├── 5-5.ipynb
├── 2-2-1.ipynb
├── check_version.ipynb
├── 3.ipynb
├── 5-1.ipynb
├── explain_loc_iloc_get_loc.ipynb
├── 7-1.ipynb
├── 5-6.ipynb
└── 2-2-2.ipynb


/5-5.ipynb:
--------------------------------------------------------------------------------
  1 | {
  2 |  "cells": [
  3 |   {
  4 |    "cell_type": "markdown",
  5 |    "metadata": {},
  6 |    "source": [
  7 |     "## 例5-5. 特徴量ハッシング（別名「ハッシングトリック」）"
  8 |    ]
  9 |   },
 10 |   {
 11 |    "cell_type": "code",
 12 |    "execution_count": null,
 13 |    "metadata": {},
 14 |    "outputs": [],
 15 |    "source": [
 16 |     "import pandas as pd\n",
 17 |     "import json\n",
 18 |     "\n",
 19 |     "# 最初の10,000件のレビューを読み込み\n",
 20 |     "with open('data/yelp/yelp_academic_dataset_review.json') as f:\n",
 21 |     "    js = []\n",
 22 |     "    for i in range(10000):\n",
 23 |     "        js.append(json.loads(f.readline()))\n",
 24 |     "\n",
 25 |     "review_df = pd.DataFrame(js)\n",
 26 |     "# mにbusiness_idのユニーク数を代入\n",
 27 |     "m = len(review_df['business_id'].unique())\n",
 28 |     "\n",
 29 |     "m"
 30 |    ]
 31 |   },
 32 |   {
 33 |    "cell_type": "code",
 34 |    "execution_count": 2,
 35 |    "metadata": {},
 36 |    "outputs": [
 37 |     {
 38 |      "data": {
 39 |       "text/plain": [
 40 |        "['9yKzy9PApeiPPOUJEtnvkg',\n",
 41 |        " 'ZRJwVLyzEJq1VAihDhYiow',\n",
 42 |        " '6oRAC4uyJCsJl1X0WZpVSA',\n",
 43 |        " '_1QQZuf4zZOyFCvXc0o6Vg',\n",
 44 |        " '6ozycU1RpktNG2-1BroVtw']"
 45 |       ]
 46 |      },
 47 |      "execution_count": 2,
 48 |      "metadata": {},
 49 |      "output_type": "execute_result"
 50 |     }
 51 |    ],
 52 |    "source": [
 53 |     "from sklearn.feature_extraction import FeatureHasher\n",
 54 |     "h = FeatureHasher(n_features=m, input_type='string')\n",
 55 |     "f = h.transform(review_df['business_id'])\n",
 56 |     "\n",
 57 |     "# 変換後の特徴量が解釈が困難であることを確認\n",
 58 |     "review_df['business_id'].unique().tolist()[0:5]"
 59 |    ]
 60 |   },
 61 |   {
 62 |    "cell_type": "code",
 63 |    "execution_count": 7,
 64 |    "metadata": {},
 65 |    "outputs": [
 66 |     {
 67 |      "data": {
 68 |       "text/plain": [
 69 |        "array([[0., 0., 0., ..., 0., 0., 0.],\n",
 70 |        "       [0., 0., 0., ..., 0., 0., 0.],\n",
 71 |        "       [0., 0., 0., ..., 0., 0., 0.],\n",
 72 |        "       ...,\n",
 73 |        "       [0., 0., 0., ..., 0., 0., 0.],\n",
 74 |        "       [0., 0., 0., ..., 0., 0., 0.],\n",
 75 |        "       [0., 0., 0., ..., 0., 0., 0.]])"
 76 |       ]
 77 |      },
 78 |      "execution_count": 7,
 79 |      "metadata": {},
 80 |      "output_type": "execute_result"
 81 |     }
 82 |    ],
 83 |    "source": [
 84 |     "f.toarray()"
 85 |    ]
 86 |   },
 87 |   {
 88 |    "cell_type": "code",
 89 |    "execution_count": 8,
 90 |    "metadata": {},
 91 |    "outputs": [
 92 |     {
 93 |      "name": "stdout",
 94 |      "output_type": "stream",
 95 |      "text": [
 96 |       "Our pandas Series, in bytes:  790104\n",
 97 |       "Our hashed numpy array, in bytes:  56\n"
 98 |      ]
 99 |     }
100 |    ],
101 |    "source": [
102 |     "# 変換後の特徴量のストレージサイズが大きく減っていることを確認\n",
103 |     "from sys import getsizeof\n",
104 |     "print('Our pandas Series, in bytes: ', getsizeof(review_df['business_id']))\n",
105 |     "print('Our hashed numpy array, in bytes: ', getsizeof(f))"
106 |    ]
107 |   },
108 |   {
109 |    "cell_type": "code",
110 |    "execution_count": null,
111 |    "metadata": {},
112 |    "outputs": [],
113 |    "source": []
114 |   }
115 |  ],
116 |  "metadata": {
117 |   "kernelspec": {
118 |    "display_name": "Python 3",
119 |    "language": "python",
120 |    "name": "python3"
121 |   },
122 |   "language_info": {
123 |    "codemirror_mode": {
124 |     "name": "ipython",
125 |     "version": 3
126 |    },
127 |    "file_extension": ".py",
128 |    "mimetype": "text/x-python",
129 |    "name": "python",
130 |    "nbconvert_exporter": "python",
131 |    "pygments_lexer": "ipython3",
132 |    "version": "3.7.0"
133 |   }
134 |  },
135 |  "nbformat": 4,
136 |  "nbformat_minor": 2
137 | }
138 | 


--------------------------------------------------------------------------------
/2-2-1.ipynb:
--------------------------------------------------------------------------------
  1 | {
  2 |  "cells": [
  3 |   {
  4 |    "cell_type": "markdown",
  5 |    "metadata": {},
  6 |    "source": [
  7 |     "## 例2-1. The Echo Nest データセットの再生回数の二値化"
  8 |    ]
  9 |   },
 10 |   {
 11 |    "cell_type": "code",
 12 |    "execution_count": 4,
 13 |    "metadata": {},
 14 |    "outputs": [
 15 |     {
 16 |      "data": {
 17 |       "text/html": [
 18 |        "<div>\n",
 19 |        "<style scoped>\n",
 20 |        "    .dataframe tbody tr th:only-of-type {\n",
 21 |        "        vertical-align: middle;\n",
 22 |        "    }\n",
 23 |        "\n",
 24 |        "    .dataframe tbody tr th {\n",
 25 |        "        vertical-align: top;\n",
 26 |        "    }\n",
 27 |        "\n",
 28 |        "    .dataframe thead th {\n",
 29 |        "        text-align: right;\n",
 30 |        "    }\n",
 31 |        "</style>\n",
 32 |        "<table border=\"1\" class=\"dataframe\">\n",
 33 |        "  <thead>\n",
 34 |        "    <tr style=\"text-align: right;\">\n",
 35 |        "      <th></th>\n",
 36 |        "      <th>0</th>\n",
 37 |        "      <th>1</th>\n",
 38 |        "      <th>2</th>\n",
 39 |        "    </tr>\n",
 40 |        "  </thead>\n",
 41 |        "  <tbody>\n",
 42 |        "    <tr>\n",
 43 |        "      <th>0</th>\n",
 44 |        "      <td>b80344d063b5ccb3212f76538f3d9e43d87dca9e</td>\n",
 45 |        "      <td>SOAKIMP12A8C130995</td>\n",
 46 |        "      <td>1</td>\n",
 47 |        "    </tr>\n",
 48 |        "    <tr>\n",
 49 |        "      <th>1</th>\n",
 50 |        "      <td>b80344d063b5ccb3212f76538f3d9e43d87dca9e</td>\n",
 51 |        "      <td>SOAPDEY12A81C210A9</td>\n",
 52 |        "      <td>1</td>\n",
 53 |        "    </tr>\n",
 54 |        "    <tr>\n",
 55 |        "      <th>2</th>\n",
 56 |        "      <td>b80344d063b5ccb3212f76538f3d9e43d87dca9e</td>\n",
 57 |        "      <td>SOBBMDR12A8C13253B</td>\n",
 58 |        "      <td>1</td>\n",
 59 |        "    </tr>\n",
 60 |        "    <tr>\n",
 61 |        "      <th>3</th>\n",
 62 |        "      <td>b80344d063b5ccb3212f76538f3d9e43d87dca9e</td>\n",
 63 |        "      <td>SOBFNSP12AF72A0E22</td>\n",
 64 |        "      <td>1</td>\n",
 65 |        "    </tr>\n",
 66 |        "    <tr>\n",
 67 |        "      <th>4</th>\n",
 68 |        "      <td>b80344d063b5ccb3212f76538f3d9e43d87dca9e</td>\n",
 69 |        "      <td>SOBFOVM12A58A7D494</td>\n",
 70 |        "      <td>1</td>\n",
 71 |        "    </tr>\n",
 72 |        "  </tbody>\n",
 73 |        "</table>\n",
 74 |        "</div>"
 75 |       ],
 76 |       "text/plain": [
 77 |        "                                          0                   1  2\n",
 78 |        "0  b80344d063b5ccb3212f76538f3d9e43d87dca9e  SOAKIMP12A8C130995  1\n",
 79 |        "1  b80344d063b5ccb3212f76538f3d9e43d87dca9e  SOAPDEY12A81C210A9  1\n",
 80 |        "2  b80344d063b5ccb3212f76538f3d9e43d87dca9e  SOBBMDR12A8C13253B  1\n",
 81 |        "3  b80344d063b5ccb3212f76538f3d9e43d87dca9e  SOBFNSP12AF72A0E22  1\n",
 82 |        "4  b80344d063b5ccb3212f76538f3d9e43d87dca9e  SOBFOVM12A58A7D494  1"
 83 |       ]
 84 |      },
 85 |      "execution_count": 4,
 86 |      "metadata": {},
 87 |      "output_type": "execute_result"
 88 |     }
 89 |    ],
 90 |    "source": [
 91 |     "import pandas as pd\n",
 92 |     "listen_count = pd.read_csv('data/millionsong/train_triplets.txt.zip', header=None, delimiter='\\t', compression='zip')\n",
 93 |     "\n",
 94 |     "# このデータはユーザID、曲ID、再生回数の3つの列で構成されます。\n",
 95 |     "# 再生回数 0 を含まないため、単に再生回数の列をすべて 1 で上書きすることで\n",
 96 |     "# 再生回数を二値化できます。\n",
 97 |     "listen_count.iloc[:, 2] = 1\n",
 98 |     "\n",
 99 |     "listen_count.head()"
100 |    ]
101 |   },
102 |   {
103 |    "cell_type": "code",
104 |    "execution_count": null,
105 |    "metadata": {},
106 |    "outputs": [],
107 |    "source": []
108 |   }
109 |  ],
110 |  "metadata": {
111 |   "kernelspec": {
112 |    "display_name": "Python 3",
113 |    "language": "python",
114 |    "name": "python3"
115 |   },
116 |   "language_info": {
117 |    "codemirror_mode": {
118 |     "name": "ipython",
119 |     "version": 3
120 |    },
121 |    "file_extension": ".py",
122 |    "mimetype": "text/x-python",
123 |    "name": "python",
124 |    "nbconvert_exporter": "python",
125 |    "pygments_lexer": "ipython3",
126 |    "version": "3.7.0"
127 |   }
128 |  },
129 |  "nbformat": 4,
130 |  "nbformat_minor": 2
131 | }
132 | 


--------------------------------------------------------------------------------
/check_version.ipynb:
--------------------------------------------------------------------------------
  1 | {
  2 |  "cells": [
  3 |   {
  4 |    "cell_type": "code",
  5 |    "execution_count": 21,
  6 |    "metadata": {},
  7 |    "outputs": [
  8 |     {
  9 |      "data": {
 10 |       "text/plain": [
 11 |        "'1.15.3'"
 12 |       ]
 13 |      },
 14 |      "execution_count": 21,
 15 |      "metadata": {},
 16 |      "output_type": "execute_result"
 17 |     }
 18 |    ],
 19 |    "source": [
 20 |     "import numpy as np\n",
 21 |     "np.version.full_version"
 22 |    ]
 23 |   },
 24 |   {
 25 |    "cell_type": "code",
 26 |    "execution_count": 22,
 27 |    "metadata": {},
 28 |    "outputs": [
 29 |     {
 30 |      "data": {
 31 |       "text/plain": [
 32 |        "'1.1.0'"
 33 |       ]
 34 |      },
 35 |      "execution_count": 22,
 36 |      "metadata": {},
 37 |      "output_type": "execute_result"
 38 |     }
 39 |    ],
 40 |    "source": [
 41 |     "import scipy as sp\n",
 42 |     "sp.version.full_version"
 43 |    ]
 44 |   },
 45 |   {
 46 |    "cell_type": "code",
 47 |    "execution_count": 23,
 48 |    "metadata": {},
 49 |    "outputs": [
 50 |     {
 51 |      "data": {
 52 |       "text/plain": [
 53 |        "'3.0.0'"
 54 |       ]
 55 |      },
 56 |      "execution_count": 23,
 57 |      "metadata": {},
 58 |      "output_type": "execute_result"
 59 |     }
 60 |    ],
 61 |    "source": [
 62 |     "import matplotlib\n",
 63 |     "matplotlib.__version__"
 64 |    ]
 65 |   },
 66 |   {
 67 |    "cell_type": "code",
 68 |    "execution_count": 24,
 69 |    "metadata": {},
 70 |    "outputs": [
 71 |     {
 72 |      "data": {
 73 |       "text/plain": [
 74 |        "'0.20.0'"
 75 |       ]
 76 |      },
 77 |      "execution_count": 24,
 78 |      "metadata": {},
 79 |      "output_type": "execute_result"
 80 |     }
 81 |    ],
 82 |    "source": [
 83 |     "import sklearn\n",
 84 |     "sklearn.__version__"
 85 |    ]
 86 |   },
 87 |   {
 88 |    "cell_type": "code",
 89 |    "execution_count": 20,
 90 |    "metadata": {},
 91 |    "outputs": [
 92 |     {
 93 |      "name": "stdout",
 94 |      "output_type": "stream",
 95 |      "text": [
 96 |       "\n",
 97 |       "INSTALLED VERSIONS\n",
 98 |       "------------------\n",
 99 |       "commit: None\n",
100 |       "python: 3.7.0.final.0\n",
101 |       "python-bits: 64\n",
102 |       "OS: Darwin\n",
103 |       "OS-release: 17.7.0\n",
104 |       "machine: x86_64\n",
105 |       "processor: i386\n",
106 |       "byteorder: little\n",
107 |       "LC_ALL: None\n",
108 |       "LANG: ja_JP.UTF-8\n",
109 |       "LOCALE: ja_JP.UTF-8\n",
110 |       "\n",
111 |       "pandas: 0.23.4\n",
112 |       "pytest: None\n",
113 |       "pip: 18.0\n",
114 |       "setuptools: 40.2.0\n",
115 |       "Cython: None\n",
116 |       "numpy: 1.15.3\n",
117 |       "scipy: 1.1.0\n",
118 |       "pyarrow: None\n",
119 |       "xarray: None\n",
120 |       "IPython: 7.0.1\n",
121 |       "sphinx: None\n",
122 |       "patsy: None\n",
123 |       "dateutil: 2.7.3\n",
124 |       "pytz: 2018.5\n",
125 |       "blosc: None\n",
126 |       "bottleneck: None\n",
127 |       "tables: None\n",
128 |       "numexpr: None\n",
129 |       "feather: None\n",
130 |       "matplotlib: 3.0.0\n",
131 |       "openpyxl: None\n",
132 |       "xlrd: None\n",
133 |       "xlwt: None\n",
134 |       "xlsxwriter: None\n",
135 |       "lxml: None\n",
136 |       "bs4: None\n",
137 |       "html5lib: None\n",
138 |       "sqlalchemy: None\n",
139 |       "pymysql: None\n",
140 |       "psycopg2: None\n",
141 |       "jinja2: 2.10\n",
142 |       "s3fs: None\n",
143 |       "fastparquet: None\n",
144 |       "pandas_gbq: None\n",
145 |       "pandas_datareader: None\n"
146 |      ]
147 |     }
148 |    ],
149 |    "source": [
150 |     "import pandas as pd\n",
151 |     "pd.show_versions()"
152 |    ]
153 |   },
154 |   {
155 |    "cell_type": "code",
156 |    "execution_count": null,
157 |    "metadata": {},
158 |    "outputs": [],
159 |    "source": []
160 |   }
161 |  ],
162 |  "metadata": {
163 |   "kernelspec": {
164 |    "display_name": "Python 3",
165 |    "language": "python",
166 |    "name": "python3"
167 |   },
168 |   "language_info": {
169 |    "codemirror_mode": {
170 |     "name": "ipython",
171 |     "version": 3
172 |    },
173 |    "file_extension": ".py",
174 |    "mimetype": "text/x-python",
175 |    "name": "python",
176 |    "nbconvert_exporter": "python",
177 |    "pygments_lexer": "ipython3",
178 |    "version": "3.7.0"
179 |   }
180 |  },
181 |  "nbformat": 4,
182 |  "nbformat_minor": 2
183 | }
184 | 


--------------------------------------------------------------------------------
/3.ipynb:
--------------------------------------------------------------------------------
  1 | {
  2 |  "cells": [
  3 |   {
  4 |    "cell_type": "markdown",
  5 |    "metadata": {},
  6 |    "source": [
  7 |     "## 例3-1. nグラムの計算"
  8 |    ]
  9 |   },
 10 |   {
 11 |    "cell_type": "code",
 12 |    "execution_count": 1,
 13 |    "metadata": {},
 14 |    "outputs": [
 15 |     {
 16 |      "name": "stdout",
 17 |      "output_type": "stream",
 18 |      "text": [
 19 |       "29222 368943 881620\n"
 20 |      ]
 21 |     },
 22 |     {
 23 |      "data": {
 24 |       "text/plain": [
 25 |        "['0', '00', '000', '007', '00a', '00am', '00pm', '01', '02', '03']"
 26 |       ]
 27 |      },
 28 |      "execution_count": 1,
 29 |      "metadata": {},
 30 |      "output_type": "execute_result"
 31 |     }
 32 |    ],
 33 |    "source": [
 34 |     "import pandas as pd\n",
 35 |     "import json\n",
 36 |     "from sklearn.feature_extraction.text import CountVectorizer\n",
 37 |     "\n",
 38 |     "# 最初の 10,000 件のレビューを読み込む\n",
 39 |     "with open('data/yelp/yelp_academic_dataset_review.json') as f:\n",
 40 |     "    js = []\n",
 41 |     "    for i in range(10000):\n",
 42 |     "        js.append(json.loads(f.readline()))\n",
 43 |     "review_df = pd.DataFrame(js)\n",
 44 |     "\n",
 45 |     "# scikit-learn の CountVectorizer を使ってユニグラム（BoW）、\n",
 46 |     "# バイグラム、トライグラムの特徴量変換器を作成する。\n",
 47 |     "# CountVectorizer はデフォルトでは1文字の単語を無視するが、\n",
 48 |     "# これは意味のない単語を除外するため実用的である。\n",
 49 |     "# ただしここでは全ての単語を含むように設定している。\n",
 50 |     "bow_converter = CountVectorizer(token_pattern='(?u)\\\\b\\\\w+\\\\b')\n",
 51 |     "bigram_converter = CountVectorizer(ngram_range=(2,2), token_pattern='(?u)\\\\b\\\\w+\\\\b')\n",
 52 |     "trigram_converter = CountVectorizer(ngram_range=(3,3), token_pattern='(?u)\\\\b\\\\w+\\\\b')\n",
 53 |     "\n",
 54 |     "# 変換器を適用し、語彙数を確認する\n",
 55 |     "bow_converter.fit(review_df['text'])\n",
 56 |     "words = bow_converter.get_feature_names()\n",
 57 |     "\n",
 58 |     "bigram_converter.fit(review_df['text'])\n",
 59 |     "bigrams = bigram_converter.get_feature_names()\n",
 60 |     "\n",
 61 |     "trigram_converter.fit(review_df['text'])\n",
 62 |     "trigrams = trigram_converter.get_feature_names()\n",
 63 |     "\n",
 64 |     "print (len(words), len(bigrams), len(trigrams))\n",
 65 |     "\n",
 66 |     "# n-グラムを確認する\n",
 67 |     "words[:10]"
 68 |    ]
 69 |   },
 70 |   {
 71 |    "cell_type": "code",
 72 |    "execution_count": 3,
 73 |    "metadata": {},
 74 |    "outputs": [
 75 |     {
 76 |      "data": {
 77 |       "text/plain": [
 78 |        "['zuzu was',\n",
 79 |        " 'zuzus room',\n",
 80 |        " 'zweigel wine',\n",
 81 |        " 'zwiebel kräuter',\n",
 82 |        " 'zy world',\n",
 83 |        " 'zzed in',\n",
 84 |        " 'éclairs napoleons',\n",
 85 |        " 'école lenôtre',\n",
 86 |        " 'ém all',\n",
 87 |        " 'òc châm']"
 88 |       ]
 89 |      },
 90 |      "execution_count": 3,
 91 |      "metadata": {},
 92 |      "output_type": "execute_result"
 93 |     }
 94 |    ],
 95 |    "source": [
 96 |     "bigrams[-10:]"
 97 |    ]
 98 |   },
 99 |   {
100 |    "cell_type": "code",
101 |    "execution_count": 4,
102 |    "metadata": {},
103 |    "outputs": [
104 |     {
105 |      "data": {
106 |       "text/plain": [
107 |        "['0 0 eye',\n",
108 |        " '0 20 less',\n",
109 |        " '0 39 oz',\n",
110 |        " '0 39 pizza',\n",
111 |        " '0 5 i',\n",
112 |        " '0 50 to',\n",
113 |        " '0 6 can',\n",
114 |        " '0 75 oysters',\n",
115 |        " '0 75 that',\n",
116 |        " '0 75 to']"
117 |       ]
118 |      },
119 |      "execution_count": 4,
120 |      "metadata": {},
121 |      "output_type": "execute_result"
122 |     }
123 |    ],
124 |    "source": [
125 |     "trigrams[:10]"
126 |    ]
127 |   },
128 |   {
129 |    "cell_type": "markdown",
130 |    "metadata": {},
131 |    "source": [
