From 1f4eca29743ea8698adab783a87785d4aef3d8a8 Mon Sep 17 00:00:00 2001 From: Peter Norvig Date: Mon, 8 Apr 2024 17:28:33 -0700 Subject: [PATCH] Add files via upload --- ipynb/Paint.ipynb | 374 +++++++++++++++++++--------------------------- 1 file changed, 157 insertions(+), 217 deletions(-) diff --git a/ipynb/Paint.ipynb b/ipynb/Paint.ipynb index fd56852..43a3479 100644 --- a/ipynb/Paint.ipynb +++ b/ipynb/Paint.ipynb @@ -104,8 +104,9 @@ "source": [ "def do(n, h=1, colors='RB') -> dict: \n", " \"\"\"A dict of {(w, h): mean-cluster-size} for grids with width w from 1 to n, and with height h.\"\"\"\n", - " return {(w, h): mean(map(mean_cluster_size, grids(w, h, colors)))\n", - " for w in range(1, n + 1)}" + " for w in range(1, n + 1):\n", + " μ = mean(map(mean_cluster_size, grids(w, h, colors)))\n", + " print(f'{w:2} × {h} grids: {μ:6.4f} mean cluster size')" ] }, { @@ -121,44 +122,32 @@ { "cell_type": "code", "execution_count": 3, - "id": "ee733740-c0b2-490b-9541-1b07e37a6056", + "id": "e004289f-0c81-46ab-8900-9312a0aa6f7c", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "CPU times: user 1.29 s, sys: 43 ms, total: 1.33 s\n", - "Wall time: 1.34 s\n" + " 1 × 1 grids: 1.0000 mean cluster size\n", + " 2 × 1 grids: 1.5000 mean cluster size\n", + " 3 × 1 grids: 1.7500 mean cluster size\n", + " 4 × 1 grids: 1.8750 mean cluster size\n", + " 5 × 1 grids: 1.9375 mean cluster size\n", + " 6 × 1 grids: 1.9688 mean cluster size\n", + " 7 × 1 grids: 1.9844 mean cluster size\n", + " 8 × 1 grids: 1.9922 mean cluster size\n", + " 9 × 1 grids: 1.9961 mean cluster size\n", + "10 × 1 grids: 1.9980 mean cluster size\n", + "11 × 1 grids: 1.9990 mean cluster size\n", + "12 × 1 grids: 1.9995 mean cluster size\n", + "13 × 1 grids: 1.9998 mean cluster size\n", + "14 × 1 grids: 1.9999 mean cluster size\n" ] - }, - { - "data": { - "text/plain": [ - "{(1, 1): 1.0,\n", - " (2, 1): 1.5,\n", - " (3, 1): 1.75,\n", - " (4, 1): 1.875,\n", - " (5, 1): 1.9375,\n", - " (6, 1): 1.96875,\n", - " (7, 1): 1.984375,\n", - " (8, 1): 1.9921875,\n", - " (9, 1): 1.99609375,\n", - " (10, 1): 1.998046875,\n", - " (11, 1): 1.9990234375,\n", - " (12, 1): 1.99951171875,\n", - " (13, 1): 1.999755859375,\n", - " (14, 1): 1.9998779296875,\n", - " (15, 1): 1.99993896484375}" - ] - }, - "execution_count": 3, - "metadata": {}, - "output_type": "execute_result" } ], "source": [ - "%time do(15)" + "do(14)" ] }, { @@ -187,31 +176,20 @@ "name": "stdout", "output_type": "stream", "text": [ - "CPU times: user 8.12 s, sys: 269 ms, total: 8.39 s\n", - "Wall time: 8.71 s\n" + " 1 × 2 grids: 1.5000 mean cluster size\n", + " 2 × 2 grids: 2.1250 mean cluster size\n", + " 3 × 2 grids: 2.4688 mean cluster size\n", + " 4 × 2 grids: 2.6682 mean cluster size\n", + " 5 × 2 grids: 2.7920 mean cluster size\n", + " 6 × 2 grids: 2.8737 mean cluster size\n", + " 7 × 2 grids: 2.9305 mean cluster size\n", + " 8 × 2 grids: 2.9717 mean cluster size\n", + " 9 × 2 grids: 3.0027 mean cluster size\n" ] - }, - { - "data": { - "text/plain": [ - "{(1, 2): 1.5,\n", - " (2, 2): 2.125,\n", - " (3, 2): 2.46875,\n", - " (4, 2): 2.6682291666666664,\n", - " (5, 2): 2.7920386904761907,\n", - " (6, 2): 2.873721168154762,\n", - " (7, 2): 2.9304973563762626,\n", - " (8, 2): 2.971709969311288,\n", - " (9, 2): 3.0027141140414764}" - ] - }, - "execution_count": 4, - "metadata": {}, - "output_type": "execute_result" } ], "source": [ - "%time do(9, 2)" + "do(9, 2)" ] }, { @@ -238,28 +216,17 @@ "name": "stdout", "output_type": "stream", "text": [ - "CPU times: user 6.52 s, sys: 170 ms, total: 6.69 s\n", - "Wall time: 6.72 s\n" + " 1 × 3 grids: 1.7500 mean cluster size\n", + " 2 × 3 grids: 2.4688 mean cluster size\n", + " 3 × 3 grids: 2.8986 mean cluster size\n", + " 4 × 3 grids: 3.1591 mean cluster size\n", + " 5 × 3 grids: 3.3266 mean cluster size\n", + " 6 × 3 grids: 3.4408 mean cluster size\n" ] - }, - { - "data": { - "text/plain": [ - "{(1, 3): 1.75,\n", - " (2, 3): 2.46875,\n", - " (3, 3): 2.8986049107142855,\n", - " (4, 3): 3.1591145833333334,\n", - " (5, 3): 3.326567058836346,\n", - " (6, 3): 3.4407984463961334}" - ] - }, - "execution_count": 5, - "metadata": {}, - "output_type": "execute_result" } ], "source": [ - "%time do(6, 3)" + "do(6, 3)" ] }, { @@ -272,27 +239,16 @@ "name": "stdout", "output_type": "stream", "text": [ - "CPU times: user 27.9 s, sys: 952 ms, total: 28.9 s\n", - "Wall time: 29.1 s\n" + " 1 × 4 grids: 1.8750 mean cluster size\n", + " 2 × 4 grids: 2.6682 mean cluster size\n", + " 3 × 4 grids: 3.1591 mean cluster size\n", + " 4 × 4 grids: 3.4616 mean cluster size\n", + " 5 × 4 grids: 3.6608 mean cluster size\n" ] - }, - { - "data": { - "text/plain": [ - "{(1, 4): 1.875,\n", - " (2, 4): 2.6682291666666664,\n", - " (3, 4): 3.1591145833333334,\n", - " (4, 4): 3.461628834083668,\n", - " (5, 4): 3.6608226758600444}" - ] - }, - "execution_count": 6, - "metadata": {}, - "output_type": "execute_result" } ], "source": [ - "%time do(5, 4)" + "do(5, 4)" ] }, { @@ -315,34 +271,21 @@ "name": "stdout", "output_type": "stream", "text": [ - "CPU times: user 13.1 s, sys: 510 ms, total: 13.6 s\n", - "Wall time: 13.7 s\n" + " 1 × 1 grids: 1.0000 mean cluster size\n", + " 2 × 1 grids: 1.3333 mean cluster size\n", + " 3 × 1 grids: 1.4444 mean cluster size\n", + " 4 × 1 grids: 1.4815 mean cluster size\n", + " 5 × 1 grids: 1.4938 mean cluster size\n", + " 6 × 1 grids: 1.4979 mean cluster size\n", + " 7 × 1 grids: 1.4993 mean cluster size\n", + " 8 × 1 grids: 1.4998 mean cluster size\n", + " 9 × 1 grids: 1.4999 mean cluster size\n", + "10 × 1 grids: 1.5000 mean cluster size\n" ] - }, - { - "data": { - "text/plain": [ - "{(1, 1): 1.0,\n", - " (2, 1): 1.3333333333333333,\n", - " (3, 1): 1.4444444444444444,\n", - " (4, 1): 1.4814814814814814,\n", - " (5, 1): 1.4938271604938271,\n", - " (6, 1): 1.4979423868312758,\n", - " (7, 1): 1.4993141289437586,\n", - " (8, 1): 1.4997713763145861,\n", - " (9, 1): 1.499923792104862,\n", - " (10, 1): 1.4999745973682874,\n", - " (11, 1): 1.4999915324560957,\n", - " (12, 1): 1.4999971774853653}" - ] - }, - "execution_count": 7, - "metadata": {}, - "output_type": "execute_result" } ], "source": [ - "%time do(12, 1, 'RGB')" + "do(10, 1, 'RGB')" ] }, { @@ -367,28 +310,17 @@ "name": "stdout", "output_type": "stream", "text": [ - "CPU times: user 10.5 s, sys: 412 ms, total: 11 s\n", - "Wall time: 11.3 s\n" + " 1 × 2 grids: 1.3333 mean cluster size\n", + " 2 × 2 grids: 1.6543 mean cluster size\n", + " 3 × 2 grids: 1.7737 mean cluster size\n", + " 4 × 2 grids: 1.8261 mean cluster size\n", + " 5 × 2 grids: 1.8535 mean cluster size\n", + " 6 × 2 grids: 1.8698 mean cluster size\n" ] - }, - { - "data": { - "text/plain": [ - "{(1, 2): 1.3333333333333333,\n", - " (2, 2): 1.654320987654321,\n", - " (3, 2): 1.7736625514403292,\n", - " (4, 2): 1.8261240664532845,\n", - " (5, 2): 1.8535078615096905,\n", - " (6, 2): 1.8697928838761029}" - ] - }, - "execution_count": 8, - "metadata": {}, - "output_type": "execute_result" } ], "source": [ - "%time do(6, 2, 'RGB')" + "do(6, 2, 'RGB')" ] }, { @@ -396,7 +328,7 @@ "id": "3c18f723-447f-4cd6-ba27-59db37e87659", "metadata": {}, "source": [ - "By the same reasoning as we had last time, this should converge to 3/2 + 3/2 × 1/3 = 2. " + "By the same reasoning as we had last time, this should converge to 3/2 + 3/2 × 1/3 = 2. (But it would take a lot of computation time to consider the larger grids that would be needed to get closer to 2.)" ] }, { @@ -446,7 +378,7 @@ { "data": { "text/plain": [ - "7.58150113722517" + "7.037297677691766" ] }, "execution_count": 10, @@ -491,10 +423,10 @@ { "data": { "text/plain": [ - "{'mean': 7.246114781372947,\n", - " 'max': 7.716049382716049,\n", - " 'min': 6.830601092896175,\n", - " 'stdev': 0.181119344027548,\n", + "{'mean': 7.239472697041692,\n", + " 'max': 7.9239302694136295,\n", + " 'min': 6.6711140760507,\n", + " 'stdev': 0.19030622926154506,\n", " 'N': 200}" ] }, @@ -504,7 +436,7 @@ }, { "data": { - "image/png": "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\n", + "image/png": "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\n", "text/plain": [ "
" ] @@ -536,10 +468,10 @@ { "data": { "text/plain": [ - "{'mean': 7.402852321356604,\n", - " 'max': 7.635044855888529,\n", - " 'min': 7.126313914127917,\n", - " 'stdev': 0.10047800378358163,\n", + "{'mean': 7.4134560772843425,\n", + " 'max': 7.651109410864575,\n", + " 'min': 7.116171499733143,\n", + " 'stdev': 0.10453350361308941,\n", " 'N': 200}" ] }, @@ -549,7 +481,7 @@ }, { "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAiwAAAGdCAYAAAAxCSikAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjYuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8o6BhiAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAilElEQVR4nO3de3CU1f3H8c+G6CYiCROEXLgkSLkI2MiAkkC5lQk0XGoLCEgloYiXEamYoTQRKfCrGrSoKYJQWkhEhos2cqlhFKgE5CIjl3hpEUNJSAqJiEoWUDcgz+8Ph61bksDCPsnZ5f2aOTM+5znnyXcPMfnM2SfPOizLsgQAAGCwkIYuAAAA4HIILAAAwHgEFgAAYDwCCwAAMB6BBQAAGI/AAgAAjEdgAQAAxiOwAAAA44U2dAH+cuHCBR0/flxNmjSRw+Fo6HIAAMAVsCxLp0+fVlxcnEJCat9HCZrAcvz4cbVu3bqhywAAAFehvLxcrVq1qvV80ASWJk2aSPr+BUdERDRwNQAA4Eq4XC61bt3a83u8NkETWC6+DRQREUFgAQAgwFzudg5uugUAAMYjsAAAAOMRWAAAgPEILAAAwHgEFgAAYDwCCwAAMB6BBQAAGI/AAgAAjEdgAQAAxiOwAAAA4xFYAACA8QgsAADAeAQWAABgPAILAAAwXmhDFwDg+pKQWeDT+NK5Q22qBEAgYYcFAAAYj8ACAACMR2ABAADGI7AAAADjEVgAAIDxfA4s27dv1/DhwxUXFyeHw6F169Z5nXc4HDW2P/7xj7VeMy8vr8Y53377rc8vCAAABB+fA8vZs2eVmJioBQsW1Hi+oqLCqy1btkwOh0MjR46s87oRERGXzA0LC/O1PAAAEIR8fg5LamqqUlNTaz0fExPjdbx+/XoNGDBAt956a53XdTgcl8wFAACQbL6H5bPPPlNBQYHuv//+y449c+aM4uPj1apVKw0bNkwHDhyoc7zb7ZbL5fJqAAAgONkaWF555RU1adJEI0aMqHNcp06dlJeXpw0bNmjVqlUKCwtT7969VVxcXOuc7OxsRUZGelrr1q39XT4AADCErYFl2bJl+tWvfnXZe1GSkpJ03333KTExUX369NFrr72mDh066KWXXqp1TlZWlqqqqjytvLzc3+UDAABD2PZZQu+++64OHTqkNWvW+Dw3JCREd955Z507LE6nU06n81pKBAAAAcK2HZalS5eqe/fuSkxM9HmuZVkqKipSbGysDZUBAIBA4/MOy5kzZ3T48GHPcUlJiYqKihQVFaU2bdpIklwul15//XU9//zzNV4jLS1NLVu2VHZ2tiRpzpw5SkpKUvv27eVyuTR//nwVFRVp4cKFV/OaAABAkPE5sOzdu1cDBgzwHGdkZEiS0tPTlZeXJ0lavXq1LMvSvffeW+M1ysrKFBLy382dU6dO6cEHH1RlZaUiIyPVrVs3bd++XXfddZev5QEAgCDksCzLaugi/MHlcikyMlJVVVWKiIho6HIA1CIhs8Cn8aVzh9pUCQATXOnvbz5LCAAAGI/AAgAAjEdgAQAAxiOwAAAA4xFYAACA8QgsAADAeAQWAABgPAILAAAwHoEFAAAYj8ACAACMR2ABAADGI7AAAADjEVgAAIDxCCwAAMB4BBYAAGA8AgsAADAegQUAABiPwAIAAIxHYAEAAMYjsAAAAOMRWAAAgPEILAAAwHihDV0AANSHhMwCn8aXzh1qUyUArgY7LAAAwHgEFgAAYDwCCwAAMB6BBQAAGI/AAgAAjEdgAQAAxiOwAAAA4xFYAACA8QgsAADAeAQWAABgPAILAAAwHoEFAAAYj8ACAACMR2ABAADGI7AAAADj+RxYtm/fruHDhysuLk4Oh0Pr1q3zOj9hwgQ5HA6vlpSUdNnr5ufnq3PnznI6nercubPWrl3ra2kAACBI+RxYzp49q8TERC1YsKDWMT/72c9UUVHhaRs3bqzzmrt379