Chapter2_Numpy_matplotlib.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 2. Manipulating and plotting data in Python: `numpy`, and `matplotlib` libraries\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "toc": true
   },
   "source": [
    "<h1>Table of Contents<span class=\"tocSkip\"></span></h1>\n",
    "<div class=\"toc\"><ul class=\"toc-item\"><li><span><a href=\"#2.-Manipulating-and-plotting-data-in-Python:-numpy,-and-matplotlib-libraries\" data-toc-modified-id=\"2.-Manipulating-and-plotting-data-in-Python:-numpy,-and-matplotlib-libraries-1\">2. Manipulating and plotting data in Python: <code>numpy</code>, and <code>matplotlib</code> libraries</a></span><ul class=\"toc-item\"><li><span><a href=\"#2.1-Functions\" data-toc-modified-id=\"2.1-Functions-1.1\">2.1 Functions</a></span></li><li><span><a href=\"#2.2-numpy\" data-toc-modified-id=\"2.2-numpy-1.2\">2.2 <code>numpy</code></a></span><ul class=\"toc-item\"><li><span><a href=\"#2.2.1-arrays\" data-toc-modified-id=\"2.2.1-arrays-1.2.1\">2.2.1 arrays</a></span></li><li><span><a href=\"#2.2.2-More-detail-about-numpy.arrays\" data-toc-modified-id=\"2.2.2-More-detail-about-numpy.arrays-1.2.2\">2.2.2 More detail about <code>numpy.arrays</code></a></span></li><li><span><a href=\"#2.2.3-Array-creators\" data-toc-modified-id=\"2.2.3-Array-creators-1.2.3\">2.2.3 Array creators</a></span></li><li><span><a href=\"#2.2.4-Summary-statistics\" data-toc-modified-id=\"2.2.4-Summary-statistics-1.2.4\">2.2.4 Summary statistics</a></span></li><li><span><a href=\"#2.2.5-Summary\" data-toc-modified-id=\"2.2.5-Summary-1.2.5\">2.2.5 Summary</a></span></li></ul></li><li><span><a href=\"#2.3-Plotting-with-Matplotlib\" data-toc-modified-id=\"2.3-Plotting-with-Matplotlib-1.3\">2.3 Plotting with Matplotlib</a></span></li><li><span><a href=\"#2.4--Indexing-and-slicing-arrays\" data-toc-modified-id=\"2.4--Indexing-and-slicing-arrays-1.4\">2.4  Indexing and slicing arrays</a></span><ul class=\"toc-item\"><li><span><a href=\"#2.4.1-Recap\" data-toc-modified-id=\"2.4.1-Recap-1.4.1\">2.4.1 Recap</a></span></li><li><span><a href=\"#2.4.1-data-mask\" data-toc-modified-id=\"2.4.1-data-mask-1.4.2\">2.4.1 data mask</a></span></li></ul></li><li><span><a href=\"#2.5--Reading-data\" data-toc-modified-id=\"2.5--Reading-data-1.5\">2.5  Reading data</a></span><ul class=\"toc-item\"><li><span><a href=\"#2.5.1-np.loadtxt\" data-toc-modified-id=\"2.5.1-np.loadtxt-1.5.1\">2.5.1 <code>np.loadtxt</code></a></span></li><li><span><a href=\"#2.5.2-Accessing-online-resources\" data-toc-modified-id=\"2.5.2-Accessing-online-resources-1.5.2\">2.5.2 Accessing online resources</a></span></li><li><span><a href=\"#2.5.4-Homework\" data-toc-modified-id=\"2.5.4-Homework-1.5.3\">2.5.4 Homework</a></span><ul class=\"toc-item\"><li><span><a href=\"#Exercise-E2.5.4\" data-toc-modified-id=\"Exercise-E2.5.4-1.5.3.1\">Exercise E2.5.4</a></span></li></ul></li><li><span><a href=\"#2.5.5-Summary\" data-toc-modified-id=\"2.5.5-Summary-1.5.4\">2.5.5 Summary</a></span></li></ul></li></ul></li></ul></div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "While Python has a rich set of modules and data types by default, for numerical computing you'll be using two main libraries that conform the backbone of the [Python scientific stack](https://scipy.org/about.html). These libraries implement a great deal of functionality related to mathematical operations and efficient computations on large data volumes. These libraries are [`numpy`](http://numpy.org) and [`scipy`](http://scipy.org). `numpy`, which we will concentrate on in this section, deals with efficient arrays, similar to lists, that simplify many common processing operations. Of course, just doing calculations isn't much fun if you can't plot some results. To do this, we use the [`matplotlib`](http://matplotlib.org) library.\n",
    "\n",
    "But first, we'll see the concept of *functions*....\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2.1 Functions\n",
    "\n",
    "A function is a collection of Python statements that do something (usually on some data). For example, you may want to convert from Fahrenheit to Centigrade. The conversion is \n",
    "\n",
    "$$\n",
    "^{\\circ}C = \\left(^{\\circ}F -32\\right)\\cdot\\frac{5}{9}\n",
    "$$\n",
    "\n",
    "A Python function will have a name (and we hope that the name is self-explanatory as to what the function does), and a set of input parameters. In the case above, the function would look like this:\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "def fahrenheit_to_centigrade(deg_fahrenheit):\n",
    "    \"\"\"A function to convert from degrees Fahrenheit to degrees Centigrade\n",
    "\n",
    "    Parameters\n",
    "    ----------\n",
    "    deg_fahrenheit: float\n",
    "        Temperature in degrees F\n",
    "\n",
    "    Returns\n",
    "    -------\n",
    "    Temperature converted to degrees C\n",
    "    \"\"\"\n",
    "    deg_c = (deg_fahrenheit - 32.)*5./9.\n",
    "    return deg_c"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We see that the function has a name (`fahrenheit_to_centigrade`), and takes one parameter (`deg_fahrenheit`). \n",
    "\n",
    "The main body of the function is indented (like `if` and `for` statements). There is first a comment string, that describes what the function does, as well as what the inputs are, and what the output is. This is just useful documentation of the code. \n",
    "\n",
    "The main body of the function calculates `deg_C` from the given input, and **returns** it back to the user. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Notice that the document string `\"\"\"A function to convert from temperature... \"\"\"` is what is printed when you request `help` on the function:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Help on function fahrenheit_to_centigrade in module __main__:\n",
      "\n",
      "fahrenheit_to_centigrade(deg_fahrenheit)\n",
      "    A function to convert from degrees Fahrenheit to degrees Centigrade\n",
      "    \n",
      "    Parameters\n",
      "    ----------\n",
      "    deg_fahrenheit: float\n",
      "        Temperature in degrees F\n",
      "    \n",
      "    Returns\n",
      "    -------\n",
      "    Temperature converted to degrees C\n",
      "\n"
     ]
    }
   ],
   "source": [
    "help(fahrenheit_to_centigrade)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**E2.1.1 Exercise**\n",
    "\n",
    "* In the vein of converting units, write functions that convert from\n",
    "\n",
    "    * inches to m (and back)\n",
    "    * kg to stones (and back)\n",
    "\n",
    "Hint: A stone is equal to 14 pounds, and a pound is equal to 0.45359237 kg.\n",
    "\n",
    "**Ensure** that your functions are clearly named, have sensible variable names, a brief docmentation string, and remember to test the functions work: just demonstrate running the function with some input pairs where you know the output and checking it makese sense."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Space for your solution"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "## 2.2 `numpy`\n",
    "\n",
    "### 2.2.1 arrays\n",
    "\n",
    "You import the `numpy` library using\n",
    "\n",
    "    import numpy as np\n",
    "    \n",
    "This means that all the functionality of `numpy` is accessed by the prefix `np.`: e.g. `np.array`. The main element of `numpy` is the numpy array. An array is like a list, but unlike a list, all the elements are of the same type, floating point numbers for example. \n",
    "\n",
    "Let's see some arrays in action..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-09-26T16:26:31.395372Z",
     "start_time": "2018-09-26T16:26:31.387781Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1. 1. 1. 1. 1.]\n",
      "<class 'numpy.ndarray'>\n",
      "[1 2 3 4]\n",
      "[1. 2. 3. 4.]\n"
     ]
    }
   ],
   "source": [
    "import numpy as np  # Import the numpy library\n",
    "\n",
    "# An array with 5 ones\n",
    "arr = np.ones(5)\n",
    "print(arr)\n",
    "print(type(arr))\n",
    "\n",
    "# An array started from a list of **integers**\n",
    "arr = np.array([1, 2, 3, 4])\n",
    "print(arr)\n",
    "\n",
    "# An array started from a list of numbers, what's the difference??\n",
    "arr = np.array([1., 2, 3, 4])\n",
    "print(arr)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In the example above we have generated an array where all the elements are `1.0`, using [`np.ones`](https://docs.scipy.org/doc/numpy/reference/generated/numpy.ones.html), and then we have been able to generate arrays from lists using the [`np.array`](https://docs.scipy.org/doc/numpy/reference/generated/numpy.array.html) function. The difference between the 2nd and 3rd examples is that in the 2nd example, all the elements of the list are integers, and in the 3rd example, one is a floating point number. `numpy` automatically makes the array floating point by converting the integers to floating point numbers.\n",
    "\n",
    "What can we do with arrays? We can efficiently operate on individual elements without loops:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[2. 2. 2. 2. 2. 2. 2. 2. 2. 2.]\n"
     ]
    }
   ],
   "source": [
    "arr = np.ones(10)\n",
    "print(2 * arr)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`numpy` is clever enough to figure out that the 2 multiplying the array is applied to all elements of the array, and returns an array of the same size as `arr` with the elements of `arr` multiplied by 2. We can also multiply two arrays of the same size. So let's create an array with the numbers 0 to 9 and one with the numbers 9 to 0 and do a times table:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[9. 9. 9. 9. 9. 9. 9. 9. 9. 9.]\n",
      "[ 1  2  3  4  5  6  7  8  9 10]\n",
      "[ 9. 18. 27. 36. 45. 54. 63. 72. 81. 90.]\n"
     ]
    }
   ],
   "source": [
    "arr1 = 9 * np.ones(10)\n",
    "arr2 = np.arange(1, 11)  # arange gives an array from 1 to 11, 11 not included\n",
    "\n",
    "print(arr1)\n",
    "print(arr2)\n",
    "\n",
    "print(arr1 * arr2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**E2.2.1 Exercise**\n",
    "\n",
    "* Using code similar to the above and a `for` loop, write the times tables for 2 to 10. The solution you're looking for should look a bit like this:\n",
    "\n",
    "        [ 2  4  6  8 10 12 14 16 18 20]\n",
    "        [ 3  6  9 12 15 18 21 24 27 30]\n",
    "        [ 4  8 12 16 20 24 28 32 36 40]\n",
    "        [ 5 10 15 20 25 30 35 40 45 50]\n",
    "        [ 6 12 18 24 30 36 42 48 54 60]\n",
    "        [ 7 14 21 28 35 42 49 56 63 70]\n",
    "        [ 8 16 24 32 40 48 56 64 72 80]\n",
    "        [ 9 18 27 36 45 54 63 72 81 90]\n",
    "        [ 10  20  30  40  50  60  70  80  90 100]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Your solution here"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If the arrays are of the same *shape*, you can do standard operations between them **element-wise**:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[27. 36. 45. 54.]\n",
      "[ 90. 160. 250. 360.]\n",
      "Array shapes:\n",
      "arr1:  (4,)\n",
      "arr2:  (4,)\n"
     ]
    }
   ],
   "source": [
    "arr1 = np.array([3, 4, 5, 6.])\n",
    "arr2 = np.array([30, 40, 50, 60.])\n",
    "\n",
    "print(arr2 - arr1)\n",
    "print(arr1 * arr2)\n",
    "\n",
    "print(\"Array shapes:\")\n",
    "print(\"arr1: \", arr1.shape)\n",
    "print(\"arr2: \", arr2.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The `numpy` documenation is huge. There's an [user's guide](https://docs.scipy.org/doc/numpy/user/index.html), as well as a reference to all the [contents of the library](https://docs.scipy.org/doc/numpy/reference/index.html). There's even [a tutorial availabe](https://docs.scipy.org/doc/numpy/user/quickstart.html) if you get bored with this one.\n",
    "\n",
    "### 2.2.2 More detail about `numpy.arrays` \n",
    "\n",
    "So far, we have seen a 1D array, which is the equivalent to a vector. But arrays can have more dimensions: a 2D array would be equivalent to a matrix (or an image, with rows and columns), and a 3D array would be a volume split into voxels, as seen below\n",
    "\n",
    "\n",
    "![numpy arrays](https://cdn-images-1.medium.com/max/1120/1*Ikn1J6siiiCSk4ivYUhdgw.png)\n",
    "\n",
    "So a 1D array has one axis, a 2D array has 2 axes, a 3D array 3, and so on. The `shape` of the array provides a tuple with the number of elements along each axis. Let's see this with some generally useful array creation options:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3*5 array:\n",
      "[[ 0  1  2  3  4]\n",
      " [ 5  6  7  8  9]\n",
      " [10 11 12 13 14]]\n",
      "2*3 array:\n",
      "[[1.5 2.  3. ]\n",
      " [4.  5.  6. ]]\n"
     ]
    }
   ],
   "source": [
    "# Create a 2D array from a list of rows. Note that the 3 rows have the same number of elements!\n",
    "arr1 = np.array([[0, 1, 2, 3, 4], [5, 6, 7, 8, 9], [10, 11, 12, 13, 14]])\n",
    "# A 2D array from a list of tuples.\n",
    "# We're specifically asking for floating point numbers\n",
    "arr2 = np.array([(1.5, 2, 3), (4, 5, 6)], dtype=np.float)\n",
    "print(\"3*5 array:\")\n",
    "print(arr1)\n",
    "print(\"2*3 array:\")\n",
    "print(arr2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.2.3 Array creators\n",
    "\n",
    "Quite often, we will want to initialise an array to be all the same number. The methods for doing this as 0,1 and unspecified in `numpy` are `np.zeros()`, `np.ones()`, `np.empty()` respectively."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3*4 array of 0s\n",
      "[[0. 0. 0. 0.]\n",
      " [0. 0. 0. 0.]\n",
      " [0. 0. 0. 0.]]\n",
      "2*3*4 array of 1s (integers)\n",
      "[[[1 1 1 1]\n",
      "  [1 1 1 1]\n",
      "  [1 1 1 1]]\n",
      "\n",
      " [[1 1 1 1]\n",
      "  [1 1 1 1]\n",
      "  [1 1 1 1]]]\n",
      "2*3 empty array (contents could be anything)\n",
      "[[1.5 2.  3. ]\n",
      " [4.  5.  6. ]]\n"
     ]
    }
   ],
   "source": [
    "# Creates a 3*4 array of 0s\n",
    "arr = np.zeros((3, 4))\n",
    "print(\"3*4 array of 0s\")\n",
    "print(arr)\n",
    "\n",
    "# Creates a 2x3x4 array of int 1's\n",
    "print(\"2*3*4 array of 1s (integers)\")\n",
    "arr = np.ones((2, 3, 4), dtype=np.int)\n",
    "print(arr)\n",
    "\n",
    "# Creates an empty (e.g. uninitialised) 2x3 array. Elements are random\n",
    "print(\"2*3 empty array (contents could be anything)\")\n",
    "arr = np.empty((2, 3))\n",
    "print(arr)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Exercise E2.2.2**\n",
    "\n",
    "* write a function that does the following:\n",
    "    * create a 2-D tuple called `indices` containing the integers `((0, 1, 2, 3, 4),(5, 6, 7, 8, 9))`\n",
    "    * create a 2-D numpy array called `data` of shape `(5,10)`, data type `int`, initialised with zero\n",
    "    * set the value of `data[r,c]` to be `1` for each of the 5 row,column pairs specified in `indices`.\n",
    "    * return the data array\n",
    "* print out the result returned\n",
    "\n",
    "The result should look like:\n",
    "\n",
    "    [[0 0 0 0 0 1 0 0 0 0]\n",
    "     [0 0 0 0 0 0 1 0 0 0]\n",
    "     [0 0 0 0 0 0 0 1 0 0]\n",
    "     [0 0 0 0 0 0 0 0 1 0]\n",
    "     [0 0 0 0 0 0 0 0 0 1]]\n",
    "\n",
    "**Hint**: You could use a `for` loop, but what does `data[indices]` give?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# do exercise here"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Exercise 2.2.3**\n",
    "\n",
    "* write a more flexible version of you function above where `indices`, the value you want to set (`1` above) and the desired shape of `data` are specified through function keyword arguments (e.g. `indices=((0, 1, 2, 3, 4),(5, 6, 7, 8, 9)),value=1,shape=(5,10)`)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# do exercise here"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As well as initialising arrays with the same number as above, we often also want to initialise with common data patterns. This includes simple integer ranges `(start, stop, skip)` in a similar fashion to slicing in the last session, or variations on this theme:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1D array of numbers from 0 to 2 in increments of 0.3\n",
      "arr of shape (7,):\n",
      "\t[0.  0.3 0.6 0.9 1.2 1.5 1.8]\n",
      "array of shape (9,) numbers equally spaced from 0 to 34:\n",
      "\t[ 0.    4.25  8.5  12.75 17.   21.25 25.5  29.75 34.  ]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([34.  , 29.75, 25.5 , 21.25, 17.  , 12.75,  8.5 ,  4.25,  0.  ])"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# array creators\n",
    "\n",
    "print(\"1D array of numbers from 0 to 2 in increments of 0.3\")\n",
    "start = 0\n",
    "stop = 2.0\n",
    "skip = 0.3\n",
    "\n",
    "arr = np.arange(start, stop, skip)\n",
    "print(f'arr of shape {arr.shape}:\\n\\t{arr}')\n",
    "\n",
    "start = 0\n",
    "stop = 34\n",
    "nsamp = 9\n",
    "arr = np.linspace(start, stop, nsamp)\n",
    "print(\n",
    "    f\"array of shape {arr.shape} numbers equally spaced from {start} to {stop}:\\n\\t{arr}\")\n",
    "\n",
    "np.linspace(stop, start, 9)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Exercise E2.2.4**\n",
    "\n",
    "* print an array of integer numbers from 100 to 1\n",
    "* print an array with 9 numbers equally spaced between 100 and 1\n",
    "\n",
    "Hint: what value of skip would be appropriate here? what about `start` and `stop`?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "# do exercise here"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.2.4 Summary statistics\n",
    "\n",
    "Below are some typical arithmetic operations that you can use on arrays. Remember that they happen **elementwise** (i.e. to the whole array)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0 1 2 3]^2 = [0 1 4 9]\n",
      "\n",
      "assuming in radians,\n",
      "10*sin([20 30 40 50]) = [ 9.12945251 -9.88031624  7.4511316  -2.62374854]\n",
      "\n",
      "Some useful numpy array methods for summary statistics...\n",
      "\n",
      "Find the maximum of an array: a.max():  50\n",
      "Find the minimum of an array: a.min():  20\n",
      "Find the sum of an array: a.sum():  140\n",
      "Find the mean of an array: a.mean():  35.0\n",
      "Find the standard deviation of an array: a.std():  11.180339887498949\n"
     ]
    }
   ],
   "source": [
    "b = np.arange(4)\n",
    "print(f'{b}^2 = {b**2}\\n')\n",
    "\n",
    "a = np.array([20, 30, 40, 50])\n",
    "print(f\"assuming in radians,\\n10*sin({a}) = {10 * np.sin(a)}\")\n",
    "\n",
    "print(\"\\nSome useful numpy array methods for summary statistics...\\n\")\n",
    "print(\"Find the maximum of an array: a.max(): \", a.max())\n",
    "print(\"Find the minimum of an array: a.min(): \", a.min())\n",
    "print(\"Find the sum of an array: a.sum(): \", a.sum())\n",
    "print(\"Find the mean of an array: a.mean(): \", a.mean())\n",
    "print(\"Find the standard deviation of an array: a.std(): \", a.std())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's access an interesting dataset on the frequency of satellite launches to illustrate this.\n",
    "\n",
    "![SpaceX landing](https://media.giphy.com/media/26DNbCqVfLJbYrXIA/giphy.gif)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "data shape (12, 62)\n",
      "some summary statistics over the period 1957 to 2019:\n",
      "The total number of launches is 43611\n",
      "The mean number of launches is 58.617 per month\n"
     ]
    }
   ],
   "source": [
    "from geog0111.nsat import nsat\n",
    "\n",
    "'''\n",
    "This dataset gives the number of \n",
    "satellites launched per month and year\n",
    "data from https://www.n2yo.com\n",
    "'''\n",
    "# We use the code supplied in nsat.py\n",
    "# to generate the dataset (takes time)\n",
    "# or to load it if it exists\n",
    "data, years = nsat().data, nsat().years\n",
    "\n",
    "print(f'data shape {data.shape}')\n",
    "\n",
    "print(f'some summary statistics over the period {years[0]} to {years[1]}:')\n",
    "print(f'The total number of launches is {data.sum():d}')\n",
    "print(f'The mean number of launches is {data.mean():.3f} per month')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Exercise E2.2.5**\n",
    "\n",
    "* copy the code above but generate a fuller set of summary statistics including the standard deviation, minimum and maximum.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "# do exercise here"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Whilst we have generated some interesting summary statistics on the dataset, it's not really enough to give us a good idea of the data characteristics.\n",
    "\n",
    "To do that, we want to be able to ask somewhat more complex questions of the data, such as, which *year* has the most/least launches? which month do most launches happen in? which month in which year had the most launches? which years had more than 100 launches?\n",
    "\n",
    "To be able to address these, we need some new concepts:\n",
    "\n",
    "* methods `argmin()` and `argmax()` that provide the *index* where the min/max occurs\n",
    "* filtering and the related method `where()`\n",
    "* `axis` methods: the dataset is two-dimensional, and for some questions we need to operate only over one of these\n",
    "\n",
    "To illustrate:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shape of data array: (12, 62)\n",
      "the year with most launches was 1999 with 4195\n",
      "the year with fewest launches was 1957 with 3\n",
      "\n",
      "The years with >= 1000 launches are:\n",
      "[1965 1975 1976 1981 1986 1987 1993 1994 1999 2006] \n",
      "values:\n",
      " [1527 1195 1264 1190 1375 1130 2131 1166 4195 1158]\n",
      "The years with <= 300 launches are:\n",
      "[1957 1958 1959 1960 1962 1996 2002 2003 2004 2005] \n",
      "values:\n",
      " [  3  11  22  52 207 246 277 243 209 192]\n"
     ]
    }
   ],
   "source": [
    "from geog0111.nsat import nsat\n",
    "import numpy as np\n",
    "\n",
    "data = nsat().data\n",
    "start_year, end_year = nsat().years\n",
    "\n",
    "print(\"Shape of data array:\", data.shape)\n",
    "\n",
    "\n",
    "# sum the data over all months (axis 0)\n",
    "sum_per_year = data.sum(axis=0)\n",
    "\n",
    "# Construct an array of years\n",
    "year_array = np.arange(start_year, end_year)\n",
    "\n",
    "# Find the location (e.g. year) with **most** launches\n",
    "imax = np.argmax(sum_per_year)\n",
    "# Find the location (e.g. year) with **least** launches\n",
    "imin = np.argmin(sum_per_year)\n",
    "\n",
    "# filtering\n",
    "# high(low) is an array set to True where the condition\n",
    "# is True, and False otherwise\n",
    "high = sum_per_year >= 1000  # True if more than 1000 launches\n",
    "low = sum_per_year <= 300  # True if less than 300 launches\n",
    "\n",
    "print(\n",
    "    f'the year with most launches was {year_array[imax]} with {sum_per_year[imax]}')\n",
    "print(\n",
    "    f'the year with fewest launches was {year_array[imin]} with {sum_per_year[imin]}')\n",
    "\n",
    "print('\\nThe years with >= 1000 launches are:')\n",
    "print(year_array[high], '\\nvalues:\\n', sum_per_year[high])\n",
    "\n",
    "print('The years with <= 300 launches are:')\n",
    "print(year_array[low], '\\nvalues:\\n', sum_per_year[low])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Exercise E2.2.6**\n",
    "\n",
    "* copy the code above, and modify it to find the total launches *per month* (over all years)\n",
    "* show these data in a table\n",
    "* which month do launches mostly take place in? which month do launches most seldom take place in?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "# do exercise here"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The form of filtering above (`high = sum_per_year >= 1000`) produces a numpy array of the same shape as that operated on (`sum_per_year` here) of `bool` data type.  It has entries of `True` where the condition is met, and `False` where it is not met."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "type(sum_per_year): <class 'numpy.ndarray'>, sum_per_year.shape: (62,), sum_per_year.dtype: int64\n",
      "type(high): <class 'numpy.ndarray'>, high.shape: (62,), high.dtype: bool\n",
      "\n",
      "sum_per_year: [   3   11   22   52  396  207  346  401 1527  786  466  690  641  906\n",
      "  636  654  875  694 1195 1264  891  783  857  637 1190  946  884  760\n",
      "  788 1375 1130  814  950  691  691  740 2131 1166  534  246  960  651\n",
      " 4195  730  582  277  243  209  192 1158  349  406  378  373  315  435\n",
      "  352  355  335  308  512  320]\n",
      "high: [False False False False False False False False  True False False False\n",
      " False False False False False False  True  True False False False False\n",
      "  True False False False False  True  True False False False False False\n",
      "  True  True False False False False  True False False False False False\n",
      " False  True False False False False False False False False False False\n",
      " False False]\n",
      "low: [ True  True  True  True False  True False False False False False False\n",
      " False False False False False False False False False False False False\n",
      " False False False False False False False False False False False False\n",
