From 0e547901533f64d71871335bf4c23a36223108c4 Mon Sep 17 00:00:00 2001 From: Sabina Firtala Date: Thu, 26 Mar 2026 20:01:29 +0100 Subject: [PATCH 1/2] Merge from main --- notebooks/02_rq_eda.ipynb | 1587 ++++++++++++++++++++++++++++++++++++- 1 file changed, 1554 insertions(+), 33 deletions(-) diff --git a/notebooks/02_rq_eda.ipynb b/notebooks/02_rq_eda.ipynb index 988df87..6ec711e 100644 --- a/notebooks/02_rq_eda.ipynb +++ b/notebooks/02_rq_eda.ipynb @@ -2,7 +2,7 @@ "cells": [ { "cell_type": "code", - "execution_count": 3, + "execution_count": 76, "id": "bbaa676b", "metadata": {}, "outputs": [], @@ -11,12 +11,12 @@ "import pandas as pd\n", "import numpy as np\n", "import matplotlib.pyplot as plt\n", - "\n" + "import seaborn as sns" ] }, { "cell_type": "code", - "execution_count": 4, + "execution_count": 2, "id": "f484ce3a", "metadata": {}, "outputs": [], @@ -30,7 +30,7 @@ }, { "cell_type": "code", - "execution_count": 5, + "execution_count": 3, "id": "8e245485", "metadata": {}, "outputs": [ @@ -229,7 +229,7 @@ "4 3.99 " ] }, - "execution_count": 5, + "execution_count": 3, "metadata": {}, "output_type": "execute_result" } @@ -264,7 +264,7 @@ }, { "cell_type": "code", - "execution_count": 6, + "execution_count": 4, "id": "2aa61a7e", "metadata": {}, "outputs": [], @@ -278,7 +278,7 @@ }, { "cell_type": "code", - "execution_count": 7, + "execution_count": 5, "id": "a245e29e", "metadata": {}, "outputs": [], @@ -290,7 +290,7 @@ }, { "cell_type": "code", - "execution_count": 8, + "execution_count": 6, "id": "e343f566", "metadata": {}, "outputs": [], @@ -304,7 +304,7 @@ }, { "cell_type": "code", - "execution_count": 9, + "execution_count": 7, "id": "b1adcdf8", "metadata": {}, "outputs": [ @@ -325,7 +325,7 @@ "Name: success, Length: 27075, dtype: float64" ] }, - "execution_count": 9, + "execution_count": 7, "metadata": {}, "output_type": "execute_result" } @@ -336,7 +336,7 @@ }, { "cell_type": "code", - "execution_count": 10, + "execution_count": 8, "id": "225e849a", "metadata": {}, "outputs": [ @@ -422,7 +422,7 @@ "4 5250 288 5538 0.947996 5250.0" ] }, - "execution_count": 10, + "execution_count": 8, "metadata": {}, "output_type": "execute_result" } @@ -433,7 +433,7 @@ }, { "cell_type": "code", - "execution_count": 11, + "execution_count": 9, "id": "185dc896", "metadata": {}, "outputs": [ @@ -815,7 +815,7 @@ "[10 rows x 21 columns]" ] }, - "execution_count": 11, + "execution_count": 9, "metadata": {}, "output_type": "execute_result" } @@ -828,7 +828,7 @@ }, { "cell_type": "code", - "execution_count": 12, + "execution_count": 10, "id": "0f631e8c", "metadata": {}, "outputs": [ @@ -855,7 +855,7 @@ }, { "cell_type": "code", - "execution_count": 13, + "execution_count": 11, "id": "e7c93631", "metadata": {}, "outputs": [ @@ -873,7 +873,7 @@ "Name: success, dtype: float64" ] }, - "execution_count": 13, + "execution_count": 11, "metadata": {}, "output_type": "execute_result" } @@ -884,7 +884,7 @@ }, { "cell_type": "code", - "execution_count": 14, + "execution_count": 12, "id": "bbe3c9ca", "metadata": {}, "outputs": [ @@ -894,7 +894,7 @@ "Text(0.5, 1.0, 'Success Distribution')" ] }, - "execution_count": 14, + "execution_count": 12, "metadata": {}, "output_type": "execute_result" }, @@ -916,7 +916,7 @@ }, { "cell_type": "code", - "execution_count": 15, + "execution_count": 13, "id": "a4c6b8f7", "metadata": {}, "outputs": [], @@ -927,7 +927,7 @@ }, { "cell_type": "code", - "execution_count": 16, + "execution_count": 14, "id": "50f728f8", "metadata": {}, "outputs": [ @@ -937,7 +937,7 @@ "Text(0.5, 1.0, 'Log Success Distribution')" ] }, - "execution_count": 16, + "execution_count": 14, "metadata": {}, "output_type": "execute_result" }, @@ -975,7 +975,7 @@ }, { "cell_type": "code", - "execution_count": 17, + "execution_count": 15, "id": "1c15bd6b", "metadata": {}, "outputs": [ @@ -1023,7 +1023,7 @@ }, { "cell_type": "code", - "execution_count": 18, + "execution_count": 16, "id": "db4a5a01", "metadata": {}, "outputs": [ @@ -1066,7 +1066,7 @@ }, { "cell_type": "code", - "execution_count": 19, + "execution_count": 17, "id": "d285cf1d", "metadata": {}, "outputs": [ @@ -1114,6 +1114,66 @@ "------------------------------------------------------------------------------------------------------------------------" ] }, + { + "cell_type": "markdown", + "id": "08e39ea4", + "metadata": {}, + "source": [ + "## Correlation Heatmap" + ] + }, + { + "cell_type": "code", + "execution_count": 78, + "id": "9d687e23", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAuwAAAJOCAYAAADs9FIZAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjgsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvwVt1zgAAAAlwSFlzAAAPYQAAD2EBqD+naQAApuVJREFUeJzs3Qd4U2UXB/B/ulu6KbSUVVbZG9nIBgcgorKUPURlCSIWlSEgKkMUUQERUVFRAUURlCEisvfeLbTQAi3de+R7ztsvaVJSLNjSW/L/PU+kublNk/ua9txzz3tenV6v14OIiIiIiDTJpqhfABERERER5Y0BOxERERGRhjFgJyIiIiLSMAbsREREREQaxoCdiIiIiEjDGLATEREREWkYA3YiIiIiIg1jwE5EREREpGEM2ImIiIiINIwBOxFRMaPT6dCuXbti/zMeZIMHD1bHMCQkpKhfChE9ABiwE5FVOHjwIIYNG4Zq1aqhRIkScHZ2RpUqVTBgwABs3rwZ1kaCcQkoiwt5rXJzdHREVFSUxX2io6PVuBr2/S+mT5+unmP79u3/6XmIiAoCA3YieqBlZWVhwoQJaNKkCb788ktUrlwZo0aNwrhx49C4cWNs2LABXbp0wcyZM4v6pWrK6dOn1fHSEjs7O6SlpWHVqlUWH5ftKSkpar+iNmfOHHUMy5YtW9QvhYgeAEX/W42IqBC98cYbeP/999GgQQP8+OOPKqtuKjk5GR999FGeWVtrVaNGDWiNjJ1er8eKFSswduzY2x7//PPPUb16dfX12bNnUZTKlCmjbkREBYEZdiJ6YF24cAHvvfceSpYsiU2bNt0WrAspoZg0aRJmzJhhtj0yMhLjx49HpUqVVBlG6dKl0bt3b5w4cSLPeuVLly5h/vz5qFWrlvoe2S4CAgLULSYmBqNHj0b58uVVFviLL74wPsexY8fQt29fFeQ5ODigYsWKGDNmTL5PJM6dO4dXX30VjRo1Uu/XyckJgYGBeO2115CQkGC2r7zWv/76y/i14WZ4vXeqYb+X4xIcHIwPP/xQnQTI98h7k+MtVz/u1pAhQ3DkyBEcOnTIbPvRo0dx+PBh9bglsbGxePfdd9G2bVv4+/urYyz/Dhw4EBcvXjTbV9634f+H9u3bG4+PjKHBv41p7hp2OdF47LHH1LbVq1eb/Tx57NFHH7X4GBGRYIadiB5YEjxlZmbi+eefh6+v7x33lUDS4ObNm2jRooUK5CR4k0Bagk7J0EsJze+//47WrVvf9hwSYO/ZswePP/44unfvroJZg9TUVHTo0EEFzz169FDBneE1rV+/XgW9NjY2eOKJJ1Twd+rUKZX5l5+1d+9eeHl53fH1r127FsuXL1cBprxmCYbltUiQKsH5jh07YG9vr/adNm2aOjaXL19WXxvIVYg7udfjIidE8hq6deuGrl274qefflI14lLeMnv2bNyNQYMGqasmkmWXkxMDee+2trYqAJfHcpPylKlTp6rj8+STT6p5DGfOnME333yjXrucAMiJhDCcuMhrlp9nCNQ9PT3NnvNOY5qbBOPyuurVq6f+f2zevLnx5y1cuFCdUMrP7dOnz10dDyKyEnoiogdUu3bt9PJrbsuWLXf1fUOGDFHfFxQUZLZ9w4YNanvVqlX1mZmZxu2DBg1S28uVK6e/fPnybc9XsWJF9XjXrl31SUlJZo9FRkbq3d3d9WXLltWHhISYPfbtt9+q7xs9erTZdtnWtm1bs21hYWH61NTU2372jBkz1P5ff/212Xb5/jv9CbD0M+71uFSqVEl/7do14/abN2/qPT099W5ubhZfc16vp3r16urrbt266b29vfUpKSnqvvwr97t3767uy36531tMTIw+Kirqtufdtm2b3sbGRj98+HCz7dOmTVPP8eeff1p8PXcaU9P3HhwcbLZ948aNep1Op2/ZsqU+IyNDf/jwYb2Dg4O+WrVq+vj4+HwdCyKyPiyJIaIHVkREhPq3XLly+f4eyfp+++23qqxEMrmmpKShc+fOqtTmn3/+sZhJrlChQp7PLeU5UoJjSiZ2xsXFqUmKhoyrgWSwJYv83Xff/evrlsmNUuaRm5RriC1btuC/+C/H5c033zSr5/bx8VFXEuLj4++p1nzo0KG4deuWytQL+Vfuy/a8eHh4wNvb+7btknGvXbv2PR8fS2N6J4888oia8Lxr1y5VrtSvXz9VEiPH1tXV9Z5eAxE9+FgSQ0RkQsokpNOIBHIuLi63HRvZLm0gpY66TZs2Zo81bdo0z2MpNeV169a9bbuUrQgpe8ldSy3ktUjduNwk0M2LYTKmlLpIPbnUbJvWiF+7dq3Ijot048nNcBIlNeB3S0prpNxIJplKCYn8K/dl+51Ii0YpP5FjLcczIyPD+Jilk51/k9eY/pt33nlHvZZ58+ap+1K2ZOkYEREZMGAnogeWn5+fCjSvXr1q7B7ybyTbLfKqRTZkig37mbpTnbwElJZ6g0tmWCxevPiOrysxMfGOAbt0TZGad6l/l3pqeZ2GunyZQCn11v/Ffzku7u7ut20ztF6UOQZ3S2rxn3vuORV8S6ZasuMvv/zyHds5/vDDDyq4lyy21NFLXbqceMiYGOr571ZeY/pvZFxkkqmc3EjQP3z48Lt+DiKyLgzYieiB1apVK5XJ3Lp1q5ocmB+G4PL69et3LLOxFITeKXjL6zHD8xw/fhx16tTBvbhx44YK+GVC4+7du80y4PJ6c3fAuRf/5bgUBlkEa8GCBWqyrlxJkPt3IpNcJTiWBbRk8SxT+Sk5suReF2eSDP/cuXNVeZF0AXrhhRfYHYaI7og17ET0wJKuG9I5ZOnSparDyZ0YMtDSelACu/379yMpKem2/QwrX/5bR5X8atasmfpXAu17Je0kpSSmU6dOt5Wr/P333xa/R47L3WS47/dx+TfSOlOOnVw9kY4rNWvWvOP+Um4k++QO1sPDw9Xx+6/HJ7+kbr9///7qaoAcs6eeegrff/+9KushIsoLA3YiemBVrVpV9SaXemUpQZAWhLlJXbZkaiUDa6hllomA8j0yEdSUtN6T1oXyvJK9LwjSN9zNzQ2vv/46Tp48edvjEhwb6tzzYpisKuUhpnXrYWFhCAoKsvg9hgmYoaGh+Xqd9/u45IcEuevWrVMtHf+NHCOZFGt6hUDGXrLb6enp//n45NeLL76oThCkfl2uqCxbtkyVMUlJk/TSJyKyhCUxRPRAmzVrlgrMZLVTqWOX0hgJlKQOWgJ4qX+WsgTZz8DQu1y2SRAsmVxZAEfqoCWDLZM7pWd6QShVqpTqEPLMM8+gfv36qouIZLMl4y8/U15Hy5YtVVCcF6kfl0ztmjVr0KRJE3Ts2FEFpr/++qv62tJkVjkO0j9dvk9OZiR7Lj9f+sfn5X4el/xm2eWWH9IjX24NGzbE008/rSacyiRZuTIh71sWXjJlWDBpypQp6kRKusxIH3ZD15178fXXX6ubHOOXXnpJbZP++rJNfp5k3uVKi6FfPhGRUVH3lSQiuh/279+vHzp0qOoV7uzsrHd0dNQHBATo+/fvr9+8efNt+0uv8LFjx6p+2/b29nofHx/9008/rT9+/Hi+e24byHPI7U7OnDmjHzZsmNpP+nJ7eXnp69atq17Dvn37/rVHuvTwnjhxonpP8t6kr/fMmTP1aWlpFvdPT0/Xv/rqq/oKFSro7ezs1D7yPu70MwryuPxbn/M79WH/N5b6sGdlZek//fRTfe3atfVOTk56Pz8/dbxv3LiRZ0/6L774Qo2BHE953HQM/21Mc7/3S5cuqX77ZcqUUccwt9dff13t/8orr+TrPRKRddHJf3LCdyIiIiIi0hLWsBMRERERaRgDdiIiIiIiDWPATkRERESkYQzYiYiIiIjyYceOHarTk7+/v+ok9dNPP/3r98iaC40aNVKrHEv7W1ld+W4xYCciIiIiyofExETVClZWl84PaR/8+OOPq9atR44cwfjx4zF8+HC1dsXdYJcYIiIiIqK7JBl2WbytZ8+eee4zefJkbNiwASdOnDBu69u3L2JiYu64vkZuzLATERERkdVKTU1FXFyc2U22FQRZDK1Tp05m27p27aq23w2udEr/2Qb76jyKGmPrzHNxLVo35Z+ifglkwem9p3hcNMbG1raoXwJZsP3HFg9kvLH/9X6YMWOG2bZp06Zh+vTp//m5IyIi4Ovra7ZN7stJQXJyMpydnfP1PAzYiYiIiMhqBQUFYcKECWbbZIKoljBgJyIiIiLN09nrCuV5JTgvrADdz88P169fN9sm993d3fOdXRe8bk5EREREVAhatGiBrVu3mm3bvHmz2n43mGEnIiIiIs2zsSucDPvdSEhIwIULF8zaNkq7Rm9vb1SoUEGV11y9ehVffvmlenzUqFH46KOP8Oqrr2Lo0KHYtm0bvv/+e9U55m4wYCciIiIizdPZF31hyIEDB1RPdQND7fugQYPUgkjh4eG4cuWK8fFKlSqp4Pzll1/GBx98gHLlyuGzzz5TnWLuBgN2IiIiIqJ8aNeuHfR6fZ6PW1rFVL7n8OHD+C8YsBMRERGR5tlooCSmqBT9tQUiIiIiIsoTM+xEREREZLVtHYsDZtiJiIiIiDSMGXYiIiIi0jwbK65hZ8BORERERJqnY0kMERERERFpETPsRERERKR5NlZcEsNJp0REREREGsYMOxERERFpns7WejPsDNiJiIiISPNsrDhgZ0kMEREREZGGMcNORERERJqns2GGnYiIiIiINIgZdiIiIiLSPJ2t9VZyW+87JyIiIiIqBphhJyIiIiLNs7HiLjEM2ImIiIhI83ScdEpERERERFrEDDsRERERaZ6NFZfEcNIpEREREZGGMcNORERERJqns+IMOwN2IiIiItI8nY31FoZY7zsnIiIiIioGmGEnIiIiIs3Tsa0jERERERFpETPsRERERKR5NlY86ZQ17EREREREGsYMOxERERFpns6Ka9gZsBMRERGR5unY1pGIiIiIiLSIGXZ6IHm3boLKE4fBo1EdOPmXxoGnXsT19Vvv/D0PN0Wtea/BtVY1pISG48KcTxD25TqzfSq+0B+VJwyDo18pxB07g5PjZyJ2//FCfjcPlgoj+qPS2KFw8PVB/IkzOD1pNmIPWj6GOjs7VJ44EmX7PwHHMr5IPB+Mc9PmI3LLTuM+tq4uqPbGOPh26wSHUt6IO3Yapye/jbhDJ+7juyre2jVyROdmTvAoYYOwG5n4bnMiQsIzLe5bxscWPdo4o4KfLXw8bPH9lkRsPZB6236erjr0aueC2lXs4WCnw83oTKz8LRGXIyw/L93uyUfLoG/PsvD2dMDFkER88NlFnD6fkPc4tiyJYf0qwq+0E66GJ+PTL0Ow51C08fEhfSqgQ2sflPZxREaGHmcvJmDZqpA7Pifdrucjvujbw1+Ny4XLifhweQjOXMj7GLZt4Y1hfSvAr5QjwsJTsOTry9h7OMbivhNGVkKPLn74aEUwftwQwcOfi86KS2I46ZQeSLYlXBB37CxOjJ2Rr/2dA8rhofVLELV9L3Y2eQLBi1ai7pJZ8Onc2rhPmWceRc25QTg/azF2Nn0S8cfOoNmG5SpIpPzx6/Uoarw9GRfeWYxdbZ5C/PGzaLJ2GRx8LB/Dam+OQ/khvXFq0mzsbNoNoZ+vRsNVi+BWr6ZxnzqLZqFk+5Y4NnIy/mnxBKK2/YOHfv4cjmVKc1jyoUkNBzzdwQUbdiZj9opYhN3IwNg+bnBzsfyH0cEOiIzJxLrtyYhNyLK4j4ujDpMGuCMzC1j0fTymfxaLH7YlITFFzzHJpw6tfPDSkEr4YvUVDJ94GBdCEjFvah14ethb3L9OdTdMnVADG7ZeV/v/vTcKs1+riUoVXIz7hF5LxsJlFzF4/CG8NOUYIm6kYP60OvBwZ+4uv9q3LIkXBwXgix/CMOLVY7gYkoS5b9SEZx7HsHZ1V0wdH4gNW29g+KRj2Ln/Fma9Wh2Vyjvftm/rpt6oVc0NN6PS8v16yHowYC9AP/74I+rWrQtnZ2eULFkSnTp1QmJiItq1a4fx48eb7duzZ08MHjzYeD81NRWTJ09G+fLl4ejoiKpVq2L58uXGx0+ePIlu3brB3d0dbm5uaNOmDS5evGh8/LPPPkPNmjXh5OSEGjVq4OOPPzY+lpaWhtGjR6NMmTLq8YoVK2LOnDnqMb1ej+nTp6NChQrq5/r7+2Ps2LEo7m7+vgPnpi3E9Z+35Gv/iiP7Ijk4DKdffRcJZy7h8serELHmd1QalzNGlcYPQejy7xG2ci0STl/E8RenITMpBeUHP1WI7+TBEjB6EEJX/oCrq9Yh8exFnBw/HZnJKSg7oJfF/f379sCl+UsR+ccOJIeEIXT5d7j5xw5UGpM9LjZOjvB9ojPOTZ2H6F0HkHTpCi7MWaz+rTC8331+d8VTp6ZO2Hk0FbuOpyE8KgurNiUhLR1oWc/R4v6SIV/zZzIOnE5DeqblALxrcydEx2WpjLpk6qNis3A6JAORMZYDfLpd7x5l8evmCGzcdgOXw5Ix/9MLSEnNxOMdfS0erqe7+WPf4Wh899NVtf/yb6/g3KUE9HqsjHGfLX/fxMFjsQi/noqQ0CSVxXUtYYcqFUtwCPLpme5lsGHLDWz686Y6zguWXkJKahYe62A5QfDUY2Ww70gMVq+/hitXk/H5d6E4H5yIJx/1M9vPx9sB44YFYNYH55EpZ7qUZ1tHm0K4FQc8rS4g4eHh6NevH9577z08+eSTiI+Px99//60C4vwYOHAgdu/ejQ8//BD169dHcHAwIiMj1WNXr17Fww8/rAL/bdu2qaD9n3/+QUZGhnp81apVmDp1Kj766CM0bNgQhw8fxogRI1CiRAkMGjRIPef69evx/fffq8A8NDRU3cSaNWvw/vvv47vvvkPt2rURERGBo0ePwtp4Nm+AyG27zbbd3LwTteZPUV/r7O3h0ag2Lr67JGcHvR6R23bBs3nD+/1yiyU5hu4NauPS/GU5G/V6RG3fDc+mDSx+j42jAzJTzMstslJS4NW8cfZz2tnCxs4uj30aFcbbeKDY2kCVtmzcnWzcJr+xzoSko3LZe//zUK+aA04Fp2NkT1dUK2+HmIQs/HUoVZ0Y0L+zs9MhsIorvl4TmjMueuDgsRjUru5m8Xtk+/frr5ltk0CxTdOSef4MKb2IT8xQ5TaUv3GpXtkV36y9aj4ux2NQK69xCXTDD7/ePi6STTfQ6YApY6riu5+vISQs57NIt9NZcUkMA/YCDNglgO7Vq5fKYAvJtufHuXPnVDC9efNmlZUXlStXNj6+ePFieHh4qKDa3j77cmhgYKDx8WnTpmH+/PnqZ4tKlSrh1KlTWLJkiQrYr1y5gmrVqqF169bQ6XTG1yfkMT8/P/Vz5bkloG/atCmsjaOvD1KvZ58gGch9ew83lcW19/JQgWHqjahc+0ShRPWcsaK8OZT0VMcw7WauY3gjCiUCK1n8nsitOxEwerAxe16yXQv4du8Mna2tejwzIQnRew+j6qsv4OjZi+q5yjzzuDoBkP3pzlxddLC10SE+0TyxEJeYBb+Slksv8qOUpw3aNnTEln0p6mQgwM8OfTq5ICNTjz0neLn/33i42cPOVofo2HSz7bdi0lGhbE6Jiympp74VY35so2PS4O1lPo4tmnhh2oQacHK0QVR0GiZOP4HY+OzkD/3buNjB1laHW7nGJVqNi3Me42Kvxs1s/9h0td2gX09/ZGbpseY31qxT3lgSU0AkK96xY0cVpD/zzDNYtmwZoqNzJvvcyZEjR2Bra4u2bdvm+biUwBiCdVNSciOlMcOGDYOrq6vxNmvWLGPJjJTeyHNUr15dlbv88ccfxu+X15qcnKxOECQrv27dOmPm3hIp3YmLizO7pet5+Y4Kx+lX30bSxRC0ObABXaKOoda8NxC2ah30WTn/z0ntuqSo2p/bgS6RR1Fx1HMI/3GD2T50f0nG8EpEBn7akYzQ65n4+2h2dr1tQycORRE7fDwWwyYcxotBx1QJzYxXauRZF0+FL7ByCTz9WBm889EFHu58tnXUFcKtOCger7IYkIBbMuQbN25ErVq1sGjRIhUgS2mLjY3NbaUx6ek5Z9xS834nd3o8ISF7ZrqcIEhQbridOHECe/bsUY81atRIvY6ZM2eq4Lx37954+umn1WNSM3/27FlV8y4/58UXX1TlN6avz5TUvku23/T2fdYtFHeSTZcsuym5nx4bj6yUVKRFRiMrIwOOpc0vLzv6lkRqhHlmnixLi4pRx9ChVK5jWLrkbVc3DNKjonG4/xhs9muEv2p3xN+NH1NZ9aSQMOM+ycGh2PfYwOx9anbAnvZ9oLOzN9uHLEtI0qvMnlsJ88vM7iVsEJt47yc8Mhk1PMq8G4zc93Lnn5x8Hb/4dHU1witXIJ2drbV8hUK2S5bdlJdk3aPNf5dLvfXViBScOhePdxdfQGamPs+6eMo9LhnqeHnnGhcvC1n0nHExz6ar/T1y9q9X002dMH3/aWNsXd1c3aTLzwsDA/Ddxyy3pBz87VmApNykVatWmDFjhqojd3BwUBnrUqVKqZIZg8zMTBVQG0hWPisrC3/99ZfF561Xr56qh7cURPv6+qqJopcuXVITVU1vUhpjIHXvffr0UYH96tWrVe36rVvZgbYE6t27d1e17tu3b1e19MePW26zFxQUhNjYWLNbb5vi3yUlZs8RlOzQ3GybT8eWiN5zRH2tT09H7KGT8OnQImcHnQ4l27dAzJ7D9/vlFktyDOOOnETJds3Nj2Hb5ojZl32c85KVmobU8BuqzaNMMr2x4fYWnZlJyUi9fhN2nu7w6djK4j6U65hlSSY8EzUDcgIKCd1rVLTHpav3XiZxMSwDvt7ZZUsGcv9WLFs65oe0XDx3MQGN63nmjIsOaFTXEyfPxlv8HtneyGR/8VB9T5w8F3fHn6WzARzsGQrkd1zOXkpAo7oeZuPSuK4HTuU1LufizfYXTep7qhMm8cdfkRg28SiGv5Jzky4xMkl10qzT+Xpd1lbDriuEW3HAGvYCsnfvXmzduhVdunRB6dKl1f2bN2+qzi0y+XPChAnYsGEDqlSpggULFiAmJqcHa0BAgKo1Hzp0qHHS6eXLl3Hjxg2VDZcOL5Kx79u3rwqYJast2XOpNZcsvpwgSKmLbH/kkUdU2cqBAwdUSY78XPl50