AMLD introduction to python.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Introduction to python\n",
    "\n",
    "Python is a programming language used in many different areas such as: network programming, webdevelopment, systems and **data science/machine learning/AI**.\n",
    "\n",
    "Python is a dynamic typed language\n",
    "\n",
    "## Good to know packages for data science:\n",
    "\n",
    "### Data science\n",
    "* numpy/scipy (most of the libraries rely on numpy's array)\n",
    "* statmodels (statistics)\n",
    "* pandas (Data frames, data manipulation)\n",
    "* scikit-learn (ML)\n",
    "* matplotlib (plotting)\n",
    "* seaborn (plotting)\n",
    "\n",
    "### ML/AI\n",
    "For ML/AI at least one of the following should be part of a data scientist repertoir\n",
    "* **keras** (very high leve API) \n",
    "* **pytorch** (lower level API, but more flexible. Good for research and maturing)\n",
    "* **tensorflow** (with many APIs, robust for production)\n",
    "* **MXNet** (API: python, julia, R, scala), low level and high level apis, but faster than keras when using the symbolic library (precompiled code)\n",
    " \n",
    "\n",
    "\n",
    "This notebook will just go through the basic topics in order:\n",
    "\n",
    "* Data types\n",
    "    * Numbers\n",
    "    * Strings\n",
    "    * Printing\n",
    "    * Lists\n",
    "    * Dictionaries\n",
    "    * Booleans\n",
    "    * Tuples \n",
    "    * Sets\n",
    "* Comparison Operators\n",
    "* if,elif, else Statements\n",
    "* for Loops\n",
    "* while Loops\n",
    "* range()\n",
    "* list comprehension\n",
    "* functions\n",
    "* lambda expressions\n",
    "* map and filter\n",
    "* methods\n",
    "____"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "def hello_world():\n",
    "    print('Hello World!')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Hello World!\n"
     ]
    }
   ],
   "source": [
    "hello_world()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data types"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Numbers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "1 + 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "1 * 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.5"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "1/2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "16"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "2 ** 4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "4 % 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "50"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(2 + 3) * (5 + 5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Printing many statements in the same cell**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1 + 1 = 2\n",
      "1 * 3 = 3\n",
      "1/2 = 0.5\n",
      "2**4 =  16\n",
      "4 % 2 = 0\n",
      "(2 + 3) * (5 + 5) = 50\n"
     ]
    }
   ],
   "source": [
    "print('1 + 1 =', 1 + 1)\n",
    "print('1 * 3 =', 1 * 3)\n",
    "print('1/2 =', 1/2)\n",
    "print('2**4 = ', 2**4)\n",
    "print('4 % 2 =', 4 % 2)\n",
    "print('(2 + 3) * (5 + 5) =', (2 + 3) * (5 + 5))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Variable assignment"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "my_variable = 5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "5"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "my_variable"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Strings\n",
    "Either with single or double quotes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'my string'"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "'my string'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'my string'"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\"my string\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Python containers\n",
    "* ### Lists\n",
    "* ### Dicts\n",
    "* ### Tuples\n",
    "* ### Sets"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Lists\n",
    "The following is true for lists:\n",
    "* they can hold heterogeneous elements\n",
    "* they are mutable\n",
    "* they can be indexed by integers\n",
    "* their index starts at 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[1, 2, 3]"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "[1,2,3]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['hi', 1, [1, 2]]"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Heterogeneous object\n",
    "['hi',1,[1,2]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "my_list = ['a','b','c']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "my_list.append('d')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['a', 'b', 'c', 'd']"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "my_list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'a'"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# access elements by index\n",
    "my_list[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['b', 'c', 'd']"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# slicing\n",
    "my_list[1:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['a']"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "my_list[:1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "my_list[0] = 'NEW'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['NEW', 'b', 'c', 'd']"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "my_list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['NEW', 'b', 'c', 'd', 'NEW', 'b', 'c', 'd', 'NEW', 'b', 'c', 'd']"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "my_list * 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Dictionaries\n",
    "Dictionaries are hashmaps with key, value pairs. Dicts are indexed by their key."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "d = {'key1':'item1','key2':'item2'}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'key1': 'item1', 'key2': 'item2'}"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "d"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'item1'"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "d['key1']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "key: key1 | value: item1\n",
      "key: key2 | value: item2\n"
     ]
    }
   ],
   "source": [
    "for k, v in d.items():\n",
    "    print('key: {} | value: {}'.format(k, v))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dict_values(['item1', 'item2'])\n"
     ]
    }
   ],
   "source": [
    "print(d.values())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dict_keys(['key1', 'key2'])\n"
     ]
    }
   ],
   "source": [
    "print(d.keys())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Tuples \n",
    "Tuples are inmutable containers. Once created, they cannot be changed in any way. \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "t = (1,2,3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "t[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "#t[0] = 'NEW'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1, 2, 3, 1, 2, 3, 1, 2, 3)"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "t * 3"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Control statemest\n",
    "Python code is structured \n",
    "* if\n",
    "* elif\n",
    "* else"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### if statement"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Hello world!\n"
     ]
    }
   ],
   "source": [
    "if 1 < 2:\n",
    "    print('Hello world!')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Please enter an integer: 0\n",
      "Zero\n"
     ]
    }
   ],
   "source": [
    "# elif can be seen as either switch or case statement used by other languages\n",
    "x = int(input('Please enter an integer: '))\n",
    "if x < 0:\n",
    "    x = 0\n",
    "    print('Negative changed to zero')\n",
    "elif x == 0:\n",
    "    print('Zero')\n",
    "else:\n",
    "    print('More')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### for statement"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cat 3\n",
      "dog 3\n",
      "house 5\n",
      "car 3\n"
     ]
    }
   ],
   "source": [
    "words = ['cat', 'dog', 'house', 'car']\n",
    "for w in words:\n",
    "    print(w, len(w))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['house', 'cat', 'dog', 'house', 'car']"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "for w in words[:]:\n",
    "    if len(w) > 3:\n",
    "        words.insert(0, w) # insert w at index 0\n",
    "words"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0\n",
      "1\n",
      "2\n",
      "3\n",
      "4\n"
     ]
    }
   ],
   "source": [
    "# range\n",
    "for i in range(5):\n",
    "    print(i)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0 Mary\n",
      "1 had\n",
      "2 a\n",
      "3 little\n",
      "4 lamb\n"
     ]
    }
   ],
   "source": [
    "a = ['Mary', 'had', 'a', 'little', 'lamb']\n",
    "for i in range(len(a)):\n",
    "    print(i, a[i])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0 Mary\n",
      "1 had\n",
      "2 a\n",
      "3 little\n",
      "4 lamb\n"
     ]
    }
   ],
   "source": [
    "for idx, value in enumerate(a):\n",
    "    print(idx, value)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### While statement"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0,1,1,2,3,5,8,13,21,34,55,89,144,233,377,610,987,"
     ]
    }
   ],
   "source": [
    "# Fibonacci series\n",
    "a, b = 0, 1\n",
    "while a < 1000:\n",
    "    print(a, end=',')\n",
    "    a, b = b, a + b"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Functions\n",
    "The keyword **def** introduces a function definition. It must be followed by the function name and the parenthesized list of formal parameters. The statements that form the body of the function start at the next line, and must be indented."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "def hello_world():\n",
    "    print('Hello world!')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Hello world!\n"
     ]
    }
   ],
   "source": [
    "hello_world()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "def fib(n):\n",
    "    a, b = 0, 1\n",
    "    while a < n:\n",
    "        print(a, end=' ')\n",
    "        a, b = b, a + b\n",
    "    print()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0 1 1 2 3 5 8 \n"
     ]
    }
   ],
   "source": [
    "fib(9)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0 1 1 2 3 5 8 13 21 34 55 89 144 233 377 610 987 1597 \n"
     ]
    }
   ],
   "source": [
    "fib(2000)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Lambda expressions\n",
    "Small anonymous functions can be created with the lambda keyword. Lambda functions can be used wherever function objects are required. They are syntactically restricted to a single expression. Semantically, they are just syntactic sugar for a normal function definition. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "def make_incrementor(n):\n",
    "    return lambda x: x + n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "f = make_incrementor(42) # Initialize incrementor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "42"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "f(0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "43"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "f(1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Numpy\n",
    "\n",
    "NumPy is the fundamental package for scientific computing with Python. It contains among other things:\n",
    "\n",
    "* a powerful N-dimensional array object\n",
    "* sophisticated (broadcasting) functions\n",
    "* tools for integrating C/C++ and Fortran code\n",
    "* useful linear algebra, Fourier transform, and random number capabilities\n",
    "\n",
    "Besides its obvious scientific uses, NumPy can also be used as an efficient multi-dimensional container of generic data. Arbitrary data-types can be defined. This allows NumPy to seamlessly and speedily integrate with a wide variety of databases.\n",
    "\n",
    "NumPy is licensed under the BSD license, enabling reuse with few restrictions.\n",
    "\n",
    "[Clickable_visible_hyperlink](http://www.numpy.org/)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Array creation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([2, 3, 4])"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# using python list\n",
    "a = np.array([2, 3, 4])\n",
    "a"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dtype('int32')"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a.dtype"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 1.2,  2.5,  6.9])"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "b = np.array([1.2, 2.5, 6.9])\n",
    "b"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dtype('float64')"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "b.dtype"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 1.5,  2. ,  3. ],\n",
       "       [ 4. ,  5. ,  6. ]])"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# matrix or 2-dim array\n",
    "b = np.array([(1.5, 2, 3), (4, 5, 6)])\n",
    "b"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2, 3)"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "b.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 1.+0.j,  2.+0.j],\n",
       "       [ 5.+0.j,  9.+0.j]])"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "c = np.array([[1, 2], [5, 9]], dtype=complex)\n",
    "c"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dtype('complex128')"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "c.dtype"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2, 2)"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "c.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Special functions\n",
    "Creating arrays with special functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 0.,  0.,  0.,  0.],\n",
       "       [ 0.,  0.,  0.,  0.],\n",
       "       [ 0.,  0.,  0.,  0.]])"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.zeros((3, 4))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[[1, 1, 1, 1],\n",
       "        [1, 1, 1, 1],\n",
       "        [1, 1, 1, 1]],\n",
       "\n",
       "       [[1, 1, 1, 1],\n",
       "        [1, 1, 1, 1],\n",
       "        [1, 1, 1, 1]]], dtype=int16)"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ones_array = np.ones((2, 3, 4), dtype=np.int16)\n",
    "ones_array"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2, 3, 4)"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ones_array.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dtype('int16')"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ones_array.dtype"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Create a sequence of numbers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([10, 15, 20, 25])"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# first index: start of sequence\n",
    "# second index: end of sequence, exclusive\n",
    "# third index: step\n",
    "np.arange(10, 30, 5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 0.   ,  1.125,  2.25 ,  3.375,  4.5  ,  5.625,  6.75 ,  7.875,  9.   ])"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# in linspace, specify no of elements instead of step\n",
    "np.linspace(0, 9, 9)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Basic math operations with arrays"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "a: [5 6 7 8] | b: [0 1 2 3]\n"
     ]
    }
   ],
   "source": [
    "a = np.array([5, 6, 7, 8])\n",
    "b = np.arange(4)\n",
    "print('a: {} | b: {}'.format(a, b))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[5 5 5 5]\n",
      "[0 1 4 9]\n",
      "[-9.58924275 -2.79415498  6.56986599  9.89358247]\n"
     ]
    }
   ],
   "source": [
    "print(a - b)\n",
    "print(b**2)\n",
    "print(10 * np.sin(a))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ True,  True,  True,  True], dtype=bool)"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a < 10"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[2, 0],\n",
       "       [0, 4]])"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Elementwise product of matrices\n",
    "\n",
    "A = np.array([[1, 1], [0, 1]])\n",
    "B = np.array([[2, 0], [3, 4]])\n",
    "A * B"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[2, 0],\n",
       "       [0, 4]])"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.multiply(A, B)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Indexing, slicing an iterating"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([  0,   1,   8,  27,  64, 125, 216, 343, 512, 729], dtype=int32)"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a = np.arange(10) ** 3\n",
    "a"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 8, 27, 64], dtype=int32)"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# range chosen [2, 5)\n",
    "a[2:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [],
   "source": [
    "a[:6:2] = -1000 # change every second element"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([-1000,     1, -1000,    27, -1000,   125,   216,   343,   512,   729], dtype=int32)"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([  729,   512,   343,   216,   125, -1000,    27, -1000,     1, -1000], dtype=int32)"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# reverse a\n",
    "a[::-1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Changing the shape of arrays"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 7.,  0.,  6.,  7.],\n",
       "       [ 0.,  4.,  4.,  0.],\n",
       "       [ 6.,  7.,  5.,  5.]])"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a = np.floor(10 * np.random.random((3, 4)))\n",
    "a"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(3, 4)"
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The shape of an array can be changed with various commands. Note that the following three commands all return a modified array, but do not change the original array:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 7.,  0.,  6.,  7.,  0.,  4.,  4.,  0.,  6.,  7.,  5.,  5.])"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a.ravel() # returns a flattened array"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 7.,  0.,  6.,  7.],\n",
       "       [ 0.,  4.,  4.,  0.],\n",
       "       [ 6.,  7.,  5.,  5.]])"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a # does not change a"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 7.,  0.],\n",
       "       [ 6.,  7.],\n",
       "       [ 0.,  4.],\n",
       "       [ 4.,  0.],\n",
       "       [ 6.,  7.],\n",
       "       [ 5.,  5.]])"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a.reshape(6, 2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 7.,  0.,  6.,  7.],\n",
       "       [ 0.,  4.,  4.,  0.],\n",
       "       [ 6.,  7.,  5.,  5.]])"
