Digit_Recognizer_Forest.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# MNIST Digit Recognizer\n",
    "\n",
    "This competition is hosted on [Kaggle](https://www.kaggle.com/c/digit-recognizer/data), and the data are stored [locally](C:\\Users\\DMacKellar\\Documents\\Python\\Kaggle\\Digit\\train.csv)\n",
    "\n",
    "This is my first foray into applying what I've learned in data science with a hands-on goal of predicting hand-written digits from the MNIST data set.  As a classical classification problem, I think that using decision trees/random forests is a decent primary approach, although I am interested in exploring this data set with other tools, which I'll carry out later in other notebooks.\n",
    "\n",
    "Please feel free to fork/comment/upbraid, etc."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "from time import clock"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Load in the test and train data sets with pandas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 11 s, sys: 1.63 s, total: 12.6 s\n",
      "Wall time: 14.4 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "\n",
    "# on the PC:\n",
    "# train = pd.read_csv('C:\\\\Users\\\\DMacKellar\\\\Documents\\\\Python\\\\Kaggle\\\\Digit\\\\train.csv')\n",
    "\n",
    "# test = pd.read_csv('C:\\\\Users\\\\DMacKellar\\\\Documents\\\\Python\\\\Kaggle\\\\Digit\\\\test.csv')\n",
    "\n",
    "# on the Mac:\n",
    "train = pd.read_csv('/Users/drew/Documents/Data/R/Kaggle_Digit_Recognizer/train.csv')\n",
    "\n",
    "test = pd.read_csv('/Users/drew/Documents/Data/R/Kaggle_Digit_Recognizer/test.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "And check the format of both:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>label</th>\n",
       "      <th>pixel0</th>\n",
       "      <th>pixel1</th>\n",
       "      <th>pixel2</th>\n",
       "      <th>pixel3</th>\n",
       "      <th>pixel4</th>\n",
       "      <th>pixel5</th>\n",
       "      <th>pixel6</th>\n",
       "      <th>pixel7</th>\n",
       "      <th>pixel8</th>\n",
       "      <th>...</th>\n",
       "      <th>pixel774</th>\n",
       "      <th>pixel775</th>\n",
       "      <th>pixel776</th>\n",
       "      <th>pixel777</th>\n",
       "      <th>pixel778</th>\n",
       "      <th>pixel779</th>\n",
       "      <th>pixel780</th>\n",
       "      <th>pixel781</th>\n",
       "      <th>pixel782</th>\n",
       "      <th>pixel783</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 785 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   label  pixel0  pixel1  pixel2  pixel3  pixel4  pixel5  pixel6  pixel7  \\\n",
       "0      1       0       0       0       0       0       0       0       0   \n",
       "1      0       0       0       0       0       0       0       0       0   \n",
       "2      1       0       0       0       0       0       0       0       0   \n",
       "3      4       0       0       0       0       0       0       0       0   \n",
       "4      0       0       0       0       0       0       0       0       0   \n",
       "\n",
       "   pixel8    ...     pixel774  pixel775  pixel776  pixel777  pixel778  \\\n",
       "0       0    ...            0         0         0         0         0   \n",
       "1       0    ...            0         0         0         0         0   \n",
       "2       0    ...            0         0         0         0         0   \n",
       "3       0    ...            0         0         0         0         0   \n",
       "4       0    ...            0         0         0         0         0   \n",
       "\n",
       "   pixel779  pixel780  pixel781  pixel782  pixel783  \n",
       "0         0         0         0         0         0  \n",
       "1         0         0         0         0         0  \n",
       "2         0         0         0         0         0  \n",
       "3         0         0         0         0         0  \n",
       "4         0         0         0         0         0  \n",
       "\n",
       "[5 rows x 785 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>pixel0</th>\n",
       "      <th>pixel1</th>\n",
       "      <th>pixel2</th>\n",
       "      <th>pixel3</th>\n",
       "      <th>pixel4</th>\n",
       "      <th>pixel5</th>\n",
       "      <th>pixel6</th>\n",
       "      <th>pixel7</th>\n",
       "      <th>pixel8</th>\n",
       "      <th>pixel9</th>\n",
       "      <th>...</th>\n",
       "      <th>pixel774</th>\n",
       "      <th>pixel775</th>\n",
       "      <th>pixel776</th>\n",
       "      <th>pixel777</th>\n",
       "      <th>pixel778</th>\n",
       "      <th>pixel779</th>\n",
       "      <th>pixel780</th>\n",
       "      <th>pixel781</th>\n",
       "      <th>pixel782</th>\n",
