pca.ipynb

{
 "metadata": {
  "name": "",
  "signature": "sha256:8cbdeac071dd0f1ed4b2d44dc67f732ab335cf7b9e82affbafbc2d1b149978e1"
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "### PCA\u548c\u6838PCA"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "### \u7b2c\u4e00\u6b65\uff1a\u8bfb\u53d6\u6570\u636e"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "%pylab inline\n",
      "import pandas as pd\n",
      "import matplotlib.pylab as plt  \n",
      "from sklearn.decomposition import PCA\n",
      "df = pd.read_csv('iris.csv')\n",
      "df.head()\n",
      "#df.info()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Populating the interactive namespace from numpy and matplotlib\n"
       ]
      },
      {
       "html": [
        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
        "<table border=\"1\" class=\"dataframe\">\n",
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
        "      <th>Sepal_Length</th>\n",
        "      <th>Sepal_Width</th>\n",
        "      <th>Petal_Length</th>\n",
        "      <th>Petal_Width</th>\n",
        "      <th>Species</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>0</th>\n",
        "      <td> 5.1</td>\n",
        "      <td> 3.5</td>\n",
        "      <td> 1.4</td>\n",
        "      <td> 0.2</td>\n",
        "      <td> setosa</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>1</th>\n",
        "      <td> 4.9</td>\n",
        "      <td> 3.0</td>\n",
        "      <td> 1.4</td>\n",
        "      <td> 0.2</td>\n",
        "      <td> setosa</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>2</th>\n",
        "      <td> 4.7</td>\n",
        "      <td> 3.2</td>\n",
        "      <td> 1.3</td>\n",
        "      <td> 0.2</td>\n",
        "      <td> setosa</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>3</th>\n",
        "      <td> 4.6</td>\n",
        "      <td> 3.1</td>\n",
        "      <td> 1.5</td>\n",
        "      <td> 0.2</td>\n",
        "      <td> setosa</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>4</th>\n",
        "      <td> 5.0</td>\n",
        "      <td> 3.6</td>\n",
        "      <td> 1.4</td>\n",
        "      <td> 0.2</td>\n",
        "      <td> setosa</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "</div>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 1,
       "text": [
        "   Sepal_Length  Sepal_Width  Petal_Length  Petal_Width Species\n",
        "0           5.1          3.5           1.4          0.2  setosa\n",
        "1           4.9          3.0           1.4          0.2  setosa\n",
        "2           4.7          3.2           1.3          0.2  setosa\n",
        "3           4.6          3.1           1.5          0.2  setosa\n",
        "4           5.0          3.6           1.4          0.2  setosa"
       ]
      }
     ],
     "prompt_number": 1
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "### \u7b2c\u4e8c\u6b65\uff1aPCA"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "X = df.ix[:,:4]\n",
      "pca = PCA(n_components='mle') # \u81ea\u52a8\u9009\u62e9\u4e3b\u6210\u5206\u4e2a\u6570\n",
      "pca.fit(X)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 2,
       "text": [
        "PCA(copy=True, n_components='mle', whiten=False)"
       ]
      }
     ],
     "prompt_number": 2
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "pca.components_  # \u4e3b\u6210\u4efd\u8d1f\u8f7d\n",
      "pca.components_[0]# \u7b2c\u4e00\u4e3b\u6210\u5206\u7684\u8d1f\u8f7d"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 3,
       "text": [
        "array([ 0.36158968, -0.08226889,  0.85657211,  0.35884393])"
       ]
      }
     ],
     "prompt_number": 3
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "pca.explained_variance_"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 4,
       "text": [
        "array([ 4.19667516,  0.24062861,  0.07800042])"
       ]
      }
     ],
     "prompt_number": 4
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# \u65b9\u5dee\u8d21\u732e\uff0c\u788e\u77f3\u56fe\n",
      "variance = pca.explained_variance_ratio_\n",
      "readable_variance = variance * (1/variance[0])\n",
      "component = pca.n_components_\n",
      "plt.plot(range(component), readable_variance)\n",
      "plt.show()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAEACAYAAABI5zaHAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nFxGJof8HBoJegZ6qOCUAAAAASUVORK5CYII=\n",
       "text": [
        "<matplotlib.figure.Figure at 0x109453f10>"
       ]
      }
     ],
     "prompt_number": 5
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "score = pca.transform(X) # \u4e3b\u6210\u5206\u5f97\u5206\n",
      "score = pd.concat([pd.DataFrame(score[:,:2]),df['Species']],axis=1)\n",
      "score.columns = ['var1','var2','species']\n",
      "score.head()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
        "<table border=\"1\" class=\"dataframe\">\n",
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
        "      <th>var1</th>\n",
        "      <th>var2</th>\n",
        "      <th>species</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>0</th>\n",
        "      <td>-2.684207</td>\n",
        "      <td>-0.326607</td>\n",
        "      <td> setosa</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>1</th>\n",
        "      <td>-2.715391</td>\n",
        "      <td> 0.169557</td>\n",
        "      <td> setosa</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>2</th>\n",
        "      <td>-2.889820</td>\n",
        "      <td> 0.137346</td>\n",
        "      <td> setosa</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>3</th>\n",
        "      <td>-2.746437</td>\n",
        "      <td> 0.311124</td>\n",
        "      <td> setosa</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>4</th>\n",
        "      <td>-2.728593</td>\n",
        "      <td>-0.333925</td>\n",
