{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-19T10:34:45.079040Z",
     "start_time": "2019-12-19T10:34:45.075853Z"
    }
   },
   "source": [
    "# Giotto-Time\n",
    "\n",
    "Welcome to `giotto-time`, our new library for time series forecasting!\n",
    "\n",
    "Let's start with an example."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-19T10:37:13.829605Z",
     "start_time": "2019-12-19T10:37:13.827033Z"
    }
   },
   "source": [
    "## First example"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-19T10:51:37.701263Z",
     "start_time": "2019-12-19T10:51:37.698686Z"
    }
   },
   "source": [
    "### Ingredients"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-19T10:43:03.249232Z",
     "start_time": "2019-12-19T10:43:03.244743Z"
    }
   },
   "source": [
    "These are the main ingredients of `giotto-time`:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-01-24T13:52:35.655125Z",
     "start_time": "2020-01-24T13:52:34.838920Z"
    }
   },
   "outputs": [],
   "source": [
    "%load_ext autoreload\n",
    "%autoreload 2\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from gtime.preprocessing import TimeSeriesPreparation\n",
    "from gtime.compose import FeatureCreation\n",
    "from gtime.feature_extraction import Shift, MovingAverage\n",
    "from gtime.feature_generation import PeriodicSeasonal, Constant, Calendar\n",
    "from gtime.model_selection import horizon_shift, FeatureSplitter\n",
    "from gtime.forecasting import GAR"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-01-23T10:33:07.963000Z",
     "start_time": "2020-01-23T10:33:07.896247Z"
    }
   },
   "source": [
    "- `TimeSeriesPreparation`: checks the input format of the time series and converts it to the expected format\n",
    "- `DataFrameTransformer`: scikit-learn's `ColumnTransformer` wrapper that returns DataFrame\n",
    "- `Shift`, `MovingAverage`: create the desired features on the time series for the forecasting\n",
    "- `FeatureSplitter`: prepares the custom `giotto-time` train-test matrices that are used in the model\n",
    "- `GAR`: Generalized Auto Regressive model, scikit-learn's `MultiOutputRegressor` wrapper. This is the only time series forecasting model available for the first release"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We also need a scikit-learn regression model. We go for a standard `LinearRegression` for this example."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-01-24T13:52:35.695082Z",
     "start_time": "2020-01-24T13:52:35.657030Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.linear_model import LinearRegression"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-19T10:51:15.298065Z",
     "start_time": "2019-12-19T10:51:15.295733Z"
    }
   },
   "source": [
    "### Data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-19T10:55:55.362286Z",
     "start_time": "2019-12-19T10:55:55.358045Z"
    }
   },
   "source": [
    "We use the `pandas.testing` module to create a testing time series"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-01-24T13:52:35.717578Z",
     "start_time": "2020-01-24T13:52:35.696874Z"
    }
   },
   "outputs": [],
   "source": [
    "def test_time_series():\n",
    "    from pandas.util import testing as testing\n",
    "\n",
    "    testing.N, testing.K = 500, 1\n",
    "    df = testing.makeTimeDataFrame( freq=\"D\" )\n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-01-24T13:52:35.741934Z",
     "start_time": "2020-01-24T13:52:35.719414Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Time series shape: (500, 1)\n",
      "Time series index type: <class 'pandas.core.indexes.datetimes.DatetimeIndex'>\n"
     ]
    }
   ],
   "source": [
    "time_series = test_time_series()\n",
    "print(f'Time series shape: {time_series.shape}')\n",
    "print(f'Time series index type: {time_series.index.__class__}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-19T13:37:09.941132Z",
     "start_time": "2019-12-19T13:37:09.938476Z"
    }
   },
   "source": [
    "### Time Series Preparation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-19T13:37:33.357619Z",
     "start_time": "2019-12-19T13:37:33.347192Z"
    }
   },
   "source": [
    "The input time series has to be a `DataFrame` with a `PeriodIndex`. Use the provided class `TimeSeriesPreparation` to convert the time series into this format."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-01-24T13:52:35.765147Z",
     "start_time": "2020-01-24T13:52:35.743900Z"
    }
   },
   "outputs": [],
   "source": [
    "time_series_preparation = TimeSeriesPreparation()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-01-24T13:52:35.798419Z",
     "start_time": "2020-01-24T13:52:35.767856Z"
    }
   },
   "outputs": [],
   "source": [
    "period_index_time_series = time_series_preparation.transform(time_series)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-01-24T13:52:35.824670Z",
     "start_time": "2020-01-24T13:52:35.801249Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Time series index type after the preprocessing: <class 'pandas.core.indexes.period.PeriodIndex'>\n"
     ]
    }
   ],
   "source": [
