code/06-graphblas/NB06-graphblas.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "43210813-e30f-400b-8613-883d95895656",
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    },
    "tags": []
   },
   "source": [
    "# Notebook #6 - GraphBLAS overview\n",
    "\n",
    "Outline:\n",
    "1. [Graphs and Sparse Matrices](#Graphs-and-Sparse-Matrices)\n",
    "2. [Whirlwind Tour of the GraphBLAS](#Whirlwind-Tour-of-the-GraphBLAS)\n",
    "3. [Looking at Code](#Looking-at-Code)\n",
    "4. [More Resources](#More-Resources)\n",
    "\n",
    "With observations on pitfalls I've seen for new users."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "99a01183-6f1f-4b51-956f-236c1acb26ff",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "Note: This **cannot** be even a serious outline of even the high-level pieces of the GraphBLAS.\n",
    "\n",
    "My intention is a quick, ***opinionated*** tour of the GraphBLAS and its C binding.  Questions, comments, and corrections welcome during the crazy tour.\n",
    "\n",
    "Other much, much longer tutorials cover a full introduction:\n",
    "- https://graphblas.org/GraphBLAS-Pointers/#tutorials\n",
    "- https://github.com/GraphBLAS-Tutorials/SIAM-Tutorial\n",
    "\n",
    "Also: The GraphBLAS is the product of many other peoples' work over a long, long time.  I'm just a sideline commenter.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "02c08234-ee11-4932-b4fc-7d87e78f12b7",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Graphs and Sparse Matrices\n",
    "\n",
    "These have been buddies for a long, long time.\n",
    "- Solving sparse systems, computing their spectrum, etc. often rely on graph analysis primitives.\n",
    "- Spectral graph theories rely on sparse matrix algebra.\n",
    "\n",
    "First: A quick tour of a few methods for mapping graphs onto sparse matrices.\n",
    "- Adjacency and variations\n",
    "- Incidence"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fcc614a5-0aee-43b4-a3a2-8a1878fd5249",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Adjacency Matrices\n",
    "\n",
    "This is what most people first consider.  The rows and columns of a square matrix $A$ represent the vertices of the graph.  Then an entry $a_{ij}$ holds the value on the **directed** edge from $i$ to $j$.\n",
    "- Matrix-vector (mxv, vxm) and matrix-matrix (mxm) product gather data from paths $i \\rightarrow k \\rightarrow j$ in the output's entry for $i \\rightarrow j$.\n",
    "  - Natural mapping to breadth-first search and other path-oriented algorithms.\n",
    "- Undirected graphs store $i \\rightarrow j$ and $j \\rightarrow i$, leading to occassional confusion over the number of edges.\n",
    "  - An entry on the diagonal is a self-loop.  Is that one edge or two?\n",
    "  \n",
    "But this is the basis for the rest here and the most commonly used (IMHO)."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "167b593c-c2d0-4db8-b7ad-844969958659",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "#### Bipartite Variation\n",
    "\n",
    "Bipartite graphs in adjacency matrix form would have a block matrix structure: $[0, A; A^T, 0]$.\n",
    "\n",
    "Often this is compressed to simply representing $A$, where the rows correspond to one side of the graph and columns to the other.\n",
    "\n",
    "Multi-partite graphs often are broken into multiple such graphs.\n",
    "\n",
    "This is similar to \"optimizing\" structured matrices for mxv and mxm."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7215acbd-8843-4148-9c07-bff01cc63fcf",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "#### Depth Variation\n",
    "\n",
    "*Personal* view, not common usage:\n",
    "\n",
    "Sometimes you'll encounter algorithms that modify the adjacency matrix to force what appear to be self-loops (adding the identity matrix).\n",
    "\n",
    "In my experience, these often are not quite adjacency matrices but a slight variation.  \n",
    "- The matrix entries encode which vertices can be *reached* in at most a fixed number of edge traversals.  \n",
    "- The plain identity becomes the edges in at most zero hops.  \n",
    "- Adjacency plus identity: In two hops.\n",
    "- $A^k$: $k$ hops, used *e.g.* when computing the transitive closure."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0c6d3b45-e6a6-42dc-8613-ff1e62aa66b1",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Incidence Matrices\n",
    "\n",
    "Perhaps the most general and not often seen in code.\n",
    "- Rows (typically) represent vertices and columns **edges**.\n",
    "- Then two vertices $i$ an $j$ connected by edge $e$ are denoted by $a_{ie} = a_{je} = 1$ for an undirected graph.\n",
    "- Directed graphs use signed entries.  \n",
    "- Multigraphs: Multiple edges between the same two vertices just use multiple edge columns.\n",
    "- *Hypergraphs* have more than two entries in a column denoting that *all* the marked vertices are in the same hyperedge.\n",
    "\n",
    "This is terribly flexible.  Other forms can be computed by interpretations of $A A^{T}$ (adjacency), $A^T A$ (line graph / dual), and more.\n",
    "\n",
    "Incidence matrices appear in mathematic programming (linear optimizaton) formulations, non-backtracking walks, and algorithms in increasing levels of generality and complexity."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5bb62d5b-55fc-4dc1-a2e5-c624d1bbef6a",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Pitfalls in Sparse Matrix Representations\n",
    "\n",
    "Although the fit between entries in a sparse matrix and edges in a graph appears natural at first, the results often have been hacky:\n",
    "- Sparse matrices assume zero is the additive identity.\n",
    "- GNU Octave / MATLAB (TM) try not to store explicit zeros.\n",
    "\n",
    "But a graph either has an edge or does not.  The value is separate.  Often takes some mental adjustment.\n",
    "\n",
    "And nomenclature...  Try to use *vertices* rather than *nodes.*  The latter is confusing when combining with distributed-memory parallelism."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f93c1cad-0e50-4483-af29-530cea51a3f8",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Whirlwind Tour of the GraphBLAS\n",
    "\n",
    "### GraphBLAS objects\n",
    "- `GrB_Matrix`\n",
    "- `GrB_Vector` (a vertex set with values)\n",
    "- *masks*, an interpretation of the above\n",
    "- `GrB_Scalar` (in v2.0, `GxB_Scalar` prior)\n",
    "- Unary operators, binary operators\n",
    "- Monoids, semirings, *oh my*\n",
    "- `GrB_Descriptor`"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bc07d5dd-8b9a-4191-8dee-c78518aee70a",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### GraphBLAS operations\n",
    "\n",
    "- `GrB_mxm`, `GrB_vxm`, `GrB_mxv`\n",
    "- `GrB_eWiseMult`, `GrB_eWiseAdd` *(hic sunt dracones)*\n",
    "- `GrB_reduce`\n",
    "- submatrix assignment, extraction\n",
    "- object creation, finalizaton, building, and *access*\n",
    "- **and more...**"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e7b32803-002d-4097-8a65-cadc288630f4",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Early-on Pitfalls for New Users\n",
    "\n",
    "- Some of the terminology appears obtuse, but it actually is very well defined and considered.\n",
    "- Matrices have rows and columns.  Vectors ***do not***.  They only have a size.\n",
    "- Many graph and stats folks think of vxm (vector-matrix), linear algebra folks of mxv.\n",
    "  - Because adj. matrix rows often are the *source* of an edge...\n",
    "- Zeros are ***not necessarily*** \"additive\" identities for the monoids and semirings.\n",
    "  - The value does not determine the edge's existence.\n",
    "  - There are some, um, rules in corner cases about this.  It's \"fun\"."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1cd6f537-7b36-47c2-86da-a091b2fc5cab",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Anatomy of a GraphBLAS C Call\n",
    "\n",
    "``` C\n",
    "GrB_Info GrB_Matrix_new(GrB_Matrix *, GrB_Type, \n",
    "                        GrB_Index nrows, GrB_Index ncols);\n",
    "\n",
    "GrB_Info GrB_eWiseAdd(GrB_Vector w /* output */,\n",
    "    const GrB_Vector mask,\n",
    "    const GrB_BinaryOp accum,\n",
    "    const GrB_Semiring op /* overloads for binary op... */,\n",
    "    const GrB_Vector u /* operands */,\n",
    "    const GrB_Vector v,\n",
    "    const GrB_Descriptor desc);\n",
    "```\n",
    "\n",
    "*Generally* users are encouraged to stick to a similar style."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f8c5e759-241f-401e-bb16-7db6fa9ea47d",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Occasional Pitfalls\n",
    "\n",
    "GrB_Matrix and GrB_Vector *do not encode their entry types in their names*.  Indeed, there is not even a method for querying the stored type (iirc).\n",
    "- How to create a matrix with the same type as another?  `GrB_Matrix_dup`.\n",
    "- Vector with the same type?  *crickets*\n",
    "\n",
    "The operations, however, **do** include the type in the name, *e.g.* `GrB_PLUS_FP64`.\n",
    "- Pre-defined ones either are uniform in their input/output types or have an output type of `GrB_BOOL`.\n",
    "\n",
    "Casting rules are supposed to follow C.  Expect possible problems in non-SuiteSparse implementations...\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d45aedf8-735b-4117-9632-bdaa564a8700",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "#### Under the Hood (SuiteSparse):\n",
    "\n",
    "The joy of overloading in standard C.\n",
    "\n",
    "``` C\n",
    "#define GB_BIND(kind,x,y,...)                                                \\\n",
    "    _Generic                                                                 \\\n",
    "    (                                                                        \\\n",
    "        (x),                                                                 \\\n",
    "        const GrB_Scalar: GB_CONCAT ( GrB,_,kind,_apply_BinaryOp1st_Scalar), \\\n",
    "              GrB_Scalar: GB_CONCAT ( GrB,_,kind,_apply_BinaryOp1st_Scalar), \\\n",
    "        GB_CASES (, GrB, GB_CONCAT ( kind, _apply_BinaryOp1st,, )) ,         \\\n",
    "        default:                                                             \\\n",
    "            _Generic                                                         \\\n",
    "            (                                                                \\\n",
    "                (y),                                                         \\\n",
    "                GB_CASES (, GrB, GB_CONCAT ( kind , _apply_BinaryOp2nd,, )), \\\n",
    "                default:  GB_CONCAT ( GrB,_,kind,_apply_BinaryOp2nd_Scalar)  \\\n",
    "            )                                                                \\\n",
    "    )\n",
    "\n",
    "#define GB_IDXOP(kind,A,y,...)                                               \\\n",
    "    _Generic                                                                 \\\n",
    "    (                                                                        \\\n",
    "        (y),                                                                 \\\n",
    "            GB_CASES (, GrB, GB_CONCAT ( kind, _apply_IndexOp,, )),          \\\n",
    "            default:  GB_CONCAT ( GrB, _, kind, _apply_IndexOp_Scalar)       \\\n",
    "    )\n",
    "\n",
    "#define GrB_apply(C,Mask,accum,op,...)                                       \\\n",
    "    _Generic                                                                 \\\n",
    "    (                                                                        \\\n",
    "        (C),                                                                 \\\n",
    "            GrB_Vector :                                                     \\\n",
    "                _Generic                                                     \\\n",
    "                (                                                            \\\n",
    "                    (op),                                                    \\\n",
    "                        GrB_UnaryOp  : GrB_Vector_apply ,                    \\\n",
    "                        GrB_BinaryOp : GB_BIND (Vector, __VA_ARGS__),        \\\n",
    "                        GrB_IndexUnaryOp : GB_IDXOP (Vector, __VA_ARGS__)    \\\n",
    "                ),                                                           \\\n",
    "            GrB_Matrix :                                                     \\\n",
    "                _Generic                                                     \\\n",
    "                (                                                            \\\n",
    "                    (op),                                                    \\\n",
    "                        GrB_UnaryOp  : GrB_Matrix_apply ,                    \\\n",
    "                        GrB_BinaryOp : GB_BIND (Matrix, __VA_ARGS__),        \\\n",
    "                        GrB_IndexUnaryOp : GB_IDXOP (Matrix, __VA_ARGS__)    \\\n",
    "                )                                                            \\\n",
    "    )                                                                        \\\n",
    "    (C, Mask, accum, op, __VA_ARGS__)\n",
    "#endif\n",
    "```\n",
    "\n",
    "Perhaps you see why there are few implementations..."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d5af0aad-fcfa-48cf-a4f7-6f0f99698656",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Looking at Code\n",
    "\n",
    "Note:  The same code runs unchanged on SuiteSparse:GraphBLAS and LucataGraphBLAS.\n",
    "\n",
    "That is by careful selection for now.  LGB is not quite at v1.3 of the spec.\n",
    "\n",
    "### BFS\n",
    "\n",
    "This was quite the starting point (iirc) for Aydin Buluç's CombBLAS thesis work under John Gilbert."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "f629f888-e686-4b9d-8224-e8d67b45e3c5",
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "import os\n",
    "import sys\n",
    "#Used to display the code file directly within the notebook\n",
    "from IPython.display import Code\n",
    "import scipy as sp\n",
    "import scipy.io\n",
    "import io\n",
    "import networkx"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "aae464eb-3376-45b8-b9fe-aa8149232a27",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>pre { line-height: 125%; }\n",
       "td.linenos .normal { color: inherit; background-color: transparent; padding-left: 5px; padding-right: 5px; }\n",
       "span.linenos { color: inherit; background-color: transparent; padding-left: 5px; padding-right: 5px; }\n",
       "td.linenos .special { color: #000000; background-color: #ffffc0; padding-left: 5px; padding-right: 5px; }\n",
       "span.linenos.special { color: #000000; background-color: #ffffc0; padding-left: 5px; padding-right: 5px; }\n",
       ".output_html .hll { background-color: #ffffcc }\n",
       ".output_html { background: #f8f8f8; }\n",
       ".output_html .c { color: #3D7B7B; font-style: italic } /* Comment */\n",
       ".output_html .err { border: 1px solid #FF0000 } /* Error */\n",
       ".output_html .k { color: #008000; font-weight: bold } /* Keyword */\n",
       ".output_html .o { color: #666666 } /* Operator */\n",
       ".output_html .ch { color: #3D7B7B; font-style: italic } /* Comment.Hashbang */\n",
       ".output_html .cm { color: #3D7B7B; font-style: italic } /* Comment.Multiline */\n",
       ".output_html .cp { color: #9C6500 } /* Comment.Preproc */\n",
       ".output_html .cpf { color: #3D7B7B; font-style: italic } /* Comment.PreprocFile */\n",
       ".output_html .c1 { color: #3D7B7B; font-style: italic } /* Comment.Single */\n",
       ".output_html .cs { color: #3D7B7B; font-style: italic } /* Comment.Special */\n",
       ".output_html .gd { color: #A00000 } /* Generic.Deleted */\n",
       ".output_html .ge { font-style: italic } /* Generic.Emph */\n",
       ".output_html .gr { color: #E40000 } /* Generic.Error */\n",
       ".output_html .gh { color: #000080; font-weight: bold } /* Generic.Heading */\n",
       ".output_html .gi { color: #008400 } /* Generic.Inserted */\n",
       ".output_html .go { color: #717171 } /* Generic.Output */\n",
       ".output_html .gp { color: #000080; font-weight: bold } /* Generic.Prompt */\n",
       ".output_html .gs { font-weight: bold } /* Generic.Strong */\n",
       ".output_html .gu { color: #800080; font-weight: bold } /* Generic.Subheading */\n",
       ".output_html .gt { color: #0044DD } /* Generic.Traceback */\n",
       ".output_html .kc { color: #008000; font-weight: bold } /* Keyword.Constant */\n",
       ".output_html .kd { color: #008000; font-weight: bold } /* Keyword.Declaration */\n",
       ".output_html .kn { color: #008000; font-weight: bold } /* Keyword.Namespace */\n",
       ".output_html .kp { color: #008000 } /* Keyword.Pseudo */\n",
       ".output_html .kr { color: #008000; font-weight: bold } /* Keyword.Reserved */\n",
       ".output_html .kt { color: #B00040 } /* Keyword.Type */\n",
       ".output_html .m { color: #666666 } /* Literal.Number */\n",
       ".output_html .s { color: #BA2121 } /* Literal.String */\n",
       ".output_html .na { color: #687822 } /* Name.Attribute */\n",
       ".output_html .nb { color: #008000 } /* Name.Builtin */\n",
       ".output_html .nc { color: #0000FF; font-weight: bold } /* Name.Class */\n",
       ".output_html .no { color: #880000 } /* Name.Constant */\n",
       ".output_html .nd { color: #AA22FF } /* Name.Decorator */\n",
       ".output_html .ni { color: #717171; font-weight: bold } /* Name.Entity */\n",
       ".output_html .ne { color: #CB3F38; font-weight: bold } /* Name.Exception */\n",
       ".output_html .nf { color: #0000FF } /* Name.Function */\n",
       ".output_html .nl { color: #767600 } /* Name.Label */\n",
       ".output_html .nn { color: #0000FF; font-weight: bold } /* Name.Namespace */\n",
       ".output_html .nt { color: #008000; font-weight: bold } /* Name.Tag */\n",
       ".output_html .nv { color: #19177C } /* Name.Variable */\n",
       ".output_html .ow { color: #AA22FF; font-weight: bold } /* Operator.Word */\n",
       ".output_html .w { color: #bbbbbb } /* Text.Whitespace */\n",
       ".output_html .mb { color: #666666 } /* Literal.Number.Bin */\n",
       ".output_html .mf { color: #666666 } /* Literal.Number.Float */\n",
       ".output_html .mh { color: #666666 } /* Literal.Number.Hex */\n",
       ".output_html .mi { color: #666666 } /* Literal.Number.Integer */\n",
       ".output_html .mo { color: #666666 } /* Literal.Number.Oct */\n",
       ".output_html .sa { color: #BA2121 } /* Literal.String.Affix */\n",
       ".output_html .sb { color: #BA2121 } /* Literal.String.Backtick */\n",
       ".output_html .sc { color: #BA2121 } /* Literal.String.Char */\n",
       ".output_html .dl { color: #BA2121 } /* Literal.String.Delimiter */\n",
       ".output_html .sd { color: #BA2121; font-style: italic } /* Literal.String.Doc */\n",
       ".output_html .s2 { color: #BA2121 } /* Literal.String.Double */\n",
       ".output_html .se { color: #AA5D1F; font-weight: bold } /* Literal.String.Escape */\n",
       ".output_html .sh { color: #BA2121 } /* Literal.String.Heredoc */\n",
       ".output_html .si { color: #A45A77; font-weight: bold } /* Literal.String.Interpol */\n",
       ".output_html .sx { color: #008000 } /* Literal.String.Other */\n",
       ".output_html .sr { color: #A45A77 } /* Literal.String.Regex */\n",
       ".output_html .s1 { color: #BA2121 } /* Literal.String.Single */\n",
       ".output_html .ss { color: #19177C } /* Literal.String.Symbol */\n",
       ".output_html .bp { color: #008000 } /* Name.Builtin.Pseudo */\n",
       ".output_html .fm { color: #0000FF } /* Name.Function.Magic */\n",
       ".output_html .vc { color: #19177C } /* Name.Variable.Class */\n",
       ".output_html .vg { color: #19177C } /* Name.Variable.Global */\n",
       ".output_html .vi { color: #19177C } /* Name.Variable.Instance */\n",
       ".output_html .vm { color: #19177C } /* Name.Variable.Magic */\n",
       ".output_html .il { color: #666666 } /* Literal.Number.Integer.Long */</style><div class=\"highlight\"><pre><span></span><span class=\"cp\">#include</span><span class=\"w\"> </span><span class=\"cpf\">&lt;GraphBLAS.h&gt;</span><span class=\"cp\"></span>\n",
       "\n",
       "<span class=\"kt\">int</span><span class=\"w\"> </span><span class=\"nf\">bfs</span><span class=\"p\">(</span><span class=\"n\">GrB_Vector</span><span class=\"w\"> </span><span class=\"o\">*</span><span class=\"n\">level</span><span class=\"p\">,</span><span class=\"w\"></span>\n",
       "<span class=\"w\">        </span><span class=\"n\">GrB_Matrix</span><span class=\"w\"> </span><span class=\"n\">A</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_Index</span><span class=\"w\"> </span><span class=\"n\">source</span><span class=\"p\">,</span><span class=\"w\"></span>\n",
       "<span class=\"w\">        </span><span class=\"k\">const</span><span class=\"w\"> </span><span class=\"n\">GrB_Index</span><span class=\"w\"> </span><span class=\"n\">max_level</span><span class=\"p\">)</span><span class=\"w\"></span>\n",
       "<span class=\"p\">{</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_Index</span><span class=\"w\"> </span><span class=\"n\">N</span><span class=\"p\">;</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_Matrix_nrows</span><span class=\"p\">(</span><span class=\"o\">&amp;</span><span class=\"n\">N</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">A</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"w\">  </span><span class=\"k\">if</span><span class=\"w\"> </span><span class=\"p\">(</span><span class=\"n\">N</span><span class=\"w\"> </span><span class=\"o\">==</span><span class=\"w\"> </span><span class=\"mi\">0</span><span class=\"p\">)</span><span class=\"w\"> </span><span class=\"k\">return</span><span class=\"w\"> </span><span class=\"mi\">0</span><span class=\"p\">;</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_Vector_new</span><span class=\"p\">(</span><span class=\"n\">level</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_INT64</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">N</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_Vector</span><span class=\"w\"> </span><span class=\"n\">q</span><span class=\"p\">;</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_Vector_new</span><span class=\"p\">(</span><span class=\"o\">&amp;</span><span class=\"n\">q</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_BOOL</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">N</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_Vector_setElement</span><span class=\"p\">(</span><span class=\"n\">q</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"mi\">1</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">source</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"w\">  </span><span class=\"kt\">int64_t</span><span class=\"w\"> </span><span class=\"n\">depth</span><span class=\"p\">;</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"k\">for</span><span class=\"w\"> </span><span class=\"p\">(</span><span class=\"n\">depth</span><span class=\"w\"> </span><span class=\"o\">=</span><span class=\"w\"> </span><span class=\"mi\">2</span><span class=\"p\">;</span><span class=\"w\"> </span><span class=\"n\">depth</span><span class=\"w\"> </span><span class=\"o\">&lt;=</span><span class=\"w\"> </span><span class=\"n\">max_level</span><span class=\"p\">;</span><span class=\"w\"> </span><span class=\"o\">++</span><span class=\"n\">depth</span><span class=\"p\">)</span><span class=\"w\"> </span><span class=\"p\">{</span><span class=\"w\"></span>\n",
       "<span class=\"w\">    </span><span class=\"n\">GrB_assign</span><span class=\"p\">(</span><span class=\"o\">*</span><span class=\"n\">level</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">q</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_NULL</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">depth</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_ALL</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">N</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_NULL</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"w\">    </span><span class=\"n\">GrB_vxm</span><span class=\"p\">(</span><span class=\"n\">q</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"o\">*</span><span class=\"n\">level</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_NULL</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_LOR_LAND_SEMIRING_BOOL</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">q</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">A</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_DESC_RC</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"w\">    </span><span class=\"kt\">bool</span><span class=\"w\"> </span><span class=\"n\">found_more</span><span class=\"p\">;</span><span class=\"w\"></span>\n",
       "<span class=\"w\">    </span><span class=\"n\">GrB_reduce</span><span class=\"p\">(</span><span class=\"o\">&amp;</span><span class=\"n\">found_more</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_NULL</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_LOR_MONOID_BOOL</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">q</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_NULL</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"w\">    </span><span class=\"k\">if</span><span class=\"w\"> </span><span class=\"p\">(</span><span class=\"o\">!</span><span class=\"n\">found_more</span><span class=\"p\">)</span><span class=\"w\"> </span><span class=\"k\">break</span><span class=\"p\">;</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"p\">}</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_free</span><span class=\"p\">(</span><span class=\"o\">&amp;</span><span class=\"n\">q</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span>\n",
       "<span class=\"w\">  </span><span class=\"k\">return</span><span class=\"w\"> </span><span class=\"n\">depth</span><span class=\"p\">;</span><span class=\"w\"></span>\n",
       "<span class=\"p\">}</span><span class=\"w\"></span>\n",
       "</pre></div>\n"
      ],
      "text/latex": [
       "\\begin{Verbatim}[commandchars=\\\\\\{\\}]\n",
       "\\PY{c+cp}{\\PYZsh{}}\\PY{c+cp}{include}\\PY{+w}{ }\\PY{c+cpf}{\\PYZlt{}GraphBLAS.h\\PYZgt{}}\n",
       "\n",
       "\\PY{k+kt}{int}\\PY{+w}{ }\\PY{n+nf}{bfs}\\PY{p}{(}\\PY{n}{GrB\\PYZus{}Vector}\\PY{+w}{ }\\PY{o}{*}\\PY{n}{level}\\PY{p}{,}\n",
       "\\PY{+w}{        }\\PY{n}{GrB\\PYZus{}Matrix}\\PY{+w}{ }\\PY{n}{A}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}Index}\\PY{+w}{ }\\PY{n}{source}\\PY{p}{,}\n",
       "\\PY{+w}{        }\\PY{k}{const}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}Index}\\PY{+w}{ }\\PY{n}{max\\PYZus{}level}\\PY{p}{)}\n",
       "\\PY{p}{\\PYZob{}}\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}Index}\\PY{+w}{ }\\PY{n}{N}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}Matrix\\PYZus{}nrows}\\PY{p}{(}\\PY{o}{\\PYZam{}}\\PY{n}{N}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{A}\\PY{p}{)}\\PY{p}{;}\n",
       "\n",
       "\\PY{+w}{  }\\PY{k}{if}\\PY{+w}{ }\\PY{p}{(}\\PY{n}{N}\\PY{+w}{ }\\PY{o}{=}\\PY{o}{=}\\PY{+w}{ }\\PY{l+m+mi}{0}\\PY{p}{)}\\PY{+w}{ }\\PY{k}{return}\\PY{+w}{ }\\PY{l+m+mi}{0}\\PY{p}{;}\n",
       "\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}Vector\\PYZus{}new}\\PY{p}{(}\\PY{n}{level}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}INT64}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{N}\\PY{p}{)}\\PY{p}{;}\n",
       "\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}Vector}\\PY{+w}{ }\\PY{n}{q}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}Vector\\PYZus{}new}\\PY{p}{(}\\PY{o}{\\PYZam{}}\\PY{n}{q}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}BOOL}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{N}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}Vector\\PYZus{}setElement}\\PY{p}{(}\\PY{n}{q}\\PY{p}{,}\\PY{+w}{ }\\PY{l+m+mi}{1}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{source}\\PY{p}{)}\\PY{p}{;}\n",
       "\n",
       "\\PY{+w}{  }\\PY{k+kt}{int64\\PYZus{}t}\\PY{+w}{ }\\PY{n}{depth}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{k}{for}\\PY{+w}{ }\\PY{p}{(}\\PY{n}{depth}\\PY{+w}{ }\\PY{o}{=}\\PY{+w}{ }\\PY{l+m+mi}{2}\\PY{p}{;}\\PY{+w}{ }\\PY{n}{depth}\\PY{+w}{ }\\PY{o}{\\PYZlt{}}\\PY{o}{=}\\PY{+w}{ }\\PY{n}{max\\PYZus{}level}\\PY{p}{;}\\PY{+w}{ }\\PY{o}{+}\\PY{o}{+}\\PY{n}{depth}\\PY{p}{)}\\PY{+w}{ }\\PY{p}{\\PYZob{}}\n",
       "\\PY{+w}{    }\\PY{n}{GrB\\PYZus{}assign}\\PY{p}{(}\\PY{o}{*}\\PY{n}{level}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{q}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}NULL}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{depth}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}ALL}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{N}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}NULL}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{    }\\PY{n}{GrB\\PYZus{}vxm}\\PY{p}{(}\\PY{n}{q}\\PY{p}{,}\\PY{+w}{ }\\PY{o}{*}\\PY{n}{level}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}NULL}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}LOR\\PYZus{}LAND\\PYZus{}SEMIRING\\PYZus{}BOOL}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{q}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{A}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}DESC\\PYZus{}RC}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{    }\\PY{k+kt}{bool}\\PY{+w}{ }\\PY{n}{found\\PYZus{}more}\\PY{p}{;}\n",
