├── .gitignore
├── 007. machine learning ensemble - random forest.ipynb
├── 008. machine learning - ensemble boosting basic.ipynb
├── 009. XGboost, LightGBM.ipynb
├── 010. credit_card_fraud_basic.ipynb
├── 011. outlier, oversampling with credit card fraud_kaggle.ipynb
├── 012. stacking ensemble.ipynb
├── 013. Time series cointegration.ipynb
└── README.md


/.gitignore:
--------------------------------------------------------------------------------
  1 | 
  2 | # Created by https://www.gitignore.io/api/python,pycharm,jupyternotebooks
  3 | # Edit at https://www.gitignore.io/?templates=python,pycharm,jupyternotebooks
  4 | 
  5 | ### JupyterNotebooks ###
  6 | # gitignore template for Jupyter Notebooks
  7 | # website: http://jupyter.org/
  8 | 
  9 | .ipynb_checkpoints
 10 | */.ipynb_checkpoints/*
 11 | 
 12 | # IPython
 13 | profile_default/
 14 | ipython_config.py
 15 | 
 16 | # Remove previous ipynb_checkpoints
 17 | #   git rm -r .ipynb_checkpoints/
 18 | 
 19 | ### PyCharm ###
 20 | # Covers JetBrains IDEs: IntelliJ, RubyMine, PhpStorm, AppCode, PyCharm, CLion, Android Studio and WebStorm
 21 | # Reference: https://intellij-support.jetbrains.com/hc/en-us/articles/206544839
 22 | 
 23 | # User-specific stuff
 24 | .idea/**/workspace.xml
 25 | .idea/**/tasks.xml
 26 | .idea/**/usage.statistics.xml
 27 | .idea/**/dictionaries
 28 | .idea/**/shelf
 29 | 
 30 | # Generated files
 31 | .idea/**/contentModel.xml
 32 | 
 33 | # Sensitive or high-churn files
 34 | .idea/**/dataSources/
 35 | .idea/**/dataSources.ids
 36 | .idea/**/dataSources.local.xml
 37 | .idea/**/sqlDataSources.xml
 38 | .idea/**/dynamic.xml
 39 | .idea/**/uiDesigner.xml
 40 | .idea/**/dbnavigator.xml
 41 | 
 42 | # Gradle
 43 | .idea/**/gradle.xml
 44 | .idea/**/libraries
 45 | 
 46 | # Gradle and Maven with auto-import
 47 | # When using Gradle or Maven with auto-import, you should exclude module files,
 48 | # since they will be recreated, and may cause churn.  Uncomment if using
 49 | # auto-import.
 50 | # .idea/modules.xml
 51 | # .idea/*.iml
 52 | # .idea/modules
 53 | # *.iml
 54 | # *.ipr
 55 | 
 56 | # CMake
 57 | cmake-build-*/
 58 | 
 59 | # Mongo Explorer plugin
 60 | .idea/**/mongoSettings.xml
 61 | 
 62 | # File-based project format
 63 | *.iws
 64 | 
 65 | # IntelliJ
 66 | out/
 67 | 
 68 | # mpeltonen/sbt-idea plugin
 69 | .idea_modules/
 70 | 
 71 | # JIRA plugin
 72 | atlassian-ide-plugin.xml
 73 | 
 74 | # Cursive Clojure plugin
 75 | .idea/replstate.xml
 76 | 
 77 | # Crashlytics plugin (for Android Studio and IntelliJ)
 78 | com_crashlytics_export_strings.xml
 79 | crashlytics.properties
 80 | crashlytics-build.properties
 81 | fabric.properties
 82 | 
 83 | # Editor-based Rest Client
 84 | .idea/httpRequests
 85 | 
 86 | # Android studio 3.1+ serialized cache file
 87 | .idea/caches/build_file_checksums.ser
 88 | 
 89 | ### PyCharm Patch ###
 90 | # Comment Reason: https://github.com/joeblau/gitignore.io/issues/186#issuecomment-215987721
 91 | 
 92 | # *.iml
 93 | # modules.xml
 94 | # .idea/misc.xml
 95 | # *.ipr
 96 | 
 97 | # Sonarlint plugin
 98 | .idea/**/sonarlint/
 99 | 
100 | # SonarQube Plugin
101 | .idea/**/sonarIssues.xml
102 | 
103 | # Markdown Navigator plugin
104 | .idea/**/markdown-navigator.xml
105 | .idea/**/markdown-navigator/
106 | 
107 | ### Python ###
108 | # Byte-compiled / optimized / DLL files
109 | __pycache__/
110 | *.py[cod]
111 | *$py.class
112 | 
113 | # C extensions
114 | *.so
115 | 
116 | # Distribution / packaging
117 | .Python
118 | build/
119 | develop-eggs/
120 | dist/
121 | downloads/
122 | eggs/
123 | .eggs/
124 | lib/
125 | lib64/
126 | parts/
127 | sdist/
128 | var/
129 | wheels/
130 | pip-wheel-metadata/
131 | share/python-wheels/
132 | *.egg-info/
133 | .installed.cfg
134 | *.egg
135 | MANIFEST
136 | 
137 | # PyInstaller
138 | #  Usually these files are written by a python script from a template
139 | #  before PyInstaller builds the exe, so as to inject date/other infos into it.
140 | *.manifest
141 | *.spec
142 | 
143 | # Installer logs
144 | pip-log.txt
145 | pip-delete-this-directory.txt
146 | 
147 | # Unit test / coverage reports
148 | htmlcov/
149 | .tox/
150 | .nox/
151 | .coverage
152 | .coverage.*
153 | .cache
154 | nosetests.xml
155 | coverage.xml
156 | *.cover
157 | .hypothesis/
158 | .pytest_cache/
159 | 
160 | # Translations
161 | *.mo
162 | *.pot
163 | 
164 | # Scrapy stuff:
165 | .scrapy
166 | 
167 | # Sphinx documentation
168 | docs/_build/
169 | 
170 | # PyBuilder
171 | target/
172 | 
173 | # pyenv
174 | .python-version
175 | 
176 | # pipenv
177 | #   According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
178 | #   However, in case of collaboration, if having platform-specific dependencies or dependencies
179 | #   having no cross-platform support, pipenv may install dependencies that don't work, or not
180 | #   install all needed dependencies.
181 | #Pipfile.lock
182 | 
183 | # celery beat schedule file
184 | celerybeat-schedule
185 | 
186 | # SageMath parsed files
187 | *.sage.py
188 | 
189 | # Spyder project settings
190 | .spyderproject
191 | .spyproject
192 | 
193 | # Rope project settings
194 | .ropeproject
195 | 
196 | # Mr Developer
197 | .mr.developer.cfg
198 | .project
199 | .pydevproject
200 | 
201 | # mkdocs documentation
202 | /site
203 | 
204 | # mypy
205 | .mypy_cache/
206 | .dmypy.json
207 | dmypy.json
208 | 
209 | # Pyre type checker
210 | .pyre/
211 | 
212 | # End of https://www.gitignore.io/api/python,pycharm,jupyternotebooks
213 | 


--------------------------------------------------------------------------------
/007. machine learning ensemble - random forest.ipynb:
--------------------------------------------------------------------------------
  1 | {
  2 |  "cells": [
  3 |   {
  4 |    "cell_type": "markdown",
  5 |    "metadata": {},
  6 |    "source": [
  7 |     "# 출처\n",
  8 |     "\n",
  9 |     "- https://www.kaggle.com/arthurtok/introduction-to-ensembling-stacking-in-python\n",
 10 |     "- https://www.kaggle.com/lsjsj92/simple-titanic-kernel-82-for-beginner-like-me\n",
 11 |     "- https://www.kaggle.com/startupsci/titanic-data-science-solutions\n",
 12 |     "- https://www.kaggle.com/ash316/eda-to-prediction-dietanic\n",
 13 |     "- https://www.kaggle.com/mjbahmani/a-comprehensive-ml-workflow-with-python"
 14 |    ]
 15 |   },
 16 |   {
 17 |    "cell_type": "code",
 18 |    "execution_count": 1,
 19 |    "metadata": {},
 20 |    "outputs": [],
 21 |    "source": [
 22 |     "import pandas as pd\n",
 23 |     "import numpy as np\n",
 24 |     "import matplotlib.pyplot as plt\n",
 25 |     "import seaborn as sns\n",
 26 |     "\n",
 27 |     "from sklearn.tree import DecisionTreeClassifier\n",
 28 |     "from sklearn.ensemble import RandomForestClassifier\n",
 29 |     "from sklearn.model_selection import train_test_split, GridSearchCV\n",
 30 |     "from sklearn.metrics import accuracy_score"
 31 |    ]
 32 |   },
 33 |   {
 34 |    "cell_type": "code",
 35 |    "execution_count": 2,
 36 |    "metadata": {},
 37 |    "outputs": [
 38 |     {
 39 |      "data": {
 40 |       "text/html": [
 41 |        "<div>\n",
 42 |        "<style scoped>\n",
 43 |        "    .dataframe tbody tr th:only-of-type {\n",
 44 |        "        vertical-align: middle;\n",
 45 |        "    }\n",
 46 |        "\n",
 47 |        "    .dataframe tbody tr th {\n",
 48 |        "        vertical-align: top;\n",
 49 |        "    }\n",
 50 |        "\n",
 51 |        "    .dataframe thead th {\n",
 52 |        "        text-align: right;\n",
 53 |        "    }\n",
 54 |        "</style>\n",
 55 |        "<table border=\"1\" class=\"dataframe\">\n",
 56 |        "  <thead>\n",
 57 |        "    <tr style=\"text-align: right;\">\n",
 58 |        "      <th></th>\n",
 59 |        "      <th>PassengerId</th>\n",
 60 |        "      <th>Survived</th>\n",
 61 |        "      <th>Pclass</th>\n",
 62 |        "      <th>Name</th>\n",
 63 |        "      <th>Sex</th>\n",
 64 |        "      <th>Age</th>\n",
 65 |        "      <th>SibSp</th>\n",
 66 |        "      <th>Parch</th>\n",
 67 |        "      <th>Ticket</th>\n",
 68 |        "      <th>Fare</th>\n",
 69 |        "      <th>Cabin</th>\n",
 70 |        "      <th>Embarked</th>\n",
 71 |        "    </tr>\n",
 72 |        "  </thead>\n",
 73 |        "  <tbody>\n",
 74 |        "    <tr>\n",
 75 |        "      <th>0</th>\n",
 76 |        "      <td>1</td>\n",
 77 |        "      <td>0</td>\n",
 78 |        "      <td>3</td>\n",
 79 |        "      <td>Braund, Mr. Owen Harris</td>\n",
 80 |        "      <td>male</td>\n",
 81 |        "      <td>22.0</td>\n",
 82 |        "      <td>1</td>\n",
 83 |        "      <td>0</td>\n",
 84 |        "      <td>A/5 21171</td>\n",
 85 |        "      <td>7.2500</td>\n",
 86 |        "      <td>NaN</td>\n",
 87 |        "      <td>S</td>\n",
 88 |        "    </tr>\n",
 89 |        "    <tr>\n",
 90 |        "      <th>1</th>\n",
 91 |        "      <td>2</td>\n",
 92 |        "      <td>1</td>\n",
 93 |        "      <td>1</td>\n",
 94 |        "      <td>Cumings, Mrs. John Bradley (Florence Briggs Th...</td>\n",
 95 |        "      <td>female</td>\n",
 96 |        "      <td>38.0</td>\n",
 97 |        "      <td>1</td>\n",
 98 |        "      <td>0</td>\n",
 99 |        "      <td>PC 17599</td>\n",
100 |        "      <td>71.2833</td>\n",
101 |        "      <td>C85</td>\n",
102 |        "      <td>C</td>\n",
103 |        "    </tr>\n",
104 |        "    <tr>\n",
105 |        "      <th>2</th>\n",
106 |        "      <td>3</td>\n",
107 |        "      <td>1</td>\n",
108 |        "      <td>3</td>\n",
109 |        "      <td>Heikkinen, Miss. Laina</td>\n",
110 |        "      <td>female</td>\n",
111 |        "      <td>26.0</td>\n",
112 |        "      <td>0</td>\n",
113 |        "      <td>0</td>\n",
114 |        "      <td>STON/O2. 3101282</td>\n",
115 |        "      <td>7.9250</td>\n",
116 |        "      <td>NaN</td>\n",
117 |        "      <td>S</td>\n",
118 |        "    </tr>\n",
119 |        "    <tr>\n",
120 |        "      <th>3</th>\n",
121 |        "      <td>4</td>\n",
122 |        "      <td>1</td>\n",
123 |        "      <td>1</td>\n",
124 |        "      <td>Futrelle, Mrs. Jacques Heath (Lily May Peel)</td>\n",
125 |        "      <td>female</td>\n",
126 |        "      <td>35.0</td>\n",
127 |        "      <td>1</td>\n",
128 |        "      <td>0</td>\n",
129 |        "      <td>113803</td>\n",
130 |        "      <td>53.1000</td>\n",
131 |        "      <td>C123</td>\n",
132 |        "      <td>S</td>\n",
133 |        "    </tr>\n",
134 |        "    <tr>\n",
135 |        "      <th>4</th>\n",
136 |        "      <td>5</td>\n",
137 |        "      <td>0</td>\n",
138 |        "      <td>3</td>\n",
139 |        "      <td>Allen, Mr. William Henry</td>\n",
140 |        "      <td>male</td>\n",
141 |        "      <td>35.0</td>\n",
142 |        "      <td>0</td>\n",
143 |        "      <td>0</td>\n",
144 |        "      <td>373450</td>\n",
145 |        "      <td>8.0500</td>\n",
146 |        "      <td>NaN</td>\n",
147 |        "      <td>S</td>\n",
148 |        "    </tr>\n",
149 |        "  </tbody>\n",
150 |        "</table>\n",
151 |        "</div>"
152 |       ],
153 |       "text/plain": [
154 |        "   PassengerId  Survived  Pclass  \\\n",
155 |        "0            1         0       3   \n",
156 |        "1            2         1       1   \n",
157 |        "2            3         1       3   \n",
158 |        "3            4         1       1   \n",
159 |        "4            5         0       3   \n",
160 |        "\n",
161 |        "                                                Name     Sex   Age  SibSp  \\\n",
162 |        "0                            Braund, Mr. Owen Harris    male  22.0      1   \n",
163 |        "1  Cumings, Mrs. John Bradley (Florence Briggs Th...  female  38.0      1   \n",
164 |        "2                             Heikkinen, Miss. Laina  female  26.0      0   \n",
165 |        "3       Futrelle, Mrs. Jacques Heath (Lily May Peel)  female  35.0      1   \n",
166 |        "4                           Allen, Mr. William Henry    male  35.0      0   \n",
167 |        "\n",
168 |        "   Parch            Ticket     Fare Cabin Embarked  \n",
169 |        "0      0         A/5 21171   7.2500   NaN        S  \n",
170 |        "1      0          PC 17599  71.2833   C85        C  \n",
171 |        "2      0  STON/O2. 3101282   7.9250   NaN        S  \n",
172 |        "3      0            113803  53.1000  C123        S  \n",
173 |        "4      0            373450   8.0500   NaN        S  "
174 |       ]
175 |      },
176 |      "execution_count": 2,
177 |      "metadata": {},
178 |      "output_type": "execute_result"
179 |     }
180 |    ],
181 |    "source": [
182 |     "data = pd.read_csv('../datas/titanic/train.csv')\n",
183 |     "data.head()"
184 |    ]
185 |   },
186 |   {
187 |    "cell_type": "code",
188 |    "execution_count": 3,
189 |    "metadata": {},
190 |    "outputs": [
191 |     {
192 |      "data": {
193 |       "text/plain": [
194 |        "PassengerId      0\n",
195 |        "Survived         0\n",
196 |        "Pclass           0\n",
197 |        "Name             0\n",
198 |        "Sex              0\n",
199 |        "Age            177\n",
200 |        "SibSp            0\n",
201 |        "Parch            0\n",
202 |        "Ticket           0\n",
203 |        "Fare             0\n",
204 |        "Cabin          687\n",
205 |        "Embarked         2\n",
206 |        "dtype: int64"
207 |       ]
208 |      },
209 |      "execution_count": 3,
210 |      "metadata": {},
211 |      "output_type": "execute_result"
212 |     }
213 |    ],
214 |    "source": [
215 |     "data.isna().sum()"
216 |    ]
217 |   },
218 |   {
219 |    "cell_type": "code",
220 |    "execution_count": 4,
221 |    "metadata": {},
222 |    "outputs": [
223 |     {
224 |      "name": "stdout",
225 |      "output_type": "stream",
226 |      "text": [
227 |       "age null값 비율 : 19.865\n",
228 |       "cabin null값 비율 : 77.104\n"
229 |      ]
230 |     }
231 |    ],
232 |    "source": [
233 |     "print(\"age null값 비율 : {0:.3f}\".format((data['Age'].isna().sum() / len(data)) * 100))\n",
234 |     "print(\"cabin null값 비율 : {0:.3f}\".format((data['Cabin'].isna().sum() / len(data)) * 100 ))"
235 |    ]
236 |   },
237 |   {
238 |    "cell_type": "code",
239 |    "execution_count": null,
240 |    "metadata": {},
241 |    "outputs": [],
242 |    "source": []
243 |   },
244 |   {
245 |    "cell_type": "code",
246 |    "execution_count": 13,
247 |    "metadata": {},
248 |    "outputs": [],
249 |    "source": [
250 |     "data['Embarked'].fillna('S', inplace = True)\n",
251 |     "data['Fare'].fillna(0, inplace=True)\n",
252 |     "data['Fare'] = data['Fare'].map(lambda x : np.log(x) if x > 0 else 0)"
253 |    ]
254 |   },
255 |   {
256 |    "cell_type": "code",
257 |    "execution_count": 14,
258 |    "metadata": {},
259 |    "outputs": [],
260 |    "source": [
261 |     "data['Initial'] = data['Name'].str.extract('([A-Za-z]+)\\.')\n",
262 |     "data['Initial'].data(['Mlle','Mme','Ms','Dr','Major','Lady','Countess','Jonkheer','Col','Rev','Capt','Sir','Don', 'Dona'],['Miss','Miss','Miss','Mr','Mr','Mrs','Mrs','Other','Other','Other','Mr','Mr','Mr','Other'],inplace=True)\n",
263 |     "mapping = {\n",
264 |     "    \"Mr\":0,\n",
265 |     "    \"Miss\":1,\n",
266 |     "    \"Mrs\" : 1,\n",
267 |     "    \"Master\":2,\n",
268 |     "    \"Other\":3\n",
269 |     "}\n",
270 |     "\n",
271 |     "data['Initial'] = data['Initial'].map(mapping)\n"
272 |    ]
273 |   },
274 |   {
275 |    "cell_type": "code",
276 |    "execution_count": 15,
277 |    "metadata": {},
278 |    "outputs": [],
279 |    "source": [
280 |     "mapping_sex = {\n",
281 |     "    'male' : 0,\n",
282 |     "    'female': 1\n",
283 |     "}\n",
284 |     "\n",
285 |     "mapping_em = {\n",
286 |     "    'S' :0,\n",
287 |     "    'C' :1,\n",
288 |     "    'Q' :2\n",
289 |     "}\n",
290 |     "\n",
291 |     "\n",
292 |     "data['Sex'] = data['Sex'].map(mapping_sex)\n",
293 |     "data['Embarked'] = data['Embarked'].map(mapping_em)\n",
294 |     "\n",
295 |     "\n",
296 |     "data.drop(['PassengerId', \"Ticket\", \"Cabin\", \"Name\"], axis = 1, inplace = True)\n"
297 |    ]
298 |   },
299 |   {
300 |    "cell_type": "code",
301 |    "execution_count": 16,
302 |    "metadata": {},
303 |    "outputs": [
304 |     {
305 |      "data": {
306 |       "text/plain": [
307 |        "Initial\n",
308 |        "0    32.739609\n",
309 |        "1    27.834615\n",
310 |        "2     4.574167\n",
311 |        "3    45.888889\n",
312 |        "Name: Age, dtype: float64"
313 |       ]
314 |      },
315 |      "execution_count": 16,
316 |      "metadata": {},
317 |      "output_type": "execute_result"
318 |     }
319 |    ],
320 |    "source": [
321 |     "data.groupby('Initial')['Age'].mean()"
322 |    ]
323 |   },
324 |   {
325 |    "cell_type": "code",
326 |    "execution_count": 17,
327 |    "metadata": {},
328 |    "outputs": [],
329 |    "source": [
330 |     "data.loc[ (data['Age'].isnull()) & (data['Initial'] == 0), 'Age' ] = 32\n",
331 |     "data.loc[ (data['Age'].isnull()) & (data['Initial'] == 1), 'Age' ] = 28\n",
332 |     "data.loc[ (data['Age'].isnull()) & (data['Initial'] == 2), 'Age' ] = 5\n",
333 |     "data.loc[ (data['Age'].isnull()) & (data['Initial'] == 3), 'Age' ] = 45"
334 |    ]
335 |   },
336 |   {
337 |    "cell_type": "code",
338 |    "execution_count": 18,
339 |    "metadata": {},
340 |    "outputs": [
341 |     {
342 |      "data": {
343 |       "text/html": [
344 |        "<div>\n",
345 |        "<style scoped>\n",
346 |        "    .dataframe tbody tr th:only-of-type {\n",
347 |        "        vertical-align: middle;\n",
348 |        "    }\n",
349 |        "\n",
