{
 "metadata": {
  "name": ""
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "#Basic principles of machine learning (1 hr)"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Here is where we start diving into the field of machine learning.\n",
      "\n",
      "By the end of this section you will\n",
      "\n",
      "- Know the basic categories of supervised learning, including classification and regression problems.\n",
      "- Know the basic categories of unsupervised learning, including dimensionality reduction and clustering.\n",
      "- Know the basic syntax of the Scikit-learn **estimator** interface.\n",
      "- Know how features are extracted from real-world data.\n",
      "\n",
      "In addition, we will go over several basic tools within scikit-learn which can be used to accomplish the above tasks."
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "##Problem setting"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "###A simple definition of machine learning"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Machine Learning (ML) is about building programs with **tunable parameters** (typically an\n",
      "array of floating point values) that are adjusted automatically so as to improve\n",
      "their behavior by **adapting to previously seen data.**\n",
      "\n",
      "In most ML applications, the data is in a 2D array of shape ``[n_samples x n_features]``,\n",
      "where the number of features is the same for each object, and each feature column refers\n",
      "to a related piece of information about each sample.\n",
      "\n",
      "Machine learning can be broken into two broad regimes:\n",
      "*supervised learning* and *unsupervised learning*.\n",
      "We\u2019ll introduce these concepts here, and discuss them in more detail below."
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "## Scikit Learn\n",
      "\n",
      "### What is it?\n",
      "\n",
      "- Simple and efficient tools for data mining and data analysis\n",
      "- Accessible to everybody, and reusable in various contexts\n",
      "- Built on NumPy, SciPy, and matplotlib\n",
      "- Open source, commercially usable - BSD license"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "*Adapted from http://scikit-learn.org/stable/tutorial/basic/tutorial.html*"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "###Introducing the scikit-learn estimator object"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Every algorithm is exposed in scikit-learn via an ''Estimator'' object. For instance a linear regression is:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "%pylab inline\n",
      "import pylab as plt\n",
      "import numpy as np"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Populating the interactive namespace from numpy and matplotlib\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stderr",
       "text": [
        "WARNING: pylab import has clobbered these variables: ['plt']\n",
        "`%pylab --no-import-all` prevents importing * from pylab and numpy\n"
       ]
      }
     ],
     "prompt_number": 3
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from sklearn.linear_model import LinearRegression"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 22
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "**Estimator parameters**: All the parameters of an estimator can be set when it is instantiated:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "model = LinearRegression(normalize=True)\n",
      "print model.normalize"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "True\n"
       ]
      }
     ],
     "prompt_number": 23
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "print model"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "LinearRegression(copy_X=True, fit_intercept=True, normalize=True)\n"
       ]
      }
     ],
     "prompt_number": 24
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "**Estimated parameters**: When data is fitted with an estimator, parameters are estimated from the data at hand. All the estimated parameters are attributes of the estimator object ending by an underscore:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "x = np.array([0, 1, 2])\n",
      "y = np.array([0, 1, 2])"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 25
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "_ = plt.plot(x, y, marker='o')"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAD9CAYAAABHnDf0AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n4EL/D49YAOUwyE9hAAAAAElFTkSuQmCC\n",
       "text": [
        "<matplotlib.figure.Figure at 0x61933f0>"
       ]
      }
     ],
     "prompt_number": 27
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "X = x[:, np.newaxis] # The input data for sklearn is 2D: (samples == 3 x features == 1)\n",
      "X"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 28,
       "text": [
        "array([[0],\n",
        "       [1],\n",
        "       [2]])"
       ]
      }
     ],
     "prompt_number": 28
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "model.fit(X, y) \n",
      "model.coef_"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 31,
       "text": [
        "array([ 1.00000003])"
       ]
      }
     ],
     "prompt_number": 31
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "###Supervised Learning: Classification and regression"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "In **Supervised Learning**, we have a dataset consisting of both features and labels.\n",
      "The task is to construct an estimator which is able to predict the label of an object\n",
      "given the set of features. A relatively simple example is predicting the species of \n",
      "iris given a set of measurements of its flower. This is a relatively simple task. \n",
      "Some more complicated examples are:\n",
      "\n",
      "- given a multicolor image of an object through a telescope, determine\n",
      "  whether that object is a star, a quasar, or a galaxy.\n",
      "- given a photograph of a person, identify the person in the photo.\n",
      "- given a list of movies a person has watched and their personal rating\n",
      "  of the movie, recommend a list of movies they would like\n",
      "  (So-called *recommender systems*: a famous example is the [Netflix Prize](http://en.wikipedia.org/wiki/Netflix_prize)).\n",
      "\n",
      "What these tasks have in common is that there is one or more unknown\n",
      "quantities associated with the object which needs to be determined from other\n",
      "observed quantities.\n",
      "\n",
      "Supervised learning is further broken down into two categories, **classification** and **regression**.\n",
      "In classification, the label is discrete, while in regression, the label is continuous. For example,\n",
      "in astronomy, the task of determining whether an object is a star, a galaxy, or a quasar is a\n",
      "classification problem: the label is from three distinct categories. On the other hand, we might\n",
      "wish to estimate the age of an object based on such observations: this would be a regression problem,\n",
      "because the label (age) is a continuous quantity."
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "**Classification**: K nearest neighbors (kNN) is one of the simplest learning strategies: given a new, unknown observation, look up in your reference database which ones have the closest features and assign the predominant class.\n",
      "\n",
      "Let's try it out on our iris classification problem:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from sklearn import neighbors, datasets\n",
      "iris = datasets.load_iris()\n",
      "X, y = iris.data, iris.target\n",
      "knn = neighbors.KNeighborsClassifier(n_neighbors=1)\n",
      "knn.fit(X, y)\n",
      "# What kind of iris has 3cm x 5cm sepal and 4cm x 2cm petal?\n",
      "print iris.target_names[knn.predict([[3, 5, 4, 2]])]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "['virginica']\n"
       ]
      }
     ],
     "prompt_number": 33
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# A plot of the sepal space and the prediction of the KNN\n",
      "from helpers import plot_iris_knn\n",
      "plot_iris_knn()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAX8AAAEICAYAAAC3Y/QeAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQmCC\n",
       "text": [
        "<matplotlib.figure.Figure at 0x6a5da10>"
       ]
      }
     ],
     "prompt_number": 36
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "**Regression**: The simplest possible regression setting is the linear regression one:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# Create some simple data\n",
      "import numpy as np\n",
      "np.random.seed(0)\n",
      "X = np.random.random(size=(20, 1))\n",
      "y = 3 * X.squeeze() + 2 + np.random.normal(size=20)\n",
      "\n",
      "# Fit a linear regression to it\n",
      "from sklearn.linear_model import LinearRegression\n",
      "model = LinearRegression(fit_intercept=True)\n",
      "model.fit(X, y)\n",
      "print \"Model coefficient: %.5f, and intercept: %.5f\" % (model.coef_, model.intercept_)\n",
      "\n",
      "# Plot the data and the model prediction\n",
      "X_test = np.linspace(0, 1, 100)[:, np.newaxis]\n",
      "y_test = model.predict(X_test)\n",
      "import pylab as pl\n",
      "print X.squeeze()\n",
      "pl.plot(X.squeeze(), y, 'o')\n",
      "pl.plot(X_test.squeeze(), y_test)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Model coefficient: 3.93491, and intercept: 1.46229\n",
        "[ 0.5488135   0.71518937  0.60276338  0.54488318  0.4236548   0.64589411\n",
        "  0.43758721  0.891773    0.96366276  0.38344152  0.79172504  0.52889492\n",
        "  0.56804456  0.92559664  0.07103606  0.0871293   0.0202184   0.83261985\n",
        "  0.77815675  0.87001215]\n"
       ]
      },
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 4,
       "text": [
        "[<matplotlib.lines.Line2D at 0x1a57430>]"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAW0AAAD9CAYAAAB3ECbVAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRU5ErkJggg==\n",
       "text": [
        "<matplotlib.figure.Figure at 0x6910470>"
       ]
      }
     ],
     "prompt_number": 4
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "**Unsupervised Learning** addresses a different sort of problem. Here the data has no labels,\n",
      "and we are interested in finding similarities between the objects in question. In a sense,\n",
      "you can think of unsupervised learning as a means of discovering labels from the data itself.\n",
      "Unsupervised learning comprises tasks such as *dimensionality reduction*, *clustering*, and\n",
      "*density estimation*. For example, in the iris data discussed above, we can used unsupervised\n",
      "methods to determine combinations of the measurements which best display the structure of the\n",
      "data. As we\u2019ll see below, such a projection of the data can be used to visualize the\n",
      "four-dimensional dataset in two dimensions. Some more involved unsupervised learning problems are:\n",
      "\n",
      "- given detailed observations of distant galaxies, determine which features or combinations of\n",
      "  features summarize best the information.\n",
      "- given a mixture of two sound sources (for example, a person talking over some music),\n",
      "  separate the two (this is called the [blind source separation](http://en.wikipedia.org/wiki/Blind_signal_separation) problem).\n",
      "- given a video, isolate a moving object and categorize in relation to other moving objects which have been seen.\n",
      "\n",
      "Sometimes the two may even be combined: e.g. Unsupervised learning can be used to find useful\n",
      "features in heterogeneous data, and then these features can be used within a supervised\n",
      "framework."
