REG03-NB01_back.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Regression Week 3: Assessing Fit (polynomial regression)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this notebook you will compare different regression models in order to assess which model fits best. We will be using polynomial regression as a means to examine this topic. In particular you will:\n",
    "* Write a function to take an SArray and a degree and return an SFrame where each column is the SArray to a polynomial value up to the total degree e.g. degree = 3 then column 1 is the SArray column 2 is the SArray squared and column 3 is the SArray cubed\n",
    "* Use matplotlib to visualize polynomial regressions\n",
    "* Use matplotlib to visualize the same polynomial degree on different subsets of the data\n",
    "* Use a validation set to select a polynomial degree\n",
    "* Assess the final fit using test data\n",
    "\n",
    "We will continue to use the House data from previous notebooks."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Fire up Turi Create"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import turicreate"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Next we're going to write a polynomial function that takes an SArray and a maximal degree and returns an SFrame with columns containing the SArray to all the powers up to the maximal degree.\n",
    "\n",
    "The easiest way to apply a power to an SArray is to use the .apply() and lambda x: functions. \n",
    "For example to take the example array and compute the third power we can do as follows: (note running this cell the first time may take longer than expected since it loads Turi Create)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1.0, 2.0, 3.0]\n",
      "[1.0, 8.0, 27.0]\n"
     ]
    }
   ],
   "source": [
    "tmp = turicreate.SArray([1., 2., 3.])\n",
    "tmp_cubed = tmp.apply(lambda x: x**3)\n",
    "print(tmp)\n",
    "print(tmp_cubed)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can create an empty SFrame using turicreate.SFrame() and then add any columns to it with ex_sframe['column_name'] = value. For example we create an empty SFrame and make the column 'power_1' to be the first power of tmp (i.e. tmp itself)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "+---------+\n",
      "| power_1 |\n",
      "+---------+\n",
      "|   1.0   |\n",
      "|   2.0   |\n",
      "|   3.0   |\n",
      "+---------+\n",
      "[3 rows x 1 columns]\n",
      "\n"
     ]
    }
   ],
   "source": [
    "ex_sframe = turicreate.SFrame()\n",
    "ex_sframe['power_1'] = tmp\n",
    "print(ex_sframe)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Polynomial_sframe function"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Using the hints above complete the following function to create an SFrame consisting of the powers of an SArray up to a specific degree:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "def polynomial_sframe(feature, degree):\n",
    "    # assume that degree >= 1\n",
    "    # initialize the SFrame:\n",
    "    poly_sframe = turicreate.SFrame()\n",
    "    # and set poly_sframe['power_1'] equal to the passed feature\n",
    "    poly_sframe['power_1'] = feature\n",
    "    # first check if degree > 1\n",
    "    if degree > 1:\n",
    "        # then loop over the remaining degrees:\n",
    "        # range usually starts at 0 and stops at the endpoint-1. We want it to start at 2 and stop at degree\n",
    "        for power in range(2, degree+1): \n",
    "            # first we'll give the column a name:\n",
    "            name = f\"power_{power}\"\n",
    "            # then assign poly_sframe[name] to the appropriate power of feature\n",
    "            poly_sframe[name] = poly_sframe['power_1'].apply(lambda x: x**power)\n",
    "    return poly_sframe"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To test your function consider the smaller tmp variable and what you would expect the outcome of the following call:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "+---------+---------+---------+---------+\n",
      "| power_1 | power_2 | power_3 | power_4 |\n",
      "+---------+---------+---------+---------+\n",
      "|   1.0   |   1.0   |   1.0   |   1.0   |\n",
      "|   2.0   |   4.0   |   8.0   |   16.0  |\n",
      "|   3.0   |   9.0   |   27.0  |   81.0  |\n",
      "+---------+---------+---------+---------+\n",
      "[3 rows x 4 columns]\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(polynomial_sframe(tmp, 4))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Visualizing polynomial regression"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's use matplotlib to visualize what a polynomial regression looks like on some real data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "sales = turicreate.SFrame('home_data.sframe/')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As in Week 3, we will use the sqft_living variable. For plotting purposes (connecting the dots), you'll need to sort by the values of sqft_living. For houses with identical square footage, we break the tie by their prices."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "sales = sales.sort(['sqft_living', 'price'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's start with a degree 1 polynomial using 'sqft_living' (i.e. a line) to predict 'price' and plot what it looks like."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "poly1_data = polynomial_sframe(sales['sqft_living'], 1)\n",
    "poly1_data['price'] = sales['price'] # add price to the data since it's the target"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "NOTE: for all the models in this notebook use validation_set = None to ensure that all results are consistent across users."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre>Linear regression:</pre>"
      ],
      "text/plain": [
       "Linear regression:"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of examples          : 21613</pre>"
      ],
      "text/plain": [
       "Number of examples          : 21613"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of features          : 1</pre>"
      ],
      "text/plain": [
       "Number of features          : 1"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of unpacked features : 1</pre>"
      ],
      "text/plain": [
       "Number of unpacked features : 1"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of coefficients    : 2</pre>"
      ],
      "text/plain": [
       "Number of coefficients    : 2"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Starting Newton Method</pre>"
      ],
      "text/plain": [
       "Starting Newton Method"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |</pre>"
      ],
      "text/plain": [
       "| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| 1         | 2        | 1.036640     | 4362074.696077     | 261440.790724                   |</pre>"
      ],
      "text/plain": [
       "| 1         | 2        | 1.036640     | 4362074.696077     | 261440.790724                   |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>SUCCESS: Optimal solution found.</pre>"
      ],
      "text/plain": [
       "SUCCESS: Optimal solution found."
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre></pre>"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "model1 = turicreate.linear_regression.create(poly1_data, target = 'price', features = ['power_1'], validation_set = None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\"><table frame=\"box\" rules=\"cols\">\n",
       "    <tr>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">name</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">index</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">value</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">stderr</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">(intercept)</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">None</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">-43579.0852514517</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">4402.689697427722</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">power_1</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">None</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">280.6227708858482</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1.9363985551321259</td>\n",
       "    </tr>\n",
       "</table>\n",
       "[2 rows x 4 columns]<br/>\n",
       "</div>"
      ],
      "text/plain": [
       "Columns:\n",
       "\tname\tstr\n",
       "\tindex\tstr\n",
       "\tvalue\tfloat\n",
       "\tstderr\tfloat\n",
       "\n",
       "Rows: 2\n",
       "\n",
       "Data:\n",
       "+-------------+-------+-------------------+--------------------+\n",
       "|     name    | index |       value       |       stderr       |\n",
       "+-------------+-------+-------------------+--------------------+\n",
       "| (intercept) |  None | -43579.0852514517 | 4402.689697427722  |\n",
       "|   power_1   |  None | 280.6227708858482 | 1.9363985551321259 |\n",
       "+-------------+-------+-------------------+--------------------+\n",
       "[2 rows x 4 columns]"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#let's take a look at the weights before we plot\n",
    "model1.coefficients"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x11f0e4f90>,\n",
       " <matplotlib.lines.Line2D at 0x11f0f1210>]"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(poly1_data['power_1'],poly1_data['price'],'.',\n",
    "         poly1_data['power_1'], model1.predict(poly1_data),'-')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's unpack that plt.plot() command. The first pair of SArrays we passed are the 1st power of sqft and the actual price we then ask it to print these as dots '.'. The next pair we pass is the 1st power of sqft and the predicted values from the linear model. We ask these to be plotted as a line '-'. \n",
    "\n",
    "We can see, not surprisingly, that the predicted values all fall on a line, specifically the one with slope 280 and intercept -43579. What if we wanted to plot a second degree polynomial?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre>Linear regression:</pre>"
      ],
      "text/plain": [
       "Linear regression:"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of examples          : 21613</pre>"
      ],
      "text/plain": [
       "Number of examples          : 21613"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of features          : 2</pre>"
      ],
      "text/plain": [
       "Number of features          : 2"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of unpacked features : 2</pre>"
      ],
      "text/plain": [
       "Number of unpacked features : 2"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of coefficients    : 3</pre>"
      ],
      "text/plain": [
       "Number of coefficients    : 3"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Starting Newton Method</pre>"
      ],
      "text/plain": [
       "Starting Newton Method"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |</pre>"
      ],
      "text/plain": [
       "| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| 1         | 2        | 0.048454     | 5913020.984255     | 250948.368758                   |</pre>"
      ],
      "text/plain": [
       "| 1         | 2        | 0.048454     | 5913020.984255     | 250948.368758                   |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>SUCCESS: Optimal solution found.</pre>"
      ],
      "text/plain": [
       "SUCCESS: Optimal solution found."
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre></pre>"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "poly2_data = polynomial_sframe(sales['sqft_living'], 2)\n",
    "my_features = poly2_data.column_names() # get the name of the features\n",
    "poly2_data['price'] = sales['price'] # add price to the data since it's the target\n",
    "model2 = turicreate.linear_regression.create(poly2_data, target = 'price', features = my_features, validation_set = None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\"><table frame=\"box\" rules=\"cols\">\n",
       "    <tr>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">name</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">index</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">value</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">stderr</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">(intercept)</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">None</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">199222.49644461478</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">7058.004835516328</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">power_1</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">None</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">67.99406406774321</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">5.287872013161801</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">power_2</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">None</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0.03858123127891487</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0.0008982465470323481</td>\n",
       "    </tr>\n",
       "</table>\n",
       "[3 rows x 4 columns]<br/>\n",
       "</div>"
      ],
      "text/plain": [
       "Columns:\n",
       "\tname\tstr\n",
       "\tindex\tstr\n",
       "\tvalue\tfloat\n",
       "\tstderr\tfloat\n",
       "\n",
       "Rows: 3\n",
       "\n",
       "Data:\n",
       "+-------------+-------+---------------------+-----------------------+\n",
       "|     name    | index |        value        |         stderr        |\n",
       "+-------------+-------+---------------------+-----------------------+\n",
       "| (intercept) |  None |  199222.49644461478 |   7058.004835516328   |\n",
       "|   power_1   |  None |  67.99406406774321  |   5.287872013161801   |\n",
       "|   power_2   |  None | 0.03858123127891487 | 0.0008982465470323481 |\n",
       "+-------------+-------+---------------------+-----------------------+\n",
       "[3 rows x 4 columns]"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model2.coefficients"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x122f37fd0>,\n",
       " <matplotlib.lines.Line2D at 0x122f43250>]"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(poly2_data['power_1'],poly2_data['price'],'.',\n",
    "         poly2_data['power_1'], model2.predict(poly2_data),'-')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The resulting model looks like half a parabola. Try on your own to see what the cubic looks like:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre>Linear regression:</pre>"
      ],
      "text/plain": [
       "Linear regression:"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of examples          : 21613</pre>"
      ],
      "text/plain": [
       "Number of examples          : 21613"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of features          : 3</pre>"
      ],
      "text/plain": [
       "Number of features          : 3"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of unpacked features : 3</pre>"
      ],
      "text/plain": [
       "Number of unpacked features : 3"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of coefficients    : 4</pre>"
      ],
      "text/plain": [
       "Number of coefficients    : 4"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Starting Newton Method</pre>"
      ],
      "text/plain": [
       "Starting Newton Method"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |</pre>"
      ],
      "text/plain": [
       "| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| 1         | 2        | 0.076623     | 3261066.736008     | 249261.286346                   |</pre>"
      ],
      "text/plain": [
       "| 1         | 2        | 0.076623     | 3261066.736008     | 249261.286346                   |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>SUCCESS: Optimal solution found.</pre>"
      ],
      "text/plain": [
       "SUCCESS: Optimal solution found."
