1 - Analysis-for-Gene-Classification.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "57a84db9-8817-4589-9ffb-c0ed70395c93",
    "_uuid": "220ae8bf2e2ecdd588a3d6281f3016ba24c47c2d"
   },
   "source": [
    "\n",
    "Gene analysis is something which is widely used in biological science.Many of the Data learners coming here will be haveing trouble with these weird transciption numbers. So why not just conver this into just two columns and then apply our analysis."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "_cell_guid": "a1222775-7e0e-4be6-8622-929b9c0524f3",
    "_execution_state": "idle",
    "_uuid": "09385d2dec4ad2c202f8851ca432cb58375532aa"
   },
   "outputs": [],
   "source": [
    "import numpy as np # linear algebra\n",
    "import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "_cell_guid": "6a8bb215-ab90-4b79-9dda-a6e6ecf01048",
    "_execution_state": "idle",
    "_uuid": "f2c57a9d0028643593a8a7d9f57232f4d861897c"
   },
   "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>Gene Description</th>\n",
       "      <th>Gene Accession Number</th>\n",
       "      <th>1</th>\n",
       "      <th>call</th>\n",
       "      <th>2</th>\n",
       "      <th>call.1</th>\n",
       "      <th>3</th>\n",
       "      <th>call.2</th>\n",
       "      <th>4</th>\n",
       "      <th>call.3</th>\n",
       "      <th>...</th>\n",
       "      <th>29</th>\n",
       "      <th>call.33</th>\n",
       "      <th>30</th>\n",
       "      <th>call.34</th>\n",
       "      <th>31</th>\n",
       "      <th>call.35</th>\n",
       "      <th>32</th>\n",
       "      <th>call.36</th>\n",
       "      <th>33</th>\n",
       "      <th>call.37</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>AFFX-BioB-5_at (endogenous control)</td>\n",
       "      <td>AFFX-BioB-5_at</td>\n",
       "      <td>-214</td>\n",
       "      <td>A</td>\n",
       "      <td>-139</td>\n",
       "      <td>A</td>\n",
       "      <td>-76</td>\n",
       "      <td>A</td>\n",
       "      <td>-135</td>\n",
       "      <td>A</td>\n",
       "      <td>...</td>\n",
       "      <td>15</td>\n",
       "      <td>A</td>\n",
       "      <td>-318</td>\n",
       "      <td>A</td>\n",
       "      <td>-32</td>\n",
       "      <td>A</td>\n",
       "      <td>-124</td>\n",
       "      <td>A</td>\n",
       "      <td>-135</td>\n",
       "      <td>A</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>AFFX-BioB-M_at (endogenous control)</td>\n",
       "      <td>AFFX-BioB-M_at</td>\n",
       "      <td>-153</td>\n",
       "      <td>A</td>\n",
       "      <td>-73</td>\n",
       "      <td>A</td>\n",
       "      <td>-49</td>\n",
       "      <td>A</td>\n",
       "      <td>-114</td>\n",
       "      <td>A</td>\n",
       "      <td>...</td>\n",
       "      <td>-114</td>\n",
       "      <td>A</td>\n",
       "      <td>-192</td>\n",
       "      <td>A</td>\n",
       "      <td>-49</td>\n",
       "      <td>A</td>\n",
       "      <td>-79</td>\n",
       "      <td>A</td>\n",
       "      <td>-186</td>\n",
       "      <td>A</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>AFFX-BioB-3_at (endogenous control)</td>\n",
       "      <td>AFFX-BioB-3_at</td>\n",
       "      <td>-58</td>\n",
       "      <td>A</td>\n",
       "      <td>-1</td>\n",
       "      <td>A</td>\n",
       "      <td>-307</td>\n",
       "      <td>A</td>\n",
       "      <td>265</td>\n",
       "      <td>A</td>\n",
       "      <td>...</td>\n",
       "      <td>2</td>\n",
       "      <td>A</td>\n",
       "      <td>-95</td>\n",
       "      <td>A</td>\n",
       "      <td>49</td>\n",
       "      <td>A</td>\n",
       "      <td>-37</td>\n",
       "      <td>A</td>\n",
       "      <td>-70</td>\n",
       "      <td>A</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>AFFX-BioC-5_at (endogenous control)</td>\n",
       "      <td>AFFX-BioC-5_at</td>\n",
       "      <td>88</td>\n",
       "      <td>A</td>\n",
       "      <td>283</td>\n",
       "      <td>A</td>\n",
       "      <td>309</td>\n",
       "      <td>A</td>\n",
       "      <td>12</td>\n",
       "      <td>A</td>\n",
       "      <td>...</td>\n",
       "      <td>193</td>\n",
       "      <td>A</td>\n",
       "      <td>312</td>\n",
       "      <td>A</td>\n",
       "      <td>230</td>\n",
       "      <td>P</td>\n",
       "      <td>330</td>\n",
       "      <td>A</td>\n",
       "      <td>337</td>\n",
       "      <td>A</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>AFFX-BioC-3_at (endogenous control)</td>\n",
       "      <td>AFFX-BioC-3_at</td>\n",
       "      <td>-295</td>\n",
       "      <td>A</td>\n",
       "      <td>-264</td>\n",
       "      <td>A</td>\n",
       "      <td>-376</td>\n",
       "      <td>A</td>\n",
       "      <td>-419</td>\n",
       "      <td>A</td>\n",
       "      <td>...</td>\n",
       "      <td>-51</td>\n",
       "      <td>A</td>\n",
       "      <td>-139</td>\n",
       "      <td>A</td>\n",
       "      <td>-367</td>\n",
       "      <td>A</td>\n",
       "      <td>-188</td>\n",
       "      <td>A</td>\n",
       "      <td>-407</td>\n",
       "      <td>A</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 78 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                      Gene Description Gene Accession Number    1 call    2  \\\n",
       "0  AFFX-BioB-5_at (endogenous control)        AFFX-BioB-5_at -214    A -139   \n",
       "1  AFFX-BioB-M_at (endogenous control)        AFFX-BioB-M_at -153    A  -73   \n",
       "2  AFFX-BioB-3_at (endogenous control)        AFFX-BioB-3_at  -58    A   -1   \n",
       "3  AFFX-BioC-5_at (endogenous control)        AFFX-BioC-5_at   88    A  283   \n",
       "4  AFFX-BioC-3_at (endogenous control)        AFFX-BioC-3_at -295    A -264   \n",
       "\n",
       "  call.1    3 call.2    4 call.3  ...   29 call.33   30 call.34   31 call.35  \\\n",
       "0      A  -76      A -135      A  ...   15       A -318       A  -32       A   \n",
       "1      A  -49      A -114      A  ... -114       A -192       A  -49       A   \n",
       "2      A -307      A  265      A  ...    2       A  -95       A   49       A   \n",
       "3      A  309      A   12      A  ...  193       A  312       A  230       P   \n",
       "4      A -376      A -419      A  ...  -51       A -139       A -367       A   \n",
       "\n",
       "    32 call.36   33 call.37  \n",
       "0 -124       A -135       A  \n",
       "1  -79       A -186       A  \n",
       "2  -37       A  -70       A  \n",
       "3  330       A  337       A  \n",
       "4 -188       A -407       A  \n",
       "\n",
       "[5 rows x 78 columns]"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df= pd.read_csv('../input/data_set_ALL_AML_train.csv')\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "_cell_guid": "ff0b95b2-4e17-4750-a599-b6c907b0cff4",
    "_execution_state": "idle",
    "_uuid": "e8605ede271058592a09f59fc3b567e71cad4ddb"
   },
   "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>Gene Description</th>\n",
       "      <th>Gene Accession Number</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>...</th>\n",
       "      <th>35</th>\n",
       "      <th>36</th>\n",
       "      <th>37</th>\n",
       "      <th>38</th>\n",
       "      <th>28</th>\n",
       "      <th>29</th>\n",
