From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail.toke.dk (mail.toke.dk [52.28.52.200]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by lists.bufferbloat.net (Postfix) with ESMTPS id 6D1F43BA8E for ; Mon, 6 Mar 2017 07:43:12 -0500 (EST) Received: from mail.toke.dk (localhost.localdomain [127.0.0.1]) by mail.toke.dk (Postfix) with ESMTPS id 215E48A89E for ; Mon, 6 Mar 2017 13:43:10 +0100 (CET) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=toke.dk; s=20161023; t=1488804190; bh=F6TVyFOhV+SHJCCX18zBfUHShXPvkbWPL5YKKiZcGRg=; h=From:To:Subject:Date:From; b=op0rWglShBmzFh6bKqFQbxxgbKg7UUDnVQLC6UkaJMmzZeYzrPoJetZVi7pzSljqu eXD0ZuKW08lrBlQQtBi4VJdUFyaoPYLPQgQscLBS36XbLnT573mvGDAcEW7bS3kylu 1mDNL10UiXNOzcQHoc6QCL3mplUKuH4/mNzipuED+4if6Yq/DPJ5Iz70ZcotTgUOSO kC3UPYCAyU6Lw+SCIGnFY/+FfkAzQV+RY+7p16RsoiW/SsrB57ljscy4sktdbu20gH mDnCV6cTzLaT4+aCBYht6z9Qri1d4XR8w6POcoGIHvMHH7wUkg6/6441d46pSAIyj1 61UGya+AH/HAw== Received: by alrua-kau.kau.toke.dk (Postfix, from userid 1000) id AA236C40C6F; Mon, 6 Mar 2017 13:43:09 +0100 (CET) From: =?utf-8?Q?Toke_H=C3=B8iland-J=C3=B8rgensen?= To: mab-wifi@lists.bufferbloat.net Date: Mon, 06 Mar 2017 13:43:09 +0100 X-Clacks-Overhead: GNU Terry Pratchett Message-ID: <87mvcyshia.fsf@alrua-kau> MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="=-=-=" X-Mailman-Approved-At: Mon, 06 Mar 2017 08:29:43 -0500 Subject: [mab-wifi] Initial algorithm simulation results X-BeenThere: mab-wifi@lists.bufferbloat.net X-Mailman-Version: 2.1.20 Precedence: list List-Id: Multi-armed-bandit WiFi rate control List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 06 Mar 2017 12:43:13 -0000 --=-=-= Content-Type: text/plain So I've finally gotten around to trying out the Python bandit algorithm implementations from https://www.math.univ-toulouse.fr/~agarivie/Telecom/bandits/ I have added a very simple WiFi arm implementation to the code base (which is basically a Bernoulli arm that succeeds with a certain probability and gives a payoff scaled by the base rate), which can be instantiated from the rate_stats_csv output from Minstrel in the kernel. Based on the data from a simple test run in my own testbed I was able to get three of the algorithms to produce something meaningful; see the attached graph. The best of the algorithms performs roughly comparable to Minstrel (I think; the numbers are not quite straight forward to compare). I have not been able to get the Thompson and BayesUCB algorithms to work with this scenario yet (they require a posterior distribution to sample from, and the included implementation doesn't work with the varying payoffs of each arm). However, perhaps sticking with the KL-UCB algorithm is better anyway (same one the "optimal rate sampling" modifies; haven't quite grokked how they modify it yet). Anyway, I do believe it is possible to extend this simulation to something that we can use to guide, say, a dynamic implementation (i.e., change the probabilities over the duration of the test run), as well as evaluating the effects of collapsing arms / defining them differently. It would probably be good with a better source of the actual probabilities for each rate, though. So yeah, bit of a brain dump of where I'm at; but I'll be away for the next couple of weeks, so though it better to get this out there. Code here: https://kau.toke.dk/git/pybandits/ -Toke --=-=-= Content-Type: image/png Content-Disposition: attachment; filename=initial-wifi.png Content-Transfer-Encoding: base64 iVBORw0KGgoAAAANSUhEUgAAA7YAAAKECAYAAAAzP6aKAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz AAAPYQAAD2EBqD+naQAAIABJREFUeJzs3Xl4TGf/x/HPJBKxJSGWEFsssddae1u0RWsrLa3Y1U4V XdRSRSlV2kdXWh5LSwm1VSmKlgelWlvE1kfsS6wJElnv3x/zmP5SSmZMjEner+vKlTNnzvme70xG m0/uc+5jMcYYAQAAAADgpjxc3QAAAAAAAPeDYAsAAAAAcGsEWwAAAACAWyPYAgAAAADcGsEWAAAA AODWCLYAAAAAALdGsAUAAAAAuDWCLQAAAADArRFsAQAAAABujWALAAAAAHBrBFsAAAAAgFsj2AIA AAAA3BrBFgAAAADg1gi2AAAAAAC3RrAFAAAAALg1gi0AAAAAwK0RbAEAAAAAbo1gCwAAAABwawRb AAAAAIBbI9gCAAAAANwawRYAAAAA4NYItgAAAAAAt0awBQAAAAC4NYItAAAAAMCtEWwBAAAAAG6N YAsAAAAAcGsEWwAAAACAWyPYAgAAAADcGsEWAAAAAODWCLYAAAAAALdGsAUAAAAAuDWCLQAAAADA rRFsAQAAAABujWALAAAAAHBrBFsAAAAAgFsj2AIAAAAA3BrBFgAAAADg1gi2AAAAAAC3RrAFAAAA ALg1gi0AAAAAwK0RbAEAAAAAbo1gCwAAAABwawRbAAAAAIBbI9gCAAAAANwawRYAAAAA4NYItgAA AAAAt0awBQAAAAC4NYItAAAAAMCtEWwBAAAAAG6NYAsAAAAAcGsEWwAAAACAWyPYAgAAAADcGsEW AAAAAODWCLYAAAAAALdGsAUAAAAAuDWCLQAAD0hycrLGjx+vwoULq1ixYvr5559d3VKGd/DgQY0e PVrNmjVzdSsAgHREsAWADOTixYs6evSoq9vAP+jRo4f8/f21Y8cO9ejRQ5UrV3Z1SxleYGCgzp07 pz179ri6FQBAOiLYAkAG0qxZM3300UeubiNTSU5O1oABA3T69Om7bnfgwAEtW7ZMffv2VaFChfT2 228rd+7cD6hLx+3Zs0cFChTQkSNHXN2KQ/z9/VW7du3b1j/++OMaN25cmmqcO3dOQUFB2rhxo7Pb AwA4CcEWADKIY8eOadeuXVq8eLFSUlJc3U6mcfLkSX322WeKi4u763YbN25UyZIl5eHhXv/rvXbt mvLly6eAgABXt+JUN2/eVMmSJdO07Y0bN5QjRw4FBQWlc1cAAEe51/9dAQD/6LvvvtOgQYN0+fJl /fLLL65uB39z6dIl+fj4uLoNu9WvX1/h4eHKkydPmvc5efJkqhFeY4z++OOPe45qP0g7duxQ+/bt 07RtyZIldfjwYYWEhKRzVwAARxFsASCDWLJkibp166bGjRtr4cKFqZ579dVXZbFY1LNnT129elWS tGzZMvn6+uq9996zbbd69WpVr15duXPnVpMmTXTq1ClJ0pYtW9S8eXN17dpV27ZtU8WKFRUQEGC7 bvHAgQN68cUXVbBgQRUtWjTVKZvR0dHq16+fgoODVbBgQQ0dOlSvv/666tSpoxs3bkiyhozHH39c uXPnVt26dbVv3747vsa5c+eqevXq+uqrr/Tll1+qYMGCKlWqlIwx96xz5swZdejQQYULF1ZwcLBG jx6tV155RU899ZSOHTumXr16qXz58jp//ryaNGminDlzasaMGZKkxMREDR8+XEWLFlX+/Pk1atQo SVJkZKQee+wxSdKjjz4qLy8vVa1a9ba+33//fU2YMEHbt2+Xj4+PSpQooeTkZEnSihUrVKVKFfn7 +6tWrVravXu3JCk2NlZjxoxRcHCw9u/frz59+ih37tzq2rVrqtpRUVF6+umn5eHhoU8//dS2ftSo UcqTJ49WrlwpSYqPj9eYMWNUsWJF5cyZU0OGDJFkDZ0ffvihypYtq3Xr1mn06NHKly+fGjVqpGPH jumNN95Q0aJFlZSUlOpn+vLLL6tQoULKly+fRowYYXt+zpw5Klu2rObMmSNJSklJ0bBhw1S9enWt Xr3aVmPkyJG2/fv27auEhITb3reFCxeqcePGGjt2rP773/+qZcuWypUrl7p06aL4+Pg09XMnH3/8 sRo0aKCZM2emWr9p0ybVq1dPfn5+KlSokL744gtdunRJo0aNUtmyZfXnn3/atv2nz8Stfp5//nkF BASoRIkSqX4uAIB0YgAAbu/MmTOmfPnyxhhj5s6da/LmzWsSExNTbVO2bFkzceJE2+Nz586ZRx99 1PZ46dKlpnbt2ubMmTMmLi7OdOjQwTRo0MD2/ODBg02dOnXMM888Y65fv25atGhhzp49a3799VdT okQJc+rUKZOYmGiaNWtmmjZtatuvcePGZs6cOcYYY8LDw03p0qVNp06dzMmTJ40xxmzfvt1UqFDB HDp0yCQkJJihQ4eaUqVKmaSkpDu+1qpVq5pnn33WDBw40MTExNh6vFudhIQEU6lSJbNx40ZjjDFb t241efPmNUOHDjXnz583xhizfPly4+vra1q2bGkOHz5sJk2aZBYuXGiMMaZNmzZmzJgxJjEx0Rw+ fNgEBgaa2bNnG2OMiYyMNJLMkSNHTFxcnLly5cod+37nnXdMvXr1TFJSkrl06ZIxxpglS5aYxx9/ 3Jw/f97Exsaa0NBQkz17dnPq1CljjDHR0dFGkmnWrJlZvHix2bFjh+nbt+9ttU+dOmU8PT3Nr7/+ alu3aNEiM3jwYGOMMQkJCaZu3brmq6++MsYY8+233xpJ5ty5c8YYY1JSUkyePHlMs2bNzOTJk82p U6dMq1atTHJyspk2bZqRZPs8JSYmmlq1apn58+ebpKQk88svvxh/f3/z6quv2o5dv359M2LEiFQ9 FipUyHb8H3/80dSoUcPExcWZiIgIU6dOHZOQkHDb60pJSTFNmzY1jRs3Ni+99JK5cOGC2bNnj8md O7cZN25cmvuZNWuWCQoKsj0+ffq08fPzs/VjjDHz5883TzzxhDl//rxJTEw0kydPNuPGjTMpKSlm 1apVtp/xLXf7TLz11ltmwIABJiUlxaxYscK0b9/+Th8JAIATEWwBIAP47LPPbEEiOjraZM2a1axZ sybVNhMnTjTlypWzPZ46dar58MMPjTHWAFG8eHGzZcsW2/M7duwwkszFixeNMdZgJskWnq5evWqM MWbNmjXm999/t+13K8AZY8z58+eNJBMdHW17/l//+pd55plnbI8ff/xxM2/ePNvjqKgoIylVzf/v iSeeMAEBAebGjRup+rhbne3bt5vcuXOnqjNo0KBUIXHjxo1Gki3837hxwyQkJJgNGzaYkJCQVPv2 69fPtGjRwhiTOtjezf9/X4yxvufFihVL9Z7HxcWZwoULm2HDhtnWSUr1h4Jbr/fvmjZtavr06WN7 3Lp1a/PHH38YY6zBd968ebY/Ftyp52LFipny5cub5OTkVMe59b7cCrYzZ840Tz/9dKpjf/LJJ8bL y8tcu3bNGGP9Gf092BYrVswWJJcuXWoKFixoDh48+M9v2P906dLFhISEmOvXr9vWjR071lSsWDHN /fw92P69nytXrhh/f38TERFxxx7+/n7d6zPx6quvmoYNG/7jzwoA4HycigwAGcCSJUv0xRdfKDAw UCEhITLG3HY6cufOnXX48GHt2LFDkhQWFma7xvDQoUM6duyYWrdurcDAQAUGBuqZZ57Ro48+qixZ sthqlCpVSrVq1ZIk+fn5SZIaN26sihUravXq1Ro2bJjt1FdJCggIUMGCBfXWW2/p6tWrio2N1Zo1 a9SmTRtJ1tNtN2/erIEDB9qOW6FCBZUpU+aukxW1bt1a2bNnt/VxrzolSpRQfHy8xo0bp9jYWF2+ fFn/+c9/1KpVq9tqd+zYUZKUPXt2eXl5ac2aNTp+/LitbmBgoObPn68aNWrY90P6m4iICB0/flwV K1a0rfPx8VHTpk21d+/eO/Z06/XeSbdu3bRw4ULFx8crOjpax48ft50WHRQUpNDQUB08eFBTpkzR iBEj7lgjNDTUNrnVPx1n9erVqXqWrD+PxMREHTx48B6v2qpVq1YKDQ1V1apV1a9fP124cOGu2z/6 6KPKkSOH7XHDhg0VGRnptH42b96sxMRElStXLk3b3+szMWrUKGXNmlUlS5bU5MmTlZiYmKa6AADH Zbn3JgCAh9nly5d15MgRXbhwwRZKpk+frmHDhmnatGny8vKSJBUsWFBNmzbV7NmzlTdvXvn6+iow