       "\\PY{+w}{    }\\PY{n}{GrB\\PYZus{}reduce}\\PY{p}{(}\\PY{o}{\\PYZam{}}\\PY{n}{found\\PYZus{}more}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}NULL}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}LOR\\PYZus{}MONOID\\PYZus{}BOOL}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{q}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}NULL}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{    }\\PY{k}{if}\\PY{+w}{ }\\PY{p}{(}\\PY{o}{!}\\PY{n}{found\\PYZus{}more}\\PY{p}{)}\\PY{+w}{ }\\PY{k}{break}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{p}{\\PYZcb{}}\n",
       "\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}free}\\PY{p}{(}\\PY{o}{\\PYZam{}}\\PY{n}{q}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{  }\n",
       "\\PY{+w}{  }\\PY{k}{return}\\PY{+w}{ }\\PY{n}{depth}\\PY{p}{;}\n",
       "\\PY{p}{\\PYZcb{}}\n",
       "\\end{Verbatim}\n"
      ],
      "text/plain": [
       "#include <GraphBLAS.h>\n",
       "\n",
       "int bfs(GrB_Vector *level,\n",
       "        GrB_Matrix A, GrB_Index source,\n",
       "        const GrB_Index max_level)\n",
       "{\n",
       "  GrB_Index N;\n",
       "  GrB_Matrix_nrows(&N, A);\n",
       "\n",
       "  if (N == 0) return 0;\n",
       "\n",
       "  GrB_Vector_new(level, GrB_INT64, N);\n",
       "\n",
       "  GrB_Vector q;\n",
       "  GrB_Vector_new(&q, GrB_BOOL, N);\n",
       "  GrB_Vector_setElement(q, 1, source);\n",
       "\n",
       "  int64_t depth;\n",
       "  for (depth = 2; depth <= max_level; ++depth) {\n",
       "    GrB_assign(*level, q, GrB_NULL, depth, GrB_ALL, N, GrB_NULL);\n",
       "    GrB_vxm(q, *level, GrB_NULL, GrB_LOR_LAND_SEMIRING_BOOL, q, A, GrB_DESC_RC);\n",
       "    bool found_more;\n",
       "    GrB_reduce(&found_more, GrB_NULL, GrB_LOR_MONOID_BOOL, q, GrB_NULL);\n",
       "    if (!found_more) break;\n",
       "  }\n",
       "\n",
       "  GrB_free(&q);\n",
       "  \n",
       "  return depth;\n",
       "}"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Code('bfs.c')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "477fefb1-9d96-43d9-ab31-37952c523921",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### PageRank\n",
    "\n",
    "The following implements a Jacobi solver for $x (I - \\alpha D^\\dagger A) = (1 - \\alpha) v$ where $v$ is an input probability distribution on vertices, and $\\alpha$ is the \"teleportation\" constant.\n",
    "- This follows [Gleich, D., L. Zhukov, and P. Berkhin. \"Fast Parallel PageRank: A Linear System Approach.\"](https://scholar.google.com/citations?view_op=view_citation&hl=en&user=5m5ds6UAAAAJ&citation_for_view=5m5ds6UAAAAJ:u5HHmVD_uO8C)\n",
    "- Dangling vertices (no outgoing edges) are handled implicitly, see reference\n",
    "  - Note: Pseudo-inverse: The empty rows are just scaled by zero.\n",
    "\n",
    "(Yes, linear algebra in the language of linear algebra.)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "0a96d95d-ae77-438e-a21d-a1a1fd8a6e53",
   "metadata": {
    "scrolled": true,
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>pre { line-height: 125%; }\n",
       "td.linenos .normal { color: inherit; background-color: transparent; padding-left: 5px; padding-right: 5px; }\n",
       "span.linenos { color: inherit; background-color: transparent; padding-left: 5px; padding-right: 5px; }\n",
       "td.linenos .special { color: #000000; background-color: #ffffc0; padding-left: 5px; padding-right: 5px; }\n",
       "span.linenos.special { color: #000000; background-color: #ffffc0; padding-left: 5px; padding-right: 5px; }\n",
       ".output_html .hll { background-color: #ffffcc }\n",
       ".output_html { background: #f8f8f8; }\n",
       ".output_html .c { color: #3D7B7B; font-style: italic } /* Comment */\n",
       ".output_html .err { border: 1px solid #FF0000 } /* Error */\n",
       ".output_html .k { color: #008000; font-weight: bold } /* Keyword */\n",
       ".output_html .o { color: #666666 } /* Operator */\n",
       ".output_html .ch { color: #3D7B7B; font-style: italic } /* Comment.Hashbang */\n",
       ".output_html .cm { color: #3D7B7B; font-style: italic } /* Comment.Multiline */\n",
       ".output_html .cp { color: #9C6500 } /* Comment.Preproc */\n",
       ".output_html .cpf { color: #3D7B7B; font-style: italic } /* Comment.PreprocFile */\n",
       ".output_html .c1 { color: #3D7B7B; font-style: italic } /* Comment.Single */\n",
       ".output_html .cs { color: #3D7B7B; font-style: italic } /* Comment.Special */\n",
       ".output_html .gd { color: #A00000 } /* Generic.Deleted */\n",
       ".output_html .ge { font-style: italic } /* Generic.Emph */\n",
       ".output_html .gr { color: #E40000 } /* Generic.Error */\n",
       ".output_html .gh { color: #000080; font-weight: bold } /* Generic.Heading */\n",
       ".output_html .gi { color: #008400 } /* Generic.Inserted */\n",
       ".output_html .go { color: #717171 } /* Generic.Output */\n",
       ".output_html .gp { color: #000080; font-weight: bold } /* Generic.Prompt */\n",
       ".output_html .gs { font-weight: bold } /* Generic.Strong */\n",
       ".output_html .gu { color: #800080; font-weight: bold } /* Generic.Subheading */\n",
       ".output_html .gt { color: #0044DD } /* Generic.Traceback */\n",
       ".output_html .kc { color: #008000; font-weight: bold } /* Keyword.Constant */\n",
       ".output_html .kd { color: #008000; font-weight: bold } /* Keyword.Declaration */\n",
       ".output_html .kn { color: #008000; font-weight: bold } /* Keyword.Namespace */\n",
       ".output_html .kp { color: #008000 } /* Keyword.Pseudo */\n",
       ".output_html .kr { color: #008000; font-weight: bold } /* Keyword.Reserved */\n",
       ".output_html .kt { color: #B00040 } /* Keyword.Type */\n",
       ".output_html .m { color: #666666 } /* Literal.Number */\n",
       ".output_html .s { color: #BA2121 } /* Literal.String */\n",
       ".output_html .na { color: #687822 } /* Name.Attribute */\n",
       ".output_html .nb { color: #008000 } /* Name.Builtin */\n",
       ".output_html .nc { color: #0000FF; font-weight: bold } /* Name.Class */\n",
       ".output_html .no { color: #880000 } /* Name.Constant */\n",
       ".output_html .nd { color: #AA22FF } /* Name.Decorator */\n",
       ".output_html .ni { color: #717171; font-weight: bold } /* Name.Entity */\n",
       ".output_html .ne { color: #CB3F38; font-weight: bold } /* Name.Exception */\n",
       ".output_html .nf { color: #0000FF } /* Name.Function */\n",
       ".output_html .nl { color: #767600 } /* Name.Label */\n",
       ".output_html .nn { color: #0000FF; font-weight: bold } /* Name.Namespace */\n",
       ".output_html .nt { color: #008000; font-weight: bold } /* Name.Tag */\n",
       ".output_html .nv { color: #19177C } /* Name.Variable */\n",
       ".output_html .ow { color: #AA22FF; font-weight: bold } /* Operator.Word */\n",
       ".output_html .w { color: #bbbbbb } /* Text.Whitespace */\n",
       ".output_html .mb { color: #666666 } /* Literal.Number.Bin */\n",
       ".output_html .mf { color: #666666 } /* Literal.Number.Float */\n",
       ".output_html .mh { color: #666666 } /* Literal.Number.Hex */\n",
       ".output_html .mi { color: #666666 } /* Literal.Number.Integer */\n",
       ".output_html .mo { color: #666666 } /* Literal.Number.Oct */\n",
       ".output_html .sa { color: #BA2121 } /* Literal.String.Affix */\n",
       ".output_html .sb { color: #BA2121 } /* Literal.String.Backtick */\n",
       ".output_html .sc { color: #BA2121 } /* Literal.String.Char */\n",
       ".output_html .dl { color: #BA2121 } /* Literal.String.Delimiter */\n",
       ".output_html .sd { color: #BA2121; font-style: italic } /* Literal.String.Doc */\n",
       ".output_html .s2 { color: #BA2121 } /* Literal.String.Double */\n",
       ".output_html .se { color: #AA5D1F; font-weight: bold } /* Literal.String.Escape */\n",
       ".output_html .sh { color: #BA2121 } /* Literal.String.Heredoc */\n",
       ".output_html .si { color: #A45A77; font-weight: bold } /* Literal.String.Interpol */\n",
       ".output_html .sx { color: #008000 } /* Literal.String.Other */\n",
       ".output_html .sr { color: #A45A77 } /* Literal.String.Regex */\n",
       ".output_html .s1 { color: #BA2121 } /* Literal.String.Single */\n",
       ".output_html .ss { color: #19177C } /* Literal.String.Symbol */\n",
       ".output_html .bp { color: #008000 } /* Name.Builtin.Pseudo */\n",
       ".output_html .fm { color: #0000FF } /* Name.Function.Magic */\n",
       ".output_html .vc { color: #19177C } /* Name.Variable.Class */\n",
       ".output_html .vg { color: #19177C } /* Name.Variable.Global */\n",
       ".output_html .vi { color: #19177C } /* Name.Variable.Instance */\n",
       ".output_html .vm { color: #19177C } /* Name.Variable.Magic */\n",
       ".output_html .il { color: #666666 } /* Literal.Number.Integer.Long */</style><div class=\"highlight\"><pre><span></span><span class=\"cp\">#include</span><span class=\"w\"> </span><span class=\"cpf\">&lt;GraphBLAS.h&gt;</span><span class=\"cp\"></span>\n",
       "\n",
       "<span class=\"k\">extern</span><span class=\"w\"> </span><span class=\"kt\">void</span><span class=\"w\"> </span><span class=\"nf\">degree_pseudoinv</span><span class=\"p\">(</span><span class=\"n\">GrB_Vector</span><span class=\"o\">*</span><span class=\"w\"> </span><span class=\"n\">v</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_Matrix</span><span class=\"w\"> </span><span class=\"n\">A</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"k\">extern</span><span class=\"w\"> </span><span class=\"kt\">void</span><span class=\"w\"> </span><span class=\"nf\">scale_vector</span><span class=\"p\">(</span><span class=\"n\">GrB_Vector</span><span class=\"w\"> </span><span class=\"n\">v</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"kt\">double</span><span class=\"w\"> </span><span class=\"n\">scl</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"kt\">int</span><span class=\"w\"> </span><span class=\"nf\">pagerank</span><span class=\"p\">(</span><span class=\"n\">GrB_Vector</span><span class=\"w\"> </span><span class=\"o\">*</span><span class=\"n\">pr</span><span class=\"p\">,</span><span class=\"w\"></span>\n",
       "<span class=\"w\">             </span><span class=\"n\">GrB_Matrix</span><span class=\"w\"> </span><span class=\"n\">A</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_Vector</span><span class=\"w\"> </span><span class=\"n\">v</span><span class=\"p\">,</span><span class=\"w\"></span>\n",
       "<span class=\"w\">             </span><span class=\"k\">const</span><span class=\"w\"> </span><span class=\"kt\">double</span><span class=\"w\"> </span><span class=\"n\">alpha</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"k\">const</span><span class=\"w\"> </span><span class=\"kt\">double</span><span class=\"w\"> </span><span class=\"n\">ctol</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"k\">const</span><span class=\"w\"> </span><span class=\"kt\">int</span><span class=\"w\"> </span><span class=\"n\">itmax</span><span class=\"p\">)</span><span class=\"w\"></span>\n",
       "<span class=\"p\">{</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_Vector</span><span class=\"w\"> </span><span class=\"n\">v_scaled</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">x</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">xprev</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">accum</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">D_pseudoinv</span><span class=\"p\">;</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"kt\">int</span><span class=\"w\"> </span><span class=\"n\">k</span><span class=\"w\"> </span><span class=\"o\">=</span><span class=\"w\"> </span><span class=\"mi\">-1</span><span class=\"p\">;</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_Index</span><span class=\"w\"> </span><span class=\"n\">N</span><span class=\"p\">;</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_Matrix_nrows</span><span class=\"p\">(</span><span class=\"o\">&amp;</span><span class=\"n\">N</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">A</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"k\">if</span><span class=\"w\"> </span><span class=\"p\">(</span><span class=\"n\">N</span><span class=\"w\"> </span><span class=\"o\">==</span><span class=\"w\"> </span><span class=\"mi\">0</span><span class=\"p\">)</span><span class=\"w\"> </span><span class=\"k\">return</span><span class=\"w\"> </span><span class=\"mi\">0</span><span class=\"p\">;</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"w\">  </span><span class=\"p\">{</span><span class=\"w\"></span>\n",
       "<span class=\"w\">    </span><span class=\"n\">GrB_Index</span><span class=\"w\"> </span><span class=\"n\">NE</span><span class=\"p\">;</span><span class=\"w\"></span>\n",
       "<span class=\"w\">    </span><span class=\"n\">GrB_Matrix_nvals</span><span class=\"p\">(</span><span class=\"o\">&amp;</span><span class=\"n\">NE</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">A</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"w\">    </span><span class=\"k\">if</span><span class=\"w\"> </span><span class=\"p\">(</span><span class=\"n\">NE</span><span class=\"w\"> </span><span class=\"o\">==</span><span class=\"w\"> </span><span class=\"mi\">0</span><span class=\"p\">)</span><span class=\"w\"> </span><span class=\"k\">return</span><span class=\"w\"> </span><span class=\"mi\">0</span><span class=\"p\">;</span><span class=\"w\"></span>\n",
       "<span class=\"w\">    </span><span class=\"n\">GrB_Vector_nvals</span><span class=\"p\">(</span><span class=\"o\">&amp;</span><span class=\"n\">NE</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">v</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"w\">    </span><span class=\"k\">if</span><span class=\"w\"> </span><span class=\"p\">(</span><span class=\"n\">NE</span><span class=\"w\"> </span><span class=\"o\">==</span><span class=\"w\"> </span><span class=\"mi\">0</span><span class=\"p\">)</span><span class=\"w\"> </span><span class=\"k\">return</span><span class=\"w\"> </span><span class=\"mi\">0</span><span class=\"p\">;</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"p\">}</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_Vector_new</span><span class=\"p\">(</span><span class=\"o\">&amp;</span><span class=\"n\">v_scaled</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_FP64</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">N</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_Vector_new</span><span class=\"p\">(</span><span class=\"o\">&amp;</span><span class=\"n\">x</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_FP64</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">N</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_Vector_new</span><span class=\"p\">(</span><span class=\"o\">&amp;</span><span class=\"n\">xprev</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_FP64</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">N</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_Vector_new</span><span class=\"p\">(</span><span class=\"o\">&amp;</span><span class=\"n\">accum</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_FP64</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">N</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_Vector_new</span><span class=\"p\">(</span><span class=\"o\">&amp;</span><span class=\"n\">D_pseudoinv</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_FP64</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">N</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"w\">  </span><span class=\"n\">degree_pseudoinv</span><span class=\"p\">(</span><span class=\"o\">&amp;</span><span class=\"n\">D_pseudoinv</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">A</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_Vector_dup</span><span class=\"p\">(</span><span class=\"o\">&amp;</span><span class=\"n\">v_scaled</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">v</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"n\">scale_vector</span><span class=\"p\">(</span><span class=\"n\">v_scaled</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"mf\">1.0</span><span class=\"w\"> </span><span class=\"o\">-</span><span class=\"w\"> </span><span class=\"n\">alpha</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"w\">  </span><span class=\"k\">for</span><span class=\"w\"> </span><span class=\"p\">(</span><span class=\"n\">k</span><span class=\"w\"> </span><span class=\"o\">=</span><span class=\"w\"> </span><span class=\"mi\">0</span><span class=\"p\">;</span><span class=\"w\"> </span><span class=\"n\">k</span><span class=\"w\"> </span><span class=\"o\">&lt;</span><span class=\"w\"> </span><span class=\"n\">itmax</span><span class=\"p\">;</span><span class=\"w\"> </span><span class=\"o\">++</span><span class=\"n\">k</span><span class=\"p\">)</span><span class=\"w\"> </span><span class=\"p\">{</span><span class=\"w\"></span>\n",
       "<span class=\"w\">    </span><span class=\"n\">GrB_assign</span><span class=\"p\">(</span><span class=\"n\">xprev</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_NULL</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_NULL</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">x</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_ALL</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">N</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_DESC_R</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"w\">    </span><span class=\"n\">GrB_vxm</span><span class=\"p\">(</span><span class=\"n\">accum</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_NULL</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_NULL</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_PLUS_TIMES_SEMIRING_FP64</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">x</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">A</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_DESC_R</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"w\">    </span><span class=\"n\">GrB_eWiseAdd</span><span class=\"p\">(</span><span class=\"n\">accum</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_NULL</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_NULL</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_TIMES_FP64</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">D_pseudoinv</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">accum</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_DESC_R</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"w\">    </span><span class=\"n\">scale_vector</span><span class=\"p\">(</span><span class=\"n\">accum</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">alpha</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"w\">    </span><span class=\"n\">GrB_eWiseAdd</span><span class=\"p\">(</span><span class=\"n\">accum</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_NULL</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_NULL</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_PLUS_FP64</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">v_scaled</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">accum</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_DESC_R</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"w\">    </span><span class=\"n\">GrB_assign</span><span class=\"p\">(</span><span class=\"n\">x</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_NULL</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_NULL</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">accum</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_ALL</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">N</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_DESC_R</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"w\">    </span><span class=\"c1\">// Re-use xprev for &quot;convergence&quot; check</span>\n",
       "<span class=\"w\">    </span><span class=\"n\">GrB_eWiseAdd</span><span class=\"p\">(</span><span class=\"n\">xprev</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_NULL</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_NULL</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_MINUS_FP64</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">accum</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">xprev</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_DESC_R</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"w\">    </span><span class=\"n\">GrB_apply</span><span class=\"p\">(</span><span class=\"n\">xprev</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_NULL</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_NULL</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_ABS_FP64</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">xprev</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_DESC_R</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"w\">    </span><span class=\"kt\">double</span><span class=\"w\"> </span><span class=\"n\">diff</span><span class=\"w\"> </span><span class=\"o\">=</span><span class=\"w\"> </span><span class=\"mf\">0.0</span><span class=\"p\">;</span><span class=\"w\"></span>\n",
       "<span class=\"w\">    </span><span class=\"n\">GrB_reduce</span><span class=\"p\">(</span><span class=\"o\">&amp;</span><span class=\"n\">diff</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_NULL</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_MAX_MONOID_FP64</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">xprev</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_NULL</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"w\">    </span><span class=\"c1\">//fprintf(stderr, &quot;norm1 diff %lg\\n&quot;, diff);</span>\n",
       "<span class=\"w\">    </span><span class=\"k\">if</span><span class=\"w\"> </span><span class=\"p\">(</span><span class=\"n\">diff</span><span class=\"w\"> </span><span class=\"o\">&lt;=</span><span class=\"w\"> </span><span class=\"n\">ctol</span><span class=\"p\">)</span><span class=\"w\"> </span><span class=\"k\">break</span><span class=\"p\">;</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"p\">}</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"w\">  </span><span class=\"kt\">double</span><span class=\"w\"> </span><span class=\"n\">sum</span><span class=\"p\">;</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_reduce</span><span class=\"p\">(</span><span class=\"o\">&amp;</span><span class=\"n\">sum</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_NULL</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_PLUS_MONOID_FP64</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">x</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_NULL</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"k\">if</span><span class=\"w\"> </span><span class=\"p\">(</span><span class=\"n\">sum</span><span class=\"w\"> </span><span class=\"o\">!=</span><span class=\"w\"> </span><span class=\"mi\">0</span><span class=\"p\">)</span><span class=\"w\"></span>\n",
       "<span class=\"w\">    </span><span class=\"n\">scale_vector</span><span class=\"p\">(</span><span class=\"n\">x</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"mf\">1.0</span><span class=\"o\">/</span><span class=\"n\">sum</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_Vector_dup</span><span class=\"p\">(</span><span class=\"n\">pr</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">x</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_free</span><span class=\"p\">(</span><span class=\"o\">&amp;</span><span class=\"n\">D_pseudoinv</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_free</span><span class=\"p\">(</span><span class=\"o\">&amp;</span><span class=\"n\">accum</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_free</span><span class=\"p\">(</span><span class=\"o\">&amp;</span><span class=\"n\">xprev</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_free</span><span class=\"p\">(</span><span class=\"o\">&amp;</span><span class=\"n\">x</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_free</span><span class=\"p\">(</span><span class=\"o\">&amp;</span><span class=\"n\">v_scaled</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"w\">  </span><span class=\"k\">return</span><span class=\"w\"> </span><span class=\"n\">k</span><span class=\"p\">;</span><span class=\"w\"></span>\n",
       "<span class=\"p\">}</span><span class=\"w\"></span>\n",
       "</pre></div>\n"
      ],
      "text/latex": [
       "\\begin{Verbatim}[commandchars=\\\\\\{\\}]\n",
       "\\PY{c+cp}{\\PYZsh{}}\\PY{c+cp}{include}\\PY{+w}{ }\\PY{c+cpf}{\\PYZlt{}GraphBLAS.h\\PYZgt{}}\n",
       "\n",
       "\\PY{k}{extern}\\PY{+w}{ }\\PY{k+kt}{void}\\PY{+w}{ }\\PY{n+nf}{degree\\PYZus{}pseudoinv}\\PY{p}{(}\\PY{n}{GrB\\PYZus{}Vector}\\PY{o}{*}\\PY{+w}{ }\\PY{n}{v}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}Matrix}\\PY{+w}{ }\\PY{n}{A}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{k}{extern}\\PY{+w}{ }\\PY{k+kt}{void}\\PY{+w}{ }\\PY{n+nf}{scale\\PYZus{}vector}\\PY{p}{(}\\PY{n}{GrB\\PYZus{}Vector}\\PY{+w}{ }\\PY{n}{v}\\PY{p}{,}\\PY{+w}{ }\\PY{k+kt}{double}\\PY{+w}{ }\\PY{n}{scl}\\PY{p}{)}\\PY{p}{;}\n",
       "\n",
       "\\PY{k+kt}{int}\\PY{+w}{ }\\PY{n+nf}{pagerank}\\PY{p}{(}\\PY{n}{GrB\\PYZus{}Vector}\\PY{+w}{ }\\PY{o}{*}\\PY{n}{pr}\\PY{p}{,}\n",
       "\\PY{+w}{             }\\PY{n}{GrB\\PYZus{}Matrix}\\PY{+w}{ }\\PY{n}{A}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}Vector}\\PY{+w}{ }\\PY{n}{v}\\PY{p}{,}\n",
       "\\PY{+w}{             }\\PY{k}{const}\\PY{+w}{ }\\PY{k+kt}{double}\\PY{+w}{ }\\PY{n}{alpha}\\PY{p}{,}\\PY{+w}{ }\\PY{k}{const}\\PY{+w}{ }\\PY{k+kt}{double}\\PY{+w}{ }\\PY{n}{ctol}\\PY{p}{,}\\PY{+w}{ }\\PY{k}{const}\\PY{+w}{ }\\PY{k+kt}{int}\\PY{+w}{ }\\PY{n}{itmax}\\PY{p}{)}\n",
       "\\PY{p}{\\PYZob{}}\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}Vector}\\PY{+w}{ }\\PY{n}{v\\PYZus{}scaled}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{x}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{xprev}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{accum}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{D\\PYZus{}pseudoinv}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{k+kt}{int}\\PY{+w}{ }\\PY{n}{k}\\PY{+w}{ }\\PY{o}{=}\\PY{+w}{ }\\PY{l+m+mi}{\\PYZhy{}1}\\PY{p}{;}\n",
       "\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}Index}\\PY{+w}{ }\\PY{n}{N}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}Matrix\\PYZus{}nrows}\\PY{p}{(}\\PY{o}{\\PYZam{}}\\PY{n}{N}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{A}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{k}{if}\\PY{+w}{ }\\PY{p}{(}\\PY{n}{N}\\PY{+w}{ }\\PY{o}{=}\\PY{o}{=}\\PY{+w}{ }\\PY{l+m+mi}{0}\\PY{p}{)}\\PY{+w}{ }\\PY{k}{return}\\PY{+w}{ }\\PY{l+m+mi}{0}\\PY{p}{;}\n",
       "\n",
       "\\PY{+w}{  }\\PY{p}{\\PYZob{}}\n",
       "\\PY{+w}{    }\\PY{n}{GrB\\PYZus{}Index}\\PY{+w}{ }\\PY{n}{NE}\\PY{p}{;}\n",
       "\\PY{+w}{    }\\PY{n}{GrB\\PYZus{}Matrix\\PYZus{}nvals}\\PY{p}{(}\\PY{o}{\\PYZam{}}\\PY{n}{NE}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{A}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{    }\\PY{k}{if}\\PY{+w}{ }\\PY{p}{(}\\PY{n}{NE}\\PY{+w}{ }\\PY{o}{=}\\PY{o}{=}\\PY{+w}{ }\\PY{l+m+mi}{0}\\PY{p}{)}\\PY{+w}{ }\\PY{k}{return}\\PY{+w}{ }\\PY{l+m+mi}{0}\\PY{p}{;}\n",
       "\\PY{+w}{    }\\PY{n}{GrB\\PYZus{}Vector\\PYZus{}nvals}\\PY{p}{(}\\PY{o}{\\PYZam{}}\\PY{n}{NE}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{v}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{    }\\PY{k}{if}\\PY{+w}{ }\\PY{p}{(}\\PY{n}{NE}\\PY{+w}{ }\\PY{o}{=}\\PY{o}{=}\\PY{+w}{ }\\PY{l+m+mi}{0}\\PY{p}{)}\\PY{+w}{ }\\PY{k}{return}\\PY{+w}{ }\\PY{l+m+mi}{0}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{p}{\\PYZcb{}}\n",
       "\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}Vector\\PYZus{}new}\\PY{p}{(}\\PY{o}{\\PYZam{}}\\PY{n}{v\\PYZus{}scaled}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}FP64}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{N}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}Vector\\PYZus{}new}\\PY{p}{(}\\PY{o}{\\PYZam{}}\\PY{n}{x}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}FP64}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{N}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}Vector\\PYZus{}new}\\PY{p}{(}\\PY{o}{\\PYZam{}}\\PY{n}{xprev}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}FP64}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{N}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}Vector\\PYZus{}new}\\PY{p}{(}\\PY{o}{\\PYZam{}}\\PY{n}{accum}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}FP64}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{N}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}Vector\\PYZus{}new}\\PY{p}{(}\\PY{o}{\\PYZam{}}\\PY{n}{D\\PYZus{}pseudoinv}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}FP64}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{N}\\PY{p}{)}\\PY{p}{;}\n",
       "\n",
       "\\PY{+w}{  }\\PY{n}{degree\\PYZus{}pseudoinv}\\PY{p}{(}\\PY{o}{\\PYZam{}}\\PY{n}{D\\PYZus{}pseudoinv}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{A}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}Vector\\PYZus{}dup}\\PY{p}{(}\\PY{o}{\\PYZam{}}\\PY{n}{v\\PYZus{}scaled}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{v}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{n}{scale\\PYZus{}vector}\\PY{p}{(}\\PY{n}{v\\PYZus{}scaled}\\PY{p}{,}\\PY{+w}{ }\\PY{l+m+mf}{1.0}\\PY{+w}{ }\\PY{o}{\\PYZhy{}}\\PY{+w}{ }\\PY{n}{alpha}\\PY{p}{)}\\PY{p}{;}\n",
       "\n",
       "\\PY{+w}{  }\\PY{k}{for}\\PY{+w}{ }\\PY{p}{(}\\PY{n}{k}\\PY{+w}{ }\\PY{o}{=}\\PY{+w}{ }\\PY{l+m+mi}{0}\\PY{p}{;}\\PY{+w}{ }\\PY{n}{k}\\PY{+w}{ }\\PY{o}{\\PYZlt{}}\\PY{+w}{ }\\PY{n}{itmax}\\PY{p}{;}\\PY{+w}{ }\\PY{o}{+}\\PY{o}{+}\\PY{n}{k}\\PY{p}{)}\\PY{+w}{ }\\PY{p}{\\PYZob{}}\n",
       "\\PY{+w}{    }\\PY{n}{GrB\\PYZus{}assign}\\PY{p}{(}\\PY{n}{xprev}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}NULL}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}NULL}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{x}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}ALL}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{N}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}DESC\\PYZus{}R}\\PY{p}{)}\\PY{p}{;}\n",
       "\n",
       "\\PY{+w}{    }\\PY{n}{GrB\\PYZus{}vxm}\\PY{p}{(}\\PY{n}{accum}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}NULL}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}NULL}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}PLUS\\PYZus{}TIMES\\PYZus{}SEMIRING\\PYZus{}FP64}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{x}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{A}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}DESC\\PYZus{}R}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{    }\\PY{n}{GrB\\PYZus{}eWiseAdd}\\PY{p}{(}\\PY{n}{accum}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}NULL}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}NULL}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}TIMES\\PYZus{}FP64}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{D\\PYZus{}pseudoinv}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{accum}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}DESC\\PYZus{}R}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{    }\\PY{n}{scale\\PYZus{}vector}\\PY{p}{(}\\PY{n}{accum}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{alpha}\\PY{p}{)}\\PY{p}{;}\n",
       "\n",
       "\\PY{+w}{    }\\PY{n}{GrB\\PYZus{}eWiseAdd}\\PY{p}{(}\\PY{n}{accum}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}NULL}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}NULL}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}PLUS\\PYZus{}FP64}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{v\\PYZus{}scaled}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{accum}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}DESC\\PYZus{}R}\\PY{p}{)}\\PY{p}{;}\n",
       "\n",
       "\\PY{+w}{    }\\PY{n}{GrB\\PYZus{}assign}\\PY{p}{(}\\PY{n}{x}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}NULL}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}NULL}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{accum}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}ALL}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{N}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}DESC\\PYZus{}R}\\PY{p}{)}\\PY{p}{;}\n",
       "\n",
       "\\PY{+w}{    }\\PY{c+c1}{// Re\\PYZhy{}use xprev for \\PYZdq{}convergence\\PYZdq{} check}\n",
       "\\PY{+w}{    }\\PY{n}{GrB\\PYZus{}eWiseAdd}\\PY{p}{(}\\PY{n}{xprev}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}NULL}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}NULL}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}MINUS\\PYZus{}FP64}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{accum}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{xprev}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}DESC\\PYZus{}R}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{    }\\PY{n}{GrB\\PYZus{}apply}\\PY{p}{(}\\PY{n}{xprev}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}NULL}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}NULL}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}ABS\\PYZus{}FP64}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{xprev}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}DESC\\PYZus{}R}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{    }\\PY{k+kt}{double}\\PY{+w}{ }\\PY{n}{diff}\\PY{+w}{ }\\PY{o}{=}\\PY{+w}{ }\\PY{l+m+mf}{0.0}\\PY{p}{;}\n",
       "\\PY{+w}{    }\\PY{n}{GrB\\PYZus{}reduce}\\PY{p}{(}\\PY{o}{\\PYZam{}}\\PY{n}{diff}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}NULL}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}MAX\\PYZus{}MONOID\\PYZus{}FP64}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{xprev}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}NULL}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{    }\\PY{c+c1}{//fprintf(stderr, \\PYZdq{}norm1 diff \\PYZpc{}lg\\PYZbs{}n\\PYZdq{}, diff);}\n",
       "\\PY{+w}{    }\\PY{k}{if}\\PY{+w}{ }\\PY{p}{(}\\PY{n}{diff}\\PY{+w}{ }\\PY{o}{\\PYZlt{}}\\PY{o}{=}\\PY{+w}{ }\\PY{n}{ctol}\\PY{p}{)}\\PY{+w}{ }\\PY{k}{break}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{p}{\\PYZcb{}}\n",
       "\n",
       "\\PY{+w}{  }\\PY{k+kt}{double}\\PY{+w}{ }\\PY{n}{sum}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}reduce}\\PY{p}{(}\\PY{o}{\\PYZam{}}\\PY{n}{sum}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}NULL}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}PLUS\\PYZus{}MONOID\\PYZus{}FP64}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{x}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}NULL}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{k}{if}\\PY{+w}{ }\\PY{p}{(}\\PY{n}{sum}\\PY{+w}{ }\\PY{o}{!}\\PY{o}{=}\\PY{+w}{ }\\PY{l+m+mi}{0}\\PY{p}{)}\n",
       "\\PY{+w}{    }\\PY{n}{scale\\PYZus{}vector}\\PY{p}{(}\\PY{n}{x}\\PY{p}{,}\\PY{+w}{ }\\PY{l+m+mf}{1.0}\\PY{o}{/}\\PY{n}{sum}\\PY{p}{)}\\PY{p}{;}\n",
       "\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}Vector\\PYZus{}dup}\\PY{p}{(}\\PY{n}{pr}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{x}\\PY{p}{)}\\PY{p}{;}\n",
       "\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}free}\\PY{p}{(}\\PY{o}{\\PYZam{}}\\PY{n}{D\\PYZus{}pseudoinv}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}free}\\PY{p}{(}\\PY{o}{\\PYZam{}}\\PY{n}{accum}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}free}\\PY{p}{(}\\PY{o}{\\PYZam{}}\\PY{n}{xprev}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}free}\\PY{p}{(}\\PY{o}{\\PYZam{}}\\PY{n}{x}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}free}\\PY{p}{(}\\PY{o}{\\PYZam{}}\\PY{n}{v\\PYZus{}scaled}\\PY{p}{)}\\PY{p}{;}\n",
       "\n",
       "\\PY{+w}{  }\\PY{k}{return}\\PY{+w}{ }\\PY{n}{k}\\PY{p}{;}\n",
       "\\PY{p}{\\PYZcb{}}\n",
       "\\end{Verbatim}\n"
      ],
      "text/plain": [
       "#include <GraphBLAS.h>\n",
       "\n",
       "extern void degree_pseudoinv(GrB_Vector* v, GrB_Matrix A);\n",
       "extern void scale_vector(GrB_Vector v, double scl);\n",
       "\n",
       "int pagerank(GrB_Vector *pr,\n",
       "             GrB_Matrix A, GrB_Vector v,\n",
       "             const double alpha, const double ctol, const int itmax)\n",
       "{\n",
       "  GrB_Vector v_scaled, x, xprev, accum, D_pseudoinv;\n",
       "  int k = -1;\n",
       "\n",
       "  GrB_Index N;\n",
       "  GrB_Matrix_nrows(&N, A);\n",
       "  if (N == 0) return 0;\n",
       "\n",
       "  {\n",
       "    GrB_Index NE;\n",
       "    GrB_Matrix_nvals(&NE, A);\n",
       "    if (NE == 0) return 0;\n",
       "    GrB_Vector_nvals(&NE, v);\n",
       "    if (NE == 0) return 0;\n",
       "  }\n",
       "\n",
       "  GrB_Vector_new(&v_scaled, GrB_FP64, N);\n",
       "  GrB_Vector_new(&x, GrB_FP64, N);\n",
       "  GrB_Vector_new(&xprev, GrB_FP64, N);\n",
       "  GrB_Vector_new(&accum, GrB_FP64, N);\n",
       "  GrB_Vector_new(&D_pseudoinv, GrB_FP64, N);\n",
       "\n",
       "  degree_pseudoinv(&D_pseudoinv, A);\n",
       "  GrB_Vector_dup(&v_scaled, v);\n",
       "  scale_vector(v_scaled, 1.0 - alpha);\n",
       "\n",
       "  for (k = 0; k < itmax; ++k) {\n",
       "    GrB_assign(xprev, GrB_NULL, GrB_NULL, x, GrB_ALL, N, GrB_DESC_R);\n",
       "\n",
       "    GrB_vxm(accum, GrB_NULL, GrB_NULL, GrB_PLUS_TIMES_SEMIRING_FP64, x, A, GrB_DESC_R);\n",
       "    GrB_eWiseAdd(accum, GrB_NULL, GrB_NULL, GrB_TIMES_FP64, D_pseudoinv, accum, GrB_DESC_R);\n",
       "    scale_vector(accum, alpha);\n",
       "\n",
       "    GrB_eWiseAdd(accum, GrB_NULL, GrB_NULL, GrB_PLUS_FP64, v_scaled, accum, GrB_DESC_R);\n",
       "\n",
       "    GrB_assign(x, GrB_NULL, GrB_NULL, accum, GrB_ALL, N, GrB_DESC_R);\n",
       "\n",
       "    // Re-use xprev for \"convergence\" check\n",
       "    GrB_eWiseAdd(xprev, GrB_NULL, GrB_NULL, GrB_MINUS_FP64, accum, xprev, GrB_DESC_R);\n",
       "    GrB_apply(xprev, GrB_NULL, GrB_NULL, GrB_ABS_FP64, xprev, GrB_DESC_R);\n",
       "    double diff = 0.0;\n",
       "    GrB_reduce(&diff, GrB_NULL, GrB_MAX_MONOID_FP64, xprev, GrB_NULL);\n",
       "    //fprintf(stderr, \"norm1 diff %lg\\n\", diff);\n",
       "    if (diff <= ctol) break;\n",
       "  }\n",
       "\n",
       "  double sum;\n",
       "  GrB_reduce(&sum, GrB_NULL, GrB_PLUS_MONOID_FP64, x, GrB_NULL);\n",
       "  if (sum != 0)\n",
       "    scale_vector(x, 1.0/sum);\n",
       "\n",
       "  GrB_Vector_dup(pr, x);\n",
       "\n",
       "  GrB_free(&D_pseudoinv);\n",
       "  GrB_free(&accum);\n",
       "  GrB_free(&xprev);\n",
       "  GrB_free(&x);\n",
       "  GrB_free(&v_scaled);\n",
       "\n",
       "  return k;\n",
       "}"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Code('pagerank.c')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1b479619",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### PItfalls: eWiseAdd and eWiseMult\n",
    "\n",
    "Slight misnomers that lead to pitfalls:\n",
    "- `eWiseAdd` is an edge set **union**\n",
    "- `eWiseMult` is an edge set **intersection**\n",
    "  \n",
    "Each can have operators that do not correspond to *add* and *mult*!  See `eWiseAdd(..., GrB_TIMES_FP64, ...)` above.\n",
    "\n",
    "Pitfall for the un-initiated:  `eWiseAdd` with `GrB_MINUS_*`.\n",
    "- If the edge is in both operands, the second's value is subtracted from the first's.\n",
    "- If the edge is in **only one** operand, the value is stored directly.\n",
    "  - The second operand's value is **not** subtracted from an additive identity.\n",
    "  - It **cannot** be.  BinaryOps don't encode the additive identity, unlike monoids.\n",
    "- IIRC, v2.0 adds other operations that will invert..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "e95884e6-ffef-4e9e-9163-bfaa875989b6",
   "metadata": {
    "scrolled": true,
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>pre { line-height: 125%; }\n",
       "td.linenos .normal { color: inherit; background-color: transparent; padding-left: 5px; padding-right: 5px; }\n",
       "span.linenos { color: inherit; background-color: transparent; padding-left: 5px; padding-right: 5px; }\n",
       "td.linenos .special { color: #000000; background-color: #ffffc0; padding-left: 5px; padding-right: 5px; }\n",
       "span.linenos.special { color: #000000; background-color: #ffffc0; padding-left: 5px; padding-right: 5px; }\n",
       ".output_html .hll { background-color: #ffffcc }\n",
       ".output_html { background: #f8f8f8; }\n",
       ".output_html .c { color: #3D7B7B; font-style: italic } /* Comment */\n",
       ".output_html .err { border: 1px solid #FF0000 } /* Error */\n",
       ".output_html .k { color: #008000; font-weight: bold } /* Keyword */\n",
       ".output_html .o { color: #666666 } /* Operator */\n",
       ".output_html .ch { color: #3D7B7B; font-style: italic } /* Comment.Hashbang */\n",
       ".output_html .cm { color: #3D7B7B; font-style: italic } /* Comment.Multiline */\n",
       ".output_html .cp { color: #9C6500 } /* Comment.Preproc */\n",
       ".output_html .cpf { color: #3D7B7B; font-style: italic } /* Comment.PreprocFile */\n",
       ".output_html .c1 { color: #3D7B7B; font-style: italic } /* Comment.Single */\n",
       ".output_html .cs { color: #3D7B7B; font-style: italic } /* Comment.Special */\n",
       ".output_html .gd { color: #A00000 } /* Generic.Deleted */\n",
       ".output_html .ge { font-style: italic } /* Generic.Emph */\n",
       ".output_html .gr { color: #E40000 } /* Generic.Error */\n",
       ".output_html .gh { color: #000080; font-weight: bold } /* Generic.Heading */\n",
       ".output_html .gi { color: #008400 } /* Generic.Inserted */\n",
       ".output_html .go { color: #717171 } /* Generic.Output */\n",
       ".output_html .gp { color: #000080; font-weight: bold } /* Generic.Prompt */\n",
       ".output_html .gs { font-weight: bold } /* Generic.Strong */\n",
       ".output_html .gu { color: #800080; font-weight: bold } /* Generic.Subheading */\n",
       ".output_html .gt { color: #0044DD } /* Generic.Traceback */\n",
       ".output_html .kc { color: #008000; font-weight: bold } /* Keyword.Constant */\n",
       ".output_html .kd { color: #008000; font-weight: bold } /* Keyword.Declaration */\n",
       ".output_html .kn { color: #008000; font-weight: bold } /* Keyword.Namespace */\n",
       ".output_html .kp { color: #008000 } /* Keyword.Pseudo */\n",
       ".output_html .kr { color: #008000; font-weight: bold } /* Keyword.Reserved */\n",
       ".output_html .kt { color: #B00040 } /* Keyword.Type */\n",
       ".output_html .m { color: #666666 } /* Literal.Number */\n",
       ".output_html .s { color: #BA2121 } /* Literal.String */\n",
       ".output_html .na { color: #687822 } /* Name.Attribute */\n",
       ".output_html .nb { color: #008000 } /* Name.Builtin */\n",
       ".output_html .nc { color: #0000FF; font-weight: bold } /* Name.Class */\n",
       ".output_html .no { color: #880000 } /* Name.Constant */\n",
       ".output_html .nd { color: #AA22FF } /* Name.Decorator */\n",
       ".output_html .ni { color: #717171; font-weight: bold } /* Name.Entity */\n",
       ".output_html .ne { color: #CB3F38; font-weight: bold } /* Name.Exception */\n",
       ".output_html .nf { color: #0000FF } /* Name.Function */\n",
       ".output_html .nl { color: #767600 } /* Name.Label */\n",
       ".output_html .nn { color: #0000FF; font-weight: bold } /* Name.Namespace */\n",
       ".output_html .nt { color: #008000; font-weight: bold } /* Name.Tag */\n",
       ".output_html .nv { color: #19177C } /* Name.Variable */\n",
       ".output_html .ow { color: #AA22FF; font-weight: bold } /* Operator.Word */\n",
       ".output_html .w { color: #bbbbbb } /* Text.Whitespace */\n",
       ".output_html .mb { color: #666666 } /* Literal.Number.Bin */\n",
       ".output_html .mf { color: #666666 } /* Literal.Number.Float */\n",
       ".output_html .mh { color: #666666 } /* Literal.Number.Hex */\n",
       ".output_html .mi { color: #666666 } /* Literal.Number.Integer */\n",
       ".output_html .mo { color: #666666 } /* Literal.Number.Oct */\n",
       ".output_html .sa { color: #BA2121 } /* Literal.String.Affix */\n",
       ".output_html .sb { color: #BA2121 } /* Literal.String.Backtick */\n",
       ".output_html .sc { color: #BA2121 } /* Literal.String.Char */\n",
       ".output_html .dl { color: #BA2121 } /* Literal.String.Delimiter */\n",
       ".output_html .sd { color: #BA2121; font-style: italic } /* Literal.String.Doc */\n",
       ".output_html .s2 { color: #BA2121 } /* Literal.String.Double */\n",
       ".output_html .se { color: #AA5D1F; font-weight: bold } /* Literal.String.Escape */\n",
       ".output_html .sh { color: #BA2121 } /* Literal.String.Heredoc */\n",
       ".output_html .si { color: #A45A77; font-weight: bold } /* Literal.String.Interpol */\n",
       ".output_html .sx { color: #008000 } /* Literal.String.Other */\n",
       ".output_html .sr { color: #A45A77 } /* Literal.String.Regex */\n",
       ".output_html .s1 { color: #BA2121 } /* Literal.String.Single */\n",
       ".output_html .ss { color: #19177C } /* Literal.String.Symbol */\n",
       ".output_html .bp { color: #008000 } /* Name.Builtin.Pseudo */\n",
       ".output_html .fm { color: #0000FF } /* Name.Function.Magic */\n",
       ".output_html .vc { color: #19177C } /* Name.Variable.Class */\n",
       ".output_html .vg { color: #19177C } /* Name.Variable.Global */\n",
       ".output_html .vi { color: #19177C } /* Name.Variable.Instance */\n",
       ".output_html .vm { color: #19177C } /* Name.Variable.Magic */\n",
       ".output_html .il { color: #666666 } /* Literal.Number.Integer.Long */</style><div class=\"highlight\"><pre><span></span><span class=\"cp\">#include</span><span class=\"w\"> </span><span class=\"cpf\">&lt;GraphBLAS.h&gt;</span><span class=\"cp\"></span>\n",
       "\n",
       "<span class=\"cp\">#if defined(__cilk)</span>\n",
       "<span class=\"cp\">#include</span><span class=\"w\"> </span><span class=\"cpf\">&lt;cilk/cilk.h&gt;</span><span class=\"cp\"></span>\n",
       "<span class=\"cp\">#else</span>\n",
       "<span class=\"cp\">#define cilk_for for</span>\n",
       "<span class=\"cp\">#endif</span>\n",
       "\n",
       "<span class=\"kt\">void</span><span class=\"w\"> </span><span class=\"nf\">degree_pseudoinv</span><span class=\"p\">(</span><span class=\"n\">GrB_Vector</span><span class=\"w\"> </span><span class=\"o\">*</span><span class=\"n\">v</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_Matrix</span><span class=\"w\"> </span><span class=\"n\">A</span><span class=\"p\">)</span><span class=\"w\"></span>\n",
       "<span class=\"p\">{</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_Index</span><span class=\"w\"> </span><span class=\"n\">N</span><span class=\"p\">;</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_Matrix_nrows</span><span class=\"p\">(</span><span class=\"o\">&amp;</span><span class=\"n\">N</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">A</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_Vector_new</span><span class=\"p\">(</span><span class=\"n\">v</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_FP64</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">N</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"w\">   </span><span class=\"c1\">// Row reduce to compute the degrees.</span>\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_reduce</span><span class=\"p\">(</span><span class=\"o\">*</span><span class=\"n\">v</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_NULL</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_NULL</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_PLUS_MONOID_FP64</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">A</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_NULL</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"w\">  </span><span class=\"c1\">// Using a SuiteSparse extension, deprecated by GrB v2.0&#39;s GrB_select.</span>\n",
       "<span class=\"w\">  </span><span class=\"n\">GxB_select</span><span class=\"p\">(</span><span class=\"o\">*</span><span class=\"n\">v</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_NULL</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_NULL</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GxB_NONZERO</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"o\">*</span><span class=\"n\">v</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_NULL</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_DESC_R</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_apply</span><span class=\"p\">(</span><span class=\"o\">*</span><span class=\"n\">v</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_NULL</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_NULL</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_MINV_FP64</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"o\">*</span><span class=\"n\">v</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_DESC_R</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span>\n",
       "<span class=\"cp\">#if 0</span><span class=\"c\"></span>\n",
       "<span class=\"c\">  // An older way for use with v1.2.</span>\n",
       "<span class=\"c\">  GrB_Index nv;</span>\n",
       "<span class=\"c\">  GrB_Vector_nvals(&amp;nv, *v);</span>\n",
       "<span class=\"c\">  GrB_Index *idx = malloc(nv * sizeof(*idx));</span>\n",
       "<span class=\"c\">  double *val = malloc(nv * sizeof(*val));</span>\n",
       "<span class=\"c\">  </span>\n",
       "<span class=\"c\">  GrB_Vector_extractTuples(idx, val, &amp;nv, *v);</span>\n",
       "<span class=\"c\">  cilk_for (GrB_Index k = 0; k &lt; nv; ++k) {</span>\n",
       "<span class=\"c\">    if (val[k] == 0.0) val[k] = 1.0;</span>\n",
       "<span class=\"c\">    else val[k] = 1.0 / val[k];</span>\n",
       "<span class=\"c\">  }</span>\n",
       "<span class=\"c\">  GrB_Vector_clear(*v);</span>\n",
       "<span class=\"c\">  GrB_Vector_build(*v, idx, val, nv, GrB_FIRST_FP64);</span>\n",
       "\n",
       "<span class=\"c\">  free(val); free(idx);</span>\n",
       "<span class=\"cp\">#endif</span>\n",
       "<span class=\"w\">  </span>\n",
       "<span class=\"cp\">#if 0</span><span class=\"c\"></span>\n",
       "<span class=\"c\">  // How it could be done using v1.3&#39;s binary apply.</span>\n",
       "<span class=\"c\">  GrB_Vector dangling_mask;</span>\n",
       "<span class=\"c\">  GrB_Vector_new(&amp;dangling_mask, GrB_BOOL, N);</span>\n",
       "<span class=\"c\">  GrB_apply(dangling_mask, GrB_NULL, GrB_NULL, GrB_EQ_FP64, 0.0, *v, GrB_NULL);</span>\n",
       "<span class=\"c\">  GrB_apply(*v, dangling_mask, GrB_NULL, GrB_MINV_FP64, *v, GrB_DESC_RSC);</span>\n",
       "<span class=\"c\">  GrB_assign(*v, dangling_mask, GrB_NULL, 1.0, GrB_ALL, N, GrB_DESC_RS);</span>\n",
       "<span class=\"c\">  GrB_free(&amp;dangling_mask);</span>\n",
       "<span class=\"cp\">#endif</span>\n",
       "<span class=\"p\">}</span><span class=\"w\"></span>\n",
       "</pre></div>\n"
      ],
      "text/latex": [
       "\\begin{Verbatim}[commandchars=\\\\\\{\\}]\n",
       "\\PY{c+cp}{\\PYZsh{}}\\PY{c+cp}{include}\\PY{+w}{ }\\PY{c+cpf}{\\PYZlt{}GraphBLAS.h\\PYZgt{}}\n",
       "\n",
       "\\PY{c+cp}{\\PYZsh{}}\\PY{c+cp}{if defined(\\PYZus{}\\PYZus{}cilk)}\n",
       "\\PY{c+cp}{\\PYZsh{}}\\PY{c+cp}{include}\\PY{+w}{ }\\PY{c+cpf}{\\PYZlt{}cilk/cilk.h\\PYZgt{}}\n",
       "\\PY{c+cp}{\\PYZsh{}}\\PY{c+cp}{else}\n",
       "\\PY{c+cp}{\\PYZsh{}}\\PY{c+cp}{define cilk\\PYZus{}for for}\n",
       "\\PY{c+cp}{\\PYZsh{}}\\PY{c+cp}{endif}\n",
       "\n",
       "\\PY{k+kt}{void}\\PY{+w}{ }\\PY{n+nf}{degree\\PYZus{}pseudoinv}\\PY{p}{(}\\PY{n}{GrB\\PYZus{}Vector}\\PY{+w}{ }\\PY{o}{*}\\PY{n}{v}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}Matrix}\\PY{+w}{ }\\PY{n}{A}\\PY{p}{)}\n",
       "\\PY{p}{\\PYZob{}}\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}Index}\\PY{+w}{ }\\PY{n}{N}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}Matrix\\PYZus{}nrows}\\PY{p}{(}\\PY{o}{\\PYZam{}}\\PY{n}{N}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{A}\\PY{p}{)}\\PY{p}{;}\n",
       "\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}Vector\\PYZus{}new}\\PY{p}{(}\\PY{n}{v}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}FP64}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{N}\\PY{p}{)}\\PY{p}{;}\n",
       "\n",
       "\\PY{+w}{   }\\PY{c+c1}{// Row reduce to compute the degrees.}\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}reduce}\\PY{p}{(}\\PY{o}{*}\\PY{n}{v}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}NULL}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}NULL}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}PLUS\\PYZus{}MONOID\\PYZus{}FP64}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{A}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}NULL}\\PY{p}{)}\\PY{p}{;}\n",
       "\n",
       "\\PY{+w}{  }\\PY{c+c1}{// Using a SuiteSparse extension, deprecated by GrB v2.0\\PYZsq{}s GrB\\PYZus{}select.}\n",
       "\\PY{+w}{  }\\PY{n}{GxB\\PYZus{}select}\\PY{p}{(}\\PY{o}{*}\\PY{n}{v}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}NULL}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}NULL}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GxB\\PYZus{}NONZERO}\\PY{p}{,}\\PY{+w}{ }\\PY{o}{*}\\PY{n}{v}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}NULL}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}DESC\\PYZus{}R}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}apply}\\PY{p}{(}\\PY{o}{*}\\PY{n}{v}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}NULL}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}NULL}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}MINV\\PYZus{}FP64}\\PY{p}{,}\\PY{+w}{ }\\PY{o}{*}\\PY{n}{v}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}DESC\\PYZus{}R}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{  }\n",
       "\\PY{c+cp}{\\PYZsh{}if 0}\n",
       "\\PY{c}{  // An older way for use with v1.2.}\n",
       "\\PY{c}{  GrB\\PYZus{}Index nv;}\n",
       "\\PY{c}{  GrB\\PYZus{}Vector\\PYZus{}nvals(\\PYZam{}nv, *v);}\n",
       "\\PY{c}{  GrB\\PYZus{}Index *idx = malloc(nv * sizeof(*idx));}\n",
       "\\PY{c}{  double *val = malloc(nv * sizeof(*val));}\n",
       "\\PY{c}{  }\n",
       "\\PY{c}{  GrB\\PYZus{}Vector\\PYZus{}extractTuples(idx, val, \\PYZam{}nv, *v);}\n",
       "\\PY{c}{  cilk\\PYZus{}for (GrB\\PYZus{}Index k = 0; k \\PYZlt{} nv; ++k) \\PYZob{}}\n",
       "\\PY{c}{    if (val[k] == 0.0) val[k] = 1.0;}\n",
       "\\PY{c}{    else val[k] = 1.0 / val[k];}\n",
       "\\PY{c}{  \\PYZcb{}}\n",
       "\\PY{c}{  GrB\\PYZus{}Vector\\PYZus{}clear(*v);}\n",
       "\\PY{c}{  GrB\\PYZus{}Vector\\PYZus{}build(*v, idx, val, nv, GrB\\PYZus{}FIRST\\PYZus{}FP64);}\n",
       "\n",
       "\\PY{c}{  free(val); free(idx);}\n",
       "\\PY{c+cp}{\\PYZsh{}endif}\n",
       "\\PY{+w}{  }\n",
       "\\PY{c+cp}{\\PYZsh{}if 0}\n",
       "\\PY{c}{  // How it could be done using v1.3\\PYZsq{}s binary apply.}\n",
       "\\PY{c}{  GrB\\PYZus{}Vector dangling\\PYZus{}mask;}\n",
       "\\PY{c}{  GrB\\PYZus{}Vector\\PYZus{}new(\\PYZam{}dangling\\PYZus{}mask, GrB\\PYZus{}BOOL, N);}\n",
       "\\PY{c}{  GrB\\PYZus{}apply(dangling\\PYZus{}mask, GrB\\PYZus{}NULL, GrB\\PYZus{}NULL, GrB\\PYZus{}EQ\\PYZus{}FP64, 0.0, *v, GrB\\PYZus{}NULL);}\n",
       "\\PY{c}{  GrB\\PYZus{}apply(*v, dangling\\PYZus{}mask, GrB\\PYZus{}NULL, GrB\\PYZus{}MINV\\PYZus{}FP64, *v, GrB\\PYZus{}DESC\\PYZus{}RSC);}\n",
       "\\PY{c}{  GrB\\PYZus{}assign(*v, dangling\\PYZus{}mask, GrB\\PYZus{}NULL, 1.0, GrB\\PYZus{}ALL, N, GrB\\PYZus{}DESC\\PYZus{}RS);}\n",
       "\\PY{c}{  GrB\\PYZus{}free(\\PYZam{}dangling\\PYZus{}mask);}\n",
       "\\PY{c+cp}{\\PYZsh{}endif}\n",
       "\\PY{p}{\\PYZcb{}}\n",
       "\\end{Verbatim}\n"
      ],
      "text/plain": [
       "#include <GraphBLAS.h>\n",
       "\n",
       "#if defined(__cilk)\n",
       "#include <cilk/cilk.h>\n",
       "#else\n",
       "#define cilk_for for\n",
       "#endif\n",
       "\n",
       "void degree_pseudoinv(GrB_Vector *v, GrB_Matrix A)\n",
       "{\n",
       "  GrB_Index N;\n",
       "  GrB_Matrix_nrows(&N, A);\n",
       "\n",
       "  GrB_Vector_new(v, GrB_FP64, N);\n",
       "\n",
       "   // Row reduce to compute the degrees.\n",
       "  GrB_reduce(*v, GrB_NULL, GrB_NULL, GrB_PLUS_MONOID_FP64, A, GrB_NULL);\n",
       "\n",
       "  // Using a SuiteSparse extension, deprecated by GrB v2.0's GrB_select.\n",
       "  GxB_select(*v, GrB_NULL, GrB_NULL, GxB_NONZERO, *v, GrB_NULL, GrB_DESC_R);\n",
       "  GrB_apply(*v, GrB_NULL, GrB_NULL, GrB_MINV_FP64, *v, GrB_DESC_R);\n",
       "  \n",
       "#if 0\n",
       "  // An older way for use with v1.2.\n",
       "  GrB_Index nv;\n",
       "  GrB_Vector_nvals(&nv, *v);\n",
       "  GrB_Index *idx = malloc(nv * sizeof(*idx));\n",
       "  double *val = malloc(nv * sizeof(*val));\n",
       "  \n",
       "  GrB_Vector_extractTuples(idx, val, &nv, *v);\n",
       "  cilk_for (GrB_Index k = 0; k < nv; ++k) {\n",
       "    if (val[k] == 0.0) val[k] = 1.0;\n",
       "    else val[k] = 1.0 / val[k];\n",
       "  }\n",
       "  GrB_Vector_clear(*v);\n",
       "  GrB_Vector_build(*v, idx, val, nv, GrB_FIRST_FP64);\n",
       "\n",
       "  free(val); free(idx);\n",
       "#endif\n",
       "  \n",
       "#if 0\n",
       "  // How it could be done using v1.3's binary apply.\n",
       "  GrB_Vector dangling_mask;\n",
       "  GrB_Vector_new(&dangling_mask, GrB_BOOL, N);\n",
       "  GrB_apply(dangling_mask, GrB_NULL, GrB_NULL, GrB_EQ_FP64, 0.0, *v, GrB_NULL);\n",
       "  GrB_apply(*v, dangling_mask, GrB_NULL, GrB_MINV_FP64, *v, GrB_DESC_RSC);\n",
       "  GrB_assign(*v, dangling_mask, GrB_NULL, 1.0, GrB_ALL, N, GrB_DESC_RS);\n",
       "  GrB_free(&dangling_mask);\n",
       "#endif\n",
       "}"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Code('degree_pseudoinv.c')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6841a1ff-c6e5-491d-a567-18cb142134e5",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Tying It Together\n",
    "\n",
    "Typical boilerplate:\n",
    "- Initialization\n",
    "- Checking for errors\n",
    "- Printing!\n",
    "- And the unusual `GxB_Scalar` / `GrB_Scalar`\n",
    "\n",
    "Note: This once could run on the stationary (boring) core and execute the methods on the Lucata side.  Ongoing name changes...\n",
    "- Worked for running RedisGraph..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "6631413f-9838-49c0-9353-9aeb1a58b02a",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>pre { line-height: 125%; }\n",
       "td.linenos .normal { color: inherit; background-color: transparent; padding-left: 5px; padding-right: 5px; }\n",
       "span.linenos { color: inherit; background-color: transparent; padding-left: 5px; padding-right: 5px; }\n",
       "td.linenos .special { color: #000000; background-color: #ffffc0; padding-left: 5px; padding-right: 5px; }\n",
       "span.linenos.special { color: #000000; background-color: #ffffc0; padding-left: 5px; padding-right: 5px; }\n",
       ".output_html .hll { background-color: #ffffcc }\n",
       ".output_html { background: #f8f8f8; }\n",