       "      <th>30</th>\n",
       "      <th>31</th>\n",
       "      <th>32</th>\n",
       "      <th>33</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>AFFX-BioB-5_at (endogenous control)</td>\n",
       "      <td>AFFX-BioB-5_at</td>\n",
       "      <td>-214</td>\n",
       "      <td>-139</td>\n",
       "      <td>-76</td>\n",
       "      <td>-135</td>\n",
       "      <td>-106</td>\n",
       "      <td>-138</td>\n",
       "      <td>-72</td>\n",
       "      <td>-413</td>\n",
       "      <td>...</td>\n",
       "      <td>7</td>\n",
       "      <td>-213</td>\n",
       "      <td>-25</td>\n",
       "      <td>-72</td>\n",
       "      <td>-4</td>\n",
       "      <td>15</td>\n",
       "      <td>-318</td>\n",
       "      <td>-32</td>\n",
       "      <td>-124</td>\n",
       "      <td>-135</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>AFFX-BioB-M_at (endogenous control)</td>\n",
       "      <td>AFFX-BioB-M_at</td>\n",
       "      <td>-153</td>\n",
       "      <td>-73</td>\n",
       "      <td>-49</td>\n",
       "      <td>-114</td>\n",
       "      <td>-125</td>\n",
       "      <td>-85</td>\n",
       "      <td>-144</td>\n",
       "      <td>-260</td>\n",
       "      <td>...</td>\n",
       "      <td>-100</td>\n",
       "      <td>-252</td>\n",
       "      <td>-20</td>\n",
       "      <td>-139</td>\n",
       "      <td>-116</td>\n",
       "      <td>-114</td>\n",
       "      <td>-192</td>\n",
       "      <td>-49</td>\n",
       "      <td>-79</td>\n",
       "      <td>-186</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>AFFX-BioB-3_at (endogenous control)</td>\n",
       "      <td>AFFX-BioB-3_at</td>\n",
       "      <td>-58</td>\n",
       "      <td>-1</td>\n",
       "      <td>-307</td>\n",
       "      <td>265</td>\n",
       "      <td>-76</td>\n",
       "      <td>215</td>\n",
       "      <td>238</td>\n",
       "      <td>7</td>\n",
       "      <td>...</td>\n",
       "      <td>-57</td>\n",
       "      <td>136</td>\n",
       "      <td>124</td>\n",
       "      <td>-1</td>\n",
       "      <td>-125</td>\n",
       "      <td>2</td>\n",
       "      <td>-95</td>\n",
       "      <td>49</td>\n",
       "      <td>-37</td>\n",
       "      <td>-70</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>AFFX-BioC-5_at (endogenous control)</td>\n",
       "      <td>AFFX-BioC-5_at</td>\n",
       "      <td>88</td>\n",
       "      <td>283</td>\n",
       "      <td>309</td>\n",
       "      <td>12</td>\n",
       "      <td>168</td>\n",
       "      <td>71</td>\n",
       "      <td>55</td>\n",
       "      <td>-2</td>\n",
       "      <td>...</td>\n",
       "      <td>132</td>\n",
       "      <td>318</td>\n",
       "      <td>325</td>\n",
       "      <td>392</td>\n",
       "      <td>241</td>\n",
       "      <td>193</td>\n",
       "      <td>312</td>\n",
       "      <td>230</td>\n",
       "      <td>330</td>\n",
       "      <td>337</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>AFFX-BioC-3_at (endogenous control)</td>\n",
       "      <td>AFFX-BioC-3_at</td>\n",
       "      <td>-295</td>\n",
       "      <td>-264</td>\n",
       "      <td>-376</td>\n",
       "      <td>-419</td>\n",
       "      <td>-230</td>\n",
       "      <td>-272</td>\n",
       "      <td>-399</td>\n",
       "      <td>-541</td>\n",
       "      <td>...</td>\n",
       "      <td>-377</td>\n",
       "      <td>-209</td>\n",
       "      <td>-396</td>\n",
       "      <td>-324</td>\n",
       "      <td>-191</td>\n",
       "      <td>-51</td>\n",
       "      <td>-139</td>\n",
       "      <td>-367</td>\n",
       "      <td>-188</td>\n",
       "      <td>-407</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 40 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                      Gene Description Gene Accession Number    1    2    3  \\\n",
       "0  AFFX-BioB-5_at (endogenous control)        AFFX-BioB-5_at -214 -139  -76   \n",
       "1  AFFX-BioB-M_at (endogenous control)        AFFX-BioB-M_at -153  -73  -49   \n",
       "2  AFFX-BioB-3_at (endogenous control)        AFFX-BioB-3_at  -58   -1 -307   \n",
       "3  AFFX-BioC-5_at (endogenous control)        AFFX-BioC-5_at   88  283  309   \n",
       "4  AFFX-BioC-3_at (endogenous control)        AFFX-BioC-3_at -295 -264 -376   \n",
       "\n",
       "     4    5    6    7    8  ...   35   36   37   38   28   29   30   31   32  \\\n",
       "0 -135 -106 -138  -72 -413  ...    7 -213  -25  -72   -4   15 -318  -32 -124   \n",
       "1 -114 -125  -85 -144 -260  ... -100 -252  -20 -139 -116 -114 -192  -49  -79   \n",
       "2  265  -76  215  238    7  ...  -57  136  124   -1 -125    2  -95   49  -37   \n",
       "3   12  168   71   55   -2  ...  132  318  325  392  241  193  312  230  330   \n",
       "4 -419 -230 -272 -399 -541  ... -377 -209 -396 -324 -191  -51 -139 -367 -188   \n",
       "\n",
       "    33  \n",
       "0 -135  \n",
       "1 -186  \n",
       "2  -70  \n",
       "3  337  \n",
       "4 -407  \n",
       "\n",
       "[5 rows x 40 columns]"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1 = [col for col in df.columns if \"call\" not in col]\n",
    "df = df[df1]\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "_cell_guid": "104d271b-e5f8-4e88-b163-6761ca6b5125",
    "_execution_state": "idle",
    "_uuid": "5c74edb93810a775d039f0de9b0bf620d3db3d56",
    "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>7119</th>\n",
       "      <th>7120</th>\n",
       "      <th>7121</th>\n",
       "      <th>7122</th>\n",
       "      <th>7123</th>\n",
       "      <th>7124</th>\n",
       "      <th>7125</th>\n",
       "      <th>7126</th>\n",
       "      <th>7127</th>\n",
       "      <th>7128</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Gene Description</th>\n",
       "      <td>AFFX-BioB-5_at (endogenous control)</td>\n",
       "      <td>AFFX-BioB-M_at (endogenous control)</td>\n",
       "      <td>AFFX-BioB-3_at (endogenous control)</td>\n",
       "      <td>AFFX-BioC-5_at (endogenous control)</td>\n",
       "      <td>AFFX-BioC-3_at (endogenous control)</td>\n",
       "      <td>AFFX-BioDn-5_at (endogenous control)</td>\n",
       "      <td>AFFX-BioDn-3_at (endogenous control)</td>\n",
       "      <td>AFFX-CreX-5_at (endogenous control)</td>\n",
       "      <td>AFFX-CreX-3_at (endogenous control)</td>\n",
       "      <td>AFFX-BioB-5_st (endogenous control)</td>\n",
       "      <td>...</td>\n",
       "      <td>Transcription factor Stat5b (stat5b) mRNA</td>\n",
       "      <td>Breast epithelial antigen BA46 mRNA</td>\n",
       "      <td>GB DEF = Calcium/calmodulin-dependent protein ...</td>\n",
       "      <td>TUBULIN ALPHA-4 CHAIN</td>\n",
       "      <td>CYP4B1 Cytochrome P450; subfamily IVB; polypep...</td>\n",
       "      <td>PTGER3 Prostaglandin E receptor 3 (subtype EP3...</td>\n",
       "      <td>HMG2 High-mobility group (nonhistone chromosom...</td>\n",
       "      <td>RB1 Retinoblastoma 1 (including osteosarcoma)</td>\n",
       "      <td>GB DEF = Glycophorin Sta (type A) exons 3 and ...</td>\n",
       "      <td>GB DEF = mRNA (clone 1A7)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Gene Accession Number</th>\n",
       "      <td>AFFX-BioB-5_at</td>\n",
       "      <td>AFFX-BioB-M_at</td>\n",
