{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "source": [
    "from femda import FEMDA"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "# 1. Using FEMDA in a sklearn.pipeline"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "source": [
    "from sklearn.datasets import load_iris, load_digits\r\n",
    "from sklearn.pipeline import make_pipeline\r\n",
    "from sklearn.model_selection import train_test_split\r\n",
    "from sklearn.decomposition import PCA\r\n",
    "\r\n",
    "X, y = load_digits(return_X_y=True)\r\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.33, random_state=42)"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "source": [
    "pipe = make_pipeline(PCA(n_components=5), FEMDA())"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "source": [
    "pipe.fit(X_train, y_train)"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "Pipeline(steps=[('pca', PCA(n_components=5)), ('femda', FEMDA())])"
      ]
     },
     "metadata": {},
     "execution_count": 3
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "source": [
    "print(pipe.score(X_test, y_test))\r\n",
    "\r\n",
    "y_pred = pipe.predict(X_test)\r\n",
    "#from sklearn.metrics import classification_report\r\n",
    "#print(classification_report(y_test, y_pred))"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "0.8855218855218855\n"
     ]
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "# 2. Testing FEMDA on other datasets"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "source": [
    "import femda.experiments.preprocessing as pre\r\n",
    "from femda._models_lda import LDA, QDA\r\n",
    "from femda._models_t_lda import t_QDA\r\n",
    "X_train, y_train, X_test, y_test = pre.ionosphere(r\"data\\Paper\\\\\") \r\n",
    "#Choose between ionosphere, statlog, ecoli, breast_cancer, spam_base"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "source": [
    "print(\"FEMDA\", FEMDA().fit(X_train, y_train).score(X_test, y_test))\r\n",
    "print(\"t_QDA\", t_QDA().fit(X_train, y_train).score(X_test, y_test))\r\n",
    "print(\"LDA\",   LDA()  .fit(X_train, y_train).score(X_test, y_test))\r\n",
    "print(\"QDA\",   QDA()  .fit(X_train, y_train).score(X_test, y_test))"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "FEMDA 0.9339622641509434\n",
      "t_QDA 0.9339622641509434\n",
      "LDA 0.8773584905660378\n",
      "QDA 0.8867924528301887\n"
     ]
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "# 3. Running experiments presented in the paper"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "source": [
    "from femda.experiments import run_experiments"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "source": [
    "run_experiments()"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "# Check estimator"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "source": [
    "from sklearn.utils.estimator_checks import check_estimator"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "source": [
    "f = FEMDA()"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "source": [
    "print(check_estimator(f) == None)"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "True\n"
     ]
    }
   ],
   "metadata": {}
  }
 ],
 "metadata": {
  "interpreter": {
   "hash": "c7a650d791d0a1d035b66682f8967f04fed3045153a1ba3c3bfeefd2541b18a6"
  },
  "kernelspec": {
   "name": "python3",
   "display_name": "Python 3.8.7 64-bit"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}