4'.deploy.prototxt

1:

神经网络中，我们通过最小化神经网络来训练网络，所以在训练时最后一层是损失函数层（LOSS），

在测试时我们通过准确率来评价该网络的优劣，因此最后一层是准确率层（ACCURACY）。

但是当我们真正要使用训练好的数据时，我们需要的是网络给我们输入结果，对于分类问题，我们需要获得分类结果，如下右图最后一层我们得到

的是概率，我们不需要训练及测试阶段的LOSS，ACCURACY层了。

下图是能过$CAFFE_ROOT/python/draw_net.py绘制$CAFFE_ROOT/models/caffe_reference_caffnet/train_val.prototxt   , $CAFFE_ROOT/models/caffe_reference_caffnet/deploy.prototxt，分别代表训练时与最后使用时的网络结构。

我们一般将train与test放在同一个.prototxt中，需要在data层输入数据的source，

而在使用时.prototxt只需要定义输入图片的大小通道数据参数即可，如下图所示，分别是

$CAFFE_ROOT/models/caffe_reference_caffnet/train_val.prototxt   , $CAFFE_ROOT/models/caffe_reference_caffnet/deploy.prototxt的data层

训练时, solver.prototxt中使用的是rain_val.prototxt

./build/tools/caffe/train -solver ./models/bvlc_reference_caffenet/solver.prototxt

使用上面训练的网络提取特征，使用的网络模型是deploy.prototxt

./build/tools/extract_features.bin models/bvlc_refrence_caffenet.caffemodel models/bvlc_refrence_caffenet/deploy.prototxt

