julia_tree.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {},
   "outputs": [],
   "source": [
    "using CSV, Statistics, Plots"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_X = CSV.read(\"boston_X.csv\", delim=\",\", header=false)\n",
    "df_y = CSV.read(\"boston_y.csv\", delim=\",\", header=false)\n",
    "X = convert(Matrix{Float64}, df_X);\n",
    "y = convert(Matrix{Float64}, df_y);\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((506, 13), (506, 1))"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "size(X), size(y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "rmse (generic function with 1 method)"
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "std_agg(cnt, s1, s2) = sqrt((s2/cnt) - (s1/cnt)^2)\n",
    "rmse(pred, truth) = sqrt(mean((pred .- truth).^2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "506×1 Array{Float64,2}:\n",
       "   1.4671936758893267\n",
       "  -0.9328063241106719\n",
       "  12.16719367588933\n",
       "  10.867193675889325\n",
       "  13.66719367588933\n",
       "   6.167193675889326\n",
       "   0.3671936758893253\n",
       "   4.567193675889328\n",
       "  -6.032806324110673\n",
       "  -3.6328063241106747\n",
       "  -7.532806324110673\n",
       "  -3.6328063241106747\n",
       "  -0.832806324110674\n",
       "   ⋮\n",
       "   1.9671936758893267\n",
       "   0.5671936758893281\n",
       "  -2.832806324110674\n",
       "  -4.232806324110673\n",
       "  -1.332806324110674\n",
       "  -5.032806324110673\n",
       "  -5.732806324110673\n",
       "  -0.13280632411067472\n",
       "  -1.9328063241106719\n",
       "   1.3671936758893253\n",
       "  -0.5328063241106733\n",
       " -10.632806324110673"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y .- mean(y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "rmse (generic function with 1 method)"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "root_val = mean(y)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2.0"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rmse(y.+2, y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "506×13 Array{Float64,2}:\n",
       " 0.00632  18.0   2.31  0.0  0.538  6.575  …  1.0  296.0  15.3  396.9    4.98\n",
       " 0.02731   0.0   7.07  0.0  0.469  6.421     2.0  242.0  17.8  396.9    9.14\n",
       " 0.02729   0.0   7.07  0.0  0.469  7.185     2.0  242.0  17.8  392.83   4.03\n",
       " 0.03237   0.0   2.18  0.0  0.458  6.998     3.0  222.0  18.7  394.63   2.94\n",
       " 0.06905   0.0   2.18  0.0  0.458  7.147     3.0  222.0  18.7  396.9    5.33\n",
       " 0.02985   0.0   2.18  0.0  0.458  6.43   …  3.0  222.0  18.7  394.12   5.21\n",
       " 0.08829  12.5   7.87  0.0  0.524  6.012     5.0  311.0  15.2  395.6   12.43\n",
       " 0.14455  12.5   7.87  0.0  0.524  6.172     5.0  311.0  15.2  396.9   19.15\n",
       " 0.21124  12.5   7.87  0.0  0.524  5.631     5.0  311.0  15.2  386.63  29.93\n",
       " 0.17004  12.5   7.87  0.0  0.524  6.004     5.0  311.0  15.2  386.71  17.1\n",
       " 0.22489  12.5   7.87  0.0  0.524  6.377  …  5.0  311.0  15.2  392.52  20.45\n",
       " 0.11747  12.5   7.87  0.0  0.524  6.009     5.0  311.0  15.2  396.9   13.27\n",
       " 0.09378  12.5   7.87  0.0  0.524  5.889     5.0  311.0  15.2  390.5   15.71\n",
       " ⋮                                 ⋮      ⋱               ⋮            \n",
       " 0.27957   0.0   9.69  0.0  0.585  5.926     6.0  391.0  19.2  396.9   13.59\n",
       " 0.17899   0.0   9.69  0.0  0.585  5.67   …  6.0  391.0  19.2  393.29  17.6\n",
       " 0.2896    0.0   9.69  0.0  0.585  5.39      6.0  391.0  19.2  396.9   21.14\n",
       " 0.26838   0.0   9.69  0.0  0.585  5.794     6.0  391.0  19.2  396.9   14.1\n",
       " 0.23912   0.0   9.69  0.0  0.585  6.019     6.0  391.0  19.2  396.9   12.92\n",
       " 0.17783   0.0   9.69  0.0  0.585  5.569     6.0  391.0  19.2  395.77  15.1\n",
       " 0.22438   0.0   9.69  0.0  0.585  6.027  …  6.0  391.0  19.2  396.9   14.33\n",
       " 0.06263   0.0  11.93  0.0  0.573  6.593     1.0  273.0  21.0  391.99   9.67\n",
       " 0.04527   0.0  11.93  0.0  0.573  6.12      1.0  273.0  21.0  396.9    9.08\n",
       " 0.06076   0.0  11.93  0.0  0.573  6.976     1.0  273.0  21.0  396.9    5.64\n",
       " 0.10959   0.0  11.93  0.0  0.573  6.794     1.0  273.0  21.0  393.45   6.48\n",
       " 0.04741   0.0  11.93  0.0  0.573  6.03   …  1.0  273.0  21.0  396.9    7.88"
      ]
     },
     "execution_count": 96,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1\n",
      "2\n"
     ]
    },
    {
     "ename": "DomainError",
     "evalue": "DomainError with -5.502442945726216e-11:\nsqrt will only return a complex result if called with a complex argument. Try sqrt(Complex(x)).",
     "output_type": "error",
     "traceback": [
      "DomainError with -5.502442945726216e-11:\nsqrt will only return a complex result if called with a complex argument. Try sqrt(Complex(x)).",
      "",
      "Stacktrace:",
      " [1] throw_complex_domainerror(::Symbol, ::Float64) at ./math.jl:33",
      " [2] sqrt at ./math.jl:557 [inlined]",
      " [3] std_agg(::Int64, ::Float64, ::Float64) at ./In[80]:1",
      " [4] top-level scope at ./In[109]:39"
     ]
    }
   ],
   "source": [
    "# for a given column, we are going to look for a split\n",
    "\n",
    "num_samples = length(y)\n",
    "num_cols = size(X, 2)\n",
    "\n",
    "x = X[:, 1]\n",
    "\n",
    "\n",
    "best_score = Inf\n",
    "best_split = 0\n",
    "lhs_idx = []\n",
    "rhs_idx = []\n",
    "split_col = 0\n",
    "\n",
    "\n",
    "for col in 1:2\n",
    "    println(col)\n",
    "    # try each split, moving an item from right to left\n",
    "    x = X[:, col]\n",
    "\n",
    "    # sort the x's\n",
    "    sort_idx = sortperm(x)\n",
    "    sorted_x = x[sort_idx]\n",
    "    sorted_y = y[sort_idx]\n",
    "    \n",
    "    # put everything in the right\n",
    "    rhs_cnt, rhs_sum, rhs_sum2 = num_samples, sum(y), sum(y.^2)\n",
    "    lhs_cnt, lhs_sum, lhs_sum2 = 0, 0.0, 0.0\n",
    "    \n",
    "    for i in 1:(num_samples - 1)\n",
    "        x_i, y_i = sorted_x[i], sorted_y[i]\n",
    "        lhs_cnt += 1\n",
    "        rhs_cnt -= 1\n",
    "        lhs_sum += y_i\n",
    "        rhs_sum -= y_i\n",
    "        lhs_sum2 += y_i^2\n",
    "        rhs_sum2 -= y_i^2\n",
    "    \n",
    "        lhs_std = std_agg(lhs_cnt, lhs_sum, lhs_sum2)\n",
    "        rhs_std = std_agg(rhs_cnt, rhs_sum, rhs_sum2)\n",
    "    \n",
    "        curr_score = lhs_std*lhs_cnt + rhs_std*rhs_cnt\n",
    "        \n",
    "        if curr_score < best_score\n",
    "            split_col = col\n",
    "            best_score = curr_score\n",
    "            best_split = x_i\n",
    "            lhs_idx = findall(<=(best_split), x)\n",
    "            rhs_idx = findall(>(best_split), x)\n",
    "        end\n",
    "    end\n",
    "end\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1, 4167.223505074033, 6.65492)"
      ]
     },
     "execution_count": 110,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "split_col, best_score, best_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/svg+xml": [
       "<?xml version=\"1.0\" encoding=\"utf-8\"?>\n",
       "<svg xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\" width=\"600\" height=\"400\" viewBox=\"0 0 2400 1600\">\n",
       "<defs>\n",
       "  <clipPath id=\"clip110\">\n",
       "    <rect x=\"0\" y=\"0\" width=\"2400\" height=\"1600\"/>\n",
       "  </clipPath>\n",
       "</defs>\n",
       "<path clip-path=\"url(#clip110)\" d=\"\n",
       "M0 1600 L2400 1600 L2400 0 L0 0  Z\n",
       "  \" fill=\"#ffffff\" fill-rule=\"evenodd\" fill-opacity=\"1\"/>\n",
       "<defs>\n",
       "  <clipPath id=\"clip111\">\n",
       "    <rect x=\"480\" y=\"0\" width=\"1681\" height=\"1600\"/>\n",
       "  </clipPath>\n",
       "</defs>\n",
       "<path clip-path=\"url(#clip110)\" d=\"\n",
       "M161.598 1486.45 L2352.76 1486.45 L2352.76 47.2441 L161.598 47.2441  Z\n",
       "  \" fill=\"#ffffff\" fill-rule=\"evenodd\" fill-opacity=\"1\"/>\n",
       "<defs>\n",
       "  <clipPath id=\"clip112\">\n",
       "    <rect x=\"161\" y=\"47\" width=\"2192\" height=\"1440\"/>\n",
       "  </clipPath>\n",
       "</defs>\n",
       "<polyline clip-path=\"url(#clip112)\" style=\"stroke:#000000; stroke-width:2; stroke-opacity:0.1; fill:none\" points=\"\n",
       "  219.519,1486.45 219.519,47.2441 \n",
       "  \"/>\n",
       "<polyline clip-path=\"url(#clip112)\" style=\"stroke:#000000; stroke-width:2; stroke-opacity:0.1; fill:none\" points=\"\n",
       "  628.851,1486.45 628.851,47.2441 \n",
       "  \"/>\n",
       "<polyline clip-path=\"url(#clip112)\" style=\"stroke:#000000; stroke-width:2; stroke-opacity:0.1; fill:none\" points=\"\n",
       "  1038.18,1486.45 1038.18,47.2441 \n",
       "  \"/>\n",
       "<polyline clip-path=\"url(#clip112)\" style=\"stroke:#000000; stroke-width:2; stroke-opacity:0.1; fill:none\" points=\"\n",
       "  1447.52,1486.45 1447.52,47.2441 \n",
