1 line
201 KiB
Plaintext
1 line
201 KiB
Plaintext
{"cells":[{"cell_type":"markdown","metadata":{"id":"17-3KDFyzGCG"},"source":["# Lab 5 - Solutions 1: Linearly separable data"]},{"cell_type":"markdown","metadata":{"id":"SJvaJXVRzGCn"},"source":["Hopefully it is now clear to you that instantiating a Perceptron in Python is\n","a matter of just a few lines provided that you are happy to just accept all default values and are not too interested in optimisation and what's happening under the hood.\n","\n","A take-away point from this lab is that if you know that your problem is linearly separable then you can set a stopping criterion based on reaching zero misclassification error. If the problem is not linearly separable then you need to look at an alternative. The default is to look at whether accuracy is improving (although you have to be wary of overfitting - not so much a problem for these simple data, but care is needed when you start fitting more complex models, like multi-layer perceptrons). In Lab 5, Solutions 2, I provide an alternative solution based on the norm of the change in weight (see notebook Lab 5 Solutions 2)."]},{"cell_type":"code","execution_count":1,"metadata":{"id":"YH17VYKBzGCq","executionInfo":{"status":"ok","timestamp":1708965761839,"user_tz":0,"elapsed":866,"user":{"displayName":"Ben Evans","userId":"05311453112213240997"}}},"outputs":[],"source":["import numpy as np\n","from scipy.stats import norm\n","import matplotlib.pyplot as plt"]},{"cell_type":"markdown","metadata":{"id":"XSHT9_eDzGCv"},"source":["Generate the data"]},{"cell_type":"code","execution_count":2,"metadata":{"id":"aDgQNqVczGCz","executionInfo":{"status":"ok","timestamp":1708965763596,"user_tz":0,"elapsed":2,"user":{"displayName":"Ben Evans","userId":"05311453112213240997"}}},"outputs":[],"source":["nbpts1 = 100 # Number of points in first cluster (vary to explore effect of number of points)\n","mu1 = [2, 0] # the mean of the first cluster (this is a vector as we are in 2D) -- move it around\n","sigma1 = [[1, 0], [0, 1]] # the covariance matrix of the first cluster -- this one is not correlated\n","input1 = np.random.multivariate_normal(mu1, sigma1, nbpts1)\n","\n","nbpts2 = 100 # Number of points in second cluster (vary to explore effect of number of points)\n","mu2 = [10, 0] # the mean of the second cluster (this is a vector as we are in 2D) -- move it around\n","sigma2 = [[1, 0], [0, 1]] # the covariance matrix of the second cluster\n","input2 = np.random.multivariate_normal(mu2, sigma2, nbpts2)\n","\n","data = np.concatenate((input1, input2), axis=0) # Concatenate the data by row\n","classes = np.concatenate([np.zeros(nbpts1, dtype=int), np.ones(nbpts2, dtype=int)]) # integers"]},{"cell_type":"markdown","metadata":{"id":"ojCaQgmBzGC3"},"source":["2D plot of the data to check linear separability"]},{"cell_type":"code","execution_count":3,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":448},"id":"nOPstskszGC5","outputId":"16c75b27-87a2-4d85-e0af-c50197b6b5f9","executionInfo":{"status":"ok","timestamp":1708965768428,"user_tz":0,"elapsed":600,"user":{"displayName":"Ben Evans","userId":"05311453112213240997"}}},"outputs":[{"output_type":"execute_result","data":{"text/plain":["<matplotlib.collections.PathCollection at 0x7fc1a8242050>"]},"metadata":{},"execution_count":3},{"output_type":"display_data","data":{"text/plain":["<Figure size 640x480 with 1 Axes>"],"image/png":"iVBORw0KGgoAAAANSUhEUgAAAiIAAAGdCAYAAAAvwBgXAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAAA060lEQVR4nO3df3BV9Z3/8de9IYkpS7LGn1CSLnG1rhUCFbFIZ6srq3W6qOO0nTpuRXE6ZAdQRC3gDmWdXU1VCh2zDtFdFcettt1Z0epOtyK1OFRFBUPxF5TiKooUxW0u4jWE5Hz/ON+b3CT33twf55zP55zzfMww11xvcj/3JOdz3p/35/35nITjOI4AAAAMSJpuAAAAiC8CEQAAYAyBCAAAMIZABAAAGEMgAgAAjCEQAQAAxhCIAAAAYwhEAACAMWNMN6CQ/v5+7du3T+PGjVMikTDdHAAAUATHcXTo0CFNmDBByWThnIfVgci+ffvU1NRkuhkAAKAMe/fu1cSJEwu+xupAZNy4cZLcD1JfX2+4NQAAoBipVEpNTU0D1/FCrA5EMtMx9fX1BCIAAIRMMWUVFKsCAABjCEQAAIAxBCIAAMAYAhEAAGAMgQgAADCGQAQAABhDIAIAAIwhEAEAAMYQiAAAAGMIRAAAgDFWb/EOAAhOOi11dkp79kgtLVJbm1RXZ7pViDoCEQCA0mlp1ixp+3apqkrq65Meflj67W8JRuAvpmYAAOrsdIOQ/n6pt9d93L7dfR7wE4EIAEB79riZkGxVVe7zgJ8IRAAAamlxp2Oy9fW5zwN+IhABAKitTWptlZJJqbrafZw61X0e8BPFqgAA1dW5hamsmkHQCEQAAJLcoOOGG0y3AnHD1AwAADCGQAQAABhDIAIAAIwhEAEAAMYQiAAAAGMIRAAAgDG+BiJr167VlClTVF9fr/r6es2cOVO//OUv/XxLAAAQIr4GIhMnTtQPf/hDbd26Va+88or+5m/+Rpdeeqlef/11P98WAACERMJxHCfIN2xsbNRdd92la6+9dtTXplIpNTQ0qLu7W/X19QG0DgAAVKqU63dgO6v29fXpP//zP3X48GHNnDkzqLcFAAAW8z0Q2bFjh2bOnKnPPvtMf/Znf6b169frjDPOyPnanp4e9fT0DHydSqX8bh4AADDI91UzX/ziF9XV1aUtW7boH/7hHzR37ly98cYbOV/b3t6uhoaGgX9NTU1+Nw8AABgUeI3I7Nmzdcopp+jee+8d8f9yZUSampqoEQEAIESsrBHJ6O/vHxJsZKutrVVtbW3ALQIAAKb4GogsX75cF198sZqbm3Xo0CE98sgj+s1vfqNf/epXfr4tAAAICV8DkQMHDuiqq67SBx98oIaGBk2ZMkW/+tWv9Ld/+7d+vi0AAAgJXwOR+++/388fDwAAQo57zQAAAGMCL1YFAMAW6bTU2Snt2SO1tEhtbVJdnelWxQuBCAAgltJpadYsaft2qapK6uuTHn5Y+u1vCUaCxNQMACCWOjvdIKS/X+rtdR+3b3efR3AIRAAAsbRnj5sJyVZV5T6P4BCIAABiqaXFnY7J1tfnPo/gEIgAAGKprU1qbZWSSam62n2cOtV9HsGhWBUAEEt1dW5hKqtmzCIQAQDEVl2ddMMNplsRb0zNAAAAYwhEAACAMQQiAADAGAIRAABgDIEIAAAwhlUziC/udgUAxhGIIJ642xUAWIGpGcQTd7sCACsQiCCeuNsVAFiBQATxxN2uAMAKBCKIJ+52BQBWoFgV8cTdrgDACgQiiC/udgUAxhGIAACswzY/8UEgAgCwCtv8xAvFqgAAq7DNT7wQiAAArMI2P/FCIAIAsArb/MQLgQgAwCps8xMvFKsCAKzCNj/xQiACALAO2/zEB1MzAADAGAIRAABgDFMzUcf2hAAAixGIRBnbEwIALMfUTJSxPSEAwHJkRKIssz1hf//gc2xPCCBkmGGONgKRKGN7QgAhxwxz9DE1E2VsTwgg5Jhhjj4yIlHG9oQAQo4Z5ugjEIm6MG5PyIQwgP+PGeboSziO45huRD6pVEoNDQ3q7u5WfX296eYgCLkmhFtbmRAGYipXlzB1qrR5M12CzUq5fvtaI9Le3q6zzz5b48aN04knnqjLLrtMO3fu9PMtEXaFJoTTaWnNGmnRIvcxnTbdWgA+y8wwr1olzZ/vPhKERIuvUzObNm3SggULdPbZZ+vo0aO65ZZbdOGFF+qNN97Q2LFj/XxrhFW+CeFduyidB2IqjDPMKJ6vgcj//M//DPl63bp1OvHEE7V161b99V//tZ9vjbDKNSHc2+sOgV5/XXKcwSAlkymhhwKGoMwKYRJosWp3d7ckqbGxMef/7+npUU9Pz8DXqVQqkHbBIm1tbqajq8sNOiQpkZBee23kaymdB0Zg3w2ETWD7iPT392vx4sWaNWuWzjzzzJyvaW9vV0NDw8C/pqamoJoHW2QmhC+5ZPC5fPXUlM4DI9i67wYlXsgnsIzIggUL9Nprr2nz5s15X7N8+XItWbJk4OtUKkUwEkd1dVJTk7sJW2/v4POJhBuUVFcPls6zORswhI37bpClQSGBBCILFy7UU089peeee04TJ07M+7ra2lrV1tYG0STYLletSCLhZkqampj4BvKwcd+N7CwNJV4YztdAxHEcLVq0SOvXr9dvfvMbTZo0yc+3Q5RkakWGbx7w6KMEH5WgijHy8p06JpOHNmZpYA9fA5EFCxbokUce0RNPPKFx48Zp//79kqSGhgbV0fmhELan9x758Viw8dSxMUsDe/i6s2oikcj5/IMPPqirr7561O9nZ1XAQ2vWSDfdNHRYmky6O0SRH4eP2B01fkq5fvs+NQPAEuTHYYiNWRrYg5veAXFBfhwGxWl3VEqxSsNN7+A/zko7kB8HfMd9O13WTM0AFEhahPw44DuWKpeOQAT+4qy0S5zy44ABlGKVjkAE+XkxpRKms5IpJAAVohSrdAQiyM2rKZWwnJVMIQHwgI0bytkusJveIWS8unNWW5tbqZVMuveISSbtPCttvVMYgLxsvJFephRr1Spp/nz3kXrwwsiIIDevplTCUiAZpikkAFYnMSnFKg2BCHLzckolDGdlWKaQAEiiDj5KmJpBbmGZUvFK3D4vEHKZJGY2kpjhREYEuYVlSsUrcfu8QMiRxIwOdlYFAIQOGwXbjZ1VAQCRRhIzOghEAAChFIY6eIyOQAQAYC02PI4+AhEAgJVs3isE3iEQAQAYlyvzkWuvkK4u6YorpKYmMiRRQSCCaCKfC4RGvszHOeeM3PDYcaQnnnC3+yFDEg0EIoge8rlAqOTbJbW5eeReIRm9vYOvYzfVcGNnVUQPN7ADQiXfLqnjxw/d8FiSEomRr2M31XAjI4Lo4QZ2QCC8mgHNt0vqaadJq1cPvsfevdKTT7rTM9mvYzfVcCMQQfSw9zPgOy9nQNva3O8dvktqJrDJTLvk202VW0KFG1u8I3rY+xkoSiUZjTVrpJtuGpp4TCalVavKq9coti3UoYcDW7yjOFE9o9n7GRhVpRkNr2dAi90lld1Uo4dAJK6ivrKE3gooKN9KlWJXoDADCq+waiaubFtZkk67ud5Fi9zHdDra7wsYlm+lSrEZjba2oStakknqNVAeMiJxZdPKElPZmahnhYACKs1oMAMKr5ARiSub8qqmsjO2ZYWAAHmR0cjMgHZ0uI8EISgHGZG4KrReLmimsjM2ZYWAgJHRgC0IROLKi17I792M/M7O2JQVAgygphs2YB8RlCdXfUVra3n1Fab2/WC/EQDwBfuIwH+Vrv3LZipHTG4aAIwjEEF5TO1m5DVy0wBgFIEIyhOl+oqo7jALWIZTDblQI4LymKzr8LIn87LWBUBenGrxQo1I2IRxmGCivsKPDci8rHUBIsCv7ohTDfkQiJgW5t09g66v8KMnYy8RYICf3VGlp1oYx2teiMPnJhAxjWFC8fwIGqJU6wJUyM/uqJJTLczjtUrE5XOzxbtpld55Kk78CBqG73OdSEjjx0u7dnETPMSOn91RJVvKh+1uDF7dSzNsn7tcZERMY0RePD+2pc+uddm1S3rySemDD6T774/u8APIw8/uqJKysjDNoHqZxQjT566ErxmR5557TnPmzNGECROUSCT0+OOP+/l24cS9tIuX6clWrZLmz3cfvVilk6l1Oe00NwiJ+vADyKPU7qjUkX+5N8kL03jNyyxGmD53JXzNiBw+fFitra2aN2+eLr/8cj/fKrzY3bOwXJVaftXOxGX4AeRRSncUZP2CTffoHI2X3UiYPnclfA1ELr74Yl188cV+vkU0sLtnbkFXasVl+IFYKnb1RbHdUZB19mEar3nZjYTpc1fCqhqRnp4e9fT0DHydSqUMtgbGBb2iKC7DD8SOHzF90AlEU+O1UpfPet2NxGGcalUg0t7erltvvdV0M2ALEz1dHIYfiB0/YnpbE4he7rtRTgBHN1I6qwKR5cuXa8mSJQNfp1IpNTU1GWwRKlJJj5BOS3v3utVe2fzu6eIw/EDs+BHT25hA9DrzU24ARzdSGqsCkdraWtXW1ppuBrxQSY+Q+d6ursHnEgn3n+meDgghP7IX5Y78/dwp1OvMD/XrwbAqEIElvOgpKukRMt+bfT9Gx5EuuUR69FFynECJ/MpelDry97v+3OvAwdbpp6jxNRD55JNPtHv37oGv3377bXV1damxsVHNzc1+vjXK5VVPUUmPkOt7q6ulpiaCEKAMttQt+F1/XihwKGd8ZeP0UxT5Goi88sorOv/88we+ztR/zJ07V+vWrfPzrVEur3qKcocSpmpDgIizoW7B76mOfIHD3Lnlja9K3VfFdKAXVr4GIuedd56c7PQ67OdVT1HOUILaECDSvJ7qyHXxzxU4VDK+KiaAi8vN6fxCjQiG8qqnKDUXnE5LV1whvfrq0OepDQEiw8upjkIX/+GBg9+ZGG6iXhkCEQzlZU9RbC4406MMD0IkakOACPGyVqWUi7/fRaesrqkMgQiGMlHVlulRcqE2BIgUr2pVSrn4+110yuqayhCIYKSgq9py9SgZ1IYAyKHYi3+mjuScc6TmZmn8ePdG216