gistfile1.txt

{
 "metadata": {
  "name": "stumbleupon"
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "import numpy as np\nimport pandas as pd\nfrom sklearn.naive_bayes import MultinomialNB\nfrom sklearn import metrics, cross_validation\nfrom sklearn.cross_validation import train_test_split\nfrom sklearn.grid_search import GridSearchCV",
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 31
    },
    {
     "cell_type": "heading",
     "level": 3,
     "metadata": {},
     "source": "Import data"
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "# import data\nX = pd.read_csv('trainMatrix.csv')\ny = pd.read_csv('labels.csv').values.ravel()\nX_train, X_test, y_train, y_test = train_test_split( X, y, test_size=0.5, random_state=0)",
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 38
    },
    {
     "cell_type": "heading",
     "level": 3,
     "metadata": {},
     "source": "Grid Search"
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "# Set the parameters by cross-validation\ntuned_parameters = [{'alpha': [0.5, 1, 2, 5, 10, 50, 100]}] #, 1, 2, 5, 10, 50, 100]\nclf = GridSearchCV(MultinomialNB(), tuned_parameters, cv=5, scoring='roc_auc', n_jobs=-1)\nclf.fit(X_train, y_train)\n\nmodel = clf.best_estimator_\nprint 'Best model = {}'.format(model)\n#model = MultinomialNB(alpha=best_alpha)\nfor params, mean_score, scores in clf.grid_scores_:\n    print(\"%0.3f (+/-%0.03f) for %r\" % (mean_score, scores.std() / 2, params))",
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": "*"
    }
   ],
   "metadata": {}
  }
 ]
}