       "      <th>pixel783</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 784 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   pixel0  pixel1  pixel2  pixel3  pixel4  pixel5  pixel6  pixel7  pixel8  \\\n",
       "0       0       0       0       0       0       0       0       0       0   \n",
       "1       0       0       0       0       0       0       0       0       0   \n",
       "2       0       0       0       0       0       0       0       0       0   \n",
       "3       0       0       0       0       0       0       0       0       0   \n",
       "4       0       0       0       0       0       0       0       0       0   \n",
       "\n",
       "   pixel9    ...     pixel774  pixel775  pixel776  pixel777  pixel778  \\\n",
       "0       0    ...            0         0         0         0         0   \n",
       "1       0    ...            0         0         0         0         0   \n",
       "2       0    ...            0         0         0         0         0   \n",
       "3       0    ...            0         0         0         0         0   \n",
       "4       0    ...            0         0         0         0         0   \n",
       "\n",
       "   pixel779  pixel780  pixel781  pixel782  pixel783  \n",
       "0         0         0         0         0         0  \n",
       "1         0         0         0         0         0  \n",
       "2         0         0         0         0         0  \n",
       "3         0         0         0         0         0  \n",
       "4         0         0         0         0         0  \n",
       "\n",
       "[5 rows x 784 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I'll also want to take a look at the data represented as images, using PyPlot's imshow:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAk4AAAJCCAYAAAA2m0iOAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10ecd07b8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot some of the numbers\n",
    "plt.figure(figsize=(6, 6))\n",
    "for digit in range(0,70):\n",
    "    plt.subplot(7,10,digit+1)\n",
    "    grid_data = train.iloc[digit,1:].as_matrix().reshape(28,28)  # reshape from 1d to 2d pixel array\n",
    "    plt.imshow(grid_data, interpolation = \"none\", cmap = \"bone_r\", label=train.iloc[digit,0])\n",
    "    title = train.iloc[digit,0]\n",
    "    plt.title(title)\n",
    "    plt.xticks([])\n",
    "    plt.yticks([])\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "These data look pretty variable (especially the first numeral of the bottom row: that's a one?); a more complicated approach like a neural net might be necessary, but I'll stick with a random forest to start."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn import model_selection\n",
    "train_labels = train.iloc[:,0]\n",
    "train_pixels = train.iloc[:,1:]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, this is essentially my first time using scikit-learn's Random Forest Classifier, and I don't yet know how all of the parameters will affect the outcome ([docs here](http://scikit-learn.org/stable/modules/ensemble.html#forests-of-randomized-trees)).  I'll probably have to play with n_estimators and n_features a bit."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[   1.    2.    5.    9.   18.   35.   68.  133.  258.  501.]\n",
      "28.0178514522\n",
      "(42000, 785)\n"
     ]
    }
   ],
   "source": [
    "print(np.round(np.logspace(0.1, 2.7, num=10)))\n",
    "print(np.sqrt(785))\n",
    "print(train.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Following [Samuel Flender's example](https://www.kaggle.com/sflender/digit-recognizer/comparing-random-forest-pca-and-knn), I'm going to try further splitting the train data to offer a means of evaluating different parameter sets and output models from sklearn.  \n",
    "\n",
    "In fact, I've basically just copied his code below for setting up a loop to split the training data set into smaller subsets, iterate through calls to RandomForestClassifier with different values for the number of estimators, then evaluate each output decision tree's accuracy classifying other subsets of values from the training data set."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "def evaluate_classifier(clf, data, target, split_ratio):\n",
    "    # train_test_split returns: splitting : list, length=2 * len(arrays)\n",
    "    trainX, testX, trainY, testY = train_test_split(data, target, train_size=split_ratio, random_state=0)\n",
    "    clf.fit(trainX, trainY)   # this is the actual call to build the decision trees; where most of the comp time comes in\n",
    "    return clf.score(testX,testY)   # outputs score only (freq pred for testX matches label testY); doesn't keep the trees"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1, 2, 5, 9, 18, 35, 68, 133, 258, 501]\n",
      "CPU times: user 215 µs, sys: 17 µs, total: 232 µs\n",
