Deep_NN.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import torch.nn as nn\n",
    "from sklearn import datasets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1000 500\n"
     ]
    }
   ],
   "source": [
    "n_pts = 500\n",
    "X, y = datasets.make_circles(n_samples=n_pts, random_state=20, noise=0.13, factor=0.2)\n",
    "print(X.size, y.size)\n",
    "x_data = torch.Tensor(X)\n",
    "y_data = torch.Tensor(y.reshape(500, 1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "def scatter_plot():\n",
    "    plt.scatter(X[y==0, 0], X[y==0, 1]) # Blue\n",
    "    plt.scatter(X[y==1, 0], X[y==1, 1]) # Orange\n",
    "    plt.axis('scaled')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "scatter_plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "class Model(nn.Module):\n",
    "  def __init__(self, input_size, H1, H2, H3, output_size):\n",
    "    super().__init__()\n",
    "    self.linear = nn.Linear(input_size, H1)\n",
    "    self.linear2 = nn.Linear(H1, H2)\n",
    "    self.linear3 = nn.Linear(H2, H3)\n",
    "    self.linear4 = nn.Linear(H3, output_size)\n",
    "  def forward(self, x):\n",
    "    x = torch.sigmoid(self.linear(x))\n",
    "    x = torch.sigmoid(self.linear2(x))\n",
    "    x = torch.sigmoid(self.linear3(x))\n",
    "    x = torch.sigmoid(self.linear4(x))\n",
    "    return x\n",
    "  def predict(self, x):\n",
    "    pred = self.forward(x)\n",
    "    if pred >= 0.5:\n",
    "      return 1\n",
    "    else: \n",
    "      return 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[Parameter containing:\n",
      "tensor([[ 0.1622, -0.1683],\n",
      "        [ 0.1939, -0.0361],\n",
      "        [ 0.3021,  0.1683],\n",
      "        [-0.0813, -0.5717]], requires_grad=True), Parameter containing:\n",
      "tensor([ 0.1614, -0.6260,  0.0929,  0.0470], requires_grad=True), Parameter containing:\n",
      "tensor([[-0.1099,  0.4088,  0.0334,  0.2073],\n",
      "        [ 0.2116, -0.2950, -0.1922,  0.4809],\n",
      "        [-0.4897, -0.0340, -0.0396,  0.3547],\n",
      "        [-0.0475,  0.1317, -0.0240, -0.2800]], requires_grad=True), Parameter containing:\n",
      "tensor([-0.2834, -0.2429, -0.4542, -0.3245], requires_grad=True), Parameter containing:\n",
      "tensor([[ 0.1177,  0.3291,  0.0246, -0.2292],\n",
      "        [ 0.2197, -0.1919, -0.1108, -0.2741],\n",
      "        [-0.1570, -0.4633,  0.2133,  0.1944]], requires_grad=True), Parameter containing:\n",
      "tensor([0.0993, 0.2455, 0.2119], requires_grad=True), Parameter containing:\n",
      "tensor([[0.0255, 0.0611, 0.0441]], requires_grad=True), Parameter containing:\n",
      "tensor([0.3081], requires_grad=True)]\n"
     ]
    }
   ],
   "source": [
    "torch.manual_seed(2)\n",
    "model = Model(2, 4, 4, 3, 1)\n",
    "print(list(model.parameters()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "criterion = nn.BCELoss()\n",
    "optimizer = torch.optim.Adam(model.parameters(), lr=0.1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 0 loss 0.7108473777770996\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 1 loss 0.6952362060546875\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 2 loss 0.6937694549560547\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 3 loss 0.6971819996833801\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 4 loss 0.6976629495620728\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 5 loss 0.6956403851509094\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 6 loss 0.6937021017074585\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 7 loss 0.6930345296859741\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 8 loss 0.6933485865592957\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 9 loss 0.693918764591217\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 10 loss 0.6942744851112366\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 11 loss 0.6942707300186157\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 12 loss 0.6939638257026672\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 13 loss 0.6934999227523804\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 14 loss 0.6930370330810547\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 15 loss 0.6926935315132141\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 16 loss 0.6925193667411804\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 17 loss 0.6924903988838196\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 18 loss 0.6925264596939087\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 19 loss 0.6925249695777893\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 20 loss 0.6924009919166565\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 21 loss 0.6921126842498779\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 22 loss 0.6916623115539551\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 23 loss 0.6910765767097473\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 24 loss 0.6903777122497559\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 25 loss 0.6895580887794495\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 26 loss 0.6885629296302795\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 27 loss 0.6872910261154175\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 28 loss 0.685608983039856\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 29 loss 0.6833812594413757\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 30 loss 0.6804977059364319\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 31 loss 0.6768878102302551\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 32 loss 0.6724549531936646\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 33 loss 0.666930615901947\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 34 loss 0.6599811911582947\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 35 loss 0.6515145897865295\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 36 loss 0.6417653560638428\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 37 loss 0.6311525702476501\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 38 loss 0.6201736927032471\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 39 loss 0.6094537973403931\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 40 loss 0.5996927618980408\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 41 loss 0.5914125442504883\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 42 loss 0.5848241448402405\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 43 loss 0.5798713564872742\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 44 loss 0.5762720108032227\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 45 loss 0.5736197829246521\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 46 loss 0.5715599656105042\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 47 loss 0.5699527263641357\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 48 loss 0.5687289237976074\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 49 loss 0.5674773454666138\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 50 loss 0.5656585693359375\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 51 loss 0.5633635520935059\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 52 loss 0.5601596236228943\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 53 loss 0.5538081526756287\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 54 loss 0.5431280732154846\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 55 loss 0.5320253968238831\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 56 loss 0.5279895663261414\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 57 loss 0.5103381276130676\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 58 loss 0.4941960871219635\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 59 loss 0.47662001848220825\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 60 loss 0.45270341634750366\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 61 loss 0.42961812019348145\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 62 loss 0.4148339331150055\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 63 loss 0.3956161141395569\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 64 loss 0.38239815831184387\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 65 loss 0.37137916684150696\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 66 loss 0.3589176833629608\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 67 loss 0.3533436357975006\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 68 loss 0.3490615487098694\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 69 loss 0.3457699120044708\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 70 loss 0.3464547097682953\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 71 loss 0.34546345472335815\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 72 loss 0.3434210419654846\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 73 loss 0.3425619900226593\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 74 loss 0.34041377902030945\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 75 loss 0.3370762765407562\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 76 loss 0.33444222807884216\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 77 loss 0.33147385716438293\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 78 loss 0.32908689975738525\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 79 loss 0.3273159861564636\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 80 loss 0.32331350445747375\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 81 loss 0.31930071115493774\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 82 loss 0.3142613172531128\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 83 loss 0.30889132618904114\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 84 loss 0.3042079508304596\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 85 loss 0.2982664108276367\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 86 loss 0.291772723197937\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 87 loss 0.2843581438064575\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 88 loss 0.2750806510448456\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 89 loss 0.2649516761302948\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 90 loss 0.25357839465141296\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 91 loss 0.2407079041004181\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 92 loss 0.22643373906612396\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 93 loss 0.20788775384426117\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 94 loss 0.18559786677360535\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 95 loss 0.16380174458026886\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 96 loss 0.13971681892871857\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 97 loss 0.119268499314785\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 98 loss 0.10557819902896881\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 99 loss 0.09483570605516434\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 100 loss 0.08469575643539429\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 101 loss 0.07564723491668701\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 102 loss 0.06860531866550446\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 103 loss 0.06371138989925385\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 104 loss 0.060220398008823395\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 105 loss 0.05704408884048462\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 106 loss 0.053745120763778687\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 107 loss 0.050421297550201416\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 108 loss 0.04705435037612915\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 109 loss 0.04366616904735565\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 110 loss 0.040404003113508224\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 