IDS-project-checkpoint.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "d7dc8df3",
   "metadata": {},
   "source": [
    "# Group B15 - Kaggle Student Alcohol consumption"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7ee6bea1",
   "metadata": {},
   "source": [
    "## Predicting student's final grade "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7cc6f3ac",
   "metadata": {},
   "source": [
    " Otto Bruno Koobakene, Liivika Koobakene"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c3786dd6",
   "metadata": {},
   "source": [
    "\n",
    "## Imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "ff5bed3a",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.linear_model import LinearRegression, Lasso, Ridge\n",
    "from sklearn.neighbors import KNeighborsRegressor\n",
    "from sklearn.metrics import mean_absolute_error\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b8397dae",
   "metadata": {},
   "source": [
    "## Processing data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e4d5b826",
   "metadata": {},
   "source": [
    "Merging Portuguese and Math student's data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "e3e5dd5a",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.concat(\n",
    "    map(pd.read_csv, ['student-mat.csv', 'student-por.csv']), ignore_index=True)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "17ba7f84",
   "metadata": {},
   "source": [
    "Adding new column for average alcohol consumption per week"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "336bfe9b",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['weeklyAlcohol'] = (df.Dalc + df.Walc)/2"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7b2b17d4",
   "metadata": {},
   "source": [
    "Deleting unnecessary columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "2d54a668",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df.drop(['school','sex','age','address','famsize','Pstatus','Medu','Fedu','Mjob','Fjob','reason','guardian','traveltime','studytime','failures','schoolsup','activities','nursery','higher','Dalc','Walc','internet','romantic','freetime','goout','health','G1','G2'\n",
    "], axis=1)\n",
    "df = df.rename(columns={'G3':'finalgrade'})"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1e84179c",
   "metadata": {},
   "source": [
    "Generating columns for \"yes or no\" questions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "46be003c",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.concat([df.drop('famsup', axis=1), pd.get_dummies(df['famsup'])], axis=1)\n",
    "df = df.rename(columns = {'no': 'no_famsup', 'yes': 'famsup'})\n",
    "df = pd.concat([df.drop('paid', axis=1), pd.get_dummies(df['paid'])], axis=1)\n",
    "df = df.rename(columns = {'no': 'no_paid', 'yes': 'paid'})\n",
    "df = df.drop(['no_famsup', 'no_paid'], axis = 1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cdb5c42f",
   "metadata": {},
   "source": [
    "Generating new merged csv file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "b922d3bd",
   "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>famrel</th>\n",
       "      <th>absences</th>\n",
       "      <th>finalgrade</th>\n",
       "      <th>weeklyAlcohol</th>\n",
       "      <th>famsup</th>\n",
       "      <th>paid</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>4</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5</td>\n",
       "      <td>4</td>\n",
       "      <td>6</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>2.5</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>15</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>10</td>\n",
       "      <td>1.5</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   famrel  absences  finalgrade  weeklyAlcohol  famsup  paid\n",
       "0       4         6           6            1.0       0     0\n",
       "1       5         4           6            1.0       1     0\n",
       "2       4        10          10            2.5       0     1\n",
       "3       3         2          15            1.0       1     1\n",
       "4       4         4          10            1.5       1     1"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.to_csv('merged.csv')\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "42e78dce",
   "metadata": {},
   "source": [
    "## Learning models"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "43d53cea",
   "metadata": {},
   "source": [
    "Training and test sets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "cac4eab3",
   "metadata": {},
   "outputs": [],
   "source": [
    "# X is the input set and y is the output set\n",
