LotteryGame.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "source": [
        "통계 방법\n",
        "1. 회차별 보너스 포함 당첨번호를 집계해서 최근 2주간, 최근 3개월간 등 원하는 기간 내에 당첨번호가 몇개 나오는지 통계\n",
        "2. 번호별로 당첨이 몇번 됐는지 확인\n",
        "\n",
        "번호뽑기\n",
        "1. 각 번호는 당첨횟수만큼 백분율로 환산하여 가중치를 갖는다. (번호가중치)\n",
        "\n",
        "ex) 100% 중에서 0부터 0.01은 번호 1,\n",
        "    0.01부터 0.02는 번호 2를 가리킨다\n",
        "\n",
        "2. 최근 한달간, 최근 3개월간, 최근 6개월간의 당첨번호를 통계내어\n",
        "1-9, 10-19, 20-29, 30-39, 40-45 각 구간의 백분율로 환산된만큼 가중치를 갖는다. (기간가중치)\n",
        "\n",
        "ex) 100% 중에서 0부터 0.1은 1-9 구간,\n",
        "0.1부터 0.3은 10-19 구간을 가리킨다\n",
        "\n",
        "3. 번호를 뽑을 때 0과 1사이의 랜덤값에 번호뽑기 1,2번에서 나온 가중치를 반영하여 번호를 뽑고 이 과정으로 총 6개의 번호를 뽑는다.\n",
        "\n",
        "4. 이 스크립트는 입력값 n을 받고 n만큼 번호를 뽑을 수 있다.\n",
        "6개의 번호를 n번 뽑는 것이다."
      ],
      "metadata": {
        "id": "rQoe6E2x1ZOS"
      }
    },
    {
      "cell_type": "code",
      "execution_count": 21,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "mid8ztgzqZPH",
        "outputId": "653de0d8-2c00-493d-e1bc-482a102190e2"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Requirement already satisfied: requests in /usr/local/lib/python3.10/dist-packages (2.32.3)\n",
            "Requirement already satisfied: beautifulsoup4 in /usr/local/lib/python3.10/dist-packages (4.12.3)\n",
            "Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.10/dist-packages (from requests) (3.4.1)\n",
            "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.10/dist-packages (from requests) (3.10)\n",
            "Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.10/dist-packages (from requests) (2.3.0)\n",
            "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.10/dist-packages (from requests) (2024.12.14)\n",
            "Requirement already satisfied: soupsieve>1.2 in /usr/local/lib/python3.10/dist-packages (from beautifulsoup4) (2.6)\n"
          ]
        }
      ],
      "source": [
        "pip install requests beautifulsoup4"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# 최근 회차까지의 데이터셋 저장"
      ],
      "metadata": {
        "id": "OBpXmxvt2AZ1"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import requests\n",
        "from bs4 import BeautifulSoup\n",
        "import re\n",
        "import pandas as pd\n",
        "import math\n",
        "import random\n",
        "from collections import defaultdict\n",
        "\n",
        "last_round = 1153\n",
        "url = \"https://dhlottery.co.kr/gameResult.do?method=allWinPrint&gubun=byWin&nowPage=&drwNoStart=1&drwNoEnd=\"+str(last_round)\n",
        "\n",
        "response = requests.get(url)\n",
        "response.encoding = 'euc-kr'  # 동행복권 사이트는 euc-kr 인코딩 사용\n",
        "html = response.text\n",
        "soup = BeautifulSoup(html, 'html.parser')\n",
        "\n",
        "table = soup.find(\"table\", {\"class\": \"tbl_data tbl_data_col\"})\n",
        "rows = table.find_all(\"tr\")[1:]  # 1행 이후부터 데이터\n",
        "\n",
        "lotto_data = []\n",
        "for row in rows:\n",
        "    cols = row.find_all(\"td\")\n",
        "    if len(cols) < 3:\n",
        "        continue\n",
        "\n",
        "    # (1) 회차\n",
        "    # \"1153회\"처럼 되어 있으면, 숫자 이외 문자는 제거하여 int 변환\n",
        "    draw_no_text = cols[0].text.strip()\n",
        "    draw_no_text = re.sub(r'[^0-9]', '', draw_no_text)  # 숫자 외 모두 제거\n",
        "    draw_no = int(draw_no_text)  # 정수 변환\n",
        "\n",
        "    # (2) 당첨번호 + 보너스번호 (하나의 리스트로 묶음)\n",
        "    # - '\\n' 제거\n",
