Hackathon.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "207ee8e6-aead-406a-b875-1dade0657d10",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import scipy.stats as stats\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "from imblearn.over_sampling import SMOTE\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.metrics import accuracy_score, f1_score, roc_auc_score, roc_curve, precision_recall_curve\n",
    "\n",
    "data = pd.read_csv(\"train.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "c602555a-535a-4b4c-8b0c-f5b965ecd1d3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Entfernte Features:  ['A1', 'I']\n"
     ]
    }
   ],
   "source": [
    "# Rauschen entfernen\n",
    "# Gemacht über eine Korrelationsmatrix und davon dann das obere Dreieck auswählen    \n",
    "corr_matrix = data.corr().abs()\n",
    "upper = corr_matrix.where((np.triu(np.ones(corr_matrix.shape), k=1) + np.tril(np.ones(corr_matrix.shape), k=-1)).astype(bool))\n",
    "\n",
    "#Finde die Features, die eine Korrelation von größer als 0.95 besitzen\n",
    "to_drop = [column for column in upper.columns if any(upper[column] > 0.95)]\n",
    "\n",
    "#Entferne gefundene Features\n",
    "print(\"Entfernte Features: \", to_drop)\n",
    "data.drop(to_drop, axis = 1, inplace=True)\n",
    "\n",
    "# Da A1, B1, C1, D1, E1 alle von I und A2, B2, C2, D2, E2 abhängen, können wir diese auch gleich entfernen, da I und A1 entfernt wurden.\n",
    "data.drop([\"B1\", \"C1\", \"D1\", \"E1\"], axis = 1, inplace=True)\n",
    "\n",
    "# print(data[\"Target\"].value_counts())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "57a78164-d633-4540-89c9-e171f32bbb6d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Oversampling der Minority Klasse, um so Informationen der Minority Klasse nicht zu verlieren\n",
    "sm = SMOTE(sampling_strategy='minority', random_state=42)\n",
    "\n",
    "# Modell fitten und die Daten generieren\n",
    "oversampled_trainX, oversampled_trainY = sm.fit_sample(data.drop('Target', axis=1), data['Target'])\n",
    "data = pd.concat([pd.DataFrame(oversampled_trainX), pd.DataFrame(oversampled_trainY)], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "1d5098b3-2a8f-4a29-8039-93f60c13c803",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['C3', 'F', 'G', 'H', 'D2', 'E2', 'A2', 'B2', 'C2']\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 9 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Ausreißer entfernen geht nicht einfach, da viele Nullen enthalten sind\n",
    "# Selektives Ausreißer entfernen durch IQR bringt Bias ins Bild, wahrscheinlich nicht ideal\n",
    "\n",
    "# Entferne alle Outlier mit IQR\n",
    "# Q1 = data.quantile(q=.25)\n",
    "# Q3 = data.quantile(q=.75)\n",
    "# IQR = Q3 - Q1\n",
    "\n",
    "# data = data[~((data < (Q1-1.5*IQR)) | (data > (Q3+1.5*IQR))).any(axis=1)]\n",
    "\n",
    "# Transformiere Daten, sodass diese nicht so nach links skewen. Hier durch einen log Transform.\n",
    "# Da log(0) = undefiniert, addieren wir 0.5 zu allen Features hinzu, um den Log darauf trotzdem verwenden zu können\n",
    "\n",
    "data_zero = data.drop(\"Target\", axis=1)\n",
    "feature_names = list(data_zero.columns)\n",
    "print(feature_names)\n",
    "data_zero[feature_names] += 0.5\n",
    "\n",
    "transformed_data = np.log(data_zero[feature_names])\n",
    "transformed_data.hist()\n",
    "\n",
    "data = pd.concat([transformed_data, data[\"Target\"]], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "f6e005e6-1c63-4f46-94a1-8f20de70cdf5",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Normalisieren (Min-Max Scaling)\n",
    "data_norm = np.divide(data - data.min(), data.max() - data.min())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "cf058aa3-bc73-49c7-832c-3ea33bf20754",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy Score:  0.999329268292683\n",
      "F1 Score:  0.9993268465822166\n"
     ]
    }
   ],
   "source": [
    "feature_names = list(data_norm.columns)\n",
    "feature_names.remove(\"Target\")\n",
    "X = data_norm[feature_names]\n",
    "y = data_norm[\"Target\"]\n",
    "\n",
    "# 80/20 Split\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y , test_size=0.2)\n",
    "\n",
    "model = RandomForestClassifier()\n",
    "model.fit(X_train, y_train)\n",
    "\n",
    "y_pred = model.predict(X_test)\n",
    "print(\"Accuracy Score: \", accuracy_score(y_pred, y_test))\n",
    "print(\"F1 Score: \", f1_score(y_pred, y_test)) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "95b4ca21-1ba9-4adb-888d-52727dfae14a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Ergebnis visualisieren\n",
    "\n",
    "fpr, tpr, thresholds = roc_curve(y_test, y_pred)\n",
    "plt.plot(fpr, tpr)\n",
    "plt.xlabel('False positive rate')\n",
    "plt.ylabel('True positive rate')\n",
    "plt.title('ROC curve')\n",
    "plt.show()\n",
    "\n",
    "precision, recall, thresholds = precision_recall_curve(y_test, y_pred)\n",
    "fig, ax = plt.subplots()\n",
    "ax.plot(recall, precision, color='purple')\n",
    "\n",
    "ax.set_title('Precision-Recall Curve')\n",
    "ax.set_xlabel('Recall')\n",
    "ax.set_ylabel('Precision')\n",
    "\n",
    "\n",
    "plt.show()"
   ]
  }
 ],
 "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": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}