQmNK_HO",
        "outputId": "5384b15e-b956-43a9-81ba-46c30bba6078"
      },
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "42,72,14,73,11,31,52,40,73,11,58,59,71,39,10,78,60,72,49,35,73,64,30,32,63,14,0,\n",
            "11,31,74,77,14,78,20,3,33,26,55,40,63,54,20,54,20,54,68,51,49,71,5,62,55,36,76,21,26,62,75,5,56,66,47,42,51,74,77,25,44,60,14,40,26,34,67,20,54,45,26,12,64,17,9,36,50,10,15,6,29,11,25,48,40,43,18,53,4,69,16,8,20,54,68,5,76,21,16,55,36,76,16,58,38,58,62,55,40,43,18,53,58,5,29,9,36,57,59,41,28,22,13,54,46,70,58,62,54,75,29,9,36,50,23,25,48,35,54,46,11,9,36,57,40,43,73,63,39,44,61,62,33,27,37,67,40,30,63,54,45,37,58,23,61,67,13,76,70,65,79,14,40,30,75,6,64,17,69,13,15,16,55,64,17,72,67,52,50,52,50,23,39,44,60,20,77,73,11,25,2,21,41,21,78,30,63,34,67,25,2,22,18,32,10,43,26,35,47,33,27,2,10,42,72,49,71,5,71,12,34,28,22,13,76,70,55,40,30,57,40,43,33,59,60,54,75,31,62,75,7,29,64,6,64,24,64,50,5,57,28,46,24,49,12,60,7,74,17,69,13,28,13,76,65,11,25,44,61,22,18,53,51,5,12,64,17,30,59,41,78,20,51,79,14,78,60,7,29,64,40,26,77,74,17,65,40,73,58,73,23,59,26,71,46,50,52,49,50,23,39,10,78,20,51,11,58,73,64,6,20,51,5,62,54,46,24,48,46,29,14,52,23,8,19,27,11,67,20,77,73,11,50,68,77,66,35,73,23,17,65,40,2,12,64,30,63,73,49,12,55,40,30,32,5,62,33,26,62,16,70,21,51,25,2,22,13,15,6,17,72,78,20,3,33,27,26,55,36,57,28,13,76,70,14,77,66,47,42,24,49,4,2,10,23,59,71,48,40,2,36,50,52,23,39,18,3,33,27,11,9,36,21,51,54,45,42,24,64,40,30,63,73,23,25,56,28,19,6,64,17,79,69,78,35,12,34,28,13,54,33,26,35,54,68,76,5,72,47,33,59,71,22,63,14,77,11,25,52,23,25,56,73,11,8,20,51,60,37,67,20,3,1,31,74,77,66,4,50,23,12,64,50,5,72,78,60,7,26,68,51,79,26,37,43,49,50,53,4,71,48,67,20,54,53,4,71,12,0,\n",
            "17,14,64,6,77,14,77,66,73,64,17,73,64,24,40,73,23,12,34,58,73,11,58,23,39,18,53,36,21,26,62,33,27,7,26,55,73,64,50,2,63,54,46,50,2,36,21,78,50,2,75,32,77,41,78,60,37,49,50,68,51,54,20,51,49,4,77,25,48,67,52,35,20,51,11,9,36,76,21,51,66,41,78,20,51,79,26,62,9,36,50,10,78,20,51,49,50,52,23,26,12,55,36,57,26,12,34,9,36,21,26,55,12,55,64,6,29,64,30,75,31,3,13,15,16,55,17,1,63,42,58,23,59,26,62,16,70,55,40,2,36,76,65,11,58,23,26,12,0,\n",
            "38,9,36,76,21,27,26,71,39,10,28,22,58,38,74,44,45,28,19,66,47,42,61,76,16,55,12,0,\n",
            "23,26,37,43,16,70,55,64,50,68,64,30,59,26,37,49,71,22,13,15,6,63,54,33,26,12,64,40,26,71,39,10,42,50,68,17,0,\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "P = torch.tensor(P, dtype=torch.float32)\n",
        "\n",
        "log_likelihood = 0.0\n",
        "n = 0\n",
        "\n",
        "epsilon = 1e-10  # Small value to avoid log(0)\n",
        "\n",
        "for i, i_group in enumerate(groups_number):\n",
        "    for j, j_group in enumerate(groups_number):\n",
        "        if i != j:\n",
        "            prob = P[i, j].clamp(min=epsilon)  # Clamp the probability to avoid log(0)\n",
        "            logprob = torch.log(prob)\n",
        "            log_likelihood += logprob\n",
        "            n += 1\n",
        "\n",
        "print(f'{log_likelihood=}')\n",
        "nll = -log_likelihood\n",
        "print(f'{nll=}')\n",
        "print(f'{nll/n}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Bg3LVbNtze0A",
        "outputId": "f230bb96-2443-4220-b90b-8f2d3e0892a6"
      },
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "log_likelihood=tensor(-133637)\n",
            "nll=tensor(133637)\n",
            "21.145099639892578\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "<ipython-input-18-d028af947055>:1: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).\n",
            "  P = torch.tensor(P, dtype=torch.float32)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "xs = torch.tensor(xs)\n",
        "ys = torch.tensor(ys)\n",
        "print(xs,ys)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "CL6gHJFt3S5A",
        "outputId": "2ba25eb7-8493-4092-a04d-3909206427b2"
      },
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "tensor([ 0,  0,  0,  ..., 79, 79, 79]) tensor([ 1,  2,  3,  ..., 76, 77, 78])\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import torch\n",
        "import torch.nn.functional as F\n",
        "import random\n",
        "\n",
        "# Initialization\n",
        "num = len(ys)\n",
        "g = torch.Generator().manual_seed(2147483647)\n",
        "W = torch.randn((80, 80), generator=g, requires_grad=True)\n",
        "out = set()\n",
        "finish = False\n",
        "\n",
        "# Function to print generated sequence\n",
        "def print_sequence(seq):\n",
        "    for i in range(4):\n",
        "        for j in range(20):\n",
        "            print(seq[i * 20 + j], end='\\t')\n",
        "        print()\n",
        "    print()\n",
        "\n",
        "# Function to calculate loss and perform backward pass\n",
        "def compute_loss(xenc, ys, W, conflict_pairs_len=1):\n",
        "    logits = xenc @ W\n",
        "    counts = logits.exp()\n",
        "    probs = counts / counts.sum(1, keepdims=True)\n",