aaarticlea/png;base64,iVBORw0KGgoAAAANSUhEUgAAAfoAAAFWCAYAAACBy1/cAAAABHNCSVQICAgIfAhkiAAAIABJREFUeF7snQdcFMcXx39wdAHpTZoKgopRFDuIxB4LGBN77KISSzRGFE2iCRprrKhYYowNY1RibwHbXyVYYknAqLEiCggq0sv+Zxc4qXd7eCLgGz+XcFPevPnu3r5pO0+FYwEUiAARIAJEgAgQgWpJQLVatooaRQSIABEgAkSACAgEyNDTjUAEiAARIAJEoBoTqLaGPjsH6DrUD+sXTEFaOtCquzfO7l7/Vi/lifN/49PPRsDdwwPtPT0wZsRAJP33T5E6xeR5q0oWEp6SnoOTZ/+H9OcJpVb5LhiWqghFEgEiQASIQLkJqJW7ZCUvqKICqKmpQUtTC+zP/L81S2h9P/YFVq4JQuT/wpGbmQZ7azMM/tQHXXwGs/kO8f2gJ4k5mD5zEro0skHAhEXIkWgi9uljaBuaSusUk6eEgoUiniTmoqu3J3o2tcH3P/4CSPIu39jZK1Dj8QksWRuqkM5Rd5/Df8YUHFoXBC0DkxJVi2VYoiBFEAEiQASIQKUhUG0NvYTZaG1dPZgZm0AiAbR0dGFUzJjdfvQKg4f3Qz1DCWZN9kMNYytE3bqH+KSHChlM/mo+jnsBSdYrTBo9BmYurfIvsGuRCy0mj5g74+Tlu7gV8Qcc23QRk73cecQwLLdwKkgEiAARIAIVQkDtu5UbcfnkUcTHxQoL9va1jDF53Gg09ewuVSA2IRO+E0Yh6dFdYXRsaaoP3yH90bHXIMEgXrv1DDMDJuD5k4eoZVYTrm5eOBx2HAaa2Qic/iVc3LsJsvip4J937ceenVuQnBALEwMdDO3nA5+BvkwOs8b5IScX8J/PRtlHtyLou5lw8ewhTVPkD3trO9jbmrPRPGBibgE9A+MidcxeugSWmsnYtC4UEoO8kXeb1i2LVCGv7SlsWaBjH2+ovUyEBtTRZ9xX0vLfjP0UnQZNhJg8fCExfPhJhmaN2mLO0gXY2tSd9WBqlEAiT05RfTTg4ztZKsPH0wlfBm6QfpfFsETFFEEEiAARIAKVj0CXIeO4of06cRFnz3Cnzpznhk8M4NzdXblX92/yb94JIS2D446eOM7duHaVu/LXDW7G3JUsTwvu6fULQvqxC/e5Nu5NuRvnznABC9eztKbcX2dOcpPnLOf69GrLBKRwubkcN2fVdu5Dr5bcoV1bub+v/82t3rSba+nelvvf3k15FeX/NyWN49r0/JjJcec2LPiySJoiX54kJHEZKS+FIg9in3DZ6anS4rHPcrjGHh7c/p8WyhQpr+18u+49fMTtP3GZa+Xeirt8Yj939+5d9vmPS3mRKMgWm0ceH15n13ZtuL9PneLa9vThTu4KFuSP+XY5N2V0T47LyRHFuWx97nKJ8bFFeMhiKBMcJRIBIkAEiEClICBM3Vub6qFFWzY6ZMHOwRW9Pz2Nv/86jxa29YQ4LQ2gc4eOwt98sK/bEEdO7MXVqxHo5JI3AlZRyUHDhg3gJbHFqf2r0LhRQ/TUqotZJzaBS03G42R1hP66BhsDJsO1Sz9BTn2W//TVa9gWuhNterI18fw1Zx0tYMHsubgecRSD+vXPr1Xx/5kbG0gL2ViYFxHwJOEFNLgcNG7YVKZgeW3n17HtrGshPlkLueyfrZUFjO3ti8gUk+dxQpZ8PvkvSRjqSzB65DQs3fAVPLsOKFKXGDkqjLM8nQuEymIoExwlEgEiQASIQKUgUGKN3khfE6paOohLeCpVMC4xA0uDVuBKxBmksilqI0Nj6GRlst3sqSUaoaqSv4GNncOjmr+ZjQOH2w/joZXLYWLgKoD/FArZRuzMnuwsqaHnk9ybOguftxVUmfUVc1KQIm1/E11F8UHBZkIOn3RpjU2/WOPU4Z1F3pEUJSe/Q/Um+lJZIkAEiAARqBoEShh6Xm0VZqxzObZQzkJWNuA7fQa4mIuYNckflrWdEPssDZMDvii1hfzoVVKKIWHzF7w0rJzzDQzqNChSVkONFdJgw/gKDKZG+kwbVUT9ewM2LTqUWrOibS9ViMhIUXxSXndNajBcI4dPQdBP02Hi+Hq2RZQckTpRNiJABIgAEaj6BEo19IWblZYBPLh9Fd8O/AjunfM26Jlbs71zGtqltt7LzRpe4Rfz0x5L8zjYmCFNRQPRUZHo78U25/E9Ahnh7OWbbOr+CIazqXsto6LT7jKKiU4yNVCFff0W2PDbbrap8DOo6huVKKto20sIUCBCHJ+icxA+HzL9N5og8fIpNK6dN9oXJydPMU11NbbYIEF66isFNKWsRIAIEAEiUJUIyDX0/Bq1pa0zth88DgtHNxgwK5/wIlt451yRUMtUHb0+8UVQyBrEvcyBmzsbRatp4d7D+/D2bI0a5jZScalsJ7v/7ACoJsVBI/khRk5brEhVovKqsU3+30z2x+gxn2DUmGEYMGgEDMysce9xHF7F/4PhI6ewvQmqSmm7GIUU4VMgT08H6Nt3DEJWz2ZReYZeETlWpjXBaepicfBaDBiqhtQcVWSlxKBTt0/FqEx5iAARIAJEoAoQkGvoNdSBoHk/YOHShZg2Zz5yszLYYJyDvYkhLKzriG4iP4D/Zvxg2FkYYt9v27HrUDgry6G2tSm6NmtQxNBrss5Fq1ZeuHh0G1q2bC+6DkUzNq1vhp83hiBozUrMWbQMKrk5MDWsgT49OzPVcqGhrqqUtovRSxE+heX17eaFzT8tkkYpIse4JjBrxlwEr1qIL9nyjJqEwyBvL6BLH4XPERDTRspDBIgAESACFU9Ahd/7X/HVUo1EgAgQASJABIhARRAQf8ZrRWhDdRABIkAEiAARIAJKJUCGXqk4SRgRIAJEgAgQgcpFgAy9jOvhwbzQUSACRIAIEAEiUJUJkKGvylePdCcCRIAIEAEiIIcAGXo5gCiZCBABIkAEiEBVJkCGvipfPdKdCBABIkAEiIAcAmqubB1aNz+TZxMzBK7cLadI9Uv+934SZn8/Cw9uXhPc8C76cgRa+Qyvfg2lFhEBIkAEiMB7R0Bt1+ZfkMPOs5/w7fes8S/eOwB8g+euXgsu8R/8tvknqNc0gr4uO3KOAhEgAkSACBCBakBAzaFObWTnsNNotWuw5rx/hj6H+e5JiI9FRxdHWNRxrAaXlJpABIgAESACROA1AblH4PJZYxMy4TthFJIe3RWmti1N9eE7pD9zBjNIOCr1SWIOuvXugK8+8cDACfzMQF4YEbAIadH7sSPkmOCdju9Q/LxrP/bs3ILkhFiYGOhgaD8f+Az0ZXLY4fMsJL7MxpoNa/FXZATinsQgl3dfi2zM8huMLgM+F/K8SMnBguUrcS78MHKYq1zeG65zHXOsXb4OKqU4p8nTpvT/8ucC8m50JfkudUvPRbFEgAgQASJABKomAVGG3lBfAxNGDUEtMzNk5Urw68FwzFmyCh841IOZS0uYGkhg5+yGsAsXMHBsJqCuAd7z261/ItG3aSPByPMGdd7aHTi1ezmm+k1g+V1x6mI05gf/CDNdbbTxGSYQfJqYjr17d2Jkjw/RpO1oaOvWxIvkZDS0t5ASDly9GVfDQvDd1ACY2TnieXIaYh/fgYp2wW4DBS8GU06FDL2C0Cg7ESACRIAIVAUCogw978Guc4fXPs/t6zbEkRN7cfVqBDoxQy9hI2rvbn3w05JwpD59AB1rBzx4koK0ZzHo0v4zgcPjhCyE/roGGwMmw7VLPyGufsMGOH31GraF7kSbnoOZI/sCdXLR80NPWDdvVyrD2LhYGOlroWWLltCUurBtWmpeeZGxCel4FnMH9Tt9JC8rpRMBIkAEiAARqHIERBn6uMQMLA1agSsRZ5D6MhFGhsbQycpEGps2Lwhd3ZtjxVI9XDhzFB8OcEDk9SjoSdLg4NJCyHL7YTy0cjlMDFwF8J9CIduIDff5KXqpoS+SXOLL1+P9MGHK3+jo7Y3Onu74qFs3NGvdXjr9X6JAGRHjZi/DjT92o7GdHjr2pl32ZWCiaCJABIgAEajCBOQa+iy2I9+XuTDlYi5i1iR/WNZ2QuyzNEwO+KJIs82N1NCkVRds2x+KD/uMxcE/DqBlwzpQNTAR8uU5ycvCyjnfwKBOgyJlNdTYyj+b3hcbnOwMcSBkG8LPReDY0WMYP302Wta3xOJFa6BW01isGHw/aQQutvfEN9+Ox52/zqFua+aelgIRIAJEgAgQgWpEQDgwR5XZWYmaOlIz2Pp6LtuGXijwa+0Pbl/FiN4fwb1zd9R1dECTxo2gqqFdJB+/xD164DD8/fAloi5exJ0b59G/16fSUbaDjRnSVDQQHRUJezs72NvbSz9W1nZgTu4VwqqhroIunq2wZN432Lh5H05FxSJs3xaFZJgY6qODuysM7Rrh5Bm2YZACESACRIAIEIFqRkAY0Qu71h1d8OeB0zh5aB9yahhBNfsZvDr1Br8+b2nrjO0Hj8PC0Q0G5tZIeJGN3My0EijcGljCsE5T+AfOgQGXhA+at5fmqWWqjl6f+CIoZA3iXubAzb0De6dPC/ce3oe3Z2vUMLcpIa+siMUbt8BBTwV1nRpCVV0bEVdvQ5PtnDc2NiurSJnxfP9C6OSkpZSZhxKIABEgAkSACFRVAtKp+ynDB2Laf9cxa8ESqKnmYmCv9vDq4A0NdVUEzfsBC5cuxLQ585GblcEG3xzsTQxhYV2nSLs11IGJoyfj+xkj0bdDM6gavja8vEH9Zvxg2FkYYt9v27HrUDgry6G2tSm6Nmsg2tDz772D7fxfu3UXUpKChfpNjXTx+WAfNOvUp9zXIW9podzFqSARIAJEgAgQgUpJQIUZOLYTTnkh+v4rDB3aCYeDV8PIyVV5gt+SJL7j0Gv0JHxo+QpfBm4sUgvvpvbMmTNvqWYSSwSIABEgAkTg7ROQuxlPjAovU7Lw6P6/7LCbLPywfBn6t28KI8cPxBR953n4VwPt7Rxw/tIuPL4dDU1DU+EIXHXNonsQ3rmipAARIAJEgAgQgXIQUIr3uhu3EzBszFjM9PdjR8ma4wv/BQq/6lYO3ZVWZMbYUdCp5Ya+w0fDx8cHlw6HKE02CSICRIAIEAEi8C4JKH3q/l02huomAkSACBABIkAEihJQyoieoBIBIkAEiAARIAKVkwAZ+sp5XUgrIkAEiAARIAJKIUCGXikYiwrJYKf5Ll/3E6L+DHsL0kkkESACRIAIEAHxBMjQy2HFu9btOtQP6xdMYWf7A626e+Ps7vUyS6WzAwY3btmEp3ejZeYrLTElPQcnz/4P6c8TSksWFVcenUUJpkxEgAgQASJQ5QiQoZdzyfiDftTU1KClqQX+kN68vzXllCp/ctTd5/CfMQUpsffKLaSidS63olSQCBABIkAE3joBpbxH/9a1fIcV8O/Za+vqwczYBBIJoKWjC6N8Rz3vUC2ZVVdFnWU2iBKJABEgAkSg3ASUYuh5PziBQRtx+eRRxDNf8fw0gX0tY0weNxpNPbtLlYtNyITvhFFIenRXGB1bmurDd0h/dOw1SDhw/9qtZ5gZMAHPnzxELbOacHXzwuGw4zDQzEbg9C/h4t5NkMVPTf+8az/27NyC5IRYmBjoYGg/H/gM9C3y/j5/6p3//CBEHt2KoO9mwsWzR7lA2TOnO/a25mw0D5iYW0DPoKiHvFdpuVgUtBZhh3YD7Ihg57oO0C1Wk7y2p7BlgY59vKHG3ABrsH8+vpOlEnw8ndipfRuE7/LkFBSSp3O5QFAhIkAEiAARqHIElGPo2SG65y5egoX6cwTM/wHpnJpgiCfN+hpHtjmihm09AYyhvgYmjBrCjLgZsth59b8eDMecJavwgUM9mLm0xNPEFCQ+uYl1C5dj+9loHNi3Bqt++BGbw65i9qK5+M3NE5ymDuat3YFTu5djqt8E2Dm74tTFaMwP/hFmutpo4zNMehF4Z3znL4RBNVcNEREny23op47oC0NtieBgb9H0KTBiXu8KAt+ZmDpvKaJO78RXY/1gaOOE0xHX8PedO0VuBnlt12GrASHs2ODrN+Mwd7YfVs+eCcO6LoKMmrqvXfjKk1NQqSydiyhGX4gAESACRKB6E+DPun/TkJXNcV2GjONmjv9YKupebDrn6u7ORfz+c5nik5I5ztXrQ+7Y1uVCnmMX7nNtPRpz3Itn3PGI/L+fx3NhkQ+5Nh5NuNxnT7hHcZmcazsP7vKREKnc3FyO6//Ft5zf0K4cl51VpL4zl6K41auXcWms7NsI/8WkCe08sX2VVPzzVxzXiMWFhwSVWWXxthdkjPwngWvh3oJL+CeyzLKFE8qSI6owZSICRIAIEIFqT0ApI/rSukJG+ppQ1dJBXMJTaXJcYgaWBq3AlYgzSGVT1EaGxtDJymS72VNLiFBVyd8nyHzuqPJ+dFng2L/bD+OhlcthYuAqtl7APoVCthGbWshm77ZJXjfLvakz+M/bCvceJ0ALmWjcuIXMKhRpuyxBypIjqw5KIwJEgAgQgepD4K0Zeh6RCjPWuRzvV5YtXWcDvtNngIu5iFmT/GFZ2wmxz9IwOeCLUmkKfuILGeyCTKzrxUvDyjnfwKBOgyJlNdRYIY3X09ylClZypAqvKNtxwPEbFcoIira9DDEKMyxLDsUTASJABIjA+0PgrRr6whjTMoAHt6/i24Efwb1z3gY9c2u2d06jdC9xXm7W8Aq/mC/isVSUg40Z0lQ0EB0Vif5ebHOeYGjLDmcv38T1iCMY3q8/tIzMy85YzpS61qZIhTrOnPsDH7N9BqUFRdquqa6GXEiQnvqqhChF5JQoTBFEgAgQASLwXhKoMEOvpcF22ds6Y/vB47BwdIMBs/IJL7KRm5mmEPhapuro9YkvgkLWIO5lDtzcO7CX27Vw7+F9eHu2Rg1zG6m8VLaT3X92AFST4qCR/BAjpy1WqC4xma3NNNDNZziWblmPV9laqN+sDeJf5LJ986+DIm23Mq3JNhzqYnHwWgwYqobUHFVkpcSgU7dPoYgcMbpTHiJABIgAEaj+BCrM0GuoA0HzfsDCpQsxbc585LLX0FRUONibGMLCuo5o0vwA/pvxg2FnYYh9v23HrkPhrCyH2mxk3bVZgyKGXpNZ21atvHDx6Da0bNledB2KZOT1mTNxBNYZ62H7nh1I2bFX0MfJyhQWtg6CKEXablwTmDVjLoJXLcSXbKlDTcJhkLcX0KUPk6OqFIaKtI/yEgEiQASIQNUmQG5qq/b1I+2JABEgAkSACMgkQEfgysRDiUSACBABIkAEqjYBMvRV+/qR9kSACBABIkAEZBIgQy8TDyUSASJABIgAEajaBMjQV+3rR9oTASJABIgAEZBJgAy9TDyUSASIABEgAkSgahMgQ1+1rx9pTwSIABEgAkRAJgGlGvqU9BycPPs/pD9PkFkpJRIBIkAEiAARIAIVQ0CpB+ZE3X0O/xlTcGhdELQMTCqmBW+xlvjnuRg+bhhePLor1GJpaoDOXu4YPmIsJDXYyTaFAu+utt+EGXh67Sxm+/VHhwGfS1MPnf0PP8wYWqqmEtVknDx0CsiXJ0tOqQIokggQASJABIiADAJKNfQy6qmSSVlZHGKYkf9uxKewaeqF69F3sX7TOpw/fxLrg7dDVc9Q2q67j1Px8J8zaNnUA1t+34MOn4wC1JmTeRY8mtqg3s8/M9c3wOGzN7Fzw9fYuCwIEgNTSFSZkx5tPVFyqiREUpoIEAEiQATeKQGlTN2nsDPlW3f3xmRfH3bGuwZ8fCfDw8ND+CyZxQweC08Sc+Dq2R7bV35dpMEjAhZhwMftgMx08A7g5qzYiI969c4v7w6/kYMQHclGvIVCdg6wIWQ/PurdV8jXu2cXhG5dA+SyhEKBHx1PnRcET3YG/o1TB4qkKfKlrm0tNGncCJ/164WN60LwV0wa9oWsLSLityPH4WqrhwkTvsLfj1Pw7PYNabqejjoc6tZFXfYxsWCefJADh9r2wnf72uyY3Hw3vHwBWXIU0ZnyEgEiQASIABHgCSjF0OuwgWtI8GrmQGYFstm/FbP9sWXLFuEz4otAgbSpgQR2zm4Iu3CBeZnNFOJ4b2y3/olEmyaNBPeyzM08zl+6BIsaqVi+ZDHm/vAjsmvWwegpU/Hor/8JZXgvtfPW7sCOdXMxYZA3gtcEo3OfMZgfvA3n9m0R8hSEDFbN+QthrAOhhoiIk0XSyvvFwUYXTs28EPrHCdYO1gAW0lk9R4/uwuBefVDb3ggmtV1x+OgehatQlhyFK6YCRIAIEAEiUG0JKMXQ845d7KxrwcLKlrlYzYWtlQXs7e2Fj6GJhQBPwmry7tYHNx8kIvXpAyHuwZMUpD2LQZf2XYsAtjLRRdMWLdHOvQWW//A9NKwaYGXwEmHE/jghC6G/rsFi/8no9skgNHBpgLFDP0Ztt47YFrqTDZaZ4/v8oMNc0y+YPRcDB36CQaO+KlJHeb/wba1X1xlPnr2Udlii7yUh69ktNG/bCWpsMaRLRx/sP3UayFDMM5+y5JS3bVSOCBABIkAEqh8BpRh6sVi6ujfHK1U9XDhzVCgSeT0KepI0OLi0KFOENpstcGvhhRt3HwmG8/bDeGixof/EwFXS5YF27Tzw6OIfiEtKBrKzishyb+qMceMmvRVf9AUV/X7iONzqmkPTtJYQ1cWjLf57lonEe9Fltqu0BGXJKU02xREBIkAEiMD7SaBCN+OZG6mhSasu2LY/FB/2GYuDfxxAy4Z1oCpnh74qW8Pmp+z5wAl/ZGHlnG9gUKdBkaumocaG22wJ4G0Gvvp/70TDwlifbbbTQAbrV5w+fQiZMUnwaO8lrVqL7VU4e/YYejm5ilJHWXJEVUaZiAARIAJE4L0hoFRDr6muxibuJUhPfVUqQH7P2eiBwzBlfCiiLl7EnRvn4T91BNspICk1Px+ZxWbir1w5B2cbc8GIO9hoIk1FA9FRkejv1Q3MqX2ZZfmEs5dv4nrEEQzv118po/r/YlJx81I4Zg7qJOyqv/8gBcmx0dj0w1zUsK6br4sqFm/ajQPhh9Fr6BRATV2mjnzi/VjlyJFbEWUgAkSACBCB94qAUg29lWlNcJq6WBy8FgOGqiE1RxVZKTHo1O1TKVS3BpYwrNMU/oFzYMAl4YPm7UsAj4yORWjIFpjZOGJf2Hmk3L0K38AAttCvhlqmQK9PfBEUsgZxL3Pg5t6BGVIt3Ht4H95sd30NcxupvFT2NoD/7ACoJsVBI/khRk5bXKIuMRF3HsQg98Y/iLr9EGs2BqORpSZ8+o8Tip6OvAITrVw0bMqWH3RevybXke07WBz2CzISHkPTwk5uNcqSI7ciykAEiAARIALvFQGlGnpjdobMrBlzEbxqIb6cPgNqEg6DvNl0dpc+0lfINNjgduLoyfh+xkj07dAMqoZmJYBLVDWwfsfveJn4FJYmegj0n4QG7XoI+fgB/DfjB8POwhD7ftuOXYfCWSyH2tam6NqsQRFDr6kBtGrlhYtHt6Fly/Yl6pEXoa6uAqta9lj40y6AfSxMamJAZ3cMGzZGeIeefx3wSPghNLC3BLRqFBHn4miPNOgi6q/zaNJVtqFXlhx57aF0IkAEiAAReP8IqLA17/zV74prfPT9Vxg6tBMOs1fyjAqtYfPvx/cY4Qc3/acIXLm74hSimogAESACRIAIVFMCSh3Ry2L0MiULj+7/i8SXWfhh+TL0b98URo4fyCpCaUSACBABIkAEiMAbEqgwQ3/jdgK+GD8W2qo58OnsgS+msBPyZGzCe8N2UXEiQASIABEgAkSAEXgnU/dEnggQASJABIgAEagYAhV6YE7FNIlqIQJEgAgQASJABAoIkKGne4EIEAEiQASIQDUmQIZewYvLn2C3fN1PiPozTMGSlJ0IEAEiQASIQMUTUKqhT0nPwcmz/0P684SKb0k5auRf5+s61A/rF0xBGjtcpxVztXt293qZkngPcxu3bMLTu4qdY88LVQaf8ugss0GUSASIABEgAtWagFINfdTd5/CfMQUpsfeqBDT+8B015m5OS1ML/EG6eX8zLzpvKSiDT0Xr/JZQkFgiQASIABGoIAIV9npdBbVHoWp417naunowMzaBhB23r6WjCyM5DnYUquAtZK6KOr8FDCSSCBABIkAERBJQiqFPYdPeHft4Q+1lIvPZpgEf38nS6n08nfBl4AbheyZb3179SwgO7N6OjOQkONexxMSxvmjUuqOQzh8F+/2qjYg4cQDJSfz0P4dG9eyEPM7NPaUy+T9yWF7/+UGIPLoVQd/NhItn3hG5RTKJ+GJvbQd7W3PBj7yJuQX0DIyLlHqVlotFQWsRdoid1JeVAee6Duxg26IhNiETvhNGIenRXWFmwNJUH75D+qNjr0HC0b9i+ciTU1CrPJ1FNJuyEAEiQASIwHtCQCnv0fOH6D6IicH1m3GYO9sPq2fPhGFdFwFhTV0tGJpYCG5mA378CWdD12Dq2HEwreOCXQfDcfHUbmxdtgh2zTxRcASuVeYdjJ/8HV5lqmDrnv2IijyCHSuXwbpJW+ll4R3WdOrbR3BYM6xny3I7rHn67DkMtSXQYA5pHj55CitDfUg0tYV6+M7E598uQdTpnfhyrB8MbZxwOuIaDoZuxbzxA9C+n5+Qj1+3P33mBGqZmSErV4JfWbvOHN6C3WtWwMylpdB2eXzEyClovCydC/LQ/4kAESACRIAICAT4s+6VFSL/SeBauLfgEv6JLCHywZMMrrG7O3dg0yJpWnomx3UcMIKbOsaH43KyuaxsjusyZBw3c/zH0jyp6Rz3Yd8h3LSxvYU8hcOZS1Hc6tXLuLRnT0rUp4yI/2LSOFem84ntq6Tinr/iuEYsLjwkqMwqkpI5ztXrQ+7Y1uVF8sjiU5qwsuSUlpfiiAARIAJEgAiURkApU/di+kx3HsVDB5lo0bydNLsm82TXzK0drh8LAjLSmL/54pPibA2d7Y1za+GFG8dX5+XRfp3Hvakz+M/bCvceJ0CL6dy4MXNBKyPEJWZgadAKXIk4g1S2fGFkaAydrEz/kBxxAAAgAElEQVS2kz9VRqmSScqSU1IyxRABIkAEiMD7SqDCDH0eYBVhDbtwYL0PuexV2Tq3iGxy5SiaQYXf4s405vjNA2WErGzAl7nk5WIuYtYkf1jWdkLsszRMDviijBKlRytLTunSKZYIEAEiQATeVwJKfb1OU10NuZAgPfVVCZ51mb/4VKgjIvK0NI0/fObypTPMn3stIH9dvHhB3gBeuXIO9WzM2Yhfq0jy2cs3sWbNcqQzv/VvIxTofObcH2WKT8tg6++3r2JE74/g3rk76jo6oEnjRlDVyFvnL1xQFh9F5JSpDCUQASJABIgAEShGQKkjeivTmuA0dbE4eC0GDFVDao4qslJi0Knbp7A200BX72GYv3EtMiQ1YVa7EXYdDkf6w+sYzzbjCZ7s2AE2fIiMjkVoyBaY2ThiX9h5pNy9Ct/AAEDyWl1+M57/7ABhM55G8sNyb8aTdUfwOnfzGY6lW9bjVbYW6jdrg/gXuey9gtdBi32xtHXG9oPHYeHoBgNzayS8yEZuJluKKBZk8VFETnG59J0IEAEiQASIQFkElLLrvrDw0D8iEbxqIV4lPIGahMMgby+MmjRHeM2Mf71u1c872Ot1W1kH4CUc7MwwedwYfNC2syCiYNe96pN/oKJlgpdspG5poofPRw6BZ/d+bBb99cQ/vyN+2vxVuHh02xu9XlcWmIJ4Xud123bhwJ4dSEmKZ9EcbK1M8fUX4+DcOk/ve49fYuHShfj70gXkslfwVFQ4mJsY4qsxn6FFl75FqpDFRxE58vSmdCJABIgAESACPAGlG/o3wVpg6N30nyJwJXtvnQIRIAJEgAgQASLwRgSUukb/RppQYSJABIgAESACREDpBMjQKx0pCSQCRIAIEAEiUHkIVKqp+8qDhTQhAkSACBABIlA9CNCIvnpcR2oFESACRIAIEIFSCZChLxULRRIBIkAEiAARqB4EyNBXwHXkD9bbEnoEu7ezY3wpEAEiQASIABGoQAJKNfQp6Tk4efZ/SH/Ou5it/IF/na/rUD+sXzCFnUsPtOrujbO710sVfxyfA1cPD3jkf2ZN6FOuRuWyU3637d2HK+fLPmFPjGCx+shrl5i6KA8RIAJEgAhUDwJKNfRRd5/Df8YUpMTeqxJ0+PN31Jgjei1NLeEM/ry/mRed/GBmJMGuzb9gw8ZfoG3t+M7bJFYfee165w0hBYgAESACRKDCCCj1CNwK01pJFUlYN0dbVw9mxiaQsBN4tXR0YWRgIpWuxuIc6tQWTuxT067B4l8oqebyiRGrj7x2la92KkUEiAARIAJVkYBSDH0Km/bu2McbasxFqwb75+M7WcrCx9MJXwZuEL7zx8mu/iWEHYG7HRnJSXCuY4mJY33RqHVHIZ1fy/5+1UZEnDiA5CR++p9Do3p2Qh7n5p5Smfwf/BG4/vODEHl06xsdgWtvbQd7W3M2mgdMzC2gZ2BcpB4xX2ITMuE7YRSSHt0VZgYsTfXhO6Q/OvYaJBz9WxAu/5uAXn36IykuBlamevAbUfRoX75D8fOu/dizcwuSE2JhYqCDof184DPQN88XgBhl8vMoo10KVEdZiQARIAJEoJISUIqh12Gz3SHBq3H9ZhzmzvbD6tkzYVjXRWhyTd08j3O8m9lvV/yEs6FrMHXsOJjWccGug+HwmxaArcvUYdfME/xa9vlLl2BVIxWBsxbjVaYKtu7Zj9FTpmLHymWwbtJWijEjk+W9EAbVXDVERJyEi2ePciGeOqIvDLUlwjH6i6ZPYb7k9RWWY6ivgQmjhqCWmRmyciX4lbVrzpJV+MChHsxcWkrl5eZmY2TfHjCzdcTvJ85j1oLlWKevi/rteghueOet3YFTu5djqt8E2Dm74tTFaMwP/hFmutpo4zNMIb2U0S6FKqTMRIAIEAEiUCkJKMXQ80bSzroW4pO1mJvaXOb0xQLG9vZFGvwoLhOHQzdh7sh+6D5ohJDWzK0Fegz9F6vWL8MiV3cWw+bKWbAy0UXTFnkGsnlzlmfIA6wMXoIFQa2kI1sd1n9YMHsurkccxaB+/YvUpcgXc2MDaXYbC+YKtxyB9zzXuUPerARf3L5uQxw5sRdXr0agUyFD7+ZsAe9+g4Ua3Jq3RM9b0Qj+JRjL2nbF40QOob+uwcaAyXDtwhz4sFC/YQOcvnoN20J3ok1PVq6Q9z4hg4ygjHbJEE9JRIAIEAEiUEUIKMXQi2nrnUfx0EEmWjRvJ82uqc4b+3a4fiwIyGBuXTV0S4jSZrMFbi28cOM4ezWNz6P9Oo97U2fwn3cd4hIzsDRoBa5EnEEqW74wMjSGTlYm28mfWqZq6oy8q2sbXD2yiq1ppOP2w5fQYlMaEwPZd/5TKGQbseF+Nlv3UMDQl1kxJRABIkAEiMB7RaDCDH0eVRVhDbtw4Pg5azlBla1zi8gmR8rbSc7KBnynzwAXcxGzJvnDsrYTYp+lYXLAF3Ir5Nte4Hk3j0MWVs75BgZ1GhQpq6HGqGnkLYHIFUoZiAARIAJEgAgUIqBUQ6/Jhqm5bPo9PfVVCch1rU2RCnVERJ5G94YthPQMNki9fOkMmtjXAjS1+b13JQJvSK9cOQdnGzatXszYnb18k03dH8FwNnWvZVS+afcSFZYSocrsrERNHan8xgB+x2ChDXZpGcCD21fx7cCP4N65u1Da3Jpl0WDtkRHSmaiLkafQtDbLzNruYKOJNBUNREdFor9XN+ZAuHiX6LUwWfrIqJKSiAARIAJE4D0koFRDb2VaE5ymLhYHr8WAoWpIzVFFVkoMOnX7FNZmGujqPQzzN65FhqQmzGo3wq7D4Uh/eB3jly3KW3tnu875EBkdi9CQLTCzccS+sPNIuXsVvoEBRaauU9lOf//ZAVBNioNG8kOMnLb4rV0+3q47O7rgzwOncfLQPuTUMIJq9jN4deoNfn3e0tYZ2w8eh4WjGwyYlU94kY3cTLbMUCzEJLzC5YgLeJHBYUfoIWQ8uoHxXy0R2l7LVIJen/giKGQN4l7mwM29A3unTwv3Ht6Ht2dr1DC3kUqTpU/xOuk7ESACRIAIvN8EJLNZUBYCfoOcWS1nHD15Fgf2/Y4zp47D1kQHTVu2hwobhrZr3hTJnB627vwVxw4fQA32Xvr3M6bBqSUzaizwu+63/34QkpREXI5+iIP79yL7xSNMnzAGzTt9XGSUyxu7qHvxiP2PdQI+GwIz+3rKakapcho51cefN25hz97dOHvqBGoZaKBZKy820ldB2xYeuPLvA2zfEYLfQ/fij+MHYG2kjy4ftkcth4aCvNsPYhEVdQf79+9DeNgxWNTIxfcz/eHglvfaID+Ab9f8A6joWeBI2Gns+30vThw/imeP/0Mn9zbQNjIroldZ+siaCSi1YRRJBIgAESAC1ZpApXJTy79H3mOEH9z0nyJw5e5qDZ4aRwSIABEgAkSgIggo9QjcilCY6iACRIAIEAEiQATEEyBDL54V5SQCRIAIEAEiUOUIVKqp+ypHjxQmAkSACBABIlDJCdCIvpJfIFKPCBABIkAEiMCbECBD/yb0qCwRIAJEgAgQgUpO4K0Zev4wnOXrfkLUn2GVHAGpRwSIABEgAkSg+hJ4a4aeP/lt45ZNeHo3WmF6Kek5OHn2f0h/zruqpVBeAv/eT8LAUZ/D3cMDHuxzgTkVokAEiAARIALvFwGlnoynLHRRd5/Df8YUHFoXBC0DE2WJfe/kzF29FlziP/ht809Qr2kEfV2d944BNZgIEAEi8L4TqJSG/n2/KMpofw47kj8hPhYdXRxhUcdRGSJJBhEgAkSACFRBAkoz9K/ScrEoaC3CDrET7bIy4FzXAcWdzsYmZMJ3wigkPboreLGzNNWH75D+6NhrkOAoJoWdX9+xjzfUmKtXDfbPx3eyFKmPpxO+DNwgfJcnR8x14H3TfL9qIyJOHEByEr9EwKFRPTtMHOsL5+Z5x9KKrSvxZTbWbFiLvyIjEPckBrm8S1lkY5bfYHQZ8Lkona/deoaZARPw/MlD1DKrCVc3LxwOOw4DzWwETv8SLu7M0Y0Cgff2x7F/kkIOeBQoTlmJABEgAkSgmhBQiqHnR49T5y1F1Omd+GqsHwxtnHA64hr+vnOnCCZDfQ1MGDWEGTIzZOVK8OvBcMxZsgofONSDmUtL6DDf8yHBq3H9ZhzmzvbD6tkzYVjXRZBRU/e1m1Z5csRcG/5c/fOXLsGqRioCZy3Gq0wVbN2zH6OnTMWOlctg3aStIEZMXU8T07F3706M7PEhmrQdDW3dmniRnIyG9hZSVeTJeZqYgsQnN7Fu4XJsPxvNfAWswaoffsTmsKuYvWgufuPPxNdScOqdd4NLhl7M7UB5iAARIALVloBSDP2DJ+n482QoFvkNRYcBwwRYH7i2xs7QrUXA8Z7eOnfoKI2zr9sQR07sxdWrEejEDD3v2MXOuhbik7WYu9tc2FpZwNjevgR8eXJKFJARYWWii6YtWgo5mjdvgR5DHmBl8BIsCGoleJUTX1cuen7oCevm7UqtTYwcFZUcNGzYAF4SW5zavwqNGzVET626mHViE7jUZKgoYOhjE9LxLOYO6nf6qFR9KJIIEAEiQATeDwJKMfT3HidAC5lo3DjPz3xZ6OISM7A0aAWuRJxBKpueNzI0hk5WJtLSU8sqUmq8suQUF67NZhTcWnjhxvHVQAZzM6utC2XVpYgcVZX8lyHYiFw1f0TOT8OX7aG+aEvGzV6GG3/sRmM7PXTsPbx4M+k7ESACRIAIvEcElGLoVfihODNDHL/wXUbIygZ8p88AF3MRsyb5w7K2E2KfpWFywBdllCg9WllySpfObxVQBb++zQdl1aWoHB6nRFL+S/P9pBG42N4T33w7Hnf+Ooe6rTuX1VyKJwJEgAgQgWpOoPzWpBCYutamSIU6zpz7Ax+zKfjSQloG8OD2VXw78CO4d+4uZDG3ZoZVQ7tEdk11NTZxL0F66qsSaYrIKVFYTgRvkK9cOQdnG3NAQ4vNNIjXWZZoRXX2crOGV/jFfJGPZYkuNc3EUB8d3F2x1K4RTp45Roa+VEoUSQSIABF4PwgoxdBbm2mgm89wLN2yHq+ytVC/WRvEv8hl++ZfB36N2tLWGdsPHoeFoxsMmJVPeJGN3Ew2RV4sWJnWBKepi8XBazFgqBpSc1SRlRKDTt0+FdbMxcopLre075HRsQgN2QIzG0fsCzuPlLtX4RsYwIbUakqrS9k6l9aO4nHCrICaOlLTUoon0XciQASIABF4jwgoxdDzRmXOxBFYZ6yH7Xt2IGXHXoaQg5OVKSxsHQScGupA0LwfsHDpQkybMx+57BU8FRUO9iaGsLCuUwS5cU1g1oy5CF61EF+y6X41CYdB3l5Alz5MjqpoOWKuo0RVA+t3/I6XiU9haaKHQP9JaNCuh8I6y6pLkbbLklOeNK5gHaI8hakMESACRIAIVHkC762b2uwcoMcIP7jpP0XgSvbufzUL/CuPvUZPwoeWr9j5AxurWeuoOUSACBABIiCWgFJG9GIrq4h8/AD2zv0HyOF3zZcRzIz0oWdkWUZq9YiWsI379nYO7KyAXXh8OxqahqbCEbjqmiX3RFSPFlMriAARIAJEoDQC1c7Q8850RkyZBi4+prT2CnFj+nhg4IR5ZaZXl4QZY0chYPZd9B0+Wng1b9GXI9CK7aWgQASIABEgAu8Pgfd26v79ucTUUiJABIgAEXifCbw1N7XvM1RqOxEgAkSACBCBykKADH1luRKkBxEgAkSACBCBt0CADP1bgEoiiQARIAJEoHISyGDORZev+wlRf4YJCvIHum4JPYLd29nR59U0kKGvhheWf3Ww61A/rF8wRTjdr1V3b5zdvb5Kt/RxfA5cPTzgkf+ZNaFPhbbnn7vJcPNogzvnjhSp99Sle2jj1R67NyyoUH0qQ2Up6Tk4efZ/SH/Ou3mWH8rD8F1fd/mtqhw5FL0Wb6J1Rdzzb/O68xu2N27ZhKd3owUMvCfTbXv34cr5P2RiqcrP1Upp6J8k5sCNeYJ7dDEcfO+rQ/9h2L/x/XuQyrzrZCTyBxipqbGT/TS1hN32eX8zjz1VOJgZSbBr8y/YsPEXaFs7VnhL9oeFw1Q7C3U/YF4NC4VHTxIgYU+AO//9W+E6vesKo+4+h/+MKUiJvSdKlfIwfNfXXVTDZGT6934SOnzcX+h0Fw+X/rmPMZOmwd2zPevAumPU0L5slBlePJuo74peC1FCy8hUEfd8ZbzuVfm5Wilfr/sr+j70VVJh7dgI959l4Hnsv2jZcmwZtx1FFyfAv0OvrasHM2MT5hyHd2OvCyMDk+LZqtR3NdYOhzq1wfeq1bRrMN1fVJj+fGfz6NHf0LNVU0DXoEi9/bu5oXGddXCubVth+lTFisrL8F1e9zfhnPA8DZt37MLO7euR18W2KiIuk91T85b/iGamEgwPnIOXmRKs+ulnjJ8+FUd27oe6adH8b6KLsstWxD1fGa97VX6uVipDH7BsK87sDs6/L7Xg0aO38LcOc3DTZ9wkhK4LgnF9N7xIycGC5StxLvwwcpiLW96Tq3Mdc6xdvg4q+kZCGf6HtPqXEBzYvR0ZyUks3RITx/qiUeuOQjr/4Ok6aAh6N9DBxNlrhbgX7Fh4j64eWDF+ANr38xPiEl9mY82GtfgrMgJxT2KQm80KIhuz/Aajy4DPhTx8CI+4gS2bNyD6xmVI2Kk9ejrqGDOoF7yH5Hnn4w3Uz7v2Y8/OLUhOiIWJgQ6G9vOBz0Bfwe99QeBPtPOfH4TIo1sR9N1MuHjmHccrzSDyD3trO9jbmrPRPGBibgE9A+MiJeXp8yguA32H9EG/1nUw6eulgo5PE7Px8Wd9MMKrPkZOYecQMPCxCZnwnTAKSY/uCrMHlqb68B3SHx17DRLSr916hpkBE/D8yUPUMqsJVzcvHA47DgPNbARO/xIu7t2ENbLvV21ExIkDSE7ip4E5NKpnJ1wv5+aeIlucl01eu/hcYu6fwpVG30vCq9hb6PPV67MX+KnF7h+3h25+Rs8mZiVOWBTTdr64PIZ8Hv5+XrtlJ/azI6bTXzxjaDnUtbeBv98oOLXsIGghr+0858Cgjbh88iji42LBT+fZ1zLG5HGj0dQzz9GUGDkpbDmoYx9vqDFX0xrsn4/vZKF+Pvh4OrGTGDdIvxf8UV6GJQSVESHvmv5+MgqLvh6OPWuWwczl9azM/A2/48T273B871Go1DRR2u89+NeDePDnISyfMwfTVxY8014rzx+LvXP1j+x479e/fT1TR0z+3AcJj+6w35E4Qy/2Woh5jsm7D8Xc82LvsTIuo0LRYn9fr9JysShoLcIOsRNQ2dHrznUdpL/bwhVe/jcBvfr0R1JcDKxM9eA3Ygg8u/djjln5J1tekPdcVagBFZi5Uhn66aP7YHy/bhjx1Qx0dVDHYL85mLv+Vzy/sQdLfgyBgYmpgCZw9WZcDQvBd1MDYGbniOfJaYh9fAcqzH88H/jT8b5d8RPOhq7B1LHjYFrHBbsOhsNvWgC2LlOHXTPxxuNpYjr27t2JkT0+RJO2o9lIuSZeJCejob2F9DL9duwylnw/CZ6uThgwYxoMjM2R8PwFnK0NpfrMW7sDp3Yvx1S/CbBzdsWpi9GYH/wjzHS10cZnmFRWBls/On8hDKq5aoiIOFluQz91RF8YakuEe3TR9CkwYh7tCgLPR54+1maaCPxuBaZ/+Rkau2xCO59RmDF/AZz00zDSb6ZgxPlgqK+BCaOGMCNuhqxcCX5lnOcsWYUPHOqxB2pL1jlIQeKTm1i3cDm2n43GgX1rsOqHH7E57CpmL5qL39w8kauuw07wuwSrGqkInLUYrzJVsHXPfoyeMhU7Vi6DdZO2Ut1l/SGmXXx5efdP8Tp2HzkCc10Otg2aSZP4qcU9W7chN0cF42bNZvHPixcT1XY23SKXYd79vEm4nyeNHIk6Ls0QE5+G7+YFIPbeTcHQi2k7vxZ57uIlWKg/R8D8H5DOqQmdz0mzvsaRbY6oYVtPlBwdNkQNCV6N6zfjMHe2H1bPngnDui5C+2vqapXgwEeUl2GpwkqJlHdNW33ghBSJNk6EH8TAfEPPd55O/PE7PJo4QUUvb4CgjN87r14Ae5ap+H4C3nMlt2pdKRqzmalCRp7P8Dgujg1pMlDTVPypnWKvhZh2yfsti7nnxdxjpcIoR6SYZ0uOhg6mzluKqNM78dVYPxjaOOF0xDX8fedOiRpzc7Mxsm8PmNk64vcT5zFrwXKs09dF/XzfJ3wBWc/VEgIrUUSlMvT6NbSZ/dBGUsJjeAzoDyMTE9x9eBeejnVgZFlLii2WjUaM9LXQskVLaBoxl7JCYNOq+eFRXCYOh27C3JH90H3QCCG2mVsL9Bj6L1atX4ZFru4s5nVPWlqwzD9y0ZPtGbBu3q5EDr5H/ePKQHi4mGP+ErZrU72wz7687I8TshD66xpsDJgM1y6sh8hC/YYNcPrqNWwL3Yk2PQcL3vL4oMOekwtmz8X1iKMY1K9/ifrERpgbv55itrEoYJRXWqw+Xs3rsFP1puHrZYHwuZOB2xGhCN34E1NST6oG75mvc4e8WRI+0r5uQxw5sRdXr0agU77LYhWVHDRk7fWS2OLU/lVo3KghemrVxawTm8ClJjProCPIszLRRVN2TfnQvDm7XkMeYGXwEiwIYiOwQrMeQoZSgth2ybt/CovmH9ThYaEY0LY5UON1Z4mfWqxtZ5u3lMCMdWmGnpcjr+0qrKw8hjHxWTjy+yZ885kPeg8bI6hXl80+ZRQ63FFU25F3j1mz0UqLtvxvALBzcEXvT0/j77/OowUz9GLkqLB71c66FuKTtZg76VzYWlnA2N5ekFdaeFOGpcksHifvmpoaqMK5STscPHUKA8ewHy1zQ33r4Uu8eHQTA/wCpB3XPLlv9nvnZagUGgUW17W077eZLj8uCWCndnaHTq26pWUpNY6vRvy1KLtdvHB596HYe56XJeseK7Uh5YyU9/t6kKiKP0+GYpHfUHQYMEyo5QPX1tgZurVEjW7OFvDux57FLLg1b4met6IR/EswlrXtKn0+y3qulhBYiSIqlaHnuTx9loac1ATY1XZCGhvdPn10C24dPyqC7Ovxfpgw5W909PZGZ093fNStG5q1bi81BncexbPp/ky0KGSYNdlUWTO3drh+LIg9Idk5+GoFk65vdjUex6ci43k8+o9jewhKMfK89NsP46HFuroTA1ex4ST7FArZRmyYxS8H5Bt6Psm9qbPweVtBrD78Q2TiZ944/+cZhO4PgT+bJTByaFRErbjEDCwNWoErEWeQyqZyjQyNoZOVyXb7p5ZQX1Ulf+8nG36q5s8IcGyavrSgzUaNbi28cOM46zzx1yt/tqa0vAVxYtsl7/4pXMf12/HITHgAnx4TZFUtN62stvOTgvIY8vezNpeJtm0+LLMeUW3n13GKBSN9TaiyzkZcwlMhRZScQvdqcXmlfVcWw9JkF8TJu6b87Tbo44H4YeZhJN6NgpGTK35jMzWWermo1zSv0yNLfkGamN+7GDmF81y7FYfPJ45Ebzc7DPl8VpGpYkVlvUl+efdheWUXv8fKK0dWubJ+X/ceJ0CL2YLGjVvIKl4iTZ39VFxd2+DqEfa8zmQdQxHPnxJCKlFEyV/+O1Tu21U7ELZzNbShjt5j8nap8goGLN/CXnxcz9bo17I1+mZwsjPEgZBtCD8XgWNHj7ENLLPRsr4lFi9aA7WaBWvRKsKaceFQ3GUrb2wyhTX38oeCnrtExogzr94srJzzDQzqNChSmYYa05KNLioyKKJPDOvIPP7vb2iy9u05dgI9+rIRZf7INisb8GVuhLmYi5g1yR+WrHMWyzpqkwPy9iUUbxPfcZAoYCT468NPSYsNYtsl7v7Jq3XXof2oxUaDVk5NSlVD7MCtrLaLYZjDNm6osA6RLHai2s7W6EsLKqwDlsvlJYqSU5oQGXHKYiijClHPBI9mTpirZ4F9B0LQ18YVxw6HwK8L299Qg/nFFhnE/N5FihKy/XnjESZOGo7POjTB51/NLXOwoIjM8uQVcx8WyBV7zxfWo/A9Vh795JUp6/eVd73Yr4ffPKBg4H8L5WmrgtVUSPb8IVaF1CW3kgmDeqJNj89gZ6GO7Zt+wufTFoJTScGW1SuwdeuvMGDTwgVBQ10FXTxbYcm8b7Bx8z6ciopF2D7WIWChrrUpUllnISLytDQ/v/nu8qUzaGDPlgCYBzd+GqoG27h399FjIIdZrHIGC2NtaLJNPL8eDM0bmZcix8HGDGkqGoiOimQe5dgmOTbNWfCxYpvmit9NZy/fxJo1y5GemDfKKkXkG0WJ1ecVG0hPDJiKdo762Lx5D/5m++TWrpiTd8IE04Cfkn1w+ypG9P4I7p27o66jA5o0bgRVjdI95Hm5WeNU+EXAIG+vhaxG8A+eK1fOoZ4NW3Yo1BFS5TsLaupI5TczFPvxim0XX6+s+6dAL779504fQI92bcrs0fOdSXWmX3Iq6/Xnsh2XZYSy2i6GobW5IdLZprfr1yPLkM7eSFDwHitLkCJyNNmwJ5ctgaWnvipLHJTFUNZ1L6hc3jWtyV7W6MaeL7uOn8Sx42eh8uI+evcZVqbupSWI+b2XVq60uJj4DHwxdSyGdGiMz/1/eCMjL+ZalKZDQZyY+7Agr9h7XlZ9YtPEXHdeVlm/rwJbcOac7Hfki+vDv2t/MfIUGta2FuxFVQ+VakRvbKCLe48foZmjPWwcHPH0wl3Ym+vD3ql+kYf94o1b4KCngrpODaGqro2Iq7fZKywcjI3NhOthbaaBrt7DMH/jWmRIasKsdiPsOhyO9IfXMX7ZImGKn1+h79HZB1uWX8C29StR380DT5Oy2ONUscD20mHcmK+wZqE//CePRefuH0PfyAwvU9Kgmv0MXp16o5apOnp94ougkDWIe8nOCHBnowg1Ldx7eB/enq1Rw9xGWilvL/xnB3zcNPEAACAASURBVEA1KQ4ayQ8xctpixRQSkVuMPvxIekEw6zg9/Qvfb9kNDRMjTPNfiCVzxqK9xxE4u38krOlZ2jpj+8HjsHB0g4G5NRJeZCM3k1nIcoTI6FiEhmyBmY0j9oWdR8rdq/ANZOunhWYB+ClYZ0cX/HngNE4e2oecGkYKc5Z3/xSofiX6MXKfx6DnR9PLbA2vTxO2OS5s50kc3bsLakZWbKovHh26iDvQRwzDOrVqoGGLjpizehNSVWtC36I2zl+JKrJzWMw1LbMRhRIUkWNlWhOcpi4WB6/FgKFqSM1RRVZKDDp1+1QqUVkMZV13vjKx13RI7574fWcQliyegW6tG0HHqo4YLNI8Yn7vfOb0zFw8enQPGRkq7PeQgcTnKbjLNoDV1NeBUf5muxW/7IAhnqFrr89w5959aR01tNVhYWWrkF5iroUsgWLuw4Lyb3rPy9KjeJq86148f/HvvC3oxjx2Lt2yHq+ytVC/WRvEv8gt9Tkfk/AKlyMu4EUGhx2hh5Dx6AbGf7VE1P6g4vVWtu+VytDzvajYh/+iRb+8TW8Xr0Wini0bgRca0fGvn4Ht7l67dRdSkvJeWzE10sXng33QrFPew5Wfbvlu0kisYpu8Vm3byh4+v8DBzgxB8wOL7Lgf5t0OzxP8sGHPNmRv28MLhnMtc5jb1FXoOg3u0QYWRiuxbctP+HrufKETUUNLDX0+codXB2+osLv1m/GD2UyFIfb9th27DoWzHBxqs5mHrs0aFDH0msx4tmrlhYtHt7GzA9orpIfYzDwfefpE/v0UR3evxua5M5mRz9sF7O3VGL/t74BvlyzATrahUYNN4QfN+wELly7EtDnzkcteXVFR4WBvYggLa8UeoLzuElUNrN/xO16ymQxLEz0E+k9Cg0I7XgvaN2X4QEz77zrbFbsEaqq5GNirvWjOYu6fgnp+PbAXtiaaMM3fUV4W3wmD++DhLdYhWhbE2pCLQUyfDqyDJybwr1nJY8jPPi1nO9vnLdVhr/AFQ8JmDurVybtHVfOXjMRcUzH6KCLHmM14z5oxF8GrFuJLtoSjJuEwyNsL4Ds5/BOahTdmmC+Hl1XWdc/hXxSU80woaLudhRbadPwUfx37GWNHTiq2CU8MIUDe751vO3+AzZhRQ4V36PkR8IGzCewzDJ92aCi8zsvPWF29FoGUFDUMHTe+SMUt6htiyVo2Q1io7fI0E3MtZMkQcx8WLl/2Pa/8SeKyrrus9hSk8ffznIkjsM5YD9vZq6kpO/ayJA5OVqawsHUQsvGzBi1dm+LC8UeYNPUrId3F0RbB7Dln29RDTDWVPg+5qa30l6j6K8i//91jhB/c9J+WeBf9XbWeP1Ohc58emPRREwycGPiu1Ciz3v9i0vFJ/47YEfgVnDy9y8z3LhMqO8N3yYbqJgIVSaBSjegrsuFUFxGQRSDyxj1IUuLQVeQUvCxZykjbsf8waulI2FkS5nj2Ig0/7fgN5vyRvI3Yq4eVNFQ2hpUUU6lq8Utnd+4/QA7/xkkZwcxIH4YKvHNfhpgKja6u7apQiOWojAx9OaBRkepNgH8Yhez/DbXNtWFUm+0PeceBn/GI/OsfrAk/BI69uqjKlgeaf1AfP6wIZnsC2GbFShgqG8NKiEimSvwy5ogp08DFx5SZb0wfDwz+otBhCmXmrDwJ1bVdlYdw6ZrQ1H3pXCiWCBABIkAEiEC1IKD8nRPVAgs1gggQASJABIhA9SBAhr56XEdqBREgAkSACBCBUgmQoS8VC0USgaIE+AOXlq/7ifkLDxMS+LN6toQewe7t7IheCu81Abo33uvLXyUaT4a+Slym10ryG7O6DvXD+gVT2Hny7J377t44u3t9hbTin7vJcPNogzvnjlRIfYpWwrvRdPXwgEf+Z9YEcYfWiKmH30S0ccsmPL0bLWTnvXRt27sPV84rduJWZWdYwCIlPQcnz/4P6c/ZcYhvOZy6dA9tvNpj94YFb62mynhvvMvf8lsDTYIrJQEy9JXyspStFH8AhBpzTqKlqSUcxJH3N38sx9sP+8PCYcq/0vVBK+EM+s++CsSn3swhSLFXgPKMmTtunvpdUCqNnTS1YPXP+LBrd8EIDx34MSLD9uf5Ey6k9v3YF5j6zTx4dujE8rnjswEf4+ieX0ocdVtWS3k3mrs2/4ING3+BtrVjWdneaXxhhu9UETmV8we++M+YgpTYe3JyvnnyoycJkDCrd+e/f99cWBkSKuO98S5/y2VgouhqSoBer6tiF1bCumbaunowMzZhTk6Ya0kdXRgZmLz1VvDTk0eP/oaerZg7YF0DoZPR1esjrL2wF2nxj5lhfX2a4JWoW8wx0SvUYQ6I+Clu/4Wr8Hf4Vsz8YiqM7epjz5EzmPztXGzUUIOTezdB99uPXmHw8H6oZyjBrMl+qGFshahb9xCf9FD0CWH8CXIOdWrnuY7VZgeb48Vb56JIBcUZKlK2Ouft380Njeusg3NtxY59VYRJZbw33tVvWRFulLd6EKh0hv5FSg4WLF+Jc+GHkcNcnQpnHdcxx9rl66DCnNDwIZMZndW/hODA7u3ISE5i6ZaYONYXjVp3FNL5B2rXQUPQu4GOcNwkH/hTujy6emDF+AFo389PiEt8mY01G9bir8gIxD2JQa7gyS4bs/wGo8uAz4U8fAiPuIEtmzcg+sZlSNhQVk9HHWMG9YL3kDwvbfwU3M+79mPPzi1IToiFiYEOhvbzgc9A3yLnJPPHr/rPD0Lk0a0I+m4mXDx7SOtQ5A975gjH3tacjeYBE3ML6BkUeOwDniTmoFvvDvjqEw8MnPC9VOyIgEVIi96PHSHHhCOFxba9QED0vSS8ir2FPl+9fm+3TZMGWMIcAv/790U0zjf0/Ej/6MnDcKljAXX2jvftx6k4e+xXBE0dh9a98s5Ad3FxwbU7t7Fs/QqsadUROarqmL10CSw1k7FpXSgk+U5v2rRuqQgWUXmv3XqGmQET8PzJQ9QyqwlXNy8cDjsOA81sBE7/Ei75HY9XablYFLQWYYd2A+xoX+e6DkXOli+o7PK/CejVpz+S4mJgxfy8+40YAs/u/Uo4KuLzl8bw95NRWPT1cOxZswxmLq8Pv5m/4Xec2P4dju89ChXmNEnePV+gj7x7VR6kFLYc1LGPN9SYy2EN9s/Hd7K0iI+nE74M3CB8F3P/xCZkwnfCKCQ9uit0DC1N9eE7pD869hokdN746fTuH7eXcvVsYlbiZES+oxgYtBGXTx5FfFwsf9gt7GsZY/K40Wjq2V1ecxRKfxf3hqzfskLKU2YiIINApTP0gas342pYCL6bGgAzO0c8T05D7OM7UMn3B8wbkm9X/ISzoWswdew4mNZxwa6D4fCbFoCty9SLnGUvo91C0tPEdOzduxMje3yIJm1Hs5FyTbxITkZDewtp0d+OXcaS7yfB09UJA2ZMg4GxORKev4CztaGQh9dn3todOLV7Oab6TYCdsytOXYzG/OAfYcY8YLTxGSaVxTtcO38hDKq5aoiIOFluQz+V+YU31JYITu8WTZ/CfMDrS+swNZAwHdwQduECBo5lFaprCF7mbv0Tib5NmS/5fL8BYtouFcr+2M18d5vrcrBt0EwabW2mBVP7Bjjwx0E07txXMG7JzA199LVzmN2/E1tXUMflf6JRAylo1tJLWo4fXXXv0hu/LDuB3JfPEA8z3Ij8A4HDfaRGvnDdyvz7aWIKEp/cxLqFy7H9bDQO7FuDVT/8iM1hVzF70Vz85uaJHA0dTJ23FFGnd+KrsX4wtHHC6Yhr+Js5JSkecnOzMbJvD5jZOuL3E+fZ+fvLsU5fF/VLOaO/NIatPnBCikQbJ8IPYmC+oeeN+ok/fodHEyeo6BkJ95iYe17evVpc99K+67BVoJDg1bh+Mw5zZ/thNTtj3zD/rP+auq/dKYu5fwz1NTBh1BDWoTJDFjuL/lf2O52zZBU+cKjHOjUtwU+n79m6jTn9U8G4WbOZOs9LqMTvhTh38RIs1J8jYP4PSOfUhE71pFlf48g2R9SwrVeiTHkj3sW9Ieu3XN52UDkiUJxApTP0sazXbqSvhZYtWkJTeuoXmy7OD4/iMnE4dBPmjuyH7oNGCLHN3Fqgx9B/sWr9MixizlaYe5Ti7ZTxPRc9P/SEdfM8RzqFM/Kjmx9XBsLDxRzzl7Dd1cxoFg+PE7IQ+usabAyYDNcubCTHQv2GDXD66jVsC92JNj0HS72v6bDn5ILZc3E94igG9etfXJTo7+bGBtK8NhZFT0bjpwO9u/XBT0vCkfr0AXSsHfDgSQrSnsWgS/vPitVRdtsLZ+Q7CuFhoRjQtrnUFz2fzryUoktHH4TtmA2kJgtp//z3FKrs6Fj3dl0FEQ8fP2T5OGjo5XWMCuTaWtkw1xHqSHmRiCdZmtDgctC44evrXLh+Zf+topKDhuwaeUlscWr/KjRu1BA9tepi1olN4Fg7HiSq4s+ToVjkNxQdBgwTqv/AtTV2hm4toYqbswW8+7FrzIJb85boeSsawb8EY1lb1v5CXvfKYmjKfN07N2mHg6dOYeAYdsOxjtithy/x4tFNDPALEEa+j57Kv+dTMiVy79USypcSwXce7axrIT5Zi7l4yoWtlQWM7e1LyclHyb5/eI9onTvkzbLxue2Zm+kjJ/bi6tUIdGKGnu/w1bazzVtq0dJhOUoa+oKKrdlsSYu2/G+b6efgit6fnsbff51HCyUael52Rd8bsn7LBW2n/xOBNyVQ6TbjfT3eD4kqVujo7Y3vZ03Fpf+xXc2F/HzfeRTPJosz0aKQYdZkHsCaubXDP/fYcZEyzoZWFNbj+FRkPI9H/57ME1kpRp6Xd/thPLTYsGNi4Crpbu927Tzw6OIfiEtixk9YDngd3Js6Y9y4SdB6i0eXdnVvjleqerhw5qhQceT1KOhJ0uDg0qKILmK/XL8dj8yEB/DpkdeRKVzuo/YeeJLCpmFvXROi9/1xjHmG0oO+jYNY8cx7lAoz+hUfVFXyb382ZFbN9xTGMU3uPU6AFrvHGjdWjBff8XF1bYN/Hz5l60vMaBcKZTHkqx308UDce5qKxLtRQonf2OyJpV4u6jXNM2xi7nkx92pFE45LzMCM7xeha49eaNfOHcMGeUOHHeGbxpbk3iQY6WtClXUM4hIY57cUKvLeeEtNILFEQEqg0o3onewMcSBkG8LPReDY0WMYP302Wta3xOJFa6BWs2AtWkVY8yscOH5+s1DgH9yZxYxssSJyv6rwwxsWJPmuQEsrkFdvFlbO+QYGdRoUyaKhxsoXcrFbWvm3EWdupIYmrbpg2/5QfNhnLA7+cQAtG9aBajk37e06tB+12MjTyqlJCXXrWOnC0JaN1I7/jsEN3PG/M4fx5UedWcco700Aaytr5pJTBZlsLwXv1rYgPGAjfRVkoUZNI5hy+uwv5trz3xuwadGhRB1vK4K/vJJCo+6CevKuO+t88AvECgb+fsi/bYqUlMXQo5kT5upZYN+BEPS1ccWxwyHw68I41GB+YKVB9j0v5l5VsClvlJ13w+rLXNdyMRcxa5I/LGs7IfZZGiYH5O1rKSy8NF7yKldhnbRcTvHrI09uQXpF3htidaJ8RKC8BCrdiJ5viIa6Crp4tsKSed9g4+Z9OBUVi7B9W4Q21mU+3FPZlG9E5Glpm/nNd5cvnUED+1qAprYwJViDbdy7++gxkMOeOOUMFsba0GQboX49yHxDl9FpcLAxQ5qKBqKjImFvxzbJsWnOgo8V2zRX/Kl/9vJNrFnzf/bOBK7G7I3jv9ttR6s20qIo25Cyl2Tfy5gZ62AYIUtjzT4hZN9KGMyQLJOlwdjHNv5oLKNhxjKMNVFRSfty/+e81VV0697cuHiOz/3ovu85z3nO97z3fc7+rEA687leXoH3Eof2HYS/2RDw9YsXcefaOfTmC+FKaLDI0uUlc5519vR+dG3ZnC/3fyMa92PdpeOXOHDmHG5ef4SM+NtozYbzC0LDWvasvirg8h8npdf44sVfD+9BHVsLqOkZgw9f29RqjPU7d7E5++dv5KHoBe5fWszWB6TyRRElGGsPF0ucOnERyF/8V5BPwTP2+1nF9sjzvfYXL5xi5bIUnsOCUBpDfbZBoFPXrxF+9CSOHD0DUdJ99Og5SJpenmdenmdVEY5abHgil02Bpae+VCSZNC6fqnhwOwqDe3SGa/susKthjwb160FN8xWXgsi8Oa3BGsTJqWwUpNDoXZkyLiWRqj0bpahLt4mAUgioXI9+8YZQ2FcSwc6hDtQ0dBAZdRtabDjV2NhUKLClqSY6eg5C4IY1yBDrw9S2HsIPnkD6w6sYtXyRYMz4DH3X9l4IXXEeYT+sQi0XNzxNyGJriBULbC0dRgybiJCFfvAbOxztu3wOPSNTvEhJg1r2M3i064GqJhro/oU3greHIPZFDlxcWU9MXRv3Ht6Hp3szVDCrJs2Uv8f8/KdCLSEWmskPMWTSYsUUUiC2S20LGFZvCL+AWTCQJOCzRq0USP0q6p83HiM3MRrdOk+Wmb6bR0tsWQ8Er14MFzsTVGTrAgqCXVVdNG/7JaYtX43pahVhbFULOw+eQvz1swicN11YsMcfwplj/TB02Bf4dtgg9GFrLwxMLdkQeixexv2Db4aMk3uLHc9X2KlRoy7+2H8aJw/sRU4FI2l9ySxEoRv8Gevk9Q2Whf6Al9naqOXcHHFJucU+P9HxL3E58jyS2FkB2yIOIOPRNYyauKRIo0oehgN6dMMvO4KxZPEUdGpWD7pVqks1kueZl+dZlafsBXGqmOhDolURi9euQZ+B6kjNUUNWSjTadcrbOVGaLD4/b2HliK2/HoV5DRcYmFkiPikbuZms5fha4PXVoK4zju84icN7wplHvips6iMObcrBRbCqPRuvs6DvRKA8CKiUoefbz8BW567ZEo6UhLVCeU2MKmJkfy84t+spfOdDarN9hyBIXxdBYVvYy2cz7K1NERwYUGTF/SDPlkiM98H63WHIDtvNBcOxqhnMqtkJcuQN/bs2h7nRKoSFbsSMuYFCI6KCtjp6dnaFRxtPiNibY+ao/rA2N8TenVsRfuAEiyGBLRt56Ohcu4ih12Ivv6ZNPXDxcBiaNGklrwplisd72mOGjsWcKUPwVRtnqBnmNZQUFfbz/j2wqqwFk/yV18Wlt7bQQZWajfD3PxexZNLQIovQ+It1gd9oLDPUxeyV65Gd9lLYHrV45iQ4unaWimtYyxQ/bdiO4JBVmLVoOUSsZ2diWAE9u7FpAGGIVrHBp3Hf9MWk/66yVfBLoM7cuvbt3kqoL3kCf8ZmjRmMdcaVsHX3NqRs28OSSdjaAxOYW+U1YnjPsIlTQ5w/+gi+EyYK9+vWsMLaJQtg1dCtSDZyMTTXFhpEV478hOFDfIs0bOR95kt7VoUWkJzBmM0aTJ8yF2uDFmI8G4JXF0vQz9ODrb5kv0M55PDnL3jefCxcthCTZgUil21RFIkksKlsCHPLV42YAnVG9++Jh/9ewZzlwWyqLBf9WH21YQ1pRetdnuKp0rMhj74Uhwi8LQFyU/u2BFU4/Y37LzFwYDscZNuljBycFNaUnz3QvmdX+HZugL5jAhROTwnyzm8ghvQkEAEi8D4JqFSP/n2C+FjyfpGShUf3b7EDTbIwf8Vy9G7VEEY1PitT8S5cuwcx2yrXsRyGUBVRiK+zvHP/AXJK2FFhaqQHQxMLRcS+k7jEsHwxf8jPRvmSIelE4BUBMvQf2dNw7XY8vhs1HDpqOfBq74bvxs0oMl8sb3H5C3T7vp2wNdOBkW0teZOVSzy+yG3wuEmQxLHtkzLCsJ5u6P/dPBl3389lYlj+3D/UZ6P8yVAOROAVARq6p6eBCBABIkAEiMBHTED+1TkfMQQqGhEgAkSACBCBj5UAGfqPtWapXESACBABIkAEGAGVMvT8bJPQiEPYtZWdK1/OgW/lm7XiBwQtnFTOORUvnh/ys2LdRlz/43jxEegqESACRIAIEAElEFAtQ88WgIXt2Ys/zyl2IllZOPCFUucu/4knD++UJflbp+GLiDaE/oind2+8tSwSQASIABEgAkRAFgGVMvSylKTrihH4YdcZ5mCHbalLSSo2IW/keI3wQ/8vW0mdr/z93ws0cnND5C8/CWniEnPRtc8AqaOerz7vhvWrFiCnkMzYhBy0+7y3NI67uytGeQ/E35E0SlEseLpIBIgAEXgPBMjQvwfo7zvLTHb8f3LCU9xn7msT7t8U1Nn723HmsY1547t/W/ielSVB9KO7mDT4S6wKWo1uvYZi86+/w3tob+QyBzU8ZDM58WzL2zTvPti48UcEBK5AktgEIyZOQOqj9zNSIihGgQgQASJABKQEVNLQX74Vj+4983qKvT7vjFPMqxcKeaeLic9EN9bbdGU9UDf2+erzLjgWwZzevObA5ETkNQz2+Q7NW7YU4nXu0Bq/bF4us/r/d+U+Grdqhe0hbD92vizugGX99n3o3OMrQUaPbh0QsSVE6nyDR5u9agO8en6FFvn6fN3bC5dP/Vokn5dpufh+8Wq4tW7D5LhiwshBeNNFDPPBw+RNmBcMd3ZO/rVT+2Xq+jY3Mtn6gOzUONSq15I5UYlgRh04eXIPWjZvi9sP7hYRbWdVVXBG8nWv7tiwbjuuRKdh7/Y1ReJYVTFDDea0xK2ZMyZPnIk0iQ6ex9x/GxUpLREgAkSACCiJgEoemJObm40hX3WFqVUN/HLsHDuvfAXW6VVErZZdhWIb6mli9LcDUNXUFFnsbPyffz2BWUuC8Jl9TZjWbSLE2XnkMpbM8YW7kwP6TJkEA2MzxCcmwdHSsFh0v52/iWlThmF8n47oNZQt0GPnefO2xbw123Bq1wpM8BkNa0cnnLp4A4Frl8KUeRFp7jWIucpk3t0uXoK5RiKmBs5HukQdP4Xvg+/0GTgUVgMVrGrmG+9luH56ByYO94FhNQecjvwLf995s9fLHa6dO38carnqiIw8ibrueWUuVukyXkxjDliy0l6gj+eXWPfjDHzW6ikqZETDvn5/RJ1k3txkeHyzr1YRDs4eiPjtGLwGMUczgjuavMBZPY1PQlj4Nljqi2BRo14ZtaNkRIAIEAEioEwCKmnoXRzN4dmrv1BOl0ZN0O3fG1i7eS2Wt+goOEzhnrHat2kr5WBjx/yhH9uDqKhItGOGPoV5iVu6KgBudc0QuISt4NeQ7bdOIlHDtl9OY+XSKQgYNRhtvxgsdS37OD4LET+HYMPUsXDq0EvIr1ad2jgd9RfCInageTeuYx5CS5NKaNzCVYhjbe+EHl+ext9XzqExM/QPnqTjj5MRWOQzEG36DBLifObUDDsitgh/Fw66bPx8gf9cXI08jH69er9+Wynfk7jf1JwsNKxljbgcI/ywfiW6NG0ILebaNzGZ32Nj8oL7nqKBO1epaeeI81fZSEMWa5Hkl32YPxsl4R8hZCBozgyIX3P9+poo+koEiAARIALviIBKGvrCZWduseHk1BxRh4LyFo4xn+ixzzOwLHgl/oz8HanMf7mRoTF0meFJS08Vkj6OS0VGYhx6jxheopHncU9HPWafaZjt0xdtvxxSBPvth3HQZl32MQEsb/4pFLKNWBeW+6hXfxOhkZ4W1LR1ERuf53P+3uN4Nv+difr1GxeRIeuLa0NH8E95hRfM0KtJclCxghhN3TxxKnwFpqwJwoloTaRwR+KCT/A3Db0sfWYM74fqTdoh4cVLHDt9ESNmzEXQzEw0zPc4KCsdXScCRIAIEIHyJ/CmlSr/PBXOQcLGhXlvkocs1tn0Zm4zJdEXMd3XDxa2Doh5loaxU7+TyhXlRxYz3/SlhZrVDJBRyQHz1v6E6jXrwc65pTQJz5fliFWzZsKgeu0iojTVmUKarPvNPagWE0QiNTasn3czTx8R87YqI3Ix6cvzEjfm3A2sWF0Nk4Z8ga87NIBJDQdoxd9BKt/3Jxj6NwPHcevODZgb6xVpQFmam8DePs/9r3OD+rhw619s3LoBDVszt7BsBIYCESACRIAIvD8CKrkYrzAObncuXjiFOraWgJYOeIfzwe0oDO7RGa7tu8COLQLji8XUNHWkycyNdaClX5nN3Ufk9bpL4GthrI0fVy5FfY+++GbsePxb6AAb+2qmSBNp4sb1C7CxtoaNjY30U8XSWjrEX4J44ZYd802fCg38fla+8wHOXL6JkJAVSH+eNyJQmnxF779MZXMbrEfPjbBeBU04OLBGDGsUabDvqXyRgIwGyX/Rqbh56QS6t27NDL2WzGxzWUNBaGsVtM5kxqQbRIAIEAEiUN4EVLK7FR3/EpcjzyOJLRrbFnEAGY+uYdTEJYIx4vPzFlaO2PrrUZjXcIGBmSXik7KRm8nmlvMDWyeHEcMmImShH/zGDkf7Lp9Dz8gUL1LYkHX2M3i061GEK7dZi6aNxQTWC/ee5Ietq0NgUdsFVU000P0LbwRvD0Hsixy4uLZhQ/XauPfwPjzZqvgKZtXkqh9LU0108voGy0J/wMtsbdRybo64pFwUt3KA22A//6lQS4iFZvJDDJm0WK483owkxt3/7iJXt5L0ljob4LC2ssWLl8nMqLM2Hitv4SAWi9mICRt1yGKtKVQQbt15EI3ca//g+u2HCNmwFvUstODVe0SRdNFP41DxwUMkJL3EvqNnEPf3OYz7blCZvOYVEUxfiAARIAJE4K0JqJShV2O9wCZODXH+6CP4TpjICidB3RpWWLtkAawaugmF1dQAgufNx8JlCzFpViBymVESiSSwqWwIc8vqUiD9uzaHudEqhIVuxIy5gcKMcwVtdfTs7AqPNmxI+bXTf/lagEA/X/R9eA8j/b5DeOgeYUHZzFH9YW1uiL07tyL8wAlBJ1vWQ+/oXFtuQ887trPGDMY640rYunsbUrbtEeQ4VDGBuZW9VGf+hxaz/k2beuDi4TA0adKqyD3FvlTEALZToHDQ1EjBb/tOt63bGQAAIABJREFUISk5CRrc6r/W4xaL2fQCI5WZngYNbRGqVLXBwo3hAPuYV9ZHn/auGDRoGNQq5e1c4MsTjIwtMDskDOAfNo9Rs7oVZk3wQctu5bOQUDEGFJsIEAEiQATITS09A0SACBABIkAEPmICKj9H/xGzp6IRASJABIgAESh3AmToyx0xZUAEiAARIAJE4P0RIEP//thTzkSACBABIkAEyp0AGfpyR0wZEAEiQASIABF4fwTI0L8/9pQzESACRIAIEIFyJ6BShp6f0xIacQi7trLz6VUgcH027TrAttqtUgFtSAUiQASIABEgAooTUC1Dz45YDduzF3+ek+8EudKKm5Keg5Nn/of0xPjSohZ7n3um27Z3P6L+OFns/fK4eOrSPTT3aIVd6xeUh3iSSQSIABEgAp8YAZU6MEfZ7K/fTYTflHE4sC4Y2gaVlS2+XOQ9ehIPcXYO7vx3q1zkK1toBvPr06n/QGQ9/q+I6LX+36F2m57Kzo7kEQEiQASIgIIEPmpDryALlYjeu5ML6ldfB0dbK5XQpzQlctiR+ZnMa6Bvv65wbvcFO1mPBwkszU1LS0r3iQARIAJE4B0QUElDf/lWPLr37I2E2GhUYX7efQYPgHsX5g8+/8jWmPhMeI/+FgmP7oL7TrEw0YP3gN5o270fO9lWTfBH37anJ9SZC1tN9s/Le6wUpZe7A8YHrJd+PxF5DaGb1uPGtcsQM/dslXQ1MKxfd3gOeOUN73RUDDs2tx2yM9NhXcUIE0YOQ4OWnRWqnqSUHCxYsQpnTxxEDjOMTE04VjfDmhXrIGJ+4B/H5aDL561QMV+qewNTBKza9UYe/sE78Nv2oi5z+dGz25bOgWWj1kJ8NiCAn8L3YfeOUCTHx6CygS4G9vKCV1/vIufP57A1CH6BwbhweAuCZ09DXfeub+RX2gUuIyP1BerXckR1uzwPdqWloftEgAgQASLw7giopKHPzc3GkK+6wtSqBn45dg7TF6zAOr2KqNUyzxAZ6mli9LcDUNXUFFm5Yual7gRmLQnCZ/Y1YVq3CXSZk5rta1fj6s1YzPX3wWr/aTC0qytQ1a/IXMvmh51HLmPJHF+4Ozmgz5RJMDA2Q3xiEhwt885yL4hXzVQXI0ZOhkirEjb+vBdjpk/H4W2O0KlaXe6aCli9CVHHt2P2hKkwta6BxOQ0xDy+A5FOnmk3NRJj95Yw5iFWhBHT/ZncxGJlj+7XGYO6NBHuvUiRsHP5x8FGJxkWNesL17gr2XlrtuHUrhWYwM66t3Z0wqmLNxC4dilMmbef5l6DpHK5o7pz549DLVcdkZEny2To05jjoZzMFLzMyEHqyyToVtSXyqc/iAARIAJE4P0TUElD7+JoDs9e/QU6Lo2aoNu/N7B281osb9FRcK3KPdi1b9NWSs/Grg4OHduDqKhItGOGnnf8rS2rIi5Zm/V1c2FVxRzGzMVs4cB7/UtXBcCtrhkCl7BV/hrF+ZLLS2FroQ9XD+a5joWqtvXx+Ven8M+Vs3BWwNDHxMbASE8bTRo3gZaRWb4qDaUqcR8zttZWQm9cXVuXXS/e0BsbVAL/8J70+LlB0E6LRlBwKMT6xoKsx/FZiPg5BBumjoVTBzYKwkKtOrVxOuovhEXsQPNujGu+j3hd1uZZ4D8XVyMPo1+vsjmheZGSyhzeVIHv3JVsYGEZLE31MGbYEDTjHgLJTa20fukPIkAEiMD7IqCShr4wDO5VzsmpOaIOseFqNnQO1gOOfZ6BZcEr8Wfk70hlw/NGhsbQzcpEGhsSlzc8jktFRmIceo8YXqKRf12esb4W69nrgBtuRcKMUT4YPe5vtPX0RHt3V3Tu1AnOzVqV2ZVr2P6zuHBkC9bP94e+tYNUldsP46DNtguMCWC8+KdQyDZi3f1stnou39DzW64NHYVPWYO9ZQUc2PWzkPzRkwSs2bQdk+YsRKihIWwaeZRVLKUjAkSACBABJRFQeUPPyylh49EFncOsbMB78hRIoi9iuq8fLGwdEPMsDWOnvppTl4eNKF+gmPm4VzSosTS5uazrrUBwsDbE/u1hOHE2EkcOH8Goyf5oUssCixeFQD2/N87FydMJvnU/GSuXToPvF+1Rq0WnIlpwVsyhPFbNmgmD6rWL3NNUZ0Mdmq+mLhRQX66olsyd7xTfETjxv2P45dcd8CVDLxc3ikQEiAARKE8CKrWPvriCprN55IsXTqGOrSVz1q6DtAzgwe0oDO7RGa7tu8Cuhj0a1K8HNU2dN5JrseGAXOZfPT315Rv3zI11oKVfmc3vR+T1ct+IofwLmhoidHBviiXzZmLDpr04dT0Gx/eGFsmILy7UYMY4OZWNXhTTmOA8Js6eDidLLXw9zO+NloF9NVOkiTRx4/oF2Fhbw4ZNWRR8qlhavxH/zOWbCAlZgfTnT5VSYKGdQYEIEAEiQARUhoBK9uij41/icuR5JLGFXtsiDiDj0TWMmrhEGObm8/MWVo7Y+utRmNdwgYGZJeKTspGbmfYG1Com+pBoVcTitWvQZ6A6UnPUkJUSjXadvgRbl4YRwyYiZKEf/MYOR/sun0PPyJQtcEuDWvYzePA5ZiWGxRtCYV9JBDuHOlDT0EFk1G1osW1oxsZFt6Hx1fgN6jrj+I6TOLwnHOpGVdiURRzadMjbkx667zSe3foDk/zG4k70k3wNJahiWhk6lQxQ1UQD3b/wRvD2EMS+yIGLK1tboK6New/vw9O9GSqYVZOWircl/PynQi0hFprJDzFk0mKFS/zDz3tgpStCZXNLJKVkYOf+o0xeNDy7jFJYFiUgAkSACBAB5RNQKUOvxrqzTZwa4vzRR/CdMJGVVoK6NaywdskCWDV0E0qvqQEEz5uPhcsWYtKsQORmZbDhbglsKhvC3LLoKnhjtgB8+pS5WBu0EOPZcL+6WIJ+nmzemBtNZlH7d20Oc6NVCAvdiBlzA1nfH6igrY6enV3ZdjpP9k05Ax584RzY7oA1W8KRkrBWKIeJUUWM7O/F9p6/eajM6P498fDfK5izPBhitVz0694KbVjDI4fps/fQHralUB2TFxQ9lnfR+MFo6vWNMPQ/c1R/WLNh9L07tyL8wAmWmwS2libo6Fy7iKHXYo2mpk09cPFwGJo0aSXopUjg5Xr2LBkRYb/gxfNYljQXdWrYYFnADNi4KC5PkbwpLhEgAkSACMhHQMTmdGmwVT5WFIsIEAEiQASIwAdHQKV69B8aPd5EunP/AXIy3pw2KCiLqZEeDE0sPrSikb5EgAgQASLwkRCgHv1bVCRfGNihT29I4qJlShnW0w39v5sn8z7dIAJEgAgQASJQngTI0JcnXZJNBIgAESACROA9E1DOarP3XAjKnggQASJABIgAESieABn64rnQVSJABIgAESACHwUBMvQfRTVSIYgAESACRIAIFE9AKavu+X5qT+/vkHTzEvjZ9J7tWsB3/PfCSXYUiAARIAJEgAgQgfdHQGmL8RKSkpHDHMv8cfUB/GeOwtb502DvqpjP9veHgXImAkSACBABIvBxElBKj56jMdSvJBBq2ZS5S9XVx9Onj2H/cTKjUhEBIkAEiAAR+GAIlMscvUikxg5dpQP3PpingBQlAkSACBCBj5aA0g09P2tdxM6Rp5N1P9pnhgpGBIgAESACHxABpRt6LeZ0xsTCFsfP/Y+5RWc+VSkQASJABIgAESAC742A0hbjFS7BlZvxGPJtD+giC9uWzYelC/MYR4EIEAEiQASIABF45wSU3qPPZVvtVof+hHpVdRG2YSPM6zV954WiDIkAESACRIAIEIE8Ako39OlstP5a1Fn069YFVjVrQ5320tOzRgSIABEgAkTgvRFQuqHPZYvt+X56HW06LOe91SplTASIABEgAkQgn4DSDT2RJQJEgAgQASJABFSHgNINvYT36HOyIRaLVaeUpAkRIAJEgAgQgU+UgNJOxis4Avdc1H2I05NgZWX3iSKlYhMBIkAEiAARUB0CSjH03KnN1+NnCE5tNNk+eu8vOqPqZ81Vp5SkCREgAkSACBCBT5RAueyj/0RZUrGJABEgAkSACKgcAaXP0atcCUkhIkAEiAARIAKfMAEy9J9w5VPRiQARIAJE4OMnQIb+469jKiERIAJEgAh8wgSUauhT0nNw8sz/kJ4Y/0EhPXbub3z59WC4urmhlbsbhg3ui4T//vmgyvCxKHvrfgL6fjtSqAs39jkf8ePHUjQqBxEgAkTgvRBQyqr7As2v302E35RxOLAuGNoGld9LgRTN9MnzHEye5osO9aph6uhFyBFrIebpY+gYmigqiuIrgcDc1Wsgef4Pdm7aCA19I+hV1FWCVBJBBIgAEfh0CSjV0H+IGB/HJkGc9RK+Q4fBtG6BAx6nD7EoH7zOfJtmfFwM2tatAfPqNT748lABiAARIAKqQEAphj4lHWjb0xPqL55Dk/3z8h4rLZuXuwPGB6wXvj9/kY2Q9Wtw5UIkYp9EIzc7i13NxnSf/ujQZ6QQJyY+E96jv0XCo7sQse8WJnrwHtAbbbv3A9TUwL3jBQRvwOWThxEXGwM+92BT1RhjRwxFQ/cuggweklJysGDFKpw9cRA56ak8KRyrm2HNinUQ6RmhqM4a6DliojTtzOFfol2/McL3TKbi6s3bsX/XVmQkJzAZFhgz3Bv1mrWVxi+tXH/9+wzTpo5G4pOHqGqqDyfmtvfg8aMw0MpGwOTxqOvaSSqrpD942ecEbUDksf1ITuDTIxLUq2kt6OPYyF2atDSGPGJpOvM4pclRVrkKFOenKkrYPzGvLApEgAgQASKgFAJKMfS6WsD2tatx9WYs5vr7YLX/NBja1RUU1K+oLVX06fN07NmzA0O6tkaDFkOhU1EfScnJqGNjLo1jqKeJ0d8OYAbRFFm5Yvz86wnMWhKEz+xrsh53E3CnOWcvXoK5RiKmBs5HukQdP4Xvg+/0GTgUVgMVrGoKsgJWb0LU8e2YPWEqTK1rIDE5DTGP70CkU1G4L1tnCUyNDIQ43PB8v3IjzkSEYMLwETCpXhfhTB+fSVOxZbkGrJ3zjGtp5Xr6PAXPn9zEuoUrsPXMDezfG4Kg+Uux6XgU/BfNxU4XJke79CFqXvZzly6hSoVUBExfjJeZImzZvQ9Dx03AtlXLYdmghaB3aQx5nNJ0lkeOssolKF0QGHQRGfoiSOgLESACROBtCCjF0ItY19vasirikrWRy/5ZVTGHsY2NDL1y0a21OywbtSz2vrYm0L7Nq96yjV0dHDq2B1FRkWjHDH1BsDSphMYtXIWv1vZO6PHlafx95Rwa5xv6GNbbN9LTRpPGTaBlZJafrKE0vTw6P4rNxEG2GGzukF7o0m+wkNbZpTG6DryFoB+WY5ETy1+t4Ez/ksslEuWgTp3a8BBb4dS+INSvVwfdtO0w/diPkKQmQySHoS9QvkrlimjIysVDo0ZMnwEPsGrtEiwIZlMPTB95GYLV1dvWhTLLFROfjmfRd1CrXWdpPdEfRIAIEAEi8HYElGLo306Foqljn2dgWfBK/Bn5O1LZVICRoTF0mdvbNDb8LisY6WlBjRnK2Pin0igzRvlg9Li/0dbTE+3dXdG5Uyc4N2tVyDDLkvbq+p1HcdBFJhoXapRosSN+nV1a4uqRYCAjDWxYonRBhWKoifKHpVnPVS2/58qHq/k0RVmCDhtNcWnsgWtHV0v1KQvD4vJWRM7