examples/paranoids.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Welcome to Paranoia\n",
    "\n",
    "Paranoia is a lovely planet located 1240.12 million light-years away from Earth.\n",
    "This is how Paranoia looks from the space:\n",
    "\n",
    "![Paranoia planet](assets/paranoia.jpg)\n",
    "\n",
    "Beautiful, isn't it?\n",
    "\n",
    "I've been there taking notes about the creatures that inhabit that country: **paranoids**.\n",
    "In this notebook I will show how I've been able to prepare the paranoids data to be processed later by machine learning models that predict who will be the winner of a supposed battle. To perform that processing I will use, obviously, **Recipipe** :)\n",
    "\n",
    "Through this tutorial, I will describe the most basic Recipipe concepts."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "\n",
    "import recipipe as r"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Read data\n",
    "\n",
    "Paranoids are normally peaceful and very colorful creatures. I don't have any pictures of them because I forgot to bring a camera. Can you believe it?! Wormholes are fast but not pleasant enough to go back for a camera and come back...\n",
    "\n",
    "I didn't forget paper and pen, so I was able to collect the data now contained in `paranoids.csv`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>name</th>\n",
       "      <th>type 1</th>\n",
       "      <th>type 2</th>\n",
       "      <th>type 3</th>\n",
       "      <th>width</th>\n",
       "      <th>height</th>\n",
       "      <th>length</th>\n",
       "      <th>weight</th>\n",
       "      <th>cute</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Scucioid</td>\n",
       "      <td>Ground</td>\n",
       "      <td>Flying</td>\n",
       "      <td>NaN</td>\n",
       "      <td>11.04000</td>\n",
       "      <td>10.11470</td>\n",
       "      <td>14.21440</td>\n",
       "      <td>843.7770</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Trelvisoid</td>\n",
       "      <td>Ground</td>\n",
       "      <td>Flying</td>\n",
       "      <td>Water</td>\n",
       "      <td>13.56950</td>\n",
       "      <td>7.07998</td>\n",
       "      <td>13.21440</td>\n",
       "      <td>1175.0500</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Dovanoid</td>\n",
       "      <td>Flying</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>20.23680</td>\n",
       "      <td>13.98820</td>\n",
       "      <td>27.60250</td>\n",
       "      <td>6691.8200</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Pilinoid</td>\n",
       "      <td>Water</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>9.66392</td>\n",
       "      <td>5.85685</td>\n",
       "      <td>13.53060</td>\n",
       "      <td>775.8360</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Arkanoid</td>\n",
       "      <td>Fire</td>\n",
       "      <td>Flying</td>\n",
       "      <td>NaN</td>\n",
       "      <td>6.35188</td>\n",
       "      <td>8.05839</td>\n",
       "      <td>7.03025</td>\n",
       "      <td>340.1620</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Pinkfloid</td>\n",
       "      <td>Water</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>7.55713</td>\n",
       "      <td>9.75967</td>\n",
       "      <td>10.94330</td>\n",
       "      <td>770.9640</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Redathoid</td>\n",
       "      <td>Fire</td>\n",
       "      <td>Ground</td>\n",
       "      <td>NaN</td>\n",
       "      <td>5.68892</td>\n",
       "      <td>7.86451</td>\n",
       "      <td>7.22494</td>\n",
       "      <td>462.4400</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Nebidoid</td>\n",
       "      <td>Ground</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>10.55570</td>\n",
       "      <td>5.92844</td>\n",
       "      <td>15.17950</td>\n",
       "      <td>698.2830</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Golukoid</td>\n",
       "      <td>Flying</td>\n",
       "      <td>Fire</td>\n",
       "      <td>NaN</td>\n",
       "      <td>9.28180</td>\n",
       "      <td>10.01900</td>\n",
       "      <td>10.58730</td>\n",
       "      <td>1208.4000</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Oid</td>\n",
       "      <td>Water</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>13.49640</td>\n",
       "      <td>8.88992</td>\n",
       "      <td>7.03398</td>\n",
       "      <td>685.8780</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>Dickydoid</td>\n",
       "      <td>Flying</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>9.94806</td>\n",
       "      <td>12.27760</td>\n",
       "      <td>13.21970</td>\n",
       "      <td>1400.6800</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>Bugoid</td>\n",
       "      <td>Flying</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>15.87650</td>\n",
       "      <td>21.23760</td>\n",
       "      <td>14.14620</td>\n",
       "      <td>2925.1100</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>Uglioid</td>\n",
       "      <td>Flying</td>\n",
       "      <td>Ground</td>\n",
       "      <td>Water</td>\n",
       "      <td>15.38910</td>\n",
       "      <td>11.15110</td>\n",
       "      <td>8.78153</td>\n",
       "      <td>800.9100</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>Viscuoid</td>\n",
       "      <td>Flying</td>\n",
       "      <td>Ground</td>\n",
       "      <td>NaN</td>\n",
       "      <td>13.92040</td>\n",
       "      <td>19.69970</td>\n",
       "      <td>9.85045</td>\n",
       "      <td>1807.0000</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>Odriloid</td>\n",
       "      <td>Electric</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>12.51670</td>\n",
       "      <td>6.27300</td>\n",
       "      <td>6.92060</td>\n",
       "      <td>319.5720</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>Raxoid</td>\n",
       "      <td>Flying</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>23.03890</td>\n",
       "      <td>23.79350</td>\n",
       "      <td>21.17430</td>\n",
       "      <td>9991.2700</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>Kikkoid</td>\n",
       "      <td>Fire</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>8.97374</td>\n",
       "      <td>13.27590</td>\n",
       "      <td>4.63856</td>\n",
       "      <td>623.9520</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>Thormolipirisodifoloid</td>\n",
       "      <td>Flying</td>\n",
       "      <td>Electric</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3.25330</td>\n",
       "      <td>2.39910</td>\n",
       "      <td>3.06284</td>\n",
       "      <td>34.6145</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>Qurikoid</td>\n",
       "      <td>Water</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>18.19000</td>\n",
       "      <td>10.53380</td>\n",
       "      <td>17.32540</td>\n",
       "      <td>3601.3000</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                      name    type 1    type 2 type 3     width    height  \\\n",
       "0                 Scucioid    Ground    Flying    NaN  11.04000  10.11470   \n",
       "1               Trelvisoid    Ground    Flying  Water  13.56950   7.07998   \n",
       "2                 Dovanoid    Flying       NaN    NaN  20.23680  13.98820   \n",