        "    loss = -probs[torch.arange(ys.nelement()), ys].log().mean() + 0.01 * conflict_pairs_len * (W**2).mean()\n",
        "    print(f\"Iteration {k + 1}: Loss = {loss.item():.4f}\")\n",
        "    W.grad = None\n",
        "    loss.backward()\n",
        "    return probs\n",
        "\n",
        "# Function to update new_xs and new_ys based on the current list_out\n",
        "def update_new_xs_ys(xs, ys, list_out):\n",
        "    new_xs, new_ys = [], []\n",
        "    for i in range(len(xs) - 1):\n",
        "        group1, group2 = xs[i], ys[i]\n",
        "        for j in range(len(list_out) - 1):\n",
        "            if group1 == int(list_out[j]) and group2 == int(list_out[j + 1]):\n",
        "                new_xs.append(group1.item())\n",
        "                new_ys.append(group2.item())\n",
        "                break\n",
        "    return torch.tensor(new_xs, dtype=torch.long), torch.tensor(new_ys, dtype=torch.long)\n",
        "\n",
        "# Function to swap conflicting pairs\n",
        "def swap_conflicts(list_out, conflict_pairs, conflict_pairs_index):\n",
        "    swap_conflict_pairs_index = conflict_pairs_index.copy()\n",
        "    random.shuffle(swap_conflict_pairs_index)\n",
        "    if len(conflict_pairs) == 0:\n",
        "      return True\n",
        "    if len(conflict_pairs) == 2:\n",
        "        swap = 0\n",
        "        while True:\n",
        "            person_last_first_group = groups_number[conflict_pairs[0]-1][4]\n",
        "            person_first_second_group = groups_number[conflict_pairs[1]-1][0]\n",
        "            if int(incompatible_pairs[person_last_first_group, person_first_second_group]) == 1:\n",
        "                print('original_group:', groups_number[conflict_pairs[0]-1])\n",
        "                random.shuffle(groups_number[conflict_pairs[0]-1])\n",
        "                print('shuffled_group:', groups_number[conflict_pairs[0]-1])\n",
        "                swap += 1\n",
        "            else:\n",
        "                return swap == 0\n",
        "    for i in range(len(conflict_pairs_index)):\n",
        "        index_group = conflict_pairs_index[i]\n",
        "        new_index_group = swap_conflict_pairs_index[i]\n",
        "        list_out[index_group], list_out[new_index_group] = list_out[new_index_group], list_out[index_group]\n",
        "    return False\n",
        "\n",
        "# Main loop\n",
        "for k in range(10):\n",
        "    if k == 0:\n",
        "        xenc = F.one_hot(torch.tensor(xs), num_classes=80).float()\n",
        "        probs = compute_loss(xenc, ys, W)\n",
        "        while len(out) < 80:\n",
        "            ix = torch.multinomial(probs[0], num_samples=1, replacement=True, generator=g).item()\n",
        "            out.add(str(ix + 1))\n",
        "        list_out = list(out).copy()\n",
        "        print(\"Generated Sequence:\")\n",
        "        print_sequence(list_out)\n",
        "    else:\n",
        "        xenc = F.one_hot(torch.tensor(new_xs), num_classes=80).float()\n",
        "        probs = compute_loss(xenc, new_ys, W, len(conflict_pairs))\n",
        "\n",
        "    # Identify conflicts\n",
        "    conflict_pairs, conflict_pairs_index = [], []\n",
        "    for i in range(len(list_out) - 1):\n",
        "        group1, group2 = int(list_out[i]), int(list_out[i + 1])\n",
        "        if group_conflicts[group1-1, group2-1].item() > 0:\n",
        "            conflict_pairs.extend([group1, group2])\n",
        "            conflict_pairs_index.extend([i, i + 1])\n",
        "    print('conflict_pairs:', conflict_pairs)\n",
        "\n",
        "    if swap_conflicts(list_out, conflict_pairs, conflict_pairs_index):\n",
        "        finish = True\n",
        "\n",
        "    print(\"Generated Sequence:\")\n",
        "    print_sequence(list_out)\n",
        "\n",
        "    if finish:\n",
        "        print(\"=\" * 50)\n",
        "        print(\"Optimal seating arrangement:\")\n",
        "        print(\"Generated Sequence:\")\n",
        "        print_sequence(list_out)\n",
        "        for i in range(4):\n",
        "            for j in range(20):\n",
        "                print(groups_name[i * 20 + j], end='\\n')\n",
        "        print()\n",
        "        break\n",
        "\n",
        "    new_xs, new_ys = update_new_xs_ys(xs, ys, list_out)\n",
        "\n",
        "    if k != 0:\n",
        "        with torch.no_grad():\n",
        "            W.data += -50 * W.grad\n",
        "\n",
        "    print(\"=\" * 50)  # Separator for each iteration\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ROYw1BUuqEAm",
        "outputId": "5ffee8d4-2d84-4551-a79a-d20de49384b6"
      },
      "execution_count": 83,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "<ipython-input-83-a9b1ad7a97c2>:70: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).\n",