aYMWM0fvx4ffDBBxo/frxGjx6tPXv2+FoeAAAIQqG+TkhNTVVqamqdY5xOp2JiYq74mjk5OUpJSVFWVpYkKSsrS9u2bVNOTo5WrVrla4kAACDI2HIPS2FhoVq0aKEOHTrogQce0IkTJ+ocv3v3bg0aNMirb/Dgwdq1a1etc9xut1wul1cDAADByecdlstJTU3VPffco/j4eJWUlGjmzJn66U9/qn379snpdNY4p7KyUtHR0V590dHRqqysrPXrZGdna86cOX6tHQCCSUJmgU/jS+cOtakS4Nr5PbCMGTPG899du3ZVjx49FB8fr4KCAo0YMaLWeQ6Hw+vYsqxL+n4oKytLGRkZnmOXy6XWrVtfQ+UAAMBUfg8s/ys2Nlbx8fEqLi6udUxMTMwluyknTpy4ZNflh5xOZ607NgAAILjY/hyWL774QuXl5YqNja11THJysjZv3uzVt2nTJvXq1cvu8gAAQADweYflzJkzOnz4sOe4pKRERUVFioqKUlRUlGbPnq2RI0cqNjZWpaWleuKJJ3TLLbfol7/8pWdOWlqaWrZsqezsbEnSY489pr59++rZZ5/V3XffrfXr12vLli3asWOHH14iAAAIdD4Hlr1792rAgAGe44v3kaSnp2vRokX66KOPtHz5cp06dUqxsbEaMGCA1qxZoyZNmnjmlJWVKSTkv5s7vXr10urVq/Xkk09q5syZateundasWaOePXtey2sDAABBwufA0r9/f1mWVev5t99++7LXKCwsvKRv1KhRGjVqlK/lAACA6wCfJQQAAIxHYAEAAMYjsAAAAOMRWAAAgPEILAAAwHgEFgAAYDwCCwAAMB6BBQAAGI/AAgAAjEdgAQAAxiOwAAAA4xFYAACA8QgsAADAeD5/WjMAXG8SMgt8nlM6d6gNlQDXL3ZYAACA8QgsAADAeAQWAABgPAILAAAwHoEFAAAYj8ACAACMR2ABAADGI7AAAADjEVgAAIDxCCwAAMB4BBYAAGA8AgsAADAegQUAABiPwAIAAIxHYAEAAMYjsAAAAOMRWAAAgPEILAAAwHihDV0AAFyphMwCn8aXzh1qUyUA6hs7LAAAwHgEFgAAYDwCCwAAMB6BBQAAGI/AAgAAjOdzYNm+fbuGDx+uuLg4ORwOrVu3znPu3Llz+t3vfqfbb79djRs3VlxcnNLS0nT8+PE6r5mXlyeHw3FJ+/bbb31+QQAAIPj4HFjOnj2rxMRELViw4JJzX3/9tfbv36+ZM2dq//79euONN/Tpp5/q5z//+WWvGxERoYqKCq8WFhbma3kAACAI+fwcltTUVKWmptZ4LjIyUps3b/bqe+mll3TXXXeprKxMbdq0qfW6DodDMTExvpYDAACuA7bfw1JVVSWHw6GmTZvWOe7MmTOKj49Xq1atNGzYMB04cKDO8W63Wy6Xy6sBAIDgZGtg+fbbb5WZmalx48YpIiKi1nGdOnVSXl6eNmzYoFWrViksLEy9e/dWcXFxrXOys7MVGRnpaa1bt7bjJQAAAAPYFljOnTunsWPH6sKFC3r55ZfrHJuUlKT77rtPiYmJ6tOnj1577TV16NBBL730Uq1zsrKyVFVV5Wnl5eX+fgkAAMAQtnyW0Llz5zR69GiVlJTonXfeqXN3pSYhISG6884769xhcTqdcjqd11oqAAAIAH7fYbkYVoqLi7VlyxY1a9bM52tYlqWioiLFxsb6uzwAABCAfN5hOXPmjA4fPuw5LikpUVFRkaKiohQXF6dRo0Zp//79evPNN/Xdd9+psrJSkhQVFaUbb7xRkpSWlqaWLVsqOztbkjRnzhwlJSWpffv2crlcmj9/voqKirRw4UJ/vEYAABDgfA4se/fu1YABAzzHGRkZkqT09HTNnj1bGzZskCTdcccdXvO2bt2q/v37S5LKysoUEvLfzZ1Tp07pwQcfVGVlpSIjI9WtWzdt375dd911l6/lAQCAIORzYOnfv78sy6r1fF3nLiosLPQ6fvHFF/Xiiy/6WgoAALhO8FlCAADAeAQWAABgPAILAAAwni3PYQEA+EdCZoFP40vnDrWpEqBhscMCAACMR2ABAADGI7AAAADjEVgAAIDxCCwAAMB4BBYAAGA8AgsAADAegQUAABiPwAIAAIxHYAEAAMYjsAAAAOMRWAAAgPEILAAAwHgEFgAAYDwCCwAAMB6BBQAAGI/AAgAAjEdgAQAAxiOwAAAA4xFYAACA8QgsAADAeAQWAABgPAILAAAwHoEFAAAYj8ACAACMR2ABAADGI7AAAADjEVgAAIDxCCwAAMB4BBYAAGA8AgsAADAegQUAABiPwAIAAIxHYAEAAMbzObBs375dw4cPV1xcnBwOh9atW+d13rIszZ49W3FxcQoPD1f//v31z3/+87LXzc/PV+fOneV0OtW5c2etXbvW19IAAECQ8jmwnD17VomJiVqwYEGN55977jm98MILWrBggd5//33FxMQoJSVFp0+frvWau3fv1pgxYzR+/Hh98MEHGj9+vEaPHq09e/b4Wh4AAAhCob5OSE1NVWpqao3nLMtSTk6OZsyYoREjRkiSXnnlFUVHR2vlypV66KGHapyXk5OjlJQUZWVlSZKysrK0bds25eTkaNWqVb6WCAAAgoxf72EpKSlRZWWlBg0a5OlzOp3q16+fdu3aVeu83bt3e82RpMGDB9c5x+12y+VyeTUAABCcfN5hqUtlZaUkKTo62qs/OjpaR48erXNeTXMuXq8m2dnZmjNnzjVUC1zfEjILfBpfOneoX+YCwNWw5a+EHA6H17FlWZf0XeucrKwsVVVVeVp5efnVFwwAAIzm1x2WmJgYSd/vmMTGxnr6T5w4cckOyv/O+9/dlMvNcTqdcjqd11gxAAAIBH7dYWnbtq1iYmK0efNmT191dbW2bdumXr161TovOTnZa44kbdq0qc45AADg+uHzDsuZM2d0+PBhz3FJSYmKiooUFRWlNm3aaOrUqXrmmWfUvn17tW/fXs8884xuuukmjRs3zjMnLS1NLVu2VHZ2tiTpscceU9++ffXss8/q7rvv1vr167Vlyxbt2LHDDy8RAAAEOp8Dy969ezVgwADPcUZGhiQpPT1deXl5mj59ur755hs98sgj+uqrr9SzZ09t2rRJTZo08cwpKytTSMh/N3d69eql1atX68knn9TMmTPVrl07rVmzRj179ryW1wYAAIKEz4Glf//+siyr1vMOh0OzZ8/W7Nmzax1TWFh4Sd+oUaM0atQoX8sBAADXAT5LCAAAGI/AAgAAjEdgAQAAxiOwAAAA4xFYAACA8QgsAADAeAQWAABgPAILAAAwHoEFAAAYj8ACAACMR2ABAADGI7AAAADjEVgAAIDxCCwAAMB4BBYAAGA8AgsAADAegQUAABiPwAIAAIxHYAEAAMYjsAAAAOMRWAAAgPEILAAAwHgEFgAAYDwCCwAAMB6BBQAAGI/AAgAAjEdgAQAAxiOwAAAA4xFYAACA8QgsAADAeAQWAABgPAILAAAwXmhDFwAEuoTMAp/nlM4detXz/TUXqAvfWzANOywAAMB4BBYAAGA8AgsAADAegQUAABjP74ElISFBDofjkjZ58uQaxxcWFtY4/pNPPvF3aQAAIED5/a+E3n//fX333Xee448//lgpKSm655576px36NAhRUREeI6bN2/u79IAAECA8ntg+d+gMXfuXLVr1079+vWrc16LFi3UtGlTf5cDAACCgK33sFRXV2vFihWaOHGiHA5HnWO7deum2NhYDRw4UFu3brWzLAAAEGBsfXDcunXrdOrUKU2YMKHWMbGxsVqyZIm6d+8ut9utV199VQMHDlRhYaH69u1b6zy32y232+05drlc/iwdAAAYxNbAsnTpUqWmpiouLq7WMR07dlTHjh09x8nJySovL9e8efPqDCzZ2dmaM2eOX+sFAABmsu0toaNHj2rLli2aNGmSz3OTkpJUXFxc55isrCxVVVV5Wnl5+dWWCgAADGfbDktubq5atGihoUN9/3yJAwcOKDY2ts4xTqdTTqfzassDAAABxJbAcuHCBeXm5io9PV2hod5fIisrS8eOHdPy5cslSTk5OUpISFCXLl08N+nm5+crPz/fjtIAAEAAsiWwbNmyRWVlZZo4ceIl5yoqKlRWVuY5rq6u1rRp03Ts2DGFh4erS5cuKigo0JAhQ+woDQAABCBbAsugQYNkWVaN5/Ly8ryOp0+frunTp9tRBgAACBJ8lhAAADAegQUAABiPwAIAAIxn64PjAABSQmaBT+NL5/r+OAiTXG+vF/WDHRYAAGA8AgsAADAegQUAABiPwAIAAIxHYAEAAMYjsAAAAOMRWAAAgPEILAAAwHgEFgAAYDwCCwAAMB6BBQAAGI/AAgAAjEdgAQAAxiOwAAAA4xFYAACA8QgsAADAeAQWAABgPAILAAAwHoEFAAAYL7ShCwD8JSGzwKfxpXOH+mUuAMB+7LAAAADjEVgAAIDxCCwAAMB4BBYAAGA8AgsAADAegQUAABiPwAIAAIxHYAEAAMYjsAAAAOMRWAAAgPEILAAAwHgEFgAAYDwCCwAAMB6BBQAAGM/vgWX27NlyOBxeLSYmps4527ZtU/fu3RUWFqZbb71Vixcv9ndZAAAggIXacdEuXbpoy5YtnuNGjRrVOrakpERDhgzRAw88oBUrVmjnzp165JFH1Lx5c40cOdKO8gAAQICxJbCEhoZedlflosWLF6tNmzbKycmRJN12223au3ev5s2bR2ABAACSbLqHpbi4WHFxcWrbtq3Gjh2rI0eO1Dp29+7dGjRokFff4MGDtXfvXp07d86O8gAAQIDxe2Dp2bOnli9frrffflt/+ctfVFlZqV69eumLL76ocXxlZaWio6O9+qKjo3X+/HmdPHmy1q/jdrvlcrm8GgAACE5+f0soNTXV89+33367kpOT1a5dO73yyivKyMiocY7D4fA6tiyrxv4fys7O1pw5c/xQMQDAFAmZBT6NL5079Krn/u98mM32P2tu3Lixbr/9dhUXF9d4PiYmRpWVlV59J06cUGhoqJo1a1brdbOyslRVVeVp5eXlfq0bAACYw5abbn/I7Xbr4MGD6tOnT43nk5OT9fe//92rb9OmTerRo4duuOGGWq/rdDrldDr9WisAADCT33dYpk2bpm3btqmkpER79uzRqFGj5HK5lJ6eLun7nZG0tDTP+IcfflhHjx5VRkaGDh48qGXLlmnp0qWaNm2av0sDAAAByu87LP/5z39077336uTJk2revLmSkpL03nvvKT4+XpJUUVGhsrIyz/i2bdtq48aNevzxx7Vw4ULFxcVp/vz5/EkzAADw8HtgWb16dZ3n8/LyLunr16+f9u/f7+9SAABAkOCzhAAAgPEILAAAwHgEFgAAYDwCCwAAMB6BBQAAGI/AAgAAjEdgAQAAxiOwAAAA4xFYAACA8QgsAADAeAQWAABgPAILAAAwHoEFAAAYz++f1gwAQCBKyCzwaXzp3KE2VVI/Au31ssMCAACMR2ABAADGI7AAAADjEVgAAIDxCCwAAMB4BBYAAGA8AgsAADAegQUAABiPwAIAAIxHYAEAAMYjsAAAAOMRWAAAgPEILAAAwHgEFgAAYDwCCwAAMB6BBQAAGI/AAgAAjEdgAQAAxiOwAAAA44U2dAGoW0JmgU/jS+cOtakSAIAd+Dl/ZdhhAQAAxiOwAAAA4xFYAACA8QgsAADAeH4PLNnZ2brzzjvVpEkTtWjRQr/4xS906NChOucUFhbK4XBc0j755BN/lwcAAAKQ3wPLtm3bNHnyZL333nvavHmzzp8/r0GDBuns2bOXnXvo0CFVVFR4Wvv27f1dHgAACEB+/7Pmt956y+s4NzdXLVq00L59+9S3b98657Zo0UJNmzb1d0kAACDA2X4PS1VVlSQpKirqsmO7deum2NhYDRw4UFu3bq1zrNvtlsvl8moAACA42RpYLMtSRkaGfvKTn6hr1661jouNjdWSJUuUn5+vN954Qx07dtTAgQO1ffv2WudkZ2crMjLS01q3bm3HSwAAAAaw9Um3jz76qD788EPt2LGjznEdO3ZUx44dPcfJyckqLy/XvHnzan0bKSsrSxkZGZ5jl8tFaAEAIEjZtsMyZcoUbdiwQVu3blWrVq18np+UlKTi4uJazzudTkVERHg1AAAQnPy+w2JZlqZMmaK1a9eqsLBQbdu2varrHDhwQLGxsX6uDgAABCK/B5bJkydr5cqVWr9+vZo0aaLKykpJUmRkpMLDwyV9/3bOsWPHtHz5cklSTk6OEhIS1KVLF1VXV2vFihXKz89Xfn6+v8sDAAAByO+BZdGiRZKk/v37e/Xn5uZqwoQJkqSKigqVlZV5zlVXV2vatGk6duyYwsPD1aVLFxUUFGjIkCH+Lg8AAAQgW94Supy8vDyv4+nTp2v69On+LgUAAAQJPksIAAAYj8ACAACMR2ABAADGs/XBccEiIbPAp/Glc4faVAkAwEQN9Xvievr9xA4LAAAwHoEFAAAYj8ACAACMR2ABAADGI7AAAADjEVgAAIDxCCwAAMB4BBYAAGA8AgsAADAegQUAABiPwAIAAIxHYAEAAMYjsAAAAOMRWAAAgPEILAAAwHgEFgAAYDwCCwAAMB6BBQAAGI/AAgAAjBfa0AXAPgmZBT6NL507NKDnAgCCFzssAADAeAQWAABgPAILAAAwHoEFAAAYj8ACAACMR2ABAADGI7AAAADjEVgAAIDxCCwAAMB4BBYAAGA8AgsAADAegQUAABiPwAIAAIxHYAEAAMazLbC8/PLLatu2rcLCwtS9e3e9++67dY7ftm2bunfvrrCwMN16661avHixXaUBAIAAY0tgWbNmjaZOnaoZM2bowIED6tOnj1JTU1VWVlbj+JKSEg0ZMkR9+vTRgQMH9MQTT+g3v/mN8vPz7SgPAAAEGFsCywsvvKD7779fkyZN0m233aacnBy1bt1aixYtqnH84sWL1aZNG+Xk5Oi2227TpEmTNHHiRM2bN8+O8gAAQIAJ9fcFq6urtW/fPmVmZnr1Dxo0SLt27apxzu7duzVo0CCvvsGDB2vp0qU6d+6cbrjhhkvmuN1uud1uz3FVVZUkyeVyXetLuMQF99c+jfdnDdfytZlr5tyG/NrMrZ+5Dfm1meubQKzbhLn+dPG6lmXVPdDys2PHjlmSrJ07d3r1P/3001aHDh1qnNO+fXvr6aef9urbuXOnJck6fvx4jXNmzZplSaLRaDQajRYErby8vM584fcdloscDofXsWVZl/RdbnxN/RdlZWUpIyPDc3zhwgV9+eWXatasmWeOy+VS69atVV5eroiIiKt6Hagba2w/1rh+sM72Y43tF4hrbFmWTp8+rbi4uDrH+T2w3HLLLWrUqJEqKyu9+k+cOKHo6Oga58TExNQ4PjQ0VM2aNatxjtPplNPp9Opr2rRpjWMjIiIC5h8uULHG9mON6wfrbD/W2H6BtsaRkZGXHeP3m25vvPFGde/eXZs3b/bq37x5s3r16lXjnOTk5EvGb9q0ST169Kjx/hUAAHB9seWvhDIyMvTXv/5Vy5Yt08GDB/X444+rrKxMDz/8sKTv385JS0vzjH/44Yd19OhRZWRk6ODBg1q2bJmWLl2qadOm2VEeAAAIMLbcwzJmzBh98cUX+r//+z9VVFSoa9eu2rhxo+Lj4yVJFRUVXs9kadu2rTZu3KjHH39cCxcuVFxcnObPn6+RI0deUx1Op1OzZs265K0j+A9rbD/WuH6wzvZjje0XzGvssKzL/R0RAABAw+KzhAAAgPEILAAAwHgEFgAAYDwCCwAAMF7ABpaEhAQ5HI5L2uTJk2scX1FRoXHjxqljx44KCQnR1KlT67fgAOTrGr/xxhtKSUlR8+bNFRERoeTkZL399tv1XHVg8XWNd+zYod69e6tZs2YKDw9Xp06d9OKLL9Zz1YHH13X+oZ07dyo0NFR33HGH/YUGMF/XuLCwsMbxn3zyST1XHjiu5vvY7XZrxowZio+Pl9PpVLt27bRs2bJ6rNp/bHs0v93ef/99fffdd57jjz/+WCkpKbrnnntqHO92u9W8eXPNmDGDH/BXyNc13r59u1JSUvTMM8+oadOmys3N1fDhw7Vnzx5169atvsoOKL6ucePGjfXoo4/qxz/+sRo3bqwdO3booYceUuPGjfXggw/WV9kBx9d1vqiqqkppaWkaOHCgPvvsM7vLDGhXu8aHDh3yeiJr8+bNbasx0F3NGo8ePVqfffaZli5dqh/96Ec6ceKEzp8/Xx/l+l3Q/Fnz1KlT9eabb6q4uLjOzyySpP79++uOO+5QTk5O/RQXJHxZ44u6dOmiMWPG6Pe//73N1QWHq1njESNGqHHjxnr11Vdtri54XOk6jx07Vu3bt1ejRo20bt06FRUV1V+RAe5ya1xYWKgBAwboq6++qvVjVVC3y63xW2+9pbFjx+rIkSOKiopqgAr9K2DfEvqh6upqrVixQhMnTrziH/LwzdWs8YULF3T69Omg+B+lPlzNGh84cEC7du1Sv379bK4ueFzpOufm5urf//63Zs2aVY/VBQdfvpe7deum2NhYDRw4UFu3bq2nCgPflazxhg0b1KNHDz333HNq2bKlOnTooGnTpumbb76p52r9I2DfEvqhdevW6dSpU5owYUJDlxK0rmaNn3/+eZ09e1ajR4+2r7Ag4ssat2rVSp9//rnOnz+v2bNna9KkSfYXGCSuZJ2Li4uVmZmpd999V6GhQfFjsl5dyRrHxsZqyZIl6t69u9xut1599VUNHDhQhYWF6tu3b/0VG6CuZI2PHDmiHTt2KCwsTGvXrtXJkyf1yCOP6MsvvwzM+1isIDBo0CBr2LBhVzy+X79+1mOPPWZfQUHI1zVeuXKlddNNN1mbN2+2sarg4ssaHzlyxPrwww+tJUuWWFFRUdbKlSttri54XG6dz58/b/Xo0cNatGiRp2/WrFlWYmJiPVQXHHz9eXHRsGHDrOHDh9tQUfC5kjVOSUmxwsLCrFOnTnn68vPzLYfDYX399dd2l+h3AR9YSktLrZCQEGvdunVXPIfA4htf13j16tVWeHi49eabb9pcWfC4mu/ji/7whz9YHTp0sKGq4HMl6/zVV19ZkqxGjRp5msPh8PT94x//qMeKA8+1fC8/9dRTVqdOnWyoKrhc6RqnpaVZ7dq18+r717/+ZUmyPv30UztLtEXA73Xm5uaqRYsWGjp0aEOXErR8WeNVq1Zp4sSJWrVqFf8mPriW72PLsuR2u22oKvhcyTpHREToo48+8up7+eWX9c477+hvf/ub2rZta3eZAe1avpcPHDig2NhYG6oKLle6xr1799brr7+uM2fO6Oabb5YkffrppwoJCVGrVq3qo1S/CujAcuHCBeXm5io9Pf2S95mzsrJ07NgxLV++3NN38Q7/M2fO6PPPP1dRUZFuvPFGde7cuT7LDii+rPGqVauUlpamP/3pT0pKSlJlZaUkKTw8XJGRkfVee6DwZY0XLlyoNm3aqFOnTpK+fy7LvHnzNGXKlHqvO9Bc6TqHhISoa9euXudbtGihsLCwS/rhzZfv5ZycHCUkJKhLly6eG0jz8/OVn5/fEKUHDF/WeNy4cfrDH/6gX//615ozZ45Onjyp3/72t5o4caLCw8MbovxrEtCBZcuWLSorK9PEiRMvOVdRUaGysjKvvh8+C2Tfvn1auXKl4uPjVVpaanepAcuXNf7zn/+s8+fPa/LkyV4PMkpPT1deXl59lBuQfFnjCxcuKCsrSyUlJQoNDVW7du00d+5cPfTQQ/VZckDy9ecFfOfLGldXV2vatGk6duyYwsPD1aVLFxUUFGjIkCH1WXLA8WWNb775Zm3evFlTpkxRjx491KxZM40ePVpPPfVUfZbsN0HzHBYAABC8guI5LAAAILgRWAAAgPEILAAAwHgEFgAAYDwCCwAAMB6BBQAAGI/AAgAAjEdgAQAAxiOwAAAA4xFYAACA8QgsAADAeAQWAABgvP8H6XaZoF5z+kYAAAAASUVORK5CYII=\n", + "image/png": "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\n", "text/plain": [ "
" ] @@ -571,10 +503,10 @@ { "data": { "text/plain": [ - "{'mean': 7.469780220918766,\n", - " 'max': 7.719358435543357,\n", - " 'min': 7.246960302761897,\n", - " 'stdev': 0.06989052018543407,\n", + "{'mean': 7.47612464655258,\n", + " 'max': 7.681147051292993,\n", + " 'min': 7.279786459597185,\n", + " 'stdev': 0.07889928048753433,\n", " 'N': 200}" ] }, @@ -584,7 +516,7 @@ }, { "data": { - "image/png": "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\n", + "image/png": "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\n", "text/plain": [ "
