baseline_02.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "827557d3",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "import seaborn as sns \n",
    "import matplotlib.pyplot as plt\n",
    "import tensorflow as tf\n",
    "\n",
    "plt.rcParams['figure.figsize']=(10,10)\n",
    "plt.rcParams['font.family']='AppleGothic'\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings(action='ignore')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "d97ce81d",
   "metadata": {},
   "outputs": [],
   "source": [
    "def read_csv_by_dir(path, index_col=None):\n",
    "    df_raw = pd.DataFrame()\n",
    "    for files in os.listdir(path):\n",
    "        if files.endswith('.csv'):\n",
    "            df = pd.read_csv('/'.join([path,files]),\n",
    "                            index_col=index_col)\n",
    "        df_raw = pd.concat((df_raw,df),axis=0)\n",
    "    return df_raw"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "b5354701",
   "metadata": {},
   "outputs": [],
   "source": [
    "path = 'Dataset'\n",
    "_df_rf_raw = read_csv_by_dir('/'.join([path,'rf_data']),\n",
    "                            index_col=0)\n",
    "\n",
    "_df_water_raw = read_csv_by_dir('/'.join([path,'water_data']),\n",
    "                               index_col=0)\n",
    "\n",
    "_submission = pd.read_csv('/'.join([path,'sample_submission.csv']),\n",
    "                             index_col=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "584f790f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# raw_data 보존하기\n",
    "df_rf=_df_rf_raw.copy()\n",
    "df_rf.name = \"rain_data\"\n",
    "\n",
    "df_water=_df_water_raw.copy()\n",
    "df_water.name = \"water_data\"\n",
    "\n",
    "submission=_submission.copy()\n",
    "submission.name = \"submission\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "27854e65",
   "metadata": {},
   "outputs": [],
   "source": [
    "def index_to_datetime(df,format):\n",
    "    df.index = pd.to_datetime(df.index,\n",
    "                              format=format)\n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "6ac6f689",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_rf=index_to_datetime(df=df_rf,format='%Y-%m-%d %H:%M')\n",
    "df_water=index_to_datetime(df=df_water,format='%Y-%m-%d %H:%M')\n",
    "submission=index_to_datetime(df=submission,format='%Y-%m-%d %H:%M')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "6ea111e5",
   "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>swl</th>\n",
       "      <th>inf</th>\n",
       "      <th>sfw</th>\n",
       "      <th>ecpc</th>\n",
       "      <th>tototf</th>\n",
       "      <th>tide_level</th>\n",
       "      <th>wl_1018662</th>\n",
       "      <th>fw_1018662</th>\n",
       "      <th>wl_1018680</th>\n",
       "      <th>fw_1018680</th>\n",
       "      <th>wl_1018683</th>\n",
       "      <th>fw_1018683</th>\n",
       "      <th>wl_1019630</th>\n",
       "      <th>fw_1019630</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ymdhm</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2012-01-01 00:00:00</th>\n",
       "      <td>24.95</td>\n",
       "      <td>232</td>\n",
       "      <td>224.23</td>\n",
       "      <td>19.77</td>\n",
       "      <td>134</td>\n",
       "      <td>NaN</td>\n",
       "      <td>278.7</td>\n",
       "      <td>269.4</td>\n",
       "      <td>273.2</td>\n",
       "      <td>0</td>\n",
       "      <td>261.0</td>\n",
       "      <td>175.16</td>\n",
       "      <td>253.3</td>\n",
       "      <td>310.31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2012-01-01 00:10:00</th>\n",
       "      <td>24.929</td>\n",
       "      <td>134.4</td>\n",
       "      <td>223.49</td>\n",
       "      <td>20.51</td>\n",
       "      <td>134.4</td>\n",
       "      <td>NaN</td>\n",
       "      <td>279.7</td>\n",
       "      <td>274.78</td>\n",
       "      <td>273.2</td>\n",
       "      <td>0</td>\n",
       "      <td>261.0</td>\n",
       "      <td>176.6</td>\n",
       "      <td>258.3</td>\n",
       "      <td>356.89</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2012-01-01 00:20:00</th>\n",