       "3                 Pilinoid     Water       NaN    NaN   9.66392   5.85685   \n",
       "4                 Arkanoid      Fire    Flying    NaN   6.35188   8.05839   \n",
       "5                Pinkfloid     Water       NaN    NaN   7.55713   9.75967   \n",
       "6                Redathoid      Fire    Ground    NaN   5.68892   7.86451   \n",
       "7                 Nebidoid    Ground       NaN    NaN  10.55570   5.92844   \n",
       "8                 Golukoid    Flying      Fire    NaN   9.28180  10.01900   \n",
       "9                      Oid     Water       NaN    NaN  13.49640   8.88992   \n",
       "10               Dickydoid    Flying       NaN    NaN   9.94806  12.27760   \n",
       "11                  Bugoid    Flying       NaN    NaN  15.87650  21.23760   \n",
       "12                 Uglioid    Flying    Ground  Water  15.38910  11.15110   \n",
       "13                Viscuoid    Flying    Ground    NaN  13.92040  19.69970   \n",
       "14                Odriloid  Electric       NaN    NaN  12.51670   6.27300   \n",
       "15                  Raxoid    Flying       NaN    NaN  23.03890  23.79350   \n",
       "16                 Kikkoid      Fire       NaN    NaN   8.97374  13.27590   \n",
       "17  Thormolipirisodifoloid    Flying  Electric    NaN   3.25330   2.39910   \n",
       "18                Qurikoid     Water       NaN    NaN  18.19000  10.53380   \n",
       "\n",
       "      length     weight   cute  \n",
       "0   14.21440   843.7770  False  \n",
       "1   13.21440  1175.0500  False  \n",
       "2   27.60250  6691.8200  False  \n",
       "3   13.53060   775.8360   True  \n",
       "4    7.03025   340.1620  False  \n",
       "5   10.94330   770.9640   True  \n",
       "6    7.22494   462.4400  False  \n",
       "7   15.17950   698.2830   True  \n",
       "8   10.58730  1208.4000   True  \n",
       "9    7.03398   685.8780   True  \n",
       "10  13.21970  1400.6800   True  \n",
       "11  14.14620  2925.1100  False  \n",
       "12   8.78153   800.9100  False  \n",
       "13   9.85045  1807.0000   True  \n",
       "14   6.92060   319.5720  False  \n",
       "15  21.17430  9991.2700  False  \n",
       "16   4.63856   623.9520   True  \n",
       "17   3.06284    34.6145  False  \n",
       "18  17.32540  3601.3000   True  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv(\"data/paranoids.csv\")\n",
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The column names are pretty descriptive.\n",
    "All the lenght measures are in meters and the weight is in kilograms.\n",
    "Perhaps the column that needs more explanation is `cute`.\n",
    "It simply indicates whether or not I found the paranoid physically beautiful ^.^\n",
    "\n",
    "As you can observe, there is **no column for the gender**. I'm not going to go into the disgusting details of paranoid reproduction, just to say that it's not an important feature."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Introduction to transformers\n",
    "\n",
    "The basic unit used in Recipipe is the transformer.\n",
    "\n",
    "All Recipipe transformers inherit from `recipipe.core.RecipipeTransformer`.\n",
    "At the same time, that class inherits from `sklearn.base.TransformerMixin` and `sklearn.base.BaseEstimator`, that is, a Recipipe transformer is an SKLearn transformer.\n",
    "The biggest difference is that Recipipe transformers only work with `pandas.DataFrames`.\n",
    "No worries, SKLearn transformers are fully supported using wrappers.\n",
    "We will talk about them later.\n",
    "\n",
    "Let's start creating the simplest transformer, a transformer that selects columns from the input."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SelectTransformer(col_format='{}', cols_init=[], cols_not_found_error=False,\n",
       "                  dtype=None, exclude=[], keep_original=False, name=None)"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "r.select()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To apply that transformation to our DataFrame:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>name</th>\n",
       "      <th>type 1</th>\n",
       "      <th>type 2</th>\n",
       "      <th>type 3</th>\n",
       "      <th>width</th>\n",
       "      <th>height</th>\n",
       "      <th>length</th>\n",
       "      <th>weight</th>\n",
       "      <th>cute</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Scucioid</td>\n",
       "      <td>Ground</td>\n",
       "      <td>Flying</td>\n",
       "      <td>NaN</td>\n",
       "      <td>11.04000</td>\n",
       "      <td>10.11470</td>\n",
       "      <td>14.21440</td>\n",
       "      <td>843.777</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Trelvisoid</td>\n",
       "      <td>Ground</td>\n",
       "      <td>Flying</td>\n",
       "      <td>Water</td>\n",
       "      <td>13.56950</td>\n",
       "      <td>7.07998</td>\n",
       "      <td>13.21440</td>\n",
       "      <td>1175.050</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Dovanoid</td>\n",
       "      <td>Flying</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>20.23680</td>\n",
       "      <td>13.98820</td>\n",
       "      <td>27.60250</td>\n",
       "      <td>6691.820</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Pilinoid</td>\n",
       "      <td>Water</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>9.66392</td>\n",
       "      <td>5.85685</td>\n",
       "      <td>13.53060</td>\n",
       "      <td>775.836</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Arkanoid</td>\n",
       "      <td>Fire</td>\n",
       "      <td>Flying</td>\n",
       "      <td>NaN</td>\n",
       "      <td>6.35188</td>\n",
       "      <td>8.05839</td>\n",
       "      <td>7.03025</td>\n",
       "      <td>340.162</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         name  type 1  type 2 type 3     width    height    length    weight  \\\n",
       "0    Scucioid  Ground  Flying    NaN  11.04000  10.11470  14.21440   843.777   \n",
       "1  Trelvisoid  Ground  Flying  Water  13.56950   7.07998  13.21440  1175.050   \n",
       "2    Dovanoid  Flying     NaN    NaN  20.23680  13.98820  27.60250  6691.820   \n",
       "3    Pilinoid   Water     NaN    NaN   9.66392   5.85685  13.53060   775.836   \n",
       "4    Arkanoid    Fire  Flying    NaN   6.35188   8.05839   7.03025   340.162   \n",
       "\n",
       "    cute  \n",
       "0  False  \n",
       "1  False  \n",
       "2  False  \n",
       "3   True  \n",
       "4  False  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "r.select().fit_transform(df).head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Almost all the Recipipe transformers are expecting a list of columns to which apply the transformation.\n",
    "In our previous example, we didn't specify any columns, so the transformer is applied over all the columns.\n",
    "Selecting all the columns give us the input DataFrame, so nothing exciting.\n",
    "\n",
    "Let's make something more interesting. Let's select a couple of columns."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>name</th>\n",
       "      <th>cute</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Scucioid</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Trelvisoid</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Dovanoid</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Pilinoid</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Arkanoid</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         name   cute\n",