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre></pre>"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "poly3_data = polynomial_sframe(sales['sqft_living'], 3)\n",
    "my_features = poly3_data.column_names() # get the name of the features\n",
    "poly3_data['price'] = sales['price'] # add price to the data since it's the target\n",
    "model3 = turicreate.linear_regression.create(poly3_data, target = 'price', features = my_features, validation_set = None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x123457e10>,\n",
       " <matplotlib.lines.Line2D at 0x123457fd0>]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXUAAAEDCAYAAADKhpQUAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+j8jraAAAgAElEQVR4nO2deXxU9bn/38+ZSYCwxrBF9iiiBDeICmoV14riXrXV9tpbLd1uW7vbjXK5vfd2r95fvVVqe729VdQCblRaq9WCVpYkiiQiCJFAWBMIELZkZs7398c5M5mZzCSTZPY879eLFzPnnDnzzEnyOc88z/N9HjHGoCiKouQHVqYNUBRFUZKHirqiKEoeoaKuKIqSR6ioK4qi5BEq6oqiKHmEirqiKEoekTJRF5Hficg+EalJ8PjbReRdEakVkSdSZZeiKEo+I6mqUxeRS4AjwO+NMdO6OHYy8DRwuTGmWURGGmP2pcQwRVGUPCZlnroxZiVwIHybiJwiIn8WkSoRWSUip7u7Pg08ZIxpdl+rgq4oitID0h1TXwR80RgzA/g68N/u9tOA00TkDRFZLSLXpNkuRVGUvMCbrjcSkUHAhcAfRSS4uV+YHZOB2cBYYKWInGmMOZgu+xRFUfKBtIk6zreCg8aYc2LsawDWGGN8wAcishlH5Nel0T5FUZScJ23hF2PMYRzBvg1AHM52dz+L46UjIsNxwjF16bJNURQlX0hlSeNi4E1giog0iMg9wF3APSKyHqgFbnQP/wuwX0TeBV4FvmGM2Z8q2xRFUfKVlJU0KoqiKOknIU9dRL7iLgqqEZHFItI/1YYpiqIo3adLT11ExgCvA1ONMcdF5GngRWPMY/FeM3z4cDNx4sRk2qkoipLXVFVVNRljRvT2PIlWv3iBASLiA4qAXZ0dPHHiRCorK3trm6IoSp9BROqTcZ4uwy/GmJ3Az4DtwG7gkDHmpRgGzRORShGpbGxsTIZtiqIoSjfpUtRFpBinSmUScDIwUEQ+Hn2cMWaRMabCGFMxYkSvv0EoiqIoPSCRROmVwAfGmEZ3cdAynJWhiqIoSpaRiKhvB2aKSJE46/uvADam1ixFURSlJyQSU18DLAGqgQ3uaxal2C5FURSlByRU/WKM+QHwgxTboiiKovQSHWenKEpOU1XfzEOvbqGqvjnTpmQF6ezSqCiKklSq6pu569HVtPltCr0Wj987kxkTijNtVkZRT11RlJxldd1+2vw2tgGf32Z1nfYBVFFXFCVnmVlWQqHXwiNQ4LWYWVaSaZMyjoZfFEXJWWZMKObxe2eyum4/M8tK+nzoBVTUFUXJcWZMKFYxD0PDL4qiKHmEirqiKEoeoaKuKIqSR6ioK4qi5BEq6oqiKHmEirqiKEoeoaKuKIqSR6ioK4qi5BEq6oqiKHmEirqiKEoeoaKuKIqSR3Qp6iIyRUTeDvt3WETuS4dxiqKkDx02kR902dDLGLMJOAdARDzATuCZFNulKEoa0WET+UN3wy9XAFuNMfWpMEZRlMygwybyh+6K+keBxbF2iMg8EakUkcrGxsbeW6YoStrQYRP5gxhjEjtQpBDYBZQbY/Z2dmxFRYWprKxMgnmKoqSKqvrmiOES0c+V9CIiVcaYit6epztDMuYA1V0JuqIo2U+8GLqKee7TnfDLx4gTelEUJbfQGHr+kpCoi8hA4CpgWWrNURQlHWgMPX9JKPxijDkK6E9dUfIEHdicv+jgaUXpo2gMPT/RNgGKoih5hIq6oihKHqGiriiKkkeoqCuKouQRKuqKoih5hIq6oihKHqGiriiKkkeoqCuKouQRKuqKoih5hIq6oihKHqGiriiKkkeoqCuKouQRKuqKoih5hIq6oihKHqGiriiKkkeoqCuKouQRiY6zGyYiS0TkPRHZKCKzUm2YoiiK0n0SnXz0IPBnY8xHRKQQKEqhTYqiKEoP6VLURWQocAnwSQBjTBvQllqzFEVRlJ6QSPhlEtAI/I+IvCUij4rIwOiDRGSeiFSKSGVjY2PSDVUURVG6JhFR9wLTgV8bY84FjgL3Rx9kjFlkjKkwxlSMGDEiyWYqiqIoiZCIqDcADcaYNe7zJTgiryiKomQZXYq6MWYPsENEpribrgDeTalViqIoSo9ItPrli8DjbuVLHfDPqTNJURRF6SkJibox5m2gIsW2KIqiKL1EV5QqiqLkESrqiqIoeYSKuqIoSh6hoq4oOUpVfTMPvbqFqvrmTJuiZBGJVr8oipJFVNU3c9ejq2nz2xR6LR6/dyYzJhRn2iwlC1BPXVFykNV1+2nz29gGfH6b1XX7M21S3+bwLqj+P/j7TzJtiXrqipKLzCwrodBr4fPbFHgtZpaVZNqkvoW/Fbavhi0vw5ZXYF+ts714Ilz8VfBkTlpV1BUlB5kxoZjH753J6rr9zCwr0dBLOjjwQbuIf7ASfEfBKoAJF8JV/wanXgkjzwCRjJqpoq4oOcqMCcUq5qmk7Shse8MV8pfhwFZne/FEOOdOR8QnXgz9BmXUzGhU1BUliVTVN6v3nKsYA43vtYt4/ZsQaAXvAJh0CVzwWTj1Cig5JdOWdoqKuqLEoCfirBUpOcjxg/DB39vDKod3OttHnAHnf9rxxsfPgoL+mbWzG6ioK0oUPRXnWBUpKupZhm3DnvXtIr5jLZgA9BsCZbPh0m853vjQsZm2tMeoqCtKFD0VZ61IyVKONsHWv7UL+bEmZ3vpOXDxVxxvfGwFeAoya2eSUFFXlCh6Ks5akZIlBPyws7I9Nr7rbcBAUQmccoUj4qdcDoPyc0KbGGOSftKKigpTWVmZ9PMqSrrQhGeOcWgnbH3FEfGtr0HrIRALxp7viPipVzieuZW96y1FpMoY0+sW5+qpK0oMelouqDeDNGLb8N5yeOMB2FnlbBt8MpTf6HjkZZfCgL73M1BRV5QkodUvaSLgg5qlsOoX0LQJiifBVQth8tUw4vSML/7JNAmJuohsA1qAAOBPxlcERck3tPolxfhOwNuPO575we0wcirc+luYelNGl+VnG925EpcZY5pSZomi5Dha/ZIiWo9A5e/gzV/Bkb0wpgLm/AQmfzirY+SZQm9vipIktPolyRw7AGsXwepfw4mDMOlSuOU3zurOPh5i6YxERd0AL4mIAR4xxiyKPkBE5gHzAMaPH588CxUlh9B+LEmgZY/jlVf+D7QdgSnXOp0Px52XactygkRF/WJjzE4RGQn8VUTeM8asDD/AFfpF4JQ0JtlORVHyneZ6eONBeOsPYPtg2q3O4qBR5Zm2LKdISNSNMTvd//eJyDPA+cDKzl+lKIqSAI2b4PVfwjtPO7Xl59wJF3056xtnZStdirqIDAQsY0yL+/hqYGHKLVMUJb/Z9Ras+jlsXA7e/nDBZ2DWv8DQMZm2LKdJxFMfBTwjTmLCCzxhjPlzSq1SFCV/2faGI+ZbX4F+Q+GSrzttbQcOz7RleUGXom6MqQPOToMtiqLkK8Y4S/hX/Ry2vwlFw+GKH8B590D/oZm2Lq/QkkZFUVKHHYCNLzhivucdGDLWqTE/9xNQWJRp6/ISFXVF6QOkvSdNwOckPl//Jex/H0pOhRsfgjNvB29h6t+/D6Oirih5Tlp70viOOyWJbzwIh3bAqDPhI/8DU28Ey5Oa91QiUFFXlDwnLT1pThyGyt/Cmw/B0UYYdwFc9wuYfJWu/kwzKuqKkuektCfNsQPOMv61j8CJQ87wiQ99DSZcpGKeIVTUlZwmVqxYe5pHkpKeNId3ty/l9x2F0+fCh74KY2b0/txKr1BRV3KWWLFiQHuaxyBpPWkOfOC0vn37Caey5cyPOEv5R57R+3MrSUFFXclZYsWKAe1pngr2vutUstQsAcsL534cLvwSnDQp05YpUaioKzlLvFix9jRPIjurnAlD7y2HgoEw8/POUv4hpZm2TImDirqSs8SLFWtP815iDGx7HVb9DOpeg/7D4NL7nd4sRSdl2jqlC8SY5HfJraioMJWVlUk/r6IoKcQY2PwXZ/Vnw1oYOBIu/Beo+BT0G5xp6/IeEalKxqhQ9dQVRXE88pe+B3s2wNDxcO3PnLh5wYBMW6Z0ExV1RenLNL0PL30fNq+AYePhpl/DmbeBpyDTlik9REVdUeKQ1/XuJw7Baz92Fg15B8CVC+CCz0FB/0xbpvQSFXVFiUFa+6WkE9uG9U/AywvgaBNM/ye4/PswaESmLVOShIq6osQgLf1S0s3OKnjxm7Cz0unNctcSOPmcTFulJBkVdUWJQUr7paSbI43wygKne+KgUXDzI3DWHdqbJU9JWNRFxANUAjuNMXNTZ5KiZJ6U9EtJNwEfrP0NvPYj8B1zVoBe+k0tT8xzuuOpfxnYCAxJkS2Kkla6SoQmrV9KJqh7DVZ8Cxrfg1OugDk/huGTM22VkgYSEnURGQtcB/w78NWUWqQoaSBvE6EHt8Nfvgsbn4fiifDRxTBljoZa+hCJeuoPAN8E4n5vE5F5wDyA8ePH994yRUkheZcI9R13pg29/ktA4PLvwawvaoliH6RLUReRucA+Y0yViMyOd5wxZhGwCJw2AUmzUFFSQN4kQo1xmm395TuOl15+M1z9Qxg6NtOWKRkiEU/9IuAGEbkW6A8MEZE/GGM+nlrTlL5CJhb55EUitOl9ePEbUPcqjJwKdy+HSR/KtFVKhulS1I0x3wa+DeB66l9XQVd6QrwpRZmIbef0alHfcafp1hsPOqtB5/wEKu4Bj1YoK1qnrqSJeOKdidh2TidJ338ZXvwaNG9zas2v/iEMGplpq5QswurOwcaY17RGXekJ8aYUBWPbHiFtse14tmQ1h3fB03fD47c6k4f+6Xm4ZZEKutIB9dSVtBAU7zafjYhQXFQIZCa2nVNJ0oAf1v0G/vbvEGiDy74HF30JvP0ybZmSpaioKwkTHYfublz6Q5NH8Lf39mEbw8LltUwZPTi0wCed4Y+cSZI2VMHy+2DPO3DqlXDtT+GkskxbpWQ5KupKQkTHoefPLWfh8tqE4tJV9c18bNGbtAXaK10zXRue1atFjx+EVxZC5e9g8Gi47TGYepMuIFISQkVdSYjwOHSrz+apddtDz9u6EOil1Q0Rgi6kL36eUxgDG/7o1Jwf2w8XfBYu+w70184cSuKoqCsJMbOsBK8ltAUMBqjZdQjb1WnbEIqRxyLavzxr7FDmX1+evZ5yJti/Ff70Vadny5gZ8PGlUHp2pq1ScpBuVb8ofZcZE4q5rWJcSKBtu32fBTQfa4v72lumj6XQayHghG5U0NsJ+GDVL+DXF8LOarju53DPX1XQlR6jnrqSMLdMH8vS6gZ8fhuPJSBCINB1BcmMCcUs/nQOJCbTTUMlPP8l2FcLZ9zgLCIaUpppq5QcR0VdSZjoqhEgIaHO6dWbqeDEYfjbvzm9zgeXwkefgNOvy7RVSp6got5H6Y7QRh8bfnz0a2OVPebs6s1U8N6f4E9fh5bdcP48p5uiJkKVJKKi3gfpjtD25tj5c8tZUbM7v1rc9pTDu5zmW+8th1HT4I7/g7EVmbZKyUNU1Psgy6obaPXZGLoW2u70Zgk/ts1nM/+5GgK2Uy1jJaENQLaEcTqzo8M+24bK38LL/wq2D65cALP+BTwFGbFdyX9U1PsYVfXN/LFyB8GqcY+nc6HtzpL68GNFBNu4gg5cdOpw7rvytB6LcW/COMm8GXRmR/S+pbcWU175fWhYC2WzYe4vdUWoknJU1PsYq+v243cLzAX4yIyxnQpdd5bUhx9bXFTIwuW1oZtBbwQ9aHdPwjjJjul3ZkdwX4Fp40vmGU5/bjkMGAY3P+J0VNQVoUoaUFHvY0R73rdO7zghJ1ZiNJj0fOjVLZ2Ke3gidcrowUnzkIOLn3wBg8eShMM48ToyBm88zcfaumVfZ99cZpaV8CHvRv5VFjHR2kvTKbcy/OafwkBdOaukDxX1PkasssRwoY7n2fbE4016fxURwHTL440W4eKiQu56dHUop2AJ3fLg435zOd7MjLe/x/96/sCh/mPZfOnjnDZLu1Qr6UdFPU/pLI4c7nknWq3SWdghHQnM1XX78QccIQ4EEg+/RItweJIY6FFVTsTNyhh491l48ZtOv5aL7mPo7PsZWjCgZx9UUXqJinoeEuyK6AsYCjzC4nmzYrbK7axaRYhMosYLO6SrDr03PdDDb2LhSWLoZVXOoZ3wp6/B5hVQeo7br+Ws7p9HUZJIl6IuIv2BlUA/9/glxpgfpNowpeeEd0VsCxiWVjcAdBDfeNUqgPO/aZe/eGGHdI2jS0YP9Ogk8ZVTR3HOuGHdP19EmaLfGSl3wed0RqiSFSTyW9gKXG6MOSIiBcDrIrLCGLM6xbYpPSQ64izEFt8vXHYqj987k6XVDTS1tPLa5kZ8/vbQhM+9IQQFL1zIg89nlpXg9Tg3hujyyGSHZRKN0cd732hv/7OXntJ9u/a9By98CXasgbLL3DLFSd39KIqSMroUdWOMAY64Twvcfyb+K5RMU37yUDyWYNuGAq/FLW6FS7zwxbLqBtr8NpbAKSMG8sH+owRs54e8pKqBW6ePDYUvosM6QLtHH+bZpyIsk8hNorP37Y633+G9/K1ON8VVP4d+g+Cmh+Hsj2qZopJ1JPR9UUQ8QBVwKvCQMWZNjGPmAfMAxo8fn0wblW5QVd/MwuW12LZT+rcgrM1tLEELTxzaBrY2Ho3QqfCkZKywzphhA/C7cfiAbRJKrPb0cyVyk+jqfRPx9qPf69nrvZy+9rvQtAnOvA0+/J8waESPP4uipJKERN0YEwDOEZFhwDMiMs0YUxN1zCJgEUBFRYV68mkm6FnuOnicNjeEYoyJ6HMeLWixEofO69qfh4dUYoV1iosKsdxSQ48l7Dp4nKr65k7DMtF2L61uQHBa+/a2XUEyhkoH32ugOcb9LOb0F1+BoePhriUw+apuny9RsqUNgpLbdCuzY4w5KCKvAtcANV0dr6SHcM/Sawlej5VQn/PwxCGAxxICduT9+NLTRoQE5pbpY/ljVUNIMMtPHsqCF2rx2wZLnBvC4rXbWVrdwPy55THDMtF2h88u/WNVA4s/HdsDT1Ssk5FQnVlWwhxvFd+3fscIDrK3/B5G3bDQCbukCO1mqSSLRKpfRgA+V9AHAFcBP065ZUrChHuxAdtwx/njGDNsQJeiFi2U8+eW89S67axvOBQ6Jtw7jx52EYzFgxO6sV1x9vltVtTsjhmWibbbl+Aw6u62K+ixIB7ezYzV32CG5wWaBp7G5sse44yK2T07VzdIVxWRkv8k4qmXAv/rxtUt4GljzPLUmqV0h1hL/3u6OnLK6MHcsehN/K7Yvra5kar65ohkY/BxsFQyiMcSME5yds60UtZtO9CpZz2zrIQCj4Q89UQmKKVM6Gwbqh+Dvy6AQCtc8QOGX/hFhqepm2IywkaKAiAmzlfj3lBRUWEqKyuTfl4lPvHiseHbIf6kovDjllU38MSa7RjAI/DVq6fwhctOjfme4dUwC26YFtFLJdFqlURi6rFs7Unvlpg0boYXvgzb/wGTLoG5D0DJKT0/Xw/RmHrfRkSqjDG9brKvop5nRIt4eKwdEfwB5/FtFeMoP3kozcfaKC4qZMEL7R0VF1xfHtFhsavBGF3dTJIpUMHYc097t0Tgb4M3HoCVP4WCIvjwv8M5d3WrTFGFWEkWyRJ1XQKXR0Qn226dPrY9ThswgBPjbgsYHl+zHXBEUYBgaLvNb1O761DEoqRlbpglXqw7ltefqqRfMPbcm94tAOxY6wx9btwI026Fa34Eg0Z2yxZNbirZiIp6jpCIRxidbDO0LzjyuJ56+IpRcEQxmuCmJZU7EqpMiSbeZKVkeLXB2HObz8amB71bThyGVxbCukdhyBi482k47cM9skWTm0o2oqKexYTHjhcur+3SI4yVML11+tiIcMzS6gaWVDXg97eLotcSbMAfMHgEpp08tENlSpvf5oGXN3c57CLeZKWq+mY+9pvVIdsSvUFEEz2Io1sx9fdedBpwteyGCz7jDH3uN7jbNgTR5KaSjaioZynhX+0tt9lWPI8wPNk4f255B6GLJXiC006gZtchBBjcz8ujr3+AbQwLl9cyf255RGUKwBtbmli37UCHEW7RnR/Da9/PHuvcINbvOBgqf2zz2ywL6ynTXbpdBdOyF1Z8A959jqaBp9I4dxlnVFzeo/eOtqO3NfGKkmxU1LOU8K/2xhhE2mvG1+84GCozjF7AUxjmBUcLbnQMeHA/L0+v24FtTIcbR/OxNhbPm8XS6gZqdx5iw85DHW4qsWLKwQlFQXvWbWumqr65Q7egmp2HIkolU4IxUP17+Ov3sX3HedC+g18fuA7ruVYeH5Gc905pmaWi9AAV9SwlOnYcLFIKGHjp3b28trkxtBAo1gIe6NhqN/xG0eqzeWRlXVjC0VkVGh2jHjNsAEP6eanddRgwEfsi+rGHdX68rWJcKBHrnLvj59uw8xB3Pbq6V8nFTksbm7Y4ZYr1r9MyeiYLmcfS+v7YBjwa/1byGBX1LCX41f6BlzfzxpamDsIYFO94C3hiJfGibxTR2MaJr8+fW86mPS0dh2a4+4JiWFxUGLLLNs5zcMI6XdHb5GK80sYn/vlcpu/4Paz8GRT0Z9tFP+aaleNp9ZnQccmOf2tZo5JNqKhnAbFEIbgtuDIzfAQbgGUJxUWFbNrTwhmlQ+jntRhWVMiIwf2A2LM5l1Y3cMnkEew9fIJ3Gg7F7J9sjKFm1yGeXrcjIjZuwvYFZ5o2H2vDnRqKBaHmYc3H2rDEEW4Bzho7lHd3H3YSsR7BEonZm6Y74hhrLN25gVomLrkfjm2D8pvhmh/zp8oW2vybQjZedOrwLpO93UHLGpVsQ0U9w8QSBYgMnXxy1kQeff0DArZx4uriCOz85zbgD3O5CzxOQ66l1Q08fu/MiCqRBc/XhLx5r0cocAU/XNiD3njtzkMdGntZ4uxbUtWAP9Bul8dyYvGF3vij7+44bzwLnnf7vxnD7CkjGTG4X8QK0u6IY3SFTTGH+U7BYm7z/J1WaxzctRQmX+naUhhhSzIFHbSsUck+VNQzTCxRACJi32/W7Y8YNYch1P88HF9YQ63wMsbaXYci4u6BgOGyqSM54QsgwOth4Z2AIeTFB0X+3osnMXhAAbsOHmfx2u2hmaaPvv5BqEPjJ2dNjKi2CQ6xnjOtlOZjbaHmXn4b/vruXvoVtA/vqKpv5oGXNycsju0VNoaPeFbyg36LKTJH2T3tc5RePx8Ki0LHprpCRcsalWxDRT3DxAqT1Ow6hGUJdsARwppdh0Kia3A8ctt2KmKiPfXgtKPiosKQ5xsMkQQxwGub9uEPmA7bgx66BVw0OTJUESydDM40DR5rG3j09Q+4qnx0qComWFe/btsB5s8tp9BrhcIl4YuSgA6x8a7EcWZZCVO8u/mBPMpMayNHhs/Ac+uvKB01NebxqaxQ0bJGJUi25FZU1DNM9GKaoBiGdx+xo7Ka54wbxuwpIykuKuTVTfvYd/gEd5znTJtaUbOb8tIhrKjZHfJ8YxHuuUcjQGGBE6oA+M4zG0INt4IeeHnpkJCnDs7NYOELtUwb4yRJw73u5mNtobYDS6oaIuLp4cv+E4p5+04wY+t/86eCX9Jm9af+/P9kwpWfBcvq/EInQE//KLWsUcmm3IqKehYQFIWHXt0SEkMLpxIl2HAtXIPXbWvmyAk/m/cdwbjxbCAUN1/1flMH77w7XDV1FJ+59BQ27Wnhe89sCFXKPLluOx7Lwh9wPPB7L54UivUbYH3DIdY3HArF38VuL4EMfsbwFa7BX/qgF49AeemQ+C0Ftr4Kf/oqHKjDOvN2+n/435nQzX4t8cimP0ol98im3IqKehYRPhqu0B1aUbPrEE+t29FhetDGPS2hx21+m6fWbY9Y/Rn0fEcO6ceew62dvu9JAwtpPtoWCn+cPW4YQISgAwRsCLhfG3x+m8EDClh44zQWrdzKtv3HQsfZBsQ2fPT88R3a6UZ7tTMmFPPJWRN5eGUdxsDDK+sAeOzNbSGBfequUzm79iew4Wk4qQw+8SycclnXF7QbZNMfpZJ7ZFNupfffWZWkEIxDBytczhozlFp3Cb8dL4biYokwakj/Dts9HuFLV5xGoccJ5ngEPntJGVZUZ9kDYYIerGJZWt0Qs5bdYwkeN+4dDBeFC3oQA5w8bAAAD726xVlVGofa3Ycjnv+5do/TUsDYfMR+iSlLLofaZ+DSb8Hn3ky6oEP7H6UnBXXsSv4TDKN+9eopGf+Wp556lhAeWw4YWLutmbXbmuOWH4Zz78WTuKp8NK+8ty+iFNG2DVNGD2bxvFkRoYw9h0/w7Nu7Is4hRMazl0VNNQKnFHJh2CCMoM2xKPRatBz3cccjb4ZKHuP9ss+ZVsqq95tCz68pH03lm68wX37LWVYdLcNn0v/W/wcjTuvyOvYUTXgqvSVbciuJzCgdB/weGIXjgC0yxjyYasP6GkFPMXqRUSBguP0CZ+Zoy3FfRHISHDEePKCAGROK+bcbp/HdZze0z3uG0NJ93Meb9rTw59o9Hd7f4Ihr9JDpoGifP7GYb805o8MvbSybr546itlTRjL/uZqQrW2dhDTuvKA9yXvjaQP4yMFHMJ7HOFZYQt2sBymbfXe3Blf0lGz5o1SU3pCIp+4HvmaMqRaRwUCViPzVGPNuim3LC+JVVDyxZnuojjsoardMH0tTSyt/c8sNgQ4zR8eXDOR3r9expfEo4IhxcHn+lNGDsaQ9qWqAluO+Dh0foxcWgRN6Ca4WDfZS+dSFE0OdG9/ZeajDa8K925bjPmp3H6a8dAiDBxRQuytyAZMl0unq0TvPG8ud3lfh5QVw4hAy8/MMnH0/Zf2H9PgaK0pfpEtRN8bsBna7j1tEZCMwBlBR74RgTXf4Csxg+OGJNdv5zjMbAFj1fhPb9x8NJQa9lnD5FKeiI7jqEpy4dNBTjxTL9uX5y6obCIRFQ4KJx+Xv7Ap507YxxPJ5bQOL3bmk4HwDCL9BtPpi91MP927Dbx5eywkb+f02liUsvHFa3NWjy24axNTqBbCzEsZfCNf9DEaVx7ymsdoppKpqRW8WSi7SrZi6iEwEzgXWxNg3D5gHMH78+CSYlrtEN5uCyESUYBEAABpUSURBVIqKFTW7I44PJgZt44yae+ndvaEWukCHc4VjiYQ89Xgx94aDJyKexzuuw0IkE/l81ftNvLl1PwtvnBb6dhFOeAVJwDbccb4TNooWxeBxg8wRvs4fOf2FV2BgCdz8CJx1R8xQSzzxTlbVSldtiqNvFir4SraSsKiLyCBgKXCfMeZw9H5jzCJgETiDp5NmYQ4SPUczSKiLYemQDonB37xeF6GqbX6bH6/YyL6W1piCbgHi9l2Z/9wGnlq3nVllJRRGDbZINn7bMP+5GqaMHtyh+VjLcV+oJDM6bBTOzEnF3F6wim/I4wyjhaYz7mbXuV/hjZ1+Zm4/GPM18cQ7vPOkhN3gukMsAe/sZqE17Uo2k5Coi0gBjqA/boxZllqTcp9YLW4DBha8UMv2/Ud59PUPACfEcfrowVRvb44ImwRZuy12GaAlUDGhmMr6ZmzjhE6CC3/OGD0YX8AOxdx7Q7DTYjTBpmHhHm34jSe6F0w4GytfZczfv8ePrBp2DzmLTbP/k+Ml5V2KZLw64GCfmfnP1YSmNoXfcBKhszbFseqOtaZdyWYSqX4R4LfARmPML1JvUu4T3gv99febQmLX5rd5ZFVdRHVK+CKizhjUz8OVZ4zihXd2E7AN1TsO4vVYHUoKEz1fIlx5xiiGD+4XGnf3m1V1BNxmYkuqGkKrQ6O/ldgGHllZR13TUT5z6SmO4LXsoem573DGlqXsM8P4lvk8t9/ydWZMLIlYSdvZoOrwkkNob1/gvGf71Kal1Q3dCo3EEvDOShyzaaGJokSTiKd+EfAJYIOIvO1u+44x5sXUmZV7RAvQjAnFzJlWyj+2NIVi0/E830Q40hrg+fW7Qq/3BwxjivuDMeyMipkni+ZjbZw9bljoM7W0+kMTjfxdDN4wOBOa3ti8iz+UVzP1/YcZavt42H89/89/EydkAOM/aGbGxJLQSlpnbJ8TQgmO6fMFDAUeYfG8WaHrGj3Cz+sRvB6LQMDu0B44kdBIPAGPV+KoNe1KNpNI9cvrELNgQnGJ1xN94fLaUB+XspGDOKmoIG5IJZwRgwo5eNzXoelW9A1hZ/NxvB7pVZ+Xzqisd+aLBj9T+EQjG6dccnXd/tCw65bjPhatqnPtNFxhVfM96w9M2rSXvwam80P/x6k3oyNWrgZX0kY0BlteyyWTR4REuy3QHu4BOozwC6/lD28P3J3QSHdr1LWmXclWdEVpEoiOsQaHNZ/wOb6rAbbsO0KhR/B6JFSDHo/GI20Jv7c/ELtEMR7DBngZ2L+Anc3HuzzWuKGWNvcz/X3Tvoj9j7h9WgrcSp1Ne1ooGz4Q07SZ+d7/41LPO2yxT+butm/xd/vs0OvGn1TEvEtOiWhiFnpP9/32Ho789hH+GWON8AsmZcPbA+vYOqUvoqKeBMJjrB5LeLpyR0zhbgsYhg8qZP/Rtuj+XL2iO6c6eNzPweP+7p3fwJOu9xvrfYOVOpu27eBL3mf4p8KXOE4/Fvo+wRPmKvxSEGHltv3HWPCCk9AMn3MaxDYwq6yEjXtaQuIcrNcHx0tePG8WS6sbQi2Bw0MmqQiNaMWLkiuoqCeB8Ek/Awo8/PXdvXGPbeqGF54tGDo0iYygEB8X7H2SRf2WMIRjPBmYzc/9t7OfoXgE7jhvHI0trVTVH+DAUR/g3AiWVTdw8rABHcJHFk7rgwXXl/PUuu0xm5VFhz9i5TSSiVa8KLmCinoPCRcRcMoVfX4bj0fwJBBiyQ8Mc6y1fLtgMePNPlbaZ/If/rt4z4zHYwmW7SQ+h/Tzsqy6IRSOCrJ5bwu3TB9Lv4L2/jHBAR2Rc1UP8eqmfTzpJkujSYcXrRUvSq6got4DokXkkskjQnFhf8Bw6oiBlI0YBMCrm/Z1OmUo3fT3WpyI01kRoNAj+KLG3AER3vTowf2YZjbx3YLHmXS8lj39y7j78D9HxM2NbUCcUsNg/5ho1m1rZtOeFj45ayKLwko9zxozlNc27YtYROULGJaFJUvDSYcXrRUvSq6got4FsZJj0SISndTb0niU+gPHuL1iXFYJOsBZY4d2WoETbzWqwZnE9LWKQia8/VOutVazzzeMZ8Z9i82lN/D3VfURx9vBFwEYg2UJglNLHv4OT63bTu2uwxEzWOPZ19gSe9hHIqtKk5Hk1IoXJRdQUe+EeKWKOw8ex+txmlWJiJPU2324g2e5YsPueKfOGImUVMZiGC18wXqeT77zF/xi8aD/Fh7xz+XYlv5YW+vjvi4YTmmf4rQ9YvXsqCH9eaehYwfIWAwf3C/m9q5WlWqSU+lLqKh3QrhHHizrW1bdEBoMbXB6ofzuH9tYcMM0fvaX9zhwzBd6ffjjXKWIE9zjWcGnvcsZxAmWBi7hZ77b2MtJoWPcSEuHkI1HiBhp99CrWyISrsFZqCvfb4zZ3yZ4TgEKPMKtYRUw0TQfa4tYVRoegtEkp9KXUFHvhJbjvlBYwDawZW9Lh2QfOIL/2qZ9HDqR+yIepBAfd3pe4QveZxkhh3kpMIOf+W9nsxkX8/iSQYXsP9IW0f/lijNGRZQbRicbL5sykqXVDVwyeQQjBvej/OSh1Lgj/MpPHkrzsbZQb/euwiadJTI1yan0JcQks2DapaKiwlRWVib9vOmkqr6ZOx55M2LKUGcUejv2YclFLGxutl7nKwVLGCtNvBmYyk/8d/CWmdzp6zyWYLvzVUuLB7Dv8An8AYPHEu69eBKDBxRQXFQYIdrt1S2EWg3H6vmSKJ29ThcOKdmOiFQZYyp6ex711F2i/+hX1+2POSEoHrkv6IYPW5V8zfs0p1k7eceexLd997LKPpNEukQEr5WBiNWqftvw8Mq6UCgl2CIAiEgiB8MiQET8O9iCoLdCrUlOpa+gok7sRFpxUSEeSxL21HMXw2XW29znXcrZVh1b7VI+1/Zl/mzOx5jktfwJXsVgfsIAHguC90IDrN9xkF0Hj7fnMXx2KPnptYTbKsaFwjnRP7P5c8tZuLxWk6FKn0dFndgJ0SVVDfjt9r4q+SftkWK+wx7BN32fZmngEgJ4enTG4YMLOeCumLUEDLHnodoGpp08FDlvfKjrIzhdHQvDOi6KSCj52RYwPLFmO0urG2IOsVhRszv0vNVnRzQAU5S+RJ8X9ar6ZpaHtbS1Dby9vTkUTslHMb/ceosve5dFiPmywIfw9/LXoamlXdAX3ngmAN97ZgPRgSkLp1olvOtjEL9t+Kg7Bq+4qJCFy2tDlTEGx3t/4OXNzJlWGpH8nDOtlDV1+2lzF04F+72rsCt9jT4t6k4y9B9Eh8Pf3Z28QRPZgyPm93mXcpb1AdvtEXzDN49nAhd3S8zPHjuUDTsPddoX3jbw2qZ9HPcFOgg6OHXrM8tKWF23P2Yp5LFWP6vr9lNeOoRbpo+lqaWV1zY34vc7Pdvf2NLEum0HOsTba3cd4gl3eHYg0LF0MZEYvCZUlVynT4v66rr9HQQ937CwmWOt5XPe55lmbeuxmAcZNaQ/m/a2dBiKEc0r7+2LGXq52q1NDwpmQdRMVdvAs2/vApxB1wL0K7BYcL3TMO2NLU2hkEvzsTa+cNmpodfeMn1s3La7iSxACh/N57Ek7oBtRclm+rSo92RIca7QjzZu8aziM57lTLT2stUu7ZWYB6ne3sxZY4bS6reZVVbC4VY/S6oaIqp/LKGDoJ86chCfumgSd14wnqr6Zh56dQszy0pCLXRrdx6K+Q3A0C7g9115Guu2HYhbb95Zf5ZEFiCtrtsfCvXEGrCtKLlAIjNKfwfMBfYZY6al3qTUEfxqHVzQsn7HwQ7HeCwQEQKB9hK9XGIQx7jL8wr3eFcwUg6y3i7js2338ZJdgY3V6/M3HWkLtQ+u3X2Yy6eMxBf1dSe6ZkaAm88dExL0aI/5P24+M+YA6+BrE5kbGiRe6WIiC5BmlpVEVDzZxujqUyXnSMRlewz4FfD71JqSWqJFI94IuIAN54wdwtbGI7S0BtJsZc8ZSTN3e//CJzwvM0SOsSowjfsCn+cfdjmpmkboDxhe3ri3w3WM9rYtS0IiGu4Nt/naPeZwwW457qN292HKS4cweEBBQnNDuyLRG8LCG6eFyigLdfWpkoMkMqN0pYhMTL0pqSHona/fcTDCC+zMA387wQZT2cCZUsenvCu4zlqNF5sV9nk87L+BDaYsLe8fLuAeS0LTNMKbPU4fPwyAh17dQstxX3vNOvD+3vakdKoXCCVy/jsvGM+U0YM1WarkLAm1CXBFfXln4RcRmQfMAxg/fvyM+vr4nfvSQXBW5R8rd2Rd+9ve4iHA1VYln/Ku4DxrM0dMf54OzOaxwIfZbkZlxiZLuP6sUvYfbaNkYCHPv70rlEj1egRLBH/AaYQW/eP47CVl3H/tGR3OqZUoSl8iWW0Ckibq4WS698sTa7Yz/7mavFsNOoQj3OF5jbu9LzFWmthuj+CxwDX8MXApLRRl2jwgdlirqwVcIrDksxd2KD9MR7tcvXEo2YL2fonDE2u2871nN3RaR51bGKbL+9zlfYXrrNX0Fx9vBqayMPAJXrZnJCX52VOC0ZbwSx3rsgd7vmDi7Dd0SEimo12u9llX8pG8EvWq+ma+/1xNXgj6YI5xk+d17vS8whnWDo6Y/iwJXMLjgSvZaCZk2jzAiaefMXowW5uOhhYGhfdAN1HHCk6YxtgmosY9VkIyHe1yE7lxqCev5BqJlDQuBmYDw0WkAfiBMea3qTasJ/x4xcZudVbMPhyv/HbPa9zgeZMiaaXGnsi3fffwfOBCjjIg0wZ2YNPeFn5405kRvc9bjvt4eeNetjQejTjWABjDxy4YH9E7/RZ3+EWwdj26GiZVgtrVjUM9eSUXSaT65WPpMKS3/OjFjT0e1ZZpxtDITZ43uMWzilOs3Rwz/Xg+MIsnAlfwjikjVSWJycAYIlZ2hpeOxjweOHnYgIiVmsEcSLCMMCie6aiG6ezGoROTlFwkL8Iv9z35Vmhpea5QxAnmWGu51bOSCz3vArDGPp2HfdezInA+R7Ik8dkZwSX84R5uUAjDvy95LOdoE6P2u6q+OSKp3ZZm8ezsxqETk5RcJOdF/UcvbswZQe9PK7Ot9cz1rOZy6y2KpJV6eyS/8H2EZ+yL2JGhcsSeYAEXTR7OfVeeFjGtqLio0JkC5WuPsXtEWHDDtJjDLqKHkVgiWSOe6QgBKUqyyXlRf+wf2zJtQqf0o43Z1nqu86zmCquagdJKkxnCM4GLeSZwEZVmCtkcXgnHa4FlWfj9NpYlzJlWGndgRXjzLX/AsKJmd+gGEM7MshL6FTg3ActtopVN4qkTk5RcIydFPegVrqnbz4ksbLNYzGEus97mcs9bzLbWM0hOcMAM4rnARSy3Z7LGPqPHgygyhdcV3O37j7JoVR0B27BweW1o9WV47Dm8+VbQYw+2y41ONqo3rCjJJedEPegVnoiTiMsMhsmykyutai73VDNd3scjhr1mGM8HZvEneyar7ak5J+ThGGOo2XWIp9ZuD5WMBuPfsWLPQbF+4OXNIY893kQi9YYVJXnknKh//9kNWSHoIzjILKuWi6xaLvLUMFaaANhgT+RXgZt5OTCdGjMRk8HFQclERNiytyViib+48e943vaMCcXcd+VpOpFIUdJITon6fU++lbGpRCUcYrr1PrOsd7nIqmGK1QDAQTOQN+2p/Ld9I68EzmUvJ2XEvlQRrFyxjaGqPrJk9IrTR0YkSWOFT2ZMKOa2inGdTiRSFCV5ZL2oBxtzvVXfzMY96RF0C5vTpIEZ1mamW5uZLu8zydoLwAlTwFr7dJ7xXcwb9jRqzcSMLtVPFQKhWvLFbsjFwlkRatuGAq/FZy49JaEFOp1NJFIUJblktahX1TfzsUVvRow7SzYDOMHpsoOpVj1TpZ6pVj1TZAdF0gpAoxlCtX0ai32XU2WfRo2ZRCv5OzEpyGfczonBm2pQkKPngj706pYuF+hoMlRR0kfWinpVfTMPvLw5KYLuxc9oOcAk2cMk2e3+7zweJ41Y4rzHYVPEu2YCTwYuY4M9iUpzGjvMSHKl5DAZhI+dg64FOdEFOpoMVZT0kFDr3e7S49a7zdtAPKzffYxP/+FtjgUsfHjx4UUwFOCnEB+FBCjER39pYyhHGSpHGeL+P5SjjJSDlMoBRskBSuUAJRwOCTfAEdOfD8xotpnRbDUnU2tPZKOZQIMZTj4KeIFH8AeckMmnLpxI7e7DCM5g5+BVEdq98+6iTa8UpffkZ+vd/54FvmOcDawtAAp6dpqDZiC7zUnsNSdRa09kL8XsNiVss0dTZ0bTyDDyUbzDOamogFNHDuJbcxyRjiW6wdBKsKlWTwVZvXBFyR6yS9Svf5DVm3fxwlv1eAlQgJ8C938bCXntbe6/VlPAIQZy2AzkEAM5ZAZxiIH4suxjpRqv5TTKCthO3/If3nRmRMMsoFtDmhVFyV2yS/3Oup2vvfgKOwOTM21J1iPAhyYP54KykoihzhoCUZS+TVaJelV9MweOtmXajIwjwKkjBnJeWQlNLa2h7SMG96P85KExG2NBbG9cUZS+RdaIenYu/08+RQUWHksYOqCAioknsf9oG3OmlQKwomY3c6aVdgidKIqiJErWiHqwKVSuUzqkH1+84jSeeauBHQeOMbOshMmjBicUFlExVxSltyQk6iJyDfAg4AEeNcb8KNmGhNc7eyzBb5usmzVa4BH3f4sRg/rht20Qobx0CJ+59JQI0VaBVhQlEyQyo9QDPARcBTQA60TkeWPMu8k0JHyRy66Dx1m8dnsyT98pFlBU6KE1YGMbw5D+XiaPHMzkUYN7VeqnKIqSbhLx1M8Hthhj6gBE5EngRiCpog7tJXbB+ulgL+7eMqSfh0kjBjGrrITDrf5e12UriqJkK4mI+hhgR9jzBuCC6INEZB4wD2D8+N6FHsK99uKiwlB8+qZzxjC+ZCBPrdvOwWM+DhxtpX+Bh6JCL63+QLfi14qiKPlI0hKlxphFwCJw2gT09nzhC2Oi49Mar1YURYlNIj1jdwLjwp6PdbcpiqIoWUYior4OmCwik0SkEPgo8HxqzVIURVF6QpfhF2OMX0T+BfgLTknj74wxtSm3TFEURek2CcXUjTEvAi+m2BZFURSll+TfHDZFUZQ+jIq6oihKHpGSyUci0gjUd3LIcKAp6W+cetTu9KJ2pxe1O71E2z3BGDOitydNiah3+aYilckY25Ru1O70onanF7U7vaTKbg2/KIqi5BEq6oqiKHlEpkR9UYbet7eo3elF7U4vand6SYndGYmpK4qiKKlBwy+Koih5hIq6oihKHpF2UReRa0Rkk4hsEZH70/3+UbaME5FXReRdEakVkS+7208Skb+KyPvu/8XudhGR/3Jtf0dEpoed6273+PdF5O402e8RkbdEZLn7fJKIrHHte8ptwIaI9HOfb3H3Tww7x7fd7ZtE5MNpsHmYiCwRkfdEZKOIzMqF6y0iX3F/R2pEZLGI9M/G6y0ivxORfSJSE7YtaddXRGaIyAb3Nf8lIpJCu3/q/p68IyLPiMiwsH0xr2M8fYn3s0qF3WH7viYiRkSGu8/Tc72NMWn7h9MQbCtQBhQC64Gp6bQhyp5SYLr7eDCwGZgK/AS4391+P/Bj9/G1wApAgJnAGnf7SUCd+3+x+7g4DfZ/FXgCWO4+fxr4qPv4YeBz7uPPAw+7jz8KPOU+nur+DPoBk9yfjSfFNv8vcK/7uBAYlu3XG2dQzAfAgLDr/MlsvN7AJcB0oCZsW9KuL7DWPVbc185Jod1XA1738Y/D7I55HelEX+L9rFJht7t9HE4TxHpgeDqvd0pFJ8YFmAX8Jez5t4Fvp9OGLux7DmcW6yag1N1WCmxyHz8CfCzs+E3u/o8Bj4RtjzguRbaOBV4BLgeWuz/0prA/gtC1dn+5ZrmPve5xEn39w49Lkc1DccRRorZn9fWmffrXSe71Ww58OFuvNzCRSHFMyvV1970Xtj3iuGTbHbXvZuBx93HM60gcfensbyNVdgNLgLOBbbSLelqud7rDL7FG441Jsw0xcb8inwusAUYZY3a7u/YAo9zH8ezPxOd6APgmhMa4lgAHjTH+GDaE7HP3H3KPT7fdk4BG4H/ECRs9KiIDyfLrbYzZCfwM2A7sxrl+VWT/9Q6SrOs7xn0cvT0dfArHU4Xu293Z30bSEZEbgZ3GmPVRu9JyvTVRCojIIGApcJ8x5nD4PuPcIrOq7lNE5gL7jDFVmbalm3hxvqr+2hhzLnAUJxwQIkuvdzHOsPVJwMnAQOCajBrVQ7Lx+naFiHwX8AOPZ9qWrhCRIuA7wPxM2ZBuUc+60XgiUoAj6I8bY5a5m/eKSKm7vxTY526PZ3+6P9dFwA0isg14EicE8yAwTESCPfLDbQjZ5+4fCuzPgN0NQIMxZo37fAmOyGf79b4S+MAY02iM8QHLcH4G2X69gyTr+u50H0dvTxki8klgLnCXe0OiC/tibd9P/J9VsjkF5+a/3v37HAtUi8joHtjds+ud7HheF7EnL04SYBLtiYzydNoQZY8AvwceiNr+UyITSz9xH19HZKJjrbv9JJxYcbH77wPgpDR9htm0J0r/SGQy6PPu4y8Qmbh72n1cTmTCqY7UJ0pXAVPcxwvca53V1xu4AKgFilxb/hf4YrZebzrG1JN2femYuLs2hXZfA7wLjIg6LuZ1pBN9ifezSoXdUfu20R5TT8v1TtkfbycX4FqcKpOtwHfT/f5RtlyM81X0HeBt99+1ODG4V4D3gZfDLrAAD7m2bwAqws71KWCL+++f0/gZZtMu6mXuL8EW95e4n7u9v/t8i7u/LOz133U/zyaSVMnQhb3nAJXuNX/W/SXO+usN/CvwHlAD/J8rKFl3vYHFOHF/H843o3uSeX2BCvcabAV+RVTSO8l2b8GJNQf/Nh/u6joSR1/i/axSYXfU/m20i3parre2CVAURckjNFGqKIqSR6ioK4qi5BEq6oqiKHmEirqiKEoeoaKuKIqSR6ioK4qi5BEq6oqiKHnE/wcVzp9BfnDQ4gAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(poly3_data['power_1'],poly3_data['price'],'.',\n",
    "         poly3_data['power_1'], model3.predict(poly3_data),'-')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now try a 15th degree polynomial:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre>Linear regression:</pre>"
      ],
      "text/plain": [
       "Linear regression:"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of examples          : 21613</pre>"
      ],
      "text/plain": [