132 |     "## 例3-2. 品詞タグ付けとチャンク化"
133 |    ]
134 |   },
135 |   {
136 |    "cell_type": "code",
137 |    "execution_count": 2,
138 |    "metadata": {},
139 |    "outputs": [
140 |     {
141 |      "name": "stdout",
142 |      "output_type": "stream",
143 |      "text": [
144 |       "['General', 'PROPN', 'NNP']\n",
145 |       "['Manager', 'PROPN', 'NNP']\n",
146 |       "['Scott', 'PROPN', 'NNP']\n",
147 |       "['Petello', 'PROPN', 'NNP']\n",
148 |       "['is', 'VERB', 'VBZ']\n",
149 |       "['a', 'DET', 'DT']\n",
150 |       "['good', 'ADJ', 'JJ']\n",
151 |       "['egg', 'NOUN', 'NN']\n",
152 |       "['!', 'PUNCT', '.']\n",
153 |       "['!', 'PUNCT', '.']\n",
154 |       "['!', 'PUNCT', '.']\n",
155 |       "['Not', 'ADV', 'RB']\n",
156 |       "['to', 'PART', 'TO']\n",
157 |       "['go', 'VERB', 'VB']\n",
158 |       "['into', 'ADP', 'IN']\n",
159 |       "['detail', 'NOUN', 'NN']\n",
160 |       "[',', 'PUNCT', ',']\n",
161 |       "['but', 'CCONJ', 'CC']\n",
162 |       "['let', 'VERB', 'VB']\n",
163 |       "['me', 'PRON', 'PRP']\n",
164 |       "['assure', 'VERB', 'VB']\n",
165 |       "['you', 'PRON', 'PRP']\n",
166 |       "['if', 'ADP', 'IN']\n",
167 |       "['you', 'PRON', 'PRP']\n",
168 |       "['have', 'VERB', 'VBP']\n",
169 |       "['any', 'DET', 'DT']\n",
170 |       "['issues', 'NOUN', 'NNS']\n",
171 |       "['(', 'PUNCT', '-LRB-']\n",
172 |       "['albeit', 'ADP', 'IN']\n",
173 |       "['rare', 'ADJ', 'JJ']\n",
174 |       "[')', 'PUNCT', '-RRB-']\n",
175 |       "['speak', 'VERB', 'VBP']\n",
176 |       "['with', 'ADP', 'IN']\n",
177 |       "['Scott', 'PROPN', 'NNP']\n",
178 |       "['and', 'CCONJ', 'CC']\n",
179 |       "['treat', 'VERB', 'VB']\n",
180 |       "['the', 'DET', 'DT']\n",
181 |       "['guy', 'NOUN', 'NN']\n",
182 |       "['with', 'ADP', 'IN']\n",
183 |       "['some', 'DET', 'DT']\n",
184 |       "['respect', 'NOUN', 'NN']\n",
185 |       "['as', 'ADP', 'IN']\n",
186 |       "['you', 'PRON', 'PRP']\n",
187 |       "['state', 'VERB', 'VBP']\n",
188 |       "['your', 'ADJ', 'PRP$']\n",
189 |       "['case', 'NOUN', 'NN']\n",
190 |       "['and', 'CCONJ', 'CC']\n",
191 |       "['I', 'PRON', 'PRP']\n",
192 |       "[\"'d\", 'VERB', 'MD']\n",
193 |       "['be', 'VERB', 'VB']\n",
194 |       "['surprised', 'ADJ', 'JJ']\n",
195 |       "['if', 'ADP', 'IN']\n",
196 |       "['you', 'PRON', 'PRP']\n",
197 |       "['do', 'VERB', 'VBP']\n",
198 |       "[\"n't\", 'ADV', 'RB']\n",
199 |       "['walk', 'VERB', 'VB']\n",
200 |       "['out', 'ADV', 'RB']\n",
201 |       "['totally', 'ADV', 'RB']\n",
202 |       "['satisfied', 'ADJ', 'JJ']\n",
203 |       "['as', 'ADP', 'IN']\n",
204 |       "['I', 'PRON', 'PRP']\n",
205 |       "['just', 'ADV', 'RB']\n",
206 |       "['did', 'VERB', 'VBD']\n",
207 |       "['.', 'PUNCT', '.']\n",
208 |       "['Like', 'INTJ', 'UH']\n",
209 |       "['I', 'PRON', 'PRP']\n",
210 |       "['always', 'ADV', 'RB']\n",
211 |       "['say', 'VERB', 'VBP']\n",
212 |       "['.....', 'PUNCT', 'NFP']\n",
213 |       "['\"', 'PUNCT', '``']\n",
214 |       "['Mistakes', 'NOUN', 'NNS']\n",
215 |       "['are', 'VERB', 'VBP']\n",
216 |       "['inevitable', 'ADJ', 'JJ']\n",
217 |       "[',', 'PUNCT', ',']\n",
218 |       "['it', 'PRON', 'PRP']\n",
219 |       "[\"'s\", 'VERB', 'VBZ']\n",
220 |       "['how', 'ADV', 'WRB']\n",
221 |       "['we', 'PRON', 'PRP']\n",
222 |       "['recover', 'VERB', 'VBP']\n",
223 |       "['from', 'ADP', 'IN']\n",
224 |       "['them', 'PRON', 'PRP']\n",
225 |       "['that', 'ADJ', 'WDT']\n",
226 |       "['is', 'VERB', 'VBZ']\n",
227 |       "['important', 'ADJ', 'JJ']\n",
228 |       "['\"', 'PUNCT', \"''\"]\n",
229 |       "['!', 'PUNCT', '.']\n",
230 |       "['!', 'PUNCT', '.']\n",
231 |       "['!', 'PUNCT', '.']\n",
232 |       "['\\n\\n', 'SPACE', '_SP']\n",
233 |       "['Thanks', 'NOUN', 'NNS']\n",
234 |       "['to', 'ADP', 'IN']\n",
235 |       "['Scott', 'PROPN', 'NNP']\n",
236 |       "['and', 'CCONJ', 'CC']\n",
237 |       "['his', 'ADJ', 'PRP$']\n",
238 |       "['awesome', 'ADJ', 'JJ']\n",
239 |       "['staff', 'NOUN', 'NN']\n",
240 |       "['.', 'PUNCT', '.']\n",
241 |       "['You', 'PRON', 'PRP']\n",
242 |       "[\"'ve\", 'VERB', 'VB']\n",
243 |       "['got', 'VERB', 'VBN']\n",
244 |       "['a', 'DET', 'DT']\n",
245 |       "['customer', 'NOUN', 'NN']\n",
246 |       "['for', 'ADP', 'IN']\n",
247 |       "['life', 'NOUN', 'NN']\n",
248 |       "['!', 'PUNCT', '.']\n",
249 |       "['!', 'PUNCT', '.']\n",
250 |       "['..........', 'PUNCT', 'NFP']\n",
251 |       "[':', 'PUNCT', ':']\n",
252 |       "['^', 'PUNCT', 'NFP']\n",
253 |       "[')', 'PUNCT', '-RRB-']\n",
254 |       "[General Manager Scott Petello, a good egg, detail, me, you, you, any issues, Scott, the guy, some respect, you, your case, I, you, I, I, Mistakes, it, we, them, Thanks, Scott, his awesome staff, You, a customer, life]\n"
255 |      ]
256 |     }
257 |    ],
258 |    "source": [
259 |     "import pandas as pd\n",
260 |     "import json\n",
261 |     "\n",
262 |     "# 最初の10レビューを読み込む\n",
263 |     "with open('data/yelp/yelp_academic_dataset_review.json') as f:\n",
264 |     "    js = []\n",
265 |     "    for i in range(10):\n",
266 |     "        js.append(json.loads(f.readline()))\n",
267 |     "review_df = pd.DataFrame(js)\n",
268 |     "\n",
269 |     "# まずは Spacy を使った方法\n",
270 |     "import spacy\n",
271 |     "# 言語モデル（英語）を読み込む\n",
272 |     "nlp = spacy.load('en')\n",
273 |     "\n",
274 |     "# spaCy の言語モデルを使ってテキストから Pandas Series を作成する\n",
275 |     "doc_df = review_df['text'].apply(nlp)\n",
276 |     "\n",
277 |     "# spaCy は細かい品詞タグを .pos_ で、粗い品詞タグを .tag_ で提供します\n",
278 |     "for doc in doc_df[4]:\n",
279 |     "    print([doc.text, doc.pos_, doc.tag_])\n",
280 |     "\n",
281 |     "# spaCy は基本的な名詞句も .noun_chunks で提供します\n",
282 |     "print([chunk for chunk in doc_df[4].noun_chunks])"
283 |    ]
284 |   },
285 |   {
286 |    "cell_type": "code",
287 |    "execution_count": 7,
288 |    "metadata": {},
289 |    "outputs": [
290 |     {
291 |      "data": {
292 |       "text/plain": [
293 |        "[('General', 'NNP'),\n",
294 |        " ('Manager', 'NNP'),\n",
295 |        " ('Scott', 'NNP'),\n",
296 |        " ('Petello', 'NNP'),\n",
297 |        " ('is', 'VBZ'),\n",
298 |        " ('a', 'DT'),\n",
299 |        " ('good', 'JJ'),\n",
300 |        " ('egg', 'NN'),\n",
301 |        " ('Not', 'RB'),\n",
302 |        " ('to', 'TO'),\n",
303 |        " ('go', 'VB'),\n",
304 |        " ('into', 'IN'),\n",
305 |        " ('detail', 'NN'),\n",
306 |        " ('but', 'CC'),\n",
307 |        " ('let', 'VB'),\n",
308 |        " ('me', 'PRP'),\n",
309 |        " ('assure', 'VB'),\n",
310 |        " ('you', 'PRP'),\n",
311 |        " ('if', 'IN'),\n",
312 |        " ('you', 'PRP'),\n",
313 |        " ('have', 'VBP'),\n",
314 |        " ('any', 'DT'),\n",
315 |        " ('issues', 'NNS'),\n",
316 |        " ('albeit', 'IN'),\n",
317 |        " ('rare', 'NN'),\n",
318 |        " ('speak', 'NN'),\n",
319 |        " ('with', 'IN'),\n",
320 |        " ('Scott', 'NNP'),\n",
321 |        " ('and', 'CC'),\n",
322 |        " ('treat', 'VB'),\n",
323 |        " ('the', 'DT'),\n",
324 |        " ('guy', 'NN'),\n",
325 |        " ('with', 'IN'),\n",
326 |        " ('some', 'DT'),\n",
327 |        " ('respect', 'NN'),\n",
328 |        " ('as', 'IN'),\n",
329 |        " ('you', 'PRP'),\n",
330 |        " ('state', 'NN'),\n",
331 |        " ('your', 'PRP$'),\n",
332 |        " ('case', 'NN'),\n",
333 |        " ('and', 'CC'),\n",
334 |        " ('I', 'PRP'),\n",
335 |        " (\"'d\", 'MD'),\n",
336 |        " ('be', 'VB'),\n",
337 |        " ('surprised', 'VBN'),\n",
338 |        " ('if', 'IN'),\n",
339 |        " ('you', 'PRP'),\n",
340 |        " ('do', 'VBP'),\n",
341 |        " (\"n't\", 'RB'),\n",
342 |        " ('walk', 'VB'),\n",
343 |        " ('out', 'RP'),\n",
344 |        " ('totally', 'RB'),\n",
345 |        " ('satisfied', 'JJ'),\n",
346 |        " ('as', 'IN'),\n",
347 |        " ('I', 'PRP'),\n",
348 |        " ('just', 'RB'),\n",
349 |        " ('did', 'VBD'),\n",
350 |        " ('Like', 'IN'),\n",
351 |        " ('I', 'PRP'),\n",
352 |        " ('always', 'RB'),\n",
353 |        " ('say', 'VBP'),\n",
354 |        " ('..', 'VBP'),\n",
355 |        " ('Mistakes', 'NNS'),\n",
356 |        " ('are', 'VBP'),\n",
357 |        " ('inevitable', 'JJ'),\n",
358 |        " ('it', 'PRP'),\n",
359 |        " (\"'s\", 'VBZ'),\n",
360 |        " ('how', 'WRB'),\n",
361 |        " ('we', 'PRP'),\n",
362 |        " ('recover', 'VBP'),\n",
363 |        " ('from', 'IN'),\n",
364 |        " ('them', 'PRP'),\n",
365 |        " ('that', 'WDT'),\n",
366 |        " ('is', 'VBZ'),\n",
367 |        " ('important', 'JJ'),\n",
368 |        " ('Thanks', 'NNS'),\n",
369 |        " ('to', 'TO'),\n",
370 |        " ('Scott', 'NNP'),\n",
371 |        " ('and', 'CC'),\n",
372 |        " ('his', 'PRP$'),\n",
373 |        " ('awesome', 'JJ'),\n",
374 |        " ('staff', 'NN'),\n",
375 |        " ('You', 'PRP'),\n",
376 |        " (\"'ve\", 'VBP'),\n",
377 |        " ('got', 'VBN'),\n",
378 |        " ('a', 'DT'),\n",
379 |        " ('customer', 'NN'),\n",
380 |        " ('for', 'IN'),\n",
381 |        " ('life', 'NN'),\n",
382 |        " ('^', 'NN')]"
383 |       ]
384 |      },
385 |      "execution_count": 7,
386 |      "metadata": {},
387 |      "output_type": "execute_result"
388 |     }
389 |    ],
390 |    "source": [
391 |     "# TextBlob ライブラリを使って同じことができる\n",
392 |     "from textblob import TextBlob\n",
393 |     "\n",
394 |     "# TextBlob はデフォルトでは PatternTagger を使ってタグ付けを行う。\n",
395 |     "# これは今回の例ではうまくいくが、文法の正しくない文章を含む場合は \n",
396 |     "# NLTKTagger を使うことをおすすめする。\n",
397 |     "blob_df = review_df['text'].apply(TextBlob)\n",
398 |     "\n",
399 |     "blob_df[4].tags"
400 |    ]
401 |   },
402 |   {
403 |    "cell_type": "code",
404 |    "execution_count": 8,
405 |    "metadata": {},
406 |    "outputs": [
407 |     {
408 |      "name": "stdout",
409 |      "output_type": "stream",
410 |      "text": [
411 |       "['general manager', 'scott petello', 'good egg', 'scott', \"n't walk\", '... ..', 'mistakes', 'thanks', 'scott', 'awesome staff', '... ... ...']\n"
412 |      ]
413 |     }
414 |    ],
415 |    "source": [
416 |     "print([np for np in blob_df[4].noun_phrases])"
417 |    ]
418 |   },
419 |   {
420 |    "cell_type": "code",
421 |    "execution_count": null,
422 |    "metadata": {},
423 |    "outputs": [],
424 |    "source": []
425 |   }
426 |  ],
427 |  "metadata": {
428 |   "kernelspec": {
429 |    "display_name": "Python 3",
430 |    "language": "python",
431 |    "name": "python3"
432 |   },
433 |   "language_info": {
434 |    "codemirror_mode": {
435 |     "name": "ipython",
436 |     "version": 3
437 |    },
438 |    "file_extension": ".py",
439 |    "mimetype": "text/x-python",
440 |    "name": "python",
441 |    "nbconvert_exporter": "python",
442 |    "pygments_lexer": "ipython3",
443 |    "version": "3.7.0"
444 |   }
445 |  },
446 |  "nbformat": 4,
447 |  "nbformat_minor": 2
448 | }
449 | 


--------------------------------------------------------------------------------
/5-1.ipynb:
--------------------------------------------------------------------------------
  1 | {
  2 |  "cells": [
  3 |   {
  4 |    "cell_type": "markdown",
  5 |    "metadata": {},
  6 |    "source": [
  7 |     "## 例5-1. One-Hotエンコーディングとダミーコーディングを利用した線形回帰モデリング"
  8 |    ]
  9 |   },
 10 |   {
 11 |    "cell_type": "code",
 12 |    "execution_count": 1,
 13 |    "metadata": {},
 14 |    "outputs": [
 15 |     {
 16 |      "data": {
 17 |       "text/plain": [
 18 |        "3333.3333333333335"
 19 |       ]
 20 |      },
 21 |      "execution_count": 1,
 22 |      "metadata": {},
 23 |      "output_type": "execute_result"
 24 |     }
 25 |    ],
 26 |    "source": [
 27 |     "import pandas as pd\n",
 28 |     "from sklearn import linear_model\n",
 29 |     "\n",
 30 |     "# 3つの都市におけるアパートの家賃のデータセットを設定\n",
 31 |     "df = pd.DataFrame({\n",
 32 |     "    'City': ['SF', 'SF', 'SF', 'NYC', 'NYC', 'NYC', 'Seattle', 'Seattle', 'Seattle'],\n",
 33 |     "    'Rent': [3999, 4000, 4001, 3499, 3500, 3501, 2499, 2500, 2501]\n",
 34 |     "})\n",
 35 |     "\n",
 36 |     "df['Rent'].mean()"
 37 |    ]
 38 |   },
 39 |   {
 40 |    "cell_type": "code",
 41 |    "execution_count": 2,
 42 |    "metadata": {},
 43 |    "outputs": [
 44 |     {
 45 |      "data": {
 46 |       "text/html": [
 47 |        "<div>\n",
 48 |        "<style scoped>\n",
 49 |        "    .dataframe tbody tr th:only-of-type {\n",
 50 |        "        vertical-align: middle;\n",
 51 |        "    }\n",
 52 |        "\n",
 53 |        "    .dataframe tbody tr th {\n",
 54 |        "        vertical-align: top;\n",
 55 |        "    }\n",
 56 |        "\n",
 57 |        "    .dataframe thead th {\n",
 58 |        "        text-align: right;\n",
 59 |        "    }\n",
 60 |        "</style>\n",
 61 |        "<table border=\"1\" class=\"dataframe\">\n",
 62 |        "  <thead>\n",
 63 |        "    <tr style=\"text-align: right;\">\n",
 64 |        "      <th></th>\n",
 65 |        "      <th>Rent</th>\n",
 66 |        "      <th>city_NYC</th>\n",
 67 |        "      <th>city_SF</th>\n",
 68 |        "      <th>city_Seattle</th>\n",
 69 |        "    </tr>\n",
 70 |        "  </thead>\n",
 71 |        "  <tbody>\n",
 72 |        "    <tr>\n",
 73 |        "      <th>0</th>\n",
 74 |        "      <td>3999</td>\n",
 75 |        "      <td>0</td>\n",
 76 |        "      <td>1</td>\n",
 77 |        "      <td>0</td>\n",
 78 |        "    </tr>\n",
 79 |        "    <tr>\n",
 80 |        "      <th>1</th>\n",
 81 |        "      <td>4000</td>\n",
 82 |        "      <td>0</td>\n",
 83 |        "      <td>1</td>\n",
 84 |        "      <td>0</td>\n",
 85 |        "    </tr>\n",
 86 |        "    <tr>\n",
 87 |        "      <th>2</th>\n",
 88 |        "      <td>4001</td>\n",
 89 |        "      <td>0</td>\n",
 90 |        "      <td>1</td>\n",
 91 |        "      <td>0</td>\n",
 92 |        "    </tr>\n",
 93 |        "    <tr>\n",
 94 |        "      <th>3</th>\n",
 95 |        "      <td>3499</td>\n",
 96 |        "      <td>1</td>\n",
 97 |        "      <td>0</td>\n",
 98 |        "      <td>0</td>\n",
 99 |        "    </tr>\n",
100 |        "    <tr>\n",
101 |        "      <th>4</th>\n",
102 |        "      <td>3500</td>\n",
103 |        "      <td>1</td>\n",
104 |        "      <td>0</td>\n",
105 |        "      <td>0</td>\n",
106 |        "    </tr>\n",
107 |        "    <tr>\n",
108 |        "      <th>5</th>\n",
109 |        "      <td>3501</td>\n",
110 |        "      <td>1</td>\n",
111 |        "      <td>0</td>\n",
112 |        "      <td>0</td>\n",
113 |        "    </tr>\n",
114 |        "    <tr>\n",
115 |        "      <th>6</th>\n",
116 |        "      <td>2499</td>\n",
117 |        "      <td>0</td>\n",
118 |        "      <td>0</td>\n",
119 |        "      <td>1</td>\n",
120 |        "    </tr>\n",
121 |        "    <tr>\n",
122 |        "      <th>7</th>\n",
123 |        "      <td>2500</td>\n",
124 |        "      <td>0</td>\n",
125 |        "      <td>0</td>\n",
126 |        "      <td>1</td>\n",
127 |        "    </tr>\n",
128 |        "    <tr>\n",
129 |        "      <th>8</th>\n",
130 |        "      <td>2501</td>\n",
131 |        "      <td>0</td>\n",
132 |        "      <td>0</td>\n",
133 |        "      <td>1</td>\n",
134 |        "    </tr>\n",
135 |        "  </tbody>\n",
136 |        "</table>\n",
137 |        "</div>"
138 |       ],
139 |       "text/plain": [
140 |        "   Rent  city_NYC  city_SF  city_Seattle\n",
141 |        "0  3999         0        1             0\n",
142 |        "1  4000         0        1             0\n",
143 |        "2  4001         0        1             0\n",
144 |        "3  3499         1        0             0\n",
145 |        "4  3500         1        0             0\n",
146 |        "5  3501         1        0             0\n",
147 |        "6  2499         0        0             1\n",
148 |        "7  2500         0        0             1\n",
149 |        "8  2501         0        0             1"
150 |       ]
151 |      },
152 |      "execution_count": 2,
153 |      "metadata": {},
154 |      "output_type": "execute_result"
155 |     }
156 |    ],
157 |    "source": [
158 |     "# One-Hotエンコーディングをカテゴリ値であるcity列に適用\n",
159 |     "# 特徴量をOne-Hotエンコーディングで生成した列に、ターゲット変数を家賃に指定し、線形回帰モデルを学習\n",
160 |     "one_hot_df = pd.get_dummies(df, prefix=['city'])\n",
161 |     "one_hot_df"
162 |    ]
163 |   },
164 |   {
165 |    "cell_type": "code",
166 |    "execution_count": 6,
167 |    "metadata": {},
168 |    "outputs": [
169 |     {
170 |      "data": {
171 |       "text/plain": [
172 |        "array([ 166.66666667,  666.66666667, -833.33333333])"
173 |       ]
174 |      },
175 |      "execution_count": 6,
176 |      "metadata": {},
177 |      "output_type": "execute_result"
178 |     }
179 |    ],
180 |    "source": [
181 |     "model = linear_model.LinearRegression()\n",
182 |     "model.fit(one_hot_df[['city_NYC', 'city_SF', 'city_Seattle']], one_hot_df['Rent'])\n",
183 |     "model.coef_"
184 |    ]
185 |   },
186 |   {
187 |    "cell_type": "code",
188 |    "execution_count": 7,
189 |    "metadata": {},
190 |    "outputs": [
191 |     {
192 |      "data": {
193 |       "text/plain": [
194 |        "3333.3333333333335"
195 |       ]
196 |      },
197 |      "execution_count": 7,
198 |      "metadata": {},
199 |      "output_type": "execute_result"
200 |     }
201 |    ],
202 |    "source": [
203 |     "model.intercept_"