       "      <td>AFFX-BioB-3_at</td>\n",
       "      <td>AFFX-BioC-5_at</td>\n",
       "      <td>AFFX-BioC-3_at</td>\n",
       "      <td>AFFX-BioDn-5_at</td>\n",
       "      <td>AFFX-BioDn-3_at</td>\n",
       "      <td>AFFX-CreX-5_at</td>\n",
       "      <td>AFFX-CreX-3_at</td>\n",
       "      <td>AFFX-BioB-5_st</td>\n",
       "      <td>...</td>\n",
       "      <td>U48730_at</td>\n",
       "      <td>U58516_at</td>\n",
       "      <td>U73738_at</td>\n",
       "      <td>X06956_at</td>\n",
       "      <td>X16699_at</td>\n",
       "      <td>X83863_at</td>\n",
       "      <td>Z17240_at</td>\n",
       "      <td>L49218_f_at</td>\n",
       "      <td>M71243_f_at</td>\n",
       "      <td>Z78285_f_at</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-214</td>\n",
       "      <td>-153</td>\n",
       "      <td>-58</td>\n",
       "      <td>88</td>\n",
       "      <td>-295</td>\n",
       "      <td>-558</td>\n",
       "      <td>199</td>\n",
       "      <td>-176</td>\n",
       "      <td>252</td>\n",
       "      <td>206</td>\n",
       "      <td>...</td>\n",
       "      <td>185</td>\n",
       "      <td>511</td>\n",
       "      <td>-125</td>\n",
       "      <td>389</td>\n",
       "      <td>-37</td>\n",
       "      <td>793</td>\n",
       "      <td>329</td>\n",
       "      <td>36</td>\n",
       "      <td>191</td>\n",
       "      <td>-37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-139</td>\n",
       "      <td>-73</td>\n",
       "      <td>-1</td>\n",
       "      <td>283</td>\n",
       "      <td>-264</td>\n",
       "      <td>-400</td>\n",
       "      <td>-330</td>\n",
       "      <td>-168</td>\n",
       "      <td>101</td>\n",
       "      <td>74</td>\n",
       "      <td>...</td>\n",
       "      <td>169</td>\n",
       "      <td>837</td>\n",
       "      <td>-36</td>\n",
       "      <td>442</td>\n",
       "      <td>-17</td>\n",
       "      <td>782</td>\n",
       "      <td>295</td>\n",
       "      <td>11</td>\n",
       "      <td>76</td>\n",
       "      <td>-14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-76</td>\n",
       "      <td>-49</td>\n",
       "      <td>-307</td>\n",
       "      <td>309</td>\n",
       "      <td>-376</td>\n",
       "      <td>-650</td>\n",
       "      <td>33</td>\n",
       "      <td>-367</td>\n",
       "      <td>206</td>\n",
       "      <td>-215</td>\n",
       "      <td>...</td>\n",
       "      <td>315</td>\n",
       "      <td>1199</td>\n",
       "      <td>33</td>\n",
       "      <td>168</td>\n",
       "      <td>52</td>\n",
       "      <td>1138</td>\n",
       "      <td>777</td>\n",
       "      <td>41</td>\n",
       "      <td>228</td>\n",
       "      <td>-41</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 7129 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                      0     \\\n",
       "Gene Description       AFFX-BioB-5_at (endogenous control)   \n",
       "Gene Accession Number                       AFFX-BioB-5_at   \n",
       "1                                                     -214   \n",
       "2                                                     -139   \n",
       "3                                                      -76   \n",
       "\n",
       "                                                      1     \\\n",
       "Gene Description       AFFX-BioB-M_at (endogenous control)   \n",
       "Gene Accession Number                       AFFX-BioB-M_at   \n",
       "1                                                     -153   \n",
       "2                                                      -73   \n",
       "3                                                      -49   \n",
       "\n",
       "                                                      2     \\\n",
       "Gene Description       AFFX-BioB-3_at (endogenous control)   \n",
       "Gene Accession Number                       AFFX-BioB-3_at   \n",
       "1                                                      -58   \n",
       "2                                                       -1   \n",
       "3                                                     -307   \n",
       "\n",
       "                                                      3     \\\n",
       "Gene Description       AFFX-BioC-5_at (endogenous control)   \n",
       "Gene Accession Number                       AFFX-BioC-5_at   \n",
       "1                                                       88   \n",
       "2                                                      283   \n",
       "3                                                      309   \n",
       "\n",
       "                                                      4     \\\n",
       "Gene Description       AFFX-BioC-3_at (endogenous control)   \n",
       "Gene Accession Number                       AFFX-BioC-3_at   \n",
       "1                                                     -295   \n",
       "2                                                     -264   \n",
       "3                                                     -376   \n",
       "\n",
       "                                                       5     \\\n",
       "Gene Description       AFFX-BioDn-5_at (endogenous control)   \n",
       "Gene Accession Number                       AFFX-BioDn-5_at   \n",
       "1                                                      -558   \n",
       "2                                                      -400   \n",
       "3                                                      -650   \n",
       "\n",
       "                                                       6     \\\n",
       "Gene Description       AFFX-BioDn-3_at (endogenous control)   \n",
       "Gene Accession Number                       AFFX-BioDn-3_at   \n",
       "1                                                       199   \n",
       "2                                                      -330   \n",
       "3                                                        33   \n",
       "\n",
       "                                                      7     \\\n",
       "Gene Description       AFFX-CreX-5_at (endogenous control)   \n",
       "Gene Accession Number                       AFFX-CreX-5_at   \n",
       "1                                                     -176   \n",
       "2                                                     -168   \n",
       "3                                                     -367   \n",
       "\n",
       "                                                      8     \\\n",
       "Gene Description       AFFX-CreX-3_at (endogenous control)   \n",
       "Gene Accession Number                       AFFX-CreX-3_at   \n",
       "1                                                      252   \n",
       "2                                                      101   \n",
       "3                                                      206   \n",
       "\n",
       "                                                      9     ...  \\\n",
       "Gene Description       AFFX-BioB-5_st (endogenous control)  ...   \n",
       "Gene Accession Number                       AFFX-BioB-5_st  ...   \n",
       "1                                                      206  ...   \n",
       "2                                                       74  ...   \n",
       "3                                                     -215  ...   \n",
