Added a markdown comment about my findings

.envrc

@@ -0,0 +1 @@
+layout python3

.gitignore

@@ -65,3 +65,8 @@ docs/_build/
 # PyBuilder
 target/
 
+# direnv
+.direnv/
+
+# Data
+spambase/

Spam_Detector.ipynb

@@ -0,0 +1,130 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
+   "source": [
+    "import sklearn\n",
+    "from sklearn.naive_bayes import MultinomialNB\n",
+    "classifier = MultinomialNB()\n",
+    "import matplotlib.pyplot as plt\n",
+    "import pandas as pd\n",
+    "from sklearn.cross_validation import train_test_split "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
+   "source": [
+    "%matplotlib inline"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
+   "source": [
+    "with open(\"spambase/spambase.names\") as file:\n",
+    "    names = file.readlines()[33:]\n",
+    "names = [name.split(\":\")[0] for name in names]\n",
+    "names.append(\"spam_flag\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
+   "source": [
+    "data = pd.read_csv(\"spambase/spambase.data\", names=names)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
+   "source": [
+    "train_data, test_data = train_test_split(data, test_size = .4)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "0.80825638240086906"
+      ]
+     },
+     "execution_count": 6,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "bayes = MultinomialNB()\n",
+    "bayes.fit(train_data[names[:-1]], train_data.spam_flag)\n",
+    "bayes.score(test_data[names[:-1]], test_data.spam_flag)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "collapsed": true
+   },
+   "source": [
+    "The Spam detector appears to be fairly accurate.  It has an $R^2$ score of .808."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.4.3"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 0
+}