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "**An example PCA for visualization** Principle Component Analysis (PCA) is a dimension reduction technique that can find the combinations of variables that explain the most variance.\n",
      "\n",
      "Consider the iris dataset. It cannot be visualized in a single 2D plot, as it has 4 features. We are going to extract 2 combinations of sepal and petal dimensions to visualize it:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "X, y = iris.data, iris.target\n",
      "from sklearn.decomposition import PCA\n",
      "pca = PCA(n_components=2)\n",
      "pca.fit(X)\n",
      "X_reduced = pca.transform(X)\n",
      "print \"Reduced dataset shape:\", X_reduced.shape\n",
      "\n",
      "import pylab as pl\n",
      "pl.scatter(X_reduced[:, 0], X_reduced[:, 1], c=y)\n",
      "\n",
      "print \"Meaning of the 2 components:\"\n",
      "for component in pca.components_:\n",
      "    print \" + \".join(\"%.3f x %s\" % (value, name)\n",
      "                     for value, name in zip(component, iris.feature_names))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Reduced dataset shape: (150, 2)\n",
        "Meaning of the 2 components:"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\n",
        "0.362 x sepal length (cm) + -0.082 x sepal width (cm) + 0.857 x petal length (cm) + 0.359 x petal width (cm)\n",
        "-0.657 x sepal length (cm) + -0.730 x sepal width (cm) + 0.176 x petal length (cm) + 0.075 x petal width (cm)\n"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAD9CAYAAACyYrxEAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
       "text": [
        "<matplotlib.figure.Figure at 0x651ba70>"
       ]
      }
     ],
     "prompt_number": 38
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "**Clustering**: Clustering groups together observations that are homogeneous with respect to a given criterion, finding ''clusters'' in the data.\n",
      "\n",
      "Note that these clusters will uncover relevent hidden structure of the data only if the criterion used highlights it."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from sklearn.cluster import KMeans\n",
      "k_means = KMeans(n_clusters=3, random_state=0) # Fixing the RNG in kmeans\n",
      "k_means.fit(X_reduced)\n",
      "y_pred = k_means.predict(X_reduced)\n",
      "\n",
      "pl.scatter(X_reduced[:, 0], X_reduced[:, 1], c=y_pred)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 39,
       "text": [
        "<matplotlib.collections.PathCollection at 0x69d5570>"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAD9CAYAAACyYrxEAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n40LPGGMmjgs9Y4yZuP8Hr2YGqLGCRuIAAAAASUVORK5CYII=\n",
       "text": [
        "<matplotlib.figure.Figure at 0x69ca850>"
       ]
      }
     ],
     "prompt_number": 39
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "##Flow chart: how do I choose what to do with my data set?"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "<a href=\"http://scikit-learn.org/dev/tutorial/machine_learning_map/index.html\"><img src=\"http://scikit-learn.org/dev/_static/ml_map.png\" width=100%></a>"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "%pylab inline\n",
      "import pylab as pl\n",
      "import numpy as np\n",
      "\n",
      "# Some nice default configuration for plots\n",
      "pl.rcParams['figure.figsize'] = 10, 7.5\n",
      "pl.rcParams['axes.grid'] = True\n",
      "pl.gray()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Populating the interactive namespace from numpy and matplotlib\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stderr",
       "text": [
        "WARNING: pylab import has clobbered these variables: ['plt']\n",
        "`%pylab --no-import-all` prevents importing * from pylab and numpy\n"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "text": [
        "<matplotlib.figure.Figure at 0x69ea950>"
       ]
      }
     ],
     "prompt_number": 48
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "##Text Feature Extraction for Classification and Clustering"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Outline of this section:\n",
      "\n",
      "- Turn a corpus of text documents into **feature vectors** using a **Bag of Words** representation,\n",
      "- Train a simple text classifier on the feature vectors,\n",
      "- Wrap the vectorizer and the classifier with a **pipeline**,\n",
      "- Cross-validation and **model selection** on the pipeline."
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "###Text Classification in 20 lines of Python"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Let's start by implementing a canonical text classification example:\n",
      "\n",
      "- The 20 newsgroups dataset: around 18000 text posts from 20 newsgroups forums\n",
      "- Bag of Words features extraction with TF-IDF weighting\n",
      "- Naive Bayes classifier or Linear Support Vector Machine for the classifier itself"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from sklearn.datasets import load_files\n",
      "from sklearn.feature_extraction.text import TfidfVectorizer\n",
      "from sklearn.naive_bayes import MultinomialNB\n",
      "\n",
      "# Load the text data\n",
      "categories = [\n",
      "    'alt.atheism',\n",
      "    'talk.religion.misc',\n",
      "    'comp.graphics',\n",
      "    'sci.space',\n",
      "]\n",
      "twenty_train_small = load_files('../data/twenty_newsgroups/20news-bydate-train/',\n",
      "    categories=categories, charset='latin-1')\n",
      "twenty_test_small = load_files('../data/twenty_newsgroups/20news-bydate-test/',\n",
      "    categories=categories, charset='latin-1')\n",
      "\n",
      "# Turn the text documents into vectors of word frequencies\n",
      "vectorizer = TfidfVectorizer(min_df=2)\n",
      "X_train = vectorizer.fit_transform(twenty_train_small.data)\n",
      "y_train = twenty_train_small.target\n",
      "\n",
      "# Fit a classifier on the training set\n",
      "classifier = MultinomialNB().fit(X_train, y_train)\n",
      "print(\"Training score: {0:.1f}%\".format(\n",
      "    classifier.score(X_train, y_train) * 100))\n",
      "\n",
      "# Evaluate the classifier on the testing set\n",
      "X_test = vectorizer.transform(twenty_test_small.data)\n",
      "y_test = twenty_test_small.target\n",
      "print(\"Testing score: {0:.1f}%\".format(\n",
      "    classifier.score(X_test, y_test) * 100))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stderr",
       "text": [
        "/Library/Frameworks/Python.framework/Versions/7.2/lib/python2.7/site-packages/scikit_learn-0.14.1-py2.7-macosx-10.5-i386.egg/sklearn/datasets/base.py:161: DeprecationWarning: The charset parameter is deprecated as of version 0.14 and will be removed in 0.16. Use encode instead.\n",
        "  DeprecationWarning)\n",
        "/Library/Frameworks/Python.framework/Versions/7.2/lib/python2.7/site-packages/scikit_learn-0.14.1-py2.7-macosx-10.5-i386.egg/sklearn/datasets/base.py:161: DeprecationWarning: The charset parameter is deprecated as of version 0.14 and will be removed in 0.16. Use encode instead.\n",
        "  DeprecationWarning)\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Training score: 95.1%\n",
        "Testing score: 85.1%"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\n"
       ]
      }
     ],
     "prompt_number": 51
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Here is a workflow diagram summary of what happened previously:"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "<img src=\"https://raw.github.com/jakevdp/sklearn_scipy2013/master/notebooks/figures/supervised_scikit_learn.png\" style=\"width: 80%\" />"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "###Loading the Dataset"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Let's explore the dataset loading utility without passing a list of categories: in this case we load the full 20 newsgroups dataset in memory. The source website for the 20 newsgroups already provides a date-based train / test split that is made available using the `subset` keyword argument: "
     ]
    },
    {
     "cell_type": "code",
     "collapsed": true,
     "input": [
      "ls -l ../data/"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "total 1744\r\n",
        "drwxr-xr-x  3 marcelcaraciolo  staff     102 Sep 26 16:54 \u001b[34mlabeled_faces_wild\u001b[m\u001b[m/\r\n",
        "drwxr-xr-x  3 marcelcaraciolo  staff     102 Sep 26 16:54 \u001b[34mlanguages\u001b[m\u001b[m/\r\n",
        "drwxr-xr-x  6 marcelcaraciolo  staff     204 Aug  4 02:30 \u001b[34mml-1m\u001b[m\u001b[m/\r\n",
        "-rw-r--r--@ 1 marcelcaraciolo  staff    1047 Sep 29 23:46 movie_rating.csv\r\n",
        "drwxr-xr-x  3 marcelcaraciolo  staff     102 Sep 26 16:54 \u001b[34mmovie_reviews\u001b[m\u001b[m/\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff  371839 Sep 29 20:27 movielens_test.csv\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff  512859 Sep 29 20:27 movielens_train.csv\r\n",
        "drwxr-xr-x  5 marcelcaraciolo  staff     170 Sep 30 17:16 \u001b[34mtwenty_newsgroups\u001b[m\u001b[m/\r\n"
       ]
      }
     ],
     "prompt_number": 55
    },
    {
     "cell_type": "code",
     "collapsed": true,
     "input": [
      "ls -lh ../data/twenty_newsgroups/20news-bydate-train"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "total 0\r\n",
        "drwxr-xr-x  482 marcelcaraciolo  staff    16K Mar 18  2003 \u001b[34malt.atheism\u001b[m\u001b[m/\r\n",
        "drwxr-xr-x  586 marcelcaraciolo  staff    19K Mar 18  2003 \u001b[34mcomp.graphics\u001b[m\u001b[m/\r\n",
        "drwxr-xr-x  593 marcelcaraciolo  staff    20K Mar 18  2003 \u001b[34mcomp.os.ms-windows.misc\u001b[m\u001b[m/\r\n",
        "drwxr-xr-x  592 marcelcaraciolo  staff    20K Mar 18  2003 \u001b[34mcomp.sys.ibm.pc.hardware\u001b[m\u001b[m/\r\n",
        "drwxr-xr-x  580 marcelcaraciolo  staff    19K Mar 18  2003 \u001b[34mcomp.sys.mac.hardware\u001b[m\u001b[m/\r\n",
        "drwxr-xr-x  595 marcelcaraciolo  staff    20K Mar 18  2003 \u001b[34mcomp.windows.x\u001b[m\u001b[m/\r\n",
        "drwxr-xr-x  587 marcelcaraciolo  staff    19K Mar 18  2003 \u001b[34mmisc.forsale\u001b[m\u001b[m/\r\n",
        "drwxr-xr-x  596 marcelcaraciolo  staff    20K Mar 18  2003 \u001b[34mrec.autos\u001b[m\u001b[m/\r\n",
        "drwxr-xr-x  600 marcelcaraciolo  staff    20K Mar 18  2003 \u001b[34mrec.motorcycles\u001b[m\u001b[m/\r\n",
        "drwxr-xr-x  599 marcelcaraciolo  staff    20K Mar 18  2003 \u001b[34mrec.sport.baseball\u001b[m\u001b[m/\r\n",
        "drwxr-xr-x  602 marcelcaraciolo  staff    20K Mar 18  2003 \u001b[34mrec.sport.hockey\u001b[m\u001b[m/\r\n",
        "drwxr-xr-x  597 marcelcaraciolo  staff    20K Mar 18  2003 \u001b[34msci.crypt\u001b[m\u001b[m/\r\n",
        "drwxr-xr-x  593 marcelcaraciolo  staff    20K Mar 18  2003 \u001b[34msci.electronics\u001b[m\u001b[m/\r\n",
        "drwxr-xr-x  596 marcelcaraciolo  staff    20K Mar 18  2003 \u001b[34msci.med\u001b[m\u001b[m/\r\n",
        "drwxr-xr-x  595 marcelcaraciolo  staff    20K Mar 18  2003 \u001b[34msci.space\u001b[m\u001b[m/\r\n",
        "drwxr-xr-x  601 marcelcaraciolo  staff    20K Mar 18  2003 \u001b[34msoc.religion.christian\u001b[m\u001b[m/\r\n",
        "drwxr-xr-x  548 marcelcaraciolo  staff    18K Mar 18  2003 \u001b[34mtalk.politics.guns\u001b[m\u001b[m/\r\n",
        "drwxr-xr-x  566 marcelcaraciolo  staff    19K Mar 18  2003 \u001b[34mtalk.politics.mideast\u001b[m\u001b[m/\r\n",
