seq2seq-time/notebooks/00.02-mc-load_data.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-10-25T07:08:04.786938Z",
     "start_time": "2020-10-25T07:08:04.731940Z"
    }
   },
   "source": [
    "Here I load multiple multivariate timeseries regression datasets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-10-26T06:24:49.597542Z",
     "start_time": "2020-10-26T06:24:49.140731Z"
    }
   },
   "outputs": [],
   "source": [
    "# OPTIONAL: Load the \"autoreload\" extension so that code can change. But blacklist large modules\n",
    "%load_ext autoreload\n",
    "%autoreload 2\n",
    "%aimport -pandas\n",
    "%aimport -torch\n",
    "%aimport -numpy\n",
    "%aimport -matplotlib\n",
    "%aimport -dask\n",
    "%aimport -tqdm\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-10-26T06:24:53.015854Z",
     "start_time": "2020-10-26T06:24:49.600031Z"
    }
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "plt.rcParams['figure.figsize'] = (12.0, 3.0)\n",
    "plt.style.use('ggplot')\n",
    "\n",
    "import holoviews as hv\n",
    "from holoviews.operation.datashader import datashade, dynspread"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-10-26T06:24:53.571950Z",
     "start_time": "2020-10-26T06:24:53.019038Z"
    }
   },
   "outputs": [],
   "source": [
    "import os\n",
    "from tqdm.auto import tqdm\n",
    "from pathlib import Path\n",
    "from torchvision.datasets.utils import download_url, extract_archive, download_and_extract_archive"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-10-26T06:24:53.998896Z",
     "start_time": "2020-10-26T06:24:53.574661Z"
    }
   },
   "outputs": [],
   "source": [
    "import sklearn\n",
    "from sklearn.preprocessing import StandardScaler, OrdinalEncoder\n",
    "from sklearn_pandas import DataFrameMapper\n",
    "\n",
    "def normalize_encode_dataframe(df, encoder=OrdinalEncoder):\n",
    "    \"\"\"Normalise numeric data, encode categorical data.\"\"\"\n",
    "    columns_input_numeric = list(df._get_numeric_data().columns)\n",
    "    columns_categorical = list(set(df.columns)-set(columns_input_numeric))\n",
    "    \n",
    "    transformers= [([n], StandardScaler()) for n in columns_input_numeric] + \\\n",
    "                  [([n], encoder()) for n in columns_categorical]\n",
    "    scaler = DataFrameMapper(transformers, df_out=True)\n",
    "    df_norm = scaler.fit_transform(df)\n",
    "    return df_norm, scaler\n",
    "    \n",
    "def timeseries_split(df, test_fraction=0.2):\n",
    "    \"\"\"Split timeseries data with test in the future\"\"\"\n",
    "    i = int(len(df)*test_fraction)\n",
    "    return df.iloc[:i], df.iloc[i:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-10-26T06:24:54.050092Z",
     "start_time": "2020-10-26T06:24:54.001084Z"
    }
   },
   "outputs": [],
   "source": [
    "from typing import List, Tuple\n",
    "from seq2seq_time.data.dataset import Seq2SeqDataSet"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-10-26T06:24:54.095015Z",
     "start_time": "2020-10-26T06:24:54.052468Z"
    }
   },
   "outputs": [],
   "source": [
    "datasets_root = Path('../data/processed/')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-10-26T06:24:54.173346Z",
     "start_time": "2020-10-26T06:24:54.097401Z"
    }
   },
   "outputs": [],
   "source": [
    "class RegressionForecastData:   \n",
    "    columns_forecast = None # The input colums which can be included in future (e.g. week or weather forecast)\n",
    "    columns_target = None # Target columns\n",
    "    \n",
    "    def __init__(self, datasets_root):        \n",
    "        self.datasets_root = datasets_root\n",
    "        \n",
    "        # Process data\n",
    "        self.df = self.download()        \n",
    "        self.df_norm, self.scaler = self.normalize(self.df)\n",
    "        self.output_scaler = next(filter(lambda r:r[0][0] in self.columns_target, self.scaler.features))[-1]\n",
    "        self.df_train, self.df_test = self.split(self.df_norm)\n",
    "        \n",
    "        # Check processing\n",
    "        self.check()\n",
    "    \n",
    "    def download(self) -> pd.DataFrame:\n",
    "        \"\"\"Implement this method to download data and return raw df\"\"\"\n",
    "        raise NotImplementedError()\n",
    "        return df\n",
    "    \n",
    "    def normalize(self, df) -> Tuple[pd.DataFrame, DataFrameMapper]:\n",
    "        df_norm, scaler = normalize_encode_dataframe(df)\n",
    "        return df_norm, scaler\n",
    "    \n",
    "    def split(self, df_norm: pd.DataFrame) -> Tuple[pd.DataFrame, pd.DataFrame]:\n",
    "        df_train, df_test = timeseries_split(df_norm)\n",
    "        return df_train, df_test \n",
    "    \n",
    "    def check(self) -> None:\n",
    "        \"\"\"Check the resulting dataframe\"\"\"\n",
    "        assert isinstance(self.df.index, pd.DatetimeIndex), 'index must be datetime'\n",
    "        assert self.df.index.freq is not None, 'df must have freq'        \n",
    "        assert self.columns_forecast is not None\n",
    "        assert self.columns_target is not None\n",
    "        assert ~set(self.columns_target).issubset(set(self.columns_forecast)), 'target columns should not be in forecast'\n",
    "        assert set(self.columns_forecast).issubset(set(self.df.columns)), 'columns_forecast must be in df'\n",
    "        assert set(self.columns_target).issubset(set(self.df.columns)), 'columns_target must be in df'\n",
    "        \n",
    "    def to_datasets(self, window_past: int, window_future: int, valid:bool=False) -> Tuple[Seq2SeqDataSet, Seq2SeqDataSet]:\n",
    "        \"\"\"Convert to torch datasets\"\"\"\n",
    "        ds_train = Seq2SeqDataSet(df_train, window_past=window_past, window_future=window_future, columns_target=self.columns_target, columns_past=self.columns_past)\n",
    "        ds_test = Seq2SeqDataSet(df_test, window_past=window_past, window_future=window_future, columns_target=self.columns_target, columns_past=self.columns_past)\n",
    "        return ds_train, ds_test\n",
    "    \n",
    "    def __repr__(self):\n",
    "        return f'<{type(self).__name__} {self.df.shape if (self.df is not None) else None}>'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-10-26T06:25:08.694514Z",
     "start_time": "2020-10-26T06:24:54.177083Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<GasSensor (303034, 6)>\n"
     ]
    },
    {
     "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>CO (ppm)</th>\n",
       "      <th>Humidity (%r.h.)</th>\n",
       "      <th>Temperature (C)</th>\n",
       "      <th>Flow rate (mL/min)</th>\n",
       "      <th>Heater voltage (V)</th>\n",
       "      <th>R1 (MOhm)</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Time (s)</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>2016-10-16 05:36:55.800</th>\n",
       "      <td>0.0</td>\n",
       "      <td>48.4700</td>\n",
       "      <td>24.6200</td>\n",
       "      <td>247.4926</td>\n",
       "      <td>0.2000</td>\n",
       "      <td>0.6831</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-10-16 05:36:56.100</th>\n",
       "      <td>0.0</td>\n",
       "      <td>48.4700</td>\n",
       "      <td>24.6200</td>\n",
       "      <td>243.8282</td>\n",
       "      <td>0.1998</td>\n",
       "      <td>0.6649</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-10-16 05:36:56.400</th>\n",
       "      <td>0.0</td>\n",
       "      <td>48.4700</td>\n",
       "      <td>24.6200</td>\n",
       "      <td>243.0668</td>\n",
       "      <td>0.2000</td>\n",
       "      <td>0.6481</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-10-16 05:36:56.700</th>\n",
       "      <td>0.0</td>\n",
       "      <td>48.4700</td>\n",
       "      <td>24.6200</td>\n",
       "      <td>242.3030</td>\n",
       "      <td>0.2000</td>\n",
       "      <td>0.6318</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-10-16 05:36:57.000</th>\n",
       "      <td>0.0</td>\n",
       "      <td>48.4702</td>\n",
       "      <td>24.6206</td>\n",
       "      <td>241.5632</td>\n",
       "      <td>0.2000</td>\n",