       "      <td>24.929</td>\n",
       "      <td>128.3</td>\n",
       "      <td>223.49</td>\n",
       "      <td>20.51</td>\n",
       "      <td>128.3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>279.7</td>\n",
       "      <td>274.78</td>\n",
       "      <td>273.2</td>\n",
       "      <td>0</td>\n",
       "      <td>261.0</td>\n",
       "      <td>179.49</td>\n",
       "      <td>261.3</td>\n",
       "      <td>386.43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2012-01-01 00:30:00</th>\n",
       "      <td>24.929</td>\n",
       "      <td>134.4</td>\n",
       "      <td>223.49</td>\n",
       "      <td>20.51</td>\n",
       "      <td>134.4</td>\n",
       "      <td>NaN</td>\n",
       "      <td>278.7</td>\n",
       "      <td>269.4</td>\n",
       "      <td>273.2</td>\n",
       "      <td>0</td>\n",
       "      <td>261.0</td>\n",
       "      <td>173.72</td>\n",
       "      <td>262.3</td>\n",
       "      <td>396.55</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2012-01-01 00:40:00</th>\n",
       "      <td>24.937</td>\n",
       "      <td>232.7</td>\n",
       "      <td>223.74</td>\n",
       "      <td>20.26</td>\n",
       "      <td>134.4</td>\n",
       "      <td>NaN</td>\n",
       "      <td>279.7</td>\n",
       "      <td>274.78</td>\n",
       "      <td>273.2</td>\n",
       "      <td>0</td>\n",
       "      <td>261.0</td>\n",
       "      <td>76.99</td>\n",
       "      <td>262.3</td>\n",
       "      <td>396.55</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-07-18 23:10:00</th>\n",
       "      <td>25.04</td>\n",
       "      <td>259.23</td>\n",
       "      <td>212.86</td>\n",
       "      <td>31.14</td>\n",
       "      <td>259.23</td>\n",
       "      <td>510.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>319.84</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-456.41</td>\n",
       "      <td>0.0</td>\n",
       "      <td>974.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-07-18 23:20:00</th>\n",
       "      <td>25.04</td>\n",
       "      <td>260.46</td>\n",
       "      <td>212.86</td>\n",
       "      <td>31.14</td>\n",
       "      <td>260.46</td>\n",
       "      <td>492.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>314.01</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-717.3</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1006.88</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-07-18 23:30:00</th>\n",
       "      <td>25.04</td>\n",
       "      <td>259.37</td>\n",
       "      <td>212.86</td>\n",
       "      <td>31.14</td>\n",
       "      <td>259.37</td>\n",
       "      <td>475.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>387.55</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-843.37</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1039.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-07-18 23:40:00</th>\n",
       "      <td>25.04</td>\n",
       "      <td>259.13</td>\n",
       "      <td>212.86</td>\n",
       "      <td>31.14</td>\n",
       "      <td>259.13</td>\n",
       "      <td>458.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>454.91</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-1023.37</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1073.46</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-07-18 23:50:00</th>\n",
       "      <td>25.04</td>\n",
       "      <td>258.16</td>\n",
       "      <td>212.86</td>\n",
       "      <td>31.14</td>\n",
       "      <td>258.16</td>\n",
       "      <td>442.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>582.21</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-1049.44</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1090.45</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>537408 rows × 14 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                        swl     inf     sfw   ecpc  tototf  tide_level  \\\n",
       "ymdhm                                                                    \n",
       "2012-01-01 00:00:00   24.95     232  224.23  19.77     134         NaN   \n",