iFGJqRKtv+HH35Qdeuenp744osv1AlEs2bN4OLigq+//loF8KZ17qakk4zcTNnLb3wNtnUsUbWC8b5LpXJwr18DabdiVY/16rMmwKmsL44Omawev7z0O1R88VnUmDMJoV+sgU/75qqN4/4ezxufI3jhCtT//F3EHDyB2P3HEDB2EOxKOCN05doieY/FUchHK1H30zmIPXwCsQeOI+DFgbB1ccbVr7P73ddd8g5Sr13HuRnvq/seTerBqYwv4o6fVv9WDXoJOp0Ngj/I6aAkwbn81ZQe7S6VK6L6zFfU14bnpDuTOvPB3UogJDxD3To2cYKDA7DrWPYEUXksJj4LP/2VbJyoKr3YhZ2NDp5uNihX2hapaXrc/H8XmC37UzB5gDsebeGkuskE+NuhTX1HfL2Jkxvz6/v1VxE0NlD1Sj99Ph7PdPOHs5Mtftt6XT0+ZWwgIm+lYunXl9X9H3+9hg9n1UWfHmWx++AtdGxdCtWruGLuJ9mlFlKzPuDp8vhn/y1Vuy712E8+5g8fb0f8uYtXCfPrh1/CETS6Ks5eTMTpCwl4+vEycHK0xcY/b6rHg8ZURWRUGpZ9kz2HZs1v4fhgRm307l4Gew5Gqz741SuXwPxPs0tW4xIy1M2UdImRKyah11Ly/browceAvYBIBnvHjh1YuHChquuWgFcmgj766KMqMy6dV6QTjJ2dHV5++WW0b9/e7Ps/+eQTTJkyRZWkREVFqcmfcl9Ii0jpDjNp0iRV5y7lNw0aNFDZfDF8+HAVbM+dO1ftIycIkrU3tJKUNpDSveb8+fPqex966CH89ttvKniXoP2dd95Rgb0E7vJ9v/zyi/qZxZlH4zposfUr4/1a87KPZeiXa3FsWBAcy5SCc/mcdmfSMlCC81rzgxAwZiBSwiJw/Pk3ELk5Z4Ge8B82qp7rgdPGZi+cdPQ09nUbjrRcE1EpbxFrN8LBxwvVpoxVJUcSiB94aqRxIqpzuTKASe25jaMjqr05Fs4B5ZGZmKRaOkrNekZsTjbLzt0NgdNfhpO/H9KiY3F9/R84/9ZC6O8wF4NyHDiTpiafymJI7v9fOOnD1fGIT8ou4/N2l5K+nP0lQH9zaE7GsEszZ3U7eyUdC76JN7Z+/GRtAp5s64zHWzmrdo7fb03CvlOccJpf2/6JhKe7PYb2rQBvLwdcCE7EK2+dME5E9S3laFZqeeJsPN56/yyG96+IEc9VRFh4Ml5/5zSCrySpx7Oy9KhYzhmPtK8BD3d7xMWnq8V+xrx+TLV4pPz5c1eUGpchfcurUhfpj//q7NM54+LjAH1WzricPJuAmR+cVwsnDe9fAVfDU/DGe2cRHMpuMPdCV0yy4YVBp89v30GiPGywr85jozG2ztq76kHAuin/8DBo0Om9p4r6JVAuNv/vBEXasv1Hk7LQInCu3yOF8ryB326C1vGvOhERERGRhrEkhoiIiIg0T1dMWjAWBut950RERERExQAz7ERERESkeTa21jvplAE7EREREWmezoq7xLAkhoiIiIhIw5hhJyIiIiLN03HSKRERERERaREz7ERERESkeTrWsBMRERERkRYxw05EREREmqez4gw7A3YiIiIi0jwdJ50SEREREZEWMcNORERERJqns+KSGC6cRERERESkYcywExEREZHm6ay4hp0BOxERERFpn44lMUREREREpEHMsBMRERGR5uk46ZSIiIiIiLSIGXYiIiIi0jydFU86td53TkRERERUDDDDTkRERESap7PiGnYG7ERERESkeTqWxBARERERkRYxw05EREREmqez4pIYTjolIiIiItIwZtiJiIiISPN0VpxhZ8BORERERNpnY72FIdb7zomIiIiIigFm2ImIiIhI83Q66y2JYYadiIiIiEjDmGEnIiIiIs3TsYadiIiIiIi0iBl2IiIiItI8Hds6EhERERFpmI31Tr203ndORERERHSXFi9ejICAADg5OaFZs2bYt2/fHfdfuHAhqlevDmdnZ5QvXx4vv/wyUlJS7upnsiSGiIiIiDRPp4GSmNWrV2PChAn49NNPVbAuwXjXrl1x9uxZlC5d+rb9v/nmG7z22mv4/PPP0bJlS5w7dw6DBw9WLSoXLFiQ75/LDDsRERERUT5IkD1ixAgMGTIEtWrVUoG7i4uLCsgt2bVrF1q1aoX+/furrHyXLl3Qr1+/f83K58YMO/1nts4879OazOSson4JZIGNLT8rWqTP0hf1S6BcdPb8rNDtdLqi/f8iLS0NBw8eRFBQkHGbjY0NOnXqhN27d1v8Hsmqf/311ypAb9q0KS5duoTffvsNAwYMuKufzYCdiIiIiLTPpnBKYlJTU9XNlKOjo7qZioyMRGZmJnx9fc22y/0zZ85YfG7JrMv3tW7dGnq9HhkZGRg1ahSmTJlyV6+Rp7BEREREZLXmzJkDDw8Ps5tsKwjbt2/H22+/jY8//hiHDh3C2rVrsWHDBsycOfOunocZdiIiIiKy2pVOg4KC1ERSU7mz68LHxwe2tra4fv262Xa57+fnZ/G533zzTVX+Mnz4cHW/bt26SExMxMiRI/H666+rkpr8YIadiIiIiKyWo6Mj3N3dzW6WAnYHBwc0btwYW7duNW7LyspS91u0aGHxuZOSkm4LyiXoF1Iik1/MsBMRERGR5uk00NZRMvGDBg1CkyZN1CRSaesoGXPpGiMGDhyIsmXLGktqunfvrjrLNGzYULWBvHDhgsq6y3ZD4J4fDNiJiIiIiPKhT58+uHnzJqZOnYqIiAg0aNAAmzZtMk5EvXLlillG/Y033lA91+Xfq1evolSpUipYnz17Nu6GTn83+XgiCza51+Rx0Ri2ddSmX9/eW9QvgSw4sfM4j4vG2Dk6FPVLIAv+/L5ZkR6X2HnjCuV5PV75AFrHDDsRERERaZ5OAyUxRYWTTomIiIiINIwZdiIiIiLSPhvrzTNb7zsnIiIiIioGmGEnIiIiIs3T6ay3hp0BOxERERFpn431FoZY7zsnIiIiIioGmGEnIiIiIs3Tsa0jERERERFpETPsRERERKR9Ouut5Lbed05EREREVAwww05ERERE2mfDto5ERERERJqlY0kMERERERFpEUtiiIiIiEj7bKy3JIaTTomIiIiINIwZdiIiIiLSPJ2N9eaZGbATERERkfbpWBJDREREREQaxAw7EREREWmfjfWWxFjvOyciIiIiKgaYYSciIiIi7dOxhp2IiIiIiDSIGXYiIiIi0jydFdewM2AnIiIiIu3TWW/Abr3vnIiIiIioGGCGnYiIiIi0z4aTTomIiIiISIOYYSciIiIizdOxhp0sGTx4MHr27FmkB2f69Olo0KBBkb4GIiIiIk2UxNgUwq0YKHYZ9nbt2qkAduHChYX6PVrxyiuvYMyYMUX9MoqlCiP6o9LYoXDw9UH8iTM4PWk2Yg8et7ivzs4OlSeORNn+T8CxjC8Szwfj3LT5iNyy07iPrasLqr0xDr7dOsGhlDfijp3G6clvI+7Qifv4roov79ZNUHniMHg0qgMn/9I48NSLuL5+652/5+GmqDXvNbjWqoaU0HBcmPMJwr5cZ7ZPxRf6o/KEYXD0K4W4Y2dwcvxMxO63PM5kWduGDuj8kBPcS+gQdiMTq7cm43JEpsV9y5S0QffWTqjga4eSHjb4YVsyth1MvW0/D1cdnmzrjNqV7OBgp8PNmCx8uTEJV65bfl663ZOPlUG/nuXg7eWAiyEJWLj0Ik6fT8jzULVr6YPhz1aEX2knhF1LxqdfBmPPwWjj40P6VkDHNqVQ2scRGRlZOHsxAcu+voxT5+J5+O9Cz66+6NO9DLw97XHxchI+/DwEZy4m5rl/2+beGNqnHPxKOSIsIgVLV13B3sOxFvd9eUQAenT2xUdfXMaa3yI4LmTELjGFJC0trUCex9XVFSVLliyQ57Imfr0eRY23J+PCO4uxq81TiD9+Fk3WLoODj7fF/au9OQ7lh/TGqUmzsbNpN4R+vhoNVy2CW72axn3qLJqFku1b4tjIyfinxROI2vYPHvr5cziWKX0f31nxZVvCBXHHzuLE2Bn52t85oBweWr8EUdv3YmeTJxC8aCXqLpkFn86tjfuUeeZR1JwbhPOzFmNn0ycRf+wMmm1Yrk6oKH8aV7fHU+2csWFXCt7+Mh5hNzMx9pkScHOxnHVysNchMiYLP+1IRmxClsV9XBx1mNTfDZmZenz0YyLeWhGPNduTkZSq57DkU4fWPhg9tDK+WH0FwyccxoXgRMyfXgeeHvYW969Tww3TXqmBDVsiMOzlQ/h7bxTeDqqFShVcjPuEXkvG+0svYtDYQ3jxtWOIuJGa/Zzulp+Tbte+hTdeGFgBK38Mw8jJJ1TA/t7rNeDpbjn/WTvQFW+Oq4rftt3EiMnHsXN/NGZOCkRAeefb9m39kBdqVXPFzVsFEz88kHQ2hXMrBorHqzQpUfnrr7/wwQcfQKfTqVtISIja1rRpUzg6OqJMmTJ47bXXkJGRccfvyczMxLBhw1CpUiU4OzujevXqap97JVn80aNHY/z48fDx8UHXrl3V9hMnTuDRRx9Vgbevry8GDBiAyMhI9djSpUvh7++PrCzzP3pPPPEEhg4dmmdJzGeffYaaNWvCyckJNWrUwMcff2x87Omnn1avw0Bej7znM2fOGE8kSpQogS1btqj7P/74I+rWrauOgZwYdOrUCYmJeWcKiouA0YMQuvIHXF21DolnL+Lk+OnITE5B2QG9LO7v37cHLs1fisg/diA5JAyhy7/DzT92oNKYwepxGydH+D7RGeemzkP0rgNIunQFF+YsVv9WGN7vPr+74unm7ztwbtpCXP85+/+9f1NxZF8kB4fh9KvvIuHMJVz+eBUi1vyOSuOyx0RUGj8Eocu/R9jKtUg4fRHHX5yGzKQUlB/8VCG+kwdLxyaO+OdYGnafSENEVBa+/SMZaelAizoOFveXzPvav1Jw4Ew6MvJIlndp5ojo+Cx8tSk7Ux8Vm4XTIRkq0Kf86fNEWfzyRwR+23odIaFJmPfJBaSkZuHxTr4W93+6e1nsO3QL3667isthyVj+zWWcu5SAXo/7G/fZsuMmDh6NQfj1FPWci5ZfgmsJO1QJKMFhyadnupXBhq03sGl7JC5fTcaCZcFIScvCo+1LWdz/qcf8sO9IDFb/Eo4rV1OwYnUYzl9KwpOPmI+jj5c9xg4NwOwPLyIzgye2VMwDdgmoW7RogREjRiA8PFzd7O3t8dhjj+Ghhx7C0aNH8cknn2D58uWYNWtWnt9Tvnx5FSSXK1cOP/zwA06dOoWpU6diypQp+P777+/59a1cuRIODg74559/8OmnnyImJgYdOnRAw4YNceDAAWzatAnXr19H79691f7PPPMMoqKi8Oeffxqf49atW2q/Z5991uLPWLVqlXqts2fPxunTp/H222/jzTffVD9btG3bFtu3bzfuLycrcgJh2LZ//36kp6ejZcuW6lj069dPnRzIc8k+vXr1gl5fvH9Z6Ozt4d6gNqL+3J2zUa9H1Pbd8GxqeT6AjaMDMlPML+tnpaTAq3nj7Oe0s4WNnV0e+zQqjLdh9TybN0Dktt3mQf/mnfBq3sA4zh6NaiNy666cHfR6RG7bBc/mDa3++OWHrQ1Qwc8WZy5n5BxCQN2v7H/vFZP1qtjjckQGhvdwwXsvumPKQFe0qmf5BIBuZ2enQ2AVNxVcG8dFDxw4GoPa1d0tHrI61d3U46b2HY5W2/P6GT26+iE+IQMXgvMusyGTY2arQ2DlEjh4PM5sXA4dj0XtQMvHuVagq9n+Yr+MYzVX432dDggaUwWr119DSFgyD/md6HSFcysGilUNu4eHhwqIXVxc4Ofnp7a9/vrrKgD/6KOPVCZZMs7Xrl3D5MmTVWBr6XuEra0tZszIuTQvmfbdu3ergN0QUN+tatWq4b333jPel5MGCdYlqDb4/PPP1es9d+4cAgMDVfb9m2++QceOHY0Zbwmw27dvb/FnTJs2DfPnz1eBteF1ywnHkiVLMGjQIJXpHzduHG7evAk7Ozv1mAT0EoyPGjVK/SsnN3I8JOsuVyLkuSpWrKieT7LtxZ1DSU8VXKfdjDLbnnojCiUCK1n8nsitOxEwerAxe16yXQv4du8Mna2tejwzIQnRew+j6qsv4OjZi+q5yjzzuDoBkP2p4Dn6+iD1eqT5GF6PhL2Hm7riYe/locZZxsJ8nyiUqF6ZQ5IPrs462NroEJdknvmW+77e9/7nwcfTBg83cMTWA6nYtCcRAX626N3BWZXI7DmZzrH5Fx7u9io4vBVjXhoRHZOGiuVuL6UQ3p4OuBVjfmzlvtS/m2rZxFuVzjg52iAqOg0Tph1HbHzOCRvdaVzsYGurQ3Su4yz3K/jnNS72iI7NtX9sOrw8c8al3xP+yMwE1my8zsNPD0aG3RLJDEsGXYJ1g1atWiEhIQFhYWF3/N7FixejcePGKFWqlCpZkRKVK1fuPfiS5zIlGX/JnstzG25yQiEuXryo/pVM+po1a5CammrMoPft2xc2NrcPjZSqyPdJKY/pc8qJgeH56tSpA29vb5VZ//vvv9UJQ7du3dR9If9KUC/q16+vThQkSJds/7JlyxAdnTNByRJ5nXFxcWa3NH3xv8x9+tW3kXQxBG0ObECXqGOoNe8NhK1aB71JuZLUrsuZePtzO9Al8igqjnoO4T9uMNuHiLITVjK59Oe/U9Qk1p3H0lTZTZsGjjw8RezQ8RgMHX8IL0w+ir2HojHj1Zp51sVT4Qus5IKnHvPFux9n/w2nf2FjUzi3YqBYZdgL0nfffac6sEi2WgJ+Nzc3zJ07F3v37r3n55TacFNy0tC9e3e8++67t+0rtfZCHpcSlA0bNqjMtwTZ77//vsXnl+cTElg3a9bM7DG5YiDkxOXhhx9WmXSp6ZfgvF69eirQlnr6Xbt2qfdt+J7NmzerbX/88QcWLVqkrljIMZDMvSVz5swxuzIhnnUoieccLdfvFYW0qBhkZWTAoZT5ZF3H0iVvy9gapEdF43D/Mao0xt7bE6nhNxA4YyKSQnJO+pKDQ7HvsYGwdXGGnZsrUq/fRP0VC8z2oYIjYyVZdrMx9PVBemw8slJSkRYZrcZZxtV8n5JIjbA8zmQuIVmPzCw93F3kD1ZOQbrcj0u899K42AQ9IqLMC9wjbmWiYSADw3wdvziZH6BXWXNTkpWNirZ8hUKy8ZLNNSX3b0WbZ+mlDv5qRIq6SXeYbz5pgm6dfPH1Gv4e+/dxyVBXibxyHWe5n/vqRs64pMMr1wmR3JerJaJuTXc16Xf1xzllfJLFl4mtTz/mh36jj/zr67IquuIRXBeGYvfOpbxFJowayORLKWUxrbuWGnIJwKVG3dL3GPaROu4XX3xRZaGrVq1qzFIXlEaNGuHkyZMICAhQz296MwT3MnFUSlIks/7tt9+qya/yfZbIpFWZpHrp0qXbns80wDbUsctNAnbJ1ksQLyckErjLFQgDCfDlvgThhw8fVsdq3TrztnmmgoKCEBsba3br7aCtLjb69HTEHTmJku2a52zU6VCybXPE7LvzL7+s1DQVrEubR5lkemPD7W0HM5OSVbBu5+kOn46tLO5D/13MniMo2aG5ealFx5aI3nPEOM6xh07Cp0ML83Fu3wIxew5zCPIhMwu4EpGJ6hVzcjdyrVLuX7p272USl65mwNc7O4lgUNrLBlFxvBqVHxkZepy7GI/G9TxzxkUHdf/kWfN6aIMTZ833F00aeKntdyItqO3ti10oUCTkJOrcpUQ0quNuNi6N6njgZB6tMU+dS0CjuubzDhrX88DJ/7fn3LwjEsMmHcfwV3Nu0iVm9fpwvDo7u1kEkSh2n1IJfiUDLJ1epNuKBNyhoaGqV7nUZP/888+qznvChAnGspLc3yMTTqXeXCaC/v7776qeXOq8ZUJmQXrppZfUJFKZ2CnPLScE8vOGDBlidgIhZTGSYZf69rwmmxpIYC1Z7g8//FC97uPHj2PFihVYsGCBcR8J0qV2XU4WWrdubdwmJwVNmjQxnizIMZH6ejkOUgq0du1aVfsuJ0F5kay9u7u72c1Bg2e8IR+tRLlBz8C//xMoEVgZtd+fpjLjV7/OPhmpu+QdBE572bi/R5N6qmZdWgl6tWiMJmuXqhXVgj9YbtxHgnOfTq3hXLGsau/Y9NcvVL92w3PSv7d1dK9fQ92ES6Vy6mun8tlXm6rPmoD6K3KuRl1e+h1cKpVHjTmTVE16xVH9VRvH4A++MO4TvHAFyg/rjbIDesK1RmXUWTwddiWcEbpyLYcjn6TOvHU9BzSvbQ8/bxv06+IMR3uorjFi0GMueKKNU8442gDlStuqm1zY83TVqa9Leeb8Hth6MBWVytjikWaOavtDNe3Rup4j/jp8e792smz1z1fRrYsfHmlfWtWtTxxVFc5ONvhtS3ad8+vjA/H8gADj/j/+chXNGnmp7jIVyjqrnus1qrhi7YZr6nGpWR/5XEXUCnSDbylHBFZxxWtjqsGnpCP+/IdXpPLrh1/D0a1jaXRt64MKZZ3w8vAAdWw3bb+pHg96qTKG9ytv3F96qTet74FnuvmhvL8TBj1TFtWrlMC6TdnjGJeQgZDQZLObdImRzHxoeAo/HrnZcOGkYkPKOWRyZa1atZCcnIzg4GD89ttvmDRpkqrJlvptqfF+44037vg9zz//vMoo9+nTR2WZJaiW4H/jxo0F9lolGy6ZfJkA26VLF5XdlsmdjzzyiFmNunSSkdd99uxZ9O/f/47POXz4cDVhVLLl8p4l+JYadGnfaCD3PT091aRWqXE3BOxykmCoXxcSbO/YsUMtKCW16PLapERIJsIWdxFrN8LBxwvVpoxVZRRxx0/jwFMjjRNRncuVAUxqz20cHVHtzbFwDiiPzMQk1dJRatYzYnOyJnbubgic/jKc/P2QFh2L6+v/wPm3FkL//xaidGcejeugxdavjPdrzZui/g39ci2ODQuCY5lScP5/8C6kveb+Hs+j1vwgBIwZiJSwCBx//g1Ebs5ZzCr8h42q53rgtLHZCycdPY193YYjLddEVMrbwbPpcHVJRrdWzsaFkxb9mIj4pOyrlt5uNqoThnEcXW3w+qCcjhidmzqp27krGXh/dXbWUFo5fvpTIno+7IzHWjohMjYLP/yZjP2nOeE0v7btjFSlEsP6V1QTR6WTyyszThonMPr6OMJ0+tCJM/GYMf8sRjxXESMHBKiFk6bMOYXgK0nq8awsPSqUc8GsDr5qUmtcfLpahGl00FHV4pHy58/dt9TxG9y7XPbCSSFJmPz2GUTHZv8dkEWpskw+LyfPJWDWhxcxtG85FchfDU/Bm3PPqcCc6G7o9MW9hx8VuU3ueWfkqWhkJrP0QIt+ffve58hQ4Tmxkyvjao2dI9uAatGf35vPn7vfUn7+qFCe1+mJnPVrtMpqJ50SERERUTGiKx490wuD9oqPNUjqu03bKOa+/ZdWkEREREREd8IMez5r0Y8cOXLHx4mIiIioENlYb56ZAXt+DpKdnWqdSERERER0vzFgJyIiIiLt01lvDTsDdiIiIiLSPp31lsRY7zsnIiIiIioGmGEnIiIiIu2zsd48s/W+cyIiIiKiYoAZdiIiIiLSPp31Tjplhp2IiIiISMOYYSciIiIi7dNZb56ZATsRERERaZ+OJTFERERERKRBzLATERERkfbZWG9JjPW+cyIiIiKiYoAZdiIiIiLSPL0V17AzYCciIiIi7dNZb2GI9b5zIiIiIqJigBl2IiIiItI+nfXmma33nRMRERERFQPMsBMRERGR5umteNIpM+xERERERBrGDDsRERERaZ/OevPMDNiJiIiISPt0LIkhIiIiIiINYoadiIiIiLTPxnpLYqz3nRMRERERFQPMsBMRERGR5umtuIadATsRERERaZ/OegtDrPedExEREREVA8ywExEREZHm6ZlhJyIiIiIiLWKGnYiIiIi0T2e9k05Zw05EREREpGHMsBMRERGR5umtuIadATsRERERaZ+OJTFERERERKRB1nttgYiIiIiKD51N4dzu0uLFixEQEAAnJyc0a9YM+/btu+P+MTExeOmll1CmTBk4OjoiMDAQv/322139TJbE0H+2bso/PIoaY2PLc3Et6jalWVG/BLLg+CNLeVw0xtbWtqhfApFFq1evxoQJE/Dpp5+qYH3hwoXo2rUrzp49i9KlS9+2f1paGjp37qwe+/HHH1G2bFlcvnwZnp6euBsM2ImIiIhI8/QaqGFfsGABRowYgSFDhqj7Erhv2LABn3/+OV577bXb9pftt27dwq5du2Bvb6+2SXb+bjENR0RERERWWxKTmpqKuLg4s5tss5QtP3jwIDp16mTcZmNjo+7v3r3b4ktev349WrRooUpifH19UadOHbz99tvIzMy8q7fOgJ2IiIiIrNacOXPg4eFhdpNtuUVGRqpAWwJvU3I/IiLC4nNfunRJlcLI90nd+ptvvon58+dj1qxZd/UaWRJDRERERJqnR+GUxAQFBam6dFMyObQgZGVlqfr1pUuXqrkZjRs3xtWrVzF37lxMmzYt38/DgJ2IiIiIrJajo2O+AnQfHx8VdF+/ft1su9z38/Oz+D3SGUZq100nUtesWVNl5KXExsHBIV+vkSUxRERERFQsVjrVF8ItvyS4lgz51q1bzTLocl/q1C1p1aoVLly4oPYzOHfunArk8xusCwbsRERERET5IKUzy5Ytw8qVK3H69Gm88MILSExMNHaNGThwoCqxMZDHpUvMuHHjVKAuHWVk0qlMQr0bLIkhIiIiIu3TFX2euU+fPrh58yamTp2qyloaNGiATZs2GSeiXrlyRXWOMShfvjx+//13vPzyy6hXr57qwy7B++TJk+/q5zJgJyIiIiLN00IfdjF69Gh1s2T79u23bZNymT179vynn1n0pypERERERJQnZtiJiIiISPP0GiiJKSrW+86JiIiIiIoBZtiJiIiISPt02qhhLwoM2ImIiIhI8/QsiSEiIiIiIi1ihp2IiIiINE8P6y2J4aRTIiIiIiINY4adiIiIiDRPzxp2IiIiIiLSImbYiYiIiEj7dNZbw86AnYiIiIg0T2/FUy+t950TERERERUDzLATERERkebprbgkhhl2IiIiIiINY4adiIiIiDRPb8VtHRmwExEREZHm6bnSKRERERERaREz7ERERESkeXorLomx3ndORERERFQMMMNORERERJqnZ1tHIiIiIiLSImbYiYiIiEjz9FbcJYYBOxERERFpnp6TTomIiIiISIuYYSciIiIizdNbcUkM2zoSEREREWkYA/YC1q5dO4wfPx6Fbfr06WjQoEGh/xwiIiIirdSw6wvhVhywJKYY0Ol0WLduHXr27Gnc9sorr2DMmDFF+rq0rl0jR3Ru5gSPEjYIu5GJ7zYnIiQ80+K+ZXxs0aONMyr42cLHwxbfb0nE1gOpt+3n6apDr3YuqF3FHg52OtyMzsTK3xJxOcLy89Lt2jZ0QOeHnOBeQqfGZfXW5DyPX5mSNuje2gkVfO1Q0sMGP2xLxraDt4+Lh6sOT7Z1Ru1KdtnjEpOFLzcm4cp1jsu/8W7dBJUnDoNHozpw8i+NA0+9iOvrt975ex5uilrzXoNrrWpICQ3HhTmfIOzLdWb7VHyhPypPGAZHv1KIO3YGJ8fPROz+4/xI3IVej/mjX6/y8PZywMXgBLy/5AJOn4/Pc//2rXww/LlK8CvthLBrSfjki2DsOXhLPWZrq8PI5wLQvIk3/P2ckZiYgQNHo/HJymBE3UrjuNyFHp1LoXd3P3h72OPilSR89EUozl5MzHP/h5t5YfAz/vAr5YirESlY9u1V7DsSa3x80qgAdG3rY/Y9+4/GIuid8xyXXPQsidGuzMxMZGVlFfXL0BxXV1eULFmyqF+GZjWp4YCnO7hgw85kzF4Ri7AbGRjbxw1uLpbr3xzsgMiYTKzbnozYBMv/v7k46jBpgDsys4BF38dj+mex+GFbEhJT9IX8bh4cjavb46l2ztiwKwVvfxmPsJuZGPtMibzHxV6HyJgs/LTjX8alvxsyM/X46MdEvLUiHmu2JyMpleOSH7YlXBB37CxOjJ2Rr/2dA8rhofVLELV9L3Y2eQLBi1ai7pJZ8Onc2rhPmWceRc25QTg/azF2Nn0S8cfOoNmG5XAo5Z2vn0FAh9alMHp4Faz4NgTDxh/EheAELHirLjw97C0enjo13DFtUi38+kc4ho47iL/3RGHO67VRqYKLetzJ0QaBVdywcvUVDB1/EK/POYkKZV3w7ht1eLjvQrvmXhg1oDy+WnMNo6acwqXLyXjntWrwdLec/6xVrQReH1MZm7ZHYlTQKfxzIAYzJlZBQDkns/0kgH9m1BHjbfaiSxwXMnPX1wE2bdqE1q1bw9PTUwWM3bp1w8WLF9VjLVu2xOTJk832v3nzJuzt7bFjxw51PzU1VWWHy5YtixIlSqBZs2bYvn27cf8vvvhCPff69etRq1YtODo64sqVK9i/fz86d+4MHx8feHh4oG3btjh06JDZzzpz5ox6bU5OTup7t2zZorLTP/30k3Gf0NBQ9O7dW/0Mb29vPPHEEwgJCcnXex88eLDKcs+YMQOlSpWCu7s7Ro0ahbS0vLMTX331FZo0aQI3Nzf4+fmhf//+uHHjhnpMr9ejatWqmDdvntn3HDlyRL3uCxcuICAgQG178skn1TbD/dwlMYbX9vbbb8PX11e9v7feegsZGRmYNGmSeq/lypXDihUrzH7WfzkeWtapqRN2Hk3FruNpCI/KwqpNSUhLB1rWc7S4v2R41/yZjAOn05CeaTnQ69rcCdFxWSqjLpn6qNgsnA7JUAEl5U/HJo7451gadp9IQ0RUFr79I1mNS4s6DnmOy9q/UnDgTDoy8kiWd2nmiOj4LHy1KTtTz3G5Ozd/34Fz0xbi+s9b8rV/xZF9kRwchtOvvouEM5dw+eNViFjzOyqNG2zcp9L4IQhd/j3CVq5FwumLOP7iNGQmpaD84Kfu8tVZr749y+GX38Px29brCAlNwtyPzyMlNQvdOvtZ3P+ZHmWx99AtfLsuDJfDkvDZqhCcu5iAp7qVVY8nJmXi5anHsG3nTYReTcbJs/FYsOQCalRzg28py78X6XZPPe6L37ZF4ve/onDlagoWLr+M1LQsPNLOPENu0OtRX5Ut//7X67hyLQVf/HANF4KT8ETX0mb7padnITo2w3hLSOTVQUv0VlwSc9evMjExERMmTMCBAwewdetW2NjYqGBSsuDPPvssvvvuOxWIGqxevRr+/v5o06aNuj969Gjs3r1b7Xfs2DE888wzeOSRR3D+fM6ln6SkJLz77rv47LPPcPLkSZQuXRrx8fEYNGgQdu7ciT179qBatWp47LHH1HZDJl4CVhcXF+zduxdLly7F66+/bvba09PT0bVrVxU8//333/jnn39Uplp+/p2CblPynk+fPq1OMr799lusXbtWBfB5kZ85c+ZMHD16VJ04SDAswbWQAHzo0KG3BdFy/+GHH1bBvJyoGLaFh4cb71uybds2XLt2TZ0cLViwANOmTVMnVF5eXuqYyMnF888/j7CwsAI7HlpkawNV2nI6JN24Tf6PPBOSjspl770KrF41BxUQjuzpirljPPH6EHe0rs8/dHc7LmcuZ5iPy+UMVPb/D+NSxR6XIzIwvIcL3nvRHVMGuqJVPcsnAPTfeTZvgMhtu8223dy8E17NsxMIOnt7eDSqjcitu3J20OsRuW0XPJs35BDkg52dDoFV3VTJiskhxIEj0ahd3T3PDLs8bmrv4Vtqe15cXWyRlaVHfELOZ5LuMC62OgRWKoFDJ+LMxkXuSybdEtluur/Yf0z2dzXbVr+WG374tD5WzK+DcUMrwN3VlkNBZu76r+RTT5lnSD7//HOVbT516pTK1MqESwmqDQH6N998g379+qngVDLlEnjKvxLEC8m2S9Zetkt22BBIfvzxx6hfv77x53To0MHs50pALlnhv/76SwWlmzdvVpl+CaQlky1mz56tsvKmJw9yYiEnAvJ6hPxceR75vi5duvzr+3dwcFDvWU4MateurbLYksGWoFxOXnKTgNygcuXK+PDDD/HQQw8hISFBBccSvE+dOhX79u1D06ZN1XuXY2bIusuxFfIaDe8rL5Ihl+eX11G9enW899576uRnypQp6vGgoCC88847anz69u1bIMdDi1xddLC10SE+0TxTHpeYBb+Sli8n50cpTxu0beiILftSsHF3MgL87NCnkwsyMvXYc6L4nuDcL67O2eMSl2R+RULu+3rfe8Du42mDhxs4qjkHm/YkIsDPFr07OKsSmT0nc07aqGA4+vog9Xqk2Ta5b+/hBhsnR9h7ecDGzg6pN6Jy7ROFEtUrcxjywcPdXgWHt6LN//+9FZOOiuWyS1xy8/Z0QHSM+e+h6Jh0tT2vcrMXBlfGlh03kJTMbG7+xsVOzQWIjjUfF8mIl/c3L3Ex8PK0V4+biomVccn5WyQZ+J37oxFxIw1lfB0xrE9ZvD05EGOnnkYWK/vM6K24hv2u/0pKJlwCTMnYRkZGGuvLJQivU6eOCvJWrVqlAvbg4GCVTV+yZIna5/jx4yoTHhgYaPacUiZjWo8tQXG9evXM9rl+/TreeOMNFUhKSYk8jwSj8nPF2bNnUb58ebOgVgJgU5LlljITySibSklJMZb1/Bs5iZBg3aBFixYq+JbSkooVK962/8GDB1X5ivzs6Ohos+MlZTty4vL444+rkwB5vb/88os6HnLl4W7JCYTpSYOUxsiYGNja2qrjbCjJuZfjIa9NbqYyM1Jha/fgZ5rlnOZyeIaqpxah1zPhX8oWbRs6MWAv6nGJyMTPf6eo+zKR1d/HFm0aODJgJ7JAgs63JteCxD7zPubExqK2fXfOlZHg0GQEX0nCVx/UU1n3wyfznmRM1uWuA/bu3burwHTZsmUq2JQAVIJCQwmFlMWMHTsWixYtUpniunXrqpuQwFaCRgli5V9Tkm02cHZ2NmZ8DaQcJioqCh988IH6+VLbLsHy3ZRuyM9v3LixOqHIzZDJLkhSPiQlJ3KTnyk/QwJ1uW/6uocPH44BAwbg/fffVxnuPn36mJ0U5JfMFTAlx9DSNsNJw70cjzlz5txWAtSo46to0sl87kJRSkjSIzNLD7cS5v8PuZewQWzivdeby6TH8CjzTJTcb1id5Rf5Gpfk7HFxd5GTypzjKPfjcl0Nubtx0SMi17hE3MpEw8B7v5pCeZNsumTZTcn99Nh4ZKWkIi0yGlkZGXAsbT4p3tG3JFIjzDPzZFlsnMzZ0MPby/z/YcnKRkVb/pt3KyYNXrmy6ZLdle25g/WZk2upTjJjXz/K7PpdiI3LUFfuvHJN/PXysFNXMyyR7fK4KZk4LFdL8hJ+Iw0xcenw93NkwJ6LPldsaE3uqoZdAmbJZEumu2PHjqhZs6bKGpuSSYuSoZUyFwnYJYA3aNiwocqMS4ZX6rNNb/9W7iH11XIiIHXrkkmWgF0y/AZSAiJZbsnEG+Su927UqJG6QiA18bl/vkxkzQ/JSicnZ2dYhdTTy8mGZPdzk0mwcsykDEWuONSoUcOY3TYl70km4H7yySfquJmW0QgJuuW4FbR7OR5SVhMbG2t2a9iu8PvO3w3p4nIlIhM1A3J+qcpHvEZFe1y6eu+1mhfDMuDrbX6iKfdvxfJy8t2MS/WKdmbjIvcvXbv3cZExzT0upb1sEBXHycCFIWbPEZTs0Nxsm0/Hlojec0R9rU9PR+yhk/Dp0CJnB50OJdu3QMyew4Xymh40GRl6nLsQj8b1vIzbJE5pXN8LJ8+a10MbnDgThyb1c/YXDzXwUttzB+vl/J0x/o1jiItn7fpdjUumHueCE9GojpvZuDSs7Y5T5y23dZTt8ripxnVl/4Q8f46Ptz3cXe3uGNRbK71eVyi3By5gl8mLUlIh9eNSSiGTHGUCqikJPGXy55tvvqkmZ0r9uoGUwkgAP3DgQDVZU0pmpHZbsrYbNmy448+WSabScUWeU8px5HkkE28gtepVqlRRmXiZzCoBvpxYCEO2Xr5HuszISYVMspSfLyU2ciJgmIj5byQzPmzYMFWz/9tvv6mJnTKR1lL9eoUKFVR5j1xtuHTpkup8I7XuucnVBqlll2BY3qdcOTAlnWFksmtERMRtJ0j/xb0cDzlRku44pjctlsNInblMCG1exwF+JW3Qv6sLHByAXceyy3kGdyuBnm2dzSZElittq252Njp4utmor6Vu3fic+1PU5MhHWzip7Q/VckCb+o7Yfuj2vuBkmdSZt67ngOa17eHnbYN+XZzhaA/VNUYMeswFT7RxsjguclFO+uDnHpetB1NRqYwtHmnmmD0uNe3Rup4j/jrMcclvW0f3+jXUTbhUKqe+dipfRt2vPmsC6q9417j/5aXfwaVSedSYM0nVpFcc1V+1cQz+4AvjPsELV6D8sN4oO6AnXGtURp3F02FXwhmhK9fyo5FP3/0Uhu5dy+CRDr6qbv2VF6vB2ckGG7ZEqMffeLk6nh9Yybj/D+uvolkjL9VdpkI5ZwztVxE1qrphza9Xs8fZVodZr9VC9aqueGveacifLMnYy00muVL+rNlwHY+1L4XOD5dEBX8njBtaUbXM3PRXdgJx8gsBGNY3uzOPWLvxOh6q746nH/dVde4Dn/JHYGUX/Px7dvJOvndk/3KoWbUEfH0c0LC2G96aWBXXrqfiwFHLJ2dkne6qJEaCUunuIgGdlMFIVlsmOcrqnrkDQckaS6cTCVpNScnHrFmzMHHiRFy9elUFjM2bN1cTR+9k+fLlGDlypMoKSzZbJqjKhFXToFe6sEh5iUzqlAmec+fOVSU80uZRSJmJdFCR1pO9evVSHWakvaRcLZDAMz9kXwmq5b1JLbeckEiNuiVSViJtKmXSpxwnee0ymbRHjx637SsnAfKehgwZcttj8+fPVydGUoYkr7eg2i4WxPHQqgNn0tTkU1kMSUphpK75w9XxiE/KLr3wdrdRs/sNJEB/c2jOVYUuzZzV7eyVdCz4JruGUOqkP1mboBboebyVs2rn+P3WJOw7xQmn+XXwbDpcXZLRrZWzceGkRT8m5oyLm/m4eLja4PVBOdmszk2d1O3clQy8vzrBOC6f/pSIng8747GWToiMzcIPfyZj/2lmp/LDo3EdtNj6lfF+rXnZk9RDv1yLY8OC4FimFJz/H7yL5JAw7O/xPGrND0LAmIFICYvA8effQOTmncZ9wn/YqHquB04bm71w0tHT2NdtONJyTUSlvEn7RSmdGP5sgFo46cKlBEycdtxYeuFbyslsQqJk0mfMO40Rz1XCyIGVEHYtGUGzT6p6aFGqpAPaNM8uZfpiUROznzUm6AgOn8hZyIfytn1PtJp8Ovhpf1VydPFyklrgKOb/E0tL+ziajYtk2N/+KBhDepfF0D5lcTUiFdPmX0RIWPacG+nSU7mCszoBcC1hi6jodBw8FocVP1xFegZnnOamv/vmhg8Mnd60B+MDRrLs0pddrgZI9v2/kix4TEyMWV/3giIZbgmUpaxHJosWJ8+/k72SHmmHjaSmSXO6TWlW1C+BLJjzyFIeF41xcrXcJpGK1pZvzU/27rfzFy8XyvNWq3J70xCtufdeahq0bt06VU8uGXAJ0seNG4dWrVoVSLBeWCRLL4tLSZZeOsMUt2CdiIiI6H7Qs63jg0FKOqS8QzqxSKlNp06dVDlJfpl2qslt48aNKAyy+JKUw8iqpV9++WWh/AwiIiKi4k5vxQH7A10Sc7ckK58Xqe02neRKOVgSoz0sidEmlsRoE0titIclMdpU1CUxZy+GFsrzVq9ye6c/rXmgSmL+K2lnSERERETao7fiDDtnphERERERaRgz7ERERESkeXpm2ImIiIiISIuYYSciIiIizdPrrbeGnQE7EREREWmeniUxRERERESkRcywExEREZHm6ZlhJyIiIiIiLWKGnYiIiIg0T2/FGXYG7ERERESkeXor7hLDlU6JiIiIiDSMGXYiIiIi0rwsKy6JYYadiIiIiEjDmGEnIiIiIs3TM8NORERERERaxAw7EREREWme3oq7xDBgJyIiIiLN07MkhoiIiIiItIgZdiIiIiLSPL0Vl8SwrSMRERERkYYxw05EREREmqe34hp2BuxEREREpHl6lsQQEREREZEWMcNORERERJqXBevFSadERERERBrGDDsRERERaZ6eNexERERERKRFzLATERERkebp2daRiIiIiEi79CyJISIiIiIiLWJJDBERERFpnt6KS2LY1pGIiIiISMOYYSciIiIizcvSw2oxYCciIiIizdNbcUkMA3b6z07vPcWjqDF6a05DaNjxR5YW9UsgC4I2jeRx0ZhPBq8r6pdApCmsYSciIiKiYtHWUV8It7u1ePFiBAQEwMnJCc2aNcO+ffvy9X3fffcddDodevbsedc/kwE7EREREVE+rF69GhMmTMC0adNw6NAh1K9fH127dsWNGzfu+H0hISF45ZVX0KZNG9wLBuxEREREpHl6feHc7saCBQswYsQIDBkyBLVq1cKnn34KFxcXfP7553l+T2ZmJp599lnMmDEDlStXvqf3zoCdiIiIiOhfpKWl4eDBg+jUqZNxm42Njbq/e/fuPL/vrbfeQunSpTFs2DDcK046JSIiIiLNyyqkLjGpqanqZsrR0VHdTEVGRqpsua+vr9l2uX/mzBmLz71z504sX74cR44c+U+vkRl2IiIiIrLaSadz5syBh4eH2U22/Vfx8fEYMGAAli1bBh8fn//0XMywExEREZHVCgoKUhNJTeXOrgsJum1tbXH9+nWz7XLfz8/vtv0vXryoJpt2797duC0rK0v9a2dnh7Nnz6JKlSr5eo0M2ImIiIhI8/SFtMSIpfIXSxwcHNC4cWNs3brV2JpRAnC5P3r06Nv2r1GjBo4fP2627Y033lCZ9w8++ADly5fP92tkwE5ERERElA+SiR80aBCaNGmCpk2bYuHChUhMTFRdY8TAgQNRtmxZVVIjfdrr1Klj9v2enp7q39zb/w0DdiIiIiLSPH0hTTq9G3369MHNmzcxdepUREREoEGDBti0aZNxIuqVK1dU55iCxoCdiIiIiDQvq5BKYu6WlL9YKoER27dvv+P3fvHFF/f0M9klhoiIiIhIw5hhJyIiIiLN0+uLviSmqDDDTkRERESkYcywExEREZHVtnUsDphhJyIiIiLSMGbYiYiIiEjzsjTQ1rGoMGAnIiIiIs3TsySGiIiIiIi0iBl2IiIiItI8Pds6EhERERGRFjHDTkRERESal2XFNewM2ImIiIhI8/RWHLCzDzsRERERkYYxw05EREREmqe34j7szLATEREREWkYM+xEREREpHlZrGEnIiIiIiItYoadiIiIiDRPb8UZdgbsRERERKR5eisO2DnplIiIiIhIw5hhJyIiIiLNy9KzrSMREREREWkQM+xEREREpHl6K65hZ8BORERERJqnt+KAnZNOiYiIiIg0jAG7ienTp6NBgwZFNxpERERElOdKp1mFcCsOWBJj4pVXXsGYMWOKbjSoQD35aBn07VkW3p4OuBiSiA8+u4jT5xPy3L9dy5IY1q8i/Eo74Wp4Mj79MgR7DkUbHx/SpwI6tPZBaR9HZGTocfZiApatCrnjc5KFcXmsDPr1LAdvLxmXBCxc+m/j4oPhz2aPS9g1GZdg7DloMi59K6Bjm1L/H5es7HH5+jJOnYvn4c+nXo/5o1+v8tljEpyA95dcwOnzeR+/9q18MPy5Sv8fkyR88oWMyS31mK2tDiOfC0DzJt7w93NGYmIGDhyNxicrgxF1K41jkk/erZug8sRh8GhUB07+pXHgqRdxff3WO3/Pw01Ra95rcK1VDSmh4bgw5xOEfbnObJ+KL/RH5QnD4OhXCnHHzuDk+JmI3X+c43IXHm/nhV5dS8LLww7BoalY8m04zoWk5Ll/q8ZueO6J0vD1sce162n4Ys0NHDiR8zvPyVGHwb180byhG9xK2OJ6ZDp+2XYLG//K+T1H9MBk2NPS/vsfAldXV5QsWRIPuoI4VlrXoZUPXhpSCV+svoLhEw/jQkgi5k2tA08Pe4v716nuhqkTamDD1utq/7/3RmH2azVRqYKLcZ/Qa8lYuOwiBo8/hJemHEPEjRTMn1YHHu487833uLT2weihlbPHZcJhXAhOxPzpdxiXGm6Y9koNbNgSgWEvH1Lj8nZQrdvG5f2lFzFo7CG8+JqMS2r2c7pbfk7KPSalMHp4Faz4NgTDxh/EheAELHir7h3GxB3TJtXCr3+EY+i4g/h7TxTmvF7bOCZOjjYIrOKGlauvYOj4g3h9zklUKOuCd9+ow0N/F2xLuCDu2FmcGDsjX/s7B5TDQ+uXIGr7Xuxs8gSCF61E3SWz4NO5tXGfMs88ippzg3B+1mLsbPok4o+dQbMNy+FQyptjk09tmrhjeG9ffPvLTYybeQnBYSl4a3xFeLjZWty/RhVnvDqiHDbvjMHYty5hz5F4vP5SeVT0dzTuM7y3HxrVccX8z67ihakX8fOWKIzq54em9V05Lrno9bpCuRUHxTZgb9euHUaPHo3x48fDx8cHXbt2xYkTJ/Doo4+qwNvX1xcDBgxAZGSk2n/p0qXw9/dHVlaW2fM88cQTGDp0aJ4lMZ999hlq1qwJJycn1KhRAx9//LHxsaefflq9BgN5LTqdDmfOnDEGxiVKlMCWLVv+9f38+OOPqFu3LpydndVJQ6dOnZCYmGh8/PPPP0ft2rXh6OiIMmXKmP3cK1euqPch79vd3R29e/fG9evXjY8b3pe8l0qVKqn3ImJiYjB8+HCUKlVKfV+HDh1w9OhRPAh69yiLXzdHYOO2G7gcloz5n15ASmomHu/oa3H/p7v5Y9/haHz301W1//Jvr+DcpQT0eqyMcZ8tf9/EwWOxCL+eipDQJHy0IhiuJexQpWKJ+/jOirc+T5TFL39E4Let19UxnPeJjEsWHu+Ux7h0L4t9h27h23X/H5dvLmePy+P+xn227LiJg0djEH49RT3nouWXssclgOOSH317lsMvv4cbx2Tux+fVmHTr7Gdx/2d6lMVeNSZhuByWhM9WheDcxQQ81a2sejwxKRMvTz2GbTtvIvRqMk6ejceCJRdQo5obfEvlBCl0Zzd/34Fz0xbi+s///vdDVBzZF8nBYTj96rtIOHMJlz9ehYg1v6PSuMHGfSqNH4LQ5d8jbOVaJJy+iOMvTkNmUgrKD36Kw5FPPTuXxO9/x2DLrliEhqdh8dfhSE3LQudWnhb379HRGwdPJmDtH1EIi0jD1z/fxMUryejWwcu4T80qzti2KwbHzyXhRlS6en45EQis5MxxsTDpVF8It+Kg2AbsYuXKlXBwcMA///yDd955RwWcDRs2xIEDB7Bp0yYVtErwKp555hlERUXhzz//NH7/rVu31H7PPvusxedftWoVpk6ditmzZ+P06dN4++238eabb6qfK9q2bYvt27cb9//rr7/UyYNh2/79+5Geno6WLVve8X2Eh4ejX79+6sRBfo58f69evaD///9Fn3zyCV566SWMHDkSx48fx/r161G1alX1mJyASLAu70V+/ubNm3Hp0iX06dPH7GdcuHABa9aswdq1a3HkyBHjMblx4wY2btyIgwcPolGjRujYsaN6ruLMzk6HwCquOHA0xrhNDuXBYzGoXd3N4vfIdgn6TO07EoPage55/oweXfwQn5ihym0ov+NifpxlXGScald3z/PKx4Hc43I4Wm3Pc1y6+iE+IUNliikfY1JVjnG0+Zgcic57TGq4q8dN7T18S23Pi6uLLbKy9GpcqHB4Nm+AyG27zbbd3LwTXs2zk1A6e3t4NKqNyK27cnbQ6xG5bRc8mzfksOSDnS1QtaITjpxONPu8yP0aVXKu+pmqUdkFR06Z/404dDJRbTc4fTEZTRu4oaRn9tXautVd4O/rgMMn+beFchTra/nVqlXDe++9p76eNWuWCtYlqDbNSpcvXx7nzp1DYGCgyr5/8803Kig1ZLUlwG7fvr3F5582bRrmz5+vgmch2elTp05hyZIlGDRokMryjxs3Djdv3oSdnZ16TAJ6CbhHjRql/n3ooYfg4mL5g2wasGdkZKifU7FiRbVNsu0G8t4mTpyofpaBPK/YunWrCuKDg4PVexVffvmlysbLCYNhP8n2y3bJpoudO3di3759KmCXrL2YN28efvrpJ3Vc5OSguPJws4edrQ7Rselm22/FpKtL85ZInfutGPNSoeiYNHh7mZcFtGjihWkTaqjL/lHRaZg4/QRi4xmE5Gtc3LPHxdJxrljO+Q7jcvs4Sq21qZZNvFXpjGFcJkw7znG5mzGJvv0YVyyX92cl+rYxTFfbLXGw1+GFwZWxZccNJCVn5udl0T1w9PVB6vXsK8oGct/eww02To6w9/KAjZ0dUm9E5donCiWqV+Yxzwd3Vzs1RyMmzvx3vtwv52f56pHUucfE376/p0dO+PXptxEYM6AMVs4NVPOjJFm36KtwnDyfxHHJJauYZMMLQ7EO2Bs3bmz8Wko5JHsuZSG5Xbx4UQXskkkfMWKEKmuRIFUy6H379oWNze0XGqQcRb5v2LBh6nsMJLD28PBQX9epUwfe3t4qsy2Zfjlh6NatGxYvXqwel+0S1P+b+vXrq5MICdKltKdLly6q3MbLy0sF1NeuXTOeZOQmGXkJ1A3BuqhVqxY8PT3VY4aAXU4EDMG64XglJCTcVrOfnJys3ndeUlNT1c1UVmYabGwt/7F+0Bw+HothEw6rQKd7Z1/MeKUGnp98FDG5Tg7o/jp0PAZDxx/KHpcufpjxak08P+kIx6WISXDz1uRagA6Y9/H5on45RJrUvYM3qld2xluLrqiSmDqBLhjV3w9RMRk4apLNJ+tWrAN2qQ83kOCze/fuePfdd2/bT2q+hTwuZ64bNmxQgezff/+N999/3+Jzy/OJZcuWoVmzZmaP2dpmTy6RevWHH35YZdLlBECC83r16qmAVurpd+3apTrP/Bt5Pillkf3/+OMPLFq0CK+//jr27t2rrgAU9LEyvD85LqYlPQYS7Odlzpw5mDHDfBJUhepDULFm9jwALYiNT0dGph5euSbNeXva35bdNZDtuTOEXpLdzZV5lNreqxEp6iZdSL5Z3FjVxa9aG1YI7+TBEhuXPS6WjnNUruNsPi4WxjE67c7j8kkTdOvki6/XcFzyNSZetx9juVKR15h43TaGt3+2JFifObmW6iQz9vWjzK4XMsmmS5bdlNxPj41HVkoq0iKjkZWRAcfS5kkaR9+SSI0wz8yTZXEJGcjM1MMzV6MBuR+dK+tuEB2bAU+32/ePic0wXoEa+GRpzP44FAeOZ8cdIVdTUam8E3p1KcmAPRe9FWfYi3UNuympvz558iQCAgJUfbfpzRCsymRLKTuRzPq3336L6tWrq++zRCatyiRVqQfP/XxSGmNgqGOXmwTskq2XIH7u3LkqcG/VqlW+Xr8E/7KvBMOHDx9WGft169bBzc1NvScpfbFEJsSGhoaqm4GU5siEUsm03+l4RUREqFKe3O/vTicJQUFBiI2NNbuVD3wOWiKXFGUSXON6OSceOh3QqK6nmgBniWxvZLK/eKi+J06ei7vjz9LZyC/cB+ZjdB/GJf62cZH7J89aPs4nzprvL5o08FLb78RGB9hzXPI3JhfkGHuZj0l9r7zH5EwcmtTP2V88JGNyJu62YL2cvzPGv3EMcSwbK3Qxe46gZIfmZtt8OrZE9J7sOUv69HTEHjoJnw4tcnbQ6VCyfQvE7Dlc+C/wAZCRCVy4nIL6NUuYfV7k/pmLlstXzlxKQgOT/UVD2f9SkvGzYm8nnUrMv0/6Y8hzExk8MJGGTMqUyZIyeVNqt6Ws4/fff8eQIUOQmZlTNyllMZJhl/r2vCabGkjwLBnlDz/8UNXBS634ihUrsGDBAuM+EqRLgCwnC61btzZuk5OCJk2a3JbZtkQy6VJ7L5NlpeOLTAyVungJxg1dXqSWXl7H+fPncejQIZWFF9JNRkpp5L3IdqlLHzhwoDqRkJ+fF/m+Fi1aoGfPniqrHxISojL8ktmX15EXuZIgHWVMb1osh/l+/VXV5eKR9qVVffTE56vA2clWdcIQU8YGYuRz2fMFxI+/XkOzhp7o06MsKpR1Vj3Xq1dxxdrfwtXjUhs94tmKqBWY3ekisHIJTB5dDT7ejvhzF7NT+bX656vo1sVkXEZVhbOTDX7bkj0ur48PxPMDAnLG5ZeraNbIS3WXUePStwJqyLhsuGYcFxlH47hUccVrY6rBp6Qj/vyH45If3/0Uhu5dy+CRDr6qbv2VF6upMZFWmuKNl6vj+YE5SYof1mePiXSXqVDOGUP7VUSNqm5Y8+tVYwAy67VaqF7VFW/NOw2pOJSMvdxkkivlv62je/0a6iZcKpVTXzuVz75iXH3WBNRfkXNF+fLS7+BSqTxqzJmkatIrjuqv2jgGf/CFcZ/ghStQflhvlB3QE641KqPO4umwK+GM0JVrOSz59NPmKHRt44kOLTxQzs8BLz5bBk4ONtjyT/bk+AlD/THoydLG/ddvvYVGtV3xZGdvtX//7qVQNcAZv27LnridnJKF42cTMfTp0qgb6KJ6tXds6aGef/dhriWRm96Ku8QU65IYU5INl24xkydPVjXgkt2Wuu1HHnnErEZdOslI3fnZs2fRv3//Oz6ntDyUCaOSLZ80aZIKviU4lvaNBnJfSkikRt5QPy8Bu5wk5Kd+XUjQu2PHDixcuBBxcXHqdUuALpNkhUxwTUlJUeU7UmIjGXCpcTdk5n/++We14JNk9uW9yns2BPR5ke/77bffVIAuJzVyguDn56eeQ64uFHfb/olUfbiH9q2gJihKv+9X3jphnIgqwZ2hC4+QjO1b75/F8P4VMeK5iggLT8br75xG8JXsLIh0uJAA85H2NVSddFx8Os5cSMCY14+pVniUz3HZmT0uw/pX/P+4JOCVGSdzxsXHEXqTzqsnzsRjxvyzakxGDghQCydNmXPKbFwqlHPBrA6+xnGRRZhGBx3luOR7TG6qnuvDnw3IHpNLCZg47biaSJr9WXEym+glmfQZ805jxHOVMHJgJTUmQbNPGsekVEkHtGmefZXui0XmSYMxQUdw+EQsPy754NG4Dlps/cp4v9a8Kerf0C/X4tiwIDiWKQXn/wfvIjkkDPt7PI9a84MQMGYgUsIicPz5NxC5eadxn/AfNqqe64HTxmYvnHT0NPZ1G460XBNRKW9/H4hTPdefe6IUvNztcCk0FVM/uIKY+OzEYClve7PPy5mLyZj7WRgG9CytSl+u3UjD7MWhuHwtZy7Yu0vDMKiXL14ZXhauJWxVHftXP93gwkkWZBWT4Low6PSmUQvRPXj4yZw/CKQNemv+raZh+lzrQJA2BG0qvl2xHlSfDDZfoZW04ddleZfa3g+f3Xmx33s23HJfD015YDLsRERERPTg0ltxLuqBqWHXMqlLl3KZvG7yOBERERGRJcyw36f6esPqonk9TkRERER5y7LiqkIG7PfjIP+/dSIRERER3Rs9S2KIiIiIiEiLmGEnIiIiIs3TM8NORERERERaxAw7EREREWleFjPsRERERESkRcywExEREZHm6QutiF0HrWPATkRERESap2dJDBERERERaREz7ERERESkeVlWvNKpTVG/ACIiIiIiyhsz7ERERESkeXorrmFnwE5EREREmpdlxQE7S2KIiIiIiDSMGXYiIiIi0jw9M+xERERERKRFzLATERERkebps6x3pVPWsBMRERERaRgz7ERERESkeVlWXMPOgJ2IiIiINE9vxQE7S2KIiIiIiDSMGXYiIiIi0rwsK66JYYadiIiIiEjDmGEnIiIiIs3TW2+CnQE7EREREWmf3ooDdpbEEBERERFpGEtiiIiIiEjzsqw4xc4MOxERERGRhjHDTkRERESap8+C1WKGnYiIiIhIw5hhJyIiIiLN01txDTsDdiIiIiLSvCyWxBARERERkRYxw05EREREmqe34pIYTjolIiIiItIwZtiJiIiISPOyrDfBzoCd/jsbW1seRo3R2fPimRbZ8rOiSZ8MXlfUL4FyeeGLJ3lMtGjZ2SL98Xorjtj5V52IiIiISMNYEkNEREREmqe33gQ7M+xERERERPm1ePFiBAQEwMnJCc2aNcO+ffvy3HfZsmVo06YNvLy81K1Tp0533D8vLIkhIiIiIs3LytIXyu1urF69GhMmTMC0adNw6NAh1K9fH127dsWNGzcs7r99+3b069cPf/75J3bv3o3y5cujS5cuuHr16l39XAbsRERERET5sGDBAowYMQJDhgxBrVq18Omnn8LFxQWff/65xf1XrVqFF198EQ0aNECNGjXw2WefISsrC1u3bsXdYA07EREREVntwkmpqanqZsrR0VHdTKWlpeHgwYMICgoybrOxsVFlLpI9z4+kpCSkp6fD29v7rl4jM+xEREREpHn6rMK5zZkzBx4eHmY32ZZbZGQkMjMz4evra7Zd7kdEROTrPUyePBn+/v4qyL8bzLATERERkdUKCgpSdemmcmfXC8I777yD7777TtW1y4TVu8GAnYiIiIg0L6uQSmIslb9Y4uPjoxbAu379utl2ue/n53fH7503b54K2Lds2YJ69erd9WtkSQwRERER0b9wcHBA48aNzSaMGiaQtmjRIs/ve++99zBz5kxs2rQJTZo0wb1ghp2IiIiIrHbS6d2Q0plBgwapwLtp06ZYuHAhEhMTVdcYMXDgQJQtW9ZYA//uu+9i6tSp+Oabb1TvdkOtu6urq7rlFwN2IiIiItK8rLvsmV4Y+vTpg5s3b6ogXIJvadcomXPDRNQrV66ozjEGn3zyieou8/TTT5s9j/Rxnz59er5/LgN2IiIiIqJ8Gj16tLpZIhNKTYWEhKAgMGAnIiIiIs3TF32Cvchw0ikRERERkYYxw05EREREmqfXQA17UWGGnYiIiIhIw5hhJyIiIiKrXTipOGDATkRERESap2dJDBERERERaREz7ERERESkeXpm2ImIiIiISIuYYSciIiIizcuy3jmnDNiJiIiISPv0Vhyxsw87EREREZGGsSSGiIiIiDRPb8V92JlhJyIiIiLSMGbYiYiIiEjzsljDTkREREREWsQMOxERERFpnt6Ka9gZsBMRERGR5ulZEkNERERERFrEDDsRERERaZ6eGXYiIiIiItIiZtiJiIiISPOyOOmUiIiIiEi79CyJoQdBSEgIdDodjhw5UtQvhYiIiIgKCEtiHiDly5dHeHg4fHx8ivqlaELPR3zRt4c/vD0dcOFyIj5cHoIzFxLy3L9tC28M61sBfqUcERaegiVfX8bewzEW950wshJ6dPHDRyuC8eOGiEJ8