350 |        "    .dataframe tbody tr th {\n",
351 |        "        vertical-align: top;\n",
352 |        "    }\n",
353 |        "\n",
354 |        "    .dataframe thead th {\n",
355 |        "        text-align: right;\n",
356 |        "    }\n",
357 |        "</style>\n",
358 |        "<table border=\"1\" class=\"dataframe\">\n",
359 |        "  <thead>\n",
360 |        "    <tr style=\"text-align: right;\">\n",
361 |        "      <th></th>\n",
362 |        "      <th>Survived</th>\n",
363 |        "      <th>Pclass</th>\n",
364 |        "      <th>Sex</th>\n",
365 |        "      <th>Age</th>\n",
366 |        "      <th>SibSp</th>\n",
367 |        "      <th>Parch</th>\n",
368 |        "      <th>Fare</th>\n",
369 |        "      <th>Embarked</th>\n",
370 |        "      <th>Initial</th>\n",
371 |        "    </tr>\n",
372 |        "  </thead>\n",
373 |        "  <tbody>\n",
374 |        "    <tr>\n",
375 |        "      <th>0</th>\n",
376 |        "      <td>0</td>\n",
377 |        "      <td>3</td>\n",
378 |        "      <td>0</td>\n",
379 |        "      <td>22.0</td>\n",
380 |        "      <td>1</td>\n",
381 |        "      <td>0</td>\n",
382 |        "      <td>1.981001</td>\n",
383 |        "      <td>0</td>\n",
384 |        "      <td>0</td>\n",
385 |        "    </tr>\n",
386 |        "    <tr>\n",
387 |        "      <th>1</th>\n",
388 |        "      <td>1</td>\n",
389 |        "      <td>1</td>\n",
390 |        "      <td>1</td>\n",
391 |        "      <td>38.0</td>\n",
392 |        "      <td>1</td>\n",
393 |        "      <td>0</td>\n",
394 |        "      <td>4.266662</td>\n",
395 |        "      <td>1</td>\n",
396 |        "      <td>1</td>\n",
397 |        "    </tr>\n",
398 |        "    <tr>\n",
399 |        "      <th>2</th>\n",
400 |        "      <td>1</td>\n",
401 |        "      <td>3</td>\n",
402 |        "      <td>1</td>\n",
403 |        "      <td>26.0</td>\n",
404 |        "      <td>0</td>\n",
405 |        "      <td>0</td>\n",
406 |        "      <td>2.070022</td>\n",
407 |        "      <td>0</td>\n",
408 |        "      <td>1</td>\n",
409 |        "    </tr>\n",
410 |        "    <tr>\n",
411 |        "      <th>3</th>\n",
412 |        "      <td>1</td>\n",
413 |        "      <td>1</td>\n",
414 |        "      <td>1</td>\n",
415 |        "      <td>35.0</td>\n",
416 |        "      <td>1</td>\n",
417 |        "      <td>0</td>\n",
418 |        "      <td>3.972177</td>\n",
419 |        "      <td>0</td>\n",
420 |        "      <td>1</td>\n",
421 |        "    </tr>\n",
422 |        "    <tr>\n",
423 |        "      <th>4</th>\n",
424 |        "      <td>0</td>\n",
425 |        "      <td>3</td>\n",
426 |        "      <td>0</td>\n",
427 |        "      <td>35.0</td>\n",
428 |        "      <td>0</td>\n",
429 |        "      <td>0</td>\n",
430 |        "      <td>2.085672</td>\n",
431 |        "      <td>0</td>\n",
432 |        "      <td>0</td>\n",
433 |        "    </tr>\n",
434 |        "  </tbody>\n",
435 |        "</table>\n",
436 |        "</div>"
437 |       ],
438 |       "text/plain": [
439 |        "   Survived  Pclass  Sex   Age  SibSp  Parch      Fare  Embarked  Initial\n",
440 |        "0         0       3    0  22.0      1      0  1.981001         0        0\n",
441 |        "1         1       1    1  38.0      1      0  4.266662         1        1\n",
442 |        "2         1       3    1  26.0      0      0  2.070022         0        1\n",
443 |        "3         1       1    1  35.0      1      0  3.972177         0        1\n",
444 |        "4         0       3    0  35.0      0      0  2.085672         0        0"
445 |       ]
446 |      },
447 |      "execution_count": 18,
448 |      "metadata": {},
449 |      "output_type": "execute_result"
450 |     }
451 |    ],
452 |    "source": [
453 |     "data.head()"
454 |    ]
455 |   },
456 |   {
457 |    "cell_type": "code",
458 |    "execution_count": 19,
459 |    "metadata": {},
460 |    "outputs": [],
461 |    "source": [
462 |     "y = data['Survived']\n",
463 |     "X = data.drop('Survived', axis = 1)"
464 |    ]
465 |   },
466 |   {
467 |    "cell_type": "code",
468 |    "execution_count": 20,
469 |    "metadata": {},
470 |    "outputs": [],
471 |    "source": [
472 |     "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2, random_state=0)"
473 |    ]
474 |   },
475 |   {
476 |    "cell_type": "code",
477 |    "execution_count": 21,
478 |    "metadata": {},
479 |    "outputs": [
480 |     {
481 |      "name": "stderr",
482 |      "output_type": "stream",
483 |      "text": [
484 |       "d:\\anaconda3\\envs\\soojin\\lib\\site-packages\\sklearn\\ensemble\\forest.py:245: FutureWarning: The default value of n_estimators will change from 10 in version 0.20 to 100 in 0.22.\n",
485 |       "  \"10 in version 0.20 to 100 in 0.22.\", FutureWarning)\n"
486 |      ]
487 |     },
488 |     {
489 |      "data": {
490 |       "text/plain": [
491 |        "RandomForestClassifier(bootstrap=True, class_weight=None, criterion='gini',\n",
492 |        "                       max_depth=None, max_features='auto', max_leaf_nodes=None,\n",
493 |        "                       min_impurity_decrease=0.0, min_impurity_split=None,\n",
494 |        "                       min_samples_leaf=1, min_samples_split=2,\n",
495 |        "                       min_weight_fraction_leaf=0.0, n_estimators=10,\n",
496 |        "                       n_jobs=None, oob_score=False, random_state=0, verbose=0,\n",
497 |        "                       warm_start=False)"
498 |       ]
499 |      },
500 |      "execution_count": 21,
501 |      "metadata": {},
502 |      "output_type": "execute_result"
503 |     }
504 |    ],
505 |    "source": [
506 |     "rf = RandomForestClassifier(random_state=0)\n",
507 |     "rf.fit(X_train, y_train)"
508 |    ]
509 |   },
510 |   {
511 |    "cell_type": "code",
512 |    "execution_count": 22,
513 |    "metadata": {},
514 |    "outputs": [
515 |     {
516 |      "name": "stdout",
517 |      "output_type": "stream",
518 |      "text": [
519 |       "정확도 :0.810\n"
520 |      ]
521 |     }
522 |    ],
523 |    "source": [
524 |     "pred = rf.predict(X_test)\n",
525 |     "print(\"정확도 :{0:.3f}\".format(accuracy_score(y_test, pred)))"
526 |    ]
527 |   },
528 |   {
529 |    "cell_type": "code",
530 |    "execution_count": 23,
531 |    "metadata": {},
532 |    "outputs": [],
533 |    "source": [
534 |     "rf_param_grid = {\n",
535 |     "    'n_estimators' : [100, 200, 300],\n",
536 |     "    'max_depth' : [4, 6, 8, 10, 12],\n",
537 |     "    'min_samples_leaf' : [3, 5, 6, 7, 10],\n",
538 |     "    'min_samples_split' : [2, 3, 5, 7, 10]\n",
539 |     "}"
540 |    ]
541 |   },
542 |   {
543 |    "cell_type": "code",
544 |    "execution_count": 24,
545 |    "metadata": {},
546 |    "outputs": [
547 |     {
548 |      "name": "stderr",
549 |      "output_type": "stream",
550 |      "text": [
551 |       "d:\\anaconda3\\envs\\soojin\\lib\\site-packages\\sklearn\\model_selection\\_split.py:1978: FutureWarning: The default value of cv will change from 3 to 5 in version 0.22. Specify it explicitly to silence this warning.\n",
552 |       "  warnings.warn(CV_WARNING, FutureWarning)\n",
553 |       "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 12 concurrent workers.\n"
554 |      ]
555 |     },
556 |     {
557 |      "name": "stdout",
558 |      "output_type": "stream",
559 |      "text": [
560 |       "Fitting 3 folds for each of 128 candidates, totalling 384 fits\n"
561 |      ]
562 |     },
563 |     {
564 |      "name": "stderr",
565 |      "output_type": "stream",
566 |      "text": [
567 |       "[Parallel(n_jobs=-1)]: Done  26 tasks      | elapsed:    4.4s\n",
568 |       "[Parallel(n_jobs=-1)]: Done 176 tasks      | elapsed:    9.7s\n",
569 |       "[Parallel(n_jobs=-1)]: Done 384 out of 384 | elapsed:   16.7s finished\n"
570 |      ]
571 |     },
572 |     {
573 |      "data": {
574 |       "text/plain": [
575 |        "GridSearchCV(cv='warn', error_score='raise-deprecating',\n",
576 |        "             estimator=RandomForestClassifier(bootstrap=True, class_weight=None,\n",
577 |        "                                              criterion='gini', max_depth=None,\n",
578 |        "                                              max_features='auto',\n",
579 |        "                                              max_leaf_nodes=None,\n",
580 |        "                                              min_impurity_decrease=0.0,\n",
581 |        "                                              min_impurity_split=None,\n",
582 |        "                                              min_samples_leaf=1,\n",
583 |        "                                              min_samples_split=2,\n",
584 |        "                                              min_weight_fraction_leaf=0.0,\n",
585 |        "                                              n_estimators=10, n_jobs=None,\n",
586 |        "                                              oob_score=False, random_state=0,\n",
587 |        "                                              verbose=0, warm_start=False),\n",
588 |        "             iid='warn', n_jobs=-1,\n",
589 |        "             param_grid={'max_depth': [6, 8, 10, 12],\n",
590 |        "                         'min_samples_leaf': [3, 5, 7, 10],\n",
591 |        "                         'min_samples_split': [2, 3, 5, 10],\n",
592 |        "                         'n_estimators': [100, 200]},\n",
593 |        "             pre_dispatch='2*n_jobs', refit=True, return_train_score=False,\n",
594 |        "             scoring='accuracy', verbose=1)"
595 |       ]
596 |      },
597 |      "execution_count": 24,
598 |      "metadata": {},
599 |      "output_type": "execute_result"
600 |     }
601 |    ],
602 |    "source": [
603 |     "rf_grid = GridSearchCV(rf, param_grid = rf_param_grid, scoring=\"accuracy\", n_jobs= -1, verbose = 1)\n",
604 |     "rf_grid.fit(X_train, y_train)"
605 |    ]
606 |   },
607 |   {
608 |    "cell_type": "code",
609 |    "execution_count": 25,
610 |    "metadata": {},
611 |    "outputs": [
612 |     {
613 |      "name": "stdout",
614 |      "output_type": "stream",
615 |      "text": [
616 |       "최고 평균 정확도 : 0.8174\n",
617 |       "최고의 파라미터 :  {'max_depth': 8, 'min_samples_leaf': 3, 'min_samples_split': 2, 'n_estimators': 100}\n"
618 |      ]
619 |     }
620 |    ],
621 |    "source": [
622 |     "print(\"최고 평균 정확도 : {0:.4f}\".format(rf_grid.best_score_))\n",
623 |     "print(\"최고의 파라미터 : \", rf_grid.best_params_)"
624 |    ]
625 |   },
626 |   {
627 |    "cell_type": "code",
628 |    "execution_count": 26,
629 |    "metadata": {},
630 |    "outputs": [
631 |     {
632 |      "data": {
633 |       "text/plain": [
634 |        "Index(['mean_fit_time', 'std_fit_time', 'mean_score_time', 'std_score_time',\n",
635 |        "       'param_max_depth', 'param_min_samples_leaf', 'param_min_samples_split',\n",
636 |        "       'param_n_estimators', 'params', 'split0_test_score',\n",
637 |        "       'split1_test_score', 'split2_test_score', 'mean_test_score',\n",
638 |        "       'std_test_score', 'rank_test_score'],\n",
639 |        "      dtype='object')"
640 |       ]
641 |      },
642 |      "execution_count": 26,
643 |      "metadata": {},
644 |      "output_type": "execute_result"
645 |     }
646 |    ],
647 |    "source": [
648 |     "result = pd.DataFrame(rf_grid.cv_results_)\n",
649 |     "result.columns"
650 |    ]
651 |   },
652 |   {
653 |    "cell_type": "code",
654 |    "execution_count": 27,
655 |    "metadata": {},
656 |    "outputs": [],
657 |    "source": [
658 |     "result.sort_values(by=['rank_test_score'], inplace=True)"
659 |    ]
660 |   },
661 |   {
662 |    "cell_type": "code",
663 |    "execution_count": 28,
664 |    "metadata": {},
665 |    "outputs": [
666 |     {
667 |      "data": {
668 |       "text/html": [
669 |        "<div>\n",
670 |        "<style scoped>\n",
671 |        "    .dataframe tbody tr th:only-of-type {\n",
672 |        "        vertical-align: middle;\n",
673 |        "    }\n",
674 |        "\n",
675 |        "    .dataframe tbody tr th {\n",
676 |        "        vertical-align: top;\n",
677 |        "    }\n",
678 |        "\n",
679 |        "    .dataframe thead th {\n",
680 |        "        text-align: right;\n",
681 |        "    }\n",
682 |        "</style>\n",
683 |        "<table border=\"1\" class=\"dataframe\">\n",
684 |        "  <thead>\n",
685 |        "    <tr style=\"text-align: right;\">\n",
686 |        "      <th></th>\n",
687 |        "      <th>params</th>\n",
688 |        "      <th>mean_test_score</th>\n",
689 |        "      <th>rank_test_score</th>\n",
690 |        "    </tr>\n",
691 |        "  </thead>\n",
692 |        "  <tbody>\n",
693 |        "    <tr>\n",
694 |        "      <th>32</th>\n",
695 |        "      <td>{'max_depth': 8, 'min_samples_leaf': 3, 'min_s...</td>\n",
696 |        "      <td>0.817416</td>\n",
697 |        "      <td>1</td>\n",
698 |        "    </tr>\n",
699 |        "    <tr>\n",
700 |        "      <th>34</th>\n",
701 |        "      <td>{'max_depth': 8, 'min_samples_leaf': 3, 'min_s...</td>\n",
702 |        "      <td>0.817416</td>\n",
703 |        "      <td>1</td>\n",
704 |        "    </tr>\n",
705 |        "    <tr>\n",
706 |        "      <th>36</th>\n",
707 |        "      <td>{'max_depth': 8, 'min_samples_leaf': 3, 'min_s...</td>\n",
708 |        "      <td>0.817416</td>\n",
709 |        "      <td>1</td>\n",
710 |        "    </tr>\n",
711 |        "    <tr>\n",
712 |        "      <th>56</th>\n",
713 |        "      <td>{'max_depth': 8, 'min_samples_leaf': 10, 'min_...</td>\n",
714 |        "      <td>0.814607</td>\n",
715 |        "      <td>4</td>\n",
716 |        "    </tr>\n",
717 |        "    <tr>\n",
718 |        "      <th>105</th>\n",
719 |        "      <td>{'max_depth': 12, 'min_samples_leaf': 5, 'min_...</td>\n",
720 |        "      <td>0.814607</td>\n",
721 |        "      <td>4</td>\n",
722 |        "    </tr>\n",
723 |        "    <tr>\n",
724 |        "      <th>107</th>\n",
725 |        "      <td>{'max_depth': 12, 'min_samples_leaf': 5, 'min_...</td>\n",
726 |        "      <td>0.814607</td>\n",
727 |        "      <td>4</td>\n",
728 |        "    </tr>\n",
729 |        "    <tr>\n",
730 |        "      <th>68</th>\n",
731 |        "      <td>{'max_depth': 10, 'min_samples_leaf': 3, 'min_...</td>\n",
732 |        "      <td>0.814607</td>\n",
733 |        "      <td>4</td>\n",
734 |        "    </tr>\n",
735 |        "    <tr>\n",
736 |        "      <th>66</th>\n",
737 |        "      <td>{'max_depth': 10, 'min_samples_leaf': 3, 'min_...</td>\n",
738 |        "      <td>0.814607</td>\n",
739 |        "      <td>4</td>\n",
740 |        "    </tr>\n",
741 |        "    <tr>\n",
742 |        "      <th>111</th>\n",
743 |        "      <td>{'max_depth': 12, 'min_samples_leaf': 5, 'min_...</td>\n",
744 |        "      <td>0.814607</td>\n",
745 |        "      <td>4</td>\n",
746 |        "    </tr>\n",
747 |        "    <tr>\n",
748 |        "      <th>109</th>\n",
749 |        "      <td>{'max_depth': 12, 'min_samples_leaf': 5, 'min_...</td>\n",
750 |        "      <td>0.814607</td>\n",
751 |        "      <td>4</td>\n",
752 |        "    </tr>\n",
753 |        "  </tbody>\n",
754 |        "</table>\n",
755 |        "</div>"
756 |       ],
757 |       "text/plain": [
758 |        "                                                params  mean_test_score  \\\n",
759 |        "32   {'max_depth': 8, 'min_samples_leaf': 3, 'min_s...         0.817416   \n",
760 |        "34   {'max_depth': 8, 'min_samples_leaf': 3, 'min_s...         0.817416   \n",
761 |        "36   {'max_depth': 8, 'min_samples_leaf': 3, 'min_s...         0.817416   \n",
762 |        "56   {'max_depth': 8, 'min_samples_leaf': 10, 'min_...         0.814607   \n",
763 |        "105  {'max_depth': 12, 'min_samples_leaf': 5, 'min_...         0.814607   \n",
764 |        "107  {'max_depth': 12, 'min_samples_leaf': 5, 'min_...         0.814607   \n",
765 |        "68   {'max_depth': 10, 'min_samples_leaf': 3, 'min_...         0.814607   \n",
766 |        "66   {'max_depth': 10, 'min_samples_leaf': 3, 'min_...         0.814607   \n",
767 |        "111  {'max_depth': 12, 'min_samples_leaf': 5, 'min_...         0.814607   \n",
768 |        "109  {'max_depth': 12, 'min_samples_leaf': 5, 'min_...         0.814607   \n",
769 |        "\n",
770 |        "     rank_test_score  \n",
771 |        "32                 1  \n",
772 |        "34                 1  \n",
773 |        "36                 1  \n",
774 |        "56                 4  \n",
775 |        "105                4  \n",
776 |        "107                4  \n",
777 |        "68                 4  \n",
778 |        "66                 4  \n",
779 |        "111                4  \n",
780 |        "109                4  "
781 |       ]
782 |      },
783 |      "execution_count": 28,
784 |      "metadata": {},
785 |      "output_type": "execute_result"
786 |     }
787 |    ],
788 |    "source": [
789 |     "result[['params', 'mean_test_score', 'rank_test_score']].head(10)"
790 |    ]
791 |   },
792 |   {
793 |    "cell_type": "code",
794 |    "execution_count": 29,
795 |    "metadata": {},
796 |    "outputs": [
797 |     {
798 |      "name": "stdout",
799 |      "output_type": "stream",
800 |      "text": [
801 |       "정확도 : 0.8603\n"
802 |      ]
803 |     }
804 |    ],
805 |    "source": [
806 |     "model = rf_grid.best_estimator_\n",
807 |     "pred = model.predict(X_test)\n",
808 |     "acc = accuracy_score(y_test, pred)\n",
809 |     "print(\"정확도 : {0:.4f}\".format(acc))\n"
810 |    ]
811 |   },
812 |   {
813 |    "cell_type": "code",
814 |    "execution_count": 30,
815 |    "metadata": {},
816 |    "outputs": [],
817 |    "source": [
