{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Non Local Means\n",
    "===============\n",
    "\n",
    "*Important:* Please read the [installation page](http://gpeyre.github.io/numerical-tours/installation_python/) for details about how to install the toolboxes.\n",
    "$\\newcommand{\\dotp}[2]{\\langle #1, #2 \\rangle}$\n",
    "$\\newcommand{\\enscond}[2]{\\lbrace #1, #2 \\rbrace}$\n",
    "$\\newcommand{\\pd}[2]{ \\frac{ \\partial #1}{\\partial #2} }$\n",
    "$\\newcommand{\\umin}[1]{\\underset{#1}{\\min}\\;}$\n",
    "$\\newcommand{\\umax}[1]{\\underset{#1}{\\max}\\;}$\n",
    "$\\newcommand{\\umin}[1]{\\underset{#1}{\\min}\\;}$\n",
    "$\\newcommand{\\uargmin}[1]{\\underset{#1}{argmin}\\;}$\n",
    "$\\newcommand{\\norm}[1]{\\|#1\\|}$\n",
    "$\\newcommand{\\abs}[1]{\\left|#1\\right|}$\n",
    "$\\newcommand{\\choice}[1]{ \\left\\{  \\begin{array}{l} #1 \\end{array} \\right. }$\n",
    "$\\newcommand{\\pa}[1]{\\left(#1\\right)}$\n",
    "$\\newcommand{\\diag}[1]{{diag}\\left( #1 \\right)}$\n",
    "$\\newcommand{\\qandq}{\\quad\\text{and}\\quad}$\n",
    "$\\newcommand{\\qwhereq}{\\quad\\text{where}\\quad}$\n",
    "$\\newcommand{\\qifq}{ \\quad \\text{if} \\quad }$\n",
    "$\\newcommand{\\qarrq}{ \\quad \\Longrightarrow \\quad }$\n",
    "$\\newcommand{\\ZZ}{\\mathbb{Z}}$\n",
    "$\\newcommand{\\CC}{\\mathbb{C}}$\n",
    "$\\newcommand{\\RR}{\\mathbb{R}}$\n",
    "$\\newcommand{\\EE}{\\mathbb{E}}$\n",
    "$\\newcommand{\\Zz}{\\mathcal{Z}}$\n",
    "$\\newcommand{\\Ww}{\\mathcal{W}}$\n",
    "$\\newcommand{\\Vv}{\\mathcal{V}}$\n",
    "$\\newcommand{\\Nn}{\\mathcal{N}}$\n",
    "$\\newcommand{\\NN}{\\mathcal{N}}$\n",
    "$\\newcommand{\\Hh}{\\mathcal{H}}$\n",
    "$\\newcommand{\\Bb}{\\mathcal{B}}$\n",
    "$\\newcommand{\\Ee}{\\mathcal{E}}$\n",
    "$\\newcommand{\\Cc}{\\mathcal{C}}$\n",
    "$\\newcommand{\\Gg}{\\mathcal{G}}$\n",
    "$\\newcommand{\\Ss}{\\mathcal{S}}$\n",
    "$\\newcommand{\\Pp}{\\mathcal{P}}$\n",
    "$\\newcommand{\\Ff}{\\mathcal{F}}$\n",
    "$\\newcommand{\\Xx}{\\mathcal{X}}$\n",
    "$\\newcommand{\\Mm}{\\mathcal{M}}$\n",
    "$\\newcommand{\\Ii}{\\mathcal{I}}$\n",
    "$\\newcommand{\\Dd}{\\mathcal{D}}$\n",
    "$\\newcommand{\\Ll}{\\mathcal{L}}$\n",
    "$\\newcommand{\\Tt}{\\mathcal{T}}$\n",
    "$\\newcommand{\\si}{\\sigma}$\n",
    "$\\newcommand{\\al}{\\alpha}$\n",
    "$\\newcommand{\\la}{\\lambda}$\n",
    "$\\newcommand{\\ga}{\\gamma}$\n",
    "$\\newcommand{\\Ga}{\\Gamma}$\n",
    "$\\newcommand{\\La}{\\Lambda}$\n",
    "$\\newcommand{\\si}{\\sigma}$\n",
    "$\\newcommand{\\Si}{\\Sigma}$\n",
    "$\\newcommand{\\be}{\\beta}$\n",
    "$\\newcommand{\\de}{\\delta}$\n",
    "$\\newcommand{\\De}{\\Delta}$\n",
    "$\\newcommand{\\phi}{\\varphi}$\n",
    "$\\newcommand{\\th}{\\theta}$\n",
    "$\\newcommand{\\om}{\\omega}$\n",
    "$\\newcommand{\\Om}{\\Omega}$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This numerical tour study image denoising using\n",
    "non-local means. This algorithm has been\n",
    "introduced for denoising purposes in [BuaCoMoA05](#biblio)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "using PyPlot\n",
    "using NtToolBox"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Patches in Images\n",
    "-----------------\n",
    "This numerical tour is dedicated to the study of the structure of patches\n",
    "in images.\n",
    "\n",
    "\n",
    "Size $N = n \\times n$ of the image."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "128"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "n = 128"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We load a noisy image $f_0\\in \\RR^N$."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "128×128 Array{Float32,2}:\n",
       " 0.362205  0.362205  0.354331  0.401575  …  0.354331  0.362205  0.362205\n",
       " 0.385827  0.354331  0.346457  0.362205     0.417323  0.385827  0.401575\n",
       " 0.338583  0.338583  0.362205  0.346457     0.377953  0.385827  0.401575\n",
       " 0.346457  0.338583  0.354331  0.362205     0.362205  0.362205  0.362205\n",
       " 0.385827  0.338583  0.338583  0.354331     0.370079  0.377953  0.338583\n",
       " 0.370079  0.362205  0.385827  0.354331  …  0.354331  0.385827  0.346457\n",
       " 0.330709  0.354331  0.385827  0.346457     0.362205  0.377953  0.346457\n",
       " 0.354331  0.362205  0.362205  0.354331     0.393701  0.362205  0.346457\n",
       " 0.346457  0.354331  0.354331  0.354331     0.330709  0.393701  0.362205\n",
       " 0.385827  0.354331  0.385827  0.354331     0.330709  0.330709  0.385827\n",
       " 0.354331  0.354331  0.338583  0.377953  …  0.354331  0.354331  0.362205\n",
       " 0.354331  0.354331  0.385827  0.362205     0.330709  0.338583  0.346457\n",
       " 0.362205  0.370079  0.370079  0.385827     0.338583  0.362205  0.330709\n",
       " ⋮                                       ⋱  ⋮                           \n",
       " 0.283465  0.291339  0.409449  0.354331     0.669291  0.669291  0.692913\n",
       " 0.291339  0.322835  0.527559  0.409449     0.677165  0.669291  0.685039\n",
       " 0.346457  0.385827  0.417323  0.377953     0.692913  0.72441   0.72441 \n",
       " 0.472441  0.370079  0.377953  0.354331     0.645669  0.72441   0.732284\n",
       " 0.582677  0.425197  0.409449  0.448819  …  0.574803  0.637795  0.700787\n",
       " 0.464567  0.401575  0.496063  0.448819     0.622047  0.566929  0.574803\n",
       " 0.464567  0.417323  0.456693  0.409449     0.677165  0.669291  0.582677\n",
       " 0.559055  0.464567  0.440945  0.385827     0.661417  0.692913  0.669291\n",
       " 0.543307  0.433071  0.401575  0.480315     0.685039  0.692913  0.608924\n",
       " 0.480315  0.330709  0.472441  0.566929  …  0.645669  0.582677  0.566929\n",
       " 0.464567  0.393701  0.496063  0.488189     0.535433  0.496063  0.653543\n",
       " 0.464567  0.503937  0.425197  0.322835     0.551181  0.608924  0.629921"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "c = [100, 200]\n",
    "f0 = load_image(\"NtToolBox/src/data/lena.png\")\n",
    "f0 = rescale(f0[c[1] - Base.div(n, 2) + 1 : c[1] + Base.div(n, 2), c[2] - Base.div(n, 2) + 1 : c[2] + Base.div(n, 2)])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Display $f_0$."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "PyPlot.Figure(PyObject <matplotlib.figure.Figure object at 0x000000001E546E80>)"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "figure(figsize = (5, 5))\n",
    "imageplot(f0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Noise level $\\si$."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.04"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sigma = .04"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Generate a noisy image $f=f_0+\\epsilon$ where $\\epsilon \\times\n",
    "\\Nn(0,\\si^2\\text{Id}_N)$."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "128×128 Array{Float64,2}:\n",
       " 0.310121  0.331575  0.421916  0.41914   …  0.404933  0.373489  0.366263\n",
       " 0.371062  0.375166  0.367454  0.298638     0.409829  0.324649  0.391277\n",
       " 0.338717  0.337497  0.350857  0.322122     0.305613  0.443315  0.328793\n",
       " 0.358469  0.408959  0.302386  0.420562     0.36485   0.386978  0.353269\n",
       " 0.366528  0.307268  0.334106  0.41304      0.394557  0.403713  0.349689\n",
       " 0.301808  0.372549  0.383443  0.387839  …  0.337633  0.398922  0.343062\n",
       " 0.271791  0.389746  0.422243  0.281307     0.326126  0.359733  0.318974\n",
       " 0.397233  0.429843  0.395194  0.403108     0.402998  0.382453  0.380637\n",
       " 0.375241  0.388268  0.349711  0.427031     0.301518  0.44367   0.255511\n",
       " 0.369202  0.334936  0.424085  0.340765     0.33009   0.321673  0.33002 \n",
       " 0.39084   0.361636  0.372712  0.314115  …  0.438289  0.320409  0.350257\n",
       " 0.341601  0.383261  0.4244    0.366063     0.322722  0.400725  0.360035\n",
       " 0.352725  0.376963  0.435233  0.383157     0.370172  0.409228  0.294399\n",
       " ⋮                                       ⋱  ⋮                           \n",
       " 0.28303   0.312103  0.44081   0.345268     0.659962  0.706894  0.771867\n",
       " 0.301504  0.379728  0.581491  0.440036     0.663047  0.651876  0.648853\n",
       " 0.345021  0.388229  0.447937  0.416182     0.657772  0.74391   0.703863\n",
       " 0.49896   0.403568  0.364329  0.346948     0.67903   0.667091  0.72686 \n",
       " 0.626079  0.425512  0.424497  0.492793  …  0.575927  0.607786  0.731195\n",
       " 0.434764  0.447411  0.443906  0.453921     0.680113  0.624357  0.586979\n",
       " 0.505774  0.428263  0.488337  0.428573     0.711711  0.630046  0.585115\n",
       " 0.528377  0.459281  0.439019  0.44862      0.666042  0.737301  0.679758\n",
       " 0.5479    0.433398  0.463395  0.496517     0.657626  0.757109  0.694726\n",
       " 0.441499  0.33134   0.540048  0.567202  …  0.644774  0.634769  0.64582 \n",
       " 0.432999  0.321467  0.467588  0.485594     0.508329  0.498877  0.628629\n",
       " 0.460332  0.480376  0.394749  0.291599     0.494321  0.5831    0.570823"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "using Distributions\n",
    "f = f0 .+ sigma.*rand(Normal(), n, n)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Display $f$."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "PyPlot.Figure(PyObject <matplotlib.figure.Figure object at 0x000000001E664320>)"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "figure(figsize = (5,5))\n",
    "imageplot(clamP(f))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We denote $w$ to be the half width of the patches,\n",
    "and $w_1=2w+1$ the full width."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "7"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "w = 3\n",
    "w1 = 2*w + 1"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We set up large $(n,n,w_1,w_1)$ matrices to index the the X and Y\n",
    "position of the pixel to extract.\n",
    "\n",
    "Location of pixels to extract."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "128×128×7×7 Array{Int64,4}:\n",
       "[:, :, 1, 1] =\n",
       "  -2   -2   -2   -2   -2   -2   -2   -2  …   -2   -2   -2   -2   -2   -2   -2\n",
       "  -1   -1   -1   -1   -1   -1   -1   -1      -1   -1   -1   -1   -1   -1   -1\n",
       "   0    0    0    0    0    0    0    0       0    0    0    0    0    0    0\n",
       "   1    1    1    1    1    1    1    1       1    1    1    1    1    1    1\n",
       "   2    2    2    2    2    2    2    2       2    2    2    2    2    2    2\n",
       "   3    3    3    3    3    3    3    3  …    3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4       4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8  …    8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9       9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 114  114  114  114  114  114  114  114     114  114  114  114  114  114  114\n",
       " 115  115  115  115  115  115  115  115     115  115  115  115  115  115  115\n",
       " 116  116  116  116  116  116  116  116     116  116  116  116  116  116  116\n",
       " 117  117  117  117  117  117  117  117     117  117  117  117  117  117  117\n",
       " 118  118  118  118  118  118  118  118  …  118  118  118  118  118  118  118\n",
       " 119  119  119  119  119  119  119  119     119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123  …  123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124     124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       "\n",
       "[:, :, 2, 1] =\n",
       "  -1   -1   -1   -1   -1   -1   -1   -1  …   -1   -1   -1   -1   -1   -1   -1\n",
       "   0    0    0    0    0    0    0    0       0    0    0    0    0    0    0\n",
       "   1    1    1    1    1    1    1    1       1    1    1    1    1    1    1\n",
       "   2    2    2    2    2    2    2    2       2    2    2    2    2    2    2\n",
       "   3    3    3    3    3    3    3    3       3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4  …    4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8       8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9  …    9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11      11   11   11   11   11   11   11\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 115  115  115  115  115  115  115  115     115  115  115  115  115  115  115\n",
       " 116  116  116  116  116  116  116  116     116  116  116  116  116  116  116\n",
       " 117  117  117  117  117  117  117  117     117  117  117  117  117  117  117\n",
       " 118  118  118  118  118  118  118  118     118  118  118  118  118  118  118\n",
       " 119  119  119  119  119  119  119  119  …  119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123     123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124  …  124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126     126  126  126  126  126  126  126\n",
       "\n",
       "[:, :, 3, 1] =\n",
       "   0    0    0    0    0    0    0    0  …    0    0    0    0    0    0    0\n",
       "   1    1    1    1    1    1    1    1       1    1    1    1    1    1    1\n",
       "   2    2    2    2    2    2    2    2       2    2    2    2    2    2    2\n",
       "   3    3    3    3    3    3    3    3       3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4       4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5  …    5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8       8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9       9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10  …   10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11      11   11   11   11   11   11   11\n",
       "  12   12   12   12   12   12   12   12      12   12   12   12   12   12   12\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 116  116  116  116  116  116  116  116     116  116  116  116  116  116  116\n",
       " 117  117  117  117  117  117  117  117     117  117  117  117  117  117  117\n",
       " 118  118  118  118  118  118  118  118     118  118  118  118  118  118  118\n",
       " 119  119  119  119  119  119  119  119     119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120  …  120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123     123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124     124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125  …  125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126     126  126  126  126  126  126  126\n",
       " 127  127  127  127  127  127  127  127     127  127  127  127  127  127  127\n",
       "\n",
       "[:, :, 4, 1] =\n",
       "   1    1    1    1    1    1    1    1  …    1    1    1    1    1    1    1\n",
       "   2    2    2    2    2    2    2    2       2    2    2    2    2    2    2\n",
       "   3    3    3    3    3    3    3    3       3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4       4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6  …    6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8       8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9       9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11  …   11   11   11   11   11   11   11\n",
       "  12   12   12   12   12   12   12   12      12   12   12   12   12   12   12\n",
       "  13   13   13   13   13   13   13   13      13   13   13   13   13   13   13\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 117  117  117  117  117  117  117  117     117  117  117  117  117  117  117\n",
       " 118  118  118  118  118  118  118  118     118  118  118  118  118  118  118\n",
       " 119  119  119  119  119  119  119  119     119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121  …  121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123     123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124     124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126  …  126  126  126  126  126  126  126\n",
       " 127  127  127  127  127  127  127  127     127  127  127  127  127  127  127\n",
       " 128  128  128  128  128  128  128  128     128  128  128  128  128  128  128\n",
       "\n",
       "[:, :, 5, 1] =\n",
       "   2    2    2    2    2    2    2    2  …    2    2    2    2    2    2    2\n",
       "   3    3    3    3    3    3    3    3       3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4       4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7  …    7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8       8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9       9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11      11   11   11   11   11   11   11\n",
       "  12   12   12   12   12   12   12   12  …   12   12   12   12   12   12   12\n",
       "  13   13   13   13   13   13   13   13      13   13   13   13   13   13   13\n",
       "  14   14   14   14   14   14   14   14      14   14   14   14   14   14   14\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 118  118  118  118  118  118  118  118     118  118  118  118  118  118  118\n",
       " 119  119  119  119  119  119  119  119     119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122  …  122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123     123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124     124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126     126  126  126  126  126  126  126\n",
       " 127  127  127  127  127  127  127  127  …  127  127  127  127  127  127  127\n",
       " 128  128  128  128  128  128  128  128     128  128  128  128  128  128  128\n",
       " 129  129  129  129  129  129  129  129     129  129  129  129  129  129  129\n",
       "\n",
       "[:, :, 6, 1] =\n",
       "   3    3    3    3    3    3    3    3  …    3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4       4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8  …    8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9       9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11      11   11   11   11   11   11   11\n",
       "  12   12   12   12   12   12   12   12      12   12   12   12   12   12   12\n",
       "  13   13   13   13   13   13   13   13  …   13   13   13   13   13   13   13\n",
       "  14   14   14   14   14   14   14   14      14   14   14   14   14   14   14\n",
       "  15   15   15   15   15   15   15   15      15   15   15   15   15   15   15\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 119  119  119  119  119  119  119  119     119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123  …  123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124     124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126     126  126  126  126  126  126  126\n",
       " 127  127  127  127  127  127  127  127     127  127  127  127  127  127  127\n",
       " 128  128  128  128  128  128  128  128  …  128  128  128  128  128  128  128\n",
       " 129  129  129  129  129  129  129  129     129  129  129  129  129  129  129\n",
       " 130  130  130  130  130  130  130  130     130  130  130  130  130  130  130\n",
       "\n",
       "[:, :, 7, 1] =\n",
       "   4    4    4    4    4    4    4    4  …    4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8       8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9  …    9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11      11   11   11   11   11   11   11\n",
       "  12   12   12   12   12   12   12   12      12   12   12   12   12   12   12\n",
       "  13   13   13   13   13   13   13   13      13   13   13   13   13   13   13\n",
       "  14   14   14   14   14   14   14   14  …   14   14   14   14   14   14   14\n",
       "  15   15   15   15   15   15   15   15      15   15   15   15   15   15   15\n",
       "  16   16   16   16   16   16   16   16      16   16   16   16   16   16   16\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123     123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124  …  124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126     126  126  126  126  126  126  126\n",
       " 127  127  127  127  127  127  127  127     127  127  127  127  127  127  127\n",
       " 128  128  128  128  128  128  128  128     128  128  128  128  128  128  128\n",
       " 129  129  129  129  129  129  129  129  …  129  129  129  129  129  129  129\n",
       " 130  130  130  130  130  130  130  130     130  130  130  130  130  130  130\n",
       " 131  131  131  131  131  131  131  131     131  131  131  131  131  131  131\n",
       "\n",
       "[:, :, 1, 2] =\n",
       "  -2   -2   -2   -2   -2   -2   -2   -2  …   -2   -2   -2   -2   -2   -2   -2\n",
       "  -1   -1   -1   -1   -1   -1   -1   -1      -1   -1   -1   -1   -1   -1   -1\n",
       "   0    0    0    0    0    0    0    0       0    0    0    0    0    0    0\n",
       "   1    1    1    1    1    1    1    1       1    1    1    1    1    1    1\n",
       "   2    2    2    2    2    2    2    2       2    2    2    2    2    2    2\n",
       "   3    3    3    3    3    3    3    3  …    3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4       4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8  …    8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9       9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 114  114  114  114  114  114  114  114     114  114  114  114  114  114  114\n",
       " 115  115  115  115  115  115  115  115     115  115  115  115  115  115  115\n",
       " 116  116  116  116  116  116  116  116     116  116  116  116  116  116  116\n",
       " 117  117  117  117  117  117  117  117     117  117  117  117  117  117  117\n",
       " 118  118  118  118  118  118  118  118  …  118  118  118  118  118  118  118\n",
       " 119  119  119  119  119  119  119  119     119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123  …  123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124     124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       "\n",
       "[:, :, 2, 2] =\n",
       "  -1   -1   -1   -1   -1   -1   -1   -1  …   -1   -1   -1   -1   -1   -1   -1\n",
       "   0    0    0    0    0    0    0    0       0    0    0    0    0    0    0\n",
       "   1    1    1    1    1    1    1    1       1    1    1    1    1    1    1\n",
       "   2    2    2    2    2    2    2    2       2    2    2    2    2    2    2\n",
       "   3    3    3    3    3    3    3    3       3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4  …    4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8       8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9  …    9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11      11   11   11   11   11   11   11\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 115  115  115  115  115  115  115  115     115  115  115  115  115  115  115\n",
       " 116  116  116  116  116  116  116  116     116  116  116  116  116  116  116\n",
       " 117  117  117  117  117  117  117  117     117  117  117  117  117  117  117\n",
       " 118  118  118  118  118  118  118  118     118  118  118  118  118  118  118\n",
       " 119  119  119  119  119  119  119  119  …  119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123     123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124  …  124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126     126  126  126  126  126  126  126\n",
       "\n",
       "[:, :, 3, 2] =\n",
       "   0    0    0    0    0    0    0    0  …    0    0    0    0    0    0    0\n",
       "   1    1    1    1    1    1    1    1       1    1    1    1    1    1    1\n",
       "   2    2    2    2    2    2    2    2       2    2    2    2    2    2    2\n",
       "   3    3    3    3    3    3    3    3       3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4       4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5  …    5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8       8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9       9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10  …   10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11      11   11   11   11   11   11   11\n",
       "  12   12   12   12   12   12   12   12      12   12   12   12   12   12   12\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 116  116  116  116  116  116  116  116     116  116  116  116  116  116  116\n",
       " 117  117  117  117  117  117  117  117     117  117  117  117  117  117  117\n",
       " 118  118  118  118  118  118  118  118     118  118  118  118  118  118  118\n",
       " 119  119  119  119  119  119  119  119     119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120  …  120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123     123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124     124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125  …  125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126     126  126  126  126  126  126  126\n",
       " 127  127  127  127  127  127  127  127     127  127  127  127  127  127  127\n",
       "\n",
       "[:, :, 4, 2] =\n",
       "   1    1    1    1    1    1    1    1  …    1    1    1    1    1    1    1\n",
       "   2    2    2    2    2    2    2    2       2    2    2    2    2    2    2\n",
       "   3    3    3    3    3    3    3    3       3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4       4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6  …    6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8       8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9       9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11  …   11   11   11   11   11   11   11\n",
       "  12   12   12   12   12   12   12   12      12   12   12   12   12   12   12\n",
       "  13   13   13   13   13   13   13   13      13   13   13   13   13   13   13\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 117  117  117  117  117  117  117  117     117  117  117  117  117  117  117\n",
       " 118  118  118  118  118  118  118  118     118  118  118  118  118  118  118\n",
       " 119  119  119  119  119  119  119  119     119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121  …  121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123     123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124     124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126  …  126  126  126  126  126  126  126\n",
       " 127  127  127  127  127  127  127  127     127  127  127  127  127  127  127\n",
       " 128  128  128  128  128  128  128  128     128  128  128  128  128  128  128\n",
       "\n",
       "[:, :, 5, 2] =\n",
       "   2    2    2    2    2    2    2    2  …    2    2    2    2    2    2    2\n",
       "   3    3    3    3    3    3    3    3       3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4       4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7  …    7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8       8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9       9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11      11   11   11   11   11   11   11\n",
       "  12   12   12   12   12   12   12   12  …   12   12   12   12   12   12   12\n",
       "  13   13   13   13   13   13   13   13      13   13   13   13   13   13   13\n",
       "  14   14   14   14   14   14   14   14      14   14   14   14   14   14   14\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 118  118  118  118  118  118  118  118     118  118  118  118  118  118  118\n",
       " 119  119  119  119  119  119  119  119     119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122  …  122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123     123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124     124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126     126  126  126  126  126  126  126\n",
