From 20e0d8a1c1c276b1166d84627ea8d750810cdd08 Mon Sep 17 00:00:00 2001 From: Jessica Ro Date: Tue, 14 Feb 2017 10:34:56 -0800 Subject: [PATCH] didn't push yesterday --- logistic-regression/demo.ipynb | 189 +++++++++++++++++++++++++++++---- 1 file changed, 167 insertions(+), 22 deletions(-) diff --git a/logistic-regression/demo.ipynb b/logistic-regression/demo.ipynb index 57d46b9..98ad48a 100644 --- a/logistic-regression/demo.ipynb +++ b/logistic-regression/demo.ipynb @@ -20,7 +20,7 @@ }, { "cell_type": "code", - "execution_count": 64, + "execution_count": 10, "metadata": { "collapsed": false }, @@ -68,22 +68,55 @@ }, { "cell_type": "code", - "execution_count": 65, + "execution_count": 30, "metadata": { "collapsed": false }, - "outputs": [], - "source": [ - "# What is the **default rate** in the dataset (# of defaults / total)\n" + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "0.0333\n" + ] + }, + { + "data": { + "text/plain": [ + "3.3300000000000005" + ] + }, + "execution_count": 30, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "# What is the **default rate** in the dataset (# of defaults / total)\n", + "print (len(df[df.default==1])/len(df.default))\n", + "\n", + "rate = df[df.default ==1].count()[0]/df.shape[0] * 100 #3% default rate\n", + "rate" ] }, { "cell_type": "code", - "execution_count": 66, + "execution_count": 34, "metadata": { "collapsed": false }, - "outputs": [], + "outputs": [ + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeoAAAHtCAYAAADWaD2CAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAAPYQAAD2EBqD+naQAAIABJREFUeJzs3XmYXMV97//36WW6Z7SgkTRIIGE2AxJYILQAQguLF8B5\nro2xU0kc/66XGHPtGMdJCL5JHJzEuV7IgrfLz7tjXyc3LifGO8iAAe0gIYlVIPZFQmIkjTSSZu3u\nun/U6dFo1N0z0+rpOd39eT0Pj6arTreqGJjPfM+pUydwziEiIiLRFBvvAYiIiEhxCmoREZEIU1CL\niIhEmIJaREQkwhTUIiIiEaagFhERiTAFtYiISIQpqEVERCJMQS0iIhJhifEeQJ4x5kzgfwNLgb3A\n16y1/xT2nQZ8C1gCvAj8qbX27kHvfQtwG3AGsB643lr7wqD+TwI3AZOAHwMft9b2jP2sREREjk8k\nKmpjTAD8CtgNzAf+B/BpY8zvh4f8DNgJLAR+CNxhjJkdvvcU4A7gO8AiYA/w00Gf/W7gFuB64Erg\nEuDWsZ+ViIjI8YtEUAMzgC3Ax6y1z1lr7wLuBZYZY64ATgdusNY+ba39Ar5q/lD43uuBjdbaL1lr\ntwEfBE4zxqwI+z8B3GatvdNa+zBwA/BHxph09aYnIiJSnkic+rbW7gL+IP/aGLMUWA58DF8Bbx5y\nqnoN/jQ4wMXAqkGf1W2M2QwsMcasARYDnxn03g1AE3AB8GDlZyMiIlI5UamoBxhjXsQH73rgJ8BJ\n+NPeg+0GZodfl+qfAqQH91trs/hr4LMRERGJuMgFNXAd8N/w16pvA1qA3iHH9AKp8OtS/S2DXhd7\nv4iISGRFLqittZuttb8G/gx/PblQqKaArvDrnhL9PYNeF3v/cFqABRwJfRERkaqJxDVqY8yJwBJr\n7c8GNT+Jv5b8GjB3yFtmhu0AO8LXQ/u34E9x94Svt4d/VxyYNuj9w5kDPDzCY0VERACCSn1QJIIa\nv6r7J8aY2dbafIAuAl7HLxz7C2NMylqbP4W9DFgdfr0hfA2AMaYFuBC4xVrrjDEbw/78grNLgT7g\nkdEMsLOzk97eoWfQ60MikaC1tZWOjg4ymcx4D2dMaI71od7nWO/zg8aaY8U+r2KfdHw2ApuA7xpj\n/gwf3LcC/4AP2FeAfzXGfBZ4B34l9wfC934XuMkYczPwS/wK7+ettflgvh34ujHmCfyistuBb452\nw5NsNkt/f3/5M6wBmUxGc6wDmmPtq/f5QWPMsVIicY3aWpsD3gkcBtYB3wS+ZK39Wtj3Dvzp603A\ne4FrrbWvhu99Cb8A7UPAQ/iV3tcO+uwfAZ8HvgGsxK8m/1R1ZiYiInJ8AufceI8h6hYAD3d0dNDd\n3T3eYxkTyWSStrY22tvb6/Y3XM2xPtT7HOt9ftBQc6zYNepIVNQiIiJSmIJaREQkwhTUIiIiEaag\nFhERiTAFtYiISIQpqEVERCJMQS0iIhJhCmoREZEIU1CLiIhEmIJaREQkwhTUIiIiEaagFhERiTAF\ntYiISIQpqEVERCJMQS0iIhJhCmoREZEIU1CLiIhEmIJaREQkwhTUIiIiEaagFhERiTAFtYiISIQp\nqEVERCJMQS0iIhJhCmoREZEIU1CLiIhEmIJaREQkwhTUIiIiEaagFhERiTAFtYiISIQpqEVERCJM\nQS0iIhJhCmoREZEIU1CLiIhEWGK8ByAiIo3B9feRfeQhOg92kJ3Uijt3PkGyabyHFXkKahERGXOu\nvw/3/a+S2b2TrkSCTCYDG1fD+29UWA9Dp75FRGTsPbYJt2vHUU1u1w54bNM4Dah2KKhFRGTMDQ3p\ngfbdO6s8ktqjoBYRkTEXzJxVuH3GyVUeSe1RUIuIyNibt+iYsA5mzoJ5i8ZpQLVDi8lERGTMBckm\neP+NxJ/cSsvB/XRNmkJWq75HREEtIiJVESSbiC9ayuS2Nnrb28n194/3kGqCTn2LiIhEmCpqERGp\nCm14Uh4FtYiIjDlteFI+nfoWEZGxpw1PyqagFhGRMacNT8qnoBYRkTGnDU/Kp6AWEZGxpw1PyhaZ\nxWTGmJOBrwBXAF2ABf7SWttnjPkycCPggCD880Zr7e3he98C3AacAawHrrfWvjDosz8J3ARMAn4M\nfNxa21OtuYmINDpteFK+yAQ18F/AXmApMA34HpABPgXMDf/8/qDjOwGMMacAdwB/A6wEPgP8FLgg\n7H83cAvwh8Dr4WfcCnxirCckIiJHaMOT8kQiqI0x5wAXATOstXvCtluAf+RIUN9qrX29wNs/DGy0\n1n4pfN8HgV3GmBXW2lX4QL7NWntn2H8D8BtjzM2qqkVEJOqico16F3B1PqRDAXCCMWYSMAvYXuS9\nlwCr8i+std3AZmCJMSYGLAZWDzp+A9BEWHGLiEh1uP4+spvW0mm/R3bTWlx/33gPqSZEoqK21h4A\n7s6/NsYEwMeBe/DVtAM+bYy5Bn96/F+stT8IDz8JGLq+fzcwG5gCpAf3W2uzxpi9Yf+DYzIhERE5\niuvvI/e9L+NefJZDuQy5WAJ32huJffBPdJ16GFGpqIf6R2A+8GlgDpADngSuAb4NfNMY887w2Bag\nd8j7e4FU2EeJfhERqQK3ZQM8uQV3YB+5QwdxB/b511s2jPfQIi8SFfVgxpgv4q8rG2vtk8CTxpif\nW2v3h4c8bow5G/go8DOgh2NDNwV0hH0U6e8azbji8TjJZHI0b6kZiUTiqD/rkeZYH+p9jvU8v77H\nHsYNXTzW30/w+MMkl755fAY1Rir9/YvUfw3GmK8CNwB/aK39ab59UEjnbcPfxgWwA5g5pH8msAV/\nmrwnfL09/Dvi+FXlr41mbJMnTx7N4TWptbV1vIcw5jTH+lDvc6zH+e1Np+gOgoHXQfh1OpVmWlvb\neA2rJkQmqI0xnwE+AvyetfaOQe1/B1xqrX3roMMvBJ4Kv94ALBt0fEvYf4u11hljNob9+QVnlwJ9\nwCOjGV9nZye9vUPPoNeHRCJBa2srHR0dfqP8OqQ51od6n2M9zy8zZz7uoTXQ13NkR4ymNH1zLqC9\nvX28h1dR+e9jxT6vYp90HIwxc/HXoz8HrDPGzBjU/Qvgfxpj/gx/f/RVwPuAy8P+7wI3GWNuBn6J\nv4/6+fDWLIDbga8bY57ALyq7HfjmaG/Nymaz9Nf5PX+ZTEZzrAOaY+2rx/nl5p4P6Wbo62UgqdPN\nZOeef+wpcTlKVBaTvQM/lk/jw3Qn/tT0TmvtJuA9wH8HHsOvBv8Da+1DANbal4DrgA8BD+FXel+b\n/2Br7Y+AzwPfwG+Ish5/b7aIiFRJ8NSjMHESQet04lNaCVqn+9dPPTreQ4u8wDk33mOIugXAwx0d\nHXR3d4/3WMZEMpmkra2N9vb2uvstPk9zrA/1Psd6nl/u1z/GbV5PEAQkwudRO+cIFl5K7Jr3jPfw\nKir8PgbDHzkyUamoRUSkjunpWeVTUIuIyNjT07PKpqAWEZExFySbcL/7QYLW6eS6DhG0TvevtSvZ\nsBTUIiIy5nJdh+DzN5N7YjO5ve3kntjsX3cdGu+hRZ6CWkRExt6vfwwHDxzddvCAb5eSInEftYyf\nXNch+u/6Ca/vfpXMjNnkrr6OWMvE8R6WiNQZ98oLgIOcwzkHQQCxAPfqi+M9tMhTUDewXNch3N9/\nkuzBA+SCAPf4FnhoFblbvqSwFpHKOvkN8NSjMPiW4FwAJ50yfmOqETr13ch+/WM4uB9yOVwmA7mc\nf61TUSJSaaecAcGQyAlivl1KUkXdwNzLz0Mmc8xvuO7lF8ZvUCJSl4L9e3Enz4YD+wky/bhEEk6Y\nQnBg33gPLfIU1I0slTo6pMG/Tul2CRGprGDmLFwsQTC17eidybThybB06ruRzTodYvGj22JxmK1T\nUSJSYfMWwYkn4Q4fIrtvD+7wITjxJG14MgKqqBtY7A2nkytwKip2ymnjPTQRqUe5HPT24DJ9kGjy\nr2VYqqgb2bxFBCef6k9FnfwGgqltBCefqt9wRaTi3JYNsG0rHOrE9XTDoU7YttW3S0mqqBtYkGyC\n999I/MmttBzcT9ekKWTPna8t/USk4tzmddB1+OjGrsO4zevhohXjM6gaoaBucEGyifiipUxua6O3\nvZ1cnT1aT0Qi4tWXCrfveLGqw6hFCuoG5/r7yD7yEJ0HO8hOasWpohaRsdDfV7i9r0i7DFBQNzDX\n34f7/lfJ7N5JV3i7BBtXw/tvVFiLSGW1ToN97YXbpSQtJmtkj23C7dpxVJPbtQMe2zROAxKRurXi\n6tG1ywAFdQMbGtID7bt3VnkkIlL3zj4XYkMiJxbz7VKSgrqBBTNnFW7XTkEiUmk/+nbhnRB/9O3x\nGU8NUVA3snmLjgnrYOYs3UctIpW34+XCQb3jlfEZTw3RYrIGpvuoRaRqWiYUaW+p7jhqkIK6wek+\nahGpitYT4aXnjm2femL1x1JjdOpbRETGXrJIXZhQvTgcBbWIiIy9iZNH1y4DFNQiIjL2OvePrl0G\n6JxDg9MWoiJSFfv3FW4/UKRdBiioG5i2EBWRqhl6a1Zerki7DNCp70amLURFpFqy2dG1ywAFdQPT\nFqIiUjXdhwu39xRplwEK6gamLURFpGqK7dGgvRuGpaBuZNpCVESqJREv3B4v0i4DtJisgWkLURGp\nmt6e0bXLAAV1g9MWoiJSFb29o2uXAQrqBqf7qEWkKoY+i3q4dhmgoG5guo9aRKom2VR45XcyVf2x\n1BgFdSN7bBNux0uwt53+TB8kmiDTT/DYJlhw6XiPTkTqSabIZbVMX3XHUYN0zqGB5V58Fna85H/L\n7e/3f+54idxLz4730ESk3ujUd9n0b6iRbdt67LZ+zsGTW8dnPCJSv1xudO0yQEHdyA52Fm4/dLC6\n4xCR+tfVNbp2GaCgbmTFTjkFQXXHISL1r9hDOYq1ywAFdSOb3FqkfWp1xyEi9a/Y7/8qDIaloG5k\nqSK3RaR0a5aIVFjzhCLtLdUdRw1SUDeyva8XaW+v7jhEpP61FAnqYu0yQEHdyLq7C7f3aHGHiFRY\nb5GfN8XaZYCCupGddmbh9lOLtIuIlKvYXSbF2mWAgrqR/cFHjl35HYv5dhERiQQFdQMLnt8OqeYj\nqy6DAFLNvl1EpKKKLvuu6ihqUWT2+jbGnAx8BbgC6AIs8JfW2j5jzGnAt4AlwIvAn1pr7x703rcA\ntwFnAOuB6621Lwzq/yRwEzAJ+DHwcWttwz8E1W19EHq6j9zH6Bz0dPv2i1aM7+BEpL4kEoX3+05E\nJoYiK0oV9X8BaWAp8PvAfwM+G/b9DNgJLAR+CNxhjJkNYIw5BbgD+A6wCNgD/DT/ocaYdwO3ANcD\nVwKXALeO/XRqwJ72Y7fvczmt+haRystmRtcuAyIR1MaYc4CLgA9Ya5+y1q7Fh+t7jTFXAKcDN1hr\nn7bWfgFfNX8ofPv1wEZr7ZestduADwKnGWPyJeEngNustXdaax8GbgD+yBiTrt4MI+rwgcLth7S4\nQ0QqTDuTlS0SQQ3sAq621u4Z0n4CvgLePORU9Rr8aXCAi4FV+Q5rbTewGVhijIkBi4HVg967AWgC\nLqjoDGrR4SJ7eh9WUIuIREUkgtpae2DINecA+DhwL3AS/rT3YLuB2eHXpfqn4E+nD/Rba7PA3kHv\nb1yZIqecMtnqjkNEGoAWk5UrEkFdwD8CFwJ/DbQAvUP6e4H8/pel+lsGvS72/saVSBZp1+IOEamw\nYnt6a6/vYUXuJ7Ix5ov468rGWvukMaYHGPqUiBR+ZThAD8eGbgroCPso0j+q7bfi8TjJZJFgq1HZ\nlomFT3+3TKy7uSbCXz4SdfxLiOZY++p5ftkS16jr9edNxT6vop92nIwxX8Uv9vpDa21+5fYO4Nwh\nh84EXhvUP7NA/xb8Ke6e8PX28O+IA9MGvX9EJk+ePJrDa8KOZJJCj2yPJZO0tbVVfTzV0Npa5Ilh\ndURzrH31OL9XEvHCl9sS8br9eVMpkQlqY8xngI8Av2etvWNQ1wbgU8aYlLU2fwp7GUcWiG0IX+c/\npwV/2vwWa60zxmwM+/MLzi4F+oBHRjO+zs5OenuHnkGvbYVCOt/e3l5ft2glEglaW1vp6OggU+za\nfI3THGtfXc8vV+QnTi5Xtz9vKvZ5Ffuk42CMmQt8GvgcsM4YM2NQ9wPAK8C/GmM+C7wDv5L7A2H/\nd4GbjDE3A78EPgM8b63NB/PtwNeNMU/gF5XdDnxztBueZLNZ+vsL3Kxfy0qs7ai7uYYymUzdzi1P\nc6x9dTm/EkFdd3MF9j73TOu0M8/qqMRnRWUx2TvwY/k0Pkx34k9N77TW5oBr8aevNwHvBa611r4K\nYK19CbgOf1/1Q/iV3tfmP9ha+yPg88A3gJX4e7A/VZVZRV3HvtG1i4hISa6vF3dgH9mH11Xs6UaR\nqKittV8Evlii/zn81qLF+lcCc0r034p2IztWb5GTCsXaRUTkGC6XJejtwXUdhv5+XDIBPd0Vu3YR\niaCWcZIrcr90sXYRERng+vsIerqhuwsXntp3zpHb8VJFT1crqBuZtvQTkWpJt0BPgbti0y3HtkWY\ny+UI+npx3Yeht5f8T0vXdQi2PojbuIbMnl00/d6HK/Z3KqhFRGTspZsLB3Vzc/XHUgaXzRB0d/nq\nOevPOjrn4MVncJvWwBNbxuwBIwpqEREZe537C7cfKNIeAc65sHrugt4eH8yAO3wItm7wAb1n95iP\nQ0EtIiJjr8TtWVHjslmC3vDac3jr2ED1vHE1PLm1ePUcixHMOb+i41FQi4jI2IvFCi9UjUXlLuFw\ncVh3F/R0H1kcdvggbAmr572vF3/zlKkEC5fBgiUkp02HLQ9WbFwK6kbWlIa+ArdiNelR3SJSYRG9\ny8QvDgtvrerrwxFWzy88jdu4BrZthWyRMcZicM48gsXL4cy5BGP0S4eCupH1FdkStVi7iEidcJkM\nQU8XdB/GZcPq+VBneO157TDV8zSCRUthwaUEk04Y87EqqBtasduwdHuWiNQfvzisJ1wc1otzzp/i\nzlfPTz1Sunqecz7BomXDVs9BMkVwymndlRq3glpEROrawOKwri5cJlwcdqgTNq/HPbwW9pV4KEjr\ntIFrzyWr53iCoLkZl24haG6mbc65T1dq/ApqERGpgoDCZ+uKPR3o+Lm+Pug5DN3dR6rn55/2C8O2\nbS2+4jxfPS