       "  \"/>\n",
       "<polyline clip-path=\"url(#clip112)\" style=\"stroke:#000000; stroke-width:2; stroke-opacity:0.1; fill:none\" points=\"\n",
       "  1856.85,1486.45 1856.85,47.2441 \n",
       "  \"/>\n",
       "<polyline clip-path=\"url(#clip112)\" style=\"stroke:#000000; stroke-width:2; stroke-opacity:0.1; fill:none\" points=\"\n",
       "  2266.18,1486.45 2266.18,47.2441 \n",
       "  \"/>\n",
       "<polyline clip-path=\"url(#clip112)\" style=\"stroke:#000000; stroke-width:2; stroke-opacity:0.1; fill:none\" points=\"\n",
       "  161.598,1445.72 2352.76,1445.72 \n",
       "  \"/>\n",
       "<polyline clip-path=\"url(#clip112)\" style=\"stroke:#000000; stroke-width:2; stroke-opacity:0.1; fill:none\" points=\"\n",
       "  161.598,1106.28 2352.76,1106.28 \n",
       "  \"/>\n",
       "<polyline clip-path=\"url(#clip112)\" style=\"stroke:#000000; stroke-width:2; stroke-opacity:0.1; fill:none\" points=\"\n",
       "  161.598,766.846 2352.76,766.846 \n",
       "  \"/>\n",
       "<polyline clip-path=\"url(#clip112)\" style=\"stroke:#000000; stroke-width:2; stroke-opacity:0.1; fill:none\" points=\"\n",
       "  161.598,427.411 2352.76,427.411 \n",
       "  \"/>\n",
       "<polyline clip-path=\"url(#clip112)\" style=\"stroke:#000000; stroke-width:2; stroke-opacity:0.1; fill:none\" points=\"\n",
       "  161.598,87.9763 2352.76,87.9763 \n",
       "  \"/>\n",
       "<polyline clip-path=\"url(#clip110)\" style=\"stroke:#000000; stroke-width:4; stroke-opacity:1; fill:none\" points=\"\n",
       "  161.598,1486.45 2352.76,1486.45 \n",
       "  \"/>\n",
       "<polyline clip-path=\"url(#clip110)\" style=\"stroke:#000000; stroke-width:4; stroke-opacity:1; fill:none\" points=\"\n",
       "  161.598,1486.45 161.598,47.2441 \n",
       "  \"/>\n",
       "<polyline clip-path=\"url(#clip110)\" style=\"stroke:#000000; stroke-width:4; stroke-opacity:1; fill:none\" points=\"\n",
       "  219.519,1486.45 219.519,1469.18 \n",
       "  \"/>\n",
       "<polyline clip-path=\"url(#clip110)\" style=\"stroke:#000000; stroke-width:4; stroke-opacity:1; fill:none\" points=\"\n",
       "  628.851,1486.45 628.851,1469.18 \n",
       "  \"/>\n",
       "<polyline clip-path=\"url(#clip110)\" style=\"stroke:#000000; stroke-width:4; stroke-opacity:1; fill:none\" points=\"\n",
       "  1038.18,1486.45 1038.18,1469.18 \n",
       "  \"/>\n",
       "<polyline clip-path=\"url(#clip110)\" style=\"stroke:#000000; stroke-width:4; stroke-opacity:1; fill:none\" points=\"\n",
       "  1447.52,1486.45 1447.52,1469.18 \n",
       "  \"/>\n",
       "<polyline clip-path=\"url(#clip110)\" style=\"stroke:#000000; stroke-width:4; stroke-opacity:1; fill:none\" points=\"\n",
       "  1856.85,1486.45 1856.85,1469.18 \n",
       "  \"/>\n",
       "<polyline clip-path=\"url(#clip110)\" style=\"stroke:#000000; stroke-width:4; stroke-opacity:1; fill:none\" points=\"\n",
       "  2266.18,1486.45 2266.18,1469.18 \n",
       "  \"/>\n",
       "<polyline clip-path=\"url(#clip110)\" style=\"stroke:#000000; stroke-width:4; stroke-opacity:1; fill:none\" points=\"\n",
       "  161.598,1445.72 187.892,1445.72 \n",
       "  \"/>\n",
       "<polyline clip-path=\"url(#clip110)\" style=\"stroke:#000000; stroke-width:4; stroke-opacity:1; fill:none\" points=\"\n",
       "  161.598,1106.28 187.892,1106.28 \n",
       "  \"/>\n",
       "<polyline clip-path=\"url(#clip110)\" style=\"stroke:#000000; stroke-width:4; stroke-opacity:1; fill:none\" points=\"\n",
       "  161.598,766.846 187.892,766.846 \n",
       "  \"/>\n",
       "<polyline clip-path=\"url(#clip110)\" style=\"stroke:#000000; stroke-width:4; stroke-opacity:1; fill:none\" points=\"\n",
       "  161.598,427.411 187.892,427.411 \n",
       "  \"/>\n",
       "<polyline clip-path=\"url(#clip110)\" style=\"stroke:#000000; stroke-width:4; stroke-opacity:1; fill:none\" points=\"\n",
       "  161.598,87.9763 187.892,87.9763 \n",
       "  \"/>\n",
       "<path clip-path=\"url(#clip110)\" d=\"M 0 0 M219.519 1508.44 Q215.908 1508.44 214.079 1512 Q212.273 1515.55 212.273 1522.67 Q212.273 1529.78 214.079 1533.35 Q215.908 1536.89 219.519 1536.89 Q223.153 1536.89 224.959 1533.35 Q226.787 1529.78 226.787 1522.67 Q226.787 1515.55 224.959 1512 Q223.153 1508.44 219.519 1508.44 M219.519 1504.73 Q225.329 1504.73 228.384 1509.34 Q231.463 1513.92 231.463 1522.67 Q231.463 1531.4 228.384 1536.01 Q225.329 1540.59 219.519 1540.59 Q213.709 1540.59 210.63 1536.01 Q207.574 1531.4 207.574 1522.67 Q207.574 1513.92 210.63 1509.34 Q213.709 1504.73 219.519 1504.73 Z\" fill=\"#000000\" fill-rule=\"evenodd\" fill-opacity=\"1\" /><path clip-path=\"url(#clip110)\" d=\"M 0 0 M592.22 1535.98 L599.859 1535.98 L599.859 1509.62 L591.548 1511.29 L591.548 1507.03 L599.812 1505.36 L604.488 1505.36 L604.488 1535.98 L612.127 1535.98 L612.127 1539.92 L592.22 1539.92 L592.22 1535.98 Z\" fill=\"#000000\" fill-rule=\"evenodd\" fill-opacity=\"1\" /><path clip-path=\"url(#clip110)\" d=\"M 0 0 M627.196 1508.44 Q623.585 1508.44 621.757 1512 Q619.951 1515.55 619.951 1522.67 Q619.951 1529.78 621.757 1533.35 Q623.585 1536.89 627.196 1536.89 Q630.831 1536.89 632.636 1533.35 Q634.465 1529.78 634.465 1522.67 Q634.465 1515.55 632.636 1512 Q630.831 1508.44 627.196 1508.44 M627.196 1504.73 Q633.007 1504.73 636.062 1509.34 Q639.141 1513.92 639.141 1522.67 Q639.141 1531.4 636.062 1536.01 Q633.007 1540.59 627.196 1540.59 Q621.386 1540.59 618.308 1536.01 Q615.252 1531.4 615.252 1522.67 Q615.252 1513.92 618.308 1509.34 Q621.386 1504.73 627.196 1504.73 Z\" fill=\"#000000\" fill-rule=\"evenodd\" fill-opacity=\"1\" /><path clip-path=\"url(#clip110)\" d=\"M 0 0 M654.21 1508.44 Q650.599 1508.44 648.77 1512 Q646.965 1515.55 646.965 1522.67 Q646.965 1529.78 648.77 1533.35 Q650.599 1536.89 654.21 1536.89 Q657.844 1536.89 659.65 1533.35 Q661.479 1529.78 661.479 1522.67 Q661.479 1515.55 659.65 1512 Q657.844 1508.44 654.21 1508.44 M654.21 1504.73 Q660.02 1504.73 663.076 1509.34 Q666.154 1513.92 666.154 1522.67 Q666.154 1531.4 663.076 1536.01 Q660.02 1540.59 654.21 1540.59 Q648.4 1540.59 645.321 1536.01 Q642.266 1531.4 642.266 1522.67 Q642.266 1513.92 645.321 1509.34 Q648.4 1504.73 654.21 1504.73 Z\" fill=\"#000000\" fill-rule=\"evenodd\" fill-opacity=\"1\" /><path clip-path=\"url(#clip110)\" d=\"M 0 0 M1005.82 1535.98 L1022.14 1535.98 L1022.14 1539.92 L1000.2 1539.92 L1000.2 1535.98 Q1002.86 1533.23 1007.44 1528.6 Q1012.05 1523.95 1013.23 1522.61 Q1015.48 1520.08 1016.36 1518.35 Q1017.26 1516.59 1017.26 1514.9 Q1017.26 1512.14 1015.31 1510.41 Q1013.39 1508.67 1010.29 1508.67 Q1008.09 1508.67 1005.64 1509.43 Q1003.21 1510.2 1000.43 1511.75 L1000.43 1507.03 Q1003.25 1505.89 1005.71 1505.31 Q1008.16 1504.73 1010.2 1504.73 Q1015.57 1504.73 1018.76 1507.42 Q1021.96 1510.11 1021.96 1514.6 Q1021.96 1516.73 1021.15 1518.65 Q1020.36 1520.54 1018.25 1523.14 Q1017.67 1523.81 1014.57 1527.03 Q1011.47 1530.22 1005.82 1535.98 Z\" fill=\"#000000\" fill-rule=\"evenodd\" fill-opacity=\"1\" /><path clip-path=\"url(#clip110)\" d=\"M 0 0 M1037.21 1508.44 Q1033.6 1508.44 1031.77 1512 Q1029.97 1515.55 1029.97 1522.67 Q1029.97 1529.78 1031.77 1533.35 Q1033.6 1536.89 1037.21 1536.89 Q1040.85 1536.89 1042.65 1533.35 Q1044.48 1529.78 1044.48 1522.67 Q1044.48 1515.55 1042.65 1512 Q1040.85 1508.44 1037.21 1508.44 M1037.21 1504.73 Q1043.02 1504.73 1046.08 1509.34 Q1049.16 1513.92 1049.16 1522.67 Q1049.16 1531.4 1046.08 1536.01 Q1043.02 1540.59 1037.21 1540.59 Q1031.4 1540.59 1028.32 1536.01 Q1025.27 1531.4 1025.27 1522.67 Q1025.27 1513.92 1028.32 1509.34 Q1031.4 1504.73 1037.21 1504.73 Z\" fill=\"#000000\" fill-rule=\"evenodd\" fill-opacity=\"1\" /><path clip-path=\"url(#clip110)\" d=\"M 0 0 M1064.23 1508.44 Q1060.61 1508.44 1058.79 1512 Q1056.98 1515.55 1056.98 1522.67 Q1056.98 1529.78 1058.79 1533.35 Q1060.61 1536.89 1064.23 1536.89 Q1067.86 1536.89 1069.67 1533.35 Q1071.49 1529.78 1071.49 1522.67 Q1071.49 1515.55 1069.67 1512 Q1067.86 1508.44 1064.23 1508.44 M1064.23 1504.73 Q1070.04 1504.73 1073.09 1509.34 Q1076.17 1513.92 1076.17 1522.67 Q1076.17 1531.4 1073.09 1536.01 Q1070.04 1540.59 1064.23 1540.59 Q1058.42 1540.59 1055.34 1536.01 Q1052.28 1531.4 1052.28 1522.67 Q1052.28 1513.92 1055.34 1509.34 Q1058.42 1504.73 1064.23 1504.73 Z\" fill=\"#000000\" fill-rule=\"evenodd\" fill-opacity=\"1\" /><path clip-path=\"url(#clip110)\" d=\"M 0 0 M1424.75 1521.29 Q1428.11 1522 1429.98 1524.27 Q1431.88 1526.54 1431.88 1529.87 Q1431.88 1534.99 1428.36 1537.79 Q1424.84 1540.59 1418.36 1540.59 Q1416.19 1540.59 1413.87 1540.15 Q1411.58 1539.73 1409.13 1538.88 L1409.13 1534.36 Q1411.07 1535.5 1413.38 1536.08 Q1415.7 1536.66 1418.22 1536.66 Q1422.62 1536.66 1424.91 1534.92 Q1427.23 1533.18 1427.23 1529.87 Q1427.23 1526.82 1425.07 1525.11 Q1422.95 1523.37 1419.13 1523.37 L1415.1 1523.37 L1415.1 1519.53 L1419.31 1519.53 Q1422.76 1519.53 1424.59 1518.16 Q1426.42 1516.77 1426.42 1514.18 Q1426.42 1511.52 1424.52 1510.11 Q1422.64 1508.67 1419.13 1508.67 Q1417.2 1508.67 1415.01 1509.09 Q1412.81 1509.5 1410.17 1510.38 L1410.17 1506.22 Q1412.83 1505.48 1415.14 1505.11 Q1417.48 1504.73 1419.54 1504.73 Q1424.87 