       "\n",
       "                                                            7119  \\\n",
       "Gene Description       Transcription factor Stat5b (stat5b) mRNA   \n",
       "Gene Accession Number                                  U48730_at   \n",
       "1                                                            185   \n",
       "2                                                            169   \n",
       "3                                                            315   \n",
       "\n",
       "                                                      7120  \\\n",
       "Gene Description       Breast epithelial antigen BA46 mRNA   \n",
       "Gene Accession Number                            U58516_at   \n",
       "1                                                      511   \n",
       "2                                                      837   \n",
       "3                                                     1199   \n",
       "\n",
       "                                                                    7121  \\\n",
       "Gene Description       GB DEF = Calcium/calmodulin-dependent protein ...   \n",
       "Gene Accession Number                                          U73738_at   \n",
       "1                                                                   -125   \n",
       "2                                                                    -36   \n",
       "3                                                                     33   \n",
       "\n",
       "                                        7122  \\\n",
       "Gene Description       TUBULIN ALPHA-4 CHAIN   \n",
       "Gene Accession Number              X06956_at   \n",
       "1                                        389   \n",
       "2                                        442   \n",
       "3                                        168   \n",
       "\n",
       "                                                                    7123  \\\n",
       "Gene Description       CYP4B1 Cytochrome P450; subfamily IVB; polypep...   \n",
       "Gene Accession Number                                          X16699_at   \n",
       "1                                                                    -37   \n",
       "2                                                                    -17   \n",
       "3                                                                     52   \n",
       "\n",
       "                                                                    7124  \\\n",
       "Gene Description       PTGER3 Prostaglandin E receptor 3 (subtype EP3...   \n",
       "Gene Accession Number                                          X83863_at   \n",
       "1                                                                    793   \n",
       "2                                                                    782   \n",
       "3                                                                   1138   \n",
       "\n",
       "                                                                    7125  \\\n",
       "Gene Description       HMG2 High-mobility group (nonhistone chromosom...   \n",
       "Gene Accession Number                                          Z17240_at   \n",
       "1                                                                    329   \n",
       "2                                                                    295   \n",
       "3                                                                    777   \n",
       "\n",
       "                                                                7126  \\\n",
       "Gene Description       RB1 Retinoblastoma 1 (including osteosarcoma)   \n",
       "Gene Accession Number                                    L49218_f_at   \n",
       "1                                                                 36   \n",
       "2                                                                 11   \n",
       "3                                                                 41   \n",
       "\n",
       "                                                                    7127  \\\n",
       "Gene Description       GB DEF = Glycophorin Sta (type A) exons 3 and ...   \n",
       "Gene Accession Number                                        M71243_f_at   \n",
       "1                                                                    191   \n",
       "2                                                                     76   \n",
       "3                                                                    228   \n",
       "\n",
       "                                            7128  \n",
       "Gene Description       GB DEF = mRNA (clone 1A7)  \n",
       "Gene Accession Number                Z78285_f_at  \n",
       "1                                            -37  \n",
       "2                                            -14  \n",
       "3                                            -41  \n",
       "\n",
       "[5 rows x 7129 columns]"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.T.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "_cell_guid": "5d3b1a7a-111e-4e06-b676-b0ac14613509",
    "_execution_state": "idle",
    "_uuid": "b7129d92fb2e29b5d83928c2ecf323d4be6d9dd7"
   },
   "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>7119</th>\n",
       "      <th>7120</th>\n",
       "      <th>7121</th>\n",
       "      <th>7122</th>\n",
       "      <th>7123</th>\n",
       "      <th>7124</th>\n",
       "      <th>7125</th>\n",
       "      <th>7126</th>\n",
       "      <th>7127</th>\n",
       "      <th>7128</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-214</td>\n",
       "      <td>-153</td>\n",
       "      <td>-58</td>\n",
       "      <td>88</td>\n",
       "      <td>-295</td>\n",
       "      <td>-558</td>\n",
       "      <td>199</td>\n",
       "      <td>-176</td>\n",
       "      <td>252</td>\n",
       "      <td>206</td>\n",
       "      <td>...</td>\n",
       "      <td>185</td>\n",
       "      <td>511</td>\n",
       "      <td>-125</td>\n",
       "      <td>389</td>\n",
       "      <td>-37</td>\n",
       "      <td>793</td>\n",
       "      <td>329</td>\n",
       "      <td>36</td>\n",
       "      <td>191</td>\n",
       "      <td>-37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-139</td>\n",
       "      <td>-73</td>\n",
       "      <td>-1</td>\n",
       "      <td>283</td>\n",
       "      <td>-264</td>\n",
       "      <td>-400</td>\n",
       "      <td>-330</td>\n",
       "      <td>-168</td>\n",
       "      <td>101</td>\n",
       "      <td>74</td>\n",
       "      <td>...</td>\n",
       "      <td>169</td>\n",
       "      <td>837</td>\n",
       "      <td>-36</td>\n",
       "      <td>442</td>\n",
       "      <td>-17</td>\n",
       "      <td>782</td>\n",
       "      <td>295</td>\n",
       "      <td>11</td>\n",
       "      <td>76</td>\n",
       "      <td>-14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-76</td>\n",
       "      <td>-49</td>\n",
       "      <td>-307</td>\n",
       "      <td>309</td>\n",
       "      <td>-376</td>\n",
       "      <td>-650</td>\n",
       "      <td>33</td>\n",
       "      <td>-367</td>\n",
       "      <td>206</td>\n",
       "      <td>-215</td>\n",
       "      <td>...</td>\n",
       "      <td>315</td>\n",
       "      <td>1199</td>\n",
       "      <td>33</td>\n",