" ] @@ -602,7 +534,7 @@ "id": "6d20f32a-8c7a-4456-aa41-d2989e33fd57", "metadata": {}, "source": [ - "I think that what's happening is that the clusters that are near the edge of the grid get arbitrarily cut off, and since the edge is a smaller percentage of the larger grid, the larger grid has a larger mean cluster size, one that is a better representative of what would happen on an infinite grid. But I can't say what the mean converges to." + "I think that what's happening is that the clusters that are near the edge of the grid get arbitrarily cut off, and since the edge is a smaller percentage of the larger grid, the larger grid has a larger mean cluster size, one that is a better representative of what would happen on an infinite grid. But I can't say exactly what the mean converges to." ] }, { @@ -612,7 +544,7 @@ "source": [ "# My First Random Cluster\n", "\n", - "The great thing about an *n* × 1 grid is that the clusters can't interfere with each other. Going left-to-right, once one cluster ends, the next one starts, and there's no looping back. But in an arbitrary 2D array there is looping back. So I want to get a feel for what the \"first\" cluster on an infinite grid looks like: start by placing the first color at the origin square, then recursively place random colors in neighboring squares, extending the cluster, but stop when the cluster is completely surrounded by squares of the second color. " + "The great thing about an *n* × 1 grid is that the clusters can't interfere with each other. Going left-to-right, once one cluster ends, the next one starts, and there's no looping back. But in an arbitrary 2D array there is looping back. So I want to get a feel for what the \"first\" cluster on an infinite grid looks like: start by placing a color at the origin square, then recursively place random colors in neighboring squares, extending the cluster, but stop when the cluster is completely surrounded by squares of the other color. " ] }, { @@ -622,16 +554,15 @@ "metadata": {}, "outputs": [], "source": [ - "def random_cluster(colors='@.') -> Grid:\n", - " \"\"\"Create just enough of a grid to cover a single random cluster of the first color.\"\"\"\n", - " c = colors[0]\n", - " cluster = {(0, 0): c}\n", - " Q = neighbors((0, 0))\n", + "def random_cluster(colors='@.', start=(0, 0)) -> Grid:\n", + " \"\"\"Create just enough of a grid to create a single random cluster.\"\"\"\n", + " cluster = {}\n", + " Q = [start] # A queue of the frontier squares of the cluster\n", " while Q:\n", " sq = Q.pop()\n", " if sq not in cluster:\n", " cluster[sq] = random.choice(colors)\n", - " if cluster[sq] == c:\n", + " if cluster[sq] == cluster[start]:\n", " Q.extend(neighbors(sq))\n", " return cluster" ] @@ -651,15 +582,17 @@ "metadata": {}, "outputs": [], "source": [ - "def show(cluster: Grid, color='@') -> int:\n", + "def show(cluster: Grid) -> int:\n", " \"\"\"Print a representation of the grid and return the size of the one cluster.\"\"\"\n", " xs = sorted({x for (x, y) in cluster})\n", " ys = sorted({y for (x, y) in cluster})\n", " for y in ys:\n", " print(*[cluster.get((x, y), ' ') for x in xs])\n", - " return cluster_size(cluster, color)\n", + " return cluster_size(cluster)\n", "\n", - "def cluster_size(cluster, color='@') -> int: return Counter(cluster.values())[color]" + "def cluster_size(cluster) -> int: \n", + " color = cluster[0, 0]\n", + " return Counter(cluster.values())[color]" ] }, { @@ -672,21 +605,16 @@ "name": "stdout", "output_type": "stream", "text": [ - " . \n", - ". @ . \n", - ". @ @ . . \n", - " . @ @ @ .\n", - " . @ . . \n", - " . @ @ . \n", - " . @ @ .\n", - " . @ . \n", - " . \n" + " . . \n", + ". @ @ .\n", + ". @ . \n", + " . \n" ] }, { "data": { "text/plain": [ - "12" + "3" ] }, "execution_count": 17, @@ -695,6 +623,7 @@ } ], "source": [ + "random.seed(37)\n", "show(random_cluster())" ] }, @@ -708,30 +637,45 @@ "name": "stdout", "output_type": "stream", "text": [ - " . . . . \n", - " . @ @ . . @ @ . \n", - ". @ @ @ @ @ @ @ @ . \n", - " . @ @ . . @ . @ @ . \n", - " . . . @ . . @ . \n", - " . @ . . @ @ @ . @ . \n", - " . @ . . @ @ . . \n", - " . @ @ @ . @ @ . \n", - " . @ @ @ @ . @ . \n", - " . @ @ @ . @ . \n", - " . @ @ . @ @ . \n", - " . @ @ @ @ . \n", - " . @ . @ . \n", - " . @ @ . . \n", - " . @ @ @ . @ .\n", - " . . @ @ @ .\n", - " . @ . . \n", - " . \n" + " . \n", + " . @ . \n", + " . @ @ . \n", + " . . @ . \n", + " . @ . . @ . \n", + " . @ @ @ @ @ . \n", + " . . @ @ . @ . @ .\n", + " . . @ . @ @ @ @ @ @ @ .\n", + " . @ . @ @ @ @ @ @ . . . \n", + " . @ @ @ @ . . @ . \n", + " . . @ . @ . @ . \n", + " . . @ . . . @ @ @ @ . \n", + " . . @ . . @ @ @ @ @ . . . \n", + " . @ @ @ @ . . @ @ . @ @ @ . \n", + " . @ @ @ . . . @ @ . @ @ . . \n", + " . . @ @ @ @ @ @ . @ . @ . @ @ @ @ . \n", + " . @ @ @ . . @ . @ @ @ @ . @ @ . . . \n", + " . @ . @ . . . @ . @ @ @ @ @ . \n", + " . . @ . . @ @ @ . . . . . \n", + " . @ @ . @ @ . @ . \n", + " . @ . @ @ @ . . \n", + " . @ @ @ @ @ @ @ . \n", + " . @ @ @ . @ @ @ . \n", + " . . . . . @ @ @ . @ @ . \n", + " . @ @ @ . . @ @ @ @ @ @ . @ . \n", + " . @ @ @ . @ @ . . @ @ . . \n", + " . @ @ @ @ @ @ . @ . \n", + " . @ @ . . . . @ @ @ . \n", + " . . . @ . . @ . \n", + ". @ @ . @ @ . . \n", + " . @ @ @ . \n", + " . @ @ @ . \n", + " . . . \n" ] }, { "data": { "text/plain": [ - "61" + "167" ] }, "execution_count": 18, @@ -753,29 +697,18 @@ "name": "stdout", "output_type": "stream", "text": [ - " . \n", - " . @ . . . . \n", - " . @ @ @ @ @ . \n", - " . . @ @ . @ @ @ . \n", - " . . @ @ @ @ . @ . . \n", - " . @ . . @ . @ @ @ . \n", - " . @ @ @ . . . @ @ . \n", - " . @ @ . . @ . . . . . \n", - " . . @ @ . . . . . @ . . @ @ @ @ .