        "    # - 숫자와 공백 외 문자는 제거\n",
        "    numbers_text = (cols[1].text+cols[2].text).strip()\n",
        "    numbers_text = re.sub(r'[^0-9\\s]', '', numbers_text) # 숫자/공백 외 제거\n",
        "\n",
        "    # 공백 기준 split → 정수화\n",
        "    all_numbers = numbers_text.split()\n",
        "    all_numbers = list(map(int, all_numbers))\n",
        "\n",
        "    lotto_data.append({\n",
        "        \"회차\": draw_no,\n",
        "        \"번호\": all_numbers\n",
        "    })\n",
        "\n",
        "# 판다스 DataFrame으로 변환\n",
        "df = pd.DataFrame(lotto_data)"
      ],
      "metadata": {
        "id": "UqiCKClHqall"
      },
      "execution_count": 22,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import matplotlib\n",
        "import matplotlib.pyplot as plt\n",
        "import matplotlib.patches as mpatches\n",
        "\n",
        "matplotlib.rcParams['font.family'] = 'DejaVu Sans'\n",
        "\n",
        "def get_number_counts_total(df):\n",
        "    \"\"\"\n",
        "    df 전체를 대상으로, 각 번호(1~45)가 몇 번 등장했는지 반환\n",
        "    return: {번호: 출현 횟수}\n",
        "    \"\"\"\n",
        "    number_counts = defaultdict(int)\n",
        "    for _, row in df.iterrows():\n",
        "        for num in row[\"번호\"]:\n",
        "            number_counts[num] += 1\n",
        "    return dict(number_counts)\n",
        "\n",
        "# 구간별 색깔 매핑\n",
        "def get_color_for_number(num):\n",
        "    if 1 <= num <= 10:\n",
        "        return \"orange\"\n",
        "    elif 11 <= num <= 20:\n",
        "        return \"skyblue\"\n",
        "    elif 21 <= num <= 30:\n",
        "        return \"red\"\n",
        "    elif 31 <= num <= 40:\n",
        "        return \"gray\"\n",
        "    else:  # 40~45\n",
        "        return \"green\"\n",
        "\n",
        "# 예: df가 (회차, 번호[]) 형태라고 가정\n",
        "counts_dict = get_number_counts_total(df)\n",
        "counts_series = pd.Series(counts_dict).sort_index()\n",
        "\n",
        "# 각 막대(bar)에 대응하는 색깔 리스트 만들기\n",
        "colors_for_bars = [get_color_for_number(num) for num in counts_series.index]\n",
        "\n",
        "# 그래프 그리기\n",
        "plt.figure(figsize=(10, 6))\n",
        "plt.bar(counts_series.index, counts_series.values, color=colors_for_bars)\n",
        "\n",
        "plt.title(\"Number statistics \"+str(last_round))\n",
        "plt.xlabel(\"Number\")\n",
        "plt.ylabel(\"Count\")\n",
        "plt.xticks(range(1, 46))\n",
        "plt.grid(axis='y', linestyle='--', alpha=0.5)\n",
        "\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 564
        },
        "id": "PLW4gsUB3GOo",
        "outputId": "da9b67a5-b3c3-47aa-ed07-0b02e9a20382"
      },
      "execution_count": 23,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x600 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# 번호 가중치 계산"
      ],
      "metadata": {
        "id": "GX-tu1s12LSB"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def get_number_counts_in_range(df, start_round=None, end_round=None):\n",
        "    \"\"\"\n",
        "    df: DataFrame (회차, 번호)\n",
        "    start_round: 조회 시작 회차 (None이면 가장 처음 회차)\n",
        "    end_round: 조회 끝 회차 (None이면 가장 마지막 회차)\n",
        "\n",
        "    return: {번호: 출현 횟수} 딕셔너리\n",
        "    \"\"\"\n",
        "    # 회차 범위 지정 (None이면 전체 범위 중 최소~최대)\n",
        "    if start_round is None:\n",
        "        start_round = df[\"회차\"].min()\n",
        "    if end_round is None:\n",
        "        end_round = df[\"회차\"].max()\n",