       "      <td>0.6178</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-10-17 06:52:04.500</th>\n",
       "      <td>0.0</td>\n",
       "      <td>63.9400</td>\n",
       "      <td>24.6200</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.2080</td>\n",
       "      <td>9.9322</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-10-17 06:52:04.800</th>\n",
       "      <td>0.0</td>\n",
       "      <td>63.9400</td>\n",
       "      <td>24.6200</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.2050</td>\n",
       "      <td>31.7887</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-10-17 06:52:05.100</th>\n",
       "      <td>0.0</td>\n",
       "      <td>63.9400</td>\n",
       "      <td>24.6200</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.2040</td>\n",
       "      <td>57.7304</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-10-17 06:52:05.400</th>\n",
       "      <td>0.0</td>\n",
       "      <td>63.9400</td>\n",
       "      <td>24.6200</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.2020</td>\n",
       "      <td>71.9176</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-10-17 06:52:05.700</th>\n",
       "      <td>0.0</td>\n",
       "      <td>63.9400</td>\n",
       "      <td>24.6200</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.2010</td>\n",
       "      <td>61.8035</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>303034 rows × 6 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                         CO (ppm)  Humidity (%r.h.)  Temperature (C)  \\\n",
       "Time (s)                                                               \n",
       "2016-10-16 05:36:55.800       0.0           48.4700          24.6200   \n",
       "2016-10-16 05:36:56.100       0.0           48.4700          24.6200   \n",
       "2016-10-16 05:36:56.400       0.0           48.4700          24.6200   \n",
       "2016-10-16 05:36:56.700       0.0           48.4700          24.6200   \n",
       "2016-10-16 05:36:57.000       0.0           48.4702          24.6206   \n",
       "...                           ...               ...              ...   \n",
       "2016-10-17 06:52:04.500       0.0           63.9400          24.6200   \n",
       "2016-10-17 06:52:04.800       0.0           63.9400          24.6200   \n",
       "2016-10-17 06:52:05.100       0.0           63.9400          24.6200   \n",
       "2016-10-17 06:52:05.400       0.0           63.9400          24.6200   \n",
       "2016-10-17 06:52:05.700       0.0           63.9400          24.6200   \n",
       "\n",
       "                         Flow rate (mL/min)  Heater voltage (V)  R1 (MOhm)  \n",
       "Time (s)                                                                    \n",
       "2016-10-16 05:36:55.800            247.4926              0.2000     0.6831  \n",
       "2016-10-16 05:36:56.100            243.8282              0.1998     0.6649  \n",
       "2016-10-16 05:36:56.400            243.0668              0.2000     0.6481  \n",
       "2016-10-16 05:36:56.700            242.3030              0.2000     0.6318  \n",
       "2016-10-16 05:36:57.000            241.5632              0.2000     0.6178  \n",
       "...                                     ...                 ...        ...  \n",
       "2016-10-17 06:52:04.500              0.0000              0.2080     9.9322  \n",
       "2016-10-17 06:52:04.800              0.0000              0.2050    31.7887  \n",
       "2016-10-17 06:52:05.100              0.0000              0.2040    57.7304  \n",
       "2016-10-17 06:52:05.400              0.0000              0.2020    71.9176  \n",
       "2016-10-17 06:52:05.700              0.0000              0.2010    61.8035  \n",
       "\n",
       "[303034 rows x 6 columns]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x216 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "class GasSensor(RegressionForecastData):\n",
    "    \"\"\"\n",
    "    See: http://archive.ics.uci.edu/ml/datasets/Gas+sensor+array+temperature+modulation\n",
    "    \"\"\"\n",
    "    \n",
    "    columns_target = ['R1 (MOhm)']\n",
    "    columns_forecast = ['Flow rate (mL/min)', 'Heater voltage (V)']\n",
    "    \n",
    "    def download(self):\n",
    "        url = 'http://archive.ics.uci.edu/ml/machine-learning-databases/00487/gas-sensor-array-temperature-modulation.zip'\n",
    "        \n",
    "        # download if needed\n",
    "        extract_path = self.datasets_root/'GasSensor'\n",
    "        files = sorted(extract_path.glob('*.csv'))\n",
    "        if len(files)<13:\n",
    "            print('download_and_extract_archive')\n",
    "            download_and_extract_archive(url, self.datasets_root, extract_path)\n",
    "        \n",
    "        # Load csv's\n",
    "        files = sorted(extract_path.glob('*.csv'))\n",
    "        dfs = []\n",
    "        for f in files:\n",
    "            now = pd.to_datetime(f.stem, format='%Y%m%d_%H%M%S')\n",
    "            df = pd.read_csv(f)\n",
    "            df.index = pd.to_timedelta(df['Time (s)'], unit='s') + now\n",
    "            dfs.append(df)\n",
    "        self.df = pd.concat(dfs).dropna(subset=self.columns_target)\n",
    "\n",
    "        df = df[[ 'CO (ppm)', 'Humidity (%r.h.)', 'Temperature (C)',\n",
    "               'Flow rate (mL/min)', 'Heater voltage (V)', 'R1 (MOhm)']]\n",
    "        df = df.resample('0.3S').first()\n",
    "        \n",
    "        return df\n",
    "dataset = GasSensor(datasets_root)\n",
    "print(dataset)\n",
    "dataset.df[dataset.columns_target].head(1000).plot()\n",
    "dataset.df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-10-26T01:01:04.653709Z",
     "start_time": "2020-10-26T01:01:04.524050Z"
    }
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-10-26T06:25:12.274174Z",
     "start_time": "2020-10-26T06:25:08.697022Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading https://archive.ics.uci.edu/ml/machine-learning-databases/00492/Metro_Interstate_Traffic_Volume.csv.gz to ../data/processed/00492_Metro_Interstate_Traffic_Volume.csv.gz\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "364da475e08247948c34e2a122b03920",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(HTML(value=''), FloatProgress(value=1.0, bar_style='info', layout=Layout(width='20px'), max=1.0…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<MetroInterstateTraffic (52551, 14)>\n",
      "\n"
     ]
    },
    {
     "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>holiday</th>\n",
       "      <th>temp</th>\n",
       "      <th>rain_1h</th>\n",
       "      <th>snow_1h</th>\n",
       "      <th>clouds_all</th>\n",
       "      <th>weather_main</th>\n",
       "      <th>weather_description</th>\n",
       "      <th>traffic_volume</th>\n",
       "      <th>month</th>\n",
       "      <th>day</th>\n",
       "      <th>week</th>\n",
       "      <th>hour</th>\n",
       "      <th>minute</th>\n",
       "      <th>dayofweek</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>date_time</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>2012-10-02 09:00:00</th>\n",
       "      <td>True</td>\n",
       "      <td>288.28</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>40.0</td>\n",
       "      <td>Clouds</td>\n",
       "      <td>scattered clouds</td>\n",
       "      <td>5545.0</td>\n",
       "      <td>10</td>\n",
       "      <td>2</td>\n",
       "      <td>40</td>\n",
       "      <td>9</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2012-10-02 10:00:00</th>\n",
       "      <td>True</td>\n",
       "      <td>289.36</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>75.0</td>\n",
       "      <td>Clouds</td>\n",
       "      <td>broken clouds</td>\n",
       "      <td>4516.0</td>\n",
       "      <td>10</td>\n",
       "      <td>2</td>\n",
       "      <td>40</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2012-10-02 11:00:00</th>\n",
       "      <td>True</td>\n",
       "      <td>289.58</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>90.0</td>\n",
       "      <td>Clouds</td>\n",
       "      <td>overcast clouds</td>\n",