111 loss 0.03731946647167206\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 112 loss 0.03440932184457779\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 113 loss 0.031739868223667145\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 114 loss 0.029430774971842766\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 115 loss 0.027605855837464333\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 116 loss 0.026257377117872238\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 117 loss 0.025131704285740852\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 118 loss 0.02400820516049862\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 119 loss 0.02292429283261299\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 120 loss 0.02190941572189331\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 121 loss 0.0208748672157526\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 122 loss 0.019858859479427338\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 123 loss 0.018996315076947212\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 124 loss 0.018298979848623276\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 125 loss 0.017687225714325905\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 126 loss 0.01713070645928383\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 127 loss 0.016648519784212112\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 128 loss 0.016237910836935043\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 129 loss 0.015863776206970215\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 130 loss 0.015503819100558758\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 131 loss 0.015166250057518482\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 132 loss 0.014856437221169472\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 133 loss 0.014557507820427418\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 134 loss 0.01425087545067072\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 135 loss 0.01393580250442028\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 136 loss 0.013622535392642021\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 137 loss 0.013314739800989628\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 138 loss 0.013007069006562233\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 139 loss 0.012697748839855194\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 140 loss 0.012391519732773304\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 141 loss 0.012088773772120476\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 142 loss 0.011783498339354992\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 143 loss 0.011473329737782478\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 144 loss 0.011163484305143356\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 145 loss 0.010858353227376938\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 146 loss 0.010556265711784363\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 147 loss 0.01025511883199215\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 148 loss 0.00995701178908348\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 149 loss 0.009663938544690609\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 150 loss 0.009374945424497128\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 151 loss 0.0090901218354702\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 152 loss 0.008813059888780117\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 153 loss 0.008546393364667892\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 154 loss 0.008289165794849396\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 155 loss 0.008040392771363258\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 156 loss 0.007801114581525326\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 157 loss 0.007571383845061064\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 158 loss 0.007349968887865543\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 159 loss 0.007137266453355551\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "epoch: 160 loss 0.006935118697583675\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 161 loss 0.006744070909917355\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 162 loss 0.006563511211425066\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 163 loss 0.00639338931068778\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 164 loss 0.006233778782188892\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 165 loss 0.006083574611693621\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 166 loss 0.005941149313002825\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 167 loss 0.005805585999041796\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 168 loss 0.005676536820828915\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 169 loss 0.005553508177399635\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 170 loss 0.0054359701462090015\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 171 loss 0.005323705729097128\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 172 loss 0.005216566380113363\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 173 loss 0.005114095285534859\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 174 loss 0.0050156693905591965\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 175 loss 0.00492088682949543\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 176 loss 0.0048295496962964535\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 177 loss 0.00474145682528615\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 178 loss 0.004656333476305008\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 179 loss 0.0045740073546767235\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 180 loss 0.004494434222579002\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 181 loss 0.004417576361447573\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 182 loss 0.00434329267591238\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 183 loss 0.004271429963409901\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 184 loss 0.004201866220682859\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 185 loss 0.0041345288045704365\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 186 loss 0.0040693203918635845\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 187 loss 0.004006110597401857\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 188 loss 0.0039447760209441185\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 189 loss 0.003885248675942421\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 190 loss 0.0038274554535746574\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 191 loss 0.003771333722397685\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 192 loss 0.003716798732057214\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 193 loss 0.003663788316771388\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 194 loss 0.0036122493911534548\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 195 loss 0.003562128869816661\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 196 loss 0.0035133794881403446\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 197 loss 0.003465929999947548\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 198 loss 0.003419724991545081\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 199 loss 0.003374716266989708\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 200 loss 0.003330860985442996\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 201 loss 0.0032881121151149273\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 202 loss 0.00324643193744123\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 203 loss 0.003205769695341587\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 204 loss 0.0031661014072597027\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 205 loss 0.0031273914501070976\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 206 loss 0.003089612117037177\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 207 loss 0.0030527161434292793\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 208 loss 0.003016677452251315\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 209 loss 0.0029814611189067364\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 210 loss 0.002947036176919937\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 211 loss 0.002913373289629817\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 212 loss 0.0028804410248994827\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 213 loss 0.002848223317414522\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 214 loss 0.00281668733805418\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 215 loss 0.0027858142275363207\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 216 loss 0.0027555772103369236\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 217 loss 0.0027259644120931625\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 218 loss 0.0026969481259584427\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 219 loss 0.002668517641723156\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 220 loss 0.0026406417600810528\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 221 loss 0.002613306976854801\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 222 loss 0.002586500486359\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 223 loss 0.002560203429311514\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 224 loss 0.002534397877752781\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 225 loss 0.002509067999199033\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 226 loss 0.0024842096026986837\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 227 loss 0.002459796378389001\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 228 loss 0.0024358280934393406\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 229 loss 0.0024122819304466248\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 230 loss 0.002389150205999613\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 231 loss 0.002366419415920973\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 232 loss 0.0023440863005816936\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 233 loss 0.0023221285082399845\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 234 loss 0.00230054697021842\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 235 loss 0.0022793221287429333\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 236 loss 0.002258450025692582\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 237 loss 0.002237922977656126\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 238 loss 0.002217728877440095\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 239 loss 0.002197858178988099\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 240 loss 0.0021783101838082075\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 241 loss 0.0021590678952634335\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 242 loss 0.002140128519386053\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 243 loss 0.002121488330885768\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 244 loss 0.002103132661432028\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 245 loss 0.0020850575529038906\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 246 loss 0.002067260444164276\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 247 loss 0.0020497306250035763\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 248 loss 0.0020324622746556997\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 249 loss 0.0020154519006609917\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 250 loss 0.001998691586777568\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 251 loss 0.001982178771868348\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 252 loss 0.0019659006502479315\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 253 loss 0.0019498602487146854\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 254 loss 