        "drwxr-xr-x  467 marcelcaraciolo  staff    16K Mar 18  2003 \u001b[34mtalk.politics.misc\u001b[m\u001b[m/\r\n",
        "drwxr-xr-x  379 marcelcaraciolo  staff    13K Mar 18  2003 \u001b[34mtalk.religion.misc\u001b[m\u001b[m/\r\n"
       ]
      }
     ],
     "prompt_number": 56
    },
    {
     "cell_type": "code",
     "collapsed": true,
     "input": [
      "ls -lh ../data/twenty_newsgroups/20news-bydate-train/alt.atheism/"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "total 4480\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff    12K Mar 18  2003 49960\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff    31K Mar 18  2003 51060\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   4.0K Mar 18  2003 51119\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.6K Mar 18  2003 51120\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   773B Mar 18  2003 51121\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   4.8K Mar 18  2003 51122\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   618B Mar 18  2003 51123\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.4K Mar 18  2003 51124\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.7K Mar 18  2003 51125\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   427B Mar 18  2003 51126\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   742B Mar 18  2003 51127\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   650B Mar 18  2003 51128\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.3K Mar 18  2003 51130\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.3K Mar 18  2003 51131\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.6K Mar 18  2003 51132\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.5K Mar 18  2003 51133\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.2K Mar 18  2003 51134\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.6K Mar 18  2003 51135\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.1K Mar 18  2003 51136\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.3K Mar 18  2003 51139\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   409B Mar 18  2003 51140\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   940B Mar 18  2003 51141\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   9.0K Mar 18  2003 51142\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   632B Mar 18  2003 51143\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.2K Mar 18  2003 51144\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   609B Mar 18  2003 51145\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   631B Mar 18  2003 51146\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.0K Mar 18  2003 51147\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.8K Mar 18  2003 51148\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   405B Mar 18  2003 51149\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   696B Mar 18  2003 51150\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   5.5K Mar 18  2003 51151\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.4K Mar 18  2003 51152\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   5.0K Mar 18  2003 51153\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.3K Mar 18  2003 51154\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.6K Mar 18  2003 51155\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   5.0K Mar 18  2003 51156\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.8K Mar 18  2003 51157\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   604B Mar 18  2003 51158\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.4K Mar 18  2003 51159\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.0K Mar 18  2003 51160\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.4K Mar 18  2003 51161\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.9K Mar 18  2003 51162\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.1K Mar 18  2003 51163\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.3K Mar 18  2003 51164\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   4.8K Mar 18  2003 51165\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.2K Mar 18  2003 51169\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   868B Mar 18  2003 51170\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   721B Mar 18  2003 51171\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   3.0K Mar 18  2003 51172\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.9K Mar 18  2003 51173\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   645B Mar 18  2003 51174\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.4K Mar 18  2003 51175\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.9K Mar 18  2003 51176\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.3K Mar 18  2003 51177\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   879B Mar 18  2003 51178\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.3K Mar 18  2003 51179\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   994B Mar 18  2003 51180\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.2K Mar 18  2003 51181\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.2K Mar 18  2003 51182\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.7K Mar 18  2003 51183\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.0K Mar 18  2003 51184\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.2K Mar 18  2003 51185\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   949B Mar 18  2003 51186\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.9K Mar 18  2003 51187\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.1K Mar 18  2003 51188\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   834B Mar 18  2003 51189\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   895B Mar 18  2003 51190\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   776B Mar 18  2003 51191\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.6K Mar 18  2003 51192\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.8K Mar 18  2003 51193\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.4K Mar 18  2003 51194\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   964B Mar 18  2003 51195\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.4K Mar 18  2003 51196\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   759B Mar 18  2003 51197\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.5K Mar 18  2003 51198\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.0K Mar 18  2003 51199\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.9K Mar 18  2003 51200\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   916B Mar 18  2003 51201\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.9K Mar 18  2003 51202\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.5K Mar 18  2003 51203\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   846B Mar 18  2003 51204\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.4K Mar 18  2003 51205\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   881B Mar 18  2003 51206\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   6.2K Mar 18  2003 51208\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.7K Mar 18  2003 51209\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.7K Mar 18  2003 51210\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff    10K Mar 18  2003 51211\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.5K Mar 18  2003 51212\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.6K Mar 18  2003 51213\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   636B Mar 18  2003 51214\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   989B Mar 18  2003 51215\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   668B Mar 18  2003 51216\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.8K Mar 18  2003 51217\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.7K Mar 18  2003 51218\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   905B Mar 18  2003 51219\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.4K Mar 18  2003 51220\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.0K Mar 18  2003 51221\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.7K Mar 18  2003 51222\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.0K Mar 18  2003 51223\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.1K Mar 18  2003 51224\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.5K Mar 18  2003 51225\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   3.4K Mar 18  2003 51226\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   704B Mar 18  2003 51227\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   949B Mar 18  2003 51228\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   714B Mar 18  2003 51229\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   966B Mar 18  2003 51230\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.9K Mar 18  2003 51231\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   871B Mar 18  2003 51232\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.0K Mar 18  2003 51233\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.5K Mar 18  2003 51234\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.4K Mar 18  2003 51235\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.2K Mar 18  2003 51236\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   564B Mar 18  2003 51237\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff    11K Mar 18  2003 51238\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.3K Mar 18  2003 51239\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   749B Mar 18  2003 51240\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   932B Mar 18  2003 51241\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.2K Mar 18  2003 51242\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.2K Mar 18  2003 51243\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   554B Mar 18  2003 51244\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.3K Mar 18  2003 51245\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.7K Mar 18  2003 51246\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.7K Mar 18  2003 51247\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.6K Mar 18  2003 51249\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.8K Mar 18  2003 51250\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   570B Mar 18  2003 51251\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.8K Mar 18  2003 51252\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   3.1K Mar 18  2003 51253\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.9K Mar 18  2003 51254\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   748B Mar 18  2003 51255\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.3K Mar 18  2003 51256\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.2K Mar 18  2003 51258\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.7K Mar 18  2003 51259\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   6.2K Mar 18  2003 51260\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.6K Mar 18  2003 51261\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.3K Mar 18  2003 51262\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.2K Mar 18  2003 51265\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   456B Mar 18  2003 51266\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   816B Mar 18  2003 51267\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.4K Mar 18  2003 51268\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.0K Mar 18  2003 51269\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   3.4K Mar 18  2003 