       "0    Scucioid  False\n",
       "1  Trelvisoid  False\n",
       "2    Dovanoid  False\n",
       "3    Pilinoid   True\n",
       "4    Arkanoid  False"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "r.select(\"name\", \"cute\").fit_transform(df).head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Note that after this operation `df` is still intact since we are not assigning the result, just printing it.\n",
    "\n",
    "If you like to keep a list of features you can use that **list as a parameter** too."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>name</th>\n",
       "      <th>width</th>\n",
       "      <th>height</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Scucioid</td>\n",
       "      <td>11.04000</td>\n",
       "      <td>10.11470</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Trelvisoid</td>\n",
       "      <td>13.56950</td>\n",
       "      <td>7.07998</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Dovanoid</td>\n",
       "      <td>20.23680</td>\n",
       "      <td>13.98820</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Pilinoid</td>\n",
       "      <td>9.66392</td>\n",
       "      <td>5.85685</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Arkanoid</td>\n",
       "      <td>6.35188</td>\n",
       "      <td>8.05839</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         name     width    height\n",
       "0    Scucioid  11.04000  10.11470\n",
       "1  Trelvisoid  13.56950   7.07998\n",
       "2    Dovanoid  20.23680  13.98820\n",
       "3    Pilinoid   9.66392   5.85685\n",
       "4    Arkanoid   6.35188   8.05839"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "features_str = [\"name\"]\n",
    "features_float = [\"width\", \"height\"]\n",
    "r.select(features_str, features_float).fit_transform(df).head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's do something more interesting, let's select the three type columns (type 1, type 2 and type 3) but without using a list, we'll use a wildcard."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>type 1</th>\n",
       "      <th>type 2</th>\n",
       "      <th>type 3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Ground</td>\n",
       "      <td>Flying</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Ground</td>\n",
       "      <td>Flying</td>\n",
       "      <td>Water</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Flying</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Water</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Fire</td>\n",
       "      <td>Flying</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   type 1  type 2 type 3\n",
       "0  Ground  Flying    NaN\n",
       "1  Ground  Flying  Water\n",
       "2  Flying     NaN    NaN\n",
       "3   Water     NaN    NaN\n",
       "4    Fire  Flying    NaN"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "r.select(\"type *\").fit_transform(df).head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "At fit time, the Recipipe transformer match the given list of columns to the columns present on the given DataFrame.\n",
    "That feature is available in all the existing Recipipe transformers.\n",
    "\n",
    "Also, all Recipipe transformers have `dtype` as parameter."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>width</th>\n",
       "      <th>height</th>\n",
       "      <th>length</th>\n",
       "      <th>weight</th>\n",
       "      <th>cute</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>11.04000</td>\n",
       "      <td>10.11470</td>\n",
       "      <td>14.21440</td>\n",
       "      <td>843.777</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>13.56950</td>\n",
       "      <td>7.07998</td>\n",
       "      <td>13.21440</td>\n",
       "      <td>1175.050</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>20.23680</td>\n",
       "      <td>13.98820</td>\n",
       "      <td>27.60250</td>\n",
       "      <td>6691.820</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>9.66392</td>\n",
       "      <td>5.85685</td>\n",
       "      <td>13.53060</td>\n",
       "      <td>775.836</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>6.35188</td>\n",
       "      <td>8.05839</td>\n",
       "      <td>7.03025</td>\n",
       "      <td>340.162</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      width    height    length    weight   cute\n",
       "0  11.04000  10.11470  14.21440   843.777  False\n",
       "1  13.56950   7.07998  13.21440  1175.050  False\n",
       "2  20.23680  13.98820  27.60250  6691.820  False\n",
       "3   9.66392   5.85685  13.53060   775.836   True\n",
       "4   6.35188   8.05839   7.03025   340.162  False"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "r.select(dtype=[bool, float]).fit_transform(df).head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Recipipe uses a lot of alias to make the creation of transformers more convenient.\n",
    "The `recipipe.select` method is defined as an alias for the `recipipe.transformers.SelectTransformer` class.\n",
    "It is far easier to type `recipipe.select` than `recipipe.SelectTransformer`.\n",
    "Using `r` as an alias of `recipipe`, as we are using in this notebook, makes things even easier.\n",
    "All the alias definitions are in `recipipe/__init__.py` (list with all the aliases in the official documentation: https://guiferviz.com/recipipe/recipipe.html#recipipe-aliases)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Create pipeline\n",
    "\n",
    "If we want to chain one transformation after the other and apply them all at the same time, we need to create a `recipipe` object.\n",
    "This is nothing but a super-vitaminized SKLearn pipeline.\n",
    "For example, in contrast to SKLearn pipelines, `recipipe` objects can be created without any transformation at all.\n",
    "\n",
    "Although we can also create a `recipipe` from a list of transformations, let's go step by step, adding one transformation at a time. So let's create an empty pipeline."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "pipe = r.recipipe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Recipipe pipelines work with `pandas.DataFrame`s as an input object.\n",
    "In the same way, all the transformers in the Recipipe pipeline should receive a `pandas.DataFrame`, that's why we need to use Recipipe transformers on it.\n",
    "\n",
    "---\n",
    "\n",
    "Although reading the names I've given to the Paranoids is not a waste of time, unfortunately I doubt that our machine learning model will be able to enjoy my originality, so the sooner we delete unnecessary data the better performance we will have in our pipeline. Let's consider that whatever I find a paranoid `cute` or not is also very uninformative for our future model.\n",
    "\n",
    "We will use the `recipipe.drop` transformer to drop columns."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "pipe += r.drop(\"name\", \"cute\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Another valid method for adding steps to the pipeline is `pipe.add(your_transformer_here)`.\n",
    "\n",
    "---\n",
    "\n",
    "Each paranoid has one or more types (3 types as maximum).\n",
    "Existing types are: Ground, Water, Flying, Electric and Fire.\n",