       "      <td>4767.0</td>\n",
       "      <td>10</td>\n",
       "      <td>2</td>\n",
       "      <td>40</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2012-10-02 12:00:00</th>\n",
       "      <td>True</td>\n",
       "      <td>290.13</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>90.0</td>\n",
       "      <td>Clouds</td>\n",
       "      <td>overcast clouds</td>\n",
       "      <td>5026.0</td>\n",
       "      <td>10</td>\n",
       "      <td>2</td>\n",
       "      <td>40</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2012-10-02 13:00:00</th>\n",
       "      <td>True</td>\n",
       "      <td>291.14</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>75.0</td>\n",
       "      <td>Clouds</td>\n",
       "      <td>broken clouds</td>\n",
       "      <td>4918.0</td>\n",
       "      <td>10</td>\n",
       "      <td>2</td>\n",
       "      <td>40</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>1</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>2018-09-30 19:00:00</th>\n",
       "      <td>True</td>\n",
       "      <td>283.45</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>75.0</td>\n",
       "      <td>Clouds</td>\n",
       "      <td>broken clouds</td>\n",
       "      <td>3543.0</td>\n",
       "      <td>9</td>\n",
       "      <td>30</td>\n",
       "      <td>39</td>\n",
       "      <td>19</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-09-30 20:00:00</th>\n",
       "      <td>True</td>\n",
       "      <td>282.76</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>90.0</td>\n",
       "      <td>Clouds</td>\n",
       "      <td>overcast clouds</td>\n",
       "      <td>2781.0</td>\n",
       "      <td>9</td>\n",
       "      <td>30</td>\n",
       "      <td>39</td>\n",
       "      <td>20</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-09-30 21:00:00</th>\n",
       "      <td>True</td>\n",
       "      <td>282.73</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>90.0</td>\n",
       "      <td>Thunderstorm</td>\n",
       "      <td>proximity thunderstorm</td>\n",
       "      <td>2159.0</td>\n",
       "      <td>9</td>\n",
       "      <td>30</td>\n",
       "      <td>39</td>\n",
       "      <td>21</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-09-30 22:00:00</th>\n",
       "      <td>True</td>\n",
       "      <td>282.09</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>90.0</td>\n",
       "      <td>Clouds</td>\n",
       "      <td>overcast clouds</td>\n",
       "      <td>1450.0</td>\n",
       "      <td>9</td>\n",
       "      <td>30</td>\n",
       "      <td>39</td>\n",
       "      <td>22</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-09-30 23:00:00</th>\n",
       "      <td>True</td>\n",
       "      <td>282.12</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>90.0</td>\n",
       "      <td>Clouds</td>\n",
       "      <td>overcast clouds</td>\n",
       "      <td>954.0</td>\n",
       "      <td>9</td>\n",
       "      <td>30</td>\n",
       "      <td>39</td>\n",
       "      <td>23</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>52551 rows × 14 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                     holiday    temp  rain_1h  snow_1h  clouds_all  \\\n",
       "date_time                                                            \n",
       "2012-10-02 09:00:00     True  288.28      0.0      0.0        40.0   \n",
       "2012-10-02 10:00:00     True  289.36      0.0      0.0        75.0   \n",
       "2012-10-02 11:00:00     True  289.58      0.0      0.0        90.0   \n",
       "2012-10-02 12:00:00     True  290.13      0.0      0.0        90.0   \n",
       "2012-10-02 13:00:00     True  291.14      0.0      0.0        75.0   \n",
       "...                      ...     ...      ...      ...         ...   \n",
       "2018-09-30 19:00:00     True  283.45      0.0      0.0        75.0   \n",
       "2018-09-30 20:00:00     True  282.76      0.0      0.0        90.0   \n",
       "2018-09-30 21:00:00     True  282.73      0.0      0.0        90.0   \n",
       "2018-09-30 22:00:00     True  282.09      0.0      0.0        90.0   \n",
       "2018-09-30 23:00:00     True  282.12      0.0      0.0        90.0   \n",
       "\n",
       "                     weather_main     weather_description  traffic_volume  \\\n",
       "date_time                                                                   \n",
       "2012-10-02 09:00:00        Clouds        scattered clouds          5545.0   \n",
       "2012-10-02 10:00:00        Clouds           broken clouds          4516.0   \n",
       "2012-10-02 11:00:00        Clouds         overcast clouds          4767.0   \n",
       "2012-10-02 12:00:00        Clouds         overcast clouds          5026.0   \n",
       "2012-10-02 13:00:00        Clouds           broken clouds          4918.0   \n",
       "...                           ...                     ...             ...   \n",
       "2018-09-30 19:00:00        Clouds           broken clouds          3543.0   \n",
       "2018-09-30 20:00:00        Clouds         overcast clouds          2781.0   \n",
       "2018-09-30 21:00:00  Thunderstorm  proximity thunderstorm          2159.0   \n",
       "2018-09-30 22:00:00        Clouds         overcast clouds          1450.0   \n",
       "2018-09-30 23:00:00        Clouds         overcast clouds           954.0   \n",
       "\n",
       "                     month  day  week  hour  minute  dayofweek  \n",
       "date_time                                                       \n",
       "2012-10-02 09:00:00     10    2    40     9       0          1  \n",
       "2012-10-02 10:00:00     10    2    40    10       0          1  \n",
       "2012-10-02 11:00:00     10    2    40    11       0          1  \n",
       "2012-10-02 12:00:00     10    2    40    12       0          1  \n",
       "2012-10-02 13:00:00     10    2    40    13       0          1  \n",
       "...                    ...  ...   ...   ...     ...        ...  \n",
       "2018-09-30 19:00:00      9   30    39    19       0          6  \n",
       "2018-09-30 20:00:00      9   30    39    20       0          6  \n",
       "2018-09-30 21:00:00      9   30    39    21       0          6  \n",
       "2018-09-30 22:00:00      9   30    39    22       0          6  \n",
       "2018-09-30 23:00:00      9   30    39    23       0          6  \n",
       "\n",
       "[52551 rows x 14 columns]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x216 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "class MetroInterstateTraffic(RegressionForecastData):\n",
    "    \"\"\"\n",
    "    See: https://archive.ics.uci.edu/ml/datasets/Metro+Interstate+Traffic+Volume\n",
    "    \"\"\"\n",
    "    \n",
    "    columns_target = ['traffic_volume']\n",
    "    columns_forecast = ['holiday', 'month', 'day', 'week', 'hour',\n",
    "       'minute', 'dayofweek']\n",
    "    \n",
    "    def download(self):\n",
    "        url = 'https://archive.ics.uci.edu/ml/machine-learning-databases/00492/Metro_Interstate_Traffic_Volume.csv.gz'\n",
    "        \n",
    "        # download if needed\n",
    "        filename = '00492_Metro_Interstate_Traffic_Volume.csv.gz'\n",
    "        local_path = self.datasets_root/filename\n",
    "        if not local_path.exists():\n",
    "            download_url(url, self.datasets_root, filename)\n",
    "        df = (pd.read_csv(local_path, index_col='date_time', parse_dates=['date_time'])\n",
    "              .dropna(subset=self.columns_target)\n",
    "              .resample('1H').first()\n",
    "             )\n",
    "        \n",
    "        # Make holiday a bool\n",
    "        df['holiday'] = ~df['holiday'].isna()\n",
    "        df['weather_main'] = df['weather_main'].fillna('none')\n",
    "        df['weather_description'] = df['weather_description'].fillna('none')\n",
    "        \n",
    "        # Add time features \n",
    "        time = df.index.to_series()\n",
    "        df[\"month\"] = time.dt.month\n",
    "        df['day'] = time.dt.day\n",
    "        df['week'] = time.dt.isocalendar().week\n",