    "#X,y = df.drop(columns=[\"finalgrade\",'weeklyAlcohol','famrel'], axis=1), df.finalgrade\n",
    "#testing all given factors combined\n",
    "X,y = df.drop(columns=['finalgrade'], axis=1), df.finalgrade\n",
    "# Creating training and test sets\n",
    "# Since we want the comparison between different mean absolute values to be fair, we use random_state = 0\n",
    "X_train, X_test, y_train, y_test = train_test_split(X.values, y.values, test_size=0.3, random_state=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f7781c21",
   "metadata": {},
   "source": [
    "##### Lasso"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "ab5a46f2",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mean absolute error for Lasso regression:  2.43529\n"
     ]
    }
   ],
   "source": [
    "lasso = Lasso(alpha = 0.1).fit(X_train, y_train)\n",
    "lasso_pred = lasso.predict(X_test)\n",
    "#print(lasso_pred)\n",
    "lasso_mae = mean_absolute_error(y_test, lasso_pred)\n",
    "print(\"Mean absolute error for Lasso regression: \" , round(lasso_mae,5))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f6a8c139",
   "metadata": {},
   "source": [
    "##### Linear Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "4d21e874",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mean absolute error for Linear regression: 2.4415\n"
     ]
    }
   ],
   "source": [
    "linear = LinearRegression().fit(X_train, y_train)\n",
    "pred_linear = linear.predict(X_test)\n",
    "#print(pred_linear)\n",
    "linear_mae = mean_absolute_error(y_test, pred_linear)\n",
    "print (\"Mean absolute error for Linear regression:\", round(linear_mae,4))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3a6f5c8c",
   "metadata": {},
   "source": [
    "##### K nearest neighbor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "6299f0dd",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mean absolute error for K nearest neighbors: 2.7573\n"
     ]
    }
   ],
   "source": [
    "KNeighbors= KNeighborsRegressor().fit(X_train, y_train)\n",
    "pred_kneighbors = KNeighbors.predict(X_test)\n",
    "#print(pred_kneighbors)\n",
    "kneighbors_mae = mean_absolute_error(y_test, pred_kneighbors)\n",
    "print (\"Mean absolute error for K nearest neighbors:\", round(kneighbors_mae,4))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "09600b9f",
   "metadata": {
    "scrolled": true
   },
   "source": [
    "##### Ridge"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "f6f6d16d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mean absolute error for Ridge:  2.44149\n"
     ]
    }
   ],
   "source": [
    "Ridge = Ridge().fit(X_train, y_train)\n",
    "Ridge_pred = Ridge.predict(X_test)\n",
    "#print(Ridge_pred)\n",
    "Ridge_mae = mean_absolute_error(y_test, Ridge_pred)\n",
    "print(\"Mean absolute error for Ridge: \", round(Ridge_mae, 5))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dbdbbc1d",
   "metadata": {},
   "source": [
    "### Comparison"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d41bd969",
   "metadata": {},
   "source": [
    "##### Visualizing the mean absolute errors"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "010b9608",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_lasso = round(lasso_mae, 5)\n",
    "df_kneighbor = round(kneighbors_mae, 5)\n",
    "df_linear = round(linear_mae, 5)\n",
    "df_ridge = round(Ridge_mae, 5)\n",
    "names = ['Lasso', 'K nearest neighbor', 'Linear Regression', 'Ridge']\n",
    "values = [df_lasso, df_kneighbor, df_linear, df_ridge]\n",
    "\n",
    "plt.rcParams['figure.figsize'] = (12,8)\n",
    "matplotlib.style.use('ggplot')\n",
    "df_plt = pd.DataFrame({'Name':names, 'Mean absolute error':values})\n",
    "df_plt.sort_values('Mean absolute error', ascending = True).plot.bar(x = 'Name', y = 'Mean absolute error', rot = 1)\n",
    "\n",
    "values.sort()\n",
    "for index, data in enumerate(values):\n",
    "    plt.text(x = index-0.25, y = data+0.03 , s = f\"{data}\", fontdict = dict(fontsize = 20))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "74c8a344",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0f9d74b9",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8170de4c",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3ea2ad7c",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "caaf56eb",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e2864b65",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "90d54f1e",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8bdd0395",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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
}