SVM.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import time\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from sklearn.pipeline import Pipeline, FeatureUnion\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.pipeline import make_pipeline\n",
    "from sklearn.model_selection import train_test_split, cross_val_score, KFold, LeaveOneOut\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.ensemble import RandomForestClassifier, BaggingClassifier\n",
    "from sklearn import preprocessing\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "from keras.utils.np_utils import to_categorical\n",
    "from sklearn.tree import DecisionTreeClassifier, ExtraTreeClassifier, DecisionTreeRegressor\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "path =os.getcwd()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "df =pd.read_csv(path +'/Kickstarter/cleaned_kick_data01.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(85233, 17)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>backers_count</th>\n",
       "      <th>category</th>\n",
       "      <th>country</th>\n",
       "      <th>created_at</th>\n",
       "      <th>currency</th>\n",
       "      <th>deadline</th>\n",
       "      <th>fx_rate</th>\n",
       "      <th>goal</th>\n",
       "      <th>location</th>\n",
       "      <th>pledged</th>\n",
       "      <th>staff_pick</th>\n",
       "      <th>status</th>\n",
       "      <th>duration</th>\n",
       "      <th>created_day</th>\n",
       "      <th>continent</th>\n",
       "      <th>city</th>\n",
       "      <th>state</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>film</td>\n",
       "      <td>GB</td>\n",
       "      <td>2017-01-11</td>\n",
       "      <td>GBP</td>\n",
       "      <td>2017-02-15</td>\n",
       "      <td>1.261110</td>\n",
       "      <td>40000.0</td>\n",
       "      <td>London, UK</td>\n",
       "      <td>0.0</td>\n",
       "      <td>False</td>\n",
       "      <td>failed</td>\n",
       "      <td>35</td>\n",
       "      <td>11</td>\n",
       "      <td>Europe</td>\n",
       "      <td>London</td>\n",
       "      <td>UK</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>12</td>\n",
       "      <td>crafts</td>\n",
       "      <td>CA</td>\n",
       "      <td>2016-02-28</td>\n",
       "      <td>CAD</td>\n",
       "      <td>2016-03-29</td>\n",
       "      <td>0.748587</td>\n",
       "      <td>5000.0</td>\n",
       "      <td>London, Canada</td>\n",
       "      <td>357.0</td>\n",
       "      <td>False</td>\n",
       "      <td>failed</td>\n",
       "      <td>30</td>\n",
       "      <td>28</td>\n",
       "      <td>America</td>\n",
       "      <td>London</td>\n",
       "      <td>Canada</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>games</td>\n",
       "      <td>AU</td>\n",
       "      <td>2016-02-13</td>\n",
       "      <td>AUD</td>\n",
       "      <td>2016-03-19</td>\n",
       "      <td>0.722499</td>\n",
       "      <td>5000.0</td>\n",
       "      <td>Perth, AU</td>\n",
       "      <td>1.0</td>\n",
       "      <td>False</td>\n",
       "      <td>failed</td>\n",
       "      <td>35</td>\n",
       "      <td>13</td>\n",
       "      <td>Oceania</td>\n",
       "      <td>Perth</td>\n",
       "      <td>AU</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>fashion</td>\n",
       "      <td>US</td>\n",
       "      <td>2015-05-25</td>\n",
       "      <td>USD</td>\n",
       "      <td>2015-07-01</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1900.0</td>\n",
       "      <td>Austin, TX</td>\n",
       "      <td>49.0</td>\n",
       "      <td>False</td>\n",
       "      <td>failed</td>\n",
       "      <td>37</td>\n",
       "      <td>25</td>\n",
       "      <td>America</td>\n",
       "      <td>Austin</td>\n",
       "      <td>TX</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2</td>\n",
       "      <td>film</td>\n",
       "      <td>US</td>\n",
       "      <td>2018-05-18</td>\n",
       "      <td>USD</td>\n",
       "      <td>2018-06-17</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>30000.0</td>\n",
       "      <td>Los Angeles, CA</td>\n",
       "      <td>2.0</td>\n",
       "      <td>False</td>\n",
       "      <td>failed</td>\n",
       "      <td>30</td>\n",
       "      <td>18</td>\n",
       "      <td>America</td>\n",
       "      <td>Los Angeles</td>\n",