    "print(f'Time series index type after the preprocessing: {period_index_time_series.index.__class__}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-01-24T13:52:36.218080Z",
     "start_time": "2020-01-24T13:52:35.827402Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "period_index_time_series.plot(figsize=(20, 5))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-19T14:15:31.332440Z",
     "start_time": "2019-12-19T14:15:31.322583Z"
    }
   },
   "source": [
    "### Feature extraction"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-19T14:20:25.312078Z",
     "start_time": "2019-12-19T14:20:25.307741Z"
    }
   },
   "source": [
    "The feature extraction part is aimed at providing a scikit-learn paradigm with a time-series forecasting perspective\n",
    "Our `DataFrameTransformer` inherits from scikit-learn's `ColumnTransformer`, it will create a feature DataFrame with the provided Transformers.\n",
    "\n",
    "For simplicity we will create only `Shift` and `MovingAverage` features. \n",
    "\n",
    "`Shift` provides a temporal shift of the time series. Adding two `Shift` features (by 1 and 2) is equivalent to an `AR(2)` model. \n",
    "\n",
    "Since the `DataFrameTransformer` is a `ColumnTransformer` wrapper, you can easily include features from `scikit-learn`, `tsfresh`, topological features from `giotto-tda` (\\o/) or your own custom features."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-01-24T13:52:36.240916Z",
     "start_time": "2020-01-24T13:52:36.219519Z"
    }
   },
   "outputs": [],
   "source": [
    "cal = Calendar(\n",
    "    start_date=\"ignored\",\n",
    "    end_date=\"ignored\",\n",
    "    region=\"america\",\n",
    "    country=\"Brazil\",\n",
    "    kernel=np.array([0, 1]),\n",
    ")\n",
    "# New API \n",
    "dft = FeatureCreation(\n",
    "    [('s1', Shift(1), ['time_series']), \n",
    "     ('s2', Shift(2), ['time_series']),\n",
    "     ('ma3', MovingAverage(window_size=3), ['time_series']),\n",
    "     # ('cal', cal, ['time_series']),\n",
    "     # ('ct', Constant(2), ['time_series']),\n",
    "    ])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-01-24T13:52:36.281807Z",
     "start_time": "2020-01-24T13:52:36.242706Z"
    }
   },
   "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>s1__time_series__Shift</th>\n",
       "      <th>s2__time_series__Shift</th>\n",
       "      <th>ma3__time_series__MovingAverage</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2000-01-01</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2000-01-02</th>\n",
       "      <td>-1.366772</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2000-01-03</th>\n",
       "      <td>-0.439480</td>\n",
       "      <td>-1.366772</td>\n",
       "      <td>-0.410613</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2000-01-04</th>\n",
       "      <td>0.574412</td>\n",
       "      <td>-0.439480</td>\n",
       "      <td>-0.171874</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2000-01-05</th>\n",
       "      <td>-0.650554</td>\n",
       "      <td>0.574412</td>\n",
       "      <td>0.097240</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2000-01-06</th>\n",
       "      <td>0.367861</td>\n",
       "      <td>-0.650554</td>\n",
       "      <td>0.634610</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            s1__time_series__Shift  s2__time_series__Shift  \\\n",
       "2000-01-01                     NaN                     NaN   \n",
       "2000-01-02               -1.366772                     NaN   \n",
       "2000-01-03               -0.439480               -1.366772   \n",
       "2000-01-04                0.574412               -0.439480   \n",
       "2000-01-05               -0.650554                0.574412   \n",
       "2000-01-06                0.367861               -0.650554   \n",
       "\n",
       "            ma3__time_series__MovingAverage  \n",
       "2000-01-01                              NaN  \n",
       "2000-01-02                              NaN  \n",
       "2000-01-03                        -0.410613  \n",
       "2000-01-04                        -0.171874  \n",
       "2000-01-05                         0.097240  \n",
       "2000-01-06                         0.634610  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = dft.fit_transform(period_index_time_series)\n",
    "X.head(6)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-01-24T13:52:36.322677Z",
     "start_time": "2020-01-24T13:52:36.283752Z"
    }
   },
   "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>y_1</th>\n",
       "      <th>y_2</th>\n",
       "      <th>y_3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2000-01-01</th>\n",
       "      <td>-0.439480</td>\n",
       "      <td>0.574412</td>\n",
       "      <td>-0.650554</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2000-01-02</th>\n",
       "      <td>0.574412</td>\n",
       "      <td>-0.650554</td>\n",
       "      <td>0.367861</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2000-01-03</th>\n",
       "      <td>-0.650554</td>\n",
       "      <td>0.367861</td>\n",
