k_means_demo_kafka.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# K-Means Algorithm Demo\n",
    "\n",
    "_Source: 🤖[Homemade Machine Learning](https://github.com/trekhleb/homemade-machine-learning) repository_\n",
    "\n",
    "> ☝Before moving on with this demo you might want to take a look at:\n",
    "> - 📗[Math behind the K-Means Algorithm](https://github.com/trekhleb/homemade-machine-learning/tree/master/homemade/k_means)\n",
    "> - ⚙️[K-Means Algorithm Source Code](https://github.com/trekhleb/homemade-machine-learning/blob/master/homemade/k_means/k_means.py)\n",
    "\n",
    "**K-means clustering** aims to partition _n_ observations into _K_ clusters in which each observation belongs to the cluster with the nearest mean, serving as a prototype of the cluster.\n",
    "\n",
    "![title](./mesarch.png)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# To make debugging of logistic_regression module easier we enable imported modules autoreloading feature.\n",
    "# By doing this you may change the code of logistic_regression library and all these changes will be available here.\n",
    "%load_ext autoreload\n",
    "%autoreload 2\n",
    "\n",
    "# Add project root folder to module loading paths.\n",
    "import sys\n",
    "sys.path.append('./')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Import Dependencies\n",
    "\n",
    "- [pandas](https://pandas.pydata.org/) - library that we will use for loading and displaying the data in a table\n",
    "- [numpy](http://www.numpy.org/) - library that we will use for linear algebra operations\n",
    "- [matplotlib](https://matplotlib.org/) - library that we will use for plotting the data\n",
    "- [k_means](https://github.com/trekhleb/homemade-machine-learning/blob/master/homemade/k_means/k_means.py) - custom implementation of K-Means algorithm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import 3rd party dependencies.\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "from mpl_toolkits.mplot3d import Axes3D\n",
    "# Import custom k-means implementation.\n",
    "from homemade.k_means import KMeans"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Load the Data\n",
    "\n",
    "数据来源于MES 制造管理过程中设备的IoT 收集。通过5G 设备收集数据流进入系统进行运算。\n",
    "平台使用KAFKA 作为前端物联网数据的收集。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Defaulting to user installation because normal site-packages is not writeable\n",
      "Requirement already satisfied: kafka-python in /home/jovyan/.local/lib/python3.7/site-packages (2.0.2)\n"
     ]
    }
   ],
   "source": [
    "import sys\n",
    "!{sys.executable} -m pip install kafka-python"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "total lines: 8520\n"
     ]
    },
    {
     "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>filename</th>\n",
       "      <th>column</th>\n",
       "      <th>avr</th>\n",
       "      <th>var</th>\n",
       "      <th>std</th>\n",
       "      <th>skew</th>\n",
       "      <th>kern</th>\n",
       "      <th>logviola</th>\n",
       "      <th>rollingvarmax</th>\n",
       "      <th>DFA</th>\n",
       "      <th>rollingvarmean</th>\n",
       "      <th>violmax</th>\n",
       "      <th>violmean</th>\n",
       "      <th>maxpeaksqt</th>\n",
       "      <th>Unnamed: 14</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5550.csv</td>\n",
       "      <td>millinletTemp</td>\n",
       "      <td>863.744110</td>\n",
       "      <td>3.087366</td>\n",
       "      <td>1.757090</td>\n",
       "      <td>0.127967</td>\n",
       "      <td>-0.118871</td>\n",
       "      <td>0.018943</td>\n",
       "      <td>5.161280</td>\n",
       "      <td>0.979514</td>\n",
       "      <td>2.479970</td>\n",
       "      <td>5.883527</td>\n",
       "      <td>1.481444</td>\n",
       "      <td>1.679927</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5551.csv</td>\n",
       "      <td>millinletTemp</td>\n",
       "      <td>863.797703</td>\n",
       "      <td>2.767410</td>\n",
       "      <td>1.663554</td>\n",
       "      <td>0.062153</td>\n",
       "      <td>-0.348086</td>\n",
       "      <td>0.017787</td>\n",
       "      <td>4.394257</td>\n",
       "      <td>0.891497</td>\n",
       "      <td>2.273234</td>\n",