    "Even if we know the types in advance, let's use a onehot encoder to find them all automatically.\n",
    "\n",
    "Under the hood, the onehot encoder that Recipipe uses is `sklearn.preprocessing.OneHotEncoder`.\n",
    "Recipipe is only a wrapper over it.\n",
    "Indeed the definition of `recipipe.onehot` is:\n",
    "\n",
    "    onehot = from_sklearn(OneHotEncoder(sparse=False, handle_unknown=\"ignore\"))\n",
    "\n",
    "`recipipe.onehot` is a creator object, when you call it you get a Recipipe transformer that wraps the `sklearn.preprocessing.OneHotEncoder` passed as an argument to `from_sklearn`.\n",
    "\n",
    "When you use `recipipe.onehot()` a *columns wrapper* over the onehot encoder is created (equivalent to `recipipe.onehot(wrapper=\"columns\")`).\n",
    "A columns wrapper fits the SKLearn transformer in all the selected columns and transform all the selected columns at the same time. That's not going to work in this example where we want to transform all the \"type \\*\" columns using the same category values.\n",
    "\n",
    "If we treat each column as a separate column we will get different encodings, specially because almost no paranoid has 3 types. Look to the unique values of each type column:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "type 1 : ['Ground' 'Flying' 'Water' 'Fire' 'Electric']\n",
      "type 2 : ['Flying' nan 'Ground' 'Fire' 'Electric']\n",
      "type 3 : [nan 'Water']\n"
     ]
    }
   ],
   "source": [
    "for i in [\"type 1\", \"type 2\", \"type 3\"]:\n",
    "    print(i, \":\", df[i].unique())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The first thing we need to do is get rid of the `nan` values. For that task we can use `recipipe.impute` that is defined as:\n",
    "\n",
    "    impute = from_sklearn(SimpleImputer(strategy=\"constant\"))\n",
    "\n",
    "Later we need to fit a onehot encoder in a long column created from the concatenation of all the \"type \\*\" columns.\n",
    "However, at the time of applying that transformation, it should be applied one \"type \\*\" column at a time.\n",
    "To do that we have the `fit_one_col` wrapper."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "pipe += r.impute(\"type *\", fill_value=\"Unknown\")  # as strategy=\"constant\" is used by default we only need\n",
    "                                                  # to specify fill_value.\n",
    "pipe += r.onehot(\"type *\", wrapper=\"fit_one_col\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Notice that if you have another kind of column (non-related to types) that you want to encode using onehot, you cannot reuse this transformer or all the categories will be mixed. You need to add a new `r.onehot` to the pipeline.\n",
    "\n",
    "---\n",
    "\n",
    "Let's do more complicate operations.\n",
    "Observe the results of the transformation pipeline we have so far."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>type 1=Electric</th>\n",
       "      <th>type 1=Fire</th>\n",
       "      <th>type 1=Flying</th>\n",
       "      <th>type 1=Ground</th>\n",
       "      <th>type 1=Unknown</th>\n",
       "      <th>type 1=Water</th>\n",
       "      <th>type 2=Electric</th>\n",
       "      <th>type 2=Fire</th>\n",
       "      <th>type 2=Flying</th>\n",
       "      <th>type 2=Ground</th>\n",
       "      <th>...</th>\n",
       "      <th>type 3=Electric</th>\n",
       "      <th>type 3=Fire</th>\n",
       "      <th>type 3=Flying</th>\n",
       "      <th>type 3=Ground</th>\n",
       "      <th>type 3=Unknown</th>\n",
       "      <th>type 3=Water</th>\n",
       "      <th>width</th>\n",
       "      <th>height</th>\n",
       "      <th>length</th>\n",
       "      <th>weight</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11.04000</td>\n",
       "      <td>10.11470</td>\n",
       "      <td>14.21440</td>\n",
       "      <td>843.777</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>13.56950</td>\n",
       "      <td>7.07998</td>\n",
       "      <td>13.21440</td>\n",
       "      <td>1175.050</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>20.23680</td>\n",
       "      <td>13.98820</td>\n",
       "      <td>27.60250</td>\n",
       "      <td>6691.820</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>9.66392</td>\n",
       "      <td>5.85685</td>\n",
       "      <td>13.53060</td>\n",
       "      <td>775.836</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.35188</td>\n",
       "      <td>8.05839</td>\n",
       "      <td>7.03025</td>\n",
       "      <td>340.162</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 22 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   type 1=Electric  type 1=Fire  type 1=Flying  type 1=Ground  type 1=Unknown  \\\n",
       "0              0.0          0.0            0.0            1.0             0.0   \n",
       "1              0.0          0.0            0.0            1.0             0.0   \n",
       "2              0.0          0.0            1.0            0.0             0.0   \n",
       "3              0.0          0.0            0.0            0.0             0.0   \n",
       "4              0.0          1.0            0.0            0.0             0.0   \n",
       "\n",
       "   type 1=Water  type 2=Electric  type 2=Fire  type 2=Flying  type 2=Ground  \\\n",
       "0           0.0              0.0          0.0            1.0            0.0   \n",
       "1           0.0              0.0          0.0            1.0            0.0   \n",
       "2           0.0              0.0          0.0            0.0            0.0   \n",
       "3           1.0              0.0          0.0            0.0            0.0   \n",
       "4           0.0              0.0          0.0            1.0            0.0   \n",
       "\n",
       "   ...  type 3=Electric  type 3=Fire  type 3=Flying  type 3=Ground  \\\n",
       "0  ...              0.0          0.0            0.0            0.0   \n",
       "1  ...              0.0          0.0            0.0            0.0   \n",
       "2  ...              0.0          0.0            0.0            0.0   \n",
       "3  ...              0.0          0.0            0.0            0.0   \n",
       "4  ...              0.0          0.0            0.0            0.0   \n",
       "\n",
       "   type 3=Unknown  type 3=Water     width    height    length    weight  \n",
       "0             1.0           0.0  11.04000  10.11470  14.21440   843.777  \n",
       "1             0.0           1.0  13.56950   7.07998  13.21440  1175.050  \n",
       "2             1.0           0.0  20.23680  13.98820  27.60250  6691.820  \n",
       "3             1.0           0.0   9.66392   5.85685  13.53060   775.836  \n",
       "4             1.0           0.0   6.35188   8.05839   7.03025   340.162  \n",
       "\n",
       "[5 rows x 22 columns]"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pipe.fit_transform(df).head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "First I don't want to keep the \"Unknown\" colums.\n",
    "Indeed they are not really unknown, it's a kind of indicator if a paranoid has a \"type \\*\" or not."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "pipe += r.drop(\"*=Unknown\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Later, I want to have only one \"type=*\\\" column with a 1 if the paranoid has that type among the three type columns and a 0 otherwise.\n",