MFCSbNe67tu3T/nz5//HYxUoUOC2dYcPH1bbtm3Vv39/DR48WFmzZtVPP/0kSfL09NTKlSvVo0cP BQUFqXLlyurXr58tqMXFxckYo1WrVqlmzZppfs1/7yMtdZYtW6Z+/fpp8uTJqlatmsaOHasmTZrc s/aNGzfUqFEjrVq1Ks39pcWta0Sjo6Pl6+trW+/v768rV67ctac7adWqlfr27avly5fr+vXrateu ne05Y4y6dOkiT09PDRkyRK1bt9b8+fNvq5GW49wKzv+fv7+/JKV5ciyLxaLJkyerV69eGjp0qKpV q6bdu3eneeblgIAA23vmjH6uX7+ulJQUpaSkpGnW6nt9JvLkyaPVq1dr/fr1Gjx4sDZs2OD0zw8A IDVGbAHAzS1fvlytW7dO9Qv5Cy+8oOvXr2vt2rWptu3WrZsWLFigWbNmqUOHDrb1ISEh8vT01Pr1 6+0+/muvvaY2bdqoV69edwzFAQEBio2NVUREhLZu3Zpq9DEgIED58uVz6Lh/P8a96uTJk0c+Pj6K jIzUhg0b9Mwzz6Spdrly5bR9+3Zb+P87i8UiSXbfYikkJEReXl5aunRpqvX79+9Xs2bN7KolSVmz ZlVoaKhmz56tBQsWKDQ01PbckiVLtGXLFv373/9WpUqV7uuWQxUqVNCaNWtS3d5o//79Kl68uMqX Ly/J+geN//9+RUdH6/r167bH27dvV3x8vEJCQrR06VKVKFHCrpm8Dx48qBYtWqS5n3upVKmS4uLi tGXLljRtf6/PxK3X8uSTT2rLli3atGmTLl++nKbaAADHEGwBwI3Fx8dr9uzZt408BgQEqG7dupo9 e3aq9S1atJCHh4c+//xzPffcc7b1uXLlUs+ePfXGG2/o999/l2QNB4MHD77r7LKSdOLECZ05c0Zn z57Vl19+qYULF+ratWu2GZX/+OMPHT58WAMGDNC4ceO0ZMkSXbp0ybb/a6+9pokTJ2rdunUyxujY sWMaMmSIbfbmtLpXnU2bNunQoUPq06ePJk6cqJUrV+ratWv3rNuxY0dlyZJFL7/8sq5evaqkpCTN mzfPNuvvrVN2N23apCNHjuiTTz5JU785c+ZU37599fbbb2vdunVKSUnRokWLFB0drU6dOtn12m/p 1q2b1q5dK2OMihQpYlt/4sQJxcbG6tSpU/rll1/02muvSbLOFP330c576d+/v65cuaKuXbsqJiZG Fy9e1FtvvaWpU6faAnPx4sW1bt06Xb58Wbt371bv3r3l4eGhkydP6tixY1q7dq1GjhypmzdvKjo6 WhcuXLhrCP3pp5+0fPlypaSk6OzZs5o6dartdOq09HMvFStWVKtWrdSjRw/bTNobN27UlClT7rj9 vT4T7733nsLCwmSM0fHjx5U3b17lzp07ze8xAMABrru8FwBwv3x8fIwk4+npaZvB1xhj2rVrZyQZ SaZNmzap9nn99dfN+PHjb6sVFxdnBgwYYHLlymUCAwNNz549bZM+ff/996Z8+fLG19fXjBo1yjYp jzHGfPPNNyZ37twmODjYTJ8+3SxatMgUKVLETJs2zRhjzO+//25Kly5tmjRpYsqWLWuyZMlismXL Zj7//HNjjDHJyclmzJgxJk+ePCYgIMC0bdvWnDlz5o6vd/Lkycbf39+UKVPGTJ8+PdVz96qzbt06 U6xYMdO4cWNTqlQp4+HhYXLnzm2WLl1qwsPDTdOmTY0k06dPn9smNQoPDzd16tQxPj4+pnz58rZZ nm/p3LmzyZkzp2nTpo0JDw+/re81a9aYRx55xPj6+ppBgwaZPXv2GGOsM/q++eabJk+ePCZ//vym e/futgmHoqOjTf/+/Y0k8/TTT5u1a9fe8T35uwYNGty27enTp03lypVNrly5TPfu3c2pU6dMkSJF TLdu3UxUVJQZOnSo8fb2NjVr1jRhYWG2/Y4cOWKaN29uJJkhQ4aYmJgYY4wxO3fuNDVq1DA5c+Y0 lStXNsuXL091vEOHDpkyZcqYHDlymM6dO5uoqCjTtm1b07hxY7N//36zbds2U7duXZMzZ05TtmxZ M2vWrH98PV26dDHPPPOM6d+/v22Cq82bN6fa5m79bNy40Tz22GPG29vbvPPOO+bGjRtm9OjRJmvW rKZevXq2ydCio6NN165dTbZs2UxwcLB55513TFJSkomKijKdO3c2kkyXLl3M6dOnjTF3/0z8+9// NiEhIcbPz8/Url071WzVAID0YTHmfzf/AwDAyX799Vd16tRJK1assE3MExMTo9dff10LFy60e7TQ UUuXLtXo0aO1evVqFSpUSJL1/q8vv/yyTpw4YbsfLx4+Xbt2VVJSkr755htXtwIAeIgxeRQAIN18 +umnql+/fqrZZn19fdWwYUPt2rXrgfUxZcoUtWnTxhZqJSl//vyqW7dumicYAgAADy+usQUApJt6 9eppwYIFmj59um2inV27dmnSpEmaOnXqA+1j6tSpWrRokRISEmSM0c8//6x58+bpvffee2B9AACA 9MGpyACAdPXVV1/p888/14EDB5Q3b149/vjjGj16tEJCQh5YD0lJSZoyZYpmz56tyMhIBQYGqkmT JnrnnXdSjeLi4TJnzhyNHDlSSUlJGjBgwD/efxcAAIItAAAAAMCtcSoyAAAAAMCtEWwBAAAAAG6N WZEzqJSUFJ05c0a5cuWSxWJxdTsAAABAhmaM0bVr11SoUCF5eDB++KARbDOoM2fOqEiRIq5uAwAA AMhUTp48qcKFC7u6jUyHYJtB5cqVS5L1H5avr6+LuwEAAAAytpiYGBUpUsT2ezgeLIJtBnXr9GNf X1+CLQAAAPCAcBmga3DyNwAAAADArRFsAQAAAABujWALAAAAAHBrXGObiaWkpCghIcHVbcABXl5e 8vT0dHUbAAAAwEOBYJtJJSQkKDIyUikpKa5uBQ7y9/dXYGAgExQAAAAg0yPYZkLGGJ09e1aenp4q UqQIN5B2M8YYxcbGKioqSpJUsGBBF3cEAAAAuBbBNhNKSkpSbGysChUqpOzZs7u6HTggW7ZskqSo qCjlz5+f05IBAACQqTFUlwklJydLkry9vV3cCe7HrT9KJCYmurgTAAAAwLUItpkY12a6N35+AAAA gBXBFgAAAADg1gi2AAAAAAC3RrAFAAAAALg1gi3cxubNm1W2bFk1btxYhw4dkiTFxsZqyJAhCgwM 1MqVKyVJP//8s5o1a6aKFSsqODhY7777rowxGjFihMqWLatq1aopJCREDRo00Pbt2135kgAAAAA4 AcEWbuOxxx5T7dq11bJlS5UpU0aSdWbggQMHqmzZsmrevLlWrVqlKVOmaOHChQoPD9eKFStUpUoV WSwWjR8/XocOHdKqVat0+PBhNWnSRCNGjHDxqwIAAABwv7iPLWSMFBvrmmNnzy45a3Lf5ORkDRw4 UJs3b1bOnDklSZUqVVKlSpVu2zYuLk5Xr15Vp06dnHNwAAAAAC5DsIViY6X/5cAH7vp1KUcO59Ta uXOn8uTJo4IFC951uyeeeEIXL15U1qxZ9emnnzrn4AAAAABchlORkWEcPXr0nqFWkn7//XdFRUXp gw8+UPv27XX48OEH0B0AAACA9MKILZQ9u3Xk1FXHtoeHh4cSEhJSrUtMTJSvr6+yZcumCxcupKmO p6enOnTooMWLF2vZsmV688037WsEAAAAwEODYAtZLM47HTi9FS1aVAcOHEi1Ljw8XKVKlVKdOnW0 d+9eHTp0yDa51L0EBwfL4qyLfAEAAPDAJSdLX30l9eghZSHdZFqcigy30rp1a4WFhenXX3+VJF27 dk0TJ05Ujx49VKBAAfXv318dOnTQ6dOnJUnHjh3T8OHD71jr+PHjWrFihZo3b/7A+gcAAIDzGCMN GCD17Su1b299jMyJv2nArVSuXFlz585V//79lZiYqBw5cujtt99W+fLlJUnjx4+XJNWoUUPFixdX jRo1bOvat28vSapevbpy5Mghb29vffLJJypXrpxrXgwAAAAclpgojR4tTZtmPQOxbVvn3W0D7sdi DH/XyIhiYmLk5+en6Oho+fr6pnru5s2bioyMVHBwsHx8fFzUIe4XP0cAAJBZnTolNW0q7d9vffzO O9aQ60p3+/0b6Y9TkQEAAAC4hZgYadIkqVixv0Lt009LQ4e6ti+4HqciAwAAAHjoGSM995y0caP1 sbe3tHu3xFVlkBixBQAAAOAGfvzxr1Dbo4e0dCmhFn9hxBYAAADAQ61LF2nuXOvykCHSlCmu7QcP H0ZsAQAAADy0wsP/CrUhIdLYsa7tBw8ngi0AAACAh9atUFumjLRzp5Qjh2v7wcOJYAsAAADgoWSM tHixdfm996RcuVzbDx5eBFsAAAAAD6WICCkyUsqaVWrSxNXd4GFGsIXbSE5O1ubNm7V3797bnouO jlZUVJQLugIAAEB6mTHD+r1hQ05Bxt0RbOE2duzYoW7dumnTpk2p1v/nP/9R+fLlFRYW5qLOAAAA 4Gy7dkkff2xdHjjQtb3g4UewhduoU6eO6tWrd9v6+vXrq1GjRi7oCAAAAM527pzUp49UrZqUkiK1 ayc984yru8LDjmALt2KxWO643tPT07Y8e/Zs5ciRQ6+88opiYmIkSUePHtWzzz6runXrat++fbZt p02bpqefflrBwcGqVKmSVqxYoRMnTqhXr14qVaqUqlatqpIlS+qll17S+fPn0/fFAQAAZHJxcdYQ O3269XHBgtK//uXanuAeCLbIcLp27ap8+fKpd+/e8vX1lSSVKFFC7dq1U+PGjVWpUiVJ0vDhw3X2 7FmtWbNGkZGRGj9+vEqUKKGiRYsqNDRUWbJk0a5du3TgwAHduHFDX331lStfFgAAQIb3ySfS7t1S QIC0ZIl04IA13AL3ksXVDcD1jDGKTYx1ybGze2X/x1HYe3nuuee0fPlyNWvWTHnz5rVr3yNHjuiH H37Q7t27bcdv2bLlHbe9cuWKsmXLpubNmzvUJwAAAO4tNlb66CPr8uTJUuvWru0H7oVgC8Umxirn hJwuOfb1YdeVw9v+Ke5SUlIUFBSkL774Qn369FHXrl3t2n/JkiVq0qTJXUP10aNHVbFiRf3555+q UKGCvLy87O4TAAAAadOvn/X62iJFpNBQV3cDd8OpyHA7N2/eVP/+/RUaGqo+ffo4VOPo0aMqeI/z Wh555BGFh4frypUrqlevnlq2bKmkpCSHjgcAAIB/tn69NGeO5OEhzZ0reXu7uiO4G0Zsoexe2XV9 2HWXHdteixcvVu7cuVWnTp1/3MbDw0MJCQmp1iUmJtquuc2WLZsuXLiQpuNly5ZNU6ZM0cKFC/XH H3+oZs2advcMAACAO0tMlF591bo8YIDUoIFL24GbYsQWslgsyuGdwyVfjlxf27FjR/n6+mro0KH/ uE3RokV14MCBVOvCw8NVqlQpSVKDBg0UFham+Pj4NB3Ty8tLQUFBDl8PDAAAgNROnZJ++UV65RVp /37rhFGjRztW63LcZaf2BvdDsIVbmjlzplatWqXZs2ff8fm2bdtqwoQJOnv2rCQpIiJCP//8s5o2 bSpJatWqlW3m5FvhdtOmTfriiy/uWG/dunWKjY1V5cqVnf9iAAAAMplJk6Rixayjs7du7TNpkpQ7 t/21fjj8g4KnBmvZwWVO7RHuhVOR4TY2bNigTZs26cyZM+rQoYOWLFmixo0b6/jx4/rtt9906tQp PfXUUypbtqz69u2r6OhoNWrUSNmzZ1eePHkUFhYm7/9dsGGxWLRs2TL1799fRYsWVYUKFdSsWTMN GTJEx44d0/Dhw7V7925VqVJFFotFBQoU0PLly237AwAAwDHXrkmjRkkpKVKBAlJ0tPVU5G7d7K+1 7eQ2tVzQUikmRbN2z1KrMq04wy6TshhjjKubgPPFxMTIz89P0dHRtutKb7l586YiIyMVHBwsHx8f F3WI+8XPEQAAuKMFC6T27aWSJaUjRyRHc+jm45vVJqyNLsZeVOuyrbXwhYXy8nTdXSzu9vs30h+n IgMAAAB4IM6fl8aMsS6/+KLjoTY2MVbtv2uvi7EXVdi3sKY3n+7SUAvX41RkAAAAAA/EsGHSwYNS UJDUt6/jdT7Z/olOXzutoFxB2t17twKyBzivSbglRmwBAAAApLsLF6T5863LYWFS4cKO1bl686om bpkoSXrvyfcItZBEsAUAAACQzhISrJNDxcdLNWpIdeo4Xmv6zum6evOqyucrrw6VOjivSbg1gi0A AACAdJGUZD3lOEcO6YcfJB8f6YsvHL+2Nj4pXlO3T5UkvVn3TXl6eDqxW7gzgi0AAACAdDFsmDRt mjXg5solLV1qHbF11Lfh3+rs9bMqlKuQ2ldq77xG4fYItgAAAACcLjpa+uwz6/K4cdLJk1LTpo7X M8Zo8tbJkqRXa70qb09vJ3SJjIJZkQEAAAA41fXr1tv5xMVJFSpIw4c7fvrxLSsOrdD+C/uV0zun elXv5ZxGkWEwYgsAAADAqfr0kdassS4PHnz/ofZa/DX1X9VfktT/0f7y9/G/zw6R0RBsAQAAADjN r79K8+ZZl2fMkF5++f5rTt46WaevnVbJ3CU16olR918QGQ7BFgAAAIBTpKRIgwZZl7t1c06oDdsf pvGbx0uSJjw5Qdm9st9/UWQ4BFsAAAAA980Y62nH27dLOXNK48fff80dp3eo+/LuSjbJ6vRIJz1f /vn7L4oMiWALt7F582aVLVtWjRs31pEjR3Tu3DlVq1ZNbdq0ue25B23SpEmyWCx69913lZycLEna vXu3atdm20/JAAAgAElEQVSurebNm+vYsWOSpKSkJI0fP16NGjVSkSJFVLNmTW3dulV79+5Vx44d FRISosqVK6tUqVLq3bu3rl+//sBfCwAAgCNmz5Y+/th6Pe3nn0sFCzpeKyE5Qd2Xd1ftGbV1I/GG GhZvqNnPzZaHhfiCO+OTAbfx2GOPqXbt2mrZsqVKly6tffv2qXDhwlq0aNFtzz1ob775piRp8ODB 8vS03ii8SpUqatq0qV544QUVL15cktSlSxcFBgZqw4YNOnHihAYMGKBixYrpkUce0VNPPaUSJUpo z5492r17t3777TctW7bsgb8WAAAAe50/Lw0caF0eN07q1MmxOsYYbYjcoIZzGmrW7lkyMnqxwota 1HYRoRZ3xacD1vNGbtxwzZcxDrV85MgRTZgwQfPnz7cFybtZsWKFnnrqKZUpU0Z9+vRRSkqKduzY oUaNGum7777TtGnTVKNGDdWrV0+xsbEaPny4ypYtq549e0qSTp8+rY4dO2rIkCFas2aNGjZsqLJl y2rXrl1p7nn9+vW6evWqXv7fxSYWi0WdO3dWUFDQbdtevHhRhQoV0hNPPJHm+gAAAA+SMdKZM9KE CVJgoPUWP48+Kg0d6njNif+ZqCfnPqmtJ7fKy8NL37f/XgteWKCA7AHOaxwZEsEWUmys9UIIV3zF xtrdbnR0tPr3769vvvlGOXPmvOf227Zt0+bNm7Vu3Trt379fu3fv1pIlS1SzZk3lz59fM2fOVK5c ubRz506dPHlSnTp10vPPP69ffvlFM2fOVEJCgoKCglS/fn2tWrVKBw4c0MaNG9WvXz/17ds3zX0v XrxYzZo1u+s2mzZtUsWKFVWqVCnFxsamKbQDAAC4wssvS0FB1nvU3vL++5Kjv77sO79Pb2982/Z4 