{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "c855b45e",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "# AutoML\n",
    "\n",
    "[![Binder](https://mybinder.org/badge_logo.svg)](https://mybinder.org/v2/gh/etna-team/etna/master?filepath=examples/205-automl.ipynb)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bca01a6c",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "This notebooks covers AutoML utilities of ETNA library.\n",
    "\n",
    "**Table of contents**\n",
    "\n",
    "- [Hyperparameters tuning](#chapter_1)\n",
    "    - [How Tune works](#section_1_1)\n",
    "    - [Example](#section_1_2)\n",
    "- [General AutoML](#chapter_2)\n",
    "    - [How Auto works](#section_2_1)\n",
    "    - [Example](#section_2_2)\n",
    "    - [Using custom pipeline pool](#section_2_3)\n",
    "- [Summary](chapter_3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "45f65253",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "!pip install \"etna[auto, prophet]\" -q"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "6f70e872",
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "import warnings\n",
    "\n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "4a371085",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "\n",
    "from etna.datasets import TSDataset\n",
    "from etna.metrics import SMAPE\n",
    "from etna.models import LinearPerSegmentModel\n",
    "from etna.models import NaiveModel\n",
    "from etna.models import ProphetModel\n",
    "from etna.pipeline import Pipeline\n",
    "from etna.transforms import DateFlagsTransform\n",
    "from etna.transforms import LagTransform"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "2f67d0d0",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "HORIZON = 14"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c30553de",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## 1. Hyperparameters tuning <a class=\"anchor\" id=\"chapter_1\"></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e1d81449",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "It is a common task to tune hyperparameters of existing pipeline to improve its quality. For this purpose there is an `etna.auto.Tune` class, which is responsible for creating [optuna](https://github.com/optuna/optuna) study to solve this problem.\n",
    "\n",
    "In the next sections we will see how it works and how to use it for your particular problems."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a79e3a34",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "### 1.1 How `Tune` works <a class=\"anchor\" id=\"section_1_1\"></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "09a9cb75",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "During init `Tune` accepts `pipeline`, its tuning parameters (`params_to_tune`), optimization metric (`target_metric`), parameters of backtest and parameters of optuna study.\n",
    "\n",
    "In `fit` the optuna study is created. During each trial the sample of parameters is generated from `params_to_tune` and applied to `pipeline`. After that, the new pipeline is checked in backtest and target metric is returned to optuna framework."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b5bcae1b",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "Let's look closer at `params_to_tune` parameter. It expects dictionary with parameter names and its distributions. But how this parameter names should be chosen?"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "416208e3",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "#### 1.1.1 `set_params`"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7b39811b",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "We are going to make a little detour to explain the `set_params` method, which is supported by ETNA pipelines, models and transforms. Given a dictionary with parameters it allows to create from existing object a new one with changed parameters."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b1388880",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "First, we define some objects for our future examples."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "19767c13",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "model = LinearPerSegmentModel()\n",
    "transforms = [\n",
    "    LagTransform(in_column=\"target\", lags=list(range(HORIZON, HORIZON + 10)), out_column=\"target_lag\"),\n",
    "    DateFlagsTransform(out_column=\"date_flags\"),\n",
    "]\n",
    "pipeline = Pipeline(model=model, transforms=transforms, horizon=HORIZON)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7aed68b5",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "Let's look at simple example, when we want to change `fit_intercept` parameter of the `model`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "4caacbcb",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'fit_intercept': True,\n",
       " 'kwargs': {},\n",
       " '_target_': 'etna.models.linear.LinearPerSegmentModel'}"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.to_dict()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "e80b47b3",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'fit_intercept': False,\n",
       " 'kwargs': {},\n",
       " '_target_': 'etna.models.linear.LinearPerSegmentModel'}"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "new_model_params = {\"fit_intercept\": False}\n",
    "new_model = model.set_params(**new_model_params)\n",
    "new_model.to_dict()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "523251f2",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "Great! On the next step we want to change the `fit_intercept` of `model` inside the `pipeline`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "6caccffb",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'model': {'fit_intercept': True,\n",
       "  'kwargs': {},\n",
       "  '_target_': 'etna.models.linear.LinearPerSegmentModel'},\n",
       " 'transforms': [{'in_column': 'target',\n",
       "   'lags': [14, 15, 16, 17, 18, 19, 20, 21, 22, 23],\n",
       "   'out_column': 'target_lag',\n",
       "   '_target_': 'etna.transforms.math.lags.LagTransform'},\n",
       "  {'day_number_in_week': True,\n",
       "   'day_number_in_month': True,\n",
       "   'day_number_in_year': False,\n",
       "   'week_number_in_month': False,\n",
       "   'week_number_in_year': False,\n",
       "   'month_number_in_year': False,\n",
       "   'season_number': False,\n",
       "   'year_number': False,\n",
       "   'is_weekend': True,\n",
       "   'special_days_in_week': (),\n",
       "   'special_days_in_month': (),\n",
       "   'out_column': 'date_flags',\n",
       "   '_target_': 'etna.transforms.timestamp.date_flags.DateFlagsTransform'}],\n",
       " 'horizon': 14,\n",
       " '_target_': 'etna.pipeline.pipeline.Pipeline'}"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pipeline.to_dict()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "cb5959d3",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'model': {'fit_intercept': False,\n",
       "  'kwargs': {},\n",
       "  '_target_': 'etna.models.linear.LinearPerSegmentModel'},\n",
       " 'transforms': [{'in_column': 'target',\n",
       "   'lags': [14, 15, 16, 17, 18, 19, 20, 21, 22, 23],\n",
       "   'out_column': 'target_lag',\n",
       "   '_target_': 'etna.transforms.math.lags.LagTransform'},\n",
       "  {'day_number_in_week': True,\n",
       "   'day_number_in_month': True,\n",
       "   'day_number_in_year': False,\n",
       "   'week_number_in_month': False,\n",
       "   'week_number_in_year': False,\n",
       "   'month_number_in_year': False,\n",
       "   'season_number': False,\n",
       "   'year_number': False,\n",
       "   'is_weekend': True,\n",
       "   'special_days_in_week': (),\n",
       "   'special_days_in_month': (),\n",
       "   'out_column': 'date_flags',\n",
       "   '_target_': 'etna.transforms.timestamp.date_flags.DateFlagsTransform'}],\n",
       " 'horizon': 14,\n",
       " '_target_': 'etna.pipeline.pipeline.Pipeline'}"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "new_pipeline_params = {\"model.fit_intercept\": False}\n",
    "new_pipeline = pipeline.set_params(**new_pipeline_params)\n",
    "new_pipeline.to_dict()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fe2ab629",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "Ok, it looks like we managed to do this. On the last step we are going to change `is_weekend` flag of `DateFlagsTransform` inside our `pipeline`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "d658e37b",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'model': {'fit_intercept': True,\n",
       "  'kwargs': {},\n",
       "  '_target_': 'etna.models.linear.LinearPerSegmentModel'},\n",
       " 'transforms': [{'in_column': 'target',\n",
       "   'lags': [14, 15, 16, 17, 18, 19, 20, 21, 22, 23],\n",
       "   'out_column': 'target_lag',\n",
       "   '_target_': 'etna.transforms.math.lags.LagTransform'},\n",
       "  {'day_number_in_week': True,\n",
       "   'day_number_in_month': True,\n",
       "   'day_number_in_year': False,\n",
       "   'week_number_in_month': False,\n",
       "   'week_number_in_year': False,\n",
       "   'month_number_in_year': False,\n",
       "   'season_number': False,\n",
       "   'year_number': False,\n",
       "   'is_weekend': False,\n",
       "   'special_days_in_week': (),\n",
       "   'special_days_in_month': (),\n",
       "   'out_column': 'date_flags',\n",
       "   '_target_': 'etna.transforms.timestamp.date_flags.DateFlagsTransform'}],\n",
       " 'horizon': 14,\n",
       " '_target_': 'etna.pipeline.pipeline.Pipeline'}"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "new_pipeline_params = {\"transforms.1.is_weekend\": False}\n",
    "new_pipeline = pipeline.set_params(**new_pipeline_params)\n",
    "new_pipeline.to_dict()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cc89ba4a",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "As we can see, we managed to do this."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "304a8015",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "#### 1.1.2 `params_to_tune`"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8686327c",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "Let's get back to our initial question about `params_to_tune`. In our optuna study we are going to sample each parameter value from its distribution and pass it into `pipeline.set_params` method. So, the keys for `params_to_tune` should be a valid for `set_params` method.\n",
    "\n",
    "Distributions are taken from `etna.distributions` and they are matching `optuna.Trial.suggest_` methods."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bad3f449",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "For example, something like this will be valid for our `pipeline` defined above:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "cc010639",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "from etna.distributions import CategoricalDistribution\n",
    "\n",
    "example_params_to_tune = {\n",
    "    \"model.fit_intercept\": CategoricalDistribution(choices=[False, True]),\n",
    "    \"transforms.1.is_weekend\": CategoricalDistribution(choices=[False, True]),\n",