9eDmfMKV49BwGkmwnSaVxTGoJgTGajoBYRkbFXpZx2uZzfd7vbLw4DcAcPHFm53bGn\n+Jtbp/tT2wuWEEycXPy4ZJKgZQKuKU0QjwNj+euGglpERKphjJfEuEy/XxzW5ffd9tXzU0euPZeq\nnufOJ1i8DE4/p3j1HI8RpFtw6WaCZBMwtuE8mIJaRETGXjxWeKFWWJGWwzlH0Jvfd7vH31p18EB4\n7XkNdOwt/uapbb56vvCS4tVzEEBTiqC5BdeU8puZlD3a8imoRURk7KWaoetQgfbR79vgshn/1Kqu\nLlw246vn557CbVoNTz1avHqOx331vGhp6eo5kfDhnG4miPuYHI+AHhjOOP7dIiLSKLoOj659iEL7\nbrvO/UdWbu8fpnpevAzml6ieYzFIp/0vFE2pMVsYVg4FtYiIVEF5F6mH7rvtcjl49km/MOzpx4av\nnhcvg9POLl49NzWFp7aPLAyLGgW1iIhEztB9t331vC6snvcVf+O0E49ce54wqfAx8ZgP51T1F4aV\nQ0EtIiKRMHTf7Vy+et64BraXqp4TcO58f9/zaWcRBAViNwgglfIrt8dxYVg5FNQiIjJ+4olj9t12\nnfvh4bB6PlCiep4+w1fP8y8hmDCx8DGJZLgwLB2JhWHlUFCLiEj1pdIEzRP807P2vk4um4Vnnjhy\n7dkVuXYdT8B5F/qALlY9R3hhWDkU1CIiMvZSKejvh5YJBE0pXE8P7kCHv7/63l/gNq+DAx3F3z99\nhj+1Pf9igpYi1XNTE0HzBFwqRRCL5sKwciioRURkzAUnTIPw+rM7dPBIRyaHu+9Xhd+USMB5C3z1\nfOobC1fP+YVh6RaCRNL/XWMw/vGkoBYRkTHhclm/c1jXIdy+dgifXDWstplHrj23TDi2f2Bh2ARc\nU1NNLQwrh4JaREQqymX6/a1V3V3k+vth++PDh3QiAect9Pc9v+HMwtVzHSwMK4eCWkREjtvQfbdz\n+/fiHl4HD6+FgwdKvjd4++/CBRcXrp7zC8PSLQRNKX/8WEwgwhTUIiJSNpfJ+J3DurrI9fXC9sf9\nyu1nnii+cnuIYMmVxzbW6cKwciioRURk1FxfL4Q7h+X27fH3PG9eN2z1XFI8TtDcXNcLw8qhoBYR\nkRHxi8PClds9PfD0Y/6JVc9uK149J5PwpkWwZX3h/qYUpJsJ0s24VLrm73keCwpqEREp6ajFYXvb\n/bOeN68vXT3PmOVXbl9wEUFzC7nBQR2LHbmfuq+XYMpUQNVzMQpqERE5hl8c1oPrPozr6sI99ai/\n9vzcMNXzvEU+oGeffuzK7XQzQboFXA7X3XX0/dRSlIJaREQGuGyGoKcbug6T27Mbtym89nyos/ib\nZszyt1VdcDFBuvnoviCAphTB9Bm4wwdxB/YdHfThtWgpTkEtItLgnHMQLg5zXYdx27b66vnZbcXf\nlGyC8xcRLFoOs049tnpOJI48SjKRIHdCK+zZfeznnHpmZSdThxTUIiINymWz/taq7i7c7p2+et6y\nvnT1PHO2r57Pv6hw9Zxu9v8MfRhGsWdIl3q2tAAKahGRhuP6+gh6/DVi9+SW8NrzU8Xf0JQ6cu25\nUPXc1OSr56Y0QbzIPc/72kfXLgMU1CIiDcDlcgT5h2K89iq5h9f6lduHSyzoOukU/8SqeYuOrZ7j\nMYJ0Cy7dTJBsAoZZtV1sAdoIN0VpZApqEZE65jIZgp7DuIOduMe3+Puen3+6+Bvy1fPi5QSzTj26\nL78wrLkF15Qa3cMwghi4XOF2KUlBLSJSZwbvu+12vkxu0xp/7fnwoeJvOvkUvzDs/MUEqfTRffmF\nYenmMXgYhu6eHo6CWkSkTvh9t3twB/fjHt3krz2/sL34G5pSPpgXLStcPRdbGFbW4Iqd+i5QZctR\nFNQiIjUuv++2e/VFchtXw5YN0FWiep51ql8YNm/RsdVzMknQMqH0wrByJBKFH3WZUAwNR/+GRERq\nkN93u5dcZwc8snH46jmVPlI9n/yGo/tisSOntkeyMKwc8SJBHdeGJ8NRUIuI1JD8vtvu5efJPbQK\ntm6ArsPF31Csej6ehWHlmDgJersLtE8cy7+1LiioRUQiLr/vdq5zP2zZ4BeHvfhM8Tek0v5hGAuX\nEZx8ytF9iWT4KMmxWBhWQlipH6MpVY2/vaYpqEVEIsplM756fuk5Xz1v2QDdJarn2af5ldvzFhIM\nDsBYDNJpSLf4wByPR0nmsoXbs5nqjqMGKahFRCIkl8vherrJ7dsDm9ePvHpetJzgpNlD+lJ+U5JU\niiBWwYVh5Sh2er6rq7rjqEEKahGRCHDZLK63h96Xn6V/5c9wm9eXrp5POZ1g4bJjq+d4PDy13UIQ\nPpkqEncqF9s//FCJZ1oLoKAWERlXrq8Xd3A/bFpH36bV7Hnx2eIHp9Iw/2K/cnvmoOo5CCCVJkg3\n41Lp8Tm1LWNGQS0iUmUDt1a98DRuw/2w9UHoLnEK+JTT/bXnNy0kaBq0KCuZDE9tpwkSVVwYJlUV\nuaA2xqSATcAfW2tXhW1fBm4EHP6/QwfcaK29Pex/C3AbcAawHrjeWvvCoM/8JHATMAn4MfBxa21P\n1SYlIgK4/j7oPIDbtJrcQ6vh5eeKH5xu9tXzwmUEM2cdaY/FfOWcbh445a1wrm+RCuowpP8vcO6Q\nrrnAp4DvD2rrDN9zCnAH8DfASuAzwE+BC8L+dwO3AH8IvB5+xq3AJ8ZqHiIief6pVWH1vO63vnru\nKXA/cSh55hxyC5aQmzP/6Oo5Nfie57jCuYFEJqiNMXOBfy/SPRe41Vr7eoG+DwMbrbVfCj/ng8Au\nY8yKsCL/BHCbtfbOsP8G4DfGmJtVVYvIWHGZDHR24DauJvfgKnjl+eIHp5vhwktIXHIZ0+bOY9++\nfbj+TLgwLNwxLEoLw6SqIhPUwGXAvcCngYGLNcaYScAsoNjeeJcAq/IvrLXdxpjNwBJjzBpgMb7K\nztsANOEr7gcrOQERaWzOOYK+HnLPj6x65tQz/bXn8y4kSDYRSyb89b10M8HEJK6pjhaGxeKF76Ue\n79vGakBkgtpa+/X818aYwV1z8dekP22MuQbYC/yLtfYHYf9JwM4hH7cbmA1MAdKD+621WWPM3rBf\nQS0ix81Xz/txD60i99AD8MoLxQ9uboH5l/iV2yeedKQ9mSSYNJn4iScT7N8P/f31EdB5EybBwf2F\n26WkyAR1CXOAHPAk8BXgcuCbxpgD1tqfAS1A75D39AKpsI8S/SIiZXN9vbjnnsKtvcdXz70lrqad\n+kaCxcvh3AsJkuGDKAYWhrUQNDURJJPEk3X6kIrDB0fXLgMiH9TW2h8YY35urc3/Kva4MeZs4KPA\nz4Aejg3dFNAR9lGkf1Tb4cTjcZJ19j9QkQ39AOpuronw1pVEHT9ST3OsDpfL4vZ3kN1wP279b3Gl\nqueWCcQWLCF+0YqjqucgmfKVdTp91I5hUZjfWMkW20I0l63bnzcV+7yKftoYGRTSeduAK8KvdwAz\nh/TPBLbgT5P3hK+3Axhj4sA04LXRjGHy5MmjG3QNeKVEX1tbW9XGUU2tra3jPYQxpzlWnnOObG8P\nfdsepevun9Pz4CpcqZXbZ51Ly4q3kV5wMUGyCQcQxAiaJxBrmUCsqYkgKH5iux6/h43486ZSIh/U\nxpi/Ay611r51UPOFwFPh1xuAZYOObwn7b7HWOmPMxrA/v+DsUqAPeGQ04+js7KS3d+gZ9PrV3t4+\n3kOoqEQiQWtrKx0dHWQy9fkQAM2x8lwuh9u/j9z6+8iu/y3u1ReLH9wyMayelxOceBLdQPfBwwTp\nrF/VnUoRZLLQWWQrTRrje1hIvf68qdjnVeyTxs4vgP9pjPkz/P3RVwHvw1+rBvgucJMx5mbgl/gV\n3s/nN0sbMhO6AAAgAElEQVQBbge+box5Ar+o7Hbgm6O9NSubzdLfX+Ch53WqXueayWTqdm55muPx\nc5kMPPsEudV3wyMPlb72fNpZ/trz3PmQTIaXlIJjdgxjFMFbl9/DIADnCrbX3VwrLKpBPfDdtNZu\nMsa8B/hs+M+LwB9Yax8K+18yxlwHfBm/scla4NpB7/+RMeZU4Bv427L+E795iojIAOccHNyPW38f\nbv19sOOl4ge3TIQLw5Xb02f4Nu0YVlo8AZkCgRyPagxFx3H9GzLGtALLgZPxATgN2G6tLfBr08hZ\na+NDXv8CX1kXO34lfnV4sf5b8buRiYgcxWWz8MyT5Fav9Cu3+0pc4jr9bIJFy+Dc+QMbkGjHsBGK\nxYECQa37qIdVdlAbY/4a+CugGV8BPwT8AzDdGPO2AgvAREQiI9e