Or1hdUxkCEZiXq0eRpEsvpTYEQE7FXPxz1ZG0tkqrV3vbrdiyPDqsEo7Fy1pSqZQaGhrU3d2t+vp6082BX3L1FlOnSps3cyYDyGu0JbNr1kg33TQ02ZpMSqtWUUTqt1Ku32REYB7DCQBlGG0WmSLScCAQgR1s2G0JQKRQRBoOSdMNAADAD21tbk1IMunuBJBMUkRqIzIisBd7JgOoALO+4UAgAn9UGkSwZzJgtXynuG3jh1L2VbSp3XFCIALveRFEsGcyIiKKF7h8p/gzz0izZ4dv/MC4xyxqROC97CCit9d9zAQRxcqUu2ej3B0hk7nA3XSTdO+97uOsWe7zYZbvFJ83r/JTv1TptLtMd9Ei97GcY+tFl4XykRGB97xYM0e5OyIgqom9Qqe4l8tlR8smeZXJYJmvWQQi8J4XQYTpPZOjmE9H4KJ6gSt0ir/+eu7nS1VMkOFVoMe4xyymZuA9L9bMZcrdV62S5s93H4PaaTWq+XQELqoXuHyn+AMPeLdctpjpklwzuJL02GOlna4s8zWLjAi859WaOVObnEU1n47AmU7s+aXQKe7Vctliskm5Ar3+frcNs2YVP0XDMl+zCETgj+FBRKaiLPssl+w886OaT0fgonyByzdO8Gr8UEw2KRPodXVJ2XdNc5zSxw5R3Nw5LDPMBCLwRqG/+FyTvQ895P6/HTvsWy83caJ09OjQ56KQT4cRUbzABaGYbFIm0LvwQun55/0dO4Tlop4RpiXJBCKo3Gh/8bmmOn73O/fRceya/kinpUcfHTq8kqQpU8KfTwcCVsnFu9hsUl2ddPnlbiCSzcuxQ5gu6hlhmmEmEEHlRvuLzzXVkUi4j9kXfBumPzo73SxNtkRCuuIKe3scwEJeXLyLzSb5XYsTpot6RphmmAlEULnR/uJzTfYOzzhI3g9hyhmK5fosY8ZIe/d60y4gJoK8ePtdixOmi3pGmFZsEYigePku7qP9xecarkyZ4gYj2TUiXg1hKhmKhensBSwW9MXb61qc7O5u797wlY2FacUWgQiKU+jiPtpffL7hiuTPEKajY7CMvtShWJjOXiAg5SQYwxzT5+rujjlG+uwzN0Eahm4hTCu2Eo6TK0duh1QqpYaGBnV3d6u+vt50c+JtzRp3Y6/s4U0y6W40dsMN3paUV/Kz0mmpuVn66KOhz1dXuxujdXT4+/5AxOS6KLe2jp5gzPV9U6cGty9hJfJ1d3PmSE1NdAvFKOX6TUYExRktz+pVXrTSCrfOTungwZHPHz1a/FCM9ZbAgHJrPcI0Ih8uX3fX1FTcWAalIRCJAy9G+EHlWSutcMv0IMMndI8/3u48KmCpSmo9whrTh3laKYwIRKLOqwXwQdVOVFrh1tIy9Hsld/ntjTfaORRjGgiWKLcWPYooFQsWgUjUebWGLqg8a6W9Xr4e5Lrr7Lvoh3GXJERSJbXoURTmaaUwIhCJOi/X0AWRZ6201yu0Qse2i34Yd0lCJI32p2jTRTmo8URYp5XCiEAk6sKUV830MOec4658GT9eOu200nuaXD3ImjX2XfTDuEsSIimoWvRKkUSMJgKRqLMtr5pvOJNvjeDq1d70MDZe9IMKEm2bkoJ1wjJeIYkYTQQiUWfTZGeh4YzfPYyNPW0QQSJDSBTBtvFKPvluW/Xgg+5/E2OHE4FIHNiSVy0UbPidsbCxpw0iSGQIiSLYNF4pJN944vXX3Q3IiLHDiUAEwSkUbPidsfCqp/V6msPvINHGKSlYyZbxSiHZ44lEYrDLIMYONwIRBKdQsBFExqLSnjaM0xw2TkkBZcoeTzz4oJsJ8TrGpqQqeNxrBsEZ7eYTtvcAo91vx0ZhvuEHUIAfp2O599XBSNxrBt7yKkAYbXrE9txwGKc5wjL5j1grtYtJp6XeXqmx0b21VOa0rDSJSkmVGQQitrEtK+D1dMTwYCOddoc2QX/eON3X3PYAD7FWahcz/PWSdOyx7l0crruusu4jjGONKCAQsYmNNQh+DBEyQcCuXdKTT0r79kljxrif98EHpS98QXrnHfcC/8AD7rDHS+UeZxtX3gAhV2oXk+v1Bw9KNTWVd5NhHWuEHYGITWzMC3o9RMgOArLL3nt73ccdO9x/mf+eOFF67z1vg5E43tccMGS05GOpXYwfXVKmfU1N0uTJbtcTh7GGLQn4QAKRe+65R3fddZf279+v1tZWdXR0aMaMGUG8dbjYmBf0eoiQHQQUI52W5s2THn+8vPfLJY73NQcMKCb5WGoX42WXlKt9kydLt9/ujn+iPNawKQGf9PsNfvazn2nJkiVauXKltm3bptbWVl100UU6cOCA328dPjbmBdva3LLxZFKqrnYfKxkiZIKAUr/HSzYeZyCCsscdvb3uY1eXdOGFbmlYOl16F+Nll5SrfTt2uNM8HR3umCOKQYiU+7NnEsNB8z0QWb16tb73ve/pmmuu0RlnnKHOzk597nOf0wMPPOD3W4eP1xd9L2SmI1atkubPdx8rWfqZKwiQ3M9a6Hu8ZONxBiIo17jDcaTnn3eX3s6a5T5XShfjZZeUq32mk9BBsemz+zo1c+TIEW3dulXLly8feC6ZTGr27Nl64YUXRry+p6dHPT09A1+nUik/m2cfW2sQvJyOyFXwOX68NGeO9PnPuznRdHroe3sdtNp6nIGIyTfuyMyKdnVJd98tLV1aWhfjVZcU5+SoTZ/d1w3N9u3bp89//vN6/vnnNXPmzIHnv//972vTpk3asmXLkNf/0z/9k2699dYRP4cNzXxgskqp0Ht//LFbE5L5f36smrGNl78LW6rP4Imw/zqz6xCk3KVhJ5zgLpLL97n8PAbZ7Usm3Qvx8ce72ZpFi8J1rEvl916HpWxoZlUgkisj0tTURCDitThsH1jODkkmenwvfxdx+L3GSBC/Tr/+7IevRHEc6amn3IvccImE9KMf5c5wBHUM7r7bbcNHHw3uJBCHU8fPbq+kndEdH/X09DhVVVXO+vXrhzx/1VVXOZdccsmo39/d3e1Icrq7u31qYUytXu04yaTjuP2D+y+ZdJ+Pgk8/dZxp09zPVF3tPk6b5j7vxeu95OXvIuq/15jx+9fp1599vp978KDjHH/80M8jua9ZuDD3zwrqT5pTx3ulXL99LVatqanRWWedpY0bNw4819/fr40bNw7JkCBgNlUp+aHUcnCT5eNe/i6i/nuNGb9/nX792ef7uQ89JN18s5sByVaoLqHYY5DZoHnRosHVOKX8f04ds3zfR2TJkiWaO3eupk+frhkzZujHP/6xDh8+rGuuucbvt44WL3NoNlUp+cGPHZL8ymF6+buI+u81Zvz+dfq1bVGhn3vnndJPf5p/c+Lhp9nEiaMfg9H2w/BjLxN4LIAMjdPR0eE0Nzc7NTU1zowZM5wXX3yxqO9jaub/8zqHmuvnffnLwUxFBKHUPOtor/dz6sbL30XUf68x4/ev06/piGJOp9Wr3emY1asHP0+uz9va6v4rdAxGe79iPmdYTp18x85GpVy/A9lZdeHChVq4cGEQb2WG34WOXm/9HvXlq6XeE2a01/u59b6Xv4uo/15jxs9fp193r02npSNHpOOOG1r4mf1z8y29zXWa7dghtbe72/2UskW85N62SnJvaTVa5sePY+31ZcGmnVC9xr1mKhXEX4cfOdQob1Veaq8y2uv93nrfy99FlH+vMeTHr9Ovu9cOXworSQ0Nbl1IMUth851me/e6u5zmk2tapb9feuMNdxnu+PHFTbt4eaxzXRYeeki64oryt46/+2533xXHsedWZF4hEKmUn6PlTEjd1SUdPTr0/5mewLR9g4NSe5VCr2cCGRHi191r8/3c5593g4nRuolCp1mh7iY7oSkNvnfmZ+3bJ02YIH3wQXA3sst3Wfjd7wazRKWMV9Npd3nx8M02ksloFNQSiFTKr9Hy8JA68xc4Zow3OVQv21borPIiYPE76Mn186XB5yZOjNctORFpQRapOo70xBPu1MpoF998my6/8YZ06qlDA4nsn5Od0HzwQff12QHNmDHuxs2nnRbcuCnfdJHjDN5ovJTxamenO9U1XFTGQwQilfJrtJwrpE4kpK98Rbr8crMZiGKzQF5MW3k59ZUv4MiVQ5WGBh6TJ7uT1cUM7QCLjdZllRv359vOXSru4psdUOzaJT35pJvNeOCBwW4mX3eTndC86aaRn+2004Kdvih0LDJKCf727HEDqsxxzDj++IiMhwIoni1bKFbN+FVuvXCh+/OK3fknSMW2zYuyfK9K+/OtfLnjjpE/P5Fw/7G7ESKoUJdVyQKx4d+bOZXK6cJynfa5fs7wVSQHD9qx+mX4sRh+HErtUnIdj0TC7b5sZd2qmUjzq7Td5rqEYtvmRQ7YqzxyvizO2LEjf35mx6XsCVl2N0JEFOqy1qwpv+Rt+M/du9fNamSfRsV2YfmmNrJ/TlNT7mTpM8+4SU2T5WvDj0VTk/TII+XP7uZb2LdokZ+fIjgEIl7wo7S91CWoQSq2bV4EU14FZPkCmszPy5br9ku2BIGAB/J1WZXG/dk/N99N1YrpwvJNbWTK5aZOdR9zBU0PPeS+RyYI6OzMH4z4WX42/BgvWlT+e0V9ZT6BiK1s/ssrtm1eBFNeBWT5Apq/+zvp8OGhP3/KFLeXozgVEVLMRdfLRGwlXdjw0/7oUXflS6botK1N+v73cwdNu3YVV1YW9L4clY5Xo7wy39e771aqpLv3wU6Z3m/XLrfsffz4wZ6klIJVL1be5LvntVR41YxNQSBQhmLvYuv3reFLbXOhU3DNGrcwNTsQSSbdYOXJJ0c+v2qVeyHP/NzHHnM/f/YVMPt1qEwp128CEfjPltvT2773CeCTfBftXBddm06TQm3JFzTNmCHdf//QFSbV1dL8+e69bjLfI42sQcm8rtAGapW0OU5KuX4zNQP/+bnpWyminNsECiil9sOW02S0qZN8Uz+dnfmnl7K7olwqLQWL8jbsfkqabkAkjHaP6bjjHtuAUTYvwssnO2jo7XUfM+OXjEzQ1NHhPtbVucFIa6ub8amudh8zJV65uiLJfU326/xsM0YiI1IpQuDR+dkLkgcFRmXzIrx8yl3BU6hINldXlEhI557rzT6Rft+WKqoIRCply7SDzUrtBdNp9w5PTz3lfj1nTu67ZhEEAkWxeRFePpWMX/JNL+Xrip5+2ptjEcbMkw0oVq3UokXSvffmroyqpOIpaorNXKTT0syZg9VkGa2t0gsvFFc2T9k74KsgEpF+reDxs+2ltjnKCV2KVYNECFycYivgOjvdW1QO97vfjcwykQcFAhdUItKvLI6fxbiltJmE7iACkUqFcfLVZnv2uJO2wxN1icTIAIMgEDFQ6ajZ61F3kLPRtqzgKUUpYy5m9V0EIpUK4+SrzVpacm+x7jgjAwyCQERcpaNmP0bdYUtE2jr9Ebbj6CcCES/YHLbbehbm09bm3iwiV43I8ACDIBARV+mo2Y9Rd5gSkTZPf4TpOPqNQCTKbD4L86mrc4tSi1k1k3m9rUEgUKFKR835vn/XLrfWu5z4PUyJSJunP8J0HP1GIBJlNp+FhdTVSUuXuv+KEbasD1CkSkfNub7/6FH3XiwffFDe+CRMiUibpz/CdBz9RiASZTafhV4JY9YHKFKlo+Zc3z9+vBuEVDI+CUsi0vbpj7AcR78RiESZ7WehF8Ka9QGKUOmoOdf379rl3hQuyuOTDKY/woFAJMricBbGIeuDUKt05rDSUfPw71+zJvrjkwymP8KBQCTK4nAWmsj6UJOCItk4cxiH8Uk2pj/sxxbvCDe/9oEu5f1aW6lJQU623oWg3FiaGBzFYot3xEfQWR9qUlACW2cOy8kS2JjdKQbBk/0IRBB+QeZebb2ywEpRqhcPYwwe1uApbpKmGwCESpSuLPBdW5s7c5dMujflTibDW4+RicGz2RCDp9PuFNiiRe5jOj34/7KDp95e9zETPMEeZESAUsSt0g8ViVK9uI0x+GgZDxKY4UAgApQiSlcWBCIqqzZsjMFHmy6yMXjCSAQiQKmicmUBSmBjDD5axsPG4AkjEYgAAIpiWww+WsbDxuAJI7GPCMKNtXlAbAW9jRCKxz4iiAfW5gGxFuaMB2OoQQQisEepZ2YYNzYA4CnbpouKwRhqKAIR2KGcM5O1eYgJRs/RwhhqKAIR2KGcMzOItXlcAWAYo+eRwn5aMoYaikAEdijnzPR7bR5XAFiA0fNQUTgt2d9kKLZ4hx3KOTMzlWqrVknz57uPXpbLsz80LGDr1uqmFDotC233bpMobf3vBd8yIrfddpv++7//W11dXaqpqdGf/vQnv94KUVBudsPPSjXyp7AAo+eh8p2Wu3aFJ1MS5tU+fvAtEDly5Ii+9a1vaebMmbr//vv9ehs7hX0C04Ryzky/jzNXAFiA3UGHyndafvBBuKawwrjaxy++b2i2bt06LV68uKyMSCg3NMs1gdnaamdYHmZBHGd2S4IlGNsMyndazpgh3X+/O12TUV3tztp2dBhrbmyFdkOznp4e9fT0DHydSqUMtqZMVJYFI9dx7uqSLrxQuvxyb3pq8qewhG2j548/lubNGzwtHnhAamwM5r3znZadnSQww8qqQKS9vV233nqr6WZUxoa6gjgMn3IdZ8eRnn/e/efV5HBd3WAvt2eP+xjF4wkU6eOPpYkTBwtBd+xwv37vvWCDkeGBGVNY4VVSILJs2TLdcccdBV/z5ptv6vTTTy+rMcuXL9eSJUsGvk6lUmpqairrZxljuq4gCmvbipHrOEv5s1DlBmdxOZ5AkebNG7kaJZ12n3/8cSNNkkQCM8xKCkRuvPFGXX311QVf01LBBbe2tla1tbVlf78VTIfl5UwN2ZZBKaY92cdZGpoZkYZmoSoJJvIdz44OdwLalmMGBCRfcteGxWSZTEmmC/n+9zk9w6CkQOSEE07QCSec4FdbosF0WF7q1JBNI/502r3A33WXdPDg4OfI1Z7s4/zYY+5/Z9ddZ2ehKqnbyXc877rLzVGbPmZAwFpa3OmYXM/bwKYuDcXxbUOzd999V11dXXr33XfV19enrq4udXV16ZNPPvHrLe2RCcs7OtzHIP/6S50asmXTrkzvsWyZ9NFHblBx9Gjh9mSO89NPu1mnfLsDVbIjVK7jefSoGyiZPmaAAQ88MLJLq6tzn7eBLV0aiudbIPKDH/xA06ZN08qVK/XJJ59o2rRpmjZtml555RW/3hJS6Vv22bJtY6b3yLWafLT2jLbDaiV1O7mO5/HH23HMAAMaG93C1EsvlSZPdh+DLFQdjS1dGorn26qZdevWad26dX79eORT6tSQ6eLajFxTIKW0p9D6xkrqdnIdzyNHpFtuKb2NQEQ0NpotTC3Eli4NxfN9Q7NKhHJDs7CxZdOuNWukm24aGYgkEtK0aZW3x8uCXFuOGWCIbfXt2Tg97VDK9ZtABHb0KsN7j6NH3SmQG2+UrrvOvh7EhmMGGBCGzaM5Pc0jEEE40XsA1suVvEwm3dIsm3Z/hVmh3eIdMWfbPtYARrBh82hEi2+rZgAA0UMxKLxGRgQoBdNHsJzff6KmN4+uBKevnagRAYoVhio9xFpQf6JhvKBz+garlOs3UzNAsdiyEZYL6k/U5ObR5eL0tReBCFAstmyE5fgTzY9jYy9qRGA3m3LAVOnBcvyJ5sexsRc1IrCXbZO6bNkIy/Enmh/HJljsI4JoyJ7UzWxakJnUNbHfSKn38QECxp9ofhwbexGIwF427pzEpmuwHH+i+XFs7ESxKuzFpC4ARB6BCOzV1ubWhCSTUnW1+xiWnZMAAEVhagb2YlIXACIvnoGITUtCUVgpk7r8XoHAcLoVj2NVWPyW79q2JBTe4PcKBIbTrXhxPVZs8V4I+/xGE79XIDCcbsXjWI0ufoEI+/xGE79XIDCcbsXjWI0ufoEIS0Kjid8rEBhOt+JxrEZHjQj7/IZLvqovfq9AYDjdihfXY1XK9Tt+gYhECXNYjVb1xe8VCAynW/HieKwIRBBNa9ZIN900dMv3ZFJataqyfZvj2EsAgI+46R2iyY97z+TKsjz8cPTX1gGAJQhEMKiczECQ2QQ/qr5su8MvAMQMgQhc5WQGgs4mtLW5P3941Vcl956x8Q6/ABAj8Vu+i9zK2XUn6J16MveeWbVKmj/ffay09Jy1dQBgFBkRuMrJDJjIJpRy75li+JFlAQAUjUAErnIyA1HIJnCHXwAwiuW7cJWz605cd+oBAI9FbRcBlu+idOVkBsgmAEDF4r6LABkRAAAM8muvRpNKuX6zagYAAIPifodeAhEAAAyKQt1/JQhEAAAwqK3NvX9nMilVV7uPcdpFgGJVxEPUStIBREbc6/4JRBB9cS9JB2A9r/dqDBOmZhB9QW9FDwAoGhkRWzB14B9ubAcA1iIQsQFTB/6Ke0k6AFjMt6mZ//3f/9W1116rSZMmqa6uTqeccopWrlypI0eO+PWW4VXM1EE67e56s2iR+5hOm2tv2MS9JB0ALOZbRuStt95Sf3+/7r33Xv3lX/6lXnvtNX3ve9/T4cOHtWrVKr/eNpxGmzogY1KZuJekA4DFfAtEvv71r+vrX//6wNctLS3auXOn1q5dSyAy3GhTB9kZk0ywksmYxLXMulRxLkkHAIsFumqmu7tbjY2Nef9/T0+PUqnUkH+xMNrUQdz3/wUARFZgxaq7d+9WR0dHwWxIe3u7br311qCaZI/Rpg4otgQARFTJd99dtmyZ7rjjjoKvefPNN3X66acPfP3+++/ra1/7ms477zz9+7//e97v6+npUU9Pz8DXqVRKTU1N3H03V43I1KnS5s3UOQAArFPK3XdLDkQ+/PBDHTx4sOBrWlpaVFNTI0nat2+fzjvvPH3lK1/RunXrlEwWPxtUygeJPPYZAQBP0J36z9dApBTvv/++zj//fJ111ln6j//4D1UNr3MYBYEIAMBLuRLMra0sQvRaKddv34pV33//fZ133nlqbm7WqlWr9OGHH2r//v3av3+/X28JAEBB3PHBPr4Vq27YsEG7d+/W7t27NXHixCH/z8ckDAAAeXHHB/v4lhG5+uqr5ThOzn8AAJjAIkT7cPddAEBscMcH+3DTOwBAbHDHB/sQiAAAYoU7PtiFqRkAAGAMgQgAADCGQAQAABhDIAIAAIwhEAEAAMYQiAAAAGNYvgtvcDtLAEAZCERQuVy3s3z4YW5nCQAYFVMzqBy3swQAlImMCEqTawqG21kCAMpEIILi5ZuC+c53uJ0lAKAsTM2gePmmYByH21kCAMpCRgTFyzcF89573M4SAFAWAhEUr6Ul/xQMt7MEAJSBqRkUr62NKRgAgKfIiKB4dXVMwQAAPEUggtIwBQMA8BCBCMKFreQBIFIIRBAebCUPAJFDsSrCg63kASByCEQQHpl9TLKxlTwAhBqBCMKj0D4mAIBQIhBBeLCPCQBEDsWqCA/2MQGAyCEQQbiwjwkARApTMwAAwBgCEQAAYAyBCAAAMIZABAAAGEMgAgAAjCEQAQAAxhCIAAAAYwhEAACAMQQiAADAGAIRAABgDIEIAAAwhkAEAAAY42sgcskll6i5uVnHHHOMxo8fr+9+97vat2+fn28JAABCxNdA5Pzzz9fPf/5z7dy5U//1X/+lP/zhD/rmN7/p51sCAIAQSTiO4wT1Zr/4xS902WWXqaenR9XV1aO+PpVKqaGhQd3d3aqvrw+ghQAAoFKlXL/HBNQmffzxx/rJT36ic889N28Q0tPTo56enoGvU6lUUM0DAAAG+F6sunTpUo0dO1bHHXec3n33XT3xxBN5X9ve3q6GhoaBf01NTX43DwAAGFRyILJs2TIlEomC/956662B199888169dVX9fTTT6uqqkpXXXWV8s0GLV++XN3d3QP/9u7dW/4nAwAA1iu5RuTDDz/UwYMHC76mpaVFNTU1I55/77331NTUpOeff14zZ84c9b2oEQEAIHx8rRE54YQTdMIJJ5TVsP7+fkkaUgcCAADiy7di1S1btujll1/WV7/6VR177LH6wx/+oBUrVuiUU04pKhsCAACiz7di1c997nN67LHHdMEFF+iLX/yirr32Wk2ZMkWbNm1SbW2tX28LAABCxLeMyOTJk/XrX//arx8PAAAigHvNAAAAYwhEAACAMYHtrApAUjotdXZKe/ZILS1SW5tUV2e6VQBgDIEIEJR0Wpo1S9q+Xaqqkvr6pIcfln77W4IRALHF1AwQlM5ONwjp75d6e93H7dvd5wEgpghEgKDs2eNmQrJVVbnPA0BMEYgAQWlpcadjsvX1uc8DQEwRiABBaWuTWlulZFKqrnYfp051nweAmKJYFQhKXZ1bmMqqGQAYQCACBKmuTrrhBtOtAABrMDUDAACMIRABAADGEIgAAABjCEQAAIAxBCIAAMAYAhEAAGAMgQgAADCGQAQAABhDIAIAAIwhEAEAAMYQiAAAAGOsvteM4ziSpFQqZbglAACgWJnrduY6XojVgcihQ4ckSU1NTYZbAgAASnXo0CE1NDQUfE3CKSZcMaS/v1/79u3TuHHjlEgkTDcnMKlUSk1NTdq7d6/q6+tNN8cIjgHHIO6fX+IYSBwDKZzHwHEcHTp0SBMmTFAyWbgKxOqMSDKZ1MSJE003w5j6+vrQ/NH5hWPAMYj755c4BhLHQArfMRgtE5JBsSoAADCGQAQAABhDIGKh2tparVy5UrW1taabYgzHgGMQ988vcQwkjoEU/WNgdbEqAACINjIiAADAGAIRAABgDIEIAAAwhkAEAAAYQyBimXvuuUd/8Rd/oWOOOUbnnHOOXnrpJdNNCkx7e7vOPvtsjRs3TieeeKIuu+wy7dy503SzjPrhD3+oRCKhxYsXm25KoN5//339/d//vY477jjV1dVp8uTJeuWVV0w3KzB9fX1asWKFJk2apLq6Op1yyin653/+56Lu2xFWzz33nObMmaMJEyYokUjo8ccfH/L/HcfRD37wA40fP151dXWaPXu2fv/735tprE8KHYPe3l4tXbpUkydP1tixYzVhwgRdddVV2rdvn7kGe4RAxCI/+9nPtGTJEq1cuVLbtm1Ta2urLrroIh04cMB00wKxadMmLViwQC+++KI2bNig3t5eXXjhhTp8+LDpphnx8ssv695779WUKVNMNyVQ//d//6dZs2apurpav/zlL/XGG2/oRz/6kY499ljTTQvMHXfcobVr1+pf//Vf9eabb+qOO+7QnXfeqY6ODtNN883hw4fV2tqqe+65J+f/v/POO3X33Xers7NTW7Zs0dixY3XRRRfps88+C7il/il0DD799FNt27ZNK1as0LZt2/TYY49p586duuSSSwy01GMOrDFjxgxnwYIFA1/39fU5EyZMcNrb2w22ypwDBw44kpxNmzaZbkrgDh065Jx66qnOhg0bnK997WvO9ddfb7pJgVm6dKnz1a9+1XQzjPrGN77hzJs3b8hzl19+uXPllVcaalGwJDnr168f+Lq/v985+eSTnbvuumvguT/96U9ObW2t8+ijjxpoof+GH4NcXnrpJUeS88477wTTKJ+QEbHEkSNHtHXrVs2ePXvguWQyqdmzZ+uFF14w2DJzuru7JUmNjY2GWxK8BQsW6Bvf+MaQv4e4+MUvfqHp06frW9/6lk488URNmzZN//Zv/2a6WYE699xztXHjRu3atUuStH37dm3evFkXX3yx4ZaZ8fbbb2v//v1DzoeGhgadc845se0fJbePTCQS+vM//3PTTamI1Te9i5OPPvpIfX19Oumkk4Y8f9JJJ+mtt94y1Cpz+vv7tXjxYs2aNUtnnnmm6eYE6qc//am2bduml19+2XRTjNizZ4/Wrl2rJUuW6JZbbtHLL7+s6667TjU1NZo7d67p5gVi2bJlSqVSOv3001VVVaW+vj7ddtttuvLKK003zYj9+/dLUs7+MfP/4uazzz7T0qVLdcUVV4TqRni5EIjASgsWLNBrr72mzZs3m25KoPbu3avrr79eGzZs0DHHHGO6OUb09/dr+vTpuv322yVJ06ZN02uvvabOzs7YBCI///nP9ZOf/ESPPPKIvvSlL6mrq0uLFy/WhAkTYnMMkF9vb6++/e1vy3EcrV271nRzKsbUjCWOP/54VVVV6Y9//OOQ5//4xz/q5JNPNtQqMxYuXKinnnpKzz77rCZOnGi6OYHaunWrDhw4oC9/+csaM2aMxowZo02bNunuu+/WmDFj1NfXZ7qJvhs/frzOOOOMIc/91V/9ld59911DLQrezTffrGXLluk73/mOJk+erO9+97u64YYb1N7ebrppRmT6QPrHwSDknXfe0YYNG0KfDZEIRKxRU1Ojs846Sxs3bhx4rr+/Xxs3btTMmTMNtiw4juNo4cKFWr9+vX79619r0qRJppsUuAsuuEA7duxQV1fXwL/p06fryiuvVFdXl6qqqkw30XezZs0asWx7165d+sIXvmCoRcH79NNPlUwO7Z6rqqrU399vqEVmTZo0SSeffPKQ/jGVSmnLli2x6R+lwSDk97//vZ555hkdd9xxppvkCaZmLLJkyRLNnTtX06dP14wZM/TjH/9Yhw8f1jXXXGO6aYFYsGCBHnnkET3xxBMaN27cwNxvQ0OD6urqDLcuGOPGjRtREzN27Fgdd9xxsamVueGGG3Tuuefq9ttv17e//W299NJLuu+++3TfffeZblpg5syZo9tuu03Nzc360pe+pFdffVWrV6/WvHnzTDfNN5988ol279498PXbb7+trq4uNTY2qrm5WYsXL9a//Mu/6NRTT9WkSZO0YsUKTZgwQZdddpm5Rnus0DEYP368vvnNb2rbtm166qmn1NfXN9BHNjY2qqamxlSzK2d62Q6G6ujocJqbm52amhpnxowZzosvvmi6SYGRlPPfgw8+aLppRsVt+a7jOM6TTz7pnHnmmU5tba1z+umnO/fdd5/pJgUqlUo5119/vdPc3Owcc8wxTktLi/OP//iPTk9Pj+mm+ebZZ5/Nef7PnTvXcRx3Ce+KFSuck046yamtrXUuuOACZ+fOnWYb7bFCx+Dtt9/O20c+++yzpptekYTjRHirPgAAYDVqRAAAgDEEIgAAwBgCEQAAYAyBCAAAMIZABAAAGEMgAgAAjCEQAQAAxhCIAAAAYwhEAACAMQQiAADAGAIRAABgDIEIAAAw5v8BISQMranjokEAAAAASUVORK5CYII=\n"},"metadata":{}}],"source":["plt.figure(1)\n","plt.scatter(input1[:, 