       "      <td>4.447802</td>\n",
       "      <td>1.412332</td>\n",
       "      <td>1.526001</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>5552.csv</td>\n",
       "      <td>millinletTemp</td>\n",
       "      <td>863.744015</td>\n",
       "      <td>2.903806</td>\n",
       "      <td>1.704056</td>\n",
       "      <td>0.084013</td>\n",
       "      <td>0.487382</td>\n",
       "      <td>0.018174</td>\n",
       "      <td>6.377349</td>\n",
       "      <td>0.996471</td>\n",
       "      <td>2.552628</td>\n",
       "      <td>5.901430</td>\n",
       "      <td>1.388344</td>\n",
       "      <td>1.781452</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>5553.csv</td>\n",
       "      <td>millinletTemp</td>\n",
       "      <td>863.882932</td>\n",
       "      <td>3.492361</td>\n",
       "      <td>1.868786</td>\n",
       "      <td>0.193423</td>\n",
       "      <td>-0.116656</td>\n",
       "      <td>0.019932</td>\n",
       "      <td>5.258999</td>\n",
       "      <td>0.966817</td>\n",
       "      <td>2.947490</td>\n",
       "      <td>6.082835</td>\n",
       "      <td>1.575984</td>\n",
       "      <td>1.599961</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5554.csv</td>\n",
       "      <td>millinletTemp</td>\n",
       "      <td>864.011664</td>\n",
       "      <td>3.143501</td>\n",
       "      <td>1.772992</td>\n",
       "      <td>0.217091</td>\n",
       "      <td>0.037112</td>\n",
       "      <td>0.018888</td>\n",
       "      <td>5.576391</td>\n",
       "      <td>1.016957</td>\n",
       "      <td>2.920016</td>\n",
       "      <td>5.584815</td>\n",
       "      <td>1.478182</td>\n",
       "      <td>1.594840</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>5555.csv</td>\n",
       "      <td>millinletTemp</td>\n",
       "      <td>864.837067</td>\n",
       "      <td>3.544050</td>\n",
       "      <td>1.882565</td>\n",
       "      <td>-0.152580</td>\n",
       "      <td>-0.281152</td>\n",
       "      <td>0.020046</td>\n",
       "      <td>5.576391</td>\n",
       "      <td>1.027948</td>\n",
       "      <td>2.877514</td>\n",
       "      <td>5.256140</td>\n",
       "      <td>1.575533</td>\n",
       "      <td>1.400383</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>5556.csv</td>\n",
       "      <td>millinletTemp</td>\n",
       "      <td>865.341135</td>\n",
       "      <td>3.867873</td>\n",
       "      <td>1.966691</td>\n",
       "      <td>-0.359354</td>\n",
       "      <td>-0.046214</td>\n",
       "      <td>0.020892</td>\n",
       "      <td>6.777299</td>\n",
       "      <td>1.087793</td>\n",
       "      <td>3.435109</td>\n",
       "      <td>6.019159</td>\n",
       "      <td>1.658891</td>\n",
       "      <td>2.556618</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>5557.csv</td>\n",
       "      <td>millinletTemp</td>\n",
       "      <td>865.719821</td>\n",
       "      <td>3.542313</td>\n",
       "      <td>1.882103</td>\n",
       "      <td>0.094590</td>\n",
       "      <td>0.230677</td>\n",
       "      <td>0.020064</td>\n",
       "      <td>6.846417</td>\n",
       "      <td>1.055729</td>\n",
       "      <td>2.957667</td>\n",
       "      <td>6.699753</td>\n",
       "      <td>1.555244</td>\n",
       "      <td>1.723979</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>5558.csv</td>\n",
       "      <td>millinletTemp</td>\n",
       "      <td>866.099481</td>\n",
       "      <td>2.565872</td>\n",
       "      <td>1.601834</td>\n",
       "      <td>-0.097358</td>\n",
       "      <td>-0.379971</td>\n",
       "      <td>0.017066</td>\n",
       "      <td>4.637326</td>\n",
       "      <td>1.027707</td>\n",
       "      <td>2.432753</td>\n",
       "      <td>4.199982</td>\n",
       "      <td>1.357467</td>\n",
       "      <td>1.318181</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>5559.csv</td>\n",
       "      <td>millinletTemp</td>\n",
       "      <td>866.593505</td>\n",
       "      <td>2.572016</td>\n",
       "      <td>1.603751</td>\n",
       "      <td>0.113377</td>\n",
       "      <td>-0.044126</td>\n",
       "      <td>0.017057</td>\n",
       "      <td>5.013850</td>\n",
       "      <td>0.974966</td>\n",
       "      <td>2.433880</td>\n",