1504.73 1427.97 1507.17 Q1431.07 1509.57 1431.07 1513.69 Q1431.07 1516.56 1429.43 1518.55 Q1427.78 1520.52 1424.75 1521.29 Z\" fill=\"#000000\" fill-rule=\"evenodd\" fill-opacity=\"1\" /><path clip-path=\"url(#clip110)\" d=\"M 0 0 M1446.95 1508.44 Q1443.34 1508.44 1441.51 1512 Q1439.7 1515.55 1439.7 1522.67 Q1439.7 1529.78 1441.51 1533.35 Q1443.34 1536.89 1446.95 1536.89 Q1450.58 1536.89 1452.39 1533.35 Q1454.22 1529.78 1454.22 1522.67 Q1454.22 1515.55 1452.39 1512 Q1450.58 1508.44 1446.95 1508.44 M1446.95 1504.73 Q1452.76 1504.73 1455.82 1509.34 Q1458.89 1513.92 1458.89 1522.67 Q1458.89 1531.4 1455.82 1536.01 Q1452.76 1540.59 1446.95 1540.59 Q1441.14 1540.59 1438.06 1536.01 Q1435.01 1531.4 1435.01 1522.67 Q1435.01 1513.92 1438.06 1509.34 Q1441.14 1504.73 1446.95 1504.73 Z\" fill=\"#000000\" fill-rule=\"evenodd\" fill-opacity=\"1\" /><path clip-path=\"url(#clip110)\" d=\"M 0 0 M1473.96 1508.44 Q1470.35 1508.44 1468.52 1512 Q1466.72 1515.55 1466.72 1522.67 Q1466.72 1529.78 1468.52 1533.35 Q1470.35 1536.89 1473.96 1536.89 Q1477.6 1536.89 1479.4 1533.35 Q1481.23 1529.78 1481.23 1522.67 Q1481.23 1515.55 1479.4 1512 Q1477.6 1508.44 1473.96 1508.44 M1473.96 1504.73 Q1479.77 1504.73 1482.83 1509.34 Q1485.91 1513.92 1485.91 1522.67 Q1485.91 1531.4 1482.83 1536.01 Q1479.77 1540.59 1473.96 1540.59 Q1468.15 1540.59 1465.07 1536.01 Q1462.02 1531.4 1462.02 1522.67 Q1462.02 1513.92 1465.07 1509.34 Q1468.15 1504.73 1473.96 1504.73 Z\" fill=\"#000000\" fill-rule=\"evenodd\" fill-opacity=\"1\" /><path clip-path=\"url(#clip110)\" d=\"M 0 0 M1832.84 1509.43 L1821.04 1527.88 L1832.84 1527.88 L1832.84 1509.43 M1831.62 1505.36 L1837.5 1505.36 L1837.5 1527.88 L1842.43 1527.88 L1842.43 1531.77 L1837.5 1531.77 L1837.5 1539.92 L1832.84 1539.92 L1832.84 1531.77 L1817.24 1531.77 L1817.24 1527.26 L1831.62 1505.36 Z\" fill=\"#000000\" fill-rule=\"evenodd\" fill-opacity=\"1\" /><path clip-path=\"url(#clip110)\" d=\"M 0 0 M1857.5 1508.44 Q1853.89 1508.44 1852.06 1512 Q1850.25 1515.55 1850.25 1522.67 Q1850.25 1529.78 1852.06 1533.35 Q1853.89 1536.89 1857.5 1536.89 Q1861.13 1536.89 1862.94 1533.35 Q1864.77 1529.78 1864.77 1522.67 Q1864.77 1515.55 1862.94 1512 Q1861.13 1508.44 1857.5 1508.44 M1857.5 1504.73 Q1863.31 1504.73 1866.36 1509.34 Q1869.44 1513.92 1869.44 1522.67 Q1869.44 1531.4 1866.36 1536.01 Q1863.31 1540.59 1857.5 1540.59 Q1851.69 1540.59 1848.61 1536.01 Q1845.55 1531.4 1845.55 1522.67 Q1845.55 1513.92 1848.61 1509.34 Q1851.69 1504.73 1857.5 1504.73 Z\" fill=\"#000000\" fill-rule=\"evenodd\" fill-opacity=\"1\" /><path clip-path=\"url(#clip110)\" d=\"M 0 0 M1884.51 1508.44 Q1880.9 1508.44 1879.07 1512 Q1877.27 1515.55 1877.27 1522.67 Q1877.27 1529.78 1879.07 1533.35 Q1880.9 1536.89 1884.51 1536.89 Q1888.15 1536.89 1889.95 1533.35 Q1891.78 1529.78 1891.78 1522.67 Q1891.78 1515.55 1889.95 1512 Q1888.15 1508.44 1884.51 1508.44 M1884.51 1504.73 Q1890.32 1504.73 1893.38 1509.34 Q1896.46 1513.92 1896.46 1522.67 Q1896.46 1531.4 1893.38 1536.01 Q1890.32 1540.59 1884.51 1540.59 Q1878.7 1540.59 1875.62 1536.01 Q1872.57 1531.4 1872.57 1522.67 Q1872.57 1513.92 1875.62 1509.34 Q1878.7 1504.73 1884.51 1504.73 Z\" fill=\"#000000\" fill-rule=\"evenodd\" fill-opacity=\"1\" /><path clip-path=\"url(#clip110)\" d=\"M 0 0 M2229.45 1505.36 L2247.8 1505.36 L2247.8 1509.3 L2233.73 1509.3 L2233.73 1517.77 Q2234.75 1517.42 2235.77 1517.26 Q2236.78 1517.07 2237.8 1517.07 Q2243.59 1517.07 2246.97 1520.24 Q2250.35 1523.42 2250.35 1528.83 Q2250.35 1534.41 2246.88 1537.51 Q2243.4 1540.59 2237.08 1540.59 Q2234.91 1540.59 2232.64 1540.22 Q2230.4 1539.85 2227.99 1539.11 L2227.99 1534.41 Q2230.07 1535.54 2232.29 1536.1 Q2234.52 1536.66 2236.99 1536.66 Q2241 1536.66 2243.33 1534.55 Q2245.67 1532.44 2245.67 1528.83 Q2245.67 1525.22 2243.33 1523.11 Q2241 1521.01 2236.99 1521.01 Q2235.12 1521.01 2233.24 1521.42 Q2231.39 1521.84 2229.45 1522.72 L2229.45 1505.36 Z\" fill=\"#000000\" fill-rule=\"evenodd\" fill-opacity=\"1\" /><path clip-path=\"url(#clip110)\" d=\"M 0 0 M2265.42 1508.44 Q2261.81 1508.44 2259.98 1512 Q2258.17 1515.55 2258.17 1522.67 Q2258.17 1529.78 2259.98 1533.35 Q2261.81 1536.89 2265.42 1536.89 Q2269.05 1536.89 2270.86 1533.35 Q2272.69 1529.78 2272.69 1522.67 Q2272.69 1515.55 2270.86 1512 Q2269.05 1508.44 2265.42 1508.44 M2265.42 1504.73 Q2271.23 1504.73 2274.28 1509.34 Q2277.36 1513.92 2277.36 1522.67 Q2277.36 1531.4 2274.28 1536.01 Q2271.23 1540.59 2265.42 1540.59 Q2259.61 1540.59 2256.53 1536.01 Q2253.47 1531.4 2253.47 1522.67 Q2253.47 1513.92 2256.53 1509.34 Q2259.61 1504.73 2265.42 1504.73 Z\" fill=\"#000000\" fill-rule=\"evenodd\" fill-opacity=\"1\" /><path clip-path=\"url(#clip110)\" d=\"M 0 0 M2292.43 1508.44 Q2288.82 1508.44 2286.99 1512 Q2285.19 1515.55 2285.19 1522.67 Q2285.19 1529.78 2286.99 1533.35 Q2288.82 1536.89 2292.43 1536.89 Q2296.07 1536.89 2297.87 1533.35 Q2299.7 1529.78 2299.7 1522.67 Q2299.7 1515.55 2297.87 1512 Q2296.07 1508.44 2292.43 1508.44 M2292.43 1504.73 Q2298.24 1504.73 2301.3 1509.34 Q2304.38 1513.92 2304.38 1522.67 Q2304.38 1531.4 2301.3 1536.01 Q2298.24 1540.59 2292.43 1540.59 Q2286.62 1540.59 2283.54 1536.01 Q2280.49 1531.4 2280.49 1522.67 Q2280.49 1513.92 2283.54 1509.34 Q2286.62 1504.73 2292.43 1504.73 Z\" fill=\"#000000\" fill-rule=\"evenodd\" fill-opacity=\"1\" /><path clip-path=\"url(#clip110)\" d=\"M 0 0 M125.654 1431.51 Q122.043 1431.51 120.214 1435.08 Q118.408 1438.62 118.408 1445.75 Q118.408 1452.86 120.214 1456.42 Q122.043 1459.96 125.654 1459.96 Q129.288 1459.96 131.094 1456.42 Q132.922 1452.86 132.922 1445.75 Q132.922 1438.62 131.094 1435.08 Q129.288 1431.51 125.654 1431.51 M125.654 1427.81 Q131.464 1427.81 134.52 1432.42 Q137.598 1437 137.598 1445.75 Q137.598 1454.48 134.52 1459.08 Q131.464 1463.67 125.654 1463.67 Q119.844 1463.67 116.765 1459.08 Q113.709 1454.48 113.709 1445.75 Q113.709 1437 116.765 1432.42 Q119.844 1427.81 125.654 1427.81 Z\" fill=\"#000000\" fill-rule=\"evenodd\" fill-opacity=\"1\" /><path clip-path=\"url(#clip110)\" d=\"M 0 0 M95.2605 1119.63 L111.58 1119.63 L111.58 1123.56 L89.6355 1123.56 L89.6355 1119.63 Q92.2975 1116.87 96.8808 1112.24 Q101.487 1107.59 102.668 1106.25 Q104.913 1103.72 105.793 1101.99 Q106.696 1100.23 106.696 1098.54 Q106.696 1095.78 104.751 1094.05 Q102.83 1092.31 99.728 1092.31 Q97.529 1092.31 95.0753 1093.07 Q92.6447 1093.84 89.867 1095.39 L89.867 1090.67 Q92.691 1089.53 95.1447 1088.95 Q97.5984 1088.38 99.6354 1088.38 Q105.006 1088.38 108.2 1091.06 Q111.395 1093.75 111.395 1098.24 Q111.395 1100.37 110.584 1102.29 Q109.797 1104.19 107.691 1106.78 Q107.112 1107.45 104.01 1110.67 Q100.909 1113.86 95.2605 1119.63 Z\" fill=\"#000000\" fill-rule=\"evenodd\" fill-opacity=\"1\" /><path clip-path=\"url(#clip110)\" d=\"M 0 0 M116.696 1089 L135.052 1089 L135.052 1092.94 L120.978 1092.94 L120.978 1101.41 Q121.996 1101.06 123.015 1100.9 Q124.033 1100.71 125.052 1100.71 Q130.839 1100.71 134.219 1103.89 Q137.598 1107.06 137.598 1112.47 Q137.598 1118.05 134.126 1121.15 Q130.654 1124.23 124.334 1124.23 Q122.158 1124.23 119.89 1123.86 Q117.645 1123.49 115.237 1122.75 L115.237 1118.05 Q117.321 1119.19 119.543 1119.74 Q121.765 1120.3 124.242 1120.3 Q128.246 1120.3 130.584 1118.19 Q132.922 1116.08 132.922 1112.47 Q132.922 1108.86 130.584 1106.76 Q128.246 1104.65 124.242 1104.65 Q122.367 1104.65 120.492 1105.07 Q118.64 1105.48 116.696 1106.36 L116.696 1089 Z\" fill=\"#000000\" fill-rule=\"evenodd\" fill-opacity=\"1\" /><path clip-path=\"url(#clip110)\" d=\"M 0 0 M89.6818 749.566 L108.038 749.566 L108.038 753.501 L93.9642 753.501 L93.9642 761.973 Q94.9827 761.626 96.0012 761.464 Q97.0197 761.279 98.0382 761.279 Q103.825 761.279 107.205 764.45 Q110.584 767.621 110.584 773.038 Q110.584 778.617 107.112 781.719 Q103.64 784.797 97.3206 784.797 Q95.1447 784.797 92.8762 784.427 Q90.6309 784.057 88.2235 783.316 L88.2235 778.617 Q90.3068 779.751 92.529 780.307 Q94.7512 780.862 97.2281 780.862 Q101.233 780.862 103.571 778.756 Q105.909 776.649 105.909 773.038 Q105.909 769.427 103.571 767.321 Q101.233 765.214 97.2281 765.214 Q95.3531 765.214 93.4781 765.631 Q91.6262 766.047 89.6818 766.927 L89.6818 749.566 Z\" fill=\"#000000\" fill-rule=\"evenodd\" fill-opacity=\"1\" /><path clip-path=\"url(#clip110)\" d=\"M 0 0 M125.654 752.645 Q122.043 752.645 120.214 756.209 Q118.408 759.751 118.408 766.881 Q118.408 773.987 120.214 777.552 Q122.043 781.094 125.654 781.094 Q129.288 781.094 131.094 777.552 Q132.922 773.987 132.922 766.881 Q132.922 759.751 131.094 756.209 Q129.288 752.645 125.654 752.645 M125.654 748.941 Q131.464 748.941 134.52 753.547 Q137.598 758.131 137.598 766.881 Q137.598 775.608 134.52 780.214 Q131.464 784.797 125.654 784.797 Q119.844 784.797 116.765 780.214 Q113.709 775.608 113.709 766.881 Q113.709 758.131 116.765 753.547 Q119.844 748.941 125.654 748.941 Z\" fill=\"#000000\" fill-rule=\"evenodd\" fill-opacity=\"1\" /><path clip-path=\"url(#clip110)\" d=\"M 0 0 M89.3577 410.131 L111.58 410.131 L111.58 412.122 L99.0336 444.691 L94.1494 444.691 L105.955 414.066 L89.3577 414.066 L89.3577 410.131 Z\" fill=\"#000000\" fill-rule=\"evenodd\" fill-opacity=\"1\" /><path clip-path=\"url(#clip110)\" d=\"M 0 0 M116.696 