blmuE/3Jc+20X6ltXQtse3xSnDl0jAkSACBCBMhBQKUOflc3OyZ88BZLoi5ju6wcLWwfEPEvD2KnflVo0ETOguRI2iZ0fHKwNsX97GE6cjcSRw0cwarI/mtSywOJFIVDXNy5V3qsIojeMsISP6Zcx8JEEsVi52HmDoUClt2FYuEiKynnbcs3xHYyLrdwx8/tRuHPlLOyatS8jYUpGBIgAESAChQkoddWTloY6GwwWIz31ZZkop2UAD25HYXCPznBt3wV2NezRoH49qGnqlEmepoYIHdybYsm8mdiwaS9OXY/B8b2hcsuyszRBKjQQeeG0NE0GW5x3+dLvqG1TFdBSXC8PF0ucOnERMFDO9j1ukP/88yxqVmPTE5raUBZDReW8bbkqG+qhjasTDK3r4eTvR+SuI4pIBIgAESACJRNQateyiok+JFoVsXjtGvQZqI7UHDVkpUSjXacvS9Yi/y6fW7awcsTWX4/CvIYLDMwsEZ+UjdxMNkSuYFi8IRT2lUSwc6gDNQ0dREbdhhYbIjc2NpVbkqWpJjp6DkLghjXIEOvD1LYewg+eQPrDqxi1fJFC0wByZypHxAs3YhCxPRSm1Wpg7/FzSLkbBe+AqWyoQF2Yn1cGQ2XJkaM40ijCqIC6BlLTUhRJRnGJABEgAkSgBAJKNfTG+sD0KXOxNmghxrMheHWxBP08PYAOPYWtcaUF7uI2eN58LFy2EJNmBSI3KwMikQQ2lQ1hblm9tOTS+3w/NtiK/TVbwpGSsFa4bmJUESP7e8G5HdNFzsANz2zfIQjS10VQ2BbWaNkMe2tTBAcGSFfcyylKqdHEapr4YdsvePH8KSwqV0KAny9qt+wq5KEshsqSU5aCv83USFnyozREgAgQgY+ZALmp/YBqNzsH6DrYBy56TxGwatcHpLl8qvIGWvehvmht8ZKdvbBBvkQUiwgQASJABEokoNQefYk50U2ZBPhCujv3HyCHr+KXEUyN9FDJyELG3Y/jspgN+thY27OzAsLx+PYNaLFjiPkRuBplWAvxcRChUhABIkAE3p4AGfq3Z/jWEtIzgcHjJkESFy1T1rCebug7ep7M+x/LjSnDv8VU/7v46puhwm6HReMHo6kXbbf7WOqXykEEiMC7J0BD9++eOeVIBIgAESACROCdESh9hdw7U4UyIgJEgAgQASJABJRNgAy9somSPCJABIgAESACKkSADL0KVQapQgSIABEgAkRA2QTI0CubaAnyTl26h+YerbBr/YISYtEtIkAEiAARIALKI0Cr7pXHslRJj57EQ8w2w9/571apcVUhwqV/7mPd2mDmFfAPiHKz4VC9CsaPHIFazIlOQSgtTko60LanJ9SZg6LXg4lBDnaHH5PLRe/raek7ESACRIAIyEeAVt3Lx0kpsfiBMNdvXoejrRXUtSsoRWZ5CclkZ/r3GeULZxMxWnfwxItMMYI2/oSUmD9xaMc+aJhUgTxxclmZ/7l1CxJ2ymFBiEvIwuTvJ2Bgm/oYOXXxeztKuLzYkVwiQASIgCoRUElDfyLyGkI3rceNa5chZqfJVNLVwLB+3eE5IM+L3fMX2QhZvwZXLkQi9kk0crOZVUI2pvv0R4c+IwW+3Ait3rwd+3dtRUZyAhyrW2DMcG/Ua9ZWuM8N0JygDYg8th/JCfHsigT1aloLcRwbuQtx5A1JKTlYsGIVzp44iBzmTpef9utY3QxrVqyDSM8Ij+Ny0OXzVihwaOvewLTYk+38g3fgt+1Br2Wbi21L58CyUWvhOj8d76fwfdi9IxTJ8TGobKCLgb284NXXu4jB5I0Kv8BgXDi8BcGzp6Gue94RufKWKS+vHHaMsVia5NxfTzF2pBf2rFwICye3fH1Kj1M4T66Xz8xFePHPr9i6OQKiigaKqERxiQARIAJEQEECKjd0v/PIZSyZ4wt3Jwf0mTIJBsZmiE9MgqOlobRoT5+nY8+eHRjStTUatBjKXMLrIyk5GXVszIU4/KS571duxJmIEEwYPgIm1esi/NcT8Jk0FVuWawjn1OeyOOcuXUKVCqkImL4YLzNF2LJ7H4aOm4Btq5bDskELuVEGrN6EqOPbMXvCVJha10BichpiHt+BKN9XvamRGLu3hCE3R4QR0/2Z3MRiZY/u1xmDujQR7r1IkWCk3zjY6CTDomZ94Rov17w123Bq1wpM8BkNa0cnnLp4A4Frl8K0og6aew2Sys1gh/CcO38carnqiIw8WSZDX9jIc8GPY2OZb8IM6Ju8OqFPnjhSpdgf56Ie4cqp3QgN/J6MfGEw9DcRIAJEoJwIqJSh5/O5S1cFwK2uGQKXrAY0mCs2mSEX3Vq7s55uyzdiPIrNxMGIHzF3SC906TdYuO/s0hhdB95C0A/LscjJlV3J66lWqVwRDRvnGddGjVicAQ+wau0SLAhuKveQckxsDIz0tNGEydEyYu5ihdAw/3+wXjFga20l9MbVtXXZ9eINvbFBJfAP7/WOnxsE7bRoBAWHQqxvLMh6HJ+FiJ9DsGHqWDh16CVcq1WnNk5H/YWwiB1o3q2/4MGOB11tYIH/XFyNPIx+vXoL194m3H74AkuXTMWwnl2gW9WuWFGlxeHlX7JmORrXMEbNJm2KlUEXiQARIAJEQLkEVMrQP45LRUZiHHqPGF6KkS8Zwp1HcdBFJhoXagRoMc94zi4tcfVIMMDPlNcsGEh/JUtHC3BhC82uHWWNDB4nv0decm7AjFE+GD3ub7T19ER7d1d07tQJzs1ayd1QeF1+2P6zuHBkC9bP94e+tYP09u2HcdBmQxFjAtjwPv8UCtlGrLvPpzDyDT2/5drQUfi8bfjr31iMHDMEPVysMWDkdDCXgm+IlCfO9btJePTPOcybM4m1eFiFUCACRIAIEIFyJ6BShl6Ub0DEaq/mhctOQCSclV44yOP+VI1NsPMhckWCg7Uh9m8Pw4mzkThy+AhGTfZHk1oWWLwoBOr5vXEurxj7+EY2t+4nY+XSafD9oj1qtehUjP5ZWDVrJgyq1y5yT1OdlVaTdeOVHP649ghjfL/B120aYOTEucU2wOSJw9Xatn8fTCtko1ajVkrWksQRASJABIiALAIqtY/e3FgHWvqV8fOvEXm9U1lal3LdztIEqdBA5IXT0pgZrLN7+dLvqG1TFZDhDS0rG/jzz7OoWY0NvytoNDU1ROjg3hRL5s3Ehk17cep6DI7vDS2iKW94aDC5yalsjiKXjWO/Frhzm4mzp8PJUgtfD/N7o2VgX80UaSJN3Lh+gXl5s4aNjY30U8XS+o34Zy7fREjICqQzv/VlCdFxGfhuwnAM4Kvj/eYXa+TlicPz5mX735mDaNWwDlBBvyzqUBoiQASIABEoAwGV6tGz9WQYMWwiQhb6wW/scLTv8jn0jEzZwrQ0qGU/g0e7HnIV0dJUEx09ByFwwxpkiPVhalsP4QdPIP3hVYxavihvSD3fzl64EYOI7aEwrVYDe4+fQ8rdKHgHTC0yBF5apos3hMK+kgh2DnWgpqGDyKjb0GKr+I2NTYsk5avxG9R1xvEdJ3F4TzjUjaqw7QFxaNOhpxAvdN9pPLv1Byb5jcWd6Cf5aSWoYloZOpUMUNVEA92/8Ebw9hDEvsiBiyub51bXxr2H9+Hp3gwVzKpJ8+NtCT//qVBLiIVm8kMMmcS2sSkYVm7eBkM8Q8fuX+POvfvS1BV0NGBexUr4Lk8cHu/p80wkP32ANoPZtAwFIkAEiAAReGcEVMrQ81L379oc5karEBa6ETPmBgpL5ipoq6NnZ1d4tPFkRrr0QQg+RD7bdwiC9HURFLYFWSmbYW9tiuDAAGHFfeEgVtPED9t+wQvW67WoXAkBfr6o3VL+rWh84RxyxVizJRwpCWsF0SZGFTGyvxec2+UZ8ML5je7fEw//vYI5y4MhVstFv+6t0IY1YHKghr2H9rDpBnVMXrCqiI4Frlp5uWaO6g9rc0Ps3bkV4QdOsHgS2LIRjI7OtYsYei22jrFpUw9cPByGJk1aFZEnzxc+uhH1VyRSUtQxcMSoIkka1zJki+oikJWrVmqcgvp6+OQZa/xkwdKq+IV88uhEcYgAESACREBxAiq5j17xYiiegq8A7zrYBy56T4vd0664REpBBIgAESACRED1CKhcj15VEPEFeXfuP0AOX30vI5ga6cGw0J5yGdHoMhEgAkSACBCB90aADL0M9Hzx2OBxkyCJi5YRA2xPuRv6fzdP5n26QQSIABEgAkTgfRP4ZIfu3zd4yp8IEAEiQASIwLsgUPrKtnehBeVBBIgAESACRIAIlAsBMvTlgpWEEgEiQASIABFQDQJk6FWjHj4aLfjBRCvWbcT1P45/NGWighABIpDn8TM04hB2bWVHhFP4oAiQoVdidR079ze+/HowXN3c0MrdDcMG90XCf/9Iczh16R6ae7TCrvUL3siVu7J1Yunc8j/TR7+5B/+NROV8gW9B7DjQBz8sGIc0dgBP0y6eOLPrhxJz5YsYN4T+iKd3b5QYr7ibKek5OHnmf0hPjC/udpmu/XM3GS5uzXHn7KEi6T+0uihT4ZWUqCwMlZS1yoopj2dVZQubrxj3+Bm2Zy/+PPdbqaqW9PsqNfEnEEEZz48i72cy9Ep6qJ48z8Hkab6oyZzLrFqyCIuXrkTXL/pBx9BEmsOjJ/EQs9q589+tN3LlrmzDN23G+g2boWNZ44377+MCP6BHXV0d2lragt+AvL+Z559yCtfvJsJvyjikxNxTWg77jp+AiU4W7D5j3ggLhQ+tLhQBcut+Atp83ltooL0eLv1zH8N8J8HVvRVrVLri24FfsdEXfvCS7FAWhrKlfRx3yuNZ/TjI5JWipN/Xx1TOspZFGc+PIu9n2l5X1pp6Ld3j2CSIs17Cd+gwmNYtMCpORWL17uSC+tXXwdE27/jYwje5K1v76rZ5rmx1KrBbSUrSrOxixKwZqFOxEkyNK0PM9NPWrQgjg8plF/iOU/JphMOHd6JbU+YyuKLBB10X8qCLT0zDpm3h2LH1B3YKIQ/siOVCIZPxmLdiKZxNxPgmYBZeZIoRtPEn5oRpAg7t2AcNk6LxedKyMiySMX355AiU9K775GCUU4EVeT+rlKH/699nmDZ1NBKfPERVU304uXjg4PGjMNDKRsDk8ajrmufNjQ9Z/BS+D7t3hCI5PgaVDXQxsJcXvPp6S13DxsRnwnv0t0h4dFfojVqY6MF7QG+07d5POEY3lx1dGxC8AZdPHkYc8yfPhzZsqhpj7IihaOjeRe7GStyVAAAgAElEQVSqSWFD2m17ekL9xXNoMkc6PUdMlKadOfxLtOs3BnxYvsvnrVDgGNe9gWmZT+OTp+xcAX40r19gMC4c3oLg2dNQ113+Y30LF96GOcuxsTJjvXmgspk5KhkYF2HzMi0Xi4LX4PiBXUBWBhzt7KXlLIhYWl0UZagJL++x0jy83B0wPmC98L00OUUUY19u3EvAy5h/0XPiq7MOlFkXSSk5WLBiFc6eOIic9FThtF/H6mZYs2IdRHpG+OXkdSya8Q12hywv1PgDAtf/gmNbZ+PonsMQMSdOz19kI2T9Gly5EInYJ9HI5e6GkY3pPv3Roc9IabFORF5D6Kb1uHHtMsTsRKdKuhoY1q87PAd8J8RZ+/OvePDHAayYNQuTV+Udx1yYiSbzDLxj9VKo81ZbfqhkUgNjR3oh/tEd9ht509CXlaG8v2V56pQ3UNaE7sC+3duQnvSMcZbAzqYa/Hy+hUMT5u+BhdJ+F/L+3kuTI++zWph7cX/Ly6c0fbjs0hjKW/bi9Czu2uVb8ejeszcSYqNRxaQSfAYPgHuXXoJTLXl+X/LqU1q5uG6l/Xb4SGunHm0w8Qs39B09R1qcwVMXIe3GPmzbfgS5zF/InKANiDy2H8kJfNpQgno1rTFmuDccG706Mr20vLhw/qyu3rwd+3dtRUZyAnsfWAhy6jVrK+Qt7/NTmpyCgpT2fi6Ip1KG/unzFDx/chPrFq7A1jM3sH9vCILmL8Wm41HwXzQXO13cIdHSxbw123Bq1wpM8BkNa0cnnLp4A4Frl8KUecVp7jVIKJuhniZGfzuANRhM2ZnsYuYR7wRmLQnCZ/Y12Uu3Cfh809mLl2CukYipgfORLlEXGg++02fgUFgNVLCqWcCoxP91Wddp+9rVuHozFnP9fbDafxoM7eqyNBKYGuX1Ivmw/O4tYcxhnQgjpvuze4klypR1k5/WJ0/ZefoMNld+7vxxqOWqIzLyZJkN/YTBX8FQRyy42F00eRyMDPWk6vHGxIR5y3D99A5MHO4Dw2oOOB35F/6+c6dIEUqrC9kMAf2Kr1zvlibndW67Dh2CWUUJrGo7S28pqy64wIDVmxB1fDtmT5gKU+saSExOQ8zjOxDp5DXpmn7mgBSxDo6d+BV980d5+A/42G+/wK2BA0SVjAS9nj5Px549OzCka2s0aDGUjaLoIyk5GXVszKV67zxyGUvm+MLdyQF9pkyCgbEZ4hOT4GhpKI0zdWhPiLy/QFoGe/qC1kmvF/6jsJHn1x/HxjJ/EhnQl3HCY1kZyvNbZkNEpf5O+TP//cofcSYiBL5DhqA6cwoVHZeG2fOmIubeTcHQy/O7kOf3Lo8ceZ/VYuEXuigPH1V71xWon5ubjSFfdYWpVQ38cuwcpi9YgXV6FVGL+QiR5/clT13wvOT5vZf22zExEDMb4YLj58+j73D2UtTQFH4f//5zAV81rCd4KeWORM9duoQqFVIRMH0xXmaKsGX3PgwdNwHbVi2HZYMWQtFLyyvvWd0oPKsTho+ASfW6CGd2x2fSVGxZriH4WZHn+ZFHTkFdlPR+LvS4sReCCoUj5+9LWrjVl0iSnkmORub/nRgnOX7hoaS5WwNJ7rMnkkexmRKnlm6Sy4e2SzXPzZVIen/3vcRnYEeJJDur2BIlJEskTh6tJUe2rBDuZ2VLJB0GjJBMG/W5NP69mHSJk6urJPKXn4qVUdLFC//ESxq7NpbE/3NBZjSeZ6dvRhXJ8/XIJcVRtOy/X7ouWb16uSSNcSuP8F90msDr2NYgqfjElxJJPXbtxPZgmVm+XhcFEeVhWFioLDk8Tmq6ROLq9ZUkeM6oYvUoiXNBgtLi9JsQIOnVw12SLoNvTg57Ln1nSvr2dJdIMtIEsdfuJElcXJtLbp7eL9WLLXaTOLNrD//4rVhdX7Kkzbr2kEwc3lMiycwoNk7hi7zsbj16SdYFji0x7r8PkiTNO3aVbFo2XSLhP6LXwtswlOe3/Hp+/PvrdfrwaaakgZubZPfaudLorz9j8vwu5Pm9yyOnQAlFn9XXyyoPH0X0KSz/dYbylP11/Yr7XpycTPa67fD1MInvkK5F3r0l/XaKkyPPu/f1cnEdS/vt8Dgb95yVuLo2lKQ8/Fco1o17L4Xf281Te4XvxenDn/3WXw2QTBreQyLJYS9uFkrL68GTDEl99u7b/+MiIT4P6ZkSSds+gyUThnlJ5fDrJT0/isgpyKe0/1WqR1+4BaImyl8nyJo3avke6ySsl3z7YRy0WZNwTEAQ61KxT6GQzRbCgQ97itUR+zwDy4JX4s/I35HKhtWNDI2hm5XJVo+nFklT+IuRnhbUWC8jNr5s/ttlClbSDXnLXpCda0NH8E95hXuP46GNTNSv37jELMpSF8UJVETO1dtxyIx/AK+uo4sTpZRrM0b5YPS4v9HW0xPt3V3RuVMnODdrJZ0+4o9tv8/7Yv60g3h+9zqMHJywk40yWFTKRc2GrnLr8DguFRmJceg9grn4ZT0SZYS//o3FyDFD0MPFGgNGTheGXV8PymIo67fMcyytTu88ioOOJBMtmrd+XT3pd7l+F3zu6bXw+u9dLjns3aLsIIuPvPqUxrA4fV8ve3Fx5LmmwXA4OTVH1CH2Ls5k85j5o1nypC0cpzh9ylKu4vLt6NoIK5dVwvnfD6N1H3tcuHodlcRpsK8r+72lw0ZqXRp74NpRtpWQ+zuRo1z8WdVl78PGjVpK1dBi02XOLi1x9UjwO5dTmIXyn9riSJfhGn/viIv5UbGWC5OWhVWzZsKgeu0ikjXVWSI2FMNdrHpPngJJ9EVM9/WDha0DYp6lYezUvLnMktQRsQZGroT7nlV+KOZdqlAm8pRdIYFvGVkkFEgE1huUKelt6qKwUEXlhB/Yh6oGaqji0KBY3d62LrhQB2tD7N8ehhNnI3Hk8BG2qM0fTWpZYPGiEKjr561lcHN2wNxK5ti7fzu+quaEIwe3w6cDm1euoF+sXsVdzOPMfg9qr+bWi4sn77U/rj3CGN9v8HWbBhg5ca7MxoOyGMr6LctTpzlsfog9YcW+CwrKK9fvQsYjWvj3LpcceSErEE8WH3n0kYehLFWU9a7jer7+e3r9uywdCl8vrM/blOv1vMyM1NGgaQeE7YtA657D8etv+9GkTnWolbKwmHcwBXOjUBAJa8IKh7x6VEgIi6wsOXn5qqyh93CxhMeJi/l0Hksp2VczRZpIEzeuX0BvD7Y4r5gnis/BPLgdhe/7doZr+7yFdWaWrKOlqaMobaXG5w+ABmuIJKeyli+fGCrmxa3GGzjqGkjlk+zcgOaPZnBF5Cl7YYXPXL6Jq5GH8E2v3tA2MlNqWbgwO0sTpLIFiL+f/Q2fs3UPxQVF6kKLdQ9y2YxxeurLN0QpIucla4CfPb0fQ1o2l9kSf9u6KFBQU0OEDu5Nhc8//41Gv4HdcXxvKNp/ndeo1GcbKDp1/RrhexfDwP4MREn30aPnwjfKV9IFc2MdaLFFez//GoH6bdn5Cuz5KGuIjsvAdxOGY0Cb+vDxmy+MfhUXlMWQy5b1W5anTi3NDJHOlrlevXoBrWq5FKeqwr+LYoWwi4r8vkp6VmXJl3VdFh959JGHoax8ZV1X5L3Bz824eOEUGtqyF6zWq/erPL8vWfnz68osF3+FDu07CONGReD6xYu4c+0c/CYMLvb9W6ATb2j8+edZOFZj7032zpYnFLwPIy+cRpc6eaMFfNfK5Uu/o4FN1SJ8Snp+FJEjj148TvG/cnlTv4d4VU000P0LbwRvD0Hsixy4uLLeEVs1ee/hfXi6N0MFs2rQZqObFlaO2PrrUZjXcIEBs/LxSdnIzZTtcvZdFIU/cA3YYqLjO07i8J5wqBuxVc6ZcWjT4dXhOMLK7Rp18cf+0zh5YC9yKhhBLfsZPNr1gDxlLygHb0v4+U+FWkIsNJMfYsikxUovoqWpJjp5fYNloT/gZbY2ajk3R1xSLnstvwqK1EUVE3222LIiFq9dgz4D1ZGao4aslGi06/SlQnX6543HyE2MRrfOk2WW+W3rggtevCEU9pVEsHOoAzUNHURG3Wbb2iQwNjYtku+AHt3wy45gLFk8BZ2a1YNuleoy9SruBltjihHDJiJkoR/8xg5H+y6fQ8/IFC9S0qTPBk+XnpmLR4/uISNDxJ71DDxPTMFdtjBSX08XRvmL7VZu3gZDPEPH7l/jzr370uwq6GjAvIqV9LuyGBZXnoJr8jwb1atWQJ3GbTFr9Y9IVdOHnrktzv15vcjODkV+FyXpo4ickp7VkvJQ5J48+sjDUJE85XlvRMe/xOXI80jKkGBbxAFkPLqGUROXFDGc8vy+StJL2eVyqW0Bw+oN4ce2lRpIEvBZo1ZvZH/hRgwitofCtFoN7D1+Dil3o+AdMFVmY/h1Afx