       ".output_html .c { color: #3D7B7B; font-style: italic } /* Comment */\n",
       ".output_html .err { border: 1px solid #FF0000 } /* Error */\n",
       ".output_html .k { color: #008000; font-weight: bold } /* Keyword */\n",
       ".output_html .o { color: #666666 } /* Operator */\n",
       ".output_html .ch { color: #3D7B7B; font-style: italic } /* Comment.Hashbang */\n",
       ".output_html .cm { color: #3D7B7B; font-style: italic } /* Comment.Multiline */\n",
       ".output_html .cp { color: #9C6500 } /* Comment.Preproc */\n",
       ".output_html .cpf { color: #3D7B7B; font-style: italic } /* Comment.PreprocFile */\n",
       ".output_html .c1 { color: #3D7B7B; font-style: italic } /* Comment.Single */\n",
       ".output_html .cs { color: #3D7B7B; font-style: italic } /* Comment.Special */\n",
       ".output_html .gd { color: #A00000 } /* Generic.Deleted */\n",
       ".output_html .ge { font-style: italic } /* Generic.Emph */\n",
       ".output_html .gr { color: #E40000 } /* Generic.Error */\n",
       ".output_html .gh { color: #000080; font-weight: bold } /* Generic.Heading */\n",
       ".output_html .gi { color: #008400 } /* Generic.Inserted */\n",
       ".output_html .go { color: #717171 } /* Generic.Output */\n",
       ".output_html .gp { color: #000080; font-weight: bold } /* Generic.Prompt */\n",
       ".output_html .gs { font-weight: bold } /* Generic.Strong */\n",
       ".output_html .gu { color: #800080; font-weight: bold } /* Generic.Subheading */\n",
       ".output_html .gt { color: #0044DD } /* Generic.Traceback */\n",
       ".output_html .kc { color: #008000; font-weight: bold } /* Keyword.Constant */\n",
       ".output_html .kd { color: #008000; font-weight: bold } /* Keyword.Declaration */\n",
       ".output_html .kn { color: #008000; font-weight: bold } /* Keyword.Namespace */\n",
       ".output_html .kp { color: #008000 } /* Keyword.Pseudo */\n",
       ".output_html .kr { color: #008000; font-weight: bold } /* Keyword.Reserved */\n",
       ".output_html .kt { color: #B00040 } /* Keyword.Type */\n",
       ".output_html .m { color: #666666 } /* Literal.Number */\n",
       ".output_html .s { color: #BA2121 } /* Literal.String */\n",
       ".output_html .na { color: #687822 } /* Name.Attribute */\n",
       ".output_html .nb { color: #008000 } /* Name.Builtin */\n",
       ".output_html .nc { color: #0000FF; font-weight: bold } /* Name.Class */\n",
       ".output_html .no { color: #880000 } /* Name.Constant */\n",
       ".output_html .nd { color: #AA22FF } /* Name.Decorator */\n",
       ".output_html .ni { color: #717171; font-weight: bold } /* Name.Entity */\n",
       ".output_html .ne { color: #CB3F38; font-weight: bold } /* Name.Exception */\n",
       ".output_html .nf { color: #0000FF } /* Name.Function */\n",
       ".output_html .nl { color: #767600 } /* Name.Label */\n",
       ".output_html .nn { color: #0000FF; font-weight: bold } /* Name.Namespace */\n",
       ".output_html .nt { color: #008000; font-weight: bold } /* Name.Tag */\n",
       ".output_html .nv { color: #19177C } /* Name.Variable */\n",
       ".output_html .ow { color: #AA22FF; font-weight: bold } /* Operator.Word */\n",
       ".output_html .w { color: #bbbbbb } /* Text.Whitespace */\n",
       ".output_html .mb { color: #666666 } /* Literal.Number.Bin */\n",
       ".output_html .mf { color: #666666 } /* Literal.Number.Float */\n",
       ".output_html .mh { color: #666666 } /* Literal.Number.Hex */\n",
       ".output_html .mi { color: #666666 } /* Literal.Number.Integer */\n",
       ".output_html .mo { color: #666666 } /* Literal.Number.Oct */\n",
       ".output_html .sa { color: #BA2121 } /* Literal.String.Affix */\n",
       ".output_html .sb { color: #BA2121 } /* Literal.String.Backtick */\n",
       ".output_html .sc { color: #BA2121 } /* Literal.String.Char */\n",
       ".output_html .dl { color: #BA2121 } /* Literal.String.Delimiter */\n",
       ".output_html .sd { color: #BA2121; font-style: italic } /* Literal.String.Doc */\n",
       ".output_html .s2 { color: #BA2121 } /* Literal.String.Double */\n",
       ".output_html .se { color: #AA5D1F; font-weight: bold } /* Literal.String.Escape */\n",
       ".output_html .sh { color: #BA2121 } /* Literal.String.Heredoc */\n",
       ".output_html .si { color: #A45A77; font-weight: bold } /* Literal.String.Interpol */\n",
       ".output_html .sx { color: #008000 } /* Literal.String.Other */\n",
       ".output_html .sr { color: #A45A77 } /* Literal.String.Regex */\n",
       ".output_html .s1 { color: #BA2121 } /* Literal.String.Single */\n",
       ".output_html .ss { color: #19177C } /* Literal.String.Symbol */\n",
       ".output_html .bp { color: #008000 } /* Name.Builtin.Pseudo */\n",
       ".output_html .fm { color: #0000FF } /* Name.Function.Magic */\n",
       ".output_html .vc { color: #19177C } /* Name.Variable.Class */\n",
       ".output_html .vg { color: #19177C } /* Name.Variable.Global */\n",
       ".output_html .vi { color: #19177C } /* Name.Variable.Instance */\n",
       ".output_html .vm { color: #19177C } /* Name.Variable.Magic */\n",
       ".output_html .il { color: #666666 } /* Literal.Number.Integer.Long */</style><div class=\"highlight\"><pre><span></span><span class=\"cp\">#include</span><span class=\"w\"> </span><span class=\"cpf\">&lt;stdlib.h&gt;</span><span class=\"cp\"></span>\n",
       "<span class=\"cp\">#include</span><span class=\"w\"> </span><span class=\"cpf\">&lt;stdio.h&gt;</span><span class=\"cp\"></span>\n",
       "<span class=\"cp\">#include</span><span class=\"w\"> </span><span class=\"cpf\">&lt;stdint.h&gt;</span><span class=\"cp\"></span>\n",
       "\n",
       "<span class=\"cp\">#include</span><span class=\"w\"> </span><span class=\"cpf\">&lt;assert.h&gt;</span><span class=\"cp\"></span>\n",
       "\n",
       "<span class=\"cp\">#include</span><span class=\"w\"> </span><span class=\"cpf\">&lt;GraphBLAS.h&gt;</span><span class=\"cp\"></span>\n",
       "\n",
       "<span class=\"k\">extern</span><span class=\"w\"> </span><span class=\"kt\">int</span><span class=\"w\"> </span><span class=\"nf\">bfs</span><span class=\"p\">(</span><span class=\"n\">GrB_Vector</span><span class=\"w\"> </span><span class=\"o\">*</span><span class=\"n\">level</span><span class=\"p\">,</span><span class=\"w\"></span>\n",
       "<span class=\"w\">               </span><span class=\"n\">GrB_Matrix</span><span class=\"w\"> </span><span class=\"n\">A</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_Index</span><span class=\"w\"> </span><span class=\"n\">source</span><span class=\"p\">,</span><span class=\"w\"></span>\n",
       "<span class=\"w\">               </span><span class=\"k\">const</span><span class=\"w\"> </span><span class=\"n\">GrB_Index</span><span class=\"w\"> </span><span class=\"n\">max_level</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"k\">extern</span><span class=\"w\"> </span><span class=\"kt\">int</span><span class=\"w\"> </span><span class=\"nf\">pagerank</span><span class=\"p\">(</span><span class=\"n\">GrB_Vector</span><span class=\"w\"> </span><span class=\"o\">*</span><span class=\"n\">pr</span><span class=\"p\">,</span><span class=\"w\"></span>\n",
       "<span class=\"w\">                    </span><span class=\"n\">GrB_Matrix</span><span class=\"w\"> </span><span class=\"n\">A</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_Vector</span><span class=\"w\"> </span><span class=\"n\">v</span><span class=\"p\">,</span><span class=\"w\"></span>\n",
       "<span class=\"w\">                    </span><span class=\"k\">const</span><span class=\"w\"> </span><span class=\"kt\">double</span><span class=\"w\"> </span><span class=\"n\">alpha</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"k\">const</span><span class=\"w\"> </span><span class=\"kt\">double</span><span class=\"w\"> </span><span class=\"n\">ctol</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"k\">const</span><span class=\"w\"> </span><span class=\"kt\">int</span><span class=\"w\"> </span><span class=\"n\">itmax</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"k\">extern</span><span class=\"w\"> </span><span class=\"kt\">void</span><span class=\"w\"> </span><span class=\"nf\">read_dumped</span><span class=\"p\">(</span><span class=\"n\">GrB_Matrix</span><span class=\"w\"> </span><span class=\"o\">*</span><span class=\"n\">A</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"k\">const</span><span class=\"w\"> </span><span class=\"kt\">char</span><span class=\"w\"> </span><span class=\"o\">*</span><span class=\"n\">fname</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"k\">extern</span><span class=\"w\"> </span><span class=\"kt\">void</span><span class=\"w\"> </span><span class=\"nf\">dump_vtcs</span><span class=\"p\">(</span><span class=\"k\">const</span><span class=\"w\"> </span><span class=\"kt\">char</span><span class=\"w\"> </span><span class=\"o\">*</span><span class=\"n\">fname</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_Vector</span><span class=\"w\"> </span><span class=\"n\">v</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"k\">static</span><span class=\"w\"> </span><span class=\"kt\">void</span><span class=\"w\"> </span><span class=\"nf\">filter_pr</span><span class=\"p\">(</span><span class=\"n\">GrB_Vector</span><span class=\"w\"> </span><span class=\"o\">*</span><span class=\"n\">filtered_pr</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_Vector</span><span class=\"w\"> </span><span class=\"n\">pr</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"kt\">double</span><span class=\"w\"> </span><span class=\"n\">thresh</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"kt\">int</span><span class=\"w\"></span>\n",
       "<span class=\"nf\">main</span><span class=\"w\"> </span><span class=\"p\">(</span><span class=\"kt\">int</span><span class=\"w\"> </span><span class=\"n\">argc</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"kt\">char</span><span class=\"o\">**</span><span class=\"w\"> </span><span class=\"n\">argv</span><span class=\"p\">)</span><span class=\"w\"></span>\n",
       "<span class=\"p\">{</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"n\">srand48</span><span class=\"p\">(</span><span class=\"mi\">11</span><span class=\"w\"> </span><span class=\"o\">*</span><span class=\"w\"> </span><span class=\"mh\">0xDEADBEEF</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_Info</span><span class=\"w\"> </span><span class=\"n\">info</span><span class=\"p\">;</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_init</span><span class=\"p\">(</span><span class=\"n\">GrB_BLOCKING</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_Matrix</span><span class=\"w\"> </span><span class=\"n\">A</span><span class=\"p\">;</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"n\">read_dumped</span><span class=\"p\">(</span><span class=\"o\">&amp;</span><span class=\"n\">A</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">argv</span><span class=\"p\">[</span><span class=\"mi\">1</span><span class=\"p\">]);</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span>\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_Index</span><span class=\"w\"> </span><span class=\"n\">N</span><span class=\"p\">;</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"n\">info</span><span class=\"w\"> </span><span class=\"o\">=</span><span class=\"w\"> </span><span class=\"n\">GrB_Matrix_nrows</span><span class=\"p\">(</span><span class=\"o\">&amp;</span><span class=\"n\">N</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">A</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"n\">assert</span><span class=\"p\">(</span><span class=\"n\">info</span><span class=\"w\"> </span><span class=\"o\">==</span><span class=\"w\"> </span><span class=\"n\">GrB_SUCCESS</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_Index</span><span class=\"w\"> </span><span class=\"n\">seed</span><span class=\"w\"> </span><span class=\"o\">=</span><span class=\"w\"> </span><span class=\"n\">N</span><span class=\"mi\">-1</span><span class=\"p\">;</span><span class=\"w\"> </span><span class=\"c1\">// Kinda works for most pre-cooked graphs.</span>\n",
       "\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_Vector</span><span class=\"w\"> </span><span class=\"n\">region</span><span class=\"p\">;</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"w\">  </span><span class=\"n\">bfs</span><span class=\"p\">(</span><span class=\"o\">&amp;</span><span class=\"n\">region</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">A</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">seed</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"mi\">3</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_Index</span><span class=\"w\"> </span><span class=\"n\">region_size</span><span class=\"p\">;</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_Vector_nvals</span><span class=\"p\">(</span><span class=\"o\">&amp;</span><span class=\"n\">region_size</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">region</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_Index</span><span class=\"w\"> </span><span class=\"o\">*</span><span class=\"n\">region_vtx</span><span class=\"w\"> </span><span class=\"o\">=</span><span class=\"w\"> </span><span class=\"n\">malloc</span><span class=\"p\">(</span><span class=\"n\">region_size</span><span class=\"w\"> </span><span class=\"o\">*</span><span class=\"w\"> </span><span class=\"k\">sizeof</span><span class=\"p\">(</span><span class=\"n\">GrB_Index</span><span class=\"p\">));</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"kt\">int64_t</span><span class=\"w\"> </span><span class=\"o\">*</span><span class=\"n\">levels</span><span class=\"w\"> </span><span class=\"o\">=</span><span class=\"w\"> </span><span class=\"n\">malloc</span><span class=\"p\">(</span><span class=\"n\">region_size</span><span class=\"w\"> </span><span class=\"o\">*</span><span class=\"w\"> </span><span class=\"k\">sizeof</span><span class=\"p\">(</span><span class=\"kt\">int64_t</span><span class=\"p\">));</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_Vector_extractTuples</span><span class=\"p\">(</span><span class=\"n\">region_vtx</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">levels</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"o\">&amp;</span><span class=\"n\">region_size</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">region</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"w\">  </span><span class=\"n\">free</span><span class=\"p\">(</span><span class=\"n\">levels</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_free</span><span class=\"p\">(</span><span class=\"o\">&amp;</span><span class=\"n\">region</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_Vector</span><span class=\"w\"> </span><span class=\"n\">pr_seeds</span><span class=\"p\">;</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_Vector_new</span><span class=\"p\">(</span><span class=\"o\">&amp;</span><span class=\"n\">pr_seeds</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_FP64</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">N</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_Index</span><span class=\"w\"> </span><span class=\"n\">n_seeds</span><span class=\"w\"> </span><span class=\"o\">=</span><span class=\"w\"> </span><span class=\"p\">(</span><span class=\"n\">region_size</span><span class=\"w\"> </span><span class=\"o\">&gt;=</span><span class=\"w\"> </span><span class=\"mi\">3</span><span class=\"o\">?</span><span class=\"w\"> </span><span class=\"mi\">3</span><span class=\"w\"> </span><span class=\"o\">:</span><span class=\"w\"> </span><span class=\"n\">region_size</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"w\">  </span><span class=\"k\">for</span><span class=\"w\"> </span><span class=\"p\">(</span><span class=\"n\">GrB_Index</span><span class=\"w\"> </span><span class=\"n\">k</span><span class=\"w\"> </span><span class=\"o\">=</span><span class=\"w\"> </span><span class=\"mi\">0</span><span class=\"p\">;</span><span class=\"w\"> </span><span class=\"n\">k</span><span class=\"w\"> </span><span class=\"o\">&lt;</span><span class=\"w\"> </span><span class=\"n\">n_seeds</span><span class=\"p\">;</span><span class=\"w\"> </span><span class=\"o\">++</span><span class=\"n\">k</span><span class=\"p\">)</span><span class=\"w\"> </span><span class=\"p\">{</span><span class=\"w\"></span>\n",
       "<span class=\"w\">    </span><span class=\"k\">const</span><span class=\"w\"> </span><span class=\"n\">GrB_Index</span><span class=\"w\"> </span><span class=\"n\">i</span><span class=\"w\"> </span><span class=\"o\">=</span><span class=\"w\"> </span><span class=\"n\">lrand48</span><span class=\"p\">()</span><span class=\"w\"> </span><span class=\"o\">%</span><span class=\"w\"> </span><span class=\"n\">region_size</span><span class=\"p\">;</span><span class=\"w\"></span>\n",
       "<span class=\"w\">    </span><span class=\"n\">GrB_Index</span><span class=\"w\"> </span><span class=\"n\">vtx_of_interest</span><span class=\"w\"> </span><span class=\"o\">=</span><span class=\"w\"> </span><span class=\"n\">region_vtx</span><span class=\"p\">[</span><span class=\"n\">i</span><span class=\"p\">];</span><span class=\"w\"></span>\n",
       "<span class=\"w\">    </span><span class=\"n\">GrB_Vector_setElement</span><span class=\"p\">(</span><span class=\"n\">pr_seeds</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"mf\">1.0</span><span class=\"w\"> </span><span class=\"o\">/</span><span class=\"w\"> </span><span class=\"n\">n_seeds</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">vtx_of_interest</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"w\">    </span><span class=\"n\">printf</span><span class=\"p\">(</span><span class=\"s\">&quot;Seed %ld: %ld</span><span class=\"se\">\\n</span><span class=\"s\">&quot;</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"p\">(</span><span class=\"kt\">long</span><span class=\"p\">)</span><span class=\"n\">k</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"p\">(</span><span class=\"kt\">long</span><span class=\"p\">)</span><span class=\"n\">vtx_of_interest</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"p\">}</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"w\">  </span><span class=\"n\">free</span><span class=\"p\">(</span><span class=\"n\">region_vtx</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_Vector</span><span class=\"w\"> </span><span class=\"n\">pr</span><span class=\"p\">;</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"n\">pagerank</span><span class=\"p\">(</span><span class=\"o\">&amp;</span><span class=\"n\">pr</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">A</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">pr_seeds</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"mf\">0.85</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"mf\">1.0e-4</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"mi\">100</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"w\">  </span><span class=\"c1\">// Really, you&#39;d filter for a statistical difference,</span>\n",
       "<span class=\"w\">  </span><span class=\"c1\">// possibly against global PageRank.</span>\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_Vector</span><span class=\"w\"> </span><span class=\"n\">filtered_pr</span><span class=\"p\">;</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"n\">filter_pr</span><span class=\"p\">(</span><span class=\"o\">&amp;</span><span class=\"n\">filtered_pr</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">pr</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"mf\">1.0e-3</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"w\">  </span><span class=\"c1\">// SuiteSparse GraphBLAS extension, but LGB also supports:</span>\n",
       "<span class=\"w\">  </span><span class=\"n\">GxB_print</span><span class=\"p\">(</span><span class=\"n\">pr</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GxB_SUMMARY</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"n\">GxB_print</span><span class=\"p\">(</span><span class=\"n\">filtered_pr</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GxB_COMPLETE</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"w\">  </span><span class=\"n\">dump_vtcs</span><span class=\"p\">(</span><span class=\"s\">&quot;out-list&quot;</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">filtered_pr</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_free</span><span class=\"p\">(</span><span class=\"o\">&amp;</span><span class=\"n\">filtered_pr</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_free</span><span class=\"p\">(</span><span class=\"o\">&amp;</span><span class=\"n\">pr</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_free</span><span class=\"p\">(</span><span class=\"o\">&amp;</span><span class=\"n\">pr_seeds</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_free</span><span class=\"p\">(</span><span class=\"o\">&amp;</span><span class=\"n\">A</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"p\">}</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"kt\">void</span><span class=\"w\"></span>\n",
       "<span class=\"nf\">filter_pr</span><span class=\"p\">(</span><span class=\"n\">GrB_Vector</span><span class=\"w\"> </span><span class=\"o\">*</span><span class=\"n\">filtered_pr</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_Vector</span><span class=\"w\"> </span><span class=\"n\">pr</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"kt\">double</span><span class=\"w\"> </span><span class=\"n\">thresh</span><span class=\"p\">)</span><span class=\"w\"></span>\n",
       "<span class=\"p\">{</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"c1\">// Included in the v2.0 spec:</span>\n",
       "<span class=\"w\">  </span><span class=\"n\">GxB_Scalar</span><span class=\"w\"> </span><span class=\"n\">tmp</span><span class=\"p\">;</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"n\">GxB_Scalar_new</span><span class=\"p\">(</span><span class=\"o\">&amp;</span><span class=\"n\">tmp</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_FP64</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"n\">GxB_Scalar_setElement</span><span class=\"p\">(</span><span class=\"n\">tmp</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">thresh</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_Vector_dup</span><span class=\"p\">(</span><span class=\"n\">filtered_pr</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">pr</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_Vector_clear</span><span class=\"p\">(</span><span class=\"o\">*</span><span class=\"n\">filtered_pr</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span>\n",
       "<span class=\"w\">  </span><span class=\"n\">GxB_select</span><span class=\"p\">(</span><span class=\"o\">*</span><span class=\"n\">filtered_pr</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_NULL</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_NULL</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GxB_GE_THUNK</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">pr</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">tmp</span><span class=\"p\">,</span><span class=\"w\"> </span><span class=\"n\">GrB_NULL</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"n\">GrB_free</span><span class=\"p\">(</span><span class=\"o\">&amp;</span><span class=\"n\">tmp</span><span class=\"p\">);</span><span class=\"w\"></span>\n",
       "<span class=\"p\">}</span><span class=\"w\"></span>\n",
       "</pre></div>\n"
      ],
      "text/latex": [
       "\\begin{Verbatim}[commandchars=\\\\\\{\\}]\n",
       "\\PY{c+cp}{\\PYZsh{}}\\PY{c+cp}{include}\\PY{+w}{ }\\PY{c+cpf}{\\PYZlt{}stdlib.h\\PYZgt{}}\n",
       "\\PY{c+cp}{\\PYZsh{}}\\PY{c+cp}{include}\\PY{+w}{ }\\PY{c+cpf}{\\PYZlt{}stdio.h\\PYZgt{}}\n",
       "\\PY{c+cp}{\\PYZsh{}}\\PY{c+cp}{include}\\PY{+w}{ }\\PY{c+cpf}{\\PYZlt{}stdint.h\\PYZgt{}}\n",
       "\n",
       "\\PY{c+cp}{\\PYZsh{}}\\PY{c+cp}{include}\\PY{+w}{ }\\PY{c+cpf}{\\PYZlt{}assert.h\\PYZgt{}}\n",
       "\n",
       "\\PY{c+cp}{\\PYZsh{}}\\PY{c+cp}{include}\\PY{+w}{ }\\PY{c+cpf}{\\PYZlt{}GraphBLAS.h\\PYZgt{}}\n",
       "\n",
       "\\PY{k}{extern}\\PY{+w}{ }\\PY{k+kt}{int}\\PY{+w}{ }\\PY{n+nf}{bfs}\\PY{p}{(}\\PY{n}{GrB\\PYZus{}Vector}\\PY{+w}{ }\\PY{o}{*}\\PY{n}{level}\\PY{p}{,}\n",
       "\\PY{+w}{               }\\PY{n}{GrB\\PYZus{}Matrix}\\PY{+w}{ }\\PY{n}{A}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}Index}\\PY{+w}{ }\\PY{n}{source}\\PY{p}{,}\n",
       "\\PY{+w}{               }\\PY{k}{const}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}Index}\\PY{+w}{ }\\PY{n}{max\\PYZus{}level}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{k}{extern}\\PY{+w}{ }\\PY{k+kt}{int}\\PY{+w}{ }\\PY{n+nf}{pagerank}\\PY{p}{(}\\PY{n}{GrB\\PYZus{}Vector}\\PY{+w}{ }\\PY{o}{*}\\PY{n}{pr}\\PY{p}{,}\n",
       "\\PY{+w}{                    }\\PY{n}{GrB\\PYZus{}Matrix}\\PY{+w}{ }\\PY{n}{A}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}Vector}\\PY{+w}{ }\\PY{n}{v}\\PY{p}{,}\n",
       "\\PY{+w}{                    }\\PY{k}{const}\\PY{+w}{ }\\PY{k+kt}{double}\\PY{+w}{ }\\PY{n}{alpha}\\PY{p}{,}\\PY{+w}{ }\\PY{k}{const}\\PY{+w}{ }\\PY{k+kt}{double}\\PY{+w}{ }\\PY{n}{ctol}\\PY{p}{,}\\PY{+w}{ }\\PY{k}{const}\\PY{+w}{ }\\PY{k+kt}{int}\\PY{+w}{ }\\PY{n}{itmax}\\PY{p}{)}\\PY{p}{;}\n",
       "\n",
       "\\PY{k}{extern}\\PY{+w}{ }\\PY{k+kt}{void}\\PY{+w}{ }\\PY{n+nf}{read\\PYZus{}dumped}\\PY{p}{(}\\PY{n}{GrB\\PYZus{}Matrix}\\PY{+w}{ }\\PY{o}{*}\\PY{n}{A}\\PY{p}{,}\\PY{+w}{ }\\PY{k}{const}\\PY{+w}{ }\\PY{k+kt}{char}\\PY{+w}{ }\\PY{o}{*}\\PY{n}{fname}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{k}{extern}\\PY{+w}{ }\\PY{k+kt}{void}\\PY{+w}{ }\\PY{n+nf}{dump\\PYZus{}vtcs}\\PY{p}{(}\\PY{k}{const}\\PY{+w}{ }\\PY{k+kt}{char}\\PY{+w}{ }\\PY{o}{*}\\PY{n}{fname}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}Vector}\\PY{+w}{ }\\PY{n}{v}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{k}{static}\\PY{+w}{ }\\PY{k+kt}{void}\\PY{+w}{ }\\PY{n+nf}{filter\\PYZus{}pr}\\PY{p}{(}\\PY{n}{GrB\\PYZus{}Vector}\\PY{+w}{ }\\PY{o}{*}\\PY{n}{filtered\\PYZus{}pr}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}Vector}\\PY{+w}{ }\\PY{n}{pr}\\PY{p}{,}\\PY{+w}{ }\\PY{k+kt}{double}\\PY{+w}{ }\\PY{n}{thresh}\\PY{p}{)}\\PY{p}{;}\n",
       "\n",
       "\\PY{k+kt}{int}\n",
       "\\PY{n+nf}{main}\\PY{+w}{ }\\PY{p}{(}\\PY{k+kt}{int}\\PY{+w}{ }\\PY{n}{argc}\\PY{p}{,}\\PY{+w}{ }\\PY{k+kt}{char}\\PY{o}{*}\\PY{o}{*}\\PY{+w}{ }\\PY{n}{argv}\\PY{p}{)}\n",
       "\\PY{p}{\\PYZob{}}\n",
       "\\PY{+w}{  }\\PY{n}{srand48}\\PY{p}{(}\\PY{l+m+mi}{11}\\PY{+w}{ }\\PY{o}{*}\\PY{+w}{ }\\PY{l+m+mh}{0xDEADBEEF}\\PY{p}{)}\\PY{p}{;}\n",
       "\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}Info}\\PY{+w}{ }\\PY{n}{info}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}init}\\PY{p}{(}\\PY{n}{GrB\\PYZus{}BLOCKING}\\PY{p}{)}\\PY{p}{;}\n",
       "\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}Matrix}\\PY{+w}{ }\\PY{n}{A}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{n}{read\\PYZus{}dumped}\\PY{p}{(}\\PY{o}{\\PYZam{}}\\PY{n}{A}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{argv}\\PY{p}{[}\\PY{l+m+mi}{1}\\PY{p}{]}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{  }\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}Index}\\PY{+w}{ }\\PY{n}{N}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{n}{info}\\PY{+w}{ }\\PY{o}{=}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}Matrix\\PYZus{}nrows}\\PY{p}{(}\\PY{o}{\\PYZam{}}\\PY{n}{N}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{A}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{n}{assert}\\PY{p}{(}\\PY{n}{info}\\PY{+w}{ }\\PY{o}{=}\\PY{o}{=}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}SUCCESS}\\PY{p}{)}\\PY{p}{;}\n",
       "\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}Index}\\PY{+w}{ }\\PY{n}{seed}\\PY{+w}{ }\\PY{o}{=}\\PY{+w}{ }\\PY{n}{N}\\PY{l+m+mi}{\\PYZhy{}1}\\PY{p}{;}\\PY{+w}{ }\\PY{c+c1}{// Kinda works for most pre\\PYZhy{}cooked graphs.}\n",
       "\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}Vector}\\PY{+w}{ }\\PY{n}{region}\\PY{p}{;}\n",
       "\n",
       "\\PY{+w}{  }\\PY{n}{bfs}\\PY{p}{(}\\PY{o}{\\PYZam{}}\\PY{n}{region}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{A}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{seed}\\PY{p}{,}\\PY{+w}{ }\\PY{l+m+mi}{3}\\PY{p}{)}\\PY{p}{;}\n",
       "\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}Index}\\PY{+w}{ }\\PY{n}{region\\PYZus{}size}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}Vector\\PYZus{}nvals}\\PY{p}{(}\\PY{o}{\\PYZam{}}\\PY{n}{region\\PYZus{}size}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{region}\\PY{p}{)}\\PY{p}{;}\n",
       "\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}Index}\\PY{+w}{ }\\PY{o}{*}\\PY{n}{region\\PYZus{}vtx}\\PY{+w}{ }\\PY{o}{=}\\PY{+w}{ }\\PY{n}{malloc}\\PY{p}{(}\\PY{n}{region\\PYZus{}size}\\PY{+w}{ }\\PY{o}{*}\\PY{+w}{ }\\PY{k}{sizeof}\\PY{p}{(}\\PY{n}{GrB\\PYZus{}Index}\\PY{p}{)}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{k+kt}{int64\\PYZus{}t}\\PY{+w}{ }\\PY{o}{*}\\PY{n}{levels}\\PY{+w}{ }\\PY{o}{=}\\PY{+w}{ }\\PY{n}{malloc}\\PY{p}{(}\\PY{n}{region\\PYZus{}size}\\PY{+w}{ }\\PY{o}{*}\\PY{+w}{ }\\PY{k}{sizeof}\\PY{p}{(}\\PY{k+kt}{int64\\PYZus{}t}\\PY{p}{)}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}Vector\\PYZus{}extractTuples}\\PY{p}{(}\\PY{n}{region\\PYZus{}vtx}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{levels}\\PY{p}{,}\\PY{+w}{ }\\PY{o}{\\PYZam{}}\\PY{n}{region\\PYZus{}size}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{region}\\PY{p}{)}\\PY{p}{;}\n",
       "\n",
