ensemble.ipynb

{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's try combining the KNN and neural net! [See scripts imported for generation of those models]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/150\n",
      "1179/1179 - 2s - loss: 87976192.0000 - val_loss: 58147300.0000 - 2s/epoch - 2ms/step\n",
      "Epoch 2/150\n",
      "1179/1179 - 2s - loss: 54713648.0000 - val_loss: 51757676.0000 - 2s/epoch - 1ms/step\n",
      "Epoch 3/150\n",
      "1179/1179 - 2s - loss: 52264576.0000 - val_loss: 50963208.0000 - 2s/epoch - 1ms/step\n",
      "Epoch 4/150\n",
      "1179/1179 - 2s - loss: 50730536.0000 - val_loss: 48525196.0000 - 2s/epoch - 2ms/step\n",
      "Epoch 5/150\n",
      "1179/1179 - 2s - loss: 48738744.0000 - val_loss: 50095156.0000 - 2s/epoch - 2ms/step\n",
      "Epoch 6/150\n",
      "1179/1179 - 3s - loss: 47810512.0000 - val_loss: 44599672.0000 - 3s/epoch - 3ms/step\n",
      "Epoch 7/150\n",
      "1179/1179 - 5s - loss: 45707868.0000 - val_loss: 51032840.0000 - 5s/epoch - 4ms/step\n",
      "Epoch 8/150\n",
      "1179/1179 - 4s - loss: 44650928.0000 - val_loss: 47185228.0000 - 4s/epoch - 4ms/step\n",
      "Epoch 9/150\n",
      "1179/1179 - 5s - loss: 43929456.0000 - val_loss: 44123616.0000 - 5s/epoch - 4ms/step\n",
      "Epoch 10/150\n",
      "1179/1179 - 5s - loss: 43361984.0000 - val_loss: 43359824.0000 - 5s/epoch - 4ms/step\n",
      "Epoch 11/150\n",
      "1179/1179 - 4s - loss: 43769112.0000 - val_loss: 42968304.0000 - 4s/epoch - 3ms/step\n",
      "Epoch 12/150\n",
      "1179/1179 - 4s - loss: 42828320.0000 - val_loss: 45778660.0000 - 4s/epoch - 4ms/step\n",
      "Epoch 13/150\n",
      "1179/1179 - 6s - loss: 43094204.0000 - val_loss: 42441604.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 14/150\n",
      "1179/1179 - 4s - loss: 41892860.0000 - val_loss: 45561368.0000 - 4s/epoch - 3ms/step\n",
      "Epoch 15/150\n",
      "1179/1179 - 5s - loss: 41641040.0000 - val_loss: 44182788.0000 - 5s/epoch - 4ms/step\n",
      "Epoch 16/150\n",
      "1179/1179 - 6s - loss: 41346416.0000 - val_loss: 43204452.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 17/150\n",
      "1179/1179 - 3s - loss: 41260108.0000 - val_loss: 43969824.0000 - 3s/epoch - 3ms/step\n",
      "Epoch 18/150\n",
      "1179/1179 - 6s - loss: 40727548.0000 - val_loss: 47934552.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 19/150\n",
      "1179/1179 - 5s - loss: 40718764.0000 - val_loss: 41275304.0000 - 5s/epoch - 4ms/step\n",
      "Epoch 20/150\n",
      "1179/1179 - 6s - loss: 39935584.0000 - val_loss: 42259872.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 21/150\n",
      "1179/1179 - 6s - loss: 40513044.0000 - val_loss: 41399868.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 22/150\n",
      "1179/1179 - 5s - loss: 40181572.0000 - val_loss: 45194116.0000 - 5s/epoch - 5ms/step\n",
      "Epoch 23/150\n",
      "1179/1179 - 4s - loss: 40193212.0000 - val_loss: 39587568.0000 - 4s/epoch - 4ms/step\n",
      "Epoch 24/150\n",
      "1179/1179 - 6s - loss: 39810396.0000 - val_loss: 42345356.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 25/150\n",
      "1179/1179 - 6s - loss: 39219968.0000 - val_loss: 42336800.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 26/150\n",
      "1179/1179 - 6s - loss: 39310328.0000 - val_loss: 40897768.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 27/150\n",
      "1179/1179 - 6s - loss: 39338396.0000 - val_loss: 40231336.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 28/150\n",
      "1179/1179 - 5s - loss: 39043360.0000 - val_loss: 40786276.0000 - 5s/epoch - 4ms/step\n",
      "Epoch 29/150\n",
      "1179/1179 - 6s - loss: 38894824.0000 - val_loss: 40127652.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 30/150\n",