0], input1[:, 1], c='r', s=12)\n","plt.scatter(input2[:, 0], input2[:, 1], c='b', s=12)"]},{"cell_type":"markdown","metadata":{"id":"F5Ux_YklzGDA"},"source":["should be linearly separable! If not, try again until it works or move means further apart."]},{"cell_type":"markdown","metadata":{"id":"HrViPzmCzGDC"},"source":["# Batch learning, default settings."]},{"cell_type":"code","execution_count":4,"metadata":{"id":"zC-nwPwszGDD","executionInfo":{"status":"ok","timestamp":1708965782892,"user_tz":0,"elapsed":740,"user":{"displayName":"Ben Evans","userId":"05311453112213240997"}}},"outputs":[],"source":["## Set up Perceptron for batch\n","from sklearn.linear_model import Perceptron\n","from sklearn.metrics import accuracy_score, hinge_loss, confusion_matrix # Python has a class of metrics -- very useful"]},{"cell_type":"code","execution_count":5,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":75},"id":"hrvOjPSizGDF","outputId":"e98140bc-c057-4c5f-f99f-f739e05bbe1e","executionInfo":{"status":"ok","timestamp":1708965794931,"user_tz":0,"elapsed":209,"user":{"displayName":"Ben Evans","userId":"05311453112213240997"}}},"outputs":[{"output_type":"execute_result","data":{"text/plain":["Perceptron()"],"text/html":["<style>#sk-container-id-1 {color: black;background-color: white;}#sk-container-id-1 pre{padding: 0;}#sk-container-id-1 div.sk-toggleable {background-color: white;}#sk-container-id-1 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-1 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-1 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-1 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-1 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-1 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-1 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-1 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-1 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-1 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-1 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-1 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-1 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-1 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-1 div.sk-item {position: relative;z-index: 1;}#sk-container-id-1 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-1 div.sk-item::before, #sk-container-id-1 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-1 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-1 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-1 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-1 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-1 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-1 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-1 div.sk-label-container {text-align: center;}#sk-container-id-1 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-1 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-1\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>Perceptron()</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-1\" type=\"checkbox\" checked><label for=\"sk-estimator-id-1\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">Perceptron</label><div class=\"sk-toggleable__content\"><pre>Perceptron()</pre></div></div></div></div></div>"]},"metadata":{},"execution_count":5}],"source":["clf = Perceptron() # Default without option: This will return a warning on max_iter and tol -- adjust to your liking\n","# You should check what the defaults are. For example, alpha=0.0001 but alpha is regularisation which is not something we discussed in the context of Perceptron. The learning rate eta0 = 1.