      ]
     },
     "execution_count": 84,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 7.,  0.,  6.],\n",
       "       [ 0.,  4.,  7.],\n",
       "       [ 6.,  4.,  5.],\n",
       "       [ 7.,  0.,  5.]])"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a.T"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(4, 3)"
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a.T.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 7.,  0.,  6.,  7.,  0.,  4.],\n",
       "       [ 4.,  0.,  6.,  7.,  5.,  5.]])"
      ]
     },
     "execution_count": 87,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a.resize((2, 6)) # resize modifies the array\n",
    "a"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "When one of the dimensions is: **-1**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 7.,  0.,  6.,  7.],\n",
       "       [ 0.,  4.,  4.,  0.],\n",
       "       [ 6.,  7.,  5.,  5.]])"
      ]
     },
     "execution_count": 88,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a.reshape(3, -1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Pandas\n",
    "Pandas is a powerful python data analysis toolkit. Panda provides fast, flexible, and expressive data structures designed to make working with “relational” or “labeled” data both easy and intuitive. It aims to be the fundamental high-level building block for doing practical, real world data analysis in Python. Additionally, it has the broader goal of becoming the most powerful and flexible open source data analysis / manipulation tool available in any language. It is already well on its way toward this goal.\n",
    "\n",
    "**Pandas** is well suited for many different kinds of data:\n",
    "* Tabular data with heterogeneously-typed columns, as in an SQL table or Excel spreadsheet\n",
    "* Ordered and unordered (not necessarily fixed-frequency) time series data.\n",
    "* Arbitrary matrix data (homogeneously typed or heterogeneous) with row and column labels\n",
    "* Any other form of observational / statistical data sets. The data actually need not be labeled at all to be placed into a pandas data structure\n",
    "\n",
    "### Data structures\n",
    "The two most important data structures in pandas are:\n",
    "* **Series**: a one-dimensional array-like object, with an index\n",
    "* **Dataframe**: A DataFrame represents a tabular, spreadsheet-like data structure containing an ordered collection of columns, each of which can be a different value type (numeric, string, boolean, etc.)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Series"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    4\n",
       "1    7\n",
       "2   -5\n",
       "3    3\n",
       "dtype: int64"
      ]
     },
     "execution_count": 90,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "obj = pd.Series([4, 7, -5, 3])\n",
    "obj"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 4,  7, -5,  3], dtype=int64)"
      ]
     },
     "execution_count": 91,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "obj.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RangeIndex(start=0, stop=4, step=1)"
      ]
     },
     "execution_count": 92,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "obj.index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "d    4\n",
       "b    7\n",
       "a   -5\n",
       "c    3\n",
       "dtype: int64"
      ]
     },
     "execution_count": 93,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "obj2 = pd.Series([4, 7, -5, 3], index=['d', 'b', 'a', 'c'])\n",
    "obj2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['d', 'b', 'a', 'c'], dtype='object')"
      ]
     },
     "execution_count": 94,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "obj2.index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-5"
      ]
     },
     "execution_count": 95,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "obj2['a']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "d    6\n",
       "b    7\n",
       "a   -5\n",
       "c    3\n",
       "dtype: int64"
      ]
     },
     "execution_count": 96,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "obj2['d'] = 6\n",
    "obj2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Many operations preserve the relation between values and index\n",
    "* filtering with booleans\n",
    "* multiplication by scalar\n",
    "* applying math operations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "d    6\n",
       "b    7\n",
       "c    3\n",
       "dtype: int64"
      ]
     },
     "execution_count": 97,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# filtering with boolean array\n",
    "obj2[obj2 > 0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "d    12\n",
       "b    14\n",
       "a   -10\n",
       "c     6\n",
       "dtype: int64"
      ]
     },
     "execution_count": 98,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "obj2 * 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "d    6\n",
       "b    7\n",
       "a   -5\n",
       "c    3\n",
       "dtype: int64"
      ]
     },
     "execution_count": 99,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "obj2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "d     403.428793\n",
       "b    1096.633158\n",
       "a       0.006738\n",
       "c      20.085537\n",
       "dtype: float64"
      ]
     },
     "execution_count": 100,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.exp(obj2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Creating a series from a python dict\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Ohio      35000\n",
       "Texas     71000\n",
       "Oregon    16000\n",
       "Utah       5000\n",
       "dtype: int64"
      ]
     },
     "execution_count": 101,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sdata = {'Ohio': 35000, 'Texas': 71000, 'Oregon': 16000, 'Utah': 5000}\n",
    "obj3 = pd.Series(sdata)\n",
    "obj3"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Dataframe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>state</th>\n",
       "      <th>year</th>\n",
       "      <th>pop</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Ohio</td>\n",
       "      <td>2000</td>\n",
       "      <td>1.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Ohio</td>\n",
       "      <td>2001</td>\n",
       "      <td>1.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Ohio</td>\n",
       "      <td>2002</td>\n",
       "      <td>3.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Nevada</td>\n",
       "      <td>2001</td>\n",
       "      <td>2.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Nevada</td>\n",
       "      <td>2002</td>\n",
       "      <td>2.9</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    state  year  pop\n",
       "0    Ohio  2000  1.5\n",
       "1    Ohio  2001  1.7\n",
       "2    Ohio  2002  3.6\n",
       "3  Nevada  2001  2.4\n",
       "4  Nevada  2002  2.9"
      ]
     },
     "execution_count": 102,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = {'state': ['Ohio', 'Ohio', 'Ohio', 'Nevada', 'Nevada'],\n",
    "        'year': [2000, 2001, 2002, 2001, 2002],\n",
    "        'pop': [1.5, 1.7, 3.6, 2.4, 2.9]}\n",
    "\n",
    "df = pd.DataFrame(data)\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {},
   "outputs": [],
   "source": [
    "from utils.utils import PROJECT_DATA_DIR\n",
    "import os"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv(os.path.join(PROJECT_DATA_DIR, 'closing_prices_tiingo.csv'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>date</th>\n",
       "      <th>AAPL</th>\n",
       "      <th>MSFT</th>\n",
       "      <th>FB</th>\n",
       "      <th>AMZN</th>\n",
       "      <th>JNJ</th>\n",
       "      <th>JPM</th>\n",
       "      <th>XOM</th>\n",
       "      <th>GOOGL</th>\n",
       "      <th>GOOG</th>\n",
       "      <th>...</th>\n",
       "      <th>WMT</th>\n",
       "      <th>MDT</th>\n",
       "      <th>GILD</th>\n",
       "      <th>CELG</th>\n",
       "      <th>HON</th>\n",
       "      <th>NVDA</th>\n",
       "      <th>AVGO</th>\n",
       "      <th>BMY</th>\n",
       "      <th>PCLN</th>\n",
       "      <th>ABT</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2001-01-02</td>\n",
       "      <td>0.941914</td>\n",
       "      <td>14.254892</td>\n",
       "      <td>NaN</td>\n",
       "      <td>13.88</td>\n",
       "      <td>32.419918</td>\n",
       "      <td>25.057661</td>\n",
       "      <td>28.026554</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>38.562248</td>\n",
       "      <td>45.207628</td>\n",
       "      <td>2.141676</td>\n",
       "      <td>4.04750</td>\n",
       "      <td>29.239666</td>\n",
       "      <td>4.621229</td>\n",
       "      <td>NaN</td>\n",
       "      <td>36.259178</td>\n",
       "      <td>8.46</td>\n",
       "      <td>13.101557</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2001-01-03</td>\n",
       "      <td>1.036232</td>\n",
       "      <td>15.753331</td>\n",
       "      <td>NaN</td>\n",
       "      <td>17.56</td>\n",
       "      <td>31.386930</td>\n",
       "      <td>29.630184</td>\n",
       "      <td>26.806504</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>41.825868</td>\n",
       "      <td>44.774547</td>\n",
       "      <td>2.230082</td>\n",
       "      <td>3.75000</td>\n",
       "      <td>31.803506</td>\n",
       "      <td>6.039702</td>\n",
       "      <td>NaN</td>\n",
       "      <td>34.020237</td>\n",
       "      <td>10.86</td>\n",
       "      <td>12.810290</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2001-01-04</td>\n",
       "      <td>1.079909</td>\n",
       "      <td>15.917634</td>\n",
       "      <td>NaN</td>\n",
       "      <td>15.50</td>\n",
       "      <td>30.713105</td>\n",
       "      <td>30.434732</td>\n",
       "      <td>26.061268</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>40.215529</td>\n",
       "      <td>41.633941</td>\n",
       "      <td>1.899571</td>\n",
       "      <td>3.16375</td>\n",
       "      <td>31.757251</td>\n",
       "      <td>6.734276</td>\n",
       "      <td>NaN</td>\n",
       "      <td>32.110705</td>\n",
       "      <td>11.64</td>\n",
       "      <td>11.637057</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2001-01-05</td>\n",
       "      <td>1.036232</td>\n",