204 |    ]
205 |   },
206 |   {
207 |    "cell_type": "code",
208 |    "execution_count": 8,
209 |    "metadata": {
210 |     "scrolled": true
211 |    },
212 |    "outputs": [
213 |     {
214 |      "data": {
215 |       "text/html": [
216 |        "<div>\n",
217 |        "<style scoped>\n",
218 |        "    .dataframe tbody tr th:only-of-type {\n",
219 |        "        vertical-align: middle;\n",
220 |        "    }\n",
221 |        "\n",
222 |        "    .dataframe tbody tr th {\n",
223 |        "        vertical-align: top;\n",
224 |        "    }\n",
225 |        "\n",
226 |        "    .dataframe thead th {\n",
227 |        "        text-align: right;\n",
228 |        "    }\n",
229 |        "</style>\n",
230 |        "<table border=\"1\" class=\"dataframe\">\n",
231 |        "  <thead>\n",
232 |        "    <tr style=\"text-align: right;\">\n",
233 |        "      <th></th>\n",
234 |        "      <th>Rent</th>\n",
235 |        "      <th>city_SF</th>\n",
236 |        "      <th>city_Seattle</th>\n",
237 |        "    </tr>\n",
238 |        "  </thead>\n",
239 |        "  <tbody>\n",
240 |        "    <tr>\n",
241 |        "      <th>0</th>\n",
242 |        "      <td>3999</td>\n",
243 |        "      <td>1</td>\n",
244 |        "      <td>0</td>\n",
245 |        "    </tr>\n",
246 |        "    <tr>\n",
247 |        "      <th>1</th>\n",
248 |        "      <td>4000</td>\n",
249 |        "      <td>1</td>\n",
250 |        "      <td>0</td>\n",
251 |        "    </tr>\n",
252 |        "    <tr>\n",
253 |        "      <th>2</th>\n",
254 |        "      <td>4001</td>\n",
255 |        "      <td>1</td>\n",
256 |        "      <td>0</td>\n",
257 |        "    </tr>\n",
258 |        "    <tr>\n",
259 |        "      <th>3</th>\n",
260 |        "      <td>3499</td>\n",
261 |        "      <td>0</td>\n",
262 |        "      <td>0</td>\n",
263 |        "    </tr>\n",
264 |        "    <tr>\n",
265 |        "      <th>4</th>\n",
266 |        "      <td>3500</td>\n",
267 |        "      <td>0</td>\n",
268 |        "      <td>0</td>\n",
269 |        "    </tr>\n",
270 |        "    <tr>\n",
271 |        "      <th>5</th>\n",
272 |        "      <td>3501</td>\n",
273 |        "      <td>0</td>\n",
274 |        "      <td>0</td>\n",
275 |        "    </tr>\n",
276 |        "    <tr>\n",
277 |        "      <th>6</th>\n",
278 |        "      <td>2499</td>\n",
279 |        "      <td>0</td>\n",
280 |        "      <td>1</td>\n",
281 |        "    </tr>\n",
282 |        "    <tr>\n",
283 |        "      <th>7</th>\n",
284 |        "      <td>2500</td>\n",
285 |        "      <td>0</td>\n",
286 |        "      <td>1</td>\n",
287 |        "    </tr>\n",
288 |        "    <tr>\n",
289 |        "      <th>8</th>\n",
290 |        "      <td>2501</td>\n",
291 |        "      <td>0</td>\n",
292 |        "      <td>1</td>\n",
293 |        "    </tr>\n",
294 |        "  </tbody>\n",
295 |        "</table>\n",
296 |        "</div>"
297 |       ],
298 |       "text/plain": [
299 |        "   Rent  city_SF  city_Seattle\n",
300 |        "0  3999        1             0\n",
301 |        "1  4000        1             0\n",
302 |        "2  4001        1             0\n",
303 |        "3  3499        0             0\n",
304 |        "4  3500        0             0\n",
305 |        "5  3501        0             0\n",
306 |        "6  2499        0             1\n",
307 |        "7  2500        0             1\n",
308 |        "8  2501        0             1"
309 |       ]
310 |      },
311 |      "execution_count": 8,
312 |      "metadata": {},
313 |      "output_type": "execute_result"
314 |     }
315 |    ],
316 |    "source": [
317 |     "# ダミーコーディングを利用して線形回帰モデルを学習\n",
318 |     "dummy_df = pd.get_dummies(df, prefix=['city'], drop_first=True)\n",
319 |     "dummy_df"
320 |    ]
321 |   },
322 |   {
323 |    "cell_type": "code",
324 |    "execution_count": 9,
325 |    "metadata": {},
326 |    "outputs": [
327 |     {
328 |      "data": {
329 |       "text/plain": [
330 |        "array([  500., -1000.])"
331 |       ]
332 |      },
333 |      "execution_count": 9,
334 |      "metadata": {},
335 |      "output_type": "execute_result"
336 |     }
337 |    ],
338 |    "source": [
339 |     "model.fit(dummy_df[['city_SF', 'city_Seattle']], dummy_df['Rent'])\n",
340 |     "model.coef_"
341 |    ]
342 |   },
343 |   {
344 |    "cell_type": "code",
345 |    "execution_count": 10,
346 |    "metadata": {},
347 |    "outputs": [
348 |     {
349 |      "data": {
350 |       "text/plain": [
351 |        "3500.0"
352 |       ]
353 |      },
354 |      "execution_count": 10,
355 |      "metadata": {},
356 |      "output_type": "execute_result"
357 |     }
358 |    ],
359 |    "source": [
360 |     "model.intercept_"
361 |    ]
362 |   },
363 |   {
364 |    "cell_type": "markdown",
365 |    "metadata": {},
366 |    "source": [
367 |     "## 例5-2. Effectコーディングを用いた線形回帰モデル"
368 |    ]
369 |   },
370 |   {
371 |    "cell_type": "code",
372 |    "execution_count": 11,
373 |    "metadata": {},
374 |    "outputs": [
375 |     {
376 |      "data": {
377 |       "text/html": [
378 |        "<div>\n",
379 |        "<style scoped>\n",
380 |        "    .dataframe tbody tr th:only-of-type {\n",
381 |        "        vertical-align: middle;\n",
382 |        "    }\n",
383 |        "\n",
384 |        "    .dataframe tbody tr th {\n",
385 |        "        vertical-align: top;\n",
386 |        "    }\n",
387 |        "\n",
388 |        "    .dataframe thead th {\n",
389 |        "        text-align: right;\n",
390 |        "    }\n",
391 |        "</style>\n",
392 |        "<table border=\"1\" class=\"dataframe\">\n",
393 |        "  <thead>\n",
394 |        "    <tr style=\"text-align: right;\">\n",
395 |        "      <th></th>\n",
396 |        "      <th>Rent</th>\n",
397 |        "      <th>city_SF</th>\n",
398 |        "      <th>city_Seattle</th>\n",
399 |        "    </tr>\n",
400 |        "  </thead>\n",
401 |        "  <tbody>\n",
402 |        "    <tr>\n",
403 |        "      <th>0</th>\n",
404 |        "      <td>3999</td>\n",
405 |        "      <td>1.0</td>\n",
406 |        "      <td>0.0</td>\n",
407 |        "    </tr>\n",
408 |        "    <tr>\n",
409 |        "      <th>1</th>\n",
410 |        "      <td>4000</td>\n",
411 |        "      <td>1.0</td>\n",
412 |        "      <td>0.0</td>\n",
413 |        "    </tr>\n",
414 |        "    <tr>\n",
415 |        "      <th>2</th>\n",
416 |        "      <td>4001</td>\n",
417 |        "      <td>1.0</td>\n",
418 |        "      <td>0.0</td>\n",
419 |        "    </tr>\n",
420 |        "    <tr>\n",
421 |        "      <th>3</th>\n",
422 |        "      <td>3499</td>\n",
423 |        "      <td>-1.0</td>\n",
424 |        "      <td>-1.0</td>\n",
425 |        "    </tr>\n",
426 |        "    <tr>\n",
427 |        "      <th>4</th>\n",
428 |        "      <td>3500</td>\n",
429 |        "      <td>-1.0</td>\n",
430 |        "      <td>-1.0</td>\n",
431 |        "    </tr>\n",
432 |        "    <tr>\n",
433 |        "      <th>5</th>\n",
434 |        "      <td>3501</td>\n",
435 |        "      <td>-1.0</td>\n",
436 |        "      <td>-1.0</td>\n",
437 |        "    </tr>\n",
438 |        "    <tr>\n",
439 |        "      <th>6</th>\n",
440 |        "      <td>2499</td>\n",
441 |        "      <td>0.0</td>\n",
442 |        "      <td>1.0</td>\n",
443 |        "    </tr>\n",
444 |        "    <tr>\n",
445 |        "      <th>7</th>\n",
446 |        "      <td>2500</td>\n",
447 |        "      <td>0.0</td>\n",
448 |        "      <td>1.0</td>\n",
449 |        "    </tr>\n",
450 |        "    <tr>\n",
451 |        "      <th>8</th>\n",
452 |        "      <td>2501</td>\n",
453 |        "      <td>0.0</td>\n",
454 |        "      <td>1.0</td>\n",
455 |        "    </tr>\n",
456 |        "  </tbody>\n",
457 |        "</table>\n",
458 |        "</div>"
459 |       ],
460 |       "text/plain": [
461 |        "   Rent  city_SF  city_Seattle\n",
462 |        "0  3999      1.0           0.0\n",
463 |        "1  4000      1.0           0.0\n",
464 |        "2  4001      1.0           0.0\n",
465 |        "3  3499     -1.0          -1.0\n",
466 |        "4  3500     -1.0          -1.0\n",
467 |        "5  3501     -1.0          -1.0\n",
468 |        "6  2499      0.0           1.0\n",
469 |        "7  2500      0.0           1.0\n",
470 |        "8  2501      0.0           1.0"
471 |       ]
472 |      },
473 |      "execution_count": 11,
474 |      "metadata": {},
475 |      "output_type": "execute_result"
476 |     }
477 |    ],
478 |    "source": [
479 |     "effect_df = dummy_df.copy()\n",
480 |     "effect_df.loc[3:5, ['city_SF', 'city_Seattle']] = -1.0\n",
481 |     "effect_df"
482 |    ]
483 |   },
484 |   {
485 |    "cell_type": "code",
486 |    "execution_count": 12,
487 |    "metadata": {},
488 |    "outputs": [
489 |     {
490 |      "data": {
491 |       "text/plain": [
492 |        "array([ 666.66666667, -833.33333333])"
493 |       ]
494 |      },
495 |      "execution_count": 12,
496 |      "metadata": {},
497 |      "output_type": "execute_result"
498 |     }
499 |    ],
500 |    "source": [
501 |     "model.fit(effect_df[['city_SF', 'city_Seattle']], effect_df['Rent'])\n",
502 |     "model.coef_"
503 |    ]
504 |   },
505 |   {
506 |    "cell_type": "code",
507 |    "execution_count": 13,
508 |    "metadata": {},
509 |    "outputs": [
510 |     {
511 |      "data": {
512 |       "text/plain": [
513 |        "3333.3333333333335"
514 |       ]
515 |      },
516 |      "execution_count": 13,
517 |      "metadata": {},
518 |      "output_type": "execute_result"
519 |     }
520 |    ],
521 |    "source": [
522 |     "model.intercept_"
523 |    ]
524 |   },
525 |   {
526 |    "cell_type": "code",
527 |    "execution_count": null,
528 |    "metadata": {},
529 |    "outputs": [],
530 |    "source": []
531 |   }
532 |  ],
533 |  "metadata": {
534 |   "kernelspec": {
535 |    "display_name": "Python 3",
536 |    "language": "python",
537 |    "name": "python3"
538 |   },
539 |   "language_info": {
540 |    "codemirror_mode": {
541 |     "name": "ipython",
542 |     "version": 3
543 |    },
544 |    "file_extension": ".py",
545 |    "mimetype": "text/x-python",
546 |    "name": "python",
547 |    "nbconvert_exporter": "python",
548 |    "pygments_lexer": "ipython3",
549 |    "version": "3.7.0"
550 |   }
551 |  },
552 |  "nbformat": 4,
553 |  "nbformat_minor": 2
554 | }
555 | 


--------------------------------------------------------------------------------
/explain_loc_iloc_get_loc.ipynb:
--------------------------------------------------------------------------------
  1 | {
  2 |  "cells": [
  3 |   {
  4 |    "cell_type": "code",
  5 |    "execution_count": 2,
  6 |    "metadata": {},
  7 |    "outputs": [
  8 |     {
  9 |      "data": {
 10 |       "text/html": [
 11 |        "<div>\n",
 12 |        "<style scoped>\n",
 13 |        "    .dataframe tbody tr th:only-of-type {\n",
 14 |        "        vertical-align: middle;\n",
 15 |        "    }\n",
 16 |        "\n",
 17 |        "    .dataframe tbody tr th {\n",
 18 |        "        vertical-align: top;\n",
 19 |        "    }\n",
 20 |        "\n",
 21 |        "    .dataframe thead th {\n",
 22 |        "        text-align: right;\n",
 23 |        "    }\n",
 24 |        "</style>\n",
 25 |        "<table border=\"1\" class=\"dataframe\">\n",
 26 |        "  <thead>\n",
 27 |        "    <tr style=\"text-align: right;\">\n",
 28 |        "      <th></th>\n",
 29 |        "      <th>b</th>\n",
 30 |        "      <th>a</th>\n",
 31 |        "      <th>b</th>\n",
 32 |        "    </tr>\n",
 33 |        "  </thead>\n",
 34 |        "  <tbody>\n",
 35 |        "    <tr>\n",
 36 |        "      <th>10</th>\n",
 37 |        "      <td>1</td>\n",
 38 |        "      <td>2</td>\n",
 39 |        "      <td>3</td>\n",
 40 |        "    </tr>\n",
 41 |        "    <tr>\n",
 42 |        "      <th>11</th>\n",
 43 |        "      <td>4</td>\n",
 44 |        "      <td>5</td>\n",
 45 |        "      <td>6</td>\n",
 46 |        "    </tr>\n",
 47 |        "    <tr>\n",
 48 |        "      <th>11</th>\n",
 49 |        "      <td>7</td>\n",
 50 |        "      <td>8</td>\n",
 51 |        "      <td>9</td>\n",
 52 |        "    </tr>\n",
 53 |        "  </tbody>\n",
 54 |        "</table>\n",
 55 |        "</div>"
 56 |       ],
 57 |       "text/plain": [
 58 |        "    b  a  b\n",
 59 |        "10  1  2  3\n",
 60 |        "11  4  5  6\n",
 61 |        "11  7  8  9"
 62 |       ]
 63 |      },
 64 |      "execution_count": 2,
 65 |      "metadata": {},
 66 |      "output_type": "execute_result"
 67 |     }
 68 |    ],
 69 |    "source": [
 70 |     "import pandas as pd\n",
 71 |     "\n",
 72 |     "# DataFrame基礎知識\n",
 73 |     "# pandasにはIndexクラスがある\n",
 74 |     "# pandas.DataFrameには、indexとcolumnsというプロパティがあり、\n",
 75 |     "# indexとcolumnsプロパティはIndexクラスが設定される\n",
 76 |     "# pandas.Index https://pandas.pydata.org/pandas-docs/stable/generated/pandas.Index.html\n",
 77 |     "\n",
 78 |     "df = pd.DataFrame([[1, 2, 3], [4, 5, 6], [7, 8, 9]], index=[10, 11, 11],columns=['b', 'a', 'b'])\n",
 79 |     "df"
 80 |    ]
 81 |   },
 82 |   {
 83 |    "cell_type": "code",
 84 |    "execution_count": 21,
 85 |    "metadata": {},
 86 |    "outputs": [
 87 |     {
 88 |      "data": {
 89 |       "text/plain": [
 90 |        "b    1\n",
 91 |        "a    2\n",
 92 |        "b    3\n",
 93 |        "Name: 10, dtype: int64"
 94 |       ]
 95 |      },
 96 |      "execution_count": 21,
 97 |      "metadata": {},
 98 |      "output_type": "execute_result"
 99 |     }
100 |    ],
101 |    "source": [
102 |     "# locは、1次元の場合、index値で指定(pd.DataFrameは、indexとcolumnsというプロパティを持っている)\n",
103 |     "# 指定する値が1つで、該当件数が1件の場合pd.Seriesで返される\n",
104 |     "df.loc[10]"
105 |    ]
106 |   },
107 |   {
108 |    "cell_type": "code",
109 |    "execution_count": 4,
110 |    "metadata": {},
111 |    "outputs": [
112 |     {
113 |      "data": {
114 |       "text/html": [
115 |        "<div>\n",
116 |        "<style scoped>\n",
117 |        "    .dataframe tbody tr th:only-of-type {\n",
118 |        "        vertical-align: middle;\n",
119 |        "    }\n",
120 |        "\n",
121 |        "    .dataframe tbody tr th {\n",
122 |        "        vertical-align: top;\n",
123 |        "    }\n",
124 |        "\n",
125 |        "    .dataframe thead th {\n",
126 |        "        text-align: right;\n",
127 |        "    }\n",
128 |        "</style>\n",
129 |        "<table border=\"1\" class=\"dataframe\">\n",
130 |        "  <thead>\n",
131 |        "    <tr style=\"text-align: right;\">\n",
132 |        "      <th></th>\n",
133 |        "      <th>b</th>\n",
134 |        "      <th>a</th>\n",
135 |        "      <th>b</th>\n",
136 |        "    </tr>\n",
137 |        "  </thead>\n",
138 |        "  <tbody>\n",
139 |        "    <tr>\n",
140 |        "      <th>10</th>\n",
141 |        "      <td>1</td>\n",
142 |        "      <td>2</td>\n",
143 |        "      <td>3</td>\n",
144 |        "    </tr>\n",
145 |        "  </tbody>\n",
146 |        "</table>\n",
147 |        "</div>"
148 |       ],
149 |       "text/plain": [
150 |        "    b  a  b\n",
151 |        "10  1  2  3"
152 |       ]
153 |      },
154 |      "execution_count": 4,
155 |      "metadata": {},
156 |      "output_type": "execute_result"
157 |     }
158 |    ],
159 |    "source": [
160 |     "# locは、1次元の場合、index値で指定\n",
161 |     "# 指定する値がリストの場合は、該当件数が1件でもpd.DataFrameで返される\n",
162 |     "df.loc[[10]]"
163 |    ]
164 |   },
165 |   {
166 |    "cell_type": "code",
167 |    "execution_count": 5,
168 |    "metadata": {},
169 |    "outputs": [
170 |     {
171 |      "data": {
172 |       "text/html": [
173 |        "<div>\n",
174 |        "<style scoped>\n",
175 |        "    .dataframe tbody tr th:only-of-type {\n",
176 |        "        vertical-align: middle;\n",
177 |        "    }\n",
178 |        "\n",
179 |        "    .dataframe tbody tr th {\n",
180 |        "        vertical-align: top;\n",
181 |        "    }\n",
182 |        "\n",
183 |        "    .dataframe thead th {\n",
184 |        "        text-align: right;\n",
185 |        "    }\n",
186 |        "</style>\n",
187 |        "<table border=\"1\" class=\"dataframe\">\n",
188 |        "  <thead>\n",
189 |        "    <tr style=\"text-align: right;\">\n",
190 |        "      <th></th>\n",
191 |        "      <th>b</th>\n",
192 |        "      <th>a</th>\n",
193 |        "      <th>b</th>\n",
194 |        "    </tr>\n",
195 |        "  </thead>\n",
196 |        "  <tbody>\n",
197 |        "    <tr>\n",
198 |        "      <th>11</th>\n",
199 |        "      <td>4</td>\n",
200 |        "      <td>5</td>\n",
201 |        "      <td>6</td>\n",
202 |        "    </tr>\n",
203 |        "    <tr>\n",
204 |        "      <th>11</th>\n",
205 |        "      <td>7</td>\n",
206 |        "      <td>8</td>\n",
207 |        "      <td>9</td>\n",
208 |        "    </tr>\n",
209 |        "  </tbody>\n",
210 |        "</table>\n",
211 |        "</div>"
212 |       ],
213 |       "text/plain": [
214 |        "    b  a  b\n",
215 |        "11  4  5  6\n",
216 |        "11  7  8  9"
217 |       ]
218 |      },
219 |      "execution_count": 5,
220 |      "metadata": {},
221 |      "output_type": "execute_result"
222 |     }
223 |    ],
224 |    "source": [
225 |     "# locは、1次元の場合、index値で指定\n",
226 |     "# 指定する値が1つでも、該当件数が2件以上の場合pd.DataFrameで返される\n",
227 |     "df.loc[11]"
228 |    ]
229 |   },
230 |   {
231 |    "cell_type": "code",
232 |    "execution_count": 6,