      "1179/1179 - 5s - loss: 38592064.0000 - val_loss: 42225940.0000 - 5s/epoch - 4ms/step\n",
      "Epoch 31/150\n",
      "1179/1179 - 7s - loss: 39355392.0000 - val_loss: 41702736.0000 - 7s/epoch - 6ms/step\n",
      "Epoch 32/150\n",
      "1179/1179 - 6s - loss: 38714432.0000 - val_loss: 45553984.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 33/150\n",
      "1179/1179 - 6s - loss: 38547820.0000 - val_loss: 57024852.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 34/150\n",
      "1179/1179 - 6s - loss: 39020336.0000 - val_loss: 42042312.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 35/150\n",
      "1179/1179 - 4s - loss: 37998788.0000 - val_loss: 40838884.0000 - 4s/epoch - 3ms/step\n",
      "Epoch 36/150\n",
      "1179/1179 - 5s - loss: 38382004.0000 - val_loss: 49911792.0000 - 5s/epoch - 4ms/step\n",
      "Epoch 37/150\n",
      "1179/1179 - 6s - loss: 38769188.0000 - val_loss: 40880204.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 38/150\n",
      "1179/1179 - 5s - loss: 38390480.0000 - val_loss: 39951632.0000 - 5s/epoch - 5ms/step\n",
      "Epoch 39/150\n",
      "1179/1179 - 5s - loss: 38485160.0000 - val_loss: 40065948.0000 - 5s/epoch - 4ms/step\n",
      "Epoch 40/150\n",
      "1179/1179 - 4s - loss: 38264700.0000 - val_loss: 39712196.0000 - 4s/epoch - 4ms/step\n",
      "Epoch 41/150\n",
      "1179/1179 - 4s - loss: 38364116.0000 - val_loss: 39921628.0000 - 4s/epoch - 4ms/step\n",
      "Epoch 42/150\n",
      "1179/1179 - 4s - loss: 37812992.0000 - val_loss: 48089664.0000 - 4s/epoch - 4ms/step\n",
      "Epoch 43/150\n",
      "1179/1179 - 4s - loss: 38052856.0000 - val_loss: 41140208.0000 - 4s/epoch - 3ms/step\n",
      "Epoch 44/150\n",
      "1179/1179 - 5s - loss: 38297016.0000 - val_loss: 41079208.0000 - 5s/epoch - 5ms/step\n",
      "Epoch 45/150\n",
      "1179/1179 - 6s - loss: 37867552.0000 - val_loss: 42830952.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 46/150\n",
      "1179/1179 - 5s - loss: 38027992.0000 - val_loss: 43562620.0000 - 5s/epoch - 4ms/step\n",
      "Epoch 47/150\n",
      "1179/1179 - 6s - loss: 38424304.0000 - val_loss: 42139112.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 48/150\n",
      "1179/1179 - 5s - loss: 37782708.0000 - val_loss: 39030216.0000 - 5s/epoch - 4ms/step\n",
      "Epoch 49/150\n",
      "1179/1179 - 5s - loss: 37682772.0000 - val_loss: 43923588.0000 - 5s/epoch - 4ms/step\n",
      "Epoch 50/150\n",
      "1179/1179 - 6s - loss: 37956972.0000 - val_loss: 39616136.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 51/150\n",
      "1179/1179 - 5s - loss: 37568848.0000 - val_loss: 42118684.0000 - 5s/epoch - 4ms/step\n",
      "Epoch 52/150\n",
      "1179/1179 - 6s - loss: 37630216.0000 - val_loss: 40034432.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 53/150\n",
      "1179/1179 - 4s - loss: 37807676.0000 - val_loss: 41041956.0000 - 4s/epoch - 3ms/step\n",
      "Epoch 54/150\n",
      "1179/1179 - 4s - loss: 37922652.0000 - val_loss: 44180528.0000 - 4s/epoch - 4ms/step\n",
      "Epoch 55/150\n",
      "1179/1179 - 4s - loss: 37731460.0000 - val_loss: 42264496.0000 - 4s/epoch - 4ms/step\n",
      "Epoch 56/150\n",
      "1179/1179 - 4s - loss: 37554176.0000 - val_loss: 40126036.0000 - 4s/epoch - 4ms/step\n",
      "Epoch 57/150\n",
      "1179/1179 - 6s - loss: 37508120.0000 - val_loss: 40796028.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 58/150\n",
      "1179/1179 - 5s - loss: 37696100.0000 - val_loss: 39815840.0000 - 5s/epoch - 4ms/step\n",
      "Epoch 59/150\n",
      "1179/1179 - 6s - loss: 37084772.0000 - val_loss: 39268232.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 60/150\n",
      "1179/1179 - 4s - loss: 37723300.0000 - val_loss: 40328152.0000 - 4s/epoch - 4ms/step\n",