    "        df['hour'] = time.dt.hour\n",
    "        df['minute'] = time.dt.minute\n",
    "        df['dayofweek'] = time.dt.dayofweek\n",
    "        \n",
    "        return df\n",
    "    \n",
    "\n",
    "dataset = MetroInterstateTraffic(datasets_root)\n",
    "dataset.df[dataset.columns_target].head(1000).plot()\n",
    "print(dataset)\n",
    "dataset.df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-10-26T06:20:52.874642Z",
     "start_time": "2020-10-26T06:20:52.003940Z"
    }
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-10-26T06:25:17.829014Z",
     "start_time": "2020-10-26T06:25:12.276854Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading https://archive.ics.uci.edu/ml/machine-learning-databases/00374/energydata_complete.csv to ../data/processed/00374_AppliancesEnergyPrediction.csv\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3a5bcd8759bb45faabdc3a9061f15abd",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(HTML(value=''), FloatProgress(value=1.0, bar_style='info', layout=Layout(width='20px'), max=1.0…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<AppliancesEnergyPrediction (19735, 34)>\n"
     ]
    },
    {
     "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>lights</th>\n",
       "      <th>T1</th>\n",
       "      <th>RH_1</th>\n",
       "      <th>T2</th>\n",
       "      <th>RH_2</th>\n",
       "      <th>T3</th>\n",
       "      <th>RH_3</th>\n",
       "      <th>T4</th>\n",
       "      <th>RH_4</th>\n",
       "      <th>T5</th>\n",
       "      <th>...</th>\n",
       "      <th>Tdewpoint</th>\n",
       "      <th>rv1</th>\n",
       "      <th>rv2</th>\n",
       "      <th>log_Appliances</th>\n",
       "      <th>month</th>\n",
       "      <th>day</th>\n",
       "      <th>week</th>\n",
       "      <th>hour</th>\n",
       "      <th>minute</th>\n",
       "      <th>dayofweek</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>date</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",
       "      <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>2016-01-11 17:00:00</th>\n",
       "      <td>30</td>\n",
       "      <td>19.890000</td>\n",
       "      <td>47.596667</td>\n",
       "      <td>19.200000</td>\n",
       "      <td>44.790000</td>\n",
       "      <td>19.790000</td>\n",
       "      <td>44.730000</td>\n",
       "      <td>19.000000</td>\n",
       "      <td>45.566667</td>\n",
       "      <td>17.166667</td>\n",
       "      <td>...</td>\n",
       "      <td>5.300000</td>\n",
       "      <td>13.275433</td>\n",
       "      <td>13.275433</td>\n",
       "      <td>4.094345</td>\n",
       "      <td>1</td>\n",
       "      <td>11</td>\n",
       "      <td>2</td>\n",
       "      <td>17</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-01-11 17:10:00</th>\n",
       "      <td>30</td>\n",
       "      <td>19.890000</td>\n",
       "      <td>46.693333</td>\n",
       "      <td>19.200000</td>\n",
       "      <td>44.722500</td>\n",
       "      <td>19.790000</td>\n",
       "      <td>44.790000</td>\n",
       "      <td>19.000000</td>\n",
       "      <td>45.992500</td>\n",
       "      <td>17.166667</td>\n",
       "      <td>...</td>\n",
       "      <td>5.200000</td>\n",
       "      <td>18.606195</td>\n",
       "      <td>18.606195</td>\n",
       "      <td>4.094345</td>\n",
       "      <td>1</td>\n",
       "      <td>11</td>\n",
       "      <td>2</td>\n",
       "      <td>17</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-01-11 17:20:00</th>\n",
       "      <td>30</td>\n",
       "      <td>19.890000</td>\n",
       "      <td>46.300000</td>\n",
       "      <td>19.200000</td>\n",
       "      <td>44.626667</td>\n",
       "      <td>19.790000</td>\n",
       "      <td>44.933333</td>\n",
       "      <td>18.926667</td>\n",
       "      <td>45.890000</td>\n",
       "      <td>17.166667</td>\n",
       "      <td>...</td>\n",
       "      <td>5.100000</td>\n",
       "      <td>28.642668</td>\n",
       "      <td>28.642668</td>\n",
       "      <td>3.912023</td>\n",
       "      <td>1</td>\n",
       "      <td>11</td>\n",
       "      <td>2</td>\n",
       "      <td>17</td>\n",
       "      <td>20</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-01-11 17:30:00</th>\n",
       "      <td>40</td>\n",
       "      <td>19.890000</td>\n",
       "      <td>46.066667</td>\n",
       "      <td>19.200000</td>\n",
       "      <td>44.590000</td>\n",
       "      <td>19.790000</td>\n",
       "      <td>45.000000</td>\n",
       "      <td>18.890000</td>\n",
       "      <td>45.723333</td>\n",
       "      <td>17.166667</td>\n",
       "      <td>...</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>45.410389</td>\n",
       "      <td>45.410389</td>\n",
       "      <td>3.912023</td>\n",
       "      <td>1</td>\n",
       "      <td>11</td>\n",
       "      <td>2</td>\n",
       "      <td>17</td>\n",
       "      <td>30</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-01-11 17:40:00</th>\n",
       "      <td>40</td>\n",
       "      <td>19.890000</td>\n",
       "      <td>46.333333</td>\n",
       "      <td>19.200000</td>\n",
       "      <td>44.530000</td>\n",
       "      <td>19.790000</td>\n",
       "      <td>45.000000</td>\n",
       "      <td>18.890000</td>\n",
       "      <td>45.530000</td>\n",
       "      <td>17.200000</td>\n",
       "      <td>...</td>\n",
       "      <td>4.900000</td>\n",
       "      <td>10.084097</td>\n",
       "      <td>10.084097</td>\n",
       "      <td>4.094345</td>\n",
       "      <td>1</td>\n",
       "      <td>11</td>\n",
       "      <td>2</td>\n",
       "      <td>17</td>\n",
       "      <td>40</td>\n",
       "      <td>0</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",
       "      <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>2016-05-27 17:20:00</th>\n",
       "      <td>0</td>\n",
       "      <td>25.566667</td>\n",
       "      <td>46.560000</td>\n",
       "      <td>25.890000</td>\n",
       "      <td>42.025714</td>\n",
       "      <td>27.200000</td>\n",
       "      <td>41.163333</td>\n",
       "      <td>24.700000</td>\n",
       "      <td>45.590000</td>\n",
       "      <td>23.200000</td>\n",
       "      <td>...</td>\n",
       "      <td>13.333333</td>\n",
       "      <td>43.096812</td>\n",
       "      <td>43.096812</td>\n",
       "      <td>4.605170</td>\n",
       "      <td>5</td>\n",
       "      <td>27</td>\n",
       "      <td>21</td>\n",
       "      <td>17</td>\n",
       "      <td>20</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-05-27 17:30:00</th>\n",
       "      <td>0</td>\n",
       "      <td>25.500000</td>\n",
       "      <td>46.500000</td>\n",
       "      <td>25.754000</td>\n",
       "      <td>42.080000</td>\n",
       "      <td>27.133333</td>\n",
       "      <td>41.223333</td>\n",
       "      <td>24.700000</td>\n",
       "      <td>45.590000</td>\n",
       "      <td>23.230000</td>\n",
       "      <td>...</td>\n",
       "      <td>13.300000</td>\n",
       "      <td>49.282940</td>\n",
       "      <td>49.282940</td>\n",
       "      <td>4.499810</td>\n",
       "      <td>5</td>\n",
       "      <td>27</td>\n",
       "      <td>21</td>\n",
       "      <td>17</td>\n",
       "      <td>30</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-05-27 17:40:00</th>\n",
       "      <td>10</td>\n",
       "      <td>25.500000</td>\n",
       "      <td>46.596667</td>\n",
       "      <td>25.628571</td>\n",
       "      <td>42.768571</td>\n",
       "      <td>27.050000</td>\n",
       "      <td>41.690000</td>\n",
       "      <td>24.700000</td>\n",
       "      <td>45.730000</td>\n",
       "      <td>23.230000</td>\n",
       "      <td>...</td>\n",
       "      <td>13.266667</td>\n",
       "      <td>29.199117</td>\n",
       "      <td>29.199117</td>\n",
       "      <td>5.598422</td>\n",
       "      <td>5</td>\n",
       "      <td>27</td>\n",
       "      <td>21</td>\n",
       "      <td>17</td>\n",
       "      <td>40</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-05-27 17:50:00</th>\n",