    "For example, if one row has \"type 1=Fire\"=0, \"type 2=Fire\"=1 and \"type 3=Fire\"=0, then I want merge all those three columns into one \"type=Fire\"=1.\n",
    "I want to repeat the same process with all the types.\n",
    "\n",
    "We can use the `recipipe.sum` transformer for this task. This transformer automatically removes any number from the input names and use that as an output name, so it's enough with:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "pipe += r.sum(\"type 1=*\", \"type 2=*\", \"type 3=*\", groups=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We need to specify the `groups=True` to sum the three groups of columns (\"type 1=\\*\", \"type 2=\\*\" and \"type 3=\\*\") one column per group at a time and not to sum all the columns together.\n",
    "That is, we want to sum \"type 1=Electric\", \"type 2=Electric\" and \"type 3=Electric\", later \"type 1=Fire\", \"type 2=Fire\" and \"type 3=Fire\" and so on.\n",
    "We want to store the results in `\"type=Electric\"`, `\"type=Fire\"`... respectively.\n",
    "\n",
    "Let's see how our pipeline goes so far."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>type=Electric</th>\n",
       "      <th>type=Fire</th>\n",
       "      <th>type=Flying</th>\n",
       "      <th>type=Ground</th>\n",
       "      <th>type=Water</th>\n",
       "      <th>width</th>\n",
       "      <th>height</th>\n",
       "      <th>length</th>\n",
       "      <th>weight</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11.04000</td>\n",
       "      <td>10.11470</td>\n",
       "      <td>14.21440</td>\n",
       "      <td>843.777</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>13.56950</td>\n",
       "      <td>7.07998</td>\n",
       "      <td>13.21440</td>\n",
       "      <td>1175.050</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>20.23680</td>\n",
       "      <td>13.98820</td>\n",
       "      <td>27.60250</td>\n",
       "      <td>6691.820</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>9.66392</td>\n",
       "      <td>5.85685</td>\n",
       "      <td>13.53060</td>\n",
       "      <td>775.836</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.35188</td>\n",
       "      <td>8.05839</td>\n",
       "      <td>7.03025</td>\n",
       "      <td>340.162</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   type=Electric  type=Fire  type=Flying  type=Ground  type=Water     width  \\\n",
       "0            0.0        0.0          1.0          1.0         0.0  11.04000   \n",
       "1            0.0        0.0          1.0          1.0         1.0  13.56950   \n",
       "2            0.0        0.0          1.0          0.0         0.0  20.23680   \n",
       "3            0.0        0.0          0.0          0.0         1.0   9.66392   \n",
       "4            0.0        1.0          1.0          0.0         0.0   6.35188   \n",
       "\n",
       "     height    length    weight  \n",
       "0  10.11470  14.21440   843.777  \n",
       "1   7.07998  13.21440  1175.050  \n",
       "2  13.98820  27.60250  6691.820  \n",
       "3   5.85685  13.53060   775.836  \n",
       "4   8.05839   7.03025   340.162  "
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pipe.fit_transform(df).head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "\n",
    "Now I want to scale using mean and std the width, height, length and weight.\n",
    "But for some reason that I don't want to disclose here I also want to scale the weight using a MinMax scaler.\n",
    "\n",
    "Although the MinMax scaler is available in `recipipe.minmax` we will define it manually using the SKLearn transformer and the `recipipe.from_sklearn` creator object."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import MinMaxScaler\n",
    "\n",
    "minmax = r.from_sklearn(MinMaxScaler())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we can use `minmax` as any other Recipipe transformer.\n",
    "We are going to use two new parameters of all the Recipipe transformers: `keep_original` and `col_format`.\n",
    "\n",
    "We want to keep the original column, the non-scaled column. That's because we want to use a standar scaler later over that column.\n",
    "If we want to keep the original column we need to rename it.\n",
    "Specify an output name with `col_format`.\n",
    "Any '{}' in `col_format` will be replace by the original column name."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "pipe += minmax(\"weight\", keep_original=True, col_format=\"{}_minmax\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`recipipe.scaler` does the job for scaling using mean and stdev.\n",
    "You can use `col_format` without `keep_original`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "pipe += r.scale(\"width\", \"height\", \"length\", \"weight\", col_format=\"{}_norm\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>type=Electric</th>\n",
       "      <th>type=Fire</th>\n",
       "      <th>type=Flying</th>\n",
       "      <th>type=Ground</th>\n",
       "      <th>type=Water</th>\n",
       "      <th>width_norm</th>\n",
       "      <th>height_norm</th>\n",
       "      <th>length_norm</th>\n",
       "      <th>weight_norm</th>\n",
       "      <th>weight_minmax</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-0.200894</td>\n",
       "      <td>-0.156701</td>\n",
       "      <td>0.405524</td>\n",
       "      <td>-0.411887</td>\n",
       "      <td>0.081269</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.312981</td>\n",
       "      <td>-0.720646</td>\n",
       "      <td>0.231961</td>\n",
       "      <td>-0.276334</td>\n",
       "      <td>0.114540</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.667463</td>\n",
       "      <td>0.563115</td>\n",
       "      <td>2.729205</td>\n",
       "      <td>1.981071</td>\n",
       "      <td>0.668619</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>-0.480449</td>\n",
       "      <td>-0.947942</td>\n",
       "      <td>0.286842</td>\n",
       "      <td>-0.439688</td>\n",
       "      <td>0.074445</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.153300</td>\n",
       "      <td>-0.538827</td>\n",
       "      <td>-0.841379</td>\n",
       "      <td>-0.617961</td>\n",
       "      <td>0.030688</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   type=Electric  type=Fire  type=Flying  type=Ground  type=Water  width_norm  \\\n",
       "0            0.0        0.0          1.0          1.0         0.0   -0.200894   \n",
       "1            0.0        0.0          1.0          1.0         1.0    0.312981   \n",
       "2            0.0        0.0          1.0          0.0         0.0    1.667463   \n",
       "3            0.0        0.0          0.0          0.0         1.0   -0.480449   \n",
       "4            0.0        1.0          1.0          0.0         0.0   -1.153300   \n",
       "\n",
       "   height_norm  length_norm  weight_norm  weight_minmax  \n",
       "0    -0.156701     0.405524    -0.411887       0.081269  \n",
       "1    -0.720646     0.231961    -0.276334       0.114540  \n",
       "2     0.563115     2.729205     1.981071       0.668619  \n",
       "3    -0.947942     0.286842    -0.439688       0.074445  \n",
       "4    -0.538827    -0.841379    -0.617961       0.030688  "
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pipe.fit_transform(df).head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Apply transformations\n",