      "Wall time: 238 µs\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "# set up a broad range of estimators to try\n",
    "n_est = np.round(np.logspace(0.1, 2.7, num=10)).tolist()\n",
    "n_estimators_array = []\n",
    "for i in n_est:\n",
    "    n_estimators_array.append(int(i))\n",
    "# set up a broad range of features to try\n",
    "# rf_1 = RandomForestClassifier(n_estimators=200)\n",
    "# rf_2 = rf_1.fit(train_pixels, train_labels)\n",
    "print(n_estimators_array)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of estimators:  1 \tTime taken:  0.317  seconds\n",
      "Number of estimators:  2 \tTime taken:  0.439  seconds\n",
      "Number of estimators:  5 \tTime taken:  0.68  seconds\n",
      "Number of estimators:  9 \tTime taken:  0.993  seconds\n",
      "Number of estimators:  18 \tTime taken:  1.738  seconds\n",
      "Number of estimators:  35 \tTime taken:  3.136  seconds\n",
      "Number of estimators:  68 \tTime taken:  5.954  seconds\n",
      "Number of estimators:  133 \tTime taken:  11.3  seconds\n",
      "Number of estimators:  258 \tTime taken:  21.955  seconds\n",
      "Number of estimators:  501 \tTime taken:  41.528  seconds\n"
     ]
    }
   ],
   "source": [
    "n_samples = 10\n",
    "n_grid = len(n_estimators_array)\n",
    "score_array_mu = np.zeros(n_grid)\n",
    "score_array_sigma = np.zeros(n_grid)\n",
    "j=0\n",
    "timer = clock()\n",
    "for n_estimators in n_estimators_array:\n",
    "    # taking 10 samples for each forest, to return average and stdev of accuracy scores\n",
    "    score_array=np.zeros(n_samples)\n",
    "    for i in range(0,n_samples):\n",
    "        clf = RandomForestClassifier(n_estimators = n_estimators, n_jobs=-1, criterion=\"gini\")\n",
    "        # split_ratio=0.8 means 80/20 split in train/test for each classifier\n",
    "        # only using 1,000 samples to build trees\n",
    "        score_array[i] = evaluate_classifier(clf, train_pixels.iloc[0:1000], train_labels.iloc[0:1000], 0.8)\n",
    "    print('Number of estimators: ', n_estimators, '\\tTime taken: ', round(clock() - timer, 3), ' seconds')\n",
    "    score_array_mu[j], score_array_sigma[j] = np.mean(score_array), np.std(score_array)\n",
    "    j=j+1\n",
    "    timer = clock()\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAnEAAAFNCAYAAABv3TlzAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1146d8c18>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# it looks like the performance saturates around 50-100 estimators\n",
    "\n",
    "plt.figure(figsize=(7,3))\n",
    "plt.errorbar(n_estimators_array, score_array_mu, yerr=score_array_sigma, fmt='k.-')\n",
    "plt.xscale(\"log\")\n",
    "plt.xlabel(\"number of estimators\",size = 20)\n",
    "plt.ylabel(\"accuracy\",size = 20)\n",
    "plt.xlim(0.9,600)\n",
    "plt.ylim(0.5, 0.95)\n",
    "plt.grid(which=\"both\")\n",
    "plt.title('Impact of number of estimators on accuracy')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "So around 100 estimators is approaching saturation, getting close to 90% accuracy.  I wonder if the number of features would affect the outcome.  The default is \"auto\", which takes the square root of the number of features (sqrt(785)~28)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1, 2, 5, 9, 18, 35, 68, 133, 258, 501, 784]\n"
     ]
    }
   ],
   "source": [
    "n_feat = np.round(np.logspace(0.1, 2.7, num=10)).tolist()\n",
    "n_features_array = []\n",
    "for i in n_est:\n",
    "    n_features_array.append(int(i))\n",
    "n_features_array.append(784)\n",
    "print(n_features_array)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of features:  1 \tTime taken:  4.216  seconds\n",
      "Number of features:  2 \tTime taken:  4.346  seconds\n",
      "Number of features:  5 \tTime taken:  4.956  seconds\n",
      "Number of features:  9 \tTime taken:  5.244  seconds\n",
      "Number of features:  18 \tTime taken:  6.948  seconds\n",
      "Number of features:  35 \tTime taken:  9.242  seconds\n",
      "Number of features:  68 \tTime taken:  13.569  seconds\n",
      "Number of features:  133 \tTime taken:  21.506  seconds\n",
      "Number of features:  258 \tTime taken:  36.087  seconds\n",
      "Number of features:  501 \tTime taken:  61.929  seconds\n",
      "Number of features:  784 \tTime taken:  91.768  seconds\n"
     ]
    }
   ],
   "source": [
    "n_samples = 10\n",
    "n_grid = len(n_features_array)\n",
    "score_array_mu2 = np.zeros(n_grid)\n",
    "score_array_sigma2 = np.zeros(n_grid)\n",
    "j=0\n",
    "timer = clock()\n",
    "for n_features in n_features_array:\n",
    "    # taking 10 samples for each forest, to return average and stdev of accuracy scores\n",
    "    score_array=np.zeros(n_samples)\n",
    "    for i in range(0, n_samples):\n",
