Untitled.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 140,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 141,
   "metadata": {},
   "outputs": [],
   "source": [
    "df=pd.read_csv(\"train__updated.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 142,
   "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>index</th>\n",
       "      <th>year</th>\n",
       "      <th>month</th>\n",
       "      <th>day</th>\n",
       "      <th>hour</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>imma_version</th>\n",
       "      <th>attm_count</th>\n",
       "      <th>time_indicator</th>\n",
       "      <th>...</th>\n",
       "      <th>platform_type</th>\n",
       "      <th>dup_status</th>\n",
       "      <th>dup_check</th>\n",
       "      <th>nightday_flag</th>\n",
       "      <th>source_exclusion_flags</th>\n",
       "      <th>release_no_primary</th>\n",
       "      <th>release_no_secondary</th>\n",
       "      <th>release_no_tertiary</th>\n",
       "      <th>release_status_indicator</th>\n",
       "      <th>intermediate_reject_flag</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>train_0</td>\n",
       "      <td>1978</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>48.8</td>\n",
       "      <td>-126.0</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>6.0</td>\n",
       "      <td>1</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>train_1</td>\n",
       "      <td>1978</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>49.7</td>\n",
       "      <td>-127.9</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>6.0</td>\n",
       "      <td>1</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>train_2</td>\n",
       "      <td>1978</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>50.9</td>\n",
       "      <td>-129.9</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>6.0</td>\n",
       "      <td>1</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>train_3</td>\n",
       "      <td>1978</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>51.4</td>\n",
       "      <td>-128.8</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>6.0</td>\n",
       "      <td>1</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>train_4</td>\n",
       "      <td>1978</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>49.3</td>\n",
       "      <td>-123.7</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>6.0</td>\n",
       "      <td>1</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 54 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     index  year  month  day  hour  latitude  longitude  imma_version  \\\n",
       "0  train_0  1978      1    1   0.0      48.8     -126.0             1   \n",
       "1  train_1  1978      1    1   0.0      49.7     -127.9             1   \n",
       "2  train_2  1978      1    1   0.0      50.9     -129.9             1   \n",
       "3  train_3  1978      1    1   0.0      51.4     -128.8             1   \n",
       "4  train_4  1978      1    1   0.0      49.3     -123.7             1   \n",
       "\n",
       "   attm_count  time_indicator            ...             platform_type  \\\n",
       "0           4             NaN            ...                       6.0   \n",
       "1           4             NaN            ...                       6.0   \n",
       "2           4             NaN            ...                       6.0   \n",
       "3           4             NaN            ...                       6.0   \n",
       "4           4             NaN            ...                       6.0   \n",
       "\n",
       "   dup_status  dup_check  nightday_flag  source_exclusion_flags  \\\n",
       "0           1        2.0              2                     NaN   \n",
       "1           1        2.0              2                     NaN   \n",
       "2           1        2.0              2                     NaN   \n",
       "3           1        2.0              2                     NaN   \n",
       "4           1        2.0              2                     NaN   \n",
       "\n",
       "   release_no_primary  release_no_secondary  release_no_tertiary  \\\n",
       "0                   3                     0                    1   \n",
       "1                   3                     0                    1   \n",
       "2                   3                     0                    1   \n",
       "3                   3                     0                    1   \n",
       "4                   3                     0                    1   \n",
       "\n",
       "   release_status_indicator  intermediate_reject_flag  \n",
       "0                         0                         1  \n",
       "1                         0                         1  \n",
       "2                         0                         1  \n",
       "3                         0                         1  \n",