Fw+enl190ad7GXh72uPi5SR8+HkIzlxMzHP/ts29MbRPuexxiUjB0lVXsPdwrMV9Xx4RgB6dffHRF5ex5jeOS3716FwKvbv7wdvDHhevJOGjL0Jx9g5j8nAzLwx+xl+NydWIFCz79ir2HckZk0mjAtC1rfnvof1HYxH0zvl8vyYCHm/nhV5dS8LLww7BoalY8m04zoWk5HloWjV2w3NPlIavjz2uXU/DF2tu4MCJnN95To46DO7li+YN3eBWwhbXI9Pxy7Zb2PhXNA93Pni3boLKE4fBo1EdOPmXxoGnXsT19Vvv/D0PN0Wtea/BtVY1pISG48KcTxD25TqzfSq+0B+VJwyDo18pxB07g5PjZyJ2/3GOiQV6Ky6JsSnqF0AFIy0tDba2tvDz84OdHc/D2rcsiRcHBeCLH8Iw4tVjuBiShLlv1ISnu+VjU7u6K6aOD8SGrTcwfNIx7Nx/C7NerY5K5Z1v27d1U2/UquaGm1Fp/N/3LrVv4Y0XBlbAyh/DMHLyCRWwv/d6jbzHJdAVb46rit+23cSIycexc380Zk4KRIClcXnIC7WqueLmLY7L3WjX3AujBpTHV2uuYdSUU7h0ORnvvFYtzzGpVa0EXh9TGZu2R2JU0Cn8cyAGMyZWQUA5J7P9JIB/ZtQR4232okt39bqsXZsm7hje2xff/nIT42ZeQnBYCt4aXxEebrYW969RxRmvjiiHzTtjMPatS9hzJB6vv1QeFf0djfsM7+2HRnVcMf+zq3hh6kX8vCUKo/r5oWl91/v4zoov2xIuiDt2FifGzsjX/s4B5fDQ+iWI2r4XO5s8geBFK1F3ySz4dG5t3KfMM4+i5twgnJ+1GDubPon4Y2fQbMNyOJTyLsR3QsURA3aNateuHUaPHq1uHh4eKmv+5ptvGs8uAwICMHPmTAwcOBDu7u4YOXKkxZKYkydPolu3bmofNzc3tGnTBhcvXjQ+/tlnn6FmzZpwcnJCjRo18PHHH+NB8Ez3Mtiw5QY2/XkTl8OSsWDpJaSkZuGxDqUt7v/UY2Ww70gMVq+/hitXk/H5d6E4H5yIJx/1M9vPx9sB44YFYNYH55GZmXWf3s2D45luZdRJkQR7l68mY8GyYKSkZeHR9qUs7v/UY37Z4/JLOK5cTcGK1WE4fykJTz7ia7afj5c9xg4NwOwPLyIzw3ozMPfiqcd98du2SPz+V5Q6xguXX0ZqWhYeaWf5Sl2vR31Vtvz7X6/jyrUUfPHDNVwITsITXc0/W+npWYiOzTDeEhIz79M7ejD07FwSv/8dgy27YhEanobFX4ercencytPi/j06euPgyQSs/SMKYRFp+Prnm7h4JRndOngZ96lZxRnbdsXg+Lkk3IhKV88vJwKBlW4/Aabb3fx9B85NW4jrP2/J1+GpOLIvkoPDcPrVd5Fw5hIuf7wKEWt+R6Vxg437VBo/BKHLv0fYyrVIOH0Rx1+chsykFJQf/BSHwIKsLH2h3IoDBuwatnLlSpUt37dvHz744AMsWLBABdgG8+bNQ/369XH48GEVzOd29epVPPzww3B0dMS2bdtw8OBBDB06FBkZGerxVatWYerUqZg9ezZOnz6Nt99+Wz2P/NzizM5Oh+qVXXHwWE45i5znHDweg1rV3Sx+T+1AN7P9hQSKtQJz9tfpgCljquK7n68hJCy5EN/Bg8nOVofAyiVw8Hic2bgcOh6rjr8ltQJdzfYX+4/GoHY1V7NxCRpTRZ1scVzuYUwqlcChE7nG5EScyqRbHJNq5vurMTkm+5tnaevXcsMPn9bHivl1MG5oBbi7Ws4Mk6VxAapWdMKR04lm4yL3a1RxsXjIalR2wZFT5mVMh04mqu0Gpy8mo2kDN5T0zL56Ure6C/x9HXD4ZN7lT3TvPJs3QOS23Wbbbm7eCa/mDdTXOnt7eDSqjcitu3J20OsRuW0XPJs35KEnM6yd0HhN+vvvv6+y5tWrV8fx48fV/REjRqjHO3TogIkTJxr3lwy7qcWLF6vs/HfffQd7e3u1LTAw0Pj4tGnTMH/+fPTq1Uvdr1SpEk6dOoUlS5Zg0KBBKK483Oxga6vDrdh0s+3RMemoUNZyJknqqW/F5No/Nl1tN+jX0x+ZWXrWRt/ruLhnj4uMw23j4p/3uMg45B4XL0+HnHF5wh+ZmcCajdfv9aVZLeOY3HaMM1De37zExcBLjUn2Sb9BTK7PimTgpXwp4kYayvg6Ylifsnh7ciDGTj2NYpLMKlLurtnjEhOX6zjHZaCcX06Jiympc4+Jv31/T4+cP/OffhuBMQPKYOXcQGRk6NUV20VfhePk+aRCeifWzdHXB6nXI822yX17DzfYODnC3ssDNnZ2SL0RlWufKJSoXvk+v9riwZq7xDBg17DmzZurYN2gRYsWKsDOlOgEQJMmTe74/VIaIyUwhmDdVGJioiqNGTZsmPEEQEj2XYL8vKSmpqqbqazMNNjY5gRQDyLJDD/9WBlVD0/aEVjJBU895qvq4Uk7tu/OmcQYHJqM4CtJ+OqDeirrfvhkfJG+NmvWvYM3qld2xluLrqiSmDqBLhjV3w9RMRk4apLNJ9IqvRVPOmXAXoyVKGH5krWBs3PedYkJCdmdA5YtW4ZmzZqZPSaTV/MyZ84czJhhPuGmYs2hCKg1HFoRG5+BzEy96niROzOYO4tuINtNM4Rqf4+c/evVdIOnhz2+/7Sx8XHJgL0wMABPP14GfV88XCjv5UESG5c9LjIOdzMuMg5m+3vYIzome2Jp3Zru8HS3x+qPG+Yalwp4+jE/9BvNFqf5GpPbjrHdbVdCDGS7PG5KPht5jaEIv5GGmLh0+Ps5MmDPh7iE7HHJPfFX7kfnyrobxyU2A55ut+8f8/+rIQ72Ogx8sjRmfxyKA8ezf/+HXE1FpfJO6NWlJAP2QiDZdMmym5L76bHxyEpJRVpkNLIyMuBYumSufUoiNcI8M0/EGnYN27t3r9n9PXv2oFq1ancMqE3Vq1cPf//9N9LTb/9D6uvrC39/f1y6dAlVq1Y1u0lpTF6CgoIQGxtrdqtQfSC0RC71nr2UgEZ1c64UyIWKxnU9cOqs5ezeyXPxZvuLJvU9cepc9v5//BWJYROPYvgrOTfpEiN105NmnS7kd/RgyMjU49ylRDSq4242Lo3qeKjjb8mpczKOOfuLxvU8cPJ8dsCxeUckhk06juGv5tykS8zq9eF4dfaZQn5HD8iYBMuYmM/VaFjbHafOW864ynZ53FTjurJ/3i1TfbztVZnHnYJ6Mh0X4MLlFNSvWcJsXOT+mYuWy1fOXEpCA5P9RUPZ/1KS8UTW3k6nauFNZWVlPzcVvJg9R1CyQ3OzbT4dWyJ6T3YiQZ+ejthDJ+HToUXODjodSrZvgZg9TAJZos/KKpRbccCAXcOuXLmCCRMm4OzZs/j222+xaNEijBs3Lt/fLx1m4uLi0LdvXxw4cADnz5/HV199pZ5PSKZcMuYffvghzp07p2rkV6xYoSa35kUmsErHGdObFsthfvglHN06+aJr21Kqbv3lEZXh5GiLjX/eVI8HjamKEf0rGPdf81s4mjbwRO/uZVDB3wmDe5dD9colsG5jhDHjpS7tm9ykS8ytmDSEXsu7LzLlGpdfw9GtY2nVo7tCWSe8PDwATo422LT9/+PyUmUM71feZFwi0LS+B57p5qdqqgc9UxbVq5TAuk3XjeMSEppsdpMuMRIYhoZzXPJjzYbreKx9KXR+uKT6f3/c0IrZY/JXdoZv8gsBGNa3rHH/tRuv46H67nj6cV81JgOf8kdgZRf8/PsN9bh878j+5VCzagn4+jigYW03vDWxKq5dT8WBo+aTVSlvP22OQtc2nujQwgPl/Bzw4rNl4ORggy3/ZE+OnzDUH4OezOnMs37rLTSq7YonO3ur/ft3L4WqAc74dVt2eVJyShaOn03E0KdLo26gi+rV3rGlh3r+3YdZppTfto7u9Wuom3CpVE597VS+jLpffdYE1F/xrnH/y0u/g0ul8qgxZ5KqSa84qr9q4xj8wRfGfYIXrkD5Yb1RdkBPuNaojDqLp8OuhDNCV67lx4PMsCRGw6RlY3JyMpo2baqy6hKsS/vG/CpZsqTqDjNp0iS0bdtWPUeDBg3QqlUr9fjw4cPh4uKCuXPnqn2kxKZu3boYP348irs/d0WpUokhfcurUpcLIYl4dfZp4+Q6CSRMJ6+cPJuAmR+cVwsnDe9fAVfDU/DGe2dVYE4FOC67b8HD3V6dEKmFk0KSMPntM8ZJjKV9HM0mJZ48l4BZH17E0L7lVCAv4/Lm3HMqMKeCsX1PtJp8Ovhpf1WeJL3xZYGjmDzGRDLsb38UjCG9y2Jon7K4GpGKafMvIiQs+wRJWqRVruCsTgBcS9giKjodB4/FYcUPV5HOlpv59veBONVz/bknSsHL3Q6XQlMx9YMriInPnsNUytvebFzOXEzG3M/CMKBnaVX6cu1GGmYvDsXlazlzjt5dGoZBvXzxyvCyamykjv2rn25w4aR88mhcBy22fmW8X2veFPVv6JdrcWxYEBzLlILz/4N3kRwShv09nket+UEIGDMQKWEROP78G4jcvNO4T/gPG1XP9cBpY7MXTjp6Gvu6DUdaromolK24tGAsDDq9NVfwa7wPuwTXCxcuhNa1e9q8bRUVPZ0NL55pUX7L2ej+cnK13CqRis4LXzzJw69Bj6dnX6EvKr0nmnfDKyjfzw+A1vGvOhERERGRhrEkhoiIiIg0T2/FJTEM2DVq+/btRf0SiIiIiEgDGLATERERkebprTjDzhp2IiIiIiINY4adiIiIiDQvS188FjkqDAzYiYiIiEjz9CyJISIiIiIiLWKGnYiIiIg0T88MOxERERERaREz7ERERESkeXq99bZ1ZMBORERERJqXlWW9XWLYh52IiIiISMOYYSciIiIizdNz0ikREREREWkRM+xEREREpHl6K17plDXsREREREQaxgw7EREREWme3opr2BmwExEREZHm6a04YGdJDBERERGRhjHDTkRERESal8VJp0REREREpEXMsBMRERGR5umtuIadATsRERERaZ4+i33YiYiIiIhIg5hhJyIiIiLN01txSQzbOhIRERERaRgz7ERERESkeXq2dSQiIiIiIi1ihp2IiIiINC/LimvYGbATERERkebp2daRiIiIiIi0iBl2IiIiItI8vRWXxLCtIxERERGRhjHDTkRERESap7fito4M2ImIiIhI8/QsiSEiIiIiIi1ihp2IiIiINE/Pto5ERERERKRFOr1eb709cohMpKamYs6cOQgKCoKjoyOPjQZwTLSJ46JNHBft4ZhQQWHATvR/cXFx8PDwQGxsLNzd3XlcNIBjok0cF23iuGgPx4QKCvuwExERERFpGAN2IiIiIiINY8BORERERKRhDNiJ/k8mmk6bNo0TTjWEY6JNHBdt4rhoD8eECgonnRIRERERaRgz7EREREREGsaAnYiIiIhIwxiwExERERFpGAN2IiIiIiINY8BORERERKRhDNjJKmRmZqp/9Xp9Ub8U4jgQERHdFQbs9MDbvn07Zs6ciejoaOh0uqJ+OVYvKyvLOA6GEykq+jEh7bGUYGDSgcg6MWCnB9qaNWvQs2dPpKWlITg42PgHj3/0ii4wtLHJ/rUzd+5cvPrqq0hKSiqiV0O5x+T3339Xn5lVq1YhNTWVB0gjJ7a3bt1CaGio+tqwjb/Dim5c7uUxov+KCyfRA2vv3r145JFH8O6772LkyJHG7SkpKXBycirS12btJFD/5ptvMHHiRDzzzDMoV65cUb8kqzd58mR89913CAgIwLlz51C1alVMnz4dHTt2tPpjc79JMG4IzN966y1s2rQJFy9exEMPPYSnn34a/fr144rMRXxy+9lnn+HQoUMqGdSwYUO89NJLRfGSyIoww04PHEPmadeuXWjatKkK1mNiYvDzzz+r4LBZs2ZYvXq12b50/0igvnLlSvzyyy94+eWXVbAuf/Ti4uKQnp7OoSgCy5YtU2Min5G//voL8+fPxz///MMSsiJiGqwvXrxYfU4kOLxy5Qref/9949VCur8MwbokHN544w14enrC1dVVjdOQIUM4HFSoGLDTAycjI0P96+vri3379mHBggXo06ePyojY2tqibdu2KkN1+fJlBiT3Qe7LxJcuXUL79u1VVur48eP44IMPUL9+fXUi9dFHH7EUowicP38eAwcORIMGDfDtt9/ixRdfVIFihw4dkJycrK5KUeEzJBDkMxMWFoYNGzZgyZIlKtEgGXb57IwZMwY1atRQ8z+YcLj/duzYgXXr1uGnn37C22+/jTZt2iAxMRGtWrWyOJZEBYUBOz1Q/v77byxatAhRUVF4+OGHMWrUKBV4yGX+oKAgFYxIdqRRo0YqEKH7l5U6efKk+rdkyZL4/vvvVQlG7969VSZXroJ06tQJ7733npocTPeHBIYSWJw4cQIeHh44ePCgGot33nkHL7zwgnpcMrpSKkOFPxaGzLp8Ztzc3NTvqMcffxy//vqr+nfevHkYPny4mvchv8uuXr3KYbnP5JhLZr158+ZYu3atyqzLFSkZl4SEBFW+JNjggAqaXYE/I1ERkclyw4YNU5eP5ZdqvXr1VAbklVdeUUGigQTwkjH08fHhWN0HEhBu27YNXbp0UVc1JBCUE6rNmzerbKFsl3ppyfLKvAP5o0eFX4MrDF8/99xz6gRq6tSpWLFihcq2CwkMJaMotdNUuJ8Rw1hIACjBugTn8ntq8ODBKtMu959//nm1j5TGfP755yhVqhTnf9zHz4uQvyWSAJLSPhkP03HZvXu3KvWrXbs2ypcvX5gvjayRnugB8M8//+i9vLz0S5cuNduekJBg/Hr79u36kSNH6r29vfWHDx8ugldpPTIzM2/b1q5dO/2jjz6qT01NVfeTkpLUv1lZWWrbI488ou/SpYu6T4U7Jn///bf+119/1cfGxqr7ly9f1vft21dfo0YN/YYNG9S28+fPq/Fq0qSJPj09nUNSSEz/f5ffSzVr1tRv2rRJ3f/ss8/0Pj4++qefftq4r3xuHn/8cX3nzp31GRkZHJf78Hn55ptv9Dt27NCnpaWpz0XJkiX1Op1Ov2jRIuM+ycnJ6nfYgAED+DuMCgVLYuiBIBPlZILpiBEjEB8fj99++w39+/dH586d1eVkmcwomcLw8HC1r9TqUuExZKWOHTtmrEmfNm2aGhtpGSgZRWdnZ3XJXzKF0s0nIiJCjZVcSmZ7tMIbE8mkP/HEExg0aJCaOyDlLjLxd8KECepzIfM9KlSogKeeekpNBJbJ23Z2duyZX0gMpRPyOZBWp9KVp2vXrmqb/CsZdrka1bdvX/X77bHHHlNXqiTrLnNy+Fkp/Amm0s3q8OHD6vMgVwN/+OEHdeyllEy+3rhxI7p3766u7Mo4ypiyhp0KGts60gPT5ULqbseOHav+uBn+CJYpU0b9ApVAXTg6OsLd3b2IX611kEvGUmoh7c7kxKlHjx6qHEYmz61fv1611pQ/cHIJWQL7Dz/8UAWGMmlY/qWCbRFoqFWXsjGpuQ0MDFTlYnIZ/7XXXsPQoUPVCZTMNZC2jhK0y0Q6CUw4JoXrxo0bKiiURINMyP7xxx+Nj8mJrIyRTD6V32cVK1ZUHUr4WSl8n376qUo0SF26lLk4ODgYP09ywjRlyhRV3iefFX9/fzWvwN7eXp3cyueGqCAxYKdiSzK3EoCLkJAQzJkzB3/88YfKUA0YMEB1g5FAUAIRmRwkv1Tp/vSOFnLM5dhLRlDGqmzZsiq7K3/4JFCUoCN3X3z+oSu8Glw5zhL8SSceqbs1kABerjpJ0C7dSGTyqSmOSeF/VoRkcD/++GN89dVXKgEhv8PuhONSsKQrT+71IGQCtgTp8pkxHG/Tz5Rk3GXOjQTpMidKxpQnt1RYmMaiYkkyHpLNuHDhgiqFkT9ukoGSTFXp0qWN+8nlfvkFWqJEiSJ9vdbAEIDIyZNMyurVq5caJylFksvG0rFHroBI+0aZ+CvZW8kmmi5ixaxUwTIEFjNmzMCWLVvUxF5pCShBhvSPFsuXL1cdLiTrLtslSHFxceGYFBLTgE86Isl9mcgobU6l/ELuz549WwWBUgYj5HeYfI/cDME+PysFR05aZSwkyWAgSYYDBw6gSZMm6r4cb8PkYFk3Qq5CVa9e3eyKrTzOq4NUWFjDTsWOLO7y5JNPokqVKmrVP1lERII/6VFsCNZ37typSjHkkuYXX3xh1iWGCo+cIMnJkwQcYuHChapbj1zSl/aNEqxLxur69etqjKhwmNbPSsZWxqFnz55o3Lgxzp49q1b/leXuDWSNglq1amHPnj1qbgEV3rgYgvVZs2apGvWWLVuqFrTy+ahWrZq60iG9vWUxHsMCbxIEGr6P7QILnpysGlqXGtrKytVbmeshv7vkc2F67OVvjVzRlYSRKY4NFarCmctKVDiioqJUt5GFCxeq+9evX9eXKVNG/9JLLxn3uXHjhv6VV17RP/bYY/pjx45xKApR7o4ue/fu1c+YMUONiXRMkM487733nn7MmDH6mzdvqn0uXbqk/+STT9h55D74/fff9RMmTNCvXr3auE0+G9L5Zfr06fpbt25Z7IzBTj2FS469dBpZsmSJ/quvvtJ36tRJX758edWNRBw/flw/atQo1dFq8+bNhfxqrJtppx0Zi1KlSunPnTun7v/555/65s2b6/v37686K4mrV6/qe/TooW/dujW79NB9xYCdipWIiAh95cqV9SdPntSHhYXpy5Urpx8xYoTx8TVr1qggRILD6OjoIn2t1tT2TE6S5GTKsE1anzVr1kzftWtXfc+ePfUVK1bUf/DBB7c9B9sFFhwJKuQEyWDnzp36evXqqcBw7dq1Zvsagva33npLHxkZmee4UsGTJEP9+vX1K1euNNv+3HPP6cuWLau/ePGiur9//379u+++y6CwEJmemEpwLn9T5HNRt25d/YULF9R2+ezI7zFprxkYGKivVauWvmHDhqrFo+Dnhe4XlsRQsWBoXSaX66UGV1ZkbN26tZrQ+MknnxgXE5G2gHIJUyYAyWp0dH8u7UsLTSm3ePbZZ1W9urQ+k0v8Mj5S4yljM378eOzbt8/seVjvWTCkzKVy5cqqvMJAVmKU8ZBFeKTkxbQERtoHdujQQU1ulI49pnIvFEP/Te72ftJiVjqLeHt7GycDG0qX5HfWggUL1H2pnZaadqmdlgmPVHgTf2XBMFnETcZGWjRKOUy3bt1U6YuUX8qcG5kzJW01p0+fjv3796s5Boa5BUT3xX07NSC6R1u2bNHPnz9ff+3aNXV/8ODBatGKZ555xmy/yZMn6+vUqaMPDQ3lsS4kuUsl3njjDZXBXbdunX7jxo3q0r6Hh4c+ODjYuL+M25tvvqlv06YNs4X3YWyk3Ojbb781Zv/mzp2rb9q0qSqxyF0C89FHH3FMCpFp9tXw+0s0aNDA7PeXYTGxXr166UePHl2YL4lykbLJbt266f/66y/jNrlCK5l2WUjMkGnPjYtW0f3GU0PStDVr1qhuI9JyKyYmRm2TpdMff/xxtaDL+++/j0WLFqn+3tIS7euvv+ZS3YXIdFJVaGgotm7dqibGyYRGeUwy6JK9lS4x0klBtknvaJlAJ60DmS0seIbsqxxr6fIivbwlYyiTsyX79/LLL6ss4dGjR1XfaMPnSMjEbI5J4XeDkQmK8jvKMNFaxkfaOMpVJyETscW1a9e4TsR9JH8zRo8erSaaypVbw7jJFVrJtMvVKfnsnDlz5rbvZZceuu/u+ykCUT4dPXpU7+fnp1++fLnF7MawYcP0rVq1UnW6ffr04QTTQiTL1i9dutRsm9Ta+vv7q8zh+vXr9a6uriq7a1imW7421OMacDJjwQoJCTFm+iSTLnM8ZHn7oUOHqlpbQ+267PPOO++oiXLyWYmLiyvgV0J5ee2119RExh9//NGYrY2JiVHL2stEU5nrMWTIEH3Lli31NWvW5LyOQpS73nzr1q36ChUq6J2cnPQbNmy4bT/JtMv8G5lfQFTU2IedNEvqnmUhC1kh07AYhSFrJdkNqcuVjKJkFuUxwyJKVLBu3ryp5gvIEummJCsoWSmZQyCrlEpmXXqtC+lRLH2/a9asqWqrDdj2rODqb6WOtkWLFmrFRcmqL126VF3paNCggcqcS+bdsDiVZAllsSpZ6CUyMpLrEtwncsVJVi2V+mdZ0M1AFqeSz5PMM5CrhDKe8rW025TfZVwUqXCveEg7Rvl7IfM4/vzzT7USs/wek7bAMndA9jNk2uXKlGHNAqIiVdRnDER5kdaN7u7uFmsGDx06ZFYTSvfH4sWL9UFBQcb7Uhct8wkmTpxo3JaQkKBaakpnBXZQKFzSIalEiRLqJl1FTB08eFA/aNAgfe3atdUcAyHjYbjKwbEpfD/99JPqZGX6u+rfjj87JxU80yt7MtdJatNl7o3Mq5HPhrSale5jMq/gwIEDFr+PNetU1FjDTprrdnHs2DH1tWQ9/Pz81CI8suqcodZWMh+S0f3pp59u68BAhdOdR8hVDslMSc36zJkz1TbJSj311FP4/PPPVVZXVjKV7gpydeSXX34xZqqoYMfE8P+9ZGWTkpLU/fDwcCQnJxv3a9SokXFlWVnJVOYQyHjIVQ7TLj9UcOOSm3QSkZtc1TAwjJ0s1LNt27bbvoedkwp+XAxX9uSYr1y5Eu+8845aLEk+G/L76++//8bmzZtx6NAhtT33QkmCNetU1PgbmzRDloXu3r27urwvk68qVqyIdu3a4ffff8fbb7+tgnaZ6Dht2jS1j1zOZIlF4TIEdfKHTlqcTZ48Wa1k+s0336jl7oW0cZSWaBKUXL16VZXPyIQ6tj0rvDGR/++///57VWYhZWHPPfcc+vXrp1o0GtoEGoL2SZMmqcBdxsWAn5vCK7eQ8pfjx48bj7+MjwSBhsm+sp/8LpPP0Pbt2wv4lVBuhnGRYy2T5KVVpqxgOmjQIDX5V0qSnn/+eURERKjfZTJZWyacEmlOUaf4iYSs5ufi4qJKLkxbz8nXslKjTKBzdHRUC1rIJCEpiaH7QyaOVq1a1TjpVFo2SpvG6tWrq1VNc7emM+Al5MIjkxbl8yArZhrIRFMpjZHJjSkpKWrb2LFj1URUjknhMS2bkAmmMhFbVveVMTJMbJRJjbKAmHyGZHG3jh07qt9lLH+5P8LDw/VVqlTRu7m56WfNmmX2mPyNkZVLDatly6Rt/u4iLdLJf4r6pIGsl/zvJ2Uuw4YNU+3MpEWjgbQFlImNkjGUyXIyiVEmoVapUgVly5Yt0tdtbWTComSfTpw4oVqdSZtNmeQoE+oks/vmm28W9Uu0OtKiUdrNSTmSYZEwKX2RcZIFXg4cOICQkBBVxsQyi8InVwElW7tp0ybUqVNHTWo0LM4jZX5ylUOuEMpYVahQQV0hkatQnGB6f8gYSItgmfArDQsaNmxofEw+N/I7TcbOgONCWsOSGCpShg4v8ofMEFQYakENvYmlzEJm78tqmg8//DCD9ftYhysnTUJKXvz9/fHFF1+ofeTESTrC9O7dWwUpsp3uz5gYSPckKQ2TFWUNJBCRuQRSvuTr64vz588bu45Q4YmPj1crLEu/dVnx9/r16/jjjz9UfbTUS8taBHJfusZIVx8p/zOUjLE2+v6oV6+eOu7yWVi4cCGOHDliHLvTp0+rkyhTHBfSGmbYqchIbaehXdajjz6qAhOpVxeGzJT84fv0009VYCgtAun+kOxfp06djMunS2AhV0FkMqm0QTOQ+xKIDBkyhH/gCplcYSpRooRq5WggJ01ysivL2kvm0BBkmH62DC1RqfDIZF/J2MokRpnj8dFHHxnbZ8p8DsngSgY+r7p3un9kPGTOx61bt1QLR0kMBQcHq4mm8rXhbw+R1vC3BRUJyTRJACi9pIWUVMhMfcPKf4ZfmJK9laxIyZIlOVL3ycmTJ9VkrGrVqqmJpbI6owR8EnBIF58PPvjAuK9kpSQY4WqZhUcCCMkASiAoJ0byr5S7yMTFvn37qpIXmTAnY5Cenq6+xxCsy/cyWC9ccoydnZ3VxFKZ1NinTx/UqlULs2bNUgkImdwon5vc1acM1ouGnFhJpysZs9jYWNWNTLrDSLAunx8G66RVzLBTkVi1ahXmzZunaj2lPrp+/frqcr4E7JL1KF++vCrHkOyttD4zrTekgpVXpk8u5Uu5hbQDlJMradcoAYm0D5SgXWrZ+cetcFjK8km9uhz7iRMnwsnJSdVIy2fo2WefVZldaVdHRUs6wUgQKB2uDJ+trl27om7duliwYAGHR0OkJEauUEmpjHSOqVq1alG/JKI7YsBORUZaBcqlYwnOJcMuWSlphyYrZkqwLqvMST0uS2HuT7AuE65kcq/UdEqALqKiorBjxw6VLZTJcnJFRMot9u7di4ceeqgQX5n1Mh0TaW8qgbt8FqTmWchnQ65GybwBuVKVmJioMoPStk4+K7ykX/TjJp8RKbGQWunLly+rMgy50sGx0RYZFwnaZTVmaRcsKzcTaRUDdrpvJEMoWcFKlSoZt0kvYll8R7q+vP766yoTZcAaz/tH+qtLdxGZHHfz5k01TitWrFB9pCUIkYm/cjIlJUrSTUGW62apRcEz/X9eSpCkC490SZLL92vWrEFAQIDZ/hK4nzp1SmXd5UrV9OnTC+FV0d0G2jIBVT4rsqjVunXrjBNM+ZnRHklCyFoF0j9ffv8RaVZR95Uk6xAaGqqvU6eOfuTIkaqPt6mVK1fqvby89P369dPv27fPYn9jKjxLlizRly5dWvUfFj/88INep9Pp//jjD4v7G8aFPaQLlulS9W+88Ybez89Pv2rVKrXmgPTsrlmzpv6ff/4xGwOD5cuX62vXrq2/cuVKAb8qMhUbG5uvA5KYmKg/c+aMcUz5WdG25OTkon4JRP+Kk07pvvS/lR7rQ4cOVZcg5TKxzMo3GDhwIGrXrq3qo6W3t0ymE6yPvj9tAmVCnFwWbtCggZqMJZNIP/74YzUZS8otDCRDaBgXeQ5mCwuGXK0Qhsy6TPKV8iS5+iStTOXqhnTjEU8++aTK3ub+bMjnR8YnrxaQ9N/J761x48bla1+5IlK9enU1ppz4q30yJ4RI6xiwU6H66aefVOAnkxRffvll1dVCaqLlvizqIuSSv9Teyh9DqSOUcgwqHBI8GAJDQwtNKVWSumiZYCoL7kgvaQngZV+ZeCrzDIRpgM4OFwXjjTfeUPM0pGWjgXSrkEC9ffv22Lx5szrRlTGRmmgvLy81RrmXtJeuMXISbFi7gP673F1dZI6NTOyVpevvRE6aDCdUMq5Sz05E9F8xYKdCIwuESOAxe/ZsDB48WG2bMGGCyuBKlvC1115TE+ek7laCxeeff14tyEOFX4cr7RolKJd2gHKFY+PGjSpAlE4WL7zwgtpHJp/KFRHZhwpHhw4dVDvGxYsXq45IomnTpurEVgK/Dz/8ULUFlDGROmjpPiLZ9vfee8+sB7icTEktLmtwC/6zYlg8rEuXLirpIOtCSCvNvL7PcDK7ZMkS9T3SkpOI6L9iwE6FQrLmko2SP3ASoEsf9XPnzqlexXKpWBZKks4WkmGUy/+y8At7rRcuQwBy8OBBnDhxQo2Pn5+fapkp/dSlxaZhDGSs+vXrpxau4kTGwg3YZXKpLLIjQbvhqocE3tKhR1YqNUzElkBQFrKSTPuvv/5qVn4hWXdpT0cF+1mRK0ytWrVSVwqlXaP0wZexMlzhMF1B1jTIl2A9KChITeSWEzAiov+KXWKoUEjW7+GHH1arMkrAJ6Uu0mVEAkHJKI4dO1YFGZLFdXFxYbB+n3z55ZfqqoZ0r/jll19QqlQptV2CQCm7kGBeaqElYJRVGmVVU8nsSmDCpboLlmmAJ3XrEuBJQD569GhVRiYkQyulLlI2I11iZNyklaOMBbsoFS6ZvzFmzBj1eZFAXbonSUcrmVsgiQfpziPjZSiBMQ3Wpa/3559/jqeeeqqQXyURWQtm2KlQSNZP/tjJYkjSxlEmzkktrvSV7tWrl8qqy2qM0oOdmfX7R5avl8ytZNgl8DNo3ry5utQvZUyyGI9ke2XBJEM7OgbrBccwMdR04mjr1q0xc+ZMtVy6zBmQz4eQqyCBgYEqUyufEzmxYrBe+DXrUgYjJ6wyr0aC8tKlS6tMu0zMlnkEsq8s8iZXCeXKh2EsZeymTJnCYJ2IChwz7FSoJAslwbpkDA0ZQckgSmZdOsJwgun9J4G49O2WLLpkAtu0aZNnr2lm1guWaVZ8/fr1KkCXE9fHHntMXWmSvuoS8EmQKCe8nTp1UvvKSZZsk/FhP+/C9fXXX6v5ADLHRj4jctIkWXPpp75r1y71uEzyDQ0NVZOCO3bsaPxd17hxY5WR79OnTyG/SiKyNgzY6b6RbiRSqy7ZWykBkJppun9MA3KplZYyJZnEKOVJkj2k+3f8ZW6HfBbc3NyMQbgE8JLBlS5KMrdDMuoyWfuJJ56w+BxUsAylLXLCJCv5Xrx4UV0hlOy6lJIZPisyr0AmyUtdu5QpmV59kvIl04XhiIgKCgN2ui+kNlrqPo8cOaJWlKtfvz6PfCG5U1bcNOCTsgvpFiOrZ8p8ApkASYXD9LhLdlbKKaS1qRx7mcQoVzqkc5LcqlSpooJ26ZokwbpMfKTCv+Jh+rVc0ZCVlyWr/uyzz6oWtHJSJXXpPj4+Zs/BKx5EdD8wYKf7NglVAhUJUKRunQreihUr0Lt3b1V7e6cJibkz7S+++KIKSt566y0OSyGThamkZEL6pf/4449qjoCQEjFZFEl6dkvJkpSKySRt6f3N+QOFw/QzIpn0Q4cOqe5WDz30kLG1qSQXpN2mLG4lyYbu3bv/ax92IqLCwEmndN8moUqtNIP1wiGTEmfNmoU333xTdRKRQCSvVS8lWDdMsOvatavqeiHlMVS4ZAVfyZxLrbN0SzIE65KhldIYaX8qmV1ppSmknaME66atA6ngGIJ1ubohJUienp5qRWb5LMjaBEJam8p9+VzJXIOYmJjbFlQiIrofGLATPQAk8yf1ztJFRGpw8xu0y+PNmjVjYHgfSNZcWmfKOMlkU0OXEcMKsjLBUbqTyNiZYoa9YJl+JuQESspepB5d+uFLUkGOv7SkNZCrgnL1Q+rTt23bpj47eX2uiIgKCwN2omJOggcnJyeVKXzkkUfyHbSbZhkN/fGpcMdJMrivvPKKanEqHWGkTl1WLpVyCwnm5QqUtHGkgicTRm/cuGFWKiZtZiWzLm1N165dq/qty1wbudoh5UmG9ppCJgEbrnjkVW5GRFRY+FuHqJiT4EGCCMngTp48Wa0i+29Bu2kdu/SObtSoES5fvlxE78A6GMZByl+kZaCcXEkdu6w0K+VMsuqstAn8t5MsunuyMqkshCStMU3JfenqImVhgwYNwty5c9VJlJAJwLK4mLRvNMUTWyIqCpx0SlRM5TWxVCbOSWcRyQ5K5lAu9UuPb8P+uZdQl8D+448/Zu/o+zxuMtH03XffVZMaZeGk9957T5XHyPjJFRMqnOMufdTbtm2rrmacPn1alcFIidKHH36o1ogQMgZSBiMrAUsfdrbSJKKixoCdqJgH6xs3blQ9o8uWLYuaNWuiRo0aqiuPBIMStLdo0QKzZ89WQbtpCzouoV704xcbG6vGRmqpZQKwTH40TEalgmH6/7x8Tnr27Kky6xK4S9AuVzVk4aphw4apRZBkcumCBQvU5F/pHCPfy/73RFTUGLATFTOmwYOUwHz33XcqEyglMXKT9oySsZWgXbK2ksGVuuhPP/3UuLKsBOtSliHt7J566qkifkfF353aaOYnaJdSDFmEp3///qojCRU8WQxJJljLJFNZvE3KyCR7XqFCBbVolRx36dIj9/39/VVLRzl54mq/RKQFDNiJiqmFCxeqTKAE7C1btlSTFqdPn47q1avj/fffV9lCCdolayvlFxKwS4Aovdelzl1aQTJYL9hgXVbAlHIKqXNu167dHb/P0KVH9r106ZJaTVM6yEhXEipYW7ZsUV155IpTuXLl1P/7UgYmJ77SF1+CdFnASlpvSo98WRzJsAKtITtPRFSUGLATFROmmT6puZXJcV26dFGrlP76669q8SP5WhZ5CQ8PxyeffKLqc6VVoGQKDVl56SV95swZVd9OBScoKEidPEk3EVm+Xk6KpCNMkyZN7niVRE66Tp48qbqTSMcSKngycVQmVo8bN06dwAq5oiETrmUcJNOee40IlsEQkZawSwxRMSDdXgzBurQBlBpcuYQvWXRZEVMmy0kt9Lx589CtWze1OM/TTz+tusVIxtB0sSRDGzsqOBL4yUqzkrmVFX2nTp2qvpaMbW6mgeDSpUvVVZFOnToxWC8ghg47hn8lSy7BuHw+pMxFTmiFXF2SVo9ydUQ+Mzdv3jR7Hk40JSItYcBOpHFSgz5jxgz19UsvvYTHH39cBfD16tVD5cqV8eeff6oyGJk0J3x9fdGjRw9V3y7LrBswACk8x44dU2Mj2fTvv/9etWmUIL5Vq1YqaDesVipBpGmHnkmTJqmSDFlRkwqGoTwpJCRE/WsoaZEMu8zhkJ73BtIJRnqvy0JJcmWEiEirGLATaZx0EJFOMFKnvnr1atW7Wzq+GEgGUTLqFy5cUAGhlGVIb++XX36ZK5gWgtw90iUgl6sc0s9bsuty4iTzCV544QU1NjLx17AAjyGYXLZsmVro6vPPP0evXr0K42VaNSkRq1q1qiqBkdVJhZxMSQmZTLaWsjADKSVbtGiRcT0DIiItYsBOpHGGRXWkvEUygoZaW0PgKJ0vpJ2jlMfUqVNHrVr6+uuvG8tguNBLwTIE3dL2T8ZAsrYSdMsxl5Mqmcw4atQotY8s1vPXX3+pgN5AOpRIMC+ZdU76LRiGz4Kh7EtOWOVKh3xm5MRI2jbKPIHnnnsOTZs2VeVLhkm/pvhZISKt4qRTIg2TCaNykzpnKYPZv3+/qneWjhdS+mKwa9cutQhMXFwcxowZo8oA2I6u8LrBSO25TBaVyYpywiR10VKCJEvdS5107dq1cfXqVTUJWCYI79y501iaIW0F09PT0bt37wJ+hdbJdFzkKpNMsJaFp+TzcePGDVUCI61O5bMhJ7uyj5SQSStHIqLiggE7UTHq6S1ZXCmPkUV2TIN2WRCmSpUqxv0YrBfemEh5y+XLl1WWvEOHDqq7S/369VUAKNlzOXmqWLGiChxlwq8E6/I1WwQWPNMJvHJSu3btWnWCK73tZ86cqfraG8rHVq1ahYMHD6p2qNWqVVMnuHfbO5+IqKgwYCfSaGAo9c2SUZeSi1q1amHkyJFqu3QgkaBdWjbKhDmpVZdgcPv27WxFdx9aN8q4yL/SKlDmE3h5eeGrr75C3bp1VVZ93759KtMu2VyZICxlFgzWC5cE51KHLquXykTfAQMGqM/D7t271YmsaS/1w4cPqwnbMi73suAVEVFRYMBOpEFSdys1zlL+ItlCWT5dOolICYaQS/yS0ZVL/mXLllV10pLNpcIjNdAyT0ACdqmJFnL8ZVVZWc5++fLlxkDQFK92FDzTQFu+fuKJJ9C3b181gfSnn37C0KFDVRtHuQpi2Df3OPAkioiKEwbsRBpcLVN6qEtHGGk3J4GFtG6Umme5STtAceLECRXMS091ZnELn0wclS4jUhIjJTBSeiEnSdI+UFoGymRGCRIbN258H14NiWnTpqmSF8muSzcYubLRvXt3zJ07V038lZV+3377bbXImKxwSkRUXPFaIJEG6nANwboE59HR0XB2djYGfnI5v3Pnzvjss8/UYjzS5lFIRxi5/C/BumQPuYR64QoICFDjI3XSQoJ1Oe6yiFVgYKCqVZeFeGRysGFcqWCZdnWRE1rp9iKLHrVv314deylB+uCDD4xdeuSzJFefpDyGiKg4Y8BOVIQkc/7NN9+oryXIkFIYyQRKqYVk2k1JuYXUsyckJNz2PGxHV3Byt/ozcHNzU6vLSpAoQaHhuEtHEhkbCQqlpab0YBdcqKrgGU5sDUH4xIkTVUceucp09uxZVbIk5TAiPj4ew4cPV2PEhamIqLjLmYlDRPeNZF8l8JbgTkorJFsoQYhkz2UBHgk8pPRFaqOlt7eQTK6Pj4/K8tL9ad145swZhIWFqRVJGzRooLqOyMmUzCGQCcE1atRQ8wtkIR4ZLxmrK1eucHgKUUREhFqcSsZhypQpxpNd6ZQkZTHSg126wMg4pKSkqHEyXIXiiS0RFVesYScqQtKjW4I8Q2ZWenkLmVAq7eckgJRFeaQcQ5a6j4qKUl1IGHgULlkNUyb99ujRA+Hh4Th06JDqDCPZWxmT33//He+++67KusuS9tIyUFo3PvrooyrbLo+ZthykgnXs2DG16FTp0qXx4YcfqvIxCcg3bNigTnylz72c+BrWJOAEUyIq7hiwExUhycxKZwvJtku5i7Sjk5uQyY0SgEjLQKmRlsBQAnkJDJktLDzSBUYy6NJtRLLqe/fuRYsWLVQ3Hul9L2UWHh4eZtl4Cc7lZEvGSgJGGS8q/KB90KBBaNKkiQrM5UTJEn5WiOhBwICdSEOX+aWrhfRWNwTtQjK8MglVgkTJ2DJbWLBMAzoJwqUURra99NJLalVSGQ/J4sqiOzLxVzrBSLcef39/Y+Ao/b+///57tb+UZND9IT3V5QRKMuzjxo1T9exERA8iBuxEGhEcHKwyhVLTLj2l/9fe/YNU+UdxHD+TS6QZhIPpIIZUNlrgn0QwCFQKqSnBlkQ0FFIjSdRRrKhEQWgpIgyiIBJEDRoiCBRt0cSl0BTEFoUC/wzxOeHlVuPvV36fx/drkXufe0V08HPPc77nqHqoTZqaBKPRdMKil79Hv2O1wOgOhj4cqf9Z870vX77sy6nUE63JPKqqDw4O+t9I1H6h7aY5OTm+LAn/PrRrbKO2y/b29norDADEDYEdCCy0t7a2+tr0jY0NnzGtyi5Lkf5/yR9+NKmnpqbGDyjujNPUAeCGhga/pnYL9bGrXUbBXNVcVeXpUw+DznXoQ5TugLC5FEAcEdiBwKgFRiF9ZWXFq+wcmvu71KuulqR9+/b90or08uVLu3Llih/2PXr0qN28edPPEWiso9AbHZadD0/chQIQRwR2IHAEw79ndnbWSkpKfMGOKrR1dXW/nBFQ28vY2JiP18zIyPDWF7XMIEzc8QAQVwR2AHs20Gm5zqtXr3whUm5uro9rFPWvayGSvH371t+jSTFqg+HQLwDgXyOwA9gTklslFLoVylU51x2M58+fe796eXm5PX369I/QvoO7HQCA3fDzvxcA7JGwrqVG2lh64sQJnwyjA4sa0zgwMOAtL7omCusK6MlYWAUA2A1U2AHsGVplr0kiCupqj7l9+7bPU3/x4oUvrhoeHrbr169bXl5eoj0GAIDd9vNkFQDE3IcPH3xTrKbCFBYW2rt37+zz588e4tPT0/01Wnf//ft3fx3TRgAAoaAlBkAsKXAn08FRtbQorD979szOnj1r9+/f99GZ37598zGO2jSrlhhV3NVC8/v3AABgNxDYAcTSTs/64uKif1Uo11SYBw8e+PjGnp4eq6+v92uTk5P25MkTr7gnL6liCQ8AIAQEdgCxpS2lGscoqqyfOnXK19i3t7dbY2OjP6+qunrZt7a27NixY7v8EwMA8Cd62AHE1pkzZ3zai1pfmpubfd76169f7datW77ZdH193d68eWPLy8s2PT2daIOhsg4ACAlTYgDEcimSRjJubm56UNcmU/Wt6zWfPn2yO3fu+KHTQ4cO2ZEjR6yvr8+3m7IUCQAQIgI7gFj58uWLHT58OPF4YmLC22IePXpkly5dSjy/trZmaWlpiceEdQBAqOhhBxAbmrFeXV1tra2tfsBU20oLCgrs6tWrNjQ0ZCsrK15ll/379yfep+dUYQcAIEQEdgCR9fvYRR0qvXDhgs9aLyoqsq6uLq+4V1VV2fz8vC0tLXnbjAJ6cp96cisNAAChIbADiKTkw6Hj4+M+O311ddU3lX78+NEuXrxoU1NTlp+fbzMzM36tra3Ne9sJ6ACAKOEeMIDISa6Qa0Tj48eP/QDp3NycV9h7e3t9IowOnQ4ODtrw8LBvMNVjJsAAAKKGQ6cAIkvB/N69e94Cc/LkSevv77empiY7f/68X8vNzfXXLSws+HSY4uJi33bK6EYAQJTQEgMgkjQ7fXZ21u7evethXS0xnZ2d1tHR4bPVb9y44dclOzvbSktLPayrJYYqOwAgSmiJARBJBw8etHPnzllZWZlNTk5aS0uLdXd3e4X9wIEDPilG89cfPnxoWVlZifcptAMAECVU2AFEkjaYVlZWejh//fq1HT9+3Gpra/1aSkqK1dTU+NfMzMzd/lEBAPhPCOwAImtndrpGNmoRkqa/aPb66OioVVRU2MjIiLe//D7+EQCAKOHQKYDIe//+vZ0+fdry8vJsY2PDq+8a6cgyJABAHBDYAcSCAroOnqamptq1a9c8rG9vbxPaAQCRR2AHEEuEdQBAXBDYAQAAgIBx6BQAAAAIGIEdAAAACBiBHQAAAAgYgR0AAAAIGIEdAAAACBiBHQAAAAgYgR0AAAAIGIEdAAAACBiBHQAAAAgYgR0AAAAIGIEdAAAAsHD9AN9H0kGkQUuyAAAAAElFTkSuQmCC", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "plt.figure(figsize=(8,6))\n", + "\n", + "sns.heatmap(\n", + " steam[[\n", + " \"success\", \n", + " \"total_reviews\", \n", + " \"average_playtime\", \n", + " \"review_score\", \n", + " \"price\"\n", + " ]].corr(),\n", + " annot=True,\n", + " fmt=\".2f\",\n", + " cmap=\"coolwarm\"\n", + ")\n", + "\n", + "plt.title(\"Correlation Matrix\", fontsize=14)\n", + "\n", + "# Rotate axis labels for readability\n", + "plt.xticks(rotation=45, ha=\"right\")\n", + "plt.yticks(rotation=0)\n", + "\n", + "plt.tight_layout()\n", + "plt.show()" + ] + }, + { + "cell_type": "markdown", + "id": "6f7018b4", + "metadata": {}, + "source": [ + "* Success vs Reviews 0.9 correlation is mathematically expected since success formula consists in total_reviews * review_score.\n", + "\n" + ] + }, { "cell_type": "markdown", "id": "4d3e4247", @@ -1141,7 +1201,7 @@ "\n", " * Do free-to-play games achieve higher engagement?\n", "\n", - " * Are premium-priced games associated with higher or lower review scores? " + " * Are higher-priced games associated with higher or lower review scores? " ] }, { @@ -1172,26 +1232,1487 @@ "* H2: Games with more reviews tend to achieve higher success.\n", "* H3: Games with higher average playtime tend to be more successful.\n", "* H4: Game success on Steam is highly concentrated among small number of titles.\n", - "* H5: There is an optimal price range that maximizes game success." + "* H5: There is an optimal price range that maximizes game success.\n", + "\n", + "------------------------------------------------------------------------------------------------\n", + "------------------------------------------------------------------------------------------------\n" ] }, { "cell_type": "markdown", - "id": "e92d454d", + "id": "d95a97c7", + "metadata": {}, + "source": [ + "### H1: Price vs Success\n", + "(Lower-priced games achieve higher success than high-priced games)" + ] + }, + { + "cell_type": "code", + "execution_count": 18, + "id": "e831bad8", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "price_category\n", + "Upper Mid 5513.165733\n", + "Free 3065.403125\n", + "Lower Mid 1853.017754\n", + "Cheap 792.838413\n", + "Expensive 159.633333\n", + "Very Cheap 143.650342\n", + "Name: success, dtype: float64" + ] + }, + "execution_count": 18, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "steam[\"price_category\"] = \"Paid\"\n", + "\n", + "steam.loc[steam[\"price\"]== 0, \"price_category\"] = \"Free\"\n", + "\n", + "price_bins = [0, 5, 10, 20, 60, 1000]\n", + "price_labels = [\"Very Cheap\", \"Cheap\", \"Lower Mid\", \"Upper Mid\", \"Expensive\"]\n", + "\n", + "steam.loc[steam[\"price\"] > 0, \"price_category\"] = pd.cut(\n", + " steam.loc[steam[\"price\"] > 0, \"price\"],\n", + " bins=price_bins,\n", + " labels=price_labels\n", + ")\n", + "\n", + "price_analysis = steam.groupby(\"price_category\")[\"success\"].mean().sort_values(ascending= False)\n", + "\n", + "price_analysis" + ] + }, + { + "cell_type": "code", + "execution_count": 19, + "id": "39af54ff", "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "price_category\n", + "Upper Mid 399.0\n", + "Lower Mid 88.0\n", + "Free 58.5\n", + "Cheap 26.0\n", + "Expensive 23.5\n", + "Very Cheap 14.0\n", + "Name: success, dtype: float64" + ] + }, + "execution_count": 19, + "metadata": {}, + "output_type": "execute_result" + } + ], "source": [ - "This hypothesis testing will be done in SQL, to do this I need to create a cleaned dataframe that contains all the columns needed \n", - "(steam.csv original columns + columns created for EDA)." + "steam.groupby(\"price_category\")[\"success\"].median().sort_values(ascending=False)" ] }, { "cell_type": "code", "execution_count": 20, - "id": "566dd8b8", + "id": "db6d5453", "metadata": {}, - "outputs": [], + "outputs": [ + { + "data": { + "text/plain": [ + "price_category\n", + "Very Cheap 13293\n", + "Cheap 6300\n", + "Lower Mid 3999\n", + "Free 2560\n", + "Upper Mid 893\n", + "Expensive 30\n", + "Name: count, dtype: int64" + ] + }, + "execution_count": 20, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "steam[\"price_category\"].value_counts()" + ] + }, + { + "cell_type": "markdown", + "id": "00c622b0", + "metadata": {}, + "source": [ + "The analysis suggests that success on Steam is highest for games in the upper-mid price range, followed by free-to-play titles. Very cheap and very expensive games show the lowest average success. This indicates that there may be an optimal pricing range rather than a simple pattern where lower prices always lead to better performance." + ] + }, + { + "cell_type": "markdown", + "id": "771c6781", + "metadata": {}, + "source": [ + "Advice for companies/publishers:\n", + "\n", + "* Avoid assuming that the lowest possible price will maximize success. Games priced in the upper-mid range appear to balance accessibility and perceived value more effectively, while free-to-play can also be a strong strategy when combined with high engagement.\n", + "\n", + "------------------------------------------------------------------------------------------------------------\n", + "------------------------------------------------------------------------------------------------------------" + ] + }, + { + "cell_type": "markdown", + "id": "e7760fd1", + "metadata": {}, + "source": [ + "### H2: Review vs Success\n", + "(Games with higher average reviews tend to be more successful)" + ] + }, + { + "cell_type": "code", + "execution_count": 21, + "id": "4cbbc8b2", + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "C:\\Users\\rodri\\AppData\\Local\\Temp\\ipykernel_11604\\2914874560.py:7: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.\n", + " review_analysis= steam.groupby(\"review_bucket\")[\"success\"].mean()\n" + ] + }, + { + "data": { + "text/plain": [ + "review_bucket\n", + "0-50 11.193771\n", + "50-200 73.988084\n", + "200+ 336.478836\n", + "1k+ 2472.255288\n", + "10k+ 21033.228571\n", + "Name: success, dtype: float64" + ] + }, + "execution_count": 21, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "steam['review_bucket'] = pd.cut(\n", + " steam[\"total_reviews\"],\n", + " bins=[0, 50, 200, 1000, 10000, 100000],\n", + " labels=[\"0-50\", \"50-200\", \"200+\", \"1k+\", \"10k+\"]\n", + ")\n", + "\n", + "review_analysis= steam.groupby(\"review_bucket\")[\"success\"].mean()\n", + "\n", + "review_analysis" + ] + }, + { + "cell_type": "markdown", + "id": "a292ae07", + "metadata": {}, + "source": [ + "Since the \"success\" metric is calculated as review_score * total reviews,\n", + "this KPI behaves roughly as an estimated positive review impact, this means:\n", + "\n", + "\n", + "The review bucket of 0 to 50 has an average game success of 11.19 (Due to low review count on games with 0 to 50 reviews)\n", + "But the review bucket of 10k+ reviews has an average game success of 21033, this is because games with huge review counts are translated to \"more success\" which is mathematically predicted.