       "2012-01-01 00:10:00  24.929   134.4  223.49  20.51   134.4         NaN   \n",
       "2012-01-01 00:20:00  24.929   128.3  223.49  20.51   128.3         NaN   \n",
       "2012-01-01 00:30:00  24.929   134.4  223.49  20.51   134.4         NaN   \n",
       "2012-01-01 00:40:00  24.937   232.7  223.74  20.26   134.4         NaN   \n",
       "...                     ...     ...     ...    ...     ...         ...   \n",
       "2022-07-18 23:10:00   25.04  259.23  212.86  31.14  259.23       510.0   \n",
       "2022-07-18 23:20:00   25.04  260.46  212.86  31.14  260.46       492.0   \n",
       "2022-07-18 23:30:00   25.04  259.37  212.86  31.14  259.37       475.0   \n",
       "2022-07-18 23:40:00   25.04  259.13  212.86  31.14  259.13       458.0   \n",
       "2022-07-18 23:50:00   25.04  258.16  212.86  31.14  258.16       442.0   \n",
       "\n",
       "                     wl_1018662 fw_1018662  wl_1018680 fw_1018680  wl_1018683  \\\n",
       "ymdhm                                                                           \n",
       "2012-01-01 00:00:00       278.7      269.4       273.2          0       261.0   \n",
       "2012-01-01 00:10:00       279.7     274.78       273.2          0       261.0   \n",
       "2012-01-01 00:20:00       279.7     274.78       273.2          0       261.0   \n",
       "2012-01-01 00:30:00       278.7      269.4       273.2          0       261.0   \n",
       "2012-01-01 00:40:00       279.7     274.78       273.2          0       261.0   \n",
       "...                         ...        ...         ...        ...         ...   \n",
       "2022-07-18 23:10:00         0.0     319.84         0.0        NaN         0.0   \n",
       "2022-07-18 23:20:00         0.0     314.01         0.0        NaN         0.0   \n",
       "2022-07-18 23:30:00         0.0     387.55         0.0        NaN         0.0   \n",
       "2022-07-18 23:40:00         0.0     454.91         0.0        NaN         0.0   \n",
       "2022-07-18 23:50:00         0.0     582.21         0.0        NaN         0.0   \n",
       "\n",
       "                    fw_1018683  wl_1019630 fw_1019630  \n",
       "ymdhm                                                  \n",
       "2012-01-01 00:00:00     175.16       253.3     310.31  \n",
       "2012-01-01 00:10:00      176.6       258.3     356.89  \n",
       "2012-01-01 00:20:00     179.49       261.3     386.43  \n",
       "2012-01-01 00:30:00     173.72       262.3     396.55  \n",
       "2012-01-01 00:40:00      76.99       262.3     396.55  \n",
       "...                        ...         ...        ...  \n",
       "2022-07-18 23:10:00    -456.41         0.0      974.4  \n",
       "2022-07-18 23:20:00     -717.3         0.0    1006.88  \n",
       "2022-07-18 23:30:00    -843.37         0.0     1039.9  \n",
       "2022-07-18 23:40:00   -1023.37         0.0    1073.46  \n",
       "2022-07-18 23:50:00   -1049.44         0.0    1090.45  \n",
       "\n",
       "[537408 rows x 14 columns]"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_water"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "681afe1e",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_rf.sort_index(inplace=True)\n",
    "df_water.sort_index(inplace=True)\n",
    "submission.sort_index(inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "008b3b68",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# data target 분리하기\n",
    "target = df_water.loc[:,submission.columns]\n",
    "data = pd.concat((df_rf,df_water.drop(submission.columns,axis=1)),axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "4958b1ef",
   "metadata": {},
   "outputs": [],
   "source": [
    "# data와 target 하나 밀어주기 (과거데이터를 사용해야 함으로)\n",
    "_target = target.reset_index(drop=True)\n",
    "_data = data.reset_index(drop=True)\n",
    "\n",
    "_data.index += 1\n",
    "\n",
    "tot=pd.concat((_data,_target),axis=1)\n",
    "tot=tot.sort_index()\n",
    "\n",
    "tot=tot.iloc[1:-1]\n",
    "\n",
    "target = tot.loc[:,submission.columns]\n",
    "data = tot.drop(submission.columns,axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "54a38ccf",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_target=target.iloc[:-len(submission),:]\n",