       "4                         0                         1  \n",
       "\n",
       "[5 rows x 54 columns]"
      ]
     },
     "execution_count": 142,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['index',\n",
       " 'year',\n",
       " 'month',\n",
       " 'day',\n",
       " 'hour',\n",
       " 'latitude',\n",
       " 'longitude',\n",
       " 'imma_version',\n",
       " 'attm_count',\n",
       " 'time_indicator',\n",
       " 'latlong_indicator',\n",
       " 'ship_course',\n",
       " 'ship_speed',\n",
       " 'national_source_indicator',\n",
       " 'id_indicator',\n",
       " 'wind_direction_indicator',\n",
       " 'wind_direction_true',\n",
       " 'wind_speed_indicator',\n",
       " 'wind_speed',\n",
       " 'visibility',\n",
       " 'present_weather',\n",
       " 'past_weather',\n",
       " 'sea_level_pressure',\n",
       " 'characteristic_of_ppp',\n",
       " 'amt_pressure_tend',\n",
       " 'indicator_for_temp',\n",
       " 'air_temperature',\n",
       " 'wbt_indicator',\n",
       " 'wetbulb_temperature',\n",
       " 'dpt_indicator',\n",
       " 'dewpoint_temperature',\n",
       " 'sst_measurement_method',\n",
       " 'sea_surface_temp',\n",
       " 'total_cloud_amount',\n",
       " 'lower_cloud_amount',\n",
       " 'wave_period',\n",
       " 'wave_height',\n",
       " 'swell_direction',\n",
       " 'swell_period',\n",
       " 'swell_height',\n",
       " 'ten_degree_box_number',\n",
       " 'one_degree_box_number',\n",
       " 'deck',\n",
       " 'source_id',\n",
       " 'platform_type',\n",
       " 'dup_status',\n",
       " 'dup_check',\n",
       " 'nightday_flag',\n",
       " 'source_exclusion_flags',\n",
       " 'release_no_primary',\n",
       " 'release_no_secondary',\n",
       " 'release_no_tertiary',\n",
       " 'release_status_indicator',\n",
       " 'intermediate_reject_flag']"
      ]
     },
     "execution_count": 143,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns.tolist()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Data Visualisation**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 144,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x1296 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "missing_df = df.isnull().sum(axis=0).reset_index()\n",
    "missing_df.columns = ['column_name', 'missing_count']\n",
    "missing_df = missing_df.loc[missing_df['missing_count']>0]\n",
    "missing_df = missing_df.sort_values(by='missing_count')\n",
    "\n",
    "ind = np.arange(missing_df.shape[0])\n",
    "width = 0.9\n",
    "fig, ax = plt.subplots(figsize=(12,18))\n",
    "rects = ax.barh(ind, missing_df.missing_count.values, color='#34495e')\n",
    "ax.set_yticks(ind)\n",
    "ax.set_yticklabels(missing_df.column_name.values, rotation='horizontal')\n",
    "ax.set_xlabel(\"Count of missing values\")\n",
    "ax.set_title(\"Number of missing values in each column\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 145,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "index                            0\n",
       "year                             0\n",
       "month                            0\n",
       "day                              0\n",
       "hour                             0\n",
       "latitude                         0\n",
       "longitude                        0\n",
       "imma_version                     0\n",
       "attm_count                       0\n",
       "time_indicator                   0\n",
       "latlong_indicator                0\n",
       "ship_course                      0\n",
       "ship_speed                       0\n",
       "national_source_indicator        0\n",
       "id_indicator                     0\n",
       "wind_direction_indicator         0\n",
       "wind_direction_true              0\n",
       "wind_speed_indicator             0\n",
       "wind_speed                       0\n",
       "visibility                       0\n",
       "present_weather                  0\n",
       "past_weather                     0\n",
       "sea_level_pressure               0\n",
       "characteristic_of_ppp            0\n",
       "amt_pressure_tend                0\n",
       "indicator_for_temp               0\n",
       "air_temperature                  0\n",
       "wbt_indicator                    0\n",
       "wetbulb_temperature              0\n",
       "dpt_indicator                50000\n",
       "dewpoint_temperature             0\n",
       "sst_measurement_method           0\n",
       "sea_surface_temp                 0\n",
       "total_cloud_amount               0\n",
       "lower_cloud_amount               0\n",
       "wave_period                      0\n",
       "wave_height                      0\n",
       "swell_direction                  0\n",
       "swell_period                     0\n",
       "swell_height                     0\n",
       "ten_degree_box_number            0\n",