410.131 L135.052 410.131 L135.052 414.066 L120.978 414.066 L120.978 422.538 Q121.996 422.191 123.015 422.029 Q124.033 421.844 125.052 421.844 Q130.839 421.844 134.219 425.015 Q137.598 428.187 137.598 433.603 Q137.598 439.182 134.126 442.284 Q130.654 445.362 124.334 445.362 Q122.158 445.362 119.89 444.992 Q117.645 444.622 115.237 443.881 L115.237 439.182 Q117.321 440.316 119.543 440.872 Q121.765 441.427 124.242 441.427 Q128.246 441.427 130.584 439.321 Q132.922 437.214 132.922 433.603 Q132.922 429.992 130.584 427.886 Q128.246 425.779 124.242 425.779 Q122.367 425.779 120.492 426.196 Q118.64 426.613 116.696 427.492 L116.696 410.131 Z\" fill=\"#000000\" fill-rule=\"evenodd\" fill-opacity=\"1\" /><path clip-path=\"url(#clip110)\" d=\"M 0 0 M63.6634 101.321 L71.3023 101.321 L71.3023 74.9555 L62.9921 76.6222 L62.9921 72.3629 L71.256 70.6963 L75.9319 70.6963 L75.9319 101.321 L83.5707 101.321 L83.5707 105.256 L63.6634 105.256 L63.6634 101.321 Z\" fill=\"#000000\" fill-rule=\"evenodd\" fill-opacity=\"1\" /><path clip-path=\"url(#clip110)\" d=\"M 0 0 M98.6401 73.775 Q95.029 73.775 93.2003 77.3398 Q91.3948 80.8814 91.3948 88.011 Q91.3948 95.1174 93.2003 98.6822 Q95.029 102.224 98.6401 102.224 Q102.274 102.224 104.08 98.6822 Q105.909 95.1174 105.909 88.011 Q105.909 80.8814 104.08 77.3398 Q102.274 73.775 98.6401 73.775 M98.6401 70.0713 Q104.45 70.0713 107.506 74.6777 Q110.584 79.261 110.584 88.011 Q110.584 96.7378 107.506 101.344 Q104.45 105.928 98.6401 105.928 Q92.8299 105.928 89.7512 101.344 Q86.6957 96.7378 86.6957 88.011 Q86.6957 79.261 89.7512 74.6777 Q92.8299 70.0713 98.6401 70.0713 Z\" fill=\"#000000\" fill-rule=\"evenodd\" fill-opacity=\"1\" /><path clip-path=\"url(#clip110)\" d=\"M 0 0 M125.654 73.775 Q122.043 73.775 120.214 77.3398 Q118.408 80.8814 118.408 88.011 Q118.408 95.1174 120.214 98.6822 Q122.043 102.224 125.654 102.224 Q129.288 102.224 131.094 98.6822 Q132.922 95.1174 132.922 88.011 Q132.922 80.8814 131.094 77.3398 Q129.288 73.775 125.654 73.775 M125.654 70.0713 Q131.464 70.0713 134.52 74.6777 Q137.598 79.261 137.598 88.011 Q137.598 96.7378 134.52 101.344 Q131.464 105.928 125.654 105.928 Q119.844 105.928 116.765 101.344 Q113.709 96.7378 113.709 88.011 Q113.709 79.261 116.765 74.6777 Q119.844 70.0713 125.654 70.0713 Z\" fill=\"#000000\" fill-rule=\"evenodd\" fill-opacity=\"1\" /><polyline clip-path=\"url(#clip112)\" style=\"stroke:#009af9; stroke-width:4; stroke-opacity:1; fill:none\" points=\"\n",
       "  223.612,1201.32 227.705,1445.72 231.799,1445.72 235.892,1445.72 239.985,1445.72 244.079,1445.72 248.172,1276 252.265,1276 256.359,1276 260.452,1276 \n",
       "  264.545,1276 268.639,1276 272.732,1276 276.825,1445.72 280.919,1445.72 285.012,1445.72 289.105,1445.72 293.199,1445.72 297.292,1445.72 301.385,1445.72 \n",
       "  305.479,1445.72 309.572,1445.72 313.665,1445.72 317.759,1445.72 321.852,1445.72 325.945,1445.72 330.039,1445.72 334.132,1445.72 338.225,1445.72 342.319,1445.72 \n",
       "  346.412,1445.72 350.505,1445.72 354.599,1445.72 358.692,1445.72 362.785,1445.72 366.879,1445.72 370.972,1445.72 375.065,1445.72 379.159,1445.72 383.252,427.411 \n",
       "  387.345,427.411 391.438,1445.72 395.532,1445.72 399.625,1445.72 403.718,1445.72 407.812,1445.72 411.905,1445.72 415.998,1445.72 420.092,1445.72 424.185,1445.72 \n",
       "  428.278,1160.59 432.372,1160.59 436.465,1160.59 440.558,1160.59 444.652,427.411 448.745,223.75 452.838,291.637 456.932,87.9763 461.025,1106.28 465.118,1106.28 \n",
       "  469.212,1106.28 473.305,1106.28 477.398,1106.28 481.492,1106.28 485.585,1208.11 489.678,359.524 493.772,359.524 497.865,1276 501.958,1276 506.052,1276 \n",
       "  510.145,1445.72 514.238,1445.72 518.332,1445.72 522.425,1445.72 526.518,1445.72 530.612,1445.72 534.705,1445.72 538.798,1445.72 542.892,1445.72 546.985,1445.72 \n",
       "  551.078,1106.28 555.172,1106.28 559.265,1106.28 563.358,1106.28 567.452,1445.72 571.545,1445.72 575.638,1445.72 579.732,1445.72 583.825,1445.72 587.918,1445.72 \n",
       "  592.011,1445.72 596.105,1445.72 600.198,1065.55 604.291,1065.55 608.385,1065.55 612.478,1445.72 616.571,1445.72 620.665,1445.72 624.758,1445.72 628.851,1445.72 \n",
       "  632.945,1445.72 637.038,1445.72 641.131,1445.72 645.225,1445.72 649.318,1445.72 653.411,1445.72 657.505,1445.72 661.598,1445.72 665.691,1445.72 669.785,1445.72 \n",
       "  673.878,1445.72 677.971,1445.72 682.065,1445.72 686.158,1445.72 690.251,1445.72 694.345,1445.72 698.438,1445.72 702.531,1445.72 706.625,1445.72 710.718,1445.72 \n",
       "  714.811,1445.72 718.905,1445.72 722.998,1445.72 727.091,1445.72 731.185,1445.72 735.278,1445.72 739.371,1445.72 743.465,1445.72 747.558,1445.72 751.651,1445.72 \n",
       "  755.745,1445.72 759.838,1445.72 763.931,1445.72 768.025,1445.72 772.118,1445.72 776.211,1445.72 780.305,1445.72 784.398,1445.72 788.491,1445.72 792.584,1445.72 \n",
       "  796.678,1445.72 800.771,1445.72 804.864,1445.72 808.958,1445.72 813.051,1445.72 817.144,1445.72 821.238,1445.72 825.331,1445.72 829.424,1445.72 833.518,1445.72 \n",
       "  837.611,1445.72 841.704,1445.72 845.798,1445.72 849.891,1445.72 853.984,1445.72 858.078,1445.72 862.171,1445.72 866.264,1445.72 870.358,1445.72 874.451,1445.72 \n",
       "  878.544,1445.72 882.638,1445.72 886.731,1445.72 890.824,1445.72 894.918,1445.72 899.011,1445.72 903.104,1445.72 907.198,1445.72 911.291,1445.72 915.384,1445.72 \n",
       "  919.478,1445.72 923.571,1445.72 927.664,1445.72 931.758,1445.72 935.851,1445.72 939.944,1445.72 944.038,1445.72 948.131,1445.72 952.224,1445.72 956.318,1445.72 \n",
       "  960.411,1445.72 964.504,1445.72 968.598,1445.72 972.691,1445.72 976.784,1445.72 980.878,1445.72 984.971,1445.72 989.064,834.733 993.157,834.733 997.251,834.733 \n",
       "  1001.34,834.733 1005.44,834.733 1009.53,834.733 1013.62,631.072 1017.72,631.072 1021.81,359.524 1025.9,359.524 1030,359.524 1034.09,359.524 1038.18,155.863 \n",
       "  1042.28,155.863 1046.37,325.581 1050.46,325.581 1054.56,155.863 1058.65,155.863 1062.74,1445.72 1066.84,1445.72 1070.93,1445.72 1075.02,1445.72 1079.12,1445.72 \n",
       "  1083.21,1445.72 1087.3,1445.72 1091.4,1445.72 1095.49,1445.72 1099.58,1445.72 1103.68,1445.72 1107.77,1445.72 1111.86,1445.72 1115.96,1445.72 1120.05,1445.72 \n",
       "  1124.14,1445.72 1128.24,1445.72 1132.33,1445.72 1136.42,1445.72 1140.52,1445.72 1144.61,1445.72 1148.7,1445.72 1152.8,1445.72 1156.89,1445.72 1160.98,1445.72 \n",
       "  1165.08,1445.72 1169.17,1445.72 1173.26,1445.72 1177.36,1445.72 1181.45,1445.72 1185.54,1445.72 1189.64,1445.72 1193.73,1445.72 1197.82,1038.39 1201.92,1038.39 \n",
       "  1206.01,1038.39 1210.1,1038.39 1214.2,1038.39 1218.29,1038.39 1222.38,1147.01 1226.48,1147.01 1230.57,1147.01 1234.66,1147.01 1238.76,1147.01 1242.85,1147.01 \n",
       "  1246.94,1147.01 1251.04,1147.01 1255.13,1147.01 1259.22,1147.01 1263.32,359.524 1267.41,359.524 1271.5,223.75 1275.6,1174.17 1279.69,1174.17 1283.78,1174.17 \n",
       "  1287.88,1174.17 1291.97,1174.17 1296.06,1174.17 1300.16,1174.17 1304.25,1174.17 1308.34,1174.17 1312.44,1174.17 1316.53,1174.17 1320.62,1174.17 1324.72,1174.17 \n",
       "  1328.81,1174.17 1332.9,1174.17 1337,1174.17 1341.09,1174.17 1345.18,902.62 1349.28,902.62 1353.37,902.62 1357.46,902.62 1361.56,902.62 1365.65,1174.17 \n",
       "  1369.74,1174.17 1373.84,1174.17 1377.93,1174.17 1382.02,223.75 1386.12,223.75 1390.21,698.959 1394.3,359.524 1398.4,732.902 1402.49,732.902 1406.58,732.902 \n",
       "  1410.68,359.524 1414.77,359.524 1418.86,359.524 1422.96,1445.72 1427.05,1445.72 1431.14,1445.72 1435.24,1445.72 1439.33,1445.72 1443.42,495.298 1447.52,495.298 \n",
       "  1451.61,495.298 1455.7,984.084 1459.8,984.084 1463.89,984.084 1467.98,997.662 1472.08,997.662 1476.17,997.662 1480.26,997.662 1484.36,1445.72 1488.45,1445.72 \n",
       "  1492.54,1445.72 1496.64,1445.72 1500.73,1445.72 1504.82,1445.72 1508.92,1445.72 1513.01,1445.72 1517.1,1445.72 1521.2,1445.72 1525.29,1445.72 1529.38,1445.72 \n",
       "  1533.48,1445.72 1537.57,1445.72 1541.66,1445.72 1545.76,1445.72 1549.85,1445.72 1553.94,1445.72 1558.04,1445.72 1562.13,1445.72 1566.22,1445.72 1570.32,1445.72 \n",
       "  1574.41,1445.72 1578.5,970.507 1582.6,970.507 1586.69,1445.72 1590.78,1445.72 1594.88,1445.72 1598.97,1445.72 1603.06,1445.72 1607.16,1445.72 1611.25,1445.72 \n",
       "  1615.34,1445.72 1619.44,970.507 1623.53,1445.72 1627.62,698.959 1631.72,698.959 1635.81,1445.72 1639.9,1445.72 1644,291.637 1648.09,359.524 1652.18,902.62 \n",
       "  1656.28,902.62 1660.37,631.072 1664.46,631.072 1668.56,223.75 1672.65,359.524 1676.74,359.524 1680.84,1445.72 1684.93,1445.72 1689.02,1445.72 1693.12,1445.72 \n",
       "  1697.21,1445.72 1701.3,1445.72 1705.4,1445.72 1709.49,1445.72 1713.58,1445.72 1717.68,1445.72 1721.77,1445.72 1725.86,1445.72 1729.96,1445.72 1734.05,1445.72 \n",
       "  1738.14,1445.72 1742.24,1445.72 1746.33,1445.72 1750.42,1445.72 1754.52,1445.72 1758.61,1445.72 1762.7,1445.72 1766.8,1445.72 1770.89,1445.72 1774.98,1445.72 \n",
