From 273238d00c6e57f4cba12622c5dc6f5fc27c96cc Mon Sep 17 00:00:00 2001 From: Thomas Wiecki Date: Mon, 20 Apr 2020 15:41:18 +0200 Subject: [PATCH 1/2] Fix ODE NB. --- .../notebooks/ODE_API_introduction.ipynb | 261 +++++++----------- 1 file changed, 94 insertions(+), 167 deletions(-) diff --git a/docs/source/notebooks/ODE_API_introduction.ipynb b/docs/source/notebooks/ODE_API_introduction.ipynb index 5544b27cf4..7533b54d74 100644 --- a/docs/source/notebooks/ODE_API_introduction.ipynb +++ b/docs/source/notebooks/ODE_API_introduction.ipynb @@ -76,14 +76,12 @@ "outputs": [ { "data": { - "image/png": "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\n", + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAnsAAAGsCAYAAACl52GdAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAASdAAAEnQB3mYfeAAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjAsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8GearUAAAgAElEQVR4nOzdd3xT1fvA8c9NmqR70cUqe4nssivoVwVEHIDj60DLFBERFRD5uhABNwqCyBZcPweKiLJEtmxQpsjeHXSP7Pv7IyU0dJeulOf9euWV5ubcm3NzmvTpuec8R1FVVUUIIYQQQlRJmoqugBBCCCGEKDsS7AkhhBBCVGES7AkhhBBCVGES7AkhhBBCVGES7AkhhBBCVGES7AkhhBBCVGES7AkhhBBCVGES7AkhhBBCVGES7AkhhBBCVGES7AkhhBBCVGEeFV2ByiA+Pq1Mj6/RKAQH+5CYmIHdLqvTVWbSVu5D2sp9SFu5D2kr9xAa6les8tKzVw4URXHeROUmbeU+pK3ch7SV+5C2qpok2BNCCCGEqMIk2BNCCCGEqMIk2BNCCCGEqMIk2BNCCCGEqMIk2BNCCCGEqMIk2BNCCCGEqMIk2BNCCCGEqMIk2BNCCCGEqMIk2BNCCCGEqMLcZrk0k8lE27ZtsVqtLtu9vb3Zu3cvAPv37+fdd9/lwIED+Pj40K9fP0aOHIler6+IKgshhBDiWhYLuh3bUJKSUIOCsHToBDpdRdeqSnObYO/o0aNYrVbee+89IiMjnds1Gkfn5NmzZxk4cCCtW7fmo48+4vjx40ybNo3k5GTefPPNiqq2EEIIIQAsFrynf4jXgrlo4uOcm21hYRgHDiVz1AsS9JURtwn2jhw5goeHB7169cqzp27u3Ln4+Pgwa9Ys9Ho93bt3x9PTk0mTJjF8+HBq1KhRAbUWQgghBBYL/k8+gmHtatRr1t3VxMfj885kPPbsInXRVxLwlQG3GbN3+PBh6tevn+8l2c2bN9O9e3eX53v16oXdbmfz5s3lVU0hhBCi9Fgs6LZsQv/Lz+i2bAKLpaJrVCLe0z/EsHY1AIqqujx35bFhzSq8Z0wr97rdCNymZ+/w4cNotVoGDRrEnj170Ov19OrVi3HjxuHh4cH58+epV6+eyz7BwcH4+vpy8uTJAo+t0Sgo1/ynUZq0WiXHfY742mLBY/ufznEL1o6d5T+aCpZvW4lKR9rKfUhblYDFgufHH+A5b47LJU97WDjGwUMxPvdimfy9KJO2sljwWjAXVVFyBXo5qYqC14K5mEaXzbndyNwi2FNVlX/++QdVVXnwwQd5+umn2b9/P5988gnHjh1j2jTHfwK+vr659vXx8SE9Pb3A4wcH+5RpsHdFQIC34weLBd5+G2bOhNjYqwUiImDECBg/vlL9oo8fP54dO3awbt26iq5KqTt37hy33347U6dOpV+/fs7tzrZyE/mdx43A3drqRiZtVUQWC9z/EPz6K+S65BmH99S38P57L/z4Y5n9rSjVtlq/HnIErPlRVBUlLpbgw/vg1ltL7/WF+wR7n376KcHBwTRq1AiA9u3bExISwtixY9m+fXuB+xcWyCUmZpR5z15AgDcpKZnYjGZ8B/wX/ZpVjv9ycpRTY2NRXnsN8+atpC/+utIEfCaTFZvNTmJiRkVXpdSlpGQBkJFhIjExw7WtbPn/B1rZXHseNwJ3basbkbRV8Xi+/zbev/7qeHBtT9iVxytWkPn6JIxjXirV1y6LttKdvoBfMcqnnb6A5Qb5Hiup4GCfYpV3i2BPo9HQsWPHXNtvzY78z507B0BGRu5fjvT0dPz8Cv41s9tVoCy/gBxd4TabimHa++jXrALyH7egX70Sw0cfkPnCuDKsU9Gp2fWy2ewVXJPSd+Wc7HY1++erbeVO55v7PG4E7tlWNyZpqyKzWPCcN6dIlzwN8+eQMXJ0KXcMlH5baQICi1XeFhAovyelzC2CvdjYWDZs2EB0dLTLrFqj0QhAaGgo4eHhnD592mW/y5cvk5GRQYMGDcq1vvkqxrgFzwVzyXz2+TLv3bPZbCxbtpSffvqec+fOERgYyJ139mLQoGEYDAaXssuWLeXzz+eTnJxMixYteeaZ52jcuCkAdrudefNms2bNShIS4gkJCeX223swZMhwPDwcv2Ymk4n582ezdu1qkpISiYyswxNPDOL223s4X+OBB+6hW7dbOX78GPv3/02PHr3444+13HNPX0aOHO0sZ7Vauf/+XtxxRy9Gjx4DwPLlP/F///cV58+fJSgomLvvvpeYmCFotVrnfhs2rGPBgrmcPXuGunXrEhMztND3KCkpienTP2D37p2kp6cRGVmXhx9+lLvu6gPAr78uZ8qUiXz22SI++GAqp06donbt2sTEDOG22+5wHqco51+W5yGEqPx0O7a5jNHLj6KqaONi0e3YhqXrLeVQs5KzdOiEPTQMJT6Owq6hqTodtjp1y6NaNxS3CPZsNhuvvvoqw4cP5/nnn3du//XXX9FqtURFRdG1a1fWr1/Pyy+/7JyRu2rVKrRaLZ06dSqzuhm++RLPb74ssIy9RUuYPROP7X9e14c44P7eeZY3/vcxTP99rHgVz/bee1NYuXIFjz8eQ6tWrTl69B8WLpzLv//+wwcfzHBe3o6Pj2Phwrk89dQz+Pj4snDhHJ599ik+//z/iIiI4MsvP+fHH79n5MjR1KhRk0OHDjBnzix0Oh2DBz+FqqpMmDCW/fv/YvDgYdStW5+NG//g9dcnYDabnYETwA8/fMt///s4jz32JN7e3qiqyrp1a3jmmeec9dm5czvJycn06uV4T5YsWcicObPo3/9hRo16gX///Yf58+cQFxfLyy+/BsDmzRt55ZWX6NGjFyNGjOLo0X+YNOnVQt+jSZNeJSkpkTFjXsbX15eVK1cwefIbhIdH0LZtlLPcSy+Npn//hxk2rDm//LKM1157mXff9aRz5+gin39ZnocQovJTkpLKtHyF0Okw9rkX74XzCi2qWCwE9r6D1EVfYs3x/Squj1sEezVq1KBfv37Mnz8fg8FAmzZt2L17N7Nnz+axxx6jXr16DBkyhBUrVjBkyBAGDhzIqVOn+PDDD3nooYfKNMee9sxp9FsLTu1iyf5X5no/xPm9jqVLdLGOe8XJkyf45ZdlPPXUSAYMiAGgfftOhISEMmnSa2zbtoXOnR3HttlsTJ36Ps2aNQegefObeeih+/j++28YOXI0e/fuoWnTZtx9970AtGnTDk9PT3x9HZfQd+3azvbtW5k4cYqzJ6tjx84YjVnMnv0Jd97Zy9kDGBFRnaefftZZT7PZzIoVP/P33/to1aoNAGvXrqJOnbo0bXoT6enpLFo0j/vu6+fs5evQoRMBAQG8/fZbPPzwY9Sv34BFi+bRrFlzXn11kvP1FQVmz/6kwPdp3749xMQMoVu3WwFo3botAQGB6K7pdX3ggf8SEzPEeexBgx5j4cJ5dO4cXaTzNxqNZXoeQojKTw0KKtPyFSIjA/2mDc6HKriOV8++2mWrURPthfNoL10k8N5epE3/FFO/B8u9uiVWiVcGcZs58BMnTmTEiBEsW7aMYcOGsWzZMkaNGsXLL78MQIMGDViwYAFGo5FRo0axcOFCYmJi+N///lem9bJF1sHcJbrAm+3mlsD1f4jzPX5knRLVfd++PQDceWdPl+23394DrVbL3r27ndtq1KjpDPQAqlUL4eabW/LXX45jtG3bjp07tzNixBC++moxJ0+eoH//h+nZ09HztmvXThRFoXPnaKxWq/PWtWt3Ll9O4OTJ485jN2rU2KU+bdq0Izw8grXZOZpMJhObNm1wHvvAgb8xmUx07drtmmN3y37t7ZhMRv755zBdr7nc8Z//3Fno+9SmTRTz53/GK6+8xC+//ERiYiLPPPMcLVq0cinXq9fV3klFUeje/T8cPnwQk8lYpPMv6/MQQlRi2VkjrlzyvDbx8LVURcEWFu4IKCo539dexuPYvwCYb+mOPSzc5Xl7aBgZ418hccdfpE19H9XDA2w27BHVK6K6xWex4P3BO1Rr3YzAvncTMOhxAvveTXCbZnh/8E6lyI3oFj17AHq9nhEjRjBixIh8y0RFRfHtt9+WY63AVIRLqFqtBk/A2rGzY9xCQnyhY/bsoWG5PsQpP/1aGlV2Sk1NASA4uJrLdg8PDwICAklLu5qyplo11zIAgYFBxMZeAuDRR5/Ay8ubFSt+5tNPZzBr1nTq1avP88+Po23bKFJTU1BVlR49uuVZl4SEeBo1agKAl5frlH9FUejR4y6WL/+J0aPHsHXrJrKyMunR4y6X8xg79rl8j52WloaqqgRcM1C4WrWQvN+cHCZOnMLixQtYt24N69f/jkajISqqI+PGTSAix5dRSIjrsQIDg1BVlbS09CKdf1paWpmehxCi8tFcuojPW2/gsXM7SRu3g8FA1qCh+LwzucD9FFXFOGhopek5yo9++U94LVkEgLlzV1K+/Qns9nx7wIyDh2G7qTnakydKfNWqXLnJyiBuE+xVCTpdpfoQ+/sHAJCYeNklaLFaraSkJBMYeDWgSE1NzbV/YuJlgrJ7HzUaDf37P0T//g+RlJTIn39uYfHiBUyYMJbly1fj6+uHl5c3M2bMzrMuNWvWLrCuPXv2ZsmShezZs4vff19Nq1ZtnHW+cqn4tdfeclk3+YqgoGD8/PzRaDQkJSW6PHclUCyIr68vI0aMYsSIUZw5c4pNmzawaNE8Pvjgbd5772OXY+UMnJOSEtFqtfj7+xfp/Pfv/6tMz0MIUYkYjXjP/gTvjz5AyXRkkvCa8ylZz44mc9QLeOzZheFKiq4cnQNXLoGa7uzpmMRXiWnOncXvhVEA2AMDSZs1F7Ra0GoLnFRi6dwVS+euLtt0Wzah+vlhbdm6TOtcXMVZGaQiM2y4zWXcqiJz1AuYsi+bXvtfwJXH5fUhbt26LQBrslPBXLF27SpsNhstW169THnmzGnOnz/nfBwbe4kDB/6mTRvHANrhwwfx0UfvA46gpHfve+jX7yHS09PIyMigdeu2ZGVloqoqTZve5LwdP36MBQvmYrPZCqxr3br1aNKkGb//vpo//9zinJgBjvGDOp2OhIQ4l2NrtVpmz/6EixcvYDAYuPnmlqxfv86ZSgZgy5aNBb7upUsX6dfvbv74Yy0AkZF1eeyxJ4mK6silSxddym7cuN75s6qqrF+/jhYtWqHX64t0/mV5HkKISkJV0S//ieDo9vhMedMZ6Jl634Opj2PMMzodqYu+ImP8K9hDw1x2t4c5LnlWdE9RoaxW/J8egiYlGYC0aTOx16xVokNpTp3Ef9DjBPbpgeH7/yvNWl6fHBk2CnIlw0ZFXs6Vnr3ylv0h9p4xDc8Fc9HGXV1Bwx4ahnHQ0HJJuQJQr1597rqrD/Pnz8ZkMtKqVRv+/fcoCxfOoW3bKDp27OIsq9freemlFxg2bAR2u41582bj7x/Agw8+AjgCx6+/XkJwcDA339yShIR4vvnmC1q3bktgYCCdO3eldeu2jB//IjExQ6hTpy6HDx9k/vzP6Nixs0svYn569uzNzJkfodV6cOutV1OaBAQE8uijTzBv3mwyMjJo06Yd8fFxzJs3G0VRaNjQMQbwqaeeYdSo4UyYMJb77uvHmTOnWbx4QYGvGRFRndDQMD766H0yMjKoWbMWR44cZtu2LTz+eIxL2VmzpmM2m4mMrMPy5T9y6tQJpk939OQV9fzL6jyEEBVPu/9vfF8d7zLZztqsOelvvY3llu6uhXU6Ml8YR+azzxc46F9z8gT2uvVyrbRR4RQF8+134rFrB8bHYzDffU+JD6XbsQ0lJQXFbsd/xFAy9/9NxqsTwaNiQxh3SpOjqGoBg8duEPHxaWV6fK1WQ3CwD4mJGa6JIivBzB2bzcYXXyxixYqfiYuLJTQ0jDvu6ElMzBBnnr3Jk9/g3Lmz3Hbb7Xz55eekp6fTrl0HnnvuRWpm/6dmtVr5/PP5rF79G/Hxcfj4+BId3Y3hw0c6x5dlZWUxb96n/PHH7yQlJRISEsYdd/Rwea0HHriHNm3a8b//vZGrrklJidx//11063Ybkya9nev5pUu/Y+nS7zh//ix+fv5ERXVg2LBniIiIcJbZuXM7c+bM5Pjx49SoUYOnnx7FSy89z4QJr9O79z15ttXlywl89tlMduzYRkpKMmFh4dx99708/ngMGo3GmWfvtdcmsXjxQi5cOE/jxk0YNmyES2qWopx/aZ3HjSDfz5WodKStQL9mJf4D/otiz06AHhxMxkuvYBwQU+KgxfOrJfiOe570iZMxDn6qVOpZ2m3lsXc31ibNwPv6ll/TrV+H/7AYNMmOnkJz99tInbMQNSj4uutYUobvvsH/mWFFLp+y4AvMV3pvr1NoaHHWJJFgD6jAYE9UOiVpqyvB3nff/Uz16mWX5ke4ks+V+5C2ArKyCI5uj+biBbIGDSVzzHjUwJKnTVHS0wjq2h7txQuoHh6k/LA81zi3kqjMbaU5dZKAJx/F4/BBAGx16pLy+dfYbmpeyJ5Fp6oqFosFs9mE0WjCZDJm38yYTEaMRhO2wwex/7YC++YNmK1WjIARMAE24BYg95pfkPzjilLr2StusCeXcYUQQojiKOyqjKqiX7MSa8vWV9OHeHmRNmM29pBQbI2bXHcVVF8/UhcsIfC+u1DMZvwHDyBpzcYSj4srFaqK94fvYnzkcew1apb64e1165G0Yg3+z43AsPwntKdPEdT7DlJnzMZ8z33OcjabjaysTLKyjBiNWWRlZWUHakZMJtM1gZwp+3Y1oLPb8w5ytefOovvjd7TZWSicp41r3sADuAZ7+WXYKE8S7AkhhBBFYbHgPf1DvBbMdRmrZQsLwzhwKJmjXkB78oRjXN4fv2N8+FHScszAL+1UItZ27Ul/dxp+o59Bk5CA/6DHSV62Ejw9S/V1ispzySJ83pmM15xZpCz+P6wdry+4UVUVo9FIVlYmRqMjcMvMzMI4fCR2f3/ULxdjzMwg9a3XSbycQJYxC6PRiNlsLqUzuqY+Hh5oYy+hAQyKgkeTZhi8vfDbsxsD4Al4A22v2a8ypMmRy7jIZVxxlbSV+5C2ch9Voq2uyafmkg7lygoQtSPRnD93dVxeQCCJ2/ai5pGntDT5jnser0XzAcfymWkfzyrxhI2StpX2nyME9eiOkpWFLaI6SX9sdTlvVVUxmUxkZmaQmZlJRkYGmZkZ2cGco/ftSi+c494R3BUUomhPHEe/YR3Ghx5B9fEtcl0NBgMGgwG93oCnp2eOnw0YDJ4YFAXf7X/iv30blqnvYfDywmDwxNPTQMhTg1BatsY0IMbRe2mx4B/zaN5pcrIfm+7sWeqzp+UyrhBCCFHKipJPTXv2DACqRoPx8Rgyxr9S5oEeQPpb7+Bx6CC6Hdvw/OZLLK3aYBxc9IkDJXWl5y0j8TK6gY9zISuLdCD2qWdI3bmdjIz07MDOcW+1WkvttRVFQd+8BV5towjy9sbLywtPTy+8DAa8MzIwREbi5eWNp4cHfocP4ZWViSE0DKVLNJocE+Jy0h7/F8/PF+L5f1+iyV6uNOXkccx39nKWMX75netOlSjDRkEk2BNCCCEKkiOfWkGrHwGoOh1JK9Zia92mnCoH6PWkzF9C0J3d0F66iO+r47G2bIW1fV7TBApmt9vJyMggLi6WtLR0Zw+c4z4z1882mw3972vQHTsKgKVjZ8wZ6fD3vmJUX+8M1jw9PfH29sbT09MRvGVv9/a++ryXl+N5JY/eS+8P38Vr5nTSZnyKx+FDBV5yR6cDsxnDb7/guXihy/q9APaQkKKtaV/ENDkVSYI9IYQQogBFzacGoFgsaDLSKThNfOlTw8MdEzbu7425Z2+szXLPULXZbKSnp5GWdvWWnp7q8jgrKwODwYOsLDN2e+GjvLTHj6HLXkfdVr065q63oNFo8PLyxsfHB29vb3x8fK+598Hb2wcfH288Pb3wKKV8edoTx/B+d4ojH1/MYyjkXrwg5xJmluhueM/4CE1CvEsZc3Q3jE8OwnRXH9Dri14Bna7C8ugVRoI9IYQQogBF6t25jvKlwWq1ktywEWcWfU1y9eqkHTxAWlqKSyCXmZlR4Bg4AI0m77F+Wq0Wb28fl2DNz2wi/LOZ+ANe3j5YF36JZ5NmeHt759nrVtZs9RuSOn8JfsMGorE4JmkUtIQZZrMz0LMHBmJ8+DGMTwzE1qhx+Va8HEiwJ4QQQhRADSpePrzili+MxWIhLS01R29cmsvjK4FcSeh0Ovz8/PD19cPPz5+AAH9q1AjFalXw9PTO7oHzyfOyqf+gARjSHBMcUz+cjilHEvmKYu7RC3x9UZMSKSjcVBUF3cH9mDt2xjggBtM994OXV7nVs7xJsCeEEEIUwNKhE/aQUJSE+EIDiJLkU1NVlYyMdJKTk0lOTiYlJTn75ySSk5PJyEgvUb0NWi3V1q3Fp34DPHvfg5+fn/Pm6+uPn59friCuOLNx09+cgpIQj71OXUz9HixRHUubbsc2NEmJhZZTVBUlIYHMuZ9X2kuvpUmCPSGEEKIgJhN2b+9C/2AWlE/NYrGQkpJCSkqSSyCXnJxMamoKFoulWFXy9PTKEbz55wjirj4OfWoQhp3bUffuJvmxJ7F26lys1yiMvVZtUpb+AsWse1lyh0vuFUGCvRvY5Mlv8NtvvxRYpnXrtnzyyZxyqpF7uXjxAg8+eO8NtR6tEDcaJT0N/0cfxOPMaee2vPKpoaok3nY7Z/s9RNKB/c7euSv36enFy+fq5+dPYGAgAQGBBAYG4u8f4BLc6YowyzPzmVHo16xEsVgIGDyApLUbsV/vko6q6prDz8OjxGv7loWKvuReWVWeFhLlLiZmCPfd19/5+PPP53H06BEmT37fuc3Hx6ciqiaEEBVOSUsl4JEH0O3YBoCp+22ktGpD2peLSbycwGUgCbjs40t823ZktmkH335VpGPrdDpnIBcYGJR9H0hAQBABAQGlMkPV2r4j6W9/gN+Lo9DExzlW2PjpN8gnz1yhVBW/50ZgbdiIrJGjQaO57jqWNkuHTthDwxyX3AuYjFIZljArTxLs3cBq1qxFzRzrKAYGBqHT6bn55hYVWCshhKhYqqqSceE8pgEPc+zAfhKACzfdzLmut5BlNsOTg9CePwdZWeDlha1mrTwDH19fP4KCgpxBXc57Hx+fcpmxahwQg8e+vXgtWYhu9y58x79I+oczSrTChuHbr/H85ksAlIx0Ml9+rbSre/10OrIGDcXnnckFFqsMS5iVJwn2RKF+/XU577zzFmPGvMzcuZ9isViYNWseY8c+R5s27fjf/95wKTtlykS+++5nqmdfLjhx4hizZ3/Cvn17AWjXrj0jR452CTSvlZSUxPTpH7B7907S09OIjKzLww8/yl139XF5nc8+W8QHH0zl1KlT1K5dm5iYIdx22x3O45hMJubPn83atatJSkokMrIOTzwxiNtv7+HyesuX/8T//d9XnD9/lpCQEHr3vocnnhiMVqt1ltmwYR0LFszl7Nkz1K1bl5iYodf93grhZLFU2oSsVZWqqqSmpnD5cgIJCZe5fDnBcTt/FuXLJWgvXQTA2qAhpp53wZU1VzUabLUjr+mdyxnMOXrninKptTykT3kXj8MH0e3agdeXi7G2bovxyUHFOobmxHF8x48BwB4SStagp8qiqqUic9QLeOzZVegSZpnPPl+BtSxfEuyJIrHZbHzzzReMH/8qKSnJ1K1br0j7nTlzmuHDB1OnTh3+9783sNlsfP75fEaMGMyiRV8TFBSc536TJr1KUlIiY8a8jK+vLytXrmDy5DcID4+gbY7p/S+9NJr+/R9m2LDm/PLLMl577WXefdeTzp2jUVWVCRPGsn//XwwePIy6deuzceMfvP76BMxmszNwXLJkIXPmzKJ//4cZPfpFzp07yfTpM7h06RIvZ//nunnzRl555SV69OjFiBGjOHr0HyZNevU631UhAIsF7+kfFp7pX5SY3W4nOTmJy5cvZwd2CSQmOn7Oa2KEftNGPK4Eeg0bozzwMDXDwwkOrka1aiGEhDju/f0DKiSfXLEZDI6Ey3d0QxsXi++EsVib3oS1YxEvYZrN+A8fhCZ7VnDajE9Rw8PLsMLXyU2WMCtPEuxdh4sXL7B16+ZCZ1EpioKvr4H0dFOhCS1LQqfT0aVLtLMnraw88cQgunSJLtY+CxfOxdPTk48+moVP9kLVUVHteeih+/jqqyU888xzee63b98eYmKG0K3brYBjokhAQGCu/5QfeOC/xMQMAaBjx84MGvQYCxfOo3PnaHbt2s727VuZOHGKsyevY8fOGI1ZzJ79CXfe2Quj0ciiRfO4775+jB49Bq1Ww1133YFO583UqW/y8MOPUb9+AxYtmkezZs159dVJzuMoCsye/Umx3g8hXFgs+D/5CIa1qwvM9F/ai6iXK4sFjz+3gzULDw8vbFEdy+xc7HY7iYmJV3vosgO7pKTEIq3L6uXlTUhICNVGPEskEOLji+fsefgEBrlHUFcAe0R1Uhd8QWDf3mC343H0SJGDPZ+pk9BlX5nJfOoZzNdcGamU3GAJs/Ikwd512LVrJ8ePHyu0nEaj4OWlL/LyMyWh1xu45577yuTYVzQqQVbx3bt30qZNWwwGT+eXrbe3Dy1btmHnzu357temTRTz53/G0aP/0KlTZzp1is4zMOzVq4/zZ0VR6N79P8yf/xkmk5Fdu3aiKAqdO0e7fNF37dqdVat+4+TJ41y+fBmTyUTXrt2wWq2oqgar1Up0dDcAdu3aTs2aNfnnn8MMGTLc5bX/8587JdgT18V7+ocY1q4GCs707z1jGpkvjCv3+l2XPHos/Sm9Hku73U5CQgKxsReJjb3EpUuXiI+PK1IKE19fP6pVq0ZISIizt65atRDXCWm393CMw6tCgYG1Q0fSPpyBvWYtLNnfcYXRrV+H98yPAbC0aEXGK2+UYQ3LQCVewqw8SbB3HaKi2mM2mypFz15UVPtSP+61vLy8i71PSkoyv/++ht9/X5PrucDA/Ke8T5w4hcWLF7Bu3RrWr/8djUZDVFRHxo2bQEREdWe5kJCQXMdUVZW0tHRSU1NQVZUePfL+UqYz288AACAASURBVEtIiCctO/v72LF59zBeKaOqKgEBgS7PVasWkuc+QhSJxYLXgrm5xhRdS1UUPBfMda/LTqXcY5kzsLt06SKxsbHExcUW2lvn7+/vDORCQhz3wcHV8LpmpQQlIQGvmR+T+eJLcGWcbklnrFZypocfLXJZJT4ev5GOsXmqtzdpny2osu9LVSfB3nWoXr0G/fs/VGi54mQkdyeKomC3uy73nZWV6fLYz8+Pdu068Mgjj+faP+fkh2v5+voyYsQoRowYxZkzp9i0aQOLFs3jgw/e5r33PnaWS01NITi4mvNxUlIiWq0Wf39/fH398PLyZsaM2Xm+Rs2atdm//y8AXnvtLSIjI9FoNPj7e5KaasRutxMUFIyfnz8ajYaka7Kyp6am5Ft/IQqj27HNZYxefhRVRRsXi27HNrfpobieHkubzUZCQgJxcZeKHNjpdDpCQ8OIiIggPDyCkJBQgoOrYShCYKLExRH4wD14HDmM5vw50qd9UilTipQVzdkzeOzbi/me+3Jdclf273eOd0uf8h62ho0quLaipCTYEyXm7e1DXJzrH6u//97n8rh167acOnWShg0bO/NGqarKxImvULt2JI0aNcl13EuXLjJixBCeffZ5brvtDiIj6/LYY3U5cGA/586dcSm7ceN67r+/v/O469evo0WLVuj1elq3bsvXXy9BVVWaNWvu3GfFip/ZsOEPJkx4nebNb0an05GQEEePHr2cgfm2bXuYNWs6MTFDCA+P4OabW7J+/TpiYoY4x+5s2bLx+t9EccMqbuZ+zZkz0LWMKlOaitFjqZs/h9iHHyX2coLLpdjCAruwsHDCw8MJD6+eHdyFoClBgKbExhLYvw8eR/9xPLbbHUmDbxC6LZvwHzwAJTUV47rHMKz+Lfcl9/v6ovr6Y8zjH3bhPiTYEyXWpUs0X3yxiCVLFtK8eQs2b97A7t27XMrExAxl+PCBjBv3PH379kevN7Bs2VI2bVrPW2+9k+dxIyKqExoaxkcfvU9GRgY1a9biyJHDbNu2hccfj3EpO2vWdMxmM5GRdVi+/EdOnTrB9OmOnrzOnbvSunVbxo9/kZiYIdSpU5fDhw8yf/5ndOzYmcBAx2XZRx99gnnzZpORkUG7dlFkZaUybdo0QKFhQ8c4xaeeeoZRo4YzYcJY7ruvH2fOnGbx4gWl+4aKG0pxM/f7jR6B59dLMPfsjbnXXdgaVM5eFt22rXn2WNqAeOBC9u2iqhIbH0f6229hqx2Z97GyA7uIiAjCwiKIiKhOtWrVShTYXUtz6SIB/frgcexfALIee4L0D6bfUL16qocOJTUVxWrF68vP87zk7rnsR0x39gSr1X2GEYhcJNgTJfbEE4NITk7mq6+WYLVa6dKlK+PHv8r48S84yzRs2IiZM+cxZ84sJk16HVVVqV+/AVOnvk90dPd8jz1lynt89tlM5s2bTUpKMmFh4QwcODRXsDdmzHgWL17IhQvnady4CdOmzaRVqzYAaDQa3nvvY+bN+5QlSxaSlJRISEgYDz/8mHMGL8DQoU9TrVoIS5d+x1dfLSYgIIB27dozdOgz+Po6ZhC3atWG99+fzpw5M5kwYSw1atTg5Zdf46WXbpw8TaJ0FTnTP6DguPyp37YV/batMPEVrA0bYe7ZG1Pf/lhbti63eudiMuGxby+6bVvQ/7kFj61bAMgCzgCns2+xQJ79dVlZgGtgd6XHrrQCu2tpLl4goO/deJw47qjCgIGkvzfthgr0AKwdO2H+z50YVv8GVMFJQsJJUctixoCbiY8v3pqFxVVVx+xVpLySN5cGaSv3URXayvuDdwrN9A+Q9egAVP8ADCtXoD110uW5jNFjyJxQzJUMrjOBs/afIxh+/B7dtq3o9uxCMRpJ52pwdwqIwxGo5kUHVM++BQ2IwX/MeKqFR5RJYHctzflzBPa92/k+ZsUMJv3tD264QA8Ai4VqrZuixMdTUGKZK0uLJe49JL17lURoqF+xykvPnhBCVBBj3wfwfv8dFJvV2YN3Rc5M/+nvfQQ6HRkTJ6P99yj6lb9iWPUrHrt2YO55l8sxfSaMRXvqpONyb49ergvflyCBs5KchCYuDlvjq+NrPQ4dwPbhu5zAEdidBhIAe0Ag9pq10B47CmYzCqDBEdTVzr6vAVTL3g7AkkXY1q4ma9BQjANiUHNMuCp16ekE3t8b7elTAGQNHkb6lPdKtHRYVeCYJBRfaDl3nCQkXEmwJ4QQFcFmw++FZ1FsjoubakAgSkqy8+k8M/0rCrbGTchq3ISsUc+jxMejVssRHNntGH7+CW1crGM27FiwtG7juNx7ew983p1caDqUtHemoduzE/2fW9D9uRXt4YNYWrfh1DdLOXfuLGfPnuX88WOYAXu1EGy1amOvXRtbrdqovn5otVpq/f0XjX/7hTo4gjx9Pm+B3csLTVYW2osX8J08EXvNWpgeeLjU3uJcfH3JihmC78RXyBz2NBmT3r5hAz0o/iSh4pYXlYdcxkUu44qrpK3ch7u3ldfM6fhOfAXIHjP29vvXnelfSUvF541X0K/6zWWJqOJScfTUXRlvd0pRuPTs86C/GrYpxixUTy90Oh01atSkdu1IatWqTfXqNdAB/jGPFro2aeqn8/H8ZRlen81CSbxM4u4DLq/hsXsn1jbtSv0Sq27TBkdS4Rs40APHbNzAvncXuXzyjyukZ6+SKO5lXAn2kGBPXCVt5T7cua00J44T3K0jitmMtV59kn7fDNmTgUqF3Y7Hvj3oV/2KYeVveBw+WGBxFccEitM5bulaLfaI6thq18ZeK9IxY1arxWAwULNmLWrViqR27dpERFTPO2emxZLn2qS2sPDcPZaqiubCeew1aznLaY/+Q3B0e6wNGpI1ZDjGhx8t0XukOXkC1c8fNUSSoOdisVCtdbPCJwnJmL1KR4K9EpBgT1whbeU+3LqtVBXPhfPweesNUr79EWtUhzJ9Of3S7wgYPvjqywOXgJM4xtydAYw5ypv/cweWlq3BwwNPTy9q165NrVq1iYysQ2hoWPEmUlgsGHZtx9+aRaqHF6Yiro3r+/IYvObPcT62+wdgfPxJsgYPw55XqpY8Jp1oz5wioN89qEHBJC9dXrbjAd1UUScJZYx/RWbjViIS7JWABHviCmkr91EV2kpJSUa9Zhm+sqD/5We8Bj3OSeBo9i2/bz1fIGzEKCLu60utWpGEhIQ4E4mXVEnaSklNwfPrL/Ca+xnaM6ec21WNBvPd95I59GmsHTuB1ZrnpBN7cDUwmdBkpAOQ9s6HGAcOufZlhMVStEvuRVzaTpQPCfZKQII9cYW0lfuQtipcamoKx48f49RvK4j76P0889wFAHWBOtm3YCCllMdmXVdb2WzoV/2G15xZ6Ldudnkq47kX8Ti43znpxCVQ4ers5owXxpE5/pXrO4mqrDiX3EWlIKlXhBCiElJSkvF77hkyXnmjzNYYtdvtXLx4gePHj3H8+DHir/R0aTR4e3tDZiY6oB7QGGgI5FzH48rYLEuHTmVSvxLRajH37oO5dx+0+//Ge+6nGJZ+h2I2o0lJzn8N3pwP9PnNBxYA6HRkvjCOzGefL9Eld1H5Sc8e0rMnrpK2ch/u1lZ+Tw/B84dvUb28SFq1HlvTZqVyXJPJxKlTJzl27F9OnDhOVlZmrjK+vn7ctH8fLX/4jnrknwoFymZsVmm3lRIXh+G3X/B5d4pMLihl7va5ulFJz54QQlQyhp9+wPOHbwEwd4nG1qTpdR0vKSnR2Xt39uwZ7Pbcf5QjIqrTsGEjGjRoSFhYOIrVin9qKvpCxmZlPlv5lwBUw8KwNWyU5xq815KEwEJIsCeEEGVKc/ECvuMcAZQ9OJj0j2YWO7+bzWbj/PlzHD9+jBMnjnH58uVcZXQ6HXXr1qNBg4bUr98AX1+/awuQuuirPMdm5ZnAuZKThMBCFJ0Ee0IIUVbsdvxGPY0m2bEyRtr707GHRxRp18zMTE6ePMGJE8c4efIERqMxV5mAgIDs4K4hkZF18PAo5Cs9x9is603gXNHUoKDCC11HeSGqEgn2hBCijHjN/wz9hj8AMP73Mcx97i2wfFJSIkePHuXEiWOcO3eWa4dUK4pCzZq1qF+/IQ0aNCx5WhSdzu0vaVo6dMIeGlbkMXuVatKJEOVMgj0hhCgD2n+O4DPpdQBskXVIn/xOnuWysrI4cuQQhw4d5Pz5c7meNxgM1KtXn/r1HZdnvb29y7TebkOnI2vQ0EITAiuqinHQULfruRSiNEmwJ4QQZcD39QkoRiOqopD2yWeofv7O56xWKydOHOfQoQMcP34Mm83msm9QUBANGjgmV9SqVTvv5cgEmaNewGPPrkITArvDpBMhypIEe0IIUQZSZ3yG34vPYmvUBEunLqiqyoUL5zl4cD9HjhzBaMxyKe/n589NNzXnpptuJjQ0tIJq7Waq2KQTIcqK5NlD8uyJq6St3IdbtJWqknT5Mof+OcyhQwdIumZGqF6vp3HjpjRvfjORkXWue1myyqpc2iqPtXElyCs+t/hcCcmzJ4QQFUZVQVEKHIen0WioW7ceN910M40aNUYnAUnpqAKTToQoKxLsCSFEKbBarVx6aiD7zRYOtWiJ7Zrnw8MjaN78Zpo2vQlfX98KqaMQ4sYkwZ4QQpRQznF4x376EXX5MgC0sZew3dHDOQ6vefMWhISEVHBthRA3Kgn2hBCimJKSEjl06KBzHJ6SkYHXimUogM7gSYMnBtLslu5VehyeEMJ9SLAnhBBFUNA4PM9Vv9I4K4uWQI1p07E/8N+KqaQQQuTBbYO9kSNHcujQIdatW+fcdvr0aaZOncquXbvQarX06tWLsWPHyvgYIUSJqKrK6dOn2LdvT5758MLDI2hz+hQdTxzHFzD2e4A0CfSEEJWMWwZ7y5YtY82aNdSsWdO5LTU1lSeffJKQkBDefvttEhMTee+99zh37hzz58+vwNoKIdyN1Wrl8OFD7Nq1g/j4OJfn/P39adbMMQ4vPCWJoNcnoAC2GjVJf/uDiqmwEEIUwO2CvdjYWCZPnkxEhOti4l9//TXJycksXbqU4OBgAMLDwxk2bBi7d++mXbt2FVFdIYQbycjI4K+/9rJ37x4yMtKd23U6HU2aNHPNh2e14vf4gyiZmQCkTf8UNTCooqouhBD5crtg75VXXqFr164YDAZ27Njh3L5582batWvnDPQAoqOj8fHxYePGjRLsCSHylZCQwO7dOzl4cD9Wq9W53dfXj7Zt29GyZetca9LqNv6Bbs9uADKfegZLt1vLs8pCCFFkbhXsfffddxw8eJBffvmFd9991+W548eP07t3b5dtWq2WWrVqcfLkyQKPq9EoZTpjTqtVctxryux1xPWTtnIf19tWqqpy6tRJdu3ayYkTx5zbNRqFsLBw2rfvSLNmN+W7Lq39zp6k/rAcz5kfY3xtIlqt/L7kRz5X7kPaqmpym2Dv/PnzTJ06lalTp7r03l2RlpaGj49Pru0+Pj6kp6fn2p5TcLBPuaRHCAjwLryQqBSkrdxHcdvKarWyf/9+tm3bRmysYy1VLy89iqLQuHFjOnfuTJ06RUyZ0q8P9OtD7m8kkRf5XLkPaauqxS2CPVVVmTBhAt27d6dnz575lslPYV/aiYkZZd6zFxDgTUpKJjbbDb8UcaUmbeU+ittWGRkZ7Nu3J9d4PA8PHS1atCQqqj3BwdUASErKzP9AWVng5XXd9b+RyOfKfUhbuYfg4NydWwVxi2Dvyy+/5J9//mH58uXO8TRXgjur1YpGo8HX15eMjIxc+6anpxMeHl7g8e12FSjLX2pHV7jNpsrC0pWetJX7KFpbJSQksGvXDg4dOpDHeLwoWrVqjVd28JbncSwWdDu2oSQloT1xDK/5c0ib/imW7reV7ulUafK5ch/SVlWRWwR7q1atIikpiejo6FzPNW/enJEjR1KvXj3OnDnj8pzNZuPcuXP06NGjvKoqhKgEro7H28HJkydcngsPjyAqqgNNmzbLdzweABYL3tM/xGvBXDTXpl95/CES9x5CDQkti+oLIUSpcotgb+LEibl67WbOnMmBAwf49NNPCQsLQ1EU5s+fT2JionNM3+bNm8nMzKRr164VUW0hRDlz5Mc7yM6dO0hIiHduVxSFBg0aEhXVgdq1IwsftmGx4P/kIxjWrkbNo6zGZMLvuRGkLvoKdLrSPg0hhChVbhHs1a9fP9e2wMBA9Ho9LVq0AODRRx/liy++YODAgYwcOZLk5GTee+89unXrRtu2bcu7ykKIcpRzPF5m5tV/DHU6x3i8tm2jnOPxisJ7+ocY1q4GQMlnPLBhzSq8Z0wj84Vx11d5IYQoY24R7BVFcHAwixcvZsqUKYwZMwYfHx969erFuHHyRSxEucgxtk0NCsLSoVOZ93rFxcWxZs16DhzIKz+e63i8IrNY8FowF1VR8g30AFRFwXPBXDKffV5694QQlZqiFjSN9QYRH59WpsfXajUEB/uQmJghA14rOWmrEshnbJstLAzjwKFkjnqh1IOh2NhLbNmyiYsXz5CVZc6eZFWM8XgF0G3ZRGDfu4tcPvnHFVi63lKi17pRyOfKfUhbuYfQUL9ila8yPXtCiApQwNg2TXw8Pu9MxmPPrlIb25acnMSmTRs5fPggGo3izI/XqFEjoqI6UKtW7RKnUVKSElGDglGSkoq5X/HKCyFEeZNgTwhRYgWNbbvyuDTGtmVmZrJt2xb27t2DzWYDQKPREBUVRbNmrfD3DyzRcTVnTmP48Xs8l34PxiyStu1FDSre+rbFLS+EEOVNgj0hRMmUw9g2i8XC7t072b79T0wmk3N706bNuPXW22jQoHaxLzcp8fEYfl6K59Lv0e3c7vKcx197sXTohD00DCUhvtDzsoeGOcYmCiFEJSbBnhCiRHQ7tuXKP5cXRVXRxsWi27GtyGPb7HY7+/f/xZYtm0lPvzqmNjKyDt2730b1kFAMu7bDXzvx8PDCFtWx4EDSYsGw9Ds8l36HbuN6lOzewSusTZpi6v8Q9uo1QKcja9BQfN6ZXOh5GQcNlckZQohKT4I9IUSJFHesms/LY7B2vQVLqzaY/vtYnmVUVeXYsX/ZuHE9ly8nOLeHhobRvfut1KsVic+MaS6TQfwpwmQQrRafKW+ivXjBuclWqzamfg9i7PsAtpuaQ46xfpmjXsBjzy4Ma1bl6rm88th0Z09Hb6UQQlRyEuwJIUqkuGPVdEcOoztyGI9rgj3N+XN4zZ7J6YjqrE1J5qzdjqrXA+Dv70/Xrt1o3vxmNDZb4ZNBdu8ka/BTGJb/hDFmMNbW2Tk2NRpMfR/A8/++xHRvX4z9HsLavgNoNPlUVkfqoq/wnjENzwVz0cbFOp+yh4ZhHDRUUq4IIdyGBHtCiBKx1Y5ENRjAbC54bBuApyfWJk3x+PcotqbNXJ5P2bSRtZ/N5HD2Y29AHxBA58ZNaNe+Exw+iE21o1+zqvDJIGtXO8uoPj5Xgz0g88VxZPzv9aIHaDodmS+MI/PZ58s9f6AQQpQmCfaEEMVmWLYU37GjUXJMmsiPAmSMHuOYjWu3o2SvcJGensaWLZs5/MtPeGg0KHY7HkBHIDolBa+dO2DnDgBsYeEoqlroZJArVJ0OxWh03ebnX9zTdNDpJI+eEMKtSbAnhCgyJTUF35fH4vndN85ttho10V44X7SxbRoNRp2eHZs2sGvXDiwWC7RsjaVFK1qEhNLd149qZ0+jHDmM9cghtMePodhs2KtXR/fXviLXM3XhF5h73FVq5y2EEO5Mgj0hRJHo/tyC38in0J49A4A9OJi096dj7nlXkca22Ww29u3bw9atW8jKynSWq1+/Ad263UZYWBgAmTlf1GRCe/wY+vW/FyvYw2y5jjMVQoiqRYI9IUTBzGZ83pmM1ycfOXvuzLfdTtr0T7GHRwAUOLZNVVWOHD7Epk3rSU5Odh62evUadO9+G5GRdfJ/bYMB203NsSYlFqvKkuhYCCGukmBPCFEg3/Ev4vXF5wConp6kvz4J46BhLqlKgDzHtp06dZKNG9dz6dJF57agoCBuueVWmjRpWuSlzSTRsRBClJwEe0KIAmWOHI3n0u+xNmhI2qfzsDVuUug+sbGxbNz4BydPnnBu8/b2oUuXrrRq1QatVlu8SkiiYyGEKDEJ9oQQLpT4eNRq1Zw56Oz1G5C8dDnWm1tCdv67/KSmprBp00YOHTqAmt0Dp9friYrqQPv2HTEYDCWulyQ6FkKIkpFgTwjhpF/+E34vjiJj3ASMQ4Y7t1vbRhW4n6qq7Nu3hw0b/sBsNgOg0Who2bIVXbrcgq+v7/VXThIdCyFEiSiqWoSkVVVcfHxa4YWug1arITjYp9gLtovyd6O2lZKagu+EcXh++zXgGJuXuOMv7BHVC903OTmJlSt/5cyZ085tTZo0JTq6O9WqVSubClssGHZtx9+aRaqHF6bC1sYVFepG/Vy5I2kr9xAa6les8tKzJ8QNTrdtK37PDLuaUiUoiLT3Py400FNVlT17drFx43pHvjwcy5v17NmbevXql3GldViju0GwD9bEDJA/SkIIkS8J9oSo6iyWvJf7MpvxeXcKXjOmXU2p0v020mbMLjTQS0pKZOXKXzmbHSACtGnTlm7dbruucXlCCCFKnwR7QlRVFgve0z/Ea8FcNPFxzs22sDBM9/ZFt+1PdAf+BkA1GMh47U2yBj/lnJiRF7vdzu7dO9m8eaOzNy8wMJCePXtTp07dMj0dIYQQJSPBnhBVkcWC/5OPYFi7GvWaXHaa+Hi85312tejNLUmbNRdb02YFHvLy5cusXLmC8+fPObe1bduObt1uQ1/ILF0hhBAVR4I9Iaog7+kfYli7GiBXEuKcj82dupDy/c8FplSx2+3s2rWTzZs3YLVaAUdi5J49exe8+oUQQohKQYI9IaoaiwWvBXNz5aK7lqooaE8cz70SRg4JCQmsXLmCCxfOA6AoCu3aRREd3V1684QQwk1IsCdEFaPbsc1ljF5+FFVFGxeLbse2XMuc2e12duzYztatm5y9ecHBwfTqdTe1atUuk3oLIYQoGxLsCVHFKElJ11U+Pj6elStXcPHiBcfzikJUVAeio7uhk1x2QgjhdiTYE6KKUT09i1c+KAgAm83Gjh3b2Lp1MzabDYBq1arRq9fd1KxZq9TrKYQQonxIsCdEFeKx7U/8Xh5TpLKqomAPDcPSoRNxcXH89tsvxMZeAhy9ee3bd6Rr11ukN08IIdycBHtCVBH6n3/Ef2hMgZMyclJUlYyYwWzduZ0//9ySozcvhLvuupsaNWqWZXWFEEKUEwn2hKgiLLf+B3vNWmji48gY/wq6LZsxrF2Va1bulcenu93Kt0HBxG3eCIBGo6FDh0506RKNh4d8NQghRFUh3+hCuCuLBaxW8PICQPUPIPWzBagBgdgaNyFr2Ai8Z0zDc8FctHGxV3cLCWXtHT3YULMW9ssJAISEhNK7dx8iClkmTQghhPuRYE8IN6Q9chi/kU9hbduO9HenObdb23e8WkinI/OFcWQ++7xzbdwLqp3lsbHEJ14GHL15nTp1oVOnLtKbJ4QQVZR8uwvhTmw2vGbPxOftSSgmE7q/92G6r1+uPHkudDqyOnZm69bN7NixDbvdDkBYWDh33XU34eER5VR5IYQQFUGCPSHchObkCfxHPY1u+58AqBoNWSNHY4nqUOB+sbGxrFjxMwkJ8Y7jaDR07tyVTp26oNVqy7zeQgghKpYEe0JUdqqK56L5+E58BSUzEwBrvfqkzfgMa4eOBe56+PAhVq5cgcViAa705vUhPDy8zKsthBCicpBgT4hKTHPhPH7PjUC/4Q/ntqzBw0h/ZSL4+OS7n91uZ9OmDWzP7gVUFIWuXW+hY8fO0psnhBA3GAn2hKjENBcvoNu0AQBbzVqkfTwLS7dbC9zHaDTyyy/LOHHiOACenp706XMf9es3KOvqCiGEqIQk2BOiIlgszhmyalAQlg6dII+VKqzt2pM56gU0sZfImDQV1T+gwMMmJCTw00/fk5iYCDgSJPfr9wBBQcFlchpCCCEqPwn2hChPFgve0z/Ea8FcNPFxzs22sDCMA4dird8AvH0w97zL+Vzmy6+CohR66GPH/mXFip8xmUwANGrUmN6978FgMJT+eQghhHAbEuwJUV4sFvyffATD2tWo1wRvmvh4fN6ZDIA9OJjETTtRQ0MdTxYS6KmqyrZtW9m8eSNq9koZXbveQpcu0ShFCBKFEEJUbRLsCVFOvKd/iGHtaoBc69e6PM7MxOP4v1iuBHsFMJvN/PbbL/zzzxEA9Ho9vXvfQ+PGTUqv4kIIIdyaBHtClAeLBa8Fc3OtU3stFcDXF0u79oUeMjk5iR9//IH47MvBQUFB3H//A4QWIUgUQghx45BgT4hyoNuxzWWMXn4UQElIQLdjW4GrYpw6dZKff/4JozELgLp163HPPffjlb1OrhBCCHGFBHtClAMlKalUyquqyu7dO1m/fp1z2bP27TvSvfttaDSa666nEEKIqkeCPSHKgRoUdN3lrVYrq1ev5MCBvwHw8PCgZ8/eNG9+c6nUUQghRNUkwZ4Q5cDSoRP20DCUhPiCx+wpCvbQMEfevRzS0lL56aelXLx4AQA/P3/69u1PRET1Mq23EEII9yfXfYQoDx4eZA0aWmCgB45ZucZBQ10SLJ8/f47Fixc5A71atWozYECMBHpCCCGKRHr2hCgHPq//D9XbG9MdPTGsXZVrVu6Vx6Y7e5L57PPO7X/9tZe1a1djs9kAaN26Dbff3kPWtxVCCFFkEuwJUca8Zs3Ae/YnAGSMGY81qj2eC+aijYt1lrGHhmEcNNQR6Ol02Gw21q1bw969ewDQarXccUcPWrVqUyHnIIQQwn1JsCdEGTJ8/3/4vvE/AGwR1TE+8jj22pFkPvt8vmvjZmRk8PPPP3L27BkAfHx8ue++vtSqhAu14wAAIABJREFUVbvCzkMIIYT7kmBPiDKi++N3/EY9DYDdP4CUb5Zirx2Z/aQuzzx6sbGX+PHH70lNTQUgIqI6ffv2x8/Pv9zqLYQQomqRYE+IMuDx1178Bw1AsVpR9XpSF3+N7abmBe5z6NBBVq36FYvFAkDz5i3o0aMXuhyTNYQQQojikmBPiFKmOXGcgEf6o8lIR1UUUj+dh6VLdL7l7XY7GzeuZ8eObQAoisKtt/6HqKgOKIpSXtUWQghRRblN6hW73c78+fPp0aMHLVu25N577+Xnn392KbN//34GDBhAmzZtiI6O5sMPP8RsNldQjcWNSElOIvC//dAkJACQPuVdzPfcn2/5rKwsfvjhW2eg5+npxQMPPEz79h0l0BNCCFEq3KZn7+OPP2b+/PmMGjWKFi1asGHDBsaOHYtGo6FPnz6cPXuWgQMH0rp1az766COOHz/OtGnTSE5O5s0336zo6osbhOofgKnX3XjP/oSM0WMwDn4q37IJCQn8+ON3JGUvjRYSEkrfvv0JCgour+oKIYS4AbhFsJeVlcXixYsZMGAAw4YNA6Bz584cPHiQJUuW0KdPH+bOnYuPjw+zZs1Cr9fTvXt3PD09mTRpEsOHD6dGjRoVfBbihqDRkPHmFCzdb8X8nzvzLXbmzGl+/PF7TCYTAI0bN6F373vQ6/XlVVMhhBA3CLe4jKvX6/n6668ZNGiQy3adTuf8Y7l582a6d+/u8seyV69e2O12Nm/eXK71FTcYux2uGS5gvr0H5HMZ9tSpk/zww7fO393o6G7cd18/CfSEEEKUCbfo2dNqtTRt2hQAVVW5fPkyS5cuZevWrbz55psYjUbOnz9PvXr1XPYLDg7G19eXkydPFnh8jUYp0/FRWq2S494t4usbVknaymvS63js3cX/t3ff4VFV+R/H3zOTSQ8hEYi0UAJBejV0EBTpKkWqDVBEURRUQBeXVRdRWRtFd6UsiwUVpIggGBSRIiAdpAiIEJCekEqSycz9/ZEwP0IgJJhkMpPP63l4SM7cO/dLTmI+nnvPOUkff45RKjjXY48e/Z3Fixdit2dgsZjp0eMe6tSp91fLLpH0c+U+1FfuQ33lmdwi7F1p+fLlPPfccwDccccd3HPPPSQmJgIQGBiY4/iAgACSkpJyfc/Q0IAieRg+ONi/0K8hBSPPffXeezDtXQBCnh8FCxde99DDhw/z7bdL8fY2Yzb70qdPH+rWzX05Frkx/Vy5D/WV+1BfeRa3C3sNGjTgk08+4eDBg7z//vs8+uijvP3227mec6MgFxubXOgje8HB/sTHp2C3Gzc+QVwmP33lvWgBgaMz97F1hIWR8LdXcMQmX/PYI0cOs3jxV9jtGZjNZnr2vI/y5asSe53j5cb0c+U+1FfuQ33lHkJDA/J1vNuFvfDwcMLDw7n99tsJDAxk3LhxHD+eua1UcnLOX5xJSUkEBQXl+p4OhwEU5jd15lC43W5gtzsK8Try1+Wtr6zr1hLwZOZkIUdgEBc/+wp7pXC4xjlHjhxiyZJF2O32rNnj91GzZi19L/xl+rlyH+or96G+8kRucUM+NjaWJUuWcOHChWztderUAeDs2bOEhYVx7NixbK9fuHCB5ORkIiIiiqxW8XyWPbsp9fAgTDYbhtVKwv8+w16/wTWPPXz4/4OexWLh3nt7ExlZq4grFhGRkswtwl5qairjxo1j4VXPQ23YsAGAWrVq0bp1a3788cdsiyivWrUKi8VCixYtirRe8VzmY39k7o6RlPmcaOKMj7C1bX/NYw8d+o2lS7MHvZo1I4uyXBEREfe4jVuhQgX69OnDjBkz8PLyok6dOmzdupWPPvqIvn37UqNGDR599FGWL1/Oo48+ypAhQ/jjjz9455136Nevn9bYk4KRkUHwA/2wnD0DQNI/3yDtvj7XPPTgwQMsW7YEh8OBxWKhV68+VK9eoyirFRERAcBkGIZbPIGZnp7O7NmzWbJkCSdPnqR8+fL069ePYcOGYTZnDlBu3bqVt956i/379xMSEsK9997LqFGjbriR/LlziYVau8ViJjQ0gNjYZD0DUczdqK+8o1dS6tGHuTTscZL/fu2dWQ4c2M833yzF4XDg5eXFfff1oXp1PUpQ0PRz5T7UV+5DfeUeypbNfS7C1dwm7BUmhT25LC99ZTlyCHv1GtdcNHn//n0sX/61M+j16tWXatWqF3bZJZJ+rtyH+sp9qK/cQ37Dnls8syfiMoaB6cyZbE32iJrXDHr79v3qHNGzWq307n2/gp6IiLicwp7IZTYbXut/gkWLMv+22Qh4/VVCO7TCa9eOXE/99de9LF/+NYZhOINe1arVcj1HRESkKLjFBA2RQmWz4T/1HfzmzMR87iwApchcP+/yrNvA0U9zcfVPYM75/0d79uxm5crlGIaBt7c3vXvfT3h4laL8F4iIiFyXwp6UbDYbpR4eiM/q7zCuujVrygp6htWbhP/Mvk7Q28XKlSucQa9Pn35UrhxeJKWLiIjkhW7jSonmP/UdfFZ/B4DpqrlKl6OfyZaO77KlOc7dvXsn336bOaLn4+ND3779FfRERKTYUdiTkstmw2/OzBwjelczTCZ858wEm83ZtnPndlauXAHgDHqVKlUu1HJFRERuhsKelFjWLZswnzubY0TvaibDwHL2DNYtmwDYsWMb3323EsgMevffP4CKFSsVer0iIiI3Q8/sSYlliovL9/Hbt29lddZtX19fX+6/fwDly2uHFhERKb4U9qTEMkJC8nX8lrNnWL1vLwC+vn706zeAW28tXxiliYiIFBjdxpUSyxbVAkfZcnl6Zm99SAjfZe2J6+vrR//+AxX0RETELSjsSclltXJp6GM3fGZvo2Gwok49MJvx8/Onf/9BhIXdWkRFioiI/DUKe1KipYwaQ1qnzgA5RvgMk4n1wLfVI7A1b3lF0AtzQaUiIiI3R8/sSclmtXJp+JNYfjuIOS4WU0KC86Ufg0sTXbcetuYt8Q8Mon//QZQtW9aFxYqIiOSfwp6UeL5ffIbXsT8wrFYSP/mCID8vVhw4zJoLF8BsJiAgkP79B1GmTBlXlyoiIpJvuo0rJZopIR6fbzJ3x0jr1pP0Lt34MTSUtfEXnUFvwIDBCnoiIuK2NLInJZrPkkWYUlMBSB34AOvX/8SOHVsACMy6dXvLLbe4skQREZG/RCN7UqL5zv8YAHuFimwtHcLGjesBCAoqxYABCnoiIuL+CnxkLy0tDZPJhLe3d0G/tUiBshw8gHXbVgAOdulG9A/RAAQEBNC79wBKlSrtyvJEREQKxF8Oe5s3b+b7779n+/btHDlyhNSsW2K+vr5ERETQuHFj7rrrLpo3b/6XixUpSL7zPwEgDvjCzx+Hw4HV6kX//v0JDAzFbne4tkAREZECcFNhz2az8cUXX/Df//6XkydPEhwcTN26denZsyfBwcEYhkFCQgInTpzg66+/5uOPP6ZChQoMHTqU/v37Y7VaC/rfIZI/Nhu+X84nDfi4WnVS/PwA6Ny5G+Hh4cTGJru2PhERkQJyU2Hv7rvvxmazcd9999G1a1fq1q2b6/F79+5l5cqV/Pvf/2bOnDn88MMPN1WsSEHx2v8rRlIiXwGnatUG4Pbbm1O/fgPXFiYiIlLAbirsPf744/Tu3TvPz+XVq1ePevXqMWrUKBYtWnQzlxQpUBkNGrHkP3PY88k8MmrdRvXqEbRv38HVZYmIiBS4mwp7AwYMuKmLeXt73/S5IgVp7949bNqzGxo24pZbytCjx72YzZqcLiIinqdAfrtFRUWxdu3agngrkUJ38uQJVq1aAYCvrx+9e/fF19fXxVWJiIgUjgIJewkJCSRcsafo1X799Vc+/fTTgriUyF+SEH+Rb/42DkdCPGazmXvv7UVISKiryxIRESk0Nx32tm3bxsqVK4mJiQHAZDJd99gjR47wz3/+82YvJVIg0tPTWfr+O2SsWIbfv2fQ1cebKlWqurosERGRQnXT6+xt2rSJadOmYTKZMJlMTJ8+nY0bN1KrVi1q1arFbbfdRunSmYvSnj17Fn9//wIrWiS/DMPg22+/4cJPP2IFokwm6vUbjOHqwkRERArZTYe9kSNH0q1bN/bu3csLL7yA1Wpl06ZNztm2JpOJsmXLcuutt3LgwAEtqiwutWHDOg7u3YP//n1UBzredTdJYWGuLktERKTQ/aUdNKpVq0a1atWYN28ew4cPp1OnTiQlJXHgwAEOHjzIwYMHOXXqFH369OGJJ54oqJpF8mX//n1s3Lger0MHKWNL534gfdBDri5LRESkSBTI3rgLFixwfhwYGEizZs1o1qxZQby1yF9y+vQpvv32GwACf93LQMCnTBmSOnV2bWEiIiJFRAuLicdKSkpk0aKFZGRkYI6Pp/+xPygDpPYdANqyT0RESoibCnvdunVjyZIlpKen5/mc9PR0vvrqK7p163YzlxTJF5vNxuLFX5GUlAhAl8REamS9ljrwAdcVJiIiUsRu6jZur169mDx5MpMmTaJjx460bNmSunXrUqlSJfyyNpRPSUnhxIkT7N27l40bN7JmzRqsVivDhg0r0H+AyNUMw2DlyhWcOvUnAA3rN6DtZx8DYGvcBHvtOq4sT0REpEjdVNh77LHHGDhwIAsXLmTx4sUsXbrUuc6exWIBwG63A5m/eGvWrMnTTz9N3759CQwMLKDSRa5t8+af2b//VwAqVw7nrrs6k2y14vvpx6Tf2cnF1YmIiBQtk2EYf3mpsZiYGHbs2MHRo0e5ePEiAKVLl6Z69eo0atSIypUr/+VCC9O5c4mF+v4Wi5nQ0ABiY5Ox2x2Feq2S7rffDrJkyVcABAcH8+CDQ/K1xqP6yn2or9yH+sp9qK/cQ9myQfk6vkBm41auXLnYBzrxfGfOnGHFimUAeHt707t3Py3mLSIiJV6BzMZ94IEH2LlzZ0G8lchNSU5OZvHiBaSnp2MymejR417Kli2LKf6iq0sTERFxqQIJe8nJyQwcOJCRI0dy5MiRgnhLkTzLyMhg6dJFJCQkANCuXQdq1KgJQHDfeyndsQ0+Cz53ZYkiIiIuUyBhb/HixUyZMoVDhw5xzz338OKLL3Lq1KmCeGuRXBmGQXT0Kk6ciAGgbt36REVlbs1n+XUv1l07sO7djeXo764sU0RExGUKbFHlHj16sGLFCl566SXWrVtH586deeONN4iLiyuoS4jksHXrFvbs2QVAhQoV6dy5q3NmuO/nnziPSx0w2CX1iYiIuFqB7qDh5eXF4MGDiY6O5vHHH2fhwoV06tSJDz74gEuXLhXkpUT4/ffD/PjjDwCUKlWK++7rg5dX1pyj9HR8s27dpre9A0d4FVeVKSIi4lKFsl2an58fI0eOZNmyZURGRjJt2jTuuusuPv30UzIyMgrjklLCnD9/nmXLlmIYBlarlV69sq/h6L3qW8yxsQCkDtSonoiIlFwFsvQKQGJiIvv27XP+2b9/P0ePHnUuruzr68vrr7/OvHnzmDhxIq1atSqoS0sJk5KSwqJFX5KWlgZAt249CQu7Ndsxl2/hOkoFk9b9niKvUUREpLgokLDXsWNH54QMwzAICwujfv369OzZk3r16lGvXj2Cg4OJiYnh7bff5tFHH+XVV1+lb9++BXF5KUHsdjtff73YuXh3mzbtqFXrtmzHmE+fwvv7aADS7usDWVv4iYiIlEQFEvYiIiLo1asX9erVo379+pQpU+aax1WuXJn33nuPv//978yYMUNhT/LFMAy+//47jh8/BkDt2nVo2bJ1juN8vpyPyZG58nvqoAeKtEYREZHipkDC3syZM/N1fFRUFF9++WVBXFpKkB07trFz5w4Abr21PF26dHfOvM3GbMEREoIj7FYyGjct4ipFRESKlwJ7Zi8/2rdvz9SpU11xaXFTf/xxlB9+WA1AYGAQvXr1wWq1XvPYS089w6XHRmA5GQPXCoMiIiIlSKHMxr2RoKAg7r77bldcWtxQbOwFvv56MQ6HAy8vL3r16kNQUKncT/LxwV69RtEUKCIiUoy5JOyJ5JXdbmfp0sWkpqYC0KVLd8qXr+DiqkRERNyHwp4Uaxs3rufcubMAREW1oE6dutc91nvZEgJe/TuWQ78VVXkiIiLFntuEPYfDwfz58+nZsyeNGzfmzjvv5PXXXycpKcl5zLFjxxgxYgTNmjWjefPmTJw4Mdvr4l5OnfqTzZt/BqBcuTDatm2f6/H+H32I//T3CO7VHbLWdxQRESnpXDJB42bMmjWL9957j2HDhtGyZUuOHj3K1KlTOXToEHPmzCExMZGHH36YMmXK8MYbbxAbG8uUKVM4ceIEs2fPdnX5kk82m40VK5bhcDiwWCx069YTi8Vy3eMtRw5hzQqGaf0GQi7HioiIlCRuEfYcDgczZ86kf//+PPfccwC0atWKkJAQRo8ezd69e9m4cSMXL15k0aJFhIaGAhAWFsbw4cPZtm0bTZtqCQ53sm7dWi5cuABA69ZtKVeuXK7H+37+mfPj1IFaW09EROQyt7iNm5SUxL333kuPHj2ytVevXh2AmJgY1q9fT9OmTZ1BD6BNmzYEBATw008/FWm98tfExBxn27ZfAKhQoSJRUS1yP8Fux+eLzLBnaxaFvWZkYZcoIiLiNtxiZK9UqVJMmDAhR/vq1ZnrrtWoUYMjR47QrVu3bK9bLBYqVarE0aNHc31/s9l07cV5C4jFYrrib7fI1y6Tnp7OqlXLMZnAy8tKjx49sVpz/za1rlmN5XTmdn3pDzyExXLzX2P1lftQX7kP9ZX7UF95JrcIe9eya9cuPvroIzp06EBkZCSJiYkEBATkOC4gIOCGkzRCQwMKNexdFhzsX+jXcHfLl/9IWloKfn7edOnShZo1q9z4pAVZt3D9/AgY8iABpXJ+H+SX+sp9qK/ch/rKfaivPItbhr1t27YxYsQIKlWqxOTJk4HMfVOv50ZBLjY2udBH9oKD/YmPT8Fuv36dJd3Ro7/z008bAKhcuQo1a9YjNjY513NMF85TeulSTEDaPb1IzrDADc7JjfrKfaiv3If6yn2or9xDaGj+BjXcLuytWLGC8ePHU7VqVWbNmkVISAgAgYGBJCfn/CWflJREWFhYru/pcBhAYX5TZw6F2+0GdrujEK/jvlJTU1m+/BscDgNvb286d+6ap37xXbEck80GwKWBDxTA11d95T7UV+5DfeU+1FeeyK1uyM+ePZsxY8bQqFEjPv3002wzNKtVq8bx48ezHW+32zlx4gQRERFFXark0w8/rCYxMQGAO+7oSOnSIXk6L3XgA8QtjyZl5DPYWrQqzBJFRETcktuEvc8//5y33nqLrl27MmvWLIKCgrK93rp1a3755RdiY2OdbevXryclJYXWrVsXdbmSD4cPH2Lv3t0AVK1ajYYNG+f9ZJOJjNubkzzxNTC7zbeziIhIkXGL27jnzp1j8uTJVKxYkcGDB7Nv375sr4eHhzNo0CA++eQThgwZwlNPPcXFixeZMmUK7dq1o0mTJi6qXG4kJSWFVau+BcDX15euXbsXyWQZERGRksItwt7atWtJTU3l5MmTDB48OMfrkydPpnfv3sybN4/XX3+d559/noCAALp06cLYsWNdULHk1ffff0dycuZs6Y4dOxEUVCpvJ6amYt2yCVubdhrRExERyYVbhL2+ffvSt2/fGx4XGRnJ3LlzC78gKRAHDuxn//7MUdqaNSOpW7dens/1WbmcUsOHYK8cTvynC7DfVruwyhQREXFrGhIRl0hKSiI6ehUAfn7+dOrUJV+3b33nfwKAKTEBe9VqhVKjiIiIJ1DYkyJnGAbfffctly6lANCpU2cCAwPzfL755AmsP/4AQFqffuDrWyh1ioiIeAKFPSlye/fu4fDhQwDUrl2H2/J5C9b3y/mYshbRTh30YIHXJyIi4kkU9qRIJSTEs2ZN5p7GAQGB3Hnn3fl7A4cD388+BiCjbn0y6jcs6BJFREQ8isKeFBnDMFi5cgWpqakAdOnSFX///O2/aN20EcuxPwBIHfRAQZcoIiLicRT2pMjs2rWDP/44CkD9+g2JiKiZ7/e4PDHD8PYmtU+/Aq1PRETEEynsSZG4eDGOH7MmVZQqVYoOHe7M93uYEhPwWbYEgLQu3TFCbynQGkVERDyRwp4UOsMw+Pbb5aSnpwPQpUt3fG9mBq3DQcrIZ7BXDidtYM7FtUVERCQnt1hUWdzb1q1biIk5DkDjxk2omp918Ww2rFs2YYqLwwgJIeXZ50l5blwhVSoiIuJ5FPakUF24cIF169YCEBISQvv2HfN2os2G/9R38JszE/O5s85me7lypA55jJRRY7RNmoiISB4o7EmhcTgcrFixjIyMDEwmE1279sDb2/vGJ9pslHp4ID6rv8O4alcN87lzBLw5Ca/tW0mY+xlYrYVUvYiIiGfQ0IgUmi1bNnHq1J8ANGsWRaVKlfN0nv/Ud/BZ/R2Ac/Hkyy5/7hO9Cv9p7xZgtSIiIp5JYU8KxdmzZ9mwYR0At9xShrZt2+ftRJsNvzkzc4zoXc0wmfCdMxNstr9aqoiIiEdT2JMCZ7fbWbFiGXa7HbPZTLduPfDyytsTA9YtmzCfO5tjRO9qJsPAcvYM1i2bCqJkERERj6Vn9uTmXDVL1hbVwvn83MaN6zl79gwAzZu3pHz5Cnl7z4wMgp58NF9lmOLi8nW8iIhISaOwJ/lzg1myR+4fwObNPwNQrlwYrVq1yX5+RgZee3dj3bQR6+ZNJL7xNkZYWOZrXl4Y/gH5KscICflL/xwRERFPp7AneXeDWbLeb05izdJFOLr1xGK10q1bTyypqVi3b8W6+Wesm37GunULppRk53mpvfqQfk8v5+eXHh9JwCsTMCUnk9tTe4bJhKNsucwRRREREbkuhT3JsxvNkl0DxB3Yj/WWMrQc9zdqjhqB9acfMWVkXPP97GG3YkpOztaW+sgwzBfOE/DmpFxrMRkGqUMf09IrIiIiN6CwJ3lzxSzZa02eOA78DBhAld07iWrcFMNiyRb0MmpGYmve0vnHUaUqXGPWbcqoMXht34pP9Koc17v8eVqnzqQ8Pbrg/50iIiIeRmFP8uTyLNlrSQcWkxn0rECfxES8tm4h7b4+2CNqOsOdUaZMHi9mJWHuZ/hPexffOTOxZE32AHCULUfq0Mcyg55G9URERG5IYU/yJLdZr9HA5VfvBMoA8XFxpN0/gLT7B9zcBa1WUsaMJeXp0ded9SsiIiI3prAneXK9Wa+/A79kfVwFaHGD4/PNasXWum3BvJeIiEgJpEWVJU9sUS1wlC3HlU/rpQJLsz72Bu4DMJmwlwvTLFkREZFiQmFP8sZq5dKgB7Mth7IKiM/6+G4gBM2SFRERKW4U9iTPUh8agiPr9uwhYEdWewTQJOtjzZIVEREpXhT2JM8clcO5sPs3Uu7txbeBgQD4APcARrkwksdPIGHuZxrVExERKUY0QUPyx8eHDSOfISaiJpaTJ2gVUQPj9ubEapasiIhIsaSRPcmdYeCzdBHY7QCkpaWxfv06MJsJrFef+k89mzlbVkFPRESkWFLYk1z5fjyXUo89QnCv7pj/PMnmzT+TkrW3bbt2HfDy0uCwiIhIcaawJ9dlPvo7gX9/CQDLsT+It2ewdesWAMqXr8Btt9V2ZXkiIiKSBwp7cm12O6WeehxT1ihe4vsf8NPu3WRk7XXbocOdmK6xr62IiIgULwp7ck1+M97H+stmAC4NG86J2nXYt28vAJGRtahUqbIryxMREZE8UtiTHCx7dhPw5iQAMmrUJHHCK6xZ8z2GYWA2m2nX7g7XFigiIiJ5prAn2aWmUuqp4ZhsNgyLhcTp/+H3039y/PgxABo3bkJo6C0uLlJERETySmFPsgl4cxJe+/cBkPLs86Q3asKPP/4AgK+vLy1btnFleSIiIpJPWjdDsrE1boIjJAR7laqkjBnL7t07uXDhAgAtWrTG39/fxRWKiIhIfijsSTbp9/QiLqoFXLpEmsORuYAyEBwcTJMmTV1cnYiIiOSXwp7k4Li1PABb1q3VAsoiIiJuTs/sCda1a/DasS1bW0JCPL9kLb2iBZRFRETcl8JeCWc6f55STzxK6W534TdjqrN93bqfnAso33FHRy2gLCIi4qZ0X64kMwyCnn8G8/lzADjKlQPgzJnT2RZQrlw53GUlioiIyF+jkb0SzOeLz/BZsQyAtJ73kda3P4ZhaAFlERERD6KwV0KZY44T+NJYAOzlwkh8610wmfj998NaQFlERMSDKOyVRA4HQaOewJyUCEDSu9MwbrkFh8PBjz+uAbSAsoiIiKdQ2CuB/D76AO8NmevnXXpwCOmdugBkLaB8HoDmzVtpAWUREREPoLBXwlgOHiBg0isA2KtUJemVSQCkpaVlW0C5adNmLqtRRERECo7CXgnjqFiR1H4DMcxmEqZ/BIGBAGzZskkLKIuIiHgghb0SxggMIuntqcRt+IWM5i0ASExM0ALKIiIiHkphr4SyR9R0fqwFlEVERDyXwl5JkJxMwKRXMGXNvr3SmTOn+fXXPYAWUBYREfFECnslQOCrL+P//tuE3NEK84kYZ7sWUBYREfF8bhn2Tp8+TbNmzdi8eXO29mPHjjFixAiaNWtG8+bNmThxIklJSS6qsniw/rAav//OAsBeoSKO8hWcr2kBZREREc/ndlMuT506xbBhw0hMzH5LMiEhgYcffpgyZcrwxhtvEBsby5QpUzhx4gSzZ892UbWuZYqLJejZkQA4AgJJnPZvsFgyP9cCyiIiIiWC24Q9h8PBkiVLePPNN6/5+vz587l48SKLFi0iNDQUgLCwMIYPH862bdto2rRpUZbrGjYb1i2bMMXFYYSE4Dt3FpbTpwBI/ucbOKpUdR6qBZRFRERKBrcJewcPHmTixIkMGjSIVq1aMXz48Gyvr1+/nqZNmzqDHkCbNm0ICAjgp59+8uywZ7PhP/VQrlvcAAAepUlEQVQd/ObMxHzubI6X0zp1JnXQg///uRZQFhERKTHcJuyVL1+e6Ohobr311hzP6gEcOXKEbt26ZWuzWCxUqlSJo0eP5vreZrOpUJcbsVhMV/xdwI9J2mwEPjII7+hVGNf5N5jS07EYDvCyArB162ZSU1Mwm03ccUcHfHy8C7YmN1aofSUFSn3lPtRX7kN95ZncJuyVLl0619cTExMJCAjI0R4QEHDDSRqhoQFFsrZccHAh3Cp97TWIXgWAyTCueYj32jWEzpwOEyaQkJDA3r078PPzpmLFirRuHaV19a6hUPpKCoX6yn2or9yH+sqzuE3YuxHjOkEHuGGYiY1NLvSRveBgf+LjU7Dbr19nvtlslJ42HZPJdN2gB2CYTBjTpnPxsadY/t1KEhNTAIiKakNcXErB1eMBCq2vpMCpr9yH+sp9qK/cQ2hozsGt3HhM2AsMDCQ5OTlHe1JSEmFhYbme63AYQGF+U2cOhdvtBna7o8De1frzxms+o3c1k2FgOnuGs8u/Yc+BfRiGQWRkLSpUqFSg9XiGwukrKQzqK/ehvnIf6itP5DE35KtVq8bx48eztdntdk6cOEFERISLqipcpri4PB9rAGs2rtMCyiIiIiWMx4S91q1b88svvxAbG+tsW79+PSkpKbRu3dqFlRUeIyQkz8ceBo5lPbuoBZRFRERKDo8Je4MGDcLHx4chQ4YQHR3NggULeOGFF2jXrh1NmjRxdXmFwhbVAkfZctedhXuZHVgZGIS9YiUtoCwiIlLCeEzYCw0NZd68eYSEhPD888/z7rvv0qVLF959911Xl1Z4rFYuDX0s18kZADuAUw0bgdmsBZRFRERKGLecoNG8eXMOHjyYoz0yMpK5c+cWfUEulDJqDF7bfsFn9Xc5XjNMJtINg+iakdiat9QCyiIiIiWQW4Y9uYLVyqVHHr1m2HOULcfqjncRW6EimM20bXsHXl7qchERkZJEv/k9gO/ihQAYPj4kzP4Y0tMxQkKIrV2HDf+bAzYb5ctXoHbtOi6uVERERIqawp6bMyUl4rNiGQBpXbuTfncX52vrVnyDzWYD4I47OmqnDBERkRLIYyZolFTe33yN6dIlANL6DXS2nzlzhl9/3QNAzZqRVK4c7pL6RERExLUU9tyc74LPAXCUKUv6HXcCmVvHrVmz2rmAcvv2HVxZooiIiLiQwp47cziwR9bCERJCap9+