       "      <td>10</td>\n",
       "      <td>25.500000</td>\n",
       "      <td>46.990000</td>\n",
       "      <td>25.414000</td>\n",
       "      <td>43.036000</td>\n",
       "      <td>26.890000</td>\n",
       "      <td>41.290000</td>\n",
       "      <td>24.700000</td>\n",
       "      <td>45.790000</td>\n",
       "      <td>23.200000</td>\n",
       "      <td>...</td>\n",
       "      <td>13.233333</td>\n",
       "      <td>6.322784</td>\n",
       "      <td>6.322784</td>\n",
       "      <td>6.040255</td>\n",
       "      <td>5</td>\n",
       "      <td>27</td>\n",
       "      <td>21</td>\n",
       "      <td>17</td>\n",
       "      <td>50</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-05-27 18:00:00</th>\n",
       "      <td>10</td>\n",
       "      <td>25.500000</td>\n",
       "      <td>46.600000</td>\n",
       "      <td>25.264286</td>\n",
       "      <td>42.971429</td>\n",
       "      <td>26.823333</td>\n",
       "      <td>41.156667</td>\n",
       "      <td>24.700000</td>\n",
       "      <td>45.963333</td>\n",
       "      <td>23.200000</td>\n",
       "      <td>...</td>\n",
       "      <td>13.200000</td>\n",
       "      <td>34.118851</td>\n",
       "      <td>34.118851</td>\n",
       "      <td>6.063785</td>\n",
       "      <td>5</td>\n",
       "      <td>27</td>\n",
       "      <td>21</td>\n",
       "      <td>18</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>19735 rows × 34 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                     lights         T1       RH_1         T2       RH_2  \\\n",
       "date                                                                      \n",
       "2016-01-11 17:00:00      30  19.890000  47.596667  19.200000  44.790000   \n",
       "2016-01-11 17:10:00      30  19.890000  46.693333  19.200000  44.722500   \n",
       "2016-01-11 17:20:00      30  19.890000  46.300000  19.200000  44.626667   \n",
       "2016-01-11 17:30:00      40  19.890000  46.066667  19.200000  44.590000   \n",
       "2016-01-11 17:40:00      40  19.890000  46.333333  19.200000  44.530000   \n",
       "...                     ...        ...        ...        ...        ...   \n",
       "2016-05-27 17:20:00       0  25.566667  46.560000  25.890000  42.025714   \n",
       "2016-05-27 17:30:00       0  25.500000  46.500000  25.754000  42.080000   \n",
       "2016-05-27 17:40:00      10  25.500000  46.596667  25.628571  42.768571   \n",
       "2016-05-27 17:50:00      10  25.500000  46.990000  25.414000  43.036000   \n",
       "2016-05-27 18:00:00      10  25.500000  46.600000  25.264286  42.971429   \n",
       "\n",
       "                            T3       RH_3         T4       RH_4         T5  \\\n",
       "date                                                                         \n",
       "2016-01-11 17:00:00  19.790000  44.730000  19.000000  45.566667  17.166667   \n",
       "2016-01-11 17:10:00  19.790000  44.790000  19.000000  45.992500  17.166667   \n",
       "2016-01-11 17:20:00  19.790000  44.933333  18.926667  45.890000  17.166667   \n",
       "2016-01-11 17:30:00  19.790000  45.000000  18.890000  45.723333  17.166667   \n",
       "2016-01-11 17:40:00  19.790000  45.000000  18.890000  45.530000  17.200000   \n",
       "...                        ...        ...        ...        ...        ...   \n",
       "2016-05-27 17:20:00  27.200000  41.163333  24.700000  45.590000  23.200000   \n",
       "2016-05-27 17:30:00  27.133333  41.223333  24.700000  45.590000  23.230000   \n",
       "2016-05-27 17:40:00  27.050000  41.690000  24.700000  45.730000  23.230000   \n",
       "2016-05-27 17:50:00  26.890000  41.290000  24.700000  45.790000  23.200000   \n",
       "2016-05-27 18:00:00  26.823333  41.156667  24.700000  45.963333  23.200000   \n",
       "\n",
       "                     ...  Tdewpoint        rv1        rv2  log_Appliances  \\\n",
       "date                 ...                                                    \n",
       "2016-01-11 17:00:00  ...   5.300000  13.275433  13.275433        4.094345   \n",
       "2016-01-11 17:10:00  ...   5.200000  18.606195  18.606195        4.094345   \n",
       "2016-01-11 17:20:00  ...   5.100000  28.642668  28.642668        3.912023   \n",
       "2016-01-11 17:30:00  ...   5.000000  45.410389  45.410389        3.912023   \n",
       "2016-01-11 17:40:00  ...   4.900000  10.084097  10.084097        4.094345   \n",
       "...                  ...        ...        ...        ...             ...   \n",
       "2016-05-27 17:20:00  ...  13.333333  43.096812  43.096812        4.605170   \n",
       "2016-05-27 17:30:00  ...  13.300000  49.282940  49.282940        4.499810   \n",
       "2016-05-27 17:40:00  ...  13.266667  29.199117  29.199117        5.598422   \n",
       "2016-05-27 17:50:00  ...  13.233333   6.322784   6.322784        6.040255   \n",
       "2016-05-27 18:00:00  ...  13.200000  34.118851  34.118851        6.063785   \n",
       "\n",
       "                     month  day  week  hour  minute  dayofweek  \n",
       "date                                                            \n",
       "2016-01-11 17:00:00      1   11     2    17       0          0  \n",
       "2016-01-11 17:10:00      1   11     2    17      10          0  \n",
       "2016-01-11 17:20:00      1   11     2    17      20          0  \n",
       "2016-01-11 17:30:00      1   11     2    17      30          0  \n",
       "2016-01-11 17:40:00      1   11     2    17      40          0  \n",
       "...                    ...  ...   ...   ...     ...        ...  \n",
       "2016-05-27 17:20:00      5   27    21    17      20          4  \n",
       "2016-05-27 17:30:00      5   27    21    17      30          4  \n",
       "2016-05-27 17:40:00      5   27    21    17      40          4  \n",
       "2016-05-27 17:50:00      5   27    21    17      50          4  \n",
       "2016-05-27 18:00:00      5   27    21    18       0          4  \n",
       "\n",
       "[19735 rows x 34 columns]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x216 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "class AppliancesEnergyPrediction(RegressionForecastData):\n",
    "    \"\"\"\n",
    "    See: https://archive.ics.uci.edu/ml/datasets/Appliances+energy+prediction\n",
    "    \"\"\"\n",
    "    \n",
    "    columns_target = ['log_Appliances']\n",
    "    columns_forecast = ['month', 'day', 'week', 'hour',\n",
    "       'minute', 'dayofweek']\n",
    "    \n",
    "    def download(self):\n",
    "        url = 'https://archive.ics.uci.edu/ml/machine-learning-databases/00374/energydata_complete.csv'\n",
    "        \n",
    "        # download if needed\n",
    "        filename = '00374_AppliancesEnergyPrediction.csv'\n",
    "        local_path = self.datasets_root/filename\n",
    "        if not local_path.exists():\n",
    "            download_url(url, self.datasets_root, filename)\n",
    "        df = pd.read_csv(local_path, index_col='date', parse_dates=['date'])\n",
    "        \n",
    "        # log target\n",
    "        df['log_Appliances'] = np.log(df['Appliances'] + 1e-5)\n",
    "        df = df.drop(columns=['Appliances'])\n",
    "        df = df.dropna(subset=self.columns_target).resample('10T').first()\n",
    "        \n",
    "        # Add time features \n",
    "        time = df.index.to_series()\n",
    "        df[\"month\"] = time.dt.month\n",
    "        df['day'] = time.dt.day\n",
    "        df['week'] = time.dt.isocalendar().week\n",
    "        df['hour'] = time.dt.hour\n",
    "        df['minute'] = time.dt.minute\n",
    "        df['dayofweek'] = time.dt.dayofweek\n",
    "        \n",
    "        return df\n",
    "dataset = AppliancesEnergyPrediction(datasets_root)\n",
    "print(dataset)\n",
    "dataset.df[dataset.columns_target].head(1000).plot()\n",
    "dataset.df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-10-26T06:19:53.791396Z",
     "start_time": "2020-10-26T06:19:53.476265Z"
    }
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-10-26T06:25:20.668980Z",