            "  xenc = F.one_hot(torch.tensor(xs), num_classes=80).float()\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Iteration 1: Loss = 4.8884\n",
            "Generated Sequence:\n",
            "35\t69\t28\t72\t64\t56\t55\t63\t12\t53\t31\t65\t38\t57\t40\t20\t68\t21\t14\t36\t\n",
            "29\t42\t4\t58\t41\t3\t50\t19\t71\t54\t1\t62\t45\t47\t30\t48\t59\t33\t26\t6\t\n",
            "9\t17\t60\t79\t16\t11\t7\t8\t74\t39\t43\t15\t77\t5\t76\t75\t73\t10\t61\t32\t\n",
            "34\t66\t37\t22\t18\t46\t24\t2\t70\t51\t52\t44\t78\t25\t27\t80\t49\t67\t23\t13\t\n",
            "\n",
            "conflict_pairs: [68, 21, 58, 41, 41, 3, 47, 30, 37, 22]\n",
            "Generated Sequence:\n",
            "35\t69\t28\t72\t64\t56\t55\t63\t12\t53\t31\t65\t38\t57\t40\t20\t58\t47\t14\t36\t\n",
            "29\t42\t4\t21\t68\t30\t50\t19\t71\t54\t1\t62\t45\t3\t22\t48\t59\t33\t26\t6\t\n",
            "9\t17\t60\t79\t16\t11\t7\t8\t74\t39\t43\t15\t77\t5\t76\t75\t73\t10\t61\t32\t\n",
            "34\t66\t41\t37\t18\t46\t24\t2\t70\t51\t52\t44\t78\t25\t27\t80\t49\t67\t23\t13\t\n",
            "\n",
            "==================================================\n",
            "Iteration 2: Loss = 4.7775\n",
            "conflict_pairs: [4, 21, 3, 22, 66, 41]\n",
            "Generated Sequence:\n",
            "35\t69\t28\t72\t64\t56\t55\t63\t12\t53\t31\t65\t38\t57\t40\t20\t58\t47\t14\t36\t\n",
            "29\t42\t4\t21\t68\t30\t50\t19\t71\t54\t1\t62\t45\t22\t3\t48\t59\t33\t26\t6\t\n",
            "9\t17\t60\t79\t16\t11\t7\t8\t74\t39\t43\t15\t77\t5\t76\t75\t73\t10\t61\t32\t\n",
            "34\t41\t66\t37\t18\t46\t24\t2\t70\t51\t52\t44\t78\t25\t27\t80\t49\t67\t23\t13\t\n",
            "\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "<ipython-input-83-a9b1ad7a97c2>:79: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).\n",
            "  xenc = F.one_hot(torch.tensor(new_xs), num_classes=80).float()\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "==================================================\n",
            "Iteration 3: Loss = 4.0851\n",
            "conflict_pairs: [4, 21, 45, 22]\n",
            "Generated Sequence:\n",
            "35\t69\t28\t72\t64\t56\t55\t63\t12\t53\t31\t65\t38\t57\t40\t20\t58\t47\t14\t36\t\n",
            "29\t42\t4\t21\t68\t30\t50\t19\t71\t54\t1\t62\t45\t22\t3\t48\t59\t33\t26\t6\t\n",
            "9\t17\t60\t79\t16\t11\t7\t8\t74\t39\t43\t15\t77\t5\t76\t75\t73\t10\t61\t32\t\n",
            "34\t41\t66\t37\t18\t46\t24\t2\t70\t51\t52\t44\t78\t25\t27\t80\t49\t67\t23\t13\t\n",
            "\n",
            "==================================================\n",
            "Iteration 4: Loss = 3.3961\n",
            "conflict_pairs: [4, 21, 45, 22]\n",
            "Generated Sequence:\n",
            "35\t69\t28\t72\t64\t56\t55\t63\t12\t53\t31\t65\t38\t57\t40\t20\t58\t47\t14\t36\t\n",
            "29\t42\t21\t4\t68\t30\t50\t19\t71\t54\t1\t62\t22\t45\t3\t48\t59\t33\t26\t6\t\n",
            "9\t17\t60\t79\t16\t11\t7\t8\t74\t39\t43\t15\t77\t5\t76\t75\t73\t10\t61\t32\t\n",
            "34\t41\t66\t37\t18\t46\t24\t2\t70\t51\t52\t44\t78\t25\t27\t80\t49\t67\t23\t13\t\n",
            "\n",
            "==================================================\n",
            "Iteration 5: Loss = 2.9525\n",
            "conflict_pairs: [21, 4]\n",
            "Generated Sequence:\n",
            "35\t69\t28\t72\t64\t56\t55\t63\t12\t53\t31\t65\t38\t57\t40\t20\t58\t47\t14\t36\t\n",
            "29\t42\t21\t4\t68\t30\t50\t19\t71\t54\t1\t62\t22\t45\t3\t48\t59\t33\t26\t6\t\n",
            "9\t17\t60\t79\t16\t11\t7\t8\t74\t39\t43\t15\t77\t5\t76\t75\t73\t10\t61\t32\t\n",
            "34\t41\t66\t37\t18\t46\t24\t2\t70\t51\t52\t44\t78\t25\t27\t80\t49\t67\t23\t13\t\n",
            "\n",
            "==================================================\n",
            "Optimal seating arrangement:\n",
            "Generated Sequence:\n",
            "35\t69\t28\t72\t64\t56\t55\t63\t12\t53\t31\t65\t38\t57\t40\t20\t58\t47\t14\t36\t\n",
            "29\t42\t21\t4\t68\t30\t50\t19\t71\t54\t1\t62\t22\t45\t3\t48\t59\t33\t26\t6\t\n",
            "9\t17\t60\t79\t16\t11\t7\t8\t74\t39\t43\t15\t77\t5\t76\t75\t73\t10\t61\t32\t\n",
            "34\t41\t66\t37\t18\t46\t24\t2\t70\t51\t52\t44\t78\t25\t27\t80\t49\t67\t23\t13\t\n",
            "\n",
            "['Sogol Hashemi', 'Samin Hashemi', 'Ahmad Niknam', 'Samin Safari', 'Mohammad Ahmadi']\n",
            "['Ahmad Karimi', 'Narges Rajabi', 'Hakan Fazeli', 'Hakan Ahmadzadeh', 'Hossein Rasoli']\n",
            "['Ali Rostami', 'Ali Niknam', 'Sara Mehrzad', 'Ahmad Mehrzad', 'Samin Moradi']\n",
            "['Sogol Rostami', 'sahar Lotfi', 'Samin Ebrahimi', 'Mohammad Ebrahimi', 'Ahmad Ebrahimi']\n",
            "['Mohammad Mohammadi', 'Zahra Fazeli', 'Hakan Rahimi', 'Sogol Rasoli', 'Maryam Ahmadzadeh']\n",
            "['Ahmad Hosseinzadeh', 'Hakan Hashemi', 'Maryam Rajabi', 'Zahra Safari', 'Raha Sadeghi']\n",
            "['Hossein Hosseini', 'Fatemeh Ebrahimi', 'Roya Hosseini', 'Sogol Sadeghi', 'Zahra Rasoli']\n",
