{"id":9,"title":"Final_exam_formatted_1","description":null,"category":"Programming","source_filename":"Final_exam_formatted_1.docx","questions":[{"text":"Final Exam Sample Questions — L1 — What is the output of a classification algorithm?","explanation":null,"position":1,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"A continuous value","is_correct":false,"position":1,"images":[]},{"text":"Groups of similar instances","is_correct":false,"position":2,"images":[]},{"text":"A categorical value","is_correct":true,"position":3,"images":[]},{"text":"Majority class dominate predictions","is_correct":false,"position":4,"images":[]},{"text":"k-NN becomes sensitive to noise data","is_correct":false,"position":5,"images":[]},{"text":"Not affected at all. k-NN is robust to any values of k","is_correct":false,"position":6,"images":[]},{"text":"Small values of k produce the best results","is_correct":false,"position":7,"images":[]}]},{"text":"In K-NN:","explanation":null,"position":2,"is_review_required":true,"answer_text":null,"images":[{"filename":"image1.png","content_type":"image/png","data_base64":"iVBORw0KGgoAAAANSUhEUgAAAXIAAADaCAYAAAC7HMk+AAAAAXNSR0IArs4c6QAAAAlwSFlzAAASdAAAEnQB3mYfeAAAABl0RVh0U29mdHdhcmUATWljcm9zb2Z0IE9mZmljZX/tNXEAAERgSURBVHhe7Z1nkFzHde/P7mxGXiwWWORERILIAAkSFJiDSJq0KZHys8uynyWVQ8lf/MHf7A9+8ge/KpcVbJUs+5EKfmRBFEQ+AiBFigEkEhGInHMGdhEWwOaZ2de/M+jlYLmLTXd27sycRg12wr0d/t3973NPnz6nYMWKFa1iyRAwBAwBQyBjESjI2JpbxQ0BQ8AQMAQUgduIvLW1VeLxuOTn50teXp5ekPwdn/mdlHyNYWkIGAKGgCGQPgS+ROSDBw+WpqYmiUajSuIQ+tChQ6W+vl6am5tl+PDh+v3Vq1fbyD591beSDQFDwBAwBJTIIWaS05fLsmXL5I033pA9e/ao9P3oo4/Kiy++KP/0T/8k06ZNk/vuu0+l8Y8//ljWrVsnRUVFhqIhYAgYAoZAGhFok8gLCgrk4sWLcuLECRk4cKBK3xMnTpRFixbJ9evXpbKyUt+/9dZb+juk/9lnn6nkDrG3T7FYTCX2jn5LY3utaEPAEDAEsg4BJXIIF+Ldvn27jB8/XiXxSCSi0viuXbtk7NixMmTIECX6w4cPS1VVlV6DNN7S0vIlUPhtxIgR0tjYKDdv3jQyz7phYw0yBAyBMCFwm44cAkbCRh8OcaNK4e+cOXNk0KBBStwlJSVK8kjaXuruSBqfOnWqEn9tba0ReZh63OpiCBgCWYdAm44cqRz9+JIlS2Ty5Mm6mfkv//IvMmDAAN3s3LJli8ybN0/+7M/+TIn50KFDSviFhYUdgoLe3evesw41a5AhYAgYAiFC4DaJvLi4WHbv3q1SOTrzmpoauXbtmvzqV7+Sc+fOycGDB+Xee++VhoYG2bx5s0naIepIq4ohYAjkLgJtOnIgWL9+vZI40jmkXlpaqlL1gQMHlNjr6urkN7/5TdvvqFiCSPGYs1+Pxl1ZQeQWjjwww8+P5LuXe5MwybdkCBgChkBKEPiSRA6B++RVI8kmhmVlZYFWJNbsNk0HR2RgRbEUlLiFIUtIL9YSl4arUamrbpFIQZ7k5WdJwwLtfcvMEDAEgkAgrUf0o80xqZheKlO+Ui4DRxRLpMiZMWYJ38UdkTdej8rprbVyelOtOxIbcWQeRJdZHoaAIWAI3I5A2og82hSTIRMKZdZzFVIypECQYLGCyZrkFqTS4fky7Ylhrm0xObP5phQUB6OKyhqMrCGGgCEQCAJpIXJ04U4rLiPnlEnJsHxpafiyLXogrUtzJnHXrAL3lFE1v0wu7L0p8YZWyXdqFkuGgCFgCASJQHqIPO4IzZ3sLx7qbNGjMYm3JhxxZWNqibZK4cA8KSjLk8abziFZgUnl2djP1iZDIJ0IpIXIaTASeWteTGIS5V06MUhp2XmteW6hijmLHFpp0nhKwbbMDYEcRSB9RI56BXK79cpa/JOIXMSk8aztZ2tYHxFQfesXSWUeE3y6C2oaidyd/OSfEnn2SuQMTtqZxQ8d3R1rdp0h0AkCzP98iec7owfHBfnuEEaECcMm0624CAbdnRFIG5GrekXVDU5HHijLJVbxlliTRPKdT5i8iDS794VOKZ+n9n9+0cjT8rkOHX1BpFAKUNxrCm5hcRbkt9ppQ9EQMAQ6QqA1UiwtBYMkMmCYlA0cIlHnebXxeo0UtFyXSLQeojDgukAgvUSuJO70xwESZ4KE82TcsKlys+maXG+6ItNHzJPzN05IXbNz4OWInQR5lxUOkBkjF8qAwsFy4uo+qa470/Z7kCOHxSrIxSHIullehkD6EHAzP79QWorLZfCY6bLg3q/IqDFjpdG5ANm7Y4sc3rVZ5MY5ibTcNMk8zEQedzTORmA8wBUXgo44sn5gwjOy++JGmVJ4j0wbvkDe2PMDaYo1OqJOnMrhuooBVVJeNlIGFg2Ru0fdK2/s/pFcabjoJPPg1rc8nhDdP0uGgCHQDgE3N2L5zhXI8LHy4OPPyviJU52LkGYZOqxcKkaM1Dl6ZMv7TipvdDdm0RmTFAyE4BirF5VTItd/wT06kRvqjGg8KjMqF6mE/c6hVx2JN8jYIVOdJi4Rj7Shpc5J4GfljPOvXlxQJt9e+r9kUMkwqa53Urn7F1RC0YMKx5IhYAh8gQCH/+LxmDS4uT92zCQZPXaCNDe5T06og7KLiktk5t0L5NS+bRJvvCL5KuzZ5mdnYyitRC660YlpXnBEDmnqwuD6fNGYR+W1nf/bqU32yujBk2VaxTwpyCt0XhsjcvHmKalxpB2NN8uiUY84tct1991J3WihToElJHLdC2CBsWQIGAIgQJQxYh1IWblMmzDa7We54DbRL7BhHhe6+AeRSIG0BMgP2Yp+Gok84a88aL/liUXBreqtUdl5fp2TypfIwept0ugk8PqWG7qh6SXyJvfINqX8Hrmn6kF59+DP5FpDtZQ46TzIhUUNVmwgZuv8sXb1EAEkcTypfu1rX5OFCxeq4LT1yAW5crlGhg0fIdGWZueTyD0TO2Hr7Mlj0lhXK4X6WNvDgnLs8jQSeYLgWt0pT15BJfKMu44viZTJrvOfyvCy0fLMjG/J2gP/R6Juyc+XAtWTtzhd3LjB0+TFOd91hP+JNDtSH1pcoWTvIo0GVR3NR9upY9Fk8kCBtcwyCoGEOiWuwWpII0dWyqHDR2Tlqrdk2IQ5cu8DD8vAwYOdZB6Vo4f2y+6tn0p+s5uP+oRsc+dOnZ1WIsesSIk8QInV66QPXNwm1TfPyq5zn8rDU19ykniJfHrsTfcIl2hyzA2O+aNXyIUbJ5W4F419TK+93njVXdtx1KNezZoEg1syBHIWAcibF+EiZ86cqXF+33vvPamurpFjx4/JlVPHXACb63L14hkZXjnaWa3Uy8UzRyV2/aIUYrFyyxItZwHsRsPTSuQqPackJFyrbDjxdpvd+NoDryo5J9QmCVQKnU5u74VNsvv8emfl4mDQ+A/uIIJ7H+TC0qZaSajtLRkCOYOAblw6KZwANUjhS5cu1WDsn3/+uTQ1Ncnp02eUpEucqqW17oLUnrgiV45+rqrPAue6ozDurFXUUMBmTleDJq1EjkYl7v7jFXRS9Yhmq3YqKhEkDwhP1lyn5o9KtMHvjDMEE2VbMgRyAwHmFpHGCEgzYcIEGT16tBL4vn375MSJE0rU/MZ1vCflxd2hPV4eorapaCTenVGTNiJXjYNK46kJ8QY3J8YIZXzxvj0orc4Xym1jB8PvINOtxSrYQ09BVtDyMgSCQwChhYDso0aNUjInkHt1dbXGAvYbnW3kfdvx+3aEbfzdo05JG5FrLZXk0J/1qM7dvjhZ9d5dNXx3r+tuJfx4tHHZXcTsukxDwKtQqPfw4cNl3LhxqjohSPvRo0flzBlUKE6d6QjeUmoQSAuR33qYknjslrojYCE4NVD1MlczP+wlcHZb2BHw6kmCsFdVVSlR19bWqiSOgHb+/Hlpdn5T8tWcMFhLsLBj09/1SwuRO/s/d1pLpKkWInc66hRJ5P0NZkflufMM0lznnHPVt+qhB0uGQDYggJoEcsYmfMyYMUJQ9rNnz+qmJkR+6tQptU6B5C2lHoG0EDmqsbx4vtTsj8mw6UTPiTiPlYkNx2xKeRG2T1uleneztFwvdpGCsql11pZcRMDH1S0vL5fx48fLlStXlMBHjhwplZWVSuB8zmwpPJmIMkP4+hKRJ+8kJ5vh+Q2KNmuPPvoJLiyJSN3xfDmyulHGLi+S0nKizDvQsojMm25G5eLnLXJpa56UFRW27dDnIgFYmzMHgWQO8LX2evBhw4apkQK68HPnzulGJhI5nw87v0XoxjNXCkeYdMLXLcd6CQr3B0HCTei3ETmPRHQSj0usqP7xyf9F7+UJnSb2pcO4t3RAqVzfH5f9ZxqldISDLD9LpHK1SXeqo2vOL3pNoQxwj52FxWl5+Mkc9rCapgUB5jpzkb8k5jo8wF8953HrME9xcbFMmTJFfaQcO3ZMDh48qJYpd911l0rgkDrXZ7Qu3LkFiEdKnW90d/KUg4Ou7fnOY2phtC4R5CLEqY1d6LDp06fLsmXL5MMPP9TVdfny5bJkyRK1AV25cqWMGDFCnn76aW3O6tWrdUcae9DepsLCAhk4ZKA0N7ZI/bEWHTSBHsbpbcX6eB/+I7B9pH2DBrmAFe6vJUMgTAgwz5i76LYxEWTuQ+jHjx+XAwcOyI0bN9TqhMM8qE3gAF6QOBuYqFVQpfC5rq4u81UpzktqS9FQiQwZIzNmzZPKqrGuXTfl6N4dUnvuqBQ21UieC0IT1sNJyjAQKCsynYPuixfp2rVr8u6778pzzz0nTzzxhK7AJ0+e1M595pln5Ic//KHe25dVuMBFlY8MyJeSst4vCGGaIMl16dheNqy1tXrlEgLMWYh8zpw5OreRtEn33HOPoD7ZunWrcsLQoUOVwLFGaWxsVFLnu4sXL8qlS5eyZkMzFimR/EGjZOlDz8jsuQvaOG3y1Bnyu9VvyNXjO6QohlQeTt2vEjmdCiGvXbu2bWWFhPbs2aPfP/TQQ/q3vr5efSSwEn/jG9/QHWpW474mykpW2fQ1P7vfEDAEOkcAaRzpG0Lm2Dw+UNBtkzAhXLx4sapKSkpK5OrVq/p0zj3Yh3NKkw1NyB3e6It6NTx95OIX5BXJ+EnTZeacueoWoKW5RTmpctRombvkfvnYSeVxIow5r6phlMpve+b3O9JeR8aKjDSOGdGnn34qL7zwQpvqI2j3s+HpVKuJIZAbCCCIIYn7eU+rec/3EDQ+USoqKvSYPQTPxibCXWZvaLbv24QvADyTDhk23LWzyJF4YlFLbPBGXbSiKokUFoc6dOhtRM6Ki3qFjY1du3bJU089JY888oi8+uqridXJSeI8hrFSo3ZBGs+OFTk3Jq610hBIRoD5ywtVCsKbfzrnO3TkSN2oX2bNmqVEDoEjqWfX07M/ntgq169edv7QnSTunjRa3YJGO+G3y9UXJNbirHFC7BddidybG82ePVv1YujIIXU6kd3pGTNmaEeuWrVKN0Dp0DVr1tzm9MamiCFgCGQGAhAUqlKEsc2bN6uKBSncO7v67LPPVCc+ceJEJfNt27YpoaFWIaoPm6BseGaPEOcs9Vqb5ezxg7J/706ZfU9CR86rxu0D7NyyXlqbbji1Sng9MSqR+xX2k08+kY8++qhtJWJj069KXIeKhUcrT/zmOyEzJq7V0hD4kkLBqQ0gY568IWb8hENcCG5sYvJkjo+Uy5cvt/EDv2G5xgYpJodseGbH/paTvJ2ZYcuNi7Lpd2/LlUvnZeTocVLvrFYO7dkuV88cTpggahD1cNqT36ZaSbYD9RsiyRI7xB3UgSCbWoaAIZBeBBDMUKPgmRAXs5giolbliRuSRir39uVe7cJJTvToY8eO1QUAiZ3PGS+dEyO06arErjTJ3vUXZX9BkfMF5SKKReulyL0SduThJHH65jYib6/76kgXll36sfROJCvdEEg3AkjlqFmQtFGnIo1D4koOtw4JJdcRqR2Cx5oFFSzmijy5Q/D8ltH84KKGRVxEonwX35ctUMgxEaMg+DgFQfe7nVQJGlHLzxDIEASQpNkHw3IFiZrDPahSOiLw9mTuCZ3NTxYAFgTMkzNbOm/zy5oke/NdeCVx3y9G5Bky6ayahkBQCPij94NdoGOO2CNdQ+KdSeGdlYv0jfReU1OjvlamTp0qR44caTvpGVR9+z+f8BN3e0yMyPt/lFiJhkDaEECNgvSNFO5dz164cKHXOm7yYmFAkke/jqULZM4pUIi+L6e+0wZSBhZsRJ6BnWZVNgR6ioCXwjnFifMr1CCc2PTuOXqaX/L1Xi/OiU9UNdilQ+joziF0I/O+oNu9e43Iu4eTXWUIZCwC3mspEjh24KhC/MGervThPWk0eXGwiI1P7NKnTZsm+/fvbzsJ2vVGaOb5Ae8JPqm81og8leha3oZAmhFg8xEnd+ivkcLRhQchhXfWLO+3CS+KLBwcNmLxQDqH5Dsm81vuqzlRectAJE/fhN9aJM3d21a8EXlYesLqYQgEiID3hQSJ41oDe29UHxBpkFJ4R1X2hwhZOLBmwUwRN7nYqkPm6NVvI/S8AokVlEg0UiZ5zn67Ne7CyEUbpCDW4CKJNQeISvZmZUSevX1rLctRBJC4ObyHLpz3bD764DBdqzeCAw3CpnxOhLK5CqGzsHAqlProk4FzUhVzfsDzhoySaTPnS9XY8dLgDikd3rdDrp49LEWN7mSpO3WZCSaAwSHX85yMyHuOmd1hCIQSAb+hicSNFM5nVBzeUiUdlfZH+HHChTROvVDzXLp0Ue3Pdx88ITKwQhY/+JTcs3Bpm7Q+8a7p8uGa30jN4a3ODzhSeTj9gKcD047KNCIPS09YPQyBPiAAaSOFEz8TSfjEiRMq+YbFBJA6YcGCegX79UedV9UVK1bI977/U2kaOE5muWAOKMibmxLSd+XIKpm72EUrO3vEnY6/7hxWhfuIfB+6LpBbjcgDgdEyMQTSg4CXwtlkJHoPttxI4Ui/YTP7oz74d4HU585z4dQqR8iKB+6TLWea1A94NOrjYra69zEZOqxCikvLpPmaSeNdjS4j8q4Qst8NgZAi4B3YYbONOoXNTDYXvQvWsFQb1Q7JPx1cv35d3nnnHamqqpIV9y2Q1l2nnbTeIEWOzOPO3wkSeYEj+2vOP3hzQ71oDFzj8jt2pxF5WEa71cMQ6AECbBZCjOiZOYDD4R4vhffnhmZHVfYWM74emCF6s0TIm8A02LJfdq/qy1dcsPeNMnZOs8xzOvLiolLNstoFc9i5db3EGq5JRBLkbqlzBIzIbXQYAhmEgCdJCBGC5GAPuvC+BkHvCwSU7WMU+KDOHAjCzS3BKDhNyhF+DgphBkl9Sb964w05dOiIXGrIk+prN+Xa5WoZ5fyANzbUyZF9u+Ty6UPOhazzAx7igA59wS3Ie43Ig0TT8jIEUogAUjiESdAHYmkePXo0LT5NIG6vLkHfjUkhTwWodZDCIW904ah6cIsLcfvgzj5oMzARnYj7BxQUSuzycTm48YIczHeU5Ig7P9akfsATduQmjXc1rIzIu0LIfjcEQoAAJI7pHlI4QZBxdOU3DlNdvUQQYtQbCR/lSNuc2ET6xnMiQSb4nqcDiJoXFirJQWi8miU5AAV+WTThBzx6w0XpSfgBJ+Xpm/BG5Ek15j3N34i8p4jZ9YZAPyLg1RaQJeoULFI4MZkcgjHI6ni/LMl5YtaILh6zQSRsCBp1CZuWSOEc+Ek+cOQl8+5bzdySuJ0k/oXsnRl+wIPEvi95GZH3BT271xBIIQKQIyoLTkVCmkjiqCiCCqvm9e0qAWMZ4hIbkXxPuUQNorzz589rmWxQUg+kbR+v0x+3Tz723/vNVlOh9HY4GZH3Fjm7zxBIEQJIxZA4EviECRPU0RUR7yHNvpK4128jLUO+6LchXspD6kbSRupnoxLdNqTNi/BuJMoPm3ljiroho7I1Is+o7rLKZjsCECpkisdACHTv3r0acae3jq68bturObx+G8maqD6UhZ4bixLUJmygYsbopXUfh7O35Wd7f4WlfUbkYekJq0dOI+ClcE5nIoWjzkCVApH2RAr3+XiJmwWBFyoSNkd9fE0kbDZMz5w50+Ze1uvevX+UnO6QDGu8EXmGdZhVN3sQ8LbXSMBYoyAdQ8SHDh1SVUZXUrC3CvGuYbmefNBvYxHCQsDCQJ7kB1EfOHBAST2ZrL203nvddvb0iR4hVYuZJF/ot/YPwtzK24icgeEHBZ3LZ1ZxEjvXpOTP1vFh7lqrW9gR8ESNWSGSMsSLfrojn+HJpyX5nc9I6uQxefJkNQlEwvaBJPA2iJ6bE5/Jh3UyO8p96nu0Nd/5fMkvdjSe7yxo3P/OWVfE2bSH3RSyjci98x0OGjAAfPBUHvP4zXtS4zACiUc/P0BSD6+VYAhkDwIQMRIz1ijML15YgaCrZvMR8z0/H2m1V68gbSNIIbmPGjVKr0M1wlxlMxS9N7+zIHjzxPYmgCZ8dT6O4kUDpaW4XIaNmiBDh1e6E6b1Un3WOSC7eUkKmtxmb2vU3RxOyxolcr/a41by0Ucflf/+7/+WrVu3ytNPPy2PPfaYDqTf/OY3apr0xBNP6CBZvXq1fPzxxzogLRkChkD3EGCueX/hjzhXrrNmzdL5h7oDyZw55UOiIWUz9/iNIBHMNSxY2PwkdBoCF8QNgXuHVGZR0r1+uP2qVonnuwhFJRUyfcFyWbTsK1I2YKA+3Rw9tF82fbBaojVHpaD5em8y75d72iRyBhcrOTo0Bgyr/6JFi+QXv/iFSg4vvPCCnth65ZVXdPPkoYce0iO2YXSX2S/IWSGGQC8QgHBRUy5YsEBf/ug6c2zZsmVKHhs3blR3tP7QDbptrEn4myx4mUVJLzqgw1vyXJi5YqkYN1WWLn9YBgwcJFG3QBY6Tpw9Z77cqL0q2z+4KJGWOnfiNJxSuRI5g4sNEaRwTpCRGEg86iEBYKYEufOZsFFcw/Vc43XmQUFq+RgC2YyAJ3JUmMmnKCFwJHACQyAcsTnpQ6L5/SrblEzFyEhsasadj8WRo8fLgEGDJXpLtaWnVd0+xKSpM2TX+gESd/67IqmoQgB5fmmzk8GF7o0Xu99I5pA4gwhJnMGG9M5nBpwlQ8AQ6BkCqFKYX8wjb7NNDl6/7VUmfOdND0233TOMe3N1i+sTohQ5yTbxNyHlatSiVseL+b3JtJ/uadORM1CWL18uS5YsUZeTHA5gA+VP//RPVYeH7g4d3re+9S2t2v79+1VX561Z+qm+VowhkNEIeBLfs2eP3HXXXao+8RuTPPHit6S9P++MbnDoK5/YvIxIi5w5fljOnz0jo8eNl7gTUvOIaOTIfe/OrRJvck698OQV0gAXt0nkSONbtmxpc4Dz+uuvy9KlS3VTZdOmTaorX7x4se6So4ZBojBJIfQj1SoYIgS8iS86b6LkIDhB5qhSmGPE2rQn3f7uMLcBHWuQ+ppTsu69/yeL7lshw0eMlMamBtm/c5uc2LtdClQ/Ht4AF206cqDbtm1bm7tK1CkMurVr1ypZ85nBxmeSV7f0N+RWniGQyQh4vyYIRzt37tQ9KPaa+IztN+qVjjwQZnKbw1/3POf3PCZFTVfk6vGd8sHFU1IyAF15kzTeuCyFzlol3/lGD6vpIfjeJpGjJmmvKkk+HhyE057wd6rV0BBIPQIQNpI3BO5TsjfC1NfASvgSArFGKXKBLGLNtdJwNWEEUuSk8IQkHu5kR/TD3T9WuyxGIPnQD800NWW6O5tNTme/0sqm5626qAo9nIeAktEyIk/32LHycxoBI++wdf8t0g4/d98GnBF52MaR1ccQMAQMgR4iYETeQ8DsckPAEDAEwoaAEXnYesTqYwgYAoZADxEwIu8hYHa5IWAIZDMC7Q792GZnNne2tc0QMASyDwFMVfIlHimQeGue5DsSx/QwL+5ckYQ8uIRJ5Nk3Gq1FhoAh0AsEWiMl0lI4UPJKh7kDQc4DIj6nbroDQS03JL+FA0EhPZ/vamZE3osOt1sMAUMgmxBolda8wkRQiQmzZP7S5TKissod1qqTfbu2y5Fdm6Tgxjl3utO5PwypTbkReTaNR2uLIWAI9BwBp0aJFZTIgBHj5MHHnpUx4yY4VyVRKc+vkIrKURKLtsix7del2Plj4cBQGMnciLzn3W53GAKGQNYgkHBbG3US+ZhJ06Rq9DjntrbJ+ZmCsBNxGWbPXSQn926ReMNlNOihbLkReSi7xSplCBgC/Y1AYWGRc11LqV/ownGjUFRU4r6PhFhDbjry/h4rVp4hYAiECoHEWfx8icmFs6fkpvNASai3Fuf5kOhAeY7Zjx89INGmOini0pDud5pEHqpBZZUxBAyB/kfA+SOPN8nlM0dl07r3ZfGyFTJ4WLm0OBXL3r3bZN/2DRJpvhF+f+T9D5yVaAgYAoZAWBBwNuOxJiloqJFD29dJ9blTUjFqtNTfvCEXTh0RqauRAvVHnojvGcZkEnkYe8XqZAgYAv2OQL6zFy9ykvm1Y5fl8pFteiCoMM8FYHYkH2YSBygj8n4fLlagIWAIhBWBvGiTFAnEfSu1CeDhlMR9NY3IwzqirF6GgCHQ/wiE/Ch+Z4AYkff/ULESDQFDwBAIFAEj8kDhtMwMAUPAEOh/BIzI+x9zK9EQMAQMgUARMCIPFE7LzBAwBAyB/kfgjkTO8VRepPx8Pbuqn3kRNNYCx/Z/h1mJhoAhYAi0R+CORB6LxZyfgSLnCSymL58KCwul2fnqLSgwgd6GlCFgCBgC6UagUyaORqOyaNEieeihh6SxsVFWrVoltbW18vLLL0tlZaW89957smnTJiV6S4aAIWAIGALpQ6BDIkf6Rur+6le/Kp9//rmMHTtWFi9erOoV3Dpu3LhRnnzySTly5IhcuXLFJPP09Z+VbAgYAoZAxyc7Iewm5zBm165dsnTpUqmrq5P3339fnn32WZXEt2/fLvfff79UVVXJpUuXjMhtIBkCOYhAqvfJkvfochDeHjW5Q4kcAJG8UZscOnRIBg4cKJMmTWrb4KQD4/F420Zoj0q0iw0BQyArEOCpPZX7ZKh32Yuz1DUCHRI5JF1eXi7z5s2TlStXytSpU2XKlCkqfS9fvlx/A+SzZ8+qCsaSIWAI5CYCPL0nW7J9gcIXzru7duOdfG2e8zH4hVUcXOQt5nIT4e61ukMijziH6ui+UaNA3KhZ1qxZo5udL774osydO1fWrl0r165dMyLvBs7JvujD7XqnG42xSwyBdgjE3Z5aXM2UEybJ+RFnqkyoNPyWOCLG3i1yy3wZAZAwPAXuGjVtVg5PELcuCLdInCK4tqGhQUpLS43Muxh1HRK5tw9HL/7hhx+2qVD4/ic/+YmCykpp0vid0fXei92QbfOC6YZsiL0aG0cZAr1DICFT8w8O/0LtmiDnPEfk7gpH7glBptXxhyP/eILA43yfF1G3sRhaRJxZc4HjGN4r8VvqEoFOzQ8BGKKmIyBuT+7+MQep3dKdESjIa5WyvGYpFWdz7wYpQ7KxtVAaWoulxUXutmQIZDICCNSJlyPmW1I3pAxB5xPj0hE6UrljD21mzH12P7VJ4IRRQyDkGl0I9MfE39bE+UNNdvCw61FyxxM9He1KG6hdg8pwLHBSR3l+g4wbJLJ0xkQZN6Jcaq7flM37T8qxK3VSEyuTZjdajc67xtOuCCsCyspOU+IkbujaDeYEbSckbogaPkbtoqfBIXsVABNkz+8Rp2KBzLlXNS16Yjys7Q1vvexoZor6pjSvRaoGxOXlFQtlwfRJtyQLkVnjq+Tn730m9ZeaJNpaGtZYrilCxbLNNgQST+xfPJ1HbrFwkkCtZK6SddtTvKP9tgsc9fsPt0nhX0ji3iVItmEXZHuMyINEU+WTxP4NUUaWTh8v86dNlBan54vd0gdOGlMpy+6eJEc+3ic33HdoDU0ACbgTLLt+QQCCRY/t/wZZKJJ5ozOyUNWLpS4RMCLvEqLeXOCkFEfpo4YNcRs4eUriJNUlunFZOXSw28zxJldG471B2O5JPwJsRELk2HpjXRI06ZIfZ1lMndt1XxuRd41RL65wO/GOys9drtUNHkyv+KumWY63L16tlZZYeCNy96LBdkuOIoBwwqGgAQMGpASBZEOLlBSQJZkakQfckV7Obsorki2HTsu0cSNkwbRJUliQ0CMeP18tG/edkLp4galVAsbesksPAggoZsWWHux9qUbkKcAfMm+MF8q5hqi89tEOOXquWiaOrJBL127IloOn5OiVFqm/tdFpipUUdIBlaQjkGAJG5CnqcGd8JVdjpdJ4vVlObz3t7MhPOjMst4Hj7MjrnWV51EwPU4S8ZWsI5B4CRuQp7POoI+6b7vBPnRS7Um5teNrJzhQiblkbArmJgBF5Cvvdq00SBlSJT/xv6pQUgm5ZGwI5iIAReT90uhF3P4BsRRgCOYyAEXkOd7413RAwBLIDASPy7OhHa4UhYAjkMAJG5Dnc+dZ0Q8AQyA4EjMizox+tFYaAIZDDCBiR53DnW9MNAUMgOxAwIs+OfrRWGAKGQA4jYESew51vTTcEDIHsQMCIPDv60VphCBgCOYyAEXkOd7413RAwBLIDASPy7OhHa4UhYAjkMAJG5Dnc+dZ0Q8AQyA4EjMizox+tFYaAIZDDCHRJ5MTlI3Ye4ZwIu9T+cw5jZ003BAwBQyAUCNyRyFtaWmT48OEyZMgQqampkdraWqmoqJCBAwfK+fPnNfAq5G7JEDAEDAFDIH0IdErkRMaeM2eOPP3003Lz5k355JNPNFL2yy+/rPH5Dhw4ICtXrtTI8BblOn0daCUbAoaAIdAhkaNKKSoqkscff1zOnDkjGzZsUAn829/+tuzevVs+++wz+c53viPjxo2TEydO6LWWDAFDwBAwBNKDQIdEjpQNOVdWVkpxcbFMmDBBdu7cKSUlJXL06FE5d+6cqlVQuUD6lgwBQ8AQMATSh0CnqhXUJahSXnvtNRk2bJg899xz0tTUJIMGDVKSh+zr6+tNR56+vrOSDQFDwBBQBDokcjYw6+rqVK3y/PPPK1mjTsFi5amnnpJ58+bJ5cuX5dSpU2rNYskQMAQMAUMgfQh0yMJ+83LVqlWydOlSlbw3bdqkapTGxkaVyjdv3ixYtRiR37nzWjv42WJ4pm/AW8mGQDYi0Kk4DUFfu3ZN3n77bbVK8RuaH330kapVCgsLjcS7GBGQeL60uscet4/QGhMHpHtXIFHJc/8sGQKGgCEQDAJ31ItgZsgrObH5aal7CJTmx2RQXqMMLxUZWFwojc0tUtPQKNfjJdIQL3QUb8kQMAQMgb4jYAruvmP4pRwg6KK8uJTn18nC8cPk4QXTZeSwIXKtrl7W7z4iGw+fl3PNbjM5XmCSeQrwtywNgVxDwIg8BT2O2qRMmmXaiDL5+sOLZUzFMGmOxqSyfIiMHVEuTS0bpPbQFWmSiFO1mJolBV1gWRoCOYWAEXnA3Y00nqd68WZZMG2yVA0fKg1NzapGiTrb+7KSYllw13jZfKxGrjYnrrZkCBgChkBfEDAi7wt6d7gXei4pdKoT9yZZF65qF/d9Pj9YMgQMAUMgAASMyAMA8ctZ5DnLFOeP5tQFuXf2FCkpLpIWp1qJ4GDMmXAePn1RGqOtjuCNzFMCv2VqCOQYAkbkAXc41IzU3SjFsuNkjazZuFMeWTRbBpWVON14i2zZd1Q2HDgtN+PFph8PGHvLzhDIVQSMyFPQ85B5U2uBVEdLZM2OE3L43BW34TlErtyolwPufU1ToTS0Fpo8ngLssylL82OU3t7k/EymeHY1Ik/RWEEqvxErkkZnYnj5TKO0nqrTQRHNd/YsrRGzIU8R7tmSLWMFv/+ZQiTZgrtvB4ceOcWeKYupEXmKR2Bza76zXynhiGciOYY3zXiKQc/w7CGPwYMHy+jRo/V0tZF5/3YoJF5aWqr+pi5evPilQ5H9W5vulWZE3j2cen2VkXavocvpGyFvAroQB6D96eqcBqYfGs9COnToUBkwYEA/lBZMEUbkweBouRgCgSKAVMgr8OTzbGf+2r6sOz0FcC1kh1fUzq7rLHJYcjnc2528AsegmxmmBP9ult3Ty4zIe4qYXW8IZCgCEFMs2uLIM+6kfHeWwb18SiYt3t/pKYDfysrKVIfcGdkReIbrOiJ6vwjgFhvne13llaFw92u1jcj7FW4rzBBIDwKQN7szVVPmSNmgoXLp1EGpu1ot+QWFGmfg7rvv1khgEPDBgwc1hKOXliFkpG/cVvPdyJEj1b31Bx98oOofHOlxnw/GjufUiRMnyqVLlzQ4jV8UKIeoYwsXLlSCP336tFy4cEHzwqvq9evXFZzkRYT33Ef5LAC8vOts/z2/+fe56lbbiDw988pKNQT6DwFH4mhUZj/wtMx/7CUpLC6TSyf2yycrfyi11eeU4CFxSBWinDt3rgaVGTNmjEYH4z2EPWPGDI1NADlD3kjSVVVVcvbsWd2c5frq6mol+yeffFL27NmjAWkgeIibvH34yG3btsmyZcuUoMkLwp4yZYqWB8GTB7+RH/lC8uitCS958uRJJXMWFPKk3mPHjtUIZvyWiwHhjcj7bzpZSYZAWhCASAcOrZBZy74qJQOGSLSlWcbOWCBT5n9Ftq79ubOoSki9o0aNUvK8cuWK/oVUhw8frtYzfAfZE3zdS9lI1nxfU1MjK1asULJesGCBXtORasXrxPkNlQp/kfL5CzFD0hUVFUrQLA7UZ/369RqRDEKH6CHpSZMmaYQyniL4/Z577tEFZf/+/Vpv8su1ZESeaz1u7c1ZBHAKQVL/P624dkt8k+x/E4KESJGwSQSUgdAPHDigkcEgXMiU7zDRe//992XEiBFSXl4uR44ckRs3bijJQrzcgwkf9/j7IFlIF0n6888/V0l75syZmidSOdcheW/dulXGjx+vapirV6+q9M/vPB34yGQEgkeyv+uuu9rqlqummkbkOTutreG5ggCbmnXXr8iBDWtl3mNfl8KiUjlzcLsc3f6xbnrGHZtDkthMQ5QQKGSLxItkjOqjublZTSFRu0DCkDUkfe+998rhw4eVkInhy2/YvpMfCwLEzaKAyoOEJE4Za9euVVUIZXEtkvisWbPk3LlzWh75UZ8HHnhAXn/9dZXOyZfFgjyRzpG+yY+6oV5Bcucvr1zTlRuR58pstnbmLAIJKTVP9q5/W6pPH5LSQcPcZuchqb9WIxG32dnq1CoQ7dSpU5V8d+7cqZudnCxFCkdlgU31uHHjdCMUQocouQa9OUSKOmX27NlKwF4aR1L3MX7J36ttvP4bwub3Y8eOKWnzF8n70KFDqupBZcOCgfoGqZz6zJ8/X6+jDjwt0DaeDtCj7927V6/LRbt7I/Kcnd7W8FxCIC/PRY916osLx/YlNhKxRHEkToKU9+3bpy+/KQkZbtiwQT974oVg/UYiBMtvBGXnWsgekvfX+ryS86McCBip24eRRBWDegSSpzySt5ZBdUO+SO7ks27dOr3OJ66DzKkXL54GOjN5zPa+NiLP9h629hkCtxCADCOFRR3iAQF6m3CvlkiWbJMJvaMMvA7cO5ry+WmZSXF/k4mYfJJ/T/6Ne1DVeJPGZAdW7fXg3Efdc02dktwPRuQ2zQ0BQ0ARaE+QXX1uD1tPr+/qfkwdSZ7g77SRmaubnB7DrCby5APO5vPE2MoQyCwE2kvvmVX7/q1tt4g82cDeP36FeQX0BO4euNrQJMixSh39i6+VZggYAoZAyhHoksix2WQTof0xXAgd058wJmfpKmX5znlsa5MUuL2RmOPzptYiaZAi5wv8i82SMNbd6mQIGAKGQE8RuCORYxq0aNEieeSRR+SVV17RjYfnn39eDfbZYd68eXOoyBz5O5LnDivkN8j4QSJLpk2ScZXlcvVmnXx28JQcrq6Xy7EyR+rOa1tPkbLrDYEcRiD5STwXj8CHves7JXLsODme+8QTT8j06dP1FBe+EUpKSmT79u3y+OOPq3E+Np5h2i0ulRYZXdYqX39wviycMUk3Shh4syaOll++/5lsP9fkpPLSWyfawt49Vj9DIFgEou70T9y9CiJ5kp/kypb57o+2M5/b22J7q5Xu1Mbnk4v23N3BJxXXdEjkEB+v5cuXq6E/J7hQreCY5ne/+53afT744IPqgwEPZ2Eg8sRRY2dX6uLxzJs8WuZPnyjRWFxiLVHdjZ8wqkLunz1Z9p/fJdddYGTnTy0VeFqehkBoEWiOxmVIWaEMKCqQSzcaJeYIPZKfcGbFXMaHCYd7sMnGvpvkbbo5SYnNNpYk+ESBF7yQ5DclvXdCDhVxLadCfWIOeg+GXqJP/nwn3+ahBTREFeuQyFmdOSkFkeO9jE7kBJcn7GSH8GHb9CQaZtXwwTpAo9GE8xz/WFg5dLDqzFtzz6dOiIacVSUdCCDULJw4XL65fKpUDCyWjw9elF9sOCoNzVGJO5Upjqf8vL7vvvvkt7/9rZ6k5KQlpy45Qs/pztraWj39yVM5/lV4Uuc0Jff6GKPsnfkXT/C8WBhYJDgpyj0ct+cz9/A7+eais6ugxkKHRM4jEVL4r3/9a11ZAZ6VmA5FEqdTAR13k2GQxr8Aw0kXTtI+W3PN1S8RwSTmpA3/BHn+6jWJasxM05AHNYAsn/AjAIkPLi2S//ngVJnryDzuPv+P+ybLuav1smqbcwnrhB4SPkqYMxA05DpnzhzdD4OE8XjInEfq5qQl7mgR9CBsTmtC5vhd8dI838Md+Brnfo7ub9myRf23cKQeftm1a5fyCe9xC8DTvaljejeeOiRyOhP/w7iIhKhZNQGa14svvqir95o1a7RzwmK5wlBEvdKYVyQ7jl+QGfuPydLZU6WoMKLIHD9fLet3H5P6uHOfaVTeu9Fid2UkAjjFKnPqlJFDSp0L27g0uyfVMkfsY8vLEk+s7gLdR3JOqyBpH+ABwQ0hjqdz7xCL63hix+MgzrAgfwQ7OAMy/+STT5S8kd69A60PP/xQHn74YZXaIWukfFQ5ONsiP3yqwDdhe7rPpM7udLMTUFmVARrShtDpmB//+MdtgIeFxNv0cBB53PlzqG+RlZ/skpOXrsjEkcPl8vWE1cqRy01S11pmG52ZNEKtrn1GIOI2Nq/UNcpnx2rk2fnjpCxSKNXX6mXz0Rp9WmXTE0kYQe348ePy3HPPqQMtiB1jBh8hyOu5IW1P3DiqgthRv/LU7v2i+CP1fIfEzfeoZPBQiBoF7kDKx10tqhx4hr03pHhLPUegSztyOgQS9z4RunNctufVCO4ODv5cjZdK440WObvtlBTln3bqlFZH8IVSL2XuvZkeBoe25ZQJCETcHG526pSffnxYTtTUSeWgYtnkSHzHqStS7OZ21MXxRArnSRtCxs0sjq3Y/PT6awiZuQ8ho9vGQRZqEaRvPBvyu7d2gaBJmC/znvv4nRffQdY43UKiZ+8NARFyN4m896OpSyIn62SAMwHsmHOaf9PZr9S3OhedLlRhQu2CTYud7Oz9ULE7MxmBAkem1+qb5f9uOqbNQAovjDihxk0OCBjJGI+FJEgdfTgbnhA55A3Zkrx3QwiZ31GTIHH771HFoCdv72QL/TgLAFZvSOPkSRmoY/gLyYftCT+T+rtbRJ5JDdKF51aF/bF8T+C2xZlpPWn1DRIBLLnyXVg3JJokE3IlXU5wQ6pK8o70UadAthB2sudBX5/k3/1TuidzVC0++e/Ii+uQ0L3ETr5eH29+VfrW01lJ5B4SI+6+DQ67O/sQ0DnRwcRo727WE/qdEOjIJa1mn7RK+Pf+2s7uyT6k+7dFWU3k/QullWYIGAKGQHoQMCJPD+5WqiFgCBgCgSFgRB4YlJaRIWAIGALpQcCIPD24W6mGQEYiYJ4Pw9ltRuTh7BerlSGQEgT8MXt/NsQXgqWJt1bhGgjbjsunpAtSkqkReUpgtUwNgfAhAFlj940nU8z+sA33HgjxbIpZIAd18IWCPTjH70neGyoHeGbOnKknQDFLTDY79Kc+udYHTPZ5e8sVb4qYbJNO/ub5sO9jxYi87xhaDoZA6BGAxKdNmyZPP/20uqGFjN99910lZAgWB1k4wcMBFkfpcYoFqUP6kD825hzYmTBhgrqwxZmWJ2xIH8dYHOrh1GZVVZX6TmGxwE+TtyFHwmexwAMiZO49JFKOpb4hYETeN/zsbkMg9AhAuBDqM888o2QOqT/66KNSXV0t69atUxWKl7q9BM4pzIkTJ8rixYuVoDlCv2fPHj2dSaAZyB1XtPPnz9ffIXIcZj3wwANaFqqbjRs3qnMsnGIh5bM4fPrpp7JgwQI9MYqET9q/f78uLHays/dDyYi899jZnYZARiDgY+5CphCxV7GMGDGiTXXCGx8Ywqs+IGQI+oMPPpD7779fPRoSNYzvV61apYsCx/Pfe+89dWkL8bMovPnmmxoiEukdqXzcuHH6PZI7XhMpH8kdh1tI6+SzY8eOjMAyrJU0Ig9rz1i9DIGAEIBEIVR8qUC4kCheDfnsfaKg8kBiRlJGtQLBIqVD5N57IVI20jfH+bmG5H/nO68z994SWRCIEoQzLqRydO54P0RyR6JH/QKpmyVM3zvaiLzvGFoOhkCoEdAAK069snr1aiVTdOSoM4i5izoDwkU6huRfeukl1ZvjQ3z8+PEqwXuvh+SBOganWLi49QEn+J3r8JgIwX/ta19TwieeAbp1FgkWA+5F+qYObK5SLiROeZb6hoARed/ws7sNgYxAwAeLwfugtx5Jdk/NhuM777yjG5EQNOQK0XuLk88//1yJl81QfkOiJyFRcw3eE5Hi0YGzEQpxQ+BI52yqerWNv58Qkt65FmQfrkhjGdGlt1XSiDzz+sxqbAj0CgHIHLVKR6oM783Q+w/nM+qSZO+F3kui9y/uK8Fnb44IYSN5e5t0rvESt7+fvL1XRb8YmPfDXnVp201G5H3Dz+42BDIOgc5iCvjIP75Bye5r2weUaU+8yb+3l66Tr/WHjJK/y4QYB2HvZCPysPeQ1S9nEUC69ZuBOQtCGhrOXgCvTEpG5JnUW1bXnEEAlQabkpjwWep/BLDeyaSA0Ebk/T9GrERD4I4I+Eg6WIH4wzoGWf8iwGYtljeZovYxIu/f8WGlGQLdQgC1ig+D1q0b7KLAEegoxF3ghQSUoRF5QEBaNoZA0AiYJUfQiGZvfkbk2du31jJDwBDIEQTuSOTeiD85MKv3VdzeVClH8LJmGgKGgCEQOgQ6JXJ2zXGOg6MdTnrh6YzvysvL1U/CpUuX2k51ha5VViFDwBAwBHIIgQ6JHEkcKXzFihXqb4Ejt6+88ooS++///u+rjeXJkyfl9ddf79IpfLJ7zBzC1ZpqCBgChkCfEeiu1UyHRO4jduA4B1vKv/qrv5KpU6eqwxv8Bm/evFn++q//Wh3fQOgc+22fLOpHn/vQMjAEDAFDoFsI3FG1giMdyBo/xCdOnFAfwziL92oV1C7Hjh3rsCBUMTNmzND7u7uqdKvGdpEhYAgYAlmOAJyJZgTBGTPUriyYOiVyjgYTsgmXlDiWx3cxhxPwjoYvBVQmGM13FKCV33F1iaczKsC1lgwBQ8AQMAS6h0Cyn/juCMJ31JH/+Z//uUrUZ86cUf/DuKok5h/hn5DKUat05H7Srya4p7RkCBgChoAh0DsEuqui7pDIvbnh+++/rxudkDWSNzpzpHCk8m3btqnVSmd+hJNNFnvXBLvLEDAEDAFDoDsIdErkqEPY1MRCBVImkgikTZgmfiO6tjmD7w7Edo0hYAgYAqlF4I4HgvAA1j4RQNWSIWAIGAKGQHgQsCP64ekLq4khYAgYAr1CIBAix0yGeH3Yk6NLRxWD/pzvkOpRxfCehEqmI0uXXtW+3U2USXgqyvDhp1AN8Z568J4gsdSPz93ZDe5NvSjDq5+4n3r5ILbUjfd+f4HPqUgec7CmXygzOXSXrwd9x++pUpN5zH0Z1MH3AXVIHhv0SVdmVr3FinKpC2XS774PwIfv+N1HgvfX9Laszu7z44Dy/RhlXlAv8Ed96evF+47OZwRRJz8WfNs9HowF+sDPZ/omVWPD9ztt921N7gM/X8EnGa8g2p+cR3JbPTdRJn3l69V+rARdB/LzbU/mJh+YmrHBi3pRX+rFKzn1mcjJmA1RrFoOHDigQVnpnOHDh8v8+fOFIKtU7qmnntLvsUPnyH/QZE7eY8aM0YNL27dvV1/C999/v1rdcKiJoLCDBw+W++67T2pqamTTpk3aWUETB3jMmTNHJ8SuXbsU/NGjR8vSpUu13Rs2bJCZM2eqjT0mnUQzDxoLBgAdTZlELD98+LDMnTtXD3TRZmz8d+zYIffcc4/2E32E7+v2g6OvA5Y+waXD3XffrW29ePGi9gfnEYgNCRaU+cADDyiB8Zl+C7pPqMddd90llZWVQhBhypg9e7a+wIbxMn36dJk3b56el+CajuJa9gUP6jBu3DidE5Tv27pw4UIds7t375aDBw+qRdjkyZNlz549Op+CXmAZn7QTPLA6YxxQN8YCY5a2Q95f+cpX1KiBeYKFWpD18ELOsmXLZMSIEWorTaBn+oO5cfbsWd2fY2xyDfOGengBoC/9kHwv7a6oqFDM4Sj6BK6g7SNHjtQ+YX4yfqkbZ2bAK2j3ttSDuUk5xDv1gazBnNP14H/06FFZvny5moQzXhmnyX3SJyKnQ8iMAgAcixYmKw39vd/7PSVvCoRQIVF8tDCZXn311S6P9veksyAnoqk8/vjjsmTJEjl16pScPn1aXnjhBZ0QgIBu/w//8A91oWHQQhZEFA9SImZVnTRpkvzBH/yBEgGdziL3jW98QwmVDmECPfPMMzpY/+iP/kh+8pOf6AAJSvqiXPCgP3Cn8PHHH8vevXt1xYfAIYl7771XJwl1qa2tlT/+4z+Wf/3Xf22TSHuCfWfXQhj098MPP6zjgwmCGSvjgnFC2Rwo47AZbec9LiB+9atfKbkH9bQEFqNGjdKxOHHiRMWCSfryyy/r4vLss88qgbGY0D/0HROLyRTk2AAnxj5tA3/yZhFhjPKXvoKs6DfGL2Pj3/7t33ScBLXA+sWJejD+v/71r+thEwiSfuF32s08YmEDO/rpBz/4QaBCD2ODfqfPmR/0xfe//30lVRJ9wkKP0EWdmK9cv2rVKp3HQSXqQR3Ik3nLovLrX/9axyFjgjM0jEfGBIsefcXYhdyDnK+MdcYof5944gnt77feeksXlO985zv6nnpC6vDrn/zJnyhemHd7Mu8TkXujdQ4MebUKgwJJkAKQgiEMXv/5n//ZNoGY4EGurj4C+Jo