    "\n",
    "Let's simulate that we have a train and a test DataFrame."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "half = len(df) // 2\n",
    "df_train = df[:half]\n",
    "df_test = df[half:]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Fit and transform the pipeline again, but this time in our train DataFrame."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>type=Fire</th>\n",
       "      <th>type=Flying</th>\n",
       "      <th>type=Ground</th>\n",
       "      <th>type=Water</th>\n",
       "      <th>width_norm</th>\n",
       "      <th>height_norm</th>\n",
       "      <th>length_norm</th>\n",
       "      <th>weight_norm</th>\n",
       "      <th>weight_minmax</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.144645</td>\n",
       "      <td>0.569747</td>\n",
       "      <td>0.162215</td>\n",
       "      <td>-0.318262</td>\n",
       "      <td>0.079289</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.752826</td>\n",
       "      <td>-0.688989</td>\n",
       "      <td>-0.011537</td>\n",
       "      <td>-0.141651</td>\n",
       "      <td>0.131444</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.355881</td>\n",
       "      <td>2.176390</td>\n",
       "      <td>2.488416</td>\n",
       "      <td>2.799491</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>-0.186214</td>\n",
       "      <td>-1.196317</td>\n",
       "      <td>0.043403</td>\n",
       "      <td>-0.354483</td>\n",
       "      <td>0.068592</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-0.982546</td>\n",
       "      <td>-0.283166</td>\n",
       "      <td>-1.086042</td>\n",
       "      <td>-0.586753</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>-0.692761</td>\n",
       "      <td>0.422488</td>\n",
       "      <td>-0.406144</td>\n",
       "      <td>-0.357080</td>\n",
       "      <td>0.067825</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.141945</td>\n",
       "      <td>-0.363583</td>\n",
       "      <td>-1.052214</td>\n",
       "      <td>-0.521563</td>\n",
       "      <td>0.019251</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.028202</td>\n",
       "      <td>-1.166623</td>\n",
       "      <td>0.329902</td>\n",
       "      <td>-0.395829</td>\n",
       "      <td>0.056382</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-0.278089</td>\n",
       "      <td>0.530052</td>\n",
       "      <td>-0.467999</td>\n",
       "      <td>-0.123871</td>\n",
       "      <td>0.136695</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   type=Fire  type=Flying  type=Ground  type=Water  width_norm  height_norm  \\\n",
       "0        0.0          1.0          1.0         0.0    0.144645     0.569747   \n",
       "1        0.0          1.0          1.0         1.0    0.752826    -0.688989   \n",
       "2        0.0          1.0          0.0         0.0    2.355881     2.176390   \n",
       "3        0.0          0.0          0.0         1.0   -0.186214    -1.196317   \n",
       "4        1.0          1.0          0.0         0.0   -0.982546    -0.283166   \n",
       "5        0.0          0.0          0.0         1.0   -0.692761     0.422488   \n",
       "6        1.0          0.0          1.0         0.0   -1.141945    -0.363583   \n",
       "7        0.0          0.0          1.0         0.0    0.028202    -1.166623   \n",
       "8        1.0          1.0          0.0         0.0   -0.278089     0.530052   \n",
       "\n",
       "   length_norm  weight_norm  weight_minmax  \n",
       "0     0.162215    -0.318262       0.079289  \n",
       "1    -0.011537    -0.141651       0.131444  \n",
       "2     2.488416     2.799491       1.000000  \n",
       "3     0.043403    -0.354483       0.068592  \n",
       "4    -1.086042    -0.586753       0.000000  \n",
       "5    -0.406144    -0.357080       0.067825  \n",
       "6    -1.052214    -0.521563       0.019251  \n",
       "7     0.329902    -0.395829       0.056382  \n",
       "8    -0.467999    -0.123871       0.136695  "
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pipe.fit_transform(df_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Note that the Electic type is not present in the output DataFrame, that's because all the electric types are in the tests dataset.\n",
    "\n",
    "Transform the test dataset using the same transformations."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "scrolled": true
   },
   "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>type=Fire</th>\n",
       "      <th>type=Flying</th>\n",
       "      <th>type=Ground</th>\n",
       "      <th>type=Water</th>\n",
       "      <th>width_norm</th>\n",
       "      <th>height_norm</th>\n",
       "      <th>length_norm</th>\n",
       "      <th>weight_norm</th>\n",
       "      <th>weight_minmax</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.735250</td>\n",
       "      <td>0.061735</td>\n",
       "      <td>-1.085394</td>\n",
       "      <td>-0.402442</td>\n",
       "      <td>0.054429</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-0.117896</td>\n",
       "      <td>1.466870</td>\n",
       "      <td>-0.010616</td>\n",
       "      <td>-0.021361</td>\n",
       "      <td>0.166967</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.307511</td>\n",
       "      <td>5.183283</td>\n",
       "      <td>0.150365</td>\n",
       "      <td>0.791354</td>\n",
       "      <td>0.406972</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.190322</td>\n",
       "      <td>0.999623</td>\n",
       "      <td>-0.781754</td>\n",
       "      <td>-0.341115</td>\n",
       "      <td>0.072540</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.837195</td>\n",
       "      <td>4.545396</td>\n",
       "      <td>-0.596028</td>\n",
       "      <td>0.195259</td>\n",
       "      <td>0.230938</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.499696</td>\n",
       "      <td>-1.023707</td>\n",
       "      <td>-1.105094</td>\n",
       "      <td>-0.597730</td>\n",
       "      <td>-0.003242</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3.029605</td>\n",
       "      <td>6.243415</td>\n",
       "      <td>1.371507</td>\n",
       "      <td>4.558518</td>\n",
       "      <td>1.519463</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-0.352157</td>\n",
       "      <td>1.880944</td>\n",
       "      <td>-1.501602</td>\n",
       "      <td>-0.435456</td>\n",
       "      <td>0.044680</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.727554</td>\n",
       "      <td>-2.630516</td>\n",
       "      <td>-1.775385</td>\n",
       "      <td>-0.749648</td>\n",
       "      <td>-0.048105</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.863758</td>\n",
       "      <td>0.743580</td>\n",
       "      <td>0.702755</td>\n",
       "      <td>1.151850</td>\n",
       "      <td>0.513431</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    type=Fire  type=Flying  type=Ground  type=Water  width_norm  height_norm  \\\n",
       "9         0.0          0.0          0.0         1.0    0.735250     0.061735   \n",
       "10        0.0          1.0          0.0         0.0   -0.117896     1.466870   \n",
       "11        0.0          1.0          0.0         0.0    1.307511     5.183283   \n",
       "12        0.0          1.0          1.0         1.0    1.190322     0.999623   \n",
