dataflows/regress/computer-RidgeRegression-READ-SPLIT-REGRESS.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Linear Regression\n",
    "\n",
    "This notebook provides an example of how to use SML to read in a dataset, split the data into training and testing data, replace troublesome values such as 'NaNs' from the dataset, and perform linear regression on the dataset. For this use-case we use publicly availiable dataset [Computer Hardware Dataset]https://archive.ics.uci.edu/ml/datasets/Computer+Hardware) and use ridge regression to predict compute performance.\n",
    "\n",
    "## SML Query\n",
    "\n",
    "### Imports\n",
    "We import the nescessary library to use SML."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sml import execute"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Query\n",
    "\n",
    "Next we create a query statement to `READ` in the data and the file is delimited by a ',', the header is not used, next we `SPLIT` the dataset and use 80% of it for training and 20% of it for testing, and lastly, we perform linear regression on the 10th column, using columns 1-9 as the predictiors."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Sml Summary:\n",
      "=============================================\n",
      "=============================================\n",
      "   Dataset Path:        ../data/computer.csv\n",
      "   Delimiter:      ,\n",
      "   Training Set Split:       80.00%\n",
      "   Testing Set Split:        20.00%\n",
      "   Predictiors:        ['1', '2', '3', '4', '5', '6', '7', '8', '9']\n",
      "   Label:         10\n",
      "   Algorithm:     ridge\n",
      "   Dataset Preview:\n",
      "   adviser  32/60  125   256   6000  256.1  16  128  198  199\n",
      "0       22    137   29  8000  32000     32   8   32  269  253\n",
      "1       22     95   29  8000  32000     32   8   32  220  253\n",
      "2       22    120   29  8000  32000     32   8   32  172  253\n",
      "3       22    161   29  8000  16000     32   8   16  132  132\n",
      "4       22     51   26  8000  32000     64   8   32  318  290\n",
      "\n",
      "=============================================\n",
      "=============================================\n",
      "\n"
     ]
    }
   ],
   "source": [
    "query = 'READ \"../data/computer.csv\" (separator = \",\", header = 0) AND \\\n",
    "SPLIT (train = .8, test = .2, validation = .0) AND \\\n",
    "REGRESS (predictors = [1,2,3,4,5,6,7,8,9], label = 10, algorithm = ridge)'\n",
    "\n",
    "execute(query, verbose=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Manually\n",
    "\n",
    "The subsequent cells below show how the same actions of a SML query can be performed manually.\n",
    "\n",
    "### Imports\n",
    "Here we import the necessary libraries needed to perform the same actions as the SML query above."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "from sklearn.base import BaseEstimator, TransformerMixin\n",
    "from sklearn.feature_extraction import DictVectorizer\n",
    "\n",
    "from sklearn.cross_validation import train_test_split\n",
    "from sklearn.metrics import r2_score\n",
    "from sklearn import linear_model\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### READ\n",
    "\n",
    "By default the Computer Hardware data does not include it's headers, so we specify it manually, and read that file into a pandas dataframe."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>vendor name</th>\n",
       "      <th>Model Name</th>\n",
       "      <th>MYCT</th>\n",
       "      <th>MMIN</th>\n",
       "      <th>MMAX</th>\n",
       "      <th>CACH</th>\n",
       "      <th>CHMN</th>\n",
       "      <th>CHMAX</th>\n",
       "      <th>PRP</th>\n",
       "      <th>ERP</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>adviser</td>\n",
       "      <td>32/60</td>\n",
       "      <td>125</td>\n",
       "      <td>256</td>\n",
       "      <td>6000</td>\n",
       "      <td>256</td>\n",
       "      <td>16</td>\n",
       "      <td>128</td>\n",
       "      <td>198</td>\n",
       "      <td>199</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>amdahl</td>\n",
       "      <td>470v/7</td>\n",
       "      <td>29</td>\n",
       "      <td>8000</td>\n",
       "      <td>32000</td>\n",
       "      <td>32</td>\n",
       "      <td>8</td>\n",
       "      <td>32</td>\n",
       "      <td>269</td>\n",
       "      <td>253</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>amdahl</td>\n",
       "      <td>470v/7a</td>\n",
       "      <td>29</td>\n",
       "      <td>8000</td>\n",
       "      <td>32000</td>\n",
       "      <td>32</td>\n",
       "      <td>8</td>\n",
       "      <td>32</td>\n",
       "      <td>220</td>\n",
       "      <td>253</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>amdahl</td>\n",
       "      <td>470v/7b</td>\n",
       "      <td>29</td>\n",
       "      <td>8000</td>\n",
       "      <td>32000</td>\n",
       "      <td>32</td>\n",
       "      <td>8</td>\n",
       "      <td>32</td>\n",
       "      <td>172</td>\n",
       "      <td>253</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>amdahl</td>\n",
       "      <td>470v/7c</td>\n",
       "      <td>29</td>\n",
       "      <td>8000</td>\n",
       "      <td>16000</td>\n",
       "      <td>32</td>\n",
       "      <td>8</td>\n",
       "      <td>16</td>\n",