233 |    "metadata": {},
234 |    "outputs": [
235 |     {
236 |      "data": {
237 |       "text/plain": [
238 |        "10    2\n",
239 |        "11    5\n",
240 |        "11    8\n",
241 |        "Name: a, dtype: int64"
242 |       ]
243 |      },
244 |      "execution_count": 6,
245 |      "metadata": {},
246 |      "output_type": "execute_result"
247 |     }
248 |    ],
249 |    "source": [
250 |     "# locは、2次元の場合、1次元目がindex値、2次元目がcolumns値で指定\n",
251 |     "# 該当件数が1件の場合は、pd.Seriesになる\n",
252 |     "df.loc[:, 'a']"
253 |    ]
254 |   },
255 |   {
256 |    "cell_type": "code",
257 |    "execution_count": 7,
258 |    "metadata": {},
259 |    "outputs": [
260 |     {
261 |      "data": {
262 |       "text/html": [
263 |        "<div>\n",
264 |        "<style scoped>\n",
265 |        "    .dataframe tbody tr th:only-of-type {\n",
266 |        "        vertical-align: middle;\n",
267 |        "    }\n",
268 |        "\n",
269 |        "    .dataframe tbody tr th {\n",
270 |        "        vertical-align: top;\n",
271 |        "    }\n",
272 |        "\n",
273 |        "    .dataframe thead th {\n",
274 |        "        text-align: right;\n",
275 |        "    }\n",
276 |        "</style>\n",
277 |        "<table border=\"1\" class=\"dataframe\">\n",
278 |        "  <thead>\n",
279 |        "    <tr style=\"text-align: right;\">\n",
280 |        "      <th></th>\n",
281 |        "      <th>a</th>\n",
282 |        "    </tr>\n",
283 |        "  </thead>\n",
284 |        "  <tbody>\n",
285 |        "    <tr>\n",
286 |        "      <th>10</th>\n",
287 |        "      <td>2</td>\n",
288 |        "    </tr>\n",
289 |        "    <tr>\n",
290 |        "      <th>11</th>\n",
291 |        "      <td>5</td>\n",
292 |        "    </tr>\n",
293 |        "    <tr>\n",
294 |        "      <th>11</th>\n",
295 |        "      <td>8</td>\n",
296 |        "    </tr>\n",
297 |        "  </tbody>\n",
298 |        "</table>\n",
299 |        "</div>"
300 |       ],
301 |       "text/plain": [
302 |        "    a\n",
303 |        "10  2\n",
304 |        "11  5\n",
305 |        "11  8"
306 |       ]
307 |      },
308 |      "execution_count": 7,
309 |      "metadata": {},
310 |      "output_type": "execute_result"
311 |     }
312 |    ],
313 |    "source": [
314 |     "# locは、2次元の場合、1次元目がindex値、2次元目がcolumns値で指定\n",
315 |     "# リストで指定した場合は、該当件数が1件でもpd.DataFrameになる\n",
316 |     "df.loc[:, ['a']]"
317 |    ]
318 |   },
319 |   {
320 |    "cell_type": "code",
321 |    "execution_count": 8,
322 |    "metadata": {},
323 |    "outputs": [
324 |     {
325 |      "data": {
326 |       "text/html": [
327 |        "<div>\n",
328 |        "<style scoped>\n",
329 |        "    .dataframe tbody tr th:only-of-type {\n",
330 |        "        vertical-align: middle;\n",
331 |        "    }\n",
332 |        "\n",
333 |        "    .dataframe tbody tr th {\n",
334 |        "        vertical-align: top;\n",
335 |        "    }\n",
336 |        "\n",
337 |        "    .dataframe thead th {\n",
338 |        "        text-align: right;\n",
339 |        "    }\n",
340 |        "</style>\n",
341 |        "<table border=\"1\" class=\"dataframe\">\n",
342 |        "  <thead>\n",
343 |        "    <tr style=\"text-align: right;\">\n",
344 |        "      <th></th>\n",
345 |        "      <th>b</th>\n",
346 |        "      <th>b</th>\n",
347 |        "    </tr>\n",
348 |        "  </thead>\n",
349 |        "  <tbody>\n",
350 |        "    <tr>\n",
351 |        "      <th>10</th>\n",
352 |        "      <td>1</td>\n",
353 |        "      <td>3</td>\n",
354 |        "    </tr>\n",
355 |        "    <tr>\n",
356 |        "      <th>11</th>\n",
357 |        "      <td>4</td>\n",
358 |        "      <td>6</td>\n",
359 |        "    </tr>\n",
360 |        "    <tr>\n",
361 |        "      <th>11</th>\n",
362 |        "      <td>7</td>\n",
363 |        "      <td>9</td>\n",
364 |        "    </tr>\n",
365 |        "  </tbody>\n",
366 |        "</table>\n",
367 |        "</div>"
368 |       ],
369 |       "text/plain": [
370 |        "    b  b\n",
371 |        "10  1  3\n",
372 |        "11  4  6\n",
373 |        "11  7  9"
374 |       ]
375 |      },
376 |      "execution_count": 8,
377 |      "metadata": {},
378 |      "output_type": "execute_result"
379 |     }
380 |    ],
381 |    "source": [
382 |     "# locは、2次元の場合、1次元目がindex値、2次元目がcolumns値で指定\n",
383 |     "# 指定する値が1つでも、該当件数が2件以上の場合pd.DataFrameで返される\n",
384 |     "\n"
385 |    ]
386 |   },
387 |   {
388 |    "cell_type": "code",
389 |    "execution_count": 17,
390 |    "metadata": {},
391 |    "outputs": [
392 |     {
393 |      "data": {
394 |       "text/plain": [
395 |        "2"
396 |       ]
397 |      },
398 |      "execution_count": 17,
399 |      "metadata": {},
400 |      "output_type": "execute_result"
401 |     }
402 |    ],
403 |    "source": [
404 |     "# locは、2次元の場合、1次元目がindex値、2次元目がcolumns値で指定\n",
405 |     "# 返り値が1つだと値の型に応じた返り値になる\n",
406 |     "df.loc[10, 'a']"
407 |    ]
408 |   },
409 |   {
410 |    "cell_type": "code",
411 |    "execution_count": 18,
412 |    "metadata": {},
413 |    "outputs": [
414 |     {
415 |      "data": {
416 |       "text/plain": [
417 |        "numpy.int64"
418 |       ]
419 |      },
420 |      "execution_count": 18,
421 |      "metadata": {},
422 |      "output_type": "execute_result"
423 |     }
424 |    ],
425 |    "source": [
426 |     "type(df.loc[10, 'a'])"
427 |    ]
428 |   },
429 |   {
430 |    "cell_type": "code",
431 |    "execution_count": 58,
432 |    "metadata": {},
433 |    "outputs": [
434 |     {
435 |      "data": {
436 |       "text/plain": [
437 |        "b    1\n",
438 |        "a    2\n",
439 |        "b    3\n",
440 |        "Name: 10, dtype: int64"
441 |       ]
442 |      },
443 |      "execution_count": 58,
444 |      "metadata": {},
445 |      "output_type": "execute_result"
446 |     }
447 |    ],
448 |    "source": [
449 |     "# ilocは、1次元の場合、行番号で指定\n",
450 |     "df.iloc[0]"
451 |    ]
452 |   },
453 |   {
454 |    "cell_type": "code",
455 |    "execution_count": 59,
456 |    "metadata": {},
457 |    "outputs": [
458 |     {
459 |      "data": {
460 |       "text/html": [
461 |        "<div>\n",
462 |        "<style scoped>\n",
463 |        "    .dataframe tbody tr th:only-of-type {\n",
464 |        "        vertical-align: middle;\n",
465 |        "    }\n",
466 |        "\n",
467 |        "    .dataframe tbody tr th {\n",
468 |        "        vertical-align: top;\n",
469 |        "    }\n",
470 |        "\n",
471 |        "    .dataframe thead th {\n",
472 |        "        text-align: right;\n",
473 |        "    }\n",
474 |        "</style>\n",
475 |        "<table border=\"1\" class=\"dataframe\">\n",
476 |        "  <thead>\n",
477 |        "    <tr style=\"text-align: right;\">\n",
478 |        "      <th></th>\n",
479 |        "      <th>b</th>\n",
480 |        "      <th>a</th>\n",
481 |        "      <th>b</th>\n",
482 |        "    </tr>\n",
483 |        "  </thead>\n",
484 |        "  <tbody>\n",
485 |        "    <tr>\n",
486 |        "      <th>10</th>\n",
487 |        "      <td>1</td>\n",
488 |        "      <td>2</td>\n",
489 |        "      <td>3</td>\n",
490 |        "    </tr>\n",
491 |        "  </tbody>\n",
492 |        "</table>\n",
493 |        "</div>"
494 |       ],
495 |       "text/plain": [
496 |        "    b  a  b\n",
497 |        "10  1  2  3"
498 |       ]
499 |      },
500 |      "execution_count": 59,
501 |      "metadata": {},
502 |      "output_type": "execute_result"
503 |     }
504 |    ],
505 |    "source": [
506 |     "df.iloc[[0]]"
507 |    ]
508 |   },
509 |   {
510 |    "cell_type": "code",
511 |    "execution_count": 60,
512 |    "metadata": {},
513 |    "outputs": [
514 |     {
515 |      "data": {
516 |       "text/plain": [
517 |        "10    1\n",
518 |        "11    4\n",
519 |        "11    7\n",
520 |        "Name: b, dtype: int64"
521 |       ]
522 |      },
523 |      "execution_count": 60,
524 |      "metadata": {},
525 |      "output_type": "execute_result"
526 |     }
527 |    ],
528 |    "source": [
529 |     "# ilocは、2次元の場合、行番号と列番号で指定\n",
530 |     "df.iloc[:, 0]"
531 |    ]
532 |   },
533 |   {
534 |    "cell_type": "code",
535 |    "execution_count": 61,
536 |    "metadata": {},
537 |    "outputs": [
538 |     {
539 |      "data": {
540 |       "text/html": [
541 |        "<div>\n",
542 |        "<style scoped>\n",
543 |        "    .dataframe tbody tr th:only-of-type {\n",
544 |        "        vertical-align: middle;\n",
545 |        "    }\n",
546 |        "\n",
547 |        "    .dataframe tbody tr th {\n",
548 |        "        vertical-align: top;\n",
549 |        "    }\n",
550 |        "\n",
551 |        "    .dataframe thead th {\n",
552 |        "        text-align: right;\n",
553 |        "    }\n",
554 |        "</style>\n",
555 |        "<table border=\"1\" class=\"dataframe\">\n",
556 |        "  <thead>\n",
557 |        "    <tr style=\"text-align: right;\">\n",
558 |        "      <th></th>\n",
559 |        "      <th>b</th>\n",
560 |        "    </tr>\n",
561 |        "  </thead>\n",
562 |        "  <tbody>\n",
563 |        "    <tr>\n",
564 |        "      <th>10</th>\n",
565 |        "      <td>1</td>\n",
566 |        "    </tr>\n",
567 |        "    <tr>\n",
568 |        "      <th>11</th>\n",
569 |        "      <td>4</td>\n",
570 |        "    </tr>\n",
571 |        "    <tr>\n",
572 |        "      <th>11</th>\n",
573 |        "      <td>7</td>\n",
574 |        "    </tr>\n",
575 |        "  </tbody>\n",
576 |        "</table>\n",
577 |        "</div>"
578 |       ],
579 |       "text/plain": [
580 |        "    b\n",
581 |        "10  1\n",
582 |        "11  4\n",
583 |        "11  7"
584 |       ]
585 |      },
586 |      "execution_count": 61,
587 |      "metadata": {},
588 |      "output_type": "execute_result"
589 |     }
590 |    ],
591 |    "source": [
592 |     "df.iloc[:, [0]]"
593 |    ]
594 |   },
595 |   {
596 |    "cell_type": "code",
597 |    "execution_count": 62,
598 |    "metadata": {},
599 |    "outputs": [
600 |     {
601 |      "data": {
602 |       "text/plain": [
603 |        "0"
604 |       ]
605 |      },
606 |      "execution_count": 62,
607 |      "metadata": {},
608 |      "output_type": "execute_result"
609 |     }
610 |    ],
611 |    "source": [
612 |     "# get_loc関数は、pd.Indexクラスが持っている関数\n",
613 |     "# Indexの値を検索して該当したインデックス番号を返す関数\n",
614 |     "\n",
615 |     "df.index.get_loc(10)"
616 |    ]
617 |   },
618 |   {
619 |    "cell_type": "code",
620 |    "execution_count": 63,
621 |    "metadata": {},
622 |    "outputs": [
623 |     {
624 |      "data": {
625 |       "text/plain": [
626 |        "slice(1, 3, None)"
627 |       ]
628 |      },
629 |      "execution_count": 63,
630 |      "metadata": {},
631 |      "output_type": "execute_result"
632 |     }
633 |    ],
634 |    "source": [
635 |     "# 連続だとsliceになる\n",
636 |     "df.index.get_loc(11)"
637 |    ]
638 |   },
639 |   {
640 |    "cell_type": "code",
641 |    "execution_count": 64,
642 |    "metadata": {},
643 |    "outputs": [
644 |     {
645 |      "data": {
646 |       "text/plain": [
647 |        "1"
648 |       ]
649 |      },
650 |      "execution_count": 64,
651 |      "metadata": {},
652 |      "output_type": "execute_result"
653 |     }
654 |    ],
655 |    "source": [
656 |     "df.columns.get_loc('a')"
657 |    ]
658 |   },
659 |   {
660 |    "cell_type": "code",
661 |    "execution_count": 65,
662 |    "metadata": {},
663 |    "outputs": [
664 |     {
665 |      "data": {
666 |       "text/plain": [
667 |        "array([ True, False,  True])"
668 |       ]
669 |      },
670 |      "execution_count": 65,
671 |      "metadata": {},
672 |      "output_type": "execute_result"
673 |     }
674 |    ],
675 |    "source": [
676 |     "# 非連続だとarrayになる\n",
677 |     "df.columns.get_loc('b')"
678 |    ]
679 |   },
680 |   {
681 |    "cell_type": "code",
682 |    "execution_count": null,
683 |    "metadata": {},
684 |    "outputs": [],
685 |    "source": []
686 |   }
687 |  ],
688 |  "metadata": {
689 |   "kernelspec": {
690 |    "display_name": "Python 3",
691 |    "language": "python",
692 |    "name": "python3"
693 |   },
694 |   "language_info": {
695 |    "codemirror_mode": {
696 |     "name": "ipython",
697 |     "version": 3
698 |    },
699 |    "file_extension": ".py",
700 |    "mimetype": "text/x-python",
701 |    "name": "python",
702 |    "nbconvert_exporter": "python",
703 |    "pygments_lexer": "ipython3",
704 |    "version": "3.7.0"
705 |   }
706 |  },
707 |  "nbformat": 4,
708 |  "nbformat_minor": 2
709 | }
710 | 


--------------------------------------------------------------------------------
/7-1.ipynb:
--------------------------------------------------------------------------------
  1 | {
  2 |  "cells": [
  3 |   {
  4 |    "cell_type": "markdown",
  5 |    "metadata": {},
  6 |    "source": [
  7 |     "## 例7-1. k-meansの適用例を実行するコード"
  8 |    ]
  9 |   },
 10 |   {
 11 |    "cell_type": "code",
 12 |    "execution_count": 1,
 13 |    "metadata": {},
 14 |    "outputs": [
 15 |     {
 16 |      "data": {
 17 |       "text/plain": [
 18 |        "(-0.07086790846959401,\n",
 19 |        " 1.0713297683670353,\n",
 20 |        " -0.08072533667047069,\n",
 21 |        " 1.082584430361733)"
 22 |       ]
 23 |      },
 24 |      "execution_count": 1,
 25 |      "metadata": {},
 26 |      "output_type": "execute_result"
 27 |     },
 28 |     {
 29 |      "data": {
 30 |       "image/png": "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\n",
 31 |       "text/plain": [
 32 |        "<Figure size 432x288 with 4 Axes>"
 33 |       ]
 34 |      },
 35 |      "metadata": {
 36 |       "needs_background": "light"
 37 |      },
 38 |      "output_type": "display_data"
 39 |     }
 40 |    ],
 41 |    "source": [
 42 |     "%matplotlib inline\n",
 43 |     "import numpy as np\n",
 44 |     "from sklearn.cluster import KMeans\n",
 45 |     "from sklearn.datasets import make_blobs\n",
 46 |     "import matplotlib.pyplot as plt\n",
 47 |     "\n",
 48 |     "n_data = 1000\n",
 49 |     "seed = 1\n",
 50 |     "n_centers = 4\n",
 51 |     "\n",
 52 |     "# 4つの二変量正規分布に従うデータを生成し、k-meansを実行する\n",
 53 |     "blobs, blob_labels = make_blobs(n_samples=n_data, n_features=2, centers=n_centers, random_state=seed)\n",
 54 |     "clusters_blob = KMeans(n_clusters=n_centers, random_state=seed).fit_predict(blobs)\n",
 55 |     "\n",
 56 |     "#　2次元の一様分布に従うデータを生成し、k-meansを実行する\n",
 57 |     "uniform = np.random.rand(n_data, 2)\n",
 58 |     "clusters_uniform = KMeans(n_clusters=n_centers, random_state=seed).fit_predict(uniform)\n",
 59 |     "\n",
 60 |     "# 結果を可視化するためのMatplotlibのおまじない\n",
 61 |     "figure = plt.figure()\n",
 62 |     "plt.subplot(221)\n",
 63 |     "plt.scatter(blobs[:, 0], blobs[:, 1], c=blob_labels, cmap='gist_rainbow')\n",
 64 |     "plt.title('(a) Four randomly generated blobs', fontsize=14)\n",
 65 |     "plt.axis('off')\n",
 66 |     "\n",
 67 |     "plt.subplot(222)\n",
 68 |     "plt.scatter(blobs[:, 0], blobs[:, 1], c=clusters_blob, cmap='gist_rainbow')\n",
 69 |     "plt.title('(b) Clusters found via K-means', fontsize=14)\n",
 70 |     "plt.axis('off')\n",
 71 |     "\n",
 72 |     "plt.subplot(223)\n",
 73 |     "plt.scatter(uniform[:, 0], uniform[:, 1])\n",
 74 |     "plt.title('(c) 1000 randomly generated points', fontsize=14)\n",
 75 |     "plt.axis('off')\n",
 76 |     "\n",
 77 |     "plt.subplot(224)\n",
 78 |     "plt.scatter(uniform[:, 0], uniform[:, 1], c=clusters_uniform, cmap='gist_rainbow')\n",
 79 |     "plt.title('(d) Clusters found via K-means', fontsize=14)\n",
 80 |     "plt.axis('off')"
 81 |    ]
 82 |   },
 83 |   {
 84 |    "cell_type": "code",
 85 |    "execution_count": null,
 86 |    "metadata": {},
 87 |    "outputs": [],
 88 |    "source": []
 89 |   }
 90 |  ],
 91 |  "metadata": {
 92 |   "kernelspec": {
 93 |    "display_name": "Python 3",
 94 |    "language": "python",
 95 |    "name": "python3"
 96 |   },
 97 |   "language_info": {
 98 |    "codemirror_mode": {
 99 |     "name": "ipython",
100 |     "version": 3
101 |    },
102 |    "file_extension": ".py",
103 |    "mimetype": "text/x-python",
104 |    "name": "python",
105 |    "nbconvert_exporter": "python",
106 |    "pygments_lexer": "ipython3",
107 |    "version": "3.7.0"
108 |   }
109 |  },
110 |  "nbformat": 4,
111 |  "nbformat_minor": 2
112 | }
113 | 


--------------------------------------------------------------------------------
/5-6.ipynb:
--------------------------------------------------------------------------------
   1 | {
   2 |  "cells": [
   3 |   {
   4 |    "cell_type": "markdown",
   5 |    "metadata": {},
   6 |    "source": [
   7 |     "## 例5-6. ビンカウンティングの例"
   8 |    ]
   9 |   },
  10 |   {
  11 |    "cell_type": "code",
  12 |    "execution_count": 10,
  13 |    "metadata": {},
  14 |    "outputs": [
  15 |     {
  16 |      "data": {
  17 |       "text/plain": [
  18 |        "(8208, 24)"
  19 |       ]
  20 |      },
  21 |      "execution_count": 10,
  22 |      "metadata": {},
  23 |      "output_type": "execute_result"
  24 |     }
  25 |    ],
  26 |    "source": [
  27 |     "import pandas as pd\n",
  28 |     "# train_subsetを読み込み(サンプルコードの対象のデータ件数が、8208件です。)\n",
  29 |     "df = pd.read_csv('data/avazu/train_subset.csv')\n",
  30 |     "\n",
  31 |     "df.shape"
  32 |    ]
  33 |   },
  34 |   {
  35 |    "cell_type": "code",
  36 |    "execution_count": 11,
  37 |    "metadata": {},
  38 |    "outputs": [
  39 |     {
  40 |      "data": {
  41 |       "text/plain": [
  42 |        "906"