wQsL1Dyk+X12icyCYAu3s3jxYlksFgUGBqZp+8mTJ+vHH39UnTp1VL9+fcXExChPnjySJB8fHxUs WFAdOnSQJOXPn19PPPGEqlevrgIFCsgYo4SEhFTbDvrfVH99+vTR4cOHFRcXl6Y+jh49qoL3uNjk 2WefVXh4uC5duqQ8efLopZdeSlNtAACAByU+3jpKO2vWX+v8/aUVK6SGDe2vF5sYq9fXvq5Hpj2i ZJOs1mVbK2VUitqUa+O8ppHhZXF1A3gIZM9uPXfEVce2U/fu3bVu3ToNHjxYU6dOvef2ERER+vLL L/XYY4/d8fmqVaumevzII4/8Y62AgL9Og/H29lapUqV07do1ZcuWTRaLxRaCb0lMTJSvr68kKVu2 bLpw4cI9+5UkPz8/ffnll8qfP7+ioqKUP3/+NO0HAACQno4elRo3lv77X+vjypWlp56SRo60hlt7 GWPUbXk3he0PkyTl8MqhKY2nyGKxOLFrZAaM2MI6dV2OHK75cuA/WhaLRV9//bVWr16tzz777J7b FyhQQDt37nTknbmnHDlyKF++fJKkokWL6sCBA6meDw8PV6lSpSRJDRo00DfffJPm2nny5JGvry// YQcAAA+Nt96yhtrAQGnGDOmPP6TJkx0LtZL0+W+fK2x/mLJ4ZNH05tMV0T9CwbmDnds0MgWCLdyS n5+fli5dqhEjRmj16tW29UlJSbblP//8UwsWLFDbtm310Ucf6fz585KkJUuWaM+ePQ4dNyIiQidP npQk/fLLL2rSpIkteLZt21YjR45UTEyMJOv1spcvX7aNAPfu3VsnTpzQiBEjlJKSYutl0aJFdzzW rFmzVLFiRVtwBgAAcKUDB6TFi63L69ZZr7H1uI808cfZP/TK6lckSWMbjFWv6r1U1K+oEzpFZsSp yHAb33//vTZu3Kj9+/erfv36qlKlimbMmKHOnTtryJAh2rRpk8LCwjRnzhx5enrqzz//1I8//qi2 bdvq6NGjql27tkqVKqV+/fqpcuXK2rJli7Zt26aTJ0+qWbNm2rNnj44ePapZs2apSpUqmvW/C0em TJmiUaNGSZL8/f31+uuv6+TJk6pSpYo++eQTW39jx47VyJEj9eijj8rX11eFCxdWWFiY7fls2bJp 3bp16tevnwoXLqyKFSuqY8eO6ty5s37//XdNmTJFR44cUfXq1ZWSkqLSpUtr4cKFD/ZNBgAAuANj pEGDrN+fe06qWPH+a76/5X0ZGT1f7nm9Vf+t+y+ITM1ijIP3W8FDLSYmRn5+foqOjrZd43nLzZs3 FRkZqeDgYPn4+LioQ/cze/ZsrVy5Uotv/anSxfg5AgCAB2XxYqltW8nbWwoPl0qXvr96Sw4s0fNh z0uS9vTZo0cK/PMcJ+7ibr9/I/1xKjIAAACAf3TxojR4sHV56ND7D7XRN6PV6/tekqRXa72aIUIt XI9gC6TB6dOntXTpUu3Zs0fff/+9q9sBAAB4IK5elerVk06dkoKDpWHD7r/m+M3jdSnuksrlLafJ jSfff0FAXGMLpElQUJCWL1/u6jYAAAAeqAkTpMOHpSJFpB9+kLJlu796v576VVO2TZEkvf/U+8ri QRyBczBiCwAAAOA24eHS1KnW5S++kMqVu/+aH2z9QCkmRaGVQtWiTIv7Lwj8D8E2E2PeMPfGzw8A AKSXuDipVSspPl5q0kR69tn7r3kl7opWHl4pSRpab+j9FwT+H4JtJuTp6SlJSkhIcHEnuB+xsbGS JC8vLxd3AgAAMpolS6SjR6VChaR58ySLxfFaxhgtO7hMFT6voITkBD1S4BEmjILTcVJ7JpQlSxZl z55dFy5ckJeXlzzu587aeOCMMYqNjVVUVJT8/f1tf6gAAABwln//2/q9d28pIMCxGtfir2n4+uH6 Zt83unrzqm39uw3fdUKHQGoE20zIYrGoYMGCioyM1PHjx13dDhzk7++vwMBAV7cBAAAymG3bpA0b rKO0Xbo4VuPgxYNq+k1THY/+63fNCvkqKLRSqFqWaemkToG/EGwzKW9vb5UuXZrTkd2Ul5cXI7UA AMDpEhOll1+2LnfuLBUr5lidwWsG63j0cRX3L64Pnv5ADYs3VEB2B4d+gTQg2DooPDxcb775po4f P66goCAtXLhQuXPn1gcffKCvv/5aiYmJ+vbbb1WlShVJ0rFjx9SzZ08dPHhQL7zwgj766CNbLUf2 cQYPDw/5+Pg4tSYAAADc19Kl0oEDUt680ocfOlZj55md+vHPH+Vh8dBPnX5SyTwlndskcAdcXOmg Pn36aPbs2QoPD1fWrFn15Zdfav369Tp27Jj27t2rgQMHqnv37rbte/furRkzZmj//v1auXKlVqxY IUkO7QMAAACkh48/tn7v10/Kk8f+/ROSEzRg1QBJUodKHQi1eGAItg46cuSIvL29ZbFYFBQUpHz5 8umzzz5Tl/9diNCzZ09FRkbq4MGD2rt3rzw9PVWsWDH5+vqqd+/emj9/viQ5tM+dxMfHKyYmJtUX AAAAkFY//yxt2SJlyWKdNMpeV+KuqNaMWtp+ertyeOXQe0++5/QegX9CsHVQt27dVKNGDb333nuy WCzq0qWLfvvtNxUtWlSSdebhatWqKSIiItV6SapVq5YiIiIkyaF97mTChAny8/OzfRUpUiQ9XjYA AAAyoJQU6Y03rMu9e1tv82PX/iZFHZd21O5zu5XbJ7fmtZmnwr6Fnd8o8A8Itg7q0aOHKlWqpJ9+ +kmbNm3S8ePHFRUVJW9vb9s2+fPn16VLl/5xvSSH9rmTYcOGKTo62vZ18uRJZ75cAAAAZGCLFkk7 d0o5c0qjRtm//7u/vKtVR1bJJ4uP1nder1ZlWzm/SeAumDzKAdevX9eLL76o9evXy9/fXwMGDNCg QYPk7++vixcvKs//uyChQIECSkpK0r59+1LVKFCggCQ5tM+dZM2aVVmzZnXGywMAAEAm8vXXUp8+ 1uU33pDy57dv/9VHVmvML2MkSdObT1fVglWd3CFwb4zYOmDdunUKDAyUv7+/JKljx446efKkypcv r507d9q2i4yMVNWqVW9b/9///lfVq1eXJIf2AQAAAJzh99+tt/WJjZVKlpSGDLFv/ytxV9R1eVcZ GfWt0VedK3dOn0aBeyDYOqBw4cLasWOHTp8+LUnatGmTnnvuOXXu3FmffvqpkpKSFBERoSJFiqhI kSJ67LHHlJSUpB9++EGS9O2336pv376S5NA+AAAAgDN88IH1++OPW0Nuzpz27T9q4yhF3YhSubzl 9FET596aErCHxRhjXN2EO3r//fc1ffp0FSpUSLVr19aECRPk5eWl/v37a9WqVapcubLmzJkjPz8/ SdKuXbvUrl07+fn5aciQIQoNDbXVcmSfe4mJiZGfn5+io6Pl6+vr3BcPAAAAtxcRIVWqZJ04avdu qXJl+/a/Fn9NBacU1I3EG1rbca2eLvl0+jTqJvj927UIthkU/7AAAADwT4yRnn5aWr9eat1aWrLE /hoz/5ipHt/3UEhAiA72PyiLxeL8Rt0Iv3+7FqciAwAAAJnML79YQ23WrNKUKfbvfznuskb9bJ0+ uUfVHpk+1ML1CLYAAABAJjN5svV79+5ScLB9+xpj1Htlb525dkZlAsqof83+zm8QsBPBFgAAAMhE IiKkH36QLBZp8GD795+7Z64WRyxWFo8smtdmnrJ7ZXd+k4CdCLYAAABAJvLhh9bvzz0nlS5t377H rh7TK6tfkSSNbTBW1QtxO0o8HAi2AAAAQCYRGSl9/bV1+fXX7d9/9M+jdS3hmuoVqac3673p3OaA +0CwBQAAADKBpCSpQwcpIUFq0ECqW9e+/SOvROqbvd9Ikj5s8qE8PTyd3yTgIIItAAAAkAl8+620 bZvk6yvNmmX//jN3zVSySdZTJZ5SzaCazm8QuA8EWwAAACCDM0b64APr8ltvScWL27d/iknR13ut 5zD3rNbTuc0BTkCwBQAAADK4jz6S9u2TcuaU+vSxb19jjMZvGq8T0Sfkm9VXLUJapE+TwH0g2AIA AAAZ2E8/Sa+9Zl0eO1bKndu+/T/Y+oFG/TxKkjS8/nBl88rm5A6B+5fF1Q0AAAAASB+JidKrr1qX e/SQBg2yb//tp7Zr6E9DJUkfPP2BXq/rwFTKwAPAiC0AAACQAS1ZIoWESBERUkCANGmSZLHYV2PE hhGSpA6VOhBq8VBjxBYAAADIYKZO/Wt0Nk8eaeZM+09B/uK3L7Q+cr28PLw0rtE45zcJOBEjtgAA AEAGcvSoNGyYdXnIEOnUKalVK/tqTNoySf1X9ZckjW04VsX9izu3ScDJGLEFAAAAMoCNG6UffpAW LZLi4qRGjaTJk+0//Xj5weW262pfqfmKhtYbmg7dAs5FsAUAAADc3OnTUosW0o0b1selSkkzZtgf amPiY9TnB+v9gIbUHqIpTaY4uVMgfXAqMgAAAODGYmOl/v2todZikd57T/r9dyk42L46xhhN2jJJ 566fU+k8pTX+yfHp0zCQDhixBQAAANzU9etSrVrWmY89PKTt26UaNeyrYYzRgFUDtChikS7EXpAk TXxqonyy+KRDx0D6INgCAAAAbur9962hNn9+68zH9oZaSdp6cqs+3/m5JMnLw0tdq3RV67Ktndwp kL4ItgAAAIAbioy0Tg4lSdOmSc2bO1Zn7p65kqQXK7yo2c/NZqQWbolrbAEAAAA3Y4zUp49086bU sKH03HOO1YlNjNXC/QslSb2r9ybUwm0RbAEAAAA3M368tHatlDWrNH26/bMf3zJv7zxFx0cr2D9Y TxR/wrlNAg8QwRYAAABwI9OmSW+/bV2eNEkqXdqxOsYYfbzjY0nSgJoD5GEhGsB98ekFAAAA3MSR I9KQIdblMWOkgQMdrzV3z1yFR4Urh1cOda/a3TkNAi5CsAUAAADcwNmzUosWUlyc9OST0siRjte6 FHtJr697XZI06olR8vfxd1KXgGsQbAEAAICHXGys1KSJdOiQVLSoNHu29b61jnrrp7d0MfaiKuav qMG1BzutT8BVCLYAAADAQ27AAGnfPikwUPr5Z6lwYcdr7TyzUzN2zZAkTWs2TV6eXs5pEnAhgi0A AADwENuyRZo1yzpCO3++FBzseC1jjIb+NFSS1PGRjqpXtJ6TugRci2ALAAAAPKRiYqRhw6zLL79s vWft/ZiybYo2RG6Qt6e33m347v03CDwksri6AQAAAAC327RJevZZ6cYN6/1qb93ix1Hz9s7TG+ve kCRNeHKCivsXv/8mgYcEI7YAAADAQyY5WerXzxpqS5SQ5syRihRxvN7a/65V1+VdJUmDaw9mwihk OIzYAgAAAA+ZTz6R9u+XcueWdu60fnfE0StHFXklUs+HPa+klCS9VPElTW48WRaLxbkNAy5GsAUA AAAeIlOmSK9bbzGrt992PNSO2zROb2/86/zlJ4Of1OxWs+Vh4aRNZDx8qgEAAICHxLZt0ptvWpff eEN69VXH6qz5c41GbRxle9wipIVWtF+hrFmyOqFL4OHDiC0AAADwELh2TerSRUpJkTp1kiZNcqzO iegT6rCkg4yMelXrpbENxyp/jvycfowMjRFbAAAAwMWMsd7O58gRqXBh6eOP7a+RlJKkFYdWqN6/ 6+lS3CVVK1hNU5+ZqgI5CxBqkeExYgsAAAC42McfS4sWSVmySGFhkr+/ffvfSLihJt800ZaTWyRJ IQEh+q7dd/LJ4pMO3QIPH0ZsAQAAABfasuWvyaKmTJHq1LFvf2OMOi/rrC0ntyiXdy69Vuc1/dbz N+5Ti0yFEVsAAADARaKipHbtpKQk6cUXpVdesb/GsoPLtOTAEnl7emtVh1WqX7S+8xsFHnKM2AIA AAAucOWK1KqVdOaMVLasNGOGZO+lsMevHtegNYMkSa/XeZ1Qi0yLEVsAAADgAYuKkp58UgoPt15P +913Us6c9tWIvBKp2jNrK+pGlIr7F9fwx4anT7OAG2DEFgAAAHiAoqKkhg2tobZgQWnTJql8eftq JKUkKXRJqKJuRKlS/kra1HWTcnjnSJ+GATfAiC0AAADwgNwKtRERUlCQtHGjVLq0/XWm7ZymX0/9 Kr+sfvq+/fcq4lfE+c0CboQRWwAAAOABcFaojU+K18T/TJQkTXhygor5F3Nyp4D7IdgCAAAA6cwY qVmz+w+1SSlJem3tazp97bQK+xZW96rdnd8s4IY4FRkAAABIZzt3Wr+yZ3c81H647UNN+M8EXYy9 KEka32i8smbJ6uROAfdEsAUAAADSWViY9XvLlo6F2jm75+i1ta9JkgKyBWha82l6ofwLTuwQcG8E WwAAACAdGfNXsG3Xzv79Z/4xU71X9pYkDas/TGMajJGXp5cTOwTcH8EWAAAASEebN0snTljvU9u0 qX37rvlzjXp830OS1KVyF41rNE4eFqbJAf6OfxUAAABAOjFGGjnSuty+vZQtW9r3jU2MVd8f+kqS elbrqVmtZhFqgX/AvwwAAAAgnSxbZh2x9fGRRo2yb9/PdnymyKuRKuxbWFMaT5HFYkmfJoEMgGAL AAAApIN166QOHazLgwZJhQunfd+klCR9+tunkqQxDcYoV9Zc6dAhkHEQbAEAAAAn++9/pdatpbg4 6Zln7B+t/f7Q9zoRfUJ5s+dVaKXQ9GkSyECYPAoAAABwouRkqVs36cYN6fHHpaVLpax23G42MTlR o362JuFe1XrJJ4tPOnUKZByM2AIAAABOcvWq1KqV9branDml2bPtC7WS9NGvHyk8KlwB2QI0pM6Q dOkTyGgItgAAAIATzJ0rlSkj/fCDdbKoOXOk4GD7aizav0jD1g+TJH3w9AcKyB6QDp0CGQ+nIgMA AAD3adYsqXt363JIiDR/vlS9un01Vh5eqdAloUoxKepepbu6Vunq9D6BjIoRWwAAAOA+bNki9epl XR48WNq3z/5Qu/vcbj0f9rySUpIUWilUX7b4ktv7AHZgxBYAAABwUFyc1LWrlJQktW0rTZ4sedg5 dGSM0ZA1Q5SQnKBnSz+rOc/NkaeHZ7r0C2RUjNgCAAAADho9WvrzT6lQIemrr+wPtZL0458/auOx jfL29Nbnz36uLB6MPQH2ItgCAAAADjh8WPrwQ+vyF19Ifn721zDGaPQvoyVJr9R8RcX8izmvQSAT IdgCAAAADhg61HoK8rPPSi1bOlZj47GN2nF6h3yy+OjNem86t0EgEyHYAgAAAHY4c0bq1Elatkzy 9LReV+uI6wnXNfSnoZKkl6u+rPw58juxSyBz4QR+AAAAII3OnpXq15ciI62Px4yRypWzv05icqJa LWilnWd2yi+rH6O1wH0i2AIAAABpEBEhvfSSNdSWKCGFhdl/W59bRm0cpQ2RG5TLO5fWdFyjon5F ndsskMkQbAEAAIB7+OwzadAg6zW1+fNLa9dKJUvaXyfySqSGbxiuBeELJEmzWs1SrcK1nNwtkPkQ bAEAAIC7CAuTBgywLrdqJf3rX1Lx4vbXSUxOVItvW2j/hf2SpCG1h+j58s87r1EgEyPYAgAAAHdw /bo0YoT06afWx6+8Ik2dKlksjtX77LfPtP/CfuXNnlc/dvhR1Qs5eB4zgNsQbAEAAIC/MUYKDZW+ /976uHt36aOPHA+1W09u1YgNIyRJE56cQKgFnIxgCwAAAPzNwoXWUOvlZf3epInjtfac26Nm85sp