       "      <td>4.784490</td>\n",
       "      <td>1.359781</td>\n",
       "      <td>1.320124</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>5560.csv</td>\n",
       "      <td>millinletTemp</td>\n",
       "      <td>866.434904</td>\n",
       "      <td>3.036685</td>\n",
       "      <td>1.742609</td>\n",
       "      <td>0.240725</td>\n",
       "      <td>0.336081</td>\n",
       "      <td>0.018526</td>\n",
       "      <td>5.261743</td>\n",
       "      <td>0.944697</td>\n",
       "      <td>2.547749</td>\n",
       "      <td>5.822851</td>\n",
       "      <td>1.446750</td>\n",
       "      <td>2.087255</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>5561.csv</td>\n",
       "      <td>millinletTemp</td>\n",
       "      <td>866.046859</td>\n",
       "      <td>3.146230</td>\n",
       "      <td>1.773762</td>\n",
       "      <td>-0.105827</td>\n",
       "      <td>0.237061</td>\n",
       "      <td>0.018844</td>\n",
       "      <td>5.040798</td>\n",
       "      <td>0.980405</td>\n",
       "      <td>2.477905</td>\n",
       "      <td>5.678154</td>\n",
       "      <td>1.472044</td>\n",
       "      <td>1.440391</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>5562.csv</td>\n",
       "      <td>millinletTemp</td>\n",
       "      <td>866.401694</td>\n",
       "      <td>2.718848</td>\n",
       "      <td>1.648893</td>\n",
       "      <td>-0.400684</td>\n",
       "      <td>0.166929</td>\n",
       "      <td>0.017525</td>\n",
       "      <td>5.040798</td>\n",
       "      <td>0.966670</td>\n",
       "      <td>2.155163</td>\n",
       "      <td>5.287476</td>\n",
       "      <td>1.356055</td>\n",
       "      <td>1.715741</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>5563.csv</td>\n",
       "      <td>millinletTemp</td>\n",
       "      <td>866.582125</td>\n",
       "      <td>2.194428</td>\n",
       "      <td>1.481360</td>\n",
       "      <td>-0.126376</td>\n",
       "      <td>0.577957</td>\n",
       "      <td>0.015722</td>\n",
       "      <td>3.702664</td>\n",
       "      <td>0.928382</td>\n",
       "      <td>1.812006</td>\n",
       "      <td>5.172229</td>\n",
       "      <td>1.207491</td>\n",
       "      <td>1.089145</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>5564.csv</td>\n",
       "      <td>millinletTemp</td>\n",
       "      <td>866.780873</td>\n",
       "      <td>2.903218</td>\n",
       "      <td>1.703883</td>\n",
       "      <td>-0.320021</td>\n",
       "      <td>0.604934</td>\n",
       "      <td>0.018078</td>\n",
       "      <td>5.341680</td>\n",
       "      <td>0.920927</td>\n",
       "      <td>2.366524</td>\n",
       "      <td>6.346094</td>\n",
       "      <td>1.397269</td>\n",
       "      <td>1.496185</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>5565.csv</td>\n",
       "      <td>millinletTemp</td>\n",
       "      <td>866.668855</td>\n",
       "      <td>3.172729</td>\n",
       "      <td>1.781216</td>\n",
       "      <td>0.078174</td>\n",
       "      <td>0.094080</td>\n",
       "      <td>0.018894</td>\n",
       "      <td>4.834593</td>\n",
       "      <td>1.075597</td>\n",
       "      <td>2.634076</td>\n",
       "      <td>5.135749</td>\n",
       "      <td>1.461496</td>\n",
       "      <td>1.607952</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>5566.csv</td>\n",
       "      <td>millinletTemp</td>\n",
       "      <td>867.685379</td>\n",
       "      <td>3.975865</td>\n",
       "      <td>1.993957</td>\n",
       "      <td>-0.087448</td>\n",
       "      <td>-0.387168</td>\n",
       "      <td>0.021143</td>\n",
       "      <td>5.907520</td>\n",
       "      <td>1.102939</td>\n",
       "      <td>2.513231</td>\n",
       "      <td>5.209100</td>\n",
       "      <td>1.672559</td>\n",
       "      <td>3.561170</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>5567.csv</td>\n",
       "      <td>millinletTemp</td>\n",
       "      <td>869.253052</td>\n",
       "      <td>2.497958</td>\n",
       "      <td>1.580493</td>\n",
       "      <td>0.057334</td>\n",
       "      <td>-0.090561</td>\n",
       "      <td>0.016745</td>\n",
       "      <td>4.614997</td>\n",
       "      <td>1.119110</td>\n",