    "test_target=target.iloc[-len(submission):,:]\n",
    "\n",
    "train_data=data.iloc[:-len(submission),:]\n",
    "test_data=data.iloc[-len(submission):,:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "471f2572",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "rf_10184100         0\n",
       "rf_10184110         0\n",
       "rf_10184140         0\n",
       "swl               707\n",
       "inf               707\n",
       "sfw               707\n",
       "ecpc              707\n",
       "tototf            707\n",
       "tide_level      13384\n",
       "fw_1018662      15164\n",
       "fw_1018680     189936\n",
       "fw_1018683       1279\n",
       "fw_1019630         59\n",
       "dtype: int64"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_data.isna().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "a999d2b8",
   "metadata": {},
   "outputs": [],
   "source": [
    "#train_target.fillna(train_target.mean(),inplace=True)\n",
    "#test_target.fillna(train_target.mean(),inplace=True)\n",
    "train_data.fillna(train_data.mean(),inplace=True)\n",
    "#test_data.fillna(train_data.mean(),inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "f968f1dd",
   "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>rf_10184100</th>\n",
       "      <th>rf_10184110</th>\n",
       "      <th>rf_10184140</th>\n",
       "      <th>swl</th>\n",
       "      <th>inf</th>\n",
       "      <th>sfw</th>\n",
       "      <th>ecpc</th>\n",
       "      <th>tototf</th>\n",
       "      <th>tide_level</th>\n",
       "      <th>fw_1018662</th>\n",
       "      <th>fw_1018680</th>\n",
       "      <th>fw_1018683</th>\n",
       "      <th>fw_1019630</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>24.95</td>\n",
       "      <td>232</td>\n",
       "      <td>224.23</td>\n",
       "      <td>19.77</td>\n",
       "      <td>134</td>\n",
       "      <td>334.386259</td>\n",
       "      <td>269.4</td>\n",
       "      <td>0</td>\n",
       "      <td>175.16</td>\n",
       "      <td>310.31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>24.929</td>\n",
       "      <td>134.4</td>\n",
       "      <td>223.49</td>\n",
       "      <td>20.51</td>\n",
       "      <td>134.4</td>\n",
       "      <td>334.386259</td>\n",
       "      <td>274.78</td>\n",
       "      <td>0</td>\n",
       "      <td>176.6</td>\n",
       "      <td>356.89</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>24.929</td>\n",
       "      <td>128.3</td>\n",
       "      <td>223.49</td>\n",
       "      <td>20.51</td>\n",
       "      <td>128.3</td>\n",
       "      <td>334.386259</td>\n",
       "      <td>274.78</td>\n",
       "      <td>0</td>\n",
       "      <td>179.49</td>\n",
       "      <td>386.43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>24.929</td>\n",
       "      <td>134.4</td>\n",
       "      <td>223.49</td>\n",
       "      <td>20.51</td>\n",
       "      <td>134.4</td>\n",
       "      <td>334.386259</td>\n",
       "      <td>269.4</td>\n",
       "      <td>0</td>\n",
       "      <td>173.72</td>\n",
       "      <td>396.55</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>24.937</td>\n",
       "      <td>232.7</td>\n",
       "      <td>223.74</td>\n",
       "      <td>20.26</td>\n",
       "      <td>134.4</td>\n",
       "      <td>334.386259</td>\n",
       "      <td>274.78</td>\n",
       "      <td>0</td>\n",
       "      <td>76.99</td>\n",
       "      <td>396.55</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>530491</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>25.18</td>\n",
       "      <td>140.89</td>\n",
       "      <td>217.99</td>\n",
       "      <td>26.01</td>\n",
       "      <td>140.89</td>\n",
       "      <td>232.000000</td>\n",
       "      <td>314.01</td>\n",
       "      <td>NaN</td>\n",
       "      <td>597.98</td>\n",
       "      <td>471.08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>530492</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>25.18</td>\n",
       "      <td>140.94</td>\n",
       "      <td>217.99</td>\n",