NjFWTUs1Vfeq3Z3XKABJ3McWAAAASOXAAetpx5I0cuT9hdrwqHA1/qaxrt68qnpF6um7dt/Jw8Kv 4ICz8a8KAAAA+J9Nm6THHpMuXpSqVJHeesuxOsYYTdg8QdWmV1PUjShVCayilaErld0ru3MbBiCJ YAsAAAAoKUnq3Vt64gnp0iXp0Ueldeskb2/7axljNGLDCA3fMFyJKYl6ptQzWtNxjfx9/J3fOABJ BFsAAABA06dLX35pXe7RQ1q/Xsqb17Fac/fM1YT/TJAkTW06VT+E/qD8OfI7qVMAd8LkUQAAAMjU Ll2S3n7buvzxx39dX+uI0zGn9eqPr0qSRj8xWgNrDXRChwDuhRFbAAAAZGqjR0tXrkiVKkl9+zpe Jz4pXqFLQhUdH62aQTU14vERTusRwN0xYgsAAIBMKzxc+uIL6/K//iVlceC34yOXjujDbR9qb9Re bT25Vbm8c2l2q9nK4sGv2sCDwr82AAAAZErGSIMGScnJUps2UqNG9tf46ehPareona7cvCJJ8rR4 anG7xSqXr5yTuwVwNwRbAAAAZDqJidLQodZJorJmlT74wL7945Pi1X9Vf83cNVOSVCuoltqWb6sG xRuoeqHq6dAxgLsh2AIAACBTOXtWevFFafNm6+MJE6QSJeyrMWrjKFuo7VO9jz5s8qGyeWVzcqcA 0opgCwAAgExj61bracfnz0u5cklz5kitW9tXY/up7Zq8bbIkKeyFMLWt0DYdOgVgD2ZFBgAAQKYQ Fma9jvb8ealiRWnnTvtD7YbIDXp2/rNKMSnq+EhHQi3wkCDYAgAAIMObM0d66SUpPl5q2VL69Vcp JMS+Gp//9rkaf91Yl+Muq2ZQTX3c9OP0aRaA3Qi2AAAAyNDCwqTu3a2zIPfpIy1ZIuXIkfb9j189 rraL2qr/qv5KNsnqUKmDfu7ys3Jny51+TQOwC9fYAgAAIMPasUPq3FlKSZF69pQ++0zysGNoZ3/U ftX9d13FxMfIw+Kh8Y3Ga2i9obJYLOnXNAC7EWwBAACQIZ09a50o6tbpx9Om2Rdqr8Rd0XMLn1NM fIxqFKqhGS1mqHJg5fRrGIDDCLYAAADIcM6ds04Udfq0VK6c9PXX9oXaa/HX1HJBS/15+U8V8yum 1R1WK2/2vOnXMID7QrAFAABAhnLunNSwoXTwoFSkiLRypeTrm/b94xLj1HReU209uVV+Wf207KVl hFrgIcfkUQAAAMgwbo3UHjwoFS4sbdwolSiR9v2NMeq1spe2ntwqfx9//R979x1f89n/cfyVIbFj b0qrFK29R83W3rVXFaVW0VqlaK0YtWpUKY0RtRW1Y9VWVSuoosQWQWSv8/vj+pVbZ3Jy4iQ57+fj cR451/ec75VP7lvu+7xzrV2dd1EyR8mEK1hEbEIjtiIiIiKS5J08CWPGwPbtZk1tnjywdy+88krs 7j977yzt1rbjysMrhESG4OLkwrrW6yibq2xCli0iNqJgKyIiIiJJmrc3dOsGYWGmXawYbNgQ+1B7 4tYJ3l5mzqf9w7S606hZoGYCVCsiCUHBVkRERESSrLlzoU8f87xBA5g0yQTb2J7GExAaQJPvmhAQ GkCF3BX4psk3pHFLQ/4M+ROsZhGxPQVbGzh79iwpU6akYMGC9i5FRERExGEsX/4s1H70kQm1Li6x v99isdBrcy9uPblF4cyF2dlpJ+nc0yVMsSKSoLR5VDxcuXKFpk2bsm3bNjJlygTAlClTKF68OEWK FOGXX355+t7ff/+dt956i7x58zJw4MDn+rHmHhERERFH9tVX0Lmzed6vH0yZErdQu+78Oop/VZzV vqtxdXZlWYtlCrUiSZhGbK108+ZNmjdvznfffUeRIkUA8PHx4ffff+f06dPMmzeP9957j59//hmA nj17snDhQjJmzEiZMmWoWbMmTZo0seoeEREREUf2+ecwerR53qMHzJgR+6nHADOPzGTA9gEApE6R mul1p2uTKJEkTiO2Vho6dCiDBg16GmoB5syZQ5cuXQDo0aMHV69e5cKFC5w+fRoXFxdeeukl0qdP T8+ePfH29rb6HhERERFH5eX1LNSOGQPz54NzHD7RTvhxwtNQ27dcX24Ousn7Zd63faEi8kIp2Foh NDSUjRs3cuHCBcqVK0fjxo25cuUKx48fJ1++fAC4urpSunRpfH19n7sOUKFCBXx9fQGsuufvhIeH ExgY+NxDREREJDk5cgTe//8MOmKECbixHam1WCyM2jOKEbtHADCm+hhm1Z9FhpQZEqhaEXmRFGyt cOHCBZydnWnevDnHjh2jYMGC9OjRg3v37uHm5vb0fdmyZePBgwf/eB2w6p6/M3HiRDw8PJ4+8ubN a1renrYAACAASURBVMsfWURERMSubtyAZs0gIsJ8/fzz2N9rsVgYtmsYY/ePBWByncmMrjEap7jM XxaRRE3B1gqRkZFky5aN8uXL4+TkRJ8+fTh69CgZMmTA39//ufdmz579H68DVt3zd4YPH87jx4+f Pvz8/OLzI4qIiIgkGk+emDB79y688QYsXRq36ccjd49k8qHJAMyqN4vBVQYnUKUiYi8KtlYoUKAA N27cICQkBDDhNEOGDBQtWpSffvrp6fuuXr1KqVKl/nL98uXLlClTBsCqe/6Ou7s76dOnf+4hIiIi klRZLHDpEkycCC+/DCdOQJYssHEjpE0b+36O3DjCxAMTAfiq4Vf0q9AvgSoWEXtSsLVC1qxZqVmz JrNmzQJgy5YttG/fns6dOzN79myioqLw9fUlb9685M2bl2rVqhEVFcUPP/wAwIoVK/jggw8ArLpH REREJDk7exZefx0KFYJPPgF/f3jlFdi0CfLnj30/kdGRvL/pfSxY6FKiCz3L9kywmkXEvpwsFovF 3kUkRffu3aN9+/bcvn2bEiVK8M0335AqVSr69OnDli1bKFGiBF5eXnh4eABw8uRJWrdujYeHB4MG DaJ9+/ZP+7Lmnv8SGBiIh4cHjx8/1uitiIiIJBlHj0L9+vDwIaRIAVWqQNeu0L49uMbxoErPA54M 9xlO5lSZudD3AllSZ0mYokXQ5297U7BNpvSLJSIiIkmJxQKLFkG/fhAaChUrwubNkDmzdf1dDrjM 6/NeJywqDK9mXnQu0dm2BYv8iT5/25emIouIiIiIXT16BG3aQPfuJtTWqwc7d1ofaqNjovnghw8I iwqjdoHadCreybYFi0iio2ArIiIiInYRHg5bt0LJkrB6tZlqPGkS/PBD3DaI+l/BEcG0WNWCnVd2 4u7izryG83Ssj4gDiONKBRERERGR+LlxAwYNMptBhYWZa6+8At7eUL689f3GWGJovKIxe37fg7uL O8taLOPVzK/apmgRSdQUbEVERETkhbBYYPFiGDgQAgPNtWzZoGVL8PSE+C5LXPjzQvb8voc0KdKw o9MOKuetHP+iRSRJULAVERERkQTn5wc9esD27aZdvjzMnQulS4MtZgrfCbrD0F1DARhbc6xCrYiD UbAVERERkQR16hTUrg0PHoC7O4wda0Zt43p8zz+JjI6k8/rOPAp7ROmcpelXoZ9tOhaRJEPBVkRE REQSzNmzUKeOCbWlS8Py5fDaa7br32Kx0G9rP3Ze2UnqFKlZ3HQxrs76iCviaLQrsoiIiIgkiG3b oGZN8PeHsmXBx8e2oRbgm5PfMP/EfJxwYkXLFRTPXty230BEkgQFWxERERGxqfBws+tx/fom1JYu bdbWZshg2+9z0f8iH277EIAJtSfQpHAT234DEUkyFGxFRERExGbOn4cKFWD6dNPu2xcOHIBMmWz7 fa4/vk6bNW0IiQyhVoFaDKkyxLbfQESSFC1AEBERERGb2LsXGjaEkBDIksUc7dOoke2/j/cZb3pt 7sWTiCdkTpWZJc2W4Oyk8RoRR6b/BRARERGReDt/Hpo3N6G2Vi04fTphQu3kg5PpsK4DTyKeUClP JQ53O0zu9Llt/41EJEnRiK2IiIiIxMvdu9CgATx6BJUrw+bNkCqV7fr3e+zHmL1jOHrzKOfunwPg 40of41nHExdnF9t9IxFJshRsRURERMRq0dHQvj38/jsULAjff2/bUBsYHki95fXwve8LgLOTM561 PRlcZbDtvomIJHkKtiIiIiJiNU9P2L0bUqeGjRvN2lpbiY6JpsO6Dvje9yVXulzMaziPcrnKkTNd Ttt9ExFJFhRsRURERMQqP/4Io0eb53PmQJEituvbYrEwYNsANv+6mZSuKdnQZgPlcpez3TcQkWRF wVZERERE4uyXX6BpUzMVuWNH6NLFdn1Hx0Qz/sfxzD4+Gyec8GrmpVArIv9KwVZERERE4uTsWahT Bx4+hEqVYN48cHKKf79RMVGM3D2Sxb8s5l7wPQBm1ptJ62Kt49+5iCRrCrYiIiIiEmsXLkDt2vDg AZQrB1u3Qtq08e/3j6nHc47PASBjyox8Uu0T+lXoF//ORSTZU7AVERERkVi5dMmcUXvvHpQsCdu3 g4eHbfqeeXQmc47PwQknFjZZSKfinUjhksI2nYtIsuds7wJEREREJPG7etWE2tu34Y03YOdOyJgx /v3GWGL4dPenDNw+EIBJdSbxXqn3FGpFJE40YisiIiIi/+r6dahZE27cMDsf79plm2N9wqLCaL+2 PesvrAdgSOUhfFz54/h3LCIOR8FWRERERP7RrVsm1F67Bq++Cj4+kC1b/PsNjQyl+crmbL+8HTcX N75u9DVdStpwa2URcSgKtiIiIiLytyIi4J134MoVePll2L0bcuaMf78hkSE0WdEEn6s+pE6Rmk3t NlGrQK34dywiDkvBVkRERET+1qBBcPgwZMhgNorKkyf+fQZFBNHIuxH7ru0jrVtatrTfQrWXqsW/ YxFxaNo8SkRERET+YskSmGNO3mHZMihYMH79RUZHsuLMCip/U5l91/aR3j09OzruUKgVEZvQiK2I iIiIPGfjRujWzTwfPRoaNoxffw9DH1JtcTXO3T8HmDNqt3fcTrnc5eJZqYiIoWArIiIiIk9t2wat WkFUFLRvD6NGxa8/i8VC1++7cu7+OTKnykz/Cv3pVbYX2dLYYAcqEZH/p2ArIiIiIoDZHKp582eb Rnl5gXM8Fq5FREcw9dBUvr/4PW4ubmzvuJ0yucrYrmARkf+nYCsiIiIiHDgAjRtDWJj56u0NrlZ+ UrwbdJcem3qw88pOwqLCAJhRd4ZCrYgkGAVbEREREQd36BA0aAAhIVC3LqxeDSlSWNfXnaA71PKq xXn/8wBkTpWZXmV70atsLxtWLCLyPAVbEREREQe2YQO0a2dGamvWhHXrwN3dur78Q/yfhtrc6XKz oe0GyuQsg5OTk22LFhH5Ex33IyIiIuKg5s2DFi1MqG3UCDZtgtSpresrOiaaDus6cN7/PHnS52Hv u3spm6usQq2IvBAKtiIiIiIOyMsLevcGiwV69oT16yFNGuv7G7d/HDsu7yCVayq2tN9CwUzxPPhW RCQOFGxFREREHMzWrc/Oqf34YzNya+1GUQBLTi3hs32fATC/0XzeyP6GDaoUEYk9rbEVERERcQCP HsG+fXDwIMyZA9HR0KkTTJoE1s4WDosK48OtH/L1z18D0LNMTzqV6GTDqkVEYkfBVkRERCSZ+/57 ePddE27/8Pbb8M038Tun9t0N77Ly3EqccGLkmyMZXX10vGsVEbGGgq2IiIhIMhUZCcOHwxdfmPbL L0Pt2lCtGrRubf2RPgBbL21l5bmVuDi5sKndJuq/Wt82RYuIWEHBVkRERCQZsliga1dYvty0Bw4E T09wc4t/36GRofTd2heAARUHKNSKiN0p2IqIiIgkQzNnmlDr4gLffQfvvGO7vsftH8eVh1fInS43 Y2qMsV3HIiJWUrAVERERSWb27jW7HYOZhmyrUGuxWBj/43gmHJgAwMx6M0nrltY2nYuIxIPDHfez bNmyv1w7e/Ys/v7+dqhGRERExLauXTPrZ6OjoUMH6N/fNv36Pfaj6/dd+XTPpwAMrTKUFkVa2KZz EZF4crhgO3DgwL9cy507N8OHD7dDNSIiIiK28/Ah1K8P9+9DyZLw9dfWH+Xzhwv+F2i5qiX5Z+bH 65QXTjgxu/5sPOt44hTfzkVEbMQhpiI/fPiQ27dvA5AtWzbOnz+PxWJ5+vr169dZv349CxYssFeJ IiIiIvESHg7Nm8P585A7N2zaBKlTW99fRHQEngc8Gf/jeCKiIwCokb8GI6uNpPbLtW1UtYiIbThE sL1+/TqtW7cmIiKC27dv06BBg+deT5UqFX369LFTdSIiIiLxExICnTrBvn2QPj1s3Qp58ljXl8Vi YY3vGj7Z/Qm/BfwGQINXGzC5zmSKZStmw6pFRGzHIYJtiRIluHjxIgBDhgxh8uTJdq5IRERExDau XTMjtSdPgqsrrF0Lb7xhXV/3g+/TbGUzDvkdAiB7muzMqDeDNsXaaNqxiCRqDrfGdtKkSaxbt469 e/c+vXbmzBn7FSQiIiJipcOHoWxZE2qzZIGdO6FOHev6CggN4K2lb3HI7xBpUqRhTPUxXOp3ibav t1WoFZFEzyFGbP9Xnz59OHnyJE+ePOHs2bMA/Pbbb1y5coWmTZvauToRERGR2Ll714zU+vtD6dKw fj3ky2ddX4/DHlNvWT1O3T1F9jTZ2ffuPgpnKWzbgkVEEpDDjdju2bOHAwcOkC5duqfXmjdvzty5 c+1YlYiIiEjsxcRAx44m3L7+Ouzfb32otVgsdFzfkeO3jpMldRZ8Ovso1IpIkuNwwTZPnjy4uLg8 N6Xm3r17XLp0yY5ViYiIiMSepyfs2mV2PV65EtKksb6vaYensfnXzbi7uLO1w1ZtECUiSZLDBdv+ /fszZMgQgoODOXjwIDNnzqR8+fK0bNnS3qWJiIiI/KclS2DUKPN8zhwoWtT6vo7cOMIwn2EAzKg3 g7K5ytqgQhGRF8/h1tg2btyYrFmzcuXKFXr06EH27NkZMmQIvXr1sndpIiIiIv8oOBj69oVvvzXt Ll3Mw1pBEUG0XdOWqJgo2hRrQ88yPW1Sp4iIPThcsPXx8SF//vysWbPG3qWIiIiIxMr9+1CzJpw7 B87OMGYMfPIJxGez4tF7RnPt8TXyZ8jP142/1s7HIpKkOdxU5A4dOnDs2DF7lyEiIiISK9HRZqOo c+cgRw7w8YFPPwUXF+v7PHn7JDOOzgBgboO5pHdPb6NqRUTsw+GCbZMmTShW7K+bIhw9etQO1YiI iIj8uwkTYMcOSJXKnFNbo0b8+ouOieb9ze8TY4mhTbE21H+1vk3qFBGxJ4ebilyrVi06duxI4cKF qV69Os7OJtuvWbOG3bt327k6ERERkWd8fGD0aPN83jxztE98TT8ynZ9u/YSHuwfT606Pf4ciIomA wwXbuXPnUrx4cVKkSMGJEyeeXr9x44YdqxIRERF53o0b0L49WCzQrVv8Nor6w5EbRxjuMxyAKW9N IWe6nPHvVEQkEXC4YPvll19SokSJv1w/cOCAHaoRERER+avQUGjeHO7dgxIl4Msv49/ng5AHtF7d +ukuyN1Ld49/pyIiiYTDrbGdPXv2X65FR0fj7e1NgwYN2LZtmx2qEhERETEsFujZE376CTJnhg0b zPra+Hgc9pj269rjF+jHq5le1S7IIpLsOFywdXd3p169ejRr1oyHDx8CsHz5crZs2ULPnj3x8vJi 165ddq5SREREHNWMGbB0qdn1eNUqyJ8/fv3tvLyTN+a9wY7LO3B3cWdVq1XaBVlEkh2HC7bXrl2j