       "Number of examples          : 21613"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of features          : 15</pre>"
      ],
      "text/plain": [
       "Number of features          : 15"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of unpacked features : 15</pre>"
      ],
      "text/plain": [
       "Number of unpacked features : 15"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of coefficients    : 16</pre>"
      ],
      "text/plain": [
       "Number of coefficients    : 16"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Starting Newton Method</pre>"
      ],
      "text/plain": [
       "Starting Newton Method"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |</pre>"
      ],
      "text/plain": [
       "| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| 1         | 2        | 0.014611     | 2662308.584335     | 245690.511190                   |</pre>"
      ],
      "text/plain": [
       "| 1         | 2        | 0.014611     | 2662308.584335     | 245690.511190                   |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>SUCCESS: Optimal solution found.</pre>"
      ],
      "text/plain": [
       "SUCCESS: Optimal solution found."
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre></pre>"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "poly15_data = polynomial_sframe(sales['sqft_living'], 15)\n",
    "my_features = poly15_data.column_names() # get the name of the features\n",
    "poly15_data['price'] = sales['price'] # add price to the data since it's the target\n",
    "model15 = turicreate.linear_regression.create(poly15_data, target = 'price', features = my_features, validation_set = None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x123117fd0>,\n",
       " <matplotlib.lines.Line2D at 0x12312a250>]"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(poly15_data['power_1'],poly15_data['price'],'.',\n",
    "         poly15_data['power_1'], model15.predict(poly15_data),'-')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\"><table frame=\"box\" rules=\"cols\">\n",
       "    <tr>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">name</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">index</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">value</th>\n",
       "        <th style=\"padding-left: 1em; padding-right: 1em; text-align: center\">stderr</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">(intercept)</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">None</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">73619.75210628664</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">593312.434331862</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">power_1</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">None</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">410.28746254848676</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">2949.9428791280775</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">power_2</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">None</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">-0.2304507144377397</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">6.08058877191724</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">power_3</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">None</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">7.588405424690668e-05</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">0.006874881623187978</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">power_4</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">None</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">-5.6570180260606194e-09</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">4.760370835953835e-06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">power_5</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">None</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">-4.570281308521954e-13</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">2.1306740836434974e-09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">power_6</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">None</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">2.6636020689135647e-17</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">6.314322654361962e-13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">power_7</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">None</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">3.38584768858826e-21</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1.2390026990918032e-16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">power_8</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">None</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1.1472310435298844e-25</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1.5616256434209065e-20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">power_9</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">None</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">-4.652935864361304e-30</td>\n",
       "        <td style=\"padding-left: 1em; padding-right: 1em; text-align: center; vertical-align: top\">1.1355498618150417e-24</td>\n",
       "    </tr>\n",
       "</table>\n",
       "[16 rows x 4 columns]<br/>Note: Only the head of the SFrame is printed.<br/>You can use print_rows(num_rows=m, num_columns=n) to print more rows and columns.\n",
       "</div>"
      ],
      "text/plain": [
       "Columns:\n",
       "\tname\tstr\n",
       "\tindex\tstr\n",
       "\tvalue\tfloat\n",
       "\tstderr\tfloat\n",
       "\n",
       "Rows: 16\n",
       "\n",
       "Data:\n",
       "+-------------+-------+-------------------------+------------------------+\n",
       "|     name    | index |          value          |         stderr         |\n",
       "+-------------+-------+-------------------------+------------------------+\n",
       "| (intercept) |  None |    73619.75210628664    |    593312.434331862    |\n",
       "|   power_1   |  None |    410.28746254848676   |   2949.9428791280775   |\n",
       "|   power_2   |  None |   -0.2304507144377397   |    6.08058877191724    |\n",
       "|   power_3   |  None |  7.588405424690668e-05  |  0.006874881623187978  |\n",
       "|   power_4   |  None | -5.6570180260606194e-09 | 4.760370835953835e-06  |\n",
       "|   power_5   |  None |  -4.570281308521954e-13 | 2.1306740836434974e-09 |\n",
       "|   power_6   |  None |  2.6636020689135647e-17 | 6.314322654361962e-13  |\n",
       "|   power_7   |  None |   3.38584768858826e-21  | 1.2390026990918032e-16 |\n",
       "|   power_8   |  None |  1.1472310435298844e-25 | 1.5616256434209065e-20 |\n",
       "|   power_9   |  None |  -4.652935864361304e-30 | 1.1355498618150417e-24 |\n",
       "+-------------+-------+-------------------------+------------------------+\n",
       "[16 rows x 4 columns]\n",
       "Note: Only the head of the SFrame is printed.\n",
       "You can use print_rows(num_rows=m, num_columns=n) to print more rows and columns."
