Part 09 - Generalized Cross Validation.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from sklearn.datasets import load_iris\n",
    "from sklearn.feature_selection import RFE\n",
    "from sklearn.grid_search import GridSearchCV\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.cross_validation import StratifiedKFold\n",
    "\n",
    "iris = load_iris()\n",
    "X, y = iris.data, iris.target\n",
    "\n",
    "rfe = RFE(LogisticRegression(random_state=0))\n",
    "param_grid = {'n_features_to_select': range(1, 5)}\n",
    "grid = GridSearchCV(rfe, param_grid, cv=StratifiedKFold(y))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "%%timeit\n",
    "grid.fit(X, y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "grid.best_score_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from sklearn.feature_selection import RFECV\n",
    "rfecv = RFECV(LogisticRegression(random_state=0), cv=StratifiedKFold(y))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "%%timeit\n",
    "rfecv.fit(X, y)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "rfecv.grid_scores_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "from sklearn import linear_model\n",
    "from sklearn import datasets\n",
    "\n",
    "diabetes = datasets.load_diabetes()\n",
    "X = diabetes.data\n",
    "y = diabetes.target\n",
    "\n",
    "print(\"Computing regularization path using the LARS ...\")\n",
    "alphas, _, coefs = linear_model.lars_path(X, y, method='lasso', verbose=True)\n",
    "\n",
    "xx = np.sum(np.abs(coefs.T), axis=1)\n",
    "xx /= xx[-1]\n",
    "\n",
    "plt.plot(xx, coefs.T)\n",
    "ymin, ymax = plt.ylim()\n",
    "plt.vlines(xx, ymin, ymax, linestyle='dashed')\n",
    "plt.xlabel('|coef| / max|coef|')\n",
    "plt.ylabel('Coefficients')\n",
    "plt.title('LASSO Path')\n",
    "plt.axis('tight')\n",
    "plt.show()\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from sklearn.cross_validation import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, y)\n",
    "grid_search = GridSearchCV(linear_model.LassoLars(), param_grid={'alpha': np.logspace(-5, 3)})\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "%%timeit\n",
    "grid_search.fit(X, y)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "print(grid_search.best_params_)\n",
    "print(grid_search.score(X_test, y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "lasso = linear_model.LassoLarsCV()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "%%timeit\n",
    "lasso.fit(X, y)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "print(lasso.alpha)\n",
    "print(lasso.score(X_test, y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "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": "3.4.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}