       "      <td>168</td>\n",
       "      <td>52</td>\n",
       "      <td>1138</td>\n",
       "      <td>777</td>\n",
       "      <td>41</td>\n",
       "      <td>228</td>\n",
       "      <td>-41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-135</td>\n",
       "      <td>-114</td>\n",
       "      <td>265</td>\n",
       "      <td>12</td>\n",
       "      <td>-419</td>\n",
       "      <td>-585</td>\n",
       "      <td>158</td>\n",
       "      <td>-253</td>\n",
       "      <td>49</td>\n",
       "      <td>31</td>\n",
       "      <td>...</td>\n",
       "      <td>240</td>\n",
       "      <td>835</td>\n",
       "      <td>218</td>\n",
       "      <td>174</td>\n",
       "      <td>-110</td>\n",
       "      <td>627</td>\n",
       "      <td>170</td>\n",
       "      <td>-50</td>\n",
       "      <td>126</td>\n",
       "      <td>-91</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>-106</td>\n",
       "      <td>-125</td>\n",
       "      <td>-76</td>\n",
       "      <td>168</td>\n",
       "      <td>-230</td>\n",
       "      <td>-284</td>\n",
       "      <td>4</td>\n",
       "      <td>-122</td>\n",
       "      <td>70</td>\n",
       "      <td>252</td>\n",
       "      <td>...</td>\n",
       "      <td>156</td>\n",
       "      <td>649</td>\n",
       "      <td>57</td>\n",
       "      <td>504</td>\n",
       "      <td>-26</td>\n",
       "      <td>250</td>\n",
       "      <td>314</td>\n",
       "      <td>14</td>\n",
       "      <td>56</td>\n",
       "      <td>-25</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 7129 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   0     1     2    3     4     5     6     7    8     9     ... 7119  7120  \\\n",
       "1  -214  -153   -58   88  -295  -558   199  -176  252   206  ...  185   511   \n",
       "2  -139   -73    -1  283  -264  -400  -330  -168  101    74  ...  169   837   \n",
       "3   -76   -49  -307  309  -376  -650    33  -367  206  -215  ...  315  1199   \n",
       "4  -135  -114   265   12  -419  -585   158  -253   49    31  ...  240   835   \n",
       "5  -106  -125   -76  168  -230  -284     4  -122   70   252  ...  156   649   \n",
       "\n",
       "   7121 7122  7123  7124 7125 7126 7127 7128  \n",
       "1  -125  389   -37   793  329   36  191  -37  \n",
       "2   -36  442   -17   782  295   11   76  -14  \n",
       "3    33  168    52  1138  777   41  228  -41  \n",
       "4   218  174  -110   627  170  -50  126  -91  \n",
       "5    57  504   -26   250  314   14   56  -25  \n",
       "\n",
       "[5 rows x 7129 columns]"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = df.T\n",
    "df2 = df.drop(['Gene Description','Gene Accession Number'],axis=0)\n",
    "df2.index = pd.to_numeric(df2.index)\n",
    "df2.sort_index(inplace=True)\n",
    "df2.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "48ea0072-e5f1-4b88-8535-2a112e555583",
    "_uuid": "3f00bf1f6e335330df93acccbcc033179f443f93"
   },
   "source": [
    "Now see here we are having just 38 columns which are our respective people"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "_cell_guid": "e78a7f04-7335-429f-8205-588379407661",
    "_execution_state": "idle",
    "_uuid": "a0e16c4988ddb26883817718257c62b08b64cd9d",
    "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>7120</th>\n",
       "      <th>7121</th>\n",
       "      <th>7122</th>\n",
       "      <th>7123</th>\n",
       "      <th>7124</th>\n",
       "      <th>7125</th>\n",
       "      <th>7126</th>\n",
       "      <th>7127</th>\n",
       "      <th>7128</th>\n",
       "      <th>cat</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-214</td>\n",
       "      <td>-153</td>\n",
       "      <td>-58</td>\n",
       "      <td>88</td>\n",
       "      <td>-295</td>\n",
       "      <td>-558</td>\n",
       "      <td>199</td>\n",
       "      <td>-176</td>\n",
       "      <td>252</td>\n",
       "      <td>206</td>\n",
       "      <td>...</td>\n",
       "      <td>511</td>\n",
       "      <td>-125</td>\n",
       "      <td>389</td>\n",
       "      <td>-37</td>\n",
       "      <td>793</td>\n",
       "      <td>329</td>\n",
       "      <td>36</td>\n",
       "      <td>191</td>\n",
       "      <td>-37</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-139</td>\n",
       "      <td>-73</td>\n",
       "      <td>-1</td>\n",
       "      <td>283</td>\n",
       "      <td>-264</td>\n",
       "      <td>-400</td>\n",
       "      <td>-330</td>\n",
       "      <td>-168</td>\n",
       "      <td>101</td>\n",
       "      <td>74</td>\n",
       "      <td>...</td>\n",
       "      <td>837</td>\n",
       "      <td>-36</td>\n",
       "      <td>442</td>\n",
       "      <td>-17</td>\n",
       "      <td>782</td>\n",
       "      <td>295</td>\n",
       "      <td>11</td>\n",
       "      <td>76</td>\n",
       "      <td>-14</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-76</td>\n",
       "      <td>-49</td>\n",
       "      <td>-307</td>\n",
       "      <td>309</td>\n",
       "      <td>-376</td>\n",
       "      <td>-650</td>\n",
       "      <td>33</td>\n",
       "      <td>-367</td>\n",
       "      <td>206</td>\n",
       "      <td>-215</td>\n",
       "      <td>...</td>\n",
       "      <td>1199</td>\n",
       "      <td>33</td>\n",
       "      <td>168</td>\n",
       "      <td>52</td>\n",
       "      <td>1138</td>\n",
       "      <td>777</td>\n",
       "      <td>41</td>\n",
       "      <td>228</td>\n",
       "      <td>-41</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3 rows × 7130 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     0    1    2    3    4    5    6    7    8    9  ...  7120  7121  7122  \\\n",
       "1 -214 -153  -58   88 -295 -558  199 -176  252  206  ...   511  -125   389   \n",
       "2 -139  -73   -1  283 -264 -400 -330 -168  101   74  ...   837   -36   442   \n",
       "3  -76  -49 -307  309 -376 -650   33 -367  206 -215  ...  1199    33   168   \n",
       "\n",
       "   7123  7124  7125  7126  7127  7128  cat  \n",
       "1   -37   793   329    36   191   -37    0  \n",
       "2   -17   782   295    11    76   -14    0  \n",
       "3    52  1138   777    41   228   -41    0  \n",
       "\n",
       "[3 rows x 7130 columns]"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2['cat'] = list(pd.read_csv('../input/actual.csv')[:38]['cancer'])\n",
    "dic = {'ALL':0,'AML':1}\n",
    "df2.replace(dic,inplace=True)\n",
    "df2.head(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "7feb4b2d-4a18-483e-a88e-42a4aff5a197",
    "_uuid": "0fe4088ffacf7fbe7e9a40257427279e54cbd47b"
   },
   "source": [
    "# PCA analysis\n",
    "#### Principal Component Analysis (or PCA) uses linear algebra to transform the dataset into a compressed form. Generally this is called a data reduction technique. A property of PCA is that you can choose the number of dimensions or principal component in the transformed result.\n",
    "\n",
    "#### In this case we will use it to analyse the feature importanace.\n",
    "\n",
    "We will reduce the data into two features."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "fb96b296-edc0-4d8f-a1f1-c5d298e16ea3",
    "_uuid": "8c8e2f4b4d092857510bad11e8f68e5534a48a84"
   },
   "source": [