       "      <td>26.01</td>\n",
       "      <td>140.94</td>\n",
       "      <td>220.000000</td>\n",
       "      <td>285.72</td>\n",
       "      <td>NaN</td>\n",
       "      <td>575.57</td>\n",
       "      <td>493.58</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>530493</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>25.18</td>\n",
       "      <td>141.07</td>\n",
       "      <td>217.99</td>\n",
       "      <td>26.01</td>\n",
       "      <td>141.07</td>\n",
       "      <td>208.000000</td>\n",
       "      <td>274.78</td>\n",
       "      <td>NaN</td>\n",
       "      <td>501.04</td>\n",
       "      <td>505.03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>530494</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>25.18</td>\n",
       "      <td>141.01</td>\n",
       "      <td>217.99</td>\n",
       "      <td>26.01</td>\n",
       "      <td>141.01</td>\n",
       "      <td>196.000000</td>\n",
       "      <td>269.4</td>\n",
       "      <td>NaN</td>\n",
       "      <td>425.89</td>\n",
       "      <td>505.03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>530495</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>25.19</td>\n",
       "      <td>755.75</td>\n",
       "      <td>218.36</td>\n",
       "      <td>25.64</td>\n",
       "      <td>140.75</td>\n",
       "      <td>184.000000</td>\n",
       "      <td>264.07</td>\n",
       "      <td>NaN</td>\n",
       "      <td>198.19</td>\n",
       "      <td>493.58</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>530495 rows × 13 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        rf_10184100  rf_10184110  rf_10184140     swl     inf     sfw   ecpc  \\\n",
       "1               0.0          0.0          0.0   24.95     232  224.23  19.77   \n",
       "2               0.0          0.0          0.0  24.929   134.4  223.49  20.51   \n",
       "3               0.0          0.0          0.0  24.929   128.3  223.49  20.51   \n",
       "4               0.0          0.0          0.0  24.929   134.4  223.49  20.51   \n",
       "5               0.0          0.0          0.0  24.937   232.7  223.74  20.26   \n",
       "...             ...          ...          ...     ...     ...     ...    ...   \n",
       "530491          0.0          0.0          0.0   25.18  140.89  217.99  26.01   \n",
       "530492          0.0          0.0          0.0   25.18  140.94  217.99  26.01   \n",
       "530493          0.0          0.0          0.0   25.18  141.07  217.99  26.01   \n",
       "530494          0.0          0.0          0.0   25.18  141.01  217.99  26.01   \n",
       "530495          0.0          0.0          0.0   25.19  755.75  218.36  25.64   \n",
       "\n",
       "        tototf  tide_level fw_1018662 fw_1018680 fw_1018683 fw_1019630  \n",
       "1          134  334.386259      269.4          0     175.16     310.31  \n",
       "2        134.4  334.386259     274.78          0      176.6     356.89  \n",
       "3        128.3  334.386259     274.78          0     179.49     386.43  \n",
       "4        134.4  334.386259      269.4          0     173.72     396.55  \n",
       "5        134.4  334.386259     274.78          0      76.99     396.55  \n",
       "...        ...         ...        ...        ...        ...        ...  \n",
       "530491  140.89  232.000000     314.01        NaN     597.98     471.08  \n",
       "530492  140.94  220.000000     285.72        NaN     575.57     493.58  \n",
       "530493  141.07  208.000000     274.78        NaN     501.04     505.03  \n",
       "530494  141.01  196.000000      269.4        NaN     425.89     505.03  \n",
       "530495  140.75  184.000000     264.07        NaN     198.19     493.58  \n",
       "\n",
       "[530495 rows x 13 columns]"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "1bec03ee",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "rf_10184100         0\n",
       "rf_10184110         0\n",
       "rf_10184140         0\n",
       "swl               707\n",
       "inf               707\n",
       "sfw               707\n",
       "ecpc              707\n",