            "['Ahmad Safari', 'Ali Rasoli', 'Roze Safavi', 'Sogol Fazeli', 'Roze Hashemi']\n",
            "['sahar Mehrzad', 'Reza Mohammadi', 'Maryam Lotfi', 'Reza Rahimi', 'Narges Safari']\n",
            "['Narges Ebrahimi', 'Sara Safavi', 'Mohammad Lotfi', 'Roze Niknam', 'Samin Rajabi']\n",
            "['Roya Safavi', 'Bahram Rasoli', 'Sogol Rahimi', 'sahar Rostami', 'Narges Rasoli']\n",
            "['Narges Ahmadzadeh', 'Sogol Safavi', 'Mohammad Moradi', 'Reza Ahmadi', 'Hossein Safavi']\n",
            "['Roya Ahmadi', 'Roya Ahmadzadeh', 'Ali Safari', 'Narges Fazeli', 'Reza Fazeli']\n",
            "['Zahra Rajabi', 'Raha Rasoli', 'sahar Rasoli', 'Reza Sadeghi', 'Roze Rahimi']\n",
            "['Samin Mehrzad', 'Sogol Hosseinzadeh', 'Roze Ahmadi', 'Roze Safari', 'Mehdi Niknam']\n",
            "['Mehdi Mohammadi', 'Roze Rostami', 'Roya Ebrahimi', 'Reza Ebrahimi', 'Narges Moradi']\n",
            "['Maryam Hosseini', 'Ali Mehrzad', 'Reza Hosseini', 'Bahram Mohammadi', 'Bahram Ahmadzadeh']\n",
            "['Mehdi Lotfi', 'Reza Niknam', 'Mehdi Karimi', 'Mehdi Ebrahimi', 'Ahmad Rostami']\n",
            "['Fatemeh Lotfi', 'Raha Niknam', 'Roze Mehrzad', 'Fatemeh Moradi', 'Narges Ahmadi']\n",
            "['Maryam Mohammadi', 'Sara Lotfi', 'Roze Ahmadzadeh', 'Ali Moradi', 'Mehdi Safari']\n",
            "['Mehdi Moradi', 'Ali Hashemi', 'Roya Lotfi', 'Fatemeh Rasoli', 'Bahram Rahimi']\n",
            "['Bahram Ebrahimi', 'Bahram Lotfi', 'Samin Hosseinzadeh', 'Sara Fazeli', 'Hamed Moradi']\n",
            "['Hossein Sadeghi', 'Zahra Hosseini', 'sahar Rajabi', 'Hamed Rasoli', 'Bahram Rajabi']\n",
            "['Maryam Rahimi', 'Zahra Ahmadzadeh', 'Raha Karimi', 'Samin Safavi', 'Zahra Rahimi']\n",
            "['Hamed Safavi', 'Mohammad Rostami', 'Mehdi Rajabi', 'Zahra Sadeghi', 'sahar Safari']\n",
            "['Hamed Rahimi', 'Sogol Karimi', 'Sogol Mohammadi', 'Roze Rajabi', 'Fatemeh Safari']\n",
            "['Roya Mohammadi', 'Narges Safavi', 'Raha Mohammadi', 'Mohammad Mehrzad', 'Fatemeh Hashemi']\n",
            "['Hamed Ebrahimi', 'Narges Karimi', 'Narges Lotfi', 'Fatemeh Ahmadi', 'Mehdi Ahmadi']\n",
            "['Mohammad Safavi', 'Mohammad Sadeghi', 'Reza Ahmadzadeh', 'Raha Safavi', 'Maryam Hashemi']\n",
            "['sahar Hosseinzadeh', 'Roya Safari', 'Sara Ahmadi', 'Hossein Moradi', 'Raha Rajabi']\n",
            "['Maryam Rasoli', 'sahar Ahmadi', 'Sara Ahmadzadeh', 'Mehdi Hosseini', 'Ahmad Ahmadi']\n",
            "['sahar Ahmadzadeh', 'Maryam Ahmadi', 'Ali Mohammadi', 'Fatemeh Safavi', 'Samin Niknam']\n",
            "['Reza Hashemi', 'Roze Mohammadi', 'Hakan Karimi', 'Sogol Rajabi', 'Mehdi Fazeli']\n",
            "['Reza Karimi', 'sahar Fazeli', 'Hossein Ahmadzadeh', 'Hakan Rajabi', 'Roze Lotfi']\n",
            "['Fatemeh Mohammadi', 'Raha Mehrzad', 'Zahra Mehrzad', 'Narges Niknam', 'Hakan Hosseini']\n",
            "['Mohammad Hashemi', 'Zahra Safavi', 'Raha Ahmadzadeh', 'Raha Hashemi', 'Roze Hosseini']\n",
            "['Maryam Sadeghi', 'Hossein Mehrzad', 'Zahra Hashemi', 'Mohammad Ahmadzadeh', 'sahar Karimi']\n",
            "['Roya Sadeghi', 'Samin Fazeli', 'Fatemeh Rostami', 'Raha Hosseini', 'Sara Hosseini']\n",
            "['Mohammad Fazeli', 'Raha Rahimi', 'Hakan Niknam', 'Ali Karimi', 'Ali Safavi']\n",
            "['Reza Rajabi', 'Zahra Ahmadi', 'Hamed Safari', 'Ahmad Hashemi', 'Samin Lotfi']\n",
            "['Hossein Rostami', 'Raha Rostami', 'Hossein Rahimi', 'Sogol Safari', 'Hossein Ebrahimi']\n",
            "['Ali Hosseini', 'Maryam Moradi', 'Reza Safavi', 'Ahmad Rahimi', 'Roya Moradi']\n",
            "['Samin Ahmadzadeh', 'Fatemeh Ahmadzadeh', 'Sara Karimi', 'Ali Lotfi', 'Zahra Hosseinzadeh']\n",
            "['Hossein Fazeli', 'Sara Niknam', 'Ali Ebrahimi', 'Bahram Sadeghi', 'Hamed Fazeli']\n",
            "['Reza Mehrzad', 'Hakan Ebrahimi', 'Hakan Sadeghi', 'Fatemeh Karimi', 'Samin Mohammadi']\n",
            "['Roya Rostami', 'Roze Moradi', 'Hakan Rostami', 'Roze Rasoli', 'Maryam Fazeli']\n",
            "['Roya Hashemi', 'Fatemeh Rahimi', 'Roya Karimi', 'Ahmad Hosseini', 'Hakan Safari']\n",
            "['Ahmad Fazeli', 'Sara Rostami', 'Hakan Mohammadi', 'Sara Rajabi', 'Raha Moradi']\n",
            "['Maryam Hosseinzadeh', 'Narges Rostami', 'Reza Lotfi', 'Bahram Niknam', 'Zahra Ebrahimi']\n",
            "['Raha Safari', 'Hamed Lotfi', 'Roya Niknam', 'Ali Fazeli', 'Fatemeh Hosseini']\n",
            "['Bahram Hosseini', 'Roze Sadeghi', 'Ali Ahmadi', 'Ali Ahmadzadeh', 'Mohammad Rajabi']\n",
            "['Sara Rasoli', 'Ali Sadeghi', 'Maryam Mehrzad', 'Ahmad Moradi', 'Sara Hashemi']\n",
            "['Hossein Mohammadi', 'Roya Mehrzad', 'Maryam Ebrahimi', 'Maryam Safari', 'Mohammad Safari']\n",
            "['Maryam Rostami', 'Hakan Ahmadi', 'sahar Niknam', 'Sogol Ahmadi', 'Roze Karimi']\n",
            "['Sogol Lotfi', 'Hossein Hashemi', 'Roya Rajabi', 'Maryam Karimi', 'Raha Fazeli']\n",
            "['Roya Rasoli', 'Samin Rasoli', 'Sara Sadeghi', 'sahar Ebrahimi', 'Hamed Niknam']\n",