0.0019340525614097714\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 255 loss 0.0019184623379260302\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 256 loss 0.0019031006377190351\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 257 loss 0.0018879467388615012\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 258 loss 0.001873005647212267\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 259 loss 0.001858268748037517\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 260 loss 0.0018437399994581938\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 261 loss 0.0018294090405106544\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 262 loss 0.0018152677221223712\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 263 loss 0.001801318139769137\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 264 loss 0.0017875549383461475\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 265 loss 0.0017739712493494153\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 266 loss 0.0017605738248676062\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 267 loss 0.0017473460175096989\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 268 loss 0.0017342949286103249\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 269 loss 0.0017214082181453705\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 270 loss 0.0017086885636672378\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 271 loss 0.0016961314249783754\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 272 loss 0.00168373784981668\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 273 loss 0.0016714956145733595\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 274 loss 0.0016594082117080688\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "epoch: 275 loss 0.001647471566684544\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 276 loss 0.0016356816049665213\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 277 loss 0.0016240396071225405\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 278 loss 0.0016125409165397286\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 279 loss 0.0016011818079277873\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 280 loss 0.001589958555996418\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 281 loss 0.0015788720920681953\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 282 loss 0.001567914499901235\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 283 loss 0.0015570919495075941\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 284 loss 0.0015463951276615262\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 285 loss 0.0015358293894678354\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 286 loss 0.0015253785531967878\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 287 loss 0.0015150585677474737\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 288 loss 0.0015048549976199865\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 289 loss 0.0014947669114917517\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 290 loss 0.001484796404838562\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 291 loss 0.0014749389374628663\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 292 loss 0.0014651915989816189\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 293 loss 0.0014555564848706126\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 294 loss 0.001446031266823411\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 295 loss 0.001436609192751348\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 296 loss 0.001427292823791504\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 297 loss 0.0014180797152221203\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 298 loss 0.0014089660253375769\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 299 loss 0.0013999552465975285\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 300 loss 0.0013910428388044238\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 301 loss 0.0013822229811921716\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 302 loss 0.0013735032407566905\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 303 loss 0.0013648727908730507\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 304 loss 0.0013563396641984582\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 305 loss 0.0013478909386321902\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 306 loss 0.0013395353453233838\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 307 loss 0.001331264153122902\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 308 loss 0.0013230859767645597\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 309 loss 0.0013149863807484508\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 310 loss 0.0013069773558527231\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 311 loss 0.0012990463292226195\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 312 loss 0.0012912015663459897\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 313 loss 0.001283430028706789\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 314 loss 0.0012757469667121768\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 315 loss 0.0012681324733421206\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 316 loss 0.00126060307957232\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 317 loss 0.0012531416723504663\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 318 loss 0.001245762687176466\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 319 loss 0.0012384506408125162\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 320 loss 0.0012312146136537194\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 321 loss 0.0012240505311638117\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 322 loss 0.001216949662193656\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 323 loss 0.0012099294690415263\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 324 loss 0.0012029700446873903\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 325 loss 0.0011960844276472926\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 326 loss 0.0011892594629898667\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 327 loss 0.0011824978282675147\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 328 loss 0.0011758108157664537\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 329 loss 0.001169179449789226\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 330 loss 0.0011626158375293016\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 331 loss 0.0011561125284060836\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 332 loss 0.0011496691731736064\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 333 loss 0.0011432900791987777\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 334 loss 0.0011369676794856787\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 335 loss 0.0011306984815746546\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 336 loss 0.0011245000641793013\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 337 loss 0.001118347398005426\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 338 loss 0.0011122567811980844\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 339 loss 0.0011062263511121273\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 340 loss 0.0011002414394170046\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 341 loss 0.0010943126399070024\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 342 loss 0.0010884471703320742\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 343 loss 0.0010826262878254056\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 344 loss 0.0010768546489998698\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 345 loss 0.0010711416834965348\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 346 loss 0.0010654781945049763\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 347 loss 0.0010598610388115048\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 348 loss 0.0010543027892708778\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 349 loss 0.0010487826075404882\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 350 loss 0.0010433156276121736\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 351 loss 0.0010378957958891988\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 352 loss 0.001032524392940104\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 353 loss 0.0010271983919665217\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 354 loss 0.001021915697492659\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 355 loss 0.0010166828287765384\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 356 loss 0.0010114978067576885\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 357 loss 0.0010063493391498923\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 358 loss 0.0010012463899329305\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 359 loss 0.0009961938485503197\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 360 loss 0.0009911776287481189\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 361 loss 0.0009862023871392012\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 362 loss 0.0009812674252316356\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 363 loss 0.0009763840935193002\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 364 loss 0.0009715357446111739\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 365 loss 0.0009667238336987793\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 366 loss 0.0009619491174817085\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 367 loss 0.0009572272538207471\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 368 loss 0.0009525361238047481\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 369 loss 0.0009478803840465844\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 370 loss 0.0009432614897377789\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 371 loss 0.0009386870660819113\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 372 loss 0.0009341544355265796\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 373 loss 0.0009296476491726935\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 374 loss 0.00092517607845366\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 375 loss 0.0009207453113049269\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 376 loss 0.000916354067157954\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 377 loss 0.0009119908208958805\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 378 loss 0.0009076623828150332\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 379 loss 0.0009033706737682223\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 380 loss 0.0008991078939288855\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 381 loss 0.0008948901668190956\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 382 loss 0.000890699215233326\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 383 loss 0.0008865398121997714\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 384 loss 0.0008824074757285416\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 385 loss 0.0008783128578215837\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 386 loss 0.000874254503287375\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 387 loss 0.000870224554091692\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 388 loss 0.0008662217296659946\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 389 loss 0.0008622499299235642\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 390 loss 0.0008583037415519357\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "epoch: 391 loss 0.0008543950389139354\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 392 loss 0.0008505209116265178\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 393 loss 0.000846670416649431\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 394 loss 0.0008428460569120944\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 395 loss 0.0008390512084588408\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 396 loss 