ZcuWdO/enU8++QSAPXv20LlzZ5o2bYqXlxczZsywc5UiIiLiiBYtgo8+Ms+/+AJq1bKun8DwQBaf XEztJbV5e9nb+AX68XLGl9nVeRclc5S0XcEiIomEw62xvXHjBj169ABgwoQJAJw4cYJGjRoB4Obm xp07d+xWn4iIiDimefOgd2/zvHdv6N8/7n2ERIYw7fA0Jh2cRFBEEABOONG7XG8m1ZlEGrc0NqxY RCTxcLhgmzdvXoKCgjh//jw3btygfv36XL58mevXrwNw7969p1OURURERF6EGTNg4EDz/MMPYfp0 iOsS2LW+a+m/rT+3ntwCoHDmwnQq3okOxTuQP0N+2xYsIpLIOFksFou9i3iRjh07Rtu2bfH398fb 2xs3Nzfc3NxYvnw5uXLlYuvWrbz55ptMnTrV3qXGS2BgIB4eHjx+/Jj06bWORkREJDGKiYHhw2Hy ZNMeOhQmTox7qN1wYQMtVrbAgoX8GfIzsfZEWhdrjbOTw606E7Ebff62L4cLtgBRUVGEhYWRNm3a p9ciIyOZOHEiLi4ufPzxx7i7u9uxwvjTL5aIiEjiFh4OXbvCihWmPXYsjBgR91B7yO8QtZfUJiwq jPdKvsfchnNxd03an2NEkiJ9/rYvhwu2oaGhzJs3j9y5c9OmTRvAbB5Vs2ZNO1dmW/rFEhERSbxC QqBRI9izB1xdYeFC6NIl7v1c9L9I5UWVCQgNoFGhRqxvsx5XZ4dbaSaSKOjzt3053PyULl26cPjw YcaOHfv0mqurKwsXLrRjVSIiIuIooqKgbVsTatOlgy1brAu1EdERtF7TmoDQAMrnLs93Lb9TqBUR h+Vwwfb8+fOsXr2adOnSPb1WrVo1VvwxD0hEREQkgVgs8MEHsGkTpExpQu1bb1nX1+f7Puf03dNk SZ2FjW03asdjEXFoDhds8+TJQ0xMDE7/s4Dl4sWLXL161Y5ViYiIiCMYM8ZMO3Z2Nmtrq1a1rp9j N48x8cBEAL5q+BXZ02a3XZEiIkmQw81XGTVqFN26dSMgIIDly5dz/PhxvLy8GDFihL1LExERkWQq PBw++gjmzDHtuXOhWTPr+gqOCKbLhi7EWGJo/0Z7WhZtabtCRUSSKIcLtpUqVSJnzpzMnj2b5cuX kz17dpYvX57kd0EWERGRxMnPD1q1gqNHTdvTE3r2tK6vw36H6byhM78F/EbOtDn5sv6XtitURCQJ c6hge/XqVW7fvk25cuWeO6fW19eXtm3bcvr0aTtWJyIiIslJRAR8+y188gk8eAAZM8KyZdCgQdz7 Co8K57N9nzHp4CRiLDHkTpeb1a1WkylVJpvXLSKSFDnMGtvVq1dTsmRJ+vfvT8mSJbl79y7h4eF8 8sknVKtWjY4dO9q7RBEREUkGIiJg/nx49VUzMvvgAZQuDSdOWBdqT989TfmF5Zl4YCIxlhg6Fe/E 2d5nqZS3ku2LFxFJohxmxHbixIns3buXUqVKsXXrVjp27Mj169epXr06586dI0eOHPYuUURERJK4 X36BTp3g7FnTzpkThg41ATdlyrj1ZbFYmHpoKiN2jyAyJpKsqbMyv9F8mhdpbvvCRUSSOIcJtlFR UZQqVQqA+vXr06tXL7Zt20aRIkUAOHfuHMWKFbNniSIiIpJERUXBpElm1+OoKMiaFT79FLp3h1Sp rOtz+pHpDNk1BIBmrzVjfqP5ZEuTzXZFi4gkIw4TbGNiYpg7d+7Ttru7O3v27GHPnj0AeHt7c+DA AXuVJyIiIknU4cPQvz/89JNpt2gBX31lwq21Nv+6mY93fAyAZ21PhlQZ8txRhSIi8jyHCbbZsmXj +PHjT9tVqlR5rn3v3j17lCUiIiJJ1PXrMGyYOY8WwMMDZs+GDh0gPhn09N3TtFvbDgsW3i/9vkKt iEgsOEyw7dmzJ23atPnH13fs2PECqxEREZGkymIxm0MNHAhhYSbEdu0K48aZNbXx4R/iT5MVTQiK CKJWgVrMbjBboVZEJBacLBaLxd5FiO0FBgbi4eHB48ePSZ8+vb3LERERSRbCw6FPH/jmG9OuXh2m T4f/38YjXqJjomno3ZDtl7dTMFNBjnY/quN8RJIQff62L4cZsRURERGJj5s3oWVLOHoUnJ1h4kQY PDh+047/17j949h+eTupXFOxrvU6hVoRkThQsBURERH5D76+8NZbcOsWZMgA330Hdevarv9tv23j s32fATC/0XzeyP6G7ToXEXEACrYiIiIi/+L4cahXDwICoEgR2LgRCha0Xf/7r+1/ullUzzI96VSi k+06FxFxEM72LkBEREQksdqzB2rVMqG2XDn48UfbhtoVZ1bw1tK3eBT2iKr5qjKj3gzbdS4i4kAU bEVERET+xsaNUL8+BAWZcOvjA5kz26Zvi8XCxB8n0n5deyKiI2hZpCU7Ou4gpWtK23wDEREHo6nI IiIiIn+ydKk5wic6Gpo2NWtqU9ooc94Pvk/X77vyw6UfABhUcRBT3p6Cs5PGG0RErKVgKyIiIvI/ vvwS+vc3zzt3Nkf7uNroE9Puq7vpuK4jt4Nu4+7izox6M+hVtpdtOhcRcWAKtiIiIiKAxQLjxsGo Uabdv785o9bZBgOpFouFyQcnM9xnOBYsFMlShO/e+Y7i2YvHv3MREVGwFREREQGYNu1ZqB0zxjy3 xRm1IZEhdN/YnRVnVwDwXsn3mFV/Fmnc0sS/cxERARRsRURERNiwAQYPNs89PWHo0Pj3GWOJYf+1 /Xy04yN+vv0zrs6uzKo3iw/KfRD/zkVE5DkKtiIiIuLQTpyADh3MVORevWDIkPj1dz/4PvNPzGfx L4u58vAKAFlSZ2FNqzVUz1/dBhWLiMifKdiKiIiIw/Lzg8aNISQE6tY1G0dZO/04Oiaa+SfmM2L3 CB6FPQIgnVs62hRrw8g3R/JShpdsWLmIiPwvBVsRERFxSAEB0LAh3L4Nr78Oq1ZZv/vxsZvH6P1D b07cPgFAiewlGFRpEC2LtNRaWhGRF0DBVkRERBzOgwdQpw6cOQM5csDmzZA+fdz7OX7zOJ/v/5zN v24GwMPdg3G1xvFB2Q9wcXaxcdUiIvJPFGxFRETEofj7m1B76hRkzw4+PvBSHGcJH71xlM/2fcbW 37YC4OzkTKfinfCs40mOtDkSoGoREfk3NjiZzbGFhYUxZ86cp+1ly5ZRpkwZChQowM6dO59eDwgI oHnz5hQoUIB27doRERERr3tEREQk7vz9oXbtZ6F2zx4oWjT29x/2O0y9ZfWo+E1Ftv62FRcnFzqX 6Mz5Puf5ttm3CrUiInaiYBtPI0aMoG/fvgBcvHiRNWvWcOLECebOnUvHjh2JjIwE4MMPP+STTz7h 0qVL+Pn58dVXX1l9j4iIiMTd+fNQrRqcPm2mH+/dC0WKxO7eKw+v0NC7IZUXVWb75e24OLnQtWRX LvS9gFczLwplLpSgtYuIyL9TsI2H8+fP8+jRo6ft+fPn0759ewDq169P1qxZ8fHxwd/fnzNnzlCu XDlcXV0ZMGAA3t7eVt8jIiIicbNqFZQrBxcuQO7cZqT2tdf++77I6Eg8D3hSbG4xtlzagquzK91L defXfr+yqOkiCmYqmPDFi4jIf1KwjYcRI0Ywfvz4p+3jx4+TL1++p+0KFSrg6+vLqVOnyJEjx1+u W3vP3wkPDycwMPC5h4iIiKOLiIABA6BNGwgOhlq14OefYxdqz9w9Q5mvyzDcZzhhUWHUKlCLc73P saDJAl7O+HLCFy8isWOxwL59ZotzcVgKtlZas2YN1apVey583rt3Dzc3t6ftbNmy8eDBg7+9/uTJ EyIjI6265+9MnDgRDw+Pp4+8efPa8scVERFJcn75BapWhZkzTXv4cNixA7Jl++97D/kd4s1v3+TM vTNkTpUZr2Ze7Oq0S1OORRKLiAhYuBDGjjXnddWoAfPm2bsqsSPtimyFkJAQFi9ezPfff//c9QwZ MuDv7//ctezZs//t9YwZM5IiRQqr7vk7w4cPZ9CgQU/bgYGBCrciIuKQnjyBUaNg1iyIiQEPD/Dy gqZNY3f/jss7aL6yOSGRIVTJW4UNbTeQJXWWhC1aROJm4kQYM+ZZO3Vq8wsvDksjtlZYuXIlV69e pWrVqlSsWBGAzp07U7RoUX766aen77t8+TJlypShSJEinD179umuxn9cB6y65++4u7uTPn365x4i IiKOxGKBtWvNhlAzZpjPuG3agK9v7EPtWt+1NPJuREhkCHVfqcuOTjsUakUSk2PHoF2750Ptl1/C rVswbpzdyhL7U7C1QteuXfH19eXIkSMcOXIEgCVLltC5c2cWLVpEYGAgd+/exd/fnypVqpA/f35K lSrFggULAFi6dCn9+vUDsOoeERERed6VK9CwIbzzDty8Ca+8Atu2wXffQa5c/31/RHQEI3xG0HpN ayJjImlVtBUb220kdYrUCV+8iPw3i8UE2CpVzC82QIECEBkJffuaqRni0BRs4+H3339n6tSpACxf vpyaNWvSrVs3ihUrRseOHVmyZMnT9y5dupSvv/6aEiVKkCVLFpo0aQJg1T0iIiJiRETAhAlQrBhs 3QpubvDpp3DmDNStG7s+zt47S/kF5ZlwYAIxlhh6lenFipYrcHNx+++bReTF6NUL+veHqCho1gzW rIEDB8BVKyvFcLJYLBZ7FyG2FxgYiIeHB48fP9a0ZBERSZaOHYN33zXn04LZ8XjuXChcOHb3WywW ZhyZwTCfYURER5A5VWbmN5pPy6ItE6xmEbHCqlVmXYGLC0ydCh9+CE5O9q7qL/T52740YisiIiJJ SmSkWV5XubIJtdmywbJlsGtX7EMtwOi9oxm0YxAR0RE0KtSIs73PKtSKJCbr1kHjxibUgplyPGBA ogy1Yn8auxcREZEk4+JF6NQJjh837bZtYc4cyJQpbv2M3TeWsfvHAjC5zmQ+rvwxTvqwLGI/gYHw +eewaRNkzQp58z5bSwvQsaM52kfkHyjYioiISKJnscA335gldqGhkCGDmXbcrl3c+/I84MmovaMA mPLWFD6u/LGNqxWRf3X7NqRMCRkzmvbZs9C69bN1Bb/++uy9KVLA+vVmdziRf6FgKyIiIolacDD0 7g1/7K9Ypw4sXgx58sS9r2mHpzHcZzgAE2pNUKgVeVGCg83XkSPNeVzOzmYUNiAANm82r+XODZMn m2sbNkBQkJmOHJutzcXhafOoZEqL10VEJDm4eBFatoRz58zn4PHjYcgQ8zyuFpxYwPub3wdgTPUx jK4x2sbVisjfmjULRowwQfWfNGgAX31lpiAnUfr8bV8asRUREZFEadUq6NbNfBbOnt0st6tRw8q+ zq2i5+aeAAypPIRR1UfZrlARed758/DLL7Bx4/PrZP8wfz68+qoZlT16FN58EyZN0qZQEi8KtiIi IpKoWCzw2WfmAVC9OqxYATlzWtfftt+20XFdRyxY6FmmJ551PLVRlEhCOHQIvv4avLz++tqwYeZr iRJm1zeAmjVfXG2S7CnYioiISKIRFgbvvWeCLMDHH8PEieBq5SeWg9cP0mJlCyJjImlTrA1zV+Uq KwAAIABJREFUGsxRqBWxBYsF9u6FefPgyhXo2tXs7hYTY16vXNmMyjZtar6+/rpdy5XkT8FWRERE EoV796BZMzh82ATZr74yU5GtEWOJYeaRmQz3GU54dDj1C9ZnSfMluDi72LZoEUd04YJZ/O7r++za iRPma+nS5lieBg3sU5s4LAVbERERsbvTp83Azu+/m6N81q6FWrWs6+tG4A3e3fAuPld9AGhUqBEr 31mJm4ub7QoWcWQDBphQmzYttG9vjuO5fx9q1zbn0KZKZe8KxQEp2IqIiIjdBAebtbTTp0NUFLzy ijn547XXrOtvje8aemzqwaOwR6RyTcW0utPoWaanph+LWCs6GhYuhDNn4NQp8/XxY3BxMRtEvfKK 2ar85Elo3txcF7EDBVsRERGxi++/N0vyrl837WbNYMECyJIl7n1ZLBYmH5zMMB+zQU3ZXGVZ1nwZ hbMUtmHFIg5o6FD44ou/Xu/UyYRaMF//eC5iJwq2IiIi8kL5+UHfvuYkEICXXoLZs6FRI+v6i7HE MGj7IGYenQnAwIoDmVRnEilcUtioYhEHc+CAWeh+8uSzUPvOO2bdbMGCZuOo8uXtW6PInyjYioiI yAthsZgZjR99BE+emM/NgwfDyJGQOrV1fUZER/DuhndZcdZsozzt7WkMrDTQhlWLOJhx4+DTT5+/ Nno0jBljl3JEYkvBVkRERBLc779Djx6wa5dpV6pkQm7Rotb3eT/4Pu3XtWfXlV2kcE7Bt82+pf0b 7W1Sr4jDCA2FkBAzlWLRIvjyy2evublBr14wapT96hOJJQVbERERSTAWC8yfb0Zmg4LMZqnjx5u1 tdbuMWOxWFh1bhV9t/bFP8SfNCnSsK7NOt5+5W3bFi+SnPn5wbRpZmF7cPDzr40dCx9+aJ6nS/fi axOxgoKtiIiIJIioKHj/fVi82LSrVYNvvoFXX7W+zztBd+j9Q2/WX1gPQPHsxVnSbAklcpSwQcUi DuDWLVi2zEwvDgt7dj1tWnjjDWjYED75BLSTuCQxCrYiIiJicyEh0LatOdLS2RmmTjUDQM7O1ve5 /vx6um/qTkBoAK7OroysNpLh1YbrfFqR/3L+PMycCYcPm0Oj/1C1qllPW60apEhhFr6LJFH61ysi IiI29fAhNG4MBw9CypSwciU0aWJ9f1ExUYzcPZJJBycBUCpHKRY3XaxRWpF/ExJiFrXv2AHz5kFM jLnu5ATlyplF7926aWRWkg0FWxEREbGZmzehbl04dw4yZDBH+lSrZn1/94Pv03ZtW3Zf3Q3AoIqD 8KzjqaN8RP5XWJgZcf1j4fqxY9CuHVy58uw9TZpAly5QpQpkz26fOkUSkIKtiIiI2MSFCybUXr8O uXLBtm1myZ61jt08RstVLbkReIM0KdKwqOkiWhdrbbuCRZKDq1fNX49SpYKtW+HePahfHwIDIXNm qFgRevY00yhEkjEFWxEREYm3Y8egQQN48AAKFYLt2yF/fuv787niQ6MVjQiLCqNQ5kKsb7Oeolnj cTaQSHLk7w+tWpmpEmD+khQRYaYdv/mmWeSePr19axR5QeKxhYOIiIiICbG1aplQW64cHDgQv1C7 7/d9NF7RmLCoMOoXrM/xHscVakX+7PRpKFYMTpwwI7NlypgpyTExZue2zZsVasWhaMRWRERErObt bZbtRUXB22/D2rXm1BBr/XjtRxp6NyQ0KpQGrzZgXet1uLu6265gkeQgMBDeecdMOy5aFJYvh+LF 4ddfzdbjhQrZu0KRF07BVkRERKwycyYMGGCet2sH334LbvE4eeeQ3yEaeDcgODKYt195m7Wt1yrU ivzZ/ftmveylS5A3L+zfb0ZsAV57zb61idiRpiKLiIhInISHw8CBz0Jt//6wbFn8Q229ZfUIigii doHabGizgZSuKW1TsEhyEBZmQmzlynD0KGTKBOvWPQu1Ig5OI7YiIiISa6dOQceOcPasaU+YAMOG WX8UpsViYfax2Xy04yMiYyKpkb8GG9ttJFWKVLYrWiSp8/c3uxtfvmza+fObHZA1QivylIKtiIiI /KfoaJgyBUaNgshIyJYNFiwwR2NaKzA8kO4bu7PadzUALYq0wKuZF6lTpLZR1SJJ0JkzsGYNBAVB 