       "      <td>16.144371</td>\n",
       "      <td>NaN</td>\n",
       "      <td>14.56</td>\n",
       "      <td>31.110408</td>\n",
       "      <td>28.645513</td>\n",
       "      <td>26.177613</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>38.605190</td>\n",
       "      <td>40.827877</td>\n",
       "      <td>1.690112</td>\n",
       "      <td>3.08625</td>\n",
       "      <td>31.017174</td>\n",
       "      <td>5.724829</td>\n",
       "      <td>NaN</td>\n",
       "      <td>32.747215</td>\n",
       "      <td>10.32</td>\n",
       "      <td>11.705110</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2001-01-08</td>\n",
       "      <td>1.048259</td>\n",
       "      <td>16.081936</td>\n",
       "      <td>NaN</td>\n",
       "      <td>14.94</td>\n",
       "      <td>31.069088</td>\n",
       "      <td>28.825636</td>\n",
       "      <td>26.061268</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>38.605190</td>\n",
       "      <td>40.592142</td>\n",
       "      <td>1.571082</td>\n",
       "      <td>2.97625</td>\n",
       "      <td>30.191194</td>\n",
       "      <td>6.271227</td>\n",
       "      <td>NaN</td>\n",
       "      <td>33.129122</td>\n",
       "      <td>9.36</td>\n",
       "      <td>11.585337</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 50 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         date      AAPL       MSFT  FB   AMZN        JNJ        JPM  \\\n",
       "0  2001-01-02  0.941914  14.254892 NaN  13.88  32.419918  25.057661   \n",
       "1  2001-01-03  1.036232  15.753331 NaN  17.56  31.386930  29.630184   \n",
       "2  2001-01-04  1.079909  15.917634 NaN  15.50  30.713105  30.434732   \n",
       "3  2001-01-05  1.036232  16.144371 NaN  14.56  31.110408  28.645513   \n",
       "4  2001-01-08  1.048259  16.081936 NaN  14.94  31.069088  28.825636   \n",
       "\n",
       "         XOM  GOOGL  GOOG    ...            WMT        MDT      GILD     CELG  \\\n",
       "0  28.026554    NaN   NaN    ...      38.562248  45.207628  2.141676  4.04750   \n",
       "1  26.806504    NaN   NaN    ...      41.825868  44.774547  2.230082  3.75000   \n",
       "2  26.061268    NaN   NaN    ...      40.215529  41.633941  1.899571  3.16375   \n",
       "3  26.177613    NaN   NaN    ...      38.605190  40.827877  1.690112  3.08625   \n",
       "4  26.061268    NaN   NaN    ...      38.605190  40.592142  1.571082  2.97625   \n",
       "\n",
       "         HON      NVDA  AVGO        BMY   PCLN        ABT  \n",
       "0  29.239666  4.621229   NaN  36.259178   8.46  13.101557  \n",
       "1  31.803506  6.039702   NaN  34.020237  10.86  12.810290  \n",
       "2  31.757251  6.734276   NaN  32.110705  11.64  11.637057  \n",
       "3  31.017174  5.724829   NaN  32.747215  10.32  11.705110  \n",
       "4  30.191194  6.271227   NaN  33.129122   9.36  11.585337  \n",
       "\n",
       "[5 rows x 50 columns]"
      ]
     },
     "execution_count": 105,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>date</th>\n",
       "      <th>AAPL</th>\n",
       "      <th>MSFT</th>\n",
       "      <th>FB</th>\n",
       "      <th>AMZN</th>\n",
       "      <th>JNJ</th>\n",
       "      <th>JPM</th>\n",
       "      <th>XOM</th>\n",
       "      <th>GOOGL</th>\n",
       "      <th>GOOG</th>\n",
       "      <th>...</th>\n",
       "      <th>WMT</th>\n",
       "      <th>MDT</th>\n",
       "      <th>GILD</th>\n",
       "      <th>CELG</th>\n",
       "      <th>HON</th>\n",
       "      <th>NVDA</th>\n",
       "      <th>AVGO</th>\n",
       "      <th>BMY</th>\n",
       "      <th>PCLN</th>\n",
       "      <th>ABT</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4435</th>\n",
       "      <td>2018-08-20</td>\n",
       "      <td>215.46</td>\n",
       "      <td>106.87</td>\n",
       "      <td>172.500</td>\n",
       "      <td>1876.71</td>\n",
       "      <td>136.88</td>\n",
       "      <td>114.62</td>\n",
       "      <td>78.81</td>\n",
       "      <td>1221.95</td>\n",
       "      <td>1207.77</td>\n",
       "      <td>...</td>\n",
       "      <td>96.00</td>\n",
       "      <td>90.03</td>\n",
       "      <td>72.36</td>\n",
       "      <td>90.88</td>\n",
       "      <td>155.47</td>\n",
       "      <td>247.84</td>\n",
       "      <td>206.72</td>\n",
       "      <td>60.32</td>\n",
       "      <td>NaN</td>\n",
       "      <td>64.64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4436</th>\n",
       "      <td>2018-08-21</td>\n",
       "      <td>215.04</td>\n",
       "      <td>105.98</td>\n",
       "      <td>172.620</td>\n",
       "      <td>1883.42</td>\n",
       "      <td>135.35</td>\n",
       "      <td>115.32</td>\n",
       "      <td>78.83</td>\n",
       "      <td>1217.41</td>\n",
       "      <td>1201.62</td>\n",
       "      <td>...</td>\n",
       "      <td>96.08</td>\n",
       "      <td>95.17</td>\n",
       "      <td>73.32</td>\n",
       "      <td>91.20</td>\n",
       "      <td>156.98</td>\n",
       "      <td>253.32</td>\n",
       "      <td>210.51</td>\n",
       "      <td>60.08</td>\n",
       "      <td>NaN</td>\n",
       "      <td>64.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4437</th>\n",
       "      <td>2018-08-22</td>\n",
       "      <td>215.05</td>\n",
       "      <td>107.06</td>\n",
       "      <td>173.640</td>\n",
       "      <td>1904.90</td>\n",
       "      <td>134.61</td>\n",
       "      <td>114.97</td>\n",
       "      <td>79.96</td>\n",
       "      <td>1221.75</td>\n",
       "      <td>1207.33</td>\n",
       "      <td>...</td>\n",
       "      <td>95.67</td>\n",
       "      <td>95.58</td>\n",
       "      <td>73.82</td>\n",
       "      <td>90.64</td>\n",
       "      <td>156.13</td>\n",
       "      <td>262.82</td>\n",
       "      <td>207.54</td>\n",
       "      <td>60.57</td>\n",
       "      <td>NaN</td>\n",
       "      <td>65.67</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4438</th>\n",
       "      <td>2018-08-23</td>\n",
       "      <td>215.49</td>\n",
       "      <td>107.56</td>\n",
       "      <td>172.900</td>\n",
       "      <td>1902.90</td>\n",
       "      <td>135.11</td>\n",
       "      <td>114.73</td>\n",
       "      <td>79.08</td>\n",
       "      <td>1221.16</td>\n",
       "      <td>1205.38</td>\n",
       "      <td>...</td>\n",
       "      <td>95.18</td>\n",
       "      <td>95.61</td>\n",
       "      <td>73.34</td>\n",
       "      <td>89.84</td>\n",
       "      <td>157.91</td>\n",
       "      <td>266.84</td>\n",
       "      <td>205.59</td>\n",
       "      <td>60.09</td>\n",
       "      <td>NaN</td>\n",
       "      <td>65.83</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4439</th>\n",
       "      <td>2018-08-24</td>\n",
       "      <td>216.16</td>\n",
       "      <td>108.40</td>\n",
       "      <td>174.645</td>\n",
       "      <td>1905.39</td>\n",
       "      <td>135.95</td>\n",
       "      <td>114.68</td>\n",
       "      <td>79.62</td>\n",
       "      <td>1236.75</td>\n",
       "      <td>1220.65</td>\n",
       "      <td>...</td>\n",
       "      <td>94.95</td>\n",
       "      <td>95.55</td>\n",
       "      <td>73.59</td>\n",
       "      <td>89.93</td>\n",
       "      <td>157.51</td>\n",
       "      <td>272.22</td>\n",
       "      <td>209.00</td>\n",
       "      <td>59.97</td>\n",
       "      <td>NaN</td>\n",
       "      <td>66.10</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 50 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            date    AAPL    MSFT       FB     AMZN     JNJ     JPM    XOM  \\\n",
       "4435  2018-08-20  215.46  106.87  172.500  1876.71  136.88  114.62  78.81   \n",
       "4436  2018-08-21  215.04  105.98  172.620  1883.42  135.35  115.32  78.83   \n",
       "4437  2018-08-22  215.05  107.06  173.640  1904.90  134.61  114.97  79.96   \n",
       "4438  2018-08-23  215.49  107.56  172.900  1902.90  135.11  114.73  79.08   \n",
       "4439  2018-08-24  216.16  108.40  174.645  1905.39  135.95  114.68  79.62   \n",
       "\n",
       "        GOOGL     GOOG  ...      WMT    MDT   GILD   CELG     HON    NVDA  \\\n",
       "4435  1221.95  1207.77  ...    96.00  90.03  72.36  90.88  155.47  247.84   \n",
       "4436  1217.41  1201.62  ...    96.08  95.17  73.32  91.20  156.98  253.32   \n",
       "4437  1221.75  1207.33  ...    95.67  95.58  73.82  90.64  156.13  262.82   \n",
       "4438  1221.16  1205.38  ...    95.18  95.61  73.34  89.84  157.91  266.84   \n",
       "4439  1236.75  1220.65  ...    94.95  95.55  73.59  89.93  157.51  272.22   \n",
       "\n",
       "        AVGO    BMY  PCLN    ABT  \n",
       "4435  206.72  60.32   NaN  64.64  \n",
       "4436  210.51  60.08   NaN  64.75  \n",
       "4437  207.54  60.57   NaN  65.67  \n",
       "4438  205.59  60.09   NaN  65.83  \n",
       "4439  209.00  59.97   NaN  66.10  \n",
       "\n",
       "[5 rows x 50 columns]"
      ]
     },
     "execution_count": 106,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(4440, 50)"
      ]
     },
     "execution_count": 107,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>AAPL</th>\n",
       "      <th>MSFT</th>\n",
       "      <th>FB</th>\n",
       "      <th>AMZN</th>\n",
       "      <th>JNJ</th>\n",
       "      <th>JPM</th>\n",
       "      <th>XOM</th>\n",
       "      <th>GOOGL</th>\n",
       "      <th>GOOG</th>\n",
       "      <th>BAC</th>\n",
       "      <th>...</th>\n",
       "      <th>WMT</th>\n",
       "      <th>MDT</th>\n",
       "      <th>GILD</th>\n",
       "      <th>CELG</th>\n",
       "      <th>HON</th>\n",
       "      <th>NVDA</th>\n",
       "      <th>AVGO</th>\n",
       "      <th>BMY</th>\n",
       "      <th>PCLN</th>\n",
       "      <th>ABT</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>4440.000000</td>\n",
       "      <td>4440.000000</td>\n",
       "      <td>1578.000000</td>\n",
       "      <td>4440.000000</td>\n",
       "      <td>4440.000000</td>\n",
       "      <td>4440.000000</td>\n",
       "      <td>4440.000000</td>\n",
       "      <td>3530.000000</td>\n",
       "      <td>1113.000000</td>\n",
       "      <td>4440.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>4440.000000</td>\n",
       "      <td>4440.000000</td>\n",
       "      <td>4440.000000</td>\n",
       "      <td>4440.000000</td>\n",