    "}"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9d401266",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "This custom dict could be passed into `Tune` class. This will be shown in the [Example](#custom_params) below."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d37306b8",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "There are some good news: it isn't necessary for our users to define `params_to_tune`, because we have a default grid for many of our classes. The default grid is available by calling `params_to_tune` method on pipeline, model or transform. Let's check our `pipeline`:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "6571b3e8",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'model.fit_intercept': CategoricalDistribution(choices=[False, True]),\n",
       " 'transforms.1.day_number_in_week': CategoricalDistribution(choices=[False, True]),\n",
       " 'transforms.1.day_number_in_month': CategoricalDistribution(choices=[False, True]),\n",
       " 'transforms.1.day_number_in_year': CategoricalDistribution(choices=[False, True]),\n",
       " 'transforms.1.week_number_in_month': CategoricalDistribution(choices=[False, True]),\n",
       " 'transforms.1.week_number_in_year': CategoricalDistribution(choices=[False, True]),\n",
       " 'transforms.1.month_number_in_year': CategoricalDistribution(choices=[False, True]),\n",
       " 'transforms.1.season_number': CategoricalDistribution(choices=[False, True]),\n",
       " 'transforms.1.year_number': CategoricalDistribution(choices=[False, True]),\n",
       " 'transforms.1.is_weekend': CategoricalDistribution(choices=[False, True])}"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pipeline.params_to_tune()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fb521dca",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "Now we are ready to use it in practice."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fb561f05",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "### 1.2 Example <a class=\"anchor\" id=\"section_1_2\"></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a3b573d6",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "#### 1.2.1 Loading data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ae581eb6",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "Let's start by loading example data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "66f053a1",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "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>timestamp</th>\n",
       "      <th>segment</th>\n",
       "      <th>target</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2019-01-01</td>\n",
       "      <td>segment_a</td>\n",
       "      <td>170</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2019-01-02</td>\n",
       "      <td>segment_a</td>\n",
       "      <td>243</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2019-01-03</td>\n",
       "      <td>segment_a</td>\n",
       "      <td>267</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2019-01-04</td>\n",
       "      <td>segment_a</td>\n",
       "      <td>287</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2019-01-05</td>\n",
       "      <td>segment_a</td>\n",
       "      <td>279</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    timestamp    segment  target\n",
       "0  2019-01-01  segment_a     170\n",
       "1  2019-01-02  segment_a     243\n",
       "2  2019-01-03  segment_a     267\n",
       "3  2019-01-04  segment_a     287\n",
       "4  2019-01-05  segment_a     279"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv(\"data/example_dataset.csv\")\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "7421a616",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x1000 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "full_ts = TSDataset(df, freq=\"D\")\n",
    "full_ts.plot()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ddc66d43",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "Let's divide current dataset into train and validation parts. We will use validation part later to check final results."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "9c76e694",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "ts, _ = full_ts.train_test_split(test_size=HORIZON * 5)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e3271b39",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "#### 1.2.2 Running `Tune`"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ae2fbaed",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "We are going to define our `Tune` object:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "db305339",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "from etna.auto import Tune\n",
    "\n",
    "tune = Tune(pipeline=pipeline, target_metric=SMAPE(), horizon=HORIZON, backtest_params=dict(n_folds=5))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4a27b5ac",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "We used mostly default parameters for this example. But for your own experiments you might want to also set up other parameters. \n",
    "\n",
    "For example, parameter `runner` allows you to run tuning in parallel on a local machine, and parameter `storage` makes it possible to store optuna results on a dedicated remote server.\n",
    "\n",
    "For a full list of parameters we advise you to check our documentation."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ae2d89ca",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "Let's hide the logs of optuna, there are too many of them for a notebook."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "22fbec9f",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "import optuna\n",
    "\n",
    "optuna.logging.set_verbosity(optuna.logging.CRITICAL)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8ad33ea7",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "Let's run the tuning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "0a4a7781",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "%%capture\n",
    "best_pipeline = tune.fit(ts=ts, n_trials=20)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fb6e5682",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "Command `%%capture` just hides the output."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cd45ff60",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "#### 1.2.3 Running `Tune` with custom `params_to_tune`<a class=\"anchor\" id=\"custom_params\"></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "14fefb85",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "Let's remember that earlier we created a dict:\n",
    "```python\n",
    "example_params_to_tune = {\n",
    "    \"model.fit_intercept\": CategoricalDistribution(choices=[False, True]),\n",
    "    \"transforms.1.is_weekend\": CategoricalDistribution(choices=[False, True]),\n",
    "}\n",
    "```\n",
    "Now we can use these parameters when initializing `Tune`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "4157e5c6",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "tune_custom_params = Tune(\n",
    "    pipeline=pipeline,\n",
    "    target_metric=SMAPE(),\n",
    "    horizon=HORIZON,\n",
    "    backtest_params=dict(n_folds=5),\n",
    "    params_to_tune=example_params_to_tune,\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dcc67b63",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "Let's run the tuning with our custom parameters."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "996e0ee0",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "%%capture\n",
    "best_pipeline_custom_params = tune_custom_params.fit(ts=ts, n_trials=20)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f8bdaf7e",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "#### 1.2.4 Analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a84cdb9e",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "In the last section dedicated to `Tune` we will look at methods for result analysis."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ea57eccc",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "First of all there is `summary` method that shows us the results of optuna trials."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "004dcb2a",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "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>MAE_mean</th>\n",
       "      <th>MAE_median</th>\n",
       "      <th>MAE_notna_size</th>\n",
       "      <th>MAE_percentile_25</th>\n",
       "      <th>MAE_percentile_5</th>\n",
       "      <th>MAE_percentile_75</th>\n",
       "      <th>MAE_percentile_95</th>\n",
       "      <th>MAE_std</th>\n",
       "      <th>MSE_mean</th>\n",
       "      <th>MSE_median</th>\n",
       "      <th>...</th>\n",
       "      <th>Sign_notna_size</th>\n",
       "      <th>Sign_percentile_25</th>\n",
       "      <th>Sign_percentile_5</th>\n",
       "      <th>Sign_percentile_75</th>\n",
       "      <th>Sign_percentile_95</th>\n",
       "      <th>Sign_std</th>\n",
       "      <th>elapsed_time</th>\n",
       "      <th>hash</th>\n",
       "      <th>pipeline</th>\n",
       "      <th>state</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>25.667904</td>\n",
       "      <td>26.788472</td>\n",
       "      <td>4.0</td>\n",
       "      <td>20.385243</td>\n",
       "      <td>15.145294</td>\n",
       "      <td>32.071133</td>\n",
       "      <td>34.621717</td>\n",
       "      <td>8.217865</td>\n",
       "      <td>1527.813269</td>\n",
       "      <td>1356.477027</td>\n",
       "      <td>...</td>\n",
       "      <td>4.0</td>\n",
       "      <td>-0.621429</td>\n",
       "      <td>-0.672857</td>\n",
       "      <td>-0.357143</td>\n",
       "      <td>-0.254286</td>\n",
       "      <td>0.177712</td>\n",
       "      <td>1.586009</td>\n",
       "      <td>f4f02e1d5f60b8f322a4a8a622dd1c1e</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>26.606482</td>\n",
       "      <td>28.272429</td>\n",
       "      <td>4.0</td>\n",
       "      <td>20.554262</td>\n",
       "      <td>15.198315</td>\n",
       "      <td>34.324649</td>\n",
       "      <td>35.682322</td>\n",
       "      <td>8.900725</td>\n",
       "      <td>1729.295705</td>\n",
       "      <td>1599.030615</td>\n",
       "      <td>...</td>\n",
       "      <td>4.0</td>\n",
       "      <td>-0.642857</td>\n",
       "      <td>-0.745714</td>\n",
       "      <td>-0.300000</td>\n",
       "      <td>-0.265714</td>\n",
       "      <td>0.209956</td>\n",
       "      <td>1.700775</td>\n",
       "      <td>3d7b7af16d71a36f3b935f69e113e22d</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>26.427670</td>\n",
       "      <td>26.809357</td>\n",
       "      <td>4.0</td>\n",
       "      <td>20.238589</td>\n",
       "      <td>11.884902</td>\n",
       "      <td>32.998438</td>\n",
       "      <td>40.436078</td>\n",
       "      <td>11.702270</td>\n",
       "      <td>2133.343967</td>\n",
       "      <td>1997.594638</td>\n",
       "      <td>...</td>\n",
       "      <td>4.0</td>\n",
       "      <td>-0.392857</td>\n",
       "      <td>-0.581429</td>\n",
       "      <td>-0.078571</td>\n",
       "      <td>-0.061429</td>\n",
       "      <td>0.229017</td>\n",
       "      <td>1.096877</td>\n",
       "      <td>7c7932114268832a5458acfecfb453fc</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>34.586733</td>\n",
       "      <td>35.616831</td>\n",
       "      <td>4.0</td>\n",
       "      <td>29.797233</td>\n",
       "      <td>16.157674</td>\n",
       "      <td>40.406330</td>\n",
       "      <td>51.573653</td>\n",
       "      <td>14.750408</td>\n",
       "      <td>3185.461093</td>\n",
       "      <td>3324.924980</td>\n",
       "      <td>...</td>\n",
       "      <td>4.0</td>\n",
       "      <td>-0.100000</td>\n",
       "      <td>-0.340000</td>\n",
       "      <td>0.014286</td>\n",
       "      <td>0.048571</td>\n",
       "      <td>0.182946</td>\n",
       "      <td>0.989273</td>\n",