     "start_time": "2020-10-26T06:25:17.831622Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading http://archive.ics.uci.edu/ml/machine-learning-databases/00381/PRSA_data_2010.1.1-2014.12.31.csv to ../data/processed/00381_BejingPM25.csv\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d479f39fd7714412a2ca8189861a74f6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(HTML(value=''), FloatProgress(value=1.0, bar_style='info', layout=Layout(width='20px'), max=1.0…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<BejingPM25 (43800, 14)>\n"
     ]
    },
    {
     "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>DEWP</th>\n",
       "      <th>TEMP</th>\n",
       "      <th>PRES</th>\n",
       "      <th>cbwd</th>\n",
       "      <th>Iws</th>\n",
       "      <th>Is</th>\n",
       "      <th>Ir</th>\n",
       "      <th>log_pm2.5</th>\n",
       "      <th>month</th>\n",
       "      <th>day</th>\n",
       "      <th>week</th>\n",
       "      <th>hour</th>\n",
       "      <th>minute</th>\n",
       "      <th>dayofweek</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2010-01-02 00:00:00+08:00</th>\n",
       "      <td>-16.0</td>\n",
       "      <td>-4.0</td>\n",
       "      <td>1020.0</td>\n",
       "      <td>SE</td>\n",
       "      <td>1.79</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4.859812</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>53</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-01-02 01:00:00+08:00</th>\n",
       "      <td>-15.0</td>\n",
       "      <td>-4.0</td>\n",
       "      <td>1020.0</td>\n",
       "      <td>SE</td>\n",
       "      <td>2.68</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4.997212</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>53</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-01-02 02:00:00+08:00</th>\n",
       "      <td>-11.0</td>\n",
       "      <td>-5.0</td>\n",
       "      <td>1021.0</td>\n",
       "      <td>SE</td>\n",
       "      <td>3.57</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>5.068904</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>53</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-01-02 03:00:00+08:00</th>\n",
       "      <td>-7.0</td>\n",
       "      <td>-5.0</td>\n",
       "      <td>1022.0</td>\n",
       "      <td>SE</td>\n",
       "      <td>5.36</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>5.198497</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>53</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-01-02 04:00:00+08:00</th>\n",
       "      <td>-7.0</td>\n",
       "      <td>-5.0</td>\n",
       "      <td>1022.0</td>\n",
       "      <td>SE</td>\n",
       "      <td>6.25</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4.927254</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>53</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>5</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>2014-12-31 19:00:00+08:00</th>\n",
       "      <td>-23.0</td>\n",
       "      <td>-2.0</td>\n",
       "      <td>1034.0</td>\n",
       "      <td>NW</td>\n",
       "      <td>231.97</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.079443</td>\n",
       "      <td>12</td>\n",
       "      <td>31</td>\n",
       "      <td>1</td>\n",
       "      <td>19</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2014-12-31 20:00:00+08:00</th>\n",
       "      <td>-22.0</td>\n",
       "      <td>-3.0</td>\n",
       "      <td>1034.0</td>\n",
       "      <td>NW</td>\n",
       "      <td>237.78</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.302586</td>\n",
       "      <td>12</td>\n",
       "      <td>31</td>\n",
       "      <td>1</td>\n",
       "      <td>20</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2014-12-31 21:00:00+08:00</th>\n",
       "      <td>-22.0</td>\n",
       "      <td>-3.0</td>\n",
       "      <td>1034.0</td>\n",
       "      <td>NW</td>\n",
       "      <td>242.70</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.302586</td>\n",
       "      <td>12</td>\n",
       "      <td>31</td>\n",
       "      <td>1</td>\n",
       "      <td>21</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2014-12-31 22:00:00+08:00</th>\n",
       "      <td>-22.0</td>\n",
       "      <td>-4.0</td>\n",
       "      <td>1034.0</td>\n",
       "      <td>NW</td>\n",
       "      <td>246.72</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.079443</td>\n",
       "      <td>12</td>\n",
       "      <td>31</td>\n",
       "      <td>1</td>\n",
       "      <td>22</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2014-12-31 23:00:00+08:00</th>\n",
       "      <td>-21.0</td>\n",
       "      <td>-3.0</td>\n",
       "      <td>1034.0</td>\n",
       "      <td>NW</td>\n",
       "      <td>249.85</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.484907</td>\n",
       "      <td>12</td>\n",
       "      <td>31</td>\n",
       "      <td>1</td>\n",
       "      <td>23</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>43800 rows × 14 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                           DEWP  TEMP    PRES cbwd     Iws   Is   Ir  \\\n",
       "2010-01-02 00:00:00+08:00 -16.0  -4.0  1020.0   SE    1.79  0.0  0.0   \n",
       "2010-01-02 01:00:00+08:00 -15.0  -4.0  1020.0   SE    2.68  0.0  0.0   \n",
       "2010-01-02 02:00:00+08:00 -11.0  -5.0  1021.0   SE    3.57  0.0  0.0   \n",
       "2010-01-02 03:00:00+08:00  -7.0  -5.0  1022.0   SE    5.36  1.0  0.0   \n",
       "2010-01-02 04:00:00+08:00  -7.0  -5.0  1022.0   SE    6.25  2.0  0.0   \n",
       "...                         ...   ...     ...  ...     ...  ...  ...   \n",
       "2014-12-31 19:00:00+08:00 -23.0  -2.0  1034.0   NW  231.97  0.0  0.0   \n",
       "2014-12-31 20:00:00+08:00 -22.0  -3.0  1034.0   NW  237.78  0.0  0.0   \n",
       "2014-12-31 21:00:00+08:00 -22.0  -3.0  1034.0   NW  242.70  0.0  0.0   \n",
       "2014-12-31 22:00:00+08:00 -22.0  -4.0  1034.0   NW  246.72  0.0  0.0   \n",
       "2014-12-31 23:00:00+08:00 -21.0  -3.0  1034.0   NW  249.85  0.0  0.0   \n",
       "\n",
       "                           log_pm2.5  month  day  week  hour  minute  \\\n",
       "2010-01-02 00:00:00+08:00   4.859812      1    2    53     0       0   \n",
       "2010-01-02 01:00:00+08:00   4.997212      1    2    53     1       0   \n",
       "2010-01-02 02:00:00+08:00   5.068904      1    2    53     2       0   \n",
       "2010-01-02 03:00:00+08:00   5.198497      1    2    53     3       0   \n",
       "2010-01-02 04:00:00+08:00   4.927254      1    2    53     4       0   \n",
       "...                              ...    ...  ...   ...   ...     ...   \n",
       "2014-12-31 19:00:00+08:00   2.079443     12   31     1    19       0   \n",
       "2014-12-31 20:00:00+08:00   2.302586     12   31     1    20       0   \n",
       "2014-12-31 21:00:00+08:00   2.302586     12   31     1    21       0   \n",
       "2014-12-31 22:00:00+08:00   2.079443     12   31     1    22       0   \n",
       "2014-12-31 23:00:00+08:00   2.484907     12   31     1    23       0   \n",
       "\n",
       "                           dayofweek  \n",
       "2010-01-02 00:00:00+08:00          5  \n",
       "2010-01-02 01:00:00+08:00          5  \n",
       "2010-01-02 02:00:00+08:00          5  \n",
       "2010-01-02 03:00:00+08:00          5  \n",
       "2010-01-02 04:00:00+08:00          5  \n",
       "...                              ...  \n",
       "2014-12-31 19:00:00+08:00          2  \n",
       "2014-12-31 20:00:00+08:00          2  \n",
       "2014-12-31 21:00:00+08:00          2  \n",
       "2014-12-31 22:00:00+08:00          2  \n",
       "2014-12-31 23:00:00+08:00          2  \n",
       "\n",
       "[43800 rows x 14 columns]"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x216 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "class BejingPM25(RegressionForecastData):\n",
    "    \"\"\"\n",
    "    See: http://archive.ics.uci.edu/ml/datasets/Beijing+PM2.5+Data\n",
    "    \"\"\"\n",
    "    \n",
    "    columns_target = ['log_pm2.5']\n",
    "    columns_forecast = ['month', 'day', 'week', 'hour',\n",