       "one_degree_box_number            0\n",
       "deck                             0\n",
       "source_id                        0\n",
       "platform_type                    0\n",
       "dup_status                       0\n",
       "dup_check                        0\n",
       "nightday_flag                    0\n",
       "source_exclusion_flags       50000\n",
       "release_no_primary               0\n",
       "release_no_secondary             0\n",
       "release_no_tertiary              0\n",
       "release_status_indicator         0\n",
       "intermediate_reject_flag         0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 145,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.fillna(method='ffill', inplace=True)\n",
    "df.fillna(method='bfill', inplace=True)\n",
    "df.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 146,
   "metadata": {},
   "outputs": [],
   "source": [
    "s=df.std()==0\n",
    "m=[]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 147,
   "metadata": {},
   "outputs": [],
   "source": [
    "for each in range(len(s)):\n",
    "    if s[each]==True:\n",
    "        m.append(s.index[each])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 148,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['year',\n",
       " 'imma_version',\n",
       " 'attm_count',\n",
       " 'latlong_indicator',\n",
       " 'national_source_indicator',\n",
       " 'wind_direction_indicator',\n",
       " 'dup_check',\n",
       " 'release_no_primary',\n",
       " 'release_no_secondary',\n",
       " 'release_no_tertiary',\n",
       " 'release_status_indicator',\n",
       " 'intermediate_reject_flag']"
      ]
     },
     "execution_count": 148,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "m"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 180,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    50000.000000\n",
       "mean         0.017840\n",
       "std          0.183963\n",
       "min          0.000000\n",
       "25%          0.000000\n",
       "50%          0.000000\n",
       "75%          0.000000\n",
       "max          2.000000\n",
       "Name: wbt_indicator, dtype: float64"
      ]
     },
     "execution_count": 180,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['wbt_indicator'].describe() "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Normalisation**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 152,
   "metadata": {},
   "outputs": [],
   "source": [
    "normalized_df=df.drop(['index', 'wave_height', 'source_exclusion_flags', 'dpt_indicator'], axis=1)\n",
    "normalized_df=normalized_df.drop(m, axis=1)\n",
    "normalized_df=(normalized_df-normalized_df.mean())/normalized_df.std()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 154,
   "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>month</th>\n",
       "      <th>day</th>\n",
       "      <th>hour</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>time_indicator</th>\n",
       "      <th>ship_course</th>\n",
       "      <th>ship_speed</th>\n",
       "      <th>id_indicator</th>\n",
       "      <th>wind_direction_true</th>\n",
       "      <th>...</th>\n",
       "      <th>swell_direction</th>\n",
       "      <th>swell_period</th>\n",
       "      <th>swell_height</th>\n",
       "      <th>ten_degree_box_number</th>\n",
       "      <th>one_degree_box_number</th>\n",
       "      <th>deck</th>\n",
       "      <th>source_id</th>\n",
       "      <th>platform_type</th>\n",
       "      <th>dup_status</th>\n",
       "      <th>nightday_flag</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-1.401968</td>\n",
       "      <td>-1.687158</td>\n",
       "      <td>-1.699598</td>\n",
       "      <td>-0.320314</td>\n",
       "      <td>0.004165</td>\n",
       "      <td>-0.016126</td>\n",
       "      <td>1.066827</td>\n",
       "      <td>0.803293</td>\n",
       "      <td>0.160768</td>\n",
       "      <td>1.211566</td>\n",
       "      <td>...</td>\n",
       "      <td>0.884043</td>\n",
       "      <td>-2.842444</td>\n",
       "      <td>2.850838</td>\n",
       "      <td>0.617359</td>\n",
       "      <td>0.676268</td>\n",
       "      <td>-0.16058</td>\n",
       "      <td>-0.160586</td>\n",
       "      <td>0.161175</td>\n",
       "      <td>0.044315</td>\n",
       "      <td>1.013805</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-1.401968</td>\n",
       "      <td>-1.687158</td>\n",
       "      <td>-1.699598</td>\n",
       "      <td>0.141247</td>\n",
       "      <td>-0.817914</td>\n",
       "      <td>-0.016126</td>\n",
       "      <td>1.066827</td>\n",
       "      <td>0.803293</td>\n",
       "      <td>0.160768</td>\n",
       "      <td>1.315335</td>\n",
       "      <td>...</td>\n",
       "      <td>0.884043</td>\n",
       "      <td>-2.842444</td>\n",
       "      <td>2.850838</td>\n",