       "      <td>b7ac5f7fcf9c8959626befe263a9d561</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>28.046631</td>\n",
       "      <td>28.279612</td>\n",
       "      <td>4.0</td>\n",
       "      <td>22.153130</td>\n",
       "      <td>14.349408</td>\n",
       "      <td>34.173114</td>\n",
       "      <td>41.417681</td>\n",
       "      <td>11.090567</td>\n",
       "      <td>1875.958439</td>\n",
       "      <td>2045.887754</td>\n",
       "      <td>...</td>\n",
       "      <td>4.0</td>\n",
       "      <td>-0.585714</td>\n",
       "      <td>-0.620000</td>\n",
       "      <td>-0.250000</td>\n",
       "      <td>-0.232857</td>\n",
       "      <td>0.179994</td>\n",
       "      <td>0.785666</td>\n",
       "      <td>e928929f89156d88ef49e28abaf55847</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>31.714578</td>\n",
       "      <td>34.650441</td>\n",
       "      <td>4.0</td>\n",
       "      <td>25.441858</td>\n",
       "      <td>14.662179</td>\n",
       "      <td>40.923160</td>\n",
       "      <td>44.656767</td>\n",
       "      <td>12.690827</td>\n",
       "      <td>2496.862613</td>\n",
       "      <td>2344.530581</td>\n",
       "      <td>...</td>\n",
       "      <td>4.0</td>\n",
       "      <td>-0.514286</td>\n",
       "      <td>-0.788571</td>\n",
       "      <td>-0.271429</td>\n",
       "      <td>0.037143</td>\n",
       "      <td>0.343749</td>\n",
       "      <td>1.141506</td>\n",
       "      <td>3b4311d41fcaab7307235ea23b6d4599</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>28.401011</td>\n",
       "      <td>30.937439</td>\n",
       "      <td>4.0</td>\n",
       "      <td>22.517391</td>\n",
       "      <td>15.064969</td>\n",
       "      <td>36.821059</td>\n",
       "      <td>38.186053</td>\n",
       "      <td>10.036415</td>\n",
       "      <td>2160.703619</td>\n",
       "      <td>1565.256313</td>\n",
       "      <td>...</td>\n",
       "      <td>4.0</td>\n",
       "      <td>-0.621429</td>\n",
       "      <td>-0.672857</td>\n",
       "      <td>-0.257143</td>\n",
       "      <td>-0.188571</td>\n",
       "      <td>0.213212</td>\n",
       "      <td>0.849209</td>\n",
       "      <td>74065ebc11c81bed6a9819d026c7cd84</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>28.834402</td>\n",
       "      <td>30.782735</td>\n",
       "      <td>4.0</td>\n",
       "      <td>23.540480</td>\n",
       "      <td>15.883965</td>\n",
       "      <td>36.076656</td>\n",
       "      <td>39.057173</td>\n",
       "      <td>9.768319</td>\n",
       "      <td>1887.330868</td>\n",
       "      <td>1671.590780</td>\n",
       "      <td>...</td>\n",
       "      <td>4.0</td>\n",
       "      <td>-0.657143</td>\n",
       "      <td>-0.725714</td>\n",
       "      <td>-0.250000</td>\n",
       "      <td>-0.232857</td>\n",
       "      <td>0.225312</td>\n",
       "      <td>0.883154</td>\n",
       "      <td>b0d0420255c6117045f8254bf8f377a0</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>27.589468</td>\n",
       "      <td>28.296973</td>\n",
       "      <td>4.0</td>\n",
       "      <td>21.668623</td>\n",
       "      <td>14.025183</td>\n",
       "      <td>34.217818</td>\n",
       "      <td>40.163245</td>\n",
       "      <td>10.745756</td>\n",
       "      <td>1861.965363</td>\n",
       "      <td>2038.946807</td>\n",
       "      <td>...</td>\n",
       "      <td>4.0</td>\n",
       "      <td>-0.671429</td>\n",
       "      <td>-0.705714</td>\n",
       "      <td>-0.242857</td>\n",
       "      <td>-0.208571</td>\n",
       "      <td>0.230350</td>\n",
       "      <td>0.875966</td>\n",
       "      <td>25dcd8bb095f87a1ffc499fa6a83ef5d</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>29.720151</td>\n",
       "      <td>28.713470</td>\n",
       "      <td>4.0</td>\n",
       "      <td>23.395341</td>\n",
       "      <td>14.726153</td>\n",
       "      <td>35.038280</td>\n",
       "      <td>46.123501</td>\n",
       "      <td>12.942420</td>\n",
       "      <td>1975.650004</td>\n",
       "      <td>1963.591771</td>\n",
       "      <td>...</td>\n",
       "      <td>4.0</td>\n",
       "      <td>-0.657143</td>\n",
       "      <td>-0.725714</td>\n",
       "      <td>-0.192857</td>\n",
       "      <td>-0.175714</td>\n",
       "      <td>0.253446</td>\n",
       "      <td>1.043901</td>\n",
       "      <td>3f1ca1759261598081fa3bb2f32fe0ac</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>25.598822</td>\n",
       "      <td>22.912185</td>\n",
       "      <td>4.0</td>\n",
       "      <td>19.472060</td>\n",
       "      <td>11.719988</td>\n",
       "      <td>29.038946</td>\n",
       "      <td>43.238949</td>\n",
       "      <td>13.357756</td>\n",
       "      <td>1822.076084</td>\n",
       "      <td>1734.654459</td>\n",
       "      <td>...</td>\n",
       "      <td>4.0</td>\n",
       "      <td>-0.328571</td>\n",
       "      <td>-0.431429</td>\n",
       "      <td>-0.014286</td>\n",
       "      <td>0.020000</td>\n",
       "      <td>0.196396</td>\n",
       "      <td>1.131200</td>\n",
       "      <td>8363309e454e72993f86f10c7fc7c137</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>25.598822</td>\n",
       "      <td>22.912185</td>\n",
       "      <td>4.0</td>\n",
       "      <td>19.472060</td>\n",
       "      <td>11.719988</td>\n",
       "      <td>29.038946</td>\n",
       "      <td>43.238949</td>\n",
       "      <td>13.357756</td>\n",
       "      <td>1822.076084</td>\n",
       "      <td>1734.654459</td>\n",
       "      <td>...</td>\n",
       "      <td>4.0</td>\n",
       "      <td>-0.328571</td>\n",
       "      <td>-0.431429</td>\n",
       "      <td>-0.014286</td>\n",
       "      <td>0.020000</td>\n",
       "      <td>0.196396</td>\n",
       "      <td>1.131200</td>\n",
       "      <td>8363309e454e72993f86f10c7fc7c137</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>25.598822</td>\n",
       "      <td>22.912185</td>\n",
       "      <td>4.0</td>\n",
       "      <td>19.472060</td>\n",
       "      <td>11.719988</td>\n",
       "      <td>29.038946</td>\n",
       "      <td>43.238949</td>\n",
       "      <td>13.357756</td>\n",
       "      <td>1822.076084</td>\n",
       "      <td>1734.654459</td>\n",
       "      <td>...</td>\n",
       "      <td>4.0</td>\n",
       "      <td>-0.328571</td>\n",
       "      <td>-0.431429</td>\n",
       "      <td>-0.014286</td>\n",
       "      <td>0.020000</td>\n",
       "      <td>0.196396</td>\n",
       "      <td>1.131200</td>\n",
       "      <td>8363309e454e72993f86f10c7fc7c137</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>25.598822</td>\n",
       "      <td>22.912185</td>\n",
       "      <td>4.0</td>\n",
       "      <td>19.472060</td>\n",
       "      <td>11.719988</td>\n",
       "      <td>29.038946</td>\n",
       "      <td>43.238949</td>\n",
       "      <td>13.357756</td>\n",
       "      <td>1822.076084</td>\n",
       "      <td>1734.654459</td>\n",
       "      <td>...</td>\n",
       "      <td>4.0</td>\n",
       "      <td>-0.328571</td>\n",
       "      <td>-0.431429</td>\n",
       "      <td>-0.014286</td>\n",
       "      <td>0.020000</td>\n",
       "      <td>0.196396</td>\n",
       "      <td>1.131200</td>\n",
       "      <td>8363309e454e72993f86f10c7fc7c137</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>25.598822</td>\n",
       "      <td>22.912185</td>\n",
       "      <td>4.0</td>\n",
       "      <td>19.472060</td>\n",
       "      <td>11.719988</td>\n",
       "      <td>29.038946</td>\n",
       "      <td>43.238949</td>\n",
       "      <td>13.357756</td>\n",
       "      <td>1822.076084</td>\n",
       "      <td>1734.654459</td>\n",
       "      <td>...</td>\n",
       "      <td>4.0</td>\n",
       "      <td>-0.328571</td>\n",
       "      <td>-0.431429</td>\n",
       "      <td>-0.014286</td>\n",
       "      <td>0.020000</td>\n",
       "      <td>0.196396</td>\n",
       "      <td>1.131200</td>\n",
       "      <td>8363309e454e72993f86f10c7fc7c137</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>25.598822</td>\n",
       "      <td>22.912185</td>\n",
       "      <td>4.0</td>\n",
       "      <td>19.472060</td>\n",
       "      <td>11.719988</td>\n",
       "      <td>29.038946</td>\n",
       "      <td>43.238949</td>\n",
       "      <td>13.357756</td>\n",
       "      <td>1822.076084</td>\n",
       "      <td>1734.654459</td>\n",
       "      <td>...</td>\n",
       "      <td>4.0</td>\n",
       "      <td>-0.328571</td>\n",
       "      <td>-0.431429</td>\n",
       "      <td>-0.014286</td>\n",
       "      <td>0.020000</td>\n",
       "      <td>0.196396</td>\n",
       "      <td>1.131200</td>\n",
       "      <td>8363309e454e72993f86f10c7fc7c137</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>25.598822</td>\n",
       "      <td>22.912185</td>\n",
       "      <td>4.0</td>\n",
       "      <td>19.472060</td>\n",
       "      <td>11.719988</td>\n",
       "      <td>29.038946</td>\n",
       "      <td>43.238949</td>\n",
       "      <td>13.357756</td>\n",
       "      <td>1822.076084</td>\n",
       "      <td>1734.654459</td>\n",
       "      <td>...</td>\n",
       "      <td>4.0</td>\n",
       "      <td>-0.328571</td>\n",
       "      <td>-0.431429</td>\n",
       "      <td>-0.014286</td>\n",
       "      <td>0.020000</td>\n",
       "      <td>0.196396</td>\n",
       "      <td>1.131200</td>\n",
       "      <td>8363309e454e72993f86f10c7fc7c137</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>25.598822</td>\n",
       "      <td>22.912185</td>\n",
       "      <td>4.0</td>\n",
       "      <td>19.472060</td>\n",
       "      <td>11.719988</td>\n",
       "      <td>29.038946</td>\n",
       "      <td>43.238949</td>\n",
       "      <td>13.357756</td>\n",
       "      <td>1822.076084</td>\n",
       "      <td>1734.654459</td>\n",
       "      <td>...</td>\n",
       "      <td>4.0</td>\n",
       "      <td>-0.328571</td>\n",
       "      <td>-0.431429</td>\n",
       "      <td>-0.014286</td>\n",
       "      <td>0.020000</td>\n",
       "      <td>0.196396</td>\n",
       "      <td>1.131200</td>\n",
       "      <td>8363309e454e72993f86f10c7fc7c137</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>25.288430</td>\n",
       "      <td>26.243869</td>\n",
       "      <td>4.0</td>\n",
       "      <td>18.913168</td>\n",
       "      <td>14.518271</td>\n",
       "      <td>32.619131</td>\n",
       "      <td>34.720975</td>\n",
       "      <td>8.694062</td>\n",
       "      <td>1441.296901</td>\n",
       "      <td>1366.199962</td>\n",
       "      <td>...</td>\n",
       "      <td>4.0</td>\n",
       "      <td>-0.685714</td>\n",
       "      <td>-0.754286</td>\n",
       "      <td>-0.164286</td>\n",
       "      <td>-0.147143</td>\n",
       "      <td>0.281668</td>\n",
       "      <td>1.071287</td>\n",
       "      <td>6f595f4f43b323804c04d4cea49c169b</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>25.598822</td>\n",
       "      <td>22.912185</td>\n",
       "      <td>4.0</td>\n",
       "      <td>19.472060</td>\n",
       "      <td>11.719988</td>\n",
       "      <td>29.038946</td>\n",
       "      <td>43.238949</td>\n",
       "      <td>13.357756</td>\n",
       "      <td>1822.076084</td>\n",
       "      <td>1734.654459</td>\n",
       "      <td>...</td>\n",
       "      <td>4.0</td>\n",
       "      <td>-0.328571</td>\n",
       "      <td>-0.431429</td>\n",
       "      <td>-0.014286</td>\n",
       "      <td>0.020000</td>\n",