    "![](https://media.giphy.com/media/QA44U3zAnH0ha/source.gif)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "_cell_guid": "8e03ed05-df24-48d7-bc14-2a79b06c4172",
    "_execution_state": "idle",
    "_uuid": "0e5e2c7dba3a40dcc749885c7bcd8950e2b9f8a6"
   },
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "X_std = StandardScaler().fit_transform(df2.drop('cat',axis=1))\n",
    "\n",
    "from sklearn.decomposition import PCA as sklearnPCA\n",
    "sklearn_pca = sklearnPCA(n_components=30)\n",
    "Y_sklearn = sklearn_pca.fit_transform(X_std)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "_cell_guid": "535a317c-a430-4629-aa66-412ea0a55f2e",
    "_uuid": "adb296628a683655e426b0c358a1e60e534114de"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "cum_sum = sklearn_pca.explained_variance_ratio_.cumsum()\n",
    "\n",
    "sklearn_pca.explained_variance_ratio_[:10].sum()\n",
    "\n",
    "cum_sum = cum_sum*100\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(8,8))\n",
    "plt.bar(range(30), cum_sum, label='Cumulative _Sum_of_Explained _Varaince', color = 'b',alpha=0.5)\n",
    "plt.title(\"Around 95% of variance is explained by the Fisrt 30 colmns \");"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "_cell_guid": "61735619-4929-4464-8064-a99c0b5d2e25",
    "_execution_state": "idle",
    "_uuid": "d64a9f74065a26fbd57c242aed643401c4b9536e"
   },
   "outputs": [],
   "source": [
    "X_reduced2 = Y_sklearn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "_cell_guid": "2da6ea95-32ae-4813-a3b2-bc259feab616",
    "_execution_state": "idle",
    "_uuid": "9b901ab1cbaa93d24687e12f7881ca7dfeabd699"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
       "       0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1])"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2.cat.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "_cell_guid": "d7addc0f-0749-460f-84a5-165e5c6446e2",
    "_execution_state": "idle",
    "_uuid": "a563f4d00368dd2951352dec54c5be2bca38daa1"
   },
   "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>21</th>\n",
       "      <th>22</th>\n",
       "      <th>23</th>\n",
       "      <th>24</th>\n",
       "      <th>25</th>\n",
       "      <th>26</th>\n",
       "      <th>27</th>\n",
       "      <th>28</th>\n",
       "      <th>29</th>\n",
       "      <th>cat</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>18.544799</td>\n",
       "      <td>5.826943</td>\n",
       "      <td>-19.442313</td>\n",
       "      <td>-11.267566</td>\n",
       "      <td>5.303304</td>\n",
       "      <td>19.100937</td>\n",
       "      <td>-9.447438</td>\n",
       "      <td>0.915309</td>\n",
       "      <td>-8.665217</td>\n",
       "      <td>-7.912754</td>\n",
       "      <td>...</td>\n",
       "      <td>6.323438</td>\n",
       "      <td>5.630107</td>\n",
       "      <td>-4.538090</td>\n",
       "      <td>-14.672156</td>\n",
       "      <td>4.899964</td>\n",
       "      <td>9.335462</td>\n",
       "      <td>-12.552358</td>\n",
       "      <td>2.484640</td>\n",
       "      <td>12.217094</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-7.341389</td>\n",
       "      <td>10.008663</td>\n",
       "      <td>11.795698</td>\n",
       "      <td>3.920823</td>\n",
       "      <td>-13.999094</td>\n",
       "      <td>-18.636026</td>\n",
       "      <td>4.687060</td>\n",
       "      <td>8.812418</td>\n",
       "      <td>1.372678</td>\n",
       "      <td>-11.690751</td>\n",
       "      <td>...</td>\n",
       "      <td>-11.423156</td>\n",
       "      <td>-10.669290</td>\n",
       "      <td>-5.364779</td>\n",
       "      <td>-20.280586</td>\n",
       "      <td>7.685347</td>\n",
       "      <td>23.676243</td>\n",
       "      <td>5.216321</td>\n",
       "      <td>17.828269</td>\n",
       "      <td>-1.015588</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>52.852146</td>\n",
       "      <td>11.390630</td>\n",
       "      <td>-29.548566</td>\n",
       "      <td>-38.800348</td>\n",
       "      <td>-22.866689</td>\n",
       "      <td>-9.458113</td>\n",
       "      <td>-22.880735</td>\n",
       "      <td>20.995112</td>\n",
       "      <td>4.038731</td>\n",
       "      <td>-5.260765</td>\n",
       "      <td>...</td>\n",
       "      <td>1.847696</td>\n",
       "      <td>5.073579</td>\n",
       "      <td>0.797486</td>\n",
       "      <td>0.995521</td>\n",
       "      <td>-0.354742</td>\n",
       "      <td>-0.093794</td>\n",
       "      <td>2.769450</td>\n",
       "      <td>-2.868344</td>\n",
       "      <td>7.354274</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3 rows × 31 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           0          1          2          3          4          5  \\\n",
       "0  18.544799   5.826943 -19.442313 -11.267566   5.303304  19.100937   \n",
       "1  -7.341389  10.008663  11.795698   3.920823 -13.999094 -18.636026   \n",
       "2  52.852146  11.390630 -29.548566 -38.800348 -22.866689  -9.458113   \n",
       "\n",
       "           6          7         8          9  ...         21         22  \\\n",
       "0  -9.447438   0.915309 -8.665217  -7.912754  ...   6.323438   5.630107   \n",
       "1   4.687060   8.812418  1.372678 -11.690751  ... -11.423156 -10.669290   \n",
       "2 -22.880735  20.995112  4.038731  -5.260765  ...   1.847696   5.073579   \n",
       "\n",
       "         23         24        25         26         27         28         29  \\\n",
       "0 -4.538090 -14.672156  4.899964   9.335462 -12.552358   2.484640  12.217094   \n",
       "1 -5.364779 -20.280586  7.685347  23.676243   5.216321  17.828269  -1.015588   \n",
       "2  0.797486   0.995521 -0.354742  -0.093794   2.769450  -2.868344   7.354274   \n",
       "\n",
       "   cat  \n",
       "0    0  \n",
       "1    0  \n",
       "2    0  \n",
       "\n",
       "[3 rows x 31 columns]"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train = pd.DataFrame(X_reduced2)\n",
    "train['cat'] =  df2['cat'].reset_index().cat\n",
    "train.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "_cell_guid": "27be69c9-5ca5-4707-a96d-e7c557908a0d",
    "_uuid": "dd2a20673859324f785a9b2563bd6dde95074590"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[]"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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": [
    "from sklearn.decomposition import PCA as sklearnPCA\n",
    "sklearn_pca = sklearnPCA(n_components=3)\n",
    "X_reduced  = sklearn_pca.fit_transform(X_std)\n",
    "Y=train['cat']\n",
    "from mpl_toolkits.mplot3d import Axes3D\n",
    "plt.clf()\n",
    "fig = plt.figure(1, figsize=(10,6 ))\n",
    "ax = Axes3D(fig, elev=-150, azim=110,)\n",
    "ax.scatter(X_reduced[:, 0], X_reduced[:, 1], X_reduced[:, 2], c=Y,cmap=plt.cm.Paired,linewidths=10)\n",
    "ax.set_title(\"First three PCA directions\")\n",
    "ax.set_xlabel(\"1st eigenvector\")\n",
    "ax.w_xaxis.set_ticklabels([])\n",
    "ax.set_ylabel(\"2nd eigenvector\")\n",
    "ax.w_yaxis.set_ticklabels([])\n",
    "ax.set_zlabel(\"3rd eigenvector\")\n",
    "ax.w_zaxis.set_ticklabels([])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "_cell_guid": "dc398fce-0912-4752-a292-7e10ddf645c8",
    "_execution_state": "idle",
    "_uuid": "d465f4a493189b72735135795c807d8c534799a4"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'This The 2D Transformation of above graph ')"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "fig = plt.figure(1, figsize=(10,6))\n",