       " 127  127  127  127  127  127  127  127  …  127  127  127  127  127  127  127\n",
       " 128  128  128  128  128  128  128  128     128  128  128  128  128  128  128\n",
       " 129  129  129  129  129  129  129  129     129  129  129  129  129  129  129\n",
       "\n",
       "[:, :, 6, 2] =\n",
       "   3    3    3    3    3    3    3    3  …    3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4       4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8  …    8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9       9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11      11   11   11   11   11   11   11\n",
       "  12   12   12   12   12   12   12   12      12   12   12   12   12   12   12\n",
       "  13   13   13   13   13   13   13   13  …   13   13   13   13   13   13   13\n",
       "  14   14   14   14   14   14   14   14      14   14   14   14   14   14   14\n",
       "  15   15   15   15   15   15   15   15      15   15   15   15   15   15   15\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 119  119  119  119  119  119  119  119     119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123  …  123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124     124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126     126  126  126  126  126  126  126\n",
       " 127  127  127  127  127  127  127  127     127  127  127  127  127  127  127\n",
       " 128  128  128  128  128  128  128  128  …  128  128  128  128  128  128  128\n",
       " 129  129  129  129  129  129  129  129     129  129  129  129  129  129  129\n",
       " 130  130  130  130  130  130  130  130     130  130  130  130  130  130  130\n",
       "\n",
       "[:, :, 7, 2] =\n",
       "   4    4    4    4    4    4    4    4  …    4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8       8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9  …    9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11      11   11   11   11   11   11   11\n",
       "  12   12   12   12   12   12   12   12      12   12   12   12   12   12   12\n",
       "  13   13   13   13   13   13   13   13      13   13   13   13   13   13   13\n",
       "  14   14   14   14   14   14   14   14  …   14   14   14   14   14   14   14\n",
       "  15   15   15   15   15   15   15   15      15   15   15   15   15   15   15\n",
       "  16   16   16   16   16   16   16   16      16   16   16   16   16   16   16\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123     123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124  …  124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126     126  126  126  126  126  126  126\n",
       " 127  127  127  127  127  127  127  127     127  127  127  127  127  127  127\n",
       " 128  128  128  128  128  128  128  128     128  128  128  128  128  128  128\n",
       " 129  129  129  129  129  129  129  129  …  129  129  129  129  129  129  129\n",
       " 130  130  130  130  130  130  130  130     130  130  130  130  130  130  130\n",
       " 131  131  131  131  131  131  131  131     131  131  131  131  131  131  131\n",
       "\n",
       "[:, :, 1, 3] =\n",
       "  -2   -2   -2   -2   -2   -2   -2   -2  …   -2   -2   -2   -2   -2   -2   -2\n",
       "  -1   -1   -1   -1   -1   -1   -1   -1      -1   -1   -1   -1   -1   -1   -1\n",
       "   0    0    0    0    0    0    0    0       0    0    0    0    0    0    0\n",
       "   1    1    1    1    1    1    1    1       1    1    1    1    1    1    1\n",
       "   2    2    2    2    2    2    2    2       2    2    2    2    2    2    2\n",
       "   3    3    3    3    3    3    3    3  …    3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4       4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8  …    8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9       9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 114  114  114  114  114  114  114  114     114  114  114  114  114  114  114\n",
       " 115  115  115  115  115  115  115  115     115  115  115  115  115  115  115\n",
       " 116  116  116  116  116  116  116  116     116  116  116  116  116  116  116\n",
       " 117  117  117  117  117  117  117  117     117  117  117  117  117  117  117\n",
       " 118  118  118  118  118  118  118  118  …  118  118  118  118  118  118  118\n",
       " 119  119  119  119  119  119  119  119     119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123  …  123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124     124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       "\n",
       "[:, :, 2, 3] =\n",
       "  -1   -1   -1   -1   -1   -1   -1   -1  …   -1   -1   -1   -1   -1   -1   -1\n",
       "   0    0    0    0    0    0    0    0       0    0    0    0    0    0    0\n",
       "   1    1    1    1    1    1    1    1       1    1    1    1    1    1    1\n",
       "   2    2    2    2    2    2    2    2       2    2    2    2    2    2    2\n",
       "   3    3    3    3    3    3    3    3       3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4  …    4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8       8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9  …    9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11      11   11   11   11   11   11   11\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 115  115  115  115  115  115  115  115     115  115  115  115  115  115  115\n",
       " 116  116  116  116  116  116  116  116     116  116  116  116  116  116  116\n",
       " 117  117  117  117  117  117  117  117     117  117  117  117  117  117  117\n",
       " 118  118  118  118  118  118  118  118     118  118  118  118  118  118  118\n",
       " 119  119  119  119  119  119  119  119  …  119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123     123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124  …  124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126     126  126  126  126  126  126  126\n",
       "\n",
       "[:, :, 3, 3] =\n",
       "   0    0    0    0    0    0    0    0  …    0    0    0    0    0    0    0\n",
       "   1    1    1    1    1    1    1    1       1    1    1    1    1    1    1\n",
       "   2    2    2    2    2    2    2    2       2    2    2    2    2    2    2\n",
       "   3    3    3    3    3    3    3    3       3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4       4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5  …    5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8       8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9       9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10  …   10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11      11   11   11   11   11   11   11\n",
       "  12   12   12   12   12   12   12   12      12   12   12   12   12   12   12\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 116  116  116  116  116  116  116  116     116  116  116  116  116  116  116\n",
       " 117  117  117  117  117  117  117  117     117  117  117  117  117  117  117\n",
       " 118  118  118  118  118  118  118  118     118  118  118  118  118  118  118\n",
       " 119  119  119  119  119  119  119  119     119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120  …  120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123     123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124     124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125  …  125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126     126  126  126  126  126  126  126\n",
       " 127  127  127  127  127  127  127  127     127  127  127  127  127  127  127\n",
       "\n",
       "[:, :, 4, 3] =\n",
       "   1    1    1    1    1    1    1    1  …    1    1    1    1    1    1    1\n",
       "   2    2    2    2    2    2    2    2       2    2    2    2    2    2    2\n",
       "   3    3    3    3    3    3    3    3       3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4       4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6  …    6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8       8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9       9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11  …   11   11   11   11   11   11   11\n",
       "  12   12   12   12   12   12   12   12      12   12   12   12   12   12   12\n",
       "  13   13   13   13   13   13   13   13      13   13   13   13   13   13   13\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 117  117  117  117  117  117  117  117     117  117  117  117  117  117  117\n",
       " 118  118  118  118  118  118  118  118     118  118  118  118  118  118  118\n",
       " 119  119  119  119  119  119  119  119     119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121  …  121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123     123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124     124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126  …  126  126  126  126  126  126  126\n",
       " 127  127  127  127  127  127  127  127     127  127  127  127  127  127  127\n",
       " 128  128  128  128  128  128  128  128     128  128  128  128  128  128  128\n",
       "\n",
       "[:, :, 5, 3] =\n",
       "   2    2    2    2    2    2    2    2  …    2    2    2    2    2    2    2\n",
       "   3    3    3    3    3    3    3    3       3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4       4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7  …    7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8       8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9       9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11      11   11   11   11   11   11   11\n",
       "  12   12   12   12   12   12   12   12  …   12   12   12   12   12   12   12\n",
       "  13   13   13   13   13   13   13   13      13   13   13   13   13   13   13\n",
       "  14   14   14   14   14   14   14   14      14   14   14   14   14   14   14\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 118  118  118  118  118  118  118  118     118  118  118  118  118  118  118\n",
       " 119  119  119  119  119  119  119  119     119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122  …  122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123     123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124     124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126     126  126  126  126  126  126  126\n",
       " 127  127  127  127  127  127  127  127  …  127  127  127  127  127  127  127\n",
       " 128  128  128  128  128  128  128  128     128  128  128  128  128  128  128\n",
       " 129  129  129  129  129  129  129  129     129  129  129  129  129  129  129\n",
       "\n",
       "[:, :, 6, 3] =\n",
       "   3    3    3    3    3    3    3    3  …    3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4       4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8  …    8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9       9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11      11   11   11   11   11   11   11\n",
       "  12   12   12   12   12   12   12   12      12   12   12   12   12   12   12\n",
       "  13   13   13   13   13   13   13   13  …   13   13   13   13   13   13   13\n",
       "  14   14   14   14   14   14   14   14      14   14   14   14   14   14   14\n",
       "  15   15   15   15   15   15   15   15      15   15   15   15   15   15   15\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 119  119  119  119  119  119  119  119     119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123  …  123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124     124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126     126  126  126  126  126  126  126\n",
       " 127  127  127  127  127  127  127  127     127  127  127  127  127  127  127\n",
       " 128  128  128  128  128  128  128  128  …  128  128  128  128  128  128  128\n",
       " 129  129  129  129  129  129  129  129     129  129  129  129  129  129  129\n",
       " 130  130  130  130  130  130  130  130     130  130  130  130  130  130  130\n",
       "\n",
       "[:, :, 7, 3] =\n",
       "   4    4    4    4    4    4    4    4  …    4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8       8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9  …    9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11      11   11   11   11   11   11   11\n",
       "  12   12   12   12   12   12   12   12      12   12   12   12   12   12   12\n",
       "  13   13   13   13   13   13   13   13      13   13   13   13   13   13   13\n",
       "  14   14   14   14   14   14   14   14  …   14   14   14   14   14   14   14\n",
       "  15   15   15   15   15   15   15   15      15   15   15   15   15   15   15\n",
       "  16   16   16   16   16   16   16   16      16   16   16   16   16   16   16\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123     123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124  …  124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126     126  126  126  126  126  126  126\n",
       " 127  127  127  127  127  127  127  127     127  127  127  127  127  127  127\n",
       " 128  128  128  128  128  128  128  128     128  128  128  128  128  128  128\n",
       " 129  129  129  129  129  129  129  129  …  129  129  129  129  129  129  129\n",
       " 130  130  130  130  130  130  130  130     130  130  130  130  130  130  130\n",
       " 131  131  131  131  131  131  131  131     131  131  131  131  131  131  131\n",
       "\n",
       "[:, :, 1, 4] =\n",
       "  -2   -2   -2   -2   -2   -2   -2   -2  …   -2   -2   -2   -2   -2   -2   -2\n",
       "  -1   -1   -1   -1   -1   -1   -1   -1      -1   -1   -1   -1   -1   -1   -1\n",
       "   0    0    0    0    0    0    0    0       0    0    0    0    0    0    0\n",
       "   1    1    1    1    1    1    1    1       1    1    1    1    1    1    1\n",
       "   2    2    2    2    2    2    2    2       2    2    2    2    2    2    2\n",
       "   3    3    3    3    3    3    3    3  …    3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4       4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8  …    8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9       9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 114  114  114  114  114  114  114  114     114  114  114  114  114  114  114\n",
       " 115  115  115  115  115  115  115  115     115  115  115  115  115  115  115\n",
       " 116  116  116  116  116  116  116  116     116  116  116  116  116  116  116\n",
       " 117  117  117  117  117  117  117  117     117  117  117  117  117  117  117\n",
       " 118  118  118  118  118  118  118  118  …  118  118  118  118  118  118  118\n",
       " 119  119  119  119  119  119  119  119     119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123  …  123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124     124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       "\n",
       "[:, :, 2, 4] =\n",
       "  -1   -1   -1   -1   -1   -1   -1   -1  …   -1   -1   -1   -1   -1   -1   -1\n",
       "   0    0    0    0    0    0    0    0       0    0    0    0    0    0    0\n",
       "   1    1    1    1    1    1    1    1       1    1    1    1    1    1    1\n",
       "   2    2    2    2    2    2    2    2       2    2    2    2    2    2    2\n",
       "   3    3    3    3    3    3    3    3       3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4  …    4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8       8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9  …    9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11      11   11   11   11   11   11   11\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 115  115  115  115  115  115  115  115     115  115  115  115  115  115  115\n",
       " 116  116  116  116  116  116  116  116     116  116  116  116  116  116  116\n",
       " 117  117  117  117  117  117  117  117     117  117  117  117  117  117  117\n",
       " 118  118  118  118  118  118  118  118     118  118  118  118  118  118  118\n",
       " 119  119  119  119  119  119  119  119  …  119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123     123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124  …  124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126     126  126  126  126  126  126  126\n",
       "\n",
       "[:, :, 3, 4] =\n",
       "   0    0    0    0    0    0    0    0  …    0    0    0    0    0    0    0\n",
       "   1    1    1    1    1    1    1    1       1    1    1    1    1    1    1\n",
       "   2    2    2    2    2    2    2    2       2    2    2    2    2    2    2\n",
       "   3    3    3    3    3    3    3    3       3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4       4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5  …    5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8       8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9       9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10  …   10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11      11   11   11   11   11   11   11\n",
       "  12   12   12   12   12   12   12   12      12   12   12   12   12   12   12\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 116  116  116  116  116  116  116  116     116  116  116  116  116  116  116\n",
       " 117  117  117  117  117  117  117  117     117  117  117  117  117  117  117\n",
       " 118  118  118  118  118  118  118  118     118  118  118  118  118  118  118\n",
       " 119  119  119  119  119  119  119  119     119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120  …  120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123     123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124     124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125  …  125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126     126  126  126  126  126  126  126\n",
       " 127  127  127  127  127  127  127  127     127  127  127  127  127  127  127\n",
       "\n",
       "[:, :, 4, 4] =\n",
       "   1    1    1    1    1    1    1    1  …    1    1    1    1    1    1    1\n",
       "   2    2    2    2    2    2    2    2       2    2    2    2    2    2    2\n",
       "   3    3    3    3    3    3    3    3       3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4       4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6  …    6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8       8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9       9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11  …   11   11   11   11   11   11   11\n",
       "  12   12   12   12   12   12   12   12      12   12   12   12   12   12   12\n",
       "  13   13   13   13   13   13   13   13      13   13   13   13   13   13   13\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 117  117  117  117  117  117  117  117     117  117  117  117  117  117  117\n",
       " 118  118  118  118  118  118  118  118     118  118  118  118  118  118  118\n",
       " 119  119  119  119  119  119  119  119     119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121  …  121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123     123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124     124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126  …  126  126  126  126  126  126  126\n",
       " 127  127  127  127  127  127  127  127     127  127  127  127  127  127  127\n",
       " 128  128  128  128  128  128  128  128     128  128  128  128  128  128  128\n",
       "\n",
       "[:, :, 5, 4] =\n",
       "   2    2    2    2    2    2    2    2  …    2    2    2    2    2    2    2\n",
       "   3    3    3    3    3    3    3    3       3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4       4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7  …    7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8       8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9       9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11      11   11   11   11   11   11   11\n",
       "  12   12   12   12   12   12   12   12  …   12   12   12   12   12   12   12\n",
       "  13   13   13   13   13   13   13   13      13   13   13   13   13   13   13\n",
       "  14   14   14   14   14   14   14   14      14   14   14   14   14   14   14\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 118  118  118  118  118  118  118  118     118  118  118  118  118  118  118\n",
       " 119  119  119  119  119  119  119  119     119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122  …  122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123     123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124     124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126     126  126  126  126  126  126  126\n",
       " 127  127  127  127  127  127  127  127  …  127  127  127  127  127  127  127\n",
       " 128  128  128  128  128  128  128  128     128  128  128  128  128  128  128\n",
       " 129  129  129  129  129  129  129  129     129  129  129  129  129  129  129\n",
       "\n",
       "[:, :, 6, 4] =\n",
       "   3    3    3    3    3    3    3    3  …    3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4       4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8  …    8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9       9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11      11   11   11   11   11   11   11\n",
       "  12   12   12   12   12   12   12   12      12   12   12   12   12   12   12\n",
       "  13   13   13   13   13   13   13   13  …   13   13   13   13   13   13   13\n",
       "  14   14   14   14   14   14   14   14      14   14   14   14   14   14   14\n",
       "  15   15   15   15   15   15   15   15      15   15   15   15   15   15   15\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 119  119  119  119  119  119  119  119     119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123  …  123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124     124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126     126  126  126  126  126  126  126\n",
       " 127  127  127  127  127  127  127  127     127  127  127  127  127  127  127\n",
       " 128  128  128  128  128  128  128  128  …  128  128  128  128  128  128  128\n",
       " 129  129  129  129  129  129  129  129     129  129  129  129  129  129  129\n",
       " 130  130  130  130  130  130  130  130     130  130  130  130  130  130  130\n",
       "\n",
       "[:, :, 7, 4] =\n",
       "   4    4    4    4    4    4    4    4  …    4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8       8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9  …    9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11      11   11   11   11   11   11   11\n",
       "  12   12   12   12   12   12   12   12      12   12   12   12   12   12   12\n",
       "  13   13   13   13   13   13   13   13      13   13   13   13   13   13   13\n",
       "  14   14   14   14   14   14   14   14  …   14   14   14   14   14   14   14\n",
       "  15   15   15   15   15   15   15   15      15   15   15   15   15   15   15\n",
       "  16   16   16   16   16   16   16   16      16   16   16   16   16   16   16\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123     123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124  …  124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126     126  126  126  126  126  126  126\n",
       " 127  127  127  127  127  127  127  127     127  127  127  127  127  127  127\n",
       " 128  128  128  128  128  128  128  128     128  128  128  128  128  128  128\n",
       " 129  129  129  129  129  129  129  129  …  129  129  129  129  129  129  129\n",
       " 130  130  130  130  130  130  130  130     130  130  130  130  130  130  130\n",
       " 131  131  131  131  131  131  131  131     131  131  131  131  131  131  131\n",
       "\n",
       "[:, :, 1, 5] =\n",
       "  -2   -2   -2   -2   -2   -2   -2   -2  …   -2   -2   -2   -2   -2   -2   -2\n",
       "  -1   -1   -1   -1   -1   -1   -1   -1      -1   -1   -1   -1   -1   -1   -1\n",
       "   0    0    0    0    0    0    0    0       0    0    0    0    0    0    0\n",
       "   1    1    1    1    1    1    1    1       1    1    1    1    1    1    1\n",
       "   2    2    2    2    2    2    2    2       2    2    2    2    2    2    2\n",
       "   3    3    3    3    3    3    3    3  …    3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4       4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8  …    8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9       9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 114  114  114  114  114  114  114  114     114  114  114  114  114  114  114\n",
       " 115  115  115  115  115  115  115  115     115  115  115  115  115  115  115\n",
       " 116  116  116  116  116  116  116  116     116  116  116  116  116  116  116\n",
       " 117  117  117  117  117  117  117  117     117  117  117  117  117  117  117\n",
       " 118  118  118  118  118  118  118  118  …  118  118  118  118  118  118  118\n",
       " 119  119  119  119  119  119  119  119     119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123  …  123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124     124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       "\n",
       "[:, :, 2, 5] =\n",
       "  -1   -1   -1   -1   -1   -1   -1   -1  …   -1   -1   -1   -1   -1   -1   -1\n",
       "   0    0    0    0    0    0    0    0       0    0    0    0    0    0    0\n",
       "   1    1    1    1    1    1    1    1       1    1    1    1    1    1    1\n",
       "   2    2    2    2    2    2    2    2       2    2    2    2    2    2    2\n",
       "   3    3    3    3    3    3    3    3       3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4  …    4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8       8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9  …    9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11      11   11   11   11   11   11   11\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 115  115  115  115  115  115  115  115     115  115  115  115  115  115  115\n",
       " 116  116  116  116  116  116  116  116     116  116  116  116  116  116  116\n",
       " 117  117  117  117  117  117  117  117     117  117  117  117  117  117  117\n",
       " 118  118  118  118  118  118  118  118     118  118  118  118  118  118  118\n",
       " 119  119  119  119  119  119  119  119  …  119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123     123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124  …  124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126     126  126  126  126  126  126  126\n",
       "\n",
       "[:, :, 3, 5] =\n",
       "   0    0    0    0    0    0    0    0  …    0    0    0    0    0    0    0\n",
       "   1    1    1    1    1    1    1    1       1    1    1    1    1    1    1\n",
       "   2    2    2    2    2    2    2    2       2    2    2    2    2    2    2\n",
       "   3    3    3    3    3    3    3    3       3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4       4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5  …    5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8       8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9       9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10  …   10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11      11   11   11   11   11   11   11\n",