    "       'minute', 'dayofweek']\n",
    "    \n",
    "    def download(self):\n",
    "        url = 'http://archive.ics.uci.edu/ml/machine-learning-databases/00381/PRSA_data_2010.1.1-2014.12.31.csv'\n",
    "        \n",
    "        # download if needed\n",
    "        filename = '00381_BejingPM25.csv'\n",
    "        local_path = self.datasets_root/filename\n",
    "        if not local_path.exists():\n",
    "            download_url(url, self.datasets_root, filename)\n",
    "        df = pd.read_csv(local_path)\n",
    "        df.index = pd.to_datetime(df[['year', 'month', 'day', 'hour']]).dt.tz_localize('Asia/Shanghai')\n",
    "        df = df.drop(columns=['year', 'month', 'day', 'hour', 'No'])\n",
    "        \n",
    "        # log target\n",
    "        df['log_pm2.5'] = np.log(df['pm2.5'] + 1e-5)\n",
    "        df = df.drop(columns=['pm2.5'])\n",
    "        \n",
    "        df.dropna(subset=self.columns_target, inplace=True)\n",
    "        df = df.resample('1H').first()\n",
    "        \n",
    "        df['cbwd'] = df['cbwd'].fillna('none')\n",
    "        \n",
    "        \n",
    "        \n",
    "        # Add time features \n",
    "        time = df.index.to_series()\n",
    "        df[\"month\"] = time.dt.month\n",
    "        df['day'] = time.dt.day\n",
    "        df['week'] = time.dt.isocalendar().week\n",
    "        df['hour'] = time.dt.hour\n",
    "        df['minute'] = time.dt.minute\n",
    "        df['dayofweek'] = time.dt.dayofweek\n",
    "        \n",
    "#         df['log_pm2.5'] = np.log(df['pm2.5']+1e-5)\n",
    "        \n",
    "        return df\n",
    "dataset = BejingPM25(datasets_root)\n",
    "print(dataset)\n",
    "dataset.df[dataset.columns_target].head(1000).plot()\n",
    "dataset.df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-10-26T06:06:15.058353Z",
     "start_time": "2020-10-26T06:06:14.990663Z"
    }
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-10-26T06:25:23.345365Z",
     "start_time": "2020-10-26T06:25:20.671318Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Using downloaded and verified file: ../data/raw/IMOS_ANMN/IMOS_ANMN-WA_AETVZ_20090715T080000Z_WATR20_FV01_WATR20-0907-Continental-194_END-20090716T181317Z_C-20191122T052830Z.nc\n",
      "Using downloaded and verified file: ../data/raw/IMOS_ANMN/IMOS_ANMN-WA_AETVZ_20100409T080000Z_WATR20_FV01_WATR20-1004-Continental-194_END-20100430T084500Z_C-20191122T053845Z.nc\n",
      "Using downloaded and verified file: ../data/raw/IMOS_ANMN/IMOS_ANMN-WA_AETVZ_20101222T080000Z_WATR20_FV01_WATR20-1012-Continental-194_END-20110518T051500Z_C-20200916T020035Z.nc\n",
      "Using downloaded and verified file: ../data/raw/IMOS_ANMN/IMOS_ANMN-WA_AETVZ_20110608T080000Z_WATR20_FV01_WATR20-1106-Continental-194_END-20111122T035000Z_C-20200916T025619Z.nc\n",
      "Using downloaded and verified file: ../data/raw/IMOS_ANMN/IMOS_ANMN-WA_AETVZ_20111221T060300Z_WATR20_FV01_WATR20-1112-Continental-194_END-20120704T050500Z_C-20200916T043212Z.nc\n",
      "Using downloaded and verified file: ../data/raw/IMOS_ANMN/IMOS_ANMN-WA_AETVZ_20120726T044000Z_WATR20_FV01_WATR20-1207-Continental-194_END-20130204T044000Z_C-20200916T032027Z.nc\n",
      "Using downloaded and verified file: ../data/raw/IMOS_ANMN/IMOS_ANMN-WA_AETVZ_20130221T080000Z_WATR20_FV01_WATR20-1302-Continental-194_END-20131003T035000Z_C-20180529T020609Z.nc\n",
      "Using downloaded and verified file: ../data/raw/IMOS_ANMN/IMOS_ANMN-WA_AETVZ_20131111T080000Z_WATR20_FV01_WATR20-1311-Continental-194_END-20140519T035000Z_C-20200114T033335Z.nc\n",
      "Using downloaded and verified file: ../data/raw/IMOS_ANMN/IMOS_ANMN-WA_AETVZ_20140710T080000Z_WATR20_FV01_WATR20-1407-Continental-194_END-20150121T021500Z_C-20180529T055902Z.nc\n",
      "Using downloaded and verified file: ../data/raw/IMOS_ANMN/IMOS_ANMN-WA_AETVZ_20150213T080000Z_WATR20_FV01_WATR20-1502-Continental-194_END-20150424T134002Z_C-20200114T035347Z.nc\n",
      "Using downloaded and verified file: ../data/raw/IMOS_ANMN/IMOS_ANMN-WA_AETVZ_20150914T080000Z_WATR20_FV01_WATR20-1509-Continental-194_END-20160331T043000Z_C-20180601T013623Z.nc\n",
      "Using downloaded and verified file: ../data/raw/IMOS_ANMN/IMOS_ANMN-WA_AETVZ_20160427T080000Z_WATR20_FV01_WATR20-1604-Continental-194_END-20160531T021800Z_C-20180531T071709Z.nc\n",
      "Using downloaded and verified file: ../data/raw/IMOS_ANMN/IMOS_ANMN-WA_AETVZ_20171204T080000Z_WATR20_FV01_WATR20-1712-Continental-194_END-20180618T030000Z_C-20180620T233149Z.nc\n",
      "Using downloaded and verified file: ../data/raw/IMOS_ANMN/IMOS_ANMN-WA_AETVZ_20180802T080000Z_WATR20_FV01_WATR20-1807-Continental-194_END-20190225T054500Z_C-20190227T001343Z.nc\n",
      "Using downloaded and verified file: ../data/raw/IMOS_ANMN/IMOS_ANMN-WA_AETVZ_20190307T080000Z_WATR20_FV01_WATR20-1903-Continental-194_END-20190911T003144Z_C-20200114T045053Z.nc\n",
      "Using downloaded and verified file: ../data/raw/IMOS_ANMN/IMOS_ANMN-WA_AETVZ_20190926T080000Z_WATR20_FV01_WATR20-1909-Continental-194_END-20200326T030000Z_C-20200420T064334Z.nc\n",
      "wrote \"../data/processed/MOS_ANMN-WA_AETVZ_WATR20_FV01_WATR20-1909-Continental-194_currents.nc\" with size 43.11 MB\n"
     ]
    },
    {
     "ename": "NameError",
     "evalue": "name 'xarray' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-12-53e5e67305d6>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m    140\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    141\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 142\u001b[0;31m \u001b[0mdataset\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mIMOSCurrentsVel\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdatasets_root\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    143\u001b[0m \u001b[0mdataset\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdf\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mdataset\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcolumns_target\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdropna\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mhead\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m1000\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mplot\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    144\u001b[0m \u001b[0mdataset\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdf\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m<ipython-input-7-9fbff7440955>\u001b[0m in \u001b[0;36m__init__\u001b[0;34m(self, datasets_root)\u001b[0m\n\u001b[1;32m      7\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      8\u001b[0m         \u001b[0;31m# Process data\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 9\u001b[0;31m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdf\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdownload\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     10\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdf_norm\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mscaler\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnormalize\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdf\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     11\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moutput_scaler\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnext\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfilter\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;32mlambda\u001b[0m \u001b[0mr\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0mr\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcolumns_target\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mscaler\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfeatures\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m<ipython-input-12-53e5e67305d6>\u001b[0m in \u001b[0;36mdownload\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    130\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    131\u001b[0m         \u001b[0;31m# made in previous notebook\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 132\u001b[0;31m         \u001b[0mxd\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mxarray\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mload_dataset\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0moutfile\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    133\u001b[0m         df = xd.to_dataframe().drop(\n\u001b[1;32m    134\u001b[0m             columns=['HEIGHT_ABOVE_SENSOR', 'NOMINAL_DEPTH'])\n",