       "      <td>4440.000000</td>\n",
       "      <td>4440.000000</td>\n",
       "      <td>2280.000000</td>\n",
       "      <td>4440.000000</td>\n",
       "      <td>4314.000000</td>\n",
       "      <td>4440.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>48.310180</td>\n",
       "      <td>30.616045</td>\n",
       "      <td>95.801032</td>\n",
       "      <td>270.666637</td>\n",
       "      <td>62.314535</td>\n",
       "      <td>41.262014</td>\n",
       "      <td>56.386740</td>\n",
       "      <td>430.834224</td>\n",
       "      <td>774.047528</td>\n",
       "      <td>21.269334</td>\n",
       "      <td>...</td>\n",
       "      <td>50.629555</td>\n",
       "      <td>46.361626</td>\n",
       "      <td>34.046486</td>\n",
       "      <td>44.703558</td>\n",
       "      <td>55.500218</td>\n",
       "      <td>32.006529</td>\n",
       "      <td>93.189420</td>\n",
       "      <td>29.663031</td>\n",
       "      <td>499.818315</td>\n",
       "      <td>25.332072</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>51.382564</td>\n",
       "      <td>19.383773</td>\n",
       "      <td>52.610191</td>\n",
       "      <td>367.991115</td>\n",
       "      <td>29.993223</td>\n",
       "      <td>22.461346</td>\n",
       "      <td>20.189544</td>\n",
       "      <td>282.774046</td>\n",
       "      <td>203.960334</td>\n",
       "      <td>9.718843</td>\n",
       "      <td>...</td>\n",
       "      <td>15.997773</td>\n",
       "      <td>17.358570</td>\n",
       "      <td>31.165455</td>\n",
       "      <td>40.640718</td>\n",
       "      <td>37.514779</td>\n",
       "      <td>53.605196</td>\n",
       "      <td>80.098052</td>\n",
       "      <td>17.387019</td>\n",
       "      <td>583.174223</td>\n",
       "      <td>13.087757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.830505</td>\n",
       "      <td>11.960377</td>\n",
       "      <td>17.729000</td>\n",
       "      <td>5.970000</td>\n",
       "      <td>26.550261</td>\n",
       "      <td>9.988996</td>\n",
       "      <td>19.688871</td>\n",
       "      <td>50.159839</td>\n",
       "      <td>492.550000</td>\n",
       "      <td>2.911450</td>\n",
       "      <td>...</td>\n",
       "      <td>31.534436</td>\n",
       "      <td>19.761432</td>\n",
       "      <td>1.561837</td>\n",
       "      <td>1.437500</td>\n",
       "      <td>13.154448</td>\n",
       "      <td>2.275114</td>\n",
       "      <td>12.724734</td>\n",
       "      <td>10.881438</td>\n",
       "      <td>6.600000</td>\n",
       "      <td>8.718111</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>5.283072</td>\n",
       "      <td>19.060846</td>\n",
       "      <td>53.535000</td>\n",
       "      <td>39.922500</td>\n",
       "      <td>41.527536</td>\n",
       "      <td>26.696893</td>\n",
       "      <td>40.870123</td>\n",
       "      <td>233.117110</td>\n",
       "      <td>575.280000</td>\n",
       "      <td>13.800224</td>\n",
       "      <td>...</td>\n",
       "      <td>38.345388</td>\n",
       "      <td>34.634424</td>\n",
       "      <td>9.527034</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>26.593259</td>\n",
       "      <td>9.858312</td>\n",
       "      <td>29.527170</td>\n",
       "      <td>15.139746</td>\n",
       "      <td>26.717500</td>\n",
       "      <td>15.235212</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>24.624587</td>\n",
       "      <td>22.308219</td>\n",
       "      <td>87.257500</td>\n",
       "      <td>112.195000</td>\n",
       "      <td>48.382821</td>\n",
       "      <td>33.847769</td>\n",
       "      <td>59.110905</td>\n",
       "      <td>304.560148</td>\n",
       "      <td>747.770000</td>\n",
       "      <td>19.197016</td>\n",
       "      <td>...</td>\n",
       "      <td>43.234824</td>\n",
       "      <td>39.209899</td>\n",
       "      <td>20.306773</td>\n",
       "      <td>29.350000</td>\n",
       "      <td>40.509620</td>\n",
       "      <td>14.063283</td>\n",
       "      <td>54.189991</td>\n",
       "      <td>20.990110</td>\n",
       "      <td>137.815000</td>\n",
       "      <td>20.183048</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>81.382612</td>\n",
       "      <td>35.518671</td>\n",
       "      <td>132.152500</td>\n",
       "      <td>321.167500</td>\n",
       "      <td>85.524969</td>\n",
       "      <td>50.898520</td>\n",
       "      <td>74.623713</td>\n",
       "      <td>581.085475</td>\n",
       "      <td>932.220000</td>\n",
       "      <td>29.489776</td>\n",
       "      <td>...</td>\n",
       "      <td>65.135675</td>\n",
       "      <td>54.229711</td>\n",
       "      <td>65.731400</td>\n",
       "      <td>79.045000</td>\n",
       "      <td>84.146170</td>\n",
       "      <td>21.161331</td>\n",
       "      <td>147.934228</td>\n",
       "      <td>45.300801</td>\n",
       "      <td>1040.520000</td>\n",
       "      <td>35.251927</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>217.580000</td>\n",
       "      <td>110.399313</td>\n",
       "      <td>217.500000</td>\n",
       "      <td>1919.650000</td>\n",
       "      <td>146.121100</td>\n",
       "      <td>117.790000</td>\n",
       "      <td>89.752731</td>\n",
       "      <td>1285.500000</td>\n",
       "      <td>1268.330000</td>\n",
       "      <td>42.241976</td>\n",
       "      <td>...</td>\n",
       "      <td>107.596715</td>\n",
       "      <td>95.610000</td>\n",
       "      <td>113.396782</td>\n",
       "      <td>146.520000</td>\n",
       "      <td>162.578866</td>\n",
       "      <td>272.220000</td>\n",
       "      <td>278.881020</td>\n",
       "      <td>72.750648</td>\n",
       "      <td>2049.000000</td>\n",
       "      <td>66.100000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 49 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "              AAPL         MSFT           FB         AMZN          JNJ  \\\n",
       "count  4440.000000  4440.000000  1578.000000  4440.000000  4440.000000   \n",
       "mean     48.310180    30.616045    95.801032   270.666637    62.314535   \n",
       "std      51.382564    19.383773    52.610191   367.991115    29.993223   \n",
       "min       0.830505    11.960377    17.729000     5.970000    26.550261   \n",
       "25%       5.283072    19.060846    53.535000    39.922500    41.527536   \n",
       "50%      24.624587    22.308219    87.257500   112.195000    48.382821   \n",
       "75%      81.382612    35.518671   132.152500   321.167500    85.524969   \n",
       "max     217.580000   110.399313   217.500000  1919.650000   146.121100   \n",
       "\n",
       "               JPM          XOM        GOOGL         GOOG          BAC  \\\n",
       "count  4440.000000  4440.000000  3530.000000  1113.000000  4440.000000   \n",
       "mean     41.262014    56.386740   430.834224   774.047528    21.269334   \n",
       "std      22.461346    20.189544   282.774046   203.960334     9.718843   \n",
       "min       9.988996    19.688871    50.159839   492.550000     2.911450   \n",
       "25%      26.696893    40.870123   233.117110   575.280000    13.800224   \n",
       "50%      33.847769    59.110905   304.560148   747.770000    19.197016   \n",
       "75%      50.898520    74.623713   581.085475   932.220000    29.489776   \n",
       "max     117.790000    89.752731  1285.500000  1268.330000    42.241976   \n",
       "\n",
       "          ...               WMT          MDT         GILD         CELG  \\\n",
       "count     ...       4440.000000  4440.000000  4440.000000  4440.000000   \n",
       "mean      ...         50.629555    46.361626    34.046486    44.703558   \n",
       "std       ...         15.997773    17.358570    31.165455    40.640718   \n",
       "min       ...         31.534436    19.761432     1.561837     1.437500   \n",
       "25%       ...         38.345388    34.634424     9.527034    10.000000   \n",
       "50%       ...         43.234824    39.209899    20.306773    29.350000   \n",
       "75%       ...         65.135675    54.229711    65.731400    79.045000   \n",
       "max       ...        107.596715    95.610000   113.396782   146.520000   \n",
       "\n",
       "               HON         NVDA         AVGO          BMY         PCLN  \\\n",
       "count  4440.000000  4440.000000  2280.000000  4440.000000  4314.000000   \n",
       "mean     55.500218    32.006529    93.189420    29.663031   499.818315   \n",
       "std      37.514779    53.605196    80.098052    17.387019   583.174223   \n",
       "min      13.154448     2.275114    12.724734    10.881438     6.600000   \n",
       "25%      26.593259     9.858312    29.527170    15.139746    26.717500   \n",
       "50%      40.509620    14.063283    54.189991    20.990110   137.815000   \n",
       "75%      84.146170    21.161331   147.934228    45.300801  1040.520000   \n",
       "max     162.578866   272.220000   278.881020    72.750648  2049.000000   \n",
       "\n",
       "               ABT  \n",
       "count  4440.000000  \n",
       "mean     25.332072  \n",
       "std      13.087757  \n",
       "min       8.718111  \n",
       "25%      15.235212  \n",
       "50%      20.183048  \n",
       "75%      35.251927  \n",
       "max      66.100000  \n",
       "\n",
       "[8 rows x 49 columns]"
      ]
     },
     "execution_count": 108,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 4440 entries, 0 to 4439\n",
      "Data columns (total 10 columns):\n",
      "date     4440 non-null object\n",
      "AAPL     4440 non-null float64\n",
      "MSFT     4440 non-null float64\n",
      "FB       1578 non-null float64\n",
      "AMZN     4440 non-null float64\n",
      "JNJ      4440 non-null float64\n",
      "JPM      4440 non-null float64\n",
      "XOM      4440 non-null float64\n",
      "GOOGL    3530 non-null float64\n",
      "GOOG     1113 non-null float64\n",
      "dtypes: float64(9), object(1)\n",
      "memory usage: 347.0+ KB\n"
     ]
    }
   ],
   "source": [
    "# Show only the first 10 columns in datafram\n",
    "df.iloc[:, 0:10].info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "date     0.000000\n",
       "AAPL     0.000000\n",
       "MSFT     0.000000\n",
       "FB       0.644595\n",
       "AMZN     0.000000\n",
       "JNJ      0.000000\n",
       "JPM      0.000000\n",
       "XOM      0.000000\n",