       "  1779.08,1445.72 1783.17,1445.72 1787.26,1445.72 1791.36,1445.72 1795.45,1445.72 1799.54,1445.72 1803.64,1445.72 1807.73,1445.72 1811.82,1445.72 1815.92,1445.72 \n",
       "  1820.01,1445.72 1824.1,1445.72 1828.2,1445.72 1832.29,1445.72 1836.38,1445.72 1840.48,1445.72 1844.57,1445.72 1848.66,1445.72 1852.76,1445.72 1856.85,1445.72 \n",
       "  1860.94,1445.72 1865.04,1445.72 1869.13,1445.72 1873.22,1445.72 1877.32,1445.72 1881.41,1445.72 1885.5,1445.72 1889.6,1445.72 1893.69,1445.72 1897.78,1445.72 \n",
       "  1901.88,1445.72 1905.97,1445.72 1910.06,1445.72 1914.16,1445.72 1918.25,1445.72 1922.34,1445.72 1926.44,1445.72 1930.53,1445.72 1934.62,1445.72 1938.72,1445.72 \n",
       "  1942.81,1445.72 1946.9,1445.72 1951,1445.72 1955.09,1445.72 1959.18,1445.72 1963.28,1445.72 1967.37,1445.72 1971.46,1445.72 1975.56,1445.72 1979.65,1445.72 \n",
       "  1983.74,1445.72 1987.84,1445.72 1991.93,1445.72 1996.02,1445.72 2000.12,1445.72 2004.21,1445.72 2008.3,1445.72 2012.4,1445.72 2016.49,1445.72 2020.58,1445.72 \n",
       "  2024.68,1445.72 2028.77,1445.72 2032.86,1445.72 2036.96,1445.72 2041.05,1445.72 2045.14,1445.72 2049.24,1445.72 2053.33,1445.72 2057.42,1445.72 2061.52,1445.72 \n",
       "  2065.61,1445.72 2069.7,1445.72 2073.8,1445.72 2077.89,1445.72 2081.98,1445.72 2086.08,1445.72 2090.17,1445.72 2094.26,1445.72 2098.36,1445.72 2102.45,1445.72 \n",
       "  2106.54,1445.72 2110.64,1445.72 2114.73,1445.72 2118.82,1445.72 2122.92,1445.72 2127.01,1445.72 2131.1,1445.72 2135.2,1445.72 2139.29,1445.72 2143.38,1445.72 \n",
       "  2147.48,1445.72 2151.57,1445.72 2155.66,1445.72 2159.76,1445.72 2163.85,1445.72 2167.94,1445.72 2172.04,1445.72 2176.13,1445.72 2180.22,1445.72 2184.32,1445.72 \n",
       "  2188.41,1445.72 2192.5,1445.72 2196.6,1445.72 2200.69,1445.72 2204.78,1445.72 2208.88,1445.72 2212.97,1445.72 2217.06,1445.72 2221.16,1445.72 2225.25,1445.72 \n",
       "  2229.34,1445.72 2233.44,1445.72 2237.53,1445.72 2241.62,1445.72 2245.72,1445.72 2249.81,1445.72 2253.9,1445.72 2258,1445.72 2262.09,1445.72 2266.18,1445.72 \n",
       "  2270.28,1445.72 2274.37,1445.72 2278.46,1445.72 2282.56,1445.72 2286.65,1445.72 2290.74,1445.72 \n",
       "  \"/>\n",
       "<path clip-path=\"url(#clip110)\" d=\"\n",
       "M1989.74 251.724 L2280.76 251.724 L2280.76 130.764 L1989.74 130.764  Z\n",
       "  \" fill=\"#ffffff\" fill-rule=\"evenodd\" fill-opacity=\"1\"/>\n",
       "<polyline clip-path=\"url(#clip110)\" style=\"stroke:#000000; stroke-width:4; stroke-opacity:1; fill:none\" points=\"\n",
       "  1989.74,251.724 2280.76,251.724 2280.76,130.764 1989.74,130.764 1989.74,251.724 \n",
       "  \"/>\n",
       "<polyline clip-path=\"url(#clip110)\" style=\"stroke:#009af9; stroke-width:4; stroke-opacity:1; fill:none\" points=\"\n",
       "  2013.74,191.244 2157.74,191.244 \n",
       "  \"/>\n",
       "<path clip-path=\"url(#clip110)\" d=\"M 0 0 M2195.58 210.931 Q2193.77 215.561 2192.06 216.973 Q2190.35 218.385 2187.48 218.385 L2184.08 218.385 L2184.08 214.82 L2186.58 214.82 Q2188.33 214.82 2189.31 213.987 Q2190.28 213.154 2191.46 210.052 L2192.22 208.107 L2181.74 182.598 L2186.25 182.598 L2194.35 202.876 L2202.46 182.598 L2206.97 182.598 L2195.58 210.931 Z\" fill=\"#000000\" fill-rule=\"evenodd\" fill-opacity=\"1\" /><path clip-path=\"url(#clip110)\" d=\"M 0 0 M2212.85 204.589 L2220.49 204.589 L2220.49 178.223 L2212.18 179.89 L2212.18 175.631 L2220.44 173.964 L2225.12 173.964 L2225.12 204.589 L2232.76 204.589 L2232.76 208.524 L2212.85 208.524 L2212.85 204.589 Z\" fill=\"#000000\" fill-rule=\"evenodd\" fill-opacity=\"1\" /></svg>\n"
      ]
     },
     "execution_count": 112,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "plot(X[:, 2])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/latex": [
       "\\begin{verbatim}\n",
       "CSV.File(source; kwargs...) => CSV.File\n",
       "\\end{verbatim}\n",
       "Read a UTF-8 CSV input (a filename given as a String or FilePaths.jl type, or any other IO source), returning a \\texttt{CSV.File} object.\n",
       "\n",
       "Opens the file and uses passed arguments to detect the number of columns and column types, unless column types are provided manually via the \\texttt{types} keyword argument. Note that passing column types manually can increase performance and reduce the memory use for each column type provided (column types can be given as a \\texttt{Vector} for all columns, or specified per column via name or index in a \\texttt{Dict}). For text encodings other than UTF-8, see the \\href{https://github.com/JuliaStrings/StringEncodings.jl}{StringEncodings.jl} package for re-encoding a file or IO stream. The returned \\texttt{CSV.File} object supports the \\href{https://github.com/JuliaData/Tables.jl}{Tables.jl} interface and can iterate \\texttt{CSV.Row}s. \\texttt{CSV.Row} supports \\texttt{propertynames} and \\texttt{getproperty} to access individual row values. \\texttt{CSV.File} also supports entire column access like a \\texttt{DataFrame} via direct property access on the file object, like \\texttt{f = CSV.File(file); f.col1}. Note that duplicate column names will be detected and adjusted to ensure uniqueness (duplicate column name \\texttt{a} will become \\texttt{a\\_1}). For example, one could iterate over a csv file with column names \\texttt{a}, \\texttt{b}, and \\texttt{c} by doing:\n",
       "\n",
       "\\begin{verbatim}\n",
       "for row in CSV.File(file)\n",
       "    println(\"a=$(row.a), b=$(row.b), c=$(row.c)\")\n",
       "end\n",
       "\\end{verbatim}\n",
       "By supporting the Tables.jl interface, a \\texttt{CSV.File} can also be a table input to any other table sink function. Like:\n",
       "\n",
       "\\begin{verbatim}\n",
       "# materialize a csv file as a DataFrame, without copying columns from CSV.File; these columns are read-only\n",
       "df = CSV.File(file) |> DataFrame!\n",
       "\n",
       "# load a csv file directly into an sqlite database table\n",
       "db = SQLite.DB()\n",
       "tbl = CSV.File(file) |> SQLite.load!(db, \"sqlite_table\")\n",
       "\\end{verbatim}\n",
       "Supported keyword arguments include:\n",
       "\n",
       "\\begin{itemize}\n",
       "\\item File layout options:\n",
       "\n",
       "\\begin{itemize}\n",
       "\\item \\texttt{header=1}: the \\texttt{header} argument can be an \\texttt{Int}, indicating the row to parse for column names; or a \\texttt{Range}, indicating a span of rows to be concatenated together as column names; or an entire \\texttt{Vector\\{Symbol\\}} or \\texttt{Vector\\{String\\}} to use as column names; if a file doesn't have column names, either provide them as a \\texttt{Vector}, or set \\texttt{header=0} or \\texttt{header=false} and column names will be auto-generated (\\texttt{Column1}, \\texttt{Column2}, etc.)\n",
       "\n",
       "\n",
       "\\item \\texttt{normalizenames=false}: whether column names should be \"normalized\" into valid Julia identifier symbols; useful when iterating rows and accessing column values of a row via \\texttt{getproperty} (e.g. \\texttt{row.col1})\n",
       "\n",
       "\n",
       "\\item \\texttt{datarow}: an \\texttt{Int} argument to specify the row where the data starts in the csv file; by default, the next row after the \\texttt{header} row is used. If \\texttt{header=0}, then the 1st row is assumed to be the start of data\n",
       "\n",
       "\n",
       "\\item \\texttt{skipto::Int}: similar to \\texttt{datarow}, specifies the number of rows to skip before starting to read data\n",
       "\n",
       "\n",
       "\\item \\texttt{footerskip::Int}: number of rows at the end of a file to skip parsing\n",
       "\n",
       "\n",
       "\\item \\texttt{limit}: an \\texttt{Int} to indicate a limited number of rows to parse in a csv file; use in combination with \\texttt{skipto} to read a specific, contiguous chunk within a file\n",
       "\n",
       "\n",
       "\\item \\texttt{transpose::Bool}: read a csv file \"transposed\", i.e. each column is parsed as a row\n",
       "\n",
       "\n",
       "\\item \\texttt{comment}: rows that begin with this \\texttt{String} will be skipped while parsing\n",
       "\n",
       "\n",
       "\\item \\texttt{use\\_mmap::Bool=!Sys.iswindows()}: whether the file should be mmapped for reading, which in some cases can be faster\n",
       "\n",
       "\n",
       "\\item \\texttt{ignoreemptylines::Bool=false}: whether empty rows/lines in a file should be ignored (if \\texttt{false}, each column will be assigned \\texttt{missing} for that empty row)\n",
       "\n",
       "\n",
       "\\item \\texttt{threaded::Bool}: whether parsing should utilize multiple threads; by default threads are used on large enough files, but isn't allowed when \\texttt{transpose=true} or when \\texttt{limit} is used; only available in Julia 1.3+\n",
       "\n",
       "\n",
       "\\item \\texttt{select}: an \\texttt{AbstractVector} of \\texttt{Int}, \\texttt{Symbol}, \\texttt{String}, or \\texttt{Bool}, or a \"selector\" function of the form \\texttt{(i, name) -> keep::Bool}; only columns in the collection or for which the selector function returns \\texttt{true} will be parsed and accessible in the resulting \\texttt{CSV.File}. Invalid values in \\texttt{select} are ignored.\n",