\n", - " . . . @ @ @ . @ @ @ @ . . @ . . . . @ . . . \n", - " . @ @ @ @ @ . . . . @ @ @ @ @ . @ @ @ @ @ . \n", - ". @ @ @ @ @ @ @ @ @ @ @ . . . @ . @ . @ @ . \n", - " . @ @ @ . @ @ @ . @ . . @ @ @ @ @ . . \n", - " . . @ . @ . . . . @ @ @ @ @ @ . \n", - " . . . @ @ @ . . . . \n", - " . @ @ . . \n", - " . . \n" + " . . \n", + ". @ @ . . \n", + " . @ @ @ .\n", + " . . @ .\n", + " . @ .\n", + " . \n" ] }, { "data": { "text/plain": [ - "100" + "7" ] }, "execution_count": 19, @@ -792,7 +725,7 @@ "id": "faf079b0-8534-4d92-aa8e-acd035dc5420", "metadata": {}, "source": [ - "We can see that there is a lof of variation in size and shape of the first clusters. We can also see that the shape will constrain other clusters: there are some small clusters of the second color that are trapped inside the cluster of the first color. Below I sample 10,000 first clusters and report on their sizes:" + "We can see that there is a lot of variation in size and shape of the first clusters. We can also see that the shape will constrain other clusters: there are some small clusters of the second color that are trapped inside the cluster of the first color. Below I sample 10,000 first clusters and report on their sizes:" ] }, { @@ -809,18 +742,12 @@ "cell_type": "code", "execution_count": 21, "id": "df7f6ae0-32aa-4002-8056-9de0dc2da07a", - "metadata": { - "collapsed": true, - "jupyter": { - "outputs_hidden": true - }, - "scrolled": true - }, + "metadata": {}, "outputs": [ { "data": { "text/plain": [ - "{'mean': 59.3772, 'max': 703, 'min': 1, 'stdev': 75.3564687329362, 'N': 10000}" + "{'mean': 59.6728, 'max': 917, 'min': 1, 'stdev': 75.3054688022268, 'N': 10000}" ] }, "execution_count": 21, @@ -829,7 +756,7 @@ }, { "data": { - "image/png": "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\n", + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjEAAAGdCAYAAADjWSL8AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjYuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8o6BhiAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAjNklEQVR4nO3dfXBU5d2H8e+alyXE5JQkJuuWgLFNEQ1aG2wIdQQFApSYdpwpanSlUwoq8rIFiqDPDLSjCdIp2E4qReqIFWycTqW11UZi1bQUAhhnW9586TRq0CyhbdgNGjcI9/NHxzNdguiGhOSO12fmzDRnf7s5h7uYaw57Nh5jjBEAAIBlzuvvAwAAAOgJIgYAAFiJiAEAAFYiYgAAgJWIGAAAYCUiBgAAWImIAQAAViJiAACAlZL7+wD6ysmTJ/Xuu+8qIyNDHo+nvw8HAAB8CsYYdXR0yO/367zzznytZdBGzLvvvqv8/Pz+PgwAANADLS0tGj58+BlnBm3EZGRkSPrvH0JmZmY/Hw0AAPg0otGo8vPz3Z/jZzJoI+ajf0LKzMwkYgAAsMyneSsIb+wFAABWImIAAICViBgAAGAlIgYAAFiJiAEAAFYiYgAAgJWIGAAAYCUiBgAAWImIAQAAVkooYlatWiWPxxO3+Xw+93FjjFatWiW/36+0tDRNnDhR+/fvj3uNWCymBQsWKCcnR+np6aqoqNChQ4fiZtrb2xUIBOQ4jhzHUSAQ0NGjR3t+lgAAYNBJ+ErMZZddptbWVnfbu3ev+9iaNWu0du1a1dTUaM+ePfL5fJoyZYo6OjrcmWAwqK1bt6q2tlbbt2/XsWPHVF5erhMnTrgzlZWVCoVCqqurU11dnUKhkAKBwFmeKgAAGFRMAlauXGmuuOKK0z528uRJ4/P5zOrVq919H3zwgXEcx/z85z83xhhz9OhRk5KSYmpra92Zd955x5x33nmmrq7OGGPMgQMHjCTT2NjozuzcudNIMq+++uqnPtZIJGIkmUgkksgpAgCAfpTIz++Er8S88cYb8vv9Kigo0E033aR//vOfkqTm5maFw2GVlZW5s16vVxMmTNCOHTskSU1NTTp+/HjcjN/vV1FRkTuzc+dOOY6jkpISd2bcuHFyHMedOZ1YLKZoNBq3AQCAwSuhiCkpKdEvf/lLPffcc9q4caPC4bDGjx+vf//73wqHw5KkvLy8uOfk5eW5j4XDYaWmpmrYsGFnnMnNze32vXNzc92Z06murnbfQ+M4jvLz8xM5NQAAYJnkRIanT5/u/u8xY8aotLRUX/jCF/TYY49p3Lhxkrr/6mxjzCf+Ou1TZ043/0mvs2LFCi1evNj9OhqN9mnIXLT8mYTm31w9o4+OBACAz6azusU6PT1dY8aM0RtvvOHepXTq1ZK2tjb36ozP51NXV5fa29vPOHP48OFu3+vIkSPdrvL8L6/Xq8zMzLgNAAAMXmcVMbFYTAcPHtSFF16ogoIC+Xw+1dfXu493dXWpoaFB48ePlyQVFxcrJSUlbqa1tVX79u1zZ0pLSxWJRLR79253ZteuXYpEIu4MAABAQv+ctHTpUl1//fUaMWKE2tradN999ykajWrWrFnyeDwKBoOqqqpSYWGhCgsLVVVVpaFDh6qyslKS5DiOZs+erSVLlig7O1tZWVlaunSpxowZo8mTJ0uSRo8erWnTpmnOnDnasGGDJGnu3LkqLy/XqFGjevn0AQCArRKKmEOHDunmm2/Wv/71L11wwQUaN26cGhsbNXLkSEnSsmXL1NnZqXnz5qm9vV0lJSXatm2bMjIy3NdYt26dkpOTNXPmTHV2dmrSpEnatGmTkpKS3JktW7Zo4cKF7l1MFRUVqqmp6Y3zBQAAg4THGGP6+yD6QjQaleM4ikQiffL+GN7YCwBA70vk5ze/OwkAAFiJiAEAAFYiYgAAgJWIGAAAYCUiBgAAWImIAQAAViJiAACAlYgYAABgJSIGAABYiYgBAABWImIAAICViBgAAGAlIgYAAFiJiAEAAFYiYgAAgJWIGAAAYCUiBgAAWImIAQAAViJiAACAlYgYAABgJSIGAABYiYgBAABWImIAAICViBgAAGAlIgYAAFiJiAEAAFYiYgAAgJWIGAAAYCUiBgAAWImIAQAAViJiAACAlYgYAABgJSIGAABYiYgBAABWImIAAICViBgAAGAlIgYAAFiJiAEAAFYiYgAAgJWIGAAAYCUiBgAAWImIAQAAViJiAACAlYgYAABgJSIGAABYiYgBAABWImIAAICViBgAAGAlIgYAAFiJiAEAAFYiYgAAgJWIGAAAYCUiBgAAWImIAQAAViJiAACAlYgYAABgJSIGAABYiYgBAABWImIAAICVzipiqqur5fF4FAwG3X3GGK1atUp+v19paWmaOHGi9u/fH/e8WCymBQsWKCcnR+np6aqoqNChQ4fiZtrb2xUIBOQ4jhzHUSAQ0NGjR8/mcAEAwCDS44jZs2ePHn74YV1++eVx+9esWaO1a9eqpqZGe/bskc/n05QpU9TR0eHOBINBbd26VbW1tdq+fbuOHTum8vJynThxwp2prKxUKBRSXV2d6urqFAqFFAgEenq4AABgkOlRxBw7dky33HKLNm7cqGHDhrn7jTF68MEHde+99+qGG25QUVGRHnvsMb3//vt64oknJEmRSESPPPKIfvzjH2vy5Mm68sortXnzZu3du1fPP/+8JOngwYOqq6vTL37xC5WWlqq0tFQbN27UH/7wB7322mu9cNoAAMB2PYqYu+66SzNmzNDkyZPj9jc3NyscDqusrMzd5/V6NWHCBO3YsUOS1NTUpOPHj8fN+P1+FRUVuTM7d+6U