       "GOOGL    0.204955\n",
       "GOOG     0.749324\n",
       "dtype: float64"
      ]
     },
     "execution_count": 110,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# proportion missing values\n",
    "# show only the first 10 columns in dataframe\n",
    "df.iloc[:, 0:10].isnull().sum()/len(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>AAPL</th>\n",
       "      <th>MSFT</th>\n",
       "      <th>FB</th>\n",
       "      <th>AMZN</th>\n",
       "      <th>JNJ</th>\n",
       "      <th>JPM</th>\n",
       "      <th>XOM</th>\n",
       "      <th>GOOGL</th>\n",
       "      <th>GOOG</th>\n",
       "      <th>BAC</th>\n",
       "      <th>...</th>\n",
       "      <th>WMT</th>\n",
       "      <th>MDT</th>\n",
       "      <th>GILD</th>\n",
       "      <th>CELG</th>\n",
       "      <th>HON</th>\n",
       "      <th>NVDA</th>\n",
       "      <th>AVGO</th>\n",
       "      <th>BMY</th>\n",
       "      <th>PCLN</th>\n",
       "      <th>ABT</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>date</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2017-01-03</th>\n",
       "      <td>113.008218</td>\n",
       "      <td>60.439210</td>\n",
       "      <td>116.86</td>\n",
       "      <td>753.67</td>\n",
       "      <td>111.386107</td>\n",
       "      <td>83.990394</td>\n",
       "      <td>84.924851</td>\n",
       "      <td>808.01</td>\n",
       "      <td>786.14</td>\n",
       "      <td>22.007337</td>\n",
       "      <td>...</td>\n",
       "      <td>65.744953</td>\n",
       "      <td>68.272158</td>\n",
       "      <td>70.967312</td>\n",
       "      <td>118.75</td>\n",
       "      <td>112.184666</td>\n",
       "      <td>101.493542</td>\n",
       "      <td>172.499458</td>\n",
       "      <td>56.728182</td>\n",
       "      <td>1477.57</td>\n",
       "      <td>37.642052</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-01-04</th>\n",
       "      <td>112.881734</td>\n",
       "      <td>60.168788</td>\n",
       "      <td>118.69</td>\n",
       "      <td>757.18</td>\n",
       "      <td>111.203412</td>\n",
       "      <td>84.144452</td>\n",
       "      <td>83.990481</td>\n",
       "      <td>807.77</td>\n",
       "      <td>786.90</td>\n",
       "      <td>22.417594</td>\n",
       "      <td>...</td>\n",
       "      <td>66.127970</td>\n",
       "      <td>68.910306</td>\n",
       "      <td>73.091738</td>\n",
       "      <td>118.55</td>\n",
       "      <td>112.493662</td>\n",
       "      <td>103.861492</td>\n",
       "      <td>171.271050</td>\n",
       "      <td>56.804520</td>\n",
       "      <td>1487.29</td>\n",
       "      <td>37.940875</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-01-05</th>\n",
       "      <td>113.455775</td>\n",
       "      <td>60.168788</td>\n",
       "      <td>120.67</td>\n",
       "      <td>780.45</td>\n",
       "      <td>112.366890</td>\n",
       "      <td>83.369909</td>\n",
       "      <td>82.738426</td>\n",
       "      <td>813.02</td>\n",
       "      <td>794.02</td>\n",
       "      <td>22.153857</td>\n",
       "      <td>...</td>\n",
       "      <td>66.271602</td>\n",
       "      <td>68.484874</td>\n",
       "      <td>72.737667</td>\n",
       "      <td>118.76</td>\n",
       "      <td>112.744721</td>\n",
       "      <td>101.224909</td>\n",
       "      <td>168.572421</td>\n",
       "      <td>57.755424</td>\n",
       "      <td>1503.70</td>\n",
       "      <td>38.268616</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3 rows × 49 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                  AAPL       MSFT      FB    AMZN         JNJ        JPM  \\\n",
       "date                                                                       \n",
       "2017-01-03  113.008218  60.439210  116.86  753.67  111.386107  83.990394   \n",
       "2017-01-04  112.881734  60.168788  118.69  757.18  111.203412  84.144452   \n",
       "2017-01-05  113.455775  60.168788  120.67  780.45  112.366890  83.369909   \n",
       "\n",
       "                  XOM   GOOGL    GOOG        BAC    ...            WMT  \\\n",
       "date                                                ...                  \n",
       "2017-01-03  84.924851  808.01  786.14  22.007337    ...      65.744953   \n",
       "2017-01-04  83.990481  807.77  786.90  22.417594    ...      66.127970   \n",
       "2017-01-05  82.738426  813.02  794.02  22.153857    ...      66.271602   \n",
       "\n",
       "                  MDT       GILD    CELG         HON        NVDA        AVGO  \\\n",
       "date                                                                           \n",
       "2017-01-03  68.272158  70.967312  118.75  112.184666  101.493542  172.499458   \n",
       "2017-01-04  68.910306  73.091738  118.55  112.493662  103.861492  171.271050   \n",
       "2017-01-05  68.484874  72.737667  118.76  112.744721  101.224909  168.572421   \n",
       "\n",
       "                  BMY     PCLN        ABT  \n",
       "date                                       \n",
       "2017-01-03  56.728182  1477.57  37.642052  \n",
       "2017-01-04  56.804520  1487.29  37.940875  \n",
       "2017-01-05  57.755424  1503.70  38.268616  \n",
       "\n",
       "[3 rows x 49 columns]"
      ]
     },
     "execution_count": 111,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# needs to set date as index in order to slice by date, \n",
    "# now the dataframe holds time series for each stock.\n",
    "df.date = pd.to_datetime(df.date) # necessary in order to show dates in plots\n",
    "df.set_index('date', inplace=True)\n",
    "df['2016-12-31':'2017-01-05'] # notice that slicing uses apostroph"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Selection"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>AAPL</th>\n",
       "      <th>MSFT</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>date</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2001-01-02</th>\n",
       "      <td>0.941914</td>\n",
       "      <td>14.254892</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2001-01-03</th>\n",
       "      <td>1.036232</td>\n",
       "      <td>15.753331</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2001-01-04</th>\n",
       "      <td>1.079909</td>\n",
       "      <td>15.917634</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2001-01-05</th>\n",
       "      <td>1.036232</td>\n",
       "      <td>16.144371</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2001-01-08</th>\n",
       "      <td>1.048259</td>\n",
       "      <td>16.081936</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                AAPL       MSFT\n",
       "date                           \n",
       "2001-01-02  0.941914  14.254892\n",
       "2001-01-03  1.036232  15.753331\n",
       "2001-01-04  1.079909  15.917634\n",
       "2001-01-05  1.036232  16.144371\n",
       "2001-01-08  1.048259  16.081936"
      ]
     },
     "execution_count": 112,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.loc[:, ['AAPL', 'MSFT']].head() "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>AAPL</th>\n",
       "      <th>MSFT</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>date</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2017-01-03</th>\n",
       "      <td>113.008218</td>\n",
       "      <td>60.439210</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-01-04</th>\n",
       "      <td>112.881734</td>\n",
       "      <td>60.168788</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-01-05</th>\n",
       "      <td>113.455775</td>\n",
       "      <td>60.168788</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  AAPL       MSFT\n",
       "date                             \n",
       "2017-01-03  113.008218  60.439210\n",
       "2017-01-04  112.881734  60.168788\n",
       "2017-01-05  113.455775  60.168788"
      ]
     },
     "execution_count": 113,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.loc['2016-12-31':'2017-01-05', ['AAPL', 'MSFT']].head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Plotting with pandas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x18f50f6c668>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df['AAPL'].plot(figsize=(10, 6))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x18f511e8e48>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot the first 5 columns in the dataframe\n",
    "df.iloc[:, :5].plot(subplots=False, figsize=(10, 10))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x18f51366160>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df.AAPL.pct_change().plot(figsize=(10, 6))\n",
    "plt.title('Apple stock returns by date')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x18f515bac18>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df.AAPL.pct_change().hist(bins=100, figsize=(10, 6))\n",
    "plt.title('Apple stock returns')\n",
    "plt.xlabel('Percentage change')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Some examples on how to use apply on dataframe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.series.Series'>\n"
     ]
    }
   ],
   "source": [
    "df_mean = df.apply(np.mean)\n",
    "print(type(df_mean))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "AAPL      48.310180\n",
       "MSFT      30.616045\n",
       "FB        95.801032\n",
       "AMZN     270.666637\n",
       "JNJ       62.314535\n",
       "JPM       41.262014\n",
       "XOM       56.386740\n",
       "GOOGL    430.834224\n",
       "GOOG     774.047528\n",
       "BAC       21.269334\n",
       "dtype: float64"
      ]
     },
     "execution_count": 120,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_mean.iloc[0:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "AAPL      51.376778\n",
       "MSFT      19.381590\n",
       "FB        52.593519\n",
       "AMZN     367.949673\n",
       "JNJ       29.989845\n",
       "JPM       22.458816\n",
       "XOM       20.187271\n",
       "GOOGL    282.733990\n",
       "GOOG     203.868687\n",
       "BAC        9.717748\n",
       "dtype: float64"
      ]
     },
     "execution_count": 121,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_std = df.apply(np.std)\n",
    "df_std.iloc[:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Python Import"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "from pylab import plt\n",