kDX54ujRIxw/fgzQAsoiIiIlnZ7Zc2dmM0mT/0XSK69jupQCgMPhYM2aHwDw8fHRAsoiIiIlnMKeJ/D2xvD2BmDPnl1cuHAegBYtWmsBZRERkRJOt3E9SHp6OuvXrwPQAsoiIiICaGTPbfm/OwVTUhKp/QZir3UbkDmql5ycBKAFlEVERARQ2HNPNht+Mz/EfP48Xrt2Er9wKQ6Hg61btwAQEhLCbbfVdnGRIiIiUhzoNq4b8l6zGvP5zOfyUvsNAODgwQPEx8cD0KxZFGazulZEREQU9tySz5eZa+sZ/v6kdeuJYRj88stmAPz8/KlXr4EryxMREZFiRGHPzZguxuGzagUAad3vgcBAYmKOc/r0KQCaNGmK1Wp1ZYkiIiJSjCjsuRmfr5dgSksDIDVre7TLo3peXl40atTEZbWJiIhI8aOw52Yub49mv7U8tjbtOH/+PEeOHAagXr36BAQEuLI8ERERKWYU9tyI+ejvWDf/DEBa3/5gsThH9UwmE82aRbmyPBERESmGFPbciM/KFc6PU+8fQFJSIvv27QWgZs1I7YErIiIiOWidPTdyacRIbE1vx3vdj9hr12H7Tz9it9sBuP325i6uTkRERIojhT13YjKREdWcjKjmpKens3PnDgAqVqxExYqVXFyciIiIFEe6jeum9uzZRWrqJUCjeiIiInJ9CnvuIC0N64Z14HAA5NgarUaNmq6sTkRERIoxhT034B29itK9uhParD6Wfb9m2xrt9tuba2s0ERERuS6lBDfg++V8AEzx8WRUrZZta7S6deu7sjQREREp5hT2ijnThQt4f/8dAGn33EfM+XPaGk1ERETyTGGvmPNZ8hUmmw2AtPsHaGs0ERERyReFvWLOd0HmLVx75XBO1YjU1mgiIiKSLwp7xZjl8CGs27cBkHp/f37Z9gugrdFEREQk7xT2ijGfrFE9gPPdemprNBEREck3hb3iyuHAd8EXANiaNmPrxYvaGk1ERETyTWGvuEpLI7XfQOzhVUjodT87d24HtDWaiIiI5I/CXnHl50fK+AnEbtnFlgYNSU1NBSAqqoWLCxMRERF3orBXzDmAbbt2ABAaGkpERA3XFiQiIiJuRWGvmLtya7RmzaK0NZqIiIjki5erC5CcgkYMwwgI4NKAwWzZ9yugrdFERETk5ijsFTOmM2fwWboIk93OsZQUzlSrDmhrNBEREbk5uidYzPguXoApa4mVn7KCnpeXF40bN3VlWSIiIuKmFPaKGZ+stfX+DK/Coay2+vUb4O/v77qiRERExG0p7BUjln2/Yt2zC4B1TW8HMrdGa5r1sYiIiEh+KewVI74LPgcgEdhZrhygrdFERETkr1HYKy7sdny++hKAjbfVISOoFKCt0UREROSvUdgrJqzr1mI5fYo0YFP1CAAqVaqsrdFERETkL1HYKyasmzYCsN1qJTkiM+xpVE9ERET+KoW9YiJl/ATOr9vC2nt7Y/j4EhoaSo0aNV1dloiIiLg5hb1iZJ/DwYWq1YDMrdFMJpOLKxIRERF3p7BXTBiGwZYtmwDw9w/Q1mgiIiJSILRdmouZz5zGa8tmDt1WmzNnTgPaGk1EREQKjkb2XMzni/kED3uQ/Z3aYUpMxGq10qhRE1eXJSIiIh5CYc+VDAPfBfM5Cxzy9cMICqJevfraGk1EREQKjMKeC3nt2YXXwQP8DGTUrYfJZKJZsyhXlyUiIiIexCPD3vr16+nTpw8NGzakY8eOzJ49G8MwXF1WDj5fzicR2AVk1KlLzZqRhISEurosERER8SAeF/Z27tzJiBEjqF69OtOmTaNnz55MmTKFmTNnurq07Gw2fBctZDNgC6+CEVSKqKgWrq5KREREPIzHzcadNm0atWvXZsqUKQC0a9eOjIwM/v3vf/PQQw/h6+vr4gozef/4Pbbz59gKZNSpR6VKlalQoaKryxIREREP41Eje+np6WzevJlOnTpla+/cuTPJycls27bNRZXl5PPl5+wALnl5YY+spa3RREREpFB41MheTEwMNpuNqlWrZmuvUqUKAEePHqV169Y5zjObTYW6W4XFYrribzOm+ItYv/2GnwF7zVqEli9PrVq1tGNGMXB1X0nxpb5yH+or96G+8kweFfYSExMBCAwMzNYeEBAAQFJS0jXPCw0NKJKgFRyctaSK1cGe0aOJnzcPryaN6NSpA7fcEpj7yVKknH0lxZ76yn2or9yH+sqzeFTYczgcub5uNl/7/1JiY5MLfWQvONif+PgU7HYDwzARXa4Clx58BH8/fypViiA2NrnQri95d3VfSfGlvnIf6iv3ob5yD6GhAfk63qPCXlBQEADJydmD0+URvatH/C5zOAygML+pM0Om3W5gtzs4duwPTp06BUCjxk0xmy3Y7bkHVSkq2ftKijP1lftQX7kP9ZUn8qgb8uHh4VgsFo4dO5at/fjx4wBERES4oqzsMjL45ZfNANoaTURERAqdR4U9Hx8fmjVrRnR0dLZFlFetWkVQUBANGjRwYXWAYWBr34KT707BEnNcW6OJiIhIofOo27gATzzxBEOGDOGZZ56hT58+7Nixg9mzZ/Pcc8/h5+fn0tos27ay5dBveAFGxcraGk1EREQKnUeN7AG0bNmSadOmcfToUUaOHMmyZcsYO3Ysjz32mKtLI/2Tuewh8+nA6t17ams0ERERKXQeN7IH0KlTpxwLK7uMzYbXz5shNZEdC77EDjiqVqNZp86urkxERERKAI8Me8WCzYb/1HfwmzMT87mzpAHbs16qFBxMhbLlXFmdiIiIlBAedxu3WLDZKPXwQALenITp/DkAdgCpZN7Cbb9rJ6UeGQQ2myurFBERkRJAYa8Q+E99B5/V3wFgMgzswM9Zr5UBagE+0avwn/auiyoUERGRkkJhr6DZbPjNmYlxxY4c+4D4rI9bASbAMJnwnTNTo3siIiJSqBT2Cph1yybM585iumKdv8vP6gUADbM+NhkGlrNnsG7ZVNQlioiISAmisFfATHFxOdqCsv7uQM4ZMdc6XkRERKSgaDZuATNCQnK03Qt04v9D342OFxERESkoGtkrYLaoFjjKlsv2zJ6FnEHPMJmwlwvDFtWiSOsTERGRkkVhr6BZrVwa+li2Z/auxWQYpA59DKzWIipMRERESiKFvUKQMmoMaVk7ZFw5wnfl52mdOpPy9Ogir01ERERKFoW9wmC1kjD3M5LHT8Bx1U4ZjrLlSB4/gYS5n2lUT0RERAqdJmgUFquVlDFjSXl6ND5bN1Mq4xIJXn6kNWuukCciIiJFRmGvsFmtZLRpB6EBZMQmg93h6opERESkBNFtXBEREREPprAnIiIi4sEU9kREREQ8mMKeiIiIiAdT2BMRERHxYAp7IiIiIh5MYU9ERETEgynsiYiIiHgwhT0RERERD2YyDMNwdREiIiIiUjg0siciIiLiwRT2RERERDyYwp6IiIiIB1PYExEREfFgCnsiIiIiHkxhr5CtX7+ePn360LBhQzp27Mjs2bPRBOjix+FwMH/+fHr27Enjxo258847ef3110lKSnJ1aXIDTz31FB07dnR1GXIdO3fu5MEHH6RRo0a0atWKcePGceHCBVeXJdfw5Zdf0r17dxo1akTXrl359NNP9fvKQyjsFaKdO3cyYsQIqlevzrRp0+jZsydTpkxh5syZri5NrjJr1ixee+017rjjDmbMmMHQoUNZunQpTz/9tP5jV4wtXbqU6OhoV5ch17F3714eeughAgICmD59Os8//zwbNmxg5MiRri5NrrJgwQJefvllWrZsyYcffki3bt147bXX+O9//+vq0qQAaJ29QjRs2DASEhJYsGCBs23KlCnMnz+fjRs34uvr68Lq5DKHw0Hz5s3p0aMHEydOdLavWLGC0aNHs3DhQurXr+/CCuVazpw5Q8+ePfHz88NisfDDDz+4uiS5ysMPP0xaWhqfffYZZnPm2MJ3333HpEmT+OSTT6hcubKLK5TLBgwYgNls5rPPPnO2jRkzhp07d+pnywNoZK+QpKens3nzZjp16pStvXPnziQnJ7Nt2zYXVSZXS0pK4t5776VHjx7Z2qtXrw5ATEyMK8qSG5gwYQKtW7emZcuWri5FriEuLo4tW7YwcOBAZ9ADuPvuu1m7dq2CXjGTlpZGYGBgtrbSpUtz8eJFF1UkBUlhr5DExMRgs9moWrVqtvYqVaoAcPToURdUJddSqlQpJkyYQNOmTbO1r169GoAaNWq4oizJxYIFC/j11195+eWXXV2KXMfBgwdxOByEhoby3HPP0bhxYxo3bszYsWNJSEhwdXlylYceeoj169ezdOlSEhMTWbduHYsXL+bee+91dWlSALxcXYCnSkxMBMjxf0oBAQEAevC/mNu1axcfffQRHTp0IDIy0tXlyBVOnjzJ5MmTmTx5MqGhoa4uR64jNjYWgJdeeol27drxwQcf8Mcff/DOO+8QExPDZ599hslkcnGVcln37t3ZsmULY8eOdba1adOGl156yYVVSUFR2CskDocj19evvK0hxcu2bdsYMWIElSpVYvLkya4uR65gGAYvvfQS7du3p3Pnzq4uR3Jhs9kAqFu3LpMmTQKgZcuWlCpVijFjxrBhwwbatGnjyhLlCk8++STbtm3jhRdeoEGDBvz2229MmzaNZ555hhkzZiiYuzmFvUISFBQEQHJycrb2yyN6V4/4SfGwYsUKxo8fT9WqVZk1axYhISGuLkmu8Omnn3Lw4EGWLVtGRkYGgHO2dEZGBmazWf8jVUxcvovRoUOHbO1t27YFYN++fQp7xcT27dtZt24d//znP7n//vsBiIqKonLlygwfPpwff/wxRz+Ke1HYKyTh4eFYLBaOHTuWrf348eMAREREuKIsycXs2bOZMmUKUVFRzJgxwxnYpfhYtWoVcXFx1wwJdevW5amnnuLpp592QWVytcvPK6enp2drvxzStRpB8fHnn38C0KRJk2ztzZo1A+DQoUMKe25OYa+Q+Pj40KxZM6Kjoxk2bJhzCHzVqlUEBQXRoEEDF1coV/r8889566236NatG2+++Sbe3t6uLkmu4ZVXXskxWj5jxgz27t3Lhx9+SLly5VxUmVwtIiKCihUrsnz5ch544AHnfwO///574P+DhLje5ZUHtm7dmm0gYvv27QCaOe0BFPYK0RNPPMGQIUN45pln6NOnDzt27GD27Nk899xz+Pn5ubo8yXLu3DkmT55MxYoVGTx4MPv27cv2enh4uCYCFBOXfyldqXTp0nh7e2stxGLGZDIxduxYnn32WUaPHk2/fv04fPgw7777Lp07d6ZOnTquLlGy1KlTh86dO/PGG28QHx9Pw4YNOXz4MNOmTaNu3bo5lhAT96NFlQtZdHQ0U6dO5ejRo4SFhTF48GCGDh3q6rLkCgsXLuRvf/vbdV+fPHkyvXv3LsKKJD/Gjx/Pli1btPBrMbVmzRpmzJjBwYMHCQ4OpmfPnowePVqj58VMeno6H374IUuXLuXs2bNUqFCBu+66i5EjRzqfvxT3pbAnIiIi4sE0bU1ERETEgynsiYiIiHgwhT0RERERD6awJyIiIuLBFPZEREREPJjCnoiIiIgHU9gTERER8WAKeyIiIiIeTGFPRERExIMp7IlIkRs/fjy1atXK9c+DDz7IokWLqFWrFidOnHBpvatWrWLQoEHZ2hYtWsSAAQNo0qQJDRs2pHv37rz//vskJSW5qMqbN378eDp27Jjn4xcuXMjw4cMLsSIRKUjaLk1Eitzx48eJjY11fv7BBx+wb98+pk+f7mwLDAwkNDSU48ePU6dOHZftpXrhwgV69uzJzJkzqVu3LgDTp0/n3//+N0OHDqVp06ZYrVb27t3LrFmzCA8PZ/78+VitVpfUezPyu7+wYRj07t2bwYMH07dv30KuTkT+Ki9XFyAiJU94eDjh4eHOz0NDQ/H29qZRo0Y5jg0NDS3K0nL48MMPadCggTPopaenM3PmTIYNG8bo0aOdx7Vq1Yrq1aszcuRIVq9eTdeuXV1VcqEzmUw8/vjjvPrqq/To0QNfX19XlyQiudBtXBEptq6+jTt+/HiGDRvGF198wV133UWDBg0YMGAAR48eZc2aNfTs2ZOGDRty//33s3///mzvtXXrVh544AEaNmxIVFQU48aNyza6eC2xsbEsXLiQHj16ONuSkpJITU3F4XDkOL59+/aMHj2aypUrO9vS0tJ46623aN++PfXq1aNnz56sWLEi23mGYTB37ly6du1KgwYN6NSpE7Nnz+bKGy8bNmxg0KBBNG3alObNm/Pcc89x6tSpbF+rOnXqsGvXLvr370/9+vXp0KEDs2fPznat+Ph4XnzxRaKiorj99tuZMmVKjn/L8ePHGTFiBM2bN6dhw4b079+ftWvXZjumQ4cOpKWl8dVXX+X6NRQR19PInoi4lR07dnD27FnGjx9PWloa//jHPxg+fDgmk4lRo0bh5+fHxIkTef7551m+fDkAv/zyC0OGDKFFixa89957xMfH8/777/PQQw+xcOHC645Mfffdd2RkZNChQwdnW2hoKA0bNmT27NmcPXuWTp060aRJE0JDQ7FarYwYMcJ5rGEYjBw5ku3btzNq1CgiIiKIjo5m9OjRpKenc9999wHw1ltv8b///Y8hQ4bQunVr9uzZw7/+9S8yMjJ4/PHHWbJkCePGjaNHjx48/vjjxMXFMXXqVPr378/ixYu55ZZbAHA4HDz77LM88sgjPPvssyxcuJC33nqLyMhI2rZti8Ph4NFHH+XkyZOMGzeO0qVLM2vWLPbs2UO5cuWc7/H4449Trlw53nrrLby8vJg3bx5PPPEE3377LVWqVAHAx8eHDh06sGzZMgYPHlzwHS0iBccQEXGxcePGGR06dMjR/tVXXxmRkZFGTEyM87jIyEjj8OHDzmP+/ve/G5GRkcbGjRudbbNnzzYiIyON+Ph4wzAMo3///kaPHj2MjIwM5zG///67Ubt2beOTTz65bl3PPPOMcc899+RoP3XqlPHggw8akZGRRmRkpFGrVi2jR48exvvvv29cvHjRedz69euNyMhIY/ny5dnOf/75543WrVsbNpvNiI+PN+rUqWNMmjQp2zGvvfaaMWzYMMNutxutW7c2hg4dmu31Y8eOGXXr1jXefPPNbF+rL7/80nlMWlqaUb9+fePVV181DMMw1qxZY0RGRhpr1651HpOcnGw0b97c+fU/e/asERkZaXz99dfOYxISEozXX3/d+O2337LVMHfuXKN27dpGYmLidb+GIuJ6uo0rIm4lODiYiIgI5+dlypQBoGHDhs620qVLA5CQkMClS5fYtWsX7du3xzAMMjIyyMjIoHLlykRERLBhw4brXismJoZKlSrlaL/11luZN28ey5cvZ9y4cbRv356TJ08yY8YMunfvzh9//AHAzz//jMlkon379s7rZmRk0LFjR86dO8ehQ4fYuXMnGRkZ3H333dmuMWHCBGbNmsXRo0c5d+5ctlvJkPncY+PGjdmyZUu29saNGzs/9vb2JjQ0lJSUFCDzVrbVaqVt27bOY/z9/Wnfvn22r2eNGjV4+eWXGTduHMuWLcPhcPDiiy9Ss2bNbNeqWLEidrud06dPX/drKCKup9u4IuJWAgMDr9nu7+9/zfaEhAQcDgczZ85k5syZOV738fG57rWSkpLw8/O77us1atSgRo0aDB06FJvNxqJFi3j11Vd55513mDp1KhcvXsQwDJo0aXLN88+ePUt8fDxw/YkoFy9eBP4/1F6pTJky7Nu3L1vb1bekzWaz89m/+Ph4SpcujclkynZM2bJlnR+bTCbmzJnDhx9+SHR0NEuWLMFqtXLXXXfxyiuvEBwc7Dz28tc8MTHxmrWLSPGgsCciHi0gIACTycQjjzxC9+7dc7yeW5gLCQnJEWT+97//8eGHH7JmzZps51qtVudEhsOHDwMQFBSEv78/8+bNu+b7V6lShe3btwOZk0GqV6/ufO3PP//k+PHjhISEAHD+/Pkc5587d875el6EhIQQFxeH3W7HYrE42y8HysvCwsL4xz/+wcSJEzlw4AArV65k5syZhISEMHHiROdxl4NqfmoQkaKn27gi4tECAwOpU6cOv//+O/Xr13f+qVmzJtOmTWPz5s3XPbdChQrZZrxC5mheXFwcH3/8cY7j7XY7MTExREZGAhAVFUVKSgqGYWS79m+//caMGTPIyMigQYMGWK1W1qxZk+295syZw5gxY6hZsyZly5blm2++yfZ6TEwMO3fuvO6o4bW0bNmSjIwMVq9e7WxLT0/Pdit7x44dtGrVit27d2MymahduzajR48mMjKSP//8M9v7nTlzBovFQlhYWJ5rEJGip5E9EfF4Y8aMYfjw4Tz33HPcc8892O125syZw65du3jyySeve17r1q359ttvSUxMJCgoyNnWo0cP3nnnHQ4ePEjnzp0JDQ3l9OnTfP7555w+fZr33nsPyFyK5fbbb+fJJ5/kySefJCIigt27dzN16lTatm3rvHX70EMPMXfuXLy9vYmKimLXrl3Mnz+fsWPHYjabGTNmDC+++KKz/ri4OKZPn05wcDBDhgzJ89ehZcuWtGnThgkTJnDhwgUqVqzIvHnziI2Ndc7orVOnDr6+vowdO5ann36aMmXKsHHjRvbv389DDz2U7f22bdtGs2bNch0dFRHXU9gTEY/Xpk0bZs+ezfTp0xk1ahRWq5W6devy3//+95oLOV/WoUMHvLy8WLduHd26dXO2T5kyhaioKL7++msmTJhASkoKoaGhtG7dmsmTJzvX2TObzXz00Ue8//77/Oc//+HChQuEhYUxZMgQRo4c6Xy/F154gVtuuYXPP/+cWbNmUalSJV5++WUGDBgAQO/evQkICOA///kPI0eOJDAwkLZt2zJmzJhsz9vlxfTp0/nXv/7F1KlTSUtLo1u3bvTr14/vv/8eyHyGcc6cObz99ttMmjSJhIQEqlatyquvvkrv3r2d75OWlsbmzZt59tln83V9ESl62i5NRCQXr732GocOHbruc3cl1ZIlS/jXv/7F6tWrtYOGSDGnZ/ZERHIxYsQIDhw4wO7du11dSrHhcDiYM2cOTz31lIKeiBtQ2BMRyUXZsmX5xz/+weuvv+7qUoqNr776irJlyzpvM4tI8abbuCIiIiIeTCN7IiIiIh5MYU9ERETEgynsiYiIiHgwhT0RERERD6awJyIiIuLBFPZEREREPJjCnoiIiIgHU9gTERER8WAKeyIiIiIeTGFPRERExIP9H2REsx9QUs5tAAAAAElFTkSuQmCC\n", "text/plain": [ "
" ] }, - "metadata": { - "needs_background": "light" - }, + "metadata": {}, "output_type": "display_data" } ], @@ -92,17 +90,17 @@ "np.random.seed(20394)\n", "\n", "def freefall(y, t, p): \n", - " return 2.0*p[1] - p[0]*y[0]\n", + " return 2.0 * p[1] - p[0] * y[0]\n", "\n", "# Times for observation\n", "times = np.arange(0,10,0.5)\n", - "gamma,g, y0, sigma = 0.4, 9.8, -2, 2\n", - "y = odeint(freefall, t=times, y0=y0, args=tuple([[gamma,g]]))\n", + "gamma, g, y0, sigma = 0.4, 9.8, -2, 2\n", + "y = odeint(freefall, t=times, y0=y0, args=tuple([[gamma, g]]))\n", "yobs = np.random.normal(y,2)\n", "\n", "fig, ax = plt.subplots(dpi=120)\n", - "plt.plot(times,yobs, label='observed speed', linestyle='dashed', marker='o', color='red')\n", - "plt.plot(times,y, label='True speed', color='k', alpha=0.5)\n", + "plt.plot(times, yobs, label='observed speed', linestyle='dashed', marker='o', color='red')\n", + "plt.plot(times, y, label='True speed', color='k', alpha=0.5)\n", "plt.legend()\n", "plt.xlabel('Time (Seconds)')\n", "plt.ylabel(r'$y(t)$');\n", @@ -139,13 +137,51 @@ "name": "stderr", "output_type": "stream", "text": [ - "Auto-assigning NUTS sampler...\n", - "Initializing NUTS using jitter+adapt_diag...\n", - "Sequential sampling (2 chains in 1 job)\n", - "NUTS: [gamma, sigma]\n", - "Sampling chain 0, 0 divergences: 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 3000/3000 [02:08<00:00, 23.28it/s]\n", - "Sampling chain 1, 0 divergences: 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 3000/3000 [01:57<00:00, 25.63it/s]\n", - "100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 4000/4000 [00:42<00:00, 93.03it/s]\n" + "INFO (theano.gof.compilelock): Waiting for existing lock by process '79996' (I am process '80216')\n", + "INFO (theano.gof.compilelock): To manually release the lock, delete /Users/twiecki/.theano/compiledir_macOS-10.15.3-x86_64-i386-64bit-i386-3.8.1-64/lock_dir\n", + "INFO (theano.gof.compilelock): Waiting for existing lock by process '79996' (I am process '80216')\n", + "INFO (theano.gof.compilelock): To manually release the lock, delete /Users/twiecki/.theano/compiledir_macOS-10.15.3-x86_64-i386-64bit-i386-3.8.1-64/lock_dir\n" + ] + }, + { + "ename": "KeyboardInterrupt", + "evalue": "", + "output_type": "error", + "traceback": [ + "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", + "\u001b[0;31mKeyboardInterrupt\u001b[0m Traceback (most recent call last)", + "\u001b[0;32m\u001b[0m in \u001b[0;36m\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m ode_model = DifferentialEquation(\n\u001b[0m\u001b[1;32m 2\u001b[0m \u001b[0mfunc\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mfreefall\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 3\u001b[0m \u001b[0mtimes\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mtimes\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 4\u001b[0m \u001b[0mn_states\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 5\u001b[0m \u001b[0mn_theta\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m2\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/working/projects/pymc/pymc3/ode/ode.py\u001b[0m in \u001b[0;36m__init__\u001b[0;34m(self, func, times, n_states, n_theta, t0)\u001b[0m\n\u001b[1;32m 86\u001b[0m \u001b[0;31m# Private\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 87\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_augmented_times\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0minsert\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtimes\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mt0\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mastype\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfloatX\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 88\u001b[0;31m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_augmented_func\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mutils\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0maugment_system\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfunc\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mn_states\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mn_theta\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 89\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_sens_ic\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mutils\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmake_sens_ic\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mn_states\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mn_theta\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfloatX\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 90\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/working/projects/pymc/pymc3/ode/utils.py\u001b[0m in \u001b[0;36maugment_system\u001b[0;34m(ode_func, n_states, n_theta)\u001b[0m\n\u001b[1;32m 119\u001b[0m \u001b[0mddt_dydp\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mJdfdy\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0mgrad_f\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mflatten\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 120\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 121\u001b[0;31m system = theano.function(\n\u001b[0m\u001b[1;32m 122\u001b[0m \u001b[0minputs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mt_y\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mt_t\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mt_p\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdydp_vec\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 123\u001b[0m \u001b[0moutputs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mt_yhat\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mddt_dydp\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/compile/function.py\u001b[0m in \u001b[0;36mfunction\u001b[0;34m(inputs, outputs, mode, updates, givens, no_default_updates, accept_inplace, name, rebuild_strict, allow_input_downcast, profile, on_unused_input)\u001b[0m\n\u001b[1;32m 304\u001b[0m \u001b[0;31m# note: pfunc will also call orig_function -- orig_function is\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 305\u001b[0m \u001b[0;31m# a choke point that all compilation must pass through\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 306\u001b[0;31m fn = pfunc(params=inputs,\n\u001b[0m\u001b[1;32m 307\u001b[0m \u001b[0moutputs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0moutputs\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 308\u001b[0m \u001b[0mmode\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mmode\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/compile/pfunc.py\u001b[0m in \u001b[0;36mpfunc\u001b[0;34m(params, outputs, mode, updates, givens, no_default_updates, accept_inplace, name, rebuild_strict, allow_input_downcast, profile, on_unused_input, output_keys)\u001b[0m\n\u001b[1;32m 481\u001b[0m \u001b[0minputs\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msi\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 482\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 483\u001b[0;31m return orig_function(inputs, cloned_outputs, mode,\n\u001b[0m\u001b[1;32m 484\u001b[0m \u001b[0maccept_inplace\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0maccept_inplace\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mname\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mname\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 485\u001b[0m \u001b[0mprofile\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mprofile\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mon_unused_input\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mon_unused_input\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/compile/function_module.py\u001b[0m in \u001b[0;36morig_function\u001b[0;34m(inputs, outputs, mode, accept_inplace, name, profile, on_unused_input, output_keys)\u001b[0m\n\u001b[1;32m 1830\u001b[0m \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1831\u001b[0m \u001b[0mMaker\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mgetattr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmode\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'function_maker'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mFunctionMaker\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1832\u001b[0;31m m = Maker(inputs,\n\u001b[0m\u001b[1;32m 1833\u001b[0m \u001b[0moutputs\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1834\u001b[0m \u001b[0mmode\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/compile/function_module.py\u001b[0m in \u001b[0;36m__init__\u001b[0;34m(self, inputs, outputs, mode, accept_inplace, function_builder, profile, on_unused_input, fgraph, output_keys, name)\u001b[0m\n\u001b[1;32m 1517\u001b[0m optimizer, inputs, outputs)\n\u001b[1;32m 1518\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1519\u001b[0;31m \u001b[0moptimizer_profile\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0moptimizer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfgraph\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1520\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1521\u001b[0m \u001b[0mend_optimizer\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtime\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtime\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/gof/opt.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, fgraph)\u001b[0m\n\u001b[1;32m 106\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 107\u001b[0m \"\"\"\n\u001b[0;32m--> 108\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moptimize\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfgraph\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 109\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 110\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0madd_requirements\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfgraph\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/gof/opt.py\u001b[0m in \u001b[0;36moptimize\u001b[0;34m(self, fgraph, *args, **kwargs)\u001b[0m\n\u001b[1;32m 95\u001b[0m \u001b[0morig\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtheano\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtensor\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbasic\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconstant\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0menable\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 96\u001b[0m \u001b[0mtheano\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtensor\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbasic\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconstant\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0menable\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mFalse\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 97\u001b[0;31m \u001b[0mret\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mapply\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfgraph\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 98\u001b[0m \u001b[0;32mfinally\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 99\u001b[0m \u001b[0mtheano\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtensor\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbasic\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconstant\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0menable\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0morig\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/gof/opt.py\u001b[0m in \u001b[0;36mapply\u001b[0;34m(self, fgraph)\u001b[0m\n\u001b[1;32m 249\u001b[0m \u001b[0mnb_nodes_before\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfgraph\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mapply_nodes\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 250\u001b[0m \u001b[0mt0\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtime\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtime\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 251\u001b[0;31m \u001b[0msub_prof\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0moptimizer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moptimize\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfgraph\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 252\u001b[0m \u001b[0ml\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfloat\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtime\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtime\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m-\u001b[0m \u001b[0mt0\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 253\u001b[0m \u001b[0msub_profs\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msub_prof\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/gof/opt.py\u001b[0m in \u001b[0;36moptimize\u001b[0;34m(self, fgraph, *args, **kwargs)\u001b[0m\n\u001b[1;32m 95\u001b[0m \u001b[0morig\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtheano\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtensor\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbasic\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconstant\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0menable\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 96\u001b[0m \u001b[0mtheano\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtensor\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbasic\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconstant\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0menable\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mFalse\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 97\u001b[0;31m \u001b[0mret\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mapply\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfgraph\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 98\u001b[0m \u001b[0;32mfinally\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 99\u001b[0m \u001b[0mtheano\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtensor\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbasic\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconstant\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0menable\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0morig\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/gof/opt.py\u001b[0m in \u001b[0;36mapply\u001b[0;34m(self, fgraph, start_from)\u001b[0m\n\u001b[1;32m 2511\u001b[0m \u001b[0mnb\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mchange_tracker\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnb_imported\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2512\u001b[0m \u001b[0mt_opt\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtime\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtime\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2513\u001b[0;31m \u001b[0mlopt_change\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mprocess_node\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfgraph\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnode\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlopt\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 2514\u001b[0m \u001b[0mtime_opts\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mlopt\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m+=\u001b[0m \u001b[0mtime\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtime\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m-\u001b[0m \u001b[0mt_opt\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2515\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0mlopt_change\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/gof/opt.py\u001b[0m in \u001b[0;36mprocess_node\u001b[0;34m(self, fgraph, node, lopt)\u001b[0m\n\u001b[1;32m 2032\u001b[0m \u001b[0mlopt\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mlopt\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlocal_opt\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2033\u001b[0m \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2034\u001b[0;31m \u001b[0mreplacements\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mlopt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtransform\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnode\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 2035\u001b[0m \u001b[0;32mexcept\u001b[0m \u001b[0mException\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2036\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfailure_callback\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/tensor/opt.py\u001b[0m in \u001b[0;36mlocal_subtensor_merge\u001b[0;34m(node)\u001b[0m\n\u001b[1;32m 3086\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mtype\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mslice1\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0mslice\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 3087\u001b[0m merged_slices.append(\n\u001b[0;32m-> 3088\u001b[0;31m merge_two_slices(slice1,\n\u001b[0m\u001b[1;32m 3089\u001b[0m \u001b[0mxshape\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mpos_1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 3090\u001b[0m \u001b[0mslices2\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mpos_2\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/tensor/opt.py\u001b[0m in \u001b[0;36mmerge_two_slices\u001b[0;34m(slice1, len1, slice2, len2)\u001b[0m\n\u001b[1;32m 3038\u001b[0m \u001b[0mstep\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mpre_greedy_local_optimizer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlist_opt\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstep\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 3039\u001b[0m \u001b[0mstart\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mpre_greedy_local_optimizer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlist_opt\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstart\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 3040\u001b[0;31m \u001b[0mstop\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mpre_greedy_local_optimizer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlist_opt\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstop\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 3041\u001b[0m \u001b[0mstep\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mpre_greedy_local_optimizer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlist_opt\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstep\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 3042\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/gof/opt.py\u001b[0m in \u001b[0;36mpre_greedy_local_optimizer\u001b[0;34m(list_optimizations, out)\u001b[0m\n\u001b[1;32m 2920\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2921\u001b[0m \u001b[0mout_index\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2922\u001b[0;31m final_outs, optimized_nodes = local_recursive_function(\n\u001b[0m\u001b[1;32m 2923\u001b[0m list_optimizations, out, {}, 0)\n\u001b[1;32m 2924\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mfinal_outs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mout_index\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/gof/opt.py\u001b[0m in \u001b[0;36mlocal_recursive_function\u001b[0;34m(list_opt, out, optimized_vars, depth)\u001b[0m\n\u001b[1;32m 2889\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2890\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0minp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mowner\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2891\u001b[0;31m outs, optimized_vars = local_recursive_function(\n\u001b[0m\u001b[1;32m 2892\u001b[0m \u001b[0mlist_opt\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2893\u001b[0m \u001b[0minp\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/gof/opt.py\u001b[0m in \u001b[0;36mlocal_recursive_function\u001b[0;34m(list_opt, out, optimized_vars, depth)\u001b[0m\n\u001b[1;32m 2889\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2890\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0minp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mowner\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2891\u001b[0;31m outs, optimized_vars = local_recursive_function(\n\u001b[0m\u001b[1;32m 2892\u001b[0m \u001b[0mlist_opt\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2893\u001b[0m \u001b[0minp\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/gof/opt.py\u001b[0m in \u001b[0;36mlocal_recursive_function\u001b[0;34m(list_opt, out, optimized_vars, depth)\u001b[0m\n\u001b[1;32m 2889\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2890\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0minp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mowner\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2891\u001b[0;31m outs, optimized_vars = local_recursive_function(\n\u001b[0m\u001b[1;32m 2892\u001b[0m \u001b[0mlist_opt\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2893\u001b[0m \u001b[0minp\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/gof/opt.py\u001b[0m in \u001b[0;36mlocal_recursive_function\u001b[0;34m(list_opt, out, optimized_vars, depth)\u001b[0m\n\u001b[1;32m 2905\u001b[0m \u001b[0mresults\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnode\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moutputs\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2906\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mopt\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mlist_opt\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2907\u001b[0;31m \u001b[0mret\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mopt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtransform\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnode\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 2908\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mret\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mFalse\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0mret\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2909\u001b[0m \u001b[0;32massert\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mret\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnode\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moutputs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mopt\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/tensor/opt.py\u001b[0m in \u001b[0;36mconstant_folding\u001b[0;34m(node)\u001b[0m\n\u001b[1;32m 6513\u001b[0m node.op.python_constant_folding(node)):\n\u001b[1;32m 6514\u001b[0m \u001b[0mimpl\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m'py'\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 6515\u001b[0;31m thunk = node.op.make_thunk(node, storage_map, compute_map,\n\u001b[0m\u001b[1;32m 6516\u001b[0m no_recycling=[], impl=impl)\n\u001b[1;32m 6517\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/gof/op.py\u001b[0m in \u001b[0;36mmake_thunk\u001b[0;34m(self, node, storage_map, compute_map, no_recycling, impl)\u001b[0m\n\u001b[1;32m 952\u001b[0m compute_map=compute_map, impl='c')\n\u001b[1;32m 953\u001b[0m \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 954\u001b[0;31m return self.make_c_thunk(node, storage_map, compute_map,\n\u001b[0m\u001b[1;32m 955\u001b[0m no_recycling)\n\u001b[1;32m 956\u001b[0m \u001b[0;32mexcept\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mNotImplementedError\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mutils\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mMethodNotDefined\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/gof/op.py\u001b[0m in \u001b[0;36mmake_c_thunk\u001b[0;34m(self, node, storage_map, compute_map, no_recycling)\u001b[0m\n\u001b[1;32m 855\u001b[0m \u001b[0;32mraise\u001b[0m \u001b[0mNotImplementedError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"float16\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 856\u001b[0m \u001b[0m_logger\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdebug\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'Trying CLinker.make_thunk'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 857\u001b[0;31m outputs = cl.make_thunk(input_storage=node_input_storage,\n\u001b[0m\u001b[1;32m 858\u001b[0m output_storage=node_output_storage)\n\u001b[1;32m 859\u001b[0m \u001b[0mthunk\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnode_input_filters\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnode_output_filters\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0moutputs\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/gof/cc.py\u001b[0m in \u001b[0;36mmake_thunk\u001b[0;34m(self, input_storage, output_storage, storage_map, keep_lock)\u001b[0m\n\u001b[1;32m 1213\u001b[0m \"\"\"\n\u001b[1;32m 1214\u001b[0m \u001b[0minit_tasks\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtasks\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget_init_tasks\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1215\u001b[0;31m cthunk, module, in_storage, out_storage, error_storage = self.