            "['Zahra Rostami', 'Mehdi Safavi', 'sahar Moradi', 'Fatemeh Mehrzad', 'Roze Ebrahimi']\n",
            "['Ahmad Mohammadi', 'Mohammad Rasoli', 'Mohammad Rahimi', 'Narges Hashemi', 'Sara Rahimi']\n",
            "['Hamed Hosseini', 'Hamed Hosseinzadeh', 'Samin Karimi', 'Hamed Ahmadi', 'Hossein Rajabi']\n",
            "['Sara Safari', 'Sara Hosseinzadeh', 'Ali Rahimi', 'Raha Lotfi', 'Hossein Safari']\n",
            "['Sogol Hosseini', 'Mehdi Mehrzad', 'Hossein Hosseinzadeh', 'Mehdi Sadeghi', 'Samin Hosseini']\n",
            "['Roya Hosseinzadeh', 'Roya Fazeli', 'Hakan Mehrzad', 'Fatemeh Rajabi', 'Mehdi Rostami']\n",
            "['Hakan Moradi', 'Hossein Lotfi', 'Zahra Mohammadi', 'Zahra Lotfi', 'Hakan Hosseinzadeh']\n",
            "['Mohammad Hosseini', 'sahar Mohammadi', 'Reza Moradi', 'Narges Hosseini', 'Sogol Ahmadzadeh']\n",
            "['Hamed Ahmadzadeh', 'Zahra Moradi', 'Mohammad Hosseinzadeh', 'Bahram Karimi', 'Zahra Karimi']\n",
            "['Ahmad Ahmadzadeh', 'Mehdi Rahimi', 'Bahram Rostami', 'Mehdi Ahmadzadeh', 'Hossein Niknam']\n",
            "['sahar Hashemi', 'Ali Rajabi', 'Samin Sadeghi', 'Bahram Hashemi', 'Hamed Karimi']\n",
            "['Sara Mohammadi', 'Narges Rahimi', 'Mohammad Niknam', 'Sara Moradi', 'Fatemeh Sadeghi']\n",
            "['Mehdi Rasoli', 'Roze Hosseinzadeh', 'Bahram Safari', 'Hossein Ahmadi', 'Ahmad Rajabi']\n",
            "['Fatemeh Fazeli', 'Fatemeh Hosseinzadeh', 'Reza Rostami', 'Narges Sadeghi', 'Mohammad Karimi']\n",
            "['sahar Safavi', 'Hamed Mohammadi', 'Ali Hosseinzadeh', 'Raha Hosseinzadeh', 'Zahra Niknam']\n",
            "['Sogol Niknam', 'Samin Rahimi', 'Reza Safari', 'Hakan Rasoli', 'Raha Ahmadi']\n",
            "['Ahmad Rasoli', 'Bahram Moradi', 'sahar Sadeghi', 'Hamed Sadeghi', 'Narges Hosseinzadeh']\n",
            "['Samin Rostami', 'Samin Ahmadi', 'Bahram Safavi', 'Bahram Mehrzad', 'Narges Mohammadi']\n",
            "['Ahmad Lotfi', 'Narges Mehrzad', 'Raha Ebrahimi', 'Hamed Rajabi', 'Ahmad Sadeghi']\n",
            "['Roya Rahimi', 'Hamed Mehrzad', 'Bahram Fazeli', 'Hakan Safavi', 'Roze Fazeli']\n",
            "['Fatemeh Niknam', 'Bahram Ahmadi', 'Mehdi Hosseinzadeh', 'Bahram Hosseinzadeh', 'sahar Hosseini']\n",
            "['Sogol Ebrahimi', 'Reza Hosseinzadeh', 'Maryam Niknam', 'Hossein Karimi', 'Hamed Rostami']\n",
            "['Hamed Hashemi', 'Mehdi Hashemi', 'Ahmad Safavi', 'Sogol Moradi', 'Sara Ebrahimi']\n",
            "['Sogol Mehrzad', 'Reza Rasoli', 'sahar Rahimi', 'Maryam Safavi', 'Hakan Lotfi']\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import torch\n",
        "import torch.nn.functional as F\n",
        "import random\n",
        "\n",
        "# Initialization\n",
        "num = len(ys)\n",
        "g = torch.Generator().manual_seed(2147483647)\n",
        "W = torch.randn((80, 80), generator=g, requires_grad=True)\n",
        "out = set()\n",
        "finish = False\n",
        "\n",
        "# Function to print generated sequence\n",
        "def print_sequence(seq):\n",
        "    for i in range(4):\n",
        "        for j in range(20):\n",
        "            print(seq[i * 20 + j], end='\\t')\n",
        "        print()\n",
        "    print()\n",
        "\n",
        "# Function to calculate loss and perform backward pass\n",
        "def compute_loss(xenc, ys, W, conflict_pairs_len=1):\n",
        "    logits = xenc @ W\n",
        "    counts = logits.exp()\n",
        "    probs = counts / counts.sum(1, keepdims=True)\n",
        "    loss = -probs[torch.arange(ys.nelement()), ys].log().mean() + 0.01 * conflict_pairs_len * (W**2).mean()\n",
        "    print(f\"Iteration {k + 1}: Loss = {loss.item():.4f}\")\n",
        "    W.grad = None\n",
        "    loss.backward()\n",
        "    return probs\n",
        "\n",
        "# Function to update new_xs and new_ys based on the current list_out\n",
        "def update_new_xs_ys(xs, ys, list_out):\n",
        "    new_xs, new_ys = [], []\n",
        "    for i in range(len(xs) - 1):\n",
        "        group1, group2 = xs[i], ys[i]\n",
        "        for j in range(len(list_out) - 1):\n",
        "            if group1 == int(list_out[j]) and group2 == int(list_out[j + 1]):\n",
        "                new_xs.append(group1.item())\n",
        "                new_ys.append(group2.item())\n",
        "                break\n",
        "    return torch.tensor(new_xs, dtype=torch.long), torch.tensor(new_ys, dtype=torch.long)\n",
        "\n",
        "# Function to swap conflicting pairs\n",
        "def swap_conflicts(list_out, conflict_pairs, conflict_pairs_index):\n",
        "    swap_conflict_pairs_index = conflict_pairs_index.copy()\n",
        "    random.shuffle(swap_conflict_pairs_index)\n",
        "    if len(conflict_pairs) == 0:\n",
        "      return True\n",
        "    if len(conflict_pairs) == 2:\n",
        "        swap = 0\n",
        "        while True:\n",