       "  12   12   12   12   12   12   12   12      12   12   12   12   12   12   12\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 116  116  116  116  116  116  116  116     116  116  116  116  116  116  116\n",
       " 117  117  117  117  117  117  117  117     117  117  117  117  117  117  117\n",
       " 118  118  118  118  118  118  118  118     118  118  118  118  118  118  118\n",
       " 119  119  119  119  119  119  119  119     119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120  …  120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123     123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124     124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125  …  125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126     126  126  126  126  126  126  126\n",
       " 127  127  127  127  127  127  127  127     127  127  127  127  127  127  127\n",
       "\n",
       "[:, :, 4, 5] =\n",
       "   1    1    1    1    1    1    1    1  …    1    1    1    1    1    1    1\n",
       "   2    2    2    2    2    2    2    2       2    2    2    2    2    2    2\n",
       "   3    3    3    3    3    3    3    3       3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4       4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6  …    6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8       8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9       9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11  …   11   11   11   11   11   11   11\n",
       "  12   12   12   12   12   12   12   12      12   12   12   12   12   12   12\n",
       "  13   13   13   13   13   13   13   13      13   13   13   13   13   13   13\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 117  117  117  117  117  117  117  117     117  117  117  117  117  117  117\n",
       " 118  118  118  118  118  118  118  118     118  118  118  118  118  118  118\n",
       " 119  119  119  119  119  119  119  119     119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121  …  121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123     123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124     124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126  …  126  126  126  126  126  126  126\n",
       " 127  127  127  127  127  127  127  127     127  127  127  127  127  127  127\n",
       " 128  128  128  128  128  128  128  128     128  128  128  128  128  128  128\n",
       "\n",
       "[:, :, 5, 5] =\n",
       "   2    2    2    2    2    2    2    2  …    2    2    2    2    2    2    2\n",
       "   3    3    3    3    3    3    3    3       3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4       4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7  …    7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8       8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9       9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11      11   11   11   11   11   11   11\n",
       "  12   12   12   12   12   12   12   12  …   12   12   12   12   12   12   12\n",
       "  13   13   13   13   13   13   13   13      13   13   13   13   13   13   13\n",
       "  14   14   14   14   14   14   14   14      14   14   14   14   14   14   14\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 118  118  118  118  118  118  118  118     118  118  118  118  118  118  118\n",
       " 119  119  119  119  119  119  119  119     119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122  …  122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123     123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124     124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126     126  126  126  126  126  126  126\n",
       " 127  127  127  127  127  127  127  127  …  127  127  127  127  127  127  127\n",
       " 128  128  128  128  128  128  128  128     128  128  128  128  128  128  128\n",
       " 129  129  129  129  129  129  129  129     129  129  129  129  129  129  129\n",
       "\n",
       "[:, :, 6, 5] =\n",
       "   3    3    3    3    3    3    3    3  …    3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4       4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8  …    8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9       9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11      11   11   11   11   11   11   11\n",
       "  12   12   12   12   12   12   12   12      12   12   12   12   12   12   12\n",
       "  13   13   13   13   13   13   13   13  …   13   13   13   13   13   13   13\n",
       "  14   14   14   14   14   14   14   14      14   14   14   14   14   14   14\n",
       "  15   15   15   15   15   15   15   15      15   15   15   15   15   15   15\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 119  119  119  119  119  119  119  119     119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123  …  123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124     124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126     126  126  126  126  126  126  126\n",
       " 127  127  127  127  127  127  127  127     127  127  127  127  127  127  127\n",
       " 128  128  128  128  128  128  128  128  …  128  128  128  128  128  128  128\n",
       " 129  129  129  129  129  129  129  129     129  129  129  129  129  129  129\n",
       " 130  130  130  130  130  130  130  130     130  130  130  130  130  130  130\n",
       "\n",
       "[:, :, 7, 5] =\n",
       "   4    4    4    4    4    4    4    4  …    4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8       8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9  …    9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11      11   11   11   11   11   11   11\n",
       "  12   12   12   12   12   12   12   12      12   12   12   12   12   12   12\n",
       "  13   13   13   13   13   13   13   13      13   13   13   13   13   13   13\n",
       "  14   14   14   14   14   14   14   14  …   14   14   14   14   14   14   14\n",
       "  15   15   15   15   15   15   15   15      15   15   15   15   15   15   15\n",
       "  16   16   16   16   16   16   16   16      16   16   16   16   16   16   16\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123     123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124  …  124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126     126  126  126  126  126  126  126\n",
       " 127  127  127  127  127  127  127  127     127  127  127  127  127  127  127\n",
       " 128  128  128  128  128  128  128  128     128  128  128  128  128  128  128\n",
       " 129  129  129  129  129  129  129  129  …  129  129  129  129  129  129  129\n",
       " 130  130  130  130  130  130  130  130     130  130  130  130  130  130  130\n",
       " 131  131  131  131  131  131  131  131     131  131  131  131  131  131  131\n",
       "\n",
       "[:, :, 1, 6] =\n",
       "  -2   -2   -2   -2   -2   -2   -2   -2  …   -2   -2   -2   -2   -2   -2   -2\n",
       "  -1   -1   -1   -1   -1   -1   -1   -1      -1   -1   -1   -1   -1   -1   -1\n",
       "   0    0    0    0    0    0    0    0       0    0    0    0    0    0    0\n",
       "   1    1    1    1    1    1    1    1       1    1    1    1    1    1    1\n",
       "   2    2    2    2    2    2    2    2       2    2    2    2    2    2    2\n",
       "   3    3    3    3    3    3    3    3  …    3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4       4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8  …    8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9       9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 114  114  114  114  114  114  114  114     114  114  114  114  114  114  114\n",
       " 115  115  115  115  115  115  115  115     115  115  115  115  115  115  115\n",
       " 116  116  116  116  116  116  116  116     116  116  116  116  116  116  116\n",
       " 117  117  117  117  117  117  117  117     117  117  117  117  117  117  117\n",
       " 118  118  118  118  118  118  118  118  …  118  118  118  118  118  118  118\n",
       " 119  119  119  119  119  119  119  119     119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123  …  123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124     124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       "\n",
       "[:, :, 2, 6] =\n",
       "  -1   -1   -1   -1   -1   -1   -1   -1  …   -1   -1   -1   -1   -1   -1   -1\n",
       "   0    0    0    0    0    0    0    0       0    0    0    0    0    0    0\n",
       "   1    1    1    1    1    1    1    1       1    1    1    1    1    1    1\n",
       "   2    2    2    2    2    2    2    2       2    2    2    2    2    2    2\n",
       "   3    3    3    3    3    3    3    3       3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4  …    4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8       8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9  …    9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11      11   11   11   11   11   11   11\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 115  115  115  115  115  115  115  115     115  115  115  115  115  115  115\n",
       " 116  116  116  116  116  116  116  116     116  116  116  116  116  116  116\n",
       " 117  117  117  117  117  117  117  117     117  117  117  117  117  117  117\n",
       " 118  118  118  118  118  118  118  118     118  118  118  118  118  118  118\n",
       " 119  119  119  119  119  119  119  119  …  119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123     123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124  …  124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126     126  126  126  126  126  126  126\n",
       "\n",
       "[:, :, 3, 6] =\n",
       "   0    0    0    0    0    0    0    0  …    0    0    0    0    0    0    0\n",
       "   1    1    1    1    1    1    1    1       1    1    1    1    1    1    1\n",
       "   2    2    2    2    2    2    2    2       2    2    2    2    2    2    2\n",
       "   3    3    3    3    3    3    3    3       3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4       4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5  …    5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8       8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9       9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10  …   10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11      11   11   11   11   11   11   11\n",
       "  12   12   12   12   12   12   12   12      12   12   12   12   12   12   12\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 116  116  116  116  116  116  116  116     116  116  116  116  116  116  116\n",
       " 117  117  117  117  117  117  117  117     117  117  117  117  117  117  117\n",
       " 118  118  118  118  118  118  118  118     118  118  118  118  118  118  118\n",
       " 119  119  119  119  119  119  119  119     119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120  …  120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123     123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124     124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125  …  125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126     126  126  126  126  126  126  126\n",
       " 127  127  127  127  127  127  127  127     127  127  127  127  127  127  127\n",
       "\n",
       "[:, :, 4, 6] =\n",
       "   1    1    1    1    1    1    1    1  …    1    1    1    1    1    1    1\n",
       "   2    2    2    2    2    2    2    2       2    2    2    2    2    2    2\n",
       "   3    3    3    3    3    3    3    3       3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4       4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6  …    6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8       8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9       9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11  …   11   11   11   11   11   11   11\n",
       "  12   12   12   12   12   12   12   12      12   12   12   12   12   12   12\n",
       "  13   13   13   13   13   13   13   13      13   13   13   13   13   13   13\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 117  117  117  117  117  117  117  117     117  117  117  117  117  117  117\n",
       " 118  118  118  118  118  118  118  118     118  118  118  118  118  118  118\n",
       " 119  119  119  119  119  119  119  119     119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121  …  121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123     123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124     124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126  …  126  126  126  126  126  126  126\n",
       " 127  127  127  127  127  127  127  127     127  127  127  127  127  127  127\n",
       " 128  128  128  128  128  128  128  128     128  128  128  128  128  128  128\n",
       "\n",
       "[:, :, 5, 6] =\n",
       "   2    2    2    2    2    2    2    2  …    2    2    2    2    2    2    2\n",
       "   3    3    3    3    3    3    3    3       3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4       4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7  …    7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8       8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9       9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11      11   11   11   11   11   11   11\n",
       "  12   12   12   12   12   12   12   12  …   12   12   12   12   12   12   12\n",
       "  13   13   13   13   13   13   13   13      13   13   13   13   13   13   13\n",
       "  14   14   14   14   14   14   14   14      14   14   14   14   14   14   14\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 118  118  118  118  118  118  118  118     118  118  118  118  118  118  118\n",
       " 119  119  119  119  119  119  119  119     119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122  …  122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123     123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124     124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126     126  126  126  126  126  126  126\n",
       " 127  127  127  127  127  127  127  127  …  127  127  127  127  127  127  127\n",
       " 128  128  128  128  128  128  128  128     128  128  128  128  128  128  128\n",
       " 129  129  129  129  129  129  129  129     129  129  129  129  129  129  129\n",
       "\n",
       "[:, :, 6, 6] =\n",
       "   3    3    3    3    3    3    3    3  …    3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4       4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8  …    8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9       9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11      11   11   11   11   11   11   11\n",
       "  12   12   12   12   12   12   12   12      12   12   12   12   12   12   12\n",
       "  13   13   13   13   13   13   13   13  …   13   13   13   13   13   13   13\n",
       "  14   14   14   14   14   14   14   14      14   14   14   14   14   14   14\n",
       "  15   15   15   15   15   15   15   15      15   15   15   15   15   15   15\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 119  119  119  119  119  119  119  119     119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123  …  123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124     124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126     126  126  126  126  126  126  126\n",
       " 127  127  127  127  127  127  127  127     127  127  127  127  127  127  127\n",
       " 128  128  128  128  128  128  128  128  …  128  128  128  128  128  128  128\n",
       " 129  129  129  129  129  129  129  129     129  129  129  129  129  129  129\n",
       " 130  130  130  130  130  130  130  130     130  130  130  130  130  130  130\n",
       "\n",
       "[:, :, 7, 6] =\n",
       "   4    4    4    4    4    4    4    4  …    4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8       8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9  …    9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11      11   11   11   11   11   11   11\n",
       "  12   12   12   12   12   12   12   12      12   12   12   12   12   12   12\n",
       "  13   13   13   13   13   13   13   13      13   13   13   13   13   13   13\n",
       "  14   14   14   14   14   14   14   14  …   14   14   14   14   14   14   14\n",
       "  15   15   15   15   15   15   15   15      15   15   15   15   15   15   15\n",
       "  16   16   16   16   16   16   16   16      16   16   16   16   16   16   16\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123     123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124  …  124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126     126  126  126  126  126  126  126\n",
       " 127  127  127  127  127  127  127  127     127  127  127  127  127  127  127\n",
       " 128  128  128  128  128  128  128  128     128  128  128  128  128  128  128\n",
       " 129  129  129  129  129  129  129  129  …  129  129  129  129  129  129  129\n",
       " 130  130  130  130  130  130  130  130     130  130  130  130  130  130  130\n",
       " 131  131  131  131  131  131  131  131     131  131  131  131  131  131  131\n",
       "\n",
       "[:, :, 1, 7] =\n",
       "  -2   -2   -2   -2   -2   -2   -2   -2  …   -2   -2   -2   -2   -2   -2   -2\n",
       "  -1   -1   -1   -1   -1   -1   -1   -1      -1   -1   -1   -1   -1   -1   -1\n",
       "   0    0    0    0    0    0    0    0       0    0    0    0    0    0    0\n",
       "   1    1    1    1    1    1    1    1       1    1    1    1    1    1    1\n",
       "   2    2    2    2    2    2    2    2       2    2    2    2    2    2    2\n",
       "   3    3    3    3    3    3    3    3  …    3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4       4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8  …    8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9       9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 114  114  114  114  114  114  114  114     114  114  114  114  114  114  114\n",
       " 115  115  115  115  115  115  115  115     115  115  115  115  115  115  115\n",
       " 116  116  116  116  116  116  116  116     116  116  116  116  116  116  116\n",
       " 117  117  117  117  117  117  117  117     117  117  117  117  117  117  117\n",
       " 118  118  118  118  118  118  118  118  …  118  118  118  118  118  118  118\n",
       " 119  119  119  119  119  119  119  119     119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123  …  123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124     124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       "\n",
       "[:, :, 2, 7] =\n",
       "  -1   -1   -1   -1   -1   -1   -1   -1  …   -1   -1   -1   -1   -1   -1   -1\n",
       "   0    0    0    0    0    0    0    0       0    0    0    0    0    0    0\n",
       "   1    1    1    1    1    1    1    1       1    1    1    1    1    1    1\n",
       "   2    2    2    2    2    2    2    2       2    2    2    2    2    2    2\n",
       "   3    3    3    3    3    3    3    3       3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4  …    4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8       8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9  …    9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11      11   11   11   11   11   11   11\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 115  115  115  115  115  115  115  115     115  115  115  115  115  115  115\n",
       " 116  116  116  116  116  116  116  116     116  116  116  116  116  116  116\n",
       " 117  117  117  117  117  117  117  117     117  117  117  117  117  117  117\n",
       " 118  118  118  118  118  118  118  118     118  118  118  118  118  118  118\n",
       " 119  119  119  119  119  119  119  119  …  119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123     123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124  …  124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126     126  126  126  126  126  126  126\n",
       "\n",
       "[:, :, 3, 7] =\n",
       "   0    0    0    0    0    0    0    0  …    0    0    0    0    0    0    0\n",
       "   1    1    1    1    1    1    1    1       1    1    1    1    1    1    1\n",
       "   2    2    2    2    2    2    2    2       2    2    2    2    2    2    2\n",
       "   3    3    3    3    3    3    3    3       3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4       4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5  …    5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8       8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9       9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10  …   10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11      11   11   11   11   11   11   11\n",
       "  12   12   12   12   12   12   12   12      12   12   12   12   12   12   12\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 116  116  116  116  116  116  116  116     116  116  116  116  116  116  116\n",
       " 117  117  117  117  117  117  117  117     117  117  117  117  117  117  117\n",
       " 118  118  118  118  118  118  118  118     118  118  118  118  118  118  118\n",
       " 119  119  119  119  119  119  119  119     119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120  …  120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123     123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124     124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125  …  125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126     126  126  126  126  126  126  126\n",
       " 127  127  127  127  127  127  127  127     127  127  127  127  127  127  127\n",
       "\n",
       "[:, :, 4, 7] =\n",
       "   1    1    1    1    1    1    1    1  …    1    1    1    1    1    1    1\n",
       "   2    2    2    2    2    2    2    2       2    2    2    2    2    2    2\n",
       "   3    3    3    3    3    3    3    3       3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4       4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6  …    6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8       8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9       9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11  …   11   11   11   11   11   11   11\n",
       "  12   12   12   12   12   12   12   12      12   12   12   12   12   12   12\n",
       "  13   13   13   13   13   13   13   13      13   13   13   13   13   13   13\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 117  117  117  117  117  117  117  117     117  117  117  117  117  117  117\n",
       " 118  118  118  118  118  118  118  118     118  118  118  118  118  118  118\n",
       " 119  119  119  119  119  119  119  119     119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121  …  121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123     123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124     124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126  …  126  126  126  126  126  126  126\n",
       " 127  127  127  127  127  127  127  127     127  127  127  127  127  127  127\n",
       " 128  128  128  128  128  128  128  128     128  128  128  128  128  128  128\n",
       "\n",
       "[:, :, 5, 7] =\n",
       "   2    2    2    2    2    2    2    2  …    2    2    2    2    2    2    2\n",
       "   3    3    3    3    3    3    3    3       3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4       4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7  …    7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8       8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9       9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11      11   11   11   11   11   11   11\n",
       "  12   12   12   12   12   12   12   12  …   12   12   12   12   12   12   12\n",
       "  13   13   13   13   13   13   13   13      13   13   13   13   13   13   13\n",
       "  14   14   14   14   14   14   14   14      14   14   14   14   14   14   14\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 118  118  118  118  118  118  118  118     118  118  118  118  118  118  118\n",
       " 119  119  119  119  119  119  119  119     119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122  …  122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123     123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124     124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126     126  126  126  126  126  126  126\n",
       " 127  127  127  127  127  127  127  127  …  127  127  127  127  127  127  127\n",
       " 128  128  128  128  128  128  128  128     128  128  128  128  128  128  128\n",
       " 129  129  129  129  129  129  129  129     129  129  129  129  129  129  129\n",
       "\n",
       "[:, :, 6, 7] =\n",
       "   3    3    3    3    3    3    3    3  …    3    3    3    3    3    3    3\n",
       "   4    4    4    4    4    4    4    4       4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8  …    8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9       9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11      11   11   11   11   11   11   11\n",
       "  12   12   12   12   12   12   12   12      12   12   12   12   12   12   12\n",
       "  13   13   13   13   13   13   13   13  …   13   13   13   13   13   13   13\n",
       "  14   14   14   14   14   14   14   14      14   14   14   14   14   14   14\n",
       "  15   15   15   15   15   15   15   15      15   15   15   15   15   15   15\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 119  119  119  119  119  119  119  119     119  119  119  119  119  119  119\n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123  …  123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124     124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126     126  126  126  126  126  126  126\n",
       " 127  127  127  127  127  127  127  127     127  127  127  127  127  127  127\n",
       " 128  128  128  128  128  128  128  128  …  128  128  128  128  128  128  128\n",
       " 129  129  129  129  129  129  129  129     129  129  129  129  129  129  129\n",
       " 130  130  130  130  130  130  130  130     130  130  130  130  130  130  130\n",
       "\n",
       "[:, :, 7, 7] =\n",
       "   4    4    4    4    4    4    4    4  …    4    4    4    4    4    4    4\n",
       "   5    5    5    5    5    5    5    5       5    5    5    5    5    5    5\n",
       "   6    6    6    6    6    6    6    6       6    6    6    6    6    6    6\n",
       "   7    7    7    7    7    7    7    7       7    7    7    7    7    7    7\n",
       "   8    8    8    8    8    8    8    8       8    8    8    8    8    8    8\n",
       "   9    9    9    9    9    9    9    9  …    9    9    9    9    9    9    9\n",
       "  10   10   10   10   10   10   10   10      10   10   10   10   10   10   10\n",
       "  11   11   11   11   11   11   11   11      11   11   11   11   11   11   11\n",
       "  12   12   12   12   12   12   12   12      12   12   12   12   12   12   12\n",
       "  13   13   13   13   13   13   13   13      13   13   13   13   13   13   13\n",
       "  14   14   14   14   14   14   14   14  …   14   14   14   14   14   14   14\n",
       "  15   15   15   15   15   15   15   15      15   15   15   15   15   15   15\n",
       "  16   16   16   16   16   16   16   16      16   16   16   16   16   16   16\n",
       "   ⋮                        ⋮            ⋱                        ⋮          \n",
       " 120  120  120  120  120  120  120  120     120  120  120  120  120  120  120\n",
       " 121  121  121  121  121  121  121  121     121  121  121  121  121  121  121\n",