      " False False False  True False False False False False  True  True  True\n",
      "  True False False False False False False False False False False False\n",
      " False False]\n"
     ]
    }
   ],
   "source": [
    "from geog0111.nsat import nsat\n",
    "# sum the data over all months (axis 0)\n",
    "sum_per_year = nsat().data.sum(axis=0)\n",
    "\n",
    "high = sum_per_year >= 1000\n",
    "low = sum_per_year <= 300\n",
    "\n",
    "print(f'type(sum_per_year): {type(sum_per_year)}, sum_per_year.shape: {sum_per_year.shape}, '\n",
    "      + f'sum_per_year.dtype: {sum_per_year.dtype}')\n",
    "print(\n",
    "    f'type(high): {type(high)}, high.shape: {high.shape}, high.dtype: {high.dtype}\\n')\n",
    "\n",
    "print(f'sum_per_year: {sum_per_year}')\n",
    "print(f'high: {high}')\n",
    "print(f'low: {low}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can think of this logical array as a 'data mask' that we use to select (filter) entries. \n",
    "\n",
    "The figure shows `log(sum_per_year)` in the top line of the image (numbers represented by colour shown in colourbar), then a representation of the `bool` arrays `high` and `low`. Where the `bool` value is shown yellow, the 'data mask' is true.\n",
    "![](images/arrayviz.png)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1527 1195 1264 1190 1375 1130 2131 1166 4195 1158]\n",
      "[  3  11  22  52 207 246 277 243 209 192]\n"
     ]
    }
   ],
   "source": [
    "print(f'{sum_per_year[high]}')\n",
    "print(f'{sum_per_year[low]}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Sometimes, instead of just applying the filter as above, we want to know the indices of the filtered values. \n",
    "\n",
    "To do this, we can use the `np.where()` method. This takes a `bool` array as its argument (such as our data masks or other conditions) and returns a tuple of the indices where this is set `True`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "indices:\n",
      "[1 3 4 4 5 5 5 6 8 8 9 9]\n",
      "[29 13 37 42 24 36 49 19 40 43  8 42]\n",
      "\n",
      "type(indices): <class 'tuple'>\n",
      "len(indices): 2, len(indices[0]): 12\n",
      "type(indices[0][0]): <class 'numpy.int64'>\n",
      "***********************\n",
      "busy months\n",
      "***********************\n",
      "0000 month 01 year 1986 (870)\n",
      "0001 month 03 year 1970 (486)\n",
      "0002 month 04 year 1994 (770)\n",
      "0003 month 04 year 1999 (3476)\n",
      "0004 month 05 year 1981 (507)\n",
      "0005 month 05 year 1993 (1699)\n",
      "0006 month 05 year 2006 (531)\n",
      "0007 month 06 year 1976 (592)\n",
      "0008 month 08 year 1997 (665)\n",
      "0009 month 08 year 2000 (491)\n",
      "0010 month 09 year 1965 (499)\n",
      "0011 month 09 year 1999 (447)\n",
      "***********************\n"
     ]
    }
   ],
   "source": [
    "from geog0111.nsat import nsat\n",
    "data = nsat().data\n",
    "# where :\n",
    "# which months in the dataset were particularly busy ..\n",
    "# we select data > 400 as a condition\n",
    "\n",
    "indices = np.where(data > 400)\n",
    "# indices is a 2-element tuple with the locations of months\n",
    "# with more than 400 launches\n",
    "print(f'indices:\\n{indices[0]}\\n{indices[1]}')\n",
    "print(f'\\ntype(indices): {type(indices)}')\n",
    "print(f'len(indices): {len(indices)}, len(indices[0]): {len(indices[0])}')\n",
    "print(f'type(indices[0][0]): {type(indices[0][0])}')\n",
    "\n",
    "start_year, end_year = nsat().years\n",
    "year_list = np.arange(start_year, end_year)\n",
    "month = np.arange(12)\n",
    "\n",
    "# Number of months where there are more than 400 launches\n",
    "n_high_launches_months = len(indices[0])\n",
    "\n",
    "# loop over the entries in the tuple\n",
    "print('*'*23)\n",
    "print('busy months')\n",
    "print('*'*23)\n",
    "for i in range(n_high_launches_months):\n",
    "    # Calculate and store data for each month\n",
    "    month_location = month[indices[0][i]]\n",
    "    year_location = year_list[indices[1][i]]\n",
    "    launches = data[indices[0][i], indices[1][i]]\n",
    "    print(f\"{i:04} month {month_location:02d} year \" +\n",
    "          f\"{year_location:d} ({launches:d})\")\n",
    "print('*'*23)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Exercise E2.2.7**\n",
    "\n",
    "* Using code from the sections above, print out a table with the busiest launch months with an additional column stating the number of launches\n",
    "\n",
    "Hint: this is just adding another column to the print statement in the for loop"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "# do exercise here"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "You might notice the indices in the tuple derived above using `where` are *ordered*, but the effect of this is that the months are in sequential order, rather than the years. We have\n",
    "\n",
    "    month[indices[0][i]]\n",
    "    year[indices[1][i]]\n",
    "    \n",
    "If we want to put the data in year order, there are several ways we could go about this. An insteresting one, following the ideas in `argmax()` and `argmin()` above is to use `argsort()`. This gives the *indices* of the sorted array, rather than the values.\n",
    "\n",
    "So here, we can find the *indices* of the year-sorted array, and apply them to both `month` and `year` datasets:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "years not in order: [1986 1970 1994 1999 1981 1993 2006 1976 1997 2000 1965 1999]\n",
      "but months are:     [1 3 4 4 5 5 5 6 8 8 9 9]\n",
      "\n",
      "year order: [10  1  7  4  0  5  2  8  3 11  9  6]\n",
      "\n",
      "years in order: [1965 1970 1976 1981 1986 1993 1994 1997 1999 1999 2000 2006]\n",
      "months in year order: [9 3 6 5 1 5 4 8 4 9 8 5]\n"
     ]
    }
   ],
   "source": [
    "# prepare data as above\n",
    "from geog0111.nsat import nsat\n",
    "\n",
    "\n",
    "def prepare_satellite_data(min_launches=400):\n",
    "    \"\"\"Prepares satellite launch data, returning\n",
    "    data as well as indices where more than `min_launches`\n",
    "    took place per month.\n",
    "    \"\"\"\n",
    "    data = nsat().data\n",
    "    start_year, end_year = nsat().years\n",
    "    indices = np.where(data > 400)\n",
    "    year_arr = np.arange(start_year, end_year)\n",
    "    month_arr = np.arange(12)\n",
    "    return data, indices, year_arr, month_arr\n",
    "\n",
    "\n",
    "data, indices, year_arr, month_arr = prepare_satellite_data()\n",
    "\n",
    "# store the months and years\n",
    "# in their unsorted (original) form\n",
    "unsorted_months = month_arr[indices[0]]\n",
    "unsorted_years = year_arr[indices[1]]\n",
    "print(f'years not in order: {unsorted_years}')\n",
    "print(f'but months are:     {unsorted_months}\\n')\n",
    "\n",
    "\n",
    "# get the indices to put years in order\n",
    "year_order = np.argsort(indices[1])\n",
    "\n",
    "# apply this to months and years\n",
    "print(f'year order: {year_order}\\n')\n",
    "print(f'years in order: {unsorted_years[year_order]}')\n",
    "print(f'months in year order: {unsorted_months[year_order]}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Exercise E2.2.8**\n",
    "\n",
    "* Use this example of `argsort()` to redo Exercise E2.2.7, putting the data in correct year order"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "# do exercise here"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.2.5 Summary\n",
    "\n",
    "In this section, you have been introduced to more detail on arrays in `numpy`. The big advantages of `numpy` are that you can easily perform array operators (such as adding two arrays together), and that `numpy` has a large number of useful functions for manipulating N-dimensional data in array form. This makes it particularly appropriate for raster geospatial data processing.\n",
    "\n",
    "We have seen how to create various forms of array (e.g. `np.ones()`, `np.arange()`), how to calculate some basic statistics (`min()`, `max()` etc), and finding the array index where some pattern occurs (e.g. `argmin()`, `argsort()` or `where()`). "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2.3 Plotting with Matplotlib\n",
    "\n",
    "There are [quite a few graphical libraries for Python](https://scipy.org/topical-software.html#plotting-data-visualization-3-d-programming), but [matplotlib](http://matplotlib.org) is probably the most famous one. It does pretty much all you need in terms of 2D plots, and simple 3D plots, and is fairly straightforward to use. Have a look at the [matplotlib gallery](https://matplotlib.org/gallery/index.html) for a fairly comprehensive list of examples of what the library can do as well as the code that was used in the examples.\n",
    "\n",
    "**Importing matplotlib**\n",
    "\n",
    "You can import matplotlib with \n",
    "\n",
    "    import matplotlib.pyplot as plt\n",
    "    \n",
    "As with `numpy`, it's custom to use the `plt` prefix to call matplotlib commands. In the notebook, you should also issue the following command just after the import\n",
    "\n",
    "    %matplotlib notebook\n",
    "    \n",
    "or\n",
    "\n",
    "    %matplotlib inline\n",
    "    \n",
    "    \n",
    "The former command will make the plots in the notebook interactive (i.e. point-and-click-ey), and the second will just stick the plots into the notebook as PNG files.\n",
    "\n",
    "**Simple 2D plots**\n",
    "\n",
    "The most basic plots are 2D plots (e.g. x and y). \n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "data shape (12, 62)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7fcc03473ef0>]"
      ]
     },
     "execution_count": 26,
     "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": [
    "import numpy as np\n",
    "from geog0111.nsat import nsat\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "'''\n",
    "This dataset gives the number of \n",
    "satellites launched per month and year\n",
    "\n",
    "data from https://www.n2yo.com\n",
    "'''\n",
    "data, years = nsat().data, nsat().years\n",
    "year = np.arange(years[0], years[1], dtype=np.int)\n",
    "# sum the data over all months (axis 0)\n",
    "sum_per_year = data.sum(axis=0)\n",
    "\n",
    "print(f'data shape {data.shape}')\n",
    "\n",
    "# plot x as year\n",
    "# plot y as the number of satellites per year\n",
    "plt.plot(year, sum_per_year, label='launches per year')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Whilst this plot is fine, there are a few simple things we could do improve it. \n",
    "\n",
    "We will go through some of the options below, but to get a taste of improved ploitting, lets use e.g.:\n",
    "\n",
    "* reset the image shape/size\n",
    "    * `plt.figure(figsize=(13,3))`\n",
    "* plot the mean value (as a red dashed line) for comparison\n",
    "    * `plt.plot([year[0],year[-1]],[mean,mean],'r--',label='mean')`\n",
    "* limit the dataset to range of variable `year`\n",
    "    * `plt.xlim(year[0],year[-1])`\n",
    "* put labels on the x and y axes\n",
    "    * `plt.xlabel('year')`\n",
    "    * `plt.ylabel('# satellite launches')`\n",
    "* set a title\n",
    "    * `plt.title('data from https://www.n2yo.com')`\n",
    "* use a legend (in conjunction with `label=` using `plot`)\n",
    "    * `plt.legend(loc='best')`\n",
    "* use a log scale in the y-axis\n",
    "    * `plt.semilogy()`\n",
    "    \n",
    "What you choose to do will depend on what you want to show on the graph, but the examples above are quite common."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[]"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 936x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "from geog0111.nsat import nsat\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "'''data as above'''\n",
    "data, years = nsat().data, nsat().years\n",
    "year = np.arange(years[0], years[1], dtype=np.int)\n",
    "sum_per_year = data.sum(axis=0)\n",
    "\n",
    "# calculate mean of sum_per_year\n",
    "mean = sum_per_year.mean()\n",
    "\n",
    "plt.figure(figsize=(13, 3))\n",
    "plt.plot(year, sum_per_year, label='launches per year')\n",
    "plt.plot([year[0], year[-1]], [mean, mean], 'r--', label='mean')\n",
    "plt.xlim(year[0], year[-1])\n",
    "plt.xlabel('year')\n",
    "plt.ylabel('# satellite launches')\n",
    "plt.title('data from https://www.n2yo.com')\n",
    "plt.legend(loc='best')\n",
    "plt.semilogy()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Exercise E2.3.1**\n",
    "\n",
    "* produce a plot showing launches per year as a function of year, showing data for selected months individually.\n",
    "\n",
    "Hint: do a simple plot first, then add some improvements gradually. You might set up a list of months to process and use a loop to go over each month."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "# do exercise here"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Exercise 2.3.2**\n",
    "\n",
    "Putting together some ideas from above to look at some turning points in a function:\n",
    "\n",
    "* generate a numpy array called `x` with 100 equally spaced numbers between 0 and 5\n",
    "* generate a numpy array called `y` which contains $x^3 - 9 x^2 + 26 x - 24$\n",
    "* plot `y` as a function of `x` with a red line\n",
    "* plot **only positive** values of `y` (as a function of `x`) with a green line\n",
    "\n",
    "Hint: to plot with red and green line `plot(x,y,'r')` and `plot(x,y,'g')` "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "# do exercise here"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2.4  Indexing and slicing arrays\n",
    "\n",
    "### 2.4.1 Recap\n",
    "\n",
    "Selecting different elements of the array to operate in them is a very common task. `numpy` has a very rich syntax for selecting different bits of the array. We have come across slicing before, but it is so important to array processing, we will go over some of it again. \n",
    "\n",
    "Similar to lists, you can refer to elements in the array by their position. You can also use the `:` symbol to specify a range (a **slice**) of positions `first_element:(last_element+1`. If you want to start counting from the end of the array, use negative numbers: `-1` refers to the last element of the array, `-2` the one before last and so on. In a slice, you can also specify a step as the third element in `first_element:(last_element+1:step`. If the step is negative you count from the back.\n",
    "\n",
    "All this probably appears mind bogging, but it's easier shown in practice. You'll get used to it quite quickly once you start using it"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2\n",
      "[2 3 4]\n",
      "10\n",
      "[0 1 2 3 4 5 6 7]\n",
      "[ 2  3  4  5  6  7  8  9 10]\n",
      "[5 4 3]\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "\n",
    "a = np.array([0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10])\n",
    "print(a[2])      # 2\n",
    "print(a[2:5])    # [2, 3, 4]\n",
    "print(a[-1])     # 10\n",
    "print(a[:8])     # [0, 1, 2, 3, 4, 5, 6, 7]\n",
    "print(a[2:])     # [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]\n",
    "print(a[5:2:-1])  # [5, 4, 3]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The concept extends cleanly to multidimensional arrays..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "23\n",
      "[ 1 11 21 31 41]\n",
      "[ 1 11 21 31 41]\n",
      "[[10 11 12 13]\n",
      " [20 21 22 23]]\n"
     ]
    }
   ],
   "source": [
    "b = np.array([[0, 1, 2, 3], [10, 11, 12, 13], [20, 21, 22, 23], [30, 31, 32, 33],\n",
    "              [40, 41, 42, 43]])\n",
    "\n",
    "print(b[2, 3])    # 23\n",
    "print(b[0:5, 1])  # each row in the second column of b\n",
    "print(b[:, 1])    # same thing as above\n",
    "print(b[1:3, :])  # each column in the second and third row of b"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Exercise 2.4.1**\n",
    "\n",
    "* generate a 2-D numpy array of integer zeros called `x`, of shape (7,7)\n",
    "* we can think of this as a square. Set the central 3 by 3 samples of the square to one\n",
    "* print the result\n",
    "\n",
    "Hint: Don't use looping, instead work out how to define the slice of the central 3 x 3 samples. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "# do exercise here"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.4.1 data mask\n",
    "\n",
    "\n",
    "A useful way to select elements is by using what's called a mask as we saw above: an array of logical (boolean) elements that only selects the elements that are `True`:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[7 8 9]\n"
     ]
    }
   ],
   "source": [
    "a = np.arange(10)\n",
    "select_me = a >= 7\n",
    "print(a[select_me])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The previous point also shows something interesting: you can apply comparisons element by element. So in the previous example, `select_me` is a 10 element array where all the elements of `a` that are equal or higher than 7 are set to True.\n",
    "\n",
    "If you want to build up element by element logical operations, it's best to use specialised functions like [`np.logical_and`](https://docs.scipy.org/doc/numpy/reference/generated/numpy.logical_and.html) and friends"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69\n",
      " 70 71 72]\n"
     ]
    }
   ],
   "source": [
    "a = np.arange(100)\n",
    "sel1 = a > 45\n",
    "sel2 = a < 73\n",
    "print(a[np.logical_and(sel1, sel2)])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Exercise 2.4.2**\n",
    "\n",
    "* generate a numpy array called `x` with 100 equally spaced numbers between 0 and 5\n",
    "* generate a numpy array called `y` which contains $x^3 - 9 x^2 + 26 x - 24$\n",
    "* print the values of `x` for which `y` is greater than or equal to zero and `x` lies between 3.5 and 4.5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "# do exercise here"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2.5  Reading data\n",
    "\n",
    "### 2.5.1 `np.loadtxt`\n",
    "\n",
    "It's a bit tedious just making up numbers to play with them, but it's easy to load up data from external files. The most common data interchange format is [CSV (comma-seperated values)](https://en.wikipedia.org/wiki/Comma-separated_values), a plain text format. Think of CSV as a plain text table. Each element in each row is separated by a comma (although other symbols, such as white space, semicolons `;`, tabs `\\t` or pipe `|` symbols are often found as delimiters). The first few lines might contain some metadata that describes the dataset, and the first line will also contain the names of the headers of the different columns.  Lines starting with `#` tend to be ignored. An example file might look like this\n",
    "```\n",
    "# Monthly transatlantic airtravel, in thousands of passengers, for 1958-1960. \n",
    "# There are 4 fields, \"Month\", \"1958\", \"1959\" and \"1960\" and 12 records, \"JAN\" through \"DEC\". \n",
    "# There is also an initial header line.\n",
    "# And some lines with comments starting with #\n",
    "# Data obtained from https://people.sc.fsu.edu/~jburkardt/data/csv/csv.html\n",
    "\"Month\", \"1958\", \"1959\", \"1960\"\n",
    "\"JAN\",  340,  360,  417\n",
    "\"FEB\",  318,  342,  391\n",
    "\"MAR\",  362,  406,  419\n",
    "\"APR\",  348,  396,  461\n",
    "\"MAY\",  363,  420,  472\n",
    "\"JUN\",  435,  472,  535\n",
    "\"JUL\",  491,  548,  622\n",
    "\"AUG\",  505,  559,  606\n",
    "\"SEP\",  404,  463,  508\n",
    "\"OCT\",  359,  407,  461\n",
    "\"NOV\",  310,  362,  390\n",
    "\"DEC\",  337,  405,  432\n",
    "```\n",
    "\n",
    "We can see the first few lines are comments or metadata, the first line without a `#` is the headers, and we note that text is entered between `\"`s. In this case, the delimiter is a comma. We can read the data as an array with [`np.loadtxt`](https://docs.scipy.org/doc/numpy/reference/generated/numpy.loadtxt.html), telling it...\n",
    "\n",
    "* to ignore lines starting by #\n",
    "* to ignore the first column as it's text\n",
    "* to note that the separator is a comma"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[340. 360. 417.]\n",
      " [318. 342. 391.]\n",
      " [362. 406. 419.]\n",
      " [348. 396. 461.]\n",
      " [363. 420. 472.]\n",
      " [435. 472. 535.]\n",
      " [491. 548. 622.]\n",
      " [505. 559. 606.]\n",
      " [404. 463. 508.]\n",
      " [359. 407. 461.]\n",
      " [310. 362. 390.]\n",
      " [337. 405. 432.]]\n",
      "(12, 3)\n"
     ]
    }
   ],
   "source": [
    "air_travel = np.loadtxt(\"data/airtravel.csv\", comments=\"#\", skiprows=6,\n",
    "                        usecols=[1, 2, 3], delimiter=\",\")\n",
    "print(air_travel)\n",
    "print(air_travel.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "While `np.loadtxt` is quite flexible for dealing with text files, [`pandas`](https://pandas.pydata.org) absolutely shines at working with tabular data. You can find a [pandas quickstart tutorial here](https://pandas.pydata.org/pandas-docs/stable/10min.html) if you are curious about it!\n",
    "\n",
    "Before we go into plotting, we can do some fun calculations (yay!) using our airtravel data\n",
    "\n",
    "**Exercise 2.5.1**\n",
    "\n",
    "* Calculate the total number of passengers per year\n",
    "* Calculate the average number of passengers per month\n",
    "* Can you spot any trends in the data?\n",
    "\n",
    "Hint: Remember the `.sum()`, `.mean()` methods for arrays? "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Space for your solution"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's plot our previous air travel dataset... We'll plot it as annual lines, so the x axis will be month number (running from 1 to 12) and the y axis will be 1000s of passengers. Different line colours will be used for every year. We'll also add x and y axes labels, as well as a