818 |     "feature_importances = model.feature_importances_"
819 |    ]
820 |   },
821 |   {
822 |    "cell_type": "code",
823 |    "execution_count": 31,
824 |    "metadata": {},
825 |    "outputs": [
826 |     {
827 |      "data": {
828 |       "image/png": "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\n",
829 |       "text/plain": [
830 |        "<Figure size 864x720 with 1 Axes>"
831 |       ]
832 |      },
833 |      "metadata": {
834 |       "needs_background": "light"
835 |      },
836 |      "output_type": "display_data"
837 |     }
838 |    ],
839 |    "source": [
840 |     "ft_importances = pd.Series(feature_importances, index = X_train.columns)\n",
841 |     "ft_importances = ft_importances.sort_values(ascending=False)\n",
842 |     "\n",
843 |     "plt.figure(figsize=(12, 10))\n",
844 |     "plt.title(\"feature importances\")\n",
845 |     "sns.barplot(x=ft_importances, y = X_train.columns)\n",
846 |     "plt.show()\n"
847 |    ]
848 |   },
849 |   {
850 |    "cell_type": "code",
851 |    "execution_count": null,
852 |    "metadata": {},
853 |    "outputs": [],
854 |    "source": []
855 |   }
856 |  ],
857 |  "metadata": {
858 |   "kernelspec": {
859 |    "display_name": "Python 3",
860 |    "language": "python",
861 |    "name": "python3"
862 |   },
863 |   "language_info": {
864 |    "codemirror_mode": {
865 |     "name": "ipython",
866 |     "version": 3
867 |    },
868 |    "file_extension": ".py",
869 |    "mimetype": "text/x-python",
870 |    "name": "python",
871 |    "nbconvert_exporter": "python",
872 |    "pygments_lexer": "ipython3",
873 |    "version": "3.6.9"
874 |   }
875 |  },
876 |  "nbformat": 4,
877 |  "nbformat_minor": 2
878 | }
879 | 


--------------------------------------------------------------------------------
/008. machine learning - ensemble boosting basic.ipynb:
--------------------------------------------------------------------------------
  1 | {
  2 |  "cells": [
  3 |   {
  4 |    "cell_type": "markdown",
  5 |    "metadata": {},
  6 |    "source": [
  7 |     "# 출처\n",
  8 |     "\n",
  9 |     "- https://www.kaggle.com/arthurtok/introduction-to-ensembling-stacking-in-python\n",
 10 |     "- https://www.kaggle.com/lsjsj92/simple-titanic-kernel-82-for-beginner-like-me\n",
 11 |     "- https://www.kaggle.com/startupsci/titanic-data-science-solutions\n",
 12 |     "- https://www.kaggle.com/ash316/eda-to-prediction-dietanic\n",
 13 |     "- https://www.kaggle.com/mjbahmani/a-comprehensive-ml-workflow-with-python"
 14 |    ]
 15 |   },
 16 |   {
 17 |    "cell_type": "code",
 18 |    "execution_count": 3,
 19 |    "metadata": {},
 20 |    "outputs": [],
 21 |    "source": [
 22 |     "import pandas as pd\n",
 23 |     "import numpy as np\n",
 24 |     "import matplotlib.pyplot as plt\n",
 25 |     "import seaborn as sns\n",
 26 |     "\n",
 27 |     "from sklearn.tree import DecisionTreeClassifier\n",
 28 |     "from sklearn.ensemble import RandomForestClassifier, GradientBoostingClassifier\n",
 29 |     "from sklearn.model_selection import train_test_split, GridSearchCV\n",
 30 |     "from sklearn.metrics import accuracy_score"
 31 |    ]
 32 |   },
 33 |   {
 34 |    "cell_type": "code",
 35 |    "execution_count": 4,
 36 |    "metadata": {},
 37 |    "outputs": [
 38 |     {
 39 |      "data": {
 40 |       "text/html": [
 41 |        "<div>\n",
 42 |        "<style scoped>\n",
 43 |        "    .dataframe tbody tr th:only-of-type {\n",
 44 |        "        vertical-align: middle;\n",
 45 |        "    }\n",
 46 |        "\n",
 47 |        "    .dataframe tbody tr th {\n",
 48 |        "        vertical-align: top;\n",
 49 |        "    }\n",
 50 |        "\n",
 51 |        "    .dataframe thead th {\n",
 52 |        "        text-align: right;\n",
 53 |        "    }\n",
 54 |        "</style>\n",
 55 |        "<table border=\"1\" class=\"dataframe\">\n",
 56 |        "  <thead>\n",
 57 |        "    <tr style=\"text-align: right;\">\n",
 58 |        "      <th></th>\n",
 59 |        "      <th>PassengerId</th>\n",
 60 |        "      <th>Survived</th>\n",
 61 |        "      <th>Pclass</th>\n",
 62 |        "      <th>Name</th>\n",
 63 |        "      <th>Sex</th>\n",
 64 |        "      <th>Age</th>\n",
 65 |        "      <th>SibSp</th>\n",
 66 |        "      <th>Parch</th>\n",
 67 |        "      <th>Ticket</th>\n",
 68 |        "      <th>Fare</th>\n",
 69 |        "      <th>Cabin</th>\n",
 70 |        "      <th>Embarked</th>\n",
 71 |        "    </tr>\n",
 72 |        "  </thead>\n",
 73 |        "  <tbody>\n",
 74 |        "    <tr>\n",
 75 |        "      <th>0</th>\n",
 76 |        "      <td>1</td>\n",
 77 |        "      <td>0</td>\n",
 78 |        "      <td>3</td>\n",
 79 |        "      <td>Braund, Mr. Owen Harris</td>\n",
 80 |        "      <td>male</td>\n",
 81 |        "      <td>22.0</td>\n",
 82 |        "      <td>1</td>\n",
 83 |        "      <td>0</td>\n",
 84 |        "      <td>A/5 21171</td>\n",
 85 |        "      <td>7.2500</td>\n",
 86 |        "      <td>NaN</td>\n",
 87 |        "      <td>S</td>\n",
 88 |        "    </tr>\n",
 89 |        "    <tr>\n",
 90 |        "      <th>1</th>\n",
 91 |        "      <td>2</td>\n",
 92 |        "      <td>1</td>\n",
 93 |        "      <td>1</td>\n",
 94 |        "      <td>Cumings, Mrs. John Bradley (Florence Briggs Th...</td>\n",
 95 |        "      <td>female</td>\n",
 96 |        "      <td>38.0</td>\n",
 97 |        "      <td>1</td>\n",
 98 |        "      <td>0</td>\n",
 99 |        "      <td>PC 17599</td>\n",
100 |        "      <td>71.2833</td>\n",
101 |        "      <td>C85</td>\n",
102 |        "      <td>C</td>\n",
103 |        "    </tr>\n",
104 |        "    <tr>\n",
105 |        "      <th>2</th>\n",
106 |        "      <td>3</td>\n",
107 |        "      <td>1</td>\n",
108 |        "      <td>3</td>\n",
109 |        "      <td>Heikkinen, Miss. Laina</td>\n",
110 |        "      <td>female</td>\n",
111 |        "      <td>26.0</td>\n",
112 |        "      <td>0</td>\n",
113 |        "      <td>0</td>\n",
114 |        "      <td>STON/O2. 3101282</td>\n",
115 |        "      <td>7.9250</td>\n",
116 |        "      <td>NaN</td>\n",
117 |        "      <td>S</td>\n",
118 |        "    </tr>\n",
119 |        "    <tr>\n",
120 |        "      <th>3</th>\n",
121 |        "      <td>4</td>\n",
122 |        "      <td>1</td>\n",
123 |        "      <td>1</td>\n",
124 |        "      <td>Futrelle, Mrs. Jacques Heath (Lily May Peel)</td>\n",
125 |        "      <td>female</td>\n",
126 |        "      <td>35.0</td>\n",
127 |        "      <td>1</td>\n",
128 |        "      <td>0</td>\n",
129 |        "      <td>113803</td>\n",
130 |        "      <td>53.1000</td>\n",
131 |        "      <td>C123</td>\n",
132 |        "      <td>S</td>\n",
133 |        "    </tr>\n",
134 |        "    <tr>\n",
135 |        "      <th>4</th>\n",
136 |        "      <td>5</td>\n",
137 |        "      <td>0</td>\n",
138 |        "      <td>3</td>\n",
139 |        "      <td>Allen, Mr. William Henry</td>\n",
140 |        "      <td>male</td>\n",
141 |        "      <td>35.0</td>\n",
142 |        "      <td>0</td>\n",
143 |        "      <td>0</td>\n",
144 |        "      <td>373450</td>\n",
145 |        "      <td>8.0500</td>\n",
146 |        "      <td>NaN</td>\n",
147 |        "      <td>S</td>\n",
148 |        "    </tr>\n",
149 |        "  </tbody>\n",
150 |        "</table>\n",
151 |        "</div>"
152 |       ],
153 |       "text/plain": [
154 |        "   PassengerId  Survived  Pclass  \\\n",
155 |        "0            1         0       3   \n",
156 |        "1            2         1       1   \n",
157 |        "2            3         1       3   \n",
158 |        "3            4         1       1   \n",
159 |        "4            5         0       3   \n",
160 |        "\n",
161 |        "                                                Name     Sex   Age  SibSp  \\\n",
162 |        "0                            Braund, Mr. Owen Harris    male  22.0      1   \n",
163 |        "1  Cumings, Mrs. John Bradley (Florence Briggs Th...  female  38.0      1   \n",
164 |        "2                             Heikkinen, Miss. Laina  female  26.0      0   \n",
165 |        "3       Futrelle, Mrs. Jacques Heath (Lily May Peel)  female  35.0      1   \n",
166 |        "4                           Allen, Mr. William Henry    male  35.0      0   \n",
167 |        "\n",
168 |        "   Parch            Ticket     Fare Cabin Embarked  \n",
169 |        "0      0         A/5 21171   7.2500   NaN        S  \n",
170 |        "1      0          PC 17599  71.2833   C85        C  \n",
171 |        "2      0  STON/O2. 3101282   7.9250   NaN        S  \n",
172 |        "3      0            113803  53.1000  C123        S  \n",
173 |        "4      0            373450   8.0500   NaN        S  "
174 |       ]
175 |      },
176 |      "execution_count": 4,
177 |      "metadata": {},
178 |      "output_type": "execute_result"
179 |     }
180 |    ],
181 |    "source": [
182 |     "data = pd.read_csv('../datas/titanic/train.csv')\n",
183 |     "data.head()"
184 |    ]
185 |   },
186 |   {
187 |    "cell_type": "code",
188 |    "execution_count": 5,
189 |    "metadata": {},
190 |    "outputs": [],
191 |    "source": [
192 |     "data['Embarked'].fillna('S', inplace = True)\n",
193 |     "data['Fare'].fillna(0, inplace=True)\n",
194 |     "data['Fare'] = data['Fare'].map(lambda x : np.log(x) if x > 0 else 0)"
195 |    ]
196 |   },
197 |   {
198 |    "cell_type": "code",
199 |    "execution_count": 6,
200 |    "metadata": {},
201 |    "outputs": [],
202 |    "source": [
203 |     "data['Initial'] = data['Name'].str.extract('([A-Za-z]+)\\.')\n",
204 |     "data['Initial'].replace(['Mlle','Mme','Ms','Dr','Major','Lady','Countess','Jonkheer','Col','Rev','Capt','Sir','Don', 'Dona'],['Miss','Miss','Miss','Mr','Mr','Mrs','Mrs','Other','Other','Other','Mr','Mr','Mr','Other'],inplace=True)\n",
205 |     "mapping = {\n",
206 |     "    \"Mr\":0,\n",
207 |     "    \"Miss\":1,\n",
208 |     "    \"Mrs\" : 1,\n",
209 |     "    \"Master\":2,\n",
210 |     "    \"Other\":3\n",
211 |     "}\n",
212 |     "\n",
213 |     "data['Initial'] = data['Initial'].map(mapping)\n"
214 |    ]
215 |   },
216 |   {
217 |    "cell_type": "code",
218 |    "execution_count": 7,
219 |    "metadata": {},
220 |    "outputs": [],
221 |    "source": [
222 |     "mapping_sex = {\n",
223 |     "    'male' : 0,\n",
224 |     "    'female': 1\n",
225 |     "}\n",
226 |     "\n",
227 |     "mapping_em = {\n",
228 |     "    'S' :0,\n",
229 |     "    'C' :1,\n",
230 |     "    'Q' :2\n",
231 |     "}\n",
232 |     "\n",
233 |     "\n",
234 |     "data['Sex'] = data['Sex'].map(mapping_sex)\n",
235 |     "data['Embarked'] = data['Embarked'].map(mapping_em)\n",
236 |     "\n",
237 |     "\n",
238 |     "data.drop(['PassengerId', \"Ticket\", \"Cabin\", \"Name\"], axis = 1, inplace = True)\n"
239 |    ]
240 |   },
241 |   {
242 |    "cell_type": "code",
243 |    "execution_count": 8,
244 |    "metadata": {},
245 |    "outputs": [
246 |     {
247 |      "data": {
248 |       "text/plain": [
249 |        "Initial\n",
250 |        "0    32.739609\n",
251 |        "1    27.834615\n",
252 |        "2     4.574167\n",
253 |        "3    45.888889\n",
254 |        "Name: Age, dtype: float64"
255 |       ]
256 |      },
257 |      "execution_count": 8,
258 |      "metadata": {},
259 |      "output_type": "execute_result"
260 |     }
261 |    ],
262 |    "source": [
263 |     "data.groupby('Initial')['Age'].mean()"
264 |    ]
265 |   },
266 |   {
267 |    "cell_type": "code",
268 |    "execution_count": 9,
269 |    "metadata": {},
270 |    "outputs": [],
271 |    "source": [
272 |     "data.loc[ (data['Age'].isnull()) & (data['Initial'] == 0), 'Age' ] = 32\n",
273 |     "data.loc[ (data['Age'].isnull()) & (data['Initial'] == 1), 'Age' ] = 28\n",
274 |     "data.loc[ (data['Age'].isnull()) & (data['Initial'] == 2), 'Age' ] = 5\n",
275 |     "data.loc[ (data['Age'].isnull()) & (data['Initial'] == 3), 'Age' ] = 45"
276 |    ]
277 |   },
278 |   {
279 |    "cell_type": "code",
280 |    "execution_count": 10,
281 |    "metadata": {},
282 |    "outputs": [
283 |     {
284 |      "data": {
285 |       "text/html": [
286 |        "<div>\n",
287 |        "<style scoped>\n",
288 |        "    .dataframe tbody tr th:only-of-type {\n",
289 |        "        vertical-align: middle;\n",
290 |        "    }\n",
291 |        "\n",
292 |        "    .dataframe tbody tr th {\n",
293 |        "        vertical-align: top;\n",
294 |        "    }\n",
295 |        "\n",
296 |        "    .dataframe thead th {\n",
297 |        "        text-align: right;\n",
298 |        "    }\n",
299 |        "</style>\n",
300 |        "<table border=\"1\" class=\"dataframe\">\n",
301 |        "  <thead>\n",
302 |        "    <tr style=\"text-align: right;\">\n",
303 |        "      <th></th>\n",
304 |        "      <th>Survived</th>\n",
305 |        "      <th>Pclass</th>\n",
306 |        "      <th>Sex</th>\n",
307 |        "      <th>Age</th>\n",
308 |        "      <th>SibSp</th>\n",
309 |        "      <th>Parch</th>\n",
310 |        "      <th>Fare</th>\n",
311 |        "      <th>Embarked</th>\n",
312 |        "      <th>Initial</th>\n",
313 |        "    </tr>\n",
314 |        "  </thead>\n",
315 |        "  <tbody>\n",
316 |        "    <tr>\n",
317 |        "      <th>0</th>\n",
318 |        "      <td>0</td>\n",
319 |        "      <td>3</td>\n",
320 |        "      <td>0</td>\n",
321 |        "      <td>22.0</td>\n",
322 |        "      <td>1</td>\n",
323 |        "      <td>0</td>\n",
324 |        "      <td>1.981001</td>\n",
325 |        "      <td>0</td>\n",
326 |        "      <td>0</td>\n",
327 |        "    </tr>\n",
328 |        "    <tr>\n",
329 |        "      <th>1</th>\n",
330 |        "      <td>1</td>\n",
331 |        "      <td>1</td>\n",
332 |        "      <td>1</td>\n",
333 |        "      <td>38.0</td>\n",
334 |        "      <td>1</td>\n",
335 |        "      <td>0</td>\n",
336 |        "      <td>4.266662</td>\n",
337 |        "      <td>1</td>\n",
338 |        "      <td>1</td>\n",
339 |        "    </tr>\n",
340 |        "    <tr>\n",
341 |        "      <th>2</th>\n",
342 |        "      <td>1</td>\n",
343 |        "      <td>3</td>\n",
344 |        "      <td>1</td>\n",
345 |        "      <td>26.0</td>\n",
346 |        "      <td>0</td>\n",
347 |        "      <td>0</td>\n",
348 |        "      <td>2.070022</td>\n",
349 |        "      <td>0</td>\n",
350 |        "      <td>1</td>\n",
351 |        "    </tr>\n",
352 |        "    <tr>\n",
353 |        "      <th>3</th>\n",
354 |        "      <td>1</td>\n",
355 |        "      <td>1</td>\n",
356 |        "      <td>1</td>\n",
357 |        "      <td>35.0</td>\n",
358 |        "      <td>1</td>\n",
359 |        "      <td>0</td>\n",
360 |        "      <td>3.972177</td>\n",
361 |        "      <td>0</td>\n",
362 |        "      <td>1</td>\n",
363 |        "    </tr>\n",
364 |        "    <tr>\n",
365 |        "      <th>4</th>\n",
366 |        "      <td>0</td>\n",
367 |        "      <td>3</td>\n",
368 |        "      <td>0</td>\n",
369 |        "      <td>35.0</td>\n",
370 |        "      <td>0</td>\n",
371 |        "      <td>0</td>\n",
372 |        "      <td>2.085672</td>\n",
373 |        "      <td>0</td>\n",
374 |        "      <td>0</td>\n",
375 |        "    </tr>\n",
376 |        "  </tbody>\n",
377 |        "</table>\n",
378 |        "</div>"
379 |       ],
380 |       "text/plain": [
381 |        "   Survived  Pclass  Sex   Age  SibSp  Parch      Fare  Embarked  Initial\n",
382 |        "0         0       3    0  22.0      1      0  1.981001         0        0\n",
383 |        "1         1       1    1  38.0      1      0  4.266662         1        1\n",
384 |        "2         1       3    1  26.0      0      0  2.070022         0        1\n",
385 |        "3         1       1    1  35.0      1      0  3.972177         0        1\n",
386 |        "4         0       3    0  35.0      0      0  2.085672         0        0"
387 |       ]
388 |      },
389 |      "execution_count": 10,
390 |      "metadata": {},
391 |      "output_type": "execute_result"
392 |     }
393 |    ],
394 |    "source": [
395 |     "data.head()"
396 |    ]
397 |   },
398 |   {
399 |    "cell_type": "code",
400 |    "execution_count": 11,
401 |    "metadata": {},
402 |    "outputs": [],
403 |    "source": [
404 |     "y = data['Survived']\n",
405 |     "X = data.drop('Survived', axis = 1)"
406 |    ]
407 |   },
408 |   {
409 |    "cell_type": "code",
410 |    "execution_count": 12,
411 |    "metadata": {},
412 |    "outputs": [],
413 |    "source": [
414 |     "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2, random_state=0)"
415 |    ]
416 |   },
417 |   {
418 |    "cell_type": "code",
419 |    "execution_count": 13,
420 |    "metadata": {},
421 |    "outputs": [
422 |     {
423 |      "name": "stderr",
424 |      "output_type": "stream",
425 |      "text": [
426 |       "d:\\anaconda3\\envs\\soojin\\lib\\site-packages\\sklearn\\ensemble\\forest.py:245: FutureWarning: The default value of n_estimators will change from 10 in version 0.20 to 100 in 0.22.\n",
427 |       "  \"10 in version 0.20 to 100 in 0.22.\", FutureWarning)\n"
428 |      ]
429 |     },
430 |     {
431 |      "data": {
432 |       "text/plain": [
433 |        "RandomForestClassifier(bootstrap=True, class_weight=None, criterion='gini',\n",
434 |        "                       max_depth=None, max_features='auto', max_leaf_nodes=None,\n",
435 |        "                       min_impurity_decrease=0.0, min_impurity_split=None,\n",
436 |        "                       min_samples_leaf=1, min_samples_split=2,\n",
437 |        "                       min_weight_fraction_leaf=0.0, n_estimators=10,\n",
