chapter 5 Mechanisms of learning.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "d0d75a39",
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "7b045d5e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(tensor([ 0.5000, 14.0000, 15.0000, 28.0000, 11.0000,  8.0000,  3.0000, -4.0000,\n",
       "          6.0000, 13.0000, 21.0000]),\n",
       " tensor([35.7000, 55.9000, 58.2000, 81.9000, 56.3000, 48.9000, 33.9000, 21.8000,\n",
       "         48.4000, 60.4000, 68.4000]))"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "t_c = [0.5,  14.0, 15.0, 28.0, 11.0,  8.0,  3.0, -4.0,  6.0, 13.0, 21.0]\n",
    "t_u = [35.7, 55.9, 58.2, 81.9, 56.3, 48.9, 33.9, 21.8, 48.4, 60.4, 68.4]\n",
    "t_c = torch.tensor(t_c)\n",
    "t_u = torch.tensor(t_u)\n",
    "t_c, t_u"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "56f2695d",
   "metadata": {},
   "outputs": [],
   "source": [
    "def loss_fn(t_p, t_c):\n",
    "    squared_diffs = (t_p - t_c)**2\n",
    "    return squared_diffs.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "7e800430",
   "metadata": {},
   "outputs": [],
   "source": [
    "def model(t_u, w, b):\n",
    "    return w * t_u + b"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "44d496d6",
   "metadata": {},
   "outputs": [],
   "source": [
    "def dloss_fn(t_p, t_c):\n",
    "    dsq_diffs = 2 * (t_p - t_c) / t_p.size(0) # 均值的导数\n",
    "    return dsq_diffs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "d4d34cbf",
   "metadata": {},
   "outputs": [],
   "source": [
    "def dmodel_dw(t_u, w, b):\n",
    "    return t_u"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "eda395e9",
   "metadata": {},
   "outputs": [],
   "source": [
    "def dmodel_db(t_u, w, b):\n",
    "    return 1.0"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5492a7b7",
   "metadata": {},
   "source": [
    "定义梯度函数"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "d39be690",
   "metadata": {},
   "outputs": [],
   "source": [
    "def grad_fn(t_u, t_c, t_p, w, b):\n",
    "    dloss_dtp = dloss_fn(t_p, t_c)\n",
    "    dloss_dw = dloss_dtp * dmodel_dw(t_u, w, b)\n",
    "    dloss_db = dloss_dtp * dmodel_db(t_u, w, b)\n",
    "    return torch.stack([dloss_dw.sum(), dloss_db.sum()])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "c3c4cf3f",
   "metadata": {},
   "outputs": [],
   "source": [
    "def training_loop(n_epochs, learning_rate, params, t_u, t_c):\n",
    "    for epoch in range(1, n_epochs + 1):\n",
    "        w, b = params\n",
    "        t_p = model(t_u, w, b) # 正向\n",
    "        loss = loss_fn(t_p, t_c)\n",
    "        grad = grad_fn(t_u, t_c, t_p, w, b) # 反向\n",
    "        params = params - learning_rate * grad\n",
    "        print('Epoch %d, Loss %f' % (epoch, float(loss)))\n",
    "    return params"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "22df4a75",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1, Loss 1763.884766\n",
      "Epoch 2, Loss 5802484.500000\n",
      "Epoch 3, Loss 19408029696.000000\n",
      "Epoch 4, Loss 64915905708032.000000\n",
      "Epoch 5, Loss 217130525461053440.000000\n",
      "Epoch 6, Loss 726257583152928129024.000000\n",
      "Epoch 7, Loss 2429183416467662896627712.000000\n",
      "Epoch 8, Loss 8125122549611731432050262016.000000\n",
      "Epoch 9, Loss 27176882120842590626938030653440.000000\n",
      "Epoch 10, Loss 90901105189019073810297959556841472.000000\n",
      "Epoch 11, Loss inf\n",
      "Epoch 12, Loss inf\n",
      "Epoch 13, Loss inf\n",
      "Epoch 14, Loss inf\n",
      "Epoch 15, Loss inf\n",
      "Epoch 16, Loss inf\n",
      "Epoch 17, Loss inf\n",
      "Epoch 18, Loss inf\n",
      "Epoch 19, Loss inf\n",
      "Epoch 20, Loss inf\n",
      "Epoch 21, Loss inf\n",
      "Epoch 22, Loss inf\n",
      "Epoch 23, Loss nan\n",
      "Epoch 24, Loss nan\n",
      "Epoch 25, Loss nan\n",
      "Epoch 26, Loss nan\n",
      "Epoch 27, Loss nan\n",
      "Epoch 28, Loss nan\n",
      "Epoch 29, Loss nan\n",
      "Epoch 30, Loss nan\n",
      "Epoch 31, Loss nan\n",
      "Epoch 32, Loss nan\n",
      "Epoch 33, Loss nan\n",
      "Epoch 34, Loss nan\n",
      "Epoch 35, Loss nan\n",
      "Epoch 36, Loss nan\n",
      "Epoch 37, Loss nan\n",
      "Epoch 38, Loss nan\n",
      "Epoch 39, Loss nan\n",
      "Epoch 40, Loss nan\n",
      "Epoch 41, Loss nan\n",
      "Epoch 42, Loss nan\n",
      "Epoch 43, Loss nan\n",
      "Epoch 44, Loss nan\n",
      "Epoch 45, Loss nan\n",
      "Epoch 46, Loss nan\n",
      "Epoch 47, Loss nan\n",
      "Epoch 48, Loss nan\n",
      "Epoch 49, Loss nan\n",
      "Epoch 50, Loss nan\n",
      "Epoch 51, Loss nan\n",
      "Epoch 52, Loss nan\n",
      "Epoch 53, Loss nan\n",
      "Epoch 54, Loss nan\n",
      "Epoch 55, Loss nan\n",
      "Epoch 56, Loss nan\n",
      "Epoch 57, Loss nan\n",
      "Epoch 58, Loss nan\n",
      "Epoch 59, Loss nan\n",
      "Epoch 60, Loss nan\n",
      "Epoch 61, Loss nan\n",
      "Epoch 62, Loss nan\n",
      "Epoch 63, Loss nan\n",
      "Epoch 64, Loss nan\n",
      "Epoch 65, Loss nan\n",
      "Epoch 66, Loss nan\n",
      "Epoch 67, Loss nan\n",
      "Epoch 68, Loss nan\n",
      "Epoch 69, Loss nan\n",
      "Epoch 70, Loss nan\n",
      "Epoch 71, Loss nan\n",
      "Epoch 72, Loss nan\n",
      "Epoch 73, Loss nan\n",
      "Epoch 74, Loss nan\n",
      "Epoch 75, Loss nan\n",
      "Epoch 76, Loss nan\n",
      "Epoch 77, Loss nan\n",
      "Epoch 78, Loss nan\n",
      "Epoch 79, Loss nan\n",
      "Epoch 80, Loss nan\n",
      "Epoch 81, Loss nan\n",
      "Epoch 82, Loss nan\n",
      "Epoch 83, Loss nan\n",
      "Epoch 84, Loss nan\n",
      "Epoch 85, Loss nan\n",
      "Epoch 86, Loss nan\n",
      "Epoch 87, Loss nan\n",
      "Epoch 88, Loss nan\n",
      "Epoch 89, Loss nan\n",
      "Epoch 90, Loss nan\n",
      "Epoch 91, Loss nan\n",
      "Epoch 92, Loss nan\n",
      "Epoch 93, Loss nan\n",
      "Epoch 94, Loss nan\n",
      "Epoch 95, Loss nan\n",
      "Epoch 96, Loss nan\n",
      "Epoch 97, Loss nan\n",
      "Epoch 98, Loss nan\n",
      "Epoch 99, Loss nan\n",
      "Epoch 100, Loss nan\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "tensor([nan, nan])"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "training_loop(n_epochs = 100, learning_rate = 1e-2, params = torch.tensor([1.0, 0.0]), t_u = t_u, t_c= t_c)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "c9dd6b63",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1, Loss 1763.884766\n",
      "Epoch 2, Loss 323.090515\n",
      "Epoch 3, Loss 78.929634\n",
      "Epoch 4, Loss 37.552845\n",
      "Epoch 5, Loss 30.540283\n",
      "Epoch 6, Loss 29.351154\n",
      "Epoch 7, Loss 29.148884\n",
      "Epoch 8, Loss 29.113848\n",
      "Epoch 9, Loss 29.107145\n",
      "Epoch 10, Loss 29.105247\n",
      "Epoch 11, Loss 29.104168\n",
      "Epoch 12, Loss 29.103222\n",
      "Epoch 13, Loss 29.102295\n",
      "Epoch 14, Loss 29.101379\n",
      "Epoch 15, Loss 29.100466\n",
      "Epoch 16, Loss 29.099548\n",
      "Epoch 17, Loss 29.098631\n",
      "Epoch 18, Loss 29.097717\n",
      "Epoch 19, Loss 29.096796\n",
      "Epoch 20, Loss 29.095881\n",
      "Epoch 21, Loss 29.094959\n",
      "Epoch 22, Loss 29.094049\n",
      "Epoch 23, Loss 29.093134\n",
      "Epoch 24, Loss 29.092216\n",
      "Epoch 25, Loss 29.091301\n",
      "Epoch 26, Loss 29.090385\n",
      "Epoch 27, Loss 29.089464\n",
      "Epoch 28, Loss 29.088551\n",
      "Epoch 29, Loss 29.087635\n",
      "Epoch 30, Loss 29.086714\n",
      "Epoch 31, Loss 29.085804\n",
      "Epoch 32, Loss 29.084888\n",
      "Epoch 33, Loss 29.083967\n",
      "Epoch 34, Loss 29.083057\n",
      "Epoch 35, Loss 29.082142\n",
      "Epoch 36, Loss 29.081221\n",
      "Epoch 37, Loss 29.080309\n",
      "Epoch 38, Loss 29.079390\n",
      "Epoch 39, Loss 29.078474\n",
      "Epoch 40, Loss 29.077562\n",
      "Epoch 41, Loss 29.076649\n",
      "Epoch 42, Loss 29.075731\n",
      "Epoch 43, Loss 29.074812\n",
      "Epoch 44, Loss 29.073895\n",
      "Epoch 45, Loss 29.072981\n",
      "Epoch 46, Loss 29.072069\n",
      "Epoch 47, Loss 29.071148\n",
      "Epoch 48, Loss 29.070234\n",
      "Epoch 49, Loss 29.069323\n",
      "Epoch 50, Loss 29.068401\n",
      "Epoch 51, Loss 29.067486\n",
      "Epoch 52, Loss 29.066566\n",
      "Epoch 53, Loss 29.065657\n",
      "Epoch 54, Loss 29.064741\n",
      "Epoch 55, Loss 29.063826\n",
      "Epoch 56, Loss 29.062910\n",
      "Epoch 57, Loss 29.061995\n",
      "Epoch 58, Loss 29.061079\n",
      "Epoch 59, Loss 29.060169\n",
      "Epoch 60, Loss 29.059248\n",
      "Epoch 61, Loss 29.058336\n",
      "Epoch 62, Loss 29.057415\n",
      "Epoch 63, Loss 29.056507\n",
      "Epoch 64, Loss 29.055586\n",
      "Epoch 65, Loss 29.054674\n",
      "Epoch 66, Loss 29.053761\n",
      "Epoch 67, Loss 29.052843\n",
      "Epoch 68, Loss 29.051929\n",
      "Epoch 69, Loss 29.051012\n",
      "Epoch 70, Loss 29.050098\n",
      "Epoch 71, Loss 29.049183\n",
      "Epoch 72, Loss 29.048273\n",
      "Epoch 73, Loss 29.047350\n",
      "Epoch 74, Loss 29.046442\n",
      "Epoch 75, Loss 29.045530\n",
      "Epoch 76, Loss 29.044611\n",
      "Epoch 77, Loss 29.043699\n",
      "Epoch 78, Loss 29.042784\n",
      "Epoch 79, Loss 29.041870\n",
      "Epoch 80, Loss 29.040955\n",
      "Epoch 81, Loss 29.040039\n",
      "Epoch 82, Loss 29.039122\n",
      "Epoch 83, Loss 29.038210\n",
      "Epoch 84, Loss 29.037294\n",
      "Epoch 85, Loss 29.036379\n",
      "Epoch 86, Loss 29.035463\n",
      "Epoch 87, Loss 29.034554\n",
      "Epoch 88, Loss 29.033636\n",
      "Epoch 89, Loss 29.032722\n",
      "Epoch 90, Loss 29.031811\n",
      "Epoch 91, Loss 29.030895\n",
      "Epoch 92, Loss 29.029976\n",
      "Epoch 93, Loss 29.029066\n",
      "Epoch 94, Loss 29.028151\n",
      "Epoch 95, Loss 29.027235\n",
      "Epoch 96, Loss 29.026323\n",
      "Epoch 97, Loss 29.025410\n",
      "Epoch 98, Loss 29.024492\n",
      "Epoch 99, Loss 29.023582\n",
      "Epoch 100, Loss 29.022667\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "tensor([ 0.2327, -0.0438])"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "training_loop(n_epochs = 100, learning_rate = 1e-4, params = torch.tensor([1.0, 0.0]), t_u = t_u, t_c= t_c)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "62a545e5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([3.5700, 5.5900, 5.8200, 8.1900, 5.6300, 4.8900, 3.3900, 2.1800, 4.8400,\n",
       "        6.0400, 6.8400])"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "t_un = 0.1 * t_u\n",
    "t_un"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "617f7c1b",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1, Loss 80.364342\n",
      "Epoch 2, Loss 79.752480\n",
      "Epoch 3, Loss 79.148026\n",
      "Epoch 4, Loss 78.550865\n",
      "Epoch 5, Loss 77.960899\n",
      "Epoch 6, Loss 77.378052\n",
      "Epoch 7, Loss 76.802254\n",
      "Epoch 8, Loss 76.233406\n",
      "Epoch 9, Loss 75.671425\n",
      "Epoch 10, Loss 75.116219\n",
      "Epoch 11, Loss 74.567734\n",
      "Epoch 12, Loss 74.025856\n",
      "Epoch 13, Loss 73.490532\n",
      "Epoch 14, Loss 72.961662\n",
      "Epoch 15, Loss 72.439178\n",
      "Epoch 16, Loss 71.923004\n",
      "Epoch 17, Loss 71.413055\n",
      "Epoch 18, Loss 70.909271\n",
      "Epoch 19, Loss 70.411568\n",
      "Epoch 20, Loss 69.919861\n",
      "Epoch 21, Loss 69.434097\n",
      "Epoch 22, Loss 68.954185\n",
      "Epoch 23, Loss 68.480080\n",
      "Epoch 24, Loss 68.011688\n",
      "Epoch 25, Loss 67.548958\n",
      "Epoch 26, Loss 67.091805\n",
      "Epoch 27, Loss 66.640167\n",
      "Epoch 28, Loss 66.193993\n",
      "Epoch 29, Loss 65.753181\n",
      "Epoch 30, Loss 65.317703\n",
      "Epoch 31, Loss 64.887482\n",
      "Epoch 32, Loss 64.462448\n",
      "Epoch 33, Loss 64.042534\n",
      "Epoch 34, Loss 63.627697\n",
      "Epoch 35, Loss 63.217857\n",
      "Epoch 36, Loss 62.812977\n",
      "Epoch 37, Loss 62.412975\n",
      "Epoch 38, Loss 62.017799\n",
      "Epoch 39, Loss 61.627377\n",
      "Epoch 40, Loss 61.241684\n",
      "Epoch 41, Loss 60.860641\n",
      "Epoch 42, Loss 60.484188\n",
      "Epoch 43, Loss 60.112278\n",
      "Epoch 44, Loss 59.744869\n",
      "Epoch 45, Loss 59.381863\n",
      "Epoch 46, Loss 59.023254\n",
      "Epoch 47, Loss 58.668968\n",
      "Epoch 48, Loss 58.318947\n",
      "Epoch 49, Loss 57.973156\n",
      "Epoch 50, Loss 57.631542\n",
      "Epoch 51, Loss 57.294033\n",
      "Epoch 52, Loss 56.960609\n",
      "Epoch 53, Loss 56.631180\n",
      "Epoch 54, Loss 56.305759\n",
      "Epoch 55, Loss 55.984241\n",
      "Epoch 56, Loss 55.666615\n",
      "Epoch 57, Loss 55.352806\n",
      "Epoch 58, Loss 55.042782\n",
      "Epoch 59, Loss 54.736511\n",
      "Epoch 60, Loss 54.433910\n",
      "Epoch 61, Loss 54.134972\n",
      "Epoch 62, Loss 53.839622\n",
      "Epoch 63, Loss 53.547840\n",
      "Epoch 64, Loss 53.259579\n",
      "Epoch 65, Loss 52.974792\n",
      "Epoch 66, Loss 52.693436\n",
      "Epoch 67, Loss 52.415470\n",
      "Epoch 68, Loss 52.140854\n",
      "Epoch 69, Loss 51.869553\n",
      "Epoch 70, Loss 51.601517\n",
      "Epoch 71, Loss 51.336720\n",
      "Epoch 72, Loss 51.075100\n",
      "Epoch 73, Loss 50.816647\n",
      "Epoch 74, Loss 50.561291\n",
      "Epoch 75, Loss 50.309025\n",
      "Epoch 76, Loss 50.059799\n",
      "Epoch 77, Loss 49.813564\n",
      "Epoch 78, Loss 49.570301\n",
      "Epoch 79, Loss 49.329971\n",
      "Epoch 80, Loss 49.092545\n",
      "Epoch 81, Loss 48.857971\n",
      "Epoch 82, Loss 48.626225\n",
      "Epoch 83, Loss 48.397266\n",
      "Epoch 84, Loss 48.171066\n",
      "Epoch 85, Loss 47.947586\n",
      "Epoch 86, Loss 47.726810\n",
      "Epoch 87, Loss 47.508690\n",
      "Epoch 88, Loss 47.293186\n",
      "Epoch 89, Loss 47.080284\n",
      "Epoch 90, Loss 46.869946\n",
      "Epoch 91, Loss 46.662144\n",
      "Epoch 92, Loss 46.456837\n",
      "Epoch 93, Loss 46.254002\n",
      "Epoch 94, Loss 46.053604\n",
      "Epoch 95, Loss 45.855618\n",
      "Epoch 96, Loss 45.660030\n",
      "Epoch 97, Loss 45.466782\n",
      "Epoch 98, Loss 45.275871\n",
      "Epoch 99, Loss 45.087246\n",
      "Epoch 100, Loss 44.900906\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "tensor([1.5845, 0.0726])"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "training_loop(n_epochs = 100, learning_rate = 1e-4, params = torch.tensor([1.0, 0.0]), t_u = t_un, t_c= t_c)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "54be6088",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1, Loss 80.364342\n",
      "Epoch 2, Loss 79.752480\n",
      "Epoch 3, Loss 79.148026\n",
      "Epoch 4, Loss 78.550865\n",
      "Epoch 5, Loss 77.960899\n",
      "Epoch 6, Loss 77.378052\n",
      "Epoch 7, Loss 76.802254\n",
      "Epoch 8, Loss 76.233406\n",
      "Epoch 9, Loss 75.671425\n",
      "Epoch 10, Loss 75.116219\n",
      "Epoch 11, Loss 74.567734\n",
      "Epoch 12, Loss 74.025856\n",
      "Epoch 13, Loss 73.490532\n",
      "Epoch 14, Loss 72.961662\n",
      "Epoch 15, Loss 72.439178\n",
      "Epoch 16, Loss 71.923004\n",
      "Epoch 17, Loss 71.413055\n",
      "Epoch 18, Loss 70.909271\n",
      "Epoch 19, Loss 70.411568\n",
      "Epoch 20, Loss 69.919861\n",
      "Epoch 21, Loss 69.434097\n",
      "Epoch 22, Loss 68.954185\n",
      "Epoch 23, Loss 68.480080\n",
      "Epoch 24, Loss 68.011688\n",
      "Epoch 25, Loss 67.548958\n",
      "Epoch 26, Loss 67.091805\n",
      "Epoch 27, Loss 66.640167\n",
      "Epoch 28, Loss 66.193993\n",
      "Epoch 29, Loss 65.753181\n",
      "Epoch 30, Loss 65.317703\n",
      "Epoch 31, Loss 64.887482\n",
      "Epoch 32, Loss 64.462448\n",
      "Epoch 33, Loss 64.042534\n",
      "Epoch 34, Loss 63.627697\n",
      "Epoch 35, Loss 63.217857\n",
      "Epoch 36, Loss 62.812977\n",
      "Epoch 37, Loss 62.412975\n",
      "Epoch 38, Loss 62.017799\n",
      "Epoch 39, Loss 61.627377\n",
      "Epoch 40, Loss 61.241684\n",
      "Epoch 41, Loss 60.860641\n",
      "Epoch 42, Loss 60.484188\n",
      "Epoch 43, Loss 60.112278\n",
      "Epoch 44, Loss 59.744869\n",
      "Epoch 45, Loss 59.381863\n",
      "Epoch 46, Loss 59.023254\n",
      "Epoch 47, Loss 58.668968\n",
      "Epoch 48, Loss 58.318947\n",
      "Epoch 49, Loss 57.973156\n",
      "Epoch 50, Loss 57.631542\n",
      "Epoch 51, Loss 57.294033\n",
      "Epoch 52, Loss 56.960609\n",
      "Epoch 53, Loss 56.631180\n",
      "Epoch 54, Loss 56.305759\n",
      "Epoch 55, Loss 55.984241\n",
      "Epoch 56, Loss 55.666615\n",
      "Epoch 57, Loss 55.352806\n",
      "Epoch 58, Loss 55.042782\n",
      "Epoch 59, Loss 54.736511\n",
      "Epoch 60, Loss 54.433910\n",
      "Epoch 61, Loss 54.134972\n",
      "Epoch 62, Loss 53.839622\n",
      "Epoch 63, Loss 53.547840\n",
      "Epoch 64, Loss 53.259579\n",
      "Epoch 65, Loss 52.974792\n",
      "Epoch 66, Loss 52.693436\n",
      "Epoch 67, Loss 52.415470\n",
      "Epoch 68, Loss 52.140854\n",
      "Epoch 69, Loss 51.869553\n",
      "Epoch 70, Loss 51.601517\n",
      "Epoch 71, Loss 51.336720\n",
      "Epoch 72, Loss 51.075100\n",
      "Epoch 73, Loss 50.816647\n",
      "Epoch 74, Loss 50.561291\n",
      "Epoch 75, Loss 50.309025\n",
      "Epoch 76, Loss 50.059799\n",
      "Epoch 77, Loss 49.813564\n",
      "Epoch 78, Loss 49.570301\n",
      "Epoch 79, Loss 49.329971\n",
      "Epoch 80, Loss 49.092545\n",
      "Epoch 81, Loss 48.857971\n",
      "Epoch 82, Loss 48.626225\n",
      "Epoch 83, Loss 48.397266\n",
      "Epoch 84, Loss 48.171066\n",
      "Epoch 85, Loss 47.947586\n",
      "Epoch 86, Loss 47.726810\n",
      "Epoch 87, Loss 47.508690\n",
      "Epoch 88, Loss 47.293186\n",
      "Epoch 89, Loss 47.080284\n",
      "Epoch 90, Loss 46.869946\n",
      "Epoch 91, Loss 46.662144\n",
      "Epoch 92, Loss 46.456837\n",
      "Epoch 93, Loss 46.254002\n",
      "Epoch 94, Loss 46.053604\n",
      "Epoch 95, Loss 45.855618\n",
      "Epoch 96, Loss 45.660030\n",
      "Epoch 97, Loss 45.466782\n",
      "Epoch 98, Loss 45.275871\n",
      "Epoch 99, Loss 45.087246\n",
      "Epoch 100, Loss 44.900906\n",
      "Epoch 101, Loss 44.716797\n",
      "Epoch 102, Loss 44.534904\n",
      "Epoch 103, Loss 44.355198\n",
      "Epoch 104, Loss 44.177654\n",
      "Epoch 105, Loss 44.002254\n",
      "Epoch 106, Loss 43.828960\n",
      "Epoch 107, Loss 43.657749\n",
      "Epoch 108, Loss 43.488598\n",
      "Epoch 109, Loss 43.321487\n",
      "Epoch 110, Loss 43.156376\n",
      "Epoch 111, Loss 42.993252\n",
      "Epoch 112, Loss 42.832096\n",
      "Epoch 113, Loss 42.672878\n",
      "Epoch 114, Loss 42.515564\n",
      "Epoch 115, Loss 42.360153\n",
      "Epoch 116, Loss 42.206604\n",
      "Epoch 117, Loss 42.054901\n",
      "Epoch 118, Loss 41.905025\n",
      "Epoch 119, Loss 41.756943\n",