        "\n",
        "    # 해당 회차 범위 내 데이터만 필터\n",
        "    mask = (df[\"회차\"] >= start_round) & (df[\"회차\"] <= end_round)\n",
        "    sub_df = df[mask]\n",
        "\n",
        "    # 번호별 출현 횟수 계산\n",
        "    from collections import defaultdict\n",
        "    number_counts = defaultdict(int)\n",
        "\n",
        "    for _, row in sub_df.iterrows():\n",
        "        # row[\"번호\"]는 7개(당첨번호6 + 보너스1)\n",
        "        for num in row[\"번호\"]:\n",
        "            number_counts[num] += 1\n",
        "\n",
        "    return dict(number_counts)\n",
        "\n",
        "def get_number_counts_total(df):\n",
        "    \"\"\"\n",
        "    df 전체를 대상으로 번호별 출현 횟수 반환\n",
        "    \"\"\"\n",
        "    return get_number_counts_in_range(df, None, None)\n",
        "\n",
        "def get_number_weights(df, start_round=None, end_round=None):\n",
        "    \"\"\"\n",
        "    start_round, end_round 범위에서\n",
        "    번호별 출현 횟수 -> 출현 비율(= 가중치)로 계산\n",
        "\n",
        "    return: {번호: 가중치} 딕셔너리 (합이 1)\n",
        "    \"\"\"\n",
        "    counts = get_number_counts_in_range(df, start_round, end_round)\n",
        "\n",
        "    # 전체 출현 횟수\n",
        "    total_count = sum(counts.values())\n",
        "    if total_count == 0:\n",
        "        # 혹시 해당 구간에 아무 데이터가 없으면 (예: 잘못된 회차)\n",
        "        # 모든 번호의 가중치를 0으로 두거나, 1/45 균등분포로 처리\n",
        "        return {n: 1/45 for n in range(1,46)}\n",
        "\n",
        "    # 번호(1~45)에 대해, 없으면 0으로 처리\n",
        "    weights = {}\n",
        "    for num in range(1, 46):\n",
        "        cnt = counts.get(num, 0)\n",
        "        weights[num] = cnt / total_count\n",
        "\n",
        "    return weights"
      ],
      "metadata": {
        "id": "bEKSyrRRsCRE"
      },
      "execution_count": 24,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# 기간 가중치 계산"
      ],
      "metadata": {
        "id": "vsNs-NwY2Pb_"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def get_range_counts_in_interval(df, start_round, end_round):\n",
        "    \"\"\"\n",
        "    start_round~end_round 범위에서\n",
        "    1~9, 10~19, 20~29, 30~39, 40~45 구간별 출현 횟수 반환\n",
        "\n",
        "    return: { '1~9': cnt, '10~19': cnt, '20~29': cnt, '30~39': cnt, '40~45': cnt }\n",
        "    \"\"\"\n",
        "    counts = get_number_counts_in_range(df, start_round, end_round)\n",
        "\n",
        "    # 구간 초기화\n",
        "    intervals = {\n",
        "        '1~9': 0,\n",
        "        '10~19': 0,\n",
        "        '20~29': 0,\n",
        "        '30~39': 0,\n",
        "        '40~45': 0\n",
        "    }\n",
        "\n",
        "    for num, c in counts.items():\n",
        "        if 1 <= num <= 9:\n",
        "            intervals['1~9'] += c\n",
        "        elif 10 <= num <= 19:\n",
        "            intervals['10~19'] += c\n",
        "        elif 20 <= num <= 29:\n",
        "            intervals['20~29'] += c\n",
        "        elif 30 <= num <= 39:\n",
        "            intervals['30~39'] += c\n",
        "        elif 40 <= num <= 45:\n",
        "            intervals['40~45'] += c\n",
        "\n",
        "    return intervals\n",
        "\n",
        "def get_interval_weights(df, start_round, end_round):\n",
        "    \"\"\"\n",
        "    start_round~end_round 범위에서,\n",
        "    1~9, 10~19, 20~29, 30~39, 40~45 구간별 비율 -> 번호별 기간 가중치\n",
        "\n",
        "    예) 1~9 구간이 전체 120회 중 20회라면 = 0.166..,\n",
        "        해당 구간 번호(1~9)는 기간 가중치 0.166..을 공통으로 갖음.\n",