       "      <td>0.617359</td>\n",
       "      <td>1.061496</td>\n",
       "      <td>-0.16058</td>\n",
       "      <td>-0.160586</td>\n",
       "      <td>0.161175</td>\n",
       "      <td>0.044315</td>\n",
       "      <td>1.013805</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-1.401968</td>\n",
       "      <td>-1.687158</td>\n",
       "      <td>-1.699598</td>\n",
       "      <td>0.756662</td>\n",
       "      <td>-1.683260</td>\n",
       "      <td>-0.016126</td>\n",
       "      <td>1.066827</td>\n",
       "      <td>0.803293</td>\n",
       "      <td>0.160768</td>\n",
       "      <td>1.522872</td>\n",
       "      <td>...</td>\n",
       "      <td>0.884043</td>\n",
       "      <td>-2.842444</td>\n",
       "      <td>2.850838</td>\n",
       "      <td>-1.619772</td>\n",
       "      <td>-2.020331</td>\n",
       "      <td>-0.16058</td>\n",
       "      <td>-0.160586</td>\n",
       "      <td>0.161175</td>\n",
       "      <td>0.044315</td>\n",
       "      <td>1.013805</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-1.401968</td>\n",
       "      <td>-1.687158</td>\n",
       "      <td>-1.699598</td>\n",
       "      <td>1.013085</td>\n",
       "      <td>-1.207320</td>\n",
       "      <td>-0.016126</td>\n",
       "      <td>1.066827</td>\n",
       "      <td>0.803293</td>\n",
       "      <td>0.160768</td>\n",
       "      <td>1.626641</td>\n",
       "      <td>...</td>\n",
       "      <td>0.884043</td>\n",
       "      <td>-2.842444</td>\n",
       "      <td>2.850838</td>\n",
       "      <td>-1.619772</td>\n",
       "      <td>-1.705144</td>\n",
       "      <td>-0.16058</td>\n",
       "      <td>-0.160586</td>\n",
       "      <td>0.161175</td>\n",
       "      <td>0.044315</td>\n",
       "      <td>1.013805</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-1.401968</td>\n",
       "      <td>-1.687158</td>\n",
       "      <td>-1.699598</td>\n",
       "      <td>-0.063891</td>\n",
       "      <td>0.999313</td>\n",
       "      <td>-0.016126</td>\n",
       "      <td>1.066827</td>\n",
       "      <td>0.803293</td>\n",
       "      <td>0.160768</td>\n",
       "      <td>-1.278883</td>\n",
       "      <td>...</td>\n",
       "      <td>0.884043</td>\n",
       "      <td>-2.842444</td>\n",
       "      <td>2.850838</td>\n",
       "      <td>0.617359</td>\n",
       "      <td>0.921413</td>\n",
       "      <td>-0.16058</td>\n",
       "      <td>-0.160586</td>\n",
       "      <td>0.161175</td>\n",
       "      <td>0.044315</td>\n",
       "      <td>1.013805</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 38 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      month       day      hour  latitude  longitude  time_indicator  \\\n",
       "0 -1.401968 -1.687158 -1.699598 -0.320314   0.004165       -0.016126   \n",
       "1 -1.401968 -1.687158 -1.699598  0.141247  -0.817914       -0.016126   \n",
       "2 -1.401968 -1.687158 -1.699598  0.756662  -1.683260       -0.016126   \n",
       "3 -1.401968 -1.687158 -1.699598  1.013085  -1.207320       -0.016126   \n",
       "4 -1.401968 -1.687158 -1.699598 -0.063891   0.999313       -0.016126   \n",
       "\n",
       "   ship_course  ship_speed  id_indicator  wind_direction_true      ...        \\\n",
       "0     1.066827    0.803293      0.160768             1.211566      ...         \n",
       "1     1.066827    0.803293      0.160768             1.315335      ...         \n",
       "2     1.066827    0.803293      0.160768             1.522872      ...         \n",
       "3     1.066827    0.803293      0.160768             1.626641      ...         \n",
       "4     1.066827    0.803293      0.160768            -1.278883      ...         \n",
       "\n",
       "   swell_direction  swell_period  swell_height  ten_degree_box_number  \\\n",
       "0         0.884043     -2.842444      2.850838               0.617359   \n",
       "1         0.884043     -2.842444      2.850838               0.617359   \n",
       "2         0.884043     -2.842444      2.850838              -1.619772   \n",
       "3         0.884043     -2.842444      2.850838              -1.619772   \n",
       "4         0.884043     -2.842444      2.850838               0.617359   \n",
       "\n",
       "   one_degree_box_number     deck  source_id  platform_type  dup_status  \\\n",
       "0               0.676268 -0.16058  -0.160586       0.161175    0.044315   \n",
       "1               1.061496 -0.16058  -0.160586       0.161175    0.044315   \n",
       "2              -2.020331 -0.16058  -0.160586       0.161175    0.044315   \n",
       "3              -1.705144 -0.16058  -0.160586       0.161175    0.044315   \n",
       "4               0.921413 -0.16058  -0.160586       0.161175    0.044315   \n",
       "\n",
       "   nightday_flag  \n",
       "0       1.013805  \n",
       "1       1.013805  \n",
       "2       1.013805  \n",