92NFzEAI3rEGGWB+mtvUQfvAE0h9exajli4rwKen5UUTO6zrI+v7BGXregZ85qj+szQ2xd+dWhB/gh31IYMt6lx2dawuGnm8jCp43HwuXLcSkWYHIZdu+RCIJbCobwtxSsZesLHBlvT66f088/PcK5iwPZkOxuejXvRXaMCNeuOc+7pu+mPTfVbaadQnUWZy+LI5HG0+I2K+ntLIX6KXFrG3Tph64eDgMTZq0Kqu6JabjdTFrzGCsM66ErWzrU8q2PUJdOFQxgbmVvZBWkbowZqPZ06fMxdqghRjPpl7UxRL08/QAWENIU0NN7jr9ef8eWFXWgomw+0F2eJu6yOEbMtlujjVbwpGSkLeVzcSoIkb294Jzu6KnGlqba6N52y9x5chPGD7Et0hdy9au6J3+XZvD3GgVwkI3YsbcQDbuwUb/tdXRs7Or8Gzw54cfwjHs24HCHnreo9p/Jp59BuHLNnUwxn+NMKUV9VckUlLUMXDEqCIZNK5liCVrIqS6vTVDOQomz7PBz5dYwXayzFumy7akroWYjYTVrG4nSFfLHxGT550ghzrC4KC8v6+SntXCv2V58pUVRx595GEoS35x10t6b/DRxiZODXH+6CP4TuDbiCWoW8MKa5csgFVDtzfEyfx9CZuZSw7KLheXN2boWMyZMgRftXGGmmHRxjjXRqymiR+2/YIXz5/ConIlBPj5ojbbSSBv4PU123cIgvR1ERS2hXVSNsPe2hTBgQHCivvCoaTnh7/n5ZUjt258tZ68kSkeEVB1AkkpQPueXeHbuQH6jglQdXVVUj9VZ/hfdDq+6N0W2wImwsGdNXIoEAE5CNy4/xIDB7bDQbYdmi+MLQj8rIKug33gove0zOebyJH9e43ywfXo3xUt3vy5c/8BcviKSxnB1EgPhjL2H8tIQpfLmcCFa/cgTolFx0LTIeWc5UcnXtUYbtt3EFV1xTCobIZnSWnYuG0nzPixxvUKTqBUnSqg94bq1AXX5EVKFh7dv8UO1snC/BXL0btVQxjV+Ey1lHwH2pChlwGZLzIZPG4SJHHRMmIAw3q6of93r05dkxmRbrwTAvwlu33fTtia6cDIttY7yfNjy0TVGPLe1oUr/yDkxAFI2PZYNTaV1eizWpi/ci1b46L8BaZvW5/03nhbgspNf+12PL4bNRw6ajnwau+G78bNKHERnnJzVx1pIhq6V53KIE2IABEgAkSACCibQOmrIpSdI8kjAkSACBABIkAE3hkBMvTvDDVlRASIABEgAkTg3RMgQ//umVOORIAIEAEiQATeGYEP1tCfunQPzT1aYdf6Be8MlqpklJKeg5Nn/of0xHhVUYn0IAJEgAgQARUl8MEa+kdP4iFmS3Lv/HdLRdGWn1r8YBS/KeOQEnOv/DIhyUSACBABIvBREPhgt9f17uSC+tXXwdH21bGdH0WNUCGIABEgAkSACCiRgEptr/vr32eYNnU0Ep88RFVTfTi5eODg8aMw0MpGwOTxqOvaCY/jctDl81bSs67dG5gWe5rR8xfZCFm/BlcuRCL2STRyuftaZGO6T3906DNSQChPnEyWbPXm7di/aysykhPgWN0CY4Z7o16zttJqkEcOj8wcccEvMBgXDm9B8OxpqOsu//GKPH0KO7++bU9PqDO3u68HL3cHjA9YL1wuTR/uaKFjvwHoUVtXOBqVB34amltHN6wc1QetevkI1/ge5p/C92H3jlAkx8egsoEuBvbygldf709yL6oAhQIRIAJE4AMjoFI9+qfPU/D8yU2sW7gCW8/cwP69IQiavxSbjkfBf9Fc7HRxh6mRLnZvCWPO30QYMd2f4U4sFvnT5+nYs2cHhnRtjQYthjJ3wvpISk5GHRtzafzS4vDDQ75fuRFnIkIwYfgImFSvi/BfT8Bn0lRsWa4hPb+4NDkFGXKHdOfOH4darjoiI08qbOh12SHm29nxjVdvxmKuvw9WszPADfPPc9ev+MrDkrz6FAsu/yIv+7w123Bq1wpM8BkNa0cnnLp4A4Frl8KUeVlp7jWopOR0jwgQASJABFSEgEoZes5EJMpBnTq14SG2wql9Qahfrw66adth+rEfIUlNZqdh6cLW2krobapr67IUxRv6PL656NbaHZaNWpaAW3acR7GZOBjxI+YO6YUu/QYLMpxdGqPrwFsI+mE5Fjm5FurZypZTkLkus8UL/Ocy17GH0Y+5jlU0cKcJ1pZVEZeszdy55sKqijmMbWxkiCldHxkJhcuP47MQ8XMINkwdC6cOvYRrtVi9nI76C2ERO9C8W3+5vTqVlA/dIwJEgAgQgfIloHKGvqC4aqL8dYKsa6nGfR6yIGH/uFeudxXuPIqDLjLRuFBDQYt5QXJ2aYmrR4IBfg6+TkWF1HFt6Aj+UfVw+2EctHMlGBMQBPBPoZBtxLr7fCpEhi9zVS8b6UcEiAAR+JQIqKyh571XcQmGhN9/N0H0RuPiY3D4xxtPmcK6heJDXhmzsGrWTBhUr10kkqY6g6/5aqqgeAl0lQgQASJABFSBgMpur/NwscSpExcBA5NiOXE7r8GMTXIqW6HGfFSXR7BjPu5ToYHIC6el4vlCtsuXfkdtm6qAlo7C2Z65fBMhISuQznwelzVoaaizgXsx0lNflkkE9/FdQc8Idx89ZisEmZPyYoJ9NVOkiTRx4/oF2Fhbw4ZNERR8qlha8zmWYlLRJSJABIgAEVA1Airboy8NFB/Nb1DXGcd3nMThPeFs7r4KkBmHNkp0T2ppqomOnoMQuGENMsT6MLWth/CDJ5D+8CpGLV+k8Mpz3ibx858KtYRYaCY/xJBJi0srZrH3q5joQ6JVEYvXrkGfgepIzVFDVko02nX6stj4r18UM3Zd23shdMV5hP2wCrVc3PA0IQuahSJWNdFA9y+8Ebw9BLEvcuDi2oYvisC9h/fh6d4MFcyqvS6WvhMBIkAEiIAKEvhgDT1nObp/Tzz89wrmLA+GmLmv7Ne9Fdq068EMsHIGKnindbbvEATp6yIobAszppthb22K4MAA6Yp7RepUi1nSpk09cPFwGJo0aaVI0iJxjfWB6VPmYm3QQoyfPAXqYgn6eXoAvJEjZ9kHebZEYrwP1u8OQ3bYbiY/F45VzWBWzU7Ii5d95qj+sDY3xN6dWxF+4AS7KoEtG+Xo6FybDH2Za48SEgEiQATeLQGV2kf/botOuREBIkAEiAAR+PgJKKfr+/FzohISASJABIgAEfggCZCh/yCrjZQmAkSACBABIiAfATL08nGiWESACBABIkAEPkgCZOg/yGojpYkAESACRIAIyEeADL18nCgWESACRIAIEIEPkgAZ+g+y2khpIkAEiAARIALyEVDKPnruftXT+zsk3bwEdmgbPNu1gO/478t0cpx8alMsIkAEiAARIAJEQB4CSttHn5CUjJysTPxx9QH8Z47C1vnTYO/aWR4dKA4RIAJEgAgQASJQTgSU0qPnuhnqVxJUbNnUGNDVx9Onj2FfTkqTWCJABIgAESACREA+AuUyRy9iLma5S1kKRIAIEAEiQASIwPsloHRDz89IF7Hz1j8GV67vt2oodyJABIgAESACb09A6YZeSwMwsbDF8XP/A9icPQUiQASIABEgAkTg/RFQ2mK8wkW4cjMeQ77tAV1kYduy+bB0YZ7VKBABIkAEiAARIALvnIDSe/S5bKvd6tCfUK+qLsI2bIR5vabvvFCUIREgAkSACBABIpBHQOmGPp2N1l+LOot+3brAqmZtqGvpEGsiQASIABEgAkTgPRFQuqHPZYvt+X56HW0y8O+pTilbIkAEiAARIAJSAko39MSWCBABIkAEiAARUB0CSjf0Et6jz8mGWCxWnVKSJkSACBABIkAEPlECSjsZr+AI3HNR9yFOT4KVld0nipSKTQSIABEgAkRAdQgoxdBzpzZfj58hOLXRZPvovb/ojKqfNVedUpImRIAIEAEiQAQ+UQLlso/+E2VJxSYCRIAIEAEioHIElD5Hr3IlJIWIABEgAkSACHzCBMjQf8KVT0UnAkSACBCBj58AGfpCdZyRBaxYtxHX/zj+Udc8P71w064DCA9d9VGXkwpHBIgAESACgFIN/alL99DcoxV2rV9Qbmwfx+XAyc0Nbvmf6aN7Ki0vfqrfhtAf8fTuDaXJLIugW/cT0PfbkXDNL+P5iB/LIkZmGn6o0ba9+xH1x0mZceiG6hDgi127fvud8My39nDDinmTgYw01VGQNCECREClCShl1X1BCR89iYc4Owd3/rtVboU2NRIjfNNmsK36GP39HJZPUrnl9b4Ez129BpLn/2Dnpo3Q0DeCXkXd96XKB5XvgTP/Yf6UgcXqLFZLxskDp/AsWw99h/RD6tOHQjx1dtyDS/1aGD1sGKxqOxdJyw1sr9FT8PSvM/D36Y02fUYWub8y7CjC18x+Lb80HArbigpWNfHkeS46erqjW8NqmLN0MyDO+7kN91+JCo+PYcmaCNbULr2tLWZRQpfMEU6c/OPqA/jPHIUuLU/A3rVzsWWli0SACBCBwgSUauh7d3JB/err4GhrVW6U+YvZvrotWHsC6joVWD4fl6HnxiU+LgZt69aAefUa5cbxYxTsxgxqzZ9+gogV7uCZm9ixfgY2LA+G2MAEYjU2jKFTCVmJEjxlRv77AV6o5dYVT+KTsG7zNnwzahh+3boL2ubWUjR3H6fi4T+/o0lDN4T+shttvvgW0NCS3u/XtRk6N/0JhwrlpW5YGboWRZ//k5fv4t/I31CjeYcyYzfUrySkbdnUGNDVZ2V4DPsyS6OERIAIfEoElGLo+XB6l89boWI+OfcGpghYtasIRz4vHBC8AZdPHkZcbIwwZ2BT1RhjRwxFQ/cuReKeiLyG0E3rcePaZYjZUXuVdDUwrF93eA74Tq664XPtHfsNQI/auhjjv0ZIk5QCuHV0w8pRfdCql49w7WVaLhYFr8HxA0zXrAw42tlLy1CQEW9Q/BS+D7t3hCI5PgaVDXQxsJcXvPp6s97Yq9P/uIH2CwzGhcNbEDx7Guq6d5VL19cj8ZMFJeyfWI6e3utp+XfOeU7QBkQe24/khHh2RYJ6Na0xZrg3HBu5F0lyOioGHm3aITszHdZVjDBh5DA0aPmqlxgTnwnv0d8i4dFdwXhamOjBe0BvtO3eT9oTTUrJwYIVq3D2xEHkpKcKHVTH6mZYs2IdRHpGQn7viiF/TirZ5R3UVPk2q3DkwN7WBmpG5oXKzQCxYF3FDI6ODnBkf9vY1cfnX7XF1Qun0KjbAGncnYeOwsmqEkaPnoi+gzvj2e1rMK71qtdvrF8R/CM7rzxMzvVaYNayBdjS0BXQ5o3TtwsikZrwjFAgAkSACMhDQCmGng+n794ShtwcEUZM92f5Jr6RN58XPnvxEsw1EjE1cD7SJeqCAfWdPoMNddYQhjp52HnkMpbM8YW7kwP6TJkEA2MzxCcmwdHS8A2Zb3OBG+YJ85bh+ukdmDjcB4bVHHA68i/8feeOVCw3uvPWbMOpXSswwWc0rB2dcOriDQSuXQrTijpo7jVIGjeDze+fO38carnqiIw8WWZDLwhkGYvKauiZzucuXUKVCqkImL4YLzNF2LJ7H4aOm4Btq5bDskELqc7VTHUxYuRkiLQqYePPezFm+nQc3uYInarVhTiGepoY/e0AVDU1RVauGD//egKzlgThM/uaMK3bRIgTsHoToo5vx+wJU2FqXQOJyWmIeXwHIp28Zt97YygtZel/6Ghpgpv/9Ix0aWS+XuPw4XDM6d0TtjZGqGzrhIOHd6N/IUNfumQeIxfeX3WH9+I7OLU/DO5fsAbiWwQRa3HxZ0PCwVIgAkSACMhBQCmGng+n21pb5Q2na/P55DcNfYEuliaV0LgF69mwYG3vhB5fnsbfV86hMTP0Kew9u3RVANzqmiFwyWo2TKopRxHKFuXBk3T8cTICi3wGsrnXQYKQz5yaYUfEFqnAx/FZiPg5BBumjoVTh17C9Vp1auN01F8Ii9iB5t36S+dddbWBBf5zcTXyMPr16i2VoegfMfHpeBZ9B7Xavd38a5XKFdGwcZ4xbtSoMboOeIBVa5dgQXBTplLeSISthT5cPdoIKla15b3aU/jnylk45xt6bYa/fZu20iLY2NXBoWN7EBUViXb5hj6Gjc4Y6WmjCctLy8gsP25DaZr3wVAR5vEJKVj5wxYYqKWhYRMPadIb9xKQ9exfNGoxF+rsV9KhrRf27Q5A/xFsEZyCrpcN9cQYOmQSlq2fCPeOfRRR7424WuzkSRMLWxw/9z+4eQ4q19/IG5nTBSJABD5IAkox9GUtuZGeFtRYwyA2/qkg4nFcKjIS49B7xPByf4HdexwPbWSifv3GMtW//TAO2mwoYkxAEOu6sk+hkG3EelTZbI4gf4EVv+Xa0FH4lDWM8F+Oa7/tQn3rSmjb45uyinkjnQ6bVnZp7IFrR1njia/W1iyYZHkV1Vhfi/XsdcANd0GIfZ6BZcEr8Wfk70h98RxGhsbQZQvC0tgQfUGYMcoHo8f9jbaenmjv7orOnTrBuVkr6bTGu2b4Rn6JpuEAAAihSURBVOFlXPANXAs2NCPc1Wej6WsXLUaFaq/WRPxy7Chc7MygZVJViNPBrQXCfsjE83s3YOTgJEOqrMsSfNGhGX7cbIlTB3e81VYX3qieM2kGhnzbA8dat8S2ZfNh6fKqgSJLA7pOBIjAp0tAqYaeDysqGvh8Y64kb95UlC9AXGjuW1F5BfHV2PBmJjfEMkJeXiJI+KS2jJA3PJqFVbNmwqB67SKxNNVZYTVZN16JYY7vYFxs5Y6Z34/CHdaztmvWXmnSOY/SRnvVGPfcXLYogYUstqvBe/IUSKIvYrqvHyxsHRDzLA1jpxZdJ+FgbYj928Nw4mwkjhw+glGT/dGklgUWLwqBur5x/hDzu2MoL7DJQ75EbXdPrN95BH8d/xHV7V410Pgaj9OnDyAzOgFurV4ZUW1o4syZI+iusKEHKrBHZcg34xC8cTIq13g1SiKvvgXx+OO6OvQn1Kuqi9mzl8Hc2lZRERSfCBCBT4xA6Xt7FADC7bwGM37JqWwMPt9gKJAc5sY60NKvzOaC2bajEow0l6nGMhOrayCVT46/Zqx5r6cCWwj2//buPSaKI44D+PfgeFq0vEHgOMJDRVErpCKlKFoVQQUfwUQxqeCraksrUapWxdSKzxhrFFOLqb0CNjRqiTW1TfHVQqutSjQRW1JEDRRFkVdB5dHZu3pIOZRTrNzxXcI/u3O7M5/dZHZnfjNTfKNUxGOJGkvH5u3uiL9hhtN5P+g4qtnl4+GEepk5Ci+fhdLTE0qlUvvf191TdJa2fbP58dwVpKXtQMMdTQtFhyfu4ICDbW+MCX0Ftp4BOHH6uw5S6b9bqrTPn8+Dn4doWu/ky0n9PeBaUQHip0QidFwUvH19MHRIAEzMrdplwNxMhvEjg7Ftwxqk78/ByctlyM1RqdP934btMtfBDldHO3h5eWLFknlo6OOH3aLL6OFzVFJWh5qyQuxJTYFKpfr3PwNDRk/FkePieXnCs9nBJREz+lWUtzjg93MnO0ryxP1S7MClgjzMmhQFhZ8/5Hp2IzzxAkxAAQoYnUCXftFL8WNDBwUi98sTOHYoG3K7vsD9WxgzvnOT2oj4Nry1YBnSNicj+b2FGBc1Fb3tnFBdVw+TxtsIHztFewPU0d2+g3DmyCkxPjoHTb3stGmkcccTx8VAteNn0dy6EwOCXkd55QPxPda6uTuZY0LMHGxX7UVtoyUGBIbgVlVzmzRujmaYLIKndh1Iw83qJgSFiv5suSWuXi9B9MgR6OXsoT2h9G6TnLISJpU3YV5zHQnLtz7VwyK9O6hfYOqlqPGn384WluHwARWcRHN0Tm4+6ooLMH/9Sk1Xg+aj/bEnl/rnXRX9kfnN93DxDcLLzu6oqGpE8/22E7VsTVfBx0YG734DYWJmhV8KimAhIsLt7Z3U538Rho8t2H8OSs/c+0vXYlXSDMyYdgYuAcE4dfY8HCybMXCY6Nax1gxrk372xqgIbM39HPcqSmEhhuHdrqpFpeh2qvjrhjhqiqLiq5DfrYWXhwIyHfElNiJ8JTZ2AQ7sThHpW4fp6ZNfKahVGk9vZdn+hUuf8zAtBSjQcwS6tKKX2N6Om4brf1zAh9L4ZZNmzJo8CmPUFXTnGg/iJobAxW4nMlT7sPqjjeqwsV6WckyLDBVDwaLbTDCydM5MLP/zIj7YtA1yca2Z4loP07wZHYa7FYvw6cEMNGYcFGdpRn83Zzh7eKvvrlShrnsnHp/Y2yDzYBbqsg6JvS3o19cRLgofbZo1S+Lg6WKLnK8ykX30uDqNl2gNiAj0b1PRi8BtBAeH49djGRg+fJT698+yPWtUtamJOfZmfY1q0brg6mCD9cmJ8A/r/JA/abnhXRtSsXn7ZixftxHNYvihTNYCpYMtXNw1UfnSyAWIaPw9X2SjrlLT3+1o9xIWx8UgcKzm5U5yflGGnfUPG+YFlwGh2PRxqohJOIxvjx+Fv9K13VC4Qb5K1IsBmJdF8OjQCE9kHMl/ZMIcayS8u0xcsnXCHF3Xj50Qjv37tug6xH0UoAAFnosAl6l9LqxPf1Kp8pw8LxGjXWuRtD5d7xNJY9Ynxi9CUO/ydnMZ6H0y/qDbCdSIOMjw6AjsXByLETHx3S5/zBAFKND9BLr8i777FdGwciR1Oyg9fcRY+GyUFhXCwtZRPQWuGftiDetGdnFuK6tq1E32+QUlMG2ogkKhaZnq4svwdBSggBEKsKLvhjd1xcK5WJlSjNg589Qz0m1JikewiCfg1jMFpFae2UmrUXXlN0hdKvOnR8JtcEjPxGCpKUABvQXYdK83GX9AAQpQgAIUMByBzkXIGU55mFMKUIACFKAABR4RYEXPx4ECFKAABShgxAKs6I345rJoFKAABShAAVlUQmKLFORjJsLyose+hsSktXov2kFGClCAAhSgAAW6p4Dszt3qFmnYzpmL15CyZgkyU1fBJ/TZVk7rnkVlrihAAQpQgAI9T0Bu20czxWdYsL2Y7rMPystLoZkXrudhsMQUoAAFKEABYxNo00cvrSTXIv64UYACFKAABShgHALail6ak1ymXsqUFb1x3FqWggIUoAAFKPDISjMWYsYtR1cv5Ob/JBYjF2thcqMABShAAQpQwOAF2syMd+FKBRLmToE1HiBreyrcg8INvoAsAAUoQAEKUKAnC2ib7pvFfNq7VZ8hwM0aGen71Otyc6MABShAAQpQwLAFtBV9g2itv1SQh1mToqDw84ecq6UZ9p1l7ilAAQpQgAJCoPWLXsTgSePprSytCEMBClCAAhSggJEIcApcI7mRLAYFKEABClBAl4C2opdG1TU1NcLU1FRXOu6jAAUoQAEKUMAABeSVVTXqJvv8ghKYNlRBofA2wGIwyxSgAAUoQAEK6BKQz05aDWlRG3Mxjn7+9Ei4DQ7RlY77KEABClCAAhQwQIE24+gNMP/MMgUoQAEKUIACjxH4B/OmfRFAGTx7AAAAAElFTkSuQmCC" alt="" />     。。aaarticlea/png;base64,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" alt="" width="250" height="330" />