      "Epoch 120, Loss 41.610645\n",
      "Epoch 121, Loss 41.466106\n",
      "Epoch 122, Loss 41.323307\n",
      "Epoch 123, Loss 41.182213\n",
      "Epoch 124, Loss 41.042820\n",
      "Epoch 125, Loss 40.905106\n",
      "Epoch 126, Loss 40.769043\n",
      "Epoch 127, Loss 40.634613\n",
      "Epoch 128, Loss 40.501797\n",
      "Epoch 129, Loss 40.370579\n",
      "Epoch 130, Loss 40.240940\n",
      "Epoch 131, Loss 40.112854\n",
      "Epoch 132, Loss 39.986301\n",
      "Epoch 133, Loss 39.861279\n",
      "Epoch 134, Loss 39.737747\n",
      "Epoch 135, Loss 39.615700\n",
      "Epoch 136, Loss 39.495121\n",
      "Epoch 137, Loss 39.375992\n",
      "Epoch 138, Loss 39.258289\n",
      "Epoch 139, Loss 39.141994\n",
      "Epoch 140, Loss 39.027100\n",
      "Epoch 141, Loss 38.913582\n",
      "Epoch 142, Loss 38.801426\n",
      "Epoch 143, Loss 38.690617\n",
      "Epoch 144, Loss 38.581135\n",
      "Epoch 145, Loss 38.472965\n",
      "Epoch 146, Loss 38.366096\n",
      "Epoch 147, Loss 38.260502\n",
      "Epoch 148, Loss 38.156181\n",
      "Epoch 149, Loss 38.053116\n",
      "Epoch 150, Loss 37.951271\n",
      "Epoch 151, Loss 37.850651\n",
      "Epoch 152, Loss 37.751236\n",
      "Epoch 153, Loss 37.653019\n",
      "Epoch 154, Loss 37.555973\n",
      "Epoch 155, Loss 37.460091\n",
      "Epoch 156, Loss 37.365356\n",
      "Epoch 157, Loss 37.271755\n",
      "Epoch 158, Loss 37.179283\n",
      "Epoch 159, Loss 37.087917\n",
      "Epoch 160, Loss 36.997639\n",
      "Epoch 161, Loss 36.908447\n",
      "Epoch 162, Loss 36.820320\n",
      "Epoch 163, Loss 36.733246\n",
      "Epoch 164, Loss 36.647213\n",
      "Epoch 165, Loss 36.562210\n",
      "Epoch 166, Loss 36.478230\n",
      "Epoch 167, Loss 36.395248\n",
      "Epoch 168, Loss 36.313259\n",
      "Epoch 169, Loss 36.232250\n",
      "Epoch 170, Loss 36.152210\n",
      "Epoch 171, Loss 36.073124\n",
      "Epoch 172, Loss 35.994991\n",
      "Epoch 173, Loss 35.917782\n",
      "Epoch 174, Loss 35.841503\n",
      "Epoch 175, Loss 35.766136\n",
      "Epoch 176, Loss 35.691666\n",
      "Epoch 177, Loss 35.618088\n",
      "Epoch 178, Loss 35.545391\n",
      "Epoch 179, Loss 35.473557\n",
      "Epoch 180, Loss 35.402580\n",
      "Epoch 181, Loss 35.332458\n",
      "Epoch 182, Loss 35.263168\n",
      "Epoch 183, Loss 35.194702\n",
      "Epoch 184, Loss 35.127052\n",
      "Epoch 185, Loss 35.060219\n",
      "Epoch 186, Loss 34.994175\n",
      "Epoch 187, Loss 34.928921\n",
      "Epoch 188, Loss 34.864445\n",
      "Epoch 189, Loss 34.800747\n",
      "Epoch 190, Loss 34.737793\n",
      "Epoch 191, Loss 34.675594\n",
      "Epoch 192, Loss 34.614136\n",
      "Epoch 193, Loss 34.553417\n",
      "Epoch 194, Loss 34.493412\n",
      "Epoch 195, Loss 34.434128\n",
      "Epoch 196, Loss 34.375546\n",
      "Epoch 197, Loss 34.317661\n",
      "Epoch 198, Loss 34.260471\n",
      "Epoch 199, Loss 34.203953\n",
      "Epoch 200, Loss 34.148117\n",
      "Epoch 201, Loss 34.092941\n",
      "Epoch 202, Loss 34.038418\n",
      "Epoch 203, Loss 33.984543\n",
      "Epoch 204, Loss 33.931316\n",
      "Epoch 205, Loss 33.878712\n",
      "Epoch 206, Loss 33.826748\n",
      "Epoch 207, Loss 33.775391\n",
      "Epoch 208, Loss 33.724644\n",
      "Epoch 209, Loss 33.674500\n",
      "Epoch 210, Loss 33.624954\n",
      "Epoch 211, Loss 33.575996\n",
      "Epoch 212, Loss 33.527615\n",
      "Epoch 213, Loss 33.479809\n",
      "Epoch 214, Loss 33.432571\n",
      "Epoch 215, Loss 33.385895\n",
      "Epoch 216, Loss 33.339771\n",
      "Epoch 217, Loss 33.294197\n",
      "Epoch 218, Loss 33.249161\n",
      "Epoch 219, Loss 33.204655\n",
      "Epoch 220, Loss 33.160679\n",
      "Epoch 221, Loss 33.117226\n",
      "Epoch 222, Loss 33.074280\n",
      "Epoch 223, Loss 33.031853\n",
      "Epoch 224, Loss 32.989925\n",
      "Epoch 225, Loss 32.948486\n",
      "Epoch 226, Loss 32.907547\n",
      "Epoch 227, Loss 32.867088\n",
      "Epoch 228, Loss 32.827110\n",
      "Epoch 229, Loss 32.787601\n",
      "Epoch 230, Loss 32.748558\n",
      "Epoch 231, Loss 32.709972\n",
      "Epoch 232, Loss 32.671848\n",
      "Epoch 233, Loss 32.634171\n",
      "Epoch 234, Loss 32.596939\n",
      "Epoch 235, Loss 32.560154\n",
      "Epoch 236, Loss 32.523788\n",
      "Epoch 237, Loss 32.487865\n",
      "Epoch 238, Loss 32.452354\n",
      "Epoch 239, Loss 32.417274\n",
      "Epoch 240, Loss 32.382595\n",
      "Epoch 241, Loss 32.348335\n",
      "Epoch 242, Loss 32.314476\n",
      "Epoch 243, Loss 32.281010\n",
      "Epoch 244, Loss 32.247936\n",
      "Epoch 245, Loss 32.215260\n",
      "Epoch 246, Loss 32.182964\n",
      "Epoch 247, Loss 32.151047\n",
      "Epoch 248, Loss 32.119511\n",
      "Epoch 249, Loss 32.088341\n",
      "Epoch 250, Loss 32.057533\n",
      "Epoch 251, Loss 32.027084\n",
      "Epoch 252, Loss 31.997004\n",
      "Epoch 253, Loss 31.967274\n",
      "Epoch 254, Loss 31.937885\n",
      "Epoch 255, Loss 31.908852\n",
      "Epoch 256, Loss 31.880150\n",
      "Epoch 257, Loss 31.851784\n",
      "Epoch 258, Loss 31.823750\n",
      "Epoch 259, Loss 31.796049\n",
      "Epoch 260, Loss 31.768675\n",
      "Epoch 261, Loss 31.741610\n",
      "Epoch 262, Loss 31.714869\n",
      "Epoch 263, Loss 31.688433\n",
      "Epoch 264, Loss 31.662321\n",
      "Epoch 265, Loss 31.636497\n",
      "Epoch 266, Loss 31.610979\n",
      "Epoch 267, Loss 31.585760\n",
      "Epoch 268, Loss 31.560833\n",
      "Epoch 269, Loss 31.536200\n",
      "Epoch 270, Loss 31.511852\n",
      "Epoch 271, Loss 31.487787\n",
      "Epoch 272, Loss 31.464001\n",
      "Epoch 273, Loss 31.440496\n",
      "Epoch 274, Loss 31.417265\n",
      "Epoch 275, Loss 31.394300\n",
      "Epoch 276, Loss 31.371605\n",
      "Epoch 277, Loss 31.349176\n",
      "Epoch 278, Loss 31.326998\n",
      "Epoch 279, Loss 31.305086\n",
      "Epoch 280, Loss 31.283424\n",
      "Epoch 281, Loss 31.262011\n",
      "Epoch 282, Loss 31.240850\n",
      "Epoch 283, Loss 31.219934\n",
      "Epoch 284, Loss 31.199255\n",
      "Epoch 285, Loss 31.178816\n",
      "Epoch 286, Loss 31.158617\n",
      "Epoch 287, Loss 31.138647\n",
      "Epoch 288, Loss 31.118914\n",
      "Epoch 289, Loss 31.099403\n",
      "Epoch 290, Loss 31.080120\n",
      "Epoch 291, Loss 31.061064\n",
      "Epoch 292, Loss 31.042219\n",
      "Epoch 293, Loss 31.023596\n",
      "Epoch 294, Loss 31.005182\n",
      "Epoch 295, Loss 30.986982\n",
      "Epoch 296, Loss 30.968996\n",
      "Epoch 297, Loss 30.951210\n",
      "Epoch 298, Loss 30.933632\n",
      "Epoch 299, Loss 30.916254\n",
      "Epoch 300, Loss 30.899076\n",
      "Epoch 301, Loss 30.882097\n",
      "Epoch 302, Loss 30.865307\n",
      "Epoch 303, Loss 30.848717\n",
      "Epoch 304, Loss 30.832308\n",
      "Epoch 305, Loss 30.816095\n",
      "Epoch 306, Loss 30.800066\n",
      "Epoch 307, Loss 30.784210\n",
      "Epoch 308, Loss 30.768547\n",
      "Epoch 309, Loss 30.753054\n",
      "Epoch 310, Loss 30.737740\n",
      "Epoch 311, Loss 30.722597\n",
      "Epoch 312, Loss 30.707636\n",
      "Epoch 313, Loss 30.692839\n",
      "Epoch 314, Loss 30.678209\n",
      "Epoch 315, Loss 30.663750\n",
      "Epoch 316, Loss 30.649445\n",
      "Epoch 317, Loss 30.635309\n",
      "Epoch 318, Loss 30.621326\n",
      "Epoch 319, Loss 30.607512\n",
      "Epoch 320, Loss 30.593849\n",
      "Epoch 321, Loss 30.580336\n",
      "Epoch 322, Loss 30.566988\n",
      "Epoch 323, Loss 30.553783\n",
      "Epoch 324, Loss 30.540724\n",
      "Epoch 325, Loss 30.527813\n",
      "Epoch 326, Loss 30.515051\n",
      "Epoch 327, Loss 30.502424\n",
      "Epoch 328, Loss 30.489946\n",
      "Epoch 329, Loss 30.477612\n",
      "Epoch 330, Loss 30.465405\n",
      "Epoch 331, Loss 30.453346\n",
      "Epoch 332, Loss 30.441418\n",
      "Epoch 333, Loss 30.429613\n",
      "Epoch 334, Loss 30.417952\n",
      "Epoch 335, Loss 30.406422\n",
      "Epoch 336, Loss 30.395020\n",
      "Epoch 337, Loss 30.383741\n",
      "Epoch 338, Loss 30.372587\n",
      "Epoch 339, Loss 30.361555\n",
      "Epoch 340, Loss 30.350647\n",
      "Epoch 341, Loss 30.339863\n",
      "Epoch 342, Loss 30.329195\n",
      "Epoch 343, Loss 30.318647\n",
      "Epoch 344, Loss 30.308222\n",
      "Epoch 345, Loss 30.297895\n",
      "Epoch 346, Loss 30.287697\n",
      "Epoch 347, Loss 30.277611\n",
      "Epoch 348, Loss 30.267628\n",
      "Epoch 349, Loss 30.257763\n",
      "Epoch 350, Loss 30.247999\n",
      "Epoch 351, Loss 30.238344\n",
      "Epoch 352, Loss 30.228796\n",
      "Epoch 353, Loss 30.219355\n",
      "Epoch 354, Loss 30.210014\n",
      "Epoch 355, Loss 30.200781\n",
      "Epoch 356, Loss 30.191645\n",
      "Epoch 357, Loss 30.182604\n",
      "Epoch 358, Loss 30.173668\n",
      "Epoch 359, Loss 30.164824\n",
      "Epoch 360, Loss 30.156078\n",
      "Epoch 361, Loss 30.147429\n",
      "Epoch 362, Loss 30.138872\n",
      "Epoch 363, Loss 30.130409\n",
      "Epoch 364, Loss 30.122040\n",
      "Epoch 365, Loss 30.113750\n",
      "Epoch 366, Loss 30.105558\n",
      "Epoch 367, Loss 30.097454\n",
      "Epoch 368, Loss 30.089436\n",
      "Epoch 369, Loss 30.081505\n",
      "Epoch 370, Loss 30.073658\n",
      "Epoch 371, Loss 30.065889\n",
      "Epoch 372, Loss 30.058210\n",
      "Epoch 373, Loss 30.050621\n",
      "Epoch 374, Loss 30.043102\n",
      "Epoch 375, Loss 30.035662\n",
      "Epoch 376, Loss 30.028307\n",
      "Epoch 377, Loss 30.021025\n",
      "Epoch 378, Loss 30.013823\n",
      "Epoch 379, Loss 30.006697\n",
      "Epoch 380, Loss 29.999641\n",
      "Epoch 381, Loss 29.992664\n",
      "Epoch 382, Loss 29.985760\n",
      "Epoch 383, Loss 29.978931\n",
      "Epoch 384, Loss 29.972170\n",
      "Epoch 385, Loss 29.965483\n",
      "Epoch 386, Loss 29.958862\n",
      "Epoch 387, Loss 29.952314\n",
      "Epoch 388, Loss 29.945829\n",
      "Epoch 389, Loss 29.939421\n",
      "Epoch 390, Loss 29.933073\n",
      "Epoch 391, Loss 29.926796\n",
      "Epoch 392, Loss 29.920580\n",
      "Epoch 393, Loss 29.914427\n",
      "Epoch 394, Loss 29.908337\n",
      "Epoch 395, Loss 29.902315\n",
      "Epoch 396, Loss 29.896349\n",
      "Epoch 397, Loss 29.890453\n",
      "Epoch 398, Loss 29.884607\n",
      "Epoch 399, Loss 29.878826\n",
      "Epoch 400, Loss 29.873108\n",
      "Epoch 401, Loss 29.867437\n",
      "Epoch 402, Loss 29.861834\n",
      "Epoch 403, Loss 29.856287\n",
      "Epoch 404, Loss 29.850792\n",
      "Epoch 405, Loss 29.845354\n",
      "Epoch 406, Loss 29.839972\n",
      "Epoch 407, Loss 29.834639\n",
      "Epoch 408, Loss 29.829369\n",
      "Epoch 409, Loss 29.824144\n",
      "Epoch 410, Loss 29.818970\n",
      "Epoch 411, Loss 29.813854\n",
      "Epoch 412, Loss 29.808788\n",
      "Epoch 413, Loss 29.803772\n",
      "Epoch 414, Loss 29.798803\n",
      "Epoch 415, Loss 29.793884\n",
      "Epoch 416, Loss 29.789015\n",
      "Epoch 417, Loss 29.784193\n",
      "Epoch 418, Loss 29.779421\n",
      "Epoch 419, Loss 29.774691\n",
      "Epoch 420, Loss 29.770014\n",
      "Epoch 421, Loss 29.765379\n",
      "Epoch 422, Loss 29.760786\n",
      "Epoch 423, Loss 29.756237\n",
      "Epoch 424, Loss 29.751738\n",
      "Epoch 425, Loss 29.747284\n",
      "Epoch 426, Loss 29.742870\n",
      "Epoch 427, Loss 29.738495\n",
      "Epoch 428, Loss 29.734163\n",
      "Epoch 429, Loss 29.729876\n",
      "Epoch 430, Loss 29.725630\n",
      "Epoch 431, Loss 29.721418\n",
      "Epoch 432, Loss 29.717257\n",
      "Epoch 433, Loss 29.713129\n",
      "Epoch 434, Loss 29.709040\n",
      "Epoch 435, Loss 29.704988\n",
      "Epoch 436, Loss 29.700983\n",
      "Epoch 437, Loss 29.697004\n",
      "Epoch 438, Loss 29.693068\n",
      "Epoch 439, Loss 29.689171\n",
      "Epoch 440, Loss 29.685303\n",
      "Epoch 441, Loss 29.681477\n",
      "Epoch 442, Loss 29.677685\n",
      "Epoch 443, Loss 29.673922\n",
      "Epoch 444, Loss 29.670204\n",
      "Epoch 445, Loss 29.666512\n",
      "Epoch 446, Loss 29.662859\n",
      "Epoch 447, Loss 29.659235\n",
      "Epoch 448, Loss 29.655647\n",
      "Epoch 449, Loss 29.652094\n",
      "Epoch 450, Loss 29.648565\n",
      "Epoch 451, Loss 29.645073\n",
      "Epoch 452, Loss 29.641615\n",
      "Epoch 453, Loss 29.638182\n",
      "Epoch 454, Loss 29.634783\n",
      "Epoch 455, Loss 29.631414\n",
      "Epoch 456, Loss 29.628069\n",
      "Epoch 457, Loss 29.624767\n",
      "Epoch 458, Loss 29.621483\n",
      "Epoch 459, Loss 29.618237\n",
      "Epoch 460, Loss 29.615007\n",
      "Epoch 461, Loss 29.611816\n",
      "Epoch 462, Loss 29.608644\n",
      "Epoch 463, Loss 29.605505\n",
      "Epoch 464, Loss 29.602392\n",
      "Epoch 465, Loss 29.599304\n",
      "Epoch 466, Loss 29.596251\n",
      "Epoch 467, Loss 29.593212\n",
      "Epoch 468, Loss 29.590204\n",
      "Epoch 469, Loss 29.587221\n",
      "Epoch 470, Loss 29.584267\n",
      "Epoch 471, Loss 29.581329\n",
      "Epoch 472, Loss 29.578424\n",
      "Epoch 473, Loss 29.575539\n",
      "Epoch 474, Loss 29.572685\n",
      "Epoch 475, Loss 29.569847\n",
      "Epoch 476, Loss 29.567034\n",
      "Epoch 477, Loss 29.564245\n",
      "Epoch 478, Loss 29.561480\n",
      "Epoch 479, Loss 29.558739\n",
      "Epoch 480, Loss 29.556017\n",
      "Epoch 481, Loss 29.553316\n",
      "Epoch 482, Loss 29.550640\n",
      "Epoch 483, Loss 29.547985\n",
      "Epoch 484, Loss 29.545347\n",
      "Epoch 485, Loss 29.542736\n",
      "Epoch 486, Loss 29.540148\n",
      "Epoch 487, Loss 29.537569\n",
      "Epoch 488, Loss 29.535021\n",
      "Epoch 489, Loss 29.532488\n",
      "Epoch 490, Loss 29.529974\n",
      "Epoch 491, Loss 29.527483\n",
      "Epoch 492, Loss 29.525013\n",
      "Epoch 493, Loss 29.522554\n",
      "Epoch 494, Loss 29.520119\n",
      "Epoch 495, Loss 29.517706\n",
      "Epoch 496, Loss 29.515306\n",
      "Epoch 497, Loss 29.512928\n",
      "Epoch 498, Loss 29.510561\n",
      "Epoch 499, Loss 29.508217\n",
      "Epoch 500, Loss 29.505890\n",
      "Epoch 501, Loss 29.503584\n",
      "Epoch 502, Loss 29.501293\n",
      "Epoch 503, Loss 29.499010\n",
      "Epoch 504, Loss 29.496754\n",
      "Epoch 505, Loss 29.494513\n",
      "Epoch 506, Loss 29.492285\n",
      "Epoch 507, Loss 29.490076\n",
      "Epoch 508, Loss 29.487879\n",
      "Epoch 509, Loss 29.485701\n",
      "Epoch 510, Loss 29.483534\n",
      "Epoch 511, Loss 29.481390\n",
      "Epoch 512, Loss 29.479259\n",
      "Epoch 513, Loss 29.477142\n",
      "Epoch 514, Loss 29.475035\n",
      "Epoch 515, Loss 29.472948\n",
      "Epoch 516, Loss 29.470869\n",
      "Epoch 517, Loss 29.468811\n",
      "Epoch 518, Loss 29.466768\n",
      "Epoch 519, Loss 29.464739\n",
      "Epoch 520, Loss 29.462719\n",
      "Epoch 521, Loss 29.460716\n",
      "Epoch 522, Loss 29.458723\n",
      "Epoch 523, Loss 29.456751\n",
      "Epoch 524, Loss 29.454781\n",
      "Epoch 525, Loss 29.452833\n",
      "Epoch 526, Loss 29.450891\n",
      "Epoch 527, Loss 29.448961\n",
      "Epoch 528, Loss 29.447052\n",
      "Epoch 529, Loss 29.445147\n",
      "Epoch 530, Loss 29.443260\n",
      "Epoch 531, Loss 29.441383\n",
      "Epoch 532, Loss 29.439516\n",
      "Epoch 533, Loss 29.437664\n",
      "Epoch 534, Loss 29.435820\n",
      "Epoch 535, Loss 29.433989\n",
      "Epoch 536, Loss 29.432171\n",
      "Epoch 537, Loss 29.430357\n",
      "Epoch 538, Loss 29.428566\n",
      "Epoch 539, Loss 29.426781\n",
      "Epoch 540, Loss 29.425005\n",
      "Epoch 541, Loss 29.423241\n",
      "Epoch 542, Loss 29.421484\n",
      "Epoch 543, Loss 29.419739\n",
      "Epoch 544, Loss 29.418007\n",
      "Epoch 545, Loss 29.416283\n",
      "Epoch 546, Loss 29.414568\n",
      "Epoch 547, Loss 29.412867\n",
      "Epoch 548, Loss 29.411171\n",
      "Epoch 549, Loss 29.409491\n",
      "Epoch 550, Loss 29.407814\n",
      "Epoch 551, Loss 29.406147\n",
      "Epoch 552, Loss 29.404491\n",
      "Epoch 553, Loss 29.402843\n",
      "Epoch 554, Loss 29.401203\n",
      "Epoch 555, Loss 29.399578\n",
      "Epoch 556, Loss 29.397955\n",
      "Epoch 557, Loss 29.396345\n",
      "Epoch 558, Loss 29.394747\n",
      "Epoch 559, Loss 29.393150\n",
      "Epoch 560, Loss 29.391569\n",
      "Epoch 561, Loss 29.389986\n",
      "Epoch 562, Loss 29.388416\n",
      "Epoch 563, Loss 29.386858\n",
      "Epoch 564, Loss 29.385302\n",
      "Epoch 565, Loss 29.383760\n",
      "Epoch 566, Loss 29.382225\n",
      "Epoch 567, Loss 29.380693\n",
      "Epoch 568, Loss 29.379179\n",
      "Epoch 569, Loss 29.377663\n",
      "Epoch 570, Loss 29.376152\n",
      "Epoch 571, Loss 29.374653\n",
      "Epoch 572, Loss 29.373161\n",
      "Epoch 573, Loss 29.371675\n",
      "Epoch 574, Loss 29.370201\n",
      "Epoch 575, Loss 29.368731\n",
      "Epoch 576, Loss 29.367268\n",
      "Epoch 577, Loss 29.365812\n",
      "Epoch 578, Loss 29.364361\n",
      "Epoch 579, Loss 29.362921\n",
      "Epoch 580, Loss 29.361486\n",
      "Epoch 581, Loss 29.360052\n",
      "Epoch 582, Loss 29.358629\n",
      "Epoch 583, Loss 29.357216\n",
      "Epoch 584, Loss 29.355803\n",
      "Epoch 585, Loss 29.354399\n",
      "Epoch 586, Loss 29.353003\n",
      "Epoch 587, Loss 29.351612\n",
      "Epoch 588, Loss 29.350225\n",
      "Epoch 589, Loss 29.348846\n",
      "Epoch 590, Loss 29.347466\n",
      "Epoch 591, Loss 29.346100\n",
      "Epoch 592, Loss 29.344738\n",
      "Epoch 593, Loss 29.343382\n",
      "Epoch 594, Loss 29.342028\n",
      "Epoch 595, Loss 29.340687\n",
      "Epoch 596, Loss 29.339342\n",
      "Epoch 597, Loss 29.338007\n",
      "Epoch 598, Loss 29.336678\n",
      "Epoch 599, Loss 29.335358\n",
      "Epoch 600, Loss 29.334038\n",
      "Epoch 601, Loss 29.332727\n",
      "Epoch 602, Loss 29.331415\n",
      "Epoch 603, Loss 29.330109\n",
      "Epoch 604, Loss 29.328810\n",
      "Epoch 605, Loss 29.327517\n",
      "Epoch 606, Loss 29.326227\n",
      "Epoch 607, Loss 29.324944\n",
      "Epoch 608, Loss 29.323664\n",
      "Epoch 609, Loss 29.322393\n",
      "Epoch 610, Loss 29.321114\n",
      "Epoch 611, Loss 29.319853\n",
      "Epoch 612, Loss 29.318590\n",
      "Epoch 613, Loss 29.317339\n",
      "Epoch 614, Loss 29.316082\n",
      "Epoch 615, Loss 29.314833\n",
      "Epoch 616, Loss 29.313597\n",
      "Epoch 617, Loss 29.312349\n",
      "Epoch 618, Loss 29.311119\n",
      "Epoch 619, Loss 29.309893\n",
      "Epoch 620, Loss 29.308655\n",
      "Epoch 621, Loss 29.307434\n",
      "Epoch 622, Loss 29.306215\n",
      "Epoch 623, Loss 29.305004\n",
      "Epoch 624, Loss 29.303789\n",
      "Epoch 625, Loss 29.302580\n",
      "Epoch 626, Loss 29.301378\n",
      "Epoch 627, Loss 29.300179\n",
      "Epoch 628, Loss 29.298983\n",
      "Epoch 629, Loss 29.297791\n",
      "Epoch 630, Loss 29.296600\n",
      "Epoch 631, Loss 29.295416\n",