\n","clf.fit(data, classes) # Learning."]},{"cell_type":"code","execution_count":6,"metadata":{"id":"MerGnj9RzGDH","executionInfo":{"status":"ok","timestamp":1708965800342,"user_tz":0,"elapsed":204,"user":{"displayName":"Ben Evans","userId":"05311453112213240997"}}},"outputs":[],"source":["pred_class = clf.predict(data) # Prediction after learning"]},{"cell_type":"markdown","metadata":{"id":"Q0_cAdImzGDK"},"source":["Print some statistics"]},{"cell_type":"code","execution_count":7,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"a4hgUqyOzGDM","outputId":"f3217753-5ade-4f38-8f58-09e54e0ce0b5","executionInfo":{"status":"ok","timestamp":1708965808510,"user_tz":0,"elapsed":215,"user":{"displayName":"Ben Evans","userId":"05311453112213240997"}}},"outputs":[{"output_type":"stream","name":"stdout","text":["Process completed in 8 epochs\n","Classification accuracy = 200\n","[[100 0]\n"," [ 0 100]]\n"]}],"source":["print(f'Process completed in {clf.n_iter_} epochs')\n","print(f'Classification accuracy = {accuracy_score(classes, pred_class, normalize=False)}')\n","print(confusion_matrix(classes, pred_class)) # Confusion matrix"]},{"cell_type":"code","execution_count":8,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"gHkX1HgxzGDR","outputId":"adbdb72e-54ec-42eb-ec5c-04760058fe4e","executionInfo":{"status":"ok","timestamp":1708965818231,"user_tz":0,"elapsed":488,"user":{"displayName":"Ben Evans","userId":"05311453112213240997"}}},"outputs":[{"output_type":"stream","name":"stdout","text":["Final weights are: 3.430 -0.349 -16.000\n","\n"]}],"source":["# Let's plot the weights\n","print(f'Final weights are: {clf.coef_[0][0]:.3f} {clf.coef_[0][1]:.3f} {clf.intercept_[0]:.3f}\\n')"]},{"cell_type":"code","execution_count":9,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":861},"id":"E9wshgVxzGDU","outputId":"80522c63-0ed8-4e64-81b6-900409769873","executionInfo":{"status":"ok","timestamp":1708965855211,"user_tz":0,"elapsed":906,"user":{"displayName":"Ben Evans","userId":"05311453112213240997"}}},"outputs":[{"output_type":"execute_result","data":{"text/plain":["<matplotlib.collections.PathCollection at 0x7fc1a118ffd0>"]},"metadata":{},"execution_count":9},{"output_type":"display_data","data":{"text/plain":["<Figure size 640x480 with 1 Axes>"],"image/png":"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\n"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":["<Figure size 640x480 with 1 Axes>"],"image/png":"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\n"},"metadata":{}}],"source":["# Plotting decision regions\n","\n","# First find out area of the graph\n","plt.figure(1)\n","plt.scatter(input1[:, 0], input1[:, 1], c='r', s=12)\n","plt.scatter(input2[:, 0], input2[:, 1], c='b', s=12)\n","axes = plt.gca()\n","(x_min, x_max) = axes.get_xlim()\n","(y_min, y_max) = axes.get_ylim()\n","# Generate a meshgrid over which to make predictions\n","xx, yy = np.meshgrid(np.arange(x_min, x_max, 0.01), np.arange(y_min, y_max, 0.01))\n","plt.figure(2)\n","Z = clf.predict(np.c_[xx.ravel(), yy.ravel()])\n","Z = Z.reshape(xx.shape)\n","plt.contourf(xx, yy, Z, alpha=0.4) # Will reveal decision boundary\n","plt.scatter(input1[:, 0], input1[:, 1], c='r', s=12)\n","plt.scatter(input2[:, 0], input2[:, 1], c='b', s=12)"]},{"cell_type":"markdown","metadata":{"id":"SnUpoJXJzGDV"},"source":["# Sequential learning, default settings."]},{"cell_type":"code","execution_count":10,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":950},"id":"X744tcPuzGDW","outputId":"fa21c386-b0d6-430c-dbb0-0266e4fdac42","executionInfo":{"status":"ok","timestamp":1708965932716,"user_tz":0,"elapsed":2156,"user":{"displayName":"Ben Evans","userId":"05311453112213240997"}}},"outputs":[{"output_type":"stream","name":"stdout","text":["Completed after 349 iterations or 2 epochs\n","Classification accuracy = 200\n","[[100 0]\n"," [ 0 100]]\n","Final weights are: 0.328 -0.143 -2.000\n"]},{"output_type":"execute_result","data":{"text/plain":["<matplotlib.collections.PathCollection at 0x7fc19fccf130>"]},"metadata":{},"execution_count":10},{"output_type":"display_data","data":{"text/plain":["<Figure size 640x480 with 1 Axes>"],"image/png":"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\n"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":["<Figure size 640x480 with 1 Axes>"],"image/png":"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\n"},"metadata":{}}],"source":["## Set up Perceptron for sequential\n","# Same code but use function partial_fit.