       "tototf            707\n",
       "tide_level          0\n",
       "fw_1018662      15164\n",
       "fw_1018680     189936\n",
       "fw_1018683       1279\n",
       "fw_1019630         59\n",
       "dtype: int64"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_data.isna().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "d19801af",
   "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>wl_1018662</th>\n",
       "      <th>wl_1018680</th>\n",
       "      <th>wl_1018683</th>\n",
       "      <th>wl_1019630</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>314.7</td>\n",
       "      <td>300.2</td>\n",
       "      <td>290.0</td>\n",
       "      <td>275.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>313.7</td>\n",
       "      <td>301.2</td>\n",
       "      <td>290.0</td>\n",
       "      <td>275.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>311.7</td>\n",
       "      <td>301.2</td>\n",
       "      <td>290.0</td>\n",
       "      <td>276.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>311.7</td>\n",
       "      <td>301.2</td>\n",
       "      <td>291.0</td>\n",
       "      <td>277.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>311.7</td>\n",
       "      <td>301.2</td>\n",
       "      <td>291.0</td>\n",
       "      <td>277.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>269419</th>\n",
       "      <td>281.7</td>\n",
       "      <td>281.2</td>\n",
       "      <td>278.0</td>\n",
       "      <td>271.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>269420</th>\n",
       "      <td>279.7</td>\n",
       "      <td>279.2</td>\n",
       "      <td>278.0</td>\n",
       "      <td>272.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>269421</th>\n",
       "      <td>278.7</td>\n",
       "      <td>277.2</td>\n",
       "      <td>277.0</td>\n",
       "      <td>272.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>269422</th>\n",
       "      <td>277.7</td>\n",
       "      <td>276.2</td>\n",
       "      <td>276.0</td>\n",
       "      <td>271.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>269423</th>\n",
       "      <td>278.7</td>\n",
       "      <td>276.2</td>\n",
       "      <td>275.0</td>\n",
       "      <td>270.3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>269423 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        wl_1018662  wl_1018680  wl_1018683  wl_1019630\n",
       "1            314.7       300.2       290.0       275.3\n",
       "2            313.7       301.2       290.0       275.3\n",
       "3            311.7       301.2       290.0       276.3\n",
       "4            311.7       301.2       291.0       277.3\n",
       "5            311.7       301.2       291.0       277.3\n",
       "...            ...         ...         ...         ...\n",
       "269419       281.7       281.2       278.0       271.3\n",
       "269420       279.7       279.2       278.0       272.3\n",
       "269421       278.7       277.2       277.0       272.3\n",
       "269422       277.7       276.2       276.0       271.3\n",
       "269423       278.7       276.2       275.0       270.3\n",
       "\n",
       "[269423 rows x 4 columns]"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_target"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "f7d6c9a1",
   "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>rf_10184100</th>\n",
       "      <th>rf_10184110</th>\n",
       "      <th>rf_10184140</th>\n",
       "      <th>swl</th>\n",
       "      <th>inf</th>\n",
       "      <th>sfw</th>\n",
       "      <th>ecpc</th>\n",
       "      <th>tototf</th>\n",
       "      <th>tide_level</th>\n",
       "      <th>fw_1018662</th>\n",
       "      <th>fw_1018680</th>\n",
       "      <th>fw_1018683</th>\n",
       "      <th>fw_1019630</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>530496</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>25.19</td>\n",
       "      <td>140.56</td>\n",
       "      <td>218.36</td>\n",
       "      <td>25.64</td>\n",
       "      <td>140.56</td>\n",
       "      <td>173.0</td>\n",
       "      <td>269.4</td>\n",