  43 |       ]
  44 |      },
  45 |      "execution_count": 11,
  46 |      "metadata": {},
  47 |      "output_type": "execute_result"
  48 |     }
  49 |    ],
  50 |    "source": [
  51 |     "\n",
  52 |     "# device_idが何種類あるか計算\n",
  53 |     "len(df['device_id'].unique())"
  54 |    ]
  55 |   },
  56 |   {
  57 |    "cell_type": "code",
  58 |    "execution_count": 12,
  59 |    "metadata": {},
  60 |    "outputs": [
  61 |     {
  62 |      "data": {
  63 |       "text/html": [
  64 |        "<div>\n",
  65 |        "<style scoped>\n",
  66 |        "    .dataframe tbody tr th:only-of-type {\n",
  67 |        "        vertical-align: middle;\n",
  68 |        "    }\n",
  69 |        "\n",
  70 |        "    .dataframe tbody tr th {\n",
  71 |        "        vertical-align: top;\n",
  72 |        "    }\n",
  73 |        "\n",
  74 |        "    .dataframe thead th {\n",
  75 |        "        text-align: right;\n",
  76 |        "    }\n",
  77 |        "</style>\n",
  78 |        "<table border=\"1\" class=\"dataframe\">\n",
  79 |        "  <thead>\n",
  80 |        "    <tr style=\"text-align: right;\">\n",
  81 |        "      <th></th>\n",
  82 |        "      <th>clicks</th>\n",
  83 |        "      <th>no_clicks</th>\n",
  84 |        "      <th>total_clicks</th>\n",
  85 |        "      <th>N+</th>\n",
  86 |        "      <th>N-</th>\n",
  87 |        "      <th>log_N+</th>\n",
  88 |        "    </tr>\n",
  89 |        "  </thead>\n",
  90 |        "  <tbody>\n",
  91 |        "    <tr>\n",
  92 |        "      <th>a99f214a</th>\n",
  93 |        "      <td>1279</td>\n",
  94 |        "      <td>5878</td>\n",
  95 |        "      <td>7157</td>\n",
  96 |        "      <td>0.178706</td>\n",
  97 |        "      <td>0.821294</td>\n",
  98 |        "      <td>0.217591</td>\n",
  99 |        "    </tr>\n",
 100 |        "    <tr>\n",
 101 |        "      <th>c357dbff</th>\n",
 102 |        "      <td>2</td>\n",
 103 |        "      <td>12</td>\n",
 104 |        "      <td>14</td>\n",
 105 |        "      <td>0.142857</td>\n",
 106 |        "      <td>0.857143</td>\n",
 107 |        "      <td>0.166667</td>\n",
 108 |        "    </tr>\n",
 109 |        "    <tr>\n",
 110 |        "      <th>25635c83</th>\n",
 111 |        "      <td>2</td>\n",
 112 |        "      <td>0</td>\n",
 113 |        "      <td>2</td>\n",
 114 |        "      <td>1.000000</td>\n",
 115 |        "      <td>0.000000</td>\n",
 116 |        "      <td>inf</td>\n",
 117 |        "    </tr>\n",
 118 |        "    <tr>\n",
 119 |        "      <th>e62f1261</th>\n",
 120 |        "      <td>2</td>\n",
 121 |        "      <td>1</td>\n",
 122 |        "      <td>3</td>\n",
 123 |        "      <td>0.666667</td>\n",
 124 |        "      <td>0.333333</td>\n",
 125 |        "      <td>2.000000</td>\n",
 126 |        "    </tr>\n",
 127 |        "    <tr>\n",
 128 |        "      <th>135f7d9a</th>\n",
 129 |        "      <td>2</td>\n",
 130 |        "      <td>0</td>\n",
 131 |        "      <td>2</td>\n",
 132 |        "      <td>1.000000</td>\n",
 133 |        "      <td>0.000000</td>\n",
 134 |        "      <td>inf</td>\n",
 135 |        "    </tr>\n",
 136 |        "    <tr>\n",
 137 |        "      <th>9af87478</th>\n",
 138 |        "      <td>2</td>\n",
 139 |        "      <td>0</td>\n",
 140 |        "      <td>2</td>\n",
 141 |        "      <td>1.000000</td>\n",
 142 |        "      <td>0.000000</td>\n",
 143 |        "      <td>inf</td>\n",
 144 |        "    </tr>\n",
 145 |        "    <tr>\n",
 146 |        "      <th>77cf1a27</th>\n",
 147 |        "      <td>1</td>\n",
 148 |        "      <td>0</td>\n",
 149 |        "      <td>1</td>\n",
 150 |        "      <td>1.000000</td>\n",
 151 |        "      <td>0.000000</td>\n",
 152 |        "      <td>inf</td>\n",
 153 |        "    </tr>\n",
 154 |        "    <tr>\n",
 155 |        "      <th>d62216cc</th>\n",
 156 |        "      <td>1</td>\n",
 157 |        "      <td>0</td>\n",
 158 |        "      <td>1</td>\n",
 159 |        "      <td>1.000000</td>\n",
 160 |        "      <td>0.000000</td>\n",
 161 |        "      <td>inf</td>\n",
 162 |        "    </tr>\n",
 163 |        "    <tr>\n",
 164 |        "      <th>fcc5c7c0</th>\n",
 165 |        "      <td>1</td>\n",
 166 |        "      <td>0</td>\n",
 167 |        "      <td>1</td>\n",
 168 |        "      <td>1.000000</td>\n",
 169 |        "      <td>0.000000</td>\n",
 170 |        "      <td>inf</td>\n",
 171 |        "    </tr>\n",
 172 |        "    <tr>\n",
 173 |        "      <th>7181509e</th>\n",
 174 |        "      <td>1</td>\n",
 175 |        "      <td>0</td>\n",
 176 |        "      <td>1</td>\n",
 177 |        "      <td>1.000000</td>\n",
 178 |        "      <td>0.000000</td>\n",
 179 |        "      <td>inf</td>\n",
 180 |        "    </tr>\n",
 181 |        "    <tr>\n",
 182 |        "      <th>2a32a3ca</th>\n",
 183 |        "      <td>1</td>\n",
 184 |        "      <td>0</td>\n",
 185 |        "      <td>1</td>\n",
 186 |        "      <td>1.000000</td>\n",
 187 |        "      <td>0.000000</td>\n",
 188 |        "      <td>inf</td>\n",
 189 |        "    </tr>\n",
 190 |        "    <tr>\n",
 191 |        "      <th>2ad16ba3</th>\n",
 192 |        "      <td>1</td>\n",
 193 |        "      <td>0</td>\n",
 194 |        "      <td>1</td>\n",
 195 |        "      <td>1.000000</td>\n",
 196 |        "      <td>0.000000</td>\n",
 197 |        "      <td>inf</td>\n",
 198 |        "    </tr>\n",
 199 |        "    <tr>\n",
 200 |        "      <th>938f494b</th>\n",
 201 |        "      <td>1</td>\n",
 202 |        "      <td>0</td>\n",
 203 |        "      <td>1</td>\n",
 204 |        "      <td>1.000000</td>\n",
 205 |        "      <td>0.000000</td>\n",
 206 |        "      <td>inf</td>\n",
 207 |        "    </tr>\n",
 208 |        "    <tr>\n",
 209 |        "      <th>24dbae83</th>\n",
 210 |        "      <td>1</td>\n",
 211 |        "      <td>0</td>\n",
 212 |        "      <td>1</td>\n",
 213 |        "      <td>1.000000</td>\n",
 214 |        "      <td>0.000000</td>\n",
 215 |        "      <td>inf</td>\n",
 216 |        "    </tr>\n",
 217 |        "    <tr>\n",
 218 |        "      <th>ca9b95aa</th>\n",
 219 |        "      <td>1</td>\n",
 220 |        "      <td>0</td>\n",
 221 |        "      <td>1</td>\n",
 222 |        "      <td>1.000000</td>\n",
 223 |        "      <td>0.000000</td>\n",
 224 |        "      <td>inf</td>\n",
 225 |        "    </tr>\n",
 226 |        "    <tr>\n",
 227 |        "      <th>7c369899</th>\n",
 228 |        "      <td>1</td>\n",
 229 |        "      <td>0</td>\n",
 230 |        "      <td>1</td>\n",
 231 |        "      <td>1.000000</td>\n",
 232 |        "      <td>0.000000</td>\n",
 233 |        "      <td>inf</td>\n",
 234 |        "    </tr>\n",
 235 |        "    <tr>\n",
 236 |        "      <th>3bf8c26c</th>\n",
 237 |        "      <td>1</td>\n",
 238 |        "      <td>0</td>\n",
 239 |        "      <td>1</td>\n",
 240 |        "      <td>1.000000</td>\n",
 241 |        "      <td>0.000000</td>\n",
 242 |        "      <td>inf</td>\n",
 243 |        "    </tr>\n",
 244 |        "    <tr>\n",
 245 |        "      <th>45378128</th>\n",
 246 |        "      <td>1</td>\n",
 247 |        "      <td>0</td>\n",
 248 |        "      <td>1</td>\n",
 249 |        "      <td>1.000000</td>\n",
 250 |        "      <td>0.000000</td>\n",
 251 |        "      <td>inf</td>\n",
 252 |        "    </tr>\n",
 253 |        "    <tr>\n",
 254 |        "      <th>023ca1f9</th>\n",
 255 |        "      <td>1</td>\n",
 256 |        "      <td>0</td>\n",
 257 |        "      <td>1</td>\n",
 258 |        "      <td>1.000000</td>\n",
 259 |        "      <td>0.000000</td>\n",
 260 |        "      <td>inf</td>\n",
 261 |        "    </tr>\n",
 262 |        "    <tr>\n",
 263 |        "      <th>3b9ab74d</th>\n",
 264 |        "      <td>1</td>\n",
 265 |        "      <td>0</td>\n",
 266 |        "      <td>1</td>\n",
 267 |        "      <td>1.000000</td>\n",
 268 |        "      <td>0.000000</td>\n",
 269 |        "      <td>inf</td>\n",
 270 |        "    </tr>\n",
 271 |        "    <tr>\n",
 272 |        "      <th>9eb9a972</th>\n",
 273 |        "      <td>1</td>\n",
 274 |        "      <td>0</td>\n",
 275 |        "      <td>1</td>\n",
 276 |        "      <td>1.000000</td>\n",
 277 |        "      <td>0.000000</td>\n",
 278 |        "      <td>inf</td>\n",
 279 |        "    </tr>\n",
 280 |        "    <tr>\n",
 281 |        "      <th>59bcd1ae</th>\n",
 282 |        "      <td>1</td>\n",
 283 |        "      <td>0</td>\n",
 284 |        "      <td>1</td>\n",
 285 |        "      <td>1.000000</td>\n",
 286 |        "      <td>0.000000</td>\n",
 287 |        "      <td>inf</td>\n",
 288 |        "    </tr>\n",
 289 |        "    <tr>\n",
 290 |        "      <th>7f8c00b4</th>\n",
 291 |        "      <td>1</td>\n",
 292 |        "      <td>0</td>\n",
 293 |        "      <td>1</td>\n",
 294 |        "      <td>1.000000</td>\n",
 295 |        "      <td>0.000000</td>\n",
 296 |        "      <td>inf</td>\n",
 297 |        "    </tr>\n",
 298 |        "    <tr>\n",
 299 |        "      <th>8d61d7eb</th>\n",
 300 |        "      <td>1</td>\n",
 301 |        "      <td>0</td>\n",
 302 |        "      <td>1</td>\n",
 303 |        "      <td>1.000000</td>\n",
 304 |        "      <td>0.000000</td>\n",
 305 |        "      <td>inf</td>\n",
 306 |        "    </tr>\n",
 307 |        "    <tr>\n",
 308 |        "      <th>b441c41f</th>\n",
 309 |        "      <td>1</td>\n",
 310 |        "      <td>0</td>\n",
 311 |        "      <td>1</td>\n",
 312 |        "      <td>1.000000</td>\n",
 313 |        "      <td>0.000000</td>\n",
 314 |        "      <td>inf</td>\n",
 315 |        "    </tr>\n",
 316 |        "    <tr>\n",
 317 |        "      <th>e317838f</th>\n",
 318 |        "      <td>1</td>\n",
 319 |        "      <td>0</td>\n",
 320 |        "      <td>1</td>\n",
 321 |        "      <td>1.000000</td>\n",
 322 |        "      <td>0.000000</td>\n",
 323 |        "      <td>inf</td>\n",
 324 |        "    </tr>\n",
 325 |        "    <tr>\n",
 326 |        "      <th>2e7d4b65</th>\n",
 327 |        "      <td>1</td>\n",
 328 |        "      <td>0</td>\n",
 329 |        "      <td>1</td>\n",
 330 |        "      <td>1.000000</td>\n",
 331 |        "      <td>0.000000</td>\n",
 332 |        "      <td>inf</td>\n",
 333 |        "    </tr>\n",
 334 |        "    <tr>\n",
 335 |        "      <th>70d97ece</th>\n",
 336 |        "      <td>1</td>\n",
 337 |        "      <td>0</td>\n",
 338 |        "      <td>1</td>\n",
 339 |        "      <td>1.000000</td>\n",
 340 |        "      <td>0.000000</td>\n",
 341 |        "      <td>inf</td>\n",
 342 |        "    </tr>\n",
 343 |        "    <tr>\n",
 344 |        "      <th>9809e6c9</th>\n",
 345 |        "      <td>1</td>\n",
 346 |        "      <td>0</td>\n",
 347 |        "      <td>1</td>\n",
 348 |        "      <td>1.000000</td>\n",
 349 |        "      <td>0.000000</td>\n",
 350 |        "      <td>inf</td>\n",
 351 |        "    </tr>\n",
 352 |        "    <tr>\n",
 353 |        "      <th>cb73ba55</th>\n",
 354 |        "      <td>1</td>\n",
 355 |        "      <td>0</td>\n",
 356 |        "      <td>1</td>\n",
 357 |        "      <td>1.000000</td>\n",
 358 |        "      <td>0.000000</td>\n",
 359 |        "      <td>inf</td>\n",
 360 |        "    </tr>\n",
 361 |        "    <tr>\n",
 362 |        "      <th>...</th>\n",
 363 |        "      <td>...</td>\n",
 364 |        "      <td>...</td>\n",
 365 |        "      <td>...</td>\n",
 366 |        "      <td>...</td>\n",
 367 |        "      <td>...</td>\n",
 368 |        "      <td>...</td>\n",
 369 |        "    </tr>\n",
 370 |        "    <tr>\n",
 371 |        "      <th>0c3bbac0</th>\n",
 372 |        "      <td>0</td>\n",
 373 |        "      <td>1</td>\n",
 374 |        "      <td>1</td>\n",
 375 |        "      <td>0.000000</td>\n",
 376 |        "      <td>1.000000</td>\n",
 377 |        "      <td>0.000000</td>\n",
 378 |        "    </tr>\n",
 379 |        "    <tr>\n",
 380 |        "      <th>34c9f908</th>\n",
 381 |        "      <td>0</td>\n",
 382 |        "      <td>1</td>\n",
 383 |        "      <td>1</td>\n",
 384 |        "      <td>0.000000</td>\n",
 385 |        "      <td>1.000000</td>\n",
 386 |        "      <td>0.000000</td>\n",
 387 |        "    </tr>\n",
 388 |        "    <tr>\n",
 389 |        "      <th>41a1ae5f</th>\n",
 390 |        "      <td>0</td>\n",
 391 |        "      <td>1</td>\n",
 392 |        "      <td>1</td>\n",
 393 |        "      <td>0.000000</td>\n",
 394 |        "      <td>1.000000</td>\n",
 395 |        "      <td>0.000000</td>\n",
 396 |        "    </tr>\n",
 397 |        "    <tr>\n",
 398 |        "      <th>5d03585e</th>\n",
 399 |        "      <td>0</td>\n",
 400 |        "      <td>1</td>\n",
 401 |        "      <td>1</td>\n",
 402 |        "      <td>0.000000</td>\n",
 403 |        "      <td>1.000000</td>\n",
 404 |        "      <td>0.000000</td>\n",
 405 |        "    </tr>\n",
 406 |        "    <tr>\n",
 407 |        "      <th>7d242dfd</th>\n",
 408 |        "      <td>0</td>\n",
 409 |        "      <td>1</td>\n",
 410 |        "      <td>1</td>\n",
 411 |        "      <td>0.000000</td>\n",
 412 |        "      <td>1.000000</td>\n",
 413 |        "      <td>0.000000</td>\n",
 414 |        "    </tr>\n",
 415 |        "    <tr>\n",
 416 |        "      <th>9ffa0563</th>\n",
 417 |        "      <td>0</td>\n",
 418 |        "      <td>1</td>\n",
 419 |        "      <td>1</td>\n",
 420 |        "      <td>0.000000</td>\n",
 421 |        "      <td>1.000000</td>\n",
 422 |        "      <td>0.000000</td>\n",
 423 |        "    </tr>\n",
 424 |        "    <tr>\n",
 425 |        "      <th>2fd58990</th>\n",
 426 |        "      <td>0</td>\n",
 427 |        "      <td>1</td>\n",
 428 |        "      <td>1</td>\n",
 429 |        "      <td>0.000000</td>\n",
 430 |        "      <td>1.000000</td>\n",
 431 |        "      <td>0.000000</td>\n",
 432 |        "    </tr>\n",
 433 |        "    <tr>\n",
 434 |        "      <th>01e47a3d</th>\n",
 435 |        "      <td>0</td>\n",
 436 |        "      <td>1</td>\n",
 437 |        "      <td>1</td>\n",
 438 |        "      <td>0.000000</td>\n",
 439 |        "      <td>1.000000</td>\n",
 440 |        "      <td>0.000000</td>\n",
 441 |        "    </tr>\n",
 442 |        "    <tr>\n",
 443 |        "      <th>bcd5195e</th>\n",
 444 |        "      <td>0</td>\n",
 445 |        "      <td>1</td>\n",
 446 |        "      <td>1</td>\n",
 447 |        "      <td>0.000000</td>\n",
 448 |        "      <td>1.000000</td>\n",
 449 |        "      <td>0.000000</td>\n",
 450 |        "    </tr>\n",
 451 |        "    <tr>\n",
 452 |        "      <th>a318236b</th>\n",
 453 |        "      <td>0</td>\n",
 454 |        "      <td>1</td>\n",
 455 |        "      <td>1</td>\n",
 456 |        "      <td>0.000000</td>\n",
 457 |        "      <td>1.000000</td>\n",
 458 |        "      <td>0.000000</td>\n",
 459 |        "    </tr>\n",
 460 |        "    <tr>\n",
 461 |        "      <th>1168ce02</th>\n",
 462 |        "      <td>0</td>\n",
 463 |        "      <td>1</td>\n",
 464 |        "      <td>1</td>\n",
 465 |        "      <td>0.000000</td>\n",
 466 |        "      <td>1.000000</td>\n",
 467 |        "      <td>0.000000</td>\n",
 468 |        "    </tr>\n",
 469 |        "    <tr>\n",
 470 |        "      <th>83c34e93</th>\n",
 471 |        "      <td>0</td>\n",
 472 |        "      <td>1</td>\n",
 473 |        "      <td>1</td>\n",
 474 |        "      <td>0.000000</td>\n",
 475 |        "      <td>1.000000</td>\n",
 476 |        "      <td>0.000000</td>\n",
 477 |        "    </tr>\n",
 478 |        "    <tr>\n",
 479 |        "      <th>c8c31032</th>\n",
 480 |        "      <td>0</td>\n",
 481 |        "      <td>1</td>\n",
 482 |        "      <td>1</td>\n",
 483 |        "      <td>0.000000</td>\n",
 484 |        "      <td>1.000000</td>\n",
 485 |        "      <td>0.000000</td>\n",
 486 |        "    </tr>\n",
 487 |        "    <tr>\n",
 488 |        "      <th>004270bf</th>\n",
 489 |        "      <td>0</td>\n",
 490 |        "      <td>1</td>\n",
 491 |        "      <td>1</td>\n",
 492 |        "      <td>0.000000</td>\n",
 493 |        "      <td>1.000000</td>\n",
 494 |        "      <td>0.000000</td>\n",
 495 |        "    </tr>\n",
 496 |        "    <tr>\n",
 497 |        "      <th>e6d0facc</th>\n",
 498 |        "      <td>0</td>\n",
 499 |        "      <td>1</td>\n",
 500 |        "      <td>1</td>\n",
 501 |        "      <td>0.000000</td>\n",
 502 |        "      <td>1.000000</td>\n",
 503 |        "      <td>0.000000</td>\n",
 504 |        "    </tr>\n",
 505 |        "    <tr>\n",
 506 |        "      <th>844524dc</th>\n",
 507 |        "      <td>0</td>\n",
 508 |        "      <td>1</td>\n",
 509 |        "      <td>1</td>\n",
 510 |        "      <td>0.000000</td>\n",
 511 |        "      <td>1.000000</td>\n",
 512 |        "      <td>0.000000</td>\n",
 513 |        "    </tr>\n",
 514 |        "    <tr>\n",
 515 |        "      <th>c90a30a1</th>\n",
 516 |        "      <td>0</td>\n",
 517 |        "      <td>1</td>\n",
 518 |        "      <td>1</td>\n",
 519 |        "      <td>0.000000</td>\n",
 520 |        "      <td>1.000000</td>\n",
 521 |        "      <td>0.000000</td>\n",
 522 |        "    </tr>\n",
 523 |        "    <tr>\n",
 524 |        "      <th>5015495b</th>\n",