        "      <td> setosa</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "</div>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 6,
       "text": [
        "       var1      var2 species\n",
        "0 -2.684207 -0.326607  setosa\n",
        "1 -2.715391  0.169557  setosa\n",
        "2 -2.889820  0.137346  setosa\n",
        "3 -2.746437  0.311124  setosa\n",
        "4 -2.728593 -0.333925  setosa"
       ]
      }
     ],
     "prompt_number": 6
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from  ggplot import *\n",
      "ggplot(score,aes('var1','var2',color='species')) \\\n",
      "    + geom_point()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAqUAAAHzCAYAAAAZ52RFAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
       "text": [
        "<matplotlib.figure.Figure at 0x10d282a10>"
       ]
      },
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 7,
       "text": [
        "<ggplot: (282231425)>"
       ]
      }
     ],
     "prompt_number": 7
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "X.corr()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
        "<table border=\"1\" class=\"dataframe\">\n",
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
        "      <th>Sepal_Length</th>\n",
        "      <th>Sepal_Width</th>\n",
        "      <th>Petal_Length</th>\n",
        "      <th>Petal_Width</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>Sepal_Length</th>\n",
        "      <td> 1.000000</td>\n",
        "      <td>-0.109369</td>\n",
        "      <td> 0.871754</td>\n",
        "      <td> 0.817954</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>Sepal_Width</th>\n",
        "      <td>-0.109369</td>\n",
        "      <td> 1.000000</td>\n",
        "      <td>-0.420516</td>\n",
        "      <td>-0.356544</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>Petal_Length</th>\n",
        "      <td> 0.871754</td>\n",
        "      <td>-0.420516</td>\n",
        "      <td> 1.000000</td>\n",
        "      <td> 0.962757</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>Petal_Width</th>\n",
        "      <td> 0.817954</td>\n",
        "      <td>-0.356544</td>\n",
        "      <td> 0.962757</td>\n",
        "      <td> 1.000000</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "</div>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 40,
       "text": [
        "              Sepal_Length  Sepal_Width  Petal_Length  Petal_Width\n",
        "Sepal_Length      1.000000    -0.109369      0.871754     0.817954\n",
        "Sepal_Width      -0.109369     1.000000     -0.420516    -0.356544\n",
        "Petal_Length      0.871754    -0.420516      1.000000     0.962757\n",
        "Petal_Width       0.817954    -0.356544      0.962757     1.000000"
       ]
      }
     ],
     "prompt_number": 40
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "### \u7b2c\u4e09\u6b65\uff1a\u6838\u4e3b\u6210\u5206"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from sklearn.decomposition import KernelPCA\n",
      "kpca = KernelPCA(n_components='mle') # \u81ea\u52a8\u9009\u62e9\u4e3b\u6210\u5206\u4e2a\u6570\n",
      "kpca.fit(X)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 44,
       "text": [
        "KernelPCA(alpha=1.0, coef0=1, degree=3, eigen_solver='auto',\n",
        "     fit_inverse_transform=False, gamma=None, kernel='linear',\n",
        "     kernel_params=None, max_iter=None, n_components='mle',\n",
        "     remove_zero_eig=False, tol=0)"
       ]
      }
     ],
     "prompt_number": 44
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "score = kpca.transform(X)\n",
      "score = pd.concat([pd.DataFrame(score[:,:2]),df['Species']],axis=1)\n",
      "score.columns = ['var1','var2','species']\n",
      "score.head()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
        "<table border=\"1\" class=\"dataframe\">\n",
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
        "      <th>var1</th>\n",
        "      <th>var2</th>\n",
        "      <th>species</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>0</th>\n",
        "      <td>-2.684207</td>\n",
        "      <td> 0.326607</td>\n",
        "      <td> setosa</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>1</th>\n",
        "      <td>-2.715391</td>\n",
        "      <td>-0.169557</td>\n",
        "      <td> setosa</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>2</th>\n",
        "      <td>-2.889820</td>\n",
        "      <td>-0.137346</td>\n",
        "      <td> setosa</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>3</th>\n",
        "      <td>-2.746437</td>\n",
        "      <td>-0.311124</td>\n",
        "      <td> setosa</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>4</th>\n",
        "      <td>-2.728593</td>\n",
        "      <td> 0.333925</td>\n",
        "      <td> setosa</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "</div>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 47,
       "text": [
        "       var1      var2 species\n",
        "0 -2.684207  0.326607  setosa\n",
        "1 -2.715391 -0.169557  setosa\n",
        "2 -2.889820 -0.137346  setosa\n",
        "3 -2.746437 -0.311124  setosa\n",
        "4 -2.728593  0.333925  setosa"
       ]
      }
     ],
     "prompt_number": 47
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from  ggplot import *\n",
      "ggplot(score,aes('var1','var2',color='species')) \\\n",
      "    + geom_point()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAqUAAAHzCAYAAAAZ52RFAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nqYiIiIgY9/8BrOz5oWaU0QwAAAAASUVORK5CYII=\n",
       "text": [
        "<matplotlib.figure.Figure at 0x10a1e7b10>"
       ]
      },
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 55,
       "text": [
        "<ggplot: (279554953)>"
       ]
      }
     ],
     "prompt_number": 55
    }
   ],
   "metadata": {}
  }
 ]
}