      "Epoch 632, Loss 29.294233\n",
      "Epoch 633, Loss 29.293055\n",
      "Epoch 634, Loss 29.291883\n",
      "Epoch 635, Loss 29.290710\n",
      "Epoch 636, Loss 29.289545\n",
      "Epoch 637, Loss 29.288378\n",
      "Epoch 638, Loss 29.287214\n",
      "Epoch 639, Loss 29.286064\n",
      "Epoch 640, Loss 29.284901\n",
      "Epoch 641, Loss 29.283752\n",
      "Epoch 642, Loss 29.282604\n",
      "Epoch 643, Loss 29.281466\n",
      "Epoch 644, Loss 29.280317\n",
      "Epoch 645, Loss 29.279175\n",
      "Epoch 646, Loss 29.278044\n",
      "Epoch 647, Loss 29.276911\n",
      "Epoch 648, Loss 29.275780\n",
      "Epoch 649, Loss 29.274652\n",
      "Epoch 650, Loss 29.273527\n",
      "Epoch 651, Loss 29.272406\n",
      "Epoch 652, Loss 29.271284\n",
      "Epoch 653, Loss 29.270164\n",
      "Epoch 654, Loss 29.269056\n",
      "Epoch 655, Loss 29.267942\n",
      "Epoch 656, Loss 29.266832\n",
      "Epoch 657, Loss 29.265728\n",
      "Epoch 658, Loss 29.264629\n",
      "Epoch 659, Loss 29.263525\n",
      "Epoch 660, Loss 29.262426\n",
      "Epoch 661, Loss 29.261335\n",
      "Epoch 662, Loss 29.260237\n",
      "Epoch 663, Loss 29.259150\n",
      "Epoch 664, Loss 29.258059\n",
      "Epoch 665, Loss 29.256975\n",
      "Epoch 666, Loss 29.255892\n",
      "Epoch 667, Loss 29.254810\n",
      "Epoch 668, Loss 29.253729\n",
      "Epoch 669, Loss 29.252655\n",
      "Epoch 670, Loss 29.251581\n",
      "Epoch 671, Loss 29.250507\n",
      "Epoch 672, Loss 29.249434\n",
      "Epoch 673, Loss 29.248369\n",
      "Epoch 674, Loss 29.247303\n",
      "Epoch 675, Loss 29.246239\n",
      "Epoch 676, Loss 29.245178\n",
      "Epoch 677, Loss 29.244116\n",
      "Epoch 678, Loss 29.243059\n",
      "Epoch 679, Loss 29.242010\n",
      "Epoch 680, Loss 29.240955\n",
      "Epoch 681, Loss 29.239901\n",
      "Epoch 682, Loss 29.238850\n",
      "Epoch 683, Loss 29.237804\n",
      "Epoch 684, Loss 29.236757\n",
      "Epoch 685, Loss 29.235712\n",
      "Epoch 686, Loss 29.234674\n",
      "Epoch 687, Loss 29.233631\n",
      "Epoch 688, Loss 29.232594\n",
      "Epoch 689, Loss 29.231560\n",
      "Epoch 690, Loss 29.230522\n",
      "Epoch 691, Loss 29.229490\n",
      "Epoch 692, Loss 29.228460\n",
      "Epoch 693, Loss 29.227428\n",
      "Epoch 694, Loss 29.226398\n",
      "Epoch 695, Loss 29.225374\n",
      "Epoch 696, Loss 29.224354\n",
      "Epoch 697, Loss 29.223328\n",
      "Epoch 698, Loss 29.222307\n",
      "Epoch 699, Loss 29.221289\n",
      "Epoch 700, Loss 29.220268\n",
      "Epoch 701, Loss 29.219250\n",
      "Epoch 702, Loss 29.218233\n",
      "Epoch 703, Loss 29.217218\n",
      "Epoch 704, Loss 29.216211\n",
      "Epoch 705, Loss 29.215199\n",
      "Epoch 706, Loss 29.214190\n",
      "Epoch 707, Loss 29.213182\n",
      "Epoch 708, Loss 29.212175\n",
      "Epoch 709, Loss 29.211170\n",
      "Epoch 710, Loss 29.210167\n",
      "Epoch 711, Loss 29.209166\n",
      "Epoch 712, Loss 29.208166\n",
      "Epoch 713, Loss 29.207165\n",
      "Epoch 714, Loss 29.206169\n",
      "Epoch 715, Loss 29.205172\n",
      "Epoch 716, Loss 29.204174\n",
      "Epoch 717, Loss 29.203180\n",
      "Epoch 718, Loss 29.202187\n",
      "Epoch 719, Loss 29.201195\n",
      "Epoch 720, Loss 29.200203\n",
      "Epoch 721, Loss 29.199217\n",
      "Epoch 722, Loss 29.198231\n",
      "Epoch 723, Loss 29.197241\n",
      "Epoch 724, Loss 29.196253\n",
      "Epoch 725, Loss 29.195271\n",
      "Epoch 726, Loss 29.194286\n",
      "Epoch 727, Loss 29.193310\n",
      "Epoch 728, Loss 29.192322\n",
      "Epoch 729, Loss 29.191339\n",
      "Epoch 730, Loss 29.190363\n",
      "Epoch 731, Loss 29.189384\n",
      "Epoch 732, Loss 29.188410\n",
      "Epoch 733, Loss 29.187433\n",
      "Epoch 734, Loss 29.186460\n",
      "Epoch 735, Loss 29.185486\n",
      "Epoch 736, Loss 29.184515\n",
      "Epoch 737, Loss 29.183540\n",
      "Epoch 738, Loss 29.182570\n",
      "Epoch 739, Loss 29.181602\n",
      "Epoch 740, Loss 29.180632\n",
      "Epoch 741, Loss 29.179663\n",
      "Epoch 742, Loss 29.178698\n",
      "Epoch 743, Loss 29.177732\n",
      "Epoch 744, Loss 29.176760\n",
      "Epoch 745, Loss 29.175804\n",
      "Epoch 746, Loss 29.174837\n",
      "Epoch 747, Loss 29.173878\n",
      "Epoch 748, Loss 29.172915\n",
      "Epoch 749, Loss 29.171955\n",
      "Epoch 750, Loss 29.170992\n",
      "Epoch 751, Loss 29.170036\n",
      "Epoch 752, Loss 29.169079\n",
      "Epoch 753, Loss 29.168119\n",
      "Epoch 754, Loss 29.167164\n",
      "Epoch 755, Loss 29.166204\n",
      "Epoch 756, Loss 29.165255\n",
      "Epoch 757, Loss 29.164295\n",
      "Epoch 758, Loss 29.163347\n",
      "Epoch 759, Loss 29.162397\n",
      "Epoch 760, Loss 29.161444\n",
      "Epoch 761, Loss 29.160494\n",
      "Epoch 762, Loss 29.159540\n",
      "Epoch 763, Loss 29.158598\n",
      "Epoch 764, Loss 29.157646\n",
      "Epoch 765, Loss 29.156694\n",
      "Epoch 766, Loss 29.155750\n",
      "Epoch 767, Loss 29.154802\n",
      "Epoch 768, Loss 29.153852\n",
      "Epoch 769, Loss 29.152916\n",
      "Epoch 770, Loss 29.151966\n",
      "Epoch 771, Loss 29.151030\n",
      "Epoch 772, Loss 29.150084\n",
      "Epoch 773, Loss 29.149139\n",
      "Epoch 774, Loss 29.148199\n",
      "Epoch 775, Loss 29.147259\n",
      "Epoch 776, Loss 29.146315\n",
      "Epoch 777, Loss 29.145380\n",
      "Epoch 778, Loss 29.144438\n",
      "Epoch 779, Loss 29.143499\n",
      "Epoch 780, Loss 29.142559\n",
      "Epoch 781, Loss 29.141624\n",
      "Epoch 782, Loss 29.140684\n",
      "Epoch 783, Loss 29.139748\n",
      "Epoch 784, Loss 29.138819\n",
      "Epoch 785, Loss 29.137878\n",
      "Epoch 786, Loss 29.136946\n",
      "Epoch 787, Loss 29.136015\n",
      "Epoch 788, Loss 29.135082\n",
      "Epoch 789, Loss 29.134144\n",
      "Epoch 790, Loss 29.133215\n",
      "Epoch 791, Loss 29.132280\n",
      "Epoch 792, Loss 29.131348\n",
      "Epoch 793, Loss 29.130419\n",
      "Epoch 794, Loss 29.129488\n",
      "Epoch 795, Loss 29.128559\n",
      "Epoch 796, Loss 29.127630\n",
      "Epoch 797, Loss 29.126698\n",
      "Epoch 798, Loss 29.125771\n",
      "Epoch 799, Loss 29.124847\n",
      "Epoch 800, Loss 29.123920\n",
      "Epoch 801, Loss 29.122993\n",
      "Epoch 802, Loss 29.122065\n",
      "Epoch 803, Loss 29.121138\n",
      "Epoch 804, Loss 29.120211\n",
      "Epoch 805, Loss 29.119291\n",
      "Epoch 806, Loss 29.118370\n",
      "Epoch 807, Loss 29.117443\n",
      "Epoch 808, Loss 29.116522\n",
      "Epoch 809, Loss 29.115599\n",
      "Epoch 810, Loss 29.114674\n",
      "Epoch 811, Loss 29.113752\n",
      "Epoch 812, Loss 29.112831\n",
      "Epoch 813, Loss 29.111910\n",
      "Epoch 814, Loss 29.110987\n",
      "Epoch 815, Loss 29.110069\n",
      "Epoch 816, Loss 29.109150\n",
      "Epoch 817, Loss 29.108229\n",
      "Epoch 818, Loss 29.107311\n",
      "Epoch 819, Loss 29.106392\n",
      "Epoch 820, Loss 29.105474\n",
      "Epoch 821, Loss 29.104553\n",
      "Epoch 822, Loss 29.103638\n",
      "Epoch 823, Loss 29.102720\n",
      "Epoch 824, Loss 29.101801\n",
      "Epoch 825, Loss 29.100885\n",
      "Epoch 826, Loss 29.099968\n",
      "Epoch 827, Loss 29.099054\n",
      "Epoch 828, Loss 29.098137\n",
      "Epoch 829, Loss 29.097223\n",
      "Epoch 830, Loss 29.096306\n",
      "Epoch 831, Loss 29.095392\n",
      "Epoch 832, Loss 29.094482\n",
      "Epoch 833, Loss 29.093569\n",
      "Epoch 834, Loss 29.092651\n",
      "Epoch 835, Loss 29.091738\n",
      "Epoch 836, Loss 29.090828\n",
      "Epoch 837, Loss 29.089916\n",
      "Epoch 838, Loss 29.089006\n",
      "Epoch 839, Loss 29.088091\n",
      "Epoch 840, Loss 29.087181\n",
      "Epoch 841, Loss 29.086262\n",
      "Epoch 842, Loss 29.085356\n",
      "Epoch 843, Loss 29.084448\n",
      "Epoch 844, Loss 29.083534\n",
      "Epoch 845, Loss 29.082623\n",
      "Epoch 846, Loss 29.081715\n",
      "Epoch 847, Loss 29.080805\n",
      "Epoch 848, Loss 29.079895\n",
      "Epoch 849, Loss 29.078987\n",
      "Epoch 850, Loss 29.078081\n",
      "Epoch 851, Loss 29.077173\n",
      "Epoch 852, Loss 29.076262\n",
      "Epoch 853, Loss 29.075354\n",
      "Epoch 854, Loss 29.074446\n",
      "Epoch 855, Loss 29.073542\n",
      "Epoch 856, Loss 29.072634\n",
      "Epoch 857, Loss 29.071724\n",
      "Epoch 858, Loss 29.070820\n",
      "Epoch 859, Loss 29.069914\n",
      "Epoch 860, Loss 29.069008\n",
      "Epoch 861, Loss 29.068098\n",
      "Epoch 862, Loss 29.067194\n",
      "Epoch 863, Loss 29.066290\n",
      "Epoch 864, Loss 29.065386\n",
      "Epoch 865, Loss 29.064478\n",
      "Epoch 866, Loss 29.063574\n",
      "Epoch 867, Loss 29.062672\n",
      "Epoch 868, Loss 29.061768\n",
      "Epoch 869, Loss 29.060869\n",
      "Epoch 870, Loss 29.059961\n",
      "Epoch 871, Loss 29.059059\n",
      "Epoch 872, Loss 29.058155\n",
      "Epoch 873, Loss 29.057257\n",
      "Epoch 874, Loss 29.056347\n",
      "Epoch 875, Loss 29.055449\n",
      "Epoch 876, Loss 29.054546\n",
      "Epoch 877, Loss 29.053638\n",
      "Epoch 878, Loss 29.052736\n",
      "Epoch 879, Loss 29.051838\n",
      "Epoch 880, Loss 29.050932\n",
      "Epoch 881, Loss 29.050035\n",
      "Epoch 882, Loss 29.049135\n",
      "Epoch 883, Loss 29.048229\n",
      "Epoch 884, Loss 29.047331\n",
      "Epoch 885, Loss 29.046431\n",
      "Epoch 886, Loss 29.045532\n",
      "Epoch 887, Loss 29.044628\n",
      "Epoch 888, Loss 29.043732\n",
      "Epoch 889, Loss 29.042828\n",
      "Epoch 890, Loss 29.041931\n",
      "Epoch 891, Loss 29.041033\n",
      "Epoch 892, Loss 29.040133\n",
      "Epoch 893, Loss 29.039234\n",
      "Epoch 894, Loss 29.038332\n",
      "Epoch 895, Loss 29.037432\n",
      "Epoch 896, Loss 29.036537\n",
      "Epoch 897, Loss 29.035633\n",
      "Epoch 898, Loss 29.034737\n",
      "Epoch 899, Loss 29.033842\n",
      "Epoch 900, Loss 29.032942\n",
      "Epoch 901, Loss 29.032045\n",
      "Epoch 902, Loss 29.031147\n",
      "Epoch 903, Loss 29.030254\n",
      "Epoch 904, Loss 29.029356\n",
      "Epoch 905, Loss 29.028456\n",
      "Epoch 906, Loss 29.027559\n",
      "Epoch 907, Loss 29.026667\n",
      "Epoch 908, Loss 29.025764\n",
      "Epoch 909, Loss 29.024870\n",
      "Epoch 910, Loss 29.023975\n",
      "Epoch 911, Loss 29.023073\n",
      "Epoch 912, Loss 29.022181\n",
      "Epoch 913, Loss 29.021278\n",
      "Epoch 914, Loss 29.020388\n",
      "Epoch 915, Loss 29.019493\n",
      "Epoch 916, Loss 29.018599\n",
      "Epoch 917, Loss 29.017702\n",
      "Epoch 918, Loss 29.016808\n",
      "Epoch 919, Loss 29.015913\n",
      "Epoch 920, Loss 29.015017\n",
      "Epoch 921, Loss 29.014124\n",
      "Epoch 922, Loss 29.013227\n",
      "Epoch 923, Loss 29.012331\n",
      "Epoch 924, Loss 29.011436\n",
      "Epoch 925, Loss 29.010548\n",
      "Epoch 926, Loss 29.009649\n",
      "Epoch 927, Loss 29.008759\n",
      "Epoch 928, Loss 29.007862\n",
      "Epoch 929, Loss 29.006969\n",
      "Epoch 930, Loss 29.006073\n",
      "Epoch 931, Loss 29.005180\n",
      "Epoch 932, Loss 29.004286\n",
      "Epoch 933, Loss 29.003399\n",
      "Epoch 934, Loss 29.002501\n",
      "Epoch 935, Loss 29.001610\n",
      "Epoch 936, Loss 29.000713\n",
      "Epoch 937, Loss 28.999828\n",
      "Epoch 938, Loss 28.998932\n",
      "Epoch 939, Loss 28.998039\n",
      "Epoch 940, Loss 28.997139\n",
      "Epoch 941, Loss 28.996254\n",
      "Epoch 942, Loss 28.995361\n",
      "Epoch 943, Loss 28.994465\n",
      "Epoch 944, Loss 28.993580\n",
      "Epoch 945, Loss 28.992683\n",
      "Epoch 946, Loss 28.991789\n",
      "Epoch 947, Loss 28.990904\n",
      "Epoch 948, Loss 28.990009\n",
      "Epoch 949, Loss 28.989117\n",
      "Epoch 950, Loss 28.988228\n",
      "Epoch 951, Loss 28.987335\n",
      "Epoch 952, Loss 28.986448\n",
      "Epoch 953, Loss 28.985558\n",
      "Epoch 954, Loss 28.984663\n",
      "Epoch 955, Loss 28.983776\n",
      "Epoch 956, Loss 28.982880\n",
      "Epoch 957, Loss 28.981989\n",
      "Epoch 958, Loss 28.981096\n",
      "Epoch 959, Loss 28.980207\n",
      "Epoch 960, Loss 28.979315\n",
      "Epoch 961, Loss 28.978430\n",
      "Epoch 962, Loss 28.977533\n",
      "Epoch 963, Loss 28.976646\n",
      "Epoch 964, Loss 28.975756\n",
      "Epoch 965, Loss 28.974865\n",
      "Epoch 966, Loss 28.973980\n",
      "Epoch 967, Loss 28.973089\n",
      "Epoch 968, Loss 28.972198\n",
      "Epoch 969, Loss 28.971308\n",
      "Epoch 970, Loss 28.970417\n",
      "Epoch 971, Loss 28.969532\n",
      "Epoch 972, Loss 28.968639\n",
      "Epoch 973, Loss 28.967751\n",
      "Epoch 974, Loss 28.966860\n",
      "Epoch 975, Loss 28.965975\n",
      "Epoch 976, Loss 28.965082\n",
      "Epoch 977, Loss 28.964195\n",
      "Epoch 978, Loss 28.963305\n",
      "Epoch 979, Loss 28.962420\n",
      "Epoch 980, Loss 28.961525\n",
      "Epoch 981, Loss 28.960644\n",
      "Epoch 982, Loss 28.959753\n",
      "Epoch 983, Loss 28.958862\n",
      "Epoch 984, Loss 28.957975\n",
      "Epoch 985, Loss 28.957085\n",
      "Epoch 986, Loss 28.956200\n",
      "Epoch 987, Loss 28.955309\n",
      "Epoch 988, Loss 28.954420\n",
      "Epoch 989, Loss 28.953533\n",
      "Epoch 990, Loss 28.952648\n",
      "Epoch 991, Loss 28.951759\n",
      "Epoch 992, Loss 28.950872\n",
      "Epoch 993, Loss 28.949984\n",
      "Epoch 994, Loss 28.949097\n",
      "Epoch 995, Loss 28.948206\n",
      "Epoch 996, Loss 28.947321\n",
      "Epoch 997, Loss 28.946428\n",
      "Epoch 998, Loss 28.945549\n",
      "Epoch 999, Loss 28.944658\n",
      "Epoch 1000, Loss 28.943773\n",
      "Epoch 1001, Loss 28.942883\n",
      "Epoch 1002, Loss 28.941998\n",
      "Epoch 1003, Loss 28.941113\n",
      "Epoch 1004, Loss 28.940222\n",
      "Epoch 1005, Loss 28.939337\n",
      "Epoch 1006, Loss 28.938452\n",
      "Epoch 1007, Loss 28.937561\n",
      "Epoch 1008, Loss 28.936678\n",
      "Epoch 1009, Loss 28.935789\n",
      "Epoch 1010, Loss 28.934904\n",
      "Epoch 1011, Loss 28.934021\n",
      "Epoch 1012, Loss 28.933134\n",
      "Epoch 1013, Loss 28.932245\n",
      "Epoch 1014, Loss 28.931360\n",
      "Epoch 1015, Loss 28.930473\n",
      "Epoch 1016, Loss 28.929588\n",
      "Epoch 1017, Loss 28.928698\n",
      "Epoch 1018, Loss 28.927814\n",
      "Epoch 1019, Loss 28.926928\n",
      "Epoch 1020, Loss 28.926043\n",
      "Epoch 1021, Loss 28.925154\n",
      "Epoch 1022, Loss 28.924267\n",
      "Epoch 1023, Loss 28.923384\n",
      "Epoch 1024, Loss 28.922497\n",
      "Epoch 1025, Loss 28.921612\n",
      "Epoch 1026, Loss 28.920727\n",
      "Epoch 1027, Loss 28.919842\n",
      "Epoch 1028, Loss 28.918953\n",
      "Epoch 1029, Loss 28.918074\n",
      "Epoch 1030, Loss 28.917189\n",
      "Epoch 1031, Loss 28.916302\n",
      "Epoch 1032, Loss 28.915417\n",
      "Epoch 1033, Loss 28.914528\n",
      "Epoch 1034, Loss 28.913647\n",
      "Epoch 1035, Loss 28.912758\n",
      "Epoch 1036, Loss 28.911877\n",
      "Epoch 1037, Loss 28.910992\n",
      "Epoch 1038, Loss 28.910101\n",
      "Epoch 1039, Loss 28.909222\n",
      "Epoch 1040, Loss 28.908333\n",
      "Epoch 1041, Loss 28.907454\n",
      "Epoch 1042, Loss 28.906569\n",
      "Epoch 1043, Loss 28.905684\n",
      "Epoch 1044, Loss 28.904797\n",
      "Epoch 1045, Loss 28.903913\n",
      "Epoch 1046, Loss 28.903028\n",
      "Epoch 1047, Loss 28.902147\n",
      "Epoch 1048, Loss 28.901258\n",
      "Epoch 1049, Loss 28.900373\n",
      "Epoch 1050, Loss 28.899488\n",
      "Epoch 1051, Loss 28.898609\n",
      "Epoch 1052, Loss 28.897724\n",
      "Epoch 1053, Loss 28.896839\n",
      "Epoch 1054, Loss 28.895954\n",
      "Epoch 1055, Loss 28.895069\n",
      "Epoch 1056, Loss 28.894190\n",
      "Epoch 1057, Loss 28.893303\n",
      "Epoch 1058, Loss 28.892424\n",
      "Epoch 1059, Loss 28.891535\n",
      "Epoch 1060, Loss 28.890650\n",
      "Epoch 1061, Loss 28.889771\n",
      "Epoch 1062, Loss 28.888884\n",
      "Epoch 1063, Loss 28.888002\n",
      "Epoch 1064, Loss 28.887117\n",
      "Epoch 1065, Loss 28.886232\n",
      "Epoch 1066, Loss 28.885353\n",
      "Epoch 1067, Loss 28.884472\n",
      "Epoch 1068, Loss 28.883583\n",
      "Epoch 1069, Loss 28.882696\n",
      "Epoch 1070, Loss 28.881819\n",
      "Epoch 1071, Loss 28.880938\n",
      "Epoch 1072, Loss 28.880053\n",
      "Epoch 1073, Loss 28.879168\n",
      "Epoch 1074, Loss 28.878290\n",
      "Epoch 1075, Loss 28.877399\n",
      "Epoch 1076, Loss 28.876520\n",
      "Epoch 1077, Loss 28.875639\n",
      "Epoch 1078, Loss 28.874758\n",
      "Epoch 1079, Loss 28.873869\n",
      "Epoch 1080, Loss 28.872988\n",
      "Epoch 1081, Loss 28.872107\n",
      "Epoch 1082, Loss 28.871223\n",
      "Epoch 1083, Loss 28.870338\n",
      "Epoch 1084, Loss 28.869457\n",
      "Epoch 1085, Loss 28.868578\n",
      "Epoch 1086, Loss 28.867695\n",
      "Epoch 1087, Loss 28.866808\n",
      "Epoch 1088, Loss 28.865931\n",
      "Epoch 1089, Loss 28.865049\n",
      "Epoch 1090, Loss 28.864164\n",
      "Epoch 1091, Loss 28.863281\n",
      "Epoch 1092, Loss 28.862398\n",
      "Epoch 1093, Loss 28.861517\n",
      "Epoch 1094, Loss 28.860634\n",
      "Epoch 1095, Loss 28.859753\n",
      "Epoch 1096, Loss 28.858870\n",
      "Epoch 1097, Loss 28.857988\n",
      "Epoch 1098, Loss 28.857109\n",
      "Epoch 1099, Loss 28.856230\n",
      "Epoch 1100, Loss 28.855345\n",
      "Epoch 1101, Loss 28.854464\n",
      "Epoch 1102, Loss 28.853582\n",
      "Epoch 1103, Loss 28.852699\n",
      "Epoch 1104, Loss 28.851818\n",
      "Epoch 1105, Loss 28.850933\n",
      "Epoch 1106, Loss 28.850056\n",
      "Epoch 1107, Loss 28.849171\n",
      "Epoch 1108, Loss 28.848289\n",
      "Epoch 1109, Loss 28.847410\n",
      "Epoch 1110, Loss 28.846527\n",
      "Epoch 1111, Loss 28.845642\n",
      "Epoch 1112, Loss 28.844763\n",
      "Epoch 1113, Loss 28.843884\n",
      "Epoch 1114, Loss 28.843004\n",
      "Epoch 1115, Loss 28.842119\n",
      "Epoch 1116, Loss 28.841236\n",
      "Epoch 1117, Loss 28.840361\n",
      "Epoch 1118, Loss 28.839478\n",
      "Epoch 1119, Loss 28.838594\n",
      "Epoch 1120, Loss 28.837715\n",
      "Epoch 1121, Loss 28.836830\n",
      "Epoch 1122, Loss 28.835955\n",
      "Epoch 1123, Loss 28.835072\n",
      "Epoch 1124, Loss 28.834187\n",
      "Epoch 1125, Loss 28.833313\n",
      "Epoch 1126, Loss 28.832428\n",
      "Epoch 1127, Loss 28.831549\n",
      "Epoch 1128, Loss 28.830666\n",
      "Epoch 1129, Loss 28.829784\n",
      "Epoch 1130, Loss 28.828901\n",
      "Epoch 1131, Loss 28.828026\n",
      "Epoch 1132, Loss 28.827143\n",
      "Epoch 1133, Loss 28.826265\n",
      "Epoch 1134, Loss 28.825384\n",
      "Epoch 1135, Loss 28.824501\n",
      "Epoch 1136, Loss 28.823622\n",
      "Epoch 1137, Loss 28.822742\n",
      "Epoch 1138, Loss 28.821863\n",
      "Epoch 1139, Loss 28.820984\n",