__compile__(\n\u001b[0m\u001b[1;32m 1216\u001b[0m \u001b[0minput_storage\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0moutput_storage\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstorage_map\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1217\u001b[0m keep_lock=keep_lock)\n", + "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/gof/cc.py\u001b[0m in \u001b[0;36m__compile__\u001b[0;34m(self, input_storage, output_storage, storage_map, keep_lock)\u001b[0m\n\u001b[1;32m 1151\u001b[0m \u001b[0minput_storage\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtuple\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minput_storage\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1152\u001b[0m \u001b[0moutput_storage\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtuple\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0moutput_storage\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1153\u001b[0;31m thunk, module = self.cthunk_factory(error_storage,\n\u001b[0m\u001b[1;32m 1154\u001b[0m \u001b[0minput_storage\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1155\u001b[0m \u001b[0moutput_storage\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/gof/cc.py\u001b[0m in \u001b[0;36mcthunk_factory\u001b[0;34m(self, error_storage, in_storage, out_storage, storage_map, keep_lock)\u001b[0m\n\u001b[1;32m 1621\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mnode\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnode_order\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1622\u001b[0m \u001b[0mnode\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mop\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mprepare_node\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnode\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstorage_map\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'c'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1623\u001b[0;31m module = get_module_cache().module_from_key(\n\u001b[0m\u001b[1;32m 1624\u001b[0m key=key, lnk=self, keep_lock=keep_lock)\n\u001b[1;32m 1625\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/gof/cmodule.py\u001b[0m in \u001b[0;36mmodule_from_key\u001b[0;34m(self, key, lnk, keep_lock)\u001b[0m\n\u001b[1;32m 1157\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mmodule\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1158\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1159\u001b[0;31m \u001b[0;32mwith\u001b[0m \u001b[0mcompilelock\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlock_ctx\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mkeep_lock\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mkeep_lock\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1160\u001b[0m \u001b[0;31m# 1) Maybe somebody else compiled it for us while we\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1161\u001b[0m \u001b[0;31m# where waiting for the lock. Try to load it again.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/contextlib.py\u001b[0m in \u001b[0;36m__enter__\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 111\u001b[0m \u001b[0;32mdel\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mkwds\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfunc\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 112\u001b[0m \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 113\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mnext\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mgen\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 114\u001b[0m \u001b[0;32mexcept\u001b[0m \u001b[0mStopIteration\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 115\u001b[0m \u001b[0;32mraise\u001b[0m \u001b[0mRuntimeError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"generator didn't yield\"\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfrom\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/gof/compilelock.py\u001b[0m in \u001b[0;36mlock_ctx\u001b[0;34m(lock_dir, keep_lock, **kw)\u001b[0m\n\u001b[1;32m 38\u001b[0m \u001b[0;34m@\u001b[0m\u001b[0mcontextmanager\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 39\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mlock_ctx\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlock_dir\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkeep_lock\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mFalse\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkw\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 40\u001b[0;31m \u001b[0mget_lock\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlock_dir\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mlock_dir\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkw\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 41\u001b[0m \u001b[0;32myield\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 42\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0mkeep_lock\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/gof/compilelock.py\u001b[0m in \u001b[0;36m_get_lock\u001b[0;34m(lock_dir, **kw)\u001b[0m\n\u001b[1;32m 84\u001b[0m \u001b[0;31m# Only really try to acquire the lock if we do not have it already.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 85\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mget_lock\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mn_lock\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 86\u001b[0;31m \u001b[0mlock\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mget_lock\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlock_dir\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkw\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 87\u001b[0m \u001b[0matexit\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mregister\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mUnlocker\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0munlock\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mget_lock\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0munlocker\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 88\u001b[0m \u001b[0;31m# Store time at which the lock was set.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/gof/compilelock.py\u001b[0m in \u001b[0;36mlock\u001b[0;34m(tmp_dir, timeout, min_wait, max_wait, verbosity)\u001b[0m\n\u001b[1;32m 271\u001b[0m \u001b[0mno_display\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mTrue\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 272\u001b[0m \u001b[0mnb_wait\u001b[0m \u001b[0;34m+=\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 273\u001b[0;31m \u001b[0mtime\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msleep\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mrandom\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0muniform\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmin_wait\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmax_wait\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 274\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 275\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mPY3\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;31mKeyboardInterrupt\u001b[0m: " ] } ], @@ -153,20 +189,21 @@ "ode_model = DifferentialEquation(\n", " func=freefall,\n", " times=times,\n", - " n_states=1, n_theta=2,\n", + " n_states=1, \n", + " n_theta=2,\n", " t0=0\n", ")\n", "\n", "with pm.Model() as model:\n", " # Specify prior distributions for soem of our model parameters\n", - " sigma = pm.HalfCauchy('sigma',1) \n", - " gamma = pm.Lognormal('gamma',0,1)\n", + " sigma = pm.HalfCauchy('sigma', 1) \n", + " gamma = pm.Lognormal('gamma',0, 1)\n", " \n", " # If we know one of the parameter values, we can simply pass the value.\n", " ode_solution = ode_model(y0=[0], theta=[gamma, 9.8])\n", " # The ode_solution has a shape of (n_times, n_states)\n", " \n", - " Y = pm.Normal('Y', mu=ode_solution, sd=sigma, observed=yobs)\n", + " Y = pm.Normal('Y', mu=ode_solution, sigma=sigma, observed=yobs)\n", " \n", " prior = pm.sample_prior_predictive()\n", " trace = pm.sample(2000, tune=1000, cores=1)\n", @@ -177,22 +214,9 @@ }, { "cell_type": "code", - "execution_count": 4, + "execution_count": null, "metadata": {}, - "outputs": [ - { - "data": { - "image/png": "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\n", - "text/plain": [ - "
" - ] - }, - "metadata": { - "needs_background": "light" - }, - "output_type": "display_data" - } - ], + "outputs": [], "source": [ "az.plot_posterior(data);" ] @@ -208,25 +232,9 @@ }, { "cell_type": "code", - "execution_count": 5, + "execution_count": null, "metadata": {}, - "outputs": [ - { - "name": "stderr", - "output_type": "stream", - "text": [ - "Auto-assigning NUTS sampler...\n", - "Initializing NUTS using jitter+adapt_diag...\n", - "Sequential sampling (2 chains in 1 job)\n", - "NUTS: [g, gamma, sigma]\n", - "Sampling chain 0, 0 divergences: 0%| | 0/3000 [00:00" - ] - }, - "metadata": { - "needs_background": "light" - }, - "output_type": "display_data" - } - ], + "outputs": [], "source": [ "az.plot_posterior(data);" ] @@ -281,34 +276,20 @@ }, { "cell_type": "code", - "execution_count": 7, + "execution_count": null, "metadata": {}, - "outputs": [ - { - "name": "stderr", - "output_type": "stream", - "text": [ - "Auto-assigning NUTS sampler...\n", - "Initializing NUTS using jitter+adapt_diag...\n", - "Sequential sampling (2 chains in 1 job)\n", - "NUTS: [y0, g, gamma, sigma]\n", - "Sampling chain 0, 0 divergences: 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 3000/3000 [11:00<00:00, 4.54it/s]\n", - "Sampling chain 1, 0 divergences: 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 3000/3000 [09:51<00:00, 5.07it/s]\n", - "100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 4000/4000 [00:41<00:00, 96.92it/s]\n" - ] - } - ], + "outputs": [], "source": [ "with pm.Model() as model3: \n", - " sigma = pm.HalfCauchy('sigma',1)\n", - " gamma = pm.Lognormal('gamma',0,1)\n", - " g = pm.Lognormal('g',pm.math.log(10),2)\n", + " sigma = pm.HalfCauchy('sigma', 1)\n", + " gamma = pm.Lognormal('gamma', 0, 1)\n", + " g = pm.Lognormal('g', pm.math.log(10), 2)\n", " # Initial condition prior. We think it is at rest, but will allow for perturbations in initial velocity.\n", " y0 = pm.Normal('y0', 0, 2)\n", " \n", " ode_solution = ode_model(y0=[y0], theta=[gamma, g])\n", " \n", - " Y = pm.Normal('Y', mu=ode_solution, sd=sigma, observed=yobs)\n", + " Y = pm.Normal('Y', mu=ode_solution, sigma=sigma, observed=yobs)\n", " \n", " prior = pm.sample_prior_predictive()\n", " trace = pm.sample(2000, tune=1000, target_accept=0.9, cores=1)\n", @@ -319,22 +300,9 @@ }, { "cell_type": "code", - "execution_count": 8, + "execution_count": null, "metadata": {}, - "outputs": [ - { - "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAA7AAAADgCAYAAAA624qoAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjAsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+17YcXAAAgAElEQVR4nOzdd3xb1fn48Y+GLe9tJ3G81yWD7MQOWSQh7B1aaMMoUFqgBFoILbPsGTaUUTYhLfy+BdoSoBASIIPsReb1imccb3kvjd8fskziLNuRfSX5eb9efVGkK+mR8Lnnnnue8xyd3W5HCCGEEEIIIYRwd3qtAxBCCCGEEEIIIXpCBrBCCCGEEEIIITyCDGCFEEIIIYQQQngEGcAKIYQQQgghhPAIMoAVQgghhBBCCOERZAArhBBCCCGEEMIjGLUOQPQvRVFuAP4MxAHbgdtVVV2nbVRCeDdFUezAnaqqPqN1LEIIIYQQ3kRmYL2YoihXA68DHwLzATPwtaIoyZoGJoT3mwos1ToIIYQQQghvo7Pb7VrHIPqBoig6YD/wlaqqN3U+5gOowDJVVW/VMj4hhBBCCCGE6C1JIfZgiqI8C1wLDFVVtf2Qx78BGoBE4L/Ox1VV7VAU5Qvg7IGOVQhvoyhKJrAYGA90ACuARaqqFnZPIVYUZTbwFHAqkA/cDnwB/FZV1fcURXkQOB94CfgrMBxYC1wFXAjcC4QCnwO/U1W1ufN9FeBR4PTO5w8AbwOPqqoqdyfFoKMoih/wNHAF4Af8P6AC+LWqqkmKooQAjwAXA8OAOuBL4DZVVc2d72EHrgfOw9Ff1nW+5r/A34HZQEnna77qfM33OJbptAHXASYcWRiLgCeAq4EW4CVVVZ84JN6zgHuACYAPsA94WFXVT13/6wjh/U50DtAwNOFCkkLs2d4HwoGznA8oijIUmAN80PlQbrfX5AOpiqIYBiRCIbyQoigBOC56S4GLgBtwXIB+dJRjTwW+AsqBS4H3cHSo3dugAvwFuBP4LZAF/IDjYvhmHBfBvwZu63zfIOB7IBK4BsfF9krgYRyDYSEGo3eA3wAP4WgvaThuGDn9A0ebvQs4E3im87j7u73P80AOcAGwDngF+BbHjaVf4BjULu08FzhdB4wEfgU8C9wIbMNxc+nyztc/rijKVABFUabgOI/s6ozpcqAZ+IeiKNEn9SsIMXid6BwgvIDMwHowVVV/UhRlB44G+nnnw1fg6FjDOv+9odvLGnDcuAgE6gciTiG80CggAsdsyjoARVGqgDmKonS/MXgXjtmaS1RVtQBfKYpiw3HhfKgg4DpVVTd0vt/5ONpzkqqqhcAyRVHOAzI7j1dw3KC6XFXVys7XrAAuAWbx8zlBiEFBUZQMHIPHa1VVfa/zsZU4ltM4Z2Z8gRtVVf1f58u+VxTlNBxt5lA/qqp6V+frSnHcfFqnqurjnY+14hiQZuCYeQWwAr9UVbUF+FZRlN/j6G+vVVXV1jlL+yscbXgdjvPIp6qq/uGQ71AEbO08ZpkrfhchBosTnQOE95ABrOf7AHhEUZRAVVWbgCtxzAJ1dD7fPY1Q1/lP2wDFJ4Q32gfUAJ8rivIRjnTglaqq/gDgyOztcjrwf52DV6f/48gBrB3YfMi/lwOVnYNXp2o6b06pqroFmKEoio+iKCNxXEiPx5GGaDqpbyeEZ3IOQv/tfEBV1ebOpTNzVFVtxTHriqIoSTjazGgcs6at3d5r4yH/v7zzn4e2z+rOf4Yd8thPnYNXpwogW1VVW2csbYqiNPJzG34XeFdRlEBgRGc8czpfK21YiN477jlAm5BEf5AUYs+3FMcd5Qs77zxNBJbgmIUFCO52fBCOwWvTgEUohJdRVbUBmIlj3es1ONIADyqKctNRDo8CKrs9Vn6U45pVVbV2f+x4cSiKck/ne+8GXgDScdy80h3vdUJ4qSigw7mW9RBd7U1RlAsVRcnDMSOzFJiHo511bzPds5fgBO2xt69RFCVQUZQPcewQsA7HWli/zqelDQvReyc8BwjvIANYD6eqajnwDXBZ5/9yVVVdj2PtDkBKt5ekOF4mBV6EOBmqqu5WVfVyHGtQzwBWA692Fnc61AGg+3q2k17f1rlN1qM4UpTDVFVNUlX11/ycfSHEYHMA8FEUJazb49EAiqKk48h+WAHEq6oararqOTiq82vhZRwzwucCQaqqjgYe1ygWIbzBcc8BwnvIANY7fICjkNN8HLOv4BjAFuOotAh0baNzHo7OWwjRR4qinK0oSoWiKNGqqrarqroCWNj5dEK3w1cB53VbG3uRC8KYCpSoqvq6qqp1nXFNwNFRy+yNGIzW4sgwutD5gKIovvxceX8CjoylJ1VVLel8PhCYjjZtZirwP1VVl6uq2tb5mDNWacNC9N6JzgHCS8gaWO/wH+B1HJ3zLwBUVbUrivIk8IqiKLU4GvUtONIrntcqUCG8xEYcF5ifKoryFNAO/BFHKuB33Y59EkeRl08URXkDxzq3RzqfO5m16JuAGxVF+SuOasUjgAdwrKUNON4LhfBGqqrmKoqyFHipc2BaCNyKY7ucQhwVga3AU4qivIajP1wEDMWx/c1A24Rj+c81QBGONXp3dj4nbViIXurBOUB4CZmB9QKdhSm+B9aqqpp/yOOv4ugMrwL+haNwxFmHHiOE6D1VVWtw3NFtxZH18BmOtWtnqKpa1e3YvTi24kjGcbPpd8CfOp9uPIkw3sOx192NONbgLsSxL+3bOLbgEWIwuglHIcPHOv9ZiCNtuFFV1Wwc+7GOwdFmnsZRmOlmIEFRlNgBjvUOYDmO9eufAnNxVDvOxjE7K4TovWOeA7QMSriWzm6XpZCernNrgFLgL6qqvqV1PEKInymKcgbQ4Nwep/OxM4GvgbGqqv6kWXBCeBFFUaJwrCn9vLPQmvPxH4GDqqpeqllwQoh+J+eAwUNSiD2YoijhOFIjZgMWHBu0CyHcSxZwp6Ioi3AUi0kEHgZWyeBVCJdqAV4FfqEoyus4+sVf4GiD87QMTAgxIOQcMEhICrFnawX+gKOy8AJVVU9U4l8IMfCeBF7CUS34G+ApHOmLF2gZlBDepnMv9DNxbBf3EfBfYCxwQWehNSGEF5NzwOAhKcRCCCGEEEIIITyCzMAKIYQQQgghhPAIMoAVQgghhBBCCOERjlvEqbKywaX5xUFBJhobtdhqrX/J9/IsnvS9oqODe7SZvavbqjvxpP9eriDf1zP1tK1C/7VXb/kttSK/38nxlN9P2qr3kN/RddzxtzxeWx3QGVij0TCQHzdg5Ht5Fm/9Xt5qsP33ku8r+kp+y5Mjv9/Jkd+v5+S3cg35HV3H035LSSEWQgghhBBCCOERZAArhBBCCCGEEMIjyABWCCGEEEIIIYRHkAGsEEIIIYQQQgiPcNwqxMKzdFhtVDa2o9dBdJAJg77HhfaEcHvVTe20W20MCTah18nfthDi5NntdswtHTS1W/ELNGkdjhCDnsVmp6qxDTsQEeCLyShzbeJIMoD1AltLzCzZVMKmIjNtFhsAwSYjs9Mj+c2UBOLD/TWOUIi+K6xu4q5Pd7K5yAzA8FA/bp6exJmnxGgcmRDCU7V2WPl/2w7w2c4ySsytAPgYdGQmhnPjaUkoQ4I0jlCIwaOwppkv91awpcjMvorGrmtZgGEhJs7IiOZXE4cTHSQ3mYSDDGA9WEuHladX5LJsdzkRAT5cMmYYaVEBWO2w80A9X++r5Ot9lfxxVgqXjYvVOlwhek2taOSm//sJHXDjtERC/Xz4766D3PvFPgpqmvndaUlahyiE8DBbis08+JXKwYY2JsaHctnYWEL8jJQ2tvOvLSVcs3Qrt85K4dcT47QOVQivtrXEzBtrC9laUodeB6OGhjB/7DCSIgLQATXNHewqq+cfW0r4144D3Dozhfljh6GTLKxBTwawHqqmuZ2F/9pJblUT12XGc21mAn4+P+/hdOmYYfxhehKPfpPDUytyKW9o4w8zkjWMWIjeqW1u57ZPdxFkMvL6L8YQG+oHwMWnDuWx5Tm8ua6IocF+XHjqUI0jFUJ4iv/uOsjjy3MYHurH3y8fy/i40K7nwsICWDAulke+yeb57/Np7bBxXVaChtEK4Z2qGtt4bHkOa/JriA7yZeGMZM4ZGXPMGdYScwtPr8jlqRW55Fc3c+ecVBnEDnIygPVAjW0Wbvp/P1Fa18pzl4xmWnLEUY+LCjLx7MWjeGpFDu9tLCY6yJdfjh8+wNEK0TeLV+ZR19LBZzedxlC/n2/OGA167j0zg4qGNhavzGVcXCgJkiYvhDiBt9YV8saPhWQmhvHkBSMJMh15CRTsZ+SJ80fw8Ncqr60tID7cn3lKtAbRCuGd1ubX8OD/VFo6rNw6M5lfjIs9bALmaOLC/Hnh0tG89MN+lm4pwd/HwMKZMikzmMnKaA9jtdm5/8t9FNa28Pwlo445eHUy6HX8ZW46M1Mjee67PHaU1g1QpEL03ZZiM8vVSq7LSuCUocFHPG/U63jgbAVfo57Hl2djt9s1iFII4Sk+3FzCGz8Wct7IGF64ZPRRB69OBr2O+87MYExsCI9+nU1ZfesARiqE9/rvroP86bNdRAf5suTKCVw1Of6Eg1cnvU7HbbOSuXTMMD7YVMy3amU/RyvcmQxgPcwbPxawJr+GRbNTmZwQ3qPXGPQ6HjpHYWiIH3/9ch+tHdZ+jlKIvrPb7by2poDoIF+umnTsNWgxwSZunJbEluI6vs+tHsAIhRCeZNnug7z4Qz5nZERz/1kKRsOJL318DHoePe8UAB77Rm6SCXGyPtlxgEe+zmZKYhjv/GocyZEBvX4PnU7HojmpjB4WzKPfZFMuN5cGLRnAepAtxWbe3VDMRacOZf7YYb16bZDJyF/PzuBAfRtvrS/qpwiFOHlbS+rYcaCe67qt6z6aS8YMIzkigNfWFGC1yQWmEOJwW4rNPPpNDlMSwnjoHKVX28sNC3FUPN9QaGZVntwkE6KvVuVV8+S3uUxPieDZi0f3eNb1aHwMeh4+5xQ6rDYe+WKvC6MUnkQGsB6itcPaVXhi0ey+LV6fEBfGeaOG8I8tJZISJdzWR1tLCfUzcv6oISc81qjX8dupCeyvaWZlTtUARCeE8BRl9a3c9fle4kL9eOrCkfj2YT/J+eNiSY4I4KVV+7HITTIheq2svpW/frmPEUOCeOL8ES7Z1zU+3J/rshL4ek85W0vMLohSeBoZwHqIdzYUUVTbwt3z0k/qztVN05LQ63S8sbbAdcEJ4SJl9a2syqvmkjHDevx3PjcjmsRwf5ZsKu7n6IQQnqK1w8qif+/GYrPx7MWjjrvm9XiMeh03T0+iqLaFldmy5k6I3rDZ7Tz4lQrAExeMOKnr1+4WTIxjSLCJV1YVSIr/ICQDWA9QWNPMB5tKOG/UEDITe7bu9ViGBJuYP3YY/9tbIbOwwu18sbscm92RGtxTBr2OX46PZW95I3sONvRjdEIIT/HCD/lkVzbxyLmnkBjR+7V2h5qZFklShD/vbyyWC2UhemHZ7nK2ltTxp9NTGB7q2t0C/HwM/GF2KjvL6tlYJLOwg40MYD3AGz8W4mvQcauLSob/asJw0On4aGupS95PCFew2+18uaecSfGhXXu+9tS5I4fgZ9TzyY4D/RSdEMJTrMyu5JMdZVw5KY7pKZEn/X56nY6rJseTXdnE+sJaF0QohPdrbLPw8qr9jI0N4YLR/bNf+6Xj44gO8uW9DVLbZbCRAaybU8sbWa5W8quJcUQE+LrkPYeG+HFGRhT/2XmQxjaLS95TiJO1q6yBYnMr5/Vg7Wt3QSYj54yM4et9ldS3dvRDdEIIT1DR0Maj3+QwcmgwN09Pctn7njMihpggX97fKEsVhOiJf2wpwdzSwR1zUtH3oW5LT5iMen41YTibi+vIrmjsl88Q7kkGsG7utbUFhPgZuXLisbcT6YsrJ8XR1G7ls5/KXPq+QvTViuwqjHodp6dF9en188fG0maxsWx3uYsjE0J4ArvdzhPf5tButfHouafg04PtcnrKx6DnignD2VJcR15Vk8veVwhvVN/awT+2lDInPYoRQ47cy92VLhw9FF+DTq5nBxkZwLqx3WX1rN1fw9WT4wn261sBimM5ZUgwk+JD+XjbAdl+RGjObrezMqeSrKTwPhdbUWKCGD0smM93yQBWiMHoq70VrMmv4ebpScSHu3a9HcD5o4Zg0OvkHCPECXy6o4ymdivXZyX0+2eF+vswNyOar/ZW0NJh7ffPE+5BBrBubOmWUoJMBi4b17s9X3vq0rGxlDe0sblYFr8LbWVXNFFW38bs9L7NvjqdMyKG3Kom8qtlhkSIwaSmuZ1nv8tjbGwIl48f3i+fER7gy8zUSL7cU47FauuXzxDC03VYbXy87QBZieFkxAQNyGdeMmYYTe1WlqtSKXywkAGsmyqrb2VldiUXnzqMQF/Xzr46zUyNJNhk5ItDUi737dvDTTddx9y50zj33LP56qtlx32PpqZGFi9+nIsuOptzzpnDww/fT21tTb/EK7zXjwWOv5npKREn9T5zMqLR62D5vt51YitWfMNvfvNrzjhjOpdffjFLlryH1XrsO7klJcXcffcdnHfeXM4//wzuu+8vHDx48KRiF0L03atrCmhqt3LvmRkY9P2z3g7gwtFDqG3p4OMV67v6yiuuuOSEfeWhdu3ayaxZmWzdurnf4hRCK6vzqqlqaueKif1zI+loxg0PISnCn3//VMb69T9y/fVXMW/eDK6++nKWL//fcV/b0tLC888/zUUXnc3ZZ5/OHXfcyv79+QMUuegrGcC6qY+3OqqpXj4+tt8+w2TUM0+J5rucKpraLdTW1nL77QvJyDiFd95ZyoIFV/Lkk4+wceP6Y77H/fffxfr1P3LPPQ/wt7+9SUtLMwsX3kh7e3u/xS28z7r9NZwSE3TShcqiAn2ZEB/GN2plj7e7WLduLQ8/fD8XXHAR77//ETfeeAtLl77PkiXvHvX4lpYWbr/9FqxWGy+++DrPPvsKdXVmFi26Vf7uhdDA3vIG/rvzIJePjyU58uS2zDmRrKQIIgxt/P3pu7v6yssuu/yEfaVTS0sLjz761+PeIBPCk/13VzkxQb5kneS2j72h0+m4ZMwwdu36iTvvvI0xY8bx1ltLuPLKa1m8+An+978vjvnaF198hs2bN/LII0/y+uvv4uvryx13LKStrW3A4he9JwNYN9TUbuHfO8uYkxHN0JDebSfSW+eNGkKrxcaK7CqWLfs3gYFB3HbbIhITk1iw4ErOOutc/vnPJUd9bU6OysaN67nrrvvJzJxKSkoq99//CFVVlaxY8U2/xi28R2ObhZ/KGshKck1nd6YSTVFtC9kVPUsj/s9/PmHWrDnMn385w4fHMXv2GVxxxQK++OLzox6/ceN6yssP8sADj5CWlo6inMJ99z1EQUE+e/bscsl3EEL0jN1u59mVeYQH+HDD1MR+/zyjXkdqw04sehNX/vZWEhOTuOyyK47bVx7q5ZefIzo6pt/jFEILFQ1trCuo6VovPpDOVKIx5nxHVKLCbbfdQWJiEmeeeTYLFlzN22+/cczXrV79PZdcchljxowjKSmZ3/3uZioqyiko2D+A0Yve6p/c1EFk+vRJPP30YpYs+RBV3Uts7HDuuuuv5OSoLFnyLk1NjWRlTePeex/E19cxu7RjxzZee+1lsrNVIiOjmDt3HtdeewMmkwmAD7/dSNv3r7Hp82JOf6aNYcNiufrq6zjnnPMBuOWW3zF69BgqK8tZvXoVgYGBzJx5OgsX3o7RaOTtt9/g3XffPGq855xzPvfe+2DXv586LJiEcH+W7S4nZMc2xo0bj17/832N8eMn8uyzT2Kz2Q57HKC42LGdwJgx47oeCwgIID4+nm3btnTFKwaH6dMn8cADj/Lpp//Xq7bw5HPPYdifwzerozDsPvuwtpCTk80777zGtm3baG1t7VFbyDxtJgZ9Ft+oFaxZtvSEbeGaa67Hz+/wgi86nY6Ghvqjvm7kyFE888yLBAb+vLbH2TaO9RohxJH6es44tP9MHz+NHcaJ3Hv2SIJMRnJysnnjjVfYteunHp8znP0n0LP+szIPW1Qqq/JrmD/WkSV1vL7Sad26Naxbt5ZnnnmJa665wsW/phCu09e2+dDTz2AszmP52mh0u846oj/va9vs6bWtf1sNzdGnYrfb0XVu3ZORoVBWdoCDBw8ydOiR+9GGhYWzYsVy5s49k6CgYJYt+w/BwSHExg5cCrToPRnAusBTTz3JnXfeS0JCAo8++iB33nkbI0eO5plnXqSoqJCHHrqPceMmcMkll5GTo3L77bdw/fU3ct99D1FefpAXXlhMTU0199zzAC0tLfzj+XsJjh/Dm889Auj46KMPefrpx8jMnEpEhGNT9o8/Xso111zPW299wKZNG3jhhcWMHDmas846l1/96iouvnj+UWM1mQ6f0dXpdJw7MobX1xaScbCc9HTlsOejoqJobW2lvr6esLCwI54DqKysIC4uHgCr1UpFRQXh4Se3llF4ppdffp4//7l3bSF55mUYlfncO28YL7/07GFt4fbbb2H27Nn8/e/vYbfT47aQdlY4K7IDWPqrK0/YFkaMGHXY401Njfz735+QmTn1qK+Ljo45Ygblww/fw8/P77CbOUKIE+vLOcPZf5YdLOPPDz1KaFQx5y+a03XOmDZtRq/PGSNHjubyyy/rUf/ZaK4mJCyNb9XKrgHs8fpKALPZzJNPPsrdd/+V4OD+3VZECFfoS9v0OfVc0mdew8JJ4Udc255M2zzWtW1oqD91dS1dbXNIdDT7ayvJq2omLToQgLIyx/Y6ZnPNUQewf/7zvZ3LiM7EYDBgMvnx3HOvSDt1c5JC7AKXXjqf6dNnkpCQxFlnnUtDQz133PEXUlLSOP30uaSlZZCfnwfAP//5IVOnTufXv76KuLh4Jk6czJ133sOXX35OVVUV6oEa2lJmcsX1C0lMTCYxMYmrrrqWjo4OiouLuj4zLS2D3/zmtyQkJDJ//i9JTU1n166dgGMWNDIy6qj/Cwo6siLc3IxoAMyNTV13ypx8fBx31trbj1wLMGLEKBITk1i8+Amqqqpoa2vljTdewWyupaOjwzU/rvAo5513Ya/bQkXsaUwemc6UyZmHtYXW1hYuv/zX3H33PSQkJPWqLYQ2l1Ja10p5K71qC62trdx99yLa2tq46aaFPfrOn332Lz755P9x000LCQ098sJVCHFsfTlnOPvPhtAUmk6dT2vOOsw11V3njD/+8c5enzN603+2tbWSOiSUrSV11DQ71r0fr68EWLz4MaZNm0FW1mn99lsK4Uq9bZujxmdSHT+D+dPGHHFt219tMyoq+rC2eemFF6Iv2c5b//dvLBYLOTkq//znhwDHvC4tKSkmIiKSxYtf4NVX32bKlCzuv/8vVFTIdlnuTGZgXSAh4ed9rvz9/dHr9Qwb9nPxJZPJREeHo5PLzlYpKSli3rwZXc87i80UFu7nx6ZISJmGb8kWnn76Y0pKisnJyQY4rOjDoZ8JEBQUhMXiaJwffPDOMQvQnHnmOdx55z2HfT5AyoLnqbIbjihC44y7e4olgI+PD48//gwPP3w/F198Nj4+PsybdzZTp07DaPQ56ucL7+aciYeetYXikiIsq1ezxaBn3t90h7WFiRMnc8kll/Hf//6HHTt29qotRAUYAHjp9TfY/f1nR43V2RaczGYzd911OwUF+Tz//N8YOvTE21e9//7bvPnma1x11bXMn3/5CY8XQhyut+eMQ/vP1g4bJuzYOfyc8b//Les8tn/6T5PJRGKoL5us8F1OFfPHxh63r/zqq2VkZ2fz/vv/7O3PI4Rmets2i4oL8WU9L35l4CUdR+3P+7ttXnbRRfxz9U7WfPwKcz56kSFDhnLFFVfy/PNPH7bsx+nAgVKefvox/va3txg9+lQAHnzwMRYsuIyPP/4HCxf+qde/mxgYMoB1AaPx8J9Rp9N15d535+Nj5JxzzmfBgmuOeC40PJK731hNwPfPsSZxONOmzeC002YQFRXNb397Vbf3ObJaq/NkcfHF85kzZ95RPz8w0JFS8e67/zjs8S+LrLxnDKa0vOKwx6uqqvD3DzjqbBVAYmISb7+9hLo6Mz4+PgQEBHLddQuYPDnrqMcL79bbtjBiyhy2BE7k+cvGEBv6c3p7ZGQUVVVV/P73v2Ho0KFkZU3rVVvwM+oZMSQIc8cU3n33F0f9fGdbACgrO8Cf/nQLLS1NvPLKm6SlpR/3e9psNp599kn+859PuemmhUdtz0KIE+tr/7mpqJbHl+fwx9NTmJUaddg5Izo6pl/7z5iYIdha6kgc4t+VRny8vvLLLz+nsrKciy4667DPWrToNs4557zDbqQJ4S562zZNqVkknXY+D5x9ymHP9WfbDAnxo76+9bD+/Mqrr+fxIdN4+YJkMpVE1q5dhcFgOOpN6X379mK1WjnllBGHfe/0dIXS0uKjflfhHmQAO8CSk1MpKNh/2J2tXbt+YunS95n1y5uoy9tCiK2Nv/3tTQwGxyzShg3revUZISGhhISEHveYQz8fYI5fE+9EprBl65bDFr9v3bqZU08de9SiFM3NTfz5z3/i9tv/TEpKGuAYCOTm5vCHP/yxVzGLwSc5OZUNe/KImHUBk0emodPputrCokV388MPK2lubuaDD5bQ0OBIy+tNW5iREsmb6xoJiBhy3O15amtruPXWG9Hr9bz22js9Ktzw3HNPs2zZf7jnngc499wLehyTEKLvDu0/H19XS8zQ4YwO0/G3v71w2Dmjv/vPMWPG8eWXnzP3lqt4b2MxNc3tx+0r//rXRw7bkqO6upo//OG33HXXfUyenNmr+IRwR9Gxiag7c5k3cRRxcY5Z2qP1565um2FhAZjNzV3//sknH5O7vxCdPotdtTqydDpWrfqe0aPHEBBw5BZbMTGOehZ5ebkoimPgbbfbKSjYL+n+bk7WwA6wBQuuYc+eXbz88nMUFhawbdsWHn30ARoaGthYbsUUEEJbSwvff7+SgwfLWLPmBxYvfhygX/eYTI0MYNjYWdTXmVm8+HEKCvazdOmHLF/+PxYsuLrruPr6Ourr6wAICA4k2BIAACAASURBVAjEarXy4ovPsX+/YwuRP//5j0ycOJmJEyf3W6zCOyxYcA11pbmEqssoKio8rC1ERkYRFhZOc3MTy5d/06e2MCM1AjuwNr/muMc9++xTmM1mHnzwMUwmE9XVVVRXV1FTU911TG1tLY2NjYCjkui///0vrr76OjIzp3YdX11dJfvGCdGPnP3nI08/xebdKtOCa3niiYeOOGf0d/95/vkXYTbXUrLiA+x15bz09vvH7Sujo2OIi4vv+t+wYY6ZoKioaCl4KLzC0MnnoqstJH/F0iOubQeybSYkJPHl558QX7+blduzWbLkXb755it++9sbu445tG2OGDGK0aPH8PjjD7Jjx3YKCwt45pknKC8/yGWXybIgdyYzsAMsNTWNxYtf5M03X+Ozz/5FYGAQ06bN4Pc3LuSypXs4ffZcYhItvPDCYlpamhk+PJ7f/Oa3fPjhe+zbt6ff7gjpdDrOHJvC+2U3sFddznXXLSA2Npb77nvosMHoPffcCcArr/wdgIceepznn3+aG2+8FpPJj1mz5vS4+I0Y3IJi4mmfegOW0pVce+2vu9rCzTffBsCcOWewd+9uHn/8cZqbm3rdFpSYIGKCfFmVV80Fo4+sPAiOYiyrVn2HzWbjhhsOTwM2GAz88MMGAG644WrGj5/Ivfc+yNdffwXAu+++eURJ//vvf5izzjq3T7+HEOL4nP3nQ88+j2/ZZ/y4OZjp02cecc7o7/4zIiKSZ555mRdfXIyv+hVrw6JO2FcK4c12Ngcx7NyF5Knfcu21nx2zP+/vtjl5ciaLFt3Nq2+9xUFzNStT03jqqecZP35i1zGHtk2DwcCTTz7HG2+8woMP3kNLSzOKMpJXX32rR3UwhHZ0ztzyo6msbDj2k33QfarfW7jie/24v4bbPt3FsxePYmZqpIsi6509Bxu4Zuk2Hjg7g/NHDZX/Xm4gOjq4RzuBu7qtDoRv9lVw7xf7+ODK8YwYcuxy9Sfz3+vx5dl8s6+Sb2+eitHgGQknnvT36Qre8n172lah/9qrt/yWx1PX0sF5f9/A2afEcN9ZGS597778fs5zzPKbp+LjIeeY/uIpf3/SVl2rsrGNc9/YwE3TkrguK+HEL3ChY/2O2RWNLFiylXvnpXPxGBmI9oQ7/k0er60O7rOtG1mRXUmgr4GsxHDNYhgxxDFj9UNu9YkPFuIkbS+tx99HT3r00QuEuUJWUgRN7VZ2ljX022cIIQbO//ZW0Gax8YvxsSc+eABMT4mkqd3KtpI6rUMRQhPrCmoBmJ7iPunw6dGBDAk2sfoES4iE55IBrBuwWG38kFvNzNRIfI3a/SfR6XTMTI1kfUEtrR3WE79AiJOwvbSOU4eFYNT3+GZ4r01JCMOgg/UF0okJ4Q2W7S5HiQlCiem/G1+9MSUhDJNRzxq5UBaD1IaCWiICfEiPDjzxwQNEp9MxIyWCjYVyPeutZADrBrYU11HXamFuRrTWoTArLZJWi42NRWatQxFerLHNQm5lE+OGH7/a58kKMhkZPSyk6w6xEMJz5VY2sa+ikfNHDdE6lC5+PgYmxYexOr+a4y3JEsIb2ex2NhaZyUwMP+YWO1qZkeq4nt0i2RFeSQawbmB1fjUmo57MxDCtQ2FifBiBvgZWSRqx6Ec/HajHDowdHtLvn5WZFM6+8kbMzR39/llCiP6zbHc5Br2Os07R/mbvoaanRFBibqWwtkXrUIQYUDkVTZhbOsjUcPnbsUyIC8Vk1LNebmB7JRnAasxut7M6v4bJCWH4+Ri0Dgcfg55pyRGsyqvGapO7yaJ/7D7YgA4YNezYxZtcZWpSOHZgY5F0YkJ4KovVxld7y5mREkH4cfZ11oJz7Z+kEYvBZkOho191hwmY7vx8DIwfHipLiLyUDGA1VlDTwoG6Vrda/D4rLZLalg62F0sasegfew42kBQRQKBv/+/kNWJIMCF+RrkLK4QH21Bopqa5g/NGuk/6sNPQED/SowNZky+ZS2JwWV9YS2pUAFFBJq1DOaqpyeEU1LRwsL5V61CEi8kAVmPODm9asvsMYE9LjsCo17F8b7nWoQgvZLfb2XOwgZFDB6YIi0GvY0pCGBsKa2WNmhAe6tvsSoJMBk5zo77yUNNTItheUkd9qyxVEINDa4eVHaV1bpk+7JSV5IhN6mB4HxnAamzt/hrSogIZGuKndShdgkxGJsWHsWJfhVzwC5crb2ijprmDkUP7P33YaVJCGBWN7RSb5S6sEJ6mw00q9R/PtOQIrHbYJAUQxSCxo7SedqudKW48gE2OCCAmyFcGsF7IPXuCQaKh1cL2kjq3Sh92mpUWSUF1MwU1UpRCuNae8kaAAR3ATox3rM/ZLGnxQnicTUVmGtoszEl3r+JNhxo1LIQgk0EulMWgsbW0Dr0Oxg1AMca+0ul0TE2OYFNRLRap6+JVZACroQ2FtVjt7rX5s9PM1EgAvs+t0jgS4W32HmzAoNeRHj1w+zgmhvsTHeTLFpkdEcLjrMyuItDX0JUO6I6Meh2TE8JZXyBLFcTgsKO0DiUmaEBqWZyMqUnhNLZZ2V1Wr3UowoVkAKuhjUW1BPoaGDXM/e5exQSbOHV4CKvypCiFcK09BxtIiwrENICpgDqdjonxYWwuNsvFpRAexGK18X1uFdNTIgb0nNEXWYlhlDe0USiZS8LLdVht7Cpr6Pe93F1hckIYeh38KNkRXsW9ewMvt7nIzIS4UIx699r82emMU4awq6yBysY2rUMRXsJut7O3vHHACjgdalJ8KDXNHeyvaR7wzxZC9M2WkjrqWi3MzXDf9GGnrCRHNtW6QrlQFt5tb3kjbRabW6cPO4X4+TBqaIjsROBlZACrkYP1rRSbW5nsxovfzxgRA8BqmYUVLlJsbqWhzcLIIQO3/tWpax1sUd2Af7YQom9W5VZjMuqZ6sbpw06xoX4khPuzQS6UhZfbXuLoR8d6wAwsOLbT2XuwAXOzVAn3FjKA1YizUuHkePfb/NkpPSaIuDA/fpABrHCRvQcbgIEt4OQ0PNSPocEmtkghJyE8gt1uZ3V+NZMTwvDzMWgdTo9kJYazpdhMu8WmdShC9JttpXUkhPsTGeirdSg9MjUpHDuO2jPCO8gAViObisyE+/uQGhWgdSjHpNPpmJkayaYiM03tFq3DEV4gp6oJo15HSuTA/93rdDomJoSxpdiMTdbBCuH28qqaKatvY0ZnUUFPkJUUTqvFxvZSyfQQ3slmt/PTgXrGe8jsK8CIIcGE+BllAOtFZACrAbvdzuZiM5MSwtDp3HP9q9OstEg6rHbW7ZdGL05ebmUTyZEBGA3anHomxYdS12ohr6pJk88XQvTc6nxH9s8MN6zUfywT48Mw6nVyoSy8Vn51M/WtFsbFuf/6VyeDXseUhDA2FEqVcG8hA1gNFNa0UNnYzuQE900fdhoTG0qYv49spyNcIqeykbSoQM0+f1LXfrAyOyKEu1udV82IIUFEB5m0DqXHAnwNjB0eIvvBCq/lXP/qCRWID5WZGE5FY7sUcvQSMoDVwEbn+lcPGMAa9Tqmp0Tw4/5aLFZZ0yP6rq6lg4rGdtKjtRvADg3xY3ion+wHK4Sbq2luZ1dZg0elDztlJYaTU9lEVVO71qEI4XLbS+uICvRleKif1qH0SmZnITipRuwdZACrgc3FZoaFmDym8Z+eFklDm4WtJTJrJfoutzNtN03DASw4ZmG3ltRhtUkakRDu6sf9NdiBmSkeOIDtvFDeKGnEwsvY7Xa2ldQxbnio2y+B625YiB+J4f6S3u8lZAA7wKw2O1uKzUyKd//1r06ZieGYjHp+yJVqxKLvcisdA9h0DVOIASYmhNLQZiGnslHTOIQQx7a+oJaIAB8yYrQ9X/RFRkwQ4f4+kkYsvE5ZfRsVje2M96D1r4fKSgpnS3GdVAn3AjKAHWA5lY3Ut1qYnOj+6cNOfj4GMhPD+SGvWha/iz7LrWoi1M+oedl9WQcrhHuz2e1sKDSTlRTuMTd6D6XX6ZiSGMbGwlqpeC68irO6tqetf3XKTAynzWJjxwHp/z2dDGAHmHP/10luvP/r0cxKi6S8oQ21QmatRN/kVjWRHh2o+QVpdJCJxHB/2Q9WCDeVXdGIuaWDzMRwrUPps6lJEdQ0d5BTIRXPhffYXlpHoK+BVI0zqfpqQnwoBr2O9QXS/3s6GcAOsE1FZpIjAjyqqiI4tjHQ65A0YtEnNrud3Mom0qKDtA4FgEkJYWwrqcMi62CFcDvOIitTPHgAm9mZZbWuoEbjSIRwne0l9YwdHoJB73mZEQCBvkbGxIbIOlgvIAPYAdRhtbGtpI5JHlB9uLvwAF/GxobwQ54MYEXvlZpbabXYNF//6jQxPoymditqeYPWoQghutlQWEt6dCBRGi83OBlRQSbSowPlQll4DXNzB/trmj02fdgpKzEctaKRmmapEu7JZAA7gHaXNdBqsXnE9jlHMzMtipzKJkrrWrQORXiYHDepQOw0Ic7RAcs6WCHcS0uHle2l9WR58OyrU1ZiONtL62lut2odihAnzbn+dbyHD2Azu6qESxqxJ5MB7ADaVGRGx88Xz55mVud+fKvyJCVK9E5uZSM6ICUyQOtQAIgM9CU5MoDNsg5WCLeytdiR2u+8yPRkmUnhWDp3HhDC020vrcfHoGPE0GCtQzkpp8QEEepnZL1kR3g0GcAOoE3FZk4ZEkSov4/WofRJfLg/KZEB/JBbpXUowsPkVjUTH+6Pn49B61C6TIoPY0dpHRarlNMXwl2sL6zFZNR7fJoiOCq1+hn1XWt6hfBk20vrGDU0GJPRs4cOBr2OyQnhbCiolZ01PJhn/xV6kJYOKzsP1Htc9eHuZqVFsr2kjrqWDq1DER4kt7KRdDdJH3aaGB9KS4eNPeVSWVsId7GhoJbxcaEef5EMYDLqmZQQJoWchMdr6bCyr6LRK24sAWQlhVHV1E5edbPWoYg+8vwewkPsKHWkRXnS/q9HMystCqsd1u6XDln0THO7lRJzK2luUsDJaWKcoy1Kep8Q7uFgfSv7a5q9Yv2r09SkcIrNrRTXSu0I4bl2HqjHarMzzkOXwHXn3KJrg2RHeCwZwA6QTUVmjHqdx9+9GjEkiOggX76X7XRED+VXN2EHt5uBDQvwIS0qUAawQrgJZ1EVb1j/6pSVFAHAOrlQFh5se2kdOmBsbIjWobjE0BA/kiL8ZR2sB5MB7ADZVGTm1GHB+LvRGsC+0Ot0zEyNZH1BDa0dUllRnFhOpXtVID7UxPhQdpTW0yHrYIXQ3PrCWqICfUl1k2JvrhAf5sfwUD/WSxqx8GDbS+tJiw4kyGTUOhSXyUwMZ1tJHW0W6f89kQxgB0B9awf7yhs9cv/Xo5mVFklLh42NRTJzJU4st7KJAB8Dw0L8tA7lCBPjw2i12NhdJvvBCqElm93OxsJaMpPC0el0WofjMjqdjqlJ4WwuNtMuF8rCA1msNnYeqPf47XO6y0oKp81i69oeSHgWGcAOgK3FddiByQnekRY1KT6MIJOBlTlSjVicWG5VE6lRgejd8KJ0QlwoOpDtdITQmFrRSF2rhUwPrxNxNFOTI2jpsLHjgFwoC8+jVjTSarF5zfpXpwlxYRj1OlkH66FkADsANhWZ8TPqGT3Ms/fOcvIx6JmZGsmq3GpJvRTHZbfbya1qcrv1r06h/j6kRQeypUQuLIXQknOrmSlecqP3UJPiHRfK6/bLhbLwPNtK6wEYN9w71r86BfgaGDs8RNbBeigZwA6ATcVmxsWF4mPwnp97Tno0DW0WmbkSx1XR2E59q8Ut1786TYoPY+eBeknvE0JDGwpryYgOJDLQV+tQXC7A18C4uFAp5CQ80vaSOoaH+hEdZNI6FJfLTAwnp7KJqqZ2rUMRveQ9Iyo3VdXYxv7qZiZ7+P6v3WUlhRPgY2BltqQRi2PL7SzglO5mW+gcamJ8GG0WGzvL6rUORYhBqbndyo7SerK8qPpwd6clhZNb1URFQ5vWoQjRYza7ne2ldV6XPuzkPOdslFlYjyMD2H62udiRmujp+792ZzLqmZ4SwQ+51Vhsdq3DEW4qp7IRgFQ3HsCOjwtBh2OtuhBi4G0tMWOx2bv2ZvRGzgvl9TILKzxIYU0Lda0WxntZ+rCTEhNEqJ+RDTKA9TgygO1nGwtrCTYZyYgO0joUl5uTEUVtSwfbZf2gOIbcqiaGBpsI9nPf0vshfj4oMUGSDi+ERtYX1GIy6hnrZVVOD5UWFUhUoK+kEQuPsq2zQu84L22bep2OzMRw1hfUYrfLZIwnkQFsP7Lb7WworGVyQhgGvftVYD1ZpyVHYDLqpRqxOKbcqia3Xv/qNDE+jJ1l9bK3sRAa2FBYy4S4UExG770kcW6ns7GoVrKWhMfYWmwmIsCHhHB/rUPpN5lJ4dQ0d5Bb1aR1KKIXvLe3cAMFNS1UNLaT6aXrevx9DJyWHMF3OVXY5M6V6KbdYqOgpsVtKxAfanJiGB1Wu+wHJ8QAO1jfSkFNi1evf3WamhxBfauFPQdl32nh/ux2O9tK6pgQF+ZVezN351y6IOn9nkUGsP3IWZrbG/e1c5qTHkVVUzs7D0gBHHG4gppmrDY7aW68/tVpQlwoPgYd6wskjViIgbShq5/0/gHslIQw9DpYu79G61CEOKHSulYqGtuZEO+d6cNOQ4JNJEcGyDpYDyMD2H60sbCW+DA/hod6b+rF9JQIfAw6SSMWR3Cm46R7wPpvfx8D44aHsr5QLiyFGEjrC8xEB/mSEhmgdSj9LtTfh7GxIazJq9Y6FCFOyFnYcIKXViA+VFZiONtK6mQZkQeRAWw/6bDa2FJs9vq7ykEmI5mJ4azMrpIF8OIwOZVN+Bp0xHvI2pmsxHDyqpqpbJRtLoQYCFabnU1FtWQmhnt1iuKhpqdEkl3ZRLlspyPc3NYSM2H+PoPi5lJmUjjtsozIo8gAtp/8dKCelg7boFjXMyc9ioMNbewpb9Q6FOFGciubSI4MxOghBcxkmwshBta+ikbqWi1kefmN3kNNT40AYG2+zMIK97a1pI7xcaGD4uaSLCPyPDKA7ScbCmsx6BzVTb3dzNRIDHodK7MljVj8zFMqEDulRQcSEeAj62CEGCAbOm8WTfHiOhHdJUcEEBvqx+p8Wa4g3FdZfStl9W2DIn0YHMuIxg4Plf7fg8gAtp+sL6hl9LAQgkzuu/+lq4T6+zA5PoyVOZWSRiwAqG1up6qpnXQPKODkpNfpyEoKZ0OhWapqCzEA1hfWckpMEOEBvlqHMmB0Oh0zUiLYVGSW9XbCbQ2m9a9OWYnh5FY1USXLiDyCDGD7gbmlg33ljV67fc7RzM6IosTcSk6l7KMlfi7g5EkzsOCohGpu6UCtkHR4IfpTU7uFnw7UD6p+0ml6SgRtFhubiyVdUbinrSVmQvyMHteHnwznUob1MgvrEWQA2w82FZmxw6Ba13N6WiR6HVKNWAB03cjwhC10DiX7wQkxMLYU12G12QdVP+k0IS6MAB8Dq/MkjVi4p60ldYwbHop+EKx/dUqPcSwjWrdf+n9PIAPYfrChoJYgk4ERQ4O1DmXARAT4Mj4uVNbBCsBRwCkiwIfIQM9KDYwM9CUjOlAGsEL0sw0FtfgZ9YyJDdE6lAHna9STmRTOmvxqWXYj3E5FQxsl5tZBlT4MjmVEU5MjWF9Yi8Um7dLdyQDWxWx2O6vzq8lKjPCY6quuMic9iv01zeRXSxrxYJdb1eRxs69OWUkR7DhQT1O7RetQhPBa6wpqmJQQhq9xcF6GTE+JoKKxnWxZdiPczNaSzvWv8YNrAAswPTmC+lYLu8vqtQ5FnMDg7Dn60d6DDdQ0dzCjs1T+YDI7PQpAZmEHOavNTn51s8eunTktORyrzc6GQlmfJkR/KK5todjcytRBuP7VaVqy4xphjWynI9zM5mIzgb4GMqKDtA5lwGUmhmPQwRqpEu72ZADrYqvyqjHofu6cBpPoIBNjYkNkHewgV2xuoc1iI91DB7BjY0MI8TOyOk8uLIXoD+s6U/SnJg2+ftIpMtCXUUOD5UJZuBW73c7GwlomJ4RhGGRZhADBfkbGDA9l7X5pl+5OBrAutjq/hjHDQwn199E6FE3MzYgip7KJ4toWrUMRGsntTIlLj/LMu7dGg57TkiNYk1+DVdbBCOFy6wpqiAvzIz7cX+tQNDU9JYLdZQ1UN7VrHYoQABSbHfu/Zg7C4mpO05MjyKlsorxBttNxZzKAdaGyesc2MjNSBu9d5a40YpmFHbRyqpow6CApMkDrUPpsZmok5pYOdsk6GCFcqt1iY3OReVDPvjrNSI3EDjLbI9zGhs4tZAbzAPa0zmv4H6VdujUZwLqQM+VwZmqkxpFoZ1iIHyOGBMkAdhDLrWwiITwAkwcXZ5maFI5Rr2OVpBEL4VLbS+totdgG9fpXp4zoQGKCfFmVK+cZ4R42FtYSG2IiLsxP61A0kxoZwNBgE2slvd+tee4Vphv6IbeahHB/EiM8d+bJFeZmRLPnYANl9a1ahyI0kFvZ6LEFnJyCTEYmxIXKAFYIF1tXUIuPQcfE+DCtQ9GcTqdjdnoU6wtraemwah2OGOQsNjubisxMSQxHN4j2f+1Op9MxLSWCjUW1tFtsWocjjkEGsC5ibu5gS7GZOZ0ptIOZ8zf4TmZhB53GNgsH6ts8toDToWamRlJQ00JhTbPWoQjhNdYV1DBueCgBvgatQ3ELp6dF0WaxdRW2EkIru8vqaWq3Dur0YadpyRG0dNjYVlqndSjiGGQA6yLf5VZhtcMZGdFah6K5+HB/0qIC+V4GsINOXpWjgJOn7gF7qFlpjqUAkg4vhGuUN7SRV9Us6cOHGBcXSqifUW74Cs1tLDSjAyYlSHbEpIQwfA06SSN2YzKAdZEV2ZXEhfmREeP5F+6uMDs9ku2l9VJdcZDJdQ5gvWAGdmiIH6OHBcu+xkK4yPoCx8Xg1EG4zdyxGPU6ZqZGsia/mg6rpCsK7WworGXE0GDCBukuGofy9zEwMT5MCqy5MRnAukBNUzubi8zMzYge1OsGDnV6WhR2kDWEg0xOZROBvgaGBpu0DsUl5mZEs6+ikRKzbAslxMlaX1BLTJAvqR5cobw/nJ4eRWOblS3FZq1DEYNUY5uFXWX1ZCbK7KvTtOQIimpbZFtINyUDWBf4dm95Z/qwrH91So8OJDbUj+9zZfZqMMmuaCIjJshrbuQ413OvkFlYIU6KxWZnQ6Fj+xxvOT+4SmZiOP4+er7LkRu+Qhtbiuuw2gf39jndTevcTkdmYd2TDGBd4ItdBxke6ocSE6R1KG5Dp9Nxelokm4rMNLZZtA5HDACrzU5uVSMZXpA+7BQb6sfIocGsyK7UOhQhPNrusnoa2ixMTZYL5O5MRj3TkiP4Ia8am92udThiENpQWIufUc+pw0K0DsVtxIX5kxTh37VFpnAvMoA9SRUNbazLr+bsETFyV7mbOelRdFjtsgh+kCgxt9DSYSPDy27knJERxd7yRkrrJI1IiL5aV1CLQQdTEmQAezSz06Oobmpn54F6rUMRg4zdbmdNfjVTEsPx9eD92/vDzNRItpTU0dAqEzHuRv5ST9LX+yqw2+HckUO0DsXtnBobQkSAj6QRDxLZlY4CTkq0dw1g53ZWFv96r8zCCtFXa/NrGD0shGA/o9ahuKXTkiPwMegkjVgMuLzqZsrq25ieIsXVupuVFoXVZudHSSN2OzKAPQl2u50v9pQzLj6UhHB/rcNxO3qdjllpkazdX0ObbAbt9bIrGjHodSR7WYGW2FA/xseF8uWecuyS3idEr5U3tLGvopGZqZFah+K2gkxGMhPDWZFdKecZMaCcKbIygD3S6GHBnRMxcmPJ3cgA9iRkVzaRV9XMxWOHax2K25qdHkVLh40NhbJJu7fLqWwiOSLAK1OQzh0RQ2FtC3sONmgdihAex3mBLAPY45unRHOwoY2dZXKeEQNndV4NI4YEER3kHbsHuJJep2NGaiTrCmpol4kYt+J9V5oD6Ms95Rj1Os49dajWobitSfFhBJkMfC+btHu97MpGr90H+QwlGl+Dji/3VGgdihAeZ1VeNQnh/iRGSKbS8cxMjcTXoGO5KssVxMCobW5nV1k9M1Lk5tKxzEqNpKndypYS2ebKncgAto/aLTa+2lPB9JQIwgN8tQ7HbfkYHNUVV+VVY7FJWpS3qmlup7KxnQwvW//qFGQyMjM1iq/3VdBhlbuwQvRUU7uFzcVmZqRESqHDEwgyGTktOYIV2ZVSjVgMiLX7a7AD01MlffhYJieE4WfU84OkEbsVGcD20Xc5VdS2dHDp2GFah+L25qRHUddqYUdpndahiH6SU+Eo4OStM7AA542Koa7Vwo/7JR1eiJ7aUFBLh9XOzDS5QO6JeUo0lY3tbJf+UgyAVXk1RAX6coqX7R7gSn4+BqYmR7BatrlyKzKA7aNPdhwgLsxPNn3uganJEZiMer6TNGKvlV3ZCEC6l87AAmQlhhPu78OXe8q1DkUIj7Eqr5pQPyNjYkO1DsUjTE+JxGTU860q/aXoXy0dVn7cX8PpaZIdcSKzUiOpaGxnb3mj1qGITlLPvg/yqprYVlrPrTOTaWlu4qGXFrNy5Ura29uZOnUaCxf+ifDwI+82WywWbrjhatLSMrj33ge7Hl+y5D0++mgJJpMfN998K2eccVbXcx9/vJT9+/O56677jxvTY489SEVFBS+++OoRzz355COUlBTzyit/B+Cyyy7g4MGyruf1ej0BAQGMGjWGG2+8hfT0DADOPPMMDhw40HWcj48PUVExzJo1m+uuu4GAgJ7Ntvn7GMhMDOf73GrumJ0qJ0ovpFY0MiTYRJi/DwBNTY28+upLrFmz6qTbhb+/PzfeuNAt2sVZI2L49Ikbmf70z7OwfW0XQng7i83OmvwapqVEYNQf+7zvyvPFyfSj9957DyUlB1zaj3Y/7kTniwBfA9NTiSzv/wAAIABJREFUHGnEd8xOxXCc302Ik7Guc4eIORlRgPu0w970292vU13VDrublhKBQQercqsYNTT4uMeKgSED2D74dEcZPgYd548awv333U5xcSH33PMA0dExvPnmqyxceCPvvPMhvr6Hr419663XycnJJi0to+ux/Pw8PvjgHZ5//hXMZjMPPngPU6ZMJSQkhKamRj76aCmvv/6Oy7/DggXX8Mtf/gpwbAdUXV3NCy88ze2338LHH3/W1ZgPPa61tZV9+/bwyisvsHv3T7z00hv4+Pj06PNmp0eyKq+aveWNjJTG73WyK5vIiP65A7j//rsoLCxwSbuwWFpYtOgOt2gX542M4VNg4hnz+evCG4CTaxdCeLOdB+qpa7WcsPqwK88X3tCPzlOiWZFdxdYSM5MTJMtL9I+VOVWE+fswPi4MkHZ4vH47zN+HscND+SGvmpumJ7v8O4jekxTiXmput/LFnnLmZkRTVbqfjRvX8/DDj5CZOZWUlFTuv/8RqqoqWbHim8Ne99NP2/nii/+Smpp22OP79+eTkpLK6NFjmD59JgEBgZSWFgPwj38sYc6cMxgyxPVVjv39/YmMjCIyMoqoqGgU5RT+8Ic/Ultbw5Ytm4563PDhccydeyZPPvkcu3bt5Isv/tPjz5uREolBh6QRe6HWDitFNc2kd66hyclR2bhxPXfddb9L2sXpp892m3ahxATho9eTU9tBRETkSbcLIbzZqrxqjHrdcZfauPp84Q396LTkCPx99HyzT6oRi/7RZrGxOq+GWWmRGPU6aYc96LdnpUWSV9VMibnF5d9B9J4MYHvpm30VNLVbuWzsMIqLHQ1zwoSJXc8HBAQQHx/Ptm1buh5rbm7m0Ucf4I9/XERY2OGpGLGxsZSUFFFTU83+/fk0NjYQEzOEmppqli37N1dddd3AfDHAYDAA4ONz/KrKinIKY8aM49tvvznucYcK9fdhfHwY3+fKANbb5Fc3Y7WD0jkD62wXY8aM6zrmZNpFXl6u27QLnU5HgK+BmuYOfjpQf9hxfWkXQngru93O97lVnVupHTvZy9XnC2/oR/18DJyeFsWK7CraZO9J0Q82FNbS3GFlbmf6sLTDE/fbs9IcmSSr8qQasTuQAWwv2O12/rWjjNSoAMbEhhAV5Wj45eUHu46xWq1UVFRgNv+8Ru6ll55lxIiRzJ175hHvOWLEKGbPPoOLLz6Ha6/9Nddf/3siI6N49923uPDCSwkLC+v/LwaUlpbw+uuvEBkZxamnjjnh8SkpaeTn5/XqM2anRVFQ08L+6ua+hincUHaFo6hBRucMrLNdVFb+vGfqybSL+fMvdat24e9jwNeg49Ofyo44vi/tQghvpFY0UmJu7bpAPhZXny+8pR89b+QQGtosrMmXi2Xheitzqgg2GZkU72gb0g5P3G8PD/UnLSpQMgndhKyB7YWNRWbUikbumZeOTqdjxIhRJCYm8fDDD3HXXQ8SHBzE22+/gdlcS0dHBwBr1qxi3bq1LFny8THfd9Giu/n972/BaDTi7+9PaWkJq1Z9xz//+QnLlv2HJUveJSQkhL/85X7S0tKP+T7bt29h3rwZRzze3t7OqaeOPeyx999/m6VL3wcci/EtFgvp6QqPPbaYwMATV5INDg6mqal31dhOT4tk8cpcvs+tIjkyoVevFe4ru7KJwP/f3n2HV1GmDRz+nZzknPTeE0ggwEAgVGkK2FHEBhZsa2/Y1ro2LLirrrh25QNcRKwodlFWmg0VkBZqhtACIb33nPr9MUkklQSSnJLnvswVmcxJninvO/N2g57YIG+AhnTx4ovP88QTz5xwuggLC8BkwmnShU4H/SP8WKXmc99pSQ0TV8HxpQsh3NFKtQC9Dk7r33YBtrPzC3d5jp7UO5gIfwPf7dSGLAnRWWrMVn5KL+CM/uF46bV2LEmH7Xtun6WEM++3DHLLa4kKMLbrM6JrSAG2A97bcJgwPwPnJUcB2ixmzz33H5599ikuvvhcvLy8OPvscxk//hQ8Pb0oLi7mhRf+xaOPPklgYNtLCAQE/DWx0YIFc7nyymuoqqrizTdf4d13l7Bz53b++c8nWbz441Z/R3LyYB5/fHaz7fPnv0VxcVGjbdOnX8a0aZcBWleLoKCgDs2eWlVV2Sjm9ogMMJISE8AqNZ8bxkoB1l3szi1nQKQ/HnWzS9eni2eeeaJT0oWvry8mU5VTpYvk6AC21dj5flcuV42Kb9h+POlCCHdjt9tZvSef0QkhjSp4WtLZ+UU9Z8ovWtNWfqH30DFlUCQfbjpCUZWJUN+2u0IK0V5r9xdRabJy7qDIhm2SDtv33D5biWTebxmsUvO5+qT4Y39AdBkpwLZTWm45Gw6VcNfEPhg9/+p5nZCQyKefLiUjIwsvLy98ff248carGT16HOvW/UZxcRFPPfVow/4mkwmAn35azcqVvzb7O3v2pLF9eyqPPfYU69b9Tq9eCURHRxMSEsxTTz1KZWVFqzVKBoM38fG9mm338/NrluADAgJb3Le9VDWNAQMGdvhzZykRvPLTfg4WVZEY6nvcf184B7PVhppXwYwRcY22JyQksnDh+5SWlrhlugjzM5ASEsjnqdlcOTKuYWmo400XQriTtLwKjpTWcGM7KyrdPb9ozbHyiynJUbz3ZyYr0vK5YmRcq/sJ0RH/251HuJ+BUb0ad+mVdHhsvUN8GBTlz0opwDqcjIFtp/f+zMTPoOeSYTEN26qqKrnrrltJT08nKCgYX18/srOz2Ls3nTFjxnHqqaezZMmXLFr0UcNXcvJgJkyYxKJFH7X4d+bNe5MbbrgFo9GITqfDbrcDWrcIAJvN3vUHewzp6XvYsWMbZ599boc/e9aACHTASlVmV3QHewsqMVvtjZZGqk8X+/fvdet0ccmwGA4VV7PxcAlwYulCCHeySs1H76FrmPSkLT0lv2iqPflFv3A/Bkb68/2u3G6MTLiz0mozvx0oYvLAiEZrDEs6bP9z+2wlgp055TIbsYNJC2w7ZJZUs3pPPtecFN9oNkVfXz+sVisvvPA8d955P9XVVTz//DOMGjWaUaNGN+xzNIPBG19fvxZrizZt+pPs7CymTDkfgAEDBnLgwD42b97Irl07SEzs0+3dE6urqyks1Aas19TUsH17KvPmvcmwYSOYPHlKh39fZICR4fFBrEzL5+ZxvRtaroRr2pVTDkBy9F+1qPXp4rXXXubeex884XSxfv06p0wXI8J1BNqreHfNVgqjq04oXQjhLux2O6vUfMYmBBN0jO7D0Pn5hbs9R6cOjuKlH/exr6CSpPD2d4sUoiWr0wuw2OxMOar7MEg67Mhz+ywlgtd/OcBKGQ7nUFKAbYcPNmai99BxZQtdeGbPfo4333yJ22+/AaPRm1NPPYOZM+8+rr8zb94b3Hzz7Xh6apclOjqaO+64hyeffISgoGBmzWo+HqCrffjh4obB8b6+fkRHRzNt2qXMmHF1wzTlHXW2EsGc1XvZV1BFvwh5ILuyndnlBPt4ERvo3Wj77NnP8corczolXbzyyitOnS62ehopjIs94XQhhDvYlVtBVlktN49PaPdnOjO/cLfn6OSBEbz6836+25nLPaf27Y5whRv7365cEkN9UCKbd92VdNg+MYHeDI0NlAKsg+nqm/Rbkp9f3qnt+8HBvpSUuNYSKrnltUxfuIEpyVHMmjygxX1c8bjao6uOq6jKxJR567h+TC9mTujT6b//WFzpekVEBLSribqz02p7XbF4I1EBRl6bntJlf8OZr1dhpYkL3l7PhUOieeSs1mdU7AhnPt6u4C7H2960Cl2XXp3hXL72836WbD7CDzPHEeh97BZYZ+IM568lD3y1k5055Sy7ZQyeeucd+eWs56+pnppWM4qquHTRRu6a2Ifrxhz/mFFn4qh77pPNR/jPj/v49PqT6BPmHvO5OGP6bSutOm9O6CTeWXcIm512T0Yhji3U18BJvYJZoebTVgWKcG5VJisHCqsYHN1zZ90N8zMwZVAky3bmUlJtdnQ4QjiUzW5npZrP2IQQlyu8OrNpQ6MprDTxyz5ZE1Ycv2925KDXad3SxYk5U4nAQwcr1bxj7yy6hBRg25BZUs3XO3K4OCW6YY1L0TkmD4wgs6SGtDxZM9NVpeWVY7PTaAKnnuiqUfHUWmx8kZrt6FCEcKhNh0vILa/lvOTIY+8s2m18YijRAUa+2CZ5jDg+FquNZTtzmdA3jHA/WZLpRIX7GRjZK5j/7c6ThhgHkQJsG97+IwNPDx03jZPW1852Wr9w9B46ftgtsxG7ql05WuVDTy/AJoX7MT4xhE+3ZmGy2BwdjhAO893OXPyNeiYlHXv2YdF+eg8dF6VEsz6jhMPFMvOp6Lhf9xdRVGXmopRoR4fiNs5PjuJwSQ1bjpQ6OpQeSQqwrdhfWMnyXXlcPjyWcH+jo8NxO0E+XoxPDGGFmofVCaZSFx23M7uc6AAjob5Sm3v1qHgKK038b7d0JxI9U5XJypr0As4aEIG3l0xk1tkuSolGr4OvtksrrOi4r7fnEO5nYHyfUEeH4jbOHBCOn0HPN9tzHB1KjyQF2FbM/y0DX4Oea91koLszumBwFPkVJtZnFDs6FNFBdrud1KxShsYGOjoUpzAmIRgl0p9FGw5hkQoZ0QOtSc+n2mxjarKMr+sKEf5GJiaF8fX2HGrMVkeHI1zIkdJqfj9QxIUp0Xh6yNKFncXbS885AyNZtaeAilqLo8PpcaQA24LUI6WsSS/gqlFxBLdjHTtxfCb0DSPI25NlO2WRdleTXVZLfoWJ4fFBjg7FKeh0Om4e15vMkhp+kFZY0QMt25lLfLA3w+KkUqurXDEyjtIaC99LHiM64POt2XjoYPrQGEeH4nYuTImm1mJjRZqkye4mBdgmbHY7L/24jwh/A9ecJK2vXcng6cE5AyP5eW8BZTUyg6sr2Vo35mO4vKw2OLVfGP0j/HhnvbTCip7lcHE1mw6Xcv7gKHQ6aeHpKiPjg1Ai/Vmy6YhMHCPapcZs5ZsdOZzaL5yoABkO19mSo/zpF+7HV9tzJE12MynANvH9rlx251Zw18Q++BpkHE9XO39IFCartvSCcB1bj5Tib9TTN8zP0aE4DZ1Ox83jEzhUXC21saJH+XJbNnodXDhEJojpSjqdjqtGxXGgqIo/DsrQG3FsK9V8SmssXDY81tGhuCWdTsf0YTHszq1ge3a5o8PpUaQAe5Qqk5W3fj3I4OgAzh0kywB0h4GR/iSF+/K1DIJ3KVuPlDE0NhC9jKdp5LR+YQyI8GP+bwdlRmLRI5gsNr7dmcvEpDAiZMLDLne2EkG4n4GPNmU6OhTh5Ox2O0u3ZtEnzJdRvWS4T1eZmhxFgNGTjzcdcXQoPYoUYI/y7oZDFFSaeOD0JDykG1S30Ol0TEvRaq925UjtlSsoqTZzoLCK4XHyQGzKQ6fjnlP7klVWy6dbsxwdjhBd7qe9BZRUm5k+TMbXdQcvvQczRsSyPqOEnfLMFG3YnFnK7twKrhgRK137u5CvQc+0odH8mJ5PTlmNo8PpMaQAW+dwcTUfbszk3EGRpMjMqt1q6uAovD09+DxVXvhdwbasMgApwLZibEIIJ/cJ4Z11hyitlrHdwr19nppNbJA3YxNCHB1Kj3Hp8FgCvT1ZtO6Qo0MRTuyDjZkE+3hxnswM3uXqu2h/ukXeY7uLFGDRulk8tyodL70H90zq4+hwehx/oyfnDIrkh7R8mczJBWzNLMVLryM5OsDRoTituyf1pdJk4e0/MhwdihBdZm9+JZszS5mWEi29lrqRv9GTK0bE8fO+QtLzKxwdjnBCBwurWLu/iMuHx8q6zN0gOtCb0/tH8MW2bHmP7SZSgAW+3ZnLxkMl3D2pj4zhcZBLh8VQa7Hx3S6Z/MbZbc4sZVBUAEZPyT5a0y/cj+lDY1i6NYvdudLNT7inJZuPYPT04GJZnqPbzRgZi59BzzvSCita8OGmTIyeHlw6XNJmd7lxXC8qTVY+krGw3aLHv4EWVZl47ef9DI8LZJo8hB1mYFQAQ2IC+GxrFlZZgsRpldWY2Z1bztiEYEeH4vTunNiHEF8Dz61Il2V1hNsprjKxfHcuU5OjZL10Bwj09mLGyDhW7SlAzZVWWPGXwkoT3+/S0maIr8HR4fQY/SP8OWtAOEs2H6FEhg91uR5fgH35x31Um608dvYA6QLlYFeNiudQcTU/7y1wdCiiFRsPlWCzI+Pd2sHf6MkDpyeRllfBp1ukRla4ly+35WCy2pkxUpbncJS/nRRPkLcnb6494OhQhBNZujULs9XOlaPiHB1Kj3Pz+ASqTFY+2CizhHe1Hl2A/e1AET+k5XPDmN70CfN1dDg93hn9w4kP9mbxn5myILSTWpdRjJ9Bz+AYmeisPc4aEM6EvqHMXXuQ/YWVjg5HiE5hsthYujWLcYkhsha0A/kbPblhbG/WHSxmQ4asCyu05SA/25rFxKQwEkPlvba7JYX7MXlgBJ9sPiIzEnexHluArTJZeWFVOn1CfbluTC9HhyMAvYeOa06KZ1dOOZsOlzo6HNGE3W5n3cFiRvcOxlPWf20XnU7H45MH4OOlZ9Z3abI2rHALy3blUlBp4ppR8Y4Opce7dHgsUQFG3vz1ADap+O3xPk/NorTGwvXyXuswd0zogx149ef9jg7FrfXYAuzrv+wnp6yWxyf3xyCT0TiNqclRhPp6sXjDYUeHIpo4XFJDdlmtdB/uoHA/A0+cM4D0/Erekq5+wsVZbHYWbzjM4OgAxshYeIczenpwx4REdudWsGxHrqPDEQ5UY9a6ro5NCJblIB0oNsibG8f2ZvWeAtYdLHJ0OG6rR5bc1h8s5vPUbK4aFc8wWcvSqXh76bnmpHjWZRSz6XCJo8MRR1lf10VtXKIUYDtqUlIYlw+P5aNNR/hht8y0LVzXirQ8skpruHFcb3Qyb4RTmDIokmGxgbz56wFZwqMH+2JbNkVVZm4al+DoUHq8a06Kp1ewNy+u2UeN2erocNxSjyvAVtRaeOYHlcRQH24/RRK5M7pseCyR/gbe+OWAjIV1IusOFhMb5E18sI+jQ3FJ957WlxFxgfxzxR6ZNVS4JKvNzqL1h+gf4cfEvqGODkfU0el0PHRmP0przCz4Xdae7olqLTbe/zOTkfFBjIiXhhlHM3h68PBZ/TlUXM1baw86Ohy31OMKsC/9uI+CShNPn6vI4s5OyttLz22nJLIzp5w16TIjsTOoMVtZn1HMydL6ety89B48f0EyQd6e3P/VDpngQbicH9MLOFhUzQ1jpfXV2SiR/g1rT+/ILnN0OKKbfbMjh4JKEzeN6+3oUESdsQkhzBgRy5LNRxp6sInO06MKsL/uK2TZzlyuH9NLZlF1clOTo+gb5stbvx6gVia+cbg/DhZTa7FxxoBwR4fi0sL8DLw8bQiVJit//2KHdPcTLsNut/PO+kMkhPhwRn/JB5zRnRP7EOFv5OnlqnRb7EFMFhuLNxxmaGwgo3vLuHRnctfEPiSG+vDM/1SKq0yODset9JgCbEm1mWdXptM/wo+bx0vXYWen99Bx/+lJHC6p4Z31hxwdTo+3Jr2AIG9PRsTLw/FEKZH+vHhRMoeKq3nw611SQSNcwpr0AtLzK7lhbG/0Mgu5U/I3evLE5AFkFFcz7zfpStxTfJaaRW55LbeOT5CeEU7G20vPv84bRGmNhYe/3Y3FKs/7ztJjCrAvrt5LabWZp89V8NL3mMN2aWMTQpiaHMniDYfZWyBraDqK2Wrj132FTEoKk+VzOsno3iHMnqKwJbOUJ75Pw2KTsd7CeVmsNuauPUjfMF/OHRTp6HBEG8YmhnDJsBg+2pRJ6hFZjs7dVdRaeGfdIcb0DmasDPFxSkqUP49P7s+WzFJe+nGfo8NxGz2iJLciLY8Vaj43j+/NgEh/R4cjOuDeU5MIMHry7Io98pLvIH8eKqHSZJXuw51s8sBI7j89iR/TC3h6eRpWub+Fk/pmRw6Hiqu5Y0IfaX11AXdP6kNMoJHZ/1OpMklXYnf2/sZMSmss3DWpj6NDEW2YMiiKa06K57PUbL7Ylu3ocNyCp6MD6Gx5ebm8/vrLbNr0J3a7jaEjx7Ih5FRS4mK4bkzLg9tvvvla0tJ2Ndp2/vkX8cgjT3RHyD3SnDnPYrPZjnmOg329eOiMJJ56/EEuWw5fvv9eN0XYsx19fdakF+Bn0DO6d+Pa3VmzHuann1Y32jZq1Bhee21ud4bqtIqKCpk793X+/HMdtbW1JCcP4a677qVv334N+1w5Mo4as5W5aw9SuOs3KratICcnm8TEvtx660xGjx7nwCMQPZnVauXtt/+P77//lqKycsITUhhy/bOt7r9s2dd8/PH7ZGdnERsbx5VX/o2pUy/sxojdX/01Wb78W6qqqhg7djz33/8woaFhjfbzM3jy5LkKdz/4AJMXpTb6meTRrqe6upp5897gp5/WUFtbw+DBQ7nrrnsJiIzno42ZnK1EMCgqoGF/eTZ3jZbeWzdsWMfcua9z6FAGvXr14vbb72b8+FNa/PxdE/uw+2AW/3nuSeaX7sfgqeeMM87i9tvvxsdHVnfoKLcqwNrtdh566F6Cg0N4/fV5WO127p41G0v1Av75wcctdn+02+1kZBzgySf/xahRJzVs9/b27s7Qewy73c7ChfP55psvOf/8i9r1maq0X9Hn7iI3PIlf6rqyiq7R9PqYLDZ+Si9gQt9QjJ6NO2wcOLCP22+/iylTzm/Y5uVl6O6QnZLNZuOxxx7Cbrfz/PMv4ePjyzvvLODvf7+DDz74lKCgv8YS3zC2N+qfP/PrkvkMOvNyPn39Wtas+ZmHH36A//znNUaOPKmNvyRE13jnnQUsX76MYdNm8sP+KsIOL2fWrIf5v/9b2Gzfn35azUsv/ZuHHnqM4cNHsmnTn8yZ8yxBQUFMmHCqA6J3T/XXZNas2QQGBvPSS//m8cf/0eI1GdUrmBBzAYWDp3LrlZdxUUo0IHm0K3rttf+wfXsq//znvwkMDGL+/Ld44IG7GXXLHMw2OzNPSWy0vzybO1dr760HDuznkUfu57rrbuK0085kxYrlPPbYgyxc+AF9+yY1/z02KxUr38BYVUP16Ou5b0oyX7/3Fo888oBULhwHt+pCXFRUSGJiIo88Mov+/QfwS4GRsl4ToDiTAF3Ls31mZR2hurqaIUNSCAsLb/jy85Ouxp3tyJFM7rnndr766jOioqLb9ZnMzMMsWDCX5MEp+HjpeXq5yv5CGQ/bFVq6Pj/tLaC0xsL5g6Ma7Ws2m8nMPExy8pBG6SYwUGb3Bti7dw87dmzj0UefJDl5CH369OWJJ56hurqK339f22z/nI3fkzDsFFIDx/JxWg3Tp1/GOedMYdGitx0QvejpzGYzS5cu4dJrbmF1aRhTTh7Ji8+/wPbtqWzfntps/5KSYm688VbOO+8CYmPjuOCCi+nbN4mNG/90QPTuqf6a3HbbnYwePQ5FGcjs2c+1ek3MZjMVhTkMTh7C21uKOVJrkDzaRf36609Mm3YpQ4cOJzGxD7feegd5ebl8+0cq01Ki6RXyV+udPJs7V1vvrUuXfkxy8hCuu+4mEhISueWWmQwZMpSlS5e0+Lt+/30tBw/u45U5LxHcawAvbzFx6/1PsXnzn2zZsqk7DsetuFUBNiwsnNmznycmJpYtmaUsXLON6IJNDBqU3Gri3b9/L0ajkejomG6OtufZuXM7sbFxLF68hJiY2GPub7Va+de/nuLqq6+lb5++JIX7YvD04J7Pd5BXXtsNEfcsLV2fb3bkEBNoZExC4+7DBw8ewGq1kpCQ6IBInV9UVDRz5rxK795/zXju4eGB3W6nvLy82f6ZmYe5/OyJzBgRy6LfM5i79iD9+w9gx45tWCyW7gxdCNLTVaqqKvmtIhSD3oO7J/YhJiaWmJhYUlO3NNv/4osv5W9/uwEAi8XCmjWryMg4yOjRY7s7dLdVf01GjBjVsK2ta1KfRz9x2UQiA4w8+u0uCiplGQ9XFBwcwurVKykuLsJsNrNs2VfojX74BEdy68mNV9WQZ3Pnauu9ddu2rY3SI8CIEaPYtq15egTIzDxEWFgYwwf24/VLUrADT/+UQ0BgEFu3bu6qQ3BbbtWFuF5BRS33PngfXpnbqA0I5OE35re67/79+/D3D2D27Fls3bqZoKAgzjvvAi6//Co8PNyqfO9wkydPYfLkKe3e//33F6HTwZVX/o05c57FoPfgtWlDuPWTVP7+xQ7+77KhBPt6dWHEPUvT61NlsrI+o4Rbxyfg0WRq/gMH9uHl5cXChfNZt+53jEYjp59+FtdddxNGo7G7Q3c6QUHBnHzyhEbbli5dgslkYsyY5uNaw8MjyMvL5YGLkkDvwbsbDjOsIE1rRamoIDhYli8S3Sc/Pw+ATQVw7+kJhPtraTo8PJy8vNxWP5eWtovbbrsBq9XKBRdc3CwNiONXf00iIhrPAt3aNanPoz/9YCH233+jtMbO9TtH8+HzjxDk59stMYvO8Y9/PM4zzzzBBRdMRq/Xo/cyUD3uFv5xRjIhvo27BsuzuXO19d6al5fXQnqMaDWPDA+PoKysjOrqahJDfXl1+hDu+GgD1rIysvLyOz12d+d2JTSL1cajy3ZjHXQuT86Zx7Chw7j33jsaMv+mDhzYT3V1NWPHjufll99k+vTLWLhwgXTdczBVTWPJkg95/PHZjSoSlCh/5lyUzOGSam5fmkqh1Ch3mUPF1eiAC4ZENfvZgQP7sdvt9O6dwIsvvsYNN9zKsmVf8+KLz3V/oC5g7dqfmT//TWbMuJrExOazRZ5zznl88cWnbNr0J0+dN5BTA4vYvW4VAGaz3OOiexWUVoBOR3JsMDNGxjVs9/IyYDK1fj/GxMTy3/++x6OPPsnq1StZsEDGdXWWmpoaPDw88PRs3O7Q2jU5Oo9+5aXXufzqGyne+SvX3feozHi2rGbVAAAgAElEQVTuYjIzDxMaGsaLL77KS6/NxxY5EJ+N73FKdPNXeHk2d5/a2hoMhsYVCF5eXq3mkePGnYyfnx9z5jxLeXk5Cf46RuSuAOAnNY/8CulZ2BEu3QL73nvv8P77ixr+fc01N5AdN5GtR8r41+WnMXlQJJNGDmH69KksX76Ma6+9sdnvmDVrNtXV1QQEaDO4JSX1o6KigvfeW8SNN94qi0I7QG1tLf/85xPccstM4uN7Nfv52IQQXr54MA98tZPbPknl9UtSiA2SSbc6kx3IKK5i7IQQogObn9tbbpnJlVdeQ2BgEKClG73eg6eeeoy7776v0SRFPd3333/LCy/8i7POmswdd9zT4j7XXHM9xcVFPPjgPdhsNhIT+5J86kXsXvUx72zO5x+TIyQvEt3CbrfzXVoR2O3MOrtvo8kPzWYT3t6tz5YZFBRMUFAw/fsrFBcXsWjR29x88+3o9fruCN2tGY1GbDYbFoulUSG2tWvSNI++J6kfpbVWfnj3Pzzx1Sb+NW1Us541wvlkZR1hzpxneeut/zJkSArP/E+leuRVxP7xCp8t/Zi7776v0f7ybO4+BoMRs7nx/Dpms7nVSWADA4P4979f5l//eprzzjsDo9HIJZfMIC+7P4d0Xsz8dBvzLh/a0ONFtM2lW2AvvvgSFi36qOGrptdovvzue64dHc85dYute3t7ExcXR35+y83znp6eDYXXeklJ/aiqqqSioqLLj0E0t2vXDg4ePMC8eW9w9tkTOfvsiSxfvoxt27Zy9tkTycnJYUxCCK9fkkJRlZkbPtrC9qwyR4ftVoqrzNSYbVwxIq7Fn3t4eDQ8IOvVLw/TVhfDnmbx4oU899xsLr74EmbNeqbVYQleXl7cf//DrFjxC2vW/Mh77y1h8uBYjP5BfLajkBdW78Vml1YT0fW+3JbNjlKtwBlor270s4KCAiIiIpp9ZsuWTaSnq422JSX1o7a2lrIyyZs7Q2Sk1hOmsLCg0fbWrklLefQ1Z44BYNWWdJ5bkS55igtIS9uN1Wpl4MBB/LKvkG935nLt2EQGDxrEkSOHm+0vz+buExUV1UJ6zG/WrfhoQ4YMZcmSL/jmmx/47rtVzJx5NyUFOcyYMIy8ilpmLt0mLbHt5NIF2MDAIOLjexEf34ud5UYWrt6K15/vc2bEXw/diooKDh3KaLHbHsCtt17Pa6+91GhbWtpuwsMjmhVsRfdITh7MkiVfNqqcmDTpdBRlEIsWfUR4eDgAI+KDeOfK4fh46bn901RWpLXcTVx0jM1uJ7e8lgCjJyf3CWlxnyeeeIRHH32w0ba0tF0YDAbi4pq3mvdEH364mLff/j9uvvl27rvvH222oC5YMJcPPngXg8FAeLj2Mvrrrz9z2imncO3oeD5PzeaRb3dTY7Z2V/iiB9qeVcaLa/YxZuhgfH39Gk0skp2dRXZ2FsOGjWz2ufp7/Wi7du0kJCRUxm93kn79BnTomrSVR//t9OF8vSOHf6+SQqyzi4zUCkObd+zm2RV76B/hx83jenPw4AHi43s321+ezd0nJWU4W7Y0nnxp8+aNDBs2osX9Dx8+xMyZN1FWVkpISChGozdbt26moqKcaWefymvTU8gtr+Xmj7dyqLi6xd8h/uLSBdh6K9V8nlqexrCUIaQMHc6cF55l164d7NmTxpNPPkJwcEjDeli1tTUUFhZgtWovgqeeejpff/05y5cv48iRTJYt+4oPP1zMTTfd5shD6nHMZjOFhQWYzWaMRu+Gion6Lz8/P4xGI/HxvRp1n0oM8+Xdq0YwKCqAx79L441fDmCR8T0nZM2eAqrNVvpH+jUUuo6+PgCnnXYma9f+zJIlH3DkSCY//riKt956jSuuuAZfX5kgZO/edBYsmMvUqRdywQUXU1hY0PBVXV3dLB+KiYnl/fcX8ccfazl8+DCvvvoiu3fv4rrrbuSuiX2477S+/JRewO2fbpNx36JLFFTU8vC3u4gMMPLshSlMm3Ypb731GuvW/Y6qpvHUU9oar0OGpDTLD2bMuIo//viNjz56j8zMwyxb9hUfffQeN90kw3A6i8Fg6NA1aSuPvuv0gVw3phdfbsvhye/TMFlsDj460ZpBgwaTPDiFR5+aRWVWOrenePP6qy+Qm5vDpZfOkGezA1166QxSUzezcOF8MjIO8t//zmPXrp1cdtmVDfsUFxc39OaMiYmloCCfV155kczMw2zevJHZs2cxdepFxMf3YkR8EPMuH0aV2cbNH28lLbf5igXiL/qnn3661R9WVZla/+Fx8Pb2oqam5fVYj9cPu/N48vs0UmIDeW36UE6fdBr79qWzaNHbfPvt1yQm9uWZZ55vqAVesWI5d955C1OnXkhAQAApKcPw9vbmk08+4oMPFrFv3z5uuuk2LrxwmkOPyxl05XF9//23BAb+tch9auoWrrvuCkaNGt3iEjtr1/5CWVkZ5513QfM4vfScMyiS4moTn2zJYktmKeP7hOJraHnclStdLz8/4+z27NdZadVitfHIt7sx71vHiD6xTJzY8vXp2zeJmJhYvvxyKYsXv8P27alMn34ZN9xwS6e/sLrS9aq3dOkStm7dTHq6ypIlHzT6MhqN5OXlNsqHFGUgVquVRYve5v33F6PX63nyyX/Sr98AdDodKbGB9I/w47PUbFam5TEsLogINxkn44rXtyXtTavQ+c/Wesd7Lkuqzcxcuo2iKhNvXJJCXLAPI0aMoqioiHfemc93333DwIHJPP74bLy9vZvlB3Fx8SQl9eOzzz5h8eKF7Nu3j9tuu4MLL5zeBUfZdZz9XuzINWkrj/bw8GB072C89B4s2ZzF5swSJiWF4e11YmOVnf381XOltKrT6dhojUfdu5+gjJ/5dcU3GAxGZs9+joSERIc+m51Bd95zTd9bQ0NDGTBgIEuXLuGDDxZRWlrKY489xeDBQxo+87e/XU5GxkEmTToNDw8PTjppDD/+uIr//nce69f/wZQp53PnnX9vGF4U4W/k1KQwVqr5fLY1GyXSv9E6v13JGdNvW2lVZ2+j+0h+fnmnNmUFB/tSUlLVKb/LbrfzzvpDzPstgxFxgbwyfQh+BsfMSdWZx+VMXPG4vtuZy/Or0vE3evLUuQMYnxjabB9XOq6IiIB2PXE6K61+tCmTV37az38uGsyp/cI641eeMFe6Xp2hrePdmVPOw9/sorDSxN2T+nDFyDiXn4jFXa5ve9MqdP6ztd7xnMuKWgt3LN3GvoJKXp0+hNG9Wx420BO4y73YESvS8pj9P5XoQG/+c9Fg+oQdfyudq5w/V0qrb/+ewYI/Mrh+TC/unNjyULiezFXuuY7KK6/l3i93sDe/kjsmJHLdmF5dXgHhjOeyrbTqkl2Iy2rMPLYsjXm/ZTBlUCRvXjrUYYVX4VymDo7i3atGEOjtyT2f7+Dfq9KpMsm4wfbIKathwe8ZjE8MYVJS84K/cLzB0QF8+LeRnNwnlFd+2s+tS1LZW1Dp6LCEi8ouq+Gmj7eyJ7+Sf1+Q3KMLrz3V5IGRzL1sKOU1Fv72wWa+3JZNWw0bovv89w+t8Dp1cBQzJyQ6OhzRjSIDjCy8cjhnKRG8tfYgjy3bLe+yTbhcAXbdwSKuXLyJH/cWcOeERGZPUTB4utxhiC7UL8KP968ZydWj4vkiNZur3tvE1sxSR4fl1Kw2O8/8sAeb3c4/zuznll2N3EWQjxf/uSiZJ84ZwMGiKq55fzPPr0wnt1xmLhTttyWzlBs+2kpeRS1vXDKEiUnO0eNCdL9hcUF8dO1IhsUG8tzKdB75djcl1c7VlbAnsVhtvLh6L/N/z2BqciRPTB7g8j1tRMf5eOl5dupA7pnUhzXpBVz13iY2Z5Y4Oiyn4TIlv8ySah76eid3f74DHy8971w5nOvH9pYXbdEio6cH957Wl3kzhmK327n1k1TmrN5LmZP173cWC34/yJ+HSnjg9CTig7tnvIU4fjqdjguHRPPZDaOZlhLNNztymLZwA0//T2W3TPwg2lBrsfH6z/u57ZNUfLw8WHjlcGl5FYT7G3nj0hTumdSHn/cVctmijSzbmSOtsd0sp6yGmUu38enWLK4aFccT5yjoPeQ9t6fS6XT8bXQv5l0+DIDbP9nGyz/ukxUJcIExsGm55bz3Zyar9+Rj0Htw47jeXD0q3qlaXZ2x33hncJfjqjRZmPvrQT5LzSLI24uHz1U4vU+IS9RodscY2E+3ZPHimr1clBLN42f3d7pKIXe5D9vreI43u6yG9zYc5rtduVSbbfQN8+UsJYKzB0SQeAJj2rqDu1xfZx9XZ7fbWb2ngDd+2U9WWS3Th8bw91P7tjrZXU/kLvfiidqbX8lzK9PZnl3G4OgA7prYh5N6H3s5JFc5f86YVi1WG59uzWLebwex22HW5AGcM6j19USFxlXuuc5QZbLyxi/7+Sw1m0h/A7edksjU5KhOq+BwxnPZVlp1ygJspcnC6j0FLNuZy5bMUvwMei4ZFsMVI+OccuZNZ7zoncHdjkvNreCF1XvZnl3GgAg/bj8lkQl9Q52uwHa0rizAmq025v+eweINh5mUFMYLFwzCU+88FUP13O0+PJYTOd6KWgvLd+exUs1na2YpdiAuyJsxCcGM6R3CSb2CCfb16tyAT5C7XF9nfCkGbXjAmvQCFm84jJpXQb9wP+47rS9jEqTVtSl3uRc7g81uZ9nOXOb/dpC8ChMj4oO4elQcE/qGtfrC7Crnz5nSapXJyne7cnn/z8Nkl9UyoW8oD53Rj9gg7674s27HVe65zrQ5s4TXfj7ArpxyksJ9uWV8Aqf2C8fzBAuyznguXaIAW15j4Y+DRfyyr5Cf9xZSY7HRO8SHi1OimTY0Bn+j807S5IwXvTO443HZ7HZ+zSjl1dV7yCypYUhMANeP6dXmQ9mROrsAe6S0mvxyE7tyy/k8NZtDxdVcnBLNP87sh5cTFl7BPe/DtnTW8eZX1PJjeiEbMorZeLiESpMVHaBE+jM8Pogh0QEMiQ0gNtDboZU47nJ9neml2G63s6+wihVpeXy3M5e8ChO9Q3y4bnQvpg7uvBp7d+Mu92JnqjFb+WJbNh9vOkJOeS2R/gbOHRTFWUo4SqR/o55MrnL+HJ1Wa8xWdhZU8/mmw/yyr5Bai42UmEBuGtebk/uEOHWlurNxlXuus9ntdlbtKWDu2gNkltQQ6W/g0uGxXDA4ivDjbOhzxnPplAXYSpOF7VllbDlSxpbMUrZllWG12Qnx8eK0/mFMTY5iaGxgpybkuXPf4MUXn6eysqLTfqfoXH5+/jz00KPcccfdXfY3goN9KSisYNnOXBauO0ROeS3RAUYuGRbD+SeQ+LtCZxZgy2ssnDX3d2x1ew6K8ue2UxI5pU/7ZhyW9NP5Wrrfu+IhYrHZ2Z1TzoZDxWzIKGFnTjm1FhsAIT5eDI4JYGCkP33D/egT5ktCiE+3VWg440PzeDj6pbigopYDZSZ+2p3L2v2FZJXV4qGD8YmhXJwSzcSkzqukk7zAOXXl89Nis/NTegHf7crljwNFWO0Q6uvFyPhgkqP9GRQVwKh+4VBrdvoCWHem1UqThUPF1RwqqiYtr4KtR0rZnVuB1WYn2MeLMweEM2VQ5HG/70padD7d8R5bz2qzs3Z/EUu3HmF9Rgk6YHhcIKcPiGBSUihxQe2f08QZn8UOLcDWmK3kV5g4UFRFdqWZXUdK2Jtfyd6CSmx28NBpLQJjEkKY2DeUITGBXVY7nJIygNzcnC753aLzREVFs337ni77/UcnUovNzi/7Clm65QgbD5eiA4bGBnLGgHBOTgwlIdTHoQ/jzm6B3Z1bTnGVmYRQnw5lbCDpp6s0vd+74yFisdnZl1/JjpwytmeXszO7jIyiaupvIr0OeoX4kBDiS2yQNzFB3sQGehMbZCQ6wBt/o77T0oUzPjSPR2e/FOeV15JZWo3Fasdis2O1ad8rTRZKqy2U1pjJLa/lcHENmSXVFNfNGmv09GBM72AmJIUxsW9olwy7kbzAeXX18xOguMrEHweL+f1AEduzysgq+2sGdH+jnvggH+KDvYkN8ibC30hkgJFIfwOR/kZC/Qwn3NXxRHV2Wv1hdx6/7CvEYrNjttqoNFkpqjJRXGWmtMbSsJ+XXsfg6ACGxQVx6qAoBoV4n/CwHUmLzqk70mFTB4uqWJmWz5r0gobl9WICjYzqFczwuEAGRPrTN8wPYytzCDnjs7ittHrC/XJLqszM+/0g5TUWTFYbFpudGrOVwiozBRUmymstjfaP9DfQN9yPG5LCGBEXxJDYgG5bw3XmzLulpsrJ+fn5M3Nm19da1fP00HFG/3DO6B/OwcIqVu7J58f0Al75aT+vsJ8gb0+GxgYyKDqAhLoX+pggIwFGT6evZW7JoKiA4/6spJ/O1933ez1PDx1KlD9KlD+XaJMbUmO2klFczf7CSvYXVLG/sIpDJdWszyimpq61tp5BryPE10CorxdhfgbCfA0E+3rh66XH28sDHy993ZcHXnoPDHoPDJ4eGPS6hn976XUYPD3Qe3thstjw0utcMk11lZs+3kpOG0sj6XXazLHxwd5M6hdG3zBfxvWPINbHE2+vrp2YSfIC59Rd+UmIr4HzkqM4LzkK0N4Dd+WWk1ttIT27jMySavbkV/LzvkLM1sblPw8dhPsZmhVsIwK075H+RiL8DV1+D3em/YWVpOVV4Omhw9NDh59BT1K4HyE+XkQGGEkI8aF3qC+9g30aJiDtrMKCpEXn46jnemKoL7ecnMAtJydwsKiK9QeL2ZRZyq/7Clm2MxfQnhsJob4khfsRE2gkKsCb6EAtHcbawFZjxs+od9ohZUc74RbYw8XVPPj1Tiw2u/ZCUvdyEupnIMLPQLi/gXA/AwmhvozoG4bVDZcxccZai87Qk48rs6SaTYdLSD1SRmpWGYeL/2qdAq0mNcTHi0BvL4ye2su5Ue+Bp16H3a6NtbWjjVOw2Wn0/0Henjx1rtKucd3dMQuxs3PX+7A1zna8drudkmozWWW1ZJfWkFteS1GVicJKE4VVZooqTRRVmSmpNmOxndhtWP8M0Qq5WmHX6Kl9eXt6YPTSa98btukb/t/o6YG3lx6DXgc6HTpAB9oYPe0/QJu8rNZqx2yxYbLWfVnsDf9vttoaWj09PXTcObFPuyZU6exWncySarJKa/DU69DrdHjqPRpejoO8vfAz6pvNpO5s946rkfN3Ypqev/q8I6/CRF55LfkVteRWmMgvryWvorZhe6Wp+ZIgQd6ehPsb8DN44l2Xtn28tDTvUfdurUNHfRLw0Om4dHgMfcP8jhmno7v7g9xrnUXO47HZ7HaOlNSwJ7+CPXkV7Mmv5EBhFXkVtc0qmOoZPT3w9dI3q3hu77/R6aDunVd7Lah7F7aDHTun9QtnRHzQMWPv0hbYXiE+fHL9Se3aN8DbixI3LMAK9xMf7EN8sA8XpcQAWutUZkkNGcVV5JTVUlRlpqjKRHmNhVqrDZPFRoXJgsVqR6fTHqbae7MOD51W66zT6dDrcHj3KSE6QqfTWltDfA0Mjm67Bd9stVFttlJtrv9uxWSxYbbaGwqHJqvWzc5k0f7f06CntKK22c/MVntD2qq12KixWKmotVBYaaPGbKW2YbutYTzv8dDraCgo1xecPfUe6D10+HrpT+h3n4j6PEgIV3V03qFE+re6X6XJQn65qa5QW0t+hYnc8loKK01UmbT8pLjaTI3ZSo3FVvci/NeLt92uvS+f1CuoXQVYIXoSD52OXiE+9Arx4cwBEQ3bbXY7RVVmcstqyK8wYfPUk19cRaVJe9ZW1T1n65/NJstflbxVZotWGWxp/Fyv/7fd/td7b/07sRaL9l4cYPRsVwG2Lc47ta8QTsTbS0+/CD/6RcjDUYjWeNW1ngZ2YAWIzqhBt9m1h2vNUQ/P+l4P2s+1Wl8AY30Lb12BVSqUhHAsP4MnfmGeTr9mtRDuxEOnI9xP6yULrteaLQVYIYQQLs1Dp8PbS+9S4+aEEEIIcXycf5SuEEIIIYQQQgiBFGCFEEIIIYQQQrgIKcAKIYQQQgghhHAJUoAVQgghhBBCCOESpAArhBBCCCGEEMI12O32bvt6+umnn+7OvyfHJcfVk47LXb962vWS45UvOZdy/lzxS86fnCs5j6775WrnsrtbYJ/q5r/XXeS4XIu7Hpe76mnXS45XHC85lydGzt+JkfPXfnKuOoecx87jUudSuhALIYQQQgghhHAJ3V2And3Nf6+7yHG5Fnc9LnfV066XHK84XnIuT4ycvxMj56/95Fx1DjmPncelzqXObrc7OgYhhBBCCCGEEOKYpAuxEEIIIYQQQgiXIAVYIYQQQgghhBAuwbOrfrGiKCcDvzXZnKqq6vBW9k8A3gZOAQ4B96uquryr4jteiqJEAq8AkwE7sAwt1pJW9l8A3NJk832qqr7apYG2g6IoRuAN4DKgFnhZVdU5rew7DJgHDAN2A7erqvpnd8XaXoqiJAGvAhOASuAT4HFVVWta2HcFcHaTzdNUVf2qywMVberIdXQXiqIMBN4ExgGFwJuqqr7o2Ki6h6Io/wX6qap6mqNjcWYdzLNvAB4H4oDNaM+dDd0VqzPqyPk76jOhaM+8h1VVfbfLg3RiHbz/5PnaAkVRQtDO4XlANfA+2rPN6tDAXETdPbgJuFdV1VV120KB+cA5QBHwlKqqix0XpfNr5TyOBF4GRgEFwALgBVVVbQ4LtA1d2QKbDGwEYo76OrOlHRVF0QFfo720jQYWA58ritKnC+M7Xh8B8WgZ83lACrCwjf2TgYdofB4WdHGM7fUicDJwFnAbMEtRlCua7qQoih+wHFiHdmP/CnynKEpAN8Z6TIqiGIBv0R6sJwNXAxcDz7bykWTgChpfG6erNOlpjuM6ujxFUbzQ7r1DwHDgTuAJRVGudmhg3UBRlDOBmxwdh4tob559NjAXrQA7BFgPLHe2PNsB2nX+mngViOzqwFxER86fPF9bNhftHXIScA1wHXC/QyNyEYqieAMfA4Ob/OhdIAytAewZYH5dI5poQUvnsa4SYDmwHe09/27gAbR3EafUZS2waJnXTlVVc9qx7+mAAkxUVbUc2KUoylloLzWzujDGDlEUJR6tED5QVVW1btvfgV8VRfFVVbWqhY8NAma18zx0m7pC6S3ABaqqbgI2KYoyB7gLWNJk9xmAGXhAVVWboij3AVPrtv+3G8M+ljFAP2CMqqoVwG5FUZ5Aq1F64OgdFUUJRGuZWO9s10a0/zq6kThgA3CnqqrVwF5FUVYBpwIfOjSyLlSXDy2geW8d0UQH8+wotFaIT+o++zRwH1qF6+/dFrQT6eD5q//MFLT8KL/bAnVSHTl/8nxt03nAdaqq7gBQFOUj4Ay0ygHRCkVRktEakHRNticBFwD9VVXdC2yvK7zeQQ/N69rS2nlEuy/NwN/rWlz3KIryMloDwhvdG2X7dHULrNrOfccBW+oKr/XWAuM7PaoTU4pWcEs/apsd7Twam+6sKEo0EEr7z0N3GoYW89qjtq0FRiuK0rRiYxzwW303AlVV7WgvnM52fVTgvLpCTz07LVwbtPuzBq3FSziXjlxHt6Cq6kFVVWeoqlqtKIpOUZRT0GroVzs6ti72LPBT3ZdoW7vzbFVVP6jv2qkoig9a4TUP2NFNsTqjjjzzqGutngfcCpi6JULn1pHzJ8/X1hUCVyuK4qsoSixwLlpXTtG2icAKmr93jgWy6wqv9Zyx/OAsWjuPPwNXNOkubAe8uyuwjurqFtgqRVF2AIFoTdP/UFW1tIV9Y4CsJtty0bpZOI26Avb3TTb/HdiuqmpxCx9JBizAP+tqcguAV5xkHE0MUNRkTGEuYAAigOwm+zYthOeidXV0Gqqq5gOr6v+tKIoHWu3wry3sngyUAEsURZkIHAaeVlW16fUV3ayD19EdZQKxaOPrP3NwLF1GUZTxaGPphuC+LeudqSN5NgCKopzDX902r1ZVtazLo3ReHT1/c4D/qar6i6Io3RSiU+vI+ZPna+vuQBv3Wo7W+LEGeNqRAbkCVVXn1/9/k/ToEuUHZ9HaeVRV9TBaOq3/mQ9aj4tl3RlfRxx3AbauD3VrN0gu0Att4ojr0Pqmv4zWbD21hf190ca7Ha0WB7S4HOu4jm4lrutKexnawPGWDKr7ngq8DpyG1je/UlXVpZ0T8XFr7ZxD8/PuNNeng14GRqCNq25qEOAPfIPWCjQN+FZRlJNVVV3ffSGKdmjrOrqji9AKsP+HNmHcPY4Np/PVTSCxEG0CiWIpILRLR/LseqnASLR76l1FUQ6oqrqui+Jzdu0+f4qinIrWLbHpWLuerCP3nzxfW9cP2ALMRmvceRP4D3CvI4NyYa3dlwZFUXR1PQZFByiKogc+APyA5xwcTqtOpAX2JFpvEbkBCAHK62dWUxTlOmCjoii9VVVt2q2kBghqss0ItDSmtKsd67jeBVAU5QG0MQt318/g1YK5wMeqqhbV/Xuboij9gZmAowuwNTR/6NT/u+l5b21fR1yfY6qbFOxVtJrOS1VV3dnCbo8Azx41e3Sqoiij0Cam6OkPWKfQzuvodlRV3QigKIovsFhRlAdVVXW3LoxPAulOUJHnSjqSZwNQN/4wB9ha1+J9O9pkfD1Ru85fXcvDf9Ge7S31GOupOnL/yfO1BUfNrp+oqmpm3babgZWKojyvqmquQwN0Ta3dl9VSeO24ukk0P0JbaeUsZx7DftwFWFVV19J8EHBbdtd9j6P5uIgjaOMrjhZNC12iulp7jktRlNloL2D3qKr6Vhu/y442pffRdqPdGI52BAhRFMVw1MtxNFrNVdOYj9T97GgOuT7HUtfddCHawPMZqqp+3dJ+dRUrTZc+2k3z+1A4QHuvo7tQFCUOGKWq6jdHbd6F1j0vEG34gTu5CohRFKV+nLMB0CuKUqGqqr8D43Jm7c6z6wqrVaqqph61eZwunVUAAAPxSURBVBcwoFsidU7tPX/1k8i9f1TPAF9gnqIo41RVvb27AnYy7b7/5PnaqlFoDTuZR23bBOiBBLTei6JjXOb91NnVVd59iTY+9lxn7y3RJZM4KYoyWlGU8roB6vVGAFYaT4BUbx0wvG6Wu3oTcMKa4rpZh58AblNVtc2ZuRRFeUlRlKb9x0cAaV0VXwdsRZuY4uipxicAm1RVtTTZdx1wcl2LWH3L2Mk44fUBXkJ7OZ6uquoXre2kKMpniqLMbbLZWa6NaOd1dCODgC8UbZ3peqOAfFVV3a3wCtpwiiFo4+iHo60BvhEnG1fvZDqSZ98J/LPJtlH8VZHcE7X3/G0A+vPXvTkcrWDxZN1XT9Xu+0+er63KAoIVRel11Lb6oWb7HRCPO1gHxCmKknjUNqcsP7iAD9EmxTpLVVWnXxmgqyZxSkWrFVmoKMqDaGNg5wPv1L+MKYoSgdbEX4E2+1UG2hidp4Hz0Wa+daq1ARVF6Q28gDY27Zu6WYbr5auqam1yXN8AaxRFuQf4Dm22uWtpZT3c7qSqapWiKIuBuYqiXI9WY/Ug2oyL9TMol9Yt6fEZ8G/gjbqH0i1orUItLj3gKIqijEMbR/IoWnf1huujqmpOk2P6BligKMpa4E+09dgmoHWxEw50rOvosMC61s9oLWTv1g1PSEJLc2659q2qqhlH/1tRlGK0fHNvKx/p8TqYZ7+FtrzbXcAPaHNRjETr0dAjdfD8NboPFUWxAnmqquZ1b9TOo4PnT56vLVuHVhGwSFGU+9Fa9ucD77tpRWWXU1V1v6IoPwDv1eV3o9DyudMdG5lrURRlBtpY9WuAw0e9d1nrJtZ0Ol3SAlvXvaR+TaHfgC/Qpm2++6jd/kTL/Oq7m1yEtlj4JrRC3jRVVQ92RXwn4EK0vvV3oHVPOPqrT90+Rx/Xz8CVaBn8TrRa8Svruik7g/vR4l2DtlzAM/XrBqId0wyAupkrp6LVvG5GWyz6vCbLHjmDS+u+P0+T61M3zf/Rx/Qe2nV6Bm3h5vOAc1RV3dfdQYtmjnUd3Y6qqma0ijsL2hix+WhjpV53ZFzC6bQ3z/4DbYLBmWj522S0/C2z2W/sWdp1/kSr2nv/yfO1BXUt1VPRulyvQXs3/hltbLA4fteidVlfj9ZL4ua6PFC032V13z+g8XvXFodFdAw6u13GOAshhBBCCCGEcH5d0gIrhBBCCCGEEEJ0NinACiGEEEIIIYRwCVKAFUIIIYQQQgjhEqQAK4QQQgghhBDCJUgBVgghhBBCCCGES5ACrBBCCCGEEEIIlyAFWCGEEEIIIYQQLkEKsEIIIYQQQgghXIIUYIUQQgghhBBCuIT/B5hr983ARHBpAAAAAElFTkSuQmCC\n", - "text/plain": [ - "
" - ] - }, - "metadata": { - "needs_background": "light" - }, - "output_type": "display_data" - } - ], + "outputs": [], "source": [ "az.plot_posterior(data, figsize=(13,3));" ] @@ -386,63 +354,35 @@ }, { "cell_type": "code", - "execution_count": 9, + "execution_count": null, "metadata": {}, - "outputs": [ - { - "data": { - "image/png": "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\n", - "text/plain": [ - "
" - ] - }, - "metadata": { - "needs_background": "light" - }, - "output_type": "display_data" - } - ], + "outputs": [], "source": [ "def SIR(y, t, p):\n", - " ds = -p[0]*y[0]*y[1]\n", - " di = p[0]*y[0]*y[1] - p[1]*y[1] \n", + " ds = -p[0] * y[0] * y[1]\n", + " di = p[0] * y[0] * y[1] - p[1] * y[1] \n", " return [ds, di]\n", "\n", - "times = np.arange(0,5,0.25)\n", + "times = np.arange(0, 5, 0.25)\n", "\n", "beta,gamma = 4,1.0\n", "# Create true curves\n", - "y = odeint(SIR, t=times, y0=[0.99, 0.01], args=((beta,gamma),), rtol=1e-8)\n", + "y = odeint(SIR, t=times, y0=[0.99, 0.01], args=((beta, gamma),), rtol=1e-8)\n", "# Observational model. Lognormal likelihood isn't appropriate, but we'll do it anyway\n", "yobs = np.random.lognormal(mean=np.log(y[1::]), sigma=[0.2, 0.3])\n", "\n", - "\n", - "plt.plot(times[1::],yobs, marker='o', linestyle='none')\n", - "plt.plot(times, y[:,0], color='C0', alpha=0.5, label=f'$S(t)$')\n", - "plt.plot(times, y[:,1], color ='C1', alpha=0.5, label=f'$I(t)$')\n", + "plt.plot(times[1::], yobs, marker='o', linestyle='none')\n", + "plt.plot(times, y[:, 0], color='C0', alpha=0.5, label=f'$S(t)$')\n", + "plt.plot(times, y[:, 1], color='C1', alpha=0.5, label=f'$I(t)$')\n", "plt.legend()\n", "plt.show()" ] }, { "cell_type": "code", - "execution_count": 10, + "execution_count": null, "metadata": {}, - "outputs": [ - { - "name": "stderr", - "output_type": "stream", - "text": [ - "Auto-assigning NUTS sampler...\n", - "Initializing NUTS using jitter+adapt_diag...\n", - "Sequential sampling (2 chains in 1 job)\n", - "NUTS: [lambda, R0, sigma]\n", - "Sampling chain 0, 0 divergences: 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 3000/3000 [20:55<00:00, 2.39it/s]\n", - "Sampling chain 1, 0 divergences: 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 3000/3000 [20:11<00:00, 2.48it/s]\n", - "100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 4000/4000 [02:19<00:00, 28.66it/s]\n" - ] - } - ], + "outputs": [], "source": [ "sir_model = DifferentialEquation(\n", " func=SIR, \n", @@ -456,41 +396,28 @@ " sigma = pm.HalfCauchy('sigma', 1, shape=2)\n", " \n", " # R0 is bounded below by 1 because we see an epidemic has occured\n", - " R0 = pm.Bound(pm.Normal, lower=1)('R0', 2,3)\n", - " lam = pm.Lognormal('lambda',pm.math.log(2),2)\n", - " beta = pm.Deterministic('beta', lam*R0)\n", + " R0 = pm.Bound(pm.Normal, lower=1)('R0', 2, 3)\n", + " lam = pm.Lognormal('lambda', pm.math.log(2), 2)\n", + " beta = pm.Deterministic('beta', lam * R0)\n", " \n", " sir_curves = sir_model(y0=[0.99, 0.01], theta=[beta, lam])\n", " \n", - " Y = pm.Lognormal('Y', mu=pm.math.log(sir_curves), sd=sigma, observed=yobs)\n", + " Y = pm.Lognormal('Y', mu=pm.math.log(sir_curves), sigma=sigma, observed=yobs)\n", "\n", " prior = pm.sample_prior_predictive()\n", " trace = pm.sample(2000,tune=1000, target_accept=0.9, cores=1)\n", " posterior_predictive = pm.sample_posterior_predictive(trace)\n", " \n", - " data = az.from_pymc3(trace=trace, prior = prior, posterior_predictive = posterior_predictive)" + " data = az.from_pymc3(trace=trace, prior=prior, posterior_predictive=posterior_predictive)" ] }, { "cell_type": "code", - "execution_count": 11, + "execution_count": null, "metadata": {}, - "outputs": [ - { - "data": { - "image/png": "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\n", - "text/plain": [ - "
" - ] - }, - "metadata": { - "needs_background": "light" - }, - "output_type": "display_data" - } - ], + "outputs": [], "source": [ - "az.plot_posterior(data,round_to=2, credible_interval=0.95);" + "az.plot_posterior(data, round_to=2, credible_interval=0.95);" ] }, { @@ -524,9 +451,9 @@ ], "metadata": { "kernelspec": { - "display_name": "Python 3", + "display_name": "pymc3", "language": "python", - "name": "python3" + "name": "pymc3" }, "language_info": { "codemirror_mode": { @@ -538,7 +465,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.6.9" + "version": "3.8.1" } }, "nbformat": 4, From b899fcbeeaa969566e19737b28a2b2761aaf775c Mon Sep 17 00:00:00 2001 From: Thomas Wiecki Date: Tue, 12 May 2020 14:57:01 +0200 Subject: [PATCH 2/2] Rerun ODE NB. --- .../notebooks/ODE_API_introduction.ipynb | 545 +++++++++++++++--- 1 file changed, 465 insertions(+), 80 deletions(-) diff --git a/docs/source/notebooks/ODE_API_introduction.ipynb b/docs/source/notebooks/ODE_API_introduction.ipynb index 7533b54d74..afc338af6e 100644 --- a/docs/source/notebooks/ODE_API_introduction.ipynb +++ b/docs/source/notebooks/ODE_API_introduction.ipynb @@ -28,10 +28,10 @@ "Ordinary differential equations (ODEs) are a convenient mathematical framework for modelling the temporal dynamics of a system in disciplines from engineering to ecology. Though most analyses focus on bifurcations and stability of fixed points, parameter and uncertainty estimates are more salient in systems of practical interest, such as population pharmacokinetics and pharmacodynamics.\n", "\n", "\n", - "Both parameter estimation and uncertainty propagation are handled elegantly by the Bayesian framework. In this notebook, I showcase how PyMC3 can be used to do inference for differential equations using the `ode` submodule. \n", + "Both parameter estimation and uncertainty propagation are handled elegantly by the Bayesian framework. In this notebook, I showcase how PyMC3 can be used to do inference for differential equations using the `ode` submodule. While the current implementation is quite flexible and well integrated, more complex models can easily become slow to estimate. A new package that integrates the much faster `sundials` package into PyMC3 called `sunode` can be found [here](https://github.com/aseyboldt/sunode). \n", "\n", "\n", - "# Catching Up On Differential Equations\n", + "## Catching Up On Differential Equations\n", "\n", "A differential equation is an equation relating an unknown function's derivative to itself. We usually write differentual equations as \n", "\n", @@ -42,7 +42,7 @@ "Only a small subset of differential equations have an analytical solution. For most differential equations of applied interest, numerical methods must be used to obtain approximate solutions.\n", "\n", "\n", - "# Doing Bayesian Inference With Differential Equations\n", + "## Doing Bayesian Inference With Differential Equations\n", "\n", "PyMC3 uses Hamiltonian Monte Carlo (HMC) to obtain samples from the posterior distribution. HMC requires derivatives of the ODE's solution with respect to the parameters $p$. The `ode` submodual automatically computes appropriate derivatives so you don't have to. All you have to do is \n", "\n", @@ -52,7 +52,7 @@ "\n", "Let's see how this is done in practice with a small example.\n", "\n", - "# A Differential Equation For Freefall\n", + "## A Differential Equation For Freefall\n", "\n", "An object of mass $m$ is brought to some height and allowed to fall freely until it reaches the ground. A differential equation describing the object's speed over time is \n", "\n", @@ -60,13 +60,7 @@ "\n", "The force the object experiences in the downwards direction is $mg$, while the force the object experiences in the opposite direction (due to air resistance) is proportional to how fast the object is presently moving. Let's assume the object starts from rest (that is, that the object's inital velocity is 0). This may or may not be the case. To showcase how to do inference on intial conditions, I will first assume the object starts from rest, and then relax that assumption later.\n", "\n", - "Data on this object's speed as a function of time is shown below. The data may be noisy because of our measurement tools, or because the object is an irregular shape, thus leading to times during freefall when the object is more/less aerodynamic. Let's use this data to estimate the proportionality constant for air resistance.\n", - "\n", - "\n", - "\n", - "\n", - "\n", - "\n" + "Data on this object's speed as a function of time is shown below. The data may be noisy because of our measurement tools, or because the object is an irregular shape, thus leading to times during freefall when the object is more/less aerodynamic. Let's use this data to estimate the proportionality constant for air resistance." ] }, { @@ -76,7 +70,7 @@ "outputs": [ { "data": { - "image/png": "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\n", + "image/png": "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\n", "text/plain": [ "
" ] @@ -137,52 +131,105 @@ "name": "stderr", "output_type": "stream", "text": [ - "INFO (theano.gof.compilelock): Waiting for existing lock by process '79996' (I am process '80216')\n", - "INFO (theano.gof.compilelock): To manually release the lock, delete /Users/twiecki/.theano/compiledir_macOS-10.15.3-x86_64-i386-64bit-i386-3.8.1-64/lock_dir\n", - "INFO (theano.gof.compilelock): Waiting for existing lock by process '79996' (I am process '80216')\n", - "INFO (theano.gof.compilelock): To manually release the lock, delete /Users/twiecki/.theano/compiledir_macOS-10.15.3-x86_64-i386-64bit-i386-3.8.1-64/lock_dir\n" + "Auto-assigning NUTS sampler...\n", + "Initializing NUTS using jitter+adapt_diag...\n", + "Sequential sampling (2 chains in 1 job)\n", + "NUTS: [gamma, sigma]\n" ] }, { - "ename": "KeyboardInterrupt", - "evalue": "", - "output_type": "error", - "traceback": [ - "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", - "\u001b[0;31mKeyboardInterrupt\u001b[0m Traceback (most recent call last)", - "\u001b[0;32m\u001b[0m in \u001b[0;36m\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m ode_model = DifferentialEquation(\n\u001b[0m\u001b[1;32m 2\u001b[0m \u001b[0mfunc\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mfreefall\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 3\u001b[0m \u001b[0mtimes\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mtimes\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 4\u001b[0m \u001b[0mn_states\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 5\u001b[0m \u001b[0mn_theta\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m2\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m~/working/projects/pymc/pymc3/ode/ode.py\u001b[0m in \u001b[0;36m__init__\u001b[0;34m(self, func, times, n_states, n_theta, t0)\u001b[0m\n\u001b[1;32m 86\u001b[0m \u001b[0;31m# Private\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 87\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_augmented_times\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0minsert\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtimes\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mt0\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mastype\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfloatX\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 88\u001b[0;31m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_augmented_func\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mutils\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0maugment_system\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfunc\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mn_states\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mn_theta\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 89\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_sens_ic\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mutils\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmake_sens_ic\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mn_states\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mn_theta\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfloatX\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 90\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m~/working/projects/pymc/pymc3/ode/utils.py\u001b[0m in \u001b[0;36maugment_system\u001b[0;34m(ode_func, n_states, n_theta)\u001b[0m\n\u001b[1;32m 119\u001b[0m \u001b[0mddt_dydp\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mJdfdy\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0mgrad_f\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mflatten\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 120\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 121\u001b[0;31m system = theano.function(\n\u001b[0m\u001b[1;32m 122\u001b[0m \u001b[0minputs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mt_y\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mt_t\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mt_p\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdydp_vec\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 123\u001b[0m \u001b[0moutputs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mt_yhat\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mddt_dydp\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/compile/function.py\u001b[0m in \u001b[0;36mfunction\u001b[0;34m(inputs, outputs, mode, updates, givens, no_default_updates, accept_inplace, name, rebuild_strict, allow_input_downcast, profile, on_unused_input)\u001b[0m\n\u001b[1;32m 304\u001b[0m \u001b[0;31m# note: pfunc will also call orig_function -- orig_function is\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 305\u001b[0m \u001b[0;31m# a choke point that all compilation must pass through\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 306\u001b[0;31m fn = pfunc(params=inputs,\n\u001b[0m\u001b[1;32m 307\u001b[0m \u001b[0moutputs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0moutputs\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 308\u001b[0m \u001b[0mmode\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mmode\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/compile/pfunc.py\u001b[0m in \u001b[0;36mpfunc\u001b[0;34m(params, outputs, mode, updates, givens, no_default_updates, accept_inplace, name, rebuild_strict, allow_input_downcast, profile, on_unused_input, output_keys)\u001b[0m\n\u001b[1;32m 481\u001b[0m \u001b[0minputs\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msi\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 482\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 483\u001b[0;31m return orig_function(inputs, cloned_outputs, mode,\n\u001b[0m\u001b[1;32m 484\u001b[0m \u001b[0maccept_inplace\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0maccept_inplace\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mname\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mname\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 485\u001b[0m \u001b[0mprofile\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mprofile\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mon_unused_input\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mon_unused_input\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/compile/function_module.py\u001b[0m in \u001b[0;36morig_function\u001b[0;34m(inputs, outputs, mode, accept_inplace, name, profile, on_unused_input, output_keys)\u001b[0m\n\u001b[1;32m 1830\u001b[0m \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1831\u001b[0m \u001b[0mMaker\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mgetattr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmode\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'function_maker'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mFunctionMaker\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1832\u001b[0;31m m = Maker(inputs,\n\u001b[0m\u001b[1;32m 1833\u001b[0m \u001b[0moutputs\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1834\u001b[0m \u001b[0mmode\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/compile/function_module.py\u001b[0m in \u001b[0;36m__init__\u001b[0;34m(self, inputs, outputs, mode, accept_inplace, function_builder, profile, on_unused_input, fgraph, output_keys, name)\u001b[0m\n\u001b[1;32m 1517\u001b[0m optimizer, inputs, outputs)\n\u001b[1;32m 1518\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1519\u001b[0;31m \u001b[0moptimizer_profile\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0moptimizer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfgraph\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1520\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1521\u001b[0m \u001b[0mend_optimizer\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtime\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtime\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/gof/opt.