fish_market.ipynb

{
  "cells": [
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "id": "LgnWwrEhDGLd"
      },
      "outputs": [],
      "source": [
        "fish_length = [25.4, 26.3, 26.5, 29.0, 29.0, 29.7, 29.7, 30.0, 30.0, 30.7, 31.0, 31.0,\n",
        "                31.5, 32.0, 32.0, 32.0, 33.0, 33.0, 33.5, 33.5, 34.0, 34.0, 34.5, 35.0,\n",
        "                35.0, 35.0, 35.0, 36.0, 36.0, 37.0, 38.5, 38.5, 39.5, 41.0, 41.0, 9.8,\n",
        "                10.5, 10.6, 11.0, 11.2, 11.3, 11.8, 11.8, 12.0, 12.2, 12.4, 13.0, 14.3, 15.0]\n",
        "\n",
        "fish_weight = [242.0, 290.0, 340.0, 363.0, 430.0, 450.0, 500.0, 390.0, 450.0, 500.0, 475.0, 500.0,\n",
        "                500.0, 340.0, 600.0, 600.0, 700.0, 700.0, 610.0, 650.0, 575.0, 685.0, 620.0, 680.0,\n",
        "                700.0, 725.0, 720.0, 714.0, 850.0, 1000.0, 920.0, 955.0, 925.0, 975.0, 950.0, 6.7,\n",
        "                7.5, 7.0, 9.7, 9.8, 8.7, 10.0, 9.9, 9.8, 12.2, 13.4, 12.2, 19.7, 19.9]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "i5_Lxhxkb5vT"
      },
      "outputs": [],
      "source": [
        "fish_length"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "b3_9oYd8b9ga"
      },
      "outputs": [],
      "source": [
        "fish_weight"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "id": "881IuQ_xDf7k"
      },
      "outputs": [],
      "source": [
        "#35개는 도미(1), 14개는 빙어(0) 문자열 코드->숫자(문자열)\n",
        "fish_target=[1]*35+[0]*14"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#fish_length(길이) fish_weight(무게)의 데이터를 (x,y) 쌍으로 묶습니다.\n",
        "fish_data=[[l,w] for l, w in zip(fish_length, fish_weight)]"
      ],
      "metadata": {
        "id": "dR9h9-a3gFch"
      },
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "lVxWFc2oEgeB"
      },
      "outputs": [],
      "source": [
        "#총 샘플 숫자\n",
        "len(fish_data)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "metadata": {
        "id": "JgTUK9NVEgZn"
      },
      "outputs": [],
      "source": [
        "from sklearn.neighbors import KNeighborsClassifier\n",
        "kn=KNeighborsClassifier()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "CauovNfbI8r0"
      },
      "outputs": [],
      "source": [
        "fish_data\n",
        "#입력된 데이터 확인하기"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(fish_data[0])"
      ],
      "metadata": {
        "id": "AwV8Fj-XfHPi"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "IwDfJ8h8EgS6"
      },
      "outputs": [],
      "source": [
        "print(fish_data[4])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "_ZQ-aomeEgHj"
      },
      "outputs": [],
      "source": [
        "print(fish_data[0:5])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Ksi4kgb3F0pE"
      },
      "outputs": [],
      "source": [
        "print(fish_data[:5])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "B19BvK0MF_zE"
      },
      "outputs": [],
      "source": [
        "print(fish_data[44:])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 17,
      "metadata": {
        "id": "tHpvuf7lGGGf"
      },
      "outputs": [],
      "source": [
        "train_input=fish_data[:35]\n",
        "#훈련세트로 0번부터 34번째 인덱스까지 데이터까지 사용"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 18,
      "metadata": {
        "id": "VMaLJ9BYGi86"
      },
      "outputs": [],
      "source": [
        "train_target=fish_target[:35]\n",
        "#훈련세트로 타깃값 중에 0번 부터 34번째 인덱스의 데이터까지 사용"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 19,
      "metadata": {
        "id": "6p0sgKE-GzwD"
      },
      "outputs": [],
      "source": [
        "test_input=fish_data[35:]\n",
        "#테스트 세트로 35번째 인덱스부터 마지막 인덱스까지 사용"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 20,
      "metadata": {
        "id": "2MFPo96qIUE6"
      },
      "outputs": [],
      "source": [
        "test_target=fish_target[35:]\n",
        "#테스트 세트로 35번째 인덱스부터 마지막 인덱스까지 사용"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "onS3QGNkIiRJ"
      },
      "outputs": [],
      "source": [
        "kn=kn.fit(train_input,train_target)\n",
        "#훈련세트, 타깃세트를 집어놓고 훈련\n",
        "kn.score(test_input,test_target)\n",
        "#이제 정확도를 평가\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "kn=kn.fit(train_input,train_target)\n",
        "#훈련세트, 타깃세트를 집어놓고 훈련\n",
        "kn.score(test_input,test_target)\n",
        "#이제 정확도를 평가\n"
      ],
      "metadata": {
        "id": "8ECPiMwBNSYT"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "execution_count": 23,
      "metadata": {
        "id": "Gy7ESpnGJ8KQ"
      },
      "outputs": [],
      "source": [
        "import numpy as np"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 26,
      "metadata": {
        "id": "LuE8YkcrKMWk"
      },
      "outputs": [],
      "source": [
        "input_arr=np.array(fish_data)\n",
        "target_arr=np.array(fish_target)\n",
        "#생선데이터와 생선타겟를 2차원으로 배열 (좌표계에 표시)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "ufbu1n0_K9UR"
      },
      "outputs": [],
      "source": [
        "print(input_arr)\n",
        "#2차원으로 배열된 생선데이터를 출력"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "4-IaUbs9LJkz"
      },
      "outputs": [],
      "source": [
        "print(fish_data)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "lFzbPkQeLn4n"
      },
      "outputs": [],
      "source": [