0.0008352804579772055\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 397 loss 0.0008315497543662786\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 398 loss 0.0008278419263660908\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 399 loss 0.0008241616887971759\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 400 loss 0.0008205047342926264\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 401 loss 0.000816873915027827\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 402 loss 0.0008132683578878641\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 403 loss 0.0008096990641206503\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 404 loss 0.0008061525295488536\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 405 loss 0.0008026316063478589\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 406 loss 0.0007991308229975402\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 407 loss 0.0007956565241329372\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 408 loss 0.0007922074291855097\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 409 loss 0.0007887851097621024\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 410 loss 0.0007853905553929508\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 411 loss 0.0007820198079571128\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 412 loss 0.0007786715868860483\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 413 loss 0.000775342108681798\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 414 loss 0.0007720365538261831\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 415 loss 0.000768756028264761\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 416 loss 0.0007654976798221469\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 417 loss 0.0007622716366313398\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 418 loss 0.0007590618915855885\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 419 loss 0.0007558773504570127\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 420 loss 0.0007527071284130216\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 421 loss 0.0007495608297176659\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 422 loss 0.000746437581256032\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 423 loss 0.0007433353457599878\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 424 loss 0.0007402540650218725\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 425 loss 0.000737202528398484\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 426 loss 0.0007341657183133066\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 427 loss 0.0007311482331715524\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 428 loss 0.0007281522266566753\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 429 loss 0.0007251744391396642\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 430 loss 0.0007222145795822144\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 431 loss 0.0007192784105427563\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 432 loss 0.0007163588306866586\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 433 loss 0.0007134688785299659\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 434 loss 0.000710593827534467\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 435 loss 0.0007077347254380584\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 436 loss 0.0007048967527225614\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 437 loss 0.0007020743796601892\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 438 loss 0.0006992697599343956\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 439 loss 0.0006964848726056516\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 440 loss 0.00069371925201267\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 441 loss 0.0006909685325808823\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 442 loss 0.0006882476736791432\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 443 loss 0.000685536244418472\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 444 loss 0.0006828453624621034\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 445 loss 0.0006801680428907275\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 446 loss 0.0006775070214644074\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 447 loss 0.0006748651503585279\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 448 loss 0.0006722388206981122\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 449 loss 0.0006696247146464884\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 450 loss 0.0006670340080745518\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 451 loss 0.0006644575041718781\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 452 loss 0.0006619053892791271\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 453 loss 0.0006593653233721852\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 454 loss 0.0006568389944732189\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 455 loss 0.0006543303607031703\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 456 loss 0.0006518354057334363\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 457 loss 0.0006493554683402181\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 458 loss 0.0006468847277574241\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 459 loss 0.0006444361642934382\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 460 loss 0.0006420051213353872\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 461 loss 0.0006395845557563007\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 462 loss 0.0006371811032295227\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 463 loss 0.0006348010501824319\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 464 loss 0.000632431183476001\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 465 loss 0.0006300724926404655\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 466 loss 0.0006277297507040203\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 467 loss 0.0006253964966163039\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 468 loss 0.0006230801809579134\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 469 loss 0.0006207764963619411\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 470 loss 0.0006184875965118408\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 471 loss 0.000616212491877377\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 472 loss 0.0006139525794424117\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 473 loss 0.0006117047159932554\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 474 loss 0.0006094707059673965\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 475 loss 0.0006072543328627944\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 476 loss 0.0006050522206351161\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 477 loss 0.0006028597708791494\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 478 loss 0.0006006823386996984\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 479 loss 0.0005985143943689764\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 480 loss 0.0005963586736470461\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 481 loss 0.0005942138959653676\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 482 loss 0.0005920861149206758\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 483 loss 0.0005899688112549484\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 484 loss 0.0005878612864762545\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 485 loss 0.0005857711657881737\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 486 loss 0.0005836902419105172\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 487 loss 0.0005816196207888424\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 488 loss 0.0005795609322376549\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 489 loss 0.000577528087887913\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 490 loss 0.0005755004822276533\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 491 loss 0.0005734755541197956\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 492 loss 0.0005714708822779357\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 493 loss 0.0005694741848856211\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 494 loss 0.0005674895364791155\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 495 loss 0.0005655150744132698\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 496 loss 0.0005635492852889001\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 497 loss 0.0005615945556201041\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 498 loss 0.0005596536211669445\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 499 loss 0.0005577229312621057\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 500 loss 0.0005558024276979268\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 501 loss 0.0005538926343433559\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 502 loss 0.0005519923288375139\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 503 loss 0.0005501045961864293\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 504 loss 0.0005482299602590501\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 505 loss 0.0005463694105856121\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 506 loss 0.000544519629329443\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 507 loss 0.0005426755524240434\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 508 loss 0.0005408411961980164\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 509 loss 0.0005390153382904828\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 510 loss 0.0005371993174776435\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 511 loss 0.0005353947635740042\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 512 loss 0.0005335978930816054\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 513 loss 0.0005318134790286422\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 514 loss 0.0005300371558405459\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 515 loss 0.0005282697966322303\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 516 loss 0.0005265131476335227\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 517 loss 0.0005247694789431989\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 518 loss 0.0005230283131822944\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 519 loss 0.0005212976248003542\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 520 loss 0.0005195779958739877\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 521 loss 0.0005178707651793957\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 522 loss 0.0005161739536561072\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 523 loss 0.0005144845927134156\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 524 loss 0.0005128054763190448\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 525 loss 0.0005111350328661501\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 526 loss 0.000509473611600697\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 527 loss 0.0005078208050690591\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 528 loss 0.0005061707342974842\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 529 loss 0.0005045330035500228\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 530 loss 0.0005029037711210549\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 531 loss 0.0005012822221033275\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 532 loss 0.0004996686475351453\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 533 loss 0.0004980669473297894\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 534 loss 0.0004964679246768355\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 535 loss 0.0004948804271407425\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 536 loss 0.0004932981682941318\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 537 loss 0.0004917283076792955\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 538 loss 