    "        clf = RandomForestClassifier(n_estimators=100, max_features=n_features, n_jobs=1, criterion=\"gini\")\n",
    "        # split_ratio=0.8 means 80/20 split in train/test for each classifier\n",
    "        # only using 1,000 samples to build trees\n",
    "        score_array[i] = evaluate_classifier(clf, train_pixels.iloc[0:1000], train_labels.iloc[0:1000], 0.8)\n",
    "    print('Number of features: ', n_features, '\\tTime taken: ', round(clock() - timer, 3), ' seconds')\n",
    "    score_array_mu2[j], score_array_sigma2[j] = np.mean(score_array), np.std(score_array)\n",
    "    j=j+1\n",
    "    timer = clock()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAnEAAAFNCAYAAABv3TlzAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n8n8iu9nKLtq4/QAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x114734080>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# it looks like the performance saturates around 50-100 estimators\n",
    "\n",
    "plt.figure(figsize=(7,3))\n",
    "plt.errorbar(n_features_array, score_array_mu2, yerr=score_array_sigma2, fmt='k.-')\n",
    "plt.xscale(\"log\")\n",
    "plt.xlabel(\"number of max features\",size = 20)\n",
    "plt.ylabel(\"accuracy\",size = 20)\n",
    "plt.xlim(0.9,600)\n",
    "plt.ylim(0.5, 0.95)\n",
    "plt.grid(which=\"both\")\n",
    "plt.title('Impact of number of max features on accuracy')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average for 18 features:  0.8795 \tStdev for 18 features:  0.0123389626793\n",
      "Average for 35 features:  0.88 \tStdev for 35 features:  0.0120415945788\n"
     ]
    }
   ],
   "source": [
    "print('Average for %s features: ' % n_features_array[4], score_array_mu2[4], '\\tStdev for 18 features: ', score_array_sigma2[4])\n",
    "print('Average for %s features: ' % n_features_array[5], score_array_mu2[5], '\\tStdev for 35 features: ', score_array_sigma2[5])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So, not a big impact on accuracy.  If anything, there's a slight improvement with fewer than sqrt(features); best accuracy around 18 features.  I could probably try playing with other parameters, but let's try a larger training set next."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.963095238095\n",
      "CPU times: user 27.6 s, sys: 755 ms, total: 28.4 s\n",
      "Wall time: 29.4 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "clf = RandomForestClassifier(n_estimators=100, max_features=20, max_depth=None, n_jobs=1, criterion=\"gini\")\n",
    "trainX1, testX1, trainY1, testY1 = train_test_split(train_pixels, train_labels, train_size=0.8, random_state=0)\n",
    "clf1 = clf.fit(trainX1, trainY1)\n",
    "print(clf1.score(testX1,testY1))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Not bad; I eked out another 6% accuracy with a larger dataset.  As of 20170226, there are about 1400 entries for this competition in Kaggle, and this accuracy score would land at around 1,038.  That's probably about as good as you could expect for this kind of classifier; I think I'll try neural networks next.\n",
    "\n",
    "But first, to retrain with the entire training data set, and prepare the predictions on the test data for submission to Kaggle:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 35.1 s, sys: 615 ms, total: 35.7 s\n",
      "Wall time: 39.6 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "\n",
    "clf2 = clf.fit(train_pixels, train_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "results=clf2.predict(test)\n",
    "\n",
    "test['Label'] = pd.Series(results)\n",
    "test['ImageId'] = test.index +1\n",
    "sub = test[['ImageId','Label']]\n",
    "\n",
    "sub.to_csv('digit_forest_submission.csv', index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Just to check the first 30 digits and predictions output:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAekAAAFlCAYAAADPvBA0AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nXOr/weyXnpSCbB0AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x111a27b00>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot some of the numbers\n",
    "plt.figure(figsize=(5, 5))\n",
    "for digit in range(0,30):\n",
    "    plt.subplot(3,10,digit+1)\n",
    "    grid_data = test.iloc[digit,:-2].as_matrix().reshape(28,28)  # reshape from 1d to 2d pixel array\n",
    "    plt.imshow(grid_data, interpolation = \"none\", cmap = \"bone_r\", label=train.iloc[digit,0])\n",
    "    title = test.iloc[digit,-2]\n",
    "    plt.title(title)\n",
    "    plt.xticks([])\n",
    "    plt.yticks([])\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Looks like it's formatted correctly, and I agree with most of the calls (except the fourth, but hey, nobody's perfect)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python [conda root]",
   "language": "python",
   "name": "conda-root-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}