4zgqKSlxZ8aNGyfHcdyZU8ViMUWj0bgNAAAMXsmJPqG2tlavvPKK9uzZ0+2xcDgsScrLy4vbn5eXp7feesudSU1NjbuC89HMR88Ph8PKzc3t9vq5ubnuzKmqq6v1gx/8INHTAQAAlkroSkxLS4sWLVqkzZs3a8iQIR875/F44r42xnTbd6pTZ043f6bXWbFihSKRiLu1tLSc8fsBAAC7JRQxTU1NamtrU3FxsZKTk5WcnKyGhgb99Kc/VXJysnsF5tSrJW1tbe5jPp9PXV1dam9vP+PM4cOHu33/I0eOdLvK8xGv16vMzMy4DQAADF4JRcykSZO0d+9ehUIhdxs7dqxuueUWhUIhXXzxxfL5fKqvr3ef09XVpYaGBo0fP16SVFxcrJSUlLiZ1tZW7du3z50pLS1VJBLR7t273Zldu3YpEom4MwAA4LMtoffEZGRkqKioKG5fenq6srOz3f3BYFBVVVUqLCxUYWGhqqqqNHToUFVWVkqSHMfR7NmztWTJEmVnZysrK0tLly7VmDFj3DcKjx49WtOmTdOcOXO0YcMGSdLcuXNVXl6uUaNGnfVJAwAA+yX8xt5PsmzZMnV2dmrevHlqb29XSUmJtm3bpoyMDHdm3bp1Sk5O1syZM9XZ2alJkyZp06ZNSkpKcme2bNmihQsXuncxVVRUqKamprcPFwAAWMpjjDH9fRB9IRqNynEcRSKRPnl/zEXLn0lo/s3VM3r9GAAAGGwS+fnN704CAABWImIAAICViBgAAGAlIgYAAFiJiAEAAFYiYgAAgJWIGAAAYCUiBgAAWImIAQAAViJiAACAlYgYAABgJSIGAABYiYgBAABWImIAAICViBgAAGAlIgYAAFiJiAEAAFYiYgAAgJWIGAAAYCUiBgAAWImIAQAAViJiAACAlYgYAABgJSIGAABYiYgBAABWImIAAICViBgAAGAlIgYAAFiJiAEAAFYiYgAAgJWIGAAAYCUiBgAAWImIAQAAViJiAACAlYgYAABgJSIGAABYiYgBAABWImIAAICViBgAAGAlIgYAAFiJiAEAAFYiYgAAgJWIGAAAYCUiBgAAWImIAQAAViJiAACAlYgYAABgJSIGAABYiYgBAABWImIAAICViBgAAGAlIgYAAFiJiAEAAFYiYgAAgJWIGAAAYCUiBgAAWImIAQAAViJiAACAlYgYAABgJSIGAABYKaGIWb9+vS6//HJlZmYqMzNTpaWl+uMf/+g+bozRqlWr5Pf7lZaWpokTJ2r//v1xrxGLxbRgwQLl5OQoPT1dFRUVOnToUNxMe3u7AoGAHMeR4zgKBAI6evRoz88SAAAMOglFzPDhw7V69Wq9/PLLevnll3XdddfpG9/4hhsqa9as0dq1a1VTU6M9e/bI5/NpypQp6ujocF8jGAxq69atqq2t1fbt23Xs2DGVl5frxIkT7kxlZaVCoZDq6upUV1enUCikQCDQS6cMAAAGA48xxpzNC2RlZelHP/qRvvOd78jv9ysYDOruu++W9N+rLnl5eXrggQd0++23KxKJ6IILLtDjjz+uG2+8UZL07rvvKj8/X88++6ymTp2qgwcP6tJLL1VjY6NKSkokSY2NjSotLdWrr76qUaNGfarjikajchxHkUhEmZmZZ3OKp3XR8mcSmn9z9YxePwYAAAabRH5+9/g9MSdOnFBtba3ee+89lZaWqrm5WeFwWGVlZe6M1+vVhAkTtGPHDklSU1OTjh8/Hjfj9/tVVFTkzuzcuVOO47gBI0njxo2T4zjuzOnEYjFFo9G4DQAADF4JR8zevXt1/vnny+v16o477tDWrVt16aWXKhwOS5Ly8vLi5vPy8tzHwuGwUlNTNWzYsDPO5Obmdvu+ubm57szpVFdXu++hcRxH+fn5iZ4aAACwSMIRM2rUKIVCITU2NurOO+/UrFmzdODAAfdxj8cTN2+M6bbvVKfOnG7+k15nxYoVikQi7tbS0vJpTwkAAFgo4YhJTU3VF7/4RY0dO1bV1dW64oor9JOf/EQ+n0+Sul0taWtrc6/O+Hw+dXV1qb29/Ywzhw8f7vZ9jxw50u0qz//yer3uXVMfbQAAYPA668+JMcYoFoupoKBAPp9P9fX17mNdXV1qaGjQ+PHjJUnFxcVKSUmJm2ltbdW+ffvcmdLSUkUiEe3evdud2bVrlyKRiDsDAACQnMjwPffco+nTpys/P18dHR2qra3VSy+9pLq6Onk8HgWDQVVVVamwsFCFhYWqqqrS0KFDVVlZKUlyHEezZ8/WkiVLlJ2draysLC1dulRjxozR5MmTJUmjR4/WtGnTNGfOHG3YsEGSNHfuXJWXl3/qO5MAAMDgl1DEHD58WIFAQK2trXIcR5dffrnq6uo0ZcoUSdKyZcvU2dmpefPmqb29XSUlJdq2bZsyMjLc11i3bp2Sk5M1c+ZMdXZ2atKkSdq0aZOSkpLcmS1btmjhwoXuXUwVFRWqqanpjfMFAACDxFl/TsxAxefEAABgn3PyOTEAAAD9iYgBAABWImIAAICViBgAAGAlIgYAAFiJiAEAAFYiYgAAgJWIGAAAYCUiBgAAWImIAQAAViJiAACAlYgYAABgJSIGAABYiYgBAABWImIAAICViBgAAGAlIgYAAFiJiAEAAFYiYgAAgJWIGAAAYCUiBgAAWImIAQAAViJiAACAlYgYAABgJSIGAABYiYgBAABWImIAAICViBgAAGAlIgYAAFiJiAEAAFYiYgAAgJWIGAAAYCUiBgAAWImIAQAAViJiAACAlYgYAABgJSIGAABYiYgBAABWImIAAICViBgAAGAlIgYAAFiJiAEAAFYiYgAAgJWIGAAAYCUiBgAAWImIAQAAViJiAACAlYgYAABgJSIGAABYiYgBAABWImIAAICViBgAAGAlIgYAAFiJiAEAAFYiYgAAgJWIGAAAYCUiBgAAWImIAQAAViJiAACAlYgYAABgpYQiprq6WldddZUyMjKUm5urb37zm3rttdfiZowxWrVqlfx+v9LS0jRx4kTt378/biYWi2nBggXKyclRenq6KioqdOjQobiZ9vZ2BQIBOY4jx3EUCAR09OjRnp0lAAAYdBKKmIaGBt11111qbGxUfX29PvzwQ5WVlem9995zZ9asWaO1a9eqpqZGe/bskc/n05QpU9TR0eHOBINBbd26VbW1tdq+fbuOHTum8vJynThxwp2prKxUKBRSXV2d6urqFAqFFAgEeuGUAQDAYOAxxpiePvnIkSPKzc1VQ0ODrrnmGhlj5Pf7FQwGdffdd0v671WXvLw8PfDAA7r99tsViUR0wQUX6PHHH9eNN94oSXr33XeVn5+vZ599VlOnTtXBgwd16aWXqrGxUSUlJZKkxsZGlZaW6tVXX9WoUaM+8dii0agcx1EkElFmZmZPT/FjXbT8mYTm31w9o9ePAQCAwSaRn99n9Z6YSCQiScrKypIkNTc3KxwOq6yszJ3xer2aMGGCduzYIUlqamrS8ePH42b8fr+KiorcmZ07d8pxHDdgJGncuHFyHMedOVUsFlM0Go3bAADA4NXjiDHGaPHixbr66qtVVFQkSQqHw5KkvLy8uNm8vDz3sXA4rNTUVA0bNuyMM7m5ud2+Z25urjtzqurqavf9M47jKD8/v6enBgAALNDjiJk/f77+/ve/61e/+lW3xzweT9zXxphu+0516szp5s/0OitWrFAkEnG3lpaWT3MaAADAUj2KmAULFujpp5/Wiy++qOHDh7v7fT6fJHW7WtLW1uZenfH5fOrq6lJ7e/sZZw4fPtzt+x45cqTbVZ6PeL1eZWZmxm0AAGDwSihijDGaP3++nnrqKb3wwgsqKCiIe7ygoEA+n0/19fXuvq6uLjU0NGj8+PGSpOLiYqWkpMTNtLa2at++fe5MaWmpIpGIdu/e7c7s2rVLkUjEnQEAAJ9tyYkM33XXXXriiSf0u9/9ThkZGe4VF8dxlJaWJo/Ho2AwqKqqKhUWFqqwsFBVVVUaOnSoKisr3dnZs2dryZIlys7OVlZWlpYuXaoxY8Zo8uTJkqTRo0dr2rRpmjNnjjZs2CBJmjt3rsrLyz/VnUkAAGDwSyhi1q9fL0maOHFi3P5HH31U3/72tyVJy5YtU2dnp+bNm6f29naVlJRo27ZtysjIcOfXrVun5ORkzZw5U52dnZo0aZI2bdqkpKQkd2bLli1auHChexdTRUWFampqenKOAABgEDqrz4kZyPicGAAA7HPOPicGAACgvxAxAADASkQMAACwEhEDAACsRMQAAAArETEAAMBKCX1ODHoHt2cDAHD2uBIDAACsRMQAAAArETEAAMBKRAwAALASEQMAAKxExAAAACsRMQAAwEpEDAAAsBIRAwAArETEAAAAKxExAADASkQMAACwEhEDAACsRMQAAAArETEAAMBKRAwAALASEQMAAKxExAAAACsRMQAAwEpEDAAAsBIRAwAArETEAAAAKxExAADASkQMAACwEhEDAACsRMQAAAArETEAAMBKRAwAALASEQMAAKxExAAAACsRMQAAwEpEDAAAsBIRAwAArETEAAAAKxExAADASkQMAACwEhEDAACsRMQAAAArETEAAMBKRAwAALASEQMAAKxExAAAACsRMQAAwEpEDAAAsBIRAwAArETEAAAAKxExAADASkQMAACwEhEDAACsRMQAAAArETEAAMBKRAwAALBSwhHz5z//Wddff738fr88Ho9++9vfxj1ujNGqVavk9/uVlpamiRMnav/+/XEzsVhMCxYsUE5OjtLT01VRUaFDhw7FzbS3tysQCMhxHDmOo0AgoKNHjyZ8ggAAYHBKOGLee+89XXHFFaqpqTnt42vWrNHatWtVU1OjPXv2yOfzacqUKero6HBngsGgtm7dqtraWm3fvl3Hjh1TeXm5Tpw44c5UVlYqFAqprq5OdXV1CoVCCgQCPThFAAAwGCUn+oTp06dr+vTpp33MGKMHH3xQ9957r2644QZJ0mOPPaa8vDw98cQTuv322xWJRPTII4/o8ccf1+TJkyVJmzdvVn5+vp5//nlNnTpVBw8eVF1dnRobG1VSUiJJ2rhxo0pLS/Xaa69p1KhRPT1fAAAwSPTqe2Kam5sVDodVVlbm7vN6vZowYYJ27NghSWpqatLx48fjZvx+v4qKityZnTt3ynEcN2Akady4cXIcx50BAACfbQlfiTmTcDgsScrLy4vbn5eXp7feesudSU1N1bBhw7rNfPT8cDis3Nzcbq+fm5vrzpwqFospFou5X0ej0