51270\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.3K Mar 18  2003 51271\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.0K Mar 18  2003 51272\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   790B Mar 18  2003 51273\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.6K Mar 18  2003 51274\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.5K Mar 18  2003 51275\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   4.4K Mar 18  2003 51276\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.5K Mar 18  2003 51277\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   6.2K Mar 18  2003 51278\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   963B Mar 18  2003 51279\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.0K Mar 18  2003 51280\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.1K Mar 18  2003 51281\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   618B Mar 18  2003 51282\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.7K Mar 18  2003 51283\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.0K Mar 18  2003 51284\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.3K Mar 18  2003 51285\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   601B Mar 18  2003 51286\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   751B Mar 18  2003 51287\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.3K Mar 18  2003 51288\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   8.0K Mar 18  2003 51290\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.0K Mar 18  2003 51291\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.9K Mar 18  2003 51292\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.2K Mar 18  2003 51293\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.8K Mar 18  2003 51294\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.9K Mar 18  2003 51295\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.7K Mar 18  2003 51296\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   4.2K Mar 18  2003 51297\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.6K Mar 18  2003 51298\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.2K Mar 18  2003 51299\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.3K Mar 18  2003 51300\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   6.3K Mar 18  2003 51301\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.3K Mar 18  2003 51302\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.9K Mar 18  2003 51303\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff    10K Mar 18  2003 51304\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.5K Mar 18  2003 51305\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.4K Mar 18  2003 51306\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   4.1K Mar 18  2003 51307\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   6.2K Mar 18  2003 51308\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.0K Mar 18  2003 51309\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   768B Mar 18  2003 51310\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.0K Mar 18  2003 51311\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   930B Mar 18  2003 51312\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   771B Mar 18  2003 51313\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   670B Mar 18  2003 51314\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.1K Mar 18  2003 51315\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   3.7K Mar 18  2003 51316\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   406B Mar 18  2003 51317\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   5.4K Mar 18  2003 51318\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   9.6K Mar 18  2003 51319\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.1K Mar 18  2003 51320\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff    29K Mar 18  2003 52499\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff    25K Mar 18  2003 52909\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   5.8K Mar 18  2003 52910\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   819B Mar 18  2003 53055\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   857B Mar 18  2003 53056\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   755B Mar 18  2003 53057\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   4.4K Mar 18  2003 53058\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.1K Mar 18  2003 53059\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.1K Mar 18  2003 53062\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.6K Mar 18  2003 53064\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   515B Mar 18  2003 53065\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   9.2K Mar 18  2003 53066\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.4K Mar 18  2003 53067\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   610B Mar 18  2003 53069\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   759B Mar 18  2003 53070\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.3K Mar 18  2003 53071\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.5K Mar 18  2003 53072\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.9K Mar 18  2003 53073\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.1K Mar 18  2003 53075\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   411B Mar 18  2003 53078\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.3K Mar 18  2003 53081\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   962B Mar 18  2003 53082\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.3K Mar 18  2003 53083\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.0K Mar 18  2003 53085\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.1K Mar 18  2003 53086\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   247B Mar 18  2003 53087\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   3.8K Mar 18  2003 53090\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.1K Mar 18  2003 53093\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.1K Mar 18  2003 53094\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.0K Mar 18  2003 53095\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   863B Mar 18  2003 53096\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.1K Mar 18  2003 53097\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.2K Mar 18  2003 53098\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.1K Mar 18  2003 53099\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.0K Mar 18  2003 53106\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   784B Mar 18  2003 53108\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.3K Mar 18  2003 53110\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   712B Mar 18  2003 53111\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.4K Mar 18  2003 53112\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.6K Mar 18  2003 53113\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.7K Mar 18  2003 53114\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.5K Mar 18  2003 53117\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.8K Mar 18  2003 53118\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   4.1K Mar 18  2003 53120\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.8K Mar 18  2003 53121\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.4K Mar 18  2003 53122\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.2K Mar 18  2003 53123\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   3.4K Mar 18  2003 53124\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.8K Mar 18  2003 53125\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.2K Mar 18  2003 53126\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   826B Mar 18  2003 53127\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   958B Mar 18  2003 53130\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.5K Mar 18  2003 53131\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.0K Mar 18  2003 53132\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   640B Mar 18  2003 53133\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.3K Mar 18  2003 53134\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.1K Mar 18  2003 53135\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   4.2K Mar 18  2003 53136\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   4.8K Mar 18  2003 53137\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.3K Mar 18  2003 53139\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   3.0K Mar 18  2003 53140\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.1K Mar 18  2003 53141\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   456B Mar 18  2003 53142\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   760B Mar 18  2003 53143\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   768B Mar 18  2003 53144\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.0K Mar 18  2003 53145\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.2K Mar 18  2003 53149\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.1K Mar 18  2003 53150\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.3K Mar 18  2003 53151\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.9K Mar 18  2003 53153\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.2K Mar 18  2003 53154\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.2K Mar 18  2003 53157\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.0K Mar 18  2003 53158\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   819B Mar 18  2003 53159\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.9K Mar 18  2003 53160\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   3.5K Mar 18  2003 53161\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.5K Mar 18  2003 53162\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.9K Mar 18  2003 53163\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.2K Mar 18  2003 53164\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.1K Mar 18  2003 53165\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   684B Mar 18  2003 53166\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   443B Mar 18  2003 53167\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.2K Mar 18  2003 53168\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.4K Mar 18  2003 53170\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.5K Mar 18  2003 53171\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   785B Mar 18  2003 53172\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.1K Mar 18  2003 53173\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.3K Mar 18  2003 53174\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   737B Mar 18  2003 53175\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.3K Mar 18  2003 53176\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.8K Mar 18  2003 53177\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.2K Mar 18  2003 53178\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.6K Mar 18  2003 53179\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.1K Mar 18  2003 53180\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   3.2K Mar 18  2003 53181\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.2K Mar 18  2003 53182\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.4K Mar 18  2003 53183\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.7K Mar 18  2003 53184\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.6K Mar 18  2003 53185\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   3.0K Mar 18  2003 53186\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   