        "            person_last_first_group = groups_number[conflict_pairs[0]-1][4]\n",
        "            person_first_second_group = groups_number[conflict_pairs[1]-1][0]\n",
        "            if int(incompatible_pairs[person_last_first_group, person_first_second_group]) == 1:\n",
        "                print('original_group:', groups_number[conflict_pairs[0]-1])\n",
        "                random.shuffle(groups_number[conflict_pairs[0]-1])\n",
        "                print('shuffled_group:', groups_number[conflict_pairs[0]-1])\n",
        "                swap += 1\n",
        "            else:\n",
        "                return swap == 0\n",
        "    for i in range(len(conflict_pairs_index)):\n",
        "        index_group = conflict_pairs_index[i]\n",
        "        new_index_group = swap_conflict_pairs_index[i]\n",
        "        list_out[index_group], list_out[new_index_group] = list_out[new_index_group], list_out[index_group]\n",
        "    return False\n",
        "\n",
        "# Main loop\n",
        "for k in range(10):\n",
        "    if k == 0:\n",
        "        xenc = F.one_hot(torch.tensor(xs), num_classes=80).float()\n",
        "        probs = compute_loss(xenc, ys, W)\n",
        "        while len(out) < 80:\n",
        "            ix = torch.multinomial(probs[0], num_samples=1, replacement=True, generator=g).item()\n",
        "            out.add(str(ix + 1))\n",
        "        list_out = list(out).copy()\n",
        "        print(\"Generated Sequence:\")\n",
        "        print_sequence(list_out)\n",
        "    else:\n",
        "        xenc = F.one_hot(torch.tensor(new_xs), num_classes=80).float()\n",
        "        probs = compute_loss(xenc, new_ys, W, len(conflict_pairs))\n",
        "\n",
        "    # Identify conflicts\n",
        "    conflict_pairs, conflict_pairs_index = [], []\n",
        "    for i in range(len(list_out) - 1):\n",
        "        group1, group2 = int(list_out[i]), int(list_out[i + 1])\n",
        "        if group_conflicts[group1-1, group2-1].item() > 0:\n",
        "            conflict_pairs.extend([group1, group2])\n",
        "            conflict_pairs_index.extend([i, i + 1])\n",
        "    print('conflict_pairs:', conflict_pairs)\n",
        "\n",
        "    if swap_conflicts(list_out, conflict_pairs, conflict_pairs_index):\n",
        "        finish = True\n",
        "\n",
        "    print(\"Generated Sequence:\")\n",
        "    print_sequence(list_out)\n",
        "\n",
        "    if finish:\n",
        "        print(\"=\" * 50)\n",
        "        print(\"Optimal seating arrangement:\")\n",
        "        print(\"Generated Sequence:\")\n",
        "        print_sequence(list_out)\n",
        "        for i in range(4):\n",
        "            for j in range(20):\n",
        "                print(groups_name[i * 20 + j], end='\\n')\n",
        "        print()\n",
        "        break\n",
        "\n",
        "    new_xs, new_ys = update_new_xs_ys(xs, ys, list_out)\n",
        "\n",
        "    if k != 0:\n",
        "        with torch.no_grad():\n",
        "            W.data += -50 * W.grad\n",
        "\n",
        "    print(\"=\" * 50)  # Separator for each iteration\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "gFm5bnZ3s4il",
        "outputId": "2a748144-20a8-48cb-f090-a6dfd42b3e92"
      },
      "execution_count": 91,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "<ipython-input-91-a9b1ad7a97c2>:70: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).\n",
            "  xenc = F.one_hot(torch.tensor(xs), num_classes=80).float()\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Iteration 1: Loss = 4.8884\n",
            "Generated Sequence:\n",
            "35\t69\t28\t72\t64\t56\t55\t63\t12\t53\t31\t65\t38\t57\t40\t20\t68\t21\t14\t36\t\n",
            "29\t42\t4\t58\t41\t3\t50\t19\t71\t54\t1\t62\t45\t47\t30\t48\t59\t33\t26\t6\t\n",
            "9\t17\t60\t79\t16\t11\t7\t8\t74\t39\t43\t15\t77\t5\t76\t75\t73\t10\t61\t32\t\n",
            "34\t66\t37\t22\t18\t46\t24\t2\t70\t51\t52\t44\t78\t25\t27\t80\t49\t67\t23\t13\t\n",
            "\n",
            "conflict_pairs: [68, 21, 58, 41, 41, 3, 47, 30, 37, 22]\n",
            "Generated Sequence:\n",
            "35\t69\t28\t72\t64\t56\t55\t63\t12\t53\t31\t65\t38\t57\t40\t20\t41\t58\t14\t36\t\n",
            "29\t42\t4\t21\t68\t3\t50\t19\t71\t54\t1\t62\t45\t22\t30\t48\t59\t33\t26\t6\t\n",
            "9\t17\t60\t79\t16\t11\t7\t8\t74\t39\t43\t15\t77\t5\t76\t75\t73\t10\t61\t32\t\n",
            "34\t66\t47\t37\t18\t46\t24\t2\t70\t51\t52\t44\t78\t25\t27\t80\t49\t67\t23\t13\t\n",
            "\n",
            "==================================================\n",
            "Iteration 2: Loss = 4.8346\n",
            "conflict_pairs: [4, 21, 45, 22]\n",
            "Generated Sequence:\n",
            "35\t69\t28\t72\t64\t56\t55\t63\t12\t53\t31\t65\t38\t57\t40\t20\t41\t58\t14\t36\t\n",
            "29\t42\t4\t21\t68\t3\t50\t19\t71\t54\t1\t62\t22\t45\t30\t48\t59\t33\t26\t6\t\n",