\n","from sklearn.utils import shuffle # will use this function to shuffle presentation of data before each epoch\n","\n","clf = Perceptron(eta0=0.1, alpha=0.0) # Here I'm going to specify a value for the learning rate (consider changing it to see effect) and also setting regularisation to 0.0 as we do not need it.\n","\n","# Shuffle dataset\n","data, classes = shuffle(data, classes, random_state=0)\n","\n","plt.figure(3)\n","plt.scatter(input1[:, 0], input1[:, 1], c='r', s=12)\n","plt.scatter(input2[:, 0], input2[:, 1], c='b', s=12)\n","\n","num_iter = 0\n","while True:\n"," clf.partial_fit(data[num_iter % len(data), :].reshape(1, -1), [classes[num_iter % len(data)]], classes=np.unique(classes))\n"," # Learning over one sample\n","\n"," # Plot the new decision boundary\n"," axes = plt.gca()\n"," (x_min, x_max) = axes.get_xlim()\n"," (y_min, y_max) = axes.get_ylim() # We want to keep the plot to the same dimension\n"," plt.plot([x_min, x_max],\n"," [(-clf.coef_[0][0] * x_min-clf.intercept_) / clf.coef_[0][1],\n"," (-clf.coef_[0][0] * x_max-clf.intercept_) / clf.coef_[0][1]],\n"," 'k-')\n"," plt.ylim([y_min, y_max]) # Restrict plot to original dimension\n"," plt.xlim([x_min, x_max])\n","\n"," pred_class = clf.predict(data) # Prediction after learning\n"," num_iter += 1\n","\n"," # Check termination criterion\n"," if accuracy_score(classes, pred_class, normalize=False) == len(data): # We are done!\n"," break\n","\n"," # Shuffle at the end of each epoch\n"," if num_iter % len(data) == 0:\n"," data, classes = shuffle(data, classes, random_state=0) # Shuffle dataset\n","\n","\n","# Highlight last boundary\n","# Plot the new decision boundary\n","axes = plt.gca()\n","(x_min, x_max) = axes.get_xlim()\n","(y_min, y_max) = axes.get_ylim() # We want to keep the plot to the same dimension\n","plt.plot([x_min, x_max],\n"," [(-clf.coef_[0][0] * x_min-clf.intercept_) / clf.coef_[0][1],\n"," (-clf.coef_[0][0] * x_max-clf.intercept_) / clf.coef_[0][1]],\n"," 'r-', linewidth=3)\n","plt.ylim([y_min, y_max]) # Restrict plot to original dimension\n","plt.xlim([x_min, x_max])\n","\n","\n","# Print some statistics\n","print(f'Completed after {num_iter} iterations or {int(np.ceil(num_iter / len(data)))} epochs')\n","print(f'Classification accuracy = {accuracy_score(classes, pred_class, normalize=False)}')\n","print(confusion_matrix(classes, pred_class)) # Confusion matrix\n","\n","# Let's plot the weights\n","print(f'Final weights are: {clf.coef_[0][0]:.3f} {clf.coef_[0][1]:.3f} {clf.intercept_[0]:.3f}')\n","\n","\n","# Plotting decision regions\n","# First find out area of the graph\n","axes = plt.gca()\n","(x_min, x_max) = axes.get_xlim()\n","(y_min, y_max) = axes.get_ylim()\n","# Generate a meshgrid over which to make predictions\n","xx, yy = np.meshgrid(np.arange(x_min, x_max, 0.01), np.arange(y_min, y_max, 0.01))\n","plt.figure(4)\n","Z = clf.predict(np.c_[xx.ravel(), yy.ravel()])\n","Z = Z.reshape(xx.shape)\n","plt.contourf(xx, yy, Z, alpha=0.4) # Will reveal decision boundary\n","plt.scatter(input1[:, 0], input1[:, 1], c='r', s=12)\n","plt.scatter(input2[:, 0], input2[:, 1], c='b', s=12)"]},{"cell_type":"markdown","metadata":{"id":"4oOYmCnpzGDX"},"source":["Why is the decision boundary hanging out for much longer amongst the red class? Smaller x values lead to smaller weight changes according to the weight update rule. This is why one should normalise the data to be centred around 0. We will talk about this in week 7."]},{"cell_type":"code","execution_count":null,"metadata":{"id":"teRBLok6zGDX"},"outputs":[],"source":[]}],"metadata":{"kernelspec":{"display_name":"Python 3","language":"python","name":"python3"},"language_info":{"codemirror_mode":{"name":"ipython","version":3},"file_extension":".py","mimetype":"text/x-python","name":"python","nbconvert_exporter":"python","pygments_lexer":"ipython3","version":"3.7.4"},"colab":{"provenance":[]}},"nbformat":4,"nbformat_minor":0} |