       "3       1.013805  \n",
       "4       1.013805  \n",
       "\n",
       "[5 rows x 38 columns]"
      ]
     },
     "execution_count": 154,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "normalized_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 155,
   "metadata": {},
   "outputs": [],
   "source": [
    "Xn = normalized_df.values\n",
    "yn = df['wave_height'].values\n",
    "\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "Xn_train, Xn_val, yn_train, yn_val = train_test_split(Xn,yn,test_size=0.25)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**MLP**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 166,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE: 45.79301256742124\n"
     ]
    }
   ],
   "source": [
    "from sklearn.neural_network import MLPClassifier\n",
    "from sklearn.model_selection import validation_curve\n",
    "MLP = MLPClassifier(alpha=0.005, max_iter=500)\n",
    "MLP.fit(Xn_train, yn_train)\n",
    "yn_pred = MLP.predict(Xn_val)\n",
    "\n",
    "from sklearn import metrics\n",
    "print('RMSE:', np.sqrt(metrics.mean_squared_error(yn_val, yn_pred)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**RF**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 237,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Amartya Sen\\Anaconda3\\lib\\site-packages\\sklearn\\ensemble\\weight_boosting.py:29: DeprecationWarning: numpy.core.umath_tests is an internal NumPy module and should not be imported. It will be removed in a future NumPy release.\n",
      "  from numpy.core.umath_tests import inner1d\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE: 30.88365263371546\n"
     ]
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "RF = RandomForestClassifier(n_estimators=500)\n",
    "RF.fit(Xn_train, yn_train)\n",
    "yn_pred = RF.predict(Xn_val)\n",
    "\n",
    "from sklearn import metrics\n",
    "print('RMSE:', np.sqrt(metrics.mean_squared_error(yn_val, yn_pred)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**XGB Regressor**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 220,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE: 46.98352317658346\n"
     ]
    }
   ],
   "source": [
    "from xgboost import XGBRegressor\n",
    "XG = XGBRegressor()\n",
    "XG.fit(Xn_train, yn_train)\n",
    "yn_pred = XG.predict(Xn_val)\n",
    "\n",
    "from sklearn import metrics\n",
    "print('RMSE:', np.sqrt(metrics.mean_squared_error(yn_val, yn_pred)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**LR without regularisation**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 235,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.cross_validation import train_test_split\n",
    "t=df.drop(['index', 'wave_height', 'source_exclusion_flags', 'dpt_indicator'], axis=1)\n",
    "t=df.drop(m, axis=1)\n",
    "X=t.values\n",
    "y=df['wave_height'].values\n",
    "X_train, X_test, y_train, y_test=train_test_split(X,y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 236,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE: 70.78572069501624\n"
     ]
    }
   ],
   "source": [
    "from sklearn.linear_model import LinearRegression\n",
    "linreg = LinearRegression()\n",
    "linreg.fit(Xn_train,yn_train)\n",
    "yn_pred = linreg.predict(Xn_val)\n",
    "\n",
    "from sklearn import metrics\n",
    "print('RMSE:', np.sqrt(metrics.mean_squared_error(yn_val, yn_pred)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**SUBMISSION**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 239,
   "metadata": {},
   "outputs": [],
   "source": [
    "test=pd.read_csv(\"test__updated.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 240,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    837.0\n",
       "mean       0.0\n",
       "std        0.0\n",
       "min        0.0\n",
       "25%        0.0\n",
       "50%        0.0\n",
       "75%        0.0\n",
       "max        0.0\n",
       "Name: wbt_indicator, dtype: float64"
      ]
     },
     "execution_count": 240,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test['wbt_indicator'].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 241,
   "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>month</th>\n",
       "      <th>day</th>\n",
       "      <th>hour</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>time_indicator</th>\n",
       "      <th>ship_course</th>\n",
       "      <th>ship_speed</th>\n",
       "      <th>id_indicator</th>\n",
       "      <th>wind_direction_true</th>\n",
       "      <th>...</th>\n",
       "      <th>swell_direction</th>\n",
       "      <th>swell_period</th>\n",
       "      <th>swell_height</th>\n",
       "      <th>ten_degree_box_number</th>\n",
       "      <th>one_degree_box_number</th>\n",
       "      <th>deck</th>\n",
       "      <th>source_id</th>\n",
       "      <th>platform_type</th>\n",
       "      <th>dup_status</th>\n",