      "Epoch 1140, Loss 28.820101\n",
      "Epoch 1141, Loss 28.819221\n",
      "Epoch 1142, Loss 28.818336\n",
      "Epoch 1143, Loss 28.817457\n",
      "Epoch 1144, Loss 28.816578\n",
      "Epoch 1145, Loss 28.815702\n",
      "Epoch 1146, Loss 28.814825\n",
      "Epoch 1147, Loss 28.813940\n",
      "Epoch 1148, Loss 28.813066\n",
      "Epoch 1149, Loss 28.812181\n",
      "Epoch 1150, Loss 28.811296\n",
      "Epoch 1151, Loss 28.810417\n",
      "Epoch 1152, Loss 28.809540\n",
      "Epoch 1153, Loss 28.808662\n",
      "Epoch 1154, Loss 28.807775\n",
      "Epoch 1155, Loss 28.806904\n",
      "Epoch 1156, Loss 28.806019\n",
      "Epoch 1157, Loss 28.805140\n",
      "Epoch 1158, Loss 28.804262\n",
      "Epoch 1159, Loss 28.803383\n",
      "Epoch 1160, Loss 28.802502\n",
      "Epoch 1161, Loss 28.801619\n",
      "Epoch 1162, Loss 28.800739\n",
      "Epoch 1163, Loss 28.799860\n",
      "Epoch 1164, Loss 28.798986\n",
      "Epoch 1165, Loss 28.798101\n",
      "Epoch 1166, Loss 28.797222\n",
      "Epoch 1167, Loss 28.796343\n",
      "Epoch 1168, Loss 28.795471\n",
      "Epoch 1169, Loss 28.794588\n",
      "Epoch 1170, Loss 28.793707\n",
      "Epoch 1171, Loss 28.792828\n",
      "Epoch 1172, Loss 28.791954\n",
      "Epoch 1173, Loss 28.791071\n",
      "Epoch 1174, Loss 28.790192\n",
      "Epoch 1175, Loss 28.789312\n",
      "Epoch 1176, Loss 28.788435\n",
      "Epoch 1177, Loss 28.787554\n",
      "Epoch 1178, Loss 28.786676\n",
      "Epoch 1179, Loss 28.785795\n",
      "Epoch 1180, Loss 28.784918\n",
      "Epoch 1181, Loss 28.784035\n",
      "Epoch 1182, Loss 28.783159\n",
      "Epoch 1183, Loss 28.782280\n",
      "Epoch 1184, Loss 28.781403\n",
      "Epoch 1185, Loss 28.780519\n",
      "Epoch 1186, Loss 28.779644\n",
      "Epoch 1187, Loss 28.778770\n",
      "Epoch 1188, Loss 28.777887\n",
      "Epoch 1189, Loss 28.777006\n",
      "Epoch 1190, Loss 28.776131\n",
      "Epoch 1191, Loss 28.775251\n",
      "Epoch 1192, Loss 28.774372\n",
      "Epoch 1193, Loss 28.773499\n",
      "Epoch 1194, Loss 28.772615\n",
      "Epoch 1195, Loss 28.771740\n",
      "Epoch 1196, Loss 28.770857\n",
      "Epoch 1197, Loss 28.769981\n",
      "Epoch 1198, Loss 28.769104\n",
      "Epoch 1199, Loss 28.768221\n",
      "Epoch 1200, Loss 28.767351\n",
      "Epoch 1201, Loss 28.766468\n",
      "Epoch 1202, Loss 28.765593\n",
      "Epoch 1203, Loss 28.764715\n",
      "Epoch 1204, Loss 28.763832\n",
      "Epoch 1205, Loss 28.762959\n",
      "Epoch 1206, Loss 28.762079\n",
      "Epoch 1207, Loss 28.761202\n",
      "Epoch 1208, Loss 28.760326\n",
      "Epoch 1209, Loss 28.759449\n",
      "Epoch 1210, Loss 28.758568\n",
      "Epoch 1211, Loss 28.757692\n",
      "Epoch 1212, Loss 28.756807\n",
      "Epoch 1213, Loss 28.755938\n",
      "Epoch 1214, Loss 28.755054\n",
      "Epoch 1215, Loss 28.754179\n",
      "Epoch 1216, Loss 28.753302\n",
      "Epoch 1217, Loss 28.752419\n",
      "Epoch 1218, Loss 28.751545\n",
      "Epoch 1219, Loss 28.750666\n",
      "Epoch 1220, Loss 28.749788\n",
      "Epoch 1221, Loss 28.748909\n",
      "Epoch 1222, Loss 28.748034\n",
      "Epoch 1223, Loss 28.747160\n",
      "Epoch 1224, Loss 28.746279\n",
      "Epoch 1225, Loss 28.745399\n",
      "Epoch 1226, Loss 28.744524\n",
      "Epoch 1227, Loss 28.743647\n",
      "Epoch 1228, Loss 28.742769\n",
      "Epoch 1229, Loss 28.741894\n",
      "Epoch 1230, Loss 28.741016\n",
      "Epoch 1231, Loss 28.740137\n",
      "Epoch 1232, Loss 28.739258\n",
      "Epoch 1233, Loss 28.738384\n",
      "Epoch 1234, Loss 28.737507\n",
      "Epoch 1235, Loss 28.736631\n",
      "Epoch 1236, Loss 28.735754\n",
      "Epoch 1237, Loss 28.734877\n",
      "Epoch 1238, Loss 28.734001\n",
      "Epoch 1239, Loss 28.733122\n",
      "Epoch 1240, Loss 28.732244\n",
      "Epoch 1241, Loss 28.731367\n",
      "Epoch 1242, Loss 28.730497\n",
      "Epoch 1243, Loss 28.729612\n",
      "Epoch 1244, Loss 28.728741\n",
      "Epoch 1245, Loss 28.727858\n",
      "Epoch 1246, Loss 28.726988\n",
      "Epoch 1247, Loss 28.726110\n",
      "Epoch 1248, Loss 28.725237\n",
      "Epoch 1249, Loss 28.724354\n",
      "Epoch 1250, Loss 28.723480\n",
      "Epoch 1251, Loss 28.722601\n",
      "Epoch 1252, Loss 28.721727\n",
      "Epoch 1253, Loss 28.720848\n",
      "Epoch 1254, Loss 28.719973\n",
      "Epoch 1255, Loss 28.719093\n",
      "Epoch 1256, Loss 28.718224\n",
      "Epoch 1257, Loss 28.717344\n",
      "Epoch 1258, Loss 28.716469\n",
      "Epoch 1259, Loss 28.715593\n",
      "Epoch 1260, Loss 28.714720\n",
      "Epoch 1261, Loss 28.713839\n",
      "Epoch 1262, Loss 28.712965\n",
      "Epoch 1263, Loss 28.712084\n",
      "Epoch 1264, Loss 28.711210\n",
      "Epoch 1265, Loss 28.710333\n",
      "Epoch 1266, Loss 28.709455\n",
      "Epoch 1267, Loss 28.708580\n",
      "Epoch 1268, Loss 28.707703\n",
      "Epoch 1269, Loss 28.706825\n",
      "Epoch 1270, Loss 28.705950\n",
      "Epoch 1271, Loss 28.705076\n",
      "Epoch 1272, Loss 28.704199\n",
      "Epoch 1273, Loss 28.703325\n",
      "Epoch 1274, Loss 28.702448\n",
      "Epoch 1275, Loss 28.701571\n",
      "Epoch 1276, Loss 28.700701\n",
      "Epoch 1277, Loss 28.699823\n",
      "Epoch 1278, Loss 28.698946\n",
      "Epoch 1279, Loss 28.698072\n",
      "Epoch 1280, Loss 28.697193\n",
      "Epoch 1281, Loss 28.696321\n",
      "Epoch 1282, Loss 28.695442\n",
      "Epoch 1283, Loss 28.694569\n",
      "Epoch 1284, Loss 28.693693\n",
      "Epoch 1285, Loss 28.692816\n",
      "Epoch 1286, Loss 28.691942\n",
      "Epoch 1287, Loss 28.691065\n",
      "Epoch 1288, Loss 28.690191\n",
      "Epoch 1289, Loss 28.689314\n",
      "Epoch 1290, Loss 28.688438\n",
      "Epoch 1291, Loss 28.687567\n",
      "Epoch 1292, Loss 28.686689\n",
      "Epoch 1293, Loss 28.685814\n",
      "Epoch 1294, Loss 28.684942\n",
      "Epoch 1295, Loss 28.684065\n",
      "Epoch 1296, Loss 28.683189\n",
      "Epoch 1297, Loss 28.682312\n",
      "Epoch 1298, Loss 28.681435\n",
      "Epoch 1299, Loss 28.680565\n",
      "Epoch 1300, Loss 28.679688\n",
      "Epoch 1301, Loss 28.678814\n",
      "Epoch 1302, Loss 28.677940\n",
      "Epoch 1303, Loss 28.677063\n",
      "Epoch 1304, Loss 28.676186\n",
      "Epoch 1305, Loss 28.675318\n",
      "Epoch 1306, Loss 28.674438\n",
      "Epoch 1307, Loss 28.673561\n",
      "Epoch 1308, Loss 28.672686\n",
      "Epoch 1309, Loss 28.671814\n",
      "Epoch 1310, Loss 28.670940\n",
      "Epoch 1311, Loss 28.670067\n",
      "Epoch 1312, Loss 28.669191\n",
      "Epoch 1313, Loss 28.668312\n",
      "Epoch 1314, Loss 28.667442\n",
      "Epoch 1315, Loss 28.666567\n",
      "Epoch 1316, Loss 28.665688\n",
      "Epoch 1317, Loss 28.664818\n",
      "Epoch 1318, Loss 28.663940\n",
      "Epoch 1319, Loss 28.663069\n",
      "Epoch 1320, Loss 28.662193\n",
      "Epoch 1321, Loss 28.661318\n",
      "Epoch 1322, Loss 28.660444\n",
      "Epoch 1323, Loss 28.659569\n",
      "Epoch 1324, Loss 28.658697\n",
      "Epoch 1325, Loss 28.657820\n",
      "Epoch 1326, Loss 28.656952\n",
      "Epoch 1327, Loss 28.656069\n",
      "Epoch 1328, Loss 28.655201\n",
      "Epoch 1329, Loss 28.654325\n",
      "Epoch 1330, Loss 28.653448\n",
      "Epoch 1331, Loss 28.652580\n",
      "Epoch 1332, Loss 28.651703\n",
      "Epoch 1333, Loss 28.650831\n",
      "Epoch 1334, Loss 28.649958\n",
      "Epoch 1335, Loss 28.649076\n",
      "Epoch 1336, Loss 28.648205\n",
      "Epoch 1337, Loss 28.647331\n",
      "Epoch 1338, Loss 28.646460\n",
      "Epoch 1339, Loss 28.645588\n",
      "Epoch 1340, Loss 28.644714\n",
      "Epoch 1341, Loss 28.643837\n",
      "Epoch 1342, Loss 28.642963\n",
      "Epoch 1343, Loss 28.642090\n",
      "Epoch 1344, Loss 28.641216\n",
      "Epoch 1345, Loss 28.640343\n",
      "Epoch 1346, Loss 28.639471\n",
      "Epoch 1347, Loss 28.638594\n",
      "Epoch 1348, Loss 28.637724\n",
      "Epoch 1349, Loss 28.636848\n",
      "Epoch 1350, Loss 28.635975\n",
      "Epoch 1351, Loss 28.635101\n",
      "Epoch 1352, Loss 28.634224\n",
      "Epoch 1353, Loss 28.633356\n",
      "Epoch 1354, Loss 28.632479\n",
      "Epoch 1355, Loss 28.631609\n",
      "Epoch 1356, Loss 28.630735\n",
      "Epoch 1357, Loss 28.629860\n",
      "Epoch 1358, Loss 28.628992\n",
      "Epoch 1359, Loss 28.628113\n",
      "Epoch 1360, Loss 28.627241\n",
      "Epoch 1361, Loss 28.626371\n",
      "Epoch 1362, Loss 28.625494\n",
      "Epoch 1363, Loss 28.624620\n",
      "Epoch 1364, Loss 28.623751\n",
      "Epoch 1365, Loss 28.622875\n",
      "Epoch 1366, Loss 28.622004\n",
      "Epoch 1367, Loss 28.621132\n",
      "Epoch 1368, Loss 28.620256\n",
      "Epoch 1369, Loss 28.619385\n",
      "Epoch 1370, Loss 28.618511\n",
      "Epoch 1371, Loss 28.617638\n",
      "Epoch 1372, Loss 28.616762\n",
      "Epoch 1373, Loss 28.615894\n",
      "Epoch 1374, Loss 28.615019\n",
      "Epoch 1375, Loss 28.614149\n",
      "Epoch 1376, Loss 28.613276\n",
      "Epoch 1377, Loss 28.612404\n",
      "Epoch 1378, Loss 28.611528\n",
      "Epoch 1379, Loss 28.610657\n",
      "Epoch 1380, Loss 28.609785\n",
      "Epoch 1381, Loss 28.608912\n",
      "Epoch 1382, Loss 28.608038\n",
      "Epoch 1383, Loss 28.607170\n",
      "Epoch 1384, Loss 28.606295\n",
      "Epoch 1385, Loss 28.605421\n",
      "Epoch 1386, Loss 28.604548\n",
      "Epoch 1387, Loss 28.603680\n",
      "Epoch 1388, Loss 28.602802\n",
      "Epoch 1389, Loss 28.601931\n",
      "Epoch 1390, Loss 28.601057\n",
      "Epoch 1391, Loss 28.600187\n",
      "Epoch 1392, Loss 28.599316\n",
      "Epoch 1393, Loss 28.598442\n",
      "Epoch 1394, Loss 28.597567\n",
      "Epoch 1395, Loss 28.596699\n",
      "Epoch 1396, Loss 28.595825\n",
      "Epoch 1397, Loss 28.594957\n",
      "Epoch 1398, Loss 28.594086\n",
      "Epoch 1399, Loss 28.593208\n",
      "Epoch 1400, Loss 28.592339\n",
      "Epoch 1401, Loss 28.591463\n",
      "Epoch 1402, Loss 28.590595\n",
      "Epoch 1403, Loss 28.589722\n",
      "Epoch 1404, Loss 28.588848\n",
      "Epoch 1405, Loss 28.587980\n",
      "Epoch 1406, Loss 28.587103\n",
      "Epoch 1407, Loss 28.586231\n",
      "Epoch 1408, Loss 28.585363\n",
      "Epoch 1409, Loss 28.584490\n",
      "Epoch 1410, Loss 28.583618\n",
      "Epoch 1411, Loss 28.582747\n",
      "Epoch 1412, Loss 28.581877\n",
      "Epoch 1413, Loss 28.581007\n",
      "Epoch 1414, Loss 28.580132\n",
      "Epoch 1415, Loss 28.579262\n",
      "Epoch 1416, Loss 28.578392\n",
      "Epoch 1417, Loss 28.577515\n",
      "Epoch 1418, Loss 28.576643\n",
      "Epoch 1419, Loss 28.575775\n",
      "Epoch 1420, Loss 28.574903\n",
      "Epoch 1421, Loss 28.574030\n",
      "Epoch 1422, Loss 28.573162\n",
      "Epoch 1423, Loss 28.572289\n",
      "Epoch 1424, Loss 28.571419\n",
      "Epoch 1425, Loss 28.570545\n",
      "Epoch 1426, Loss 28.569674\n",
      "Epoch 1427, Loss 28.568806\n",
      "Epoch 1428, Loss 28.567932\n",
      "Epoch 1429, Loss 28.567066\n",
      "Epoch 1430, Loss 28.566191\n",
      "Epoch 1431, Loss 28.565321\n",
      "Epoch 1432, Loss 28.564447\n",
      "Epoch 1433, Loss 28.563581\n",
      "Epoch 1434, Loss 28.562706\n",
      "Epoch 1435, Loss 28.561836\n",
      "Epoch 1436, Loss 28.560963\n",
      "Epoch 1437, Loss 28.560097\n",
      "Epoch 1438, Loss 28.559223\n",
      "Epoch 1439, Loss 28.558352\n",
      "Epoch 1440, Loss 28.557482\n",
      "Epoch 1441, Loss 28.556612\n",
      "Epoch 1442, Loss 28.555738\n",
      "Epoch 1443, Loss 28.554869\n",
      "Epoch 1444, Loss 28.553997\n",
      "Epoch 1445, Loss 28.553125\n",
      "Epoch 1446, Loss 28.552259\n",
      "Epoch 1447, Loss 28.551386\n",
      "Epoch 1448, Loss 28.550514\n",
      "Epoch 1449, Loss 28.549644\n",
      "Epoch 1450, Loss 28.548779\n",
      "Epoch 1451, Loss 28.547907\n",
      "Epoch 1452, Loss 28.547033\n",
      "Epoch 1453, Loss 28.546160\n",
      "Epoch 1454, Loss 28.545294\n",
      "Epoch 1455, Loss 28.544426\n",
      "Epoch 1456, Loss 28.543550\n",
      "Epoch 1457, Loss 28.542683\n",
      "Epoch 1458, Loss 28.541811\n",
      "Epoch 1459, Loss 28.540947\n",
      "Epoch 1460, Loss 28.540071\n",
      "Epoch 1461, Loss 28.539202\n",
      "Epoch 1462, Loss 28.538332\n",
      "Epoch 1463, Loss 28.537462\n",
      "Epoch 1464, Loss 28.536592\n",
      "Epoch 1465, Loss 28.535719\n",
      "Epoch 1466, Loss 28.534851\n",
      "Epoch 1467, Loss 28.533983\n",
      "Epoch 1468, Loss 28.533113\n",
      "Epoch 1469, Loss 28.532238\n",
      "Epoch 1470, Loss 28.531372\n",
      "Epoch 1471, Loss 28.530500\n",
      "Epoch 1472, Loss 28.529633\n",
      "Epoch 1473, Loss 28.528761\n",
      "Epoch 1474, Loss 28.527887\n",
      "Epoch 1475, Loss 28.527025\n",
      "Epoch 1476, Loss 28.526154\n",
      "Epoch 1477, Loss 28.525282\n",
      "Epoch 1478, Loss 28.524414\n",
      "Epoch 1479, Loss 28.523542\n",
      "Epoch 1480, Loss 28.522669\n",
      "Epoch 1481, Loss 28.521807\n",
      "Epoch 1482, Loss 28.520935\n",
      "Epoch 1483, Loss 28.520067\n",
      "Epoch 1484, Loss 28.519196\n",
      "Epoch 1485, Loss 28.518324\n",
      "Epoch 1486, Loss 28.517456\n",
      "Epoch 1487, Loss 28.516584\n",
      "Epoch 1488, Loss 28.515718\n",
      "Epoch 1489, Loss 28.514849\n",
      "Epoch 1490, Loss 28.513979\n",
      "Epoch 1491, Loss 28.513109\n",
      "Epoch 1492, Loss 28.512243\n",
      "Epoch 1493, Loss 28.511370\n",
      "Epoch 1494, Loss 28.510498\n",
      "Epoch 1495, Loss 28.509636\n",
      "Epoch 1496, Loss 28.508764\n",
      "Epoch 1497, Loss 28.507896\n",
      "Epoch 1498, Loss 28.507027\n",
      "Epoch 1499, Loss 28.506157\n",
      "Epoch 1500, Loss 28.505281\n",
      "Epoch 1501, Loss 28.504417\n",
      "Epoch 1502, Loss 28.503548\n",
      "Epoch 1503, Loss 28.502684\n",
      "Epoch 1504, Loss 28.501812\n",
      "Epoch 1505, Loss 28.500944\n",
      "Epoch 1506, Loss 28.500072\n",
      "Epoch 1507, Loss 28.499207\n",
      "Epoch 1508, Loss 28.498339\n",
      "Epoch 1509, Loss 28.497467\n",
      "Epoch 1510, Loss 28.496595\n",
      "Epoch 1511, Loss 28.495728\n",
      "Epoch 1512, Loss 28.494862\n",
      "Epoch 1513, Loss 28.493994\n",
      "Epoch 1514, Loss 28.493126\n",
      "Epoch 1515, Loss 28.492254\n",
      "Epoch 1516, Loss 28.491388\n",
      "Epoch 1517, Loss 28.490515\n",
      "Epoch 1518, Loss 28.489653\n",
      "Epoch 1519, Loss 28.488783\n",
      "Epoch 1520, Loss 28.487913\n",
      "Epoch 1521, Loss 28.487043\n",
      "Epoch 1522, Loss 28.486174\n",
      "Epoch 1523, Loss 28.485308\n",
      "Epoch 1524, Loss 28.484442\n",
      "Epoch 1525, Loss 28.483576\n",
      "Epoch 1526, Loss 28.482702\n",
      "Epoch 1527, Loss 28.481834\n",
      "Epoch 1528, Loss 28.480968\n",
      "Epoch 1529, Loss 28.480101\n",
      "Epoch 1530, Loss 28.479231\n",
      "Epoch 1531, Loss 28.478363\n",
      "Epoch 1532, Loss 28.477489\n",
      "Epoch 1533, Loss 28.476624\n",
      "Epoch 1534, Loss 28.475758\n",
      "Epoch 1535, Loss 28.474892\n",
      "Epoch 1536, Loss 28.474022\n",
      "Epoch 1537, Loss 28.473150\n",
      "Epoch 1538, Loss 28.472284\n",
      "Epoch 1539, Loss 28.471418\n",
      "Epoch 1540, Loss 28.470549\n",
      "Epoch 1541, Loss 28.469683\n",
      "Epoch 1542, Loss 28.468813\n",
      "Epoch 1543, Loss 28.467943\n",
      "Epoch 1544, Loss 28.467079\n",
      "Epoch 1545, Loss 28.466209\n",
      "Epoch 1546, Loss 28.465340\n",
      "Epoch 1547, Loss 28.464478\n",
      "Epoch 1548, Loss 28.463612\n",
      "Epoch 1549, Loss 28.462740\n",
      "Epoch 1550, Loss 28.461872\n",
      "Epoch 1551, Loss 28.461010\n",
      "Epoch 1552, Loss 28.460138\n",
      "Epoch 1553, Loss 28.459272\n",
      "Epoch 1554, Loss 28.458406\n",
      "Epoch 1555, Loss 28.457537\n",
      "Epoch 1556, Loss 28.456671\n",
      "Epoch 1557, Loss 28.455803\n",
      "Epoch 1558, Loss 28.454933\n",
      "Epoch 1559, Loss 28.454071\n",
      "Epoch 1560, Loss 28.453203\n",
      "Epoch 1561, Loss 28.452333\n",
      "Epoch 1562, Loss 28.451466\n",
      "Epoch 1563, Loss 28.450598\n",
      "Epoch 1564, Loss 28.449736\n",
      "Epoch 1565, Loss 28.448864\n",
      "Epoch 1566, Loss 28.447996\n",
      "Epoch 1567, Loss 28.447130\n",
      "Epoch 1568, Loss 28.446259\n",
      "Epoch 1569, Loss 28.445391\n",
      "Epoch 1570, Loss 28.444525\n",
      "Epoch 1571, Loss 28.443659\n",
      "Epoch 1572, Loss 28.442793\n",
      "Epoch 1573, Loss 28.441927\n",
      "Epoch 1574, Loss 28.441063\n",
      "Epoch 1575, Loss 28.440197\n",
      "Epoch 1576, Loss 28.439329\n",
      "Epoch 1577, Loss 28.438459\n",
      "Epoch 1578, Loss 28.437593\n",
      "Epoch 1579, Loss 28.436726\n",
      "Epoch 1580, Loss 28.435863\n",
      "Epoch 1581, Loss 28.434996\n",
      "Epoch 1582, Loss 28.434124\n",
      "Epoch 1583, Loss 28.433264\n",
      "Epoch 1584, Loss 28.432396\n",
      "Epoch 1585, Loss 28.431530\n",
      "Epoch 1586, Loss 28.430664\n",
      "Epoch 1587, Loss 28.429790\n",
      "Epoch 1588, Loss 28.428926\n",
      "Epoch 1589, Loss 28.428062\n",
      "Epoch 1590, Loss 28.427197\n",
      "Epoch 1591, Loss 28.426331\n",
      "Epoch 1592, Loss 28.425463\n",
      "Epoch 1593, Loss 28.424599\n",
      "Epoch 1594, Loss 28.423729\n",
      "Epoch 1595, Loss 28.422859\n",
      "Epoch 1596, Loss 28.421997\n",
      "Epoch 1597, Loss 28.421131\n",
      "Epoch 1598, Loss 28.420269\n",
      "Epoch 1599, Loss 28.419401\n",
      "Epoch 1600, Loss 28.418530\n",
      "Epoch 1601, Loss 28.417664\n",
      "Epoch 1602, Loss 28.416803\n",
      "Epoch 1603, Loss 28.415937\n",
      "Epoch 1604, Loss 28.415070\n",
      "Epoch 1605, Loss 28.414202\n",
      "Epoch 1606, Loss 28.413342\n",
      "Epoch 1607, Loss 28.412470\n",
      "Epoch 1608, Loss 28.411604\n",
      "Epoch 1609, Loss 28.410736\n",
      "Epoch 1610, Loss 28.409874\n",
      "Epoch 1611, Loss 28.409010\n",
      "Epoch 1612, Loss 28.408144\n",
      "Epoch 1613, Loss 28.407276\n",
      "Epoch 1614, Loss 28.406410\n",
      "Epoch 1615, Loss 28.405542\n",
      "Epoch 1616, Loss 28.404676\n",
      "Epoch 1617, Loss 28.403814\n",
      "Epoch 1618, Loss 28.402946\n",
      "Epoch 1619, Loss 28.402082\n",
      "Epoch 1620, Loss 28.401217\n",
      "Epoch 1621, Loss 28.400352\n",
      "Epoch 1622, Loss 28.399487\n",
      "Epoch 1623, Loss 28.398619\n",
      "Epoch 1624, Loss 28.397755\n",
      "Epoch 1625, Loss 28.396889\n",
      "Epoch 1626, Loss 28.396027\n",
      "Epoch 1627, Loss 28.395164\n",
      "Epoch 1628, Loss 28.394299\n",
      "Epoch 1629, Loss 28.393427\n",
      "Epoch 1630, Loss 28.392567\n",
      "Epoch 1631, Loss 28.391699\n",
      "Epoch 1632, Loss 28.390837\n",
      "Epoch 1633, Loss 28.389973\n",
      "Epoch 1634, Loss 28.389107\n",
      "Epoch 1635, Loss 28.388237\n",
      "Epoch 1636, Loss 28.387379\n",
      "Epoch 1637, Loss 28.386505\n",