      "Epoch 61/150\n",
      "1179/1179 - 5s - loss: 37223336.0000 - val_loss: 39612980.0000 - 5s/epoch - 4ms/step\n",
      "Epoch 62/150\n",
      "1179/1179 - 4s - loss: 37500260.0000 - val_loss: 41590488.0000 - 4s/epoch - 3ms/step\n",
      "Epoch 63/150\n",
      "1179/1179 - 4s - loss: 37333932.0000 - val_loss: 42454304.0000 - 4s/epoch - 3ms/step\n",
      "Epoch 64/150\n",
      "1179/1179 - 5s - loss: 37554280.0000 - val_loss: 39886128.0000 - 5s/epoch - 4ms/step\n",
      "Epoch 65/150\n",
      "1179/1179 - 5s - loss: 37319164.0000 - val_loss: 40137060.0000 - 5s/epoch - 4ms/step\n",
      "Epoch 66/150\n",
      "1179/1179 - 5s - loss: 37328216.0000 - val_loss: 39579528.0000 - 5s/epoch - 5ms/step\n",
      "Epoch 67/150\n",
      "1179/1179 - 5s - loss: 37188464.0000 - val_loss: 45527984.0000 - 5s/epoch - 5ms/step\n",
      "Epoch 68/150\n",
      "1179/1179 - 6s - loss: 37397212.0000 - val_loss: 40841320.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 69/150\n",
      "1179/1179 - 5s - loss: 37111208.0000 - val_loss: 42619012.0000 - 5s/epoch - 5ms/step\n",
      "Epoch 70/150\n",
      "1179/1179 - 6s - loss: 37588264.0000 - val_loss: 43495812.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 71/150\n",
      "1179/1179 - 6s - loss: 37272824.0000 - val_loss: 39022356.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 72/150\n",
      "1179/1179 - 6s - loss: 36944384.0000 - val_loss: 40061868.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 73/150\n",
      "1179/1179 - 6s - loss: 37084264.0000 - val_loss: 40707812.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 74/150\n",
      "1179/1179 - 5s - loss: 37243272.0000 - val_loss: 39267184.0000 - 5s/epoch - 4ms/step\n",
      "Epoch 75/150\n",
      "1179/1179 - 3s - loss: 37150272.0000 - val_loss: 43487292.0000 - 3s/epoch - 3ms/step\n",
      "Epoch 76/150\n",
      "1179/1179 - 5s - loss: 37173804.0000 - val_loss: 39039160.0000 - 5s/epoch - 4ms/step\n",
      "Epoch 77/150\n",
      "1179/1179 - 6s - loss: 37107280.0000 - val_loss: 40666180.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 78/150\n",
      "1179/1179 - 5s - loss: 37234916.0000 - val_loss: 41254948.0000 - 5s/epoch - 5ms/step\n",
      "Epoch 79/150\n",
      "1179/1179 - 6s - loss: 37009020.0000 - val_loss: 40650488.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 80/150\n",
      "1179/1179 - 5s - loss: 37002840.0000 - val_loss: 42089608.0000 - 5s/epoch - 4ms/step\n",
      "Epoch 81/150\n",
      "1179/1179 - 6s - loss: 37117312.0000 - val_loss: 42013724.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 82/150\n",
      "1179/1179 - 5s - loss: 36860268.0000 - val_loss: 45011772.0000 - 5s/epoch - 5ms/step\n",
      "Epoch 83/150\n",
      "1179/1179 - 5s - loss: 37054828.0000 - val_loss: 43114668.0000 - 5s/epoch - 4ms/step\n",
      "Epoch 84/150\n",
      "1179/1179 - 4s - loss: 36855304.0000 - val_loss: 40243904.0000 - 4s/epoch - 3ms/step\n",
      "Epoch 85/150\n",
      "1179/1179 - 4s - loss: 36831696.0000 - val_loss: 44233152.0000 - 4s/epoch - 3ms/step\n",
      "Epoch 86/150\n",
      "1179/1179 - 3s - loss: 36924448.0000 - val_loss: 42196860.0000 - 3s/epoch - 3ms/step\n",
      "Epoch 87/150\n",
      "1179/1179 - 4s - loss: 36976928.0000 - val_loss: 41932348.0000 - 4s/epoch - 3ms/step\n",
      "Epoch 88/150\n",
      "1179/1179 - 3s - loss: 36887904.0000 - val_loss: 40614720.0000 - 3s/epoch - 3ms/step\n",
      "Epoch 89/150\n",
      "1179/1179 - 3s - loss: 36973700.0000 - val_loss: 40557324.0000 - 3s/epoch - 3ms/step\n",
      "Epoch 90/150\n",
      "1179/1179 - 3s - loss: 36926356.0000 - val_loss: 40006188.0000 - 3s/epoch - 3ms/step\n",
      "Epoch 91/150\n",
      "1179/1179 - 3s - loss: 36992668.0000 - val_loss: 44421628.0000 - 3s/epoch - 3ms/step\n",