       "      <td>132</td>\n",
       "      <td>132</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  vendor name Model Name  MYCT  MMIN   MMAX  CACH  CHMN  CHMAX  PRP  ERP\n",
       "0     adviser      32/60   125   256   6000   256    16    128  198  199\n",
       "1      amdahl     470v/7    29  8000  32000    32     8     32  269  253\n",
       "2      amdahl    470v/7a    29  8000  32000    32     8     32  220  253\n",
       "3      amdahl    470v/7b    29  8000  32000    32     8     32  172  253\n",
       "4      amdahl    470v/7c    29  8000  16000    32     8     16  132  132"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "names = ['vendor name', 'Model Name', 'MYCT', 'MMIN', 'MMAX', 'CACH', 'CHMN', 'CHMAX', 'PRP', 'ERP']\n",
    "\n",
    "df = pd.read_csv('../data/computer.csv', header = None, names=names)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Preprocessing \n",
    "\n",
    "Here we iterate through the dataframe and convert columns of dtype object to categorical."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>vendor name</th>\n",
       "      <th>Model Name</th>\n",
       "      <th>MYCT</th>\n",
       "      <th>MMIN</th>\n",
       "      <th>MMAX</th>\n",
       "      <th>CACH</th>\n",
       "      <th>CHMN</th>\n",
       "      <th>CHMAX</th>\n",
       "      <th>PRP</th>\n",
       "      <th>ERP</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>13</td>\n",
       "      <td>27</td>\n",
       "      <td>125</td>\n",
       "      <td>256</td>\n",
       "      <td>6000</td>\n",
       "      <td>256</td>\n",
       "      <td>16</td>\n",
       "      <td>128</td>\n",
       "      <td>198</td>\n",
       "      <td>199</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>23</td>\n",
       "      <td>138</td>\n",
       "      <td>29</td>\n",
       "      <td>8000</td>\n",
       "      <td>32000</td>\n",
       "      <td>32</td>\n",
       "      <td>8</td>\n",
       "      <td>32</td>\n",
       "      <td>269</td>\n",
       "      <td>253</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>23</td>\n",
       "      <td>96</td>\n",
       "      <td>29</td>\n",
       "      <td>8000</td>\n",
       "      <td>32000</td>\n",
       "      <td>32</td>\n",
       "      <td>8</td>\n",
       "      <td>32</td>\n",
       "      <td>220</td>\n",
       "      <td>253</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>23</td>\n",
       "      <td>121</td>\n",
       "      <td>29</td>\n",
       "      <td>8000</td>\n",
       "      <td>32000</td>\n",
       "      <td>32</td>\n",
       "      <td>8</td>\n",
       "      <td>32</td>\n",
       "      <td>172</td>\n",
       "      <td>253</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>23</td>\n",
       "      <td>162</td>\n",
       "      <td>29</td>\n",
       "      <td>8000</td>\n",
       "      <td>16000</td>\n",
       "      <td>32</td>\n",
       "      <td>8</td>\n",
       "      <td>16</td>\n",
       "      <td>132</td>\n",
       "      <td>132</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   vendor name  Model Name  MYCT  MMIN   MMAX  CACH  CHMN  CHMAX  PRP  ERP\n",
       "0           13          27   125   256   6000   256    16    128  198  199\n",
       "1           23         138    29  8000  32000    32     8     32  269  253\n",
       "2           23          96    29  8000  32000    32     8     32  220  253\n",
       "3           23         121    29  8000  32000    32     8     32  172  253\n",
       "4           23         162    29  8000  16000    32     8     16  132  132"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def encode_categorical(df, cols=None):\n",
    "  categorical = list()\n",
    "  if cols is not None:\n",
    "    categorical = cols\n",
    "  else:\n",
    "    for col in df.columns:\n",
    "        if df[col].dtype == 'object':\n",
    "            categorical.append(col)\n",
    "\n",
    "  for feature in categorical:\n",
    "      l = list(df[feature])\n",
    "      s = set(l)\n",
    "      l2 = list(s)\n",
    "      numbers = list()\n",
    "      for i in range(0,len(l2)):\n",
    "          numbers.append(i)\n",
    "      df[feature] = df[feature].replace(l2, numbers)\n",
    "  return df\n",
    "\n",
    "df2 =  encode_categorical(df)\n",
    "df2.head()\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### SPLIT\n",
    "We then seperate our labels from our features and use a sklearn function to perform a 80%/20% split our training and testing dataset respectively.# SPLIT"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "features = df2.drop('PRP',1)\n",
    "label = df2['PRP']\n",
    "X_train, X_test, y_train, y_test = train_test_split(features, label, test_size=0.2, random_state=42)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### REGRESS\n",
    "\n",
    "We fit our Ridge regression model with our training dataset and make predictions on our testing dataset and display the accuracy."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9607123194827184"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ridge = linear_model.Ridge(alpha=0.1)\n",
    "ridge.fit(X_train, y_train)\n",
    "pred = ridge.predict(X_test)\n",
    "r2_score(pred, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python [Root]",
   "language": "python",
   "name": "Python [Root]"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}