variability of active set in cross validation

CV_variability.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/jonathantaylor/anaconda/lib/python2.7/site-packages/selection/algorithms/lasso.py:32: UserWarning: cvx not available\n",
      "  warnings.warn('cvx not available')\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Loading required package: Matrix\n",
       "Loading required package: foreach\n",
       "foreach: simple, scalable parallel programming from Revolution Analytics\n",
       "Use Revolution R for scalability, fault tolerance and more.\n",
       "http://www.revolutionanalytics.com\n",
       "Loaded glmnet 2.0-2\n",
       "\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "%load_ext rpy2.ipython\n",
    "from selection.algorithms.lasso import instance\n",
    "%R library(glmnet)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def active_set(X, Y ):\n",
    "    \n",
    "    %R -i X,Y\n",
    "    %R lam.min = cv.glmnet(X,as.matrix(Y))$lambda.min\n",
    "    %R -o B B = as.matrix(glmnet(X, as.matrix(Y), lambda=lam.min)$beta)\n",
    "    return tuple(np.nonzero(np.squeeze(B) != 0)[0])\n",
    "\n",
    "X, Y = instance(n=100, p=200, rho=0.3)[:2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of different active sets in 50 assignments to folds: 4\n"
     ]
    }
   ],
   "source": [
    "X, Y = instance(n=100, p=200, rho=0.3)[:2]\n",
    "active_sets = set([active_set(X, Y) for _ in range(50)])\n",
    "print 'Number of different active sets in 50 assignments to folds:', len(active_sets)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of different active sets in 50 assignments to folds: 7\n"
     ]
    }
   ],
   "source": [
    "X, Y = instance(n=100, p=200, rho=0.3)[:2]\n",
    "active_sets = set([active_set(X, Y) for _ in range(50)])\n",
    "print 'Number of different active sets in 50 assignments to folds:', len(active_sets)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of different active sets in 50 assignments to folds: 2\n"
     ]
    }
   ],
   "source": [
    "X, Y = instance(n=100, p=200, rho=0.3)[:2]\n",
    "active_sets = set([active_set(X, Y) for _ in range(50)])\n",
    "print 'Number of different active sets in 50 assignments to folds:', len(active_sets)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}