legend:\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7fcc031f8588>"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAm0AAADbCAYAAAA/B6b/AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nOzdeXyV5Z3//9d9Z9/3fSeQO2EPO0F2wQWXuutMp24d62hnqs7Qxfl1+q2/+c7UabWtbadabcd21LqiqKAioOyCrAGSXCRkIwvZ9+XknHNf3z/ugIBAAicnOQnX8/HgIblPzrk+CEne57qv63NpUkoURVEURVEUz6aPdAGKoiiKoijKwFRoUxRFURRFGQVUaFMURVEURRkFVGhTFEVRFEUZBVRoUxRFURRFGQVUaFMURVEURRkFvEe6AHczDEMKIUa6DEVRFEVRlMHQLvSAmmlTFEVRFEUZBVRoUxRFURRFGQVUaFMURVEURRkFVGhTFEVRFEUZBQbciGAYxnjg+0D6uZ8vhFjmnrIURVEURblSSSlpbm7GNM2RLsVtdF0nMjISTbvgvoOvGczu0TXAAeBlwHl5pSmKoigDcTpNtu6rorquk/BQP6LCAogK9ycqPIDwED+8vNTNEeXK0NzcTFBQEP7+/iNditv09vbS3NxMVFTUoJ8zmNA2HpgjhOi97MoURVGUi7L1Ofng8xIqazsAqG/uBlpOP65rmhXkwvuDXFgAkeH+RIT6463CnDLGmKY5pgMbgL+/Px0dHZf0nMGEtnXATcCbl1OUoiiKcnFdPXbe3VhMfXM3gf7eXD0/DVufk6a2Xppae2hu7aGts4/mtl6a23oprvjquZoG4SH+ZwW56PAAIsJUmFOUseaCoc0wjI8ACQQArxuG8SRQ238NACHE9W6vUFEUZQxraetlzcZjtHX2ER7ix61XTyA89OszDHa7k+b2Xppa+4Ncf6Br7bDR0t5LS3svJbSe/nxNg7Bga2YuMsz/9AxdZJg/Pt5ew/lHVJRRbe3atezatYuf/exn5Ofn88ADD9De3s6CBQt48cUX8fb25tFHH2Xjxo2n16fl5+djmibf/OY3OXr0KIGBgbz88stMmTLFpVouNtP2+hm/f9mlURRFUZSvqW3o5L1NJfTYHMRFBXLL8gkEBvic93N9fLyIiwoiLirorOt2h0lrey+NrT00t/bS1NZDU2svrR29tHbYaO2wcfzE2a8VFuxrhbnwAKL6A11kmD++PirMKcopUkqeeOIJXnvtNe6//34AHn74YZ577jny8vJYvXo1r7zyCvfddx9CCI4cOYKPz1dfvy+++CLJycm8/fbbrFu3jieffJIPPvjApZouGNqEEH8GMAzjN0KIfzz3ccMwfgn82aXRFUVRrlClVa18uKUUh8MkPSmUGxZnXlZo8vHWiYkMJCYy8KzrDqdJa7uNptae00GuubWHlnYbbZ19tHX2UVrVdtZzQoN8iey/zfrVzFwAfr4qzClXpuXLl5/1cXl5OXl5eQAsW7bsdGiz2WxnBTawdod2dnYipaSlpYWQkBCX67nY7dH/7v/tQ4ZhnPsVGwWsBB53uQJFUZQrzNGSRjbsLEdKmJgZxYq8NLz0oV1/5u2lEx0RQHREwFnXneaZYc4Kck1tvbS09dLe1Ud7Vx/l1e1nPSc40Kc/xJ0xMxfuj7/vmD++WvEAzmcfdMvrej3xx4s+rmkaN9xwA42NjRQVFQGQlJTEjh07mD9/PmvXrqWhoQEpJeXl5SxYsIDW1la+973v8dBDD3HnnXfy7//+70RHR9Pa2sq2bdtcrvliX3F1p+o+4/enVAO/dnl0RVGUK4iUkj2Ha9lxoAaAOVPiWZCbdEl9mlzlpeunA9iZTFPS2mHrXy9nzcw1tVm3XDu77XR226moOTvMBQX4nG5JEhUWQEJM0Ndm/BRlLHnppZd45JFH6OzsJC8vj6CgILq7u7nnnnt48skncTqdLF26lNmzZ/PGG29w11138bOf/YyNGzfy8MMPk5+f79L4F7s9+lMAwzAKhBBvuTKIYRgPAT8C7MBjQBHwNhANvC2EeKL/8x4Efgz0Af8ghNjkyriKoiiewjQln+2p5JBoAGDpnFRyc2JHuKqv6LpGZJi1UQEiTl83TUlbp+2s9XKnNkJ09djp6rGfblMCsGROCjNy4kbgT6CMVQPNiA2njo4OduzYAcCrr75KQUEBvr6+PPXUU6dblFx77bUcPXqUQ4cO8R//8R8AXH311TQ2NiKldOlN2mDmtvcahvFHIOXczx/MiQiGYWQA/whMAWKBT4AjwE+AD4EPDcNYChwFVgNTgUjgIyBn0H8SRVEUD+VwmqzfWkpJZSteusZ1CzPISo8c6bIGRdc1IkKtfnCZhJ++LqWkvbPvdJCrb+5GlDXz+Z4T6JrG9GzPCaSKMlRWr17Ns88+y4wZM3j++ef59a9/zb59+3j88cfZunUrdrudzZs388ADD5Cfn8/HH39Mbm4u+fn5pKSkuDyrPpjQ9hZQDLwKXE6D3ZuAl4UQnUCnYRh3A+8CtwohpGEYb2Gtj4sH1gkh2oF2wzDqDMMwhBDiMsZUFEXxCL02B2s3l1Bd34mfjxc3LRtPSrzrC5JHmqZphIX4ERbix7hk61pSbDCbd1eyeXclGjBNBTdljHnmmWd48MEH6e7u5pFHHmHGjBkALF26lKysLIKCgnj00UfJzs7mX//1X7n//vvJzs4mJCSEF1980eXxNSnlRT/BMIwmIOlyT0QwDOM3gAnMA/yA/wSeFEJM6398JXAP1i3TPiHEL/uvvwa8IITYcjnjnjG+VLlPUZSR0NHVx5qNx2hq7SU40Idbrp5ATMTYXvN1oLCOz/ZYPUaunpfGVCNmhCtSRqOGhgZiYsb+v50L/DkvOB03mJm29bh2IkIAkAgsxrrFeggoPONxiXWmqQQc57l+XoZhLAGWXGZNiqIobtXY0sOajcfo7LYTGebPrVdPIDTYb6TLcrvcnDikhM+/PMHGLypAg6lZY/+Hr6IMh8GEtlDgNcMw/gVoPPOBQZ6I0ALs65+pKzYM4ygw44zHE4FKrB2pk89z/byEEJ8Dnw80uGEYPxlEjYqiKEOmuq6D9zaXYOtzkhgbzM3LxhPgd+W0x5gxMQ4pJVv2VrFxVwW6pjF5QvRIl6Uoo95gvous6f91uTYDDxuG8QcgDggB1vdvPtgKfBP4IVZA+4FhGE8BqUCwEOKCoU1RFMUTFVe0sH5rKU5TkpkSzvWLxuHjfeWdATpzUjxSwtZ9VWzYWY6mwaTxKrgpiisGDG1CiD8bhuED5GLt/iwXQhwZ7ABCiI8Mw5iHtTu0B/gOUIN1TFYU1iaFfQCGYfwOOIzV8uP+S/yzKIqijKhDRfVs2m2915yaFcOyuano+vD1YPM0sybHI5Fs21fNJzus4DYxUwU3Rblcg9mIMBF4H2t9WSMwDuvg+L8RQhS5vUIXqY0IiqK4m5SSnQdq2H24FoC86YnMnZowrE1zPdmew7Vs318NwLVXZTAxM2qEK1I8ndqIcH6DmbP/HfCMEMIQQizAWmv2GvDSZVepKIoyRpimZMPOCnYfrkXTYMX8NOZNS1SB7QxzpiSwIDcRgE92lFFY2jTCFSnK6DSY0DYDePnUB0IIiRXkZlzoCYqiKFcCu93J2s9KOFrSiLeXzs1LxzNF7ZQ8r7lTE5k/PREp4ePtZRSVNY90SYoyKGvXruWHP/whAPn5+cyaNYusrCzuv/9+HA6r6cWjjz6KYRhkZ2eTnZ1NX18fra2t3HjjjRiGwaJFi6itrXW5lsGEtkJg1TnXlgGlLo+uKIoySnX32nlrwzHKqtrw9/Pm9muyGJcSPvATr2DzpyUyb1oCUsJH20oR5Sq4KZ5LSsnjjz/OQw89dPraww8/zHPPPcexY8eIjo7mlVdeAUAIwZEjRygqKqKoqAhfX1+efvpp8vLyEELwgx/8gB/96Ecu1zSY3aPfBdYZhvFdoASr19oc4A6XR1cURRmF2jpsvLPxGK3tNkKDfLl1RVb/mZ3KQOZPs2bbdufXsn5rKRqMmiO9lCvP8uXLz/q4vLycvLw8AJYtW8Yrr7zCfffdh81mw8fH56zPPfPs0WXLlp0V/i7XYHaP7jUMIxNrti0N2AF8SwhR5/LoiqIoo0x9Uzfvbiqmq8dOTEQAt1w9geBA35Eua9TQNI286YlIKdlz+CTrt5ahaRoT0iIGfrJyRfrOttfc8rovLPybiz6uaRo33HADjY2NFBVZ+y6TkpLYsWMH8+fPZ+3atTQ0NCClpLy8nAULFtDa2sr3vvc9HnroISZOnMj69euZPn06a9asoaGhweWaB9s8KBVoAg5iNcGd1n/8lKIoyhWjoqadNz8poqvHTkp8CHdea6jAdhk0TWNBbhKzJ8djSsm6LaWUVLaMdFmKMqCXXnqJ73//++Tm5qJpGkFBQXR3d3PPPfewbt06tm7dym9/+1sOHDjAk08+yf79+5k4cSJbt24lMTHR5fEHnGkzDONZ4F6+voZNAhtcrkBRFGUUKCpt4uMd5ZimxEiP5Jqr0vH2uvKa5g4VTdO4akYSUkr2Hq3jwy2l3Lg4k8xUtS5QOdtAM2LDqaOjgx07dgDw6quvUlBQgK+vL0899RT+/tYSiWuvvZajR4/i4+PDSy+9RHh4ONXV1Rw5MugWtxc0mO843wImCiFmn/NrjsujK4qijAL7jp5k/bYyTFMyY2Ic1y/KUIFtCGiaxsKZycycGIdpSj7YcpzSE60jXZaiXNDq1avZtWsXNpuN559/nttuu419+/axdOlS7HY73d3dbN68mVmzZvHqq6/y+9//HoBf/epX3H777S6PP5jvOiVYx08piqJcUaSUbPnyBFv2VgGwaFYyS2anqB5sQ0jTNBbNSmZGTqwV3D4/TmmVCm6KZ3rmmWd48MEHMQyDG2+8kRkzZjBv3jyWLl1KVlYWc+bM4cEHHyQ7O5vHHnuMjz76iIyMDE6cOMGjjz7q8viDORHheuCvwKdA+5mPCSEecLkCN1MnIiiKcjmcTpOPd5QjyprRNY1rrkonZ5zq5O8uUko+//IEBwrr8dI1blo2noyksJEuSxkh6kSE8xtMy4+fYwW2fYB5+aUpiqKMDrY+Jx98XkJlbQc+3jo3Lc0kLVEFCHfSNI0ls1MwTckh0cD7m0u4edl40lVwU5TTBhPakoBZQogedxejKIoy0rp67KzZeIyG5h4C/b255eoJxEUFjXRZVwRN01g2NxWAQ6KBtZ+V8I1lE0hLDB3hyhTFMwxmTdvPgR8bhhHi7mIURVFGUktbL39dX0hDcw/hIX7cfX2OCmzD7FRwm5oVg9MpeW9zMZW17QM/UVGuAIOZafshEAT8wDCMPqxWHxoghRCB7ixOURRluNQ2dPLephJ6bA7iogK55eoJBPr7DPzEK4SUkm6HnZa+Llps3TTbummxddNi6zr9+wBvX+7NmktykGuNcjVNY/m8VEwpOVLcyHubSvjG8vGkJqgZN+XKNpjQNtntVSiKooyg0qpWPtxSisNhkp4Uyg2LM/H18RrpsoaVzek4K4CdGcpabN0093VjczoGfJ1n8zfz+JRlpAS7HtxWzE9DSjha0sh7m0u4ZfkEUuLVTR/lyjXg7tGhYBjGXiC4/8ODwG+BV4De/ms/EUK8YRjGg8CPgT7gH4QQm4ZgbLV7VFGUCzpS3Minu8qREiZlRnF1Xhpe+tjqwWY3nbSeEcSaT4Wxvq8+7nb0Dfg6fro3EX6BRPgFEtn/3wi/ICL9Agn3DWBN+UEON9cQ5O3LY1OWkRrs+pmiUko27Cjn6PEmvL11br16AslxKriNdWr36Pm5PbQZhuEFfCqEWHbGtXsBLyHEn864FgtsxTqMPhL4SAiRMwTjq9CmKMrXSCnZnV/LzoM1AMyZEs+C3KRR14PNKU3abD2ng1hz36kZsq9CWoe9d8DX8db0cwJZ0NnhzDeIQG+fi/7/sZtO/lC4nfzmagK9fXhs8nLSQlwPbqYp2bCznILjTfj0B7ckFdzGNE8KbWvXrmXXrl387Gc/Iz8/nwceeID29nYWLFjAiy++iLe3N7t27eLb3/42NpuNb33rW/zbv/0bNpuNe+65h71795KTk8Obb75JWNjZu6Hd0fLDVYlA7TnXUoEvzrm2HFgnhGgH2g3DqDMMwxAqcSmKMsRMU/LZnkoOCesA52VzU5meHTvCVX2dKSUd9l6abeeuI+s+fa2trxfJxd9862iE+wWcnhWzQtjZ4SzEx8/lwOqje/GdnKt4sWgHB5uq+OXhTTw2ZRnpIa71t9N1jZV56UgpKSxtZs3GYm5bkUVibPDAT1aUyySl5IknnuC1117j/vvvB+Dhhx/mueeeIy8vj9WrV/PKK6/wrW99i7//+7/nzTffZPz48cyYMYN7772X9957j4yMDNasWcPTTz/Ns88+y09/+lOXarpoaDMMQweeE0J814Ux0rAOmD+EdTv0if5ryw3DeA6r/9sjWEGu6ozn1QDxwHlDm2EYS4AlLtSlKMoVyO4w+WhbKSWVrXjpGtctzCAr3fXZoKFgSsnmGsGBxhO02Lpp7evBKS/eHlMDQn38zz875hdIpF8QYb7+6Nrw3PL11r14KNsKbgeaTvDLw5t5bPJSMkKjXXpdXde4ZkEGUkJRWTNrNh7j1hVZJMao4Ka4z/Lly8/6uLy8nLy8PACWLVvGK6+8Qm5uLhkZGUyaNAmA999/n8jISNavX88vfvELAO666y7uuece94Y2IYRpGMY0wzDmCiF2X+YYXcBLwO+wNjW8BzwLvAzsAP4v8BRWSDtzlasEnBep7XPg84EGNwzjJ5dVtaIoY06vzcHazSVU13fi5+PFzcvGk+whC9u7HX38j9hFfnP1WdeDvP3OCWFfX0vmrXvWpgkvXefvsxfwktjB/sYT/OrIZ/zT5CVkhrp2u0vXNa69ygpuoryZNZ8Wc9uKCSSo4DamPfvnvW553SfunXXRxzVN44YbbqCxsZGioiIAkpKS2LFjB/Pnz2ft2rU0NDRQUlKCr68v1113HeXl5TzwwAOsXr2ayspKkpOTAUhMTOTkyZMu1zyY26MlwAbDMNYBZx0IJ4R4ZBDPLwIOCyEcwAHDMOqAd4UQlQCGYbwCPAd8ydk7VROBykG8vqIoyoA6uvpYs/EYTa29BAf6cOvVWURHBIx0WQBUd7XyfMFW6ns7CfT24e7MWaQHRxHhF4iv13CsYhl6XrrOt40F/JGd7Gus5Lkjn/FPk5cOSXC7bmEGUkqOVbRYwW1lFvHRqp+e4n4vvfQSjzzyCJ2dneTl5REUFERPTw+7d+9mz549hIaGsmjRIhYvXoymaXh7W1+/mqbh5eX6m6vBfDcox5oZu1zfA2KAfzYMIxMIAz4xDONuIcQh4GZgN7ABqxfcU1i3SoNPBTtFURRXNLb0sGbjMTq77USF+XPriixCgnxHuiwA9tSX85fi3dhNJylBEXwnZyExAWNj5shL13kwOw9daHzZUMGvj3zGP01awvgw19YP6rrGdYsykFuhuKKFdz49xu0rs1Qj5DFqoBmx4dTR0cGOHTsAePXVVykoKCAiIoIFCxaQmJgIwIoVKxBCkJSURE1NDYZhUFNTQ2pqqsvjD7jIQQjxUyHET4FfYR0c/9QZ1wbjt8AEwzBKgDeBB4GHgVcNwygEZgBPCyEasG6hHgbeBVxZR6coigJAVV0Hb3xcRGe3naTYYO68LtsjApvDdPL68b38UezEbjqZF5vB96etGDOB7RQvTed+Yz5zYtKxOR08d+RzjrXVu/66us71izIYnxqOrc/J2xuOUdfUNQQVK8qFrV69ml27dmGz2Xj++ee57bbbmD9/PgcPHqSlpQW73c727duZOXMmq1at4rXXXgPg5ZdfZtWqVS6PP+BMm2EYKcCfgXlYIW+hYRj/DdwlhCgd6PlCiE7gpvM89LWmvUKIF4AXBnpNRVGUwSiuaGH91lKcpiQzJZzrF43Dx3vke7C12rr5Q9EOjrc34KXp3DVuJosSxo+6diODZQW3eegafFFfzm+OfMZ3Jy3BCI9z7XV1nVWLxvHhllKOn2jl7Q3HuOMag9hIdViP4h7PPPMMDz74IN3d3TzyyCPMmDEDgKeffpqrrroKgIceeoiJEyeSkZHB3XffTWZmJlOmTOGvf/2ry+MP2KfNMIxPsNabPQW0CSECDMP4IXCNEGKpyxW4merTpihXHiklBwrr+fzLEwBMzYph2dxUdH3kQ1FxWz1/KNxOu72XcN8AvpOzkHEu7qwcLUxp8pfiPeyqK8VH9+K7kxaTHR7v8us6nSYffH6c0qo2/P28uGOlQYwKbqOaJ/Vpc6dL7dM2mLececB/CCFOnTsK1q3S2ZdVoaIoihu1dth4e8Ox04Etb3oiy+eNfGCTUrKxuohn8zfRbu8lKyyWf8297ooJbAC6pvOtCXNZEDcOu+nkt0e3UNji+o46Ly+dG5ZkkpEcRq/NyVsbjtHQ0j0EFSuKZxlMaDsCnDujNg0oG/pyFEVRLo9pSvYdPclf1h7lxMkO/P28WbVoHPOmJY74bcdep50/ip28VbofE8nK5Bwem7KMUF//Ea1rJOiaxjcnzOWq+EzsppPfFWyhoOXc/uuXzttL58YlmWQkhdFrc/D2J8dobOkZgooVxXMMJrT9A/CCYRgfAD6GYfwVWAP8k1srUxRFGaTGlh5e/6iILXurcDhNjIxI7vvGJIyMkW+aW9fdztMHN/BlQwV+Xt48lH0Vt2Xk4jVMzW49ka5p/O34OSyKH28Ft6NbONJc4/Lrenvp3Lg0k/SkUHpsDt7aIGhqVcFNGTsGdfaoYRiBwA1YJxnUAR8LIVzf/jMM1Jo2RRm7nE6TPUdOsju/FtOUBAf6sHxeGpkp4SNdGgAHm6r4H7GLXqed+IBQHp64kITAsIGfeIWQUvLX43vZUluMt6bz8MSFTIlMcvl1HU6TtZtLqKhpJ9DfmzuuMYgK94yefMrgqDVt5zfYt3pTsHZ7ZgLjsBrfKoqijJiTjV288mEhuw7WYJqSqVkx3HvzJI8IbKY0ebf8IL8v2Eqv086MqBR+NP0aFdjOoWka92TOYklCFg5p8nzBNvKbqgd+4gC8vXRuWjqe1IQQunsdvL3hGM1tvUNQsTJcdF2nt3ds/5319vai65c24z6Y3aOPAf+K1aOtAWu27XbgCSHEny6v1OGjZtoUZWyxO5zsPFDD/sI6pISwED9WzE8jNSF0pEsDoNPey0tFOylsPYmGxi0Z01iZlDPi6+o8mZSSN0v3s7lG4KXpPJRzFdOjkl1+XbvDyXubSjhxsoOgAB/uvMYgIuzKW0c4GkkpaW5uxjQvfvbuaKbrOpGRkef73nDBbxaDCW01wNVCiIIzrs0F1gghXJ/HdjMV2hRl7KisbefTXRW0ddjQNJgxMY686Yn4eHvG2ZvlHU28ULiNZls3IT5+fDt7wZC0tLgSSCl5u+wAG6uL0DWNh7KvIjc6xeXXtTucvLuphKpTwe1ag4hQFdwUj+bS7VEbUHXOtYJBPldRFMVltj4Hn+4q5+0Nx2jrsBEdEcA91+eweFaKxwS27SeP8/NDn9Js6yY9JIonc69Vge0SaJrG7Rm5rEzOwZSSPxRtZ3+j6ycZ+nh7ccuy8STFBdPVY+etTwQt7WP7tpsydg1mpm01cB3wDFZ4SwEeAQqBN059nhBij/vKvHxqpk1RRrfjJ1rZuKuCrh47uq4xd2oCcybH4+XlGe8b7f3HUW0/eRyARfHjuTNzJj66Z4TJ0UZKybvlh/ikqgAdjQez85gVk+by6/bZnby7sZjq+k6CA32489pswkP8hqBiRRlyLt0eHUw/NimEGHepVQ0HFdoUZXTq7rXz2Z4TiLJmAOKjg1iZl050hOfsAmzq7eKFwm1UdDbjo3vxN+Nnkxfnkd8KRxUpJWsr8vnoxFF0NB4w5jM7Nt3l1+2zO1mzsZia+k5Cgny58xqDMBXcFM9z+aFttFOhTVFGFyklRWXNfLbnBL02B97eOlflJjE9O3bETzU4U0FLLS8V7aTLYSPKL4iHJy4kNXjk+8KNFVJKPqg4zLoTR9D6g9ucIQpu73x6jNqGLkKDfLnjWoOwYBXcFI+iQpuiKJ6vo6uPjV9UUFbVBkBqQghXz0/3qNtYppR8UlXA2vJ8JJJJEQk8aOQR5OM5NY4lH1Qc5sPKw2ho3GfMY15shsuvaetzsmZjf3ALtmbcQlVwUzyHCm2KonguKSX5xxrYtq+KPruJn48Xi2enMGl8lEe1yuhx9PE/x77gUJO1N2tV6mRuSJ2MfgWfbjAc1lUe5v2Kw2jAvVnzmD8Et6BtfQ7e+bSYk41dhAX7siIvnZT4EI/696ZcsVRoUxTFM7W09/LpznKq6joByEwJZ/m8VIIDfUe4srNVd7XyfOE26ns6CPT24f6sPKZGeXzXo2Eh7TZorQdff7Qw93SxX195hLUV+WjA302Yy4L4TJdfs7fPwTsbjlHXZB0unxATxNypCWQkhanwpowk10KbYRjXApuBCOCfsZrs/kYIMah904Zh7AWC+z88CDwJvA1EA28LIZ7o/7wHgR8DfcA/CCE2Deb1BxhbhTZF8UCmKdlXUMfOg9U4nZJAf2+WzU1lQlqEx/3A/LK+nL8U76bPdJIcFM7DOQuJCQgZ6bKGlXTYrWDWWodsOfXfOutaZ8vpz9OW/S369GVuqeHjE0d5t/wQAH83YQ5XxY93+TX77E4OFNazr6COXpsDgNjIQOZOTWB8arjH/VtUrggu7R79T+AWYAbwav9lf6BOCHHfQCMbhuEFfCqEWHbGtXeBPwEf9v/6BXAU2ArMASKBj4QQOQO9/iDGV6FNUTxMQ3M3G3aWn57hmJgZxeJZKQT4e49wZWdzmibvlB1gU431PWRubDrfHD8HXy/PqnOoSIcd2hqgtf50IJOtddBSBx0twAV+XuheEBplBThAW3kf+uSFbqnxk6oC1pQdBLAOnU9wPbiBFd4OH2tg79E6unrsAESF+TNnagJGeqRHbYJRxrwL/mMbzHeee4HpWF+ty4Hk/hcsHeTgiUDtqQ/6Q9xM4FYhhDQM4y1gJRAPrE4GPNEAACAASURBVBNCtAPthmHUGYZhCJW4FGXMcDhNdufX8uXhk5hSEhLky9Xz08hI8rwzOdv6evhD4XZK2hvQNY07x81kScKEUT/zIp0OaG+Eljpka33/f+ugpR46muBCb+Q1HcJiIDwWLTwOIuLQwmMhIg5Co9B0L8x9G5Bb3kBu+DOmty969twhr/+a5InoaLxddoBXS/YgpWRx4gSXX9fXx4uZk+KZasRytKSRL4+cpKmtl4+2lbHzYA1zpsQzcVyUx/QHVK5MgwltvkA3cD1wQAjRbhhGPBd8y/U1acA0wzAOAb3A94EWIcSp59cAi4FWzj55oQYryJ03tBmGsQRYMsgaFEUZYTX1nWzYWX764O5pRgwLZybj6+N5TWhL2hr4Q9F22vp6CPcN4KGcq8gMdc9aLXeQphPaGs8zY1ZvBTZ5gfMcNQ3CoiE8rj+YfRXQCI1CG2CGUZ+5EtPRh9zxLvKjl5BePmgTZgz5n29FsnWW61ul+3nt+JeYSJYmZg3Ja/t460zPjmXKhGgKS5vZfbiWtg4bn+6s4ItDtcyeHM/kCdF4q/CmjIDBhLYXgS+BWOAxwzAmA38E1g5yjC7gJeB3wGRgJ9YxWKdIwNn/X8d5rp+XEOJz4POBBjcM4yeDrFNRFDfoszvZcaCaA4XWrbOIUD9W5KWTHOd5a8KklHxWc4y3yvZjSsmE0Fj+PmcBYb6e09D3FGma1szY+WbM2hvBvNC3T826lRl+aqbszGAWjebt41Jd+twbMO19yD3rMNc9j37zP6JlTHHpNc/n6qRsdDTeKN3H68f3IqVkWZIxZK/v5aUzeUI0EzOjEOXN7Mmvpamtl827K9mdX8vMSXFMzYrxyDcdytg1YGgTQvzIMIyP+3+/xTCMqcAfgL8Mcowi4LAQwgEcMAzjCJB7xuOJQCVQjRXqzr2uKMooVVHTxqe7Kmjv7EPTYPbkeOZNS/TIWQqb08ErxXvY01AOWKHg1vTpeOkjV6uUJnQ0Q8tXa8tka721dqytAZyOCz85OKL/FuY5M2ZhMS4Hs4FoC24BRx9y/6eY7/8O/ZbvoaW6vET5a5YlGWiaxuvH9/JG6T5MJFcnZQ/pGLqukTMuiuyMSEoqW9mdX0t9czdb91ax5/BJZk6MZXp2LH6+Y3Odo+JZLroRwTAMHdgHzBJCXHDW62IMw/ghECOE+GfDMDKB9Vi3PH+JtfHgY+CHWAFtEzAXSAVeEULMvpwxzxlfbURQlGHWa3OwZW8VR0saAYiJDGBlXjpxUUEjXNn51fd08HzBNqq7W/HTvfm7rLnMHoLzLi+VNJ3II9uRZfn9OzXrLx7MgsK/WlsWHosWEQfhcRAegzbCzX6llMhN/4vM3wI+fui3Po6W5Pras/PZUlvMayVfAnB7Ri4rkoc+IJ4ipaSsuo3d+bXUNnQB4OfjxfScWGbkxBLg795ArFwRXNo9+hKwVwjx/OWMbBhGMPAaMBFoA74H1AGvA1HAy0KI/9P/ud8BVmO1/LhfCLH7csY8Z3wV2hRlGBVXtLB5dyVdPXa8dI350xOZOSluRGesLuZQUxX/I3bR47QTFxDCwzkLSQwKH/Y6ZGUh5uevQ2PV2Q8Ehp616P/0jFl47IgHs4FIaSI//hOycBf4BqDf/i9o8eluGWtrbQmvluwB4Nb06VyTMtEt45wipeTEyQ5259dy4mQHYK2Hm2rEMGtSPEEBKrwpl82l0LYHa7dnHdZmgdOEEO79qhgCKrQpyvDo6rGzeXclxRVWz67E2GBW5qURGeZ568EATGnyQcVh1p84CsD0qGTuy5pHgPfwNvWVrQ2YW9+Ekv3WhdBotLk3oMWmWsHMzzP//w2WNJ3I9X9AHtsLfoHod34fLSbFLWNtP3mcV4p3I4FvpE/jupRJbhnnXNX1Vngrr24HwEvXmJIVw+zJ8YQEeVaTaGVUcCm0Lb7QY0KILS4UNSxUaFMU95JSUnC8ic+/PIGtz4mPt87CmclMM2I8tj1Gp93GH8VOClpq0dD4Rvo0runfkThcZF8Pcvd65P4N1i1QHz+0uavQZqx0+5qz4SadDswP/htKD0FgqBXcIhPcMtbOulL+cuwLJHBT2lRWpU4e8DlDpa6pi935tZRUWvMbuq4xMTOKOZPjCQ/1H7Y6lFHP5RMRAoCrsVpwvAH4CyHqh6w8N1KhTVHcp73Txqe7KqiosWYY0hJDWTE/zaMP367sbOb5gm002boI9vbj29kLyImIH7bxpTSRBbuQ29+BrjYAtIl5aFfdihYcMWx1DDfpsGOufQ4qCiAoHP2uH1i3fN1gV10pf+4PbjekTuHGtKHfvXoxjS097D5cy7HyZqS0OqlkZ0QyZ0oCUeGje+ZUGRYuzbQtxTpyKh+Yj7Xzcxvwt0KIT4awSLdQoU1Rhp6UkoNFDWzfX4XdYeLn68XSOSnkjPOsA97PtePkcV4r+RKHNEkPjuQ7OQuJ9B++zRGypgTzs9ehrsy6ED8Ofek9aAmuH4A+Gki7DXPNr6D6GIRGod/1Q7SQSLeM9UV9GS+LL5BIVqVO5sbUKcP+b7OlrZc9R2opPN6M2f+zdkJaBHOnJhAbGTistSijikuhLR/4sRBirWEYPUKIAMMwrgZ+LYQYngUDLlChTVGGVnNbDxt2VlBTbx3wnpUWwdK5qR698NpuOnnj+D62nSwBYGH8eO7KnImPPjw9tmRHM3LbO8iiL6wLQeFoC29Hy5mLpnnmBg13kX09mG8/CydLITwW/c4foAW7Z+PHnvpy/iR2IZFcnzKJm9KmjsibirZOG3uPnORIcSNO0/qZOy45jDlTE0iMCR7g2coVyKVjrFKAUzNqpxLeVqyTDhRFuUI4TZO9R+r44lANTlMSFOBz+oB3T9Zs6+KFwu2UdzThren8zfjZLIjPHJaxpb0Pue8T5J714OgDL2+0Wdegzb4ezffKXOOk+Qag3/oY5ls/h4YTmO88g37H99ECh77Z8pzYdDRN409FO1l/4igmkm+kTRv24BYW7MfyeWnMnZrA3qN15IsGSqvaKK1qIzUhhLlTE0mOC/boWWrFMwxmpu0toBD4CdAlhAjsb81xuxBixTDU6BI106Yorqtv6uaTnWU0NPcAMHlCNItmJuPv59kNRYtaT/Ji4Q46HTai/IL4Ts5C0tx0O+5MUkoo3ou59S1ob7IuTpiJvugOtLDRcxyWO8nuDsy3/guaaiAmBf2O1WhuulW9r6GSl4p2YCJZmZzDrenTRzQgdffY2V9Yx8Gievrs1pFiSbHBzJ2aQFpiqApviku3RyOBPwOLgGCgHigH7hFClA9ZiW6iQpuiDJ5pSrp77XR09Z3+1dTWy9GSRqSE0GBfVsxPJy0xdKRLvag+p4NPqwv5oOIIEsnE8HgezF5A8DD0NZP1FZif/RWqi60LMSnoS+5BSxm6I5bGCtnZivnm01YT4fhx6Lc/geamI8P2N1byYtEOTClZkZTNbRm5Ix6Oem0ODhTVs7+gDluf1b8+LiqQuVMTyEwJH/H6lBHj2u5RAMMwYrFuidYLISqGqDC3U6FNUSxSSnpsjjMCmZ2Obuv3nf3XOrvtpxdMn2tGTiwLcpPw8eCzFp3S5Iu6Mj6oOExLXzcA16dM4sa0KehuXjsmu9utg9IPbwMkBASjLbgVbfJCNA9tLOwJZHuTFdzamyApC/3Wx9zWNPhA4wn+ULQdU0qWJxnckTHDI4JRn93JIVHPvqN1dPdaJ2BERwQwd0oCE9Ii0PWRr1EZVpce2gzD+O+BXlUI8YgLRQ0LFdqUK4GUEpvdeTqQdXb1z5Z1fzVj1tHdh9M58Ju0AH9vQgJ9CQk69cuH1IRQjz2CCqw//6GmKt4rP0Rtj9V+JDkonNszZri9nYd0OpAHNiG/+AD6ekD3Qpu+DG3eTWj+aofgYMjWesw3noauVkibiH7zP7mtV92hpipeKNyOU5osSzS4c5xnBDcAu8PJ4eJG9h45SWe3HYCIUD/mTEkge1ykx54qogy5ywptPxnoVYUQP3WhqGGhQpsyFtjtzjMC2HkCWVcfdoc54Ov4+XpZQSzwq0AWEuRL8BkhzRMPc7+YY231vFt2kNIO65zTaP8gbk6bxqyYNHQ3/jCWUkJZPubnb0BrnXUxYwr64rvRIoev79tYIZtrMd/8L+huh3HT0G98BM3LPWsm85uqeaFwGw5psiQhi7szZ3pMcANwOE0Kjjex53At7Z19gLU0YfbkeCaNjx51X6PKJXP99uhopUKb4ukcTtO6Pdl9RiA7J5SdWu9yMT7e+ulAFnxGIDtz1szXg29tXqrqrlbeLT/I4eYaAEJ8/Lg+ZTKLEsbj7eZWHrKpxgprFUesCxHx6IvvQhs31a3jjnWy4YS1q7S3Cy1rFtr1D6G56e/ycHM1zxdYwW1xwgTuzpzp9lvol8ppmoiyZnbnn6SlvReAoAAfZk+OZ0pWND7eY+frWTnLZc209fBVi4/zEkJ4/Ny/Cm2KJzBNSXlNG02tvV/dwuwPZafWsFyMl65Zs2HBp0KYz1e3L/tDmZ+vl0fNFrhLY28nH1Tks7u+HAn4eXmzIimHFUnZ+Lv5+CfZ24Xc9T7y4GaQJvgFoM2/GW3aUrfNCl1p5MlyzLd/AX091kkR19zvtl52R5pr+H3BVhzSJCUogjszZ5IV5p5TGlxhmpLiyhZ259fS2GLt4A7w92bmxDhyc+Lw8fassKm47LJC24B92EbDhgQV2pSR1NPr4HBxAweL6k+vUTmXpmEFssD+W5XnhLGQIF8C/b2viEB2MR19vaw/cZSttcU4pImXprMoYTzXp0wm1M09z6TpRB7eitzxHvR2gqahTVmElneLW/qLXelkTQnmO8+C3YY2dTHa8r9z27//otaTvHzsC1ps1saVmdGp3JaRS9QwnpQxWFJKSk+08UV+DXVNVr1R4QGsWjSO6Ah1PNYY4vLZo+nAjUAc8GsgXghxeKiqcycV2pSR0NjSw4HCOgpLm3E4rbVmEaF+pCeFfW2GLCjAR+0Ou4hep51N1UVsqCqk1+lAA2bHpHNT2lRiAtzfTV5WFmJ+/jo0VlkXkg3r6KmYFLePfSWTlYWY7/4anHa0GSvQFt/ltuDW53TwSVUhn1QVYDed+OherEzK4ZqUifh54AyqlJKKmnY+21NJS7sNby+dJXNSmDIh+op/czdGuNSn7U7gv4EPgHuAicAO4AdCiL8MtgLDMHRgJ/B/gE7gFaC3/+GfCCHeMAzjQeDHQB/wD0KITYN9/YuMq0KbMixMU1JW1cb+wjpOnOw4fT09KZTcnDjSVdPMS+IwnWw7eZz1lUdot1vfKiZHJPCN9OmkDMPB6rK1AXPrm1Cy37oQGo2+6A6Y4FmL1scyWXYYc+1vwHSizV2FvuBWt47XbOtiTdlBvmywbiKF+wZwa8Z05sSke+TfeZ/dyWe7Kzl63GrgnJUWwdV5afj7el7QVC6JS6GtBLhXCLHjjLNHZwJvCSEGfcqxYRiPA/8f8LdYM3ZeQog/nfF4LNbxWHOASOAjIUTOYF//IuOq0Ka4la3PwZHiRg4WNdDWaQOsTQETM6PIzYklMkzdtrgUppTsbahgbUU+jb3W+abpIVHcmj4dIzzO7ePLvh7knvXIfRvA6QAfP7Q5q9BmrnRbGwrlwmTxfswPfw/SRFtwC/rcG9w+ZklbPW+U7qeysxmAcSHR3JU5k/SQKLePfTkKS5vYuKsCu8MkNNiX6xeNU2eajm4unT0aDhzo//2phHcUGPS/XsMwMoAVwIf9l1KBL875tOXAOiFEO9BuGEadYRiGUIlL8VAtbb0cKKrnaEnj6XYbocG+5GbHMmlCtHq3e4mklBS01vJu2SFOdLUAEBcQyjfSp5Eblez2mQ4pTWTBLuT2d6CrDQAtZz7aVbehhXj2+apjmTZhBtp130aufxG5411Mb1/0mSvdOub4sFh+NP0adtWV8l75IUo7GvnPg58wP24ct6RPI8xNpzZcrpxxUcRHB7F+ayl1Td288VERC3KTmD053iNnCJXLN5ifKu8CvzUM4/sAhmH4Y82YbbiEcX4DPA78qP/jNGC5YRjPAfuAR7CCXNUZz6kB4gEV2hSPcWotyf7COsqr209fT44PYUZOLOOSw9X6tMtQ3tHEmrKDiDar31m4bwA3pk1hftw4vIahDYOsKcH87HWoK7MuxGegL/0btIRB30xQ3EjPnovp6ENueBm55Q0ruE1b4t4xNY0F8ZnMiE7loxNH2VRdxK66UvY3VnJ9yiSWJ2Xj4+bWMpciItSfu6/LZvv+avYV1LF9fzWVte1ct3AcQQFqhnisGExo+0fgF1jnjfoDrcB7wMODGcAwjAeA3UIIYRinz97bD7yMtTbu/wJPYYW0M3sfSOCCzakMw1gCLBlMDYriKrvdScHxJg4U1dPcZq2v8vLSyBkXRW52LDGRHt/9xiPVdbfzXsUh9jeeACDQ24drkyexNDEL32FYAC47mpHb3kEW9U/8B4WhLbwdLWee29pMKJdHn7wQ02FHbn4Vuel/Mb190CctcPu4Ad4+3JoxnaviM3m77ACHmqp4t/wQ204e5/aMXKYPwyzwYHl56SyenUJqQigfby+jsraDv7x/lGuvyiAjKWyky1OGwGDWtM0QQuzv/30M0CyEGLjT51fP/wswDzCBBKAd+K4QYm3/4xOB54D/ASYLIX7Uf/0zrLV0lZf8pzp7fLWmTblsbZ02DhbWc6Sk8XSD2+BAH6ZnxzJlQjQB/uod7OVotXXzYeURdpw8jonER/diaWIW1yZPIsjH1+3jS3sfct8nyD3rwdEHXt5os65Bm309mpvbhyiuMfd+jNz6ltV25frvoBuzh3X8wpaTvFm6j5pu6xa6ERbHXZkzSQoKH9Y6BtLZ3cdH28pOb4qaOSmOq3KT8FKnKYwGLm1EaMCaXXsDeF0IceRyqzAM42XgdeCXwN1CiEOGYfwICAZ+BWwC5mLdKn1FCOHyV6MKbcqlklJSVdfJgcI6jp9o5dSXSEJMEDNy4hifFq7OALxM3Y4+PqkqYFO1wG460dBYED+OG1KnEOHn/tlKKSUU78Xc+pZ1QDnAhJnoi+5AC4tx+/jK0DB3vY/ctRZ0L+u4q8zpwzq+U5psrS3h/Yp8uh19aGgsShjPTWlTCPbxnNBvmpIvj5xk58FqpIS4qEBWLRpHeKjn1Kicl0uhTQeuAm7q/+XACnBvCiEKL6WKM0JbD/A7wAs4AjwohGg3DOM7wGqslh/3CyF2X8rrX2BMFdqUQXE4TYpKmzlQWEdDf9dxXdcw0iPIzYkjPtrzmm2OFnbTyWc1x/joxFG6HdZZirlRKXwjfSrxgcNz20bWV2J+9leoPmZdiE62+q2lZA/L+MrQkVIit7+D/PIj8PJGv/kf0dInD3sdnXYbH1QcZmttMSaSQG8fbkydyuKECR71xq6mvpN1W0vp6OrD10fn6nlpZI/zzJ2wCjCUZ48ahrESa2PBeCGE56zCvAAV2pSBdHb3caiogfxjDfTYrGWVgf7eTDVimJoVQ3Cg+2/XjVWmNNlVV8YHlYdPd5zPCovllvTpjAuNHpYaZHc7cse7yMPbAAkBwWgLbkGbvAjNg36wKpdGSon87K/Ig5vA2xf9lsfQUoyBn+gGNV2tvFm6n8LWkwAkBIRyR+YMJkUkjkg959Nrc/DprgqKK6yd2ZPGR7FsTio+Y+g84jHEpZk2DZgPrOr/lQy8j9Wn7aMhLNItVGhTLqSmoZMDhfUUl7dg9n8dxEYGkpsTi5ERibda+3HZpJQcaq7mvfJD1Pav/UkOCueW9OlMikgYloXb0ulAHtyE3PUB9PWA7oU2fRnavJvQ/NXGkbFAShP56V+QR7aBjx/6bf+Mlpg5QrVI8pureat0Pw39/QWnRCZyR8YM4gJDR6Smc0kpOXyskc++rMTplESE+rNq8Thi1UYqT+NSaGvE2kTwPvAWsEkIMfAJ1x5ChTblTE6nybGKFg4U1nOysQuwzv4cnxrBjJxYEmODPWYn2GhV3FbPu+UHOd7eCECUXxA3pU9lTkw6+jD9v5WNVZjrXoCmGutC+hT0JXehRSYMy/jK8JGmifz4j9YOYL8A9NtXo8UNeHS229hNJ5trBOsrj9DrdOCl6SxLzGJV6mQCvD1j1r6xpYd1W47T1NaLl66xaFYK07Nj1Pc+z+FSaFsBbD53x6hhGFNGw/mjKrQpAN09dvKPNXBINNDVYx3c7ufrxZSsGKYbMYQG+41whaNfdVcr75Yf5HCzFZSCvf1YlTqZhQnjh62flZQSeWQbcvNr4LRDeBz6krvRxk0dlvGVkSFNpxXSi/eBfzD6nd9Hi04a0Zra+npYW36InXWlSCDEx49vpE8jL24cuge0k7E7nHz+5QkOH7PeXGWmhLMyL50Af9UU/Hya23o4UFhPfHQQk8a7fWmHS6EtGLiXs09A8AEeF0J4/DkZKrRd2eqbuzlQUEdRWTNO0/q3HhXmT+7EOHLGReLjrdZzuKqpt4v3K/LZXV+GBPx0b1YkZ7MiKQf/YTz2Sdp6kBv/F9m/f0mbtABt2d+i+ahAfiWQTgfm+7+FssMQGIp+1w/QIuJHuiwqOpp5o3Qfx9sbAEgJiuCuzJlMCIsd4cosx8qb+XRnBTa7k+BAH65fOI7k+JCRLssjSCkpq27jQGE9FTVWM/WoMH++dfMkd89KuhTa3gJisQ53DwdKgWuAR4UQfx3CIt1ChbYrj2lKSk60cqCgjur6ztPXxyWHkZsTR2pCiLoNMAQ67b2srzzKltpiHNLES9NZGD+eVamTCB3mY35kXQXmuuehtd46K3T536FPnD+sNSgjTzrsmO/9GioLITgC/a4fooUNz4aXi9bVf57uO2UHaemzNuTMjE7ltoxcovxHfld6W6eN9VtLqW3oQtNg3tRE5k5NuGJPd+mzOzla0sTBojpa2q3zpL29dHIyI8nNjiM6wu3f31wKbS1Yx04lAD8XQtxkGMYNWH3WvjmkZbqBCm1Xjl6bg8PFjRwsqqejy2or4eujM2l8NNOzY4lQvYmGRK/TzqbqIjZUFdLrtJa3zolJ46a0acQEDO/ku5QSeXAzcuub1uHuMSnoqx5Gixz5GRZlZEi7DXPNL6G6GMKi0e/8ocecHdvndPBJVSGfVBVgN5346F6sTM7hmuSJ+A3DCSAX4zRNdh2sYc9hawdsclww1y0cR0iQZ6zDGw6tHTYOFtZxpKSJPru1IiwkyJfp2TFMnhBDgN+w/R25vBEhC+gEjgohJhiG4Qc0CiE8fg5Vhbaxr6nVWmtQUNqEo//g9vAQP6bnxDIpMxo/X3ULdCg4TZNtJ0tYV3mEdrt1lNekiARuSZ9OSvDw/1CUvV2YG16Gkv0AaNOWoi2+C20Yb8kqnknaejDffsY6SzYi3lrjFuQ5xzg193axpvwgXzZUANZZu7dl5DI7Jm3E7wJU1rbz0bYyunrs+Pt5cU1eBpmpnnXaw1CSUnLiZAcHCus5fqL19PWk2GByc2IZnxoxEjOOLoW254BFwA3Aq1jnhfoBK4UQU4awSLdQoW1sOr3WoKCeitqvDm5PSwgld2IsGUlhI/7Nb6wwpWRfQwVrK/JPtzJID4ni1vTpGOFxI1KTrDmOuf4F61QD3wD0lfehZc0akVoUzyR7OjHf+jk0VlmNlO9YjTbMM8EDKWmr543SfVR2Wr3TMkOjuXPcTNJDRrbxbXePnY93lFFebX1vnZ4dy6JZyWOqDZLd4aSwtJkDhfU0tVrN1L10DSMjktycWOKiRvS2tct92q4DdgNhWIe7ewP/vxDi6BAW6RYqtI0t7Z02SipbOVhUT2vHV2sNJmZGkZsTS1T48K6lGqs6+nopbq/nWFs9BS0nqeuxvnnHBYTwjfRp5EaljEgoltJE7tuA3L4GTCfEZaCv+g5auDqCSvk62d2O+eZ/QXMtxKWh3/4vaMNwXNqlMKVkV10p75UfOj2DPT9uHLekTyNsmNeGnklKyb6COrbvr8Y0JTGRAaxalElk2OheZtLR1cfBonoOFzfQa7NugQYF+Jxuph4U4BEz9ZcX2vqPsHpOCPFdd1Q1HNwd2qrqOqiu6yQtMZTYyMArduGmuzhNk5r6Tsqq2iirbj/9jghOrTWIZfKE6OFcazAmtff1cKzNCmnFbfWnD8M+Jcw3gBtTp5AXPw6vEWpXIHs6MD/+o7U7ENBmrkS76ja0EV4LpHg22dmC+cZ/QVs9JI5Hv/VxNF/PCx49DjvrTxxhU7XAKU38vLy5PmUyy5OMYWuZcz4nG7tYt7WUtg4bPt46y+amMjEzalTdyZBSWs3UC+oprvx/7d13eJzXdeD/7zvovZdBZwEuqyg2kZRkURLVlcgpayeOY3kTxZbtzf78ixPv2ptN7M1u9nmSnx1vkp83LptE7o4lO5GtRpqUKKqRYhcbLgmCBIjeexlg5u4f98UAoMAGYDAzwPk8Dx8A7zuY92IIzJy5955zuoP9pAtyktm4ugBVkUVMZM0izmmm7Q3gT+ajD2g4hDpo23ewjpPapnInxMdQ5k2nzJtGuTddmvLO0sCQj8uNfVxq7KWuqS+4IRQgLtZDmTed1StyWFmaKUHyLPX6hjnf0xoM1FqG+6adj/PEsDwtl8qMfFRGPsvSc8P6wmEaNIEXvwWDPZCYgufh31/wJuEiepm+DgL/8lfQ3wWlq/D82mdx4iJzg33rcB8/rT3Oya5GAHITU/nQso1syCkJW6A06vOz72Ad1Ze6AFi1LJtd28sjfr/wuD/A+cvdHD/XSmunzdp1HKgqt/2kvXkpkRp8zilo+2fgN4AXgZ6p57TWn5mP0YVSqIO2uqZeLtT1UNfUR+/A6LRzGanxlHnTKS9Kp7QwXYoWXkMgYGjuGOBygw3U2rqGpp3PzkhkWXEGy0oyKM5PjbR3RFGhe3SI872TQVrbcP+08/GeGFak5wWDtPK0nLAGaRNM+1TKBwAAIABJREFUIIB590XMO8+DMXam5PGncdKywz00EWVMdyuBn/wVDPbaDhlP/IeITlo5293MT2qPBdvArcos4MPLN1OcEp6kAGMMZy92su9QPePjATLSEnj8nuUU5oa/ZMnVBofHOKnbeE+3MzRiM9wTE2K5rSqXDSo/GjJi5xS0fekap4zW+i/mMqqFsJB72nr6R6lv7qO+qY/6lr7gevmEgpzkYBBXlJ+6qDZ13qqh4TEuN/VxqaGXy029jPomH6vYGA9l3jSWFWdQUZJBhnQruGWdI4Oc723lQm8bureNjpGBaecTPLGsyMijKiOfqox8ylOziY2AIG0qM9BD4OVvw5VqwMG54zGcOz+IE2HjFNHDdDbZwG14AFZuwvMrn4ro3ye/CXCg+QI/rzvF0LgPB4d7vCt5ovw2UsNUNLqrd5gXX6+lvXsYj8fh7k3FbF5TEBEzVq2dgxw724a+3EXALaaem5XExtX5rFqWQ1xs1Lzmzilo+xOt9VdmOP5ZrfXfzsPgQipciQiBgKGta4j65j7qmvpoahsIVuQHiIlxKMlPo6wonXJvOnnZSRHxSx8qxhhaOga51NjL5ca+YN/PCRlpCSwvyWBZcQYlhWlLOqC9VcYYOkcH0T02SDvf20bn6PTHNzEmlpXp+cEgrSw1mxhP5D7Gpu4MgZf/Dwz12er2jzyFU7Eu3MMSi4Bpq7dZpaNDOKu24TzyBzgR/LcAMDA2yi/qTnGg+QIBDMmx8fxq2Xp2eivD8nc87g9w4EgDJ6rbAKgoTueRu5aRHIZN/P5AgJp6W0y9qX3yeW9FaSYbV+dTWhiVxdRvPWhTSv0n99O/BP70qtO5wGekjdXNGxv309g6YIO45j7au4annU9KiKXMOxnELYZemMOj49RNmU0bdqepwaZWlxSmBZc9b6XwbetwH8ZAZkISiTGRu7wRKsYY2kcGpi13do9OX1JOioljZUYeKqOAyox8SlOzwpZAcCtMwI95+3nMuy8Bxu4/evQTOKmLt06UWHimuZbAT78KvhGcdR/AefBJnCj4+2gc7OHZ2mOc67EFcL3JGTxeto7NuaVh6WdaU9/NnrcvMzLqJyUpjkc/sIwyb/qCXHt4ZJxTF9o5Ud3GwJDbTzouhnWVtph6RlpoXkMDxuAJfRA4q6Dtn91PnwS+e9XpcWCP1vrZmx2Bm4n6NvBl4DzwHDb4e05r/Tn3Nk8Bfwb4gE9rrffd7P1f57oREbRdbWh4jPqWfuqa+qhv7gtW8J+QmZ5AuTedMm86pd40EuMjfz+cMYb2rmEuNfZwqbGP5vYBpv56pafEs8ydTSstTCMu7taWJfp9I/zo4hGOdtQHjyXGxJIZn0xGfBKZCUlkxieTGZ807euM+MSIW/q7FcYYWof73QDNzqb1+KYH/cmx8VRmTM6klaRkRkRT6lth+rts0++mGnAcnB0fxLnj8YifBRHRyTSct50Txn04G+7Hue8jUfG7Zozhva5Gnq09FqybmJuYykPFq9lRsIz4Bc6m7h/08dIbtTS22rHcsb6QHbcXhWwGsL17iOPn2jhX24nfb19gstIT2Li6gDUrcoi/xdeVmxEwAU52NrK3sZp12UU8Wrp23q9xlTktj35ea/3/zXUESqk/Av4r8FHgaeCfgBfcf18BzgAHgDuAbOBlrfXqebhuRAZtUxlj6Okbpc5dSr3S0j8tY9JxoCAnhXJ3Fs6blxIxm/FHfePUNffb2bTGXgaHx4LnPI5DcUFqcDYtOyNxVtPUxhiOdNTz45ojDIyPEu+JISM+iR7fMGMB/43vAEiLS7CBXHwymQlJk58Hg7skUuMSF+Id1A0ZY2ge6nOXOu1s2kT9pgmpsQnBIK0yI5/ilMyIGPtsmYsnCOz+JxgZhJRMPI9/EqdEhXtYYpEzdWdtr1L/OGQV4Gx+GGfNnRGdoDBhLODn7dZa9jScC+5ZTYtL5P6iKnZ6q0hZwOzYQMBw6L1mDr7XhDHgzUvhsXuWz9t+5EDAcKmhl2PnWrnSMplEVVGczsbVBVQUpYdkCXRkfIy3W2vZ16SDj7E3KZ0vbX48chvGzwel1DLg60A78BPgH4ByrbVRSv17QAHvAVu01n/sfs9+4Gk9x4grGoK2qwUChpbOQeqb7FJqc9sggSn/T3GxHkoK0mxpkaIMcjJnFwzNhjGGzp4RLjX2cqmxl6bWgWljS0mKC86mlXnT55wS3usb5oc1hznR2QDYDKqPVW4jNzEVYwxD42P0+obo8Q3bf6PD9PiG6A1+PUSfb4QAN/499zgOGXFJU4K6JDIT3MBuymxeUkzcvD7eAWNoHuoNBmgXetvpvypIS4tLoMpd6qzKyMebnBHVQdoE4x/HvPEc5tgv7YGK9Xb/WnLEd8gTi4S5fJrA3u9BX4c9kJKBs/EBnA33Rlwh3pkETIBjHVfY3XA22FkhwRPL3d4VPFC8iuyEhcvubGjp56U3ahkYGiMhLoYH7yynqmL2md6jvnFOX+jgRHV7sDpDXOxkMfXsjNAUH+4aGeTVpvO82VLDsN9OROQmprCraBV3FiwnMfRBfdiDtheAPwa+CLwGfE5rvcE99xDwEaAa8Gmtv+Ye/yHwTa3163O8dtQFbVfzjflpaO0PBnGdPdNf0FOS4twAzpYWme90Zt+YnyvN/cFAbepSruNAUV5qMFDLzZqfhApjDIfaL/OTi0cZHPeRGBPLby7bxAcKV9zy/QdMgP6x0WBA1+MbpnfUDeomArzRYQbGR298Z9jyGHYJNjk4S5fxvgAv8ZrLFAFjaBzsmRakDV517fS4RLvUmVlAVUY+hUmheScZTqan3S6Htl4CTwzO3b9hC+ZG2bKuiH4m4Mfow5gjr0D7FXswPhHntntxNj2AE4beurfKGEN1Tyu7G84G97x5HIdteRU8VLKaogUqFTI8Ms6ety8H+3jeVpXHzq2lt5S52d07wvHqNs7UdDDm9pNOT41n46p81lbmhmy70KW+DvY2VnOs40rwjf7K9DweKF7Fhpzihdxycs0n+5Avfiulfh84pLXWSikAg90TN8EA/uscv9b93gvcO8/DjUjxcTEsL8lkeYn9oxsY8lHfPLkfbnB4jHO1XZyrtYUPczISgwkNJYVpt7zGb4yhu2+Uy429XGropaG1f1rma3JiLBXukme5N53Eee5G0D06xA9q3uVUVxNgm5L/buUds37H6HE8ZLiBVTnXftc3FvAHA7he32SAN+3r0WFGA+O0jQzQdlUZjaslx8a7QVwSGQnJpMcl0jLUy4W+dobGr9rDGJ/k7kezQVp+UlRmPN00c/6IbfbuG4b0HDyPPY1TtCLcwxJLlOOJwVm9HbNqG9SdIXD4ZbhSjTnyCub4XpzVO3C2PIKTXRjuoV6T4ziszipkdVYh9QNd7Gk4x5H2et5pu8Q7bZdYn13EIyVrWJGeF9LnlqTEWJ64bwUnqts5cOQK751vp7Gtn8fvWUFu1rVnxowx1DX1cexca7DnKUBJYRqbVuezvCQ0xdT9JsCJjgb2NVVz0Z1t9TgOd+SWs6t4Vdj7wF7teokIr2qt75/4ONsLKKW+C2wHAoAX6APQWpe65z8OlAMXgXVa6y+6x18DPq61rp/pfm/h+lE/03Y9E8uVE1mpDS39wXcmYH/5vHkpwSCuMDdlxl/8sfEADS3ubFpD7/sKBRfmpgRn0wpykkPyR2+M4e3WWp6tPcawf4zk2Dg+tHwzO/KXRVQAMzx1SXbqjN202bsR/CZwzfvITkh296MVoDLyyU1MjaifMVTM+Bjm9R9jTu63B1Zuss3eEyOvQKdY2kzLJRu8XTiGnUNwYMXteLY+GjVvMNqHB/hl4znebq0N7v9dnpbLw6VruC27OORbLNq6hnjx9Vq6+0aIiXG4b2sZ66typz3X+cb8nLvYyfHqNrp67SpSTIzD6uU5bFyVT152aJaoh8fHeKv1Iq826mCJpOTYOD5QWMl9RVVkhXdpfFbZozVACzbg2jPTbbTWj93KKJRSzwA/Bj4F/C028eAV4AtAPbAP2AaUAd/XWm+9lfu/xjUXddB2Nb8/QHPHYHAWrqVjcFoGZ3xcDKWFaZR50ynMTaa1c4hLDb3Ut/QFM3EAEhNiqChyZ9OK0klODO0aftfIIN+reZez3c0A3JZdzEdXbiUzCvaUzCRgDINjo9OWYHt9w2QlpFDlBmlLjelqIfDiN+zyU0wszj0fxrn9/iURrIroZbpbMUd3Y868ZRMWAIqr8Gx9BJbdFhW/v/2+EV5rOs9rzeeDs/yFSek8VLKaO/IrQtr9ZGzMz6vv1nOmphOAyvIsHryznFGfnxPn2jh9oYNRN/EuNTmODSqf26pySQrRa07HyACvNmnearnIiPv/mZ+Yyv3Fq9hRsCxSykjNKmhbjc3k/EfgqZluo7X+zq2MYkrQdtH9mAM8o7X+snv+aeDz2JIfvzcf/U6XWtB2tVHfOFeCpUX66e4bueZtC3KS7bJnccY1Z+TmW8AY3myp4aeXjjPiHyclNp7fXrGFrXnlUfFkKG5O4Ow7mH3fg7FRyMzH8/incArKwz0sIW6aGezFHN+LOfkajLold3KKcbY+gqPuwFngUhuzMeIf482Wi+xtrA7WdsyMT2JX8So+ULiSpBBusK+u7WTvwTp8YwESE2IZ9Y0HJxS8eSlsWl3AyvLMkJQKMcZQ22/3qx3vaMC4+9WqMvJ5oHgV67OLIq1E0pxKftyjtT7gfp6qtb7+Rp4Is9SDtqv1DYza/XDNfbR1DpKXlcyykgwqijNIWeBq1u3DA3zvwiF0bysAm3JK+cjKLaTHhyYjSCw8MzaKefUHdpYC7IvbA0/iJMj/sYhOZnQYc+qAzXh2szVJy7ZJNOs+gBN/84XCw8UfCHC4vY7dDWdpcnubJsXEsdNbyf3FiowQPQf39I3w4oFaWjuH8Hgcqiqy2LS6IGT9S/1uZu3exmou99uZPo/jcEee3a9WlhqxPYznFLRlAv8/8OtAIrZp/DPAn2utB6/zrRFBgrbIEzCG/U3n+dfLJ/AF/KTFJfCRFVvZnFcW7qGJeWQ6Ggi88A3oaoaYOJz7f8e+qMkMqlgEjH8cc+6gzTjtsts6SEi2S/4bd+EkL0xngLkwxnC6u4ndV85xoc+2pIp1POwoWM6DJasoSJr/n8HvD3C5sY+C3GRSk0NTS25o3MebLRd5tUkHZxRTYuO5x1vJvd7KaNh2M6eg7bvYLM4/xdZZqwL+HBjXWn90HgcZEhK0RZbW4T6+e/4QNX3tAGzNK+e3lm8mLQrenYqbY4zBnH4D8+oPwT8G2V7bmDu3JNxDE2LeGROA2vds0kJTjT0YE4ez7m5brDczL7wDvEm1fR3sbjjLyc6GibQLNuaU8nDpmojLoLyW9uF+u1+ttZZRd79aQVIau4pXsSN/4btFzMGcgrZOYLnWunfKsUygXmsd8W8lJGiLDAETYF+j5vm69xgL+EmPS+SjlXdwe468kC8mZnQYs/d7GHc7qrP2Lpz7P4oTF/29dIW4EdN4wQZvtSftAcfBqdqCs/VRnPzo2MPZMtTHnoZzHGq7xLibAV+Vkc/DJWtYm+WNuJlyYww1fe3sbawOBpxgC7E/ULyKtVlF0ViIfE512jqx2ZynphzzAr0z31yI6ZqHevnO+YNccvcUbM9fxoeXbyJFXsgXFdNaZ7NDe9ogLgFn18fwrNkR7mEJsWCc4kpiiisxHY22xlv1IVu0Vx+G8jV4tjwKZasjLvCZqjA5nSertvFE+Xr2NWkONNe4fY/bKEnJ5OGSNWzOKyMmzBv3/YEARzvq2dtYTd2ArVEa63jYml/BriJFaRQURJ6Nm5lpexL4K+CbQA1Qgi3Z8fda66+GfIRzJDNt4eM3AX7ZcI5f1J1i3ATIjE/idyvvYH12cbiHJuaRMQZz4lXMgZ/Ykgh5pTY7NIILkQqxEEx/F+bYLzHvvW4zpwEKynG2PIpTuTkqGtQPj/t4vbmGfY3VwR7IOQkpPFC8irsKV5CwwEuOg2M+3mip4bUmTY/PZvGmxiaw01vJzqLKkCVRLLC5tbFSSm0HJorgtgPPaq1fmLfhhZAEbeHRONjDM+cPUu++A7qrYAUfWr6RpNiFa2IsQs+MDBLY/c9w8TgAzob7cHb+VlQ03BZioZiRQczJ/Zjje2HIrfafkYez5WGcNXfhLGBz99kaC/g51HaJPQ3naB22TdtTYhO4r6iK+4oqSY0L7b7k1uE+Xm3UvN1ai88tFOxNSmdX8Sq25VdE0361mxHe3qPhJEHbwvIHArx85QwvXTmD3wTITkjmY5XbWJPlDffQxDwzTRcJvPRN6OuE+CTb2aBqS7iHJUTEMmM+zNm3MUd2Q6/N1iQ53Wab3nYvTlLkF90OmAAnOxvZ3XA2uOUl3hPDXYW2Qf18Fg43xnC+t419jdW819UY3K+2JrOQXcWrWJPljcb9ajdDgjYRevUDXXzn/EEaBm2j4J3eSn6j4nYSZdZlUTEmgDmyB/PWzyDgh4JleB5/Omqy5IQINxMIQM1Rm7TQWmcPxiXgrL/H1ntLi9j6YUHGGC70trG74Syn3U42Hhy25JXxUMmaOe0pGw/4OdJu96tdGbS18GIdD9vyl7GrWFGckjkvP0MEk6BNhM5YwM+L9afZfeUsAQy5iak8WbkNlVkQ7qGJeWaG+gm88o9w2eYlOZsfwrn7N6OiGrwQkcYYA1fOETj8CtSdsQc9MTirttkG9bnRsf+3YbCbPQ3nONxWR8CdD1ub5eXhkjVUZeTfdOLFwNgoB5pr2N98nl53v1paXKLdr+atJH3plIaav6BNKZUKDGut/XMd1UKQoC20Lvd38p3zB2ka6sUB7itS/FrFhgXfnCpCz1zRBF76Fgz2QGIKnkeewlm+IdzDEmJRMG11mMOvYM4fJtjfafkGPFsegeLKiM44ndA5MsjexmrebKkJ7jurSM3m4dI13J5Tcs1WUS1DfexrrOadtkvBxvZFyRk8ULwq5L1RI9Sc6rStB74NPAL8O+AbwADwYa31jI3kI4kEbaHh84/zi/pT/LKhGoMhPymNj1duY2VGfriHJuaZCQQw776Ieed5+2JSXInnsU9GxRKOENHG9LRjju7BnH7DFqcG8K7As/VRWLEBJ7J6ZM5