        "print(input_arr.shape)\n",
        "#샘플의 숫자와 특성 수(변수)를 출력합니다."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 116,
      "metadata": {
        "id": "D9zHkLpxMuTQ"
      },
      "outputs": [],
      "source": [
        "np.random.seed()\n",
        "#무작위로 섞기\n",
        "index=np.arange(49)\n",
        "#0부터 49까지 1씩 증가하는 변수를 만듭니다.\n",
        "np.random.shuffle(index)\n",
        "#인덱스 섞기"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "GncCAUcZOG_M"
      },
      "outputs": [],
      "source": [
        "print(index)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "avlQbI0BOvek"
      },
      "outputs": [],
      "source": [
        "print(input_arr[[1,3]])\n",
        "#1번째와 3번째 값을 불러오기"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 33,
      "metadata": {
        "id": "dgBAODHjPErp"
      },
      "outputs": [],
      "source": [
        "train_input=input_arr[index[:35]]\n",
        "train_target=target_arr[index[:35]]\n",
        "#35개의 훈련 데이터 셋, 타겟 셋"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "mIcEiT-TQOs8"
      },
      "outputs": [],
      "source": [
        "print(input_arr[13],train_input[0])\n",
        "#input_arr의 13번째 값과 학습(train)에 넣은 0번째 수치를 비교"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 35,
      "metadata": {
        "id": "feFBuGMSRh8F"
      },
      "outputs": [],
      "source": [
        "test_input=input_arr[index[35:]]\n",
        "test_target=target_arr[index[35:]]\n",
        "#35번부터 49번(끝)까지 값들을 테스트 타겟으로 함"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 36,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 467
        },
        "id": "ORykMdVcSIh9",
        "outputId": "0cb6328f-db94-40bb-bec4-2f90179887b9"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<function matplotlib.pyplot.show(close=None, block=None)>"
            ]
          },
          "metadata": {},
          "execution_count": 36
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "import matplotlib.pyplot as plt\n",
        "#그래프를 그리는 함수를 불러오기\n",
        "\n",
        "# Convert train_input to a NumPy array\n",
        "train_input = np.array(train_input)\n",
        "\n",
        "\n",
        "plt.scatter(train_input[:,0], train_input[:,1])\n",
        "plt.scatter(test_input[:,0], test_input[:,1])\n",
        "plt.xlabel('length')\n",
        "plt.ylabel('weight')\n",
        "plt.show\n",
        "#파란색이 훈련세트 주황색이 테스트 세트\n",
        "#데이터에 도미와 빙어가 모두 섞여 있음"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "J-krrAVPT0wi"
      },
      "outputs": [],
      "source": [
        "kn=kn.fit(train_input, train_target)\n",
        "#테스트 데이터를 입력\n",
        "kn.score(test_input,test_target)"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "kn.score(train_input,train_target)"
      ],
      "metadata": {
        "id": "ttyrmVvBFP53"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "len(test_input)"
      ],
      "metadata": {
        "id": "t22iYadNkVrK"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "test_input"
      ],
      "metadata": {
        "id": "KYruZRK4kazh"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "u3b4ycLvUIum"
      },
      "outputs": [],
      "source": [
        "kn.predict(test_input)\n",
        "#자 한번 test_input을 놓고 난 뒤에 예측값 확인"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "q9XiKTSnURom"
      },
      "outputs": [],
      "source": [
        "test_target"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "W3xCz0vr8nHy"
      },
      "source": [
        "## 2번째 프로젝트"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 88,
      "metadata": {
        "id": "nV4ujl_M8rpx"
      },
      "outputs": [],
      "source": [
        "fish_length = [25.4, 26.3, 26.5, 29.0, 29.0, 29.7, 29.7, 30.0, 30.0, 30.7, 31.0, 31.0,\n",
        "                31.5, 32.0, 32.0, 32.0, 33.0, 33.0, 33.5, 33.5, 34.0, 34.0, 34.5, 35.0,\n",
        "                35.0, 35.0, 35.0, 36.0, 36.0, 37.0, 38.5, 38.5, 39.5, 41.0, 41.0, 9.8,\n",
        "                10.5, 10.6, 11.0, 11.2, 11.3, 11.8, 11.8, 12.0, 12.2, 12.4, 13.0, 14.3, 15.0]\n",
        "#생선의 길이 입력"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 89,
      "metadata": {
        "id": "1nYDhAgk9oLy"
      },
      "outputs": [],
      "source": [
        "fish_weight = [242.0, 290.0, 340.0, 363.0, 430.0, 450.0, 500.0, 390.0, 450.0, 500.0, 475.0, 500.0,\n",
        "                500.0, 340.0, 600.0, 600.0, 700.0, 700.0, 610.0, 650.0, 575.0, 685.0, 620.0, 680.0,\n",
        "                700.0, 725.0, 720.0, 714.0, 850.0, 1000.0, 920.0, 955.0, 925.0, 975.0, 950.0, 6.7,\n",
        "                7.5, 7.0, 9.7, 9.8, 8.7, 10.0, 9.9, 9.8, 12.2, 13.4, 12.2, 19.7, 19.9]\n",
        "#생선의 무게 입력"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 90,
      "metadata": {
        "id": "Tf1qG68j976V"
      },
      "outputs": [],
      "source": [
        "import numpy as np"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "7xJOUaMA-MWW"
      },
      "outputs": [],
      "source": [
        "np.column_stack(([1,2,3],[4,5,6]))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 92,
      "metadata": {
        "id": "D6YZgUJJ-lRi"
      },
      "outputs": [],
      "source": [
        "fish_data=np.column_stack((fish_length,fish_weight))\n",
        "#데이터를 2개를 합치고 이것을 배열의 형태로한다. 즉 2개의 변수를 연결한 데이터로 만든다."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Mcm4nwA-_G86"