       "      <td>CA</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   backers_count category country  created_at currency    deadline   fx_rate  \\\n",
       "0              0     film      GB  2017-01-11      GBP  2017-02-15  1.261110   \n",
       "1             12   crafts      CA  2016-02-28      CAD  2016-03-29  0.748587   \n",
       "2              1    games      AU  2016-02-13      AUD  2016-03-19  0.722499   \n",
       "3              3  fashion      US  2015-05-25      USD  2015-07-01  1.000000   \n",
       "4              2     film      US  2018-05-18      USD  2018-06-17  1.000000   \n",
       "\n",
       "      goal         location  pledged  staff_pick  status  duration  \\\n",
       "0  40000.0       London, UK      0.0       False  failed        35   \n",
       "1   5000.0   London, Canada    357.0       False  failed        30   \n",
       "2   5000.0        Perth, AU      1.0       False  failed        35   \n",
       "3   1900.0       Austin, TX     49.0       False  failed        37   \n",
       "4  30000.0  Los Angeles, CA      2.0       False  failed        30   \n",
       "\n",
       "   created_day continent         city   state  \n",
       "0           11    Europe       London      UK  \n",
       "1           28   America       London  Canada  \n",
       "2           13   Oceania        Perth      AU  \n",
       "3           25   America       Austin      TX  \n",
       "4           18   America  Los Angeles      CA  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "#对标称属性进行独热编码，转变为数值属性\n",
    "le =LabelEncoder()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['backers_count'] =le.fit_transform(df['backers_count'].values)\n",
    "df['category'] =le.fit_transform(df['category'].values)\n",
    "df['country'] =le.fit_transform(df['country'].values)\n",
    "df['currency'] =le.fit_transform(df['currency'].values)\n",
    "df['staff_pick'] =le.fit_transform(df['staff_pick'].values)\n",
    "df['status'] =le.fit_transform(df['status'].values)\n",
    "df['duration'] =le.fit_transform(df['duration'].values)\n",
    "df['created_day'] =le.fit_transform(df['created_day'].values)\n",
    "df['continent'] =le.fit_transform(df['continent'].values)\n",
    "df['city'] =le.fit_transform(df['city'].values)\n",
    "df['state'] =le.fit_transform(df['state'].values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "#对类别信息进行处理\n",
    "df['continent'] = to_categorical(df['continent']) \n",
    "df['category'] =to_categorical(df['category'])\n",
    "df['country'] =to_categorical(df['country'])\n",
    "df['city'] =to_categorical(df['city'])\n",
    "df['state'] =to_categorical(df['state'])\n",
    "df['currency'] =to_categorical(df['currency'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>backers_count</th>\n",
       "      <th>category</th>\n",
       "      <th>country</th>\n",
       "      <th>created_at</th>\n",
       "      <th>currency</th>\n",
       "      <th>deadline</th>\n",
       "      <th>fx_rate</th>\n",
       "      <th>goal</th>\n",
       "      <th>location</th>\n",
       "      <th>pledged</th>\n",
       "      <th>staff_pick</th>\n",
       "      <th>status</th>\n",
       "      <th>duration</th>\n",
       "      <th>created_day</th>\n",
       "      <th>continent</th>\n",
       "      <th>city</th>\n",
       "      <th>state</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2017-01-11</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2017-02-15</td>\n",
       "      <td>1.261110</td>\n",
       "      <td>40000.0</td>\n",
       "      <td>London, UK</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>34</td>\n",
       "      <td>10</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>12</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2016-02-28</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2016-03-29</td>\n",
       "      <td>0.748587</td>\n",
       "      <td>5000.0</td>\n",
       "      <td>London, Canada</td>\n",
       "      <td>357.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>29</td>\n",
       "      <td>27</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2016-02-13</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2016-03-19</td>\n",
       "      <td>0.722499</td>\n",
       "      <td>5000.0</td>\n",
       "      <td>Perth, AU</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>34</td>\n",
