{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "from sklearn.datasets import load_iris\n",
    "from sklearn.preprocessing import KBinsDiscretizer as skKBinsDiscretizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "class KBinsDiscretizer():\n",
    "    def __init__(self, n_bins=5, strategy=\"quantile\"):\n",
    "        self.n_bins = n_bins\n",
    "        self.strategy = strategy\n",
    "\n",
    "    def fit(self, X):\n",
    "        self.n_bins_ = np.full(X.shape[1], self.n_bins)\n",
    "        self.bin_edges_ = np.empty(X.shape[1], dtype=object)\n",
    "        for i in range(X.shape[1]):\n",
    "            if self.strategy == \"uniform\":\n",
    "                self.bin_edges_[i] = np.linspace(X[:, i].min(), X[:, i].max(),\n",
    "                                                 self.n_bins_[i] + 1)\n",
    "            elif self.strategy == \"quantile\":\n",
    "                quantiles = np.linspace(0, 100, self.n_bins_[i] + 1)\n",
    "                self.bin_edges_[i] = np.percentile(X[:, i], quantiles)\n",
    "        return self\n",
    "\n",
    "    def transform(self, X):\n",
    "        Xt = np.empty_like(X)\n",
    "        for i in range(X.shape[1]):\n",
    "            # similar to scikit-learn solution\n",
    "            Xt[:, i] = np.digitize(X[:, i] + np.finfo(float).eps, self.bin_edges_[i][1:])\n",
    "        Xt = np.clip(Xt, 0, self.n_bins_ - 1)\n",
    "        return Xt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "X, _ = load_iris(return_X_y=True)\n",
    "trans1 = KBinsDiscretizer(n_bins=5).fit(X)\n",
    "trans2 = skKBinsDiscretizer(n_bins=5, encode=\"ordinal\").fit(X)\n",
    "for i in range(X.shape[1]):\n",
    "    assert np.allclose(trans1.bin_edges_[i], trans2.bin_edges_[i])\n",
    "Xt1 = trans1.transform(X)\n",
    "Xt2 = trans2.transform(X)\n",
    "assert np.array_equal(Xt1, Xt2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "X, _ = load_iris(return_X_y=True)\n",
    "trans1 = KBinsDiscretizer(n_bins=5, strategy=\"uniform\").fit(X)\n",
    "trans2 = skKBinsDiscretizer(n_bins=5, encode=\"ordinal\", strategy=\"uniform\").fit(X)\n",
    "for i in range(X.shape[1]):\n",
    "    assert np.allclose(trans1.bin_edges_[i], trans2.bin_edges_[i])\n",
    "Xt1 = trans1.transform(X)\n",
    "Xt2 = trans2.transform(X)\n",
    "assert np.array_equal(Xt1, Xt2)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "dev",
   "language": "python",
   "name": "dev"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}