diff --git a/examples/samplers/SMC2_gaussians.ipynb b/examples/samplers/SMC2_gaussians.ipynb
index 1f032162c..77a73bef9 100644
--- a/examples/samplers/SMC2_gaussians.ipynb
+++ b/examples/samplers/SMC2_gaussians.ipynb
@@ -7,7 +7,7 @@
"# Sequential Monte Carlo\n",
"\n",
":::{post} Oct 19, 2021\n",
- ":tags: SMC, pymc3.Model, pymc3.Potential, pymc3.Uniform, pymc3.sample_smc\n",
+ ":tags: SMC\n",
":category: beginner\n",
":::"
]
@@ -21,17 +21,15 @@
"name": "stdout",
"output_type": "stream",
"text": [
- "Running on PyMC v3.11.4\n"
+ "Running on PyMC v4.0.0b6\n"
]
}
],
"source": [
- "%matplotlib inline\n",
+ "import aesara.tensor as at\n",
"import arviz as az\n",
- "import matplotlib.pyplot as plt\n",
"import numpy as np\n",
- "import pymc3 as pm\n",
- "import theano.tensor as tt\n",
+ "import pymc as pm\n",
"\n",
"print(f\"Running on PyMC v{pm.__version__}\")"
]
@@ -62,7 +60,7 @@
"A summary of the algorithm is:\n",
"\n",
"1. Initialize $\\beta$ at zero and stage at zero.\n",
- "2. Generate N samples $S_\\text{\\beta}$ from the prior (because when $\\beta = 0$ the tempered posterior is the prior).\n",
+ "2. Generate N samples $S_{\\beta}$ from the prior (because when $\\beta = 0$ the tempered posterior is the prior).\n",
"3. Increase $\\beta$ in order to make the effective sample size equals some predefined value (we use $Nt$, where $t$ is 0.5 by default).\n",
"4. Compute a set of N importance weights $W$. The weights are computed as the ratio of the likelihoods of a sample at stage $i+1$ and stage $i$.\n",
"5. Obtain $S_{w}$ by re-sampling according to $W$.\n",
@@ -135,16 +133,16 @@
"\n",
"def two_gaussians(x):\n",
" log_like1 = (\n",
- " -0.5 * n * tt.log(2 * np.pi)\n",
- " - 0.5 * tt.log(dsigma)\n",
+ " -0.5 * n * at.log(2 * np.pi)\n",
+ " - 0.5 * at.log(dsigma)\n",
" - 0.5 * (x - mu1).T.dot(isigma).dot(x - mu1)\n",
" )\n",
" log_like2 = (\n",
- " -0.5 * n * tt.log(2 * np.pi)\n",
- " - 0.5 * tt.log(dsigma)\n",
+ " -0.5 * n * at.log(2 * np.pi)\n",
+ " - 0.5 * at.log(dsigma)\n",
" - 0.5 * (x - mu2).T.dot(isigma).dot(x - mu2)\n",
" )\n",
- " return pm.math.logsumexp([tt.log(w1) + log_like1, tt.log(w2) + log_like2])"
+ " return pm.math.logsumexp([at.log(w1) + log_like1, at.log(w2) + log_like2])"
]
},
{
@@ -157,22 +155,55 @@
"output_type": "stream",
"text": [
"Initializing SMC sampler...\n",
- "Sampling 4 chains in 4 jobs\n",
- "/u/32/martino5/unix/anaconda3/envs/pymcv3/lib/python3.9/site-packages/pymc3/sampling.py:1925: UserWarning: The effect of Potentials on other parameters is ignored during prior predictive sampling. This is likely to lead to invalid or biased predictive samples.\n",
- " warnings.warn(\n",
- "/u/32/martino5/unix/anaconda3/envs/pymcv3/lib/python3.9/site-packages/pymc3/sampling.py:1925: UserWarning: The effect of Potentials on other parameters is ignored during prior predictive sampling. This is likely to lead to invalid or biased predictive samples.\n",
- " warnings.warn(\n",
- "/u/32/martino5/unix/anaconda3/envs/pymcv3/lib/python3.9/site-packages/pymc3/sampling.py:1925: UserWarning: The effect of Potentials on other parameters is ignored during prior predictive sampling. This is likely to lead to invalid or biased predictive samples.\n",
- " warnings.warn(\n",
- "/u/32/martino5/unix/anaconda3/envs/pymcv3/lib/python3.9/site-packages/pymc3/sampling.py:1925: UserWarning: The effect of Potentials on other parameters is ignored during prior predictive sampling. This is likely to lead to invalid or biased predictive samples.\n",
- " warnings.warn(\n",
- "Stage: 0 Beta: 0.010\n",
- "Stage: 1 Beta: 0.029\n",
- "Stage: 2 Beta: 0.065\n",
- "Stage: 3 Beta: 0.141\n",
- "Stage: 4 Beta: 0.307\n",
- "Stage: 5 Beta: 0.638\n",
- "Stage: 6 Beta: 1.000\n"
+ "Sampling 4 chains in 4 jobs\n"
+ ]
+ },
+ {
+ "data": {
+ "text/html": [
+ "\n",
+ "\n"
+ ],
+ "text/plain": [
+ ""
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/html": [
+ "\n",
+ "
\n",
+ " "
+ ],
+ "text/plain": [
+ ""
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " "
]
}
],
"source": [
- "def two_gaussians(x):\n",
- " log_like1 = (\n",
- " -0.5 * n * tt.log(2 * np.pi)\n",
- " - 0.5 * tt.log(dsigma)\n",
- " - 0.5 * (x - mu1).T.dot(isigma).dot(x - mu1)\n",
- " )\n",
- " log_like2 = (\n",
- " -0.5 * n * tt.log(2 * np.pi)\n",
- " - 0.5 * tt.log(dsigma)\n",
- " - 0.5 * (x - mu2).T.dot(isigma).dot(x - mu2)\n",
- " )\n",
- " return pm.math.logsumexp([tt.log(w1) + log_like1, tt.log(w2) + log_like2])\n",
- "\n",
- "\n",
"with pm.Model() as model:\n",
" X = pm.Uniform(\n",
" \"X\",\n",
" shape=n,\n",
" lower=-2.0 * np.ones_like(mu1),\n",
" upper=2.0 * np.ones_like(mu1),\n",
- " testval=-1.0 * np.ones_like(mu1),\n",
+ " initval=-1.0 * np.ones_like(mu1),\n",
" )\n",
" llk = pm.Potential(\"llk\", two_gaussians(X))\n",
- " trace_80 = pm.sample_smc(2000, parallel=True)\n",
- " idata_80 = az.from_pymc3(trace_80)"
+ " idata_80 = pm.sample_smc(2000)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
- "We see that SMC recognizes this is a harder problem and increases the number of stages. We can see that SMC still sample from both modes but now the model with less weight is being subsampled (we get a relative weight way lower than 0.1). Notice how the rank plot looks worse than when n=4."
+ "We see that SMC recognizes this is a harder problem and increases the number of stages. We can see that SMC still sample from both modes but now the model with higher weight is being oversampled (we get a relative weight of 0.99 instead of 0.9). Notice how the rank plot looks worse than when n=4."
]
},
{
@@ -374,7 +404,7 @@
{
"data": {
"text/plain": [
- "'Estimated w1 = 0.008'"
+ "'Estimated w1 = 0.991'"
]
},
"execution_count": 8,
@@ -383,7 +413,7 @@
},
{
"data": {
- "image/png": "iVBORw0KGgoAAAANSUhEUgAABLsAAADTCAYAAABp7hHfAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8rg+JYAAAACXBIWXMAAA9hAAAPYQGoP6dpAADLE0lEQVR4nOzdd4AcZfnA8e/M9nK990u99HqpJCQhhSSEAAmEoogooKIoKFJUEEUEEQR/oiggoEgVCD2QhFQgpLdLvVyu5Xrf2927bTO/P/ZKru9NEkLi+/lDudv3npmd3Z3sPPO8zyupqqoiCIIgCIIgCIIgCIIgCOcB+WzvgCAIgiAIgiAIgiAIgiCcLiLZJQiCIAiCIAiCIAiCIJw3RLJLEARBEARBEARBEARBOG+IZJcgCIIgCIIgCIIgCIJw3hDJLkEQBEEQBEEQBEEQBOG8IZJdgiAIgiAIgiAIgiAIwnlDJLsEQRAEQRAEQRAEQRCE84ZIdgmCIAiCIAiCIAiCIAjnDZHsEgRBEARBEARBEARBEM4bItklCMI557bbbmPKlCnU1tZ2+L3H4+Hiiy/myiuvJBAInKW9EwRBEARBOLeI71aCIJxvRLJLEIRzzn333YeiKDz00EMdfv/nP/+ZkpISHnroIXQ63VnaO0EQBEEQhHOL+G4lCML5RiS7BEE458THx3PXXXfxwQcfsHHjRgBycnJ48cUXufnmm8nKyjrLeygIgiAIgnDuEN+tBEE430iqqqpneycEQRC0uOGGGygsLOTdd9/lW9/6Fj6fj3feeQej0Xi2d00QBEEQBOGcI75bCYJwvhDJLkEQzllFRUVceumlxMbGUlJSwiuvvMKECRPO9m4JgiAIgiCck8R3K0EQzhdiGqMgCOes9PR0rr76ak6cOMHy5cvFlzFBEARBEIRTIL5bCYJwvhDJLkEQzlkNDQ18+OGHSJLEzp078Xq9Z3uXBEEQBEEQzlniu5UgCOcLkewSBOGc9fDDD+PxeHj66acpLi7m6aefPtu7JAiCIAiCcM4S360EQThfiGSXIAjnpM8//5yVK1dy5513MmfOHG644QaeffZZcnNzz/auCYIgCIIgnHPEdytBEM4nokG9IAjnHLfbzZIlS0hOTuall15CkiSam5tZsmQJsbGxvPLKK8iyyOULgiAIgiCEQny3EgThfCPOWIIgnHOeeOIJqqurefDBB5EkCQCz2cxvfvMbdu/ezauvvnqW91AQBEEQBOHcIb5bCYJwvhHJLkEQzil79+7lP//5D7feeisDBgzo8NgFF1zA5ZdfzuOPP055eflZ2kNBEARBEIRzh/huJQjC+UhMYxQEQRAEQRAEQRAEQRDOG6KySxAEQRAEQRAEQRAEQThviGSXIAiCIAiCIAiCIAiCcN4QyS5BEARBEARBEARBEAThvCGSXYIgCIIgCIIgCIIgCMJ5QyS7BEEQBEEQBEEQBEEQhPOGSHYJgiAIgiAIgiAIgiAI5w2R7BIEQRAEQRAEQRAEQRDOG/r+/kFdXd1p23hERAQNDQ2nLd75SByj3onj0ztxfHonjk/vxPHpnTg+fftfPUZRUVH9Gn8q363+V49xX8Rx6Uock+6J49I9cVy6J45LV+KYdE8cl+5pPS79/W4FZ7myS5ZFYVlfxDHqnTg+vRPHp3fi+PROHJ/eiePTN3GMzjxxjLsnjktX4ph0TxyX7onj0j1xXLoSx6R74rh076s8LuIVEARBEARBEARBEARBEM4bItklCIIgCIIgCIIgCIIgnDf63bNLEL6u/HXNNO2pQRdpxDI2FkmWzvYuCYIgCIIgCIIgCILwFRPJLuGcp3gCHL9/J643jiIHFAAsY2JIe+pCTJlhZ3nvBEEQBEEQBEEQBEH4Kolkl3DOqKxUOVTcxL6YPeRbCin0VWMvt5C5NsCctXoOWEew9ImB2BsaKfvdTvKv+YRBKxdhSLKd7V0XBEEQBEEQBEEQBOErInp2CV97LpfKb36nsPxqlbufLeVZ9SM2l5Rh8kVTV+nlw4klRP1yHFeun0raggS8swbxyUXz8Df4KP7p56iqerafgiAIgiAIgiAIgiAIXxFR2SV8rTU2qnz/F04KInL5xTcGUZDYxGsHpzN+2GKmljo5nmllxLZiRi1Jw6+HzdVHCaxLZPXhCCYsGI26chf1K/OJWjbwbD8VQRAEQRAEQRAEQRC+AiLZJXytNSguDl//HP6Yeur3XU5ThMSU8YuJK3YzMreAyC3lxEcO49+P/4y1V11Emfcgo1LSuO7OIYQXJMOOCMof3U3EJRnIJt3ZfjqCIAiCIAiCIAiCIJxhItklfG0V1dZwXfU/8MW6kRV4YcRqmp11rPiXgxlfbOfIkhJ2Xu1mQt4sMuY28fPc/WxxT2N87SD+O3ALeReUMuvwYGJf30ndm3nEfGPo2X5KgiAIgiAIgiAIgiCcYSLZJXwt7c0v5h8FG3Bkurl12xgOVf+LAZ8nM2PbZIzSCQIXz+Kh39xEbZyPTU/FMnxTJoZJlSx7rp6nrvmIBlMzD3x2FQdsJiKHxVP+l/1EXzsESZbO9lMTBEEQBEEQBEEQBOEMEsku4WulvsDF1v2vUXN0N59fEcHleWai8mYw740wBtvzCH/3EwJSMqWfj6D4lSKohJk/WcD1xwvYsNjGS7MOMnytgdevdbBh4FGmMhjlqtH4Pz1G/aclRM1PPdtPURAEQRAEQRAEQRCEM0gku4SvDVeFi+NraolzRbF523RGucvY5z9OcUk90d9q4L20ar4r19BcdQKbuajt72avOMjRbY2M+byUofmJNF5awKVHB3JR8TAODi3AVetn0ryhFO9wEDFHRdaL6i5BEARBEARBEARBOF+JZJfwtaAEAhx6/ygG1UKq7v+wxL5AatF/ue0mF9Yhd/LgmwojnRNwvXklDYlDMHgHAFMAsHsnY8/ajzLwnzj2NGP8cDTzp3lpMvyX9KpjvJcylNgCHwMyhpH7fhVDLo0TCS9BEARBEARBEARBOE+JZJdw1qhKDQH/ZvTGyzm06kt0DGRf5ct8Z8V0pqW9y+/SdmJIb+TxLXVMOjqSrMHfQUq0Ei1JNEnutjiGRBeDx+cQlpyBuyadoi3T2LD9RaZPD/CKKZzF+/cTNWM3+ueW4r54EgXr6xgwN0r07xIEQRAEQRAEQRCE85BIdglnjd/7CgHfBmTdQhy1iTQ0BPhbZjQBXQM/Sd2FIb2RP+SPY27hVYQPSwQg4HOCuQ5Vb2+LI7vNnNj8U3SxHxCd7EJ1LWLOnFQ2THydosoB6Gpm0LT9FVJ/+QL1z5hpkMZQsq2R1KnhZ+upC4IgCIIgCIIgCIJwhshneweE/116060YbX+lyWugWV+Ca+hbNE3L5+JclZRB1ex4dwWX7boJu2xHCTTT7Crj4NrLqLjsELb5f26L09wENe69BKqXUFmYhmP0WygVo5m941rSq8poMhkJ6MPYttZI3Lf+gTuqjqocF44Sz1l89oIgCIIgCIIgCIIgnAmiskv4ygX825HkNJASeflQLc7cvTROjWaLIQM81Uzbns7OLS7CUufi91WhNOfga9yG03iIOrvEx4f3khsZ1havwq7yr3EHmJp7nCX1V7ChcQfzR63EknMFs2LquX/KG0yQS/luTgxlDWXg+Q6l9bei33QxI5bHoTOKnK8gCIIgCIIgCIIgnC/EVb7wlVJVF77mR/E3/w1FBXPuCQ6tHIFuo4Fq5z5me2aQkDgFXdh4fEozBksSbjKom2/mT8uncvDaRWQf8ZNaEWiLudnyH6JrS3l79Ak2J+xnaUE2xf5qpIwdDLCYsXl95KUn06yPZH1VOp/MGIPz6G+p2PMfyvc4z+LREARBEARBEARBEAThdBOVXcJXKuBbhaI24jPeSJgMGU1m5GUKEQdeJvvLMMKVETQ2O/hEP5r5aeV8HhbF6oQcSg1mZEXl7m0u9DJYIrPZyL8AWLqrCbPBh6JXcEQfoTDSTurhqzm++2YWPmTlfl8Enr3fxRqXTK3peQ5WW7ni2lQOPvo4hrAU4oZfiTFMfBQEQRAEQRAEQRAE4XwgKruEr4yqBgh432eL7nYe8w3haGEZ+4dOYOu8gTSMnonduwRVlTk2VI86swT7iVTMR1/hBysPkRgI47mww2RdW82W8SN4fuDhtrjySDPOmHhUVCIqDxKV+xJUfkjkqB/wcVkkSe5osgw2YlKeY+T2z/jhO262BBrwT8lglfwvCj4rPotHRRAEQRAEQRAEQRCE00kku4SvjBLYRqVq5FPpEsyVJ9j3zpdUhdUQKN5DfM1IJF0iEjIN/mJymrZTVfMUIwvr0afG8N+wdaQ31PDqs7lkyf9ima+0La4/pxmaTZQPGUNDXCYqMkpTLobSTRjeT2DtlkhclUYauZQh34ykIVBO/d5Iji8eysdXjWP7Z3/D2+g/i0dGEARBEARBEARBEITTRczdEr4yfu97fKT7GSYkFm94mVJ8jFx9GEP4Uux+KwAFyXsZvHoHY+xV+Ewgz3YzfeBOcteEc6zIyaxxEcSoFsLfac/TViWMpcFUjMtjwHH5j4hu0jPpvXdQGzcT7tYx6WA9TerzSBXReIsvJemizzF9FsW26u38fNMkpN3rKdx8A0MWjzpbh0YQBEEQBEEQBEEQhNNEJLuEr4Sq1JCryBzXjWZpcyUBVz7uwd8mz3ec+dXRKCg49dUMKC2kMbwCZJUBU2uIG1XHRzubiM9NJEmOwXtIogxQ/e2VWBNHlhEemYtsPkytYS8fF67AGjadZiUZS1Q6mwZs4GhgH5ftN2OtOIy/IooIu4eEA5kMGtTA8U161m7/KykX/Rmr2Xj2DpIgCIIgCIIgCIIgCKdMTGMUvhIB/yYKpDHY8FG58RN8k5L5y8D1jHKOwiv7CcgS6tY3qS/aC6ofg9VHSa2Zu/91kGfLjnEweR1KdgMAhigD5thFbbErdu7n+AcJHHszmePP6Bjyzqscdv4HS/QgGmU/kwunML5C4ZFFCv6BP8CdNgqzBUY7MpBqbeTOmsDT34zmmSOrztbhEQRBEARBEARBEAThNBGVXcJXIuDbyHx8RFZfgLpzI1u9WVxmG0VScxQAesuLNMVVI6lQNSGO8rVrWR2VzzdnJTJvt5/w0deh/7gQc3QEOvuPcFrK2mJ/svQxEnRNpH25GUOzmwilEGtzNUWjfaQXW/E15TPkhJ/BQ+t5aNKHGKxV/FMpoD43kpov45llDEN98XMGI6GOWYokSWfrMAmCIAiCIAiCIAiCcIpEsks441RVwa2bTpgukuz9f+efF6UxpmAhYY5wVBSU+L1UVnqwqLBp3gykcgcbxh3novgkFg6O4+9qJpNXFaKLA688DYNSz44sU1t82RyGVGzAefFE3A4fx+pmYmxy4y/7kGTdeHz1n2KNWEylNZcaWwXLjs4k5/Ay3oq6h8UpCjGlCUxTBlB/YCXOsrsJSw47i0dLEARBEARBEARBEIRTIaYxCmdcAzJ/VK9jXWAeX0h5OLzjCQ+EIyMh690Exr7Gidj5FIy7Gr8cR+yOT5h7LImLs8birDEz/mM9DWaZZyJ2kHRNFtk3jWRR3ra2+OFRUfxgVibfjI/get2zzNn/FBO2vMaNsz8gM/sfgIKn9j1+strL9IPNuNUKwqxxXOX6I4cP5JAyuwyrbMI4YDDfLP8zud6Ks3ewBEEQBEEQBEEQBEE4JSLZJZxxW701yE0ulKfuZ5M8nrknRoHkpybiCE1VT/DppjD+OWgVNWWbSH76Vi65uobrb8hikL2K3DfTMSoWwm1Xc23eeOI9VmSdxKAl17TF/55uB4r7BnxNdyAnVJM2s5nYcQVU5QRwVzeizkvFGDkHgyOPFVvdjDiyj1Xj84gxpDN39HdImughfmIDsfZk0gqqqPI0nsWjJQiCIAiCIAiCIAjCqRDJLuGMCihN7ArAgNojuCwqSY4h6FQZSTVwrPFjfA6V8YUwpWYoSV9UMO+mOGKGWKk/tpzNLw3C3xTgSFSA6BgvU79zN7aktJbIzrZt6OQ4ZDkJp/Hn/MPwJn/MfI/8ET+jLr+RKoeOqr/+hTJTM6a4FciqTGZ1gPKatZyIzEdSr2PtXY00m/MwR3tZvrWBE883n52DJQiCIAiCIAiCIAjCKRPJLuGMKlBN1EuJDP70IM8tTWd/ihsVaPTsZWR+A2rkJBpTF2IbcgUDrqsmeVI4lfvnUrZ3PRaPj8ZZE/n29x9m9HXzicgYjCRJqKqCz/3Ttm3IulSM1j8Qa5rHD8025uoDbExcwsZLPsa/N46kMcNxffQrPkvLxRz/LQyRNpZvbyKlJhEVCWv6Hex4oZSMBZXojRb0x//FuqITZ++gCYIgCIIgCIIgCIKgmUh2CWfUnoCepOP7yYkro8RfzExrMXKgGX/VKsqj9fxxgQvHvpVc9bvfMWZmHY1leo7sd2L0qkRdfDlL596EziQR8H2Br/nPqGoASZLRmW7pdntGCRYYFG4xBnDHRHFg5PUUrd+FFGbDFRaLb2AAvfVaLF6Fw55neGPwVpyRLlJTwwhP9xIx2ElkRCMP1D5Dg9/9FR8tQRAEQRAEQRAEQRBOlViNUThjVFXlosoX2PfefvKjdCwumkZaWToNyiF0ihe7GkdWTQzDNkeStXI3igrHPhuApbSK6EXRDBm7EY/rTVTlBKCAFIHOsBRJNwCdflKv287UqfzYHMB0STab3xpFVe1WJn/0X3w2M+aMh9CbY0irquEzz1Z+v9DAxc+bGJPXROrMWuqP2fjRZjv20eav5kAJgiAIgiAIgiAIgnDaiMou4YxRfMWY7K/QKDnYP3ggV+dOweJMojK5FFkFf0QW+ik3EP9gCYRD6efR+PMDrJpgJmZUJarqRpLT0Bmvw2D5Aybbq8i6ASFvP1IGi1nP6N8/gy9xKGqTG1N5M1WO1RijrkUFrtrSxJKDo8lbeCM71ozAGucnOstJYmU+qs9z5g6OIAiCIAiCIAiCIAhnhKjsEs6YT8pr0QW+wceLapl8bByK4kXxFGIt3o/XYKJh78dcWPI5aT+po2BNDFV7Ijkx4UIuXzCDRFta3xsIkT4tmhP3/Qv57oXUuH04rrqasONlGMIzwFHI1IPlDMqYSXjsYgp3/YCEifXUHQnjb689T9L4pVw19vTtiyAIgiAIgiAIwnlLCaAr2YnkqkK1xRFImQiy7mzvlfA/SCS7hDMioMLB5jCGv3ec6tlhjKxLxefYhK9hE5HAthgng9YUMvw3PmoPxVG5J4qyIWPIXfANllmV07ovcTqJ6wek85+fP0fEuy8gff4BAZ0RnT4LCRmj4xh7ojwMc9zIphPXcf2E17GmNDGw9CAv6oaLZJcgCIIgnGUBJcD+ii+pdVcSbY1ndMJUdOLiSein7t5HgiCcPrrc1ZjWP4zsLG/7nWJPxDPnXgJDFpzFPRP+F4lkl3BGHDxWi76xEdxOhtSOBF8DvobPUGQTPr1M3bIHcGy+kNqVZpyRaZTG6PnPdBvvWBSM0unfnwxZZenYafxn1EwG/epSqqyFDM26jOaaWVR517M9vpoJtQcY1jSH7TXrGHpBHe43LCS9byF/psqAzDOwU4IgCIIg9GlzwYf89ctfUe0ua/tdrDWJH079HTMzLzmLeyacS3p6H90970nGx846i3smCOcHXe5qzO/fDqgdfi85KzC/fzvNlz4pEl7nETWg4NpWib+yCX28BdvkeCTd16tL1tdrb4Tzxp6mSioHjeadq7MY7hiAp241ICErHmp0JtI/eZLxKRL+2MnorHb2LPse34ybh/EMviNHm2F2qYvKS36Ma1c5tkEvIxtiiA6k8Zvt17Avvgm738ymvCWY0j2YIj1ckLKON98OnLmdEgRBEISvgBpQcG4pp/7dfJxbylEDp7eK+kzZXPAhv1l3U4cEBUC1u5zfrLuJzQUfnqU9E84lvb2Pfv7eNeJ9JAinSglgWv8woNK5REBqSX6ZNjwMiriu+qoEAiq7dqus+TT4/4GA2vcfhahhVSGHpr5F/tWrKb5tM/lXr+bQ1LdoWFV42rZxOojKLuG0U1TIj4lj3FtP8cLsSjIqBxBoOkKzyUyjEXybPmXZnUmUBibTVGZA/sZtyHGZzDP6z/i+LRpiIrFxLId0BnboJzEu9S3KvzgC9RvIVMI4pLgosnr5Z1E0lw134Kk/wUuD/8IPnD/Bbhe5YUEQBOHc07CqkLIHtuMrc7f9zpBkJemBSUQsyjiLe9a7gBLgr1/+is5VAkHB3/1t631MT18opjQSvLDZuw9qaiEmGsaOAZ1OVKb3/T6STtv76FyodDjbxJTk85OuZGeHqYudSahIjeXoSnYSSJv8Fe7Z/6aNm1Se/ItKVVX77+Li4PbbYNaFp/bvQsOqQoq+t5HW82erQIWbou9tJP0fs7423y1Esks47SryjzLh/f8SqKthfOFMjAETXkMiZk85OwYnM6J+OI5tHhoabbyWHU1MaSnXJqcR/xX9Ozd+QjS7J/wK3+8exflgKubCe/HW7GFYcT4r531BTWkNr41J5yJLHqqssqDYQ50rIJJdgiAIwjmnYVUhRd/f2OU631fupuj7G0n/++n5UnomLmD3V3zZpRKnsypXKfsrvmRc0gWntK1z3cZNKk/+n0JVdfuFR1ysyu0/lk/5wuardCYSdn2/j9TT8j46V5PKXyUxJfn8Jbmq+h7Uj3GCdhs3qfzy/q7J/aoq+OX9Kg/9VnvCSw0onPjp53R/8wBA5cRPPyd8QdrXItEvkl3CaVfw3DPIhih2ZcazLH8yOosKuiXs0b/FoOZK+GEUxQe8HDbFUzx+JKkRqYzVn76yylBcdM21vPxpPZ5f/Inhk1/BFv8Q7pI/M2ubxKcZyaSXjOGtxAauGeIg+3gJyVFf7f4JgiAIwqlSAwplD2zvraCFsge2n/KX0jN1AVvr6rlKQMu4vqg+H75PP0WtLkOKTcIwdy6SwXBaYsOZq2gJXtgodL7LXlUd/P1Dvz03El4bN6k88Rc/tfJWsFRAUwLRyhTuuE1/Svtf6648reO681UlleHcrR5rnUra+SC1Tkn+9UXPiYRXi3OxSlO1xZ3WcWfdObqiZCAQrOjqzZ+fUplxgbb3VOMXZUg+J2CCbiasgorkc9L4RRnhM1P6Hf90E8ku4bSq23MEZ7iOfWNGE9BVEdj/GXL4NN7LOoE3p4Qboyqpzh1IjTeKA9dczuz4bJYbAvScHT4zMkfYSfl3Na4RC/hsylIW1+xGLXIR7VCIsk/i8mNzeNrkobD+JcJ94Wxa/ynDxi0gKeHrf5ITBEEQBADXtsq2KhNFUjiekY/D7iDcGc7AwgHIqoyvzI1rWyX2aYmatnEmL2CjnbWndVxvPK+/TFju/xFmdQR/UQneHeE0Dvkxpqu/ccrxNxd8yFNf3E9lQTo0JYClgvjMIn40/bendIEfCKg89Ec3YKZrK14ZUHjoj25mXGA9hb0/aXtnMGF333MfYrjwPoy29qRpoyuJ+557kAe5RHPCK9oaf1rHddaeVO6YbAw+GPyf05FUhmBSreSB7RyqC6NBZyEi0MTwqEZSTmP1WGuipdnjwWxST0ui5eSppLIK4/x6YlSJGkllj96PIp2+qaTnujM5/exMCqRMRLEnIjkr2np0nUxFQg1LCCaNToMmt5cXn9lJWSkkJcO3b5mIxWo8LbGDK0r+HtlZ0fY7xZ6AZ84vTluD/TOV0Ny7jw7vne5UVgbHTRjf//j1b+US8Jp7GSER8JqpfytXJLuE88+htR+QO3gy9UVuxvlP4HVU48dPoWc7F3+0Dr41A7VJ4c2hSWRFBhgmKwzSnZ2qqbnXX8OGuy9nxKU+Moap5BwehuotZ8yJYpr8ZcwvGsO6S4dyycteCo59xBrHYH5/1ZCzsq+CIAiC0F/+yiYA9g3fz8oF7xNTH024MxyH3UFNZC1XrL6UMYdGt43rr5MvYFVFRqmc0pbIkeO3IsnqKV3ARrsbiFckKiW16w1kABUSVIlod4Om/W/lef1lok/8Diwdf2+wOIg+8TtqX+eUEl6bCz7k/hfeJLDtbSZYy4g1V1HdHMeuzUncf+QBfnsjmhNeO3cHcDdaehkh4260sHN3gIvna9v/VpsLPuTpz37JwBMBLKqOJinAo6k6fjDjoVNO2P3xjQ8xzLqZLjc/reUYZt3MY/99lhkXXKLpYnBU7GQinZHU2+qR6SbRAkS6IhkVq62PUHtSuad9k045qQzBRNe7Pz3OGzEXU59sa/t9pN/Fip/u4DI45YRXx0SLEzg9iZbWqaSzvQZud9mwnMjA57JjsDlpSi3kSZuLDZyeKclNHi//WPsGJbVlpEQn8b15K7CYTk8SpNWZSlS0Tz/rVKVZpfLL+zml6Wcn8zuduP71O4zuIrzWdGw3/Aq93X5qQWUdnjn3Yn7/djyqzJv1YznhiSPVVMWVkXsxSiqe2feeluqo3923hdWbs1GY0va7V98JsGDmFn714LRTih1cUfInXR9wVmB+/yc0X/rnU054BT9nClVVJ007j1O5/bZTr8KtCfHeT6jjOjM17gZigj9ICuEpxRhsTnwuO46SNFDlk8bN1raR00gku4TTxudyUhprYfTetZjsUaQ469CHTcPbuIX5m6tImZxIdW4EgQjImV1PqsnDUuPZWw0qbeR4whIGU/Sf3Ux5KI2oUUZqd0FEoJ7mMTUMPjyKl+uGU1H+EUOaBrJ2qwn/FSp6/df3roogCIIgtJJjTewbvp8do3fz4xduJcoR2fZYXXg9Kxe+B0BG7DxN8VsvYAOFi/Bv+y2Da3VEBJpo0Fk4Fh1AP/l+qjJWab6ArdNJ3NFk5V6rq/O1X1tO5PYmK3U6iXRNzyA4dTEs9//AAlKnf94lKVisYz/6F7y+FZqmNAaUAI+99Akzj17NXTNuINHSXilQ3pTAozn38Nh/PmL6L7QlBFd9kQtkhTTu4vnapw9tLviQDa/8AqJrqHKltyVNcRYFf3+d9oTd7r0BmrLuA1T0isSi0lSSvGbKjM2sSj6BXwb30PvZvXch2RP6f+nSvKOGKz68lPwrXuOOJisJant1VYWk8ITFzYAPL6V5Uo2mZJS/rLHtv3uqoGwfpy3ZpQYUPvhtCc/EX9jlsXqdlWfiL0T3261cdwrVY2eyz0+tu5LZXgM/3zeBgg3z8TrD2x4z2h38fPYaGLPrlKaSAvzq1f9jS80/kazBz9nuCnj/2T8yLea7/O7aH59S7FZr1np46PcBjKX1GFU/XkmPNzmS23+iO6VERfv0s24qBFumh/35KTRPP2vlevw7xLOFSAnQAZ7dqP94l0qmYfvZ85rjAgSGLOAXnt/wyaZpKE3Jbb9/3FLKxRdu4ZenoSrqd/dt4eNNLYnpkw6DosrB3993CgkvJQBr70dFRe70GgTrZFVY+2sYNFdz0q592nlH1VWnZ9p5TPTpHdeZ4vYCED3oCJmz12AKaz//eRrDKNgwn9q8rLZxZ5tIdgmnTeHqTyiKmcxHqwfwzRGbkPSRmCJn89wFjSx46GPsi2fQXAd/zfs9s5Is3Hia77JoMel7v2DDr6+nZFcsqVM3UXNoMqayWo7HB7DqAiwoGsOuhdtJ+UJlinE127Z/l+mndsNAEARBEL4SR1NzOTj4KDe+cX2XxyIdEdz4xvW8vuRtslJzySa5mwi9q3VXEihcxMiP7mdF9VaiAu2NuesqrbxRfT8HFmvvhaSmTibjXxu5MS+rS2VabWQtl6++lPRBRyhbpn1lL9+nn7ZNXfSjsFcfoEZSiVElxvp16CUZk60B16efYly4sN/x95Z8wdijs3gs+6d0rlqKN1fyWPZPuXPPb9lb8gUT0mb2O/7xsipCSXYFx2kTUAJ8/tqvqa2N4wcvf5fIxsi2x+rD6nl70Xt8/toDTL9LW8Jub9lWJFsZ3zo+iBuNHixhTlqrin5WlsILXhP/HpjH3rKtZNP/pKm/sokrBx0h2m3r8licKvGw20btoCM0aaxwNDd+AUjsG76fdy5+jwE2FzGqxAFJ5RWXjcs/WcqYQ6NbxmmbIeD4spJXdaODP/SQlX1NN5olX1YScUH/E2pnus9PtDmWW/aN5+gHy7o85nWGcfSDZdyMStmlsf2O3epXr/4fW9wPo5oklPJpbVWmUtw2trgf5levcsoJr42bVJ67Kx+P3kaVqT1bEF7h5tm7XPBovOZERfv0s54rBE9l+hm0J7q6E88WKh//ziklvH7x5HY2fdL1NVaaklj1yTJclu38/vZJmuM3ub2s29Ty5Hv4HKzfNJ6fub3apjQWb8PS1ABIBFSZXTUTqW6OI9ZcxYSYnegkBUtTPc7ibZDR/wvCQEDlz080AUYkFQY3V7TfIDLHI0nBx2dcYNGc0BwzUsWmenBhREJlcHNlh22oSNjxMmZkdz23+iapzUQPOsLQJW93ecxob2Tokrc5+sEyJPXsX+eDSHYJp1FdfiEby0ZxzbwnMeSHoSYuptLiJPJYKckXpNNUYWG3PJbAYjfDJImh+t7m+341kiddhDztUo4WG0jLzkUa4IWDfgYe2sPeCVHElfl4fnYaywtLGK7bxov7ZjN92uCzvduCIAiC0KdVW3O4eONcAKROX2olJFRULt50Eatm55CdMavf8SPN8Yxcexu3VGzu+ljAzS0Vm3lm7W1E3qItiTAqfiq7N13MGIed0YdGdXgOKioSUFSSwfg/T9UUH0CtKgFgvcHDE5YmKuX2C/54ReKOJgtzfKa2cf21e2Mxdw/7B6Aid7qukCUVRZW4O+spXtt4CxO+2f/49vDm9ngEmBCzs32aZM1EFHRdxvX7ORRtpLkymhvf7DqVM6IxmDR9+8qX2V20kezMi/q/AXMF3zo+iB9EdZ1XY7a5+IHNBccH4cuo6OaP+6aPgehBuaBKKJLKHr2vLaE5zq9HUiWiB+VSEaMpPGEpxRwcX07+onf5p9tCmcfcFj8Jlf+74jX0ZjcTUpLwadsEe3cp1Ou7JuvaSBJ1eht7dzVyoYZZgB36/MhedENfRAorRG3MIHD026AYTynRMtaj4+iG1qqe7quW6jfMZ+yd2qplmjxettT8k0DVQpTtDzDBctJ04aYk5Em/YUvcP2nyfF/zlMZAQOXlByrJN3VNyDl0Fhw6Cy8/UMmMNfGaEhVl5QG69t3raVz/L+H9Tiex/u2g77mKNda/nUanU9OUxqZmP5s+GhSMp6pdEjmqJLHpo4E0fd+PxawtBfH80zvwSr2c7yUJD0aef/pLfviz6f2Of6J0C8OAT0vn8WjOvVQ0tyeOE8zl3DXqYeYmr+VE6RZSNSS79u5VqKwzM85VxJXVW3HGKjiMZsJ9zdiLZN6MncIe0tm7V2HCBG2fBdf2SlQlnHFNRVxZs4MaQ3hbf78Yn4M3Y7LJtSTg2q4tMa7q9GTO/rDlB4mGE2ltU5LDU4pBUsmcvYaio5dp2v/TTSS7hNOiqbaKz6pjmV27hibLOGoTrcQ3luBp3sbMNdvRzxqIKa6JFy6oQxf1PDeZf3m2dxkASZIYtfRm6t96jsIKD3LcTnzGseg8hQyX17D1+Ha8c0ZxZKSBgaU6fMaVBAJ3fu1XRBEEQRCEwEYXUY6ev8xKSEQ5oghsLIcV/Y/vLR7PitKPWmJ1jh2sY7qq9Dje4sWgoU9t844azI6wHpawCbZANjvCad6hbfoZgN7kZL3By71Wd5fHKiWVe61uHnZLTDA5NcW35jeRYOu5sk2WVBKsFVjztSUEp4+LY+8ncFHSGu4a9XA30yTvZV3ZfKaP0z6F8dDq57n4wyuAXpKmH17BoWHPk31L/5Nd4wfHMulzTzBeDxfhNxo9bB+sreonuvklJAnWG7w8YXF3k9C0MsdnJLr5Jbzc3+/4gcgUyuf/k4t9Br4f3thtwvSL+R8RiOx/7FaVpaFNCQp1XGdV1cF91k14EP2IfyDJ7dOs9Nm/xX/wewR23dcyrv/fgZu2VrRPXZT9JI7dhTmijuaGKMr3TgBFj9cZQdPWCqwD+7///1j7BkrVBC48ejV3XdD9dOHNqPxj7RvcfomGrDKwe6dCgS88mI/q4Y1a4Atn906F7Mn9T1R8vrcQGBDSuEsWDep3fOcLDxJp8Pf4uCSBzuAPjrvtD/2O/7dXDoB3NONcRayo3tGx0ldn5Y3YbPaQzt9e2c/PvjO23/EBTnwZ2nENdVxnNbLKp6XzuHPHE13+3alsjufOHU/wWPYdmGWVVC3xjx5nnMvATN1OHk9dRJ2uJalohCizk2/61oALao76YIK2KtC9uxWGNlcyyVnAYykLOyTJI/0urqreAZLE3t1GTYnx6LhdmMIaqTk2tNspyZmz1xAz+CjRcbs07f/p9vVfp1Y4J+x79s+UlKYzISkH9FnEGKfid+2n1qIjfrYev9tAw9jBKHpYYB+OXv76vPVGjh6PbLHx6Yte8p8/RIVUHHxgax31K37KuHw7LybEgyHA1Qf34hV5LkEQBOEckFAVWqlKqOM62/Xhp0QF3L1MuoHogJtdH36qKb6n1Im/5atqd8k0AD8ynlJtiSiAGmcpT1jcHYN22siTFjc1zlJN8Qd4Q+sCHOq4zgaXRTM/5R0ey76DeHPHyqfgNMk7mJ/yDoPLNDZoAUxHTdibbF0SXa0kJOxNNkxHTZriDz/gwGJ3dckftMWXwGJ3MfyAQ1N8uaGoJaHpogqVIX6ZUX4dQ/wyVajca3Wx3uBFbijSFH9bTDoTVYl7re7gYgonaU2YTlAltsVo7SwHas3B0zqus/r6lkTXyKdRVZVA7QiUymwCtSNQVRX9yKfRTXiQ+npN4SnOD17Up1+wjqk/eowBsz4ladwuBsz6lKk/eoz0C9Z1GNfv+JVlzMlbzGPZPyXGVMX26kmsOrGY7dWTiDFV81j2T5mTt4jiyrK+g/Vgz3t1uHWm4BtSCiAnfIGcuRI54QuQAiBJuHUm9rxXpym+Wy1HspQDPfUzVpAsZbjVck3xlYLQppOHOq6zI9sVxrmKuKViE5GBjjcPgpW+mxjnKuLIdu39msP8gbb/llSFIU3lZDvzGdJUjqQq3Y7rD1/ceB7Zf2+wa5pKh/it2a8/7L8XX5y2eaRKbQUX6HfyF9Pl1EuWDvHrJQt/MV3OBfqdKLXaqlgBGiQzI5oLeDbxQur1HVfhrddbeDbxQkY0F9AgaZxh5fNQc2woRz9YhtcZ1uGh1inJNceGgs+j9SmcVqKySzhlSiDAnuNplDbq8VljGF07lED9Z3glCcveHNTUVMJTG3l2xAIk5/v8OFpbI9wzRdbpsKdnEPHyqxy7/rvM2J2AYjmE7HczaXs1y7mBp2yf4Mz6DNt+CzXb15M6RUOZviAIgiB8hcLV0O5uhzqus6rSg0BSiOMu7Xf8gmOHMPZ44RfMRelRKDh2iBi0tRj4vHEXlWG99CqSoEJS+bxxF1doiB9uDe3mXqjjOjtc+Tm/HvFnepsm+esRf+BflT9hIeM0bcNeGVoNQ6jjOqvO30Vr3VmA4AqJJ08z1LUk2arzd5HGkn7Hb7In8YTFzUi/jnKdQq6+/T0Vo0gk+mWetLjJtidpqFmCvC/f5l1LsDJPUiUGFgxo6y13PCMfVVJ52tLEZV++TfaAuRq2AMbmY6BzQcCCLPkZOvglLGGFNDVmcPTY9SiqHvRNwXEa2O1eDFkvEqgaj2wvQY5uT5qp7ngUZyqGrH9ht98N9P8iuTx8BOMueJnk7K0EUNl70ms81qcnOXsrAHvCHwqhA11XdUVJ/Gb4X/m0bC6PdTP97M5Rj3DHsKf5SdF3NUQPyj0anAosp3+EYdKvkGztSSfVlYhv++9Qiha3jes3736kmHrUEwvpfkUOCSlmD3iLQUPvuqLaAaSnfAnQYz+q1nFaUuNqYxgrqncAvVT6Vu/gP43jNEQPGmGS+MLvIrO5mitrv8QzfiPOqBrsdTGYds/izeipFJpjGWHSVplw8AWJKk9in9VpB184QvaD/Y8fd7icp/QLGOcuYmn1NlanJFFnthMTKOHa4s28FzuZV8zz+M3hNZr2HyBe7+DvMeMBFVRI8da1LaRQYowESeXDmPHcr68DDa+04pco2ty6tG/3r3TBhnmkz1yt+TmcTiLZJZyyiq37eL9iMd8e+iwmr5dmyQnO3Xg8dtLiFBSfzBvSaHYe2sWcrOEk6iPP9i53kbHsJo6t+jcT9r3H1F8lcuB34/CYHcSWH6MhbRSLjg/n/cF7uDbHy+OHt/LTYXNIixAlXoIgCMLX145oF0NNRkweY7dVOSoqHpOXHdFeunZj6lt95XCgPsRx/Xdgq5dQ7p8f2OploqYtQJ7adfriqYzrrKl5H5hae4x1/xpISMFxGiRU7MGe1nPFkyyp2C0OEor3aIoPYDO1VyQpQIkhCq+sw6gESPHVtU0TOXlcf+QeKWJ4Zt/TDHOPFJGmIf7jdeXEBWQO6LtWe9RIKjX6ACP9Oh6vK+dODfGjq7dQaVMZfXAUV3y8tNtVT/ePyCG6uvvG4KH4yF6GftKvGdkUTfWov1Oga5+OljT5N8TmfJ8Dljo+chjR0innQNnfCNQNQ47bQ+eFFLBUIVuqCFSN44Dhb1zCT/sd3zbBRlLzNjYYvDxh7eY1dluZNXEbuRN66UvWi+mVlRywjOauHqaf3bXjTzyWfQfTm7Sv9uhVgoku46ybujwmWcsxzroJ78bn8LrHaYqfQhk7SlqTcd0nEZSS+aQMeUhT/A9d1zGp8QM2N07hj90kBH8+6mFmhm3lQ9d1mtLi0x3H25JDChLHzPFtvaIGN1cioxIdcDPdcRw03pyIGPUh395XgT77v/xjxgaqTvpMx138Pld/Nhv/jquIWLgD6H9PrT0n6tqq0zprrU57JuFC9pyo41sa9r8gRyajuRa/uZoH05ejSi1nTz1sTVcYqewmo1miIEdmgob4AAc+zaGBxWQ2V1Grt1Fy8kIKfjfRfhcF5jgOfLqZKd/L7Hf8qsPDOkxd7ErC64yg6vAwzaskn05fn7lkwjlJVVXefjmPefGfkmCqwhR5EfqGHEAh/HgDATWOgFrDlyNjGZ5p4sdR8/uMeTZEWq1kLvo2jTk1VDn8JF2bg9ceh4SES+9hsDuD4+Y03BEq2Y0VPLVeewmuIAiCIHwVomolAmPLaO1vdbLgz8HHo2q13bwp9w2jTmftoadWcBu1OivlvmGa4itVoVWchTquO74Q2yqEOq6zCqrYrwu2JVc6HanWn/frfFSgbbXEZKWx70H9GNedUn8EiqRw3BiDQ2cmzVfHIE81ab46HDozx40xKJJCqT9CU/wDtZGs8eu41+rqYRqgizV+HQdqIzXFlys+p1zX8r2th6mqFToFueJzTfGbvDKjD47ixjeuJ9LR8Ri0rno6+uAomrzaL7sqoyoZE7OdI2Oeokbu2HepRvZzZMxTjInZRmWUtmROeWk5sr0EUNEBE3x65nsNTPDp0bW8T2V7KeWl2qbQJRz6KRtNHu619fAa21xsNHlIONT/RBpApJzHozn3tu5ph8fUlp8fzbmHSDlPU3yApHAXEdNaVnPs4X0UMe3HJIW7NMXPOzYNVF03wU/aiKoPjtMgckgO/954Bz/f8QQVzfEdHqtsjufnO57g3xvvIHJIjqb4gw35AOy2pvHL9Mt5Ink+zyfM4Ink+fwy/XJ2W9M6jNPiiLkYw5RXeWrWp1TpOiavq3QBnpr1KYYpr3LEXKwpviM6v9fqNAhWpzmitT2HaqMXr7WKHHlCyxIr7VQkcuQJeK1VVBu19d4DKFF8ZDZXUWCOxdFpMTiH3kyBOZbM5ipKFG3LZfhcoVWDhTruTBPJLuGU1OQe4cPS2YyI3QmSiYbwBPyNO6mPzsQ6uhJZB6a4YayZfiNPplzDAKP2Bqln2sgl1xKemEXRy2XEDfFTE16FFDaChDonASnA/JIxFGRJpFd40L3/NFUi3yUIgiB8jQ01F/JH5008kzCTetnS4bE62cKzCTP5o/O7DDUXaoqf6nXyRmw2AAFJ4VhmHrtG7eZYZh6Blikx/43NJtWrraeWxefHq/ei9pBOU1Hx6r1YfD03Xe7LBV4PEY1hXYpZTtoIEY5wLvBq6z9SEVD5pc0V7BfVQyLnVzYXFYFeplL2Qg4LrZdYqOO6c8JdSYEhjgHeGsIUd4fXOUxxM8BbQ4EhjhNubYkWk8vIk/aWZFxPfdNsTkwubavoJeCnRlZ7zSFUyyoJaHsfRXiNXPHx0pZQXRv4A1zx8aVEeLXtP4AUVkZFZPACW6dILDo2hG/tnsiiY0PQKcFtVEbmI4Vp60lVXh6PZK1kjs/AyoYI/uYK40G3nb+5wljZEMEcnx7JWkF5eXzfwbpRUeBt640n0zGZ1nox+qTFTUWBtov8PNeAlkql7i9tVWQqmpPIc/XdAL4nEbb1NJncIIGkSAzKH8j4/eMYlD8QSZFAgiaTmwjbek3xXY6+p4T3Z1xnSmUdz/iXtyRZuiYEVSSe8S9HqdTWc0zvbWS3NY1nEi6kQTZ36EfVIJt5JuFCdlvT0Hu1J95TT0Tw2rTPgj/0cK54fdpnpJ7QlnhPtRWF1Icy1aatv1+MsZRDrfXK3S1yABxiPDFG7efrpogCGsxGgv+odf48yEioOMwGmiIKNMWX9KF9RkMdd6aJaYzCKfn3q1UMt5zAovNjjrsad/1mUAMUxHiJrI3HlOJh++gr+FfFv/lDnIalnr5C5sgYohZ8g5rV/6a2OZxZM6o49OYaFF0kpF7P5LJMHpgYyQPb6sg2vc8nhd/hmwOsfcYVBEEQhLMhvDmOiuZExnpzUCWVo+aEtmklsb4G4ryN7G4eRXizthtRcxu38zvrTTw+NUDjtD/giKhveyysIZLwLXeTV5HGrxo/AQ2TMiYkruffkwq46t0V3XawAXj3kpVcX5wJXKnpOcTsnMqyWhMvrHip+zY5wLJVlxETrS3Ztd2VRqW8m4RjWez8ZAkZYY0YbE58LjuFjWHUXPwBFSNy2O5K40YN8RuKB1AVfYQYgpcxnSmoVLeM06pBKiI6EMu+4ftZufA9GiIa2h6LaIjgio+XknF0Mg2StgvA/Mgaasy9HF8JaizN5EfWaIpvbrRDWN/N7c2N2pqjR+cNhJOmLnbWuuopeRqWGWyR5VfYrvNz7b6xTFt9KYGTphEtsjvYsuB9Xh2zl0l+bXdivUo9s70GHnZ3nUYYj8TDbhv34mKPUq8p/kFTGZWyymyvgR+77ZRUT2rrF5USu53/szrZYPRx0FRGtob4FZ724yGpCoObK4kINNGgs3DMHN82Xezkcf3VGLcSoM/pqsFx3+93/LjwfeQ3jA5pHPR/NcOjNUPwyb1d+kv4ZD1Ha7StAvhxtI8dsZMY5y5iRfXObvpdTeT12ElkR3+pYYJhkFMp7jB1sQsJKvUBnIq2yq7UEBN9oY7rbKNnWPvUxe5IEioSGz3DWK5pCxBxIoI6ek72qcjUEkmExoSgJ6yh70H9GHemiWSXoJmnycPq3MkkJv8VQ8xi6qwyloIjNMSOYOzBQ2Dy8UTYEA5FrGKQP5zwTneVv45iL1xMYP8Ojm+OZuzMHSjNRlRjJSZHAcdH51IdJlOVCmrYYN7b3cDVmVYMonWXIAiC8DXk9IWxoC6HDE8Nf0xZ1O0S5AvqcnBGhvUSpWc+jx7D+H9yYsqvodOkDEd4A46Lf4Fhqw/fl9q+bh4yNuNwjKC3kpwGxwgOGWvRdnkGBY3RjDk0kBvfuJ63OyVyIh3BRM6YQ6MpGHOckRri62Rv2xQ3AEdD+9SOSFRufON6XljxEn5Z213wfKOT9VYXD7ttBBRwlqTjc9kx2JzYU4qQZfiT1UWmUfuKlQHHIIqGbufFFS91eawhvIEXV7zEjW9AwDFUU/zw+OPtP/h1DNi7mPD6BByRFeSP/QhaLm47jOuHfbVjIbnvSol9tWM1XWC662MI5danu17bqqcACX6Za/eNZfLb19H5Uj/gDGPy29cB0BipbdKOWa7jHnfwWQQCEhUN6chyAEXRkRBRhE6ncrfbyo2ytov8Y6rCbK+Beccu49s591LZ3F6dFG8u465RD8PgdzmmakvWJUfnQAm9Nxa3pQfHaRQruTp8lk/WOl31hRUvERuhbRpjVtwnHKm+iIamBLqvUFOItJSTFfcJ0HUf+uQ/abq3FECO3wqWCmhKQKmc0jKFstO4fqjxDGaAp5pbKjZ3eSzY72ozzyRcSI1HW78uAMUa2qq1oY7rzNqYeVrHddbclNj+0vbyGjQ3JfYcpA+SxQ+t9w562YZk0VbJWqULEBXiOG3dOk8vkewSNPv7KydwnsjkmoF5KE06SuKaGGpMIj/BxLhqhZKSSirLv48y/iNui56P1NOa0l8jWYmJRF/9XfRWC58/tJGklSOpvukAsnM7aUVXMzbu36xPs7LiRBPjow5w3BdPllF7rxBBEARBOFMsaiNxvkaeTbiwy2P1OivPJlzIdVVfYlG1TSsp1kciT74NULuZkaGiqiBPfoDiHX/RFN+VWM7yDxYAzd1MDwtOY1z+WTYll/xbU3yALXob6cCYQ6MZdXgkxzPycdgdhDvDGVg4AFmV28ZdoiG+L/oIV716bcs+d53ipqJyxceX8sq1r2na/0qdjw1GH698nk3Wjmn4XO2JS4OtkSPZW9gwZy2Lddr6swB4VBNvL3yvdac7allmbeXC90hf/UtN8fW1w4H9jNx4Pcs3TyTK35r4G0rdqrm8NXMnB2a91DKu//ILLyRu2Opgj5/uvoqqEBfQkV/Y9XMSinJ/FKHUbJX7Q7lE7F6VI5IFq6e3JLq6fxGmrV7Cywu1NcHPjtxFJDKldakkRZ4gJbq9Sk9VoawujeSoE2RH7tIU31oew3jHSO7c8WSXxyqbE7hzx5P8SoI9bm09wbJi9jPeVcDNFZ91eay1sfizCTPIitmvKT6AoTaT5Z9MAXr+LC9fdRmFC7Zqiu9zWTBMug82PUNwKYiTE17BJKBh0v34KrQVDyi6YAIouJrkfUi29imvqisJ3/YHUYoWt43rL0d9JjeFsBrjP2MyNcUHKKnMhqFHQhunQbPVjMHWiNdl73FBEaPNSbO1/yuSAph9Cpj6fg3MPu29cgxh5VDf9zYMYdo+ayfC60JKdp0I15YYP91Ezy5BE58KG7bHcduFv8Voz8IbMYxkZQyGxBsZVbgdSQ9/y3sLzzA306PTmWXRspDw2RE3aiKB9Hjcso6mNBVdnQHFU4TBVcmPdl1PZdwAVJ3KuPefwvlx17ucgiAIgvB10OzS8WH0mOAPPfQHWRU9hmaXtps2X47bhyQrXUKfvAlJVvhynLaVBn3FE4nye7q96ICW6WE+D75irWsxgi7ej9HuAFRkVWZwwSAm5IxncMGglkSXitHegC5e213wqdVhRDkie38OjiimVmurrnN6E1myejEDNs7H6+o4Dc/rsjNg43yWrF6M06u9UsAXvyNY8dZLI5v6iAZ88Ts0xVfrJjB6/Xf57vpRRPk7TmeM8nv47vpRjF7/XdQ6beuTZRgPcc1ns1s21nnjwf+75rPZZBgPaYq/NzMPRVK6Cd6+EUVS2JupvTn6+Hpry9TFnl+EgDOC8fXa2mvMlYzBRFdUMXTqLadKKklRxZTWpTJX0tZ3LJMo/i/n3pafuvYRAngq5x4yQ7qM7upQ2UKuqg4m4npqLH5l9S4OlS3UFB9Ad2wIEc6IXj/LEY0R6I5pqzO1exScmR9hmHUzWDslIqxlGGbdTGPmR9g92hIhyxLeIDLz9Zb4nXq7WcsxzLqZyMzXWJbwhqb4Q7z5IfW7GuLV3qDefWgJqisJVe1+K6oqobqScR9aoil+slqLc2wFvS3q4hxbQbKqLSF4ieUzoge93OtrED3oZS6xdE3ahqo0whfSNkojtN0A2Tv9M+rC63vtpVkXXsfe6dqfw+kkkl2CJi+udTDS9RnJaiUFow0ctO8jtsmM37UDvVOHOsiN7/oPcC1ez8+jF58TVV0ne/9YHcbB85HvPkz6vHxUwFu3BoPfyuVFk8gdKGELT2Dj+ldQVW1NZQVBEAThTDrgzA5OXZQkdKqPJYYNfNewkiWGDehUH0gSdXobB5za7oJ7Ikra/rvbhs3djOsPkye0C/dQx3VnmNdA5uw1LT91f3mTOXstw7wGTfGbKkKb/BjquM6Gu33M2RKsSOqpOfqcLRcy3K29ssuaGdrUr1DHdaYzOLhm28CWve2aqpCAa7YNRGfou+9WdzI89QzccDG3bZxLXKBjYjc+oOO2jXMZuOFiMjz1muJHShUtidGeL/NlVSZSqtAUHyDGE9r7L9RxnakBE1FRJ0CVUICdeh+rDV526n2oBJMIUVEnUAMmTfErmwfR0JQEyMGpVQlfIGeuRE74AqQAIFPflExl8yBN8StKU0JKtFSUpmiKDxDm7ntMf8Z1ptjqAdBlrMK0bDKGBcsxzLwVw4LlmJZNQZexqsO4/kqV3Mjjf0tPlbgA8vjfkSppewKJUmhJrFDHdSfc58W3/cFgKqqb07WEim/7bwn3aZsWblfcPFz//eCiLrpOi7rorDybMJOH67+HXdF2jDJSjqCffD89vwYqhsn3k5HSd/VaT3Y1DkWe9EAv2wDdpAfY1aht2rkroGdlS6Vv54RX688rF76PK/D1mED49dgL4ZziVeHjVTpuTv0URR9HdEksMYX7CMTlozrWYUlo5tUTNShX5fBIwjIGGbWt3HI2zbLoqFHh8MEhjP7BHsq2RhHwluKVGsiqS+dPQ+IYmltF46BYPlz3HkvmXna2d1kQBEEQOqj3xwJwvfF9ZhaXEGir/HGx1PZPNqel8JL30rZx/dXc0nenr4bNzc0aVw+LzgeSQxynjc5pJmbMUTzT82jeNpKIkyqLGvRmzJMPEDP4KLov5muKbzSHdl851HGdWYsHt0217I6EhKRKWIu198kJhNicNNRxXdVjbYrs5XEJa5MBqNcU3aIEj8/ADRdzz6a5lE/chjOqBntdDIk7JyMr+g7j+svXTVP3UxnXnUa9Qiif0ka9tqqfKMWIBVhvDK6aWCm3X8TGKxJ3NFmZ4zMSpWir7KrwBhPSfU2tah3XX2ZdMdB3Iis4ThuvPrRka6jjOkuwtFcLSbKCLrH7Kaknj+uPnICEO6yu54SgpOIOqyUnIKFlPo7RHFoCKNRx3THZK7lI0rPYZedJq4vKk6oQ41WJ2902PpIaMdm1VeLmm01UNCdSNfwjDk98mEHVZsKd4TjsDvJim/Hu/C1K0WLyzSZmaIhfnJqHw9DUy2sADYYmilPzGKbpGQCJW3EaXb2+zo1GFyRuRUvvt0lNZl4akcMLK17q8u9+fXg9Kxe+z/4ROVxf+vW4/hfJLqHfVh6BoVWfYct0g9/NkfDjzEy5CcVVgOLzEeA4aTuyWPPzOwnXff2b0ndnYMZABtz6C0Y0FOHa/xoZiyrI/zCeUs8GMo2XMatuPg3W15ia76fss1/jmz4fg0WszCgIgiB8fRjUZq43vs/0w/UE6HihHXDZmH64Hoa9zzE1XVP8sJJxDNTP5MZ3u04ZaWvYfNkHVJWM0xT/WPoRUsJHEunofuqQikp9eD3H0rXfBd9XPZw3C67kofL7IU1hcHN12ypueeZY1PLl/LJAZV/1cLSsKd0cW4fR7sDrDKOnhlFGu4PmWG39TVR3aN89Qh3XHcmdgEk9EOx53EPPK1PLOC3i5DogM8Rx/ZcSXt/2GsiKnuTt0zuNCL4GKeH1muJXWJpP67juHI8pJSuE99HxmL4b8XfnhMlLucHLvdauzdUrJZV7rS4edoPf5CVNQ/xGb1Mw0TXrZrqU5LRMrfJtfJZGT5Om/U/yhtZ3MNRx3amOrCYurIHwxvAez0eOMAfVkdWa4kf4DMQrUjCB08PnLEGViPBpq947EWKSKdRxnSW09BXrvKhtq9bft47TwjtyCz8f8DlxPgOzHBHs0fupkVRiVIlxfj2SKjNi9CM8Z78ALYmcMjms/X2qAJy0iu1J79MyWdu0892RlW3/LSkSAwsHtCXTjmfko7YkmXdHVqLt9grozCUhHWGdWVvF9Xyfjlf9OvYPzyFn2IGuz0FS0ft1zPd9PXpai2SX0C+KCitfdfPtzHU0xg4gWppAdUw55roonI0vEZ7pYsfmUo7cMpMwWVvzvq8PiXcUK9bANVw06H0AYssP4Emdx4TqgXw00MRsdzQmj4sNz/2W+bc9cpb3VxAEQRDayQZHsKILGz21DJ5ZXMLxiEhN8T0RZVy1aklLtO4bNl+5aglPTivr7s/7VKYLNj6/8Y3rUVE7bOPk6RKeU/hO7cTAH3PuDUaT9ORaOva2klD4Y849jGanpvhmtZnM2Ws4+sEyul4Gtk+TNKvaqn7k6Gqo6fsOuhyt7QIcINlQw8VuK49Y3T09BW53W8kx1GiKr48IbXpfqOM6i4ivCuk1cFu1rVjZpJow2Brxuez0dJlvsDlpUrVNAQSYHl0c0nOYHq2tcqm0Poq30oJJDkmVGFjQ9QL2SYubZfXaemoFIvdiGP0ywd54XeMrEhgm/5LAfm3TeVGNeCUdBjXQY6LFJ+lA1VaZBlBsMXFw0bu9no/eWvQuYdWhLFfQVRU67miycq/VhRSQGFh00jFKDyZCbm+ycgJtJzyL3Hkdz1Mb11lNcxopnNztquN7tHWack2zlnRpUHxUMQltK9dKTPR3SvxJKomSl/gobZ8Dk9SEYdJ9jDk0ssdq5X2T7sd0dK6m+AUtVbh9VUMX9FKt25eY+iiqMkIbp8UBkw9/ywq5qqSSN6DrKrl+fYADJl8Iddlnnkh2Cf1SXe6nbCMcvnYgo3R+fIVvcUHhYhpc69ArHvKsNfzlV1dxS/Tsc65PV2eSBCkb11DyWS2Hw6YTO/Qg1UetlLvfIMN+I7JtErCJxjmT8G/6kOoFVxObNf5s77YgCIJwjomOju7xMZ1OR1VVVUhjAWpr26e4DIj6F86SaUDX/iUyEibJSMBlZ8aENURHXxFy3NZ9uCrje+g9Azm5XkVCwtzSxFpCwuTxEp1T0uN+dxe31fJZUzl0QTHPXPEiS9dcQpKzPalTGVbNO/M+4EBiNYM/T+vytyf304yLiyMQ6P4CLnvgj3FbpqHTBS8uvN4qoGNlgwc46F1HdPTlPP/881x++eUApKWl4XJ1rYRpdcsttzCxcDyB8W9Tt/AVGjfNIswZ0fa4w96AacZ6wqLq8O29tO33Q4cOpbq65+TUggULeO214OqNB+ybiFIvRVbVbr93GVQ9sqyjNnUfQ4YM4dixYz3GHTduHOvWrQNgwoQJFBQUAPDQwlnMtxgxeeEvFjdVkoqsB0knkaBK3Oa0MK3ZyLFca7evc2ZmJrt2BZuHX3TRRezZs6fD47dfOoER1mo8LjtGDOil4MV8QFXw4kNFxWRr5JirskP82NhYjh49CsA111zD6tWru31eP599PeMnf8zQJW9TsGE+boeVRoIVRAabg/QLNhFIPM7HexYyNzq6Q9x77rmHZ555psdjZrPZePDi+QyYs5oj719BMx66S0SlzlzF8Jr2BQTeeecdvvOd7/QYt/PnvuKYDfOwI6Qu/C+Fmy7Ce9Kqm2Z7I1lzNhIz+Cimkq6fhc66+8xdtmgBZT6FkYdHsHTNJUQ2RrSdI1ovwvcO3s8H1RI3a/gsT7742yi6Txm5bwSXrbmExJM+y+VhVbw77wMODDtInmNul7/tLW6ruzNuwuGvJUYX0ZYK9Kg+FJS21Is70Iyr3N0W4+S4vZ0jAJ5//nksYV6+GJHDc8v/xZJPFhPZ2P5Zrg+r5735H