       "\n",
       "\n",
       "\\item \\texttt{drop}: inverse of \\texttt{select}; an \\texttt{AbstractVector} of \\texttt{Int}, \\texttt{Symbol}, \\texttt{String}, or \\texttt{Bool}, or a \"drop\" function of the form \\texttt{(i, name) -> drop::Bool}; columns in the collection or for which the drop function returns \\texttt{true} will ignored in the resulting \\texttt{CSV.File}. Invalid values in \\texttt{drop} are ignored.\n",
       "\n",
       "\\end{itemize}\n",
       "\n",
       "\\item Parsing options:\n",
       "\n",
       "\\begin{itemize}\n",
       "\\item \\texttt{missingstrings}, \\texttt{missingstring}: either a \\texttt{String}, or \\texttt{Vector\\{String\\}} to use as sentinel values that will be parsed as \\texttt{missing}; by default, only an empty field (two consecutive delimiters) is considered \\texttt{missing}\n",
       "\n",
       "\n",
       "\\item \\texttt{delim=','}: a \\texttt{Char} or \\texttt{String} that indicates how columns are delimited in a file; if no argument is provided, parsing will try to detect the most consistent delimiter on the first 10 rows of the file\n",
       "\n",
       "\n",
       "\\item \\texttt{ignorerepeated::Bool=false}: whether repeated (consecutive) delimiters should be ignored while parsing; useful for fixed-width files with delimiter padding between cells\n",
       "\n",
       "\n",
       "\\item \\texttt{quotechar='\"'}, \\texttt{openquotechar}, \\texttt{closequotechar}: a \\texttt{Char} (or different start and end characters) that indicate a quoted field which may contain textual delimiters or newline characters\n",
       "\n",
       "\n",
       "\\item \\texttt{escapechar='\"'}: the \\texttt{Char} used to escape quote characters in a quoted field\n",
       "\n",
       "\n",
       "\\item \\texttt{dateformat::Union\\{String, Dates.DateFormat, Nothing\\}}: a date format string to indicate how Date/DateTime columns are formatted for the entire file\n",
       "\n",
       "\n",
       "\\item \\texttt{decimal='.'}: a \\texttt{Char} indicating how decimals are separated in floats, i.e. \\texttt{3.14} used '.', or \\texttt{3,14} uses a comma ','\n",
       "\n",
       "\n",
       "\\item \\texttt{truestrings}, \\texttt{falsestrings}: \\texttt{Vectors of Strings} that indicate how \\texttt{true} or \\texttt{false} values are represented; by default only \\texttt{true} and \\texttt{false} are treated as \\texttt{Bool}\n",
       "\n",
       "\\end{itemize}\n",
       "\n",
       "\\item Column Type Options:\n",
       "\n",
       "\\begin{itemize}\n",
       "\\item \\texttt{type}: a single type to use for parsing an entire file; i.e. all columns will be treated as the same type; useful for matrix-like data files\n",
       "\n",
       "\n",
       "\\item \\texttt{types}: a Vector or Dict of types to be used for column types; a Dict can map column index \\texttt{Int}, or name \\texttt{Symbol} or \\texttt{String} to type for a column, i.e. Dict(1=>Float64) will set the first column as a Float64, Dict(:column1=>Float64) will set the column named column1 to Float64 and, Dict(\"column1\"=>Float64) will set the column1 to Float64; if a \\texttt{Vector} if provided, it must match the \\# of columns provided or detected in \\texttt{header}\n",
       "\n",
       "\n",
       "\\item \\texttt{typemap::Dict\\{Type, Type\\}}: a mapping of a type that should be replaced in every instance with another type, i.e. \\texttt{Dict(Float64=>String)} would change every detected \\texttt{Float64} column to be parsed as \\texttt{String}\n",
       "\n",
       "\n",
       "\\item \\texttt{pool::Union\\{Bool, Float64\\}=0.1}: if \\texttt{true}, \\emph{all} columns detected as \\texttt{String} will be internally pooled; alternatively, the proportion of unique values below which \\texttt{String} columns should be pooled (by default 0.1, meaning that if the \\# of unique strings in a column is under 10\\%, it will be pooled)\n",
       "\n",
       "\n",
       "\\item \\texttt{categorical::Bool=false}: whether pooled columns should be copied as CategoricalArray instead of PooledArray; note that in \\texttt{CSV.read}, by default, columns are not copied, so pooled columns will have type \\texttt{CSV.Column\\{String, PooledString\\}}; to get \\texttt{CategoricalArray} columns, also pass \\texttt{copycols=true}\n",
       "\n",
       "\n",
       "\\item \\texttt{strict::Bool=false}: whether invalid values should throw a parsing error or be replaced with \\texttt{missing}\n",
       "\n",
       "\n",
       "\\item \\texttt{silencewarnings::Bool=false}: if \\texttt{strict=false}, whether invalid value warnings should be silenced\n",
       "\n",
       "\\end{itemize}\n",
       "\\end{itemize}\n"
      ],
      "text/markdown": [
       "```\n",
       "CSV.File(source; kwargs...) => CSV.File\n",
       "```\n",
       "\n",
       "Read a UTF-8 CSV input (a filename given as a String or FilePaths.jl type, or any other IO source), returning a `CSV.File` object.\n",
       "\n",
       "Opens the file and uses passed arguments to detect the number of columns and column types, unless column types are provided manually via the `types` keyword argument. Note that passing column types manually can increase performance and reduce the memory use for each column type provided (column types can be given as a `Vector` for all columns, or specified per column via name or index in a `Dict`). For text encodings other than UTF-8, see the [StringEncodings.jl](https://github.com/JuliaStrings/StringEncodings.jl) package for re-encoding a file or IO stream. The returned `CSV.File` object supports the [Tables.jl](https://github.com/JuliaData/Tables.jl) interface and can iterate `CSV.Row`s. `CSV.Row` supports `propertynames` and `getproperty` to access individual row values. `CSV.File` also supports entire column access like a `DataFrame` via direct property access on the file object, like `f = CSV.File(file); f.col1`. Note that duplicate column names will be detected and adjusted to ensure uniqueness (duplicate column name `a` will become `a_1`). For example, one could iterate over a csv file with column names `a`, `b`, and `c` by doing:\n",
       "\n",
       "```julia\n",
       "for row in CSV.File(file)\n",
       "    println(\"a=$(row.a), b=$(row.b), c=$(row.c)\")\n",
       "end\n",
       "```\n",
       "\n",
       "By supporting the Tables.jl interface, a `CSV.File` can also be a table input to any other table sink function. Like:\n",
       "\n",
       "```julia\n",
       "# materialize a csv file as a DataFrame, without copying columns from CSV.File; these columns are read-only\n",
       "df = CSV.File(file) |> DataFrame!\n",
       "\n",
       "# load a csv file directly into an sqlite database table\n",
       "db = SQLite.DB()\n",
       "tbl = CSV.File(file) |> SQLite.load!(db, \"sqlite_table\")\n",
       "```\n",
       "\n",
       "Supported keyword arguments include:\n",
       "\n",
       "  * File layout options:\n",
       "\n",
       "      * `header=1`: the `header` argument can be an `Int`, indicating the row to parse for column names; or a `Range`, indicating a span of rows to be concatenated together as column names; or an entire `Vector{Symbol}` or `Vector{String}` to use as column names; if a file doesn't have column names, either provide them as a `Vector`, or set `header=0` or `header=false` and column names will be auto-generated (`Column1`, `Column2`, etc.)\n",
       "      * `normalizenames=false`: whether column names should be \"normalized\" into valid Julia identifier symbols; useful when iterating rows and accessing column values of a row via `getproperty` (e.g. `row.col1`)\n",
       "      * `datarow`: an `Int` argument to specify the row where the data starts in the csv file; by default, the next row after the `header` row is used. If `header=0`, then the 1st row is assumed to be the start of data\n",
       "      * `skipto::Int`: similar to `datarow`, specifies the number of rows to skip before starting to read data\n",
       "      * `footerskip::Int`: number of rows at the end of a file to skip parsing\n",
       "      * `limit`: an `Int` to indicate a limited number of rows to parse in a csv file; use in combination with `skipto` to read a specific, contiguous chunk within a file\n",
       "      * `transpose::Bool`: read a csv file \"transposed\", i.e. each column is parsed as a row\n",
       "      * `comment`: rows that begin with this `String` will be skipped while parsing\n",
       "      * `use_mmap::Bool=!Sys.iswindows()`: whether the file should be mmapped for reading, which in some cases can be faster\n",
       "      * `ignoreemptylines::Bool=false`: whether empty rows/lines in a file should be ignored (if `false`, each column will be assigned `missing` for that empty row)\n",
