{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data Gathering\n",
    "\n",
    "This recipe shows how we will be accessing the datasets necessary for the rest of the book.\n",
    "\n",
    "We start by loading the necessary libraries and resetting the computational graph."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "from tensorflow.python.framework import ops\n",
    "ops.reset_default_graph()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## The Iris Dataset (R. Fisher / Scikit-Learn)\n",
    "\n",
    "One of the most frequently used ML datasets is the iris flower dataset.  We will use the easy import tool, `datasets` from scikit-learn.  You can read more about it here: http://scikit-learn.org/stable/modules/generated/sklearn.datasets.load_iris.html#sklearn.datasets.load_iris"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "150\n",
      "150\n",
      "[ 5.1  3.5  1.4  0.2]\n",
      "{0, 1, 2}\n"
     ]
    }
   ],
   "source": [
    "from sklearn.datasets import load_iris\n",
    "\n",
    "iris = load_iris()\n",
    "print(len(iris.data))\n",
    "print(len(iris.target))\n",
    "print(iris.data[0])\n",
    "print(set(iris.target))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Low Birthrate Dataset (Hosted on Github)\n",
    "\n",
    "The 'Low Birthrate Dataset' is a dataset from a famous study by Hosmer and Lemeshow in 1989 called, \"Low Infant Birth Weight Risk Factor Study\".  It is a very commonly used academic dataset mostly for logistic regression.  We will host this dataset on the public Github here:\n",
    "https://github.com/nfmcclure/tensorflow_cookbook/raw/master/01_Introduction/07_Working_with_Data_Sources/birthweight_data/birthweight.dat"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "189\n",
      "9\n"
     ]
    }
   ],
   "source": [
    "import requests\n",
    "\n",
    "birthdata_url = 'https://github.com/nfmcclure/tensorflow_cookbook/raw/master/01_Introduction/07_Working_with_Data_Sources/birthweight_data/birthweight.dat'\n",
    "birth_file = requests.get(birthdata_url)\n",
    "birth_data = birth_file.text.split('\\r\\n')\n",
    "birth_header = birth_data[0].split('\\t')\n",
    "birth_data = [[float(x) for x in y.split('\\t') if len(x)>=1] for y in birth_data[1:] if len(y)>=1]\n",
    "print(len(birth_data))\n",
    "print(len(birth_data[0]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Housing Price Dataset (UCI)\n",
    "\n",
    "We will also use a housing price dataset from the University of California at Irvine (UCI) Machine Learning Database Repository. It is a great regression dataset to use.  You can read more about it here:\n",
    "https://archive.ics.uci.edu/ml/datasets/Housing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "506\n",
      "14\n"
     ]
    }
   ],
   "source": [
    "import requests\n",
    "\n",
    "housing_url = 'https://archive.ics.uci.edu/ml/machine-learning-databases/housing/housing.data'\n",
    "housing_header = ['CRIM', 'ZN', 'INDUS', 'CHAS', 'NOX', 'RM', 'AGE', 'DIS', 'RAD', 'TAX', 'PTRATIO', 'B', 'LSTAT', 'MEDV']\n",
    "housing_file = requests.get(housing_url)\n",
    "housing_data = [[float(x) for x in y.split(' ') if len(x)>=1] for y in housing_file.text.split('\\n') if len(y)>=1]\n",
    "print(len(housing_data))\n",
    "print(len(housing_data[0]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## MNIST Handwriting Dataset (Yann LeCun)\n",
    "\n",
    "The MNIST Handwritten digit picture dataset is the `Hello World` of image recognition. The famous scientist and researcher, Yann LeCun, hosts it on his webpage here, http://yann.lecun.com/exdb/mnist/ .  But because it is so commonly used, many libraries, including TensorFlow, host it internally.  We will use TensorFlow to access this data as follows.\n",
    "\n",
    "If you haven't downloaded this before, please wait a bit while it downloads"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully downloaded train-images-idx3-ubyte.gz 9912422 bytes.\n",
      "Extracting MNIST_data/train-images-idx3-ubyte.gz\n",
      "Successfully downloaded train-labels-idx1-ubyte.gz 28881 bytes.\n",
      "Extracting MNIST_data/train-labels-idx1-ubyte.gz\n",
      "Successfully downloaded t10k-images-idx3-ubyte.gz 1648877 bytes.\n",
      "Extracting MNIST_data/t10k-images-idx3-ubyte.gz\n",
      "Successfully downloaded t10k-labels-idx1-ubyte.gz 4542 bytes.\n",