            "9\t17\t60\t79\t16\t11\t7\t8\t74\t39\t43\t15\t77\t5\t76\t75\t73\t10\t61\t32\t\n",
            "34\t66\t47\t37\t18\t46\t24\t2\t70\t51\t52\t44\t78\t25\t27\t80\t49\t67\t23\t13\t\n",
            "\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "<ipython-input-91-a9b1ad7a97c2>:79: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).\n",
            "  xenc = F.one_hot(torch.tensor(new_xs), num_classes=80).float()\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "==================================================\n",
            "Iteration 3: Loss = 4.0607\n",
            "conflict_pairs: [4, 21]\n",
            "Generated Sequence:\n",
            "35\t69\t28\t72\t64\t56\t55\t63\t12\t53\t31\t65\t38\t57\t40\t20\t41\t58\t14\t36\t\n",
            "29\t42\t4\t21\t68\t3\t50\t19\t71\t54\t1\t62\t22\t45\t30\t48\t59\t33\t26\t6\t\n",
            "9\t17\t60\t79\t16\t11\t7\t8\t74\t39\t43\t15\t77\t5\t76\t75\t73\t10\t61\t32\t\n",
            "34\t66\t47\t37\t18\t46\t24\t2\t70\t51\t52\t44\t78\t25\t27\t80\t49\t67\t23\t13\t\n",
            "\n",
            "==================================================\n",
            "Optimal seating arrangement:\n",
            "Generated Sequence:\n",
            "35\t69\t28\t72\t64\t56\t55\t63\t12\t53\t31\t65\t38\t57\t40\t20\t41\t58\t14\t36\t\n",
            "29\t42\t4\t21\t68\t3\t50\t19\t71\t54\t1\t62\t22\t45\t30\t48\t59\t33\t26\t6\t\n",
            "9\t17\t60\t79\t16\t11\t7\t8\t74\t39\t43\t15\t77\t5\t76\t75\t73\t10\t61\t32\t\n",
            "34\t66\t47\t37\t18\t46\t24\t2\t70\t51\t52\t44\t78\t25\t27\t80\t49\t67\t23\t13\t\n",
            "\n",
            "['Sogol Hashemi', 'Samin Hashemi', 'Ahmad Niknam', 'Samin Safari', 'Mohammad Ahmadi']\n",
            "['Ahmad Karimi', 'Narges Rajabi', 'Hakan Fazeli', 'Hakan Ahmadzadeh', 'Hossein Rasoli']\n",
            "['Ali Rostami', 'Ali Niknam', 'Sara Mehrzad', 'Ahmad Mehrzad', 'Samin Moradi']\n",
            "['Sogol Rostami', 'sahar Lotfi', 'Samin Ebrahimi', 'Mohammad Ebrahimi', 'Ahmad Ebrahimi']\n",
            "['Mohammad Mohammadi', 'Zahra Fazeli', 'Hakan Rahimi', 'Sogol Rasoli', 'Maryam Ahmadzadeh']\n",
            "['Ahmad Hosseinzadeh', 'Hakan Hashemi', 'Maryam Rajabi', 'Zahra Safari', 'Raha Sadeghi']\n",
            "['Hossein Hosseini', 'Fatemeh Ebrahimi', 'Roya Hosseini', 'Sogol Sadeghi', 'Zahra Rasoli']\n",
            "['Ahmad Safari', 'Ali Rasoli', 'Roze Safavi', 'Sogol Fazeli', 'Roze Hashemi']\n",
            "['sahar Mehrzad', 'Reza Mohammadi', 'Maryam Lotfi', 'Reza Rahimi', 'Narges Safari']\n",
            "['Narges Ebrahimi', 'Sara Safavi', 'Mohammad Lotfi', 'Roze Niknam', 'Samin Rajabi']\n",
            "['Roya Safavi', 'Bahram Rasoli', 'Sogol Rahimi', 'sahar Rostami', 'Narges Rasoli']\n",
            "['Narges Ahmadzadeh', 'Sogol Safavi', 'Mohammad Moradi', 'Reza Ahmadi', 'Hossein Safavi']\n",
            "['Roya Ahmadi', 'Roya Ahmadzadeh', 'Ali Safari', 'Narges Fazeli', 'Reza Fazeli']\n",
            "['Zahra Rajabi', 'Raha Rasoli', 'sahar Rasoli', 'Reza Sadeghi', 'Roze Rahimi']\n",
            "['Samin Mehrzad', 'Sogol Hosseinzadeh', 'Roze Ahmadi', 'Roze Safari', 'Mehdi Niknam']\n",
            "['Mehdi Mohammadi', 'Roze Rostami', 'Roya Ebrahimi', 'Reza Ebrahimi', 'Narges Moradi']\n",
            "['Maryam Hosseini', 'Ali Mehrzad', 'Reza Hosseini', 'Bahram Mohammadi', 'Bahram Ahmadzadeh']\n",
            "['Mehdi Lotfi', 'Reza Niknam', 'Mehdi Karimi', 'Mehdi Ebrahimi', 'Ahmad Rostami']\n",
            "['Fatemeh Lotfi', 'Raha Niknam', 'Roze Mehrzad', 'Fatemeh Moradi', 'Narges Ahmadi']\n",
            "['Maryam Mohammadi', 'Sara Lotfi', 'Roze Ahmadzadeh', 'Ali Moradi', 'Mehdi Safari']\n",
            "['Mehdi Moradi', 'Ali Hashemi', 'Roya Lotfi', 'Fatemeh Rasoli', 'Bahram Rahimi']\n",
            "['Bahram Ebrahimi', 'Bahram Lotfi', 'Samin Hosseinzadeh', 'Sara Fazeli', 'Hamed Moradi']\n",
            "['Hossein Sadeghi', 'Zahra Hosseini', 'sahar Rajabi', 'Hamed Rasoli', 'Bahram Rajabi']\n",
            "['Maryam Rahimi', 'Zahra Ahmadzadeh', 'Raha Karimi', 'Samin Safavi', 'Zahra Rahimi']\n",
            "['Hamed Safavi', 'Mohammad Rostami', 'Mehdi Rajabi', 'Zahra Sadeghi', 'sahar Safari']\n",
            "['Hamed Rahimi', 'Sogol Karimi', 'Sogol Mohammadi', 'Roze Rajabi', 'Fatemeh Safari']\n",
            "['Roya Mohammadi', 'Narges Safavi', 'Raha Mohammadi', 'Mohammad Mehrzad', 'Fatemeh Hashemi']\n",
            "['Hamed Ebrahimi', 'Narges Karimi', 'Narges Lotfi', 'Fatemeh Ahmadi', 'Mehdi Ahmadi']\n",
            "['Mohammad Safavi', 'Mohammad Sadeghi', 'Reza Ahmadzadeh', 'Raha Safavi', 'Maryam Hashemi']\n",
            "['sahar Hosseinzadeh', 'Roya Safari', 'Sara Ahmadi', 'Hossein Moradi', 'Raha Rajabi']\n",
            "['Maryam Rasoli', 'sahar Ahmadi', 'Sara Ahmadzadeh', 'Mehdi Hosseini', 'Ahmad Ahmadi']\n",
            "['sahar Ahmadzadeh', 'Maryam Ahmadi', 'Ali Mohammadi', 'Fatemeh Safavi', 'Samin Niknam']\n",
            "['Reza Hashemi', 'Roze Mohammadi', 'Hakan Karimi', 'Sogol Rajabi', 'Mehdi Fazeli']\n",