       "      <td>0.196396</td>\n",
       "      <td>1.131200</td>\n",
       "      <td>8363309e454e72993f86f10c7fc7c137</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>20 rows × 44 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     MAE_mean  MAE_median  MAE_notna_size  MAE_percentile_25  \\\n",
       "0   25.667904   26.788472             4.0          20.385243   \n",
       "1   26.606482   28.272429             4.0          20.554262   \n",
       "2   26.427670   26.809357             4.0          20.238589   \n",
       "3   34.586733   35.616831             4.0          29.797233   \n",
       "4   28.046631   28.279612             4.0          22.153130   \n",
       "5   31.714578   34.650441             4.0          25.441858   \n",
       "6   28.401011   30.937439             4.0          22.517391   \n",
       "7   28.834402   30.782735             4.0          23.540480   \n",
       "8   27.589468   28.296973             4.0          21.668623   \n",
       "9   29.720151   28.713470             4.0          23.395341   \n",
       "10  25.598822   22.912185             4.0          19.472060   \n",
       "11  25.598822   22.912185             4.0          19.472060   \n",
       "12  25.598822   22.912185             4.0          19.472060   \n",
       "13  25.598822   22.912185             4.0          19.472060   \n",
       "14  25.598822   22.912185             4.0          19.472060   \n",
       "15  25.598822   22.912185             4.0          19.472060   \n",
       "16  25.598822   22.912185             4.0          19.472060   \n",
       "17  25.598822   22.912185             4.0          19.472060   \n",
       "18  25.288430   26.243869             4.0          18.913168   \n",
       "19  25.598822   22.912185             4.0          19.472060   \n",
       "\n",
       "    MAE_percentile_5  MAE_percentile_75  MAE_percentile_95    MAE_std  \\\n",
       "0          15.145294          32.071133          34.621717   8.217865   \n",
       "1          15.198315          34.324649          35.682322   8.900725   \n",
       "2          11.884902          32.998438          40.436078  11.702270   \n",
       "3          16.157674          40.406330          51.573653  14.750408   \n",
       "4          14.349408          34.173114          41.417681  11.090567   \n",
       "5          14.662179          40.923160          44.656767  12.690827   \n",
       "6          15.064969          36.821059          38.186053  10.036415   \n",
       "7          15.883965          36.076656          39.057173   9.768319   \n",
       "8          14.025183          34.217818          40.163245  10.745756   \n",
       "9          14.726153          35.038280          46.123501  12.942420   \n",
       "10         11.719988          29.038946          43.238949  13.357756   \n",
       "11         11.719988          29.038946          43.238949  13.357756   \n",
       "12         11.719988          29.038946          43.238949  13.357756   \n",
       "13         11.719988          29.038946          43.238949  13.357756   \n",
       "14         11.719988          29.038946          43.238949  13.357756   \n",
       "15         11.719988          29.038946          43.238949  13.357756   \n",
       "16         11.719988          29.038946          43.238949  13.357756   \n",
       "17         11.719988          29.038946          43.238949  13.357756   \n",
       "18         14.518271          32.619131          34.720975   8.694062   \n",
       "19         11.719988          29.038946          43.238949  13.357756   \n",
       "\n",
       "       MSE_mean   MSE_median  ...  Sign_notna_size  Sign_percentile_25  \\\n",
       "0   1527.813269  1356.477027  ...              4.0           -0.621429   \n",
       "1   1729.295705  1599.030615  ...              4.0           -0.642857   \n",
       "2   2133.343967  1997.594638  ...              4.0           -0.392857   \n",
       "3   3185.461093  3324.924980  ...              4.0           -0.100000   \n",
       "4   1875.958439  2045.887754  ...              4.0           -0.585714   \n",
       "5   2496.862613  2344.530581  ...              4.0           -0.514286   \n",
       "6   2160.703619  1565.256313  ...              4.0           -0.621429   \n",
       "7   1887.330868  1671.590780  ...              4.0           -0.657143   \n",
       "8   1861.965363  2038.946807  ...              4.0           -0.671429   \n",
       "9   1975.650004  1963.591771  ...              4.0           -0.657143   \n",
       "10  1822.076084  1734.654459  ...              4.0           -0.328571   \n",
       "11  1822.076084  1734.654459  ...              4.0           -0.328571   \n",
       "12  1822.076084  1734.654459  ...              4.0           -0.328571   \n",
       "13  1822.076084  1734.654459  ...              4.0           -0.328571   \n",
       "14  1822.076084  1734.654459  ...              4.0           -0.328571   \n",
       "15  1822.076084  1734.654459  ...              4.0           -0.328571   \n",
       "16  1822.076084  1734.654459  ...              4.0           -0.328571   \n",
       "17  1822.076084  1734.654459  ...              4.0           -0.328571   \n",
       "18  1441.296901  1366.199962  ...              4.0           -0.685714   \n",
       "19  1822.076084  1734.654459  ...              4.0           -0.328571   \n",
       "\n",
       "    Sign_percentile_5  Sign_percentile_75  Sign_percentile_95  Sign_std  \\\n",
       "0           -0.672857           -0.357143           -0.254286  0.177712   \n",
       "1           -0.745714           -0.300000           -0.265714  0.209956   \n",
       "2           -0.581429           -0.078571           -0.061429  0.229017   \n",
       "3           -0.340000            0.014286            0.048571  0.182946   \n",
       "4           -0.620000           -0.250000           -0.232857  0.179994   \n",
       "5           -0.788571           -0.271429            0.037143  0.343749   \n",
       "6           -0.672857           -0.257143           -0.188571  0.213212   \n",
       "7           -0.725714           -0.250000           -0.232857  0.225312   \n",
       "8           -0.705714           -0.242857           -0.208571  0.230350   \n",
       "9           -0.725714           -0.192857           -0.175714  0.253446   \n",
       "10          -0.431429           -0.014286            0.020000  0.196396   \n",
       "11          -0.431429           -0.014286            0.020000  0.196396   \n",
       "12          -0.431429           -0.014286            0.020000  0.196396   \n",
       "13          -0.431429           -0.014286            0.020000  0.196396   \n",
       "14          -0.431429           -0.014286            0.020000  0.196396   \n",
       "15          -0.431429           -0.014286            0.020000  0.196396   \n",
       "16          -0.431429           -0.014286            0.020000  0.196396   \n",
       "17          -0.431429           -0.014286            0.020000  0.196396   \n",
       "18          -0.754286           -0.164286           -0.147143  0.281668   \n",
       "19          -0.431429           -0.014286            0.020000  0.196396   \n",
       "\n",
       "    elapsed_time                              hash  \\\n",
       "0       1.586009  f4f02e1d5f60b8f322a4a8a622dd1c1e   \n",
       "1       1.700775  3d7b7af16d71a36f3b935f69e113e22d   \n",
       "2       1.096877  7c7932114268832a5458acfecfb453fc   \n",
       "3       0.989273  b7ac5f7fcf9c8959626befe263a9d561   \n",
       "4       0.785666  e928929f89156d88ef49e28abaf55847   \n",
       "5       1.141506  3b4311d41fcaab7307235ea23b6d4599   \n",
       "6       0.849209  74065ebc11c81bed6a9819d026c7cd84   \n",
       "7       0.883154  b0d0420255c6117045f8254bf8f377a0   \n",
       "8       0.875966  25dcd8bb095f87a1ffc499fa6a83ef5d   \n",
       "9       1.043901  3f1ca1759261598081fa3bb2f32fe0ac   \n",
       "10      1.131200  8363309e454e72993f86f10c7fc7c137   \n",
       "11      1.131200  8363309e454e72993f86f10c7fc7c137   \n",
       "12      1.131200  8363309e454e72993f86f10c7fc7c137   \n",
       "13      1.131200  8363309e454e72993f86f10c7fc7c137   \n",
       "14      1.131200  8363309e454e72993f86f10c7fc7c137   \n",
       "15      1.131200  8363309e454e72993f86f10c7fc7c137   \n",
       "16      1.131200  8363309e454e72993f86f10c7fc7c137   \n",
       "17      1.131200  8363309e454e72993f86f10c7fc7c137   \n",
       "18      1.071287  6f595f4f43b323804c04d4cea49c169b   \n",
       "19      1.131200  8363309e454e72993f86f10c7fc7c137   \n",
       "\n",
       "                                             pipeline  state  \n",
       "0   Pipeline(model = LinearPerSegmentModel(fit_int...      1  \n",
       "1   Pipeline(model = LinearPerSegmentModel(fit_int...      1  \n",
       "2   Pipeline(model = LinearPerSegmentModel(fit_int...      1  \n",
       "3   Pipeline(model = LinearPerSegmentModel(fit_int...      1  \n",
       "4   Pipeline(model = LinearPerSegmentModel(fit_int...      1  \n",
       "5   Pipeline(model = LinearPerSegmentModel(fit_int...      1  \n",
       "6   Pipeline(model = LinearPerSegmentModel(fit_int...      1  \n",
       "7   Pipeline(model = LinearPerSegmentModel(fit_int...      1  \n",
       "8   Pipeline(model = LinearPerSegmentModel(fit_int...      1  \n",
       "9   Pipeline(model = LinearPerSegmentModel(fit_int...      1  \n",
       "10  Pipeline(model = LinearPerSegmentModel(fit_int...      1  \n",
       "11  Pipeline(model = LinearPerSegmentModel(fit_int...      1  \n",
       "12  Pipeline(model = LinearPerSegmentModel(fit_int...      1  \n",
       "13  Pipeline(model = LinearPerSegmentModel(fit_int...      1  \n",
       "14  Pipeline(model = LinearPerSegmentModel(fit_int...      1  \n",
       "15  Pipeline(model = LinearPerSegmentModel(fit_int...      1  \n",
       "16  Pipeline(model = LinearPerSegmentModel(fit_int...      1  \n",
       "17  Pipeline(model = LinearPerSegmentModel(fit_int...      1  \n",
       "18  Pipeline(model = LinearPerSegmentModel(fit_int...      1  \n",
       "19  Pipeline(model = LinearPerSegmentModel(fit_int...      1  \n",
       "\n",
       "[20 rows x 44 columns]"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tune.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a95dc86a",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "Let's show only the columns we are interested in."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "88427207",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "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>hash</th>\n",
       "      <th>pipeline</th>\n",
       "      <th>SMAPE_mean</th>\n",
       "      <th>elapsed_time</th>\n",
       "      <th>state</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>8363309e454e72993f86f10c7fc7c137</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>8.556535</td>\n",
       "      <td>1.131200</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>8363309e454e72993f86f10c7fc7c137</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>8.556535</td>\n",
       "      <td>1.131200</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>8363309e454e72993f86f10c7fc7c137</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>8.556535</td>\n",
       "      <td>1.131200</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>8363309e454e72993f86f10c7fc7c137</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>8.556535</td>\n",
       "      <td>1.131200</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>8363309e454e72993f86f10c7fc7c137</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>8.556535</td>\n",