665B Mar 18  2003 53187\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.0K Mar 18  2003 53188\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.3K Mar 18  2003 53190\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.9K Mar 18  2003 53191\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.8K Mar 18  2003 53192\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.4K Mar 18  2003 53193\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   792B Mar 18  2003 53194\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.0K Mar 18  2003 53195\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.6K Mar 18  2003 53196\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.6K Mar 18  2003 53197\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.1K Mar 18  2003 53198\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.4K Mar 18  2003 53199\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.3K Mar 18  2003 53201\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.3K Mar 18  2003 53203\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   3.7K Mar 18  2003 53208\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.1K Mar 18  2003 53209\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.5K Mar 18  2003 53210\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.7K Mar 18  2003 53211\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.4K Mar 18  2003 53212\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.3K Mar 18  2003 53213\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.9K Mar 18  2003 53214\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   919B Mar 18  2003 53215\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   868B Mar 18  2003 53216\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.3K Mar 18  2003 53217\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.0K Mar 18  2003 53218\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.1K Mar 18  2003 53219\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   640B Mar 18  2003 53220\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.1K Mar 18  2003 53221\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.0K Mar 18  2003 53222\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.0K Mar 18  2003 53223\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   3.4K Mar 18  2003 53224\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   808B Mar 18  2003 53225\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.0K Mar 18  2003 53226\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   640B Mar 18  2003 53228\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   856B Mar 18  2003 53229\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   967B Mar 18  2003 53230\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   781B Mar 18  2003 53231\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.2K Mar 18  2003 53232\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.2K Mar 18  2003 53235\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.7K Mar 18  2003 53237\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.2K Mar 18  2003 53238\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.4K Mar 18  2003 53239\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.2K Mar 18  2003 53240\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.3K Mar 18  2003 53243\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.3K Mar 18  2003 53248\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.4K Mar 18  2003 53249\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.8K Mar 18  2003 53250\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.5K Mar 18  2003 53251\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.4K Mar 18  2003 53252\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.2K Mar 18  2003 53256\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   806B Mar 18  2003 53258\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   4.2K Mar 18  2003 53266\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   3.5K Mar 18  2003 53267\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.8K Mar 18  2003 53269\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   3.2K Mar 18  2003 53271\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.3K Mar 18  2003 53274\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.1K Mar 18  2003 53275\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.0K Mar 18  2003 53281\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   958B Mar 18  2003 53282\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   3.2K Mar 18  2003 53283\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   872B Mar 18  2003 53284\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   387B Mar 18  2003 53285\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   3.1K Mar 18  2003 53286\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   3.5K Mar 18  2003 53287\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.6K Mar 18  2003 53288\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   956B Mar 18  2003 53289\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.6K Mar 18  2003 53290\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff    10K Mar 18  2003 53292\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   5.4K Mar 18  2003 53298\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   945B Mar 18  2003 53303\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.2K Mar 18  2003 53304\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.5K Mar 18  2003 53305\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.4K Mar 18  2003 53306\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   590B Mar 18  2003 53307\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   663B Mar 18  2003 53308\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   907B Mar 18  2003 53309\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.0K Mar 18  2003 53311\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.5K Mar 18  2003 53312\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   576B Mar 18  2003 53314\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff    15K Mar 18  2003 53323\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.0K Mar 18  2003 53334\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   783B Mar 18  2003 53347\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   5.8K Mar 18  2003 53351\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.6K Mar 18  2003 53366\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   698B Mar 18  2003 53370\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   600B Mar 18  2003 53371\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   5.6K Mar 18  2003 53373\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.8K Mar 18  2003 53374\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.1K Mar 18  2003 53375\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   849B Mar 18  2003 53376\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   621B Mar 18  2003 53377\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   270B Mar 18  2003 53380\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.1K Mar 18  2003 53381\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.2K Mar 18  2003 53382\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.6K Mar 18  2003 53383\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.6K Mar 18  2003 53387\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   759B Mar 18  2003 53389\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   396B Mar 18  2003 53390\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   669B Mar 18  2003 53391\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.8K Mar 18  2003 53434\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.6K Mar 18  2003 53435\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   708B Mar 18  2003 53436\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   887B Mar 18  2003 53437\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   838B Mar 18  2003 53438\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.4K Mar 18  2003 53439\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.3K Mar 18  2003 53440\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   384B Mar 18  2003 53441\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   857B Mar 18  2003 53442\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.6K Mar 18  2003 53443\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.4K Mar 18  2003 53445\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.3K Mar 18  2003 53449\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.4K Mar 18  2003 53459\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.4K Mar 18  2003 53460\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.0K Mar 18  2003 53465\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.3K Mar 18  2003 53466\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.0K Mar 18  2003 53467\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.4K Mar 18  2003 53468\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.1K Mar 18  2003 53471\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.9K Mar 18  2003 53477\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   718B Mar 18  2003 53478\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   781B Mar 18  2003 53483\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.6K Mar 18  2003 53509\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   910B Mar 18  2003 53510\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   781B Mar 18  2003 53512\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.8K Mar 18  2003 53515\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.1K Mar 18  2003 53518\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff    50K Mar 18  2003 53519\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   6.0K Mar 18  2003 53521\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.0K Mar 18  2003 53522\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.8K Mar 18  2003 53523\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   338B Mar 18  2003 53524\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.4K Mar 18  2003 53525\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   489B Mar 18  2003 53526\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.6K Mar 18  2003 53527\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.4K Mar 18  2003 53528\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   228B Mar 18  2003 53529\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.1K Mar 18  2003 53531\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.3K Mar 18  2003 53532\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.2K Mar 18  2003 53533\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   356B Mar 18  2003 53534\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   614B Mar 18  2003 53535\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   895B Mar 18  2003 53571\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.0K Mar 18  2003 53572\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   697B Mar 18  2003 53573\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.0K Mar 18  2003 53574\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.8K Mar 18  2003 53654\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.3K Mar 18  2003 53655\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.5K Mar 18  2003 53656\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.1K Mar 18  2003 53660\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   6.8K Mar 18  2003 53661\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.8K Mar 18  2003 53753\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   698B Mar 18  2003 53754\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   779B Mar 18  2003 53755\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   3.9K Mar 18  2003 53756\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.3K Mar 18  2003 53757\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.2K Mar 18  2003 53758\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   745B Mar 18  2003 53759\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.9K Mar 18  2003 53760\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   592B Mar 18  2003 53761\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   658B Mar 18  2003 53762\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   756B Mar 18  2003 53763\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.7K Mar 18  2003 53764\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.1K Mar 18  2003 53765\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   906B Mar 18  2003 53766\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   535B Mar 18  2003 53780\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.3K Mar 18  2003 53785\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.3K Mar 18  2003 54165\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.8K Mar 18  2003 54166\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   547B Mar 18  2003 54167\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.4K Mar 18  2003 54168\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   4.7K Mar 18  2003 54178\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.8K Mar 18  2003 54179\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   4.4K Mar 18  2003 54180\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.3K Mar 18  2003 54181\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   3.0K Mar 18  2003 54182\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.4K Mar 18  2003 54198\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.8K Mar 18  2003 54199\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   2.5K Mar 18  2003 54200\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.7K Mar 18  2003 54201\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.0K Mar 18  2003 54202\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.2K Mar 18  2003 54203\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   565B Mar 18  2003 54204\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   641B Mar 18  2003 54227\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.0K Mar 18  2003 54228\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   877B Mar 18  2003 54470\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   1.0K Mar 18  2003 54471\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   993B Mar 18  2003 54472\r\n",
        "-rw-r--r--  1 marcelcaraciolo  staff   434B Mar 18  2003 54473\r\n"
       ]
      }
     ],
     "prompt_number": 57
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "The `load_files` function can load text files from a 2 levels folder structure assuming folder names represent categories:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "#print(load_files.__doc__)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 58
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "all_twenty_train = load_files('../data/twenty_newsgroups/20news-bydate-train/',\n",
      "  charset='latin-1', random_state=42)\n",
      "all_twenty_test = load_files('../data/twenty_newsgroups/20news-bydate-test/',\n",
      "    charset='latin-1', random_state=42)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stderr",
       "text": [
        "/Library/Frameworks/Python.framework/Versions/7.2/lib/python2.7/site-packages/scikit_learn-0.14.1-py2.7-macosx-10.5-i386.egg/sklearn/datasets/base.py:161: DeprecationWarning: The charset parameter is deprecated as of version 0.14 and will be removed in 0.16. Use encode instead.\n",
        "  DeprecationWarning)\n",
        "/Library/Frameworks/Python.framework/Versions/7.2/lib/python2.7/site-packages/scikit_learn-0.14.1-py2.7-macosx-10.5-i386.egg/sklearn/datasets/base.py:161: DeprecationWarning: The charset parameter is deprecated as of version 0.14 and will be removed in 0.16. Use encode instead.\n",
        "  DeprecationWarning)\n"
       ]
      }
     ],
     "prompt_number": 60
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "all_target_names = all_twenty_train.target_names\n",
      "all_target_names"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 61,
       "text": [
        "['alt.atheism',\n",
        " 'comp.graphics',\n",
        " 'comp.os.ms-windows.misc',\n",
        " 'comp.sys.ibm.pc.hardware',\n",
        " 'comp.sys.mac.hardware',\n",
        " 'comp.windows.x',\n",
        " 'misc.forsale',\n",
        " 'rec.autos',\n",
        " 'rec.motorcycles',\n",
        " 'rec.sport.baseball',\n",
        " 'rec.sport.hockey',\n",
        " 'sci.crypt',\n",
        " 'sci.electronics',\n",
        " 'sci.med',\n",
        " 'sci.space',\n",
        " 'soc.religion.christian',\n",
        " 'talk.politics.guns',\n",
        " 'talk.politics.mideast',\n",
        " 'talk.politics.misc',\n",
        " 'talk.religion.misc']"
       ]
      }
     ],
     "prompt_number": 61
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "all_twenty_train.target"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 62,
       "text": [
        "array([12,  6,  9, ...,  9,  1, 12])"
       ]
      }
     ],
     "prompt_number": 62
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "all_twenty_train.target.shape"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 63,
       "text": [
        "(11314,)"
       ]
      }
     ],
     "prompt_number": 63
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "all_twenty_test.target.shape"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 64,
       "text": [
        "(7532,)"
       ]
      }
     ],
     "prompt_number": 64
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "len(all_twenty_train.data)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 65,
       "text": [
        "11314"
       ]
      }
     ],
     "prompt_number": 65
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "type(all_twenty_train.data[0])"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 66,
       "text": [
        "unicode"
       ]
      }
     ],
     "prompt_number": 66
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def display_sample(i, dataset):\n",
      "    print(\"Class name: \" + dataset.target_names[dataset.target[i]])\n",
      "    print(\"Text content:\\n\")\n",
      "    print(dataset.data[i])"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 67
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "display_sample(0, all_twenty_train)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Class name: sci.electronics\n",
        "Text content:\n",
        "\n",
        "From: wtm@uhura.neoucom.edu (Bill Mayhew)\n",
        "Subject: Re: How to the disks copy protected.\n",
        "Organization: Northeastern Ohio Universities College of Medicine\n",
        "Lines: 23\n",
        "\n",
        "Write a good manual to go with the software.  The hassle of\n",
        "photocopying the manual is offset by simplicity of purchasing\n",
        "the package for only $15.  Also, consider offering an inexpensive\n",
        "but attractive perc for registered users.  For instance, a coffee\n",
        "mug.  You could produce and mail the incentive for a couple of\n",
        "dollars, so consider pricing the product at $17.95.\n",
        "\n",
        "You're lucky if only 20% of the instances of your program in use\n",
        "are non-licensed users.\n",
        "\n",
        "The best approach is to estimate your loss and accomodate that into\n",
        "your price structure.  Sure it hurts legitimate users, but too bad.\n",
        "Retailers have to charge off loss to shoplifters onto paying\n",
        "customers; the software industry is the same.\n",
        "\n",
        "Unless your product is exceptionally unique, using an ostensibly\n",
        "copy-proof disk will just send your customers to the competetion.\n",
        "\n",
        "\n",
        "-- \n",
        "Bill Mayhew      NEOUCOM Computer Services Department\n",
        "Rootstown, OH  44272-9995  USA    phone: 216-325-2511\n",
        "wtm@uhura.neoucom.edu (140.220.1.1)    146.580: N8WED\n",
        "\n"
       ]
      }
     ],
     "prompt_number": 68
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "display_sample(1, all_twenty_train)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Class name: misc.forsale\n",
        "Text content:\n",
        "\n",
        "From: andy@SAIL.Stanford.EDU (Andy Freeman)\n",
        "Subject: Re: Catalog of Hard-to-Find PC Enhancements (Repost)\n",
        "Organization: Computer Science Department,  Stanford University.\n",
        "Lines: 33\n",
        "\n",
        ">andy@SAIL.Stanford.EDU (Andy Freeman) writes:\n",
        ">> >In article <C5ELME.4z4@unix.portal.com> jdoll@shell.portal.com (Joe Doll) wr\n",
        ">> >>   \"The Catalog of Personal Computing Tools for Engineers and Scien-\n",
        ">> >>   tists\" lists hardware cards and application software packages for \n",
        ">> >>   PC/XT/AT/PS/2 class machines.  Focus is on engineering and scien-\n",
        ">> >>   tific applications of PCs, such as data acquisition/control, \n",
        ">> >>   design automation, and data analysis and presentation.  \n",
        ">> >\n",
        ">> >>   If you would like a free copy, reply with your (U. S. Postal) \n",
        ">> >>   mailing address.\n",
        ">> \n",
        ">> Don't bother - it never comes.  It's a cheap trick for building a\n",
        ">> mailing list to sell if my junk mail flow is any indication.\n",
        ">> \n",
        ">> -andy sent his address months ago\n",
        ">\n",
        ">Perhaps we can get Portal to nuke this weasal.  I never received a \n",
        ">catalog either.  If that person doesn't respond to a growing flame, then \n",
        ">we can assume that we'yall look forward to lotsa junk mail.\n",
        "\n",
        "I don't want him nuked, I want him to be honest.  The junk mail has\n",
        "been much more interesting than the promised catalog.  If I'd known\n",
        "what I was going to get, I wouldn't have hesitated.  I wouldn't be\n",
        "surprised if there were other folks who looked at the ad and said\n",
        "\"nope\" but who would be very interested in the junk mail that results.\n",
        "Similarly, there are people who wanted the advertised catalog who\n",
        "aren't happy with the junk they got instead.\n",
        "\n",
        "The folks buying the mailing lists would prefer an honest ad, and\n",
        "so would the people reading it.\n",
        "\n",
        "-andy\n",
        "--\n",
        "\n"
       ]
      }
     ],
     "prompt_number": 69
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Let's compute the (uncompressed, in-memory) size of the training and test sets in MB assuming an 8 bit encoding (in this case, all chars can be encoded using the latin-1 charset)."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def text_size(text, charset='iso-8859-1'):\n",
      "    return len(text.encode(charset)) * 8 * 1e-6\n",
      "\n",
      "train_size_mb = sum(text_size(text) for text in all_twenty_train.data) \n",