       "      <td>12</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2015-05-25</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2015-07-01</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1900.0</td>\n",
       "      <td>Austin, TX</td>\n",
       "      <td>49.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>36</td>\n",
       "      <td>24</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2018-05-18</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2018-06-17</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>30000.0</td>\n",
       "      <td>Los Angeles, CA</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>29</td>\n",
       "      <td>17</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   backers_count  category  country  created_at  currency    deadline  \\\n",
       "0              0       0.0      0.0  2017-01-11       0.0  2017-02-15   \n",
       "1             12       0.0      0.0  2016-02-28       0.0  2016-03-29   \n",
       "2              1       0.0      0.0  2016-02-13       1.0  2016-03-19   \n",
       "3              3       0.0      0.0  2015-05-25       0.0  2015-07-01   \n",
       "4              2       0.0      0.0  2018-05-18       0.0  2018-06-17   \n",
       "\n",
       "    fx_rate     goal         location  pledged  staff_pick  status  duration  \\\n",
       "0  1.261110  40000.0       London, UK      0.0           0       0        34   \n",
       "1  0.748587   5000.0   London, Canada    357.0           0       0        29   \n",
       "2  0.722499   5000.0        Perth, AU      1.0           0       0        34   \n",
       "3  1.000000   1900.0       Austin, TX     49.0           0       0        36   \n",
       "4  1.000000  30000.0  Los Angeles, CA      2.0           0       0        29   \n",
       "\n",
       "   created_day  continent  city  state  \n",
       "0           10        0.0   0.0    0.0  \n",
       "1           27        1.0   0.0    0.0  \n",
       "2           12        0.0   0.0    0.0  \n",
       "3           24        1.0   0.0    0.0  \n",
       "4           17        1.0   0.0    0.0  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "X =df.drop(columns=[ 'created_at', 'deadline', 'pledged',  'location',  'status'], axis=1)\n",
    "y =df['status']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train,y_test =train_test_split(X, y, test_size =0.2, random_state =1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "使用默认的 gamma 和 C来训练svm模型"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "ss = StandardScaler()\n",
    "clf =SVC(kernel= 'rbf', gamma=0.1, C=1)\n",
    "clf_pipe =Pipeline([('sscale', ss), ('svm', clf)])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/zhenghongsheng/anaconda3/lib/python3.6/site-packages/sklearn/preprocessing/data.py:645: DataConversionWarning: Data with input dtype int64, float32, float64 were all converted to float64 by StandardScaler.\n",
      "  return self.partial_fit(X, y)\n",
      "/Users/zhenghongsheng/anaconda3/lib/python3.6/site-packages/sklearn/base.py:467: DataConversionWarning: Data with input dtype int64, float32, float64 were all converted to float64 by StandardScaler.\n",
      "  return self.fit(X, y, **fit_params).transform(X)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Pipeline(memory=None,\n",
       "     steps=[('sscale', StandardScaler(copy=True, with_mean=True, with_std=True)), ('svm', SVC(C=1, cache_size=200, class_weight=None, coef0=0.0,\n",
       "  decision_function_shape='ovr', degree=3, gamma=0.1, kernel='rbf',\n",
       "  max_iter=-1, probability=False, random_state=None, shrinking=True,\n",
       "  tol=0.001, verbose=False))])"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "start_time =time.time()\n",
    "clf_pipe.fit(X_train, y_train)\n",
    "end_time =time.time()\n",
    "dtime =end_time- start_time\n",
    "print('运行时间: %.8f s' %dtime)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/zhenghongsheng/anaconda3/lib/python3.6/site-packages/sklearn/pipeline.py:511: DataConversionWarning: Data with input dtype int64, float32, float64 were all converted to float64 by StandardScaler.\n",
      "  Xt = transform.transform(Xt)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0.8516410993459068"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf_pipe.score(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/zhenghongsheng/anaconda3/lib/python3.6/site-packages/sklearn/pipeline.py:511: DataConversionWarning: Data with input dtype int64, float32, float64 were all converted to float64 by StandardScaler.\n",