       "      <td>1.131200</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>8363309e454e72993f86f10c7fc7c137</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>8.556535</td>\n",
       "      <td>1.131200</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>8363309e454e72993f86f10c7fc7c137</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>8.556535</td>\n",
       "      <td>1.131200</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>8363309e454e72993f86f10c7fc7c137</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>8.556535</td>\n",
       "      <td>1.131200</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>8363309e454e72993f86f10c7fc7c137</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>8.556535</td>\n",
       "      <td>1.131200</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>7c7932114268832a5458acfecfb453fc</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>9.210183</td>\n",
       "      <td>1.096877</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>25dcd8bb095f87a1ffc499fa6a83ef5d</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>9.943658</td>\n",
       "      <td>0.875966</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>e928929f89156d88ef49e28abaf55847</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>9.946866</td>\n",
       "      <td>0.785666</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>f4f02e1d5f60b8f322a4a8a622dd1c1e</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>9.957781</td>\n",
       "      <td>1.586009</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>6f595f4f43b323804c04d4cea49c169b</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>10.061742</td>\n",
       "      <td>1.071287</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3d7b7af16d71a36f3b935f69e113e22d</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>10.306909</td>\n",
       "      <td>1.700775</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>3f1ca1759261598081fa3bb2f32fe0ac</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>10.554444</td>\n",
       "      <td>1.043901</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>3b4311d41fcaab7307235ea23b6d4599</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>10.756703</td>\n",
       "      <td>1.141506</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>74065ebc11c81bed6a9819d026c7cd84</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>10.917164</td>\n",
       "      <td>0.849209</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>b7ac5f7fcf9c8959626befe263a9d561</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>11.255320</td>\n",
       "      <td>0.989273</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>b0d0420255c6117045f8254bf8f377a0</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>11.478760</td>\n",
       "      <td>0.883154</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                hash  \\\n",
       "19  8363309e454e72993f86f10c7fc7c137   \n",
       "17  8363309e454e72993f86f10c7fc7c137   \n",
       "16  8363309e454e72993f86f10c7fc7c137   \n",
       "15  8363309e454e72993f86f10c7fc7c137   \n",
       "14  8363309e454e72993f86f10c7fc7c137   \n",
       "13  8363309e454e72993f86f10c7fc7c137   \n",
       "12  8363309e454e72993f86f10c7fc7c137   \n",
       "10  8363309e454e72993f86f10c7fc7c137   \n",
       "11  8363309e454e72993f86f10c7fc7c137   \n",
       "2   7c7932114268832a5458acfecfb453fc   \n",
       "8   25dcd8bb095f87a1ffc499fa6a83ef5d   \n",
       "4   e928929f89156d88ef49e28abaf55847   \n",
       "0   f4f02e1d5f60b8f322a4a8a622dd1c1e   \n",
       "18  6f595f4f43b323804c04d4cea49c169b   \n",
       "1   3d7b7af16d71a36f3b935f69e113e22d   \n",
       "9   3f1ca1759261598081fa3bb2f32fe0ac   \n",
       "5   3b4311d41fcaab7307235ea23b6d4599   \n",
       "6   74065ebc11c81bed6a9819d026c7cd84   \n",
       "3   b7ac5f7fcf9c8959626befe263a9d561   \n",
       "7   b0d0420255c6117045f8254bf8f377a0   \n",
       "\n",
       "                                             pipeline  SMAPE_mean  \\\n",
       "19  Pipeline(model = LinearPerSegmentModel(fit_int...    8.556535   \n",
       "17  Pipeline(model = LinearPerSegmentModel(fit_int...    8.556535   \n",
       "16  Pipeline(model = LinearPerSegmentModel(fit_int...    8.556535   \n",
       "15  Pipeline(model = LinearPerSegmentModel(fit_int...    8.556535   \n",
       "14  Pipeline(model = LinearPerSegmentModel(fit_int...    8.556535   \n",
       "13  Pipeline(model = LinearPerSegmentModel(fit_int...    8.556535   \n",
       "12  Pipeline(model = LinearPerSegmentModel(fit_int...    8.556535   \n",
       "10  Pipeline(model = LinearPerSegmentModel(fit_int...    8.556535   \n",
       "11  Pipeline(model = LinearPerSegmentModel(fit_int...    8.556535   \n",
       "2   Pipeline(model = LinearPerSegmentModel(fit_int...    9.210183   \n",
       "8   Pipeline(model = LinearPerSegmentModel(fit_int...    9.943658   \n",
       "4   Pipeline(model = LinearPerSegmentModel(fit_int...    9.946866   \n",
       "0   Pipeline(model = LinearPerSegmentModel(fit_int...    9.957781   \n",
       "18  Pipeline(model = LinearPerSegmentModel(fit_int...   10.061742   \n",
       "1   Pipeline(model = LinearPerSegmentModel(fit_int...   10.306909   \n",
       "9   Pipeline(model = LinearPerSegmentModel(fit_int...   10.554444   \n",
       "5   Pipeline(model = LinearPerSegmentModel(fit_int...   10.756703   \n",
       "6   Pipeline(model = LinearPerSegmentModel(fit_int...   10.917164   \n",
       "3   Pipeline(model = LinearPerSegmentModel(fit_int...   11.255320   \n",
       "7   Pipeline(model = LinearPerSegmentModel(fit_int...   11.478760   \n",
       "\n",
       "    elapsed_time  state  \n",
       "19      1.131200      1  \n",
       "17      1.131200      1  \n",
       "16      1.131200      1  \n",
       "15      1.131200      1  \n",
       "14      1.131200      1  \n",
       "13      1.131200      1  \n",
       "12      1.131200      1  \n",
       "10      1.131200      1  \n",
       "11      1.131200      1  \n",
       "2       1.096877      1  \n",
       "8       0.875966      1  \n",
       "4       0.785666      1  \n",
       "0       1.586009      1  \n",
       "18      1.071287      1  \n",
       "1       1.700775      1  \n",
       "9       1.043901      1  \n",
       "5       1.141506      1  \n",
       "6       0.849209      1  \n",
       "3       0.989273      1  \n",
       "7       0.883154      1  "
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tune.summary()[[\"hash\", \"pipeline\", \"SMAPE_mean\", \"elapsed_time\", \"state\"]].sort_values(\"SMAPE_mean\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "08736c75",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "As we can see, we have duplicate lines according to the `hash` column. Some trials have the same sampled hyperparameters and they have the same results. We have a special handling for such duplicates: they are skipped during optimization and the previously computed metric values are returned.\n",
    "\n",
    "Duplicates on the summary can be eliminated using `hash` column."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "6433880c",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "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>hash</th>\n",
       "      <th>pipeline</th>\n",
       "      <th>SMAPE_mean</th>\n",
       "      <th>elapsed_time</th>\n",
       "      <th>state</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>8363309e454e72993f86f10c7fc7c137</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>8.556535</td>\n",
       "      <td>1.131200</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>7c7932114268832a5458acfecfb453fc</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>9.210183</td>\n",
       "      <td>1.096877</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>25dcd8bb095f87a1ffc499fa6a83ef5d</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>9.943658</td>\n",
       "      <td>0.875966</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>e928929f89156d88ef49e28abaf55847</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>9.946866</td>\n",
       "      <td>0.785666</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>f4f02e1d5f60b8f322a4a8a622dd1c1e</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>9.957781</td>\n",
       "      <td>1.586009</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>6f595f4f43b323804c04d4cea49c169b</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>10.061742</td>\n",
       "      <td>1.071287</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3d7b7af16d71a36f3b935f69e113e22d</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>10.306909</td>\n",
       "      <td>1.700775</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>3f1ca1759261598081fa3bb2f32fe0ac</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>10.554444</td>\n",
       "      <td>1.043901</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>3b4311d41fcaab7307235ea23b6d4599</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>10.756703</td>\n",
       "      <td>1.141506</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>74065ebc11c81bed6a9819d026c7cd84</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>10.917164</td>\n",
       "      <td>0.849209</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>b7ac5f7fcf9c8959626befe263a9d561</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>11.255320</td>\n",
       "      <td>0.989273</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>b0d0420255c6117045f8254bf8f377a0</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>11.478760</td>\n",
       "      <td>0.883154</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                hash  \\\n",
       "19  8363309e454e72993f86f10c7fc7c137   \n",
       "2   7c7932114268832a5458acfecfb453fc   \n",
       "8   25dcd8bb095f87a1ffc499fa6a83ef5d   \n",
       "4   e928929f89156d88ef49e28abaf55847   \n",
       "0   f4f02e1d5f60b8f322a4a8a622dd1c1e   \n",
       "18  6f595f4f43b323804c04d4cea49c169b   \n",
       "1   3d7b7af16d71a36f3b935f69e113e22d   \n",
       "9   3f1ca1759261598081fa3bb2f32fe0ac   \n",
       "5   3b4311d41fcaab7307235ea23b6d4599   \n",
       "6   74065ebc11c81bed6a9819d026c7cd84   \n",
       "3   b7ac5f7fcf9c8959626befe263a9d561   \n",
       "7   b0d0420255c6117045f8254bf8f377a0   \n",
       "\n",
       "                                             pipeline  SMAPE_mean  \\\n",
       "19  Pipeline(model = LinearPerSegmentModel(fit_int...    8.556535   \n",
       "2   Pipeline(model = LinearPerSegmentModel(fit_int...    9.210183   \n",
       "8   Pipeline(model = LinearPerSegmentModel(fit_int...    9.943658   \n",
       "4   Pipeline(model = LinearPerSegmentModel(fit_int...    9.946866   \n",
       "0   Pipeline(model = LinearPerSegmentModel(fit_int...    9.957781   \n",
       "18  Pipeline(model = LinearPerSegmentModel(fit_int...   10.061742   \n",
       "1   Pipeline(model = LinearPerSegmentModel(fit_int...   10.306909   \n",
       "9   Pipeline(model = LinearPerSegmentModel(fit_int...   10.554444   \n",
       "5   Pipeline(model = LinearPerSegmentModel(fit_int...   10.756703   \n",
       "6   Pipeline(model = LinearPerSegmentModel(fit_int...   10.917164   \n",
       "3   Pipeline(model = LinearPerSegmentModel(fit_int...   11.255320   \n",
       "7   Pipeline(model = LinearPerSegmentModel(fit_int...   11.478760   \n",
       "\n",
       "    elapsed_time  state  \n",
       "19      1.131200      1  \n",
       "2       1.096877      1  \n",
       "8       0.875966      1  \n",
       "4       0.785666      1  \n",
       "0       1.586009      1  \n",
       "18      1.071287      1  \n",
       "1       1.700775      1  \n",
       "9       1.043901      1  \n",
       "5       1.141506      1  \n",
       "6       0.849209      1  \n",
       "3       0.989273      1  \n",
       "7       0.883154      1  "
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tune.summary()[[\"hash\", \"pipeline\", \"SMAPE_mean\", \"elapsed_time\", \"state\"]].sort_values(\"SMAPE_mean\").drop_duplicates(\n",
    "    subset=\"hash\"\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "19c46c87",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "The second method `top_k` is useful when you want to check out best tried pipelines without duplicates."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "2f77e3d5",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "top_3_pipelines = tune.top_k(k=3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "7e39db49",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Pipeline(model = LinearPerSegmentModel(fit_intercept = True, ), transforms = [LagTransform(in_column = 'target', lags = [14, 15, 16, 17, 18, 19, 20, 21, 22, 23], out_column = 'target_lag', ), DateFlagsTransform(day_number_in_week = False, day_number_in_month = True, day_number_in_year = False, week_number_in_month = True, week_number_in_year = False, month_number_in_year = False, season_number = False, year_number = False, is_weekend = True, special_days_in_week = (), special_days_in_month = (), out_column = 'date_flags', in_column = None, )], horizon = 14, ),\n",
       " Pipeline(model = LinearPerSegmentModel(fit_intercept = True, ), transforms = [LagTransform(in_column = 'target', lags = [14, 15, 16, 17, 18, 19, 20, 21, 22, 23], out_column = 'target_lag', ), DateFlagsTransform(day_number_in_week = False, day_number_in_month = True, day_number_in_year = False, week_number_in_month = True, week_number_in_year = False, month_number_in_year = False, season_number = False, year_number = False, is_weekend = False, special_days_in_week = (), special_days_in_month = (), out_column = 'date_flags', in_column = None, )], horizon = 14, ),\n",
       " Pipeline(model = LinearPerSegmentModel(fit_intercept = False, ), transforms = [LagTransform(in_column = 'target', lags = [14, 15, 16, 17, 18, 19, 20, 21, 22, 23], out_column = 'target_lag', ), DateFlagsTransform(day_number_in_week = True, day_number_in_month = False, day_number_in_year = True, week_number_in_month = False, week_number_in_year = False, month_number_in_year = False, season_number = False, year_number = True, is_weekend = False, special_days_in_week = (), special_days_in_month = (), out_column = 'date_flags', in_column = None, )], horizon = 14, )]"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "top_3_pipelines"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "69161b23",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## 2. General AutoML <a class=\"anchor\" id=\"chapter_2\"></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "07081c2c",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "Hyperparameters tuning is useful, but can be too narrow. In this section we move our attention to general AutoML pipeline.\n",
    "In ETNA we have an `etna.auto.Auto` class for making automatic pipeline selection. It can be useful to quickly create a good baseline for your forecasting task."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5dfe064b",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "### 2.1 How `Auto` works <a class=\"anchor\" id=\"section_2_1\"></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ad5c1149",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "`Auto` init has similar parameters to `Tune`, but instead of `pipeline` it works with `pool`. Pool, in general, is just a list of pipelines.\n",
    "\n",
    "During `fit` there are two stages:\n",
    "\n",
    "- pool stage,\n",
    "- tuning stage.\n",
    "\n",
    "Pool stage is responsible for checking every pipeline suggested in a given `pool`. For each pipeline we run a backtest and compute `target_metric`. Results are saved in optuna study.\n",
    "\n",
    "Tuning stage takes `tune_size` best pipelines according to the results of the pool stage. And then runs `Tune` with default `params_to_tune` for them sequentially from best to the worst.\n",
    "\n",
    "Limit parameters `n_trials` and `timeout` are shared between pool and tuning stages. First, we run pool stage with given `n_trials` and `timeout`. After that, the remaining values are divided equally among `tune_size` tuning steps."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e339e19e",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "### 2.2 Example <a class=\"anchor\" id=\"section_2_2\"></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0d8d046e",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "We will move straight to the example."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d1e4202e",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "We have prepared pools for different frequencies and with different duration time. Frequency of our dataset is `D`, so we can choose `D_super_fast` pool. See docs of `Pool` class to know how to choose prepared pools correctly."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "dc2c6c0f",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "from etna.auto import Auto\n",
    "from etna.auto import Pool\n",
    "\n",
    "auto = Auto(\n",
    "    target_metric=SMAPE(), horizon=HORIZON, pool=Pool.D_super_fast, generate_params={}, backtest_params=dict(n_folds=1)\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b5297e51",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "Let's have a closer look at `generate_params`. It helps to set dynamic parameters for pipelines. Default pools contain `timestamp_column`, `chronos_device` and `timesfm_device` parameters.\n",
    "\n",
    "Parameter `timestamp_column` can be met in some models and transform (e.g. `statsforecast` models or `DateFlagTransform`) and defines column with timestamps. We use such columns in misaligned datasets (see `307-working_with_misaligned_data` notebook).\n",
    "\n",
    "- If you have regular `TSDataset` with `freq` != `None`, set `generate_params={}` or `generate_params=None`\n",
    "- If you have `TSDataset` with int index and exogenous column (e.g.`external_timestamp`) with timestamps, set `generate_params={\"timestamp_column\": \"external_timestamp\"}`\n",
    "\n",
    "Parameters `chronos_device` and `timesfm_device` specify `device` parameters for corresponding pretrained model. By default `Chronos`-like models use `device=\"auto\"` and `TimesFMModel` use `device=\"gpu\"`. As all models are initialized together, to reduce workload on gpu you can change device for one model type setting `generate_params={\"timesfm_device\": \"cpu\"}`"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ea98123c",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "After initialization `Auto` you can see pipelines which will be fitted."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "89ab5f4a",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[{'model': {'lag': 1, '_target_': 'etna.models.naive.NaiveModel'},\n",
       "  'transforms': [],\n",
       "  'horizon': 14,\n",
       "  '_target_': 'etna.pipeline.pipeline.Pipeline'},\n",
       " {'model': {'lag': 7, '_target_': 'etna.models.naive.NaiveModel'},\n",
       "  'transforms': [],\n",
       "  'horizon': 14,\n",
       "  '_target_': 'etna.pipeline.pipeline.Pipeline'},\n",
       " {'model': {'lag': 14, '_target_': 'etna.models.naive.NaiveModel'},\n",
       "  'transforms': [],\n",
       "  'horizon': 14,\n",
       "  '_target_': 'etna.pipeline.pipeline.Pipeline'},\n",
       " {'model': {'window': 3,\n",
       "   'seasonality': 7,\n",
       "   '_target_': 'etna.models.seasonal_ma.SeasonalMovingAverageModel'},\n",
       "  'transforms': [{'in_column': 'target',\n",
       "    'strategy': 'forward_fill',\n",
       "    'window': -1,\n",
       "    'seasonality': 1,\n",
       "    'constant_value': 0,\n",
       "    '_target_': 'etna.transforms.missing_values.imputation.TimeSeriesImputerTransform'},\n",
       "   {'in_column': 'target',\n",
       "    'window_size': 21,\n",
       "    'distance_coef': 3,\n",
       "    'n_neighbors': 3,\n",
       "    'distance_func': 'absolute_difference',\n",
       "    '_target_': 'etna.transforms.outliers.point_outliers.DensityOutliersTransform'},\n",
       "   {'in_column': 'target',\n",
       "    'strategy': 'mean',\n",
       "    'window': -1,\n",
       "    'seasonality': 1,\n",
       "    'constant_value': 0,\n",
       "    '_target_': 'etna.transforms.missing_values.imputation.TimeSeriesImputerTransform'}],\n",
       "  'horizon': 14,\n",
       "  '_target_': 'etna.pipeline.pipeline.Pipeline'},\n",
       " {'model': {'window': 3,\n",
       "   'seasonality': <SeasonalityMode.month: 'month'>,\n",
       "   '_target_': 'etna.models.deadline_ma.DeadlineMovingAverageModel'},\n",
       "  'transforms': [{'in_column': 'target',\n",
       "    'strategy': 'forward_fill',\n",
       "    'window': -1,\n",
       "    'seasonality': 1,\n",
       "    'constant_value': 0,\n",
       "    '_target_': 'etna.transforms.missing_values.imputation.TimeSeriesImputerTransform'},\n",
       "   {'in_column': 'target',\n",
       "    'window_size': 21,\n",
       "    'distance_coef': 3,\n",
       "    'n_neighbors': 3,\n",
       "    'distance_func': 'absolute_difference',\n",
       "    '_target_': 'etna.transforms.outliers.point_outliers.DensityOutliersTransform'},\n",
       "   {'in_column': 'target',\n",
       "    'strategy': 'mean',\n",
       "    'window': -1,\n",
       "    'seasonality': 1,\n",
       "    'constant_value': 0,\n",
       "    '_target_': 'etna.transforms.missing_values.imputation.TimeSeriesImputerTransform'}],\n",
       "  'horizon': 14,\n",
       "  '_target_': 'etna.pipeline.pipeline.Pipeline'},\n",
       " {'pipelines': [{'model': {'window': 1,\n",
       "     'seasonality': 14,\n",
       "     '_target_': 'etna.models.seasonal_ma.SeasonalMovingAverageModel'},\n",
       "    'transforms': [],\n",
       "    'horizon': 14,\n",
       "    '_target_': 'etna.pipeline.pipeline.Pipeline'},\n",
       "   {'model': {'window': 2,\n",
       "     'seasonality': 7,\n",
       "     '_target_': 'etna.models.seasonal_ma.SeasonalMovingAverageModel'},\n",
       "    'transforms': [],\n",
       "    'horizon': 14,\n",
       "    '_target_': 'etna.pipeline.pipeline.Pipeline'},\n",
       "   {'model': {'window': 7,\n",
       "     'seasonality': 7,\n",
       "     '_target_': 'etna.models.seasonal_ma.SeasonalMovingAverageModel'},\n",
       "    'transforms': [],\n",
       "    'horizon': 14,\n",
       "    '_target_': 'etna.pipeline.pipeline.Pipeline'}],\n",
       "  'regressor': {'_target_': 'sklearn.ensemble._forest.RandomForestRegressor',\n",
       "   'bootstrap': True,\n",
       "   'ccp_alpha': 0.0,\n",
       "   'criterion': 'squared_error',\n",
       "   'max_depth': None,\n",
       "   'max_features': 1.0,\n",
       "   'max_leaf_nodes': None,\n",
       "   'max_samples': None,\n",
       "   'min_impurity_decrease': 0.0,\n",
       "   'min_samples_leaf': 1,\n",
       "   'min_samples_split': 2,\n",
       "   'min_weight_fraction_leaf': 0.0,\n",
       "   'monotonic_cst': None,\n",
       "   'n_estimators': 5,\n",
       "   'n_jobs': None,\n",
       "   'oob_score': False,\n",
       "   'random_state': None,\n",
       "   'verbose': 0,\n",
       "   'warm_start': False},\n",
       "  'n_folds': 3,\n",
       "  'n_jobs': 1,\n",