      "\u001b[0;31mNameError\u001b[0m: name 'xarray' is not defined"
     ]
    }
   ],
   "source": [
    "import uptide\n",
    "import xarray as xr\n",
    "\n",
    "# https://en.wikipedia.org/wiki/Theory_of_tides#Harmonic_analysis\n",
    "default_tidal_constituents = [\n",
    "    'M2',\n",
    "    'S2',\n",
    "    'N2',\n",
    "    'K2',  # Semi-diurnal\n",
    "    'K1',\n",
    "    'O1',\n",
    "    'P1',\n",
    "    'Q1',  # Diurnal\n",
    "    'M4',\n",
    "    'M6',\n",
    "    'S4',\n",
    "    'MK3',  # Short period\n",
    "    'MM',\n",
    "    'SSA',\n",
    "    'SA'  # Long period\n",
    "]\n",
    "\n",
    "\n",
    "def generate_tidal_periods(t: pd.Series,\n",
    "                           constituents: list = default_tidal_constituents):\n",
    "    tide = uptide.Tides(constituents)\n",
    "    t0 = t[0]\n",
    "    td = t - t0\n",
    "    td = td.dt.total_seconds().to_numpy().astype(int)\n",
    "    tide.set_initial_time(t0)\n",
    "\n",
    "    # calc tides\n",
    "    amplitudes = np.ones_like(td)\n",
    "    phases = np.zeros_like(td)\n",
    "    eta = {}\n",
    "    for name, f, amplitude, omega, phase, phi, u in zip(\n",
    "            tide.constituents, tide.f, amplitudes, tide.omega, phases,\n",
    "            tide.phi, tide.u):\n",
    "        eta[name] = f * amplitude * np.cos(omega * td - phase + phi + u)\n",
    "    df_eta = pd.DataFrame(eta, index=t)\n",
    "    return df_eta\n",
    "\n",
    "\n",
    "# 'ANMN Two Rocks, WA, 204m mooring, Jul2009 - Dec2009. Preprocessed with DepthPP.'\n",
    "\n",
    "\n",
    "def get_current_timeseries(\n",
    "        cache_folder=Path(\"../data/raw/IMOS_ANMN/\"),\n",
    "        outfile=Path(\n",
    "            '../data/processed/currents/MOS_ANMN-WA_AETVZ_WATR20_FV01_WATR20-1909-Continental-194_currents.nc'\n",
    "        )):\n",
    "    \"\"\"\n",
    "    Download Current data from the IMOS and pre-process.\n",
    "    \"\"\"\n",
    "    if not outfile.exists():\n",
    "\n",
    "        files = [\n",
    "            \"IMOS_ANMN-WA_AETVZ_20090715T080000Z_WATR20_FV01_WATR20-0907-Continental-194_END-20090716T181317Z_C-20191122T052830Z.nc\",\n",
    "            \"IMOS_ANMN-WA_AETVZ_20100409T080000Z_WATR20_FV01_WATR20-1004-Continental-194_END-20100430T084500Z_C-20191122T053845Z.nc\",\n",
    "            \"IMOS_ANMN-WA_AETVZ_20101222T080000Z_WATR20_FV01_WATR20-1012-Continental-194_END-20110518T051500Z_C-20200916T020035Z.nc\",\n",
    "            \"IMOS_ANMN-WA_AETVZ_20110608T080000Z_WATR20_FV01_WATR20-1106-Continental-194_END-20111122T035000Z_C-20200916T025619Z.nc\",\n",
    "            \"IMOS_ANMN-WA_AETVZ_20111221T060300Z_WATR20_FV01_WATR20-1112-Continental-194_END-20120704T050500Z_C-20200916T043212Z.nc\",\n",
    "            \"IMOS_ANMN-WA_AETVZ_20120726T044000Z_WATR20_FV01_WATR20-1207-Continental-194_END-20130204T044000Z_C-20200916T032027Z.nc\",\n",
    "            \"IMOS_ANMN-WA_AETVZ_20130221T080000Z_WATR20_FV01_WATR20-1302-Continental-194_END-20131003T035000Z_C-20180529T020609Z.nc\",\n",
    "            \"IMOS_ANMN-WA_AETVZ_20131111T080000Z_WATR20_FV01_WATR20-1311-Continental-194_END-20140519T035000Z_C-20200114T033335Z.nc\",\n",
    "            \"IMOS_ANMN-WA_AETVZ_20140710T080000Z_WATR20_FV01_WATR20-1407-Continental-194_END-20150121T021500Z_C-20180529T055902Z.nc\",\n",
    "            \"IMOS_ANMN-WA_AETVZ_20150213T080000Z_WATR20_FV01_WATR20-1502-Continental-194_END-20150424T134002Z_C-20200114T035347Z.nc\",\n",
    "            \"IMOS_ANMN-WA_AETVZ_20150914T080000Z_WATR20_FV01_WATR20-1509-Continental-194_END-20160331T043000Z_C-20180601T013623Z.nc\",\n",
    "            \"IMOS_ANMN-WA_AETVZ_20160427T080000Z_WATR20_FV01_WATR20-1604-Continental-194_END-20160531T021800Z_C-20180531T071709Z.nc\",\n",
    "            #     \"IMOS_ANMN-WA_AETVZ_20170512T080000Z_WATR20_FV01_WATR20-1705-Continental-194_END-20170717T014558Z_C-20190805T004647Z.nc\",\n",
    "            \"IMOS_ANMN-WA_AETVZ_20171204T080000Z_WATR20_FV01_WATR20-1712-Continental-194_END-20180618T030000Z_C-20180620T233149Z.nc\",\n",
    "            \"IMOS_ANMN-WA_AETVZ_20180802T080000Z_WATR20_FV01_WATR20-1807-Continental-194_END-20190225T054500Z_C-20190227T001343Z.nc\",\n",
    "            \"IMOS_ANMN-WA_AETVZ_20190307T080000Z_WATR20_FV01_WATR20-1903-Continental-194_END-20190911T003144Z_C-20200114T045053Z.nc\",\n",
    "            \"IMOS_ANMN-WA_AETVZ_20190926T080000Z_WATR20_FV01_WATR20-1909-Continental-194_END-20200326T030000Z_C-20200420T064334Z.nc\",\n",
    "        ]\n",
    "        base = \"http://thredds.aodn.org.au/thredds/fileServer/IMOS/ANMN/WA/WATR20/Velocity/\"\n",
    "\n",
    "        # Download files\n",
    "        [download_url(base + f, cache_folder) for f in files]\n",
    "\n",
    "        # load and merge\n",
    "        xds = [xr.open_dataset(cache_folder / f) for f in files]\n",
    "        vars = [\n",
    "            'VCUR', 'UCUR', 'WCUR', 'TEMP', 'PRES_REL', 'DEPTH', 'ROLL',\n",
    "            'PITCH'\n",
    "        ]\n",
    "        xds2 = [x[vars].isel(HEIGHT_ABOVE_SENSOR=18) for x in xds]\n",
    "        xd = xr.concat(xds2, dim='TIME')\n",
    "        xd = xd.where(xd.DEPTH > 150)  # remove outliers\n",
    "\n",
    "        xd['TIME'] = xd['TIME'].dt.round('10T')\n",
    "        xd = xd.dropna(dim='TIME', subset=['VCUR', 'UCUR', 'WCUR'])\n",
    "\n",
    "        # Generate tidal freqs\n",
    "        t = xd.TIME.to_series()\n",
    "        df_eta = generate_tidal_periods(t)\n",
    "\n",
    "        # Add tidal freqs\n",
    "        xd = xd.merge(df_eta)\n",
    "\n",
    "        # Cache to nc\n",
    "        xd.to_netcdf(outfile)\n",
    "        print(\n",
    "            f'wrote \"{outfile}\" with size {outfile.stat().st_size*1e-6:2.2f} MB'\n",
    "        )\n",
    "    return outfile\n",
    "\n",
    "\n",
    "class IMOSCurrentsVel(RegressionForecastData):\n",
    "    \"\"\"\n",
    "    \n",
    "    Current Speed at ANMN Two Rocks, WA, 204m mooring\n",
    "    \n",
    "    see:\n",
    "    - http://thredds.aodn.org.au/thredds/fileServer/IMOS/ANMN/WA/WATR20/Velocity/\n",
    "    from https://catalogue-imos.aodn.org.au/geonetwork/srv/api/records/ae86e2f5-eaaf-459e-a405-e654d85adb9c\n",
    "    and http://thredds.aodn.org.au/thredds/catalog/IMOS/ANMN/WA/WATR20/Velocity/catalog.html\n",
    "    And https://en.wikipedia.org/wiki/Theory_of_tides\n",
    "    \"\"\"\n",
    "\n",
    "    columns_target = ['SPD']\n",
    "    columns_forecast = [\n",
    "        'M2', 'S2', 'N2', 'K2', 'K1', 'O1', 'P1', 'Q1', 'M4', 'M6', 'S4',\n",
    "        'MK3', 'MM', 'SSA', 'SA'\n",
    "    ]\n",
    "\n",
    "    def download(self):\n",
    "        outfile = self.datasets_root / 'MOS_ANMN-WA_AETVZ_WATR20_FV01_WATR20-1909-Continental-194_currents.nc'\n",
    "        get_current_timeseries(outfile=outfile)\n",
    "\n",
    "        # made in previous notebook\n",
    "        xd = xarray.load_dataset(outfile)\n",
    "        df = xd.to_dataframe().drop(\n",
    "            columns=['HEIGHT_ABOVE_SENSOR', 'NOMINAL_DEPTH'])\n",
    "        df['SPD'] = np.sqrt(df.VCUR**2 + df.UCUR**2)\n",
    "        df.dropna(subset=self.columns_target, inplace=True)\n",
    "        df = df.resample('30T').first()\n",
    "\n",
    "        return df\n",
    "\n",
    "\n",
    "dataset = IMOSCurrentsVel(datasets_root)\n",
    "dataset.df[dataset.columns_target].dropna().head(1000).plot()\n",
    "dataset.df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "seq2seq-time",
   "language": "python",
   "name": "seq2seq-time"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.8"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {
    "height": "755px",
    "left": "10px",
    "top": "146px",
    "width": "165px"
   },
   "toc_section_display": true,
   "toc_window_display": true
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}