    "plt.style.use('seaborn')\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Random Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {},
   "outputs": [],
   "source": [
    "rn = np.random.standard_normal(size=50)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.1985803987697192"
      ]
     },
     "execution_count": 124,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rn.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.89882691547484284"
      ]
     },
     "execution_count": 125,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rn.std()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x18f51382518>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(rn, 'ro')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 127,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x18f515331d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(rn.cumsum() + 100)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## OLS Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "metadata": {},
   "outputs": [],
   "source": [
    "x = np.linspace(-5, 5, 25)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 129,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = 1.5 * x ** 2 - 2 * x + 4 + np.random.standard_normal(len(x))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x18f51486780>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(x, y, 'r.')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## numpy.polyfit\n",
    "\n",
    "numpy.polyfit(x, y, deg, rcond=None, full=False, w=None, cov=False)\n",
    "* Least squares polynomial fit.\n",
    "\n",
    "    Fit a polynomial p(x) = p[0] * x**deg + ... + p[deg] of degree deg to points (x, y). Returns a vector of coefficients p that minimises the squared error."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "metadata": {},
   "outputs": [],
   "source": [
    "reg = np.polyfit(x, y, deg=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 1.52070571, -2.03942006,  3.75409397])"
      ]
     },
     "execution_count": 132,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "reg"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "17.482687160473571"
      ]
     },
     "execution_count": 133,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y.mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## numpy.polyval\n",
    "\n",
    "numpy.polyval(p, x)[source]\n",
    " * Evaluate a polynomial at specific values.\n",
    "\n",
    "    If p is of length N, this function returns the value:\n",
    "\n",
    "        p[0]*x**(N-1) + p[1]*x**(N-2) + ... + p[N-2]*x + p[N-1]\n",
    "\n",
    "    If x is a sequence, then p(x) is returned for each element of x. If x is another polynomial then the composite polynomial p(x(t)) is returned."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 134,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 1.52070571 -2.03942006  3.75409397]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x18f515e8320>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(reg)\n",
    "yr = np.polyval(reg, x)\n",
    "plt.plot(x, y, 'r.')\n",
    "plt.plot(x, yr, 'b')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Regression with scikit-learn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "from sklearn.ensemble import GradientBoostingRegressor\n",
    "from sklearn.metrics import mean_squared_error"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Linear model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.198682703959\n",
      "mean-squared-error: 151.439333118\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x18f53337b70>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "reg_ln = LinearRegression()\n",
    "reg_ln.fit(x.reshape(-1, 1), y)\n",
    "print(reg_ln.score(x.reshape(-1, 1), y))\n",
    "yr = reg_ln.predict(x.reshape(-1, 1))#np.polyval(reg_sklearn.coef_, x)\n",
    "print('mean-squared-error:', mean_squared_error(yr, y))\n",
    "plt.plot(x, y, 'r.')\n",
    "plt.plot(x, yr, 'b')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Gradient Boosting regressor\n",
    "Gradient Tree Boosting or Gradient Boosted Regression Trees (GBRT) is a generalization of boosting to arbitrary differentiable loss functions. GBRT is an accurate and effective off-the-shelf procedure that can be used for both regression and classification problems. Gradient Tree Boosting models are used in a variety of areas including Web search ranking and ecology.\n",
    "\n",
    "The advantages of GBRT are:\n",
    "*  Natural handling of data of mixed type (= heterogeneous features)\n",
    "*  Predictive power\n",
    "*  Robustness to outliers in output space (via robust loss functions)\n",
    "\n",
    "The disadvantages of GBRT are:\n",
    "* Scalability, due to the sequential nature of boosting it can hardly be parallelized.\n",
    "\n",
    "The module sklearn.ensemble provides methods for both classification and regression via gradient boosted regression trees."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 137,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "score: 0.999997084953\n",
      "mean-squared-error: 0.000550908879883\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x18f53395320>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "reg_gb = GradientBoostingRegressor()\n",
    "reg_gb.fit(x.reshape(-1, 1), y) # we have to reshape to a matrix of dim (25, 1)\n",
    "#print(reg_sklearn.coef_)\n",
    "print('score:', reg_gb.score(x.reshape(-1, 1), y))\n",
    "yr = reg_gb.predict(x.reshape(-1, 1))\n",
    "print('mean-squared-error:', mean_squared_error(yr, y))\n",
    "plt.plot(x, y, 'r.')\n",
    "plt.plot(x, yr, 'b')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Random Forest Regressor\n",
    "A random forest is a meta estimator that fits a number of classifying decision trees on various sub-samples of the dataset and uses averaging to improve the predictive accuracy and control over-fitting. The sub-sample size is always the same as the original input sample size but the samples are drawn with replacement if bootstrap=True (default).\n",
    "\n",
    "In random forests (see RandomForestClassifier and RandomForestRegressor classes), each tree in the ensemble is built from a sample drawn with replacement (i.e., a bootstrap sample) from the training set. In addition, when splitting a node during the construction of the tree, the split that is chosen is no longer the best split among all features. Instead, the split that is picked is the best split among a random subset of the features. As a result of this randomness, the bias of the forest usually slightly increases (with respect to the bias of a single non-random tree) but, due to averaging, its variance also decreases, usually more than compensating for the increase in bias, hence yielding an overall better model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 138,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "score: 0.999997084953\n",
      "mean-squared-error: 2.64942398723\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x18f53384da0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "reg_rf = RandomForestRegressor()\n",
    "reg_rf.fit(x.reshape(-1, 1), y) # we have to reshape to a matrix of dim (25, 1)\n",
    "#print(reg_sklearn.coef_)\n",
    "print('score:', reg_gb.score(x.reshape(-1, 1), y))\n",
    "yr = reg_rf.predict(x.reshape(-1, 1))\n",
    "print('mean-squared-error:', mean_squared_error(yr, y))\n",
    "plt.plot(x, y, 'r.')\n",
    "plt.plot(x, yr, 'b')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Regression with financial data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 139,
   "metadata": {},
   "outputs": [],
   "source": [
    "import quandl\n",
    "quandl.ApiConfig.api_key = 'LcV7TTWCgXqKHco3f52D'\n",
    "from sklearn import svm\n",
    "from sklearn.model_selection import cross_validate\n",
    "from sklearn.linear_model import LinearRegression\n",
    "import math\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.metrics import mean_squared_error, r2_score"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Download data from quandl"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 140,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = quandl.get(\"WIKI/GOOGL\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 141,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Open</th>\n",
       "      <th>High</th>\n",
       "      <th>Low</th>\n",
       "      <th>Close</th>\n",
       "      <th>Volume</th>\n",
       "      <th>Ex-Dividend</th>\n",
       "      <th>Split Ratio</th>\n",
       "      <th>Adj. Open</th>\n",
       "      <th>Adj. High</th>\n",
       "      <th>Adj. Low</th>\n",
       "      <th>Adj. Close</th>\n",
       "      <th>Adj. Volume</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Date</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2004-08-19</th>\n",
       "      <td>100.01</td>\n",
       "      <td>104.06</td>\n",
       "      <td>95.96</td>\n",
       "      <td>100.335</td>\n",
       "      <td>44659000.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>50.159839</td>\n",
       "      <td>52.191109</td>\n",
       "      <td>48.128568</td>\n",
       "      <td>50.322842</td>\n",
       "      <td>44659000.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2004-08-20</th>\n",
       "      <td>101.01</td>\n",
       "      <td>109.08</td>\n",
       "      <td>100.50</td>\n",
       "      <td>108.310</td>\n",
       "      <td>22834300.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>50.661387</td>\n",
       "      <td>54.708881</td>\n",
       "      <td>50.405597</td>\n",
       "      <td>54.322689</td>\n",
       "      <td>22834300.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2004-08-23</th>\n",
       "      <td>110.76</td>\n",
       "      <td>113.48</td>\n",
       "      <td>109.05</td>\n",
       "      <td>109.400</td>\n",
       "      <td>18256100.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>55.551482</td>\n",
       "      <td>56.915693</td>\n",
       "      <td>54.693835</td>\n",
       "      <td>54.869377</td>\n",
       "      <td>18256100.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2004-08-24</th>\n",
       "      <td>111.24</td>\n",
       "      <td>111.60</td>\n",
       "      <td>103.57</td>\n",
       "      <td>104.870</td>\n",
       "      <td>15247300.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>55.792225</td>\n",