585gscD48SbUenhA7dowZw7kyGHvSkUSFQVbERER+VeXL5vP14cOmXazZvD115A1q/V9nrpzilar W3Ep4BKuzq5MfWsq/Sv0x8naoV+RpOzhQ5g3D/buhZ07n12fNu3Zc2dn2LEDypd/4eWJJAUKtiIi IvKPli6FDz6A4GBIlw6+/NKEXGvzZ3hUODOPzmT03tGERYWRN31eVrVaRcU8FW1buEhSER4O9eqZ w6DBBNgmTcx5tNu3mzW0+fJBv34KtSL/QsFWRERE/iIkxGwK9c03pl2jhtn1+KWXrO/zh19/YMD2 AfwW8BsA9QvWZ2nzpWROrc1vxEFZLPDhhybUZsgAI0ZA06bw6qvm9ZgYE3RF5D8p2IqIiMhzLlyA Vq3MBlFOTjBmjPm87eJiXX+/PviVgdsHsuXSFgBypM2BZ21POpXohLOTPrSLg/LzM2tl5883v2jL lkHDhs+/R6FWJNYUbEVEROQpb294/30z9Th7dtOuVcu6vi74X2DywcksPb2UqJgoUjinYEDFAYx8 cyTp3dPbtnCRpGTOHDMlIibGtKdP/2uoFZE4UbAVERER/P1hyBBYvNi0a9Y0odaajVeP3zyO50FP 1p9fjwULAA1ebcD0utMplLmQDasWSWKCgswOx56epl22LHTrBr162bcukWRAwVZERMSBRUWZmZCf fmo2ZnVyMs9HjYrb1OOomCh++PUHZh2bxe6ru59eb1q4KcOqDtPmUOJ4oqLg5Em4cweuXDG/aBcu mHW1YA6AnjDB+p3YROQ5CrYiIiIOat8+Mxvy9GnTLl4cZs+GatVi38eNwBss/HkhC39eyM0nNwFw dXalwxsdGFJlCEWzFk2AykUSoYAAE1RdXCB/fvMXogcP/vq+l182h0K3aPHCSxRJzhRsRUREHMyN GzB4MHz3nWlnzAjjxpm1ta6x/GRw4PoBphyawuZfNxNjMesEs6TOQteSXelbvi/5PPIlUPUiidDB gyao3rv3/PUMGcwOxylSQMeO0Ly5WbyuUVoRm1OwFRERcRBhYTBtmlniFxJiPlv37GlCbeZYnrhz 68ktBu8cjPcZ76fXqr9UnV5le9H8tea4u7onUPUiiUxwsFkr6+ICM2bA48dQtChkywbHj8PIkfDx x7H/a5GIxIt+00RERJI5iwU2b4aBA+HyZXOtalWYNQtKlYpdH5HRkcw6Oosx+8YQFBGEE050K9WN QZUGUSRrkYQrXsReIiPNmVc//WTWyhYrBr17m78IhYebEdodO569v2pV006VSufPitiBgq2IiEgy 9uuvMGAAbN1q2rlymeV97drFfjbknqt76Lu1L773fQGokLsCsxvMpmyusglUtYid/fADdO9uNn76 X7/8AoUKwZIlJvSmTg0NGphR2blzTagFhVoRO1CwFRERSYZCQuDzz83U48hIs8Rv0CAYMQLSpYtd H6GRoQzeOZg5x+cAZg3tpDqTeLfkuzg76YO7JDOhoeYXZvNmOHLEXPPwMEfy5MkDXl6wcOGz96dJ A+vXw1tv2adeEXmOgq2IiEgys3+/ORrzt99Mu0EDswTw1Vdj34fvfV/armnLmXtnAOhdtjfjao0j Y6qMCVCxiJ0dPQpdusDFi8+uffihWUObMqVpN2kC8+aZsFutGnTubHZeE5FEQcFWREQkmQgKMkdj zjEDrOTObWZHNmkS+z4sFgsLfl7AgG0DCI0KJVuabCxptoS6BesmTNEi9hQZCWPGmAAbEwM5c8Jn n0G9epA37/PvbdFCR/SIJGIKtiIiIsnAzp3Qowdcu2baPXqYtbQeHrHvIyA0gB6berDu/DoA6r5S F69mXmRPmz0BKhaxs5s3oXVrOHTItDt0MDuqZcpk37pExCoKtiIiIknY48fw0UfwzTemnT8/LFgA derEvo8HIQ+Yc3wOXx77Ev8Qf1I4p2Bi7YkMrDRQa2kl+QkNNVMZPv8cAgPNX38WLoR33rF3ZSIS Dwq2IiIiSZSvLzRt+mwtbb9+MGECpE0bu/v9Hvsx7fA0Fvy8gODIYAAKZy7MshbLtOOxJD9+fmaN 7Ndfw4MH5lq5crBiBbzyin1rE5F4U7AVERFJgr7/Hjp1gidPIF8+WLbM7GcTG78++JUJP05g+Znl RMVEAVAyR0mGVhnKO0XfwdVZHw8kGXn0CAYPhsWLITraXMuXD0aPhnff1dE8IsmE/p9LREQkCYmJ gXHjzGdygOrVYfVqyJo1dvd7n/Gm+8buhEaFAlAzf02GVhnK26+8jVNsD7YVSQosFli71hzkfPOm uVazppna0LixOXtWRJIN/UaLiIgkEU+emAGmdWZvJ/r2Ncdupkjx3/dGRkcyeOdgZh6dCUCtArWY WHsi5XOXT7iCRezl8WOzMdSOHaZdsCAsWhT7aQ0ikuQo2IqIiCQBly+b9bTnzoGbm1kq+N57sbv3 btBd2qxpw75r+wAYUW0En9X4DBdnlwSsWMROAgPNcT1HjpgzaIcMgaFDIXVqe1cmIglIwVZERCSR 27DBhNiHD80xm2vXQqVKsbv32M1jtFjZgptPbpLOLR1ezbxoXqR5whYskpDu3IHZs+HYMYiKgvTp 4eWXza7GhQpBo0Zw9Kg5tsfHB0qWtHfFIvICKNiKiIgkUkFBMHCgOYkEoEIFMw05V67Y3b/w54X0 2dKHiOgICmcuzIa2G3gty2sJV7BIQrJYwNMTxoyBiIi/vj59ugmzAQGQMaM53FmhVsRhKNiKiIgk QseOQYcO5igfJyczm/Lzz8005P8SHBFMny198DrlBUCz15rh1cyL9O7pE7hqkQRy7Rp8+iksXWra lSpB167mbKuAADNCu3y5ef7SS7B1KxQpYt+aReSFUrAVERFJRKKjYeJEMygVHQ158pjP8jVqxO7+ 8/fP887qd/C974uzkzNja45lWNVhODvpSBNJYqKizBE9np5w5Yq55uJipiH36vX8e/v0gY8+gi1b TODNkePF1ysidqVgKyIikkhcvWrOpj140LTbtDGbRGXMGLv7l51eRs/NPQmJDCFH2hysaLmCGvlr JFi9IjZjsZjDmc+dMxs+/fabmUp8+bJ53dUVypWDsWOhdu2/76NECfMQEYekYCsiImJnISEwZ445 nzYwENKlM+2OHc005P8SGhlK/639WXjSLMatXaA2y1ssJ3va7AlcuUgcxcTAqVPmXFl3d/DwMIcw Dx8OK1f+9f2ZM8PIkdC9u5l2LCLyDxRsRURE7CQ8HBYsgPHjzUavAFWqmKnHBQrEro9fH/xKq9Wt OH33NE44Mbr6aEa+OVJH+cjfCwuDM2fg9GkzKnrlitmNbPDg2O9KFhsBAabvmBhIk8Z8v61bYft2 uHfv7+9xdTVnz4KZSlylCtSpY3Y9FhH5Dwq2IiIiL1hUFHh5mc2grl831/LnN+tqO3Y0ywhjY8mp JfT+oTfBkcFkS5ON5S2WU+flOglVtiQ1f4yO7t8PP/8MJ0+Cr69ZvP1n8+fDBx+Yey5cgCxZzMLu 8uVNME2VCrJnB2fn5/tfsQJWrTJThm/cMH3HxJhpCP8kbVooXBgiI+HBA7h1CwoWhCVLoGJFm//H ICKOwclisVjsXYTYXmBgIB4eHjx+/Jj0+kuniEiiEBNjZluOHg2XLplruXKZmZbdusVux2OAJ+FP 6L2lN8tOLwOgRv4aLG+xnFzpbDjiJknLtWuwejWcPQuhofDkidla+8GDv743SxZzDE7hwmYH4Q0b 4NCh//4eqVObnYZfeQVy5zZnxJ4+/c/vz5kTUqQw51blyQP16plHlSrP/2OPijJ/zYnNvHuRREyf v+1LwTaZ0i+WiEjicvYsvPsunDhh2lmymGWFH3xgBsNi68StE7Rd25bfAn7DxcmFz2p8xrCqwzT1 2JFcvGg2UXr0yATC27fh+PG/f2/atFC9uhl5LVXKPHLnfj5EWiwm3K5ebaYAFyliphLs2QPnz5s5 86Gh5i8zf+bhYQ5brlgR8uUz62bB9JM6te1/dpFETJ+/7UtTkUVERBJQdDRMnw4jRkBEhMkBH38M H35oNomKrRhLDDOOzGDYrmFExkSSzyMf3i28qZKvSsIVL4nPypVmI6WgoOevOzmZqcN16ph/WKlS mYBavrwZNf03Tk7QvLl5/JOoKLNm9tw5Mzrs52c2durVCzJlivePJSISXwq2IiIiCeTqVTNKu3+/ aTdsaDaLypkzbv3cDLxJj0092PrbVgBaFmnJgsYLyJgqlucASeL0+DHcv2/+QaRJ89fX7941a1hv 3jTtGzfgu+/M8+rVoXNn85cTd3d46624/8OKC1dXKFTIPEREEiEFWxERERuzWGDRIhgwwAyspU1r Rm27dYvbMsLomGi++ukrhvsM50nEE1K6pmRG3Rm8X+Z9nLQeMWn79lvo3dtM8QWz82/+/PD661C0 qNno6fvvzUjpnw0bZqYiu+pjnIjIH/S/iCIiIjZ05w706AGbN5t2tWomw7z8ctz6OXP3DO9vfp8j N44AUDFPRRY2XkixbMVsW7C8WKGh0Lev+csHmE2UIiLMAcanT/91M6YKFaBqVfPcyQnq14datV5s zSIiSYCCrYiIiA1YLObUkz59zEa0bm7mfNqBA2N/fA9AaGQon+/7nKmHpxIVE0U6t3R41vGkZ5me 2iAqKQsOhjVrYOpUs5OYs7M572n4cDOsf+uW2Sr77FlzJE/WrNC1K7zxhr0rFxFJEhRsRURE4un0 aejfH/btM+2SJWHpUjOrNC52Xt5Jrx96ceXhFQBaFGnBrHqzyJ0+t40rlhciOhoOHIDly83a2CdP zPWsWc3a2dq1TTt9evN47TVo3Nh+9YqIJGEKtiIiIlZ68AA+/RTmzzcnoaRMaQbghg2L/Zm0ABf9 LzLcZzjrL6wHIE/6PMyuP5umrzVNoMrFpmJiYO9es2twZKR5XL5sjs+5ffvZ+15+Gd57z8xVz5bN buWKiCRHCrYiIiJxFBUFX30Fo0bBw4fmWqtWMGUKvPRS7Pu5G3SXz/d9zvwT84m2ROPs5Ezfcn0Z V2sc6dzjcBaQ2Mf167B4sXlcu/b378mQwRyj07kzvPmmmYIsIiI2p2ArIiISB3v2mGnHZ8+advHi MHOmOUI0toIjgpl2eBqTD00mKMKcR9q4UGM863hSNGtR2xcttnXunNmVeNUqs7gaTICtUsUcvePm Zg4sbtQI3n47bsP3IiJiFQVbERGRWLhxw2wEtWaNaWfKBOPGmVmlsT11JTommsW/LGbUnlHcDjJT VMvmKsuUt6ZQI3+NhClcbOf8eRg92vwj+CPQ1qplznFq3hxSpbJvfSIiDkzBVkRE5F9ERcGsWSbP BAWZmaQffGA2tM2UKfb9nLt3jo7rO/LLnV8AKJChABNqT6B1sdY4O2l6aqK3cKE5pic83LRbtjRz 0YsXt29dIiICKNiKiIj8o8OHoVevZ0eLVqoE8+ZBiRKx7yPGEsPsY7MZsnMI4dHhZEyZkU/f/JTe 5Xrj7uqeMIWL7fz53Nm6dc1iah3DIyKSqCjYioiI/ElAgNnZeMEC086UCSZNMhvaxmXvn1tPbvHe 9++x/fJ2AOoXrM+ipovIkTZHAlQtNhUTY47qGTgQfv7Z/Bc/dqz5h6ENoEREEh0FWxERkf9nsYCX FwweDP7+5lrXribUZs0at77Wn19Pj009eBD6gJSuKfni7S/4oOwHODk52b5wsQ2LBS5cAG9vWLYM fv/dXM+SxZw7W6eOXcsTEZF/pmArIiKC2eW4d2/48UfTLlbMTDuuVi1u/TwJf8KAbQNY9IuZulo6 Z2mWNV9GkaxFbFyx2MTVq7BjB+zbZx63bj17LV06c47TmDGQN6/dShQRkf+mYCsiIg7t6lWTW5Yu NQN2qVObjaIGDoQUKeLW18HrB+myoQuXH17GCSeGVhnKZzU/w81Fx70kKsHBsHatWTe7b9/zr7m5 mZHZTp2gSRPzD0JERBI9BVsREXFIt2+b43oWLIDISHOtZUv44gt46aW49eX32I9hPsPwPuMNQD6P fCxtvpQ3X3rTxlWL1QIDYedO2LQJ1q2DJ0/MdScnMyxfsyZUrw4VK+rYHhGRJEjBVkREHMqDBzB5 Mnz5pdnwFuDtt03ILVcubn0FRwQz+eBkphyaQmhUKE440bVkV76o+wUZUmawffESew8ewIkT5rF7 txmZ/eMvGAAvv2x2A+vcWdOMRUSSAQVbERFxCE+ewIwZMHWqGbwDqFwZxo+HGjXi1leMJYblp5cz zGcYt56YNZlvvvQm0+tOp3TO0rYtXGLv1i347DPYvh2uXfvr64UKQaNG0LQpVK2q3Y1FRJIRBVsR EUnWHjyAb78FT89nOx2XKAETJkD9+mYmalwc8jvEgG0DOH7rOAAFMhRgyltTaFGkhXY8tpfQUJg2 DSZONOtn/1CwIJQpAxUqQMOGJtiKiEiypGArIiLJTlAQfP+9OaFl+3aIijLXCxUyR5G+807cB+vO 3D3D2P1jWe27GoB0bukYUW0EH1b8kJSuKW38E0isREXB6tXmbNnr1821SpVg1CizVjaDpoOLiDgK BVsREUkWwsNh2zYTZjdufLZ+FqBkSejbF7p0Adc4/j/fsZvHGP/jeDZe3AiAE050L92dsTXHkj1t dhv+BBJrFy7A4sWwZAncuWOu5c1rDhxu2zbuw/AiIpLkKdiKiEiSFR0Ne/eCt7fZ6PbRo2evFSwI 7dubnFMkjkfIWiwW9l/bz7gfx7Hryi7ABNpWxVoxstpI3sj+hu1+CImdJ09g1SpzRM+hQ8+uZ80K /frBRx/paB4REQemYCsiIkmOxWIG64YPN8f2/CFXLhNk27UzSyvjOnBnsVjY9ts2xv84noN+BwFw cXKhU4lODKsyjMJZCtvwp5D/ZLHAgQMmzK5aBSEh5rqLCzRoAF27mrWzbjonWETE0SnYiohIknLr Frz/Pvzwg2lnymTWzLZrZ44jdXGJe5/RMdGsv7CeiQcm8vPtnwFwd3HnvVLvMaTKEPJnyG+7H0D+ 3fXrsGePeezeDX5+z14rXNgc0dOpE+TMab8aRUQk0VGwFRGRJMFigWXLoH9/M+XYzc2c7DJokPUD dmFRYSw9tZQph6ZwKeASAKlTpKZXmV58VPkjcqXLZcOfwIFZLBAQYELq9evmcfMmPH5sphgHBpqv 167BlSvP35s2LbRpYwJtpUpaPysiIn9LwVZERBK9O3egZ0+zKRSYacZeXlCsmHX9PQ57zFc/fcWM ozO4E2Q2H8qQMgN9yvVhQMUBZEmdxUaVOwiLxYTSn36C06fN/PB79/6vvXuPy7q8/zj+vjkqIiAq eAI84gkEAw+ZZ80Dq9TNbGtLXZsr1+yxXG21lq1trW3VbPvl3Dppumyug1ppaVGpeUZFOaigYuIJ xAMIKKf7/v1xyX17Jy6gG25ufT0fj+8D7u/x+mJ9H7z5fK/rkvLyHMuVo3n9L97eUmKiNGqUWYYO pe8sAOBrEWwBAE2WzWZGOZ4zxxT8fH2l3/5W+uUv6z66sSSdvHBSL2x9Qf/c+U8VlRVJkjoFddLc wXP145t+rJb+LV17A9cLq9WUyQsKnJfDh02YTUkxEwZ/nbAwKTLSjGDcqZPUqpUUFCS1bGmWtm2l gQPNOgAA6oBgCwBokvLypNmzpRUrzOebbpIWL5Zi6zEgcdaZLD276Vkt2btE5VXlkqQ+bfvol0N+ qe/Ffk9+3gw+JJtN2rrVhNTcXLMcO2a+Hj/umAz4Wnx8pH79zD9UZKQJseHhjq8dOkjNmzfOvQAA bjgEWwBAk7N8ufTAA6YI6OMjzZsnPfqoqdjWxemS0/rNp7/Ry7telk02SdKQiCH61S2/0m3Rt8nL 4tUArfcw+/ZJb7xhliNH/ve+QUFSmzaOpUMHE2QTE81fHJo1a5QmAwDwVQRbAECTkZEh/eY30sqV 5nN8vKnSxsXV7TwVVRVasGOBfvv5b1VYVihJui36Nv3qll9paORQ1zbaE335pfTOOybM7trlWN+y pTRmjNS5s3lVOCLC8dpweDjT6gAAmiyCLQDA7TIypN//3kxVarOZKu3jj0u//nXds9Tag2v187U/ 1/6C/ZKk/u36628T/qZhUcMaoOUewGaTsrOlDRvMsn69GZW4mo+PNHGi9P3vS7ffzkBNAACPRLAF ALhNRob0u99Jb71l8pckffvbZoCouvalzT6Trbnr5uqDrA8kSW0D2uqPY/6oH8b/UN5e9Zjc1hNV VkoHD0rp6WZJS5M2bzbDSl/J29tMnXP33dKdd5rXigEA8GAEWwBAo0tPNxXaKwPtd75j+tL261e3 cxWVFekPG/6gF7a+oAprhXy8fPTgwAf1xIgnFNIsxPWNb0pycqR166SNG80Pdd8+qbz86v38/KRB g6QRI6Thw02oDQxs/PYCANBACLb1dPLkST366KPatWuXQkJC9N5776lVq1Z69tlntXTpUlVUVOjN N99UfHy8JOnIkSOaNWuW9u/fr6lTp2r+/Pn2c9XnGADwROnpjgpttfoGWqvNqtdTX9djyY8pryRP kjSh+wTNHz9fvdr0cmGrm5CiIumzz0yYXbfOVGe/qkULqU8fKSbGLAkJJtQysBMA4DpGsK2nSZMm adGiRerdu7fGjx+vxYsXq1+/fjpy5Ij27t2rhQsX6t5779Wuy4Ny3HfffXrllVfUqlUrJSQkaNSo UbrjjjuUnJxc52MAwNOkpZlA+/bbjnVTp0pPPFH3QCtJyYeT9fDHDyv1VKokKbp1tOaPn6+kHkku arEbVFVJqanSyZNXzxdbUCCdOGEGeqqqchzj4yMNGWIGfOrf3wTZqCjJi9GeAQA3FoLbaw0XAAAg AElEQVRtPZw7d06TJk1S3759JUkJCQm6ePGiFixYoEcffVSSNGvWLP3617/W/v37VV5eLm9vb0VF RUkygXXZsmW644476nUMAHiKawXaefPqNx9t5ulMPfLxI1qTvUaSFOQfpCeGP6EHBz3omXPR2mzS jh3SsmVmjqOv9oWtSXS0NG6cWUaONCMZAwBwgyPY1kOrVq30+OOP2z+vX79eixYt0sKFCxUZGSlJ 8vHx0U033aTMzEydO3fOvl6SBg0apCVLlkiSduzYUedjalJWVqaysjL756KiItfcLADUw969JtC+ 845j3Z13mgptfQJtXnGenvz8Sb2862VZbVb5ePloduJszRsxT20CPHDgo8xM6c03zXLokGN9cLDU o4fzXLFXLjfdJHXp4r52AwDQRBFsv6Hly5dr0KBB6tWrl/Lz8+V3xbwUYWFhOnPmjAoKCmpcL6le x9TkmWee0VNPPeXKWwOAOvtqoLVYHIE2Jqbu5yutKNVft/xVf970ZxWXF0uSpvSaoj+N/ZOiW0e7 sOUNyGYzrxGnpJjl/felPXsc2wMCpEmTzAjF48YxVywAAPVAsP0Gjh07pn/+859as8a8EhcSEqKC ggKFhoba9wkPD1dlZaXS0tKcjg0PD6/3MTV57LHHNHfuXPvnoqIiRURE1P/mAKAO9uwxgfbdd83n bxpoq6xVWrp3qX7z6W90/MJxSdKADgP0/Ljnm/58tCdOSDt3mhBb/TUvz3kfX19pwgQTZm+/3Qz4 BAAA6o1gW08VFRWaNWuW/vWvf6l58+ayWq3q06ePUlJSFB1tqgg5OTnq37+/goODnUY0PnTokBIS EiSpXsfUxN/fX/7+/g1xqwBwTTUF2mnTTKC9PAxBnX1y+BM98vEj9oGhooKj9Kexf9K0vtPkZWlC gyJdumSm19m717GkpV0dYiUzb2yfPlJionTLLdKUKdIVf9AEAADfDMG2nh566CH9+Mc/tgfSRYsW afr06XrxxRc1bdo0ZWVlKSIiQhEREerYsaMqKyu1evVqfetb39Kbb76phx56SJLqdQwAuFtqqgm0 K1aYzxaLdNddJtD26VO/c249tlWPf/q4Ps35VJIU7B+sx4c9rjmD5qiZj5unqrHZpAMHpDVrpO3b TYA9cMB5hOJqXl7mh5CQYIJsQoIUF2deOQYAAA3CYrPZbO5uhKfZtm2bRowYoZjL79eVlZVpzJgx euGFF/TAAw9ozZo1iouL0+uvv67g4GBJ0u7duzVt2jQFBwdr7ty5uvvuu+3nq88xX6eoqEjBwcEq LCxUUFCQC+8ewI1szx7pqadcG2jT8tL0xGdPaNWBVZIkXy9fzU6crSdGPOHegaEuXZI+/1xavdoE 2sOHr94nNNTMV1S9xMaaUjWvFgPADYffv92LYHud4n8sAK6Unm4CbfW0PRaL9N3vmkDbu3f9znno 7CE9+fmTWpa2TDbZ5GXx0oy4GXpyxJOKColyXePrIjfXhNjVq6XkZKm01LHNz89MrzN6tKnA9usn tW9vfhgAgBsev3+7F68iAwCuad8+E2j/+1/zNm51hXbevPoH2uNFx/X7Db/Xq7tfVaW1UpJ0Z587 9btRv1OvNr1c2PpaOHlS2rBBWr/eLJmZzts7dpS+9S2zjB4tBQY2bvsAAECtEGwBAFfJyjJ9aJct M4FWMqMcP/lk/QeFKigt0J+/+LNe3PGiLlVekiRN6D5BT49+Wje1v8lFLf8aubmOELt+vZSd7bzd y0saPNgRZvv1oyILAIAHINgCAOwOHZJ+/3tp6VLJajXrpkyRfvtbk/Hqo6isSPO3zNfzW57XhfIL kqShkUP19OinNTxquGsaXpOLF81IxSkp0o4dpjKbk+O8j8ViXiseMcIsw4dLrVs3XJsAAECDINgC AJSTI/3hD9LrrzsG+r39dhNob6pnMfXkhZP6+7a/a2HKQhWWFUqS4tvF64+j/6gJ3SfI4spK6KVL JsReOX9sevrVoxZ7eZkbqg6yQ4dKrVq5rh0AAMAtCLYAcAM7elR6+mnptdekStPdVUlJJtAOGFC/ cx4oOKDnNj+nJXuXqLyqXJLUq00vPTXyKU3tM7V+c9FWVUn5+dKJE6ZfbPXXo0elXbtMiK2+gSu1 beuYcmfIEDOHLAN6AABw3SHYAsANKDtbmj9feuUVqaLCrBs3zgwUNXhw/c65JXeL/rL5L1q1f5Vs Mh1zb4m4Rb+85Ze6Lfq2rw+0FRWm0rpxo3TwoHOIzctzvBt9LW3aOM8dm5godepEH1kAAG4ABFsA uEFUVkrvvy/94x/SJ5841o8ebQLt0KF1P6fVZtXqrNX6y+a/6IujX9jXT+o5SY8MeUS3RN5y7YNL S6Vt20zf1w0bpK1bnafX+SovLyk83Eyx06GD42u/fibIRkYSYgEAuEERbAHgOnfypPTyy9JLL0nH j5t1Fot55fjhh83UrHVVVlmmZWnL9OzmZ7WvYJ8kydfLV9PjpusXN/9CvdvWMBdQYaG0aZMjyKak OMrF1Vq3loYNMwM6dejgHGDDwiRv77o3FgAAXPcItgBwHbLZpM8/N9XZlSsd3U/btpV+9CPpvvuk zp3rft6C0gIt2r1IL2x7QScunJAkBfkHaXbibD046EF1aNnBsbPVakYjXrFCWrtW2rPHMXdQtY4d zUjEw4ebQNu7t6nMAgAA1AHBFgCuI+fPS0uWSAsXSvv3O9YPHSrNni195zuSv3/dzlleVa4Psz/U 4j2LtTprtSqspsraoWUHPTT4If0k4ScK8r88IFNFhUnUK1ZIq1aZ/rFX6t7dEWSHDzfpmteHAQDA N0SwBYDrwK5dJswuW+bophoYKN1zjwm0sbF1O5/NZtPuU7v1eurrWpa+TAWlBfZtN7W/ST8b8DN9 v9/35eftJ5WUSB+8Y0rDH3xg0nW1li3NO8+TJpl3ntu3/+Y3CwAA8BUEWwDwUJcuSf/9r3ndeNs2 x/qYGBNmf/CDus9sc6r4lN7Y+4Ze3/O60vLT7OvbBbbTD2J/oBnxMxQTFiOdOSMtXWYqs+vWmcZU CwszQXbKFDMyVV1LxAAAAHVEsAUAD3P0qKnOvvyyyZeS5OsrTZ1qAu3QoXV7u/dS5SW9f+B9Ld6z WGsPrlWVrUqS5O/tr0m9JmlG3AyN6zZOPgVnpXfekd560Az+VFXlOEmXLibITpki3XwzgzwBAIBG RbAFAA9QPRjUiy+aN36rp3SNiDBh9t57zUw4tT+fTduPb9fi1MX6T8Z/dP6S4/XhwZ0Ga2bcTE3r O02tSq3Su+9KsydIn33mPJdsXJwjzMbG0lcWAAC4DcEWAJqw4mLp3/82gTYjw7F+9GjpZz+Tbr9d 8qnDk7ykvETL0pbpHyn/UOqpVPv6TkGdNL3fdE2Pm66ePuEmPT9xt5nwtnpIZUlKTJTuukv69rel rl1dcIcAAADfHMEWAJqggwelBQukRYvM9K+SFBAgTZ9uAm3fvnU7X/aZbP1jxz+0KHWRCsvMCZv5 NNPUPlM1I26GRrVOlPcHq6W/PWKm5ikvdxwcH2/C7J13St26uegOAQAAXIdgCwBNhNVqMuX//Z/0 4YeO9d27Sw88IM2cKYWE1P58VdYqrc5erQU7FmjdoXX29V1bddVPE3+qH0ZPU2jyZumRhdKaNc4D QPXta8LstGlSz57f/OYAAAAaEMEWANyssNBUZhcsMJXaaklJpjo7frzk5VX78xWUFujVXa9qYcpC fVn4pSTJIouSeiRpTuyPdWtWpbz+9pb0wTzH3ECSFB1twuxdd9W9JAwAAOBGBFsAcJOMDNN3dulS MxWsJAUHm4GgfvpTU6mti+3Ht2vBjgVanr5cZVVlkqTQ5qGa1We6fn6hj9qtXi/dd4/puFutSxdH mI2LYwAoAADgkQi2ANCIioult9+WXntN2rjRsb5vX2nOHOn735cCA2t/vosVF7U8Y7kW7FiglBMp 9vUDw27S76tGaPSO0/L54xUddSUzlPK0aSbMJiYSZgEAgMcj2AJAA7PZpC1bTJhdvtxRMPXykiZP NoF2xIi65cuM/Ay9uvtVLdmzRGcumslsm8tX8zRC92a1UNv/+0KWM/MdB7RvbwZ/uusuafDgur3b DAAA0MQRbAGggZw8aV4zfu016cABx/ru3c3rxtOnSx071v58xeXF+m/Gf/XKrle05dgWSZLFKk09 G6ZHj3dV/y8Oyuv0J44DwsKkqVNNmL3lFsnb20V3BgAA0LQQbAHAhSoqpNWrTZhds0aqqjLrAwLM 27/33isNHVr76qzNZtP249v1yq5X9J+M/8haXKybTkoPH7fojgsdNODgRTU7nS8p3xwQGip95zsm zI4YUbdJbgEAADwUv/EAgAtkZpowu3SplJ/vWD9kiAmz06ZJLVvW/nxnSs/ojV2va/2H/1Bo2iEN PC797LgUky952yTJJum42Tk4WJoyxYTZMWMkX18X3hkAAEDTR7AFgHoqLDR9Zl97Tdq2zbE+PFya MUP64Q+lXr1qeTKbTdacw8r4YJG+/PgthaRl68fHbXqwsoZ927eXBg2SBg40y9Chkr+/K24JAADA IxFsAaAOrFZpwwYTZt9+W7p40az38ZFuu81UZydMqEXR9NQpafduaft2Xdy8QdZtW9WisFSxkmKv 2K2sRTN5DRgo38FDHEG2Lh1zAQAAbgAEWwCohdxc6fXXpUWLpMOHHet795Z+9CPpBz8wldqrVFaa kaNSU6U9exxLXp59l+aXv5Z7SWkdvFUc11tR4+5U53HT5B8dzQjGAAAAX4NgCwDXUFYmrVplqrPr 1plpeyTTV/Z73zPV2YEDrxgI6vx55/CamiplZJgTfYXVImWFSts7mqUysb+G3zFHk+PuUoBvQOPd JAAAwHWAYAsAX5GaasLsG29IZ8861o8cacLsd6ZYFZCXY8Lrk1dUYr/8ssbz2QIDdb5npLaEluq9 Zl9qd7hN6WFSy1bhmhk/Uw/2v1fRraMb5+YAAACuQwRbAJAJsMuWmUC7e7djfe8OhXpobJomd0tT 21Np0j/3SD/dKxUX13yiqCgpLk6Ki1Nhry5a7rNffznxlg4VZtp3GdV5lBYl3q/JvSbLz9uvge8M AADg+kewBXDDqqqSkpNNmP3g3XJ1qTigWKXpu15pGhOepj5VaWp+4qi0pIaD/f2lvn2l+Hh7kFW/ frKFhGhT7iYtTFmotzP/rPKqcklSsH+wZsbP1P2J96tXm9oOlQwAAIDaINgCuLHYbDq6KVcbFqTp 2Idpiizcq8eVpiU6ID9VmH2skk5ecUynTlJsrFn69TNhtmdPMxTyZUVlRVq6Z6n+ufOfSs9Pt68f 0GGA7k+8X9+N+S59ZwEAABoIwRbA9ev8eSktTUpLU8XuNJ3bkKaAw+mKrCzUD2raPyhIiokx4bU6 yMbESK1aXfMSqadStXDHQr2R9oZKKkokSc19muvu2Ls1O3G2EjokNMy9AQAAwI5gC+D6YLVKu3aZ 4Ys3bTKBNjfXvtlXUtjl7yvko2OBveTVL1YdJsTKt//lEBsZecUQx9d2seKi3sp8SwtTFmrrsa32 9b3b9NbsxNm6J+4ehTQLcfENAgAA4FoItgA81/Hj0scfS2vXmq9nzly1y1FFKE2xSlOs8trGqte0 fpr4857q0r3ugzYdKDigl3a+pMV7FuvsRTNcsq+Xr77d+9uanThbw6OGy1KLYAwAAADXItgC8BwX L0obN5ogu26dlJ7utLnMv6V2Bo3Wf8+MUYq1v9IVo7JmIZo61UzTM2KE5OVV+8tdqrykDV9u0IfZ H2rNwTXKOpNl3xYVHKWfJPxEP+r/I4UHhrvqDgEAAFAPBFsATZfNZsLrunVm2bBBunTJsdliUU7r RL1XNl7vXBinrWWDVXnaV5I0cKD053ulu+6SQurwVvCX57/Uhwc/1JrsNUrOSVZpRal9m4+Xj8Z3 G6/ZibM1ofsEeXt5u+xWAQAAUH8EWwBNy+nT0iefOKqyJ086bS5q2VHrm43XsoJxWmcbq7MFrSVJ zZpJY0dK48dLEyeaQYtro6KqQptyN2lN9hqtyV6jjNMZTtvbB7ZXUo8kJfVI0tiuYxXkH+SKuwQA AIALEWwBuFd5ubRliyPI7tplKrWXVfo1V1roCL1VOF4rL47Tvgu9pQumH2ufPtLMCSbMDhsmNW9e u0ueuHBCHx38SGuy1+jjwx+rqKzIvs3L4qWbO91sD7Nx4XH0mwUAAGjiCLYAGpfNJmVnmxC7dq30 2WdSSYnTLifa9tNHNlOV/aJ8qMpONZNkXimeOlaaMEEaN06KiKjdJSutldp2bJupyh5co9RTqU7b 2wa01cQeEzWx+0SN6zZOoc1DXXKrAAAAaBwEWwAN7+JFM2rx6tUm0B454rS5NLCttgWN0xunx2l1 xa06dbq9JDPzzoABJsiOH2/6zfrU8qmVX5KvtQfXas3BNVp7cK3OXTpn32aRRQM6DlBSd1OVTeiQ IC9LHUaVAgAAQJNCsAXQMM6eNUF2xQpTmS11DMJk9fFVVthQrbw4XsvPjdOe4jjZik2wbN9emjne