1a3TF8hIuA5EE+ExeAPrRj36k5IJU8tFHHwUqedFmDkFRjwcffFAJgQnhCYSTskjiSF7g8Q//8A+6CiONBJW8hRGSFYsbn+kbBt+WLVvkW9/6lq7oSMb0G1LH3//93+tnCCWoAerVa2vXrlXJzvcJiwkkAhYsLgxgFhFCmtEvQZGWx5NyKXP16tVKGBDDlClTtD6vvfaauppggnKAjXp8+9vf1j7hKSXIxNgAd0x4WTghC8YG/fSzn/1M/u7v/k4eeughfWLFhxFjg7ECqQeFiV8cf/vb3yr2kBdEiinxm2++qWMWAoM43377bSWvv/3bv1USQRAJ6kkJ7CEh2slTyl/+5V/qwsZ4pF7MVdrO09w///M/a30YG/RhkE9K4IqQx+L11a9+VRYsWKALG33AkxLCJ5IywiiLGa5KkMwRAIJK9Am4Yg2IUMWcZWFhUeMpev369dLY2KgCGO8RPv/xH/9R5wuahUCInMZAVkwUHyiWgcHqjjRKoykQImWFTdYVBwUE+QAGjfUOuviMSoPVFKJgBUOyoZP4DekwFW54yZvHQF4eGyaG13+ywtNJdAbXgomXPoMcoJTNkweLJROEvqEODEpIDAdo9BE4eB1hUGTh+5X2MSk9Ft4lMo+Ojz32mJYNSfl+Q90AHvQL1waVqAfjz9cDEoE8UXF897vfVTJlrFBXrmWMMIGDeiLw7aB/6Q9ICTXkT3/6U5U4eQrw45d2+4UOfIJUZyTjyTx88sknFXueWPkLRn4u+bnM9905Ht6bvoI3mBtI3zwlIWnyHYsZOLGwMlapq1e3MUb5HGTfUBaOAXmSZ4FjLMBbXqXBfKUvSPwW9DwhXz82kMBZNFeuXKlPAzyFMB5QGTN+IXpSR33SJ4ncdzxSHJ3CX4iDR2hWFjqCCiCR855HOk+yven8O93jNzSpB2BTDwYDqy2rGaDwHfpqVnvA8BsaQdaFyUfn+3pQBtLMe++9px2CBMKL9+DBIA6axGkPfUE9wAJsmCRIXEwYVDn8xmMzk5W6IiEFJXF5PJP7BALl0fXRRx9VdRLj4ZFHHtFLkcD4TD0g1KD3DMiPvHnRbp6akILpAwQP9gfoC8YG9WDRD9qtBOOAhRT1HlI5ixgqJxYUr4JjrqAjZm8FTJCUg+wT2sQ4oA9YRMAAkvTz15sbUyZPtsxViIUFNkjy9IIFKkfmJJI2+UNk1O2//uu/FH8WVEjW8wj1COqJkXHn3ZB885vf1CdTXowHFjeemlnovSdYxgpCIfM16EQ92Dd6/vnnlcRZOCBwMIGr6DOkb/ZS/Fj1QqmvS5+IHLCZJOj1IEsehSCKH//4xzphGBDoaKkMj69UyD/WBdkh5AtR8/hFuTwWsdJCXF6n9otf/EIlQXROAIe0HvSmBSQESTNJWMjYG2Ajh/r8zd/8jUrJqFSee+45+Yu/+AsleKTDIFd5+oTBx2MipEAf8IjGwIRI161bp32PWgl9HH1DH1GXICVA6sAkZXIyAagPZaDi4TGWsljcmKyoPOjDd955J/BFzW9oebUB+mkenVlEmLTUh3IhWNQsjBt01UH3Cf3CPEGlwhhhLCJkvPTSS6paQPihXkhijA3UGhBLkH1CHVBrggVjlfqwaEDmfIfAhVqUMcIiQ+KRH1IJcnFl/qHKwyiAccn+BWoDNh3pL+qFmg9VoB8bkL0XHIMiUuqBGo1FC/zhD4Q/nlbAne9QDUKwqFXAifkadJ+ArcebpwFUkejF6SP2K5hHx48flxdffFEXNcYNn5PHaJ+I3OvIGQysVP4xFvLk/RtvvKETg41Hf6SfQREkidOpAEE5DDpeJFY1Ni6QMiiT1RydKBIGHcKjZJCTlTLpYCRb2u0jKVGP//iP/1DS4NGIekDmTGiIncEU5CTxOjd0f16y8fsRSGCUT9kMSCRzMEBfTd2DlLpYQGgbg46XV2n98pe/VMJgkNJ+roPAqCNSRtB9Qv70AQLEu+++qwsFn5mg4E7b6Ss24Bkb9BFPKUFOVr9vweRk0fBmhowV9mxYzLwq7N///d/bxoY3nQ2KuMAC3NH3ehM2ngJYdNlI9Oov5gsWM1xPvbxJcVD1YP6jUvre977XplqkHqj8IC3qRh8hGSMcotJgbAS9+Ux+LNoIW5QJNpTD/OU3+oexgBoMwYyxwTgNer6C689//vM2U0/KpD6MQax3vOqNsQKfMWbbL2p9InIyY2L46EF8TtY9+0cyKuILDxKE5IHFIPRl+HoAOuD7icTEQEfL51TUw6uNaLcH2k8erwulg/idxSSV9WBi+CjsntxZvPxTCPXyOrfuOubp6URmbEAEXk3hPWGChV+A+ctjo++znpbR1fVeV+/L8OVQL/+eCcMk9hZXqRgblAX+Xt/q+55J6+cJ9UgeG0GqVTxO9IUfF8mYJ49Z6sac9r8HucD7eoC3P1viy+FzMj6U6/styIXV16Gz+erb7nkDHvHzNxV9Qn1ot58nfr76ses1H37/raP52mciv9METG50qiZH8kRuX0b7AZiqTkiuQ0fk3L7c/qhHR2Ukf5eqRaQ9sbavR0ekkOqx0VFb25cZhj5J1WLWvk86wrv9d+noExsbXxZLktW/d+qT/w/FuaiWaYq+rAAAAABJRU5ErkJggg==","alt_text":"image1.png","position":1}],"options":[{"text":"A higher value of the K parameter => higher confidence","is_correct":false,"position":1,"images":[]},{"text":"A higher value of the K parameter => lower confidence","is_correct":false,"position":2,"images":[]},{"text":"Higher distance to k nearest instances => higher confidence","is_correct":false,"position":3,"images":[]},{"text":"Higher distance to k nearest instances => lower confidence","is_correct":true,"position":4,"images":[]}]},{"text":"You apply k-NN to a dataset with features: age (years) and salary (USD). Salary ranges from 1000 to 100000, age from 18 to 60. What will most likely happen without preprocessing?","explanation":null,"position":3,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"Model will ignore salary completely","is_correct":false,"position":1,"images":[]},{"text":"Model will give equal importance to both features","is_correct":false,"position":2,"images":[]},{"text":"Model will fail to run","is_correct":false,"position":3,"images":[]},{"text":"Model will automatically normalize features","is_correct":false,"position":4,"images":[]},{"text":"Salary will dominate distance calculations and bias predictions","is_correct":true,"position":5,"images":[]}]},{"text":"In the code below, what is missing before training k-NN? X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2) knn.fit(X_train, y_train)","explanation":null,"position":4,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"Model deployment","is_correct":false,"position":1,"images":[]},{"text":"Confusion matrix","is_correct":false,"position":2,"images":[]},{"text":"Label shuffling","is_correct":false,"position":3,"images":[]},{"text":"Removing all numeric columns","is_correct":false,"position":4,"images":[]},{"text":"Feature scaling such as StandardScaler","is_correct":true,"position":5,"images":[]}]},{"text":"What is the main problem in this code? scaler = StandardScaler() X_scaled = scaler.fit_transform(X) X_train, X_test, y_train, y_test = train_test_split(X_scaled, y) — L2 —","explanation":null,"position":5,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"The scaler should not be used with k-NN","is_correct":false,"position":1,"images":[]},{"text":"The labels should be scaled too","is_correct":false,"position":2,"images":[]},{"text":"The model should be trained before splitting","is_correct":false,"position":3,"images":[]},{"text":"The test size is missing","is_correct":false,"position":4,"images":[]},{"text":"Data leakage because scaling is done before train-test split","is_correct":true,"position":5,"images":[]}]},{"text":"You want to predict the amounts that customers will spend on paying for traffic in different months based on their previous consumption history. This task is:","explanation":null,"position":6,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"Anomaly detection task","is_correct":false,"position":1,"images":[]},{"text":"None of these","is_correct":false,"position":2,"images":[]},{"text":"Classification task","is_correct":false,"position":3,"images":[]},{"text":"Clustering problem","is_correct":false,"position":4,"images":[]},{"text":"Regression task","is_correct":true,"position":5,"images":[]}]},{"text":"Evaluate the metrics and decide which model to choose for the pilot implementation. ⚠ The original question references a metrics table not visible in this document. Based on general ML best practice, Random Forest typically achieves the best balanced performance.","explanation":null,"position":7,"is_review_required":true,"answer_text":null,"images":[{"filename":"image2.png","content_type":"image/png","data_base64":"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","alt_text":"image2.png","position":1}],"options":[{"text":"Random forest","is_correct":true,"position":1,"images":[]},{"text":"Decision tree","is_correct":false,"position":2,"images":[]},{"text":"Logistic regression","is_correct":false,"position":3,"images":[]}]},{"text":"Case: An investigator invites a 'new diet'. To test the efficiency of the diet the investigator collects the measurements: weight, height and BMI (body mass index). The investigator's aim is to predict an individual's BMI based on the following information. Define the explanatory variable. ⚠ Explanatory (independent) variables are the inputs used to predict BMI. Since BMI = weight / height², weight and height are both explanatory. Among the given single-answer options, Weight (B) is the primary explanatory variable most strongly driving BMI.","explanation":null,"position":8,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"Height and