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model15.coefficients"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "What do you think of the 15th degree polynomial? Do you think this is appropriate? If we were to change the data do you think you'd get pretty much the same curve? Let's take a look."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## A\n",
    "1 obvious outlier"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Changing the data and re-learning"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We're going to split the sales data into four subsets of roughly equal size. Then you will estimate a 15th degree polynomial model on all four subsets of the data. Print the coefficients (you should use .print_rows(num_rows = 16) to view all of them) and plot the resulting fit (as we did above). The quiz will ask you some questions about these results.\n",
    "\n",
    "To split the sales data into four subsets, we perform the following steps:\n",
    "* First split sales into 2 subsets with `.random_split(0.5, seed=0)`. \n",
    "* Next split the resulting subsets into 2 more subsets each. Use `.random_split(0.5, seed=0)`.\n",
    "\n",
    "We set `seed=0` in these steps so that different users get consistent results.\n",
    "You should end up with 4 subsets (`set_1`, `set_2`, `set_3`, `set_4`) of approximately equal size. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "ss1,ss2 = sales.random_split(.5,seed=0)\n",
    "set_1,set_2 = ss1.random_split(.5,seed=0)\n",
    "set_3,set_4 = ss2.random_split(.5,seed=0)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Fit a 15th degree polynomial on set_1, set_2, set_3, and set_4 using sqft_living to predict prices. Print the coefficients and make a plot of the resulting model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre>Linear regression:</pre>"
      ],
      "text/plain": [
       "Linear regression:"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of examples          : 5404</pre>"
      ],
      "text/plain": [
       "Number of examples          : 5404"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of features          : 15</pre>"
      ],
      "text/plain": [
       "Number of features          : 15"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of unpacked features : 15</pre>"
      ],
      "text/plain": [
       "Number of unpacked features : 15"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of coefficients    : 16</pre>"
      ],
      "text/plain": [
       "Number of coefficients    : 16"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Starting Newton Method</pre>"
      ],
      "text/plain": [
       "Starting Newton Method"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |</pre>"
      ],
      "text/plain": [
       "| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| 1         | 2        | 0.036502     | 2195218.932306     | 248858.822200                   |</pre>"
      ],
      "text/plain": [
       "| 1         | 2        | 0.036502     | 2195218.932306     | 248858.822200                   |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>SUCCESS: Optimal solution found.</pre>"
      ],
      "text/plain": [
       "SUCCESS: Optimal solution found."
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre></pre>"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "poly15_data_1 = polynomial_sframe(set_1['sqft_living'], 15)\n",
    "my_features_1 = poly15_data_1.column_names() # get the name of the features\n",
    "poly15_data_1['price'] = set_1['price'] # add price to the data since it's the target\n",
    "model15_1 = turicreate.linear_regression.create(poly15_data_1, target = 'price', features = my_features_1, validation_set = None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre>Linear regression:</pre>"
      ],
      "text/plain": [
       "Linear regression:"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of examples          : 5398</pre>"
      ],
      "text/plain": [
       "Number of examples          : 5398"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of features          : 15</pre>"
      ],
      "text/plain": [
       "Number of features          : 15"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of unpacked features : 15</pre>"
      ],
      "text/plain": [
       "Number of unpacked features : 15"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of coefficients    : 16</pre>"
      ],
      "text/plain": [
       "Number of coefficients    : 16"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Starting Newton Method</pre>"
      ],
      "text/plain": [
       "Starting Newton Method"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |</pre>"
      ],
      "text/plain": [
       "| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| 1         | 2        | 0.070043     | 2069212.978547     | 234840.067186                   |</pre>"
      ],
      "text/plain": [
       "| 1         | 2        | 0.070043     | 2069212.978547     | 234840.067186                   |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>SUCCESS: Optimal solution found.</pre>"
      ],
      "text/plain": [
       "SUCCESS: Optimal solution found."
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre></pre>"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "poly15_data_2 = polynomial_sframe(set_2['sqft_living'], 15)\n",
    "my_features_2 = poly15_data_2.column_names() # get the name of the features\n",
    "poly15_data_2['price'] = set_2['price'] # add price to the data since it's the target\n",
    "model15_2 = turicreate.linear_regression.create(poly15_data_2, target = 'price', features = my_features_2, validation_set = None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre>Linear regression:</pre>"
      ],
      "text/plain": [
       "Linear regression:"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of examples          : 5409</pre>"
      ],
      "text/plain": [
       "Number of examples          : 5409"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of features          : 15</pre>"
      ],
      "text/plain": [
       "Number of features          : 15"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of unpacked features : 15</pre>"
      ],
      "text/plain": [
       "Number of unpacked features : 15"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of coefficients    : 16</pre>"
      ],
      "text/plain": [
       "Number of coefficients    : 16"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Starting Newton Method</pre>"
      ],
      "text/plain": [
       "Starting Newton Method"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |</pre>"
      ],
      "text/plain": [
       "| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| 1         | 2        | 0.026227     | 2269769.506521     | 251460.072754                   |</pre>"
      ],
      "text/plain": [
       "| 1         | 2        | 0.026227     | 2269769.506521     | 251460.072754                   |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>SUCCESS: Optimal solution found.</pre>"
      ],
      "text/plain": [
       "SUCCESS: Optimal solution found."
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre></pre>"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "poly15_data_3 = polynomial_sframe(set_3['sqft_living'], 15)\n",
    "my_features_3 = poly15_data_3.column_names() # get the name of the features\n",
    "poly15_data_3['price'] = set_3['price'] # add price to the data since it's the target\n",
    "model15_3 = turicreate.linear_regression.create(poly15_data_3, target = 'price', features = my_features_3, validation_set = None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre>Linear regression:</pre>"
      ],
      "text/plain": [
       "Linear regression:"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of examples          : 5402</pre>"
      ],
      "text/plain": [
       "Number of examples          : 5402"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of features          : 15</pre>"
      ],
      "text/plain": [
       "Number of features          : 15"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of unpacked features : 15</pre>"
      ],
      "text/plain": [
       "Number of unpacked features : 15"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of coefficients    : 16</pre>"
      ],
      "text/plain": [
       "Number of coefficients    : 16"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Starting Newton Method</pre>"
      ],
      "text/plain": [
       "Starting Newton Method"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |</pre>"
      ],
      "text/plain": [
       "| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| 1         | 2        | 0.029828     | 2314893.173820     | 244563.136754                   |</pre>"
      ],
      "text/plain": [
       "| 1         | 2        | 0.029828     | 2314893.173820     | 244563.136754                   |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>SUCCESS: Optimal solution found.</pre>"
      ],
      "text/plain": [
       "SUCCESS: Optimal solution found."
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre></pre>"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "poly15_data_4 = polynomial_sframe(set_4['sqft_living'], 15)\n",
    "my_features_4 = poly15_data_4.column_names() # get the name of the features\n",
    "poly15_data_4['price'] = set_4['price'] # add price to the data since it's the target\n",
    "model15_4 = turicreate.linear_regression.create(poly15_data_4, target = 'price', features = my_features_4, validation_set = None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "+-------------+-------+-------------------------+------------------------+\n",
      "|     name    | index |          value          |         stderr         |\n",
      "+-------------+-------+-------------------------+------------------------+\n",
      "| (intercept) |  None |    223312.75025161862   |    1258662.3010632     |\n",
      "|   power_1   |  None |    118.08612758116162   |   5686.404018565508    |\n",
      "|   power_2   |  None |   -0.04734820112974076  |   10.534620637936902   |\n",
      "|   power_3   |  None |  3.253103424545136e-05  |  0.010581059901433858  |\n",
      "|   power_4   |  None | -3.3237215254879214e-09 | 6.414049437103749e-06  |\n",
      "|   power_5   |  None |  -9.758304575852731e-14 | 2.4645041242327048e-09 |\n",
      "|   power_6   |  None |  1.1544030339104844e-17 |  6.10749213904811e-13  |\n",
      "|   power_7   |  None |  1.0514586941801688e-21 | 9.718727089652212e-17  |\n",
      "|   power_8   |  None |  3.460496164266092e-26  | 1.0040117095444741e-20 |\n",
      "|   power_9   |  None | -1.0965445392318163e-30 |  9.1048764883969e-25   |\n",
      "|   power_10  |  None | -2.4203181209029504e-34 | 1.0932032428897115e-28 |\n",
      "|   power_11  |  None | -1.9960120687077438e-38 | 1.0960500556968315e-32 |\n",
      "|   power_12  |  None | -1.0770990389195788e-42 |  7.21942686419834e-37  |\n",
      "|   power_13  |  None | -2.7286281749405743e-47 | 3.4338546571724185e-41 |\n",
      "|   power_14  |  None |  2.447826934034351e-51  | 1.6000215116540342e-45 |\n",
      "|   power_15  |  None |  5.019752326753699e-55  | 3.662366553408393e-50  |\n",
      "+-------------+-------+-------------------------+------------------------+\n",
      "[16 rows x 4 columns]\n",
      "\n",
      "+-------------+-------+-------------------------+------------------------+\n",
      "|     name    | index |          value          |         stderr         |\n",
      "+-------------+-------+-------------------------+------------------------+\n",
      "| (intercept) |  None |    89836.50773595303    |          nan           |\n",
      "|   power_1   |  None |     319.806946756886    |          nan           |\n",
      "|   power_2   |  None |   -0.10331539703578418  |          nan           |\n",
      "|   power_3   |  None |  1.0668247604826077e-05 |          nan           |\n",
      "|   power_4   |  None |  5.7557709774540684e-09 |          nan           |\n",
      "|   power_5   |  None |  -2.546634647386611e-13 |          nan           |\n",
      "|   power_6   |  None | -1.0964134507286332e-16 |          nan           |\n",
      "|   power_7   |  None |  -6.364584415492852e-21 |          nan           |\n",
      "|   power_8   |  None |   5.52560416854716e-25  |          nan           |\n",
      "|   power_9   |  None |  1.3508203898197794e-28 | 1.5028169008795258e-23 |\n",
      "|   power_10  |  None |  1.1840818822582313e-32 | 3.0388810230834194e-27 |\n",
      "|   power_11  |  None |   1.9834800045379e-37   |          nan           |\n",
      "|   power_12  |  None |  -9.925335900144669e-41 |          nan           |\n",
      "|   power_13  |  None | -1.6083484705056515e-44 |          nan           |\n",
      "|   power_14  |  None |  -9.120060245117016e-49 |          nan           |\n",
      "|   power_15  |  None |  1.6863665833555978e-52 |          nan           |\n",
      "+-------------+-------+-------------------------+------------------------+\n",
      "[16 rows x 4 columns]\n",
      "\n",
      "+-------------+-------+-------------------------+------------------------+\n",
      "|     name    | index |          value          |         stderr         |\n",
      "+-------------+-------+-------------------------+------------------------+\n",
      "| (intercept) |  None |     87317.9795623491    |   4198773.372675237    |\n",
      "|   power_1   |  None |    356.3049110285302    |   22695.50682332979    |\n",
      "|   power_2   |  None |   -0.16481744279635122  |   50.723659288791815   |\n",
      "|   power_3   |  None |  4.4042499265816884e-05 |   0.0620939361703543   |\n",
      "|   power_4   |  None |  6.482348764851361e-10  | 4.6350367438038446e-05 |\n",
      "|   power_5   |  None |  -6.752532265310062e-13 | 2.2089835159786524e-08 |\n",
      "|   power_6   |  None | -3.3684259272719256e-17 |  6.75795389917903e-12  |\n",
      "|   power_7   |  None |  3.609997041976729e-21  | 1.2608093541095756e-15 |\n",
      "|   power_8   |  None |  6.469997256308929e-25  | 1.1093889544447856e-19 |\n",
      "|   power_9   |  None |  4.236393888725579e-29  |          nan           |\n",
      "|   power_10  |  None | -3.6214942529140553e-34 | 1.9288193386355234e-27 |\n",
      "|   power_11  |  None |  -4.271195274189792e-37 | 1.4347615943175153e-31 |\n",
      "|   power_12  |  None |  -5.614459716503116e-41 | 1.1574135876180139e-35 |\n",
      "|   power_13  |  None | -3.8745277299646824e-45 | 2.170011494582633e-39  |\n",
      "|   power_14  |  None |  4.694303605050596e-50  | 1.2866570614406508e-43 |\n",
      "|   power_15  |  None |  6.390458859794287e-53  | 2.7649163956545246e-48 |\n",
      "+-------------+-------+-------------------------+------------------------+\n",
      "[16 rows x 4 columns]\n",
      "\n",
      "+-------------+-------+-------------------------+------------------------+\n",
      "|     name    | index |          value          |         stderr         |\n",
      "+-------------+-------+-------------------------+------------------------+\n",
      "| (intercept) |  None |    259020.87943774887   |   2010537.9579847155   |\n",
      "|   power_1   |  None |   -31.727716169103356   |   13559.203733845043   |\n",
      "|   power_2   |  None |   0.10970276958759548   |   37.66697929983927    |\n",
      "|   power_3   |  None | -1.5838384723249074e-05 |  0.05754484704892946   |\n",
      "|   power_4   |  None |  -4.476606238930388e-09 | 5.436487704570491e-05  |\n",
      "|   power_5   |  None |  1.1397657347328584e-12 | 3.3750130497803066e-08 |\n",
      "|   power_6   |  None |  1.9766912053871528e-16 | 1.4242154071029366e-11 |\n",
      "|   power_7   |  None | -6.1578367781481515e-21 | 4.166161047857958e-15  |\n",