       "      <th>nightday_flag</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-2.305067</td>\n",
       "      <td>1.568219</td>\n",
       "      <td>-1.003066</td>\n",
       "      <td>-0.365400</td>\n",
       "      <td>1.289117</td>\n",
       "      <td>-0.014797</td>\n",
       "      <td>-0.606596</td>\n",
       "      <td>0.482889</td>\n",
       "      <td>0.259229</td>\n",
       "      <td>0.095535</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.062208</td>\n",
       "      <td>-1.180758</td>\n",
       "      <td>-0.408347</td>\n",
       "      <td>0.595272</td>\n",
       "      <td>0.656660</td>\n",
       "      <td>-0.258886</td>\n",
       "      <td>-0.258911</td>\n",
       "      <td>0.259932</td>\n",
       "      <td>0.038822</td>\n",
       "      <td>-1.312399</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-2.305067</td>\n",
       "      <td>1.568219</td>\n",
       "      <td>-0.978467</td>\n",
       "      <td>0.103593</td>\n",
       "      <td>1.063673</td>\n",
       "      <td>-0.014797</td>\n",
       "      <td>-0.606596</td>\n",
       "      <td>0.482889</td>\n",
       "      <td>0.259229</td>\n",
       "      <td>-1.543299</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.062208</td>\n",
       "      <td>-1.180758</td>\n",
       "      <td>-0.408347</td>\n",
       "      <td>0.595272</td>\n",
       "      <td>1.015602</td>\n",
       "      <td>-0.258886</td>\n",
       "      <td>-0.258911</td>\n",
       "      <td>0.259932</td>\n",
       "      <td>0.038822</td>\n",
       "      <td>-1.312399</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-2.305067</td>\n",
       "      <td>1.568219</td>\n",
       "      <td>-0.978467</td>\n",
       "      <td>0.384988</td>\n",
       "      <td>0.477519</td>\n",
       "      <td>-0.014797</td>\n",
       "      <td>-0.606596</td>\n",
       "      <td>0.482889</td>\n",
       "      <td>0.259229</td>\n",
       "      <td>1.149071</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.062208</td>\n",
       "      <td>-1.180758</td>\n",
       "      <td>-0.408347</td>\n",
       "      <td>0.595272</td>\n",
       "      <td>1.087390</td>\n",
       "      <td>-0.258886</td>\n",
       "      <td>-0.258911</td>\n",
       "      <td>0.259932</td>\n",
       "      <td>0.038822</td>\n",
       "      <td>-1.312399</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-2.305067</td>\n",
       "      <td>1.568219</td>\n",
       "      <td>-0.978467</td>\n",
       "      <td>2.214060</td>\n",
       "      <td>-1.235853</td>\n",
       "      <td>-0.014797</td>\n",
       "      <td>-0.606596</td>\n",
       "      <td>0.482889</td>\n",
       "      <td>0.259229</td>\n",
       "      <td>-0.372703</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.062208</td>\n",
       "      <td>-1.180758</td>\n",
       "      <td>-0.408347</td>\n",
       "      <td>-1.679859</td>\n",
       "      <td>-0.958578</td>\n",
       "      <td>-0.258886</td>\n",
       "      <td>-0.258911</td>\n",
       "      <td>0.259932</td>\n",
       "      <td>0.038822</td>\n",
       "      <td>-1.312399</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-2.305067</td>\n",
       "      <td>1.568219</td>\n",
       "      <td>-0.978467</td>\n",
       "      <td>-1.021990</td>\n",
       "      <td>0.477519</td>\n",
       "      <td>-0.014797</td>\n",
       "      <td>-0.606596</td>\n",
       "      <td>0.482889</td>\n",
       "      <td>0.259229</td>\n",
       "      <td>0.446714</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.062208</td>\n",
       "      <td>-1.180758</td>\n",
       "      <td>-0.408347</td>\n",
       "      <td>0.595272</td>\n",
       "      <td>0.010565</td>\n",
       "      <td>-0.258886</td>\n",
       "      <td>-0.258911</td>\n",
       "      <td>0.259932</td>\n",
       "      <td>0.038822</td>\n",
       "      <td>-1.312399</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 38 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      month       day      hour  latitude  longitude  time_indicator  \\\n",
       "0 -2.305067  1.568219 -1.003066 -0.365400   1.289117       -0.014797   \n",
       "1 -2.305067  1.568219 -0.978467  0.103593   1.063673       -0.014797   \n",
       "2 -2.305067  1.568219 -0.978467  0.384988   0.477519       -0.014797   \n",
       "3 -2.305067  1.568219 -0.978467  2.214060  -1.235853       -0.014797   \n",
       "4 -2.305067  1.568219 -0.978467 -1.021990   0.477519       -0.014797   \n",
       "\n",
       "   ship_course  ship_speed  id_indicator  wind_direction_true      ...        \\\n",
       "0    -0.606596    0.482889      0.259229             0.095535      ...         \n",
       "1    -0.606596    0.482889      0.259229            -1.543299      ...         \n",
       "2    -0.606596    0.482889      0.259229             1.149071      ...         \n",
