cars.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Dense, Dropout\n",
    "from tensorflow.keras import optimizers\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.metrics import mean_squared_error\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "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>mpg</th>\n",
       "      <th>cylinders</th>\n",
       "      <th>displacement</th>\n",
       "      <th>horsepower</th>\n",
       "      <th>weight</th>\n",
       "      <th>acceleration</th>\n",
       "      <th>model</th>\n",
       "      <th>origin</th>\n",
       "      <th>car_name</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>18.0</td>\n",
       "      <td>8</td>\n",
       "      <td>307.0</td>\n",
       "      <td>130.0</td>\n",
       "      <td>3504.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>70</td>\n",
       "      <td>1</td>\n",
       "      <td>chevrolet chevelle malibu</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    mpg  cylinders  displacement horsepower  weight  acceleration  model  \\\n",
       "0  18.0          8         307.0      130.0  3504.0          12.0     70   \n",
       "\n",
       "   origin                   car_name  \n",
       "0       1  chevrolet chevelle malibu  "
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv(\"cars_multivariate.csv\")\n",
    "df.head(1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['horsepower'] = pd.to_numeric(df.horsepower, errors='coerce')\n",
    "df.horsepower.fillna(df.horsepower.mean(), inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = df.mpg\n",
    "X = df.drop(columns=['mpg', 'car_name', 'origin'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.33, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "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>cylinders</th>\n",
       "      <th>displacement</th>\n",
       "      <th>horsepower</th>\n",
       "      <th>weight</th>\n",
       "      <th>acceleration</th>\n",
       "      <th>model</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>324</th>\n",
       "      <td>4</td>\n",
       "      <td>85.0</td>\n",
       "      <td>65.0</td>\n",
       "      <td>2110.0</td>\n",
       "      <td>19.2</td>\n",
       "      <td>80</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     cylinders  displacement  horsepower  weight  acceleration  model\n",
       "324          4          85.0        65.0  2110.0          19.2     80"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.head(1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/chyld/.conda/envs/py37/lib/python3.7/site-packages/sklearn/preprocessing/data.py:645: DataConversionWarning: Data with input dtype int64, float64 were all converted to float64 by StandardScaler.\n",
      "  return self.partial_fit(X, y)\n",
      "/Users/chyld/.conda/envs/py37/lib/python3.7/site-packages/ipykernel_launcher.py:2: DataConversionWarning: Data with input dtype int64, float64 were all converted to float64 by StandardScaler.\n",
      "  \n",
      "/Users/chyld/.conda/envs/py37/lib/python3.7/site-packages/ipykernel_launcher.py:3: DataConversionWarning: Data with input dtype int64, float64 were all converted to float64 by StandardScaler.\n",
      "  This is separate from the ipykernel package so we can avoid doing imports until\n"
     ]
    }
   ],
   "source": [
    "scalar = StandardScaler().fit(X_train)\n",
    "X_train = scalar.transform(X_train)\n",
    "X_test = scalar.transform(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [],
   "source": [
    "regression = LinearRegression().fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "9.668178518310656"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mean_squared_error(y_test, regression.predict(X_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = Sequential()\n",
    "model.add(Dense(units=16, activation='relu', input_shape=(6,), name='L1'))\n",
    "model.add(Dense(units=8, activation='relu', name='L3'))\n",
    "model.add(Dense(units=1, activation='linear', name='Output'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_10\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "L1 (Dense)                   (None, 16)                112       \n",
      "_________________________________________________________________\n",
      "L3 (Dense)                   (None, 8)                 136       \n",
      "_________________________________________________________________\n",
      "Output (Dense)               (None, 1)                 9         \n",
      "=================================================================\n",
      "Total params: 257\n",
      "Trainable params: 257\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "opt = optimizers.SGD()\n",
    "model.compile(loss='mse', optimizer=opt, metrics=['mse'])\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [],
   "source": [
    "history = model.fit(X_train, y_train, epochs=100, batch_size=25, verbose=0, validation_split=0.3, shuffle=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x504 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(12, 7))\n",
    "plt.plot(history.epoch, history.history['mse'], color='purple', alpha=0.5, label='train')\n",
    "plt.plot(history.epoch, history.history['val_mse'], color='green', alpha=0.5, label='validation')\n",
    "plt.xlabel('epochs')\n",
    "plt.ylabel('mse')\n",
    "\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "6.035461385016611"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mean_squared_error(y_test, model.predict(X_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "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": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}