       " 122  122  122  122  122  122  122  122     122  122  122  122  122  122  122\n",
       " 123  123  123  123  123  123  123  123     123  123  123  123  123  123  123\n",
       " 124  124  124  124  124  124  124  124  …  124  124  124  124  124  124  124\n",
       " 125  125  125  125  125  125  125  125     125  125  125  125  125  125  125\n",
       " 126  126  126  126  126  126  126  126     126  126  126  126  126  126  126\n",
       " 127  127  127  127  127  127  127  127     127  127  127  127  127  127  127\n",
       " 128  128  128  128  128  128  128  128     128  128  128  128  128  128  128\n",
       " 129  129  129  129  129  129  129  129  …  129  129  129  129  129  129  129\n",
       " 130  130  130  130  130  130  130  130     130  130  130  130  130  130  130\n",
       " 131  131  131  131  131  131  131  131     131  131  131  131  131  131  131"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "include(\"ndgrid.jl\") # Il ne faut pas oublier de mettre ndgrid.jl dans le package et exporter la fonction meshgrid.\n",
    "(X, Y) = meshgrid(1 : n, 1 : n)\n",
    "(dX, dY) = meshgrid(-w : w, -w : w)\n",
    "dX = reshape(dX, (1, 1, w1, w1))\n",
    "dY = reshape(dY, (1, 1, w1, w1))\n",
    "X = repeat(X, inner = [1, 1, w1, w1]) + repeat(dX, inner = [n, n, 1, 1])\n",
    "Y = repeat(Y, inner = [1, 1, w1, w1]) + repeat(dY, inner = [n, n, 1, 1])\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We handle boundary condition by reflexion"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "5376-element Array{Int64,1}:\n",
       " 127\n",
       " 127\n",
       " 127\n",
       " 127\n",
       " 127\n",
       " 127\n",
       " 127\n",
       " 127\n",
       " 127\n",
       " 127\n",
       " 127\n",
       " 127\n",
       " 127\n",
       "   ⋮\n",
       " 127\n",
       " 126\n",
       " 125\n",
       " 127\n",
       " 126\n",
       " 125\n",
       " 127\n",
       " 126\n",
       " 125\n",
       " 127\n",
       " 126\n",
       " 125"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X[X .< 1] = 2 .- X[X .< 1] \n",
    "Y[Y .< 1] = 2 .- Y[Y .< 1]\n",
    "X[X .> n] = 2*n .- X[X .> n]\n",
    "Y[Y .> n] = 2*n .- Y[Y .> n]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Patch extractor operator."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "patch (generic function with 1 method)"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "I = X .+ (Y .- 1)*n\n",
    "for i in 1 : Base.div(n, w)\n",
    "    for j in 1 : Base.div(n, w)\n",
    "        I[i, j, :, :] = transpose(I[i, j, :, :])\n",
    "    end\n",
    "end\n",
    "        \n",
    "#function patch(f)\n",
    "#    P = zeros(n, n, w1, w1)\n",
    "#    for i in 1:length(f[:, 1])\n",
    "#        for j in 1:length(f[1, :])\n",
    "#            P[i, j, :, :] = f[I[i, j, :, :]]\n",
    "#        end\n",
    "#    end\n",
    "#    return P\n",
    "#end\n",
    "function patch(f)\n",
    "    return f[I]\n",
    "end"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Define the patch matrix $P$ of size $(n,n,w_1,w_1)$.\n",
    "Each $P(i,j,:,:)$ represent an $(w_1,w_1)$ patch extracted around pixel\n",
    "$(i,j)$ in the image."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "128×128×7×7 Array{Float64,4}:\n",
       "[:, :, 1, 1] =\n",
       " 0.420562  0.322122  0.298638  0.41914   …  0.428573  0.44862   0.496517\n",
       " 0.302386  0.350857  0.367454  0.421916     0.488337  0.439019  0.463395\n",
       " 0.408959  0.337497  0.375166  0.331575     0.428263  0.459281  0.433398\n",
       " 0.358469  0.338717  0.371062  0.310121     0.505774  0.528377  0.5479  \n",
       " 0.408959  0.337497  0.375166  0.331575     0.428263  0.459281  0.433398\n",
       " 0.302386  0.350857  0.367454  0.421916  …  0.488337  0.439019  0.463395\n",
       " 0.420562  0.322122  0.298638  0.41914      0.428573  0.44862   0.496517\n",
       " 0.314689  0.435777  0.426165  0.384639     0.386027  0.432706  0.462501\n",
       " 0.353112  0.374979  0.379442  0.39974      0.491674  0.4753    0.440267\n",
       " 0.411506  0.357536  0.324394  0.411828     0.382523  0.478623  0.35104 \n",
       " 0.28269   0.387966  0.392045  0.384901  …  0.288662  0.340924  0.319882\n",
       " 0.343019  0.361621  0.385085  0.343604     0.31774   0.343886  0.280956\n",
       " 0.310331  0.372574  0.361015  0.324292     0.231817  0.140795  0.167252\n",
       " ⋮                                       ⋱  ⋮                           \n",
       " 0.311386  0.404238  0.378485  0.391925     0.678462  0.628793  0.558204\n",
       " 0.335594  0.425717  0.348991  0.366602     0.605747  0.701104  0.535028\n",
       " 0.399874  0.34165   0.318672  0.358747     0.676218  0.685461  0.607727\n",
       " 0.400662  0.341967  0.414842  0.401868     0.592331  0.697516  0.758275\n",
       " 0.376995  0.348963  0.397058  0.386276  …  0.643912  0.553756  0.658808\n",
       " 0.46093   0.392605  0.314303  0.369719     0.654399  0.564135  0.66532 \n",
       " 0.394611  0.366495  0.345775  0.415111     0.664339  0.64814   0.638867\n",
       " 0.433254  0.423843  0.420242  0.396661     0.654127  0.646679  0.598602\n",
       " 0.352568  0.335699  0.456157  0.515873     0.764299  0.699749  0.633151\n",
       " 0.321294  0.37938   0.366803  0.391863  …  0.774945  0.734123  0.750186\n",
       " 0.416089  0.38532   0.400952  0.439291     0.630456  0.653461  0.70249 \n",
       " 0.353797  0.407809  0.443295  0.459563     0.572147  0.629985  0.629921\n",
       "\n",
       "[:, :, 2, 1] =\n",
       " 0.322122  0.298638  0.41914   0.298638  …  0.488337  0.439019  0.463395\n",
       " 0.350857  0.367454  0.421916  0.367454     0.428263  0.459281  0.433398\n",
       " 0.337497  0.375166  0.331575  0.375166     0.505774  0.528377  0.5479  \n",
       " 0.338717  0.371062  0.310121  0.371062     0.428263  0.459281  0.433398\n",
       " 0.337497  0.375166  0.331575  0.375166     0.488337  0.439019  0.463395\n",
       " 0.350857  0.367454  0.421916  0.367454  …  0.428573  0.44862   0.496517\n",
       " 0.322122  0.298638  0.41914   0.298638     0.386027  0.432706  0.462501\n",
       " 0.435777  0.426165  0.384639  0.426165     0.491674  0.4753    0.440267\n",
       " 0.374979  0.379442  0.39974   0.379442     0.382523  0.478623  0.35104 \n",
       " 0.357536  0.324394  0.411828  0.324394     0.288662  0.340924  0.319882\n",
       " 0.387966  0.392045  0.384901  0.392045  …  0.31774   0.343886  0.280956\n",
       " 0.361621  0.385085  0.343604  0.385085     0.231817  0.140795  0.167252\n",
       " 0.372574  0.361015  0.324292  0.361015     0.202242  0.232939  0.217961\n",
       " ⋮                                       ⋱  ⋮                           \n",
       " 0.335594  0.425717  0.348991  0.366602     0.605747  0.701104  0.535028\n",
       " 0.399874  0.34165   0.318672  0.358747     0.676218  0.685461  0.607727\n",
       " 0.400662  0.341967  0.414842  0.401868     0.592331  0.697516  0.758275\n",
       " 0.376995  0.348963  0.397058  0.386276     0.643912  0.553756  0.658808\n",
       " 0.46093   0.392605  0.314303  0.369719  …  0.654399  0.564135  0.66532 \n",
       " 0.394611  0.366495  0.345775  0.415111     0.664339  0.64814   0.638867\n",
       " 0.433254  0.423843  0.420242  0.396661     0.654127  0.646679  0.598602\n",
       " 0.352568  0.335699  0.456157  0.515873     0.764299  0.699749  0.633151\n",
       " 0.321294  0.37938   0.366803  0.391863     0.774945  0.734123  0.750186\n",
       " 0.416089  0.38532   0.400952  0.439291  …  0.630456  0.653461  0.70249 \n",
       " 0.353797  0.407809  0.443295  0.459563     0.572147  0.629985  0.629921\n",
       " 0.36485   0.305613  0.409829  0.404933     0.711711  0.666042  0.657626\n",
       "\n",
       "[:, :, 3, 1] =\n",
       " 0.298638  0.41914   0.298638  0.322122  …  0.428263  0.459281  0.433398\n",
       " 0.367454  0.421916  0.367454  0.350857     0.505774  0.528377  0.5479  \n",
       " 0.375166  0.331575  0.375166  0.337497     0.428263  0.459281  0.433398\n",
       " 0.371062  0.310121  0.371062  0.338717     0.488337  0.439019  0.463395\n",
       " 0.375166  0.331575  0.375166  0.337497     0.428573  0.44862   0.496517\n",
       " 0.367454  0.421916  0.367454  0.350857  …  0.386027  0.432706  0.462501\n",
       " 0.298638  0.41914   0.298638  0.322122     0.491674  0.4753    0.440267\n",
       " 0.426165  0.384639  0.426165  0.435777     0.382523  0.478623  0.35104 \n",
       " 0.379442  0.39974   0.379442  0.374979     0.288662  0.340924  0.319882\n",
       " 0.324394  0.411828  0.324394  0.357536     0.31774   0.343886  0.280956\n",
       " 0.392045  0.384901  0.392045  0.387966  …  0.231817  0.140795  0.167252\n",
       " 0.385085  0.343604  0.385085  0.361621     0.202242  0.232939  0.217961\n",
       " 0.361015  0.324292  0.361015  0.372574     0.149172  0.188722  0.150437\n",
       " ⋮                                       ⋱  ⋮                           \n",
       " 0.399874  0.34165   0.318672  0.358747     0.676218  0.685461  0.607727\n",
       " 0.400662  0.341967  0.414842  0.401868     0.592331  0.697516  0.758275\n",
       " 0.376995  0.348963  0.397058  0.386276     0.643912  0.553756  0.658808\n",
       " 0.46093   0.392605  0.314303  0.369719     0.654399  0.564135  0.66532 \n",
       " 0.394611  0.366495  0.345775  0.415111  …  0.664339  0.64814   0.638867\n",
       " 0.433254  0.423843  0.420242  0.396661     0.654127  0.646679  0.598602\n",
       " 0.352568  0.335699  0.456157  0.515873     0.764299  0.699749  0.633151\n",
       " 0.321294  0.37938   0.366803  0.391863     0.774945  0.734123  0.750186\n",
       " 0.416089  0.38532   0.400952  0.439291     0.630456  0.653461  0.70249 \n",
       " 0.353797  0.407809  0.443295  0.459563  …  0.572147  0.629985  0.629921\n",
       " 0.36485   0.305613  0.409829  0.404933     0.711711  0.666042  0.657626\n",
       " 0.386978  0.443315  0.324649  0.373489     0.630046  0.737301  0.757109\n",
       "\n",
       "[:, :, 4, 1] =\n",
       " 0.41914   0.298638  0.322122  0.420562  …  0.505774   0.528377  0.5479  \n",
       " 0.421916  0.367454  0.350857  0.302386     0.428263   0.459281  0.433398\n",
       " 0.331575  0.375166  0.337497  0.408959     0.488337   0.439019  0.463395\n",
       " 0.310121  0.371062  0.338717  0.358469     0.428573   0.44862   0.496517\n",
       " 0.331575  0.375166  0.337497  0.408959     0.386027   0.432706  0.462501\n",
       " 0.421916  0.367454  0.350857  0.302386  …  0.491674   0.4753    0.440267\n",
       " 0.41914   0.298638  0.322122  0.420562     0.382523   0.478623  0.35104 \n",
       " 0.384639  0.426165  0.435777  0.314689     0.288662   0.340924  0.319882\n",
       " 0.39974   0.379442  0.374979  0.353112     0.31774    0.343886  0.280956\n",
       " 0.411828  0.324394  0.357536  0.411506     0.231817   0.140795  0.167252\n",
       " 0.384901  0.392045  0.387966  0.28269   …  0.202242   0.232939  0.217961\n",
       " 0.343604  0.385085  0.361621  0.343019     0.149172   0.188722  0.150437\n",
       " 0.324292  0.361015  0.372574  0.310331     0.0651754  0.174348  0.272203\n",
       " ⋮                                       ⋱  ⋮                            \n",
       " 0.400662  0.341967  0.414842  0.401868     0.592331   0.697516  0.758275\n",
       " 0.376995  0.348963  0.397058  0.386276     0.643912   0.553756  0.658808\n",
       " 0.46093   0.392605  0.314303  0.369719     0.654399   0.564135  0.66532 \n",
       " 0.394611  0.366495  0.345775  0.415111     0.664339   0.64814   0.638867\n",
       " 0.433254  0.423843  0.420242  0.396661  …  0.654127   0.646679  0.598602\n",
       " 0.352568  0.335699  0.456157  0.515873     0.764299   0.699749  0.633151\n",
       " 0.321294  0.37938   0.366803  0.391863     0.774945   0.734123  0.750186\n",
       " 0.416089  0.38532   0.400952  0.439291     0.630456   0.653461  0.70249 \n",
       " 0.353797  0.407809  0.443295  0.459563     0.572147   0.629985  0.629921\n",
       " 0.36485   0.305613  0.409829  0.404933  …  0.711711   0.666042  0.657626\n",
       " 0.386978  0.443315  0.324649  0.373489     0.630046   0.737301  0.757109\n",
       " 0.353269  0.328793  0.391277  0.366263     0.585115   0.679758  0.694726\n",
       "\n",
       "[:, :, 5, 1] =\n",
       " 0.298638  0.322122  0.420562  0.41304   …  0.428263   0.459281  0.433398\n",
       " 0.367454  0.350857  0.302386  0.334106     0.488337   0.439019  0.463395\n",
       " 0.375166  0.337497  0.408959  0.307268     0.428573   0.44862   0.496517\n",
       " 0.371062  0.338717  0.358469  0.366528     0.386027   0.432706  0.462501\n",
       " 0.375166  0.337497  0.408959  0.307268     0.491674   0.4753    0.440267\n",
       " 0.367454  0.350857  0.302386  0.334106  …  0.382523   0.478623  0.35104 \n",
       " 0.298638  0.322122  0.420562  0.41304      0.288662   0.340924  0.319882\n",
       " 0.426165  0.435777  0.314689  0.329906     0.31774    0.343886  0.280956\n",
       " 0.379442  0.374979  0.353112  0.381944     0.231817   0.140795  0.167252\n",
       " 0.324394  0.357536  0.411506  0.352719     0.202242   0.232939  0.217961\n",
       " 0.392045  0.387966  0.28269   0.377169  …  0.149172   0.188722  0.150437\n",
       " 0.385085  0.361621  0.343019  0.330815     0.0651754  0.174348  0.272203\n",
       " 0.361015  0.372574  0.310331  0.374705     0.125428   0.182513  0.163869\n",
       " ⋮                                       ⋱  ⋮                            \n",
       " 0.376995  0.348963  0.397058  0.386276     0.643912   0.553756  0.658808\n",
       " 0.46093   0.392605  0.314303  0.369719     0.654399   0.564135  0.66532 \n",
       " 0.394611  0.366495  0.345775  0.415111     0.664339   0.64814   0.638867\n",
       " 0.433254  0.423843  0.420242  0.396661     0.654127   0.646679  0.598602\n",
       " 0.352568  0.335699  0.456157  0.515873  …  0.764299   0.699749  0.633151\n",
       " 0.321294  0.37938   0.366803  0.391863     0.774945   0.734123  0.750186\n",
       " 0.416089  0.38532   0.400952  0.439291     0.630456   0.653461  0.70249 \n",
       " 0.353797  0.407809  0.443295  0.459563     0.572147   0.629985  0.629921\n",
       " 0.36485   0.305613  0.409829  0.404933     0.711711   0.666042  0.657626\n",
       " 0.386978  0.443315  0.324649  0.373489  …  0.630046   0.737301  0.757109\n",
       " 0.353269  0.328793  0.391277  0.366263     0.585115   0.679758  0.694726\n",
       " 0.386978  0.443315  0.324649  0.373489     0.630046   0.737301  0.757109\n",
       "\n",
       "[:, :, 6, 1] =\n",
       " 0.322122  0.420562  0.41304   0.387839  …  0.488337   0.439019  0.463395\n",
       " 0.350857  0.302386  0.334106  0.383443     0.428573   0.44862   0.496517\n",
       " 0.337497  0.408959  0.307268  0.372549     0.386027   0.432706  0.462501\n",
       " 0.338717  0.358469  0.366528  0.301808     0.491674   0.4753    0.440267\n",
       " 0.337497  0.408959  0.307268  0.372549     0.382523   0.478623  0.35104 \n",
       " 0.350857  0.302386  0.334106  0.383443  …  0.288662   0.340924  0.319882\n",
       " 0.322122  0.420562  0.41304   0.387839     0.31774    0.343886  0.280956\n",
       " 0.435777  0.314689  0.329906  0.379633     0.231817   0.140795  0.167252\n",
       " 0.374979  0.353112  0.381944  0.36836      0.202242   0.232939  0.217961\n",
       " 0.357536  0.411506  0.352719  0.337017     0.149172   0.188722  0.150437\n",
       " 0.387966  0.28269   0.377169  0.386111  …  0.0651754  0.174348  0.272203\n",
       " 0.361621  0.343019  0.330815  0.305207     0.125428   0.182513  0.163869\n",
       " 0.372574  0.310331  0.374705  0.38899      0.21213    0.112225  0.147525\n",
       " ⋮                                       ⋱  ⋮                            \n",
       " 0.46093   0.392605  0.314303  0.369719     0.654399   0.564135  0.66532 \n",
       " 0.394611  0.366495  0.345775  0.415111     0.664339   0.64814   0.638867\n",
       " 0.433254  0.423843  0.420242  0.396661     0.654127   0.646679  0.598602\n",
       " 0.352568  0.335699  0.456157  0.515873     0.764299   0.699749  0.633151\n",
       " 0.321294  0.37938   0.366803  0.391863  …  0.774945   0.734123  0.750186\n",
       " 0.416089  0.38532   0.400952  0.439291     0.630456   0.653461  0.70249 \n",
       " 0.353797  0.407809  0.443295  0.459563     0.572147   0.629985  0.629921\n",
       " 0.36485   0.305613  0.409829  0.404933     0.711711   0.666042  0.657626\n",
       " 0.386978  0.443315  0.324649  0.373489     0.630046   0.737301  0.757109\n",
       " 0.353269  0.328793  0.391277  0.366263  …  0.585115   0.679758  0.694726\n",
       " 0.386978  0.443315  0.324649  0.373489     0.630046   0.737301  0.757109\n",
       " 0.36485   0.305613  0.409829  0.404933     0.711711   0.666042  0.657626\n",
       "\n",
       "[:, :, 7, 1] =\n",
       " 0.420562  0.41304   0.387839  0.281307  …  0.428573   0.44862    0.496517\n",
       " 0.302386  0.334106  0.383443  0.422243     0.386027   0.432706   0.462501\n",
       " 0.408959  0.307268  0.372549  0.389746     0.491674   0.4753     0.440267\n",
       " 0.358469  0.366528  0.301808  0.271791     0.382523   0.478623   0.35104 \n",
       " 0.408959  0.307268  0.372549  0.389746     0.288662   0.340924   0.319882\n",
       " 0.302386  0.334106  0.383443  0.422243  …  0.31774    0.343886   0.280956\n",
       " 0.420562  0.41304   0.387839  0.281307     0.231817   0.140795   0.167252\n",
       " 0.314689  0.329906  0.379633  0.375243     0.202242   0.232939   0.217961\n",
       " 0.353112  0.381944  0.36836   0.332942     0.149172   0.188722   0.150437\n",
       " 0.411506  0.352719  0.337017  0.356497     0.0651754  0.174348   0.272203\n",
       " 0.28269   0.377169  0.386111  0.328073  …  0.125428   0.182513   0.163869\n",
       " 0.343019  0.330815  0.305207  0.365998     0.21213    0.112225   0.147525\n",
       " 0.310331  0.374705  0.38899   0.381824     0.115181   0.0791808  0.076828\n",
       " ⋮                                       ⋱  ⋮                             \n",
       " 0.394611  0.366495  0.345775  0.415111     0.664339   0.64814    0.638867\n",
       " 0.433254  0.423843  0.420242  0.396661     0.654127   0.646679   0.598602\n",
       " 0.352568  0.335699  0.456157  0.515873     0.764299   0.699749   0.633151\n",
       " 0.321294  0.37938   0.366803  0.391863     0.774945   0.734123   0.750186\n",
       " 0.416089  0.38532   0.400952  0.439291  …  0.630456   0.653461   0.70249 \n",
       " 0.353797  0.407809  0.443295  0.459563     0.572147   0.629985   0.629921\n",
       " 0.36485   0.305613  0.409829  0.404933     0.711711   0.666042   0.657626\n",
       " 0.386978  0.443315  0.324649  0.373489     0.630046   0.737301   0.757109\n",
       " 0.353269  0.328793  0.391277  0.366263     0.585115   0.679758   0.694726\n",
       " 0.386978  0.443315  0.324649  0.373489  …  0.630046   0.737301   0.757109\n",
       " 0.36485   0.305613  0.409829  0.404933     0.711711   0.666042   0.657626\n",
       " 0.353797  0.407809  0.443295  0.459563     0.572147   0.629985   0.629921\n",
       "\n",
       "[:, :, 1, 2] =\n",
       " 0.302386  0.350857  0.367454  0.421916  …  0.44862   0.496517  0.567202 \n",
       " 0.408959  0.337497  0.375166  0.331575     0.439019  0.463395  0.540048 \n",
       " 0.358469  0.338717  0.371062  0.310121     0.459281  0.433398  0.33134  \n",
       " 0.408959  0.337497  0.375166  0.331575     0.528377  0.5479    0.441499 \n",
       " 0.302386  0.350857  0.367454  0.421916     0.459281  0.433398  0.33134  \n",
       " 0.420562  0.322122  0.298638  0.41914   …  0.439019  0.463395  0.540048 \n",
       " 0.314689  0.435777  0.426165  0.384639     0.44862   0.496517  0.567202 \n",
       " 0.353112  0.374979  0.379442  0.39974      0.432706  0.462501  0.573065 \n",
       " 0.411506  0.357536  0.324394  0.411828     0.4753    0.440267  0.415227 \n",
       " 0.28269   0.387966  0.392045  0.384901     0.478623  0.35104   0.365848 \n",
       " 0.343019  0.361621  0.385085  0.343604  …  0.340924  0.319882  0.25511  \n",
       " 0.310331  0.372574  0.361015  0.324292     0.343886  0.280956  0.16697  \n",
       " 0.360588  0.318909  0.373246  0.374508     0.140795  0.167252  0.0874964\n",
       " ⋮                                       ⋱  ⋮                            \n",
       " 0.404238  0.378485  0.391925  0.378485     0.628793  0.558204  0.607852 \n",
       " 0.425717  0.348991  0.366602  0.348991     0.701104  0.535028  0.520682 \n",
       " 0.34165   0.318672  0.358747  0.318672     0.685461  0.607727  0.557041 \n",
       " 0.341967  0.414842  0.401868  0.414842     0.697516  0.758275  0.738772 \n",
       " 0.348963  0.397058  0.386276  0.397058  …  0.553756  0.658808  0.724762 \n",
       " 0.392605  0.314303  0.369719  0.314303     0.564135  0.66532   0.707989 \n",
       " 0.366495  0.345775  0.415111  0.345775     0.64814   0.638867  0.702256 \n",
       " 0.423843  0.420242  0.396661  0.420242     0.646679  0.598602  0.55668  \n",
       " 0.335699  0.456157  0.515873  0.456157     0.699749  0.633151  0.602578 \n",
       " 0.37938   0.366803  0.391863  0.366803  …  0.734123  0.750186  0.571596 \n",
       " 0.38532   0.400952  0.439291  0.400952     0.653461  0.70249   0.642772 \n",
       " 0.407809  0.443295  0.459563  0.443295     0.629985  0.629921  0.606939 \n",
       "\n",
       "[:, :, 2, 2] =\n",
       " 0.350857  0.367454  0.421916  0.367454  …  0.439019  0.463395  0.540048 \n",
       " 0.337497  0.375166  0.331575  0.375166     0.459281  0.433398  0.33134  \n",
       " 0.338717  0.371062  0.310121  0.371062     0.528377  0.5479    0.441499 \n",
       " 0.337497  0.375166  0.331575  0.375166     0.459281  0.433398  0.33134  \n",
       " 0.350857  0.367454  0.421916  0.367454     0.439019  0.463395  0.540048 \n",
       " 0.322122  0.298638  0.41914   0.298638  …  0.44862   0.496517  0.567202 \n",
       " 0.435777  0.426165  0.384639  0.426165     0.432706  0.462501  0.573065 \n",
       " 0.374979  0.379442  0.39974   0.379442     0.4753    0.440267  0.415227 \n",
       " 0.357536  0.324394  0.411828  0.324394     0.478623  0.35104   0.365848 \n",
       " 0.387966  0.392045  0.384901  0.392045     0.340924  0.319882  0.25511  \n",
       " 0.361621  0.385085  0.343604  0.385085  …  0.343886  0.280956  0.16697  \n",
       " 0.372574  0.361015  0.324292  0.361015     0.140795  0.167252  0.0874964\n",
       " 0.318909  0.373246  0.374508  0.373246     0.232939  0.217961  0.26696  \n",
       " ⋮                                       ⋱  ⋮                            \n",
       " 0.425717  0.348991  0.366602  0.348991     0.701104  0.535028  0.520682 \n",
       " 0.34165   0.318672  0.358747  0.318672     0.685461  0.607727  0.557041 \n",
       " 0.341967  0.414842  0.401868  0.414842     0.697516  0.758275  0.738772 \n",
       " 0.348963  0.397058  0.386276  0.397058     0.553756  0.658808  0.724762 \n",
       " 0.392605  0.314303  0.369719  0.314303  …  0.564135  0.66532   0.707989 \n",
       " 0.366495  0.345775  0.415111  0.345775     0.64814   0.638867  0.702256 \n",
       " 0.423843  0.420242  0.396661  0.420242     0.646679  0.598602  0.55668  \n",
       " 0.335699  0.456157  0.515873  0.456157     0.699749  0.633151  0.602578 \n",
       " 0.37938   0.366803  0.391863  0.366803     0.734123  0.750186  0.571596 \n",
       " 0.38532   0.400952  0.439291  0.400952  …  0.653461  0.70249   0.642772 \n",
       " 0.407809  0.443295  0.459563  0.443295     0.629985  0.629921  0.606939 \n",
       " 0.305613  0.409829  0.404933  0.409829     0.666042  0.657626  0.644774 \n",
       "\n",
       "[:, :, 3, 2] =\n",
       " 0.367454  0.421916  0.367454  0.350857  …  0.459281  0.433398  0.33134  \n",
       " 0.375166  0.331575  0.375166  0.337497     0.528377  0.5479    0.441499 \n",
       " 0.371062  0.310121  0.371062  0.338717     0.459281  0.433398  0.33134  \n",
       " 0.375166  0.331575  0.375166  0.337497     0.439019  0.463395  0.540048 \n",
       " 0.367454  0.421916  0.367454  0.350857     0.44862   0.496517  0.567202 \n",
       " 0.298638  0.41914   0.298638  0.322122  …  0.432706  0.462501  0.573065 \n",
       " 0.426165  0.384639  0.426165  0.435777     0.4753    0.440267  0.415227 \n",
       " 0.379442  0.39974   0.379442  0.374979     0.478623  0.35104   0.365848 \n",
       " 0.324394  0.411828  0.324394  0.357536     0.340924  0.319882  0.25511  \n",
       " 0.392045  0.384901  0.392045  0.387966     0.343886  0.280956  0.16697  \n",
       " 0.385085  0.343604  0.385085  0.361621  …  0.140795  0.167252  0.0874964\n",
       " 0.361015  0.324292  0.361015  0.372574     0.232939  0.217961  0.26696  \n",
       " 0.373246  0.374508  0.373246  0.318909     0.188722  0.150437  0.159322 \n",
       " ⋮                                       ⋱  ⋮                            \n",
       " 0.34165   0.318672  0.358747  0.318672     0.685461  0.607727  0.557041 \n",
       " 0.341967  0.414842  0.401868  0.414842     0.697516  0.758275  0.738772 \n",
       " 0.348963  0.397058  0.386276  0.397058     0.553756  0.658808  0.724762 \n",
       " 0.392605  0.314303  0.369719  0.314303     0.564135  0.66532   0.707989 \n",
       " 0.366495  0.345775  0.415111  0.345775  …  0.64814   0.638867  0.702256 \n",
       " 0.423843  0.420242  0.396661  0.420242     0.646679  0.598602  0.55668  \n",
       " 0.335699  0.456157  0.515873  0.456157     0.699749  0.633151  0.602578 \n",
       " 0.37938   0.366803  0.391863  0.366803     0.734123  0.750186  0.571596 \n",
       " 0.38532   0.400952  0.439291  0.400952     0.653461  0.70249   0.642772 \n",
       " 0.407809  0.443295  0.459563  0.443295  …  0.629985  0.629921  0.606939 \n",
       " 0.305613  0.409829  0.404933  0.409829     0.666042  0.657626  0.644774 \n",
       " 0.443315  0.324649  0.373489  0.324649     0.737301  0.757109  0.634769 \n",
       "\n",
       "[:, :, 4, 2] =\n",
       " 0.421916  0.367454  0.350857  0.302386  …  0.528377  0.5479    0.441499 \n",
       " 0.331575  0.375166  0.337497  0.408959     0.459281  0.433398  0.33134  \n",
       " 0.310121  0.371062  0.338717  0.358469     0.439019  0.463395  0.540048 \n",
       " 0.331575  0.375166  0.337497  0.408959     0.44862   0.496517  0.567202 \n",
       " 0.421916  0.367454  0.350857  0.302386     0.432706  0.462501  0.573065 \n",
       " 0.41914   0.298638  0.322122  0.420562  …  0.4753    0.440267  0.415227 \n",
       " 0.384639  0.426165  0.435777  0.314689     0.478623  0.35104   0.365848 \n",
       " 0.39974   0.379442  0.374979  0.353112     0.340924  0.319882  0.25511  \n",
       " 0.411828  0.324394  0.357536  0.411506     0.343886  0.280956  0.16697  \n",
       " 0.384901  0.392045  0.387966  0.28269      0.140795  0.167252  0.0874964\n",
       " 0.343604  0.385085  0.361621  0.343019  …  0.232939  0.217961  0.26696  \n",
       " 0.324292  0.361015  0.372574  0.310331     0.188722  0.150437  0.159322 \n",
       " 0.374508  0.373246  0.318909  0.360588     0.174348  0.272203  0.16047  \n",
       " ⋮                                       ⋱  ⋮                            \n",
       " 0.341967  0.414842  0.401868  0.414842     0.697516  0.758275  0.738772 \n",
       " 0.348963  0.397058  0.386276  0.397058     0.553756  0.658808  0.724762 \n",
       " 0.392605  0.314303  0.369719  0.314303     0.564135  0.66532   0.707989 \n",
       " 0.366495  0.345775  0.415111  0.345775     0.64814   0.638867  0.702256 \n",
       " 0.423843  0.420242  0.396661  0.420242  …  0.646679  0.598602  0.55668  \n",
       " 0.335699  0.456157  0.515873  0.456157     0.699749  0.633151  0.602578 \n",
       " 0.37938   0.366803  0.391863  0.366803     0.734123  0.750186  0.571596 \n",
       " 0.38532   0.400952  0.439291  0.400952     0.653461  0.70249   0.642772 \n",
       " 0.407809  0.443295  0.459563  0.443295     0.629985  0.629921  0.606939 \n",