        "\n",
        "    return: {번호: 해당 구간의 비율} (합계=1이 아님. 구간 비율이므로)\n",
        "    \"\"\"\n",
        "    intervals_count = get_range_counts_in_interval(df, start_round, end_round)\n",
        "    total = sum(intervals_count.values())\n",
        "    if total == 0:\n",
        "        # 해당 범위에 데이터가 없으면, 모든 번호 동일 취급\n",
        "        return {n: 1 for n in range(1,46)}\n",
        "\n",
        "    # 구간별 비율\n",
        "    interval_ratios = {}\n",
        "    for k in intervals_count:\n",
        "        interval_ratios[k] = intervals_count[k] / total\n",
        "\n",
        "    # 번호별로 매핑\n",
        "    interval_weight = {}\n",
        "    for num in range(1, 46):\n",
        "        if 1 <= num <= 9:\n",
        "            interval_weight[num] = interval_ratios['1~9']\n",
        "        elif 10 <= num <= 19:\n",
        "            interval_weight[num] = interval_ratios['10~19']\n",
        "        elif 20 <= num <= 29:\n",
        "            interval_weight[num] = interval_ratios['20~29']\n",
        "        elif 30 <= num <= 39:\n",
        "            interval_weight[num] = interval_ratios['30~39']\n",
        "        else:\n",
        "            interval_weight[num] = interval_ratios['40~45']\n",
        "\n",
        "    return interval_weight\n",
        "\n",
        "\n",
        "def combine_weights(num_weight, interval_weight):\n",
        "    \"\"\"\n",
        "    두 종류의 가중치(번호별 dict)를 곱한 뒤, 전체 합이 1이 되도록 정규화\n",
        "    num_weight: {번호: w1}\n",
        "    interval_weight: {번호: w2}\n",
        "    return: {번호: 최종가중치}\n",
        "    \"\"\"\n",
        "    combined = {}\n",
        "    for n in range(1, 46):\n",
        "        w1 = num_weight.get(n, 0)\n",
        "        w2 = interval_weight.get(n, 0)\n",
        "        combined[n] = w1 * w2\n",
        "\n",
        "    total = sum(combined.values())\n",
        "    # 모두 0이라면(특정 구간에 데이터가 없어서 0이 된 경우) → 균등분포\n",
        "    if math.isclose(total, 0.0):\n",
        "        return {n: 1/45 for n in range(1,46)}\n",
        "\n",
        "    # 정규화\n",
        "    for n in combined:\n",
        "        combined[n] /= total\n",
        "\n",
        "    return combined"
      ],
      "metadata": {
        "id": "2u89RJVmwUk1"
      },
      "execution_count": 25,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def pick_one_number(weight_dict):\n",
        "    \"\"\"\n",
        "    weight_dict: {번호: 가중치} (가중치 합=1)\n",
        "    0~1 난수를 발생시키고, 해당 범위에 속하는 번호를 반환\n",
        "    \"\"\"\n",
        "    r = random.random()  # 0 <= r < 1\n",
        "    cumulative = 0.0\n",
        "\n",
        "    for num, w in weight_dict.items():\n",
        "        cumulative += w\n",
        "        if r <= cumulative:\n",
        "            return num\n",
        "\n",
        "    # 혹시 부동소수점 문제 등으로 마지막에 걸리는 경우\n",
        "    return list(weight_dict.keys())[-1]\n",
        "\n",
        "def pick_six_numbers(final_weights):\n",
        "    \"\"\"\n",
        "    final_weights: {번호: 최종가중치} (1~45)\n",
        "    6개 번호를 중복 없이 뽑는다.\n",
        "    return: 오름차순 정렬된 6개의 번호 리스트\n",
        "    \"\"\"\n",
        "    picked = []\n",
        "    temp_weights = final_weights.copy()\n",
        "\n",
        "    for _ in range(6):\n",
        "        num = pick_one_number(temp_weights)\n",
        "        picked.append(num)\n",
        "        # 뽑힌 번호의 가중치를 0으로 변경\n",
        "        temp_weights[num] = 0.0\n",
        "        # 다시 합계를 구해 정규화\n",