       "      * `threaded::Bool`: whether parsing should utilize multiple threads; by default threads are used on large enough files, but isn't allowed when `transpose=true` or when `limit` is used; only available in Julia 1.3+\n",
       "      * `select`: an `AbstractVector` of `Int`, `Symbol`, `String`, or `Bool`, or a \"selector\" function of the form `(i, name) -> keep::Bool`; only columns in the collection or for which the selector function returns `true` will be parsed and accessible in the resulting `CSV.File`. Invalid values in `select` are ignored.\n",
       "      * `drop`: inverse of `select`; an `AbstractVector` of `Int`, `Symbol`, `String`, or `Bool`, or a \"drop\" function of the form `(i, name) -> drop::Bool`; columns in the collection or for which the drop function returns `true` will ignored in the resulting `CSV.File`. Invalid values in `drop` are ignored.\n",
       "  * Parsing options:\n",
       "\n",
       "      * `missingstrings`, `missingstring`: either a `String`, or `Vector{String}` to use as sentinel values that will be parsed as `missing`; by default, only an empty field (two consecutive delimiters) is considered `missing`\n",
       "      * `delim=','`: a `Char` or `String` that indicates how columns are delimited in a file; if no argument is provided, parsing will try to detect the most consistent delimiter on the first 10 rows of the file\n",
       "      * `ignorerepeated::Bool=false`: whether repeated (consecutive) delimiters should be ignored while parsing; useful for fixed-width files with delimiter padding between cells\n",
       "      * `quotechar='\"'`, `openquotechar`, `closequotechar`: a `Char` (or different start and end characters) that indicate a quoted field which may contain textual delimiters or newline characters\n",
       "      * `escapechar='\"'`: the `Char` used to escape quote characters in a quoted field\n",
       "      * `dateformat::Union{String, Dates.DateFormat, Nothing}`: a date format string to indicate how Date/DateTime columns are formatted for the entire file\n",
       "      * `decimal='.'`: a `Char` indicating how decimals are separated in floats, i.e. `3.14` used '.', or `3,14` uses a comma ','\n",
       "      * `truestrings`, `falsestrings`: `Vectors of Strings` that indicate how `true` or `false` values are represented; by default only `true` and `false` are treated as `Bool`\n",
       "  * Column Type Options:\n",
       "\n",
       "      * `type`: a single type to use for parsing an entire file; i.e. all columns will be treated as the same type; useful for matrix-like data files\n",
       "      * `types`: a Vector or Dict of types to be used for column types; a Dict can map column index `Int`, or name `Symbol` or `String` to type for a column, i.e. Dict(1=>Float64) will set the first column as a Float64, Dict(:column1=>Float64) will set the column named column1 to Float64 and, Dict(\"column1\"=>Float64) will set the column1 to Float64; if a `Vector` if provided, it must match the # of columns provided or detected in `header`\n",
       "      * `typemap::Dict{Type, Type}`: a mapping of a type that should be replaced in every instance with another type, i.e. `Dict(Float64=>String)` would change every detected `Float64` column to be parsed as `String`\n",
       "      * `pool::Union{Bool, Float64}=0.1`: if `true`, *all* columns detected as `String` will be internally pooled; alternatively, the proportion of unique values below which `String` columns should be pooled (by default 0.1, meaning that if the # of unique strings in a column is under 10%, it will be pooled)\n",
       "      * `categorical::Bool=false`: whether pooled columns should be copied as CategoricalArray instead of PooledArray; note that in `CSV.read`, by default, columns are not copied, so pooled columns will have type `CSV.Column{String, PooledString}`; to get `CategoricalArray` columns, also pass `copycols=true`\n",
       "      * `strict::Bool=false`: whether invalid values should throw a parsing error or be replaced with `missing`\n",
       "      * `silencewarnings::Bool=false`: if `strict=false`, whether invalid value warnings should be silenced\n"
      ],
      "text/plain": [
       "\u001b[36m  CSV.File(source; kwargs...) => CSV.File\u001b[39m\n",
       "\n",
       "  Read a UTF-8 CSV input (a filename given as a String or FilePaths.jl type,\n",
       "  or any other IO source), returning a \u001b[36mCSV.File\u001b[39m object.\n",
       "\n",
       "  Opens the file and uses passed arguments to detect the number of columns and\n",
       "  column types, unless column types are provided manually via the \u001b[36mtypes\u001b[39m\n",
       "  keyword argument. Note that passing column types manually can increase\n",
       "  performance and reduce the memory use for each column type provided (column\n",
       "  types can be given as a \u001b[36mVector\u001b[39m for all columns, or specified per column via\n",
       "  name or index in a \u001b[36mDict\u001b[39m). For text encodings other than UTF-8, see the\n",
       "  StringEncodings.jl (https://github.com/JuliaStrings/StringEncodings.jl)\n",
       "  package for re-encoding a file or IO stream. The returned \u001b[36mCSV.File\u001b[39m object\n",
       "  supports the Tables.jl (https://github.com/JuliaData/Tables.jl) interface\n",
       "  and can iterate \u001b[36mCSV.Row\u001b[39ms. \u001b[36mCSV.Row\u001b[39m supports \u001b[36mpropertynames\u001b[39m and \u001b[36mgetproperty\u001b[39m to\n",
       "  access individual row values. \u001b[36mCSV.File\u001b[39m also supports entire column access\n",
       "  like a \u001b[36mDataFrame\u001b[39m via direct property access on the file object, like \u001b[36mf =\n",
       "  CSV.File(file); f.col1\u001b[39m. Note that duplicate column names will be detected\n",
       "  and adjusted to ensure uniqueness (duplicate column name \u001b[36ma\u001b[39m will become \u001b[36ma_1\u001b[39m).\n",
       "  For example, one could iterate over a csv file with column names \u001b[36ma\u001b[39m, \u001b[36mb\u001b[39m, and \u001b[36mc\u001b[39m\n",
       "  by doing:\n",
       "\n",
       "\u001b[36m  for row in CSV.File(file)\u001b[39m\n",
       "\u001b[36m      println(\"a=$(row.a), b=$(row.b), c=$(row.c)\")\u001b[39m\n",
       "\u001b[36m  end\u001b[39m\n",
       "\n",
       "  By supporting the Tables.jl interface, a \u001b[36mCSV.File\u001b[39m can also be a table input\n",
       "  to any other table sink function. Like:\n",
       "\n",
       "\u001b[36m  # materialize a csv file as a DataFrame, without copying columns from CSV.File; these columns are read-only\u001b[39m\n",
       "\u001b[36m  df = CSV.File(file) |> DataFrame!\u001b[39m\n",
       "\u001b[36m  \u001b[39m\n",
       "\u001b[36m  # load a csv file directly into an sqlite database table\u001b[39m\n",
       "\u001b[36m  db = SQLite.DB()\u001b[39m\n",
       "\u001b[36m  tbl = CSV.File(file) |> SQLite.load!(db, \"sqlite_table\")\u001b[39m\n",
       "\n",
       "  Supported keyword arguments include:\n",
       "\n",
       "    •    File layout options:\n",
       "      \n",
       "          •    \u001b[36mheader=1\u001b[39m: the \u001b[36mheader\u001b[39m argument can be an \u001b[36mInt\u001b[39m, indicating\n",
       "              the row to parse for column names; or a \u001b[36mRange\u001b[39m,\n",
       "              indicating a span of rows to be concatenated together as\n",
       "              column names; or an entire \u001b[36mVector{Symbol}\u001b[39m or\n",
       "              \u001b[36mVector{String}\u001b[39m to use as column names; if a file doesn't\n",
       "              have column names, either provide them as a \u001b[36mVector\u001b[39m, or\n",
       "              set \u001b[36mheader=0\u001b[39m or \u001b[36mheader=false\u001b[39m and column names will be\n",
       "              auto-generated (\u001b[36mColumn1\u001b[39m, \u001b[36mColumn2\u001b[39m, etc.)\n",
       "      \n",
       "          •    \u001b[36mnormalizenames=false\u001b[39m: whether column names should be\n",
       "              \"normalized\" into valid Julia identifier symbols; useful\n",
       "              when iterating rows and accessing column values of a row\n",
       "              via \u001b[36mgetproperty\u001b[39m (e.g. \u001b[36mrow.col1\u001b[39m)\n",
       "      \n",
       "          •    \u001b[36mdatarow\u001b[39m: an \u001b[36mInt\u001b[39m argument to specify the row where the\n",
       "              data starts in the csv file; by default, the next row\n",
       "              after the \u001b[36mheader\u001b[39m row is used. If \u001b[36mheader=0\u001b[39m, then the 1st\n",
       "              row is assumed to be the start of data\n",
       "      \n",
       "          •    \u001b[36mskipto::Int\u001b[39m: similar to \u001b[36mdatarow\u001b[39m, specifies the number of\n",
       "              rows to skip before starting to read data\n",
       "      \n",
       "          •    \u001b[36mfooterskip::Int\u001b[39m: number of rows at the end of a file to\n",
       "              skip parsing\n",
       "      \n",