      "Epoch 1638, Loss 28.385643\n",
      "Epoch 1639, Loss 28.384783\n",
      "Epoch 1640, Loss 28.383913\n",
      "Epoch 1641, Loss 28.383049\n",
      "Epoch 1642, Loss 28.382185\n",
      "Epoch 1643, Loss 28.381319\n",
      "Epoch 1644, Loss 28.380457\n",
      "Epoch 1645, Loss 28.379589\n",
      "Epoch 1646, Loss 28.378723\n",
      "Epoch 1647, Loss 28.377861\n",
      "Epoch 1648, Loss 28.376997\n",
      "Epoch 1649, Loss 28.376131\n",
      "Epoch 1650, Loss 28.375263\n",
      "Epoch 1651, Loss 28.374407\n",
      "Epoch 1652, Loss 28.373537\n",
      "Epoch 1653, Loss 28.372675\n",
      "Epoch 1654, Loss 28.371807\n",
      "Epoch 1655, Loss 28.370943\n",
      "Epoch 1656, Loss 28.370079\n",
      "Epoch 1657, Loss 28.369219\n",
      "Epoch 1658, Loss 28.368349\n",
      "Epoch 1659, Loss 28.367485\n",
      "Epoch 1660, Loss 28.366621\n",
      "Epoch 1661, Loss 28.365761\n",
      "Epoch 1662, Loss 28.364893\n",
      "Epoch 1663, Loss 28.364031\n",
      "Epoch 1664, Loss 28.363165\n",
      "Epoch 1665, Loss 28.362299\n",
      "Epoch 1666, Loss 28.361433\n",
      "Epoch 1667, Loss 28.360573\n",
      "Epoch 1668, Loss 28.359705\n",
      "Epoch 1669, Loss 28.358845\n",
      "Epoch 1670, Loss 28.357983\n",
      "Epoch 1671, Loss 28.357119\n",
      "Epoch 1672, Loss 28.356251\n",
      "Epoch 1673, Loss 28.355391\n",
      "Epoch 1674, Loss 28.354523\n",
      "Epoch 1675, Loss 28.353661\n",
      "Epoch 1676, Loss 28.352800\n",
      "Epoch 1677, Loss 28.351931\n",
      "Epoch 1678, Loss 28.351068\n",
      "Epoch 1679, Loss 28.350203\n",
      "Epoch 1680, Loss 28.349346\n",
      "Epoch 1681, Loss 28.348478\n",
      "Epoch 1682, Loss 28.347618\n",
      "Epoch 1683, Loss 28.346752\n",
      "Epoch 1684, Loss 28.345884\n",
      "Epoch 1685, Loss 28.345024\n",
      "Epoch 1686, Loss 28.344160\n",
      "Epoch 1687, Loss 28.343294\n",
      "Epoch 1688, Loss 28.342436\n",
      "Epoch 1689, Loss 28.341570\n",
      "Epoch 1690, Loss 28.340710\n",
      "Epoch 1691, Loss 28.339842\n",
      "Epoch 1692, Loss 28.338978\n",
      "Epoch 1693, Loss 28.338116\n",
      "Epoch 1694, Loss 28.337250\n",
      "Epoch 1695, Loss 28.336390\n",
      "Epoch 1696, Loss 28.335529\n",
      "Epoch 1697, Loss 28.334658\n",
      "Epoch 1698, Loss 28.333799\n",
      "Epoch 1699, Loss 28.332941\n",
      "Epoch 1700, Loss 28.332069\n",
      "Epoch 1701, Loss 28.331209\n",
      "Epoch 1702, Loss 28.330347\n",
      "Epoch 1703, Loss 28.329485\n",
      "Epoch 1704, Loss 28.328619\n",
      "Epoch 1705, Loss 28.327761\n",
      "Epoch 1706, Loss 28.326899\n",
      "Epoch 1707, Loss 28.326036\n",
      "Epoch 1708, Loss 28.325171\n",
      "Epoch 1709, Loss 28.324305\n",
      "Epoch 1710, Loss 28.323444\n",
      "Epoch 1711, Loss 28.322578\n",
      "Epoch 1712, Loss 28.321718\n",
      "Epoch 1713, Loss 28.320856\n",
      "Epoch 1714, Loss 28.319994\n",
      "Epoch 1715, Loss 28.319130\n",
      "Epoch 1716, Loss 28.318266\n",
      "Epoch 1717, Loss 28.317406\n",
      "Epoch 1718, Loss 28.316545\n",
      "Epoch 1719, Loss 28.315676\n",
      "Epoch 1720, Loss 28.314817\n",
      "Epoch 1721, Loss 28.313953\n",
      "Epoch 1722, Loss 28.313091\n",
      "Epoch 1723, Loss 28.312227\n",
      "Epoch 1724, Loss 28.311363\n",
      "Epoch 1725, Loss 28.310509\n",
      "Epoch 1726, Loss 28.309645\n",
      "Epoch 1727, Loss 28.308783\n",
      "Epoch 1728, Loss 28.307917\n",
      "Epoch 1729, Loss 28.307055\n",
      "Epoch 1730, Loss 28.306190\n",
      "Epoch 1731, Loss 28.305334\n",
      "Epoch 1732, Loss 28.304468\n",
      "Epoch 1733, Loss 28.303612\n",
      "Epoch 1734, Loss 28.302746\n",
      "Epoch 1735, Loss 28.301886\n",
      "Epoch 1736, Loss 28.301020\n",
      "Epoch 1737, Loss 28.300159\n",
      "Epoch 1738, Loss 28.299294\n",
      "Epoch 1739, Loss 28.298433\n",
      "Epoch 1740, Loss 28.297575\n",
      "Epoch 1741, Loss 28.296711\n",
      "Epoch 1742, Loss 28.295849\n",
      "Epoch 1743, Loss 28.294989\n",
      "Epoch 1744, Loss 28.294128\n",
      "Epoch 1745, Loss 28.293262\n",
      "Epoch 1746, Loss 28.292402\n",
      "Epoch 1747, Loss 28.291540\n",
      "Epoch 1748, Loss 28.290678\n",
      "Epoch 1749, Loss 28.289818\n",
      "Epoch 1750, Loss 28.288958\n",
      "Epoch 1751, Loss 28.288097\n",
      "Epoch 1752, Loss 28.287233\n",
      "Epoch 1753, Loss 28.286371\n",
      "Epoch 1754, Loss 28.285505\n",
      "Epoch 1755, Loss 28.284645\n",
      "Epoch 1756, Loss 28.283781\n",
      "Epoch 1757, Loss 28.282923\n",
      "Epoch 1758, Loss 28.282061\n",
      "Epoch 1759, Loss 28.281200\n",
      "Epoch 1760, Loss 28.280340\n",
      "Epoch 1761, Loss 28.279474\n",
      "Epoch 1762, Loss 28.278614\n",
      "Epoch 1763, Loss 28.277754\n",
      "Epoch 1764, Loss 28.276892\n",
      "Epoch 1765, Loss 28.276033\n",
      "Epoch 1766, Loss 28.275169\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1767, Loss 28.274309\n",
      "Epoch 1768, Loss 28.273449\n",
      "Epoch 1769, Loss 28.272589\n",
      "Epoch 1770, Loss 28.271727\n",
      "Epoch 1771, Loss 28.270866\n",
      "Epoch 1772, Loss 28.270006\n",
      "Epoch 1773, Loss 28.269146\n",
      "Epoch 1774, Loss 28.268282\n",
      "Epoch 1775, Loss 28.267424\n",
      "Epoch 1776, Loss 28.266558\n",
      "Epoch 1777, Loss 28.265703\n",
      "Epoch 1778, Loss 28.264839\n",
      "Epoch 1779, Loss 28.263979\n",
      "Epoch 1780, Loss 28.263121\n",
      "Epoch 1781, Loss 28.262260\n",
      "Epoch 1782, Loss 28.261396\n",
      "Epoch 1783, Loss 28.260536\n",
      "Epoch 1784, Loss 28.259676\n",
      "Epoch 1785, Loss 28.258812\n",
      "Epoch 1786, Loss 28.257952\n",
      "Epoch 1787, Loss 28.257092\n",
      "Epoch 1788, Loss 28.256231\n",
      "Epoch 1789, Loss 28.255369\n",
      "Epoch 1790, Loss 28.254515\n",
      "Epoch 1791, Loss 28.253654\n",
      "Epoch 1792, Loss 28.252790\n",
      "Epoch 1793, Loss 28.251930\n",
      "Epoch 1794, Loss 28.251066\n",
      "Epoch 1795, Loss 28.250208\n",
      "Epoch 1796, Loss 28.249348\n",
      "Epoch 1797, Loss 28.248491\n",
      "Epoch 1798, Loss 28.247627\n",
      "Epoch 1799, Loss 28.246765\n",
      "Epoch 1800, Loss 28.245905\n",
      "Epoch 1801, Loss 28.245045\n",
      "Epoch 1802, Loss 28.244190\n",
      "Epoch 1803, Loss 28.243328\n",
      "Epoch 1804, Loss 28.242462\n",
      "Epoch 1805, Loss 28.241606\n",
      "Epoch 1806, Loss 28.240746\n",
      "Epoch 1807, Loss 28.239885\n",
      "Epoch 1808, Loss 28.239025\n",
      "Epoch 1809, Loss 28.238165\n",
      "Epoch 1810, Loss 28.237305\n",
      "Epoch 1811, Loss 28.236448\n",
      "Epoch 1812, Loss 28.235590\n",
      "Epoch 1813, Loss 28.234730\n",
      "Epoch 1814, Loss 28.233870\n",
      "Epoch 1815, Loss 28.233009\n",
      "Epoch 1816, Loss 28.232147\n",
      "Epoch 1817, Loss 28.231291\n",
      "Epoch 1818, Loss 28.230427\n",
      "Epoch 1819, Loss 28.229567\n",
      "Epoch 1820, Loss 28.228712\n",
      "Epoch 1821, Loss 28.227850\n",
      "Epoch 1822, Loss 28.226992\n",
      "Epoch 1823, Loss 28.226133\n",
      "Epoch 1824, Loss 28.225269\n",
      "Epoch 1825, Loss 28.224409\n",
      "Epoch 1826, Loss 28.223555\n",
      "Epoch 1827, Loss 28.222689\n",
      "Epoch 1828, Loss 28.221830\n",
      "Epoch 1829, Loss 28.220972\n",
      "Epoch 1830, Loss 28.220116\n",
      "Epoch 1831, Loss 28.219252\n",
      "Epoch 1832, Loss 28.218395\n",
      "Epoch 1833, Loss 28.217535\n",
      "Epoch 1834, Loss 28.216675\n",
      "Epoch 1835, Loss 28.215818\n",
      "Epoch 1836, Loss 28.214960\n",
      "Epoch 1837, Loss 28.214098\n",
      "Epoch 1838, Loss 28.213243\n",
      "Epoch 1839, Loss 28.212379\n",
      "Epoch 1840, Loss 28.211519\n",
      "Epoch 1841, Loss 28.210665\n",
      "Epoch 1842, Loss 28.209806\n",
      "Epoch 1843, Loss 28.208942\n",
      "Epoch 1844, Loss 28.208092\n",
      "Epoch 1845, Loss 28.207228\n",
      "Epoch 1846, Loss 28.206371\n",
      "Epoch 1847, Loss 28.205511\n",
      "Epoch 1848, Loss 28.204651\n",
      "Epoch 1849, Loss 28.203794\n",
      "Epoch 1850, Loss 28.202936\n",
      "Epoch 1851, Loss 28.202076\n",
      "Epoch 1852, Loss 28.201216\n",
      "Epoch 1853, Loss 28.200356\n",
      "Epoch 1854, Loss 28.199499\n",
      "Epoch 1855, Loss 28.198641\n",
      "Epoch 1856, Loss 28.197783\n",
      "Epoch 1857, Loss 28.196922\n",
      "Epoch 1858, Loss 28.196068\n",
      "Epoch 1859, Loss 28.195210\n",
      "Epoch 1860, Loss 28.194353\n",
      "Epoch 1861, Loss 28.193489\n",
      "Epoch 1862, Loss 28.192638\n",
      "Epoch 1863, Loss 28.191774\n",
      "Epoch 1864, Loss 28.190916\n",
      "Epoch 1865, Loss 28.190056\n",
      "Epoch 1866, Loss 28.189198\n",
      "Epoch 1867, Loss 28.188341\n",
      "Epoch 1868, Loss 28.187483\n",
      "Epoch 1869, Loss 28.186621\n",
      "Epoch 1870, Loss 28.185768\n",
      "Epoch 1871, Loss 28.184908\n",
      "Epoch 1872, Loss 28.184052\n",
      "Epoch 1873, Loss 28.183191\n",
      "Epoch 1874, Loss 28.182335\n",
      "Epoch 1875, Loss 28.181477\n",
      "Epoch 1876, Loss 28.180616\n",
      "Epoch 1877, Loss 28.179760\n",
      "Epoch 1878, Loss 28.178900\n",
      "Epoch 1879, Loss 28.178045\n",
      "Epoch 1880, Loss 28.177185\n",
      "Epoch 1881, Loss 28.176331\n",
      "Epoch 1882, Loss 28.175468\n",
      "Epoch 1883, Loss 28.174610\n",
      "Epoch 1884, Loss 28.173759\n",
      "Epoch 1885, Loss 28.172901\n",
      "Epoch 1886, Loss 28.172041\n",
      "Epoch 1887, Loss 28.171185\n",
      "Epoch 1888, Loss 28.170321\n",
      "Epoch 1889, Loss 28.169466\n",
      "Epoch 1890, Loss 28.168610\n",
      "Epoch 1891, Loss 28.167753\n",
      "Epoch 1892, Loss 28.166895\n",
      "Epoch 1893, Loss 28.166039\n",
      "Epoch 1894, Loss 28.165178\n",
      "Epoch 1895, Loss 28.164324\n",
      "Epoch 1896, Loss 28.163465\n",
      "Epoch 1897, Loss 28.162605\n",
      "Epoch 1898, Loss 28.161751\n",
      "Epoch 1899, Loss 28.160894\n",
      "Epoch 1900, Loss 28.160034\n",
      "Epoch 1901, Loss 28.159180\n",
      "Epoch 1902, Loss 28.158319\n",
      "Epoch 1903, Loss 28.157463\n",
      "Epoch 1904, Loss 28.156605\n",
      "Epoch 1905, Loss 28.155750\n",
      "Epoch 1906, Loss 28.154890\n",
      "Epoch 1907, Loss 28.154039\n",
      "Epoch 1908, Loss 28.153179\n",
      "Epoch 1909, Loss 28.152321\n",
      "Epoch 1910, Loss 28.151464\n",
      "Epoch 1911, Loss 28.150612\n",
      "Epoch 1912, Loss 28.149750\n",
      "Epoch 1913, Loss 28.148893\n",
      "Epoch 1914, Loss 28.148033\n",
      "Epoch 1915, Loss 28.147184\n",
      "Epoch 1916, Loss 28.146324\n",
      "Epoch 1917, Loss 28.145466\n",
      "Epoch 1918, Loss 28.144609\n",
      "Epoch 1919, Loss 28.143755\n",
      "Epoch 1920, Loss 28.142895\n",
      "Epoch 1921, Loss 28.142040\n",
      "Epoch 1922, Loss 28.141182\n",
      "Epoch 1923, Loss 28.140326\n",
      "Epoch 1924, Loss 28.139467\n",
      "Epoch 1925, Loss 28.138613\n",
      "Epoch 1926, Loss 28.137756\n",
      "Epoch 1927, Loss 28.136902\n",
      "Epoch 1928, Loss 28.136042\n",
      "Epoch 1929, Loss 28.135185\n",
      "Epoch 1930, Loss 28.134327\n",
      "Epoch 1931, Loss 28.133472\n",
      "Epoch 1932, Loss 28.132616\n",
      "Epoch 1933, Loss 28.131762\n",
      "Epoch 1934, Loss 28.130903\n",
      "Epoch 1935, Loss 28.130053\n",
      "Epoch 1936, Loss 28.129192\n",
      "Epoch 1937, Loss 28.128334\n",
      "Epoch 1938, Loss 28.127480\n",
      "Epoch 1939, Loss 28.126623\n",
      "Epoch 1940, Loss 28.125765\n",
      "Epoch 1941, Loss 28.124910\n",
      "Epoch 1942, Loss 28.124054\n",
      "Epoch 1943, Loss 28.123201\n",
      "Epoch 1944, Loss 28.122343\n",
      "Epoch 1945, Loss 28.121489\n",
      "Epoch 1946, Loss 28.120634\n",
      "Epoch 1947, Loss 28.119776\n",
      "Epoch 1948, Loss 28.118916\n",
      "Epoch 1949, Loss 28.118059\n",
      "Epoch 1950, Loss 28.117205\n",
      "Epoch 1951, Loss 28.116350\n",
      "Epoch 1952, Loss 28.115496\n",
      "Epoch 1953, Loss 28.114637\n",
      "Epoch 1954, Loss 28.113783\n",
      "Epoch 1955, Loss 28.112926\n",
      "Epoch 1956, Loss 28.112074\n",
      "Epoch 1957, Loss 28.111217\n",
      "Epoch 1958, Loss 28.110365\n",
      "Epoch 1959, Loss 28.109509\n",
      "Epoch 1960, Loss 28.108650\n",
      "Epoch 1961, Loss 28.107796\n",
      "Epoch 1962, Loss 28.106939\n",
      "Epoch 1963, Loss 28.106085\n",
      "Epoch 1964, Loss 28.105230\n",
      "Epoch 1965, Loss 28.104376\n",
      "Epoch 1966, Loss 28.103521\n",
      "Epoch 1967, Loss 28.102661\n",
      "Epoch 1968, Loss 28.101809\n",
      "Epoch 1969, Loss 28.100952\n",
      "Epoch 1970, Loss 28.100098\n",
      "Epoch 1971, Loss 28.099241\n",
      "Epoch 1972, Loss 28.098381\n",
      "Epoch 1973, Loss 28.097528\n",
      "Epoch 1974, Loss 28.096668\n",
      "Epoch 1975, Loss 28.095818\n",
      "Epoch 1976, Loss 28.094965\n",
      "Epoch 1977, Loss 28.094109\n",
      "Epoch 1978, Loss 28.093250\n",
      "Epoch 1979, Loss 28.092396\n",
      "Epoch 1980, Loss 28.091545\n",
      "Epoch 1981, Loss 28.090687\n",
      "Epoch 1982, Loss 28.089832\n",
      "Epoch 1983, Loss 28.088978\n",
      "Epoch 1984, Loss 28.088129\n",
      "Epoch 1985, Loss 28.087273\n",
      "Epoch 1986, Loss 28.086414\n",
      "Epoch 1987, Loss 28.085558\n",
      "Epoch 1988, Loss 28.084705\n",
      "Epoch 1989, Loss 28.083851\n",
      "Epoch 1990, Loss 28.082996\n",
      "Epoch 1991, Loss 28.082142\n",
      "Epoch 1992, Loss 28.081287\n",
      "Epoch 1993, Loss 28.080433\n",
      "Epoch 1994, Loss 28.079578\n",
      "Epoch 1995, Loss 28.078724\n",
      "Epoch 1996, Loss 28.077869\n",
      "Epoch 1997, Loss 28.077015\n",
      "Epoch 1998, Loss 28.076164\n",
      "Epoch 1999, Loss 28.075306\n",
      "Epoch 2000, Loss 28.074451\n",
      "Epoch 2001, Loss 28.073597\n",
      "Epoch 2002, Loss 28.072748\n",
      "Epoch 2003, Loss 28.071892\n",
      "Epoch 2004, Loss 28.071037\n",
      "Epoch 2005, Loss 28.070177\n",
      "Epoch 2006, Loss 28.069324\n",
      "Epoch 2007, Loss 28.068468\n",
      "Epoch 2008, Loss 28.067621\n",
      "Epoch 2009, Loss 28.066765\n",
      "Epoch 2010, Loss 28.065910\n",
      "Epoch 2011, Loss 28.065058\n",
      "Epoch 2012, Loss 28.064203\n",
      "Epoch 2013, Loss 28.063347\n",
      "Epoch 2014, Loss 28.062494\n",
      "Epoch 2015, Loss 28.061640\n",
      "Epoch 2016, Loss 28.060785\n",
      "Epoch 2017, Loss 28.059935\n",
      "Epoch 2018, Loss 28.059080\n",
      "Epoch 2019, Loss 28.058228\n",
      "Epoch 2020, Loss 28.057373\n",
      "Epoch 2021, Loss 28.056517\n",
      "Epoch 2022, Loss 28.055662\n",
      "Epoch 2023, Loss 28.054810\n",
      "Epoch 2024, Loss 28.053955\n",
      "Epoch 2025, Loss 28.053101\n",
      "Epoch 2026, Loss 28.052248\n",
      "Epoch 2027, Loss 28.051392\n",
      "Epoch 2028, Loss 28.050539\n",
      "Epoch 2029, Loss 28.049688\n",
      "Epoch 2030, Loss 28.048834\n",
      "Epoch 2031, Loss 28.047985\n",
      "Epoch 2032, Loss 28.047131\n",
      "Epoch 2033, Loss 28.046272\n",
      "Epoch 2034, Loss 28.045422\n",
      "Epoch 2035, Loss 28.044567\n",
      "Epoch 2036, Loss 28.043713\n",
      "Epoch 2037, Loss 28.042860\n",
      "Epoch 2038, Loss 28.042009\n",
      "Epoch 2039, Loss 28.041155\n",
      "Epoch 2040, Loss 28.040302\n",
      "Epoch 2041, Loss 28.039448\n",
      "Epoch 2042, Loss 28.038593\n",
      "Epoch 2043, Loss 28.037743\n",
      "Epoch 2044, Loss 28.036888\n",
      "Epoch 2045, Loss 28.036034\n",
      "Epoch 2046, Loss 28.035181\n",
      "Epoch 2047, Loss 28.034330\n",
      "Epoch 2048, Loss 28.033480\n",
      "Epoch 2049, Loss 28.032623\n",
      "Epoch 2050, Loss 28.031771\n",
      "Epoch 2051, Loss 28.030916\n",
      "Epoch 2052, Loss 28.030067\n",
      "Epoch 2053, Loss 28.029213\n",
      "Epoch 2054, Loss 28.028357\n",
      "Epoch 2055, Loss 28.027508\n",
      "Epoch 2056, Loss 28.026655\n",
      "Epoch 2057, Loss 28.025801\n",
      "Epoch 2058, Loss 28.024946\n",
      "Epoch 2059, Loss 28.024097\n",
      "Epoch 2060, Loss 28.023243\n",
      "Epoch 2061, Loss 28.022392\n",
      "Epoch 2062, Loss 28.021540\n",
      "Epoch 2063, Loss 28.020685\n",
      "Epoch 2064, Loss 28.019836\n",
      "Epoch 2065, Loss 28.018980\n",
      "Epoch 2066, Loss 28.018129\n",
      "Epoch 2067, Loss 28.017279\n",
      "Epoch 2068, Loss 28.016426\n",
      "Epoch 2069, Loss 28.015570\n",
      "Epoch 2070, Loss 28.014721\n",
      "Epoch 2071, Loss 28.013866\n",
      "Epoch 2072, Loss 28.013018\n",
      "Epoch 2073, Loss 28.012157\n",
      "Epoch 2074, Loss 28.011309\n",
      "Epoch 2075, Loss 28.010460\n",
      "Epoch 2076, Loss 28.009605\n",
      "Epoch 2077, Loss 28.008751\n",
      "Epoch 2078, Loss 28.007902\n",
      "Epoch 2079, Loss 28.007050\n",
      "Epoch 2080, Loss 28.006197\n",
      "Epoch 2081, Loss 28.005346\n",
      "Epoch 2082, Loss 28.004494\n",
      "Epoch 2083, Loss 28.003643\n",
      "Epoch 2084, Loss 28.002790\n",
      "Epoch 2085, Loss 28.001942\n",
      "Epoch 2086, Loss 28.001085\n",
      "Epoch 2087, Loss 28.000233\n",
      "Epoch 2088, Loss 27.999382\n",
      "Epoch 2089, Loss 27.998529\n",
      "Epoch 2090, Loss 27.997679\n",
      "Epoch 2091, Loss 27.996826\n",
      "Epoch 2092, Loss 27.995977\n",
      "Epoch 2093, Loss 27.995123\n",
      "Epoch 2094, Loss 27.994270\n",
      "Epoch 2095, Loss 27.993425\n",
      "Epoch 2096, Loss 27.992567\n",
      "Epoch 2097, Loss 27.991716\n",
      "Epoch 2098, Loss 27.990868\n",
      "Epoch 2099, Loss 27.990013\n",
      "Epoch 2100, Loss 27.989161\n",
      "Epoch 2101, Loss 27.988312\n",
      "Epoch 2102, Loss 27.987463\n",
      "Epoch 2103, Loss 27.986610\n",
      "Epoch 2104, Loss 27.985756\n",
      "Epoch 2105, Loss 27.984907\n",
      "Epoch 2106, Loss 27.984053\n",
      "Epoch 2107, Loss 27.983202\n",
      "Epoch 2108, Loss 27.982353\n",
      "Epoch 2109, Loss 27.981501\n",
      "Epoch 2110, Loss 27.980652\n",
      "Epoch 2111, Loss 27.979797\n",
      "Epoch 2112, Loss 27.978943\n",
      "Epoch 2113, Loss 27.978094\n",
      "Epoch 2114, Loss 27.977245\n",
      "Epoch 2115, Loss 27.976393\n",
      "Epoch 2116, Loss 27.975542\n",
      "Epoch 2117, Loss 27.974693\n",
      "Epoch 2118, Loss 27.973841\n",
      "Epoch 2119, Loss 27.972994\n",
      "Epoch 2120, Loss 27.972137\n",
      "Epoch 2121, Loss 27.971289\n",
      "Epoch 2122, Loss 27.970434\n",
      "Epoch 2123, Loss 27.969587\n",
      "Epoch 2124, Loss 27.968737\n",
      "Epoch 2125, Loss 27.967884\n",
      "Epoch 2126, Loss 27.967035\n",
      "Epoch 2127, Loss 27.966181\n",
      "Epoch 2128, Loss 27.965338\n",
      "Epoch 2129, Loss 27.964479\n",
      "Epoch 2130, Loss 27.963631\n",
      "Epoch 2131, Loss 27.962780\n",
      "Epoch 2132, Loss 27.961931\n",
      "Epoch 2133, Loss 27.961079\n",
      "Epoch 2134, Loss 27.960228\n",
      "Epoch 2135, Loss 27.959381\n",
      "Epoch 2136, Loss 27.958532\n",
      "Epoch 2137, Loss 27.957680\n",
      "Epoch 2138, Loss 27.956831\n",
      "Epoch 2139, Loss 27.955975\n",
      "Epoch 2140, Loss 27.955126\n",
      "Epoch 2141, Loss 27.954277\n",
      "Epoch 2142, Loss 27.953424\n",
      "Epoch 2143, Loss 27.952574\n",
      "Epoch 2144, Loss 27.951727\n",
      "Epoch 2145, Loss 27.950880\n",
      "Epoch 2146, Loss 27.950024\n",
      "Epoch 2147, Loss 27.949177\n",
      "Epoch 2148, Loss 27.948326\n",
      "Epoch 2149, Loss 27.947477\n",
      "Epoch 2150, Loss 27.946625\n",
      "Epoch 2151, Loss 27.945778\n",
      "Epoch 2152, Loss 27.944923\n",
      "Epoch 2153, Loss 27.944078\n",
      "Epoch 2154, Loss 27.943230\n",
      "Epoch 2155, Loss 27.942381\n",
      "Epoch 2156, Loss 27.941523\n",
      "Epoch 2157, Loss 27.940674\n",
      "Epoch 2158, Loss 27.939821\n",
      "Epoch 2159, Loss 27.938972\n",
      "Epoch 2160, Loss 27.938124\n",
      "Epoch 2161, Loss 27.937279\n",
      "Epoch 2162, Loss 27.936424\n",
      "Epoch 2163, Loss 27.935579\n",
      "Epoch 2164, Loss 27.934729\n",
      "Epoch 2165, Loss 27.933880\n",
      "Epoch 2166, Loss 27.933027\n",
      "Epoch 2167, Loss 27.932182\n",
      "Epoch 2168, Loss 27.931330\n",
      "Epoch 2169, Loss 27.930481\n",
      "Epoch 2170, Loss 27.929632\n",
      "Epoch 2171, Loss 27.928785\n",
      "Epoch 2172, Loss 27.927937\n",
      "Epoch 2173, Loss 27.927080\n",
      "Epoch 2174, Loss 27.926233\n",
      "Epoch 2175, Loss 27.925383\n",
      "Epoch 2176, Loss 27.924532\n",
      "Epoch 2177, Loss 27.923683\n",
      "Epoch 2178, Loss 27.922834\n",
      "Epoch 2179, Loss 27.921986\n",
      "Epoch 2180, Loss 27.921137\n",
      "Epoch 2181, Loss 27.920286\n",
      "Epoch 2182, Loss 27.919439\n",
      "Epoch 2183, Loss 27.918591\n",
      "Epoch 2184, Loss 27.917742\n",
      "Epoch 2185, Loss 27.916893\n",
      "Epoch 2186, Loss 27.916044\n",
      "Epoch 2187, Loss 27.915194\n",
      "Epoch 2188, Loss 27.914345\n",
      "Epoch 2189, Loss 27.913500\n",
      "Epoch 2190, Loss 27.912645\n",
      "Epoch 2191, Loss 27.911797\n",
      "Epoch 2192, Loss 27.910948\n",
      "Epoch 2193, Loss 27.910097\n",
      "Epoch 2194, Loss 27.909248\n",
      "Epoch 2195, Loss 27.908403\n",
      "Epoch 2196, Loss 27.907555\n",
      "Epoch 2197, Loss 27.906708\n",
      "Epoch 2198, Loss 27.905851\n",
      "Epoch 2199, Loss 27.905006\n",
      "Epoch 2200, Loss 27.904158\n",
      "Epoch 2201, Loss 27.903309\n",
      "Epoch 2202, Loss 27.902464\n",
      "Epoch 2203, Loss 27.901611\n",
      "Epoch 2204, Loss 27.900768\n",
      "Epoch 2205, Loss 27.899920\n",
      "Epoch 2206, Loss 27.899071\n",
      "Epoch 2207, Loss 27.898216\n",
      "Epoch 2208, Loss 27.897367\n",
      "Epoch 2209, Loss 27.896523\n",
      "Epoch 2210, Loss 27.895672\n",
      "Epoch 2211, Loss 27.894819\n",
      "Epoch 2212, Loss 27.893976\n",
      "Epoch 2213, Loss 27.893127\n",
      "Epoch 2214, Loss 27.892281\n",
      "Epoch 2215, Loss 27.891432\n",
      "Epoch 2216, Loss 27.890583\n",
      "Epoch 2217, Loss 27.889740\n",
      "Epoch 2218, Loss 27.888887\n",
      "Epoch 2219, Loss 27.888037\n",
      "Epoch 2220, Loss 27.887194\n",
      "Epoch 2221, Loss 27.886345\n",
      "Epoch 2222, Loss 27.885496\n",
      "Epoch 2223, Loss 27.884649\n",
      "Epoch 2224, Loss 27.883802\n",
      "Epoch 2225, Loss 27.882948\n",
      "Epoch 2226, Loss 27.882109\n",
      "Epoch 2227, Loss 27.881254\n",
      "Epoch 2228, Loss 27.880407\n",
      "Epoch 2229, Loss 27.879560\n",
      "Epoch 2230, Loss 27.878712\n",
      "Epoch 2231, Loss 27.877863\n",
      "Epoch 2232, Loss 27.877020\n",
      "Epoch 2233, Loss 27.876171\n",
      "Epoch 2234, Loss 27.875319\n",
      "Epoch 2235, Loss 27.874479\n",
      "Epoch 2236, Loss 27.873627\n",
      "Epoch 2237, Loss 27.872780\n",
      "Epoch 2238, Loss 27.871935\n",
      "Epoch 2239, Loss 27.871086\n",
      "Epoch 2240, Loss 27.870239\n",
      "Epoch 2241, Loss 27.869387\n",
      "Epoch 2242, Loss 27.868544\n",
      "Epoch 2243, Loss 27.867695\n",
      "Epoch 2244, Loss 27.866850\n",
      "Epoch 2245, Loss 27.866003\n",
      "Epoch 2246, Loss 27.865154\n",
      "Epoch 2247, Loss 27.864307\n",
      "Epoch 2248, Loss 27.863459\n",
      "Epoch 2249, Loss 27.862616\n",
      "Epoch 2250, Loss 27.861769\n",
      "Epoch 2251, Loss 27.860920\n",
      "Epoch 2252, Loss 27.860071\n",
      "Epoch 2253, Loss 27.859224\n",
      "Epoch 2254, Loss 27.858376\n",
      "Epoch 2255, Loss 27.857533\n",
      "Epoch 2256, Loss 27.856684\n",
      "Epoch 2257, Loss 27.855835\n",
      "Epoch 2258, Loss 27.854988\n",
      "Epoch 2259, Loss 27.854137\n",
      "Epoch 2260, Loss 27.853296\n",
      "Epoch 2261, Loss 27.852448\n",
      "Epoch 2262, Loss 27.851601\n",
      "Epoch 2263, Loss 27.850752\n",
      "Epoch 2264, Loss 27.849909\n",
      "Epoch 2265, Loss 27.849062\n",
      "Epoch 2266, Loss 27.848217\n",
      "Epoch 2267, Loss 27.847364\n",
      "Epoch 2268, Loss 27.846519\n",
      "Epoch 2269, Loss 27.845678\n",
      "Epoch 2270, Loss 27.844831\n",
      "Epoch 2271, Loss 27.843983\n",
      "Epoch 2272, Loss 27.843134\n",
      "Epoch 2273, Loss 27.842291\n",
      "Epoch 2274, Loss 27.841444\n",
      "Epoch 2275, Loss 27.840595\n",
      "Epoch 2276, Loss 27.839752\n",
      "Epoch 2277, Loss 27.838903\n",
      "Epoch 2278, Loss 27.838060\n",
      "Epoch 2279, Loss 27.837214\n",
      "Epoch 2280, Loss 27.836365\n",
      "Epoch 2281, Loss 27.835522\n",
      "Epoch 2282, Loss 27.834673\n",
      "Epoch 2283, Loss 27.833830\n",
      "Epoch 2284, Loss 27.832979\n",
      "Epoch 2285, Loss 27.832134\n",
      "Epoch 2286, Loss 27.831287\n",
      "Epoch 2287, Loss 27.830439\n",
      "Epoch 2288, Loss 27.829596\n",
      "Epoch 2289, Loss 27.828753\n",
      "Epoch 2290, Loss 27.827906\n",
      "Epoch 2291, Loss 27.827057\n",
      "Epoch 2292, Loss 27.826210\n",
      "Epoch 2293, Loss 27.825367\n",
      "Epoch 2294, Loss 27.824518\n",
      "Epoch 2295, Loss 27.823671\n",
      "Epoch 2296, Loss 27.822828\n",
      "Epoch 2297, Loss 27.821983\n",
      "Epoch 2298, Loss 27.821138\n",
      "Epoch 2299, Loss 27.820293\n",
      "Epoch 2300, Loss 27.819447\n",
      "Epoch 2301, Loss 27.818598\n",
      "Epoch 2302, Loss 27.817757\n",
      "Epoch 2303, Loss 27.816914\n",
      "Epoch 2304, Loss 27.816065\n",
      "Epoch 2305, Loss 27.815218\n",
      "Epoch 2306, Loss 27.814373\n",
      "Epoch 2307, Loss 27.813530\n",
      "Epoch 2308, Loss 27.812681\n",
      "Epoch 2309, Loss 27.811840\n",
      "Epoch 2310, Loss 27.810993\n",
      "Epoch 2311, Loss 27.810150\n",
      "Epoch 2312, Loss 27.809303\n",
      "Epoch 2313, Loss 27.808455\n",
      "Epoch 2314, Loss 27.807611\n",
      "Epoch 2315, Loss 27.806763\n",
      "Epoch 2316, Loss 27.805916\n",
      "Epoch 2317, Loss 27.805071\n",
      "Epoch 2318, Loss 27.804230\n",
      "Epoch 2319, Loss 27.803381\n",
      "Epoch 2320, Loss 27.802540\n",
      "Epoch 2321, Loss 27.801697\n",
      "Epoch 2322, Loss 27.800850\n",
      "Epoch 2323, Loss 27.800001\n",
      "Epoch 2324, Loss 27.799158\n",
      "Epoch 2325, Loss 27.798315\n",
      "Epoch 2326, Loss 27.797466\n",
      "Epoch 2327, Loss 27.796625\n",
      "Epoch 2328, Loss 27.795778\n",
      "Epoch 2329, Loss 27.794933\n",
      "Epoch 2330, Loss 27.794086\n",
      "Epoch 2331, Loss 27.793245\n",
      "Epoch 2332, Loss 27.792400\n",
      "Epoch 2333, Loss 27.791553\n",
      "Epoch 2334, Loss 27.790710\n",
      "Epoch 2335, Loss 27.789869\n",
      "Epoch 2336, Loss 27.789024\n",
      "Epoch 2337, Loss 27.788179\n",
      "Epoch 2338, Loss 27.787334\n",
      "Epoch 2339, Loss 27.786488\n",
      "Epoch 2340, Loss 27.785645\n",
      "Epoch 2341, Loss 27.784798\n",
      "Epoch 2342, Loss 27.783949\n",
      "Epoch 2343, Loss 27.783110\n",
      "Epoch 2344, Loss 27.782263\n",
      "Epoch 2345, Loss 27.781420\n",
      "Epoch 2346, Loss 27.780577\n",
      "Epoch 2347, Loss 27.779732\n",
      "Epoch 2348, Loss 27.778887\n",
      "Epoch 2349, Loss 27.778044\n",
      "Epoch 2350, Loss 27.777199\n",
      "Epoch 2351, Loss 27.776356\n",
      "Epoch 2352, Loss 27.775511\n",
      "Epoch 2353, Loss 27.774666\n",
      "Epoch 2354, Loss 27.773821\n",
      "Epoch 2355, Loss 27.772980\n",
      "Epoch 2356, Loss 27.772137\n",
      "Epoch 2357, Loss 27.771290\n",
      "Epoch 2358, Loss 27.770447\n",
      "Epoch 2359, Loss 27.769604\n",
      "Epoch 2360, Loss 27.768757\n",
      "Epoch 2361, Loss 27.767916\n",
      "Epoch 2362, Loss 27.767071\n",
      "Epoch 2363, Loss 27.766228\n",
      "Epoch 2364, Loss 27.765383\n",
      "Epoch 2365, Loss 27.764538\n",
      "Epoch 2366, Loss 27.763695\n",
      "Epoch 2367, Loss 27.762850\n",
      "Epoch 2368, Loss 27.762011\n",
      "Epoch 2369, Loss 27.761168\n",
      "Epoch 2370, Loss 27.760321\n",
      "Epoch 2371, Loss 27.759474\n",
      "Epoch 2372, Loss 27.758631\n",
      "Epoch 2373, Loss 27.757788\n",
      "Epoch 2374, Loss 27.756947\n",
      "Epoch 2375, Loss 27.756104\n",
      "Epoch 2376, Loss 27.755257\n",
      "Epoch 2377, Loss 27.754412\n",
      "Epoch 2378, Loss 27.753569\n",
      "Epoch 2379, Loss 27.752728\n",
      "Epoch 2380, Loss 27.751881\n",
      "Epoch 2381, Loss 27.751040\n",
      "Epoch 2382, Loss 27.750195\n",
      "Epoch 2383, Loss 27.749357\n",
      "Epoch 2384, Loss 27.748510\n",
      "Epoch 2385, Loss 27.747667\n",
      "Epoch 2386, Loss 27.746824\n",
      "Epoch 2387, Loss 27.745983\n",
      "Epoch 2388, Loss 27.745140\n",
      "Epoch 2389, Loss 27.744295\n",
      "Epoch 2390, Loss 27.743452\n",
      "Epoch 2391, Loss 27.742607\n",
      "Epoch 2392, Loss 27.741766\n",
      "Epoch 2393, Loss 27.740923\n",
      "Epoch 2394, Loss 27.740082\n",
      "Epoch 2395, Loss 27.739235\n",
      "Epoch 2396, Loss 27.738392\n",
      "Epoch 2397, Loss 27.737551\n",
      "Epoch 2398, Loss 27.736712\n",
      "Epoch 2399, Loss 27.735865\n",
      "Epoch 2400, Loss 27.735022\n",
      "Epoch 2401, Loss 27.734173\n",
      "Epoch 2402, Loss 27.733332\n",
      "Epoch 2403, Loss 27.732491\n",
      "Epoch 2404, Loss 27.731647\n",
      "Epoch 2405, Loss 27.730804\n",
      "Epoch 2406, Loss 27.729963\n",
      "Epoch 2407, Loss 27.729126\n",
      "Epoch 2408, Loss 27.728279\n",
      "Epoch 2409, Loss 27.727436\n",
      "Epoch 2410, Loss 27.726593\n",
      "Epoch 2411, Loss 27.725752\n",
      "Epoch 2412, Loss 27.724909\n",
      "Epoch 2413, Loss 27.724068\n",
      "Epoch 2414, Loss 27.723225\n",
      "Epoch 2415, Loss 27.722383\n",
      "Epoch 2416, Loss 27.721539\n",
      "Epoch 2417, Loss 27.720697\n",
      "Epoch 2418, Loss 27.719854\n",
      "Epoch 2419, Loss 27.719011\n",
      "Epoch 2420, Loss 27.718172\n",
      "Epoch 2421, Loss 27.717333\n",
      "Epoch 2422, Loss 27.716490\n",
      "Epoch 2423, Loss 27.715645\n",
      "Epoch 2424, Loss 27.714806\n",
      "Epoch 2425, Loss 27.713961\n",
      "Epoch 2426, Loss 27.713118\n",
      "Epoch 2427, Loss 27.712275\n",
      "Epoch 2428, Loss 27.711428\n",
      "Epoch 2429, Loss 27.710588\n",
      "Epoch 2430, Loss 27.709745\n",
      "Epoch 2431, Loss 27.708906\n",
      "Epoch 2432, Loss 27.708069\n",
      "Epoch 2433, Loss 27.707220\n",
      "Epoch 2434, Loss 27.706379\n",
      "Epoch 2435, Loss 27.705538\n",
      "Epoch 2436, Loss 27.704695\n",
      "Epoch 2437, Loss 27.703857\n",
      "Epoch 2438, Loss 27.703011\n",
      "Epoch 2439, Loss 27.702171\n",
      "Epoch 2440, Loss 27.701330\n",
      "Epoch 2441, Loss 27.700489\n",
      "Epoch 2442, Loss 27.699646\n",
      "Epoch 2443, Loss 27.698803\n",
      "Epoch 2444, Loss 27.697966\n",
      "Epoch 2445, Loss 27.697124\n",
      "Epoch 2446, Loss 27.696283\n",
      "Epoch 2447, Loss 27.695436\n",
      "Epoch 2448, Loss 27.694597\n",
      "Epoch 2449, Loss 27.693754\n",
      "Epoch 2450, Loss 27.692915\n",
      "Epoch 2451, Loss 27.692072\n",
      "Epoch 2452, Loss 27.691235\n",
      "Epoch 2453, Loss 27.690388\n",
      "Epoch 2454, Loss 27.689550\n",
      "Epoch 2455, Loss 27.688707\n",
      "Epoch 2456, Loss 27.687866\n",
      "Epoch 2457, Loss 27.687019\n",
      "Epoch 2458, Loss 27.686182\n",
      "Epoch 2459, Loss 27.685339\n",
      "Epoch 2460, Loss 27.684502\n",
      "Epoch 2461, Loss 27.683657\n",
      "Epoch 2462, Loss 27.682819\n",
      "Epoch 2463, Loss 27.681976\n",
      "Epoch 2464, Loss 27.681135\n",
      "Epoch 2465, Loss 27.680296\n",
      "Epoch 2466, Loss 27.679457\n",
      "Epoch 2467, Loss 27.678614\n",
      "Epoch 2468, Loss 27.677773\n",
      "Epoch 2469, Loss 27.676935\n",
      "Epoch 2470, Loss 27.676092\n",
      "Epoch 2471, Loss 27.675251\n",
      "Epoch 2472, Loss 27.674408\n",
      "Epoch 2473, Loss 27.673567\n",
      "Epoch 2474, Loss 27.672729\n",
      "Epoch 2475, Loss 27.671886\n",
      "Epoch 2476, Loss 27.671049\n",
      "Epoch 2477, Loss 27.670204\n",
      "Epoch 2478, Loss 27.669365\n",
      "Epoch 2479, Loss 27.668530\n",
      "Epoch 2480, Loss 27.667683\n",
      "Epoch 2481, Loss 27.666845\n",
      "Epoch 2482, Loss 27.666008\n",
      "Epoch 2483, Loss 27.665163\n",
      "Epoch 2484, Loss 27.664318\n",
      "Epoch 2485, Loss 27.663481\n",
      "Epoch 2486, Loss 27.662640\n",
      "Epoch 2487, Loss 27.661798\n",
      "Epoch 2488, Loss 27.660955\n",
      "Epoch 2489, Loss 27.660118\n",
      "Epoch 2490, Loss 27.659279\n",
      "Epoch 2491, Loss 27.658440\n",
      "Epoch 2492, Loss 27.657595\n",
      "Epoch 2493, Loss 27.656761\n",
      "Epoch 2494, Loss 27.655916\n",
      "Epoch 2495, Loss 27.655073\n",
      "Epoch 2496, Loss 27.654236\n",
      "Epoch 2497, Loss 27.653399\n",
      "Epoch 2498, Loss 27.652555\n",
      "Epoch 2499, Loss 27.651716\n",
      "Epoch 2500, Loss 27.650877\n",
      "Epoch 2501, Loss 27.650036\n",
      "Epoch 2502, Loss 27.649199\n",
      "Epoch 2503, Loss 27.648357\n",
      "Epoch 2504, Loss 27.647520\n",
      "Epoch 2505, Loss 27.646681\n",
      "Epoch 2506, Loss 27.645836\n",
      "Epoch 2507, Loss 27.644995\n",
      "Epoch 2508, Loss 27.644163\n",
      "Epoch 2509, Loss 27.643318\n",
      "Epoch 2510, Loss 27.642479\n",
      "Epoch 2511, Loss 27.641638\n",
      "Epoch 2512, Loss 27.640802\n",
      "Epoch 2513, Loss 27.639965\n",
      "Epoch 2514, Loss 27.639124\n",
      "Epoch 2515, Loss 27.638283\n",
      "Epoch 2516, Loss 27.637440\n",
      "Epoch 2517, Loss 27.636602\n",
      "Epoch 2518, Loss 27.635765\n",
      "Epoch 2519, Loss 27.634924\n",
      "Epoch 2520, Loss 27.634083\n",
      "Epoch 2521, Loss 27.633245\n",
      "Epoch 2522, Loss 27.632408\n",
      "Epoch 2523, Loss 27.631567\n",
      "Epoch 2524, Loss 27.630730\n",
      "Epoch 2525, Loss 27.629890\n",
      "Epoch 2526, Loss 27.629051\n",
      "Epoch 2527, Loss 27.628216\n",
      "Epoch 2528, Loss 27.627369\n",
      "Epoch 2529, Loss 27.626532\n",
      "Epoch 2530, Loss 27.625694\n",
      "Epoch 2531, Loss 27.624855\n",
      "Epoch 2532, Loss 27.624018\n",
      "Epoch 2533, Loss 27.623180\n",
      "Epoch 2534, Loss 27.622339\n",
      "Epoch 2535, Loss 27.621498\n",
      "Epoch 2536, Loss 27.620661\n",
      "Epoch 2537, Loss 27.619820\n",
      "Epoch 2538, Loss 27.618982\n",
      "Epoch 2539, Loss 27.618141\n",
      "Epoch 2540, Loss 27.617308\n",
      "Epoch 2541, Loss 27.616463\n",
      "Epoch 2542, Loss 27.615629\n",
      "Epoch 2543, Loss 27.614788\n",
      "Epoch 2544, Loss 27.613951\n",
      "Epoch 2545, Loss 27.613115\n",
      "Epoch 2546, Loss 27.612276\n",
      "Epoch 2547, Loss 27.611439\n",
      "Epoch 2548, Loss 27.610596\n",
      "Epoch 2549, Loss 27.609760\n",
      "Epoch 2550, Loss 27.608919\n",
      "Epoch 2551, Loss 27.608080\n",
      "Epoch 2552, Loss 27.607244\n",
      "Epoch 2553, Loss 27.606407\n",
      "Epoch 2554, Loss 27.605562\n",
      "Epoch 2555, Loss 27.604731\n",
      "Epoch 2556, Loss 27.603889\n",
      "Epoch 2557, Loss 27.603050\n",
      "Epoch 2558, Loss 27.602205\n",
      "Epoch 2559, Loss 27.601374\n",
      "Epoch 2560, Loss 27.600531\n",
      "Epoch 2561, Loss 27.599693\n",
      "Epoch 2562, Loss 27.598854\n",
      "Epoch 2563, Loss 27.598017\n",
      "Epoch 2564, Loss 27.597179\n",
      "Epoch 2565, Loss 27.596342\n",
      "Epoch 2566, Loss 27.595503\n",
      "Epoch 2567, Loss 27.594666\n",
      "Epoch 2568, Loss 27.593828\n",
      "Epoch 2569, Loss 27.592989\n",
      "Epoch 2570, Loss 27.592155\n",
      "Epoch 2571, Loss 27.591314\n",
      "Epoch 2572, Loss 27.590479\n",
      "Epoch 2573, Loss 27.589638\n",
      "Epoch 2574, Loss 27.588799\n",
      "Epoch 2575, Loss 27.587965\n",
      "Epoch 2576, Loss 27.587126\n",
      "Epoch 2577, Loss 27.586292\n",
      "Epoch 2578, Loss 27.585449\n",
      "Epoch 2579, Loss 27.584614\n",
      "Epoch 2580, Loss 27.583773\n",
      "Epoch 2581, Loss 27.582935\n",
      "Epoch 2582, Loss 27.582100\n",
      "Epoch 2583, Loss 27.581257\n",
      "Epoch 2584, Loss 27.580425\n",
      "Epoch 2585, Loss 27.579588\n",
      "Epoch 2586, Loss 27.578753\n",
      "Epoch 2587, Loss 27.577911\n",
      "Epoch 2588, Loss 27.577074\n",
      "Epoch 2589, Loss 27.576239\n",
      "Epoch 2590, Loss 27.575397\n",
      "Epoch 2591, Loss 27.574562\n",
      "Epoch 2592, Loss 27.573723\n",
      "Epoch 2593, Loss 27.572887\n",
      "Epoch 2594, Loss 27.572050\n",
      "Epoch 2595, Loss 27.571215\n",
      "Epoch 2596, Loss 27.570375\n",
      "Epoch 2597, Loss 27.569538\n",
      "Epoch 2598, Loss 27.568697\n",
      "Epoch 2599, Loss 27.567865\n",
      "Epoch 2600, Loss 27.567028\n",
      "Epoch 2601, Loss 27.566191\n",
      "Epoch 2602, Loss 27.565351\n",
      "Epoch 2603, Loss 27.564514\n",
      "Epoch 2604, Loss 27.563683\n",
      "Epoch 2605, Loss 27.562841\n",
      "Epoch 2606, Loss 27.562004\n",
      "Epoch 2607, Loss 27.561169\n",
      "Epoch 2608, Loss 27.560331\n",
      "Epoch 2609, Loss 27.559492\n",
      "Epoch 2610, Loss 27.558657\n",
      "Epoch 2611, Loss 27.557823\n",
      "Epoch 2612, Loss 27.556984\n",
      "Epoch 2613, Loss 27.556147\n",
      "Epoch 2614, Loss 27.555309\n",
      "Epoch 2615, Loss 27.554476\n",
      "Epoch 2616, Loss 27.553638\n",
      "Epoch 2617, Loss 27.552801\n",
      "Epoch 2618, Loss 27.551970\n",
      "Epoch 2619, Loss 27.551125\n",
      "Epoch 2620, Loss 27.550291\n",
      "Epoch 2621, Loss 27.549458\n",
      "Epoch 2622, Loss 27.548620\n",
      "Epoch 2623, Loss 27.547783\n",
      "Epoch 2624, Loss 27.546942\n",
      "Epoch 2625, Loss 27.546110\n",
      "Epoch 2626, Loss 27.545275\n",
      "Epoch 2627, Loss 27.544434\n",
      "Epoch 2628, Loss 27.543604\n",
      "Epoch 2629, Loss 27.542763\n",
      "Epoch 2630, Loss 27.541925\n",
      "Epoch 2631, Loss 27.541094\n",
      "Epoch 2632, Loss 27.540253\n",
      "Epoch 2633, Loss 27.539421\n",