2：

（1）介绍 *_train_test.prototxt文件与 *_deploy.prototxt文件的不http://blog.csdn.net/sunshine_in_moon/article/details/49472901

（2）生成deploy文件的Python代码：http://www.cnblogs.com/denny402/p/5685818.html

*_train_test.prototxt文件：这是训练与测试网络配置文件

 

*_deploy.prototxt文件：这是模型构造文件

 

在博文http://www.cnblogs.com/denny402/p/5685818.html     中给出了生成 deploy.prototxt文件的Python源代码，但是每个网络不同，修改起来比较麻烦，下面给出该博文中以mnist为例生成deploy文件的源代码，可根据自己网络的设置做出相应修改：（下方代码未测试）

# -*- coding: utf-8 -*-

from caffe import layers as L,params as P,to_proto
root='/home/xxx/'
deploy=root+'mnist/deploy.prototxt'    #文件保存路径

def create_deploy():
    #少了第一层，data层
    conv1=L.Convolution(bottom='data', kernel_size=5, stride=1,num_output=20, pad=0,weight_filler=dict(type='xavier'))
    pool1=L.Pooling(conv1, pool=P.Pooling.MAX, kernel_size=2, stride=2)
    conv2=L.Convolution(pool1, kernel_size=5, stride=1,num_output=50, pad=0,weight_filler=dict(type='xavier'))
    pool2=L.Pooling(conv2, pool=P.Pooling.MAX, kernel_size=2, stride=2)
    fc3=L.InnerProduct(pool2, num_output=500,weight_filler=dict(type='xavier'))
    relu3=L.ReLU(fc3, in_place=True)
    fc4 = L.InnerProduct(relu3, num_output=10,weight_filler=dict(type='xavier'))
    #最后没有accuracy层，但有一个Softmax层
    prob=L.Softmax(fc4)
    return to_proto(prob)
def write_deploy():
    with open(deploy, 'w') as f:
        f.write('name:"Lenet"\n')
        f.write('input:"data"\n')
        f.write('input_dim:1\n')
        f.write('input_dim:3\n')
        f.write('input_dim:28\n')
        f.write('input_dim:28\n')
        f.write(str(create_deploy()))
if __name__ == '__main__':
    write_deploy()

用代码生成deploy文件还是比较麻烦。我们在构建深度学习网络时，肯定会先定义好训练与测试网络的配置文件——*_train_test.prototxt文件，我们可以通过修改*_train_test.prototxt文件 来生成 deploy 文件。以cifar10为例先简单介绍一下两者的区别。

（1）deploy 文件中的数据层更为简单，即将*_train_test.prototxt文件中的输入训练数据lmdb与输入测试数据lmdb这两层删除，取而代之的是，

1. layer {
2. name: "data"
3. type: "Input"
4. top: "data"
5. input_param { shape: { dim: 1 dim: 3 dim: 32 dim: 32 } }
6. }

 

注：shape: { dim: 1 dim: 3 dim: 32 dim: 32 }代表含义：

 

shape {
  dim: 1  #num，可自行定义
  dim: 3  #通道数，表示RGB三个通道
  dim: 32   #图像的长和宽，通过 *_train_test.prototxt文件中数据输入层的crop_size获取
  dim: 32

 

（2）卷积层和全连接层中weight_filler{}与bias_filler{}两个参数不用再填写，因为这两个参数的值，由已经训练好的模型*.caffemodel文件提供。如下所示代码，将*_train_test.prototxt文件中的weight_filler、bias_filler全部删除。

layer {  # weight_filler、bias_filler删除

name: "ip2"

type: "InnerProduct"

bottom: "ip1"   top: "ip2"

param {

lr_mult: 1   #权重w的学习率倍数

}

param {     lr_mult: 2    #偏置b的学习率倍数

}

inner_product_param {     num_output: 10

weight_filler {       type: "gaussian"       std: 0.1     }    

bias_filler {       type: "constant"     }

}

}

删除后变为

1. layer {
2. name: "ip2"
3. type: "InnerProduct"
4. bottom: "ip1"
5. top: "ip2"
6. param {
7. lr_mult: 1
8. }
9. param {
10. lr_mult: 2
11. }
12. inner_product_param {
13. num_output: 10
14. }
15. }

 

（3）输出层的变化  

     1）没有了test模块测试精度 ，将该层删除     

     2）输出层

 

1）*_deploy.prototxt文件的构造和*_train_test.prototxt文件的构造最为明显的不同点是，deploy文件没有test网络中的test模块，只有训练模块，即将*_train_test.prototxt中最后部分的test模块测试精度删除，即将如下代码删除。

 

1. layer {                                  #删除该层
2. name: "accuracy"
3. type: "Accuracy"
4. bottom: "ip2"
5. bottom: "label"
6. top: "accuracy"
7. include {
8. phase: TEST
9. }
10. }

2） 输出层

*_train_test.prototxt文件

1. layer{
2. name: "loss"   #注意此处层名称与下面的不同
3. type: "SoftmaxWithLoss"  #注意此处与下面的不同
4. bottom: "ip2"
5. bottom: "label"    #注意标签项在下面没有了，因为下面的预测属于哪个标签，因此不能提供标签
6. top: "loss"
7. }

 

*_deploy.prototxt文件

[python]

1. layer {
2. name: "prob"
3. type: "Softmax"
4. bottom: "ip2"
5. top: "prob"
6. }

注意在两个文件中输出层的类型都发生了变化一个是SoftmaxWithLoss，另一个是Softmax。另外为了方便区分训练与应用输出，训练是输出时是loss，应用时是prob。

下面给出CIFAR10中的配置文件cifar10_quick_train_test.prototxt与其模型构造文件  cifar10_quick.prototxt 直观展示两者的区别。

cifar10_quick_train_test.prototxt文件代码

cifar10_quick_train_test.prototxt文件代码

name: "CIFAR10_quick"
 layer {               #该层去掉
  name: "cifar"
   type: "Data"
   top: "data"
   top: "label"
   include {
     phase: TRAIN
   }
   transform_param {
     mean_file: "examples/cifar10/mean.binaryproto"
   }
   data_param {
     source: "examples/cifar10/cifar10_train_lmdb"
     batch_size: 100
     backend: LMDB
   }
 }
 layer {             #该层去掉
  name: "cifar"
   type: "Data"
   top: "data"
   top: "label"
   include {
     phase: TEST
   }
   transform_param {
     mean_file: "examples/cifar10/mean.binaryproto"
   }
   data_param {
     source: "examples/cifar10/cifar10_test_lmdb"
     batch_size: 100
     backend: LMDB
   }
 }
 layer {                        #将下方的weight_filler、bias_filler全部删除
  name: "conv1"
   type: "Convolution"
   bottom: "data"
   top: "conv1"
   param {
     lr_mult: 1
   }
   param {
     lr_mult: 2
   }
   convolution_param {
     num_output: 32
     pad: 2
     kernel_size: 5
     stride: 1
     weight_filler {
       type: "gaussian"
       std: 0.0001
     }
     bias_filler {
       type: "constant"
     }
   }
 }
 layer {
   name: "pool1"
   type: "Pooling"
   bottom: "conv1"
   top: "pool1"
   pooling_param {
     pool: MAX
     kernel_size: 3
     stride: 2
   }
 }
 layer {
   name: "relu1"
   type: "ReLU"
   bottom: "pool1"
   top: "pool1"
 }
 layer {                         #weight_filler、bias_filler删除
  name: "conv2"
   type: "Convolution"
   bottom: "pool1"
   top: "conv2"
   param {
     lr_mult: 1
   }
   param {
     lr_mult: 2
   }
   convolution_param {
     num_output: 32
     pad: 2
     kernel_size: 5
     stride: 1
     weight_filler {
       type: "gaussian"
       std: 0.01
     }
     bias_filler {
       type: "constant"
     }
   }
 }
 layer {
   name: "relu2"
   type: "ReLU"
   bottom: "conv2"
   top: "conv2"
 }
 layer {
   name: "pool2"
   type: "Pooling"
   bottom: "conv2"
   top: "pool2"
   pooling_param {
     pool: AVE
     kernel_size: 3
     stride: 2
   }
 }
 layer {                         #weight_filler、bias_filler删除
  name: "conv3"
   type: "Convolution"
   bottom: "pool2"
   top: "conv3"
   param {
     lr_mult: 1
   }
   param {
     lr_mult: 2
   }
   convolution_param {
     num_output: 64
     pad: 2
     kernel_size: 5
     stride: 1
     weight_filler {
       type: "gaussian"
       std: 0.01
     }
     bias_filler {
       type: "constant"
     }
   }
 }
 layer {
   name: "relu3"
   type: "ReLU"
   bottom: "conv3"
   top: "conv3"
 }
 layer {
   name: "pool3"
   type: "Pooling"
   bottom: "conv3"
   top: "pool3"
   pooling_param {
     pool: AVE
     kernel_size: 3
     stride: 2
   }
 }
 layer {                       #weight_filler、bias_filler删除
  name: "ip1"
   type: "InnerProduct"
   bottom: "pool3"
   top: "ip1"
   param {
     lr_mult: 1
   }
   param {
     lr_mult: 2
   }
   inner_product_param {
     num_output: 64
     weight_filler {
       type: "gaussian"
       std: 0.1
     }
     bias_filler {
       type: "constant"
     }
   }
 }
 layer {                              # weight_filler、bias_filler删除
  name: "ip2"
   type: "InnerProduct"
   bottom: "ip1"
   top: "ip2"
   param {
     lr_mult: 1
   }
   param {
     lr_mult: 2
   }
   inner_product_param {
     num_output: 10
     weight_filler {
       type: "gaussian"
       std: 0.1
     }
     bias_filler {
       type: "constant"
     }
   }
 }
 layer {                                  #将该层删除
  name: "accuracy"
   type: "Accuracy"
   bottom: "ip2"
   bottom: "label"
   top: "accuracy"
   include {
     phase: TEST
   }
 }
 layer {                                 #修改
  name: "loss"       #---loss  修改为  prob
   type: "SoftmaxWithLoss"             # SoftmaxWithLoss 修改为 softmax
   bottom: "ip2"
   bottom: "label"          #去掉
  top: "loss"
 }

以下为cifar10_quick.prototxt

layer {               #将两个输入层修改为该层
  name: "data"
   type: "Input"
   top: "data"
   input_param { shape: { dim: 1 dim: 3 dim: 32 dim: 32 } }      #注意shape中变量值的修改，CIFAR10中的 *_train_test.protxt文件中没有 crop_size
 }
 layer {
   name: "conv1"
   type: "Convolution"
   bottom: "data"
   top: "conv1"
   param {
     lr_mult: 1   #权重W的学习率倍数
}
   param {
     lr_mult: 2   #偏置b的学习率倍数
  }
   convolution_param {
     num_output: 32
     pad: 2   #加边为2
    kernel_size: 5
     stride: 1
   }
 }
 layer {
   name: "pool1"
   type: "Pooling"
   bottom: "conv1"
   top: "pool1"
   pooling_param {
     pool: MAX    #Max Pooling
    kernel_size: 3
     stride: 2
   }
 }
 layer {
   name: "relu1"
   type: "ReLU"
   bottom: "pool1"
   top: "pool1"
 }
 layer {
   name: "conv2"
   type: "Convolution"
   bottom: "pool1"
   top: "conv2"
   param {
     lr_mult: 1
   }
   param {
     lr_mult: 2
   }
   convolution_param {
     num_output: 32
     pad: 2
     kernel_size: 5
     stride: 1
   }
 }
 layer {
   name: "relu2"
   type: "ReLU"
   bottom: "conv2"
   top: "conv2"
 }
 layer {
   name: "pool2"
   type: "Pooling"
   bottom: "conv2"
   top: "pool2"
   pooling_param {
     pool: AVE   #均值池化
    kernel_size: 3
     stride: 2
   }
 }
 layer {
   name: "conv3"
   type: "Convolution"
   bottom: "pool2"
   top: "conv3"
   param {
     lr_mult: 1
   }
   param {
     lr_mult: 2
   }
   convolution_param {
     num_output: 64
     pad: 2
     kernel_size: 5
     stride: 1
   }
 }
 layer {
   name: "relu3"
   type: "ReLU"  #使用ReLU激励函数，这里需要注意的是，本层的bottom和top都是conv3>
   bottom: "conv3"
   top: "conv3"
 }
 layer {
   name: "pool3"
   type: "Pooling"
   bottom: "conv3"
   top: "pool3"
   pooling_param {
     pool: AVE
 kernel_size: 3
     stride: 2
   }
 }
 layer {
   name: "ip1"
   type: "InnerProduct"
   bottom: "pool3"
   top: "ip1"
   param {
     lr_mult: 1
   }
   param {
     lr_mult: 2
   }
   inner_product_param {
     num_output: 64
   }
 }
 layer {
   name: "ip2"
   type: "InnerProduct"
   bottom: "ip1"
   top: "ip2"
   param {
     lr_mult: 1
   }
   param {
     lr_mult: 2
   }
   inner_product_param {
     num_output: 10
   }
 }
layer {
   name: "prob"
   type: "Softmax"
   bottom: "ip2"
   top: "prob"
 }

3：

将train_val.prototxt 转换成deploy.prototxt

1.删除输入数据（如：type:data...inckude{phase: TRAIN}），然后添加一个数据维度描述。

1. input: "data"
2. input_dim: 1
3. input_dim: 3
4. input_dim: 224
5. input_dim: 224
6. force_backward: true

2.移除最后的“loss” 和“accuracy” 层，加入“prob”层。

[plain]

1. layers {
2. name: "prob"
3. type: SOFTMAX
4. bottom: "fc8"
5. top: "prob"
6. }

如果train_val文件中还有其他的预处理层，就稍微复杂点。如下，在'data'层，在‘data’层和‘conv1’层(with bottom:”data”  / top:”conv1″). 插入一个层来计算输入数据的均值。

1. layer {
2. name: “mean”
3. type: “Convolution”
4. <strong>bottom: “data”
5. top: “data”</strong>
6. param {
7. lr_mult: 0
8. decay_mult: 0
9. }
10. …}

在deploy.prototxt文件中，“mean” 层必须保留，只是容器改变，相应的‘conv1’也要改变 ( bottom:”mean”/ top:”conv1″ )。

[plain]

1. layer {
2. name: “mean”
3. type: “Convolution”
4. <strong>bottom: “data”
5. top: “mean“</strong>
6. param {
7. lr_mult: 0
8. decay_mult: 0
9. }
10. …}

4：

生成deploy文件

如果要把训练好的模型拿来测试新的图片，那必须得要一个deploy.prototxt文件，这个文件实际上和test.prototxt文件差不多，只是头尾不相同而也。deploy文件没有第一层数据输入层，也没有最后的Accuracy层，但最后多了一个Softmax概率层。

这里我们采用代码的方式来自动生成该文件，以mnist为例。

deploy.py

# -*- coding: utf-8 -*-

from caffe import layers as L,params as P,to_proto
root=‘/home/xxx/‘
deploy=root+‘mnist/deploy.prototxt‘    #文件保存路径

def create_deploy():
    #少了第一层，data层
    conv1=L.Convolution(bottom=‘data‘, kernel_size=5, stride=1,num_output=20, pad=0,weight_filler=dict(type=‘xavier‘))
    pool1=L.Pooling(conv1, pool=P.Pooling.MAX, kernel_size=2, stride=2)
    conv2=L.Convolution(pool1, kernel_size=5, stride=1,num_output=50, pad=0,weight_filler=dict(type=‘xavier‘))
    pool2=L.Pooling(conv2, pool=P.Pooling.MAX, kernel_size=2, stride=2)
    fc3=L.InnerProduct(pool2, num_output=500,weight_filler=dict(type=‘xavier‘))
    relu3=L.ReLU(fc3, in_place=True)
    fc4 = L.InnerProduct(relu3, num_output=10,weight_filler=dict(type=‘xavier‘))
    #最后没有accuracy层，但有一个Softmax层
    prob=L.Softmax(fc4)
    return to_proto(prob)
def write_deploy():
    with open(deploy, ‘w‘) as f:
        f.write(‘name:"Lenet"\n‘)
        f.write(‘input:"data"\n‘)
        f.write(‘input_dim:1\n‘)
        f.write(‘input_dim:3\n‘)
        f.write(‘input_dim:28\n‘)
        f.write(‘input_dim:28\n‘)
        f.write(str(create_deploy()))
if __name__ == ‘__main__‘:
    write_deploy()

运行该文件后，会在mnist目录下，生成一个deploy.prototxt文件。

这个文件不推荐用代码来生成，反而麻烦。大家熟悉以后可以将test.prototxt复制一份，修改相应的地方就可以了，更加方便。

 5：

Convert train_val.prototxt to deploy.prototxt

1 Reply

1. Remove input datalayer and insert a description of input data dimension
2. Remove “loss” and “accuracy” layer and insert “prob” layer at the end

Here is a google groups link.

If you have preprocessing layers, things get a bit more tricky.

For example, in train_val.prototxt, which includes the “data” layer, I insert a layer to calculate the mean over the channels of input data,

layer { name: “mean” type: “Convolution” bottom: “data” top: “data” param { lr_mult: 0 decay_mult: 0 }

…}

between “data” layer and “conv1” layer (with bottom:”data” / top:”conv1″).

In deploy.prototxt, the “mean” layer has to be retained, yet its container needs to be changed! i.e.

layer { name: “mean” type: “Convolution” bottom: “data” top: “mean“ param { lr_mult: 0 decay_mult: 0 }

…}

and the “conv1” layer needs to be changed accordingly, ( bottom:”mean”/ top:”conv1″ ).

It’s fine to use train_val.prototxt with “mean” layer using “data” container in the training phase, and use deploy.prototxt with “mean” layer using “mean” container in the testing phase in python. The learned caffemodel can be loaded correctly.