438 |        "                       n_jobs=None, oob_score=False, random_state=0, verbose=0,\n",
439 |        "                       warm_start=False)"
440 |       ]
441 |      },
442 |      "execution_count": 13,
443 |      "metadata": {},
444 |      "output_type": "execute_result"
445 |     }
446 |    ],
447 |    "source": [
448 |     "rf = RandomForestClassifier(random_state=0)\n",
449 |     "rf.fit(X_train, y_train)"
450 |    ]
451 |   },
452 |   {
453 |    "cell_type": "code",
454 |    "execution_count": 14,
455 |    "metadata": {},
456 |    "outputs": [
457 |     {
458 |      "name": "stdout",
459 |      "output_type": "stream",
460 |      "text": [
461 |       "정확도 :0.810\n"
462 |      ]
463 |     }
464 |    ],
465 |    "source": [
466 |     "pred = rf.predict(X_test)\n",
467 |     "print(\"정확도 :{0:.3f}\".format(accuracy_score(y_test, pred)))"
468 |    ]
469 |   },
470 |   {
471 |    "cell_type": "code",
472 |    "execution_count": 11,
473 |    "metadata": {},
474 |    "outputs": [
475 |     {
476 |      "data": {
477 |       "text/plain": [
478 |        "GradientBoostingClassifier(criterion='friedman_mse', init=None,\n",
479 |        "                           learning_rate=0.1, loss='deviance', max_depth=3,\n",
480 |        "                           max_features=None, max_leaf_nodes=None,\n",
481 |        "                           min_impurity_decrease=0.0, min_impurity_split=None,\n",
482 |        "                           min_samples_leaf=1, min_samples_split=2,\n",
483 |        "                           min_weight_fraction_leaf=0.0, n_estimators=100,\n",
484 |        "                           n_iter_no_change=None, presort='auto',\n",
485 |        "                           random_state=0, subsample=1.0, tol=0.0001,\n",
486 |        "                           validation_fraction=0.1, verbose=0,\n",
487 |        "                           warm_start=False)"
488 |       ]
489 |      },
490 |      "execution_count": 11,
491 |      "metadata": {},
492 |      "output_type": "execute_result"
493 |     }
494 |    ],
495 |    "source": [
496 |     "gb = GradientBoostingClassifier(random_state=0)\n",
497 |     "gb.fit(X_train, y_train)"
498 |    ]
499 |   },
500 |   {
501 |    "cell_type": "code",
502 |    "execution_count": 12,
503 |    "metadata": {},
504 |    "outputs": [
505 |     {
506 |      "name": "stdout",
507 |      "output_type": "stream",
508 |      "text": [
509 |       "정확도 :0.832\n"
510 |      ]
511 |     }
512 |    ],
513 |    "source": [
514 |     "pred = gb.predict(X_test)\n",
515 |     "print(\"정확도 :{0:.3f}\".format(accuracy_score(y_test, pred)))"
516 |    ]
517 |   },
518 |   {
519 |    "cell_type": "code",
520 |    "execution_count": 13,
521 |    "metadata": {},
522 |    "outputs": [],
523 |    "source": [
524 |     "gb_param_grid = {\n",
525 |     "    'n_estimators' : [100, 200],\n",
526 |     "    'max_depth' : [6, 8, 10, 12],\n",
527 |     "    'min_samples_leaf' : [3, 5, 7, 10],\n",
528 |     "    'min_samples_split' : [2, 3, 5, 10],\n",
529 |     "    'learning_rate' : [0.05, 0.1, 0.2]\n",
530 |     "}"
531 |    ]
532 |   },
533 |   {
534 |    "cell_type": "code",
535 |    "execution_count": 14,
536 |    "metadata": {},
537 |    "outputs": [
538 |     {
539 |      "name": "stderr",
540 |      "output_type": "stream",
541 |      "text": [
542 |       "d:\\anaconda3\\envs\\soojin\\lib\\site-packages\\sklearn\\model_selection\\_split.py:1978: FutureWarning: The default value of cv will change from 3 to 5 in version 0.22. Specify it explicitly to silence this warning.\n",
543 |       "  warnings.warn(CV_WARNING, FutureWarning)\n",
544 |       "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 12 concurrent workers.\n"
545 |      ]
546 |     },
547 |     {
548 |      "name": "stdout",
549 |      "output_type": "stream",
550 |      "text": [
551 |       "Fitting 3 folds for each of 384 candidates, totalling 1152 fits\n"
552 |      ]
553 |     },
554 |     {
555 |      "name": "stderr",
556 |      "output_type": "stream",
557 |      "text": [
558 |       "[Parallel(n_jobs=-1)]: Done  26 tasks      | elapsed:    4.6s\n",
559 |       "[Parallel(n_jobs=-1)]: Done 176 tasks      | elapsed:    9.4s\n",
560 |       "[Parallel(n_jobs=-1)]: Done 426 tasks      | elapsed:   20.0s\n",
561 |       "[Parallel(n_jobs=-1)]: Done 776 tasks      | elapsed:   34.5s\n",
562 |       "[Parallel(n_jobs=-1)]: Done 1152 out of 1152 | elapsed:   49.2s finished\n"
563 |      ]
564 |     },
565 |     {
566 |      "data": {
567 |       "text/plain": [
568 |        "GridSearchCV(cv='warn', error_score='raise-deprecating',\n",
569 |        "             estimator=GradientBoostingClassifier(criterion='friedman_mse',\n",
570 |        "                                                  init=None, learning_rate=0.1,\n",
571 |        "                                                  loss='deviance', max_depth=3,\n",
572 |        "                                                  max_features=None,\n",
573 |        "                                                  max_leaf_nodes=None,\n",
574 |        "                                                  min_impurity_decrease=0.0,\n",
575 |        "                                                  min_impurity_split=None,\n",
576 |        "                                                  min_samples_leaf=1,\n",
577 |        "                                                  min_samples_split=2,\n",
578 |        "                                                  min_weight_fraction_leaf=0.0,\n",
579 |        "                                                  n_estimators=100,\n",
580 |        "                                                  n_it...\n",
581 |        "                                                  random_state=0, subsample=1.0,\n",
582 |        "                                                  tol=0.0001,\n",
583 |        "                                                  validation_fraction=0.1,\n",
584 |        "                                                  verbose=0, warm_start=False),\n",
585 |        "             iid='warn', n_jobs=-1,\n",
586 |        "             param_grid={'learning_rate': [0.05, 0.1, 0.2],\n",
587 |        "                         'max_depth': [6, 8, 10, 12],\n",
588 |        "                         'min_samples_leaf': [3, 5, 7, 10],\n",
589 |        "                         'min_samples_split': [2, 3, 5, 10],\n",
590 |        "                         'n_estimators': [100, 200]},\n",
591 |        "             pre_dispatch='2*n_jobs', refit=True, return_train_score=False,\n",
592 |        "             scoring='accuracy', verbose=1)"
593 |       ]
594 |      },
595 |      "execution_count": 14,
596 |      "metadata": {},
597 |      "output_type": "execute_result"
598 |     }
599 |    ],
600 |    "source": [
601 |     "gb_grid = GridSearchCV(gb, param_grid = gb_param_grid, scoring=\"accuracy\", n_jobs= -1, verbose = 1)\n",
602 |     "gb_grid.fit(X_train, y_train)"
603 |    ]
604 |   },
605 |   {
606 |    "cell_type": "code",
607 |    "execution_count": 15,
608 |    "metadata": {},
609 |    "outputs": [
610 |     {
611 |      "name": "stdout",
612 |      "output_type": "stream",
613 |      "text": [
614 |       "최고 평균 정확도 : 0.8202\n",
615 |       "최고의 파라미터 :  {'learning_rate': 0.05, 'max_depth': 6, 'min_samples_leaf': 7, 'min_samples_split': 2, 'n_estimators': 100}\n"
616 |      ]
617 |     }
618 |    ],
619 |    "source": [
620 |     "print(\"최고 평균 정확도 : {0:.4f}\".format(gb_grid.best_score_))\n",
621 |     "print(\"최고의 파라미터 : \", gb_grid.best_params_)"
622 |    ]
623 |   },
624 |   {
625 |    "cell_type": "code",
626 |    "execution_count": 16,
627 |    "metadata": {},
628 |    "outputs": [
629 |     {
630 |      "data": {
631 |       "text/plain": [
632 |        "Index(['mean_fit_time', 'std_fit_time', 'mean_score_time', 'std_score_time',\n",
633 |        "       'param_learning_rate', 'param_max_depth', 'param_min_samples_leaf',\n",
634 |        "       'param_min_samples_split', 'param_n_estimators', 'params',\n",
635 |        "       'split0_test_score', 'split1_test_score', 'split2_test_score',\n",
636 |        "       'mean_test_score', 'std_test_score', 'rank_test_score'],\n",
637 |        "      dtype='object')"
638 |       ]
639 |      },
640 |      "execution_count": 16,
641 |      "metadata": {},
642 |      "output_type": "execute_result"
643 |     }
644 |    ],
645 |    "source": [
646 |     "result = pd.DataFrame(gb_grid.cv_results_)\n",
647 |     "result.columns"
648 |    ]
649 |   },
650 |   {
651 |    "cell_type": "code",
652 |    "execution_count": 17,
653 |    "metadata": {},
654 |    "outputs": [],
655 |    "source": [
656 |     "result.sort_values(by=['rank_test_score'], inplace=True)"
657 |    ]
658 |   },
659 |   {
660 |    "cell_type": "code",
661 |    "execution_count": 18,
662 |    "metadata": {},
663 |    "outputs": [
664 |     {
665 |      "data": {
666 |       "text/html": [
667 |        "<div>\n",
668 |        "<style scoped>\n",
669 |        "    .dataframe tbody tr th:only-of-type {\n",
670 |        "        vertical-align: middle;\n",
671 |        "    }\n",
672 |        "\n",
673 |        "    .dataframe tbody tr th {\n",
674 |        "        vertical-align: top;\n",
675 |        "    }\n",
676 |        "\n",
677 |        "    .dataframe thead th {\n",
678 |        "        text-align: right;\n",
679 |        "    }\n",
680 |        "</style>\n",
681 |        "<table border=\"1\" class=\"dataframe\">\n",
682 |        "  <thead>\n",
683 |        "    <tr style=\"text-align: right;\">\n",
684 |        "      <th></th>\n",
685 |        "      <th>params</th>\n",
686 |        "      <th>mean_test_score</th>\n",
687 |        "      <th>rank_test_score</th>\n",
688 |        "    </tr>\n",
689 |        "  </thead>\n",
690 |        "  <tbody>\n",
691 |        "    <tr>\n",
692 |        "      <th>22</th>\n",
693 |        "      <td>{'learning_rate': 0.05, 'max_depth': 6, 'min_s...</td>\n",
694 |        "      <td>0.820225</td>\n",
695 |        "      <td>1</td>\n",
696 |        "    </tr>\n",
697 |        "    <tr>\n",
698 |        "      <th>16</th>\n",
699 |        "      <td>{'learning_rate': 0.05, 'max_depth': 6, 'min_s...</td>\n",
700 |        "      <td>0.820225</td>\n",
701 |        "      <td>1</td>\n",
702 |        "    </tr>\n",
703 |        "    <tr>\n",
704 |        "      <th>18</th>\n",
705 |        "      <td>{'learning_rate': 0.05, 'max_depth': 6, 'min_s...</td>\n",
706 |        "      <td>0.820225</td>\n",
707 |        "      <td>1</td>\n",
708 |        "    </tr>\n",
709 |        "    <tr>\n",
710 |        "      <th>20</th>\n",
711 |        "      <td>{'learning_rate': 0.05, 'max_depth': 6, 'min_s...</td>\n",
712 |        "      <td>0.820225</td>\n",
713 |        "      <td>1</td>\n",
714 |        "    </tr>\n",
715 |        "    <tr>\n",
716 |        "      <th>58</th>\n",
717 |        "      <td>{'learning_rate': 0.05, 'max_depth': 8, 'min_s...</td>\n",
718 |        "      <td>0.818820</td>\n",
719 |        "      <td>5</td>\n",
720 |        "    </tr>\n",
721 |        "    <tr>\n",
722 |        "      <th>60</th>\n",
723 |        "      <td>{'learning_rate': 0.05, 'max_depth': 8, 'min_s...</td>\n",
724 |        "      <td>0.818820</td>\n",
725 |        "      <td>5</td>\n",
726 |        "    </tr>\n",
727 |        "    <tr>\n",
728 |        "      <th>62</th>\n",
729 |        "      <td>{'learning_rate': 0.05, 'max_depth': 8, 'min_s...</td>\n",
730 |        "      <td>0.818820</td>\n",
731 |        "      <td>5</td>\n",
732 |        "    </tr>\n",
733 |        "    <tr>\n",
734 |        "      <th>135</th>\n",
735 |        "      <td>{'learning_rate': 0.1, 'max_depth': 6, 'min_sa...</td>\n",
736 |        "      <td>0.818820</td>\n",
737 |        "      <td>5</td>\n",
738 |        "    </tr>\n",
739 |        "    <tr>\n",
740 |        "      <th>56</th>\n",
741 |        "      <td>{'learning_rate': 0.05, 'max_depth': 8, 'min_s...</td>\n",
742 |        "      <td>0.818820</td>\n",
743 |        "      <td>5</td>\n",
744 |        "    </tr>\n",
745 |        "    <tr>\n",
746 |        "      <th>120</th>\n",
747 |        "      <td>{'learning_rate': 0.05, 'max_depth': 12, 'min_...</td>\n",
748 |        "      <td>0.817416</td>\n",
749 |        "      <td>10</td>\n",
750 |        "    </tr>\n",
751 |        "  </tbody>\n",
752 |        "</table>\n",
753 |        "</div>"
754 |       ],
755 |       "text/plain": [
756 |        "                                                params  mean_test_score  \\\n",
757 |        "22   {'learning_rate': 0.05, 'max_depth': 6, 'min_s...         0.820225   \n",
758 |        "16   {'learning_rate': 0.05, 'max_depth': 6, 'min_s...         0.820225   \n",
759 |        "18   {'learning_rate': 0.05, 'max_depth': 6, 'min_s...         0.820225   \n",
760 |        "20   {'learning_rate': 0.05, 'max_depth': 6, 'min_s...         0.820225   \n",
761 |        "58   {'learning_rate': 0.05, 'max_depth': 8, 'min_s...         0.818820   \n",
762 |        "60   {'learning_rate': 0.05, 'max_depth': 8, 'min_s...         0.818820   \n",
763 |        "62   {'learning_rate': 0.05, 'max_depth': 8, 'min_s...         0.818820   \n",
764 |        "135  {'learning_rate': 0.1, 'max_depth': 6, 'min_sa...         0.818820   \n",
765 |        "56   {'learning_rate': 0.05, 'max_depth': 8, 'min_s...         0.818820   \n",
766 |        "120  {'learning_rate': 0.05, 'max_depth': 12, 'min_...         0.817416   \n",
767 |        "\n",
768 |        "     rank_test_score  \n",
769 |        "22                 1  \n",
770 |        "16                 1  \n",
771 |        "18                 1  \n",
772 |        "20                 1  \n",
773 |        "58                 5  \n",
774 |        "60                 5  \n",
775 |        "62                 5  \n",
776 |        "135                5  \n",
777 |        "56                 5  \n",
778 |        "120               10  "
779 |       ]
780 |      },
781 |      "execution_count": 18,
782 |      "metadata": {},
783 |      "output_type": "execute_result"
784 |     }
785 |    ],
786 |    "source": [
787 |     "result[['params', 'mean_test_score', 'rank_test_score']].head(10)"
788 |    ]
789 |   },
790 |   {
791 |    "cell_type": "code",
792 |    "execution_count": 20,
793 |    "metadata": {},
794 |    "outputs": [
795 |     {
796 |      "name": "stdout",
797 |      "output_type": "stream",
798 |      "text": [
799 |       "정확도 : 0.8492\n"
800 |      ]
801 |     }
802 |    ],
803 |    "source": [
804 |     "model = gb_grid.best_estimator_\n",
805 |     "pred = model.predict(X_test)\n",
806 |     "acc = accuracy_score(y_test, pred)\n",
807 |     "print(\"정확도 : {0:.4f}\".format(acc))\n"
808 |    ]
809 |   },
810 |   {
811 |    "cell_type": "code",
812 |    "execution_count": 21,
813 |    "metadata": {},
814 |    "outputs": [],
815 |    "source": [
816 |     "feature_importances = model.feature_importances_"
817 |    ]
818 |   },
819 |   {
820 |    "cell_type": "code",
821 |    "execution_count": 22,
822 |    "metadata": {},
823 |    "outputs": [
824 |     {
825 |      "data": {
826 |       "image/png": "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\n",
827 |       "text/plain": [
828 |        "<Figure size 864x720 with 1 Axes>"
829 |       ]
830 |      },
831 |      "metadata": {
832 |       "needs_background": "light"
833 |      },
834 |      "output_type": "display_data"
835 |     }
836 |    ],
837 |    "source": [
838 |     "ft_importances = pd.Series(feature_importances, index = X_train.columns)\n",
839 |     "ft_importances = ft_importances.sort_values(ascending=False)\n",
840 |     "\n",
841 |     "plt.figure(figsize=(12, 10))\n",
842 |     "plt.title(\"feature importances\")\n",
843 |     "sns.barplot(x=ft_importances, y = X_train.columns)\n",
844 |     "plt.show()\n"
845 |    ]
846 |   },
847 |   {
848 |    "cell_type": "code",
849 |    "execution_count": null,
850 |    "metadata": {},
851 |    "outputs": [],
852 |    "source": []
853 |   }
854 |  ],
855 |  "metadata": {
856 |   "kernelspec": {
857 |    "display_name": "Python 3",
858 |    "language": "python",
859 |    "name": "python3"
860 |   },
861 |   "language_info": {
862 |    "codemirror_mode": {
863 |     "name": "ipython",
864 |     "version": 3
865 |    },
866 |    "file_extension": ".py",
867 |    "mimetype": "text/x-python",
868 |    "name": "python",
869 |    "nbconvert_exporter": "python",
870 |    "pygments_lexer": "ipython3",
871 |    "version": "3.6.9"
872 |   }
873 |  },
874 |  "nbformat": 4,
875 |  "nbformat_minor": 2
876 | }
877 | 


--------------------------------------------------------------------------------
/009. XGboost, LightGBM.ipynb:
--------------------------------------------------------------------------------
   1 | {
   2 |  "cells": [
   3 |   {
   4 |    "cell_type": "markdown",
   5 |    "metadata": {},
   6 |    "source": [
   7 |     "# 출처 \n",
   8 |     "\n",
   9 |     "## xgboost\n",
  10 |     "\n",
  11 |     "- https://apple-rbox.tistory.com/6\n",
  12 |     "- https://brunch.co.kr/@snobberys/137\n"
  13 |    ]
  14 |   },
  15 |   {
  16 |    "cell_type": "markdown",
  17 |    "metadata": {},
  18 |    "source": [
  19 |     "## lightgbm\n",
  20 |     "\n",
  21 |     "- https://ko.raw3h.net/page/what-is-lightgbm-how-to-implement-it-how-to-fine-tune-the-parameters-5295f7/\n",
  22 |     "\n",
  23 |     "\n",
  24 |     "## 그 외\n",
  25 |     "\n",
  26 |     "- https://www.kaggle.com/shep312/applying-lightgbm-to-titanic-dataset\n",
  27 |     "- https://www.kaggle.com/suniliitb96/titanic-survival-prediction-using-xgboost\n"
  28 |    ]
  29 |   },
  30 |   {
  31 |    "cell_type": "code",
  32 |    "execution_count": 1,
  33 |    "metadata": {},
  34 |    "outputs": [],
  35 |    "source": [
  36 |     "import pandas as pd\n",