       "      <td>2.336501</td>\n",
       "      <td>4.560658</td>\n",
       "      <td>1.296803</td>\n",
       "      <td>1.303943</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>5568.csv</td>\n",
       "      <td>millinletTemp</td>\n",
       "      <td>868.906937</td>\n",
       "      <td>2.045604</td>\n",
       "      <td>1.430246</td>\n",
       "      <td>0.000778</td>\n",
       "      <td>0.448541</td>\n",
       "      <td>0.015106</td>\n",
       "      <td>3.250630</td>\n",
       "      <td>0.889760</td>\n",
       "      <td>1.804514</td>\n",
       "      <td>5.381321</td>\n",
       "      <td>1.185978</td>\n",
       "      <td>1.162005</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>5569.csv</td>\n",
       "      <td>millinletTemp</td>\n",
       "      <td>868.108070</td>\n",
       "      <td>2.415520</td>\n",
       "      <td>1.554194</td>\n",
       "      <td>-0.083070</td>\n",
       "      <td>-0.065428</td>\n",
       "      <td>0.016476</td>\n",
       "      <td>4.072921</td>\n",
       "      <td>0.906219</td>\n",
       "      <td>1.991125</td>\n",
       "      <td>4.537639</td>\n",
       "      <td>1.278630</td>\n",
       "      <td>1.222104</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>5570.csv</td>\n",
       "      <td>millinletTemp</td>\n",
       "      <td>868.161901</td>\n",
       "      <td>2.660426</td>\n",
       "      <td>1.631081</td>\n",
       "      <td>0.056719</td>\n",
       "      <td>-0.013766</td>\n",
       "      <td>0.017301</td>\n",
       "      <td>3.904818</td>\n",
       "      <td>0.885570</td>\n",
       "      <td>2.186491</td>\n",
       "      <td>5.712833</td>\n",
       "      <td>1.361266</td>\n",
       "      <td>1.651429</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>5571.csv</td>\n",
       "      <td>millinletTemp</td>\n",
       "      <td>869.086220</td>\n",
       "      <td>2.924593</td>\n",
       "      <td>1.710144</td>\n",
       "      <td>-0.099482</td>\n",
       "      <td>0.150748</td>\n",
       "      <td>0.018114</td>\n",
       "      <td>5.679605</td>\n",
       "      <td>1.090505</td>\n",
       "      <td>2.469832</td>\n",
       "      <td>5.072283</td>\n",
       "      <td>1.417011</td>\n",
       "      <td>1.680579</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>5572.csv</td>\n",
       "      <td>millinletTemp</td>\n",
       "      <td>869.370831</td>\n",
       "      <td>2.895887</td>\n",
       "      <td>1.701730</td>\n",
       "      <td>-0.087676</td>\n",
       "      <td>-0.191609</td>\n",
       "      <td>0.018024</td>\n",
       "      <td>5.679605</td>\n",
       "      <td>1.033093</td>\n",
       "      <td>2.332446</td>\n",
       "      <td>4.948880</td>\n",
       "      <td>1.454200</td>\n",
       "      <td>1.746521</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>5573.csv</td>\n",
       "      <td>millinletTemp</td>\n",
       "      <td>869.310623</td>\n",
       "      <td>3.318874</td>\n",
       "      <td>1.821778</td>\n",
       "      <td>-0.190288</td>\n",
       "      <td>-0.355473</td>\n",
       "      <td>0.019285</td>\n",
       "      <td>6.840115</td>\n",
       "      <td>1.043302</td>\n",
       "      <td>3.050518</td>\n",
       "      <td>5.526879</td>\n",
       "      <td>1.537351</td>\n",
       "      <td>1.421699</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>5574.csv</td>\n",
       "      <td>millinletTemp</td>\n",
       "      <td>869.507853</td>\n",
       "      <td>2.910251</td>\n",
       "      <td>1.705946</td>\n",
       "      <td>-0.204506</td>\n",
       "      <td>0.324207</td>\n",
       "      <td>0.018014</td>\n",
       "      <td>6.840115</td>\n",
       "      <td>1.076852</td>\n",
       "      <td>2.692759</td>\n",
       "      <td>5.696494</td>\n",
       "      <td>1.391327</td>\n",
       "      <td>2.359679</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>5575.csv</td>\n",
       "      <td>millinletTemp</td>\n",
       "      <td>869.965120</td>\n",
       "      <td>2.387022</td>\n",
       "      <td>1.544999</td>\n",
       "      <td>0.209613</td>\n",
       "      <td>0.539019</td>\n",
       "      <td>0.016282</td>\n",
       "      <td>4.558924</td>\n",