       "\\PY{+w}{  }\\PY{n}{free}\\PY{p}{(}\\PY{n}{levels}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}free}\\PY{p}{(}\\PY{o}{\\PYZam{}}\\PY{n}{region}\\PY{p}{)}\\PY{p}{;}\n",
       "\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}Vector}\\PY{+w}{ }\\PY{n}{pr\\PYZus{}seeds}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}Vector\\PYZus{}new}\\PY{p}{(}\\PY{o}{\\PYZam{}}\\PY{n}{pr\\PYZus{}seeds}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}FP64}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{N}\\PY{p}{)}\\PY{p}{;}\n",
       "\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}Index}\\PY{+w}{ }\\PY{n}{n\\PYZus{}seeds}\\PY{+w}{ }\\PY{o}{=}\\PY{+w}{ }\\PY{p}{(}\\PY{n}{region\\PYZus{}size}\\PY{+w}{ }\\PY{o}{\\PYZgt{}}\\PY{o}{=}\\PY{+w}{ }\\PY{l+m+mi}{3}\\PY{o}{?}\\PY{+w}{ }\\PY{l+m+mi}{3}\\PY{+w}{ }\\PY{o}{:}\\PY{+w}{ }\\PY{n}{region\\PYZus{}size}\\PY{p}{)}\\PY{p}{;}\n",
       "\n",
       "\\PY{+w}{  }\\PY{k}{for}\\PY{+w}{ }\\PY{p}{(}\\PY{n}{GrB\\PYZus{}Index}\\PY{+w}{ }\\PY{n}{k}\\PY{+w}{ }\\PY{o}{=}\\PY{+w}{ }\\PY{l+m+mi}{0}\\PY{p}{;}\\PY{+w}{ }\\PY{n}{k}\\PY{+w}{ }\\PY{o}{\\PYZlt{}}\\PY{+w}{ }\\PY{n}{n\\PYZus{}seeds}\\PY{p}{;}\\PY{+w}{ }\\PY{o}{+}\\PY{o}{+}\\PY{n}{k}\\PY{p}{)}\\PY{+w}{ }\\PY{p}{\\PYZob{}}\n",
       "\\PY{+w}{    }\\PY{k}{const}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}Index}\\PY{+w}{ }\\PY{n}{i}\\PY{+w}{ }\\PY{o}{=}\\PY{+w}{ }\\PY{n}{lrand48}\\PY{p}{(}\\PY{p}{)}\\PY{+w}{ }\\PY{o}{\\PYZpc{}}\\PY{+w}{ }\\PY{n}{region\\PYZus{}size}\\PY{p}{;}\n",
       "\\PY{+w}{    }\\PY{n}{GrB\\PYZus{}Index}\\PY{+w}{ }\\PY{n}{vtx\\PYZus{}of\\PYZus{}interest}\\PY{+w}{ }\\PY{o}{=}\\PY{+w}{ }\\PY{n}{region\\PYZus{}vtx}\\PY{p}{[}\\PY{n}{i}\\PY{p}{]}\\PY{p}{;}\n",
       "\\PY{+w}{    }\\PY{n}{GrB\\PYZus{}Vector\\PYZus{}setElement}\\PY{p}{(}\\PY{n}{pr\\PYZus{}seeds}\\PY{p}{,}\\PY{+w}{ }\\PY{l+m+mf}{1.0}\\PY{+w}{ }\\PY{o}{/}\\PY{+w}{ }\\PY{n}{n\\PYZus{}seeds}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{vtx\\PYZus{}of\\PYZus{}interest}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{    }\\PY{n}{printf}\\PY{p}{(}\\PY{l+s}{\\PYZdq{}}\\PY{l+s}{Seed \\PYZpc{}ld: \\PYZpc{}ld}\\PY{l+s+se}{\\PYZbs{}n}\\PY{l+s}{\\PYZdq{}}\\PY{p}{,}\\PY{+w}{ }\\PY{p}{(}\\PY{k+kt}{long}\\PY{p}{)}\\PY{n}{k}\\PY{p}{,}\\PY{+w}{ }\\PY{p}{(}\\PY{k+kt}{long}\\PY{p}{)}\\PY{n}{vtx\\PYZus{}of\\PYZus{}interest}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{p}{\\PYZcb{}}\n",
       "\n",
       "\\PY{+w}{  }\\PY{n}{free}\\PY{p}{(}\\PY{n}{region\\PYZus{}vtx}\\PY{p}{)}\\PY{p}{;}\n",
       "\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}Vector}\\PY{+w}{ }\\PY{n}{pr}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{n}{pagerank}\\PY{p}{(}\\PY{o}{\\PYZam{}}\\PY{n}{pr}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{A}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{pr\\PYZus{}seeds}\\PY{p}{,}\\PY{+w}{ }\\PY{l+m+mf}{0.85}\\PY{p}{,}\\PY{+w}{ }\\PY{l+m+mf}{1.0e\\PYZhy{}4}\\PY{p}{,}\\PY{+w}{ }\\PY{l+m+mi}{100}\\PY{p}{)}\\PY{p}{;}\n",
       "\n",
       "\\PY{+w}{  }\\PY{c+c1}{// Really, you\\PYZsq{}d filter for a statistical difference,}\n",
       "\\PY{+w}{  }\\PY{c+c1}{// possibly against global PageRank.}\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}Vector}\\PY{+w}{ }\\PY{n}{filtered\\PYZus{}pr}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{n}{filter\\PYZus{}pr}\\PY{p}{(}\\PY{o}{\\PYZam{}}\\PY{n}{filtered\\PYZus{}pr}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{pr}\\PY{p}{,}\\PY{+w}{ }\\PY{l+m+mf}{1.0e\\PYZhy{}3}\\PY{p}{)}\\PY{p}{;}\n",
       "\n",
       "\\PY{+w}{  }\\PY{c+c1}{// SuiteSparse GraphBLAS extension, but LGB also supports:}\n",
       "\\PY{+w}{  }\\PY{n}{GxB\\PYZus{}print}\\PY{p}{(}\\PY{n}{pr}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GxB\\PYZus{}SUMMARY}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{n}{GxB\\PYZus{}print}\\PY{p}{(}\\PY{n}{filtered\\PYZus{}pr}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GxB\\PYZus{}COMPLETE}\\PY{p}{)}\\PY{p}{;}\n",
       "\n",
       "\\PY{+w}{  }\\PY{n}{dump\\PYZus{}vtcs}\\PY{p}{(}\\PY{l+s}{\\PYZdq{}}\\PY{l+s}{out\\PYZhy{}list}\\PY{l+s}{\\PYZdq{}}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{filtered\\PYZus{}pr}\\PY{p}{)}\\PY{p}{;}\n",
       "\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}free}\\PY{p}{(}\\PY{o}{\\PYZam{}}\\PY{n}{filtered\\PYZus{}pr}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}free}\\PY{p}{(}\\PY{o}{\\PYZam{}}\\PY{n}{pr}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}free}\\PY{p}{(}\\PY{o}{\\PYZam{}}\\PY{n}{pr\\PYZus{}seeds}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}free}\\PY{p}{(}\\PY{o}{\\PYZam{}}\\PY{n}{A}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{p}{\\PYZcb{}}\n",
       "\n",
       "\\PY{k+kt}{void}\n",
       "\\PY{n+nf}{filter\\PYZus{}pr}\\PY{p}{(}\\PY{n}{GrB\\PYZus{}Vector}\\PY{+w}{ }\\PY{o}{*}\\PY{n}{filtered\\PYZus{}pr}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}Vector}\\PY{+w}{ }\\PY{n}{pr}\\PY{p}{,}\\PY{+w}{ }\\PY{k+kt}{double}\\PY{+w}{ }\\PY{n}{thresh}\\PY{p}{)}\n",
       "\\PY{p}{\\PYZob{}}\n",
       "\\PY{+w}{  }\\PY{c+c1}{// Included in the v2.0 spec:}\n",
       "\\PY{+w}{  }\\PY{n}{GxB\\PYZus{}Scalar}\\PY{+w}{ }\\PY{n}{tmp}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{n}{GxB\\PYZus{}Scalar\\PYZus{}new}\\PY{p}{(}\\PY{o}{\\PYZam{}}\\PY{n}{tmp}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}FP64}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{n}{GxB\\PYZus{}Scalar\\PYZus{}setElement}\\PY{p}{(}\\PY{n}{tmp}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{thresh}\\PY{p}{)}\\PY{p}{;}\n",
       "\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}Vector\\PYZus{}dup}\\PY{p}{(}\\PY{n}{filtered\\PYZus{}pr}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{pr}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}Vector\\PYZus{}clear}\\PY{p}{(}\\PY{o}{*}\\PY{n}{filtered\\PYZus{}pr}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{  }\n",
       "\\PY{+w}{  }\\PY{n}{GxB\\PYZus{}select}\\PY{p}{(}\\PY{o}{*}\\PY{n}{filtered\\PYZus{}pr}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}NULL}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}NULL}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GxB\\PYZus{}GE\\PYZus{}THUNK}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{pr}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{tmp}\\PY{p}{,}\\PY{+w}{ }\\PY{n}{GrB\\PYZus{}NULL}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{+w}{  }\\PY{n}{GrB\\PYZus{}free}\\PY{p}{(}\\PY{o}{\\PYZam{}}\\PY{n}{tmp}\\PY{p}{)}\\PY{p}{;}\n",
       "\\PY{p}{\\PYZcb{}}\n",
       "\\end{Verbatim}\n"
      ],
      "text/plain": [
       "#include <stdlib.h>\n",
       "#include <stdio.h>\n",
       "#include <stdint.h>\n",
       "\n",
       "#include <assert.h>\n",
       "\n",
       "#include <GraphBLAS.h>\n",
       "\n",
       "extern int bfs(GrB_Vector *level,\n",
       "               GrB_Matrix A, GrB_Index source,\n",
       "               const GrB_Index max_level);\n",
       "extern int pagerank(GrB_Vector *pr,\n",
       "                    GrB_Matrix A, GrB_Vector v,\n",
       "                    const double alpha, const double ctol, const int itmax);\n",
       "\n",
       "extern void read_dumped(GrB_Matrix *A, const char *fname);\n",
       "extern void dump_vtcs(const char *fname, GrB_Vector v);\n",
       "static void filter_pr(GrB_Vector *filtered_pr, GrB_Vector pr, double thresh);\n",
       "\n",
       "int\n",
       "main (int argc, char** argv)\n",
       "{\n",
       "  srand48(11 * 0xDEADBEEF);\n",
       "\n",
       "  GrB_Info info;\n",
       "  GrB_init(GrB_BLOCKING);\n",
       "\n",
       "  GrB_Matrix A;\n",
       "  read_dumped(&A, argv[1]);\n",
       "  \n",
       "  GrB_Index N;\n",
       "  info = GrB_Matrix_nrows(&N, A);\n",
       "  assert(info == GrB_SUCCESS);\n",
       "\n",
       "  GrB_Index seed = N-1; // Kinda works for most pre-cooked graphs.\n",
       "\n",
       "  GrB_Vector region;\n",
       "\n",
       "  bfs(&region, A, seed, 3);\n",
       "\n",
       "  GrB_Index region_size;\n",
       "  GrB_Vector_nvals(&region_size, region);\n",
       "\n",
       "  GrB_Index *region_vtx = malloc(region_size * sizeof(GrB_Index));\n",
       "  int64_t *levels = malloc(region_size * sizeof(int64_t));\n",
       "  GrB_Vector_extractTuples(region_vtx, levels, &region_size, region);\n",
       "\n",
       "  free(levels);\n",
       "  GrB_free(&region);\n",
       "\n",
       "  GrB_Vector pr_seeds;\n",
       "  GrB_Vector_new(&pr_seeds, GrB_FP64, N);\n",
       "\n",
       "  GrB_Index n_seeds = (region_size >= 3? 3 : region_size);\n",
       "\n",
       "  for (GrB_Index k = 0; k < n_seeds; ++k) {\n",
       "    const GrB_Index i = lrand48() % region_size;\n",
       "    GrB_Index vtx_of_interest = region_vtx[i];\n",
       "    GrB_Vector_setElement(pr_seeds, 1.0 / n_seeds, vtx_of_interest);\n",
       "    printf(\"Seed %ld: %ld\\n\", (long)k, (long)vtx_of_interest);\n",
       "  }\n",
       "\n",
       "  free(region_vtx);\n",
       "\n",
       "  GrB_Vector pr;\n",
       "  pagerank(&pr, A, pr_seeds, 0.85, 1.0e-4, 100);\n",
       "\n",
       "  // Really, you'd filter for a statistical difference,\n",
       "  // possibly against global PageRank.\n",
       "  GrB_Vector filtered_pr;\n",
       "  filter_pr(&filtered_pr, pr, 1.0e-3);\n",
       "\n",
       "  // SuiteSparse GraphBLAS extension, but LGB also supports:\n",
       "  GxB_print(pr, GxB_SUMMARY);\n",
       "  GxB_print(filtered_pr, GxB_COMPLETE);\n",
       "\n",
       "  dump_vtcs(\"out-list\", filtered_pr);\n",
       "\n",
       "  GrB_free(&filtered_pr);\n",
       "  GrB_free(&pr);\n",
       "  GrB_free(&pr_seeds);\n",
       "  GrB_free(&A);\n",
       "}\n",
       "\n",
       "void\n",
       "filter_pr(GrB_Vector *filtered_pr, GrB_Vector pr, double thresh)\n",
       "{\n",
       "  // Included in the v2.0 spec:\n",
       "  GxB_Scalar tmp;\n",
       "  GxB_Scalar_new(&tmp, GrB_FP64);\n",
       "  GxB_Scalar_setElement(tmp, thresh);\n",
       "\n",
       "  GrB_Vector_dup(filtered_pr, pr);\n",
       "  GrB_Vector_clear(*filtered_pr);\n",
       "  \n",
       "  GxB_select(*filtered_pr, GrB_NULL, GrB_NULL, GxB_GE_THUNK, pr, tmp, GrB_NULL);\n",
       "  GrB_free(&tmp);\n",
       "}"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Code('main.c')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "36205920",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "G = networkx.from_scipy_sparse_array(sp.io.mmread('1138_bus.mtx'))\n",
    "networkx.draw(G)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "e927bef5-9d6e-4b6c-bd4a-daa22e8d398a",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "gcc -g -Wall -fPIE -I.  -c -o bfs.o bfs.c\n",
      "gcc   main.o bfs.o pagerank.o degree_pseudoinv.o scale_vector.o read_dumped.o  libgraphblas.so -lm -o main\n",
      "Seed 0: 804\n",
      "Seed 1: 1137\n",
      "\n",
      "  1138x1 GraphBLAS double vector, full by col\n",
      "  pr, 1138 entries, memory: 16.2 KB\n",
      "\n",
      "\n",
      "  1138x1 GraphBLAS double vector, sparse by col\n",
      "  filtered_pr, 20 entries, memory: 784 bytes\n",
      "\n",
      "    (493,0)    0.00366982\n",
      "    (496,0)    0.00140997\n",
      "    (497,0)    0.00517853\n",
      "    (500,0)    0.00493444\n",
      "    (529,0)    0.23086\n",
      "    (568,0)    0.118781\n",
      "    (570,0)    0.0433558\n",
      "    (574,0)    0.150774\n",
      "    (577,0)    0.00229479\n",
      "    (580,0)    0.0265988\n",
      "    (582,0)    0.00429026\n",
      "    (774,0)    0.0433593\n",
      "    (775,0)    0.00764716\n",
      "    (777,0)    0.0130288\n",
      "    (778,0)    0.00229479\n",
      "    (779,0)    0.00580447\n",
      "    (784,0)    0.0209466\n",
      "    (796,0)    0.0283395\n",
      "    (804,0)    0.171704\n",
      "    (1137,0)    0.0987297\n",
      "\n"
     ]
    }
   ],
   "source": [
    "%%bash   \n",
    "make main\n",
    "./main 1138_bus.bin"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "49da43fa",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAApQAAAHzCAYAAACe1o1DAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjYuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy89olMNAAAACXBIWXMAAA9hAAAPYQGoP6dpAAEAAElEQVR4nOzdd1zN7f8H8NcZbSFFRgsNqdDQsFJKVvYoOyKZKW6bcNsze5WRPQp39kh2OInKqFBKZhFpn3P9/vDrfO/u1jl1Rqeu5+Nx/3H3+Xyu6x369D7XeF8MQggBRVEURVEURVUSU9oBUBRFURRFUbKNJpQURVEURVFUldCEkqIoiqIoiqoSmlBSFEVRFEVRVUITSoqiKIqiKKpKaEJJURRFURRFVQlNKCmKoiiKoqgqoQklRVEURVEUVSU0oaQoiqIoiqKqhCaUFEVRFEVRVJXQhJKiKIqiKIqqEppQUhRFURRFUVVCE0qKoiiKoiiqSmhCSVEURVEURVUJTSgpiqIoiqKoKqEJJUVRFEVRFFUlNKGkKIqiKIqiqoQmlBRFURRFUVSV0ISSoiiKoiiKqhKaUFIURVEURVFVQhNKiqIoiqIoqkpoQklRFEVRFEVVCU0oKYqiKIqiqCqhCSVFURRFURRVJTShpCiKoiiKoqqEJpQURVEURVFUldCEkqIoiqIoiqoSmlBSFEVRFEVRVUITSoqiKIqiKKpK2NIOgKIoStZxuVwkJiYiJiaG/9+7d++Qm5uLV69eAQCMjY2hpKSEli1bok2bNjAzM4OZmRn09PTAZNLP9hRFyTYGIYRIOwiKoihZQwjBnTt3EBQUhDNnziArKwsA0KhRI5iamsLAwADKysrgcDioU6cOjIyMkJWVhdevXyMmJgbfv38HAKipqcHNzQ0eHh6wsrICg8GQ5rdFURRVKTShpCiKEsKvX7+wdetW7N+/H4mJiWjRogXGjBmDjh07wszMDI0aNaqwDUIIPn78iJiYGERERODQoUP48OEDTE1N4enpCS8vLygqKkrgu6EoihINmlBSFEUJKCwsDN7e3vj27RuGDRsGDw8PdO7cucpT1lwuF9euXUNQUBBCQ0PRsmVL7N27F507dxZR5BRFUeJFF+5QFEVV4MuXL3Bzc4OrqytMTEzw8uVLHDhwAPb29iJZ/8hisdCjRw+cPHkS0dHRaNCgAbp06QIvLy9kZmaK4DugKIoSLzpCSVEUVY7k5GQ4ODjg169f2Lx5M4YPHy72dY48Hg87d+7EvHnz0KJFC1y/fh0aGhpi7ZOiKKoqaEJJURRVhnfv3sHBwQEsFgs3b96Erq6uRPuPjY1Ft27doKmpievXrwu0PpOiKEoaaEJJURRVivfv36NTp05QUFBAeHg4tLS0pBLHy5cv4ejoiAYNGuD27dtQV1eXShwURVHloWsoKYqi/oPH42Hs2LFgMBiIiIiQWjIJ/KlfeevWLXz8+BHTpk2TWhwURVHloQklRVHUf+zatQvh4eEICgpC06ZNpR0OjIyMsHXrVhw7dgyhoaHSDoeiKKoEOuVNUdUcIQRpaWmIjY1FRkYGcnJycP36dTg5OUFJSQkNGjSAqakpmjZtSotii0BSUhJMTU0xcuRI7Nq1S9rh8BFC0L9/fzx8+BAvXrygU98URVUrNKGkqGomPz8fYWFhuHnzJv8Yv6JTVcqjpqbGP86vW7du6N27N+Tl5SUQcc0ybdo0nDx5EomJiVBVVZV2OMV8/PgRLVu2xMKFCzF//nxph0NRFMVHE0qKqiZiYmIQFBSEw4cP49u3bzAyMkLbtm35SaKpqSk0NTWhqKgINpuNwsJC5Obm4vPnz4iNjeUnn8+ePcPr16+hoaGBUaNGwcPDA2ZmZtL+9mTC79+/0bRpU0ydOhUrVqyQdjilGjduHG7evIk3b96AxWJJOxyKoigANKGkKKlLSEjA5MmTcf36dTRs2BCjR4+Gh4cHTExMKt1mbGws9u/fj+DgYHz9+hVOTk7YuXMn9PX1RRh5zRMYGIgJEybg3bt3Ei8RJKjHjx/D2toaFy5cQK9evaQdDkVRFACaUFKU1BQUFGD9+vVYtmwZmjRpgrVr16Jfv36Qk5MTWR/5+fk4d+4c/vrrL3z69An+/v7w9fUVaR81iYODAxQVFXHp0iVph1ImQggsLCxgbGyMo0ePSjsciqIoAHSXN0VJxYcPH2BtbY2FCxdi6tSpiImJweDBg0We6MnLy2PIkCGIjY3FlClTMH/+fNjY2CAtLU2k/dQEPB4PHA4HXbt2lXYo5WIwGLC3t8eTJ0+kHQpFURQfTSgpSsJSUlJgb2+PjIwMPHr0COvWrYOKiopY+1RRUcH69evx6NEjfPv2DV27dkVqaqpY+5Q1b968wa9fv2BhYSHtUCpkYWGBhIQE/Pz5U9qhUBRFAaAJJUVJVHJyMuzt7cHlchEREQFLS0uJ9m9paYmIiAjk5+fD3t4e79+/l2j/1VlUVBQAwNzcXMqRVKwo6Y2OjpZuIBRFUf+PJpQUJSE5OTlwcXHhn76ip6cnlTiaN2+OiIgIEELg4uKCnJwcqcRR3Xz9+hUKCgrQ0NCQdigVKjq558uXL1KOhKIo6g+2tAOgKEkoLCzEu3fv8PPnT+Tm5iI1NRW6urpQUFBAkyZNoKmpKfai4AsXLkRSUhKio6Oho6Mj1r4qoquriwsXLsDc3BxLlizB2rVrpRpPdSErheFlJU6KomoPmlBSNQ4hBA8ePMC9e/cQExOD2NhYvHjxAnl5eWU+o66uzq/3aGZmhh49ekBbW1tkMd27dw+bNm3CmjVr0KpVK5G1WxXGxsZYunQp5s+fj4EDB8LW1lbaIVEURVEyipYNomqMtLQ0HDp0CEFBQUhISICysjJMTU2LFQZXU1ODgoIC8vPzwWazkZeXh5SUFH5R8JiYGCQkJIAQgu7du2PcuHHo168fFBQUKh0XIQRt27aFiooK7t69W62KURcWFqJjx47Izc1FdHR0rR752rlzJ6ZPn47c3Nxq9XdUmi9fvkBTUxMhISEYMGCAtMOhKIqiI5SU7EtJSYGvry9CQkIgLy+PQYMGYdeuXejatSuYzIqXCVtZWRX7pfzz50+cPHkSQUFBGDZsGBo0aIBZs2Zh1qxZlSrrc+fOHcTExODmzZvVLlFhs9lYsWIFnJ2dce/ePXTq1EnaIUmNkZERCgsLkZCQUG1GkcsSFxcHANU+Toqiag+6KYeSWVwuF1u3bkXr1q1x//59bNu2DR8/fsThw4fh6OgoUDJZmrp168LT0xP379/HixcvMGLECCxatAiWlpZ49OiR0O3t3LkThoaG1ba+oaOjI/T19bFz505phyJVRbu7i3Z7V2dRUVFQVlaGoaGhtEOhKIoCQBNKSkalpqaiU6dOmD59OkaNGoUXL17A29sb9evXF2k/xsbG2LJlCx4/fgw5OTnY2tpi9uzZ4PF4Aj3/9etXnDlzBpMmTaq208lMJhNeXl44ffo0vn37Ju1wpEZNTQ16enoyk1C2bdu22o14UxRVe9GEkpI5ycnJ6NKlC9LS0nDnzh3s2LED9erVE2uf5ubmiIyMxOrVq7Fx40aMGzcOXC63wufu3r2LgoICDBkyRKzxVdWwYcOQn5+Pe/fuSTsUqXJwcEBISIjAHxikITs7GxcvXqy2I94URdVONKGkZMq7d+9gb28PBoOB27dvS3TNH5vNxl9//YXg4GAEBwdjzJgxKCwsLPeZp0+fQlNTE82aNZNQlJWjpaUFDQ0NPH36VNqhSNXEiRPx7t07XL16VSTt/c4rRFxaJp6+/464tEz8ziv/34sgTpw4gczMTHh6eoogQoqiKNGgu7wpmfHjxw+0a9cOcnJyCA8P5xd3loZTp07B3d0dEydOxI4dO8q8r0+fPuDxeLh48aIEo6scFxcXKCgo4Pz589IORWoIITA3N4eenh7Onj1bqTYSPv/Ckcj3CH/9Be8zsvHvFywDgE4DZTgYNcIIGx0YaKoK3b6NjQ0aNGiAS5cuVSo+iqIocaAjlJTM8PX1xffv33H9+nWpJpMAMGTIEGzduhU7d+4sdzTr5cuXMDU1lWBklWdmZoYXL15IOwypYjAYmDx5Ms6fP48HDx4I9WxKRjZGBUbCefNtBEcmI/k/ySQAEADJGdkIjkyG8+bbGBUYiZSMbIH7OH/+PB49eoTJkycLFRtFUZS40YSSkgmXLl3C/v37sXHjRujq6ko7HADApEmT0K1bN3h6eiIzM7PUewoLC6GoqCjhyCpHUVGxwin82mDcuHFo3749PDw8BD6W8vjj93DaFIH7b9MBAFxe+RM/Rdfvv02H06YIHH9c8ZnqGRkZ8PLyQu/evdGnTx+B4qIoipIUmlBS1V5eXh4mTpwIFxcXjBs3Ttrh8DEYDAQGBuL79+9YvHixtMOhRITNZmP//v149+4d/P39K7x/W3gC5obEIK+QV2Ei+V9cHkFeIQ9zQ2KwLTyh3Ht9fHyQm5uLPXv2VNuKARRF1V40oaSqvTNnziA1NRWbN2+udr9IdXV14efnh6CgIPz69Uva4VQJXU79P61bt8by5cuxbt06BAUFlXnf8cfvsf5qvEj6XH81HifKGKlctWoVgoODERAQgKZNm4qkP4qiKFGiCSVV7e3cuRMODg7V9lSQCRMmICcnB4cPHy5xTV1dHR8/fpRCVML7+PEj1NXVpR1GtTFr1ixMmjQJ48ePx549e0pcT8nIxpLzcSLtc/H5uBJrKpctW4b58+dj6dKlGD16tEj7oyiKEhWaUFLVWmxsLO7evQtvb29ph1KmZs2awdXVFTt37iwxymdubi4ThbKBP8WyLSwspB1GtcFkMrF9+3ZMnz4dXl5eWLt2bbH6lPNDY1Ao5BR3RQp5BPNDYwAABQUF+Ouvv7BkyRKsWLGCLqugKKpaowklVa1dunQJKioq6Nevn7RDKdfIkSMRExODDx8+FPu6hYUFYmJikJ+fL6XIBJOXl4e4uDiaUP4Hg8HA5s2bMXfuXMyZMwedO3fGixcvkPD5F+4kfhN6zWRFuDyCO4nfcOb6PVhaWmLjxo3YsGED5s+fL9J+KIqiRI0mlFS1FhUVhXbt2kFeXl7aoZSrffv2AFCiMLilpSUKCgoQGRkpjbAE9vDhQxQWFsLS0lLaoVQ7DAYDq1atwu3bt5Geno527dph2paTYIlpOS+D8DBxTTDk5eXx5MkT+Pr6iqcjiqIoEaIJJVWtPX36FObm5tIOo0La2tpQV1cvkVBaW1ujRYsW2Lt3r0j6EcfJKwCwd+9etGzZElZWViJprybq3LkzoqOjMXfuXMSmE3DFtIeJMJjQsnbBw4cP0a5dO/F0QlEUJWJsaQdAUWXJzc1FfHw8Zs2aJe1QKsRgMNCuXTs8f/682NeZTCa8vLywePFibNq0qVKbXsR98srXr19x6tQprFixAkwm/YxZHkVFRfy1YDGC/a+UKFouSj958sjjAmz6hqYoSkbQ1xXFx+PxkJycjJiYGMTGxiIlJQW5ubk4cOAA2rVrB0tLS9SpUwdGRkYwMzODqakp6tevL7Z4CgoKQAiBqqrwSZI0qKqqIjc3t8TXPTw8sHDhQuzZswfz5s0TuL2UjGzMD43BncRvYDEZpa7X+/fJKwceJKGzvgZWDjCDdgNlgfvZvXs3AGDs2LECP1ObJaf/FmsyCfz5e01K/w2TpvXE3BNFUZRo0ISylvv06ROCg4MRGhqKmJgYZGVlAQDq16+P5s2b8095efXqFeTl5fHz509s376df6KKtrY2rK2tMXr0aPTs2RNycnJS+16qq4YNG2LSpElYvnw5hgwZAn19/QqfOf74PZacj+PvIhb25JWlfU3g1l6nwn7i4+OxYsUKTJ48GRoaGgJ8N1R+Ia/im2SoH4qiKFGgCWUtxOVyERYWhqCgIFy4cAFsNht9+vRB//79YWZmBjMzMzRr1qzMIuL5+fl49eoVYmJiEBMTg6tXr6Jfv37Q1NTEqFGj4OnpCSMjIwl/V9XbypUrERYWhnHjxuHWrVvlTi1vC0+odLFsLo+AyyOYGxKDb1l5mOpgUPa9XC7GjRuHZs2a4e+//65Uf7WRPFsyywIk1Q9FUZQo0DdWLfPs2TPY2tqif//++PDhA7Zs2YKPHz/i9OnT+Ouvv9CzZ09oaWmVeyKNvLw82rRpgxEjRmD16tWIiorC06dPMWzYMAQFBaF169bw8/PD79+/qxSrgoIC2Gw20tPTq9SOpHz79q3M6fk6deogKCgId+7cwerVq8tsQ1InrwB/Tl+5f/8+9u/fDxUVFZH0WRvoqatA3Oc1Mf6/H4qiKFlBE8paIicnB/PmzYOlpSVycnJw9+5dPHnyBJMnT4aamlqV22/Xrh0CAgKQlpaGlStXYseOHTA1NcXVq1cr3aa8vDxat26N6OjoKscnbjweD9HR0eXuyu3atSsWL16MBQsWYN26dSWuS+rkFQBYs2YNFi1ahCVLlqBz584i7bOmU1FgQ0eINaqVoaOuDBUFOoFEUZTsoAllLfDlyxfY2tpi48aN8Pf3R1RUFDp27CiWvhQUFDBnzhzExMSgRYsWcHFxwZIlSyp9TrSFhYVMnDSTmJiIrKysCguD+/v7Y+HChfjrr7+watWqYtfEffJKkRUrVmDu3LlYsmQJPX2lkhyMGoHFFM84JYvJgINhI7G0TVEUJS70I3AN9+nTJ3Tr1g3p6el48uQJzMzMJNKvvr4+rl+/jtWrV2P+/PkoKCjAihUryp1KL42FhQWOHj2KrKws1KlTR0zRVt3Dhw8BoMKamQwGA8uXLwebzcb8+fMRExODzZs3I5Mo4U7iN5HHVXTySuKXX1Al2fDx8cHx48exbNkyLFq0SOT91RYjbHRw4EGSWNrm8ghG2la8oYqiKKo6YZDKDh1R1d7Hjx/h6OiInz9/4ubNm1LbKLNx40b4+flh9uzZWLNmjVBJ5fv379G8eXPs3LkTEydOrHIsv/MKkZT+G/mFPMizmdBTVxHJ1KKDgwN4PB4iIiIEfubw4cPw8fEBIQROc3bjSaayyI/yA/6MeFnW/Y3ra7zAZDIREBCAESNGiLyf2mZUYCTuv00X6d8Zi8lAhxbqCB5vI7I2KYqiJIEmlDUUj8dDt27dEB8fj4iICIFK1YhTQEAAfHx8cPToUbi7uwv1rKurKz58+AAOhyP0CCcg/sLgr169grGxcaW+t69fv2LmzJkIr2MPObWmQvctqIKMNHTLuYONGzfS8kAikpKRDadNEcgTYXkfBTYT12faC1VHlKIoqjqgCWUNtXPnTkyePBnXrl2Dk5OTtMMBALi5ueHatWt48eIFNDU1BX7u4sWL6N27N+7du4cOHToI/JwghcGLFF2vTGHw6dOn4/jx40hJSYGCgoLAzxXJyiuEqf8VoZ8TVpy/C93oIWLHH7/H3JCYim8U0JqBZhgmQP1QiqKo6oZuyqmB3r17h9mzZ8PLy6vaJJMAsHXrVrBYLHh7ewu1ScfFxQVmZmaYMmUKCgoKBHrm+OP3cNoUgftv/5QcErYw+PFyyu38W1RUFHbs2IEZM2ZUKpkE/py8IglJEuqnNnFrr4NZ3Q1F0tbs7kY0maQoSmbRhLIGWrJkCdTU1EotTSNNDRs2xPbt2xEaGoq7d+8K/ByLxcL+/fsRExNTYmd0abaFJ2BuSAzyCnlCr2/j8gjyCnmYGxKDbeEJ5d6bn5+PsWPHwtTUFLNnzxaqn2Lt0JNXZNpUBwOsHmgGBTZT6J3fLCYDCmwm1gw0wxQH6S5LoSiKqgqaUNYw3759w8mTJzFt2rRqeQb2oEGDYGBggF27dgn1nKWlJebNm4fly5fjyZMnZd4nycLgixcvxsuXL3HgwAHIy8tXuh968orsc2uvg+sz7dGhhToAgFVBXlmUeHZooY7rM+3pyCRFUTKP/oapYQ4ePAhCCDw8PKQdSqmYTCa8vLxw+vRpfP36VahnFy1aBAsLC3Tv3h0cDqfEdUkWBl+xYgXWrFmDFStWlFvMXBC6EtiAQU9eET/tBsoIHm+Daz5d0KUpAwUZaSXuIYSgLiMPo2x0cX1mFwSPt6EbcCiKqhFoQlnDBAYGYtCgQWjYsKG0QynT2LFjwWAwcOTIEaGek5eXx5UrV2BoaIhu3brh0aNHxa5LqjD40qVLsXDhQixduhR//fVXpdr9+vUrjh49irFjx8KwhS4KvpdMPkSJnrwiOQaaqjD4GY3fJ2Yjdkl3uMrFovDiKjSKCkT+kaloHncQ/n1NoN+o+s0gUBRFVRZNKGuQ79+/4+XLl+jVq5e0QymXuro6bGxs8ODBA6GfrV+/Pq5evQoTExM4OTkhMDAQhBAkfP6FO4nfRF7H8d+Fwb9//45x48bB398fK1asEOqUmfz8fNy6dQvz58+HpaUlGjVqhBEjRuDp06cYMWIEXMy0wapESSRB0JNXJC8qKgoWFhaooyiHxgqFyP8YD5Om9cDkFeDhw4eVPjmKoiiquqIJZQ1SdOZ1Rcf/VQfm5uaVPlKxbt26uHz5MgYMGABPT09069YN268+F+tReEsO34CxsTHOnDmDffv2Yf78+eU+QwhBQkICtm3bBldXVzRo0AAODg7Yt28fWrVqhYMHDyItLQ3Pnj3DunXrMG9wR3DFlGTQk1ckj8PhwNLSEgAgJyeHwsJCGBgY4OfPn/jy5QvevXsn5QgpiqJEi86B1SBRUVFQUlKS2ok4wrCwsEBAQAAyMzNRr149oZ9XVVXFwYMHMXLkSHh5eSH+/iuw6zcWQ6R/ErJbr7+iQ4cO2Lp1K5o1a1bqfZmZmbhx4wauXr2KK1euICkpCXJycujYsSMWLlyI7t27o127dmAyS36OM9BURWd9DbGdvEKnVyUnPT0dSUlJ/A92bDYbBQUFMDAwQFZWFoA/R3W2aNFCmmFSFEWJFE0oa5CkpCS0aNECLBZL2qFUyNDwT+2+5ORktGnTptLtODs7I5ITjfZrb4sqtFLJNWiC4C0ni61D5HK5ePz4MT+BjIyMBJfLhaGhIfr06QMXFxd07dpV4DPIVw4wg9OmCJEmlGwmAysHSOb8duqPp0+fAkCxEcqihBIAtLS08ODBAwwfPlxqMVIURYkaTShrGFlIJgHRxvklh+DPPmZxYiAp/Tfqcn/iypUruHLlCm7cuIHv37+jXr166NatG3bs2IHu3btDT0+vUj1oN1DG0r4mIj15ZVlfE7qLWMI4HA5UVVX5x52y2WwUFhby/19XVxcPHz6UZogURVEiRxNKSuZJqmD3wMFDEf/gKphMJqytrTF9+nR0794d1tbWYLNF86Pk1l4H37LyRFJLk568Ih0cDgfm5ub8pQ1ycnLgcrlQUVFBkyZNULduXVy7dg05OTlQUlKScrQURVGiQRPKGobHk43TUEQZp6QKdpu3NcMK3wno1q0b1NTUxNbPVAcDaNRRwJLzcSjkEaGmwFlMBthMBpb1NaHJpJRERUWhb9++/P+Xk5MDAP7GHB6Ph8LCQnA4HHTq1ElaYVIURYkU3eVdg+jq6uLt27fgcrnSDqVCCQl/jjXU0al60qOnriKBCW8gcPNqDB48WKzJZJESJ69UsIO96LK+Ko+evCJFP378wJs3b/jrJwHwR6+L1lF+/foVSkpKdNqboqgahSaUNYiFhQWys7MRHy+aowfFKSoqCi1btkT9+vWr3JaKAhs6Yl4nKI3C4P8+eWWUjS501ZVLJM4MALrqyhhtq4eGj3ch99I6umZSiopKYf27dFfRCGVRQpmYmAgrK6tK1WGlKIqqruiUdw1SdATg06dPYWxsLN1gKhAVFQVzc3ORtedg1AjBkckiL2wOSL8wuIGmKvz7msAfJvidV4ik9N/IL+RBns2EnroKP9E1KxiLwYMH48mTJ7CyspJavLVZVFQUVFRU+FUMgP+NUP67FmXbtm1x5swZEELAEFNBe4qiKEmiI5Q1SIMGDWBkZIRLly5JO5Ryff/+HZGRkbC1tRVZmyNsdMSSTALVqzC4igIbJk3rwVxHDSZN6xUbNe3fvz+aN2+OjRs3SjHC2q1oQ86/qxj8d4QSAJo2bYqPHz8iNTVVKnFSFEWJGk0oa5hx48bh1KlTSE9PF0l7v/MKEZeWiafvvyMuLRO/8wqr3ObBgwfB5XIxcuRIEUT4R3N1JZhpsMGAaJNKFpOBzvoaMlEYnMViwcfHBydPnsT79++lHU6txOFwSpxU9e9NOS1btgQAKCv/WZZAp70piqopaEJZw3h4eIDH4+HAgQOVbiPh8y/4n4+D/bpwmPpfQe+tdzFg53303noXpv5XYL8uHP7n45Dw+ZfQbRNCsGvXLgwcOBCampqVjhH4U8h9z549GDRoEDQ0NHB5+RjwCgsAER5hKGuFwceNGwdVVVVs2bJF2qHUOj9//kRCQkKxDTlA8U05ysrK0NLSwpcvX6Cnp0c35lAUVWPQhLKGadiwIQYPHowtW7bwj3kTVEpGNkYFRsJ5820ERyYjOSO7xHgfAZCckY3gyGQ4b76NUYGRSMnIFriPc+fO4fXr15g0aZJQsQFAVlYWwsLCMG3aNBgZGaF58+aYPHkyPn36BF9fX9y9cg6rBpsDIlyTJmuFwevUqQMvLy/s3bsXP3/+lHY4tcp/T8gp8u8pbwAwMDBAQkICbG1t6QglRVE1Bt2UUwMtW7YMbdq0wbx587B161aBnjn++D2/7iGACtcjFl2//zYdTpsisLSvCdwqKFWTkZEBb29v9O7dG/b29hXGxOPxEB0dzT/a8N69eygoKICenh5cXFywatUqODo6FtspbgMgI7ugSoXBizZKyGph8KlTp2LDhg0IDAzEzJkzpR1OrcHhcKCkpAQjI6NiX//3phzgT0L56NEjeHh4ICQkBHl5eVBQUJB4vFTV5eTkICsrC3l5efj9+zdUVVWhqKiIevXqieQ0sIyMDMTExBT77/Pnz8jNzUVaWhqAP+Xi6tatC2NjY5iZmfH/09PT4xfXpyhJoAllDaSvr49Vq1bBx8cHgwcPrjB52xaeUOkEjPv/hbfnhsTgW1YepjoYlHnvjBkzkJubiz179pS5s/XTp0+4evUqrl69imvXruHLly9QUVGBo6MjNm7cCBcXF+jr65e7M7ZKhcEZAJdbCIXYc/BYtF3g56oTLS0tuLm5ISAgANOmTRPZKT5U+aKiotCuXbsSf96ljVAeOXIENjY2yM/PR3R0NGxsbCQeLyWc5ORkPHjwoFhyl5SUVOq9ioqKxRK8Nm3aoFOnTvy1s+X5+fMnTpw4gf379/NHsOXk5Pjt2dvbQ0FBAVevXkXHjh2hoKCAjIwMxMXF4dq1a/j+/TsAoEmTJhg9ejQ8PDxKfMihKHFgECLCBWdUtcHj8dC1a1ckJSUhIiICzZs3R2FhIeLj4/kvw5cvXyJVXgef9ZxE1u+agWaljurt2rUL3t7eOHjwIEaPHs3/em5uLu7evcsfhXz+/DmAP3X8XFxc4OLiAjs7O8jLywsdS0pGNuaHxuBO4jewmIxyE8ui6531NTDWVAF9HTti2LBhCAoKErrf6uDp06ewsLDAiRMnMHToUGmHUysYGxvDycmpxKxA0d/FkydPYGlpiXPnzqF///5ISkqCkZERVq9eDR8fH+kETZUrOzsbISEhCAoKQnh4OIA/O/TNzMxgamoKExMTNGjQAPLy8vj06RMaNWqE3NxcpKSk8N+zcXFxyM7ORt26deHm5oZx48bB2tq6xIfi2NhYrFu3DqdOnUJeXh5cXFzg7u4OCwsLGBoa8j+YlIcQgrS0NMTExODChQs4cuQIvn//jg4dOmDKlClwd3enZaoosaEJZQ2WmpqKLl264MePH2jZsiWeP3+O/Px8AH8+vRqa2yHZdDQIgyWydYcKbCauz7Qvtu5w+/btmDp1KqZNm4bNmzfj1atX/AQyIiICOTk5aNKkCbp3747u3bvD2dkZDRs2FEk8wJ9NRkci3yM8/gvepxdfF8rAn6LlDoaNMNJWh7+bOzg4GKNHj0ZQUBA8PDxEFoskOTo64vfv33j48CH9JSJmv379Qr169RAYGFji30tMTAzatGmDBw8ewNbWFi9evICJiQkiIiIwb948aGtr4/jx41KKnCrNjx8/sGTJEhw4cAA/f/5E165dMW7cOPTq1Qvq6upCtcXj8fD69WscPXoUBw4cQGpqKlq3bo0FCxbA3d0deXl5+Pvvv7FmzRpoa2vD09MTo0ePhpaWVpW/j9zcXJw7dw6BgYG4du0aHBwcsHv3bn75KooSJZpQ1lAPHz7Ezp07cfLkSeTm5kJBQQGzZs2Cs7MzTE1Noa6ujlGBkbj/Nl2k9RtZTAY6tFBH8HgbEEKwadMm+Pn5oXfv3tDU1MTVq1eRmpoKBQUFdOnSBd27d4eLiwtMTU0lkvSUVxj8vyZMmIAjR44gMjISZmays9O7yIULF9CnTx/cuXOHnhktZnfv3kXnzp3x7NkztGnTpti1V69ewdjYGLdv30bnzp2Rm5sLZWVl7NmzBy9fvsTp06eRnJwspcip/woJCcHUqVORlZWF6dOnY9y4cWjRooVI2uZyubhx4wa2b9+O8+fPw9raGl+/fkVqaioWLFiAuXPnim097dWrVzFp0iR8/PgRS5YswezZs0WyzpOiitCEsob5/v07Zs2ahaCgILRs2RLjxo2Di4sLRo0ahaSkJCxduhQzZ87Eu/QcOG++LbY4ZrfOxeZlc5GYmMj/momJCT+B7Ny5s0DriaQpJycHtra2yMvLw+PHj6GqWv1rUf4bj8eDiYkJjI2NERISIu1warSAgADMnTsXv379KrGGMjExEQYGBrh58yYcHBwAAHp6ehg2bBjat2+PIUOG4MOHD2jatKk0Qqf+X0ZGBsaPH4+zZ8+ib9++2L59u0hGCcsyb948rF69GkwmE/Pnz8fy5cvF1leR379/w9/fHxs3bsSwYcNw6NAhusaaEhn6L6mGIITg9OnTmDZtGnJzc7F79254enryd/k9evQIixcvxty5c3Hs2DHYTtlQ4brCSsfC42LRoavIfPsWnTp1wrhx4+Ds7CzWl7M4KCkp4dSpU7CyssLEiRNx9OhRmZo6ZjKZ8PX1hZeXFxITE6Gvry/tkGosDoeDNm3alPrL+b+bcoD/lQ6aNm0aACAyMhIDBgyQTLBUCd++fYOzszNSUlJw6tQpDBo0SKw/6ydOnMC6dev4NXT//vtvsFgsLFmyRKz9qqioYN26dbC1tYWbmxsKCwtx5MgRgdZnUlRFaE2BGoAQgrlz52Lo0KHo2LEjXrx4gYkTJxYrGVGnTh1s3LgRDx8+BJfLxbnHiWI7qpDBZKFhm6749u0b7ty5Aw8PD5lLJosYGhpi3759OH78OHbt2iXtcIQ2cuRIaGhoYPPmzdIOpUaLiooqUX+yyL9PyilSlFBqaWlBS0uL1qOUoi9fvsDR0REfPnzArVu3MHjwYLEmdadPn8bw4cPh7u7Of6+sWrUKS5cuxcKFCyGJScNBgwbh1KlTOHv2LIYPHy6RPqmajyaUMo4QAj8/P6xduxabNm3CmTNnyp06a9++PW7dewg5tSZijStfvi7klWVrirgsQ4cOxdSpU+Hj4wMOhyPtcISipKSEyZMnY//+/cjIyJB2ODXS79+/8fLlyzITyn+flFPEwMAAiYmJ4PF4sLW1pSfmSEl2dja6deuGL1++4NatWzA1NRVrfykpKRg/fjwGDx6MAwcO8P9tzJ07F+vXr8fKlSuxatUqscZQpH///jh58iROnz6N3bt3S6RPqmajCaUMI4Rg5syZ2LRpE7Zv3y5w6ZG0n/n4s79ZjLEBSEr/LdY+JGn9+vVo27YthgwZgh8/fkg7HKFMnjwZXC6X/tIQk2fPnoHH45U4w7tIWVPeubm5+PDhA2xtbfHkyZNi1ynJWLBgARITE3Hjxg20bt1arH0RQjBhwgTUqVMHu3fvLrEhxs/PDwsWLMCSJUv4py6JW//+/TFx4kTMnj27zJqaFCUomlDKsFOnTiEgIAA7duzA5MmTBX4uv5Anxqgk348kKCgo4OTJk/j+/Ts8PDxkaoqoUaNGGDVqFLZu3covG0WJTlRUFOTl5WFiYlLq9f+elAOAX7YlISEBdnZ2yMnJ4ddgpSTjzp07CAgIwIoVK8r8uxOlAwcO4MqVK9i7d2+x073+bfHixTAxMcHYsWMl9rO6bt06NGjQAJ6enjL1XqOqH5pQyqgvX75gypQpGDJkCLy9vYV6Vp4tmb92SfUjKXp6ejh48CDOnj0rc2sSfX198fHjRxw7dkzaodQ4RRtyyiq+X9oIZfPmzcFkMpGQkAALCwvIycnRaW8JKigowPjx42FnZ4cZM2aIvT8ej4cVK1Zg6NCh6NWrV5n3ycvL48CBA3jx4gXWrFkj9rgAoG7duti6dStu3LhB1/JSVVKzfuPXIkUjktu2bRP6WT11FTFPeP+vn5qmb9++mD17Nv766y+ZevkaGxujV69e2LhxIx2FEDEOh1PmdDdQ+qYceXl56OnpISEhAYqKijA3N5epf0+y7p9//kFCQgK2b98ukVqM169fx5s3bwRKXtu1awdvb29s3boVeXl5Yo8NAPr06YMWLVrI5MZDqvqgCaUMioqKwpkzZ7B582Y0atRI6OdVFNjQaSDeGpCs7IwyC4bLuhUrVsDGxgZDhw7Ft2/fpB2OwHx9ffH8+XPcuHFD2qHUGDk5OXjx4kWZG3IA8Kst/HeNZNFObwB0Y46E7dy5Ex06dEC7du0k1l+bNm1gZ2cn0P3e3t74+vWrxOrHMplMeHl54eTJk0hPT5dIn1TNQxNKGbRz505oaWlh2LBhlW7DwagRWEzxjFMyQfD9xV1ER0eLpX1pk5OTw/Hjx5Gbm4tRo0aBx5ONtaKOjo5o27YtNmzYIO1Qaoznz5+Dy+WWm1AyGAzIyclVmFC+efMGX758EWu81J91q9evX8ekSZMk0t+vX7/wzz//wNPTU+ByRMbGxrC3t8fOnTvFHN3/eHh4oLCwkB6CQFUaTShlTGZmJo4ePYoJEyZU6YSDETY6YqtDyQMDSh+e1OhdxVpaWjhy5AiuXLmC1atXSzscgTAYDPj5+eHy5cuIi4uTdjg1AofDgZycXIXlZuTk5IpNeQN/Eso3b96Ay+XyR64iIyPFFiv1x/nz56GsrIzBgwdLpL9nz56By+WiS5cuQj3n4eGBO3fuSKzcV8OGDWFsbCxzpdGo6oMmlDImJCQEubm58PT0rFI7Bpqq6KyvIfJRShaTgc76GvAc0geHDx8u8Uu0JunevTsWLVqERYsW4datW9IORyDDhg1D06ZNsWnTJmmHUiNERUXB1NS0wvOX2Wx2qSOU+fn5SElJga6uLjQ1Nem0twRERUWhbdu2UFJSkkh/T58+hby8vNBliWxsbPjPS4q5ublE+6NqFppQyphHjx6hdevWIjn3d+UAM7BFnFCymQysHGCGbt26ISsrC69evRJp+9XN4sWL0bVrV7i7u+PTp0/SDqdC8vLymDZtGoKDg/H582dphyPzOBxOudPdRUoboTQ0NATwZwqWwWDAzs6ObsyRgKioqHI3UYmjvzZt2gh9vKGBgQFUVFQQFRUlpshKsrCwwLNnz2r0QAAlPjShlDGifBlqN1DG0r6irb+2rK8JtBso8xe7S/JlKA0sFgtHjx4FAAwfPhxcLlfKEVXMy8sLcnJy2L59u7RDkWm5ubmIjY0V6OextBFKXV1dsNnsYusoHz16JBP/hmRVTk4OXr9+DXNzc4n1mZ6ejiZNhD+ZjMVioW3btnj27JkYoipd06ZNkZeXh9+/a86hFJTk0IRShhQWFuL58+ci/XTt1l4Hs7obVqmNojI0s7sbYVh7HQB/apsZGBjU+IQSADQ1NXH8+HFERETA399f2uFUSE1NDePGjcOOHTuQk5Mj7XBkVmxsLAoLCwUeofxvQslms9GiRYtiCeXv378RGxsrlnipPx8CCCFQU1OTaL+VPRtcTU0NWVlZIo6mbOI8w5yq+WhCKUMyMjKQm5sLPT09kbY71cEAqweaQYHNFHpNJZMBkMJ8pF/aAsfGxX9hNm/eHB8+fBBlqNWWvb09/v77b/z999+4fPmytMOp0IwZM5CRkYFDhw5JOxSZxeFwwGKx0KZNmwrvLW3KGyi+09vKygosFouuo6QoSibRhFIGieNTpFt7HVyfaY8OLdQBoMLEsuh6x5YacK+biKxnV2FlZVWsEG9t+7Q7Z84c9OzZEyNHjkRKSoq0wylXy5YtMWDAAGzcuFFmyh5VNxwOByYmJlBUVKzw3tKmvIHiCaWKigratGlD11HWMGw2G7m5udIOQyBFcValgghVe9GEkuLTbqCM4PE2uObTBaNsdKGrrlziRB0GAF11ZYyy0cX1mV0QPN4Gaxb/hQ4dOiA7OxtWVlbSCL1aYDKZCA4OhrKyMtzc3EpNIKoTPz8/xMfH48KFC9IORSZFRUUJNN0NlD9C+fbtW/41Ozs7OkIpRkW78X/+/CmxPlu1alXpZQyZmZkS240O/FnGoa2tDRWVmnfKGSV+NKGUIUUjfuLegWegqQr/viaImOWAWH8XmH++jMbPg3FhWifE+rsgYpYD/PuaQL+RKv+Zmzdvol69eoiNjcW0adP4cda2UUp1dXWcPHkSjx49wvz586UdTrk6dOgAW1tbbNy4UdqhyJz8/HzExMQInFCWN0JZWFiI5ORkAH/WUb5+/VpitQdrG2VlZRgYGEj00AULCwukpaUJXVWBx+Ph2bNnaNu2rZgiK0nSO+CpmoUmlDJEQ0MD9erVk2gpHhUFNtTwG8zv72HStF6ZxykqKCjg8ePHYDAY2LZtG86fP4+XL1/CwMBAYrFWF7a2tli3bh3Wr1+Pc+fOSTuccvn6+uLWrVu1YvOUKMXGxiI/P1/gX76lbcoBwP/5KJr2pgXOxc/c3FzipXgACF0w/O3bt/j165fEdqTzeDyaUFJVQhNKGcJgMGBhYSHxX/5l/TL8LwMDAwQGBgIABgwYgLS0tFr7cpoxYwYGDBiAsWPH4t27d9IOp0wDBgyAnp4ePY5RSFFRUWAymQKPHpU15a2trQ15eXl+QtmyZUuoq6vTaW8xsrCwwNOnT5Gfny+R/po3bw49PT2cOnVKqOcePXoEABJLKMPDw5GRkQFHR0eJ9EfVPDShlDGS/nQNCJ5QAn+OCxs4cCB/o4eZmZk4Q6u2GAwGgoKC0KBBAwwdOrTYZqXqhM1mw8fHBydPnqz2G4mqEw6Hg9atW0NZWVmg+8ua8maxWGjZsiU/oWQwGLC1taUbc8SoT58+yMrKwtmzZyXSH4PBwMSJE3H8+HF8//5d4OcOHToEa2trNGrUqMS133mFiEvLxNP33xGXlonfeVVfBrVz5060bt0aHTt2rHJbVO1EE0oZ06lTJ7x79w4xMTES61OYhBIAjh8/zv9F6+XlJa6wqr369evj1KlTeP78Ofz8/KQdTpnGjRsHFRUVbN26VdqhyAwOhyPU6HtZI5RA8Z3ewJ9p78jISLr7XkxMTEzQuXNn7Nq1SyTtCZLcjRs3DlwuV+AyXW/fvsWVK1fg7e3N/1rC51/wPx8H+3XhMPW/gt5b72LAzvvovfUuTP2vwH5dOPzPxyHh8y+hv4e0tDScPXsWkyZNqnXr3inRoQmljOnTpw8aN26M3bt3i6Q9QV6GwiaUubm5YDAYYDAYuHXrFlauXCmSWGWRhYUFAgICsH37dpw4cULa