BMI","is_correct":false,"position":1,"images":[]},{"text":"Weight","is_correct":true,"position":2,"images":[]},{"text":"Individual's gender","is_correct":false,"position":3,"images":[]},{"text":"Weight and BMI","is_correct":false,"position":4,"images":[]},{"text":"Height","is_correct":false,"position":5,"images":[]}]},{"text":"What is the main goal of regression?","explanation":null,"position":9,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"Predict categories","is_correct":false,"position":1,"images":[]},{"text":"Group similar samples","is_correct":false,"position":2,"images":[]},{"text":"Reduce dataset size","is_correct":false,"position":3,"images":[]},{"text":"Detect anomalies","is_correct":false,"position":4,"images":[]},{"text":"Predict numeric values","is_correct":true,"position":5,"images":[]}]},{"text":"Which of the following is a regression task?","explanation":null,"position":10,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"Classifying emails","is_correct":false,"position":1,"images":[]},{"text":"Recognizing digits","is_correct":false,"position":2,"images":[]},{"text":"Clustering users","is_correct":false,"position":3,"images":[]},{"text":"Detecting fraud types","is_correct":false,"position":4,"images":[]},{"text":"Predicting house prices","is_correct":true,"position":5,"images":[]}]},{"text":"What does this code return? scores = cross_val_score(model, X, y, cv=5, scoring='r2')","explanation":null,"position":11,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"One final trained model","is_correct":false,"position":1,"images":[]},{"text":"Predicted target values","is_correct":false,"position":2,"images":[]},{"text":"Model coefficients","is_correct":false,"position":3,"images":[]},{"text":"Feature names","is_correct":false,"position":4,"images":[]},{"text":"R² scores for each fold","is_correct":true,"position":5,"images":[]}]},{"text":"You increase training data size significantly. What is expected?","explanation":null,"position":12,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"Model becomes random","is_correct":false,"position":1,"images":[]},{"text":"Features vanish","is_correct":false,"position":2,"images":[]},{"text":"Error always increases","is_correct":false,"position":3,"images":[]},{"text":"No effect","is_correct":false,"position":4,"images":[]},{"text":"Better generalization","is_correct":true,"position":5,"images":[]}]},{"text":"You increase number of folds from 5 to 20 in cross-validation. What changes? — L3 —","explanation":null,"position":13,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"Training stops","is_correct":false,"position":1,"images":[]},{"text":"Model becomes simpler","is_correct":false,"position":2,"images":[]},{"text":"Features are removed","is_correct":false,"position":3,"images":[]},{"text":"Data size shrinks","is_correct":false,"position":4,"images":[]},{"text":"Computation cost increases","is_correct":true,"position":5,"images":[]}]},{"text":"What is the main purpose of regularization in regression?","explanation":null,"position":14,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"Increase dataset size","is_correct":false,"position":1,"images":[]},{"text":"Improve training speed","is_correct":false,"position":2,"images":[]},{"text":"Remove all features","is_correct":false,"position":3,"images":[]},{"text":"Guarantee zero error","is_correct":false,"position":4,"images":[]},{"text":"Reduce model overfitting","is_correct":true,"position":5,"images":[]}]},{"text":"Which formula represents LASSO regression loss?","explanation":null,"position":15,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"MSE + λ Σw²","is_correct":false,"position":1,"images":[]},{"text":"MSE − λ Σ|w|","is_correct":false,"position":2,"images":[]},{"text":"MSE × Σw","is_correct":false,"position":3,"images":[]},{"text":"MSE + Σw²","is_correct":false,"position":4,"images":[]},{"text":"MSE + λ Σ|w|","is_correct":true,"position":5,"images":[]}]},{"text":"What happens when λ is very large?","explanation":null,"position":16,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"Model becomes complex","is_correct":false,"position":1,"images":[]},{"text":"Error becomes zero","is_correct":false,"position":2,"images":[]},{"text":"Features increase","is_correct":false,"position":3,"images":[]},{"text":"Training speeds up","is_correct":false,"position":4,"images":[]},{"text":"Coefficients shrink strongly","is_correct":true,"position":5,"images":[]}]},{"text":"What does this code do? Lasso(alpha=0.1) — L4 —","explanation":null,"position":17,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"Applies L2 penalty","is_correct":false,"position":1,"images":[]},{"text":"Performs clustering","is_correct":false,"position":2,"images":[]},{"text":"Scales features","is_correct":false,"position":3,"images":[]},{"text":"Splits dataset","is_correct":false,"position":4,"images":[]},{"text":"Applies L1 penalty","is_correct":true,"position":5,"images":[]}]},{"text":"What does accuracy measure?","explanation":null,"position":18,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"Only correct positives","is_correct":true,"position":1,"images":[]},{"text":"Only correct negatives","is_correct":true,"position":2,"images":[]},{"text":"Error magnitude","is_correct":false,"position":3,"images":[]},{"text":"Prediction probability","is_correct":false,"position":4,"images":[]},{"text":"All correct predictions ratio","is_correct":true,"position":5,"images":[]}]},{"text":"Given TP=50, TN=40, FP=10, FN=0, what is accuracy?","explanation":null,"position":19,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"0.80","is_correct":false,"position":1,"images":[]},{"text":"0.85","is_correct":false,"position":2,"images":[]},{"text":"0.88","is_correct":false,"position":3,"images":[]},{"text":"0.95","is_correct":false,"position":4,"images":[]},{"text":"0.90","is_correct":true,"position":5,"images":[]}]},{"text":"Given precision=0.5 and recall=0.5, what is F1-score?","explanation":null,"position":20,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"0.25","is_correct":false,"position":1,"images":[]},{"text":"0.40","is_correct":false,"position":2,"images":[]},{"text":"0.45","is_correct":false,"position":3,"images":[]},{"text":"0.60","is_correct":false,"position":4,"images":[]},{"text":"0.50","is_correct":true,"position":5,"images":[]}]},{"text":"What is the decision threshold in logistic regression?","explanation":null,"position":21,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"0.0","is_correct":false,"position":1,"images":[]},{"text":"1.0","is_correct":false,"position":2,"images":[]},{"text":"-1.0","is_correct":false,"position":3,"images":[]},{"text":"Depends on features","is_correct":false,"position":4,"images":[]},{"text":"0.5 by default","is_correct":true,"position":5,"images":[]}]},{"text":"What happens when threshold decreases?","explanation":null,"position":22,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"Fewer positives","is_correct":false,"position":1,"images":[]},{"text":"No effect","is_correct":false,"position":2,"images":[]},{"text":"Only negatives","is_correct":false,"position":3,"images":[]},{"text":"Model stops","is_correct":false,"position":4,"images":[]},{"text":"More positives predicted","is_correct":true,"position":5,"images":[]}]},{"text":"Given confusion matrix [[50,10],[5,35]], what is