       " 0.305613  0.409829  0.404933  0.409829  …  0.666042  0.657626  0.644774 \n",
       " 0.443315  0.324649  0.373489  0.324649     0.737301  0.757109  0.634769 \n",
       " 0.328793  0.391277  0.366263  0.391277     0.679758  0.694726  0.64582  \n",
       "\n",
       "[:, :, 5, 2] =\n",
       " 0.367454  0.350857  0.302386  0.334106  …  0.459281  0.433398  0.33134  \n",
       " 0.375166  0.337497  0.408959  0.307268     0.439019  0.463395  0.540048 \n",
       " 0.371062  0.338717  0.358469  0.366528     0.44862   0.496517  0.567202 \n",
       " 0.375166  0.337497  0.408959  0.307268     0.432706  0.462501  0.573065 \n",
       " 0.367454  0.350857  0.302386  0.334106     0.4753    0.440267  0.415227 \n",
       " 0.298638  0.322122  0.420562  0.41304   …  0.478623  0.35104   0.365848 \n",
       " 0.426165  0.435777  0.314689  0.329906     0.340924  0.319882  0.25511  \n",
       " 0.379442  0.374979  0.353112  0.381944     0.343886  0.280956  0.16697  \n",
       " 0.324394  0.357536  0.411506  0.352719     0.140795  0.167252  0.0874964\n",
       " 0.392045  0.387966  0.28269   0.377169     0.232939  0.217961  0.26696  \n",
       " 0.385085  0.361621  0.343019  0.330815  …  0.188722  0.150437  0.159322 \n",
       " 0.361015  0.372574  0.310331  0.374705     0.174348  0.272203  0.16047  \n",
       " 0.373246  0.318909  0.360588  0.420681     0.182513  0.163869  0.144812 \n",
       " ⋮                                       ⋱  ⋮                            \n",
       " 0.348963  0.397058  0.386276  0.397058     0.553756  0.658808  0.724762 \n",
       " 0.392605  0.314303  0.369719  0.314303     0.564135  0.66532   0.707989 \n",
       " 0.366495  0.345775  0.415111  0.345775     0.64814   0.638867  0.702256 \n",
       " 0.423843  0.420242  0.396661  0.420242     0.646679  0.598602  0.55668  \n",
       " 0.335699  0.456157  0.515873  0.456157  …  0.699749  0.633151  0.602578 \n",
       " 0.37938   0.366803  0.391863  0.366803     0.734123  0.750186  0.571596 \n",
       " 0.38532   0.400952  0.439291  0.400952     0.653461  0.70249   0.642772 \n",
       " 0.407809  0.443295  0.459563  0.443295     0.629985  0.629921  0.606939 \n",
       " 0.305613  0.409829  0.404933  0.409829     0.666042  0.657626  0.644774 \n",
       " 0.443315  0.324649  0.373489  0.324649  …  0.737301  0.757109  0.634769 \n",
       " 0.328793  0.391277  0.366263  0.391277     0.679758  0.694726  0.64582  \n",
       " 0.443315  0.324649  0.373489  0.324649     0.737301  0.757109  0.634769 \n",
       "\n",
       "[:, :, 6, 2] =\n",
       " 0.350857  0.302386  0.334106  0.383443  …  0.439019  0.463395  0.540048 \n",
       " 0.337497  0.408959  0.307268  0.372549     0.44862   0.496517  0.567202 \n",
       " 0.338717  0.358469  0.366528  0.301808     0.432706  0.462501  0.573065 \n",
       " 0.337497  0.408959  0.307268  0.372549     0.4753    0.440267  0.415227 \n",
       " 0.350857  0.302386  0.334106  0.383443     0.478623  0.35104   0.365848 \n",
       " 0.322122  0.420562  0.41304   0.387839  …  0.340924  0.319882  0.25511  \n",
       " 0.435777  0.314689  0.329906  0.379633     0.343886  0.280956  0.16697  \n",
       " 0.374979  0.353112  0.381944  0.36836      0.140795  0.167252  0.0874964\n",
       " 0.357536  0.411506  0.352719  0.337017     0.232939  0.217961  0.26696  \n",
       " 0.387966  0.28269   0.377169  0.386111     0.188722  0.150437  0.159322 \n",
       " 0.361621  0.343019  0.330815  0.305207  …  0.174348  0.272203  0.16047  \n",
       " 0.372574  0.310331  0.374705  0.38899      0.182513  0.163869  0.144812 \n",
       " 0.318909  0.360588  0.420681  0.289038     0.112225  0.147525  0.0775752\n",
       " ⋮                                       ⋱  ⋮                            \n",
       " 0.392605  0.314303  0.369719  0.314303     0.564135  0.66532   0.707989 \n",
       " 0.366495  0.345775  0.415111  0.345775     0.64814   0.638867  0.702256 \n",
       " 0.423843  0.420242  0.396661  0.420242     0.646679  0.598602  0.55668  \n",
       " 0.335699  0.456157  0.515873  0.456157     0.699749  0.633151  0.602578 \n",
       " 0.37938   0.366803  0.391863  0.366803  …  0.734123  0.750186  0.571596 \n",
       " 0.38532   0.400952  0.439291  0.400952     0.653461  0.70249   0.642772 \n",
       " 0.407809  0.443295  0.459563  0.443295     0.629985  0.629921  0.606939 \n",
       " 0.305613  0.409829  0.404933  0.409829     0.666042  0.657626  0.644774 \n",
       " 0.443315  0.324649  0.373489  0.324649     0.737301  0.757109  0.634769 \n",
       " 0.328793  0.391277  0.366263  0.391277  …  0.679758  0.694726  0.64582  \n",
       " 0.443315  0.324649  0.373489  0.324649     0.737301  0.757109  0.634769 \n",
       " 0.305613  0.409829  0.404933  0.409829     0.666042  0.657626  0.644774 \n",
       "\n",
       "[:, :, 7, 2] =\n",
       " 0.302386  0.334106  0.383443  0.422243  …  0.44862    0.496517  0.567202 \n",
       " 0.408959  0.307268  0.372549  0.389746     0.432706   0.462501  0.573065 \n",
       " 0.358469  0.366528  0.301808  0.271791     0.4753     0.440267  0.415227 \n",
       " 0.408959  0.307268  0.372549  0.389746     0.478623   0.35104   0.365848 \n",
       " 0.302386  0.334106  0.383443  0.422243     0.340924   0.319882  0.25511  \n",
       " 0.420562  0.41304   0.387839  0.281307  …  0.343886   0.280956  0.16697  \n",
       " 0.314689  0.329906  0.379633  0.375243     0.140795   0.167252  0.0874964\n",
       " 0.353112  0.381944  0.36836   0.332942     0.232939   0.217961  0.26696  \n",
       " 0.411506  0.352719  0.337017  0.356497     0.188722   0.150437  0.159322 \n",
       " 0.28269   0.377169  0.386111  0.328073     0.174348   0.272203  0.16047  \n",
       " 0.343019  0.330815  0.305207  0.365998  …  0.182513   0.163869  0.144812 \n",
       " 0.310331  0.374705  0.38899   0.381824     0.112225   0.147525  0.0775752\n",
       " 0.360588  0.420681  0.289038  0.321463     0.0791808  0.076828  0.0790067\n",
       " ⋮                                       ⋱  ⋮                             \n",
       " 0.366495  0.345775  0.415111  0.345775     0.64814    0.638867  0.702256 \n",
       " 0.423843  0.420242  0.396661  0.420242     0.646679   0.598602  0.55668  \n",
       " 0.335699  0.456157  0.515873  0.456157     0.699749   0.633151  0.602578 \n",
       " 0.37938   0.366803  0.391863  0.366803     0.734123   0.750186  0.571596 \n",
       " 0.38532   0.400952  0.439291  0.400952  …  0.653461   0.70249   0.642772 \n",
       " 0.407809  0.443295  0.459563  0.443295     0.629985   0.629921  0.606939 \n",
       " 0.305613  0.409829  0.404933  0.409829     0.666042   0.657626  0.644774 \n",
       " 0.443315  0.324649  0.373489  0.324649     0.737301   0.757109  0.634769 \n",
       " 0.328793  0.391277  0.366263  0.391277     0.679758   0.694726  0.64582  \n",
       " 0.443315  0.324649  0.373489  0.324649  …  0.737301   0.757109  0.634769 \n",
       " 0.305613  0.409829  0.404933  0.409829     0.666042   0.657626  0.644774 \n",
       " 0.407809  0.443295  0.459563  0.443295     0.629985   0.629921  0.606939 \n",
       "\n",
       "[:, :, 1, 3] =\n",
       " 0.408959  0.337497  0.375166  0.331575  …  0.496517  0.567202   0.485594\n",
       " 0.358469  0.338717  0.371062  0.310121     0.463395  0.540048   0.467588\n",
       " 0.408959  0.337497  0.375166  0.331575     0.433398  0.33134    0.321467\n",
       " 0.302386  0.350857  0.367454  0.421916     0.5479    0.441499   0.432999\n",
       " 0.420562  0.322122  0.298638  0.41914      0.433398  0.33134    0.321467\n",
       " 0.314689  0.435777  0.426165  0.384639  …  0.463395  0.540048   0.467588\n",
       " 0.353112  0.374979  0.379442  0.39974      0.496517  0.567202   0.485594\n",
       " 0.411506  0.357536  0.324394  0.411828     0.462501  0.573065   0.297878\n",
       " 0.28269   0.387966  0.392045  0.384901     0.440267  0.415227   0.597624\n",
       " 0.343019  0.361621  0.385085  0.343604     0.35104   0.365848   0.503564\n",
       " 0.310331  0.372574  0.361015  0.324292  …  0.319882  0.25511    0.381689\n",
       " 0.360588  0.318909  0.373246  0.374508     0.280956  0.16697    0.169959\n",
       " 0.301081  0.382493  0.355147  0.406666     0.167252  0.0874964  0.151843\n",
       " ⋮                                       ⋱  ⋮                            \n",
       " 0.378485  0.391925  0.378485  0.404238     0.558204  0.607852   0.638634\n",
       " 0.348991  0.366602  0.348991  0.425717     0.535028  0.520682   0.698036\n",
       " 0.318672  0.358747  0.318672  0.34165      0.607727  0.557041   0.577087\n",
       " 0.414842  0.401868  0.414842  0.341967     0.758275  0.738772   0.547219\n",
       " 0.397058  0.386276  0.397058  0.348963  …  0.658808  0.724762   0.63977 \n",
       " 0.314303  0.369719  0.314303  0.392605     0.66532   0.707989   0.595353\n",
       " 0.345775  0.415111  0.345775  0.366495     0.638867  0.702256   0.760678\n",
       " 0.420242  0.396661  0.420242  0.423843     0.598602  0.55668    0.655217\n",
       " 0.456157  0.515873  0.456157  0.335699     0.633151  0.602578   0.684568\n",
       " 0.366803  0.391863  0.366803  0.37938   …  0.750186  0.571596   0.526541\n",
       " 0.400952  0.439291  0.400952  0.38532      0.70249   0.642772   0.524163\n",
       " 0.443295  0.459563  0.443295  0.407809     0.629921  0.606939   0.60571 \n",
       "\n",
       "[:, :, 2, 3] =\n",
       " 0.337497  0.375166  0.331575  0.375166  …  0.463395  0.540048   0.467588\n",
       " 0.338717  0.371062  0.310121  0.371062     0.433398  0.33134    0.321467\n",
       " 0.337497  0.375166  0.331575  0.375166     0.5479    0.441499   0.432999\n",
       " 0.350857  0.367454  0.421916  0.367454     0.433398  0.33134    0.321467\n",
       " 0.322122  0.298638  0.41914   0.298638     0.463395  0.540048   0.467588\n",
       " 0.435777  0.426165  0.384639  0.426165  …  0.496517  0.567202   0.485594\n",
       " 0.374979  0.379442  0.39974   0.379442     0.462501  0.573065   0.297878\n",
       " 0.357536  0.324394  0.411828  0.324394     0.440267  0.415227   0.597624\n",
       " 0.387966  0.392045  0.384901  0.392045     0.35104   0.365848   0.503564\n",
       " 0.361621  0.385085  0.343604  0.385085     0.319882  0.25511    0.381689\n",
       " 0.372574  0.361015  0.324292  0.361015  …  0.280956  0.16697    0.169959\n",
       " 0.318909  0.373246  0.374508  0.373246     0.167252  0.0874964  0.151843\n",
       " 0.382493  0.355147  0.406666  0.355147     0.217961  0.26696    0.181298\n",
       " ⋮                                       ⋱  ⋮                            \n",
       " 0.348991  0.366602  0.348991  0.425717     0.535028  0.520682   0.698036\n",
       " 0.318672  0.358747  0.318672  0.34165      0.607727  0.557041   0.577087\n",
       " 0.414842  0.401868  0.414842  0.341967     0.758275  0.738772   0.547219\n",
       " 0.397058  0.386276  0.397058  0.348963     0.658808  0.724762   0.63977 \n",
       " 0.314303  0.369719  0.314303  0.392605  …  0.66532   0.707989   0.595353\n",
       " 0.345775  0.415111  0.345775  0.366495     0.638867  0.702256   0.760678\n",
       " 0.420242  0.396661  0.420242  0.423843     0.598602  0.55668    0.655217\n",
       " 0.456157  0.515873  0.456157  0.335699     0.633151  0.602578   0.684568\n",
       " 0.366803  0.391863  0.366803  0.37938      0.750186  0.571596   0.526541\n",
       " 0.400952  0.439291  0.400952  0.38532   …  0.70249   0.642772   0.524163\n",
       " 0.443295  0.459563  0.443295  0.407809     0.629921  0.606939   0.60571 \n",
       " 0.409829  0.404933  0.409829  0.305613     0.657626  0.644774   0.508329\n",
       "\n",
       "[:, :, 3, 3] =\n",
       " 0.375166  0.331575  0.375166  0.337497  …  0.433398  0.33134    0.321467\n",
       " 0.371062  0.310121  0.371062  0.338717     0.5479    0.441499   0.432999\n",
       " 0.375166  0.331575  0.375166  0.337497     0.433398  0.33134    0.321467\n",
       " 0.367454  0.421916  0.367454  0.350857     0.463395  0.540048   0.467588\n",
       " 0.298638  0.41914   0.298638  0.322122     0.496517  0.567202   0.485594\n",
       " 0.426165  0.384639  0.426165  0.435777  …  0.462501  0.573065   0.297878\n",
       " 0.379442  0.39974   0.379442  0.374979     0.440267  0.415227   0.597624\n",
       " 0.324394  0.411828  0.324394  0.357536     0.35104   0.365848   0.503564\n",
       " 0.392045  0.384901  0.392045  0.387966     0.319882  0.25511    0.381689\n",
       " 0.385085  0.343604  0.385085  0.361621     0.280956  0.16697    0.169959\n",
       " 0.361015  0.324292  0.361015  0.372574  …  0.167252  0.0874964  0.151843\n",
       " 0.373246  0.374508  0.373246  0.318909     0.217961  0.26696    0.181298\n",
       " 0.355147  0.406666  0.355147  0.382493     0.150437  0.159322   0.177689\n",
       " ⋮                                       ⋱  ⋮                            \n",
       " 0.318672  0.358747  0.318672  0.34165      0.607727  0.557041   0.577087\n",
       " 0.414842  0.401868  0.414842  0.341967     0.758275  0.738772   0.547219\n",
       " 0.397058  0.386276  0.397058  0.348963     0.658808  0.724762   0.63977 \n",
       " 0.314303  0.369719  0.314303  0.392605     0.66532   0.707989   0.595353\n",
       " 0.345775  0.415111  0.345775  0.366495  …  0.638867  0.702256   0.760678\n",
       " 0.420242  0.396661  0.420242  0.423843     0.598602  0.55668    0.655217\n",
       " 0.456157  0.515873  0.456157  0.335699     0.633151  0.602578   0.684568\n",
       " 0.366803  0.391863  0.366803  0.37938      0.750186  0.571596   0.526541\n",
       " 0.400952  0.439291  0.400952  0.38532      0.70249   0.642772   0.524163\n",
       " 0.443295  0.459563  0.443295  0.407809  …  0.629921  0.606939   0.60571 \n",
       " 0.409829  0.404933  0.409829  0.305613     0.657626  0.644774   0.508329\n",
       " 0.324649  0.373489  0.324649  0.443315     0.757109  0.634769   0.498877\n",
       "\n",
       "[:, :, 4, 3] =\n",
       " 0.331575  0.375166  0.337497  0.408959  …  0.5479    0.441499   0.432999\n",
       " 0.310121  0.371062  0.338717  0.358469     0.433398  0.33134    0.321467\n",
       " 0.331575  0.375166  0.337497  0.408959     0.463395  0.540048   0.467588\n",
       " 0.421916  0.367454  0.350857  0.302386     0.496517  0.567202   0.485594\n",
       " 0.41914   0.298638  0.322122  0.420562     0.462501  0.573065   0.297878\n",
       " 0.384639  0.426165  0.435777  0.314689  …  0.440267  0.415227   0.597624\n",
       " 0.39974   0.379442  0.374979  0.353112     0.35104   0.365848   0.503564\n",
       " 0.411828  0.324394  0.357536  0.411506     0.319882  0.25511    0.381689\n",
       " 0.384901  0.392045  0.387966  0.28269      0.280956  0.16697    0.169959\n",
       " 0.343604  0.385085  0.361621  0.343019     0.167252  0.0874964  0.151843\n",
       " 0.324292  0.361015  0.372574  0.310331  …  0.217961  0.26696    0.181298\n",
       " 0.374508  0.373246  0.318909  0.360588     0.150437  0.159322   0.177689\n",
       " 0.406666  0.355147  0.382493  0.301081     0.272203  0.16047    0.189698\n",
       " ⋮                                       ⋱  ⋮                            \n",
       " 0.414842  0.401868  0.414842  0.341967     0.758275  0.738772   0.547219\n",
       " 0.397058  0.386276  0.397058  0.348963     0.658808  0.724762   0.63977 \n",
       " 0.314303  0.369719  0.314303  0.392605     0.66532   0.707989   0.595353\n",
       " 0.345775  0.415111  0.345775  0.366495     0.638867  0.702256   0.760678\n",
       " 0.420242  0.396661  0.420242  0.423843  …  0.598602  0.55668    0.655217\n",
       " 0.456157  0.515873  0.456157  0.335699     0.633151  0.602578   0.684568\n",
       " 0.366803  0.391863  0.366803  0.37938      0.750186  0.571596   0.526541\n",
       " 0.400952  0.439291  0.400952  0.38532      0.70249   0.642772   0.524163\n",
       " 0.443295  0.459563  0.443295  0.407809     0.629921  0.606939   0.60571 \n",
       " 0.409829  0.404933  0.409829  0.305613  …  0.657626  0.644774   0.508329\n",
       " 0.324649  0.373489  0.324649  0.443315     0.757109  0.634769   0.498877\n",
       " 0.391277  0.366263  0.391277  0.328793     0.694726  0.64582    0.628629\n",
       "\n",
       "[:, :, 5, 3] =\n",
       " 0.375166  0.337497  0.408959  0.307268  …  0.433398  0.33134    0.321467\n",
       " 0.371062  0.338717  0.358469  0.366528     0.463395  0.540048   0.467588\n",
       " 0.375166  0.337497  0.408959  0.307268     0.496517  0.567202   0.485594\n",
       " 0.367454  0.350857  0.302386  0.334106     0.462501  0.573065   0.297878\n",
       " 0.298638  0.322122  0.420562  0.41304      0.440267  0.415227   0.597624\n",
       " 0.426165  0.435777  0.314689  0.329906  …  0.35104   0.365848   0.503564\n",
       " 0.379442  0.374979  0.353112  0.381944     0.319882  0.25511    0.381689\n",
       " 0.324394  0.357536  0.411506  0.352719     0.280956  0.16697    0.169959\n",
       " 0.392045  0.387966  0.28269   0.377169     0.167252  0.0874964  0.151843\n",
       " 0.385085  0.361621  0.343019  0.330815     0.217961  0.26696    0.181298\n",
       " 0.361015  0.372574  0.310331  0.374705  …  0.150437  0.159322   0.177689\n",
       " 0.373246  0.318909  0.360588  0.420681     0.272203  0.16047    0.189698\n",
       " 0.355147  0.382493  0.301081  0.427255     0.163869  0.144812   0.163406\n",
       " ⋮                                       ⋱  ⋮                            \n",
       " 0.397058  0.386276  0.397058  0.348963     0.658808  0.724762   0.63977 \n",
       " 0.314303  0.369719  0.314303  0.392605     0.66532   0.707989   0.595353\n",
       " 0.345775  0.415111  0.345775  0.366495     0.638867  0.702256   0.760678\n",
       " 0.420242  0.396661  0.420242  0.423843     0.598602  0.55668    0.655217\n",
       " 0.456157  0.515873  0.456157  0.335699  …  0.633151  0.602578   0.684568\n",
       " 0.366803  0.391863  0.366803  0.37938      0.750186  0.571596   0.526541\n",
       " 0.400952  0.439291  0.400952  0.38532      0.70249   0.642772   0.524163\n",
       " 0.443295  0.459563  0.443295  0.407809     0.629921  0.606939   0.60571 \n",
       " 0.409829  0.404933  0.409829  0.305613     0.657626  0.644774   0.508329\n",
       " 0.324649  0.373489  0.324649  0.443315  …  0.757109  0.634769   0.498877\n",
       " 0.391277  0.366263  0.391277  0.328793     0.694726  0.64582    0.628629\n",
       " 0.324649  0.373489  0.324649  0.443315     0.757109  0.634769   0.498877\n",
       "\n",
       "[:, :, 6, 3] =\n",
       " 0.337497  0.408959  0.307268  0.372549  …  0.463395  0.540048   0.467588\n",
       " 0.338717  0.358469  0.366528  0.301808     0.496517  0.567202   0.485594\n",
       " 0.337497  0.408959  0.307268  0.372549     0.462501  0.573065   0.297878\n",
       " 0.350857  0.302386  0.334106  0.383443     0.440267  0.415227   0.597624\n",
       " 0.322122  0.420562  0.41304   0.387839     0.35104   0.365848   0.503564\n",
       " 0.435777  0.314689  0.329906  0.379633  …  0.319882  0.25511    0.381689\n",
       " 0.374979  0.353112  0.381944  0.36836      0.280956  0.16697    0.169959\n",
       " 0.357536  0.411506  0.352719  0.337017     0.167252  0.0874964  0.151843\n",
       " 0.387966  0.28269   0.377169  0.386111     0.217961  0.26696    0.181298\n",
       " 0.361621  0.343019  0.330815  0.305207     0.150437  0.159322   0.177689\n",
       " 0.372574  0.310331  0.374705  0.38899   …  0.272203  0.16047    0.189698\n",
       " 0.318909  0.360588  0.420681  0.289038     0.163869  0.144812   0.163406\n",
       " 0.382493  0.301081  0.427255  0.391195     0.147525  0.0775752  0.274339\n",
       " ⋮                                       ⋱  ⋮                            \n",
       " 0.314303  0.369719  0.314303  0.392605     0.66532   0.707989   0.595353\n",
       " 0.345775  0.415111  0.345775  0.366495     0.638867  0.702256   0.760678\n",
       " 0.420242  0.396661  0.420242  0.423843     0.598602  0.55668    0.655217\n",
       " 0.456157  0.515873  0.456157  0.335699     0.633151  0.602578   0.684568\n",
       " 0.366803  0.391863  0.366803  0.37938   …  0.750186  0.571596   0.526541\n",
       " 0.400952  0.439291  0.400952  0.38532      0.70249   0.642772   0.524163\n",
       " 0.443295  0.459563  0.443295  0.407809     0.629921  0.606939   0.60571 \n",
       " 0.409829  0.404933  0.409829  0.305613     0.657626  0.644774   0.508329\n",
       " 0.324649  0.373489  0.324649  0.443315     0.757109  0.634769   0.498877\n",
       " 0.391277  0.366263  0.391277  0.328793  …  0.694726  0.64582    0.628629\n",
       " 0.324649  0.373489  0.324649  0.443315     0.757109  0.634769   0.498877\n",
       " 0.409829  0.404933  0.409829  0.305613     0.657626  0.644774   0.508329\n",
       "\n",
       "[:, :, 7, 3] =\n",
       " 0.408959  0.307268  0.372549  0.389746  …  0.496517  0.567202   0.485594\n",
       " 0.358469  0.366528  0.301808  0.271791     0.462501  0.573065   0.297878\n",
       " 0.408959  0.307268  0.372549  0.389746     0.440267  0.415227   0.597624\n",
       " 0.302386  0.334106  0.383443  0.422243     0.35104   0.365848   0.503564\n",
       " 0.420562  0.41304   0.387839  0.281307     0.319882  0.25511    0.381689\n",
       " 0.314689  0.329906  0.379633  0.375243  …  0.280956  0.16697    0.169959\n",
       " 0.353112  0.381944  0.36836   0.332942     0.167252  0.0874964  0.151843\n",
       " 0.411506  0.352719  0.337017  0.356497     0.217961  0.26696    0.181298\n",
       " 0.28269   0.377169  0.386111  0.328073     0.150437  0.159322   0.177689\n",
       " 0.343019  0.330815  0.305207  0.365998     0.272203  0.16047    0.189698\n",
       " 0.310331  0.374705  0.38899   0.381824  …  0.163869  0.144812   0.163406\n",
       " 0.360588  0.420681  0.289038  0.321463     0.147525  0.0775752  0.274339\n",
       " 0.301081  0.427255  0.391195  0.385622     0.076828  0.0790067  0.12026 \n",
       " ⋮                                       ⋱  ⋮                            \n",
       " 0.345775  0.415111  0.345775  0.366495     0.638867  0.702256   0.760678\n",
       " 0.420242  0.396661  0.420242  0.423843     0.598602  0.55668    0.655217\n",
       " 0.456157  0.515873  0.456157  0.335699     0.633151  0.602578   0.684568\n",
       " 0.366803  0.391863  0.366803  0.37938      0.750186  0.571596   0.526541\n",
       " 0.400952  0.439291  0.400952  0.38532   …  0.70249   0.642772   0.524163\n",
       " 0.443295  0.459563  0.443295  0.407809     0.629921  0.606939   0.60571 \n",
       " 0.409829  0.404933  0.409829  0.305613     0.657626  0.644774   0.508329\n",
       " 0.324649  0.373489  0.324649  0.443315     0.757109  0.634769   0.498877\n",
       " 0.391277  0.366263  0.391277  0.328793     0.694726  0.64582    0.628629\n",
       " 0.324649  0.373489  0.324649  0.443315  …  0.757109  0.634769   0.498877\n",
       " 0.409829  0.404933  0.409829  0.305613     0.657626  0.644774   0.508329\n",
       " 0.443295  0.459563  0.443295  0.407809     0.629921  0.606939   0.60571 \n",
       "\n",
       "[:, :, 1, 4] =\n",
       " 0.358469  0.338717  0.371062  0.310121  …  0.567202   0.485594  0.291599\n",
       " 0.408959  0.337497  0.375166  0.331575     0.540048   0.467588  0.394749\n",
       " 0.302386  0.350857  0.367454  0.421916     0.33134    0.321467  0.480376\n",
       " 0.420562  0.322122  0.298638  0.41914      0.441499   0.432999  0.460332\n",
       " 0.314689  0.435777  0.426165  0.384639     0.33134    0.321467  0.480376\n",
       " 0.353112  0.374979  0.379442  0.39974   …  0.540048   0.467588  0.394749\n",
       " 0.411506  0.357536  0.324394  0.411828     0.567202   0.485594  0.291599\n",
       " 0.28269   0.387966  0.392045  0.384901     0.573065   0.297878  0.360733\n",
       " 0.343019  0.361621  0.385085  0.343604     0.415227   0.597624  0.561067\n",
       " 0.310331  0.372574  0.361015  0.324292     0.365848   0.503564  0.354022\n",
       " 0.360588  0.318909  0.373246  0.374508  …  0.25511    0.381689  0.28482 \n",
       " 0.301081  0.382493  0.355147  0.406666     0.16697    0.169959  0.289525\n",
       " 0.316193  0.33284   0.357185  0.293279     0.0874964  0.151843  0.324038\n",
       " ⋮                                       ⋱  ⋮                            \n",
       " 0.391925  0.378485  0.404238  0.311386     0.607852   0.638634  0.57404 \n",
       " 0.366602  0.348991  0.425717  0.335594     0.520682   0.698036  0.709138\n",
       " 0.358747  0.318672  0.34165   0.399874     0.557041   0.577087  0.592634\n",
       " 0.401868  0.414842  0.341967  0.400662     0.738772   0.547219  0.586609\n",
       " 0.386276  0.397058  0.348963  0.376995  …  0.724762   0.63977   0.560576\n",
       " 0.369719  0.314303  0.392605  0.46093      0.707989   0.595353  0.678479\n",
       " 0.415111  0.345775  0.366495  0.394611     0.702256   0.760678  0.743508\n",
       " 0.396661  0.420242  0.423843  0.433254     0.55668    0.655217  0.699603\n",
       " 0.515873  0.456157  0.335699  0.352568     0.602578   0.684568  0.659011\n",
       " 0.391863  0.366803  0.37938   0.321294  …  0.571596   0.526541  0.463265\n",
       " 0.439291  0.400952  0.38532   0.416089     0.642772   0.524163  0.464796\n",
       " 0.459563  0.443295  0.407809  0.353797     0.606939   0.60571   0.46617 \n",
       "\n",
       "[:, :, 2, 4] =\n",
       " 0.338717  0.371062  0.310121  0.371062  …  0.540048   0.467588  0.394749\n",
       " 0.337497  0.375166  0.331575  0.375166     0.33134    0.321467  0.480376\n",
       " 0.350857  0.367454  0.421916  0.367454     0.441499   0.432999  0.460332\n",
       " 0.322122  0.298638  0.41914   0.298638     0.33134    0.321467  0.480376\n",
       " 0.435777  0.426165  0.384639  0.426165     0.540048   0.467588  0.394749\n",
       " 0.374979  0.379442  0.39974   0.379442  …  0.567202   0.485594  0.291599\n",
       " 0.357536  0.324394  0.411828  0.324394     0.573065   0.297878  0.360733\n",
       " 0.387966  0.392045  0.384901  0.392045     0.415227   0.597624  0.561067\n",
       " 0.361621  0.385085  0.343604  0.385085     0.365848   0.503564  0.354022\n",
       " 0.372574  0.361015  0.324292  0.361015     0.25511    0.381689  0.28482 \n",
       " 0.318909  0.373246  0.374508  0.373246  …  0.16697    0.169959  0.289525\n",
       " 0.382493  0.355147  0.406666  0.355147     0.0874964  0.151843  0.324038\n",
       " 0.33284   0.357185  0.293279  0.357185     0.26696    0.181298  0.340597\n",
       " ⋮                                       ⋱  ⋮                            \n",
       " 0.366602  0.348991  0.425717  0.335594     0.520682   0.698036  0.709138\n",
       " 0.358747  0.318672  0.34165   0.399874     0.557041   0.577087  0.592634\n",
       " 0.401868  0.414842  0.341967  0.400662     0.738772   0.547219  0.586609\n",
       " 0.386276  0.397058  0.348963  0.376995     0.724762   0.63977   0.560576\n",
       " 0.369719  0.314303  0.392605  0.46093   …  0.707989   0.595353  0.678479\n",
       " 0.415111  0.345775  0.366495  0.394611     0.702256   0.760678  0.743508\n",
       " 0.396661  0.420242  0.423843  0.433254     0.55668    0.655217  0.699603\n",
       " 0.515873  0.456157  0.335699  0.352568     0.602578   0.684568  0.659011\n",
       " 0.391863  0.366803  0.37938   0.321294     0.571596   0.526541  0.463265\n",
       " 0.439291  0.400952  0.38532   0.416089  …  0.642772   0.524163  0.464796\n",
       " 0.459563  0.443295  0.407809  0.353797     0.606939   0.60571   0.46617 \n",
       " 0.404933  0.409829  0.305613  0.36485      0.644774   0.508329  0.494321\n",
       "\n",
       "[:, :, 3, 4] =\n",
       " 0.371062  0.310121  0.371062  0.338717  …  0.33134    0.321467  0.480376\n",
       " 0.375166  0.331575  0.375166  0.337497     0.441499   0.432999  0.460332\n",
       " 0.367454  0.421916  0.367454  0.350857     0.33134    0.321467  0.480376\n",