      "Epoch 92/150\n",
      "1179/1179 - 3s - loss: 36835020.0000 - val_loss: 40464612.0000 - 3s/epoch - 3ms/step\n",
      "Epoch 93/150\n",
      "1179/1179 - 3s - loss: 36998884.0000 - val_loss: 39506508.0000 - 3s/epoch - 3ms/step\n",
      "Epoch 94/150\n",
      "1179/1179 - 3s - loss: 36734208.0000 - val_loss: 44879076.0000 - 3s/epoch - 3ms/step\n",
      "Epoch 95/150\n",
      "1179/1179 - 4s - loss: 36716972.0000 - val_loss: 40997012.0000 - 4s/epoch - 4ms/step\n",
      "Epoch 96/150\n",
      "1179/1179 - 3s - loss: 36731800.0000 - val_loss: 40070004.0000 - 3s/epoch - 3ms/step\n",
      "Epoch 97/150\n",
      "1179/1179 - 4s - loss: 37005444.0000 - val_loss: 41370636.0000 - 4s/epoch - 3ms/step\n",
      "Epoch 98/150\n",
      "1179/1179 - 3s - loss: 36735040.0000 - val_loss: 45851604.0000 - 3s/epoch - 3ms/step\n",
      "Epoch 99/150\n",
      "1179/1179 - 4s - loss: 36998508.0000 - val_loss: 41590560.0000 - 4s/epoch - 3ms/step\n",
      "Epoch 100/150\n",
      "1179/1179 - 6s - loss: 36888076.0000 - val_loss: 42617376.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 101/150\n",
      "1179/1179 - 6s - loss: 36604520.0000 - val_loss: 41765684.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 102/150\n",
      "1179/1179 - 5s - loss: 36849472.0000 - val_loss: 42999600.0000 - 5s/epoch - 4ms/step\n",
      "Epoch 103/150\n",
      "1179/1179 - 6s - loss: 36779200.0000 - val_loss: 41369012.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 104/150\n",
      "1179/1179 - 4s - loss: 36380364.0000 - val_loss: 41267348.0000 - 4s/epoch - 3ms/step\n",
      "Epoch 105/150\n",
      "1179/1179 - 4s - loss: 36656144.0000 - val_loss: 43519320.0000 - 4s/epoch - 3ms/step\n",
      "Epoch 106/150\n",
      "1179/1179 - 4s - loss: 36629748.0000 - val_loss: 39499672.0000 - 4s/epoch - 3ms/step\n",
      "Epoch 107/150\n",
      "1179/1179 - 5s - loss: 36567868.0000 - val_loss: 44356408.0000 - 5s/epoch - 4ms/step\n",
      "Epoch 108/150\n",
      "1179/1179 - 6s - loss: 36802476.0000 - val_loss: 42546952.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 109/150\n",
      "1179/1179 - 6s - loss: 36518816.0000 - val_loss: 39977492.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 110/150\n",
      "1179/1179 - 6s - loss: 36547064.0000 - val_loss: 41839164.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 111/150\n",
      "1179/1179 - 6s - loss: 36609988.0000 - val_loss: 41723220.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 112/150\n",
      "1179/1179 - 6s - loss: 36623956.0000 - val_loss: 39947556.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 113/150\n",
      "1179/1179 - 6s - loss: 36439544.0000 - val_loss: 39287176.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 114/150\n",
      "1179/1179 - 5s - loss: 36174908.0000 - val_loss: 39907296.0000 - 5s/epoch - 4ms/step\n",
      "Epoch 115/150\n",
      "1179/1179 - 5s - loss: 36571284.0000 - val_loss: 41632088.0000 - 5s/epoch - 4ms/step\n",
      "Epoch 116/150\n",
      "1179/1179 - 4s - loss: 36469716.0000 - val_loss: 43381416.0000 - 4s/epoch - 4ms/step\n",
      "Epoch 117/150\n",
      "1179/1179 - 5s - loss: 36496044.0000 - val_loss: 42556056.0000 - 5s/epoch - 4ms/step\n",
      "Epoch 118/150\n",
      "1179/1179 - 4s - loss: 36577896.0000 - val_loss: 43620692.0000 - 4s/epoch - 3ms/step\n",
      "Epoch 119/150\n",
      "1179/1179 - 6s - loss: 36425764.0000 - val_loss: 39745412.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 120/150\n",
      "1179/1179 - 6s - loss: 37081276.0000 - val_loss: 44847260.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 121/150\n",
      "1179/1179 - 6s - loss: 36554704.0000 - val_loss: 41665712.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 122/150\n",
