flying-sheep/knee.ipynb

## knee.ipynb
{
 "metadata": {
  "name": ""
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "%matplotlib inline\n",
      "import numpy as np\n",
      "import matplotlib.pyplot as plt"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 6
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "aics = np.array([18.95321392, 18.95340564, 18.95299673, 18.95276907, 18.95294959, 18.95240419, 18.95161582, 18.89662475])"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 8
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "coords = np.c_[0:len(aics), aics]\n",
      "\n",
      "line_vec = coords[-1] - coords[0]\n",
      "line_vec /= np.sqrt(np.sum(line_vec ** 2))\n",
      "\n",
      "from_first = coords - coords[0]\n",
      "\n",
      "scalar_prods = np.sum(from_first * line_vec, 1)\n",
      "from_first_parallel = np.outer(scalar_prods, line_vec)\n",
      "vec_to_line = from_first - from_first_parallel\n",
      "\n",
      "dists = np.sqrt(np.sum(vec_to_line ** 2, 1))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 15
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "colors = plt.cm.Set1(np.linspace(0, 1, 4))\n",
      "\n",
      "def normalize(x):\n",
      "    x = x - x.min()\n",
      "    return x / x.max()\n",
      "\n",
      "plt.figure()\n",
      "ax = plt.subplot(111)\n",
      "ax.set_xlim(-.5, 7.5)\n",
      "ax.set_ylim(-.1, 1.1)\n",
      "ax.plot(normalize(aics), color=colors[0])\n",
      "ax.plot(normalize(dists), color=colors[1])\n",
      "ax.axvline(dists.argmax(), color=colors[2])\n",
      "ax.axvline(aics.argmin(), color=colors[3])"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 33,
       "text": [
        "<matplotlib.lines.Line2D at 0x7f9168e3b790>"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAW8AAAD7CAYAAAClvBX1AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\niYgkiOVNRCRBLG8iIglieRMRSdD/B9c9MogeonhmAAAAAElFTkSuQmCC\n",
       "text": [
        "<matplotlib.figure.Figure at 0x7f9168f0f110>"
       ]
      }
     ],
     "prompt_number": 33
    }
   ],
   "metadata": {}
  }
 ]
}
	{
	"metadata": {
	"name": ""
	},
	"nbformat": 3,
	"nbformat_minor": 0,
	"worksheets": [
	{
	"cells": [
	{
	"cell_type": "code",
	"collapsed": false,
	"input": [
	"%matplotlib inline\n",
	"import numpy as np\n",
	"import matplotlib.pyplot as plt"
	],
	"language": "python",
	"metadata": {},
	"outputs": [],
	"prompt_number": 6
	},
	{
	"cell_type": "code",
	"collapsed": false,
	"input": [
	"aics = np.array([18.95321392, 18.95340564, 18.95299673, 18.95276907, 18.95294959, 18.95240419, 18.95161582, 18.89662475])"
	],
	"language": "python",
	"metadata": {},
	"outputs": [],
	"prompt_number": 8
	},
	{
	"cell_type": "code",
	"collapsed": false,
	"input": [
	"coords = np.c_[0:len(aics), aics]\n",
	"\n",
	"line_vec = coords[-1] - coords[0]\n",
	"line_vec /= np.sqrt(np.sum(line_vec ** 2))\n",
	"\n",
	"from_first = coords - coords[0]\n",
	"\n",
	"scalar_prods = np.sum(from_first * line_vec, 1)\n",
	"from_first_parallel = np.outer(scalar_prods, line_vec)\n",
	"vec_to_line = from_first - from_first_parallel\n",
	"\n",
	"dists = np.sqrt(np.sum(vec_to_line ** 2, 1))"
	],
	"language": "python",
	"metadata": {},
	"outputs": [],
	"prompt_number": 15
	},
	{
	"cell_type": "code",
	"collapsed": false,
	"input": [
	"colors = plt.cm.Set1(np.linspace(0, 1, 4))\n",
	"\n",
	"def normalize(x):\n",
	" x = x - x.min()\n",
	" return x / x.max()\n",
	"\n",
	"plt.figure()\n",
	"ax = plt.subplot(111)\n",
	"ax.set_xlim(-.5, 7.5)\n",
	"ax.set_ylim(-.1, 1.1)\n",
	"ax.plot(normalize(aics), color=colors[0])\n",
	"ax.plot(normalize(dists), color=colors[1])\n",
	"ax.axvline(dists.argmax(), color=colors[2])\n",
	"ax.axvline(aics.argmin(), color=colors[3])"
	],
	"language": "python",
	"metadata": {},
	"outputs": [
	{
	"metadata": {},
	"output_type": "pyout",
	"prompt_number": 33,
	"text": [
	"<matplotlib.lines.Line2D at 0x7f9168e3b790>"
	]
	},
	{
	"metadata": {},
	"output_type": "display_data",
	"png": "iVBORw0KGgoAAAANSUhEUgAAAW8AAAD7CAYAAAClvBX1AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\niYgkiOVNRCRBLG8iIglieRMRSdD/B9c9MogeonhmAAAAAElFTkSuQmCC\n",
	"text": [
	"<matplotlib.figure.Figure at 0x7f9168f0f110>"
	]
	}
	],
	"prompt_number": 33
	}
	],
	"metadata": {}
	}
	]
	}