       "      <td>0.957362</td>\n",
       "      <td>2.155662</td>\n",
       "      <td>5.346453</td>\n",
       "      <td>1.264072</td>\n",
       "      <td>1.388368</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>5576.csv</td>\n",
       "      <td>millinletTemp</td>\n",
       "      <td>870.372403</td>\n",
       "      <td>2.769363</td>\n",
       "      <td>1.664140</td>\n",
       "      <td>-0.165226</td>\n",
       "      <td>0.391603</td>\n",
       "      <td>0.017586</td>\n",
       "      <td>6.405404</td>\n",
       "      <td>0.982233</td>\n",
       "      <td>2.330465</td>\n",
       "      <td>5.688714</td>\n",
       "      <td>1.379775</td>\n",
       "      <td>1.528666</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>5577.csv</td>\n",
       "      <td>millinletTemp</td>\n",
       "      <td>868.498298</td>\n",
       "      <td>4.139657</td>\n",
       "      <td>2.034615</td>\n",
       "      <td>0.300409</td>\n",
       "      <td>-0.053456</td>\n",
       "      <td>0.021510</td>\n",
       "      <td>6.405404</td>\n",
       "      <td>1.043725</td>\n",
       "      <td>2.328389</td>\n",
       "      <td>6.812547</td>\n",
       "      <td>1.718772</td>\n",
       "      <td>3.737583</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>5578.csv</td>\n",
       "      <td>millinletTemp</td>\n",
       "      <td>866.387635</td>\n",
       "      <td>3.076718</td>\n",
       "      <td>1.754058</td>\n",
       "      <td>-0.213950</td>\n",
       "      <td>0.236203</td>\n",
       "      <td>0.018582</td>\n",
       "      <td>5.528051</td>\n",
       "      <td>1.059554</td>\n",
       "      <td>2.424239</td>\n",
       "      <td>5.388023</td>\n",
       "      <td>1.439811</td>\n",
       "      <td>1.656716</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>5579.csv</td>\n",
       "      <td>millinletTemp</td>\n",
       "      <td>866.245708</td>\n",
       "      <td>2.400508</td>\n",
       "      <td>1.549357</td>\n",
       "      <td>-0.108982</td>\n",
       "      <td>-0.347534</td>\n",
       "      <td>0.016468</td>\n",
       "      <td>3.381707</td>\n",
       "      <td>0.923013</td>\n",
       "      <td>2.002648</td>\n",
       "      <td>4.461931</td>\n",
       "      <td>1.322507</td>\n",
       "      <td>1.494718</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    filename         column         avr       var       std      skew  \\\n",
       "0   5550.csv  millinletTemp  863.744110  3.087366  1.757090  0.127967   \n",
       "1   5551.csv  millinletTemp  863.797703  2.767410  1.663554  0.062153   \n",
       "2   5552.csv  millinletTemp  863.744015  2.903806  1.704056  0.084013   \n",
       "3   5553.csv  millinletTemp  863.882932  3.492361  1.868786  0.193423   \n",
       "4   5554.csv  millinletTemp  864.011664  3.143501  1.772992  0.217091   \n",
       "5   5555.csv  millinletTemp  864.837067  3.544050  1.882565 -0.152580   \n",
       "6   5556.csv  millinletTemp  865.341135  3.867873  1.966691 -0.359354   \n",
       "7   5557.csv  millinletTemp  865.719821  3.542313  1.882103  0.094590   \n",
       "8   5558.csv  millinletTemp  866.099481  2.565872  1.601834 -0.097358   \n",
       "9   5559.csv  millinletTemp  866.593505  2.572016  1.603751  0.113377   \n",
       "10  5560.csv  millinletTemp  866.434904  3.036685  1.742609  0.240725   \n",
       "11  5561.csv  millinletTemp  866.046859  3.146230  1.773762 -0.105827   \n",
       "12  5562.csv  millinletTemp  866.401694  2.718848  1.648893 -0.400684   \n",
       "13  5563.csv  millinletTemp  866.582125  2.194428  1.481360 -0.126376   \n",
       "14  5564.csv  millinletTemp  866.780873  2.903218  1.703883 -0.320021   \n",
       "15  5565.csv  millinletTemp  866.668855  3.172729  1.781216  0.078174   \n",
       "16  5566.csv  millinletTemp  867.685379  3.975865  1.993957 -0.087448   \n",
       "17  5567.csv  millinletTemp  869.253052  2.497958  1.580493  0.057334   \n",
       "18  5568.csv  millinletTemp  868.906937  2.045604  1.430246  0.000778   \n",
       "19  5569.csv  millinletTemp  868.108070  2.415520  1.554194 -0.083070   \n",
       "20  5570.csv  millinletTemp  868.161901  2.660426  1.631081  0.056719   \n",
       "21  5571.csv  millinletTemp  869.086220  2.924593  1.710144 -0.099482   \n",