      },
      "outputs": [],
      "source": [
        "print(fish_data[:5])\n",
        "#내용을 확인"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "-0O-i9Nv_XNc"
      },
      "outputs": [],
      "source": [
        "#물고기 종류를 0과 1로 했는데, 보다 쉽게게 np.one라는 함수를 사용\n",
        "print(np.ones(5))\n",
        "#1이 5개"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 95,
      "metadata": {
        "id": "JC5SagEH_oxH"
      },
      "outputs": [],
      "source": [
        "fish_target=np.concatenate((np.ones(35),np.zeros(14)))\n",
        "#1이 35개 0이 14인 데이터를 만듦\n",
        "#stack과는 달리 길게 데이터를 늘여뜨림 이런 것을 튜플이라고 한다."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "f2C-PkAb_zrZ"
      },
      "outputs": [],
      "source": [
        "print(fish_target)\n",
        "#내용확인"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 97,
      "metadata": {
        "id": "Gqg96nN4_8DC"
      },
      "outputs": [],
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "#train_test_split함수 -> 적정한 비율로 훈련세트와 테스트 세트를 나누어 준다."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 98,
      "metadata": {
        "id": "o71g57F7A2Ck"
      },
      "outputs": [],
      "source": [
        "train_input, test_input, train_target, test_target = train_test_split(fish_data, fish_target,stratify=fish_target , random_state=40)\n",
        "#4개 값으로 된 데이터 셑을 묶어서 훈련(train)데이터 셑 테스트 데이터 셑으로 나눈다. 그리고 랜덤하게 바꾸는데 42라는 특정한 패턴으로 진행"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "test_target"
      ],
      "metadata": {
        "id": "-J-P-ODEZqyS"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "fish_data"
      ],
      "metadata": {
        "id": "DwfdiKsFtWYs"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "9dfjn41jBRzU"
      },
      "outputs": [],
      "source": [
        "print(train_input.shape, test_input.shape)\n",
        "#훈련(train)과 테스트할 데이터의 특징\n",
        "#각각 36,13개 값, 공통적으로  2개 변수(길이와 무게)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "YVTFciH9CCyK"
      },
      "outputs": [],
      "source": [
        "print(train_target.shape, test_target.shape)\n",
        "#훈련할 타겟 값과 테스트 값 특징\n",
        "#각각 36개, 13개 값"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "QosA4P-7CSEW"
      },
      "outputs": [],
      "source": [
        "print(test_target)\n",
        "#테스트 타켓의 값들 빙어(0)가 3개, 도미(1)가 10개로 빙어가 다소 부족"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 106,
      "metadata": {
        "id": "XfdIGWeBCY_s"
      },
      "outputs": [],
      "source": [
        "train_input, test_input, train_target, test_target = train_test_split(fish_data, fish_target, random_state=100)\n",
        "# 타겟값을 조절 : stratify-> 실제 클래스의 비율에 맞게 그룹을 나눠준다.\n",
        "# random_state=42를 35~42까지 입력하고 모형의 정확도가 달라지는지 확인"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "pPtjT326Cp_E"
      },
      "outputs": [],
      "source": [
        "test_target\n",
        "# 빙어4마리 도미 8마리"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_target"
      ],
      "metadata": {
        "id": "ntxHFZh5JuyD"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.neighbors import KNeighborsClassifier\n",
        "kn=KNeighborsClassifier()\n"
      ],
      "metadata": {
        "id": "pCzND6zozaQX"
      },
      "execution_count": 110,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "kn=kn.fit(train_input, train_target)\n",
        "#테스트 데이터를 입력\n",
        "kn.score(test_input,test_target)"
      ],
      "metadata": {
        "id": "y1Ep_cESLbdA"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "kn.score(train_input,train_target)"
      ],
      "metadata": {
        "id": "RMUOrUj5LkHW"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_input, test_input, train_target, test_target = train_test_split(fish_data, fish_target, stratify=fish_target, random_state=30)\n",
        "# 타겟값을 조절 : stratify-> 실제 클래스의 비율에 맞게 그룹을 나눠준다.\n",
        "# random_state=42를 35~42까지 입력하고 모형의 정확도가 달라지는지 확인"
      ],
      "metadata": {
        "id": "LNwhEo-6n3TW"
      },
      "execution_count": 113,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "test_target\n",
        "# 빙어4마리 도미 8마리"
      ],
      "metadata": {
        "id": "DoBlFmWBn7yU"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "aMUU83ZxDA_g"
      },
      "source": [
        "##수상한 도미 한 마리##\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 65,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "gN0CbyeuDFyq",
        "outputId": "fbe80dad-af5b-4e55-a425-a73aef4ec1fb"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "1.0"
            ]
          },
          "metadata": {},
          "execution_count": 65
        }
      ],
      "source": [
        "from sklearn.neighbors import KNeighborsClassifier\n",
        "\n",
        "kn = KNeighborsClassifier()\n",
        "kn.fit(train_input, train_target)\n",
        "kn.score(test_input, test_target)\n",
        "#도미/빙어 분류모델을 불러오고 성능을 검증"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 66,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "z_Wd1Ue5Du4T",
        "outputId": "19213a59-9c7e-46d5-8dc7-7a00138589c3"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[1.]\n"
          ]
        }
      ],
      "source": [
        "print(kn.predict([[2500, 1500]]))\n",
        "#길어 25cm 무게 150g인 것은 도미(1)인가? 빙어(0)인가? : 빙어"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 67,
      "metadata": {