       " 0.298638  0.41914   0.298638  0.322122     0.540048   0.467588  0.394749\n",
       " 0.426165  0.384639  0.426165  0.435777     0.567202   0.485594  0.291599\n",
       " 0.379442  0.39974   0.379442  0.374979  …  0.573065   0.297878  0.360733\n",
       " 0.324394  0.411828  0.324394  0.357536     0.415227   0.597624  0.561067\n",
       " 0.392045  0.384901  0.392045  0.387966     0.365848   0.503564  0.354022\n",
       " 0.385085  0.343604  0.385085  0.361621     0.25511    0.381689  0.28482 \n",
       " 0.361015  0.324292  0.361015  0.372574     0.16697    0.169959  0.289525\n",
       " 0.373246  0.374508  0.373246  0.318909  …  0.0874964  0.151843  0.324038\n",
       " 0.355147  0.406666  0.355147  0.382493     0.26696    0.181298  0.340597\n",
       " 0.357185  0.293279  0.357185  0.33284      0.159322   0.177689  0.141649\n",
       " ⋮                                       ⋱  ⋮                            \n",
       " 0.358747  0.318672  0.34165   0.399874     0.557041   0.577087  0.592634\n",
       " 0.401868  0.414842  0.341967  0.400662     0.738772   0.547219  0.586609\n",
       " 0.386276  0.397058  0.348963  0.376995     0.724762   0.63977   0.560576\n",
       " 0.369719  0.314303  0.392605  0.46093      0.707989   0.595353  0.678479\n",
       " 0.415111  0.345775  0.366495  0.394611  …  0.702256   0.760678  0.743508\n",
       " 0.396661  0.420242  0.423843  0.433254     0.55668    0.655217  0.699603\n",
       " 0.515873  0.456157  0.335699  0.352568     0.602578   0.684568  0.659011\n",
       " 0.391863  0.366803  0.37938   0.321294     0.571596   0.526541  0.463265\n",
       " 0.439291  0.400952  0.38532   0.416089     0.642772   0.524163  0.464796\n",
       " 0.459563  0.443295  0.407809  0.353797  …  0.606939   0.60571   0.46617 \n",
       " 0.404933  0.409829  0.305613  0.36485      0.644774   0.508329  0.494321\n",
       " 0.373489  0.324649  0.443315  0.386978     0.634769   0.498877  0.5831  \n",
       "\n",
       "[:, :, 4, 4] =\n",
       " 0.310121  0.371062  0.338717  0.358469  …  0.441499   0.432999  0.460332\n",
       " 0.331575  0.375166  0.337497  0.408959     0.33134    0.321467  0.480376\n",
       " 0.421916  0.367454  0.350857  0.302386     0.540048   0.467588  0.394749\n",
       " 0.41914   0.298638  0.322122  0.420562     0.567202   0.485594  0.291599\n",
       " 0.384639  0.426165  0.435777  0.314689     0.573065   0.297878  0.360733\n",
       " 0.39974   0.379442  0.374979  0.353112  …  0.415227   0.597624  0.561067\n",
       " 0.411828  0.324394  0.357536  0.411506     0.365848   0.503564  0.354022\n",
       " 0.384901  0.392045  0.387966  0.28269      0.25511    0.381689  0.28482 \n",
       " 0.343604  0.385085  0.361621  0.343019     0.16697    0.169959  0.289525\n",
       " 0.324292  0.361015  0.372574  0.310331     0.0874964  0.151843  0.324038\n",
       " 0.374508  0.373246  0.318909  0.360588  …  0.26696    0.181298  0.340597\n",
       " 0.406666  0.355147  0.382493  0.301081     0.159322   0.177689  0.141649\n",
       " 0.293279  0.357185  0.33284   0.316193     0.16047    0.189698  0.101248\n",
       " ⋮                                       ⋱  ⋮                            \n",
       " 0.401868  0.414842  0.341967  0.400662     0.738772   0.547219  0.586609\n",
       " 0.386276  0.397058  0.348963  0.376995     0.724762   0.63977   0.560576\n",
       " 0.369719  0.314303  0.392605  0.46093      0.707989   0.595353  0.678479\n",
       " 0.415111  0.345775  0.366495  0.394611     0.702256   0.760678  0.743508\n",
       " 0.396661  0.420242  0.423843  0.433254  …  0.55668    0.655217  0.699603\n",
       " 0.515873  0.456157  0.335699  0.352568     0.602578   0.684568  0.659011\n",
       " 0.391863  0.366803  0.37938   0.321294     0.571596   0.526541  0.463265\n",
       " 0.439291  0.400952  0.38532   0.416089     0.642772   0.524163  0.464796\n",
       " 0.459563  0.443295  0.407809  0.353797     0.606939   0.60571   0.46617 \n",
       " 0.404933  0.409829  0.305613  0.36485   …  0.644774   0.508329  0.494321\n",
       " 0.373489  0.324649  0.443315  0.386978     0.634769   0.498877  0.5831  \n",
       " 0.366263  0.391277  0.328793  0.353269     0.64582    0.628629  0.570823\n",
       "\n",
       "[:, :, 5, 4] =\n",
       " 0.371062  0.338717  0.358469  0.366528  …  0.33134    0.321467  0.480376\n",
       " 0.375166  0.337497  0.408959  0.307268     0.540048   0.467588  0.394749\n",
       " 0.367454  0.350857  0.302386  0.334106     0.567202   0.485594  0.291599\n",
       " 0.298638  0.322122  0.420562  0.41304      0.573065   0.297878  0.360733\n",
       " 0.426165  0.435777  0.314689  0.329906     0.415227   0.597624  0.561067\n",
       " 0.379442  0.374979  0.353112  0.381944  …  0.365848   0.503564  0.354022\n",
       " 0.324394  0.357536  0.411506  0.352719     0.25511    0.381689  0.28482 \n",
       " 0.392045  0.387966  0.28269   0.377169     0.16697    0.169959  0.289525\n",
       " 0.385085  0.361621  0.343019  0.330815     0.0874964  0.151843  0.324038\n",
       " 0.361015  0.372574  0.310331  0.374705     0.26696    0.181298  0.340597\n",
       " 0.373246  0.318909  0.360588  0.420681  …  0.159322   0.177689  0.141649\n",
       " 0.355147  0.382493  0.301081  0.427255     0.16047    0.189698  0.101248\n",
       " 0.357185  0.33284   0.316193  0.286588     0.144812   0.163406  0.288774\n",
       " ⋮                                       ⋱  ⋮                            \n",
       " 0.386276  0.397058  0.348963  0.376995     0.724762   0.63977   0.560576\n",
       " 0.369719  0.314303  0.392605  0.46093      0.707989   0.595353  0.678479\n",
       " 0.415111  0.345775  0.366495  0.394611     0.702256   0.760678  0.743508\n",
       " 0.396661  0.420242  0.423843  0.433254     0.55668    0.655217  0.699603\n",
       " 0.515873  0.456157  0.335699  0.352568  …  0.602578   0.684568  0.659011\n",
       " 0.391863  0.366803  0.37938   0.321294     0.571596   0.526541  0.463265\n",
       " 0.439291  0.400952  0.38532   0.416089     0.642772   0.524163  0.464796\n",
       " 0.459563  0.443295  0.407809  0.353797     0.606939   0.60571   0.46617 \n",
       " 0.404933  0.409829  0.305613  0.36485      0.644774   0.508329  0.494321\n",
       " 0.373489  0.324649  0.443315  0.386978  …  0.634769   0.498877  0.5831  \n",
       " 0.366263  0.391277  0.328793  0.353269     0.64582    0.628629  0.570823\n",
       " 0.373489  0.324649  0.443315  0.386978     0.634769   0.498877  0.5831  \n",
       "\n",
       "[:, :, 6, 4] =\n",
       " 0.338717  0.358469  0.366528  0.301808  …  0.540048   0.467588  0.394749\n",
       " 0.337497  0.408959  0.307268  0.372549     0.567202   0.485594  0.291599\n",
       " 0.350857  0.302386  0.334106  0.383443     0.573065   0.297878  0.360733\n",
       " 0.322122  0.420562  0.41304   0.387839     0.415227   0.597624  0.561067\n",
       " 0.435777  0.314689  0.329906  0.379633     0.365848   0.503564  0.354022\n",
       " 0.374979  0.353112  0.381944  0.36836   …  0.25511    0.381689  0.28482 \n",
       " 0.357536  0.411506  0.352719  0.337017     0.16697    0.169959  0.289525\n",
       " 0.387966  0.28269   0.377169  0.386111     0.0874964  0.151843  0.324038\n",
       " 0.361621  0.343019  0.330815  0.305207     0.26696    0.181298  0.340597\n",
       " 0.372574  0.310331  0.374705  0.38899      0.159322   0.177689  0.141649\n",
       " 0.318909  0.360588  0.420681  0.289038  …  0.16047    0.189698  0.101248\n",
       " 0.382493  0.301081  0.427255  0.391195     0.144812   0.163406  0.288774\n",
       " 0.33284   0.316193  0.286588  0.386774     0.0775752  0.274339  0.268564\n",
       " ⋮                                       ⋱  ⋮                            \n",
       " 0.369719  0.314303  0.392605  0.46093      0.707989   0.595353  0.678479\n",
       " 0.415111  0.345775  0.366495  0.394611     0.702256   0.760678  0.743508\n",
       " 0.396661  0.420242  0.423843  0.433254     0.55668    0.655217  0.699603\n",
       " 0.515873  0.456157  0.335699  0.352568     0.602578   0.684568  0.659011\n",
       " 0.391863  0.366803  0.37938   0.321294  …  0.571596   0.526541  0.463265\n",
       " 0.439291  0.400952  0.38532   0.416089     0.642772   0.524163  0.464796\n",
       " 0.459563  0.443295  0.407809  0.353797     0.606939   0.60571   0.46617 \n",
       " 0.404933  0.409829  0.305613  0.36485      0.644774   0.508329  0.494321\n",
       " 0.373489  0.324649  0.443315  0.386978     0.634769   0.498877  0.5831  \n",
       " 0.366263  0.391277  0.328793  0.353269  …  0.64582    0.628629  0.570823\n",
       " 0.373489  0.324649  0.443315  0.386978     0.634769   0.498877  0.5831  \n",
       " 0.404933  0.409829  0.305613  0.36485      0.644774   0.508329  0.494321\n",
       "\n",
       "[:, :, 7, 4] =\n",
       " 0.358469  0.366528  0.301808  0.271791  …  0.567202   0.485594  0.291599\n",
       " 0.408959  0.307268  0.372549  0.389746     0.573065   0.297878  0.360733\n",
       " 0.302386  0.334106  0.383443  0.422243     0.415227   0.597624  0.561067\n",
       " 0.420562  0.41304   0.387839  0.281307     0.365848   0.503564  0.354022\n",
       " 0.314689  0.329906  0.379633  0.375243     0.25511    0.381689  0.28482 \n",
       " 0.353112  0.381944  0.36836   0.332942  …  0.16697    0.169959  0.289525\n",
       " 0.411506  0.352719  0.337017  0.356497     0.0874964  0.151843  0.324038\n",
       " 0.28269   0.377169  0.386111  0.328073     0.26696    0.181298  0.340597\n",
       " 0.343019  0.330815  0.305207  0.365998     0.159322   0.177689  0.141649\n",
       " 0.310331  0.374705  0.38899   0.381824     0.16047    0.189698  0.101248\n",
       " 0.360588  0.420681  0.289038  0.321463  …  0.144812   0.163406  0.288774\n",
       " 0.301081  0.427255  0.391195  0.385622     0.0775752  0.274339  0.268564\n",
       " 0.316193  0.286588  0.386774  0.359338     0.0790067  0.12026   0.210369\n",
       " ⋮                                       ⋱  ⋮                            \n",
       " 0.415111  0.345775  0.366495  0.394611     0.702256   0.760678  0.743508\n",
       " 0.396661  0.420242  0.423843  0.433254     0.55668    0.655217  0.699603\n",
       " 0.515873  0.456157  0.335699  0.352568     0.602578   0.684568  0.659011\n",
       " 0.391863  0.366803  0.37938   0.321294     0.571596   0.526541  0.463265\n",
       " 0.439291  0.400952  0.38532   0.416089  …  0.642772   0.524163  0.464796\n",
       " 0.459563  0.443295  0.407809  0.353797     0.606939   0.60571   0.46617 \n",
       " 0.404933  0.409829  0.305613  0.36485      0.644774   0.508329  0.494321\n",
       " 0.373489  0.324649  0.443315  0.386978     0.634769   0.498877  0.5831  \n",
       " 0.366263  0.391277  0.328793  0.353269     0.64582    0.628629  0.570823\n",
       " 0.373489  0.324649  0.443315  0.386978  …  0.634769   0.498877  0.5831  \n",
       " 0.404933  0.409829  0.305613  0.36485      0.644774   0.508329  0.494321\n",
       " 0.459563  0.443295  0.407809  0.353797     0.606939   0.60571   0.46617 \n",
       "\n",
       "[:, :, 1, 5] =\n",
       " 0.408959  0.337497  0.375166  0.331575  …  0.485594  0.291599  0.485594\n",
       " 0.302386  0.350857  0.367454  0.421916     0.467588  0.394749  0.467588\n",
       " 0.420562  0.322122  0.298638  0.41914      0.321467  0.480376  0.321467\n",
       " 0.314689  0.435777  0.426165  0.384639     0.432999  0.460332  0.432999\n",
       " 0.353112  0.374979  0.379442  0.39974      0.321467  0.480376  0.321467\n",
       " 0.411506  0.357536  0.324394  0.411828  …  0.467588  0.394749  0.467588\n",
       " 0.28269   0.387966  0.392045  0.384901     0.485594  0.291599  0.485594\n",
       " 0.343019  0.361621  0.385085  0.343604     0.297878  0.360733  0.297878\n",
       " 0.310331  0.372574  0.361015  0.324292     0.597624  0.561067  0.597624\n",
       " 0.360588  0.318909  0.373246  0.374508     0.503564  0.354022  0.503564\n",
       " 0.301081  0.382493  0.355147  0.406666  …  0.381689  0.28482   0.381689\n",
       " 0.316193  0.33284   0.357185  0.293279     0.169959  0.289525  0.169959\n",
       " 0.352788  0.346671  0.335288  0.353417     0.151843  0.324038  0.151843\n",
       " ⋮                                       ⋱  ⋮                           \n",
       " 0.378485  0.404238  0.311386  0.364438     0.638634  0.57404   0.638634\n",
       " 0.348991  0.425717  0.335594  0.356753     0.698036  0.709138  0.698036\n",
       " 0.318672  0.34165   0.399874  0.394656     0.577087  0.592634  0.577087\n",
       " 0.414842  0.341967  0.400662  0.383294     0.547219  0.586609  0.547219\n",
       " 0.397058  0.348963  0.376995  0.321868  …  0.63977   0.560576  0.63977 \n",
       " 0.314303  0.392605  0.46093   0.325514     0.595353  0.678479  0.595353\n",
       " 0.345775  0.366495  0.394611  0.335777     0.760678  0.743508  0.760678\n",
       " 0.420242  0.423843  0.433254  0.393593     0.655217  0.699603  0.655217\n",
       " 0.456157  0.335699  0.352568  0.463578     0.684568  0.659011  0.684568\n",
       " 0.366803  0.37938   0.321294  0.400054  …  0.526541  0.463265  0.526541\n",
       " 0.400952  0.38532   0.416089  0.404387     0.524163  0.464796  0.524163\n",
       " 0.443295  0.407809  0.353797  0.352219     0.60571   0.46617   0.60571 \n",
       "\n",
       "[:, :, 2, 5] =\n",
       " 0.337497  0.375166  0.331575  0.375166  …  0.467588  0.394749  0.467588\n",
       " 0.350857  0.367454  0.421916  0.367454     0.321467  0.480376  0.321467\n",
       " 0.322122  0.298638  0.41914   0.298638     0.432999  0.460332  0.432999\n",
       " 0.435777  0.426165  0.384639  0.426165     0.321467  0.480376  0.321467\n",
       " 0.374979  0.379442  0.39974   0.379442     0.467588  0.394749  0.467588\n",
       " 0.357536  0.324394  0.411828  0.324394  …  0.485594  0.291599  0.485594\n",
       " 0.387966  0.392045  0.384901  0.392045     0.297878  0.360733  0.297878\n",
       " 0.361621  0.385085  0.343604  0.385085     0.597624  0.561067  0.597624\n",
       " 0.372574  0.361015  0.324292  0.361015     0.503564  0.354022  0.503564\n",
       " 0.318909  0.373246  0.374508  0.373246     0.381689  0.28482   0.381689\n",
       " 0.382493  0.355147  0.406666  0.355147  …  0.169959  0.289525  0.169959\n",
       " 0.33284   0.357185  0.293279  0.357185     0.151843  0.324038  0.151843\n",
       " 0.346671  0.335288  0.353417  0.335288     0.181298  0.340597  0.181298\n",
       " ⋮                                       ⋱  ⋮                           \n",
       " 0.348991  0.425717  0.335594  0.356753     0.698036  0.709138  0.698036\n",
       " 0.318672  0.34165   0.399874  0.394656     0.577087  0.592634  0.577087\n",
       " 0.414842  0.341967  0.400662  0.383294     0.547219  0.586609  0.547219\n",
       " 0.397058  0.348963  0.376995  0.321868     0.63977   0.560576  0.63977 \n",
       " 0.314303  0.392605  0.46093   0.325514  …  0.595353  0.678479  0.595353\n",
       " 0.345775  0.366495  0.394611  0.335777     0.760678  0.743508  0.760678\n",
       " 0.420242  0.423843  0.433254  0.393593     0.655217  0.699603  0.655217\n",
       " 0.456157  0.335699  0.352568  0.463578     0.684568  0.659011  0.684568\n",
       " 0.366803  0.37938   0.321294  0.400054     0.526541  0.463265  0.526541\n",
       " 0.400952  0.38532   0.416089  0.404387  …  0.524163  0.464796  0.524163\n",
       " 0.443295  0.407809  0.353797  0.352219     0.60571   0.46617   0.60571 \n",
       " 0.409829  0.305613  0.36485   0.394557     0.508329  0.494321  0.508329\n",
       "\n",
       "[:, :, 3, 5] =\n",
       " 0.375166  0.331575  0.375166  0.337497  …  0.321467  0.480376  0.321467\n",
       " 0.367454  0.421916  0.367454  0.350857     0.432999  0.460332  0.432999\n",
       " 0.298638  0.41914   0.298638  0.322122     0.321467  0.480376  0.321467\n",
       " 0.426165  0.384639  0.426165  0.435777     0.467588  0.394749  0.467588\n",
       " 0.379442  0.39974   0.379442  0.374979     0.485594  0.291599  0.485594\n",
       " 0.324394  0.411828  0.324394  0.357536  …  0.297878  0.360733  0.297878\n",
       " 0.392045  0.384901  0.392045  0.387966     0.597624  0.561067  0.597624\n",
       " 0.385085  0.343604  0.385085  0.361621     0.503564  0.354022  0.503564\n",
       " 0.361015  0.324292  0.361015  0.372574     0.381689  0.28482   0.381689\n",
       " 0.373246  0.374508  0.373246  0.318909     0.169959  0.289525  0.169959\n",
       " 0.355147  0.406666  0.355147  0.382493  …  0.151843  0.324038  0.151843\n",
       " 0.357185  0.293279  0.357185  0.33284      0.181298  0.340597  0.181298\n",
       " 0.335288  0.353417  0.335288  0.346671     0.177689  0.141649  0.177689\n",
       " ⋮                                       ⋱  ⋮                           \n",
       " 0.318672  0.34165   0.399874  0.394656     0.577087  0.592634  0.577087\n",
       " 0.414842  0.341967  0.400662  0.383294     0.547219  0.586609  0.547219\n",
       " 0.397058  0.348963  0.376995  0.321868     0.63977   0.560576  0.63977 \n",
       " 0.314303  0.392605  0.46093   0.325514     0.595353  0.678479  0.595353\n",
       " 0.345775  0.366495  0.394611  0.335777  …  0.760678  0.743508  0.760678\n",
       " 0.420242  0.423843  0.433254  0.393593     0.655217  0.699603  0.655217\n",
       " 0.456157  0.335699  0.352568  0.463578     0.684568  0.659011  0.684568\n",
       " 0.366803  0.37938   0.321294  0.400054     0.526541  0.463265  0.526541\n",
       " 0.400952  0.38532   0.416089  0.404387     0.524163  0.464796  0.524163\n",
       " 0.443295  0.407809  0.353797  0.352219  …  0.60571   0.46617   0.60571 \n",
       " 0.409829  0.305613  0.36485   0.394557     0.508329  0.494321  0.508329\n",
       " 0.324649  0.443315  0.386978  0.403713     0.498877  0.5831    0.498877\n",
       "\n",
       "[:, :, 4, 5] =\n",
       " 0.331575  0.375166  0.337497  0.408959  …  0.432999  0.460332  0.432999\n",
       " 0.421916  0.367454  0.350857  0.302386     0.321467  0.480376  0.321467\n",
       " 0.41914   0.298638  0.322122  0.420562     0.467588  0.394749  0.467588\n",
       " 0.384639  0.426165  0.435777  0.314689     0.485594  0.291599  0.485594\n",
       " 0.39974   0.379442  0.374979  0.353112     0.297878  0.360733  0.297878\n",
       " 0.411828  0.324394  0.357536  0.411506  …  0.597624  0.561067  0.597624\n",
       " 0.384901  0.392045  0.387966  0.28269      0.503564  0.354022  0.503564\n",
       " 0.343604  0.385085  0.361621  0.343019     0.381689  0.28482   0.381689\n",
       " 0.324292  0.361015  0.372574  0.310331     0.169959  0.289525  0.169959\n",
       " 0.374508  0.373246  0.318909  0.360588     0.151843  0.324038  0.151843\n",
       " 0.406666  0.355147  0.382493  0.301081  …  0.181298  0.340597  0.181298\n",
       " 0.293279  0.357185  0.33284   0.316193     0.177689  0.141649  0.177689\n",
       " 0.353417  0.335288  0.346671  0.352788     0.189698  0.101248  0.189698\n",
       " ⋮                                       ⋱  ⋮                           \n",
       " 0.414842  0.341967  0.400662  0.383294     0.547219  0.586609  0.547219\n",
       " 0.397058  0.348963  0.376995  0.321868     0.63977   0.560576  0.63977 \n",
       " 0.314303  0.392605  0.46093   0.325514     0.595353  0.678479  0.595353\n",
       " 0.345775  0.366495  0.394611  0.335777     0.760678  0.743508  0.760678\n",
       " 0.420242  0.423843  0.433254  0.393593  …  0.655217  0.699603  0.655217\n",
       " 0.456157  0.335699  0.352568  0.463578     0.684568  0.659011  0.684568\n",
       " 0.366803  0.37938   0.321294  0.400054     0.526541  0.463265  0.526541\n",
       " 0.400952  0.38532   0.416089  0.404387     0.524163  0.464796  0.524163\n",
       " 0.443295  0.407809  0.353797  0.352219     0.60571   0.46617   0.60571 \n",
       " 0.409829  0.305613  0.36485   0.394557  …  0.508329  0.494321  0.508329\n",
       " 0.324649  0.443315  0.386978  0.403713     0.498877  0.5831    0.498877\n",
       " 0.391277  0.328793  0.353269  0.349689     0.628629  0.570823  0.628629\n",
       "\n",
       "[:, :, 5, 5] =\n",
       " 0.375166  0.337497  0.408959  0.307268  …  0.321467  0.480376  0.321467\n",
       " 0.367454  0.350857  0.302386  0.334106     0.467588  0.394749  0.467588\n",
       " 0.298638  0.322122  0.420562  0.41304      0.485594  0.291599  0.485594\n",
       " 0.426165  0.435777  0.314689  0.329906     0.297878  0.360733  0.297878\n",
       " 0.379442  0.374979  0.353112  0.381944     0.597624  0.561067  0.597624\n",
       " 0.324394  0.357536  0.411506  0.352719  …  0.503564  0.354022  0.503564\n",
       " 0.392045  0.387966  0.28269   0.377169     0.381689  0.28482   0.381689\n",
       " 0.385085  0.361621  0.343019  0.330815     0.169959  0.289525  0.169959\n",
       " 0.361015  0.372574  0.310331  0.374705     0.151843  0.324038  0.151843\n",
       " 0.373246  0.318909  0.360588  0.420681     0.181298  0.340597  0.181298\n",
       " 0.355147  0.382493  0.301081  0.427255  …  0.177689  0.141649  0.177689\n",
       " 0.357185  0.33284   0.316193  0.286588     0.189698  0.101248  0.189698\n",
       " 0.335288  0.346671  0.352788  0.384075     0.163406  0.288774  0.163406\n",
       " ⋮                                       ⋱  ⋮                           \n",
       " 0.397058  0.348963  0.376995  0.321868     0.63977   0.560576  0.63977 \n",
       " 0.314303  0.392605  0.46093   0.325514     0.595353  0.678479  0.595353\n",
       " 0.345775  0.366495  0.394611  0.335777     0.760678  0.743508  0.760678\n",
       " 0.420242  0.423843  0.433254  0.393593     0.655217  0.699603  0.655217\n",
       " 0.456157  0.335699  0.352568  0.463578  …  0.684568  0.659011  0.684568\n",
       " 0.366803  0.37938   0.321294  0.400054     0.526541  0.463265  0.526541\n",
       " 0.400952  0.38532   0.416089  0.404387     0.524163  0.464796  0.524163\n",
       " 0.443295  0.407809  0.353797  0.352219     0.60571   0.46617   0.60571 \n",
       " 0.409829  0.305613  0.36485   0.394557     0.508329  0.494321  0.508329\n",
       " 0.324649  0.443315  0.386978  0.403713  …  0.498877  0.5831    0.498877\n",
       " 0.391277  0.328793  0.353269  0.349689     0.628629  0.570823  0.628629\n",
       " 0.324649  0.443315  0.386978  0.403713     0.498877  0.5831    0.498877\n",
       "\n",
       "[:, :, 6, 5] =\n",
       " 0.337497  0.408959  0.307268  0.372549  …  0.467588  0.394749  0.467588\n",
       " 0.350857  0.302386  0.334106  0.383443     0.485594  0.291599  0.485594\n",
       " 0.322122  0.420562  0.41304   0.387839     0.297878  0.360733  0.297878\n",
       " 0.435777  0.314689  0.329906  0.379633     0.597624  0.561067  0.597624\n",
       " 0.374979  0.353112  0.381944  0.36836      0.503564  0.354022  0.503564\n",
       " 0.357536  0.411506  0.352719  0.337017  …  0.381689  0.28482   0.381689\n",
       " 0.387966  0.28269   0.377169  0.386111     0.169959  0.289525  0.169959\n",
       " 0.361621  0.343019  0.330815  0.305207     0.151843  0.324038  0.151843\n",
       " 0.372574  0.310331  0.374705  0.38899      0.181298  0.340597  0.181298\n",
       " 0.318909  0.360588  0.420681  0.289038     0.177689  0.141649  0.177689\n",
       " 0.382493  0.301081  0.427255  0.391195  …  0.189698  0.101248  0.189698\n",
       " 0.33284   0.316193  0.286588  0.386774     0.163406  0.288774  0.163406\n",
       " 0.346671  0.352788  0.384075  0.353846     0.274339  0.268564  0.274339\n",
       " ⋮                                       ⋱  ⋮                           \n",
       " 0.314303  0.392605  0.46093   0.325514     0.595353  0.678479  0.595353\n",
       " 0.345775  0.366495  0.394611  0.335777     0.760678  0.743508  0.760678\n",
       " 0.420242  0.423843  0.433254  0.393593     0.655217  0.699603  0.655217\n",
       " 0.456157  0.335699  0.352568  0.463578     0.684568  0.659011  0.684568\n",
       " 0.366803  0.37938   0.321294  0.400054  …  0.526541  0.463265  0.526541\n",
       " 0.400952  0.38532   0.416089  0.404387     0.524163  0.464796  0.524163\n",
       " 0.443295  0.407809  0.353797  0.352219     0.60571   0.46617   0.60571 \n",
       " 0.409829  0.305613  0.36485   0.394557     0.508329  0.494321  0.508329\n",
       " 0.324649  0.443315  0.386978  0.403713     0.498877  0.5831    0.498877\n",
       " 0.391277  0.328793  0.353269  0.349689  …  0.628629  0.570823  0.628629\n",
       " 0.324649  0.443315  0.386978  0.403713     0.498877  0.5831    0.498877\n",
       " 0.409829  0.305613  0.36485   0.394557     0.508329  0.494321  0.508329\n",
       "\n",
       "[:, :, 7, 5] =\n",
       " 0.408959  0.307268  0.372549  0.389746  …  0.485594  0.291599  0.485594\n",
       " 0.302386  0.334106  0.383443  0.422243     0.297878  0.360733  0.297878\n",
       " 0.420562  0.41304   0.387839  0.281307     0.597624  0.561067  0.597624\n",
       " 0.314689  0.329906  0.379633  0.375243     0.503564  0.354022  0.503564\n",
       " 0.353112  0.381944  0.36836   0.332942     0.381689  0.28482   0.381689\n",
       " 0.411506  0.352719  0.337017  0.356497  …  0.169959  0.289525  0.169959\n",
       " 0.28269   0.377169  0.386111  0.328073     0.151843  0.324038  0.151843\n",
       " 0.343019  0.330815  0.305207  0.365998     0.181298  0.340597  0.181298\n",
       " 0.310331  0.374705  0.38899   0.381824     0.177689  0.141649  0.177689\n",
       " 0.360588  0.420681  0.289038  0.321463     0.189698  0.101248  0.189698\n",
       " 0.301081  0.427255  0.391195  0.385622  …  0.163406  0.288774  0.163406\n",
       " 0.316193  0.286588  0.386774  0.359338     0.274339  0.268564  0.274339\n",
       " 0.352788  0.384075  0.353846  0.369752     0.12026   0.210369  0.12026 \n",
       " ⋮                                       ⋱  ⋮                           \n",
       " 0.345775  0.366495  0.394611  0.335777     0.760678  0.743508  0.760678\n",
       " 0.420242  0.423843  0.433254  0.393593     0.655217  0.699603  0.655217\n",
       " 0.456157  0.335699  0.352568  0.463578     0.684568  0.659011  0.684568\n",
       " 0.366803  0.37938   0.321294  0.400054     0.526541  0.463265  0.526541\n",
       " 0.400952  0.38532   0.416089  0.404387  …  0.524163  0.464796  0.524163\n",
       " 0.443295  0.407809  0.353797  0.352219     0.60571   0.46617   0.60571 \n",
       " 0.409829  0.305613  0.36485   0.394557     0.508329  0.494321  0.508329\n",
       " 0.324649  0.443315  0.386978  0.403713     0.498877  0.5831    0.498877\n",
       " 0.391277  0.328793  0.353269  0.349689     0.628629  0.570823  0.628629\n",
       " 0.324649  0.443315  0.386978  0.403713  …  0.498877  0.5831    0.498877\n",
       " 0.409829  0.305613  0.36485   0.394557     0.508329  0.494321  0.508329\n",
       " 0.443295  0.407809  0.353797  0.352219     0.60571   0.46617   0.60571 \n",
       "\n",
       "[:, :, 1, 6] =\n",
       " 0.302386  0.350857  0.367454  0.421916  …  0.291599  0.485594  0.567202 \n",
       " 0.420562  0.322122  0.298638  0.41914      0.394749  0.467588  0.540048 \n",
       " 0.314689  0.435777  0.426165  0.384639     0.480376  0.321467  0.33134  \n",
       " 0.353112  0.374979  0.379442  0.39974      0.460332  0.432999  0.441499 \n",
       " 0.411506  0.357536  0.324394  0.411828     0.480376  0.321467  0.33134  \n",
       " 0.28269   0.387966  0.392045  0.384901  …  0.394749  0.467588  0.540048 \n",
       " 0.343019  0.361621  0.385085  0.343604     0.291599  0.485594  0.567202 \n",
       " 0.310331  0.372574  0.361015  0.324292     0.360733  0.297878  0.573065 \n",
       " 0.360588  0.318909  0.373246  0.374508     0.561067  0.597624  0.415227 \n",
       " 0.301081  0.382493  0.355147  0.406666     0.354022  0.503564  0.365848 \n",
       " 0.316193  0.33284   0.357185  0.293279  …  0.28482   0.381689  0.25511  \n",
       " 0.352788  0.346671  0.335288  0.353417     0.289525  0.169959  0.16697  \n",
       " 0.258723  0.332644  0.354601  0.432126     0.324038  0.151843  0.0874964\n",