        "        s = sum(temp_weights.values())\n",
        "        if s > 0:\n",
        "            for k in temp_weights:\n",
        "                temp_weights[k] /= s\n",
        "\n",
        "    return sorted(picked)\n",
        "\n",
        "def draw_numbers_n_times(df, n,\n",
        "                         num_weight_range=None,   # (start,end) for 번호가중치 구간\n",
        "                         interval_weight_range=None, # (start,end) for 기간가중치 구간\n",
        "                         interval_name=\"최근3개월\"):\n",
        "    \"\"\"\n",
        "    n번(각 6개씩) 뽑아서 결과를 리스트로 반환\n",
        "    num_weight_range: (start_round, end_round)\n",
        "    interval_weight_range: (start_round, end_round)\n",
        "    interval_name: 디버깅/출력용 설명\n",
        "    \"\"\"\n",
        "    # 1) 번호 가중치\n",
        "    if num_weight_range is None:\n",
        "        # 전체 범위\n",
        "        w_num = get_number_weights(df, None, None)\n",
        "    else:\n",
        "        w_num = get_number_weights(df, num_weight_range[0], num_weight_range[1])\n",
        "\n",
        "    # 2) 기간(구간) 가중치\n",
        "    if interval_weight_range is None:\n",
        "        # 예: 최근 12회 라고 가정\n",
        "        latest_round = df[\"회차\"].max()\n",
        "        start_round = latest_round - 11  # 12회 분량\n",
        "        w_interval = get_interval_weights(df, start_round, latest_round)\n",
        "    else:\n",
        "        w_interval = get_interval_weights(df, interval_weight_range[0], interval_weight_range[1])\n",
        "\n",
        "    # 3) 최종 가중치\n",
        "    final_weights = combine_weights(w_num, w_interval)\n",
        "\n",
        "    # 4) n번 뽑기\n",
        "    results = []\n",
        "    for i in range(n):\n",
        "        picked_6 = pick_six_numbers(final_weights)\n",
        "        results.append(picked_6)\n",
        "\n",
        "    # 예시 출력\n",
        "    print(f\"[{interval_name}] 번호뽑기 결과 (총 {n}번):\")\n",
        "    for idx, nums in enumerate(results, start=1):\n",
        "        print(f\"{idx}회차 뽑기: {nums}\")\n",
        "\n",
        "    return results"
      ],
      "metadata": {
        "id": "VzMeJ2Ixweaj"
      },
      "execution_count": 26,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# 번호 뽑기"
      ],
      "metadata": {
        "id": "0byVKDRO2UTN"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# 예: 전체 번호가중치 × 최근 12회(3개월 가정) 구간가중치\n",
        "res = draw_numbers_n_times(df, n=5,\n",
        "                           num_weight_range=None,  # 전체 범위\n",
        "                           interval_weight_range=None,  # 최근 12회\n",
        "                           interval_name=\"최근12회\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "BEUK3UjWw3dC",
        "outputId": "078e1862-4ac9-4b01-b562-0cbd3c3bc249"
      },
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[최근12회] 번호뽑기 결과 (총 5번):\n",
            "1회차 뽑기: [5, 23, 32, 35, 37, 42]\n",
            "2회차 뽑기: [6, 16, 18, 23, 31, 34]\n",
            "3회차 뽑기: [1, 8, 31, 33, 38, 41]\n",
            "4회차 뽑기: [4, 6, 13, 15, 33, 38]\n",
            "5회차 뽑기: [1, 2, 25, 27, 37, 38]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# 여러 기간 가중치 합산하여 계산"
      ],
      "metadata": {
        "id": "X4UQyvTRynH9"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def get_interval_weights_multiple(df, intervals_configs):\n",
        "    \"\"\"\n",
        "    intervals_configs: 리스트 형태로, 각 요소에\n",
        "       {\n",
        "         'start_round': int,\n",
        "         'end_round': int,\n",