oYGyU/U3nea3pPAPjNms2PymNh4pXs71gGXGeGIwxVPe0sq+pmlNdTcHvXZflZVfxKlZnFkZFoBoicwraDgLfB74OXAD+AIgD/pfWeu08DjIkJGibfxf72vnO+UO0Dvfh4PBA8SqeKF+/INk7xj8OjTUQGwdZBVGxcTeamEAA+juhqwXT3QJdzZjmWmi/Ajg42x7D2fFBnKX3zleIBWWG+jHH92FO7AO3FRPZXptxump7VGRo+/zjvNVayy8bztE5artepsclsr1gGWe7m4P7n+M8MWx396t5kzPCOeRIMaegrQ/IAlYAr2mti5VScUCP1jo0XV7nUaiDtkv9HTQN9lKamoU3OWNRT+P6/OP82+WTvNqkMdh0649XbWdZem7Ir218w5hTb9gmzf1dkycSkm3wllkAWfmQWYCTVWA/JkZ8f7mwMb5h6GrFdDVDd4v7sRW6W2yttaslp+N59A9wyiP+fZoQi4rxjdi2cEf3TD73pWTibH4QZ/1OnITIr0vmNwGOtdezu+FcMLEAbAB3b1EVO70rQ14yJMrMqSNCNfD7wE7gF+6xjwCX5j6u6He4rY59TTYo9DgO3qQMSlIzKU3JoiQli5KUzEXRV/N8TyvfvXCI9pEBPDg8UrqGx8vWhTxINf3dmBP7MO/th1G7MZWMfIhPsJX3R4eg5RKmZfLXMfg2JCl1ShCXPy24cxZHAcbrMiZgn+S7WjBdLdDdbD92tdh9adeSkgnZhbY4bpbXfvSuiIoXByEWGyc+EWfTg5gN92H0u5jDr0BnI+bAs5hDL+BsfABn00MR/SY1xu1UsCWvnHM9LZzobGBZWg5b8soX9URHKNzMTNttwN+6X34MuBP4CvBxrfVroR3e3IV6pu1w22VOdDbQMNhD63AfMz2amfFJNoCbEszlJ6Vec1NmJBnxj/Gvl06wv/kCAMXJmXy8ajvlId7PZNobMEd3Y6oP2bISYPdSbXkElt+G43hs5tVgL/S0Yrrb3I+tNpjraYNx37UvkJwOmfmTs3LuLB2Z+ThRFmQb3wh0twaXM+3MWYudObvWYxATC1mFNpDN9kJWofuxQIIzISKYMQYunbLlQhrP24MJSTibHrT/EiI3eBM3bV6zRx2t9U19k1IqDfghUAUMAp8AkrB75Ebcm31Ja/0vSqmngD8DfMCntdb7bmlg1x7Dgu1p8/nHaRzq4cpADw2D3e6/HkZnWG6K98RQnJJJSUqWDeRSMylOySQxJnL2KZzrbuF7Fw7ROTqIx3F4rHQtj5auJTZE74yMMVB/jsCR3VB32h50HNvuZfPDON7lt3BfARjogZ62YCBnP7pB3UxLgBNSMqbN0E0EdmTmha3xuTEGBrrdWbMpgVlXsz1+LSkZbkBWOBmYZRdCWk5UtNARQlybabxA4O1/gyvV9kBCsi3Ds/EBefMV3W49aFNKeYA/BH4bWAmkA93ASeAHWuvv3eiqSqkvAeNa679USj0GfAZ4FojRWv/TlNvlAweAO4Bs4GWt9eqb+9luOIawJiIEjKFjZICGwW6uDNgg7spgN90TG0uncIC8xFRKUrOmBXNZ8ckLmkUzPO7juUvHebPlIgClKVl8vGp7yBrwGv+4bap8dLe74R2IjcdZ9wH7znGei7baZcPumWfoetuvH9ClZrnLrPnvn6Gbh43BZmzUBphdzXYZs6sF0+3uN5voXXi1mFh3+dcNzrK9OFlukBbBSyZCiPlhrmgC7zwPDe5rXWKKfaO78f4lsRVkEZrVnra/B27DLoWeAfqBZGAt8F+VUmu01l+8wYX3ALXu55lAL1AGHLzqdruAF7XWfUCfUqpVKaX0Ikj79DgO+Ulp5CelsSm3LHh8cGw0GMA1uMFc01AvbSMDtI0McKzjSvC2KbHx71te9Sanh2TG63RXE9+/8C7dviFiHQ+Pl63n4ZLVxIRgVsaMDmFOHcAc2zs5W7QALV0cxwPpOZCeg1O2ZvqYAu4+sCmBXHCGrrfDjnOgG+O+s518y+NAWtb0fXOZBZBVYHsMTsmstcu6PdMzNLtabBJAX+e1B56UZveaZbmBmTt7RkauZHMKsYQ5pYqY0v+Eqa8m8M6/QeMFzFs/wxzdY7NNb78/6rZ9iJldb6atB1ipte6Y4VwRcFprfVMbm5RSx4EN2FpvH8bO3BUAR7Gzb58GfFrrr7m3/yHwTa3169e5z3uBe2/i8l+KlthvPOCnZbhvckbO/Tg4/v4ZlhjHgzc5PZjsUOrOzqXOcvlucMzHs5eO8U6rjbEr0nL4eOU2ikLQLsT0d2GO/RJz6gD43FXybK99Z7g6clPZTcBvg6ruVkxPm/txYoauA0xg5m90HBskZuTbxInulsmf+2qemMmkCXcpMzhrJuVNhBA3ENxm8s7z0FRjDyal4mx9FGfDfWHb4iFuyaxm2vqxZT7eF7QBGUzuSbshrfVGpdRO4J+ArwLPAG8Bfwn8BdAETF2TMsB1Oy5orfcD+290bXeJNirEemLcIGxyGdIYQ49v2F1endwr1zbcT8NgT7DOzYSs+GRKUqfvlctLTLtuU92TnQ38oOYwvb5hYh0PT1TcxgPFq4iZ50QJ01Zne1aePzyZXFCi8Gx5GJatj/iikc5EQJWZ/76/KOMfnxLQtUJ3mxvQtdrjvR3234TElMllzOwpe83Sc6UllBBi1hzHgfI1eMpWQ90ZG7w119ps0yO7cbY+YlcyJHiLStd7dfgvwMtKqeeA49ilzSRgPXaf25/e6M6VUl8Dvqq1btBav66USsTuh+t1z38f+DvgMLBuyrcWAfWz+HkWHcdxyEpIJishmfXZk33oRvxjNA32BpdXrwx20zjYQ7dviO6uoWkVphM8sW7Sw+SMXHFKJmMBP/9y8QjvttuGxSvSc3mycjuFyenzNn5jjH3iOPIK1J9zfygPjrrDzqwVVszbtcLJiYm1S6FZBe8P6MbHoK/DzsglJNsgLSktLOMUQiwNjuNAxTo85Wvh8mkCbz8PrZcwr//EDd4etXXe4uLDPVRxC66bPaqUqsIGaLdhExF6sHXbfqy1PnujO1dK/R1wWWv9N0qpHdj9cdnAb2utTyqlvgikAv8L2Adsw+55+77WeuucfrLJMYQ0EcEEAhGThRcwAdqHByb3yrnJDz1u492pHOzM3ljAT5wnhl+v2MB9RVXzVobE+Mcx1YcwR3ZDZ6M9GJeAs/4em9mUEfqCvEIIISxbKuQ9O/PWat+ok5KBc8fj9nk5QrelLFFz6ohQxfTs0fe01s03c1WlVAG25Ecpdpn1E0AutiVWDHAaeEpr3aeUehr4PLbkx+9prQ/dzDVuYgwhDdoCR3djju3FKVoJRStxilZAbklELXENjI28rwxJ01AvAWOoysjnycpt5M3TzI8ZGcK8tx9zfN9kAdeUDBuo3bYTJzHim2gIIcSiZYyB2pM2eGtzF7RSs3DueMxm7EvwFglmVfJjFfAjoBw7uzaAzR5VwClscd0rM35zBAl50PbSt2wB2Kli46FwGc5EEOddEXGbyMcCfvrHRuatnIjp68Ac22uTCyZKU+QUu33ytkVUECuEEEudMQYuHifwzs8nSy2lZduZt3V3y3P2FMYYW1Hg8hmc1Cycyk2hvuSsgrZ3gJ9i96SZKccd4M+B+7TW987vOOdfyJdHTQA6mzBNF6Gpxn7saX3/DbO9ON4Vk7Nx2YURv/H+ZpjWy5gjuzHnj0xmT5attp0LytcuaH05IYQQt8aYAFw4ZoO3ia0sadk4234FZ+1dSzZ4M6PDcOUc5vJpzOUzdl8yQNkaYv7dH4f68rMK2oaAfK31wAznEoGtGRxyAAANUElEQVROaRg/MzPUD80XMU01mKYaaLkM/rHpN0pIhqIVdjbOuxK8y6Imm8eYgG2jcmT3ZDHHieSCLQ/h5JeHd4BCCCFuiQ3ejrrBm5vIlp5rg7c1OxZ98GZMAFrrMXWnMZdPQ3PtZJUDsAWLy9fA8tvxrN4e6uHMKmh7A3gN+O9a67Grzn0O+LDWOuQjn6twd0QAtxxEWz2m2Z2Na6x5f8NuxwN5pXYWbmI2Li0nomaqzPgY5txB27mgy93WGJ9oN7FuehAnxP1IhRBChJYJBDDnD2MO/mLyeT4jD2f7r+Cs3rGoCnmbwV5M3Rm4fBpTdxaG+ydPOh7wLsepWIdTvhYKKhYy6XBWQVsF8F3gduAckyU/1gAaeFJrXTPfI51vkRC0Xc0YA/1ddhau6SKmuQbarry/OGtK5uRsXNFKyC8Ly7sdMzyAee91zPG9MNRnD6Zm4Wx6wAZs0qBYCCEWFRMIYPS7mIM/t230wNao3P6rdp9yFAZvxj8OjTWTs2ntV23LT8smGKSVrQlnG8A5ZY+WYrsZTJT80Frri/M6vBCKxKBtJmZsFFouYZouusFcja2eP1VMHBRW4HhXuNmqK3Dmsaba+8bU0247F5x+A8Z99mBeqa2vprYu+ulyIYRY6kzAb8s3HfyFrTUJth7l9ifslpgIKXl1Laa7FVN3xgZpV6qn93COjYeSKoKBWrY3Ula3Zhe0KaU2AL/F+xvG/0RrXXvNb4wg0RK0Xc2YgO1NOZHg0Hxxcqp6qsz8yQDOuxJyiub8R2SaazFHd2MuHIWJ34/ydbZzQdnqSPmlFkIIsUBMwG+3xxz8BfS224PZXjvzVrU1YoI34xuGKxpz+ZRNIJgY64ScIpyKdTgV66C4KlJLnMxqefRp4H9gy35MbRi/Bvgd4LNa63+Z96HOs2gN2mZihgfcBAd3Nq7l0uQM2IT4JLsO7y6rUrgcJyHpxvdtAlD7nu1c0HjBHvTE2GnwzQ/h5JWG4CcSQggRTYx/HHPuHczBFyYzKnOK7Mxb1eYFr4pgTADarrhLnmfsKtXUBIKEZDuLVrEWp3xttOy9nlXQ1gA8orU+PcO5bdjZtohPE1xMQdvVjH8cOhqmlBupgf6u6TdyHFvst2gFeN29cRm5wdkyM+azf4BHd0/uW0hIsr3pbt+Fk5aFEEIIMZXxj2POvm1n3iZed3KK8ex4Aio3hTR4M0N9NkCrcxMIJvZag33NK1yOU7EWp3ydrZkaIbOAt2BWQVsPsEpr3TLDuSygTmsdug1V82QxB20zMf1d02fj2uqnv+sASE63BX8zcjHnDk5mzKRl21m1dXfjxN94dk4IIcTSZvzjmDNvYg69OBm85Zbg2fFBWLlxfoq3+8dt0t7EbFpb3fQbpGZhlzwnEggivhrZjcwqaPsb4H7gr5neMH4d8CfASa31H877UOfZUgvarmbGfNB62daMc0uOMHxV6b38ctu5oGpLVGYECSGECC8zPjYZvA1024N5pTZ4W3H7LQdvpqd9Msuz/tz0BIKYWChROBVrcSrWR1ICwXyZdSLCU8BHmWwY341tafVD4Nta68A1vzlCLPWg7Wq2HUebnYXrarFr/aVqsf3CCyGECAMzPoY5dQDz7kuT9UgLym3wtuy2a77WGN8INGjMpdOYutOTmaoTsr1MSyCIiw/pzxFmsy/5Ee0kaBNCCCEWlhkfs/U9D78Eg732YMEyu+dt2Xr7dUcD5tIpW+C28cJVCQRJdqnTDdSiJIFgvkjQJoQQQoiFZcZ8mPf2Yw6/PJkwkFti91JPBHMAODZpoGItTsVEAsGS3a4zqz1td9zoXrXW785hUAtCgjYhhBAivMzYKOakG7xNJL+lZOJUrIWKdThla3CSUsM6xghyzaDteiXtvwLcBQwAXTOcN8DyuY1LCCGEEIudE5eAs+VhzIZ7oe4sZOZBTrHsp75FN0pE+Abg0Vp/cjZ3rpRKwyYtVAGDwCewyQzPAbnAc1rrz7m3fQr4M8AHfFprvW8215xhDDLTJoQQQohoMauZNoCfAXfP4cKfAw5qrX9VKfUY8N+AMeBLwAvAC0qp+7AdFz6PzVLNBl4GVs/hukIIIYQQi8p1gzat9R5gzxzufw8w0aM0E9sK6y7gN7TWRin1LPAQUAi8qLXuA/qUUq1KKaVlikwIIYQQArjxTNucaK3fAVBKHQc2AB8B1mitJ9Zkm4CdQA/QMOVbm7CB3DWDNqXUvcC98z5oIYQQQogIFNKgbYLWeqNSaifwDNOTGgzgdz+Oz3D8eve5H9h/o2srpb50a6MVQgghhIg8Ie2iqpT6mlKqBEBr/TqQiE1AmFAE1AON7udXHxdCCCGEEIR+pi0G+DDwN0qpHdj9be1u8sEB4HeBL2ADtP+slPoLoAxI1VrPW9CmlJqvuxJCCCGECCWjtZ4xgzTUQdtfAj9USn0K6MCW/PABPwZygGe01kcBlFJfB065539vvgZwrR9cWEqpL2utvxzucSwV8ngvLHm8F5Y83gtLHu+FF+7HPNSJCK3ArhlObZ7htt8EvhnK8QghhBBCRKuQ7mkTQgghhBDzQ4I2IYQQQogoIEGbEEIIIUQUkKBNCCGEECIKSNAmhBBCCBEFJGgTQgghhIgCErQJIYQQQkQBCdrE/nAPYInZH+4BLDH7wz2AJWZ/uAewxOwP9wCWoP3hvLhjjAnn9YUQQgghxE2QmTYhhBBCiCggQZsQQgghRBSQoE0IIYQQIgpI0CaEEEIIEQViwz0AsfCUUh7gH4FdQDfwGa31W+Ed1eLnPu5vA1/WWr8S7vEsdkqpTwJfBMaA/1dr/VKYh7RoKaUc4H8D9wHjwH/UWr8W3lEtXkqpDwI7tNZfUEotB54DcoHntNafC+/oFp+rHu/VwI+AFOA08DGt9cBCjUVm2pamXweygXLgI8A/hHc4S8ZngcpwD2IpUEotA/4jsB54BPjb8I5o0XsAKNRarwJ+C/i7MI9nUVJKOUqprwHfmnL4q8CXsM/nSil1X1gGtwhd4/H+a+AvtNaVQDXw6YUckwRtS1Mh8B2ttdFanwXy3HfKIkTcIOJB4IVwj2WJeAJ4Rms9oLWuBT4U7gEtcgEgSSkVB2QB/WEez2K2D/ghgFIqBtgMvKC1NsCzwENhHNtiFHy8XYnA8+7nbwGrFnIwsjy6BGmtvz7xuVLqE8AF9w9ehM7fA3+EXa4TobcSCCilDgEJwH8O83gWu9eA/wm0AenAR8M7nMXJfZ5+QSmViw0WcoHuKc/fTcDOcI1vsZnh8UZr/SCAUioJu3ryo4Uck8y0LVFKqRSl1Lexv3T/PszDWdSUUr8PHNJa63CPZQlJwi5F78TOsv2zOwskQuOTQA2QB2wA/lopFR/eIS0JBruHcOrX/jCNZclQSm0EjgAXgGcW8toStC1BSqlk4A1gANjiLh+J0LkX+JhSqhq7n/DbSqld4R3SotcN/EJrPaK1vgBcwW4LEKGxAbsJflxrfRpoRh7vhdAJ5Ez5ugioD9NYlgT3ufvnwOe11p/RWgcW8vqyPLo0fQp4U2v9R+EeyFKgtX5y4nOl1DPAj7XW+8I3oiXhVeBTSqlvAQVABnbpSITGKWzCx78qpYqBTKAxvENa/LTWfqXUCTf54ADwu8AXwjysxe5vgF/TWh8Nx8UlaFuaNgP3KKWCG1bdrC8hFgWt9ctKqe3AGWAYeFprLctGofNt4B+UUhoYBf6DPN4L5vPAj7Ezbs+EK5hYCtwtFmuBHyilJg7/q9Z6wfYqS8N4IYQQQogoIHvahBBCCCGigARtQgghhBBRQII2IYQQQogoIEGbEEIIIUQUkKBNCCGEECIKSNAmhFiSlFIVSimjlHpxhnM/c89VzPK+n1FKfcH9/MtKqW/McbhCCCFBmxBiSRsH7lRKZUwccDuGfIDp7YGEECLspLiuEGIp82O7J/wK8AP32GPYNm+PASilHga+ApS5xz+ttb6ilPoyUIrtuHA3cAL4TeAp4EnAKKUmWtzkKqVeBe4EjgEf1Fq3h/ynE0IsKjLTJoRY6n6KDbYm/KZ7DKAcW23+s9gA7STwoym3/R3gq0AJ9k3wH2qt/xr4LvCn7udgg8L/DuTidmgIyU8ihFjUZKZNCLHUvQB8XSmVgl0SfRDbnxdgB7bx/Ktg96cBPUqpiWboL2qtX3PP7cUGdjPZM+V2rwLLQvGDCCEWN5lpE0IsaVrrPuBN4FHgIeCw1rrXPb0TqJly2zGgF8h3D3VMuasRrv1GuHnK5/7r3E4IIa5JnjiEEGJyiXQU+NmU4+3YpU8A3ISFXKD+Fu9fmjwLIeZMZtqEEAJ+jp1lexx4fsrxrwIfUkrdqZRKBb4MvKS17rnB/Y1jgzshhJg3ErQJIZY8rXUXcBQ4q7Vum3KqF/h/gO9hZ91WA5+5ibvcA3xGKfVH8z1WIcTS5Rgjs/ZCCCGEEJFOZtqEEEIIIaKABG1CCCGEEFFAgjYhhBBCiCggQZsQQgghRBSQoE0IIYQQIgpI0CaEEEIIEQUkaBNCCCGEiAIStAkhhBBCRAEJ2oQQQgghosD/BQMfSyxtqhrNAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 720x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# You can probably just put this at the top of every notebook you write\n",
    "# Adding it here for completeness\n",
    "%matplotlib inline\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "\n",
    "# Load airtravel data\n",
    "air_travel = np.loadtxt(\"data/airtravel.csv\", skiprows=6, \\\n",
    "                        unpack=True, usecols=[1,2,3], delimiter=\",\")\n",
    "\n",
    "mths = np.arange(1, 13)\n",
    "plt.figure(figsize=(10,3))\n",
    "plt.plot(mths, air_travel[0], '-', label=\"1958\")\n",
    "plt.plot(mths, air_travel[1], '-', label=\"1959\")\n",
    "plt.plot(mths, air_travel[2], '-', label=\"1960\")\n",
    "plt.xlabel(\"Month\")\n",
    "plt.ylabel(\"1000s of travellers per month\")\n",
    "plt.legend(loc=\"best\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "You may not want to use lines to join the data points, but symbols like dots, crosses, etc. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7fcc031df860>"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,3))\n",
    "plt.plot(mths, air_travel[0], 'x', label=\"1958\")\n",
    "plt.plot(mths, air_travel[1], '+', label=\"1959\")\n",
    "plt.plot(mths, air_travel[2], 'o', label=\"1960\")\n",
    "plt.xlabel(\"Month\")\n",
    "plt.ylabel(\"1000s of travellers per moth\")\n",
    "plt.legend(loc=\"best\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can also use dots **and** lines. Moreover, we can change the type of line: from full lines to dashed to dash-dot..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7fcc03148ba8>"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,3))\n",
    "plt.plot(mths, air_travel[0], 'x-', label=\"1958\")\n",
    "plt.plot(mths, air_travel[1], '+--', label=\"1959\")\n",
    "plt.plot(mths, air_travel[2], 'o-.', label=\"1960\")\n",
    "plt.xlabel(\"Month\")\n",
    "plt.ylabel(\"1000s of travellers per moth\")\n",
    "plt.legend(loc=\"best\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Exercise 2.5.2**\n",
    "\n",
    "The file [`NOAA.csv`](data/NOAA.csv) contains data from [NOAA](http://www.aoml.noaa.gov/hrd/tcfaq/E11.html) on the number of storms and hurricanes in the Atlantic basin from 1851 to 2015. The data columns are described in the first row of the file. The year is in column 1 and the number of hurricanes in column 3.\n",
    "\n",
    "For those interested, the data is pulled from the website with [getNOAA.py](geog0111/getNOAA.py).\n",
    "\n",
    "* load the year and hurricane data from the file [`NOAA.csv`](data/NOAA.csv) into a numpy array\n",
    "* produce a plot showing the number of hurricanes as a function of year, with the data plotted in a blue line\n",
    "* put a dashed red line on the graph showing the mean number of hurricanes\n",
    "* plot circle symbols for all years where the number of hurricanes is greater than the mean + 1.96 standard deviations.\n",
    "\n",
    "Hint: the options on `np.loadtxt` you probably want to use are: `skiprows`, `delimiter`, `usecols` and  maybe `unpack`. You may want to investigate the use of the [`axhline`](https://matplotlib.org/3.1.0/api/_as_gen/matplotlib.pyplot.axhline.html) matplotlib function and see whether you can use it here."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "# do exercise here"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.5.2 Accessing online resources"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can use `np.loadtxt` or similar functions to load tabular data that we have stored locally in e.g. csv format.\n",
    "\n",
    "Sometimes we will need pull a data file from a [URL](https://en.wikipedia.org/wiki/URL). We have used this idea previously to 'scrape' data from a web page, but often the task is more straightforward, and we effectively need only to 'download' the data in the file.\n",
    "\n",
    "We will use the standard Python `urllib.request` package to do this and pull the data. The data is returned as a string, which needs to be decoded. We can then split it by lines (`\\n` character) and pass the resulting list of strings to [`np.loadtxt`](https://docs.scipy.org/doc/numpy/reference/generated/numpy.loadtxt.html), which allows us to read in the data from memory without first saving it to disk.\n",
    "\n",
    "First, let's look at how we acquire the data and visualise the contents of the file. You can obvioulsy open the URL on the browser and check for your yourself that you get the same.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Latitude: 45.4  Longitude: -115.0534\n",
      "X & Y on Lambert Conformal Conic: -1118364.52 408458.12\n",
      "Tile: 12093\n",
      "Elevation: 1474 meters\n",
      "All years; all variables; Daymet Software Version 3.0; Daymet Data Version 3.0.\n",
      "How to cite: Thornton; P.E.; M.M. Thornton; B.W. Mayer; Y. Wei; R. Devarakonda; R.S. Vose; and R.B. Cook. 2016. Daymet: Daily Surface Weather Data on a 1-km Grid for North America; Version 3. ORNL DAAC; Oak Ridge; Tennessee; USA. http://dx.doi.org/10.3334/ORNLDAAC/1328 \n",
      "\n",
      "year,yday,tmax (deg c)\n",
      "2000.0,1.0,-1.5\n",
      "2000.0,2.0,-2.5\n",
      "2000.0,3.0,-1.5\n",
      "2000.0,4.0,-1.5\n",
      "2000.0,5.0,-1.0\n",
      "2000.0,6.0,-0.5\n",
      "2000.0,7.0,-1.5\n",
      "2000.0,8.0,0.5\n",
      "2000.0,9.0,0.0\n",
      "2000.0,10.0,0.5\n",
      "2000.0,11.0,0.0\n",
      "2000.0,12.0,-1.0\n",
      "2000.0,13.0,-1.0\n",
      "2000.0,14.0,0.5\n",
      "2000.0,15.0,1.5\n",
      "2000.0,16.0,3.0\n",
      "2000.0,17.0,3.5\n",
      "2000.0,18.0,2.0\n",
      "2000.0,19.0,1.0\n",
      "2000.0,20.0,2.0\n",
      "2000.0,21.0,2.0\n",
      "2000.0,22.0,1.5\n",
      "2000.0,23.0,-0.5\n",
      "2000.0,24.0,2.0\n",
      "2000.0,25.0,3.5\n",
      "2000.0,26.0,3.0\n",
      "2000.0,27.0,-0.5\n",
      "2000.0,28.0,0.5\n",
      "2000.0,29.0,-2.0\n",
      "2000.0,30.0,-1.0\n",
      "2000.0,31.0,-0.5\n",
      "2000.0,32.0,0.5\n",
      "2000.0,33.0,5.0\n",
      "2000.0,34.0,7.5\n",
      "2000.0,35.0,8.0\n",
      "2000.0,36.0,4.5\n",
      "2000.0,37.0,4.5\n",
      "2000.0,38.0,5.5\n",
      "2000.0,39.0,7.5\n",
      "2000.0,40.0,6.0\n",
      "2000.0,41.0,7.0\n",
      "2000.0,42.0,5.0\n",
      "2000.0,43.0,3.5\n",
      "2000.0,44.0,2.5\n",
      "2000.0,45.0,3.5\n",
      "2000.0,46.0,3.5\n",
      "2000.0,47.0,4.0\n",
      "2000.0,48.0,4.5\n",
      "2000.0,49.0,3.0\n",
      "2000.0,50.0,3.5\n",
      "2000.0,51.0,5.5\n",
      "2000.0,52.0,6.5\n",
      "2000.0,53.0,7.0\n",
      "2000.0,54.0,8.0\n",
      "2000.0,55.0,3.5\n",
      "2000.0,56.0,3.0\n",
      "2000.0,57.0,4.0\n",
      "2000.0,58.0,6.0\n",
      "2000.0,59.0,5.5\n",
      "2000.0,60.0,6.5\n",
      "2000.0,61.0,7.0\n",
      "2000.0,62.0,9.5\n",
      "2000.0,63.0,10.5\n",
      "2000.0,64.0,12.0\n",
      "2000.0,65.0,13.0\n",
      "2000.0,66.0,11.5\n",
      "2000.0,67.0,11.0\n",
      "2000.0,68.0,9.0\n",
      "2000.0,69.0,7.0\n",
      "2000.0,70.0,4.5\n",
      "2000.0,71.0,6.5\n",
      "2000.0,72.0,6.0\n",
      "2000.0,73.0,8.0\n",
      "2000.0,74.0,9.0\n",
      "2000.0,75.0,7.5\n",
      "2000.0,76.0,6.0\n",
      "2000.0,77.0,4.0\n",
      "2000.0,78.0,6.0\n",
      "2000.0,79.0,5.0\n",
      "2000.0,80.0,4.5\n",
      "2000.0,81.0,7.0\n",
      "2000.0,82.0,12.5\n",
      "2000.0,83.0,12.5\n",
      "2000.0,84.0,7.0\n",
      "2000.0,85.0,12.0\n",
      "2000.0,86.0,14.0\n",
      "2000.0,87.0,16.5\n",
      "2000.0,88.0,13.0\n",
      "2000.0,89.0,6.0\n",
      "2000.0,90.0,7.0\n",
      "2000.0,91.0,11.0\n",
      "2000.0,92.0,13.5\n",
      "2000.0,93.0,12.0\n",
      "2000.0,94.0,15.0\n",
      "2000.0,95.0,19.5\n",
      "2000.0,96.0,16.5\n",
      "2000.0,97.0,9.0\n",
      "2000.0,98.0,7.5\n",
      "2000.0,99.0,15.5\n",
      "2000.0,100.0,18.0\n",
      "2000.0,101.0,17.0\n",
      "2000.0,102.0,18.5\n",
      "2000.0,103.0,20.0\n",
      "2000.0,104.0,17.0\n",
      "2000.0,105.0,13.5\n",
      "2000.0,106.0,12.5\n",
      "2000.0,107.0,12.0\n",
      "2000.0,108.0,13.0\n",
      "2000.0,109.0,17.5\n",
      "2000.0,110.0,16.0\n",
      "2000.0,111.0,15.5\n",
      "2000.0,112.0,18.0\n",
      "2000.0,113.0,17.0\n",
      "2000.0,114.0,10.5\n",
      "2000.0,115.0,9.5\n",
      "2000.0,116.0,10.5\n",
      "2000.0,117.0,12.0\n",
      "2000.0,118.0,19.0\n",
      "2000.0,119.0,19.5\n",
      "2000.0,120.0,10.5\n",
      "2000.0,121.0,15.5\n",
      "2000.0,122.0,22.5\n",
      "2000.0,123.0,21.5\n",
      "2000.0,124.0,18.0\n",
      "2000.0,125.0,17.0\n",
      "2000.0,126.0,12.5\n",
      "2000.0,127.0,13.0\n",
      "2000.0,128.0,13.0\n",
      "2000.0,129.0,14.0\n",
      "2000.0,130.0,12.0\n",
      "2000.0,131.0,9.0\n",
      "2000.0,132.0,5.5\n",
      "2000.0,133.0,8.0\n",
      "2000.0,134.0,13.0\n",
      "2000.0,135.0,17.0\n",
      "2000.0,136.0,20.0\n",
      "2000.0,137.0,19.5\n",
      "2000.0,138.0,18.0\n",
      "2000.0,139.0,17.5\n",
      "2000.0,140.0,19.0\n",
      "2000.0,141.0,20.0\n",
      "2000.0,142.0,22.5\n",
      "2000.0,143.0,24.0\n",
      "2000.0,144.0,20.0\n",
      "2000.0,145.0,18.0\n",
      "2000.0,146.0,19.5\n",
      "2000.0,147.0,19.5\n",
      "2000.0,148.0,17.5\n",
      "2000.0,149.0,17.0\n",
      "2000.0,150.0,17.0\n",
      "2000.0,151.0,15.5\n",
      "2000.0,152.0,13.0\n",
      "2000.0,153.0,14.0\n",
      "2000.0,154.0,23.0\n",
      "2000.0,155.0,25.5\n",
      "2000.0,156.0,26.5\n",
      "2000.0,157.0,28.5\n",
      "2000.0,158.0,26.5\n",
      "2000.0,159.0,29.0\n",
      "2000.0,160.0,25.5\n",
      "2000.0,161.0,17.0\n",
      "2000.0,162.0,17.0\n",
      "2000.0,163.0,13.5\n",
      "2000.0,164.0,13.0\n",
      "2000.0,165.0,15.0\n",
      "2000.0,166.0,21.5\n",
      "2000.0,167.0,22.0\n",
      "2000.0,168.0,19.5\n",
      "2000.0,169.0,21.5\n",
      "2000.0,170.0,23.5\n",
      "2000.0,171.0,21.5\n",
      "2000.0,172.0,20.0\n",
      "2000.0,173.0,24.5\n",
      "2000.0,174.0,26.0\n",
      "2000.0,175.0,25.5\n",
      "2000.0,176.0,25.5\n",
      "2000.0,177.0,25.5\n",
      "2000.0,178.0,26.0\n",
      "2000.0,179.0,27.0\n",
      "2000.0,180.0,27.5\n",
      "2000.0,181.0,28.5\n",
      "2000.0,182.0,29.5\n",
      "2000.0,183.0,29.5\n",
      "2000.0,184.0,27.0\n",
      "2000.0,185.0,23.0\n",
      "2000.0,186.0,19.0\n",
      "2000.0,187.0,21.5\n",
      "2000.0,188.0,20.5\n",
      "2000.0,189.0,23.0\n",
      "2000.0,190.0,26.5\n",
      "2000.0,191.0,26.0\n",
      "2000.0,192.0,26.0\n",
      "2000.0,193.0,29.0\n",
      "2000.0,194.0,30.5\n",
      "2000.0,195.0,33.5\n",
      "2000.0,196.0,32.5\n",
      "2000.0,197.0,29.0\n",
      "2000.0,198.0,27.5\n",
      "2000.0,199.0,29.0\n",
      "2000.0,200.0,27.5\n",
      "2000.0,201.0,26.0\n",
      "2000.0,202.0,26.0\n",
      "2000.0,203.0,29.0\n",
      "2000.0,204.0,32.5\n",
      "2000.0,205.0,32.0\n",
      "2000.0,206.0,29.5\n",
      "2000.0,207.0,30.0\n",
      "2000.0,208.0,31.0\n",
      "2000.0,209.0,31.5\n",
      "2000.0,210.0,32.0\n",
      "2000.0,211.0,34.0\n",
      "2000.0,212.0,35.0\n",
      "2000.0,213.0,35.5\n",
      "2000.0,214.0,34.5\n",
      "2000.0,215.0,33.5\n",
      "2000.0,216.0,32.0\n",
      "2000.0,217.0,30.5\n",
      "2000.0,218.0,30.0\n",
      "2000.0,219.0,30.0\n",
      "2000.0,220.0,31.0\n",
      "2000.0,221.0,31.5\n",
      "2000.0,222.0,34.0\n",
      "2000.0,223.0,32.5\n",
      "2000.0,224.0,30.0\n",
      "2000.0,225.0,28.0\n",
      "2000.0,226.0,29.0\n",
      "2000.0,227.0,29.0\n",
      "2000.0,228.0,28.5\n",
      "2000.0,229.0,28.0\n",
      "2000.0,230.0,29.0\n",
      "2000.0,231.0,27.5\n",
      "2000.0,232.0,26.0\n",
      "2000.0,233.0,23.0\n",
      "2000.0,234.0,24.0\n",
      "2000.0,235.0,25.0\n",
      "2000.0,236.0,29.0\n",
      "2000.0,237.0,28.0\n",
      "2000.0,238.0,28.5\n",
      "2000.0,239.0,29.0\n",
      "2000.0,240.0,26.5\n",
      "2000.0,241.0,25.5\n",
      "2000.0,242.0,25.0\n",
      "2000.0,243.0,26.0\n",
      "2000.0,244.0,23.5\n",
      "2000.0,245.0,22.5\n",
      "2000.0,246.0,17.5\n",
      "2000.0,247.0,13.0\n",
      "2000.0,248.0,15.0\n",
      "2000.0,249.0,16.0\n",
      "2000.0,250.0,15.5\n",
      "2000.0,251.0,19.0\n",
      "2000.0,252.0,22.5\n",
      "2000.0,253.0,19.5\n",
      "2000.0,254.0,15.5\n",
      "2000.0,255.0,15.5\n",
      "2000.0,256.0,23.0\n",
      "2000.0,257.0,27.5\n",
      "2000.0,258.0,29.5\n",
      "2000.0,259.0,30.5\n",
      "2000.0,260.0,30.0\n",
      "2000.0,261.0,27.5\n",
      "2000.0,262.0,24.5\n",
      "2000.0,263.0,21.5\n",
      "2000.0,264.0,19.0\n",
      "2000.0,265.0,15.5\n",
      "2000.0,266.0,11.5\n",
      "2000.0,267.0,10.5\n",
      "2000.0,268.0,15.5\n",
      "2000.0,269.0,20.0\n",
      "2000.0,270.0,21.5\n",
      "2000.0,271.0,23.0\n",
      "2000.0,272.0,24.0\n",
      "2000.0,273.0,20.5\n",
      "2000.0,274.0,16.5\n",
      "2000.0,275.0,15.5\n",
      "2000.0,276.0,16.0\n",
      "2000.0,277.0,15.5\n",
      "2000.0,278.0,15.0\n",
      "2000.0,279.0,13.5\n",
      "2000.0,280.0,14.5\n",
      "2000.0,281.0,17.5\n",
      "2000.0,282.0,19.5\n",
      "2000.0,283.0,20.5\n",
      "2000.0,284.0,19.0\n",
      "2000.0,285.0,12.0\n",
      "2000.0,286.0,9.0\n",
      "2000.0,287.0,5.0\n",
      "2000.0,288.0,6.0\n",
      "2000.0,289.0,11.0\n",
      "2000.0,290.0,14.0\n",
      "2000.0,291.0,16.5\n",
      "2000.0,292.0,16.5\n",
      "2000.0,293.0,15.0\n",
      "2000.0,294.0,15.0\n",
      "2000.0,295.0,11.0\n",
      "2000.0,296.0,6.0\n",
      "2000.0,297.0,10.0\n",
      "2000.0,298.0,11.5\n",
      "2000.0,299.0,12.5\n",
      "2000.0,300.0,11.5\n",
      "2000.0,301.0,10.5\n",
      "2000.0,302.0,11.0\n",
      "2000.0,303.0,10.0\n",
      "2000.0,304.0,7.0\n",
      "2000.0,305.0,7.5\n",
      "2000.0,306.0,5.5\n",
      "2000.0,307.0,5.5\n",
      "2000.0,308.0,6.5\n",
      "2000.0,309.0,8.0\n",
      "2000.0,310.0,5.5\n",
      "2000.0,311.0,2.5\n",
      "2000.0,312.0,3.5\n",
      "2000.0,313.0,2.0\n",
      "2000.0,314.0,0.0\n",
      "2000.0,315.0,-1.5\n",
      "2000.0,316.0,-3.5\n",
      "2000.0,317.0,-3.5\n",
      "2000.0,318.0,-1.5\n",
      "2000.0,319.0,-2.5\n",
      "2000.0,320.0,-3.0\n",
      "2000.0,321.0,0.0\n",
      "2000.0,322.0,0.5\n",
      "2000.0,323.0,0.0\n",
      "2000.0,324.0,1.0\n",
      "2000.0,325.0,2.0\n",
      "2000.0,326.0,3.5\n",
      "2000.0,327.0,4.0\n",
      "2000.0,328.0,3.0\n",
      "2000.0,329.0,2.0\n",
      "2000.0,330.0,2.0\n",
      "2000.0,331.0,2.0\n",
      "2000.0,332.0,2.0\n",
      "2000.0,333.0,1.0\n",
      "2000.0,334.0,0.5\n",
      "2000.0,335.0,1.5\n",
      "2000.0,336.0,1.0\n",
      "2000.0,337.0,1.0\n",
      "2000.0,338.0,1.5\n",
      "2000.0,339.0,1.0\n",
      "2000.0,340.0,2.0\n",
      "2000.0,341.0,2.5\n",
      "2000.0,342.0,1.5\n",
      "2000.0,343.0,0.5\n",
      "2000.0,344.0,0.0\n",
      "2000.0,345.0,-1.5\n",
      "2000.0,346.0,-3.5\n",
      "2000.0,347.0,-6.0\n",
      "2000.0,348.0,-5.5\n",
      "2000.0,349.0,-2.0\n",
      "2000.0,350.0,-0.5\n",
      "2000.0,351.0,0.0\n",
      "2000.0,352.0,-0.5\n",
      "2000.0,353.0,-0.5\n",
      "2000.0,354.0,-2.5\n",
      "2000.0,355.0,-3.0\n",
      "2000.0,356.0,-2.5\n",
      "2000.0,357.0,-0.5\n",
      "2000.0,358.0,1.0\n",
      "2000.0,359.0,1.0\n",
      "2000.0,360.0,1.5\n",
      "2000.0,361.0,1.5\n",
      "2000.0,362.0,0.0\n",
      "2000.0,363.0,0.5\n",
      "2000.0,364.0,-0.5\n",
      "2000.0,365.0,-0.5\n",
      "2001.0,1.0,2.0\n",
      "2001.0,2.0,0.0\n",
      "2001.0,3.0,1.5\n",
      "2001.0,4.0,3.5\n",
      "2001.0,5.0,4.0\n",
      "2001.0,6.0,3.5\n",
      "2001.0,7.0,0.5\n",
      "2001.0,8.0,-0.5\n",
      "2001.0,9.0,1.5\n",
      "2001.0,10.0,2.0\n",
      "2001.0,11.0,2.0\n",
      "2001.0,12.0,-0.5\n",
      "2001.0,13.0,0.0\n",
      "2001.0,14.0,-1.0\n",
      "2001.0,15.0,-1.0\n",
      "2001.0,16.0,-3.0\n",
      "2001.0,17.0,-2.5\n",
      "2001.0,18.0,-2.5\n",
      "2001.0,19.0,-0.5\n",
      "2001.0,20.0,0.5\n",
      "2001.0,21.0,0.5\n",
      "2001.0,22.0,3.5\n",
      "2001.0,23.0,2.5\n",
      "2001.0,24.0,2.5\n",
      "2001.0,25.0,1.0\n",
      "2001.0,26.0,2.0\n",
      "2001.0,27.0,1.0\n",
      "2001.0,28.0,0.5\n",
      "2001.0,29.0,-2.5\n",
      "2001.0,30.0,-3.0\n",
      "2001.0,31.0,0.5\n",
      "2001.0,32.0,2.0\n",
      "2001.0,33.0,2.0\n",
      "2001.0,34.0,2.5\n",
      "2001.0,35.0,3.5\n",
      "2001.0,36.0,2.5\n",
      "2001.0,37.0,2.5\n",
      "2001.0,38.0,-2.0\n",
      "2001.0,39.0,-4.5\n",
      "2001.0,40.0,-4.0\n",
      "2001.0,41.0,-1.0\n",
      "2001.0,42.0,1.0\n",
      "2001.0,43.0,2.0\n",
      "2001.0,44.0,1.0\n",
      "2001.0,45.0,0.0\n",
      "2001.0,46.0,-1.0\n",
      "2001.0,47.0,1.5\n",
      "2001.0,48.0,5.0\n",
      "2001.0,49.0,5.5\n",
      "2001.0,50.0,6.0\n",
      "2001.0,51.0,6.0\n",
      "2001.0,52.0,4.0\n",
      "2001.0,53.0,5.0\n",
      "2001.0,54.0,4.0\n",
      "2001.0,55.0,2.5\n",
      "2001.0,56.0,3.5\n",
      "2001.0,57.0,5.0\n",
      "2001.0,58.0,5.0\n",
      "2001.0,59.0,5.0\n",
      "2001.0,60.0,6.5\n",
      "2001.0,61.0,5.0\n",
      "2001.0,62.0,3.0\n",
      "2001.0,63.0,7.0\n",
      "2001.0,64.0,10.0\n",
      "2001.0,65.0,12.5\n",
      "2001.0,66.0,13.0\n",