       "13        0.0          1.0          1.0         0.0    0.837195     4.545396   \n",
       "14        0.0          0.0          0.0         0.0    0.499696    -1.023707   \n",
       "15        0.0          1.0          0.0         0.0    3.029605     6.243415   \n",
       "16        1.0          0.0          0.0         0.0   -0.352157     1.880944   \n",
       "17        0.0          1.0          0.0         0.0   -1.727554    -2.630516   \n",
       "18        0.0          0.0          0.0         1.0    1.863758     0.743580   \n",
       "\n",
       "    length_norm  weight_norm  weight_minmax  \n",
       "9     -1.085394    -0.402442       0.054429  \n",
       "10    -0.010616    -0.021361       0.166967  \n",
       "11     0.150365     0.791354       0.406972  \n",
       "12    -0.781754    -0.341115       0.072540  \n",
       "13    -0.596028     0.195259       0.230938  \n",
       "14    -1.105094    -0.597730      -0.003242  \n",
       "15     1.371507     4.558518       1.519463  \n",
       "16    -1.501602    -0.435456       0.044680  \n",
       "17    -1.775385    -0.749648      -0.048105  \n",
       "18     0.702755     1.151850       0.513431  "
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pipe.transform(df_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Plotting transformations\n",
    "\n",
    "If you have SKLearn >= 0.23 you can easily plot a diagram of a pipeline."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>div.sk-top-container {color: black;background-color: white;}div.sk-toggleable {background-color: white;}label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.2em 0.3em;box-sizing: border-box;text-align: center;}div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}div.sk-estimator {font-family: monospace;background-color: #f0f8ff;margin: 0.25em 0.25em;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;}div.sk-estimator:hover {background-color: #d4ebff;}div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 2em;bottom: 0;left: 50%;}div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;}div.sk-item {z-index: 1;}div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;}div.sk-parallel-item {display: flex;flex-direction: column;position: relative;background-color: white;}div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}div.sk-parallel-item:only-child::after {width: 0;}div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0.2em;box-sizing: border-box;padding-bottom: 0.1em;background-color: white;position: relative;}div.sk-label label {font-family: monospace;font-weight: bold;background-color: white;display: inline-block;line-height: 1.2em;}div.sk-label-container {position: relative;z-index: 2;text-align: center;}div.sk-container {display: inline-block;position: relative;}</style><div class=\"sk-top-container\"><div class=\"sk-container\"><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"84ce2f3d-091d-4dad-90b8-9485f5bc808e\" type=\"checkbox\" ><label class=\"sk-toggleable__label\" for=\"84ce2f3d-091d-4dad-90b8-9485f5bc808e\">Recipipe</label><div class=\"sk-toggleable__content\"><pre>Recipipe(steps=[('step00',\n",
       "                 DropTransformer(col_format='{}', cols_init=['name', 'cute'],\n",
       "                                 cols_not_found_error=False, dtype=None,\n",
       "                                 exclude=[], keep_original=False, name=None)),\n",
       "                ('step01',\n",
       "                 SklearnColumnsWrapper(col_format='{}', cols_init=['type *'],\n",
       "                                       cols_not_found_error=False, dtype=None,\n",
       "                                       exclude=[], keep_original=False,\n",
       "                                       name=None,\n",
       "                                       sk_transformer=SimpleImputer(add_indic...\n",
       "                                       exclude=[], keep_original=True,\n",
       "                                       name=None,\n",
       "                                       sk_transformer=MinMaxScaler(copy=True,\n",
       "                                                                   feature_range=(0,\n",
       "                                                                                  1)))),\n",
       "                ('step06',\n",
       "                 SklearnColumnsWrapper(col_format='{}_norm',\n",
       "                                       cols_init=['width', 'height', 'length',\n",
       "                                                  'weight'],\n",
       "                                       cols_not_found_error=False, dtype=None,\n",
       "                                       exclude=[], keep_original=False,\n",
       "                                       name=None,\n",
       "                                       sk_transformer=StandardScaler(copy=True,\n",
       "                                                                     with_mean=True,\n",
       "                                                                     with_std=True)))])</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"70fd7f24-9c0b-4db7-97ee-feb9be931751\" type=\"checkbox\" ><label class=\"sk-toggleable__label\" for=\"70fd7f24-9c0b-4db7-97ee-feb9be931751\">DropTransformer</label><div class=\"sk-toggleable__content\"><pre>DropTransformer(col_format='{}', cols_init=['name', 'cute'],\n",
       "                cols_not_found_error=False, dtype=None, exclude=[],\n",
       "                keep_original=False, name=None)</pre></div></div></div><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"f28a93f9-8ea4-4303-bd76-c27cfb7a5db1\" type=\"checkbox\" ><label class=\"sk-toggleable__label\" for=\"f28a93f9-8ea4-4303-bd76-c27cfb7a5db1\">step01: SklearnColumnsWrapper</label><div class=\"sk-toggleable__content\"><pre>SklearnColumnsWrapper(col_format='{}', cols_init=['type *'],\n",
       "                      cols_not_found_error=False, dtype=None, exclude=[],\n",
       "                      keep_original=False, name=None,\n",
       "                      sk_transformer=SimpleImputer(add_indicator=False,\n",
       "                                                   copy=True,\n",
       "                                                   fill_value='Unknown',\n",
       "                                                   missing_values=nan,\n",
       "                                                   strategy='constant',\n",
       "                                                   verbose=0))</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"c68f3ef6-dbde-4b2e-820f-a051778fc74d\" type=\"checkbox\" ><label class=\"sk-toggleable__label\" for=\"c68f3ef6-dbde-4b2e-820f-a051778fc74d\">SimpleImputer</label><div class=\"sk-toggleable__content\"><pre>SimpleImputer(add_indicator=False, copy=True, fill_value='Unknown',\n",
       "              missing_values=nan, strategy='constant', verbose=0)</pre></div></div></div></div></div></div></div></div><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"153032d8-6ac9-4285-829a-d9b89a5c1828\" type=\"checkbox\" ><label class=\"sk-toggleable__label\" for=\"153032d8-6ac9-4285-829a-d9b89a5c1828\">step02: SklearnFitOneWrapper</label><div class=\"sk-toggleable__content\"><pre>SklearnFitOneWrapper(col_format='{}', cols_init=['type *'],\n",
       "                     cols_not_found_error=False, dtype=None, exclude=[],\n",