3Fg0EEucrUnzEM5RzzySPBm+Zrdk1kSlc+wY8PbXoP2+A18tGgV/oQKdh9byN39OEe0nk/unz8Ke7KdhrCO1W2yASRZAhXC6+24y1w9HueezhEAVw/9FeMzD5GUlUvh5otQXVFtn2XZVkfyjLWUHRnCZ640busUv69zRKtHrx+BL0LFoAt+xgKKSnM3MxbdjXVEd/os93aOgOD5ZNGy8YwpjOH6166lkSbqaD9War3M5a9dTtPSN9gnHWj7fSjniOLiYPKtsXQ8wwozueaD4EIJTbRPmzc1mPnG69fy8uUfU2+I7fc5ovU1m7xgPrpmFe9JpyFJBrl1mrkKeo9KXUXPr3Nvn7lvzcrCO7sSZNAbQDa2J+YUj4LfDyjw4fYwrtbwWe5trJY+2ZLaz7+qqzt9y0hGRUWd1njno6/TMVJVePCmL1mdN5HLlv+aqVVOXHY9sWHfxVn+V2KGOdjx0RY2fudy/jbrYayy9rsnfXG5XKSlBe8MFBcXY7Np74PQG4+jnh1P3k9NwUAyN0TQsGIHilGhZPwVDKzP4rjrcWxNPhp2bCIqMo5L/roGWa+tvPhM+Dq9f76OxPHpnTg+vRPHp2//q8coKqp/d0z7ujmk9Qtiamw8bqX73iVZchrPWH8GgHHORqa9v/K07EOiFMXrtl+3/XyL+3GOKD3fZQ81rmyA5bcMa6t0+OTTozgO9zxZQ1XVtvdeX8ds8LCdGI0ZyHipK8qmpL73qWCt+9xXXIB7LvonhSkHeXf947hP9LIPcZPYduSTkOO27kNCTDw+tef+NE+Yf0iyeTw7J23l0TXPhhy3r324Z4WNy0aYGOfXs/Kgh+vebjgtcX9iXM4y40wAdvtzub35r6cl7pRUKxtvCEdS4fUv7HxrQ89Jv/7EBfj7lVfynWGbOZKTyeh3tvY4bmiMlS9zT4QcN9TPxsJBJt69JhKvM4zPtl7Ioi+fPy1xW88RrVVLl3A3LmfPjb9DjZsgRfHGGThHAGywP0FrnendTc/wZeDgaYkL8OSPZvLvzK0UvOandkfP5567rx3C3X/dGnLc1v1IjEnBq/Y8jfMFy11sW/45xp0/4s+Hrg05bl/7MOT7esIGBRMW6r9N7N7f+1TPUOM+N2sQi+PSMdicrHHkct3K+tMS988Lw/hBdnBa9sYCL/P/0/t3jP58lufetIjbPppCTk0Rv/L0/DkyhOmpKKwMOW7rPsyadQf79/+rx3EX6yfx/Yhv8p8JDt5ac1/Icfvaj/BhEoNvMrRV4h74hRdPL2uWhBrXmiox7Pb2696ch7x4e3k5WuNGRUVp+t6jJdklVmMUQpbjVjneGMF1c15lisOCqvqoTL4AT8P6YFNDfylSo5+/TvrlGU10fZVM4ZGkXbiQqKQjeCMlTOXhyCqkHD2CIaBHb8km2gHRkQb0oy9A0oliSUEQBOHrIdDD0uCdhZu1NTz+KklA3oDj7B69h7wBx9E+Iax7Fw14iQ/nLmJGjKfvwf1Q54rm8OA3MXSaHtmZT9HW+qGv7/77qCXR76Tao21qVU8GFGUyOH8Qssam7j1psmjr2dSXYldwXo8qgTWuqo/R/dNYEzy2B4ce7nWcw6O9V1FfVCTWb7kG+Qx8D22drmdtOj19eM/8vI/gh6JQKT+tUYcrPuIVqesqgJ1keLU1R9eH9z198OJ183DF5mqK35vwxjDST6RRqPZchdZf1phqYocdJCKtCKT+Jyl64vDJJ9XrnV5DCgYS5Yjs8z1q9mm7zh1S2XtkCYj2qIwtOzP9rhJUiYfdNo0TYc9N4spcCNnbfz3AjJSNDJSPIzU3UZsRzfCGTJpda4kaauXPSiQXLL8Km73vDPe5JOWCeeTv/pK8dB2jsxrxuEF2HiIQ5WYQF+LUbyFh1GB2lRV1msEvCIIgCKdGp9P+tfQ/16YQs+ZbeLvpJSS39Owy2h00JlV2+/d92XLlCJo2LesQX+o0hevvcTdgnvk2097sucKiJ9+bdQePTHkZCCbucp2BtmbEwy7XgRLc1j1bv8E/Nj7RYxydTtfjFKXxU25ieuIafmH+iMJXl3NTo5UbbcGxRlsj6TPX82jzXLZXl7J763M8/3z73X6bzdbnFCUKd6K4Esn8zm5UxdBtFxulYgwTcya1/T42NrbPKUqtpkz4KbceMiKroEoyLn04SDpQA9j8DqJUGwpQazAzePDgPqcxtsrMzGybonTN8pn8NqUGuSSKos1z8LrCMOzXczBHh9HuYMbMNRz9cS33l0Tz2lubu8TNzMzssI3OU5S+c+kgdk4tAmCHvJa6E8cJd4bTYG1gXIoRteVCeeKX6Tz/fl6H43TyMelpitI9w77POm7jzu17+fnIPzDDoifnUisGq5PGCC9/O/Ij9pVOYp7uOR45/PcOcW+55ZY+pyiVuYazUrXySGwtYx/qvrr/jrpkVn3UPiXp+eef73OK0sl+c8lUPjj+TyYkbeXHox6nwl5KraQSrUokOVO4c/sv2Omaxg+zfwCf9xi2R1dO+g43Hey4Bp980rtVQuI/5l/y7IjDvLm998qx7tw8fQnL985si3Wyv1hua0tevDV2M89+8UG/41+SPorYwVeR7N9KyZ5s8tIL+ZHlYqxOK+nFKcSM2MML9olUH/svHxbldPn73s4REHy9Kj54n9vCi/jllVWkLev+vHxfdQyOvFFtP4d0jmixYuL1XLN1aI9jTRiQHTKGsOP9Oke0jp130SjKZx5mVsDAD71WyvUKtZJKok4iVZX4v4RyCq6TmbB2BGvXdT1G0PM5AuChRZdz6+gv2vf3pCzD5cNM1N0VnLr6t/3T+eWqdzrE7esc0eqiC39AqTQFSb0bFZiRbmiL25r4v2fngxzz72HdpqdDPkdA8DgNLgrOGpqqG8E71gd7HLsjpn2qeyjniFaXZzZxi/MPPY7VtaShBkce6Pc5otUXVw9heEaw0jaAyj69n3pZJckpM86vR4dEyR3xHCqMYPrr/U+cLrwulbKRwf6JUqd7HSN+bmh7HZIOJPDxK72UM39FRLJLCImqqhxf7+ai7BwI2JF0FipipxGbswZkG5/sb+bgdZnMjZ53tnf1tNMZjIy+7JscPrSOxrcmEnP7FmqO6CjVHyBdmUxz2BA89eD6/rf5+B+/I8kkM/7GX5zt3RYEQRDOESeX6/d3bG9TRZPCohl00caWptbQU1PrPK9J0z6s/e04ZvYRf9icjWxOdIYUv/OYT+9ahs3YGlNiov+kb9YnfYOdN2Q3D6/sOf7JfU06W/bdR/gRn5H7YbBnmVGS2oO7jZR8ciW3z1/HrcnDu+xfax+W3jz+o28j20tAD7Iqd1m5SpFAF1tBamJ7Eqq1x0woxhijSG1Z+VqRFI5n5OOwOwh3hpNUOAC5pdFxkquZlbm5IU8rbu2fA/DAbb/mrS8mMP1wPTrActL0E68zjPxVy/hiWCTxWbuorX2317it/X5O9sStV7LH1H4sT254HNz74PZSbBnU1m7vNm5rf6LuvDr/Oa4rPsw2bxa7t95KVKC9eqxOZ6E5JptvOgswpA3s8ho/8sgjbT2VevL4opU8YXEjSRK67koOVfh7hIe5Yaltv7r88svber+FosQwA5chgc+terbFOJFs7QkU1dSIz+hH0YWxqWEUtbWvhBy39fk+dPFTWKTe6yXNkpEMMjV9ljfceG2P8U1Se5XMojgff+glfk/bvv+7v+GH7OCTE/G8/ZM/0BDRPq02oiGCZR8v5e6Infxt4lxq92zq8ve9nSNaPfHK54xfdRl/uOI1nrC6qZTbq4viFYk73FaS113BG3HtUz1DOUe0ivKn9vkaAJidUf06R7SO/fZtt7DAlM/DbhuKXiKuJB2fy47B5sSeUsTDbhv3WuFY0jBqa7seo85OPkdERUXx6LeWn3S+7kgvS+hbXmZ9rNzre6i7c0SrX/zgQRp2jOFI1TIGzF6DKawRW0tcT2MY+Rvm43ROYmj2Yd7stI3ezhGt3lxwP5QORi/riaLnFRfTE9oXy+jtHNH532drZCCk19ho87Cwn+eI1mNactsl2IyOlt9KzKDlAJ1UcGgzSoTpY6mt7Xnadee4rX76xwlUmrp/neWTXv/EDH2vr/OpfO/pD5HsEkLy+Id7GZmchwLIASdqeAJjiu34PIVgm0jKsfe4/5MYFty1oM9Y56LogUOZsXg7NTGfE1E4HMOgSKKVUSioRBun0VyXywV7n2FHyqU0PPMThi65AVtcytnebUEQBOF/2OcqZAw5zJAlb1O4YT5eZ3jbY0a7g4zZa/ENOcznlSmM0BB/o2rgwgHHGHrJ2xRs7Bo/c9ZapMw8NjZZ0HIrLN0Q2pSaUMd1Z5KxmfqNk1t+6n7FyoYt2UyapW2lwdxwL5K1ss/Vt3LDe+6F1JtEaxkwkH3D9/P2wve6vcgfc2h0yzht/MUjGJ7r66F6Pbj22rBcH8VWLe8iSHX1fFGpZVxnHlMjE93FjHN3TTxEBpq4pXIzErDT1Huvop7UWovxmntJIkrgtdRSa9W2QhxAdUMqcvpHGGbdTJd5dNZyDLNuxrfxWaobUrv9+77YLKFNsQx1XGcpBm+HhSx6G6dFktnLhhNhvLDipS6PNYQ38MKKl9C/t5ykVG3xAQbrvdTmjSZ55TW8Nns1h6Lq2ipNh9dFUbxhAbV5WQxO0rZyqxxiEW+o4zrzeNK4o8lKzbGsHv49WMPtQw5zs0fbaon7AgYqFCNxkg+5m2mLiipRqRrYF9De2zixWmZR9TZqK7KoOz6E8JRiDDYnPpcdR0kaqipztW47u6u1Ta82j9mB8UQ8HmdYt2c7FRWT3YF5zA5N8atj3KTYHXidYfS0cqvR7qA6xt3NY6F5vzKLzNgqjPZGumuLparBmxTvV2bxfQ3xp7ht7IsJbdzXgUh2CX1yq/Dl+gH8MPmfSJIOVWejMjGOxPx1SPoY1Jg8Um+/gJnZd53zKzD2JmvRcr44spNS20CM5Rsx4MdttGALm4NitGMuryR64y8hKokDbzzF5B8+fLZ3WRAEQfgfts3gpsLi5uHBRwgfeIS95Sk0uu2EWZ2MTizBIMO9FjeFBjc3a4hvq87kkXgvjwzOJXLgMZylqe2VAsknkGSVe9xDsFVr628iN1tP67jupDVIHS76upLwOiNIa9B2Z9njtjL64ChufOP6Lo9FOiK48Y3reWHFS9T7tT0Ho62afcNdvV7k3/jG9Rht2nthxRfHExXoOVEjAdEBN/HF2i6SpepsVNcesJYjdXORrKoSuJOQqrM1xY931IEUhaRKdL7ADP6kgqQGx2kg63ueTqZlXLfcURhm3weoXS5gJUlFVSUMk+6Hj36jKbySWEOdLpnIgLuHS3Co01lREvO6ebRvYWY/SggX+WFmbf2uUmskXpy9JvhD9zlr/jt7LTdsvUpTfIDwlP2o9sHU5A2ltiXRktmSaNlbkhZ8n4Y1EJ6yX1N8V+o66vZdSaRH6THRUmeScaWuA37S7/gzdV70ucNOqsRt53WGkfvBMoYueZuZOm0JwWZZ5Ql7LQ+77CgqyCc9BUUNPoMn7bU0y9r7bZmPZxEZaEkEqTKOExkdHpcIJrDNx7O6/nEIXCaZzNlrWo5R67qerYITzzNnr2WPSVsyLTzM22t8WuLnh2nvSumOdFKwYT5Dl7yNqtLhfNE6xbBgw3zcQ7X1LhxZH0t8yjEqJbWnjzIJqsTI+thuHvzqiQb1Qp/+tvUozn0q7nALkqqiTEonuTYd1VcFsdO4b6qOzQMGn9eJLgC9ycz4b/6EA2/bMRwx4/cVQv2XHLcWY7ZMxFFkJX3JABqXfI+8j1/FXdX7ak6CIAiCcCb5rQ42GH3cZGvkisgGbhtxgF9kb+W2EQe4IrKBm2yNbDD68FsdfQfrxuTiwaze8lfu3PEnqrxxRKQVtTUkrvTEc+eOP7F6y1NMLh6sKb7BHOwlpqqtF0vtFLX9i3vrOE08fTeF7te4TmJqw7ji46VA115FrT9f8fGlxNRqq1pqTDzCW4vfaQ3YUcvPby16h8bEI5riAyR6Qkv0hTqus1I5At/2YH8cVe34JFp/9m3/LaVyhKb4kQZAlem5NboEqhwcp0GSObRFDUId151mWy2SrazbSg0IJrwkWynNNm2vQVRA4Y3YYDKxcyqi9ef/xmYTFdBW2XWgZjSZrcmoHraQOXstB2pGa4pfpviDVY29vMT1EQ2U9bA6bUjbCOjJmr0WgICqY0f1ZD4+cQk7qicTUIPlVlmz1lIW0FZLotRP5IApHaBtBcxWrT8fNKWj1E/UFD9LX0fBhvktP3V/sijYMI8svbak70C3jQM6hXttTqrkju+TSlnhFzYnB3QKA0+h4icqMrRka6jjOtNVxxMz+ChDl7yN0d6x0tNodzB0ydvEDD6KrjpeU/xoc1VI8aPNp7CIRoSPvKoEjnxwRUtyuZ3XGcaRD64gryoeInpZjrEXB5D5bkPLgis9nCxubEjkwNckzSQqu4Re+RWVL9ek4zI340xZRIxchas+Hl3dhxTEmSD//7igws3Cmy8/27v6lQhLyWDMY5GYri7Gp1MhaRu5uk+Iir4UY8MmcptTqZ15FfZ3n+LAf//KpFsfOtu7LAiCIPyPGrNzGmXzPuGAvusFXo2kUqMPoPfpGbNzmqb47jATFCezzp3MhrK5TIjZSay5iurmOHbVTERpabbrjtF2lzq/MZMyzwQ+LLmEu0Y9QqKlvU9KRVMifzxwN5ekfEizt5Z0TVuABHdoib5Qx3U272hMh6mLnUlIRDmimHdU2xS6o+H1OMJ6+VsJHOGNHHXVa4oPMDzrI9g+K7RxLO93fJcrCqVxEb6Nz2KYdB/YTppy6U7Ct/23KEWLccmr+h0boMYQ2Uv3nY7jtLApkcQrUp+VDjZFW3yAqKRdhLK2YFTSLuCGfscfnBPD+9EmnmUmV1XvJCrQPo2qTmflzdiJlEUbGZwTwvylbmwpGMXCBat7nfJsTyhjy95rmashvjsitKq5UMd152DlKKKSvPwn+WJq9HYc+vZqzHC/mxi/k2/YtnOwTFsla3ZJPIOGvUxWXAX5G+fjO+kYmVqOUaAqgbiSCZriu2qsIVWxumq0VZleHXWYec1mPjF4uyRBWgs2f9psJiaq91VLe+PNOAy5SaGN06DQYcPTGEb0oKNED8zFUZLWVq0cnlIMkoqnMYxCh7aE3e6CC7ho9FvBhFY38aWWqrfdBRcwuY9YPUmviGPdxe8R+d/rqT0+hIiUE21TPRtKUkGVWX/VS4ys0PY+erH8aiaVR/D7aT/kSasreN5rEa9K3O628cGOR9hOA1drfA6nk0h2Cb164dAOrHvCyL72c9JLw/E1rCfcP4lAwMnWLB263FuZttzGxDRtpeXnojEDXTgm7EJan43jxmRmHKgkf2Q+Q6zDGFB0jKxXfs3+i77B8CXaS6UFQRAE4VTZ3DZQjKj4u1aESC2VUQFjcJwGx8Lbq50UdOyo6f7r+cnj+sMZiObRA/fQ0JTQQzJNYlf9KH4wsOfVrfoSb/Sh6API/p4qf1QUvUK8Udtd8BpfFin0PYWwxqdt2o27t0SXhnHdGn4A/f5J+JqtPU5xM5jd+Icf0BbfZSXa0kBt0UI8xRcjx28FSwU0JaBUTgFVIoZ6cGm7CHf6e7vA7/+4zg7o4rmjycq9VldPM5O4vcnKf3XaqkEAZhkPcyjEcVootdF807SOJ2xXs9eawqDmaiICTTToLOSZY1ElmTs8r6PUaltxPU6ubZtaFTUwl8bS9ov8sOTgRf7RD5YRJ2urTEuJC23Vt1DHdSeyYiR/rLicZnNkl2SOQ2fBobfw2N7fsFh9R1P8eLmG7DG7MNtcRA/qPtGSnVRK42cZfQfrRonbyIAQx2kRbWxijM/IbF/XEsk4JB5225CQOGHUPqXamHYIp2UytiZrj1M9XRY3xrRQPi1dNcfmtb1PA5LCsQF5bX3ZxvqCKxkWbJhP86jdmuIXVKXiaQwL9tOSVSLSijruf0s/rYIqbb33ABSPmYvcLZ/TLlM9g2/cOa5oyj1mTfGlmjGs8w6FLX/l6VG/pzyspO0YJThS+dOBe1lXNp9wY+iLKJxJX4/6MuFrya/42PBJGpePeInLDu8kEBZFYPAUAq49yJHpLKq5Bs/3/EycPL/vYOeRqMhkUu67jKZhleApRcJP5uFN6Kzjkb0+jOnHidj3Ac5m59neVUEQBOF/2IHMMvwmdy9Tn8BvdnMgU1vzcrkxtPYFoY7rLK8pnYamJEBuS6Z9XHIJO2omt1SNydQ3JZPXpLWuC0rCHMh+XcslQNc5GSog+3WUhGmr7FItofWnCXVcZ1Eh5hFDHdcdb2UyA+etamlF3/UYSagMnLcKb2WypvjNmV9wZfkeghlYCaViOkrBFSgV06Glz9by8r00Z36hKf4+qwNHeH2XqWHtz0ClIbyefRqn85YETAx3pPF7l434TtMwE1SJ37tsDGtIpySgvQ/PoBODMDbGdpnm2UpVJUyNcQw6MUhTfF14CUOOB7iD14k115BrSWSHfQC5lkTizNXcwesMOR5AF16iKb6StJvavCyOfrAMn9veYcqzz23n6AfLqM3LQknSlkQYVRpGnF/X9e3ZSoV4v45RpdqmCwMUyRk0+6IAiW4apwESzd4oimRtyajYgVux2F1IEm2JkNZjJMnBXm0Wu5PYgX2voNedUAvONBam4Zfbp+l2TkTJJ/188rj+ytN7eOPSt4Cep3q+celb5Om1bcNpbqKscCBvFWWwLLyBH9qd3G9z8UO7k2XhDbxVlEFZ4UCcZm0Ju3i3vm0qqdrpvXpyP614t/Z6pBGRe7CsXQL0PHXe/OkSRkTu0bYBQz0A68rmc+matTy98T+s+/JJnt74H5auXcO6svkdxp1tItkl9Oj5Q29h21FFHLU4bSaiBx3GVO5HxY/ZeimHjIXUG5sZdAqNBs9V+lFzqUsairx1Av6ITOSAFyVQg6SPobYuHWtzAZ+ufp31919PU90pzLsWBEEQBI1UXeZpHddZnFJGpN/V9Vt7W2CVKL+LOEVbMi2nKbTkQKjjuvNuchEGW7DqqadeRQZbI+8mF6GFNaaUOp21t2twanVWrDHa+nwmHfgGqiup1ySI6kom6cA3NMUH8B4djj2hjKGXdNdnppGhl7yNPaEM79HhmuKXDtxHpn4/N5dvJNLf8SIyyt/EzeUbydTvo3TgPk3x5di9vLnwPaDnC+S3Fr6HHLtXU/yMiIM8mnMvs30m3mqI5K9OO7912fir086bDZHM9pn444F7yIg4qCk+wPryCXzjo0VIqN32NZNQue6jhawv1zY1KXdSMJE45HiAXx9ezS99L/Fj3Wv80vcS9x9ew5DjgQ7j+qsuoRYVlZq8LHY9fysH3ryOo6uWcuDN69j1/K3U5GUFG7AnaKvs8jsS+WlrpUoPH+Q7PGb8jkRN8QEwhZgFCnVcJ/4QF5EIdVxnefLElnNRz0nfWp2VPFlbT7B9vmBlpIREAJWdeh+rDV526n0ETlrLtXWcFnvMAfaPyOGFFS/REN7Q4bH6lgU59o/IYY9ZW3Z/1o4rKZmxjsfG7KGy0/Vtpazy2Jg9lMxYx6wdV2qKf4Fa1Jb07a6fVmvS9wJV2783ADV1cZg9XRNdrSQkLJ7gOC0clvbj3v1NqK7jziYxjVHolsvbyLo9Gcwd9C5Os44IORJHTiI+5zqqkzOwGyKIXfNfvrXotQ6rbfwvif2uSvGqsVR4jjLKnoezeRth9kmYq1eTlzqLuCNbKTl2lCPvPse4b997tndXEARB+B8jhfhlM9RxnalWLyuqd/BMwoX0tOzTVdU7UEdoW92rKbIEQsiTNUVqqzYBiCgdzZHsLQzc2LVKvfXZHMneQkSptsbZgwyH+G/SbFac2NfTDDfWJA3iKsOabv66b7b6gfi2P4hh1s3BpMdJ/VNObu5uc2u/8DB5okOagmbyaJviZvaFsXLhe9z4RgRji06QZ06gQWchItDEoOYKZFReWPE+Zl+mpvgBvdp2gXzFx0s79FCrD29g5cL32D8ih+wvtVWmWe21bDmSzZ07/sRdox5hotTeW66spbfcrupspmas1BQfoNo/gOyjkXz7DVi58L1gM/YWUY4ILv/4UlKPTuFAcpSm+JbaKW3/LasSqcUAfjpP7T15XH+EVwxvv/juYRU9kAiv0JYw1et8zPEZedgNT1jcHfoIJagStzdZmeMzsl2nbToyQLMhtJv7oY7rzOUNbVpZqOM60xm9/HeYgZsPBBNbJydDWhNgbw6T0Rm1na+tgeDfrTd4g6/BScmieEXijpbXoHWcFhFWhWpg/4gccoYdYGDhAMKd4TjsDo5n5KO2bDPCqm0hBZc3nDdnbAj+0MOqnq/P2MDyfy3U9gQ8wUU2avOy2lb0bO2n5ShJa1lIo32cFgWSlVDOxAWSlb47MXbDE2LCONRxZ5hIdgnden7v3wjbvJRhcSXIhglIgUICzQfx6WXijJdSXfcZQ753J6Os2k6454MJ0SZ0po84dHwG3iGxmJvqUI1uQE9q8kQ8uY/RdPG3Ofr+i4y46ocYbdrvZAiCIAhCf8WrfkLpohSvaluhzBvTwED9fhYYClmrn4/S3N44WLaWM8+3hkx9A8djtCVaEqkg31oK7kS6n4yggLWMRCq6eSw0ftnEXy5cy21A1o5p+Fztd9sNNidHsrfwl1lrGbNJW2WUoyaehRNe4VnfD3ps/H31hKdxHNDWzynO60OpXtxnc/e48H9rig/gDofaQ8FqhMzZazr0mfE0hlGwYT61eVm4h2urZLc6I9kxsqYtGTX0pNmEdeF1rFz4Pvsyask+ME5T/HR87AD22NLYkXYFg2t1bf2ojkUH0Nt2oyOHdLQlQlwlI7ht5KP8bs/v2VB2ERNidp3UW24CCjp+Ne6XbCwZoSk+QJQ+wBux2dxyyM2owyPJz8jHYXcQ7gxnQOEAZFXmmYRsMvTa+uToascBfVdVBcf139CG0BKJoY7rbHd8OZOAOT4jF/oM7NH72/oIjfMHey21jtOWTgM5PA9qZoY2TkMa4WhjDDOcZZhtrm6nnqsqNLvsHG2MQcuSIhnhR/li0tusixjPnC0XIp1UIahKKuunbeLgBXuYfuQKDdGhIryR9QZvsHddJ5WSyr1WFw+7g+O0uiyg408AKqiySt6A4x0HtNxRuCyg6+av+1Y2bC9V+l6qwiSo1AcoG7YXuL7f8Y84hzHJVoDPZe826QsqBlsje5zDNDeoDyO03oahjuvCnXJ6x51hItkldFHpLOHo+yeI9gZL6sNSc2lsuAyl9D8cHJbKjKYoXsss5K8Trj3Le3p26YYuIO3SGMY8fgRZjUDRB/A3fAamdCIKDzPgjsFs/e9e3N5mcj/8FyNX3Ha2d1kQBEH4H2JVQ7vJEuq4zhqrRvL6lS9zJK0Qg/ILlKop0JQAlgrkuK1skhUqizPIOjFPU/yl1RPJzb6f6k3PAAodE17BO/dx2b9m6SFt024AbGnNjD48isbKJH79nadYlJdFgiOSivB6Ph50hCVrlzD68Chsac2a4jc3hTHigMI3Zv2Jx/bfQ1i91JZoaYxUuHP0H0jbGc7BJm29hMaHHybWU0lND83dJVUizlTJ+HDtK6C54sqIOBTXZzWCK07bdFVbTD3K3jvYN+1u9mcdZFBRZlu1Rl56Aaqsomx5BFvCZk3xa+OSMR5biuOLvwGQaznpfdSk4Nv4LJbpt1Ibp+01GF4ZzRf+LP6YfTuP59zV4bEEczk/HfVHVpUsYXit9rYWEwYVsPZ4Ns8kzGRF9U4GF7T35qrVWflv7ESO2cK4YmCBpvh2ayN1OiuRAXePixDU6azYrdoSFelDwBdCz/D0IZrCs8usUCEpxKkSOiQm+js2SVdQqZRUdpkVrtO2Cbwj90F+iOM0KLQ1sGPfBGZM29xToSw79o2nMEHb+yg2ys+Ywhgu+qJrIk5SJS76YhaFqUXERmm7+ZFjC+cVS3FLwM4bAFR43OJmtG8Al2naAozx2ghXwNESr7tS2QglOE6LuuTQen2FOq6zo4OTWR71InlrltDTE0ifvpGX6y7QFB8goSG0nnGhjutC54NQZomeQhXl6SSSXUIXz375O5o8FzNlwsfI9Soe57dQq1fSaNXxwiQ37k8eJWXat/CqYPwfncIIgCQTM3IC9sxV1H4xHfPdeYR/sge/7wR6VWH70dHEWT8hd/61HF75LFmX3YTeZDnbey0IgiD8j1B9F6C6NoK1rMdKAdzJqD5tX6wzjVfwSXKwYkjSKegSt3TaABQkl3JxpbZKgeQB+fwybiN3zboZ3/YHwX1S1Ye1DMOk+/lF7CaimyM1xQeITTMy/6mlRDoiGH14JMcz8qm0O4g4kcG9qy5HUiXSS1LZ/aMCTfEVCeryB5HGMd6etZwD3qFUN8cRa65ipPEoxRsvoi5/EEpStab4uis9XP3y+/yt6kakluburaSWhOCK8PfRXam9KbR93BEMuzLxOcN6rkawO7CPO6Ip/iBzJOtT1uPb+A8Mk37doVpDdSXj2/4AugErGaSP0RR/RNJcvvhgXMtPnSsEZUBB2vUAI5bs0RQ/VvLzTNl8Rkft5YN5C9HL7VOo/IrMS8dvYF3ZfKaatVfXpcWZMSbcy54tf2WvNYXBJ62WeKxltcToaT8kTdH2WZ4y9Ah/KLqGb1Rv63G67aqoUdw19DVN8VMuGsrR9yswqIEek2k+SUfmRUM1xU82DeIJy24edttQUDs0RFdQkYAnLW6STdoa+APMnTuEY6tr8XtbmtR3oWIw1jF3rraMXcTAVLzrhvMZkD1mFxZ7e4VUs8vOjn3j8R4cR8Q0bSsN+sK9XLFyKdB943IVlSs+vpRtN5Rril8Un0iNN6fnARJUSypF8dqnt7l96dzbVNFt9Vire5psuH3aFi1JSwxtFcRQx3U2ddB+4pNy0Jm8FGyYj9fZfqPJaHeQOXstMYOPMrVsPzBe0zbMrghCyUYFx2kQvR/KE1p+6O5sIbWPI4mzTSS7hA5yq/eh7A0jOu4EGRVOVL0JT/kbqL4qNk0zEBloojxvHGk3T8KAtsz/+USWZbJicnkjbChpBTI2ixldUzPIVmIOehn03eHUPFdIwrIfIfW0HJYgCIIgnAFDI07wTls/J7WbSgEJ3/bfMnRArqb4SXe6aN7Qy91bCZr1PpLu7PnCpDf+OCNzfEYejd3Eny6fQnn15LbKsaTYbdzhMTPHZ2RnnMblw4DUqpS2Hk6SKnWomGkV5YgitUrbXWp/goJaoj+pKuoEg2xOfK4h7C2Z01IVJeFP0NZjxq3O54Yx9zOgYR9/zLmXiub2C8l4cwU/H/UIF4ZvYZv6W03xARJj3KTOXsPRD5bRUzXCgNlrMcS4u/vzPi1JHcJKx7vUAJ5V7yCHFbdXpzWmYpj0G2KjPmdJxM80xdeZ0nE093aBLdPQnIjOpO0COWtACktr3+eGQS/SuTu6LKncMOhF8muGkhWtfVrPiYESTYZ3MRi9+LY/SK50cuK3BMOk+3FnrOKEbwZa0kVy7ABGust4Jn4mK2q6Trf9b8xEJjsLkGMHaNp/+6WzsNzzPP5mS4/JNIvJif3SFZri33LXA1zx5mvca3VxR5OVhJOm6FVKKk9a3GzU+1l51wOa4gMsH30zz0z7Hmx8tuU3nZ+FCtN+zvLR/9AU/7oJt/P4/Nu47p2r2bV9Cv5R+9DbG/E7w9DnjEGn6ll5+ev8bMJfNMWPKAjv0K+uMwmJKEcUEQXaPsdx9iTyQlhfIM6uPQGSGD6ZsZ49POy28oSpibDi9p5dzrR8bvdYuNBnoCJc2yTAb46PZdUBPTWyv6d8JrEBPd8cH6spfooabCwQM/go0QNzcZS09z8MTwn2Pzx5nBbmSG+HRvG9jdNCTt2IUj4PwvKgMRM6bEuBsAJoHIScuhFYoGkbp5NIdgltVFXl1fVPYa2+kln6lSAZMMZdRXPlSsqj9UysvZ6Yqt2c+PF3uFAf6HEp8/8pkkTygm8wrupTju0ZQJzFFzw3KsEycPe+KGLS88hTG5hs+N/tbyYIgiB89eanNPDC1glUb/wH+km/7tTPKRn/9geIqxzP/Bk7NMV3qKGtnBbquM4qjONo9n7MLFUK9uGx76cmbF+wD48zDEmVaQ4YqTCNI0vTFiBKCa2peqjjOvNOM8AuAKmbqqj2y37vNEM3f9239GM1GG0u5tnWMidpHbtqJrZVjk2I2YlOUtrGaeXVScQMPsrQJW/3Wo1wXKctmVMa6ebW3ER+P2gVcuonHabDGuK2IskKP8gbQmmGGy21CHVVIU5NCnFcZ/ax8dxVfQugdlm0SZZUFFXirtGPUBGrLQkC4Ip0gQt0GauQ0z5BqTxpynB88Bi1jdNAiklivDs4Be2PyQuI9TvbKsdq9HaurNnJeHcxUoy2RIVkMJB2TxIFD9S11Fl1ehyVtHuSkAzaPge2sDCm7pzNhuz1bNI3MC6gJ0aVqJFU9uj8KBJM2zEH243apqoCGPVGrr1kAK9yM75tD0LTSQlHSymGyfdz7SUDMOq1Jd8NDplUt5UXVrzEslVLidwfnJ5tBOrC61m58D0G1UdicHTXv7BvQ+pCq4wMdVxn42IMfBnCqX5cjLbXGMCoN7H/0CDGKHru3zAf30nnIoPdwYDZa9gv+0kdrW2FXmtzGRcfmcErwzf0OE1ywdEZWAeXoSVVFGlt73coyWqH/oc9jesvdbQKe6DrE2gbAUjBcRrIjUNQY3eiVndXeSZB4wCk2J3IjRrnJJ9mItkltNla9Cmeagtpts1IHjBEzERtqkAKuFHMcWQ5Ujlc6yb+colxOm0fkPPSsIUMvcTMtn978Gdl43PtQgqUYnPJlO1qIOXmOKK3BPjo339gQtY4UqdefLb3WBAEQfhfkJrNXaMe5j8bf8ryz+6lLvVAW1PrqBMjeStmNN+c9Qikavt3Kdoc2tLloY7rzDj5Ev7wtMR9Y3+DpMod+vAoLZUbj+bcw/QfXKIpPkD0hDT4Z9+9pqInpGmK759vxfm8C1uTtZul4INTh1wWF/752poFy3WV0FJkoJMUJsVu73mcRpURCVBd0mc1QmVEgqbm31uTXVwVV8kv8obwt4Ry6k+aDhvlCOcHlQlMjqvkzWSXpvixiaG1kAh1XGdh4cexux09Pi5LKnZrA+7w48BYTduIjUmBljyWJHczZfjkcRroho0CyhnvLmas+wTHzPFtK2IObq5EbrnSD47TJuw7S8nkPUoeO4HPaW/7vTHMScrPUgn7zlLNsV3bKlnx4SJUBbZO2sAuQ/vsE0mRmLptNitWLcL1rUrs07RPo7tl8v2YzY/y77QpBCrbK0118dtYMeZ73DL5fs2x9XFG7kgr4jmfnidveorY2ti2qqXqqGqW1sZyU5qDOo2VrMkDrHg+67uCNHmAtnPRspgpPHP0+eDk6R5yLHLLOK28+iE0vXc1R+l6HepzhrVUn0p4xyeipcRAsadxwUeRRFfKvD5jY4dm9fEBHSs+m8XwHdNQVmjrc+kdkoKzIB+r1YEsdX0OiirhdkfgHaK9CtQ0KwvvawdRfEZ6ytjJBg/GWdoWzDA6xuN2tS6o0v20cNWVgtGgveL6dBLJLgGAgOJnx/qt7B1tY8En1SDpULw1+JsOcixNxxDj1TQ4i6m6bA7fMgTQi6quDhLHTmBK3G/4xHMNV9kv48O091lwbCxK+Ysc/jKOiWGH2H3Uyo4tH5MyeT6SrO2ujCAIgiCEyhs+jtF1/+Lmik2ARMxJU/RUmrm5YhOD68rxho9Dy33wAfkZRDRE0BDe0OPFTaQjggH5GaDhu3v1iU2sLLqKBl8kd416mERL+6qLFU0J/PHAPawrm8/QE6sAbQmvUQtnszX8GcIc9m6SUaCi4ghvZOpCbYvyRIXH88alT3PjG9ejonbYhtpy4fHGpW9zSfgPNMVXIhL6HtSPcd0JG38TFce/T5wqIct0qUZobf4dNv4mbRtwVfKExc3DsVUs9On48vggqgI64nQBpkbWYIit4l6Lm1SXtoTdgmsG8fwb5VQ1x6N2s6qnhEK8uYIF12jr52QMc0MIbY6MYdqmhwFMmrUQ69MxuG01PX7WbK4YJs1aqCm+bWoihjgZb1UAGRja3HGFUxUVY5wO21TtiSIIJryyrvfR9NFn+Evr0CdHYVk8Q3NFVyt/ZRMAV69axBWfzGPtjC+pi6gjqiGKeZ9NxagYOow7FT+Z9TDXjLid9w6/QFljIUlh41k67HXNFV2twlOKMYU18kPgpoDMW5JCha2RBElhuSJjiq5rG6doOKEO+fYgjq7ciccZ1uO5zmR3MOTb2Zr23xCWxLUeEy+bPD1WRV3rMWEISwqpv3m3Bg8n+GHr7kMgdRrXf/Xei/A61zNow0Lu2TSP8onbcEbVYK+LIXHnZGRFjxeo987B3me0ruLTb8W1bxc2awOKSodKUEUN3v5wOizEp9+qaf8BGocrpM7ZRMXqnheGiZ2zmZLhwzTFN9macJf19u+JDE2JmFK0V6edTuKKWwBgze532RpXzZKcRmRAZ8og4MlDUX2Upk3A7o/mvbICFiY4mCCqurqQjWay445hPVxIXfMWZn22B7djA+hjCd8NuuGNDKjZi6/mGMVb15zt3RUEQRD+B/irvBxftxCJ7hfHkoDj6xbir9LWu0M6eoRlH7dUY3T+atDy8xUfL0U6qq1xeW1FMDmwrmw+i9es4abPX+CenY9y0+cvcMna1awrm99hnBZ6gwHdz5Nadrnjk2j9Wf/zZPQaL8ZjbYnsH5HDCyteoj6socNj9WH1vLDiJfaPyCHWpi2JUJgwC09jWNtqbZ2pKngawyhM6LoCW6jGpl7IM3YFiWBi62Stzb+ftSuMTb1QU/zx1jQ2GH3ca3VRp1O5MLqW5XFVXBhdS61O5V6riw1GH+Ot2qrrLDV7uXvUw0B70/5WrT/fNeoRLDV7NcU3jgktSRbquO5Ish7lwO9RkVDVjp9mVZVQkQgc+D2SrK2OQdLJ1P84WNHT0+eg/sdWJN2pXzpKBgPWy+YQ/oNlWC+bc8qJLgB9fHtVnlExsHjTTL7x/lIWb5rZlujqPO5UGPVGrhz1PW6b9nuuHPW9U050AchN7VONTbLEdZEN3BFdy3WRDZhOyoqcPK4/pEGTSZq9reXfgq4nbAlImr0daZC2fleBlIn8UD+Qb3jM3db7fMNj5oeGgQRStK+e23y05wpKLeM689e0VwTKip7k7dMZuvpSkrdPR1b03Y7rl3qJveXJOMuT8Lk6pst8LjvO8iT2lSdDvfaqknpvNX/PcDDkkpUYbB1XTzXYGhlyyUr+kdFAvVfboih+NbSViUMdd6aJyi4Bl7eRnH1F7J9p5YZddUiyGUPkPPzlz7B1ZAQzqi7kQFguU+o+ZnGktsaR5z2dEdv3XmPqHX9mnXkSl9gn4G3eC7oodH6Fkk+jSJlRTL15HNte+wtpUxeIhvWCIAjCGeWvaUYN9HYhKaEGDPhrtH0pNdicjDk0mhvfuJ63F75HQ0R7MifSEcEVHy9lzKHRGGxOTfFjE9qn0yjo2FHT/UXYyeO0mHrDMr7kbQKPlRHe0N7TxxHhRH9nElNvWKY59uiEqcRak9g/IoecYQcYWNjeUPl4Rj6qrBJnS2Z0wlRN8atTksn5zyVMWPQaqko3ixBAzuZLqJ6e3H2AEOhkHdnzn+Tej7/fY/PvWfP/jk7uuylyd8YmzyA853E2GHw99luKUILjtJBcVcxNXstj2XfwaDdN/O8a9Qhzk9fS7NJWFaWkTcKvj0Xnq+5x1dOAIQ4lbZKm+AB790H9kUuRm3QYJt3Xqf9eEr7tv8VTtJi9+2CChkXcAkqAJ00PkrQihis+XtqhkXl9eD0rF75PubmWScplml/nM8k2OR5DkhVfubtrHgdAAkOiFdvk+G4e/HpQbaFN9w51XBeyjqg7vomJv/bYe896xw8JaH19ZR2eOffyo/dv55ZmM2+bmimRVFJUiWUeM0ZkmuffC6fw/vEVhfZvSajjOgs1Gao1aaqPtxC2cQFbZq1mXMJOrJXxKD4DssGHK7aSvblZhG1cgP5W7UnZaGs8m1K2AlO44xtvYK2xtE07d8c086gayaaUrSy13qUpfky0nsa80MZ9HXw99kI4qz7euIr3R5dgCFhJnp6JY98YPPWrcZtkakaY+L38DgProrj9hz8827v69WawMPSHN7H/xZepGjeXcG82fx72AT/9QIfjmJH88bGMTUpjy453yNv7BYPHaVseWhAEQRBCoY8JbXJiqOM6s01Jwmg/wZhDoxh1eCTHM/LbeoINLByArEoY7Q3YpmShZa3BeUvn88w/y6npZfpZrKWCeUvna9r/k029YRn+63zkfLIRV2kttuRopl48S3NF1/+3d9/xUVVpA8d/905LL6TROwmQAIFIifQiIoIC9sLq6oLoursi+oKurq5iB3YVVxcb6toFC4pSFJAmTZDeawIppGcyfea8f8SMhASlJKQ934/5YO4998y5z5y5M3nmnHPLGHQDf+4znX8u+xNKVxxsc+i0Ehr39H7yvBMIUdE6swtvY9rXLtoMWool9Ndv8l3WUA6vuIzZ1tu4I/rCRuT0b30ljPgvE378O82LcvzJqONh0UxKnVW6/3w178U01YyHteP4NMqtt1SWuJiqmkHz8xtxUpYcGNr0Owb9xiL+F5JEcF/xKIav/lZ5wlED9xWPXNAf+bm/LPztOzYSZ9rl6LHrf71jZXZvUIZy5c7V9qx15NgyyOmcccakLCWl5ZKb1L7Pr5pBp8njPTk26YfSIaunJrx+eT6aPN6zSkamVRdvsxR8IY3RrFlolWTsFBoqNO6CRkZ5OwwnaDIkd30G616zPwkSkuDGNXQa3g4Xdvc8b4fhOEb/G8vyZ7jJ+uvcXl9oYxyDHrrg+s2tzu4GA2db7nTVnTQtqz905XD2rxpCSY+NEJ4HhY0I/ug2QpURU5MLS8qWfcGystkGVnk1uhqTiHYpcsyKbVH7UYYL+4Il3N3rl2tPDJVPEvRBYHZpuVpAkl0NXFZeBt8VryUw0sItHkXRlhjchV8BPrYlx7LZ5yTHUEJKi9tp1er8brPakLRq5mT43gxeHbCFPx5L4W+LrdjCWxBUeITg3Ykci4yi7U2prDAF0u60D0RCCCFEVTI1Dq7ScqfztbyEViNeZf+8wehKo/2RU6dplf6l0GrET/jOc/0Rs8XMDeN+5JUPrkbDVy7hVTb97Pqx6zBbrjmv+k9nNJlIHnXmdU7OV//WV/LYkDf4z7pHyLH9OiInJrgp9/R+8oISRd26wvHWLXn2yASuf6cbrRrvxRRsxV0SwpHMBD6N6sWJNi3p1rVqzuPSliPYnrWOPFs2jYJi6RLX58JH+ugGUgc9wzOL7mFWoJWTp3w2ilFwvz2EPiOeOe8RJ6cmESpfxF/DVwVJBMfoF7EsfxrN+ut6Vyq0Mc7BF/5HftSpNwNVBnxZl/5+uXOQZ/t1PbTKk7IVy9U24Ve0ouV/B5Lx+EbcGb9ObTY1DqLJ4z0Jv6LVbxxdC/wyMirgq/t+Wd3v12xL2R0snYMubGQUlPZVe7uhmI//hKXkJCo4BnuzlAuu99T6be2GYjj+E9ov9XurqP6oP8STOf2n0gWuzkTXiPpD/HnVX91J01Pr15WR0E2pVVo/nPYFi0GxNXbbqS0AuKAvWGxWI1pwOsoeA/gon/AqfV/Wgo9js17Y+n5VRZJdDdyXiz/nx55BPPlpIREOhSHuOhz2reRbfAwquB2VvowlSnFls/0EtpZk1+9RYc2Iuqo7l6zeyZZeLUnIgqDidDRTEwIdGfiamWHfYZJiFnJC9aCZJLuEEEJUE/+31BlnXtPqgr5F1g0E330z8Y5XOLJi2GnTYopLp8Xcfc/5T4sBbpp4DTCfjz9LJeeU6WfRgVlcP3bdL/trv+pKFBkMGvf9Bf7+j5ZsDWpOe0c24fl2Cg2BHGgRi9J0nrpXw2Comg8cBt1QLSN7vB2G04dX+Gz502yzpZOrKaKURtfA5nhGPHxhyaLfSSJoVF0Sobr+yO/WFWJi4OTJM5eJjeW8k5qNgs7uGnC25WpK+BWtCBvegpIN2Xiy7RhjAwnuFVurR3Sd6tSRUdopI6NUaBzOKhgZ5acb8LaoxpE31VS/bjYSPaETOXN2nbFM9IRO6ObzT3FUd9L0YiRlz/wFS5ML/oLF7QGVk4IW/ROqpHnpHUnLBGajBR9H5aTgPp8V/KuBJLsasC3bNvBRwhEG7dGIsPkwBDTG692GweUkrOllmDQnJzI/5Vp9GH1Hnd9Qx4ao8R030G34F/yr9w5atbiagCPvooe0RBkC8W4+xP5+PWn1+mKsKSYYMa2mmyuEEKKeKvctNVTL1J5fp8U8i3Wv6ddpMR3duIZc+LQYKE14XXObi+8WfEtOlo3ouCCGXXVZlY3ouliqK1E0cIDGU0/Av2fr7D95ynpUsfC3ezUGDqgb36x5OwzH224onU9JFjmrKFn0W0kE45VP4W1a+Uipc1ZNf+T/mtQ884iWv11AUrNs6lOOLZMzzd+KCW5y3lOfLibNoBOSWjtGlZyP6kya1gdN/l56t8ic13eXH+Gla0RP6OTffyGqO2l6MZKy1fUFS1QjOHKkNOEFXrTIHWCwgzcQld8JZW/sL1cbSLKrgfJ5fXy+7j8EJ7dh1OZsTMHBeFzgzd5ISXgzYn19WOZ5lvD4qxkb3U0WUz9HbSbF8+dHTuJ57WuKjYrAgxuxhA3E6T1B5Bad/L152JO/xr1jEElJtf+DgxBCiLrpYnyLXN3TYqB0SuPI6y5gXah6buAAjX59Sxcyz80r/UOjW1eqbETXRVONI07OlESIjIqG/Pxqecyq9GtSU5Ub4VUVSc1Tpz6daf7WhUx9Eueoukde1XFN/n4JcQ8mk/vuPlxHizG3CiXqD/EXNKLrdNWdNL0YSdnq+IKlZUv4abP/EVD5SWcsVxtIsquB+nzh27TZ7aRPHljcCoIvxedcjcJNVMjV/By1g/cSTKxsPZHQYHljO1fho+IJe3Qtm35K4FizCAaob/HYdhEY0AVj8UYcSYNpcnItOwunYuz4Ax3llSiEEKKaXJSpPfLHWY0zGLTzuhNfg1LH+2l1JjWrc+qTEFVNNxuJ+VPnmm5Gg3PPXfD5F2dXrjaQP7EbIFuxg49dSxgffim6cR8AujkAzWNnU8coUswOXktaR5dVoYQmSqLrfGhmI9F/SyVx4/u8PzCKQXs743HswxQ6GlWyhcBiDwc7JWIKOoZt0+vQZ0JNN1kIIUQ9Vten9gghSlVnUvPUqU9OrQSLCq6amxAIIeqFwECd/n19rFpz5jL9+5aWqw1qRyvERfXc0qfYm9yW1YknCC7yoltaYytaTVGQGW//PP7e4WM8div3euWN7UJEjU8gaEsXbp2bz6HWofg0D+muLzAHJQMQsTGUzk4T3xbM4qS19t7dRgghhBBCNAxlU59GdLqB5CZ9JdElhCjnmad0+p9hdmT/vqX7awsZ2dXAbNu4msTle0jcHU+MwQ4Y8TqPYAC8rYdyoiiDovAsHvBcQ88/davp5tZpeqCR6Hu7E/+igaOpblxdLMT+nINHK8Id0gaj9TD529tw82Venl1zFXcOXEbngKCabrYQQgghhBBCCFGpZ57Ssdt9vDIH0tOhefPSqYu1ZURXmdrVGlGtnC4H25fNp3mLCcTZ8zCWuDCG9gUNSiJaE2foSf7e7URmhDC+fTKaLt3jQjW6qQOhEWZ8gS7+1TocjKFohjDCI66hMCQQZ4mJw5vb8ZekIGatnojd66npJgshhBBCCCGEEGcUGKgz5T6df80o/be2JbpAkl0NyhNfTuH5a4M47l6BpoFmjMbr3IvbaCQm5BpyD37C5QdTedSRKndfrCK6xUDLZ1NJ+j87sd6T/K93MJaYIeiGIKJCRvFFz1BMR80c39CESW0P8+jaqSh15ttKCyGEEEIIIYQQ4rdJsquB2LxvJataaSQctdHYkV96U2FPAcqZSWDElWSZCtnUszGOnAgG90ut6ebWK8G944i8LoWJM+LY2N7Np6lbcNv24s38lAF5CRyPNODbHIr5eDjFmfN49ed/1XSThRBCCCGEEEKIOkuSXQ2AT/l4atd/KYwO54+rHADoAe1QeCiObIk5sAuvdV/Fd80yueqeLhjM5hpucf3T+KEUaDmcEdbGLIrII8uyF00PoYUtlVCnEa+mUfhDCy6J68BnW57n+4Nf1HSThRBCCCGEEEKIOkmSXQ3Ac/M/5lintrTdFwmNotAMofjc2ZSEGpk9BOZ12cmR8BwGRI8mqnX7mm5uvaQHGun9f8kMfTWRHvs8vJl4FFtEC3RjCNFh1+PTwOvSiV7djt4dEtl47DGO5u+t6WYLIYQQQgghhBB1jiS76rlVq+ysC1+NUhr9thcSYg1BoYO3hPWdulISHc03cSsZm9aaR5on1HRz6zVTbCD9H0rh7qfbcTLUyD8uy2Vxs/W4sr/EHNAeg4Jou06nb2KJDo/k0e/vpMRVXNPNFkIIIYQQQggh6hRJdtVj27Yr1ux4nsPtIvjHR3n0Co5EdzvBW4izaV+6l/TBbcumdVorpl5yEwEGWZS+ullah9H2k5sZaLgRW6CP49GHQTkxGKLJbBWHR4dWxgh6FsYRac5jxuq/yYL1QgghhBBCCCHEOZBkVz22fOU3LO7t47J1TiK9PrSsNHzuTGzdIDDgUjwBAUzaF0+3qBuwBAXXdHMbjOjYYPo3b82oVW3YHOJED26Fp3gdTe1teX1UGPuamslb0Yjm2S24rsMBvt0zo6abLIQQQgghhBBC1BmS7KqHlFLsWLaarMi9uCxurtzvQAHK58AU2ZpMwy2YfDrG3RvJ7ziGOzuGo8mgrotG0+CKiCAeu+FWxr6bwKeFPTAGJeLL/pG7FxlpEngLbqNOz4PtWbvHS8fIRezK/Kammy2EEEIIIYQQQtQJkuyqZ3bs8PD2X+eyJLwZA4u78dAXJZg8XjSA6Fhc0aPomtuO+W03stDWn46tLLQ2yDS5mmAJNHPJxGR6BSeyp20ndHMzcOURmLkEm0mhG3102Z5AhsNICDPJse6s6SYLIYQQQgghhBC1niS76pGf1mSzYcYTNInRcX67D1fOCsIdLgDMkQkcGHOIE4EODoRlYs3TibusOcOMvhpudcPWtWUHmvbP4NOOKylodw2G6CswlmRgwYjVouPKt2BY1pQjTiM26/24PRk13WQhhBBCCCGEEKJWk2RXPbFs3s/8/MnTNBl+J/NbHcLTajERjhwUYIwYgmlwCS/6WvNC9wV8F7UZ68GrmNZfwyjTF2tUtFFjYP/OPNHUxNK929nqdWFvfjkhJT5CXBpKM+I+HED0jzFkOSErdyLKl1fTzRZCCCGEEEIIIWotSXbVA0tem8+h9bM5/I8XKMnOZ9xONwP32QENPbw/AW0b81VBIhaPCUt2BGlv3ML0P+sES6KrVtA0A93DJ9G3fU8OksueCBvoARjMjdEi+4EeQO7OUGIOh2HKd5CbMxHlK6jpZgshhBBCCCGEELWSsaYbIM6fz+vl57ef5nDaMY4+OpeWH31JdNb3GCxOAAJibyW67w/M9mRy8/Yb6Jhu4c0vU/n742ZaSKarVtFNUaQ0ceDrFMJblhP0KboX3WfEZrajhfVAZb5P2vcQl2qiSUY2WQl30zjwSYhsX9NNF0IIIYQQQgghahVJdtVRHo/if2/m4XK5OTplDpe8NZuo3B1ovzyjpvAhWGLN/FvTuGHnNTjc+axe2pI7x4fTN0kSXbWNpmm0GdqM1r6eHP3xO15KWsxfd43EXLAbt+MImteD0i1k/RhOUeNAstxbKSx5gduaJOPufRdoMkhTCCGEEEIIIYQAmcZYJ2UeSueBB138ZPCRf8XdXP7SP4jJ3YEGaIAe0AZHqxZ81mIr1229gSKzjdd+NNKyZyduuFoSXbWVpmnohnj+1GoEtnlDmJG4AHwuDMqHsfGt6M3uRLNEYc80E/lTDy5NT8e6+R1Mn94KjsKabr4QQgghhBBCCFErSLKrjjmwI4dv/jKCkH6bSY3dSt+3p+OgEJc5pDTRFRiPo1tf/p78NV3SUig021nr+p7YuHgm/00SXXVBcERLpvUKx/rVMN7s5ySjzUAcPh+BPguGRlej9EC8NgNHclozv/XDaJm7sbw+GP3o2ppuuhBCCCGEEEIIUeMk2VWHbNuuePbRLBz3PcXgvStoufFzCkIiCDK2wuzVMcfciNbkSh5vvwLd4ealrt/yo/k9Hhh6Bc8/q2GUWy/WCZZQI71uiOXft7cm/cVBzIn7iqmDPuGE8ys8WR8SGHMTmEzoPi/N133LnKSnOG5phvnzSZg2vgleV02fghBCCCGEEEIIUWMk2VUHeF1ONv7nUawfzeWK/v+ly/LvMTpO4AloRITNCuFNCGz8B3Kio7EZFQ9s60NhiInIHQHcGjWSsJbt0HVJdNU18R0sTP1DNKbn/8xtW/oTFTQKe/MrMQY0J7jxFBxBcWgK2i//irXOPnwQ9gimFbMIemsEhkMrarr5QgghhBBCCCFEjZAF6mu5E/sO8dHEUVh7dqCz10Rsvgdfo0YYPU3xlexG4eOkyUZwsE4TWzg+lUvA/g38oaiEnuZUOg7uV9OnIC7AoBEWzAEaM//VE/t12yjMS8PQ3cSw7T4CbdkYQnvhsW4iIjsdSOd/jaeQePhHUj6/G3fSWLjx1Zo+BSGEEEIIIYQQ4qKSZFcttvLVT1i94VFaxfUmuSgcTBkYAhuh2Ux43TvwGgwcSBhEV0cvTA4j2xptwrp7J93UMW5IGEvzfpfV9CmIKnDpIDORTUw88GgfKOnNH03rmNVnO8N3diHFOQRjaA+cme+Cz0Zs5kYygoN4y/QEhateJib4Xq7uNQ1dKbCE1PSpCFFvKKVQSlVLvVB6wwohhBBCCCHE+ZFkVy1UknaS156+m5jA4Qw1X48rJgfd4sPnDsBrz0MBx5u0ob3hGuK6zsF+KJJZTbaxLyKfW0rcdOoxgejEHjV9GqIKdUrQ+OhNA+sOeDDvaMfEbVF82Xw/b7V5i1inmQe+DcZQpMBnx+hw0dSxjOAWPbFvasfC5UMYF+HB+afv0QMjavpUhKjzlFJcccUVmEwmFixYUGWJqRKfkyvTZvFps3uJMYZWSZ1CCCGEEEI0RJLsqkFe91qUL50TXo3PXM3JtIM66SVixXGGF3bB3ngLrmBAM+JzevDqGiVBZiIa3UmCOZg9qXN50WzAnvQD8RkeHt8QwYirJ2EJi6jpUxPVIDRU47LuJkraxfDyHCOmlwN5MEkx+6E9PDLSxOPrbiHCEYC7ZA+eohWEpdsJZz0HW4/mjZIWHHvhCqJbDuaPPQYTkJAqI72EOE82m40NGzb4/z84OPi86zrqzmGT4wjXhF5CsG5hfPilePFWVVOFEEIIIYRokCTZVZ2UD60wHYwBKKMFjAFgMOMtdpP5zGZKCrdQWOAgLyOES09YWX9jOhHhRbQtLsbWTkN5QANQHlwBUZiir6YxkWiGACLbLOVEQQGXFJkYqidyZe/rsVwaVtNnLC6C4DATUx+M5eekHHZldeb5Vcn8I/kzpgz8ggEnOjFuXxIWdzY+51FAo+2R48Bx2kReRlQzE7kF79J44cMctw4jdMCfiWzdvKZPSYhax1vooviH4wR2i8LULBTl9ODcX4g7PAgVev7JKJ/DiyfHjq/Eja/Ew/uNVrNI283lwUmE6AHcGTHwrOpxe12YDGaUcmHbXIjnpJ3wES3Llcnasg6vy0nT3mdXpxBCCCGEEPWFJLuqmFIK+9Zc3HYbHls6rbbeWX4/4HAFUvTFn7BGO/lhqB3bgCxicNA6101EPnjR0AANHYUPb0QKjcKvRKHICdzOZ9qT9DQkMmBTcyaMupHw1h1q5FxFzUq+Ipr4rFCy1mSTuqsJpv1tadzHzBcJ21AdIhh5dABReTk4cr8CXwnknyRvJeQZmnEk+grC2ppw77if7LdKONJ4KH1u7EBoeEc0vQWaFlDTpyfEOXOXeCk6UUxE6zAMJp2srVspOHiAmKvHYdE1CtYto/DwfuJv+hM2nwstIws0jUNROuZtJYR/f5yIa90Eth/E0XtW4dhzEMMf4oga3omc79wc3JJLy64OIi8L9D+mNd9JYFAgu3ak4zTbSEnoSFGR4vWZJSQ1O8nwCa0B2Hn5h/gyivEWmdgy6iSXhG8ndGkS1183kLvuG0yIwwZBFV93Sik0TSOzOJ2/TjvJbbcHkR9ynCbrHXRucoDlHyWQGhiIa2cugf2jyNm5iSYp/dC8DuyF+eRmnaDAuZ2TPSfRI8CIIc+NPcdNVEIQPq+XwmN7MVqCCG3a+pxifWL9CrwOBy0GjriQp0wIIYQQQohqIcmuKpJ7sITcfSXkFmRw8tgRCLBh1Kws8UQRuKsFkU0iKGltJl0zEntYJ2LcbtwBGiluN8bjpXX4zLFoBgcYIrGE9UIPageuk9g1DV/QT2yzvEJ+uoV/3/0WAeZ46F+jpyxqgaA4C23GteCf3IQr38a2/+4hoTCAI202sXT4Kgatvwt77I00T9+Lp/gnNJ8dvIX4slZRkAUnTS0oaRRGkDGUQ9+YiWyxkqCYAyjLQYwqmFz9KmKb9Cc8IBAwSBJMXBCfdxdKFWEw9vnl9wzQwtH1oDMe41FeMj2FNDKEEKSbsftcFPrsRHiMlBw9yfHNNvLztuLN3UB408Ho9GT+R0vZk/QtKcZRdM67lKWfvs8HbfPoYGvHw7m7+fHRx3h6WCva2n28tu8VMg61YU58B0aYixjs3cOGb75m5+iBNH7IR8viJWxrkocK6Yw+xkd+wI+sC/314rvhqzdRPYYQ+m0Y2oE1fBb3GgUJzSnc2o/jS77n48UriMwYRtP0OHy6jjPQyds3b+en3HDGdWmEy+BFzXuHloc3s7fjZezr1peOjgIGup385Yco4q5qScKe12n0o49i0xBWHEzj+55ruGFnGCfT43nXNJyA2E8ITz2MfVsoxd/MZ03uCfa360XhJVeCweBva7PDq/BsLyL/gIfUhJF4ix0sv/N22t9yI13vvL9i7L0ebJ5iwiyRFGels3zNG1w17nEAitOO4HU6zrkPKKXwuV0YzJZzPlYIIYQQQoizpalzvJ1Ufn5+lT14ZGRkldZ3vpTy4vNuRvfEoPnCUAYzJV4r2fY8zEYLRQd8pO3ykJvSHHPRYbTAMDKKC2ieth2fxYVHMxJY0JSgkCTMednoBbtAc6HhpVDbjdlmx+jV0JWnkkfXQTNhDO2FOSQZrzMNzRCEIbAtpsBMXC1XgXstmOJJ6j4NS4isswRQUlJCixYtAEhLS7ugNXPqE6UU1o3Z7PtiCRknVhM7+E9Y2+2hTYfvcObFkf/DGPaxjPa5ZjyeHDRPMeWW1jbGYDA3Q9M8eDUXjiQDWoQVs3LSpuVhTJ5CbAYvLkcE2pFkAgKuIiwuEotZYYwNxRgcUKfuIldbrkG1jSPfTUFOMU27ROJzaGRsLmbtkcV07deett5QMg6EUKBl0bGVA+PxfRQ1HUtwk0CMjiyUy8F66wFCTxygpfEQqqWZ/3xzgMFuC+17FWKMU6yaZqa7J4zwh9PxRhg59p/2xKUrvJ3s5DZqyyJbFL1yVlCgWVjf7U6K09xo2XYodNMtIoLirIMU5u/FFxPNld2vY/++r3mjXyDmCBt/OH6Uk5GFLGzVjLDMxkzcdYKFPbIpjjHR+6dBJBe35dPOH+JyGel6MoaAwCjSYws57DHRymqgkTcUR0QUms2HrSCLdAowXdKJ61e1xvndajak7CC8x3AeHnczAAseW4rRuRVTaC90UxQe+0EcJz8hMO52Njez4AxKJ377epzRl5GrH8TuyiLPtYH+uS0wFyjWtHPyg2khdx+8E6+rGT8PPUJ65nKSD7fEMSAZ1ToZtwrGGtYaV2gz0DRGzhqFZjCy+rpnSFr0Ke6S/ZiDray/8S0CSuwUR0YRlZ5JROZmdhatwu05RNuCFrQ5FITvZBBRJ0KIygpCzy5NXv8vrhubQzsw8abvCegeTvsF7/PCd49gHvhvVNuDXHawF8E/5bI7cQm6bqGpFkejHoMpvvpKjnl9RFo6kYOZQHxMMjjxecBZrLDmOfHle8jJsFIyuBlZwUYK7+mHtUs/Wk68j1CVR4whFKUg3tIMU925dPyuyMjIcyp/IdchuY5VTuJSkcSkchKXyklcKidxqUhiUjmJS+XONy7n+tkKammySynFsvTVZE67Bz0gEC0gEENwCMawCBqNuAbd7kBlZ+I+nkZAaAQBYVFYA72kGbIJCm5EQEAIuisYkxZFp4herHcepOTEWkryinDZdaI6ZOLUdezuYI5aiyhZ+AiRIUW4vRm0tO6hQ5cYGrcoRMPNfncEuWYDSlOUmDXCCnJpcvwIx8eZCNqcRfjhQEJKXByJ0jECjQuNmN3OM5yxDnoIujkOwrpiDYtBMwRQFFDMwcjNtM/JJt4YycmMNIoi4rBEjeD663pRbC2uspjXF5LsOjth5hDWrDzEvk9+oPXlazHa+1LSYiPNTRa0kgCOr4mnyLGaQKcPcJ9FjRpoJtAD8RoMeA06LjOYVTCaMQg9IARHiAvltWHQDTh9ZmxeJ6GUYNLcaJoHQ4gbn0vDnuOlICuQYEMJjWIVmtmMVw/AWGIEt47Vq3C4QvEpHQyBaLoXH+B0gdsYg8HowK05MJsV5qggzEYdd4mGZjKiB+sERRehGYwYVCOMxiCCghQmgwk3FgKNgVjCjUQ1iqKkwIlZM2IymTAaDJiMJkwmIyaLEZTC51WAojSHp35J5inwXzpVuX9QPnDZwFWC5v7lx1WCctl++bcEHFaUswTNZUXzOvAq8OkmlE+hDEbcuhm3KQhdGdGMAVh9Jqy6CadXw6UZcDhNOG3g8YFXc+HzWfAUGPA5SvD6CgnwlhAQ5qPAZsGRbSVGHcYSUoTmKkHXfXg1Hx6zF2XUwGCgOMhKkWYkw9meaGcUuY324bA46ZTVCrMtlmORO7EF6rTO8KLTiP2xCnuQEXQz7gATByMDMWoRJJwsQTPD0fAYmhcFMiItE0v4AVbag2haEky0IY6S0D0cNhpokx1OcHAHDEFBWN1uzLlHsTZvjssUhCqxElSYTUCoGUOxi3SjE02z4TAG0Ck7inxjJko5MHs0XAHBOC2heI2hxGYfxKv5MLldGL2/9mcF/F4eRTc1wefOILDxXfjcGXgcx1CefNCDUJ5cDOYmeANiyLRkEX50Cze8Nh+A9/75EcE5CymOak+AIYqi8HyC8jMw2I0EeRRuXwm614PBEAfYcHkKS9umW9D0AHyaF81txRURj8foxu1Kx+L0oZuMBJucOPFgbQT2mADiToZTlOvC6CnGHRFCqCEGrz2dzMaBRNo7EFIYREHgfjSHh/CSADRPKHvCs2jvDiM48CS2XRaszYKJcAPKzWrvMfY1ymdgQDht0gL4oXEOR6JtjN3bjkifC5fNhtfrwRSkYbAoSrKdgBfdoAgP8pHS5igrA02cbDKc9mE29KNGnJ6WmN0tiAzLx3WiJSfybGQXFNIudis2+xbWpbdlhboOvBouCkjo+xYtIsYS2jgau3cpVwHpn21nczC0yXbQtLGdpJvb4I41YfcEER7cH6PWBrxmdIsP3dAW+95ASo5lYiEKb4GT4rRD2HIycBfZ8FhL8JTYMYSbSZk77SyucVVDkl01T+JSkcSkchKXyklcKidxqUhiUjmJS+UuZrKrVk5j9OLj2YJFJHeKwuz0YHY4CLAVEph9iP+l61z67VaaHMspd8y+1oo1PSvm7R7yfY3rm7s42DuCVTG7+VvvFlzWppF//9FjLZl2MohjmcEEuVwM7LCf6JCr8fwS/9a//JRx2w/idhfycpGDwbYAwpURDEZcRiclJg2LL4gQZzjOgFCcAQHYLAasgSZywy0UhvpwmmzYnFY4uYCYw4UkaBBZ3JjIoL4UNx9L0vXR5UbGGE218ikSdYQh2ESXPrF06XMdcB2eAicl21I5uPEHDjTZirdVAU5HY5oGWYjWm+DOiMTm3oo1rBXBdgPmk1tQnjwUOuBDQ4FygdeFwQsGwGwH+PX1eGraMRCIOEPbzEC4wVr6SzaAHSOF/v0hv/wAcPqgSC9wak457/dj4TzlkLJ/syotWfb6O6fvASqpo3qP1ymN4RmPy4fosk2aEWyNUDTy3+fPeEr8Gv3y0xoNKKRDftwvexSQRaQ9utyjxB0p+z/XLz8lnNoHIAMofdo8x02k4AYKgAIiMjS64yttYE7pHQ3Lbq0RkJ9e/nQyS/9p699QWNqeU8vYSvilA1VCB3Q0vKAZ0YwxKG9R6Rp2v6yMWNqZDGjGEMwhI7Bnzqm0Jo87G0ogDjh18p7p+KcENbsdY+FKlC8HS+avbfFR+hrRTbGAQnm9GH0Kj27AqBQoMOjhGCL6o1nXEWiKwecKQnltKKcTN6BjJCwfwg56gFx/rLBaAStGcxPi9SuwZ76FZoggwlvgf3wFdHCArhfjtnmxNSmmyGzFZwjDbdT5dkgseaFNyM1ykxDtZmnXeDwGjUda7sJxPACvS0N5Dfg8GsqrkaV7iQoIJkg3opsUrhtclE5I3Vz6VLT1YlJbMXsU3kI3V+Xs45kb2hHqcPN/648B8MYEH7ebXkDTQNfApDR0fT664df3PWX1sPn9Y9iKNbrN7AJ4MOAhwGQH3xf4yuLrBKN5PJsfX8+x4vl02fkMGhppTT8lP/Kn8j3BGUAKFy/ZJYQQQgghalatzKQYNQMLEh/D2smJW3lw4cWtvLiUh37Kja1PDrbiPBwnM3EV5uAqzKd1US5xBbk4WjfDa9DwlRShFzs56nsfY7sWNHLCAEc/1h9rws8bd+PFiw8v0c4Shl5xH8pUiM9o54imSDNMwqACMCgjOINwqI6YLQqbOx+3vRivrRnXrCpED3KyLyiAbGN/ggLzcVtOcDRUw2wIoGXLfIKMGvaiVrTQYxnVpgkRYc0wGgPQDRqaBgEB1KkpX6LuM0ZYCB/QjB4DbqYHN5fb53HYyd62jvyClhhtOexqppHRvClhhY3QDAaCipyEF59E8zo4adJwWewYTSaaFGn4LGFY9AB8yoTmVTiDwjH6vJgdJQQ4ijDowRg1hcvjQvcpvEGNwVsATitGrwfdGI5BD8Hnc+PVFZpuRlcGlKcAvA40NDQUPuVG00z4jEFonmLwOdFV6WgzzRBcOiVZN6PhQ/P5UL4SQJUm6dDwKS+aIRCfbkT3lIDPjaYZQTOgGULwKTdKt6Ch0JQX5S355XYR6pf/FJoxGK/mQ/fYfzne/EsdZpSuoTRTaUB9DvCVtR2U8qE0HYxh+JQdzetAU95fjjeDpuHTzaAZSkeH+WzgdaFpBjTAp3ygm1CmYPjl3DVF6fG6pfT8dAtoGprPhc9nQ1MKTdPR0PDiA0MIyhQErlzwOtE1Y+nxGPHhRZnCQTeju/Pwea1o/BIbwIsXTBH4dIXuLAJlR9cC0DUTKB2vDl5LIzQ86M5CfMqGpgeia0bwKtwBAbhNQRg8DnRnCV7diclnRMeIpptwhEajOYvA50Hhwae8WDwKXdfxAZolAt0SjEvLx+e0Y/CAHhqH1xKIchSilBt7XBz2QCOm/Dx0fNgaN8WrFOaTWXh1DXtkFLpSBOVlEugtJjrAhTdEx+myExiRTkjwAdzpOt5cGxZvPoFBYTjNYZjaujBFgPdEJIWHPPhsRb++cJJDKWy3krzAMBzFUXgLY3ErL82tkfgMAWQEBmA36Hg0yLEUcyIgF90NbYoj8Boc2E0uQjy5hLi6EuiyoOkdsbgsBJY4MZnz2dXSS1JmODGuDKyOYra1jKVVZiARrlx8dg/FzZpi9G4iPDAWb0422U2bEKafoIWrhOLADDJahpNKARz3kX/cQ3G0GUNcCBFxBh7XfQQ4dSxRRkxRXq7PcYHXjMVkISROcULLxRfqJMBiJOOEl2zdhqcxtIwNISjIyL6jYPQYCXUFo3xGDntzCVPhBFmDObHNyzBvAfkH3Hi8Xq5yRRLVSKMoy0Wxw4UBnUCLhZCgIDxOA4V7zERFJmCKtoLHQ2JwCloIHPsuD0eOBWd2AOZOkYQ2UjiPgCvLS8I9EzGYm9F+QluiD3Wl9cwrMTYKonnxJbjshRgsARjMFgwmi6wRJoQQQgjRwNTKaYziVxKjysk0xrMj/ee3SXx+m8SncnL9OXsNtQ/JNMaaJ3GpSGJSOYlL5SQulZO4VCQxqZzEpXIXcxqjfs5HCCGEEEIIIYQQQghRS0mySwghhBBCCCGEEELUG5LsEkIIIYQQQgghhBD1hiS7hBBCCCGEEEIIIUS9cc4L1AshhBBCCCGEEEIIUVvJyC4hhBBCCCGEEEIIUW9IsksIIYQQQgghhBBC1BuS7BJCCCGEEEIIIYQQ9YYku4QQQgghhBBCCCFEvSHJLiGEEEIIIYQQQghRb1RrsuvVV1/lxhtvpFu3blxyySVndYxSitmzZ9OvXz+6du3K+PHj2b9/f7kyLpeLJ598kt69e5OcnMykSZPIzMysjlOoVoWFhTz44IOkpKSQkpLCgw8+SFFR0W8ek5CQUOnPG2+84S8zfvz4CvsnT55c3adT5c4nPtOmTatw7tdff325Mg21/7jdbl544QVGjx5NcnIy/fr14//+7//IysoqV66u9p/333+fIUOG0KVLF8aNG8emTZt+s/yGDRsYN24cXbp0YejQoXz44YcVyixevJiRI0eSlJTEyJEjWbp0aXU1/6I4lxgtWbKEP/7xj/Tp04cePXpwww03sGrVqnJlPvvss0qvR06ns7pPpVqcS3zWr19f6bkfPHiwXLn61IfOJT6VXYsTEhK48sor/WXqW/+52M71mleXbNy4kUmTJtGvXz8SEhL47rvvyu2vqs+K5/M5oybNmTOHa665hu7du5Oamso999zDoUOHypVpaLH54IMPGD16ND169PC/V/3www/+/Q0tHmcyZ84cEhISeOqpp/zbGmJsZs+eXeE9p2/fvv79DTEmZbKysnjggQfo3bs33bp14+qrr2bHjh3+/Q0xNkOGDKn0c8o///lPoGHGxOPx8K9//YshQ4bQtWtXhg4dyssvv4zP5/OXqVVxUdXoxRdfVHPnzlXPPPOMSklJOatj5syZo7p3764WL16s9u7dq+677z7Vt29fVVxc7C/zj3/8Q/Xv31+tWbNG7dy5U40fP15dddVVyuPxVNepVIs777xTjRo1Sm3evFlt3rxZjRo1St11112/eUx2dna5n3nz5qmEhAR17Ngxf5lbb71VPfLII+XKFRUVVffpVLnzic/UqVPVnXfeWe7c8/Pzy5VpqP2nqKhI3X777WrhwoXq4MGDasuWLeq6665TY8eOLVeuLvafhQsXqsTERPXJJ5+oAwcOqOnTp6vk5GR1/PjxSssfO3ZMdevWTU2fPl0dOHBAffLJJyoxMVEtWrTIX2bz5s2qU6dO6r///a86cOCA+u9//6s6d+6sfv7554t1WlXqXGM0ffp09dprr6mtW7eqw4cPq5kzZ6rExES1c+dOf5n58+erHj16VLgu1UXnGp9169ap+Ph4dejQoXLnfup1pD71oXONT1FRUbm4ZGRkqF69eqmXXnrJX6Y+9Z+L7Vyfj7pmxYoVatasWWrx4sUqPj5eLV26tNz+qvqseD6fM2rSHXfcoebPn6/27dundu/erSZOnKgGDRqkSkpK/GUaWmy+//57tWLFCnXo0CF16NAhNWvWLJWYmKj27dunlGp48ajM1q1b1eDBg9Xo0aPV9OnT/dsbYmxeeukldeWVV5Z7z8nNzfXvb4gxUUqpgoICNXjwYDVt2jS1detWlZaWptauXauOHj3qL9MQY5Obm1uur6xZs0bFx8erdevWKaUaZkxeeeUV1atXL7V8+XKVlpamvv32W5WcnKzefvttf5naFJdqTXaVmT9//lklu3w+n+rbt6+aM2eOf5vT6VQpKSnqww8/VEqVfoBOTExUCxcu9JfJzMxUHTt2VCtXrqz6xleTAwcOqPj4+HJ/9GzZskXFx8ergwcPnnU9d999t/rDH/5Qbtutt95a7s2sLjrf+EydOlXdfffdZ9wv/ae8rVu3qvj4+HJ/INXF/nPttdeqf/zjH+W2jRgxQs2YMaPS8s8//7waMWJEuW2PPvqouv766/2//+1vf1N33nlnuTJ33HGHmjx5chW1+uI61xhVZuTIkWr27Nn+38/22l4XnGt8ypJdhYWFZ6yzPvWhC+0/S5cuVQkJCSo9Pd2/rT71n4utKl7PdcXpya6q+qxYVe+jNSk3N1fFx8erDRs2KKUkNmV69uypPvnkE4mHUspqtarhw4erNWvWlPt811Bj89JLL6mrrrqq0n0NNSZKKfXCCy+om2666Yz7G3JsTjV9+nQ1bNgw5fP5GmxMJk6cqB566KFy2+699171wAMPKKVqX1+pVWt2paenc/LkSfr16+ffZjab6dmzJ1u2bAFgx44duN3uckNO4+Li6NChg79MXbBlyxZCQ0Pp1q2bf1tycjKhoaFnfR45OTn88MMPXHvttRX2ffXVV/Tu3Zsrr7yS5557DqvVWmVtvxguJD4bNmwgNTWVyy+/nEceeYTc3Fz/Puk/5VmtVjRNIywsrNz2utR/XC4XO3fuLHfdAOjbt+8ZY/Hzzz+X6wMA/fv39/ePsjKn19m/f/861U/KnE+MTufz+SgpKSEiIqLcdpvNxuDBgxkwYAB33XUXu3btqqpmXzQXEp8xY8bQr18/brvtNtatW1duX33pQ1XRf+bNm8ell15Ks2bNym2vD/3nYquK56Muq6rPilX1PlqTiouLAQgPDwckNl6vl4ULF2Kz2ejevXuDjwfAE088wcCBA7n00kvLbW/IsTl69Cj9+vVjyJAhTJ48mbS0NKBhx2TZsmUkJSXx17/+ldTUVMaMGcMnn3zi39+QY1PG5XKxYMECrrnmGjRNa7AxSUlJYd26dRw+fBiAPXv28NNPPzFw4ECg9vUV4/mfatU7efIkAFFRUeW2R0dHc+LECaA0wWMymfxv7KeWycnJuTgNrQI5OTkVzhNKz/1sz+Pzzz8nODiY4cOHl9s+evRomjdvTnR0NPv372fmzJns2bOHuXPnVknbL4bzjc+AAQMYMWIETZs2JT09nRdffJHbbruNzz77DLPZLP3nFE6nkxkzZjBq1ChCQkL82+ta/8nPz8fr9VZ63Si7ppwuJyeH6OjoctuioqLweDzk5+cTGxtbaYyjoqLOWGdtdj4xOt1bb72F3W7niiuu8G9r27YtzzzzDAkJCVitVt59911uuukmvvzyS1q3bl2Vp1Ctzic+MTExPPnkkyQmJuJyufjyyy+5/fbb+d///kfPnj2Byl+ndbEPXWj/yc7OZuXKlcyYMaPc9vrSfy62qng912VV9VmxKt5Ha5JSimeeeYaUlBTi4+OBhhubvXv3cuONN+J0OgkKCuI///kP7du3Z/PmzUDDi0eZhQsXsmvXLubNm1dhX0PtK127duW5556jdevW5Obm+teX/vrrrxtsTADS0tL48MMP+eMf/8ikSZPYtm0b06dPx2w2M2bMmAYdmzLfffcdxcXFjB07Fmi4r6EJEyZQXFzMFVdcgcFgwOv1MnnyZEaNGgXUvricc7Jr9uzZvPzyy79ZZt68eXTp0uVcq/bTNK3c70qp3z3mbMpcDGcbnzNRSlU4/zOZP38+o0ePxmKxlNt+6oLs8fHxtGrVimuuuYadO3eSmJh4VnVXl+qOz8iRI/3/Hx8fT1JSEkOGDGHFihUVkoKn11sbXKz+43a7mTx5MkopHn/88XL7anP/+S2VXTd+KxZnus6cuv1c66ztzvd8vv76a15++WVeeeWVcm88ycnJJCcn+3/v0aMHY8eO5b333uORRx6psnZfLOcSn7Zt29K2bVv/7927dyczM5M333zTn+w61zpru/M9l88//5zQ0FCGDRtWbnt96z8XW33qW+ejuj4r1pU4PvHEE+zbt48PPvigwr6GFps2bdrwxRdfUFRUxJIlS5g6dSrvvfeef39DiwdARkYGTz31FG+99VaFvxNO1dBiUzb6pExycjKXXXYZX3zxhX8ESUOLCZS2LSkpifvvvx+Azp07c+DAAT788EPGjBnjL9cQY1Nm/vz5DBgwgLi4uHLbG1pMvvnmGxYsWMDMmTNp3749u3fv5plnniE2NtafCITaE5dzTnbdcsst5RIKlWnevPm5VguUflMOpZm82NhY//bc3Fz/KIzo6GjcbjeFhYXlsoG5ubl07979vB63Kp1tfPbu3Vtuel2ZvLy8SrOYp9u0aROHDx/m3//+9++WTUxMxGQycfTo0RpPVlys+JSJjY2ladOmHDlyBJD+A6WJrvvuu4/09HTeeeedcqO6KlOb+k9lIiMjMRgMFbL8p143TlfZCIi8vDyMRqN/ml5lo/3y8vLOWGdtdj4xKvPNN9/w97//nRdffLHCNIjT6bpOly5d/K+3uuJC4nOqbt26sWDBAv/v9aUPXUh8lFLMnz+fq6++GrPZ/Jtl62r/udiqqr/WVVX1WTE6OrpKPmfUhCeffJJly5bx3nvv0bhxY//2hhobs9lMq1atAOjSpQvbt2/n3XffZcKECUDDiwfAzp07yc3NZdy4cf5tXq+XjRs38v7777No0SKgYcbmVEFBQcTHx3PkyBH/FzINMSYxMTG0a9eu3La2bduyePFi/35omLEBOH78OGvXrmX27Nn+bQ01Js8//zwTJ0703107ISGBEydOMGfOHMaOHVvr4nLOa3Y1atSIdu3a/ebPb32D8FuaN29OTEwMa9as8W9zuVxs3LjRf+JJSUmYTKZyZbKzs9m/f3+tSFacbXy6d+9OcXEx27Zt8x+7detWiouLz+o85s2bR2JiIh07dvzdsvv378ftdvs7X026WPEpk5+fT0ZGhv/F1tD7T1mi6+jRo7z99ttERkb+bptqU/+pjNlsJjExsdxzCrB27dozxiI5OZm1a9eW27Z69Wp//ygrc3qdq1evrhX95FydT4ygdETXtGnTmDlzJoMGDfrdx1FKsXv37lrbV87kfONzutPPvb70oQuJz4YNGzh69Gila0uerq72n4utqvprXVVVnxWr6nPGxaSU4oknnmDJkiW88847tGjRotz+hhybUymlcLlcDToeffr04auvvuKLL77w/yQlJTF69Gi++OILWrRo0WBjcyqXy8XBgweJiYlp0P2lR48e/jWYyhw5csS/zmZDjg3AZ599RlRUVLnPwg01Jg6Ho8LIKoPB4B+VVevictZL2Z+H48ePq127dqnZs2er5ORktWvXLrVr1y5ltVr9ZS6//HK1ZMkS/+9z5sxRKSkpasmSJWrv3r3q/vvvr/RWlQMGDFBr165VO3fuVH/4wx8q3KqyLrjzzjvV6NGj1ZYtW9SWLVsqvZ3m6fFRSqni4mLVrVs39cEHH1So8+jRo2r27Nlq27ZtKi0tTa1YsUKNGDFCjRkzpt7Hx2q1qmeffVZt3rxZpaWlqXXr1qkbbrhB9e/fX/qPUsrtdqtJkyapAQMGqN27d5e7la7T6VRK1d3+s3DhQpWYmKg+/fRTdeDAAfXUU0+p5ORk/53fZsyYoR588EF/+WPHjqlu3bqpp59+Wh04cEB9+umnKjExUS1atMhf5qefflKdOnVSc+bMUQcOHFBz5sxRnTt3LndXkLrkXGP01Vdfqc6dO6v33nuvXF8pKiryl5k9e7ZauXKlOnbsmNq1a5eaNm2a6ty5s9q6detFP78Lda7xmTt3rlq6dKk6fPiw2rdvn5oxY4aKj49Xixcv9pepT33oXONT5oEHHlDXXXddpXXWp/5zsf3e81HXWa1W/2fG+Ph4NXfuXLVr1y7/nYOr6rPi2byP1iaPPfaYSklJUevXry93Xbbb7f4yDS02M2fOVBs3blRpaWlqz549atasWapjx45q9erVSqmGF4/fcvrdthtibJ599lm1fv16dezYMfXzzz+ru+66S3Xv3t1/7WyIMVGq9O7snTt3Vq+++qo6cuSIWrBggerWrZv68ssv/WUaamy8Xq8aNGiQeuGFFyrsa4gxmTp1qurfv79avny5SktLU0uWLFG9e/dWzz//vL9MbYpLtSa7pk6dquLj4yv8rFu3zl8mPj5ezZ8/3/+7z+dTL730kurbt69KSkpSt9xyi9q7d2+5eh0Oh3riiSdUr169VNeuXdVdd92lTpw4UZ2nUi3y8/PVlClTVPfu3VX37t3VlClTKtzG/vT4KKXURx99pLp27Vruj84yJ06cULfccovq1auXSkxMVMOGDVNPPvmkys/Pr85TqRbnGh+73a7uuOMO1adPH5WYmKgGDRqkpk6dWqFvNNT+k5aWVunr8dTXZF3uP++9954aPHiwSkxMVGPHjvXfil2p0mvRrbfeWq78+vXr1ZgxY1RiYqIaPHhwpcnjb7/9Vl1++eUqMTFRjRgxolwioy46lxjdeuutlfaVqVOn+ss89dRTatCgQSoxMVH16dNH3XHHHWrz5s0X9Zyq0rnE57XXXlPDhg1TXbp0UT179lQ33XSTWrFiRYU661MfOtfXWFFRkeratav6+OOPK62vvvWfi+23no+6bt26db95/amqz4pn8z5am5zpPbw6PkfXldg89NBD/tdBnz591G233eZPdCnV8OLxW05PdjXE2Nx3332qb9++KjExUfXr10/de++9av/+/f79DTEmZZYtW6ZGjRqlkpKS1IgRIyq8dzfU2KxatUrFx8erQ4cOVdjXEGNSXFyspk+frgYNGqS6dOmihg4dqmbNmuUfOKFU7YqLplQtWZlbCCGEEEIIIYQQQogLdM5rdgkhhBBCCCGEEEIIUVtJsksIIYQQQgghhBBC1BuS7BJCCCGEEEIIIYQQ9YYku4QQQgghhBBCCCFEvSHJLiGEEEIIIYQQQghRb0iySwghhBBCCCGEEELUG5LsEkIIIYQQQgghhBD1hiS7hBBCCCGEEEIIIUS9IckuIYQQQgghRJ2VkJBQ7qdjx46kpKRw/fXX8/bbb+N2u2usbZ999hkJCQnMnj27yuo8evQoSUlJzJw584LrGj9+PAkJCaSnp5fbPmTIEBISEi64/sqkp6eTkJDA+PHjq6X+i/04p1u6dCkJCQl8++23F/VxhRDlGWu6AUIIIYQQQghxocaOHQuA1+vl+PHjbNmyha1bt7JixQreeOMNjMb68afPzJkzMZlM/PGPf6zppohKDBs2jI4dOzJr1iyGDh2K2Wyu6SYJ0SDVjyu+EEIIIYQQokF79tlny/2+detWxo8fz48//sjChQu5+uqra6hlVWfnzp0sXryY2267jUaNGtV0c2q1uLg4vvnmGwIDAy/q42qaxsSJE7n//vuZN28eN99880V9fCFEKZnGKIQQQgghhKh3unXr5h/ttXr16hpuTdX48MMPARgzZkzNNqQOMJlMtGvXjqZNm170xx46dCjBwcF89NFHF/2xhRClJNklhBBCCCGEqJc6dOgAQF5eXrntSim+/vprJk+ezOWXX05ycjLdu3fn2muv5f3338fn81Woa/bs2SQkJPDZZ5+xd+9eJk2aRM+ePUlOTubWW29l8+bN59S2t956i44dOzJy5EiysrJ+t3xJSQkLFy6kXbt2dO7cucL+7OxsXn/9dW699Vb69+9PUlISffv25d5772Xbtm3n1Lbz9cMPP3DXXXeRmppKUlISgwYN4p577mHFihWVlnc4HMyYMYPBgweTlJTEZZddxmuvvYZSqkLZTZs28cQTTzB69Gh69uxJ165dGTFiBDNmzKCoqKhC+TOt2XXqOmonTpxgypQp9OnTh65duzJu3DiWLVtWaVu3bt3Kn//8Z39b+/bty7XXXsvMmTMpKSkpVzYgIIBhw4axd+9etm7depbRE0JUJUl2CSGEEEIIIeqlsiTE6VP+XC4XU6ZMYc2aNTRq1IjBgwfTrVs3Dhw4wBNPPMHDDz98xjp37NjBDTfcwOHDh0lNTaVVq1Zs3LiR22+/nX379p1Vu2bNmsVzzz1HUlIS77//PnFxcb97zMaNG7HZbPTq1avS/d9//z0zZswgOzub+Ph4hg4dSmxsLEuXLuXmm2+u9tFtzz77LBMnTmTVqlW0adOG4cOH07x5c9avX8+bb75Zobzb7eaOO+7gk08+oW3btvTu3ZusrCxmzpzJv//97wrln3/+eT799FNMJhN9+vQhNTUVq9XK66+/zs0331wh4fR7jh8/zrXXXsvmzZtJSUmhc+fO7Ny5kz//+c8VYrVixQpuvPFGli9fTrNmzRg+fDgdO3YkPz+f1157jfz8/Ar1lz1PZ0r0CSGql6zZJYQQQgghhKiXVq1aBUD//v3LbTcYDMyePZtBgwaVW0A8Ly+PCRMm8Pnnn3PNNdfQs2fPCnW+//77PPDAA0yYMMG/7emnn+add97hjTfe4Pnnnz9je3w+H48//jgff/wxffr04ZVXXiE4OPiszmXTpk0AdOnSpdL9PXr04Msvv6Rjx47ltq9atYq7776bf/7znyxZsgRN087q8c7Fl19+ydy5c2ncuDFz5swp1wabzVbp6KYtW7ZwySWXsGjRIn8ycvv27dx444288847TJw4sVxs/vznP5OcnEx4eLh/m8vlYvr06Xz88cfMnTuXe++996zb/PnnnzN+/HimTZvmv3nBO++8w9NPP82rr75Kv379/GXffPNNlFJ8+umnJCUllatn27ZtREREVKi/a9euwK/PmxDi4pKRXUIIIYQQQoh6w+fzcezYMR577DE2btzIkCFDGDlyZLkyRqOR4cOHV7hTXqNGjZgyZQpQOlKqMikpKeUSXQB333038NuJDZfLxeTJk/n444+57LLLeP3118860QWwd+9eANq0aVPp/oSEhAqJLihN9I0YMYJjx46d9cizczVnzhwAHn744QptCAoKIjU1tcIxuq4zffr0cqPuunTpQv/+/bHb7ezYsaNc+YEDB5ZLdAGYzWYefvhhjEbjGacfnkmLFi2YOnVqubt03nLLLYSHh7N161ZcLpd/e25uLqGhoRUSXVCa1AoJCamwvW3btsCvz5sQ4uKSkV1CCCGEEEKIOi8hIaHCtmuvvZYnn3wSXa/8O/7du3ezevVqTpw4gcPhQCnlnw535MiRSo/p27dvhW2RkZFERESQnZ1d6TE2m41JkyaxZs0axo0bx/Tp0zEYDGd5ZqVyc3MBKiR8TuVyuVi5ciXbt28nLy8Pt9sN4E9yHT16tNI4XYisrCwOHjxIREQEl19++Vkf16xZs0oTd23atGH58uWcPHmy0sdatmwZhw4dwmq1+tf2MplMZ3y+zqRXr16YTKZy24xGI82bN2fnzp0UFBQQGxsLQGJiIgsWLODhhx/m9ttvJz4+/nfrNxqNBAcHU1RUhMfjKZdUE0JUP3nFCSGEEEIIIeq8sjsvOp1Odu/ezeHDh5k3bx7Jyclcd9115cq6XC4eeughvv766zPWd6Y1oBo3blzp9uDgYAoKCird9+677+LxeBg4cCBPP/30eU0ltFqt/sepzN69e7n77rs5fvz4Ges413WtzkZmZiYALVu2PKfjzhTHoKAggHIjqwDmzp3LzJkz/Qm8C/Vbz+Ppj3///fezb98+5s+fz/z584mMjKR79+4MGzaM0aNHVxghWCYkJISSkhKsVmulUx2FENVHkl1CCCGEEEKIOu/ZZ58t9/vrr7/OjBkzmD59OpdeeinNmjXz73v77bf5+uuviY+P58EHHyQxMZGwsDBMJhOHDx9mxIgRZ3yc80lU9e/fn02bNrFmzRoWL178m/WfSdlUubKk16mUUtx3330cP36cG2+8kZtuuonmzZsTHByMpmnMmjWLOXPmVHqXw6pyrnE5l/I///wzzz77LKGhoTz55JP06