getting_started.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import seaborn as sns\n",
    "sns.set_style('whitegrid')\n",
    "sns.set_color_codes()\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "324 patients, 35 features\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>age</th>\n",
       "      <th>gender</th>\n",
       "      <th>city</th>\n",
       "      <th>asbestos exposure</th>\n",
       "      <th>type of MM</th>\n",
       "      <th>duration of asbestos exposure</th>\n",
       "      <th>diagnosis method</th>\n",
       "      <th>keep side</th>\n",
       "      <th>cytology</th>\n",
       "      <th>duration of symptoms</th>\n",
       "      <th>...</th>\n",
       "      <th>pleural lactic dehydrogenise</th>\n",
       "      <th>pleural protein</th>\n",
       "      <th>pleural albumin</th>\n",
       "      <th>pleural glucose</th>\n",
       "      <th>dead or not</th>\n",
       "      <th>pleural effusion</th>\n",
       "      <th>pleural thickness on tomography</th>\n",
       "      <th>pleural level of acidity (pH)</th>\n",
       "      <th>C-reactive protein (CRP)</th>\n",
       "      <th>class of diagnosis</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>47.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>20.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>24.0</td>\n",
       "      <td>...</td>\n",
       "      <td>289.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>79.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>34</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>55.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>45.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>7541.0</td>\n",
       "      <td>1.6</td>\n",
       "      <td>0.80</td>\n",
       "      <td>6.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>42</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>29.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>23.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>480.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>90.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>43</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>39.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>...</td>\n",
       "      <td>459.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>2.80</td>\n",
       "      <td>45.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>21</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>47.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1.5</td>\n",
       "      <td>...</td>\n",
       "      <td>213.0</td>\n",
       "      <td>3.6</td>\n",
       "      <td>1.95</td>\n",
       "      <td>53.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 35 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    age  gender  city  asbestos exposure  type of MM  \\\n",
       "0  47.0       1     0                  1         0.0   \n",
       "1  55.0       1     0                  1         0.0   \n",
       "2  29.0       1     1                  1         0.0   \n",
       "3  39.0       1     0                  1         0.0   \n",
       "4  47.0       1     0                  1         0.0   \n",
       "\n",
       "   duration of asbestos exposure  diagnosis method  keep side  cytology  \\\n",
       "0                           20.0                 1          0         1   \n",
       "1                           45.0                 1          0         0   \n",
       "2                           23.0                 0          1         0   \n",
       "3                           10.0                 1          0         0   \n",
       "4                           10.0                 1          1         1   \n",
       "\n",
       "   duration of symptoms         ...          pleural lactic dehydrogenise  \\\n",
       "0                  24.0         ...                                 289.0   \n",
       "1                   1.0         ...                                7541.0   \n",
       "2                   1.0         ...                                 480.0   \n",
       "3                   3.0         ...                                 459.0   \n",
       "4                   1.5         ...                                 213.0   \n",
       "\n",
       "   pleural protein  pleural albumin  pleural glucose  dead or not  \\\n",
       "0              0.0             0.00             79.0            1   \n",
       "1              1.6             0.80              6.0            1   \n",
       "2              0.0             0.00             90.0            1   \n",
       "3              5.0             2.80             45.0            1   \n",
       "4              3.6             1.95             53.0            1   \n",
       "\n",
       "   pleural effusion  pleural thickness on tomography  \\\n",
       "0               0.0                              0.0   \n",
       "1               1.0                              1.0   \n",
       "2               0.0                              0.0   \n",
       "3               1.0                              0.0   \n",
       "4               1.0                              0.0   \n",
       "\n",
       "   pleural level of acidity (pH)  C-reactive protein (CRP)  class of diagnosis  \n",
       "0                            0.0                        34                   1  \n",
       "1                            1.0                        42                   1  \n",
       "2                            0.0                        43                   2  \n",
       "3                            0.0                        21                   1  \n",
       "4                            0.0                        11                   1  \n",
       "\n",
       "[5 rows x 35 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = pd.read_csv('data.csv')\n",
    "print('%s patients, %s features' % data.shape)\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "age                                 False\n",
       "gender                              False\n",
       "city                                False\n",
       "asbestos exposure                   False\n",
       "type of MM                          False\n",
       "duration of asbestos exposure       False\n",
       "diagnosis method                    False\n",
       "keep side                           False\n",
       "cytology                            False\n",
       "duration of symptoms                False\n",
       "dyspnoea                            False\n",
       "ache on chest                       False\n",
       "weakness                            False\n",
       "habit of cigarette                  False\n",
       "performance status                  False\n",
       "white blood                         False\n",
       "cell count (WBC)                    False\n",
       "hemoglobin (HGB)                    False\n",
       "platelet count (PLT)                False\n",
       "sedimentation                       False\n",
       "blood lactic dehydrogenise (LDH)    False\n",
       "alkaline phosphatise (ALP)          False\n",
       "total protein                       False\n",
       "albumin                             False\n",
       "glucose                             False\n",
       "pleural lactic dehydrogenise        False\n",
       "pleural protein                     False\n",
       "pleural albumin                     False\n",
       "pleural glucose                     False\n",
       "dead or not                         False\n",
       "pleural effusion                    False\n",
       "pleural thickness on tomography     False\n",
       "pleural level of acidity (pH)       False\n",
       "C-reactive protein (CRP)            False\n",
       "class of diagnosis                  False\n",
       "dtype: bool"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.isnull().any()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAhoAAAF9CAYAAABCoMFZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f771f94d978>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = sns.violinplot(x='class of diagnosis', y='age', hue='gender', data=data, split=True)\n",
    "ax.set_xticklabels(['Healthy', 'Mesothelioma']);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Healthy patients:   MM type 0: 97.37%;  MM type 1: 2.63%;  MM type 2: 0.00%\n",
      "Mesothelioma patients:   MM type 0: 91.67%;  MM type 1: 5.21%;  MM type 2: 3.12%\n"
     ]
    }
   ],
   "source": [
    "for label, (_, table) in zip(['Healthy', 'Mesothelioma'], data.groupby('class of diagnosis')):\n",
    "    \n",
    "    percent, mm_type = np.histogram(table['type of MM'], bins=range(4), normed=True)\n",
    "    \n",
    "    histogram = ';'.join(('  MM type %s: %.2f%%' % (t, p)) for t, p in zip(mm_type, (percent*100)))\n",
    "    \n",
    "    print(label, 'patients:', histogram)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "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": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}