       "                     keep_original=False, name=None,\n",
       "                     sk_transformer=OneHotEncoder(categories='auto', drop=None,\n",
       "                                                  dtype=<class 'numpy.float64'>,\n",
       "                                                  handle_unknown='ignore',\n",
       "                                                  sparse=False))</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"5b79e9ff-a65a-4fa2-89ef-15acdc0b1c86\" type=\"checkbox\" ><label class=\"sk-toggleable__label\" for=\"5b79e9ff-a65a-4fa2-89ef-15acdc0b1c86\">OneHotEncoder</label><div class=\"sk-toggleable__content\"><pre>OneHotEncoder(categories='auto', drop=None, dtype=<class 'numpy.float64'>,\n",
       "              handle_unknown='ignore', sparse=False)</pre></div></div></div></div></div></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"066e16d5-188d-40d1-b72d-cc5e19f6344d\" type=\"checkbox\" ><label class=\"sk-toggleable__label\" for=\"066e16d5-188d-40d1-b72d-cc5e19f6344d\">DropTransformer</label><div class=\"sk-toggleable__content\"><pre>DropTransformer(col_format='{}', cols_init=['*=Unknown'],\n",
       "                cols_not_found_error=False, dtype=None, exclude=[],\n",
       "                keep_original=False, name=None)</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"d7e4f5cd-9652-411d-a555-183b7977dbe0\" type=\"checkbox\" ><label class=\"sk-toggleable__label\" for=\"d7e4f5cd-9652-411d-a555-183b7977dbe0\">SumTransformer</label><div class=\"sk-toggleable__content\"><pre>SumTransformer(col_format='{}', cols_init=['type 1=*', 'type 2=*', 'type 3=*'],\n",
       "               cols_not_found_error=False, dtype=None, exclude=[], groups=True,\n",
       "               keep_original=False, name=None)</pre></div></div></div><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"8c7b32e6-2c36-46ae-a8c5-aa6396479e7b\" type=\"checkbox\" ><label class=\"sk-toggleable__label\" for=\"8c7b32e6-2c36-46ae-a8c5-aa6396479e7b\">step05: SklearnColumnsWrapper</label><div class=\"sk-toggleable__content\"><pre>SklearnColumnsWrapper(col_format='{}_minmax', cols_init=['weight'],\n",
       "                      cols_not_found_error=False, dtype=None, exclude=[],\n",
       "                      keep_original=True, name=None,\n",
       "                      sk_transformer=MinMaxScaler(copy=True,\n",
       "                                                  feature_range=(0, 1)))</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"2982e243-8069-4424-8c4b-db6bbd92c11b\" type=\"checkbox\" ><label class=\"sk-toggleable__label\" for=\"2982e243-8069-4424-8c4b-db6bbd92c11b\">MinMaxScaler</label><div class=\"sk-toggleable__content\"><pre>MinMaxScaler(copy=True, feature_range=(0, 1))</pre></div></div></div></div></div></div></div></div><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"62345fb1-60ef-423c-8265-ccb741df6914\" type=\"checkbox\" ><label class=\"sk-toggleable__label\" for=\"62345fb1-60ef-423c-8265-ccb741df6914\">step06: SklearnColumnsWrapper</label><div class=\"sk-toggleable__content\"><pre>SklearnColumnsWrapper(col_format='{}_norm',\n",
       "                      cols_init=['width', 'height', 'length', 'weight'],\n",
       "                      cols_not_found_error=False, dtype=None, exclude=[],\n",
       "                      keep_original=False, name=None,\n",
       "                      sk_transformer=StandardScaler(copy=True, with_mean=True,\n",
       "                                                    with_std=True))</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"5034d807-0e2b-4f95-b9cb-7f461a5d2e60\" type=\"checkbox\" ><label class=\"sk-toggleable__label\" for=\"5034d807-0e2b-4f95-b9cb-7f461a5d2e60\">StandardScaler</label><div class=\"sk-toggleable__content\"><pre>StandardScaler(copy=True, with_mean=True, with_std=True)</pre></div></div></div></div></div></div></div></div></div></div></div></div>"
      ],
      "text/plain": [
       "Recipipe(steps=[('step00',\n",
       "                 DropTransformer(col_format='{}', cols_init=['name', 'cute'],\n",
       "                                 cols_not_found_error=False, dtype=None,\n",
       "                                 exclude=[], keep_original=False, name=None)),\n",
       "                ('step01',\n",
       "                 SklearnColumnsWrapper(col_format='{}', cols_init=['type *'],\n",
       "                                       cols_not_found_error=False, dtype=None,\n",
       "                                       exclude=[], keep_original=False,\n",
       "                                       name=None,\n",
       "                                       sk_transformer=SimpleImputer(add_indic...\n",
       "                                       exclude=[], keep_original=True,\n",
       "                                       name=None,\n",
       "                                       sk_transformer=MinMaxScaler(copy=True,\n",
       "                                                                   feature_range=(0,\n",
       "                                                                                  1)))),\n",
       "                ('step06',\n",
       "                 SklearnColumnsWrapper(col_format='{}_norm',\n",
       "                                       cols_init=['width', 'height', 'length',\n",
       "                                                  'weight'],\n",
       "                                       cols_not_found_error=False, dtype=None,\n",
       "                                       exclude=[], keep_original=False,\n",
       "                                       name=None,\n",
       "                                       sk_transformer=StandardScaler(copy=True,\n",
       "                                                                     with_mean=True,\n",
       "                                                                     with_std=True)))])"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from packaging import version\n",
    "import sklearn\n",
    "\n",
    "if version.parse(sklearn.__version__) >= version.parse('0.23'):\n",
    "    sklearn.set_config(display=\"diagram\")\n",
    "pipe"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Conclusions\n",
    "\n",
    "Recipipe is easy to use and easy to extend. It takes care of many of the annoying things about handling pipelines with SKLearn. This is just a silly example of what can be done with Recipipe, but there are thousands of other possibilities.\n",
    "\n",
    "Enjoy and keep safe from the Paranoids!\n",
    "\n",
    "\n",
    "## What's next?\n",
    "\n",
    "You can continue learning about Recipipe by:\n",
    "\n",
    "* Reading the [Titanic example](https://github.com/guiferviz/recipipe/blob/master/examples/titanic.ipynb).\n",
    "* Reading a list of [features with code](https://github.com/guiferviz/recipipe/blob/master/examples/features.ipynb).\n",
    "* Reading the [official documentation](https://guiferviz.com/recipipe/).\n",
    "* Reading the source code from the [GitHub repository](https://github.com/guiferviz/recipipe) :).\n",
    "\n",
    "\n",
    "## Resources\n",
    "\n",
    "* Planet generator: https://topps.diku.dk/torbenm/maps.msp.\n",
    "Yes, I've used a random planet generator... Paranoia is not real :(\n",
    "* Paranoia planet and paranoids: my insane brain, copyright ©"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}