0.0004901658976450562\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 539 loss 0.0004886144306510687\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 540 loss 0.00048706590314395726\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 541 loss 0.00048552468069829047\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 542 loss 0.00048399402294307947\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 543 loss 0.00048247206723317504\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 544 loss 0.00048096582759171724\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 545 loss 0.00047946179984137416\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 546 loss 0.0004779648152180016\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 547 loss 0.0004764724289998412\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 548 loss 0.0004749900836031884\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 549 loss 0.0004735130933113396\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 550 loss 0.0004720431752502918\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 551 loss 0.0004705797182396054\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 552 loss 0.00046912700054235756\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 553 loss 0.00046768077299930155\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 554 loss 0.00046623655362054706\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 555 loss 0.00046480653691105545\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 556 loss 0.0004633789067156613\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 557 loss 0.0004619597748387605\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 558 loss 0.0004605488502420485\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 559 loss 0.0004591411561705172\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 560 loss 0.0004577445797622204\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 561 loss 0.0004563530383165926\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 562 loss 0.00045497019891627133\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 563 loss 0.00045359513023868203\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 564 loss 0.00045222259359434247\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 565 loss 0.0004508586134761572\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 566 loss 0.00044949993025511503\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 567 loss 0.00044814811553806067\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 568 loss 0.0004468102124519646\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 569 loss 0.000445479090558365\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 570 loss 0.00044414924923330545\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 571 loss 0.00044283142779022455\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 572 loss 0.00044150862959213555\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 573 loss 0.0004402005288284272\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 574 loss 0.0004388954257592559\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 575 loss 0.00043759707477875054\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 576 loss 0.00043630358413793147\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 577 loss 0.00043501591426320374\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 578 loss 0.0004337344435043633\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 579 loss 0.0004324612091295421\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 580 loss 0.00043119321344420314\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 581 loss 0.00042992798262275755\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 582 loss 0.0004286740149836987\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 583 loss 0.00042742295772768557\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 584 loss 0.0004261754220351577\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 585 loss 0.0004249415942467749\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 586 loss 0.00042370602022856474\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 587 loss 0.00042247812962159514\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 588 loss 0.0004212585627101362\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 589 loss 0.00042004339047707617\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 590 loss 0.0004188338352832943\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 591 loss 0.0004176288493908942\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 592 loss 0.0004164307320024818\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 593 loss 0.00041523590334691107\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 594 loss 0.00041404765215702355\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 595 loss 0.00041286638588644564\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 596 loss 0.00041168963070958853\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 597 loss 0.0004105194238945842\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 598 loss 0.0004093518073204905\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 599 loss 0.0004081958904862404\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 600 loss 0.0004070464929100126\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 601 loss 0.0004058987833559513\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 602 loss 0.0004047568654641509\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 603 loss 0.00040361948776990175\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 604 loss 0.00040248531149700284\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 605 loss 0.0004013579455204308\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 606 loss 0.0004002333153039217\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 607 loss 0.0003991136036347598\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 608 loss 0.0003979997127316892\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 609 loss 0.00039689469849690795\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 610 loss 0.00039578607538715005\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 611 loss 0.00039468571776524186\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 612 loss 0.00039359359652735293\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 613 loss 0.0003925045020878315\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 614 loss 0.0003914200933650136\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 615 loss 0.0003903410106431693\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 616 loss 0.00038926652632653713\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 617 loss 0.0003881965531036258\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 618 loss 0.00038712946115992963\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 619 loss 0.00038606900488957763\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 620 loss 0.0003850154753308743\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 621 loss 0.0003839645069092512\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 622 loss 0.0003829170891549438\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 623 loss 0.0003818772966042161\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 624 loss 0.00038083959952928126\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 625 loss 0.0003798057441599667\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 626 loss 0.00037877875729463995\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 627 loss 0.00037775232340209186\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 628 loss 0.00037673418410122395\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 629 loss 0.0003757187514565885\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 630 loss 0.00037470849929377437\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 631 loss 0.0003737027000170201\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 632 loss 0.0003726994327735156\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 633 loss 0.00037170370342209935\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 634 loss 0.0003707088471855968\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 635 loss 0.0003697201900649816\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 636 loss 0.00036873354110866785\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 637 loss 0.0003677553031593561\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 638 loss 0.00036678279866464436\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 639 loss 0.0003658139321487397\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 640 loss 0.00036484963493421674\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 641 loss 0.0003638904890976846\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 642 loss 0.0003629325656220317\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 643 loss 0.0003619754279498011\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 644 loss 0.00036102571175433695\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 645 loss 0.00036007698508910835\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 646 loss 0.00035913288593292236\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 647 loss 0.0003581948112696409\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 648 loss 0.0003572596178855747\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 649 loss 0.0003563282953109592\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 650 loss 0.0003553975257091224\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 651 loss 0.00035447659320198\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 652 loss 0.0003535565047059208\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 653 loss 0.0003526380460243672\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 654 loss 0.0003517255245242268\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 655 loss 0.00035081797977909446\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 656 loss 0.00034991453867405653\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 657 loss 0.0003490136587060988\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 658 loss 0.0003481140302028507\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 659 loss 0.00034722164855338633\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 660 loss 0.00034632987808436155\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 661 loss 0.00034544256050139666\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 662 loss 0.00034456144203431904\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 663 loss 0.00034368247725069523\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 664 loss 0.0003428095660638064\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 665 loss 0.0003419352578930557\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 666 loss 0.00034106592647731304\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 667 loss 0.0003401995054446161\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 668 loss 0.0003393419901840389\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 669 loss 0.00033848374732770026\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 670 loss 0.00033762745442800224\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 671 loss 0.0003367782337591052\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 672 loss 0.0003359282563906163\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 673 loss 0.00033508631167933345\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 674 loss 0.000334243755787611\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 675 loss 0.0003334058856125921\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 676 loss 0.0003325720317661762\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 677 loss 0.0003317413211334497\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 678 loss 0.0003309127641841769\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 679 loss 