    "plt.scatter(X_reduced[:, 0],  X_reduced[:, 1], c=df2['cat'],cmap=plt.cm.Paired,linewidths=10)\n",
    "plt.annotate('See The Brown Cluster',xy=(20,-20),xytext=(9,8),arrowprops=dict(facecolor='black', shrink=0.05))\n",
    "#plt.scatter(test_reduced[:, 0],  test_reduced[:, 1],c='r')\n",
    "plt.title(\"This The 2D Transformation of above graph \")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "a83b6281-3068-488d-aa37-23d6a6ad214d",
    "_uuid": "373ce7025addb2a5a63bd25a5bdf8a9c8a7b50ae",
    "collapsed": true
   },
   "source": [
    "PCA played a very important role here as we got our clusters and we differentiated cancer disease from it. We can think of this as a cluster.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "_cell_guid": "0d1166d5-8c30-462a-afb5-295b9cfe40e8",
    "_execution_state": "idle",
    "_uuid": "91153b20bf72975234779908face0bb72158a0e3"
   },
   "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>Gene Description</th>\n",
       "      <th>Gene Accession Number</th>\n",
       "      <th>39</th>\n",
       "      <th>call</th>\n",
       "      <th>40</th>\n",
       "      <th>call.1</th>\n",
       "      <th>42</th>\n",
       "      <th>call.2</th>\n",
       "      <th>47</th>\n",
       "      <th>call.3</th>\n",
       "      <th>...</th>\n",
       "      <th>65</th>\n",
       "      <th>call.29</th>\n",
       "      <th>66</th>\n",
       "      <th>call.30</th>\n",
       "      <th>63</th>\n",
       "      <th>call.31</th>\n",
       "      <th>64</th>\n",
       "      <th>call.32</th>\n",
       "      <th>62</th>\n",
       "      <th>call.33</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>AFFX-BioB-5_at (endogenous control)</td>\n",
       "      <td>AFFX-BioB-5_at</td>\n",
       "      <td>-342</td>\n",
       "      <td>A</td>\n",
       "      <td>-87</td>\n",
       "      <td>A</td>\n",
       "      <td>22</td>\n",
       "      <td>A</td>\n",
       "      <td>-243</td>\n",
       "      <td>A</td>\n",
       "      <td>...</td>\n",
       "      <td>-62</td>\n",
       "      <td>A</td>\n",
       "      <td>-58</td>\n",
       "      <td>A</td>\n",
       "      <td>-161</td>\n",
       "      <td>A</td>\n",
       "      <td>-48</td>\n",
       "      <td>A</td>\n",
       "      <td>-176</td>\n",
       "      <td>A</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>AFFX-BioB-M_at (endogenous control)</td>\n",
       "      <td>AFFX-BioB-M_at</td>\n",
       "      <td>-200</td>\n",
       "      <td>A</td>\n",
       "      <td>-248</td>\n",
       "      <td>A</td>\n",
       "      <td>-153</td>\n",
       "      <td>A</td>\n",
       "      <td>-218</td>\n",
       "      <td>A</td>\n",
       "      <td>...</td>\n",
       "      <td>-198</td>\n",
       "      <td>A</td>\n",
       "      <td>-217</td>\n",
       "      <td>A</td>\n",
       "      <td>-215</td>\n",
       "      <td>A</td>\n",
       "      <td>-531</td>\n",
       "      <td>A</td>\n",
       "      <td>-284</td>\n",
       "      <td>A</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>AFFX-BioB-3_at (endogenous control)</td>\n",
       "      <td>AFFX-BioB-3_at</td>\n",
       "      <td>41</td>\n",
       "      <td>A</td>\n",
       "      <td>262</td>\n",
       "      <td>A</td>\n",
       "      <td>17</td>\n",
       "      <td>A</td>\n",
       "      <td>-163</td>\n",
       "      <td>A</td>\n",
       "      <td>...</td>\n",
       "      <td>-5</td>\n",
       "      <td>A</td>\n",
       "      <td>63</td>\n",
       "      <td>A</td>\n",
       "      <td>-46</td>\n",
       "      <td>A</td>\n",
       "      <td>-124</td>\n",
       "      <td>A</td>\n",
       "      <td>-81</td>\n",
       "      <td>A</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3 rows × 70 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                      Gene Description Gene Accession Number   39 call   40  \\\n",
       "0  AFFX-BioB-5_at (endogenous control)        AFFX-BioB-5_at -342    A  -87   \n",
       "1  AFFX-BioB-M_at (endogenous control)        AFFX-BioB-M_at -200    A -248   \n",
       "2  AFFX-BioB-3_at (endogenous control)        AFFX-BioB-3_at   41    A  262   \n",
       "\n",
       "  call.1   42 call.2   47 call.3  ...   65 call.29   66 call.30   63 call.31  \\\n",
       "0      A   22      A -243      A  ...  -62       A  -58       A -161       A   \n",
       "1      A -153      A -218      A  ... -198       A -217       A -215       A   \n",
       "2      A   17      A -163      A  ...   -5       A   63       A  -46       A   \n",
       "\n",
       "    64 call.32   62 call.33  \n",
       "0  -48       A -176       A  \n",
       "1 -531       A -284       A  \n",
       "2 -124       A  -81       A  \n",
       "\n",
       "[3 rows x 70 columns]"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test = pd.read_csv('../input/data_set_ALL_AML_independent.csv')\n",
    "\n",
    "test.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "_cell_guid": "f7641b76-54af-46cd-994d-bd878983dc2b",
    "_execution_state": "idle",
    "_uuid": "2fc675af96c21fcdeac6879faf5104276b0369eb",
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "test1 = [col for col in test.columns if \"call\" not in col]\n",
    "test = test[test1]\n",
    "test = test.T\n",
    "test2 = test.drop(['Gene Description','Gene Accession Number'],axis=0)\n",
    "test2.index = pd.to_numeric(test2.index)\n",
    "test2.sort_index(inplace=True)\n",
    "#test2['cat'] = list(pd.read_csv('actual.csv')[39:63]['cancer'])\n",
    "#dic = {'ALL':0,'AML':1}\n",
    "#test2.replace(dic,inplace=True)\n",
    "#test2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "_cell_guid": "cfb541b3-8868-455c-beda-b8f03b6d9553",
    "_execution_state": "idle",
    "_uuid": "6cb7fab8fac34c26de031ec1ea6dac0ec3dba7cc"
   },
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "Y_std = StandardScaler().fit_transform(test2)\n",
    "\n",
    "from sklearn.decomposition import PCA as sklearnPCA\n",
    "sklearn_pca = sklearnPCA(n_components=30)\n",
    "test_reduced = sklearn_pca.fit_transform(Y_std)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "_cell_guid": "6534db57-7887-498c-a376-f93f0953c6ec",
    "_execution_state": "idle",
    "_uuid": "f1b31636267639403912b2cbddf2757f5af63828"
   },
   "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>20</th>\n",
       "      <th>21</th>\n",
       "      <th>22</th>\n",
       "      <th>23</th>\n",
       "      <th>24</th>\n",
       "      <th>25</th>\n",
       "      <th>26</th>\n",
       "      <th>27</th>\n",
       "      <th>28</th>\n",
       "      <th>29</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>15.954485</td>\n",
       "      <td>-19.998344</td>\n",
       "      <td>34.450819</td>\n",
       "      <td>17.107265</td>\n",
       "      <td>-14.499998</td>\n",
       "      <td>-8.547010</td>\n",
       "      <td>-27.011502</td>\n",
       "      <td>0.893890</td>\n",
       "      <td>28.760512</td>\n",
       "      <td>-11.761661</td>\n",
       "      <td>...</td>\n",
       "      <td>12.181201</td>\n",
       "      <td>-1.930483</td>\n",
       "      <td>-5.577590</td>\n",
       "      <td>0.894264</td>\n",
       "      <td>1.708536</td>\n",