       "22  5572.csv  millinletTemp  869.370831  2.895887  1.701730 -0.087676   \n",
       "23  5573.csv  millinletTemp  869.310623  3.318874  1.821778 -0.190288   \n",
       "24  5574.csv  millinletTemp  869.507853  2.910251  1.705946 -0.204506   \n",
       "25  5575.csv  millinletTemp  869.965120  2.387022  1.544999  0.209613   \n",
       "26  5576.csv  millinletTemp  870.372403  2.769363  1.664140 -0.165226   \n",
       "27  5577.csv  millinletTemp  868.498298  4.139657  2.034615  0.300409   \n",
       "28  5578.csv  millinletTemp  866.387635  3.076718  1.754058 -0.213950   \n",
       "29  5579.csv  millinletTemp  866.245708  2.400508  1.549357 -0.108982   \n",
       "\n",
       "        kern  logviola  rollingvarmax       DFA  rollingvarmean   violmax  \\\n",
       "0  -0.118871  0.018943       5.161280  0.979514        2.479970  5.883527   \n",
       "1  -0.348086  0.017787       4.394257  0.891497        2.273234  4.447802   \n",
       "2   0.487382  0.018174       6.377349  0.996471        2.552628  5.901430   \n",
       "3  -0.116656  0.019932       5.258999  0.966817        2.947490  6.082835   \n",
       "4   0.037112  0.018888       5.576391  1.016957        2.920016  5.584815   \n",
       "5  -0.281152  0.020046       5.576391  1.027948        2.877514  5.256140   \n",
       "6  -0.046214  0.020892       6.777299  1.087793        3.435109  6.019159   \n",
       "7   0.230677  0.020064       6.846417  1.055729        2.957667  6.699753   \n",
       "8  -0.379971  0.017066       4.637326  1.027707        2.432753  4.199982   \n",
       "9  -0.044126  0.017057       5.013850  0.974966        2.433880  4.784490   \n",
       "10  0.336081  0.018526       5.261743  0.944697        2.547749  5.822851   \n",
       "11  0.237061  0.018844       5.040798  0.980405        2.477905  5.678154   \n",
       "12  0.166929  0.017525       5.040798  0.966670        2.155163  5.287476   \n",
       "13  0.577957  0.015722       3.702664  0.928382        1.812006  5.172229   \n",
       "14  0.604934  0.018078       5.341680  0.920927        2.366524  6.346094   \n",
       "15  0.094080  0.018894       4.834593  1.075597        2.634076  5.135749   \n",
       "16 -0.387168  0.021143       5.907520  1.102939        2.513231  5.209100   \n",
       "17 -0.090561  0.016745       4.614997  1.119110        2.336501  4.560658   \n",
       "18  0.448541  0.015106       3.250630  0.889760        1.804514  5.381321   \n",
       "19 -0.065428  0.016476       4.072921  0.906219        1.991125  4.537639   \n",
       "20 -0.013766  0.017301       3.904818  0.885570        2.186491  5.712833   \n",
       "21  0.150748  0.018114       5.679605  1.090505        2.469832  5.072283   \n",
       "22 -0.191609  0.018024       5.679605  1.033093        2.332446  4.948880   \n",
       "23 -0.355473  0.019285       6.840115  1.043302        3.050518  5.526879   \n",
       "24  0.324207  0.018014       6.840115  1.076852        2.692759  5.696494   \n",
       "25  0.539019  0.016282       4.558924  0.957362        2.155662  5.346453   \n",
       "26  0.391603  0.017586       6.405404  0.982233        2.330465  5.688714   \n",
       "27 -0.053456  0.021510       6.405404  1.043725        2.328389  6.812547   \n",
       "28  0.236203  0.018582       5.528051  1.059554        2.424239  5.388023   \n",
       "29 -0.347534  0.016468       3.381707  0.923013        2.002648  4.461931   \n",
       "\n",
       "    violmean  maxpeaksqt  Unnamed: 14  \n",
       "0   1.481444    1.679927          NaN  \n",
       "1   1.412332    1.526001          NaN  \n",
       "2   1.388344    1.781452          NaN  \n",
       "3   1.575984    1.599961          NaN  \n",
       "4   1.478182    1.594840          NaN  \n",
       "5   1.575533    1.400383          NaN  \n",
       "6   1.658891    2.556618          NaN  \n",
       "7   1.555244    1.723979          NaN  \n",
       "8   1.357467    1.318181          NaN  \n",
       "9   1.359781    1.320124          NaN  \n",
       "10  1.446750    2.087255          NaN  \n",
       "11  1.472044    1.440391          NaN  \n",
       "12  1.356055    1.715741          NaN  \n",
       "13  1.207491    1.089145          NaN  \n",