0.0003300912503618747\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 680 loss 0.0003292684268672019\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 681 loss 0.00032844956149347126\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 682 loss 0.0003276378265582025\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 683 loss 0.0003268239088356495\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 684 loss 0.00032601860584691167\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 685 loss 0.00032521403045393527\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 686 loss 0.00032441195799037814\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 687 loss 0.00032361314515583217\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 688 loss 0.0003228172427043319\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 689 loss 0.0003220262879040092\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 690 loss 0.00032123702112585306\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 691 loss 0.00032045887201093137\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 692 loss 0.0003196811012458056\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 693 loss 0.00031890306854620576\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 694 loss 0.00031812881934456527\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 695 loss 0.00031735660741105676\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 696 loss 0.0003165871894452721\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 697 loss 0.0003158200706820935\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 698 loss 0.0003150551638100296\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 699 loss 0.0003142945352010429\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 700 loss 0.0003135339939035475\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 701 loss 0.0003127800009679049\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 702 loss 0.00031202612444758415\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 703 loss 0.0003112789709120989\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 704 loss 0.0003105287323705852\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 705 loss 0.00030978594440966845\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 706 loss 0.0003090447571594268\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 707 loss 0.00030830514151602983\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 708 loss 0.0003075716085731983\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 709 loss 0.000306838599499315\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 710 loss 0.0003061080351471901\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 711 loss 0.00030537761631421745\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 712 loss 0.00030465624877251685\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 713 loss 0.00030393325141631067\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 714 loss 0.0003032147651538253\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 715 loss 0.00030249758856371045\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 716 loss 0.00030178570887073874\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 717 loss 0.0003010752552654594\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 718 loss 0.00030036739190109074\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 719 loss 0.00029966110014356673\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 720 loss 0.00029895661282353103\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 721 loss 0.0002982569858431816\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 722 loss 0.0002975577372126281\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 723 loss 0.00029686151538044214\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 724 loss 0.000296169426292181\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 725 loss 0.0002954798110295087\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 726 loss 0.00029479165095835924\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 727 loss 0.00029410625575110316\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 728 loss 0.00029342295601963997\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 729 loss 0.00029274201369844377\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 730 loss 0.00029206782346591353\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 731 loss 0.0002913918870035559\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 732 loss 0.00029071865719743073\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 733 loss 0.0002900497929658741\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 734 loss 0.00028938238392584026\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 735 loss 0.00028871462563984096\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 736 loss 0.00028805239708162844\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 737 loss 0.000287391128949821\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 738 loss 0.00028673437191173434\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 739 loss 0.00028607912827283144\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 740 loss 0.0002854270569514483\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 741 loss 0.0002847741707228124\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 742 loss 0.00028412678511813283\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 743 loss 0.0002834806509781629\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 744 loss 0.0002828363503795117\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 745 loss 0.00028219501837156713\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 746 loss 0.0002815572661347687\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 747 loss 0.00028091922285966575\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 748 loss 0.0002802837989293039\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 749 loss 0.0002796549815684557\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 750 loss 0.0002790222060866654\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 751 loss 0.0002783959498628974\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 752 loss 0.00027776797651313245\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 753 loss 0.00027714745374396443\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 754 loss 0.00027652556309476495\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 755 loss 0.0002759065537247807\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 756 loss 0.00027529022190719843\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 757 loss 0.0002746751415543258\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 758 loss 0.0002740636409725994\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 759 loss 0.00027345132548362017\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 760 loss 0.00027284424868412316\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 761 loss 0.0002722394419834018\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 762 loss 0.0002716351591516286\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 763 loss 0.00027103497995994985\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 764 loss 0.0002704366052057594\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 765 loss 0.0002698379976209253\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 766 loss 0.0002692416019272059\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 767 loss 0.0002686477964743972\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 768 loss 0.00026806662208400667\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 769 loss 0.0002674825955182314\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 770 loss 0.00026689679361879826\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 771 loss 0.00026631346554495394\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 772 loss 0.00026573159266263247\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 773 loss 0.0002651536196935922\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 774 loss 0.0002645722415763885\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 775 loss 0.00026399752823635936\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 776 loss 0.0002634227857924998\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 777 loss 0.00026285057538188994\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 778 loss 0.00026227947091683745\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 779 loss 0.0002617117715999484\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 780 loss 0.0002611449745018035\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 781 loss 0.00026057942886836827\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 782 loss 0.00026001702644862235\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 783 loss 0.0002594554389361292\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 784 loss 0.00025889527751132846\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 785 loss 0.0002583398891147226\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 786 loss 0.0002577841514721513\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 787 loss 0.0002572291123215109\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 788 loss 0.00025668140733614564\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 789 loss 0.0002561303263064474\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 790 loss 0.0002555797400418669\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 791 loss 0.00025503389770165086\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 792 loss 0.0002544937015045434\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 793 loss 0.00025395056582055986\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 794 loss 0.0002534082159399986\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 795 loss 0.00025286851450800896\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 796 loss 0.00025233361520804465\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 797 loss 0.0002517988905310631\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 798 loss 0.00025126500986516476\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 799 loss 0.00025073427241295576\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 800 loss 0.00025020301109179854\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 801 loss 0.0002496779488865286\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 802 loss 0.0002491514605935663\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 803 loss 0.0002486246230546385\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 804 loss 0.00024810386821627617\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 805 loss 0.00024758290965110064\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 806 loss 0.0002470610779710114\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 807 loss 0.00024654463049955666\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 808 loss 0.0002460297546349466\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 809 loss 0.00024551519891247153\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 810 loss 0.00024500073050148785\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 811 loss 0.00024449251941405237\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 812 loss 0.00024398189270868897\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 813 loss 0.0002434749621897936\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 814 loss 0.00024296747869811952\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 815 loss 0.00024246389511972666\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 816 loss 0.0002419611846562475\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 817 loss 0.00024146023497451097\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 818 loss 0.00024096216657198966\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 819 loss 0.00024046313774306327\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 820 loss 