  37 |     "import numpy as np\n",
  38 |     "import matplotlib.pyplot as plt\n",
  39 |     "import seaborn as sns\n",
  40 |     "from xgboost import plot_importance\n",
  41 |     "from xgboost import XGBClassifier\n",
  42 |     "from sklearn.datasets import load_breast_cancer\n",
  43 |     "from sklearn.model_selection import train_test_split, GridSearchCV\n",
  44 |     "from sklearn.metrics import accuracy_score, f1_score, precision_score, recall_score, roc_auc_score, confusion_matrix"
  45 |    ]
  46 |   },
  47 |   {
  48 |    "cell_type": "code",
  49 |    "execution_count": 10,
  50 |    "metadata": {},
  51 |    "outputs": [],
  52 |    "source": [
  53 |     "def metrics(y_test, pred):\n",
  54 |     "    accuracy = accuracy_score(y_test, pred)\n",
  55 |     "    precision = precision_score(y_test, pred)\n",
  56 |     "    recall = recall_score(y_test, pred)\n",
  57 |     "    f1 = f1_score(y_test, pred)\n",
  58 |     "    roc_score = roc_auc_score(y_test, pred)\n",
  59 |     "    print('정확도 : {0:.2f}, 정밀도 : {1:.2f}, 재현율 : {2:.2f}'.format(accuracy, precision, recall))\n",
  60 |     "    print('f1-score : {0:.2f}, auc : {1:.2f}'.format(f1, roc_score))"
  61 |    ]
  62 |   },
  63 |   {
  64 |    "cell_type": "code",
  65 |    "execution_count": 3,
  66 |    "metadata": {},
  67 |    "outputs": [
  68 |     {
  69 |      "data": {
  70 |       "text/html": [
  71 |        "<div>\n",
  72 |        "<style scoped>\n",
  73 |        "    .dataframe tbody tr th:only-of-type {\n",
  74 |        "        vertical-align: middle;\n",
  75 |        "    }\n",
  76 |        "\n",
  77 |        "    .dataframe tbody tr th {\n",
  78 |        "        vertical-align: top;\n",
  79 |        "    }\n",
  80 |        "\n",
  81 |        "    .dataframe thead th {\n",
  82 |        "        text-align: right;\n",
  83 |        "    }\n",
  84 |        "</style>\n",
  85 |        "<table border=\"1\" class=\"dataframe\">\n",
  86 |        "  <thead>\n",
  87 |        "    <tr style=\"text-align: right;\">\n",
  88 |        "      <th></th>\n",
  89 |        "      <th>mean radius</th>\n",
  90 |        "      <th>mean texture</th>\n",
  91 |        "      <th>mean perimeter</th>\n",
  92 |        "      <th>mean area</th>\n",
  93 |        "      <th>mean smoothness</th>\n",
  94 |        "      <th>mean compactness</th>\n",
  95 |        "      <th>mean concavity</th>\n",
  96 |        "      <th>mean concave points</th>\n",
  97 |        "      <th>mean symmetry</th>\n",
  98 |        "      <th>mean fractal dimension</th>\n",
  99 |        "      <th>...</th>\n",
 100 |        "      <th>worst texture</th>\n",
 101 |        "      <th>worst perimeter</th>\n",
 102 |        "      <th>worst area</th>\n",
 103 |        "      <th>worst smoothness</th>\n",
 104 |        "      <th>worst compactness</th>\n",
 105 |        "      <th>worst concavity</th>\n",
 106 |        "      <th>worst concave points</th>\n",
 107 |        "      <th>worst symmetry</th>\n",
 108 |        "      <th>worst fractal dimension</th>\n",
 109 |        "      <th>target</th>\n",
 110 |        "    </tr>\n",
 111 |        "  </thead>\n",
 112 |        "  <tbody>\n",
 113 |        "    <tr>\n",
 114 |        "      <th>0</th>\n",
 115 |        "      <td>17.99</td>\n",
 116 |        "      <td>10.38</td>\n",
 117 |        "      <td>122.80</td>\n",
 118 |        "      <td>1001.0</td>\n",
 119 |        "      <td>0.11840</td>\n",
 120 |        "      <td>0.27760</td>\n",
 121 |        "      <td>0.3001</td>\n",
 122 |        "      <td>0.14710</td>\n",
 123 |        "      <td>0.2419</td>\n",
 124 |        "      <td>0.07871</td>\n",
 125 |        "      <td>...</td>\n",
 126 |        "      <td>17.33</td>\n",
 127 |        "      <td>184.60</td>\n",
 128 |        "      <td>2019.0</td>\n",
 129 |        "      <td>0.1622</td>\n",
 130 |        "      <td>0.6656</td>\n",
 131 |        "      <td>0.7119</td>\n",
 132 |        "      <td>0.2654</td>\n",
 133 |        "      <td>0.4601</td>\n",
 134 |        "      <td>0.11890</td>\n",
 135 |        "      <td>0</td>\n",
 136 |        "    </tr>\n",
 137 |        "    <tr>\n",
 138 |        "      <th>1</th>\n",
 139 |        "      <td>20.57</td>\n",
 140 |        "      <td>17.77</td>\n",
 141 |        "      <td>132.90</td>\n",
 142 |        "      <td>1326.0</td>\n",
 143 |        "      <td>0.08474</td>\n",
 144 |        "      <td>0.07864</td>\n",
 145 |        "      <td>0.0869</td>\n",
 146 |        "      <td>0.07017</td>\n",
 147 |        "      <td>0.1812</td>\n",
 148 |        "      <td>0.05667</td>\n",
 149 |        "      <td>...</td>\n",
 150 |        "      <td>23.41</td>\n",
 151 |        "      <td>158.80</td>\n",
 152 |        "      <td>1956.0</td>\n",
 153 |        "      <td>0.1238</td>\n",
 154 |        "      <td>0.1866</td>\n",
 155 |        "      <td>0.2416</td>\n",
 156 |        "      <td>0.1860</td>\n",
 157 |        "      <td>0.2750</td>\n",
 158 |        "      <td>0.08902</td>\n",
 159 |        "      <td>0</td>\n",
 160 |        "    </tr>\n",
 161 |        "    <tr>\n",
 162 |        "      <th>2</th>\n",
 163 |        "      <td>19.69</td>\n",
 164 |        "      <td>21.25</td>\n",
 165 |        "      <td>130.00</td>\n",
 166 |        "      <td>1203.0</td>\n",
 167 |        "      <td>0.10960</td>\n",
 168 |        "      <td>0.15990</td>\n",
 169 |        "      <td>0.1974</td>\n",
 170 |        "      <td>0.12790</td>\n",
 171 |        "      <td>0.2069</td>\n",
 172 |        "      <td>0.05999</td>\n",
 173 |        "      <td>...</td>\n",
 174 |        "      <td>25.53</td>\n",
 175 |        "      <td>152.50</td>\n",
 176 |        "      <td>1709.0</td>\n",
 177 |        "      <td>0.1444</td>\n",
 178 |        "      <td>0.4245</td>\n",
 179 |        "      <td>0.4504</td>\n",
 180 |        "      <td>0.2430</td>\n",
 181 |        "      <td>0.3613</td>\n",
 182 |        "      <td>0.08758</td>\n",
 183 |        "      <td>0</td>\n",
 184 |        "    </tr>\n",
 185 |        "    <tr>\n",
 186 |        "      <th>3</th>\n",
 187 |        "      <td>11.42</td>\n",
 188 |        "      <td>20.38</td>\n",
 189 |        "      <td>77.58</td>\n",
 190 |        "      <td>386.1</td>\n",
 191 |        "      <td>0.14250</td>\n",
 192 |        "      <td>0.28390</td>\n",
 193 |        "      <td>0.2414</td>\n",
 194 |        "      <td>0.10520</td>\n",
 195 |        "      <td>0.2597</td>\n",
 196 |        "      <td>0.09744</td>\n",
 197 |        "      <td>...</td>\n",
 198 |        "      <td>26.50</td>\n",
 199 |        "      <td>98.87</td>\n",
 200 |        "      <td>567.7</td>\n",
 201 |        "      <td>0.2098</td>\n",
 202 |        "      <td>0.8663</td>\n",
 203 |        "      <td>0.6869</td>\n",
 204 |        "      <td>0.2575</td>\n",
 205 |        "      <td>0.6638</td>\n",
 206 |        "      <td>0.17300</td>\n",
 207 |        "      <td>0</td>\n",
 208 |        "    </tr>\n",
 209 |        "    <tr>\n",
 210 |        "      <th>4</th>\n",
 211 |        "      <td>20.29</td>\n",
 212 |        "      <td>14.34</td>\n",
 213 |        "      <td>135.10</td>\n",
 214 |        "      <td>1297.0</td>\n",
 215 |        "      <td>0.10030</td>\n",
 216 |        "      <td>0.13280</td>\n",
 217 |        "      <td>0.1980</td>\n",
 218 |        "      <td>0.10430</td>\n",
 219 |        "      <td>0.1809</td>\n",
 220 |        "      <td>0.05883</td>\n",
 221 |        "      <td>...</td>\n",
 222 |        "      <td>16.67</td>\n",
 223 |        "      <td>152.20</td>\n",
 224 |        "      <td>1575.0</td>\n",
 225 |        "      <td>0.1374</td>\n",
 226 |        "      <td>0.2050</td>\n",
 227 |        "      <td>0.4000</td>\n",
 228 |        "      <td>0.1625</td>\n",
 229 |        "      <td>0.2364</td>\n",
 230 |        "      <td>0.07678</td>\n",
 231 |        "      <td>0</td>\n",
 232 |        "    </tr>\n",
 233 |        "  </tbody>\n",
 234 |        "</table>\n",
 235 |        "<p>5 rows × 31 columns</p>\n",
 236 |        "</div>"
 237 |       ],
 238 |       "text/plain": [
 239 |        "   mean radius  mean texture  mean perimeter  mean area  mean smoothness  \\\n",
 240 |        "0        17.99         10.38          122.80     1001.0          0.11840   \n",
 241 |        "1        20.57         17.77          132.90     1326.0          0.08474   \n",
 242 |        "2        19.69         21.25          130.00     1203.0          0.10960   \n",
 243 |        "3        11.42         20.38           77.58      386.1          0.14250   \n",
 244 |        "4        20.29         14.34          135.10     1297.0          0.10030   \n",
 245 |        "\n",
 246 |        "   mean compactness  mean concavity  mean concave points  mean symmetry  \\\n",
 247 |        "0           0.27760          0.3001              0.14710         0.2419   \n",
 248 |        "1           0.07864          0.0869              0.07017         0.1812   \n",
 249 |        "2           0.15990          0.1974              0.12790         0.2069   \n",
 250 |        "3           0.28390          0.2414              0.10520         0.2597   \n",
 251 |        "4           0.13280          0.1980              0.10430         0.1809   \n",
 252 |        "\n",
 253 |        "   mean fractal dimension  ...  worst texture  worst perimeter  worst area  \\\n",
 254 |        "0                 0.07871  ...          17.33           184.60      2019.0   \n",
 255 |        "1                 0.05667  ...          23.41           158.80      1956.0   \n",
 256 |        "2                 0.05999  ...          25.53           152.50      1709.0   \n",
 257 |        "3                 0.09744  ...          26.50            98.87       567.7   \n",
 258 |        "4                 0.05883  ...          16.67           152.20      1575.0   \n",
 259 |        "\n",
 260 |        "   worst smoothness  worst compactness  worst concavity  worst concave points  \\\n",
 261 |        "0            0.1622             0.6656           0.7119                0.2654   \n",
 262 |        "1            0.1238             0.1866           0.2416                0.1860   \n",
 263 |        "2            0.1444             0.4245           0.4504                0.2430   \n",
 264 |        "3            0.2098             0.8663           0.6869                0.2575   \n",
 265 |        "4            0.1374             0.2050           0.4000                0.1625   \n",
 266 |        "\n",
 267 |        "   worst symmetry  worst fractal dimension  target  \n",
 268 |        "0          0.4601                  0.11890       0  \n",
 269 |        "1          0.2750                  0.08902       0  \n",
 270 |        "2          0.3613                  0.08758       0  \n",
 271 |        "3          0.6638                  0.17300       0  \n",
 272 |        "4          0.2364                  0.07678       0  \n",
 273 |        "\n",
 274 |        "[5 rows x 31 columns]"
 275 |       ]
 276 |      },
 277 |      "execution_count": 3,
 278 |      "metadata": {},
 279 |      "output_type": "execute_result"
 280 |     }
 281 |    ],
 282 |    "source": [
 283 |     "data = load_breast_cancer()\n",
 284 |     "\n",
 285 |     "cancer = pd.DataFrame(data.data, columns = data.feature_names)\n",
 286 |     "cancer['target'] = data.target\n",
 287 |     "cancer.head()"
 288 |    ]
 289 |   },
 290 |   {
 291 |    "cell_type": "code",
 292 |    "execution_count": 5,
 293 |    "metadata": {},
 294 |    "outputs": [
 295 |     {
 296 |      "data": {
 297 |       "text/plain": [
 298 |        "1    357\n",
 299 |        "0    212\n",
 300 |        "Name: target, dtype: int64"
 301 |       ]
 302 |      },
 303 |      "execution_count": 5,
 304 |      "metadata": {},
 305 |      "output_type": "execute_result"
 306 |     }
 307 |    ],
 308 |    "source": [
 309 |     "cancer['target'].value_counts()"
 310 |    ]
 311 |   },
 312 |   {
 313 |    "cell_type": "code",
 314 |    "execution_count": 7,
 315 |    "metadata": {},
 316 |    "outputs": [
 317 |     {
 318 |      "name": "stdout",
 319 |      "output_type": "stream",
 320 |      "text": [
 321 |       "(455, 30) (114, 30)\n"
 322 |      ]
 323 |     }
 324 |    ],
 325 |    "source": [
 326 |     "y = cancer['target']\n",
 327 |     "X = cancer.drop('target', axis = 1)\n",
 328 |     "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2, random_state = 11)\n",
 329 |     "print(X_train.shape, X_test.shape)"
 330 |    ]
 331 |   },
 332 |   {
 333 |    "cell_type": "code",
 334 |    "execution_count": 8,
 335 |    "metadata": {},
 336 |    "outputs": [],
 337 |    "source": [
 338 |     "xgb = XGBClassifier(n_estimators=500, learning_rate = 0.1, max_depth = 4)\n",
 339 |     "xgb.fit(X_train, y_train)\n",
 340 |     "xgb_pred = xgb.predict(X_test)"
 341 |    ]
 342 |   },
 343 |   {
 344 |    "cell_type": "code",
 345 |    "execution_count": 12,
 346 |    "metadata": {},
 347 |    "outputs": [
 348 |     {
 349 |      "name": "stdout",
 350 |      "output_type": "stream",
 351 |      "text": [
 352 |       "정확도 : 0.99, 정밀도 : 0.99, 재현율 : 1.00\n",
 353 |       "f1-score : 0.99, auc : 0.99\n"
 354 |      ]
 355 |     }
 356 |    ],
 357 |    "source": [
 358 |     "metrics(y_test, xgb_pred)"
 359 |    ]
 360 |   },
 361 |   {
 362 |    "cell_type": "code",
 363 |    "execution_count": null,
 364 |    "metadata": {},
 365 |    "outputs": [],
 366 |    "source": []
 367 |   },
 368 |   {
 369 |    "cell_type": "code",
 370 |    "execution_count": null,
 371 |    "metadata": {},
 372 |    "outputs": [],
 373 |    "source": []
 374 |   },
 375 |   {
 376 |    "cell_type": "code",
 377 |    "execution_count": null,
 378 |    "metadata": {},
 379 |    "outputs": [],
 380 |    "source": []
 381 |   },
 382 |   {
 383 |    "cell_type": "code",
 384 |    "execution_count": 2,
 385 |    "metadata": {},
 386 |    "outputs": [
 387 |     {
 388 |      "data": {
 389 |       "text/html": [
 390 |        "<div>\n",
 391 |        "<style scoped>\n",
 392 |        "    .dataframe tbody tr th:only-of-type {\n",
 393 |        "        vertical-align: middle;\n",
 394 |        "    }\n",
 395 |        "\n",
 396 |        "    .dataframe tbody tr th {\n",
 397 |        "        vertical-align: top;\n",
 398 |        "    }\n",
 399 |        "\n",
 400 |        "    .dataframe thead th {\n",
 401 |        "        text-align: right;\n",
 402 |        "    }\n",
 403 |        "</style>\n",
 404 |        "<table border=\"1\" class=\"dataframe\">\n",
 405 |        "  <thead>\n",
 406 |        "    <tr style=\"text-align: right;\">\n",
 407 |        "      <th></th>\n",
 408 |        "      <th>PassengerId</th>\n",
 409 |        "      <th>Survived</th>\n",
 410 |        "      <th>Pclass</th>\n",
 411 |        "      <th>Name</th>\n",
 412 |        "      <th>Sex</th>\n",
 413 |        "      <th>Age</th>\n",
 414 |        "      <th>SibSp</th>\n",
 415 |        "      <th>Parch</th>\n",
 416 |        "      <th>Ticket</th>\n",
 417 |        "      <th>Fare</th>\n",
 418 |        "      <th>Cabin</th>\n",
 419 |        "      <th>Embarked</th>\n",
 420 |        "    </tr>\n",
 421 |        "  </thead>\n",
 422 |        "  <tbody>\n",
 423 |        "    <tr>\n",
 424 |        "      <th>0</th>\n",
 425 |        "      <td>1</td>\n",
 426 |        "      <td>0</td>\n",
 427 |        "      <td>3</td>\n",
 428 |        "      <td>Braund, Mr. Owen Harris</td>\n",
 429 |        "      <td>male</td>\n",
 430 |        "      <td>22.0</td>\n",
 431 |        "      <td>1</td>\n",
 432 |        "      <td>0</td>\n",
 433 |        "      <td>A/5 21171</td>\n",
 434 |        "      <td>7.2500</td>\n",
 435 |        "      <td>NaN</td>\n",
 436 |        "      <td>S</td>\n",
 437 |        "    </tr>\n",
 438 |        "    <tr>\n",
 439 |        "      <th>1</th>\n",
 440 |        "      <td>2</td>\n",
 441 |        "      <td>1</td>\n",
 442 |        "      <td>1</td>\n",
 443 |        "      <td>Cumings, Mrs. John Bradley (Florence Briggs Th...</td>\n",
 444 |        "      <td>female</td>\n",
 445 |        "      <td>38.0</td>\n",
 446 |        "      <td>1</td>\n",
 447 |        "      <td>0</td>\n",
 448 |        "      <td>PC 17599</td>\n",
 449 |        "      <td>71.2833</td>\n",
 450 |        "      <td>C85</td>\n",
 451 |        "      <td>C</td>\n",
 452 |        "    </tr>\n",
 453 |        "    <tr>\n",
 454 |        "      <th>2</th>\n",
 455 |        "      <td>3</td>\n",
 456 |        "      <td>1</td>\n",
 457 |        "      <td>3</td>\n",
 458 |        "      <td>Heikkinen, Miss. Laina</td>\n",
 459 |        "      <td>female</td>\n",
 460 |        "      <td>26.0</td>\n",
 461 |        "      <td>0</td>\n",
 462 |        "      <td>0</td>\n",
 463 |        "      <td>STON/O2. 3101282</td>\n",
 464 |        "      <td>7.9250</td>\n",
 465 |        "      <td>NaN</td>\n",
 466 |        "      <td>S</td>\n",
 467 |        "    </tr>\n",
 468 |        "    <tr>\n",
 469 |        "      <th>3</th>\n",
 470 |        "      <td>4</td>\n",
 471 |        "      <td>1</td>\n",
 472 |        "      <td>1</td>\n",
 473 |        "      <td>Futrelle, Mrs. Jacques Heath (Lily May Peel)</td>\n",
 474 |        "      <td>female</td>\n",
 475 |        "      <td>35.0</td>\n",
 476 |        "      <td>1</td>\n",
 477 |        "      <td>0</td>\n",
 478 |        "      <td>113803</td>\n",
 479 |        "      <td>53.1000</td>\n",
 480 |        "      <td>C123</td>\n",
 481 |        "      <td>S</td>\n",
 482 |        "    </tr>\n",
 483 |        "    <tr>\n",
 484 |        "      <th>4</th>\n",
 485 |        "      <td>5</td>\n",
 486 |        "      <td>0</td>\n",
 487 |        "      <td>3</td>\n",
 488 |        "      <td>Allen, Mr. William Henry</td>\n",
 489 |        "      <td>male</td>\n",