      "1179/1179 - 6s - loss: 36334112.0000 - val_loss: 40808676.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 123/150\n",
      "1179/1179 - 5s - loss: 36413404.0000 - val_loss: 38690508.0000 - 5s/epoch - 5ms/step\n",
      "Epoch 124/150\n",
      "1179/1179 - 5s - loss: 36511352.0000 - val_loss: 45004224.0000 - 5s/epoch - 4ms/step\n",
      "Epoch 125/150\n",
      "1179/1179 - 6s - loss: 36376164.0000 - val_loss: 40897232.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 126/150\n",
      "1179/1179 - 6s - loss: 36187872.0000 - val_loss: 40938856.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 127/150\n",
      "1179/1179 - 4s - loss: 36356008.0000 - val_loss: 44793424.0000 - 4s/epoch - 3ms/step\n",
      "Epoch 128/150\n",
      "1179/1179 - 4s - loss: 36766148.0000 - val_loss: 41551660.0000 - 4s/epoch - 4ms/step\n",
      "Epoch 129/150\n",
      "1179/1179 - 5s - loss: 36082836.0000 - val_loss: 39343832.0000 - 5s/epoch - 4ms/step\n",
      "Epoch 130/150\n",
      "1179/1179 - 5s - loss: 36357196.0000 - val_loss: 42245460.0000 - 5s/epoch - 4ms/step\n",
      "Epoch 131/150\n",
      "1179/1179 - 6s - loss: 35994556.0000 - val_loss: 40942976.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 132/150\n",
      "1179/1179 - 5s - loss: 36128756.0000 - val_loss: 41061648.0000 - 5s/epoch - 4ms/step\n",
      "Epoch 133/150\n",
      "1179/1179 - 6s - loss: 36520944.0000 - val_loss: 42390884.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 134/150\n",
      "1179/1179 - 6s - loss: 36269280.0000 - val_loss: 43373276.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 135/150\n",
      "1179/1179 - 6s - loss: 36679896.0000 - val_loss: 41865512.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 136/150\n",
      "1179/1179 - 5s - loss: 36459756.0000 - val_loss: 42349372.0000 - 5s/epoch - 4ms/step\n",
      "Epoch 137/150\n",
      "1179/1179 - 6s - loss: 36316972.0000 - val_loss: 41735432.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 138/150\n",
      "1179/1179 - 5s - loss: 35979368.0000 - val_loss: 42912156.0000 - 5s/epoch - 4ms/step\n",
      "Epoch 139/150\n",
      "1179/1179 - 5s - loss: 36477816.0000 - val_loss: 42925904.0000 - 5s/epoch - 4ms/step\n",
      "Epoch 140/150\n",
      "1179/1179 - 5s - loss: 36029836.0000 - val_loss: 40673348.0000 - 5s/epoch - 4ms/step\n",
      "Epoch 141/150\n",
      "1179/1179 - 6s - loss: 36495952.0000 - val_loss: 41457324.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 142/150\n",
      "1179/1179 - 6s - loss: 36329660.0000 - val_loss: 39787364.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 143/150\n",
      "1179/1179 - 6s - loss: 36218440.0000 - val_loss: 39986992.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 144/150\n",
      "1179/1179 - 6s - loss: 35697700.0000 - val_loss: 43494832.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 145/150\n",
      "1179/1179 - 6s - loss: 36215248.0000 - val_loss: 40811236.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 146/150\n",
      "1179/1179 - 5s - loss: 36098780.0000 - val_loss: 41676040.0000 - 5s/epoch - 5ms/step\n",
      "Epoch 147/150\n",
      "1179/1179 - 6s - loss: 36269200.0000 - val_loss: 41738724.0000 - 6s/epoch - 5ms/step\n",
      "Epoch 148/150\n",
      "1179/1179 - 5s - loss: 36209520.0000 - val_loss: 41210828.0000 - 5s/epoch - 4ms/step\n",
      "Epoch 149/150\n",
      "1179/1179 - 4s - loss: 36148792.0000 - val_loss: 41931992.0000 - 4s/epoch - 4ms/step\n",
      "Epoch 150/150\n",
      "1179/1179 - 4s - loss: 36445776.0000 - val_loss: 42039972.0000 - 4s/epoch - 3ms/step\n",
      "KNeighborsRegressor(n_neighbors=8)\n",
      "\tTraining time: 0.006s\n",
      "\tPrediction time: 3.378s\n",
      "\tExplained variance: 0.7775701324880491\n",