      "|   power_8   |  None |  -4.880123041178975e-24 |  8.56799685470776e-19  |\n",
      "|   power_9   |  None |  -6.621867814406893e-28 | 1.2501959511372632e-22 |\n",
      "|   power_10  |  None |  -2.706315832321753e-32 | 1.2424348669648102e-26 |\n",
      "|   power_11  |  None |  6.723704116682092e-36  | 8.548073571895184e-31  |\n",
      "|   power_12  |  None |  1.7411564630563372e-39 | 1.5358343902014186e-34 |\n",
      "|   power_13  |  None |  2.091883756704146e-43  | 1.9797891129483579e-38 |\n",
      "|   power_14  |  None |  4.780155658059687e-48  | 1.2298424771111609e-42 |\n",
      "|   power_15  |  None | -4.7453533305902726e-51 | 3.060319998871967e-47  |\n",
      "+-------------+-------+-------------------------+------------------------+\n",
      "[16 rows x 4 columns]\n",
      "\n"
     ]
    }
   ],
   "source": [
    "model15_1.coefficients.print_rows(16)\n",
    "model15_2.coefficients.print_rows(16)\n",
    "model15_3.coefficients.print_rows(16)\n",
    "model15_4.coefficients.print_rows(16)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x1282136d0>,\n",
       " <matplotlib.lines.Line2D at 0x128213990>]"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(poly15_data_1['power_1'],poly15_data_1['price'],'.',\n",
    "         poly15_data_1['power_1'], model15_1.predict(poly15_data_1),'-')\n",
    "plt.plot(poly15_data_2['power_1'],poly15_data_2['price'],'.',\n",
    "         poly15_data_2['power_1'], model15_2.predict(poly15_data_2),'-')\n",
    "plt.plot(poly15_data_3['power_1'],poly15_data_3['price'],'.',\n",
    "         poly15_data_3['power_1'], model15_3.predict(poly15_data_3),'-')\n",
    "plt.plot(poly15_data_4['power_1'],poly15_data_4['price'],'.',\n",
    "         poly15_data_4['power_1'], model15_4.predict(poly15_data_4),'-')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Some questions you will be asked on your quiz:\n",
    "\n",
    "**Quiz Question: Is the sign (positive or negative) for power_15 the same in all four models?**\n",
    "\n",
    "no, all over the place, but really small for all\n",
    "\n",
    "**Quiz Question: (True/False) the plotted fitted lines look the same in all four plots**\n",
    "\n",
    "false"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "actual= 165000.0 [292215.0328569543]\n"
     ]
    }
   ],
   "source": [
    "i = 101\n",
    "print(\"actual=\", poly15_data_4[i]['price'], model15_4.predict(poly15_data_4[i]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Selecting a Polynomial Degree"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Whenever we have a \"magic\" parameter like the degree of the polynomial there is one well-known way to select these parameters: validation set. (We will explore another approach in week 4).\n",
    "\n",
    "We split the sales dataset 3-way into training set, test set, and validation set as follows:\n",
    "\n",
    "* Split our sales data into 2 sets: `training_and_validation` and `testing`. Use `random_split(0.9, seed=1)`.\n",
    "* Further split our training data into two sets: `training` and `validation`. Use `random_split(0.5, seed=1)`.\n",
    "\n",
    "Again, we set `seed=1` to obtain consistent results for different users."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {},
   "outputs": [],
   "source": [
    "training_and_validation,testing = sales.random_split(0.9, seed=1)\n",
    "training,validation = training_and_validation.random_split(0.5, seed=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [],
   "source": [
    "#training.explore()\n",
    "#validation.explore()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Next you should write a loop that does the following:\n",
    "* For degree in [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15] (to get this in python type range(1, 15+1))\n",
    "    * Build an SFrame of polynomial data of train_data['sqft_living'] at the current degree\n",
    "    * hint: my_features = poly_data.column_names() gives you a list e.g. ['power_1', 'power_2', 'power_3'] which you might find useful for turicreate.linear_regression.create( features = my_features)\n",
    "    * Add train_data['price'] to the polynomial SFrame\n",
    "    * Learn a polynomial regression model to sqft vs price with that degree on TRAIN data\n",
    "    * Compute the RSS on VALIDATION data (here you will want to use .predict()) for that degree and you will need to make a polynmial SFrame using validation data.\n",
    "* Report which degree had the lowest RSS on validation data (remember python indexes from 0)\n",
    "\n",
    "(Note you can turn off the print out of linear_regression.create() with verbose = False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre>Linear regression:</pre>"
      ],
      "text/plain": [
       "Linear regression:"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of examples          : 9761</pre>"
      ],
      "text/plain": [
       "Number of examples          : 9761"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of features          : 1</pre>"
      ],
      "text/plain": [
       "Number of features          : 1"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of unpacked features : 1</pre>"
      ],
      "text/plain": [
       "Number of unpacked features : 1"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of coefficients    : 2</pre>"
      ],
      "text/plain": [
       "Number of coefficients    : 2"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Starting Newton Method</pre>"
      ],
      "text/plain": [
       "Starting Newton Method"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |</pre>"
      ],
      "text/plain": [
       "| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| 1         | 2        | 0.015507     | 4274505.747987     | 262315.114947                   |</pre>"
      ],
      "text/plain": [
       "| 1         | 2        | 0.015507     | 4274505.747987     | 262315.114947                   |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>SUCCESS: Optimal solution found.</pre>"
      ],
      "text/plain": [
       "SUCCESS: Optimal solution found."
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre></pre>"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "676709775198047.5 -- 1\n",
      "129030470094700.17 -- 1\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<pre>Linear regression:</pre>"
      ],
      "text/plain": [
       "Linear regression:"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of examples          : 9761</pre>"
      ],
      "text/plain": [
       "Number of examples          : 9761"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of features          : 2</pre>"
      ],
      "text/plain": [
       "Number of features          : 2"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of unpacked features : 2</pre>"
      ],
      "text/plain": [
       "Number of unpacked features : 2"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of coefficients    : 3</pre>"
      ],
      "text/plain": [
       "Number of coefficients    : 3"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Starting Newton Method</pre>"
      ],
      "text/plain": [
       "Starting Newton Method"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |</pre>"
      ],
      "text/plain": [
       "| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| 1         | 2        | 0.020515     | 4869005.244131     | 255076.149120                   |</pre>"
      ],
      "text/plain": [
       "| 1         | 2        | 0.020515     | 4869005.244131     | 255076.149120                   |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>SUCCESS: Optimal solution found.</pre>"
      ],
      "text/plain": [
       "SUCCESS: Optimal solution found."
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre></pre>"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "607090530698014.1 -- 2\n",
      "125190117212860.06 -- 2\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<pre>Linear regression:</pre>"
      ],
      "text/plain": [
       "Linear regression:"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of examples          : 9761</pre>"
      ],
      "text/plain": [
       "Number of examples          : 9761"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of features          : 3</pre>"
      ],
      "text/plain": [
       "Number of features          : 3"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of unpacked features : 3</pre>"
      ],
      "text/plain": [
       "Number of unpacked features : 3"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of coefficients    : 4</pre>"
      ],
      "text/plain": [
       "Number of coefficients    : 4"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Starting Newton Method</pre>"
      ],
      "text/plain": [
       "Starting Newton Method"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |</pre>"
      ],
      "text/plain": [
       "| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| 1         | 2        | 0.032988     | 3271232.649557     | 249640.623557                   |</pre>"
      ],
      "text/plain": [
       "| 1         | 2        | 0.032988     | 3271232.649557     | 249640.623557                   |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>SUCCESS: Optimal solution found.</pre>"
      ],
      "text/plain": [
       "SUCCESS: Optimal solution found."
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre></pre>"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "616714574532761.8 -- 3\n",
      "125939457007406.9 -- 3\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<pre>Linear regression:</pre>"
      ],
      "text/plain": [
       "Linear regression:"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of examples          : 9761</pre>"
      ],
      "text/plain": [
       "Number of examples          : 9761"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of features          : 4</pre>"
      ],
      "text/plain": [
       "Number of features          : 4"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of unpacked features : 4</pre>"
      ],
      "text/plain": [
       "Number of unpacked features : 4"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of coefficients    : 5</pre>"
      ],
      "text/plain": [
       "Number of coefficients    : 5"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Starting Newton Method</pre>"
      ],
      "text/plain": [
       "Starting Newton Method"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |</pre>"
      ],
      "text/plain": [
       "| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| 1         | 2        | 0.046243     | 2676547.198434     | 248689.572032                   |</pre>"
      ],
      "text/plain": [
       "| 1         | 2        | 0.046243     | 2676547.198434     | 248689.572032                   |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>SUCCESS: Optimal solution found.</pre>"
      ],
      "text/plain": [
       "SUCCESS: Optimal solution found."
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre></pre>"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "609129230654383.5 -- 4\n",
      "127147719016434.84 -- 4\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<pre>Linear regression:</pre>"
      ],
      "text/plain": [
       "Linear regression:"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of examples          : 9761</pre>"
      ],
      "text/plain": [
       "Number of examples          : 9761"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of features          : 5</pre>"
      ],
      "text/plain": [
       "Number of features          : 5"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of unpacked features : 5</pre>"
      ],
      "text/plain": [
       "Number of unpacked features : 5"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of coefficients    : 6</pre>"
      ],
      "text/plain": [
       "Number of coefficients    : 6"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Starting Newton Method</pre>"
      ],
      "text/plain": [
       "Starting Newton Method"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |</pre>"
      ],
      "text/plain": [
       "| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| 1         | 2        | 0.046820     | 2330678.377343     | 248281.665797                   |</pre>"
      ],
      "text/plain": [
       "| 1         | 2        | 0.046820     | 2330678.377343     | 248281.665797                   |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>SUCCESS: Optimal solution found.</pre>"
      ],
      "text/plain": [
       "SUCCESS: Optimal solution found."
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre></pre>"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "599177138583675.6 -- 5\n",
      "127037730210195.31 -- 5\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<pre>Linear regression:</pre>"
      ],
      "text/plain": [
       "Linear regression:"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of examples          : 9761</pre>"
      ],
      "text/plain": [
       "Number of examples          : 9761"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of features          : 6</pre>"
      ],
      "text/plain": [
       "Number of features          : 6"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of unpacked features : 6</pre>"
      ],
      "text/plain": [
       "Number of unpacked features : 6"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of coefficients    : 7</pre>"
      ],
      "text/plain": [
       "Number of coefficients    : 7"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Starting Newton Method</pre>"
      ],
      "text/plain": [
       "Starting Newton Method"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |</pre>"
      ],
      "text/plain": [
       "| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| 1         | 2        | 0.055990     | 2344070.143280     | 247280.891725                   |</pre>"
      ],
      "text/plain": [
       "| 1         | 2        | 0.055990     | 2344070.143280     | 247280.891725                   |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>SUCCESS: Optimal solution found.</pre>"
      ],
      "text/plain": [
       "SUCCESS: Optimal solution found."
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre></pre>"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "589182477809739.2 -- 6\n",
      "125529337848156.22 -- 6\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<pre>Linear regression:</pre>"
      ],
      "text/plain": [
       "Linear regression:"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of examples          : 9761</pre>"
      ],
      "text/plain": [
       "Number of examples          : 9761"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of features          : 7</pre>"
      ],
      "text/plain": [
       "Number of features          : 7"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of unpacked features : 7</pre>"
      ],
      "text/plain": [
       "Number of unpacked features : 7"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of coefficients    : 8</pre>"
      ],
      "text/plain": [
       "Number of coefficients    : 8"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Starting Newton Method</pre>"
      ],
      "text/plain": [
       "Starting Newton Method"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |</pre>"
      ],
      "text/plain": [
       "| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| 1         | 2        | 0.066924     | 2452711.310608     | 246772.313122                   |</pre>"
      ],
      "text/plain": [
       "| 1         | 2        | 0.066924     | 2452711.310608     | 246772.313122                   |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>SUCCESS: Optimal solution found.</pre>"
      ],
      "text/plain": [
       "SUCCESS: Optimal solution found."