      "test_size_mb = sum(text_size(text) for text in all_twenty_test.data)\n",
      "\n",
      "print(\"Training set size: {0} MB\".format(int(train_size_mb)))\n",
      "print(\"Testing set size: {0} MB\".format(int(test_size_mb)))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Training set size: 176 MB\n",
        "Testing set size: 110 MB\n"
       ]
      }
     ],
     "prompt_number": 70
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "If we only consider a small subset of the 4 categories selected from the initial example:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "train_small_size_mb = sum(text_size(text) for text in twenty_train_small.data) \n",
      "test_small_size_mb = sum(text_size(text) for text in twenty_test_small.data)\n",
      "\n",
      "print(\"Training set size: {0} MB\".format(int(train_small_size_mb)))\n",
      "print(\"Testing set size: {0} MB\".format(int(test_small_size_mb)))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Training set size: 31 MB\n",
        "Testing set size: 22 MB\n"
       ]
      }
     ],
     "prompt_number": 71
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "###Extracting Text Features"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from sklearn.feature_extraction.text import TfidfVectorizer\n",
      "\n",
      "TfidfVectorizer()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 73,
       "text": [
        "TfidfVectorizer(analyzer=u'word', binary=False, charset=None,\n",
        "        charset_error=None, decode_error=u'strict',\n",
        "        dtype=<type 'numpy.int64'>, encoding=u'utf-8', input=u'content',\n",
        "        lowercase=True, max_df=1.0, max_features=None, min_df=1,\n",
        "        ngram_range=(1, 1), norm=u'l2', preprocessor=None, smooth_idf=True,\n",
        "        stop_words=None, strip_accents=None, sublinear_tf=False,\n",
        "        token_pattern=u'(?u)\\\\b\\\\w\\\\w+\\\\b', tokenizer=None, use_idf=True,\n",
        "        vocabulary=None)"
       ]
      }
     ],
     "prompt_number": 73
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "vectorizer = TfidfVectorizer(min_df=1)\n",
      "\n",
      "%time X_train_small = vectorizer.fit_transform(twenty_train_small.data)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "CPU times: user 1.09 s, sys: 65.3 ms, total: 1.15 s\n",
        "Wall time: 1.3 s\n"
       ]
      }
     ],
     "prompt_number": 74
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "The results is not a `numpy.array` but instead a `scipy.sparse` matrix. This datastructure is quite similar to a 2D numpy array but it does not store the zeros."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "X_train_small"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 76,
       "text": [
        "<2034x34118 sparse matrix of type '<type 'numpy.float64'>'\n",
        "\twith 323433 stored elements in Compressed Sparse Row format>"
       ]
      }
     ],
     "prompt_number": 76
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "scipy.sparse matrices also have a shape attribute to access the dimensions:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "n_samples, n_features = X_train_small.shape"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 77
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "This dataset has around 2000 samples (the rows of the data matrix):"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "n_samples"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 78,
       "text": [
        "2034"
       ]
      }
     ],
     "prompt_number": 78
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "This is the same value as the number of strings in the original list of text documents:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "len(twenty_train_small.data)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 79,
       "text": [
        "2034"
       ]
      }
     ],
     "prompt_number": 79
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "The columns represent the individual token occurrences:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "n_features"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 80,
       "text": [
        "34118"
       ]
      }
     ],
     "prompt_number": 80
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "This number is the size of the vocabulary of the model extracted during fit in a Python dictionary:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "type(vectorizer.vocabulary_)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 81,
       "text": [
        "dict"
       ]
      }
     ],
     "prompt_number": 81
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "len(vectorizer.vocabulary_)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 82,
       "text": [
        "34118"
       ]
      }
     ],
     "prompt_number": 82
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "The keys of the `vocabulary_` attribute are also called feature names and can be accessed as a list of strings."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "len(vectorizer.get_feature_names())\n"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 83,
       "text": [
        "34118"
       ]
      }
     ],
     "prompt_number": 83
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Here are the first 10 elements (sorted in lexicographical order):"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "vectorizer.get_feature_names()[:10]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 91,
       "text": [
        "[u'00',\n",
        " u'000',\n",
        " u'0000',\n",
        " u'00000',\n",
        " u'000000',\n",
        " u'000005102000',\n",
        " u'000021',\n",
        " u'000062david42',\n",
        " u'0000vec',\n",
        " u'0001']"
       ]
      }
     ],
     "prompt_number": 91
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Let's have a look at the features from the middle:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "vectorizer.get_feature_names()[n_features / 2:n_features / 2 + 10]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 89,
       "text": [
        "[u'inadequate',\n",
        " u'inala',\n",
        " u'inalienable',\n",
        " u'inane',\n",
        " u'inanimate',\n",
        " u'inapplicable',\n",
        " u'inappropriate',\n",
        " u'inappropriately',\n",
        " u'inaudible',\n",
        " u'inbreeding']"
       ]
      }
     ],
     "prompt_number": 89
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Now that we have extracted a vector representation of the data, it's a good idea to project the data on the first 2D of a Principal Component Analysis to get a feel of the data. Note that the `RandomizedPCA` class can accept `scipy.sparse` matrices as input (as an alternative to numpy arrays):"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from sklearn.decomposition import RandomizedPCA\n",
      "\n",
      "%time X_train_small_pca = RandomizedPCA(n_components=2).fit_transform(X_train_small)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "CPU times: user 164 ms, sys: 15.5 ms, total: 179 ms\n",
        "Wall time: 393 ms\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stderr",
       "text": [
        "/Library/Frameworks/Python.framework/Versions/7.2/lib/python2.7/site-packages/scikit_learn-0.14.1-py2.7-macosx-10.5-i386.egg/sklearn/decomposition/pca.py:512: DeprecationWarning: Sparse matrix support is deprecated and will be dropped in 0.16. Use TruncatedSVD instead.\n",
        "  DeprecationWarning)\n"
       ]
      }
     ],
     "prompt_number": 92
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from itertools import cycle\n",
      "\n",
      "colors = ['b', 'g', 'r', 'c', 'm', 'y', 'k']\n",
      "for i, c in zip(np.unique(y_train), cycle(colors)):\n",
      "    pl.scatter(X_train_small_pca[y_train == i, 0],\n",
      "               X_train_small_pca[y_train == i, 1],\n",
      "               c=c, label=twenty_train_small.target_names[i], alpha=0.5)\n",
      "    \n",
      "_ = pl.legend(loc='best')"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAl4AAAHACAYAAABzpWYoAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAElFTkSuQmCC\n",
       "text": [
        "<matplotlib.figure.Figure at 0x6d2cad0>"
       ]
      }
     ],
     "prompt_number": 93
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "We can observe that there is a large overlap of the samples from different categories. This is to be expected as the PCA linear projection projects data from a 34118 dimensional space down to 2 dimensions: data that is linearly separable in 34118D is often no longer linearly separable in 2D.\n",
      "    \n",
      "Still we can notice an interesting pattern: the newsgroups on religion and atheism occupy the much the same region and computer graphics and space science / space overlap more together than they do with the religion or atheism newsgroups."
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "###Training a Classifier on Text Features"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "We have previously extracted a vector representation of the training corpus and put it into a variable name `X_train_small`. To train a supervised model, in this case a classifier, we also need "
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "y_train_small = twenty_train_small.target"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 94
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "y_train_small.shape"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 95,
       "text": [
        "(2034,)"
       ]
      }
     ],
     "prompt_number": 95
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "y_train_small"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 96,
       "text": [
        "array([1, 2, 2, ..., 2, 1, 1])"
       ]
      }
     ],
     "prompt_number": 96
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "We can shape that we have the same number of samples for the input data and the labels:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "X_train_small.shape[0] == y_train_small.shape[0]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 97,
       "text": [
        "True"
       ]
      }
     ],
     "prompt_number": 97
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "We can now train a classifier, for instance a Multinomial Naive Bayesian classifier:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from sklearn.naive_bayes import MultinomialNB\n",
      "\n",
      "clf = MultinomialNB(alpha=0.1)\n",
      "clf"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 99,
       "text": [
        "MultinomialNB(alpha=0.1, class_prior=None, fit_prior=True)"
       ]
      }
     ],
     "prompt_number": 99
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "clf.fit(X_train_small, y_train_small)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 100,
       "text": [
        "MultinomialNB(alpha=0.1, class_prior=None, fit_prior=True)"
       ]
      }
     ],
     "prompt_number": 100
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "We can now evaluate the classifier on the testing set. Let's first use the builtin score function, which is the rate of correct classification in the test set:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "X_test_small = vectorizer.transform(twenty_test_small.data)\n",
      "y_test_small = twenty_test_small.target"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 101
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "X_test_small.shape"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 102,