       "14  1.397269    1.496185          NaN  \n",
       "15  1.461496    1.607952          NaN  \n",
       "16  1.672559    3.561170          NaN  \n",
       "17  1.296803    1.303943          NaN  \n",
       "18  1.185978    1.162005          NaN  \n",
       "19  1.278630    1.222104          NaN  \n",
       "20  1.361266    1.651429          NaN  \n",
       "21  1.417011    1.680579          NaN  \n",
       "22  1.454200    1.746521          NaN  \n",
       "23  1.537351    1.421699          NaN  \n",
       "24  1.391327    2.359679          NaN  \n",
       "25  1.264072    1.388368          NaN  \n",
       "26  1.379775    1.528666          NaN  \n",
       "27  1.718772    3.737583          NaN  \n",
       "28  1.439811    1.656716          NaN  \n",
       "29  1.322507    1.494718          NaN  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Load Data from kafka\n",
    "from kafka import KafkaConsumer\n",
    "consumer = KafkaConsumer(bootstrap_servers='my-cluster-kafka-bootstrap:9092')\n",
    "consumer.subscribe(['streamdata'])\n",
    "msg = next(consumer)\n",
    "\n",
    "from io import StringIO\n",
    "import pandas as pd\n",
    "s=str(msg.value,'utf-8')\n",
    "transdata = StringIO(s) \n",
    "data=pd.read_csv(transdata)\n",
    "\n",
    "# Load the data.\n",
    "#data = pd.read_csv('./dataresult12316079millinletTemp.csv')\n",
    "\n",
    "# Print the data table.\n",
    "print(\"total lines:\", data.size);\n",
    "data.head(30)\n",
    "# dftest = pd.read_json('[{\"solarcalhub.post_std\":\"1.756868546549593\",\"solarcalhub.post_maxsqt\":\"5.883527537856933\",\"column\":\"OUTDB_MIAN_DB_RAYTEK\",\"solarcalhub.post_rollingvarmax\":\"5.16128000993738\",\"solarcalhub.post_maxpeaksqt\":\"1.6799265331950723\",\"solarcalhub.post_logVolatility\":\"0.018942948994278525\",\"solarcalhub.post_dfa\":\"0.9795138496780909\"}, {\"solarcalhub.post_std\":\"3.1593507287581324\",\"solarcalhub.post_maxsqt\":\"10.972852487968265\",\"column\":\"MILLINLETTEMP\",\"solarcalhub.post_rollingvarmax\":\"13.852365591986425\",\"solarcalhub.post_maxpeaksqt\":\"3.290883111104079\",\"solarcalhub.post_logVolatility\":\"0.03236536445490581\",\"solarcalhub.post_dfa\":\"0.9471421533154122\"}, {\"solarcalhub.post_std\":\"0.09275625557712425\",\"solarcalhub.post_maxsqt\":\"0.2254905958673933\",\"column\":\"SR_POSITION\",\"solarcalhub.post_rollingvarmax\":\"0.013432607580830268\",\"solarcalhub.post_maxpeaksqt\":\"0\",\"solarcalhub.post_logVolatility\":\"0.28548045647562886\",\"solarcalhub.post_dfa\":\"1.5463645641102886\"}, {\"solarcalhub.post_std\":\"0.6722473149021259\",\"solarcalhub.post_maxsqt\":\"1.8000868835800616\",\"column\":\"CASTER_POOL_LEVEL\",\"solarcalhub.post_rollingvarmax\":\"0.7737658818281329\",\"solarcalhub.post_maxpeaksqt\":\"0.30843405116102063\",\"solarcalhub.post_logVolatility\":\"0.18638939239566696\",\"solarcalhub.post_dfa\":\"1.559439957307513\"}, {\"solarcalhub.post_std\":\"0.0015988026828093313\",\"solarcalhub.post_maxsqt\":\"0.004896569043872487\",\"column\":\"CASTER_LINESPEED\",\"solarcalhub.post_rollingvarmax\":\"4.107023411384234e-06\",\"solarcalhub.post_maxpeaksqt\":\"0\",\"solarcalhub.post_logVolatility\":\"0.003801594278138662\",\"solarcalhub.post_dfa\":\"0.7239542305546263\"}]')\n",
    "# dftest.head(30)\n",
    "# data.to_json(orient='index')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Plot the Data\n",
    "\n",
    "原生产数据特征分布"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 15840x10800 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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": [
    "# List of suppported Iris classes.\n",
    "# iris_types = ['SETOSA', 'VERSICOLOR', 'VIRGINICA']\n",
    "\n",
    "# Pick the Iris parameters for consideration.\n",
    "x_axis = 'DFA'\n",
    "y_axis = 'violmax'\n",
    "z_axis = 'maxpeaksqt'\n",
    "\n",
    "# Make the plot a little bit bigger than default one.\n",
    "plt.figure(figsize=(220, 150))\n",
    "fig = plt.figure()\n",
    "ax = fig.add_subplot(111, projection='3d')\n",
    "# Plot the scatter for every type of Iris.\n",
    "# This is the case when we know flower labels in advance.\n",
    "# plt.subplot(1, 2, 1)\n",
    "ax.scatter(data[x_axis][:], data[y_axis][:], data[z_axis][:], c='r', marker='o')\n",