 525 |        "      <td>0</td>\n",
 526 |        "      <td>1</td>\n",
 527 |        "      <td>1</td>\n",
 528 |        "      <td>0.000000</td>\n",
 529 |        "      <td>1.000000</td>\n",
 530 |        "      <td>0.000000</td>\n",
 531 |        "    </tr>\n",
 532 |        "    <tr>\n",
 533 |        "      <th>a12e8a45</th>\n",
 534 |        "      <td>0</td>\n",
 535 |        "      <td>1</td>\n",
 536 |        "      <td>1</td>\n",
 537 |        "      <td>0.000000</td>\n",
 538 |        "      <td>1.000000</td>\n",
 539 |        "      <td>0.000000</td>\n",
 540 |        "    </tr>\n",
 541 |        "    <tr>\n",
 542 |        "      <th>7b5c2c3b</th>\n",
 543 |        "      <td>0</td>\n",
 544 |        "      <td>1</td>\n",
 545 |        "      <td>1</td>\n",
 546 |        "      <td>0.000000</td>\n",
 547 |        "      <td>1.000000</td>\n",
 548 |        "      <td>0.000000</td>\n",
 549 |        "    </tr>\n",
 550 |        "    <tr>\n",
 551 |        "      <th>4a6b5af3</th>\n",
 552 |        "      <td>0</td>\n",
 553 |        "      <td>1</td>\n",
 554 |        "      <td>1</td>\n",
 555 |        "      <td>0.000000</td>\n",
 556 |        "      <td>1.000000</td>\n",
 557 |        "      <td>0.000000</td>\n",
 558 |        "    </tr>\n",
 559 |        "    <tr>\n",
 560 |        "      <th>2c8ae68c</th>\n",
 561 |        "      <td>0</td>\n",
 562 |        "      <td>1</td>\n",
 563 |        "      <td>1</td>\n",
 564 |        "      <td>0.000000</td>\n",
 565 |        "      <td>1.000000</td>\n",
 566 |        "      <td>0.000000</td>\n",
 567 |        "    </tr>\n",
 568 |        "    <tr>\n",
 569 |        "      <th>84565c92</th>\n",
 570 |        "      <td>0</td>\n",
 571 |        "      <td>1</td>\n",
 572 |        "      <td>1</td>\n",
 573 |        "      <td>0.000000</td>\n",
 574 |        "      <td>1.000000</td>\n",
 575 |        "      <td>0.000000</td>\n",
 576 |        "    </tr>\n",
 577 |        "    <tr>\n",
 578 |        "      <th>2a6fe2a5</th>\n",
 579 |        "      <td>0</td>\n",
 580 |        "      <td>1</td>\n",
 581 |        "      <td>1</td>\n",
 582 |        "      <td>0.000000</td>\n",
 583 |        "      <td>1.000000</td>\n",
 584 |        "      <td>0.000000</td>\n",
 585 |        "    </tr>\n",
 586 |        "    <tr>\n",
 587 |        "      <th>e0f70006</th>\n",
 588 |        "      <td>0</td>\n",
 589 |        "      <td>1</td>\n",
 590 |        "      <td>1</td>\n",
 591 |        "      <td>0.000000</td>\n",
 592 |        "      <td>1.000000</td>\n",
 593 |        "      <td>0.000000</td>\n",
 594 |        "    </tr>\n",
 595 |        "    <tr>\n",
 596 |        "      <th>cef4c8cc</th>\n",
 597 |        "      <td>0</td>\n",
 598 |        "      <td>1</td>\n",
 599 |        "      <td>1</td>\n",
 600 |        "      <td>0.000000</td>\n",
 601 |        "      <td>1.000000</td>\n",
 602 |        "      <td>0.000000</td>\n",
 603 |        "    </tr>\n",
 604 |        "    <tr>\n",
 605 |        "      <th>7f4b1f1e</th>\n",
 606 |        "      <td>0</td>\n",
 607 |        "      <td>1</td>\n",
 608 |        "      <td>1</td>\n",
 609 |        "      <td>0.000000</td>\n",
 610 |        "      <td>1.000000</td>\n",
 611 |        "      <td>0.000000</td>\n",
 612 |        "    </tr>\n",
 613 |        "    <tr>\n",
 614 |        "      <th>7efe14f0</th>\n",
 615 |        "      <td>0</td>\n",
 616 |        "      <td>1</td>\n",
 617 |        "      <td>1</td>\n",
 618 |        "      <td>0.000000</td>\n",
 619 |        "      <td>1.000000</td>\n",
 620 |        "      <td>0.000000</td>\n",
 621 |        "    </tr>\n",
 622 |        "    <tr>\n",
 623 |        "      <th>02b99e77</th>\n",
 624 |        "      <td>0</td>\n",
 625 |        "      <td>1</td>\n",
 626 |        "      <td>1</td>\n",
 627 |        "      <td>0.000000</td>\n",
 628 |        "      <td>1.000000</td>\n",
 629 |        "      <td>0.000000</td>\n",
 630 |        "    </tr>\n",
 631 |        "    <tr>\n",
 632 |        "      <th>cbb50c1c</th>\n",
 633 |        "      <td>0</td>\n",
 634 |        "      <td>1</td>\n",
 635 |        "      <td>1</td>\n",
 636 |        "      <td>0.000000</td>\n",
 637 |        "      <td>1.000000</td>\n",
 638 |        "      <td>0.000000</td>\n",
 639 |        "    </tr>\n",
 640 |        "  </tbody>\n",
 641 |        "</table>\n",
 642 |        "<p>906 rows × 6 columns</p>\n",
 643 |        "</div>"
 644 |       ],
 645 |       "text/plain": [
 646 |        "         clicks no_clicks  total_clicks        N+        N-    log_N+\n",
 647 |        "a99f214a   1279      5878          7157  0.178706  0.821294  0.217591\n",
 648 |        "c357dbff      2        12            14  0.142857  0.857143  0.166667\n",
 649 |        "25635c83      2         0             2  1.000000  0.000000       inf\n",
 650 |        "e62f1261      2         1             3  0.666667  0.333333  2.000000\n",
 651 |        "135f7d9a      2         0             2  1.000000  0.000000       inf\n",
 652 |        "9af87478      2         0             2  1.000000  0.000000       inf\n",
 653 |        "77cf1a27      1         0             1  1.000000  0.000000       inf\n",
 654 |        "d62216cc      1         0             1  1.000000  0.000000       inf\n",
 655 |        "fcc5c7c0      1         0             1  1.000000  0.000000       inf\n",
 656 |        "7181509e      1         0             1  1.000000  0.000000       inf\n",
 657 |        "2a32a3ca      1         0             1  1.000000  0.000000       inf\n",
 658 |        "2ad16ba3      1         0             1  1.000000  0.000000       inf\n",
 659 |        "938f494b      1         0             1  1.000000  0.000000       inf\n",
 660 |        "24dbae83      1         0             1  1.000000  0.000000       inf\n",
 661 |        "ca9b95aa      1         0             1  1.000000  0.000000       inf\n",
 662 |        "7c369899      1         0             1  1.000000  0.000000       inf\n",
 663 |        "3bf8c26c      1         0             1  1.000000  0.000000       inf\n",
 664 |        "45378128      1         0             1  1.000000  0.000000       inf\n",
 665 |        "023ca1f9      1         0             1  1.000000  0.000000       inf\n",
 666 |        "3b9ab74d      1         0             1  1.000000  0.000000       inf\n",
 667 |        "9eb9a972      1         0             1  1.000000  0.000000       inf\n",
 668 |        "59bcd1ae      1         0             1  1.000000  0.000000       inf\n",
 669 |        "7f8c00b4      1         0             1  1.000000  0.000000       inf\n",
 670 |        "8d61d7eb      1         0             1  1.000000  0.000000       inf\n",
 671 |        "b441c41f      1         0             1  1.000000  0.000000       inf\n",
 672 |        "e317838f      1         0             1  1.000000  0.000000       inf\n",
 673 |        "2e7d4b65      1         0             1  1.000000  0.000000       inf\n",
 674 |        "70d97ece      1         0             1  1.000000  0.000000       inf\n",
 675 |        "9809e6c9      1         0             1  1.000000  0.000000       inf\n",
 676 |        "cb73ba55      1         0             1  1.000000  0.000000       inf\n",
 677 |        "...         ...       ...           ...       ...       ...       ...\n",
 678 |        "0c3bbac0      0         1             1  0.000000  1.000000  0.000000\n",
 679 |        "34c9f908      0         1             1  0.000000  1.000000  0.000000\n",
 680 |        "41a1ae5f      0         1             1  0.000000  1.000000  0.000000\n",
 681 |        "5d03585e      0         1             1  0.000000  1.000000  0.000000\n",
 682 |        "7d242dfd      0         1             1  0.000000  1.000000  0.000000\n",
 683 |        "9ffa0563      0         1             1  0.000000  1.000000  0.000000\n",
 684 |        "2fd58990      0         1             1  0.000000  1.000000  0.000000\n",
 685 |        "01e47a3d      0         1             1  0.000000  1.000000  0.000000\n",
 686 |        "bcd5195e      0         1             1  0.000000  1.000000  0.000000\n",
 687 |        "a318236b      0         1             1  0.000000  1.000000  0.000000\n",
 688 |        "1168ce02      0         1             1  0.000000  1.000000  0.000000\n",
 689 |        "83c34e93      0         1             1  0.000000  1.000000  0.000000\n",
 690 |        "c8c31032      0         1             1  0.000000  1.000000  0.000000\n",
 691 |        "004270bf      0         1             1  0.000000  1.000000  0.000000\n",
 692 |        "e6d0facc      0         1             1  0.000000  1.000000  0.000000\n",
 693 |        "844524dc      0         1             1  0.000000  1.000000  0.000000\n",
 694 |        "c90a30a1      0         1             1  0.000000  1.000000  0.000000\n",
 695 |        "5015495b      0         1             1  0.000000  1.000000  0.000000\n",
 696 |        "a12e8a45      0         1             1  0.000000  1.000000  0.000000\n",
 697 |        "7b5c2c3b      0         1             1  0.000000  1.000000  0.000000\n",
 698 |        "4a6b5af3      0         1             1  0.000000  1.000000  0.000000\n",
 699 |        "2c8ae68c      0         1             1  0.000000  1.000000  0.000000\n",
 700 |        "84565c92      0         1             1  0.000000  1.000000  0.000000\n",
 701 |        "2a6fe2a5      0         1             1  0.000000  1.000000  0.000000\n",
 702 |        "e0f70006      0         1             1  0.000000  1.000000  0.000000\n",
 703 |        "cef4c8cc      0         1             1  0.000000  1.000000  0.000000\n",
 704 |        "7f4b1f1e      0         1             1  0.000000  1.000000  0.000000\n",
 705 |        "7efe14f0      0         1             1  0.000000  1.000000  0.000000\n",
 706 |        "02b99e77      0         1             1  0.000000  1.000000  0.000000\n",
 707 |        "cbb50c1c      0         1             1  0.000000  1.000000  0.000000\n",
 708 |        "\n",
 709 |        "[906 rows x 6 columns]"
 710 |       ]
 711 |      },
 712 |      "execution_count": 12,
 713 |      "metadata": {},
 714 |      "output_type": "execute_result"
 715 |     }
 716 |    ],
 717 |    "source": [
 718 |     "def click_counting(x, bin_column):\n",
 719 |     "    clicks = pd.Series(x[x['click'] > 0][bin_column].value_counts(), name='clicks')\n",
 720 |     "    no_clicks = pd.Series(x[x['click'] < 1][bin_column].value_counts(), name='no_clicks')\n",
 721 |     "\n",
 722 |     "    counts = pd.DataFrame([clicks,no_clicks]).T.fillna('0')\n",
 723 |     "    counts['total_clicks'] = counts['clicks'].astype('int64') + counts['no_clicks'].astype('int64')\n",
 724 |     "    return counts\n",
 725 |     "\n",
 726 |     "def bin_counting(counts):\n",
 727 |     "    counts['N+'] = counts['clicks'].astype('int64').divide(counts['total_clicks'].astype('int64'))\n",
 728 |     "    counts['N-'] = counts['no_clicks'].astype('int64').divide(counts['total_clicks'].astype('int64'))\n",
 729 |     "    counts['log_N+'] = counts['N+'].divide(counts['N-'])\n",
 730 |     "    # Bin Countingのプロパティを返すだけの場合、ここでフィルタリングを実行\n",
 731 |     "    bin_counts = counts.filter(items= ['N+', 'N-', 'log_N+'])\n",
 732 |     "    return counts, bin_counts\n",
 733 |     "\n",
 734 |     "# device_idを対象としたビンカウンティング\n",
 735 |     "bin_column = 'device_id'\n",
 736 |     "device_clicks = click_counting(df.filter(items=[bin_column, 'click']), bin_column)\n",
 737 |     "device_all, device_bin_counts = bin_counting(device_clicks)\n",
 738 |     "\n",
 739 |     "device_all"
 740 |    ]
 741 |   },
 742 |   {
 743 |    "cell_type": "code",
 744 |    "execution_count": 13,
 745 |    "metadata": {},
 746 |    "outputs": [
 747 |     {
 748 |      "data": {
 749 |       "text/html": [
 750 |        "<div>\n",
 751 |        "<style scoped>\n",
 752 |        "    .dataframe tbody tr th:only-of-type {\n",
 753 |        "        vertical-align: middle;\n",
 754 |        "    }\n",
 755 |        "\n",
 756 |        "    .dataframe tbody tr th {\n",
 757 |        "        vertical-align: top;\n",
 758 |        "    }\n",
 759 |        "\n",
 760 |        "    .dataframe thead th {\n",
 761 |        "        text-align: right;\n",
 762 |        "    }\n",
 763 |        "</style>\n",
 764 |        "<table border=\"1\" class=\"dataframe\">\n",
 765 |        "  <thead>\n",
 766 |        "    <tr style=\"text-align: right;\">\n",
 767 |        "      <th></th>\n",
 768 |        "      <th>N+</th>\n",
 769 |        "      <th>N-</th>\n",
 770 |        "      <th>log_N+</th>\n",
 771 |        "    </tr>\n",
 772 |        "  </thead>\n",
 773 |        "  <tbody>\n",
 774 |        "    <tr>\n",
 775 |        "      <th>a99f214a</th>\n",
 776 |        "      <td>0.178706</td>\n",
 777 |        "      <td>0.821294</td>\n",
 778 |        "      <td>0.217591</td>\n",
 779 |        "    </tr>\n",
 780 |        "    <tr>\n",
 781 |        "      <th>c357dbff</th>\n",
 782 |        "      <td>0.142857</td>\n",
 783 |        "      <td>0.857143</td>\n",
 784 |        "      <td>0.166667</td>\n",
 785 |        "    </tr>\n",
 786 |        "    <tr>\n",
 787 |        "      <th>25635c83</th>\n",
 788 |        "      <td>1.000000</td>\n",
 789 |        "      <td>0.000000</td>\n",
 790 |        "      <td>inf</td>\n",
 791 |        "    </tr>\n",
 792 |        "    <tr>\n",
 793 |        "      <th>e62f1261</th>\n",
 794 |        "      <td>0.666667</td>\n",
 795 |        "      <td>0.333333</td>\n",
 796 |        "      <td>2.000000</td>\n",
 797 |        "    </tr>\n",
 798 |        "    <tr>\n",
 799 |        "      <th>135f7d9a</th>\n",
 800 |        "      <td>1.000000</td>\n",
 801 |        "      <td>0.000000</td>\n",
 802 |        "      <td>inf</td>\n",
 803 |        "    </tr>\n",
 804 |        "    <tr>\n",
 805 |        "      <th>9af87478</th>\n",
 806 |        "      <td>1.000000</td>\n",
 807 |        "      <td>0.000000</td>\n",
 808 |        "      <td>inf</td>\n",
 809 |        "    </tr>\n",
 810 |        "    <tr>\n",
 811 |        "      <th>77cf1a27</th>\n",
 812 |        "      <td>1.000000</td>\n",
 813 |        "      <td>0.000000</td>\n",
 814 |        "      <td>inf</td>\n",
 815 |        "    </tr>\n",
 816 |        "    <tr>\n",
 817 |        "      <th>d62216cc</th>\n",
 818 |        "      <td>1.000000</td>\n",
 819 |        "      <td>0.000000</td>\n",
 820 |        "      <td>inf</td>\n",
 821 |        "    </tr>\n",
 822 |        "    <tr>\n",
 823 |        "      <th>fcc5c7c0</th>\n",
 824 |        "      <td>1.000000</td>\n",
 825 |        "      <td>0.000000</td>\n",
 826 |        "      <td>inf</td>\n",
 827 |        "    </tr>\n",
 828 |        "    <tr>\n",
 829 |        "      <th>7181509e</th>\n",
 830 |        "      <td>1.000000</td>\n",
 831 |        "      <td>0.000000</td>\n",
 832 |        "      <td>inf</td>\n",
 833 |        "    </tr>\n",
 834 |        "    <tr>\n",
 835 |        "      <th>2a32a3ca</th>\n",
 836 |        "      <td>1.000000</td>\n",
 837 |        "      <td>0.000000</td>\n",
 838 |        "      <td>inf</td>\n",
 839 |        "    </tr>\n",
 840 |        "    <tr>\n",
 841 |        "      <th>2ad16ba3</th>\n",
 842 |        "      <td>1.000000</td>\n",
 843 |        "      <td>0.000000</td>\n",
 844 |        "      <td>inf</td>\n",
 845 |        "    </tr>\n",
 846 |        "    <tr>\n",
 847 |        "      <th>938f494b</th>\n",
 848 |        "      <td>1.000000</td>\n",
 849 |        "      <td>0.000000</td>\n",
 850 |        "      <td>inf</td>\n",
 851 |        "    </tr>\n",
 852 |        "    <tr>\n",
 853 |        "      <th>24dbae83</th>\n",
 854 |        "      <td>1.000000</td>\n",
 855 |        "      <td>0.000000</td>\n",
 856 |        "      <td>inf</td>\n",
 857 |        "    </tr>\n",
 858 |        "    <tr>\n",
 859 |        "      <th>ca9b95aa</th>\n",
 860 |        "      <td>1.000000</td>\n",
 861 |        "      <td>0.000000</td>\n",
 862 |        "      <td>inf</td>\n",
 863 |        "    </tr>\n",
 864 |        "    <tr>\n",
 865 |        "      <th>7c369899</th>\n",
 866 |        "      <td>1.000000</td>\n",
 867 |        "      <td>0.000000</td>\n",
 868 |        "      <td>inf</td>\n",
 869 |        "    </tr>\n",
 870 |        "    <tr>\n",
 871 |        "      <th>3bf8c26c</th>\n",
 872 |        "      <td>1.000000</td>\n",
 873 |        "      <td>0.000000</td>\n",
 874 |        "      <td>inf</td>\n",
 875 |        "    </tr>\n",
 876 |        "    <tr>\n",
 877 |        "      <th>45378128</th>\n",
 878 |        "      <td>1.000000</td>\n",
 879 |        "      <td>0.000000</td>\n",
 880 |        "      <td>inf</td>\n",
 881 |        "    </tr>\n",
 882 |        "    <tr>\n",
 883 |        "      <th>023ca1f9</th>\n",
 884 |        "      <td>1.000000</td>\n",
 885 |        "      <td>0.000000</td>\n",
 886 |        "      <td>inf</td>\n",
 887 |        "    </tr>\n",
 888 |        "    <tr>\n",
 889 |        "      <th>3b9ab74d</th>\n",
 890 |        "      <td>1.000000</td>\n",
 891 |        "      <td>0.000000</td>\n",
 892 |        "      <td>inf</td>\n",
 893 |        "    </tr>\n",
 894 |        "    <tr>\n",
 895 |        "      <th>9eb9a972</th>\n",
 896 |        "      <td>1.000000</td>\n",
 897 |        "      <td>0.000000</td>\n",
 898 |        "      <td>inf</td>\n",
 899 |        "    </tr>\n",
 900 |        "    <tr>\n",
 901 |        "      <th>59bcd1ae</th>\n",
 902 |        "      <td>1.000000</td>\n",
 903 |        "      <td>0.000000</td>\n",
 904 |        "      <td>inf</td>\n",
 905 |        "    </tr>\n",