      "Extracting MNIST_data/t10k-labels-idx1-ubyte.gz\n",
      "55000\n",
      "10000\n",
      "5000\n",
      "[ 0.  0.  0.  1.  0.  0.  0.  0.  0.  0.]\n"
     ]
    }
   ],
   "source": [
    "from tensorflow.examples.tutorials.mnist import input_data\n",
    "\n",
    "mnist = input_data.read_data_sets(\"MNIST_data/\", one_hot=True)\n",
    "print(len(mnist.train.images))\n",
    "print(len(mnist.test.images))\n",
    "print(len(mnist.validation.images))\n",
    "print(mnist.train.labels[1,:])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## CIFAR-10 Data\n",
    "\n",
    "The CIFAR-10 data ( https://www.cs.toronto.edu/~kriz/cifar.html ) contains 60,000 32x32 color images of 10 classes collected by Alex Krizhevsky, Vinod Nair, and Geoffrey Hinton.  Alex Krizhevsky maintains the page referenced here.  This is such a common dataset, that there are built in functions in TensorFlow to access this data (the keras wrapper has these commands). Note that the keras wrapper for these functions automatically splits the images into a 50,000 training set and a 10,000 test set."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading data from http://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz\n"
     ]
    }
   ],
   "source": [
    "from PIL import Image\n",
    "# Running this command requires an internet connection and a few minutes to download all the images.\n",
    "(X_train, y_train), (X_test, y_test) = tf.contrib.keras.datasets.cifar10.load_data()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The ten categories are (in order):\n",
    "\n",
    "<ol start=\"0\">\n",
    "  <li>Airplane</li>\n",
    "  <li>Automobile</li>\n",
    "  <li>Bird</li>\n",
    "  <li>Car</li>\n",
    "  <li>Deer</li>\n",
    "  <li>Dog</li>\n",
    "  <li>Frog</li>\n",
    "  <li>Horse</li>\n",
    "  <li>Ship</li>\n",
    "  <li>Truck</li>\n",
    "</ol>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(50000, 32, 32, 3)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(50000, 1)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([6], dtype=uint8)"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train[0,] # this is a frog"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7ffb48a47400>"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAP4AAAD9CAYAAACcAsr/AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n/cPnR97oVgEAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7ffb4aa61160>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot the 0-th image (a frog)\n",
    "%matplotlib inline\n",
    "img = Image.fromarray(X_train[0,:,:,:])\n",
    "plt.imshow(img)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Ham/Spam Texts Dataset (UCI)\n",
    "\n",
    "We will use another UCI ML Repository dataset called the SMS Spam Collection.  You can read about it here: https://archive.ics.uci.edu/ml/datasets/SMS+Spam+Collection .  As a sidenote about common terms, when predicting if a data point represents 'spam' (or unwanted advertisement), the alternative is called 'ham' (or useful information).\n",
    "\n",
    "This is a great dataset for predicting a binary outcome (spam/ham) from a textual input.  This will be very useful for short text sequences for Natural Language Processing (Ch 7) and Recurrent Neural Networks (Ch 9)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "5574\n",
      "{'spam', 'ham'}\n",
      "Ok lar... Joking wif u oni...\n"
     ]
    }
   ],
   "source": [
    "import requests\n",
    "import io\n",
    "from zipfile import ZipFile\n",
    "\n",
    "# Get/read zip file\n",
    "zip_url = 'http://archive.ics.uci.edu/ml/machine-learning-databases/00228/smsspamcollection.zip'\n",
    "r = requests.get(zip_url)\n",
    "z = ZipFile(io.BytesIO(r.content))\n",
    "file = z.read('SMSSpamCollection')\n",
    "# Format Data\n",
    "text_data = file.decode()\n",
    "text_data = text_data.encode('ascii',errors='ignore')\n",
    "text_data = text_data.decode().split('\\n')\n",
    "text_data = [x.split('\\t') for x in text_data if len(x)>=1]\n",
    "[text_data_target, text_data_train] = [list(x) for x in zip(*text_data)]\n",
    "print(len(text_data_train))\n",
    "print(set(text_data_target))\n",
    "print(text_data_train[1])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Movie Review Data (Cornell)\n",
    "\n",