5H9bfh1t3L+x8ufiBLRNhwRKC\nhUuPVM/xxKBHSephGCMytQ12FVhSNlULyYZTVlAbYz4O/B3wOXylm9845KvAD/GnqG+sxABFRCrF\n5bLwzDZyq+4avno+4xy/a9jc8331nH+UZHNzfe0YVi1nzS0c1GfNrf5Yaky5FfWNwOettbeEtzsB\nYK29M6y0/xIFtYhERO7gAVj32+Gr5wmTjlTP0070bcmkP7WdaiaI+x93CugytO8eXbsMKDeoTwUe\nKNL3FDCjzM8VEamIXC5H8MyTI6uez5zjrz3PucCfyo7FjoRz+AQrhfNxeunZwu0vF2mXAeUG9SvA\nEuCeAn2LKH1vu4jImBmontfeg3utxI+ifPW8aBlB/jrp4FPbsZjCuZJcrnB7rki7DCg3qL8D/K0x\npht/7zLAxPDZz38F/HMlBiciMhLOOdz2x3GrVsKWDdDfV/zgN8711fM55/vqOZ4gaAmrZy0MGzsT\nJhV+ktiE+tv1sdLKDeovAqeHf34xbLsP/9/3D4HPH//QRERKyx3uhLX34lbfDbteLX7gxMlw4RKC\nRUt99RwEkG4mSKe1MKxaiv3y1N84Oz6Wq6ygDu+TvsEY80/Alfj7p/cDq6y1j1dwfCIiR8nlcmSf\nepTcvb/Ebd0AxXa1CoLw2vNymHO+XwiWTBK0TMA1pbUwrNr6igW1diUbzvHcR30msNxa+43w9Rzg\nQ8aYr1lrSyyrFBEZvdyhTjLr72P3unvJlNpze+JkWHipX7ndOl0Lw6KiqQl6Cjy0sM6enDUWyr2P\n+hLgN/gnV303bG4F/j98WF+uylpEjpdzDvfMk/DAr3FbNpApVT2/ca6vns+Z5681Nw3eMUwLw8Zd\nqgV/4nWIdEvVh1Jryq2ovwCsAa7LN1hr1xtjTgd+AvwjcM3xD09EGlHu8EFYe8/w154nnQAL8tXz\nNEgkfDinmwniWhgWKfHY6NplQLlBvQB419AnUFlre4wxXwJ+dNwjE5GGMlA9339neO25yDXNICA4\n+02wcCmc/SaCZBLSaUg1Q1NKC8Oiqtj3s9i1axlQblB34x/EUch0QDfGiciIuMOHcGvuxq35Deza\nUfzASScQW7ycaW/9HQ4ECTKxePgoyRSBHuwQfcU2nCm1EY0A5Qf1XcBnjTFbrbWP5RuNMXPxe4D/\nuhKDE5H65KvnJ3z1vGVD4ccfgr/2fNZ5/r7ns99EoqWF+MmzCLp6CAJfN6t6rhVFvlOBvoPDKTeo\nPwWsA7YYY14AXgfagDOAF4C/qMzwRKSeuMMHcWvuwa1eCbt3Fj9w8hRYuJRgwaX+2nMqRZCeQDBx\nIokpUwkye3RbT63RzmRlK/c+6l3GmHnAB4Gl+Puod+CfnvU9a+2hyg1RRGrZQPV836/9tedMpvCB\nQeCvOS9eDm88lyCVDheGpY8sDAuCgUpaakxXkVgo1i4Dyr6P2lp7GPha+I+IyFHcoU7cuntxD6yE\n10tVz60ECy/19z63TtfCsHrl3OjaZcDxbHhyNvA7wARg6Pp6Z6397PEMTERqj3MOnnmC3P2/9ntu\nj6R6Pus8gnSaoHmCFobVs2RT4YVjyabqj6XGlLvhyfuA71N8HYcDFNQiDcId7MStu8c/FOP114of\neEIrwcKl/t7nqdP9k6rSLQQJvzuVquc69oYz4NlthdulpHIr6r8B7gauB14N9/4WkQbinIPtj5O7\n/05fPWeLVM+x2NHVc3MzQdpXzzq13UA6D4yuXQaUG9SnAh+11uq50yINxh3s9M96Xj1c9TyVYFFY\nPU9rC+95TutRko2q+3Dh9p4i7TKg3KB+GnhDJQciItHlnIOnHyP3wF3DV8/nzBu47znWMsFv59mU\nAhTODS2dhoMFqudUuvpjqTHlBvVfAl81xrwIbBi6laiI1Ad38MCRldvtJarnKVMJFi6DBUsI2mYM\n7BhGLK5wFm/yNGjffWz7CdOrP5YaU25QfxmYAdwLYIwZ2u+stWWvKBeR8eNyOV89r1oJW9ZDNlv4\nwFjMP+d50XI4+zxiEyaGC8N0alsKmDSxcPvECdUdRw0qN0x/WNFRiMi4c537ffW8aiW07yp+4JRp\nfuX2oqUE02f4ldtNaS0Mk9JOnFW4fUaRdhlQ7s5kf1fpgYhI9Q1Uzw/cBVs3DF89L/bPe45NmIhr\n0sIwGYWdLxVu31GkXQYcz4YnaeB8IMWR/09j+A1Qlltr/+fxD09ExoLr7MCt/a1fuV2qem6d5k9t\nL7yUWNsMf2pbC8OkHDuL3CRUrF0GlLvhyeXAj4GpRQ45CCioRSLE5XLw1CPkHlgJWx+EXInqee58\nv3J7zjxiEyZpYZgcv2J7tGvv9mGVW1H/L2AP8BHgfUAW+B7wduCjwDUVGZ2IHDd3oOPItec9BVbd\n5k1tC689LyN24gztGCaVdcFF8NtfFm6XksoN6guAD1tr7zDGnAD8D2vtncCdxpgm4NP4fcBFZBy4\nXA62PUJu1V2w9aHi1XM8fqR6nns+sZZJ2jFMxkTwzvfitj4I+/fhd5kO/G1973zveA8t8soN6hj+\nsZYAzwDnDer7T+AHxzMoESlPdt8eMr/4D7+t597Xix84tc2H80XLiU2fefSjJKs0VmkssZaJ5D7z\nZWJ3/YTE7h1kZswid/V1xFqK3LYlA8oN6ueAecBq/C5lE4wx51hrnwaSwKQKjU9EhuFyOXhyC31r\n7mbnlhLXnvPV80UrYO75BC0TtTBMqipINhE7cw7pE2fQNakVpydnjcjx3Ef9RWNMzFr7NWPMJuBr\nxpivAH8NPFGxEYpIQW7/Xtzae/3K7b3txQ+cdqLfc/uiy4i1zdSjJGVcuP4+ct/+Z3LbHuVApg8S\nTf5yy4f/nECBXVK5Qf2PwHTgYuBrwMeAO4GfAZ3AOyoyOhE5istl4YmtftewRx+CXK7wgfEEnDuf\n4OLLCM69AJonaGGYjKvcg/f7feJd+LDF/n7YsoHcg/cTX/a2cR1b1JW74UkO+ItBrzcZY84A5gBP\nW2s7KzQ+EQFcx17c2rtxq++GfcWr5/iMk2HhUnKLlxGbfpKvnnX7i0TBb356JKTznPPtCuqSKrYf\nt7X2ILCxUp8n0uh89bzF7xr26CZwJarn8y4kcemVTFv2Zvb39JIJfyAqoiUyDh8aXbsMGHFQG2Ny\n+DX1I6GHcoiUye3bEz7v+TfQsaf4gdNn+C09L7mcYPoMEhMm0tQ6laC93Z9WFImSN5wBj28u3C4l\njSZM/56RB7WIjILLZeGxzeRWryxdPScScO4CgiWXE5x7IaTTWhgmteGCiwoHtTY8GdaIg9pa+7dj\nOA6RhuT2tePW3INbc3fp6rltJsFFywmWXAHTZgwsDBOpGduf8NuFDr5OHQS+/fK3j9+4asDxPJTj\nZGAZxR/K8fvHPzyR+uOyWXj8Yb9y+7GHS1fP5y0kWHolwZz5kEoRxGLVHaxIpezZXXgxWaltbQUo\n/6Ec7wH+Db+5Sf7ffDDo66eOf2gi9cXtbfcrt9fcM3z1fPFlBEveDNOmD+wYJlLTit2tsK/E/wsC\nlF9R/zWwGX//9B+Hn/NF/EM5Pgd8siKjE6lxLpuFxzb56vnxzSWq5yTMW0hs6Ztxcy4glkpXd6Ai\nY63Ys86zmeqOowaVG9TnAO+11m4xxtwH3GSt3QZsM8bMwAf53ZUapEitcXtfx62521973r+v+IEn\nnkRwyeW+ep46jUCPkpR6lW6GQwW22Eg3V38sNabcoM4B+Z8+zwJzwu1Ec/gdyj5QgbGJ1BSXzcKj\nG331/MTmY6/H5eWr5+VvxZ1zPrFwv22Ruja1rfD16KknVn8sNabcoN4GLAVW4a9Hp/CPvtwCtIav\nRRqC27M7rJ7vgQMlqucZJxNccgXBpW/2j/eLxVQ9S8MITj8L98J2yPT7X2KDABJJgtPfON5Di7xy\ng/obwNeNMROttX9tjPkt8D1jzHeAjwMPV2yEIhHkMhlfPa9eCU9sKV49J5Nw/mJiy96GmzOPmG6r\nkgbl3vpOuO/X/kV+W9sg8O1SUrl7fX/bGJMCTgubPgL8Cvgy8CLwJ5UYnEjUuPZdvnpeew8c6Ch+\n4IxZBJdeCUuuJJgylSAIVD1LQ4s99xS5GSfBgf0EmX5cIgknTCH23FOw4NLxHl6kHc99H7cD1xhj\nvoA/3f0TYIu19j8rMjKRiPDV80PkHlgJ27aWqJ6b4IKLiK24CnfWecQSuq1KJM/t2gGxBMHUNhKJ\nBJlMBuccbvdO/RI7jHLvo56Or6AXAVlgD/6xl3FjzG+A66y13WV+dgrYBPyxtXZV2HYa8C1gCb5i\n/1Nr7d2D3vMW4DbgDGA9cL219oVB/Z8EbgImAT8GPm6t7SlnfNI4XPsu3Orf4NbdW7p6njmbYMmV\nsPTNxE5oBfQwDJGhgpmzCu5BHcw4uepjqTXlbnP0T/hQfBeQstaeDKSB9wKX4O+pHrUwpP8vcO6Q\nrp8CO4GFwA+BO4wxs8P3nALcAXwH/4vDnvD4/Ge+G7gFuB64MhzfreWMT+qfy2RwD68le9st5P7q\nI7g7/7NwSCebYPFyYn/+DwR/+xVib3/PQEiLSAHzFsH0Gbh97WR2vIzb1w7TZ/h2Kancc3PvBP7c\nWvvzfEN4a9aPjDFT8Q/w+MRoPtAYMxf49wLtV+J/KbgkrIK/YIx5M/Ch8O+5Hthorf1SePwHgV3G\nmBVhRf4J4DZr7Z1h/w3Ab4wxN6uqljz3+mu4Nb/Brb0XOvcXP/CkUwiWvhmWXEls8hRA1bPISLj+\nPnj+aTh0EOcc9PbA80/j+vsIkk3jPbxIKzeoHfB6kb7tlHd71mXAvcCnga5B7RcDm4eE6hr8afB8\n/6p8h7W22xizGVhijFkDLAY+M+i9G4Am/O1kD5YxTqkTLtMPWx/09z1ve6T4gU0pmH8xscuuwZ05\nh1hcT6sSGS33i//wty/m13g4Bwf2+fbf+/D4Di7iyg3q/wPcbIy5Z3CAGmNiwI3409ejYq39+qDP\nGdx1Ev6092C7gdkj6J+CPyU/0G+tzRpj9ob9CuoG5F7fiVsVXns+eKD4gSe/gWDpW+HSK4lNnASo\nehYp21OPFn4ox9OPjc94asiIg9oY891BL5P4ivZ5Y8yvgF3AVOBt+OC8vYJjbAF6h7T1cqRqL9Xf\nMuh1sfdLA3CZfrpW303fz/6D3HDV84WXEKy4muCscwkCRbNIRTS3FG7XFqLDGk1FfSUctWjv1fDP\ntww5rh14D3DzcYxrsB78LwGDpThyeryHY0M3BXSEfRTp72IU4vE4yWR9bVZRZIt8gLqZa27Xq2Qf\nWEnf2nvYW+LaczDrVOIrriK29C0D1XOtSYS3gyXq+Lawep9jPc8vuPztZF7YfvTDOeJxEpe/nUSd\n/LzJq/T3b8SfZq09raJ/88jt4NhV4DOB1wb1zyzQvwXYiw/rmfhr5xhj4sC0Qe8fkcmTJ49q0LXg\nlRJ9bW1tVRtHpbn+PrrX3cehu+6g79FNRY8LUmnSS69kwlXXkT7vgrqpnltb63/1eb3PsR7n597+\nLvZu20rv45txfb0ETSlSb1rAtLe/i0D73ZdUC7+2bQA+ZYxJWWvzp7CXAasH9S/LH2yMaQEuBG6x\n1jpjzMawP7/g7FKgDyhx/vNYnZ2d9PYOPYNev9rbizw7NsJyu14le/9dZNfeDQcLPKUnFJxyBvHl\nbyW27C24lokcAg7tqf1n4iYSCVpbW+no6CCTqc9HB9b7HOt9fu6/30jTEw+T2r+P3ilTyZ23kD0H\niv+/Wqvy38eKfV7FPmnsPIAv/v7VGPNZ4B34ldwfCPu/C9xkjLkZ+CV+hffz+c1S8NfLv26MeQK/\nqOx24JujvTUrm83S399/vHOpGbUyV9ffj9uyHrdqZelFKak0zcveTGbJW8ideiYOf+o/WyPzHI1M\nJlMz379y1fsc63V+2c4OYqvupn/3q7gZs8mdcgbxE4Ze2ZShohrUA9fCrbU5Y8w78RuabMI/VvNa\na+2rYf9Lxpjr8PuM3wKsBa4d9P4fGWNOxT9IpAn4T+BT1ZqIjA2369Uju4YdOlj8wFPOIFj+VpLL\n3sq0N5zKnj17yNXhD0CRqMse2Ad/dQO5/j7f8Nqr8MRmsp/7hsJ6GIErtm+x5C0AHu7o6KC7u6xd\nUSMre/07ivbFv/Xzon3jxfX34TaH1fP2x4sfmGomWLwULruG2GlnAX5xXFtbG+3t7XVZqYDmWA/q\neX7Z//0P8MjGYzsuWEz8jz9d/QGNofD7WLFFL1GtqEUGuNde8fc9r/8tHB6mel7xNrjkCmK65UMk\nWnYWWb5arF0GKKglklx/H+7hdbjVK2H7E8UPTDcTLF4OK64aqJ5FJIJOPgXadxVul5IU1BIpbufL\n4bXn30LXoeIHnnomwfKr4OLLiaXT1RugiJTnfR+DJ2+A/DVq8A+3ed/Hxm9MNUJBLePO9fX66nnV\nSnj2yeIHppsJLloOl72d2BvOqN4AReS4xU+YSvZz3yD2b98g2L0DN2MWuT+8QQvJRkBBLePG7XgZ\nt3olbv19pavn084iWPY2gksuJ0hpYwSRWhU/YSrJP7mlbhfMjRUFtVSV6+3FPbzWX3t+dlvxA5tb\nCC5aAZddTewUVc8i0rgU1FIVbsdLuFUrcRvug67DxQ88/WyCFVcRLF6h6llEBAW1jCFfPa/x156f\ne6r4gc0TCC5e4e97nn1a1cYnIlILFNRSce7VF8Pq+X7oHqZ6vuwagsXLtCm/iEgRCmqpCNfbg9sU\nVs/PP138wOYJBBdfBpdfQ2zWqdUboIhIjVJQy3Fxr7zgq+cH74fuEo/4PuMcgsuuJli8nCDZVLXx\niYjUOgW1jJrr7cFtXO2r5xe2Fz+wZQLBJZf7+561+5CISFkU1DJi7uXn/X3PG+6HnhIPKHnjuX7l\n9qKlqp5FRI6TglpKcj3dR6rnF58pfmDLRF89X34NsZNUPYuIVIqCWgpyLz0XXnt+AHpLVc9z/crt\nhUsJksnqDVBEpEEoqKWg3D/8afHOCZP8yu0r3k5s5uzqDUpEpAEpqGXkzjrPX3tW9SwiUjUK6gbl\nekrcSjXYhEmw5AqCy65W9SwiMg4U1A3EOQcvPeuvPT+0qvTBZ59HsOJqggWXqnoWkYpw/X1kH3mI\nzoMdZCe14s6drztDRkBB3QBcdxfuwQf8E6tefr74gUEATSlwOeJ/8fnqDVBE6p7r78N9/6tkdu+k\nK5Egk8nAxtXw/hsV1sNQUNcp5xy8+MyR6rmvt/jBiSQk4tDbC7091RukiDSOxzbhdu0gCIKBJrdr\nB8Fjm2DBpeM4sOhTUNcZ13UY99AD/r7nV14ofuDEydDfBy4HfX2Q0QPcRWTsuF07Crfv3klQsEfy\nFNR1wDkHL2z31fPG1aWr53Pm+ZXbC5aQ++i7qzdIEWlowcxZuELtM06u+lhqjYK6hrmuw/7a86q7\n4NUXix84cTLBpVf65z2feFLVxiciMmDeIoLN62H3zoGmYOYsmLdoHAdVGxTUNcY5B88/7ffc3rja\nn7YuJAjC6vlqggsvIUjoWy0i4ydINsH7byT+5FZaDu6na9IUslr1PSL66V0jXNchXz0/cBfseKn4\ngZNOOFI9t82s3gBFRIYRJJuIL1rK5LY2etvbyfVrbcxIKKgjbKB6XrUSt2kE1fNlVxPMV/UsIlJP\n9BM9gtzhQ7gN9/v7nktVz5OnECy5Ei67WtWziEidUlBHhHMOntsWVs9r/a1ThQQBzDnfP7Fq/sUE\n8Xh1ByoiUibtTFYeBfU489Xzff6+550vFz/whFZfPa+4StWziNQc7UxWPgX1OHDOwbNh9fzwMNXz\n3Av8ym1VzyJSy7QzWdkU1FXkDh/Erf8tbtVv4LVXih84uZVg6ZWw4mpi02dUb4AiImNEO5OVT0E9\nxpxz8MwTYfW8rvhWnfnq+bKrCS5Q9Swi9UU7k5VPQT1G3KFO3Prw2vOuV4sfmL/2fJmqZxGpY9qZ\nrGwK6gpyzsH2sHrevBYymcIHDq6e519MEFP1LCL1LUg24d57A7G7fkJs96vEZswmd/V1Wkg2Agrq\nCnAHO/2159Uroch1GABOmBpWz1epehaRhuL6++Dfv0Fu905yiQS5fXthXztOq76HpaAuk6+eHw+r\n53XFq+dYDOZeQGzFVbgLLiIW179yEWlAWvVdNqXGKPnq+V6/cnt3iep5ytQj9z2H1bNWNopIo9Kq\n7/IpqEco+8J2cj//D9yW9cNXz8uvwl2wmFgiWd1BiohElFZ9l09BPUKZf/0K7rmnC3dOmUaw5ApV\nzyIixWjVd9kU1OWKxeDc+b56Pn+RqmcRkRL0POryKahHq3U6wSWXw/KriLWpehYRGSk9j7o8CuoR\nCt54LrG3XYebt4hYUtWziIhUh4J6hJLv+yjZnh5VzyIiUlWx8R5ArRh875+IiEi1KKhFREQiTEEt\nIiISYQpqERGRCKuZxWTGmGuBnwAOf0eUA/7LWmuMMacB3wKWAC8Cf2qtvXvQe98C3AacAawHrrfW\nvlDVCYiIiJShlirqc4GfAzPDf04CPhz2/QzYCSwEfgjcYYyZDWCMOQW4A/gOsAjYA/y0qiMXEREp\nU81U1MBc4HFrbfvgRmPMlcDpwMXW2h7gC8aYNwMfAv4euB7YaK39Unj8