       "3    -0.606596    0.482889      0.259229            -0.372703      ...         \n",
       "4    -0.606596    0.482889      0.259229             0.446714      ...         \n",
       "\n",
       "   swell_direction  swell_period  swell_height  ten_degree_box_number  \\\n",
       "0        -2.062208     -1.180758     -0.408347               0.595272   \n",
       "1        -2.062208     -1.180758     -0.408347               0.595272   \n",
       "2        -2.062208     -1.180758     -0.408347               0.595272   \n",
       "3        -2.062208     -1.180758     -0.408347              -1.679859   \n",
       "4        -2.062208     -1.180758     -0.408347               0.595272   \n",
       "\n",
       "   one_degree_box_number      deck  source_id  platform_type  dup_status  \\\n",
       "0               0.656660 -0.258886  -0.258911       0.259932    0.038822   \n",
       "1               1.015602 -0.258886  -0.258911       0.259932    0.038822   \n",
       "2               1.087390 -0.258886  -0.258911       0.259932    0.038822   \n",
       "3              -0.958578 -0.258886  -0.258911       0.259932    0.038822   \n",
       "4               0.010565 -0.258886  -0.258911       0.259932    0.038822   \n",
       "\n",
       "   nightday_flag  \n",
       "0      -1.312399  \n",
       "1      -1.312399  \n",
       "2      -1.312399  \n",
       "3      -1.312399  \n",
       "4      -1.312399  \n",
       "\n",
       "[5 rows x 38 columns]"
      ]
     },
     "execution_count": 241,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test.drop(['index', 'source_exclusion_flags', 'dpt_indicator'], axis=1, inplace=True)\n",
    "test.drop(m, axis=1, inplace=True)\n",
    "test.fillna(method='ffill', inplace=True)\n",
    "test.fillna(method='bfill', inplace=True)\n",
    "test=(test-test.mean())/test.std()\n",
    "test.isnull().sum()\n",
    "test.fillna(0, inplace=True)\n",
    "test.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 242,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_test=test.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 243,
   "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>month</th>\n",
       "      <th>day</th>\n",
       "      <th>hour</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>time_indicator</th>\n",
       "      <th>ship_course</th>\n",
       "      <th>ship_speed</th>\n",
       "      <th>id_indicator</th>\n",
       "      <th>wind_direction_true</th>\n",
       "      <th>...</th>\n",
       "      <th>swell_direction</th>\n",
       "      <th>swell_period</th>\n",
       "      <th>swell_height</th>\n",
       "      <th>ten_degree_box_number</th>\n",
       "      <th>one_degree_box_number</th>\n",
       "      <th>deck</th>\n",
       "      <th>source_id</th>\n",
       "      <th>platform_type</th>\n",
       "      <th>dup_status</th>\n",
       "      <th>nightday_flag</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-2.305067</td>\n",
       "      <td>1.568219</td>\n",
       "      <td>-1.003066</td>\n",
       "      <td>-0.365400</td>\n",
       "      <td>1.289117</td>\n",
       "      <td>-0.014797</td>\n",
       "      <td>-0.606596</td>\n",
       "      <td>0.482889</td>\n",
       "      <td>0.259229</td>\n",
       "      <td>0.095535</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.062208</td>\n",
       "      <td>-1.180758</td>\n",
       "      <td>-0.408347</td>\n",
       "      <td>0.595272</td>\n",
       "      <td>0.656660</td>\n",
       "      <td>-0.258886</td>\n",
       "      <td>-0.258911</td>\n",
       "      <td>0.259932</td>\n",
       "      <td>0.038822</td>\n",
       "      <td>-1.312399</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-2.305067</td>\n",
       "      <td>1.568219</td>\n",
       "      <td>-0.978467</td>\n",
       "      <td>0.103593</td>\n",
       "      <td>1.063673</td>\n",
       "      <td>-0.014797</td>\n",
       "      <td>-0.606596</td>\n",
       "      <td>0.482889</td>\n",
       "      <td>0.259229</td>\n",
       "      <td>-1.543299</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.062208</td>\n",
       "      <td>-1.180758</td>\n",
       "      <td>-0.408347</td>\n",
       "      <td>0.595272</td>\n",
       "      <td>1.015602</td>\n",
       "      <td>-0.258886</td>\n",
       "      <td>-0.258911</td>\n",
       "      <td>0.259932</td>\n",
       "      <td>0.038822</td>\n",
       "      <td>-1.312399</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-2.305067</td>\n",
       "      <td>1.568219</td>\n",
       "      <td>-0.978467</td>\n",
       "      <td>0.384988</td>\n",
       "      <td>0.477519</td>\n",
       "      <td>-0.014797</td>\n",
       "      <td>-0.606596</td>\n",
       "      <td>0.482889</td>\n",
       "      <td>0.259229</td>\n",
       "      <td>1.149071</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.062208</td>\n",