      "2001.0,67.0,13.0\n",
      "2001.0,68.0,8.5\n",
      "2001.0,69.0,4.5\n",
      "2001.0,70.0,3.5\n",
      "2001.0,71.0,5.5\n",
      "2001.0,72.0,9.5\n",
      "2001.0,73.0,7.5\n",
      "2001.0,74.0,4.0\n",
      "2001.0,75.0,4.0\n",
      "2001.0,76.0,5.0\n",
      "2001.0,77.0,8.5\n",
      "2001.0,78.0,9.5\n",
      "2001.0,79.0,11.5\n",
      "2001.0,80.0,13.0\n",
      "2001.0,81.0,15.0\n",
      "2001.0,82.0,17.5\n",
      "2001.0,83.0,16.5\n",
      "2001.0,84.0,12.5\n",
      "2001.0,85.0,8.0\n",
      "2001.0,86.0,6.5\n",
      "2001.0,87.0,8.0\n",
      "2001.0,88.0,10.0\n",
      "2001.0,89.0,8.5\n",
      "2001.0,90.0,6.5\n",
      "2001.0,91.0,9.0\n",
      "2001.0,92.0,7.0\n",
      "2001.0,93.0,6.5\n",
      "2001.0,94.0,7.0\n",
      "2001.0,95.0,9.5\n",
      "2001.0,96.0,9.5\n",
      "2001.0,97.0,7.0\n",
      "2001.0,98.0,4.0\n",
      "2001.0,99.0,3.5\n",
      "2001.0,100.0,4.5\n",
      "2001.0,101.0,5.5\n",
      "2001.0,102.0,5.0\n",
      "2001.0,103.0,5.0\n",
      "2001.0,104.0,6.5\n",
      "2001.0,105.0,10.0\n",
      "2001.0,106.0,15.0\n",
      "2001.0,107.0,18.5\n",
      "2001.0,108.0,17.0\n",
      "2001.0,109.0,12.5\n",
      "2001.0,110.0,7.0\n",
      "2001.0,111.0,7.0\n",
      "2001.0,112.0,9.5\n",
      "2001.0,113.0,11.5\n",
      "2001.0,114.0,16.5\n",
      "2001.0,115.0,21.5\n",
      "2001.0,116.0,24.0\n",
      "2001.0,117.0,22.5\n",
      "2001.0,118.0,19.5\n",
      "2001.0,119.0,13.5\n",
      "2001.0,120.0,12.0\n",
      "2001.0,121.0,9.0\n",
      "2001.0,122.0,6.5\n",
      "2001.0,123.0,11.0\n",
      "2001.0,124.0,17.5\n",
      "2001.0,125.0,16.0\n",
      "2001.0,126.0,13.0\n",
      "2001.0,127.0,18.5\n",
      "2001.0,128.0,21.5\n",
      "2001.0,129.0,20.0\n",
      "2001.0,130.0,19.0\n",
      "2001.0,131.0,22.5\n",
      "2001.0,132.0,26.5\n",
      "2001.0,133.0,24.5\n",
      "2001.0,134.0,21.0\n",
      "2001.0,135.0,16.5\n",
      "2001.0,136.0,13.0\n",
      "2001.0,137.0,15.0\n",
      "2001.0,138.0,16.5\n",
      "2001.0,139.0,18.0\n",
      "2001.0,140.0,16.5\n",
      "2001.0,141.0,17.5\n",
      "2001.0,142.0,24.0\n",
      "2001.0,143.0,29.0\n",
      "2001.0,144.0,30.0\n",
      "2001.0,145.0,28.5\n",
      "2001.0,146.0,27.0\n",
      "2001.0,147.0,26.0\n",
      "2001.0,148.0,23.5\n",
      "2001.0,149.0,20.5\n",
      "2001.0,150.0,19.0\n",
      "2001.0,151.0,23.5\n",
      "2001.0,152.0,28.0\n",
      "2001.0,153.0,23.0\n",
      "2001.0,154.0,11.5\n",
      "2001.0,155.0,8.5\n",
      "2001.0,156.0,10.0\n",
      "2001.0,157.0,15.0\n",
      "2001.0,158.0,18.5\n",
      "2001.0,159.0,23.0\n",
      "2001.0,160.0,25.0\n",
      "2001.0,161.0,22.0\n",
      "2001.0,162.0,19.0\n",
      "2001.0,163.0,13.5\n",
      "2001.0,164.0,9.5\n",
      "2001.0,165.0,13.5\n",
      "2001.0,166.0,18.5\n",
      "2001.0,167.0,23.0\n",
      "2001.0,168.0,24.5\n",
      "2001.0,169.0,21.5\n",
      "2001.0,170.0,22.0\n",
      "2001.0,171.0,26.5\n",
      "2001.0,172.0,30.0\n",
      "2001.0,173.0,31.0\n",
      "2001.0,174.0,30.5\n",
      "2001.0,175.0,28.0\n",
      "2001.0,176.0,24.5\n",
      "2001.0,177.0,24.5\n",
      "2001.0,178.0,26.0\n",
      "2001.0,179.0,25.0\n",
      "2001.0,180.0,27.5\n",
      "2001.0,181.0,30.5\n",
      "2001.0,182.0,31.0\n",
      "2001.0,183.0,32.0\n",
      "2001.0,184.0,34.0\n",
      "2001.0,185.0,34.5\n",
      "2001.0,186.0,28.5\n",
      "2001.0,187.0,28.0\n",
      "2001.0,188.0,29.5\n",
      "2001.0,189.0,28.5\n",
      "2001.0,190.0,27.5\n",
      "2001.0,191.0,29.0\n",
      "2001.0,192.0,29.5\n",
      "2001.0,193.0,28.0\n",
      "2001.0,194.0,27.5\n",
      "2001.0,195.0,28.5\n",
      "2001.0,196.0,26.0\n",
      "2001.0,197.0,20.5\n",
      "2001.0,198.0,19.0\n",
      "2001.0,199.0,21.0\n",
      "2001.0,200.0,23.0\n",
      "2001.0,201.0,25.5\n",
      "2001.0,202.0,26.0\n",
      "2001.0,203.0,25.0\n",
      "2001.0,204.0,26.0\n",
      "2001.0,205.0,28.0\n",
      "2001.0,206.0,29.0\n",
      "2001.0,207.0,28.5\n",
      "2001.0,208.0,30.0\n",
      "2001.0,209.0,27.5\n",
      "2001.0,210.0,24.0\n",
      "2001.0,211.0,19.5\n",
      "2001.0,212.0,17.0\n",
      "2001.0,213.0,23.5\n",
      "2001.0,214.0,30.0\n",
      "2001.0,215.0,32.0\n",
      "2001.0,216.0,29.0\n",
      "2001.0,217.0,28.5\n",
      "2001.0,218.0,33.0\n",
      "2001.0,219.0,34.5\n",
      "2001.0,220.0,32.5\n",
      "2001.0,221.0,30.5\n",
      "2001.0,222.0,31.0\n",
      "2001.0,223.0,30.5\n",
      "2001.0,224.0,31.5\n",
      "2001.0,225.0,32.0\n",
      "2001.0,226.0,30.0\n",
      "2001.0,227.0,30.5\n",
      "2001.0,228.0,31.5\n",
      "2001.0,229.0,33.0\n",
      "2001.0,230.0,32.5\n",
      "2001.0,231.0,29.0\n",
      "2001.0,232.0,29.0\n",
      "2001.0,233.0,29.5\n",
      "2001.0,234.0,29.0\n",
      "2001.0,235.0,26.5\n",
      "2001.0,236.0,25.5\n",
      "2001.0,237.0,28.0\n",
      "2001.0,238.0,32.0\n",
      "2001.0,239.0,33.0\n",
      "2001.0,240.0,30.5\n",
      "2001.0,241.0,30.0\n",
      "2001.0,242.0,31.0\n",
      "2001.0,243.0,30.0\n",
      "2001.0,244.0,29.0\n",
      "2001.0,245.0,29.0\n",
      "2001.0,246.0,30.0\n",
      "2001.0,247.0,30.0\n",
      "2001.0,248.0,26.0\n",
      "2001.0,249.0,19.0\n",
      "2001.0,250.0,14.5\n",
      "2001.0,251.0,17.0\n",
      "2001.0,252.0,22.5\n",
      "2001.0,253.0,26.5\n",
      "2001.0,254.0,28.5\n",
      "2001.0,255.0,28.0\n",
      "2001.0,256.0,26.5\n",
      "2001.0,257.0,25.0\n",
      "2001.0,258.0,26.0\n",
      "2001.0,259.0,25.0\n",
      "2001.0,260.0,24.0\n",
      "2001.0,261.0,23.5\n",
      "2001.0,262.0,23.0\n",
      "2001.0,263.0,23.5\n",
      "2001.0,264.0,24.0\n",
      "2001.0,265.0,25.5\n",
      "2001.0,266.0,28.0\n",
      "2001.0,267.0,29.0\n",
      "2001.0,268.0,29.0\n",
      "2001.0,269.0,24.5\n",
      "2001.0,270.0,24.5\n",
      "2001.0,271.0,23.0\n",
      "2001.0,272.0,22.0\n",
      "2001.0,273.0,24.5\n",
      "2001.0,274.0,26.0\n",
      "2001.0,275.0,24.0\n",
      "2001.0,276.0,22.5\n",
      "2001.0,277.0,19.0\n",
      "2001.0,278.0,19.0\n",
      "2001.0,279.0,19.5\n",
      "2001.0,280.0,18.5\n",
      "2001.0,281.0,14.0\n",
      "2001.0,282.0,9.0\n",
      "2001.0,283.0,9.0\n",
      "2001.0,284.0,9.0\n",
      "2001.0,285.0,6.5\n",
      "2001.0,286.0,8.0\n",
      "2001.0,287.0,9.5\n",
      "2001.0,288.0,11.0\n",
      "2001.0,289.0,14.5\n",
      "2001.0,290.0,14.0\n",
      "2001.0,291.0,11.0\n",
      "2001.0,292.0,11.0\n",
      "2001.0,293.0,13.5\n",
      "2001.0,294.0,13.5\n",
      "2001.0,295.0,11.0\n",
      "2001.0,296.0,7.5\n",
      "2001.0,297.0,5.0\n",
      "2001.0,298.0,7.0\n",
      "2001.0,299.0,11.5\n",
      "2001.0,300.0,14.5\n",
      "2001.0,301.0,12.5\n",
      "2001.0,302.0,11.0\n",
      "2001.0,303.0,11.0\n",
      "2001.0,304.0,9.5\n",
      "2001.0,305.0,7.5\n",
      "2001.0,306.0,8.0\n",
      "2001.0,307.0,10.0\n",
      "2001.0,308.0,12.0\n",
      "2001.0,309.0,12.0\n",
      "2001.0,310.0,11.0\n",
      "2001.0,311.0,9.0\n",
      "2001.0,312.0,9.0\n",
      "2001.0,313.0,10.0\n",
      "2001.0,314.0,10.0\n",
      "2001.0,315.0,10.0\n",
      "2001.0,316.0,10.5\n",
      "2001.0,317.0,10.5\n",
      "2001.0,318.0,8.5\n",
      "2001.0,319.0,10.5\n",
      "2001.0,320.0,11.5\n",
      "2001.0,321.0,9.0\n",
      "2001.0,322.0,6.0\n",
      "2001.0,323.0,8.5\n",
      "2001.0,324.0,9.0\n",
      "2001.0,325.0,7.5\n",
      "2001.0,326.0,4.5\n",
      "2001.0,327.0,3.5\n",
      "2001.0,328.0,1.0\n",
      "2001.0,329.0,0.5\n",
      "2001.0,330.0,0.5\n",
      "2001.0,331.0,0.5\n",
      "2001.0,332.0,-1.0\n",
      "2001.0,333.0,0.5\n",
      "2001.0,334.0,1.0\n",
      "2001.0,335.0,0.5\n",
      "2001.0,336.0,0.5\n",
      "2001.0,337.0,0.5\n",
      "2001.0,338.0,-0.5\n",
      "2001.0,339.0,-1.5\n",
      "2001.0,340.0,-1.0\n",
      "2001.0,341.0,1.0\n",
      "2001.0,342.0,1.5\n",
      "2001.0,343.0,1.5\n",
      "2001.0,344.0,-1.5\n",
      "2001.0,345.0,-3.5\n",
      "2001.0,346.0,-2.0\n",
      "2001.0,347.0,-1.0\n",
      "2001.0,348.0,0.5\n",
      "2001.0,349.0,-0.5\n",
      "2001.0,350.0,-2.0\n",
      "2001.0,351.0,0.0\n",
      "2001.0,352.0,-0.5\n",
      "2001.0,353.0,-0.5\n",
      "2001.0,354.0,1.5\n",
      "2001.0,355.0,1.0\n",
      "2001.0,356.0,0.0\n",
      "2001.0,357.0,-1.5\n",
      "2001.0,358.0,-3.5\n",
      "2001.0,359.0,-4.0\n",
      "2001.0,360.0,-6.5\n",
      "2001.0,361.0,-3.5\n",
      "2001.0,362.0,-2.0\n",
      "2001.0,363.0,-1.0\n",
      "2001.0,364.0,-1.5\n",
      "2001.0,365.0,-1.0\n",
      "2002.0,1.0,-0.5\n",
      "2002.0,2.0,2.0\n",
      "2002.0,3.0,1.5\n",
      "2002.0,4.0,2.0\n",
      "2002.0,5.0,0.0\n",
      "2002.0,6.0,1.5\n",
      "2002.0,7.0,4.0\n",
      "2002.0,8.0,5.0\n",
      "2002.0,9.0,4.0\n",
      "2002.0,10.0,2.5\n",
      "2002.0,11.0,1.5\n",
      "2002.0,12.0,4.0\n",
      "2002.0,13.0,3.0\n",
      "2002.0,14.0,1.5\n",
      "2002.0,15.0,-2.0\n",
      "2002.0,16.0,-2.0\n",
      "2002.0,17.0,-2.5\n",
      "2002.0,18.0,-3.5\n",
      "2002.0,19.0,-2.5\n",
      "2002.0,20.0,-2.0\n",
      "2002.0,21.0,-1.5\n",
      "2002.0,22.0,-2.5\n",
      "2002.0,23.0,-3.5\n",
      "2002.0,24.0,-1.5\n",
      "2002.0,25.0,0.5\n",
      "2002.0,26.0,1.0\n",
      "2002.0,27.0,0.5\n",
      "2002.0,28.0,-2.5\n",
      "2002.0,29.0,-7.0\n",
      "2002.0,30.0,-7.5\n",
      "2002.0,31.0,-3.5\n",
      "2002.0,32.0,-1.5\n",
      "2002.0,33.0,1.0\n",
      "2002.0,34.0,1.0\n",
      "2002.0,35.0,2.0\n",
      "2002.0,36.0,1.5\n",
      "2002.0,37.0,4.0\n",
      "2002.0,38.0,4.0\n",
      "2002.0,39.0,2.5\n",
      "2002.0,40.0,3.0\n",
      "2002.0,41.0,3.5\n",
      "2002.0,42.0,4.5\n",
      "2002.0,43.0,4.5\n",
      "2002.0,44.0,3.0\n",
      "2002.0,45.0,3.5\n",
      "2002.0,46.0,5.0\n",
      "2002.0,47.0,5.5\n",
      "2002.0,48.0,7.5\n",
      "2002.0,49.0,6.5\n",
      "2002.0,50.0,4.5\n",
      "2002.0,51.0,5.0\n",
      "2002.0,52.0,5.5\n",
      "2002.0,53.0,7.5\n",
      "2002.0,54.0,7.0\n",
      "2002.0,55.0,3.0\n",
      "2002.0,56.0,0.5\n",
      "2002.0,57.0,-1.5\n",
      "2002.0,58.0,0.0\n",
      "2002.0,59.0,0.5\n",
      "2002.0,60.0,-2.5\n",
      "2002.0,61.0,-1.5\n",
      "2002.0,62.0,-0.5\n",
      "2002.0,63.0,4.5\n",
      "2002.0,64.0,7.0\n",
      "2002.0,65.0,6.5\n",
      "2002.0,66.0,4.5\n",
      "2002.0,67.0,-2.0\n",
      "2002.0,68.0,-1.0\n",
      "2002.0,69.0,2.0\n",
      "2002.0,70.0,5.0\n",
      "2002.0,71.0,5.0\n",
      "2002.0,72.0,3.5\n",
      "2002.0,73.0,2.5\n",
      "2002.0,74.0,2.5\n",
      "2002.0,75.0,1.0\n",
      "2002.0,76.0,1.5\n",
      "2002.0,77.0,-0.5\n",
      "2002.0,78.0,2.0\n",
      "2002.0,79.0,4.5\n",
      "2002.0,80.0,10.0\n",
      "2002.0,81.0,15.0\n",
      "2002.0,82.0,12.5\n",
      "2002.0,83.0,6.5\n",
      "2002.0,84.0,8.0\n",
      "2002.0,85.0,9.0\n",
      "2002.0,86.0,8.5\n",
      "2002.0,87.0,7.5\n",
      "2002.0,88.0,8.0\n",
      "2002.0,89.0,9.0\n",
      "2002.0,90.0,13.5\n",
      "2002.0,91.0,12.5\n",
      "2002.0,92.0,10.0\n",
      "2002.0,93.0,11.0\n",
      "2002.0,94.0,15.0\n",
      "2002.0,95.0,18.5\n",
      "2002.0,96.0,14.5\n",
      "2002.0,97.0,9.5\n",
      "2002.0,98.0,12.0\n",
      "2002.0,99.0,12.5\n",
      "2002.0,100.0,10.0\n",
      "2002.0,101.0,9.5\n",
      "2002.0,102.0,11.5\n",
      "2002.0,103.0,13.5\n",
      "2002.0,104.0,13.0\n",
      "2002.0,105.0,8.5\n",
      "2002.0,106.0,5.5\n",
      "2002.0,107.0,5.0\n",
      "2002.0,108.0,4.0\n",
      "2002.0,109.0,7.5\n",
      "2002.0,110.0,10.5\n",
      "2002.0,111.0,11.5\n",
      "2002.0,112.0,13.5\n",
      "2002.0,113.0,12.0\n",
      "2002.0,114.0,8.0\n",
      "2002.0,115.0,13.5\n",
      "2002.0,116.0,14.5\n",
      "2002.0,117.0,10.5\n",
      "2002.0,118.0,12.0\n",
      "2002.0,119.0,16.0\n",
      "2002.0,120.0,17.0\n",
      "2002.0,121.0,16.0\n",
      "2002.0,122.0,16.5\n",
      "2002.0,123.0,15.0\n",
      "2002.0,124.0,13.5\n",
      "2002.0,125.0,12.0\n",
      "2002.0,126.0,9.5\n",
      "2002.0,127.0,7.0\n",
      "2002.0,128.0,5.0\n",
      "2002.0,129.0,7.5\n",
      "2002.0,130.0,8.5\n",
      "2002.0,131.0,12.5\n",
      "2002.0,132.0,17.5\n",
      "2002.0,133.0,21.5\n",
      "2002.0,134.0,19.5\n",
      "2002.0,135.0,14.5\n",
      "2002.0,136.0,15.5\n",
      "2002.0,137.0,19.0\n",
      "2002.0,138.0,21.0\n",
      "2002.0,139.0,26.0\n",
      "2002.0,140.0,24.0\n",
      "2002.0,141.0,10.0\n",
      "2002.0,142.0,7.0\n",
      "2002.0,143.0,9.5\n",
      "2002.0,144.0,14.5\n",
      "2002.0,145.0,19.0\n",
      "2002.0,146.0,19.5\n",
      "2002.0,147.0,20.5\n",
      "2002.0,148.0,22.0\n",
      "2002.0,149.0,23.5\n",
      "2002.0,150.0,24.5\n",
      "2002.0,151.0,24.5\n",
      "2002.0,152.0,22.5\n",
      "2002.0,153.0,17.5\n",
      "2002.0,154.0,19.0\n",
      "2002.0,155.0,21.0\n",
      "2002.0,156.0,22.0\n",
      "2002.0,157.0,22.0\n",
      "2002.0,158.0,20.0\n",
      "2002.0,159.0,11.5\n",
      "2002.0,160.0,8.5\n",
      "2002.0,161.0,11.0\n",
      "2002.0,162.0,15.0\n",
      "2002.0,163.0,20.5\n",
      "2002.0,164.0,25.0\n",
      "2002.0,165.0,28.5\n",
      "2002.0,166.0,28.5\n",
      "2002.0,167.0,29.5\n",
      "2002.0,168.0,25.5\n",
      "2002.0,169.0,18.0\n",
      "2002.0,170.0,17.5\n",
      "2002.0,171.0,23.0\n",
      "2002.0,172.0,25.5\n",
      "2002.0,173.0,24.5\n",
      "2002.0,174.0,23.0\n",
      "2002.0,175.0,26.0\n",
      "2002.0,176.0,30.0\n",
      "2002.0,177.0,32.0\n",
      "2002.0,178.0,31.5\n",
      "2002.0,179.0,29.5\n",
      "2002.0,180.0,24.5\n",
      "2002.0,181.0,23.5\n",
      "2002.0,182.0,24.0\n",
      "2002.0,183.0,26.5\n",
      "2002.0,184.0,28.5\n",
      "2002.0,185.0,27.5\n",
      "2002.0,186.0,27.5\n",
      "2002.0,187.0,30.5\n",
      "2002.0,188.0,30.0\n",
      "2002.0,189.0,25.5\n",
      "2002.0,190.0,27.5\n",
      "2002.0,191.0,34.0\n",
      "2002.0,192.0,37.5\n",
      "2002.0,193.0,38.5\n",
      "2002.0,194.0,38.0\n",
      "2002.0,195.0,35.0\n",
      "2002.0,196.0,33.5\n",
      "2002.0,197.0,31.5\n",
      "2002.0,198.0,31.0\n",
      "2002.0,199.0,31.5\n",
      "2002.0,200.0,29.0\n",
      "2002.0,201.0,27.0\n",
      "2002.0,202.0,29.0\n",
      "2002.0,203.0,30.5\n",
      "2002.0,204.0,31.0\n",
      "2002.0,205.0,32.0\n",
      "2002.0,206.0,30.0\n",
      "2002.0,207.0,28.0\n",
      "2002.0,208.0,25.0\n",
      "2002.0,209.0,26.0\n",
      "2002.0,210.0,28.0\n",
      "2002.0,211.0,29.0\n",
      "2002.0,212.0,26.5\n",
      "2002.0,213.0,28.0\n",
      "2002.0,214.0,28.0\n",
      "2002.0,215.0,28.0\n",
      "2002.0,216.0,27.0\n",
      "2002.0,217.0,25.0\n",
      "2002.0,218.0,22.5\n",
      "2002.0,219.0,20.0\n",
      "2002.0,220.0,19.0\n",
      "2002.0,221.0,23.0\n",
      "2002.0,222.0,28.0\n",
      "2002.0,223.0,27.5\n",
      "2002.0,224.0,26.0\n",
      "2002.0,225.0,28.0\n",
      "2002.0,226.0,29.5\n",
      "2002.0,227.0,30.0\n",
      "2002.0,228.0,27.5\n",
      "2002.0,229.0,27.5\n",
      "2002.0,230.0,27.0\n",
      "2002.0,231.0,28.0\n",
      "2002.0,232.0,25.5\n",
      "2002.0,233.0,20.5\n",
      "2002.0,234.0,17.0\n",
      "2002.0,235.0,21.5\n",
      "2002.0,236.0,24.5\n",
      "2002.0,237.0,24.0\n",
      "2002.0,238.0,23.5\n",
      "2002.0,239.0,22.5\n",
      "2002.0,240.0,25.0\n",
      "2002.0,241.0,23.0\n",
      "2002.0,242.0,22.5\n",
      "2002.0,243.0,24.5\n",
      "2002.0,244.0,25.5\n",
      "2002.0,245.0,27.0\n",
      "2002.0,246.0,28.0\n",
      "2002.0,247.0,28.0\n",
      "2002.0,248.0,26.5\n",
      "2002.0,249.0,21.5\n",
      "2002.0,250.0,15.5\n",
      "2002.0,251.0,16.0\n",
      "2002.0,252.0,21.0\n",
      "2002.0,253.0,25.0\n",
      "2002.0,254.0,28.0\n",
      "2002.0,255.0,29.0\n",
      "2002.0,256.0,28.5\n",
      "2002.0,257.0,29.0\n",
      "2002.0,258.0,29.0\n",
      "2002.0,259.0,26.5\n",
      "2002.0,260.0,19.0\n",
      "2002.0,261.0,15.0\n",
      "2002.0,262.0,20.0\n",
      "2002.0,263.0,20.5\n",
      "2002.0,264.0,18.5\n",
      "2002.0,265.0,20.5\n",
      "2002.0,266.0,24.0\n",
      "2002.0,267.0,24.0\n",
      "2002.0,268.0,22.5\n",
      "2002.0,269.0,20.5\n",
      "2002.0,270.0,16.0\n",
      "2002.0,271.0,17.0\n",
      "2002.0,272.0,18.0\n",
      "2002.0,273.0,12.0\n",
      "2002.0,274.0,9.0\n",
      "2002.0,275.0,11.0\n",
      "2002.0,276.0,11.0\n",
      "2002.0,277.0,11.0\n",
      "2002.0,278.0,11.0\n",
      "2002.0,279.0,15.5\n",
      "2002.0,280.0,18.5\n",
      "2002.0,281.0,19.0\n",
      "2002.0,282.0,18.0\n",
      "2002.0,283.0,17.5\n",
      "2002.0,284.0,11.5\n",
      "2002.0,285.0,10.5\n",
      "2002.0,286.0,15.5\n",
      "2002.0,287.0,16.5\n",
      "2002.0,288.0,17.0\n",
      "2002.0,289.0,18.5\n",
      "2002.0,290.0,19.0\n",
      "2002.0,291.0,18.0\n",
      "2002.0,292.0,17.0\n",
      "2002.0,293.0,14.0\n",
      "2002.0,294.0,14.5\n",
      "2002.0,295.0,12.5\n",
      "2002.0,296.0,11.0\n",
      "2002.0,297.0,11.0\n",
      "2002.0,298.0,11.5\n",
      "2002.0,299.0,11.5\n",
      "2002.0,300.0,10.0\n",
      "2002.0,301.0,6.5\n",
      "2002.0,302.0,2.5\n",
      "2002.0,303.0,-1.5\n",
      "2002.0,304.0,-1.5\n",
      "2002.0,305.0,1.5\n",
      "2002.0,306.0,3.5\n",
      "2002.0,307.0,5.0\n",
      "2002.0,308.0,6.5\n",
      "2002.0,309.0,6.5\n",
      "2002.0,310.0,9.0\n",
      "2002.0,311.0,10.5\n",
      "2002.0,312.0,9.5\n",
      "2002.0,313.0,7.0\n",
      "2002.0,314.0,4.5\n",
      "2002.0,315.0,5.0\n",
      "2002.0,316.0,4.5\n",
      "2002.0,317.0,5.0\n",
      "2002.0,318.0,5.0\n",
      "2002.0,319.0,6.0\n",
      "2002.0,320.0,6.5\n",
      "2002.0,321.0,6.5\n",
      "2002.0,322.0,5.0\n",
      "2002.0,323.0,5.0\n",
      "2002.0,324.0,9.0\n",
      "2002.0,325.0,10.5\n",
      "2002.0,326.0,10.5\n",
      "2002.0,327.0,6.5\n",
      "2002.0,328.0,2.5\n",
      "2002.0,329.0,2.5\n",
      "2002.0,330.0,3.0\n",
      "2002.0,331.0,4.0\n",
      "2002.0,332.0,5.5\n",
      "2002.0,333.0,5.5\n",
      "2002.0,334.0,7.0\n",
      "2002.0,335.0,5.5\n",
      "2002.0,336.0,4.0\n",
      "2002.0,337.0,4.0\n",
      "2002.0,338.0,3.0\n",
      "2002.0,339.0,3.5\n",
      "2002.0,340.0,3.5\n",
      "2002.0,341.0,3.0\n",
      "2002.0,342.0,2.5\n",
      "2002.0,343.0,1.0\n",
      "2002.0,344.0,1.0\n",
      "2002.0,345.0,1.0\n",
      "2002.0,346.0,4.5\n",
      "2002.0,347.0,4.5\n",
      "2002.0,348.0,6.0\n",
      "2002.0,349.0,7.0\n",
      "2002.0,350.0,5.5\n",
      "2002.0,351.0,4.0\n",
      "2002.0,352.0,0.5\n",
      "2002.0,353.0,-1.0\n",
      "2002.0,354.0,-0.5\n",
      "2002.0,355.0,0.0\n",
      "2002.0,356.0,-0.5\n",
      "2002.0,357.0,-1.0\n",
      "2002.0,358.0,-1.0\n",
      "2002.0,359.0,-2.0\n",
      "2002.0,360.0,-1.5\n",
      "2002.0,361.0,1.0\n",
      "2002.0,362.0,3.5\n",
      "2002.0,363.0,2.5\n",
      "2002.0,364.0,0.5\n",
      "2002.0,365.0,1.5\n",
      "2003.0,1.0,0.5\n",
      "2003.0,2.0,1.0\n",
      "2003.0,3.0,2.5\n",
      "2003.0,4.0,4.0\n",
      "2003.0,5.0,4.0\n",
      "2003.0,6.0,4.5\n",
      "2003.0,7.0,3.0\n",
      "2003.0,8.0,3.0\n",
      "2003.0,9.0,1.0\n",
      "2003.0,10.0,0.5\n",
      "2003.0,11.0,0.0\n",
      "2003.0,12.0,3.0\n",
      "2003.0,13.0,4.0\n",
      "2003.0,14.0,3.0\n",
      "2003.0,15.0,1.5\n",
      "2003.0,16.0,1.0\n",
      "2003.0,17.0,1.5\n",
      "2003.0,18.0,3.0\n",
      "2003.0,19.0,3.0\n",
      "2003.0,20.0,1.5\n",
      "2003.0,21.0,2.5\n",
      "2003.0,22.0,3.0\n",
      "2003.0,23.0,3.0\n",
      "2003.0,24.0,4.5\n",
      "2003.0,25.0,4.5\n",
      "2003.0,26.0,5.0\n",
      "2003.0,27.0,4.5\n",
      "2003.0,28.0,4.0\n",
      "2003.0,29.0,3.0\n",
      "2003.0,30.0,3.0\n",
      "2003.0,31.0,6.0\n",
      "2003.0,32.0,6.0\n",
      "2003.0,33.0,2.0\n",
      "2003.0,34.0,1.5\n",
      "2003.0,35.0,1.0\n",
      "2003.0,36.0,-1.0\n",
      "2003.0,37.0,0.0\n",
      "2003.0,38.0,0.5\n",
      "2003.0,39.0,-1.0\n",
      "2003.0,40.0,1.0\n",
      "2003.0,41.0,4.0\n",
      "2003.0,42.0,6.0\n",
      "2003.0,43.0,6.5\n",
      "2003.0,44.0,3.5\n",
      "2003.0,45.0,2.0\n",
      "2003.0,46.0,4.0\n",
      "2003.0,47.0,4.0\n",
      "2003.0,48.0,2.5\n",
      "2003.0,49.0,4.0\n",
      "2003.0,50.0,2.5\n",
      "2003.0,51.0,3.5\n",
      "2003.0,52.0,3.0\n",
      "2003.0,53.0,2.5\n",
      "2003.0,54.0,0.0\n",
      "2003.0,55.0,-1.5\n",
      "2003.0,56.0,-1.0\n",
      "2003.0,57.0,1.0\n",
      "2003.0,58.0,3.0\n",
      "2003.0,59.0,2.0\n",
      "2003.0,60.0,2.0\n",
      "2003.0,61.0,4.5\n",
      "2003.0,62.0,3.5\n",
      "2003.0,63.0,1.5\n",
      "2003.0,64.0,1.0\n",
      "2003.0,65.0,1.0\n",
      "2003.0,66.0,2.0\n",
      "2003.0,67.0,4.0\n",
      "2003.0,68.0,5.5\n",
      "2003.0,69.0,7.5\n",
      "2003.0,70.0,10.0\n",
      "2003.0,71.0,12.0\n",
      "2003.0,72.0,14.5\n",
      "2003.0,73.0,14.0\n",
      "2003.0,74.0,12.0\n",
      "2003.0,75.0,10.5\n",
      "2003.0,76.0,7.0\n",
      "2003.0,77.0,7.0\n",
      "2003.0,78.0,10.5\n",
      "2003.0,79.0,11.5\n",
      "2003.0,80.0,7.5\n",
      "2003.0,81.0,9.0\n",
      "2003.0,82.0,6.0\n",
      "2003.0,83.0,6.0\n",
      "2003.0,84.0,7.0\n",
      "2003.0,85.0,5.0\n",
      "2003.0,86.0,3.5\n",
      "2003.0,87.0,6.0\n",
      "2003.0,88.0,10.0\n",
      "2003.0,89.0,14.5\n",
      "2003.0,90.0,16.0\n",
      "2003.0,91.0,12.5\n",
      "2003.0,92.0,6.0\n",
      "2003.0,93.0,5.0\n",
      "2003.0,94.0,5.5\n",
      "2003.0,95.0,6.0\n",
      "2003.0,96.0,5.0\n",
      "2003.0,97.0,8.0\n",
      "2003.0,98.0,14.5\n",
      "2003.0,99.0,18.0\n",
      "2003.0,100.0,19.5\n",
      "2003.0,101.0,19.5\n",
      "2003.0,102.0,18.5\n",
      "2003.0,103.0,14.0\n",
      "2003.0,104.0,10.5\n",
      "2003.0,105.0,9.0\n",
      "2003.0,106.0,10.5\n",
      "2003.0,107.0,12.0\n",
      "2003.0,108.0,9.0\n",
      "2003.0,109.0,11.0\n",
      "2003.0,110.0,17.0\n",
      "2003.0,111.0,18.5\n",
      "2003.0,112.0,18.5\n",
      "2003.0,113.0,14.0\n",
      "2003.0,114.0,14.5\n",
      "2003.0,115.0,13.0\n",
      "2003.0,116.0,8.5\n",
      "2003.0,117.0,10.5\n",
      "2003.0,118.0,13.0\n",
      "2003.0,119.0,12.0\n",
      "2003.0,120.0,9.5\n",
      "2003.0,121.0,11.0\n",
      "2003.0,122.0,14.5\n",
      "2003.0,123.0,13.0\n",
      "2003.0,124.0,11.0\n",
      "2003.0,125.0,9.0\n",
      "2003.0,126.0,9.5\n",
      "2003.0,127.0,10.5\n",
      "2003.0,128.0,13.5\n",
      "2003.0,129.0,13.5\n",
      "2003.0,130.0,13.5\n",
      "2003.0,131.0,13.5\n",
      "2003.0,132.0,11.5\n",
      "2003.0,133.0,15.0\n",
      "2003.0,134.0,21.0\n",
      "2003.0,135.0,21.0\n",
      "2003.0,136.0,15.5\n",
      "2003.0,137.0,11.0\n",
      "2003.0,138.0,9.0\n",
      "2003.0,139.0,11.0\n",
      "2003.0,140.0,17.0\n",
      "2003.0,141.0,19.5\n",
      "2003.0,142.0,22.0\n",
      "2003.0,143.0,26.0\n",
      "2003.0,144.0,30.0\n",
      "2003.0,145.0,28.5\n",
      "2003.0,146.0,23.0\n",
      "2003.0,147.0,26.0\n",
      "2003.0,148.0,30.5\n",
      "2003.0,149.0,32.0\n",
      "2003.0,150.0,27.5\n",
      "2003.0,151.0,23.0\n",
      "2003.0,152.0,22.5\n",
      "2003.0,153.0,21.5\n",
      "2003.0,154.0,20.5\n",
      "2003.0,155.0,22.0\n",
      "2003.0,156.0,24.0\n",
      "2003.0,157.0,25.0\n",
      "2003.0,158.0,25.5\n",
      "2003.0,159.0,26.5\n",
      "2003.0,160.0,26.0\n",
      "2003.0,161.0,26.0\n",
      "2003.0,162.0,24.5\n",
      "2003.0,163.0,26.0\n",
      "2003.0,164.0,25.0\n",
      "2003.0,165.0,24.5\n",
      "2003.0,166.0,27.0\n",
      "2003.0,167.0,28.5\n",
      "2003.0,168.0,31.0\n",
      "2003.0,169.0,33.0\n",
      "2003.0,170.0,29.5\n",
      "2003.0,171.0,21.5\n",
      "2003.0,172.0,15.5\n",
      "2003.0,173.0,16.0\n",
      "2003.0,174.0,17.0\n",
      "2003.0,175.0,18.5\n",
      "2003.0,176.0,20.0\n",
      "2003.0,177.0,24.0\n",
      "2003.0,178.0,26.5\n",
      "2003.0,179.0,28.0\n",
      "2003.0,180.0,31.0\n",
      "2003.0,181.0,31.5\n",
      "2003.0,182.0,29.5\n",
      "2003.0,183.0,27.5\n",
      "2003.0,184.0,26.5\n",
      "2003.0,185.0,27.5\n",
      "2003.0,186.0,28.0\n",
      "2003.0,187.0,28.5\n",
      "2003.0,188.0,30.0\n",
      "2003.0,189.0,27.5\n",
      "2003.0,190.0,28.0\n",
      "2003.0,191.0,33.5\n",
      "2003.0,192.0,34.5\n",
      "2003.0,193.0,34.5\n",
      "2003.0,194.0,31.0\n",
      "2003.0,195.0,29.5\n",
      "2003.0,196.0,32.5\n",
      "2003.0,197.0,34.5\n",
      "2003.0,198.0,34.5\n",
      "2003.0,199.0,35.0\n",
      "2003.0,200.0,36.5\n",
      "2003.0,201.0,35.5\n",
      "2003.0,202.0,35.5\n",
      "2003.0,203.0,36.5\n",
      "2003.0,204.0,37.0\n",
      "2003.0,205.0,34.0\n",
      "2003.0,206.0,29.0\n",
      "2003.0,207.0,27.5\n",
      "2003.0,208.0,28.5\n",
      "2003.0,209.0,31.5\n",
      "2003.0,210.0,33.0\n",
      "2003.0,211.0,34.5\n",
      "2003.0,212.0,33.5\n",
      "2003.0,213.0,33.5\n",
      "2003.0,214.0,32.5\n",
      "2003.0,215.0,27.0\n",
      "2003.0,216.0,26.0\n",
      "2003.0,217.0,28.5\n",
      "2003.0,218.0,28.0\n",
      "2003.0,219.0,29.0\n",
      "2003.0,220.0,31.0\n",
      "2003.0,221.0,33.0\n",
      "2003.0,222.0,33.0\n",
      "2003.0,223.0,32.5\n",
      "2003.0,224.0,30.5\n",
      "2003.0,225.0,31.0\n",
      "2003.0,226.0,33.0\n",
      "2003.0,227.0,33.5\n",
      "2003.0,228.0,29.5\n",
      "2003.0,229.0,27.0\n",
      "2003.0,230.0,27.5\n",
      "2003.0,231.0,32.0\n",
      "2003.0,232.0,32.0\n",
      "2003.0,233.0,30.0\n",
      "2003.0,234.0,23.5\n",
      "2003.0,235.0,22.5\n",
      "2003.0,236.0,27.0\n",
      "2003.0,237.0,30.0\n",
      "2003.0,238.0,30.0\n",
      "2003.0,239.0,26.5\n",
      "2003.0,240.0,26.5\n",
      "2003.0,241.0,27.0\n",
      "2003.0,242.0,27.5\n",