hNmxY6XWrWt3OavNqpQTKVqTvUYfHvxQO47vkE2OV5pbNWul8d3HK6l7ksZ3H6+wFmH/42wAAADw JARbAK6TmyutXGmW9eulqir7psKQSCUH3KHFpybo08oRKjkRKEny85NGDzMV2fHjpdhYU6mtjbMX z2rdoXVak71GHx38SKdLTztt79+uv72v7MCOA+XjxSMPAADgesRveQDqz2aTMjIcYXbnTqfNx1rF 6u3KyVpyYbJ2n+8vnTeJtUcPx+vFI0dKLVrU9nI27cnbYx/BeMuxLbLarPbtQf5BurXrrUrqkaQJ 3SeoQ8sOrrpTAAAANGEEWwB1U1Ulbd1qguyKFdKhQ/ZNNotF6UG3aMmFKXrXOkmHz3WTZEYrvm2M mYZnwgSpa9faX67wUqE+OfyJ/RXjk8XO0//EhMXY+8oOiRgiX29fl9wmAAAAPAfBFsDXu3RJSk42 Yfa996T8fPumCm9/bfC7VcsuTtb7ttt1utD0Xe3WTXrwW1JSkjRihJlntjZsNpsyT2faRzD+4ugX qrRW2rcH+AZobNexSuqepIk9JioyONKltwoAAADPQ7AFULPz56U1a0yY/fBDqbjYvqnYJ1jvW2/T 29YpWls1XiUXA+XnZ14rTkoyS48etb9USXmJPs351B5mjxYeddres3VPJfVI0sTuEzU8arj8ffxd dJMAAAC4HhBsATgcP24qsitXSp9+KlU6KqWnfDrq7crJWqnJWl85QpXyVWSkdM/lIDt6dN36ymaf zbb3lV3/5XqVV5XbtzfzaaZRnUfZw2y30G6uvlMAAABcRwi2wI1u/37TV3blSmn7dqdN+yx99K7N hNmUykT5+Fg0dKj0x8thtk+f2o9gfLHiotZ/ud4eZg+dO+S0vUtIF/sIxiM7j1SAb4Cr7hAAAADX OYItcKOxWk2ArR7J+MABxyZZtEU3a6Uma5UmKdsWrXbtTIh9NMnMKxscXLvLVForteP4DiXnJCs5 J1mbczc7VWV9vXw1ovMI+8BP0a2jZaltSgYAAACuQLAFbgSFhdKmTdL770urVkknHSMLl8lPyRqj lZqs93SHTnu10+DB0ozLVdm4OMnL6+svYbPZlHE6Q8mHTZBd/+V6FZUVOe3TKaiTPciO7jJaLf1b uvpOAQAAcAMi2ALXG5tNys6WNm+WtmyRNm+WLSNDFpvNvkuRWmq1vqUVmqKPNEF+rYM0YYI0/1vS uHFS69a1u9SR80fsQfbTnE+VV5LntD20eahGdR6lMV3GaEzXMeoR2oOqLAAAAFyOYAt4upISKSXF BNnqMHvmjNMuFkmH1FXrNE4rNVmfa6RiE/yVlCTNTZIGDJC8vb/+UqdLTuuzI58p+XCyPsn5RIfP HXba3tynuYZFDTNBtssYxbeLl7dXLU4MAAAAfAMEW8CT2GzS0aNO1VilpkpVVU67XZK/dmiANmuI NmuItmqwKkPDNWaM9L0k6fUJUrt2X3+54vJibfhyg70quydvj9N2b4u3BnUaZA+ygzsNZioeAAAA NDqCLdCUlZVJu3Y5QuyWLdKJE1ftdkwd7SF2s4YoVfFqF+GnYcOkicOkPw6Tevf++r6y5VXl2nps qz3Ibju+TZXWSqd9+oX3swfZ4VHD6ScLAAAAtyPYAk3JyZPOITYlRSovd9qlQj7arf7arCHaopu1 WUN0TBHq1UsaNkyaM0waPlyKivr6y1ltVu05tUefHP5EyTnJ2nh0o0orSp326RLSRWO6jNHYrmM1 qssohbUIc+UdAwAAAN8YwRZwl8pKae9eR5DdvFk6cuSq3U5b2mqz7WZ7NXanElTuHaD+/U2Q/fsw aehQqW3br7+kzWbTwbMH7VPwfJbzmc5cdO6PG9YiTKO7jLZXZbu06uKiGwYAAAAaBsEWaCxnzkhb tzqqsdu2SaXO1VGrLMrwitUXVkc19pCtm5o1s2jQIGnkMGnecGnwYKnl17wBbLPZlFeSp7S8NKXl p2lP3h59lvOZcotynfYL9AvUyM4j7UE2JiyGkYsBAADgUQi2QEOwWqV9+5yrsQcOXLXbBe9gbbYO 1iabqcZu10BdsAYpJES65RZp1jBTlU1IkPz/x5hMxeXFysjPUFp+mj3IpuWnqaC04Kp9/bz9dHOn m+1T8AzoMEC+3r6uvHsAAACgURFsAVcoKpK2b3eE2K1bpcLCq3bL8e+pz8uGaPPlauy+qt6yyUvt 25sA+8xw8zUmpuaBniqtlco6k+UUXtPy0pRzPqfGZnlZvNQ9tLtiw2IVGxarmyNu1tDIoQrwDXD1 TwAAAABwG4ItUFs2mxncKTvbsWRlmWXfPrP9CmU+AdrtM1DJlxxT7pwtay1J6tHDBNhHLldku3aV rnz712az6VjR8asC7L6CfSqvch5Mqlq7wHb2ABsbbr72adtHzX2bN9iPBAAAAGgKCLbAlWw26fRp 5/BaHWAPHpRKSq55aH5AZ220DtFnl0w1dm9lP1VV+shikeLipB9crsYOHeo8h2zhpUJtzk2/6jXi 85fO13idFr4tFBMW4xRgY8Nj1Sagjat/GgAAAIBHINjixnTunHNovTLE1vAKcTWrxUv5AZ112KuH 0sqilV7eQ9nqob3qp5OlHSRJfn7SwIHSLy9XY4cMkYKDzRyx+wv2KzkvTWnpjirsVwdzquZt8VbP Nj0VGxbrFGQ7h3SWl+VrJqQFAAAAbiAEW1y/LlyoufKanW1GKL4Gqyw67R+hbEsPpV/qoSz1UJai la0eyrF1UUWJn9P+7dqZiuzPLldkExNtyiv7Uml5adqdn6Yln5gAe+DMAVVaK2u8ZqegTle9Rtyr TS/5+/yPEaMAAAAASCLYwtOVlppXhL8aYLOzpVOn/ueh+T7tdcAWrf1VpupavRxSN10qc/RLbd7c 9IHt2VWa2NV8X70EhhUop9gxGvEHWWlK/yJdF8ov1HjNIP+gqwJsTFiMWjVv5dIfCwAAAHAjIdii 6Ssrkw4frvnV4WPH/uehBV5tdcBqAmt11TVbPXRQ3VVSGWjfr2NHE1QTu0rTvhJew8OlC+VFyjyd qfT8dKXnp+u9/HSl709XXklejdf19fJVrza9HH1gLwfZiKAI5ogFAAAAXIxgi6ahokI6cqTmV4eP HjXzwl7DOYU4hdbqEHtQ3VVoDZEkBQQ4guqortKPukrdupnPUVGmKitJFysuan/BfqXnp2tVfrrS PzVB9mjh0Wtev3NIZ0cf2MsBtmfrnswNCwAAADQSgi0aT1WVlJtb44BNtpwcWSpr7n8qSRcUeFXV tXo5o9ayWCz2qmvXrtKkr1Rdw8Kcp9OpqKpQ9tls7c1P17Kt6fZK7KFzh2S11Ryi2we2V0xYjNPS p20fBfoF1rg/AAAAgMZBsEXDWbRIysy0h1jboUOylNc8B6tFUqma66C6O1Vdq7/PU7gCAy1OYXXQ Fd9HRUnNml19XqvNqpxzOdp64HJ4PW2+Hig4oAprRY1tCW0eaoJrW0eA7RvWV6HNQ134wwEAAADg KgRbNJgzv/yTWhdk2T9bJJXJT4fU7aqqa7Z66IQ6qlOklz2sdusq3XpFeG3TxrnqeiWbzaZjRcft ldfqJfN0pi5WXqzxmEC/QPVt2/eqKmx4i3D6wQIAAAAehGCLBvOm5W5V6axT5fVcYKS6dPe2h9X4 rtK3L38fGSn512J2m9Mlp50D7Ol0ZeRnqLCs5vln/b391btt76uqsBHBEcwHCwAAAFwHCLZoMLYn nlRpsTS8qzTzcngNDb121fWrCi8VKuN0hlOIzTidofyS/Br397Z4K7p19FUV2K6tusrHi//UAQAA gOsVv+2jwcyZU7v9SitKte/0vqsqsLlFuTXub5FFXVt1Vd+wvk4V2OjW0fL3qUXJFwAAAMB1hWCL RlNeVa6sM1lX9YM9fO6wbLLVeEynoE5X9YPt3aa3Wvi1aOTWAwAAAGiqCLZoMKv2r9LevL32kYiz zmSp0lrzlD5tAtrUOBJxSLOQRm41AAAAAE9DsEWDeWjtQ8o5n+O0Lsg/qMaRiMNahLmplQAAAAA8 HcEWDWZKrykquFjgVIXtFNSJqXQAAAAAuBTBFg3m+fHPu7sJAAAAAG4ATOIJAAAAAPBoBFsAAAAA gEcj2AIAAAAAPBrBFgAAAADg0Qi2AAAAAACPRrAFAAAAAHg0gi0AAAAAwKMRbAEAAAAAHo1gCwAA AADwaARbAAAAAIBHI9gCAAAAADwawRYAAAAA4NEItgAAAAAAj0awBQAAAAB4NIItAAAAAMCjEWwB AAAAAB6NYNvEHTlyRLfeeqsiIiL00EMPubs5AAAAANDkEGybuPvuu0+vvPKKMjIy9MEHH+i9995z d5MAAAAAoEkh2DZhe/fulbe3t6KiohQUFKT77rtPy5Ytq3HfsrIyFRUVOS0AAIIxnHoAAAmhSURB VAAAcCPwcXcDcG07duxQZGSk/fOgQYO0ZMmSGvd95pln9NRTT121noALAAAANLzq37ttNpubW3Jj Itg2Yfn5+fLz87N/DgsL05kzZ2rc97HHHtPcuXPtn//yl7/o6aefVkRERIO3EwAAAIBx4cIFBQcH u7sZNxyCbRMWEhKitLQ0p3Xh4eE17uvv7y9/f3/757ffflu5ublq2bKlLBZLva4/YMAA7dixo17H FhUVKSIiQrm5uQoKCqrXOeB63+Tf1NN4yr26u52Nef2Guparz+uK89X3HDw7myZ3/3/amDzlXptC OxurDQ15HVee253PTqlpPD9tNpsuXLigDh06uOX6NzqCbRPWp08fzZ8/3/750KFDSkhIqNWxXl5e 6tSp0ze6vre39zd+MAQFBfHLWRPiin9TT+Ep9+rudjbm9RvqWq4+ryvO903PwbOzaXH3/6eNyVPu tSm0s7Ha0JDXceW5m8KzU3L/85NKrfsweFQTNmzYMFVWVmr16tWSpDfffFOzZ8+u1bEPPPDAN76+ K86BpuVG+jf1lHt1dzsb8/oNdS1Xn5fnJ77qRvr39JR7bQrtbKw2NOR1XHlunp1wN4uN3s1N2u7d uzVt2jQFBwdr7ty5uvvuu93dpFopKipScHCwCgsL3f4XVQDwFDw7AaB+eH6CV5GbuP79+ys7O9vd zagzf39/Pfnkk079fgEA/xvPTgCoH56foGILAAAAAPBo9LEFAAAAAHg0gi0AAAAAwKMRbAEAAAAA Ho1gCwAAAADwaARbNKrc3FyNHDlSffr0Ub9+/fTWW2+5u0kA4BHOnz+vxMRExcfHKyYmRi+//LK7 mwQAHqO0tFRRUVF6+OGH3d0UNBBGRUajOnnypPLy8hQfH69Tp04pISFBWVlZatGihbubBgBNWlVV lcrKyhQQEKCSkhLFxMQoJSVFrVu3dnfTAKDJe/zxx5Wdna3IyEg999xz7m4OGgAVWzSq9u3bKz4+ XpLUrl07tWnTRmfPnnVzqwCg6fP29lZAQIAkqaysTDabTfxtGgC+XnZ2tvbv36+kpCR3NwUNiGCL OtmwYYNuv/12dejQQRaLRStXrrxqnwULFqhz585q1qyZBg0apO3bt9d4rp07d6qqqkoREREN3WwA cDtXPD/Pnz+vuLg4derUSY888ojatGnTWM0HALdwxbPz4Ycf1jPPPNNYTYabEGxRJyUlJYqLi9OL L75Y4/bly5dr7ty5evLJJ7Vr1y7FxcVp/Pjxys/Pd9rv7Nmzmj59ul566aXGaDYAuJ0rnp8hISHa s2ePcnJytGzZMuXl5TVW8wHALb7ps3PVqlWKjo5WdHR0YzYbbkAfW9SbxWLRihUrNHnyZPu6QYMG acCAAfaHj9VqVUREhObMmaNHH31UknmF7tZbb9WsWbN0zz33uKXtAOBO9X1+Xmn27NkaM2aMpk6d 2mjtBgB3qs+z87HHHtO///1veXt7q7i4WBUVFfrFL36hefPmues20ECo2MJlysvLtXPnTo0dO9a+ zsvLS2PHjtWWLVskSTabTTNnztTo0aMJtQBwWW2en6dOndKFCxckSYWFhdq4caN69uzplvYCQFNQ m2fnM888o9zcXB05ckTPPfecZs2aRai9ThFs4TIFBQWqqqpSeHi40/rw8HCdOnVKkrRp0yYtX75c K1euVHx8vOLj45WWluaO5gJAk1Gb5+fRo0c1bNgwxcXFadiwYZozZ45iY2Pd0VwAaBJq8+zEjcPH 3Q3A9c9ms8lisUiShg4dKqvV6uYWAYBnuPL5OXDgQKWmprq5RQDQ9F357LzSzJkzG78xaDRUbOEy bdq0kbe391WDmeTn51/1lzQAgAPPTwCoO56duBLBFi7j5+enhIQEJScn29dZrVYlJyfr5ptvdmPL AKBp4/kJAHXHsxNX4lVk1ElxcbEOHjxo/5yTk6PU1FSFhoYqMjJSc+fO1fTp05WYmKiBAwfqhRde UElJiX74wx+6sdUA4H48PwGg7nh2oraY7gd18vnnn2vUqFFXrZ8xY4YWL14sSXrxxRf17LPP6tSp U4qPj9ff//53DRo0qJFbCgBNC89PAKg7np2oLYItAAAAAMCj0ccWAAAAAODRCLYAAAAAAI9GsAUA AAAAeDSCLQAAAADAoxFsAQAAAAAejWALAAAAAPBoBFsAAAAAgEcj2AIAAAAAPBrBFgAAAADg0Qi2 AAAAAACPRrAFAKCRlZeXKzo6Wvv373d3UwAAuC74uLsBAABcjyZOnKj8/HxZLBYdOHBA7dq1U3Bw sI4dO6Z//etf6tSpk8LCwtzdTAAArgtUbAEAaAAjRozQzp07lZKSooSEBM2fP18pKSlaunSpwsLC 9Omnnyo0NNTdzQQA4LpAsAUAoAH86le/qnH92LFjtXnzZkVHR6u4uFiFhYX6+c9/ru9973vKzMzU 7bffrsjISH3++ef6+OOPNWLECPXs2VPHjh2zn+O9997T2LFj1bNnT91///2yWq2NdVsAADRJBFsA ABqAxWK55vr77rtP2dnZkqTg4GBNmDBBmZmZ+tvf/qZ33nlHM2bM0Lx587R161atX79eXbt21Ucf fSRJ2rJlizZu3KiPP/5YGRkZSk1N1bvvvtto9wUAQFNEH1sAABpZYGCg0+dmzZrpxIkTeuaZZ+Tn 56fw8HCVlpbq8ccflySFh4fr0qVLkqTnnntOWVlZ2rhxoySpqKiIV5oBADc8gi0AAE1AVFSUU0CN iYmRl9fVL1ZlZmbqpZde0rBhwxqzeQAANGm8igwAgAcJDw9XSkqKu5sBAECTQrAFAMCD3HnnnZo/ f77y8vIkSe+++6727Nnj5lYBAOBeBFsAABrQsmXLlJmZqddee02HDh2SJD3//POSpL/+9a86d+6c Xn31VR0+fFgrV65UTk6OVqxYoa1bt2rr1q3atm2btmzZopUrV+rgwYO6//77deedd2rw4MG69dZb ZbFYFBcX585bBADA7Sw2m83m7kYAAAAAAFBfVGwBAAAAAB6NYAsAAAAA8GgEWwAAAACARyPYAgAA AAA8GsEWAAAAAODRCLYAAAAAAI9GsAUAAAAAeDSCLQAAAADAoxFsAQAAAAAejWALAAAAAPBoBFsA AAAAgEcj2AIAAAAAPBrBFgAAAADg0Qi2AAAAAACPRrAFAAAAAHg0gi0AAAAAwKMRbAEAAAAAHo1g CwAAAADwaARbAAAAAIBHI9gCAAAAADwawRYAAAAA4NEItgAAAAAAj0awBQAAAAB4NIItAAAAAMCj EWwBAAAAAB7t/wHh9q4MM490JgAAAABJRU5ErkJggg== --=-=-=--