{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Transforming Target\n",
    "One of the common requirements is to transform the target variable itself. \n",
    "hcrystalball implements ``TargetTransformer`` transformer for such a cases, that internally scales input data and return un-scaled again."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "plt.style.use('seaborn')\n",
    "plt.rcParams['figure.figsize'] = [12, 6]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from hcrystalball.utils import get_sales_data\n",
    "\n",
    "df = get_sales_data(n_dates=100, \n",
    "                    n_assortments=1, \n",
    "                    n_states=1, \n",
    "                    n_stores=1)\n",
    "X, y = pd.DataFrame(index=df.index), df['Sales']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from hcrystalball.wrappers import get_sklearn_wrapper\n",
    "from hcrystalball.compose import TSColumnTransformer\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.ensemble import RandomForestRegressor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from hcrystalball.preprocessing import TargetTransformer"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## With Wrapper"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "rf_model = get_sklearn_wrapper(RandomForestRegressor)\n",
    "scaled = TargetTransformer(rf_model, StandardScaler())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "(scaled.fit(X[:-10], y[:-10])\n",
    "       .predict(X[-10:])\n",
    "       .merge(y, left_index=True, right_index=True, how='outer')\n",
    "       .tail(50)\n",
    "       .plot()\n",
    ");"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## With Pipeline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "X['trend'] = np.arange(len(X))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# add standard scaler to the trend and let the date column pass in raw form\n",
    "preprocessing = TSColumnTransformer(\n",
    "    transformers=[        \n",
    "        ('scaler', StandardScaler(), ['trend'])\n",
    "    ])    \n",
    "# define random forest model\n",
    "rf_model = get_sklearn_wrapper(RandomForestRegressor)\n",
    "# glue it together\n",
    "sklearn_model_pipeline = Pipeline([\n",
    "    ('preprocessing', preprocessing), \n",
    "    ('model', rf_model)\n",
    "]) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "scaled_pipeline = TargetTransformer(sklearn_model_pipeline, StandardScaler())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "(scaled_pipeline.fit(X[:-10], y[:-10])\n",
    "                .predict(X[-10:])\n",
    "                .merge(y, left_index=True, right_index=True, how='outer')\n",
    "                .tail(50)\n",
    "                .plot()\n",
    ");"
   ]
  }
 ],
 "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": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}