       "  'joblib_params': {'verbose': 11,\n",
       "   'backend': 'multiprocessing',\n",
       "   'mmap_mode': 'c'},\n",
       "  '_target_': 'etna.ensembles.voting_ensemble.VotingEnsemble'},\n",
       " {'model': {'path_or_url': 'http://etna-github-prod.cdn-tinkoff.ru/chronos/chronos-bolt-tiny.zip',\n",
       "   'encoder_length': 2048,\n",
       "   'device': 'auto',\n",
       "   'dtype': 'float32',\n",
       "   'limit_prediction_length': False,\n",
       "   'batch_size': 128,\n",
       "   'cache_dir': '/Users/e.a.baturin/.etna/chronos-models/chronos-bolt',\n",
       "   '_target_': 'etna.models.nn.chronos.chronos_bolt.ChronosBoltModel'},\n",
       "  'transforms': [{'in_column': 'target',\n",
       "    'strategy': 'mean',\n",
       "    'window': -1,\n",
       "    'seasonality': 1,\n",
       "    'constant_value': 0,\n",
       "    '_target_': 'etna.transforms.missing_values.imputation.TimeSeriesImputerTransform'}],\n",
       "  'horizon': 14,\n",
       "  '_target_': 'etna.pipeline.pipeline.Pipeline'},\n",
       " {'model': {'path_or_url': 'http://etna-github-prod.cdn-tinkoff.ru/chronos/chronos-bolt-mini.zip',\n",
       "   'encoder_length': 2048,\n",
       "   'device': 'auto',\n",
       "   'dtype': 'float32',\n",
       "   'limit_prediction_length': False,\n",
       "   'batch_size': 128,\n",
       "   'cache_dir': '/Users/e.a.baturin/.etna/chronos-models/chronos-bolt',\n",
       "   '_target_': 'etna.models.nn.chronos.chronos_bolt.ChronosBoltModel'},\n",
       "  'transforms': [{'in_column': 'target',\n",
       "    'strategy': 'mean',\n",
       "    'window': -1,\n",
       "    'seasonality': 1,\n",
       "    'constant_value': 0,\n",
       "    '_target_': 'etna.transforms.missing_values.imputation.TimeSeriesImputerTransform'}],\n",
       "  'horizon': 14,\n",
       "  '_target_': 'etna.pipeline.pipeline.Pipeline'},\n",
       " {'model': {'path_or_url': 'http://etna-github-prod.cdn-tinkoff.ru/chronos/chronos-bolt-small.zip',\n",
       "   'encoder_length': 2048,\n",
       "   'device': 'auto',\n",
       "   'dtype': 'float32',\n",
       "   'limit_prediction_length': False,\n",
       "   'batch_size': 128,\n",
       "   'cache_dir': '/Users/e.a.baturin/.etna/chronos-models/chronos-bolt',\n",
       "   '_target_': 'etna.models.nn.chronos.chronos_bolt.ChronosBoltModel'},\n",
       "  'transforms': [{'in_column': 'target',\n",
       "    'strategy': 'mean',\n",
       "    'window': -1,\n",
       "    'seasonality': 1,\n",
       "    'constant_value': 0,\n",
       "    '_target_': 'etna.transforms.missing_values.imputation.TimeSeriesImputerTransform'}],\n",
       "  'horizon': 14,\n",
       "  '_target_': 'etna.pipeline.pipeline.Pipeline'}]"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "auto.get_configs()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ae89f2e6",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "Let's start the fitting. We can start by running only pool stage."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "f2214a9e",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Passing a tuple of `past_key_values` is deprecated and will be removed in Transformers v4.48.0. You should pass an instance of `EncoderDecoderCache` instead, e.g. `past_key_values=EncoderDecoderCache.from_legacy_cache(past_key_values)`.\n"
     ]
    }
   ],
   "source": [
    "%%capture\n",
    "best_pool_pipeline = auto.fit(ts=ts, tune_size=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "1fd516e0",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "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>hash</th>\n",
       "      <th>pipeline</th>\n",
       "      <th>SMAPE_mean</th>\n",
       "      <th>elapsed_time</th>\n",
       "      <th>state</th>\n",
       "      <th>study</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>5545ff3ef3cb3b2fc879b100afd9ca30</td>\n",
       "      <td>Pipeline(model = ChronosBoltModel(path_or_url ...</td>\n",
       "      <td>7.897397</td>\n",
       "      <td>0.446171</td>\n",
       "      <td>1</td>\n",
       "      <td>pool</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>d8215d95e2c6c9a4b4fdacf3fa77dddc</td>\n",
       "      <td>Pipeline(model = NaiveModel(lag = 7, ), transf...</td>\n",
       "      <td>8.266340</td>\n",
       "      <td>0.035717</td>\n",
       "      <td>1</td>\n",
       "      <td>pool</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>8b532534025945b07dedac76ad1e6207</td>\n",
       "      <td>Pipeline(model = ChronosBoltModel(path_or_url ...</td>\n",
       "      <td>8.512619</td>\n",
       "      <td>0.419310</td>\n",
       "      <td>1</td>\n",
       "      <td>pool</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>07fbe9c7aeaf9f56701630c5897db4f2</td>\n",
       "      <td>Pipeline(model = SeasonalMovingAverageModel(wi...</td>\n",
       "      <td>9.134800</td>\n",
       "      <td>0.120682</td>\n",
       "      <td>1</td>\n",
       "      <td>pool</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>add1079869e9d79f001e8c2df6c64279</td>\n",
       "      <td>VotingEnsemble(pipelines = [Pipeline(model = S...</td>\n",
       "      <td>9.351593</td>\n",
       "      <td>0.075331</td>\n",
       "      <td>1</td>\n",
       "      <td>pool</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>7899a017183d4dc042a70d5b38e7ed96</td>\n",
       "      <td>Pipeline(model = ChronosBoltModel(path_or_url ...</td>\n",
       "      <td>9.369298</td>\n",
       "      <td>1.402410</td>\n",
       "      <td>1</td>\n",
       "      <td>pool</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>8b47488c737903536374d375f7c66f40</td>\n",
       "      <td>Pipeline(model = NaiveModel(lag = 14, ), trans...</td>\n",
       "      <td>9.580548</td>\n",
       "      <td>0.046769</td>\n",
       "      <td>1</td>\n",
       "      <td>pool</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>de05e6c98895bc7d9c54fefe8dedea7b</td>\n",
       "      <td>Pipeline(model = DeadlineMovingAverageModel(wi...</td>\n",
       "      <td>14.151098</td>\n",
       "      <td>0.142992</td>\n",
       "      <td>1</td>\n",
       "      <td>pool</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>53e90ae4cf7f1f71e6396107549c25ef</td>\n",
       "      <td>Pipeline(model = NaiveModel(lag = 1, ), transf...</td>\n",
       "      <td>22.155640</td>\n",
       "      <td>0.033679</td>\n",
       "      <td>1</td>\n",
       "      <td>pool</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                               hash  \\\n",
       "8  5545ff3ef3cb3b2fc879b100afd9ca30   \n",
       "3  d8215d95e2c6c9a4b4fdacf3fa77dddc   \n",
       "6  8b532534025945b07dedac76ad1e6207   \n",
       "0  07fbe9c7aeaf9f56701630c5897db4f2   \n",
       "7  add1079869e9d79f001e8c2df6c64279   \n",
       "2  7899a017183d4dc042a70d5b38e7ed96   \n",
       "1  8b47488c737903536374d375f7c66f40   \n",
       "4  de05e6c98895bc7d9c54fefe8dedea7b   \n",
       "5  53e90ae4cf7f1f71e6396107549c25ef   \n",
       "\n",
       "                                            pipeline  SMAPE_mean  \\\n",
       "8  Pipeline(model = ChronosBoltModel(path_or_url ...    7.897397   \n",
       "3  Pipeline(model = NaiveModel(lag = 7, ), transf...    8.266340   \n",
       "6  Pipeline(model = ChronosBoltModel(path_or_url ...    8.512619   \n",
       "0  Pipeline(model = SeasonalMovingAverageModel(wi...    9.134800   \n",
       "7  VotingEnsemble(pipelines = [Pipeline(model = S...    9.351593   \n",
       "2  Pipeline(model = ChronosBoltModel(path_or_url ...    9.369298   \n",
       "1  Pipeline(model = NaiveModel(lag = 14, ), trans...    9.580548   \n",
       "4  Pipeline(model = DeadlineMovingAverageModel(wi...   14.151098   \n",
       "5  Pipeline(model = NaiveModel(lag = 1, ), transf...   22.155640   \n",
       "\n",
       "   elapsed_time  state study  \n",
       "8      0.446171      1  pool  \n",
       "3      0.035717      1  pool  \n",
       "6      0.419310      1  pool  \n",
       "0      0.120682      1  pool  \n",
       "7      0.075331      1  pool  \n",
       "2      1.402410      1  pool  \n",
       "1      0.046769      1  pool  \n",
       "4      0.142992      1  pool  \n",
       "5      0.033679      1  pool  "
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "auto.summary()[[\"hash\", \"pipeline\", \"SMAPE_mean\", \"elapsed_time\", \"state\", \"study\"]].sort_values(\"SMAPE_mean\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "5fadbb8b",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "%%capture\n",
    "best_pool_metrics = best_pool_pipeline.backtest(ts=full_ts, metrics=[SMAPE()], n_folds=5)[\"metrics\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "161471e0",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best pool SMAPE: 5.691\n"
     ]
    }
   ],
   "source": [
    "best_pool_smape = best_pool_metrics[\"SMAPE\"].mean()\n",
    "print(f\"Best pool SMAPE: {best_pool_smape:.3f}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4496dea3",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "### 2.3 Using custom pipeline pool <a class=\"anchor\" id=\"section_2_3\"></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4fa5bd32",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "We can define our own set of pipelines for the search."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "9c7a153c",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "custom_pool = [\n",
    "    Pipeline(model=NaiveModel(lag=1), transforms=(), horizon=HORIZON),\n",
    "    Pipeline(\n",
    "        model=LinearPerSegmentModel(),\n",
    "        transforms=[LagTransform(in_column=\"target\", lags=list(range(HORIZON, 2 * HORIZON)), out_column=\"target_lag\")],\n",
    "        horizon=HORIZON,\n",
    "    ),\n",
    "    Pipeline(\n",
    "        model=ProphetModel(),\n",
    "        transforms=[],\n",
    "        horizon=HORIZON,\n",
    "    ),\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "15c0ed98",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "%%capture\n",
    "custom_auto = Auto(\n",
    "    target_metric=SMAPE(),\n",
    "    horizon=HORIZON,\n",
    "    pool=custom_pool,\n",
    "    backtest_params=dict(n_folds=1),\n",
    "    storage=\"sqlite:///etna-auto-custom.db\",\n",
    ")\n",
    "best_custom_pool_pipeline = custom_auto.fit(ts=ts, tune_size=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "fe5977f8",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "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>hash</th>\n",
       "      <th>pipeline</th>\n",
       "      <th>SMAPE_mean</th>\n",
       "      <th>elapsed_time</th>\n",
       "      <th>state</th>\n",
       "      <th>study</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>6e2eb71d033b6d0607f5b6d0a7596ce9</td>\n",
       "      <td>Pipeline(model = ProphetModel(growth = 'linear...</td>\n",
       "      <td>7.784737</td>\n",
       "      <td>0.652359</td>\n",
       "      <td>1</td>\n",
       "      <td>pool</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>d4b50dc4c1b7debb0355ebfbd9c39ffb</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>8.587004</td>\n",
       "      <td>0.156382</td>\n",
       "      <td>1</td>\n",
       "      <td>pool</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>53e90ae4cf7f1f71e6396107549c25ef</td>\n",
       "      <td>Pipeline(model = NaiveModel(lag = 1, ), transf...</td>\n",
       "      <td>22.155640</td>\n",
       "      <td>0.034337</td>\n",
       "      <td>1</td>\n",
       "      <td>pool</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                               hash  \\\n",