       "      <td>55.972783</td>\n",
       "      <td>51.945350</td>\n",
       "      <td>52.597363</td>\n",
       "      <td>15247300.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2004-08-25</th>\n",
       "      <td>104.76</td>\n",
       "      <td>108.00</td>\n",
       "      <td>103.88</td>\n",
       "      <td>106.000</td>\n",
       "      <td>9188600.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>52.542193</td>\n",
       "      <td>54.167209</td>\n",
       "      <td>52.100830</td>\n",
       "      <td>53.164113</td>\n",
       "      <td>9188600.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              Open    High     Low    Close      Volume  Ex-Dividend  \\\n",
       "Date                                                                   \n",
       "2004-08-19  100.01  104.06   95.96  100.335  44659000.0          0.0   \n",
       "2004-08-20  101.01  109.08  100.50  108.310  22834300.0          0.0   \n",
       "2004-08-23  110.76  113.48  109.05  109.400  18256100.0          0.0   \n",
       "2004-08-24  111.24  111.60  103.57  104.870  15247300.0          0.0   \n",
       "2004-08-25  104.76  108.00  103.88  106.000   9188600.0          0.0   \n",
       "\n",
       "            Split Ratio  Adj. Open  Adj. High   Adj. Low  Adj. Close  \\\n",
       "Date                                                                   \n",
       "2004-08-19          1.0  50.159839  52.191109  48.128568   50.322842   \n",
       "2004-08-20          1.0  50.661387  54.708881  50.405597   54.322689   \n",
       "2004-08-23          1.0  55.551482  56.915693  54.693835   54.869377   \n",
       "2004-08-24          1.0  55.792225  55.972783  51.945350   52.597363   \n",
       "2004-08-25          1.0  52.542193  54.167209  52.100830   53.164113   \n",
       "\n",
       "            Adj. Volume  \n",
       "Date                     \n",
       "2004-08-19   44659000.0  \n",
       "2004-08-20   22834300.0  \n",
       "2004-08-23   18256100.0  \n",
       "2004-08-24   15247300.0  \n",
       "2004-08-25    9188600.0  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "display(df.head())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 142,
   "metadata": {},
   "outputs": [],
   "source": [
    "def hl_pct(high, low, close):\n",
    "    return 100.0 * (high - low)/close"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "metadata": {},
   "outputs": [],
   "source": [
    "def pct_change(cl, op):\n",
    "    return 100.0 * (cl - op)/op"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 144,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df[['Adj. Open', 'Adj. High', 'Adj. Low', 'Adj. Close', 'Adj. Volume']]\n",
    "df['HL_PCT'] = np.vectorize(hl_pct)(df['Adj. High'], df['Adj. Low'], df['Adj. Close'])\n",
    "df['PCT_change'] = np.vectorize(pct_change)(df['Adj. Close'], df['Adj. Open'])\n",
    "df = df[['Adj. Close', 'HL_PCT', 'PCT_change', 'Adj. Volume']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 145,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "df[forecast_col].head():\n",
      "Date\n",
      "2004-08-19    50.322842\n",
      "2004-08-20    54.322689\n",
      "2004-08-23    54.869377\n",
      "2004-08-24    52.597363\n",
      "2004-08-25    53.164113\n",
      "Name: Adj. Close, dtype: float64\n",
      "\n",
      "df.label.head()\n",
      "Date\n",
      "2004-08-19    69.078238\n",
      "2004-08-20    67.839414\n",
      "2004-08-23    68.912727\n",
      "2004-08-24    70.668146\n",
      "2004-08-25    71.219849\n",
      "Name: label, dtype: float64\n",
      "\n",
      "Confidence: 0.978027090006\n",
      "Mean-squared-error: 1417.74535774\n",
      "R2-score: 0.977422757687\n"
     ]
    }
   ],
   "source": [
    "\n",
    "forecast_col = 'Adj. Close'\n",
    "\n",
    "df.fillna(value=-99999, inplace=True)\n",
    "forecast_out = int(math.ceil(0.01 * len(df)))\n",
    "# print(forecast_out)\n",
    "df['label'] = df[forecast_col].shift(-forecast_out)\n",
    "\n",
    "print('df[forecast_col].head():')\n",
    "print(df[forecast_col].head())\n",
    "print('')\n",
    "print('df.label.head()')\n",
    "print(df.label.head())\n",
    "\n",
    "# Before transformation split in train and test\n",
    "X = df.drop('label', axis=1)\n",
    "#X = X[:-forecast_out]\n",
    "\n",
    "X_lately = X[-forecast_out:]\n",
    "X = X[:-forecast_out]\n",
    "df.dropna(inplace=True)\n",
    "y = df['label']\n",
    "\n",
    "scaler = StandardScaler()\n",
    "\n",
    "xtrain, xtest, ytrain, ytest = train_test_split(X, y, test_size=0.2)\n",
    "xtrain = scaler.fit_transform(xtrain)\n",
    "xtest = scaler.transform(xtest)\n",
    "\n",
    "lr = LinearRegression(n_jobs=-1)\n",
    "lr.fit(xtrain, ytrain)\n",
    "pred = lr.predict(xtest)\n",
    "# Independent data-set\n",
    "independent_pred = lr.predict(X_lately)\n",
    "\n",
    "confidence = lr.score(xtest, ytest)\n",
    "print('')\n",
    "print('Confidence:', confidence)\n",
    "print('Mean-squared-error:', mean_squared_error(pred, ytest))\n",
    "print('R2-score:', r2_score(pred, ytest))\n",
    "#print(independent_pred, confidence, forecast_out)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 146,
   "metadata": {},
   "outputs": [],
   "source": [
    "import datetime\n",
    "import matplotlib.pyplot as plt\n",
    "from matplotlib import style"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 147,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Date\n",
      "2004-08-19    50.322842\n",
      "2004-08-20    54.322689\n",
      "2004-08-23    54.869377\n",
      "2004-08-24    52.597363\n",
      "2004-08-25    53.164113\n",
      "Name: Adj. Close, dtype: float64\n",
      "[-8264531.88439293 -5554481.17800889 -6762342.38214948 -9884264.59171988\n",
      " -6130209.22489999 -3334483.25818717 -4372644.70346699 -3856018.29688938\n",
      " -4234086.94724098 -3487360.12149218 -4349825.20216558 -2890313.10432169\n",
      " -2538965.0979733  -3117412.21275857 -4509766.44172228 -5282215.67444618\n",
      " -6058699.51190678 -5463637.33859167 -2999948.57880496 -2396054.3452216\n",
      " -3235168.54320609 -3169892.07823741 -4442342.65708937 -4576310.91434313\n",
      " -4583202.32089058 -4361912.90484603 -3423931.89811459 -5785987.13460291\n",
      " -6759473.36745916 -5920889.84198406 -4276987.1667155  -7553254.29245491\n",
      " -5261706.63763537 -7219602.47975086 -6473266.66978915]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x18f5140eb00>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "style.use('ggplot')\n",
    "df['Forecast'] = np.nan\n",
    "last_date = df.iloc[-1].name\n",
    "last_unix = last_date.timestamp()\n",
    "one_day = 86400\n",
    "next_unix = last_unix + one_day\n",
    "\n",
    "for i in independent_pred:\n",
    "    next_date = datetime.datetime.fromtimestamp(next_unix)\n",
    "    next_unix += 86400\n",
    "    \n",
    "    df.loc[next_date] = [np.nan for _ in range(len(df.columns) - 1)] + [i]\n",
    "    df.loc['Forecast'] = i\n",
    "\n",
    "print(df['Adj. Close'].head())\n",
    "print(independent_pred)\n",
    "df['Adj. Close'].plot(figsize=(10, 6))\n",
    "df['Forecast'].plot()\n",
    "plt.legend(loc=4)\n",
    "plt.xlabel('Date')\n",
    "plt.ylabel('Price')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 169,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x18f50226dd8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn import preprocessing, cross_validation\n",
    "style.use('ggplot')\n",
    "\n",
    "df = quandl.get(\"WIKI/GOOGL\")\n",
    "df = df[['Adj. Open',  'Adj. High',  'Adj. Low',  'Adj. Close', 'Adj. Volume']]\n",
    "df['HL_PCT'] = (df['Adj. High'] - df['Adj. Low']) / df['Adj. Close'] * 100.0\n",
    "df['PCT_change'] = (df['Adj. Close'] - df['Adj. Open']) / df['Adj. Open'] * 100.0\n",
    "\n",
    "df = df[['Adj. Close', 'HL_PCT', 'PCT_change', 'Adj. Volume']]\n",
    "forecast_col = 'Adj. Close'\n",
    "df.fillna(value=-99999, inplace=True)\n",
    "forecast_out = int(math.ceil(0.01 * len(df)))\n",
    "df['label'] = df[forecast_col].shift(-forecast_out)\n",
    "\n",
    "X = np.array(df.drop(['label'], 1))\n",
    "#X = preprocessing.scale(X)\n",
    "X = scaler.fit_transform(X)\n",
    "X_lately = X[-forecast_out:]\n",
    "X = X[:-forecast_out]\n",
    "\n",
    "df.dropna(inplace=True)\n",
    "\n",
    "y = np.array(df['label'])\n",
    "# Ok it makes difference when you normalize data\n",
    "X_train, X_test, y_train, y_test = cross_validation.train_test_split(X, y, test_size=0.2)\n",
    "#X_train = scaler.fit_transform(X_train)\n",
    "#X_test = scaler.transform(X_test)\n",
    "\n",
    "clf = LinearRegression(n_jobs=-1)\n",
    "clf.fit(X_train, y_train)\n",
    "confidence = clf.score(X_test, y_test)\n",
    "\n",
    "forecast_set = clf.predict(X_lately)\n",
    "df['Forecast'] = np.nan\n",
    "\n",
    "last_date = df.iloc[-1].name\n",
    "last_unix = last_date.timestamp()\n",
    "one_day = 86400\n",
    "next_unix = last_unix + one_day\n",
    "\n",
    "for i in forecast_set:\n",
    "    next_date = datetime.datetime.fromtimestamp(next_unix)\n",
    "    next_unix += 86400\n",
    "    df.loc[next_date] = [np.nan for _ in range(len(df.columns)-1)]+[i]\n",
    "\n",
    "df['Adj. Close'].plot(figsize=(15, 10))\n",
    "df['Forecast'].plot()\n",
    "plt.legend(loc=4)\n",
    "plt.xlabel('Date')\n",