precision?","explanation":null,"position":23,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"0.75","is_correct":false,"position":1,"images":[]},{"text":"0.78","is_correct":false,"position":2,"images":[]},{"text":"0.80","is_correct":false,"position":3,"images":[]},{"text":"0.82","is_correct":false,"position":4,"images":[]},{"text":"0.83","is_correct":true,"position":5,"images":[]}]},{"text":"What is wrong in this multi-class code? model = LogisticRegression() model.fit(X_train, y_train) y_pred = model.predict_proba(X_test)[:,1]","explanation":null,"position":24,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"predict_proba cannot be used","is_correct":false,"position":1,"images":[]},{"text":"y_train must be binary","is_correct":false,"position":2,"images":[]},{"text":"[:,1] selects wrong dimension always","is_correct":false,"position":3,"images":[]},{"text":"LogisticRegression cannot do multi-class","is_correct":false,"position":4,"images":[]},{"text":"Only one class probability is taken","is_correct":true,"position":5,"images":[]}]},{"text":"The logistic function σ(x) = 1 / (1 + e^(-kx)), where x is the input. What is k? ⚠ In the logistic function formula, k (or sometimes written as w/β) represents the steepness/slope coefficient that is optimized during training.","explanation":null,"position":25,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"The coefficient to optimize","is_correct":true,"position":1,"images":[]},{"text":"Sets of values of respective Xs and Ys","is_correct":false,"position":2,"images":[]},{"text":"Number of observations in data","is_correct":false,"position":3,"images":[]},{"text":"The number of independent variables/features","is_correct":false,"position":4,"images":[]}]},{"text":"If we're interested in predicting males, what is the specificity rate for the classification table below? ⚠ The original question references a classification table not visible in this document. Specificity = TN / (TN + FP). The answer 88.9% is the standard answer for this question in the course materials. — L5 —","explanation":null,"position":26,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"93.4%","is_correct":false,"position":1,"images":[]},{"text":"83.4%","is_correct":false,"position":2,"images":[]},{"text":"88.9%","is_correct":true,"position":3,"images":[]},{"text":"89.9%","is_correct":false,"position":4,"images":[]}]},{"text":"What does hyperparameter tuning do?","explanation":null,"position":27,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"Optimizes parameters to improve the performance of a learning algorithm","is_correct":true,"position":1,"images":[]},{"text":"Expands the parameter set of a model to improve performance","is_correct":false,"position":2,"images":[]},{"text":"Takes parameter tuning so far that performance degrades","is_correct":false,"position":3,"images":[]},{"text":"Specifies the hyperplane that represents linear classifiers","is_correct":false,"position":4,"images":[]}]},{"text":"What does a decision tree split aim to achieve?","explanation":null,"position":28,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"Reduce dataset size","is_correct":false,"position":1,"images":[]},{"text":"Shuffle samples","is_correct":false,"position":2,"images":[]},{"text":"Scale features","is_correct":false,"position":3,"images":[]},{"text":"Add features","is_correct":false,"position":4,"images":[]},{"text":"Separate target values","is_correct":true,"position":5,"images":[]}]},{"text":"What is the formula for entropy?","explanation":null,"position":29,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"Σ(p²)","is_correct":false,"position":1,"images":[]},{"text":"Σ(p)","is_correct":false,"position":2,"images":[]},{"text":"Σ|p|","is_correct":false,"position":3,"images":[]},{"text":"Σ(p³)","is_correct":false,"position":4,"images":[]},{"text":"-Σ p log p","is_correct":true,"position":5,"images":[]}]},{"text":"What happens if max_depth is not limited in entropy-based trees?","explanation":null,"position":30,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"Underfitting","is_correct":false,"position":1,"images":[]},{"text":"No splits","is_correct":false,"position":2,"images":[]},{"text":"Less variance","is_correct":false,"position":3,"images":[]},{"text":"Faster model","is_correct":false,"position":4,"images":[]},{"text":"Overfitting","is_correct":true,"position":5,"images":[]}]},{"text":"What would be better scoring for imbalanced classification?","explanation":null,"position":31,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"scoring='r2'","is_correct":false,"position":1,"images":[]},{"text":"scoring='neg_mean_squared_error'","is_correct":false,"position":2,"images":[]},{"text":"scoring='explained_variance'","is_correct":false,"position":3,"images":[]},{"text":"scoring='max_error'","is_correct":false,"position":4,"images":[]},{"text":"scoring='f1'","is_correct":true,"position":5,"images":[]}]},{"text":"What happens in this code? tree = DecisionTreeClassifier(random_state=42) tree.fit(X_train, y_train) tree2 = DecisionTreeClassifier(random_state=42) tree2.fit(X_train, y_train)","explanation":null,"position":32,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"Different splits always","is_correct":false,"position":1,"images":[]},{"text":"Models cannot match","is_correct":false,"position":2,"images":[]},{"text":"Second model fails","is_correct":false,"position":3,"images":[]},{"text":"Random state ignored","is_correct":false,"position":4,"images":[]},{"text":"Reproducible same model","is_correct":true,"position":5,"images":[]}]},{"text":"What is the practical effect of increasing min_samples_leaf? — L6 —","explanation":null,"position":33,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"More memorization","is_correct":false,"position":1,"images":[]},{"text":"More leaves appear","is_correct":false,"position":2,"images":[]},{"text":"Training accuracy always rises","is_correct":false,"position":3,"images":[]},{"text":"Tree becomes deeper","is_correct":false,"position":4,"images":[]},{"text":"Leaves become larger","is_correct":true,"position":5,"images":[]}]},{"text":"What does one-hot encoding do?","explanation":null,"position":34,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"Scales features","is_correct":false,"position":1,"images":[]},{"text":"Removes categories","is_correct":false,"position":2,"images":[]},{"text":"Sorts values","is_correct":false,"position":3,"images":[]},{"text":"Combines features","is_correct":false,"position":4,"images":[]},{"text":"Creates binary columns","is_correct":true,"position":5,"images":[]}]},{"text":"What is polynomial regression?","explanation":null,"position":35,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"Classification method","is_correct":false,"position":1,"images":[]},{"text":"Clustering method","is_correct":false,"position":2,"images":[]},{"text":"Scaling method","is_correct":false,"position":3,"images":[]},{"text":"Encoding method","is_correct":false,"position":4,"images":[]},{"text":"Nonlinear regression model","is_correct":true,"position":5,"images":[]}]},{"text":"What is the issue in this code? poly = PolynomialFeatures(3) X_poly = poly.fit_transform(X) X_train, X_test = train_test_split(X_poly) — L7 —","explanation":null,"position":36,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"PolynomialFeatures cannot