      "  Xt = transform.transform(Xt)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0.8533466299055552"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf_pipe.score(X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import roc_curve,f1_score, classification_report, precision_score, recall_score,auc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/zhenghongsheng/anaconda3/lib/python3.6/site-packages/sklearn/pipeline.py:401: DataConversionWarning: Data with input dtype int64, float32, float64 were all converted to float64 by StandardScaler.\n",
      "  Xt = transform.transform(Xt)\n"
     ]
    }
   ],
   "source": [
    "fpr_svm, tpr_svm, thresholds_svm =roc_curve(y_test, clf_pipe.decision_function(X_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x1a351cb400>"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(fpr_svm, tpr_svm, label ='ROC Curve')\n",
    "plt.xlabel('FPR')\n",
    "plt.ylabel('TPR')\n",
    "# cloze_zero =np.argmin(np.abs(thresholds))\n",
    "# plt.plot(fpr[cloze_zero], tpr[cloze_zero], 'o', markersize =10,\n",
    "#          label = 'threshold zero', fillstyle ='none', c= 'k', mew = 2)\n",
    "plt.legend(loc =4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "auc of SVM 0.938900123905148\n"
     ]
    }
   ],
   "source": [
    "print('auc of SVM', auc(fpr_svm, tpr_svm))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/zhenghongsheng/anaconda3/lib/python3.6/site-packages/sklearn/pipeline.py:331: DataConversionWarning: Data with input dtype int64, float32, float64 were all converted to float64 by StandardScaler.\n",
      "  Xt = transform.transform(Xt)\n"
     ]
    }
   ],
   "source": [
    "clfpre =clf_pipe.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "f1 socre of SVM: 0.8455264458724666\n",
      "precison of SVM: 0.8856957928802589\n",
      "recall of SVM: 0.8088426527958388\n"
     ]
    }
   ],
   "source": [
    "#计算SVM的f1scores\n",
    "print('f1 socre of SVM:', f1_score(y_test, clfpre))\n",
    "print('precison of SVM:', precision_score(y_test, clfpre))\n",
    "print('recall of SVM:', recall_score(y_test, clfpre))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "      falied       0.72      0.93      0.81      8588\n",
      "  successful       0.90      0.63      0.74      8459\n",
      "\n",
      "   micro avg       0.78      0.78      0.78     17047\n",
      "   macro avg       0.81      0.78      0.78     17047\n",
      "weighted avg       0.81      0.78      0.78     17047\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test, clfpre, target_names=['falied', 'successful']))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "通过网格搜索方式进行调参"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "#调用GridSearchCV方法实现来寻找SVM最好的参数 gamma和 C\n",
    "# #\n",
    "\n",
    "# param_grid =[\n",
    "#     {'kernel' :['rbf'],\n",
    "#      'C' : [0.001, 0.01, 0.1, 1, 10, 100],\n",
    "#      'gamma'  : [0.001, 0.01, 0.1, 1, 10, 100]},\n",
    "#     {\n",
    "#         'kernel' : ['linear'],\n",
    "#         'C' : [0.001, 0.01, 0.1, 1, 10, 100]}]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "# grid_search =GridSearchCV(SVC(), param_grid, cv=3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "# grid_pipe =Pipeline([('sscale', ss), ('grid_search', grid_search)])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# grid_search.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#返回网格搜索后的最佳参数\n",
    "# print('Best parameters:{}'.format(grid_pipe.best_params_))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#返回交叉验证后的最佳精度\n",
    "# print('Best cross_validation:{:.2f}'.format(grid_pipe.best_score_))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#返回网格搜索后的最佳模型\n",
    "# print('Best estimator:\\n{}'.format(grid_pipe.best_estimator_))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#返回网格搜索的results\n",
    "# results =pd.DataFrame(grid_pipe.cv_results_)\n",
    "# display(results.T)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#对交叉验证的结果进行热力图可视化\n",
    "# scores =np.array(results.mean_test_score).reshape(6 , 6)\n",
    "# sns.heatmap(scores, xlabel ='gamma', xticklabels =param_grid['gamma'], ylabel ='C', yticklabels =param_grid['C'], cmap=\"YlGnBu\")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}