    "plt.ylabel('Price')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Simple Classification Problem"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 149,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXIAAAD3CAYAAAAALt/WAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvNQv5yAAAEbdJREFUeJzt3W9MlfX/x/EX53CQg2GWU0+HxGl20K2MJmOa1tTRnxs1c6U11o2WSsryjmvmBAvTs9lsbaapOVaGurWmcziWd2o6s+aGDqUZiNHEEktLSdSDHjjnd6Ov5ydpXnDkcM5bn49bnOs6f96fC3pydSGctGg0GhUAwCxXsgcAANweQg4AxhFyADCOkAOAcYQcAIwj5ABgXHoyXrS1tTWux/n9/rgfmypYQ/JZn19iDamgv+f3+/3/uY8zcgAwjpADgHGEHACMI+QAYBwhBwDjCDl6LLO6WkOLiqT0dA0tKlJmdXWyR0o5147RA7m5HKP/YP3rKJ75E/110aN/fnj8+HFt27ZNFRUV3bYfPHhQO3bskMvl0rRp01RUVNSnwyF1ZFZX6/7S0thtT0OD7i8t1TlJHTNmJG+wFMIxcmb9GMUzf3+s2fGMvLq6Whs3blQ4HO62vbOzU1988YXKysq0fPlyffvtt2pra+uToZB6steu7dX2uxHHyJn1YxTP/P2x5jSnv0d+4MABjRw5UuvWrVMwGIxtb2lp0datW1VWViZJ2rx5s/Ly8jRp0qQ+Gw4pJD1d6uq6+fZ/fZO/a3GMnFk/RvHM3w9rdry0MnHiRJ05c+aG7aFQSFlZWbHbXq9Xly9f7tGL8pud9tYwNBCQp6Hhhu3hhx/WWWPrSdTnoD+PEV9HyRHP/H215oT8ZqfX61VHR0fsdigU0sCBA+N9OqS49oULe7X9bsQxcmb9GMUzf3+sOe6Q5+Tk6PTp07p48aI6OzvV0NCgQCDQZ4MhtXTMmKFz69crPG6clJ6u8LhxOrd+vYkfUPWX649RlGN0U9a/juKZvz++LhyvkUvSmTNntGbNGgWDQe3fv18dHR0qKiqK/auVSCSiadOm6bnnnuvRi3JphTUkk/X5JdaQClLpj2b16J8fDhs2LPaDzilTpsS2FxQUqKCg4DbHAwDcDn4hCACMI+QAYBwhBwDjCDkAGEfIAcA4Qg4AxhFyADCOkAOAcYQcAIwj5ABgHCEHAOMIOQAYR8gBwDhCDgDGEXIAMI6QA4BxhBwAjCPkAGAcIQcA4wg5ABhHyAHAOEIOAMYRcgAwjpADgHGEHACMI+QAYBwhBwDjCDkAGEfIAcA4Qg4AxhFyADAu3ekOkUhElZWVamlpkcfj0fz58+Xz+WL7d+3ape+//14ul0szZ85UYWFhQgcGAHTneEZeW1urcDisYDCo4uJiVVVVxfZdunRJu3fvVjAYVFlZmTZv3pzIWQEAN+EY8sbGRuXn50uSAoGAmpubY/sGDBigoUOHqqOjQ1euXFFaWlriJgUA3JTjpZVQKKSsrKzYbZfLpa6uLrndbknSkCFDtGjRIkUiEb344os9elG/3x/nuLf32FTBGpLP+vwSa0gFqTK/Y8i9Xq9CoVDsdjQajUX88OHDamtr07p16yRJwWBQY8eO1ZgxY275nK2trXEN6/f7435sqmANyWd9fok1pIL+nv9W3zQcL63k5eWprq5OktTU1KTc3NzYvoEDByojI0Mej0cZGRkaOHCgLl261AcjAwB6yvGMvLCwUPX19SovL1c0GlVpaalqamrk8/lUUFCgH3/8UWVlZUpLS9PYsWM1fvz4/pgbAPA/jiF3uVwqKSnpti0nJyf28ezZszV79uy+nwwA0CP8QhAAGEfIAcA4Qg4AxhFyADCOkAOAcYQcAIwj5ABgHCEHAOMIOQAYR8gBwDhCDgDGEXIAMI6QA4BxhBwAjCPkAGAcIQcA4wg5ABhHyAHAOEIOAMYRcgAwjpADgHGEHACMI+QAYBwhBwDjCDkAGEfIAcA4Qg4AxhFyADCOkAOAcYQcAIwj5ABgHCEHAOPSne4QiURUWVmplpYWeTwezZ8/Xz6fL7a/rq5O27dvlySNGjVKc+bMUVpaWuImBgB043hGXltbq3A4rGAwqOLiYlVVVcX2hUIhbd26Ve+8846CwaCGDh2q9vb2hA4MAOjO8Yy8sbFR+fn5kqRAIKDm5ubYvmPHjmnEiBGqqqrSmTNnNH36dA0aNMjxRf1+f9wD385jUwVrSD7r80usIRWkyvyOIQ+FQsrKyorddrlc6urqktvtVnt7u44eParVq1crMzNT7777rgKBgOPiWltb4xrW7/fH/dhUwRqSz/r8EmtIBf09/6266nhpxev1KhQKxW5Ho1G53W5JUnZ2th566CENHjxYmZmZGjdunE6cOHH7EwMAeswx5Hl5eaqrq5MkNTU1KTc3N7Zv9OjR+vXXX3XhwgV1dXXp+PHjevDBBxM3LQDgBo6XVgoLC1VfX6/y8nJFo1GVlpaqpqZGPp9PBQUFKi4uVjAYlCRNmjSpW+gBAInnGHKXy6WSkpJu23JycmIfT548WZMnT+77yQAAPcIvBAGAcYQcAIwj5ABgHCEHAOMIOQAYR8gBwDhCDgDGEXIAMI6QA4BxhBwAjCPkAGAcIQcA4wg5ABhHyAHAOEIOAMYRcgAwjpADgHGEHACMI+QAYBwhBwDjCDkAGEfIAcA4Qg4AxhFyADCOkAOAcYQcAIwj5ABgHCEHAOMIOQAYR8gBwDhCDgDGpTvdIRKJqLKyUi0tLfJ4PJo/f758Pt8N91m1apUKCgr0zDPPJGxYAMCNHM/Ia2trFQ6HFQwGVVxcrKqqqhvu8+WXX+rixYsJGRAAcGuOIW9sbFR+fr4kKRAIqLm5udv+AwcOyOVyxe4DAOhfjpdWQqGQsrKyYrddLpe6urrkdrt18uRJ7d+/X4sWLdL27dt7/KJ+vz++aW/zsamCNSSf9fkl1pAKUmV+x5B7vV6FQqHY7Wg0KrfbLUnat2+fzp07p/fff19nz55Venq6hg0b5nh23traGtewfr8/7semCtaQfNbnl1hDKujv+W/1TcMx5Hl5eTp06JCeeOIJNTU1KTc3N7bvtddei3381VdfafDgwVxiAYB+5hjywsJC1dfXq7y8XNFoVKWlpaqpqZHP51NBQUF/zAgAuAXHkLtcLpWUlHTblpOTc8P9Zs+e3XdTAQB6jF8IAgDjCDkAGEfIAcA4Qg4AxhFyADCOkAOAcYQcAIwj5ABgHCEHAOMIOQAYR8gBwDhCDgDGEXIAMI6QA4BxhBwAjCPkAGAcIQcA4wg5ABhHyAHAOEIOAMYRcgAwjpADgHGEHACMI+QAYBwhBwDjCDkAGEfIAcA4Qg4AxhFyADCOkAOAcYQcAIwj5ABgXLrTHSKRiCorK9XS0iKPx6P58+fL5/PF9tfU1OiHH36QJD3++OOaNWtW4qYFANzA8Yy8trZW4XBYwWBQxcXFqqqqiu37448/tH//fq1cuVIrV65UfX29WlpaEjowAKA7xzPyxsZG5efnS5ICgYCam5tj+4YMGaKlS5fK5frn+0FnZ6c8Ho/ji/r9/njnva3HpgrWkHzW55dYQypIlfkdQx4KhZSVlRW77XK51NXVJbfbrfT0dA0aNEjRaFRbtmzRqFGjerSw1tbWuIb1+/1xPzZVsIbksz6/xBpSQX/Pf6u2Oobc6/UqFArFbkejUbnd7tjtq1evasOGDfJ6vZo7d+5tjgoA6C3Ha+R5eXmqq6uTJDU1NSk3Nze2LxqNavXq1Ro5cqRKSkpil1gAAP3H8Yy8sLBQ9fX1Ki8vVzQaVWlpqWpqauTz+RSJRPTTTz8pHA7r8OHDkqTi4mIFAoGEDw4A+IdjyF0ul0pKSrpty8nJiX28bdu2vp8KANBjXAsBAOMIOQAYR8gBwDhCDgDGEXIAMI6QA4BxhBwAjCPkAGAcIQcA4wg5ABhHyAHAOEIOAMYRcgAwjpADgHGEHACMI+QAYBwhBwDjCDkAGEfIAcA4Qg4AxhFyADCOkAOAcYQcAIwj5ABgHCEHAOMIOQAYR8gBwDhCDgDGEXIAMI6QA4BxhBwAjHMMeSQS0aZNm1RWVqaKigr9/vvv3fZ/8803WrJkicrKynTo0KGEDJlZXa2hRUVSerqGFhUps7o6Ia+TSIlew7XnfyA3t8fPH89jLLsTvo6Am0l3ukNtba3C4bCCwaCamppUVVWlxYsXS5La2tq0e/durVq1SuFwWMuWLdP48ePl8Xj6bMDM6mrdX1oau+1paND9paU6J6ljxow+e51ESvQa4nn+O+G49sbdtl7cXRzPyBsbG5Wfny9JCgQCam5uju37+eeflZeXJ4/Ho6ysLPl8PrW0tPTpgNlr1/ZqeypK9Brief474bj2xt22XtxdHM/IQ6GQsrKyYrddLpe6urrkdrt1+fLlbvu8Xq8uX77s+KJ+v7/nEzY13XSz5/jx3j1PMiV6DfE8/23OZObYX3MnfB39i9W5r2d9Dakyv2PIvV6vQqFQ7HY0GpXb7ZYkZWVlqaOjI7YvFApp4MCBji/a2tra4wGHBgLyNDTcsD388MM624vnSaZEryGe57+dmfx+f68+h6ngTvg6up7Fz8G/WV9Df89/q28ajpdW8vLyVFdXJ0lqampSbm5ubN+YMWPU0NCgq1ev6vLlyzp16pRGjBjRByP/v/aFC3u1PRUleg3xPP+dcFx7425bL+4u7oqKiopb3cHv9+vIkSPauXOnDh8+rHnz5mnfvn26cOGCRo8eLbfbrc8++0x79+7VrFmzNHLkSMcXbW9v7/GAnWPHKjxmjNJ/+UXutjaF8/L09/vvm/oBVaLXcP3zu9ra1NmD54/nMddkZ2f36nOYCu6Er6PrWfwc/Jv1NfT3/NnZ2f+5Ly0ajUb7bZL/ifd/R6z/r5jEGlKB9fkl1pAKTF1aAQCkNkIOAMYRcgAwjpADgHGEHACMS8q/WgEA9B3OyAHAOEIOAMYRcgAwjpADgHGEHACMI+QAYBwhBwDjHN9YIpUcP35c27Ztk8Nf3k05nZ2d2rBhg86ePatwOKyXXnpJBQUFyR6rVyKRiDZu3KjTp0/L5XJpwYIF8vl8yR4rLn///beWLFmi8vJy5eTkJHucXlu8eHHsnbmGDRum0uvei9SCnTt36uDBg+rs7NSzzz6r6dOnJ3ukXtm7d6/27t0rSQqHwzpx4oQ2bdrUozfVSRQzIa+urta+ffuUmZmZ7FF67bvvvlN2drYWLlyo9vZ2LV682FzIDx48KElasWKFjh492u1NuC3p7OzUpk2blJGRkexR4nL16lVJMncyc83Ro0d17NgxrVixQlevXtWuXbuSPVKvTZ06VVOnTpUkVVZWatq0aUmNuGTo0srw4cP19ttvJ3uMuEyaNEmvvPJK7Pa1t8qzpLCwUG+++aYk6ezZs7r33nuTPFF8tmzZoqefflr33XdfskeJS0tLi65cuaKVK1dq+fLlavqP9yJNVUeOHFFubq4+/PBDffDBB5owYUKyR4pbc3OzfvvtNxUVFSV7FDshnzhxoskASlJmZmbsvU8/+ugjvfrqq8keKS5ut1vr1q3T559/rokTJyZ7nF7bu3evBg0apPz8/GSPErcBAwbohRdeUFlZmebNm6e1a9eqq6sr2WP12IULF/TLL79o0aJFmjdvnj7++GNZ/SshO3fu1Msvv5zsMSQZCrl1f/75p5YvX64nn3xSU6ZMSfY4cXvrrbe0Zs0affrpp93eeNuCPXv2qL6+XhUVFTpx4oTWrVuntra2ZI/VKw888ICeeuoppaWlye/365577tH58+eTPVaPZWdn67HHHlN6err8fr8yMjJ04cKFZI/Va5cuXdKpU6f0yCOPJHsUSYaukVvW1tamYDCoN954Q48++miyx4nLvn379Ndff2nmzJnKyMhQWlqaXC5b5wHLly+PfVxRUaF58+Zp8ODBSZyo9/bs2aOTJ09q7ty5OnfunEKhkKnLRGPHjtXXX3+t559/XufPn1dHR8ct34syVTU0NKTUf8uEvB/s3LlTFy9e1I4dO7Rjxw5J0tKlS039wK2wsFDr16/Xe++9p87OTr3++uum5r9TTJ8+XZ988omWLVumtLQ0LViwwNQlxwkTJqihoUFLly5VJBLRnDlzzJ0QSP+87/Dw4cOTPUYMf8YWAIyz960QANANIQcA4wg5ABhHyAHAOEIOAMYRcgAwjpADgHH/B+5bHLdxPYNsAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x18f5333a5c0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "i = np.array([1, 2, 2.5, 2.75, 3, 4, 4.5, 6, 7, 7.25])\n",
    "o = np.array([0, 0, 0, 0, 1, 0, 1, 1, 1, 1])\n",
    "plt.plot(i, o, 'ro')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### OLS Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 150,
   "metadata": {},
   "outputs": [],
   "source": [
    "reg = np.polyfit(i, o, deg=3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 151,
   "metadata": {},
   "outputs": [],
   "source": [
    "oreg = np.polyval(reg, i)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 152,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([-0.02871811,  0.06731748,  0.16709993,  0.22604232,  0.28953689,\n",
       "        0.56680847,  0.70386016,  1.00198139,  1.01761939,  0.98845208])"
      ]
     },
     "execution_count": 152,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "oreg"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 153,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x18f533ac080>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(i, o, 'ro')\n",
    "plt.plot(i, oreg, 'b')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Logistic Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 154,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn import linear_model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 155,
   "metadata": {},
   "outputs": [],
   "source": [
    "lm = linear_model.LogisticRegression()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 156,
   "metadata": {},
   "outputs": [],
   "source": [
    "ir = i.reshape((1, -1)).T  # due to sklearn reqs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 157,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=True,\n",
       "          intercept_scaling=1, max_iter=100, multi_class='ovr', n_jobs=1,\n",
       "          penalty='l2', random_state=None, solver='liblinear', tol=0.0001,\n",
       "          verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 157,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lm.fit(ir, o)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 158,
   "metadata": {},
   "outputs": [],
   "source": [
    "pred = lm.predict(ir)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 159,
   "metadata": {},
   "outputs": [],
   "source": [
    "prob = lm.predict_proba(ir)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 160,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x18f538edeb8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(i, o, 'ro', label='data')\n",
    "plt.plot(i, pred, 'b', label='pred') # predicition 0 or 1\n",
    "plt.plot(i, prob, 'g--', label='prob') # probabilities\n",
    "plt.legend(loc=0) # loc=0 is translated to best location in matplotlib, there are other codes, see docs \n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Using Deep Neural Networks"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 161,
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 162,
   "metadata": {},
   "outputs": [],
   "source": [
    "tf.logging.set_verbosity(tf.logging.ERROR)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 163,
   "metadata": {},
   "outputs": [],
   "source": [
    "fc = [tf.contrib.layers.real_valued_column('i', dimension=1)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 164,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = tf.contrib.learn.DNNClassifier(hidden_units=[30], feature_columns=fc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 165,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_data():\n",
    "    fc = {'i': tf.constant(i)}\n",
    "    la = tf.constant(o, shape=[len(o), 1])\n",
    "    return fc, la"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 166,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DNNClassifier(params={'head': <tensorflow.contrib.learn.python.learn.estimators.head._BinaryLogisticHead object at 0x0000018F565B3588>, 'hidden_units': [30], 'feature_columns': (_RealValuedColumn(column_name='i', dimension=1, default_value=None, dtype=tf.float32, normalizer=None),), 'optimizer': None, 'activation_fn': <function relu at 0x0000018F55EB9730>, 'dropout': None, 'gradient_clip_norm': None, 'embedding_lr_multipliers': None, 'input_layer_min_slice_size': None})"
      ]
     },
     "execution_count": 166,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit(input_fn=get_data, steps=50)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 167,
   "metadata": {},
   "outputs": [],
   "source": [
    "dnnpred = list(model.predict(input_fn=get_data))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 168,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x190c4a18630>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(i, o, 'ro', label='data')\n",
    "plt.plot(i, pred, 'b-.', label='log')\n",
    "plt.plot(i, dnnpred, 'm--', label='dnn')\n",
    "plt.legend(loc=0)\n",
    "plt.show()"
   ]
  }
 ],
 "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": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}