4ZRKVVUVEydOxJ49e/Drl/C/jGqbgoICPH/+XOANOUDZI5RAyYTS1tYWP3/+rPHHlkrTpEmTEB4ejpcvX1bqeWGTO01NTQwZMgSrV68WaMPVrl27UL9+fQwdOhQpGdkYFRgJ5823ERyZjOSMbJD/3E8AJGdkIzgyGc6bb2NUYCRSMrIF/n4WL14MZWVljBo1Sog/BYoqjiaUMkZOTg7jx4/HoUOHKn2CgrAvQ2ETyiNHjiAnJwf79u0DACxYsAB3796tVKw1gZeXF4YPHw5PT0/Ex8dLO5xSTZ8+Hb9//+avgaXK9uLFC+Tl5QmVUJb3M2RgYICkpCT+dWtrazAYDDrtLUaDBg2CtrY2pk+fzj/pSxBVSe7Wrl2LnJwc+Pj4lNvH69evsXXrVkyaNAnn477BaVME7r9NBwBweeXHWnT9/tt0OG2KwPHH7yv8nq5cuYLAwECsX78e9evXr/B+iioLTShl0IQJE5CdnY3NmzcL9VxlX4a5bBWBE8rs7GysX78erq6uGDduHH/KxsnJqdaWQmEwGNi9eze0tLQwePDganncoZaWFoYNG4aAgACxl6WSdRwOBwwGQ6hyLhVNeXO5XP7mLVVVVZiamtKNOWKkoKCAvXv34vr169i7d69Azxx//L5Kyd2dNC42b96M4OBgnD9/vvRnuFx4eHhAW1sbDbuOwtyQGOQV8irsq7S+8wp5mBsSg23hCWXel5mZCU9PTzg7O2PChAlC9UFR/0UTShmkq6uL2bNnY9myZQIXzK3Ky/Bsvhl4zW0F6mf+/Pn48OED1q5dCwDYsmULjI2NkZeXBysrK6FGA2qSOnXq4NSpU0hMTOTX6axufH19kZSUhNDQUGmHUq1FRUWhVatWqFOnjsDPVDTlDaDEtDcdoRQvFxcXeHp6ws/Pr8JKDNvCE0SS3P3S6Yg+ffpg1KhRpX5g2LhxIx4+fIjRS3dhyy3RVIdYfzUeJ0oZqczKykK/fv2QmZmJffv20bWTVJXRhFJGLVmyBAYGBhg7dmyFI0pVfRlywYB8J49yP+kCwJ07d7BlyxasWLEChoaGAP78Ir1x4wYUFRXx7t07uLu7C9V/TWJqaoqdO3ciMDAQBw8elHY4JVhYWKBr167YsGFDrU38BcHhcISa7gbKn/Ju1qwZFBUVS2zMefHiBTIzM6sUK1W+9evXo1GjRnB0dERSUlKp9xx//B7rr4pmqcqGa/EY/NcGtGnTBt27d8f9+/f514KCgjBnzhx4z16E4BeirQqx+HxcsTWVv379Qs+ePREVFYXLly9DR0dHpP1RtRNNKGWUoqIiDhw4gKdPn8Lb27vMHaGieRn++eRa1idd4M/oiru7O+zs7DBjxoxi15o0acI/3u/EiRPYs2dPFeORXWPGjOEvBajscWzi5Ofnh8jISDo6VobCwkI8e/ZM6PqqbDa7zA9+TCYT+vr6JUYoCSF4/PhxleKlylevXj2Eh4eDxWLB3t4eb9++LXY9JSMbS87HibTPlVffYO+xEFhYWMDFxQVhYWHYs2cPxo8fj0mTJiHTsCcKhfzgX5FCHsH80D+VQVJSUuDi4oKYmBhcu3YNHTp0EGlfVO1FE0oZ1r59e+zfvx+BgYEYP348uFxusevieBn+95Mu8GcRub29PVRVVXHq1CmwWKwSzzk6OmLhwoUAAG9vbzx//lykccmSrVu3Ql9fH4MHD670xipx6dWrF4yMjGih8zK8evUKOTk5Ih2hBEru9DYyMkL9+vVpYi8BOjo6iIiIgIKCArp06YIbN27wr80PjRFLcrfiyltcuHABXbp0gaurK7y8vODp6QmfxatxNzFd6JmkinB5BHcSv2Hppt1o3bo1kpOTcf36ddjY2Ii0H6p2owmljBs9ejQOHz6MQ4cOYcyYMcU2fIjrZVj0SRcAnj59Cnt7e6ipqeHWrVto2rRpmc8uXboUtra24PF46NSpE37//i3S2GSFsrIyTp06hQ8fPsDLy6taTS8zmUzMnDkToaGhePPmjbTDqXaKjs8T9vSS8jblACUTSiaTCRsbG7oxR0KaNWuGiIgI6Ovrw8nJCR4eHnj8OgV3Er+JLbkLu/0Eb968AYvFgrKyMs6cOYN5gRfBEtdSRh4XARejMGrUKLx48QJWVlZi6oiqrWhCWQMMHz4cx44dw+nTp2FmZoabN28i4fMvsb4MY99/w/z582FtbY0mTZogPDwcmpqa5T7LZDJx/vx51KtXD79+/UKnTp2qVTIlSUZGRti3bx+OHj1a7ZYAjB49Gurq6kJXEagNOBwODA0NoaqqKtRz5W3KAf4klO/fvy92opKtrS0ePnxYa39GJK3oPbZv3z6cPXsW/WdvAKNEHQwRITx4bzwKNTU1PHv2DMnJyejTpw8epmSBK66/biYLBl36Y8eOHahXr56YOqFqM5pQ1hBDhw7Fs2fP0KxZM3Tr1g0eK4PE9kmXCaCv3zps2LABixYtQmRkZKnHg5WmYcOG/PWU0dHR1XbHsyQMGzYMU6ZMwfTp0yV+nGZ5lJSUMHnyZAQFBQl1VFxtEBUVJfR0NyDYlDePxyu2hs/Ozg4ZGRnFRi4p8WIwGBg/fjxevnwJVSNbEIjpJcpgQsemJ+7evQsTExNoaGhgx94gyKs1EU9//+9LNk8kxzRSVGloQlmDGBkZITw8HLt370ZygarYPunyADCamSI6OhqLFy+GvLy8UM937NiRX1Zo+/btOHPmjBiilA0bNmyAmZkZhgwZUq129E6ePBlcLldkJzLVBFwuF0+fPq1UQlnephyg9NJB1tbWAECnvaWgjpoG8tjCjUILK5Mrh9zC/72kk9N/i2s8lI8ASEqvnUuNKPGjCWUNw2QyMXzMOLDrlT/9XFU8pQbQaWFQ6ednzZoFFxcXAICbm1uJ3ZW1hYKCAk6dOoX09HR4eHhUm+lNTU1NjBw5Elu3bkV+fr60w6kWXr9+jezsbKF3eAMVj1A2adIEKioqxRJKNTU1GBsb0405UiCN5C6/UDJnt0uqH6r2oQllDSQLn3QZDAaOHTsGTU1NFBYWwtbWttj6sdqkefPmOHjwIEJDQxEQECDtcPhmzpyJtLS0ansGuaQVLUuobEJZ3gglg8EoUToI+N86SkqypJHcybMl8+tYUv1QtQ/9l1UDyconXTU1NYSFhYHFYuHr16/o2bOniCKTPf369YOfnx9mz55dbRIIExMT9OjRgxY6/38cDgf6+vqV2tBQ0aYcoOROb+BPQvn8+fNqV16qppNGcqenriKuFZt8jP/vh6LEgSaUNZAsfdK1srLij8qFh4fD39+/ym3KqlWrVsHa2hpDhw5Fenq6tMMB8KfQ+bNnz3Dz5k1phyJ1HA6nUqOTQMVT3kDpCaWdnR14PB6ePHlSqX6pypFGcqeiwIamSskavqKko64MFQW2WPugai+aUNZAsvZJd/LkyRg0aBAYDAaWLl1arLBwbSInJ4cTJ04gOzsbo0ePLvP0I0nq1q0b2rRpU+sLnfN4vEpvyAEq3pQD/EkoU1JSitWSbd26NVRVVavNqHVtoaLAhk4DZbH2UZTccblchISEwN7eHvERoSA8bsUPVwKLyYCDoWDVOCiqMmhCWQNJ8mUoCgwGA0FBQfzzZHv37o3Pnz+LpG1Zo6WlhcOHD+PSpUv8nfDSxGAw4Ofnh0uXLuHFixfSDkdqEhISkJWVVemEUtARSgDFCsqzWCxYW1vTjTlS4GDUCCymeD6as5gMdNCrhw0bNkBfXx+DBg0Cl8vFslHOYDDFM0rJ5RGMtKVndlPiQxPKGkrcL0NRf9KtW7cuzp07Bzk5OeTl5cHW1rbEUZK1RY8ePbBgwQIsWLAAERER0g4Hbm5uaNKkCTZt2iTtUKSmsifkFKloUw5QeukggBY4l5YRNjoiPxiiCJdHsGOmO+bNm4fOnTvjyZMnuHv3LqaOHozO+hoif3ezmAx01teAfiPxlkKiajeaUNZQ4n4ZiuOTbtu2bbFz504AQFJSEoYMGSLyPmSFv78/7O3t4ebmJvXRWnl5eUybNg3BwcFSj0VaoqKi0Lx5czRo0KBSzwuyKadRo0ZQVVUtNaH88uUL3r17V6m+qcox0FRFx5bqYBDRLj0h3EJwU2PhM84NycnJOHToULGR75UDzMAWcULJZjKwcoCZSNukqP+iCWUNZaCpKpOfdMeNG4fRo0eDwWAgNDQUW7duFUs/1R2LxcLRo0dBCMHw4cOlPlrr5eUFFovFT/hrGw6HU+npbuDPCCUhpNy/RwaDUeZOb4AWOJe0z58/493JleAVFgCiKsRGCOTYLISvnYilS5eiSZOSJ+NoN1DG0r4mounv/y3rawJtMS+DoiiaUNZgov+kS8T+SZfBYGDHjh0wMjICk8mEj49Prd3h2rhxYxw/fhy3bt3CsmXLpBpLgwYN4OHhge3btxfbNFIb8Hg8REVFVXqHN/BnhBKAQNPe/00oNTQ0oK+vTxNKCXr06BEsLS3xLuYxJljWB0S1zZHBwIoBbaDfRK3c29za62BWd0ORdDm7uxGGtadrJynxowllDSb6T7oMDG0JsX/SVVFRQUhICBQUFMDj8eDg4FCtjiWUpK5du2L58uVYvnw5rl69KtVYfHx8kJ6ejuDgYKnGIWlv377Fz58/qzxCCaBSpYOAP+WD6MYcyQgMDETnzp2ho6MDDoeDhe6OUknupjoYYPVAMyiwmULPNLGYDCiwmVgz0AxTHPQrEypFCY0mlDWcKD/pqr4Nx1qv/jh37pxI2iuPsbEx9u3bBwDIyspCx44da+2mhLlz58LFxQUjRozAhw8fpBaHvr4++vfvj02bNlWLkkaSUrQhpyojlEUJpSAjlGlpafj9u/gpVLa2toiOjq51o8OSlJ+fD29vb3h6esLDwwPh4eFo2rQpAOkld27tdXB9pj06tFD/84UK1nMWxdahhTquz7SnI5OURNGEshao6stQnsVA+sUALHfvhL59+2Lw4ME4efKkmKL9n+HDh8PLywtMJhNxcXGYMGGC2PusjphMJoKDg6GoqAg3N7cKR7nEyc/PD69evcKlS5ekFoOkcTgc6OjoQENDo9JtFE15C1o6KDExsdjX7ezsUFhYyE9uKdH6+PEjHBwcEBQUhL1792LXrl1QUFAods9/kztWBa9SUSV32g2UsbybJtL2eqNDw0LoqiuXmIBnANBVV8YoG11cn9kFweNt6JpJSuJoyfxawq29Djq21MD80BjcSfwGFpNR7i7wousdWqhjVtdmaLfiGoDJOHbsGDw8PODu7o7c3FyMHj1arHFv3rwZkZGRiI2NRWBgILp27YqRI0eKtc/qSENDAydOnIC9vT0WLFggtRqVHTp0gLW1NTZs2IDevXtLJQZJi4qKqtJ0NyDclDfwp3RQ27Zt+V83MzODkpISHj58iE6dOlUpFqq4+/fvY/DgwWAwGLh9+zZsbGzKvFe7gTJ8LeSR++AybqVyodDcAux6jQHG/1I8Bv7U6XUwbISRtjoi2cB49uxZMH59xu6JTlBVVcXvvEIkpf9GfiEP8mwm9NRV6Ak4lNTRf4G1iHYDZQSPt0HC5184Evke4fFf8D49u9j+xdJehkXrFwsKCsBms3Hw4EEoKipi7NixyM3NxcSJE8UWs6KiIk6dOgVzc3NkZ2dj7NixaN++PYyMjMTWZ3XVoUMHrFmzBn5+fujU6c9osaQVFTofNmwYnj59Wum6jLKCEIKoqCj4+flVqR1BN+Woq6ujfv36JdZRstlstG/fnm7MESFCCPbs2YNp06bBxsYGp06dQuPGjUu9t7CwEOfOnUNAQADu3LkDHR0d/DV1Kjw9B0FeWVXsyV1ISAicnZ2hqvonOVVRYMOkqfBnylOUWBGqVsvKLSCxH36QqOQMEvvhB8nKLSh5T1YWAUCOHDnC/xqXyyVTp04lAEhAQIDY4zx9+jTBn9odRF1dnWRnZ4u9z+qIx+ORfv36kfr165N3795JJYaCggKiq6tLRo4cKZX+JenNmzcEALl48WKV2rlx4wYBQBITEyu8t3379sTDw6PE1+fMmUOaNm1KeDxelWKhCMnNzSWenp4EAJk6dSrJy8sr9b6MjAyybt06oqurSwCQTp06kdOnT5OCgpLvSXH59OkTYTAYJDAwUGJ9UlRl0DWUtVzRJ11zHTWYNK1X6ifr0qbrmEwmtmzZgtmzZ2PGjBlYs2aNWOMcNGgQZsyYASaTifT0dDg5OdXKTToMBgMHDhyAmpoahg4diry8PInHwGazMWPGDBw/fhypqakS71+SoqKiAEBkU94VjVACZe/0trW1RVpaWo3/Mxe31NRU2NvbIzg4GPv378fWrVshLy9f7J5Xr15h8uTJ0NLSwoIFC2Bvbw8Oh4M7d+5g0KBB/BFnSTh//jwYDAZcXV0l1idFVQZNKKkKlfXLkMFgYM2aNViyZAnmzp2LpUuXijXJW7t2LaysrKCgoID79+9jzpw5YuurOqtfvz5OnTqFZ8+eYfbs2VKJYfz48VBWVq7xhec5HA60tLTQqFHVjhoVdFMOUH5CCYCWD6qCO3fuwNLSEmlpabh79y7Gjh3Lv8bj8XD58mX07NkTxsbGOHPmDGbPno3k5GQcPHiwSrv8qyI0NBRdunRBw4YNpdI/RQmKJpRUhRgMBlgsVqm/DBkMBvz9/bFq1Sr4+/tj3rx5Yksq5eXlcfLkSSgrK0NeXh7r1q3DhQsXxNJXdWdpaYnNmzdj69atOHXqlMT7r1u3LiZOnIjdu3cjKytL4v1LCofDEUkiIeimHOBPQvn582f8/Pmz2NcbN24MPT09uo6yEggh2LZtGxwdHWFsbIwnT57AysoKAPD792/s3LkTJiYm6NmzJz5//oyDBw/i/fv38Pf3L3NdpSRkZmbi+vXrGDBggNRioChB0YSSEoicnFy5vwznzp2LzZs3Y82aNfDx8RFbUqmrq4vg4GDk5+eDyWRi4MCBtXYKcNKkSXBzc8P48eMRHx8v8f6nTZuGrKwsBAUFSbxvSSD/vyGnqtPdgOCbcoCySwcBf0YpaUIpnJycHHh4eGDatGmYOnUqrl27hkaNGiE5ORmzZ8+GlpYWpk6dChMTE9y+fRscDgejR48uUTZIGi5evIiCggKaUFIygSaUlEAqSigBYMaMGdi1axe2bNkCLy8vsRW/7t27N+bOnQtCCPLz82FnZ4f8/Hyx9FWdMRgM7NmzB02aNMGQIUMkXvRaR0cHQ4cOxebNm6V+1rg4vH//Hunp6SJJKIUdoQRQ6ocEOzs7cDgcqaydlUXv379H586dceLECRw+fBgbN27Ew4cPMXjwYLRo0QL79u2Dp6cn3r59i9OnT6Nz585gMER5XG3VhISEwMrKCtra2tIOhaIqRBNKSiCCJJQA4OXlhQMHDiAwMBBjx44VaESmMpYvX45OnTpBWVkZqamptfYTvKqqKk6fPo34+HhMnz5d4v37+fnh3bt3CA0NlXjf4iaKE3KKCLMpR01NDerq6mWuo8zPz0d0dHSVY6rpwsPDYWlpiW/fvuHWrVvgcrmwtLREly5dEBcXh23btiE1NRXr1q2Drq6utMMtIScnB5cuXcLAgQOlHQpFCYQmlJRABE0oAWDMmDE4evQojh49iuHDh4vlZBc2m43jx49DRUUFqqqquHjxotSKfUubmZkZduzYgX379uHQoUMS7dvS0hL29vbYuHGjRPuVhKioKDRp0gRNmjSpclvCbMoByt6Y065dOygoKNCNOeUghGDz5s1wdnZGq1atMHToUPTr1w9jxoyBpqYmLl26hLi4OHh7e0NFRUXa4Zbp2rVr+P37d639sEzJHppQUgIRJqEEgGHDhuH06dM4e/YsBg8eLJYpuqZNm+Lo0aP49esXFBQUMHfuXNy9e1fk/cgCDw8PeHh4wNvbG3FxcRLt29fXFw8ePKhxSQ6HwxHJdDcg3JQ3UHZCKS8vD0tLS7qOsgzZ2dkYOXIkZs6ciVatWuHRo0fYvn07Bg0ahJcvX+LSpUvo0aMHmMzq/6svJCQExsbGaNWqlbRDoSiBVP+fKqpaEDahBID+/fvj/PnzuHr1Kvr164fs7GyRx+Xk5AR/f3/+GkoXFxd8+/ZN5P3Igm3btqFFixYYMmSIRHde9+nTB4aGhtiwYYPE+hQ3QojIdngDwk15A2UnlADdmFOWhIQEmJiY4Pjx4wCArKwsrFixAqmpqdi+fbtMJWYFBQX4559/6OgkJVNoQkkJhM1mV2rqukePHrhw4QLu3LmD3r17iyXRWbhwIbp164Y6deogOzsbHTt2rJGbRCqirKyM06dPIyUlBZMmTZJY4Xcmk4mZM2ciNDQUb9++lUif4vbhwwd8/fpVZCOUlZnyTk9Px/fv30tcs7OzQ3JyMj5+/CiS2GTd9+/fMWHCBLRq1QpJSUlo164dzpw5g8TERMyaNQtqamrSDlFot2/fRkZGBl0/SckUmlBSAqnMCGURR0dHXL16FRwOBy4uLvyzwUWFxWLhyJEjqFOnDho2bIj4+HiMGTNGpH3ICiMjI+zZswdHjhzB3r17Jdbv6NGjoaamhoCAAIn1KU5FG3JEPeUt6AiloaEhAJRb4Ly2j1K+evUK3t7e0NTUxL59+9C4cWOEh4eDw+Fg4MCBEj3NRtRCQ0Oho6MjtWLqFFUZNKGkBFKVhBIAOnbsiBs3buDly5dwcnJCRkaGCKMDGjVqhOPHjyMjIwP169eXeEJVnbi7u8Pb2xvTp0/H06dPJdKnsrIyvL29ERgYWOqomqzhcDho1KgRmjZtKpL2KjNCCZSeUGppaUFLS6vGrVkVBI/H46+DNDY2xoEDB1BQUIDp06fj/fv36Nq1q7RDrDIej4fQ0FAMGDCgWpUwoqiK0ISSEoicnFyVSwC1b98e4eHhSEpKgoODA758+SKi6P7o0qULVqxYgR8/fkBeXh6TJk3Cs2fPRNqHrNi4cSNMTEwwZMgQkY8Il2XKlCkoKCjAnj17JNKfOBUVNBfVL3RhN+WoqqpCU1OTrqP8f1lZWdixYwdat26NXr16ITU1FVpaWmAymTh9+jQCAgLAYrGkHaZIPH78GGlpaXT9JCVzaEJJCaSqI5RF2rZti4iICHz58gVdu3ZFWlqaCKL7n9mzZ6NPnz5QUFAAj8dD165dSxxhVxsoKiri1KlT+PbtG8aPHy+R9ZSNGzfGyJEjsWXLFpkvNC/KHd6AcCflFKloY86TJ0/EUpKrOklKSsKsWbOgpaWFadOmwczMDOvWrUNqaiqUlJTw6NEjDBo0SNphilRISAgaNmyITp06STsUihIKTSgpgYgqoQSA1q1b4/bt28jKykKXLl3w/v17kbQL/NkgcvDgQaipqUFLSws/fvyAg4OD2E7tqc5atGiBAwcO4MyZM9i6datE+pw5cybS0tJw8uRJifQnDmlpafj06ZNI168xmUwwmUyhfobKSyjt7OyQk5OD58+fiyrEaoMQgtu3b2PQoEFo2bIlAgMDMXHiRLx58wbm5ub466+/0KlTJzx69AgmJibSDlekCCEICQlB3759a8yIK1V70ISSEogoE0rgzy/L27dvg8fjoUuXLnjz5o3I2m7QoAFOnjyJz58/Q1tbG1FRUZgxY4bI2pcl/fv3h6+vL2bNmoXIyEix92dqagoXFxds2LBBYrvMRU3UG3KKCLtspCihLO3P0cLCAnJycjVq2jsvLw8HDx7kF8t/8eIFtm/fjtTUVCxatAh+fn5YsGABFi1ahPPnz6N+/frSDlnk4uLikJiYSHd3UzKJJpSUQESdUAKAnp4ebt++DQUFBXTp0gWvXr0SWds2NjZYv349UlJSUL9+fWzbtg1nzpwRWfuyZPXq1bCyssLQoUORnp4u9v78/PwQHR2NW7duib0vcYiKioKGhobIz08WtvSWgYEBfvz4UerfmaKiIszNzWvExpxPnz5hyZIl0NHRwdixY9G4cWNcvnwZcXFxmDRpEj58+AAbGxtcu3YNZ8+exdKlS2WiMHllhIaGQlVVFd26dZN2KBQltJr5U0mJnDgSSuDPjtWIiAioqanB3t4eMTExImt72rRpGDRoEAoLC8Fms+Hu7i7SkVBZIScnhxMnTuD3798YPXq02Kf/nZycYGZmJrOFzosKmot6h62wP0Pl7fQGZH9jDofDwejRo6Gjo4MNGzZg8ODBePXqFS5evAgXFxcwmUyEhYWhffv2IITg0aNH6Nevn7TDFquQkBD07t0bCgoK0g6FooRGE0pKIOJKKIE/mzlu3bqFZs2aoWvXroiKihJJuwwGA4GBgdDU1IS2tjYKCgrQqVMnsZzYU91pa2sjODgYFy9exLp168TaF4PBgK+vLy5cuICXL1+KtS9xEPWGnCLCTnnr6+sDKDuhtLOzw5s3b0ReLUGcCgsLcfr0aXTq1AlWVla4ffs2Vq1axT/NxsjICMCf0jnLli2Dq6srHB0dERkZKVMn3VTGu3fvEB0dTXd3UzKLJpSUQMSZUAKAhoYGbty4AQMDAzg6OopsKq9evXo4deoU0tLSYGpqik+fPqFPnz4yu76vKnr27In58+djwYIFuH37tlj7cnd3R+PGjbFp0yax9iNqnz59QlpamlgSSmGnvFVUVNC0adNyRygBSGRtbFVlZGRg7dq1aNmyJYYMGQIWi8U/zcbPz6/YesjMzEwMGDAA/v7+WL58Oc6cOYO6detKL3gJCQ0NhYKCAnr27CntUCiqUmhCSQmkskcvCkNNTQ3Xrl1DmzZt0L17d5ElPebm5tiyZQtiY2Ohp6eH8PBw+Pv7i6RtWbN06VJ07twZbm5uYh3ZUlBQwLRp03Do0CF8/fpVbP2IWtHouDhOKKlMLdfydnrr6upCU1OzWk97v3z5Et7e3tDW1saiRYvg6OiIqKgoRERElHqazcuXL2FjY4OIiAj8888/WLhwYY1dL/lfoaGhcHZ2hqqqqrRDoahKqR0/qVSViXuEsoiqqiouXboEW1tb9OjRA9euXRNJuxMmTMCIESPw9etXqKqqYtmyZbh+/bpI2pYlbDYbR48eBY/Hw4gRI8R65rmXlxeYTCZ27Nghtj5EjcPhQE1NDXp6eiJvuzIfyspLKBkMBuzs7Krdxpyi02xcXFzQunVrnD17FnPmzMH79++xf/9+mJubl/rc2bNnYW1tDRaLhcePH6N3794Sjlx6Pn/+jHv37tWq3d08Hg9v3rzB2bNnsXz5cgwbNgzOzs789csWFhZwdnbGsGHDsHz5cpw9exZv3ryplSXgZAVNKCmBCJNQ8ng8fPnyBcnJyYiLi0NMTAySk5Px+fNngV4GKioq+Oeff+Do6Ig+ffogLCysquGDwWBg165d0NbWRqNGjcBkMuHq6ooPHz5UuW1Z06RJExw7dgw3b97E33//LbZ+1NXV4eHhge3btyMnJ0ds/YiSqE/I+bfKfCgrr3QQ8Gfa+9GjR2L9YCCorKwsbN++HcbGxujVqxfS09MRHByM5ORkLF68GJqamqU+x+VysWjRIgwYMAA9evRAZGQkf0NSbXHu3DkwGAy4urpKOxSxys/Px5kzZ9CnTx/UrVsX+vr6GDBgADZv3oxPnz5BTU2Nv3ZYX18fampq+PTpEzZv3owBAwZAX18fdevWhaurK0JCQmT+AIUah1CUACZPnkzatWtX4utcLpfcvXuXbNmyhUyYMIHY2tqSOnXqEACl/qeiokJsbGyIp6cnCQgIILdv3yaFhYWl9pmXl0cGDhxI2Gw2OXXqlEi+j9jYWKKkpETs7OwIANK8eXOSl5cnkrZlzfLlywmDwSBXr14VWx/x8fGEwWCQvXv3iq0PUdLW1iZ//fWXWNo2NTUl06dPF+qZkJAQAoB8+vSp1OsREREEAImOjhZFiJXy7t074ufnR+rVq0dYLBYZMmQIuXv3LuHxeBU++/37d9KrVy/CYDDI6tWrBXqmJnJxcSEODg7SDkNs4uPjiY+PD9HQ0CAAiI2NDVmzZg25fPky+fDhQ4V/7zwej3z48IFcvnyZrFmzhlhbWxMARENDg/j4+JD4+HgJfSdUeWhCSQlkxowZxMTEhP//b9++JYsXLyY6OjoEAJGXlydt27YlI0eOJGvWrCGhoaHk6tWr5OLFiyQsLIxcvXqVnD17lqxdu5aMGjWKtGvXjigoKBAAREtLiyxYsIAkJiaW6LegoIC4u7sTJpNJDh8+LJLv5eDBgwQA/6U0YMAAkbQra7hcLnFxcSENGzYkqampYuunX79+xNjYmHC5XLH1IQpfvnwhAMiJEyfE0n67du2It7e3UM/ExMQQAOTOnTulXs/KyiIsFovs2rVLFCEKjMfjkVu3bpEBAwYQJpNJ1NTUyF9//UWSk5MFbiMmJobo6+sTNTU1cvnyZTFGW719//6dyMnJkS1btkg7FJHLy8sjS5cuJfLy8qRhw4bEz8+PxMbGiqTtmJgY4uvrSxo2bEjk5eXJsmXLau3gQHVBE0pKILNmzSKGhoYkKiqKdOvWjQAgqqqqZMKECeTevXskPz9f6DYLCgrIgwcPyKRJk0jdunUJANK1a1fy6NGjYvcVFhYSDw8PwmAwyL59+0Ty/Xh6ehIlJSV+Qrxp0yaRtCtrvnz5Qpo1a0Y6depECgoKxNLH7du3CQBy4cIFsbQvKpcvXyYASv1gIwrt27cnEyZMEOqZ7OxsAoAEBQWVeY+5uTkZO3ZsVcMTSE5ODtm/fz9p164dAUCMjY3Jrl27SFZWllDtnDx5kqioqJA2bdqQN2/eiCla2XD48GECgLx//17aoYjU/fv3SevWrQmbzSbz5s0j2dnZYuknOzubzJs3j7BYLGJiYkLu378vln6oitGEkhLIrFmz+FNaZmZm5ODBg0L/EinP79+/yeHDh0m7du0Ik8kkM2fOJL9+/eJf53K5xNvbmwAg27Ztq3J/2dnZpE2bNqRFixZESUmJMBgMEhkZWeV2ZdHdu3cJi8Uic+bMEUv7PB6PtG/fnjg6OoqlfVFZsWIFqVevntimXe3s7CqV+Glra5N58+aVeX3y5MnEyMioKqFVKC0tjSxatIg0bNiQACC9evUiV65cEfrPqrCwkMyZM4cAIG5ubiJ9h8iqQYMGkfbt20s7DJEKDg4mTCaTtG/fnjx79kwifUZHRxMrKyvCZDLJkSNHJNInVRxNKKkK3b9/n6ipqREAZNWqVZUajRRUQUEBWbt2LVFSUiK6urokIiKCf43H4xFfX18CgKxbt67Kfb1+/ZqoqqoSBwcH/ojr169fq9yuLFq/fj0BQP755x+xtH/8+HECgDx9+lQs7YvCwIEDxbqOrUuXLmTEiBFCP+fo6EgGDx5c5vVDhw4RACQ9Pb0q4ZXq8ePHZOTIkUROTo6oqKiQKVOmkFevXlWqrfT0dNK9e3fCZDLJ+vXra+16yX/Lzs4mysrKZNWqVdIORWQOHDhAGAwG8fDwKHN9vLgUFhaSsWPHEiaTSQ4ePCjRvimaUFIVuHHjBlFWViY6OjpEXV1dYv0mJiaSLl26EEVFRXLlyhX+13k8Hlm4cCEBQJYvX17lfk6cOEEAEFdXVwKAmJqaSvwlWB3weDzSr18/oqamRpKSkkTefkFBAdHR0SGjRo0SeduioqenR2bNmiW29rt160aGDRsm9HNeXl6kbdu2ZV5PSEggAMjFixerEN3/FBQUkJMnT5KOHTsSAERPT49s2LCBfP/+vdJtRkdHk+bNmxN1dXVy/fp1kcRZE5w9e5YAqHSSXt3s37+fMBgMMmHCBKmtmeZyucTT05MwGAxy4MABqcRQW9GEkirT1atXiaKiInFxcSHLly8nDRo0kGj/OTk5pHfv3kRBQaHE+ru///6bACDz58+v8kjH1KlTiby8PH+Tzrhx46rUnqzKyMggenp6xNraWiyL2zds2EDYbLZYNwBV1rdv3wgAcvToUbH14eLiQgYOHCj0c+vXrycqKipl/jvn8XhEXV2dLF68uErxpaenk9WrVxNtbW0CgNjb25OQkJAqf8A6evQoUVJSIubm5uTdu3dVaqumGT16NGndurW0wxCJ2NhYIi8vTzw9PaW+AY/L5ZLx48cTeXl5EhcXJ9VYahOaUFKlevjwIVFQUCC9evUiOTk5ZMOGDURVVVXiceTm5pJ+/foReXn5EjtdN2zYQAAQHx+fKiWVubm5xMrKiujp6ZHGjRsTALV2uuTx48dEXl5e6PI2gsjMzCSqqqpk7ty5Im+7qq5evUoAkNevX4utjz59+hBXV1ehnzt37hwBQD58+FDmPb179ybdu3evVFxxcXHEy8uLKCkpEXl5eTJ27FiRLE0oKCggfn5+BAAZOXKk2DZlyKr8/HyipqZGFixYIO1QqqygoIBYWVkRY2NjkpOTI+1wCCF/lhMYGRkRa2trsW04pIqjCSVVwr9/EHNzcwkhhAQEBBBFRUWpxJOXl0c6duxIWrRoUWIR/44dOwgAMmnSpCp9Kn779i2pX78+cXJyInJycoTFYpGYmJiqhi6Ttm3bRgCIrPbnv/n6+pL69esX23BVHaxevZqoqqqKdWSlf//+pGfPnkI/9+LFCwKA3Lp1q8x7/v77b1KvXj2B4+dyueTChQvE2dmZACCNGzcmy5YtI58/fxY6vtJ8/fqVODo6EhaLRQICAuh6yVJcu3aNACAcDkfaoVTZqlWrCJPJrHYbGx88eECYTCZZvXq1tEOpFWhCSZUwe/ZsoqCgQF68eMH/2o4dOwiLxZJaTPHx8URJSanUkbPAwEDCYDDI2LFjqzQ9VzQSNGbMGAKAqKurkx8/flQlbJnE4/HI0KFDiaqqKklISBBp28nJyYTFYlW7mntDhgwhXbp0EXsfTk5OQj+Xm5tLmExmucXhr1+/TgBUOL3369cvsnXrVmJoaEgAECsrKxIcHCzSJQ4cDofo6OiQhg0blpsE13aTJ08murq6Mp9s//z5k6iqqpKZM2dKO5RSzZgxg6iqqla7D7E1EU0oqWI4HE6pn+j27t1LAEj15bdp0yYCgDx48KDEtaNHjxIWi0Xc3NyqtAt91qxZhM1mk8GDBxMAxM7OTurrgaQhMzOTGBgYkHbt2ol8qtLNzY20aNGiWm1+atGihdh/IQ4fPpx07dq1Us/q6emVe4LPz58/y63T+vbtW+Lr60vq1q1LWCwWGTp0KLl3757If54PHTpEFBUViZWVVY2rqyhKXC6XNGnShPj4+Eg7lCrbtWsXYTKZ1fbvOykpiTAYDLJnzx5ph1Lj0YSSKmbUqFGkefPmJdacHDhwgAAQa8mgihQWFhIjI6Myd8qeOXOGyMnJkQEDBvCn6oWVn59POnbsSLS0tEjbtm0JAOLn51eVsGXWs2fPiKKiIpk4caJI2338+DEBQM6cOSPSdisrIyODABDZSUxlGT16NOnYsWOlnnV2dq7wRCczMzPi6enJ//+i02z69+/PP81mzpw5YvnFn5+fT6ZPn04AEA8Pj2qzjq66evDgAQFQrCyaLOLxeKRt27akb9++0g6lXL179ybm5uYyPxpc3dGEkuL79u0bUVBQKHW9yZEjRwgA8vv3bylE9j+bNm0icnJyZZ5tHBYWRhQUFEjPnj0rPbKWkpJCNDQ0iJOTE6lfvz4BQM6ePVuVsGVWYGAgAUCCg4NF2m7nzp1Jhw4dRNpmZd24cYMAKLbEQxzGjx9PrK2tK/Xs5MmTiampabn3TJgwgZiampKcnBwSFBTE/0DUunVrsnv3brH97H769Il06dKFsNlssn37dvpLWwCzZ88mDRs2rFaj9JXx7NkzAoCEhYVJO5Ry/fPPPwRArV0XLylMUNT/O3jwIHg8HsaNG1fimpycHACgoKBA0mEVM3r0aLBYLAQFBZV6vXfv3ggLC8OtW7fQp08f/P79W+g+tLS0cOTIEdy4cQMjR44Ek8nE0KFD8ebNm6qGL3M8PDwwZswYeHl54cWLFyJr18/PD/fv38fDhw9F1mZlcTgcqKiowNDQUKz9sNlsFBYWVupZAwMDJCYmgsfjlXmPsbExYmNjoa2tjXHjxkFLSwtXr15FbGwsJk6cCGVl5cqGXqbHjx/DysoKr1+/Rnh4OCZPngwGgyHyfmoSQghCQ0PRr18/sFgsaYdTJY8ePQKTyUTXrl2lHUq5unbtCgaDgUePHkk7lBqNJpQU37lz59CrVy80bNiwxLXqklA2aNAAffv2xblz58q8x8nJCVeuXMGjR4/Qo0cP/Pz5U+h+unfvjoULF2LHjh2YMWMG8vPz0bVrV2RnZ1clfJnDYDCwfft2NG/eHIMHD65Ugl4aV1dX6OvrY+PGjSJpryqioqJgbm4u9l/ucnJylf75MTAwQG5uLj58+FDi2uPHjzFy5Ej89ddfAICOHTvi9evXCAsLg7Ozs9gSvKCgIHTu3BlaWlrgcDjo1KmTWPqpaWJjY5GYmIiBAwdKO5Qqe/r0KVq1agUVFRVph1KuOnXqwNDQEE+fPpV2KDUaTSgpAACPx8PTp09ha2tb6vXqklACgI2NDZ4/fw4ul1vmPZ07d8b169cRGxsLZ2dnfP/+Xeh+lixZAnt7exw7dgz9+/dHamoqBg4cCEJIVcKXOSoqKjh16hTev38Pb29vkXz/TCYTM2fOxJkzZ/Du3TsRRFl5HA4HFhYWYu9HTk6uSiOUAJCQkAAAKCwsxMmTJ9GxY0dYW1vj3r17WLNmDerVqwdLS0uxjrbm5+dj8uTJGD9+PMaMGYNbt26hWbNmYuuvpgkNDYWqqiocHR2lHUqVRUVFSeRnRxQsLCzA4XCkHUaNRhNKCgDw9u1b/Pr1C+bm5qVer04JpYWFBXJycvD69ety77OxscHNmzfx5s0bODo64uvXr0L1w2KxcPToUQDAjx8/YGBggCtXrmDlypWVjl1WGRsbY8+ePQgODkZgYKBI2hw7dizq16+PgIAAkbRXGZmZmUhISIClpaXY+2Kz2ZX++WnevDlYLBaePn2K1atXo3nz5hg2bBjk5eURGhqKxMRE+Pr6wtbWVqzLCD5+/AhHR0fs27cPu3fvxu7du6GgoCC2/mqikJAQ9OnTp0b8uaWlpUFPT0/aYQikefPmSEtLk3YYNRpNKCkAwLNnzwAA7dq1K/V6dUooi2KMjo6u8F5zc3PcunULHz9+RNeuXfHx40eh+mrcuDGOHTuG27dvo3fv3lBWVsaiRYtw69Yt4QOXccOHD8ekSZMwdepUgf7sK6KsrAxvb28EBgbix48fVW6vMoq+D0kklFWZ8o6Pj4eKigrmzp0Lf39/ODs7Izo6GuHh4ejfvz9/ur4ooRTHKPqDBw9gaWmJd+/eISIiAhMnThR5HzXd27dv8ezZMwwYMEDaoYiMrKyZlZU4ZRlNKCkA4K8NrF+/fqnXq1NCWRSjoOsZTU1NERERgczMTNjb2yMlJUWo/rp27Yrly5cjICAAixcvBiEEvXv3LnU9W023adMmtG7dGkOGDKnU2tT/mjJlCvLz87F3714RRCc8DocDJSUlGBkZib0vYTfl8Hg8XLhwAd27d4epqSny8vKgr6+PlJQUBAUFoW3btiWesbOzQ0ZGBn9qXFT27NkDe3t7tGjRAhwOB3Z2diJtv7YIDQ2FgoICevbsKe1QKErkaEJJCYTNZgOoHgllZRgZGeH27dsoKChAly5dhF63N3fuXLi4uGD9+vWYNWsWsrOz4eDggLy8PDFFXD0pKiri1KlT+PLlC8aPH1/lkbAmTZpg+PDh2LJli1T+bXE4HLRr147/71ucBB2h/PXrF7Zu3YpWrVqhT58++PHjBw4fPozx48eDwWCUummuiLW1NQCIbNo7Ly8PEydOhJeXFyZOnIibN2+icePGImm7NgoNDUX37t1Rp04daYciEgwGQ2Z+JxQUFNBRSjGjCSUlkOo0QllZLVq0wO3bt8Fms9G5c2fEx8cL/CyTyURwcDAUFRXx4MEDdOvWDQkJCRgzZowYI66eWrZsif379+P06dPYtm1bldvz9fVFamoqTp48KYLohBMVFSWR6W6g4k05b9++ha+vL7S0tDBz5kyYm5vj/v37iIyMxIgRI9CqVSu8efOm3M1oampqMDY2xoMHD6oc74cPH2Bvb49Dhw4hKCgI27Ztg7y8fJXbra0+ffqE+/fv14jd3UX09fXx6tUraYchkFevXvE3t1HiQRNKCgD4NeoyMzNLvV6dEsqiGCtTV09bWxu3b99GvXr10KVLF8TFxQn8rIaGBk6ePInIyEiYmpqiSZMmOHHiBHbs2CF0HLJu4MCB8PHxgZ+fX5Vru5mZmaF79+7YuHGjRHfQ//r1C69fv5bYLtXSNuUQQnDr1i30798f+vr6OHjwICZPnox3797hxIkTsLOz44+qGBgYID8/v8IlG6LYmHP37l1YWlriw4cPuHPnDjw8PKrUHvWnLBuTyYSrq6u0QxEZCwsLmSnFI0s70mUVTSgpAECbNm0AlL3RpSihrGzZE1EqirG0NWSCaNKkCW7duoXGjRuja9euQr0Q7ezssHbtWv56Sjk5OUybNg1PnjypVCyybM2aNbCwsMDQoUORkZFRpbZ8fX0RFRWFiIgIEUVXsejoaBBCJDpCWZRQ5ubmIigoCO3atYODgwMSEhKwa9cupKSkYNWqVdDW1i7x/H9LB5XFzs4Oz58/r1TNUEIItm/fDgcHB7Rq1QocDgft27cXuh2qpJCQENjb20NdXV3aoYiMhYUFkpOTkZ6eLu1QyvX161ekpqbShFLMaEJJAfgzjamqqoqoqKhSr1enEcqoqKgqb6Ro2LAhbt68iebNm8PR0RGRkZECP+vj44P+/ftj3rx52Lx5M3g8HpydnYUuSyTr5OXlcfLkSfz69Qtjxowp9xSXihRtPNmwYYMIIyxfVFQUFBUV0bp1a4n0VzRCuWjRImhra2P8+PHQ0dHBtWvXBDrNRldXF2w2u8KE0tbWFjweD48fPxYqvtzcXIwbNw5Tp07FlClTcO3aNTRq1EioNqjS/fjxAzdv3qxRu7uBP0X0GQwGzp8/L+1QynXu3DkwGAx06NBB2qHUaDShpAD8WSNobm5eZmJVnRLKyMhItGnTpsobKRo0aIDr16/DxMQEzs7OuHPnjkDPMRgM7N+/H2pqajhw4AC8vb3x48cPuLi4VIsRXEnS0dFBcHAwwsLCqpQMMhgM+Pr6IiwsrML6oqLC4XBE8u9IEI8ePcLBgwfB5XKxefNmuLu7Iz4+Hv/88w+cnJwE2izAZrPRokWLChPK1q1bQ1VVVahp75SUFHTu3BnHjx9HcHAwNm/ezP+Zp6ouLCwMhYWF6N+/v7RDESltbW24uLhg165dImnvd14h4tIy8fT9d8SlZeJ3nmjep7t27ULPnj2hpaUlkvao0tGEkuLr27cvLl68WOpIW2USSnG8HL5//45z586hb9++VW4LAOrWrYvLly/DysoKPXr0wI0bNwR6rn79+jh16hSePXsGFosFKysrPH36FNOmTRNJXLKkV69emDdvHubNmydwUl6a4cOHQ1NTE5s2bRJhdGXjcDhine4uKCjAiRMn0KFDB9jY2PArCyQlJWHLli2V2iBgYGBQYULJYrFgbW0t8MacW7duwdLSEl+/fsW9e/cwcuRIoeOiyhcaGgpra+samdB4e3vj0aNHZc5uVSTh8y/4n4+D/bpwmPpfQe+tdzFg53303noXpv5XYL8uHP7n45Dw+Vel2n/8+DE4HA68vb0r9TwlOJpQUnxjxozhj779l6AJpbhfDocOHUJhYSHGjx9fqedLU6dOHVy4cAFdunRB7969cfHiRYGes7S0xObNm7Ft2zZ4e3tDTU0Nu3btwpEjR0QWm6xYtmwZOnbsCDc3N3z58qVSbSgoKGDq1Kk4ePCg2JcP/P79G69evRJLQpmeno7Vq1ejRYsWcHNzg6KiIs6ePYv169cDqNxmsiKCJJSAYAXOCSEICAiAk5MT2rRpgydPntA1ZmKQnZ2NS5cu1ajd3f/Wq1cvaGtrY+nSpUJtqkvJyMaowEg4b76N4MhkJGdk479PEwDJGdkIjkyG8+bbGBUYiZQMweoPA3/+jS9duhQ6Ojq09qcE0ISS4tPQ0MCQIUOwe/fuElO3FSWUkng58Hg87Nq1CwMHDoSmpqYw31qFlJSUcPbsWfTo0QP9+/dHaGioQM9NmjQJbm5u8PHxwb59+8BkMjF27Fi8ePFCpPFVd2w2G8eOHUNhYSFGjBhRbmmb8nh7e4PBYGDnzp0ijrC4Z8+egcfjiTSBiouLw8SJE6GtrQ1/f390794dz549w82bN9GvXz/+UXtVWTZiYGCAt2/fVri0ws7ODl++fEFSUlKp17OzszFq1Cj4+Phg5syZuHz5MjQ0NCodF1W2q1evIicnp8atnyzCZrMREBCA8+fP84+qrcjxx+/htCkC99/+2czD5ZWfiBZdv/82HU6bInD88XuB+jl8+DAuXLiALVu28E+TosSHJpRUMT4+Pnj37l2JacfyEkpJvRy2bduGV69ewcfHR6D7haWgoIBTp05h4MCBGDJkCI4dO1bhMwwGA3v27EGTJk2wdOlSbNq0CYWFhXB0dCyzBFNN1bRpUxw9ehQ3btzAihUrKtWGuro6xo4di+3btyM3N1fEEf4Ph8OBvLw8TExMqtQOj8dDWFgYnJ2dYWpqirCwMCxYsAApKSkIDAzkV08A/nc4QFXW2RoYGKCwsBDJycnl3mdjYwMApU57JyUloWPHjggJCcGxY8ewbt06iawjra1CQkJgYmICQ0NDaYciNgMGDIC7uzumTZuGT58+lXvvtvAEzA2JQV4hr8LfFf/F5RHkFfIwNyQG28LLH6n/+PEjpk+fjhEjRqBfv35C9UNVDk0oqWIsLS3h6+uLRYsWFStYW9ZJOZJ6OSQmJmLu3LmYOnWqWI99k5OTw5EjRzBy5EiMGDECBw4cqPAZVVVVnD59GvHx8YiJiYGbmxs+f/6Mvn37Vmnnsyzq1q0b/P394e/vL/B61P/y8fHB169fxbp0ICoqCm3atKl0oe5fv35hy5YtMDIygqurK37+/IkjR44gKSkJCxYsKPU0G1FsbBO0dJCGhgb09fVLbMy5fv06rKyskJmZiQcPHsDNza3SsVAVKygowD///FNjRyf/bcuWLZCTk8PAgQPLPJb1+OP3WH9V8AMlyrP+ajxOlDEYkZmZiYEDB0JBQQEBAQEi6Y+qGE0oqRKWL18OXV1djB07Fvn5+QBK/2UoqZdDQUEBxo0bh8aNG2PVqlUi6a88LBYLQUFBmDhxIjw8PASafjUzM8OOHTuwb98+ODk5wdDQELdv38aCBQvEHm91s3DhQjg7O2P48OFIS0sT+nlDQ0O4urqKtdA5h8Op1HT327dvMXPmTGhpacHX1xeWlpZ48OABIiMjMXz48HITVFHUctXW1oa8vLxA6yjt7Oz4I5SEEKxfvx4uLi6wtLTEkydPKl3HlRLcrVu38OPHjxq7fvLfNDQ0EBYWhpcvX6J79+748eNHsespGdlYcl7wgyQEsfh8XIllUz9+/ICzszNev36NsLCwGlX3s7qjCSVVgpKSEg4ePIjo6GgMGjQIeXl5YDKZYDKZ/IRSUi+H/Px8DBs2DA8fPsSBAwckdgYuk8nEzp07MWPGDEyePFmgncceHh4YO3Yspk+fju3bt0NZWRmrV69GWFiYBCKuPphMJg4fPgw5OTm4u7tXKoHy8/PDixcvcOXKFZHHl5OTgxcvXgi8IYcQgvDwcP5pNocOHcKUKVOQlJSE48ePw9bWVqB2yhrlFwaLxULLli0F3pgTHR2Nb9++wd3dHbNnz8acOXNw8eJFNGjQoNIxUIILDQ2Frq4u2rVrJ+1QJKJ9+/a4ceMG4uPj4ezsjI8fP/KvzQ+NQaGQs1gVKeQRzA+N4f//x48f4eTkhDdv3uDGjRuwsrISaX9U+WhCSZXK1tYW58+fx/Xr1zFgwADk5uYWO+lDEi+HvLw8DB48GBcuXEBoaCi6dOki0v4qwmAwsGnTJsybNw++vr5YuXJlhc9s374dLVq0wPTp03H8+HEAwODBg/HmzRtxh1utNGzYEMePH8e9e/ewaNEioZ/v3LkzLC0txVLo/Pnz5+ByuRUmlDk5OQgMDES7du3g6OiIxMRE7N69GykpKVi5cqXQJWBEVctV0J3ednZ2KCwshK2tLcLCwnDq1CmsXLmSbk6QEB6Ph7Nnz2LgwIEC1RmtKSwsLBAeHo6UlBQYGxtj7969eP0pE3cSvwm9LKoiXB7BncRviP/0E3v27IGxsTE+fPiAmzdvwtzcXKR9URWjCSVVpu7duyMsLAy3bt2Cs7MzWCwWCgoKkPD5l1hfDolffuHdu3fo0aMHrl69inPnzqF3794i7UtQDAYDK1aswLJly7BgwQIsWrSo3GlYZWVlnD59GikpKThx4gT8/f2Rl5eHbt26ITtb8B3tNUGnTp2watUqrF69GhcuXBDqWQaDAT8/P1y/fh3Pnj0TaVwcDgdycnIwNTUt9XpaWhoWLlwIHR0dTJgwATo6Orh+/TpiYmIwYcKESpf9EdXxpYImlEWjQ5mZmYiMjMTgwYOr1C8lnMjISHz8+LFWrJ/8r7Zt2+LFixcYOHAgJk6ciMFzA8AUU07NZAAD52yCl5cXBg0ahLi4OLqcQ0poQkmVq1u3brh27Ro+fPiA7OxsXL9+HYceJIElprcDi8mA786zMDU1xdu3b3HlyhX06NFDLH0JisFgYNGiRVi7di3+/vtvzJ49u9yk0sjICHv27MGRI0fQpEkTODs7Izk5GcOGDRPbmsDqys/PD66urhg1alSFO5P/a/DgwdDW1sbGjRtFGhOHw4GpqSm/jE+RR48eYfjw4dDV1UVAQACGDx/OP82mW7duVR5lEsWUN/AnoUxKSiqzHUIIVq1ahT59+qBBgwaws7Or8m52SnghISFo1KhRrT3ur0GDBggKCsKNGzfwS1UHIh5/4OMRIEtVFzdv3kRgYCBdziFFNKGkKtSxY0fExMRASUkJly9fxpGbT0U+OlmEyyPgpOViwoQJiIuLg729vVj6qYzZs2dj69at2LBhA6ZOnVruDm53d3d4e3tj+vTpWLJkCZo2bYqwsDCsW7dOghFLH5PJxIEDB1C3bl0MGzaMv8lLEHJycpgxYwaOHTtWqc09ZYmKiuJPdxcUFOD48eOws7ODjY0NIiMjsW7dOqSmpiIgIAD6+voi61dUI5SGhobgcrn8k3f+7devXxgyZAjmz5+PhQsXwtPTExwOp9Z9kJE2QghCQ0PRr1+/Wr/EwLpjFzDqlKx6IEoM1Yaw7tBZrH1QFaMJJSUQFRUVNGjQAGM9J4GnLN5PgHINmmLFmvUS24AjjKlTp2Lv3r3YuXMnJkyYUG4B740bN8LExARjxozB2bNnIScnh7lz5+LWrVuSC7gaaNCgAU6ePImoqCjMmTNHqGc9PT2hqKiIbdu2iSSW3NxcxMbGwtDQEKtWrULz5s3h7u4OZWVlnDt3DvHx8fDx8UG9evVE0t+/iXKEEihZOighIQG2tra4evUqQkNDsWzZMtjZ2SEtLQ2pqalV6pMSTkxMDN68eVMrdndXJDn9d4lDLkSNAEhK/y3mXqiK0ISSEpicnBwUNJoBElhgXp1fDp6enjh06BAOHDiA0aNHlznipKioiFOnTuHbt29Ys2YNgoODQQiBq6srPnz4IOGopcva2hobN27E5s2bERISIvBz9erVg6enJ3bt2oXfv6v+b+Ls2bMoLCzEokWLsHTpUvTo0QPPnj3DjRs30LdvX7GOJolqU07Tpk2hpKSE+Pj/ley6cOEC2rdvDy6Xi8jISPTv3x8A+DvQ/1uPkhKv0NBQ1K1bF46OjtIOReryCyVTi1dS/VBlowklJTA5OTn6cvh/I0eOxIkTJ3Dy5Em4ubmVOZXbokUL7N+/H2fOnMHnz58xZcoUZGVloXv37sjLy5Nw1NI1ZcoUDBkyBB4eHkhMTBT4uRkzZiAzM1OgIvOl4fF4+Oeff+Dk5AR3d3cAwPz585Gamop9+/YVO81GnEQ15c1kMqGvr4+EhATweDwsW7YMrq6usLe3x6NHj2BsbMy/t3HjxtDT0yv1xBxKfEJCQtCnT59KF86vSeTZkkkzJNUPVTb6N0AJTE5ODrxCwdfAVYUsvBwGDx6MkJAQ/PPPPxg4cGCZRwUOGDAAM2fOxKxZs+Du7g4rKyu8ePECnp6eEo5YuhgMBvbt24dGjRphyJAhAh+tqKuri8GDB2PTpk1CnRH+8+dPbNmyBYaGhujbty+ysrLg4OAAMzMzLF68WOJnV4tqyhv4M+398uVLDBw4EP7+/li6dCl/VOy/bG1t6QilBL158wbPnz+vlbu7S6OnrgJxz2kx/r8fSrqq/29tqtpgs9lQyP9JXw7/4urqin/++Qc3b95E3759yywNtHr1alhaWmL48OE4cuQI1NTUcPjwYezevVvCEUtX3bp1cfr0abx8+VKoM9n9/Pzw5s0bnD9/vsJ737x5Ax8fH2hpacHPzw/t27fHgwcP8PDhQ2RmZkqt2LGoRiiBP+tS7969i/DwcJw/fx6LFi0Ck1n669zOzg4cDqfWjYhLS2hoKBQVFaVenaK6UFFgQ6dB5UptCUpHXRkqCvQ8emmjCSUlMDk5OZCCXPpy+I/u3bvj0qVLuH//Pnr27Ilfv36VuEdeXh4nTpxAVlYW/Pz8cPHiRTCZTEyZMgVPnjyRQtTS07ZtW2zbtg27d+8W+Lxua2trdOrUqcxC54QQ3Lx5E/369YOBgQEOHz6MadOmISkpCceOHYOtrS3y8/MRExMj8Ak5oiaqEcqzZ8/i8OHDKCwsxN27d9GnT59y7y/63qOjo6vULyWY0NBQdO/evVpuKpQWB6NGYi0152DYSCxtU8KhCSUlsKKTcujLoSR7e3tcu3YN0dHRpZ5jCwA6Ojo4fPgwwsLCcPv2bWzZsgVcLhcuLi74+vWr5IOWovHjx2PUqFHw8vLCy5cvBXrGz88P9+7dQ2RkJP9rOTk52LdvH9q2bYtu3brh7du32LNnD1JSUrBixQo0a9aMf29sbCwKCgoqdYa3KFR1Uw6Px8PixYsxYMAA2NjYAIBAm4jatWsHBQUFuo5SAj5+/Ij79+/T3d3/McJGR6yl5kba6oilbUo4NKGkBFaUUNKXQ+ns7Oxw8+ZNxMfHw9HREd++fStxT8+ePTF//nzMnz8fZmZmGDZsGDIyMtC7d2+RTIXKCgaDgZ07d0JXVxdDhgwRaAe3q6srWrZsiY0bN+LDhw9YsGABtLW1MXHiROjp6eHGjRt4/vw5PD09oaSkVOJ5DocDJpMptVM0qjLl/ePHD7i6uuLvv//GypUrcfToUQAlSweVRl5eHpaWlnQdpQScO3cOLBYLrq6u0g6lWjHQVEVnfQ2RD0SwmAx01teAfiNVkbZLVQ5NKCmBFSWU9OVQNktLS4SHhyM1NRUODg74/PlziXuWLl2KTp06wd3dHWvXroWhoSEeP34MX19fKUQsPSoqKjh9+jTevXuHyZMnV1h8m8ViYcCAATh58iT09PSwdetWjBw5EvHx8Th//jwcHR3LPc0mKioKrVu3rvTRiVVV2SnvuLg4tG/fHvfv38fFixcxb948NGnSBHXq1BEooQToxhxJCQkJQdeuXelpLaVYOcAMbBH/zmAzGVg5wEykbVKVRxNKSmBycnL80RX6cihbmzZtcPv2bWRkZMDe3r5EzUk2m41jx46hsLAQ48ePx6VLl6CsrIytW7fixIkTUopaOoyNjbF7924cOnQIQUFBpd5TUFDAXwe5fv16MJlMdOzYEampqdi8ebPAp9lwOBypTXcDlRuhPH36NGxsbKCkpIQnT57wN3owGAx+6SBB2NnZITk5mX++NyV6379/R3h4ON3dXQbtBspY2le0R4Au62sCbTGv6acERxNKSmBFI5QAfTlUpFWrVrh9+zZycnLQpUsXJCUlFbvepEkTHDt2DDdu3EBwcDBCQ0MBAKNGjUJcXJwUIpaekSNHYuLEiZg6dSqePXvG//q3b9+wcuVKNG/eHMOHD0edOnVw/vx5zJ49G1FRUUIdJ1hQUIDnz59LbUMO8L/1joKMUHK5XMybNw9DhgxBnz598ODBA7Rs2bLYPQYGBkKNUAK0wLk4hYWFobCwkF9UnirJrb0OZnU3FElbs7sbYVh72VweVVPRhJIS2L8TSoC+HCrSsmVL3L59GwwGA126dCnxy9/R0RFLly7F0qVLwWQysWTJEhQUFJS5qacmCwgIQKtWrTBkyBA8ePAAnp6e0NbWxvLly9GzZ088f/4c169fh6urK6ZPn47c3Fzs3btX4PZfvHiBvLw8qY5QMhgMsNnsChPKojW1a9euxbp163Ds2DGoqJQsoyVMQqmlpQUtLS26MUeMQkNDYWNjU2wjGFXSVAcDrB5oBgU2U+hlUywmAwpsJtYMNMMUB8FmJijJoQklJbD/JpQAfTlURFdXFxEREVBRUYG9vT1evHhR7PqCBQvg7OyM4cOHw9PTE05OTkhLS8OgQYPA41Xv04JESU5ODpMmTcLbt2/RoUMHXLp0CYsWLUJKSgr27t0LM7P/LYVo2rQphg8fjoCAAIHXI3I4HDAYDLRr105M34Fg/r1spDTPnz+HlZUVHj9+jCtXrmDWrFllrgs1MDBAampqmbVP/4uuoxSf7OxsXL58me7uFpBbex1cn2mPDi3UAaDC3x1F1zu0UMf1mfY1bvChpqAJJSWw0hJKgL4cKtKsWTNERERAQ0MDXbt2LTaty2QycfjwYcjLy2P48OE4fvw4mjZtips3b2LJkiVSjFoyfv78iYCAABgaGmLSpEn8ad05c+Zg/vz5ZZ5m4+vri9TUVJw+fVqgfjgcDlq1aiX12oDljVAeP34cdnZ2qFevHjgcDpycnMpty8DAAMCfQu6CsLOzw5MnT0RyUg9V3JUrV5CTk0PXTwpBu4Eygsfb4JpPF4yy0YWuunKJQzMYAHTVlTHKRhfXZ3ZB8HibGrMsqkYiFCWgsWPHkg4dOpR7T/ynn2TJuVjSZd1Nojc3jOj+6z+9uWGky7qbZMm5WJLw+aeEoq4+vn37RiwtLYmamhp59OhRsWt37twhLBaLzJ07l8TGxhI5OTkCgISFhUkpWvFKSEgg06dPJ6qqqoTNZhN3d3fy8OFDQgghM2bMIHJyciX+jP7LycmJWFpaEh6PV2F/tra2ZMSIESKJvSoaNGhAVq9eXexrBQUFZNasWQQAGTFiBPn9+7dAbX3+/JkAIGfOnBHo/nv37hEA5MmTJ0LHTZVv5MiRxNTUVNphyLys3AIS++EHiUrOILEffpCs3AJph0QJgSaUlMA8PT1J+/btBb6fvhxK+vHjB7GzsyN169Yld+/eLXZt7dq1/CTy6NGjBABRUlIiiYmJUopWtHg8Hrl+/TpxdXUlDAaDqKurkwULFpDU1NRi9+Xl5RFra2uip6dHMjIyymzv0qVLBAC5detWuf0WFBQQJSUlsnHjRpF8H1XRqFEjsnz5cv7/f/36lXTr1o2wWCyyefNmgZLjIjwej9StW7dEglqWnJwcIicnR7Zt2yZ03FTZ8vLySP369cmiRYukHQpFSRWd8qYEVtaUd1lUFNgwaVoP5jpqMGlaT6aOUxSXevXq4cqVKzA3N4eLiwvCw8P512bNmgVXV1eMGjUKHTp0wOTJk5GTk4Pu3bsLVPi7uio6zaZNmzZwcnJCUlIS9u7di5SUFPz9998lNjHIy8vj5MmTyMzMxJgxY8rcze3i4oLWrVtj48aN5fb/8uVL5OTkSHWHd5F//wxFRUXBysqKv+FoxowZ5dbR/C8GgyHUxhxFRUWYm5vTjTkiduvWLfz48YOun6RqPZpQUgITNqGkSqeqqoqLFy+iY8eO6NWrFy5fvgzgT4Jw8OBB1KtXD8OGDcPatWthaWmJd+/eYeTIkUKVyakOUlNTMX/+fP5pNi1atMDNmzfx7NkzjB8/vtTTbIro6uri0KFD+Oeff8o8v5vBYMDX1xf//PMP4uPjy2wrKioKAKS+IQf436ac4OBgdOzYERoaGnjy5Am6du1aqfaESSgBujFHHEJDQ6Gnpye1E5goqrqgCSUlMJpQio6ysjLOnz8PZ2dn9OvXD+fOnQMAqKmp4eTJk4iKisLChQtx4cIF1K9fH2fPnq1wJK66ePjwIdzd3dG8eXNs27YNo0aNQkJCAs6dOwcHBweBR+H69OmDOXPmYO7cubh3716p94wYMQINGzbEpk2bymyHw+HA0NAQdevWrdT3I0osFgtXrlzB6NGj4ebmhjt37kBHp/Kb0oRNKO3s7PDmzZtad3a8uPB4PJw9exYDBw4UanSZomokac+5U7Jj7ty5pHnz5tIOo0bJy8sjgwcPJmw2m5w4cYL/9a1btxIA5PTp0+T+/fuEyWQSBoNR4XpBacnLyyNHjhwh1tbWBABp2bIlCQgIIJmZmVVqt6CggHTu3Jk0a9aMfPnypdR7li1bRpSUlMjXr19Lvd6hQwfi7u5epThE4fPnz0RZWZkwmUyyfft2odZLluXQoUMEAPn165dA9797944AIOfPn69y39T/NjrduXNH2qFQlNTREUpKYBXV0KOEJy8vj2PHjsHNzQ3u7u44dOgQAGDKlCkYMmQIxo0bh4YNG2Lz5s0ghKBv375ITU2VctT/8/XrV6xYsQLNmzfHiBEjoKqqyp+Cnj59epVHBYuOqczPz8fIkSNLrc3p7e0NQgh27dpV4hqXy0V0dLRUC5oDwOPHj2FpaYn8/Hz0798fkydPFsmIVlHpoMTERIHu19XVhaamJp32FpGQkBBoamrCzs5O2qFQlNTRXRKUwOiUt3iw2WwcPHgQioqKGDt2LHJzczFx4kTs27cPlpaWGDJkCO7fv487d+7g9OnT6NWrFx4/fgwFBQWpxRwTE4OAgAAcPnwYDAYDo0aNwvTp02Fqairyvpo1a4ajR4+ie/fuWLlyJRYuXFjsuoaGBsaMGYNt27Zh1qxZUFRU5F97/fo1srOzpboh58CBA5g0aRLatWuH+vXro2HDhiJruyihTEhIEGiNKIPBgJ2dHd2YIwKEEISGhqJfv378YzWpmosQgk+fPiEmJgYxMTGIi4vD9+/f8eXLF9y/fx/dunWDqqoqNDQ0YGpqCjMzM5iZmYn0573ak+4AKSVLVq1aRdTV1aUdRo3F5XLJ1KlTCQASEBBACCEkOjqaKCgoEC8vL/L7929iYGBAGAwGGTNmjMTjKywsJGfPniUODg4EAGnWrBlZuXJlmVPNoubv70+YTCa5ceNGiWuvXr0iAEhgYGCxrxdNCX///l0iMf5bXl4emTx5MgFAPD09SW5uLrGxsSHjx48XaT9qampkxYoVAt+/evVqoqKiQgoLC0UaR20THR1NAJDLly9LOxRKTAoKCsiFCxfI4MGDibq6OgHAL+dmZWVFevXqRZydnQkA0rt3b9KzZ09ibm5OFBQU+PdqamqSESNGkBs3bhAulyvtb0msaEJJCWz9+vWkbt260g6jRuPxeGT27NkEAL++4N69ewkAcvjwYfLmzRuirKxMAJDdu3dLJKbMzEyyadMm0qJFCwKA2NnZkePHj5P8/HyJ9F+ksLCQODk5kUaNGpG0tLQS111dXYmJiUmxtYk+Pj6kZcuWkgyTEELIx48fSceOHYmcnFyxv6eOHTuS0aNHi7Qva2trMnbsWIHvj4iIIADIs2fPRBpHbbNkyRJSr149kpeXJ+1QKBF79+4dmTt3LmnatCkBQMzMzMiSJUtISEgISUhIqPDDWEFBAXn58iU5deoUWbBgATE0NCQAiJ6eHvH39ycfPnyQ0HciWTShpAQWEBBAlJSUpB1Gjcfj8ciSJUsIALJkyRLC5XLJqFGjiIqKCnnx4gW5ePEiAUBYLBaJjIwUWxzx8fFk2rRppE6dOoTNZpPhw4eLtT9BfP78mTRp0oTY29uTgoLihfLDw8NLjBh17tyZDB06VKIxPnjwgDRt2pQ0adKE3L9/v9i1rl27inyD0IgRI