       "          •    \u001b[36mlimit\u001b[39m: an \u001b[36mInt\u001b[39m to indicate a limited number of rows to\n",
       "              parse in a csv file; use in combination with \u001b[36mskipto\u001b[39m to\n",
       "              read a specific, contiguous chunk within a file\n",
       "      \n",
       "          •    \u001b[36mtranspose::Bool\u001b[39m: read a csv file \"transposed\", i.e. each\n",
       "              column is parsed as a row\n",
       "      \n",
       "          •    \u001b[36mcomment\u001b[39m: rows that begin with this \u001b[36mString\u001b[39m will be\n",
       "              skipped while parsing\n",
       "      \n",
       "          •    \u001b[36muse_mmap::Bool=!Sys.iswindows()\u001b[39m: whether the file should\n",
       "              be mmapped for reading, which in some cases can be\n",
       "              faster\n",
       "      \n",
       "          •    \u001b[36mignoreemptylines::Bool=false\u001b[39m: whether empty rows/lines\n",
       "              in a file should be ignored (if \u001b[36mfalse\u001b[39m, each column will\n",
       "              be assigned \u001b[36mmissing\u001b[39m for that empty row)\n",
       "      \n",
       "          •    \u001b[36mthreaded::Bool\u001b[39m: whether parsing should utilize multiple\n",
       "              threads; by default threads are used on large enough\n",
       "              files, but isn't allowed when \u001b[36mtranspose=true\u001b[39m or when\n",
       "              \u001b[36mlimit\u001b[39m is used; only available in Julia 1.3+\n",
       "      \n",
       "          •    \u001b[36mselect\u001b[39m: an \u001b[36mAbstractVector\u001b[39m of \u001b[36mInt\u001b[39m, \u001b[36mSymbol\u001b[39m, \u001b[36mString\u001b[39m, or\n",
       "              \u001b[36mBool\u001b[39m, or a \"selector\" function of the form \u001b[36m(i, name) ->\n",
       "              keep::Bool\u001b[39m; only columns in the collection or for which\n",
       "              the selector function returns \u001b[36mtrue\u001b[39m will be parsed and\n",
       "              accessible in the resulting \u001b[36mCSV.File\u001b[39m. Invalid values in\n",
       "              \u001b[36mselect\u001b[39m are ignored.\n",
       "      \n",
       "          •    \u001b[36mdrop\u001b[39m: inverse of \u001b[36mselect\u001b[39m; an \u001b[36mAbstractVector\u001b[39m of \u001b[36mInt\u001b[39m,\n",
       "              \u001b[36mSymbol\u001b[39m, \u001b[36mString\u001b[39m, or \u001b[36mBool\u001b[39m, or a \"drop\" function of the\n",
       "              form \u001b[36m(i, name) -> drop::Bool\u001b[39m; columns in the collection\n",
       "              or for which the drop function returns \u001b[36mtrue\u001b[39m will ignored\n",
       "              in the resulting \u001b[36mCSV.File\u001b[39m. Invalid values in \u001b[36mdrop\u001b[39m are\n",
       "              ignored.\n",
       "\n",
       "    •    Parsing options:\n",
       "      \n",
       "          •    \u001b[36mmissingstrings\u001b[39m, \u001b[36mmissingstring\u001b[39m: either a \u001b[36mString\u001b[39m, or\n",
       "              \u001b[36mVector{String}\u001b[39m to use as sentinel values that will be\n",
       "              parsed as \u001b[36mmissing\u001b[39m; by default, only an empty field (two\n",
       "              consecutive delimiters) is considered \u001b[36mmissing\u001b[39m\n",
       "      \n",
       "          •    \u001b[36mdelim=','\u001b[39m: a \u001b[36mChar\u001b[39m or \u001b[36mString\u001b[39m that indicates how columns\n",
       "              are delimited in a file; if no argument is provided,\n",
       "              parsing will try to detect the most consistent delimiter\n",
       "              on the first 10 rows of the file\n",
       "      \n",
       "          •    \u001b[36mignorerepeated::Bool=false\u001b[39m: whether repeated\n",
       "              (consecutive) delimiters should be ignored while\n",
       "              parsing; useful for fixed-width files with delimiter\n",
       "              padding between cells\n",
       "      \n",
       "          •    \u001b[36mquotechar='\"'\u001b[39m, \u001b[36mopenquotechar\u001b[39m, \u001b[36mclosequotechar\u001b[39m: a \u001b[36mChar\u001b[39m (or\n",
       "              different start and end characters) that indicate a\n",
       "              quoted field which may contain textual delimiters or\n",
       "              newline characters\n",
       "      \n",
       "          •    \u001b[36mescapechar='\"'\u001b[39m: the \u001b[36mChar\u001b[39m used to escape quote characters\n",
       "              in a quoted field\n",
       "      \n",
       "          •    \u001b[36mdateformat::Union{String, Dates.DateFormat, Nothing}\u001b[39m: a\n",
       "              date format string to indicate how Date/DateTime columns\n",
       "              are formatted for the entire file\n",
       "      \n",
       "          •    \u001b[36mdecimal='.'\u001b[39m: a \u001b[36mChar\u001b[39m indicating how decimals are\n",
       "              separated in floats, i.e. \u001b[36m3.14\u001b[39m used '.', or \u001b[36m3,14\u001b[39m uses a\n",
       "              comma ','\n",
       "      \n",
       "          •    \u001b[36mtruestrings\u001b[39m, \u001b[36mfalsestrings\u001b[39m: \u001b[36mVectors of Strings\u001b[39m that\n",
       "              indicate how \u001b[36mtrue\u001b[39m or \u001b[36mfalse\u001b[39m values are represented; by\n",
       "              default only \u001b[36mtrue\u001b[39m and \u001b[36mfalse\u001b[39m are treated as \u001b[36mBool\u001b[39m\n",
       "\n",
       "    •    Column Type Options:\n",
       "      \n",
       "          •    \u001b[36mtype\u001b[39m: a single type to use for parsing an entire file;\n",
       "              i.e. all columns will be treated as the same type;\n",
       "              useful for matrix-like data files\n",
       "      \n",
       "          •    \u001b[36mtypes\u001b[39m: a Vector or Dict of types to be used for column\n",
       "              types; a Dict can map column index \u001b[36mInt\u001b[39m, or name \u001b[36mSymbol\u001b[39m\n",
       "              or \u001b[36mString\u001b[39m to type for a column, i.e. Dict(1=>Float64)\n",
       "              will set the first column as a Float64,\n",
       "              Dict(:column1=>Float64) will set the column named\n",
       "              column1 to Float64 and, Dict(\"column1\"=>Float64) will\n",
       "              set the column1 to Float64; if a \u001b[36mVector\u001b[39m if provided, it\n",
       "              must match the # of columns provided or detected in\n",
       "              \u001b[36mheader\u001b[39m\n",
       "      \n",
       "          •    \u001b[36mtypemap::Dict{Type, Type}\u001b[39m: a mapping of a type that\n",
       "              should be replaced in every instance with another type,\n",
       "              i.e. \u001b[36mDict(Float64=>String)\u001b[39m would change every detected\n",
       "              \u001b[36mFloat64\u001b[39m column to be parsed as \u001b[36mString\u001b[39m\n",
       "      \n",
       "          •    \u001b[36mpool::Union{Bool, Float64}=0.1\u001b[39m: if \u001b[36mtrue\u001b[39m, \u001b[4mall\u001b[24m columns\n",
       "              detected as \u001b[36mString\u001b[39m will be internally pooled;\n",
       "              alternatively, the proportion of unique values below\n",
       "              which \u001b[36mString\u001b[39m columns should be pooled (by default 0.1,\n",
       "              meaning that if the # of unique strings in a column is\n",
       "              under 10%, it will be pooled)\n",
       "      \n",
       "          •    \u001b[36mcategorical::Bool=false\u001b[39m: whether pooled columns should\n",
       "              be copied as CategoricalArray instead of PooledArray;\n",
       "              note that in \u001b[36mCSV.read\u001b[39m, by default, columns are not\n",
       "              copied, so pooled columns will have type\n",
       "              \u001b[36mCSV.Column{String, PooledString}\u001b[39m; to get\n",
       "              \u001b[36mCategoricalArray\u001b[39m columns, also pass \u001b[36mcopycols=true\u001b[39m\n",
       "      \n",
       "          •    \u001b[36mstrict::Bool=false\u001b[39m: whether invalid values should throw\n",
       "              a parsing error or be replaced with \u001b[36mmissing\u001b[39m\n",
       "      \n",
       "          •    \u001b[36msilencewarnings::Bool=false\u001b[39m: if \u001b[36mstrict=false\u001b[39m, whether\n",
       "              invalid value warnings should be silenced"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "?CSV.File"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Julia 1.4.1",
   "language": "julia",
   "name": "julia-1.4"
  },
  "language_info": {
   "file_extension": ".jl",
   "mimetype": "application/julia",
   "name": "julia",
   "version": "1.4.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}