B4FdxpgV1tpVVZ2BiIjI\nKNVaRb29QPvFwOYwpPPW4E+D5/sHAtla2w1sHtQvIiISWbVUUZ8DXG2M+WsgDvwYuAV/CnznkGN3\nA7PDr4frFxERiayaCGpjzBuAZqAb+F38qe6vhG0tQO+Qt/QCqfDr4fpFREQiqyaC2lr7sjFmmrV2\nf9j0qDEmjl849j2gdchbUkBX+HUPx4ZyCugYzRji8TjJOtvjO1uir97mmkgkjvqzHmmOta/e5weN\nNceKfV5FP20MDQrpvG1AGtiFX2g22EzgtfDrHeHrof1bRvP3T548eTSH14RXSvS1tbVVbRzV1No6\n9He6+qM51r56nx80xhwrpSaC2hjzNuDfgdmDFo1diL/VajVwkzEmZa3Nn+JeFrYDbAhf5z+rJXzv\nZ0Yzhs7OTnp7h55Br1/t7e3DH1RDEokEra2tdHR0kMlkxns4Y0JzrH31Pj9orDlW7PMq9kljax3+\nVPa3jTF/D5wJ3Ap8Eb+i+xXgX40xnwXeASwGPhC+97v4IL8Z+CU+oJ+z1j4wmgFks1n6G+jZqfU6\n10wmU7dzy9Mca1+9zw8aY46VUhO3Z1lrDwFXAW3ARvwuZF+31v6ztTaHD+eZwCbgvcC11tpXw/e+\nBFyHv6/6IWAK8K6qT0JERKQMtVJRY63dhg/rQn3PA1eUeO9KYM4YDU1ERGTM1ERFLSIi0qgU1CIi\nIhGmoBYREYkwBbWIiEiEKahFREQiTEEtIiISYQpqERGRCFNQi4iIRJiCWkREJMIU1CIiIhGmoBYR\nEYkwBbWIiEiEKahFREQiTEEtIiISYQpqERGRCFNQi4iIRJiCWkREJMIU1CIiIhGmoBYREYkwBbWI\niEiEKahFREQiTEEtIiISYQpqERGRCFNQi4iIRJiCWkREJMIU1CIiIhGmoBYREYkwBbWIiEiEKahF\nREQiTEEtIiISYQpqERGRCFNQi4iIRJiCWkREJMIU1CIiIhGmoBYREYkwBbWIiEiEKahFREQiTEEt\nIiISYQpqERGRCFNQi4iIRJiCWkREJMIU1CIiIhGmoBYREYkwBbWIiEiEKahFREQiTEEtIiISYQpq\nERGRCFNQi4iIRJiCWkREJMIS4z2AajDGpIDbgeuALuCfrbX/Mr6jEhERGV6jVNT/BCwALgc+BnzG\nGHPduI5IRERkBOo+qI0xLcAfAZ+w1j5irf0ZcCvw8fEdmYiIyPDqPqiBC/Cn+NcPalsDXDw+wxER\nERm5Rgjqk4A91trMoLbdQNoYM22cxiQiIjIijbCYrAXoHdKWf50a6YfE43GSyWTFBhUF2RJ99TbX\nRCJx1J/1SHOsffU+P2isOVbs8yr6adHUw7GBnH/dNdIPmTx5csUGFBWvlOhra2ur2jiqqbW1dbyH\nMOY0x9pX7/ODxphjpTRCUO8AphtjYtbaXNg2E+i21u4f6Yd0dnbS2zu0MK9f7e3t4z2EikokErS2\nttLR0UEmkxn+DTVIc6x99T4/aKw5VuzzKvZJ0bUV6AcuAdaFbcuBjaP5kGw2S39/f4WHFl31OtdM\nJlO3c8vTHGtfvc8PGmOOlVL3i8mstd3AD4CvG2MWGWOuBf4c+NL4jmz8xb/181G1i4hI9TVCRQ3w\nZ/idyX4LHAD+JryfuuHFv/VzkskkbW1ttLe36zdcEZGIaYigDqvqD4b/iIiI1Iy6P/UtIiJSyxTU\nIiIiEaagFhERiTAFtYiISIQpqEVERCJMQS0iIhJhCmoREZEIU1CLiIhEmIJaREQkwhTUIiIiEaag\nFhERiTAFtYiISIQpqEVERCJMQS0iIhJhCmoREZEIU1CLiIhEmIJaREQkwhTUIiIiEaagFhERiTAF\ntYiISIQpqEVERCJMQS0iIhJhCmoREZEIU1CLiIhEmIJaREQkwhTUIiIiEaagFhERiTAFtYiISIQp\nqMwI9nAAAA0MSURBVEVERCJMQS0iIhJhCmoREZEIU1CLiIhEmIJaREQkwhTUIiIiEaagFhERiTAF\ntYiISIQpqEVERCJMQS0iIhJhCmoREZEIU1CLiIhEmIJaREQkwhTUIiIiEaagFhERiTAFtYiISIQp\nqEVERCJMQS0iIhJhCmoREZEIU1CLiIhEWGK8BzASxpj5wGbAAUHYvMlae1HYPxX4FvBWoB24xVr7\nb4PefyHw/wPzgMeBj1prN1dvBiIiIuWplYr6XGALMHPQP1cN6v8+MAm4GPhfwLeNMYsAjDEtwK+A\nB4AFwHrgV8aY5qqNXkREpEw1UVEDc4Ft1tr2oR3GmDOA3wFOtda+AmwzxiwBPgZ8CPh9oMta+6nw\nLZ80xrwd+F3gB1UZvYiISJlqqaLeXqTvYuDlMKTz1gBLBvWvGfKetYP6RUREIquWKuqYMeZR4ATg\nTuAma+0h4CRg55DjdwOzw69Pwl+XHtp/3tgNV0REpDIiEdTGmDQwq0h3O3Am8BzwAaAV+BLwf4B3\nAS1A75D39AKp8Ovh+oeTBvh/7d170BV1HcfxN6JhI+FMMIGXFB3zrmQYkoqoWaaY1/zmZcxLmnmL\nwho0E0k0IdBynBRviY5afkc0yysOmoKKIqNoM6JmKoqMBSkQYqY+/fHdA8t5zgO7z7mw55zPa+bM\nc86e3X1+32fP8/ue3f3tfjfYYIOMszefnj17AtCrVy/WX78QH4maU4ytodVjbPX4oK1i/AowD/ig\n2vUV5a+0B/AoMaq73BFAX2CFu38CYGYnArPNbADwIZ2Tbi9W/XHW9v7aDATo3bt3xtmbV58+fdZ1\nE+pOMbaGVo+x1eODtohxDjCYuGKpKoVI1O7+GPnOl79EXKa1GbCAGAWeNgBYmDxf2/tr8xBwPPAG\nkfRFRESymFeLlRQiUa+Jme0APA3s4u5vJpN3A/4H/B34N7ClmW3q7qVz1XsDs5Lns4DRrG4v4JKM\nTVgM3N7N5ouIiFSl8Ima+EbyKnC9mf2EOEc9GbjO3ZcAS8zsIeBWMxsJDAGOBfZJlr8TuMzMfgNc\nB/yQOG/tjQ1DREQkv8JfnuXuHcChwFLgceBu4GFgVGq27yXvzwLOB0529znJ8suAQ4jE/SyRyA9y\n9xWNikFERKS7enR0VBq/JSIiIkVQ+D1qERGRdqZELSIiUmBK1CIiIgWmRC0iIlJgzXB5VsOZ2Xii\n8tZ6wI2pyluV5h0KXA7sCrwNTHL3GxvS0BzMrBdwNXAkcVe2y939ii7mbcr63TljHEFcS78NcXva\nC939L41qa3fliTG1zEDgRWCEuz9e90ZWIec23CWZdzBxCedId/9rg5rabTljPIIo3ftFotTvSHd/\nrlFtrVYS67PAWV199pq1vynJGGNV/Y32qMuY2blEaczDgKOA481sVBfz9gfuBx4BvgyMBa4ys4Ma\n09pcJhH3nt2XKAF6kZkdWT5Tk9fvzhrjrsBU4AZgEHF9/Z1Jx190mWIscw1x74BmkHUb9gGmER37\nzsRlm3ebWb/GNbXbssa4I3Abkah3BeYS/4sbNq6p3ZcksD8Q1Q+7mqeZ+5usMVbd32iPurMfAb9w\n96cAzGw0MA6o9I33cGChu1+YvH7NzPYDjiMqfBVC8s/wfeBAd58LzDWzXwNnA3eVzd6U9btzxngs\nMN3df5e8vtrMDgWM2PMspJwxlpY5HmiKG9XnjO8kYJm7n5G8Hpt8Qd4deLBBTc4tZ4zfBP7m7rcl\ny54PnEUkhULvcSZ3lMxyR8em7G8gV4xV9zfao04xs02IQ0wzUpNnErco7V9hkQeAkytM37gOzavG\nIOJL2VOpaTOJYijlmrV+d54YpwDnVZhetO1WLk+MmFlfYDzwA+Le+EWXJ77hwD3pCe6+h7sXNkkn\n8sS4GNjJzPY0sx7E6bglxKHTohsOTCf6jTV99pq1v4HsMU6hyv5Ge9Sr24So4JWub/0usRE2T56v\n5O7zgfml12b2BeIb4pi6tzSfTYBF7v5xatq7wIZm1tfdF5fN24z1uzPH6O4vpxc0s52ArxPnDYss\nz3aEOAo0xd1fMrOGNbIKeeLbGnjGzK4l7lz4OlGj/snGNbdb8sR4BxHbTOCT5DEiuXVyobn75NLz\ntXz2mrW/yRxjLfqbtkvUa6l93RvA3T9KTSvVsl5j/epkvVOJJH9dlc2sta5qckPnuKqt372u5Ilx\npeSc5lRghrv/uU5tq5XMMZrZAcCewGkNaFet5NmGvYliO1cC3yIOL04zs+3cfUFdW1mdPDH2JSr9\nnUkUJjoDmGJmu7n7orq2snGatb/plu72N+146HsPYoToKxUeQwDM7DOp+UsfmC7rV5vZRsSAiG2A\nQ9y9aOUwu6rJDZ3jqrZ+97qSJ0Zg5WDAR4ijKEfXr2k1kynG5EvjZODMsi+dRZdnG34MPOfuv3T3\nue5+HvE/fEKd21itPDFOAF5w98nJSO/TgeVUPt3WrJq1v8mtmv6m7RK1uz/m7uu5e8/yBzHCElav\nXz2A+MNWrF9tZp8jRp/uCOzn7v+oZ/u7aQHQz8zS23sAsMLd368wbzX1u9eVPDFiZpsRRV7WB/at\ncNi4iLLGOATYCphqZsvMbFky/QEzK/Lh/TzbcCGda/2+QowxKbI8MQ4mRnoDKwsUzQW2rHsrG6dZ\n+5tcqu1v2i5Rr4m7LwTeIupZlwwD5rv7u+XzJwM87gYGAvu4e02KhNfB80T97qGpacOA2RXmnUUc\nMk3bi1X1vYsqc4zJyNsHk/mHV9q2BZU1xqeBLxGXDA5KHhCjjYs2fiIt7+d0UNm07YE36tKy2skT\n4zt0vuxnO+J8fKto1v4ms1r0N213jjqDa4AJZraAGER2GTCx9GZyjmGFuy8HTiWuhfw2sDQ1Mvwj\nd3+voa1eA3dfYWa3AJPN7BRiYNy5wImw8pDMkuSQfVPW784Z4wXEHue+wHqp7bbC3Zc2vPEZ5Yxx\ntSM7yWCXd4p8bjNnfJOBs81sDHEk7ERim966ThqfUc4YrwduMrNniVHipwFbADevk8bXSCv0N2tT\n6/5Ge9SdTSRGW96V/LzZ3a9MvT+b+MeCuLNQD+Be4ttv6TG1Ya3NbhQwhzhHchVxZ5zS5S0LiWv6\nmr1+d6YYie32WWLPM73dftvQ1nZP1hjLNUs926yf0/nAgcSo6BeBEcDByVGxossaoxPXV/+cuG76\na8TptcJ+2epC+WevVfqbtC5jpAb9jepRi4iIFJj2qEVERApMiVpERKTAlKhFREQKTIlaRESkwJSo\nRURECkyJWkREpMCUqEVERApMiVpERKTAlKhFREQKTPf6FmlDZnYS8HtgYHI7zizLHAOMJ6ob3eTu\nZ9SoLcOBR4mqQo8nlYauJcp0ZmqbSCtTohZpTx3kv//3Vayq+fxOHdpTcgBwUI3XL9K0lKhFJKu+\nwDR3n1GHdffo4rlI21NRDpEWl9RNv4Aok9gPmEYUsZ8EbOXu881sZ+Kw9rBksenAue7+eurQdAeR\nRDtSy50KnA7sQIx5eRm41N3vTH73WGCMu682HsbMPgXGuvvFyfofAfYjygHexKo97Jvd/ZRa/01E\nmokGk4m0vonAhUS938OBRURSBsDMtgWeIJL4CcApwNbAE0n99TnAUCJJ35A8X2hmZxF1oe8CDgaO\nAz4EbjOzTZPV5z3Efi9wSfL8CGBczlhFWo4OfYu0MDPbGDgHmOTulyaTH04GbB2YvL4IWA583d2X\nJ8tNB14Hfubuo4FnzAzgbXefncyzFTDB3S9L/b43icS+N+A5mtoDwN0Xm9lrybTnNZhMRIlapNUN\nJf7P7y2b7qxK1PsTh7Y/NLOeybT/ADOAbwCjK63Y3X8KK78MbA9sQxy+7gB61S4EkfamRC3S2j6f\n/FxUNn1h6nlf4LvAMWXzdAD/7GrFZrY1cTh9f+C/wDxgbvK2BoSJ1IjOUYu0tkVE0uxfNr1v6vn7\nwB+BwcDuqcdXWbXXvZpkgNr9xHntwcBG7r4bMIHVk3RHav7Ssht1PxyR9qM9apHW9iSwAjgamJma\nfmjq+WPAjsBcd/+0NNHMbidGcb9QYb39gG2Bke7+XGr6wURyLu0ELE1+bg68lTwfRmfpAWefrCEe\nkbajRC3Swtx9uZmNA8aZ2QfEZVAjgENSs11MJPT7zOwa4jD26UQy/04X6/2Xmb0BnG1mC4D3iJuU\njExmKe013wdcAVxvZhOBLYAxrErgJem98PeT10eZ2f3u/nLuwEVaiA59i7Q4dx8P/JhIuvcAOwOj\nUu+/SOzlfgrcQgw06w8c5u5/Sq2q/FKrw4AFxHXPdwBDiC8A85L14e6vEpd8bUkMaDsHOJXOdzZL\nr/dR4GHgV8S13iJtTTc8ERERKTDtUYuIiBSYErWIiEiBKVGLiIgUmBK1iIhIgSlRi4iIFJgStYiI\nSIEpUYuIiBSYErWIiEiBKVGLiIgUmBK1iIhIgSlRi4iIFJgStYiISIH9H2KDH/bU4/pTAAAAAElF\nTkSuQmCC\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], "source": [ "# What is the distribution of balances (for those who default, and those who don't)?\n", "\n", @@ -129,13 +162,49 @@ }, { "cell_type": "code", - "execution_count": 68, + "execution_count": 38, "metadata": { - "collapsed": true + "collapsed": false }, - "outputs": [], - "source": [ - "# Fit a linear model of the dependent variable (default) on balance\n" + "outputs": [ + { + "ename": "AttributeError", + "evalue": "'DataFrame' object has no attribute 'sl'", + "output_type": "error", + "traceback": [ + "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", + "\u001b[0;31mAttributeError\u001b[0m Traceback (most recent call last)", + "\u001b[0;32m\u001b[0m in \u001b[0;36m\u001b[0;34m()\u001b[0m\n\u001b[1;32m 10\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 11\u001b[0m \u001b[0;31m# Draw a scatterplot with your best fit line to show how well the model fits our data\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 12\u001b[0;31m \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mscatter\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdefault\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msl\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 13\u001b[0m \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mplot\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdefault\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpredictions\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 14\u001b[0m \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mshow\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m//anaconda/lib/python3.5/site-packages/pandas/core/generic.py\u001b[0m in \u001b[0;36m__getattr__\u001b[0;34m(self, name)\u001b[0m\n\u001b[1;32m 2742\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mname\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_info_axis\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2743\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mname\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2744\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mobject\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__getattribute__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mname\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 