       "      <td>9.909458</td>\n",
       "      <td>1.429599</td>\n",
       "      <td>-1.368022</td>\n",
       "      <td>1.928401</td>\n",
       "      <td>1.395269</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>22.690962</td>\n",
       "      <td>-17.487396</td>\n",
       "      <td>-28.042792</td>\n",
       "      <td>26.803369</td>\n",
       "      <td>-12.457292</td>\n",
       "      <td>12.704222</td>\n",
       "      <td>35.329067</td>\n",
       "      <td>20.194094</td>\n",
       "      <td>43.420264</td>\n",
       "      <td>28.037366</td>\n",
       "      <td>...</td>\n",
       "      <td>3.671538</td>\n",
       "      <td>3.183151</td>\n",
       "      <td>0.897244</td>\n",
       "      <td>0.113003</td>\n",
       "      <td>0.708739</td>\n",
       "      <td>1.036280</td>\n",
       "      <td>4.038658</td>\n",
       "      <td>0.821032</td>\n",
       "      <td>1.411593</td>\n",
       "      <td>1.400042</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-36.249932</td>\n",
       "      <td>13.988727</td>\n",
       "      <td>-8.697312</td>\n",
       "      <td>10.075214</td>\n",
       "      <td>6.067020</td>\n",
       "      <td>-3.396009</td>\n",
       "      <td>3.167661</td>\n",
       "      <td>13.144338</td>\n",
       "      <td>-3.474793</td>\n",
       "      <td>-5.183876</td>\n",
       "      <td>...</td>\n",
       "      <td>-4.041182</td>\n",
       "      <td>-8.274781</td>\n",
       "      <td>13.887254</td>\n",
       "      <td>-6.468673</td>\n",
       "      <td>-6.302675</td>\n",
       "      <td>-11.033458</td>\n",
       "      <td>-5.795327</td>\n",
       "      <td>-7.775089</td>\n",
       "      <td>-8.975263</td>\n",
       "      <td>11.639648</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3 rows × 30 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          0          1          2          3          4          5   \\\n",
       "0  15.954485 -19.998344  34.450819  17.107265 -14.499998  -8.547010   \n",
       "1  22.690962 -17.487396 -28.042792  26.803369 -12.457292  12.704222   \n",
       "2 -36.249932  13.988727  -8.697312  10.075214   6.067020  -3.396009   \n",
       "\n",
       "          6          7          8          9   ...         20        21  \\\n",
       "0 -27.011502   0.893890  28.760512 -11.761661  ...  12.181201 -1.930483   \n",
       "1  35.329067  20.194094  43.420264  28.037366  ...   3.671538  3.183151   \n",
       "2   3.167661  13.144338  -3.474793  -5.183876  ...  -4.041182 -8.274781   \n",
       "\n",
       "          22        23        24         25        26        27        28  \\\n",
       "0  -5.577590  0.894264  1.708536   9.909458  1.429599 -1.368022  1.928401   \n",
       "1   0.897244  0.113003  0.708739   1.036280  4.038658  0.821032  1.411593   \n",
       "2  13.887254 -6.468673 -6.302675 -11.033458 -5.795327 -7.775089 -8.975263   \n",
       "\n",
       "          29  \n",
       "0   1.395269  \n",
       "1   1.400042  \n",
       "2  11.639648  \n",
       "\n",
       "[3 rows x 30 columns]"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_set = pd.DataFrame(test_reduced)\n",
    "\n",
    "test_set.head(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "43055fec-2184-464e-bbc6-ea66669a13c8",
    "_uuid": "7b858d1164d7db14407109cfa8abc1eba6bf0bfc"
   },
   "source": [
    "### As clear from graph it is useful to use KNN to find the required weather that it is ALL or AML.\n",
    "Depending upon distance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "_cell_guid": "17debc2f-3b5e-480a-b7a1-2928e4fd452c",
    "_uuid": "f15b5fccae415388d190b2d222838c5e6a7927c6"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f543390e390>"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "train.drop('cat',axis=1).plot(kind='hist',figsize=(8,10))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "_cell_guid": "5f8b206f-37be-410f-a97d-786e99f556e2",
    "_execution_state": "idle",
    "_uuid": "b6abf986dbce7f82d4cb19eaa197227e9d16f2f1"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KNeighborsClassifier(algorithm='auto', leaf_size=30, metric='minkowski',\n",
       "                     metric_params=None, n_jobs=None, n_neighbors=10, p=2,\n",
       "                     weights='uniform')"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "clf= KNeighborsClassifier(n_neighbors=10,)\n",
    "clf.fit(train.drop('cat',axis=1),train['cat'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "_cell_guid": "21939081-d78a-4ca9-8964-b2874e187e3b",
    "_execution_state": "idle",
    "_uuid": "51e0837807e7efd8b00c6188489c80a47462e6f7"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[19,  1],\n",
       "       [13,  1]])"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pred = clf.predict(test_set)\n",
    "\n",
    "pateint = pd.read_csv('../input/actual.csv')['cancer'][38:]\n",
    "\n",
    "true = pateint.replace(dic)\n",
    "\n",
    "import sklearn\n",
    "sklearn.metrics.confusion_matrix(true, pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "_cell_guid": "ffbd3c3b-37ac-4eb2-a22a-0322e2dd7729",
    "_uuid": "8bd59e923d83c2407dd6099836d6136c5cefd342"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[13  7]\n",
      " [10  4]]\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.tree import DecisionTreeClassifier\n",
    "clf = DecisionTreeClassifier(min_samples_split=2)\n",
    "clf.fit(train.drop('cat',axis=1),train['cat'])\n",
    "pred = clf.predict(test_set)\n",
    "true = pateint.replace(dic)\n",
    "print(sklearn.metrics.confusion_matrix(true, pred))\n",
    "print()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "_cell_guid": "e4083947-f596-40f8-94d1-e6a6fc91b9a1",
    "_uuid": "55899c7ce4faaddf81a317794b2f71acf2227ccf"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[16  4]\n",
      " [13  1]]\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn import svm\n",
    "\n",
    "clf=svm.SVC(kernel='linear')\n",
    "clf.fit(train.drop('cat',axis=1),train['cat'])\n",
    "pred = clf.predict(test_set)\n",
    "\n",
    "pateint = pd.read_csv('../input/actual.csv')['cancer'][38:]\n",
    "\n",
    "true = pateint.replace(dic)\n",
    "\n",
    "print(sklearn.metrics.confusion_matrix(true, pred))\n",
    "print()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "efe5bdd7-73f7-46fe-a80f-035e312a1245",
    "_uuid": "096b00b187d77c41b082a6fd1924669271348f75"
   },
   "source": [
    "# Here the Red Points(means test set) is plotted on the training set to just give the illusion that which one falls in which category"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "_cell_guid": "a9594a44-03cc-4b9d-aad5-00d5d365431c",
    "_execution_state": "idle",
    "_uuid": "f0b6966fa9c28b70bc49f59dded6b38d635baf6c"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x7f5433b008d0>"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(1, figsize=(14,6))\n",
    "plt.scatter(X_reduced[:, 0],  X_reduced[:, 1], c=df2['cat'],cmap=plt.cm.Paired,alpha=0.7,linewidths=7)\n",
    "plt.scatter(test_reduced[:, 0],  test_reduced[:, 1],c='r',linewidths=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "_cell_guid": "dc1767ec-357e-4a4a-a432-30c85b33743e",
    "_uuid": "5545753cfa4bf6aaa276e95b7eff6373fb14b92a"
   },
   "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": "3.6.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}