       "text": [
        "(1353, 34118)"
       ]
      }
     ],
     "prompt_number": 102
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "y_test_small.shape"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 103,
       "text": [
        "(1353,)"
       ]
      }
     ],
     "prompt_number": 103
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "clf.score(X_test_small, y_test_small)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 104,
       "text": [
        "0.89652623798965259"
       ]
      }
     ],
     "prompt_number": 104
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "We can also compute the score on the test set and observe that the model is both overfitting and underfitting a bit at the same time:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "clf.score(X_train_small, y_train_small)\n"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 105,
       "text": [
        "0.99262536873156337"
       ]
      }
     ],
     "prompt_number": 105
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "The text vectorizer has many parameters to customize it's behavior, in particular how it extracts tokens:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "TfidfVectorizer()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 107,
       "text": [
        "TfidfVectorizer(analyzer=u'word', binary=False, charset=None,\n",
        "        charset_error=None, decode_error=u'strict',\n",
        "        dtype=<type 'numpy.int64'>, encoding=u'utf-8', input=u'content',\n",
        "        lowercase=True, max_df=1.0, max_features=None, min_df=1,\n",
        "        ngram_range=(1, 1), norm=u'l2', preprocessor=None, smooth_idf=True,\n",
        "        stop_words=None, strip_accents=None, sublinear_tf=False,\n",
        "        token_pattern=u'(?u)\\\\b\\\\w\\\\w+\\\\b', tokenizer=None, use_idf=True,\n",
        "        vocabulary=None)"
       ]
      }
     ],
     "prompt_number": 107
    },
    {
     "cell_type": "code",
     "collapsed": true,
     "input": [
      "print(TfidfVectorizer.__doc__)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Convert a collection of raw documents to a matrix of TF-IDF features.\n",
        "\n",
        "    Equivalent to CountVectorizer followed by TfidfTransformer.\n",
        "\n",
        "    Parameters\n",
        "    ----------\n",
        "    input : string {'filename', 'file', 'content'}\n",
        "        If filename, the sequence passed as an argument to fit is\n",
        "        expected to be a list of filenames that need reading to fetch\n",
        "        the raw content to analyze.\n",
        "\n",
        "        If 'file', the sequence items must have 'read' method (file-like\n",
        "        object) it is called to fetch the bytes in memory.\n",
        "\n",
        "        Otherwise the input is expected to be the sequence strings or\n",
        "        bytes items are expected to be analyzed directly.\n",
        "\n",
        "    encoding : string, 'utf-8' by default.\n",
        "        If bytes or files are given to analyze, this encoding is used to\n",
        "        decode.\n",
        "\n",
        "    decode_error : {'strict', 'ignore', 'replace'}\n",
        "        Instruction on what to do if a byte sequence is given to analyze that\n",
        "        contains characters not of the given `encoding`. By default, it is\n",
        "        'strict', meaning that a UnicodeDecodeError will be raised. Other\n",
        "        values are 'ignore' and 'replace'.\n",
        "\n",
        "    strip_accents : {'ascii', 'unicode', None}\n",
        "        Remove accents during the preprocessing step.\n",
        "        'ascii' is a fast method that only works on characters that have\n",
        "        an direct ASCII mapping.\n",
        "        'unicode' is a slightly slower method that works on any characters.\n",
        "        None (default) does nothing.\n",
        "\n",
        "    analyzer : string, {'word', 'char'} or callable\n",
        "        Whether the feature should be made of word or character n-grams.\n",
        "\n",
        "        If a callable is passed it is used to extract the sequence of features\n",
        "        out of the raw, unprocessed input.\n",
        "\n",
        "    preprocessor : callable or None (default)\n",
        "        Override the preprocessing (string transformation) stage while\n",
        "        preserving the tokenizing and n-grams generation steps.\n",
        "\n",
        "    tokenizer : callable or None (default)\n",
        "        Override the string tokenization step while preserving the\n",
        "        preprocessing and n-grams generation steps.\n",
        "\n",
        "    ngram_range : tuple (min_n, max_n)\n",
        "        The lower and upper boundary of the range of n-values for different\n",
        "        n-grams to be extracted. All values of n such that min_n <= n <= max_n\n",
        "        will be used.\n",
        "\n",
        "    stop_words : string {'english'}, list, or None (default)\n",
        "        If a string, it is passed to _check_stop_list and the appropriate stop\n",
        "        list is returned. 'english' is currently the only supported string\n",
        "        value.\n",
        "\n",
        "        If a list, that list is assumed to contain stop words, all of which\n",
        "        will be removed from the resulting tokens.\n",
        "\n",
        "        If None, no stop words will be used. max_df can be set to a value\n",
        "        in the range [0.7, 1.0) to automatically detect and filter stop\n",
        "        words based on intra corpus document frequency of terms.\n",
        "\n",
        "    lowercase : boolean, default True\n",
        "        Convert all characters to lowercase befor tokenizing.\n",
        "\n",
        "    token_pattern : string\n",
        "        Regular expression denoting what constitutes a \"token\", only used\n",
        "        if `tokenize == 'word'`. The default regexp select tokens of 2\n",
        "        or more letters characters (punctuation is completely ignored\n",
        "        and always treated as a token separator).\n",
        "\n",
        "    max_df : float in range [0.0, 1.0] or int, optional, 1.0 by default\n",
        "        When building the vocabulary ignore terms that have a term frequency\n",
        "        strictly higher than the given threshold (corpus specific stop words).\n",
        "        If float, the parameter represents a proportion of documents, integer\n",
        "        absolute counts.\n",
        "        This parameter is ignored if vocabulary is not None.\n",
        "\n",
        "    min_df : float in range [0.0, 1.0] or int, optional, 1 by default\n",
        "        When building the vocabulary ignore terms that have a term frequency\n",
        "        strictly lower than the given threshold.\n",
        "        This value is also called cut-off in the literature.\n",
        "        If float, the parameter represents a proportion of documents, integer\n",
        "        absolute counts.\n",
        "        This parameter is ignored if vocabulary is not None.\n",
        "\n",
        "    max_features : optional, None by default\n",
        "        If not None, build a vocabulary that only consider the top\n",
        "        max_features ordered by term frequency across the corpus.\n",
        "\n",
        "        This parameter is ignored if vocabulary is not None.\n",
        "\n",
        "    vocabulary : Mapping or iterable, optional\n",
        "        Either a Mapping (e.g., a dict) where keys are terms and values are\n",
        "        indices in the feature matrix, or an iterable over terms. If not\n",
        "        given, a vocabulary is determined from the input documents.\n",
        "\n",
        "    binary : boolean, False by default.\n",
        "        If True, all non zero counts are set to 1. This is useful for discrete\n",
        "        probabilistic models that model binary events rather than integer\n",
        "        counts.\n",
        "\n",
        "    dtype : type, optional\n",
        "        Type of the matrix returned by fit_transform() or transform().\n",
        "\n",
        "    norm : 'l1', 'l2' or None, optional\n",
        "        Norm used to normalize term vectors. None for no normalization.\n",
        "\n",
        "    use_idf : boolean, optional\n",
        "        Enable inverse-document-frequency reweighting.\n",
        "\n",
        "    smooth_idf : boolean, optional\n",
        "        Smooth idf weights by adding one to document frequencies, as if an\n",
        "        extra document was seen containing every term in the collection\n",
        "        exactly once. Prevents zero divisions.\n",
        "\n",
        "    sublinear_tf : boolean, optional\n",
        "        Apply sublinear tf scaling, i.e. replace tf with 1 + log(tf).\n",
        "\n",
        "    See also\n",
        "    --------\n",
        "    CountVectorizer\n",
        "        Tokenize the documents and count the occurrences of token and return\n",
        "        them as a sparse matrix\n",
        "\n",
        "    TfidfTransformer\n",
        "        Apply Term Frequency Inverse Document Frequency normalization to a\n",
        "        sparse matrix of occurrence counts.\n",
        "\n",
        "    \n"
       ]
      }
     ],
     "prompt_number": 108
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "The easiest way to introspect what the vectorizer is actually doing for a given test of parameters is call the `vectorizer.build_analyzer()` to get an instance of the text analyzer it uses to process the text:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "analyzer = TfidfVectorizer().build_analyzer()\n",
      "analyzer(\"I love scikit-learn: this is a cool Python lib!\")"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 109,
       "text": [
        "[u'love', u'scikit', u'learn', u'this', u'is', u'cool', u'python', u'lib']"
       ]
      }
     ],
     "prompt_number": 109
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "You can notice that all the tokens are lowercase, that the single letter word \"I\" was dropped, and that hyphenation is used. Let's change some of that default behavior:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "analyzer = TfidfVectorizer(\n",
      "    preprocessor=lambda text: text,  # disable lowercasing\n",
      "    token_pattern=ur'(?u)\\b[\\w-]+\\b', # treat hyphen as a letter\n",
      "                                      # do not exclude single letter tokens\n",
      ").build_analyzer()\n",
      "\n",
      "analyzer(\"I love scikit-learn: this is a cool Python lib!\")"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 110,
       "text": [
        "[u'I',\n",
        " u'love',\n",
        " u'scikit-learn',\n",
        " u'this',\n",
        " u'is',\n",
        " u'a',\n",
        " u'cool',\n",
        " u'Python',\n",
        " u'lib']"
       ]
      }
     ],
     "prompt_number": 110
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "## References and further reading\n",
      "\n",
      "- Scikit-learn tutorial, http://scikit-learn.org/stable/tutorial/index.html#"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [],
     "language": "python",
     "metadata": {},
     "outputs": []
    }
   ],
   "metadata": {}
  }
 ]
}