       "      <td>NaN</td>\n",
       "      <td>70.09</td>\n",
       "      <td>482.26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>530497</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>25.19</td>\n",
       "      <td>140.6</td>\n",
       "      <td>218.36</td>\n",
       "      <td>25.64</td>\n",
       "      <td>140.6</td>\n",
       "      <td>162.0</td>\n",
       "      <td>269.4</td>\n",
       "      <td>NaN</td>\n",
       "      <td>8.23</td>\n",
       "      <td>471.08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>530498</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>25.19</td>\n",
       "      <td>140.78</td>\n",
       "      <td>218.36</td>\n",
       "      <td>25.64</td>\n",
       "      <td>140.78</td>\n",
       "      <td>151.0</td>\n",
       "      <td>280.22</td>\n",
       "      <td>NaN</td>\n",
       "      <td>28.82</td>\n",
       "      <td>449.12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>530499</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>25.2</td>\n",
       "      <td>755.9</td>\n",
       "      <td>218.73</td>\n",
       "      <td>25.27</td>\n",
       "      <td>140.9</td>\n",
       "      <td>141.0</td>\n",
       "      <td>296.87</td>\n",
       "      <td>NaN</td>\n",
       "      <td>12.35</td>\n",
       "      <td>417.17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>530500</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>25.19</td>\n",
       "      <td>0.0</td>\n",
       "      <td>218.36</td>\n",
       "      <td>25.64</td>\n",
       "      <td>140.94</td>\n",
       "      <td>130.0</td>\n",
       "      <td>302.53</td>\n",
       "      <td>NaN</td>\n",
       "      <td>53.52</td>\n",
       "      <td>386.43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>537403</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>25.04</td>\n",
       "      <td>865.96</td>\n",
       "      <td>212.86</td>\n",
       "      <td>31.14</td>\n",
       "      <td>259.3</td>\n",
       "      <td>528.0</td>\n",
       "      <td>319.84</td>\n",
       "      <td>NaN</td>\n",
       "      <td>25.27</td>\n",
       "      <td>926.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>537404</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>25.04</td>\n",
       "      <td>259.23</td>\n",
       "      <td>212.86</td>\n",
       "      <td>31.14</td>\n",
       "      <td>259.23</td>\n",
       "      <td>510.0</td>\n",
       "      <td>319.84</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-456.41</td>\n",
       "      <td>974.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>537405</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>25.04</td>\n",
       "      <td>260.46</td>\n",
       "      <td>212.86</td>\n",
       "      <td>31.14</td>\n",
       "      <td>260.46</td>\n",
       "      <td>492.0</td>\n",
       "      <td>314.01</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-717.3</td>\n",
       "      <td>1006.88</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>537406</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>25.04</td>\n",
       "      <td>259.37</td>\n",
       "      <td>212.86</td>\n",
       "      <td>31.14</td>\n",
       "      <td>259.37</td>\n",
       "      <td>475.0</td>\n",
       "      <td>387.55</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-843.37</td>\n",
       "      <td>1039.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>537407</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>25.04</td>\n",
       "      <td>259.13</td>\n",
       "      <td>212.86</td>\n",
       "      <td>31.14</td>\n",
       "      <td>259.13</td>\n",
       "      <td>458.0</td>\n",
       "      <td>454.91</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1023.37</td>\n",
       "      <td>1073.46</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>6912 rows × 13 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        rf_10184100  rf_10184110  rf_10184140    swl     inf     sfw   ecpc  \\\n",
       "530496          0.0          0.0          0.0  25.19  140.56  218.36  25.64   \n",