      "\tMean absolute error: 4930.236393778267\n",
      "\tR2 score: 0.7772572273066429\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from neural_net import ann_model, mse, test_dataset, test_labels, train_dataset, train_labels\n",
    "from KNN import knn_model, pipe95\n",
    "import pandas as pd"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's define a model class that's extensible for ensembling other models. [All my own code]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "class Model:\n",
    "    def __init__(self, knn_weight, ann_weight, bayes_weight, logreg_weight) -> None:\n",
    "        self.knn_weight = knn_weight\n",
    "        self.ann_weight = ann_weight\n",
    "        # self.bayes_weight = bayes_weight\n",
    "        # self.logreg_weight = logreg_weight\n",
    "        assert(abs(1 - (knn_weight + ann_weight + bayes_weight + logreg_weight)) < 1e-3)\n",
    "        self.knn = knn_model\n",
    "        self.ann = ann_model\n",
    "        #self.bayes = bayes_model\n",
    "        #self.logreg = logreg_model\n",
    "    \n",
    "    def predict(self, x):\n",
    "        return self.knn.predict(pipe95.transform(x))*self.knn_weight + self.ann.predict(x)*self.ann_weight # + self.bayes_model.predict(x)*self.bayes_weight + self.logreg.predict(x)*self.logreg_weight"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, let's see how to weight the KNN in conjunction with the neural net."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "228/228 [==============================] - 1s 3ms/step\n",
      "228/228 [==============================] - 1s 5ms/step\n",
      "228/228 [==============================] - 1s 4ms/step\n",
      "228/228 [==============================] - 1s 5ms/step\n",
      "228/228 [==============================] - 1s 3ms/step\n",
      "228/228 [==============================] - 1s 4ms/step\n",
      "228/228 [==============================] - 1s 3ms/step\n",
      "228/228 [==============================] - 1s 5ms/step\n",
      "228/228 [==============================] - 1s 5ms/step\n",
      "228/228 [==============================] - 1s 3ms/step\n",
      "228/228 [==============================] - 1s 3ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[274648761.8839384,\n",
       " 230611196.3233572,\n",
       " 191192253.04992944,\n",
       " 156391929.18792108,\n",
       " 126210233.36453895,\n",
       " 100647160.36900961,\n",
       " 79702699.33894704,\n",
       " 63376876.83562347,\n",
       " 51669661.43961608,\n",
       " 44581078.5941247,\n",
       " 42111110.37458343]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "neural_net_weights = []\n",
    "mses = []\n",
    "\n",
    "for weight in [0.0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0]:\n",
    "    _model = Model(ann_weight=weight, knn_weight=1.0-weight, bayes_weight=0.0, logreg_weight=0.0)   \n",
    "    y_prime = _model.predict(test_dataset.drop(columns=['IncomePerCap']))\n",
    "    y = test_labels.to_numpy()\n",
    "    # print(y_prime, y, y_prime.shape, y.shape)\n",
    "    neural_net_weights.append(100*weight)\n",
    "    mses.append(mse(y, y_prime))\n",
    "\n",
    "mses"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import seaborn as sns\n",
    "sns.set_style('whitegrid')\n",
    "plt.plot(neural_net_weights, mses, color='orange')\n",
    "plt.xlabel('Percentage of weight on ANN prediction')\n",
    "plt.ylabel('MSE')\n",
    "plt.show()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As you can see, the neural net is so much better than KNN that there's no benefit combining them."
   ]
  }
 ],
 "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": "3.11.2"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}