      "2003.0,243.0,29.0\n",
      "2003.0,244.0,30.0\n",
      "2003.0,245.0,31.0\n",
      "2003.0,246.0,31.0\n",
      "2003.0,247.0,30.5\n",
      "2003.0,248.0,29.5\n",
      "2003.0,249.0,27.5\n",
      "2003.0,250.0,26.5\n",
      "2003.0,251.0,21.0\n",
      "2003.0,252.0,14.5\n",
      "2003.0,253.0,13.5\n",
      "2003.0,254.0,16.5\n",
      "2003.0,255.0,16.5\n",
      "2003.0,256.0,16.0\n",
      "2003.0,257.0,21.5\n",
      "2003.0,258.0,21.5\n",
      "2003.0,259.0,17.0\n",
      "2003.0,260.0,10.5\n",
      "2003.0,261.0,14.5\n",
      "2003.0,262.0,19.5\n",
      "2003.0,263.0,21.0\n",
      "2003.0,264.0,20.5\n",
      "2003.0,265.0,23.5\n",
      "2003.0,266.0,24.0\n",
      "2003.0,267.0,25.5\n",
      "2003.0,268.0,26.0\n",
      "2003.0,269.0,25.5\n",
      "2003.0,270.0,27.0\n",
      "2003.0,271.0,27.0\n",
      "2003.0,272.0,26.0\n",
      "2003.0,273.0,27.0\n",
      "2003.0,274.0,26.0\n",
      "2003.0,275.0,26.0\n",
      "2003.0,276.0,25.0\n",
      "2003.0,277.0,25.5\n",
      "2003.0,278.0,25.0\n",
      "2003.0,279.0,24.5\n",
      "2003.0,280.0,21.5\n",
      "2003.0,281.0,20.5\n",
      "2003.0,282.0,17.0\n",
      "2003.0,283.0,10.5\n",
      "2003.0,284.0,9.0\n",
      "2003.0,285.0,10.0\n",
      "2003.0,286.0,11.0\n",
      "2003.0,287.0,10.5\n",
      "2003.0,288.0,8.5\n",
      "2003.0,289.0,10.0\n",
      "2003.0,290.0,16.5\n",
      "2003.0,291.0,22.5\n",
      "2003.0,292.0,21.0\n",
      "2003.0,293.0,21.5\n",
      "2003.0,294.0,23.0\n",
      "2003.0,295.0,23.0\n",
      "2003.0,296.0,17.0\n",
      "2003.0,297.0,11.0\n",
      "2003.0,298.0,12.5\n",
      "2003.0,299.0,15.0\n",
      "2003.0,300.0,16.5\n",
      "2003.0,301.0,14.0\n",
      "2003.0,302.0,10.5\n",
      "2003.0,303.0,1.0\n",
      "2003.0,304.0,0.0\n",
      "2003.0,305.0,1.0\n",
      "2003.0,306.0,0.5\n",
      "2003.0,307.0,1.5\n",
      "2003.0,308.0,-1.5\n",
      "2003.0,309.0,-1.5\n",
      "2003.0,310.0,-0.5\n",
      "2003.0,311.0,2.0\n",
      "2003.0,312.0,4.5\n",
      "2003.0,313.0,5.5\n",
      "2003.0,314.0,4.5\n",
      "2003.0,315.0,4.5\n",
      "2003.0,316.0,7.0\n",
      "2003.0,317.0,6.0\n",
      "2003.0,318.0,4.5\n",
      "2003.0,319.0,4.0\n",
      "2003.0,320.0,3.5\n",
      "2003.0,321.0,3.0\n",
      "2003.0,322.0,7.0\n",
      "2003.0,323.0,8.5\n",
      "2003.0,324.0,7.0\n",
      "2003.0,325.0,0.5\n",
      "2003.0,326.0,-3.5\n",
      "2003.0,327.0,-3.0\n",
      "2003.0,328.0,-1.5\n",
      "2003.0,329.0,-0.5\n",
      "2003.0,330.0,-0.5\n",
      "2003.0,331.0,0.5\n",
      "2003.0,332.0,3.0\n",
      "2003.0,333.0,3.0\n",
      "2003.0,334.0,3.5\n",
      "2003.0,335.0,5.0\n",
      "2003.0,336.0,6.0\n",
      "2003.0,337.0,5.0\n",
      "2003.0,338.0,4.0\n",
      "2003.0,339.0,4.0\n",
      "2003.0,340.0,5.0\n",
      "2003.0,341.0,4.0\n",
      "2003.0,342.0,2.5\n",
      "2003.0,343.0,-0.5\n",
      "2003.0,344.0,-1.0\n",
      "2003.0,345.0,-0.5\n",
      "2003.0,346.0,0.5\n",
      "2003.0,347.0,2.0\n",
      "2003.0,348.0,2.0\n",
      "2003.0,349.0,0.0\n",
      "2003.0,350.0,-1.5\n",
      "2003.0,351.0,0.5\n",
      "2003.0,352.0,2.5\n",
      "2003.0,353.0,3.0\n",
      "2003.0,354.0,0.5\n",
      "2003.0,355.0,1.5\n",
      "2003.0,356.0,2.0\n",
      "2003.0,357.0,2.0\n",
      "2003.0,358.0,1.5\n",
      "2003.0,359.0,1.0\n",
      "2003.0,360.0,-1.5\n",
      "2003.0,361.0,-2.5\n",
      "2003.0,362.0,-4.0\n",
      "2003.0,363.0,-3.0\n",
      "2003.0,364.0,-3.5\n",
      "2003.0,365.0,-4.5\n",
      "2004.0,1.0,-2.5\n",
      "2004.0,2.0,-2.0\n",
      "2004.0,3.0,-3.0\n",
      "2004.0,4.0,-5.0\n",
      "2004.0,5.0,-7.0\n",
      "2004.0,6.0,-9.5\n",
      "2004.0,7.0,-4.5\n",
      "2004.0,8.0,0.0\n",
      "2004.0,9.0,4.0\n",
      "2004.0,10.0,4.5\n",
      "2004.0,11.0,5.0\n",
      "2004.0,12.0,4.5\n",
      "2004.0,13.0,4.0\n",
      "2004.0,14.0,4.0\n",
      "2004.0,15.0,2.5\n",
      "2004.0,16.0,2.0\n",
      "2004.0,17.0,2.5\n",
      "2004.0,18.0,0.5\n",
      "2004.0,19.0,1.5\n",
      "2004.0,20.0,3.0\n",
      "2004.0,21.0,2.0\n",
      "2004.0,22.0,1.5\n",
      "2004.0,23.0,0.0\n",
      "2004.0,24.0,0.0\n",
      "2004.0,25.0,-1.5\n",
      "2004.0,26.0,-1.0\n",
      "2004.0,27.0,1.0\n",
      "2004.0,28.0,2.0\n",
      "2004.0,29.0,2.5\n",
      "2004.0,30.0,2.5\n",
      "2004.0,31.0,0.0\n",
      "2004.0,32.0,0.0\n",
      "2004.0,33.0,-1.0\n",
      "2004.0,34.0,0.5\n",
      "2004.0,35.0,0.5\n",
      "2004.0,36.0,2.5\n",
      "2004.0,37.0,2.5\n",
      "2004.0,38.0,0.5\n",
      "2004.0,39.0,0.5\n",
      "2004.0,40.0,1.5\n",
      "2004.0,41.0,2.5\n",
      "2004.0,42.0,3.5\n",
      "2004.0,43.0,4.0\n",
      "2004.0,44.0,4.0\n",
      "2004.0,45.0,2.0\n",
      "2004.0,46.0,1.0\n",
      "2004.0,47.0,1.5\n",
      "2004.0,48.0,4.5\n",
      "2004.0,49.0,5.0\n",
      "2004.0,50.0,5.0\n",
      "2004.0,51.0,5.0\n",
      "2004.0,52.0,5.5\n",
      "2004.0,53.0,5.5\n",
      "2004.0,54.0,7.0\n",
      "2004.0,55.0,6.5\n",
      "2004.0,56.0,4.5\n",
      "2004.0,57.0,5.5\n",
      "2004.0,58.0,7.0\n",
      "2004.0,59.0,6.0\n",
      "2004.0,60.0,6.5\n",
      "2004.0,61.0,6.0\n",
      "2004.0,62.0,4.0\n",
      "2004.0,63.0,3.5\n",
      "2004.0,64.0,2.5\n",
      "2004.0,65.0,3.5\n",
      "2004.0,66.0,3.0\n",
      "2004.0,67.0,6.5\n",
      "2004.0,68.0,11.5\n",
      "2004.0,69.0,13.5\n",
      "2004.0,70.0,10.5\n",
      "2004.0,71.0,11.5\n",
      "2004.0,72.0,12.5\n",
      "2004.0,73.0,12.5\n",
      "2004.0,74.0,10.0\n",
      "2004.0,75.0,11.0\n",
      "2004.0,76.0,11.5\n",
      "2004.0,77.0,12.0\n",
      "2004.0,78.0,14.5\n",
      "2004.0,79.0,13.5\n",
      "2004.0,80.0,14.5\n",
      "2004.0,81.0,18.5\n",
      "2004.0,82.0,19.0\n",
      "2004.0,83.0,18.5\n",
      "2004.0,84.0,17.0\n",
      "2004.0,85.0,13.5\n",
      "2004.0,86.0,10.5\n",
      "2004.0,87.0,8.0\n",
      "2004.0,88.0,10.5\n",
      "2004.0,89.0,17.5\n",
      "2004.0,90.0,21.0\n",
      "2004.0,91.0,17.0\n",
      "2004.0,92.0,7.5\n",
      "2004.0,93.0,7.5\n",
      "2004.0,94.0,15.5\n",
      "2004.0,95.0,18.5\n",
      "2004.0,96.0,19.0\n",
      "2004.0,97.0,17.0\n",
      "2004.0,98.0,17.5\n",
      "2004.0,99.0,16.0\n",
      "2004.0,100.0,13.0\n",
      "2004.0,101.0,13.5\n",
      "2004.0,102.0,17.0\n",
      "2004.0,103.0,21.5\n",
      "2004.0,104.0,22.5\n",
      "2004.0,105.0,15.5\n",
      "2004.0,106.0,9.5\n",
      "2004.0,107.0,10.0\n",
      "2004.0,108.0,11.0\n",
      "2004.0,109.0,10.0\n",
      "2004.0,110.0,9.5\n",
      "2004.0,111.0,7.5\n",
      "2004.0,112.0,9.0\n",
      "2004.0,113.0,12.0\n",
      "2004.0,114.0,16.5\n",
      "2004.0,115.0,15.5\n",
      "2004.0,116.0,16.5\n",
      "2004.0,117.0,21.5\n",
      "2004.0,118.0,22.5\n",
      "2004.0,119.0,14.0\n",
      "2004.0,120.0,11.5\n",
      "2004.0,121.0,18.5\n",
      "2004.0,122.0,22.0\n",
      "2004.0,123.0,24.0\n",
      "2004.0,124.0,23.5\n",
      "2004.0,125.0,23.0\n",
      "2004.0,126.0,23.0\n",
      "2004.0,127.0,22.5\n",
      "2004.0,128.0,23.0\n",
      "2004.0,129.0,20.5\n",
      "2004.0,130.0,17.5\n",
      "2004.0,131.0,15.5\n",
      "2004.0,132.0,9.5\n",
      "2004.0,133.0,7.0\n",
      "2004.0,134.0,11.0\n",
      "2004.0,135.0,13.5\n",
      "2004.0,136.0,16.0\n",
      "2004.0,137.0,13.5\n",
      "2004.0,138.0,14.0\n",
      "2004.0,139.0,13.0\n",
      "2004.0,140.0,12.5\n",
      "2004.0,141.0,16.0\n",
      "2004.0,142.0,15.5\n",
      "2004.0,143.0,12.5\n",
      "2004.0,144.0,9.5\n",
      "2004.0,145.0,12.0\n",
      "2004.0,146.0,14.5\n",
      "2004.0,147.0,14.5\n",
      "2004.0,148.0,16.5\n",
      "2004.0,149.0,13.5\n",
      "2004.0,150.0,10.0\n",
      "2004.0,151.0,11.5\n",
      "2004.0,152.0,15.5\n",
      "2004.0,153.0,19.5\n",
      "2004.0,154.0,23.0\n",
      "2004.0,155.0,27.0\n",
      "2004.0,156.0,27.5\n",
      "2004.0,157.0,27.0\n",
      "2004.0,158.0,21.0\n",
      "2004.0,159.0,19.0\n",
      "2004.0,160.0,17.5\n",
      "2004.0,161.0,18.5\n",
      "2004.0,162.0,15.0\n",
      "2004.0,163.0,15.0\n",
      "2004.0,164.0,17.5\n",
      "2004.0,165.0,19.0\n",
      "2004.0,166.0,19.5\n",
      "2004.0,167.0,17.5\n",
      "2004.0,168.0,19.5\n",
      "2004.0,169.0,21.5\n",
      "2004.0,170.0,21.0\n",
      "2004.0,171.0,19.5\n",
      "2004.0,172.0,20.5\n",
      "2004.0,173.0,24.0\n",
      "2004.0,174.0,27.5\n",
      "2004.0,175.0,28.5\n",
      "2004.0,176.0,30.0\n",
      "2004.0,177.0,28.0\n",
      "2004.0,178.0,26.5\n",
      "2004.0,179.0,26.5\n",
      "2004.0,180.0,25.5\n",
      "2004.0,181.0,25.0\n",
      "2004.0,182.0,25.5\n",
      "2004.0,183.0,26.0\n",
      "2004.0,184.0,25.0\n",
      "2004.0,185.0,24.5\n",
      "2004.0,186.0,23.5\n",
      "2004.0,187.0,24.5\n",
      "2004.0,188.0,26.0\n",
      "2004.0,189.0,24.0\n",
      "2004.0,190.0,23.0\n",
      "2004.0,191.0,26.0\n",
      "2004.0,192.0,27.5\n",
      "2004.0,193.0,26.5\n",
      "2004.0,194.0,29.5\n",
      "2004.0,195.0,32.5\n",
      "2004.0,196.0,34.0\n",
      "2004.0,197.0,33.5\n",
      "2004.0,198.0,34.0\n",
      "2004.0,199.0,34.0\n",
      "2004.0,200.0,27.5\n",
      "2004.0,201.0,28.0\n",
      "2004.0,202.0,26.5\n",
      "2004.0,203.0,27.0\n",
      "2004.0,204.0,28.5\n",
      "2004.0,205.0,29.0\n",
      "2004.0,206.0,30.5\n",
      "2004.0,207.0,30.5\n",
      "2004.0,208.0,29.5\n",
      "2004.0,209.0,28.0\n",
      "2004.0,210.0,28.0\n",
      "2004.0,211.0,28.5\n",
      "2004.0,212.0,29.5\n",
      "2004.0,213.0,31.0\n",
      "2004.0,214.0,32.5\n",
      "2004.0,215.0,29.0\n",
      "2004.0,216.0,27.5\n",
      "2004.0,217.0,29.0\n",
      "2004.0,218.0,29.5\n",
      "2004.0,219.0,28.5\n",
      "2004.0,220.0,25.0\n",
      "2004.0,221.0,26.5\n",
      "2004.0,222.0,29.5\n",
      "2004.0,223.0,31.0\n",
      "2004.0,224.0,31.5\n",
      "2004.0,225.0,32.0\n",
      "2004.0,226.0,33.0\n",
      "2004.0,227.0,34.0\n",
      "2004.0,228.0,33.0\n",
      "2004.0,229.0,31.0\n",
      "2004.0,230.0,26.5\n",
      "2004.0,231.0,24.0\n",
      "2004.0,232.0,26.0\n",
      "2004.0,233.0,26.5\n",
      "2004.0,234.0,27.5\n",
      "2004.0,235.0,23.5\n",
      "2004.0,236.0,18.5\n",
      "2004.0,237.0,15.5\n",
      "2004.0,238.0,15.0\n",
      "2004.0,239.0,15.0\n",
      "2004.0,240.0,17.5\n",
      "2004.0,241.0,21.5\n",
      "2004.0,242.0,25.0\n",
      "2004.0,243.0,27.0\n",
      "2004.0,244.0,30.0\n",
      "2004.0,245.0,29.5\n",
      "2004.0,246.0,21.5\n",
      "2004.0,247.0,17.0\n",
      "2004.0,248.0,19.0\n",
      "2004.0,249.0,19.5\n",
      "2004.0,250.0,22.0\n",
      "2004.0,251.0,24.0\n",
      "2004.0,252.0,25.5\n",
      "2004.0,253.0,24.5\n",
      "2004.0,254.0,24.5\n",
      "2004.0,255.0,25.5\n",
      "2004.0,256.0,19.0\n",
      "2004.0,257.0,15.5\n",
      "2004.0,258.0,12.5\n",
      "2004.0,259.0,13.5\n",
      "2004.0,260.0,16.5\n",
      "2004.0,261.0,19.0\n",
      "2004.0,262.0,14.0\n",
      "2004.0,263.0,8.0\n",
      "2004.0,264.0,8.5\n",
      "2004.0,265.0,9.0\n",
      "2004.0,266.0,14.0\n",
      "2004.0,267.0,18.0\n",
      "2004.0,268.0,20.5\n",
      "2004.0,269.0,23.5\n",
      "2004.0,270.0,24.0\n",
      "2004.0,271.0,24.5\n",
      "2004.0,272.0,24.0\n",
      "2004.0,273.0,23.5\n",
      "2004.0,274.0,21.5\n",
      "2004.0,275.0,20.5\n",
      "2004.0,276.0,21.0\n",
      "2004.0,277.0,21.5\n",
      "2004.0,278.0,22.0\n",
      "2004.0,279.0,21.5\n",
      "2004.0,280.0,20.0\n",
      "2004.0,281.0,18.5\n",
      "2004.0,282.0,21.0\n",
      "2004.0,283.0,15.5\n",
      "2004.0,284.0,12.5\n",
      "2004.0,285.0,15.5\n",
      "2004.0,286.0,16.5\n",
      "2004.0,287.0,18.0\n",
      "2004.0,288.0,19.0\n",
      "2004.0,289.0,18.0\n",
      "2004.0,290.0,16.5\n",
      "2004.0,291.0,15.0\n",
      "2004.0,292.0,10.0\n",
      "2004.0,293.0,10.5\n",
      "2004.0,294.0,11.5\n",
      "2004.0,295.0,10.0\n",
      "2004.0,296.0,7.5\n",
      "2004.0,297.0,5.5\n",
      "2004.0,298.0,4.5\n",
      "2004.0,299.0,6.0\n",
      "2004.0,300.0,8.5\n",
      "2004.0,301.0,11.0\n",
      "2004.0,302.0,9.0\n",
      "2004.0,303.0,6.5\n",
      "2004.0,304.0,6.0\n",
      "2004.0,305.0,4.0\n",
      "2004.0,306.0,3.5\n",
      "2004.0,307.0,6.5\n",
      "2004.0,308.0,6.0\n",
      "2004.0,309.0,6.0\n",
      "2004.0,310.0,7.5\n",
      "2004.0,311.0,10.0\n",
      "2004.0,312.0,11.0\n",
      "2004.0,313.0,11.5\n",
      "2004.0,314.0,11.5\n",
      "2004.0,315.0,9.5\n",
      "2004.0,316.0,11.0\n",
      "2004.0,317.0,11.0\n",
      "2004.0,318.0,9.0\n",
      "2004.0,319.0,7.0\n",
      "2004.0,320.0,7.5\n",
      "2004.0,321.0,6.5\n",
      "2004.0,322.0,7.5\n",
      "2004.0,323.0,7.0\n",
      "2004.0,324.0,2.5\n",
      "2004.0,325.0,0.5\n",
      "2004.0,326.0,1.5\n",
      "2004.0,327.0,0.5\n",
      "2004.0,328.0,1.5\n",
      "2004.0,329.0,2.5\n",
      "2004.0,330.0,2.5\n",
      "2004.0,331.0,1.5\n",
      "2004.0,332.0,0.0\n",
      "2004.0,333.0,-0.5\n",
      "2004.0,334.0,-0.5\n",
      "2004.0,335.0,-4.0\n",
      "2004.0,336.0,-2.5\n",
      "2004.0,337.0,0.5\n",
      "2004.0,338.0,1.5\n",
      "2004.0,339.0,0.5\n",
      "2004.0,340.0,1.0\n",
      "2004.0,341.0,-0.5\n",
      "2004.0,342.0,1.0\n",
      "2004.0,343.0,2.5\n",
      "2004.0,344.0,4.0\n",
      "2004.0,345.0,5.5\n",
      "2004.0,346.0,6.0\n",
      "2004.0,347.0,4.5\n",
      "2004.0,348.0,5.5\n",
      "2004.0,349.0,5.5\n",
      "2004.0,350.0,4.5\n",
      "2004.0,351.0,5.0\n",
      "2004.0,352.0,3.5\n",
      "2004.0,353.0,5.5\n",
      "2004.0,354.0,5.0\n",
      "2004.0,355.0,2.0\n",
      "2004.0,356.0,-0.5\n",
      "2004.0,357.0,-2.5\n",
      "2004.0,358.0,-3.5\n",
      "2004.0,359.0,-2.0\n",
      "2004.0,360.0,0.0\n",
      "2004.0,361.0,2.5\n",
      "2004.0,362.0,4.0\n",
      "2004.0,363.0,3.0\n",
      "2004.0,364.0,3.5\n",
      "2004.0,365.0,2.5\n",
      "2005.0,1.0,-0.5\n",
      "2005.0,2.0,-0.5\n",
      "2005.0,3.0,-1.5\n",
      "2005.0,4.0,-3.0\n",
      "2005.0,5.0,-4.5\n",
      "2005.0,6.0,-7.0\n",
      "2005.0,7.0,-4.5\n",
      "2005.0,8.0,-1.5\n",
      "2005.0,9.0,-0.5\n",
      "2005.0,10.0,0.0\n",
      "2005.0,11.0,0.0\n",
      "2005.0,12.0,-1.5\n",
      "2005.0,13.0,-1.0\n",
      "2005.0,14.0,0.0\n",
      "2005.0,15.0,-0.5\n",
      "2005.0,16.0,1.0\n",
      "2005.0,17.0,1.5\n",
      "2005.0,18.0,4.0\n",
      "2005.0,19.0,8.0\n",
      "2005.0,20.0,6.5\n",
      "2005.0,21.0,8.0\n",
      "2005.0,22.0,7.0\n",
      "2005.0,23.0,8.0\n",
      "2005.0,24.0,8.5\n",
      "2005.0,25.0,7.5\n",
      "2005.0,26.0,6.5\n",
      "2005.0,27.0,6.5\n",
      "2005.0,28.0,6.0\n",
      "2005.0,29.0,5.5\n",
      "2005.0,30.0,4.5\n",
      "2005.0,31.0,5.0\n",
      "2005.0,32.0,5.0\n",
      "2005.0,33.0,7.5\n",
      "2005.0,34.0,8.5\n",
      "2005.0,35.0,7.5\n",
      "2005.0,36.0,3.5\n",
      "2005.0,37.0,1.0\n",
      "2005.0,38.0,2.5\n",
      "2005.0,39.0,3.5\n",
      "2005.0,40.0,4.5\n",
      "2005.0,41.0,6.0\n",
      "2005.0,42.0,7.0\n",
      "2005.0,43.0,5.5\n",
      "2005.0,44.0,3.5\n",
      "2005.0,45.0,1.0\n",
      "2005.0,46.0,0.0\n",
      "2005.0,47.0,1.0\n",
      "2005.0,48.0,3.0\n",
      "2005.0,49.0,4.5\n",
      "2005.0,50.0,5.0\n",
      "2005.0,51.0,5.5\n",
      "2005.0,52.0,7.5\n",
      "2005.0,53.0,10.0\n",
      "2005.0,54.0,9.5\n",
      "2005.0,55.0,11.0\n",
      "2005.0,56.0,11.5\n",
      "2005.0,57.0,11.5\n",
      "2005.0,58.0,11.5\n",
      "2005.0,59.0,10.0\n",
      "2005.0,60.0,8.0\n",
      "2005.0,61.0,10.5\n",
      "2005.0,62.0,10.5\n",
      "2005.0,63.0,11.5\n",
      "2005.0,64.0,12.5\n",
      "2005.0,65.0,12.5\n",
      "2005.0,66.0,14.0\n",
      "2005.0,67.0,15.5\n",
      "2005.0,68.0,16.5\n",
      "2005.0,69.0,16.5\n",
      "2005.0,70.0,17.0\n",
      "2005.0,71.0,12.0\n",
      "2005.0,72.0,7.0\n",
      "2005.0,73.0,6.5\n",
      "2005.0,74.0,8.0\n",
      "2005.0,75.0,7.5\n",
      "2005.0,76.0,5.0\n",
      "2005.0,77.0,4.5\n",
      "2005.0,78.0,7.5\n",
      "2005.0,79.0,9.5\n",
      "2005.0,80.0,7.5\n",
      "2005.0,81.0,6.5\n",
      "2005.0,82.0,4.5\n",
      "2005.0,83.0,3.0\n",
      "2005.0,84.0,3.5\n",
      "2005.0,85.0,4.0\n",
      "2005.0,86.0,8.0\n",
      "2005.0,87.0,7.5\n",
      "2005.0,88.0,6.0\n",
      "2005.0,89.0,3.5\n",
      "2005.0,90.0,7.5\n",
      "2005.0,91.0,10.5\n",
      "2005.0,92.0,12.0\n",
      "2005.0,93.0,11.0\n",
      "2005.0,94.0,7.5\n",
      "2005.0,95.0,8.0\n",
      "2005.0,96.0,16.5\n",
      "2005.0,97.0,19.0\n",
      "2005.0,98.0,13.5\n",
      "2005.0,99.0,7.5\n",
      "2005.0,100.0,9.5\n",
      "2005.0,101.0,12.0\n",
      "2005.0,102.0,12.5\n",
      "2005.0,103.0,10.0\n",
      "2005.0,104.0,7.0\n",
      "2005.0,105.0,9.5\n",
      "2005.0,106.0,15.5\n",
      "2005.0,107.0,13.5\n",
      "2005.0,108.0,7.0\n",
      "2005.0,109.0,6.5\n",
      "2005.0,110.0,9.0\n",
      "2005.0,111.0,12.5\n",
      "2005.0,112.0,16.5\n",
      "2005.0,113.0,17.0\n",
      "2005.0,114.0,16.5\n",
      "2005.0,115.0,19.0\n",
      "2005.0,116.0,22.0\n",
      "2005.0,117.0,15.0\n",
      "2005.0,118.0,10.5\n",
      "2005.0,119.0,11.0\n",
      "2005.0,120.0,13.0\n",
      "2005.0,121.0,16.5\n",
      "2005.0,122.0,17.0\n",
      "2005.0,123.0,16.0\n",
      "2005.0,124.0,15.0\n",
      "2005.0,125.0,15.5\n",
      "2005.0,126.0,16.0\n",
      "2005.0,127.0,13.0\n",
      "2005.0,128.0,16.0\n",
      "2005.0,129.0,16.5\n",
      "2005.0,130.0,11.5\n",
      "2005.0,131.0,10.0\n",
      "2005.0,132.0,14.0\n",
      "2005.0,133.0,19.0\n",
      "2005.0,134.0,20.0\n",
      "2005.0,135.0,17.0\n",
      "2005.0,136.0,13.0\n",
      "2005.0,137.0,12.0\n",
      "2005.0,138.0,14.5\n",
      "2005.0,139.0,15.5\n",
      "2005.0,140.0,16.5\n",
      "2005.0,141.0,16.0\n",
      "2005.0,142.0,17.5\n",
      "2005.0,143.0,16.5\n",
      "2005.0,144.0,15.0\n",
      "2005.0,145.0,17.5\n",
      "2005.0,146.0,21.5\n",
      "2005.0,147.0,25.0\n",
      "2005.0,148.0,26.5\n",
      "2005.0,149.0,25.5\n",
      "2005.0,150.0,23.5\n",
      "2005.0,151.0,22.0\n",
      "2005.0,152.0,15.0\n",
      "2005.0,153.0,11.0\n",
      "2005.0,154.0,11.0\n",
      "2005.0,155.0,17.5\n",
      "2005.0,156.0,19.0\n",
      "2005.0,157.0,15.0\n",
      "2005.0,158.0,12.0\n",
      "2005.0,159.0,11.5\n",
      "2005.0,160.0,14.0\n",
      "2005.0,161.0,16.5\n",
      "2005.0,162.0,17.5\n",
      "2005.0,163.0,14.5\n",
      "2005.0,164.0,15.5\n",
      "2005.0,165.0,22.5\n",
      "2005.0,166.0,24.0\n",
      "2005.0,167.0,23.0\n",
      "2005.0,168.0,19.0\n",
      "2005.0,169.0,16.5\n",
      "2005.0,170.0,20.0\n",
      "2005.0,171.0,26.5\n",
      "2005.0,172.0,31.0\n",
      "2005.0,173.0,30.0\n",
      "2005.0,174.0,27.0\n",
      "2005.0,175.0,26.5\n",
      "2005.0,176.0,25.5\n",
      "2005.0,177.0,21.0\n",
      "2005.0,178.0,18.0\n",
      "2005.0,179.0,16.5\n",
      "2005.0,180.0,18.0\n",
      "2005.0,181.0,23.0\n",
      "2005.0,182.0,27.0\n",
      "2005.0,183.0,25.0\n",
      "2005.0,184.0,24.0\n",
      "2005.0,185.0,26.0\n",
      "2005.0,186.0,29.0\n",
      "2005.0,187.0,30.0\n",
      "2005.0,188.0,30.5\n",
      "2005.0,189.0,31.0\n",
      "2005.0,190.0,23.5\n",
      "2005.0,191.0,18.5\n",
      "2005.0,192.0,24.5\n",
      "2005.0,193.0,31.5\n",
      "2005.0,194.0,32.5\n",
      "2005.0,195.0,30.5\n",
      "2005.0,196.0,32.0\n",
      "2005.0,197.0,30.5\n",
      "2005.0,198.0,27.5\n",
      "2005.0,199.0,30.0\n",
      "2005.0,200.0,32.0\n",
      "2005.0,201.0,32.5\n",
      "2005.0,202.0,33.5\n",
      "2005.0,203.0,33.0\n",
      "2005.0,204.0,30.5\n",
      "2005.0,205.0,30.5\n",
      "2005.0,206.0,27.0\n",
      "2005.0,207.0,26.5\n",
      "2005.0,208.0,30.0\n",
      "2005.0,209.0,30.5\n",
      "2005.0,210.0,29.5\n",
      "2005.0,211.0,31.0\n",
      "2005.0,212.0,31.5\n",
      "2005.0,213.0,31.5\n",
      "2005.0,214.0,27.5\n",
      "2005.0,215.0,27.0\n",
      "2005.0,216.0,30.0\n",
      "2005.0,217.0,33.0\n",
      "2005.0,218.0,34.0\n",
      "2005.0,219.0,33.5\n",
      "2005.0,220.0,32.0\n",
      "2005.0,221.0,31.0\n",
      "2005.0,222.0,30.5\n",
      "2005.0,223.0,28.0\n",
      "2005.0,224.0,27.0\n",
      "2005.0,225.0,25.0\n",
      "2005.0,226.0,26.0\n",
      "2005.0,227.0,29.0\n",
      "2005.0,228.0,29.5\n",
      "2005.0,229.0,26.5\n",
      "2005.0,230.0,24.5\n",
      "2005.0,231.0,26.0\n",
      "2005.0,232.0,30.0\n",
      "2005.0,233.0,32.5\n",
      "2005.0,234.0,30.5\n",
      "2005.0,235.0,27.5\n",
      "2005.0,236.0,23.5\n",
      "2005.0,237.0,24.0\n",
      "2005.0,238.0,28.0\n",
      "2005.0,239.0,29.5\n",
      "2005.0,240.0,31.0\n",
      "2005.0,241.0,29.0\n",
      "2005.0,242.0,21.5\n",
      "2005.0,243.0,23.0\n",
      "2005.0,244.0,25.5\n",
      "2005.0,245.0,28.5\n",
      "2005.0,246.0,29.0\n",
      "2005.0,247.0,27.0\n",
      "2005.0,248.0,24.5\n",
      "2005.0,249.0,25.5\n",
      "2005.0,250.0,28.0\n",
      "2005.0,251.0,29.5\n",
      "2005.0,252.0,26.0\n",
      "2005.0,253.0,16.5\n",
      "2005.0,254.0,14.0\n",
      "2005.0,255.0,15.5\n",
      "2005.0,256.0,16.0\n",
      "2005.0,257.0,19.0\n",
      "2005.0,258.0,22.5\n",
      "2005.0,259.0,21.5\n",
      "2005.0,260.0,16.0\n",
      "2005.0,261.0,15.5\n",
      "2005.0,262.0,21.5\n",
      "2005.0,263.0,24.5\n",
      "2005.0,264.0,24.0\n",
      "2005.0,265.0,22.0\n",
      "2005.0,266.0,16.5\n",
      "2005.0,267.0,9.5\n",
      "2005.0,268.0,14.0\n",
      "2005.0,269.0,19.0\n",
      "2005.0,270.0,20.5\n",
      "2005.0,271.0,22.0\n",
      "2005.0,272.0,22.0\n",
      "2005.0,273.0,21.0\n",
      "2005.0,274.0,17.0\n",
      "2005.0,275.0,12.0\n",
      "2005.0,276.0,9.0\n",
      "2005.0,277.0,7.0\n",
      "2005.0,278.0,9.5\n",
      "2005.0,279.0,14.5\n",
      "2005.0,280.0,15.5\n",
      "2005.0,281.0,12.5\n",
      "2005.0,282.0,11.0\n",
      "2005.0,283.0,14.0\n",
      "2005.0,284.0,9.5\n",
      "2005.0,285.0,8.0\n",
      "2005.0,286.0,14.0\n",
      "2005.0,287.0,19.0\n",
      "2005.0,288.0,19.5\n",
      "2005.0,289.0,17.0\n",
      "2005.0,290.0,19.0\n",
      "2005.0,291.0,20.5\n",
      "2005.0,292.0,17.0\n",
      "2005.0,293.0,13.5\n",
      "2005.0,294.0,14.5\n",
      "2005.0,295.0,16.5\n",
      "2005.0,296.0,18.0\n",
      "2005.0,297.0,19.0\n",
      "2005.0,298.0,19.0\n",
      "2005.0,299.0,16.0\n",
      "2005.0,300.0,12.0\n",
      "2005.0,301.0,11.0\n",
      "2005.0,302.0,9.0\n",
      "2005.0,303.0,7.5\n",
      "2005.0,304.0,7.5\n",
      "2005.0,305.0,10.5\n",
      "2005.0,306.0,8.0\n",
      "2005.0,307.0,5.0\n",
      "2005.0,308.0,3.5\n",
      "2005.0,309.0,3.5\n",
      "2005.0,310.0,3.5\n",
      "2005.0,311.0,3.0\n",
      "2005.0,312.0,4.0\n",
      "2005.0,313.0,4.0\n",
      "2005.0,314.0,6.0\n",
      "2005.0,315.0,4.5\n",
      "2005.0,316.0,2.5\n",
      "2005.0,317.0,2.5\n",
      "2005.0,318.0,2.0\n",
      "2005.0,319.0,0.0\n",
      "2005.0,320.0,1.0\n",
      "2005.0,321.0,4.5\n",
      "2005.0,322.0,3.5\n",
      "2005.0,323.0,6.0\n",
      "2005.0,324.0,6.0\n",
      "2005.0,325.0,6.5\n",
      "2005.0,326.0,6.5\n",
      "2005.0,327.0,5.5\n",
      "2005.0,328.0,4.5\n",
      "2005.0,329.0,2.0\n",
      "2005.0,330.0,0.5\n",
      "2005.0,331.0,-0.5\n",
      "2005.0,332.0,-1.5\n",
      "2005.0,333.0,-3.0\n",
      "2005.0,334.0,-2.5\n",
      "2005.0,335.0,-2.0\n",
      "2005.0,336.0,-1.0\n",
      "2005.0,337.0,-1.5\n",
      "2005.0,338.0,-3.0\n",
      "2005.0,339.0,-3.5\n",
      "2005.0,340.0,-3.5\n",
      "2005.0,341.0,-6.0\n",
      "2005.0,342.0,-9.0\n",
      "2005.0,343.0,-5.5\n",
      "2005.0,344.0,-2.5\n",
      "2005.0,345.0,-0.5\n",
      "2005.0,346.0,-1.5\n",
      "2005.0,347.0,-3.0\n",
      "2005.0,348.0,-1.5\n",
      "2005.0,349.0,-2.0\n",
      "2005.0,350.0,-3.0\n",
      "2005.0,351.0,-6.0\n",
      "2005.0,352.0,-6.5\n",
      "2005.0,353.0,-4.5\n",
      "2005.0,354.0,0.0\n",
      "2005.0,355.0,3.5\n",
      "2005.0,356.0,4.0\n",
      "2005.0,357.0,5.0\n",
      "2005.0,358.0,7.0\n",
      "2005.0,359.0,5.5\n",
      "2005.0,360.0,4.0\n",
      "2005.0,361.0,2.0\n",
      "2005.0,362.0,3.0\n",
      "2005.0,363.0,2.5\n",
      "2005.0,364.0,2.0\n",
      "2005.0,365.0,3.0\n",
      "2006.0,1.0,2.5\n",
      "2006.0,2.0,3.5\n",
      "2006.0,3.0,3.0\n",
      "2006.0,4.0,3.0\n",
      "2006.0,5.0,5.0\n",
      "2006.0,6.0,3.5\n",
      "2006.0,7.0,3.0\n",
      "2006.0,8.0,1.0\n",
      "2006.0,9.0,0.5\n",
      "2006.0,10.0,1.5\n",
      "2006.0,11.0,2.5\n",
      "2006.0,12.0,2.0\n",
      "2006.0,13.0,3.5\n",
      "2006.0,14.0,6.0\n",
      "2006.0,15.0,2.0\n",
      "2006.0,16.0,0.0\n",
      "2006.0,17.0,0.5\n",
      "2006.0,18.0,1.0\n",
      "2006.0,19.0,0.5\n",
      "2006.0,20.0,-0.5\n",
      "2006.0,21.0,0.5\n",
      "2006.0,22.0,-0.5\n",
      "2006.0,23.0,1.5\n",
      "2006.0,24.0,4.0\n",
      "2006.0,25.0,4.5\n",
      "2006.0,26.0,2.0\n",
      "2006.0,27.0,1.0\n",
      "2006.0,28.0,0.5\n",
      "2006.0,29.0,0.5\n",
      "2006.0,30.0,2.5\n",
      "2006.0,31.0,2.0\n",
      "2006.0,32.0,2.5\n",
      "2006.0,33.0,3.0\n",
      "2006.0,34.0,4.5\n",
      "2006.0,35.0,3.5\n",
      "2006.0,36.0,1.0\n",
      "2006.0,37.0,1.5\n",
      "2006.0,38.0,3.5\n",
      "2006.0,39.0,6.0\n",
      "2006.0,40.0,4.5\n",
      "2006.0,41.0,3.5\n",
      "2006.0,42.0,3.5\n",
      "2006.0,43.0,1.5\n",
      "2006.0,44.0,3.5\n",
      "2006.0,45.0,1.5\n",
      "2006.0,46.0,-1.5\n",
      "2006.0,47.0,-3.5\n",
      "2006.0,48.0,-7.0\n",
      "2006.0,49.0,-7.0\n",
      "2006.0,50.0,-4.0\n",
      "2006.0,51.0,-2.5\n",
      "2006.0,52.0,-1.5\n",
      "2006.0,53.0,0.5\n",
      "2006.0,54.0,4.0\n",
      "2006.0,55.0,5.5\n",