    "The Movie Review database, collected by Bo Pang and Lillian Lee (researchers at Cornell), serves as a great dataset to use for predicting a numerical number from textual inputs.\n",
    "\n",
    "You can read more about the dataset and papers using it here:\n",
    "https://www.cs.cornell.edu/people/pabo/movie-review-data/"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "5331\n",
      "5331\n",
      "simplistic , silly and tedious . \n",
      "\n"
     ]
    }
   ],
   "source": [
    "import requests\n",
    "import io\n",
    "import tarfile\n",
    "\n",
    "movie_data_url = 'http://www.cs.cornell.edu/people/pabo/movie-review-data/rt-polaritydata.tar.gz'\n",
    "r = requests.get(movie_data_url)\n",
    "# Stream data into temp object\n",
    "stream_data = io.BytesIO(r.content)\n",
    "tmp = io.BytesIO()\n",
    "while True:\n",
    "    s = stream_data.read(16384)\n",
    "    if not s:  \n",
    "        break\n",
    "    tmp.write(s)\n",
    "stream_data.close()\n",
    "tmp.seek(0)\n",
    "# Extract tar file\n",
    "tar_file = tarfile.open(fileobj=tmp, mode=\"r:gz\")\n",
    "pos = tar_file.extractfile('rt-polaritydata/rt-polarity.pos')\n",
    "neg = tar_file.extractfile('rt-polaritydata/rt-polarity.neg')\n",
    "# Save pos/neg reviews\n",
    "pos_data = []\n",
    "for line in pos:\n",
    "    pos_data.append(line.decode('ISO-8859-1').encode('ascii',errors='ignore').decode())\n",
    "neg_data = []\n",
    "for line in neg:\n",
    "    neg_data.append(line.decode('ISO-8859-1').encode('ascii',errors='ignore').decode())\n",
    "tar_file.close()\n",
    "\n",
    "print(len(pos_data))\n",
    "print(len(neg_data))\n",
    "print(neg_data[0])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## The Complete Works of William Shakespeare (Gutenberg Project)\n",
    "\n",
    "For training a TensorFlow Model to create text, we will train it on the complete works of William Shakespeare.  This can be accessed through the good work of the Gutenberg Project.  The Gutenberg Project frees many non-copyright books by making them accessible for free from the hard work of volunteers.\n",
    "\n",
    "You can read more about the Shakespeare works here:\n",
    "http://www.gutenberg.org/ebooks/100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "5582212\n"
     ]
    }
   ],
   "source": [
    "# The Works of Shakespeare Data\n",
    "import requests\n",
    "\n",
    "shakespeare_url = 'http://www.gutenberg.org/cache/epub/100/pg100.txt'\n",
    "# Get Shakespeare text\n",
    "response = requests.get(shakespeare_url)\n",
    "shakespeare_file = response.content\n",
    "# Decode binary into string\n",
    "shakespeare_text = shakespeare_file.decode('utf-8')\n",
    "# Drop first few descriptive paragraphs.\n",
    "shakespeare_text = shakespeare_text[7675:]\n",
    "print(len(shakespeare_text))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## English-German Sentence Translation Database (Manythings/Tatoeba)\n",
    "\n",
    "The Tatoeba Project is also run by volunteers and is set to make the most bilingual sentence translations available between many different languages.  `Manythings.org` compiles the data and makes it accessible.\n",
    "\n",
    "http://www.manythings.org/corpus/about.html#info\n",
    "\n",
    "More bilingual sentence pairs: http://www.manythings.org/bilingual/"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "147788\n",
      "147788\n",
      "['I won!', 'Ich hab gewonnen!']\n"
     ]
    }
   ],
   "source": [
    "# English-German Sentence Translation Data\n",
    "import requests\n",
    "import io\n",
    "from zipfile import ZipFile\n",
    "sentence_url = 'http://www.manythings.org/anki/deu-eng.zip'\n",
    "r = requests.get(sentence_url)\n",
    "z = ZipFile(io.BytesIO(r.content))\n",
    "file = z.read('deu.txt')\n",
    "# Format Data\n",
    "eng_ger_data = file.decode()\n",
    "eng_ger_data = eng_ger_data.encode('ascii',errors='ignore')\n",
    "eng_ger_data = eng_ger_data.decode().split('\\n')\n",
    "eng_ger_data = [x.split('\\t') for x in eng_ger_data if len(x)>=1]\n",
    "[english_sentence, german_sentence] = [list(x) for x in zip(*eng_ger_data)]\n",
    "print(len(english_sentence))\n",
    "print(len(german_sentence))\n",
    "print(eng_ger_data[10])"
   ]
  }
 ],
 "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": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}