 490 |        "      <td>35.0</td>\n",
 491 |        "      <td>0</td>\n",
 492 |        "      <td>0</td>\n",
 493 |        "      <td>373450</td>\n",
 494 |        "      <td>8.0500</td>\n",
 495 |        "      <td>NaN</td>\n",
 496 |        "      <td>S</td>\n",
 497 |        "    </tr>\n",
 498 |        "  </tbody>\n",
 499 |        "</table>\n",
 500 |        "</div>"
 501 |       ],
 502 |       "text/plain": [
 503 |        "   PassengerId  Survived  Pclass  \\\n",
 504 |        "0            1         0       3   \n",
 505 |        "1            2         1       1   \n",
 506 |        "2            3         1       3   \n",
 507 |        "3            4         1       1   \n",
 508 |        "4            5         0       3   \n",
 509 |        "\n",
 510 |        "                                                Name     Sex   Age  SibSp  \\\n",
 511 |        "0                            Braund, Mr. Owen Harris    male  22.0      1   \n",
 512 |        "1  Cumings, Mrs. John Bradley (Florence Briggs Th...  female  38.0      1   \n",
 513 |        "2                             Heikkinen, Miss. Laina  female  26.0      0   \n",
 514 |        "3       Futrelle, Mrs. Jacques Heath (Lily May Peel)  female  35.0      1   \n",
 515 |        "4                           Allen, Mr. William Henry    male  35.0      0   \n",
 516 |        "\n",
 517 |        "   Parch            Ticket     Fare Cabin Embarked  \n",
 518 |        "0      0         A/5 21171   7.2500   NaN        S  \n",
 519 |        "1      0          PC 17599  71.2833   C85        C  \n",
 520 |        "2      0  STON/O2. 3101282   7.9250   NaN        S  \n",
 521 |        "3      0            113803  53.1000  C123        S  \n",
 522 |        "4      0            373450   8.0500   NaN        S  "
 523 |       ]
 524 |      },
 525 |      "execution_count": 2,
 526 |      "metadata": {},
 527 |      "output_type": "execute_result"
 528 |     }
 529 |    ],
 530 |    "source": [
 531 |     "data = pd.read_csv('./datas/titanic/train.csv')\n",
 532 |     "data.head()"
 533 |    ]
 534 |   },
 535 |   {
 536 |    "cell_type": "code",
 537 |    "execution_count": 3,
 538 |    "metadata": {},
 539 |    "outputs": [],
 540 |    "source": [
 541 |     "data['Embarked'].fillna('S', inplace = True)\n",
 542 |     "data['Fare'].fillna(0, inplace=True)\n",
 543 |     "data['Initial'] = data['Name'].str.extract('([A-Za-z]+)\\.')\n",
 544 |     "data['Initial'].replace(['Mlle','Mme','Ms','Dr','Major','Lady','Countess','Jonkheer','Col','Rev','Capt','Sir','Don', 'Dona'],['Miss','Miss','Miss','Mr','Mr','Mrs','Mrs','Other','Other','Other','Mr','Mr','Mr','Other'],inplace=True)\n"
 545 |    ]
 546 |   },
 547 |   {
 548 |    "cell_type": "code",
 549 |    "execution_count": 3,
 550 |    "metadata": {},
 551 |    "outputs": [],
 552 |    "source": [
 553 |     "\n",
 554 |     "mapping = {\n",
 555 |     "    \"Mr\":0,\n",
 556 |     "    \"Miss\":1,\n",
 557 |     "    \"Mrs\" : 1,\n",
 558 |     "    \"Master\":2,\n",
 559 |     "    \"Other\":3\n",
 560 |     "}\n",
 561 |     "\n",
 562 |     "\n",
 563 |     "\n",
 564 |     "mapping_sex = {\n",
 565 |     "    'male' : 0,\n",
 566 |     "    'female': 1\n",
 567 |     "}\n",
 568 |     "\n",
 569 |     "mapping_em = {\n",
 570 |     "    'S' :0,\n",
 571 |     "    'C' :1,\n",
 572 |     "    'Q' :2\n",
 573 |     "}\n"
 574 |    ]
 575 |   },
 576 |   {
 577 |    "cell_type": "code",
 578 |    "execution_count": 3,
 579 |    "metadata": {},
 580 |    "outputs": [],
 581 |    "source": [
 582 |     "\n",
 583 |     "data['Initial'] = data['Initial'].map(mapping)\n",
 584 |     "data['Fare'] = data['Fare'].map(lambda x : np.log(x) if x > 0 else 0)\n",
 585 |     "data['Sex'] = data['Sex'].map(mapping_sex)\n",
 586 |     "data['Embarked'] = data['Embarked'].map(mapping_em)\n"
 587 |    ]
 588 |   },
 589 |   {
 590 |    "cell_type": "code",
 591 |    "execution_count": 3,
 592 |    "metadata": {},
 593 |    "outputs": [],
 594 |    "source": [
 595 |     "data.drop(['PassengerId', \"Ticket\", \"Cabin\", \"Name\"], axis = 1, inplace = True)\n",
 596 |     "\n",
 597 |     "data.loc[ (data['Age'].isnull()) & (data['Initial'] == 0), 'Age' ] = 32\n",
 598 |     "data.loc[ (data['Age'].isnull()) & (data['Initial'] == 1), 'Age' ] = 28\n",
 599 |     "data.loc[ (data['Age'].isnull()) & (data['Initial'] == 2), 'Age' ] = 5\n",
 600 |     "data.loc[ (data['Age'].isnull()) & (data['Initial'] == 3), 'Age' ] = 45"
 601 |    ]
 602 |   },
 603 |   {
 604 |    "cell_type": "code",
 605 |    "execution_count": 4,
 606 |    "metadata": {},
 607 |    "outputs": [],
 608 |    "source": [
 609 |     "y = data['Survived']\n",
 610 |     "X = data.drop('Survived', axis = 1)\n",
 611 |     "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2, random_state=10)"
 612 |    ]
 613 |   },
 614 |   {
 615 |    "cell_type": "code",
 616 |    "execution_count": 13,
 617 |    "metadata": {},
 618 |    "outputs": [
 619 |     {
 620 |      "name": "stderr",
 621 |      "output_type": "stream",
 622 |      "text": [
 623 |       "d:\\anaconda3\\envs\\soojin\\lib\\site-packages\\sklearn\\model_selection\\_split.py:1978: FutureWarning: The default value of cv will change from 3 to 5 in version 0.22. Specify it explicitly to silence this warning.\n",
 624 |       "  warnings.warn(CV_WARNING, FutureWarning)\n",
 625 |       "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 12 concurrent workers.\n"
 626 |      ]
 627 |     },
 628 |     {
 629 |      "name": "stdout",
 630 |      "output_type": "stream",
 631 |      "text": [
 632 |       "Fitting 3 folds for each of 100 candidates, totalling 300 fits\n"
 633 |      ]
 634 |     },
 635 |     {
 636 |      "name": "stderr",
 637 |      "output_type": "stream",
 638 |      "text": [
 639 |       "[Parallel(n_jobs=-1)]: Done  26 tasks      | elapsed:    3.5s\n",
 640 |       "[Parallel(n_jobs=-1)]: Done 176 tasks      | elapsed:    8.8s\n",
 641 |       "[Parallel(n_jobs=-1)]: Done 300 out of 300 | elapsed:   12.8s finished\n"
 642 |      ]
 643 |     },
 644 |     {
 645 |      "data": {
 646 |       "text/plain": [
 647 |        "GridSearchCV(cv='warn', error_score='raise-deprecating',\n",
 648 |        "             estimator=XGBClassifier(base_score=0.5, booster='gbtree',\n",
 649 |        "                                     colsample_bylevel=1, colsample_bynode=1,\n",
 650 |        "                                     colsample_bytree=1, gamma=0,\n",
 651 |        "                                     learning_rate=0.1, max_delta_step=0,\n",
 652 |        "                                     max_depth=3, min_child_weight=1,\n",
 653 |        "                                     missing=None, n_estimators=100, n_jobs=1,\n",
 654 |        "                                     nthread=None, objective='binary:logistic',\n",
 655 |        "                                     random_state=0, reg_alpha=0, reg_lambda=1,\n",
 656 |        "                                     scale_pos_weight=1, seed=None, silent=None,\n",
 657 |        "                                     subsample=1, verbosity=1),\n",
 658 |        "             iid='warn', n_jobs=-1,\n",
 659 |        "             param_grid={'learning_rate': [0.01, 0.05, 0.1, 0.15, 0.2],\n",
 660 |        "                         'max_depth': [4, 6, 8, 10, 12],\n",
 661 |        "                         'n_estimators': [100, 200, 400, 600]},\n",
 662 |        "             pre_dispatch='2*n_jobs', refit=True, return_train_score=False,\n",
 663 |        "             scoring='accuracy', verbose=1)"
 664 |       ]
 665 |      },
 666 |      "execution_count": 13,
 667 |      "metadata": {},
 668 |      "output_type": "execute_result"
 669 |     }
 670 |    ],
 671 |    "source": [
 672 |     "xgb = XGBClassifier()\n",
 673 |     "\n",
 674 |     "xgb_param_grid = {\n",
 675 |     "    'n_estimators' : [100, 200, 400, 600],\n",
 676 |     "    'learning_rate' : [0.01, 0.05, 0.1, 0.15, 0.2],\n",
 677 |     "    'max_depth' : [4, 6, 8, 10, 12],\n",
 678 |     "}\n",
 679 |     "\n",
 680 |     "xgb_grid = GridSearchCV(xgb, param_grid = xgb_param_grid, scoring=\"accuracy\", n_jobs= -1, verbose = 1)\n",
 681 |     "xgb_grid.fit(X_train, y_train)\n"
 682 |    ]
 683 |   },
 684 |   {
 685 |    "cell_type": "code",
 686 |    "execution_count": 14,
 687 |    "metadata": {},
 688 |    "outputs": [
 689 |     {
 690 |      "name": "stdout",
 691 |      "output_type": "stream",
 692 |      "text": [
 693 |       "최고 평균 정확도 : 0.8244\n",
 694 |       "최고의 파라미터 :  {'learning_rate': 0.01, 'max_depth': 10, 'n_estimators': 100}\n"
 695 |      ]
 696 |     }
 697 |    ],
 698 |    "source": [
 699 |     "print(\"최고 평균 정확도 : {0:.4f}\".format(xgb_grid.best_score_))\n",
 700 |     "print(\"최고의 파라미터 : \", xgb_grid.best_params_)"
 701 |    ]
 702 |   },
 703 |   {
 704 |    "cell_type": "code",
 705 |    "execution_count": 15,
 706 |    "metadata": {},
 707 |    "outputs": [],
 708 |    "source": [
 709 |     "result = pd.DataFrame(xgb_grid.cv_results_)\n",
 710 |     "result.sort_values(by=['rank_test_score'], inplace=True)"
 711 |    ]
 712 |   },
 713 |   {
 714 |    "cell_type": "code",
 715 |    "execution_count": 16,
 716 |    "metadata": {},
 717 |    "outputs": [
 718 |     {
 719 |      "data": {
 720 |       "text/html": [
 721 |        "<div>\n",
 722 |        "<style scoped>\n",
 723 |        "    .dataframe tbody tr th:only-of-type {\n",
 724 |        "        vertical-align: middle;\n",
 725 |        "    }\n",
 726 |        "\n",
 727 |        "    .dataframe tbody tr th {\n",
 728 |        "        vertical-align: top;\n",
 729 |        "    }\n",
 730 |        "\n",
 731 |        "    .dataframe thead th {\n",
 732 |        "        text-align: right;\n",
 733 |        "    }\n",
 734 |        "</style>\n",
 735 |        "<table border=\"1\" class=\"dataframe\">\n",
 736 |        "  <thead>\n",
 737 |        "    <tr style=\"text-align: right;\">\n",
 738 |        "      <th></th>\n",
 739 |        "      <th>params</th>\n",
 740 |        "      <th>mean_test_score</th>\n",
 741 |        "      <th>rank_test_score</th>\n",
 742 |        "    </tr>\n",
 743 |        "  </thead>\n",
 744 |        "  <tbody>\n",
 745 |        "    <tr>\n",
 746 |        "      <th>16</th>\n",
 747 |        "      <td>{'learning_rate': 0.01, 'max_depth': 12, 'n_es...</td>\n",
 748 |        "      <td>0.824438</td>\n",
 749 |        "      <td>1</td>\n",
 750 |        "    </tr>\n",
 751 |        "    <tr>\n",
 752 |        "      <th>12</th>\n",
 753 |        "      <td>{'learning_rate': 0.01, 'max_depth': 10, 'n_es...</td>\n",
 754 |        "      <td>0.824438</td>\n",
 755 |        "      <td>1</td>\n",
 756 |        "    </tr>\n",
 757 |        "    <tr>\n",
 758 |        "      <th>10</th>\n",
 759 |        "      <td>{'learning_rate': 0.01, 'max_depth': 8, 'n_est...</td>\n",
 760 |        "      <td>0.823034</td>\n",
 761 |        "      <td>3</td>\n",
 762 |        "    </tr>\n",
 763 |        "    <tr>\n",
 764 |        "      <th>6</th>\n",
 765 |        "      <td>{'learning_rate': 0.01, 'max_depth': 6, 'n_est...</td>\n",
 766 |        "      <td>0.823034</td>\n",
 767 |        "      <td>3</td>\n",
 768 |        "    </tr>\n",
 769 |        "    <tr>\n",
 770 |        "      <th>11</th>\n",
 771 |        "      <td>{'learning_rate': 0.01, 'max_depth': 8, 'n_est...</td>\n",
 772 |        "      <td>0.821629</td>\n",
 773 |        "      <td>5</td>\n",
 774 |        "    </tr>\n",
 775 |        "    <tr>\n",
 776 |        "      <th>14</th>\n",
 777 |        "      <td>{'learning_rate': 0.01, 'max_depth': 10, 'n_es...</td>\n",
 778 |        "      <td>0.820225</td>\n",
 779 |        "      <td>6</td>\n",
 780 |        "    </tr>\n",
 781 |        "    <tr>\n",
 782 |        "      <th>7</th>\n",
 783 |        "      <td>{'learning_rate': 0.01, 'max_depth': 6, 'n_est...</td>\n",
 784 |        "      <td>0.820225</td>\n",
 785 |        "      <td>6</td>\n",
 786 |        "    </tr>\n",
 787 |        "    <tr>\n",
 788 |        "      <th>8</th>\n",
 789 |        "      <td>{'learning_rate': 0.01, 'max_depth': 8, 'n_est...</td>\n",
 790 |        "      <td>0.820225</td>\n",
 791 |        "      <td>6</td>\n",
 792 |        "    </tr>\n",
 793 |        "    <tr>\n",
 794 |        "      <th>15</th>\n",
 795 |        "      <td>{'learning_rate': 0.01, 'max_depth': 10, 'n_es...</td>\n",
 796 |        "      <td>0.818820</td>\n",
 797 |        "      <td>9</td>\n",
 798 |        "    </tr>\n",
 799 |        "    <tr>\n",
 800 |        "      <th>20</th>\n",
 801 |        "      <td>{'learning_rate': 0.05, 'max_depth': 4, 'n_est...</td>\n",
 802 |        "      <td>0.818820</td>\n",
 803 |        "      <td>9</td>\n",
 804 |        "    </tr>\n",
 805 |        "  </tbody>\n",
 806 |        "</table>\n",
 807 |        "</div>"
 808 |       ],
 809 |       "text/plain": [
 810 |        "                                               params  mean_test_score  \\\n",
 811 |        "16  {'learning_rate': 0.01, 'max_depth': 12, 'n_es...         0.824438   \n",
 812 |        "12  {'learning_rate': 0.01, 'max_depth': 10, 'n_es...         0.824438   \n",
 813 |        "10  {'learning_rate': 0.01, 'max_depth': 8, 'n_est...         0.823034   \n",
 814 |        "6   {'learning_rate': 0.01, 'max_depth': 6, 'n_est...         0.823034   \n",
 815 |        "11  {'learning_rate': 0.01, 'max_depth': 8, 'n_est...         0.821629   \n",
 816 |        "14  {'learning_rate': 0.01, 'max_depth': 10, 'n_es...         0.820225   \n",
 817 |        "7   {'learning_rate': 0.01, 'max_depth': 6, 'n_est...         0.820225   \n",
 818 |        "8   {'learning_rate': 0.01, 'max_depth': 8, 'n_est...         0.820225   \n",
 819 |        "15  {'learning_rate': 0.01, 'max_depth': 10, 'n_es...         0.818820   \n",
 820 |        "20  {'learning_rate': 0.05, 'max_depth': 4, 'n_est...         0.818820   \n",
 821 |        "\n",
 822 |        "    rank_test_score  \n",
 823 |        "16                1  \n",
 824 |        "12                1  \n",
 825 |        "10                3  \n",
 826 |        "6                 3  \n",
 827 |        "11                5  \n",
 828 |        "14                6  \n",
 829 |        "7                 6  \n",
 830 |        "8                 6  \n",
 831 |        "15                9  \n",
 832 |        "20                9  "
 833 |       ]
 834 |      },
 835 |      "execution_count": 16,
 836 |      "metadata": {},
 837 |      "output_type": "execute_result"
 838 |     }
 839 |    ],
 840 |    "source": [
 841 |     "result[['params', 'mean_test_score', 'rank_test_score']].head(10)"
 842 |    ]
 843 |   },
 844 |   {
 845 |    "cell_type": "code",
 846 |    "execution_count": 5,
 847 |    "metadata": {},
 848 |    "outputs": [],
 849 |    "source": []
 850 |   },
 851 |   {
 852 |    "cell_type": "code",
 853 |    "execution_count": 5,
 854 |    "metadata": {},
 855 |    "outputs": [],
 856 |    "source": []
 857 |   },
 858 |   {
 859 |    "cell_type": "code",
 860 |    "execution_count": 7,
 861 |    "metadata": {},
 862 |    "outputs": [
 863 |     {
 864 |      "name": "stdout",
 865 |      "output_type": "stream",
 866 |      "text": [
 867 |       "[0]\tvalidation_0-logloss:0.643237\n",
 868 |       "Will train until validation_0-logloss hasn't improved in 100 rounds.\n",
 869 |       "[1]\tvalidation_0-logloss:0.600544\n",
 870 |       "[2]\tvalidation_0-logloss:0.567278\n",
 871 |       "[3]\tvalidation_0-logloss:0.539616\n",
 872 |       "[4]\tvalidation_0-logloss:0.515783\n",
 873 |       "[5]\tvalidation_0-logloss:0.493083\n",
 874 |       "[6]\tvalidation_0-logloss:0.475519\n",
 875 |       "[7]\tvalidation_0-logloss:0.460413\n",
 876 |       "[8]\tvalidation_0-logloss:0.447876\n",
 877 |       "[9]\tvalidation_0-logloss:0.434248\n",
 878 |       "[10]\tvalidation_0-logloss:0.424888\n",
 879 |       "[11]\tvalidation_0-logloss:0.417043\n",
 880 |       "[12]\tvalidation_0-logloss:0.410522\n",
 881 |       "[13]\tvalidation_0-logloss:0.40404\n",
 882 |       "[14]\tvalidation_0-logloss:0.399429\n",
 883 |       "[15]\tvalidation_0-logloss:0.393991\n",
 884 |       "[16]\tvalidation_0-logloss:0.390694\n",
 885 |       "[17]\tvalidation_0-logloss:0.386906\n",
 886 |       "[18]\tvalidation_0-logloss:0.383186\n",
 887 |       "[19]\tvalidation_0-logloss:0.379069\n",
 888 |       "[20]\tvalidation_0-logloss:0.375775\n",
 889 |       "[21]\tvalidation_0-logloss:0.374222\n",
 890 |       "[22]\tvalidation_0-logloss:0.372794\n",
 891 |       "[23]\tvalidation_0-logloss:0.373341\n",
 892 |       "[24]\tvalidation_0-logloss:0.370527\n",
 893 |       "[25]\tvalidation_0-logloss:0.369523\n",
 894 |       "[26]\tvalidation_0-logloss:0.369086\n",
 895 |       "[27]\tvalidation_0-logloss:0.369067\n",
 896 |       "[28]\tvalidation_0-logloss:0.367292\n",
 897 |       "[29]\tvalidation_0-logloss:0.366029\n",
 898 |       "[30]\tvalidation_0-logloss:0.365949\n",
 899 |       "[31]\tvalidation_0-logloss:0.364792\n",
 900 |       "[32]\tvalidation_0-logloss:0.365043\n",
 901 |       "[33]\tvalidation_0-logloss:0.365255\n",
 902 |       "[34]\tvalidation_0-logloss:0.364502\n",
 903 |       "[35]\tvalidation_0-logloss:0.36495\n",
 904 |       "[36]\tvalidation_0-logloss:0.3653\n",
 905 |       "[37]\tvalidation_0-logloss:0.365692\n",
 906 |       "[38]\tvalidation_0-logloss:0.364633\n",
 907 |       "[39]\tvalidation_0-logloss:0.365394\n",
 908 |       "[40]\tvalidation_0-logloss:0.366006\n",
 909 |       "[41]\tvalidation_0-logloss:0.364511\n",
 910 |       "[42]\tvalidation_0-logloss:0.362128\n",
 911 |       "[43]\tvalidation_0-logloss:0.363265\n",
 912 |       "[44]\tvalidation_0-logloss:0.362809\n",
 913 |       "[45]\tvalidation_0-logloss:0.361418\n",