       "1  6e2eb71d033b6d0607f5b6d0a7596ce9   \n",
       "0  d4b50dc4c1b7debb0355ebfbd9c39ffb   \n",
       "2  53e90ae4cf7f1f71e6396107549c25ef   \n",
       "\n",
       "                                            pipeline  SMAPE_mean  \\\n",
       "1  Pipeline(model = ProphetModel(growth = 'linear...    7.784737   \n",
       "0  Pipeline(model = LinearPerSegmentModel(fit_int...    8.587004   \n",
       "2  Pipeline(model = NaiveModel(lag = 1, ), transf...   22.155640   \n",
       "\n",
       "   elapsed_time  state study  \n",
       "1      0.652359      1  pool  \n",
       "0      0.156382      1  pool  \n",
       "2      0.034337      1  pool  "
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "custom_auto.summary()[[\"hash\", \"pipeline\", \"SMAPE_mean\", \"elapsed_time\", \"state\", \"study\"]].sort_values(\"SMAPE_mean\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e51b46e3",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "We can continue our training. The pool stage is over and there will be only the tuning stage. If we don't want to wait forever we should limit the tuning by fixing `n_trials` or `timeout`. \n",
    "\n",
    "We also set some parameters for `optuna.Study.optimize`: \n",
    "\n",
    "- `gc_after_trial=True`: to prevent `fit` from increasing memory consumption\n",
    "- `catch=(Exception,)`: to prevent failing if some trials are erroneous."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "80b27cda",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "%%capture\n",
    "best_custom_tuning_pipeline = custom_auto.fit(ts=ts, tune_size=2, n_trials=10, gc_after_trial=True, catch=(Exception,))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a9c8a6cf",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "Let's look at the results."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "1dda3b16",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "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>hash</th>\n",
       "      <th>pipeline</th>\n",
       "      <th>SMAPE_mean</th>\n",
       "      <th>elapsed_time</th>\n",
       "      <th>state</th>\n",
       "      <th>study</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0da73fb28ad659adcd67e73d6a12ffcb</td>\n",
       "      <td>Pipeline(model = ProphetModel(growth = 'linear...</td>\n",
       "      <td>6.350313</td>\n",
       "      <td>0.681777</td>\n",
       "      <td>1</td>\n",
       "      <td>tuning/6e2eb71d033b6d0607f5b6d0a7596ce9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>06056568d43cc286f099fd767f9c4e01</td>\n",
       "      <td>Pipeline(model = ProphetModel(growth = 'linear...</td>\n",
       "      <td>6.827116</td>\n",
       "      <td>0.946632</td>\n",
       "      <td>1</td>\n",
       "      <td>tuning/6e2eb71d033b6d0607f5b6d0a7596ce9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>6e2eb71d033b6d0607f5b6d0a7596ce9</td>\n",
       "      <td>Pipeline(model = ProphetModel(growth = 'linear...</td>\n",
       "      <td>7.784737</td>\n",
       "      <td>0.652359</td>\n",
       "      <td>1</td>\n",
       "      <td>pool</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>d4b50dc4c1b7debb0355ebfbd9c39ffb</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>8.587004</td>\n",
       "      <td>0.156382</td>\n",
       "      <td>1</td>\n",
       "      <td>pool</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>d4b50dc4c1b7debb0355ebfbd9c39ffb</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>8.587004</td>\n",
       "      <td>0.181814</td>\n",
       "      <td>1</td>\n",
       "      <td>tuning/d4b50dc4c1b7debb0355ebfbd9c39ffb</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>8eacbbe63deeee69548eb62734a428e1</td>\n",
       "      <td>Pipeline(model = ProphetModel(growth = 'linear...</td>\n",
       "      <td>8.728627</td>\n",
       "      <td>1.918130</td>\n",
       "      <td>1</td>\n",
       "      <td>tuning/6e2eb71d033b6d0607f5b6d0a7596ce9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>3a875591627f904e0d2f5b633ec986b1</td>\n",
       "      <td>Pipeline(model = LinearPerSegmentModel(fit_int...</td>\n",
       "      <td>9.145437</td>\n",
       "      <td>0.122133</td>\n",
       "      <td>1</td>\n",
       "      <td>tuning/d4b50dc4c1b7debb0355ebfbd9c39ffb</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>53e90ae4cf7f1f71e6396107549c25ef</td>\n",
       "      <td>Pipeline(model = NaiveModel(lag = 1, ), transf...</td>\n",
       "      <td>22.155640</td>\n",
       "      <td>0.034337</td>\n",
       "      <td>1</td>\n",
       "      <td>pool</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                               hash  \\\n",
       "8  0da73fb28ad659adcd67e73d6a12ffcb   \n",
       "7  06056568d43cc286f099fd767f9c4e01   \n",
       "1  6e2eb71d033b6d0607f5b6d0a7596ce9   \n",
       "0  d4b50dc4c1b7debb0355ebfbd9c39ffb   \n",
       "4  d4b50dc4c1b7debb0355ebfbd9c39ffb   \n",
       "9  8eacbbe63deeee69548eb62734a428e1   \n",
       "5  3a875591627f904e0d2f5b633ec986b1   \n",
       "2  53e90ae4cf7f1f71e6396107549c25ef   \n",
       "\n",
       "                                            pipeline  SMAPE_mean  \\\n",
       "8  Pipeline(model = ProphetModel(growth = 'linear...    6.350313   \n",
       "7  Pipeline(model = ProphetModel(growth = 'linear...    6.827116   \n",
       "1  Pipeline(model = ProphetModel(growth = 'linear...    7.784737   \n",
       "0  Pipeline(model = LinearPerSegmentModel(fit_int...    8.587004   \n",
       "4  Pipeline(model = LinearPerSegmentModel(fit_int...    8.587004   \n",
       "9  Pipeline(model = ProphetModel(growth = 'linear...    8.728627   \n",
       "5  Pipeline(model = LinearPerSegmentModel(fit_int...    9.145437   \n",
       "2  Pipeline(model = NaiveModel(lag = 1, ), transf...   22.155640   \n",
       "\n",
       "   elapsed_time  state                                    study  \n",
       "8      0.681777      1  tuning/6e2eb71d033b6d0607f5b6d0a7596ce9  \n",
       "7      0.946632      1  tuning/6e2eb71d033b6d0607f5b6d0a7596ce9  \n",
       "1      0.652359      1                                     pool  \n",
       "0      0.156382      1                                     pool  \n",
       "4      0.181814      1  tuning/d4b50dc4c1b7debb0355ebfbd9c39ffb  \n",
       "9      1.918130      1  tuning/6e2eb71d033b6d0607f5b6d0a7596ce9  \n",
       "5      0.122133      1  tuning/d4b50dc4c1b7debb0355ebfbd9c39ffb  \n",
       "2      0.034337      1                                     pool  "
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "custom_auto.summary()[[\"hash\", \"pipeline\", \"SMAPE_mean\", \"elapsed_time\", \"state\", \"study\"]].sort_values(\n",
    "    \"SMAPE_mean\"\n",
    ").drop_duplicates(subset=(\"hash\", \"study\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "b1d4a0d4",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Pipeline(model = ProphetModel(growth = 'linear', changepoints = None, n_changepoints = 25, changepoint_range = 0.9187587557123996, yearly_seasonality = 'auto', weekly_seasonality = 'auto', daily_seasonality = 'auto', holidays = None, seasonality_mode = 'multiplicative', seasonality_prior_scale = 7.780155576901417, holidays_prior_scale = 0.3860866271460545, changepoint_prior_scale = 0.01083670267174957, mcmc_samples = 0, interval_width = 0.8, uncertainty_samples = 1000, stan_backend = None, additional_seasonality_params = (), timestamp_column = None, ), transforms = [], horizon = 14, ),\n",
       " Pipeline(model = ProphetModel(growth = 'linear', changepoints = None, n_changepoints = 25, changepoint_range = 0.8635482199008357, yearly_seasonality = 'auto', weekly_seasonality = 'auto', daily_seasonality = 'auto', holidays = None, seasonality_mode = 'multiplicative', seasonality_prior_scale = 0.6431172050131991, holidays_prior_scale = 0.8663279761354559, changepoint_prior_scale = 0.029554483012804774, mcmc_samples = 0, interval_width = 0.8, uncertainty_samples = 1000, stan_backend = None, additional_seasonality_params = (), timestamp_column = None, ), transforms = [], horizon = 14, ),\n",
       " Pipeline(model = ProphetModel(growth = 'linear', changepoints = None, n_changepoints = 25, changepoint_range = 0.8, yearly_seasonality = 'auto', weekly_seasonality = 'auto', daily_seasonality = 'auto', holidays = None, seasonality_mode = 'additive', seasonality_prior_scale = 10.0, holidays_prior_scale = 10.0, changepoint_prior_scale = 0.05, mcmc_samples = 0, interval_width = 0.8, uncertainty_samples = 1000, stan_backend = None, additional_seasonality_params = (), timestamp_column = None, ), transforms = [], horizon = 14, ),\n",
       " Pipeline(model = LinearPerSegmentModel(fit_intercept = True, ), transforms = [LagTransform(in_column = 'target', lags = [14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27], out_column = 'target_lag', )], horizon = 14, ),\n",
       " Pipeline(model = ProphetModel(growth = 'linear', changepoints = None, n_changepoints = 25, changepoint_range = 0.8030327596160489, yearly_seasonality = 'auto', weekly_seasonality = 'auto', daily_seasonality = 'auto', holidays = None, seasonality_mode = 'multiplicative', seasonality_prior_scale = 0.0163345876110695, holidays_prior_scale = 3.146730406166005, changepoint_prior_scale = 0.001718538897359816, mcmc_samples = 0, interval_width = 0.8, uncertainty_samples = 1000, stan_backend = None, additional_seasonality_params = (), timestamp_column = None, ), transforms = [], horizon = 14, )]"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "custom_auto.top_k(k=5)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "488a749b",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "If we look at `study` column we will see that best trial from tuning stage is better then best trial from pool stage. It means, that tuning stage was successful and improved the final result. \n",
    "\n",
    "Let's compare best pipeline on pool and tuning stages on hold-out part of initial `ts`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "5b5b5d22",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "%%capture\n",
    "best_custom_pool_metrics = best_custom_pool_pipeline.backtest(ts=full_ts, metrics=[SMAPE()], n_folds=5)[\"metrics\"]\n",
    "best_custom_tuning_metrics = best_custom_tuning_pipeline.backtest(ts=full_ts, metrics=[SMAPE()], n_folds=5)[\"metrics\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "ae1f4609",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best pool SMAPE: 7.796\n",
      "Best tuning SMAPE: 7.712\n"
     ]
    }
   ],
   "source": [
    "best_custom_pool_smape = best_custom_pool_metrics[\"SMAPE\"].mean()\n",
    "best_custom_tuning_smape = best_custom_tuning_metrics[\"SMAPE\"].mean()\n",
    "print(f\"Best pool SMAPE: {best_custom_pool_smape:.3f}\")\n",
    "print(f\"Best tuning SMAPE: {best_custom_tuning_smape:.3f}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d490acff",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "As we can see, the results after the tuning stage are a little bit better."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3322d9c2",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## 3. Summary <a class=\"anchor\" id=\"chapter_3\"></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "39b4a081",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "In this notebook we discussed how AutoML works in ETNA library and how to use it. There are two supported scenarios:\n",
    "\n",
    "- Tuning your existing pipeline;\n",
    "- Automatic search of the pipeline for your forecasting task."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.16"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}