       " ⋮                                       ⋱  ⋮                            \n",
       " 0.404238  0.311386  0.364438  0.41035      0.57404   0.638634  0.607852 \n",
       " 0.425717  0.335594  0.356753  0.293823     0.709138  0.698036  0.520682 \n",
       " 0.34165   0.399874  0.394656  0.39226      0.592634  0.577087  0.557041 \n",
       " 0.341967  0.400662  0.383294  0.453116     0.586609  0.547219  0.738772 \n",
       " 0.348963  0.376995  0.321868  0.369058  …  0.560576  0.63977   0.724762 \n",
       " 0.392605  0.46093   0.325514  0.415748     0.678479  0.595353  0.707989 \n",
       " 0.366495  0.394611  0.335777  0.346071     0.743508  0.760678  0.702256 \n",
       " 0.423843  0.433254  0.393593  0.383522     0.699603  0.655217  0.55668  \n",
       " 0.335699  0.352568  0.463578  0.350728     0.659011  0.684568  0.602578 \n",
       " 0.37938   0.321294  0.400054  0.367017  …  0.463265  0.526541  0.571596 \n",
       " 0.38532   0.416089  0.404387  0.391201     0.464796  0.524163  0.642772 \n",
       " 0.407809  0.353797  0.352219  0.388549     0.46617   0.60571   0.606939 \n",
       "\n",
       "[:, :, 2, 6] =\n",
       " 0.350857  0.367454  0.421916  0.367454  …  0.394749  0.467588  0.540048 \n",
       " 0.322122  0.298638  0.41914   0.298638     0.480376  0.321467  0.33134  \n",
       " 0.435777  0.426165  0.384639  0.426165     0.460332  0.432999  0.441499 \n",
       " 0.374979  0.379442  0.39974   0.379442     0.480376  0.321467  0.33134  \n",
       " 0.357536  0.324394  0.411828  0.324394     0.394749  0.467588  0.540048 \n",
       " 0.387966  0.392045  0.384901  0.392045  …  0.291599  0.485594  0.567202 \n",
       " 0.361621  0.385085  0.343604  0.385085     0.360733  0.297878  0.573065 \n",
       " 0.372574  0.361015  0.324292  0.361015     0.561067  0.597624  0.415227 \n",
       " 0.318909  0.373246  0.374508  0.373246     0.354022  0.503564  0.365848 \n",
       " 0.382493  0.355147  0.406666  0.355147     0.28482   0.381689  0.25511  \n",
       " 0.33284   0.357185  0.293279  0.357185  …  0.289525  0.169959  0.16697  \n",
       " 0.346671  0.335288  0.353417  0.335288     0.324038  0.151843  0.0874964\n",
       " 0.332644  0.354601  0.432126  0.354601     0.340597  0.181298  0.26696  \n",
       " ⋮                                       ⋱  ⋮                            \n",
       " 0.425717  0.335594  0.356753  0.293823     0.709138  0.698036  0.520682 \n",
       " 0.34165   0.399874  0.394656  0.39226      0.592634  0.577087  0.557041 \n",
       " 0.341967  0.400662  0.383294  0.453116     0.586609  0.547219  0.738772 \n",
       " 0.348963  0.376995  0.321868  0.369058     0.560576  0.63977   0.724762 \n",
       " 0.392605  0.46093   0.325514  0.415748  …  0.678479  0.595353  0.707989 \n",
       " 0.366495  0.394611  0.335777  0.346071     0.743508  0.760678  0.702256 \n",
       " 0.423843  0.433254  0.393593  0.383522     0.699603  0.655217  0.55668  \n",
       " 0.335699  0.352568  0.463578  0.350728     0.659011  0.684568  0.602578 \n",
       " 0.37938   0.321294  0.400054  0.367017     0.463265  0.526541  0.571596 \n",
       " 0.38532   0.416089  0.404387  0.391201  …  0.464796  0.524163  0.642772 \n",
       " 0.407809  0.353797  0.352219  0.388549     0.46617   0.60571   0.606939 \n",
       " 0.305613  0.36485   0.394557  0.337633     0.494321  0.508329  0.644774 \n",
       "\n",
       "[:, :, 3, 6] =\n",
       " 0.367454  0.421916  0.367454  0.350857  …  0.480376  0.321467  0.33134  \n",
       " 0.298638  0.41914   0.298638  0.322122     0.460332  0.432999  0.441499 \n",
       " 0.426165  0.384639  0.426165  0.435777     0.480376  0.321467  0.33134  \n",
       " 0.379442  0.39974   0.379442  0.374979     0.394749  0.467588  0.540048 \n",
       " 0.324394  0.411828  0.324394  0.357536     0.291599  0.485594  0.567202 \n",
       " 0.392045  0.384901  0.392045  0.387966  …  0.360733  0.297878  0.573065 \n",
       " 0.385085  0.343604  0.385085  0.361621     0.561067  0.597624  0.415227 \n",
       " 0.361015  0.324292  0.361015  0.372574     0.354022  0.503564  0.365848 \n",
       " 0.373246  0.374508  0.373246  0.318909     0.28482   0.381689  0.25511  \n",
       " 0.355147  0.406666  0.355147  0.382493     0.289525  0.169959  0.16697  \n",
       " 0.357185  0.293279  0.357185  0.33284   …  0.324038  0.151843  0.0874964\n",
       " 0.335288  0.353417  0.335288  0.346671     0.340597  0.181298  0.26696  \n",
       " 0.354601  0.432126  0.354601  0.332644     0.141649  0.177689  0.159322 \n",
       " ⋮                                       ⋱  ⋮                            \n",
       " 0.34165   0.399874  0.394656  0.39226      0.592634  0.577087  0.557041 \n",
       " 0.341967  0.400662  0.383294  0.453116     0.586609  0.547219  0.738772 \n",
       " 0.348963  0.376995  0.321868  0.369058     0.560576  0.63977   0.724762 \n",
       " 0.392605  0.46093   0.325514  0.415748     0.678479  0.595353  0.707989 \n",
       " 0.366495  0.394611  0.335777  0.346071  …  0.743508  0.760678  0.702256 \n",
       " 0.423843  0.433254  0.393593  0.383522     0.699603  0.655217  0.55668  \n",
       " 0.335699  0.352568  0.463578  0.350728     0.659011  0.684568  0.602578 \n",
       " 0.37938   0.321294  0.400054  0.367017     0.463265  0.526541  0.571596 \n",
       " 0.38532   0.416089  0.404387  0.391201     0.464796  0.524163  0.642772 \n",
       " 0.407809  0.353797  0.352219  0.388549  …  0.46617   0.60571   0.606939 \n",
       " 0.305613  0.36485   0.394557  0.337633     0.494321  0.508329  0.644774 \n",
       " 0.443315  0.386978  0.403713  0.398922     0.5831    0.498877  0.634769 \n",
       "\n",
       "[:, :, 4, 6] =\n",
       " 0.421916  0.367454  0.350857  0.302386  …  0.460332  0.432999  0.441499 \n",
       " 0.41914   0.298638  0.322122  0.420562     0.480376  0.321467  0.33134  \n",
       " 0.384639  0.426165  0.435777  0.314689     0.394749  0.467588  0.540048 \n",
       " 0.39974   0.379442  0.374979  0.353112     0.291599  0.485594  0.567202 \n",
       " 0.411828  0.324394  0.357536  0.411506     0.360733  0.297878  0.573065 \n",
       " 0.384901  0.392045  0.387966  0.28269   …  0.561067  0.597624  0.415227 \n",
       " 0.343604  0.385085  0.361621  0.343019     0.354022  0.503564  0.365848 \n",
       " 0.324292  0.361015  0.372574  0.310331     0.28482   0.381689  0.25511  \n",
       " 0.374508  0.373246  0.318909  0.360588     0.289525  0.169959  0.16697  \n",
       " 0.406666  0.355147  0.382493  0.301081     0.324038  0.151843  0.0874964\n",
       " 0.293279  0.357185  0.33284   0.316193  …  0.340597  0.181298  0.26696  \n",
       " 0.353417  0.335288  0.346671  0.352788     0.141649  0.177689  0.159322 \n",
       " 0.432126  0.354601  0.332644  0.258723     0.101248  0.189698  0.16047  \n",
       " ⋮                                       ⋱  ⋮                            \n",
       " 0.341967  0.400662  0.383294  0.453116     0.586609  0.547219  0.738772 \n",
       " 0.348963  0.376995  0.321868  0.369058     0.560576  0.63977   0.724762 \n",
       " 0.392605  0.46093   0.325514  0.415748     0.678479  0.595353  0.707989 \n",
       " 0.366495  0.394611  0.335777  0.346071     0.743508  0.760678  0.702256 \n",
       " 0.423843  0.433254  0.393593  0.383522  …  0.699603  0.655217  0.55668  \n",
       " 0.335699  0.352568  0.463578  0.350728     0.659011  0.684568  0.602578 \n",
       " 0.37938   0.321294  0.400054  0.367017     0.463265  0.526541  0.571596 \n",
       " 0.38532   0.416089  0.404387  0.391201     0.464796  0.524163  0.642772 \n",
       " 0.407809  0.353797  0.352219  0.388549     0.46617   0.60571   0.606939 \n",
       " 0.305613  0.36485   0.394557  0.337633  …  0.494321  0.508329  0.644774 \n",
       " 0.443315  0.386978  0.403713  0.398922     0.5831    0.498877  0.634769 \n",
       " 0.328793  0.353269  0.349689  0.343062     0.570823  0.628629  0.64582  \n",
       "\n",
       "[:, :, 5, 6] =\n",
       " 0.367454  0.350857  0.302386  0.334106  …  0.480376  0.321467  0.33134  \n",
       " 0.298638  0.322122  0.420562  0.41304      0.394749  0.467588  0.540048 \n",
       " 0.426165  0.435777  0.314689  0.329906     0.291599  0.485594  0.567202 \n",
       " 0.379442  0.374979  0.353112  0.381944     0.360733  0.297878  0.573065 \n",
       " 0.324394  0.357536  0.411506  0.352719     0.561067  0.597624  0.415227 \n",
       " 0.392045  0.387966  0.28269   0.377169  …  0.354022  0.503564  0.365848 \n",
       " 0.385085  0.361621  0.343019  0.330815     0.28482   0.381689  0.25511  \n",
       " 0.361015  0.372574  0.310331  0.374705     0.289525  0.169959  0.16697  \n",
       " 0.373246  0.318909  0.360588  0.420681     0.324038  0.151843  0.0874964\n",
       " 0.355147  0.382493  0.301081  0.427255     0.340597  0.181298  0.26696  \n",
       " 0.357185  0.33284   0.316193  0.286588  …  0.141649  0.177689  0.159322 \n",
       " 0.335288  0.346671  0.352788  0.384075     0.101248  0.189698  0.16047  \n",
       " 0.354601  0.332644  0.258723  0.305229     0.288774  0.163406  0.144812 \n",
       " ⋮                                       ⋱  ⋮                            \n",
       " 0.348963  0.376995  0.321868  0.369058     0.560576  0.63977   0.724762 \n",
       " 0.392605  0.46093   0.325514  0.415748     0.678479  0.595353  0.707989 \n",
       " 0.366495  0.394611  0.335777  0.346071     0.743508  0.760678  0.702256 \n",
       " 0.423843  0.433254  0.393593  0.383522     0.699603  0.655217  0.55668  \n",
       " 0.335699  0.352568  0.463578  0.350728  …  0.659011  0.684568  0.602578 \n",
       " 0.37938   0.321294  0.400054  0.367017     0.463265  0.526541  0.571596 \n",
       " 0.38532   0.416089  0.404387  0.391201     0.464796  0.524163  0.642772 \n",
       " 0.407809  0.353797  0.352219  0.388549     0.46617   0.60571   0.606939 \n",
       " 0.305613  0.36485   0.394557  0.337633     0.494321  0.508329  0.644774 \n",
       " 0.443315  0.386978  0.403713  0.398922  …  0.5831    0.498877  0.634769 \n",
       " 0.328793  0.353269  0.349689  0.343062     0.570823  0.628629  0.64582  \n",
       " 0.443315  0.386978  0.403713  0.398922     0.5831    0.498877  0.634769 \n",
       "\n",
       "[:, :, 6, 6] =\n",
       " 0.350857  0.302386  0.334106  0.383443  …  0.394749  0.467588  0.540048 \n",
       " 0.322122  0.420562  0.41304   0.387839     0.291599  0.485594  0.567202 \n",
       " 0.435777  0.314689  0.329906  0.379633     0.360733  0.297878  0.573065 \n",
       " 0.374979  0.353112  0.381944  0.36836      0.561067  0.597624  0.415227 \n",
       " 0.357536  0.411506  0.352719  0.337017     0.354022  0.503564  0.365848 \n",
       " 0.387966  0.28269   0.377169  0.386111  …  0.28482   0.381689  0.25511  \n",
       " 0.361621  0.343019  0.330815  0.305207     0.289525  0.169959  0.16697  \n",
       " 0.372574  0.310331  0.374705  0.38899      0.324038  0.151843  0.0874964\n",
       " 0.318909  0.360588  0.420681  0.289038     0.340597  0.181298  0.26696  \n",
       " 0.382493  0.301081  0.427255  0.391195     0.141649  0.177689  0.159322 \n",
       " 0.33284   0.316193  0.286588  0.386774  …  0.101248  0.189698  0.16047  \n",
       " 0.346671  0.352788  0.384075  0.353846     0.288774  0.163406  0.144812 \n",
       " 0.332644  0.258723  0.305229  0.397625     0.268564  0.274339  0.0775752\n",
       " ⋮                                       ⋱  ⋮                            \n",
       " 0.392605  0.46093   0.325514  0.415748     0.678479  0.595353  0.707989 \n",
       " 0.366495  0.394611  0.335777  0.346071     0.743508  0.760678  0.702256 \n",
       " 0.423843  0.433254  0.393593  0.383522     0.699603  0.655217  0.55668  \n",
       " 0.335699  0.352568  0.463578  0.350728     0.659011  0.684568  0.602578 \n",
       " 0.37938   0.321294  0.400054  0.367017  …  0.463265  0.526541  0.571596 \n",
       " 0.38532   0.416089  0.404387  0.391201     0.464796  0.524163  0.642772 \n",
       " 0.407809  0.353797  0.352219  0.388549     0.46617   0.60571   0.606939 \n",
       " 0.305613  0.36485   0.394557  0.337633     0.494321  0.508329  0.644774 \n",
       " 0.443315  0.386978  0.403713  0.398922     0.5831    0.498877  0.634769 \n",
       " 0.328793  0.353269  0.349689  0.343062  …  0.570823  0.628629  0.64582  \n",
       " 0.443315  0.386978  0.403713  0.398922     0.5831    0.498877  0.634769 \n",
       " 0.305613  0.36485   0.394557  0.337633     0.494321  0.508329  0.644774 \n",
       "\n",
       "[:, :, 7, 6] =\n",
       " 0.302386  0.334106  0.383443  0.422243  …  0.291599  0.485594  0.567202 \n",
       " 0.420562  0.41304   0.387839  0.281307     0.360733  0.297878  0.573065 \n",
       " 0.314689  0.329906  0.379633  0.375243     0.561067  0.597624  0.415227 \n",
       " 0.353112  0.381944  0.36836   0.332942     0.354022  0.503564  0.365848 \n",
       " 0.411506  0.352719  0.337017  0.356497     0.28482   0.381689  0.25511  \n",
       " 0.28269   0.377169  0.386111  0.328073  …  0.289525  0.169959  0.16697  \n",
       " 0.343019  0.330815  0.305207  0.365998     0.324038  0.151843  0.0874964\n",
       " 0.310331  0.374705  0.38899   0.381824     0.340597  0.181298  0.26696  \n",
       " 0.360588  0.420681  0.289038  0.321463     0.141649  0.177689  0.159322 \n",
       " 0.301081  0.427255  0.391195  0.385622     0.101248  0.189698  0.16047  \n",
       " 0.316193  0.286588  0.386774  0.359338  …  0.288774  0.163406  0.144812 \n",
       " 0.352788  0.384075  0.353846  0.369752     0.268564  0.274339  0.0775752\n",
       " 0.258723  0.305229  0.397625  0.349684     0.210369  0.12026   0.0790067\n",
       " ⋮                                       ⋱  ⋮                            \n",
       " 0.366495  0.394611  0.335777  0.346071     0.743508  0.760678  0.702256 \n",
       " 0.423843  0.433254  0.393593  0.383522     0.699603  0.655217  0.55668  \n",
       " 0.335699  0.352568  0.463578  0.350728     0.659011  0.684568  0.602578 \n",
       " 0.37938   0.321294  0.400054  0.367017     0.463265  0.526541  0.571596 \n",
       " 0.38532   0.416089  0.404387  0.391201  …  0.464796  0.524163  0.642772 \n",
       " 0.407809  0.353797  0.352219  0.388549     0.46617   0.60571   0.606939 \n",
       " 0.305613  0.36485   0.394557  0.337633     0.494321  0.508329  0.644774 \n",
       " 0.443315  0.386978  0.403713  0.398922     0.5831    0.498877  0.634769 \n",
       " 0.328793  0.353269  0.349689  0.343062     0.570823  0.628629  0.64582  \n",
       " 0.443315  0.386978  0.403713  0.398922  …  0.5831    0.498877  0.634769 \n",
       " 0.305613  0.36485   0.394557  0.337633     0.494321  0.508329  0.644774 \n",
       " 0.407809  0.353797  0.352219  0.388549     0.46617   0.60571   0.606939 \n",
       "\n",
       "[:, :, 1, 7] =\n",
       " 0.420562  0.322122  0.298638  0.41914   …  0.485594  0.567202   0.496517\n",
       " 0.314689  0.435777  0.426165  0.384639     0.467588  0.540048   0.463395\n",
       " 0.353112  0.374979  0.379442  0.39974      0.321467  0.33134    0.433398\n",
       " 0.411506  0.357536  0.324394  0.411828     0.432999  0.441499   0.5479  \n",
       " 0.28269   0.387966  0.392045  0.384901     0.321467  0.33134    0.433398\n",
       " 0.343019  0.361621  0.385085  0.343604  …  0.467588  0.540048   0.463395\n",
       " 0.310331  0.372574  0.361015  0.324292     0.485594  0.567202   0.496517\n",
       " 0.360588  0.318909  0.373246  0.374508     0.297878  0.573065   0.462501\n",
       " 0.301081  0.382493  0.355147  0.406666     0.597624  0.415227   0.440267\n",
       " 0.316193  0.33284   0.357185  0.293279     0.503564  0.365848   0.35104 \n",
       " 0.352788  0.346671  0.335288  0.353417  …  0.381689  0.25511    0.319882\n",
       " 0.258723  0.332644  0.354601  0.432126     0.169959  0.16697    0.280956\n",
       " 0.38013   0.356753  0.306655  0.410987     0.151843  0.0874964  0.167252\n",
       " ⋮                                       ⋱  ⋮                            \n",
       " 0.311386  0.364438  0.41035   0.338129     0.638634  0.607852   0.558204\n",
       " 0.335594  0.356753  0.293823  0.385394     0.698036  0.520682   0.535028\n",
       " 0.399874  0.394656  0.39226   0.385214     0.577087  0.557041   0.607727\n",
       " 0.400662  0.383294  0.453116  0.357385     0.547219  0.738772   0.758275\n",
       " 0.376995  0.321868  0.369058  0.421469  …  0.63977   0.724762   0.658808\n",
       " 0.46093   0.325514  0.415748  0.380033     0.595353  0.707989   0.66532 \n",
       " 0.394611  0.335777  0.346071  0.342621     0.760678  0.702256   0.638867\n",
       " 0.433254  0.393593  0.383522  0.340241     0.655217  0.55668    0.598602\n",
       " 0.352568  0.463578  0.350728  0.371003     0.684568  0.602578   0.633151\n",
       " 0.321294  0.400054  0.367017  0.435339  …  0.526541  0.571596   0.750186\n",
       " 0.416089  0.404387  0.391201  0.376272     0.524163  0.642772   0.70249 \n",
       " 0.353797  0.352219  0.388549  0.359756     0.60571   0.606939   0.629921\n",
       "\n",
       "[:, :, 2, 7] =\n",
       " 0.322122  0.298638  0.41914   0.298638  …  0.467588  0.540048   0.463395\n",
       " 0.435777  0.426165  0.384639  0.426165     0.321467  0.33134    0.433398\n",
       " 0.374979  0.379442  0.39974   0.379442     0.432999  0.441499   0.5479  \n",
       " 0.357536  0.324394  0.411828  0.324394     0.321467  0.33134    0.433398\n",
       " 0.387966  0.392045  0.384901  0.392045     0.467588  0.540048   0.463395\n",
       " 0.361621  0.385085  0.343604  0.385085  …  0.485594  0.567202   0.496517\n",
       " 0.372574  0.361015  0.324292  0.361015     0.297878  0.573065   0.462501\n",
       " 0.318909  0.373246  0.374508  0.373246     0.597624  0.415227   0.440267\n",
       " 0.382493  0.355147  0.406666  0.355147     0.503564  0.365848   0.35104 \n",
       " 0.33284   0.357185  0.293279  0.357185     0.381689  0.25511    0.319882\n",
       " 0.346671  0.335288  0.353417  0.335288  …  0.169959  0.16697    0.280956\n",
       " 0.332644  0.354601  0.432126  0.354601     0.151843  0.0874964  0.167252\n",
       " 0.356753  0.306655  0.410987  0.306655     0.181298  0.26696    0.217961\n",
       " ⋮                                       ⋱  ⋮                            \n",
       " 0.335594  0.356753  0.293823  0.385394     0.698036  0.520682   0.535028\n",
       " 0.399874  0.394656  0.39226   0.385214     0.577087  0.557041   0.607727\n",
       " 0.400662  0.383294  0.453116  0.357385     0.547219  0.738772   0.758275\n",
       " 0.376995  0.321868  0.369058  0.421469     0.63977   0.724762   0.658808\n",
       " 0.46093   0.325514  0.415748  0.380033  …  0.595353  0.707989   0.66532 \n",
       " 0.394611  0.335777  0.346071  0.342621     0.760678  0.702256   0.638867\n",
       " 0.433254  0.393593  0.383522  0.340241     0.655217  0.55668    0.598602\n",
       " 0.352568  0.463578  0.350728  0.371003     0.684568  0.602578   0.633151\n",
       " 0.321294  0.400054  0.367017  0.435339     0.526541  0.571596   0.750186\n",
       " 0.416089  0.404387  0.391201  0.376272  …  0.524163  0.642772   0.70249 \n",
       " 0.353797  0.352219  0.388549  0.359756     0.60571   0.606939   0.629921\n",
       " 0.36485   0.394557  0.337633  0.326126     0.508329  0.644774   0.657626\n",
       "\n",
       "[:, :, 3, 7] =\n",
       " 0.298638  0.41914   0.298638  0.322122  …  0.321467  0.33134    0.433398\n",
       " 0.426165  0.384639  0.426165  0.435777     0.432999  0.441499   0.5479  \n",
       " 0.379442  0.39974   0.379442  0.374979     0.321467  0.33134    0.433398\n",
       " 0.324394  0.411828  0.324394  0.357536     0.467588  0.540048   0.463395\n",
       " 0.392045  0.384901  0.392045  0.387966     0.485594  0.567202   0.496517\n",
       " 0.385085  0.343604  0.385085  0.361621  …  0.297878  0.573065   0.462501\n",
       " 0.361015  0.324292  0.361015  0.372574     0.597624  0.415227   0.440267\n",
       " 0.373246  0.374508  0.373246  0.318909     0.503564  0.365848   0.35104 \n",
       " 0.355147  0.406666  0.355147  0.382493     0.381689  0.25511    0.319882\n",
       " 0.357185  0.293279  0.357185  0.33284      0.169959  0.16697    0.280956\n",
       " 0.335288  0.353417  0.335288  0.346671  …  0.151843  0.0874964  0.167252\n",
       " 0.354601  0.432126  0.354601  0.332644     0.181298  0.26696    0.217961\n",
       " 0.306655  0.410987  0.306655  0.356753     0.177689  0.159322   0.150437\n",
       " ⋮                                       ⋱  ⋮                            \n",
       " 0.399874  0.394656  0.39226   0.385214     0.577087  0.557041   0.607727\n",
       " 0.400662  0.383294  0.453116  0.357385     0.547219  0.738772   0.758275\n",
       " 0.376995  0.321868  0.369058  0.421469     0.63977   0.724762   0.658808\n",
       " 0.46093   0.325514  0.415748  0.380033     0.595353  0.707989   0.66532 \n",
       " 0.394611  0.335777  0.346071  0.342621  …  0.760678  0.702256   0.638867\n",
       " 0.433254  0.393593  0.383522  0.340241     0.655217  0.55668    0.598602\n",
       " 0.352568  0.463578  0.350728  0.371003     0.684568  0.602578   0.633151\n",
       " 0.321294  0.400054  0.367017  0.435339     0.526541  0.571596   0.750186\n",
       " 0.416089  0.404387  0.391201  0.376272     0.524163  0.642772   0.70249 \n",
       " 0.353797  0.352219  0.388549  0.359756  …  0.60571   0.606939   0.629921\n",
       " 0.36485   0.394557  0.337633  0.326126     0.508329  0.644774   0.657626\n",
       " 0.386978  0.403713  0.398922  0.359733     0.498877  0.634769   0.757109\n",
       "\n",
       "[:, :, 4, 7] =\n",
       " 0.41914   0.298638  0.322122  0.420562  …  0.432999  0.441499   0.5479  \n",
       " 0.384639  0.426165  0.435777  0.314689     0.321467  0.33134    0.433398\n",
       " 0.39974   0.379442  0.374979  0.353112     0.467588  0.540048   0.463395\n",
       " 0.411828  0.324394  0.357536  0.411506     0.485594  0.567202   0.496517\n",
       " 0.384901  0.392045  0.387966  0.28269      0.297878  0.573065   0.462501\n",
       " 0.343604  0.385085  0.361621  0.343019  …  0.597624  0.415227   0.440267\n",
       " 0.324292  0.361015  0.372574  0.310331     0.503564  0.365848   0.35104 \n",
       " 0.374508  0.373246  0.318909  0.360588     0.381689  0.25511    0.319882\n",
       " 0.406666  0.355147  0.382493  0.301081     0.169959  0.16697    0.280956\n",
       " 0.293279  0.357185  0.33284   0.316193     0.151843  0.0874964  0.167252\n",
       " 0.353417  0.335288  0.346671  0.352788  …  0.181298  0.26696    0.217961\n",
       " 0.432126  0.354601  0.332644  0.258723     0.177689  0.159322   0.150437\n",
       " 0.410987  0.306655  0.356753  0.38013      0.189698  0.16047    0.272203\n",
       " ⋮                                       ⋱  ⋮                            \n",
       " 0.400662  0.383294  0.453116  0.357385     0.547219  0.738772   0.758275\n",
       " 0.376995  0.321868  0.369058  0.421469     0.63977   0.724762   0.658808\n",
       " 0.46093   0.325514  0.415748  0.380033     0.595353  0.707989   0.66532 \n",
       " 0.394611  0.335777  0.346071  0.342621     0.760678  0.702256   0.638867\n",
       " 0.433254  0.393593  0.383522  0.340241  …  0.655217  0.55668    0.598602\n",
       " 0.352568  0.463578  0.350728  0.371003     0.684568  0.602578   0.633151\n",
       " 0.321294  0.400054  0.367017  0.435339     0.526541  0.571596   0.750186\n",
       " 0.416089  0.404387  0.391201  0.376272     0.524163  0.642772   0.70249 \n",
       " 0.353797  0.352219  0.388549  0.359756     0.60571   0.606939   0.629921\n",
       " 0.36485   0.394557  0.337633  0.326126  …  0.508329  0.644774   0.657626\n",
       " 0.386978  0.403713  0.398922  0.359733     0.498877  0.634769   0.757109\n",
       " 0.353269  0.349689  0.343062  0.318974     0.628629  0.64582    0.694726\n",
       "\n",
       "[:, :, 5, 7] =\n",
       " 0.298638  0.322122  0.420562  0.41304   …  0.321467  0.33134    0.433398\n",
       " 0.426165  0.435777  0.314689  0.329906     0.467588  0.540048   0.463395\n",
       " 0.379442  0.374979  0.353112  0.381944     0.485594  0.567202   0.496517\n",
       " 0.324394  0.357536  0.411506  0.352719     0.297878  0.573065   0.462501\n",
       " 0.392045  0.387966  0.28269   0.377169     0.597624  0.415227   0.440267\n",
       " 0.385085  0.361621  0.343019  0.330815  …  0.503564  0.365848   0.35104 \n",
       " 0.361015  0.372574  0.310331  0.374705     0.381689  0.25511    0.319882\n",
       " 0.373246  0.318909  0.360588  0.420681     0.169959  0.16697    0.280956\n",
       " 0.355147  0.382493  0.301081  0.427255     0.151843  0.0874964  0.167252\n",
       " 0.357185  0.33284   0.316193  0.286588     0.181298  0.26696    0.217961\n",
       " 0.335288  0.346671  0.352788  0.384075  …  0.177689  0.159322   0.150437\n",
       " 0.354601  0.332644  0.258723  0.305229     0.189698  0.16047    0.272203\n",
       " 0.306655  0.356753  0.38013   0.386695     0.163406  0.144812   0.163869\n",
       " ⋮                                       ⋱  ⋮                            \n",
       " 0.376995  0.321868  0.369058  0.421469     0.63977   0.724762   0.658808\n",
       " 0.46093   0.325514  0.415748  0.380033     0.595353  0.707989   0.66532 \n",
       " 0.394611  0.335777  0.346071  0.342621     0.760678  0.702256   0.638867\n",
       " 0.433254  0.393593  0.383522  0.340241     0.655217  0.55668    0.598602\n",
       " 0.352568  0.463578  0.350728  0.371003  …  0.684568  0.602578   0.633151\n",
       " 0.321294  0.400054  0.367017  0.435339     0.526541  0.571596   0.750186\n",
       " 0.416089  0.404387  0.391201  0.376272     0.524163  0.642772   0.70249 \n",
       " 0.353797  0.352219  0.388549  0.359756     0.60571   0.606939   0.629921\n",
       " 0.36485   0.394557  0.337633  0.326126     0.508329  0.644774   0.657626\n",
       " 0.386978  0.403713  0.398922  0.359733  …  0.498877  0.634769   0.757109\n",
       " 0.353269  0.349689  0.343062  0.318974     0.628629  0.64582    0.694726\n",
       " 0.386978  0.403713  0.398922  0.359733     0.498877  0.634769   0.757109\n",
       "\n",
       "[:, :, 6, 7] =\n",
       " 0.322122  0.420562  0.41304   0.387839  …  0.467588  0.540048   0.463395\n",
       " 0.435777  0.314689  0.329906  0.379633     0.485594  0.567202   0.496517\n",
       " 0.374979  0.353112  0.381944  0.36836      0.297878  0.573065   0.462501\n",
       " 0.357536  0.411506  0.352719  0.337017     0.597624  0.415227   0.440267\n",
       " 0.387966  0.28269   0.377169  0.386111     0.503564  0.365848   0.35104 \n",
       " 0.361621  0.343019  0.330815  0.305207  …  0.381689  0.25511    0.319882\n",
       " 0.372574  0.310331  0.374705  0.38899      0.169959  0.16697    0.280956\n",
       " 0.318909  0.360588  0.420681  0.289038     0.151843  0.0874964  0.167252\n",
       " 0.382493  0.301081  0.427255  0.391195     0.181298  0.26696    0.217961\n",
       " 0.33284   0.316193  0.286588  0.386774     0.177689  0.159322   0.150437\n",
       " 0.346671  0.352788  0.384075  0.353846  …  0.189698  0.16047    0.272203\n",
       " 0.332644  0.258723  0.305229  0.397625     0.163406  0.144812   0.163869\n",
       " 0.356753  0.38013   0.386695  0.312281     0.274339  0.0775752  0.147525\n",
       " ⋮                                       ⋱  ⋮                            \n",
       " 0.46093   0.325514  0.415748  0.380033     0.595353  0.707989   0.66532 \n",
       " 0.394611  0.335777  0.346071  0.342621     0.760678  0.702256   0.638867\n",
       " 0.433254  0.393593  0.383522  0.340241     0.655217  0.55668    0.598602\n",
       " 0.352568  0.463578  0.350728  0.371003     0.684568  0.602578   0.633151\n",
       " 0.321294  0.400054  0.367017  0.435339  …  0.526541  0.571596   0.750186\n",
       " 0.416089  0.404387  0.391201  0.376272     0.524163  0.642772   0.70249 \n",
       " 0.353797  0.352219  0.388549  0.359756     0.60571   0.606939   0.629921\n",