        "id": "ynxbtcmhDw1z"
      },
      "outputs": [],
      "source": [
        "import matplotlib.pyplot as plt\n",
        "#차트로 내용을 확인"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 68,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 449
        },
        "id": "fl7uvmPNDzRH",
        "outputId": "b2e008bd-a3e6-4165-894e-455451c9ad21"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "plt.scatter(train_input[:,0], train_input[:,1])\n",
        "plt.scatter(25, 150, marker='^')\n",
        "plt.xlabel('length')\n",
        "plt.ylabel('weight')\n",
        "plt.show()\n",
        "#25, 150에 해당하는 것을 표시한다."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 70,
      "metadata": {
        "id": "lThsLzU5D2OW"
      },
      "outputs": [],
      "source": [
        "distances, indexes = kn.kneighbors([[25, 1150]])\n",
        "#왜 25cm, 150g인 도미를 빙어로 판단했는지 살펴본다."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 71,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 449
        },
        "id": "52-1mkQAD31c",
        "outputId": "5fb60bfc-b063-45e6-98e9-3ba0bbc976a0"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "plt.scatter(train_input[:,0], train_input[:,1])\n",
        "plt.scatter(25, 150, marker='^')\n",
        "plt.scatter(train_input[indexes,0], train_input[indexes,1], marker='D')\n",
        "plt.xlabel('length')\n",
        "plt.ylabel('weight')\n",
        "plt.show()\n",
        "#삼각형(마커)에 가까운 5개에 해당하는 것들의 학습결과 : 빙어로 판단"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "p-UaZuLMD68W"
      },
      "outputs": [],
      "source": [
        "print(train_input[indexes])\n",
        "# 마커 앞에 있는 4개 생선"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "O2RU3VDfD8s_"
      },
      "outputs": [],
      "source": [
        "print(train_target[indexes])\n",
        "# 실제로 보니까. 4개는 빙어이고 나머지 1개는 도미"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "ytERtiqpD-oo"
      },
      "outputs": [],
      "source": [
        "print(distances)\n",
        "# 거리를 측정\n",
        "# 실제로 빙어에 해당하는 값들은 130이다."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "r_svFTcAEBUY"
      },
      "source": [
        "기준 맞추기"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 75,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 449
        },
        "id": "qR-xTBYKLjAW",
        "outputId": "8735d9af-d43c-4ec0-ad7b-328ab4e5f4f5"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAkQAAAGwCAYAAABIC3rIAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAAA5S0lEQVR4nO3de3RU5b3/8c8kIQm3DCSSTCIBUkAlAl5AwyCCLSmglJ8oRw8VlSoHKgUrAiq0JeFSDdLWKmrF2wFXAW9VVNBSKZqgJVwEkZsFisGg5MIBmeHShJDZvz/SGZmQhEmYzOxkv19rzSqz95OZZ++1l/Pps5/nu22GYRgCAACwsIhwdwAAACDcCEQAAMDyCEQAAMDyCEQAAMDyCEQAAMDyCEQAAMDyCEQAAMDyosLdgabA4/Ho0KFDatu2rWw2W7i7AwAAAmAYho4fP66UlBRFRNQ9BkQgCsChQ4eUmpoa7m4AAIAGOHjwoDp27FhnGwJRANq2bSup6oTGxcWFuTcAACAQbrdbqampvt/xuhCIAuC9TRYXF0cgAgCgiQlkuguTqgEAgOURiAAAgOURiAAAgOURiAAAgOURiAAAgOURiAAAgOURiAAAgOURiAAAMLEKT0W4u2AJBCIAAExqxb4VyliWoRX7VoS7K80elaoBADCht/a+rdn52ZKkrPVZ8hiGRl1ya5h71XyFdYRo3bp1GjFihFJSUmSz2fTOO+/47TcMQ1lZWUpOTlbLli2VmZmpffv2+bU5evSoxowZo7i4OLVr107jxo3TiRMn/Nps375d119/vWJjY5WamqoFCxY09qEBANBgc3MXa/b6bBlG1XvDkGavz9bc3MXh7VgzFtZAdPLkSV1xxRV69tlna9y/YMECLVy4UIsWLdLGjRvVunVrDR06VGVlZb42Y8aM0a5du7RmzRqtWrVK69at04QJE3z73W63hgwZos6dO2vLli363e9+p9mzZ+uFF15o9OMDAKC+5uYu1hsHnpAhyfsILptNMiS9ceAJQlEjsRmGN3+Gl81m04oVKzRy5EhJVaNDKSkpmjZtmqZPny5JcrlcSkpK0pIlSzR69Gh9+eWXSk9P1+bNm9W3b19J0urVq3XTTTfpm2++UUpKip577jn9+te/VnFxsaKjoyVJM2bM0DvvvKN//vOfAfXN7XbLbrfL5XLxcFcAQKN5a+/bVSND+j4Mnc0wJJuk2f3ncPssAPX5/TbtpOqCggIVFxcrMzPTt81utysjI0P5+fmSpPz8fLVr184XhiQpMzNTERER2rhxo6/NwIEDfWFIkoYOHao9e/bou+++q/G7y8vL5Xa7/V4AADSmFftWaHZ+7WFI+n6kaHZ+NhOtg8y0gai4uFiSlJSU5Lc9KSnJt6+4uFiJiYl++6OiohQfH+/XpqbPOPs7qsvJyZHdbve9UlNTL/yAAACoRYWnQvM2zJNUexjy8u6ft2EeS/KDyLSBKJxmzpwpl8vlex08eDDcXQIANGMtIlpoVr9ZkqTzTWTx7p/Vb5ZaRLRo5J5Zh2kDkcPhkCSVlJT4bS8pKfHtczgcKi0t9dt/5swZHT161K9NTZ9x9ndUFxMTo7i4OL8XAACN6Zbut2i2c45sqj0U+eYQOefolu63hLJ7zZ5pA1FaWpocDofWrl3r2+Z2u7Vx40Y5nU5JktPp1LFjx7RlyxZfm48++kgej0cZGRm+NuvWrVNFxffDimvWrNGll16q9u3bh+hoAAA4v1GX3KrbukyVdG4o8r6/rctUJlQ3grAGohMnTmjbtm3atm2bpKqJ1Nu2bVNhYaFsNpumTJmi3/72t3rvvfe0Y8cO3X333UpJSfGtROvRo4eGDRum8ePHa9OmTfrHP/6hyZMna/To0UpJSZEk3XHHHYqOjta4ceO0a9cuvf7663rqqac0derUMB01AAC1y7rhHt3eZarfSJF3ZOj2LlOVdcM94exe82WE0ccff2yoasK832vs2LGGYRiGx+MxZs2aZSQlJRkxMTHG4MGDjT179vh9xpEjR4yf/vSnRps2bYy4uDjjnnvuMY4fP+7X5osvvjAGDBhgxMTEGBdffLExf/78evXT5XIZkgyXy3VBxwsAQKD+sucto+eSnr7XX/a8Fe4uNTn1+f02TR0iM6MOEQAgHFbsW6F5G+ZpVr9ZzBlqgPr8fhOIAkAgAgCES4WngtVkDdQsCjMCAAARhkKEQAQAACyPQAQAAMIu3FW3CUQAACCsVuxboYxlGWF9PltU2L4ZAABY3lt739bs/GxJUtb6LHkMIyyFJxkhAgAAYTE3d7Fmr8/2K0A5e3225uYuDnlfCEQAACDk5uYu1hsHnpAhyWar2mazVVVofuPAEyEPRQQiAAAQUm/tfVtvHnhC0vdhyMv7/s0DT+itvW+HrE8EIgAAEDIr9q3Q7Pxsv5Gh6rwjRbPzs0M20ZpABAAAQqLCU6F5G+ZJqj0MeXn3z9swLyRL8glEAAAgJFpEtNCsfrMkSed7cJh3/6x+s0JSrZtABAAAQuaW7rdotnOObKo9FBmGZJM02zknZA+1JRABAGBB4awMPeqSW3Vbl6mSzg1F3ve3dZka0npEBCIAACzGDJWhs264R7d3meo3UuQdGbq9y1Rl3XBPSPtDIAIAwELe2vu2stZnqcJToaz1WSFd2l5d1g33aHb/OX51iGb3nxPyMCTx6A4AACxjbu5ivXlWMURvZehdh1xhCSFS1e2zCJtN8zbM06x+s0I2Z6g6m2Gcb5433G637Ha7XC6X4uLiwt0dAADqzVsZWvJf8u5NAeG4TXW2Ck9F0FeT1ef3m1tmAAA0c2asDF1dKJbW14VABABAM2bWytBmQyACAKCZMnNlaLMhEAEA0EyZuTK02RCIAABoxsxaGdpsCEQAADRzZqwMbTYEIgAALMBslaHNhkAEAIBFmKkytNlQqRoAAAsxS2Vos6FSdQCoVA0AaG4aozK02VCpGgAA1Km5h6H6IhABAADLIxABAADLIxABAADLIxABAADLIxABAADLIxABAADLIxABAADLIxABAADLIxABAADLIxABAADLIxABAADLIxABAADLIxABAADLIxABAADLIxABAADLIxABAADLIxABAADLIxABAADLIxABAADLIxABAADLIxABAADLIxABAADLIxABAADLIxABAADLIxABAADLIxABAADLIxABAADLIxABAADLM3Ugqqys1KxZs5SWlqaWLVuqa9eumjdvngzD8LUxDENZWVlKTk5Wy5YtlZmZqX379vl9ztGjRzVmzBjFxcWpXbt2GjdunE6cOBHqwwEAACZl6kD0+OOP67nnntMzzzyjL7/8Uo8//rgWLFigp59+2tdmwYIFWrhwoRYtWqSNGzeqdevWGjp0qMrKynxtxowZo127dmnNmjVatWqV1q1bpwkTJoTjkAAAgAnZjLOHW0zmJz/5iZKSkvTyyy/7to0aNUotW7bU0qVLZRiGUlJSNG3aNE2fPl2S5HK5lJSUpCVLlmj06NH68ssvlZ6ers2bN6tv376SpNWrV+umm27SN998o5SUlPP2w+12y263y+VyKS4urnEOFgAABFV9fr9NPULUv39/rV27Vnv37pUkffHFF/r000914403SpIKCgpUXFyszMxM39/Y7XZlZGQoPz9fkpSfn6927dr5wpAkZWZmKiIiQhs3bqzxe8vLy+V2u/1eAACg+YoKdwfqMmPGDLndbl122WWKjIxUZWWlHn30UY0ZM0aSVFxcLElKSkry+7ukpCTfvuLiYiUmJvrtj4qKUnx8vK9NdTk5OZozZ06wDwcAAJiUqUeI3njjDS1btkzLly/X1q1b9corr+j3v/+9XnnllUb93pkzZ8rlcvleBw8ebNTvAwAA4WXqEaKHHnpIM2bM0OjRoyVJvXr10tdff62cnByNHTtWDodDklRSUqLk5GTf35WUlOjKK6+UJDkcDpWWlvp97pkzZ3T06FHf31cXExOjmJiYRjgiAABgRqYeITp16pQiIvy7GBkZKY/HI0lKS0uTw+HQ2rVrffvdbrc2btwop9MpSXI6nTp27Ji2bNnia/PRRx/J4/EoIyMjBEcBAADMztQjRCNGjNCjjz6qTp066fLLL9fnn3+uJ554Qvfee68kyWazacqUKfrtb3+r7t27Ky0tTbNmzVJKSopGjhwpSerRo4eGDRum8ePHa9GiRaqoqNDkyZM1evTogFaYAQCA5s/Ugejpp5/WrFmz9Itf/EKlpaVKSUnRz3/+c2VlZfnaPPzwwzp58qQmTJigY8eOacCAAVq9erViY2N9bZYtW6bJkydr8ODBioiI0KhRo7Rw4cJwHBIAADAhU9chMgvqEAEA0PQ0mzpEAAAAoUAgAgAAlkcgAgAAlkcgAgAAlkcgAgAAlkcgAgAAlkcgAgAAlkcgAgAAlkcgAgAAlkcgAgAAlkcgAgAAlkcgAgAgQBWeinB3AY2EQAQAQABW7FuhjGUZWrFvRbi7gkYQFe4OAABgdm/tfVuz87MlSVnrs+QxDI265NYw9wrBxAgRAAB1mJu7WLPXZ8swqt4bhjR7fbbm5i4Ob8cQVAQiAABqMTd3sd448IQMSTZb1TabTTIkvXHgCUJRM0IgAgCgBm/tfVtvHnhC0vdhyMv7/s0DT+itvW+HuGdoDMwhAgCgmhX7Vmh2frbfyFB1Ntt/bp/lZyvCZtMt3W+psV2lx9CmgqMqPV6mxLaxujYtXpERtXwowsZmGN67oqiN2+2W3W6Xy+VSXFxcuLsDAGhEFZ4KZSzLqNcS+xYRLbRxzEa1iGjht331ziLNWblbRa4y37Zke6yyR6RrWM/koPUZNavP7ze3zAAAOEuLiBYanjZcknS+IQPv/uFpw2sMQxOXbvULQ5JU7CrTxKVbtXpnUdD6jAtHIAIA4CwVngq9X/C+pNpvl3l5979f8L7fiFKlx9CclbtVU57ybpuzcrcqPdykMQsCEQCgSTl9xqOXP/lKWe/u1MuffKXTZzxB/fwWES00q98sSYGPEM3qN8tvhGhTwdFzRob8/k5SkatMmwqOXmh3ESRMqgYANBk5H+zWi58U6OyBlUc/+FLjr0/TzJvSg/Y9t3S/RR7D8NUfqmmkyDAkm6TZzjnnTKguPV57GGpIOzQ+RogAAE1Czge79fw6/zAkSR5Den5dgXI+2B3U7xt1ya3KaHufpHNHirzvM9reV2PF6sS2sQF9R6Dt0PgIRAAA0zt9xqMXPymos82LnxQE9fZZpcfQrr2XqaxolCT5VaqWpLKiUdq197Ia5wFdmxavZHusapuCZFPVarNr0+KD1l9cGAIRAMD0/px/4JyRoeo8RlW7YPHOAzrjuuacUFRWNEpnXNfUOg8oMsKm7BFVt/CqhyLv++wR6dQjMhECEQDA9L4+eioo7So9hvL3H9G7275V/v4jda7yOnt+jy8UGZG+MFRTu7MN65ms5+68Wg67/20xhz1Wz915NXWITIZJ1QAA0+sc3+qC29W3SGL1+T1nXNfohOtqSZF1tjvbsJ7J+nG6g0rVTQAjRAAA07vL2UXnyxARtqp2NWlIkcSa5wF9H4YCnQcUGWGTs2uCbr7yYjm7JhCGTIpABAAwveioCI2/Pq3ONuOvT1N01Lk/aw0tkuidB1TbTTVDzANqTghEAIAmYeZN6fr5wLRzRooibNLPB9Zeh4giiQgEc4gAAE3GzJvSNW3IZfpz/gF9ffSUOse30l3OLjWODHk1tEiid2SpNjZVjSz9ON3BKFEzQCACADQp0VERGnf9DwJu39AiifUZWXJ2TQi4PzAnbpkBAJq1hhZJ5PEb1kIgAgA0aw0tksjjN6yFQAQAaPYaUiSRx29YC3OIAACWUN8iid6RpYlLt8om+S2/5/EbzY/NMKo/wxfVud1u2e12uVwuxcXFhbs7AIAQqm+Fa5hHfX6/GSECADQblR4j6I/J4PEb1kAgAgA0C405kuN9/AaaLyZVAwCavIY8qww4G4EIANCkNfRZZcDZCEQAgCaNZ5UhGAhEAIAmjYrSCAYCEQCgSaOiNIKBQAQAaNKoKI1gIBABAJq0hj6rDDgbgQgA0OQ15FllwNkozAgAaBYutKJ0Y1S5RtNBIAIANBsNrSjN88rALTMAgKVR5RoSgQgAYGFUuYYXgQgA0GxUegzl7z+id7d9q/z9R84bZKhyDS/mEAEAmoWGzAOiyjW8GCECADR5DZ0HRJVreBGIAABhcfqMRy9/8pWy3t2plz/5SqfPeBr0ORcyD4gq1/DilhkAIORyPtitFz8p0NkZ5dEPvtT469M086b0en1WfeYBVV+S761yPXHpVtkkv1BFlWtrMf0I0bfffqs777xTCQkJatmypXr16qXPPvvMt98wDGVlZSk5OVktW7ZUZmam9u3b5/cZR48e1ZgxYxQXF6d27dpp3LhxOnHiRKgPBQCgqjD0/Dr/MCRJHkN6fl2Bcj7YXa/Pu9B5QFS5hmTyEaLvvvtO1113nX74wx/qr3/9qzp06KB9+/apffv2vjYLFizQwoUL9corrygtLU2zZs3S0KFDtXv3bsXGVl3cY8aMUVFRkdasWaOKigrdc889mjBhgpYvXx6uQwMASzp9xqMXPymos82LnxRo2pDLFB0V2P9nD8Y8oAutco2mz2YYhmmLK8yYMUP/+Mc/9Mknn9S43zAMpaSkaNq0aZo+fbokyeVyKSkpSUuWLNHo0aP15ZdfKj09XZs3b1bfvn0lSatXr9ZNN92kb775RikpKed8bnl5ucrLy33v3W63UlNT5XK5FBcX1whHCgDW8PInX2ne+1+et92s4T007vofBPSZlR5DAx7/SMWushrnEdlUNdrz6SM/IuBYjNvtlt1uD+j329S3zN577z317dtXt912mxITE3XVVVfpxRdf9O0vKChQcXGxMjMzfdvsdrsyMjKUn58vScrPz1e7du18YUiSMjMzFRERoY0bN9b4vTk5ObLb7b5XampqIx0hAFjL10dPBbWdxNPuERymDkRfffWVnnvuOXXv3l1/+9vfNHHiRP3yl7/UK6+8IkkqLi6WJCUlJfn9XVJSkm9fcXGxEhMT/fZHRUUpPj7e16a6mTNnyuVy+V4HDx4M9qEBgCV1jm8V1HZezAPChTL1HCKPx6O+ffvqsccekyRdddVV2rlzpxYtWqSxY8c22vfGxMQoJiam0T4fAKzqLmcXPfrBl+dMqD5bhK2qXX0xDwgXwtQjRMnJyUpP919+2aNHDxUWFkqSHA6HJKmkpMSvTUlJiW+fw+FQaWmp3/4zZ87o6NGjvjYAgNCIjorQ+OvT6mwz/vq0gCdUV+d92v3NV14sZ9cEwhACZupAdN1112nPnj1+2/bu3avOnTtLktLS0uRwOLR27VrffrfbrY0bN8rpdEqSnE6njh07pi1btvjafPTRR/J4PMrIyAjBUQAAzjbzpnT9fGCaqmeVCJv084H1r0MEBIOpV5lt3rxZ/fv315w5c3T77bdr06ZNGj9+vF544QWNGTNGkvT4449r/vz5fsvut2/f7rfs/sYbb1RJSYkWLVrkW3bft2/fgJfd12eWOgAgMKfPePTn/AP6+ugpdY5vpbucXRo8MgTUpD6/36YORJK0atUqzZw5U/v27VNaWpqmTp2q8ePH+/YbhqHs7Gy98MILOnbsmAYMGKA//elPuuSSS3xtjh49qsmTJ2vlypWKiIjQqFGjtHDhQrVp0yagPhCIAABoeppVIDIDAhEAAE1Ps6lDBAAAEAoEIgAAYHkEIgAAYHkEIgAAYHkNCkRz587VqVPnPmfm3//+t+bOnXvBnQIAAAilBq0yi4yMVFFR0TnPCDty5IgSExNVWVkZtA6aAavMAABoehp9lZlhGLLZzi2H/sUXXyg+Pr4hHwkAABA29Xq4a/v27WWz2WSz2XTJJZf4haLKykqdOHFC9913X9A7CQAA0JjqFYiefPJJGYahe++9V3PmzJHdbvfti46OVpcuXXzPEAMAAGgq6hWIxo4dK6nqoar9+/dXixYtGqVTAAAAoVSvQOQ1aNAgeTwe7d27V6WlpfJ4PH77Bw4cGJTOAQAAhEKDAtGGDRt0xx136Ouvv1b1RWo2m63ZrTIDAADNW4MC0X333ae+ffvq/fffV3Jyco0rzgAAAJqKBgWiffv26S9/+Yu6desW7P4AAACEXIPqEGVkZOhf//pXsPsCAAAQFgGPEG3fvt337/vvv1/Tpk1TcXGxevXqdc5qs969ewevhwAAAI0s4Ed3REREyGaznTOJ2vdB/9nXHCdV8+gOAACanvr8fgc8QlRQUHDBHQMAADCjgANR586dG7MfAAAAYdOgVWbvvfdejdttNptiY2PVrVs3paWlXVDHAAAAQqVBgWjkyJE1zic6ex7RgAED9M4776h9+/ZB6SgAAEBjadCy+zVr1uiaa67RmjVr5HK55HK5tGbNGmVkZGjVqlVat26djhw5ounTpwe7vwCAMKr0GMrff0TvbvtW+fuPqNJT97qc+rYHwqVBI0QPPPCAXnjhBfXv39+3bfDgwYqNjdWECRO0a9cuPfnkk7r33nuD1lEAQHit3lmkOSt3q8hV5tuWbI9V9oh0DeuZfMHtgXBq0AjR/v37a1y+FhcXp6+++kqS1L17d/3f//3fhfUOAGAKq3cWaeLSrX7hRpKKXWWauHSrVu8suqD2QLg1KBD16dNHDz30kA4fPuzbdvjwYT388MO65pprJFU93iM1NTU4vQQAhE2lx9CclbtV080u77Y5K3f7bofVtz1gBg0KRC+//LIKCgrUsWNHdevWTd26dVPHjh114MABvfTSS5KkEydO6De/+U1QOwsACL1NBUfPGek5myGpyFWmTQVHG9QeMIMGzSG69NJLtXv3bn344Yfau3evb9uPf/xjRURUZayRI0cGrZMAgPApPV57uKmpXX3bA2bQoEAkVT3KY9iwYRo2bFgw+wMAMJnEtrH1alff9oAZBByIFi5cqAkTJig2NlYLFy6ss+0vf/nLC+4YAMAcrk2LV7I9VsWushrnBdkkOeyxujYtvkHtATMI+OGuaWlp+uyzz5SQkFBnFWqbzeZbadZc8HBXAFbnXTUmyS/k2P7zv8/debXfUvr6tgcaQ31+vwMORFZGIAKAmusKOeJi9NNrO6nLRa2V2LZq1CcywlZre+oQIZRCFohOnz6tgoICde3aVVFRDZ6OZHoEIgCoUukxtKngqEqPl+nA/53Sq5sKVeyuPfCc3b56YAIaW31+vxu07P7UqVMaN26cWrVqpcsvv1yFhYWSpPvvv1/z589vyEcCAJqAyAibnF0TFBMVoSf/vtcvDEnnFl70tr/5yovl7JpAGIJpNSgQzZw5U1988YVyc3MVG/v9KoHMzEy9/vrrQescAMB8KLyI5qhBgeidd97RM888owEDBshm+z7tX3755dq/f3/QOgcAMB8KL6I5alAgOnz4sBITE8/ZfvLkSb+ABABofii8iOaoQYGob9++ev/9933vvSHopZdektPpDE7PAACmROFFNEcNWhr22GOP6cYbb9Tu3bt15swZPfXUU9q9e7fWr1+vvLy8YPcRAGAiFF5Ec9SgEaIBAwboiy++0JkzZ9SrVy99+OGHSkxMVH5+vvr06RPsPgIAGlmlx1D+/iN6d9u3yt9/pM4J0ZERNmWPSJf0faFFL+/77BHprChDk9KgOkR33323fvjDH2rgwIHq2rVrY/TLVKhDBKA5a2gBRQovwuwavTDj//zP/2jdunXav3+/UlJSNGjQIN1www0aNGiQunfv3uCOmxWBCEBz5X3ERvUfgkAfsUHhRZhZyCpVf/vtt1q3bp3y8vKUl5envXv3Kjk5Wd98801DP9KUCEQAmqNKj6EBj39U6xJ671ygTx/5ESEHTVKjV6r2at++vRISEtS+fXu1a9dOUVFR6tChw4V8JAAgRKgnBHyvQYHoV7/6lfr376+EhATNmDFDZWVlmjFjhoqLi/X5558Hu48AgEZAPSHgew1adj9//nx16NBB2dnZuvXWW3XJJZcEu18AgEZGPSHgew0KRJ9//rny8vKUm5urP/zhD4qOjvZNrL7hhhsISAAQBvWd4Ew9IeB7FzSp2uuLL77QH//4Ry1btkwej0eVlZXB6JtpMKkagNldyNL5iUu3SpJfKAp0lRlgZvX5/W7QCJFhGPr888+Vm5ur3Nxcffrpp3K73erdu7cGDRrUoE4DABqmtqXzxa4yTVy6tc5QM6xnsp678+pzwpSDekKwmAaNELVv314nTpzQFVdc4btVdv3116tdu3aN0MXwY4QIgFkFa+k89YTQHDX6CNHSpUt1/fXXEw4AIMzqs3Te2TWh1naREbY69wPNXYMC0fDhw4PdDwBAA7B0HgiOCyrMCAAIL5bOA8FBIAKAJsy7dL622T42Va02Y+k8UDcCEQA0YZERNmWPSJekc0KR9332iHQmSAPnQSACgCbOu3TeYfe/Leawx1JHCAhQgyZVAwDMZVjPZP043cHSeaCBmtQI0fz582Wz2TRlyhTftrKyMk2aNEkJCQlq06aNRo0apZKSEr+/Kyws1PDhw9WqVSslJibqoYce0pkzZ0LcewBoXN6l8zdfebGcXRMIQ0A9NJlAtHnzZj3//PPq3bu33/YHH3xQK1eu1Jtvvqm8vDwdOnRIt956q29/ZWWlhg8frtOnT2v9+vV65ZVXtGTJEmVlZYX6EAAAgEk1iUB04sQJjRkzRi+++KLat2/v2+5yufTyyy/riSee0I9+9CP16dNHixcv1vr167VhwwZJ0ocffqjdu3dr6dKluvLKK3XjjTdq3rx5evbZZ3X69OlwHRIAmEqlx1D+/iN6d9u3yt9/RJWeC37MJdCkNIlANGnSJA0fPlyZmZl+27ds2aKKigq/7Zdddpk6deqk/Px8SVJ+fr569eqlpKQkX5uhQ4fK7XZr165dNX5feXm53G633wsAmqvVO4s04PGP9NMXN+iB17bppy9u0IDHP9LqnUXh7hoQMqYPRK+99pq2bt2qnJycc/YVFxcrOjr6nGeoJSUlqbi42Nfm7DDk3e/dV5OcnBzZ7XbfKzU1NQhHAgDm430wbPXHf3gfDEsoglWYOhAdPHhQDzzwgJYtW6bY2NBVWZ05c6ZcLpfvdfDgwZB9NwCESqXH0JyVu1XTzTHvtjkrd3P7DJZg6kC0ZcsWlZaW6uqrr1ZUVJSioqKUl5enhQsXKioqSklJSTp9+rSOHTvm93clJSVyOBySJIfDcc6qM+97b5vqYmJiFBcX5/cCgOamPg+GBZo7UweiwYMHa8eOHdq2bZvv1bdvX40ZM8b37xYtWmjt2rW+v9mzZ48KCwvldDolSU6nUzt27FBpaamvzZo1axQXF6f09PSQHxMAmAUPhgW+Z+rCjG3btlXPnj39trVu3VoJCQm+7ePGjdPUqVMVHx+vuLg43X///XI6nerXr58kaciQIUpPT9ddd92lBQsWqLi4WL/5zW80adIkxcTEhPyYAMAseDAs8D1TB6JA/PGPf1RERIRGjRql8vJyDR06VH/60598+yMjI7Vq1SpNnDhRTqdTrVu31tixYzV37tww9hoAws/7YNhiV1mN84hsqnr8Bw+GhRXYDMNgttx5uN1u2e12uVwu5hMBaFa8q8wk+YUib41rnoWGpqw+v9+mnkMEAAhcQ4or8mBYoEqTv2UGAKga6ZmzcrffqrFke6yyR6SfN9TwYFiAW2YB4ZYZADPz3vaq/h9zbnvB6rhlBgAWQXFFIDgIRADQhFFcEQgOAhEANGEUVwSCg0AEAE0YxRWB4CAQAUAT5i2uWNt6MJuqVptRXBGoG4EIABqoIXV/gi0ywqbsEVXPZaweirzvs0ekm3IJvRnOH+BFHSIAaIALqfsTbN7iitX74whTfwJhpvMHSNQhCgh1iACczax1fyo9RpMormjW84fmhzpEANBIzFz3JzLCJmfXBN185cVydk0wZRgy8/mDtRGIAKAeqPtzYTh/MCsCEQDUA3V/LgznD2ZFIAKAeqDuz4Xh/MGsCEQAUA/U/bkwnD+YFYEIAOqhKdf9MQPOH8yKQAQA9eSt++Ow+9/WcdhjWTIeAM4fzIg6RAGgDhGAmjSVuj9mxflDY6vP7zeVqgGggbx1f9AwnD+YCbfMAACA5RGIAACA5RGIAACA5RGIAACA5RGIAACA5bHKDADqgaXiQPNEIAKAAK3eWaQ5K3f7Pa092R6r7BHpFBMEmjhumQFAAFbvLNLEpVv9wpAkFbvKNHHpVq3eWRSmngEIBgIRAJxHpcfQnJW7VVNZf++2OSt3q9JD4X+gqSIQAcB5bCo4es7I0NkMSUWuMm0qOBq6TgEIKgIRAJxH6fHaw1BD2gEwHwIRAJxHYtvY8zeqRzsA5kMgAoDzuDYtXsn2WNW2uN6mqtVm16bFh7JbAIKIQAQA5xEZYVP2iHRJOicUed9nj0inHhHQhBGIACAAw3om67k7r5bD7n9bzGGP1XN3Xk0dIqCJozAjAARoWM9k/TjdQaVqoBkiEAFAPURG2OTsmhDubgAIMm6ZAQAAyyMQAQAAyyMQAQAAyyMQAQAAyyMQAQAAyyMQAQAAyyMQAQAAyyMQAQAAyyMQAQAAyyMQAQAAyyMQAQAAyyMQAQAAyyMQAQAAyyMQAQAAyyMQAQAAyyMQAQAAyyMQAQAAyyMQAQAAyyMQAQAAyyMQAQAAyzN1IMrJydE111yjtm3bKjExUSNHjtSePXv82pSVlWnSpElKSEhQmzZtNGrUKJWUlPi1KSws1PDhw9WqVSslJibqoYce0pkzZ0J5KAAAwMRMHYjy8vI0adIkbdiwQWvWrFFFRYWGDBmikydP+to8+OCDWrlypd58803l5eXp0KFDuvXWW337KysrNXz4cJ0+fVrr16/XK6+8oiVLligrKyschwQAAEzIZhiGEe5OBOrw4cNKTExUXl6eBg4cKJfLpQ4dOmj58uX6r//6L0nSP//5T/Xo0UP5+fnq16+f/vrXv+onP/mJDh06pKSkJEnSokWL9Mgjj+jw4cOKjo4+7/e63W7Z7Xa5XC7FxcU16jECAIDgqM/vt6lHiKpzuVySpPj4eEnSli1bVFFRoczMTF+byy67TJ06dVJ+fr4kKT8/X7169fKFIUkaOnSo3G63du3aVeP3lJeXy+12+70AAEDz1WQCkcfj0ZQpU3TdddepZ8+ekqTi4mJFR0erXbt2fm2TkpJUXFzsa3N2GPLu9+6rSU5Ojux2u++Vmpoa5KMBAABm0mQC0aRJk7Rz50699tprjf5dM2fOlMvl8r0OHjzY6N8JAADCJyrcHQjE5MmTtWrVKq1bt04dO3b0bXc4HDp9+rSOHTvmN0pUUlIih8Pha7Np0ya/z/OuQvO2qS4mJkYxMTFBPgoAAGBWph4hMgxDkydP1ooVK/TRRx8pLS3Nb3+fPn3UokULrV271rdtz549KiwslNPplCQ5nU7t2LFDpaWlvjZr1qxRXFyc0tPTQ3MgAADA1Ew9QjRp0iQtX75c7777rtq2beub82O329WyZUvZ7XaNGzdOU6dOVXx8vOLi4nT//ffL6XSqX79+kqQhQ4YoPT1dd911lxYsWKDi4mL95je/0aRJkxgFAgAAkky+7N5ms9W4ffHixfrZz34mqaow47Rp0/Tqq6+qvLxcQ4cO1Z/+9Ce/22Fff/21Jk6cqNzcXLVu3Vpjx47V/PnzFRUVWB5k2T0AAE1PfX6/TR2IzIJABABA09Ns6xABAAA0BgIRAACwPAIRAACwPAIRAACwPAIRAOswDOnbLVX/CwBnIRABsI7tr0sv/kja/ka4ewLAZAhEAKyh8oz08WNV/859rOo9APwHgQiANez8i3Ts66p/f3dA2vlWWLsDwFwIRACaP9/okLf6fQSjRAD8EIgANH++0SHvZGoPo0QA/BCIADRv54wOeTFKBOB7BCIAzds5o0NejBIB+B6BCEDzVevokBejRACqEIgANF+F+bWMDnn9Z5SoMD+EnQJgRlHh7gAANJrUa6XblkhnTtfeJiq6qh0ASyMQAWi+omKky28Jdy8ANAHcMgMAAJZHIAIAAJZHIAIAAJZHIAIAAJZHIAIAAJZHIAIAAJZHIAIAAJZHIAIAAJZHIAIAAJZHIAIAAJZHIAIAAJZHIAIAAJZHIAIAAJZHIAIAAJZHIAIAAJZHIAIAAJZHIAIAAJZHIAIAAJZHIAIAAJZHIAIAAJZHIAIAAJZHIAIAAJZHIAIAAJZHIAIAAJZHIAIAAJZHIAIAAJZHIAIAAJZHIAIAAJZHIAIAAJZHIAIAAJZHIAIAAJZHIAIAAJZHIAIAAJYXFe4OwNoqPYY2FRxV6fEyJbaN1bVp8YqMsIW7WwAAiyEQIWxW7yzSnJW7VeQq821Ltscqe0S6hvVMDmPPAABWQyCCJP+Rmotax0g26f9OlNc4ahOMUZ3VO4s0celWGdW2F7vKNHHpVj1359WEIgBAyBCImriagkzp8XIdPVGu+NbRcthb1hpYvH/7993FWrHtWx09WVHjd5w9ahOMUZ1Kj6E5K3efE4YkyZBkkzRn5W79ON3B7TMAQEgQiMIo0JGW2trVFE5qUlNgCfRvpe9HbSYMTNML6woueFRnU8HROr/XkFTkKtOmgqNydk047+cBAHChCERhEuhIS23t/t8VyTWGk5oUVQsstd2uqo131ObFT2r+vvqO6pQeP38Iq087AAAuFMvuw8AbSKqPknhHWlbvLKqzXZGrTM8HGIa8DFUFltNnPLXerjrf33vq+KOzR3XOJ7FtbEDfGWg7AAAuFIEoxM43f0a6sOBSlyJXmf6cfyCg22QNFciozrVp8Uq2x6q2cSSbqkbBrk2LD2rfAACojaUC0bPPPqsuXbooNjZWGRkZ2rRpU8j7EOj8mcYKLl8fPRX0zzxbIKM6kRE2ZY9Il6RzQpH3ffaIdCZUAwBCxjKB6PXXX9fUqVOVnZ2trVu36oorrtDQoUNVWloa0n4EOi+msYJL5/hWDfo7m6S68kl9R3WG9UzWc3deLYfdP0A57LEsuQcAhJxlJlU/8cQTGj9+vO655x5J0qJFi/T+++/rf//3fzVjxgy/tuXl5SovL/e9d7vdQetHoPNiGhpc6pJsj9Vdzi566dMCFbvKAr4d581B46+vWmUmye9vGzqqM6xnsn6c7qBSNQAg7CwxQnT69Glt2bJFmZmZvm0RERHKzMxUfn7+Oe1zcnJkt9t9r9TU1KD1JdD5M3c5u9TZ7uz2gbCpKrBER0XUeruqNt5Rm5k3pQd9VCcywiZn1wTdfOXFcnZNIAwBAMLCZhhGMOftmtKhQ4d08cUXa/369XI6nb7tDz/8sPLy8rRx40a/9jWNEKWmpsrlcikuLu6C++NdPSbVPNJSfXl8be0mDEzTe18UBa0OUULraN18ZYoGX5bU6JWqAQBobG63W3a7PaDfb8vcMquPmJgYxcTENNrne+fPVA8kjmrBJZB2Dw/r0aBK1Rdyu8o7qgMAQHNhiUB00UUXKTIyUiUlJX7bS0pK5HA4wtKnQAPJ+dpdSDgh2AAAUMUSgSg6Olp9+vTR2rVrNXLkSEmSx+PR2rVrNXny5LD1K9BAQnABAKBxWSIQSdLUqVM1duxY9e3bV9dee62efPJJnTx50rfqDAAAWJdlAtF///d/6/Dhw8rKylJxcbGuvPJKrV69WklJSeHuGgAACDNLrDK7UPWZpQ4AAMyhPr/flqhDBAAAUBcCEQAAsDwCEQAAsDwCEQAAsDwCEQAAsDwCEQAAsDzL1CG6EN7KBG63O8w9AQAAgfL+bgdSYYhAFIDjx49LklJTU8PcEwAAUF/Hjx+X3W6vsw2FGQPg8Xh06NAhtW3bVjbb+Z8G31But1upqak6ePAgBSBrwPmpG+fn/DhHdeP81I3zUzcznh/DMHT8+HGlpKQoIqLuWUKMEAUgIiJCHTt2DNn3xcXFmeZiMiPOT904P+fHOaob56dunJ+6me38nG9kyItJ1QAAwPIIRAAAwPIIRCYSExOj7OxsxcTEhLsrpsT5qRvn5/w4R3Xj/NSN81O3pn5+mFQNAAAsjxEiAABgeQQiAABgeQQiAABgeQQiAABgeQSiMFi3bp1GjBihlJQU2Ww2vfPOO377DcNQVlaWkpOT1bJlS2VmZmrfvn3h6WwYnO/8/OxnP5PNZvN7DRs2LDydDYOcnBxdc801atu2rRITEzVy5Ejt2bPHr01ZWZkmTZqkhIQEtWnTRqNGjVJJSUmYehxagZyfG2644Zxr6L777gtTj0PrueeeU+/evX3F85xOp/7617/69lv52pHOf36sfO3UZP78+bLZbJoyZYpvW1O9hghEYXDy5EldccUVevbZZ2vcv2DBAi1cuFCLFi3Sxo0b1bp1aw0dOlRlZWUh7ml4nO/8SNKwYcNUVFTke7366qsh7GF45eXladKkSdqwYYPWrFmjiooKDRkyRCdPnvS1efDBB7Vy5Uq9+eabysvL06FDh3TrrbeGsdehE8j5kaTx48f7XUMLFiwIU49Dq2PHjpo/f762bNmizz77TD/60Y908803a9euXZKsfe1I5z8/knWvneo2b96s559/Xr179/bb3mSvIQNhJclYsWKF773H4zEcDofxu9/9zrft2LFjRkxMjPHqq6+GoYfhVf38GIZhjB071rj55pvD0h8zKi0tNSQZeXl5hmFUXS8tWrQw3nzzTV+bL7/80pBk5Ofnh6ubYVP9/BiGYQwaNMh44IEHwtcpk2nfvr3x0ksvce3Uwnt+DINrx+v48eNG9+7djTVr1vidk6Z8DTFCZDIFBQUqLi5WZmamb5vdbldGRoby8/PD2DNzyc3NVWJioi699FJNnDhRR44cCXeXwsblckmS4uPjJUlbtmxRRUWF3zV02WWXqVOnTpa8hqqfH69ly5bpoosuUs+ePTVz5kydOnUqHN0Lq8rKSr322ms6efKknE4n10411c+PF9eONGnSJA0fPtzvWpGa9n9/eLiryRQXF0uSkpKS/LYnJSX59lndsGHDdOuttyotLU379+/Xr371K914443Kz89XZGRkuLsXUh6PR1OmTNF1112nnj17Sqq6hqKjo9WuXTu/tla8hmo6P5J0xx13qHPnzkpJSdH27dv1yCOPaM+ePXr77bfD2NvQ2bFjh5xOp8rKytSmTRutWLFC6enp2rZtG9eOaj8/EteOJL322mvaunWrNm/efM6+pvzfHwIRmpzRo0f7/t2rVy/17t1bXbt2VW5urgYPHhzGnoXepEmTtHPnTn366afh7oop1XZ+JkyY4Pt3r169lJycrMGDB2v//v3q2rVrqLsZcpdeeqm2bdsml8ulv/zlLxo7dqzy8vLC3S3TqO38pKenW/7aOXjwoB544AGtWbNGsbGx4e5OUHHLzGQcDocknTMjv6SkxLcP/n7wgx/ooosu0r/+9a9wdyWkJk+erFWrVunjjz9Wx44dfdsdDodOnz6tY8eO+bW32jVU2/mpSUZGhiRZ5hqKjo5Wt27d1KdPH+Xk5OiKK67QU089xbXzH7Wdn5pY7drZsmWLSktLdfXVVysqKkpRUVHKy8vTwoULFRUVpaSkpCZ7DRGITCYtLU0Oh0Nr1671bXO73dq4caPfPWx875tvvtGRI0eUnJwc7q6EhGEYmjx5slasWKGPPvpIaWlpfvv79OmjFi1a+F1De/bsUWFhoSWuofOdn5ps27ZNkixzDVXn8XhUXl5u+WunNt7zUxOrXTuDBw/Wjh07tG3bNt+rb9++GjNmjO/fTfUa4pZZGJw4ccLv/00UFBRo27Ztio+PV6dOnTRlyhT99re/Vffu3ZWWlqZZs2YpJSVFI0eODF+nQ6iu8xMfH685c+Zo1KhRcjgc2r9/vx5++GF169ZNQ4cODWOvQ2fSpElavny53n33XbVt29Z3X95ut6tly5ay2+0aN26cpk6dqvj4eMXFxen++++X0+lUv379wtz7xne+87N//34tX75cN910kxISErR9+3Y9+OCDGjhw4DnLh5ujmTNn6sYbb1SnTp10/PhxLV++XLm5ufrb3/5m+WtHqvv8WP3akaS2bdv6zceTpNatWyshIcG3vcleQ+Fe5mZFH3/8sSHpnNfYsWMNw6haej9r1iwjKSnJiImJMQYPHmzs2bMnvJ0OobrOz6lTp4whQ4YYHTp0MFq0aGF07tzZGD9+vFFcXBzubodMTedGkrF48WJfm3//+9/GL37xC6N9+/ZGq1atjFtuucUoKioKX6dD6Hznp7Cw0Bg4cKARHx9vxMTEGN26dTMeeughw+VyhbfjIXLvvfcanTt3NqKjo40OHToYgwcPNj788EPffitfO4ZR9/mx+rVTm+qlCJrqNWQzDMMIZQADAAAwG+YQAQAAyyMQAQAAyyMQAQAAyyMQAQAAyyMQAQAAyyMQAQAAyyMQAQAAyyMQAQAAyyMQAWhybrjhBk2ZMiXc3VBubq5sNts5D7IE0PQQiAAgAGYJYQAaB4EIAABYHoEIQJNWXl6u6dOn6+KLL1br1q2VkZGh3Nxc3/4lS5aoXbt2+tvf/qYePXqoTZs2GjZsmIqKinxtzpw5o1/+8pdq166dEhIS9Mgjj2js2LEaOXKkJOlnP/uZ8vLy9NRTT8lms8lms+nAgQO+v9+yZYv69u2rVq1aqX///tqzZ0+Ijh5AsBCIADRpkydPVn5+vl577TVt375dt912m4YNG6Z9+/b52pw6dUq///3v9ec//1nr1q1TYWGhpk+f7tv/+OOPa9myZVq8eLH+8Y9/yO1265133vHtf+qpp+R0OjV+/HgVFRWpqKhIqampvv2//vWv9Yc//EGfffaZoqKidO+994bk2AEET1S4OwAADVVYWKjFixersLBQKSkpkqTp06dr9erVWrx4sR577DFJUkVFhRYtWqSuXbtKqgpRc+fO9X3O008/rZkzZ+qWW26RJD3zzDP64IMPfPvtdruio6PVqlUrORyOc/rx6KOPatCgQZKkGTNmaPjw4SorK1NsbGzjHDiAoCMQAWiyduzYocrKSl1yySV+28vLy5WQkOB736pVK18YkqTk5GSVlpZKklwul0pKSnTttdf69kdGRqpPnz7yeDwB9aN3795+ny1JpaWl6tSpU/0PCkBYEIgANFknTpxQZGSktmzZosjISL99bdq08f27RYsWfvtsNpsMwwhaP87+fJvNJkkBhykA5sAcIgBN1lVXXaXKykqVlpaqW7dufq+abm3VxG63KykpSZs3b/Ztq6ys1NatW/3aRUdHq7KyMqj9B2AejBABaLIuueQSjRkzRnfffbf+8Ic/6KqrrtLhw4e1du1a9e7dW8OHDw/oc+6//37l5OSoW7duuuyyy/T000/ru+++8432SFKXLl20ceNGHThwQG3atFF8fHxjHRaAMGCECECTtnjxYt19992aNm2aLr30Uo0cOVKbN2+u1/ydRx55RD/96U919913y+l0qk2bNho6dKjfpOjp06crMjJS6enp6tChgwoLCxvjcACEic0I5o10AGgGPB6PevToodtvv13z5s0Ld3cAhAC3zABY3tdff60PP/xQgwYNUnl5uZ555hkVFBTojjvuCHfXAIQIt8wAWF5ERISWLFmia665Rtddd5127Nihv//97+rRo0e4uwYgRLhlBgAALI8RIgAAYHkEIgAAYHkEIgAAYHkEIgAAYHkEIgAAYHkEIgAAYHkEIgAAYHkEIgAAYHn/H/be9fLe/6G7AAAAAElFTkSuQmCC\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "#원래 그림\n",