0.00023996748495846987\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 821 loss 0.00023947485897224396\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 822 loss 0.00023898253857623786\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 823 loss 0.00023848899581935257\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 824 loss 0.0002380016667302698\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 825 loss 0.0002375108888372779\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 826 loss 0.0002370268921367824\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 827 loss 0.00023654122196603566\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 828 loss 0.00023605664318893105\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 829 loss 0.00023557756503578275\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 830 loss 0.00023509381571784616\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 831 loss 0.00023461856471840292\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 832 loss 0.00023413766757585108\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 833 loss 0.00023366475943475962\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 834 loss 0.00023318949388340116\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 835 loss 0.00023271777899935842\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 836 loss 0.0002322461223229766\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 837 loss 0.00023177942784968764\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 838 loss 0.00023130842600949109\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 839 loss 0.0002308417606400326\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 840 loss 0.00023037716164253652\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 841 loss 0.00022991215519141406\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 842 loss 0.00022945209639146924\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 843 loss 0.00022899152827449143\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 844 loss 0.00022853221162222326\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 845 loss 0.00022807388450019062\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 846 loss 0.000227617216296494\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 847 loss 0.0002271632110932842\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 848 loss 0.000226709627895616\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 849 loss 0.0002262563502881676\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 850 loss 0.00022580564836971462\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 851 loss 0.00022535890457220376\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 852 loss 0.0002249107346870005\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 853 loss 0.00022446301591116935\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 854 loss 0.00022401746537070721\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 855 loss 0.00022357226407621056\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 856 loss 0.0002231341350125149\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 857 loss 0.0002226926590083167\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 858 loss 0.00022225115390028805\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 859 loss 0.00022181333042681217\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 860 loss 0.00022137569612823427\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 861 loss 0.000220941670704633\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 862 loss 0.00022050640836823732\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 863 loss 0.00022007577354088426\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 864 loss 0.00021964465850032866\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 865 loss 0.00021921342704445124\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 866 loss 0.00021878410188946873\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 867 loss 0.00021835515508428216\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 868 loss 0.00021793013729620725\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 869 loss 0.0002175027912016958\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 870 loss 0.0002170806546928361\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 871 loss 0.00021665905660483986\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 872 loss 0.00021623677457682788\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 873 loss 0.00021582012414000928\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 874 loss 0.00021540104353334755\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 875 loss 0.00021498437854461372\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 876 loss 0.00021456842659972608\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 877 loss 0.00021415237279143184\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 878 loss 0.00021373858908191323\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 879 loss 0.00021332944743335247\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 880 loss 0.0002129163476638496\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 881 loss 0.00021250726422294974\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 882 loss 0.00021209764236118644\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 883 loss 0.00021169085812289268\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 884 loss 0.00021128398657310754\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 885 loss 0.00021088043285999447\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 886 loss 0.00021047737391199917\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 887 loss 0.00021007739997003227\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 888 loss 0.00020967279851902276\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 889 loss 0.00020927295554429293\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 890 loss 0.00020887705613858998\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 891 loss 0.00020847601990681142\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 892 loss 0.0002080810663755983\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 893 loss 0.00020768778631463647\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 894 loss 0.00020728819072246552\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 895 loss 0.0002069001639029011\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 896 loss 0.000206506738322787\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 897 loss 0.00020611740183085203\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 898 loss 0.00020572793437168002\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 899 loss 0.00020533884526230395\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 900 loss 0.00020495821081567556\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 901 loss 0.00020457578648347408\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 902 loss 0.00020419682550709695\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 903 loss 0.00020381189824547619\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 904 loss 0.0002034348581219092\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 905 loss 0.00020305112411733717\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 906 loss 0.0002026745496550575\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 907 loss 0.0002022940170718357\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 908 loss 0.00020191658404655755\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 909 loss 0.00020154050434939563\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 910 loss 0.00020116358064115047\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 911 loss 0.00020079308887943625\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 912 loss 0.00020041628158651292\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 913 loss 0.00020004533871542662\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 914 loss 0.00019967280968558043\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 915 loss 0.0001993029291043058\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 916 loss 0.0001989342417800799\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 917 loss 0.00019856615108437836\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 918 loss 0.00019819920998997986\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 919 loss 0.00019783206516876817\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 920 loss 0.0001974699553102255\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 921 loss 0.00019710544438567013\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 922 loss 0.00019674078794196248\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 923 loss 0.00019637891091406345\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 924 loss 0.00019601843087002635\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 925 loss 0.0001956605410668999\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "epoch: 926 loss 0.00019529912970028818\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 927 loss 0.00019494007574394345\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 928 loss 0.0001945857802638784\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 929 loss 0.00019422892364673316\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 930 loss 0.00019387746579013765\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 931 loss 0.00019352023082319647\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 932 loss 0.00019316835096105933\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 933 loss 0.00019281767890788615\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 934 loss 0.00019246508600190282\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 935 loss 0.00019211653852835298\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 936 loss 0.0001917653135024011\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 937 loss 0.0001914190361276269\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 938 loss 0.00019107469415757805\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 939 loss 0.000190727470908314\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 940 loss 0.00019038205209653825\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 941 loss 0.0001900384231703356\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 942 loss 0.00018969454686157405\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 943 loss 0.00018935328989755362\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 944 loss 0.00018901228031609207\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 945 loss 0.00018867271137423813\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 946 loss 0.000188331920071505\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 947 loss 0.00018799486861098558\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 948 loss 0.00018765577988233417\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 949 loss 0.00018732095486484468\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 950 loss 0.0001869852130766958\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 951 loss 0.0001866524398792535\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 952 loss 0.00018631716375239193\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 953 loss 0.00018598747556097806\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 954 loss 0.00018565529899206012\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 955 loss 0.0001853236899478361\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 956 loss 0.00018499293946661055\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 957 loss 0.0001846655213739723\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 958 loss 0.00018433691002428532\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 959 loss 0.00018400880799163133\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 960 loss 0.00018368379096500576\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 