       "      <td>-1.180758</td>\n",
       "      <td>-0.408347</td>\n",
       "      <td>0.595272</td>\n",
       "      <td>1.087390</td>\n",
       "      <td>-0.258886</td>\n",
       "      <td>-0.258911</td>\n",
       "      <td>0.259932</td>\n",
       "      <td>0.038822</td>\n",
       "      <td>-1.312399</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-2.305067</td>\n",
       "      <td>1.568219</td>\n",
       "      <td>-0.978467</td>\n",
       "      <td>2.214060</td>\n",
       "      <td>-1.235853</td>\n",
       "      <td>-0.014797</td>\n",
       "      <td>-0.606596</td>\n",
       "      <td>0.482889</td>\n",
       "      <td>0.259229</td>\n",
       "      <td>-0.372703</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.062208</td>\n",
       "      <td>-1.180758</td>\n",
       "      <td>-0.408347</td>\n",
       "      <td>-1.679859</td>\n",
       "      <td>-0.958578</td>\n",
       "      <td>-0.258886</td>\n",
       "      <td>-0.258911</td>\n",
       "      <td>0.259932</td>\n",
       "      <td>0.038822</td>\n",
       "      <td>-1.312399</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-2.305067</td>\n",
       "      <td>1.568219</td>\n",
       "      <td>-0.978467</td>\n",
       "      <td>-1.021990</td>\n",
       "      <td>0.477519</td>\n",
       "      <td>-0.014797</td>\n",
       "      <td>-0.606596</td>\n",
       "      <td>0.482889</td>\n",
       "      <td>0.259229</td>\n",
       "      <td>0.446714</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.062208</td>\n",
       "      <td>-1.180758</td>\n",
       "      <td>-0.408347</td>\n",
       "      <td>0.595272</td>\n",
       "      <td>0.010565</td>\n",
       "      <td>-0.258886</td>\n",
       "      <td>-0.258911</td>\n",
       "      <td>0.259932</td>\n",
       "      <td>0.038822</td>\n",
       "      <td>-1.312399</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 38 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      month       day      hour  latitude  longitude  time_indicator  \\\n",
       "0 -2.305067  1.568219 -1.003066 -0.365400   1.289117       -0.014797   \n",
       "1 -2.305067  1.568219 -0.978467  0.103593   1.063673       -0.014797   \n",
       "2 -2.305067  1.568219 -0.978467  0.384988   0.477519       -0.014797   \n",
       "3 -2.305067  1.568219 -0.978467  2.214060  -1.235853       -0.014797   \n",
       "4 -2.305067  1.568219 -0.978467 -1.021990   0.477519       -0.014797   \n",
       "\n",
       "   ship_course  ship_speed  id_indicator  wind_direction_true      ...        \\\n",
       "0    -0.606596    0.482889      0.259229             0.095535      ...         \n",
       "1    -0.606596    0.482889      0.259229            -1.543299      ...         \n",
       "2    -0.606596    0.482889      0.259229             1.149071      ...         \n",
       "3    -0.606596    0.482889      0.259229            -0.372703      ...         \n",
       "4    -0.606596    0.482889      0.259229             0.446714      ...         \n",
       "\n",
       "   swell_direction  swell_period  swell_height  ten_degree_box_number  \\\n",
       "0        -2.062208     -1.180758     -0.408347               0.595272   \n",
       "1        -2.062208     -1.180758     -0.408347               0.595272   \n",
       "2        -2.062208     -1.180758     -0.408347               0.595272   \n",
       "3        -2.062208     -1.180758     -0.408347              -1.679859   \n",
       "4        -2.062208     -1.180758     -0.408347               0.595272   \n",
       "\n",
       "   one_degree_box_number      deck  source_id  platform_type  dup_status  \\\n",
       "0               0.656660 -0.258886  -0.258911       0.259932    0.038822   \n",
       "1               1.015602 -0.258886  -0.258911       0.259932    0.038822   \n",
       "2               1.087390 -0.258886  -0.258911       0.259932    0.038822   \n",
       "3              -0.958578 -0.258886  -0.258911       0.259932    0.038822   \n",
       "4               0.010565 -0.258886  -0.258911       0.259932    0.038822   \n",
       "\n",
       "   nightday_flag  \n",
       "0      -1.312399  \n",
       "1      -1.312399  \n",
       "2      -1.312399  \n",
       "3      -1.312399  \n",
       "4      -1.312399  \n",
       "\n",
       "[5 rows x 38 columns]"
      ]
     },
     "execution_count": 243,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 244,
   "metadata": {},
   "outputs": [],
   "source": [
    "abc=pd.read_csv(\"test__updated.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 245,
   "metadata": {},
   "outputs": [],
   "source": [
    "pred=RF.predict(X_test)\n",
    "dfans=pd.DataFrame({\n",
    "    'index':abc['index'],\n",
    "    'wave_height':pred\n",
    "})\n",
    "dfans.to_csv('predictions.csv',index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.7.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}