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, fgraph)\u001b[0m\n\u001b[1;32m 106\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 107\u001b[0m \"\"\"\n\u001b[0;32m--> 108\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moptimize\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfgraph\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 109\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 110\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0madd_requirements\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfgraph\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/gof/opt.py\u001b[0m in \u001b[0;36moptimize\u001b[0;34m(self, fgraph, *args, **kwargs)\u001b[0m\n\u001b[1;32m 95\u001b[0m \u001b[0morig\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtheano\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtensor\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbasic\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconstant\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0menable\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 96\u001b[0m \u001b[0mtheano\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtensor\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbasic\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconstant\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0menable\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mFalse\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 97\u001b[0;31m \u001b[0mret\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mapply\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfgraph\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 98\u001b[0m \u001b[0;32mfinally\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 99\u001b[0m \u001b[0mtheano\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtensor\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbasic\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconstant\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0menable\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0morig\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/gof/opt.py\u001b[0m in \u001b[0;36mapply\u001b[0;34m(self, fgraph)\u001b[0m\n\u001b[1;32m 249\u001b[0m \u001b[0mnb_nodes_before\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfgraph\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mapply_nodes\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 250\u001b[0m \u001b[0mt0\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtime\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtime\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 251\u001b[0;31m \u001b[0msub_prof\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0moptimizer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moptimize\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfgraph\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 252\u001b[0m \u001b[0ml\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfloat\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtime\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtime\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m-\u001b[0m \u001b[0mt0\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 253\u001b[0m \u001b[0msub_profs\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msub_prof\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/gof/opt.py\u001b[0m in \u001b[0;36moptimize\u001b[0;34m(self, fgraph, *args, **kwargs)\u001b[0m\n\u001b[1;32m 95\u001b[0m \u001b[0morig\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtheano\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtensor\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbasic\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconstant\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0menable\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 96\u001b[0m \u001b[0mtheano\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtensor\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbasic\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconstant\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0menable\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mFalse\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 97\u001b[0;31m \u001b[0mret\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mapply\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfgraph\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 98\u001b[0m \u001b[0;32mfinally\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 99\u001b[0m \u001b[0mtheano\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtensor\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbasic\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconstant\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0menable\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0morig\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/gof/opt.py\u001b[0m in \u001b[0;36mapply\u001b[0;34m(self, fgraph, start_from)\u001b[0m\n\u001b[1;32m 2511\u001b[0m \u001b[0mnb\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mchange_tracker\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnb_imported\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2512\u001b[0m \u001b[0mt_opt\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtime\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtime\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2513\u001b[0;31m \u001b[0mlopt_change\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mprocess_node\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfgraph\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnode\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlopt\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 2514\u001b[0m \u001b[0mtime_opts\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mlopt\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m+=\u001b[0m \u001b[0mtime\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtime\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m-\u001b[0m \u001b[0mt_opt\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2515\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0mlopt_change\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/gof/opt.py\u001b[0m in \u001b[0;36mprocess_node\u001b[0;34m(self, fgraph, node, lopt)\u001b[0m\n\u001b[1;32m 2032\u001b[0m \u001b[0mlopt\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mlopt\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlocal_opt\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2033\u001b[0m \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2034\u001b[0;31m \u001b[0mreplacements\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mlopt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtransform\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnode\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 2035\u001b[0m \u001b[0;32mexcept\u001b[0m \u001b[0mException\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2036\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfailure_callback\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/tensor/opt.py\u001b[0m in \u001b[0;36mlocal_subtensor_merge\u001b[0;34m(node)\u001b[0m\n\u001b[1;32m 3086\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mtype\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mslice1\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0mslice\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 3087\u001b[0m merged_slices.append(\n\u001b[0;32m-> 3088\u001b[0;31m merge_two_slices(slice1,\n\u001b[0m\u001b[1;32m 3089\u001b[0m \u001b[0mxshape\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mpos_1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 3090\u001b[0m \u001b[0mslices2\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mpos_2\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/tensor/opt.py\u001b[0m in \u001b[0;36mmerge_two_slices\u001b[0;34m(slice1, len1, slice2, len2)\u001b[0m\n\u001b[1;32m 3038\u001b[0m \u001b[0mstep\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mpre_greedy_local_optimizer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlist_opt\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstep\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 3039\u001b[0m \u001b[0mstart\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mpre_greedy_local_optimizer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlist_opt\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstart\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 3040\u001b[0;31m \u001b[0mstop\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mpre_greedy_local_optimizer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlist_opt\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstop\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 3041\u001b[0m \u001b[0mstep\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mpre_greedy_local_optimizer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlist_opt\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstep\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 3042\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/gof/opt.py\u001b[0m in \u001b[0;36mpre_greedy_local_optimizer\u001b[0;34m(list_optimizations, out)\u001b[0m\n\u001b[1;32m 2920\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2921\u001b[0m \u001b[0mout_index\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2922\u001b[0;31m final_outs, optimized_nodes = local_recursive_function(\n\u001b[0m\u001b[1;32m 2923\u001b[0m list_optimizations, out, {}, 0)\n\u001b[1;32m 2924\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mfinal_outs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mout_index\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/gof/opt.py\u001b[0m in \u001b[0;36mlocal_recursive_function\u001b[0;34m(list_opt, out, optimized_vars, depth)\u001b[0m\n\u001b[1;32m 2889\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2890\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0minp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mowner\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2891\u001b[0;31m outs, optimized_vars = local_recursive_function(\n\u001b[0m\u001b[1;32m 2892\u001b[0m \u001b[0mlist_opt\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2893\u001b[0m \u001b[0minp\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/gof/opt.py\u001b[0m in \u001b[0;36mlocal_recursive_function\u001b[0;34m(list_opt, out, optimized_vars, depth)\u001b[0m\n\u001b[1;32m 2889\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2890\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0minp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mowner\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2891\u001b[0;31m outs, optimized_vars = local_recursive_function(\n\u001b[0m\u001b[1;32m 2892\u001b[0m \u001b[0mlist_opt\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2893\u001b[0m \u001b[0minp\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/gof/opt.py\u001b[0m in \u001b[0;36mlocal_recursive_function\u001b[0;34m(list_opt, out, optimized_vars, depth)\u001b[0m\n\u001b[1;32m 2889\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2890\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0minp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mowner\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2891\u001b[0;31m outs, optimized_vars = local_recursive_function(\n\u001b[0m\u001b[1;32m 2892\u001b[0m \u001b[0mlist_opt\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2893\u001b[0m \u001b[0minp\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/gof/opt.py\u001b[0m in \u001b[0;36mlocal_recursive_function\u001b[0;34m(list_opt, out, optimized_vars, depth)\u001b[0m\n\u001b[1;32m 2905\u001b[0m \u001b[0mresults\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnode\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moutputs\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2906\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mopt\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mlist_opt\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2907\u001b[0;31m \u001b[0mret\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mopt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtransform\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnode\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 2908\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mret\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mFalse\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0mret\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2909\u001b[0m \u001b[0;32massert\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mret\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnode\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moutputs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mopt\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/tensor/opt.py\u001b[0m in \u001b[0;36mconstant_folding\u001b[0;34m(node)\u001b[0m\n\u001b[1;32m 6513\u001b[0m node.op.python_constant_folding(node)):\n\u001b[1;32m 6514\u001b[0m \u001b[0mimpl\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m'py'\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 6515\u001b[0;31m thunk = node.op.make_thunk(node, storage_map, compute_map,\n\u001b[0m\u001b[1;32m 6516\u001b[0m no_recycling=[], impl=impl)\n\u001b[1;32m 6517\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/gof/op.py\u001b[0m in \u001b[0;36mmake_thunk\u001b[0;34m(self, node, storage_map, compute_map, no_recycling, impl)\u001b[0m\n\u001b[1;32m 952\u001b[0m compute_map=compute_map, impl='c')\n\u001b[1;32m 953\u001b[0m \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 954\u001b[0;31m return self.make_c_thunk(node, storage_map, compute_map,\n\u001b[0m\u001b[1;32m 955\u001b[0m no_recycling)\n\u001b[1;32m 956\u001b[0m \u001b[0;32mexcept\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mNotImplementedError\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mutils\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mMethodNotDefined\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/gof/op.py\u001b[0m in \u001b[0;36mmake_c_thunk\u001b[0;34m(self, node, storage_map, compute_map, no_recycling)\u001b[0m\n\u001b[1;32m 855\u001b[0m \u001b[0;32mraise\u001b[0m \u001b[0mNotImplementedError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"float16\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 856\u001b[0m \u001b[0m_logger\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdebug\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'Trying CLinker.make_thunk'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 857\u001b[0;31m outputs = cl.make_thunk(input_storage=node_input_storage,\n\u001b[0m\u001b[1;32m 858\u001b[0m output_storage=node_output_storage)\n\u001b[1;32m 859\u001b[0m \u001b[0mthunk\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnode_input_filters\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnode_output_filters\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0moutputs\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/gof/cc.py\u001b[0m in \u001b[0;36mmake_thunk\u001b[0;34m(self, input_storage, output_storage, storage_map, keep_lock)\u001b[0m\n\u001b[1;32m 1213\u001b[0m \"\"\"\n\u001b[1;32m 1214\u001b[0m \u001b[0minit_tasks\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtasks\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget_init_tasks\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1215\u001b[0;31m cthunk, module, in_storage, out_storage, error_storage = self.__compile__(\n\u001b[0m\u001b[1;32m 1216\u001b[0m \u001b[0minput_storage\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0moutput_storage\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstorage_map\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1217\u001b[0m keep_lock=keep_lock)\n", - "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/gof/cc.py\u001b[0m in \u001b[0;36m__compile__\u001b[0;34m(self, input_storage, output_storage, storage_map, keep_lock)\u001b[0m\n\u001b[1;32m 1151\u001b[0m \u001b[0minput_storage\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtuple\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minput_storage\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1152\u001b[0m \u001b[0moutput_storage\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtuple\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0moutput_storage\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1153\u001b[0;31m thunk, module = self.cthunk_factory(error_storage,\n\u001b[0m\u001b[1;32m 1154\u001b[0m \u001b[0minput_storage\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1155\u001b[0m \u001b[0moutput_storage\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/gof/cc.py\u001b[0m in \u001b[0;36mcthunk_factory\u001b[0;34m(self, error_storage, in_storage, out_storage, storage_map, keep_lock)\u001b[0m\n\u001b[1;32m 1621\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mnode\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnode_order\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1622\u001b[0m \u001b[0mnode\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mop\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mprepare_node\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnode\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstorage_map\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'c'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1623\u001b[0;31m module = get_module_cache().module_from_key(\n\u001b[0m\u001b[1;32m 1624\u001b[0m key=key, lnk=self, keep_lock=keep_lock)\n\u001b[1;32m 1625\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/gof/cmodule.py\u001b[0m in \u001b[0;36mmodule_from_key\u001b[0;34m(self, key, lnk, keep_lock)\u001b[0m\n\u001b[1;32m 1157\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mmodule\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1158\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1159\u001b[0;31m \u001b[0;32mwith\u001b[0m \u001b[0mcompilelock\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlock_ctx\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mkeep_lock\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mkeep_lock\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1160\u001b[0m \u001b[0;31m# 1) Maybe somebody else compiled it for us while we\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1161\u001b[0m \u001b[0;31m# where waiting for the lock. Try to load it again.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/contextlib.py\u001b[0m in \u001b[0;36m__enter__\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 111\u001b[0m \u001b[0;32mdel\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mkwds\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfunc\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 112\u001b[0m \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 113\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mnext\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mgen\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 114\u001b[0m \u001b[0;32mexcept\u001b[0m \u001b[0mStopIteration\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 115\u001b[0m \u001b[0;32mraise\u001b[0m \u001b[0mRuntimeError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"generator didn't yield\"\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfrom\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/gof/compilelock.py\u001b[0m in \u001b[0;36mlock_ctx\u001b[0;34m(lock_dir, keep_lock, **kw)\u001b[0m\n\u001b[1;32m 38\u001b[0m \u001b[0;34m@\u001b[0m\u001b[0mcontextmanager\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 39\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mlock_ctx\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlock_dir\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkeep_lock\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mFalse\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkw\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 40\u001b[0;31m \u001b[0mget_lock\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlock_dir\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mlock_dir\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkw\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 41\u001b[0m \u001b[0;32myield\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 42\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0mkeep_lock\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/gof/compilelock.py\u001b[0m in \u001b[0;36m_get_lock\u001b[0;34m(lock_dir, **kw)\u001b[0m\n\u001b[1;32m 84\u001b[0m \u001b[0;31m# Only really try to acquire the lock if we do not have it already.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 85\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mget_lock\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mn_lock\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 86\u001b[0;31m \u001b[0mlock\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mget_lock\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlock_dir\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkw\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 87\u001b[0m \u001b[0matexit\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mregister\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mUnlocker\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0munlock\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mget_lock\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0munlocker\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 88\u001b[0m \u001b[0;31m# Store time at which the lock was set.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m~/anaconda3/envs/pymc3/lib/python3.8/site-packages/theano/gof/compilelock.py\u001b[0m in \u001b[0;36mlock\u001b[0;34m(tmp_dir, timeout, min_wait, max_wait, verbosity)\u001b[0m\n\u001b[1;32m 271\u001b[0m \u001b[0mno_display\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mTrue\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 272\u001b[0m \u001b[0mnb_wait\u001b[0m \u001b[0;34m+=\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 273\u001b[0;31m \u001b[0mtime\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msleep\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mrandom\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0muniform\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmin_wait\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmax_wait\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 274\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 275\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mPY3\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;31mKeyboardInterrupt\u001b[0m: " + "data": { + "text/html": [ + "\n", + "
\n", + " \n", + " \n", + " 100.00% [3000/3000 03:37<00:00 Sampling chain 0, 0 divergences]\n", + "
\n", + " " + ], + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "\n", + "
\n", + " \n", + " \n", + " 100.00% [3000/3000 04:36<00:00 Sampling chain 1, 0 divergences]\n", + "
\n", + " " + ], + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Sampling 2 chains for 1_000 tune and 2_000 draw iterations (2_000 + 4_000 draws total) took 493 seconds.\n" ] + }, + { + "data": { + "text/html": [ + "\n", + "
\n", + " \n", + " \n", + " 100.00% [4000/4000 01:46<00:00]\n", + "
\n", + " " + ], + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" } ], "source": [ @@ -214,9 +261,20 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 4, "metadata": {}, - "outputs": [], + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], "source": [ "az.plot_posterior(data);" ] @@ -232,9 +290,114 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 5, "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Auto-assigning NUTS sampler...\n", + "Initializing NUTS using jitter+adapt_diag...\n", + "Sequential sampling (2 chains in 1 job)\n", + "NUTS: [g, gamma, sigma]\n" + ] + }, + { + "data": { + "text/html": [ + "\n", + "
\n", + " \n", + " \n", + " 100.00% [3000/3000 18:00<00:00 Sampling chain 0, 0 divergences]\n", + "
\n", + " " + ], + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "\n", + "
\n", + " \n", + " \n", + " 100.00% [3000/3000 18:41<00:00 Sampling chain 1, 0 divergences]\n", + "
\n", + " " + ], + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Sampling 2 chains for 1_000 tune and 2_000 draw iterations (2_000 + 4_000 draws total) took 2202 seconds.\n" + ] + }, + { + "data": { + "text/html": [ + "\n", + "
\n", + " \n", + " \n", + " 100.00% [4000/4000 01:40<00:00]\n", + "
\n", + " " + ], + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], "source": [ "with pm.Model() as model2: \n", " sigma = pm.HalfCauchy('sigma',1)\n", @@ -256,9 +419,20 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 6, "metadata": {}, - "outputs": [], + "outputs": [ + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAABdoAAAFTCAYAAAAugr49AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+j8jraAAAgAElEQVR4nOzdd3gU9drG8e9ueu89hIQWEkLvighIE1ApdkSx+9q7x+M5VqwcFRX1qGAv2EVpUu0iLdRAEgKBFEjvPdl9/0ByjKC0JLOb3J/r8rp0d3bmnllkZp79zfMzWa1WKyIiIiIiIiIiIiIiclLMRgcQEREREREREREREbFnKrSLiIiIiIiIiIiIiJwCFdpFRERERERERERERE6BCu0iIiIiIiIiIiIiIqdAhXYRERERERERERERkVOgQruIiIiIiIiIiIiIyClQoV3EjsTGxjJ//nyjY4iIiIiIiIiIiMgfmKxWq9XoECJyfDZv3kx4eDjBwcFGRxEREREREREREZHfqdAuIiIiIiIiIiIiInIK1DpGxMZs2bKF6dOn07dvXwYNGsStt95KVlYWcGTrmLVr13L++efTq1cvJk6cyI8//kh8fDxffPEFAC+99BJTp07lq6++YsyYMfTq1YuZM2eSm5vLggULGDFiBP379+fuu++mqqqqcb179uzh1ltvZciQISQkJDBq1Chefvll9LuciIjIkWpqapg1axZDhw6lX79+PPDAAzz33HOMGjUKgPLycmbNmsXIkSNJSEhgyJAh3HfffZSWljauIzY2ls8++4xbbrmFPn36MGzYMD788ENycnK47rrr6N27N+PGjeP7779v/MyMGTN4/PHHmT17NkOGDKFfv3489NBDVFZWMmvWLAYMGMCwYcN47bXXmuT98ccfueyyy+jbty89e/bkvPPOY/ny5a1zsERERNqhY10riEjboEK7iA2pqqriuuuuIyQkhFdeeYXHHnuMpKQk7rzzziOWTU5O5tprryUwMJCXXnqJKVOmcPvtt9PQ0NBkub179/LGG29w7733MmvWLLZs2cKMGTP4/PPPeeihh7j++utZtGgR7777LgAVFRVcfvnlFBcX8/TTT/Paa68xePBgXnzxRdasWdMqx0FERMSe/POf/+SLL77g5ptv5tlnn2X//v289dZbje/fddddrF69mrvuuov58+dz1VVXsWjRIl555ZUm63nyySfp2LEjr776Kn379uWxxx5j5syZ9OvXjxdeeAFPT0/uueeeJj+Of/7556SlpfHcc89x1VVXsWDBAqZMmUJZWRnPP/88Q4cO5bnnniMxMRGArVu3ct1119G1a1deeeUVnn/+edzc3LjrrrsoLCxsnQMmIiLSzhzrWkFE2gZHowOIyP+kpqZSXFzMjBkz6Nu3LwB+fn6sXbsWi8XSZNnXX3+d0NBQ5s6di6OjI2eeeSZms5mnn366yXKVlZU88cQT9O7dG4DvvvuOxYsXs3r1aiIiIhg5ciTfffcdW7ZsAQ4V5qOiopgzZw7+/v4ADB06lJUrV7J+/Xr94i4iIvIHe/fuZdGiRTz55JNMnToVgCFDhnDWWWcBh0aw1dXV8fDDDzN8+HAABg8eTGJiIuvWrWuyrr59+3L33XcDEBISwvLly+nTpw833HADAC4uLsycOZP09HTi4uIAcHBwYM6cObi6unLaaaexYMECLBYLTz75JGazmcGDB7N48WK2bNlC3759SU1NZcyYMTz00EON2w0PD2fKlCls2bKFkSNHtuwBExERaWeOda0gIm2HCu0iNqRTp074+vpyww03MHHiRM4880yGDh3KoEGDjlh23bp1jB8/HkfH//1vPH78+CMK7SaTiYSEhMb/DggIwN/fn4iIiMbXfH19KSsrAyAhIYEPP/yQuro6du/eTXp6OklJSdTX11NbW9vcuywiImLX1q9fD8Do0aMbX3Nzc+PMM8/kt99+w8XFhTfffBOAzMxM0tPTSU1NJS0tDRcXlybr6tWrV+O/BwYGAjQ5h/v6+gIc0XLG1dW18b8DAgKIjo7GbD704KqzszPu7u6N5/lp06Yxbdo0KisrSUtLIz09nbVr1wLoPC8iItICjnWtICJthwrtIjbE09OT999/n5dffpkvv/ySDz74AG9vb+644w4uvfTSJssWFRU1jjg/7PBN+R+5ubnh4OBwxGt/57///S/z5s2jrKyMiIgI+vbti6Ojo3q0i4iI/ElRURFOTk54e3s3ef2P5+RVq1bx5JNPkpGRgZ+fHwkJCbi6uh7xtJqHh8cR6z/WOftEP1NZWcmDDz7I0qVLAYiJiaF79+4AOs+LiIi0gOO5VhCRtkGFdhEb07VrV+bMmUNtbS0bN27knXfe4ZFHHqFHjx5NlgsODj6il2pz9Fb96quvmDNnDg899BCTJk3Cy8sLONQ+RkRERJoKDg6mrq6O0tLSJjfQh8/J6enp3HbbbUyZMoX333+f0NBQAG677TbS0tJaPe9jjz3Gzz//zOuvv87AgQNxdnZm9+7dfPPNN62eRUREpD041rWCiLQdmgxVxIb88MMPDB06lMLCQpydnRk6dCj//ve/AcjOzm6y7MCBA/n++++bjIZbtWrVKWdITEwkNDSUSy65pLHIvmPHDgoLCzXSTURE5E/69euH2Wxm9erVja/V1tby448/ApCUlERdXR3XXXddY5G9srKSjRs3GnJe3bx5M2eccQann346zs7OAI1ZdZ4XERFpfse6VhCRtkMj2kVsSK9evbBardx8881ce+21ODk58c477+Dt7c3gwYObLHvddddx3nnnccstt3DRRReRnp7OCy+8ANDYl/Vk9OzZkwULFjB37lwGDRpEWloaL7/8MiaTierq6lPaPxERkbamY8eOnHPOOcyaNYvKykoiIiJ49913ycvLIzw8nLi4OBwcHJg9ezaXXHIJRUVFvPnmm+Tn5zcWultTz549Wb16NV9++SVhYWGsXbuW+fPnA+g8LyIi0gKOda0gIm2HCu0iNsTX15d58+bx7LPPcu+991JXV0evXr146623jujH3rlzZ/773/8ye/ZsbrzxRqKjo7n//vt54IEHcHd3P+kMU6dOZe/evSxYsIB58+YRERHB1VdfTVpaGhs3bjzVXRQREWlzHn74YVxdXZkzZw719fVMmjSJ8ePHs3v3bmJiYnj66aeZO3cu1113HUFBQQwfPpxp06bx6KOPkpOTQ0hISKtl/cc//kF1dTVPPPEEcOh6Yu7cuTzxxBMkJiYyZcqUVssiIiLSXvzdtYKItB0mq54RFbFLv/zyCx4eHvTu3bvxtZ9++omrr76ahQsXNk5sJiIiIi2nsLCQn3/+mZEjR+Lp6dn4+sUXX0xgYCBz5841MJ2IiIgYTdcKIu2HRrSL2KnNmzczf/587rvvPmJiYsjKyuLFF19k4MCBKrKLiIi0EldXVx555BGWLVvGxRdfjKOjI0uXLmXz5s289dZbRscTERERg+laQaT90Ih2ETtVX1/P3LlzWbRoETk5Ofj4+DBmzBjuuuuuJr+Si4iISMvaunUrzz//PNu3b6euro7Y2Fj+7//+jxEjRhgdTURERGyArhVE2gcV2kVEREREREREREREToHZ6AAiIiIiIiIiIiIiIvZMhXYRERERERERERERkVPwt5Oh5uWVtVaOE+bp6UJ5eY3RMeyOjtuJ0zE7OTpuJ07H7OS0h+MWFOR1Sp+35fO5EdrDnxlbpONuDB13Y+i4G8eWj73O583Llr/r9kzfi23S92Kb9L3YrmN9N391TrfbEe2Ojg5GR7BLOm4nTsfs5Oi4nTgds5Oj4yYnSn9mjKHjbgwdd2PouBtHx7790Hdtm/S92CZ9L7ZJ34vtOtnvxm4L7SIiIiIiIiIiIiIitkCFdhERERERERERERGRU6BCu4iIiIiIiIiIiIjIKVChXURERERERERERETkFKjQLiIiIiIiIiIiIiJyClRoFxERERERERERERE5BSq0i4iIiIiIiIiIiIicAhXaRUREREREREREREROgQrtIiIiIiIiIiIiIiKnwNHoAGI8q9VKSXU9DiYTni4OmEwmoyOJiIiItDlVdQ1U1Dbg5eKIi6PGu4iIiBitoraesup6ALxdnXB3djA4kYjYMxXa26nskmq+3ZXL6pR80gsrqa63AODu5EBMgDvDOvkzsmsgnQM9DE4qIiIiYr/S8iv4bHM2mzJLSC+sxGI99LqfmxOjY4OYGB9MjzBvY0OKiIi0I/UNFpYk5bIiOY/1GcU0/H5ydjBBXKgXQzr6cUHfcPzdnQ1OKiL2RoX2dqagopanlqewcNtBrECvcG+m9g4jxMsFixUOllaTdLCM137Zx2u/7GNwR1+uHBxF/w6+RkcXERERsRv7Cit5dk0av6YX4eJoZkAHX87qFoifuzPlNfWk5Jbz9faDfLo5m9HdArlrZGcCPV2Mji0iItKmbc0u5amVqaTmVdDB15VL+0UQ7e+OFSvZpTVs3F/Mm7/t5/0NmVzYN4Krh0RplLuIHDcV2tuR1Sl5PL4ilYraBi7uF8FF/cKJ8HE76rL5FbUs3pHDhxszueGTrYzoEsA9o7oQ7KUbQBEREZG/YrFa+Tgxm5d/3IuLo5kbh0UzpVcYvm5ORyxbXlPPx4lZvLl2P2v3FfHI2d0Z3jnAgNQiIiJt3/sbMnnh+z0EezrzzLnxjOgScGTr3NMhvbCSeb/u4731GXy/O5/HJ8URG+xpTGgRsStqDtkOWK1W3v5tP/d9s5NOQR58eHk/7hzZ+S+L7ACBHs5cMagDC68ZxE3Dovk1vYgL397A19sOYrVaWzG9iIiIiH2oqbfwwKKdPLcmjYFRviy4oj9XDo46apEdwNPFkauHdOTDy/vTwdeNexfu4JvtB1s5tYiISNtmtVp59ae9vPD9Hs7qFsinVw5kZNfAv5yfLtrfnVkT43jlgl5U1DZw5YeJLEnKaeXUImKPVGhv46xWK8+uSePln9IZ1z2I968cRKeA4++77urkwMzBUSy4oj/dQzx5bHkK/16yi4ra+hZMLSIiImJfiqvquOnTraxMyefW4TE8N7kHQcfZCqajvzv/vbA3A6J8efTbFD5JzGrhtCIiIu3Hf3/Zx5u/ZXBeQiiPT4w77lYwA6J8+fDyfvQO9+ahpcl8uDGzhZOKiL1Tob2Ne3tdBh8nZnNJvwgem9AdF6eT6y0W6evGy+f34obTO7IiOY+ZHySSWVzVzGlFRERE7E9pdR03frqVnTllPHVOHDMGdvjLUXJ/xd3ZgecmJ3Bm5wD+szqN73cXtFBaERGR9mNFch5vrt3PeQmhPDC2Kw7mEzs/+7k788LUnozqGsjz3+3hjV/2tVBSEWkLVGhvwxbtOMgrP6UzPi6Y20d0OuEbvj9zMJu4ekhHXrmgF0WVdcz8IJHEzJJmSisiIiJifyprG7j9ix2kF1by7OQenNUt6KTX5exoZtbE7sSFevGvxTvZlVPWjElFRETal+Tcch5dlkzvcG/uG93lpGsizo5mnpgUx6QeIbz+6z4+3qQnz0Tk6FRob6OSc8t5YkUqA6N8eXBcN8ynWGT/o/4dfHnz0r74uDlx82db+XlPYbOtW0RERMRe1Fus/OObJHYcLGXWxDiGRPuf8jpdnRx4dnIPfN2cuHthEmXVatcnIiJyomrqLfxz0U68XR156tx4nBxOrfzlYDbxwNhujOgSwLNr0li+K7eZkopIW6JCextUVdfAA4t24uPqxBMT4075hHI0UX5uzL+kD50CPLh74Q6+353f7NsQERERsWUv/bCHX9OL+MforozqGths6w30cOapc+PJL6/hmdW7m229IiIi7cWba/exv6iKB8fFEujh3CzrdDSbeGxCd/pE+vDIsmR2HChtlvWKSNuhQnsb9OyaNPYXVfHohFh83Z1abDu+bk68ckEvuod4ct83O1mZnNdi2xIRERGxJV9vP8iHG7O4qG84U3qFNfv6e4R6cfXQjizbmatRcyIiIidgd14F76zPZGJ8MIOj/Zp13a5ODjxzbjyBHs7c83US+RW1zbp+EbFvKrS3Mb/tK2LhtoNcPqgDA6Oa94RyNF6ujrw0rSc9w7x4YPFOlu3UjaCIiIi0bbtyynhq5aEWfbeP6Nxi27lycBQJYV48vWo3BbqRFxEROSaL1crjK1LwcnHk9jNb5hzt6+bE7PN6UFZdz31fJ1HXYGmR7YiI/VGhvQ2pa7Awe9VuIn1duXZox1bbrqeLIy9M7Um/SB8eXLJLxXYRERFps8pr6rl/0U783A616HM0N988OH/maDbx0LhYKmsbeOmHPS22HRERkbZiZXIe2w+UcevwmBZ9wr9bsCf/HteNrdml/PfnfS22HRGxLyq0tyEfbcxiX1EVd4/sgotj63617s4OPD8lgX4dfHh4WTK/7NUEqSIiItK2WK1WHl+ewoGSap6YFNeiN/CHRQe4c9mASBYn5ZKYWdLi2xMREbFXtfUWXv4pna5BHkyID2nx7Y3tHszknqG8uz6D3/YVtfj2RMT2qdDeRuSV1zBv7T6Gdw7g9E7+hmRwdXLgP+f1oEugB/d9ncS2bE0MIiIiIm3HN9tzWJmSz/8Ni6F3hE+rbffqIVGEebvw9KpU6vV4uoiIyFF9vvUA2SXV3DI8BocWfOLsj+4a2ZkYf3ceWppMUaXavIm0dyq0txHz1+6nrsHKHSM6GZrjUBuZBAI9nbnjy+3sKagwNI+IiIhIc8guqea579IY0MGHGQMjW3Xbrk4O3DWyM2n5lXy57WCrbltERMQelNfUM//XfQyK8mVIx5afr+4wVycHZk3sTklVHc+sSmu17YqIbVKhvQ3IKqniq20HmdIrjEhfN6PjEODhzEvTeuLoYOaWz7ZxsLTa6EgiIiIiJ81itfLIsmQAHhwfi9nUOqPk/mh45wD6Rvow79d9VNY2tPr2RUREbNnnWw5QUl3PTWfEYGrl83S3YE+uHdqRlSl5rE7Ja9Vti4htUaG9DXjj1/04mk1cObiD0VEaRfq68eLUBCpqG7j1i+1U1NYbHUlERETkpHyamM2mzBLuHNGZMG9XQzKYTCZuOSOGwso6PtyYaUgGERERW1RTb+HDjZkM7uhLfKiXIRkuH9SBuBBPnl61m+LKOkMyiIjxVGi3c+kFlSxNymFa7zCCPF2MjtNEt2BPZp8Xz/7CSh5emozFajU6koiIiMgJOVhazSs/pTMk2o9zElp+YrW/0zPcmxFdAnh/Q6b6wIqIiPxu8Y6DFFbWccUg4wYfOppNPDgultLqeuZ8rxYyIu2VCu127u11+3F2MBt6Qvk7A6P8uG1EZ77bXcD8X/cbHUdERETkuFmtVp5etRuL1cr9o7u2+qPoR3PTsBiq6hp4Z51GtYuIiNRbrLy7PpMeoV4M6OBraJYuQR7MGBjJ4qRcEjNLDM0iIsZQod2O5ZTVsGxXHuf1DMXf3dnoOH/p4r7hTOwRwuu/7uP73flGxxERERE5LqtS8vlpTyE3nB5NuI8xLWP+LDrAnbHdg/l8