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre></pre>"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "591717038418095.9 -- 7\n",
      "124608773610110.22 -- 7\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<pre>Linear regression:</pre>"
      ],
      "text/plain": [
       "Linear regression:"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of examples          : 9761</pre>"
      ],
      "text/plain": [
       "Number of examples          : 9761"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of features          : 8</pre>"
      ],
      "text/plain": [
       "Number of features          : 8"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of unpacked features : 8</pre>"
      ],
      "text/plain": [
       "Number of unpacked features : 8"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of coefficients    : 9</pre>"
      ],
      "text/plain": [
       "Number of coefficients    : 9"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Starting Newton Method</pre>"
      ],
      "text/plain": [
       "Starting Newton Method"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |</pre>"
      ],
      "text/plain": [
       "| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| 1         | 2        | 0.051593     | 2504989.234017     | 246671.859042                   |</pre>"
      ],
      "text/plain": [
       "| 1         | 2        | 0.051593     | 2504989.234017     | 246671.859042                   |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>SUCCESS: Optimal solution found.</pre>"
      ],
      "text/plain": [
       "SUCCESS: Optimal solution found."
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre></pre>"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "601558237778332.8 -- 8\n",
      "124458964534463.83 -- 8\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<pre>Linear regression:</pre>"
      ],
      "text/plain": [
       "Linear regression:"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of examples          : 9761</pre>"
      ],
      "text/plain": [
       "Number of examples          : 9761"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of features          : 9</pre>"
      ],
      "text/plain": [
       "Number of features          : 9"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of unpacked features : 9</pre>"
      ],
      "text/plain": [
       "Number of unpacked features : 9"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of coefficients    : 10</pre>"
      ],
      "text/plain": [
       "Number of coefficients    : 10"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Starting Newton Method</pre>"
      ],
      "text/plain": [
       "Starting Newton Method"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |</pre>"
      ],
      "text/plain": [
       "| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| 1         | 2        | 0.053790     | 2525802.177255     | 246663.399621                   |</pre>"
      ],
      "text/plain": [
       "| 1         | 2        | 0.053790     | 2525802.177255     | 246663.399621                   |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>SUCCESS: Optimal solution found.</pre>"
      ],
      "text/plain": [
       "SUCCESS: Optimal solution found."
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre></pre>"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "612563853986526.5 -- 9\n",
      "124525299776505.03 -- 9\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<pre>Linear regression:</pre>"
      ],
      "text/plain": [
       "Linear regression:"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of examples          : 9761</pre>"
      ],
      "text/plain": [
       "Number of examples          : 9761"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of features          : 10</pre>"
      ],
      "text/plain": [
       "Number of features          : 10"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of unpacked features : 10</pre>"
      ],
      "text/plain": [
       "Number of unpacked features : 10"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of coefficients    : 11</pre>"
      ],
      "text/plain": [
       "Number of coefficients    : 11"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Starting Newton Method</pre>"
      ],
      "text/plain": [
       "Starting Newton Method"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |</pre>"
      ],
      "text/plain": [
       "| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| 1         | 2        | 0.058558     | 2532693.511975     | 246670.636994                   |</pre>"
      ],
      "text/plain": [
       "| 1         | 2        | 0.058558     | 2532693.511975     | 246670.636994                   |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>SUCCESS: Optimal solution found.</pre>"
      ],
      "text/plain": [
       "SUCCESS: Optimal solution found."
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre></pre>"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "621744288936945.1 -- 10\n",
      "124602928485231.25 -- 10\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<pre>Linear regression:</pre>"
      ],
      "text/plain": [
       "Linear regression:"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of examples          : 9761</pre>"
      ],
      "text/plain": [
       "Number of examples          : 9761"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of features          : 11</pre>"
      ],
      "text/plain": [
       "Number of features          : 11"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of unpacked features : 11</pre>"
      ],
      "text/plain": [
       "Number of unpacked features : 11"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of coefficients    : 12</pre>"
      ],
      "text/plain": [
       "Number of coefficients    : 12"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Starting Newton Method</pre>"
      ],
      "text/plain": [
       "Starting Newton Method"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |</pre>"
      ],
      "text/plain": [
       "| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| 1         | 2        | 0.076369     | 2534201.088396     | 246675.476971                   |</pre>"
      ],
      "text/plain": [
       "| 1         | 2        | 0.076369     | 2534201.088396     | 246675.476971                   |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>SUCCESS: Optimal solution found.</pre>"
      ],
      "text/plain": [
       "SUCCESS: Optimal solution found."
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre></pre>"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "627012012706161.9 -- 11\n",
      "124641733743998.38 -- 11\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<pre>Linear regression:</pre>"
      ],
      "text/plain": [
       "Linear regression:"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of examples          : 9761</pre>"
      ],
      "text/plain": [
       "Number of examples          : 9761"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of features          : 12</pre>"
      ],
      "text/plain": [
       "Number of features          : 12"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of unpacked features : 12</pre>"
      ],
      "text/plain": [
       "Number of unpacked features : 12"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of coefficients    : 13</pre>"
      ],
      "text/plain": [
       "Number of coefficients    : 13"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Starting Newton Method</pre>"
      ],
      "text/plain": [
       "Starting Newton Method"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |</pre>"
      ],
      "text/plain": [
       "| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| 1         | 2        | 0.075822     | 2534257.195240     | 246676.033273                   |</pre>"
      ],
      "text/plain": [
       "| 1         | 2        | 0.075822     | 2534257.195240     | 246676.033273                   |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>SUCCESS: Optimal solution found.</pre>"
      ],
      "text/plain": [
       "SUCCESS: Optimal solution found."
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre></pre>"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "627757914773691.8 -- 12\n",
      "124646108317116.19 -- 12\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<pre>Linear regression:</pre>"
      ],
      "text/plain": [
       "Linear regression:"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of examples          : 9761</pre>"
      ],
      "text/plain": [
       "Number of examples          : 9761"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of features          : 13</pre>"
      ],
      "text/plain": [
       "Number of features          : 13"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of unpacked features : 13</pre>"
      ],
      "text/plain": [
       "Number of unpacked features : 13"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of coefficients    : 14</pre>"
      ],
      "text/plain": [
       "Number of coefficients    : 14"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Starting Newton Method</pre>"
      ],
      "text/plain": [
       "Starting Newton Method"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |</pre>"
      ],
      "text/plain": [
       "| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| 1         | 2        | 0.077719     | 2534342.869092     | 246674.389430                   |</pre>"
      ],
      "text/plain": [
       "| 1         | 2        | 0.077719     | 2534342.869092     | 246674.389430                   |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>SUCCESS: Optimal solution found.</pre>"
      ],
      "text/plain": [
       "SUCCESS: Optimal solution found."
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre></pre>"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "624738503276231.1 -- 13\n",
      "124632209805192.69 -- 13\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<pre>Linear regression:</pre>"
      ],
      "text/plain": [
       "Linear regression:"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of examples          : 9761</pre>"
      ],
      "text/plain": [
       "Number of examples          : 9761"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of features          : 14</pre>"
      ],
      "text/plain": [
       "Number of features          : 14"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of unpacked features : 14</pre>"
      ],
      "text/plain": [
       "Number of unpacked features : 14"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Number of coefficients    : 15</pre>"
      ],
      "text/plain": [
       "Number of coefficients    : 15"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>Starting Newton Method</pre>"
      ],
      "text/plain": [
       "Starting Newton Method"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>--------------------------------------------------------</pre>"
      ],
      "text/plain": [
       "--------------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |</pre>"
      ],
      "text/plain": [
       "| Iteration | Passes   | Elapsed Time | Training Max Error | Training Root-Mean-Square Error |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>| 1         | 2        | 0.081889     | 2534786.244127     | 246672.360649                   |</pre>"
      ],
      "text/plain": [
       "| 1         | 2        | 0.081889     | 2534786.244127     | 246672.360649                   |"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>+-----------+----------+--------------+--------------------+---------------------------------+</pre>"
      ],
      "text/plain": [
       "+-----------+----------+--------------+--------------------+---------------------------------+"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre>SUCCESS: Optimal solution found.</pre>"
      ],
      "text/plain": [
       "SUCCESS: Optimal solution found."
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre></pre>"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "619369705907999.5 -- 14\n",
      "124612578389346.47 -- 14\n",
      "validation rss  [('1', 676709775198047.5), ('12', 627757914773691.8), ('11', 627012012706161.9), ('13', 624738503276231.1), ('10', 621744288936945.1), ('14', 619369705907999.5), ('3', 616714574532761.8), ('9', 612563853986526.5), ('4', 609129230654383.5), ('2', 607090530698014.1), ('8', 601558237778332.8), ('5', 599177138583675.6), ('7', 591717038418095.9), ('6', 589182477809739.2)]\n",
      "testing rss  [('1', 129030470094700.17), ('4', 127147719016434.84), ('5', 127037730210195.31), ('3', 125939457007406.9), ('6', 125529337848156.22), ('2', 125190117212860.06), ('12', 124646108317116.19), ('11', 124641733743998.38), ('13', 124632209805192.69), ('14', 124612578389346.47), ('7', 124608773610110.22), ('10', 124602928485231.25), ('9', 124525299776505.03), ('8', 124458964534463.83)]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#poly_data = {}\n",
    "#models = {}\n",
    "v_rss = {}\n",
    "t_rss = {}\n",
    "\n",
    "for power in range(1, 15):\n",
    "    '''\n",
    "    poly_data = polynomial_sframe(training['sqft_living'], power)\n",
    "    f = poly_data.column_names() # get the name of the features\n",
    "    print(f)\n",
    "    poly_data['price'] = training['price'] # add price to the data since it's the target\n",
    "    print(poly_data[9])\n",
    "    models = turicreate.linear_regression.create(poly_data, target = 'price', features = f, validation_set = None)\n",
    "    models.coefficients.print_rows(16)\n",
    "    \n",
    "    training_output = models.predict(training)\n",
    "    print(\"training actual=\", training['price'][9], \"training predicted=\" ,training_output[9])\n",
    "    \n",
    "    validation_output = models.predict(validation)\n",
    "    \n",
    "    print(\"validation actual=\", validation['price'][9], \"validation predicted=\" ,validation_output[9])\n",
    "    #rss[str(power)] = ((validation_output - validation['price'])**2).sum()\n",
    "    #print(f\"{power} {rss[str(power)]}\")\n",
    "    plt.plot(poly_data['power_' + str(power)],poly_data['price'],'.',\n",
    "         poly_data['power_' + str(power)], models.predict(poly_data),'-')\n",
    "    '''\n",
    "    \n",
    "    data = polynomial_sframe(training['sqft_living'], power)\n",
    "    my_features = data.column_names() # get the name of the features\n",
    "    data['price'] = training['price'] # add price to the data since it's the target\n",
    "    model = turicreate.linear_regression.create(data, target = 'price', features = my_features, validation_set = None)\n",
    "    \n",
    "    validation_data = polynomial_sframe(validation['sqft_living'], power)\n",
    "    validation_data['price'] = validation['price'] \n",
    "    validation_output = model.predict(polynomial_sframe(validation['sqft_living'], power))\n",
    "    \n",
    "    diff = validation_output - validation_data['price']\n",
    "    rss_v_i = (diff * diff).sum()\n",
    "    print(f\"{rss_v_i} -- {power}\")\n",
    "    v_rss[str(power)] = rss_v_i\n",
    "\n",
    "    testing_data = polynomial_sframe(testing['sqft_living'], power)\n",
    "    testing_data['price'] = testing['price'] \n",
    "    testing_output = model.predict(polynomial_sframe(testing['sqft_living'], power))\n",
    "    \n",
    "    diff = testing_output - testing_data['price']\n",
    "    testing_v_i = (diff * diff).sum()\n",
    "    print(f\"{testing_v_i} -- {power}\")\n",
    "    t_rss[str(power)] = testing_v_i\n",
    "    \n",
    "    plt.plot(data['power_1'],data['price'],'.',\n",
    "    data['power_1'], model.predict(data),'-', label=str(power))\n",
    "    \n",
    "print(\"validation rss \",sorted(v_rss.items(), key=lambda x: x[1], reverse=True))\n",
    "print(\"testing rss \",sorted(t_rss.items(), key=lambda x: x[1], reverse=True))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Quiz Question: Which degree (1, 2, …, 15) had the lowest RSS on Validation data?**\n",
    "\n",
    "('5', 1327985055990722.0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now that you have chosen the degree of your polynomial using validation data, compute the RSS of this model on TEST data. Report the RSS on your quiz."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "5 252901572839579.44\n"
     ]
    }
   ],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Quiz Question: what is the RSS on TEST data for the model with the degree selected from Validation data?**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "('7', 252901572839579.25)"
      ]
     },
     "execution_count": 113,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "('7', 252901572839579.25)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}