2745\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2746\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0m__setattr__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mname\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mvalue\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;31mAttributeError\u001b[0m: 'DataFrame' object has no attribute 'sl'" + ] + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeoAAAHtCAYAAADWaD2CAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nERHJMQVqERGRHFOgFhERybF/AOKU9nxM72N0AAAAAElFTkSuQmCC\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "# Fit a linear model of the dependent variable (default) on balance\n", + "sns.lmplot(x='balance', y='default', data=df);\n", + "\n", + "\n", + "# Create a simple linear model that assesses the relationship between years in current position with salary\n", + "lm = smf.ols(formula='balance ~ default', data=df).fit()\n", + "lm.summary()\n", + "\n", + "df['predictions'] = lm.predict()\n", + "\n", + "# Draw a scatterplot with your best fit line to show how well the model fits our data\n", + "plt.scatter(df.default, df.sl)\n", + "plt.plot(df.default, df.predictions)\n", + "plt.show()\n" ] }, { @@ -146,7 +215,8 @@ }, "outputs": [], "source": [ - "# Generate predictions using your model (this can be interpreted as probability of default)\n" + "# Generate predictions using your model (this can be interpreted as probability of default)\n", + "\n" ] }, { @@ -269,13 +339,87 @@ }, { "cell_type": "code", - "execution_count": 77, + "execution_count": 42, "metadata": { "collapsed": false }, - "outputs": [], - "source": [ - "# Use the `smf.glm` function to fit `default` to `balance` (hint: use the `binomial` family)\n" + "outputs": [ + { + "data": { + "text/html": [ + "\n", + "\n", + "\n", + " \n", + "\n", + "\n", + " \n", + "\n", + "\n", + " \n", + "\n", + "\n", + " \n", + "\n", + "\n", + " \n", + "\n", + "\n", + " \n", + "\n", + "\n", + " \n", + "\n", + "\n", + " \n", + "\n", + "
Generalized Linear Model Regression Results
Dep. Variable: default No. Observations: 10000
Model: GLM Df Residuals: 9998
Model Family: Binomial Df Model: 1
Link Function: logit Scale: 1.0
Method: IRLS Log-Likelihood: -798.23
Date: Mon, 13 Feb 2017 Deviance: 1596.5
Time: 14:28:06 Pearson chi2: 7.15e+03
No. Iterations: 11
\n", + "\n", + "\n", + " \n", + "\n", + "\n", + " \n", + "\n", + "\n", + " \n", + "\n", + "
coef std err z P>|z| [95.0% Conf. Int.]
Intercept -10.6513 0.361 -29.491 0.000 -11.359 -9.943
balance 0.0055 0.000 24.952 0.000 0.005 0.006
" + ], + "text/plain": [ + "\n", + "\"\"\"\n", + " Generalized Linear Model Regression Results \n", + "==============================================================================\n", + "Dep. Variable: default No. Observations: 10000\n", + "Model: GLM Df Residuals: 9998\n", + "Model Family: Binomial Df Model: 1\n", + "Link Function: logit Scale: 1.0\n", + "Method: IRLS Log-Likelihood: -798.23\n", + "Date: Mon, 13 Feb 2017 Deviance: 1596.5\n", + "Time: 14:28:06 Pearson chi2: 7.15e+03\n", + "No. Iterations: 11 \n", + "==============================================================================\n", + " coef std err z P>|z| [95.0% Conf. Int.]\n", + "------------------------------------------------------------------------------\n", + "Intercept -10.6513 0.361 -29.491 0.000 -11.359 -9.943\n", + "balance 0.0055 0.000 24.952 0.000 0.005 0.006\n", + "==============================================================================\n", + "\"\"\"" + ] + }, + "execution_count": 42, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "# Use the `smf.glm` function to fit `default` to `balance`\n", + "#(hint: use the `binomial` family)\n", + "\n", + "\n", + "mod1 = smf.glm(formula='default ~ balance', data=df, family=sm.families.Binomial()).fit()\n", + "mod1.summary()\n" ] }, { @@ -624,22 +768,23 @@ } ], "metadata": { + "anaconda-cloud": {}, "kernelspec": { - "display_name": "Python 2", + "display_name": "Python [conda root]", "language": "python", - "name": "python2" + "name": "conda-root-py" }, "language_info": { "codemirror_mode": { "name": "ipython", - "version": 2 + "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", - "pygments_lexer": "ipython2", - "version": "2.7.11" + "pygments_lexer": "ipython3", + "version": "3.5.2" } }, "nbformat": 4,