       "      <td>2.186524</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2000-01-04</th>\n",
       "      <td>0.367861</td>\n",
       "      <td>2.186524</td>\n",
       "      <td>0.001702</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2000-01-05</th>\n",
       "      <td>2.186524</td>\n",
       "      <td>0.001702</td>\n",
       "      <td>-0.173121</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 y_1       y_2       y_3\n",
       "2000-01-01 -0.439480  0.574412 -0.650554\n",
       "2000-01-02  0.574412 -0.650554  0.367861\n",
       "2000-01-03 -0.650554  0.367861  2.186524\n",
       "2000-01-04  0.367861  2.186524  0.001702\n",
       "2000-01-05  2.186524  0.001702 -0.173121"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y = horizon_shift(period_index_time_series, horizon=3)\n",
    "y.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-19T14:42:32.549572Z",
     "start_time": "2019-12-19T14:42:32.547124Z"
    }
   },
   "source": [
    "### Train-Test split"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-20T10:59:18.112521Z",
     "start_time": "2019-12-20T10:59:18.108823Z"
    }
   },
   "source": [
    "We use `FeatureSplitter` to split the matrices X and y in train and test. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-01-24T13:52:36.345799Z",
     "start_time": "2020-01-24T13:52:36.324880Z"
    }
   },
   "outputs": [],
   "source": [
    "feature_splitter = FeatureSplitter()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-01-24T13:52:36.381232Z",
     "start_time": "2020-01-24T13:52:36.347825Z"
    }
   },
   "outputs": [],
   "source": [
    "X_train, y_train, X_test, y_test = feature_splitter.transform(X, y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-19T14:44:02.820817Z",
     "start_time": "2019-12-19T14:44:02.818276Z"
    }
   },
   "source": [
    "### Training"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-20T11:01:12.922844Z",
     "start_time": "2019-12-20T11:01:12.919591Z"
    }
   },
   "source": [
    "We rewrapped scikit-learn's `MultiOutputRegressor` as `GAR` (Generalized Auto Regressive) model to better fit time series forecasting frameworks.\n",
    "\n",
    "The traditional *AR* model is equivalent to the `GAR` model that uses only `Shift` columns in the `X` matrix.\n",
    "`GAR` supports all the features compatible with the feature extraction step.\n",
    "\n",
    "*AR*: https://en.wikipedia.org/wiki/Autoregressive_model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-01-24T13:52:36.405295Z",
     "start_time": "2020-01-24T13:52:36.383404Z"
    }
   },
   "outputs": [],
   "source": [
    "lr = LinearRegression()\n",
    "model = GAR(lr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-01-24T13:52:36.433394Z",
     "start_time": "2020-01-24T13:52:36.407392Z"
    }
   },
   "outputs": [],
   "source": [
    "model = model.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-20T11:01:01.280526Z",
     "start_time": "2019-12-20T11:01:01.278125Z"
    }
   },
   "source": [
    "### Forecasting"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-12-20T11:10:02.544672Z",
     "start_time": "2019-12-20T11:10:02.540859Z"
    }
   },
   "source": [
    "We forecast 3 time steps of the time series (we set this parameter in `horizon_shift` method).\n",
    "\n",
    "The format of the output is the following:\n",
    "- the index is the step at which the prediction is made.\n",
    "- the column `y_1` is the prediction one time step after and so on for `y_2` and `y_3`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-01-24T13:52:36.460582Z",
     "start_time": "2020-01-24T13:52:36.435583Z"
    }
   },
   "outputs": [],
   "source": [
    "predictions = model.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-01-24T13:52:36.489352Z",
     "start_time": "2020-01-24T13:52:36.462210Z"
    }
   },
   "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>y_1</th>\n",
       "      <th>y_2</th>\n",
       "      <th>y_3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2001-05-12</th>\n",
       "      <td>0.016174</td>\n",
       "      <td>-0.099421</td>\n",
       "      <td>-0.119078</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2001-05-13</th>\n",
       "      <td>-0.146184</td>\n",
       "      <td>-0.009442</td>\n",
       "      <td>-0.095851</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2001-05-14</th>\n",
       "      <td>0.007748</td>\n",
       "      <td>-0.052065</td>\n",
       "      <td>-0.107488</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 y_1       y_2       y_3\n",
       "2001-05-12  0.016174 -0.099421 -0.119078\n",
       "2001-05-13 -0.146184 -0.009442 -0.095851\n",
       "2001-05-14  0.007748 -0.052065 -0.107488"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predictions"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "venv-gt",
   "language": "python",
   "name": "venv-gt"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  },
  "pycharm": {
   "stem_cell": {
    "cell_type": "raw",
    "source": [],
    "metadata": {
     "collapsed": false
    }
   }
  },
  "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": {},
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}