S7YeTRcRkXZvTWo+WSXVXDGog038IH7V4ChCvVx4elUq9Q0Wo+OISCtTod2OfZKYhdVq5ZL+EUZH+Vsmk4n7R3clPtSLB5cks6egwuhIIiIiIn+rorae575Lo3uwJxf1s61rrSsHdzjUj3aTRrWLiEj7tmBTFpG+rpzZJcDoKAC4Ojlw18jOpOVX8snmbKPjiEgrU6HdTpXX1PP5lgOM6hpEhI+b0XGOycXRzDPnxuPqZObur3ZQVq3JUUVERMR2vbl2P3nltdx7VhcczcaPkPujTgEejOoWyCeJ2ZRWa1S7iIi0T7tyytiaXcoFfcIx28Bo9sPO7BLA6TH+vP7LPvLKa4yOIyKtSIV2O/X19oNU1DZw2cBIo6MctxAvF545N57s0hoeX5GCVZOjioiIiA1KL6jkg41ZnJsQQs9wb6PjHNVVg6OoqG3g40SNlhMRkfbp48RsXB3NnNMj1OgoTZhMJu4e1Zm6BgsvfL/H6Dgi0opUaLdDFquVTxKz6RvhTY9QL6PjnJDeET7ceHo0q1Ly+XzLAaPjiIiIiDRhtVp5dk0a7k4O3HRGjNFx/lK3YE+GdfLnk8RsqusajI4jIiLSqooqa1m+K5eJPULwcnU0Os4RIn3duGJQB77dlcf6/UVGxxGRVqJCux36bV8RWSXVnN8n3OgoJ+WygZEMjfbj+e/SSMktNzqOiIiISKO1+4pYu6+Ia4ZG2fRk8wCXDYikuKqOJUk5RkcRERFpVQu3HaS2wcoFNlwXuXxgB8J9XJm9Ko16i57oF2kPVGi3Q59vPoCfmxMjuwYaHeWkmE0mHjk7Fi9XJx5cuouaes3ELSIiIsZrsFh56Ye9RPi42vSN+2H9In2IC/Hkg41ZWNSST0RE2gmL1cpX2w7SL9KHzoEeRsf5S65ODtx+Zif2FlayeMdBo+OISCtQod3O5JTV8OOeAs7tGYqTg/1+fX7uzvx7XDfS8it55ae9RscRERERYenOHFLzKrhxWLRdXGeZTCYuGxDJ/qIqfkwrNDqOiIhIq9iYUUxWSTWTe9lWb/ajGdElgJ5hXrz+yz61ehNpB2z/DkKa+GrrAaxWmGIHJ5RjOT3Gn2m9w/hwY5Z6lomIiIihqusaePWndOJDvRgTG2R0nOM2qlsQoV4ufLAhw+goIiIirWLhtoN4uTgysovtP+VvMpm46YwYcstr+XSzJjAXaetUaLcjDRYrC7cfZEi0HxE+bkbHaRa3ndmJKD83Zn2bQmWtft0VERERYyzYlEVueS23Do/BZDIZHee4OZpNXNQvgsSsUlLzNPeNiIi0bcVVdaxOzWdCfDCuTg5Gxzku/Tv4MjTaj7fXZVBWXW90HBFpQSq025F1+4vIK69lck/7H81+mJuTA/8e243s0hq1kBERERFDFFfW8fa6DM7o5E//Dr5Gxzlh5/QIwcXRrJFyIiLS5i3dmUtdg5Xz7KwuctOwGEqr63lfT6CJtGkqtNuRxTty8HJxZFinAKOjNKs+kT5c0CecTxKz2ZJVYnQcERERaWfm/7afqroGbh4eY3SUk+Lj5sS47kEsTcrVSDkREWmzrFYrX287SFyIJ12DPI2Oc0JiQzwZGxvEhxuzyC+vMTqOiLQQFdrtRHlNPd/tLmBs9yCcHdve13bTGdGEeLnw+PJU6hosRscRERGRdiKzuIrPNmdzbkIonQI8jI5z0i7oE051vYVFSTlGRxEREWkRKXkV7M6v4JwE+xrNftgNp0dTZ7Eyf+1+o6OISAtxNDqAHJ/VKfnU1FuYEB9idJQW4eHsyL1ndeHOr3bwSWI20wdEHrHMqlXLee+9t8nM3E9AQCCTJk3m0ktn4OBw9L5shYUFvPLKi6xfv5aamhri4xO4+ebb6dSpS0vvjoiIiNiJ+Wv342A2cd1pHY2O0sSuXUm88MJ/SElJJigomCuuuJqzz570l8v7myoJ2PoBryzeyTvOZgYPHsott9yJr+//9mv16pW8/fYbZGdnERoaxiWXzGDixHNbY3dERERO2eIdOTiaTYy1k0nL1679hTfeeJX9+9MJCwtnxowrmdyzM19uO8j0AZFE+jade2/PnjQuv/yiI9bz8svz6N27T2vFFpFT0PaGRrdRi5NyiPJzo2eYl9FRWsywTv6cFuPHG7/uI7+itsl7v/76M48++m/OOec83nlnATfccDMffPAO77331lHXZbFY+Oc/7yEjYz9PPvksr776Jh4entx2242UlBS3xu6IiIiIjcsoqmJpUg5Te4UR5OlidJxGRUVF3HnnLXTr1p033/yA88+/iKeeeox169YedXmr1co999xOoHM9NcP+j+vv/w8FBQXce+8djcts2ZLIo4/+i6lTL+SddxZw/vkX88wzj/PLLz+11m6JiIictPoGC9/uyuWMzgH4uDkZHeeYtm7dzD333EavXn2YN+89LrvsSmbPfpLO5Uk4mk28eZRR7Xv3puHr68vChcua/NOjR4IBeyAiJ0OFdjtwoLSaTZklnB0XjMlkMjpOizGZTNw5ojM19RZe/rHpxKgLF37OmWeOYtq0i4iIiGTkyNFcfPF0Fi/+5qjr2r07he3bt3L//Q8SH59ATEwn/v3vR6mqqtQNpYiIiAAwf+0+HB3MXD6og9FRmli06KvfBwjcTceO0Zx//sWMGzeBjz5676jLFxYWEB0dzRMPPoJvaDQ/F7lx0UWXkpKyi5KSQ/Pf/Pjj93Tq1IXJk6cRERHJ5MnT6NYtlnXrfm3NXRMRETkpv6YXUVhZx0Q7ecr/o4/eIyGhF7fddhcdO0Yzdux4pk+/nE8+mM+UXmEsScohs7iqyWf27EkjOroTAQGBTf5xdFQzChF7of9b7cDK5DwAxscFn/K6EhLieeihWXzxxackJ+8kPDyCf/zjQVJTk3nvvbeoqChnyJDTeeCBh3F2dgYOjYB69dWXSElJJiAgkLPOGsOVV16Li8uhkV+pqSm89tpctm/fSnV1NWFh4Vx++VWNjzfffPN1JCT0Ii8vhx9//AEPDw+GDx/BLbfciaOjI/Pnv8Zbb73RmNEBWAGsmAVnnz2JBx54mCuuuBpX16aPVZlMJsrKSo+6nyEhoTzzzByiov73uLTZbMZqtVJWVnbKx1FERERs37BhA/7yuuetd96koLiUDnH98HYa1PiZ1r7u+aPD1z1btiTSp09fzOb/jYnp27c/zz77FBaLpcnrAAEBgTzyyJMATO5VxzvfbaMmfzlxcfH4+PhQXFyJr68v6el72LRpA3379mfLlkT27Elj6tQLm/WYi4iINKfD5/Ln572Dy4G9vL6pAwH321YN448On8szMjIYNmx4k/e6dYvlwIFszu7oyBdbTLy9LoN/je3W+P6ePWl07BjdAkdRRFqLCu12YEVyHnEhnkf07zpZL730PPfe+wBRUVHMmvUw99xzG/HxCfznPy+wf/8+HnnkX/Tp048pU84nNTWZO++8mauvvoF//esRcnIOMmfObAoLC/jnPx+iqqqKO++8mdNPP4PXX38bqxUWLHifZ555nMGDh+LvHwDAxx9/wBVXXM28ee+yfv1vzJkzm/j4BMaNm8All8xg8uRpjdH8nzIAACAASURBVPkq6xq45qMtBHo6c8vF/QGIi+vRZB8qKsr56qvPGTx46FH30cfHl9NOG9bktU8/XUBtbS2DBg1pluMoIiIitu+vrntip97Jr9uSObjhfRYv/tqw654/cnFxBSAvL5euXWObvBcYGEh1dTWlpaX4+vr+5f6mfvUiTmt/ZIebB6+/Oq/x9alTL2Tbti3ceusNODg40NDQwCWXzPjbvu8iIiK24MUXn6Mkbgqjp/0fB1a+aXM1jD86fC4PDAwkN7fpBOUHDhwAwFxbweSeYXy+9QBXD4kizPvQZ/buTaO2tpbrrpvJwYMHiInpzPXX30h8vFrHiNgLtY6xcRlFVezMKWdMM072MXHiuQwbNpyoqGjGjZtAWVkpd911H506dWHEiLPo0qUbe/akAfDRR+8zdOgwLr10BpGRHejffyD33PNPliz5hvz8fKqrq7jooku5/fZ7iIqKpmPHaGbMuJK6ujoyMv7Xc6xLl27MnHkNUVEdmTbtQjp37sr27dsAcHd3b/JYVIfQEG4f35uUUjPfpVcckb+6upr777+bmpoa/u//bjmuff7pp+957bW5XHTRdKKjY5rhKIqIiIg9ONp1z+XX385PBS5MPnscXQ2+7vnjP56ensCha53Do+4Oc3I6NEqvtrbmb/f3phtuJOGyB2nwj+H2228kJ+fQTX5RUSGFhYXceOOtzJv3LrfffjdffPEJixYtbJ4DLSIi0kLihoymNqQH5w/rY5M1jKOdy8eNm8CqVctZtWoF9fX1pKYm89FH7wNQV1fH5YM6YDbBO+syAKipqSY7O4uKinJuuuk2nnrqWQIDA7n55utJT2/aWldEbJdGtNu4lSmH2saMbsZCe2Tk//qQurm5YTabCQsLb3zNxcWFurpDk5GmpCSTmbmfMWPOaHzfarUCsG/fXvr3H8iUKeezbNmi35fNIDU1BYCGhobGz0RFRTXJ4OnpSX19HQDvvvvmUSc1da2z8Ny2AYz877N4uR76o1pcXMw//nEn6el7eP75lwkNDTvm/i5Z8g1PPz2L0aPHcuONtx5zeREREWk7jnbdsyrLgtVqZfqASB7/wjauewDGjj2be+75Jy4uLtTWNp0Y/nDGP7fS+7MuXbpypVMgN5e447z6cRYu/IoLL5zBM888TrdusVx66eUAdO0aS3FxMa+++iITJ57bpucBEhER+5bV4EmYtwsJYV5k2HANA/53Lj/77Enk5BzkiSce5pFHHiAkJJSLL76M559/Bg8PT0K8XJgQF8KiHTlcf1pH/NxdWbp0NU5Ozo0tcLp3jyc5eRdffvkpd9xx76keRhFpBSq027gVyXn0DPNqfJSoOfx5Ig2TyfSXN1dOTo6cffYkpk+/4oj3AgICyc/P5/rrZxIUFMzpp5/BaaedQWBgENdcM+NP63E+4vOHT3aTJ09j1KgxR7y/J7+Cu5ek8fa6DG4ZHsOBA9ncccfNVFVVMHfuG3Tp0vWY+/rOO/N5441XmTbtQm6//R7dRIqIiLQzR04gZuKrbQcZFxd8xPWVkdc9AB4eHgAEB4dQUJDf5L38/Hzc3NwbR8r9UWFhAZs2bWD06HEADIzyJSLAmwrPQHJzcwHYsWNb4/uHxccn8Pbb8ygrK8Pb2/uomURERIy2p6iGS0cENd7P22oNA/53LgeYOfMaLrtsJsXFRQQEBPLzzz/g4ODQOGBw+oBIFm4/yGdbDnDt0I54eDQ9x5vNZmJiOh3RgkZEbJcK7TYsvbCS1LwK7hjRybAMMTGdSU/f22Q02PbtW/ngg3e4++77+f771VRWVvLyy2/g4OAAwG+//XpC2/D29sHb2+eI1yMj4ewD8HFiFuNi3Hjgzhswm828+uqbhIdHHHO9H3zwDm+88SrXXHMDM2dec0KZREREpG2yYqW6zsKMgR2OeM/I654/6tWrD0uWfIPVam0sJGzatIGePXsfMREqwMGDB3j44QeIjOxA9+7xmE0mxnf24r2CA/iFHtqXoKBg0tJSm3xuz57d+Pj4qMguIiI2zWKxHnc7XVs5l3/++cdkZmZy2213ERh4KPsPP3xHQkIv3N3dD2UNcGdYJ38+TcxmsFcZd995Ey+99Bqxsd2BQyPsU1NTGDnyrBPKJyLGUY92G7Ym9dBIprO6NV/bmBM1ffoVJCVt56WXnmPfvnQSEzcya9ZDlJWVERAQiK+vH5WVFXz33WoOHjzATz99z+zZTwAc8cjzybj+tGgsViv3P/ooxcXFPPzw47i4uFBQkE9BQT6FhQWNy5aWllBaWgLA7t2pvP76K0yceC7nnDO5cfmCgnyqqqpOOZeIiIjYn9oGC1YrDOvkT5dAjyPeN/q657BJk86juLiI2bOfID19L599toAVK5Yxffrljcv88bqne/d4evfuy1NPzSIpaTspKbvY/uVL4OJBeVhfAC644BIWLvyCr776jOzsLJYvX8Z7773FZZdd2Wy5RUREWkKAhxOxwUc+0XU0tnIuj4qK5osvPuHbb5dw4EA27733FsuXL+Waa25oXKa0tIQp3b0pqqojpc6XsLAwnnnmcXbs2M6ePWk88cQjlJQUc+GFlzRbLhFpWRrRbsPWpObTI9SLEC+XYy/cQjp37sLs2S/wxhuv8uWXn+Hh4cnpp5/BjTfeBsCoUaPZuXMHc+bMpqqqkoiIDsyceQ3vv/82u3YlMWTIaae0/XAfV6b2COTLz9Zjwsq11zZ9/MvBwYHvv/8NgH/+8x4A5s59nVWrltPQ0MDixV+zePHXTT6jEe4iIiLt0+bMEqzAzEFHjmYH4697DvP3D+A//3mJF16YzVVXTSckJJR//esR+vcf2LjMH697zGYzjz8+m5dfnsO9995BbW0tgwYNofel/+KbHYVcPdTK1KkX4OTkxCeffMjcuXMIDQ3nuutuYurUC5ols4iISHMrqDhU+E4I8z7uNrC2ci4fOHAwd999P2+++Tr5+Xl06tSZp59+nr59+zcuc/hcHjfkBj5KPMArs1/gv6++xH333UF1dRU9e/bm5Zdfx8/Pv1kyiUjLM1kPN5k6iry8stbMckJ8fd0pLq40OkaLOVhazTlvrOOmYdHMHBx17A8cJ3s8bsVVdUyet45BHf145tz4Vt++PR4zW6DjduJ0zE5OezhuQUFep/R5Wz6fG6E9/JmxRUYf9/oGC1PmryfU24U3Lu5jWI7WtDo1n/u+TuL5KT0Y1inA6DjtitF/3tszWz72Op83L1v+rtuz5vxePt2czTOrdvPRFf2P+iRaW7F8Vy4PLN7Ff86L58wugS2yDf3/Ypv0vdiuY303f3VOV+sYG/VD2qGWKCNa6C9Ze+Lr5sTF/SJYk5pPal650XFERETEDi1PzuNgWQ1X/MVo9rbojE7+BHg4s3DbQaOjiIiInLAVyXnEBLi36SI7wKhuQYR7u/D+hkyjo4jIKVKh3Uat2V1AtL8b0QHuRkexCZf0i8DD2YH5a/cbHUVERETsjNVq5d31GXQOdOf0mPbz+LWTg5nJfcL5cU9h4+P3IiIi9iC3rIbNmSXHPQmqPXM0m7ikfySbs0rZll1qdBwROQUqtNugkqo6EjOKW+yRIXvk8/uo9lUp+ezOqzA6joiIiNiRDRnFpOVXMr1/5HH3eG0rLugfSYPFypKkHKOjiIiIHLeVKXlYoV0U2gHOTQjF29VRo9pF7JwK7Tbopz2FNFhhZBf10vyj/41q32d0FBEREbEjH2/Kxs/NibHdg42O0uo6B3nSK9ybhdsO8jdTM4mIiNiUlcl5dAvyINq/fTzl7+7swNReYaxJzSe7pNroOCJyklRot0E/pBUQ6OFMXOipTZbT1vi4OXFRvwhWpuSzO1+j2kVEROTYskqq+CGtgCm9w3BxbJ+Xvuf1DGVfURVb9Ti6iIjYgeySarYdKGs3o9kPm9Y7DJMJvtx6wOgoInKS2ufdhg2ra7CwNr2IYZ38MbezR5uPx6WHR7X/ql7tIiIicmyfJh7AbDYxrVeY0VEMM7pbEO5ODnylSVFFRMQOrEzOA2B0Oyu0h3q7MrxzAAu3HaS23mJ0HBE5CSq025hNmSVU1jUwrJPaxhyNj5sTF/UNZ1VKHmka1S4iIiJ/o6qugYXbDzCqayDBXi5GxzGMu7MDY2KDWJWSR2Vtg9FxRERE/taK5DziQ72I9HUzOkqrO793OEVVdaxOzTc6ioicBBXabcxPewpxcTQzuKOv0VFs1iX9I3FzcmCeRrWLiIjI31i+K5fymgYu7BNudBTDTeoRQlWdhTW6cRcRERuWUVTFrtxyxraz0eyHDezoS5SfG59tzjY6ioicBBXabYjVauXHtAIGdPDF1cnB6Dg2y9fNiQt/H9W+v6jK6DgiIiJio77cepCYAHd6R3gbHcVwvSO8ifR1ZdEOtY8RERHbteL3tjFndQs0OIkxzCYT03qHsSW7lJTccqPjiMgJUqHdhuwrrCKrpJphnfyNjmLzLu4XgZODifc3ZBgdRURERGxQcm45Ow6WMbVXGCbNe4PJZGJifAgbMko4UFptdBwREZGjWp6cS+9wb0K9XY2OYphJPUJwcTTz+RZNiipib1RotyE/7ikAUKH9OAR4ODOpRyiLd+SQX1FrdBwRERGxMV9uPYCLo5kJ8cFGR7EZE+JDAFi8I8fgJCIiIkfaU1BBWn4lY9pp25jDvF2dGNc9iKU7cyivqTc6joicABXabcjPewvpEujRrn+5PRHTB0RS12Dlk8Qso6OIiIiIDamsbWDZzlxGdwvE29XJ6Dg2I9zHlQEdfFiclIPVajU6joiISBMrduVhNsFZ7bzQDnB+n3Cq6iz6cVzEzqjQbiMqauvZnFXKaTF+RkexG1F+bozsGshnmw9QUatfeUVEROSQ5btyqahtYEqvMKOj2JyJPULILK5mS1ap0VFEREQaWa1WViTn0S/Sh0APZ6PjGC4uxIseoV58tiVbP46L2BEV2m3Ehv0lNFisDIlWof1EXD6oA2U19Xy1VRN7iYiIyCFfbD1ApwB3eoVrEtQ/G9U1CDcnM4uSNEJORERsR0peBfuKqtp925g/Or9PGOmFVWzMKDE6iogcJxXabcTa9EJcHc30DvcxOopd6RHqRf8OPny4MZO6BovRcURERMRgu3LK2JlTrklQ/4K7swOjugWxMjmP6roGo+OIiIgAsCI5DwfToR+E5ZDR3YLwcnHkq22aFFXEXqjQbiN+21fEgChfnB31lZyoGQM7kFtey/JdeUZHEREREYN90TgJaojRUWzWOT1CqKht4LvdBUZHERERaWwbM7CjH77umlvlMFcnB8bHBbMmNZ/S6jqj44jIcVBV1wZkFleRUVzNkI5qG3MyTov2o0ugB++uz1DvMhERkXasoraeb3fmMTo2CC9XR6Pj2Ky+kT6EebtogjUREbEJSTnlZJdUq23MUZyXEEptg5VlO3ONjiIix0GFdhuwNr0IQP3ZT5LJZGLGwEj2FFTy895Co+OIiIiIQb7dlUdlXQNTNQnq3zKbTEyMD+G3fUXklNUYHUdERNq55btycTSbGNElwOgoNic2xJPYYE8WbtO8dCL2QIV2G7A2vYhwbxei/NyMjmK3xsYGEeLlwnvrM42OIiIiIgZZtD2HmAB3eoZ5GR3F5k3sEYIVWKpJUUVExEAWq5WVyXkMifbD21VtY47m3IRQUvIqSM4pNzqKiByDCu0Gq2+wsCGjmCHR/pqw6xQ4Opi5pF8EmzJL2HGwzOg4IiIi0sr2FVay7UApk+JDdE11HCJ93egb4c2iHTlqvSciIobZll1KbnktY7urbcxfGR8XhLODiYXbNapdxNap0G6wrQdKqahtUNuYZjC5VyieLg68vz7D6CgiIiLSypbszMVsgrPjg42OYjcm9ghhX1GVBimIiIhhViTn4eJoZnhntY35K96uTozsGsiynblU1zUYHUdE/oYK7QZbm16EgwkGRvkaHcXueTg7MrVXOKtT88ksrjI6joiIiLQSi9XK0qQcBkX5EeTpYnQcu3FWtyBcHM0s0qSoIiJigAaLlZUp+ZwW44+HsyYx/zvn9QylrKae73YXGB1FRP6GCu0GW5teREKYN54uOqk0h4v7hWM2mfhwY5bRUURERKSVJGaWcKC0hok9QoyOYlc8XRwZ2TWQ5bvyqKm3GB1HRETamcTMEgoqahkTq7Yxx9K/gy/hPq5qHyNi41RoN1BRZS27csrVNqYZBXm6cHZcMF9vP0hxZZ3RcURERKQVLN6Rg7uTAyO66LHzEzWpRwhlNfX8kKYRciIi0rpWJOfh5mRmWCd/o6PYPLPJxDk9Qtiwv1hP8IvYMBXaDfTbvmKswFAV2pvV9AGR1NRb+GxLttFRREREpIVV1zWwKiWfs7oF4urkYHQcuzOggy/Bns4sSVL7GBERaT31DRZWpeRxRqcA3HT+Pi6TeoRgArV8E7FhKrQbaG16IT6ujnQP8TI6SpvSOdCD02P8+SQxWxOFiIiItHFrdudTWdegtjEnycFsYnxcML+mF1FYWWt0HBERaSfWZxRTUl2vtjEnINTblSHRfizakYPFajU6jogchQrtBrFarazdV8ygjn44mE1Gx2lzZgyMpKiqTqOzRERE2rglO3IJ83ahb6SP0VHs1tnxITRYrCzflWd0FBERaSdW7MrDw9mBoTFqG3MiJvUIIaesho0ZxUZHEZGjUKHdILvzKyioqFV/9hbSL9KHuBBPPtiYRYNFv/SKiIi0RbllNazbX8SE+BDMJg1cOFldAj2IDfbUAAUREWkVtfUWvttdwIguAbg4qix1IoZ3DsDD2YHFSblGRxGRo9DfaAZZv//Qr4+DonwNTtI2mUwmZgzswP6iKk3uJSIi0kYt25mLxQoT4tU25lRNiA9mZ045ewsqjY4iIiJt3Np9RZTV1DMmNtjoKHbH1cmB0bFBrE7Jo0qtckVsjgrtBtmwv5gOvq6EersaHaXNGtk1kHAfV95bn2l0FBEREWlmVquVxUk59AzzJsrPzeg4dm9s92AcTGhUu4iItLgVyXl4uzoyqKMGHp6MifEhVNVZWJOab3QUEfkTFdoNUG+xsimzhAEazd6iHM0mpvePYNuBUrZklRgdR0RERJrR7vwK9hRUcna8RsM1h0APZwZH+7F0Z64mWBMRkRZTXdfAD7sLGNklECcHlaRORu8Ib8J9XPXjuIgN0t9qBkjOKaOitoEBHVRob2nnJITi4+qoUe0iIiJtzIrkPBxMcFa3QKOjtBkT4g5NsLYpQwMURESkZfyyt5DKugbGxAYZHcVumU0mJsQFs25fMbllNUbHEZE/UKHdABt+v3npr0J7i3NzcuD8PuH8kFZAeqF6joqIiLQFVquV5bvyGBDli7+7s9Fx2owzuxyaYE0j5EREpKUs25WHv7sT/fWE/ymZEB+ClUPz1YiI7VCh3QAb9hfTOdCdAA/dGLaGC/uG4+xo5oMNGtUuIiLSFuzMKSerpFqj4ZqZq5MDo7oGsjo1n2pNsCYiIs2srLqen/cUMLZ7MI5mk9Fx7FoHPzd6hXuzKCkHq1q+idgMFdpbWV2DhcSsErWNaUX+7s5MjA9hSVIOBRW1RscRERGRU7R8Vx6OZhMjuqhtTHOb2COEitoGvt9dYHQUERFpY9bszqe2wcr47vqhvDlMjA9mb0ElybnlRkcRkd+p0N7Kth8oo6beokJ7K5s+IJK6BiufJGYZHUVEREROgcVqZWVKHkOi/fBxczI6TpvTN9KHUC8XluxU+xgREWle3+7MJdLXlfhQL6OjtAmjY4NwcjCxOEntY0RshQrtrWzD/mJMQL8OPkZHaVei/Nw4s0sAn205QJUehRYREbFb27JLySmrUduYFmI2mRgfF8xv6UXk60lAERFpJvnlNWzIKGZc92BMJrWNaQ7erk4M7xzAtztzqW+wGB1HRFChvdWtzyime4gn3q4agdXaLhsQSWl1PV9vO2h0FBERETlJK5LzcHYwMbxzgNFR2qwJ8SE0WGH5Lo2QExGR5rEiJR+LFcZ1DzY6SpsyIT6Eoqo6fkkvMjqKiKBCe6uqrmtgW3ap2sYYpHeED73CvflwYyb1Fk0WIiIiYm8aLFZWpuRzeqcAPF0cjY7TZsUEuBMX4skSPYouIiLN5NuducQGexIT4G50lDbltGg/fN2cWJKklm8itkCF9la0JauUeouVAVEqtBtlxoBIsktrWJ2SZ3QUEREROUGJmSUUVNSqbUwrmBAfQnJuObvzK4yOIiIidi6jqIodB8sYp0lQm52jg5lx3YP4Ia2A0uo6o+OItHsqtLei9RnFOJhN9IlQf3ajDO8SQJSfG+9vyMRq1ah2ERERe7IiOQ83JzPDOvkbHaXNG9s9CAeziaUaISciIqdo2a5cTMBYtY1pERN7hFDXYGVlsgYUihhNhfZWtGF/MQmhXrg7Oxgdpd0ym0xMHxDJzpxyNmaUGB1HREREjlN9g4VVKXmc0SkANyddS7U0f3dnhkb7sWxnLg1quSciIifJarXy7c5c+kb6EOLlYnScNqn77y15Fqvlm4jhVGhvJeU19ezMKVPbGBswMT4Ef3cn3tuQYXQUEREROU7rM4opqa5nrB47bzUT4kPILa9lQ0ax0VFERMROJeeWs6+oivFxGs3eUkwmE5PiQ9iaXUpGUZXRcUTaNRXaW0liZgkWKwxUod1wLo5mLuwbzi97i9R3VERExE4s35WHh7MDQ6PVNqa1nNHJH08XB7WPERGRk7ZsZx6OZhOjugYaHaVNGx8XjAlYrHO2iKFUaG8lGzKKcXE0kxDmbXQUAab1DsfV0cz7GzKNjiIiIiLHUFtv4bvd+YzoGoizoy5fW4urkwNndQtidWo+VXUNRscRERE702Cxsjw5l9Ni/PFxczI6TpsW7OXCwChflu7M1Xx0IgbSnUorWb+/mJ7h3rjo5tAm+Lo5cV7PUL7dmUtuWY3RcURERORv/JpeRHlNA2Ni1TamtU2MD6GqzsKa1Hyjo4iIiJ3ZnFVCXnkt49T2rVVMiA8hu6SaLVmlRkcRabdU9W0FxZV1pOZVMLCD2sbYkkv6R2CxWvloU5bRUURERORvrEjOxcfVkcFqwdfqekd4E+7twlJNsCYiIido2c5c3JzMDO8cYHSUdmFk10BcHc0s2an2MSJGUaG9FWzMPDSBlCZCtS0RPm6MiQ3i8y3ZFFfVGR1HREREjqK6roEf0goY2TUQRwddurY2s8nE+PgQ1u0vIq9cTwGKiMjxqa23sDo1nxFdAnF1cjA6Trvg7uzAyK6BrEjOo6beYnQckXZJdyutYP3+YtydHIgP8TQ6ivzJlYOjqKqzaFS7iIiIjfp5byFVdRbG6rFzw0yIC8ZiPTQyUURE5Hj8ml5EaXU94+KCjY7SrkyID6a8poGf9hQYHUWkXVKhvRVs2F9M30gfjcKyQZ0DPRjVNZCPN2VRVl1vdBwRERH5kxXJefi7O9EvUk8GGqWjvzsJYV4sVaFdRESO07e7cvF1c1Lbt1Y2MMqPQA9nlqjlm4ghVPltYbllNewrqlLbGBt21eAoKmob+GSzRrWLiIjYkoraen7aU8hZ3YJwMJuMjtOunR0XQmpeBSm55UZHERERG1deU88PaQWM7qa2b63NwWxifFwwP+8tpKiy1ug4Iu2O/sZrYRsyDvVn10Sotis2xJNhnfz5aGMWFbUa1S4iImIrfkwrpKbewthYtY0x2tjYIBzNJo2QExGRY1qdkk9NvYWJPUKMjtIuTYgPpsFiZUVyntFRRNodFdpb2MaMYrxdHeka7GF0FPkb1wyJoqS6ns83HzA6ioiIiPxu+a5cgj2d6RXhbXSUds/X3YnTY/xZtiuXeovV6DgiImLDFiflEOXnRo9QL6OjtEtdgzzpGuShH8dFDKBCewvbsL+YfpE+mE163NmW9QjzZkhHPz7YmEl1XYPRcURERNq90uo6fk0vYnRskK6jbMSE+GAKKmpZv7/I6CgiImKjMosq2ZRZwsT4EEw6fxtmQnwIOw6WkV5YaXQUkXZFhfYWlF1STXZpDQPUNsYuXD0kisLKOr7YqlHtIiIiRvtudwH1FqvaxtiQYZ0C8HJx1Ag5ERH5Swu3HLqfPjs+2OAk7dv47kGYTbA0KcfoKCLtigrtLehwf/b+mgjVLvSJ9KFfpA/vrc+kpt5idBwREZF2bUVyHuE+rsTrsXOb4exoZkxsEGtS8zWvjYiIHMFqtfLV5iz6d/AhzNvV6DjtWqCnC4M6+rF0Zy4Wq1q+ibQWFdpb0KaMYnzdnOgc4G50FDlOVw+JIr+ilm+2HzQ6ioiISLtVVFnL+n1FjIkN0mPnNmZCfDA19RbWpOYbHUVERGzM9gNlpBdUMiFek6DaggnxwRworSExs8ToKCLthgrtLcRqtbIho4T+HXx0g2hHBkb50jPMm3fWZVDXoFHtIiIiRliTmk+DFbWNsUG9wr2J9HVV+xgRETnC4qQcXJ3MjOoaaHQUAUZ0CcTNycxSnbNFWo0K7S0kq6SanLIa+qs/u10xmUxcPTSKg2U1LFEvMxEREUMsT86jo58bXYM8jI4if2IymTg7LpgN+4vJKasxOo6IiNiI2noLK5LzGBMXgqeLo9FxBHBzcmBUtyBWpuRRXddgdByRdkGF9hay8XB/9g4+BieRE3VatB9xIZ689VsG9Rb1MhMREWlN+eU1bMooYWx3tY35f/buOzzKMmsD+D01M+l90hshIZ1epCm9hSa2xQZr75+urq6oq6Ki7q66a0HFwmLBglJVqiA9ECCNkEJ6nbRJb1O+P6KsCKGlPFPu33VxXbthGO9Eyfu+J+c5x1zNitbABOCnTHbIERFRl315NWho02PBYD/RUeh3ZkV5o7nDgJ2neM0m6g8stPeRo8X1cLdXINSd89ktjUQiwdJRQSitb8NWPkASERH1qx3Z1TABmBrpLToKdSPAVY14P2dsOVkJExesERERgC0ntfB0D9bfIAAAIABJREFUUGJMmIfoKPQ7wwJd4e2oxPoTZaKjENkEFtr7gMlkwrFiHYYFurITy0JNCPfAQC8HfHy4CAZ2tRMREfWb7VlVGOjlgFAukzdrs6K9kV/Tgixtk+goREQkWF1LB/bn12JGlDfkMpaZzIlMKsGMKA325lajtqVDdBwiq8fvgH2gWNcGbVMHx8ZYMKlEgjtGB6GorhWb08pFxyEiIrIJFQ1tSC1rwFQuQTV7UyK8oJBJuBSViIiw9VQVDEYTZkdrREeh85gV7Q2D0YStp6pERyGyeiy094GjZ+azcxGqJbt6oCcGeNrj3d2n2dVORETUD7ZndT0AstBu/lzUCowL88DWU1rutCEisnGb0isQpXFEOJeYm6UBng6I8XPGjycrRUchsnostPeBY8U6eDooEeymFh2FeqCrqz0YedXN2JHFn/wSERH1te1ZVYjSOCLAlfdQlmBWlDdqWzpxuKBOdBQiIhIkq7IJ2VXNmBPjIzoKXcC8BD9kVjYhr6ZZdBQiq8ZCey8zmUw4WlyPYYEunM9uBSZFeGKgtyM+OsRZ7URERH2puK4VmZVNmDaIS1Atxdgwd7io5PiBHXJERDZrU0YFFDIJpg/iaTRzlhjvC5kEHPlG1MdYaO9lhbWtqGnu4NgYKyGVSHD/1QOQX9uCndnsaiciIuorO369zk6J8BSchC6VQibF1Egv7Dldg6Z2veg4RETUzzr0RvyUqcXEAZ5wUStEx6EL8HS0w+gQd/x4shJGE5sIifoKC+29LLmkaz77cBbarcaMGB+Eutvjo0NFvCARERH1kW2nqhDv5wwfZ5XoKHQZZsdo0K43Yld2tegoRETUz/bm1aC+TY/EWC5BtQSzor2hbepA8q97BYmo97HQ3suOFtXD21GJAFc+JFoLmVSCP48OQl5NCx8iiYiI+kB+TQtyq5sxjUtQLU6MjxOC3NT4IZPjY4iIbM2m9Ep4OyoxKthNdBS6BBMGeMDRTobNGbxmE/UVFtp7kclkwrESHYYFunI+u5WZEumFYDc1u9qJiIj6wNZTWkglwGQW2i2ORCLBzChvJBfXo7yhTXQcIiLqJ1VN7ThYUItZ0RrIpKx/WAKVQoZpkd7YmV3NkW9EfYSF9l6UX9uC2pZOjo2xQjKpBEtHByG3uhm7c2tExyEiIrIaJpMJ205pMSzQFZ4OStFx6ArMjO5aYPtTJhesERHZih9OamE0AYmxPqKj0GVIjO0a+bYjizvoiPoCC+296GhRPQBgaKCL4CTUF6YN8kaQmxqrDhayq52IiKiXnNI2oVjXxrExFszfRY0h/s7YklEJE++RiIisnslkwqb0Cgz2d0aQm1p0HLoMMT5OCHW3x8Z0jo8h6gsstPei5GIdfJzs4O/C+ezWSC6VYOmoIORUNeMXdrUTERH1iq2ZVZBLJZgU4Sk6CvXArGgNCutacbKiUXQUIiLqY2nljSisa0ViDLvZLY1EIkFirAZp5Q0oqGkRHYfI6rDQ3kuMJhOOldRjWKAL57NbselR3gh0VWHVoSJ2bBEREfWQ0WTC9iwtxoS4wVmlEB2HemBKpBfs5FJs4oI1IiKrtym9Aiq5FJMj+UNySzQzWgOZBLxmE/UBFtp7SV51C3StnRjG+exWTS6VYMmoIGRpm/DL6VrRcYiIiCzaidJ6aJs6MG2Qt+go1EOOdnJcM9AT205VoV1vFB2HiIj6SFunAduzqjA50gsOSrnoOHQFPB2UGBPqjh9OVkJvZAMhUW9iob2XHC3WAQAL7TZgZrQG/i4qrDpYyK52IiKiHth2qgp2cikmDPAQHYV6wZwYDRrb9diTWy06ChER9ZFdOdVo7jAgMUYjOgr1QGKsD6qbO3C4oE50FCKrwkJ7L0ku1sHP2Q5+nM9u9X6b1X5K24R9eexqJyIiuhJ6gxE7s6sxYYAH7JUy0XGoF4wIcoXGyQ6beRSdiMhqbUqvgL+LCkMDXERHoR4YH+YOV7UCmzIqREchsiostPcCo8mE4yX17Ga3IbOiveHnosKH7GonIiK6IklFOuhaOzF9kJfoKNRLpBIJZsdocLiwDtrGdtFxiIiol5XWt+JocT0SYzXcTWfhFDIpZkR545fTNdC1doqOQ2Q1WGjvBblVzahv02N4EAvttkIuk2LJyEBkVjbhAI9aERERXbZtWVVwtJNhTIi76CjUi+ZEa2A0AT+cZFc7EZG12ZJRCQmA2dEcG2MNEmM06DSYsDVTKzoKkdVgob0X/DafnUenbMvsGA18ne04q52IiOgyteuN2J1TjWvCPaGU83bUmgS6qTHY3xmbMyp5f0REZEWMJhM2Z1RiZLArfJw5MtcaRHg7ItLbEZs48o2o1/DJphckF9cjwFXFi42NUcikuH1UENLLG3GokF3tREREl2p/fi2aOwyYPshbdBTqA4kxPiisa0V6eaPoKERE1EuSi3Uob2hHYoyP6CjUixJjNMjSNiFb2yQ6CpFVYKG9hwxGzme3ZYkxGmic7PDhgSJ2bREREV2ibae0cLdXYBjH7lmlyZGeUMmlXLBGRGRFNqVXwtFOhonhHqKjUC+aHuUNhUzCrnaiXsJCew/lVDWhsV2PYYEcG2OLFDIpbh8ZiLTyBiQV6kTHISIiMnuNbXrsPV2DqZFekEu5SM0aOSjlmBThiW2nqtDWaRAdh4iIeqipXY9dOdWYPsgbKoVMdBzqRa5qBSYM8MBPmVp0Goyi4xBZPBbae+hocT0AYDg72m3W3FgfeDsq8SFntRMREV3UzuwqdBhMmMlFalZtTowGzR0G7MmtER2FiIh6aHtWFdr1RiTG8NptjRJjfaBr7cTevFrRUYgsHgvtPZRcrEOQmxpejnaio5AgSrkUt40MQkpZA44UsaudiIjoQn7M1CLYTY1ojaPoKNSHhgW6wtfZDpt5FJ2IyOJtSq9EmIc9on2cREehPjA62A1ejkpsSufIN6KeYqG9B/Rn5rNzbIytmxfnAy9HJVaxq52IiKhb5Q1tOFZSj5nR3pBIODbGmkklEsyO1uBwYR0qGtpExyEioitUUNOCtPIGJMb68NptpWRSCWZFa3AwvxZVTe2i4xBZNBbaeyBL24TmDgPHxhDs5FLcNiIQx0sbcKykXnQcIiIis/RTphYAMCPKW3AS6g+zYzQwoesUAxERWaYN6RW/FmJ57bZm82J9YDB1nV4goivHQnsPHCvuGhMyNIAd7QTMj/eFp0PXrHYiIiI6m8lkwo8ntRjs7wx/F7XoONQPAlzVGBrggs0ZlTzxR0RkgfQGI344WYnxYe5wt1eKjkN9KNBNjRFBrlifVg4jr9lEV4yF9h44UqRDiLsanpzPTujqar91ZCCSi+txrISz2omIiH7vZHkD8mtbuATVxsyJ0aCorhWpZQ2ioxAR0WXam1eL2pZOzI/zFR2F+sH8OB+UN7TjcGGd6ChEFouF9ivUaTDieEk9Rga5iY5CZmRBnA88HJT48GCR6ChERERmZf2JMihkEkyJ8BQdhfrR5AgvqBVSHkUnIrJAG9Iq4O2oxOgQ1j1swdXhnnBVK7A+lUtRia4UC+1XKK28AW16I0YEcT47/Y9KIcOtIwJwtEiHE5zVTkREBKBrgfzmtHKMDXWHs0ohOg71I3ulDJMjvLA9qwotHQbRcYiI6BJpG9txsKAWc2I0kEm5BNUWKOVSzInRYM/pGlQ3d4iOQ2SRWGi/QkmFOkglwDAuQqU/WBjvC3d7BWe1ExER/SqpsA7VTR2YxbExNml+nA9aOg3YkV0lOgoREV2izRmVMJqAxFgf0VGoH82P84HBaMLmdHa1E10JFtqvUFKhDtE+TnBSyUVHITOjUshw8/AAJBXpkFLKrnYiIqIfM7VwUSswNtRddBQSIN7PGcFuamxI40M7EZElMJpM2JBegeFBrghw5QJzWxLsbo+hAS5Yn1bBpahEV4CF9ivQ1K7HyYoGjOTYGOrGosF+cFMrsIqz2omIyMY1d+ixO6caM2N9oJTz1tMWSSQSzIvzQWpZA/JrWkTHISKii0gu1qGsvg3z2M1ukxbE+6K0vg1Hi3SioxBZHD7tXIFjJfUwmICRwVwIQuen/rWr/VBhHdLKGkTHISIiEmZnVjXa9EYsGOwnOgoJNPvXGb/saiciMn8b0irgZCfH1eEeoqOQANcM9ISLSo7vuRSV6LKx0H4FkgrrYCeXIs7XWXQUMmOLBvvBVa3AqkOc1U5ERLZrU0YFgt3UGMK9NjbN3V6JCQM88MPJSnQajKLjEBFRNxraOvFzTjVmRnlDpZCJjkMC2MmlmBWtwe7catS2cCkq0eVgof0KHCnSYYi/C48/0wXZK2VYPMwfB/LrkFnZKDoOERFRvyuqa8WJ0gbMidFAIpGIjkOCzYv1QV1rJ/aerhEdhYiIuvFTphYdBhPmxnFsjC2bH+8DvdGELRmVoqMQWRRWii9TdVM78mpaMDKYXVl0cYsG+8HRToZPDheLjkJERNTvNmdUQCrpGhtCNDrEDd6OSmxI51F0IiJzZDKZsD6tAoO8HRHp7Sg6DgkU5uGAwf7OWJ9WAROXohJdMhbaL1PSr8sgRgZxPjtdnKOdHNcP8cfPOdXIq2kWHYeIiKjfGH7tghoT4g4vRzvRccgMyKQSzIn1wcH8OlQ0tImOQ0REf3BK24ScqmbMYzc7AZgf54uiulYcK6kXHYXIYrDQfpmSinRwUckx0NtBdBSyEDcN8YdKLsXqJHa1ExGR7ThcWAdtUwcSY9nNTv8zN1YDE4DNPIpORGR2NqRVwE4uxfRB3qKjkBmYHOEJJzs5vk8tFx2FyGKw0H4ZTCYTjhTWYUSQK6ScM0qXyNVegYUJvtiaqUWJrlV0HCIion6xKb0SLio5xod5iI5CZsTfRY0RQa7YlF4BI4+iExGZjbZOA7ae0mLSQE84qeSi45AZUClkmBXtjV051dC1doqOQ2QRWGi/DIV1rdA2dWBEMMfG0OVZPCwAUqkEnx0tER2FiIioz9W3dmLP6WrMiPLm8ng6x/w4H5Q1tCOpsE50FCIi+tWunGo0tRs4NobOMj/OF50GLkUlulR88rkMSYW/zWfnIlS6PN5OdpgTo8HG9ApUNbWLjkNERNSntp6qQqfBhMQYPqzTua4O94SrWoHvUrkUlYjIXGxIq0CgqwpDA1xERyEzEu7lgDhfZ3yXWs6TaESXgIX2y3CkqA5+LioEuKpFRyELdOuIQBiNJnx+tFR0FCIioj61OaMCEV4OiNQ4io5CZkgplyIxRoNfcqvZgEBEZAZ+W3iZGOsDCcfk0h8sGty1FPVIkU50FCKzx0L7JTIYTTharMMIdrPTFQpwVWPaIG98l1rG+WZERGS1cqqakFnZhMRYdrNT9xbE+8JgAjams6udiEi0jekVkEqAOTFcYE7nmhzhBVe1At+eKBMdhcjssdB+iU5WNKKp3cCxMdQjt40MRGunEV8dY1c7ERFZp43plZBLJZgxyFt0FDJjgW5dS1HXp1bAYORRdCIiUfTGrvnbY0Pd4eVoJzoOmSE7uRRzY32w93QNKht5Eo3oQlhov0QHC2ohlQCjuAiVemCApwOuDvfAV8fL0NSuFx2HiIioV7V1GrAloxLXDPSEq71CdBwycwvjfVHR2I5DBVyKSkQkyoH8WlQ3d3AJKl3QwgQfGE3A+tRy0VGIzBoL7ZfoYEEdYnyc4aLmQyP1zJJRQWhs12NdCi9QRERkXXZmV6OxXY+F8b6io5AFmBjuAXd7Bb7jQzsRkTAb0irgbq/A2FB30VHIjPm7qHFVqDu+T6uA3mAUHYfIbLHQfgl0rZ3IKG/EmFB2s1PPRfs4YXSwG75ILkFbp0F0HCIiol7zfWo5gtzUGBboIjoKWQCFrOso+r68GlQ0tImOQ0Rkc6qa2rE/rwZzYjSQy1geogtbNNgXNc0d2J1bIzoKkdnid9JLkFRYBxOAq0JYaKfecfuoQNS2dHIBGBERWY3T1c1IKWvAgnhfSCQS0XHIQsyP94GJS1GJiITYmF4BgwmYH8eTaHRxY0Lc4edsh29TuBSVqDsstF+CAwV1cFHJMUjjJDoKWYmhAS5I8HPGf4+UoJPHroiIyAp8n1oOhUyCOdEa0VHIgvi7qDEqxA0b0iqg51JUIqJ+YzCasD61AiOCXBHophYdhyyATCrBwgQ/JBfXI6+mWXQcIrPEQvtFGE0mHMyvxahgN8ik7M6i3iGRSLBkVBAqG9vxY6ZWdBwiIqIeaes0YMvJSkziElS6AgvjfaFt6sD+vFrRUYiIbMahgjpUNLZzrwpdlrmxGihkEqw7wf0qROfDQvtF5FQ1o7alk/PZqdddFeqGCC8HrE4qhoEdXEREZMG2Z1Whqd2ABXxYpyswPswdXo5KHkUnIupH36eWw91egYnhHqKjkAVxs1diSoQXtpysREsHd84R/REL7RdxML+rs2Z0MAvt1Lt+62ovqmvFrpxq0XGIiIiu2Pep5QhxV2NoAJeg0uWTy6RYGO+LQwV1KKhtER2HiMjqaRvbsS+vBomxPlBwCSpdpkWD/dDcYcBPp3g6n+iP+B31Ig4W1CHCywGejnaio5AVumagJ4Ld1PjkcBFMJna1ExGR5cmpakJaeSOXoFKPLIj3hUImwbcn2NVORNTXNpxZguojOgpZoDhfJ0R4OeDbE2WsYxD9AQvtF9DUrkdKWQPGhLqLjkJWSiaV4LaRgcipasb+fM4lJSIiy/N9agWUMglmcwkq9YCHQ9dR9M0ZlWju0IuOQ0RktQxGEzakVWBUsCsCXLkElS6fRCLBosF+yKlqRmpZg+g4RGaFhfYLOFqkg8FowpgQjo2hvjMzyhs+Tnb4+FAxfxpMREQWpaXDgB9OVmJShBdc1FyCSj1zw5Cuo+hbMipFRyEisloH8mtRySWo1EMzorzhoJTh2xQuRSX6PRbaL+BgQR0clDLE+zmLjkJWTC6T4pYRgUgrb8CxknrRcYiIiC7Zj5mVaO4w4LrBfqKjkBWI8XVGjI8Tvj5eBiObD4iI+sR3qeXwcFBiwgAuQaUrp1bIMCdGg53ZVaht6RAdh8hssNDeDZPJhIMFtRgR5MrlINTn5sZq4G6vwMeHikRHISIiuiQmkwlfHy9DlMYRcb5OouOQlbh+iB8K61qRVFgnOgoRkdWpaGjDgfxazI3VQM46B/XQogQ/dBpM+D6VXe1Ev+F31m4U1raivKGdY2OoX6gUMtw8PABJRTqklLKrnYiIzF9ycT3yalpw3WA/LkGlXjMlwgvu9gp8dZxLUYmIetvG9AqYTMA8LkGlXhDiYY/RIW749kQ5Og1G0XGIzAIL7d04UNC1mHJ0CBehUv9YNNgPbmoFPjhQKDoKERHRRX19ogwuKjmmRnqJjkJWRCmXYkG8L/bn1aJE1yo6DhGR1dD/tgQ1xA3+LlyCSr3jpqH+qG7uwI7sKtFRiMwCC+3d2JtXi1B3e/i5qERHIRuhVshw68hAJBXpcJyz2omIyIxVNLRhT2415sX5QqWQiY5DVubaBF9IpRJ8c4Jd7UREvWV/Xi20TR1cgkq9anSIG4Ld1PgyuRQm7lchYqH9fJra9TheUo/xXA5C/WxRgi/c7RX44ECB6ChERETdWpfSNYtz0WA+rFPv83K0w6SBntiYXoHmDr3oOEREVuH71HJ4OigxPoyn9qn3SCUS3DjUH5mVTUgtaxAdh0g4FtrP40B+LQxGEyYM4AWI+pdKIcNtIwNxtLgeycU60XGIiIjO0a43Yn1aBSYM8ICvM0/+Ud/40zB/NLUbsCGtQnQUIiKLV/7bEtQ4Hy5BpV43O0YDJzs51h7jSTQifoc9j19O18BVrUCsr7PoKGSDFsb7wtNBifcPFPLoFRERmZ0dWVXQtXbiusF+oqOQFYv1dcYQf2d8mVwKPResERH1yPpff2g5n0tQqQ+oFTLMi/PBzzlVqGhoEx2HSCi56ADmRm804WBBHcYP8IBMKhEdp981Nzfh3Xf/jX37fkFHRwfGjBmLBx/8P7i5ndvdr9frceedtyI8PAJPP/33Mx9fs+ZTrF27BnZ2Ktx330OYMmX6md/76qvPkZ+fhyeffOaCOV566e/QarV46613z/m9FSteRElJMd5++wMAwKJFiaioKD/z+1KpFPb29oiJicc99zyAgQMjzvs6hUIBT09vTJx4DZYuvRP29g6X9kXqYyqFDEtGBeL1XadxtFiHEUFuoiMRERGd8fWJMoS4qzEiyFV0lB7jfY95u2VEIB5dn4Ed2dWYEeUtOg4RkUXqNBixPrUcV4W6W81JNF6/zc/1Q/zwRXIJvjlRjgcnhIqOQyQMC+1/kFJaj4Y2PSbY6NyyZ555EoWFBfjb356Dl5c3PvzwXTz44D34+OPPoFQqz3rtqlUrkZOTjfDwiDMfy8s7jf/+92O88cbb0Ol0+Pvf/4aRI8fA2dkZzc1NWLv2c6xc+XGv5168+DZcf/1NAACTyYSamhq8+eZrePTRB/DVV9+fuRj9/nVtbW04deok3n77TWRkpOLf/34fCoWi17NdiXlxvlidVIz39xdieKArJBLb+6EPERGZn/TyBpysaMTjk8Kt4tpkC/c9d965FPX1rWZ939OdsWHuCHW3x5ojxZg+yMsq/psjIupvu7KrUdvSieuHWM9JNFu4fltC3eL3fJ1VuDrcE+vTynHHmCCoFTLRkYiE4OiYP/jldA0UMglGhdheF3FOThaSkg7hySefwahRYxAWNgDPPPMiqqursHPntrNem5p6Alu2bMSAAeFnfTw/Pw9hYQMQGxuPceMmwN7eAaWlxQCAL75Yg0mTpkCj6f3jamq1Gh4envDw8ISnpxciIwfh/vsfQV1dLZKTj5z3df7+AZg8eRpWrPgX0tPTsGXLhl7PdaXs5FIsGRWElLIGHC6sEx2HiIgIAPBFcikclDLMjrH87mJbue/x9PQy+/ue7kglEtw8PADZVc1IKuLuGiKiK/H1iTIEuqow2kpqHLZy/baEusUf3TTUHw1tevxwslJ0FCJhWGj/g315tRgW6AoHpe01+xcXd11Y4uMHn/mYvb09AgMDcfx48pmPtbS0YPny5/DII3+Bq+vZnf9+fn4oKSlCbW0N8vPz0NTUCG9vDWpra7B583rccsvS/vlkAMhkXT9BVSiUF3xdZOQgxMcPxo4d2y74uv42N9YHvs52eGdvAYyc1U5ERIKVN7RhV3YVFsT7WsV9Eu97zOu+pzszorzh6aDEZ0dKREchIrI4pyobkVrWgEWD/SC1klNBvH6b7/U7wd8ZURpHfJFcCoORNQyyTSy0/05BbQuK6loxPsxDdBQhPD09AQBVVdozHzMYDNBqtdDp/tdV/e9//xNRUdGYPHnaOe8RFRWDa66ZgvnzZ2LJkj/hz3++Gx4envjkk1WYO3chXF37Z55raWkJVq58Gx4enoiLi7/o68PCwpGXd7ofkl06pVyKe8aG4JS2CTuyqkTHISIiG7f2WCkgkeAGKzl6zvse87rv6Y5SLsUNQ/xwqLAO2dom0XGIiCzKNyfKoJJLkRhjPUtQef023+u3RCLBLSMCUVTXij2na0THIRLC8tuRetHeX78RTBhgm/PZo6JiEBwcgtdffwXPPPMCnJwc8dFH70Onq0NnZycAYN++X3Dw4H6sWfNVt+/zl788hbvvfgByuRxqtRqlpSX45Zef8eWX67B58wasWfMJnJ2d8de/PoPw8IHdvs+JE8mYOnX8OR/v6OhAXFzCWR9bvfojfP75agBdy070ej0GDozESy+9DgcHx4t+7k5OTmhuNr+HtxlR3vjsaAne3VeAawZ6QiHjz8aIiKj/NbXrsSGtAlMiPOFjJYvUeN9jfvc93bk2wQ+fHC7GZ0dL8MKsQaLjEBFZBF1rJ7aeqsKcGA2cVNZT+uH127yv35MGeiLAVYX/JhXjmnAP7lchm2M93217wd7TNRjo5WA1D5CXS6FQ4OWX/4EXXngG8+fPgEKhwNSpMzBmzFjI5QrU1dXh1VeX46mnnoWzs8sF38vJyenM//7gg3dx0003o6WlBW+//QY+/XQtMjLS8OKLz2L16i+7fY/o6Bg8/fTz53z8/fffQV1d7VkfW7jwOixYcB2ArqNXLi4ul7WNu6Wl+azM5kIqkeCB8aF4+Lt0fJdSjhuG+ouORERENmh9WgWaOwxYPDxAdJRew/se87vv6Y6TSo758T746ngZ7hsXYrP36kREl2NjWgXa9UYsGmwdJ9F+w+u3eV+/ZdKu/SorduTiWEk9hgX2z+kAInPBQvuvdK2dSClrwO2jgkRHESo4OAQffbQG9fU6KBQK2Ns7YOnSxRgxYjQOHdqPurpaPPfcU2de39HRAQDYvXsntm/fe877ZWefQlpaCv72t+dw6NABBAYGw8fHB25urnjuuafQ3NzU7U9ulUoVAgICz/m4g4PDORcsJyfn8772UmVlnUJEhHl2SI0JccPwQBesOlSE2TEaONrxry0REfUfvcGItcdKMTTABVEa8364u1y877EcNw31x1fHy7DmSAkenxx+8T9ARGTDDEYTvk0pw7BAF4R7Xnoh11Lw+m3eZkdr8MGBQnyaVMxCO9kczqH41YH8WhhNwIQBtjmfHej66egDD9yFvLxcuLi4wt7eAeXlZcjNzcHIkaMxceI1WLv2e3zyyRdnfkVHx2DcuAn45JMvzvueK1e+jSVL7oSdnR0kEglMvy711Ov1AACjGSzIyMnJRnp6KqZOnSE6ynlJJBI8MCEMutZOfH6Ui8CIiKh/7cyuRmVju1V1swO87zHX+57u+DirkBijwfdp5dA2touOQ0Rk1vbl1aK8oR3XW1k3O8DrtyVcv1UKGW4c6o9DBXXI4n4VsjFsjf3V3tM18HBQIkpz8blY1sre3gEGgwFvvfUvPPLIX9Da2oJXXnkBw4aNwLBhI8685veUShXs7R3O+1PZ5OQjKC8vw8yZcwAAERGDkJ9/GseOHcUYXpvWAAAgAElEQVTJk+kICQnt92NPra2tqKmpBgC0tbUhLS0FK1e+jYSEIZg2bWa/ZrkcMT5OmBLhhc+TS3DtYD94Olx4IzkREVFvMJlM+Dy5BEFuaowLs64dNrZy31NdXYX6+laLuu/pzpJRQdiUUYnVScXsaiciuoBvTpTC21GJCeGeoqP0Olu5flti3eL3FiX4YXVSMdYcKcby2VGi4xD1GxbaAbTrjThYUIcpkV6Q2viihueffxlvvPEa7rlnCezsVJg4cRLuvffBK3qvlSv/gzvuuAdyedd/Zj4+Prjvvofw7LNPwsXFFcuWnTvHrK99/vnqM8tH7O0d4OPjgwULFuGGGxZDJpP1e57Lce+4EPycW41VBwvx5JTul7EQERH1luOl9cisbMKTU8Kt8h6J9z2Wxc/lf13tt44MhMbJTnQkIiKzk1fTjMOFOtw7NgRyqfVduwFevy2Bk0qOBfG++CK5BPeOC4G/i1p0JKJ+ITH9dibmPKqqGvszy2VxdbWHTtfSK+/1y+kaPLY+A28tjMVVodbVrfVHvfl1sxXm9DV7bWcuvkspw2e3DEO4l3nP2jOnr5ul4NfsytjC183Lq2ddNOZ8PRfBFv6b6S2Prc9ASmk9Nt81CipFzx7s+HUXw9q+7mX1bVj48REsjPfFE2bc1W5tX3dLYs5fe17Pe5c5/7sWafm2bPyUqcXmO0fB1V7R7/98/nsxTyL+vWgb2zFvVRIWmPk1WyT+fTFfF/t30901nTPaAezMroKzSo6RQVzSQObtrquC4Wgnx+u7cnGBn5ERERH1WGFtC/aersGiwX49LrIT9ZbfutrXp5WjkrPaiYjOUtvSgR9PVmJ2tEZIkZ3o97yd7DAr2hsb0ytQ09whOg5Rv7D5QnuH3og9uTWYOMADcpnNfznIzLmqFbh3XAiOldRje1aV6DhERGTFvjxWCrlMgkVWuEiNLNvS0UEwmoDVScWioxARmZV1J8rRYTDhpqH+oqMQAQBuGxkEvcHIazbZDJuvLB8urENzhwGTI71ERyG6JPPjfBHp7Yi39uShtdMgOg4REVkhXWsnNmdUYmaUNxdwk9nxdVZhbmxXV3tFQ5voOEREZqFdb8Q3J8owLswdIR72ouMQAQCC3NSYFa3BupQyaHkSjWyAzRfad2ZXwcmOY2PIcsikEjw+aQC0TR349HCR6DhERGSF1qWUoV1vxJ+GBYiOQnReS0Z1dbV/yg45IiIAwI8nK1HX2onFvHaTmfnzmCAYTMAnrF+QDbDpQnuH3og9p2swMdwDCo6NIQuS4O+CmVHeWHO0BCW6VtFxiIjIirTrjfj6eBlGh7hhgKd5L94m2+XrrML8OB+sT6tAUR3vhYjItplMJnyRXIoILwcMC3QRHYfoLP4uasyL7bpm8yQaWTubri4nFdWhqd2AKREcG0OW56EJoVBIpfjXz6dFRyEiIiuyKb0CtS2duHUEO+LIvN0xJhhKmQTv7M0XHYWISKiDBXXIr23B4uEBkEgkouMQnWPJqEBIJMBHh9jVTtbNpgvtO7Kru8bGBHNsDFkeT0c73DEmCHvzavHL6RrRcYiIyAroDUasOVKMOF8nDA/k/RGZN08HJW4ZEYhdOdVIKa0XHYeISJjPj5bAy1GJqdw9R2bKx1mFhfG+2JRRyVP5ZNVsttDeaTDil9waTODYGLJgNw71R7inA1bsyEFjm150HCIisnBbT1WhrKEdt48KYkccWYSbhwfA00GJt/bkw2QyiY5DRNTvcqqakFSkw/WD/VjbILN2+8hAyKUSrGJXO1kxm/0unFSoQ2O7HpMHeoqOQnTFFDIpnpkegZrmDrz1S57oOEREZMGMJhM+TSrCQC8HjA9zFx2H6JKoFTLcfVUw0sob8HNOteg4RET9bs2REqjkUiyI9xUdheiCPB3tcG2CL348WYmC2hbRcYj6hM0W2ndkV8FBKcOoYDfRUYh6JNrHCTcPD8SGtAocLqwTHYeIiCzU7pxqFNS24vaRgexmJ4uSGOuDAZ72+M/efHQajKLjEBH1mxJdK7ad0mJhgi9c1ArRcYgu6raRgbCTS7Fyf4HoKER9wiYL7Z0GI/bk1mBiuAeUcpv8EpCVuXNMEILd1Hh5WzZaOgyi4xARkYUxmUz45HAxAl1VmMwl8WRhZFIJHpwQhhJdG9allIuOQ0TUb9YcKYFUKsHNw7nAnCyDu33XfpWd2dyvQtbJJqvMSUW/jo3hgyRZCZVChmemR6C8oR3v7ssXHYeIiCzMwYI6nNI24baRgZBJ2c1OlueqEDeMCHLFqoOF3FtDRDZB29iOTRkVSIzxgZejneg4RJfs5uEB8HJU4s09edyvQlbHJgvtWzO1cLKTYzTHxpAVSfB3wfVD/PDV8TKcKOFPhomI6NJ9ergIGic7zIrWiI5CdEUkEgkenhiGxnY9mw6IyCZ8nlwCo9GEW0awm50si1ohw71jQ5Be3ojtWVWi4xD1KpsrtLd0GPBzTjWmRHpybAxZnfvHh8LPRYXnt2ahuYPdXEREdHHHS+pxvLQBtwwPgELGeyOyXJHejrhhiD/WpZQjraxBdBwioj6ja+nEdynlmDbIGwGuatFxiC7brGgNIrwc8O9f8tHayfG3ZD1s7mlqz+lqtOmNmBnFji2yPmqFDH+fEYmy+jb86+fTouMQEZEF+OBgIdztFZgX5yM6ClGP3T02GF6OSryyIwd6LkYlIiv15bEStOmNuH1UoOgoRFdEJpXg8UnhqGxsx6eHi0THIeo1Nldo//GkFr7OdkjwdxYdhahPDAlwwW0jA7ExvRK7cqpFxyEiIjOWXKzD0SIdbhsZCJVCJjoOUY85KOV4fFI4cqqa8eWxUtFxiIh6XVO7Hl+fKMM1Az0R5uEgOg7RFRsc4IKZUd5Yc7QERXWtouMQ9QqbKrTXNHfgcGEdZkR5Qyrhoi+yXneNCUaUxhEvbctGWX2b6DhERGSGTCYT3j9QCC9HJRbG+4qOQ9Rrrh7oiYkDPPD+gULeBxGR1fnmRBma2g1Ywm52sgIPTQiFUibFP3/O5WJUsgo2VWjfllUFowmYEeUtOgpRn5LLpHh5ThSMJhOe3HQS7XoenSYiorMdKdLheEk9bh8ZxG52sjp/mTQAUgnw2k4+uBOR9WjpMOCL5FKMDnFDlMZJdByiHvN0tMPdY0NwIL8OW09xMSpZPpsqtG9Or0CUxpHHq8gmBLiq8fzMQcisbMLru3JFxyEiIjNiMpmwcn8hvB2VmM/Z7GSFfJxVuGdsCPbn12JHNkfpEZF1WHusFLrWTtx9VbDoKES95vrBfojzdcI/duWitqVDdByiHrGZQntWZROyq5qRGMuHSbIdEwZ4YMmoQGxIq8CGtHLRcYiIyEwcLKhDWnkD/jw6CEq5zdwOko25fog/ojSOeG1nLqqb+eBORJatoa0Ta44WY3yYO2J9uXOOrIdMKsGy6RFo6TTg9Z2nRcch6hGbebLamF4BpUyC6YO8REch6ld3XxWCkUGueG1nLjIrG0XHISIiwX6bze7rbMcGBLJqcqkEf58ZidZOA5ZvzeYIGSKyaJ8dLUFTuwH3jA0RHYWo14V5OOCO0cHYkV2F7VkcIUOWyyYK7e16I346pcU1Az3hrFKIjkPUr2RSCZbPHgQ3eyWe3HgSutZO0ZGIiEigfXm1OFnRiD+PDoJCZhO3gmTDwjwc8OD4UOzPr8V3qTzdR0SWqbalA2uPlWJqpBcivB1FxyHqE7eOCECsrxNW7MhBZWO76DhEV8Qmnq725FajoU3Pri2yWW72SryaGIWq5g48sfEkOrgclYjIJhlNJqzcX4AAVxVmR2tExyHqF9cN8cPoYDe8sTsPBbUtouMQEV22Tw8Xo11vxF2czU5WTC6T4vmZg9BpMOL5n7Jg5Ek0skA2UWjfmF4BHyc7jAhyFR2FSJgYX2c8Oz0Sx0vqsXwbj08TEdminzK1yK5qxt1XhUDObnayEVKJBM/OiIBKLsWzP5yC3sCGAyKyHJWN7ViXUoY5MRqEuNuLjkPUp4Lc1Hj06gE4UqTDf5OKRcchumxW/4RVXNeKw4U6zI3zgVQiER2HSKgZUd64Z2wwfszU4sODhaLjEBFRP2rrNODdfQWI0jhiGnfWkI3xcrTDU1MHIrOyCasOFYmOQ0R0yT46VAijCbhjDLvZyTbMi/PBlAgvvLe/AMnFOtFxiC6L1Rfav08th0wCzI/j2BgiAFg6KghzYjT48GARtmRUio5DRET9ZO2xUlQ2tuPhiWFsPiCbNDnCC7NjNPjkcBGOFNWJjkNEdFElulZsTK/Ewnhf+DqrRMch6hcSiQTLpg9EoKsaf9ucieomzmsny2HVhfZ2vREb0yswMdwTXo52ouMQmQWJRIK/TR2I4YEuWL4tmz8hJiKyAXUtHfg0qRgTBnhgWCBH6ZHtenzSAAS72ePpzadQ0dAmOg4R0QX955d8KKQSLBkVKDoKUb9yUMqxYm40mjsM+OumTO6ZI4th1YX2XTlVqG/TY2GCr+goRGZFIZPi1bnRCHRV47H1GcisbBQdiYiI+tCHB4vQ1mnAg+NDRUchEspBKcdrc6PRYTDiST64E5EZSy7WYVdONW4fFQhPNg6SDQr3dMBzMyKRWtaAl3fkcM8cWQSrLrSvO1GOQFcVl6ASnYezSoF/XxsLZ5UcD36bhtzqZtGRiIioDxTUtuC71HLMj/dFiAeXqBGFeNjj2RmRyKhoxL92nxYdh4joHAajCW/szoPGyQ6LhwWIjkMkzNRIL9w5JghbMiqx5kiJ6DhEF2W1hfasyiaklDVgYYIf55ASdcPHWYV3r4uHUi7F/d+koqiuVXQkIiLqZe/szYdKLsVdV3GJGtFvJg30xK0jArAupRyb0itExyEiOsuWjEpkaZvw0IRQqBQy0XGIhLpjTDCmRnrhP3vz8VOmVnQcoguy2kL7F8dKoFZIMS+WS1CJLiTAVY13FsXDZALu+yYV5ZxXSkRkNZKLddidW4NbRwTC3V4pOg6RWbl3XCiGB7lixY4cjtEjIrPR3KHHO/vyEefrjKmRXqLjEAknlUjw3IxIDA1wwfM/ZSGpkAvNyXxZZaG9qqkd205VYW6sD5xUctFxiMxeqIc9/rMoDi0dBtz3TSqXgxERWYFOgxGv7siFn7Md/jTMX3QcIrMjl0rw8uxBcLdX4rH1Gbz/ISKz8OnhYtS2dOKxa8Ig4el8IgCAnVyKf8yLQbC7Go9vOImU0nrRkYjOyyoL7d+cKIPBaMKNQ/lQSXSpIr0d8e9rY1HX0ok716agsLZFdCQiIuqBL5JLkV/bgr9MCuexc6JuuNkr8caCWLR0GPB/32egqV0vOhIR2bDS+lZ8kVyCWdHeiPF1Fh2HyKw4qeT4z7Vx8HRU4uHv0pFa1iA6EtE5rK7Q3tZpwHcp5ZgY7oEAV7XoOEQWJdbXGe9fn4B2vRF3fZWCLG2T6EhERHQFyhvasOpgIa4O98D4AR6i4xCZtXAvB7yaGI382hY8tTkTeoNRdCQislH/3pMPqUSC+8aFio5CZJa8HO3w3nXx8HBQ4qF1aSy2k9mxukL7hrQK1Lfp8Sdu5ia6IpEaR3xwYwIUMinu+TqFR7KIiCzQv34+DQB47JoBgpMQWYZRIW54ako4DhXU4dWduTCZTKIjEZGN2ZdXg1051VgyKggaJzvRcYjMlrdTV7Hd3V6Bh9alIY3FdjIjVlVo7zQY8d8jxUjwc8Zgfx6zIrpSIe72WHVjAtztlbj/2zTsz6sVHYmIiC7R3tM12J1bgzvHBMPHWSU6DpHFmBfni9tHBmJ9WgVWJxWLjkNENqSlw4BXd+QizMMet4xg0yDRxXg72eG96xPgZq/Ag+vS2CBIZsOqCu0/nKyEtqkDS0cHcWkIUQ/5OKvw4Y0JCHW3x6Pr0/H50RJ2dxERmbm2TgP+sSsXoR72uIkLUIku273jQjAt0gvv7CvAxrQK0XGIyEa8f6AAFY3t+NvUgVDIrKpMQ9RnNE52WHl9AjwcuhoE9+RWi45EZD2Fdr3RhNVJxYjSOGJMiJvoOERWwd1eiQ9uTMDV4Z54c08eXtiajQ4955YSEZmrjw4VoayhHX+dHM4HdaIrIJVI8NyMSIwOdsNL27OxM7tKdCQisnKZlY1Ye6wU1yb4IsHfRXQcIouicbLDqhsTEO7pgCc2nsR3KWWiI5GNs5onsB1ZVSjWteH2UexmJ+pNaoUMryRG4a4xwdicUYl7vk5FdXOH6FhERPQHmZWNWHOkGHNiNBgW6Co6DpHFUsqleG1eNGJ9nbFsyykcLOAIPSLqG3qjCS9ty+ka2ckFqERXxM1eifeuj8dVoe54ZUcu3ttfwNP4JIxVFNr1BiPeP1CAcE8HXB3uIToOkdWRSiS486pgrEiMQk5VE2777BiOFulExyIiol91Gox4/qcsuDso8ejVXIBK1FNqhQxvLohFqIc9Ht9wkrNfiahPrD1WiixtEx6fNABOKrnoOEQWS62Q4fV5MZgbq8HHh4rwwtZsdBp4Gp/6n1UU2jekV6BE14b7x4dAym52oj4zOcILq24aDJVChvu+ScW/9+RxlAwRkRlYdagIp6tb8PTUCD6oE/USJ5Uc/7k2DhonOzzyfToyKxtFRyIiK1Kia8X7+wswPswd1wz0FB2HyOLJpRIsmxaBO8cEYXNGJe7/Ng26lk7RscjGWHyhva3TgFUHizDY3xljQ91FxyGyepHejvjslqFYmOCLNUdLcPsXx3G6ull0LCIim5VSWo9PDxdhTowGY8N4L0TUmzwclHhnURwclXLc/00aMsobREciIitgMJrw9x+zIJNK8MTkcI6/JeolEokEd10VguWzBiGjvAG3sV5B/cziC+1rj5WiurkD948L5cWJqJ+oFTI8OWUg/jk/BtVNHbjt8+NYnVTMo1lERP2sqV2PZ384BV9nFR67hiNjiPqCj7MKH9yYACeVHPd/m8YxMkTUY58dLUFKWQOemBwOH2eV6DhEVmd6lDc+uCEB7Xoj/vzlCezLqxEdiWyERRfaq5ra8cnhYowPc8fgAG7nJupvEwZ44MvbhmFMiBve3puPm1Yn43BBnehYREQ2wWQyYcWOHFQ2tuPFWYPgaMeRMUR9xddZhQ9uSICHgxIPrUvHsRLuqiGiK5OlbcLK/QWYNNATM6O8Rcchsloxvs5YvXgIAlzVePT7DHx+tIRLUqnPWXSh/T+/5KPTaMSj7OAiEsbDQYnX58XgzYWxMJpMeGBdGv668STKdK2ioxERWbUNaRXYeqoKd14VjDg/Z9FxiKyexskOK6+Ph7eTEg+vS8eRIjYXENHlaekw4G+bM+GqVuCpKQN5Kp+oj2mc7PDhjQm4ZqAn3tyTh+XbuCSV+pbFFtqPFtbhx0wtbhkegABXteg4RDZvbKg7vrxtOO4dG4L9+bWY+tZe/Ovn06hp7hAdjYjI6mRWNuL1XbkYHeyG20cGiY5DZDO8HO2w8voE+Lmo8Mh36diVUy06EhFZiN9OopXoWrF89iC42itERyKyCWqFDK8kRmHp6CBsTK/EvV+nopp1CuojFllo7zQY8fymk9A42eH2UXy4JDIXdnIplo4OwjdLhiMx3hdfHy/F/FVJ+M8vedz2TUTUS+pbO/HkxpNws1fixVmDIJOyG46oP3k4KPH+DQmI9HbCkxtPYl1KmehIRGQBNmVU4sdMLe4YHYxhga6i4xDZFKlEgnvHhuDlOVHI0jbh1s+OIZ0LzqkPWGShfdWhIpyqbMTjk8KhVshExyGiP/B1VmHFgjh8vWQErh7oiTVHSjBvVRLe3J2HioY20fGIiCyW3mDEk5szUdXcgRWJUeyGIxLEVa3Au9fFYWyYO1bsyMUHBwo495WIupVZ2YhXd+RgeJArlo5msyCRKFMjvfDxnwZDIZPirq9SsCGtXHQksjIWV2jPqGjE6sNFWDjEHxPDPUTHIaILCHJT48VZg7D29mEYP8Ada4+VYP5HR7BsSyZOVTaKjkdEZFFMJhNe33UaR4t0eHpqBGJ9OZedSCSVQobX58VgbqwGHx4swis7cqA3sthORGerbenA4xtOwt1eiZdn8yQakWgDvRzx38VDMDTABcu35WDFjhzObadeIxcd4HK0dhrw9x9PwdPRDstmDYKhjaMoiCxBmIcDls+Owv3jQ7H2WCnWp3Yt8Bse6IIbhwZgXJg7bziJiC7iy2Ol+C61HLeOCMDsGI3oOEQEQC6VYNm0CHg4KPHJ4WKUN7Tj5dlR4FAIIgK6xt4+tSkTutZOrLoxAW72StGRiAiAi1qBNxfG4d29+VhztAS5Vc1YMTcang78O0o9YzEd7SaTCS9vz0FhbSuenR4BJxWPShNZGl9nFf7v6gHYfNcoPDQhFMW6NvxlQwYWfXIEa4+VorlDLzoiEZFZ+ilTizd25+HqcA/cNy5UdBwi+h2JRIL7xoVi2bSBOFKkw9Ivj6Owpll0LCISzGQy4ZXtOThWUo+npw3EII2T6EhE9DtyqQQPTQzDS7MHcW479RqLKbR/c6IcP2VqcffYYIwMduv2da+99hJWrHjxgu+l1VZi2bInMHXqBMyZMxX/+McKtLVxbjTRb2pra7B8+XOYN286Zsy4Go8++gDy8nK7ff2hQwfw5z/fgqlTx+PWW2/A9u0/XfD9nVRy3DIiEOvvGIlX5kTB3V6Jf/58GrPfP4w3dp9GaX1rb39KREQW62BBLf7+Yyb8i3Yg66NHMWP6BCxb9gRqa2u6/TObN2/A4sWLMGnSVbj55uuwZcvGs34/Pz8Pjz76IKZNm4g5c6bg9ddfRlNTU19/KkRCGQwGrFz5NubNm46pU8df9O/R7z3xxCN44IG7uv39n3/egdfvTcQLE71Q19KJRe8fwtEiXW9FJyILtDqpGJsyKnHnmCDMjOo6idba2oo33ngN8+bNwIwZV+Oxxx5Cfn5et++xbNlfMW7c8LN+Pfzwff31KRBZlO7qgdu3/4Qbbpjf7Z+bNsgbH93Eue3UOyyi0J5crMMbu09jXJg7low6/+IQk8mEVatWYuPG7y/4Xh0dHfi//7sfDQ0NeO+9j/DCC6/gwIG9ePfdt/oiOpHFMRqN+NvfHkdxcRFeeeWfeO+9j+Hg4IiHH74P9fXnPjCmpp7A448/jPj4wVi1ag1uvnkJXn/9FWzcuPE87342uVSCKZFe+Oimwfj0T4MxLswdXx0vw8KPjuDxDRk4VqLjYjEismlHiuq65rrm74SxIAnPPPM83n77Q2i1Wjz99BPn/TO7d+/EP/+5AosX34bPPvsGN9ywGK+99hL27dsDAGhpacEjj9wHZ2dnfPjhaqxY8S+kpJzAyy8/35+fGlG/+/jjD/Djj5uxbNnF/x793vr163DgwL5uf7+6uhr/+McrAIA4X2d8ungIPB3t8MC6NHxzooz3MkQ2aGN6Bd7ZV4Dpg7xw55jgMx9/661/4OjRJLz44gqsXPkJlEolHnvsQbS3t5/3ffLzT+Oeex7Ahg0/nfn14osr+uvTILIIF6oH7t