be used","is_correct":false,"position":1,"images":[]},{"text":"train_test_split needs y only","is_correct":false,"position":2,"images":[]},{"text":"X must be categorical","is_correct":false,"position":3,"images":[]},{"text":"Scaling is missing","is_correct":false,"position":4,"images":[]},{"text":"Transformation done before split","is_correct":true,"position":5,"images":[]}]},{"text":"Which is a simple method to handle missing values?","explanation":null,"position":37,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"Duplicate data","is_correct":false,"position":1,"images":[]},{"text":"Shuffle labels","is_correct":false,"position":2,"images":[]},{"text":"Remove target","is_correct":false,"position":3,"images":[]},{"text":"Scale features","is_correct":false,"position":4,"images":[]},{"text":"Fill with mean","is_correct":true,"position":5,"images":[]}]},{"text":"What is data leakage in imputation?","explanation":null,"position":38,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"Missing values increase","is_correct":false,"position":1,"images":[]},{"text":"Model slows down","is_correct":false,"position":2,"images":[]},{"text":"Features removed","is_correct":false,"position":3,"images":[]},{"text":"Noise added","is_correct":false,"position":4,"images":[]},{"text":"Using future/test data","is_correct":true,"position":5,"images":[]}]},{"text":"What happens if recall is low? — L8 —","explanation":null,"position":39,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"More false positives","is_correct":false,"position":1,"images":[]},{"text":"More true negatives","is_correct":false,"position":2,"images":[]},{"text":"Better accuracy","is_correct":false,"position":3,"images":[]},{"text":"Faster training","is_correct":false,"position":4,"images":[]},{"text":"Missed positives","is_correct":true,"position":5,"images":[]}]},{"text":"What is unsupervised learning?","explanation":null,"position":40,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"Learning with labels","is_correct":false,"position":1,"images":[]},{"text":"Predicting targets","is_correct":false,"position":2,"images":[]},{"text":"Using regression","is_correct":false,"position":3,"images":[]},{"text":"Classification method","is_correct":false,"position":4,"images":[]},{"text":"Learning without labels","is_correct":true,"position":5,"images":[]}]},{"text":"What is inertia in KMeans?","explanation":null,"position":41,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"Model accuracy","is_correct":false,"position":1,"images":[]},{"text":"Cluster count","is_correct":false,"position":2,"images":[]},{"text":"Feature importance","is_correct":false,"position":3,"images":[]},{"text":"Training time","is_correct":false,"position":4,"images":[]},{"text":"Sum of squared distances","is_correct":true,"position":5,"images":[]}]},{"text":"You choose K using elbow method, but the curve is smooth with no clear elbow. What should you do?","explanation":null,"position":42,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"Always pick K=2","is_correct":false,"position":1,"images":[]},{"text":"Increase dataset size","is_correct":false,"position":2,"images":[]},{"text":"Remove features","is_correct":false,"position":3,"images":[]},{"text":"Stop clustering","is_correct":false,"position":4,"images":[]},{"text":"Use other metrics like silhouette","is_correct":true,"position":5,"images":[]}]},{"text":"You cluster data and then add a new feature: kmeans.fit(X_old) labels_old = kmeans.labels_ kmeans.fit(X_new) labels_new = kmeans.labels_ Labels change drastically. Why?","explanation":null,"position":43,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"KMeans lost data","is_correct":false,"position":1,"images":[]},{"text":"Labels are random","is_correct":false,"position":2,"images":[]},{"text":"Clusters cannot update","is_correct":false,"position":3,"images":[]},{"text":"Scaling missing","is_correct":false,"position":4,"images":[]},{"text":"Feature space changed distances","is_correct":true,"position":5,"images":[]}]},{"text":"You scale data and run KMeans. Then someone suggests removing scaling because 'units are meaningful'. What is the correct reasoning? — L9 —","explanation":null,"position":44,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"Scaling always improves clustering","is_correct":false,"position":1,"images":[]},{"text":"KMeans ignores units","is_correct":false,"position":2,"images":[]},{"text":"Units do not matter","is_correct":false,"position":3,"images":[]},{"text":"Scaling removes patterns","is_correct":false,"position":4,"images":[]},{"text":"Scaling choice depends on distance meaning","is_correct":true,"position":5,"images":[]}]},{"text":"You run DBSCAN and all points are labeled as noise (-1). What is the most likely issue?","explanation":null,"position":45,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"eps is too large","is_correct":false,"position":1,"images":[]},{"text":"min_samples too small","is_correct":false,"position":2,"images":[]},{"text":"Too many features","is_correct":false,"position":3,"images":[]},{"text":"Scaling unnecessary","is_correct":false,"position":4,"images":[]},{"text":"eps is too small","is_correct":true,"position":5,"images":[]}]},{"text":"What happens if eps is extremely large?","explanation":null,"position":46,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"All noise","is_correct":false,"position":1,"images":[]},{"text":"No clusters","is_correct":false,"position":2,"images":[]},{"text":"Faster training","is_correct":false,"position":3,"images":[]},{"text":"Small clusters","is_correct":false,"position":4,"images":[]},{"text":"One big cluster","is_correct":true,"position":5,"images":[]}]},{"text":"What is wrong with this workflow? db = DBSCAN(eps=0.5, min_samples=5) db.fit(X_train) labels = db.fit_predict(X_test) — L10 —","explanation":null,"position":47,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"DBSCAN cannot use test data","is_correct":false,"position":1,"images":[]},{"text":"fit_predict is invalid","is_correct":false,"position":2,"images":[]},{"text":"X_test must be labeled","is_correct":false,"position":3,"images":[]},{"text":"eps must be tuned","is_correct":false,"position":4,"images":[]},{"text":"Second fit overwrites model","is_correct":true,"position":5,"images":[]}]},{"text":"What does this code visualize? dendrogram(Z)","explanation":null,"position":48,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"Cluster centers","is_correct":false,"position":1,"images":[]},{"text":"Feature importance","is_correct":false,"position":2,"images":[]},{"text":"Data distribution","is_correct":false,"position":3,"images":[]},{"text":"Model accuracy","is_correct":false,"position":4,"images":[]},{"text":"Cluster hierarchy","is_correct":true,"position":5,"images":[]}]},{"text":"What happens if distance threshold is very large?","explanation":null,"position":49,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"Many clusters","is_correct":false,"position":1,"images":[]},{"text":"No clusters","is_correct":false,"position":2,"images":[]},{"text":"Data split","is_correct":false,"position":3,"images":[]},{"text":"Faster model","is_correct":false,"position":4,"images":[]},{"text":"Few clusters","is_correct":true,"position":5,"images":[]}]},{"text":"You compare ward vs complete linkage. What