\n", + "\n", + " Example:\n", + "\n", + " * If game A has 10,000 reviews:\n", + "\n", + " But only 20% reviews are positive -> The review_score is 0.2.\n", + "\n", + " So our success metric (review_score * total_reviews) will be: success = 0.2 * 10 000 which equals to 2 000.\n", + "\n", + " Comparing this 10k+ bin to a game that only has 50 reviews but they're all positive, how can we affirm that the 10k+ reviewed game has higher success than the 50 reviewed game?\n", + "\n", + " * Both games can be classified as unsuccessful since one lacks popularity and the other lacks positive ratings.\n", + "\n", + "For this reason, review count alone is insufficient to determine true game success. Additional metrics such as playtime are required to better capture player engagement and retention.\n", + "\n", + "------------------------------------------------------------------------------------------------------------\n", + "------------------------------------------------------------------------------------------------------------" + ] + }, + { + "cell_type": "markdown", + "id": "1b86e2d7", + "metadata": {}, + "source": [ + "### H3: Playtime vs Success \n", + "(Games with higher average playtime tend to be more successful)" + ] + }, + { + "cell_type": "code", + "execution_count": 22, + "id": "b8430f46", + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "C:\\Users\\rodri\\AppData\\Local\\Temp\\ipykernel_11604\\2336436708.py:7: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.\n", + " playtime_analysis = steam.groupby(\"playtime_bucket\")[\"success\"].mean().sort_values(ascending=False)\n" + ] + }, + { + "data": { + "text/plain": [ + "playtime_bucket\n", + "83+hrs 77294.811321\n", + "16-83hrs 15688.212810\n", + "5-16hrs 3937.003610\n", + "1-5hrs 1347.782686\n", + "1-60min 546.889377\n", + "0 61.721980\n", + "Name: success, dtype: float64" + ] + }, + "execution_count": 22, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "steam[\"playtime_bucket\"] = pd.cut(\n", + " steam[\"average_playtime\"],\n", + " bins= [-1, 0, 60, 300, 1000, 5000, steam[\"average_playtime\"].max()],\n", + " labels= [\"0\", \"1-60min\", \"1-5hrs\", \"5-16hrs\", \"16-83hrs\", \"83+hrs\"]\n", + ")\n", + "\n", + "playtime_analysis = steam.groupby(\"playtime_bucket\")[\"success\"].mean().sort_values(ascending=False)\n", + "\n", + "playtime_analysis" + ] + }, + { + "cell_type": "markdown", + "id": "9a91d4b7", + "metadata": {}, + "source": [ + "As predicted, the more the game is played, the more and positive reviews it gets" + ] + }, + { + "cell_type": "code", + "execution_count": 23, + "id": "280f605c", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "playtime_bucket\n", + "0 20905\n", + "1-60min 1365\n", + "1-5hrs 2830\n", + "5-16hrs 1385\n", + "16-83hrs 484\n", + "83+hrs 106\n", + "Name: count, dtype: int64" + ] + }, + "execution_count": 23, + "metadata": {}, + "output_type": "execute_result" + } + ], "source": [ - "steam.to_csv(\"../data/processed/steam_cleaned.csv\", index= False)" + "# Check counts per bucket\n", + "\n", + "steam[\"playtime_bucket\"].value_counts().sort_index()" + ] + }, + { + "cell_type": "markdown", + "id": "07e111ac", + "metadata": {}, + "source": [ + "20,905 out of 27,075 games = ~77%, have 0 hours of playtime\n", + "\n", + "This means the majority of Steam games fail to achieve meaningful player engagement, so Steam is a highly skewed market.\n", + "\n", + "Out of 27k games, only ~600 games reach a high playtime, higher playtime buckets contain significantly fewer games, indicating that strong engagement is rare but potentially associated with higher success." + ] + }, + { + "cell_type": "code", + "execution_count": 25, + "id": "3d94293f", + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "C:\\Users\\rodri\\AppData\\Local\\Temp\\ipykernel_11604\\2315270432.py:3: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.\n", + " steam.groupby(\"playtime_bucket\")[\"review_score\"].mean().sort_values(ascending=False)\n" + ] + }, + { + "data": { + "text/plain": [ + "playtime_bucket\n", + "16-83hrs 0.799012\n", + "83+hrs 0.777106\n", + "5-16hrs 0.770286\n", + "1-5hrs 0.721621\n", + "1-60min 0.715650\n", + "0 0.707462\n", + "Name: review_score, dtype: float64" + ] + }, + "execution_count": 25, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "# Check average review score with playtime\n", + "\n", + "steam.groupby(\"playtime_bucket\")[\"review_score\"].mean().sort_values(ascending=False)" + ] + }, + { + "cell_type": "markdown", + "id": "ed3b9442", + "metadata": {}, + "source": [ + "High-playtime games are not only more successful, but also tend to receive better player ratings, the increase is not massive but its significant (+9%)." + ] + }, + { + "cell_type": "code", + "execution_count": 26, + "id": "35dba839", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "playtime_analysis.plot(kind=\"bar\", title=\"Average Success by Playtime Bucket\")\n", + "plt.ylabel(\"Average Success\")\n", + "plt.show()" + ] + }, + { + "cell_type": "markdown", + "id": "5ae79587", + "metadata": {}, + "source": [ + "The analysis shows a strong positive relationship between playtime and success. Games with higher average playtime achieve significantly higher success scores, indicating that player engagement and retention are key drivers of performance on Steam. However, high-playtime games represent a relatively small portion of the dataset, suggesting that strong engagement is rare but highly impactful. This highlights the importance of designing games that encourage sustained player interaction." + ] + }, + { + "cell_type": "markdown", + "id": "c2ac13f9", + "metadata": {}, + "source": [ + "The results from the third hypothesis suggest that game success on Steam may be driven by a relatively small number of highly engaging titles. However, this conclusion is based on the relationship between playtime and success. To confirm whether success is truly concentrated among a limited number of games, a more direct analysis is required, which is addressed in the fourth hypothesis.\n", + "\n", + "----------------------------------------------------------------------------------------\n", + "----------------------------------------------------------------------------------------" + ] + }, + { + "cell_type": "markdown", + "id": "f3caef70", + "metadata": {}, + "source": [ + "### H4: Success Concentration\n", + "\n", + "(Game success on Steam is highly concentrated among small number of titles.)" + ] + }, + { + "cell_type": "code", + "execution_count": 31, + "id": "4971aa8e", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\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", + " \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", + " \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", + "
namesuccesscumulative_percentage
25Counter-Strike: Global Offensive2644404.00.097615
22Dota 2863507.00.129490
19Team Fortress 2515879.00.148533
12836PLAYERUNKNOWN'S BATTLEGROUNDS496184.00.166850
121Garry's Mod363721.00.180276
2478Grand Theft Auto V329061.00.192423
1467PAYDAY 2308657.00.203816
3362Unturned292574.00.214616
1120Terraria255600.00.224052
21Left 4 Dead 2251789.00.233346
5235Tom Clancy's Rainbow Six® Siege251178.00.242618
2031Rocket League®242561.00.251572
1025The Elder Scrolls V: Skyrim237303.00.260332
1634Warframe226541.00.268694
2016Rust220370.00.276829
2964The Witcher® 3: Wild Hunt202930.00.284320
1596Euro Truck Simulator 2176769.00.290845
8129Paladins®169580.00.297105
4712ARK: Survival Evolved145035.00.302459
903Borderlands 2144595.00.307796
\n", + "
" + ], + "text/plain": [ + " name success cumulative_percentage\n", + "25 Counter-Strike: Global Offensive 2644404.0 0.097615\n", + "22 Dota 2 863507.0 0.129490\n", + "19 Team Fortress 2 515879.0 0.148533\n", + "12836 PLAYERUNKNOWN'S BATTLEGROUNDS 496184.0 0.166850\n", + "121 Garry's Mod 363721.0 0.180276\n", + "2478 Grand Theft Auto V 329061.0 0.192423\n", + "1467 PAYDAY 2 308657.0 0.203816\n", + "3362 Unturned 292574.0 0.214616\n", + "1120 Terraria 255600.0 0.224052\n", + "21 Left 4 Dead 2 251789.0 0.233346\n", + "5235 Tom Clancy's Rainbow Six® Siege 251178.0 0.242618\n", + "2031 Rocket League® 242561.0 0.251572\n", + "1025 The Elder Scrolls V: Skyrim 237303.0 0.260332\n", + "1634 Warframe 226541.0 0.268694\n", + "2016 Rust 220370.0 0.276829\n", + "2964 The Witcher® 3: Wild Hunt 202930.0 0.284320\n", + "1596 Euro Truck Simulator 2 176769.0 0.290845\n", + "8129 Paladins® 169580.0 0.297105\n", + "4712 ARK: Survival Evolved 145035.0 0.302459\n", + "903 Borderlands 2 144595.0 0.307796" + ] + }, + "execution_count": 31, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "steam_sorted = steam.sort_values(\"success\", ascending = False)\n", + "\n", + "steam_sorted[\"cumulative_success\"] = steam_sorted[\"success\"].cumsum()\n", + "\n", + "total_success = steam[\"success\"].sum()\n", + "\n", + "steam_sorted[\"cumulative_percentage\"] = steam_sorted[\"cumulative_success\"] / total_success\n", + "\n", + "steam_sorted[[\"name\", \"success\", \"cumulative_percentage\"]].head(20)\n" + ] + }, + { + "cell_type": "markdown", + "id": "c5de40c2", + "metadata": {}, + "source": [ + "This Top 20 games list are ~0.07% of all Steam games (20 games out of 27k) yet they generate 30.7% of total success." + ] + }, + { + "cell_type": "code", + "execution_count": 34, + "id": "b4e5a093", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "(736, 0.027183748845798706)" + ] + }, + "execution_count": 34, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "top_games = steam_sorted[steam_sorted[\"cumulative_percentage\"] <= 0.8]\n", + "len(top_games), len(top_games) / len(steam)" + ] + }, + { + "cell_type": "markdown", + "id": "b574d16d", + "metadata": {}, + "source": [ + "The analysis shows that game success on Steam is highly concentrated among a very small number of titles. Approximately 2.7% of games account for 80% of total success, indicating a strong winner-takes-most dynamic. Additionally, the top 20 games alone generate over 30% of total success, despite representing less than 0.1% of all titles. This highlights the extreme inequality in game performance on the platform, where only a small fraction of games achieve significant visibility and impact.\n", + "\n", + "This concentration suggests that achieving high success on Steam requires reaching the top-performing segment, as the majority of games receive minimal attention and engagement." + ] + }, + { + "cell_type": "markdown", + "id": "b3915b2d", + "metadata": {}, + "source": [ + "To complete our business recommendations we analyze the optimal price range with Hypothesis 5\n", + "\n", + "--------------------------------------------------------------------------------------------------------\n", + "--------------------------------------------------------------------------------------------------------" + ] + }, + { + "cell_type": "markdown", + "id": "35a7fe03", + "metadata": {}, + "source": [ + "### H5: Optimal Price range in Steam games\n", + "\n", + "(There is an optimal price range that maximizes game success)" + ] + }, + { + "cell_type": "code", + "execution_count": 41, + "id": "0b8f37c3", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\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", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
successtotal_reviewsreview_scoreaverage_playtime
price_category
Cheap792.838413894.7422220.74500388.038730
Expensive159.633333185.9333330.75477774.033333
Free3065.4031253695.5667970.719060463.128516
Lower Mid1853.0177542161.7194300.757227193.546387
Upper Mid5513.1657337487.3124300.741594851.816349
Very Cheap143.650342178.2660800.68435658.589558
\n", + "