Z6fyAB20fJnEpp/c/WMPjoSAAD6V5/cneTxeOK+NsZ023eqU2dON3+m16murnbfBOw4jvLz83tw5AAAwBa9GjE+n0+Sul0taWtrc6/O+Hw+dXV1qb29/Ywzhw8f7vb6R44c6XaV5yMrVqxQJBJxt5aWlrM+HwAAMHD1asQUFBTI5/Opvr7e3dfV1aWGhgaNHz9eklRcXKyUlJS4mdbWVu3bt8+dKS0tVSQS0e7du92ZXbt2KRKJuDOn8nq9yszMjNsAAMDglfB7Yo4dO6Z//OMf7tfNzc0KhULKysrSiBEjFAwGVVVVpcLCQhUWFqqqqkpDhw5VZWWlJMlxHM2ePVtLlixRdna2srKytHTpUo0ZM8a9W2n06NGaNm2a5syZow0bNkiS5s6dq/Lycu5MAgAAknoQMS+//LKuvfZa9+vFixdLkmbNmqVNmzZp2bJl6uzs1Lx589Te3q6SkhJt27ZNGRkZ7nPWrVun5ORkzZw5U52dnZo0aZI2bdqkpKQkd2bLli1auHChexdTRUXFx342DQAA+OzxGGNMfx9EX4hGo3IcR5FIpE/+aels7hLqr+cCADDQJfLzm9+dBAAArETEAAAAKxExAADASkQMAACwEhEDAACsRMQAAAArETEAAMBKRAwAALASEQMAAKxExAAAACsRMQAAwEpEDAAAsBIRAwAArJTc3weAc4ffgA0AGEy4EgMAAKxExAAAACsRMQAAwEpEDAAAsBIRAwAArETEAAAAKxExAADASkQMAACwEhEDAACsRMQAAAArETEAAMBKRAwAALASEQMAAKxExAAAACsRMQAAwEpEDAAAsBIRAwAArETEAAAAKxExAADASkQMAACwUnJ/HwDscNHyZxKaf3P1jD46EgAA/osrMQAAwEpEDAAAsBIRAwAArETEAAAAKxExAADAStydhD6X6J1NEnc3AQA+GVdiAACAlYgYAABgJSIGAABYiYgBAABWImIAAICViBgAAGAlIgYAAFiJz4nBgMdv0AYAnA5XYgAAgJWIGAAAYCUiBgAAWImIAQAAViJiAACAlYgYAABgJW6xxqDG7dkAMHhxJQYAAFiJKzHAx+AqDgAMbFyJAQAAVuJKDNAHuIoDAH1vwF+Jeeihh1RQUKAhQ4aouLhYf/nLX/r7kAAAwAAwoK/EPPnkkwoGg3rooYf0ta99TRs2bND06dN14MABjRgxor8PD+gTXMUBgE9nQEfM2rVrNXv2bH33u9+VJD344IN67rnntH79elVXV/fz0QEDDwEE4LNkwEZMV1eXmpqatHz58rj9ZWVl2rFjR7f5WCymWCzmfh2JRCRJ0Wi0T47vZOz9hOb/9zh47sD93p/l5xatfC6h5+77wdQeP/fU5wPARz7675Ix5pOHzQD1zjvvGEnmr3/9a9z++++/33zpS1/qNr9y5UojiY2NjY2NjW0QbC0tLZ/YCgP2SsxHPB5P3NfGmG77JGnFihVavHix+/XJkyf1n//8R9nZ2aed723RaFT5+flqaWlRZmZmn38/fDzWYmBhPQYO1mLgYC0+njFGHR0d8vv9nzg7YCMmJydHSUlJCofDcfvb2tqUl5fXbd7r9crr9cbt+9znPteXh3hamZmZ/B9ygGAtBhbWY+BgLQYO1uL0HMf5VHMD9hbr1NRUFRcXq76+Pm5/fX29xo8f309HBQAABooBeyVGkhYvXqxAIKCxY8eqtLRUDz/8sN5++23dcccd/X1oAACgnw3oiLnxxhv173//Wz/84Q/V2tqqoqIiPfvssxo5cmR/H1o3Xq9XK1eu7PZPWjj3WIuBhfUYOFiLgYO16B0eYz7NPUwAAAADy4B9TwwAAMCZEDEAAMBKRAwAALASEQMAAKxExPSShx56SAUFBRoyZIiKi4v1l7/8pb8PaVCprq7WVVddpYyMDOXm5uqb3/ymXnvttbgZY4xWrVolv9+vtLQ0TZw4Ufv374+bicViWrBggXJycpSenq6KigodOnToXJ7KoFNdXS2Px6NgMOjuYy3OnXfeeUe33nqrsrOzNXToUH35y19WU1OT+zhrcW58+OGH+r//+z8VFBQoLS1NF198sX74wx/q5MmT7gxr0QfO+pccwdTW1pqUlBSzceNGc+DAAbNo0SKTnp5u3nrrrf4+tEFj6tSp5tFHHzX79u0zoVDIzJgxw4wYMcIcO3bMnVm9erXJyMgwv/nNb8zevXvNjTfeaC688EITjUbdmTvuuMN8/vOfN/X19eaVV14x1157rbniiivMhx9+2B+nZb3du3ebiy66yFx++eVm0aJF7n7W4tz4z3/+Y0aOHGm+/e1vm127dpnm5mbz/PPPm3/84x/uDGtxbtx3330mOzvb/OEPfzDNzc3m17/+tTn//PPNgw8+6M6wFr2PiOkFX/3qV80dd9wRt++SSy4xy5cv76cjGvza2tqMJNPQ0GCMMebkyZPG5/OZ1atXuzMffPCBcRzH/PznPzfGGHP06FGTkpJiamtr3Zl33nnHnHfeeaauru7cnsAg0NHRYQoLC019fb2ZMGGCGzGsxblz9913m6uvvvpjH2ctzp0ZM2aY73znO3H7brjhBnPrrbcaY1iLvsI/J52lrq4uNTU1qaysLG5/WVmZduzY0U9HNfhFIhFJUlZWliSpublZ4XA4bh28Xq8mTJjgrkNTU5OOHz8eN+P3+1VUVMRa9cBdd92lGTNmaPLkyXH7WYtz5+mnn9bYsWP1rW99S7m5ubryyiu1ceNG93HW4ty5+uqr9ac//Umvv/66JOlvf/ubtm/frq9//euSWIu+MqA/sdcG//rXv3TixIluv5QyLy+v2y+vRO8wxmjx4sW6+uqrVVRUJEnun/Xp1uGtt95yZ1JTUzVs2LBuM6xVYmpra/XKK69oz5493R5jLc6df/7zn1q/fr0WL16se+65R7t379bChQvl9Xp12223sRbn0N13361IJKJLLrlESUlJOnHihO6//37dfPPNkvh70VeImF7i8XjivjbGdNuH3jF//nz9/e9/1/bt27s91pN1YK0S09LSokWLFmnbtm0aMmTIx86xFn3v5MmTGjt2rKqqqiRJV155pfbv36/169frtttuc+dYi7735JNPavPmzXriiSd02WWXKRQKKRgMyu/3a9asWe4ca9G7+Oeks5STk6OkpKRuldzW1tatuHH2FixYoKefflovvviihg8f7u73+XySdMZ18Pl86urqUnt7+8fO4JM1NTWpra1NxcXFSk5OVnJyshoaGvTTn/5UycnJ7p8la9H3LrzwQl166aVx+0aPHq23335bEn8vzqXvf//7Wr58uW666SaNGTNGgUBA3/ve91RdXS2JtegrRMxZSk1NVXFxserr6+P219fXa/z48f10VIOPMUbz58/XU089pRdeeEEFBQVxjxcUFMjn88WtQ1dXlxoaGtx1KC4uVkpKStxMa2ur9u3bx1olYNKkSdq7d69CoZC7jR07VrfccotCoZAuvvhi1uIc+drXvtbtowZef/1195fk8vfi3Hn//fd13nnxP1KTkpLcW6xZiz7ST28oHlQ+usX6kUceMQcOHDDBYNCkp6ebN998s78PbdC48847jeM45qWXXjKtra3u9v7777szq1evNo7jmKeeesrs3bvX3Hzzzae9fXH48OHm+eefN6+88oq57rrruH2xF/zv3UnGsBbnyu7du01ycrK5//77zRtvvGG2bNlihg4dajZv3uzOsBbnxqxZs8znP/959xbrp556yuTk5Jhly5a5M6xF7yNiesnPfvYzM3LkSJOammq+8pWvuLf+ondIOu326KOPujMnT540K1euND6fz3i9XnPNNdeYvXv3xr1OZ2enmT9/vsnKyjJpaWmmvLzcvP322+f4bAafUyOGtTh3fv/735uioiLj9XrNJZdcYh5++OG4x1mLcyMajZpFixaZESNGmCFDhpiLL77Y3HvvvSYWi7kzrEXv8xhjTH9eCQIAAOgJ3hMDAACsRMQAAAArETEAAMBKRAwAALASEQMAAKxExAAAACsRMQAAwEpEDAAAsBIRAwAArETEAAAAKxExAADASkQMAACw0v8DjbzBwXX5iLQAAAAASUVORK5CYII=\n", "text/plain": [ "
" ] @@ -853,13 +780,17 @@ { "cell_type": "code", "execution_count": 22, - "id": "2f1dacba-d47c-46ab-be44-b5129e317ef4", + "id": "3dcbff13-ddb1-4ba0-9650-3b6b6122f76a", "metadata": {}, "outputs": [ { "data": { "text/plain": [ - "{'mean': 11.6556, 'max': 32, 'min': 1, 'stdev': 9.159878236846971, 'N': 5000}" + "{'mean': 7.851026539809714,\n", + " 'max': 20,\n", + " 'min': 1,\n", + " 'stdev': 5.816812011424741,\n", + " 'N': 3994}" ] }, "execution_count": 22, @@ -868,7 +799,7 @@ }, { "data": { - "image/png": "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\n", + "image/png": "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\n", "text/plain": [ "
" ] @@ -878,7 +809,16 @@ } ], "source": [ - "report(sorted(sizes)[:5000])" + "smaller = [s for s in sizes if s <= 20]\n", + "report(smaller, bins=range(22))" + ] + }, + { + "cell_type": "markdown", + "id": "e93bc126-d2c9-415d-97cf-ee07bd36502d", + "metadata": {}, + "source": [ + "We see that there are over 600 clusters of size 1. That makes sense; we start with one square, and about once every 24 times all 4 neighbors will be the other color, and 10000 / 16 = 625." ] }, {