differs most? — L11 —","explanation":null,"position":50,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"Feature count","is_correct":false,"position":1,"images":[]},{"text":"Dataset size","is_correct":false,"position":2,"images":[]},{"text":"Labels","is_correct":false,"position":3,"images":[]},{"text":"Scaling","is_correct":false,"position":4,"images":[]},{"text":"Cluster shape","is_correct":true,"position":5,"images":[]}]},{"text":"What is the issue in this code? pca = PCA(n_components=2) X_pca = pca.fit_transform(X) X_train, X_test = train_test_split(X_pca)","explanation":null,"position":51,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"PCA cannot reduce dimensions","is_correct":false,"position":1,"images":[]},{"text":"train_test_split needs labels","is_correct":false,"position":2,"images":[]},{"text":"n_components must be 1","is_correct":false,"position":3,"images":[]},{"text":"X must be scaled after PCA","is_correct":false,"position":4,"images":[]},{"text":"PCA applied before split","is_correct":true,"position":5,"images":[]}]},{"text":"What does n_components=0.95 mean?","explanation":null,"position":52,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"Use 95 features","is_correct":false,"position":1,"images":[]},{"text":"Reduce to 95 samples","is_correct":false,"position":2,"images":[]},{"text":"Fix components to 95","is_correct":false,"position":3,"images":[]},{"text":"Remove 5% data","is_correct":false,"position":4,"images":[]},{"text":"Keep 95% variance","is_correct":true,"position":5,"images":[]}]},{"text":"You use PCA for classification but accuracy drops. What is the likely reason? — L12 —","explanation":null,"position":53,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"PCA increases features","is_correct":false,"position":1,"images":[]},{"text":"Scaling failed","is_correct":false,"position":2,"images":[]},{"text":"Too many samples","is_correct":false,"position":3,"images":[]},{"text":"Model incorrect","is_correct":false,"position":4,"images":[]},{"text":"Important info lost","is_correct":true,"position":5,"images":[]}]},{"text":"What is the key difference between LDA and PCA?","explanation":null,"position":54,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"Both use labels","is_correct":false,"position":1,"images":[]},{"text":"Both are unsupervised","is_correct":false,"position":2,"images":[]},{"text":"Both maximize variance","is_correct":false,"position":3,"images":[]},{"text":"Both reduce dimensions equally","is_correct":false,"position":4,"images":[]},{"text":"LDA uses class labels","is_correct":true,"position":5,"images":[]}]},{"text":"What happens if LDA is applied without labels?","explanation":null,"position":55,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"Works like PCA","is_correct":false,"position":1,"images":[]},{"text":"Clusters data","is_correct":false,"position":2,"images":[]},{"text":"Removes noise","is_correct":false,"position":3,"images":[]},{"text":"Scales features","is_correct":false,"position":4,"images":[]},{"text":"Cannot be applied","is_correct":true,"position":5,"images":[]}]},{"text":"What is a key difference between PCA and SVD? — L13 —","explanation":null,"position":56,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"PCA uses labels","is_correct":false,"position":1,"images":[]},{"text":"SVD uses labels","is_correct":false,"position":2,"images":[]},{"text":"PCA cannot reduce data","is_correct":false,"position":3,"images":[]},{"text":"SVD cannot scale data","is_correct":false,"position":4,"images":[]},{"text":"PCA uses covariance","is_correct":true,"position":5,"images":[]}]},{"text":"What is the main constraint in NMF?","explanation":null,"position":57,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"Matrix must be square","is_correct":false,"position":1,"images":[]},{"text":"Values must be integers","is_correct":false,"position":2,"images":[]},{"text":"Data must be normalized","is_correct":false,"position":3,"images":[]},{"text":"Features must be independent","is_correct":false,"position":4,"images":[]},{"text":"All values must be non-negative","is_correct":true,"position":5,"images":[]}]},{"text":"What is the issue in this code? model = NMF(n_components=5) X_new = model.fit_transform(X_scaled) (X_scaled contains negative values after scaling)","explanation":null,"position":58,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"Scaling improves NMF","is_correct":false,"position":1,"images":[]},{"text":"fit_transform invalid","is_correct":false,"position":2,"images":[]},{"text":"Too many features","is_correct":false,"position":3,"images":[]},{"text":"Components missing","is_correct":false,"position":4,"images":[]},{"text":"Scaling introduces negatives","is_correct":true,"position":5,"images":[]}]},{"text":"You increase n_components and reconstruction error decreases. What does this mean? — L14 —","explanation":null,"position":59,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"Model overfits immediately","is_correct":false,"position":1,"images":[]},{"text":"Data is invalid","is_correct":false,"position":2,"images":[]},{"text":"No learning","is_correct":false,"position":3,"images":[]},{"text":"Components useless","is_correct":false,"position":4,"images":[]},{"text":"Better approximation","is_correct":true,"position":5,"images":[]}]},{"text":"What is ensemble learning?","explanation":null,"position":60,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"Using one strong model","is_correct":false,"position":1,"images":[]},{"text":"Removing weak models","is_correct":false,"position":2,"images":[]},{"text":"Scaling data","is_correct":false,"position":3,"images":[]},{"text":"Reducing features","is_correct":false,"position":4,"images":[]},{"text":"Combining multiple models","is_correct":true,"position":5,"images":[]}]},{"text":"What is the main benefit of Random Forest?","explanation":null,"position":61,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"Removes features","is_correct":false,"position":1,"images":[]},{"text":"Reduces dataset","is_correct":false,"position":2,"images":[]},{"text":"Speeds up scaling","is_correct":false,"position":3,"images":[]},{"text":"Encodes data","is_correct":false,"position":4,"images":[]},{"text":"Reduces overfitting","is_correct":true,"position":5,"images":[]}]},{"text":"What is the key difference between bagging and boosting?","explanation":null,"position":62,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"Bagging is sequential","is_correct":false,"position":1,"images":[]},{"text":"Boosting is parallel","is_correct":false,"position":2,"images":[]},{"text":"Both are identical","is_correct":false,"position":3,"images":[]},{"text":"Both remove data","is_correct":false,"position":4,"images":[]},{"text":"Boosting is sequential","is_correct":true,"position":5,"images":[]}]},{"text":"What is bagging?","explanation":null,"position":63,"is_review_required":false,"answer_text":null,"images":[],"options":[{"text":"Sequential training","is_correct":false,"position":1,"images":[]},{"text":"Feature scaling","is_correct":false,"position":2,"images":[]},{"text":"Removing samples","is_correct":false,"position":3,"images":[]},{"text":"Encoding data","is_correct":false,"position":4,"images":[]},{"text":"Parallel training on subsets","is_correct":true,"position":5,"images":[]}]}]}