Untitled.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "from sklearn.metrics.pairwise import rbf_kernel\n",
    "import matplotlib.pyplot as plt\n",
    "from scipy.optimize import fmin_l_bfgs_b"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "N = 1000\n",
    "X = (2 * np.random.rand(N) - 1).reshape(N, 1)\n",
    "# y = np.ones(N).reshape(N, 1)\n",
    "y = np.random.rand(N).reshape(N, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAD8CAYAAACMwORRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11d2d5828>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(X, y)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "K = rbf_kernel(X, gamma=1000000.)\n",
    "# lbda = 2. * 1. * np.mean(y ** 2) / 1 ** 2\n",
    "lbda = 1."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "# C = np.dot(np.linalg.pinv(K + lbda * K.shape[0] / 2 * np.eye(N)), y)\n",
    "o = fmin_l_bfgs_b(lambda C: np.linalg.norm(np.dot(K, C) - y.T) ** 2 / K.shape[0], np.ones(N), approx_grad=True, bounds=((-.1, .1),) * N)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "85.267964883534447"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.linalg.norm(np.dot(K, o[0]) - y) ** 2 / K.shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.49356496810430994"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.dot(o[0].T, np.dot(K, o[0]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "max norm:  0.666959299049\n",
      "max val:  0.816675761272\n"
     ]
    }
   ],
   "source": [
    "# np.sqrt(2 * 1. * np.mean(y ** 2) / lbda)\n",
    "print('max norm: ', 2 * np.mean(y ** 2) / lbda)\n",
    "print('max val: ', np.sqrt(2 * np.mean(y ** 2) / lbda))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAD8CAYAAACMwORRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x112342198>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(X, np.dot(K, o[0]))\n",
    "plt.scatter(X, y)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "fmin_l_bfgs_b?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 245,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1000.0"
      ]
     },
     "execution_count": 245,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(lambda C: np.linalg.norm(np.dot(K, C) - y) ** 2 / K.shape[0] + lbda / 2 * np.dot(C, np.dot(K, C)))(np.zeros(N))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,\n",
       "        0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001,  0.001])"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "o[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [conda root]",
   "language": "python",
   "name": "conda-root-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}