0jHjh0Fvj8rK4uwWCyya9cukcZR25iZmVWL05co0cnPzyerVq0iioqKpH79+mTy5MnkyZMnVd5Ax+PxyL1798j48eNJnTp1iKqqKtm2bVuNG7GkCSUlsB07dhA2my3tMGqNVatWEQDkr7/+Ir9+/SKtW7cmJiYmJCsriyxatIgAIBoaGuTz588i65PH45Fr166RPn36EAaDQTQ0NMiCBQuq1SfqiIgIwmKxyPz584t9ncfjEQsLC+Ls7EwI+bOEoE6dOmTNmjUSi23Pnj1ETk6OdOjQodRRVCcnJzJkyBCR9unv708a/R97Zx4P1f/98deMsRMhS4kS7T5ljQqlfY9CSqGN9tC+a/tooaR9o7RSpn0vSTtDRbQnS4s1iqzz/v3Rz3w/ImbGjLHc5+PRH9173+d97pjldc/7fc5RVeVpjIGBAU9RTYqKvH37llOFgaJx8PTpU9KtWzdCp9OJl5cX+fnzp1DmycnJIW5ubpzVnvj4eKHMIwooQUnBNfv37ycABFLuhII7tm/fTgCQOXPmkPj4eCIjI0MmTZpESktLibW1NaHRaMTMzKxStI5X8vPzyf79+0mXLl0IAPLPP/+QQ4cOkYKCAgHdiWDx8fEhAMiVK1cqHD9+/DgBQF68eMHZV3nz5k2h+1NYWEimT59OAJAZM2b8dRl0yJAhZPTo0QKdu/yeeSnRNHPmTNKhQweB+tGU2Lx5M5GSkhKa6KCoW06cOEHExMSIgYFBnfW6j4iIIB06dCCSkpLk8uXLdTKnsKEEJQXXBAYGEgB1vneuqbN3714CgEybNo0cOXKEk3ySnZ1NNDQ0CAAyb948vmynpKSQJUuWECUlJUKj0cioUaPInTt36v1DQ1lZGRk+fDhRUlIiycnJnOPFxcVEU1OTuLi4cIRWVlaWUH1JS0sjZmZmREJColJS0J+MHDmSDB8+XKDzP336lADg6YewPFlJ2K9NY8Xc3JyMGjVK1G5QCICjR48SOp1OnJ2da/1gziu/fv0io0aNIhISEo2iNiwlKCm45tixYwQAyc/PF7UrTY6goCBCp9PJxIkTydSpU4mUlBR5/vw5ef78OREXFycAyMmTJ7myxWazycOHD4mDgwMRExMjzZo1Ix4eHuT9+/dCvgvBkpWVRbS0tIiZmVmFiODmzZuJuLg4mT59utAL8UdGRhI1NTXSqlUrrvaX2trakkGDBgnUh5ycHJ7+/oT8b8n26tWrAvWlKZCWlkYAkCNHjojaFYpaEhQURGg0GpkyZYrI9jMWFRURW1tbIi4uTs6dOycSHwQFJSgpuOb06dMEAPn+/buoXWmSnDp1ioiJiRFbW1vSrVs3oqenR3Jzc0lwcDABQCQkJEhcXNxfx5d3szExMSEAiK6uLtmxYwfJy8urw7sQLI8fPybi4uLEw8ODcywnJ4fIyckRLS0tMmbMGKHMy2azya5duwiDwSCWlpbk69evXI1zcHAg1tbWAvdHRUWFrF27luvr2Ww2UVZWJqtWrRK4L42d3bt3EzExMSq628CJi4sj4uLiZPLkySJPjikuLia2trZETk6OJCUlidSX2kB1yqHgGnFxcQCgipuLCAcHB5w5cwYXL16EiooKvnz5gqlTp2LChAlwd3dHcXExhgwZgu/fv1cYl5GRgfXr16NNmzaYMGECFBQUcOnSJbx+/Rpz5syBvLy8aG5IAPTo0QNbt27Ftm3bwGQyAQCKioqYMmUKUlJSoK+vL/A5CwsLMWXKFMyaNQszZ87ErVu3oKamxtVYBoMhlG5TvPb0ptFoMDMzozrm8EFYWBj69u0LJSUlUbtCwSelpaVwdXWFnp4edu3aBTpdtFJIXFwchw8fhqKiIqZNmwZCiEj94RdKUFJwTbmgpNovio7Ro0fjwoULePDgAdq1a4fQ0FDs3r0b/v7+MDQ0RFpaGuzt7cFms/HixQtMmTIFrVu3xoYNGzBixAjEx8fj5s2bGDZsmMi/RAXFnDlzMGbMGLi6uuL9+/cAgJEjR4IQgszMTIHOlZKSAgsLC5w8eRJHjx6Fv78/53PBDcLqNsWroAQAc3NzPHnypMp2lhRVk52djbt378LGxkbUrlDUgi1btiAmJgaBgYEVOmuJEgUFBRw8eBA3b97EgQMHRO0Of4g6RErRcLh27RoBUCEJgkI03L59m8jIyJBWrVoRBoNBnj59StLS0oiioiKngC4AoqmpSf7991+SmZkpapeFyvfv30m7du2IoaEh+fXrFzl16hQBQNq2bSuwjjB3794lLVq0IFpaWoTFYvFlY9q0acTExEQg/vyXdevW8dzF6tatWwQAefnypcD9aayUJ8XVpzJaFLyRm5tLZGVliZeXl6hdqRIXFxeiqqraIAvmN44QBUWdQC151x+sra1x48YN5OXlQUpKCra2tggMDIS0tDQAICkpCYsXL8aHDx+wZMkSKCsri9hj4aKgoIAzZ87g5cuX8PT0BIvFgpqaGj5+/IhLly7VyjYhBDt27EC/fv2gr6+P6OhoGBoa8mWLwWAILUKZlZWFnJwcrseYmpqCRqNRy948wGQyYWZmhpYtW4raFQo+CQ4ORmFhITw9PUXtSpUsXLgQ6enpnC08DQlKUFJwDSUo6xe9evVCYGAgiouLkZqaihUrVsDa2hrz588HAAQEBODTp0+idbIO6d69OwICArBnzx5cvnwZPXv2RM+ePeHr68u3zV+/fsHZ2Rnz5s3D/Pnzcf36dbRo0YJve8Jc8gbA07K3vLw8unbtikePHgncn8ZIfn4+rl27BltbW1G7QsEnhBDs3bsXo0aNqrcPBZ07d4alpSX27t0rald4hhKUFFxDCcr6ASEEN2/exPDhw2FnZwcZGRlISkoCAHR1deHn54exY8fi169fGDp0KPLz80Xscd0xdepUODk5ISEhAa1bt4aXlxciIyMRFRXFs62kpCT06tULZ86cwYkTJ7B161YwGIxa+SfMpByAN0EJgErM4YHr16+jsLCQ2j/ZgImJiUF8fDzc3NxE7Uq1uLm54e7du0hKShK1KzxBCUoKrin/MaUEpWgoKCjA/v37oa+vj4EDByI1NRWHDh3Cly9fEBMTA1lZWXh7e+PChQsICgpCu3bt8O7dOzg7OzfYrEFeodFoWLRoEQDg/PnzGDBgANq1a8dzlPL27dswNjbG9+/f8ejRIzg6OgrEP2FFKOXl5aGmpsZXYs7Lly+Rm5srcJ8aG2FhYdDX14eurq6oXaHgkydPnkBcXBxWVlaidqVaBgwYAAB4+vSpiD3hDUpQUnANFaEUDSkpKViyZAlat24Nd3d36OnpITw8HLGxsXB1dYWUlBQ6d+6MqKgoSEpKYsyYMXj9+jWuXLkCKSkpnD17Ftu2bRP1bdQZr169AgCkp6dj7ty5mD9/Ps6cOcPV8j8hBL6+vhg4cCAMDQ0RFRWFbt26Ccw3cXFxoVVJ4CfT28zMDIQQviK4TYni4mJcunSJik42cGJiYtClSxfOik59pUWLFtDU1ERMTIyoXeEJSlBScA0lKOsOQggePnwIBwcHtG3bFnv27IGzszPevXsHJpOJPn36gEajVRjTqVMn3L17F4QQ9OrVCwBw+vRpAL83et+9e7eub0MkxMTEoFWrVti3bx+CgoIgJiYGeXl57Nixo9px+fn5GD9+PBYsWICFCxfi6tWrAk9mElZSDsCfoOzQoQMUFRWpZe8aCA8PR25uLrV/soETExPDd0JdXWNoaEgJSorGCyUohU9xcTGOHz8OU1NT9OrVC7Gxsdi+fTtSU1Ph5+cHHR2dasebmZlxshiNjIygp6eHZcuWgc1mY/To0UhNTa2jOxEdLBYLRkZGmDhxIqZOnQpPT0/Y2triwIEDf13a/fDhA3r27ImLFy8iJCQEPj4+EBMTE7hvwlryBv4nKHnZ3kCn09GjRw8qMacGmEwm2rZti3/++UfUrlDUgtTUVLRt21bUbnCFjo4OUlJSRO0GT1CCkoJrKEEpPNLT07Fu3Tq0adMGTk5OaN68OS5duoRXr15h9uzZPHWzGT9+PLy8vPDjxw+Ym5vDzs4Offv2xY8fPzB8+HAUFRUJ8U5ECyGEIygBYMeOHejQoQPu3r2LwsJCHDp0qNKY69evw9jYGPn5+Xj8+DHs7OyE5p+wknKA34Ly+/fvyMrK4mmcubk5Hj9+3GT22fJKWVkZzp07B1tb20qrAhQNj4bS0KGh+PlfGp7HFCKDEpSC5/nz55g8eTK0tLTw77//YuTIkXj58iVu3LhRq242mzZtgoWFBX7+/Alra2usWLECampqePHiBWbOnCngu6g/JCcnIzs7m7OsJS0tjdDQUGRkZKBly5bw9/fnCDpCCHx8fDBkyBCYmZkhKioKXbt2Fap/wo5QAvxlemdnZ/M8rqnw+PFjfPv2jdo/SUFRA5SgpOAaqvWiYCiPePTp0wfdu3fHzZs34e3tjdTUVOzduxedO3eu9RxiYmIIDQ2FsrIySktLYWtri40bN0JMTAyHDx9uuK29aoDFYgEAJ0IJ/BZahw4dwqdPn5CcnIwzZ87g58+fsLe3x9KlS7F8+XJcvHgRzZs3F7p/wkzKKc8+5lUYmpqaAgC1j/IvhIWFQV1dHebm5qJ2haKW0Gi0BvP7VVpa2uAi4pSgpOAaKkJZO3Jzc+Hn5wc9PT3Y2NigpKQEISEh+PjxIxYvXgwlJSWBzqempoaQkBD8/PkTzZo1w+zZs+Hl5QUAmDlzZoMrScENLBYLGhoa0NDQqHDczs4Oc+bMAY1Gw4oVK9CjRw9cu3YNYWFhWLdunVD2S1aFMJNyZGVl0bJlS54FZfPmzdGpUydKUFYBIQRMJhOjRo1qkEuQFBXR0dHBmzdvRO0GV7x+/Rrt2rUTtRs8QX1CKLiGEpT88ebNG8yZMwetWrXCkiVL0KtXL0RFReHBgwews7OrdbHs6rC0tMTGjRuRkpICPT097NixA0OGDEFZWRlGjBiB9PR0oc0tCqrL4tyyZQs0NDTw/v17fP/+HU+fPq3zZUxxcXGw2Wyw2Wyh2Ocn0xv4vexNJeZU5vnz5/j48SOV3d1IaCiZ04SQBpWRXg4lKCm4hhKU3EMI4eyD7NChA06fPg1PT098+vQJwcHBMDY2rjNfFi1ahKFDhyI5ORnm5ua4ffs22rZti8zMTNja2jaYJaCa+DMh57+w2Wxs2bIFnz9/BvB720HHjh3r2kXOw0N9qkUJ/E7MefHiRZPqqsQNTCYTCgoK6NOnj6hdoRAABgYGeP36db1/n3/+/BkZGRkwMDAQtSs8QQlKCq6hBGXNFBQUYN++fejatSsGDRqEtLQ0BAYGIjk5GWvXrq20FFsX0Ol0HD16FHJycsjPz+eISykpKTx8+BALFy6sc5+EQWpqKjIyMioJyry8PIwZMwYrV66Et7c3Zs2ahW/fvmHZsmV17qOwP0P8lA4Cfkco2Ww2VeD8D8LCwjBixAhISEiI2hUKAdC7d2+w2WxcvXpV1K5Uy6VLl0Cn02FmZiZqV3iCEpQUXEOn00Gj0ShBWQXl3Ww0NTUxc+ZMTqma2NhYuLi4QEpKSqT+KSsrIyQkBDExMWjbti3s7e1RWFgIQgi2b9+OU6dOidQ/QVC+lPXfZaJXr16hR48euHPnDi5cuIBVq1Zh69atkJGRwaZNm/Dw4cM69VHYiW3t27fHjx8/eN7K0LlzZ8jLy1P7KP/D27dvER8fT2V3NyI6duyIXr16Ye/evQKxl19UipefcxGbnIOXn3ORX1T7zzUhBHv27MGIESOgrq4uAC/rDuFt3qJolAiz7ElDo7ybjb+/P8LCwiAnJ4cpU6Zg9uzZ9bJ4bo8ePbBlyxbMnz8fZ8+ehZSUFAIDA0EIgYuLC7p06QJ9fX1Ru8k3LBYLqqqqaNWqFYDfvbwnTpwITU1NREVFoX379gAAKSkpeHl5YcOGDRg7dixevHgBFRWVOvGxfMm7LkoHqampcT1OTEwMpqamlKD8D0wmE9LS0hg0aJCoXaEQIO7u7pg4cSLevHnD+U7ghbfffuD4k2SEv05HcnYB/rsWQAOgpSSDvh1UMaGHFvTUuK8fXM6TJ0/w/Plz+Pj48DxW1FARSgqeoATl7242x44dg6mpKXr37o3nz5/D398fqamp8PX1rZdispy5c+fC1tYWkydPxtKlS+Hm5gbg9z0NHz4c379/F62DtaB8/yQhBKtXr8bo0aMxYMAAPHnypNIPx5w5cyAuLo7c3Fw4OTkJLUnmT4S95N2uXTvQaLRaJeZQBc5/w2QyMWjQIMjKyoraFQoBMnbsWCgrK2PDhg08jUvJLsDEQ08wYPs9BD/5hE9/iEkAIAA+ZRcg+MknDNh+DxMPPUFKdgHXcxBCsH79erRt2xYDBw7kyb/6ACUoKXiiKQvK9PR0rF27Ftra2pg4cSKUlJRw5coVJCYmYtasWZCTkxO1izVCo9Fw+PBhKCsrw8HBAX5+fpg9ezYIIUhJScG4cePqTFwJmpiYGHTu3BmjRo3CunXrsGHDBpw5c6bKLkMtWrSAs7MzpKSkcP36dfz777914qOwl7ylpKTQunVrvhNz0tPTkZSUJHjHGhhpaWl4/Pgxld3dCJGSksLmzZtx9OhRXL58masxp6KS0X9bBB5++N2Fqoxd/UNX+fmHH7LQf1sETkUlczVPcHAwLl++jO3btzfIMlUNz2MKkdIUBeWzZ8/g6uqK1q1bY9OmTRg9ejQSEhJw/fp1DBkypMF98BUUFBAaGoqXL19iwYIF2LFjB9zd3UEIwfXr1+Ht7S1qF3nm8+fP+Pr1K06cOIH79+/j8uXLWLZsWbWFgT08PJCdnY2RI0di1apVCA8PF7qfwl7yBvjP9O7RowcAUOWD8Hu7BIPBwPDhw0XtCoUQcHV1xeDBgzF9+nTk5ORUe+3O8LdYEhaHolJ2jULyT8rYBEWlbCwJi8PO8Oo/k58/f8a8efPg5OSEkSNH8jRPfaFh/RJSiJymIijLysrAZDJhZWUFAwMD3L59G+vWrUNKSgr27NmDTp06idrFWmFoaAh/f3/s3r0bp0+fxp49e+Do6AgAWLt2LS5evChiD3kjICAAANCsWTNER0djyJAhNY7p2LEjhg0bho8fP6JPnz5wdHTE169fhepnXXSb4ldQqqioQFdXl9pHid/Z3X379q2T7kkUdQ+NRsOBAwfw8+dPODo6orCwsMrrTkUlY+sNwRRC33rjDU7/JVKZl5eHsWPHQlJSEv7+/gKZTxRQgpKCJ4TZOq4+8P37d/j6+kJXVxe2trZgs9kIDQ3Fhw8fsGjRIoF3sxEl06dPx/jx4zFt2jS8fv0aJ06cQN++fQEAtra2DaKjRFlZGZYtWwYfHx9ISEggKiqKp+4SXl5eiIuLw/Tp00Gn0+Ho6IiysjKh+VtXEcp3797xtRfS3Ny8yQvK7Oxs3L17l8rubuRoamoiLCwM9+7dw8iRI/Hr168K51OyC7D6wkuBzrnqwstKeypzc3MxcOBAJCYm4sKFCw36N4YSlBQ80VgjlG/evMHs2bOhqamJpUuXwsLCAtHR0YiMjMTYsWOF2s1GVNBoNOzbtw+ampqws7NDQUEBbty4AT09PZSWlsLU1BR5eXmidvOvZGdnY9iwYdi0aRM6deoEKyurKvdLVkefPn1gYGCAwMBAnDp1Cvfu3cPq1auF5HHd1HLV09NDfn4+vnz5wvNYMzMzxMbGVvpxbUpcvHgRbDYbo0ePFrUrFEKmX79+uHLlCh48eFApKXEZMw6lPC5x10Qpm2AZM47z//T0dAwYMABv3rzBrVu3YGpqKtD56hpKUFLwRGMSlOXdbIYOHYoOHTogNDQUXl5eSE5OxtGjR6vsuNLYkJOTQ2hoKN69e4c5c+aAwWDgwYMHaNasGXJzc9GtW7d6GZF+8eIFTExMEBUVhWvXriE3N5ev7kM0Gg2enp64fv06lJSUsGHDBmzYsEFohY/raskbAN+JOaWlpQ2iPZ2wYDKZMDMzE0kTAoq6p0+fPrh27Rqio6PRuXNnnD17Fm++5iHyXSbPeyZrooxNEPkuE2+//UBQUBA6deqEpKQk3Llzp1H83lCCkoInGoOgzM/Px969e9GlSxcMGjQIX79+RVBQEJKTk+Ht7d3gisnWlq5du2LPnj04fPgwgoKC0KJFC9y6dQt0Oh1JSUkwMTGpV3/z06dPw9zcHPLy8oiOjoa+vj4+f/7Md99bBwcHtGrVCn5+fpw2lRMnTkRKSoqAPa+bJW8dHR3Q6XS+tizo6+tDWlq6ySbm5Ofn4/r161R2dxPDwsICCQkJMDU1xdixYzF+9R6I/T2fr1bQacDYpf5wdXXF0KFD8fLlS3Tv3l04k9UxlKCk4ImGLCiTk5OxePFitG7dGrNmzUKnTp0QEREBFosFZ2dnSEpKitpFkeHs7IwpU6Zg5syZiI+Ph4mJCaebxLNnz9C3b18UFRWJ1MfS0lIsXLgQ48aNw+jRo/Hw4UO0bduWE03j9wlfXFwcc+fOxfHjx5Geno6jR49CRkYGDg4OAn+v10WEUkJCAtra2nxFKBkMBkxMTJrsPspr166hsLCQ2j/ZBGnVqhWYTCbOnDmDdIYqyoRUjpVNgFyZVrh27RqCg4PRokUL4UwkAihBScETDAajQQlKQgju378POzs76OjoYN++fZgyZQrev3+Ps2fPwtLSstrSMk2JgIAA6OnpYezYsfj58yemTZsGV1dXAMCDBw8wZMgQke2ty8rKwpAhQ7Bt2zb4+fnh2LFjkJGRAfC7oHnz5s3Rpk0bvu1Pnz4d4uLi2LVrF6dNZVRUFJYuXSqgO/hNXUQoAf4zvYHfy95NtcB5WFgY/vnnH54SuygaDzQaDYOGj4JYM1XhziOvit59+gl1DlFACUoKnmgoEcqioiIEBwfDxMQEFhYWiIuLw44dO5CamootW7bUSnw0VqSlpREaGoq0tDRMnz4dhBDs3r0b3bp1A51Ox927dzF06FDk5+fXqV+xsbEwNjbGs2fPcPPmTXh4eFR4CIiJiYGhoWGtHgwUFRUxZcoU7NmzBwUFBTAzM8OWLVvg6+uL8+fPC+I2ANRNUg5QO0FpZmaGz58/IzU1VcBe1W+Ki4tx6dIlKjrZxPmUlV+pA46gIQCSsur2e7QuoAQlBU/Ud0H57ds3TjebSZMmQUVFBVevXkVCQgJmzpzZILrZiJL27dvj4MGDOHnyJPbt2wcpKSmcP3+ekz394MEDDBo0qM6yv48fP46ePXtCWVkZLBaLU9bov5S3XKwt8+fPR05ODo4cOQIAmDdvHmxtbeHs7IwPHz7U2j5QN0vewG9B+f79e766HpmZmQFAk1v2vnPnDvLy8qj9k02c4tK66RRWV/PUJZSgpOCJ+iooY2Nj4eLiAi0tLWzatAm2trZISEjAtWvXMHjw4AbXzUaUODg4YNasWZg3bx5iYmKgra2NM2fOAPgdWYuNjUX//v2RnZ0tNB9KSkrg4eEBJycn2NvbIzIyElpaWpWuy8jIQEpKikAEZdu2bWFra4tt27aBzWZXaFNpb2//1+LHvFCXS96FhYV8RRnV1dXRpk2bJpeYw2QyoaOjA319fVG7QiFCJBh181tRV/PUJY3vjiiESn0SlGVlZQgLC4OVlRUMDQ0RHh6O9evXIzU1Fbt3727w3WxEia+vL/T19WFnZ4fv37+jf//+2LhxI4Df2wnevHkDa2trZGRkCHzu9PR0DBw4EDt37kRAQACCgoIgLS1d5bXlCTn8Znj/iaenJ96+fYtLly4B+F+byvj4eHh5edXafl1GKAH+SgcBv6OUTSlCWVZWhnPnzsHW1pbaU93EaaMsC2G/A2j/P09jgxKUFDxRHwTl9+/fsXXrVrRr1w5jxowBm83GmTNn8P79eyxcuJBqlyYAJCUlERoaiqysLEyePBmEECxevJjTY5bBYODz58/o06cPXwW0/0Z0dDSMjY2RkJCAO3fuYPbs2dX+wLNYLCgoKAgsicLc3Bzm5ubw9fXlHPtvm8pTp07Vyn5dRSjbtGkDMTGxWiXmsFgskWf21xWPHj1Ceno6tX+SArKSDGgpyQh1Di1lGchKNr5mGZSgpOAJUQrK169fY9asWdDU1MTy5cthZWUFFouFyMhIjBkzplF2sxElbdu2xZEjR8BkMuHv7w8ajYbg4GBoa2sjNzcXWlpayM3NhZWVlUBqNgYFBaF3797Q0NAAi8WChYVFjWNYLFatE3L+xMvLC/fu3UN0dDTn2J9tKvmlrpJyxMXF0bZt21pFKIuLi/Hs2TPBOlZPCQsLg7q6Omf/KEXTpm8HVYjRhROnFKPT0Le9cLPIRQUlKCl4oq57eRNCcP36dQwZMgQdO3bEmTNnsGDBAnz69AlHjhwR2FInRdWMGjUKXl5eWLhwIR4/foxmzZrh4sWLYDAYiImJwcCBA1FSUgJLS0t8/PiRrzmKi4sxe/ZsuLq6wsnJCffu3YOmpiZXY8szvAXJ6NGj0bZtW/j5+XGOVdWmkh/qaskbqF2md/fu3SEpKdkklr0JIWAymRg9ejS115oCADChh5bAu+SUU8YmcDKrvB+8MUB9eih4oq4ilPn5+dizZw86d+6MwYMH49u3bzhy5AiSk5OxZs2aJtfNRpT8+++/MDU1hb29PbKystC5c2ccPXoUhBAEBgbCy8sLDAYDFhYWPHdn+fr1K/r164f9+/dj7969OHDgANcF5rOyspCUlCTwlmViYmKYP38+QkJCkJyczDn+Z5tKfqDT6aDRaHXyGaqNoJSQkICRkVGTSMx59uwZkpKSqOxuCg56avKw0FUReJRSjE6Dha4KdFXlBWq3vkAJSgqeELag/PTpExYtWgRNTU3Mnj0bXbp0wb1798BisTBp0qQm3c1GVIiLi+P06dMoKCjApEmTwGazYWdnB09PT9BoNCxYsAB79+6FgoICLC0tER8fz5Xdx48fw8jICO/evcPdu3fh5ubG09J1bTvkVMfkyZMhLy+PHTt2VDj+Z5tKfqirhzI9PT18+PABZWVlfI03NzdvEhFKJpMJRUVF9OnTR9SuUNQjNtrogyFgQcmg07DRpvFWEaAEJQVPCOPHsLybzdixY6Gjo4MDBw5g6tSp+PDhA86cOQMLCwsq81LEaGpq4tixY7h69So2b94MANi0aRN69eqFsrIyTJ06FefOnYO6ujr69OmD2NjYau0dPHgQVlZW0NbWBovFQs+ePXn2KSYmBvLy8tDV1eXrnqpDTk4Obm5uOHDgQKWam/9tUxkXF8ezbQaDUWdL3sXFxRWirLxgZmaGT58+CTTpqj4SFhaGESNGcLYjUFAAQGslGXiP7CJQm2tHdkFrISf8iBJKUFLwhCBbLxYVFeHo0aMwNjaGhYUFXr58iZ07d3K62WhrawtkHgrBMHjwYCxfvhzLly9HREQEGAwGzpw5A2VlZaSmpmLu3Lm4desWdHR0YG1tjSdPnlSyUVRUBDc3N0ybNg2TJ0/G3bt30bJlS778YbFYMDAwENq+t9mzZ6OgoACHDh2qdK68TaWdnR1+/PjBk926jFACtSsdBDTuAudv3rzBy5cvqexuiioZZ6KFBQPbC8TWwoEd4GDSOPdOlkMJSgqeEMSP4bdv3+Dt7Q1tbW04OztDVVUVV69excuXLzFjxgzIyja++lyNhTVr1sDKygrjxo3Dt2/foKamhnPnzoEQgmvXriEgIAC3bt1Cly5d0L9/f0RGRnLGlpcZCgoKwsGDB7Fnzx5ISEjw7YugOuT8DU1NTYwbNw7+/v6VIopVtanklrpKbNPS0oK4uDjfglJTUxOampqNWlAymUxIS0tj0KBBonaFop4yu68efGz1Icmg87ynUoxOgySDjk22+pjVV/ArKfUNSlBS8ERtBGVMTAycnZ2hpaWFzZs3Y8yYMUhMTMTVq1epbjYNBDExMZw4cQKEEIwfPx5lZWUwMzNDQEAAAGDt2rWIiIjAtWvXYGpqisGDB+P27dt48OABjIyMkJKSgnv37mHKlCm18iMnJwcfPnwQepa/p6cnPn36hLCwsErnyttUnjp1Cvv27ePapiCj/DXNo6Ojw7egBH5HKRtzYg6TycTgwYMhI9N4lyEpas84Ey3c8rBCTx1lAKhRWJaf76mjjFseVo0+MlkO9QtOwRO8CsrS0lKcPXsWlpaWMDIyQkREBDZs2IDU1FTs2rULHTt2FKK3FMJAXV0dJ0+exN27d7F27VoAgLu7OyZNmgQ6nY7x48fjy5cvuHTpEqysrDB48GBYWVlBT08PLBYLPXr0qLUP5Xs0hRmhBAADAwP07dsXvr6+VUYh/2xTyQ11Wcu1NpnewO/EnOjoaJE3MxAGaWlpePLkCZXdTcEVrZVkEDylB27Ot8TEHtrQVpap1FGHBkBbWQYTe2jjloclgqf0aNR7Jv+kSVSCZrPZeP/+PeLi4jj/4uPjkZmZiezsbM4PhbKyMpSUlNClSxfo6+tDX18fXbt2Rfv27SEmJibiu6gfcPtjmJOTg0OHDiEgIADJycmwsLDA2bNnMXLkSKoAeSOgb9++WLt2LVauXIlevXph4MCB2Lt3L2JjY/Hq1SuMHDkSkZGRUFVVRWlpKeh0OubMmQM1NTWBzM9isSArK4v27QWzv6k6vLy8MHz4cDx48AC9e/eudN7X1xdPnjyBnZ0dWCwWFBUVq7VXV0k5wG9BWd5Gkh/MzMzw69cvvHjxQujiva45d+4cGAwGhg0bJmpXKBoQemryWDOyC9agC/KLSpGUlY/iUjYkGHS0UZZtlB1wuKVR3/mnT58QFBSEwMBAfPr0CQCgoqICfX19DBo0COrq6pCQkMDdu3dhYWGBsrIyZGRkID4+HgcOHMDXr18BAC1btoSzszNcXV05G92bKjUJylevXmHHjh04cuQISktL4ejoiHnz5sHAwKAOvaSoC5YuXYrIyEhMmDABz549Q6tWrXD+/Hl0794db9++Rfv27ZGfn4+DBw/i5s2bcHR05LwnaktMTAy6d+9eJw965UX1fX19qxSUkpKSCAkJgaGhIVxdXREWFlZtVYK6jlB+/PgRpaWlfD3IGRoaQlxcnFPiqTERFhYGa2trqlUrBd/ISjLQpaWCqN2oP5BGyNWrV0n//v0JjUYjcnJyZOrUqeTatWvk69evPNnJyMggt27dIrNmzSKKiooEAOnduzc5c+YMYbPZQvK+frNlyxbSrFmzCsfKysrI1atXyeDBgwkAoqamRtasWcPz603R8MjIyCCampqkd+/epLi4mBBCyObNmwkAAoDMmzePEEJIaWkpcXZ2JjQajRw+fLjW8+rp6ZG5c+fW2g637N+/n9BoNPL27du/XnP+/HkCgPj5+VVrq0uXLpzXRdjcvHmTAKjW75owNTUlTk5OAvRK9GRmZhIxMTGyZ88eUbtCQdFoaFSC8uvXr8TBwYEAIObm5iQoKIj8/PlTILZ//fpFTp48SaytrQkAMnz4cJKcnCwQ2w2J7du3ExkZGUIIIT9+/CC7du0iHTp0IACIoaEhOXr0KCksLBSxlxR1yYMHDwiDwSALFy4k/v7+hMFgkLZt2xIAhE6nk9u3bxNCfj94uLm5EQBk9+7dfM/3/ft3AoAcOXJEULdQIwUFBaRFixZk1qxZ1V63YMECwmAwyMOHD/96Tbdu3cjMmTMF7WKVJCUlEQDkypUrfNuYO3cuadeunQC9Ej2BgYGERqORz58/i9oVCopGQ6MRlEFBQaR58+ZERUWFHDt2TKgRRCaTSVq2bEnk5ORIQEBAk4pW7tq1izAYDLJgwQKiqKhI6HQ6GTt2LImMjGxSrwNFRXx8fDhRSU9PT1JUVESGDRtGGAwGUVJSIp8+fSKEEMJms8m8efO4iuT9jbt37xIAJC4uTpC3UCOrV68mMjIyJCsr66/XFBcXk549exJNTU2SkZFR5TXGxsZk2rRpwnKzAmVlZURSUpL4+/vzbePkyZMEAElPTxegZ6Jl5MiRpGfPnqJ2g4KiUdHgBSWbzSZr1qwhAIiTk9Nfv8QFzffv34m7uztnWa+xiyk2m03u3btHDAwMCACiqKhIFi5cSJKSkkTtGoWISUpKIgYGBoROpxNZWVny8eNHQgghOTk5pE2bNkRcXJwYGhqSX79+EUJ+v5eWLl1KAJD169fzPJ+vry+RlpYmJSUlgryNGvn27RuRlJQkGzdurPa6lJQUoqysTIYMGULKysoqnTczMyOurq7CcrMSnTt3JrNnz+Z7/MePHwkAcuHCBQF6JTp+/PhBJCUlydatW0XtCgVFo6JBC0o2m01WrFhBANT4JS8s9uzZQwCQmTNnVvnj0dApLCwkQUFBHCGpoaFBAJDv37+L2jWKesDt27eJsrIyadOmDYmIiCBt27YlJiYmnG0Pz58/J5KSkkRMTIy4urpyHrzYbDZZu3YtAUBWrFjB0wPZ+PHjibm5uVDupyamTp1KNDQ0SFFRUbXXXbt2jdBotCq/lywsLOp0T+KoUaPIoEGD+B7PZrOJmpoaWbZsmQC9Eh2hoaEEAHn//r2oXaGgaFQ0aEG5evVqAoBs3rxZpH4cOHCA0Gg0MmfOHJH6IUi+fPlCVq9eTVRVVQkAMmTIEHLt2jVy5MgRAoAUFBSI2kUKEcJms4mvry+h0+mkf//+JDMzkxBCSFRUFJGQkKjwWTh+/DhnOXzfvn0V7JQn8Hh5eXEtKjt27FjjXkZhkZCQQACQoKCgGq9duXIlodPpJDw8vMJxa2trMm7cOCF5WJkFCxYQHR2dWtkYPXo0sba2FpBHosXR0ZF069ZN1G5QUDQ6GqygvHfvHqHRaGTt2rWidoUQ8ntvYWNYFoqOjiYTJ04k4uLiRFZWlsycOZO8evWKc/7UqVMEAMnNzRWhlxSiJD8/nzg6OhIAZNGiRZWWnss/CyEhIZxjc+fOJTQajTAYDPLo0aMK1wcEBHAd5c/LyxNYpji/DB06lOjr69cogEtLS0nfvn2Juro6+fLlC+f4wIEDyZgxY4TtJod9+/YROp1eY1S1Onx8fIicnBwpLS0VoGd1T2FhIWnWrBnx9vYWtSsUFI2OBiko8/Pzia6uLunZs2e9+YJjs9lk6NChRENDg2RnZ4vaHZ4oKSkhoaGhpHfv3gQA0dbWJlu3biU5OTmVrj179iwBwIlIUTQt3r9/T/755x8iIyNDTp8+XeU1bDabODg4EHl5efLmzRtCyO9kFXNzcyIuLk7U1dUrlZQqj/JPnjy52s/0vXv3CADy/Plzwd0Uj9y+fZsAIDdv3qzx2i9fvhB1dXXSt29fzn0NGzaMjBw5Uthucrhz5w4BUOHBkFciIiJE/roLgitXrhAA5MWLF6J2hYKi0dEgWy+uWrUKqampCAwMrDcdbGg0Gvbv34+CggJ4enqK2h2uyMnJwZYtW9CuXTvY2dmBTqfj7NmzePfuHby8vKrs+CEuLg4AjbIVG0X13LhxA8bGxsjPz8fjx49hb29f5XU0Gg0HDhyAhoYG7Ozs8OvXL4iLi+Ps2bNQVFREdnY27OzsKryHpk6diqNHjyIoKAiTJk36ayeZmJgYSEpKolOnTkK5R27o27cvunXrBl9f3xqvLW9TGRERAW9vbwB12ykHAKcZQ21aMBoZGUFMTKzB9/VmMpnQ1dVF165dRe0KBUXjQ9SKlleysrKIlJRUvV2yCAgIIHQ6nVMmpT6SkJBA3N3diYyMDJGQkCAuLi4kJiaGq7FXr14lAJpkDc6mCpvNJj4+PoROp5MhQ4ZwHYF//vw5kZKSqlAiJzIykoiJiRE6nU7mz59faUxoaChhMBjE1ta2yiXaiRMnElNTU/5vRkAcPXqUACDx8fFcXb9hwwZCo9HItWvXyNixY8mAAQOE7OH/KCsrI9LS0nyXaSrHwMCAuLi4CMiruqe0tJS0aNGCLFy4UNSuUFA0ShqcoNy2bRsRFxevt11Y8vLyiLy8PFm5cqWoXalAWVkZuXLlChk0aBABQNTV1Ym3tzfPr+OtW7eoDMkmxI8fP4idnR0BQJYvX87zFpNDhw4RAOTo0aOcY/7+/pwknePHj1cac+HCBSIhIUGGDRvGKTVUTufOncmMGTP4uxkBUlRURFq2bEmmTJnC1fVlZWVkyJAhREVFhYwaNYr07dtXyB5WRF9fv9av28yZM0nHjh0F5FHdU75d4s89vBQUFIKhQQlKNptNOnToQBwcHETtSrXMmDGDaGhocFrRiZI/u9kYGRmR4OBgvjfol++lqs1+LIqGwdu3b0mXLl2InJwcOXv2LN92XFxciIyMDHn58iUh5Pfn2NHRkYiJiREpKSny7NmzSmOuX79OpKWlSf/+/Tndrn7+/EnodDo5cOAA374Ikn///ZdISEhw/VCWmZlJWrduTVq0aFHnRbVtbW1J//79a2WjPCpbXWH3+sz8+fOJhoZGoyzvRkFRH2hQgjI2NpbrzfCiJCYmhgAgt27dEpkPHz9+JF5eXkRBQYGIiYkROzs7cv/+/VoXYH/48CFPS30UDZPLly8TRUVF0r59e44Q5Jf8/HzStWtX0qlTpwrisEuXLkRSUpJoa2tXKVLu3r1LZGVliYWFBcnLyyMPHjwgALjeniFssrOziaysLE+rEQ8fPiQ0Go1oaGgI0bPKLF68mGhra9fKxtu3bwkAcvXqVcE4VYew2Wyira1dZy0vKSiaIg0qKYfFYoFGo8Hc3FzUrlRLt27dICsri5iYmDqdlxCCe/fuYcyYMWjXrh0OHz4MNzc3fPjwASEhIejVqxdoNFqt5qCScho3bDYbGzZswPDhw2FhYYGnT5+ic+fOtbIpIyOD0NBQJCcnw93dHYQQyMrK4ty5c5CQkMDnz58xfvx4lJWVVRhnZWWFmzdv4vnz5xgwYAAiIyMhISGBLl261MofQdG8eXNMnjwZu3fvRkFBAVdjzM3N0aNHD3z58gXnz58Xsof/Q09PD8nJySgsLOTbRrt27aCsrNwgE3NiY2Px6dMn2NjYiNoVCopGS4MSlDExMejYsSNkZWVF7Uq10Ol0dO/evc4EZWFhIYKCgmBoaAgrKyskJiZi9+7dSElJwaZNm6ClpSWwuShB2XjJy8vDmDFjsGLFCqxevRrnzp2DgoKCQGx37NgR+/fvx7Fjx3Dw4EEAgK6uLk6cOIGSkhLcuHGDkwX9X8zNzXHnzh28ffsWmzdvRqdOnSAhISEQnwTB/PnzkZ2djeDgYK7HGBoaolmzZnBxccHHjx+F6N3/0NPTAyEEHz584NsGjUaDmZkZHj9+LEDP6gYmk4nmzZvDyspK1K5QUDRaGpSgjI2NhYGBgajd4AoDAwOhC8qvX79i9erV0NbWhqurK1q2bInr16/j5cuXcHNzE4rwZjAYAChB2dh4/fo1evTogTt37uD8+fNYvXo16HTBfj2MHz8e7u7umDNnDp49ewYAGD58OFauXAlCCNatW4cLFy5UGmdkZITw8HDk5eUhOTkZ3759E6hftUFHRwc2Njbw8/MDm83maoyEhAQ0NDTQvHlz2Nvbo6ioSMheCqZ0EPBb4D958oTre60vhIWFYcSIEZwHYgoKCsHToARlTk4O1NXVRe0GV2hoaCAnJ0cotlksFiZOnAgtLS34+vrCzs4Or169wuXLlzFw4MBaL2tXBxWhbHxcuHABpqamAICnT59i5MiRQptr27Zt6Ny5M+zs7JCbmwsAWL16NQYNGgRxcXE4OTnhzZs3lcbp6emBzWaDEAJLS0ukpqYKzUde8fLywps3b3D58mWurmcwGGCz2QgNDcWLFy+wYMECIXv4ux6mnJxcrQWlmZkZcnNz8erVKwF5Jnxev36NhIQE2NraitoVCopGTYMSlE2Z0tJShIaGonfv3jA2Nsb9+/fh4+OD1NRU7Ny5Ex06dKgTPyhB2Xhgs9lYs2YNRo0aBWtrazx58kTo7yMpKSmEhoYiPT0dU6dOBSEEYmJiOHHiBFq2bImSkhKMHj0aP378qDDuxYsXYLPZCAwMRGFhISwtLZGUlCRUX7mlZ8+eMDMzg5+fH1fXi4uLo6SkBEZGRti+fTt27tyJkJAQofpIo9Ggq6tba0FpamoKGo3WoJa9mUwmZGRkMHDgQFG7QkHRqKEEZT0nOzsbmzdvho6ODuzt7cFgMBAWFoZ3797B09Ozym42woQSlI2D3NxcjBo1CmvXrsX69etx9uxZNGvWrE7mbteuHQIDA3HmzBns3LkTAKCkpAQmkwk2m413795h8uTJIIRwxrBYLIiLi2PIkCG4d+8e6HQ6LC0tay2QBIWnpyfu3r3L1TaX/3bKcXd3h6OjI6ZOnVplZFaQ6Onp1fr1kpeXR9euXRtUYg6TycSQIUMgLS0talcoKBo1DUpQiouL49evX6J2gyvK283xS0JCAtzd3aGpqYmVK1eiX79+iI2Nxd27d2FjYyOylpOUoGz4JCQkwMTEBJGRkbh06RKWL18u8P2SNWFra4v58+fDy8sLT58+BfB73/H+/ftRUlKCM2fOYOvWrZzrWSwWunbtCklJSWhra+PevXuQlZWFpaUlEhIS6tT3qrCxsUGbNm24asdYHqEEfkcO9+3bh5YtW2Ls2LFC/X4ThKAEfu+jbCgRytTUVDx9+pTK7qagqAMalKDs3Lkz4uLiRO0GV7x48YLn8iZsNhtXrlzBoEGD0KVLF5w/fx5Lly5FSkoKAgMD0b17d+E4ywOUoGzYhIWFoUePHpCUlER0dDSGDh0qMl82bdoEIyMj2NvbIzs7GwDg7OyMGTNmgE6nY8mSJbh9+zaA3xUeDA0NOWNbtmyJiIgItGjRAlZWVnj+/LlI7qEcBoOB+fPn4/Tp00hJSan2WnFx8Qq9vOXl5REaGop3795hzpw5QvOxffv2SE1N5brE0d8wMzPDy5cvkZeXJyDPhMe5c+cgLi6OYcOGidoVCopGT4MSlIaGhnj27FmDyDCMjY2t8ANYHT9//sSuXbvQqVMnDBs2jFOG5NOnT1i5ciVUVVWF7C33UIKyYVJWVobly5djzJgxGDJkCB49egRdXV2R+iQhIYHTp0/jx48fcHZ25nyut2/fDhMTE4iLi8Pe3h6vX79GfHw8jIyMKoxXVVVFeHg4tLW10bdvX0RFRYniNjhMnjwZcnJyCAgIqPY6BoNR6fOjr6+P3bt349ChQzhy5IhQ/CvP9H7//n2t7JiZmYEQwoks12fCwsJgbW1d51uDKCiaIg1KUBoYGODnz59C32tUW9LT05GSklJjiaOPHz/Cy8sLmpqamDdvHrp3744HDx7g6dOncHJyqlf19sqhBGXDIycnB8OHD4ePjw82bdqE06dPQ05OTtRuAQC0tLQQHByMS5cucZaLJSQkcPbsWcjLy6OgoACjR49GaWlpJUEJAMrKyrh9+zY6duyIfv364cGDB3V9Cxzk5eUxffp07N+/v1JS0X/575L3f3FxcYGrqytmzJiB+Ph4gfsnqNJBHTp0gKKiYr1f9s7MzMS9e/eo7G4KijqiQQlKMzMzyMrK4vTp06J2pVpOnz4NBoNRZRFdQggiIiJga2sLXV1dBAYGwt3dHR8/fsTp06fRs2dPoZb9qS3lgvK/S3YU9Ze4uDiYmJjgyZMnuHr1KhYtWlTv3l9Dhw7F0qVLsXTpUkRGRgIAWrVqhdDQUBQXF+PNmzeg0Wjo2rVrleMVFBRw/fp1GBoaYtCgQQgPD69L9yswd+5c5Ofn49ChQ3+95s8l7/+yc+dO6Orqws7ODj9//hSobyoqKlBQUKi1oKTT6ejRo0e9T8y5ePEi2Gw2Ro0aJWpXKCiaBA1KUMrLy2PChAk4cOCAwARNflEpXn7ORWxyDl5+zkV+Ue3sEkKwd+9ejB49ukLNzPJuNgYGBujTpw9ev36NPXv2IDU1FT4+PmjdunVtb6VOEBMTA41GoyKUDYCQkBCYmZlBTk4O0dHR9bpsytq1a9GrVy+MGzcO6enpAIA+ffpg8+bNnPqTR48e/et4eXl5XLlyBb169cLQoUNx7dq1unK9ApqamnBwcIC/v/9fv6PKl7z/m8VeTnmbytTUVLi5uVV5Db/QaDSBJ+YI0j9Bw2Qy0atXL6ipqYnaFQqKpoEoGojXhpiYGAKAMJlMvm28+ZpHVp+PJ5ab75A2Sy4R7f/8a7PkErHcfIesPh9P3nzN49n2vXv3CABy69YtQgghnz9/JitXriQtWrQgAMiwYcPIjRs3CJvN5tt/USMhIUF27twpajco/kJJSQlZtGgRAUAcHR1Jfn6+qF3iirS0NKKqqkr69+9PSktLCSGEsNlsoqioSGg0GmEwGOThw4fV2igsLCQjRowg4uLi5Ny5c3XhdiVYLBYBQEJCQqo8f/jwYQKAc49VcerUKQKA7N27V6C+OTo6EktLy1rbuXbtGgFA3rx5IwCvBE9eXh6RlJQkvr6+onaFgqLJ0KAilMDvfZS9evXCihUreG5ZlpJdgImHnmDA9nsIfvIJn7IL8OfzNQHwKbsAwU8+YcD2e5h46AlSsrnLiiwrK8PixYvRuXNnNGvWDBMnToS2tjb8/Pzg4OCA169f49KlSxgwYEC9W3bkhaqSCijqB1lZWRgyZAi2bt0KX19fHD9+HDIyMqJ2iytatmyJEydO4Pbt29iwYQOA33t1f/78CVVVVYiLi2PMmDH4+vXrX21ISkrizJkzGD16NMaOHSuS7TGGhobo06cPfH19q4zgcdO+1MHBATNnzsTcuXMF2sJVUBHK8s5K9XXZ+9q1aygqKqLKBVFQ1CENTlACwJ49e/DmzRusXbuW6zGnopLRf1sEHn7IAgCUsatfqik///BDFvpvi8CpqOQa5/Dz88Pjx49Bp9NhamqK+/fvY9OmTUhLS0NAQADat2/Ptb/1mb8lFVCIlmfPnsHY2BjPnj3DzZs34enp2eAeXPr164c1a9ZgzZo1uH37NuLj41FaWort27eDTqcjJycH9vb21b7/JCQkcOLECTg6OmL8+PFCy5quDi8vLzx58qRKwcVtYpufnx/09fUrtKmsLXp6evjy5Uut92c2b94cnTp1qreJOWFhYejevTvatm0ralcoKJoMDVJQ6uvrY9WqVdi0aRNXpUJ2hr/FkrA4FJWyaxSSf1LGJigqZWNJWBx2hlf9ZJ+dnY0FCxZg8eLFIIRARUUFTCYT7969g4eHBxQUFHias75DCcr6x4kTJ9CzZ08oKSkhOjoa1tbWonaJb5YvX47+/ftj/PjxuH37Nuh0OkaOHIng4GAUFhbi4cOHWLhwYbU2GAwGgoKCMHnyZLi4uGD//v115P1vhg4dig4dOlRZ6JzbxDZJSUmEhIQgKyurUucgfinP9H737l2tbZmZmdXLCGVRUREuX75MZXdTUNQxDVJQAsDixYvRvXt3jBw5EomJiX+97lRUMrbeEEyZoa033uD0fyKVCQkJcHNzQ8uWLeHr6wt5eXk8evQI4eHhGD16tMi62QgbSlDWH0pLS+Hp6YkJEybAzs4O9+/fh7a2tqjdqhViYmI4duwYGAwGtm3bho4dO0JGRgY2NjZYsmQJ2Gw2/P39cezYsWrt0Ol07Nu3D3PmzIGbmxv8/f3r6A5+z+3h4QEmk1mp7iM3S97l6OjoICgoCGFhYdixY0et/RJU6SDgd2LOixcvkJ+fX2tbguT27dv48eMHtdxNQVHHNFhBKS4ujitXrkBFRQV9+vTBy5cvK12Tkl2A1RcqH68Nqy68xJGzlzFw4EB06dIFTCYTUlJS0NbWRlxcHMzMzAQ6X32EEpT1g4yMDAwYMAA7duzAjh07EBQU1Gj6FauqquL06dP48uVLhbaQ69evh7W1NSQlJTFt2jQ8e/asWjt0Oh3+/v5YtGgR5s+fDx8fHyF7/j8mTZoEZWVlbN++vcJxXktvjR49Gp6enliwYEGtl5iVlJSgpKQkEEFpZmYGNpst8oLyf8JkMqGnp8dzpzIKCora0WAFJfC/ThkaGhro06dPpfpzy5hxKOVxibsmiopLsCg0Bt+/f4e3tzekpKTQokUL3L9/H1paWgKdq75CCUrRw2KxYGRkhISEBNy+fRtz5sxpcPsla6JHjx4QExNDfHw8rly5AuB39PLUqVNo0aIF6HQ6bGxsOG0b/waNRoOPjw9Wr16NpUuXYs2aNXVS7kZaWhozZ87E4cOHkZOTwznOS4SyHB8fH5iYmMDe3h5ZWVm18ktQiTmdO3eGvLx8vdpHWVZWhvPnz8PGxqbRfR4oKOo7DVpQAr+L9d6+fRudO3eGtbU1pk6dipycHLz99gOR7zJ53jNZI3QxSLc1RNee/bFmzRqoqqoiIiICmpqagp2nHkMJStFy5MgR9OrVC+rq6mCxWFUW0G8MJCQkoKysDL169cLEiRORnPx7u0n5HuXS0lJ8+fIF48ePR1lZWbW2aDQa1qxZg3///Rfe3t5YsmRJnYjKmTNnoqysDPv27eMc46fblLi4OE6fPo2CggJMmjSpVu1nBSUoxcTEYGpqWq8E5YMHD5CRkUHtn6SgEAENXlACv9uvhYeHY+/evQgNDUWnTp2wLPAqxIT0gErYZbj0OhdbtmzB48eP0bJlS+FMVE+hBKVoKCkpwZw5c+Di4oIJEybg3r17jfpBhsVigUaj4cSJE5CXl4e9vT2Ki4sBAMbGxti9ezeKiopw48YNrFmzhiubS5Yswfbt27F582bMmzevVsKMG9TU1ODk5ISAgACO7/x2m2rdujWOHTuGK1euYPPmzXz7JChBCfwvMacuxDk3hIWFoWXLljAxMRG1KxQUTY5GISiB33ul3NzckJCQgN69e+NhUi7KhPQdR6OLQc9yFLy8vDjLV02J6lrHUQiHb9++oV+/fti3bx/27NmDgwcPQkpKStRuCRUWi4WOHTtCS0sLISEhiImJweLFiznnp0yZgqlTp4JOp2P9+vU4f/48V3bnzZuHffv2YefOnXBzc6sxullbPDw88PnzZ05NTH6WvMsZPHgwli9fjuXLlyMiIoIvf/T09JCeno68vDy+xv8Xc3NzpKenIykpqda2agshBEwmEzY2NhX23VJQUNQNje5T16pVKwQdPwXx5sKNGn7OK6l1m8aGChWhrFuePHkCIyMjvH37FuHh4XB3d28S+8NiYmJgaGgI4HchbV9fX2zfvh1hYWGcawICAtC9e3fIyMjAyckJr1+/5sr29OnTERQUhMOHD8PFxUWoD0hdunTB4MGDOYXO+Y1QlrNmzRpYWlrC0dER375943m8IDO9e/ToAaB+FDiPiYlBcnIyld1NQSEiGp2gBIBPWcIvY0EAJNXBPPURSlDWHYcOHYKlpSW0tLTAYrHQq1cvUbtUJ5SWluL58+cwMjLiHJs9ezbGjh0LV1dXTikeKSkphIWFQUpKCoQQ2NjY4MePH1zNMWnSJJw4cQInT57E+PHjhfqe9vLywvPnz3Hnzp1aRSiB3xHOEydOgM1mY8KECTxHWMsF5Zs3tS+npqKiAl1d3Xqxj5LJZEJJSQmWlpaidoWCoknSKAVlcalw90XV9Tz1Dar1ovApLi7GjBkzMHXqVLi6uiI8PLxJ7dVNTEzEr1+/KghKGo2GgwcPQlVVFXZ2digsLAQAzpJ4QUEB3r9/D1dXV6739Dk4OODs2bM4f/48xowZw7EpaPr164d//vkHvr6+fCXl/ImGhgZOnjyJ8PBwnjqGAYCCggJatGghsH2U5ubm9UJQhoWFYcSIEZzXl4KCom5plIJSglE3t1VX89Q3qAilcPn8+TP69u2Lw4cP48CBA9i7dy8kJSVF7VadUt6/unv37hWOKygoIDQ0FAkJCfDw8OAc79evH/79918UFxfj7Nmz2LJlC9dzjRo1CufPn8fNmzcxatQoFBQUCOQe/guNRoOXlxeuXr2KT58+AeB/ybucvn37Yu3atVi3bh1u3LjB01hBJ+bExsbi169fArHHD69evUJiYiKV3U1BIUIapSJqoywLYe8wo/3/PE0RSlAKj4cPH8LIyAhJSUm4d+8epk6dKmqXRAKLxUL79u3RrFmzSue6d++OgIAA7N27FydOnOAcX7RoEWxsbCAhIYGlS5fi1q1bXM83ePBgXL58Gffv38ewYcNq3eu6KsaNGwcNDQ1Ob3FBfIaWLl2KgQMHYsKECUhLS+N6nCAFpbm5OUpLSzkPAaKAyWRCVlYWAwYMEJkPFBRNnUYpKGUlGdBSkhHqHFrKMpCVbHoZ3gAlKIUBIQR79+5Fnz59oKurCxaLxUl4aIqUF27/G1OnToWTkxOmT5+OV69eAfgdBQwKCkKbNm0gIyMDBwcHTjSQG6ytrXHjxg2wWCwMHDgQubm5tb6P/yIhIYE5c+bg3LlzAAQjKOl0Oo4dOwZJSUmMGzeOa5uCFJT6+vqQlpYWaWIOk8nEkCFDGk2nKAqKhkijFJQA0LeDKsTowolTitFp6NteVSi2GwKUoBQshYWFmDZtGmbMmAE3Nzfcvn0b6urqonZLZJSVleHZs2ecDO+qoNFo2LNnD7S0tDB27FjOMnWzZs3AZDLBZrNRVFQEW1tbnpZie/Xqhdu3b+PVq1fo169frbvS/ImbmxunpI2gMstVVFQQEhKCx48fY8WKFVyN0dPTQ3Z2do1dhriBwWDAxMREZPsoU1JSEBUVRWV3U1CImEYrKCf00BJ8l5z/p4xN4GTWNNosVgUlKAVHamoqrKyscOzYMQQGBiIgIAASEhKidkukvH79GgUFBdVGKAFATk4OZ86cwcePHzFr1izO8c6dOyMoKAj5+fmIi4vDjBkzeCq8bWJigvDwcHz69Al9+/ZFeno63/fyJ0pKShg/fjwACHRZvWfPnvDx8cHmzZtx8eLFGq8XZOkgQLSJOefOnYO4uDiGDRsmkvkpKCh+02gFpZ6aPCx0VQQepRSj02ChqwJdVXmB2m1IUIJSMNy7dw9GRkb48uUL7t+/DxcXF1G7VC9gsVgAAAMDgxqv7dy5M/bt28epKVmOnZ0dFixYgNLSUhw5cgR79+7lyYdu3bohIiICmZmZsLKy4ml/Yk2Ui9/w8HCB2QQAT09PjBo1Cs7OzjUWGtfV1QUgOEFpZmaGtLQ0pKSkCMQeL4SFhaFfv35QUFCo87kpKCj+R6MVlACw0UYfDAELSgadho02+gK12dCgBGXtIIQgICAA/fr1Q+fOnREdHQ1jY2NRu1VviImJQbt27aCoqMjV9U5OTpg2bRpmzZqFFy9ecI7/+++/sLKygrS0NObNm4eHDx/y5Efnzp0RERGB/Px8WFlZ8bQfszo6dOgAALh06ZJAWz/SaDQEBgZCUVER9vb2KCoq+uu18vLyUFdXF6igBFDnUcqMjAzcu3ePyu4WAGw2G0lJSbhx4waYTCZ27dqFtWvXIiwsDFeuXMHLly+p732KaqGR+tKEVUicikrGkrA4gdnbZKsPB5Omu9wNADNmzMCTJ09EmtXZUPn16xfc3d1x9OhReHh4YPPmzU2yfWd1WFpaQkNDg9OqkBt+/foFc3NzFBQUIDo6mpMdnp6eDgMDA/z48QNycnKIiYnheX9qUlISrK2tUVZWhjt37qBdu3Y8jf+TkpISzraGS5cuCXypNjo6Gr169YKbmxt27Njx1+ssLS2hqalZIVO+NrRt2xY2Njbw8/MTiD1uOHz4MKZOnYovX75ATU2tzuZtDGRnZ+PMmTOIjo5GXFwc4uPja9yGISEhgY4dO6Jr167o3r07bG1ta/15oGg8NOoIJQCMM9HCgoHtBWJr4cAOTV5MAlQvb35JTk5G7969ERoaimPHjsHPz48Sk3/AZrMRGxtb4/7JP5GWlkZoaCi+fv2K6dOnc/ZMqqqq4uzZs/j16xdyc3NhZ2eH4uJinmy3adMGkZGRkJKSgoWFBSernF/K/+Zt27aFr69vrWxVhbGxMbZt24aAgACEhob+9TpBZnoDotlHGRYWht69e1NikkvKyspw/fp1ODg4QENDAzNnzsTTp0+hq6uLlStX4sqVK/jw4QMyMjKQl5eHvLw8ZGZmIiUlBXfv3oWvry/MzMzw8eNHrFmzBrq6uujbty+Cg4OFUr+VomHR6AUlAMzuqwcfW31IMug876kUo9MgyaBjk60+ZvXVFZKHDQtqyZt3wsPDYWRkhKysLDx48AATJkwQtUv1krdv3+Lnz5/VZnj/DT09PRw6dAinT5/Gnj17OMfNzMwQEBCAgoICPHr0CAsWLODZdqtWrRAREQElJSVYWVkhLo7/VQ8ajQYxMTFYW1sjPDwcsbGxfNv6GzNmzICDgwOmTJnyV9FYLigFtUhlZmaGmJiYapfaBcmPHz9w8+ZNKrubS86fPw9dXV0MHjwY8fHx2LhxI9LS0vDs2TMEBwdj0aJFGDJkCNq2bQsVFRXIy8tDXl4eysrK0NTUhJWVFWbPno19+/bh4cOHyMjIQHBwMGg0GiZNmgQNDQ1s3bqVCjY0YZqEoAR+RypveVihp44yANQoLMvP99RRxi0PKyoy+R+o1ovcQwjBtm3bMGDAAHTv3h3R0dFcJZs0VcoTcvgRlMDvZJw5c+bAw8MD0dHRnONubm5wcXEBnU5HQEAAgoODebatrq6Ou3fvolWrVujTpw/HV34QFxdH165doa2tLZQlYhqNhgMHDkBDQwN2dnZVlk7S09NDbm4uMjMzBTKnmZkZioqK8OzZM4HYq4krV66guLiYEpQ18PXrV9jZ2WH06NHo1KkTHj9+jPj4eHh5edUqsisjIwMnJyfcuXMH79+/x6RJk7Bo0SL06NFDKA9JFPWfJiMoAaC1kgyCp/TAzfmWmNhDG9rKMpU66tAAaCvLYGIPbdzysETwlB5oLeQi6Q0NKkLJHQUFBXBycoKnpyc8PT1x9epVqKioiNqteg2LxULbtm2hpKTEt40tW7agW7dusLOzQ05ODoDfAmv37t3o0qUL5OTkMH36dL5+9FRUVHDnzh3o6enB2tqa72LeDAYDbDYb8+bNw6lTp5CamsqXneqQl5dHaGgoXr9+jXnz5lU6L+jSQd27d4ekpGSdLXszmUwYGBigTZs2dTJfQ+TixYvo1KkTIiIicOLECVy+fBk9evQAjSbYZFUdHR0EBATg0aNHKCkpgYmJCdavXy+w6DdFA4E0cX4WlpD4tO8k5lM2iU/7Tn4WlojapXrPmjVriIaGhqjdqNd8+PCBdOvWjcjIyJBTp06J2p0Gg5WVFRkzZkyt7Xz8+JEoKiqSUaNGETabzTn+4cMHoqioSJo1a0batGlDsrKy+LKfl5dHLCwsiKysLAkPD+d5fPPmzYmPjw/Jzc0lzZo1I4sWLeLLD244dOgQAUCOHj1a4Xh+fj4BQIKCggQ2V8+ePYmDg4PA7P2NX79+ETk5ObJu3Tqhz9VQCQ0NJQwGg4wePZpkZmbW2bzFxcVk5cqVBABZtGhRhc8fReOmyQtKCt7ZsGEDUVFREbUb9ZYbN24QJSUloqOjQ54/fy5qdxoMZWVlpFmzZmTjxo0CsXfhwgUCgPj6+lY4fu3aNQKASEtLk0GDBpHS0lK+7P/8+ZP079+fSElJkevXr/M0tkWLFmT9+vWEEEIWLFhAFBQUSF5eHl9+cIOLiwuRkZEh8fHxFY63atWKLF++XGDzeHl5EW1tbYHZ+xuXLl0iACrdD8VvTp06RcTExIijoyMpKRFNkGTbtm0EAPHw8KBEZROBEpQUPLN582aioKAgajfqHWw2m2zatInQ6XQyaNAgvqNfTZU3b94QADyLs+pYtGgRYTAY5MGDBxWOr1u3jgAgdDqdLFu2jG/7v379IsOGDSMSEhLkwoULXI9r2bIlWb16NSGEkOTkZCImJkb8/f359qMm8vPzSdeuXUmnTp3Ijx8/OMf79OlD7O3tBTZPaGgoAUA+f/4sMJtVMXnyZNK+fXtKqFRBREQEodPpZOLEiXw/LAmKgIAAAoBs2rRJpH5Q1A2UoKTgmW3bthEZGRlRu1Gv+PHjB7G3tycAyLJly0T+Rd4QOXnyJAFAMjIyBGazuLiY9O7dm2hqalawW1ZWRkaMGEGkpKQIABIWFsb3HEVFRWTMmDGEwWCQkJAQrsZoa2tXiAw6OjqStm3bCvV9k5iYSGRlZcmECRM4QmzatGnEwMBAYHOkpKTU+vWsiZKSEqKiokIWL14stDkaKj9//iTt2rUjvXr1qjffQQsWLCASEhLk5cuXonaFQsg0qaQcCsFAJeVU5N27dzA3N8fly5dx5swZbNiwAWJiYqJ2q8ERExMDLS0tgSYuiYuL49SpUygsLISTkxOnMw2dTsfRo0fRqlUrKCgowNnZme/6khISEjh16hTs7e0xbtw4HDt2rMYxf1ZK8PLywsePH8FkMvnygRs6duyI/fv34/jx4zh48CAAwZcO0tTUhKamplATc+7fv4/MzEyqO04VLFu2DJ8/f0ZgYGC9+Q5au3Yt2rZtC1dXV6qkUCOHEpQUPFMuKAX1I9SQuXr1KkxMTFBUVIQnT55gzJgxonapwcJisXguaM4NrVq1wvHjx3Hjxg38+++/nOOKiopgMpkoLi4GnU6HjY0Nfvz4wdccDAYDR48ehYuLCyZNmsQRbH/jz4cyIyMjWFlZCb3LzPjx4+Hu7o45c+bg2bNn0NPTw8+fP/Ht2zeBzWFmZsZ39js3MJlMtGrVimpX+gexsbHYsWMHNm7cyMngrw9IS0sjKCgI0dHR2L9/v6jdoRAilKCk4BlxcXEAv7suNFUIIdiwYQOGDRuG3r174+nTp+jSpYuo3WqwEEIQExPDd/3Jmhg4cCBWrlyJVatWITw8nHNcX18fhw4dQm5uLpKSkuDi4sL3g5KYmBgOHDiAGTNmYNq0adi5c+dfr2UwGJWiNZ6ennj06JFQxRgAbNu2DV26dIGdnR2nDaWgO+ZER0cLZRWDEAImkwkbGxvQ6dTP13/ZtWsXWrdujdmzZ4valUqYmZlhzJgx2LlzJxWIaMyIcLmdooESHBxMAJBfv36J2hWRkJeXR2xsbAgAsmrVKlJWViZqlxo879+/JwDIlStXhDZHaWkpsba2JmpqauTLly8Vzs2bN4/Q6XQCgPj4+NRqHjabTby8vAgAsnnz5iqvMTAwIO7u7hWOlZWVkfbt2wukbFJNvH//nigoKJDRo0cTAOTQoUMCs/3gwQMCgLBYLIHZLCcqKooAILdv3xa47YZMTk4OkZGRqddllG7fvk0AkIiICFG7QiEkqEc8Cp4p70XcFPdRvn79Gj169MCtW7dw/vx5eHt7U5ESAVDbDjncICYmhhMnToBGo8HR0bFChH3Lli3o1asX5OTksGzZMty8eZPveWg0GrZs2YIVK1Zg0aJFWLduXaWojLi4eKUIJZ1Oh4eHB5hMJj58+MD3/Nygo6ODwMBAnDt3Ds2bNxdohNLQ0BDi4uJCibSGhYVBSUkJlpaWArfdkDl27BiKioowZcoUUbvyV/r27Yv27dtj7969onaFQkhQv4QUPFO+5N3UBOXFixdhamoKQgiePn2KkSNHitqlRgOLxUKrVq1q1QqOG9TU1HDq1Cncu3cPa9as4RwXFxdHSEgI5OTkoKCggHHjxiEpKYnveWg0GtatW4f169dj1apVWL58eQVR+bf2pZMmTULz5s3h7+/P99zcYmNjAw8PD3z//h1PnjwRmF0pKSkYGBgIJTGHyWRi5MiRnIdait/cvHkTffr0gYaGhqhd+Ss0Gg0ODg64efMmtezdSKEEJQXPNDVByWaz4e3tjZEjR8La2hpPnjxBx44dRe1WoyImJkYoCTlVYWVlhfXr12P9+vW4du0a57i6ujrOnDmDvLw8lJaWwtbWtsoe2LywfPly+Pn54d9//4WHhwfnh/RvlRJkZGQwY8YMHDp0iNM2Upj4+PhAVVUV9+/fR3Z2tsDsCiMxJzExEa9evaKyu6sgNja2zj4/tcHY2BiZmZlIS0sTtSsUQoASlBQ805QEZW5uLkaPHg1vb2+sW7cOZ8+eRbNmzUTtVqOCECK0DO+/sXjxYgwdOhROTk5ISUnhHO/Vqxe2bduGvLw8vHz5Eu7u7rWOpnh4eGD37t3w9/