" + ], + "text/plain": [ + " success total_reviews review_score average_playtime\n", + "price_category \n", + "Cheap 792.838413 894.742222 0.745003 88.038730\n", + "Expensive 159.633333 185.933333 0.754777 74.033333\n", + "Free 3065.403125 3695.566797 0.719060 463.128516\n", + "Lower Mid 1853.017754 2161.719430 0.757227 193.546387\n", + "Upper Mid 5513.165733 7487.312430 0.741594 851.816349\n", + "Very Cheap 143.650342 178.266080 0.684356 58.589558" + ] + }, + "execution_count": 41, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "# For each price category, what is the average performance across all key metrics?\n", + "\n", + "steam.groupby(\"price_category\")[[\n", + " \"success\", \n", + " \"total_reviews\", \n", + " \"review_score\", \n", + " \"average_playtime\"\n", + "]].mean()" + ] + }, + { + "cell_type": "markdown", + "id": "cf4b7877", + "metadata": {}, + "source": [ + "This Hypothesis shows us the full performance profile on each price category." + ] + }, + { + "cell_type": "markdown", + "id": "6959038a", + "metadata": {}, + "source": [ + "The analysis indicates that games priced in the upper-mid range achieve the highest overall performance across all key metrics, including success, popularity, and player engagement. While free-to-play games also demonstrate strong engagement and reach, they tend to have slightly lower review scores. In contrast, very cheap and expensive games show significantly lower performance, suggesting that both underpricing and overpricing can negatively impact success. These findings support the existence of an optimal price range that balances accessibility, perceived value, and player engagement.\n", + "\n", + "Developers should aim to position their games within the upper-mid price range, as it appears to offer the best balance between attracting players, maintaining engagement, and achieving strong overall performance." + ] + }, + { + "cell_type": "markdown", + "id": "9acdb1d9", + "metadata": {}, + "source": [ + "----------------------------------------------------------------------------------------------------------------------------------\n", + "----------------------------------------------------------------------------------------------------------------------------------" + ] + }, + { + "cell_type": "markdown", + "id": "eb71b7c1", + "metadata": {}, + "source": [ + "## Market and Business Insights" + ] + }, + { + "cell_type": "markdown", + "id": "3bacbd12", + "metadata": {}, + "source": [ + "After analyzing game performance from price, success and playtime we now look at Steam games as a market and business, we want to answer questions like:\n", + "\n", + "* What kind of games win?\n", + "* Who makes them?\n", + "* Where should Steam invest and why?" + ] + }, + { + "cell_type": "markdown", + "id": "eea187f9", + "metadata": {}, + "source": [ + "For this we come up with two new hypothesis:\n", + "\n", + "* H6: Which genre of games perform best?\n", + "* H7: Who creates successful games?\n", + "---------------------------------------------------------------------------------------------\n", + "---------------------------------------------------------------------------------------------" + ] + }, + { + "cell_type": "markdown", + "id": "080473fb", + "metadata": {}, + "source": [ + "### H6: Genre Analysis\n", + "\n", + "(Which genre of games perform best)" + ] + }, + { + "cell_type": "code", + "execution_count": 42, + "id": "890f1dd4", + "metadata": {}, + "outputs": [], + "source": [ + "# Since genre column has multiple genres in a row, first we split the genres.\n", + "\n", + "genres_expanded = steam.assign(\n", + " genre=steam[\"genres\"].str.split(\";\")\n", + ").explode(\"genre\")" + ] + }, + { + "cell_type": "code", + "execution_count": 48, + "id": "2da220a6", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\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", + " \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", + " \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", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
successtotal_reviewsreview_scoreaverage_playtime
genre
Free to Play4346.6273475247.4436620.699454554.437793
Massively Multiplayer3468.1120335352.2005530.611893725.484094
Action1581.9314461938.6939430.709888144.016634
RPG1346.5768961624.9635820.716261276.985850
Animation & Modeling1126.9493671199.1139240.73873388.139241
Design & Illustration880.712644932.8735630.752982112.229885
Adventure869.7240831102.7748210.713753151.656699
Strategy863.9925671044.2056410.692914193.130170
Nudity822.2180451034.2669170.70908099.973684
Simulation819.659607986.1771270.659835154.240855
Racing663.785156802.3212890.667196142.220703
Utilities592.404110658.7465750.688467112.561644
Sexual Content571.975510714.7387760.73021384.395918
Video Production542.710526611.1052630.65296626.973684
Indie540.172236631.5459040.721458112.836105
Sports519.477307655.6202720.666040115.518154
Early Access388.052471509.7549090.70238981.300948
Web Publishing338.892857389.6428570.750886136.214286
Gore323.681564447.2607080.63552747.648045
Violent294.763938416.1613290.63101348.144721
\n", + "
" + ], + "text/plain": [ + " success total_reviews review_score \\\n", + "genre \n", + "Free to Play 4346.627347 5247.443662 0.699454 \n", + "Massively Multiplayer 3468.112033 5352.200553 0.611893 \n", + "Action 1581.931446 1938.693943 0.709888 \n", + "RPG 1346.576896 1624.963582 0.716261 \n", + "Animation & Modeling 1126.949367 1199.113924 0.738733 \n", + "Design & Illustration 880.712644 932.873563 0.752982 \n", + "Adventure 869.724083 1102.774821 0.713753 \n", + "Strategy 863.992567 1044.205641 0.692914 \n", + "Nudity 822.218045 1034.266917 0.709080 \n", + "Simulation 819.659607 986.177127 0.659835 \n", + "Racing 663.785156 802.321289 0.667196 \n", + "Utilities 592.404110 658.746575 0.688467 \n", + "Sexual Content 571.975510 714.738776 0.730213 \n", + "Video Production 542.710526 611.105263 0.652966 \n", + "Indie 540.172236 631.545904 0.721458 \n", + "Sports 519.477307 655.620272 0.666040 \n", + "Early Access 388.052471 509.754909 0.702389 \n", + "Web Publishing 338.892857 389.642857 0.750886 \n", + "Gore 323.681564 447.260708 0.635527 \n", + "Violent 294.763938 416.161329 0.631013 \n", + "\n", + " average_playtime \n", + "genre \n", + "Free to Play 554.437793 \n", + "Massively Multiplayer 725.484094 \n", + "Action 144.016634 \n", + "RPG 276.985850 \n", + "Animation & Modeling 88.139241 \n", + "Design & Illustration 112.229885 \n", + "Adventure 151.656699 \n", + "Strategy 193.130170 \n", + "Nudity 99.973684 \n", + "Simulation 154.240855 \n", + "Racing 142.220703 \n", + "Utilities 112.561644 \n", + "Sexual Content 84.395918 \n", + "Video Production 26.973684 \n", + "Indie 112.836105 \n", + "Sports 115.518154 \n", + "Early Access 81.300948 \n", + "Web Publishing 136.214286 \n", + "Gore 47.648045 \n", + "Violent 48.144721 " + ] + }, + "execution_count": 48, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "genre_analysis = genres_expanded.groupby(\"genre\")[[\n", + " \"success\",\n", + " \"total_reviews\",\n", + " \"review_score\",\n", + " \"average_playtime\"\n", + "]].mean().sort_values(by=\"success\", ascending = False)\n", + "\n", + "genre_analysis.head(20)" + ] + }, + { + "cell_type": "markdown", + "id": "642eae2f", + "metadata": {}, + "source": [ + "Free to play, multiplayer and action genres on top 3" + ] + }, + { + "cell_type": "code", + "execution_count": 54, + "id": "e16f684a", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "genres\n", + "Action;Free to Play;Strategy 108938.000000\n", + "Action;Adventure;Massively Multiplayer 102626.833333\n", + "Action;Free to Play 93925.475000\n", + "Action;Free to Play;Indie;Massively Multiplayer;RPG;Simulation 80360.000000\n", + "Action;Adventure;Free to Play;Massively Multiplayer 55847.500000\n", + "Nudity;Violent;Gore;Action;Adventure;Indie;RPG;Simulation;Strategy;Early Access 30363.500000\n", + "Action;Adventure;Indie;Massively Multiplayer;RPG 29585.846154\n", + "Adventure;Casual;Free to Play;Massively Multiplayer;Simulation;Sports;Early Access 25818.000000\n", + "Action;Free to Play;Massively Multiplayer;Simulation 25491.500000\n", + "Massively Multiplayer;RPG 21763.500000\n", + "Free to Play;RPG;Simulation 21481.000000\n", + "Action;Adventure;Casual;Free to Play;Indie 20951.785714\n", + "Action;Adventure;Massively Multiplayer;RPG;Simulation;Strategy 19493.666667\n", + "Action;Adventure;Casual;Free to Play;Indie;Massively Multiplayer;RPG;Simulation 19114.000000\n", + "Action;Adventure;Indie;Massively Multiplayer;RPG;Simulation;Strategy 17255.000000\n", + "Action;Massively Multiplayer;Simulation;Strategy 17201.000000\n", + "Action;Free to Play;Massively Multiplayer;Early Access 16625.666667\n", + "Action;Adventure;Casual;Free to Play;Indie;Massively Multiplayer;RPG 15666.000000\n", + "Action;Adventure;Free to Play;Indie;Massively Multiplayer;RPG 15071.000000\n", + "Action;Adventure;Casual;Free to Play;Massively Multiplayer;RPG 13900.000000\n", + "Name: success, dtype: float64" + ] + }, + "execution_count": 54, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "# Lets have a look at genres without splitting them:\n", + "\n", + "steam.groupby(\"genres\")[\"success\"].mean().sort_values(ascending=False).head(20)" + ] + }, + { + "cell_type": "markdown", + "id": "ab1dacd1", + "metadata": {}, + "source": [ + "### H6 Visualizations" + ] + }, + { + "cell_type": "code", + "execution_count": 63, + "id": "05a53afa", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "# Top genres success(reviews) without splitting them\n", + "\n", + "top_genres_combo = steam.groupby(\"genres\")[\"success\"] \\\n", + " .mean() \\\n", + " .sort_values(ascending=False) \\\n", + " .head(10) # Only top 10 genres\n", + " \n", + "top_genres_combo.sort_values().plot(kind=\"barh\", figsize=(10,6))\n", + "\n", + "plt.title(\"Top successful genres\")\n", + "plt.xlabel(\"Average Success\")\n", + "plt.ylabel\n", + "\n", + "plt.show()" + ] + }, + { + "cell_type": "markdown", + "id": "1481eb26", + "metadata": {}, + "source": [ + "While Free-to-Play and multiplayer games appear among the highest-performing categories, this does not imply that monetization model alone determines success. Free-to-Play represents a pricing strategy rather than a game genre. The results show that high-performing games typically combine Free-to-Play with engaging genres such as Action, RPG and Massively Multiplayer. This suggests that success is driven by the interaction between monetization strategy and gameplay design, rather than by pricing alone as concluded in previous hypothesis.\n", + "\n", + "Steam should promote games that combine strong engagement mechanics (such as multiplayer and action/rpg systems) with effective monetization strategies(Free or upper mid pricing).\n", + "\n", + "--------------------------------------------------------------------------------\n", + "--------------------------------------------------------------------------------" + ] + }, + { + "cell_type": "markdown", + "id": "5214adc0", + "metadata": {}, + "source": [ + "### H7: Publisher performance\n", + "\n", + "(Who creates successful games)" + ] + }, + { + "cell_type": "code", + "execution_count": 71, + "id": "377041fe", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "count 14353.000000\n", + "mean 1.885390\n", + "std 4.697589\n", + "min 1.000000\n", + "25% 1.000000\n", + "50% 1.000000\n", + "75% 1.000000\n", + "max 212.000000\n", + "Name: count, dtype: float64" + ] + }, + "execution_count": 71, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "steam[\"publisher\"].value_counts().describe()" + ] + }, + { + "cell_type": "markdown", + "id": "ccd3871b", + "metadata": {}, + "source": [ + "~14k publishers out of 27k games which gives a ratio of about 2 games per publisher, this means many publishers have published only 1 or 2 game in Steam which also indicates that we have few top publishers that might dominate the platform (as seen before in H4)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "fbdcc78a", + "metadata": {}, + "outputs": [], + "source": [ + "dev_counts = steam[\"developer\"].value_counts()\n", + "\n", + "# Keeping only developers with at least 5 games\n", + "valid_devs = dev_counts[dev_counts >= 5].index\n", + "\n", + "steam_dev = steam[steam[\"developer\"].isin(valid_devs)]" + ] + }, + { + "cell_type": "code", + "execution_count": 75, + "id": "9ca369d2", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\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", + " \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", + " \n", + "
successtotal_reviewsaverage_playtimereview_scorenum_games
developer
Valve86672.11538595159.8846152663.5384620.89077926
Bethesda Game Studios47296.90000056177.4000002001.8000000.82771310
Tripwire Interactive28663.00000031387.2000001059.6000000.8514555
Klei Entertainment19098.00000020019.571429621.7142860.8956757
Ubisoft Montreal19070.15789523274.4736841027.3684210.78594619
Obsidian Entertainment18309.14285719555.7142861331.4285710.8448677
SCS Software17348.33333318161.250000648.5000000.76474212
Bohemia Interactive15939.81250022225.1875002260.1875000.65784416
Avalanche Studios14870.66666718750.166667735.1666670.7403296
Arkane Studios12659.00000014036.285714591.4285710.8066157
\n", + "
" + ], + "text/plain": [ + " success total_reviews average_playtime \\\n", + "developer \n", + "Valve 86672.115385 95159.884615 2663.538462 \n", + "Bethesda Game Studios 47296.900000 56177.400000 2001.800000 \n", + "Tripwire Interactive 28663.000000 31387.200000 1059.600000 \n", + "Klei Entertainment 19098.000000 20019.571429 621.714286 \n", + "Ubisoft Montreal 19070.157895 23274.473684 1027.368421 \n", + "Obsidian Entertainment 18309.142857 19555.714286 1331.428571 \n", + "SCS Software 17348.333333 18161.250000 648.500000 \n", + "Bohemia Interactive 15939.812500 22225.187500 2260.187500 \n", + "Avalanche Studios 14870.666667 18750.166667 735.166667 \n", + "Arkane Studios 12659.000000 14036.285714 591.428571 \n", + "\n", + " review_score num_games \n", + "developer \n", + "Valve 0.890779 26 \n", + "Bethesda Game Studios 0.827713 10 \n", + "Tripwire Interactive 0.851455 5 \n", + "Klei Entertainment 0.895675 7 \n", + "Ubisoft Montreal 0.785946 19 \n", + "Obsidian Entertainment 0.844867 7 \n", + "SCS Software 0.764742 12 \n", + "Bohemia Interactive 0.657844 16 \n", + "Avalanche Studios 0.740329 6 \n", + "Arkane Studios 0.806615 7 " + ] + }, + "execution_count": 75, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "dev_analysis = steam_dev.groupby(\"developer\").agg({\n", + " \"success\": \"mean\",\n", + " \"total_reviews\": \"mean\",\n", + " \"average_playtime\": \"mean\",\n", + " \"review_score\": \"mean\",\n", + " \"appid\": \"count\"\n", + "}).rename(columns={\"appid\":\"num_games\"}).sort_values(by=\"success\", ascending= False)\n", + "\n", + "dev_analysis.head(10)" + ] + }, + { + "cell_type": "markdown", + "id": "f54d79f7", + "metadata": {}, + "source": [ + "The most successful developers are not one-hit wonders but studios that consistently produce high-performing games across multiple titles." + ] + }, + { + "cell_type": "markdown", + "id": "191ad7ad", + "metadata": {}, + "source": [ + "The analysis identifies a group of developers that consistently produce highly successful games on Steam. Studios such as Valve, Bethesda Game Studios, and Ubisoft Montreal achieve strong performance across multiple titles, indicating sustained success rather than isolated hits. These developers also exhibit high levels of player engagement and strong review scores, suggesting that both retention and player satisfaction contribute to their success. This reinforces the idea that successful games result from a combination of quality, engagement, and scale.