        "         'weight': float  # 이 구간에 부여할 가중치(예: 0.5)\n",
        "       }\n",
        "    를 담아둔다.\n",
        "\n",
        "    해당 구간들의 interval_weights를 구해 각각 weight만큼 곱해 합산한 뒤,\n",
        "    최종적으로 정규화하여 {번호: 최종기간가중치}를 반환한다.\n",
        "    \"\"\"\n",
        "\n",
        "    # (1) 일단 0으로 초기화\n",
        "    combined = {num: 0.0 for num in range(1, 46)}\n",
        "\n",
        "    # (2) 구간별로 loop\n",
        "    for cfg in intervals_configs:\n",
        "        st = cfg[\"start_round\"]\n",
        "        en = cfg[\"end_round\"]\n",
        "        w = cfg[\"weight\"]\n",
        "\n",
        "        # 2-1) 단일 구간에 대한 interval_weights 구함\n",
        "        interval_w = get_interval_weights(df, st, en)  # {번호: 구간 비율}\n",
        "\n",
        "        # 2-2) 구한 값에 weight를 곱해 combined에 누적\n",
        "        for num in range(1, 46):\n",
        "            combined[num] += interval_w[num] * w\n",
        "\n",
        "    # (3) 정규화 (합이 0이면(=모두 0이면) 균등분포 가정)\n",
        "    total = sum(combined.values())\n",
        "    if math.isclose(total, 0.0):\n",
        "        # 아무 데이터가 없다면 균등분포\n",
        "        return {n: 1/45 for n in range(1,46)}\n",
        "\n",
        "    for num in range(1, 46):\n",
        "        combined[num] /= total\n",
        "\n",
        "    return combined\n",
        "\n",
        "\n",
        "def draw_numbers_n_times_custom(df, n):\n",
        "\n",
        "    w_num = get_number_weights(df, None, None)\n",
        "\n",
        "    latest_round = df[\"회차\"].max() # 최근 1개월, 3개월, 6개월 가중치 합산\n",
        "    intervals_configs = [\n",
        "        {\"start_round\": latest_round-3, \"end_round\": latest_round, \"weight\": 0.5},   # 최근 1개월\n",
        "        {\"start_round\": latest_round-11, \"end_round\": latest_round, \"weight\": 0.3},  # 최근 3개월\n",
        "        {\"start_round\": latest_round-23, \"end_round\": latest_round, \"weight\": 0.2}   # 최근 6개월\n",
        "    ]\n",
        "    multi_interval_weight = get_interval_weights_multiple(df, intervals_configs)\n",
        "\n",
        "    final_weights = combine_weights(w_num, multi_interval_weight)\n",
        "\n",
        "    results = []\n",
        "    for i in range(n):\n",
        "        picked_6 = pick_six_numbers(final_weights)\n",
        "        results.append(picked_6)\n",
        "\n",
        "    print(f\"커스텀 번호뽑기 결과 (총 {n}번):\")\n",
        "    for idx, nums in enumerate(results, start=1):\n",
        "        print(f\"{nums}\")\n",
        "\n",
        "    return results"
      ],
      "metadata": {
        "id": "C5rNjzY5ymis"
      },
      "execution_count": 31,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# 여러 가중치 합산한 번호 뽑기"
      ],
      "metadata": {
        "id": "DMS6sVJG2hpN"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "res = draw_numbers_n_times_custom(df, n=10)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "fUnPF3kLzzIj",
        "outputId": "5863b81f-b453-455c-d31f-3015545921f7"
      },
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "커스텀 번호뽑기 결과 (총 10번):\n",
            "[17, 23, 30, 36, 38, 39]\n",
            "[6, 24, 25, 30, 32, 38]\n",
            "[1, 9, 16, 26, 28, 39]\n",
            "[1, 3, 4, 12, 30, 41]\n",
            "[7, 11, 33, 38, 39, 43]\n",
            "[16, 25, 30, 34, 35, 45]\n",
            "[5, 7, 12, 16, 24, 40]\n",
            "[6, 7, 16, 30, 36, 38]\n",
            "[4, 6, 8, 26, 34, 39]\n",
            "[1, 3, 4, 6, 16, 33]\n"
          ]
        }
      ]
    }
  ]
}