+/96LNuAAQ4e2I1b+b2758WzbaOg19EZesnNkX2jMrG/HY+gwEuqnx/MxISCXnznEuLS3BQw/dg/Xrv4VG43PB99u+/SfU1FRj+fLXEB4+EEOHDsfSpXchMzOjrz4FIouSm5uN9PRUPPXUs4iOjkVoaBieeeYFtLa2nPch88sv1yA2Nh4PP/wYgoNDMG3aDCxefCveeec/l/XPjfF1xvLZUdh4x0jcNjIQx0vqcfdXqbj9ixPYnVMNIx9SicjGJBfr8H/fZ8DPUY7OrN245+77MWLEaERGDsLzz7+MtLQUpKWlnPPndLo6LF16F2bNSoSfnz8SE+cjLGwAjh49AgCoqChHfPxg/PWvyxAcHILY2HjMnbsAyclJ/f0pEvWbzs5OfPPNWtx9iX+PflNSUowPPngXsbHx3b7mlVdeQFhY+Jn/H+Cqxtd3jcKYEDe8tjMXT285xfF4RDZkZ3YVXtqWjdHBbnh2eiQkv6th7N27GwsWLEJ8/GCEhITirrvug1ZbiYKC/HPep7OzEyUlxYiOjoWHh+eZX87OXIZO9Jvu6oHt7W149dWX8PTTjyMg4PwNu3/kqlbgrYVxWDIqEBvSKrD0yxMoquNpe7o8Zl1oP13djIfWpcNZJcfb18bBuZu57BkZafDz88fq1Wvh6+t3wfdMSjqI4cNHnXVxmjNnHj788L+9mp3IUmk0PnjttTcRFPS/zgupVAqTyYTGxsZzXl9cXIyEhCFnfSwiIhKlpaWoqKi47H++t5Md7hsXis13jcJTU8JR39qJxzeexE2rk/FjZiX0Rhbcicj6HcivxSPfpcPPRYVHh6jQ2tKCIUOGnfl9X18/+Pr6ISXl+Dl/dv78RbjlliUAAL1ej127dqCwsAAjRowCAISFDcCLL66AWq0GABQVFWLr1h8wYsTofvjMiMTIyclCS0vzJf89ArpGzSxf/hwWL74VISHn343w3XffoKamGrfffsdZH3dSKfDP+TG4b1wIdmZX4dbPjiNby/FMRNZu7+kaLNtyCnG+znhtXjSU8rNLLq6ubti5czvq6mrR2dmJzZs3wMnJGX5+/ue8V0FBPgwGA4KDQ/opPZHl6a4eWFdXh7KyEqxc+QkmTLj6kt9PJu3au/LmglhoG9tx62fHsDO7qg+Sk7Uy20L7kaI63LH2BGRSCd5eFA9vJ7tuXztt2kw89dSz8PDwvOj7FhcXwcfHFx9++B6uu24urrtuHt5++81uj2oR2RoXF1dcddU4SKX/+/bwzTdr0dHRgZEjzy3CeHp6QqutPOtj5eVdM810utorzqFSyLAwwQ/fLh2BF2ZFwgTg2R+ycP0nR7A1U8sOdyKyWj9lavHo+gwEu9vjvevi0d7Y9b3Uy8v7rNed7/vv7506dRKTJ4/Fs88+ienTZ+Gqq8ad85rbb/8T/vSna1Ffr8NDDz3au58IkXiiGpcAACAASURBVBmpqtICuLy/R2vWfAKJBLjpplvO+/tFRYX44IN3sWzZ85DLz20IkkokWDIqCO9dH4+WDgOWfnkC36eWc5QMkZXak1uNJzaexEAvB7yxIBZqheyc1zzxxNPQaiuRmPj/7d15XFTl/gfwz+zsO8ii4gKIsuOVTFxLM61bVreytM2uN1u0a5lZ5s2y0rKyTSttT83sZ7m2uVSamRbiLggIIovIvgwMsz2/PwZGERBGlhmYz/v1mtfMnDkMz3zP98w55zvPec51GDt2OLZs2YilS9+Gq2vjC6VmZmZAoVDg448/xK233oC77roVK1euYO2C6CLN1QP9/QPw9tvvIzx84BW9b2I/L6y+Jx59vZ0wb8tJvL4rHTqDsT2aTN2czRXajULgm0N5mLnhGHxdVPjkrlj09nRst/dXq9XYunUTcnNzsGjREsyaNRu7dm3H0qWvtNv/IOpOfv/9N3z44Xu4884pTfbmGj9+Inbu/Bk7d26HXq9HWloqvvpqNQDT6Y5tJZdKMGFgD6y7bzBeu2kQHBQyPPd9CqZ+eRB7T5fwYJWIug0hBD7bn40F36cgJtANH9wRDW9nJTQaDaRSKeRyeYP5FQoltFpts+8XEBCIjz76As888z/s3LkdK1euaDTPM8/8D8uXr4KPjy9mzZrBi8NTt2XpepSamoJ169Zg/vwXGnQ+qKfX67Fo0f8wZcq9CAkJvez/ju/pgTX3xiMm0A2vbE/DU5tOoKS6+XWXiLqe7amFeHrLSYT3cMHyf0XD1UHe5Hw5OWfh5eWNpUvfwooVHyMhYSgWLHi6yR/8MjNPQwiB3r2DsXTp23jggf9g69ZNrF0QdRJ/NwesvDMGd8UH4evkPExfdxj5FdxXpsuzqUL7mZJqPLz+CF7bmY4hvT3wyV2xCHR3aNf/IZfL4ebmhgULXkR4+CCMGDEaM2fOxo8/bmvyQo9E9uz777dg/vy5uPbacXjkkVlNzjNhwo144IHpeOWVhbjmmmF49tmnMHnyFACAs7NLu7VFKpFgTKgPVt8Tj0UTw1GtNeC/3x3Df74+jCN5Fe32f4iIrEGrN+KFH1Ox/PcsXDfAF+/cFgUXlekgXaVSwWg0Qq9vOMazTqeFg0PznRHc3T0QGjoAN9xwE+699wGsX78WBkPDizoNGBCOmJg4vPTSa8jLy8Xu3b+2+2cjsgWWrEe1tbVYtGgBpk9/GD179mry/b744hNIpVLcffe9rfr/Xk5KvPuvKPx3VD/syyrBnZ8l4Ze0oiv7MERkU9YdzMX8rScR6e+Kd2+LarbInpeXi9deexmPPz4HV189HBERkVi48GUolUp8/fXaRvNPn/4wNm/+CZMnT0X//iG47rrr8fjjT7J2QdSJFDIpnhjTH6/+cyCySqox5YuD+DnlvLWbRTas6S1AJ6rWGpCcW47vDudjz+liOClleO66UNwU6d/goiHtxcfHDyqVEjLZhdO4+vTpB8A03IW7u0e7/0+irujzzz/GqlXv47bb7sB///vUZdfH++//N6ZOvR9lZaXw9vbB3r27IZPJ4O8f0O7tkkokuH6gH8aG+WDTsXNYtS8bD351CGPDfPHoiD7o6dF+Z8AQEXWG3PIaPLPlJE4WVOE/Vwfj31f3bvCd6+fXAwBQXFzU4CJPRUVFGD7ct9H7JScnwcXFBaGhA8zT+vcPQW1tLSoqKqDR1CA9/RRGjBhtft3Hxwdubu4oKuKBA3VPlqxHJ04cQ1ZWJj744F188IHp4u5arRZCCIwbNwJffvkNfvhhK4qKCnH99aMBAEaj6XTye+65A/feOw2zZj3WqA1SiQRT/tETQ/t4YuEPqZi7+QRuGOSHJ8eENFuYIyLbZTAKvLs7E2uScjA6xBuLJobDoYnhYuqlpJyEwWBoMJSFXC5HaOgA5OaebTS/VCqFm5t7g2n1F14+f76AtQuiTnRNmC9CfV3wvx9SMH9bCvacLsHca7j9psY6PSMO5pThvd2ZMAigRmvAmdJqGIXp6r73JfTCHXFB8HFWdtj/j4mJxZYtG6HX682njp4+nQGZTIaAgPYvChJ1RWvWfI5Vq97Hv/89o9HFvS61YcPXyMnJweOPPwkfH9OB6u7dvyI2NhZOTk4d1ka5TIrbYgIxYWAPrP77LL78Kwe/ZRThjtggTBvaq9mLJxMR2ZIdqYV4efspAMBrNw3CmNDG15sJCQmDk5MzDh06iPHjJwIA8vPzkJ+fh5iY+Ebzr1nzOaRSKV577S3ztBMnjsPT0wseHh7YtetvvPjic/juu+/h5eUNwNTLrqys1Nz5gKi7sWQ9GjQoAuvWfddg2ocfLse5c/l4/vmX4OPjg3ff/bBB7/jU1JN4/vlnsXTp2+jfP+Sybenv44xP747Fx39m49P92diXVYpZI/th4iC/DuloRETtr1Kjx/xtJ7EvqxS3xwbiyTH9IZNefv318zNdIyIjIx0DBoQDMA0bl5WViaFDhzWaf8GCedDr9Vi8+HXztJSUE1AqlQgKavpsGyLqOL08HbFqciw+/TMbH/95Bsk55XhhwgAM7sUfveiCTi+0q+QyuDsqYBQCvs5KjB3gg0H+rhjS2xMqefuPZKPT6VBRUQ43N3coFApMmnQbNmz4Gi+99DweeGA6zp8vwIoVb2P8+In8RZgIQHp6GlauXIEbbrgJ//znJBQXXzit2cnJGXK5vME61bt3H7zzzpsIDx+I6OhY7NjxE37++QesWvVxp7TXSSnDf4b1wS3RAfhgbxbWJuVg6/FzePDqYPwrJgAKmU2NkEVEBAAoqdZi6c507DhVhEH+rnj5hvBmz8hRKpW45ZZ/Yfnyt+Hu7gFPTy+88cYSxMbGIzIyqtG+zp133o0nnpiJtWu/wMiRY3DoUBLWrv0CM2fOhkQiQWLiCAQEBOGFFxZg1qwnUF2txrJlSxEZGd3kgT5Rd2DJeqRSOTQaMsbZ2Rkqlco8/dKz9oqLi83TL+2B2hS5TIqHEvtgdIgPluxMw8IfU7HpaD7mjg1FiI9zO31qIuoIqQVVeGbrCeRV1OKZcaG4Nbp1HfYGDoxAZGQ0XnllIZ54Yh48PDywfv1aFBScw7/+dWej7fno0ddi4cJnsW7daowYMRqnTqVg+fK3MXny1A7t0EREzZNLJZg+LBhX9/XE8z+k4uH1R3B7bCAeGdEHzkr2bicrFNoj/F2x7JbITvt/R48exqxZM/DOOx8gPv4f8PLyxnvvrcK7776JadOmwNHRCePHT8BDDzU+vZPIHu3c+TMMBgO2bduMbds2N3jt3/+egejo2Abr1JAhV2HOnGfwyScrUVRUiH79+uPVV5dhyJAhKCur7rR2+7qosGD8ANwZF4S3fzuNN3/JwDfJuXhsZD+MCfFmDzEishnbUwvx2s50qLV6PDK8D+4Z0gvyFnrBTZ/+cN3FFxdAr9fjqquG4YknngbQeF9nyJCheOmlV/HJJ6vw0Ucfws+vB2bPfgo33jgJAODg4IBly97DO++8iUcfnQ6JRIKRI0dj5swnmrzoI1F3Ycl61FkG9HDBx3fFYtPRc1i+JxNTv0jCHXFBmDa0NzwceXYekS0RQmB9ch7e3n0ano4KfHB7NGJ7tvzDWj2ZTIYlS97Ehx++h4ULn0VNTTUGDBiEFSs+gr9/AA4e/LvB99C1146DVluLr776EitXvg9PT0/cfvtk3HPPAx34KYmoNSID3LD6nni8tzsT3xzKw6/pRXh6bChG9ve2dtPIyiRCCNHci4WFlZ3ZFot4eDh1ahGvu2DcLMeYXRlrxk0IgT+ySvH2b6eRWVyNuCA3PD6qHyIC3KzSntZirl0Ze4ibr69rm/7elrfn1mCtnDlXocGyX09jV1oRBvZwwf+uH2BXPVftYV21RYy7dVxJ3MuqdXjv90xsPnoOzioZ7hvSC5Pjgy475jM1Zss5z+15++rMZZ1focFLP53CgewyDO/nhefHD4CHE38Ma4otr4P2jMul4xzNq8DL208ho6gaY8N88OSY/vBxUbXqb7lcbFdLy6a5bTrPayCibkcikSCxrxeuCvbE5mPn8OHeLNy/9hDGh/vi0RF9EeDmYO0mEpEdqarVY/XfOVj9dw4AtLoXOxHZFw8nBZ67LgyT44Kw/PdMLP89C+sP5eE/VwfjxogekHM4PKJOpzcK/N+hPHywNwtCAM+MDcEt0QE8W5aIzKIC3fDl1His/jsHH+07gz8yS3FfQi/cPZg/ltsjFtqJqNuSSyW4NToA48N98cWBs1iTlItf0oowOT4ID1zVGy4qfgUSUcep1hrwf4fy8MVfZ1Gu0WN8uC8eG9EX/vyxj4guI8TXGctuicTBnDK8tzsTL29Pw6f7s3FfQi/cGOEPZQdc14qIGvs7uwxv/pqBtEI1hgZ74plxoQh05zaciBpTyKR44KreGBvmi3d2n8b7e7Ow4XAeHhneFxMG+UHKH+fsBqtMRNTtOSvleHh4X9waE4j3f8/El3/lYPOxAky/Ohi3RvuzhxgRtauCylqsT87Ft0fyUVVrwNV9PPFQYh9E+LdtyAAisi/xPT3w8V2x2HO6BJ/uz8biHen46M9sTP1HT9wSHQBH9pIj6hBH8yqwYm8W/s4ug5+LEq/+cyDGhPqwFzsRtaiXpyOW3hyB5JxyvPXbaSz8MRVrk3Iw/epgjAzxZsHdDrDQTkR2o4erCgsnhGNyvOmCqUt3pWPdwRzcO6QXJg7qwR5iRHTFhBA4ca4SXx3MxY5TRRBC4JpQX0z5RxAibfz6EERkuyQSCUb298aIfl74K7sMn+7PxrJfT+PjP7Nxc6Q/bo8L5JB4RO0ktaAKH/yRhd9Pl8DTUYHZo/vh1ugADv1ARBaL6+mOT++OxY7UQqz4PQtPbT6Bft5OuP+qXhg3wI9DSHZjLLQTkd0J7+GKFbdHY8/pEny07wxe3p6GD/84g7sHB+GW6AAOKUNErVZQWYsfThTg+xPnkVlSDWelDHfGBeLOuCCeXk5E7UYikSAh2BMJwZ44kleBr5JysDYpB2uScjCyvzfujAvC4F7u7HFLZCGDUWB3RjG+Ts5F0tlyuKrkeGR4H9wZFwQnJQvsRHTlpBIJrgv3wzVhvtiRWohP92fjf9+n4sO9ptrDxEE94GHtRlK7YzWJiOzSxT3EDmSX4fMDZ/HO7kx8sj8bNwzqgUnRAQjxcbZ2M4nIBhVV1WJ3RjF2pRXhwJkyCAAxgW54dlwoxg3w5Y91RNShogPdEB04COcqNPi/w/nYeCQfv6YXo7enI26K9McNET3g46y0djOJbFpptRbbTpzHN8m5yKuohb+rCrNG9sWkqAC4OnA7TkTtRy6V4PqBfrgu3Be704vx6YGzWLorA+/uzsRNMYG4caAvBvbgEJPdhUQIIZp7sbCwsjPbYhEPDyeUlVVbuxldDuNmOcbsynTFuJ04V4k1f+dgV1oR9EaBqAA3TIr2x7gBvp0yDmpXjJktsIe4+fq2bcfLlrfn1mBpzgghcLq4GrszivFbejGOnzPFM8jdARMG+mHioB7o5enYUc3tNuxhXbVFjLt1dGbcNToDdpwqxOaj55CcWwGZBEjs540bI3pgWF8vqOxsaDxbznluz9uXpctardXjt/Ri/JRyHvuzSmEQpuEdJscHYWR/bw7l0E5seR20Z1wutuXEuUp8ezgfP6Weh0ZnRLifCyYM8sN1A3zh46KydvMILa8zzW3TWWi3M4yb5RizK9OV41bfw2XjkXycKa2BSi7F0GBPjA71xvB+3vBwVHTI/+3KMbMme4gbD8zbV2typkZnwMGccuzPKsWe08XIKdMAACL8XTGyvzdGhnijv7cTh2mwgD2sq7aIcbcOa8U9q6QaW46dw9bjBSip1sFZKcOYUB+MD/fFP3p72kUh0ZZzntvz9tWaZV1SrcUfmSXYk1GCvZklqNUb4e+qwviBfrg+3A8hvjyDtb3Z8jpoz7hcbJNUpcC6P7Ow6eg5nCpUQyoBBvfywPXhfhgZ0nG1B2oZC+3UKoyb5RizK9Md4iaEwKHcCuxILcSv6UU4X6WFTALE9nTHkN4eiO/pgQh/13a7iGp3iJk12EPceGDevprKGYNRIOV8FQ6cKcX+M6U4klcBnUFAKZPgH709MKq/N0b094Yve5hcMXtYV20R424d1o673ijwd3YpfkopxC9pRVBrDfB0VGB4Py+MCvHGVcGe3fYCj9aO/eVwe96+mlrWOoMRJ85V4q/sMuzNLMHx/EoIAD7OSowK8cb14X6IDnKDlD+UdxhbXgftGZeLbbp4uWQWV+OnlPP4OeU8zpZpIJMAMUHupk4+/b15Bm0nY6GdWoVxsxxjdmW6W9yEEDhZUIVf04vw++kSpBWqAQAquRRRAa6ICnRDuJ8LwvxcEOTucEW9XLtbzDqLPcSNB+bty8PDCSWlamQUqXHwbDmScspx8GwZyjV6AECYrzOuCvbEVcGeiAly67bFqM5mD+uqLWLcrcOW4l6rN+KPzBLsPFWIvZklqKo1QCWXIqG3B4b19UJCsCd6eVzZvostsqXYX4rb8/bl4eGE/MJKnCyoxOHcCiSdLcPh3Apo9EZIAEQEuCKxrxeG9/PCAD+XbpPjts6W10F7xuVim5paLvW1h98yirEno9hcewj2dMRVwZ5ICPbA4F4evC5UB7vSQjuXChFRK0gkEgzyd8Ugf1c8Mrwvymp0OJRTjoN1ty/+yoHBaPrd0lUlR4ivM3p7OKKnhwN6eTqiZ91jZyW/domswSgE0grVOJhTjqPnKnEgs8RcWA90U2FEf2/zjquXEy8iSETdh0ouxZhQH4wJ9YHeYMTBnHLszijG7oxi7DldAgDwd1WZD9yjA92uuNMAUUcyCoGskmocy6vEsXMVOHlejbSCShjqug6G+Djj5ih/xPfyQHyQOzycOOQCEXU9F9ceHk7sg7xyDXZnFGNfVgk2HzuH9YfyIJMAEQFuSOjtgauCPRER4AqFzL6uyWKr7KZH+2uvvYLXX1/SgS0isi9z5szD3LnPNvu6vf1iXqs3IqNIjdTzVUg9X4X0QjVyyjUoVmsbzOflpECAmwN8XZTwcVbCz1UFH2clfF2U6OvvDpnBAA8HOeTcSLaaPeQae8BZzmAUSCuswsGcciSdLceh3HJU1BXWe3k6Ii7IDfE9PRDfyx0Bbg5Wbm3H4L4PUdu1tL/TnK6wbRJCIKdMgwPZpThwpgx/ny0zf096OSkQGeCGyABXhPm6IMTXGX4uyi5RfLfl2HN73no1OgMyitQ4VahGeqEaaYVVSCtUQ601AABcVDLE9PTAAF9nRPq7IirArdsU1rn9Jmq7K91+dzZLt1lavRFH8yvqhrssw8mCShgFoJRJMLCHK6ID3RBVd/NxZuehtmCPdiIiK1LJpeZfnS+m1uqRW6ZBTlkNzpZpcLa0BgWVtThbVoPknHJzj9pLOStl8HBUwN1RAQ9HOdwdFPBwVMDNQQ4XlRzOSlndTQ5nVd29UgZnlQyOChnHnSS7U1hVi2P5lTiWX4nj5ypw8lwVqnWmg/FeHg4YE+KD+F7uiO/pjvDeXjZbhCEi6iwSiQS9PB3Ry9MRt8UEwmAUOF2sxtG8ChzNr8TRvArszig2z++qkiPExwnBXk7o6eGIIHcH9PRwQKC7A1xV8i5RhCfbU1Wrx5nSGpwpqcaZkmpkldQgvUiNs6U1qO8R6KyUIcTHGdcP9ENkgCsi/d3Q28sRXp7O3J4TkV1RyqUY3Mt0BtrDw4EKjc7cqehoXiXWJefiy79zAACB7g6ICnDFAD/TD+ahPs7wdu4aP5p3ZSy0ExF1IGelHGF1Y7c3pVZvRJG6FoWVWlQLIK9YjbIaHcpqdCjX6FFWo0NptQ5ZxdUoq9GbC4ctUcgkUMqkUMmlUMqkUMoveSyTQiGTQFF3Xz9dIZNCWTddedFrioumKeVSOCllcFHJ4VJ/r5JD1U4XhSW6HINRILdcg4wi0xjr6UVqHMuvREFlLQBALpUgzM8FN0b0QFSgG+J7usPPlRcwJSJqiUwqQaivC0J9XXBrjGlapUaPjCI10uq/cwvV+C29GKU1ugZ/q5JL4e2sNJ+l5+GogItKDjeVHC4OpntXBzlcVXI4KWVwUshM90p2Duju9AYjCqpqca7CdMuv0OBcRS1yy2uQVVKDoovO/pRJgCAPR/TzdsL14X4I9XVGiK8zAt0dmCdERE1wc1CYh4cDTD3eU85X4WheBY7kVSA5pxw/pRSa5/dwVJiL7v28ncxD3Pq5qvg9207sZugYMmHcLMeYXRnGzXKtiZnOYIRaa4Baq0e11gB1rcH83HRvQI3WgFqDEVq9EVqDEbV60+NavbHBdK3eCJ1BQGswQmdo+FhraHbT0CyFTAIXpRwuKlPx3c1BDjcHRd193eP66Y5yuKkuvNaWi03aQ67Zy6nmRiFQUaNHaY0OpTValFbrzAflueUa5FdokFOmQa3eaP6bIHcHRPi7IiLAFZEBbhjg59Lijz72kDO2iHG3DsbdOrp73OvP2Mst1yCvXIMitdZ0q6pFkVqL8ho9Kmr15uvXXI5D3Q/4lxbgnRRyOCmlcFLK4aQw3TsqTGf0XTyvY91ZfvXPfb1dbDb2XWV7LoSAQZh+3DYKAYNRQG8U0OgMqNEZUa0z1D02oFprQGWtvq6jiB6l1ablX1qjQ2m1KS8uTQMvJwWC3B3Q28sJfTwd0cer/kwJB4vGGO7u61lXxeVim7hcbFNnLJfyGh3S634sT6u7Ty9SNzimUsokpiFu64a2bXBzUcLTSVF3rC+Ho0JqF73iOXQMEZEdUMik8HCUwsOxY8egFMJ0QKU1GKHTm+619cV4vRFqnR5VtQZU1Zru1Vq9+XFVrR5VWj0qNXrkV9SiQqNHpUaHy9XulTJJ46K8+bEcrioF3M0F+guv80rr1pNTVoP3f89Crd4IAVORHHX3QsB0g4BRmPLJNA8AUTcNpjM6auoO1OsP1puqyTgrZQh0d0BPd0dcFeyJ/j7O6F/XC8OxDT/SEBHRlWnpjD3A9N1fozOista0T1BRq0Olpv773tRBoEZnMN9Xa+tuOgNKq3XI1WpQfdH01nYBUMgkF4rwFxXm6x87KmRQyKSQSyWQyySme2nD5z7OSowO9bGL3n3ZpTWYvu4Qqmr1MAjAaBStjvWlnJUyeDqZhjv0dVEi1NcZPVxVCHBTwd/NAf6upnueBUlE1HncHRXm4WbqGYwCBZW1yCmrqbuZfjwvrNLiSG45itTaZjvfySSAc93Z7c4q04/g9WfN159JX//YdJa9BFKJ6SaRABIJTI+Bi6ZJIK2/By4p5JvacXE3cdHoVdN+x6UcFTJMGOjXpo59lmKFgoiIGpFIJObhYtAO11ARQkBd1+OpokaPco0OlbV6lGv0qKi58LhSo0eFRof8Cg1OndejQnP54XIkABwUMijqDozrh76RXzT0jUIqgVR6+QNlIS703qrvuWVs8Nj0/OKeXcYm5vVzUeGr+wZD1sL/6w5qdAacLauB3iggwYWdI+CiHSZI6nakTK9LAEilEsjqHrs5mHbMnBQyOCikcFbJ4emogKejAh5Opnt/NxXH/iUi6oIkEom5d3qPNg7hJYRArd7YoDhfX5SvrjubT60z3RukEpRUaBoV8YvUWnPRXl/XQ1tvMDbZEUApk2DjvxPg69L9hx7zclLgX7GB0OgMpkKIVAJZXRFEVrfNlkpNj50UUjgqZBduShkcFVK4quTwcFRY1BudiIisRyaVINDddJ2VhGDPRq8LIVBZq0eRWotitelM4yqtAVUaU6e6C53u9NDUnUGvrtWbzqY3XHRGfd3zVpzg1iEUMgki/F0v2zGgvbHQTkREHU4ikZjHcg9ws+xvdQajuUBfUWsqxFdo9HU3HYxSKaqqtQ2GvtEbTT3vdUYBXSs37FKpBHJJ/QFl3QHmJQecjR+bissyqQQSiQQBbirYQY0dABDq64IvpsZbuxlERGQHJBIJHBSyVvVIs/Q0/IuHRtEbBPRGIxQyqd2cNeeikmP61cHWbgYREdkQiaT+jHMF+nk7t8t7CtH4bOf6aRefHV3//OLDakn9M8nF0y5u7yXz1U2TSyWd2psdYKGdiIhsnEImhZeTEl5OTXet53iDREREdKWkEgmkMgkUMgAdOzIfERGR3ZJITJ3WGpbIux+e20VERERERERERERE1AYstBMRERERERERERERtQEL7UREREREREREREREbcBCOxERERERERERERFRG7DQTkRERERERERERETUBiy0ExERERERERERERG1AQvtRERERERERERERERtwEI7EREREREREREREVEbsNBORERERERERERERNQGLLQTEREREREREREREbUBC+1ERERERERERERERG0gEUIIazfiSvz6668YPXq0tZvR5TBulmPMrgzjZjnG7MowbmQp5ox1MO7WwbhbB+NuPYy9/eCytk1cLraJy8U2cbnYritdNl22R/tvv/1m7SZ0SYyb5RizK8O4WY4xuzKMG1mKOWMdjLt1MO7WwbhbD2NvP7isbROXi23icrFNXC6260qXTZcttBMRERERERERERER2QLZwoULF1q7EVeqT58+1m5Cl8S4WY4xuzKMm+UYsyvDuJGlmDPWwbhbB+NuHYy79TD29oPL2jZxudgmLhfbxOViu65k2XTZMdqJiIiIiIiIiIiIiGwBh44hIiIiIiIiIiIiImoDFtqJiIiIiIiIiIiIiNrA5gvtWq0WN954I/74449m55k2bRoGDBjQ4LZjx45ObKVtyM7OxowZMzBkyBCMHDkSS5YsQW1tbZPzpqSk4M4770RMTAxuvfVWHDlypJNbazssiRtz7YKMjAzcf//9iIuLw5gxY/DRRx81Oy/zzcSSmDHXGps/fz7uueeeZl9nntkvrVaLBQsWYMiQIUhMTMSqVauanXfDhg0YN24coqOjMXny5EZ58v334VvhsQAAD3dJREFU32PcuHGIiYnBww8/jOLi4o5ufpfVXnE3Go2Ijo5u9J1XUVHRGR+jy7Ek7vXKysowbNgwfPvttw2mM98t016xZ85bxpK4t7T/xJzvHsrLyzFnzhwkJCRgxIgReP3112EwGKzdLLtXUlKC2bNnIyEhAWPGjMFnn31m7SbZtaZqabm5uZg2bRpiY2MxYcIE/Pbbb1Zsof1qrs555swZREdHQ6/XW6ll9qupZbJv3z7cdtttiIuLw/jx4/HNN9+07s2EDdNoNOLRRx8VYWFhYu/evc3ON2LECLFt2zZx/vx58622trYTW2p9tbW1YsKECWLmzJkiPT1d7N+/X1x77bVi8eLFjeZVq9UiMTFRvPzyyyI9PV289NJLYujQoaKystIKLbcuS+ImBHOtnlarFWPGjBHz5s0TWVlZYteuXSIuLk5s2rSp0bzMNxNLYiYEc+1Sf/zxhwgLCxNTp05t8nXmmX1btGiRuOGGG8TRo0fF9u3bRVxcnNi6dWuj+X7//XcRFRUltm3bJs6cOSNeeeUVkZCQYM6Tw4cPi6ioKLFhwwZx8uRJMXXqVDFt2rTO/jhdRnvFPSsrSwwYMEDk5OQ0+M4zGo2d/ZG6hNbG/WJPPfWUCAsLExs2bDBPY75brr1iz5y3jCVxv9z+E3O++5g9e7aYOnWqSE1NFfv27ROJiYli1apV1m6W3bvrrrvELbfcIo4cOSL27dsnRo0aJT777DNrN8suNVVLMxqN4qabbhKzZ88WaWlp4sMPPxTR0dEiOzvbyq21L83VOfPy8sT48eNFWFiY0Ol0Vmyh/WlqmWRmZoqoqCjx/vvvi6ysLLFp0yYRGRkpdu7c2eL72WyhPS0tTdx0003in//852UL7ZWVlSIsLEycPXu2k1toW/766y8REREhqqqqzNM2b94shg0b1mjeb775RowePVoYDAYhhOkLd9y4cWL9+vWd1l5bYUncmGsXnD17Vjz++OOipqbGPO3RRx8Vzz33XKN5mW8mlsSMudaQWq0W1157rZg8eXKzhXbmmf1Sq9UiKiqqwX7C8uXLxeTJkxvNu3HjRrFy5Urz8/p1LSkpSQhhKog9+eST5tfz8vJEWFiYyMrK6sBP0DW1Z9x37NghxowZ0/GN7gYsiXu9X3/9VYwfP14MHTq0QbGX+W6Z9ow9c771LIl7S/tPzPnuIz4+Xmzfvt38fPHixeLBBx+0Yovo6NGjIiwsTGRkZJinbd26VSQmJlqxVfapuVraH3/8IaKiohp0RLrvvvvEm2++aa2m2p3mls327dvF0KFDzdNZaO88zS2T5cuXizvuuKPBvM8995z473//2+J72uzQMX///TcSExPx9ddfX3a+9PR0qFQqBAYGdlLLbFO/fv2wcuVKODs7m6dJJBJotdpG8x4+fBjx8fGQSqXm+eLj45GcnNxp7bUVlsSNuXZBz5498dZbb8HBwQFCCCQlJeGvv/7C1Vdf3Whe5puJJTFjrjW0bNkyJCQkICEhodl5mGf2KyUlBVqtFoMHDzZPGzx4MI4ePdrotMubb74Z06dPBwBoNBp89tln8Pb2RlhYGABTHg0ZMsQ8f0BAAIKCgphHTWjPuGdkZKBv376d1/guzJK4A0BVVRUWLlyIRYsWQaFQNHiN+W6Z9ow9c771LIl7S/tPzPnuw8PDA5s3b0ZNTQ0KCgqwZ88eREREWLtZdu3s2bNwd3dHv379zNPCw8NRWFiInJwcK7bM/jRXSzt8+DAGDRoEFxcX87TBgwfj0KFDnd1Eu9XcstmzZw9mz56N+fPnW6ll9qu5ZTJhwgQsWLCgwTSJRNLsMNMXk7drC9vR5MmTWzVfeno63NzcMHv2bCQlJcHf3x8zZ87EqFGjOriFtsXLywvDhg0zPzcajVi9enWDndJ6hYWFjXbuvb29kZKS0uHttDWWxI251rSRI0fi/PnzGDNmDMaPH9/odeZbYy3FjLl2QXJyMn788Uds3boVn3zySbPzMc/sV2FhIdzd3aFSqczTfHx8oNPpUFJSAj8/v0Z/s2fPHnPh9/XXXzcfcJw/f77R/N7e3igoKOjAT9A1tWfc09PToVarMWXKFJw5cwYDBw7EM8880+BgnUwsjfvSpUsxYsSIBsXFesx3y7Rn7JnzrWdJ3Fvaf2LOdx/PP/885s6di/j4eBiNRgwdOhQzZ860drPsmo+PD6qqqlBVVWXevufm5gIASktL0bNnT2s2z640V0srLCxs8jvw3LlzndEsQvPL5oUXXgAA7N+/vzObQ2h+mVxaWygqKsK2bdvw2GOPtfieNtujvbUyMjKgVqtxzTXX4KOPPsKoUaMwY8YMHD582NpNs6rFixfj5MmTePLJJxu9VlNTA6VS2WCaUqlsshe3vblc3JhrTVuxYgVWrFiB48ePY/HixY1eZ7411lLMmGsmWq0W8+fPx7PPPgt3d/fLzss8s1/NLXsAzS7/8PBwfPfdd3jssccwb948c08ejUbDPGql9ox7RkYGysvL8dhjj2HFihVQqVS49957UVlZ2bEfoguyJO4HDhzAL7/8gqeeeqrJ92K+W6Y9Y8+cbz1L4t7S/hNzvvvIzs7GoEGDsHr1aqxcuRK5ubl49dVXrd0suxYTEwN/f38sXLgQVVVVKCgowHvvvQcA0Ol0Vm4dAabv00vPsFIqlVw+RC2orq7GY489Bj8/v1Z1CrfZHu2tNWfOHDz88MNwc3MDYDqIO378ONatW4eYmBgrt67zCSHw8ssv46uvvsLbb7+N0NDQRvOoVKpGO5RarRYODg6d1Uyb05q4MdeaFhUVBcB08PL0009j7ty5DQ5imG+NtRQz5prJ8uXLERwcjAkTJrQ4L/PMfjW37AHA0dGxyb/x9fWFr68vBg4ciOTkZKxbtw6xsbHMIwu0Z9zXrl0Lg8EAJycnAMAbb7yBUaNGYefOnZg0aVLHfpAuprVx12g0eO6557BgwQK4urpa9F7M96a1Z+yZ861nyXdNS/tPzPnuITs7G6+88gp27doFf39/AKY8mTZtGh566CH4+PhYuYX2SalU4t1338UTTzyBIUOGwMXFBXPmzMHhw4cbDFVC1qNSqVBVVdVgGr8DiS6vsrISDz30EHJycrB27dpmj3Mu1uUL7TKZzLwzVa9fv35ITU21Uousx2g0Yv78+diyZQuWLVuGsWPHNjlfjx49UFhY2GBaUVERfH19O6OZNqe1cWOuXVBQUIBjx47h2muvNU/r378/dDodqqqq4OXlZZ7OfDOxJGbMNZMtW7agsLAQcXFxAEy9YQwGA+Li4hqNp8o8s189evRARUUFtFqt+QerwsJCKJXKRmdCJCcnw9HREeHh4eZpISEhyMrKMr9XUVFRg79hHjWtPeN+8ZAQ9c979uzJ4Rya0Nq4HzlyBGfOnMHcuXPN02pqavD888/j0KFDePHFF5nvFmrP2DPnW8+S75qW9p+Y893DsWPH4OzsbC6yA0BkZCQMBgPy8vJYaLeiiIgI/PTTTygqKoKbmxuys7MhlUp53Skb0aNHj0bDavI7kKh5JSUlePDBB1FUVIQvvvgCvXv3btXfdfmhY2bNmoWFCxc2mHby5Em7vMDQkiVLsGXLFrz77ru47rrrmp0vJiYGycnJEEIAMPXmTk5ORmxsbGc11aa0Nm7MtQsyMjIwc+ZMFBcXm6cdP34cXl5eDQrGAPOtniUxY66ZfPnll9i6dSs2btyIjRs34vbbb0dkZCQ2btzYaF7mmf0aOHAgFApFgx9fkpKSEBERAbm8YX+CNWvW4K233mow7fjx4+ZxkWNiYpCUlGR+LT8/H3l5ecyjJrRX3PV6PUaMGIFt27aZX1Or1Thz5gzHq25Ca+MeHR2Nn3/+2fz9uXHjRvj4+GDWrFl4/PHHATDfLdVesWfOW8aS75qW9p+Y892Dn58fKioqkJ+fb56WkZEBABwH3IrKy8tx9913o7i4GD4+PlAqldi1a1eji2+S9cTExCAlJQXV1dXmaUlJSfwOJGqCVqvFjBkzUFpaijVr1li0j9YlC+2FhYXQaDQAgGuuuQYbNmzAli1bkJWVhXfeeQdJSUm49957rdzKznXo0CF8/vnnmDVrFiIjI1FYWGi+AQ1jdv3116O6uhqLFi1Ceno6Fi9ejKqqKkycONGaH8EqLIkbc+2CIUOGoH///pg3bx4yMjLwyy+/4I033sCMGTMAMN+aYknMmGsmQUFBCA4ONt/c3Nzg4OCA4OBgAMwzMnF0dMSkSZPwwgsv4MiRI9i5cyc++eQT8/pycZ5MmTIFu3fvxurVq5GVlYVly5bh+PHjuO+++wAAd911F7Zu3Yr169cjNTUVTz/9NEaOHIk+ffpY6+PZrPaKu1wux/Dhw7Fs2TL89ddfOHXqFObMmQNfX1+MGTPGmh/RJrU27vXflRffpFIpvL294e3tDYD5bqn2ij1z3jKWfNe0tP/EnO8eYmNjzRcQTklJwaFDh7BgwQLcfPPNjTqvUOdxd3eHRqPBkiVLkJ2djR9++AErVqzAI488Yu2mUZ2EhAQEBgZi3rx5SEtLw8qVK3H48GHcfvvt1m4akc357LPPzNfUc3R0NNcJy8rKWv5j0QWEhYWJvXv3Nni+YcMG8/Mvv/xSjB07VkRGRopbb71VHDhwwBrNtKolS5aIsLCwJm86na5RzA4fPiwmTZokIiMjxW233SaOHj1qxdZbj6VxY65dkJubKx566CERFxcnhg8fLj744ANhNBqFEI3XUeabiSUxY6419uabb4qpU6eanzPPqF51dbWYO3euiI2NFYmJieLjjz82v3Zpnvz8889i4sSJ5jxJSkpq8F7ffvutGD16tIiNjRWPPPKIKC4u7rTP0dW0V9zVarV48cUXRWJiooiJiREzZswQubm5nfpZuhJL4n6xESNGNHqN+W6Z9oo9c94ylsS9pf0n5nz3cO7cOTFr1iyRkJAgEhMTxaJFi0RNTY21m2X3MjMzxT333CNiYmLEddddJzZt2mTtJtm9S2tpWVlZYsqUKSIyMlJMnDhR7Nmzx4qts2+XLhshhPjzzz/NtSjqfBcvk1tuuaXJOuHkyZNbfB+JEHXn2RMRERERERERERERkcW65NAxRERERERERERERES2goV2IiIiIiIiIiIiIqI2YKGdiIiIiIiIiIiIiKgNWGgnIiIiIiIiIiIiImoDFtqJiIiIiIiIiIiIiNqAhXYiIiIiIiIiIiIiojZgoZ2IiIiIiIiIiIiIqA1YaCciIiIiIiIiIiIiagMW2omIiIiIiIiIiIiI2uD/ARRkL/2kmFFtAAAAAElFTkSuQmCC\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], "source": [ "az.plot_posterior(data);" ] @@ -276,9 +450,114 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 7, "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Auto-assigning NUTS sampler...\n", + "Initializing NUTS using jitter+adapt_diag...\n", + "Sequential sampling (2 chains in 1 job)\n", + "NUTS: [y0, g, gamma, sigma]\n" + ] + }, + { + "data": { + "text/html": [ + "\n", + "
\n", + " \n", + " \n", + " 100.00% [3000/3000 22:27<00:00 Sampling chain 0, 0 divergences]\n", + "
\n", + " " + ], + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "\n", + "
\n", + " \n", + " \n", + " 100.00% [3000/3000 13:37<00:00 Sampling chain 1, 0 divergences]\n", + "
\n", + " " + ], + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Sampling 2 chains for 1_000 tune and 2_000 draw iterations (2_000 + 4_000 draws total) took 2165 seconds.\n" + ] + }, + { + "data": { + "text/html": [ + "\n", + "
\n", + " \n", + " \n", + " 100.00% [4000/4000 01:01<00:00]\n", + "
\n", + " " + ], + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], "source": [ "with pm.Model() as model3: \n", " sigma = pm.HalfCauchy('sigma', 1)\n", @@ -300,9 +579,20 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 8, "metadata": {}, - "outputs": [], + "outputs": [ + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAA7AAAADgCAYAAAA624qoAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+j8jraAAAgAElEQVR4nOzdd3hUZfbA8e/MpNeZ9JCeAKGESCgp9KayigJiXdTFxqI/2yruWlddsWJBxYJixYKudQULCor0TpAeUkiD9ElPZjKZ3x9hIoEUSsidcj7Pw/NocmdyMrn3vvfce877qsxmsxkhhBBCCCGEEMLKqZUOQAghhBBCCCGEOBWSwAohhBBCCCGEsAmSwAohhBBCCCGEsAmSwAohhBBCCCGEsAmSwAohhBBCCCGEsAmSwAohhBBCCCGEsAmSwNq5zz//nAsuuIDExESuuuoqduzYoXRIQti9+Ph43nnnHaXDEEIIIYSwO5LA2rFvvvmGRx99lEsvvZRXX30Vb29vbrrpJvLy8pQOTQi79tlnn3HJJZcoHYYQQgghhN1Rmc1ms9JBiO5nNpuZOHEio0eP5vHHHwfAaDQyefJkxo8fz8MPP6xwhEIIIYQQQghxeuQJrA175plnSE5OxmAwtPn6jTfeyJ133klBQQETJkxo/bqzszPjxo1jzZo1PR2qEHYnPT2dmTNnkpSURHJycusxByeXEG/cuJHLL7+cxMRELr74YtasWcOAAQP46quvAHj11Ve57LLL+Oabbzj//PNJTExk1qxZFBcXs3TpUsaNG8fQoUOZO3cu9fX1re+blZXFnXfeSWpqKgkJCUyYMIHXXnsNuS8pHFVjYyPz5s0jLS2NIUOG8NBDD/Hiiy+2joU1NTXMmzeP8ePHk5CQQGpqKv/617+oqqpqfY/4+Hi++OIL7rjjDgYPHsyoUaP45JNPKCoqYvbs2Zx33nlceOGFrF69uvU11113HU8++STz588nNTWVIUOG8Oijj1JXV8e8efMYNmwYo0aNYtGiRW3iXbNmDddeey1JSUkMGjSIqVOnsmLFip75sISwQ12dA4R9kATWhk2bNo3KykrWrl3b+rWSkhI2btzI1KlTAYiKimrzmoiICHJzczGZTD0aqxD2pL6+ntmzZxMcHMzrr7/OE088wd69e7nnnntO2vbAgQPccsstBAQE8OqrrzJ9+nTuvvvuk47B7Oxs3n77bf75z38yb9480tPTue666/jyyy959NFH+fvf/86yZcv48MMPAaitreX6669Hr9fz7LPPsmjRIlJSUnjllVf49ddfe+RzEMLaPPjgg3z11VfcfvvtvPDCC+Tm5vLee++1fv/ee+9l1apV3HvvvbzzzjvceOONLFu2jNdff73N+zz99NNERUXxxhtvkJSUxBNPPMGsWbMYMmQIL7/8Ml5eXtx3331tbih9+eWXZGZm8uKLL3LjjTeydOlSpk+fTnV1NS+99BJpaWm8+OKLrXNR7Nq1i9mzZ9OnTx9ef/11XnrpJdzd3bn33nspLy/vmQ9MCDvT1TlA2AcnpQMQZ65fv37069ePZcuWtd5ZWr58Od7e3lRXVwPg6enZ5jWenp40NzdTX1+Pl5dXj8cshD3IyMhAr9dz3XXXkZSUBIBOp2Pjxo00Nze32fatt94iJCSEhQsX4uTkxNixY1Gr1Tz77LNttqurq+Opp57ivPPOA+C3335j+fLlrFq1irCwMMaPH89vv/1Geno60JLwRkZGsmDBAvz8/ABIS0vjl19+YcuWLXK3WTic7Oxsli1bxtNPP81ll10GQGpqKhMnTgRanswYjUYee+wxxowZA0BKSgo7duxg8+bNbd4rKSmJuXPnAhAcHMyKFSsYPHgwc+bMAcDV1ZVZs2aRk5ND//79AdBoNCxYsAA3NzdGjBjB0qVLaW5u5umnn0atVpOSksLy5ctJT08nKSmJjIwMzj//fB599NHWn9urVy+mT59Oeno648ePP7cfmBB2pqtzgLAfksDauGnTpvHyyy9TV1eHh4cH//vf/7joootwcmr506pUqjbbW0oLT/y6EOLUxcbGotVqmTNnDhdffDFjx44lLS2N5OTkk7bdvHkzkydPbj0mASZPnnxSAqtSqUhISGj9f39/f/z8/AgLC2v9mlarbb05lZCQwCeffILRaOTQoUPk5OSwd+9empqaTmorEMIRbNmyBYBJkya1fs3d3Z2xY8eyadMmXF1deffddwHIz88nJyeHjIwMMjMzcXV1bfNeiYmJrf8dEBAA0Ob41Gq1ACeVHru5ubX+v7+/P9HR0ajVLcVuLi4ueHh4tB7DM2bMYMaMGdTV1ZGZmUlOTg4bN24EkGNYiDPQ1TlA2A8pIbZxl1xyCUajkVWrVpGdnc2ePXuYOnUq3t7eQEuZ4fHq6upQq9V4eHgoEa4QdsHLy4uPPvqItLQ0vv76a2bPns3IkSP55JNPTtq2oqKi9QmpheWC+Hju7u5oNJqTvtaZN998k7S0NC6++GKeeuopDh8+jJOTk/TACodUUVGBs7MzPj4+bb5+/PG2cuVKJk2axMSJE5k7dy7r1q3Dzc3tpGPmxOol6Pp4PN3X1NXVMXfuXIYPH87VV1/NokWLaGxsBJBjWIgzcCrnAGEf5AmsjQsICGDkyJH89NNP5OfnExUVxeDBg8nOzgYgLy+vTR9sXl4eMTEx8gRWiLPUp08fFixYgMFgYNu2bXzwwQc8/vjjDBw4sM12QUFBJ/WzdUd/2zfffMOCBQt49NFHmTJlSutNq7S0tLN+byFsUVBQEEajkaqqqjYXsJbjLScnh7vuuovp06fz0UcfERISAsBdd91FZmZmj8f7xBNPsG7dOt566y2GDx+Oi4sLhw4d4rvvvuvxWISwB12dA4T9kCewdmDatGmsXbuWFStWcOmllwIQHR1NaGgov/zyS+t2RqOR3377TS5whThLv//+O2lpaZSXl+Pi4kJaWhqPPPIIAIWFhW22HT58OKtXr27TG7ty5cqzjmHHjh2EhIRwzTXXtCave/bsoby8XJ7eCIc0ZMgQ1Go1q1atav2awWBonXl/7969GI1GZs+e3Zq81tXVsW3bNkWOmZ07dzJ69GhGjhyJi4sLQGuscgwLcfq6OgcI+yFPYO3AxIkTefTRR9mzZw8vv/wy0NJPd8stt/DEE0/g6+vLkCFD+Oijj6ioqGDWrFnKBiyEjUtMTMRsNnP77bdzyy234OzszAcffICPjw8pKSlttp09ezZTp07ljjvu4KqrriInJ6f1OLX0xp2JQYMGsXTpUhYuXEhycjKZmZm89tprqFQqGhoazur3E8IWRUVFcckllzBv3jzq6uoICwvjww8/pKSkhF69etG/f380Gg3z58/nmmuuoaKignfffZfS0tLWBLInDRo0iFWrVvH1118TGhrKxo0bW5ffkmNYiNPX1TlA2A9JYO2Aq6srycnJlJeXExER0fr1mTNn0tjYyIcffsj7779P//79eeedd9psI4Q4fVqtlsWLF/PCCy/wz3/+E6PRSGJiIu+9995J/a5xcXG8+eabzJ8/n9tuu43o6GgeeOABHnroobPqRb/sssvIzs5m6dKlLF68mLCwMG666SYyMzPZtm3b2f6KQtikxx57DDc3NxYsWEBTUxNTpkxh8uTJHDp0iJiYGJ599lkWLlzI7NmzCQwMZMyYMcyYMYP//Oc/FBUVERwc3GOx3n///TQ0NPDUU08BLeeKhQsX8tRTT7Fjxw6mT5/eY7EIYS86OwcI+6EyS52KzWtsbGTMmDHMnTuXK664QulwhBDHWb9+PZ6enq3L4wCsXbuWm266iW+//ZZ+/fopGJ0Q9qO8vJx169Yxfvz4NsvEXX311QQEBLBw4UIFoxNCnGtyDnAc8gTWhlVWVrJkyRI2bdqERqNhypQpSockhDjBzp07eeedd/jXv/5FTEwMBQUFvPLKKwwfPlySVyG6kZubG48//jg//vgjV199NU5OTvzwww/s3LmT9957T+nwhBDnmJwDHIc8gbVhDQ0NjB8/HldXV5566ilGjBihdEhCiBM0NTWxcOFCli1bRlFREb6+vpx//vnce++9be4QCyHO3q5du3jppZfYvXs3RqOR+Ph4br31VsaNG6d0aEKIHiDnAMcgCawQQgghhBBCCJsgy+gIIYQQQgghhLAJksAKIYQQQgghhLAJnU7iVFJS3VNxdAsvL1dqahqVDuOM2HLsYNvxW3PsgYHep7SdrR2rZ8Oa/15Kks/lZD35mZzqsQo9f7zKvtE++VzaZ++fizUfq+2x97/HuSSf3Zmxls+ts2PVrp7AOjlplA7hjNly7GDb8dty7I5I/l7tk8/lZPKZtJDPoX3yubRPPhfrIn+PMyef3Zmxhc/NrhJYIYQQQgghhBD2SxJYIYQQQgghhBA2QRJYIYQQQgghhBA2QRJYIYQQQgghhBA2QRJYQVOzmXx9PaW1BqVDEeKUGZqaKaxsoKzWgNlsVjocIYQVM5qaOVLVQE1jk9KhCCFOk9lsprzOwJGqBppMzUqHI6xAp8voCPtWbzTxzsZcvko/QvWxQX1AiDd3jY1hSLhW4eiEaF9mSQ3P/3iANVllNDa1DGQh3q5MHRTCzGHhuDtb/+x5Qoie0WA08faGXL7aVUhNowmApDAfbh0Vw3ith8LRCSE6Y2o289WuIyzdXkBuRT0Ani4apgwM5pa0KHzdnRWOUChFElgHVVTdyD++3k1GSS2T+gaQGq2jsr6J/+4sZM5nu5g7IY4rk8KUDlOINr7YWciLv2XiolFzaUII/YK8qDOaWJ9dzqL1h1m+t4hnpgwgPthL6VCFEAorrzNw+xd/kFFSy/nxgSRHaimpNfDtH0eZ83k691c1Mn1AkNJhCiHaoa8z8s/v9rIjv5LEXj7MOC8UD2cNOwoq+WJnIb8cLOWFaQMZGHLq6/oK+yEJrAPS1xm57b+7KKs18PJlCYyI8Wv93hVJvXh4+X7mr8rE3VnDJQkhCkYqxJ8Wrcth8cZcxvYN5KGJceg8XFq/d/WQMLbl6fn39/uZ/Vk6L102UKoIhHBgtYYm/u+/f5Cnr2fB9ARGxv45zv11aBj/+fEgT/2wn/p6A38dGq5gpEKIE1XWG/n75+nk6+t5dHJfLh4QjEqlAmBaYih/HRrOP7/dw5zP0ll4+SDOC/NVOGLR06QH1sE0NZv513d7KapuPCl5BXB31vDsJf0ZHqnlqZ8z2F9UrVCkQvzpi52FLN6Yy6UJwSyaOaRN8moxNELL+zOTCPJ24Z6v93CopFaBSIUQSjObzTy5IoOsslqenzqgTfIK4OnixJNT+nPBgGBeXp3FltwKhSIVQpzI1Gzm/u/2ttx8uiyBKQNDWpNXi/ggL975axJB3q784+s9reXFwnFIAutg3t+Uy/b8Sh6Y1KfDO1ZOGjVPT+mPzsOZf/9wAEOTNMwL5ew5UsXzv2YyMsaPB8/vi0at6nDbQC9XFl6eiIeLhnu/2U11g0zYIoSj+eVgKT8fKGHOyGhSo/3a3cZJreK5ywYRpfPg398foKrB2MNRCiHa89HWfLbmVXL/xD4Mj9R1uF2ApwuvzhiEWgX3fbuHBqOpB6MUSpME1oEcLq/j3U25XNgvkIsHBne6ra+7Mw9d0Jfssjo+2ZbfQxEK0VaD0cQj3+8n0NOFJy7q12nyahHs7cpzlw6gqMbAUz9n9ECUQghrUWto4vlVh+gf7MV1wyM63dbT1Yn/XBRPRZ2BBb9l9VCEQoiO7Cuq5o11OUzqG8AlCZ1fpwL08nVj3sX9yCqr4+0Nh3sgQmEtJIF1EGazmedWHsLVSc3d4+JO6TUjY/wY19ufdzflyhI7QhEfbc0nT9/Avyf3xdvt1Fv2E0J9uCUtkl8OlrAhp/wcRiiEsCafbiugvM7Ivyb2xukUbnj1C/Zm5rAIvttTxO4jVT0QoRCiPU3NZh7/8QD+Hs7cP6nPSWXDHUmN9mPqoBA+2prP3qPS9uYoJIF1ED8fKGFzrp5bR8YQ4Hly/2BH7hgTS2NTMx9vlaewomcVVTfyweY8JvQJ6LSMqCPXDYsgUufO86sypQxeCAegrzfy0dZ8xvX2Z2Cozym/7sbUCPw9XXjptyxZU1oIhXyz6wiZpXXMndD7tJfHuWtMLP6eLsxbcVDWiXUQksA6AKOpmYVrsokP8mLGeaGn9dpInTsX9Aviy/RC9HXSIyR6zsI12TSbzdw5NuaMXu/ipGbuhDhyK+r5WMrghbB7S7bkUWcwMWdk9Gm9ztPFiZtTI9lVWMXWPP25CU4I0aHqhibeXJfD0AhfxvX2P+3Xe7s5cd+E3mSU1PK/3UfPQYTC2kgC6wC+21PEkapGbh0VfUo9hCe6ISWCBmMzn26XJED0jEMltfy4r5iZw8IJ83U/4/dJi24pg39/Ux6V9XIDRgh7VVFn4LMdhUzuH0RcgOdpv/6ShBACPF14d2PuOYhOCNGZdzbmUtXQxD/GxZ1y6fCJxvX2Z3CYD29tyKVeJnSye5LA2jmjqZn3NuaSEOrNiOjTL8MEiPX3ZELfAD7bUSizuooe8fG2fNyc1MzshvUZ/z4imjqjic93FHZDZEIIa/TNH0dpbGrmhpTIM3q9q5Oa64aHszWvkvSCym6OTgjRkdKaRv67s4ApA4OJD/I64/dRqVTcPjqGslqDTD7qACSBtXP/232Uo9WNzB4RdcZ3tQBmJUdQazCxbG9RN0YnxMlKaxr5cV8xlySEnHYfTHt6B3oyNs6fpTsKqDXIDRgh7E2TqZkvdhaSEqUlxt/jjN9nemIoWndn3tuU143RCSE6s2RrPqZmMzemntnNp+OdF9ZSgrxkS760vdk5SWDtWFOzmQ835zEo1JvUqDN7+mrRL9ibhFBvvtxZKJNciHPqv+lHMDWbuXpIWLe95w2pkVQ1NPHlziPd9p5CCOvw26EyimsMXJl0ducMd2cNVw7uxbrscvIq6rspOiFERyrqDHyVfoQL+wcRrj3zdqHjzRkZTa3BxNIdBd3yfsI6SQJrx37LKKWwqpHrh0ec1dNXixnnhXK4ol4muRDnTIPRxJc7CxkT50+krnsGM4CBIS03cT7elo9RZigUwq58tqOAXr5ujIzxO+v3mpYYgkYFX++Sm11CnGufbCtoKf1PPvunrxZxAZ6M6+3P59L2ZtckgbVjn2zLJ1zrxui405/RrT2T+gbi4+bEl+kysItzY/WhMiobmrgyqVe3v/fVQ8MorzPy26Gybn9vIYQyDpXUsrOgiisG9zqjSQpPFOjlypjeAfxvd0tPrRDi3KgzmPjvzkIm9g0g+ixK/9tzY2ok1Y1NfLJZJmWzV5LA2qn0gkr+OFLNNUPCu2VQB3Bz1jBlYDC/bdnBzbNvYOLEkVx99XR++GFZp6+rr6/npZeeY+rUyUyePI57772T7OysNtusWvUL119/FZMmjeLaa69g+fL/dUvMwrZ8v6+IYG9XhkVqu/29U6N09PJx5av0U5/MaeXKFcya9VcmTRrFVVdNY8mS9zGZOp7dsLy8jHnzHmXq1AuZPHkc99xzO1lZh7ojfCFEO77bcxQntYopA4K77T1nnBdKZUMTKw+WtH5t//693Hrrjac87h1v9+4/GDs2he3bt3ZbjELYuuV7i6g1mJg57OwnazxR/2BvUqN1vLs+h4bjZiTeuHE9N910HeefP5rrr7+Kn3/+scP32L59K6NGDWv33513zun2mMXpkQTWTn2yrQAfNycuSei+QR1gYpQ7mjWL0PhF8O67H3P55VfxzDNPsG7dug5f8/LLz7N162aeeOIZ3nzzPVxcXLj33jtobGwEID19B//5z8NcdtmVfPDBUi6//Gqee+5J1q9f262xC+tWVmtgU04Fk/sHoe6GkvcTadQqpiWGsjWvkpyyui6337BhHf/5zyNccslUPvhgKXPm3M7HH3/AkiXvtbt9c3MzDz54H3l5uTz99Au88ca7eHp6cdddt1FZKWX3QnS3JlMzP+wtZkycP1qPs5/wzWJ4pJZInTtfHas2qqio4J577qBv335txr3Nmzd2+V719fXMm/fvTm98CeFoms1mPt9RwMAQbxJCfc7Jz5iVHEF5rYHv9xUDsGvXTu677y4SEwezePESrr32BubPf5off1ze7usHDTqPb7/9sc2/hx9+HLVazcyZfzsnMYtTJwmsHSqubmT1oVKmDQrB3VnTre+dvvYnnN08qOp/CVFR0Vx++dVceOFFvP9++xf1AGvW/Mb06ZeTmDiY6OgYZs++jeLiInJyso99fzWxsb2ZNm0GYWHhTJs2g75949m8eUO3xi6s24oDJZjMcNGAoHP2My5NCMFJreLrP7oug//22y8ZO3YCM2ZcRVhYOOPHT+Lqq2eyfPl37W5/6NBBdu/exQMP/JsBAxKIiYnlkUf+Q319ndyMEeIcWJtVTkW9sdtv1KpVKqYmhJBeWEVeRT3Lln1z7GbU3Dbj3qefLunyvV599UUCA8/dOU0IW7T5cAU55fVcNaT724UshoT7MiDUh6XbCzCbzXz66RISEhK56657iYqK5oILJjNz5vW8886idl/v7OyMv39A6z9XVzfeeOMVrrnmOlJS0s5Z3OLUOCkdgCMYNWoYjz46j6+++i8HDuyjV68w7r//32RkHGDJkveora1hzJix3Hffw7i4uAAtTyXfeONVDh48gL9/ABMnns8NN9yCq6srABkZB1m0aCG7d++ioaGB0NBeXH/9jfzlL1P4dvdR1L+/RoM6lSdWl7Nmze94enoyZsw47rjjHpycnHjnnUW8997b7cb7l79M4aGHHmv3e+npO+jTP5GdpfVklNTQJ9CLpKShvPjiszQ3N6NWn3xPRKvVsXLlz0yceAFeXt4sW/Yt3t4+9OoVduz7WnJysti+fStJSUNJT99BVlYml112ZTd8+kJpp7L/p6aOJCt2KvFBXsT6e57V/g9w++2zSUhIpKSk6KT9f1zvAL5Z+h5fPtx+6ZBl///b327Cza3tRFIqlYrq6qp2XxccHMJzzy0gMjKq9WtqtRqz2Ux1dXV3fJRC2LRTPRc89NBjpzQWfrenCK2hmK9f+4J5e07/XAB0OhY6RQ7nh7RIDqbvYPDgpDbjW1LSUF544ZkOxz2ADRvWsmHDOp5//hX+9reru/OjFEIx3XEcmyMG4xc3iUl9A4GzG9O7uqY1RQ5n87g48vLyGDVqTJvv9e0bz5EjhRw9epSQkJBOf+/331+Ms7MLN9xwy9l+hKIbSALbQ1599SX++c+HiIyMZN68x7jvvrsYMCCB559/mdzcwzz++MMMGJDI9OmXk5FxgHvuuZ2bbprDww8/TlHRURYsmE95eRkPPvgo9fX13HPP7YwcOZq33nofsxmWLv2I5557kqHDU/lm1xF8XJ34/tvP+dvfbmLx4g/ZsmUTCxbMZ8CABC688CKuueY6pk2b0W6srq5uHf4eJSXFDE0eyR8GFd/vLeausV4EBARQX19PVVUVWu3JvYv//OdDx0oxL0Cj0eDq6saLLy7E29sbgMsuu5I//kjnzjvnoNFoMJlMXHPNda0nLmH7utr/H3vsYepLvbnrxuu63P/r6uo63P9TUtLw82uZtOyzzz5ud/+fkpDKL3vH8u+br2dEO7OWWvb//v0Htvl6bW0N33zzZYd3Xn19tYwYMarN1/7736UYDAaSk1PP/kMUwg6cylg4ePCQLsfCOXc/wLqMI3j+9gaB48ed0bngqqsu73QsfOCHQ/ywrxjv4mL69Ilv872AgAAaGho6HPf0ej3PPDOPBx74d+tYJ4S9OJvjeG9WLo89/TS9TXU4a0Z3ek17KsdxZ9e0bp5uTF+0maU7CggICKC4uKjN948caanG0uvLO01gKyrK+eqrz7n33vtxc+v4Gln0HCkh7iEXX3wpo0aNITIymgsvvIjq6iruvfdfxMb2Zty4ifTr14+srEwAPv30I9LSRvHXv15HeHgEQ4cO5777HuT777+jtLSUhoZ6rrrqr9x9931ERkYTFRXNddfdgNFo5IfNuymuMRDg5ULv3n2ZNetmIiOjmDHjSuLi+rB79x8AeHh4tCmNOP6fl5dXh79HQ0MDPp7ujIjW8dP+YkzNZpydW+6wGQyN7b4mPz8PPz9/5s9fwOuvv0NyciqPPPKv1hNJRUU55eXl3HbbnSxe/CF33z2Xr776nGXLvu3OP4FQUFf7v1+vaNRVR5nQJ6DL/b++vuP9Py/vzxkHO9r/U6J06Ly9WHek6ZT3/4aGBh54YC6NjY3ceusdp/Q7r127mkWLFnLVVTOJjo7pts9SCFvW1bmgd+++pzQWfrv5ICZjI5fOuPqMzwXQ+Vg4ZXA0+foGqmvrWqs/LLoa9+bPf5KRI0eTmjriXHyMQijqbI7jw07hmJKuIGvbr11e057tcRwWEsQVw+NYm1XOsFGTWLlyBStX/kxTUxMZGQf49NOPADAajZ3+vl9//QVarY4LL7zoHH2i4nTJE9geEh4e0frf7u7uqNVqQkP/rP13dXXDaDQAcPDgAfLzczn//NGt3zebzQAcPpzN0KHDmT79cn78cdmxbfPIyDgIwO+HSgjwjMDX3ZmQ4LZ9QV5eXjQ1tRykH374boeT0VxwwV+4774H2/x8gJ9/XoOrqysGg4GLBgSzJqucrXl61MfiPrHcEqCwsIDnnnuS115bTELCIAAee+xJZs68nM8++4Q77vgHzz33JH37xvPXv14PQJ8+8ej1et544xUuvvjSblnDViirq/2/pkmFjzuE+Lh1uf/37j22w/3/+IlSIiPbritn2f+d1Cqiitay5rOvmfSyhhN3L8v+b6HX67n//nvIycnipZdeIyQktMvf9/vvv+PZZ+cxadIF3HbbnafwCQnhGLoeC11PaSz8ZesfxPbqw81XjD3jcwF0PhaOnzgZV/exNJg1GAyGNt8zdjLu/fDDMg4ePMgHH3zaxachhG06m+O43mjCFWim62va7jiOx4y/EI37WMoCB3PDDbfw1FOP8fjjDxEcHMLVV1/LSy89h6dnxw9uAFas+IGLL74UJydJm6yF/CV6yIk7vUql6jAxc3Z24i9/mdLuLGf+/gGUlpby97/PIjAwiJEjRzNixGgCAm7b96cAACAASURBVAK5+ebr2F1YzaxLQtiz+887xMezDP7Tps1gwoTz2/35np6eALz33icnfS8oKJiyslKuj/XDw1nDLwdKGGIqxcPDo90nV/v378NkMtGvX/82n0WfPvEUFOQBsGfPH0yadGGb1w0YkMD77y+muroaH59zM0Od6Dmd7f+V9UZqDCb6+LSU5XS1/5eUlHD99Ve3u/8fr7P9/+aZ13C7OpbZo2OYeKwHx8Ky/wMcOVLIP/5xO/X1tSxc+Da9e/fp8nf94IN3ePvtN5gx40ruvvs+uQEjxHG6YywsrTVwyzdZXBfqdNbngq7Gwpp1Raxd601paUmb75WWluLu3v649/3331FSUsTUqRe2+Vlz597FX/5ycZsbZELYojM9jnfm63n8p4PMHd+bkbF+XV7Ttn2f0z+OfXzcMJk01Kw5yrI9RSyffSPXXjsLvb4Cf/8A1q37HY1G0+mN6aysTPLz85g48YIuPxfRcySBtUIxMXHk5GS3ucO1e/cuPv74A+bOfYDVq1dRV1fHa6+9jUbTMsvwpk2WGXvNTBsUwp4ubvz6+Pji4+Pb6TbH/3yLxMTBfP/9d7g6qRkV68dvh8poLt9CUtKQdieyCApqmX0xM/MQ8fH9WiI0m8nJyW4trQoMDCIzM6PN67KyDuHr6yvJqwNYn1MOQIh3S4leV/v/5s1rO9n/T01ynzDCwyPYWObE39rZz6GltP3OO+egVqt54413Wycd68zHH3/A22+/wc03z2HWrJtPKyYhRFsdnQteeO1NCL8I9+LdZ30u6GosPD++mVW6aDZv24bZbG69SN++fSuDBp3X7rj3738/0bpMHEBZWRn/9383c//9DzN8eMppxSeErTv+OH5tRzW+AaH09TXz2msLTuGa9tR0dBxrtR7o9XVMT4TffvyaB+b9wguPPUJAQMuN699//42EhEQ8PDw6fO9du3bg7+8vrUBWRnpgrdDMmX9j797dvPrqixw+nMOOHduYN+9Rqqur8fcPQKvVUVdXy2+/reLo0SOsXbua5+Y/BcCAQHdCfc5dg/mUKVPR6yuYP/8pEr3rqdq9il9+/okbb7yxdZuqqkqqqiqBlolwEhISeeqpx0hP38nhwzk8//zTFBUd5fLLrwLgiiuu4dtvv+Kbb76gsLCAFSt+ZMmS97j22hvO2e8hrMfazHKc1Cq07i3rOHa1/+t0J+//84/t/yeW+XVEpVIxuV8Q23L1lNW2/5oXXngWvV7PY489iaurK2VlpZSVlVJeXta6zfH7+qFDGbz11utcfPGlXHLJtNbty8pa+naFEKeno3PB0TI98ZG9iO0VfNbngq6kRetw7p2GXq9n/vynyMnJ5osvlvLzzz8yc+b1rdsdfy4IDAwiPDyi9V9oaMvTnYCAQHS6kyeOE8KeWY7j5196ntU79jLcvYxnn36802va7j6Ok6N0+IVEsGnVMn766XuOHClkyZL3WLHiB26+eU7rdscfxxYHDx4gNrZ3t8Qhuo88gbVCcXG9mT//Zd5++w2+/voLPD29GDlyNLfddhcAEyZMYt++PSxYMJ/6+jrCwiIYPeUaPv/0QyIo6eLdz46fnz/PP/8qL788nx9+WI6zmy+J024lJSUVvb4OgAcfvA+AhQvfQqPR8MwzL7Jo0UIee+xB6uvriI8fwOuvL24t2bjssitwdnbm888/YeHCBYSE9GL27P/jssuuOKe/i1Bek6mZ9Tnl+Lo7t/aidrX/X3jhZLZu3dFm/58162Y++uh99u/fe8qTpkzoG8DijbmszizjssS25UONjQ38/vuvNDc3c8stbcsXNRoNq1dvAtru6ytXrsBkMrF8+f9Yvvx/bV4jT2SFOH3tnQsGD08j0zmNa+MDmTA86aSx8EzOBZ1xc9Ywqn80WxrncPDgj9x440yCg0N4+OHHGTp0eOt2x58LhBB/shzHz73yCqrcL9jp7c3Y0WM6vabt7uNYrVJx5YXjeLPkKG+9vQh9RSmxsXE8++xLJCUNbd2uveO4rKy0y4pF0fNUZksBeTtKSmxr7UJLqYAtOtvYH/huH1tyK/hhTirOmp57sP7gsn1szdWz/l/jqalu6LGf252seb8JDDy15Rds7Vi12JanZ87nu3ju0gGM7xNwSq/prr+X2Wzmsne3EK5159UZg876/ZRmzfuxUnryMznVYxV6/ni1p33jg815LFyTzTc3DyfM9+QJlE7HqX4uvxwo4YFl+3jzykSGRpy8ZI69saf9pT3WfKy2x17+Hrd+nk5xjYEvbhjWY3NDHP/ZldYamPLWJq5OCuPucbE98vNtlbXsc50dq1JCbAeqGoyszixlcv+gHk1eASbFB1JRb2TL4Yoe/bnCPmzN1aNWwfDInr8oVKlUjO8dwNZcPdUNTT3+84UQp2/F/mIGhnifdfJ6OkbE+OHqpGbVwdIe+5lC2JPi6ka25VUyuV+QYhMbBni6MDbOn2V7jmJoalYkBtF9JIG1Ayv2l2A0mZkyMLjrjbvZiGgdrk5qftlX3OM/W9i+bXl64oO88HJVppthfJ8AmprNrM0u63pjIYSiDpfXcbCklgv6BXa9cTfycNEwIsaPVRmlNHdctCaE6MCKAyWYgQv7Bykax/TEECobmvjtkNyMsnWSwNqB5XuL6B3gSXxQ5+tYnQtuzhpSonSs3F9EJ9XoQpykwWjijyPVDFOwJG9gqDeBXi78miEJrBDW7rdDLcfphFNsN+hO43r7U1prYN9R5UtKhbA1P+0rZkCIN5G6nqucaE9ylI5gb1eW7y1SNA5x9iSBtXHZZXXsPlLNlIHBipVljInzo0DfwKHSWkV+vrBN6YVVNDWbGapA+bCFWqViXO8A1meX02A0df0CIYRiVh8qpV+QFyHncKb9joyI8UOtgt+zynv8Zwthy/Iq6tlfXMMF8T1bOdEetUrFX/oHsTGngtKaxq5fIKyWJLA2btmeIjQqmKxgWcaoWH9UKlh9SJ5iiVO3LU+PRgWDw5Rd63d8H38am5rZmCN93EJYq9JaA7uPVDO2t78iP1/r7sx5Yb6syZRxTojTYSnXndC35ysn2nPxwGCazfCDtL7ZNElgbZip2cwP+4oYEeOHv6eLYnH4e7owOFzL7zKwi9OwNbeSASHeeLoou5pXUrgWXzcnfpWeGCGs1prMMsygWAILMDrWj4ySWo5U2eaM+0Io4deMlsqJUAUqJ9oT7edBQqg3y/dK65stkwTWhm06XEFJjUGRyZtONLFfEPuKaiiulpIM0bU6g4m9RdVWsSSFk1rF6Dh/fs8sw2iSmQmFsEa/Z5bRy8eV3gGeisUwOq4leV6TKWXEQpyK4upG/jhSfcrL5PWUiwYEk1lax8FiaX2zVZLA2rBle4rwdXNiVKxyd6QtJvZrKWFekyVPYUXX0gsrMTWbFZ3A6XjjegdQ02hiR36l0qEIIU5QZzCx+XAFY3sHKDbXA7Q8uYnUuUsZsRCnyDLxmrUlsBfEB+KsUbFMJnOyWZLA2qiaxiZWHyrlgn5BuDgp/2eMC/QkQusmfbDilKQXVKFWwaBeyva/WqREaXF1UksZvBBWaGNOOQaTWdHyYYvRsf5sy9dTa5C1o4Xoyq+HSon2cyfG30PpUNrwdXdmdKw/P+0rpkkqr2yS8pmPOCO/ZpRiMJm5aICya2pZqFQtZZhb82RgF137o7CKuABPPFw0SocC/Lkc1O+ZZdITI4SVWZ1Zhq+bE+eF+SodCqPj/DCazGySSd+E6FRlvZEdeXrG9baup68WFw0IpqLeyHo5lm2SJLA26qf9xYT5ujEwxFvpUFqN7e2P0WSW2VxFp0zNZvYcrSbRSp6+WoyJ8+NIVaMsByWEFWlqNrM2q5xRsX44qZUrH7Y4L8wXHzcnqdYQogsbcyowmWFMnPKVE+0ZGaPD182JFftlNmJbJAmsDSqtNbAlV8+F/QIV7Qc6UWIvX7xdnVgn6+SJTmSX1VFrMFldAjsq1h8VyIWpEFYkvaCSqoYmxljJUxwntYoRMX6szSrH1CzVGkJ0ZF12OVp3ZwZY0YOW4zlp1IzvE8CaTFkH3hZJAmuDfjlQQrMZLlRw7df2OKlVpETpWJ9TIWWYokO7jlQBMCjUuhJYf08XEkK9pY9bCCuyLqscZ42K1Cid0qG0Gh3rR2VDE3uPVisdihBWydRsZkNOBWnROjRWUDnRkUnxgdQZTazPlgcvtkYSWBv00/5i+gR6Euuv3HICHRkZq6Os1iBTk4sO7SqsQuvuTLjWOtaEO96YOH9ZDkoIK7I+p5zBYb5W0y8PkBylQwVsyJGLXiHas6+oGn29kZExfkqH0qmhEVp07s78fEDWgbc1ksDamHx9PbuPVDO5n3U9fbVIi245Wa2XgV104I/CKgaFeltV+bvFmGOznK6V5aCEUFxRdSOZpXWkRVvP01cArbszA0O9Zb4HITqwLqsctQpSrOzYPZGTWsWEvgGszSqjXsqIbYoksDZmxf4SAC7oF6hwJO3z93ShX5CXlGOIdunrjeRW1Ftd/6tFjJ8H4Vo3fs+U/VcIpW04No6MsMKnOGnROvYcbXnKJIRoa112OQmhPmjdnZUOpUvnxwfS0NQs6zvbGElgbYjZbObHfcUkhfkQ4mN95ZcWI2J0/FFYRVWDDOyird2W/lcrTWBVKhVj4vzZkltBnUHuxgqhpPU5FQR5uRBrZWtIQku1UbMZNh+Wp7BCHK+s1sC+ohqrLx+2GBzmi7+nCz8fKFE6FHEaJIG1IZmldWSX13G+lZYPW4yI8cNkhk2H9UqHIqzMvqM1qID+wdY5KyG09MEaTGY2yoWpEIppMjWz+XAFI2L8rLLdYECINz5uTmyQMmIh2tia23Ltl2rl5cMWGrWKSX0DWJ9dTq2hSelwxCmSBNaGrDxYggqY0Mc6lhPoSEKoDz5uTlJGLE5yoLiGKD93q5qQ5USyzqMQytt1pIpag4k0K32Kozk26/5GmXVfiDa25OrxdnUiPshL6VBO2fnxgRhMZhn3bYgksDZkZUYpSeEtpQ7WzDKwb8ipoFkGdnGc/cU1Vj+oOalVjIzxY21mmazzKIRCNmRXoFGrSI7UKh1Kh9KidZTWGjhUKrPuCwEtrW6bcysYGuFr1cvnnGhQLx8CvVz4NUMSWFshCayNyCqrJbusjol9rXPyphONiLEsp1OjdCjCSlTUGSiqbqSfFZcPW4yJ86eyoYk/CquUDkUIh7Q+u5zEXj54uTopHUqHLCWSG7KljFgIgILKBo5UNTI80jbKhy3Ux+a/2JhTTmNTs9LhiFMgCayNWHmwFBUwvo+/0qGcktbldGRgF8ccOHYzIz7I+tYvPlFqtA4ntYrVUk4kRI8rrWnkYEktI6y8hy7Qy5U+gZ6yHqwQx2w+1v9qzZUTHRkT50+9sZktuXLdagskgbURqw6Wcl6YD4FerkqHckr8PV2ID/Jik0yEI47ZX2RJYK27hBjAy9WJYZFa6YcRQgGWCdSstf/1eGnROnYWVMnkL0IAWw7rCfRyIcrPXelQTtuwCC2eLhpWH5Jx3xZIAmsDcsrrOFRaazPlwxYpUVp2FVbJciQCaHkC28vXDR83618XDlruxuZW1JNTXqd0KEI4lK25erTuzvQNtP5qjbRoP5qazWzNrVQ6FCEU1Ww2szVPz/BIrVXOHN4VFyc1adE61mSVy/wtNkASWBuw6mApAOOtfPbhEyVH6WhqNrM9X5bTES0TOPWzgaevFqNjW57+yOLmQvQcs9nMtrxKhkb42sRF8HlhPrg7q9koZcTCwR0qqUVfb2S4DZYPW4zp7U9ZrYE9R6qVDkV0QRJYG7DyYAmDQn0I9raN8mGLwWG+uDqpZT1YQU1jE/n6BvoF204CG+LjRnyQl5QTCdGDCqsaOFrdyJBw27gIdtaoGR6pY70spyMc3JZj/a+2NoHT8UbG+KFRIfNf2ABJYK1cvr6egyW1TOxrW09fAVyd1CSF+UofrDhuAifbSWABxsT5sauwioo6g9KhCOEQth0rxR0a4atwJKcuNVpHYWUDefoGpUMRQjFbcvVE6txt7mHL8XzcnEmK0PK73Li2epLAWjnLJDJje9vG7MMnSo7Skl1WR1F1o9KhCAXZagI7Ni4AM7A2S8oDhegJ2/L16NydifX3UDqUU5Ya1fLEabPcrBUOqsnUzI78SpsuH7YYG+dPdnkduRX1SociOiEJrJVbk1lGXIAH4Vrbm9EN/lwnTwZ2x7a/qIZALxf8PV2UDuW09A3yJMjLRWYjFqIH2Fr/q0W41o1QH1epNhIOa8/RauqMJvtIYI89MFp9qFThSERnJIG1YlUNRnbkVzImzjafvgL0DvDEz8NZBnYHt7+4xuaevgKoWhc3r6DBKLNpC3EuFVQ2UFTdyJAI27oIVqlUJEfp2Jqnx9QsfbDC8ewsqAIgKdx2Sv87EurjRp9AT9ZnS+WVNZME1oqtz67AZMamE1jLwL75sF6mJXdQ9UYTh8vrbGoG4uON6e1PQ1MzW/NkMjIhzqVtx46xYTaWwAIkR2qpaTSxv0hmLxWOZ0d+JdF+7vh52FaVVUdkfWfrJwmsFfs9sww/D2cGhHgrHcpZSY3SUVFvJKOkVulQhAIySmppNmNTMxAfb2h4y+LmUkYsxLm1La8SPw9nov1sr2XGUjops+4LR2NqNpNeWMngMNt/+moxIsayvrMcz9ZKElgrZTQ1sz67nNGx/qhtqBeoPclRxwb2HCkjdkT7i2xzAieL1sXNM2VxcyHOlZb+Vz1DwrU21f9qofNwIT7Ii825Ms4Jx5JZWktNo8kuyoctEnv54OGsYYNct1otSWCt1Pb8SmoNJkbbcPmwRaCXK7H+HtIH66AOFtegdXe26an1R8f5U1prYN9RKQ8U4lzI1zdQXGOwqeVzTpQcqWVXYRX10i8vHMiO/Jalr+wpgXXWqEmO0rI+u1zWd7ZSksBaqTWZZbg6qUmJsr1eoPakRuvYWVApE+E4oP3FNfQL8rLJpyoWlsXNpYxYiHPDlvtfLZKjtBhN5tYLeiEcwc6CSoK9XQn1cVM6lG6VFq3jSFUjh8tlOR1rJAmsFTKbzazJLCM5Uoubs0bpcLpFcpQOg8nMzgIZ2B2JoamZzNJa+tpo+bCFr7sz54X58numzEooxLmwLb+l/zXKBvtfLQaH+eKiUbFZ+mCFgzCbzewoqGJwmI/SoXS7tBg/ANbnyLhvjSSBtUKZpXUUVjXaRfmwxZBwX5w1KpngwsFkldXS1Gy22Qmcjje2tz+HSmspqJS7sUJ0J7PZzPY8PUMjbLP/1cLNWUNimK/0wQqHka9voKzWwBA7Kh+2CPVxI8bPgw3ZcjxbI0lgrZBl7alRsX4KR9J93J01nNfLR/pgHYxlAidbXULneJblrNbIU1ghulXesf7XYTbc/2qRHKklo6SWslqD0qEIcc5ZyuUH22ECC5AWo2N7vl7a36yQJLBWaENOOX0CPQn0st1Jb9qTHKWTgd3B7C+uwdNFQ5jW9ntjwrXuxPh7sFr6YIXoVpb+1yE23P9qkRylA5DlN4RD2FFQia+bEzF+HkqHck6kRbe0v23Lk/Y3ayMJrJWpNTSxs6CKtGid0qF0u5RjA7uUVzmOA8U1xAd52fxSUBZj4vzZkaenqsGodChC2I1teXr8PV2I0tlu/6tFvyAvfNycZJwTDmFnQSVJ4b42XfrfmaRwLa5O6tbKSGE9JIG1Mltz9TQ1mxkRYz/lwxbxQV74ujnJerAOoqnZTEZJrV30v1qMifPHZEZ6YoToJi3rv1Yy1E4ugjVqFcMitGw6rJflN4RdK6lpJF/fwOAw+ywfBnB1UjM0wpeN0v5mdSSBtTIbcirwcNaQ2Mv+ZnTTqFUMj9TJwO4gDpfX0djUTLwd9L9aJIR64+fhLMvpCNFNcivqKa01MDTS9suHLZKjtBRVN5JbIRO+Cftl7/2vFqnRfuRW1FNY2aB0KOI4ksBaEbPZzIbscoZHanHW2OefJiVKS2mtgayyOqVDEefYgeJjEzjZ0RNYtUrF6Fh/1mWXYzQ1Kx2OEDZv27GL4KF2dBH8Z7uM9MEK+7WzoAp3Z7Vd3aRuT+qx41kmIbUu9pkl2ajD5fUUVjWSFmN//a8WKdFyInAU+4tqcHVSE6Wzr8kdRsf5U2swsT1fJnUQ4mxtz9MT4OlCpB30v1qE+brRy8eVzTLOCTu2s6CSxF4+OKltv/S/M9F+7gR5uch1q5WRBNaKWBZLTou2v/5Xi1AfNyJ17nIicAD7i2voG+iJxs4Gt5Solkkd1kgZsRBnxWw2szWvkqER9tH/aqFSqRgepWNrXsucFkLYm6oGI4dKau26/9VCpVKRGq1j82E9JjmerYYksFZkQ04F0X7u9PK1/SVHOpMSpWN7XiWGJinBtFfNZjMHj81AbG/cnDWkROn4PbNMermFOAuHK+opqzUw1A6WzzlRcqSWmkYT+45WKx2KEN0uvaAKM5BkR6X/nUmJ0lHd2MS+IjmerYUksFai3mBie57erp++WqRE6WhoauaPI1VKhyLOkXx9A7UGk131vx5vTJwfR6oaySipVToUIWzW9mPrv9pnAivLxgn7tbOgEie1ioEh3kqH0iOSI3WogI2yiobVkATWSmzOKcdgMtt1/6vF0AhfNCo5Ediz/cfuUvYLss/BbVSsP2oVrMwoVToUIWzW1rxKAr1ciNDaX9WR1sOZ+CAvNh+WiZyE/dmRX8mAEG/cnDVKh9IjtB7O9Av2kvY3KyIJrJX4PaMUVyc1Q8Lt7070ibxcnUgI9ZETgR07UFyLk1pFbIB9TeBk4e/pwrAILSv2F0sZsRBnoGX9Vz1DI7R21f96vJQoLbsKq6g3mpQORYhu02A0sbeoxmHKhy1So3X8UVhFTWOT0qEIJIG1Gr9nlDA0whdXJ8f4k6RE69hfVIO+3qh0KOIcOFBcTe8AT7tdDgrgwn5B5Osb2Cs9bkKctsPl9ZTXGe1q+ZwTJUfqaGo2y4zlwq7sPlKNqdlMkgNM4HS8lCgdJjNsy5OqCmtgv1eXNiRfX09OWZ1D9L9apETpMANbZJ08u2M2m9lfZJ8TOB1vfJ8AnDUqftpfonQoQticbfn22/9qcV6YDy4alSynI+zKjvxKVEBiLx+lQ+lRib18cHdWS/ublZAE1gpsOHYwpEXbf/+rxYAQb7xcNVJGbIeKqhupbGgi3k4ncLLwdnNiZIwfKw6UyNT6QpymrbmVBHm5EG6H/a8Wbs4azgvzlT5YYVe2F1TSN8gLbzcnpUPpUc4aNUMjtHLdaiUkgbUCG7LLCde529VC7l1xUqsYFqFlU06F9BDamf1FNQD0s/MnsACT+wdRVmuQkiIhToPZbGZ7vn33v1okR2o5VFpLaa1B6VCEOGtGUzN/FFY5XP+rRWqUjjx9A/n6eqVDcXiSwCrMaGpma56eMX0C7H4gP1FKlI6j1Y3kVsiJwJ7sL65BrYI+gZ5Kh3LOjYzxw9NFw0/7i5UORQibkWPpf42w/4vg5KiWyqqt0i4j7MC+ohoam5odNoFNOVYpKW0BypMEVmHpBVXUG5sZ0ztQ6VB6XMqxgX2TlFfZlf1FNUT7eTjE9PpuzhrG9QlgVUYpjU3NSocjhE3Yasfrv54oPsgLXzcnKTsUdmHHsQnJksIcq//VIkrnToi3KxvlulVxksAqbENOORq1ipRYx5nAySJc60YvXzcZ2O2I2WxmX1E1/R1kcXOAyf0CqWk0sS67XOlQhLAJ2/P0BHu7EuZrv/2vFhq1imGRWjYflnYZYft25FcS4+eBzsNF6VAUoVKpSInWsSW3giaZ+0JRksAqbENOBYPDfPBydaxmeDh2IojSsi1PT5NJnl7Zg+IaA+V1Rvo7QP+rxbBIHf6eLizfU6R0KEJYvZb1XysZGuHrMG0zyVE6imsMHC6Xdhlhu0zNZnYWVDps+bBFapSOmkaTLKGnMElgFVRS00hGSa1DLZ9zotQoHbUGE3vkRGAX9he1/B372fkMxMdzUqu4eEAw67LKKK1pVDocIaxadnkdFfVGhobbf/mwRXJky++6OVeqjYTtOlRSS63B5PAJ7PBILSpgkyynoyhJYBVkWUsq1YGWzznRsEgtahWyrpad2FfUMoGTva8Be6JLEoIxmWH5XpnMSYjObM1t6aEbGuk4F8HhWnd6+brJcjrCpm0vONb/6uAJrK+7MwNCvNko7W+KkgRWQRtzKvD3dKGvA8zW2hEfN2f6B3vLRE52Yn9RDTH+jjGB0/Gi/TwYHObD/3YflT43ITqxPb+l/7WXj/33vx4vJUrL1jy99M0Jm7Ujv5IwXzeCvV2VDkVxKdE69hyporqhSelQHJYksAoxNZvZdLiC1Gidw/QBdSQlWseeo3IisHWWCZz6BTvOBE7HuzQhhNyKenYWVCkdihBWqdlsZmuuvqUEz8HGveTIlnYZ6ZsTtshsNrMjX/pfLVKjdJjMf86oLnqeJLAK2VdUTWVDE2lRjls+bJESpaVZTgQ2zxEncDrepPhAPF00fLXriNKhCGGVMkpqqWxoYnik4/S/Wgw71jcn60cKW5RdXoe+3igJ7DGDQr3xcNbIKhoKkgRWIRtyKlDx51qojmxQqA/uzmo5Edg4R5zA6XjuzhqmDAzmlwMllNYalA5HCKuzJbflJuUwB1j/9URad2f6BXtJAits0vY8y/qvksACOGnUDIvUyvwtCpIEViEbsisYEOKN1sNZ6VAU56xRMzRCKwmsjXPUCZyOd2VSGE3NZr5Ol6ewQpxoa66eKJ07QQ7aQzc8UseuI9XUGUxKhyLEadmS29K7Hq51rN71zqRE6SiobCBfL8tjKUESWAVUNRjZc7TKoWcfPlFKlI58fQMFlXIisFWOOoHT8SJ17oyM8eOL9EKMsraxEK2aTM3syK9kmAOWD1ukRGkxNZvZni/tMsJ2NJvNbMtzzN71zliu4TfIU1hFSAKrgM2H9TSba16TGQAAIABJREFUIU0S2FaWUmqZjdg2OfoETse7MqkX5XVGfjlYonQoQliNPUerqTOaWtdEdUTnhfni6qSWskNhUw4W1zhs73pnIrRu9PJxlfVgFSIJrALWZpXh4+bEwFAfpUOxGtF+7gR5uciJwEY5+gROx0uN1hHt586SLfmypI4Qx1gm6RvigP2vFq5OaoZFaFmfXa50KEKcMkvvuiSwbalUKlKidS3LY0nFVY+TBLaHmZrNrM0qZ2SMH05qKcWwUKlUpEX7selwhZwIbJBlAqf+IfIEVq1Scd3wCDJKaqW0SIhjtuTq6Rvoidbdsed9GBHjR56+gdwKaZcRtmFzrp4YPw8CvRyzd70zqVEty2PtPiLLY/U0SWB72B+FVVQ2NDE6zl/pUKzOyFg/ag0m0gtlHU1bY5nAqW+gp9KhWIW/9A8iyMuF9zfnKR2KEIprMJrYVVjF8EhpmxkR0/IZyFNYYQsMTc3szK+Up68dGBapRa2CjTIJaY+TBLaHrckqQ6NWSf9rO5KjtDhrVKzJlIHd1uw5Wk2sv6dDT+B0PGeNmpnDwtmRX0l6QaXS4QihqPTCKowms1wEA+Fad6J07qyTBFbYgD+OVNHQ1CzHbgd83JwZGOItq2gowEnpABzN75llDA504Y1XnmPt2t8xGAykpY3koYceRKNxP2n7pqYmbrnlenr37stDDz3W+vUlS95n6dIluLq6cdttdzJp0oWt3/vss4/Jzs7i/vsf6TSWJ598jOLiYl5++fWTvvfMM0+Qn5/HwoVvAXD55Zdw9OifS4Oo1Wo8PDwYODCROXNuZ/jwwe1u5+zsTEBAEGPHjufGG2/Bw6PjJ3SeLk4MCfdlbVYZd4+L7TR2YT2azWb2Hq1mQp+ALretra3h9ddfabPv33HHP9Dp/E7atqN9f/Hit3nvvXetZt/v06dvu9s5Ozvj5uLD47nD+OjpBzvd94WwZ1tz9WjUKgaH+3TLOaC7xr+KilKef37hSd/rznNAe+PfiBg/vkwvpMFokpt+wqptydWjVsHQE3rXrek47mwsP3q0kAUL3gC6/zi2SInS8e6mXCrrjfg6eItET5IEtgflVdSTU16P074P2VhxlAcffJTAwCDefvt1brxxFm+/vQQXF5c2r1m8+E0yMg7Su3ff1q9lZWXy4Yfv8tJLC9Hr9Tz22IMkJ6fh49NycbB06ce8+ea73R7/zJl/48orrwFaZp0tKytjwYLnuOee2/nxx58A1UnbNTQ0sH//XhYuXMCePbt45ZVFODt3fICPivXnhV8zyauoJ0J3ckIvrE9uRT1VDU0khHbd//rII/dz+HBOm33/jjvm8O67H53yvv/WW4t48UXr2fc/++zr1gHtxH3/reVr+OXzxfz99tt4d9HiTvd9IezV5lw9A0O88XRx4p77z/4cYG3jX2fngPbGvxExOj7dXsC2vEpGxp58wS+EtdiSq6dfsDfebm3The4Yy239OLZIjdaxeGMuW/P0TOwb2O0xi/ZJCXEPWpNVhkpfQN7+ndx//yOkpKQRGxvHI488QUlJCStXrmiz/a5dO1m+/H/ExfVu8/Xs7CxiY+NISEhk1KgxeHh4UlDQ0mv3ySdLmDBhEsHBId0ev7u7O/7+Afj7BxAQEEh8fD/+7//upqKinE2bNra7XVhYOBMnXsAzz7zI7t1/sHz5t53+jFHHBvO1Ul5lM/Ycm7wgoYtZtTMyDrB588aT9v3S0tPb9/v06WNV+/62bVva3S4sLJwHbrgS7wlzyD64l2XLOt/3hbBHFXUG9h2tJi1a123nAGsb/zo7B7Q3/iX9f3v3Hd9U1cYB/JfdprtNJ12UbtrSgi1LtmwZAooLJwj6AqIiimxBQBwICiIoyBIUBzKUvfcqpRQK3XvvNmmz7vtHoFKSlrY0uUn7fD/kA7m5CU9Ozr3nnHvPcLeFGZ9L3YiJUauSKxGXW6G19BUdx3XLcs3FOR4tj2Vg1IA1oFNJRXDhaiYoCgsLr90uFovh6emF6OirtdukUimWLFmAGTNmwta27hVaNzc3ZGamo7i4CCkpyaisrICTkzOKi4uwb99uTJjwhmG+EAAeT9P96eErbg8LCAhEWFg4jhw51OB+7rbm8LY3x5mkohaLkehXbE45LIQ8eNuLG9wvI0NTOD2c9z08PJqU99PS0owq7wsE9ed9MwEPk59+EmqH9vhz3z+GCo0Qo3EhrQQMgO7t7VvsHGBs5V9D5wBAu/wT8bl4wlOznA4ttUWMVXRmGVRq7bHrdBzXrcfyeVxEetrifGoJHc8GRA1YAymvVuB6Zhk6+3oCAAoK8mtfU6lUyMvLRWnpf1dvVq/+CkFBwRgwYJDWZwUFdUS/fk9h9OiheP31F/Hmm5Ph4CDBpk0/YuTIMbC1Ncxg+6ysTKxb9x0cHCQID4945P4+Pr5ITk565H5P+jjgWmYZquTKlgiT6FlcTgWCXazAe8SyUBKJZozsw3k/Pz+/SXl/0KDBRpX3Q0PDGtx3RIgLLBzdkZaaBJWaCjfStpxPKYGtuQBBzpYtdg4wtvLvUecAQLv869neHlll1Uij5XSIkbqQWgIRn4swt7q9q+g41q7H9vJxQF5FDe7mV+kjTKIDjYE1kPMpJVAxwMg+UUg44I0vvliGefM+hZWVJX766QeUlJRAoVAAAM6cOYXz589i69Zf6/28mTNnY/LkqeDz+TA3N0dWViZOnTqOHTv+wL59f2Pr1k2wtrbGRx/Ng6+vX72fc/36VQwc2Etru1wuR2hopzrbNm/+Cdu3bwagGZSvVCrh5xeAzz77ApaWligtlTaYBlZWVqiqqmxwH0DTjXjblUxcTCtt1MRAhD3VChUSCirxSpTHI/cNCuoILy/tvF9a2rS8P3/+Arz++hSjyfsWFpYNfm8+l4MuHVxx+o4MB+PzMSzYucH9CWkt1AyDC6kl6OZtBy6H02LngJYq/y5fvmyQcwCgXf71aK+5I3UupfiRvVcIYcO5lGJ08bDRmmjM2I7jhsryiIjOdba19HF8X68O9uBygBOJhQhwfvTnkMdHDVgDOZ5YCHuxAJ3c7bF06Zf49NN5GD16CAQCAQYOHIJevXqBYbgoKSnB558vwezZ82FtbdPgZ1pZ/Tdpzvr1a/HCCy9DKpXiu+9W4uefdyIuLhaLF8/H5s076v2M4OCOmDNnkdb2H35Yg5KSuuNzxox5Fs888ywATZcLGxubJs2sKpVW1Ym5Pp3crGEl4uNMUhE1YI1cfF4lVAzQ0aXh8a+AZiY/XXm/e/ee4PMFrTrvO5sBPJEYa8+kor+fhGYeJW1CfF4lSmSK2mXjjO0cEBoaho8/nq+13RDln5uNGXwcxDiVVIQXu7g3+nMIMYT0EhkySqvxfGftvGlsx3FDZXlFRd1l7PRVj7UTC9HJzRonk4owuad3oz+PNB81YA1AKlfhTHIxRoa4gMflwMvLGz/9tBVlZaUQCAQQiy0wadIEdO4chQsXzqKkpBgLFsyufb9cLgcAnDhxFIcPn9b6/Lt34xEbG4NPPlmACxfOwcPDCy4uLrCzs8WCBbNRVVVZ75UlodAM7u7ad88sLCy0CnArK2ud+zbWnTvx8PcPfOR+fB4X3b3tcDalGCo188iuqYQ9sTmaMd2NmYEYgM68/8YbLyEyslurzvt372ryfkxFDTZfyqACjrQJ51OLwQHqrHtuTOcAMzMRq+VfX18H/HwpA6UyBWxp+Q1iRM4ka+Yh6dHeTufrxnQcN1SWP9yA1Wc9to+vBN+cTEZWmQztbGgVDX2jBqwBnEkuQo1SjQH+EkilVZg16z28//4s+PhoZmXLycnGnTt3MGXKdAQFBWt1XfrsswWwt3fA229P1/n569Z9h9dfnwSRSAQOh1M7iFyp1IwhVRvBuLuEhLu4efMGPvlkQaP27+snwaE7BYjJLkNnd1pA21jdzKmAm7UIDhYNT34AoN68n5iYgP/9b0abyPsuakdsvZKJESEucLMxYzs0QvTqXEoJAp0tYSfWnB/oHFC3/OvjK8HGixk4m1yM4R1paAExHpqu7eZwt9VuiNFxrLse28fXAd+cTMbJROpVYQjUgDWAI3cL4WAhRHg7G/C4HKhUKqxa9TVmzJgJmUyKZcs+Rdeu3dClSyQAaHVnEArNIBZb6LxqdPXqZeTkZGPo0KcBAP7+gUhJScK1a1dw69ZNeHu3b1S33ZYkk8lQVFQIQLN+VmxsDNat+w6dOkVg0KChjfqMnu3tIeJzcexuITVgjRTDMLieVYauXrqv0D5MLLbQmfe7dIlsM3m/s1SJk4lFWHUyGZ+PDDZobIQYUlGVHLHZ5ZjUw6t2G50D6pZ/Qc6WcLIU4kRiITVgidGQylW4llmG58Lb6XydjmPd9Vh3W3P4SiyoAWsg1IDVM6lchXMpxRh1r/swACxatBQrV67AlCmvQyQyQ58+/TF79izc62HRJOvWfYuJEzUT2gCAi4sL3nlnOubP/xg2NraYO1d7XIC+bd++uXaQvFhsARcXFzzzzDiMH/9S7XTljyIW8tDNyw7HEwrxfr8O4HKoG7GxySitRrFUgfB2jx7/ep+uvP/229Oa9f+bYt53tuLh9a6e+P5sKi6nlyDSs3GNf0JMzemkIjAA+nRwqLO9rZ8DHsThcNC7gwP2xeWhWqGisfHEKFxOL4VCxaCnT/3lEx3HuvXxdcCmi+kolSpgK6ZhAfrEYRpYtKigoMKQsTw2W1vxI2fCNbRD8fmYsz8e68d3QoR7/YPZjTH2ptBH/P/cysOCf+/gpxfCtaZxb0nGnPaOjo276sjGsbrnZi4WH7yLX1/rAh+Hxk+C8LiM+fdqjBqlGs/9fAVmfC62v9IF/BYa423q6aIPhkyTxh6rgOGPVzbyxnt/3URyYRV2T4wCx0gvQBrDMXMxtQRT/4jFV6M7ovdDjX22GEO66JMxH6u6GPr3WHLoLo7cKcDhd7pDwDPt1TYNnXbxeRWYsC0a8wf7Y0SIi8H+35ZmLOeAho5V086ZJuDA7Xw4WgrRqQl3qYhGLx8H8LkcHLtbyHYoRIfrmWWwMePTEhBNJOJz8X5fHyQXSfH79Wy2wyGkxUnlKlxKK0FvX4nRNl6NRWcPG1iKeDiWQOUcYZ9KzeBkYhGe9LE3+cYrGwKcLOFsJcLJxCK2Q2n1KHfqUbFUjnOpJRga5ERdYJvByoyPKC9bHEsoQAMdBQhLYrLL0amdDeXtZujdwQHdvOzww7lUFFY1Y+wAIUbsQloJ5CoGfX2N446iMRPwuOjjK8HJxELIlWq2wyFt3PWsMpTKFOjrS0sYNgeHw0FfXwdcSCtBlVzJdjitGjVg9ejA7Xyo1AyGBdPkDM01MMAROeU1uJFdznYo5AGFVXKkl8iaNP6V/IfD4eDDAb6QK9VYeTyJ7XAIaVEnEwthY8ZHp3YNrwFJNAYFOKKyRoXzqSVsh0LauBOJRRDyOOjR3p7tUEzWU/6OqFGq6S6snlEDVo/2x+UhyNkSHSSGGx/Y2vTzk0DE5+KfW/lsh0IecCNLs7ZaOFVQm83TzhyvdfXEoTsFuJBa/Og3EGIC5Eo1TicV40kf+xYb393aRXnawsaMj8N3qJwj7GEYBscTCtHN2x5iIU0o1lxh7azhbCXCofgCtkNp1agBqycJBZW4W1CF4XT39bFYCPno6+uAI3cLqHuVEYnOKoeIz0Wgs+6FxUnjvBrpAS87cyw/kohqhYrtcAh5bOdTS1BRo8TAQCe2QzEZfB4X/fwkOJ1UTOcBwprbeZXIq6hBPz/q+v84uBwOBgU44kJaCUplCrbDabWoAasn++LywONyMCjQke1QTN6wYGeUVytxNoXuUhmLaxmlCHG1okkeHpOQz8XHT/khq6wamy6msx0OIY/tUHw+bMz46OpJ63c3xcAAR0gVmmX3CGHDsYRC8DiaCTTJ4xkc6ASVmqHJ2fSIap96UK1QYX9cHvp0cICdWMh2OCYvyssO9mIB/r1N3auMQalUgbsFVYikCmqLeMLTFsODnbDlciaSi6rYDoeQZpMpVDiVVIQB/o7g08WtJunsYUvlHGENwzA4FJ+PSC872JjT+qWPy9/JAl525jgUT8ezvlAJoweH7xSgrFqJZ8Pd2A6lVeBzORgS5IQzyUUokdKMrWy7klEKAIj0rH+Rc9I07/bxgYWQh+WHE6CmGbeJiTqdVIRqpZp6HjXDf+VcMUql1O2QGNaN7HLklNdgaBB1/W8JHA4Hg4OccC2jDDnl1WyH0ypRA1YPfo/JQXt7Mbp40AQ3LWVUqAsUKgZ7b+axHUqbdzm9FBZCHoJdGr8YPGmYnViIab3bIzqrHPsojxMTdSi+AI6WQkS4U9nXHE93dIZSzeAg3bUhBnbgdj5EfC760NJXLeb+HDj74qhM1wdqwLawuNwK3MqtwLhwV1rAvQX5OFggwt0Gf97IoTtULLuSUYoIdxuaYbSFjQhxQXg7a6w+lUw9DYjJKZHKcTalGAMDHGlt6Gbyc7REoJMl9lKFlxiQUqXGkbuF6OXjAAshn+1wWg03GzNEetpi381cqrfqATVgW9iu69kwF3Bp7Vc9GNfJFVll1bhAa+WxJre8GuklMhr/qgdcDgcfP+WHSrkKq06lsB0OIU3yz618KNUMRoa4sB2KSRsR4ow7+ZW4m1/JdiikjbiYXopSmQJDqPtwixsZ4oLs8hpcSS9lO5RWhxqwLSi/ogYHb+djREcXWIroKlZL6+cngZ25AH/G5LAdSpt1Of3++FdqwOpDB4kFXol0x/64PJqNlJgMhmHw981chLha0brnj2lQoBP4XA7dhSUG8++tPFib8dHdm+a1aGl9/SSwNuNjz81ctkNpdagB24J2XsuCmmHw4hPt2A6lVRLwuBgV6oLTyUXILJWxHU6bdDm9FHbmAqqk6tGb3bzQ3kGMzw7dRUW1ku1wCHmkmzkVSCmS0t3XFmBrLkA/Pwn2x+VBRmvCEj0rlSlwPKEQQwKdIORTk6ClifhcDAl0wvGEQpqcrYVRbm0hlTVK/HkjB0/5O6KdjTnb4bRaz0W4gcflYPuVTLZDaXPUDIMLqSWI8rKlMW56JOJzsWBIAIqq5Fh5IontcAh5pL9v5sKMz8XAAJp9uCWMj3BDRY2SltQhevfPrTzIVQxGh9HFJ30Z08kVchWDv2Kp92BLogZsC/nrRg6q5CpMiHRnO5RWzdFShGHBztgbl4eiKproxpBu5VagRKbAk7TIud51dLHCK1Ee2BuXhzPJRWyHQ0i9KmuUOBxfgKcCHGnoTAsJc7OGv6MFdkVng6HJX4ieMAyD3bG56OhiBT9HS7bDabU6SCzQ1csWu65nQ6lSsx1Oq0EN2BYgU6iw7UomIj1tEehMS4vo24Qn3CFXqvFrdBbbobQpZ5KLweWAxskYyMRuXuggEeOzQwkor6auR8Q47bmZC6lCheciaN3zlsLhcPBchBsSC6twLbOM7XBIK3UjuxwpRVKMDqW7r/r2fOd2KKiU4+jdQrZDaTWoAdsCfr+ejWKpAm9192I7lDbBy16M/v4S7LqeTWMEDehMcjE6uVnDxlzAdihtgvBeV+ISqRxfHaeuxMT4qNQMfr2WhfB21giii7ctanCgE2zM+Nh5jS7UEv34KzYXYgEPgwJp9mF969HeHp525thBx3OLoQbsY5LKVdhyORPdvOwQTou3G8zrXT1RWaPC1isZbIfSJuRX1OBOfiV1HzawIGcrvNbVE//cyseheBoPR4zL6aQiZJfX4IXONHFhSzMT8DC2kytOJhYhpUjKdjiklSmskuNQfD6GBTtBLOSxHU6rx+VwMD6iHeJyK3A1g5bUaQnUgH1Mv0VnoVSmwFs96O6rIQU4WWJwoCN2XM1CYWUN2+G0emfuLenS08ee5UjanondPBHqaoWlhxNo9m1iVHZcy4KrtQi9fSVsh9IqvdDZHSI+F5svpbMdCmllNOMxGbzQheZtMZSRIc5wsBDix/NpbIfSKlAD9jGUyhTYcjkTPdvbI9TNmu1w2pzJPbyhUDP46QIV7vp2NrkYbtYi+DiI2Q6lzeHzuFgyPAhcDgdz9sdDQZNAECMQm12Oa5llGB/RDnwuzUquD7ZiAZ4Jc8WB2/nIKqOLV6RlVCtU+ON6Nnp3cICnHa2aYShmAh5eiXTHlYwyXMuku7CPixqwj+HH82mokisxrXd7tkNpkzzszDE61AV/xeYitZi6WOlLZY0SF1KL0dtXAg4tn8MKNxszzB3kh1u5FVh7JpXtcAjBTxfSYWPGxzNhrmyH0qq9/IQ7uFwOtlyipeNIy/jnVh7KqpV46Qm6+2poY8JcYS8WYMN5uvHyuKgB20xpxVL8HpODUaEu6CCxYDucNmtSdy+YC7j46lgSLTegJycTiyBXMRhEazyyqr+/I8Z2csW2K5k4ereA7XBIG3Y7rwJnU4rx0hPuNH5Oz5ysRBgZ4oK/b+YivYTuwpLHo1SpsfVKJoJdrBDejnoOGprmLqwHrqSX4lJaCdvhmDRqwDbTd6dTIOJxMbmHN9uhtGkOFkJM7uGNC2klOJ5I62Xqw6E7+XC1FiHElWYZZdv7fTsgzM0aC/69g1u5FWyHQ9qojRfSYSXi49lwWjrHECZ294KQx8H3Z1LYDoWYuH9u5SOztBoTu3lSjyqWjAt3g5uNGb4+kQSlmm68NBc1YJvhakYpTiQW4dUoDzhYCNkOp80bF+4GP0cLfH08CTKFiu1wWpVSqQIX00oxMMCJCjsjIORz8cWoYNiZCzDz7zjkV9AEZsSwbuVW4ERiEcZHuMFSxGc7nDZBYiHES13cceRuIW7mlLMdDjFRCpUaP11IQ7CLFZ6kCRlZI+JzMaOPD5IKpfgzJoftcEwWNWCbSKVm8M2JZDhbifBiF1o6wBjwuRx8NMAX+RU1+PYUXaFuSccSC6FSMxgUSN2HjYW9WIiVz4SgqkaFmX/HQSqnizbEMBiGwepTybAzF9D4OQN7OdId9mIBVp1MpuEypFn2xeUhu7wGb/XwogvSLOvr64AnPG3xw7lUlEoVbIdjkujyaT3y8/OwevXXuHr1MhhGja5du2PatPdxJF2O+PxKLH06CGYCno73fIWLFy9AJBKhb98BmDp1BszMzFj6FqZrxYrPoFar8fHH8xq1//ZVC+CWXYxdmIi+vg6I8rLTc4St2/30T/UZCS87c/g7ao/znjv3I5w4cbTOti5dorBq1VpDhdmqFBcXYe3a1bh8+QJqamoQHByCqVNnwMfHV2tfX0cLTPCowMafvsCQnwrg6e6Od96egh49+rEQOWktVCoVNmz4Hv/+uxdSqRRdu3bH++9/BHt7zfrPZ1OKcTWjDB/294WliI99+/7Gjh1bkZOTDTe3dnjhhQkYPnxk7eelpCTj229X4ubNGxAKBejTpz/efns6LC0t2fqKrHtUGtfHQsiHc8xmxOWWYG/IKowMcQGguaiwbdvP2L37D5SVlSIgIAgzZsyEn1+AIb4OMWIymQzr1n2LEyeOobq6GtU2nvDvPR49vHXXj6gOqz8P12k5HA5m9uuACUt/xLjnluDIvn9YjtD00B1YHRiGwYcfzkBFRQVWr16Hb79dj6KiIrw/cwa+P5OK7t52eMq/7rp3crkc7733P5SXl+P773/Cp58uw7lzp7F27SqWvoVpYhgGP/64Dnv2/NXo9+ze/QfOnTsDNxszeNqZY/HBu6isUeoxytbrwfSvrFHiakYZhgbr7j6ckpKEKVOm4u+/D9Q+Fi9ezkLUpk+tVuOTTz5ERkY6li37Ct9/vxEWFpZ49913UFamPd3+jRvXsfnreejapQuq+8wAJ3AAFi1aiAMH9rMQPWktNm5cj3//3Ye5cxfhu+82ID8/H3PmzAIAKNUMVp9KgaedOcaEueDEiaP46qvleOmlV7Ft2y6MH/8SVqz4DGfOnAQASKVSzJjxDqytrbFhw2YsX/41YmKuY+nSRWx+RdY1lMYN2b37DyTfvAILIQ+rTybX3rXZtGkDtm/fjHffnYmNG7dBInHEzJnvQiqt0vdXIUZu1aovceXKJSxevBy9Jy2GnOGi/PAayOVyrX2pDqsfDdVpc+9cgyD6V8gUKhy+QxMzNhU1YHUoLi6Ct7c3Pv54Lvz8/OHn54/x419EcuIdqGqqMGuAr1aF/vDhAygqKsSSJSvg6+uHzp2fwBtvvIXbt+NY+hamJysrE9OnT8Hu3b/D2dmlUe/JzMzA+vVrERISBi4HWDgkAPmVNfjmRLKeo219Hk7/1GIpeFwORoVqL5OhUCiQmZmB4OAQODhIah/W1jSrYXMkJt7FzZs3MHv2fAQHh6B9ex/Mm/cpZDIpzp07o7X/jh1bERIShpWL5mH2mCcRbxYI6/DB2PDjOhaiJ62BQqHArl07MXny/xAZ2Q0BAYFYtGgpYmNjEBsbg1+vZSGlSIppvdqDz+OitLQEb7zxFoYNGwE3t3YYMWI0fHw64MqVywCA3NwchIWF46OP5sLLyxshIWEYOfIZXL16ieVvyp5HpXF9HiznvOzEqJSr8M2pZEilUvzyyxZMnfoeevfuC09Pb8ya9QmEQiHu3Ik34Dcjxuj06RN45plxsHb3x74M4MmRE1BSlI/UVO2hVlSHbXn11Wlraqrx+eefYc6cD+Ht6QUhj4vPjySgoJLmtGgKasDq4OAgwaJFy+DqqplhMT8/D5t2/Aq1nQcm9g6Eu632ws+XLp3HE090rVOBf/rpUdiwYYvB4jZ1cXGxcHNrh82bd9amfUNUKhWWLFmAl156Bd7emrV4Q92sMSHSA3/fzMWJhEJ9h9yqPJj+zi5uSC+Rob+fBBIdE5WlpqZApVLBy8vb8IG2Qs7OLlix4ht4enrVbuNyuWAYBhUV2rMNZ2RkoFOnCADAM2GuWDgkAJmMA/Jyc3A7Oc1gcZPWIyHhDqTSKkREdKnd5urqBldXN5y5dBk/nEvFkz726OOr6eo6evT/mWXiAAAgAElEQVQ4TJjwOgBAqVTi2LEjSEtLRWRkVwCAj08HLF68HObmmvIyPT0NBw/+g8jIbgb+ZsajoTSOiYnW+Z6HyzkzARevRrpjf1weth04Bblcjn79BtTub2FhiV279tT5P0jbZGtrh6NHD2PZvmsQctRwKoiGlZU13Ny052+hOmzLq69OW1JSguzsTKxbtwm9e/eFnVgIuUqNj/bchkKlZjFi00IN2EeYPfsDjBkzHAnxcXDp+ypermfiioyMdLi4uGLDhu/x7LMj8eyzo/Ddd9+gpoauqDTWoEFDMXv2fDg4SB69M4CtWzeBwwFeeGFCne1vdfdCkLMlFh28g8xSWjevsR5M/xKpHAoVg3Hh2ndfAU33YYFAgJ9++gFjxgzHCy+Mwfr1aym/N5ONjS169HgSXO5/p+Rdu3ZCLpcjKkq7wi+RSJCfn1f7fHhHZ7zQUTNOefqOC7ieWab/oEmrUlCQDwBwdHSqs10ikeDI9QSoGeDD/tq9j+Ljb2HAgJ6YP/9jDB48DD16PKn12a+99iJefHEsyspKMX36+/r7EkauoTR+8Hh+kK5ybmJ3LwQ6WeKXE9GwtrHFrVs38dZbr2HEiEF4//1pSEmhHkgEmDVrDlIzsxG3fgbUf36Iwwf24osvVsHKSntJPKrDtrz66rQuLq5Ytep7BAYGAQAEPA7mDQ5AbE45vjqexEaoJokasI/w5ptTEPX6p1DZt4fs6HcoKdZ9V6+qqgr79v2NrKxMLF68HNOnv4djxw7jiy+WGjjituHOnXjs3Lkdc+YsqlPpBzRLjSwbEQQOOJi99zZqlHRFqykYhkFBpRxWIj4i2tno3CclRTMTpqenF774YhVef/0t7Nv3N+X3FnLmzEn88MN3GD/+pdreBQ8aPHgYjh49hKNHD0OpVCIh4Q7O/fs7AMCMy2DKrhvYeCGd1pgjjVZdXQ0ulws+v+7cjhUKILu4Em9194KbjfZkLq6ubvjxxy2YPXs+jh49jPXrtSdxmz17Ptas2QCJxBHTp09BdXW13r6HMasvjQUCoc5xifWVcwIeF4uHBUJZI0N5RSVWrvwCr7zyBlasWAlzczNMnToJJSUlev8+xLjF3ElCBcccPqNm4Pu1PyEqqhvmzftI58USqsOya2CAI16JdMcfMTnYeS2L7XBMAjVgAWzZshEDB/aqfWzZsrH2tSzY4XSpJV743yfggMG//+7T+Rl8Ph/W1taYN+9TBAYGo1evvpg27T0cOLBf5yQspPlqamqwePE8TJr0NtzdPXTu087GHAuGBCA+vxLLDt+lZQea4FJaKaQKFXwk4nqn2p806W3s2XMQzz//Mjp08MWgQUPw7rsfUH5vAf/8sxdz5szCgAED8c4703XuM3To03j99UlYunQh+vfvgU8++RCvvvoaAGDZmM7o7yfB92dT8cYv0UgqpMlcyKOJRCKo1Woolf9NgJdTXo20wgo4WFvixXp6H9nY2MLPLwDDh4/EK6+8jt9++wUqVd2lnQICAtGpUwSWLFmB7OwsnDp1Qp9fxWjpSmMAUCjkMDOrOzTpUeWct4MYvfwcoVbK4TPwVTz5ZG8EBXXE/PlLwOFwcPAgTejWlqWkZ2D9tysg6DIWX739LEJCQrFw4WcQCoX49ddftPanOiz73nmyPfr6OuCr40n497buHhnkP7SMDoDRo8eif/+Btc/FYjGOHDmIsK598dmhBAQ4WWJKbz/c+KUdCgp0zxQmkThBJBKCx/tvaR1vbx8AQE5ODmxsbPX7JdqQW7duIjU1BevWfYt1674FoJlBj2EYDBzYC1u37oKLiwv6+Dpgcg8v/HAuDe0dLPBqlO7GLvkPwzD48UIaBDwOPHWM9b6Py+XC2rru3dn7y73k5+dRfm+mzZt/woYN32Ps2OcwY8aHDa7V99prE/Hyy6+htLQEDg4SXL9+ETweDz4e7lgaIEZ/Pwk+P5qICduu4dVID7wa5aG19Bch9zk5OQMAiooK4ezsApWawYJ/4sHIyjCsix/43Lp5MTr6KiwtLess19Khgy9qampQXl6O6moZEhPvolevvrWvSyQSWFvboLAw3yDfydg8nMb3FRYW4skn66613Zhyrm9oB5zeDRzM4aPn7TwMDXKGSCSCq6sbcnKyDffFiFFhGAbLdx0DGDXmPf8UHC1FADSNVD+/AGRlZWi9h+qw7ONxOVgyPAjv/hmLRQfuwozPQz+/xg2pa4voDiwAa2sbuLt71D5yc3OwcOEcvLfxX1QrVVg8LBA1MinS09N0ducDgE6dwpGQcLfOldXk5CTweDy4uuoeR0iaJzi4I3bu/AubNv1S++jdux8CAoKwadMvkEj+O+Df7OaJwYGOWHM6haYpb4RTScW4nlUOFyszcLn1N57mzfsYs2fPrLMtPv4WhEIh2rWjCwXNsX37ZmzY8D0mTpyC996b1WDj9Y8/fsWqVV+Bz+dDInEEh8PB0aNHERISBrFYDAB4KsARv77WBf39JPjxQjqe3XQFx+4WUG8EopOvrz/EYgtcv34NALD6VDKi76YCVcXo3a2r1v738+uDbt2Kg52dPWxtbXHrVhzmzv0IxcVFta9nZ2ehtLSktmLc1jycxgCQk5ONnJxsdOrUuc6+jSnnwsLCAQB+vGIsOXgXV9JLUVNTg6ysLLRrp/uOOWn9fr6UgegiTfnRDv91JWcYBqmpKXB399R6D9VhjYOIz8WXozoi2NkSs/fewv44uhNbH2rA6hAYGAyJVxBSDm3E674MFEXpmD//Y9ja2mHo0KcBaKbDLyoqhEKhWYtt9OixkMtrsGTJAqSlpeLy5YtYu3YVBg8eRleuWsCD6S0SmdW54ODu7gELCwuIRCK4u3vUGV/E4XAwd5A/wttZY94/8TibXMzitzBucqUaq08lw8vOHBLLujMPP5zf+/YdgDNnTmLnzm3IysrE8eNHsGbNKjz//Mu1DSjSeImJCVi/fi2GDx+JESNGo6iosPYhk8m00t/T0xt//vkbDh78Bzk52di6dRP27duLiROn1Plce7EQS4YH4YfxYbAy4+Ojvbfxzu+xSCygbsWkLqFQiGeeGYc1a1Zh1a/7sePoRbjc2onw8M4ICQnVyoPjx7+I8+fP4pdftiAzMwP79u3GL79swZtvvgUOh4OePXvB1bUdFi2ah6SkRMTGxmDu3I8QEhKGbt16sPxt2fFgGl+4cA537sRjwYJPdKZxY8o5V1c3DB48FNWXdsJRmor3txzDx/PngsvlYtCgYWx/XcKCXdezsfZMKgZ274KQkFAsXboQMTHXkZaWii+/XIa8vFyMGzee6rBGzFLEx3fjwtDZwxYLD9zBzxfT6cKzDtSA1WHPzTxkdXwBXu198cfaTzF16mSIxRb47rv1tZXz2NgYjBo1pHbtNnt7B3z33QaUl5fhjTdewqJFc9G3b3/MnDmbza/Sajyc3k1hJuDh62dC4O9ogY/23sKFVGrE6rLxYjrSS2T4oH8HPHzv7+H0HzBgID75ZAH++WcvJkwYj++++wbPPvu8VgOKNM7Ro4egUqmwf/8ejBo1pM7j11+3a6V/ZGRXzJw5Gxs3rsfLLz+LU6eOY82atfUundHZ3RZbXu6MWQN8cTe/Ei9uuYpFB+4gt7xtTqZDdJs06W2Edu2L335YAbOz3yPczxtLlqwAoH0OiIzshiVLPsfBg//ilVeex7ZtW/Deex9i9OhxAAAzMzOsXPkdxGIx/ve/SZg16z34+fnjyy9Xa02815ZMmvQ2Bg4cgsWL52H69MlwcXGtN40b46OP5qF/v6egPL8V6sNf4EpiFqbN+xK2ttToaGt2Xc/GiqOJ6NPBAQuHBWH58pUIDg7BwoWfYPLk15CZmYm1a3+Ei4sr1WGNnFjIw8pnQjAwwBFrzqRi7v54VCtUj35jG8JhGmjWFxRorz9ozGxtxSgtlT7WZ5xNLsYHu28iyssOX4/uCD7PMAVtS8TOJlOIv1SqwDu/30BKkRQLhgRgSJBmKQNjjt3RUXu6e10e91iNySrD5N9uYHCgIxYNDXysz9I3Y/692NTYdCmTKfDzpQz8Gp0FLoeD8RHt8FqUB6zMWt+UCIbMK409VgHDl62NTYerGaV498+b8LA1x7rnwmBjLjBAdOxpbeeSrDIZpv4ei8JKOVaMCkZ3b/tmfU5rS5eHGfOxqsujfg+VmsHaM6nYcjkDvXzs8fnIYAgMVHc1dqaelxmGwZbLmVhzOgV+jhb4fGQw3BuYn6SlGEu6NXSs8hYuXLiwvhelUu1p3Y2ZmZkA1dWKZr//YmoJZu25hQ4SC6waEwqRASc8edzY2WYK8ZsJeBgc6IQb2eXYfjULHA4Q3s4GYnOh0cZuYSFq1H6Pc6zmlldj+p83YWsuwJejOkLIN+6CzxTyGhsamy5mAh66edthWLAziqVy/BGTg92xOQAAP0fLVlXxMWReaeyxChi+bG1MOlxILcbMv+PgamOG758Lg51Y2OD+rUFrO5dYmwnwVIAjzqcUY8fVLJgLeAh1tWpwPL0urS1dHmbMx6ouDf0epTIF5uy/jb1xeRjbyRULhgS0qnP44zL1vMzhcBDezgbBrlb4+2Yufr+eDQshH0Eulk0+rpvCWNKtoWOVGrD3HE8oxKy9t+BpJ8bqsaEGv/JsLJmluUwlfiGfi4EBTsitqMbOa9mIzSlHLz9H8Ix0fIG+G7DZZdWY+kcsKmqUWD02FK7W2us8GhtTyWuG1tR0sRLx0c9Pgt6+DkgpluLPGzn460YOFGo1/CSWEBn5hYzGoAasxqPS4ffr2Zj/Tzy87MX4blwYHCxaf+MVaJ3nErFQc6E2tUSGndeykFBQhUhPW5g34YJ8a0yXBxnzsapLfb/HuZRiTP/jJhILq/BBP19M7und4OSLbVFrycueduYYEuiEuwVV+C06G9GZZQhzs9FbW8VY0o0asA1QMww2XczA50cTEOxshTXPhrJy5dlYMktzmVL8fC4HfX0d4GQpwp83crHjcgYsRTwEOuv3ilZz6KsBq1IzOBRfgFl7bkEqV+GbZ0IQ7GLdnBANzpTymiE1N10kFkIMC3ZGj/Z2yC6rxp83cvF7TDYqalTwsDU36a7F1IDVqC8dKmuUWHokAZsuZqCnjz2+GRMC21bebfhBrfVcIuRz8ZS/BBZCPn6PycHu2FzYiQXwc7RoVBnXWtPlPmM+VnV5+PfIKa/G0kN3sfZMKlysRVg9NhS9fR1YjNB4taa8bCniY2iQE5ytRNgXl4cd17JQKlMgyNmySReoGsNY0q2hY7VNj4HNLa/GZ4cTcCG1BIMDHfHJQH+Iheysk2gs/c2by1TjTy2W4uuTyTifXAxfiQUmdfdEXz8JuEbSkG3JMbAKlRqrT6Ugt7wat3IrkF8pR5CzJT4dFghve9OZOdhU85q+tVS63MmvxKaL6TieUAiGAXr62GNcuBu6e9sZzXHRWDQGVuPhdGAYBmdTirHiaCLyKmrweldPTOruBV4bu3vTFs4lSYVVWHo4ATeyy+HnaIHJPbzRq4N9g8dya08XYz5Wdbn/e+SWV2PblUzsjs0FALwW5YEJkR6toreMvrTWvFxYJceP59Ow+0YOzAQ8vNC5HZ6NcIN9C92AM5Z0a+hYbZMNWKlchV+uZuLnS5rFnN/v64NnwlxZvftmLJmluUw5fhsbc+y6mIb159KQViKDu60ZRoa4YERHZ0gsG3+lVh9asgFbrVDh3T9vokSqgLeDGIMCHNHPT2JylVZTzmv61NLpkltejb9ic7H7Rg6KpQq4WInwVIAjBgU6ItDJ+Hor6EINWI0H0+FmTjm+P5OKS+ml8LQzx4IhAQhzM43eFy2trZxL1AyDg/H52HAuDRml1XC3NcPoUFeMCHHWWeFt7elizMeqLrnVKvx4Kgn74/LAABgW5IRJPbxMYsgP21p7Xk4pkmLtmRScSCyCkMfB0CBnPBvhBv9G9raoj7GkGzVg78ktr8bu2Fzsup6N8molBvhLMKOPD1yM4CRgLJmluUw5/vuxq9QMjt4twB8xObiWWQYeB4j0skMvHwf07mDPSj4x1CzEpsSU85o+6StdFCo1jicU4p9b+biQVgKVmoG7rRn6+zmiu7cdwtysjXbiL2rAaphbmmHftQz8Fp2Na5llsDHjY2J3L4zt5NqmJ3xpa+cSpZrBkTsF+PNGDqIzy8DnctDFwwa9OzigVweH2gZRa08XYz5W76usUeJYQiF238hBbE4FhDwORoa44JUoD2q4NkFrz8v3pRZLsfNaFvbF5aFGqUZ7BzGGBjnhKX9HeNg1fdZiY0m3Nt2ALZUqcCq5CP/eysPVjDIwAPr6OuDVKA+EuDbvqvOKFUvx5ZfLWyBi0lQzZ36MWbM+adHP1JVv0oql2HMzF8cTCpFRqlkr08/RAt297RHhbo1ObjYGGRtobA1Yyvvsaij/G6LAKZMpcCKxEIfiC3A1oxQqBjDjcxHuboNQVyt0dLFGkIsl7MwFRnGHti03YGUKFS6mluBMcjFOJhWhVKaAs5UIL3Zph1GhLrAQtsz5i84J+qePci+lSIp9cbk4mViEtBIZAM1EMRHtbNDT3xHtrUXwsDM3uR46jWFsx+p9VXIlziQV48jdApxLKYZcxcDb3hwvdvVCv/Z2ehmfTsev4enjeL6vVKbAkTsFOHA7HzHZ5QAAD1sz9Ghvj0hPO4S6WTWqm7EpNGBNd3aOeijVDOLzKnAupRjnUkpwK7cCDAB3WzNM6uGFoUFOBllDiZg2L3sxpvX2wdRe7ZFWLMPp5CKcTi7GL1czseUyAw4AX0cLhLlZI8DJEgFOluggsaCxKKRVszEXYFSoK0aFuqKyRomrGWW4mFaCa5ml+PF8Ce5fDbUS8eFpZw4PO3N42JpBYimCg1gAe7EQ9hYCOIiFMDPgMmVtQXm1ArdzK3E9qwzXs8pwI7scchUDCyEP/QKcMMjfAVGedq2yQUKarr2Dpoyb1tsHqcVSnE0uxtWMUhxLKMTfNzVjLM0FXPg7WiLQ2RL+TpbwlVjAy968xS5+tHVqhkFiQRUupZfiYloJojPLUKNUw9FSiDGd3PCUvwRhbtaws7MwisYEMX625gKMC3fDuHA3ZJdV40xyEc6nlmB3bC5+jc4GALhZi9DR1RrBLlZoby+Gl705XK3NTK5sMOmzULVChcyyaiQXViEutwJ3CqpwM7scNUo1OAA6ulphUncv9PCxR7ARzjBLjB+Hw4G3gxjeDmJMiPRAtUKFmzkViM4qw/XMMhy4nY8/YjRraPK4HHjZmcPD1hztbM3gbmsON2sz2JjzYW0mgLUZH1YivsmdJAjRxVLERx9fB/S5N/tllVyJ+LxK3MmvRHqJDOklMlzPLMPB2/nQ1c1HLODBViyAjRkfNuaav23NBff+LYCtOR82ZgLYmGtetzUXwIzPbZPncYZhoFIzKJUpUFAlR0GlHAWVNUgvkSG5UIqkoioUVGpmS+VyAH9HS4wLd8OTPvaIaGcDiYMlVYBJvbztxfC2F+OlJ9yhZhgU1KhxKbEAd/IrEZ9XiT03cyFTqGv3d7IUwuvee9rZmMHRUggnSxEcrYSwMxfCXNA2j9P6MAyDMpkS6aUyZJbKkFRYhVu5FbidV4kquQoA0N5ejDFhrujvJ0FYO2uTmzCPGB83GzM8F9EOz0W0Q41Sjdu5FbiZW4G4nHLcyC7H4TsFtfsKeRx42JnD004MJ0shPCSWsOJz4GgphMRCBCszPiyFPIiMqAx+7C7EcqUah+7kQypXgWEABvce9z6WYTRXme7/+/5r9/9TzTam9jU89Jy5t5NMoUZZtQJlMiXKqhXIr6hBfuV/05uL+Fx0dLNGgMQCIa5WiPLST3cLfTGW2/XNZcrxP07saoZBdll1bUGfVFiFrLJqZJVVo0ap1vkee7EAv7zSpVFrLRpbF2JjYMp5TZ+MNV0UKjWKpQoUS+UorlKgSCpHcZUcxVIFSmWKOuf1UpkClTWqej9LyOPUNnBtzDUNXnMBDzwOBxwOwL339/3nQhEfNdXK2vf/V+4wWtsedP9lIZ+LVyPdGzWZW0t3S3z3z1jcyC6HQsVArlTrjFPE56K9vRgdJGJ0kFjA39ESIW5WWnfIjDVvsI3SRbeH00WlZpBRIkNysRRp9x6pxTKkFktrG2AP4nI069GKBTxYiPgwF/Ag5HHA53Eh4HIg5HEh4HEgqPO35jUB/799+DxNRVnN3Ks3PlSHvF+/vF/HHBbs3Kgedi19rP5zKw+nk4qgVGsuNCnVDKoVKpRWK1FerUR5tQIK1X9HMJ/LgZ+jBYJdrBDiaoVITzs4W9V/jqF82nyUdvUrlSqQViJF2r1jObVYioxSGQoq5TqPa0CTd61EfFiIeJpjlKs5fvlcDvg8DvhcDjjg1Lbj1ABw73hVM//9e1SoC57u6PLIGPXahTilSIolB+9Cpat0bQEcABwOYMbnaa7Em2muxHt5msPd1hyedubwstMU4HSVmRgal8OBu60mLw7wd6zdzjAMCqvkyC2vQXm1pnJeXq1ERbUSYiEPViKT7vxASKMJeFw4W4karKA9SKlmUH6vUVsqU6DsgUbuww3epMIqyBRqMAwDdW1lFrXPuVwOGEbT5f9hD15F5tRuq7uPiM/F08HszEbey8cBHrbmmsr+vUq9jbkAjhZCOFqJ4GghhIOFkHp0EL3jcf/rifQghmFQUaNE/r0eAQUVcpTKFKhSqFBVo4RUrkKVXAWZQgWFmoFCqYZUzUChUt97MP/9rVY3eLHmUbgc1JbFhpZRIkNSoRQ8Lqf2YcbnwtteDBszTQ8sBwsBPGw1wyra2Zi16cnTiHGwFQtgK7ZBp3Y2Wq8JzIVIzCpFYZUchZVyVNQoUVGjRGWNClVyJSprlJCrGChVaijVDBRqBiqVGtJ7x6+m7capbcPV/pur+bslyq3HrkUHOFviyP96QKFSgwMO7v3RBAzOvcAf+Pe999X5YpoNdZ4byy1qQpqDw+HA0VIER5aX4SHE1PC5HM1Y2RZYz86Ur76PC3djOwRCGsThcO4NjxHAV2LRYp+rVGsqxgoVU3tR6X7viocrxlwOB1wOu3XGyT29MbmnN2v/PyEtzULEh5e9GF724kfvzJIWuQ1kSXeTCCGEEELIY+JzOeBzeTAznVFghBADoz4MhBBCCCGEEEJMAjVgCSGEEEIIIYSYBGrAEkIIIYQQQggxCdSAJYQQQgghhBBiEqgBSwghhBBCCCHEJHCYB1d0N3EnTpxA37592Q6jWUw5dsC04zfl2Nsi+r10o3TRRmmiQemgG6WLbpQuxoV+j+ajtGseU0i3VnUH9uTJk2yH0GymHDtg2vGbcuxtEf1eulG6aKM00aB00I3SRTdKF+NCv0fzUdo1jymkW6tqwBJCCCGEEEIIab14CxcuXMh2EC3J29ub7RCazZRjB0w7flOOvS2i30s3ShdtlCYalA66UbroRuliXOj3aD5Ku+Yx9nRrVWNgCSGEEEIIIYS0XtSFmBBCCCGEEEKISaAGLCGEEEIIIYQQk2CyDdhr164hICCgzmPUqFH17p+VlYU33ngD4eHhGDp0KKszbBUVFeGDDz5At27d0L17d8yePRvl5eX17j9v3jyt7/rzzz8bLmAAcrkc8+bNQ2RkJHr27IkNGzbUu298fDzGjx+PTp06YcyYMbhx44YBI9WWnp6OKVOmIDIyEr1798by5ctRU1Ojc9833nhDK62PHDli4IjJw5ryG7YlSUlJeO211xAREYF+/frhxx9/ZDskozNnzhxMmDCB7TD0qinn5z/++AMDBw5EWFgYnn/+edbPz/rUlHS5r7S0FD169MCff/5pgAjZ0ZR0oTKRPWVlZZg5cyaioqLQq1cvfPnll1CpVGyHZdTkcjmefvppnDt3rnabMdX/jZWudDt//jzGjh2LiIgIDB48GLt27WIxQm18tgNorqSkJISEhGDdunW12/h83V+HYRi888476NChA37//XccO3YM06dPx759++Dh4WGokGt98MEHUCqV2LRpE5RKJRYuXIg5c+bg22+/1bl/YmIiZs2ahZEjR9Zus7S0NFS4AIAVK1YgOjoamzZtQm5uLmbNmgU3NzcMHz68zn5SqRQTJ07EsGHDsHTpUuzcuROTJ0/G4cOHDR4zoDkop0yZAl9fX+zcuRNFRUX45JNPAAAff/yx1v6JiYlYuXIlIiMja7fZ2NgYLF6iram/YVuhUCgwadIkdO3aFYsWLUJycjI++OADODk51TlXtGXnz5/H77//jqioKLZD0avGnp/Pnj2LRYsWYfny5QgJCcH27dsxadIkHD16lJXzs741Nl0etHTpUhQVFRkwSsNrSrpQmcieRYsWoaCgANu2bUNxcTFmzpwJW1tbTJw4ke3QjFJNTQ0++OADJCQk1G4ztvq/MdKVbqmpqZg8eTLeeecdDB06FDExMZgzZw4cHBzQv39/FqN9AGOili5dynz00UeN2vfcuXNMaGgoU1FRUbvt1VdfZb7++mt9hVevnJwcxt/fn0lKSqrddvXqVSYwMJCRSqU63xMVFcVcuHDBUCFqqaqqYkJDQ5mzZ8/WbluzZg3z/PPPa+27a9cupm/fvoxKpWIYhmHUajUzcOBA5rfffjNYvA+6fPky07FjR6aysrJ22549e5gePXpo7VtRUcH4+/szGRkZhgyRPEJTfsO2JCMjg3n33XcZmUxWu+1///sfM3fuXBajMh5VVVXMgAEDmOeff555+eWX2Q5Hb5pyft69ezezfv362uf3z3lXr141SKyG1JR0ue/EiRPM4MGDmW7dujF//PGHIcI0uKakC5WJ7OrcuTNz+PDh2ufLli1j3nzzTRYjMl4JCQnMyJEjmREjRjD+/v61+duY6v/GqL50W7NmDfPcc8/V2Xfu3LnMjBkz2AhTJ5PtQpyYmIj27ds3at+YmBgEBwfXucLcpUsXXL9+XV/h1f0PCKoAAApeSURBVMvS0hLr16+vMz01h8OBWq2GXC7X2r+goAClpaWN/q76EB8fD7lcji5dutRu69KlC2JjY6FUKuvsGxMTg86dO4PL1WQtDoeDzp07Izo62qAx3+fj44P169fDwsKidhuHw9GZ1omJiRCJRHBzczNkiOQRmvIbtiXu7u745ptvYGZmBoZhcPXqVVy+fBndu3dnOzSjsHLlSkRFRbX6u69NOT+PGjUKkyZNAgBUV1fj559/hoODA/z9/Q0asyE0JV0AoLKyEgsXLsTixYshEAgMGapBNSVdqExkl62tLfbs2QOZTIa8vDycPn0aHTt2ZDsso3TlyhX07NkTv/76a53txlT/N0b1pdvQoUMxb968Ots4HI5RDd0y6S7E5ubmePrpp1FZWYnevXvjww8/hJWVlda+BQUFcHJyqrPNwcEBubm5hgq3lqWlJfr06VNn25YtW+Dv76+zW05iYiL4fD5WrVqFU6dOwc7ODq+99hrGjBljqJBRUFAAGxsbiESi2m0SiQQKhQLFxcV10ragoECrse3g4ID4+HiDxfsge3t79OjRo/a5Wq3Gtm3b6hTe9yUmJsLa2hrvvfcerl69ChcXF0ybNk3r9yKG1ZTfsK3q3bs38vPz0a9fPwwePJjtcFgXHR2NAwcOYN++fdi4cSPb4ehVU87P950+fbq2Ifvll1+2yu7DTU2XL774Ar169arTVbY1akq6UJnIrgULFmDWrFno3Lkz1Go1unXrhmnTprEdllF6/vnndW43pvq/Maov3R6uxxcWFmL//v2YOnWqIcJqFKO9A1tTU4O0tDSdj8rKSuTk5ECtVmP58uVYsmQJrl27hg8++EDnZ8lkMq0rqkKhEAqFgpXYH/Tzzz/jwIEDmD17ts7PSk5OBgAEBgZiw4YNGDduHObPn49///1XL7HrIpPJIBQK62y7//zhu2D17Wssd8uWLVuG27dv68wrSUlJqKqqQv/+/fHjjz+iT58+mDJlCmJiYliIlNSnod+wrVq7di3Wrl2LuLg4LFu2jO1wWCWXyzFnzhx88sknbWKsXlPOz/cFBgbir7/+wtSpU/Hxxx+3yrsRTUmXS5cu4fjx4/jwww8NFh9bmpIuVCayKz09HcHBwdi2bRvWr1+PrKwsfP7552yHZVIMXf9vjaRSKaZOnQonJ6d6G7xsMNo7sLGxsXjppZd0vrZs2TJcvnwZFhYW4PF4AIDly5dj7NixyM7O1uruIhKJtBqOcrkcZmZmrMR+/+7pxo0bsWLFCsybN6/OHaYHvfjiixg+fDhsbW0BaCoeaWlp2LFjB4YOHaqX+B8mEom0Crb7z83NzRu1r77SurEYhsFnn32GHTt2YNWqVfDz89PaZ+bMmXj77bdhbW0NQJPWcXFx2LlzJzp16mTokMlDGvMbtlWhoaEANN1CP/roI8yaNUurktpWrFmzBl5eXgY7P7KtKefn+xwdHeHo6IigoCBER0dj586dCA8P13ushtTYdKmursbcuXMxb948nT24Wpum5BcqE9mTnp6OpUuX4tixY3BxcQGg+e3eeOMNTJ48GRKJhOUITYOh6/+tTUVFBSZPnozMzEz88ssv9ZYpbDDaBuwTTzyBO3fuNHr/Dh06AADy8vK0GrDOzs5aXVgLCwvh6Oj4+IHq0JjYV69ejTVr1mDu3Ln1NnYBTZ/z+43X+3x8fHDmzJkWibUxnJ2dUV5eDrlcXlspLigogFAo1LrD4ezsjIKCgjrb9JnWjaFWqzFnzhzs3bsXK1euxFNPPaVzPx6PV1tQ3+fj49OkfEj0o7G/YVuSl5eHmzdvYsCAAbXbOnToAIVCgcrKStjb27MYHXv27t2LgoICREREANDM1qxSqRAREcHaWHx9asr5OTo6Gubm5ggMDKzd5uvri9TUVEOGbBCNTZcbN24gLS0Ns2bNqt0mk8mwYMECXL9+HZ9++qnBY9enpuQXKhPZc/PmTVhYWNQ2XgEgJCQEKpUK2dnZ1IBtJEPX/1uT4uJivPnmmygsLMSWLVvg6enJdkh1GG0X4obcuHEDERERyMvLq91269Yt8Hg8eHl5ae3fqVMnxMfHQyqV1m67evUqa1ecN2/ejLVr1+LTTz995PqEy5cvx+TJk+tsu337Nnx8fPQZYh1BQUEQCAR1Kn9Xr15Fx44dtZYu6tSpE6Kjo8EwDADNXbPo6GhWr+4vX74ce/fuxbfffotBgwbVu9/06dOxcOHCOttu377N6gRaRKOxv2FbkpSUhGnTptVZ8iMuLg729vZttvEKAFu3bsW+ffuwe/du7N69G88++yxCQkKwe/dutkPTi6acn7dv345vvvmmzra4uDiDlieG0th0CQsLw6FDh2rzy+7duyGRSDB9+nS8++67bISuV03JL1QmssfJyQnl5eXIycmp3ZaUlARAM4EfaRxjq/+bivvLF5aUlGD79u3GWUawOgdyM9XU1DCDBw9m3nzzTebu3bvMpUuXmCFDhtRZPqKoqKh22Q2lUskMGzaMmTZtGnP37l3mhx9+YDp16sTK1PBZWVlMSEgIs3DhQiY/P7/OQ6lUasV+8eJFJjAwkNm8eTOTlpbGbNu2jenYsSNz+fJlg8Y9b948ZujQoUxMTAxz5MgRpnPnzsz+/fsZhmGY/Pz82qU8KioqmG7dujGLFi1iEhISmM8++4zp3r17nSnMDSk6Oprx9/dnfvjhB630fjj2v/76iwkJCWH27NnDpKSkMKtWrWJCQ0OZtLQ0VmInGo/6DdsquVzOPP3008zEiROZxMRE5tixY0z37t2Zn3/+me3QjMrXX3/dqpfRYZjGn5+vXbvGBAUFMVu3bmVSUlKYr7/+mgkPD2dycnLYDF9vGpsuD+vVq1erXUaHYRqfLlQmskehUDCjRo1iXn31Veb27dtMdHQ0M2LECObDDz9kOzSj9+ByMMZU/zd2D6bbDz/8wAQHBzPnzp2rU+cqKSlhOcr/mGQDlmEYJj09nZk8eTLTpUsXJioqilm8eDFTU1NT+3q/fv2Y1atX1z5PTU1lXnrpJSYkJIQZNmwYc/r0aTbCZrZu3cr4+/vrfKSmpuqMff/+/czw4cOZkJAQZujQoczBgwcNHrdUKmVmzZrFhIeHMz179mR++umn2tf8/f3rFPYxMTHM6NGjmZCQEGbs2LFMbGysweO9b/ny5fWmt0Kh0Ip969atzFNPPcWEhIQwY8aMYS5dusRa7ETjUb9hW5aVlcVMnjyZiYiIYJ588klm3bp1jFqtZjsso9IWGrBNOT8fOnSIGTZsWO35uTWuAXtfU9LlQa29AduUdKEykT25ubnM9OnTmaioKKZnz57M4sWL673oQv7zYEOMYYyn/m/sHky3Z555Rmedq6F1tA2NwzD3+noSQgghhBBCCCFGzCTHwBJCCCGEEEIIaXuoAUsIIYQQQgghxCRQA5YQQgghhBBCiEmgBiwhhBBCCCGEEJNADVhCCCGEEEIIISaBGrCEEEIIIYQQQkwCNWAJIYQQQgghhJgEasASQgghhBBCCDEJ1IAlhBBCCCGEEGIS/g8GKr9qoNHazgAAAABJRU5ErkJggg==\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], "source": [ "az.plot_posterior(data, figsize=(13,3));" ] @@ -318,7 +608,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "# Non-linear Differential Equations\n", + "## Non-linear Differential Equations\n", "\n", "The example of an object in free fall might not be the most appropriate since that differential equation can be solved exactly. Thus, `DifferentialEquation` is not needed to solve that particular problem. There are, however, many examples of differential equations which cannot be solved exactly. Inference for these models is where `DifferentialEquation` truly shines.\n", "\n", @@ -354,9 +644,20 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 9, "metadata": {}, - "outputs": [], + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], "source": [ "def SIR(y, t, p):\n", " ds = -p[0] * y[0] * y[1]\n", @@ -380,9 +681,85 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 14, "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Auto-assigning NUTS sampler...\n", + "Initializing NUTS using jitter+adapt_diag...\n", + "Sequential sampling (2 chains in 1 job)\n", + "NUTS: [lambda, R0, sigma]\n" + ] + }, + { + "data": { + "text/html": [ + "\n", + "
\n", + " \n", + " \n", + " 100.00% [3000/3000 36:42<00:00 Sampling chain 0, 0 divergences]\n", + "
\n", + " " + ], + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "\n", + "
\n", + " \n", + " \n", + " 100.00% [3000/3000 26:11<00:00 Sampling chain 1, 0 divergences]\n", + "
\n", + " " + ], + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Sampling 2 chains for 1_000 tune and 2_000 draw iterations (2_000 + 4_000 draws total) took 3774 seconds.\n" + ] + } + ], "source": [ "sir_model = DifferentialEquation(\n", " func=SIR, \n", @@ -403,21 +780,29 @@ " sir_curves = sir_model(y0=[0.99, 0.01], theta=[beta, lam])\n", " \n", " Y = pm.Lognormal('Y', mu=pm.math.log(sir_curves), sigma=sigma, observed=yobs)\n", - "\n", - " prior = pm.sample_prior_predictive()\n", - " trace = pm.sample(2000,tune=1000, target_accept=0.9, cores=1)\n", - " posterior_predictive = pm.sample_posterior_predictive(trace)\n", " \n", - " data = az.from_pymc3(trace=trace, prior=prior, posterior_predictive=posterior_predictive)" + " trace = pm.sample(2000, tune=1000, target_accept=0.9, cores=1) \n", + " data = az.from_pymc3(trace=trace)" ] }, { "cell_type": "code", - "execution_count": null, + "execution_count": 15, "metadata": {}, - "outputs": [], + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], "source": [ - "az.plot_posterior(data, round_to=2, credible_interval=0.95);" + "az.plot_posterior(data, round_to=2, credible_interval=0.95, compact=True);" ] }, { @@ -431,7 +816,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "# Conclusions & Final Thoughts\n", + "## Conclusions & Final Thoughts\n", "\n", "ODEs are a really good model for continuous temporal evolution. With the addition of `DifferentialEquation` to PyMC3, we can now use bayesian methods to estimate the parameters of ODEs.\n", "\n", @@ -442,7 +827,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "# References\n", + "## References\n", "\n", "1. Earn, D. J., et al. Mathematical epidemiology. Berlin: Springer, 2008.\n", "2. Britton, Nicholas F. Essential mathematical biology. Springer Science & Business Media, 2012.\n" @@ -465,7 +850,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.8.1" + "version": "3.7.6" } }, "nbformat": 4,