 906 |        "    <tr>\n",
 907 |        "      <th>7f8c00b4</th>\n",
 908 |        "      <td>1.000000</td>\n",
 909 |        "      <td>0.000000</td>\n",
 910 |        "      <td>inf</td>\n",
 911 |        "    </tr>\n",
 912 |        "    <tr>\n",
 913 |        "      <th>8d61d7eb</th>\n",
 914 |        "      <td>1.000000</td>\n",
 915 |        "      <td>0.000000</td>\n",
 916 |        "      <td>inf</td>\n",
 917 |        "    </tr>\n",
 918 |        "    <tr>\n",
 919 |        "      <th>b441c41f</th>\n",
 920 |        "      <td>1.000000</td>\n",
 921 |        "      <td>0.000000</td>\n",
 922 |        "      <td>inf</td>\n",
 923 |        "    </tr>\n",
 924 |        "    <tr>\n",
 925 |        "      <th>e317838f</th>\n",
 926 |        "      <td>1.000000</td>\n",
 927 |        "      <td>0.000000</td>\n",
 928 |        "      <td>inf</td>\n",
 929 |        "    </tr>\n",
 930 |        "    <tr>\n",
 931 |        "      <th>2e7d4b65</th>\n",
 932 |        "      <td>1.000000</td>\n",
 933 |        "      <td>0.000000</td>\n",
 934 |        "      <td>inf</td>\n",
 935 |        "    </tr>\n",
 936 |        "    <tr>\n",
 937 |        "      <th>70d97ece</th>\n",
 938 |        "      <td>1.000000</td>\n",
 939 |        "      <td>0.000000</td>\n",
 940 |        "      <td>inf</td>\n",
 941 |        "    </tr>\n",
 942 |        "    <tr>\n",
 943 |        "      <th>9809e6c9</th>\n",
 944 |        "      <td>1.000000</td>\n",
 945 |        "      <td>0.000000</td>\n",
 946 |        "      <td>inf</td>\n",
 947 |        "    </tr>\n",
 948 |        "    <tr>\n",
 949 |        "      <th>cb73ba55</th>\n",
 950 |        "      <td>1.000000</td>\n",
 951 |        "      <td>0.000000</td>\n",
 952 |        "      <td>inf</td>\n",
 953 |        "    </tr>\n",
 954 |        "    <tr>\n",
 955 |        "      <th>...</th>\n",
 956 |        "      <td>...</td>\n",
 957 |        "      <td>...</td>\n",
 958 |        "      <td>...</td>\n",
 959 |        "    </tr>\n",
 960 |        "    <tr>\n",
 961 |        "      <th>0c3bbac0</th>\n",
 962 |        "      <td>0.000000</td>\n",
 963 |        "      <td>1.000000</td>\n",
 964 |        "      <td>0.000000</td>\n",
 965 |        "    </tr>\n",
 966 |        "    <tr>\n",
 967 |        "      <th>34c9f908</th>\n",
 968 |        "      <td>0.000000</td>\n",
 969 |        "      <td>1.000000</td>\n",
 970 |        "      <td>0.000000</td>\n",
 971 |        "    </tr>\n",
 972 |        "    <tr>\n",
 973 |        "      <th>41a1ae5f</th>\n",
 974 |        "      <td>0.000000</td>\n",
 975 |        "      <td>1.000000</td>\n",
 976 |        "      <td>0.000000</td>\n",
 977 |        "    </tr>\n",
 978 |        "    <tr>\n",
 979 |        "      <th>5d03585e</th>\n",
 980 |        "      <td>0.000000</td>\n",
 981 |        "      <td>1.000000</td>\n",
 982 |        "      <td>0.000000</td>\n",
 983 |        "    </tr>\n",
 984 |        "    <tr>\n",
 985 |        "      <th>7d242dfd</th>\n",
 986 |        "      <td>0.000000</td>\n",
 987 |        "      <td>1.000000</td>\n",
 988 |        "      <td>0.000000</td>\n",
 989 |        "    </tr>\n",
 990 |        "    <tr>\n",
 991 |        "      <th>9ffa0563</th>\n",
 992 |        "      <td>0.000000</td>\n",
 993 |        "      <td>1.000000</td>\n",
 994 |        "      <td>0.000000</td>\n",
 995 |        "    </tr>\n",
 996 |        "    <tr>\n",
 997 |        "      <th>2fd58990</th>\n",
 998 |        "      <td>0.000000</td>\n",
 999 |        "      <td>1.000000</td>\n",
1000 |        "      <td>0.000000</td>\n",
1001 |        "    </tr>\n",
1002 |        "    <tr>\n",
1003 |        "      <th>01e47a3d</th>\n",
1004 |        "      <td>0.000000</td>\n",
1005 |        "      <td>1.000000</td>\n",
1006 |        "      <td>0.000000</td>\n",
1007 |        "    </tr>\n",
1008 |        "    <tr>\n",
1009 |        "      <th>bcd5195e</th>\n",
1010 |        "      <td>0.000000</td>\n",
1011 |        "      <td>1.000000</td>\n",
1012 |        "      <td>0.000000</td>\n",
1013 |        "    </tr>\n",
1014 |        "    <tr>\n",
1015 |        "      <th>a318236b</th>\n",
1016 |        "      <td>0.000000</td>\n",
1017 |        "      <td>1.000000</td>\n",
1018 |        "      <td>0.000000</td>\n",
1019 |        "    </tr>\n",
1020 |        "    <tr>\n",
1021 |        "      <th>1168ce02</th>\n",
1022 |        "      <td>0.000000</td>\n",
1023 |        "      <td>1.000000</td>\n",
1024 |        "      <td>0.000000</td>\n",
1025 |        "    </tr>\n",
1026 |        "    <tr>\n",
1027 |        "      <th>83c34e93</th>\n",
1028 |        "      <td>0.000000</td>\n",
1029 |        "      <td>1.000000</td>\n",
1030 |        "      <td>0.000000</td>\n",
1031 |        "    </tr>\n",
1032 |        "    <tr>\n",
1033 |        "      <th>c8c31032</th>\n",
1034 |        "      <td>0.000000</td>\n",
1035 |        "      <td>1.000000</td>\n",
1036 |        "      <td>0.000000</td>\n",
1037 |        "    </tr>\n",
1038 |        "    <tr>\n",
1039 |        "      <th>004270bf</th>\n",
1040 |        "      <td>0.000000</td>\n",
1041 |        "      <td>1.000000</td>\n",
1042 |        "      <td>0.000000</td>\n",
1043 |        "    </tr>\n",
1044 |        "    <tr>\n",
1045 |        "      <th>e6d0facc</th>\n",
1046 |        "      <td>0.000000</td>\n",
1047 |        "      <td>1.000000</td>\n",
1048 |        "      <td>0.000000</td>\n",
1049 |        "    </tr>\n",
1050 |        "    <tr>\n",
1051 |        "      <th>844524dc</th>\n",
1052 |        "      <td>0.000000</td>\n",
1053 |        "      <td>1.000000</td>\n",
1054 |        "      <td>0.000000</td>\n",
1055 |        "    </tr>\n",
1056 |        "    <tr>\n",
1057 |        "      <th>c90a30a1</th>\n",
1058 |        "      <td>0.000000</td>\n",
1059 |        "      <td>1.000000</td>\n",
1060 |        "      <td>0.000000</td>\n",
1061 |        "    </tr>\n",
1062 |        "    <tr>\n",
1063 |        "      <th>5015495b</th>\n",
1064 |        "      <td>0.000000</td>\n",
1065 |        "      <td>1.000000</td>\n",
1066 |        "      <td>0.000000</td>\n",
1067 |        "    </tr>\n",
1068 |        "    <tr>\n",
1069 |        "      <th>a12e8a45</th>\n",
1070 |        "      <td>0.000000</td>\n",
1071 |        "      <td>1.000000</td>\n",
1072 |        "      <td>0.000000</td>\n",
1073 |        "    </tr>\n",
1074 |        "    <tr>\n",
1075 |        "      <th>7b5c2c3b</th>\n",
1076 |        "      <td>0.000000</td>\n",
1077 |        "      <td>1.000000</td>\n",
1078 |        "      <td>0.000000</td>\n",
1079 |        "    </tr>\n",
1080 |        "    <tr>\n",
1081 |        "      <th>4a6b5af3</th>\n",
1082 |        "      <td>0.000000</td>\n",
1083 |        "      <td>1.000000</td>\n",
1084 |        "      <td>0.000000</td>\n",
1085 |        "    </tr>\n",
1086 |        "    <tr>\n",
1087 |        "      <th>2c8ae68c</th>\n",
1088 |        "      <td>0.000000</td>\n",
1089 |        "      <td>1.000000</td>\n",
1090 |        "      <td>0.000000</td>\n",
1091 |        "    </tr>\n",
1092 |        "    <tr>\n",
1093 |        "      <th>84565c92</th>\n",
1094 |        "      <td>0.000000</td>\n",
1095 |        "      <td>1.000000</td>\n",
1096 |        "      <td>0.000000</td>\n",
1097 |        "    </tr>\n",
1098 |        "    <tr>\n",
1099 |        "      <th>2a6fe2a5</th>\n",
1100 |        "      <td>0.000000</td>\n",
1101 |        "      <td>1.000000</td>\n",
1102 |        "      <td>0.000000</td>\n",
1103 |        "    </tr>\n",
1104 |        "    <tr>\n",
1105 |        "      <th>e0f70006</th>\n",
1106 |        "      <td>0.000000</td>\n",
1107 |        "      <td>1.000000</td>\n",
1108 |        "      <td>0.000000</td>\n",
1109 |        "    </tr>\n",
1110 |        "    <tr>\n",
1111 |        "      <th>cef4c8cc</th>\n",
1112 |        "      <td>0.000000</td>\n",
1113 |        "      <td>1.000000</td>\n",
1114 |        "      <td>0.000000</td>\n",
1115 |        "    </tr>\n",
1116 |        "    <tr>\n",
1117 |        "      <th>7f4b1f1e</th>\n",
1118 |        "      <td>0.000000</td>\n",
1119 |        "      <td>1.000000</td>\n",
1120 |        "      <td>0.000000</td>\n",
1121 |        "    </tr>\n",
1122 |        "    <tr>\n",
1123 |        "      <th>7efe14f0</th>\n",
1124 |        "      <td>0.000000</td>\n",
1125 |        "      <td>1.000000</td>\n",
1126 |        "      <td>0.000000</td>\n",
1127 |        "    </tr>\n",
1128 |        "    <tr>\n",
1129 |        "      <th>02b99e77</th>\n",
1130 |        "      <td>0.000000</td>\n",
1131 |        "      <td>1.000000</td>\n",
1132 |        "      <td>0.000000</td>\n",
1133 |        "    </tr>\n",
1134 |        "    <tr>\n",
1135 |        "      <th>cbb50c1c</th>\n",
1136 |        "      <td>0.000000</td>\n",
1137 |        "      <td>1.000000</td>\n",
1138 |        "      <td>0.000000</td>\n",
1139 |        "    </tr>\n",
1140 |        "  </tbody>\n",
1141 |        "</table>\n",
1142 |        "<p>906 rows × 3 columns</p>\n",
1143 |        "</div>"
1144 |       ],
1145 |       "text/plain": [
1146 |        "                N+        N-    log_N+\n",
1147 |        "a99f214a  0.178706  0.821294  0.217591\n",
1148 |        "c357dbff  0.142857  0.857143  0.166667\n",
1149 |        "25635c83  1.000000  0.000000       inf\n",
1150 |        "e62f1261  0.666667  0.333333  2.000000\n",
1151 |        "135f7d9a  1.000000  0.000000       inf\n",
1152 |        "9af87478  1.000000  0.000000       inf\n",
1153 |        "77cf1a27  1.000000  0.000000       inf\n",
1154 |        "d62216cc  1.000000  0.000000       inf\n",
1155 |        "fcc5c7c0  1.000000  0.000000       inf\n",
1156 |        "7181509e  1.000000  0.000000       inf\n",
1157 |        "2a32a3ca  1.000000  0.000000       inf\n",
1158 |        "2ad16ba3  1.000000  0.000000       inf\n",
1159 |        "938f494b  1.000000  0.000000       inf\n",
1160 |        "24dbae83  1.000000  0.000000       inf\n",
1161 |        "ca9b95aa  1.000000  0.000000       inf\n",
1162 |        "7c369899  1.000000  0.000000       inf\n",
1163 |        "3bf8c26c  1.000000  0.000000       inf\n",
1164 |        "45378128  1.000000  0.000000       inf\n",
1165 |        "023ca1f9  1.000000  0.000000       inf\n",
1166 |        "3b9ab74d  1.000000  0.000000       inf\n",
1167 |        "9eb9a972  1.000000  0.000000       inf\n",
1168 |        "59bcd1ae  1.000000  0.000000       inf\n",
1169 |        "7f8c00b4  1.000000  0.000000       inf\n",
1170 |        "8d61d7eb  1.000000  0.000000       inf\n",
1171 |        "b441c41f  1.000000  0.000000       inf\n",
1172 |        "e317838f  1.000000  0.000000       inf\n",
1173 |        "2e7d4b65  1.000000  0.000000       inf\n",
1174 |        "70d97ece  1.000000  0.000000       inf\n",
1175 |        "9809e6c9  1.000000  0.000000       inf\n",
1176 |        "cb73ba55  1.000000  0.000000       inf\n",
1177 |        "...            ...       ...       ...\n",
1178 |        "0c3bbac0  0.000000  1.000000  0.000000\n",
1179 |        "34c9f908  0.000000  1.000000  0.000000\n",
1180 |        "41a1ae5f  0.000000  1.000000  0.000000\n",
1181 |        "5d03585e  0.000000  1.000000  0.000000\n",
1182 |        "7d242dfd  0.000000  1.000000  0.000000\n",
1183 |        "9ffa0563  0.000000  1.000000  0.000000\n",
1184 |        "2fd58990  0.000000  1.000000  0.000000\n",
1185 |        "01e47a3d  0.000000  1.000000  0.000000\n",
1186 |        "bcd5195e  0.000000  1.000000  0.000000\n",
1187 |        "a318236b  0.000000  1.000000  0.000000\n",
1188 |        "1168ce02  0.000000  1.000000  0.000000\n",
1189 |        "83c34e93  0.000000  1.000000  0.000000\n",
1190 |        "c8c31032  0.000000  1.000000  0.000000\n",
1191 |        "004270bf  0.000000  1.000000  0.000000\n",
1192 |        "e6d0facc  0.000000  1.000000  0.000000\n",
1193 |        "844524dc  0.000000  1.000000  0.000000\n",
1194 |        "c90a30a1  0.000000  1.000000  0.000000\n",
1195 |        "5015495b  0.000000  1.000000  0.000000\n",
1196 |        "a12e8a45  0.000000  1.000000  0.000000\n",
1197 |        "7b5c2c3b  0.000000  1.000000  0.000000\n",
1198 |        "4a6b5af3  0.000000  1.000000  0.000000\n",
1199 |        "2c8ae68c  0.000000  1.000000  0.000000\n",
1200 |        "84565c92  0.000000  1.000000  0.000000\n",
1201 |        "2a6fe2a5  0.000000  1.000000  0.000000\n",
1202 |        "e0f70006  0.000000  1.000000  0.000000\n",
1203 |        "cef4c8cc  0.000000  1.000000  0.000000\n",
1204 |        "7f4b1f1e  0.000000  1.000000  0.000000\n",
1205 |        "7efe14f0  0.000000  1.000000  0.000000\n",
1206 |        "02b99e77  0.000000  1.000000  0.000000\n",
1207 |        "cbb50c1c  0.000000  1.000000  0.000000\n",
1208 |        "\n",
1209 |        "[906 rows x 3 columns]"
1210 |       ]
1211 |      },
1212 |      "execution_count": 13,
1213 |      "metadata": {},
1214 |      "output_type": "execute_result"
1215 |     }
1216 |    ],
1217 |    "source": [
1218 |     "device_bin_counts"
1219 |    ]
1220 |   },
1221 |   {
1222 |    "cell_type": "code",
1223 |    "execution_count": 14,
1224 |    "metadata": {},
1225 |    "outputs": [
1226 |     {
1227 |      "data": {
1228 |       "text/plain": [
1229 |        "906"
1230 |       ]
1231 |      },
1232 |      "execution_count": 14,
1233 |      "metadata": {},
1234 |      "output_type": "execute_result"
1235 |     }
1236 |    ],
1237 |    "source": [
1238 |     "len(device_bin_counts)"
1239 |    ]
1240 |   },
1241 |   {
1242 |    "cell_type": "code",
1243 |    "execution_count": 15,
1244 |    "metadata": {},
1245 |    "outputs": [
1246 |     {
1247 |      "data": {
1248 |       "text/html": [
1249 |        "<div>\n",
1250 |        "<style scoped>\n",
1251 |        "    .dataframe tbody tr th:only-of-type {\n",
1252 |        "        vertical-align: middle;\n",
1253 |        "    }\n",
1254 |        "\n",
1255 |        "    .dataframe tbody tr th {\n",
1256 |        "        vertical-align: top;\n",
1257 |        "    }\n",
1258 |        "\n",
1259 |        "    .dataframe thead th {\n",
1260 |        "        text-align: right;\n",
1261 |        "    }\n",
1262 |        "</style>\n",
1263 |        "<table border=\"1\" class=\"dataframe\">\n",
1264 |        "  <thead>\n",
1265 |        "    <tr style=\"text-align: right;\">\n",
1266 |        "      <th></th>\n",
1267 |        "      <th>clicks</th>\n",
1268 |        "      <th>no_clicks</th>\n",
1269 |        "      <th>total_clicks</th>\n",
1270 |        "      <th>N+</th>\n",
1271 |        "      <th>N-</th>\n",
1272 |        "      <th>log_N+</th>\n",
1273 |        "    </tr>\n",
1274 |        "  </thead>\n",
1275 |        "  <tbody>\n",
1276 |        "    <tr>\n",
1277 |        "      <th>a99f214a</th>\n",
1278 |        "      <td>1279</td>\n",
1279 |        "      <td>5878</td>\n",
1280 |        "      <td>7157</td>\n",
1281 |        "      <td>0.178706</td>\n",
1282 |        "      <td>0.821294</td>\n",
1283 |        "      <td>0.217591</td>\n",
1284 |        "    </tr>\n",
1285 |        "    <tr>\n",
1286 |        "      <th>c357dbff</th>\n",
1287 |        "      <td>2</td>\n",
1288 |        "      <td>12</td>\n",
1289 |        "      <td>14</td>\n",
1290 |        "      <td>0.142857</td>\n",
1291 |        "      <td>0.857143</td>\n",
1292 |        "      <td>0.166667</td>\n",
1293 |        "    </tr>\n",
1294 |        "    <tr>\n",
1295 |        "      <th>a167aa83</th>\n",
1296 |        "      <td>0</td>\n",
1297 |        "      <td>7</td>\n",
1298 |        "      <td>7</td>\n",
1299 |        "      <td>0.000000</td>\n",
1300 |        "      <td>1.000000</td>\n",
1301 |        "      <td>0.000000</td>\n",
1302 |        "    </tr>\n",
1303 |        "    <tr>\n",
1304 |        "      <th>d2bbb640</th>\n",
1305 |        "      <td>0</td>\n",
1306 |        "      <td>6</td>\n",
1307 |        "      <td>6</td>\n",
1308 |        "      <td>0.000000</td>\n",
1309 |        "      <td>1.000000</td>\n",
1310 |        "      <td>0.000000</td>\n",
1311 |        "    </tr>\n",
1312 |        "  </tbody>\n",
1313 |        "</table>\n",
1314 |        "</div>"
1315 |       ],
1316 |       "text/plain": [
1317 |        "         clicks no_clicks  total_clicks        N+        N-    log_N+\n",
1318 |        "a99f214a   1279      5878          7157  0.178706  0.821294  0.217591\n",
1319 |        "c357dbff      2        12            14  0.142857  0.857143  0.166667\n",
1320 |        "a167aa83      0         7             7  0.000000  1.000000  0.000000\n",
1321 |        "d2bbb640      0         6             6  0.000000  1.000000  0.000000"
1322 |       ]
1323 |      },
1324 |      "execution_count": 15,
1325 |      "metadata": {},
1326 |      "output_type": "execute_result"
1327 |     }
1328 |    ],
1329 |    "source": [
1330 |     "device_all.sort_values(by = 'total_clicks', ascending=False).head(4)"
1331 |    ]
1332 |   },
1333 |   {
1334 |    "cell_type": "code",
1335 |    "execution_count": null,
1336 |    "metadata": {},
1337 |    "outputs": [],
1338 |    "source": []
1339 |   }
1340 |  ],
1341 |  "metadata": {
1342 |   "kernelspec": {
1343 |    "display_name": "Python 3",
1344 |    "language": "python",
1345 |    "name": "python3"
1346 |   },
1347 |   "language_info": {
1348 |    "codemirror_mode": {
1349 |     "name": "ipython",
1350 |     "version": 3
1351 |    },
1352 |    "file_extension": ".py",
1353 |    "mimetype": "text/x-python",
1354 |    "name": "python",
1355 |    "nbconvert_exporter": "python",
1356 |    "pygments_lexer": "ipython3",
1357 |    "version": "3.7.0"
1358 |   }
1359 |  },
1360 |  "nbformat": 4,
1361 |  "nbformat_minor": 2
1362 | }
1363 | 


--------------------------------------------------------------------------------