fHjBkzwGazq0zKKWfWrFkoKSmpk4xYCQkJTJs2DSUlJXB2duaUVaot5ubmeP/+PTIyMgRiD/gdnZSVlcWAAQMEZrMxkJmZiZSUFKFuFxEU5T7GxMSI2BMKYUAJSgqeaSqCMjExEaamprh37x4uXryIFStWUPslhcCnT5+QnZ1dpz+I5XUoZWRkMG7cuArv5dmzZ2PChAkAgKNHj2LPnj21nm/GjBk4fPgw9u/fj8mTJ/91yRv4HSl1cnLCjh07UFxcXOu5a8LU1BQAcOnSJWzdulUgNs3MzABAoEvpYWFhGDp0KKSkpARmszGQkJAA4HfFgvpOq1at0Lx5c7x8+VLUrlAIAerXkYJnmoKgZDKZMDU1BYPBQFRUFIYNGyZqlxot5dGKul6yU1ZWRkhICJ4+fYqlS5dyjtNoNOzfvx8dO3aEgoIC5s2bh4cPH9Z6PldXVxw/fhzHjh1DXFxctWLRw8MDnz9/RkhISK3nrYnymoXjx4/HsmXLcO/evVrb1NbWhpqamsCWvZOTk8FisWBjYyMQe42J8u9haWlpEXtSMzQaDdLS0o36t6MpQwlKCp5pzIKyrKwMK1asgK2tLQYPHownT57UqyLBjREWiwUNDQ2RJBSYmZlhy5Yt8PX1xfnz5znHZWRkEBYWBkIImjVrhjFjxuDLly+1ns/R0REhISH48uULoqKiUFRUVOV1Xbt2xaBBg+Dr6yv0BAYdHR3Q6XRYWFigd+/eGDduHNLT02tlk0ajwdzcXGCJOUwmExISEtSDHQVFPYYSlBQ801gFZU5ODkaMGIGNGzfCx8eHk/VLIVxYLJZI93/NmzcPNjY2cHZ2rlCup127djh+/Diys7ORn58POzs7gSxB29raonfv3sjMzMTo0aP/mvjj5eWFZ8+eVSjELgwkJCSgra2NDx8+4OTJkygrK8OECRNq3bjAzMwMT58+FUgDBCaTiX79+lH7lyko6jGUoKTgmcYoKOPj42FiYoLHjx/j6tWrWLx4MWg0mqjdavSQ/++QI8oMVRqNhsOHD0NZWRn29vYVoobDhw/HqlWr8OPHDzx58gReXl4CmbN169bo2rUrIiIiMHz4cOTn51e6pn///tDX1xd6O0bgfz29NTQ0cPLkSdy+fRvr16+vlU1zc3P8/Pmz1vvlMjIyEBkZSWV3/wUlJSUAwNevX0XsSc2UlJQgMzMTzZs3F7UrFEKAEpQUPNPYBGVoaCjMzMwgKyuL6OhoDBo0SNQuNRlSU1ORkZEh8pInioqKCA0NRVxcXCXRuHr1agwZMgQSEhLYuXMnjh49Wuv5xMXFIScnh+vXr+Pp06cYNGgQ8vLyKlxDo9Hg6emJy5cvIzExsdZzVke5oAQAa2treHt7w9vbG7du3eLbppGREcTExGq97H3hwgUAoOq+/oXOnTuDwWAgNjZW1K7USEJCAoqLixtERjoF71CCkoJnygXl38qeNBTKysqwePFi2NvbY8SIEXj48CF0dHRE7VaToi465HCLoaEh/P39sWvXLpw+fZpznE6n49ixY1BVVYWysjLc3Nxq/eNd3inHwsICt27dwsuXL9G/f/9KtSAdHR2hrq6Obdu21Wq+mtDT08P79+85ZYOWL1+OAQMGYPz48fj8+TNfNmVlZfHPP//UOjEnLCwMvXv3hqqqaq3sNFYkJSXRtWvXBiEoY2JiQKPR0K1bN1G7QiEEKEFJwRPFxcUoKCgAAKSkpCAjIwM/fvwQWP26uiIrKwtDhgzB1q1b4evrixMnTkBWVlbUbjU5YmJioKqqilatWonaFQCAm5sbHB0dMXXqVLx584ZzXElJCUwmEz9//oScnBxsbW2RlZXF9zz/LRvUo0cP3LlzBx8+fIC1tXWF2o2SkpKYM2cOjh49KtCajn/Svn17FBYWIjU1FcD/RLSEhATGjRvH98NjbRNz8vLycOvWLSq7uwaMjIxw//79et+B5v79++jQoQO1N72RQglKir/y+fNnhIWFwdvbG2PHjkWHDh0gLS0NAwMDAICTkxNUVVXRrFkzNGvWDD169MDUqVPh7++P8PBwFBYWivgOqub58+cwMTFBTEwMbty4AU9PT2q/pIgoL2heX15/Go2Gffv2oVWrVhg7dmyFhJnu3btj//79yMzMRGZmJsaPH893wsmfnXIMDAxw9+5dfP36FVZWVhUyyt3c3ECn07F7927+b6wGyisZ/Lend4sWLXDq1Ck8fPgQK1as4MuumZkZXr16xXcXnitXrqC4uJgSlDXg4OCAV69eCaS8FQDkF5Xi5edcxCbn4OXnXOQX1X41Ki8vD6dPn8a4ceME4CFFvYRQUPyHX79+kVOnTpFBgwYRGo1GABAVFRXSt29fMnfuXLJv3z5y4MABAoC4u7sTJpNJTp48STZv3kwmTZpEDAwMiKSkJAFAmjdvTmbNmkVYLBZhs9mivjVCCCHHjx8n0tLSxMDAgCQlJYnanSYNm80mampqZPny5aJ2pRJxcXFEWlqaTJkypdK5mTNnEjExMUKn08nSpUv5su/p6Uk6duxY6fjr169Jq1atiK6uLklOTq4wZ4sWLUhBQQFf89VESUkJYTAYZM+ePZXObdmyhQAgFy9e5Nnu27dvCQBy9epVvvyyt7cnhoaGfI1tSpSVlZF27dqRCRMm8G3jzdc8svp8PLHcfIe0WXKJaP/nX5sll4jl5jtk9fl48uZrHl/2d+/eTcTExEhqairfPlLUbyhBSUEIISQvL48sXLiQNG/enAAgPXv2JAcPHiRfvnypdG1mZiYBQM6ePVulrZKSEvL8+XOyZMkSoqGhQQCQf/75h4SEhIhMWJaUlBBPT08CgDg5OQnth5mCe9LS0ggAEhYWJmpXqiQoKIgAIEFBQRWOFxUVEXNzc9KsWbNqPwfVsWjRItKuXbsqz71//560adOGtGnThrx//54Q8luY0Wg0cuDAAd5vhEv09PSIp6dnpeNsNpuMGDGCNG/enOeHMDabTVRUVMjq1at59ufXr19ETk6OrF+/nuexTZHNmzcTCQkJkpaWxtO45Kx84nTwMdFeconoLLtcQUj++a/8vNPBxyQ5K5/rOUpLS0mXLl3I6NGjeb0tigYEteRNgUuXLqFLly7YtWsXpk2bhlevXuHBgweYMmUK1NXVK11fU5Y3g8HAP//8g3///RfJycm4dOkSWrVqBXt7e4wcObJC7+S6ICMjA4MGDYK/vz/8/f1x9OjRBtFVorFTnpAj6gzvv+Hs7IzJkydjxowZiI+P5xyXkJBAaGgoJCUl0aJFCzg7O+PVq1c82S5PyqkKHR0d3Lt3D+Li4rC0tMTr16+hq6uLUaNGwc/PT2j7lf+b6f1faDQajhw5AgUFBdjb2/NUi5NGo8HMzIyvxJxbt27h58+f1HI3l0ydOhWKioqYOXMm13spT0Ulo/+2CDz88Hs/cBm7+nHl5x9+yEL/bRE4FZXM1Tx+fn5ISEjAkiVLuLqeomFCCcomTG5uLsaNG4cRI0agS5cuePnyJTZt2oQOHTpUO46XskEMBgPDhg3DlStXEBYWBhaLhc6dO2Pfvn0CuYeaiImJgbGxMeLi4nD79m3MnTu33uzXa+qwWCwoKyujdevWonblrwQEBEBXVxdjx47Fz58/OcdbtWqF0NBQZGVlQVJSEjY2NpXK/lRHdb28gd91KiMiIqCgoAArKyvEx8fD09MTiYmJuHbtWq3u6W/8TVACQPPmzRESEoLY2FgsWrSIJ7tmZmZ48uQJz0I4LCwMHTp0QKdOnXga11Rp3rw59u7di/Pnz+PEiRM1Xr8z/C2WhMWhqJRdo5D8kzI2QVEpG0vC4rAzvOr3TDmJiYlYuXIlPD090aNHD57moWhYUIKyiZKdnY3+/fvj+vXrOH78OK5cuYI2bdpwNZbfOpQ2NjZITEyEo6Mj3N3dsXbtWl7d5omjR4+iV69eUFVVBYvFgpWVlVDno+CN8oLm9Vngy8jIIDQ0FGlpaXBzc6sQ+bGyssLWrVuRlZWFT58+wcXFhevI0J9JOVWhoaGBu3fvQl1dHX369IGsrCxMTEzg6+tbq3v6G3p6evjw4cNfE41MTEzg5+cHf39/nD17lmu7ZmZmyM3N5SmKW1paigsXLsDGxqZevz/qGzY2NnB0dMScOXMqdH36k1NRydh6481fz/PC1htvcPovkcr8/Hw4OztDW1sb69atE8h8FPUXSlA2QbKystCvXz98/PgR4eHhGD9+PE9f2mJiYgD4K2yuoKCA/fv3Y926dVi9ejVWrVol8FIXJSUlmDt3LpydneHo6IjIyMh6HQVrqpRneNd3OnTogIMHD+LEiRPYv39/hXPz58+Hg4MDCCFgMpnw8fHhymZ1S97/pUWLFggPD4eOjg6sra0xcuRI3LlzB8+ePePnVqpFT08PxcXFSE7++zLmrFmzYGdnh8mTJ+Pdu3dc2TU1NQWNRuOpfFBkZCSysrKo7jh8EBAQABUVFVhZWVX5N0rJLsDqC7XrXvQnqy68REp2QYVjP3/+xNChQ5GQkIBjx45R24yaAJSgbGIUFhaif//+SEtLQ3h4OLp3786zDRqNxlWEpTpWrFgBHx8frFu3Dps2beLbzp98+/YN/fv3x549e7B7924cOnQIUlJSArNPIRi+fv2Kz58/14uC5tzg4OCAmTNnYu7cuYiJieEcp9FoOHjwINq1awdlZWUsX74cN27cqNFeTUve/6V58+a4desWunbtik2bNkFVVVUo7RirKh30J+X3q6amBjs7u7/2If8v8vLy6Nq1K0+CkslkQlNTE8bGxlyPofiNsrIy7t69Czk5OVhZWeH169cVzi9jxqGUxyXumihlEyxjxnH+n5eXh8GDByM2NhY3btyAiYmJQOejqJ9QgrKJsWrVKiQkJODmzZvQ19fn205tBSUALF68GEuXLsXKlSsFEnF5+vQpjIyM8Pr1a4SHh2PGjBnUclk9pVyUNYQIZTl+fn7Q19eHnZ0dcnNzOcfl5OQQFhaG4uJiqKqqwtHRER8/fqzWFrcRynKaNWuGa9euwdTUFDk5OTh+/DjS0tL4vpeq0NLSgoSERLWCstyX0NBQvHr1CvPnz+fKtrm5OdeJOeXR3vLl7qysLDx8+BA3b95ESEgI/P39ce3aNdy9excfPnxocE0V6oKWLVvi7t27UFRURM+ePREcHAxCCN5++4HId5k875msiTI2QeS7TLxL/4GHDx/C3Nwc8fHxuHnzJnr27CnQuSjqL5SgbEI8fvwYvr6+WLt2ba1bX/H6g/g31qxZg86dO8PZ2Zmn7NE/OXToECwsLNC6dWuwWCz07t271r5RCA8Wi4XmzZtzvW+3PiApKYmQkBBkZWVh8uTJFbZqtG/fHsHBwfj27RsAwNbWltNRqip4iVCWIycnh0uXLqFPnz5gs9lcizluERMTg46OTo2CEgC6deuGnTt3Yv/+/Th27FiN15uZmeHly5c1Ji4VFBTg33//RWpqKh49eoSWLVtCRUUFvXr1wsCBA+Hg4ID58+djyJAh6Nu3L9q1a1ehqcKOHTuqXbJvSqipqeHevXsYNGgQJk2ahMGDB2P3zXiI0YXzkC1GA6ZuOorevXtDVlYW9+/fp5JwmhiUoGwilJaWwtXVFcbGxvDy8qq1PX5+EKtCQkICQUFBSEhIwObNm3keX1xcjBkzZmDq1KlwcXHB3bt3600bP4q/w2KxYGho2OAiyDo6OggKCkJYWBh27NhR4dyoUaOwbNky5OTkIDExEe7u7n/dHywuLo6ysjKe9w9LS0vj4sWLaNeuHc6cOYPjx4/zfS9VUV2m959MnjwZkyZNgpubGxISEqq91szMDIQQPH36tNK58uPu7u7Q0NDA8uXLQafT0bJlS0yZMgWnT5/Gixcv8PHjR3z+/BlpaWlITk7GmzdvcO3aNaxZswadOnXiZKC3adMGAwcOxKlTp+ptt666QllZGSdOnMDly5eRmJiI0AcJAo9OllNGgPcFkvDz88OjR4/QtWtXocxDUX+hBGUT4fLly3j16hV27doFBoNRa3uCWPIux8DAAO7u7tixYwdPUcovX76gb9++OHz4MPbv3499+/ZBUlJSID5RCJfyDO+GyOjRo+Hp6YkFCxZU2he4du1a9O/fH+Li4ggODsauXbuqtFFeKYGfKL+kpCSuX78OGo2GSZMmcVUihlt4EZQ0Gg27d+9GmzZtYGdnh/z8/L9e26FDBygqKlZ6ve7fvw9DQ0P06NEDV65cwdy5c9GmTRs4Ozvj/PnzWLduHezt7aGvr482bdpAQ0MDLVu2ROvWraGnp4dBgwZhwYIFCAoKAovFQmZmJg4dOoTCwkI4OjqiZcuW8PPzE8hqSkNm6NCheBrzHAxFNaHOw1DUwLQZszmJmxRNC0pQNhH27NkDU1NTgW1yF6SgBICZM2ciIyMDYWFhXF3/8OFDGBkZISkpCREREZg2bZrAfKHgDUII0tLScO3aNezZswd+fn6ws7PDhAkTsG3bNuzduxc3btzAly9fQAhBRkYGUlJSGqygBAAfHx+YmJjA3t4eWVlZnONiYmI4ceIElJSUoKamBg8PD9y/f7/S+PKHOn4/Q+3atYOdnR1kZGTg5OSEwMBA/m7kD/T09PDx40euBZisrCzOnDmDpKQkzJgx468RVzqdjh49enD2Uebm5mLGjBmwsLCAlJQUrl+/jo8fP8LR0RFJSUl8Z3fLycnB1dUV9+7dw+vXrzF+/HgsWLAAZmZmQsmMb0hkFdMBCHdFgABIyvr7gwVF44YSlE2ADx8+4Pr163B3dxeYTUELyk6dOsHKygp79+6t8dp9+/ahT58+0NHRAYvFgpmZmcD8oKgZQggeP36M+fPnw8rKCioqKtDU1MSQIUMwZ84crF69GmfOnMGJEyewatUqzJ49G4MGDULLli3RokUL9O/fH8Dv95CgS0bVFeLi4jh9+jQKCgowadKkCokhKioqCAsLQ05ODlRUVGBnZ4fPnz9XGg/wF6EsZ+HChfj58ycGDBiAyZMnY/fu3XzbKkdPTw+lpaVISkriekynTp2wf/9+BAcH49ChQ3+9ztzcHI8fP0ZkZCQ6d+6MY8eOISAgAPfv38fAgQMhJiYGJpMJOTk5znukNrRv3x47d+7Eo0ePUFxcDGNjY6xbt67BvudqS3Fp3SQv1dU8FPUPSlA2Ac6fPw8pKSk4ODgIzKagBSUAuLi4ICIiAtnZ2VWeLyoqwrRp0+Du7o7p06fjzp07VbaGpBAOX79+xZYtW9C5c2eYm5vj7NmzUFdXh4eHB86dO4d3796huLgYP378ACEEhBD8+PEDRUVFePPmDc6ePYu5c+eCzWaDRqPB1tYW3bp1w/bt25GRkSHq2+OZ1q1b49ixY7hy5Uql/b9GRkbYs2cPvn79isLCQtjZ2VXYzlHbCCUAGBsbw8LCAj9+/MD8+fMxa9asWpcT4qZ0UFVMmDABbm5umD179l8jgWZmZsjOzsagQYOgq6uLhIQEzJ5dcXk0LCwMQ4cOFWiprx49eoDFYmH58uVYtWoVFi5c2CRFpQSjbn7u62oeivoH9ZdvAsTExKBbt26QkZERmE1hCEpzc3MAQGxsbKVzaWlpsLKyQnBwMA4fPoydO3dCQkJCoPNTVM2vX7+wbNkytG7dGitXroSBgQFu3ryJT58+4fTp01ixYgVGjRqFdu3agU6v/JUiJiYGPT092NraYtWqVWjfvj0sLS1x5coVdOjQAYsWLYKmpia8vb1RVFQkgjvkn8GDB2P58uVYvnw5IiIiKpybPHkypk+fjvz8fDx9+hSenp6cc/x2m/oTLy8vPHr0CPb29li6dCm8vLywYcMGvu1pampCSkqKZ0EJANu3b0enTp1gZ2dXZTZ3eda7jo4Orl69WqnZwKdPnxATEyOU3t3i4uLw9vZGQEAAfH194eHh0eREZRtlWSEveP9eUG+jLCvkWSjqK5SgbALExsYKvIC0MASlnp4eZGVlKxSOBn53zTAyMkJaWhoiIyPh6uoq0Hkp/k54eDj++ecf+Pr6Yvny5fjy5QtOnDiB/v37VykeuYHFYsHY2BhDhgxBaGgoPn/+DC8vL6xfvx6GhoZ4+PChgO9CuKxZswaWlpZwdHTklA0qZ8eOHejevTvk5eWxa9cuHDlyBIBglrwBYMSIEdDV1cW2bduwYcMGrF27FitWrMCKFSv4Ekx0Oh3t2rXjS1BKSUkhNDQU6enpmDp1aoX54+LiMH78eMjKyqJnz55VPtwymUxISEhg6NChPM/NLbNnz8aePXvg7+9fK+HdEJGVZEBLSXBBharQUpaBrGTtkz4pGiaUoGzkFBUVITExka+OONUhDEFJp9PRrVs3zpIZIQS7du2CtbU1OnbsCBaLRXVcqCMIIVi8eDGsra2hrq6OZ8+eYc2aNWjevHmt7Jb3vf5vQo6Kigo2btwIFosFWVlZ9O7dG97e3g0mgsRgMHDixAmw2WxMmDChQi9sSUlJnD17llMGx93dHTExMQJZ8gZ+f2Y8PDxw9uxZJCUlYeXKldiyZQs2bNiABQsW8PUa8pLp/Se6uro4fPgwQkNDORnuJSUlcHFxgY6ODmxtbREVFVXlWCaTif79+6NZs2Z8zc0t7u7uWL58Oby9vfH8+XOhzlXf6NtBVXh1KOk09G2vKhTbFA0DSlA2cn79+gU2mw0lJSWB2hWGoAR+1037+fMnfv36BVdXV8yePRuzZs3CzZs3oapKfVnVBYQQzJs3D5s3b8aWLVsQERGBTp06CcR2dR1y/vnnHzx69Ahr1qzBmjVrsGzZsgYjKjU0NHDy5EmEh4dj3bp1Fc61bt0aISEh+PLlCxQVFWFra8spsSOIz5CLiwsUFRXh7+8PAFiwYAECAgLg5+eH2bNn89xJpjaCEgDGjBmDefPmwdPTE1FRUdi8eTOeP3+OoKAgWFhY4MWLF5VKDKWnpyMyMrLOenevXLkSHTt2hIuLi1C+x+orE3poCa8OJZvAyUxLKLYpGgaUoKTgC2EJSuC3CLawsMDp06cRHByM7du3c5YIKYQLm83GrFmzEBAQgH379mHBggV8L21XBYvFgry8PHR1das8LyYmhlWrVsHPzw8+Pj5YtGhRgxGVffv2hbe3N9auXYubN29WOGdtbQ0fHx98/foVOTk5WL9+PYDaL3kDgIyMDGbMmIFDhw7h+/fvAH4v7R44cAB79uzB1KlTK0RNa0JPTw9JSUm16ly1efNmGBoaYvTo0fD29saiRYtgYmICMzMzsNlsREdHV7j+woULoNFoGDlyJN9z8oKkpCSCgoIQFxdXqUB9Y0ZPTR4WuioCj1KK0Wmw0FWBrqq8QO1SNCwoQUnBF4JqvfgnmZmZiIiIQGZmJh48eAAnJyeBz0Hxd3x9fbF3714cOnQI06dPF7j9mJgYGBgY1ChSPTw8sGPHDmzduhV79uwRuB/CYtmyZRg4cCAmTJhQqdf2woULYWtri9LSUrBYLACCiVACvwVkcXExDhw4wDk2depUHD16FEeOHMGkSZO4/rzq6emBzWbX2I+8OiQkJBASEsKp2LBs2TIAQOfOnSEvL1+pr3dYWBgsLCzQokULvufkFSMjIzg5OWHnzp1Nqh/4Rht9MAQsKBl0Gjba6AvUJkXDgxKUjRxJSUnQaLQae+jyiqBaL5ZDCMH27dvx6NEjKCoqIjo6WuCJRBTVk5iYiJUrV8LT0xOTJ08WyhwsFovrguZz5syBm5sbFi1aVCtxU5fQ6XQEBwdDQkIC48aNqyDiaDQaAgMDoaWlxdmLev36dYHMq66ujvHjx2PHjh0VPpdOTk44ffo0QkJC4ODgwFXUkd/SQX+ipqYGCQkJlJSUcB4KxMTEYGpqWqFjTm5uLm7fvi2U7O6amDFjBpKSkgT2d2gItFaSgffILgK1uXZkF7QWcsIPRf2HEpSNHGlpaejq6gp887kgl7wLCgrg5OQEDw8PSEpKYs6cOVBRURGIbQruKCsrg6urFhGaiAAANINJREFUK9q0aVNpD6CgyMnJwYcPH3h6UNiyZQtUVFQwZcqUBhNFatGiBUJCQvD48WMsX768wrlmzZohLCwMv379AgB4e3sjMTFRIPN6enoiNTUVISEhFY6PHTsWYWFhuHTpEmxtbWvsb92yZUvIyMjgzZs3tfInLCwMeXl5mDZtGpYuXcrpGGRmZoZHjx5xtjJcuXIFxcXFIhGUpqam6N69O1cNFRoT40y0sGBge4HYWjiwAxxMqL2TFJSgbBIYGhpWWduxNghKUH78+BG9evUCk8nE9u3bUVRU1KBb8jVUjhw5gqdPnyIwMBDS0tJCmaP8PcjL31deXh6HDh1CeHg4zpw5IxS/hEHPnj3h4+ODzZs34+LFixXOderUCf/++y+A36V2bGxsBLKCoK+vj4EDB8LPz6/SvtMRI0bg4sWLuHPnDkaMGFFt320ajQZdXd1aRyhv3LgBAwMD7N69Gz179oSDgwMyMjJgbm6O9PR0TjceJpMJIyMjaGnVvSih0WhwdHTErVu3eNpn2hiY3VcPPrb6kGTQed5TKUanQZJBxyZbfczqW/V+aIqmByUomwAGBgaIjY2t1Sb7PxGEoLx58yaMjY2Rl5eHx48fQ01NDcBvfynqlt27d2Po0KGc4vLCoLwsUPv2vEVG+vXrBysrK4G0FqxLPD09MWrUKDg7O1dqZThs2DAAv6O2qampldo31mbOmJiYSkXWAWDgwIG4evUqHj16hCFDhlQrYmub6Q383i9rbGwMBoOBU6dOoaSkBE5OTjA2NgYAPH78GL9+/cKVK1fqLLu7KoyNjVFQUFDriGxDZJyJFm55WKGnjjIA1Cgsy8/31FHGLQ8rKjJJUQFKUDYBhg0bhp8/f+L8+fMCs1kbQUkIwZYtWzB48GCYmJggKioK//zzD44dOwYjIyOqnWIdExUVBRaLhRkzZgh1HhaLhe7du1dotcctM2bMQEREBBISEoTgmXAo3zOpoKAAe3v7Cg905VUL/vnnHzAYDJw/fx4+Pj61nnPgwIHo2rUrfH19qzxvZWWFmzdv4vnz5xg4cCAnK/xPaisoCwsL8fLlS872hpYtW+LEiRO4efMm9u/fD11dXTx69Ai3bt1Cfn6+SJa7yyl/gP2zoUJTobWSDIKn9MDN+ZaY2EMb2soylTrq0ABoK8tgYg9t3PKwRPCUHtSeSYrKEIomQe/evYm1tbVAbP0sLCF20z2IwQAbEp/2nfwsLOF+7M+fxN7engAgS5cuJaWlpYQQQj5+/EhoNBo5ePCgQHyk4J6pU6cSLS0tzt9CWOjp6ZG5c+fyNbaoqIi0aNGCzJ8/X8BeCZ+oqCgiISFR4d7T0tIIAHL8+HGiqalJWrVqRWg0Grl27Vqt5zt8+DABQBITE/96TXR0NFFSUiIGBgYkIyOj0vlDhw4RGo1Gfv36xZcPz58/JwDIgwcPKhxfs2YNodFopH///sTY2JjY29uT1q1bE19fX+Li4kKMjY1J27ZtiYyMDAFAZGRkSNu2bYmxsTFxcXEhW7duJdevXyefP38mbDabL9+qonXr1mTJkiUCs9fQ+VlYQuLTvpOYT9k8f8dTNF1ohDSQIm8UteL48eNwcnJCYmIiOnbsyPP4t99+4PiTZIS/TkdydgH++6ahAdBSkkHfDqqY0EMLempV1yJ7//49bGxs8OHDBwQFBWHs2LGcc8uWLcOuXbvw+fNnyMpSvWDrkvbt22Pw4MFCrceXm5sLRUVFTgkbfpgyZQpiY2MbZCRp165dmD17NkJDQzF27Fikp6dDTU0N586dg7q6OiwsLKChoYEfP36AxWKhbdu2fM9VVFQEbW1tjB49utpkkxcvXmDAgAFo0aIFbt26VWFlIDIyEpaWloiPj0eXLrxnBEdHR8PExASxsbEVunSVlZXB2toaT548qdC3XVpaGl26dIG+vj7U1NQgJiaGS5cuYfjw4SgrK8O3b98QHx+Ply9fcnqCd+zYEZMnT8bEiRNrvarRqVMnDB069K+RXQoKCi4QtaKlqBt+/fpFNDU1ycCBA3l6sk/OyidOBx8T7SWXiM6yy0R7yaW//is/73TwMUnOyq9g5+rVq0RRUZHo6uqS+Pj4Cufevn1LpKWlyYIFCwRyrxTck5ubSwCQoKAgoc4THh5OAJC4uDi+bezcuZOIi4uToqIiAXpWN7DZbGJvb0/k5eXJ27dvSXZ2NgFAQkNDCSGE7N27lwAgqqqqpHv37iQ/P78Gi9Wzbt06IiUlRdLT06u9LjExkbRs2ZK0b9+epKSkcI5/+fKFACBMJpOv+aOioggAEhsbyzkWHh5ORo4cScTExAgAzr+zZ89yHR0vLS0lb9++JWfOnCGOjo5EUlKSiImJkZEjR5Lw8HC+fCWEkI4dOxJPT0++x1NQUBBC7aFsIkhJSWH//v24ceMGDh06xNWYU1HJ6L8tAg8/ZAFAjS27ys8//JCF/tsicCoqGYQQbNy4EUOHDkXPnj0RFRVVIeLBZrMxefJkqKurY/Xq1XzeHQW/lJeTEnYiVExMDKSlpfmKjpdjYGCAkpISvHz5UoCe1Q00Gg0HDhyAuro67OzsOPuPy+tUTp8+Ha6urvj+/TtevXoFNze3WnUImjFjBmg0Wo1F4Tt27Ih79+6hsLAQlpaWnOQhNTU1yMnJ1ToxB/jdv93V1RV9+/ZFcnIyduzYgQsXLgD4XSfXxsaG6321YmJi0NXVxZgxY3DixAl8+fIFO3bsQEpKCvr27QsXFxdkZWXV2mcKCgreoQRlE2LIkCGYPHkyPD09K2Wd/snO8LdYEhaHolI2z71fy9gERaVsLAmLQ6/p67B8+XKsWLECFy9ehKKiYoVrd+zYgcjISBw+fBhycnI83hFFbXn9+jUACKxX998oT8hhMBh82yh/EHn16pWg3KpTmjVrhtDQULx69YrTOaZcWNJoNOzatQtdu3aFvLw8jh07hp07d/I9l7KyMlxcXLBr164a6062a9cOkZGRoNPpsLCwwNu3b0Gj0WqVmKOgoAAACAkJQadOnXDu3DkcPHgQMTExmDlzJoYNGwYxMTEUFRXh6tWrfM0BAM2bN8fMmTPBYrFw8OBBnD9/Hp06dcKpU6e4tkEIQXp6OsdnCgoK/qAEZRPDz88PKioq6NevH5KTk6u85lRUMrbeEEwJjc8qRli87xzWrl1bqd1ecHAwvLy8MH/+fPTp00cg81HwBpvNBo1GE3qvdBaLVevORxISEgDQYAqcV0W3bt2wc+dOzirBfzvpSEtL4+zZsygrK0ObNm3g6emJyMhIvueaP38+MjIycPz48Rqv1dLSwr179yAnJwdLS0skJCTUSlC2a9cOEhIS+Pfff2FlZYXExERMmTIFNNrv/OHo6GiUlZVBXFwcEydO/Ot3EbfQaDRMmTIFiYmJsLKygqOjIzZs2MDV2OTkZGRnZ1OduSgoagklKJsYCgoKCA8PByEEVlZWlSKVKdkFWH1BsEuKzBQJpGQXVDgWGBgIZ2dnuLq6UhvhGzk/fvzAmzdvqIL1/8/kyZPh7OwMAPj06VOFc23atMHJkyeRlJSEVq1awc7ODp8/f+Zrnvbt22PEiBFVFjqvipYtWyIiIgKqqqqwsrKCvLw834Jy/fr1KC4uRteuXREaGlopaSYsLAzNmjVDSUkJZGRkKpVV4hd1dXWEhobC29sbK1asgLe3d433Xl5wn6p/S0FROyhB2QTR1tZGREQEGAwGLC0tKxRBXsaMQymPS9w1UcomWMaMAwAUFxdj3bp1mDx5Mtzc3LB///5KkUuKuoUQUqv9ejXx7NkzEEJqLSiF6WNdUr68DQD79u2r1LVm4MCBWL9+PT59+oTS0lKMHTuWb7Hl5eWFhIQEXLt2javrVVVVER4eDm1tbZw8eRJpaWmcrGpuWb9+PVavXg1zc3Pk5uZWiigTQhAWFsYp7j5z5kzExMRg8eLFPM1THatWrcKGDRuwZs2aGluJPnz4EKqqqmjZsqXA5qegaIpQv+RNlNatW+Pu3bvQ1tZGnz59MG3aNES/TUPku0ye90zWRBmbIPJdJs7eegAjIyN4e3tj5cqV2L17NyUmRUx5d6K0tDShzcFisSApKVnrfZrly6LlPjdkZGVlISkpiezsbMyYMaOSWF66dClGjhyJwsJCREdHw8PDg695LCwsYGxsDD8/P67HKCkp4fbt29DR0QGASr3Bq+P+/ftYtWoVVq9ejS1btiAlJQW3b9+ucE1CQgLevn2LCRMmoE2bNsjMzISvry+2b9+OsLCwv9rOLyrFy8+5iE3OwcvPucgvKv3rtcDvUmRr1qzBmjVr/rp1oLi4GEePHoW9vT1nOZ6CgoI/qDqUTRw2m40DBw5g0aJFkO8zBeKdrUEq9UmoPTTCRh7rMnS+s3DgwIEKtekoREdycjK0tbVx4cIFjBgxQihzTJo0Ca9fv8aTJ09qZefkyZMYP348srKyoKSkJCDvRIeCggKGDx+OEydO4MCBA5g6dWqF87m5uTAxMUFBQQHS0tIQGBgIFxcXnucpf92ePXuGbt26cT3u48eP0NHRgaSkJK5cuQJra+tqry8oKED37t2hoqLCSfLp1q0b9PT0cPbsWc5169evx6ZNm5CRkYHJkycjKSkJDx48gL29PW7cuIGYmBi0a9cOQO3r35aVlcHS0hLp6el4/vw5ZGQqdncJDQ2Fvb093/U2KSgo/gcVHmri0Ol0uLm5ISEhAbJ6pkIRkwBAaHRo9hiER48eUWKyHtG6dWsoKysLtVg4i8USyP7JmJgYaGtrNwoxCfxuv6ivrw83NzfMnj0bz549q3BeQUEBYWFhyMnJQbt27eDu7g4Wi8XzPGPHjkXr1q15ilICv/dzNmvWDFpaWhg2bFiNy+YrV65ESkoKAgMDISYmBhqNBnd3d5w/f77CXtHy5W4pKSmYmZkhJiYGxcXFOHjwIFRVVWFnZ4d3X3Iw8dATDNh+D8FPPuHTH2IS+F3E8lN2AYKffMKA7fcw8dCTSnu1xcTEEBgYiNTUVKxYsaLieEIQEBAACwsLSkxSUAgASlBSAAAUVNRQLF51hxtBkVcmgaIyoU5BwSM0Gg0mJiYIDw8Xiv38/Hy8evVKIBm0d+/ehbGxsQC8qh8wGAyUlJRg+/bt6NSpE+zs7JCXl1fhmq5du+LQoUN4//491NTUYGtri8zMTJ7mERcXx7Rp03D8+HFs3LgRc+bMgaurK/755x/QaDRMmjQJ06dPx9q1a8FkMvHu3TuUlZWBRqOhffv26NmzJwYMGICRI0fi3LlzVc6RmZmJXbt2YdmyZejQoQPn+MSJE6GmpgZ3d3cQQpCUlITY2FhO724zMzMUFRXh2bNnUFBQQGhoKD7SNTDQP7JW9W//S/v27bF8+XLs3r27wmt39OhRREZGYsmSJTy9nhQUFFVDCUoKAMCnrPxKEQBBQwAkZeXXeB1F3eLk5ISIiAhOTcra8t+9bufvsUDEJGodoYyJiUF0dDQmTpwoEB/rA+Li4igpKYGUlBRCQ0ORnp6OqVOnVtpPOW7cOMyfPx9paWnIy8vDuHHjKpQbqori4mIwmUzY2tqiVatWWLVqFcrKyrBy5UqEh4fjzZs3SElJAQCOyAsICICtrS309PQgLy8PU1NTFBcXIzExEWfOnMHo0aMxduxYnD59utJ8QUFBIITA3d29wnF5eXkcOHAA165dQ2BgIJhMJiQkJDB06FAAQPfu3SEpKYnHjx8DAO7nyEJhwEyUgV6r+rc7wytmp5cL2qCgIAC/9wzPmzcPEydO5PhCQUFRO6g9lBQAgNjkHNjseSj0eZgzesJAq7nQ56HgnsLCQmhqasLZ2ZnvEk7V7XUjhEBbSQbWHdWq7fVeHW5ubrhy5Qo+fvxYq+Lo9Ym2bdvC0dERGzduBACcPXsWY8eORUBAAGbPnl3h2pKSEvTv3x9xcXHIy8vDggUL4OPjU8lmYmIiDh48iODgYGRkZMDY2BiDBg1C165dceXKFVy8eBGpqamQlZWtNJYQgm/fviEuLg5xcXF4+vQpzp49i9LSUgwZMgTOzs64ePEiTp48icOHD3NKH7HZbHTo0AGmpqZ/rXnp6uqKsLAwtG/fHmpqarh06RLnXK9evaClpYVRnpuwJCyO79fzTzbZ6sPBRIvzfycnJzx58gTx8fEYNWoUXrx4gZcvX6J5c+r7iIJCEFARSgoAgASjbt4KdTUPBfdISUlh8uTJCAwMRHZ2Nk9jU7ILatzrRqPRkJzzq9q9btXx9etXHD9+HNOmTWs0YhL4X4SynDFjxmDevHnw9PREVFRUpWtPnz4NaWlpaGlpYdOmTRUSXX7+/AkPDw907doVR48ehZOTE168eIGoqCisX78e48aNw7p16/Djxw9OlO5PaDQa1NXVMWDAAHh6euLUqVOc8kaZmZkYN24cEhISMGrUKLi4uGDfvn0Afhcpf/fuHaZNm/bXe922bRtUVVURHR2Nnj17Vjhnbm6ORy/eCLz+7aoLLyu8z6ZNm4Z3795h8ODBuHPnDgIDAykxSUEhQKhfdwoAQBtlWSGl4/wP2v/PQ1H/8PDwACEE8+bN43qMIHq91wQhBDNmzICsrCxmzpzJtW8NAXFx8UpL15s3b4ahoSHs7OyQk5NT4Vx50e7k5GS0b98eLi4unBqTXbt2xb59++Dj44O0tDT4+flBX1+/wnhtbW2MHTsW27ZtQ1kZd5uZy7PC9+3bhydPnoDNZuP8+fMwMDCAu7s7/P39wWKxwGAwYG5u/lc7ioqKcHNzAwD4+/sjMTGRc87MzAy/9EejtEywHZD+W/8W+F24nEajITIyEmFhYRg0aJBA56OgaOpQgpICACAryYCWkkzNF9YCLWUZyEo2nghTY0JDQwP+/v44duwYLly4UOP1gur1/udetz85deoUzp07hz179kBFRYWneeo75Uk5/0VCQgIhISH48eMHnJ2dKxUF79mzJ7Zv3443b95AUVERFhYWGDJkCPT09BAfH4+FCxdyWlRWhaenJ96/f8/V3xgA9PT0AABv376FqakpoqKisHHjRiQmJkJFRQXz589HYGAgunTpAklJyWpt3b59Gz179oSqqiosLCxw/PhxEEKg1r47pNsaokzAm6/K69++S/+BhIQEDB48GADQv39/DB8+XLCTUVBQUIKS4n/07aAKMbpw4pRidBr6tlcVim0KwTBx4kQMGzYM06dPx8ePH/96nSB7vW+98Qan/xKpfP36NWbPng0HBwfY2toKZL76xJ9L3uVoaWnh6NGjuHjxYpV7WmfNmoUJEybgy5cvyM7ORqdOnXDt2jVOIfLqMDU1Re/evbneK6ukpAQlJSW8evUKiYmJuHDhAlRUVDBnzhz8+PEDsrKyiIqKQl5eHj59+vTXbka5ubm4ffs2HBwccPfuXfTv3x9OTk4YOnQomHEZIGzhlH8Qo9MwZ0counfvjqysLAwdOhQZGRlCmYuCoqlDCUoKDhN6aAm8S045ZWwCJzOtmi+kEBk0Gg0HDx6EvLw8rKys8P79+0rXCKPX+5973YDf3VSsrKygrq6OnTt3CnS++gKDwfhrtvawYcOwZMkSLF26FPfv3690XkVFBWVlZZCXl0diYmKVCTp/w8vLCw8ePKi20Hxubi4OHDiASZMmobCwEN7e3ujcuTPGjv2/9u48rspqbeP4bzMWogQ4JAUkiYqIimmWA4hTzoqZSkZl5ZSaotnBSjHNOqaJUzmc9zhkFqVGaDZYEYhZakoqpKipiIojZgqJAvv9gwNFgIB7bzS9vn/yPHutBz4lF+tZ67778dxzzzFz5kyys7ML20YePnyY++67D2dnZ9q0acOIESNYt25dYWDesGEDV69eJTg4GFdXV6Kioli/fj1JSUl8vPkXDFbW5X7+isjNM7LnbC4vvfQSu3btwt/fv8Jll0SkfBQopZB3raq0rVvd7KuU1lYG2tatTt2alq1zKaa7++67iYuL48477yQwMJD9+4uuRFq61ztAUlISQUFB1KxZk7i4uFvuVXeB0lYoC0ybNo1WrVoxYMCAIqtqH374IXPnzmXSpElYW1tTr149Jk2aVO5+3T179uT+++8vVug8Ly+P7777jtDQUGrXrs3w4cNJSUnBzc0NT09PYmNjOX36NLm5uYX93w8dOoS1tTXVqlUDwMfHp7Cta+/evXF3d2fChAmsWLGC5s2b4+7uXjhfjx492J64G5u7LNtK08bZjYmTpnDHHXeovaKIBSlQShFvBPthY+ZAaWNl4I1gv7JvlJvCPffcQ1xcHNWqVePBBx9kyZIl5OXlceDURYv2ek9Jv8C8efN4+OGHcXNzIzY2lho1aph1rptJSYdy/srGxoaoqCiuXr3KE088QW5uLidPnmT06NEMGDCAqVOnsmrVKvbv30+9evV4/PHHOXToUJnzWltbExYWxpo1azhy5AgAW7ZsoUmTJrRv356tW7cyefJkjh49ytatWwkNDeXSpUsEBQVRo0YNrKz+/LVRp04dqlevzu+//86AAQPYunUrjo6OJCUlkZiYyIABA1i6dCkbN24kMzOzsPZlgXNXrMDixwFV/1akMihQShHuLg681su8bcim9vLF3cIHfsS8ateuzffff0+/fv0YNmwY7dq1Y8FXuyy2x9bKAMEvzWbs2LE8+eSTxMfH37IrkwVKOpTzd25ubnzwwQd8/fXXTJ8+neHDh2NtbV24DaBbt25ERESQkpLCHXfcQXBwMFlZZZdkevrpp3FycmLmzJmMHDmSNm3acOeddxYWuA8PD+eee+4B8jvNnDlzhgsXLpQ4lqOjI02bNmXdunXMmDGDpUuX8vTTT9OoUSPmzp3LkiVLgPzSQw0bNmTBggWFp8yv5Jj3ZHdpCua5fPnyLVV6SuRmokApxQxs4cGLneuZZawJnesXKS4s/xzOzs783//9H7GxsaSnp7N2yz6L7bHNM8Ifd3mxefNm3nnnncJXqLeysl55F+jYsSMRERFEREQQExPDggULioTtyZMn061bNzIzMzlw4ABDhgwp9XBMgSpVqhAcHMzChQtZsWIFkZGR/PDDDwQEBBR7LfzXk96QX8opLS2Nbdu2kZCQgKOjI9bW1jg5OREfH8/KlSuJiooiJCSEK1eusH79enx8fNi/fz+DBg1i9OjRBAUFcf78+Uqvf5uUlISPj0+lzClyu1GnHClV1PajRKxLJifPWKEgYW1lwMbKwNRevgqTt4izv12k+Yx4LPl60gAkTXnktikt1adPH3Jycop0jSlNbm4u9957L2fOnCE1NbVw9bDA+fPnad68Obm5uaSmpjJ37lxeeOGFUsfbsWMHHTt25MKFC0ycOJHp06eXeu/u3btp0qQJHTp04PLlyyQlJZW6Wgn5K69ubm4cO3YMb29vjh8/zujRows7Am3atIm+ffvi4eFBzIYvCZi33aJtXwv+u3Kws+buu+9m2LBhTJ061YIzityetEIppRrYwoNvwgJp5eUKUObrzoLrrbxc+SYsUGHyFnIqKw9L73W73Xq9l+eVd4Hz58+TkZGBg4MDgwYNKrb30tnZmU8++YTTp0/TsGFDxo8fz6ZNm0oca9u2bXTo0IH69eszcOBA3nvvvWLPkZWVxfvvv0/79u0Li5snJSXh6elJeHg4n332GT///DN79+5l7ty5QH7rSFtbW3r06EG3bt3w8fEhJSWFS5cusWLFCl5//XXS0tIICAggNjaWtLQ0enTpxD1O165faaqC+rcnTpzg9OnT+Pv7W3Q+kduVAqVck7uLAyufbcnXYwMIbemJp6tDsVhhADxdHQht6ck3YQGsfLal9kzeYip7r9vtoKxDOX/1wQcfALBq1So2b95MREREsXuaNGnCkiVL+OWXX/Dy8qJ///4cP368yD3p6el07doVX19fNm7cSHh4OMeOHWPNmjUAZGdnExERQe3atQkNDQXgvffe46GHHqJTp06sWrWK8PBwunfvTpMmTWjQoAEhISHY2tqSmppa2J5x4cKFJCUl0atXLyB/hfXNN9/E09OT4OBgnJyciIuL4+TJk5z/ZXOl1L9dvXo1tra2tGnTxiJzidzu9MpbKiwzO4cj5zK5kpOHnY0V97lWuW1eU96ukk9coPv84vUQzW3D6Db4ujlZfJ6bQWhoKKmpqaWuJP7VwIEDOX78OAkJCbz11lv861//YsOGDXTr1q3YvaNGjWLx4sW4uLjg5eVFXFwc9vb2GI1GevfuzbZt20hOTsbVNf/NQ6dOnTh//jyRkZEMHTqUX3/9lbFjxzJ8+PDCYumDBw9m7969/Pjjj6U+X2JiIhEREQwaNIizZ8/i7OyMu7s7rVu35ssvv8THx4cnnniCt956i/PnzzN9+nSaNGlC58eewu25d034SV7bN2EB3F/DER8fH5o2bUpUVJTF5hK5nSkFSIVVsbe5bX7pS76CXu+W3ut2q/d6P336NHv27OHYsWMcOnSIlJQUpk2bhpubG46OjjRo0IAGDRoUa2O4c+dOunbtCsCLL75IQkICoaGhJCYm4uFRdGvJ7NmzSUxM5MCBA+zcuZOxY8eycOFCVq1axfr164mOji4MkwBjx46lR48eBAQE8PDDD5OYmIivb9FKD97e3tds1zhixAjatWtX+Oo8MTGRqlWrcuLECUaOHMmECRN45JFHWLFiBQkJCbz99tuEhYXRokULQnt15PPUn7nTsynmXJ+2tjLQysuVujWrEhcXR0pKCosXLzbjDCLyV1qhFJFyCZz5HakZZZekuV6erg7EvxhksfErm9FoZNOmTcTExLB792727NnD6dOny/ycjY0N9erVw8/Pj5YtW9KjRw/q1avH8uXLeeqppwDIyMjA398fNzc34uPji/XvPnHiBM2aNcPJyYn9+/fz7rvvMmnSJB555BFWrVpVeF9OTg6hoaFERUXRqFEjdu3aVaTOZIHVq1fTv39/zp07h4uLS4nfa+PGjalatSq7du1iypQpnD17lmXLlpGeno61tTW7du2iY8eO1K5dm6+//ppDhw7x+OOPY3RwgYeehJp1yf+TxTyvv+1trPgmLBA3J3sCAgK4cOECe/bsUXFzEQvRHkoRKRf1ei+fY8eOMX36dLy9vWnXrh1r167FycmJESNGsHr1avbt28fly5cZMWIETZo0wWg0kpubS0ZGBps3b2b+/PkEBQWRnp7OxIkTadCgAZC//7FgBdDFxYWPP/6YHTt2EB4eXuwZ3Nzc+Pjjj/n1119p3LgxL7zwAhkZGUybNq3wnqtXrxISEsKaNWsYMWIEycnJHDx4sMTv6e+lg/7OYDCwaNEifvzxR5ycnDh06BCffPIJvXr1wto6v61ikyZNiI+P5+zZs7Rr1w4PDw8mLIyGbq9irF6nYKTr+pmXpKD+7Zw5c/jhhx9YtGiRwqSIBSlQiki5qNf7tf32228MGzYMT09P3njjDdq0aUN8fDxHjhxh7dq1TJkyhX79+lG/fn3s7e2xs7MrPJRjZWWFs7MzrVu3Zvjw4SxYsID4+PjCUAkwceJEvLy8iImJAaBly5bMmjWLyMhIoqOjiz1PQEAAs2bNYvfu3VhbW2Nvb0/Vqn+2Px0+fDgxMTGsWbOG2bNnU6NGDSIjI0v83urWrQuUHigBWrduzdixYzl58iSpqakcPHiQvn37FrmnYcOGxMfHk5mZSdthr/NW/HGwsTN7L++C+rcpKSm8+uqrjBkzRodxRCxMgVJEykW93kv3ySef0LBhQz788ENmz57NyZMnWb58eYmFwguUp2yQs7MzISEhAKxcuZKmTZvSp08f+vXrR3p6OqNHj+bRRx9l8ODB/Prrr8U+P2bMGDp37kx2djY2NjaEhISQk5PD+vXrWbp0KYsWLaJ3797ccccdjBo1iuXLl3P27Nli4zg6OlK7du1rBkqA119/HTs7O2JjY6lSpQodOnQodo+3tzcvLY4hz6/HNceqKGsrA/Y2Vszo68fIoLqkpaXRvXt33N3dr1lnU0TMQ4FSRMpNvd6LunLlCoMGDeLRRx/lwQcf5JdffmHMmDFFVgJLU95OOQW8vLxYt24dH330EQkJCfj4+PDVV1/x3//+l+rVq9O/f38uX75c5DMGg4GgoCAMBkNhqZ5x48YxbNgwunXrxuDBgwvvHTFiBACLFi0qcX5vb+8yA6WDgwN16tQp/L5OnjxZ7J60jCzmbyn+dVM1c7+rsP5tamoqgYGB5OTk8NVXX+HgoDJmIpamQCki5aZe73/Kzs6mX79+rFmzhg8++IDo6Gjuvffecn++vHUo3d3dgfzXzQaDgf79+7N3717atm1L79692bRpE6tXryY5OZlx48YV+3xycjKNGzfm4sWLNGzYkPnz53PhwgWWLFlSZPW0evXqPPXUUyxYsKBYMIXyBcrc3FyOHj1KXl4ejo6OBAYG8vPPPxe55+XoPeSYeeuEtQHusLXG3cWBn3/+mcDAQAwGA/Hx8dSpU6fsAUTEZAqUIlIh6vUOly9fpm/fvmzcuJGYmBhCQkIqfOCjvJ1yqlWrhre3Nzt37iz8mouLC2vXrqVHjx48+uijHDlyhHnz5rFw4UI+/PDDIp/fuXMnrVq1YuXKlYWnnK9evcpvv/1WbK6wsDBOnTpVWEj9rwoC5bUKgxw8eJCsrCxsbW2Ji4vDycmJ5s2bEx4ezh9//MGBUxdJOHjW7Htxc42QcPAsI8Jfo3nz5lSrVo24uDg8PT3NOo+IlE6BUkQqbFSQN//u64e9jVWF91T+fa/bP9Hzzz9PbGws69evp0uXLtc1RkU65TRr1qxIoASws7MjKiqK4OBg+vfvzwMPPMCgQYMYMmQI+/btK7xv3759+Pr60qtXLzp16oTRaKRWrVoEBwcX68ldv359evbsyezZs4sFR29vby5cuFDiHssCBc8YFBREgwYN2L59O6+99hpz5szBz8+PqR9+Z7FKAeTlsubnU7z22mv89NNPhSu7IlI5FChF5Lrcrr3eN2zYwLJly1iwYAGdOnW67nEq0sv74YcfZtu2bcXCnK2tLe+//z6NGjVi8ODBzJs3Dw8PD/r160dWVn7N0Ly8vMJOOWlpadSqVYvMzExOnTrFk08+SV5e0XLi48aNIzk5mY0bNxb5elmlg4DC0+YDBw4E8kPvK6+8wq5du3B3dyd23ymLVQrAyhrvgD688sorxepyiojlKVCKyHW73Xq9nz9/nqFDh9KlSxeeeeYZk8aqyKGcxx9/HIAVK1aUOM6yZcvYu3cvc+bMYc2aNRw+fJiRI0cWuS81NZV9+/Yxa9YsqrnUoIa3P1/9lML4aZFkZv+5UhoYGEizZs14++23i3z+/vvvB0oPlOfOnSM6OhqDwUDPnj2LXKtfvz7rv/waW2e3cn2/1yv94tUi34uIVB61XhQRk3nXqsqUXr5MwfeW7vU+bdo0Ll26VOxAy/WoyCvvGjVq0K9fPxYvXkxYWFixbjZNmzbl1VdfZdq0aTz99NMsWrSIJ598koCAACC/k83n3yfi3HEoC4/VgsdmkwPcDURfhugpX+Hp4kBQ/ZoMaunB+PHjGTRoELt376Zx48ZA/gnue++9t9RAuWLFCnJzc2nVqhXVq1cvdj31XGb5fzjXyQgcOZep1rAiN4BWKEXErAp6vft7OOPr5nTLhMnMzEyWLl3KyJEjzbI/ryKvvCG/rM+BAwdYu3ZtidcnTJiAo6MjS5YsITQ0lOeee47nn38eZ3dvlh2uwltJdlRt1p3jF66U+PnUjCxWbk2l05xNbMjy4t76TYoVOi/tpPelS5eYM2cOAAMGDChx/Cs55uzUXbrKmkdEilKgFBEph48++ojff/+dIUOGmGU8W1tbjEZjsT2MpWndujW9e/dm1KhRJR6McXBw4KmnnmLp0qVkZ2czb9487mv/OFUHzOBEXn5dzLI60hTsb/zxcAZ2wdOI3n2K9PT0wuulBcqJEydy6tQpjEYjffr0KXHs1MPFC69bgp2Nfq2J3Aj6P09EpBwWL15Mly5dzFbX0MYmf+W2vKuUBf2yc3JyGD16dIn3DB8+nDNnzvDpp5/y3x+P8UfjvmBtC4aK/VOfm2ckFyvu6jyS4fP/bOtYUumg+Ph4FixYgI+PDy1atChx9TY9PZ0XBofANUoOmYMBuM+1ikXnEJGSKVCKiJTh4sWLbN++nX79+pltTFtbW6D8gRLg7rvvZv78+URFRTF//vxi1318fPD19WXlll+ZtXE/gMl7PXcZPXlv80EgP1BeunSJU6dOAXD48GFCQ0Np3bo1+/fvL9a7G/K3CvTs2ZPc7CzucbI36VnK4uHqcMtssRD5p1GgFBEpw65duzAajTzwwANmG7MgUJb3YE6BkJAQxo8fzwsvvFC4b/GvGj4YQJJ9Q3M8IpB/oGfqhr2kZWTh7e2NwfYOvtmxj/Vb9hDYZxB2DlUZPHgwf/zxB8HBwUU+m5eXR2hoKPv27eOzzz6jk6+bxepQWlsZCKpX0yJji0jZ9KeciEgZdu7ciZ2dHQ0bmi+oVfSVdwGDwcDMmTOxsbEhLCyM7OxsXnrppcKVyJPuQRgvWxUr33S9DAYDOXl59F+yBVsrK9zHrebVzZlAJlbdXiEXmLnvEvf3D8fqrqJlgf71r3/x6aefEhMTg7+/P45uF1n+wxEzPVlRuXlGnnjon1nbVORWoEApIlKGffv20aBBg8JVRXO4nlfeBQwGA2+++Sa2traEh4fz7bffsmjRInKr1ODoFYeKbpksm5U16ReyC+f+KyNw2cYRg1drOs3ZRNu61Xkj2I8v1rzPrFmziIyMLKxLefH4AQynUsirUbfMA0IVYW1loJWXK3VrVjXbmCJSMXrlLSJShtzcXOztzbv/r2CFsqKvvAsYDAamTZvG559/TkpKCo0aNWLcwmgs1dmwLMb/rYluOXSO9m9/x4sLo3n++ecZM2YMeXl5REZG8tBDD1Et5TPsbc27lmFjZeCNYD+zjikiFaNAKSJyA5iyQvlXXbt2JTk5mWHDhrHz5BUs1dmwvHLzjFzJNeLSZRT1g1/gzJkz9OjRg3HjxjF69Gi2xX7B1N6NzDrn1F6+/9juSyK3CgVKEZFyMJq55M31HsopiaOjI9P+PRNb59omj2UOBa/FZ397EP9Hn2fHjh188cUXvP3229jb2zOwhQfjO+b3Bjf15zqhc/1/bF94kVuJAqWISBmqV6/OiRMnzDrm9R7KKU1ltDasKKPRiF2rUL7YtI0uXboUfj0vL49dH77FuS/mY2tFhU9+W1sZsLexYkZfP0YG1TX3Y4vIdVCgFBEpQ7NmzThx4gQnT54025jmeuVd4GZsOWgwGLCytiFy858/t9zcXJ555hmWLVvGu+MH8d2L7Wnl5QpQ5v7PguDZysuVb8ICtTIpchPRKW8RkTL4+/sDkJiYSNeuXc0ypqmHcv7uZm05mGuEhINnOXj6IjkZxxk6dCjff/8977//PiEhIQCsfLYlyWnn6PPi21CrAcYqrvz1RbiB/KLlQfVq8sRDHjrNLXITUqAUESlDnTp1cHFxIT4+3myB0twrlPe5VsEA3OAzOaUKmfs5u2c9iYeHB99++y3t2rUrcv2DRbNJjZ7Fjh078Krnw5FzmVzJycPOxor7XKuoA47ITe7m/JNWROQmYjAYCAkJYfny5WYLgDlYY1uzDilns0k+cYHMbNNWKqvY23CXba5Zns0SzuQ50mfcDHbv3l0sTG7ZsoUZM2YwefJk/Pz8qGJvg6+bE/4ezvi6OSlMivwDGIzmProoInILSkpKws/Pj48//pjHHnvsusY4cOoiq7Ye5buU06RmZBW5ZgA8XBwIql+TQS098K5V8de6DUMjyLqnOeavbG4edtYGvh3XrkiJnx9++IEuXbrQuHFjYmNjzVo8XkQqjwKliEg5tW3bFqPRSEJCQrGOMdeSlpHFy9F7SDh4FmsrA7nXKBZZcL2g40x56ytu3bqVtt0fw23IwnI/V2WzMkDr+6uz8tmWAGzevJmuXbvi7+/Phg0bqFpVeyNF/qluzj9jRURuQpMnT+b777/nP//5T7k/E7X9KB0j49ly6BzANcPkX69vOXSOjpHxRG0/WuYc2dnZPPfcc/i6u9L6ftcKl+GpLHn/O6Cz/+TvvPvuuzzyyCM0b96cL774QmFS5B9OK5QiIhUwZMgQoqKiSEpKwtPT85r3LvjuALM27jd5zhc712NUkHep1ydNmsSMGTP46aefcL63Lh0j48m+CcsIQf4q5Z3HtvPLytcYOnQokZGRODioy43IP50CpYhIBfz+++80atSIOnXq8OWXX3LnnXeWeF/U9qOEf7LHbPPO6OtXYt3F+Ph4OnToQEREBJMmTbLI3GZ36QzvPeZFQEDAjX4SETETvfIWEamAatWqsWrVKrZv306vXr3Iysoqdk9aRhYR65LNOu/kdcmk/e0gT2xsLN26daNdu3aEh4cXfn1gCw9e7FzPrPObk8GxBg+0bHWjH0NEzEiBUkSkgtq2bcvnn3/Oli1b6NGjB5mZRdsevhy9h5wy9kpWVE6ekZej/1x1/Prrr+nevTtt2rRh/fr1xU5Hjwry5t99/bC3sbrp9lQagSM3YatIEbl+CpQiItehXbt2fPnll2zfvp0WLVqwZcsWIL80UMLBs2Uevqmo3DwjCQfP8suxDCZPnkz37t1p3749MTExpb52H9jCg2/CAgtbG95MwfJmbBUpItdPeyhFREyQlJTEs88+y/bt2xkxYgROQc/xUWK62QMl5B9oMRxMIO3T2YSHh/Pqq69iZ2dXrs8W1sDcf5qj57JKbG14NSeP9AuXK6XbzobRbfB1c6qEmUSkMihQioiYKDc3l3feeYeXX34Zl6fmYVWtlsXmss7KIGaIP40aNbruMTKzc0psbZiWkVUpJ8QNQNKUR9QBR+QWokApImImew8cout/f4EKFD2vKEuHsco4Ie7p6kD8i0EWnUNEKpf2UIqImEleFVeLhkmw/IGWgS088He/y2LjW1sZCKpX02Lji8iNoUApImImlXXQxNLzvPVoY4uNnZtn5ImHitfTFJF/NgVKEREzsbOpnH9SLT2Pd62qtK1bHXMfCre2MtC2bnXq1lSbRZFbjQKliIiZ3OdaBUsX5jH8bx5LeyPYD1tr8/6KsLEy8Eawn1nHFJGbgwKliIiZVLG3wcPFsn2pPVwdKuV0tLuLA6/18jXrmFN7+eJu4Z+PiNwYCpQiImYUVL+mxQqIV/aBFnO2cJzQuX6JvchF5NagQCkiYkaDWnpYpKg53JgDLaa0cLS2MmBvY8WMvn6MDKproScUkZuBAqWIiBkVHGgx9yrljTzQUtEWjgXXW3m58k1YoFYmRW4DKmwuImJmlug4Y29jxTdhgTd8D2J5WjgG1avJEw956DS3yG1EgVJExALM3XFmRl+/m26lr7QWjiJy+1GgFBGxkAXfHWDWxv0mjzOhc33tQRSRm5oCpYiIBUVtP0rEumRy8owVOqxjbWXAxsrA1F6+N93KpIjI3ylQiohYWFpGFi9H7yHh4FmsrQzXDJYF19vWrc4bwX43fM+kiEh5KFCKiFQSHWgRkVuVAqWIyA2gAy0icitRoBQRERERk6iwuYiIiIiYRIFSREREREyiQCkiIiIiJlGgFBERERGTKFCKiIiIiEkUKEVERETEJAqUIiIiImISBUoRERERMYkCpYiIiIiYRIFSREREREyiQCkiIiIiJlGgFBERERGTKFCKiIiIiEkUKEVERETEJAqUIiIiImISBUoRERERMYkCpYiIiIiYRIFSREREREyiQCkiIiIiJlGgFBERERGTKFCKiIiIiEkUKEVERETEJAqUIiIiImISBUoRERERMYkCpYiIiIiYRIFSREREREyiQCkiIiIiJlGgFBERERGTKFCKiIiIiEkUKEVERETEJAqUIiIiImISBUoRERERMYkCpYiIiIiYRIFSREREREyiQCkiIiIiJlGgFBERERGT/D9jD09RB/ys9wAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "with open('out-list') as f:\n",
    "    array = list(map(int, f.readlines()))\n",
    "\n",
    "Gsub = networkx.subgraph(G, array)\n",
    "\n",
    "networkx.draw(Gsub)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0edae6f4-328d-4c34-a951-bf2fcf9f3244",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## More Resources\n",
    "\n",
    "- [GraphBLAS homepage](https://graphblas.org)\n",
    "  - [Great list of resources](https://graphblas.org/GraphBLAS-Pointers/)\n",
    "  - [Early mathematical description](http://www.mit.edu/~kepner/GraphBLAS/GraphBLAS-Math-release.pdf)\n",
    "- [LAGraph: higher level library](https://github.com/GraphBLAS/LAGraph)\n",
    "  - Released v1.0 on Tuesday (20 Sep 2022), congrats!\n",
    "  - Good source for more hardnened design and implementation\n",
    "- [RedisGraph: property graphs built on GraphBLAS](https://redis.io/docs/stack/graph/)\n",
    "- [GBTL: work towards a C++ interface and implementation](https://github.com/cmu-sei/gbtl/)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "07885e08",
   "metadata": {},
   "source": [
    "## Addendum: More Pitfalls\n",
    "\n",
    "The HPEC GraphBLAS BoF on 20 Sep 2022 has a talk from John Gilbert on \"What did the GraphBLAS get wrong?\"\n",
    "- [slides](https://docs.google.com/presentation/d/1Py8s_6XZo7USTN79K_m-Ls4XpABo0sblFZgKJ3FPoIg/edit#slide=id.p)\n",
    "- [wider survey responses](https://docs.google.com/document/d/1fMmm-Bmew0wpgJRrjyMHy6G-zPq6R6kQlRum560_4S0/edit)\n",
    "\n",
    "And another:  If you want to print types like `GrB_Index`, there are no macro helpers.  So in C you need casts to be safe, *e.g.* `printf(\"GrB_Index: %lu\\n\", (unsigned long) idx);`.\n",
    "\n"
   ]
  }
 ],
 "metadata": {
  "celltoolbar": "Slideshow",
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.6"
  },
  "rise": {
   "enable_chalkboard": true,
   "scroll": true,
   "transition": "none"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}