        "plt.scatter(train_input[:,0], train_input[:,1])\n",
        "plt.scatter(25, 150, marker='^')\n",
        "plt.scatter(train_input[indexes,0], train_input[indexes,1], marker='D')\n",
        "plt.xlabel('length')\n",
        "plt.ylabel('weight')\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 76,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 449
        },
        "id": "QpwzKWF_EBJE",
        "outputId": "96a202b1-e4b9-45bb-e861-0bf0992d4f05"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "#바꾼그림\n",
        "plt.scatter(train_input[:,0], train_input[:,1])\n",
        "plt.scatter(25, 150, marker='^')\n",
        "plt.scatter(train_input[indexes,0], train_input[indexes,1], marker='D')\n",
        "plt.xlim((0, 1000))\n",
        "#구간을 1000으로 동일하게 해서 본다\n",
        "plt.xlabel('length')\n",
        "plt.ylabel('weight')\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 77,
      "metadata": {
        "id": "E4TgX5sHD_u6"
      },
      "outputs": [],
      "source": [
        "#데이터의 원래 값이 아니라 표준화된 값으로 계산해서 분류한다.\n",
        "mean = np.mean(train_input, axis=0)\n",
        "std = np.std(train_input, axis=0)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "gc6wIYO0ELAt"
      },
      "outputs": [],
      "source": [
        "print(mean, std)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 79,
      "metadata": {
        "id": "S7sGpUMmENud"
      },
      "outputs": [],
      "source": [
        "train_scaled = (train_input - mean) / std\n",
        "#표준화 예시"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "pgJQOq1IENin"
      },
      "outputs": [],
      "source": [
        "train_scaled"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "zpm9nI09EyA1"
      },
      "source": [
        "전처리 데이터로 모델 훈련하기"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 81,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 449
        },
        "id": "Or96upp-E2Z1",
        "outputId": "f0005e68-4c0a-46df-e25f-373c3d5069d1"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "plt.scatter(train_scaled[:,0], train_scaled[:,1])\n",
        "plt.scatter(25, 150, marker='^')\n",
        "plt.xlabel('length')\n",
        "plt.ylabel('weight')\n",
        "plt.show()\n",
        "#표준화된 데이터를 가지고 데이터를 봄\n",
        "#25,150의 값도 마찬가지로 변환을 해야함"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 82,
      "metadata": {
        "id": "0Pa-sEjNE4X4"
      },
      "outputs": [],
      "source": [
        "new = ([25, 150] - mean) / std\n",
        "#수상한 도미(마커)에 해당 하는 값에 대해서 표준화"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "new"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "pk0n93knTDEN",
        "outputId": "ecef870c-447b-4123-87f5-18b094e86f99"
      },
      "execution_count": 83,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([-0.20160666, -0.93437391])"
            ]
          },
          "metadata": {},
          "execution_count": 83
        }
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 84,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 449
        },
        "id": "AtF_FMfsE6R0",
        "outputId": "39ff2ff6-16e5-4aeb-8153-7f61ca171c8a"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "plt.scatter(train_scaled[:,0], train_scaled[:,1])\n",
        "plt.scatter(new[0], new[1], marker='^')\n",
        "plt.xlabel('length')\n",
        "plt.ylabel('weight')\n",
        "plt.show()\n",
        "#표준화 한 값끼리 비교"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "fzhCNHmKE8Ga"
      },
      "outputs": [],
      "source": [
        "kn.fit(train_scaled, train_target)\n",
        "#표준화한 데이터를 가지고 학습"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 120,
      "metadata": {
        "id": "FFc6xeBUE9xr"
      },
      "outputs": [],
      "source": [
        "test_scaled = (test_input - mean) / std\n",
        "#테스트 할 값들에 대해서도 표준화"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "HsZtbfJgE_w5"
      },
      "outputs": [],
      "source": [
        "kn.score(test_scaled, test_target)\n",
        "#정확도는 불변"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "zY_-MSG9FCMx"
      },
      "outputs": [],
      "source": [
        "print(kn.predict([new]))\n",
        "#63번째 행과 비교"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "oqoxWY68FDt1"
      },
      "outputs": [],
      "source": [
        "distances, indexes = kn.kneighbors([new])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "j3ZIC549FFlD"
      },
      "outputs": [],
      "source": [
        "plt.scatter(train_scaled[:,0], train_scaled[:,1])\n",
        "plt.scatter(new[0], new[1], marker='^')\n",
        "plt.scatter(train_scaled[indexes,0], train_scaled[indexes,1], marker='D')\n",
        "plt.xlabel('length')\n",
        "plt.ylabel('weight')\n",
        "plt.show()\n",
        "#인접한 물고기들의 정확도 비교"
      ]
    }
  ],
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}