/2-2-2.ipynb:
--------------------------------------------------------------------------------
  1 | {
  2 |  "cells": [
  3 |   {
  4 |    "cell_type": "markdown",
  5 |    "metadata": {},
  6 |    "source": [
  7 |     "## 例2-2. Yelp データセット内の店舗に対するレビュー件数の可視化"
  8 |    ]
  9 |   },
 10 |   {
 11 |    "cell_type": "code",
 12 |    "execution_count": 7,
 13 |    "metadata": {},
 14 |    "outputs": [
 15 |     {
 16 |      "data": {
 17 |       "text/html": [
 18 |        "<div>\n",
 19 |        "<style scoped>\n",
 20 |        "    .dataframe tbody tr th:only-of-type {\n",
 21 |        "        vertical-align: middle;\n",
 22 |        "    }\n",
 23 |        "\n",
 24 |        "    .dataframe tbody tr th {\n",
 25 |        "        vertical-align: top;\n",
 26 |        "    }\n",
 27 |        "\n",
 28 |        "    .dataframe thead th {\n",
 29 |        "        text-align: right;\n",
 30 |        "    }\n",
 31 |        "</style>\n",
 32 |        "<table border=\"1\" class=\"dataframe\">\n",
 33 |        "  <thead>\n",
 34 |        "    <tr style=\"text-align: right;\">\n",
 35 |        "      <th></th>\n",
 36 |        "      <th>business_id</th>\n",
 37 |        "      <th>categories</th>\n",
 38 |        "      <th>city</th>\n",
 39 |        "      <th>full_address</th>\n",
 40 |        "      <th>latitude</th>\n",
 41 |        "      <th>longitude</th>\n",
 42 |        "      <th>name</th>\n",
 43 |        "      <th>neighborhoods</th>\n",
 44 |        "      <th>open</th>\n",
 45 |        "      <th>review_count</th>\n",
 46 |        "      <th>stars</th>\n",
 47 |        "      <th>state</th>\n",
 48 |        "      <th>type</th>\n",
 49 |        "    </tr>\n",
 50 |        "  </thead>\n",
 51 |        "  <tbody>\n",
 52 |        "    <tr>\n",
 53 |        "      <th>0</th>\n",
 54 |        "      <td>rncjoVoEFUJGCUoC1JgnUA</td>\n",
 55 |        "      <td>[Accountants, Professional Services, Tax Servi...</td>\n",
 56 |        "      <td>Peoria</td>\n",
 57 |        "      <td>8466 W Peoria Ave\\nSte 6\\nPeoria, AZ 85345</td>\n",
 58 |        "      <td>33.581867</td>\n",
 59 |        "      <td>-112.241596</td>\n",
 60 |        "      <td>Peoria Income Tax Service</td>\n",
 61 |        "      <td>[]</td>\n",
 62 |        "      <td>True</td>\n",
 63 |        "      <td>3</td>\n",
 64 |        "      <td>5.0</td>\n",
 65 |        "      <td>AZ</td>\n",
 66 |        "      <td>business</td>\n",
 67 |        "    </tr>\n",
 68 |        "    <tr>\n",
 69 |        "      <th>1</th>\n",
 70 |        "      <td>0FNFSzCFP_rGUoJx8W7tJg</td>\n",
 71 |        "      <td>[Sporting Goods, Bikes, Shopping]</td>\n",
 72 |        "      <td>Phoenix</td>\n",
 73 |        "      <td>2149 W Wood Dr\\nPhoenix, AZ 85029</td>\n",
 74 |        "      <td>33.604054</td>\n",
 75 |        "      <td>-112.105933</td>\n",
 76 |        "      <td>Bike Doctor</td>\n",
 77 |        "      <td>[]</td>\n",
 78 |        "      <td>True</td>\n",
 79 |        "      <td>5</td>\n",
 80 |        "      <td>5.0</td>\n",
 81 |        "      <td>AZ</td>\n",
 82 |        "      <td>business</td>\n",
 83 |        "    </tr>\n",
 84 |        "    <tr>\n",
 85 |        "      <th>2</th>\n",
 86 |        "      <td>3f_lyB6vFK48ukH6ScvLHg</td>\n",
 87 |        "      <td>[]</td>\n",
 88 |        "      <td>Phoenix</td>\n",
 89 |        "      <td>1134 N Central Ave\\nPhoenix, AZ 85004</td>\n",
 90 |        "      <td>33.460526</td>\n",
 91 |        "      <td>-112.073933</td>\n",
 92 |        "      <td>Valley Permaculture Alliance</td>\n",
 93 |        "      <td>[]</td>\n",
 94 |        "      <td>True</td>\n",
 95 |        "      <td>4</td>\n",
 96 |        "      <td>5.0</td>\n",
 97 |        "      <td>AZ</td>\n",
 98 |        "      <td>business</td>\n",
 99 |        "    </tr>\n",
100 |        "    <tr>\n",
101 |        "      <th>3</th>\n",
102 |        "      <td>usAsSV36QmUej8--yvN-dg</td>\n",
103 |        "      <td>[Food, Grocery]</td>\n",
104 |        "      <td>Phoenix</td>\n",
105 |        "      <td>845 W Southern Ave\\nPhoenix, AZ 85041</td>\n",
106 |        "      <td>33.392210</td>\n",
107 |        "      <td>-112.085377</td>\n",
108 |        "      <td>Food City</td>\n",
109 |        "      <td>[]</td>\n",
110 |        "      <td>True</td>\n",
111 |        "      <td>5</td>\n",
112 |        "      <td>3.5</td>\n",
113 |        "      <td>AZ</td>\n",
114 |        "      <td>business</td>\n",
115 |        "    </tr>\n",
116 |        "    <tr>\n",
117 |        "      <th>4</th>\n",
118 |        "      <td>PzOqRohWw7F7YEPBz6AubA</td>\n",
119 |        "      <td>[Food, Bagels, Delis, Restaurants]</td>\n",
120 |        "      <td>Glendale Az</td>\n",
121 |        "      <td>6520 W Happy Valley Rd\\nSte 101\\nGlendale Az, ...</td>\n",
122 |        "      <td>33.712797</td>\n",
123 |        "      <td>-112.200264</td>\n",
124 |        "      <td>Hot Bagels &amp; Deli</td>\n",
125 |        "      <td>[]</td>\n",
126 |        "      <td>True</td>\n",
127 |        "      <td>14</td>\n",
128 |        "      <td>3.5</td>\n",
129 |        "      <td>AZ</td>\n",
130 |        "      <td>business</td>\n",
131 |        "    </tr>\n",
132 |        "  </tbody>\n",
133 |        "</table>\n",
134 |        "</div>"
135 |       ],
136 |       "text/plain": [
137 |        "              business_id                                         categories  \\\n",
138 |        "0  rncjoVoEFUJGCUoC1JgnUA  [Accountants, Professional Services, Tax Servi...   \n",
139 |        "1  0FNFSzCFP_rGUoJx8W7tJg                  [Sporting Goods, Bikes, Shopping]   \n",
140 |        "2  3f_lyB6vFK48ukH6ScvLHg                                                 []   \n",
141 |        "3  usAsSV36QmUej8--yvN-dg                                    [Food, Grocery]   \n",
142 |        "4  PzOqRohWw7F7YEPBz6AubA                 [Food, Bagels, Delis, Restaurants]   \n",
143 |        "\n",
144 |        "          city                                       full_address   latitude  \\\n",
145 |        "0       Peoria         8466 W Peoria Ave\\nSte 6\\nPeoria, AZ 85345  33.581867   \n",
146 |        "1      Phoenix                  2149 W Wood Dr\\nPhoenix, AZ 85029  33.604054   \n",
147 |        "2      Phoenix              1134 N Central Ave\\nPhoenix, AZ 85004  33.460526   \n",
148 |        "3      Phoenix              845 W Southern Ave\\nPhoenix, AZ 85041  33.392210   \n",
149 |        "4  Glendale Az  6520 W Happy Valley Rd\\nSte 101\\nGlendale Az, ...  33.712797   \n",
150 |        "\n",
151 |        "    longitude                          name neighborhoods  open  review_count  \\\n",
152 |        "0 -112.241596     Peoria Income Tax Service            []  True             3   \n",
153 |        "1 -112.105933                   Bike Doctor            []  True             5   \n",
154 |        "2 -112.073933  Valley Permaculture Alliance            []  True             4   \n",
155 |        "3 -112.085377                     Food City            []  True             5   \n",
156 |        "4 -112.200264             Hot Bagels & Deli            []  True            14   \n",
157 |        "\n",
158 |        "   stars state      type  \n",
159 |        "0    5.0    AZ  business  \n",
160 |        "1    5.0    AZ  business  \n",
161 |        "2    5.0    AZ  business  \n",
162 |        "3    3.5    AZ  business  \n",
163 |        "4    3.5    AZ  business  "
164 |       ]
165 |      },
166 |      "execution_count": 7,
167 |      "metadata": {},
168 |      "output_type": "execute_result"
169 |     }
170 |    ],
171 |    "source": [
172 |     "import pandas as pd\n",
173 |     "import json\n",
174 |     "\n",
175 |     "# 店舗についてのデータを読み込む\n",
176 |     "with open('data/yelp/yelp_academic_dataset_business.json') as biz_file:\n",
177 |     "    biz_df = pd.DataFrame([json.loads(x) for x in biz_file.readlines()])\n",
178 |     "\n",
179 |     "biz_df.head()"
180 |    ]
181 |   },
182 |   {
183 |    "cell_type": "code",
184 |    "execution_count": 8,
185 |    "metadata": {},
186 |    "outputs": [
187 |     {
188 |      "data": {
189 |       "text/plain": [
190 |        "Text(0, 0.5, 'Occurrence')"
191 |       ]
192 |      },
193 |      "execution_count": 8,
194 |      "metadata": {},
195 |      "output_type": "execute_result"
196 |     },
197 |     {
198 |      "data": {
199 |       "image/png": "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\n",
200 |       "text/plain": [
201 |        "<Figure size 432x288 with 1 Axes>"
202 |       ]
203 |      },
204 |      "metadata": {},
205 |      "output_type": "display_data"
206 |     }
207 |    ],
208 |    "source": [
209 |     "%matplotlib inline\n",
210 |     "import matplotlib.pyplot as plt\n",
211 |     "import seaborn as sns\n",
212 |     "\n",
213 |     "# レビュー件数のヒストグラムを描画\n",
214 |     "sns.set_style('whitegrid')\n",
215 |     "fig, ax = plt.subplots()\n",
216 |     "biz_df['review_count'].hist(ax=ax, bins=100)\n",
217 |     "ax.set_yscale('log')\n",
218 |     "ax.tick_params(labelsize=14)\n",
219 |     "ax.set_xlabel('Review Count', fontsize=14)\n",
220 |     "ax.set_ylabel('Occurrence', fontsize=14)"
221 |    ]
222 |   },
223 |   {
224 |    "cell_type": "markdown",
225 |    "metadata": {},
226 |    "source": [
227 |     "## 例 2-3. 固定幅によるカウントの離散化"
228 |    ]
229 |   },
230 |   {
231 |    "cell_type": "code",
232 |    "execution_count": 9,
233 |    "metadata": {},
234 |    "outputs": [
235 |     {
236 |      "data": {
237 |       "text/plain": [
238 |        "array([37, 12, 72,  9, 75,  5, 79, 64, 16,  1, 76, 71,  6, 25, 50, 20, 18,\n",
239 |        "       84, 11, 28])"
240 |       ]
241 |      },
242 |      "execution_count": 9,
243 |      "metadata": {},
244 |      "output_type": "execute_result"
245 |     }
246 |    ],
247 |    "source": [
248 |     "import numpy as np\n",
249 |     "\n",
250 |     "np.random.seed(seed=1)\n",
251 |     "\n",
252 |     "# 0から99までの整数を一様分布からランダムに20個生成する\n",
253 |     "small_counts = np.random.randint(0, 100, 20)\n",
254 |     "\n",
255 |     "small_counts"
256 |    ]
257 |   },
258 |   {
259 |    "cell_type": "code",
260 |    "execution_count": 10,
261 |    "metadata": {},
262 |    "outputs": [
263 |     {
264 |      "data": {
265 |       "text/plain": [
266 |        "array([3, 1, 7, 0, 7, 0, 7, 6, 1, 0, 7, 7, 0, 2, 5, 2, 1, 8, 1, 2])"
267 |       ]
268 |      },
269 |      "execution_count": 10,
270 |      "metadata": {},
271 |      "output_type": "execute_result"
272 |     }
273 |    ],
274 |    "source": [
275 |     "# 除算により 0-9 までの階級を割り当てる\n",
276 |     "np.floor_divide(small_counts, 10)"
277 |    ]
278 |   },
279 |   {
280 |    "cell_type": "code",
281 |    "execution_count": 11,
282 |    "metadata": {},
283 |    "outputs": [
284 |     {
285 |      "data": {
286 |       "text/plain": [
287 |        "array([2., 3., 4., 1., 0., 2., 2., 3., 3., 4., 4., 1., 1., 3., 2., 2., 4.])"
288 |       ]
289 |      },
290 |      "execution_count": 11,
291 |      "metadata": {},
292 |      "output_type": "execute_result"
293 |     }
294 |    ],
295 |    "source": [
296 |     "# 複数の桁にまたがるカウントデータの配列\n",
297 |     "large_counts = [296, 8286, 64011, 80, 3, 725, 867, 2215, 7689, 11495, 91897, \n",
298 |     "                44, 28, 7971, 926, 122, 22222]\n",
299 |     "# 対数変換により指数幅の階級を割り当てる\n",
300 |     "np.floor(np.log10(large_counts))"
301 |    ]
302 |   },
303 |   {
304 |    "cell_type": "markdown",
305 |    "metadata": {},
306 |    "source": [
307 |     "## 例 2-4. Yelp ビジネスレビュー件数の十分位数を計算する"
308 |    ]
309 |   },
310 |   {
311 |    "cell_type": "code",
312 |    "execution_count": 12,
313 |    "metadata": {},
314 |    "outputs": [
315 |     {
316 |      "data": {
317 |       "text/plain": [
318 |        "0.1     3.0\n",
319 |        "0.2     3.0\n",
320 |        "0.3     4.0\n",
321 |        "0.4     5.0\n",
322 |        "0.5     6.0\n",
323 |        "0.6     8.0\n",
324 |        "0.7    12.0\n",
325 |        "0.8    23.0\n",
326 |        "0.9    50.0\n",
327 |        "Name: review_count, dtype: float64"
328 |       ]
329 |      },
330 |      "execution_count": 12,
331 |      "metadata": {},
332 |      "output_type": "execute_result"
333 |     }
334 |    ],
335 |    "source": [
336 |     "deciles = biz_df['review_count'].quantile([.1, .2, .3, .4, .5, .6, .7, .8, .9])\n",
337 |     "deciles"
338 |    ]
339 |   },
340 |   {
341 |    "cell_type": "code",
342 |    "execution_count": 13,
343 |    "metadata": {},
344 |    "outputs": [
345 |     {
346 |      "data": {
347 |       "text/plain": [
348 |        "Text(0, 0.5, 'Occurrence')"
349 |       ]
350 |      },
351 |      "execution_count": 13,
352 |      "metadata": {},
353 |      "output_type": "execute_result"
354 |     },
355 |     {
356 |      "data": {
357 |       "image/png": "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\n",
358 |       "text/plain": [
359 |        "<Figure size 432x288 with 1 Axes>"
360 |       ]
361 |      },
362 |      "metadata": {},
363 |      "output_type": "display_data"
364 |     }
365 |    ],
366 |    "source": [
367 |     "# ヒストグラムに十分位数を上書きする\n",
368 |     "sns.set_style('whitegrid')\n",
369 |     "fig, ax = plt.subplots()\n",
370 |     "biz_df['review_count'].hist(ax=ax, bins=100)\n",
371 |     "for pos in deciles:\n",
372 |     "    handle = plt.axvline(pos, color='r')\n",
373 |     "ax.legend([handle], ['deciles'], fontsize=14)\n",
374 |     "ax.set_yscale('log')\n",
375 |     "ax.set_xscale('log')\n",
376 |     "ax.tick_params(labelsize=14)\n",
377 |     "ax.set_xlabel('Review Count', fontsize=14)\n",
378 |     "ax.set_ylabel('Occurrence', fontsize=14)"
379 |    ]
380 |   },
381 |   {
382 |    "cell_type": "markdown",
383 |    "metadata": {},
384 |    "source": [
385 |     "## 例 2-5. 分位数によるカウントの離散化"
386 |    ]
387 |   },
388 |   {
389 |    "cell_type": "code",
390 |    "execution_count": 14,
391 |    "metadata": {},
392 |    "outputs": [
393 |     {
394 |      "data": {
395 |       "text/plain": [
396 |        "array([1, 2, 3, 0, 0, 1, 1, 2, 2, 3, 3, 0, 0, 2, 1, 0, 3])"
397 |       ]
398 |      },
399 |      "execution_count": 14,
400 |      "metadata": {},
401 |      "output_type": "execute_result"
402 |     }
403 |    ],
404 |    "source": [
405 |     "# 例2-3の large_count を引き続き使う\n",
406 |     "import pandas as pd\n",
407 |     "\n",
408 |     "# 四分位数に変換\n",
409 |     "pd.qcut(large_counts, 4, labels=False)"
410 |    ]
411 |   },
412 |   {
413 |    "cell_type": "code",
414 |    "execution_count": 15,
415 |    "metadata": {},
416 |    "outputs": [
417 |     {
418 |      "data": {
419 |       "text/plain": [
420 |        "0.25     122.0\n",
421 |        "0.50     926.0\n",
422 |        "0.75    8286.0\n",
423 |        "dtype: float64"
424 |       ]
425 |      },
426 |      "execution_count": 15,
427 |      "metadata": {},
428 |      "output_type": "execute_result"
429 |     }
430 |    ],
431 |    "source": [
432 |     "# 分位数の計算\n",
433 |     "large_counts_series = pd.Series(large_counts)\n",
434 |     "large_counts_series.quantile([0.25, 0.5, 0.75])"
435 |    ]
436 |   },
437 |   {
438 |    "cell_type": "markdown",
439 |    "metadata": {},
440 |    "source": [
441 |     "## 例 2-6. 対数変換の前後でレビュー件数のヒストグラムを比較する"
442 |    ]
443 |   },
444 |   {
445 |    "cell_type": "code",
446 |    "execution_count": 16,
447 |    "metadata": {},
448 |    "outputs": [
449 |     {
450 |      "data": {
451 |       "text/plain": [
452 |        "Text(0, 0.5, 'Occurrence')"
453 |       ]
454 |      },
455 |      "execution_count": 16,
456 |      "metadata": {},
457 |      "output_type": "execute_result"
458 |     },
459 |     {
460 |      "data": {
461 |       "image/png": "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\n",
462 |       "text/plain": [
463 |        "<Figure size 432x288 with 2 Axes>"
464 |       ]
465 |      },
466 |      "metadata": {},
467 |      "output_type": "display_data"
468 |     }
469 |    ],
470 |    "source": [
471 |     "import numpy as np\n",
472 |     "\n",
473 |     "# 例2-2で読み込んだ Yelp データセットの\n",
474 |     "# データフレーム biz_df を使用して、レビュー件数を対数変換する。\n",
475 |     "# レビュー件数 0 を対数変換してマイナス無限大になるのを防ぐために\n",
476 |     "# 対数変換の前に生データに 1 を加算していることに注意。\n",
477 |     "biz_df['log_review_count'] = np.log10(biz_df['review_count'] + 1)\n",
478 |     "\n",
479 |     "fig, (ax1, ax2) = plt.subplots(2,1)\n",
480 |     "biz_df['review_count'].hist(ax=ax1, bins=100)\n",
481 |     "ax1.tick_params(labelsize=14)\n",
482 |     "ax1.set_xlabel('review_count', fontsize=14)\n",
483 |     "ax1.set_ylabel('Occurrence', fontsize=14)\n",
484 |     "\n",
485 |     "biz_df['log_review_count'].hist(ax=ax2, bins=100)\n",
486 |     "ax2.tick_params(labelsize=14)\n",
487 |     "ax2.set_xlabel('log10(review_count))', fontsize=14)\n",
488 |     "ax2.set_ylabel('Occurrence', fontsize=14)"
489 |    ]
490 |   },
491 |   {
492 |    "cell_type": "code",
493 |    "execution_count": null,
494 |    "metadata": {},
495 |    "outputs": [],
496 |    "source": []
497 |   }
498 |  ],
499 |  "metadata": {
500 |   "kernelspec": {
501 |    "display_name": "Python 3",
502 |    "language": "python",
503 |    "name": "python3"
504 |   },
505 |   "language_info": {
506 |    "codemirror_mode": {
507 |     "name": "ipython",
508 |     "version": 3
509 |    },
510 |    "file_extension": ".py",
511 |    "mimetype": "text/x-python",
512 |    "name": "python",
513 |    "nbconvert_exporter": "python",
514 |    "pygments_lexer": "ipython3",
515 |    "version": "3.7.0"
516 |   }
517 |  },
518 |  "nbformat": 4,
519 |  "nbformat_minor": 2
520 | }
521 | 


--------------------------------------------------------------------------------