      "Epoch 2634, Loss 27.538580\n",
      "Epoch 2635, Loss 27.537748\n",
      "Epoch 2636, Loss 27.536909\n",
      "Epoch 2637, Loss 27.536074\n",
      "Epoch 2638, Loss 27.535242\n",
      "Epoch 2639, Loss 27.534401\n",
      "Epoch 2640, Loss 27.533567\n",
      "Epoch 2641, Loss 27.532728\n",
      "Epoch 2642, Loss 27.531893\n",
      "Epoch 2643, Loss 27.531061\n",
      "Epoch 2644, Loss 27.530224\n",
      "Epoch 2645, Loss 27.529388\n",
      "Epoch 2646, Loss 27.528547\n",
      "Epoch 2647, Loss 27.527712\n",
      "Epoch 2648, Loss 27.526875\n",
      "Epoch 2649, Loss 27.526043\n",
      "Epoch 2650, Loss 27.525208\n",
      "Epoch 2651, Loss 27.524370\n",
      "Epoch 2652, Loss 27.523531\n",
      "Epoch 2653, Loss 27.522699\n",
      "Epoch 2654, Loss 27.521864\n",
      "Epoch 2655, Loss 27.521029\n",
      "Epoch 2656, Loss 27.520191\n",
      "Epoch 2657, Loss 27.519360\n",
      "Epoch 2658, Loss 27.518522\n",
      "Epoch 2659, Loss 27.517687\n",
      "Epoch 2660, Loss 27.516848\n",
      "Epoch 2661, Loss 27.516014\n",
      "Epoch 2662, Loss 27.515181\n",
      "Epoch 2663, Loss 27.514345\n",
      "Epoch 2664, Loss 27.513508\n",
      "Epoch 2665, Loss 27.512672\n",
      "Epoch 2666, Loss 27.511841\n",
      "Epoch 2667, Loss 27.511005\n",
      "Epoch 2668, Loss 27.510168\n",
      "Epoch 2669, Loss 27.509333\n",
      "Epoch 2670, Loss 27.508495\n",
      "Epoch 2671, Loss 27.507666\n",
      "Epoch 2672, Loss 27.506830\n",
      "Epoch 2673, Loss 27.505993\n",
      "Epoch 2674, Loss 27.505157\n",
      "Epoch 2675, Loss 27.504322\n",
      "Epoch 2676, Loss 27.503487\n",
      "Epoch 2677, Loss 27.502653\n",
      "Epoch 2678, Loss 27.501820\n",
      "Epoch 2679, Loss 27.500982\n",
      "Epoch 2680, Loss 27.500147\n",
      "Epoch 2681, Loss 27.499311\n",
      "Epoch 2682, Loss 27.498480\n",
      "Epoch 2683, Loss 27.497648\n",
      "Epoch 2684, Loss 27.496809\n",
      "Epoch 2685, Loss 27.495974\n",
      "Epoch 2686, Loss 27.495140\n",
      "Epoch 2687, Loss 27.494301\n",
      "Epoch 2688, Loss 27.493469\n",
      "Epoch 2689, Loss 27.492638\n",
      "Epoch 2690, Loss 27.491802\n",
      "Epoch 2691, Loss 27.490969\n",
      "Epoch 2692, Loss 27.490131\n",
      "Epoch 2693, Loss 27.489296\n",
      "Epoch 2694, Loss 27.488459\n",
      "Epoch 2695, Loss 27.487633\n",
      "Epoch 2696, Loss 27.486794\n",
      "Epoch 2697, Loss 27.485960\n",
      "Epoch 2698, Loss 27.485125\n",
      "Epoch 2699, Loss 27.484291\n",
      "Epoch 2700, Loss 27.483459\n",
      "Epoch 2701, Loss 27.482622\n",
      "Epoch 2702, Loss 27.481789\n",
      "Epoch 2703, Loss 27.480957\n",
      "Epoch 2704, Loss 27.480120\n",
      "Epoch 2705, Loss 27.479286\n",
      "Epoch 2706, Loss 27.478449\n",
      "Epoch 2707, Loss 27.477617\n",
      "Epoch 2708, Loss 27.476784\n",
      "Epoch 2709, Loss 27.475950\n",
      "Epoch 2710, Loss 27.475115\n",
      "Epoch 2711, Loss 27.474276\n",
      "Epoch 2712, Loss 27.473448\n",
      "Epoch 2713, Loss 27.472618\n",
      "Epoch 2714, Loss 27.471779\n",
      "Epoch 2715, Loss 27.470947\n",
      "Epoch 2716, Loss 27.470112\n",
      "Epoch 2717, Loss 27.469276\n",
      "Epoch 2718, Loss 27.468445\n",
      "Epoch 2719, Loss 27.467613\n",
      "Epoch 2720, Loss 27.466780\n",
      "Epoch 2721, Loss 27.465942\n",
      "Epoch 2722, Loss 27.465111\n",
      "Epoch 2723, Loss 27.464275\n",
      "Epoch 2724, Loss 27.463440\n",
      "Epoch 2725, Loss 27.462610\n",
      "Epoch 2726, Loss 27.461775\n",
      "Epoch 2727, Loss 27.460943\n",
      "Epoch 2728, Loss 27.460108\n",
      "Epoch 2729, Loss 27.459276\n",
      "Epoch 2730, Loss 27.458443\n",
      "Epoch 2731, Loss 27.457605\n",
      "Epoch 2732, Loss 27.456774\n",
      "Epoch 2733, Loss 27.455940\n",
      "Epoch 2734, Loss 27.455103\n",
      "Epoch 2735, Loss 27.454273\n",
      "Epoch 2736, Loss 27.453438\n",
      "Epoch 2737, Loss 27.452606\n",
      "Epoch 2738, Loss 27.451771\n",
      "Epoch 2739, Loss 27.450939\n",
      "Epoch 2740, Loss 27.450104\n",
      "Epoch 2741, Loss 27.449274\n",
      "Epoch 2742, Loss 27.448437\n",
      "Epoch 2743, Loss 27.447603\n",
      "Epoch 2744, Loss 27.446772\n",
      "Epoch 2745, Loss 27.445936\n",
      "Epoch 2746, Loss 27.445105\n",
      "Epoch 2747, Loss 27.444275\n",
      "Epoch 2748, Loss 27.443438\n",
      "Epoch 2749, Loss 27.442608\n",
      "Epoch 2750, Loss 27.441772\n",
      "Epoch 2751, Loss 27.440943\n",
      "Epoch 2752, Loss 27.440107\n",
      "Epoch 2753, Loss 27.439276\n",
      "Epoch 2754, Loss 27.438444\n",
      "Epoch 2755, Loss 27.437611\n",
      "Epoch 2756, Loss 27.436775\n",
      "Epoch 2757, Loss 27.435946\n",
      "Epoch 2758, Loss 27.435112\n",
      "Epoch 2759, Loss 27.434282\n",
      "Epoch 2760, Loss 27.433449\n",
      "Epoch 2761, Loss 27.432617\n",
      "Epoch 2762, Loss 27.431782\n",
      "Epoch 2763, Loss 27.430952\n",
      "Epoch 2764, Loss 27.430117\n",
      "Epoch 2765, Loss 27.429289\n",
      "Epoch 2766, Loss 27.428450\n",
      "Epoch 2767, Loss 27.427620\n",
      "Epoch 2768, Loss 27.426786\n",
      "Epoch 2769, Loss 27.425957\n",
      "Epoch 2770, Loss 27.425123\n",
      "Epoch 2771, Loss 27.424288\n",
      "Epoch 2772, Loss 27.423460\n",
      "Epoch 2773, Loss 27.422630\n",
      "Epoch 2774, Loss 27.421795\n",
      "Epoch 2775, Loss 27.420965\n",
      "Epoch 2776, Loss 27.420130\n",
      "Epoch 2777, Loss 27.419298\n",
      "Epoch 2778, Loss 27.418465\n",
      "Epoch 2779, Loss 27.417633\n",
      "Epoch 2780, Loss 27.416803\n",
      "Epoch 2781, Loss 27.415968\n",
      "Epoch 2782, Loss 27.415138\n",
      "Epoch 2783, Loss 27.414307\n",
      "Epoch 2784, Loss 27.413475\n",
      "Epoch 2785, Loss 27.412640\n",
      "Epoch 2786, Loss 27.411816\n",
      "Epoch 2787, Loss 27.410980\n",
      "Epoch 2788, Loss 27.410145\n",
      "Epoch 2789, Loss 27.409315\n",
      "Epoch 2790, Loss 27.408480\n",
      "Epoch 2791, Loss 27.407652\n",
      "Epoch 2792, Loss 27.406818\n",
      "Epoch 2793, Loss 27.405989\n",
      "Epoch 2794, Loss 27.405153\n",
      "Epoch 2795, Loss 27.404327\n",
      "Epoch 2796, Loss 27.403492\n",
      "Epoch 2797, Loss 27.402662\n",
      "Epoch 2798, Loss 27.401827\n",
      "Epoch 2799, Loss 27.400995\n",
      "Epoch 2800, Loss 27.400166\n",
      "Epoch 2801, Loss 27.399336\n",
      "Epoch 2802, Loss 27.398510\n",
      "Epoch 2803, Loss 27.397675\n",
      "Epoch 2804, Loss 27.396845\n",
      "Epoch 2805, Loss 27.396013\n",
      "Epoch 2806, Loss 27.395180\n",
      "Epoch 2807, Loss 27.394350\n",
      "Epoch 2808, Loss 27.393518\n",
      "Epoch 2809, Loss 27.392689\n",
      "Epoch 2810, Loss 27.391857\n",
      "Epoch 2811, Loss 27.391027\n",
      "Epoch 2812, Loss 27.390192\n",
      "Epoch 2813, Loss 27.389360\n",
      "Epoch 2814, Loss 27.388533\n",
      "Epoch 2815, Loss 27.387703\n",
      "Epoch 2816, Loss 27.386869\n",
      "Epoch 2817, Loss 27.386042\n",
      "Epoch 2818, Loss 27.385210\n",
      "Epoch 2819, Loss 27.384377\n",
      "Epoch 2820, Loss 27.383547\n",
      "Epoch 2821, Loss 27.382715\n",
      "Epoch 2822, Loss 27.381886\n",
      "Epoch 2823, Loss 27.381056\n",
      "Epoch 2824, Loss 27.380220\n",
      "Epoch 2825, Loss 27.379393\n",
      "Epoch 2826, Loss 27.378563\n",
      "Epoch 2827, Loss 27.377729\n",
      "Epoch 2828, Loss 27.376900\n",
      "Epoch 2829, Loss 27.376068\n",
      "Epoch 2830, Loss 27.375238\n",
      "Epoch 2831, Loss 27.374409\n",
      "Epoch 2832, Loss 27.373573\n",
      "Epoch 2833, Loss 27.372753\n",
      "Epoch 2834, Loss 27.371916\n",
      "Epoch 2835, Loss 27.371088\n",
      "Epoch 2836, Loss 27.370256\n",
      "Epoch 2837, Loss 27.369427\n",
      "Epoch 2838, Loss 27.368597\n",
      "Epoch 2839, Loss 27.367765\n",
      "Epoch 2840, Loss 27.366932\n",
      "Epoch 2841, Loss 27.366106\n",
      "Epoch 2842, Loss 27.365273\n",
      "Epoch 2843, Loss 27.364441\n",
      "Epoch 2844, Loss 27.363615\n",
      "Epoch 2845, Loss 27.362785\n",
      "Epoch 2846, Loss 27.361956\n",
      "Epoch 2847, Loss 27.361122\n",
      "Epoch 2848, Loss 27.360292\n",
      "Epoch 2849, Loss 27.359465\n",
      "Epoch 2850, Loss 27.358631\n",
      "Epoch 2851, Loss 27.357801\n",
      "Epoch 2852, Loss 27.356970\n",
      "Epoch 2853, Loss 27.356142\n",
      "Epoch 2854, Loss 27.355314\n",
      "Epoch 2855, Loss 27.354485\n",
      "Epoch 2856, Loss 27.353651\n",
      "Epoch 2857, Loss 27.352825\n",
      "Epoch 2858, Loss 27.351992\n",
      "Epoch 2859, Loss 27.351164\n",
      "Epoch 2860, Loss 27.350334\n",
      "Epoch 2861, Loss 27.349504\n",
      "Epoch 2862, Loss 27.348671\n",
      "Epoch 2863, Loss 27.347841\n",
      "Epoch 2864, Loss 27.347013\n",
      "Epoch 2865, Loss 27.346186\n",
      "Epoch 2866, Loss 27.345356\n",
      "Epoch 2867, Loss 27.344524\n",
      "Epoch 2868, Loss 27.343697\n",
      "Epoch 2869, Loss 27.342867\n",
      "Epoch 2870, Loss 27.342035\n",
      "Epoch 2871, Loss 27.341209\n",
      "Epoch 2872, Loss 27.340376\n",
      "Epoch 2873, Loss 27.339550\n",
      "Epoch 2874, Loss 27.338720\n",
      "Epoch 2875, Loss 27.337891\n",
      "Epoch 2876, Loss 27.337059\n",
      "Epoch 2877, Loss 27.336231\n",
      "Epoch 2878, Loss 27.335400\n",
      "Epoch 2879, Loss 27.334572\n",
      "Epoch 2880, Loss 27.333746\n",
      "Epoch 2881, Loss 27.332914\n",
      "Epoch 2882, Loss 27.332090\n",
      "Epoch 2883, Loss 27.331257\n",
      "Epoch 2884, Loss 27.330431\n",
      "Epoch 2885, Loss 27.329597\n",
      "Epoch 2886, Loss 27.328772\n",
      "Epoch 2887, Loss 27.327942\n",
      "Epoch 2888, Loss 27.327116\n",
      "Epoch 2889, Loss 27.326286\n",
      "Epoch 2890, Loss 27.325453\n",
      "Epoch 2891, Loss 27.324627\n",
      "Epoch 2892, Loss 27.323797\n",
      "Epoch 2893, Loss 27.322968\n",
      "Epoch 2894, Loss 27.322142\n",
      "Epoch 2895, Loss 27.321318\n",
      "Epoch 2896, Loss 27.320482\n",
      "Epoch 2897, Loss 27.319658\n",
      "Epoch 2898, Loss 27.318823\n",
      "Epoch 2899, Loss 27.317995\n",
      "Epoch 2900, Loss 27.317167\n",
      "Epoch 2901, Loss 27.316339\n",
      "Epoch 2902, Loss 27.315510\n",
      "Epoch 2903, Loss 27.314680\n",
      "Epoch 2904, Loss 27.313854\n",
      "Epoch 2905, Loss 27.313026\n",
      "Epoch 2906, Loss 27.312197\n",
      "Epoch 2907, Loss 27.311369\n",
      "Epoch 2908, Loss 27.310541\n",
      "Epoch 2909, Loss 27.309711\n",
      "Epoch 2910, Loss 27.308882\n",
      "Epoch 2911, Loss 27.308052\n",
      "Epoch 2912, Loss 27.307222\n",
      "Epoch 2913, Loss 27.306396\n",
      "Epoch 2914, Loss 27.305567\n",
      "Epoch 2915, Loss 27.304743\n",
      "Epoch 2916, Loss 27.303911\n",
      "Epoch 2917, Loss 27.303082\n",
      "Epoch 2918, Loss 27.302254\n",
      "Epoch 2919, Loss 27.301428\n",
      "Epoch 2920, Loss 27.300604\n",
      "Epoch 2921, Loss 27.299772\n",
      "Epoch 2922, Loss 27.298944\n",
      "Epoch 2923, Loss 27.298119\n",
      "Epoch 2924, Loss 27.297289\n",
      "Epoch 2925, Loss 27.296459\n",
      "Epoch 2926, Loss 27.295631\n",
      "Epoch 2927, Loss 27.294802\n",
      "Epoch 2928, Loss 27.293976\n",
      "Epoch 2929, Loss 27.293152\n",
      "Epoch 2930, Loss 27.292322\n",
      "Epoch 2931, Loss 27.291496\n",
      "Epoch 2932, Loss 27.290663\n",
      "Epoch 2933, Loss 27.289839\n",
      "Epoch 2934, Loss 27.289009\n",
      "Epoch 2935, Loss 27.288179\n",
      "Epoch 2936, Loss 27.287354\n",
      "Epoch 2937, Loss 27.286528\n",
      "Epoch 2938, Loss 27.285698\n",
      "Epoch 2939, Loss 27.284870\n",
      "Epoch 2940, Loss 27.284040\n",
      "Epoch 2941, Loss 27.283215\n",
      "Epoch 2942, Loss 27.282387\n",
      "Epoch 2943, Loss 27.281561\n",
      "Epoch 2944, Loss 27.280733\n",
      "Epoch 2945, Loss 27.279905\n",
      "Epoch 2946, Loss 27.279078\n",
      "Epoch 2947, Loss 27.278252\n",
      "Epoch 2948, Loss 27.277424\n",
      "Epoch 2949, Loss 27.276598\n",
      "Epoch 2950, Loss 27.275768\n",
      "Epoch 2951, Loss 27.274944\n",
      "Epoch 2952, Loss 27.274111\n",
      "Epoch 2953, Loss 27.273287\n",
      "Epoch 2954, Loss 27.272459\n",
      "Epoch 2955, Loss 27.271635\n",
      "Epoch 2956, Loss 27.270805\n",
      "Epoch 2957, Loss 27.269976\n",
      "Epoch 2958, Loss 27.269152\n",
      "Epoch 2959, Loss 27.268328\n",
      "Epoch 2960, Loss 27.267500\n",
      "Epoch 2961, Loss 27.266674\n",
      "Epoch 2962, Loss 27.265850\n",
      "Epoch 2963, Loss 27.265022\n",
      "Epoch 2964, Loss 27.264193\n",
      "Epoch 2965, Loss 27.263367\n",
      "Epoch 2966, Loss 27.262541\n",
      "Epoch 2967, Loss 27.261711\n",
      "Epoch 2968, Loss 27.260887\n",
      "Epoch 2969, Loss 27.260059\n",
      "Epoch 2970, Loss 27.259233\n",
      "Epoch 2971, Loss 27.258406\n",
      "Epoch 2972, Loss 27.257580\n",
      "Epoch 2973, Loss 27.256752\n",
      "Epoch 2974, Loss 27.255926\n",
      "Epoch 2975, Loss 27.255098\n",
      "Epoch 2976, Loss 27.254271\n",
      "Epoch 2977, Loss 27.253448\n",
      "Epoch 2978, Loss 27.252619\n",
      "Epoch 2979, Loss 27.251793\n",
      "Epoch 2980, Loss 27.250969\n",
      "Epoch 2981, Loss 27.250145\n",
      "Epoch 2982, Loss 27.249315\n",
      "Epoch 2983, Loss 27.248486\n",
      "Epoch 2984, Loss 27.247662\n",
      "Epoch 2985, Loss 27.246832\n",
      "Epoch 2986, Loss 27.246010\n",
      "Epoch 2987, Loss 27.245180\n",
      "Epoch 2988, Loss 27.244356\n",
      "Epoch 2989, Loss 27.243532\n",
      "Epoch 2990, Loss 27.242704\n",
      "Epoch 2991, Loss 27.241880\n",
      "Epoch 2992, Loss 27.241053\n",
      "Epoch 2993, Loss 27.240227\n",
      "Epoch 2994, Loss 27.239399\n",
      "Epoch 2995, Loss 27.238575\n",
      "Epoch 2996, Loss 27.237743\n",
      "Epoch 2997, Loss 27.236919\n",
      "Epoch 2998, Loss 27.236095\n",
      "Epoch 2999, Loss 27.235271\n",
      "Epoch 3000, Loss 27.234444\n",
      "Epoch 3001, Loss 27.233614\n",
      "Epoch 3002, Loss 27.232794\n",
      "Epoch 3003, Loss 27.231964\n",
      "Epoch 3004, Loss 27.231138\n",
      "Epoch 3005, Loss 27.230314\n",
      "Epoch 3006, Loss 27.229490\n",
      "Epoch 3007, Loss 27.228662\n",
      "Epoch 3008, Loss 27.227837\n",
      "Epoch 3009, Loss 27.227013\n",
      "Epoch 3010, Loss 27.226185\n",
      "Epoch 3011, Loss 27.225359\n",
      "Epoch 3012, Loss 27.224531\n",
      "Epoch 3013, Loss 27.223711\n",
      "Epoch 3014, Loss 27.222883\n",
      "Epoch 3015, Loss 27.222059\n",
      "Epoch 3016, Loss 27.221233\n",
      "Epoch 3017, Loss 27.220409\n",
      "Epoch 3018, Loss 27.219580\n",
      "Epoch 3019, Loss 27.218752\n",
      "Epoch 3020, Loss 27.217928\n",
      "Epoch 3021, Loss 27.217104\n",
      "Epoch 3022, Loss 27.216278\n",
      "Epoch 3023, Loss 27.215454\n",
      "Epoch 3024, Loss 27.214628\n",
      "Epoch 3025, Loss 27.213804\n",
      "Epoch 3026, Loss 27.212984\n",
      "Epoch 3027, Loss 27.212152\n",
      "Epoch 3028, Loss 27.211327\n",
      "Epoch 3029, Loss 27.210508\n",
      "Epoch 3030, Loss 27.209675\n",
      "Epoch 3031, Loss 27.208851\n",
      "Epoch 3032, Loss 27.208025\n",
      "Epoch 3033, Loss 27.207203\n",
      "Epoch 3034, Loss 27.206377\n",
      "Epoch 3035, Loss 27.205553\n",
      "Epoch 3036, Loss 27.204729\n",
      "Epoch 3037, Loss 27.203905\n",
      "Epoch 3038, Loss 27.203077\n",
      "Epoch 3039, Loss 27.202248\n",
      "Epoch 3040, Loss 27.201433\n",
      "Epoch 3041, Loss 27.200605\n",
      "Epoch 3042, Loss 27.199785\n",
      "Epoch 3043, Loss 27.198956\n",
      "Epoch 3044, Loss 27.198137\n",
      "Epoch 3045, Loss 27.197309\n",
      "Epoch 3046, Loss 27.196486\n",
      "Epoch 3047, Loss 27.195660\n",
      "Epoch 3048, Loss 27.194838\n",
      "Epoch 3049, Loss 27.194012\n",
      "Epoch 3050, Loss 27.193184\n",
      "Epoch 3051, Loss 27.192360\n",
      "Epoch 3052, Loss 27.191540\n",
      "Epoch 3053, Loss 27.190712\n",
      "Epoch 3054, Loss 27.189886\n",
      "Epoch 3055, Loss 27.189062\n",
      "Epoch 3056, Loss 27.188240\n",
      "Epoch 3057, Loss 27.187414\n",
      "Epoch 3058, Loss 27.186592\n",
      "Epoch 3059, Loss 27.185766\n",
      "Epoch 3060, Loss 27.184942\n",
      "Epoch 3061, Loss 27.184118\n",
      "Epoch 3062, Loss 27.183294\n",
      "Epoch 3063, Loss 27.182470\n",
      "Epoch 3064, Loss 27.181643\n",
      "Epoch 3065, Loss 27.180824\n",
      "Epoch 3066, Loss 27.179998\n",
      "Epoch 3067, Loss 27.179174\n",
      "Epoch 3068, Loss 27.178350\n",
      "Epoch 3069, Loss 27.177521\n",
      "Epoch 3070, Loss 27.176702\n",
      "Epoch 3071, Loss 27.175875\n",
      "Epoch 3072, Loss 27.175055\n",
      "Epoch 3073, Loss 27.174227\n",
      "Epoch 3074, Loss 27.173407\n",
      "Epoch 3075, Loss 27.172588\n",
      "Epoch 3076, Loss 27.171764\n",
      "Epoch 3077, Loss 27.170935\n",
      "Epoch 3078, Loss 27.170116\n",
      "Epoch 3079, Loss 27.169289\n",
      "Epoch 3080, Loss 27.168465\n",
      "Epoch 3081, Loss 27.167646\n",
      "Epoch 3082, Loss 27.166821\n",
      "Epoch 3083, Loss 27.165997\n",
      "Epoch 3084, Loss 27.165173\n",
      "Epoch 3085, Loss 27.164351\n",
      "Epoch 3086, Loss 27.163525\n",
      "Epoch 3087, Loss 27.162703\n",
      "Epoch 3088, Loss 27.161879\n",
      "Epoch 3089, Loss 27.161055\n",
      "Epoch 3090, Loss 27.160231\n",
      "Epoch 3091, Loss 27.159412\n",
      "Epoch 3092, Loss 27.158583\n",
      "Epoch 3093, Loss 27.157759\n",
      "Epoch 3094, Loss 27.156939\n",
      "Epoch 3095, Loss 27.156116\n",
      "Epoch 3096, Loss 27.155293\n",
      "Epoch 3097, Loss 27.154467\n",
      "Epoch 3098, Loss 27.153643\n",
      "Epoch 3099, Loss 27.152821\n",
      "Epoch 3100, Loss 27.152000\n",
      "Epoch 3101, Loss 27.151173\n",
      "Epoch 3102, Loss 27.150358\n",
      "Epoch 3103, Loss 27.149536\n",
      "Epoch 3104, Loss 27.148712\n",
      "Epoch 3105, Loss 27.147886\n",
      "Epoch 3106, Loss 27.147064\n",
      "Epoch 3107, Loss 27.146240\n",
      "Epoch 3108, Loss 27.145418\n",
      "Epoch 3109, Loss 27.144592\n",
      "Epoch 3110, Loss 27.143774\n",
      "Epoch 3111, Loss 27.142950\n",
      "Epoch 3112, Loss 27.142126\n",
      "Epoch 3113, Loss 27.141304\n",
      "Epoch 3114, Loss 27.140484\n",
      "Epoch 3115, Loss 27.139660\n",
      "Epoch 3116, Loss 27.138838\n",
      "Epoch 3117, Loss 27.138018\n",
      "Epoch 3118, Loss 27.137190\n",
      "Epoch 3119, Loss 27.136370\n",
      "Epoch 3120, Loss 27.135548\n",
      "Epoch 3121, Loss 27.134724\n",
      "Epoch 3122, Loss 27.133904\n",
      "Epoch 3123, Loss 27.133080\n",
      "Epoch 3124, Loss 27.132256\n",
      "Epoch 3125, Loss 27.131432\n",
      "Epoch 3126, Loss 27.130610\n",
      "Epoch 3127, Loss 27.129786\n",
      "Epoch 3128, Loss 27.128967\n",
      "Epoch 3129, Loss 27.128145\n",
      "Epoch 3130, Loss 27.127327\n",
      "Epoch 3131, Loss 27.126501\n",
      "Epoch 3132, Loss 27.125683\n",