      "2006.0,56.0,6.0\n",
      "2006.0,57.0,8.5\n",
      "2006.0,58.0,9.5\n",
      "2006.0,59.0,8.0\n",
      "2006.0,60.0,7.0\n",
      "2006.0,61.0,7.0\n",
      "2006.0,62.0,7.5\n",
      "2006.0,63.0,5.0\n",
      "2006.0,64.0,5.0\n",
      "2006.0,65.0,5.0\n",
      "2006.0,66.0,5.0\n",
      "2006.0,67.0,5.0\n",
      "2006.0,68.0,2.0\n",
      "2006.0,69.0,0.5\n",
      "2006.0,70.0,1.5\n",
      "2006.0,71.0,3.0\n",
      "2006.0,72.0,3.5\n",
      "2006.0,73.0,3.5\n",
      "2006.0,74.0,4.5\n",
      "2006.0,75.0,5.5\n",
      "2006.0,76.0,8.0\n",
      "2006.0,77.0,6.0\n",
      "2006.0,78.0,4.0\n",
      "2006.0,79.0,5.5\n",
      "2006.0,80.0,7.0\n",
      "2006.0,81.0,7.5\n",
      "2006.0,82.0,11.0\n",
      "2006.0,83.0,12.0\n",
      "2006.0,84.0,11.5\n",
      "2006.0,85.0,7.5\n",
      "2006.0,86.0,9.5\n",
      "2006.0,87.0,10.0\n",
      "2006.0,88.0,7.5\n",
      "2006.0,89.0,9.0\n",
      "2006.0,90.0,10.0\n",
      "2006.0,91.0,8.5\n",
      "2006.0,92.0,9.5\n",
      "2006.0,93.0,11.5\n",
      "2006.0,94.0,11.5\n",
      "2006.0,95.0,10.5\n",
      "2006.0,96.0,7.0\n",
      "2006.0,97.0,9.5\n",
      "2006.0,98.0,11.5\n",
      "2006.0,99.0,11.5\n",
      "2006.0,100.0,10.5\n",
      "2006.0,101.0,11.5\n",
      "2006.0,102.0,13.5\n",
      "2006.0,103.0,14.0\n",
      "2006.0,104.0,13.0\n",
      "2006.0,105.0,11.5\n",
      "2006.0,106.0,7.5\n",
      "2006.0,107.0,5.0\n",
      "2006.0,108.0,7.0\n",
      "2006.0,109.0,11.0\n",
      "2006.0,110.0,17.0\n",
      "2006.0,111.0,20.0\n",
      "2006.0,112.0,15.5\n",
      "2006.0,113.0,11.5\n",
      "2006.0,114.0,12.0\n",
      "2006.0,115.0,11.5\n",
      "2006.0,116.0,15.0\n",
      "2006.0,117.0,17.0\n",
      "2006.0,118.0,18.5\n",
      "2006.0,119.0,22.0\n",
      "2006.0,120.0,18.0\n",
      "2006.0,121.0,16.0\n",
      "2006.0,122.0,14.5\n",
      "2006.0,123.0,14.0\n",
      "2006.0,124.0,16.5\n",
      "2006.0,125.0,19.0\n",
      "2006.0,126.0,19.5\n",
      "2006.0,127.0,14.0\n",
      "2006.0,128.0,10.5\n",
      "2006.0,129.0,11.5\n",
      "2006.0,130.0,16.0\n",
      "2006.0,131.0,20.0\n",
      "2006.0,132.0,20.5\n",
      "2006.0,133.0,21.5\n",
      "2006.0,134.0,25.5\n",
      "2006.0,135.0,28.0\n",
      "2006.0,136.0,30.0\n",
      "2006.0,137.0,29.5\n",
      "2006.0,138.0,29.5\n",
      "2006.0,139.0,26.5\n",
      "2006.0,140.0,24.5\n",
      "2006.0,141.0,23.0\n",
      "2006.0,142.0,23.5\n",
      "2006.0,143.0,21.5\n",
      "2006.0,144.0,21.5\n",
      "2006.0,145.0,22.5\n",
      "2006.0,146.0,17.5\n",
      "2006.0,147.0,11.5\n",
      "2006.0,148.0,8.5\n",
      "2006.0,149.0,10.5\n",
      "2006.0,150.0,16.0\n",
      "2006.0,151.0,21.5\n",
      "2006.0,152.0,25.0\n",
      "2006.0,153.0,26.5\n",
      "2006.0,154.0,23.0\n",
      "2006.0,155.0,20.0\n",
      "2006.0,156.0,22.5\n",
      "2006.0,157.0,26.5\n",
      "2006.0,158.0,25.5\n",
      "2006.0,159.0,22.5\n",
      "2006.0,160.0,19.5\n",
      "2006.0,161.0,18.5\n",
      "2006.0,162.0,19.0\n",
      "2006.0,163.0,23.5\n",
      "2006.0,164.0,27.0\n",
      "2006.0,165.0,21.0\n",
      "2006.0,166.0,16.5\n",
      "2006.0,167.0,19.0\n",
      "2006.0,168.0,21.0\n",
      "2006.0,169.0,23.0\n",
      "2006.0,170.0,23.0\n",
      "2006.0,171.0,22.0\n",
      "2006.0,172.0,22.0\n",
      "2006.0,173.0,24.5\n",
      "2006.0,174.0,26.0\n",
      "2006.0,175.0,28.0\n",
      "2006.0,176.0,30.5\n",
      "2006.0,177.0,31.5\n",
      "2006.0,178.0,31.5\n",
      "2006.0,179.0,31.0\n",
      "2006.0,180.0,27.0\n",
      "2006.0,181.0,25.5\n",
      "2006.0,182.0,27.5\n",
      "2006.0,183.0,29.5\n",
      "2006.0,184.0,30.5\n",
      "2006.0,185.0,31.5\n",
      "2006.0,186.0,27.0\n",
      "2006.0,187.0,24.5\n",
      "2006.0,188.0,27.0\n",
      "2006.0,189.0,29.5\n",
      "2006.0,190.0,31.0\n",
      "2006.0,191.0,30.5\n",
      "2006.0,192.0,31.0\n",
      "2006.0,193.0,29.5\n",
      "2006.0,194.0,27.0\n",
      "2006.0,195.0,31.0\n",
      "2006.0,196.0,32.5\n",
      "2006.0,197.0,32.5\n",
      "2006.0,198.0,33.0\n",
      "2006.0,199.0,32.0\n",
      "2006.0,200.0,32.0\n",
      "2006.0,201.0,33.0\n",
      "2006.0,202.0,34.5\n",
      "2006.0,203.0,36.5\n",
      "2006.0,204.0,36.0\n",
      "2006.0,205.0,34.5\n",
      "2006.0,206.0,33.0\n",
      "2006.0,207.0,33.0\n",
      "2006.0,208.0,34.0\n",
      "2006.0,209.0,34.5\n",
      "2006.0,210.0,33.5\n",
      "2006.0,211.0,28.5\n",
      "2006.0,212.0,25.0\n",
      "2006.0,213.0,24.5\n",
      "2006.0,214.0,26.0\n",
      "2006.0,215.0,28.0\n",
      "2006.0,216.0,29.5\n",
      "2006.0,217.0,30.0\n",
      "2006.0,218.0,30.5\n",
      "2006.0,219.0,31.0\n",
      "2006.0,220.0,30.0\n",
      "2006.0,221.0,28.0\n",
      "2006.0,222.0,28.5\n",
      "2006.0,223.0,27.0\n",
      "2006.0,224.0,23.5\n",
      "2006.0,225.0,25.0\n",
      "2006.0,226.0,28.5\n",
      "2006.0,227.0,28.5\n",
      "2006.0,228.0,24.5\n",
      "2006.0,229.0,20.0\n",
      "2006.0,230.0,23.5\n",
      "2006.0,231.0,28.0\n",
      "2006.0,232.0,30.5\n",
      "2006.0,233.0,31.5\n",
      "2006.0,234.0,30.5\n",
      "2006.0,235.0,29.5\n",
      "2006.0,236.0,25.5\n",
      "2006.0,237.0,18.5\n",
      "2006.0,238.0,20.5\n",
      "2006.0,239.0,27.0\n",
      "2006.0,240.0,29.5\n",
      "2006.0,241.0,30.0\n",
      "2006.0,242.0,23.0\n",
      "2006.0,243.0,18.5\n",
      "2006.0,244.0,23.0\n",
      "2006.0,245.0,27.5\n",
      "2006.0,246.0,30.0\n",
      "2006.0,247.0,30.5\n",
      "2006.0,248.0,29.0\n",
      "2006.0,249.0,27.5\n",
      "2006.0,250.0,26.0\n",
      "2006.0,251.0,25.5\n",
      "2006.0,252.0,23.5\n",
      "2006.0,253.0,22.5\n",
      "2006.0,254.0,27.0\n",
      "2006.0,255.0,28.5\n",
      "2006.0,256.0,26.5\n",
      "2006.0,257.0,20.5\n",
      "2006.0,258.0,15.0\n",
      "2006.0,259.0,11.0\n",
      "2006.0,260.0,10.0\n",
      "2006.0,261.0,16.5\n",
      "2006.0,262.0,17.5\n",
      "2006.0,263.0,13.0\n",
      "2006.0,264.0,9.5\n",
      "2006.0,265.0,10.5\n",
      "2006.0,266.0,13.5\n",
      "2006.0,267.0,18.0\n",
      "2006.0,268.0,21.0\n",
      "2006.0,269.0,22.0\n",
      "2006.0,270.0,22.5\n",
      "2006.0,271.0,23.5\n",
      "2006.0,272.0,24.5\n",
      "2006.0,273.0,24.0\n",
      "2006.0,274.0,23.0\n",
      "2006.0,275.0,19.5\n",
      "2006.0,276.0,18.0\n",
      "2006.0,277.0,19.0\n",
      "2006.0,278.0,18.5\n",
      "2006.0,279.0,16.0\n",
      "2006.0,280.0,14.0\n",
      "2006.0,281.0,13.0\n",
      "2006.0,282.0,9.5\n",
      "2006.0,283.0,11.0\n",
      "2006.0,284.0,14.0\n",
      "2006.0,285.0,17.5\n",
      "2006.0,286.0,18.5\n",
      "2006.0,287.0,19.0\n",
      "2006.0,288.0,13.5\n",
      "2006.0,289.0,9.5\n",
      "2006.0,290.0,8.0\n",
      "2006.0,291.0,8.0\n",
      "2006.0,292.0,8.5\n",
      "2006.0,293.0,7.5\n",
      "2006.0,294.0,9.0\n",
      "2006.0,295.0,12.0\n",
      "2006.0,296.0,12.0\n",
      "2006.0,297.0,13.0\n",
      "2006.0,298.0,7.0\n",
      "2006.0,299.0,6.0\n",
      "2006.0,300.0,10.5\n",
      "2006.0,301.0,13.5\n",
      "2006.0,302.0,13.0\n",
      "2006.0,303.0,7.0\n",
      "2006.0,304.0,1.5\n",
      "2006.0,305.0,3.5\n",
      "2006.0,306.0,5.0\n",
      "2006.0,307.0,7.5\n",
      "2006.0,308.0,9.5\n",
      "2006.0,309.0,11.5\n",
      "2006.0,310.0,11.0\n",
      "2006.0,311.0,10.5\n",
      "2006.0,312.0,11.0\n",
      "2006.0,313.0,6.0\n",
      "2006.0,314.0,3.5\n",
      "2006.0,315.0,3.5\n",
      "2006.0,316.0,4.0\n",
      "2006.0,317.0,4.5\n",
      "2006.0,318.0,4.0\n",
      "2006.0,319.0,5.0\n",
      "2006.0,320.0,4.5\n",
      "2006.0,321.0,4.0\n",
      "2006.0,322.0,5.5\n",
      "2006.0,323.0,5.5\n",
      "2006.0,324.0,7.0\n",
      "2006.0,325.0,8.0\n",
      "2006.0,326.0,8.0\n",
      "2006.0,327.0,4.5\n",
      "2006.0,328.0,2.5\n",
      "2006.0,329.0,2.0\n",
      "2006.0,330.0,1.5\n",
      "2006.0,331.0,1.0\n",
      "2006.0,332.0,-3.5\n",
      "2006.0,333.0,-4.0\n",
      "2006.0,334.0,-3.0\n",
      "2006.0,335.0,-2.0\n",
      "2006.0,336.0,-1.5\n",
      "2006.0,337.0,-1.0\n",
      "2006.0,338.0,-0.5\n",
      "2006.0,339.0,0.5\n",
      "2006.0,340.0,2.5\n",
      "2006.0,341.0,5.5\n",
      "2006.0,342.0,5.0\n",
      "2006.0,343.0,4.0\n",
      "2006.0,344.0,2.5\n",
      "2006.0,345.0,2.5\n",
      "2006.0,346.0,3.5\n",
      "2006.0,347.0,4.0\n",
      "2006.0,348.0,4.0\n",
      "2006.0,349.0,4.0\n",
      "2006.0,350.0,0.5\n",
      "2006.0,351.0,-2.0\n",
      "2006.0,352.0,-4.5\n",
      "2006.0,353.0,-4.0\n",
      "2006.0,354.0,-1.5\n",
      "2006.0,355.0,-2.0\n",
      "2006.0,356.0,-1.5\n",
      "2006.0,357.0,-0.5\n",
      "2006.0,358.0,0.0\n",
      "2006.0,359.0,1.0\n",
      "2006.0,360.0,2.5\n",
      "2006.0,361.0,3.0\n",
      "2006.0,362.0,0.5\n",
      "2006.0,363.0,0.0\n",
      "2006.0,364.0,-1.0\n",
      "2006.0,365.0,-2.0\n",
      "2007.0,1.0,-1.0\n",
      "2007.0,2.0,2.0\n",
      "2007.0,3.0,5.0\n",
      "2007.0,4.0,2.0\n",
      "2007.0,5.0,-1.5\n",
      "2007.0,6.0,-3.0\n",
      "2007.0,7.0,0.0\n",
      "2007.0,8.0,1.0\n",
      "2007.0,9.0,5.0\n",
      "2007.0,10.0,4.5\n",
      "2007.0,11.0,-4.5\n",
      "2007.0,12.0,-8.0\n",
      "2007.0,13.0,-9.5\n",
      "2007.0,14.0,-8.5\n",
      "2007.0,15.0,-5.0\n",
      "2007.0,16.0,-4.0\n",
      "2007.0,17.0,-5.5\n",
      "2007.0,18.0,-5.0\n",
      "2007.0,19.0,-2.5\n",
      "2007.0,20.0,-1.5\n",
      "2007.0,21.0,0.0\n",
      "2007.0,22.0,1.5\n",
      "2007.0,23.0,4.0\n",
      "2007.0,24.0,6.0\n",
      "2007.0,25.0,5.0\n",
      "2007.0,26.0,4.5\n",
      "2007.0,27.0,2.5\n",
      "2007.0,28.0,2.0\n",
      "2007.0,29.0,-0.5\n",
      "2007.0,30.0,0.5\n",
      "2007.0,31.0,1.0\n",
      "2007.0,32.0,-2.0\n",
      "2007.0,33.0,-0.5\n",
      "2007.0,34.0,2.0\n",
      "2007.0,35.0,5.5\n",
      "2007.0,36.0,10.0\n",
      "2007.0,37.0,10.5\n",
      "2007.0,38.0,7.5\n",
      "2007.0,39.0,6.0\n",
      "2007.0,40.0,6.5\n",
      "2007.0,41.0,6.0\n",
      "2007.0,42.0,5.0\n",
      "2007.0,43.0,5.0\n",
      "2007.0,44.0,3.0\n",
      "2007.0,45.0,3.0\n",
      "2007.0,46.0,3.5\n",
      "2007.0,47.0,5.0\n",
      "2007.0,48.0,9.0\n",
      "2007.0,49.0,6.5\n",
      "2007.0,50.0,2.0\n",
      "2007.0,51.0,3.5\n",
      "2007.0,52.0,3.0\n",
      "2007.0,53.0,5.0\n",
      "2007.0,54.0,2.5\n",
      "2007.0,55.0,1.0\n",
      "2007.0,56.0,2.0\n",
      "2007.0,57.0,2.0\n",
      "2007.0,58.0,2.0\n",
      "2007.0,59.0,1.0\n",
      "2007.0,60.0,-2.0\n",
      "2007.0,61.0,-1.0\n",
      "2007.0,62.0,2.0\n",
      "2007.0,63.0,6.5\n",
      "2007.0,64.0,10.0\n",
      "2007.0,65.0,12.0\n",
      "2007.0,66.0,13.0\n",
      "2007.0,67.0,10.0\n",
      "2007.0,68.0,9.0\n",
      "2007.0,69.0,10.5\n",
      "2007.0,70.0,14.0\n",
      "2007.0,71.0,15.5\n",
      "2007.0,72.0,13.0\n",
      "2007.0,73.0,10.5\n",
      "2007.0,74.0,10.5\n",
      "2007.0,75.0,13.5\n",
      "2007.0,76.0,18.0\n",
      "2007.0,77.0,17.5\n",
      "2007.0,78.0,16.0\n",
      "2007.0,79.0,13.0\n",
      "2007.0,80.0,7.0\n",
      "2007.0,81.0,9.0\n",
      "2007.0,82.0,12.5\n",
      "2007.0,83.0,16.5\n",
      "2007.0,84.0,15.5\n",
      "2007.0,85.0,13.5\n",
      "2007.0,86.0,7.5\n",
      "2007.0,87.0,4.0\n",
      "2007.0,88.0,7.5\n",
      "2007.0,89.0,13.0\n",
      "2007.0,90.0,13.0\n",
      "2007.0,91.0,11.0\n",
      "2007.0,92.0,6.5\n",
      "2007.0,93.0,8.0\n",
      "2007.0,94.0,12.0\n",
      "2007.0,95.0,15.5\n",
      "2007.0,96.0,18.5\n",
      "2007.0,97.0,19.5\n",
      "2007.0,98.0,18.0\n",
      "2007.0,99.0,13.0\n",
      "2007.0,100.0,5.0\n",
      "2007.0,101.0,6.5\n",
      "2007.0,102.0,8.5\n",
      "2007.0,103.0,12.5\n",
      "2007.0,104.0,16.5\n",
      "2007.0,105.0,12.0\n",
      "2007.0,106.0,11.5\n",
      "2007.0,107.0,13.0\n",
      "2007.0,108.0,5.5\n",
      "2007.0,109.0,4.0\n",
      "2007.0,110.0,7.5\n",
      "2007.0,111.0,10.5\n",
      "2007.0,112.0,11.0\n",
      "2007.0,113.0,12.0\n",
      "2007.0,114.0,15.0\n",
      "2007.0,115.0,16.0\n",
      "2007.0,116.0,13.5\n",
      "2007.0,117.0,17.5\n",
      "2007.0,118.0,22.5\n",
      "2007.0,119.0,23.0\n",
      "2007.0,120.0,21.5\n",
      "2007.0,121.0,22.5\n",
      "2007.0,122.0,20.5\n",
      "2007.0,123.0,10.5\n",
      "2007.0,124.0,7.0\n",
      "2007.0,125.0,8.5\n",
      "2007.0,126.0,13.5\n",
      "2007.0,127.0,17.5\n",
      "2007.0,128.0,22.5\n",
      "2007.0,129.0,24.0\n",
      "2007.0,130.0,23.0\n",
      "2007.0,131.0,24.0\n",
      "2007.0,132.0,25.0\n",
      "2007.0,133.0,21.5\n",
      "2007.0,134.0,18.5\n",
      "2007.0,135.0,22.5\n",
      "2007.0,136.0,25.5\n",
      "2007.0,137.0,26.5\n",
      "2007.0,138.0,26.5\n",
      "2007.0,139.0,24.5\n",
      "2007.0,140.0,18.0\n",
      "2007.0,141.0,10.0\n",
      "2007.0,142.0,7.5\n",
      "2007.0,143.0,13.0\n",
      "2007.0,144.0,15.0\n",
      "2007.0,145.0,17.5\n",
      "2007.0,146.0,21.0\n",
      "2007.0,147.0,20.0\n",
      "2007.0,148.0,14.0\n",
      "2007.0,149.0,15.0\n",
      "2007.0,150.0,21.0\n",
      "2007.0,151.0,24.5\n",
      "2007.0,152.0,27.0\n",
      "2007.0,153.0,28.0\n",
      "2007.0,154.0,28.5\n",
      "2007.0,155.0,28.0\n",
      "2007.0,156.0,23.0\n",
      "2007.0,157.0,12.5\n",
      "2007.0,158.0,11.0\n",
      "2007.0,159.0,18.0\n",
      "2007.0,160.0,21.0\n",
      "2007.0,161.0,18.5\n",
      "2007.0,162.0,17.0\n",
      "2007.0,163.0,21.0\n",
      "2007.0,164.0,23.0\n",
      "2007.0,165.0,22.5\n",
      "2007.0,166.0,24.0\n",
      "2007.0,167.0,25.5\n",
      "2007.0,168.0,20.5\n",
      "2007.0,169.0,19.5\n",
      "2007.0,170.0,25.0\n",
      "2007.0,171.0,28.5\n",
      "2007.0,172.0,29.5\n",
      "2007.0,173.0,30.0\n",
      "2007.0,174.0,28.0\n",
      "2007.0,175.0,24.5\n",
      "2007.0,176.0,20.5\n",
      "2007.0,177.0,23.0\n",
      "2007.0,178.0,30.0\n",
      "2007.0,179.0,32.5\n",
      "2007.0,180.0,30.0\n",
      "2007.0,181.0,28.5\n",
      "2007.0,182.0,30.0\n",
      "2007.0,183.0,30.0\n",
      "2007.0,184.0,29.5\n",
      "2007.0,185.0,32.5\n",
      "2007.0,186.0,35.5\n",
      "2007.0,187.0,37.0\n",
      "2007.0,188.0,34.0\n",
      "2007.0,189.0,30.5\n",
      "2007.0,190.0,30.5\n",
      "2007.0,191.0,31.5\n",
      "2007.0,192.0,33.0\n",
      "2007.0,193.0,34.0\n",
      "2007.0,194.0,34.0\n",
      "2007.0,195.0,35.0\n",
      "2007.0,196.0,35.0\n",
      "2007.0,197.0,34.5\n",
      "2007.0,198.0,32.5\n",
      "2007.0,199.0,32.5\n",
      "2007.0,200.0,32.0\n",
      "2007.0,201.0,31.0\n",
      "2007.0,202.0,32.0\n",
      "2007.0,203.0,33.5\n",
      "2007.0,204.0,34.0\n",
      "2007.0,205.0,28.5\n",
      "2007.0,206.0,30.0\n",
      "2007.0,207.0,29.5\n",
      "2007.0,208.0,30.5\n",
      "2007.0,209.0,33.5\n",
      "2007.0,210.0,34.0\n",
      "2007.0,211.0,31.5\n",
      "2007.0,212.0,31.0\n",
      "2007.0,213.0,32.0\n",
      "2007.0,214.0,32.5\n",
      "2007.0,215.0,31.0\n",
      "2007.0,216.0,30.0\n",
      "2007.0,217.0,27.5\n",
      "2007.0,218.0,25.5\n",
      "2007.0,219.0,26.0\n",
      "2007.0,220.0,26.5\n",
      "2007.0,221.0,29.0\n",
      "2007.0,222.0,27.0\n",
      "2007.0,223.0,27.5\n",
      "2007.0,224.0,29.0\n",
      "2007.0,225.0,29.0\n",
      "2007.0,226.0,28.5\n",
      "2007.0,227.0,28.5\n",
      "2007.0,228.0,30.0\n",
      "2007.0,229.0,28.0\n",
      "2007.0,230.0,28.5\n",
      "2007.0,231.0,26.0\n",
      "2007.0,232.0,20.0\n",
      "2007.0,233.0,19.0\n",
      "2007.0,234.0,22.5\n",
      "2007.0,235.0,23.5\n",
      "2007.0,236.0,25.0\n",
      "2007.0,237.0,28.5\n",
      "2007.0,238.0,28.5\n",
      "2007.0,239.0,25.5\n",
      "2007.0,240.0,25.5\n",
      "2007.0,241.0,28.5\n",
      "2007.0,242.0,30.5\n",
      "2007.0,243.0,28.5\n",
      "2007.0,244.0,29.0\n",
      "2007.0,245.0,30.5\n",
      "2007.0,246.0,31.5\n",
      "2007.0,247.0,27.0\n",
      "2007.0,248.0,23.0\n",
      "2007.0,249.0,21.5\n",
      "2007.0,250.0,24.0\n",
      "2007.0,251.0,23.0\n",
      "2007.0,252.0,21.0\n",
      "2007.0,253.0,23.0\n",
      "2007.0,254.0,25.0\n",
      "2007.0,255.0,25.0\n",
      "2007.0,256.0,25.0\n",
      "2007.0,257.0,23.5\n",
      "2007.0,258.0,23.5\n",
      "2007.0,259.0,23.0\n",
      "2007.0,260.0,18.5\n",
      "2007.0,261.0,13.5\n",
      "2007.0,262.0,10.0\n",
      "2007.0,263.0,14.0\n",
      "2007.0,264.0,20.5\n",
      "2007.0,265.0,19.0\n",
      "2007.0,266.0,10.0\n",
      "2007.0,267.0,8.5\n",
      "2007.0,268.0,14.0\n",
      "2007.0,269.0,18.0\n",
      "2007.0,270.0,21.0\n",
      "2007.0,271.0,20.5\n",
      "2007.0,272.0,13.0\n",
      "2007.0,273.0,11.0\n",
      "2007.0,274.0,11.0\n",
      "2007.0,275.0,10.0\n",
      "2007.0,276.0,9.5\n",
      "2007.0,277.0,9.0\n",
      "2007.0,278.0,6.0\n",
      "2007.0,279.0,6.0\n",
      "2007.0,280.0,9.5\n",
      "2007.0,281.0,15.0\n",
      "2007.0,282.0,21.0\n",
      "2007.0,283.0,17.5\n",
      "2007.0,284.0,14.5\n",
      "2007.0,285.0,14.0\n",
      "2007.0,286.0,14.0\n",
      "2007.0,287.0,17.5\n",
      "2007.0,288.0,18.0\n",
      "2007.0,289.0,15.0\n",
      "2007.0,290.0,9.5\n",
      "2007.0,291.0,6.0\n",
      "2007.0,292.0,6.5\n",
      "2007.0,293.0,5.5\n",
      "2007.0,294.0,5.0\n",
      "2007.0,295.0,8.0\n",
      "2007.0,296.0,15.0\n",
      "2007.0,297.0,18.0\n",
      "2007.0,298.0,10.0\n",
      "2007.0,299.0,8.5\n",
      "2007.0,300.0,11.0\n",
      "2007.0,301.0,14.5\n",
      "2007.0,302.0,13.5\n",
      "2007.0,303.0,10.0\n",
      "2007.0,304.0,11.5\n",
      "2007.0,305.0,9.5\n",
      "2007.0,306.0,11.0\n",
      "2007.0,307.0,10.0\n",
      "2007.0,308.0,12.0\n",
      "2007.0,309.0,13.5\n",
      "2007.0,310.0,13.0\n",
      "2007.0,311.0,12.5\n",
      "2007.0,312.0,12.0\n",
      "2007.0,313.0,13.0\n",
      "2007.0,314.0,9.5\n",
      "2007.0,315.0,5.5\n",
      "2007.0,316.0,6.0\n",
      "2007.0,317.0,6.0\n",
      "2007.0,318.0,3.5\n",
      "2007.0,319.0,5.5\n",
      "2007.0,320.0,7.0\n",
      "2007.0,321.0,6.0\n",
      "2007.0,322.0,7.0\n",
      "2007.0,323.0,7.0\n",
      "2007.0,324.0,1.0\n",
      "2007.0,325.0,1.0\n",
      "2007.0,326.0,0.0\n",
      "2007.0,327.0,0.0\n",
      "2007.0,328.0,0.5\n",
      "2007.0,329.0,2.0\n",
      "2007.0,330.0,2.0\n",
      "2007.0,331.0,1.0\n",
      "2007.0,332.0,0.0\n",
      "2007.0,333.0,-1.5\n",
      "2007.0,334.0,-0.5\n",
      "2007.0,335.0,-3.0\n",
      "2007.0,336.0,-1.0\n",
      "2007.0,337.0,2.5\n",
      "2007.0,338.0,5.0\n",
      "2007.0,339.0,4.5\n",
      "2007.0,340.0,3.5\n",
      "2007.0,341.0,2.0\n",
      "2007.0,342.0,0.0\n",
      "2007.0,343.0,-2.0\n",
      "2007.0,344.0,-2.5\n",
      "2007.0,345.0,-1.0\n",
      "2007.0,346.0,-2.5\n",
      "2007.0,347.0,-2.0\n",
      "2007.0,348.0,-1.5\n",
      "2007.0,349.0,-1.5\n",
      "2007.0,350.0,0.5\n",
      "2007.0,351.0,0.0\n",
      "2007.0,352.0,1.0\n",
      "2007.0,353.0,1.5\n",
      "2007.0,354.0,1.0\n",
      "2007.0,355.0,-2.0\n",
      "2007.0,356.0,-4.0\n",
      "2007.0,357.0,-2.0\n",
      "2007.0,358.0,0.0\n",
      "2007.0,359.0,-1.0\n",
      "2007.0,360.0,-4.0\n",
      "2007.0,361.0,-4.0\n",
      "2007.0,362.0,-3.5\n",
      "2007.0,363.0,-2.5\n",
      "2007.0,364.0,-3.5\n",
      "2007.0,365.0,-2.5\n",
      "2008.0,1.0,-2.0\n",
      "2008.0,2.0,-2.5\n",
      "2008.0,3.0,2.5\n",
      "2008.0,4.0,3.0\n",
      "2008.0,5.0,2.0\n",
      "2008.0,6.0,-0.5\n",
      "2008.0,7.0,-2.5\n",
      "2008.0,8.0,-3.5\n",
      "2008.0,9.0,-1.5\n",
      "2008.0,10.0,0.5\n",
      "2008.0,11.0,0.5\n",
      "2008.0,12.0,1.5\n",
      "2008.0,13.0,3.0\n",
      "2008.0,14.0,2.0\n",
      "2008.0,15.0,-1.0\n",
      "2008.0,16.0,-4.5\n",
      "2008.0,17.0,-4.0\n",
      "2008.0,18.0,-0.5\n",
      "2008.0,19.0,2.5\n",
      "2008.0,20.0,-1.5\n",
      "2008.0,21.0,-5.0\n",
      "2008.0,22.0,-7.5\n",
      "2008.0,23.0,-5.0\n",
      "2008.0,24.0,-5.0\n",
      "2008.0,25.0,-3.0\n",
      "2008.0,26.0,1.0\n",
      "2008.0,27.0,2.5\n",
      "2008.0,28.0,0.5\n",
      "2008.0,29.0,-3.0\n",
      "2008.0,30.0,-2.5\n",
      "2008.0,31.0,-1.0\n",
      "2008.0,32.0,-1.0\n",
      "2008.0,33.0,-1.0\n",
      "2008.0,34.0,-1.5\n",
      "2008.0,35.0,-0.5\n",
      "2008.0,36.0,-1.0\n",
      "2008.0,37.0,-2.5\n",
      "2008.0,38.0,0.0\n",
      "2008.0,39.0,0.5\n",
      "2008.0,40.0,4.0\n",
      "2008.0,41.0,5.0\n",
      "2008.0,42.0,5.0\n",
      "2008.0,43.0,7.0\n",
      "2008.0,44.0,4.0\n",
      "2008.0,45.0,1.5\n",
      "2008.0,46.0,5.5\n",
      "2008.0,47.0,4.5\n",
      "2008.0,48.0,5.5\n",
      "2008.0,49.0,7.0\n",
      "2008.0,50.0,8.5\n",
      "2008.0,51.0,8.5\n",
      "2008.0,52.0,8.0\n",
      "2008.0,53.0,7.0\n",
      "2008.0,54.0,6.5\n",
      "2008.0,55.0,6.0\n",
      "2008.0,56.0,5.0\n",
      "2008.0,57.0,7.0\n",
      "2008.0,58.0,6.0\n",
      "2008.0,59.0,8.0\n",
      "2008.0,60.0,10.5\n",
      "2008.0,61.0,7.5\n",
      "2008.0,62.0,3.0\n",
      "2008.0,63.0,4.0\n",
      "2008.0,64.0,2.0\n",
      "2008.0,65.0,2.5\n",
      "2008.0,66.0,6.5\n",
      "2008.0,67.0,6.5\n",
      "2008.0,68.0,5.0\n",
      "2008.0,69.0,8.5\n",
      "2008.0,70.0,12.5\n",
      "2008.0,71.0,9.5\n",
      "2008.0,72.0,9.5\n",
      "2008.0,73.0,6.5\n",
      "2008.0,74.0,4.5\n",
      "2008.0,75.0,4.5\n",
      "2008.0,76.0,3.5\n",
      "2008.0,77.0,5.5\n",
      "2008.0,78.0,7.0\n",
      "2008.0,79.0,7.0\n",
      "2008.0,80.0,6.0\n",
      "2008.0,81.0,3.5\n",
      "2008.0,82.0,7.0\n",
      "2008.0,83.0,8.5\n",
      "2008.0,84.0,7.0\n",
      "2008.0,85.0,6.5\n",
      "2008.0,86.0,5.5\n",
      "2008.0,87.0,2.5\n",
      "2008.0,88.0,1.5\n",
      "2008.0,89.0,1.5\n",
      "2008.0,90.0,3.5\n",
      "2008.0,91.0,4.0\n",
      "2008.0,92.0,3.0\n",
      "2008.0,93.0,4.5\n",
      "2008.0,94.0,9.0\n",
      "2008.0,95.0,10.5\n",
      "2008.0,96.0,7.5\n",
      "2008.0,97.0,6.0\n",
      "2008.0,98.0,6.5\n",
      "2008.0,99.0,6.0\n",
      "2008.0,100.0,6.0\n",
      "2008.0,101.0,6.5\n",
      "2008.0,102.0,9.0\n",
      "2008.0,103.0,16.5\n",
      "2008.0,104.0,20.0\n",
      "2008.0,105.0,19.5\n",
      "2008.0,106.0,10.5\n",
      "2008.0,107.0,5.0\n",
      "2008.0,108.0,11.5\n",
      "2008.0,109.0,14.5\n",
      "2008.0,110.0,9.0\n",
      "2008.0,111.0,4.5\n",
      "2008.0,112.0,2.0\n",
      "2008.0,113.0,8.5\n",
      "2008.0,114.0,9.5\n",
      "2008.0,115.0,8.5\n",
      "2008.0,116.0,6.5\n",
      "2008.0,117.0,10.5\n",
      "2008.0,118.0,16.0\n",
      "2008.0,119.0,19.5\n",
      "2008.0,120.0,16.5\n",
      "2008.0,121.0,8.0\n",
      "2008.0,122.0,6.0\n",
      "2008.0,123.0,12.0\n",
      "2008.0,124.0,15.5\n",
      "2008.0,125.0,17.5\n",
      "2008.0,126.0,19.5\n",
      "2008.0,127.0,19.0\n",
      "2008.0,128.0,13.5\n",
      "2008.0,129.0,12.0\n",
      "2008.0,130.0,11.0\n",
      "2008.0,131.0,13.0\n",
      "2008.0,132.0,12.5\n",
      "2008.0,133.0,10.0\n",
      "2008.0,134.0,13.0\n",
      "2008.0,135.0,13.0\n",
      "2008.0,136.0,18.0\n",
      "2008.0,137.0,25.5\n",
      "2008.0,138.0,28.5\n",
      "2008.0,139.0,28.0\n",
      "2008.0,140.0,26.0\n",
      "2008.0,141.0,25.0\n",
      "2008.0,142.0,15.5\n",
      "2008.0,143.0,7.0\n",
      "2008.0,144.0,12.5\n",
      "2008.0,145.0,15.0\n",
      "2008.0,146.0,15.5\n",
      "2008.0,147.0,16.5\n",
      "2008.0,148.0,18.0\n",
      "2008.0,149.0,16.5\n",
      "2008.0,150.0,14.0\n",
      "2008.0,151.0,16.0\n",
      "2008.0,152.0,19.5\n",
      "2008.0,153.0,19.5\n",
      "2008.0,154.0,16.0\n",
      "2008.0,155.0,14.0\n",
      "2008.0,156.0,9.0\n",
      "2008.0,157.0,13.5\n",
      "2008.0,158.0,12.5\n",
      "2008.0,159.0,11.0\n",
      "2008.0,160.0,13.5\n",
      "2008.0,161.0,17.5\n",
      "2008.0,162.0,11.0\n",
      "2008.0,163.0,7.0\n",
      "2008.0,164.0,11.0\n",
      "2008.0,165.0,19.0\n",
      "2008.0,166.0,21.5\n",
      "2008.0,167.0,23.5\n",
      "2008.0,168.0,25.5\n",
      "2008.0,169.0,25.5\n",
      "2008.0,170.0,24.5\n",
      "2008.0,171.0,24.0\n",
      "2008.0,172.0,26.5\n",
      "2008.0,173.0,29.5\n",
      "2008.0,174.0,27.0\n",
      "2008.0,175.0,25.5\n",
      "2008.0,176.0,25.5\n",
      "2008.0,177.0,26.5\n",
      "2008.0,178.0,25.0\n",
      "2008.0,179.0,25.5\n",
      "2008.0,180.0,30.0\n",
      "2008.0,181.0,33.5\n",
      "2008.0,182.0,33.5\n",
      "2008.0,183.0,28.0\n",
      "2008.0,184.0,27.5\n",
      "2008.0,185.0,32.0\n",
      "2008.0,186.0,31.5\n",
      "2008.0,187.0,27.0\n",
      "2008.0,188.0,26.0\n",
      "2008.0,189.0,27.0\n",
      "2008.0,190.0,27.5\n",
      "2008.0,191.0,29.0\n",
      "2008.0,192.0,30.0\n",
      "2008.0,193.0,26.5\n",
      "2008.0,194.0,26.5\n",
      "2008.0,195.0,28.5\n",
      "2008.0,196.0,29.0\n",
      "2008.0,197.0,29.5\n",
      "2008.0,198.0,28.0\n",
      "2008.0,199.0,27.0\n",
      "2008.0,200.0,27.5\n",
      "2008.0,201.0,28.0\n",
      "2008.0,202.0,31.0\n",
      "2008.0,203.0,32.5\n",
      "2008.0,204.0,27.0\n",
      "2008.0,205.0,24.0\n",
      "2008.0,206.0,28.0\n",
      "2008.0,207.0,30.5\n",
      "2008.0,208.0,31.5\n",
      "2008.0,209.0,30.0\n",
      "2008.0,210.0,29.0\n",
      "2008.0,211.0,28.5\n",
      "2008.0,212.0,26.5\n",
      "2008.0,213.0,28.0\n",
      "2008.0,214.0,29.5\n",
      "2008.0,215.0,27.5\n",
      "2008.0,216.0,26.0\n",
      "2008.0,217.0,27.5\n",
      "2008.0,218.0,31.0\n",
      "2008.0,219.0,33.0\n",
      "2008.0,220.0,32.0\n",
      "2008.0,221.0,27.5\n",
      "2008.0,222.0,27.0\n",
      "2008.0,223.0,25.5\n",
      "2008.0,224.0,25.0\n",
      "2008.0,225.0,26.5\n",
      "2008.0,226.0,28.5\n",
      "2008.0,227.0,29.5\n",
      "2008.0,228.0,30.5\n",
      "2008.0,229.0,32.0\n",
      "2008.0,230.0,33.5\n",
      "2008.0,231.0,34.0\n",