tWLmJgYf5KpX79+lY4Eq6rHb9KkCfPnz2fdunWsWLGCDRs2sHz5cpYtW8Ybb7zBRx99VOmIu+LiYjRNq3SaoxCiesmaXUIIIYQQQoh6Z8KECfTr1w+Hw8HLL79cbt/SpUsBmDlzJgMGDCAqKso/pS0tLa3K25KYmMibb75JQEAAU6ZM8T/+uYiKigKodPTYoUOHOHToEElJSfzzn/+kY8eOhISE+BM61XFOZcpGSB09erTaHqMsXvfddx9jx46lWbNm/kSXw+EgJyen2h67jNFopF+/fjzyyCMsWLCAZcuW0adPHw4dOsRrr71Wobzb7cZmsxEWFiZTGIWoAZLsEkIIIYQQQtRLDzzwAJqmsWDBgnLT+4qKioDSETun+/bbb6ulLd26dePNN9/EYrEwefLkMy6AfyZlC78fPny4wr7CwkKg8ql5hYWFrF279jxafHbi4uJo164dBQUFLFmypFoeo+z5quz8Fi1aVK0j1s6kadOm/hsVVLbw/6FDhwAqvWmAEKL6SbJLCCGEEEIIUS916tSJoUOH4vF4eOONN/zbW7duDcCHH35YrvyiRYv48ssvq609ycnJvPHGG5hMJv72t7+xYsWKsz72kksuAWDbtm0V9rVq1Qpd11m3bl25hdqdTiePPfbYGdcSqyoTJ04E4Omnn2b//v3l9tlsNn788ccLqr/s+Zo3b165NbsOHDjAjBkzLqjus/H2229XOnps1apVQOVJ07Lnqex5E0JcXDKeUgghhBBCCFFv/eUvf+H7779n/vz53HPPPcTExPCnP/2JVatWMXPmTBYtWkSbNm04cuQIO3bs4I477uCtt96qtvb06NGD119/nQkTJvCXv/yF//znPwwYMOB3j7vkkksICgpi/fr1FfZFRUVx7bXX8sknn3D11VfTp08fLBYLP/30E16vl3HjxvHZZ59Vx+kAMGbMGLZv3857773H1VdfTffu3WncuDHZ2dns2rWLzp07k5qaet71jxs3jrlz57J8+XJGjBhBly5dKCwsZOPGjQwdOpTt27f/5sL8F+rll1/mueeeo2PHjrRq1QqlFHv37uXw4cNERkbypz/9qcIxGzZsAGDgwIHV1i4hxJnJyC4hhBBCCCFEvdWxY0cuu+wynE4nc+fOBaBnz5588MEH9OnTh/T0dJYvX47JZGL27Nnccsst1d6mSy65hNdeew2j0ci9997LmjVrfveY4OBgRo0axdGjRysd3fX4448zbdo0mjdvzo8//shPP/1Eamoq8+fPp2nTptVxGuU8+uij/Oc//yE1NZX9+/ezePFi0tPTSU1NrTQZdC4iIyOZN28eo0aNwu12s2zZMrKysvjrX//KrFmzqugMzuyRRx5h5MiR2O12Vq5cyapVqzAYDNxxxx0sWLCgwp0oHQ4H33//PfHx8XTr1q3a2yeEqEhTNTHBWQghhBBCCCHEOdm9ezdjxoxh/PjxPPLIIzXdHHEGX3/9NVOmTOGxxx7j5ptvrunmCNEgycguIYQQQgghhKgDOnXqxIgRI5g/fz55eXk13RxRCaUUr7/+Oi1btuTaa6+t6eYI0WBJsksIIYQQQggh6ogpU6bgdrurdV0xcf6+//579uzZw+TJkzGbzTXdHCEaLJnGKIQQQgghhBBCCCHqDRnZJYQQQgghhBBCCCHqDUl2CSGEEEIIIYQQQoh6Q5JdQgghhBBCCCGEEKLekGSXEEIIIYQQQgghhKg3JNklhBBCCCGEEEIIIeoNSXYJIYQQQgghhBBCiHpDkl1CCCGEEEIIIYQQot6QZJcQQgghhBBCCCGEqDck2SWEEEIIIYQQQggh6o3/Bxd/2FjIGmaZAAAAAElFTkSuQmCC\n",
+ "image/png": "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",
"text/plain": [
""
]
@@ -396,7 +426,7 @@
"ax = az.plot_trace(idata_80, compact=True, kind=\"rank_vlines\")\n",
"ax[0, 0].axvline(-0.5, 0, 0.9, color=\"k\")\n",
"ax[0, 0].axvline(0.5, 0, 0.1, color=\"k\")\n",
- "f'Estimated w1 = {np.mean(idata_80.posterior[\"X\"] > 0).item():.3f}'"
+ "f'Estimated w1 = {np.mean(idata_80.posterior[\"X\"] < 0).item():.3f}'"
]
},
{
@@ -415,21 +445,22 @@
"name": "stdout",
"output_type": "stream",
"text": [
- "Last updated: Sat Oct 23 2021\n",
+ "Last updated: Tue May 31 2022\n",
"\n",
"Python implementation: CPython\n",
- "Python version : 3.9.6\n",
- "IPython version : 7.26.0\n",
+ "Python version : 3.9.7\n",
+ "IPython version : 8.3.0\n",
"\n",
- "xarray: 0.19.0\n",
+ "xarray: 2022.3.0\n",
"\n",
- "numpy : 1.20.3\n",
- "arviz : 0.11.3\n",
- "matplotlib: 3.4.3\n",
- "theano : 1.1.2\n",
- "pymc3 : 3.11.4\n",
+ "sys : 3.9.7 (default, Sep 16 2021, 13:09:58) \n",
+ "[GCC 7.5.0]\n",
+ "arviz : 0.12.0\n",
+ "numpy : 1.21.5\n",
+ "aesara: 2.6.2\n",
+ "pymc : 4.0.0b6\n",
"\n",
- "Watermark: 2.2.0\n",
+ "Watermark: 2.3.0\n",
"\n"
]
}
@@ -441,8 +472,11 @@
}
],
"metadata": {
+ "interpreter": {
+ "hash": "d4ca51fc2fdee62b1a00ff5126f64ae66836e25d3ba6f45d8551026256283997"
+ },
"kernelspec": {
- "display_name": "Python 3 (ipykernel)",
+ "display_name": "Python 3.9.7 ('base')",
"language": "python",
"name": "python3"
},
diff --git a/myst_nbs/samplers/SMC2_gaussians.myst.md b/myst_nbs/samplers/SMC2_gaussians.myst.md
index 0c226ddf6..6e4dc18f3 100644
--- a/myst_nbs/samplers/SMC2_gaussians.myst.md
+++ b/myst_nbs/samplers/SMC2_gaussians.myst.md
@@ -6,7 +6,7 @@ jupytext:
format_version: 0.13
jupytext_version: 1.13.7
kernelspec:
- display_name: Python 3 (ipykernel)
+ display_name: Python 3.9.7 ('base')
language: python
name: python3
---
@@ -14,17 +14,15 @@ kernelspec:
# Sequential Monte Carlo
:::{post} Oct 19, 2021
-:tags: SMC, pymc3.Model, pymc3.Potential, pymc3.Uniform, pymc3.sample_smc
+:tags: SMC
:category: beginner
:::
```{code-cell} ipython3
-%matplotlib inline
+import aesara.tensor as at
import arviz as az
-import matplotlib.pyplot as plt
import numpy as np
-import pymc3 as pm
-import theano.tensor as tt
+import pymc as pm
print(f"Running on PyMC v{pm.__version__}")
```
@@ -46,7 +44,7 @@ When $\beta=0$ we have that $p(\theta \mid y)_{\beta=0}$ is the prior distributi
A summary of the algorithm is:
1. Initialize $\beta$ at zero and stage at zero.
-2. Generate N samples $S_\text{\beta}$ from the prior (because when $\beta = 0$ the tempered posterior is the prior).
+2. Generate N samples $S_{\beta}$ from the prior (because when $\beta = 0$ the tempered posterior is the prior).
3. Increase $\beta$ in order to make the effective sample size equals some predefined value (we use $Nt$, where $t$ is 0.5 by default).
4. Compute a set of N importance weights $W$. The weights are computed as the ratio of the likelihoods of a sample at stage $i+1$ and stage $i$.
5. Obtain $S_{w}$ by re-sampling according to $W$.
@@ -110,16 +108,16 @@ w2 = 1 - w1 # the other mode with 0.9 of the mass
def two_gaussians(x):
log_like1 = (
- -0.5 * n * tt.log(2 * np.pi)
- - 0.5 * tt.log(dsigma)
+ -0.5 * n * at.log(2 * np.pi)
+ - 0.5 * at.log(dsigma)
- 0.5 * (x - mu1).T.dot(isigma).dot(x - mu1)
)
log_like2 = (
- -0.5 * n * tt.log(2 * np.pi)
- - 0.5 * tt.log(dsigma)
+ -0.5 * n * at.log(2 * np.pi)
+ - 0.5 * at.log(dsigma)
- 0.5 * (x - mu2).T.dot(isigma).dot(x - mu2)
)
- return pm.math.logsumexp([tt.log(w1) + log_like1, tt.log(w2) + log_like2])
+ return pm.math.logsumexp([at.log(w1) + log_like1, at.log(w2) + log_like2])
```
```{code-cell} ipython3
@@ -129,11 +127,10 @@ with pm.Model() as model:
shape=n,
lower=-2.0 * np.ones_like(mu1),
upper=2.0 * np.ones_like(mu1),
- testval=-1.0 * np.ones_like(mu1),
+ initval=-1.0 * np.ones_like(mu1),
)
llk = pm.Potential("llk", two_gaussians(X))
- trace_04 = pm.sample_smc(2000, parallel=True)
- idata_04 = az.from_pymc3(trace_04)
+ idata_04 = pm.sample_smc(2000)
```
We can see from the message that PyMC is running four **SMC chains** in parallel. As explained before this is useful for diagnostics. As with other samplers one useful diagnostics is the `plot_trace`, here we use `kind="rank_vlines"` as rank plots as generally more useful than the classical "trace"
@@ -142,7 +139,7 @@ We can see from the message that PyMC is running four **SMC chains** in parallel
ax = az.plot_trace(idata_04, compact=True, kind="rank_vlines")
ax[0, 0].axvline(-0.5, 0, 0.9, color="k")
ax[0, 0].axvline(0.5, 0, 0.1, color="k")
-f'Estimated w1 = {np.mean(idata_04.posterior["X"] > 0).item():.3f}'
+f'Estimated w1 = {np.mean(idata_04.posterior["X"] < 0).item():.3f}'
```
From the KDE we can see that we recover the modes and even the relative weights seems pretty good. The rank plot on the right looks good too. One SMC chain is represented in blue and the other in orange. The vertical lines indicate deviation from the ideal expected value, which is represented with a black dashed line. If a vertical line is above the reference black dashed line we have more samples than expected, if the vertical line is below the sampler is getting less samples than expected. Deviations like the ones in the figure above are fine and not a reason for concern.
@@ -155,7 +152,7 @@ As previously said SMC internally computes an estimation of the ESS (from import
SMC is not free of problems, sampling can deteriorate as the dimensionality of the problem increases, in particular for multimodal posterior or _weird_ geometries as in hierarchical models. To some extent increasing the number of draws could help. Increasing the value of the argument `p_acc_rate` is also a good idea. This parameter controls how the number of steps is computed at each stage. To access the number of steps per stage you can check `trace.report.nsteps`. Ideally SMC will take a number of steps lower than `n_steps`. But if the actual number of steps per stage is `n_steps`, for a few stages, this may be signaling that we should also increase `n_steps`.
-Let's see the performance of SMC when we run the same model as before, but increasing the dimensionality from 4 to 80.
+Let's see the performance of SMC when we run the same model as before, but increasing the dimensionality from 4 to 80.
```{code-cell} ipython3
n = 80
@@ -168,45 +165,44 @@ sigma = np.power(stdev, 2) * np.eye(n)
isigma = np.linalg.inv(sigma)
dsigma = np.linalg.det(sigma)
-w1 = 0.1
-w2 = 1 - w1
-```
+w1 = 0.1 # one mode with 0.1 of the mass
+w2 = 1 - w1 # the other mode with 0.9 of the mass
+
-```{code-cell} ipython3
def two_gaussians(x):
log_like1 = (
- -0.5 * n * tt.log(2 * np.pi)
- - 0.5 * tt.log(dsigma)
+ -0.5 * n * at.log(2 * np.pi)
+ - 0.5 * at.log(dsigma)
- 0.5 * (x - mu1).T.dot(isigma).dot(x - mu1)
)
log_like2 = (
- -0.5 * n * tt.log(2 * np.pi)
- - 0.5 * tt.log(dsigma)
+ -0.5 * n * at.log(2 * np.pi)
+ - 0.5 * at.log(dsigma)
- 0.5 * (x - mu2).T.dot(isigma).dot(x - mu2)
)
- return pm.math.logsumexp([tt.log(w1) + log_like1, tt.log(w2) + log_like2])
-
+ return pm.math.logsumexp([at.log(w1) + log_like1, at.log(w2) + log_like2])
+```
+```{code-cell} ipython3
with pm.Model() as model:
X = pm.Uniform(
"X",
shape=n,
lower=-2.0 * np.ones_like(mu1),
upper=2.0 * np.ones_like(mu1),
- testval=-1.0 * np.ones_like(mu1),
+ initval=-1.0 * np.ones_like(mu1),
)
llk = pm.Potential("llk", two_gaussians(X))
- trace_80 = pm.sample_smc(2000, parallel=True)
- idata_80 = az.from_pymc3(trace_80)
+ idata_80 = pm.sample_smc(2000)
```
-We see that SMC recognizes this is a harder problem and increases the number of stages. We can see that SMC still sample from both modes but now the model with less weight is being subsampled (we get a relative weight way lower than 0.1). Notice how the rank plot looks worse than when n=4.
+We see that SMC recognizes this is a harder problem and increases the number of stages. We can see that SMC still sample from both modes but now the model with higher weight is being oversampled (we get a relative weight of 0.99 instead of 0.9). Notice how the rank plot looks worse than when n=4.
```{code-cell} ipython3
ax = az.plot_trace(idata_80, compact=True, kind="rank_vlines")
ax[0, 0].axvline(-0.5, 0, 0.9, color="k")
ax[0, 0].axvline(0.5, 0, 0.1, color="k")
-f'Estimated w1 = {np.mean(idata_80.posterior["X"] > 0).item():.3f}'
+f'Estimated w1 = {np.mean(idata_80.posterior["X"] < 0).item():.3f}'
```
You may want to repeat the SMC sampling for n=80, and change one or more of the default parameters too see if you can improve the sampling and how much time the sampler takes to compute the posterior.