 914 |       "[46]\tvalidation_0-logloss:0.361081\n",
 915 |       "[47]\tvalidation_0-logloss:0.362271\n",
 916 |       "[48]\tvalidation_0-logloss:0.360343\n",
 917 |       "[49]\tvalidation_0-logloss:0.360147\n",
 918 |       "[50]\tvalidation_0-logloss:0.359536\n",
 919 |       "[51]\tvalidation_0-logloss:0.360105\n",
 920 |       "[52]\tvalidation_0-logloss:0.359863\n",
 921 |       "[53]\tvalidation_0-logloss:0.360054\n",
 922 |       "[54]\tvalidation_0-logloss:0.360457\n",
 923 |       "[55]\tvalidation_0-logloss:0.359963\n",
 924 |       "[56]\tvalidation_0-logloss:0.359591\n",
 925 |       "[57]\tvalidation_0-logloss:0.360042\n",
 926 |       "[58]\tvalidation_0-logloss:0.358606\n",
 927 |       "[59]\tvalidation_0-logloss:0.35847\n",
 928 |       "[60]\tvalidation_0-logloss:0.358429\n",
 929 |       "[61]\tvalidation_0-logloss:0.358046\n",
 930 |       "[62]\tvalidation_0-logloss:0.357865\n",
 931 |       "[63]\tvalidation_0-logloss:0.356589\n",
 932 |       "[64]\tvalidation_0-logloss:0.356376\n",
 933 |       "[65]\tvalidation_0-logloss:0.357027\n",
 934 |       "[66]\tvalidation_0-logloss:0.356924\n",
 935 |       "[67]\tvalidation_0-logloss:0.357237\n",
 936 |       "[68]\tvalidation_0-logloss:0.358427\n",
 937 |       "[69]\tvalidation_0-logloss:0.358904\n",
 938 |       "[70]\tvalidation_0-logloss:0.356838\n",
 939 |       "[71]\tvalidation_0-logloss:0.355709\n",
 940 |       "[72]\tvalidation_0-logloss:0.356185\n",
 941 |       "[73]\tvalidation_0-logloss:0.357439\n",
 942 |       "[74]\tvalidation_0-logloss:0.356952\n",
 943 |       "[75]\tvalidation_0-logloss:0.356894\n",
 944 |       "[76]\tvalidation_0-logloss:0.357164\n",
 945 |       "[77]\tvalidation_0-logloss:0.35748\n",
 946 |       "[78]\tvalidation_0-logloss:0.357296\n",
 947 |       "[79]\tvalidation_0-logloss:0.357984\n",
 948 |       "[80]\tvalidation_0-logloss:0.357816\n",
 949 |       "[81]\tvalidation_0-logloss:0.358238\n",
 950 |       "[82]\tvalidation_0-logloss:0.358398\n",
 951 |       "[83]\tvalidation_0-logloss:0.358424\n",
 952 |       "[84]\tvalidation_0-logloss:0.358912\n",
 953 |       "[85]\tvalidation_0-logloss:0.360025\n",
 954 |       "[86]\tvalidation_0-logloss:0.359234\n",
 955 |       "[87]\tvalidation_0-logloss:0.359403\n",
 956 |       "[88]\tvalidation_0-logloss:0.358514\n",
 957 |       "[89]\tvalidation_0-logloss:0.359621\n",
 958 |       "[90]\tvalidation_0-logloss:0.359716\n",
 959 |       "[91]\tvalidation_0-logloss:0.360305\n",
 960 |       "[92]\tvalidation_0-logloss:0.359297\n",
 961 |       "[93]\tvalidation_0-logloss:0.35923\n",
 962 |       "[94]\tvalidation_0-logloss:0.35925\n",
 963 |       "[95]\tvalidation_0-logloss:0.359636\n",
 964 |       "[96]\tvalidation_0-logloss:0.358746\n",
 965 |       "[97]\tvalidation_0-logloss:0.359995\n",
 966 |       "[98]\tvalidation_0-logloss:0.358856\n",
 967 |       "[99]\tvalidation_0-logloss:0.359269\n",
 968 |       "[100]\tvalidation_0-logloss:0.359495\n",
 969 |       "[101]\tvalidation_0-logloss:0.359534\n",
 970 |       "[102]\tvalidation_0-logloss:0.359903\n",
 971 |       "[103]\tvalidation_0-logloss:0.360073\n",
 972 |       "[104]\tvalidation_0-logloss:0.360139\n",
 973 |       "[105]\tvalidation_0-logloss:0.360796\n",
 974 |       "[106]\tvalidation_0-logloss:0.359293\n",
 975 |       "[107]\tvalidation_0-logloss:0.359956\n",
 976 |       "[108]\tvalidation_0-logloss:0.360043\n",
 977 |       "[109]\tvalidation_0-logloss:0.359125\n",
 978 |       "[110]\tvalidation_0-logloss:0.359315\n",
 979 |       "[111]\tvalidation_0-logloss:0.3594\n",
 980 |       "[112]\tvalidation_0-logloss:0.359811\n",
 981 |       "[113]\tvalidation_0-logloss:0.359921\n",
 982 |       "[114]\tvalidation_0-logloss:0.360095\n",
 983 |       "[115]\tvalidation_0-logloss:0.35926\n",
 984 |       "[116]\tvalidation_0-logloss:0.359522\n",
 985 |       "[117]\tvalidation_0-logloss:0.35992\n",
 986 |       "[118]\tvalidation_0-logloss:0.359175\n",
 987 |       "[119]\tvalidation_0-logloss:0.358587\n",
 988 |       "[120]\tvalidation_0-logloss:0.358692\n",
 989 |       "[121]\tvalidation_0-logloss:0.359066\n",
 990 |       "[122]\tvalidation_0-logloss:0.359215\n",
 991 |       "[123]\tvalidation_0-logloss:0.358593\n",
 992 |       "[124]\tvalidation_0-logloss:0.35855\n",
 993 |       "[125]\tvalidation_0-logloss:0.35841\n",
 994 |       "[126]\tvalidation_0-logloss:0.358248\n",
 995 |       "[127]\tvalidation_0-logloss:0.358388\n",
 996 |       "[128]\tvalidation_0-logloss:0.358489\n",
 997 |       "[129]\tvalidation_0-logloss:0.358913\n",
 998 |       "[130]\tvalidation_0-logloss:0.359169\n",
 999 |       "[131]\tvalidation_0-logloss:0.358706\n",
1000 |       "[132]\tvalidation_0-logloss:0.358846\n",
1001 |       "[133]\tvalidation_0-logloss:0.35899\n",
1002 |       "[134]\tvalidation_0-logloss:0.358574\n",
1003 |       "[135]\tvalidation_0-logloss:0.358431\n",
1004 |       "[136]\tvalidation_0-logloss:0.358572\n",
1005 |       "[137]\tvalidation_0-logloss:0.357526\n",
1006 |       "[138]\tvalidation_0-logloss:0.3576\n",
1007 |       "[139]\tvalidation_0-logloss:0.358176\n",
1008 |       "[140]\tvalidation_0-logloss:0.357707\n",
1009 |       "[141]\tvalidation_0-logloss:0.357483\n",
1010 |       "[142]\tvalidation_0-logloss:0.357542\n",
1011 |       "[143]\tvalidation_0-logloss:0.357489\n",
1012 |       "[144]\tvalidation_0-logloss:0.357366\n",
1013 |       "[145]\tvalidation_0-logloss:0.358119\n",
1014 |       "[146]\tvalidation_0-logloss:0.358145\n",
1015 |       "[147]\tvalidation_0-logloss:0.35822\n",
1016 |       "[148]\tvalidation_0-logloss:0.35805\n",
1017 |       "[149]\tvalidation_0-logloss:0.35899\n",
1018 |       "[150]\tvalidation_0-logloss:0.35882\n",
1019 |       "[151]\tvalidation_0-logloss:0.357895\n",
1020 |       "[152]\tvalidation_0-logloss:0.358154\n",
1021 |       "[153]\tvalidation_0-logloss:0.357417\n",
1022 |       "[154]\tvalidation_0-logloss:0.359365\n",
1023 |       "[155]\tvalidation_0-logloss:0.358782\n",
1024 |       "[156]\tvalidation_0-logloss:0.358195\n",
1025 |       "[157]\tvalidation_0-logloss:0.357697\n",
1026 |       "[158]\tvalidation_0-logloss:0.358491\n",
1027 |       "[159]\tvalidation_0-logloss:0.358627\n",
1028 |       "[160]\tvalidation_0-logloss:0.358216\n",
1029 |       "[161]\tvalidation_0-logloss:0.358591\n",
1030 |       "[162]\tvalidation_0-logloss:0.358682\n",
1031 |       "[163]\tvalidation_0-logloss:0.358732\n",
1032 |       "[164]\tvalidation_0-logloss:0.358995\n",
1033 |       "[165]\tvalidation_0-logloss:0.359204\n",
1034 |       "[166]\tvalidation_0-logloss:0.358358\n",
1035 |       "[167]\tvalidation_0-logloss:0.359008\n",
1036 |       "[168]\tvalidation_0-logloss:0.358891\n",
1037 |       "[169]\tvalidation_0-logloss:0.357869\n",
1038 |       "[170]\tvalidation_0-logloss:0.357907\n",
1039 |       "[171]\tvalidation_0-logloss:0.3576\n",
1040 |       "Stopping. Best iteration:\n",
1041 |       "[71]\tvalidation_0-logloss:0.355709\n",
1042 |       "\n"
1043 |      ]
1044 |     },
1045 |     {
1046 |      "data": {
1047 |       "text/plain": [
1048 |        "XGBClassifier(base_score=0.5, booster='gbtree', colsample_bylevel=1,\n",
1049 |        "              colsample_bynode=1, colsample_bytree=1, gamma=0,\n",
1050 |        "              learning_rate=0.1, max_delta_step=0, max_depth=3,\n",
1051 |        "              min_child_weight=1, missing=None, n_estimators=400, n_jobs=1,\n",
1052 |        "              nthread=None, objective='binary:logistic', random_state=0,\n",
1053 |        "              reg_alpha=0, reg_lambda=1, scale_pos_weight=1, seed=None,\n",
1054 |        "              silent=None, subsample=1, verbosity=1)"
1055 |       ]
1056 |      },
1057 |      "execution_count": 7,
1058 |      "metadata": {},
1059 |      "output_type": "execute_result"
1060 |     }
1061 |    ],
1062 |    "source": [
1063 |     "xgb = XGBClassifier(n_estimators=400, learning_rate = 0.1, max_depth = 3)\n",
1064 |     "evals = [(X_test, y_test)]\n",
1065 |     "xgb.fit(X_train, y_train, early_stopping_rounds = 100, eval_metric = \"logloss\", eval_set = evals, verbose = 1)"
1066 |    ]
1067 |   },
1068 |   {
1069 |    "cell_type": "code",
1070 |    "execution_count": 17,
1071 |    "metadata": {},
1072 |    "outputs": [
1073 |     {
1074 |      "name": "stdout",
1075 |      "output_type": "stream",
1076 |      "text": [
1077 |       "[0]\tvalidation_0-logloss:0.643237\n",
1078 |       "Will train until validation_0-logloss hasn't improved in 3 rounds.\n",
1079 |       "[1]\tvalidation_0-logloss:0.600544\n",
1080 |       "[2]\tvalidation_0-logloss:0.567278\n",
1081 |       "[3]\tvalidation_0-logloss:0.539616\n",
1082 |       "[4]\tvalidation_0-logloss:0.515783\n",
1083 |       "[5]\tvalidation_0-logloss:0.493083\n",
1084 |       "[6]\tvalidation_0-logloss:0.475519\n",
1085 |       "[7]\tvalidation_0-logloss:0.460413\n",
1086 |       "[8]\tvalidation_0-logloss:0.447876\n",
1087 |       "[9]\tvalidation_0-logloss:0.434248\n",
1088 |       "[10]\tvalidation_0-logloss:0.424888\n",
1089 |       "[11]\tvalidation_0-logloss:0.417043\n",
1090 |       "[12]\tvalidation_0-logloss:0.410522\n",
1091 |       "[13]\tvalidation_0-logloss:0.40404\n",
1092 |       "[14]\tvalidation_0-logloss:0.399429\n",
1093 |       "[15]\tvalidation_0-logloss:0.393991\n",
1094 |       "[16]\tvalidation_0-logloss:0.390694\n",
1095 |       "[17]\tvalidation_0-logloss:0.386906\n",
1096 |       "[18]\tvalidation_0-logloss:0.383186\n",
1097 |       "[19]\tvalidation_0-logloss:0.379069\n",
1098 |       "[20]\tvalidation_0-logloss:0.375775\n",
1099 |       "[21]\tvalidation_0-logloss:0.374222\n",
1100 |       "[22]\tvalidation_0-logloss:0.372794\n",
1101 |       "[23]\tvalidation_0-logloss:0.373341\n",
1102 |       "[24]\tvalidation_0-logloss:0.370527\n",
1103 |       "[25]\tvalidation_0-logloss:0.369523\n",
1104 |       "[26]\tvalidation_0-logloss:0.369086\n",
1105 |       "[27]\tvalidation_0-logloss:0.369067\n",
1106 |       "[28]\tvalidation_0-logloss:0.367292\n",
1107 |       "[29]\tvalidation_0-logloss:0.366029\n",
1108 |       "[30]\tvalidation_0-logloss:0.365949\n",
1109 |       "[31]\tvalidation_0-logloss:0.364792\n",
1110 |       "[32]\tvalidation_0-logloss:0.365043\n",
1111 |       "[33]\tvalidation_0-logloss:0.365255\n",
1112 |       "[34]\tvalidation_0-logloss:0.364502\n",
1113 |       "[35]\tvalidation_0-logloss:0.36495\n",
1114 |       "[36]\tvalidation_0-logloss:0.3653\n",
1115 |       "[37]\tvalidation_0-logloss:0.365692\n",
1116 |       "Stopping. Best iteration:\n",
1117 |       "[34]\tvalidation_0-logloss:0.364502\n",
1118 |       "\n"
1119 |      ]
1120 |     },
1121 |     {
1122 |      "data": {
1123 |       "text/plain": [
1124 |        "XGBClassifier(base_score=0.5, booster='gbtree', colsample_bylevel=1,\n",
1125 |        "              colsample_bynode=1, colsample_bytree=1, gamma=0,\n",
1126 |        "              learning_rate=0.1, max_delta_step=0, max_depth=3,\n",
1127 |        "              min_child_weight=1, missing=None, n_estimators=400, n_jobs=1,\n",
1128 |        "              nthread=None, objective='binary:logistic', random_state=0,\n",
1129 |        "              reg_alpha=0, reg_lambda=1, scale_pos_weight=1, seed=None,\n",
1130 |        "              silent=None, subsample=1, verbosity=1)"
1131 |       ]
1132 |      },
1133 |      "execution_count": 17,
1134 |      "metadata": {},
1135 |      "output_type": "execute_result"
1136 |     }
1137 |    ],
1138 |    "source": [
1139 |     "xgb = XGBClassifier(n_estimators=400, learning_rate = 0.1, max_depth = 3)\n",
1140 |     "evals = [(X_test, y_test)]\n",
1141 |     "xgb.fit(X_train, y_train, early_stopping_rounds = 3, eval_metric = \"logloss\", eval_set = evals, verbose = 1)"
1142 |    ]
1143 |   },
1144 |   {
1145 |    "cell_type": "code",
1146 |    "execution_count": 9,
1147 |    "metadata": {},
1148 |    "outputs": [
1149 |     {
1150 |      "data": {
1151 |       "text/plain": [
1152 |        "<matplotlib.axes._subplots.AxesSubplot at 0x24da5671cf8>"
1153 |       ]
1154 |      },
1155 |      "execution_count": 9,
1156 |      "metadata": {},
1157 |      "output_type": "execute_result"
1158 |     },
1159 |     {
1160 |      "data": {
1161 |       "image/png": "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\n",
1162 |       "text/plain": [
1163 |        "<Figure size 432x288 with 1 Axes>"
1164 |       ]
1165 |      },
1166 |      "metadata": {
1167 |       "needs_background": "light"
1168 |      },
1169 |      "output_type": "display_data"
1170 |     }
1171 |    ],
1172 |    "source": [
1173 |     "fig, ax = plt.subplots()\n",
1174 |     "plot_importance(xgb, ax=ax)"
1175 |    ]
1176 |   },
1177 |   {
1178 |    "cell_type": "code",
1179 |    "execution_count": null,
1180 |    "metadata": {},
1181 |    "outputs": [],
1182 |    "source": []
1183 |   },
1184 |   {
1185 |    "cell_type": "code",
1186 |    "execution_count": 6,
1187 |    "metadata": {},
1188 |    "outputs": [],
1189 |    "source": [
1190 |     "from lightgbm import LGBMClassifier, plot_importance"
1191 |    ]
1192 |   },
1193 |   {
1194 |    "cell_type": "code",
1195 |    "execution_count": 11,
1196 |    "metadata": {},
1197 |    "outputs": [
1198 |     {
1199 |      "data": {
1200 |       "text/plain": [
1201 |        "LGBMClassifier(boosting_type='gbdt', class_weight=None, colsample_bytree=1.0,\n",
1202 |        "               importance_type='split', learning_rate=0.1, max_depth=-1,\n",
1203 |        "               min_child_samples=20, min_child_weight=0.001, min_split_gain=0.0,\n",
1204 |        "               n_estimaotrs=400, n_estimators=100, n_jobs=-1, num_leaves=31,\n",
1205 |        "               objective=None, random_state=None, reg_alpha=0.0, reg_lambda=0.0,\n",
1206 |        "               silent=True, subsample=1.0, subsample_for_bin=200000,\n",
1207 |        "               subsample_freq=0)"
1208 |       ]
1209 |      },
1210 |      "execution_count": 11,
1211 |      "metadata": {},
1212 |      "output_type": "execute_result"
1213 |     }
1214 |    ],
1215 |    "source": [
1216 |     "lgb = LGBMClassifier(n_estimaotrs = 400)\n",
1217 |     "lgb.fit(X_train, y_train)"
1218 |    ]
1219 |   },
1220 |   {
1221 |    "cell_type": "code",
1222 |    "execution_count": 12,
1223 |    "metadata": {},
1224 |    "outputs": [
1225 |     {
1226 |      "name": "stdout",
1227 |      "output_type": "stream",
1228 |      "text": [
1229 |       "정확도 : 0.84, 정밀도 : 0.79, 재현율 : 0.73\n",
1230 |       "f1-score : 0.76, auc : 0.81\n"
1231 |      ]
1232 |     }
1233 |    ],
1234 |    "source": [
1235 |     "lgb_pred = lgb.predict(X_test)\n",
1236 |     "metrics(y_test, lgb_pred)"
1237 |    ]
1238 |   },
1239 |   {
1240 |    "cell_type": "code",
1241 |    "execution_count": 13,
1242 |    "metadata": {},
1243 |    "outputs": [
1244 |     {
1245 |      "name": "stdout",
1246 |      "output_type": "stream",
1247 |      "text": [
1248 |       "[1]\tvalid_0's binary_logloss: 0.605701\n",
1249 |       "Training until validation scores don't improve for 100 rounds\n",
1250 |       "[2]\tvalid_0's binary_logloss: 0.569461\n",
1251 |       "[3]\tvalid_0's binary_logloss: 0.540251\n",
1252 |       "[4]\tvalid_0's binary_logloss: 0.5147\n",
1253 |       "[5]\tvalid_0's binary_logloss: 0.493662\n",
1254 |       "[6]\tvalid_0's binary_logloss: 0.47569\n",
1255 |       "[7]\tvalid_0's binary_logloss: 0.45573\n",
1256 |       "[8]\tvalid_0's binary_logloss: 0.442288\n",
1257 |       "[9]\tvalid_0's binary_logloss: 0.427343\n",
1258 |       "[10]\tvalid_0's binary_logloss: 0.41478\n",
1259 |       "[11]\tvalid_0's binary_logloss: 0.404568\n",
1260 |       "[12]\tvalid_0's binary_logloss: 0.394087\n",
1261 |       "[13]\tvalid_0's binary_logloss: 0.384579\n",
1262 |       "[14]\tvalid_0's binary_logloss: 0.377022\n",
1263 |       "[15]\tvalid_0's binary_logloss: 0.372698\n",
1264 |       "[16]\tvalid_0's binary_logloss: 0.367266\n",
1265 |       "[17]\tvalid_0's binary_logloss: 0.364566\n",
1266 |       "[18]\tvalid_0's binary_logloss: 0.362322\n",
1267 |       "[19]\tvalid_0's binary_logloss: 0.35638\n",
1268 |       "[20]\tvalid_0's binary_logloss: 0.352956\n",
1269 |       "[21]\tvalid_0's binary_logloss: 0.351149\n",
1270 |       "[22]\tvalid_0's binary_logloss: 0.350341\n",
1271 |       "[23]\tvalid_0's binary_logloss: 0.348923\n",
1272 |       "[24]\tvalid_0's binary_logloss: 0.348176\n",
1273 |       "[25]\tvalid_0's binary_logloss: 0.34714\n",
1274 |       "[26]\tvalid_0's binary_logloss: 0.346754\n",
1275 |       "[27]\tvalid_0's binary_logloss: 0.347015\n",
1276 |       "[28]\tvalid_0's binary_logloss: 0.347799\n",
1277 |       "[29]\tvalid_0's binary_logloss: 0.348623\n",
1278 |       "[30]\tvalid_0's binary_logloss: 0.349346\n",
1279 |       "[31]\tvalid_0's binary_logloss: 0.350961\n",
1280 |       "[32]\tvalid_0's binary_logloss: 0.352158\n",
1281 |       "[33]\tvalid_0's binary_logloss: 0.352746\n",
1282 |       "[34]\tvalid_0's binary_logloss: 0.353988\n",
1283 |       "[35]\tvalid_0's binary_logloss: 0.35563\n",
1284 |       "[36]\tvalid_0's binary_logloss: 0.357587\n",
1285 |       "[37]\tvalid_0's binary_logloss: 0.357775\n",
1286 |       "[38]\tvalid_0's binary_logloss: 0.359317\n",
1287 |       "[39]\tvalid_0's binary_logloss: 0.360177\n",
1288 |       "[40]\tvalid_0's binary_logloss: 0.359158\n",
1289 |       "[41]\tvalid_0's binary_logloss: 0.360159\n",
1290 |       "[42]\tvalid_0's binary_logloss: 0.359884\n",
1291 |       "[43]\tvalid_0's binary_logloss: 0.360693\n",
1292 |       "[44]\tvalid_0's binary_logloss: 0.361518\n",
1293 |       "[45]\tvalid_0's binary_logloss: 0.361417\n",
1294 |       "[46]\tvalid_0's binary_logloss: 0.36477\n",
1295 |       "[47]\tvalid_0's binary_logloss: 0.366563\n",
1296 |       "[48]\tvalid_0's binary_logloss: 0.367413\n",
1297 |       "[49]\tvalid_0's binary_logloss: 0.370403\n",
1298 |       "[50]\tvalid_0's binary_logloss: 0.370454\n",