\n", + "\n", + "Steam should prioritize partnerships and promotional visibility for developers that consistently deliver high-performing and engaging games. Supporting these studios can drive sustained platform growth and maximize user engagement." ] } ], From 66825fc734257941524bfa423b55141e90f4fe2b Mon Sep 17 00:00:00 2001 From: Sabina Firtala Date: Thu, 26 Mar 2026 20:43:06 +0100 Subject: [PATCH 2/2] Finalize feedback --- notebooks/02_rq_eda.ipynb | 79 +++++++++++++++++++++++++++++++++++---- 1 file changed, 71 insertions(+), 8 deletions(-) diff --git a/notebooks/02_rq_eda.ipynb b/notebooks/02_rq_eda.ipynb index 6ec711e..4eba372 100644 --- a/notebooks/02_rq_eda.ipynb +++ b/notebooks/02_rq_eda.ipynb @@ -1249,7 +1249,7 @@ }, { "cell_type": "code", - "execution_count": 18, + "execution_count": null, "id": "e831bad8", "metadata": {}, "outputs": [ @@ -1278,6 +1278,7 @@ "\n", "price_bins = [0, 5, 10, 20, 60, 1000]\n", "price_labels = [\"Very Cheap\", \"Cheap\", \"Lower Mid\", \"Upper Mid\", \"Expensive\"]\n", + "# SABINA COMMENT: Would make a Very Expensive category too, to keep the same naming conventions everywhere!\n", "\n", "steam.loc[steam[\"price\"] > 0, \"price_category\"] = pd.cut(\n", " steam.loc[steam[\"price\"] > 0, \"price\"],\n", @@ -1318,6 +1319,18 @@ "steam.groupby(\"price_category\")[\"success\"].median().sort_values(ascending=False)" ] }, + { + "cell_type": "markdown", + "id": "e708bb33", + "metadata": {}, + "source": [ + "SABINA COMMENT: I think I would've gone for a scatterplot + correlation coefficient for this as both variables are continuous. I assume you were thinking about making the analysis easier to understand to someone else?\n", + "\n", + "I think you can do the binning after you check the correlation with the appropriate mathematical method as it's always going to be more accurate :)\n", + "\n", + "Also, that's a very big jump in Upper Mid... isn't this just an outlier? Would add a boxplot here to look at the success ranges across the different price bins." + ] + }, { "cell_type": "code", "execution_count": 20, @@ -1351,7 +1364,11 @@ "id": "00c622b0", "metadata": {}, "source": [ - "The analysis suggests that success on Steam is highest for games in the upper-mid price range, followed by free-to-play titles. Very cheap and very expensive games show the lowest average success. This indicates that there may be an optimal pricing range rather than a simple pattern where lower prices always lead to better performance." + "The analysis suggests that success on Steam is highest for games in the upper-mid price range, followed by free-to-play titles. \n", + "\n", + "SABINA COMMENT: Isn't it upper-mid, then lower-mid, then free-to-play?\n", + "\n", + "Very cheap and very expensive games show the lowest average success. This indicates that there may be an optimal pricing range rather than a simple pattern where lower prices always lead to better performance." ] }, { @@ -1363,6 +1380,8 @@ "\n", "* Avoid assuming that the lowest possible price will maximize success. Games priced in the upper-mid range appear to balance accessibility and perceived value more effectively, while free-to-play can also be a strong strategy when combined with high engagement.\n", "\n", + "SABINA COMMENT: Can you also see some player attributes per price range?\n", + "\n", "------------------------------------------------------------------------------------------------------------\n", "------------------------------------------------------------------------------------------------------------" ] @@ -1373,6 +1392,7 @@ "metadata": {}, "source": [ "### H2: Review vs Success\n", + "\n", "(Games with higher average reviews tend to be more successful)" ] }, @@ -1427,6 +1447,8 @@ "Since the \"success\" metric is calculated as review_score * total reviews,\n", "this KPI behaves roughly as an estimated positive review impact, this means:\n", "\n", + "SABINA COMMENT: If success is calculated from review_score * total_reviews, you already have the answer to your hypothesis :D \n", + "\n", "\n", "The review bucket of 0 to 50 has an average game success of 11.19 (Due to low review count on games with 0 to 50 reviews)\n", "But the review bucket of 10k+ reviews has an average game success of 21033, this is because games with huge review counts are translated to \"more success\" which is mathematically predicted.\n", @@ -1445,6 +1467,8 @@ "\n", "For this reason, review count alone is insufficient to determine true game success. Additional metrics such as playtime are required to better capture player engagement and retention.\n", "\n", + "SABINA COMMENT: I really like that you wrote this out and I think it's a good exercise to understand what you're dealing with. However, as soon as we know *for sure* that our variable is equal to a combination of variables, there's no point in making a hypothesis. A hypothesis is something you are trying to prove/disprove when you don't already have the answer.\n", + "\n", "------------------------------------------------------------------------------------------------------------\n", "------------------------------------------------------------------------------------------------------------" ] @@ -1549,6 +1573,8 @@ "\n", "This means the majority of Steam games fail to achieve meaningful player engagement, so Steam is a highly skewed market.\n", "\n", + "SABINA COMMENT: That's really fascinating. How come? Can you check also other characteristics like price, game type, game owner, etc.? Also, can anyone just submit a game or what's the process?\n", + "\n", "Out of 27k games, only ~600 games reach a high playtime, higher playtime buckets contain significantly fewer games, indicating that strong engagement is rare but potentially associated with higher success." ] }, @@ -1595,7 +1621,9 @@ "id": "ed3b9442", "metadata": {}, "source": [ - "High-playtime games are not only more successful, but also tend to receive better player ratings, the increase is not massive but its significant (+9%)." + "High-playtime games are not only more successful, but also tend to receive better player ratings, the increase is not massive but its significant (+9%).\n", + "\n", + "SABINA COMMENT: Makes sense, right? If you don't like a game, you're not going to play it for long." ] }, { @@ -1626,7 +1654,9 @@ "id": "5ae79587", "metadata": {}, "source": [ - "The analysis shows a strong positive relationship between playtime and success. Games with higher average playtime achieve significantly higher success scores, indicating that player engagement and retention are key drivers of performance on Steam. However, high-playtime games represent a relatively small portion of the dataset, suggesting that strong engagement is rare but highly impactful. This highlights the importance of designing games that encourage sustained player interaction." + "The analysis shows a strong positive relationship between playtime and success. Games with higher average playtime achieve significantly higher success scores, indicating that player engagement and retention are key drivers of performance on Steam. However, high-playtime games represent a relatively small portion of the dataset, suggesting that strong engagement is rare but highly impactful. This highlights the importance of designing games that encourage sustained player interaction.\n", + "\n", + "SABINA COMMENT: \"designing games that encourage sustained player interaction.\" Does it? I can't help but wonder whether those games are done by big businesses with many millions to spend on marketing like Riot Games. Any ownership data on these games?" ] }, { @@ -1853,7 +1883,9 @@ "id": "c5de40c2", "metadata": {}, "source": [ - "This Top 20 games list are ~0.07% of all Steam games (20 games out of 27k) yet they generate 30.7% of total success." + "This Top 20 games list are ~0.07% of all Steam games (20 games out of 27k) yet they generate 30.7% of total success.\n", + "\n", + "SABINA COMMENT: I like this, well done! " ] }, { @@ -1885,7 +1917,9 @@ "source": [ "The analysis shows that game success on Steam is highly concentrated among a very small number of titles. Approximately 2.7% of games account for 80% of total success, indicating a strong winner-takes-most dynamic. Additionally, the top 20 games alone generate over 30% of total success, despite representing less than 0.1% of all titles. This highlights the extreme inequality in game performance on the platform, where only a small fraction of games achieve significant visibility and impact.\n", "\n", - "This concentration suggests that achieving high success on Steam requires reaching the top-performing segment, as the majority of games receive minimal attention and engagement." + "This concentration suggests that achieving high success on Steam requires reaching the top-performing segment, as the majority of games receive minimal attention and engagement.\n", + "\n", + "SABINA COMMENT: Very cool and sad. Divya found similar things when looking at football teams. Same questions about ownership of the games as before!" ] }, { @@ -2455,10 +2489,16 @@ "id": "1481eb26", "metadata": {}, "source": [ - "While Free-to-Play and multiplayer games appear among the highest-performing categories, this does not imply that monetization model alone determines success. Free-to-Play represents a pricing strategy rather than a game genre. The results show that high-performing games typically combine Free-to-Play with engaging genres such as Action, RPG and Massively Multiplayer. This suggests that success is driven by the interaction between monetization strategy and gameplay design, rather than by pricing alone as concluded in previous hypothesis.\n", + "While Free-to-Play and multiplayer games appear among the highest-performing categories, this does not imply that monetization model alone determines success. Free-to-Play represents a pricing strategy rather than a game genre. \n", + "\n", + "SABINA COMMENT: \"Free-to-Play represents a pricing strategy rather than a game genre\" What do you mean here? As in, you can do purchases in-game or you have a Freemium model?\n", + "\n", + "The results show that high-performing games typically combine Free-to-Play with engaging genres such as Action, RPG and Massively Multiplayer. This suggests that success is driven by the interaction between monetization strategy and gameplay design, rather than by pricing alone as concluded in previous hypothesis.\n", "\n", "Steam should promote games that combine strong engagement mechanics (such as multiplayer and action/rpg systems) with effective monetization strategies(Free or upper mid pricing).\n", "\n", + "SABINA COMMENT: I would try to separate the monetization strategy from the gametype in the analysis and see what happens when we only look at gametype.\n", + "\n", "--------------------------------------------------------------------------------\n", "--------------------------------------------------------------------------------" ] @@ -2507,7 +2547,9 @@ "id": "ccd3871b", "metadata": {}, "source": [ - "~14k publishers out of 27k games which gives a ratio of about 2 games per publisher, this means many publishers have published only 1 or 2 game in Steam which also indicates that we have few top publishers that might dominate the platform (as seen before in H4)" + "~14k publishers out of 27k games which gives a ratio of about 2 games per publisher, this means many publishers have published only 1 or 2 game in Steam which also indicates that we have few top publishers that might dominate the platform (as seen before in H4)\n", + "\n", + "SABINA COMMENT: I'm a bit confused as to what the publisher variable is" ] }, { @@ -2714,6 +2756,27 @@ "\n", "Steam should prioritize partnerships and promotional visibility for developers that consistently deliver high-performing and engaging games. Supporting these studios can drive sustained platform growth and maximize user engagement." ] + }, + { + "cell_type": "markdown", + "id": "5c57b56a", + "metadata": {}, + "source": [ + "SABINA COMMENT: As per usual, your data analysis is very punctual and good. To improve on this I would recommend you trying a shift in mindset from *formulating hypotheses* to *understanding the business, what works, what doesn't and what can be changed*. \n", + "\n", + "The idea is to move from a fixed structure (hypotheses) to something more fluid, where you start from a key question (\"What drives the success of a game?\") and you answer it by decomposing it, step-by-step:\n", + "1. What is success of a game?\n", + "2. How many games are \"successful\"?\n", + "3. What do those games have in common...\n", + " 1. ...in terms of price?\n", + " 2. ...in terms of owners?\n", + " 3. ...in terms of game type?\n", + "4. What about unsuccessful games? Same questions as before.\n", + "\n", + "Also, you can start asking yourself: \"Are there any other goals the company might have with regards to games?\" For example, maybe they want to spend less on partnerships with the big game developers and find some indie game developers to invest in. Whom should they invest in, then?\n", + "\n", + "You can also use Claude/ChatGPT to come up with potential business goals a company might have and identify how you can reach them using the data. I trust you'll do well ;)" + ] } ], "metadata": {