       "530497          0.0          0.0          0.0  25.19   140.6  218.36  25.64   \n",
       "530498          0.0          0.0          0.0  25.19  140.78  218.36  25.64   \n",
       "530499          0.0          0.0          0.0   25.2   755.9  218.73  25.27   \n",
       "530500          0.0          0.0          0.0  25.19     0.0  218.36  25.64   \n",
       "...             ...          ...          ...    ...     ...     ...    ...   \n",
       "537403          0.0          0.0          0.0  25.04  865.96  212.86  31.14   \n",
       "537404          0.0          0.0          0.0  25.04  259.23  212.86  31.14   \n",
       "537405          0.0          0.0          0.0  25.04  260.46  212.86  31.14   \n",
       "537406          0.0          0.0          0.0  25.04  259.37  212.86  31.14   \n",
       "537407          0.0          0.0          0.0  25.04  259.13  212.86  31.14   \n",
       "\n",
       "        tototf  tide_level fw_1018662 fw_1018680 fw_1018683 fw_1019630  \n",
       "530496  140.56       173.0      269.4        NaN      70.09     482.26  \n",
       "530497   140.6       162.0      269.4        NaN       8.23     471.08  \n",
       "530498  140.78       151.0     280.22        NaN      28.82     449.12  \n",
       "530499   140.9       141.0     296.87        NaN      12.35     417.17  \n",
       "530500  140.94       130.0     302.53        NaN      53.52     386.43  \n",
       "...        ...         ...        ...        ...        ...        ...  \n",
       "537403   259.3       528.0     319.84        NaN      25.27      926.7  \n",
       "537404  259.23       510.0     319.84        NaN    -456.41      974.4  \n",
       "537405  260.46       492.0     314.01        NaN     -717.3    1006.88  \n",
       "537406  259.37       475.0     387.55        NaN    -843.37     1039.9  \n",
       "537407  259.13       458.0     454.91        NaN   -1023.37    1073.46  \n",
       "\n",
       "[6912 rows x 13 columns]"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "145274cc",
   "metadata": {},
   "source": [
    "### LightGBM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "cf94fde5",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_target_1018662 = train_target['wl_1018662']\n",
    "train_target_1018680 = train_target['wl_1018680']\n",
    "train_target_1018683 = train_target['wl_1018683']\n",
    "train_target_1019630 = train_target['wl_1019630']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d003d106",
   "metadata": {},
   "outputs": [],
   "source": [
    "from lightgbm import LGBMRegressor\n",
    "\n",
    "lgbm =  LGBMRegressor(\n",
    "            n_estimators=1000,\n",
    "            n_jobs=-1,\n",
    "            learning_rate=0.001,\n",
    "            num_leaves = 32,\n",
    "            max_depth = 128,\n",
    "            min_child_samples = 100,\n",
    "            subsample = 0.8,\n",
    "            silent=-1,\n",
    "            verbose=-1)\n",
    "with tf.device(\"/device:GPU:0\"):\n",
    "    print('=====lgbm_1018662=====')\n",
    "    lgbm.fit(train_data, train_target_1018662,verbose= 100)\n",
    "    pred_1018662 = lgbm.predict(test_data)\n",
    "    print('=====lgbm_1018680=====')\n",
    "    lgbm.fit(train_data, train_target_1018680,verbose= 100)\n",
    "    pred_1018680 = lgbm.predict(test_data)\n",
    "    print('=====lgbm_1018683=====')\n",
    "    lgbm.fit(train_data, train_target_1018683,verbose= 100)\n",
    "    pred_1018683 = lgbm.predict(test_data)\n",
    "    print('=====lgbm_1019630=====')\n",
    "    lgbm.fit(train_data, train_target_1019630,verbose= 100)\n",
    "    pred_1019630 = lgbm.predict(test_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a23f7759",
   "metadata": {},
   "outputs": [],
   "source": [
    "_submission[\"wl_1018662\"] = pred_1018662\n",
    "_submission[\"wl_1018680\"] = pred_1018680\n",
    "_submission[\"wl_1018683\"] = pred_1018683\n",
    "_submission[\"wl_1019630\"] = pred_1019630\n",
    "\n",
    "sample_submission.to_csv('lgbm_test2.csv',index=False)"
   ]
  }
 ],
 "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": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}