            "['Reza Karimi', 'sahar Fazeli', 'Hossein Ahmadzadeh', 'Hakan Rajabi', 'Roze Lotfi']\n",
            "['Fatemeh Mohammadi', 'Raha Mehrzad', 'Zahra Mehrzad', 'Narges Niknam', 'Hakan Hosseini']\n",
            "['Mohammad Hashemi', 'Zahra Safavi', 'Raha Ahmadzadeh', 'Raha Hashemi', 'Roze Hosseini']\n",
            "['Maryam Sadeghi', 'Hossein Mehrzad', 'Zahra Hashemi', 'Mohammad Ahmadzadeh', 'sahar Karimi']\n",
            "['Roya Sadeghi', 'Samin Fazeli', 'Fatemeh Rostami', 'Raha Hosseini', 'Sara Hosseini']\n",
            "['Mohammad Fazeli', 'Raha Rahimi', 'Hakan Niknam', 'Ali Karimi', 'Ali Safavi']\n",
            "['Reza Rajabi', 'Zahra Ahmadi', 'Hamed Safari', 'Ahmad Hashemi', 'Samin Lotfi']\n",
            "['Hossein Rostami', 'Raha Rostami', 'Hossein Rahimi', 'Sogol Safari', 'Hossein Ebrahimi']\n",
            "['Ali Hosseini', 'Maryam Moradi', 'Reza Safavi', 'Ahmad Rahimi', 'Roya Moradi']\n",
            "['Samin Ahmadzadeh', 'Fatemeh Ahmadzadeh', 'Sara Karimi', 'Ali Lotfi', 'Zahra Hosseinzadeh']\n",
            "['Hossein Fazeli', 'Sara Niknam', 'Ali Ebrahimi', 'Bahram Sadeghi', 'Hamed Fazeli']\n",
            "['Reza Mehrzad', 'Hakan Ebrahimi', 'Hakan Sadeghi', 'Fatemeh Karimi', 'Samin Mohammadi']\n",
            "['Roya Rostami', 'Roze Moradi', 'Hakan Rostami', 'Roze Rasoli', 'Maryam Fazeli']\n",
            "['Roya Hashemi', 'Fatemeh Rahimi', 'Roya Karimi', 'Ahmad Hosseini', 'Hakan Safari']\n",
            "['Ahmad Fazeli', 'Sara Rostami', 'Hakan Mohammadi', 'Sara Rajabi', 'Raha Moradi']\n",
            "['Maryam Hosseinzadeh', 'Narges Rostami', 'Reza Lotfi', 'Bahram Niknam', 'Zahra Ebrahimi']\n",
            "['Raha Safari', 'Hamed Lotfi', 'Roya Niknam', 'Ali Fazeli', 'Fatemeh Hosseini']\n",
            "['Bahram Hosseini', 'Roze Sadeghi', 'Ali Ahmadi', 'Ali Ahmadzadeh', 'Mohammad Rajabi']\n",
            "['Sara Rasoli', 'Ali Sadeghi', 'Maryam Mehrzad', 'Ahmad Moradi', 'Sara Hashemi']\n",
            "['Hossein Mohammadi', 'Roya Mehrzad', 'Maryam Ebrahimi', 'Maryam Safari', 'Mohammad Safari']\n",
            "['Maryam Rostami', 'Hakan Ahmadi', 'sahar Niknam', 'Sogol Ahmadi', 'Roze Karimi']\n",
            "['Sogol Lotfi', 'Hossein Hashemi', 'Roya Rajabi', 'Maryam Karimi', 'Raha Fazeli']\n",
            "['Roya Rasoli', 'Samin Rasoli', 'Sara Sadeghi', 'sahar Ebrahimi', 'Hamed Niknam']\n",
            "['Zahra Rostami', 'Mehdi Safavi', 'sahar Moradi', 'Fatemeh Mehrzad', 'Roze Ebrahimi']\n",
            "['Ahmad Mohammadi', 'Mohammad Rasoli', 'Mohammad Rahimi', 'Narges Hashemi', 'Sara Rahimi']\n",
            "['Hamed Hosseini', 'Hamed Hosseinzadeh', 'Samin Karimi', 'Hamed Ahmadi', 'Hossein Rajabi']\n",
            "['Sara Safari', 'Sara Hosseinzadeh', 'Ali Rahimi', 'Raha Lotfi', 'Hossein Safari']\n",
            "['Sogol Hosseini', 'Mehdi Mehrzad', 'Hossein Hosseinzadeh', 'Mehdi Sadeghi', 'Samin Hosseini']\n",
            "['Roya Hosseinzadeh', 'Roya Fazeli', 'Hakan Mehrzad', 'Fatemeh Rajabi', 'Mehdi Rostami']\n",
            "['Hakan Moradi', 'Hossein Lotfi', 'Zahra Mohammadi', 'Zahra Lotfi', 'Hakan Hosseinzadeh']\n",
            "['Mohammad Hosseini', 'sahar Mohammadi', 'Reza Moradi', 'Narges Hosseini', 'Sogol Ahmadzadeh']\n",
            "['Hamed Ahmadzadeh', 'Zahra Moradi', 'Mohammad Hosseinzadeh', 'Bahram Karimi', 'Zahra Karimi']\n",
            "['Ahmad Ahmadzadeh', 'Mehdi Rahimi', 'Bahram Rostami', 'Mehdi Ahmadzadeh', 'Hossein Niknam']\n",
            "['sahar Hashemi', 'Ali Rajabi', 'Samin Sadeghi', 'Bahram Hashemi', 'Hamed Karimi']\n",
            "['Sara Mohammadi', 'Narges Rahimi', 'Mohammad Niknam', 'Sara Moradi', 'Fatemeh Sadeghi']\n",
            "['Mehdi Rasoli', 'Roze Hosseinzadeh', 'Bahram Safari', 'Hossein Ahmadi', 'Ahmad Rajabi']\n",
            "['Fatemeh Fazeli', 'Fatemeh Hosseinzadeh', 'Reza Rostami', 'Narges Sadeghi', 'Mohammad Karimi']\n",
            "['sahar Safavi', 'Hamed Mohammadi', 'Ali Hosseinzadeh', 'Raha Hosseinzadeh', 'Zahra Niknam']\n",
            "['Sogol Niknam', 'Samin Rahimi', 'Reza Safari', 'Hakan Rasoli', 'Raha Ahmadi']\n",
            "['Ahmad Rasoli', 'Bahram Moradi', 'sahar Sadeghi', 'Hamed Sadeghi', 'Narges Hosseinzadeh']\n",
            "['Samin Rostami', 'Samin Ahmadi', 'Bahram Safavi', 'Bahram Mehrzad', 'Narges Mohammadi']\n",
            "['Ahmad Lotfi', 'Narges Mehrzad', 'Raha Ebrahimi', 'Hamed Rajabi', 'Ahmad Sadeghi']\n",
            "['Roya Rahimi', 'Hamed Mehrzad', 'Bahram Fazeli', 'Hakan Safavi', 'Roze Fazeli']\n",
            "['Fatemeh Niknam', 'Bahram Ahmadi', 'Mehdi Hosseinzadeh', 'Bahram Hosseinzadeh', 'sahar Hosseini']\n",
            "['Sogol Ebrahimi', 'Reza Hosseinzadeh', 'Maryam Niknam', 'Hossein Karimi', 'Hamed Rostami']\n",
            "['Hamed Hashemi', 'Mehdi Hashemi', 'Ahmad Safavi', 'Sogol Moradi', 'Sara Ebrahimi']\n",
            "['Sogol Mehrzad', 'Reza Rasoli', 'sahar Rahimi', 'Maryam Safavi', 'Hakan Lotfi']\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "1n4Sg7CesfTD"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}