961 loss 0.00018335900676902384\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 962 loss 0.00018303588149137795\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 963 loss 0.00018271230510435998\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 964 loss 0.00018238896154798567\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 965 loss 0.00018206905224360526\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 966 loss 0.00018174546130467206\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 967 loss 0.00018142462067771703\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 968 loss 0.00018110759265255183\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 969 loss 0.0001807923399610445\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 970 loss 0.00018047232879325747\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 971 loss 0.00018015327805187553\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 972 loss 0.00017983803991228342\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 973 loss 0.0001795265998225659\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 974 loss 0.000179210546775721\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 975 loss 0.00017889722948893905\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 976 loss 0.00017858449427876621\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 977 loss 0.0001782723848009482\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 978 loss 0.0001779628510121256\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 979 loss 0.00017765465599950403\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 980 loss 0.0001773450057953596\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 981 loss 0.00017703778576105833\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 982 loss 0.0001767298235790804\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 983 loss 0.00017642087186686695\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 984 loss 0.00017611791554372758\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 985 loss 0.00017581316933501512\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 986 loss 0.0001755094708641991\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 987 loss 0.00017520553956273943\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 988 loss 0.00017490355821792036\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 989 loss 0.00017460047092754394\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 990 loss 0.00017430156003683805\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 991 loss 0.00017400232900399715\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 992 loss 0.00017370021669194102\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 993 loss 0.00017340491467621177\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 994 loss 0.00017310483963228762\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 995 loss 0.00017281049804296345\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 996 loss 0.00017251276585739106\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 997 loss 0.00017221998132299632\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 998 loss 0.00017192467930726707\n",
      "torch.Size([500, 1]) torch.Size([500, 1])\n",
      "epoch: 999 loss 0.00017163113807328045\n"
     ]
    }
   ],
   "source": [
    "epochs = 1000\n",
    "losses = []\n",
    "for i in range(epochs):\n",
    "  y_pred = model.forward(x_data)\n",
    "  print(y_pred.size(), y_data.size())\n",
    "  loss = criterion(y_pred, y_data)\n",
    "  print(\"epoch:\", i, \"loss\", loss.item())\n",
    "  losses.append(loss.item())\n",
    "  optimizer.zero_grad()\n",
    "  loss.backward()\n",
    "  optimizer.step()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'epoch')"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(range(epochs), losses)\n",
    "plt.ylabel('Loss')\n",
    "plt.xlabel('epoch')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_decision_boundary(X, y):\n",
    "    x_span = np.linspace(min(X[:,0])-0.25, max(X[:,0])+0.25) # Created equally spaced elements in x_span (will use for plotting)\n",
    "    y_span = np.linspace(min(X[:,1])-0.25, max(X[:,1])+0.25) \n",
    "    # Created grid (returns two grid): https://www.geeksforgeeks.org/numpy-meshgrid-function/\n",
    "    xx, yy = np.meshgrid(x_span, y_span)\n",
    "    grid = torch.Tensor(np.c_[xx.ravel(), yy.ravel()])\n",
    "    pred_func = model.forward(grid)\n",
    "    z = pred_func.view(xx.shape).detach().numpy()\n",
    "    plt.contourf(xx, yy, z)\n",
    "    plt.axis('scaled')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_decision_boundary(X, y)\n",
    "scatter_plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Prediction is 1\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAARwAAAD4CAYAAAA+VlnAAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+j8jraAAATfklEQVR4nO3dX4xcZ33G8e+zXpaYeDHrJDjBic3GDS5pLgqyIjBVFf40JLmI+U9AgqRKZKCNuOhNjZCoBBcN3FSCplArRQ1c5A8pBNMaHEKIuEChMRXBdtIFO8GJF5MEsjLrxM3G2V8v5sz67NmZ2V3PmXfOmX0+kuUzM++e89vR7rPv+573nFFEYGaWwlC/CzCzlcOBY2bJOHDMLBkHjpkl48Axs2SG+11AOyNDq2P18Kv7XYaZdfDHl575fUSct9T2lQ2c1cOvZttrP9zvMsysgx9MfuXIctp7SGVmyThwzCwZB46ZJePAMbNkHDhmlowDx8ySceCYWTIOHDNLxoFjZsk4cMwsGQeOmSXjwDGzZBw4ZpaMA8fMknHgmFkyDhwzS8aBY2bJOHDMLBkHjpkl48Axs2QcOGaWjAPHzJJx4JhZMg4cM0vGgWNmyThwzCwZB46ZJePAMbNkHDhmlkwpgSPp65KekXSgzeuS9GVJhyT9UtKbyziumdVLWT2cfweu6vD61cAl2b8dwFdLOq6Z1UgpgRMRPwGe69BkO/CNaHgIeI2kC8o4tpnVR6o5nA3AU7nHR7Pn5pG0Q9I+SftmZk8mKs3MUqnUpHFE7IqIrRGxdWRodb/LMbOSpQqcSeCi3OMLs+fMbAVJFTi7gY9nZ6veAhyPiGOJjm1mFTFcxk4k3QFcAZwr6SjwD8ArACLia8Ae4BrgEPAC8NdlHNfM6qWUwImIjyzyegB/W8axzKy+KjVpbGaDzYFjZsk4cMwsGQeOmSXjwDGzZBw4ZpaMA8fMknHgmFkyDhwzS8aBY2bJOHDMLBkHjpkl48Axs2QcOGaWTCm3pzBbqpOXLbiV9TxTW0bmPR6bmOnYfvUB3ziyThw41nP5kCkGyvTm2ULr2cLrC39ERw83OuZjEzNz+3bw1IMDx3qqGDb5gFkzfpw1Z7DPadZmW43wagaPQ6f6HDjWE62DZpY148cBuHLjxLz2Hxv76ZL2+82pbbAR7ntyy7zgcejUgwPHStNu6DS9ufugadX+PrY09s9apjcPM3p4iKkt44xNzDh4KsqBY11rBk0zZJq9maY148e7Dpqi5tff9+SWuTBr9nimNw8zmgUPeH6nShw4dsZaBU1xXqYYNNB92LTbT7PHA83wGfHEcsU4cGzZWs3PrBk/zvt6GC7t5Pf/sbGfNuZ4oDHPM76FJzevZfTwkOd4KsKBY0vWao6mGTbNnkyvA2YxrXo9xd6OQ6d/HDi2qOLQCU6vn6lS2BTNzfO0CB3wEKsfHDjW0cnLNhQmgxuqHDR5rUIHvHanXxw41lYzbPK9GZg/EVzlsGlaGDrgIVZ/OHCspXzYFIOmDiFTlA8dWHgWy6GThgPH5uTnavIrg+vWo2mnuGjQoZOeb09hnLxsw4KwWTN+fKDCpqn5PVy5cWKu51a8oNR6p5TAkXSVpAlJhyTtbPH6DZKelfSL7N9NZRzXulMMmnzYXLlxYuDCpikfOguvVrde6npIJWkVcCvwV8BR4GFJuyPi0ULTuyLi5m6PZ+VYbPHeIAVMK81FgmvGj+fOXnlY1WtlzOFcDhyKiMcBJN0JbAeKgWMVsNjivUEPmrzi2auxhQulrWRlDKk2AE/lHh/Nnit6v6RfSrpH0kUlHNeWqdXwaaWGTStTW0YWvSOhdSfVWarvAXdExIuSPgHcDryj2EjSDmAHwFmrRhOVtjK0WsCXP929ksPmyo0ThYWBHlr1ShmBMwnkeywXZs/NiYg/5B7eBnyp1Y4iYhewC2DtyPoooTZj8QV8Kzls8hd8ej6n98oInIeBSySN0wia64CP5htIuiAijmUPrwUeK+G4tgSDtoCvV67cODF3bx2HTu90PYcTEaeAm4G9NILk7og4KOnzkq7Nmn1a0kFJjwCfBm7o9rjW2cnLNvDb9487bJYgf5rcequUOZyI2APsKTz3udz2Z4DPlHEsW1yxV7OG+l3/lNq8e+lYz3il8YDpNIQCh00nHxv7qXs5PeZrqQZIMWwcNGemOY/jdTnlcw9nQDhsrA4cOAPAYdMbvqizfA6cmms1Z9PksLGqceDUmNfY9J4vdSiXA6em8pcqNDlsuuf3rrccODVXtY9pMevEgTNAHDZWdQ6cAbDSr/YuW3Hxn+dxyuPAqbnimSkrx5rx40xvnp2bJ3PolMOBY2bJOHDMLBkHTo35EwfS8LCqPA4csw4c6uVy4Ji1kZ+Q93VV5XDgmLXg++L0hgPHbBEeVpXHgWPWgdc5lcuBY7YMPlPVHQdOTXkSs3f8KQ6943sa15D/yqaV/6yqsYmZfpdTa+7h1IzDJi33csrlwKkRh43VnQOnhoo3TPetKawuHDg10ezdOGzS8PvaG540rpm6hs3IkVOct2uasXtfYOj5YPZsMfWeV/HsjlFmNvnHcKVwD6dGpraM1DJsRn98kje8+2nW3fk8q04EClh1Ilh35/O84d1PM/rjk/0u0RJx4NRA/uNg6hY2I0dOselTz7HqZDB0av5rQ6dg1clg06eeY+TIqdY76KM6vc91UUrgSLpK0oSkQ5J2tnj9lZLuyl7/maTXl3HclaDTB93VwXm7ptFL0bGNXgrOu206UUXWT10HjqRVwK3A1cClwEckXVpodiMwFRF/AvwT8MVuj7sSdPoI37oYu/eFBT2boqFTMPbtF9IUdAbq+L5XVRk9nMuBQxHxeETMAHcC2wtttgO3Z9v3AO+UpBKOPbDyH3SX79nUrZs/9Hzn3s1y26VWt/e76soInA3AU7nHR7PnWraJiFPAceCcEo49kPI9m+atEer6V3b27KX9XVlqO6u3Sk0aS9ohaZ+kfTOzPnMB1P5TNafe8ypmFznrPTsMU+97VZqCrK/KCJxJ4KLc4wuz51q2kTQMrAX+UNxRROyKiK0RsXVkaHUJpdVPu3mbOoYNwLM7RolXdO69xCvEszeNJqrozNRxwr6Kygich4FLJI1LGgGuA3YX2uwGrs+2PwA8EBHVHLRXUF3DBmBm0zBHvrqOl1drQU9ndhheXi2OfHWdF/+tEF0HTjYnczOwF3gMuDsiDkr6vKRrs2b/Bpwj6RDwd8CCU+fW+uLMOodN0/TbV/Orvet57qNn8/IaEYKX14jnPno2v9q7num3r8ze7EpUyp+ViNgD7Ck897nc9v8BHyzjWIOq3bVSg2Jm0zCTXxhj8gtj/S7F+qhSk8Yr3aCGjVmTA6cC8kOpQZgoNmvHgVMRzd6N2SBz4PTZoJ0GN+vEgWNmyThwzCwZB06F+MyUDToHjpkl48CpCF+rYyuBA6ciTjyxlvue3NLvMsx6yoFTUd+c2tbvEsxK58CpgLGJGUYPD7mXYwPPgdNnqw+cvnVQMXTcy7FB48CpiLGJmblt93RsUPmuRxWw+sAkJy/bkIVO48bp06zlPhqh48scbFC4h1MRzaFVs6eTH155aNV/J55Y2+8SBoJ7OBWS7+lMbRlh9PDQvJ4OuLdj9ebAqZj5wyuAEYdOH7l3WS4HTgU1h1fz5nUOjzG9eXZe8DQ5gHqjGTaewC+PA6fCltLbyXPwWNU5cCquGTrQmFAem4CpLWPz7g5YDCAHT/fyvZsTT6xl9PAQYxMz89ZN2fL5LFUNrD4wOfcPGsGzcc8pRg8PzZ3N8oLB8nkoVT73cGpm/l/Y5s3XF67daXJvZ/m+ObVtLmyavRsrhwOnxlqdRi9OLjd/cdrd3MuB1F4+bDycKocDp+YWTixD/qxW07efuBw4fd+dZgA1h18OnoZWw1GHTXkcOAMg/8uQD5/GBPPIvLbTNFbMNntA+eBx6DR4orh3HDgDpvjLsfpA4//TZ7qyz8BiLWvGj88bcq300Cn2bhw25XPgrBALJ5vnD7vare3JG7QwKgZMfqJ4455TDpsecOCsQI1fpIVnuKAx17PU+yvXNYBarSDOX5zpsOkdB84K1WoV85zDY3Ob7RYYNodgUJ/gKS7mA+bOQo1mbTyM6i1FxJl/sbQOuAt4PfAb4EMRMdWi3cvA/uzhkxFx7WL7XjuyPra99sNnXJstX3Oep5Xm5HPzI4mb6vLRxO1WDhc5bJbnB5Nf+XlEbF1q+257ODuBH0XELZJ2Zo//vkW7kxHx510ey3qs8y/b6SFYc/jVnHTO93agOuHTao7GZ5/6q9vA2Q5ckW3fDjxI68Cxmjv9y5lNONN+3qd4ScBinyh6pgHV7hKOdpckNMPmdf/xxBkdz7rXbeCsj4hj2fbvgPVt2p0laR9wCrglIu5t1UjSDmAHwFmrRls1sT5rNeEMLFhomFdcdNhUXHx4Jorh0urOfM15mo3u1fTdooEj6X7g/BYvfTb/ICJCUrsJoU0RMSnpYuABSfsj4nCxUUTsAnZBYw5n0eqtL+b3dk4bK3Rk2i06BOatAYIz+1z1/GnspvylCO3rtn5ZNHAi4l3tXpP0tKQLIuKYpAuAZ9rsYzL7/3FJDwJvAhYEjtXLYr/A7RYdQvvwWUrwFIOmGDIOlurqdki1G7geuCX7/7vFBpLGgBci4kVJ5wJvA77U5XGtRhbewRDyw7F8+CxlHVAxaMCns+ui28C5Bbhb0o3AEeBDAJK2Ap+MiJuANwL/KmmWxv13bomIR7s8rtVQ8WZiwOmr3HPyAVQ0enho3pqZ/L6t+rpah9NLXocz+Dqt+4FGGLWaiyly2PRP6nU4ZmdsqXNANjh8KzMzS8aBY2bJOHDMLBkHjpkl48Axs2QcOGaWjAPHzJJx4JhZMg4cM0vGgWNmyThwzCwZB46ZJePAMbNkHDhmlowDx8ySceCYWTIOHDNLxoFjZsk4cMwsGQeOmSXjwDGzZBw4ZpaMA8fMknHgmFkyDhwzS8aBY2bJOHDMLBkHjpkl01XgSPqgpIOSZiVt7dDuKkkTkg5J2tnNMc2svrrt4RwA3gf8pF0DSauAW4GrgUuBj0i6tMvjmlkNDXfzxRHxGICkTs0uBw5FxONZ2zuB7cCj3RzbzOonxRzOBuCp3OOj2XMLSNohaZ+kfTOzJxOUZmYpLdrDkXQ/cH6Llz4bEd8ts5iI2AXsAlg7sj7K3LeZ9d+igRMR7+ryGJPARbnHF2bPmdkKk2JI9TBwiaRxSSPAdcDuBMc1s4rp9rT4eyUdBd4K/Jekvdnzr5O0ByAiTgE3A3uBx4C7I+Jgd2WbWR11e5bqO8B3Wjz/W+Ca3OM9wJ5ujmVm9eeVxmaWjAPHzJJx4JhZMg4cM0vGgWNmyThwzCwZB46ZJePAMbNkHDhmlowDx8ySceCYWTIOHDNLxoFjZsk4cMwsGQeOmSXjwDGzZBw4ZpaMA8fMknHgmFkyDhwzS8aBY2bJOHDMLBkHjpkl48Axs2QcOGaWjCKi3zW0JOlZ4Ej28Fzg930spx3XtTyua3nqUNemiDhvqV9Y2cDJk7QvIrb2u44i17U8rmt5BrEuD6nMLBkHjpklU5fA2dXvAtpwXcvjupZn4OqqxRyOmQ2GuvRwzGwAOHDMLJlKBo6kD0o6KGlWUtvTb5J+I2m/pF9I2lehuq6SNCHpkKSdCepaJ+mHkn6d/T/Wpt3L2Xv1C0m7e1hPx+9f0isl3ZW9/jNJr+9VLcus6wZJz+beo5sS1fV1Sc9IOtDmdUn6clb3LyW9uSJ1XSHpeO79+tyiO42Iyv0D3ghsAR4EtnZo9xvg3CrVBawCDgMXAyPAI8ClPa7rS8DObHsn8MU27U4keI8W/f6BvwG+lm1fB9xVkbpuAP451c9T7rh/CbwZONDm9WuA7wMC3gL8rCJ1XQH853L2WckeTkQ8FhET/a6jaIl1XQ4ciojHI2IGuBPY3uPStgO3Z9u3A+/p8fE6Wcr3n6/3HuCdklSBuvoiIn4CPNehyXbgG9HwEPAaSRdUoK5lq2TgLEMA90n6uaQd/S4mswF4Kvf4aPZcL62PiGPZ9u+A9W3anSVpn6SHJPUqlJby/c+1iYhTwHHgnB7Vs5y6AN6fDVvukXRRj2taqn78TC3VWyU9Iun7kv5sscbDKSpqRdL9wPktXvpsRHx3ibv5i4iYlPRa4IeS/jdL5X7XVbpOdeUfRERIarfWYVP2fl0MPCBpf0QcLrvWGvsecEdEvCjpEzR6Ye/oc01V9j80fqZOSLoGuBe4pNMX9C1wIuJdJexjMvv/GUnfodFt7ipwSqhrEsj/Zbwwe64rneqS9LSkCyLiWNbVfqbNPprv1+OSHgTeRGNeo0xL+f6bbY5KGgbWAn8ouY5l1xUR+RpuozE3VgU9+ZnqVkT8Mbe9R9K/SDo3ItpecFrbIZWksyWNNreBK4GWs+mJPQxcImlc0giNSdGenRHK7Aauz7avBxb0xCSNSXpltn0u8Dbg0R7UspTvP1/vB4AHIpuF7KFF6yrMi1wLPNbjmpZqN/Dx7GzVW4DjuSF030g6vzn3JulyGnnS+Q9H6hn5Jc6Ov5fGOPVF4Glgb/b864A92fbFNM40PAIcpDHk6Xtdcfqswq9o9B5S1HUO8CPg18D9wLrs+a3Abdn2NmB/9n7tB27sYT0Lvn/g88C12fZZwLeAQ8B/Axcn+rlarK5/zH6WHgF+DPxporruAI4BL2U/XzcCnwQ+mb0u4Nas7v10OHObuK6bc+/XQ8C2xfbpSxvMLJnaDqnMrH4cOGaWjAPHzJJx4JhZMg4cM0vGgWNmyThwzCyZ/wdCJ5SzG+5HSgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "x = 0.025\n",
    "y = 0.025\n",
    "point = torch.Tensor([x, y])\n",
    "prediction = model.predict(point)\n",
    "plt.plot([x], [y], marker='o', markersize=10, color=\"red\")\n",
    "print(\"Prediction is\", prediction)\n",
    "plot_decision_boundary(X, y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "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": "3.8.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}