    "\n",
    "ax.set_xlabel('DFA')\n",
    "ax.set_ylabel('violmax')\n",
    "ax.set_zlabel('maxpeaksqt')\n",
    "\n",
    "\n",
    "# Plot all subplots.\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Prepara the Data for Training\n",
    "\n",
    "Let's extract `petal_length` and `petal_width` data and form a training feature set."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Get total number of Iris examples.\n",
    "num_examples = data.shape[0]\n",
    "\n",
    "# Get features.\n",
    "x_train = data[[x_axis, y_axis, z_axis]].values.reshape((num_examples, 3))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Init and Train Logistic Regression Model\n",
    "\n",
    "> ☝🏻This is the place where you might want to play with model configuration.\n",
    "\n",
    "- `num_clusters` - number of clusters into which we want to split our training dataset.\n",
    "- `max_iterations` - maximum number of training iterations."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[  1.33486139  31.45011595  28.29669943]\n",
      " [  1.16601553  14.93333339  10.53802302]\n",
      " [  0.96853523   5.74722845   2.12218706]\n",
      " [  1.97639419 250.33266607   4.90527455]]\n"
     ]
    }
   ],
   "source": [
    "# Set K-Means parameters.\n",
    "num_clusters = 4  # Number of clusters into which we want to split our training dataset.\n",
    "max_iterations = 50  # maximum number of training iterations.\n",
    "\n",
    "# Init K-Means instance.\n",
    "k_means = KMeans(x_train, num_clusters)\n",
    "\n",
    "# Train K-Means instance.\n",
    "(centroids, closest_centroids_ids) = k_means.train(max_iterations)\n",
    "data_frame = pd.DataFrame(centroids,columns=[x_axis,y_axis,z_axis]) \n",
    "A = list(item for item in range(0,len(centroids)))\n",
    "data_frame['A'] = pd.Series(A, index=data_frame.index)\n",
    "dfsort = data_frame.sort_values(by=[x_axis])\n",
    "print(centroids)\n",
    "\n",
    "L=[chr(i) for i in range(97,97+len(centroids))]\n",
    "dfsort['L'] = pd.Series(L, index=dfsort.index)\n",
    "Level = list()\n",
    "for closeid, cetroid in enumerate(closest_centroids_ids):\n",
    "#     print(closeid)\n",
    "#     print(cetroid)\n",
    "    Level.append(dfsort[dfsort.A==cetroid[0]]['L'].values[0][0])\n",
    "\n",
    "\n",
    "data['L'] =  pd.Series(Level, index=data.index) \n",
    "data.to_csv('./dataresult12316079caster_pool_levelmark.csv')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Plot the Clustering Results\n",
    "\n",
    "Now let's plot the original Iris flow classification along with our unsupervised K-Means clusters to see how the algorithm performed."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 8640x3600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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": [
    "# List of suppported Iris classes.\n",
    "# iris_types = ['SETOSA', 'VERSICOLOR', 'VIRGINICA']\n",
    "\n",
    "# Pick the Iris parameters for consideration.\n",
    "x_axis = 'DFA'\n",
    "y_axis = 'violmax'\n",
    "z_axis = 'maxpeaksqt'\n",
    "\n",
    "# Make the plot a little bit bigger than default one.\n",
    "plt.figure(figsize=(120, 50))\n",
    "\n",
    "# Plot ACTUAL Iris flower classification.\n",
    "fig = plt.figure()\n",
    "ax = fig.add_subplot(111, projection='3d')\n",
    "\n",
    "for centroid_id, centroid in enumerate(centroids):\n",
    "    current_examples_indices = (closest_centroids_ids == centroid_id).flatten()\n",
    "#     print(current_examples_indices)\n",
    "    ax.scatter(\n",
    "        data[x_axis][current_examples_indices],\n",
    "        data[y_axis][current_examples_indices],\n",
    "        data[z_axis][current_examples_indices],\n",
    "        label='Cluster #' + str(centroid_id)\n",
    "    )\n",
    "\n",
    "# Plot clusters centroids.\n",
    "for centroid_id, centroid in enumerate(centroids):\n",
    "    ax.scatter(centroid[0], centroid[1], centroid[2], c='black', marker='x', linewidth=8)\n",
    "    \n",
    "ax.set_xlabel('DFA')\n",
    "ax.set_ylabel('violmax')\n",
    "ax.set_zlabel('maxpeaksqt')\n",
    "plt.title('K-Means Clusters')\n",
    "ax.legend()\n",
    "\n",
    "# Show all subplots.\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}