      "2008.0,232.0,30.5\n",
      "2008.0,233.0,24.5\n",
      "2008.0,234.0,20.5\n",
      "2008.0,235.0,22.0\n",
      "2008.0,236.0,27.0\n",
      "2008.0,237.0,31.5\n",
      "2008.0,238.0,32.0\n",
      "2008.0,239.0,26.0\n",
      "2008.0,240.0,20.0\n",
      "2008.0,241.0,22.5\n",
      "2008.0,242.0,27.0\n",
      "2008.0,243.0,28.5\n",
      "2008.0,244.0,19.5\n",
      "2008.0,245.0,12.5\n",
      "2008.0,246.0,16.0\n",
      "2008.0,247.0,20.0\n",
      "2008.0,248.0,18.5\n",
      "2008.0,249.0,19.0\n",
      "2008.0,250.0,20.0\n",
      "2008.0,251.0,21.0\n",
      "2008.0,252.0,23.0\n",
      "2008.0,253.0,24.0\n",
      "2008.0,254.0,20.5\n",
      "2008.0,255.0,20.5\n",
      "2008.0,256.0,24.0\n",
      "2008.0,257.0,24.5\n",
      "2008.0,258.0,25.5\n",
      "2008.0,259.0,28.0\n",
      "2008.0,260.0,29.5\n",
      "2008.0,261.0,29.5\n",
      "2008.0,262.0,28.5\n",
      "2008.0,263.0,28.0\n",
      "2008.0,264.0,22.5\n",
      "2008.0,265.0,14.5\n",
      "2008.0,266.0,13.5\n",
      "2008.0,267.0,15.0\n",
      "2008.0,268.0,19.5\n",
      "2008.0,269.0,21.0\n",
      "2008.0,270.0,22.5\n",
      "2008.0,271.0,23.0\n",
      "2008.0,272.0,24.0\n",
      "2008.0,273.0,25.0\n",
      "2008.0,274.0,26.5\n",
      "2008.0,275.0,26.0\n",
      "2008.0,276.0,23.5\n",
      "2008.0,277.0,20.5\n",
      "2008.0,278.0,17.0\n",
      "2008.0,279.0,13.0\n",
      "2008.0,280.0,14.0\n",
      "2008.0,281.0,16.0\n",
      "2008.0,282.0,13.5\n",
      "2008.0,283.0,8.0\n",
      "2008.0,284.0,4.0\n",
      "2008.0,285.0,2.5\n",
      "2008.0,286.0,4.0\n",
      "2008.0,287.0,8.5\n",
      "2008.0,288.0,10.5\n",
      "2008.0,289.0,8.5\n",
      "2008.0,290.0,8.5\n",
      "2008.0,291.0,15.5\n",
      "2008.0,292.0,18.0\n",
      "2008.0,293.0,17.5\n",
      "2008.0,294.0,15.0\n",
      "2008.0,295.0,9.5\n",
      "2008.0,296.0,8.5\n",
      "2008.0,297.0,11.5\n",
      "2008.0,298.0,12.5\n",
      "2008.0,299.0,12.5\n",
      "2008.0,300.0,14.5\n",
      "2008.0,301.0,15.5\n",
      "2008.0,302.0,17.0\n",
      "2008.0,303.0,16.5\n",
      "2008.0,304.0,16.0\n",
      "2008.0,305.0,13.5\n",
      "2008.0,306.0,13.5\n",
      "2008.0,307.0,10.0\n",
      "2008.0,308.0,7.5\n",
      "2008.0,309.0,6.0\n",
      "2008.0,310.0,4.5\n",
      "2008.0,311.0,4.5\n",
      "2008.0,312.0,7.0\n",
      "2008.0,313.0,9.0\n",
      "2008.0,314.0,7.5\n",
      "2008.0,315.0,7.0\n",
      "2008.0,316.0,5.5\n",
      "2008.0,317.0,7.0\n",
      "2008.0,318.0,8.5\n",
      "2008.0,319.0,8.0\n",
      "2008.0,320.0,9.0\n",
      "2008.0,321.0,10.5\n",
      "2008.0,322.0,10.5\n",
      "2008.0,323.0,12.5\n",
      "2008.0,324.0,10.0\n",
      "2008.0,325.0,9.0\n",
      "2008.0,326.0,7.5\n",
      "2008.0,327.0,3.0\n",
      "2008.0,328.0,3.0\n",
      "2008.0,329.0,5.0\n",
      "2008.0,330.0,5.0\n",
      "2008.0,331.0,5.5\n",
      "2008.0,332.0,5.5\n",
      "2008.0,333.0,3.5\n",
      "2008.0,334.0,5.0\n",
      "2008.0,335.0,6.5\n",
      "2008.0,336.0,8.0\n",
      "2008.0,337.0,5.0\n",
      "2008.0,338.0,3.0\n",
      "2008.0,339.0,2.0\n",
      "2008.0,340.0,0.5\n",
      "2008.0,341.0,2.0\n",
      "2008.0,342.0,2.5\n",
      "2008.0,343.0,1.0\n",
      "2008.0,344.0,0.5\n",
      "2008.0,345.0,2.0\n",
      "2008.0,346.0,3.5\n",
      "2008.0,347.0,1.0\n",
      "2008.0,348.0,-0.5\n",
      "2008.0,349.0,-7.5\n",
      "2008.0,350.0,-9.0\n",
      "2008.0,351.0,-9.0\n",
      "2008.0,352.0,-7.5\n",
      "2008.0,353.0,-5.0\n",
      "2008.0,354.0,-5.0\n",
      "2008.0,355.0,-7.0\n",
      "2008.0,356.0,-5.5\n",
      "2008.0,357.0,-4.0\n",
      "2008.0,358.0,-4.5\n",
      "2008.0,359.0,-4.0\n",
      "2008.0,360.0,-3.5\n",
      "2008.0,361.0,-4.0\n",
      "2008.0,362.0,-4.0\n",
      "2008.0,363.0,0.5\n",
      "2008.0,364.0,4.0\n",
      "2008.0,365.0,2.5\n",
      "2009.0,1.0,1.5\n",
      "2009.0,2.0,1.0\n",
      "2009.0,3.0,-1.0\n",
      "2009.0,4.0,-5.5\n",
      "2009.0,5.0,-6.0\n",
      "2009.0,6.0,-1.5\n",
      "2009.0,7.0,2.0\n",
      "2009.0,8.0,3.5\n",
      "2009.0,9.0,2.0\n",
      "2009.0,10.0,2.0\n",
      "2009.0,11.0,2.0\n",
      "2009.0,12.0,3.5\n",
      "2009.0,13.0,4.0\n",
      "2009.0,14.0,5.0\n",
      "2009.0,15.0,4.5\n",
      "2009.0,16.0,4.0\n",
      "2009.0,17.0,4.0\n",
      "2009.0,18.0,3.5\n",
      "2009.0,19.0,3.5\n",
      "2009.0,20.0,4.0\n",
      "2009.0,21.0,2.5\n",
      "2009.0,22.0,1.0\n",
      "2009.0,23.0,4.0\n",
      "2009.0,24.0,1.0\n",
      "2009.0,25.0,-1.5\n",
      "2009.0,26.0,-4.5\n",
      "2009.0,27.0,-7.5\n",
      "2009.0,28.0,-4.5\n",
      "2009.0,29.0,1.0\n",
      "2009.0,30.0,3.5\n",
      "2009.0,31.0,2.0\n",
      "2009.0,32.0,1.0\n",
      "2009.0,33.0,0.5\n",
      "2009.0,34.0,6.5\n",
      "2009.0,35.0,7.5\n",
      "2009.0,36.0,5.0\n",
      "2009.0,37.0,3.0\n",
      "2009.0,38.0,5.5\n",
      "2009.0,39.0,5.5\n",
      "2009.0,40.0,2.5\n",
      "2009.0,41.0,1.0\n",
      "2009.0,42.0,1.0\n",
      "2009.0,43.0,2.0\n",
      "2009.0,44.0,0.5\n",
      "2009.0,45.0,1.0\n",
      "2009.0,46.0,1.5\n",
      "2009.0,47.0,3.5\n",
      "2009.0,48.0,5.0\n",
      "2009.0,49.0,4.5\n",
      "2009.0,50.0,5.0\n",
      "2009.0,51.0,6.5\n",
      "2009.0,52.0,9.0\n",
      "2009.0,53.0,8.0\n",
      "2009.0,54.0,8.0\n",
      "2009.0,55.0,7.5\n",
      "2009.0,56.0,5.0\n",
      "2009.0,57.0,3.0\n",
      "2009.0,58.0,1.5\n",
      "2009.0,59.0,3.5\n",
      "2009.0,60.0,7.0\n",
      "2009.0,61.0,10.5\n",
      "2009.0,62.0,9.5\n",
      "2009.0,63.0,4.0\n",
      "2009.0,64.0,1.5\n",
      "2009.0,65.0,1.0\n",
      "2009.0,66.0,2.5\n",
      "2009.0,67.0,1.0\n",
      "2009.0,68.0,0.0\n",
      "2009.0,69.0,-2.5\n",
      "2009.0,70.0,0.0\n",
      "2009.0,71.0,5.0\n",
      "2009.0,72.0,7.5\n",
      "2009.0,73.0,9.5\n",
      "2009.0,74.0,6.5\n",
      "2009.0,75.0,4.5\n",
      "2009.0,76.0,3.5\n",
      "2009.0,77.0,7.0\n",
      "2009.0,78.0,8.5\n",
      "2009.0,79.0,12.0\n",
      "2009.0,80.0,13.0\n",
      "2009.0,81.0,8.0\n",
      "2009.0,82.0,4.5\n",
      "2009.0,83.0,5.0\n",
      "2009.0,84.0,2.5\n",
      "2009.0,85.0,2.5\n",
      "2009.0,86.0,5.5\n",
      "2009.0,87.0,7.0\n",
      "2009.0,88.0,4.0\n",
      "2009.0,89.0,2.0\n",
      "2009.0,90.0,3.0\n",
      "2009.0,91.0,3.0\n",
      "2009.0,92.0,4.0\n",
      "2009.0,93.0,4.5\n",
      "2009.0,94.0,6.0\n",
      "2009.0,95.0,11.0\n",
      "2009.0,96.0,15.0\n",
      "2009.0,97.0,18.5\n",
      "2009.0,98.0,14.5\n",
      "2009.0,99.0,10.0\n",
      "2009.0,100.0,11.5\n",
      "2009.0,101.0,11.5\n",
      "2009.0,102.0,12.5\n",
      "2009.0,103.0,11.0\n",
      "2009.0,104.0,6.5\n",
      "2009.0,105.0,4.5\n",
      "2009.0,106.0,7.5\n",
      "2009.0,107.0,13.0\n",
      "2009.0,108.0,16.0\n",
      "2009.0,109.0,17.0\n",
      "2009.0,110.0,20.5\n",
      "2009.0,111.0,23.0\n",
      "2009.0,112.0,22.5\n",
      "2009.0,113.0,15.0\n",
      "2009.0,114.0,7.5\n",
      "2009.0,115.0,8.5\n",
      "2009.0,116.0,9.0\n",
      "2009.0,117.0,9.5\n",
      "2009.0,118.0,9.0\n",
      "2009.0,119.0,7.0\n",
      "2009.0,120.0,8.5\n",
      "2009.0,121.0,10.5\n",
      "2009.0,122.0,14.5\n",
      "2009.0,123.0,11.0\n",
      "2009.0,124.0,11.5\n",
      "2009.0,125.0,12.0\n",
      "2009.0,126.0,12.0\n",
      "2009.0,127.0,9.0\n",
      "2009.0,128.0,8.5\n",
      "2009.0,129.0,11.5\n",
      "2009.0,130.0,16.0\n",
      "2009.0,131.0,17.0\n",
      "2009.0,132.0,11.5\n",
      "2009.0,133.0,11.5\n",
      "2009.0,134.0,13.0\n",
      "2009.0,135.0,16.0\n",
      "2009.0,136.0,20.0\n",
      "2009.0,137.0,25.0\n",
      "2009.0,138.0,28.0\n",
      "2009.0,139.0,25.5\n",
      "2009.0,140.0,18.5\n",
      "2009.0,141.0,17.5\n",
      "2009.0,142.0,23.0\n",
      "2009.0,143.0,25.5\n",
      "2009.0,144.0,22.0\n",
      "2009.0,145.0,19.5\n",
      "2009.0,146.0,22.0\n",
      "2009.0,147.0,23.0\n",
      "2009.0,148.0,25.5\n",
      "2009.0,149.0,27.0\n",
      "2009.0,150.0,26.5\n",
      "2009.0,151.0,26.0\n",
      "2009.0,152.0,22.0\n",
      "2009.0,153.0,19.0\n",
      "2009.0,154.0,21.5\n",
      "2009.0,155.0,23.5\n",
      "2009.0,156.0,21.5\n",
      "2009.0,157.0,18.0\n",
      "2009.0,158.0,13.5\n",
      "2009.0,159.0,14.5\n",
      "2009.0,160.0,17.0\n",
      "2009.0,161.0,19.5\n",
      "2009.0,162.0,20.0\n",
      "2009.0,163.0,19.0\n",
      "2009.0,164.0,19.0\n",
      "2009.0,165.0,20.0\n",
      "2009.0,166.0,17.5\n",
      "2009.0,167.0,21.0\n",
      "2009.0,168.0,20.5\n",
      "2009.0,169.0,19.5\n",
      "2009.0,170.0,20.5\n",
      "2009.0,171.0,17.0\n",
      "2009.0,172.0,16.0\n",
      "2009.0,173.0,13.5\n",
      "2009.0,174.0,18.0\n",
      "2009.0,175.0,25.5\n",
      "2009.0,176.0,27.5\n",
      "2009.0,177.0,24.0\n",
      "2009.0,178.0,24.0\n",
      "2009.0,179.0,27.5\n",
      "2009.0,180.0,29.0\n",
      "2009.0,181.0,28.5\n",
      "2009.0,182.0,28.5\n",
      "2009.0,183.0,29.5\n",
      "2009.0,184.0,28.5\n",
      "2009.0,185.0,27.5\n",
      "2009.0,186.0,28.5\n",
      "2009.0,187.0,25.5\n",
      "2009.0,188.0,22.5\n",
      "2009.0,189.0,19.5\n",
      "2009.0,190.0,18.5\n",
      "2009.0,191.0,23.0\n",
      "2009.0,192.0,27.5\n",
      "2009.0,193.0,27.5\n",
      "2009.0,194.0,23.0\n",
      "2009.0,195.0,21.0\n",
      "2009.0,196.0,27.0\n",
      "2009.0,197.0,29.5\n",
      "2009.0,198.0,31.5\n",
      "2009.0,199.0,32.5\n",
      "2009.0,200.0,31.5\n",
      "2009.0,201.0,29.0\n",
      "2009.0,202.0,30.0\n",
      "2009.0,203.0,32.0\n",
      "2009.0,204.0,33.5\n",
      "2009.0,205.0,32.5\n",
      "2009.0,206.0,31.5\n",
      "2009.0,207.0,31.0\n",
      "2009.0,208.0,27.5\n",
      "2009.0,209.0,26.5\n",
      "2009.0,210.0,26.5\n",
      "2009.0,211.0,27.5\n",
      "2009.0,212.0,29.0\n",
      "2009.0,213.0,30.5\n",
      "2009.0,214.0,32.0\n",
      "2009.0,215.0,32.5\n",
      "2009.0,216.0,31.5\n",
      "2009.0,217.0,30.5\n",
      "2009.0,218.0,24.0\n",
      "2009.0,219.0,18.0\n",
      "2009.0,220.0,17.5\n",
      "2009.0,221.0,20.5\n",
      "2009.0,222.0,24.5\n",
      "2009.0,223.0,29.0\n",
      "2009.0,224.0,29.5\n",
      "2009.0,225.0,24.5\n",
      "2009.0,226.0,17.5\n",
      "2009.0,227.0,15.5\n",
      "2009.0,228.0,16.0\n",
      "2009.0,229.0,20.5\n",
      "2009.0,230.0,24.5\n",
      "2009.0,231.0,27.5\n",
      "2009.0,232.0,31.5\n",
      "2009.0,233.0,33.0\n",
      "2009.0,234.0,32.0\n",
      "2009.0,235.0,27.0\n",
      "2009.0,236.0,25.5\n",
      "2009.0,237.0,28.5\n",
      "2009.0,238.0,30.5\n",
      "2009.0,239.0,32.0\n",
      "2009.0,240.0,33.0\n",
      "2009.0,241.0,29.5\n",
      "2009.0,242.0,23.5\n",
      "2009.0,243.0,21.0\n",
      "2009.0,244.0,25.5\n",
      "2009.0,245.0,29.5\n",
      "2009.0,246.0,29.0\n",
      "2009.0,247.0,29.5\n",
      "2009.0,248.0,28.5\n",
      "2009.0,249.0,24.5\n",
      "2009.0,250.0,20.0\n",
      "2009.0,251.0,20.0\n",
      "2009.0,252.0,23.5\n",
      "2009.0,253.0,25.5\n",
      "2009.0,254.0,28.0\n",
      "2009.0,255.0,29.5\n",
      "2009.0,256.0,29.0\n",
      "2009.0,257.0,27.5\n",
      "2009.0,258.0,26.0\n",
      "2009.0,259.0,28.0\n",
      "2009.0,260.0,27.5\n",
      "2009.0,261.0,27.0\n",
      "2009.0,262.0,28.5\n",
      "2009.0,263.0,21.0\n",
      "2009.0,264.0,19.0\n",
      "2009.0,265.0,24.5\n",
      "2009.0,266.0,29.0\n",
      "2009.0,267.0,30.5\n",
      "2009.0,268.0,29.0\n",
      "2009.0,269.0,27.0\n",
      "2009.0,270.0,25.5\n",
      "2009.0,271.0,25.5\n",
      "2009.0,272.0,22.0\n",
      "2009.0,273.0,9.5\n",
      "2009.0,274.0,7.5\n",
      "2009.0,275.0,10.5\n",
      "2009.0,276.0,12.0\n",
      "2009.0,277.0,9.0\n",
      "2009.0,278.0,8.5\n",
      "2009.0,279.0,12.0\n",
      "2009.0,280.0,11.5\n",
      "2009.0,281.0,10.5\n",
      "2009.0,282.0,9.0\n",
      "2009.0,283.0,7.0\n",
      "2009.0,284.0,6.0\n",
      "2009.0,285.0,7.5\n",
      "2009.0,286.0,12.0\n",
      "2009.0,287.0,13.5\n",
      "2009.0,288.0,12.0\n",
      "2009.0,289.0,15.5\n",
      "2009.0,290.0,19.0\n",
      "2009.0,291.0,17.5\n",
      "2009.0,292.0,13.0\n",
      "2009.0,293.0,9.0\n",
      "2009.0,294.0,8.5\n",
      "2009.0,295.0,9.0\n",
      "2009.0,296.0,10.5\n",
      "2009.0,297.0,7.0\n",
      "2009.0,298.0,9.0\n",
      "2009.0,299.0,10.5\n",
      "2009.0,300.0,6.0\n",
      "2009.0,301.0,1.0\n",
      "2009.0,302.0,1.5\n",
      "2009.0,303.0,4.0\n",
      "2009.0,304.0,8.0\n",
      "2009.0,305.0,7.5\n",
      "2009.0,306.0,8.5\n",
      "2009.0,307.0,9.0\n",
      "2009.0,308.0,11.5\n",
      "2009.0,309.0,13.0\n",
      "2009.0,310.0,12.0\n",
      "2009.0,311.0,7.0\n",
      "2009.0,312.0,7.0\n",
      "2009.0,313.0,8.5\n",
      "2009.0,314.0,8.5\n",
      "2009.0,315.0,7.0\n",
      "2009.0,316.0,2.5\n",
      "2009.0,317.0,2.0\n",
      "2009.0,318.0,0.5\n",
      "2009.0,319.0,2.0\n",
      "2009.0,320.0,3.5\n",
      "2009.0,321.0,8.5\n",
      "2009.0,322.0,6.5\n",
      "2009.0,323.0,4.5\n",
      "2009.0,324.0,5.0\n",
      "2009.0,325.0,2.0\n",
      "2009.0,326.0,1.0\n",
      "2009.0,327.0,2.0\n",
      "2009.0,328.0,3.0\n",
      "2009.0,329.0,6.5\n",
      "2009.0,330.0,5.5\n",
      "2009.0,331.0,4.5\n",
      "2009.0,332.0,3.0\n",
      "2009.0,333.0,2.5\n",
      "2009.0,334.0,4.0\n",
      "2009.0,335.0,2.0\n",
      "2009.0,336.0,1.0\n",
      "2009.0,337.0,-3.0\n",
      "2009.0,338.0,-3.0\n",
      "2009.0,339.0,-5.0\n",
      "2009.0,340.0,-8.0\n",
      "2009.0,341.0,-10.0\n",
      "2009.0,342.0,-11.0\n",
      "2009.0,343.0,-12.0\n",
      "2009.0,344.0,-9.5\n",
      "2009.0,345.0,-6.5\n",
      "2009.0,346.0,-4.0\n",
      "2009.0,347.0,-2.0\n",
      "2009.0,348.0,-0.5\n",
      "2009.0,349.0,1.0\n",
      "2009.0,350.0,3.0\n",
      "2009.0,351.0,4.0\n",
      "2009.0,352.0,2.0\n",
      "2009.0,353.0,1.0\n",
      "2009.0,354.0,2.0\n",
      "2009.0,355.0,4.5\n",
      "2009.0,356.0,1.5\n",
      "2009.0,357.0,-1.5\n",
      "2009.0,358.0,-3.0\n",
      "2009.0,359.0,-1.5\n",
      "2009.0,360.0,-1.5\n",
      "2009.0,361.0,-2.0\n",
      "2009.0,362.0,-4.0\n",
      "2009.0,363.0,-1.5\n",
      "2009.0,364.0,-3.0\n",
      "2009.0,365.0,-2.0\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# import urllib\n",
    "import urllib.request\n",
    "\n",
    "# Define the URL\n",
    "url = \"https://daymet.ornl.gov/single-pixel/api/\" + \\\n",
    "    \"data?lat=45.4&lon=-115.0534&vars=tmax&start=2000-01-01&end=2009-12-31\"\n",
    "\n",
    "# define a request object that points to the desired URL\n",
    "req = urllib.request.Request(url)\n",
    "# Open the URL and return a handle to it in\n",
    "#       variable response\n",
    "with urllib.request.urlopen(req) as response:\n",
    "    # Read the contents of the url, and convert to text\n",
    "    the_data = response.read().decode(\"utf-8\")\n",
    "print(the_data)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see the contents of the file: it's got a few lines describing the data (\"metadata\"), a header and then there's the actual data block: three variables (year, day of year and maximum temperature in degrees Celsius).\n",
    "\n",
    "We can read this data as a numpy array. Note that variable `the_data` is a string, so we can split it by lines using the `.split` method"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(3650, 3)\n"
     ]
    }
   ],
   "source": [
    "data = np.loadtxt(the_data.split(\"\\n\"), delimiter=\",\", skiprows=8)\n",
    "print(data.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In the previous example, we have just used the data from the server. In most cases, you'll want to keep a local copy of the data (e.g. in case the server goes down next time, or to work if you have no network connectivity). To store the data in a file, we can open a local file (giving it the local path and name), and tell python to open it for `w`riting. We get a handle to the file (here it's variable `fp`), and we can use the `write` method to write the data to it.\n",
    "\n",
    "In general, when you open a file, you should also close it (e.g. `fp.close()`). However, if you open the file within a context manager (the `with open(...) as fp`), Python will close the file once it exists the block:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(3650, 3)\n"
     ]
    }
   ],
   "source": [
    "# We open the output file, `daymet.csv`\n",
    "with open(\"data/daymet_tmax.csv\", 'w') as fp:\n",
    "    # make the HTTPS connection and pull text\n",
    "    # then write to file\n",
    "    fp.write(the_data)\n",
    "# The file is now automatically closed after the context manager block\n",
    "\n",
    "# You can check the text file to see its contents, but we now\n",
    "# (i) it's separated by commas\n",
    "# (ii) the first 9 lines are metadata that we're not interested in.\n",
    "\n",
    "temperature = np.loadtxt(\"data/daymet_tmax.csv\", skiprows=8, delimiter=\",\")\n",
    "# We expect to get ~10 years of data here, so 3650 daily records\n",
    "print(temperature.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The data columns are: `Year`, `day of year` (1 to 365) and `Tmax` ($C$).\n",
    "\n",
    "How can we plot such data? the technical issue we face is needing to use the first *two* columns of data (day of year and year) to describe the x-axis location. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Year  [ 2.0e+03  1.0e+00 -1.5e+00]\n",
      "DOY   [2000.     2.    -2.5]\n",
      "T_max [ 2.0e+03  3.0e+00 -1.5e+00]\n"
     ]
    }
   ],
   "source": [
    "print('Year ', temperature[0])\n",
    "print('DOY  ', temperature[1])\n",
    "print('T_max', temperature[2])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "\n",
    "A simple way of doing this, that would suffice here, would be to convert day of year to year fraction, then we could write:\n",
    "\n",
    "    year,doy,tmax = temperature\n",
    "    dates = year + (doy-1)/365.\n",
    "    \n",
    "A more elegant way might be to use [`datetime`](https://docs.python.org/3/library/datetime.html). This contains a set of methods that allow you to manipulate date formats. `matplotlib` understands the format used, and so it is generally appropriate to use `datetime` for date information when plotting. One way of doing this is to use a [`timedelta` object](https://docs.python.org/3/library/datetime.html#timedelta-objects) and just add the day of year to the first date on the dataset:\n",
    "\n",
    "```python\n",
    "    # Loop over all rows in the dataset\n",
    "    # Define the start date to the day before the dataset starts\n",
    "    the_date = datetime.datetime(1999, 12, 31)\n",
    "    dates = []\n",
    "    for i in range(temperature.shape[0]):\n",
    "        the_date = the_date + datetime.timedelta(days=1)\n",
    "        dates.append(the_date)\n",
    "```\n",
    "\n",
    "Other more generic approaches would build a string with the date (e.g. `\"2000/137\"`) and use the [`strptime` function](https://docs.python.org/3/library/datetime.html#datetime.datetime.strptime) to interpret the string. We will use the format `\"%Y/j\"`, which according [to this webpage](https://docs.python.org/3/library/datetime.html#strftime-strptime-behavior) stands for `Year/day of year`. Note that in order to build the string, we need to convert the year and day of year to be integers:\n",
    "```python\n",
    "    dates = []\n",
    "    for i in range(temperature.shape[0]):\n",
    "        the_date = f\"{int(temperature[i][0]):d}/{int(temperature[i][1]):d}\"\n",
    "        dates.append(datetime.datetime.strptime(the_date, \"%Y/%j\"))\n",
    "    \n",
    "```\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'year')"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import datetime\n",
    "year = temperature[:, 0]\n",
    "doy = temperature[:, 1]\n",
    "tmax = temperature[:, 2]\n",
    "\n",
    "dates = year + (doy-1)/365.\n",
    "# using the simple way here\n",
    "\n",
    "plt.figure(figsize=(14, 3))\n",
    "plt.plot(dates, tmax)\n",
    "plt.title(url)\n",
    "plt.ylabel('$T_{max} / C$')\n",
    "plt.xlabel('year')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Exercise E2.5.3**\n",
    "\n",
    "* use the datetime approach to plot the dataset\n",
    "* print out the value of `dates` for the first 10 entries to see what the format looks like"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "# do exercise here"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Although we have used this as a one-dimensional dataset (temperature as a function of time) we could also think of it as two-dimensional (temperature as a function of (year,doy)). Recall that the `shape` of the `temperature` dataset was `(3,3650)`. We could put the temperature column into a gridded dataset of shape `(10,365)` which would then emphasise the 2-D nature.\n",
    "\n",
    "We can do this with the `numpy` method `reshape()`. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1, 366)"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "year = temperature[:, 0].reshape(10, 365)\n",
    "doy = temperature[:, 1].reshape(10, 365)\n",
    "tmax = temperature[:, 2].reshape(10, 365)\n",
    "\n",
    "plt.figure(figsize=(14, 3))\n",
    "plt.plot(doy, tmax, 'x')\n",
    "plt.title(url)\n",
    "plt.ylabel('$T_{max} / C$')\n",
    "plt.xlabel('day of year')\n",
    "plt.xlim([1, 366])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Plotting this, we can visualise the year-on-year variations in temperature for any particular day.\n",
    "\n",
    "**Exercise E2.5.3**\n",
    "\n",
    "* using the reshaped datasets above, calculate and plot the mean value of `tmax` as a function of day of year\n",
    "* calculate standard deviation of `tmax` as a function of day of year, and plot dashed lines at mean +/- 1.96 standard deviations\n",
    "* in another plot, show the mean, +/- 1.96 standard deviations of `tmax` as a function of year (i.e. the annual average and standard deviation)\n",
    "\n",
    "Hint: use `axis=0` when calculating the mean/std over `doy` of `tmax` and `axis=1` for processing over `year`."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.5.4 Homework\n",
    "\n",
    "#### Exercise E2.5.4\n",
    "\n",
    "Select 4 locations in different regions of North America (e.g. Anchorage, Albuquerque, Seattle, Chicago). Request data on maximum temperature, precipitation and incident solar radiation for the years between 1981 to 2010, and plot in 3 different figures:\n",
    "\n",
    "* Figure 1: The *mean daily temperature* and the variation (a shaded area around the mean going from mean value **minus** 1.96 times the standard deviation to mean value **plus** 1.96 times the standard deviation). Use a subplot or panel for each site\n",
    "* Figure 2: The *mean daily precipitation* and the variation (a shaded area around the mean going from mean value **minus** 1.96 times the standard deviation to mean value **plus** 1.96 times the standard deviation). Use a subplot or panel for each site\n",
    "* Figure 3: The *mean daily incident solar radiation* and the variation (a shaded area around the mean going from mean value **minus** 1.96 times the standard deviation to mean value **plus** 1.96 times the standard deviation). Use a subplot or panel for each site\n",
    "\n",
    "In each plot, the mean value should be a full line, and the variation should be an envelope, visually similar to the plot shown below (clearly not identical!!!!)\n",
    "\n",
    "![la niña plot](https://www.metoffice.gov.uk/binaries/content/gallery/mohippo/images/news/charts-and-images/compare_datasets_new_logo_1850-1900-20122017a-630.jpg)\n",
    "\n",
    "\n",
    "Label each plot with a title, units and so on. Some useful functions to consider\n",
    "\n",
    "* [`plt.subplots`](https://matplotlib.org/api/_as_gen/matplotlib.pyplot.subplots.html) Allows you to split a figure into several panels or subplots. In particular, pay attention to the `sharex` and `sharey` options that allow you to have the same scales for all plots so they can be directly compared.\n",
    "* [`plt.fill_between`](https://matplotlib.org/api/_as_gen/matplotlib.pyplot.fill_between.html) Allows you to fill the space between two curves. You may want to give the option `color=0.8` for a nice grey effect."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "# do exercise here"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.5.5 Summary\n",
    "\n",
    "In this section, we have learned about reading data from csv files from the local disc or that we have pulled from the web (given a URL). We have gone into a little more detail and sophistication on plotting graphs, and you now should be able to produce sensible plots of datasets or summaries of datasets (e.g. mean standard deviation)."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.7"
  },
  "latex_envs": {
   "LaTeX_envs_menu_present": true,
   "autoclose": false,
   "autocomplete": true,
   "bibliofile": "biblio.bib",
   "cite_by": "apalike",
   "current_citInitial": 1,
   "eqLabelWithNumbers": true,
   "eqNumInitial": 1,
   "hotkeys": {
    "equation": "Ctrl-E",
    "itemize": "Ctrl-I"
   },
   "labels_anchors": false,
   "latex_user_defs": false,
   "report_style_numbering": false,
   "user_envs_cfg": false
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": false,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": true,
   "toc_position": {
    "height": "438px",
    "left": "10px",
    "top": "150px",
    "width": "165px"
   },
   "toc_section_display": true,
   "toc_window_display": true
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}