       " 0.36485   0.394557  0.337633  0.326126     0.508329  0.644774   0.657626\n",
       " 0.386978  0.403713  0.398922  0.359733     0.498877  0.634769   0.757109\n",
       " 0.353269  0.349689  0.343062  0.318974  …  0.628629  0.64582    0.694726\n",
       " 0.386978  0.403713  0.398922  0.359733     0.498877  0.634769   0.757109\n",
       " 0.36485   0.394557  0.337633  0.326126     0.508329  0.644774   0.657626\n",
       "\n",
       "[:, :, 7, 7] =\n",
       " 0.420562  0.41304   0.387839  0.281307  …  0.485594  0.567202   0.496517\n",
       " 0.314689  0.329906  0.379633  0.375243     0.297878  0.573065   0.462501\n",
       " 0.353112  0.381944  0.36836   0.332942     0.597624  0.415227   0.440267\n",
       " 0.411506  0.352719  0.337017  0.356497     0.503564  0.365848   0.35104 \n",
       " 0.28269   0.377169  0.386111  0.328073     0.381689  0.25511    0.319882\n",
       " 0.343019  0.330815  0.305207  0.365998  …  0.169959  0.16697    0.280956\n",
       " 0.310331  0.374705  0.38899   0.381824     0.151843  0.0874964  0.167252\n",
       " 0.360588  0.420681  0.289038  0.321463     0.181298  0.26696    0.217961\n",
       " 0.301081  0.427255  0.391195  0.385622     0.177689  0.159322   0.150437\n",
       " 0.316193  0.286588  0.386774  0.359338     0.189698  0.16047    0.272203\n",
       " 0.352788  0.384075  0.353846  0.369752  …  0.163406  0.144812   0.163869\n",
       " 0.258723  0.305229  0.397625  0.349684     0.274339  0.0775752  0.147525\n",
       " 0.38013   0.386695  0.312281  0.454245     0.12026   0.0790067  0.076828\n",
       " ⋮                                       ⋱  ⋮                            \n",
       " 0.394611  0.335777  0.346071  0.342621     0.760678  0.702256   0.638867\n",
       " 0.433254  0.393593  0.383522  0.340241     0.655217  0.55668    0.598602\n",
       " 0.352568  0.463578  0.350728  0.371003     0.684568  0.602578   0.633151\n",
       " 0.321294  0.400054  0.367017  0.435339     0.526541  0.571596   0.750186\n",
       " 0.416089  0.404387  0.391201  0.376272  …  0.524163  0.642772   0.70249 \n",
       " 0.353797  0.352219  0.388549  0.359756     0.60571   0.606939   0.629921\n",
       " 0.36485   0.394557  0.337633  0.326126     0.508329  0.644774   0.657626\n",
       " 0.386978  0.403713  0.398922  0.359733     0.498877  0.634769   0.757109\n",
       " 0.353269  0.349689  0.343062  0.318974     0.628629  0.64582    0.694726\n",
       " 0.386978  0.403713  0.398922  0.359733  …  0.498877  0.634769   0.757109\n",
       " 0.36485   0.394557  0.337633  0.326126     0.508329  0.644774   0.657626\n",
       " 0.353797  0.352219  0.388549  0.359756     0.60571   0.606939   0.629921"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "P = patch(f)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Display some example of patches."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "PyPlot.Figure(PyObject <matplotlib.figure.Figure object at 0x000000001EB1DEB8>)"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "figure(figsize = (5,5))\n",
    "for i in 1:16\n",
    "    x = rand(1 : n)\n",
    "    y = rand(1 : n)\n",
    "    imageplot(P[x, y, :, :], \"\", [4, 4, i])\n",
    "end"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Dimensionality Reduction with PCA\n",
    "---------------------------------\n",
    "Since NL-means type algorithms require the computation of many distances\n",
    "between patches, it is advantagous to reduce the dimensionality of the\n",
    "patch while keeping as much as possible of information.\n",
    "\n",
    "\n",
    "Target dimensionality $d$."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "25"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "d = 25"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A linear dimensionality reduction is obtained by Principal Component\n",
    "Analysis (PCA) that projects the data on a small number of leading\n",
    "direction of the covariance matrix of the patches.\n",
    "\n",
    "\n",
    "Turn the patch matrix into an $(w_1*w_1,n*n)$ array, so that each $P(:,i)$\n",
    "is a $w_1*w_1$ vector representing a patch."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(::#3) (generic function with 1 method)"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "resh = P -> transpose((reshape(P, (n*n,w1*w1))))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Operator to remove the mean of the patches to each patch."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(::#5) (generic function with 1 method)"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "remove_mean = Q -> Q - repeat(mean(Q, 1), inner = (w1*w1, 1))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Compute the mean and the covariance of the points cloud representing the\n",
    "patches."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "49×49 Array{Float64,2}:\n",
       " 131.936     77.2095     46.5651   …  -25.4735   -23.785    -17.5533  \n",
       "  77.2095   121.381      67.3999      -25.9776   -26.7635   -24.1005  \n",
       "  46.5651    67.3999    112.653       -25.7255   -26.301    -26.0259  \n",
       "  27.8702    37.7467     59.4592      -28.8331   -25.0024   -24.7251  \n",
       "  18.1669    20.245      31.3374      -29.9974   -26.6962   -21.9355  \n",
       "  10.1857    12.4336     14.8341   …  -30.2334   -26.4507   -22.151   \n",
       "   2.25873    6.15092     8.88626     -25.7245   -24.7144   -19.9362  \n",
       "  67.2198    68.495      54.8283      -21.4129   -22.696    -22.7255  \n",
       "  44.0538    55.9557     57.9443      -25.0261   -23.1991   -23.5449  \n",
       "  25.3588    34.2606     46.9429      -29.7959   -25.3353   -22.4983  \n",
       "  13.7735    16.8541     26.6187   …  -32.7065   -28.6216   -23.1829  \n",
       "   4.45361    7.02014    10.8829      -30.2134   -29.5932   -24.6672  \n",
       "  -3.04818   -0.227483    2.91176     -25.7234   -25.0746   -23.5566  \n",
       "   ⋮                               ⋱                                  \n",
       " -23.8409   -29.1243    -30.1578       11.0099     6.84582    4.26891 \n",
       " -22.5349   -28.259     -32.66         27.1568    17.4338    14.2956  \n",
       " -22.1599   -25.2158    -30.0699       48.5088    35.6748    26.8608  \n",
       " -23.2678   -23.3726    -25.5313   …   59.2582    58.6027    46.8494  \n",
       " -22.7891   -22.9862    -22.1232       56.2704    71.0329    71.3181  \n",
       " -17.5671   -22.3694    -23.8187        6.58808    4.35664    0.887412\n",
       " -20.2215   -25.1622    -28.9253       13.7246    11.0812     9.33479 \n",
       " -20.6776   -25.6028    -29.4684       30.9147    19.5402    17.544   \n",
       " -23.7535   -24.355     -28.3453   …   60.3797    38.0947    27.9536  \n",
       " -25.4735   -25.9776    -25.7255      115.687     69.4826    48.1777  \n",
       " -23.785    -26.7635    -26.301        69.4826   125.946     80.862   \n",
       " -17.5533   -24.1005    -26.0259       48.1777    80.862    138.289   "
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "P1 = remove_mean(resh(P))\n",
    "C = P1*transpose(P1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Extract the eigenvectors, sorted by decreasing amplitude."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "49×49 Array{Float64,2}:\n",
       " -0.142857   0.00288788   0.022342     0.115746    …  -0.201372     0.166028 \n",
       " -0.142857  -0.0775719   -0.039154    -0.237704       -0.215737     0.188468 \n",
       " -0.142857  -0.0466216    0.0304623    0.259052       -0.193864     0.201698 \n",
       " -0.142857  -0.00291281   0.0788866   -0.167343       -0.129294     0.203078 \n",
       " -0.142857   0.0375051    0.0861557    0.0362725      -0.0335826    0.188389 \n",
       " -0.142857   0.0581363    0.0609241   -0.00981668  …   0.0672203    0.156301 \n",
       " -0.142857   0.112538    -0.0793377   -0.0184382       0.133927     0.108932 \n",
       " -0.142857   0.0871356   -0.0240147   -0.0699827      -0.214902     0.171969 \n",
       " -0.142857   0.14265      0.0539251    0.207983       -0.19236      0.18962  \n",
       " -0.142857   0.1124      -0.151321    -0.26095        -0.118351     0.191938 \n",
       " -0.142857  -0.0294305   -0.205379     0.190192    …  -0.00851431   0.17653  \n",
       " -0.142857  -0.113902    -0.175673    -0.0191166       0.101116     0.141956 \n",
       " -0.142857  -0.159439    -0.0894712    0.047253        0.169903     0.0913398\n",
       "  ⋮                                                ⋱                         \n",
       " -0.142857   0.0725956   -0.119268     0.178503        0.109989    -0.13773  \n",
       " -0.142857   0.00742462  -0.255754    -0.134852        0.00273088  -0.176623 \n",
       " -0.142857  -0.117737    -0.052603     0.268887       -0.110743    -0.197004 \n",
       " -0.142857  -0.146842    -0.00787113  -0.17119     …  -0.19071     -0.198347 \n",
       " -0.142857  -0.0755484   -0.0136396    0.0525918      -0.217508    -0.182695 \n",
       " -0.142857  -0.0763572   -0.0337713    0.00799349      0.136338    -0.0957285\n",
       " -0.142857  -0.0269051    0.0526934    0.0340267       0.0764488   -0.148111 \n",
       " -0.142857  -0.0278964    0.0566297   -0.10423        -0.022017    -0.18472  \n",
       " -0.142857   0.00434924   0.106746     0.131244    …  -0.120879    -0.203753 \n",
       " -0.142857   0.0440216   -0.0214087   -0.221072       -0.191356    -0.206737 \n",
       " -0.142857   0.0768078    0.0183171    0.190107       -0.218507    -0.196315 \n",
       " -0.142857  -0.00464626  -0.00183453  -0.0903946      -0.207341    -0.175641 "
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(D, V) = eig(C)\n",
    "D = sort(D, rev = true)\n",
    "I = sortperm(D)[end : -1 : 1]\n",
    "V = V[I, :]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Display the decaying amplitude of the eigenvalues."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "PyPlot.Figure(PyObject <matplotlib.figure.Figure object at 0x00000000235C7F98>)"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot(D, linewidth = 2)\n",
    "ylim(0, maximum(D))\n",
    "show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Display the leading eigenvectors - they look like Fourier modes."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "PyPlot.Figure(PyObject <matplotlib.figure.Figure object at 0x0000000022F42B00>)"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "figure(figsize = (5,5))\n",
    "for i in 1:16\n",
    "    imageplot(abs(reshape(V[:,i], (w1,w1))), \"\", [4, 4, i])\n",
    "end"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Patch dimensionality reduction operator."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "16384-element Array{Float64,1}:\n",
       "  0.0628797 \n",
       " -0.0601156 \n",
       "  0.0433663 \n",
       " -0.0076114 \n",
       "  0.0415792 \n",
       " -0.063938  \n",
       "  0.0552831 \n",
       " -0.0501452 \n",
       " -0.00654325\n",
       "  0.0581593 \n",
       " -0.0670696 \n",
       " -0.00293932\n",
       " -0.0351177 \n",
       "  ⋮         \n",
       " -0.0810145 \n",
       " -0.10699   \n",
       " -0.040066  \n",
       "  0.108777  \n",
       "  0.0189314 \n",
       "  0.0347488 \n",
       "  0.0185005 \n",
       " -0.00987702\n",
       "  0.020824  \n",
       "  0.140613  \n",
       "  0.0938184 \n",
       "  0.0223174 "
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "iresh = Q -> reshape(Q', (n, n, d)) # order = \"F\"\n",
    "descriptor = f -> iresh(V[: , 1 : d]'*remove_mean(resh(P)))\n",
    "remove_mean(resh(P))[1, :]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Each $H(i,j,:)$ is a $d$-dimensional descriptor\n",
    "of a patch."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "25-element Array{Float64,1}:\n",
       "  1.12757e-16\n",
       "  0.00303514 \n",
       " -0.057321   \n",
       "  0.00279343 \n",
       "  0.0252618  \n",
       " -0.00510602 \n",
       " -0.0100753  \n",
       " -0.00124721 \n",
       "  0.0147514  \n",
       "  0.0228865  \n",
       "  0.00793989 \n",
       "  0.0926927  \n",
       "  0.0949608  \n",
       "  0.0191861  \n",
       " -0.0162093  \n",
       "  0.0132067  \n",
       " -0.00514667 \n",
       "  0.0530679  \n",
       " -0.0316565  \n",
       " -0.00451908 \n",
       " -0.0086364  \n",
       " -0.00777439 \n",
       " -0.146048   \n",
       " -0.00399402 \n",
       " -0.00891363 "
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "H = descriptor(f)\n",
    "H[1, 1, :]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Non-local Filter\n",
    "----------------\n",
    "NL-means applies, an adaptive averaging kernel\n",
    "is computed from patch distances to each pixel location.\n",
    "\n",
    "\n",
    "We denote $H_{i} \\in \\RR^d$ the descriptor at pixel $i$.\n",
    "We define the distance matrix\n",
    "$$ D_{i,j} = \\frac{1}{w_1^2}\\norm{H_i-H_j}^2. $$\n",
    "\n",
    "\n",
    "\n",
    "Operator to compute the distances $(D_{i,j})_j$ between the patch around $i=(i_1,i_2)$\n",
    "and all the other ones."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(::#11) (generic function with 1 method)"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "distance = i -> sum((H - repeat(reshape(H[i[1], i[2], :], (1, 1, length(H[i[1], i[2], :]))), inner = [n, n, 1])).^2, 3)./(w1*w1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The non-local mean filter computes a denoised image $\\tilde f$ as :\n",
    "\n",
    "$$ \\tilde f_i = \\sum_j K_{i,j} f_j $$\n",
    "where the weights $K$ are computed as :\n",
    "$$ K_{i,j} = \\frac{ \\tilde K_{i,j} }{ \\sum_{j'} \\tilde K_{i,j'} }\n",
    "      \\qandq\n",
    "   \\tilde K_{i,j} = e^{-\\frac{D_{i,j}}{2\\tau^2}} . $$\n",
    "\n",
    "\n",
    "\n",
    "The width $\\tau$ of the Gaussian is very important and should be adapted to match\n",
    "the noise level.\n",
    "\n",
    "\n",
    "\n",
    "Compute and normalize the weight."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(::#15) (generic function with 1 method)"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "normalize = K -> K./sum(K)\n",
    "kernel = (i, tau) -> normalize(exp(-distance(i)./(2*tau^2)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Compute a typical example of kernel for some pixel position $(x,y)$."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "128×128×1 Array{Float64,3}:\n",
       "[:, :, 1] =\n",
       " 0.00295118  0.00227457  0.00312287  …  0.00485491  0.00488994  0.00503266\n",
       " 0.00259518  0.00283335  0.00222588     0.00378126  0.00465444  0.00269188\n",
       " 0.00309054  0.00297622  0.00237054     0.00343972  0.00366405  0.00546961\n",
       " 0.00230125  0.00349517  0.00192803     0.00550008  0.00633937  0.00564258\n",
       " 0.00321561  0.00167929  0.00294033     0.00402107  0.00260499  0.00340563\n",
       " 0.00397462  0.00264592  0.00322254  …  0.00382952  0.00600861  0.00397644\n",
       " 0.00245938  0.0020279   0.00312583     0.00476732  0.00374555  0.00489222\n",
       " 0.00274398  0.00164827  0.00266606     0.00464598  0.00539079  0.00396139\n",
       " 0.002302    0.00240738  0.0030592      0.00523817  0.00437711  0.00538495\n",
       " 0.00295479  0.00121129  0.00368635     0.00277393  0.00259435  0.00276076\n",
       " 0.00185107  0.002613    0.00297119  …  0.00439295  0.00463488  0.0042115 \n",
       " 0.00240684  0.00180074  0.00291328     0.00338667  0.00283772  0.00312385\n",
       " 0.00251598  0.00311237  0.002596       0.00341123  0.00399084  0.00378708\n",
       " ⋮                                   ⋱  ⋮                                 \n",
       " 0.00235609  0.00259432  0.00318434     0.00279742  0.00293299  0.00300645\n",
       " 0.00247655  0.00230713  0.00195436     0.00427163  0.00408478  0.00350959\n",
       " 0.0024835   0.00290693  0.00297222     0.00214986  0.00222044  0.00256405\n",
       " 0.00237134  0.00260013  0.00280712     0.00384583  0.00397182  0.00404456\n",
       " 0.00246556  0.00277591  0.00223755  …  0.00250236  0.00198627  0.00181474\n",
       " 0.00227777  0.00247946  0.00304085     0.00340078  0.00327795  0.00414402\n",
       " 0.00309858  0.0028201   0.00185638     0.00252003  0.00266765  0.00246318\n",
       " 0.00150065  0.00261661  0.00231011     0.00248716  0.00203596  0.0029757 \n",
       " 0.00338383  0.00242381  0.00292186     0.00297141  0.00291344  0.00332062\n",
       " 0.0020981   0.00319493  0.00149648  …  0.00308626  0.00277732  0.0030444 \n",
       " 0.00385313  0.00293589  0.00348093     0.00218025  0.00279784  0.00236452\n",
       " 0.00173234  0.00263913  0.00185182     0.00290722  0.00248422  0.00288259"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tau = .05\n",
    "i = [83, 72]\n",
    "D = distance(i)\n",
    "H[i[1], i[2], :]\n",
    "K = kernel(i, tau)\n",
    "D"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Display the squared distance and the kernel."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "PyPlot.Figure(PyObject <matplotlib.figure.Figure object at 0x0000000022F9D908>)"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "PyObject <matplotlib.text.Text object at 0x000000001EE5BF60>"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "figure(figsize = (10,10))\n",
    "imageplot(D[:, :], \"D\", [1, 2, 1])\n",
    "imageplot(K[:, :], \"K\", [1, 2, 2])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Localizing the Non-local Means\n",
    "------------------------------\n",
    "We set a \"locality constant\" $q$ that set the maximum distance between\n",
    "patches to compare. This allows to speed up computation, and makes\n",
    "NL-means type methods semi-global (to avoid searching in all the image)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "14"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "q = 14"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Using this locality constant, we compute the distance between patches\n",
    "only within a window.\n",
    "Once again, one should be careful about boundary conditions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(::#19) (generic function with 1 method)"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#selection = i -> np.array((clamP(np.arange(i[1]-q,i[1] + q + 1), 0, n-1), clamP(np.arange(i[1]-q,i[1] + q + 1), 0, n-1)))\n",
    "selection = i -> [clamP(collect(i[1] - q + 1 : i[1] + q + 1), 1, n)'; clamP(collect(i[1] - q + 1 : i[1] + q + 1), 1, n)']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Compute distance and kernel only within the window."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(::#23) (generic function with 1 method)"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "function distance_0(i, sel)\n",
    "    H1 = (H[sel[1, :],:,:])\n",
    "    H2 = (H1[:,sel[2, :],:])\n",
    "    return sum((H2 - repeat(reshape(H[i[1], i[2], :], (1, 1, length(H[i[1], i[2], :]))), inner = [length(sel[1, :]), length(sel[2, :]), 1])), \n",
    "    3)/(w1*w1)\n",
    "end\n",
    "\n",
    "distance = i -> distance_0(i, selection(i))\n",
    "kernel = (i, tau) -> normalize(exp(-distance(i)./(2*tau^2)))\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Compute a typical example of kernel for some pixel position $(x,y)$."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Display the squared distance and the kernel."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "29×29×1 Array{Float64,3}:\n",
       "[:, :, 1] =\n",
       " 0.00150359   0.00331627   6.83253e-5   …  0.00239348   0.00041119 \n",
       " 0.000467514  0.000236781  0.00251782      0.000850111  0.00193867 \n",
       " 0.00075818   0.00360707   0.000962478     0.000245312  0.000591722\n",
       " 0.000267653  0.000326701  0.000258885     0.000398922  0.00233789 \n",
       " 0.00084619   0.000563018  0.00307985      0.00269301   0.000650247\n",
       " 0.000846278  0.00179653   0.000351704  …  0.000408738  0.000373752\n",
       " 0.00141865   0.000229179  0.000374389     0.00198729   0.00115064 \n",
       " 0.000221269  0.00258376   0.000605844     0.00121756   0.000417504\n",
       " 0.000918029  0.000436057  0.00128788      0.00127099   0.000482956\n",
       " 0.000516951  0.000268027  0.00166792      0.000316963  0.00505622 \n",
       " 0.00169658   0.000419005  0.00126345   …  0.000606645  0.000626355\n",
       " 0.00180485   0.000572404  0.000599983     0.00140198   0.00151399 \n",
       " 0.000583443  0.00231071   0.000851272     0.000184051  0.000926002\n",
       " ⋮                                      ⋱                          \n",
       " 8.48065e-5   0.00193428   0.000548713     0.000547021  0.00155817 \n",
       " 0.000667296  0.000865719  0.000783577     0.000304843  0.00126702 \n",
       " 0.000615073  0.00132741   0.000298352     0.00206906   0.000285545\n",
       " 0.000262269  0.0018164    0.0030514    …  0.000155753  0.00280487 \n",
       " 0.000379655  0.000154351  0.00500062      0.0013116    0.000273136\n",
       " 0.00285961   0.000776159  0.000525234     0.000862562  0.000633648\n",
       " 0.00242004   0.000150042  0.00169735      0.000391434  0.00240604 \n",
       " 0.00107573   0.000451066  0.000258555     0.000806164  0.000150238\n",
       " 0.000748652  0.00573865   0.000229333  …  0.00143951   0.00234482 \n",
       " 0.00066037   0.000361907  0.000626649     0.000703353  0.000698092\n",
       " 0.000435973  0.000642139  0.00095164      0.000179389  0.00101129 \n",
       " 0.000239842  0.00197698   0.000370909     0.000643037  0.00420599 "
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sel = selection(i)\n",
    "D = distance(i)\n",
    "K = kernel(i, tau)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "PyPlot.Figure(PyObject <matplotlib.figure.Figure object at 0x0000000022CFE7B8>)"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "PyObject <matplotlib.text.Text object at 0x0000000022CA5E10>"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "figure(figsize = (10,10))\n",
    "\n",
    "imageplot(D[:, :], \"D\", [1, 2, 1])\n",
    "imageplot(K[:, :], \"K\", [1, 2, 2])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The NL-filtered value at pixel $(x,y)$ is obtained by averaging the values\n",
    "of $f$ with the weight $K$."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(::#25) (generic function with 1 method)"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "function NLval_0(K,sel)\n",
    "    f_temp = f[sel[1, :], :]\n",
    "    return sum(K.*f_temp[:, sel[1, :]])\n",
    "end\n",
    "\n",
    "NLval = (i, ta) -> NLval_0(kernel(i, tau), selection(i))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We apply the filter to each pixel location\n",
    "to perform the NL-means algorithm."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(::#27) (generic function with 1 method)"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(Y, X) = meshgrid(0 : n - 1, 0 : n - 1)\n",
    "\n",
    "function arrayfun(f, X, Y)\n",
    "    n = size(X)[1]\n",
    "    p = size(Y)[1]\n",
    "    R = zeros(n, p)\n",
    "    for k in 1:n\n",
    "        for l in 1:p\n",
    "            R[k,l] = f(k,l)\n",
    "        end\n",
    "    end\n",
    "    return R\n",
    "end\n",
    "\n",
    "NLmeans = tau -> arrayfun((i1, i2) -> NLval([i1,i2], tau), X, Y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Display the result for some value of $\\tau$."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "PyPlot.Figure(PyObject <matplotlib.figure.Figure object at 0x000000001EF332B0>)"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "tau = .03\n",
    "\n",
    "figure(figsize = (5,5))\n",
    "imageplot(NLmeans(tau))\n",
    "#NLmeans(tau)\n",
    "#n = size(X)[1]\n",
    "#p = size(Y)[1]\n",
    "#R = zeros(n, p)\n",
    "#for k in 1:n\n",
    "#    for l in 1:p\n",
    "#        R[k,l] = f(k,l)\n",
    "#    end\n",
    "#end\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "tau = .03\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "__Exercise 1__\n",
    "\n",
    "Compute the denoising result for several values of $\\tau$ in order to\n",
    "determine the optimal denoising that minimizes $\\norm{\\tilde f - f_0}$."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "ename": "LoadError",
     "evalue": "LoadError: BoundsError: attempt to access 128×128×25 Array{Float64,3} at index [[0,0,0,0,0,0,0,0,0,0,0,0,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16],Colon(),Colon()]\nwhile loading C:\\Users\\Ayman\\.julia\\v0.5\\Exos\\denoisingadv_6_nl_means\\exo1.jl, in expression starting on line 6",
     "output_type": "error",
     "traceback": [
      "LoadError: BoundsError: attempt to access 128×128×25 Array{Float64,3} at index [[0,0,0,0,0,0,0,0,0,0,0,0,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16],Colon(),Colon()]\nwhile loading C:\\Users\\Ayman\\.julia\\v0.5\\Exos\\denoisingadv_6_nl_means\\exo1.jl, in expression starting on line 6",
      "",
      " in throw_boundserror(::Array{Float64,3}, ::Tuple{Array{Int64,1},Colon,Colon}) at .\\abstractarray.jl:355",
      " in checkbounds at .\\abstractarray.jl:284 [inlined]",
      " in _getindex at .\\multidimensional.jl:270 [inlined]",
      " in getindex(::Array{Float64,3}, ::Array{Int64,1}, ::Colon, ::Colon) at .\\abstractarray.jl:752",
      " in distance_0 at .\\In[30]:2 [inlined]",
      " in (::##19#20)(::Array{Int64,1}) at .\\In[30]:8",
      " in (::##21#22)(::Array{Int64,1}, ::Float64) at .\\In[30]:9",
      " in (::##23#24)(::Array{Int64,1}, ::Float64) at .\\In[33]:6",
      " in arrayfun(::##26#28{Float64}, ::Array{Int64,2}, ::Array{Int64,2}) at .\\In[34]:9",
      " in (::##25#27)(::Float64) at .\\In[34]:15",
      " in macro expansion; at C:\\Users\\Ayman\\.julia\\v0.5\\Exos\\denoisingadv_6_nl_means\\exo1.jl:8 [inlined]",
      " in anonymous at .\\<missing>:?",
      " in include_from_node1(::String) at .\\loading.jl:488"
     ]
    }
   ],
   "source": [
    "#run -i nt_solutions/denoisingadv_6_nl_means/exo1\n",
    "include(\"Exos\\\\denoisingadv_6_nl_means\\\\exo1.jl\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "## Insert your code here."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Display the best result."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAATwAAAE3CAYAAAA3wv3FAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nGdo/eJs2bboZ+v8Aw3k+ZQohUZAAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2633ea18748>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize = (5,5))\n",
    "imageplot(clamp(fNL))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "__Exercise 2__\n",
    "\n",
    "Explore the influence of the $q$ and $w$ parameters."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlMAAAJZCAYAAACN2rCOAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n6QAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2633c7fecc0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "run -i nt_solutions/denoisingadv_6_nl_means/exo2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "## Insert your code here."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Bibliography\n",
    "------------\n",
    "<html><a name=\"biblio\"></a></html>\n",
    "\n",
    "\n",
    "* [BuaCoMoA05] Buades, B. Coll, J.f Morel, [A review of image denoising algorithms, with a new one][1], SIAM Multiscale Modeling and Simulation, Vol 4 (2), pp: 490-530, 2005.\n",
    "\n",
    "[1]:http://dx.doi.org/10.1137/040616024"
   ]
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Julia 0.5.0",
   "language": "julia",
   "name": "julia-0.5"
  },
  "language_info": {
   "file_extension": ".jl",
   "mimetype": "application/julia",
   "name": "julia",
   "version": "0.5.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}