1299 |       "[51]\tvalid_0's binary_logloss: 0.3713\n",
1300 |       "[52]\tvalid_0's binary_logloss: 0.373395\n",
1301 |       "[53]\tvalid_0's binary_logloss: 0.371452\n",
1302 |       "[54]\tvalid_0's binary_logloss: 0.370792\n",
1303 |       "[55]\tvalid_0's binary_logloss: 0.369311\n",
1304 |       "[56]\tvalid_0's binary_logloss: 0.368987\n",
1305 |       "[57]\tvalid_0's binary_logloss: 0.372909\n",
1306 |       "[58]\tvalid_0's binary_logloss: 0.371587\n",
1307 |       "[59]\tvalid_0's binary_logloss: 0.371235\n",
1308 |       "[60]\tvalid_0's binary_logloss: 0.371714\n",
1309 |       "[61]\tvalid_0's binary_logloss: 0.372394\n",
1310 |       "[62]\tvalid_0's binary_logloss: 0.371164\n",
1311 |       "[63]\tvalid_0's binary_logloss: 0.371928\n",
1312 |       "[64]\tvalid_0's binary_logloss: 0.372314\n",
1313 |       "[65]\tvalid_0's binary_logloss: 0.372502\n",
1314 |       "[66]\tvalid_0's binary_logloss: 0.376302\n",
1315 |       "[67]\tvalid_0's binary_logloss: 0.378364\n",
1316 |       "[68]\tvalid_0's binary_logloss: 0.378404\n",
1317 |       "[69]\tvalid_0's binary_logloss: 0.381327\n",
1318 |       "[70]\tvalid_0's binary_logloss: 0.380973\n",
1319 |       "[71]\tvalid_0's binary_logloss: 0.382481\n",
1320 |       "[72]\tvalid_0's binary_logloss: 0.38136\n",
1321 |       "[73]\tvalid_0's binary_logloss: 0.383008\n",
1322 |       "[74]\tvalid_0's binary_logloss: 0.381861\n",
1323 |       "[75]\tvalid_0's binary_logloss: 0.382796\n",
1324 |       "[76]\tvalid_0's binary_logloss: 0.38258\n",
1325 |       "[77]\tvalid_0's binary_logloss: 0.384473\n",
1326 |       "[78]\tvalid_0's binary_logloss: 0.383581\n",
1327 |       "[79]\tvalid_0's binary_logloss: 0.385198\n",
1328 |       "[80]\tvalid_0's binary_logloss: 0.383797\n",
1329 |       "[81]\tvalid_0's binary_logloss: 0.383937\n",
1330 |       "[82]\tvalid_0's binary_logloss: 0.383372\n",
1331 |       "[83]\tvalid_0's binary_logloss: 0.384661\n",
1332 |       "[84]\tvalid_0's binary_logloss: 0.383799\n",
1333 |       "[85]\tvalid_0's binary_logloss: 0.384108\n",
1334 |       "[86]\tvalid_0's binary_logloss: 0.383364\n",
1335 |       "[87]\tvalid_0's binary_logloss: 0.384795\n",
1336 |       "[88]\tvalid_0's binary_logloss: 0.384702\n",
1337 |       "[89]\tvalid_0's binary_logloss: 0.386003\n",
1338 |       "[90]\tvalid_0's binary_logloss: 0.386621\n",
1339 |       "[91]\tvalid_0's binary_logloss: 0.387986\n",
1340 |       "[92]\tvalid_0's binary_logloss: 0.390496\n",
1341 |       "[93]\tvalid_0's binary_logloss: 0.389984\n",
1342 |       "[94]\tvalid_0's binary_logloss: 0.391477\n",
1343 |       "[95]\tvalid_0's binary_logloss: 0.391917\n",
1344 |       "[96]\tvalid_0's binary_logloss: 0.392326\n",
1345 |       "[97]\tvalid_0's binary_logloss: 0.392731\n",
1346 |       "[98]\tvalid_0's binary_logloss: 0.392586\n",
1347 |       "[99]\tvalid_0's binary_logloss: 0.394479\n",
1348 |       "[100]\tvalid_0's binary_logloss: 0.397251\n",
1349 |       "Did not meet early stopping. Best iteration is:\n",
1350 |       "[26]\tvalid_0's binary_logloss: 0.346754\n"
1351 |      ]
1352 |     },
1353 |     {
1354 |      "data": {
1355 |       "text/plain": [
1356 |        "LGBMClassifier(boosting_type='gbdt', class_weight=None, colsample_bytree=1.0,\n",
1357 |        "               importance_type='split', learning_rate=0.1, max_depth=-1,\n",
1358 |        "               min_child_samples=20, min_child_weight=0.001, min_split_gain=0.0,\n",
1359 |        "               n_estimaotrs=400, n_estimators=100, n_jobs=-1, num_leaves=31,\n",
1360 |        "               objective=None, random_state=None, reg_alpha=0.0, reg_lambda=0.0,\n",
1361 |        "               silent=True, subsample=1.0, subsample_for_bin=200000,\n",
1362 |        "               subsample_freq=0)"
1363 |       ]
1364 |      },
1365 |      "execution_count": 13,
1366 |      "metadata": {},
1367 |      "output_type": "execute_result"
1368 |     }
1369 |    ],
1370 |    "source": [
1371 |     "lgb = LGBMClassifier(n_estimaotrs = 400)\n",
1372 |     "evals = [(X_test, y_test)]\n",
1373 |     "lgb.fit(X_train, y_train, early_stopping_rounds = 100, eval_metric = \"logloss\", eval_set = evals, verbose = True)"
1374 |    ]
1375 |   },
1376 |   {
1377 |    "cell_type": "code",
1378 |    "execution_count": 15,
1379 |    "metadata": {},
1380 |    "outputs": [
1381 |     {
1382 |      "data": {
1383 |       "text/plain": [
1384 |        "<matplotlib.axes._subplots.AxesSubplot at 0x1c37516ff28>"
1385 |       ]
1386 |      },
1387 |      "execution_count": 15,
1388 |      "metadata": {},
1389 |      "output_type": "execute_result"
1390 |     },
1391 |     {
1392 |      "data": {
1393 |       "image/png": "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\n",
1394 |       "text/plain": [
1395 |        "<Figure size 720x432 with 1 Axes>"
1396 |       ]
1397 |      },
1398 |      "metadata": {
1399 |       "needs_background": "light"
1400 |      },
1401 |      "output_type": "display_data"
1402 |     }
1403 |    ],
1404 |    "source": [
1405 |     "fig, ax = plt.subplots(figsize=(10, 6))\n",
1406 |     "plot_importance(lgb, ax = ax)"
1407 |    ]
1408 |   },
1409 |   {
1410 |    "cell_type": "code",
1411 |    "execution_count": null,
1412 |    "metadata": {},
1413 |    "outputs": [],
1414 |    "source": []
1415 |   },
1416 |   {
1417 |    "cell_type": "code",
1418 |    "execution_count": null,
1419 |    "metadata": {},
1420 |    "outputs": [],
1421 |    "source": []
1422 |   }
1423 |  ],
1424 |  "metadata": {
1425 |   "kernelspec": {
1426 |    "display_name": "Python 3",
1427 |    "language": "python",
1428 |    "name": "python3"
1429 |   },
1430 |   "language_info": {
1431 |    "codemirror_mode": {
1432 |     "name": "ipython",
1433 |     "version": 3
1434 |    },
1435 |    "file_extension": ".py",
1436 |    "mimetype": "text/x-python",
1437 |    "name": "python",
1438 |    "nbconvert_exporter": "python",
1439 |    "pygments_lexer": "ipython3",
1440 |    "version": "3.6.9"
1441 |   }
1442 |  },
1443 |  "nbformat": 4,
1444 |  "nbformat_minor": 2
1445 | }
1446 | 


--------------------------------------------------------------------------------
/010. credit_card_fraud_basic.ipynb:
--------------------------------------------------------------------------------
  1 | {
  2 |  "cells": [
  3 |   {
  4 |    "cell_type": "code",
  5 |    "execution_count": 1,
  6 |    "metadata": {},
  7 |    "outputs": [],
  8 |    "source": [
  9 |     "import pandas as pd\n",
 10 |     "import numpy as np\n",
 11 |     "import matplotlib.pyplot as plt\n",
 12 |     "import seaborn as sns\n",
 13 |     "from xgboost import XGBClassifier, plot_importance as xg_importance\n",
 14 |     "from lightgbm import LGBMClassifier, plot_importance as lgb_importance\n",
 15 |     "from sklearn.datasets import load_breast_cancer\n",
 16 |     "from sklearn.model_selection import train_test_split, GridSearchCV\n",
 17 |     "from sklearn.metrics import accuracy_score, f1_score, precision_score, recall_score, roc_auc_score, confusion_matrix\n",
 18 |     "\n",
 19 |     "import warnings\n",
 20 |     "warnings.filterwarnings('ignore')\n"
 21 |    ]
 22 |   },
 23 |   {
 24 |    "cell_type": "code",
 25 |    "execution_count": 1,
 26 |    "metadata": {},
 27 |    "outputs": [],
 28 |    "source": [
 29 |     "def metrics(y_test, pred):\n",
 30 |     "    accuracy = accuracy_score(y_test, pred)\n",
 31 |     "    precision = precision_score(y_test, pred)\n",
 32 |     "    recall = recall_score(y_test, pred)\n",
 33 |     "    f1 = f1_score(y_test, pred)\n",
 34 |     "    roc_score = roc_auc_score(y_test, pred, average='macro')\n",
 35 |     "    print('정확도 : {0:.2f}, 정밀도 : {1:.2f}, 재현율 : {2:.2f}'.format(accuracy, precision, recall))\n",
 36 |     "    print('f1-score : {0:.2f}, auc : {1:.2f}'.format(f1, roc_score))"
 37 |    ]
 38 |   },
 39 |   {
 40 |    "cell_type": "code",
 41 |    "execution_count": 4,
 42 |    "metadata": {},
 43 |    "outputs": [],
 44 |    "source": [
 45 |     "data = pd.read_csv('../datas/credit card fraud/creditcard.csv')"
 46 |    ]
 47 |   },
 48 |   {
 49 |    "cell_type": "code",
 50 |    "execution_count": 5,
 51 |    "metadata": {},
 52 |    "outputs": [
 53 |     {
 54 |      "data": {
 55 |       "text/plain": [
 56 |        "(284807, 31)"
 57 |       ]
 58 |      },
 59 |      "execution_count": 5,
 60 |      "metadata": {},
 61 |      "output_type": "execute_result"
 62 |     }
 63 |    ],
 64 |    "source": [
 65 |     "data.shape"
 66 |    ]
 67 |   },
 68 |   {
 69 |    "cell_type": "code",
 70 |    "execution_count": 6,
 71 |    "metadata": {},
 72 |    "outputs": [
 73 |     {
 74 |      "name": "stdout",
 75 |      "output_type": "stream",
 76 |      "text": [
 77 |       "<class 'pandas.core.frame.DataFrame'>\n",
 78 |       "RangeIndex: 284807 entries, 0 to 284806\n",
 79 |       "Data columns (total 31 columns):\n",
 80 |       "Time      284807 non-null float64\n",
 81 |       "V1        284807 non-null float64\n",
 82 |       "V2        284807 non-null float64\n",
 83 |       "V3        284807 non-null float64\n",
 84 |       "V4        284807 non-null float64\n",
 85 |       "V5        284807 non-null float64\n",
 86 |       "V6        284807 non-null float64\n",
 87 |       "V7        284807 non-null float64\n",
 88 |       "V8        284807 non-null float64\n",
 89 |       "V9        284807 non-null float64\n",
 90 |       "V10       284807 non-null float64\n",
 91 |       "V11       284807 non-null float64\n",
 92 |       "V12       284807 non-null float64\n",
 93 |       "V13       284807 non-null float64\n",
 94 |       "V14       284807 non-null float64\n",
 95 |       "V15       284807 non-null float64\n",
 96 |       "V16       284807 non-null float64\n",
 97 |       "V17       284807 non-null float64\n",
 98 |       "V18       284807 non-null float64\n",
 99 |       "V19       284807 non-null float64\n",
100 |       "V20       284807 non-null float64\n",
101 |       "V21       284807 non-null float64\n",
102 |       "V22       284807 non-null float64\n",
103 |       "V23       284807 non-null float64\n",
104 |       "V24       284807 non-null float64\n",
105 |       "V25       284807 non-null float64\n",
106 |       "V26       284807 non-null float64\n",
107 |       "V27       284807 non-null float64\n",
108 |       "V28       284807 non-null float64\n",
109 |       "Amount    284807 non-null float64\n",
110 |       "Class     284807 non-null int64\n",
111 |       "dtypes: float64(30), int64(1)\n",
112 |       "memory usage: 67.4 MB\n"
113 |      ]
114 |     }
115 |    ],
116 |    "source": [
117 |     "data.info()"
118 |    ]
119 |   },
120 |   {
121 |    "cell_type": "code",
122 |    "execution_count": 7,
123 |    "metadata": {},
124 |    "outputs": [
125 |     {
126 |      "data": {
127 |       "text/plain": [
128 |        "Time      0\n",
129 |        "V1        0\n",
130 |        "V2        0\n",
131 |        "V3        0\n",
132 |        "V4        0\n",
133 |        "V5        0\n",
134 |        "V6        0\n",
135 |        "V7        0\n",
136 |        "V8        0\n",
137 |        "V9        0\n",
138 |        "V10       0\n",
139 |        "V11       0\n",
140 |        "V12       0\n",
141 |        "V13       0\n",
142 |        "V14       0\n",
143 |        "V15       0\n",
144 |        "V16       0\n",
145 |        "V17       0\n",
146 |        "V18       0\n",
147 |        "V19       0\n",
148 |        "V20       0\n",
149 |        "V21       0\n",
150 |        "V22       0\n",
151 |        "V23       0\n",
152 |        "V24       0\n",
153 |        "V25       0\n",
154 |        "V26       0\n",
155 |        "V27       0\n",
156 |        "V28       0\n",
157 |        "Amount    0\n",
158 |        "Class     0\n",
159 |        "dtype: int64"
160 |       ]
161 |      },
162 |      "execution_count": 7,
163 |      "metadata": {},
164 |      "output_type": "execute_result"
165 |     }
166 |    ],
167 |    "source": [
168 |     "data.isna().sum()"
169 |    ]
170 |   },
171 |   {
172 |    "cell_type": "code",
173 |    "execution_count": 8,
174 |    "metadata": {},
175 |    "outputs": [
176 |     {
177 |      "data": {
178 |       "text/plain": [
179 |        "0    284315\n",
180 |        "1       492\n",
181 |        "Name: Class, dtype: int64"
182 |       ]
183 |      },
184 |      "execution_count": 8,
185 |      "metadata": {},
186 |      "output_type": "execute_result"
187 |     }
188 |    ],
189 |    "source": [
190 |     "data.iloc[:, -1].value_counts()"
191 |    ]
192 |   },
193 |   {
194 |    "cell_type": "code",
195 |    "execution_count": 9,
196 |    "metadata": {},
197 |    "outputs": [
198 |     {
199 |      "data": {
200 |       "text/plain": [
201 |        "<matplotlib.axes._subplots.AxesSubplot at 0x17db00cd668>"
202 |       ]
203 |      },
204 |      "execution_count": 9,
205 |      "metadata": {},
206 |      "output_type": "execute_result"
207 |     },
208 |     {
209 |      "data": {
210 |       "image/png": "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\n",
211 |       "text/plain": [
212 |        "<Figure size 432x288 with 1 Axes>"
213 |       ]
214 |      },
215 |      "metadata": {
216 |       "needs_background": "light"
217 |      },
218 |      "output_type": "display_data"
219 |     }
220 |    ],
221 |    "source": [
222 |     "data.iloc[:, -1].value_counts().plot(kind='bar')"
223 |    ]
224 |   },
225 |   {
226 |    "cell_type": "code",
227 |    "execution_count": 10,
228 |    "metadata": {},
229 |    "outputs": [
230 |     {
231 |      "data": {
232 |       "text/plain": [
233 |        "0    99.827251\n",
234 |        "1     0.172749\n",
235 |        "Name: Class, dtype: float64"
236 |       ]
237 |      },
238 |      "execution_count": 10,
239 |      "metadata": {},
240 |      "output_type": "execute_result"
241 |     }
242 |    ],
243 |    "source": [
244 |     "data.iloc[:, -1].value_counts() / data.iloc[:, -1].count() * 100"
245 |    ]
246 |   },
247 |   {
248 |    "cell_type": "code",
249 |    "execution_count": 11,
250 |    "metadata": {},
251 |    "outputs": [],
252 |    "source": [
253 |     "X = data.iloc[:, :-1]\n",
254 |     "y = data.iloc[:, -1]"
255 |    ]
256 |   },
257 |   {
258 |    "cell_type": "code",
259 |    "execution_count": 13,
260 |    "metadata": {},
261 |    "outputs": [],
262 |    "source": [
263 |     "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.25, random_state = 10)"
264 |    ]
265 |   },
266 |   {
267 |    "cell_type": "code",
268 |    "execution_count": 14,
269 |    "metadata": {},
270 |    "outputs": [
271 |     {
272 |      "name": "stdout",
273 |      "output_type": "stream",
274 |      "text": [
275 |       "0    99.826315\n",
276 |       "1     0.173685\n",
277 |       "Name: Class, dtype: float64\n",
278 |       "0    99.830061\n",
279 |       "1     0.169939\n",
280 |       "Name: Class, dtype: float64\n"
281 |      ]
282 |     }
283 |    ],
284 |    "source": [
285 |     "print(y_train.value_counts() / y_train.count() * 100)\n",
286 |     "print(y_test.value_counts() / y_test.count() * 100)"
287 |    ]
288 |   },
289 |   {
290 |    "cell_type": "code",
291 |    "execution_count": 16,
292 |    "metadata": {},
293 |    "outputs": [],
294 |    "source": [
295 |     "def modeling(model, x_train, x_test, y_train, y_test):\n",
296 |     "    model.fit(x_train, y_train)\n",
297 |     "    pred = model.predict(x_test)\n",
298 |     "    metrics(y_test, pred)"
299 |    ]
300 |   },
301 |   {
302 |    "cell_type": "code",
303 |    "execution_count": 20,
304 |    "metadata": {},
305 |    "outputs": [
306 |     {
307 |      "name": "stdout",
308 |      "output_type": "stream",
309 |      "text": [
310 |       "정확도 : 1.00, 정밀도 : 0.77, 재현율 : 0.53\n",
311 |       "f1-score : 0.63, auc : 0.76\n"
312 |      ]
313 |     }
314 |    ],
315 |    "source": [
316 |     "from sklearn.linear_model import LogisticRegression\n",
317 |     "lr = LogisticRegression()\n",
318 |     "modeling(lr, X_train, X_test, y_train, y_test)"
319 |    ]
320 |   },
321 |   {
322 |    "cell_type": "code",
323 |    "execution_count": 21,
324 |    "metadata": {},
325 |    "outputs": [
326 |     {
327 |      "name": "stdout",
328 |      "output_type": "stream",
329 |      "text": [
330 |       "정확도 : 1.00, 정밀도 : 0.95, 재현율 : 0.83\n",
331 |       "f1-score : 0.88, auc : 0.91\n"
332 |      ]
333 |     }
334 |    ],
335 |    "source": [
336 |     "lgb = LGBMClassifier(n_estimators = 1000, num_leaves = 64, n_jobs = -1, boost_from_average = False)\n",
337 |     "modeling(lgb, x_train = X_train, x_test = X_test, y_train = y_train, y_test = y_test)"
338 |    ]
339 |   },
340 |   {
341 |    "cell_type": "code",
342 |    "execution_count": null,
343 |    "metadata": {},
344 |    "outputs": [],
345 |    "source": []
346 |   },
347 |   {
348 |    "cell_type": "code",
349 |    "execution_count": null,
350 |    "metadata": {},
351 |    "outputs": [],
352 |    "source": []
353 |   }
354 |  ],
355 |  "metadata": {
356 |   "kernelspec": {
357 |    "display_name": "Python 3",
358 |    "language": "python",
359 |    "name": "python3"
360 |   },
361 |   "language_info": {
362 |    "codemirror_mode": {
363 |     "name": "ipython",
364 |     "version": 3
365 |    },
366 |    "file_extension": ".py",
367 |    "mimetype": "text/x-python",
368 |    "name": "python",
369 |    "nbconvert_exporter": "python",
370 |    "pygments_lexer": "ipython3",
371 |    "version": "3.6.9"
372 |   }
373 |  },
374 |  "nbformat": 4,
375 |  "nbformat_minor": 2
376 | }
377 | 


--------------------------------------------------------------------------------
/README.md:
--------------------------------------------------------------------------------
1 | # machine_learning_basic
2 | Repo for everyone who wants a machine learning basic
3 | 


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