      "Epoch 3133, Loss 27.124859\n",
      "Epoch 3134, Loss 27.124037\n",
      "Epoch 3135, Loss 27.123217\n",
      "Epoch 3136, Loss 27.122393\n",
      "Epoch 3137, Loss 27.121571\n",
      "Epoch 3138, Loss 27.120750\n",
      "Epoch 3139, Loss 27.119926\n",
      "Epoch 3140, Loss 27.119104\n",
      "Epoch 3141, Loss 27.118284\n",
      "Epoch 3142, Loss 27.117460\n",
      "Epoch 3143, Loss 27.116640\n",
      "Epoch 3144, Loss 27.115816\n",
      "Epoch 3145, Loss 27.114992\n",
      "Epoch 3146, Loss 27.114174\n",
      "Epoch 3147, Loss 27.113354\n",
      "Epoch 3148, Loss 27.112526\n",
      "Epoch 3149, Loss 27.111706\n",
      "Epoch 3150, Loss 27.110889\n",
      "Epoch 3151, Loss 27.110065\n",
      "Epoch 3152, Loss 27.109241\n",
      "Epoch 3153, Loss 27.108421\n",
      "Epoch 3154, Loss 27.107603\n",
      "Epoch 3155, Loss 27.106779\n",
      "Epoch 3156, Loss 27.105957\n",
      "Epoch 3157, Loss 27.105139\n",
      "Epoch 3158, Loss 27.104317\n",
      "Epoch 3159, Loss 27.103497\n",
      "Epoch 3160, Loss 27.102678\n",
      "Epoch 3161, Loss 27.101851\n",
      "Epoch 3162, Loss 27.101027\n",
      "Epoch 3163, Loss 27.100208\n",
      "Epoch 3164, Loss 27.099388\n",
      "Epoch 3165, Loss 27.098564\n",
      "Epoch 3166, Loss 27.097746\n",
      "Epoch 3167, Loss 27.096922\n",
      "Epoch 3168, Loss 27.096102\n",
      "Epoch 3169, Loss 27.095282\n",
      "Epoch 3170, Loss 27.094460\n",
      "Epoch 3171, Loss 27.093639\n",
      "Epoch 3172, Loss 27.092817\n",
      "Epoch 3173, Loss 27.091997\n",
      "Epoch 3174, Loss 27.091179\n",
      "Epoch 3175, Loss 27.090357\n",
      "Epoch 3176, Loss 27.089537\n",
      "Epoch 3177, Loss 27.088715\n",
      "Epoch 3178, Loss 27.087891\n",
      "Epoch 3179, Loss 27.087072\n",
      "Epoch 3180, Loss 27.086245\n",
      "Epoch 3181, Loss 27.085426\n",
      "Epoch 3182, Loss 27.084606\n",
      "Epoch 3183, Loss 27.083788\n",
      "Epoch 3184, Loss 27.082966\n",
      "Epoch 3185, Loss 27.082151\n",
      "Epoch 3186, Loss 27.081329\n",
      "Epoch 3187, Loss 27.080505\n",
      "Epoch 3188, Loss 27.079687\n",
      "Epoch 3189, Loss 27.078865\n",
      "Epoch 3190, Loss 27.078047\n",
      "Epoch 3191, Loss 27.077223\n",
      "Epoch 3192, Loss 27.076408\n",
      "Epoch 3193, Loss 27.075584\n",
      "Epoch 3194, Loss 27.074762\n",
      "Epoch 3195, Loss 27.073942\n",
      "Epoch 3196, Loss 27.073126\n",
      "Epoch 3197, Loss 27.072298\n",
      "Epoch 3198, Loss 27.071480\n",
      "Epoch 3199, Loss 27.070660\n",
      "Epoch 3200, Loss 27.069841\n",
      "Epoch 3201, Loss 27.069019\n",
      "Epoch 3202, Loss 27.068201\n",
      "Epoch 3203, Loss 27.067381\n",
      "Epoch 3204, Loss 27.066561\n",
      "Epoch 3205, Loss 27.065741\n",
      "Epoch 3206, Loss 27.064919\n",
      "Epoch 3207, Loss 27.064098\n",
      "Epoch 3208, Loss 27.063276\n",
      "Epoch 3209, Loss 27.062462\n",
      "Epoch 3210, Loss 27.061640\n",
      "Epoch 3211, Loss 27.060820\n",
      "Epoch 3212, Loss 27.059999\n",
      "Epoch 3213, Loss 27.059181\n",
      "Epoch 3214, Loss 27.058361\n",
      "Epoch 3215, Loss 27.057545\n",
      "Epoch 3216, Loss 27.056721\n",
      "Epoch 3217, Loss 27.055897\n",
      "Epoch 3218, Loss 27.055079\n",
      "Epoch 3219, Loss 27.054260\n",
      "Epoch 3220, Loss 27.053444\n",
      "Epoch 3221, Loss 27.052624\n",
      "Epoch 3222, Loss 27.051802\n",
      "Epoch 3223, Loss 27.050982\n",
      "Epoch 3224, Loss 27.050159\n",
      "Epoch 3225, Loss 27.049345\n",
      "Epoch 3226, Loss 27.048525\n",
      "Epoch 3227, Loss 27.047701\n",
      "Epoch 3228, Loss 27.046883\n",
      "Epoch 3229, Loss 27.046064\n",
      "Epoch 3230, Loss 27.045244\n",
      "Epoch 3231, Loss 27.044428\n",
      "Epoch 3232, Loss 27.043608\n",
      "Epoch 3233, Loss 27.042788\n",
      "Epoch 3234, Loss 27.041967\n",
      "Epoch 3235, Loss 27.041151\n",
      "Epoch 3236, Loss 27.040327\n",
      "Epoch 3237, Loss 27.039513\n",
      "Epoch 3238, Loss 27.038691\n",
      "Epoch 3239, Loss 27.037874\n",
      "Epoch 3240, Loss 27.037060\n",
      "Epoch 3241, Loss 27.036232\n",
      "Epoch 3242, Loss 27.035418\n",
      "Epoch 3243, Loss 27.034599\n",
      "Epoch 3244, Loss 27.033781\n",
      "Epoch 3245, Loss 27.032959\n",
      "Epoch 3246, Loss 27.032141\n",
      "Epoch 3247, Loss 27.031322\n",
      "Epoch 3248, Loss 27.030504\n",
      "Epoch 3249, Loss 27.029686\n",
      "Epoch 3250, Loss 27.028864\n",
      "Epoch 3251, Loss 27.028044\n",
      "Epoch 3252, Loss 27.027229\n",
      "Epoch 3253, Loss 27.026405\n",
      "Epoch 3254, Loss 27.025587\n",
      "Epoch 3255, Loss 27.024769\n",
      "Epoch 3256, Loss 27.023949\n",
      "Epoch 3257, Loss 27.023130\n",
      "Epoch 3258, Loss 27.022310\n",
      "Epoch 3259, Loss 27.021496\n",
      "Epoch 3260, Loss 27.020674\n",
      "Epoch 3261, Loss 27.019855\n",
      "Epoch 3262, Loss 27.019041\n",
      "Epoch 3263, Loss 27.018217\n",
      "Epoch 3264, Loss 27.017401\n",
      "Epoch 3265, Loss 27.016582\n",
      "Epoch 3266, Loss 27.015764\n",
      "Epoch 3267, Loss 27.014948\n",
      "Epoch 3268, Loss 27.014130\n",
      "Epoch 3269, Loss 27.013317\n",
      "Epoch 3270, Loss 27.012489\n",
      "Epoch 3271, Loss 27.011677\n",
      "Epoch 3272, Loss 27.010857\n",
      "Epoch 3273, Loss 27.010038\n",
      "Epoch 3274, Loss 27.009222\n",
      "Epoch 3275, Loss 27.008406\n",
      "Epoch 3276, Loss 27.007580\n",
      "Epoch 3277, Loss 27.006763\n",
      "Epoch 3278, Loss 27.005949\n",
      "Epoch 3279, Loss 27.005125\n",
      "Epoch 3280, Loss 27.004311\n",
      "Epoch 3281, Loss 27.003492\n",
      "Epoch 3282, Loss 27.002674\n",
      "Epoch 3283, Loss 27.001854\n",
      "Epoch 3284, Loss 27.001040\n",
      "Epoch 3285, Loss 27.000216\n",
      "Epoch 3286, Loss 26.999401\n",
      "Epoch 3287, Loss 26.998585\n",
      "Epoch 3288, Loss 26.997765\n",
      "Epoch 3289, Loss 26.996946\n",
      "Epoch 3290, Loss 26.996130\n",
      "Epoch 3291, Loss 26.995312\n",
      "Epoch 3292, Loss 26.994493\n",
      "Epoch 3293, Loss 26.993677\n",
      "Epoch 3294, Loss 26.992859\n",
      "Epoch 3295, Loss 26.992043\n",
      "Epoch 3296, Loss 26.991222\n",
      "Epoch 3297, Loss 26.990410\n",
      "Epoch 3298, Loss 26.989592\n",
      "Epoch 3299, Loss 26.988771\n",
      "Epoch 3300, Loss 26.987957\n",
      "Epoch 3301, Loss 26.987139\n",
      "Epoch 3302, Loss 26.986320\n",
      "Epoch 3303, Loss 26.985504\n",
      "Epoch 3304, Loss 26.984682\n",
      "Epoch 3305, Loss 26.983864\n",
      "Epoch 3306, Loss 26.983049\n",
      "Epoch 3307, Loss 26.982237\n",
      "Epoch 3308, Loss 26.981417\n",
      "Epoch 3309, Loss 26.980597\n",
      "Epoch 3310, Loss 26.979780\n",
      "Epoch 3311, Loss 26.978966\n",
      "Epoch 3312, Loss 26.978149\n",
      "Epoch 3313, Loss 26.977327\n",
      "Epoch 3314, Loss 26.976513\n",
      "Epoch 3315, Loss 26.975697\n",
      "Epoch 3316, Loss 26.974876\n",
      "Epoch 3317, Loss 26.974060\n",
      "Epoch 3318, Loss 26.973244\n",
      "Epoch 3319, Loss 26.972425\n",
      "Epoch 3320, Loss 26.971611\n",
      "Epoch 3321, Loss 26.970789\n",
      "Epoch 3322, Loss 26.969976\n",
      "Epoch 3323, Loss 26.969160\n",
      "Epoch 3324, Loss 26.968346\n",
      "Epoch 3325, Loss 26.967524\n",
      "Epoch 3326, Loss 26.966707\n",
      "Epoch 3327, Loss 26.965899\n",
      "Epoch 3328, Loss 26.965076\n",
      "Epoch 3329, Loss 26.964262\n",
      "Epoch 3330, Loss 26.963442\n",
      "Epoch 3331, Loss 26.962627\n",
      "Epoch 3332, Loss 26.961809\n",
      "Epoch 3333, Loss 26.960993\n",
      "Epoch 3334, Loss 26.960173\n",
      "Epoch 3335, Loss 26.959358\n",
      "Epoch 3336, Loss 26.958540\n",
      "Epoch 3337, Loss 26.957726\n",
      "Epoch 3338, Loss 26.956909\n",
      "Epoch 3339, Loss 26.956093\n",
      "Epoch 3340, Loss 26.955273\n",
      "Epoch 3341, Loss 26.954456\n",
      "Epoch 3342, Loss 26.953646\n",
      "Epoch 3343, Loss 26.952831\n",
      "Epoch 3344, Loss 26.952015\n",
      "Epoch 3345, Loss 26.951197\n",
      "Epoch 3346, Loss 26.950378\n",
      "Epoch 3347, Loss 26.949562\n",
      "Epoch 3348, Loss 26.948750\n",
      "Epoch 3349, Loss 26.947931\n",
      "Epoch 3350, Loss 26.947115\n",
      "Epoch 3351, Loss 26.946297\n",
      "Epoch 3352, Loss 26.945484\n",
      "Epoch 3353, Loss 26.944670\n",
      "Epoch 3354, Loss 26.943848\n",
      "Epoch 3355, Loss 26.943031\n",
      "Epoch 3356, Loss 26.942219\n",
      "Epoch 3357, Loss 26.941404\n",
      "Epoch 3358, Loss 26.940584\n",
      "Epoch 3359, Loss 26.939772\n",
      "Epoch 3360, Loss 26.938959\n",
      "Epoch 3361, Loss 26.938143\n",
      "Epoch 3362, Loss 26.937328\n",
      "Epoch 3363, Loss 26.936510\n",
      "Epoch 3364, Loss 26.935694\n",
      "Epoch 3365, Loss 26.934881\n",
      "Epoch 3366, Loss 26.934065\n",
      "Epoch 3367, Loss 26.933247\n",
      "Epoch 3368, Loss 26.932428\n",
      "Epoch 3369, Loss 26.931616\n",
      "Epoch 3370, Loss 26.930803\n",
      "Epoch 3371, Loss 26.929981\n",
      "Epoch 3372, Loss 26.929171\n",
      "Epoch 3373, Loss 26.928356\n",
      "Epoch 3374, Loss 26.927534\n",
      "Epoch 3375, Loss 26.926725\n",
      "Epoch 3376, Loss 26.925909\n",
      "Epoch 3377, Loss 26.925097\n",
      "Epoch 3378, Loss 26.924278\n",
      "Epoch 3379, Loss 26.923468\n",
      "Epoch 3380, Loss 26.922649\n",
      "Epoch 3381, Loss 26.921833\n",
      "Epoch 3382, Loss 26.921015\n",
      "Epoch 3383, Loss 26.920202\n",
      "Epoch 3384, Loss 26.919390\n",
      "Epoch 3385, Loss 26.918573\n",
      "Epoch 3386, Loss 26.917753\n",
      "Epoch 3387, Loss 26.916943\n",
      "Epoch 3388, Loss 26.916126\n",
      "Epoch 3389, Loss 26.915312\n",
      "Epoch 3390, Loss 26.914492\n",
      "Epoch 3391, Loss 26.913679\n",
      "Epoch 3392, Loss 26.912870\n",
      "Epoch 3393, Loss 26.912050\n",
      "Epoch 3394, Loss 26.911242\n",
      "Epoch 3395, Loss 26.910425\n",
      "Epoch 3396, Loss 26.909607\n",
      "Epoch 3397, Loss 26.908791\n",
      "Epoch 3398, Loss 26.907978\n",
      "Epoch 3399, Loss 26.907164\n",
      "Epoch 3400, Loss 26.906353\n",
      "Epoch 3401, Loss 26.905531\n",
      "Epoch 3402, Loss 26.904722\n",
      "Epoch 3403, Loss 26.903906\n",
      "Epoch 3404, Loss 26.903090\n",
      "Epoch 3405, Loss 26.902277\n",
      "Epoch 3406, Loss 26.901461\n",
      "Epoch 3407, Loss 26.900646\n",
      "Epoch 3408, Loss 26.899836\n",
      "Epoch 3409, Loss 26.899017\n",
      "Epoch 3410, Loss 26.898205\n",
      "Epoch 3411, Loss 26.897394\n",
      "Epoch 3412, Loss 26.896576\n",
      "Epoch 3413, Loss 26.895760\n",
      "Epoch 3414, Loss 26.894947\n",
      "Epoch 3415, Loss 26.894133\n",
      "Epoch 3416, Loss 26.893316\n",
      "Epoch 3417, Loss 26.892506\n",
      "Epoch 3418, Loss 26.891691\n",
      "Epoch 3419, Loss 26.890877\n",
      "Epoch 3420, Loss 26.890059\n",
      "Epoch 3421, Loss 26.889250\n",
      "Epoch 3422, Loss 26.888433\n",
      "Epoch 3423, Loss 26.887617\n",
      "Epoch 3424, Loss 26.886808\n",
      "Epoch 3425, Loss 26.885992\n",
      "Epoch 3426, Loss 26.885176\n",
      "Epoch 3427, Loss 26.884369\n",
      "Epoch 3428, Loss 26.883551\n",
      "Epoch 3429, Loss 26.882738\n",
      "Epoch 3430, Loss 26.881926\n",
      "Epoch 3431, Loss 26.881111\n",
      "Epoch 3432, Loss 26.880297\n",
      "Epoch 3433, Loss 26.879478\n",
      "Epoch 3434, Loss 26.878670\n",
      "Epoch 3435, Loss 26.877851\n",
      "Epoch 3436, Loss 26.877043\n",
      "Epoch 3437, Loss 26.876226\n",
      "Epoch 3438, Loss 26.875410\n",
      "Epoch 3439, Loss 26.874601\n",
      "Epoch 3440, Loss 26.873779\n",
      "Epoch 3441, Loss 26.872969\n",
      "Epoch 3442, Loss 26.872156\n",
      "Epoch 3443, Loss 26.871346\n",
      "Epoch 3444, Loss 26.870533\n",
      "Epoch 3445, Loss 26.869719\n",
      "Epoch 3446, Loss 26.868908\n",
      "Epoch 3447, Loss 26.868092\n",
      "Epoch 3448, Loss 26.867279\n",
      "Epoch 3449, Loss 26.866463\n",
      "Epoch 3450, Loss 26.865656\n",
      "Epoch 3451, Loss 26.864836\n",
      "Epoch 3452, Loss 26.864025\n",
      "Epoch 3453, Loss 26.863209\n",
      "Epoch 3454, Loss 26.862402\n",
      "Epoch 3455, Loss 26.861589\n",
      "Epoch 3456, Loss 26.860773\n",
      "Epoch 3457, Loss 26.859957\n",
      "Epoch 3458, Loss 26.859142\n",
      "Epoch 3459, Loss 26.858332\n",
      "Epoch 3460, Loss 26.857521\n",
      "Epoch 3461, Loss 26.856707\n",
      "Epoch 3462, Loss 26.855894\n",
      "Epoch 3463, Loss 26.855083\n",
      "Epoch 3464, Loss 26.854271\n",
      "Epoch 3465, Loss 26.853455\n",
      "Epoch 3466, Loss 26.852638\n",
      "Epoch 3467, Loss 26.851830\n",
      "Epoch 3468, Loss 26.851019\n",
      "Epoch 3469, Loss 26.850203\n",
      "Epoch 3470, Loss 26.849394\n",
      "Epoch 3471, Loss 26.848577\n",
      "Epoch 3472, Loss 26.847765\n",
      "Epoch 3473, Loss 26.846949\n",
      "Epoch 3474, Loss 26.846142\n",
      "Epoch 3475, Loss 26.845329\n",
      "Epoch 3476, Loss 26.844515\n",
      "Epoch 3477, Loss 26.843702\n",
      "Epoch 3478, Loss 26.842890\n",
      "Epoch 3479, Loss 26.842077\n",
      "Epoch 3480, Loss 26.841270\n",
      "Epoch 3481, Loss 26.840454\n",
      "Epoch 3482, Loss 26.839642\n",
      "Epoch 3483, Loss 26.838825\n",
      "Epoch 3484, Loss 26.838015\n",
      "Epoch 3485, Loss 26.837202\n",
      "Epoch 3486, Loss 26.836393\n",
      "Epoch 3487, Loss 26.835579\n",
      "Epoch 3488, Loss 26.834761\n",
      "Epoch 3489, Loss 26.833954\n",
      "Epoch 3490, Loss 26.833141\n",
      "Epoch 3491, Loss 26.832325\n",
      "Epoch 3492, Loss 26.831514\n",
      "Epoch 3493, Loss 26.830711\n",
      "Epoch 3494, Loss 26.829897\n",
      "Epoch 3495, Loss 26.829084\n",
      "Epoch 3496, Loss 26.828270\n",
      "Epoch 3497, Loss 26.827459\n",
      "Epoch 3498, Loss 26.826647\n",
      "Epoch 3499, Loss 26.825834\n",
      "Epoch 3500, Loss 26.825020\n",
      "Epoch 3501, Loss 26.824207\n",
      "Epoch 3502, Loss 26.823397\n",
      "Epoch 3503, Loss 26.822588\n",
      "Epoch 3504, Loss 26.821772\n",
      "Epoch 3505, Loss 26.820961\n",
      "Epoch 3506, Loss 26.820152\n",
      "Epoch 3507, Loss 26.819336\n",
      "Epoch 3508, Loss 26.818523\n",
      "Epoch 3509, Loss 26.817717\n",
      "Epoch 3510, Loss 26.816906\n",
      "Epoch 3511, Loss 26.816093\n",
      "Epoch 3512, Loss 26.815285\n",
      "Epoch 3513, Loss 26.814472\n",
      "Epoch 3514, Loss 26.813660\n",
      "Epoch 3515, Loss 26.812847\n",
      "Epoch 3516, Loss 26.812035\n",
      "Epoch 3517, Loss 26.811224\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 3518, Loss 26.810413\n",
      "Epoch 3519, Loss 26.809601\n",
      "Epoch 3520, Loss 26.808788\n",
      "Epoch 3521, Loss 26.807978\n",
      "Epoch 3522, Loss 26.807165\n",
      "Epoch 3523, Loss 26.806353\n",
      "Epoch 3524, Loss 26.805542\n",
      "Epoch 3525, Loss 26.804731\n",
      "Epoch 3526, Loss 26.803919\n",
      "Epoch 3527, Loss 26.803112\n",
      "Epoch 3528, Loss 26.802299\n",
      "Epoch 3529, Loss 26.801491\n",
      "Epoch 3530, Loss 26.800682\n",
      "Epoch 3531, Loss 26.799868\n",
      "Epoch 3532, Loss 26.799059\n",
      "Epoch 3533, Loss 26.798246\n",
      "Epoch 3534, Loss 26.797432\n",
      "Epoch 3535, Loss 26.796623\n",
      "Epoch 3536, Loss 26.795815\n",
      "Epoch 3537, Loss 26.795000\n",
      "Epoch 3538, Loss 26.794189\n",
      "Epoch 3539, Loss 26.793379\n",
      "Epoch 3540, Loss 26.792566\n",
      "Epoch 3541, Loss 26.791754\n",
      "Epoch 3542, Loss 26.790943\n",
      "Epoch 3543, Loss 26.790136\n",
      "Epoch 3544, Loss 26.789324\n",
      "Epoch 3545, Loss 26.788515\n",
      "Epoch 3546, Loss 26.787706\n",
      "Epoch 3547, Loss 26.786894\n",
      "Epoch 3548, Loss 26.786083\n",
      "Epoch 3549, Loss 26.785273\n",
      "Epoch 3550, Loss 26.784462\n",
      "Epoch 3551, Loss 26.783648\n",
      "Epoch 3552, Loss 26.782839\n",
      "Epoch 3553, Loss 26.782026\n",
      "Epoch 3554, Loss 26.781218\n",
      "Epoch 3555, Loss 26.780409\n",
      "Epoch 3556, Loss 26.779600\n",
      "Epoch 3557, Loss 26.778786\n",
      "Epoch 3558, Loss 26.777977\n",
      "Epoch 3559, Loss 26.777168\n",
      "Epoch 3560, Loss 26.776360\n",
      "Epoch 3561, Loss 26.775545\n",
      "Epoch 3562, Loss 26.774736\n",
      "Epoch 3563, Loss 26.773920\n",
      "Epoch 3564, Loss 26.773113\n",
      "Epoch 3565, Loss 26.772303\n",
      "Epoch 3566, Loss 26.771496\n",
      "Epoch 3567, Loss 26.770683\n",
      "Epoch 3568, Loss 26.769873\n",
      "Epoch 3569, Loss 26.769066\n",
      "Epoch 3570, Loss 26.768255\n",
      "Epoch 3571, Loss 26.767443\n",
      "Epoch 3572, Loss 26.766632\n",
      "Epoch 3573, Loss 26.765825\n",
      "Epoch 3574, Loss 26.765015\n",
      "Epoch 3575, Loss 26.764204\n",
      "Epoch 3576, Loss 26.763395\n",
      "Epoch 3577, Loss 26.762587\n",
      "Epoch 3578, Loss 26.761774\n",
      "Epoch 3579, Loss 26.760963\n",
      "Epoch 3580, Loss 26.760155\n",
      "Epoch 3581, Loss 26.759346\n",
      "Epoch 3582, Loss 26.758537\n",
      "Epoch 3583, Loss 26.757727\n",
      "Epoch 3584, Loss 26.756916\n",
      "Epoch 3585, Loss 26.756105\n",
      "Epoch 3586, Loss 26.755299\n",
      "Epoch 3587, Loss 26.754488\n",
      "Epoch 3588, Loss 26.753679\n",
      "Epoch 3589, Loss 26.752863\n",
      "Epoch 3590, Loss 26.752056\n",
      "Epoch 3591, Loss 26.751251\n",
      "Epoch 3592, Loss 26.750444\n",
      "Epoch 3593, Loss 26.749632\n",
      "Epoch 3594, Loss 26.748823\n",
      "Epoch 3595, Loss 26.748014\n",
      "Epoch 3596, Loss 26.747206\n",
      "Epoch 3597, Loss 26.746393\n",
      "Epoch 3598, Loss 26.745583\n",
      "Epoch 3599, Loss 26.744774\n",
      "Epoch 3600, Loss 26.743965\n",
      "Epoch 3601, Loss 26.743153\n",
      "Epoch 3602, Loss 26.742348\n",
      "Epoch 3603, Loss 26.741539\n",
      "Epoch 3604, Loss 26.740734\n",
      "Epoch 3605, Loss 26.739916\n",
      "Epoch 3606, Loss 26.739111\n",
      "Epoch 3607, Loss 26.738298\n",
      "Epoch 3608, Loss 26.737490\n",
      "Epoch 3609, Loss 26.736681\n",
      "Epoch 3610, Loss 26.735878\n",
      "Epoch 3611, Loss 26.735065\n",
      "Epoch 3612, Loss 26.734255\n",
      "Epoch 3613, Loss 26.733452\n",
      "Epoch 3614, Loss 26.732637\n",
      "Epoch 3615, Loss 26.731834\n",
      "Epoch 3616, Loss 26.731024\n",
      "Epoch 3617, Loss 26.730207\n",
      "Epoch 3618, Loss 26.729401\n",
      "Epoch 3619, Loss 26.728594\n",
      "Epoch 3620, Loss 26.727785\n",
      "Epoch 3621, Loss 26.726978\n",
      "Epoch 3622, Loss 26.726168\n",
      "Epoch 3623, Loss 26.725359\n",
      "Epoch 3624, Loss 26.724554\n",
      "Epoch 3625, Loss 26.723743\n",
      "Epoch 3626, Loss 26.722939\n",
      "Epoch 3627, Loss 26.722130\n",
      "Epoch 3628, Loss 26.721319\n",
      "Epoch 3629, Loss 26.720512\n",
      "Epoch 3630, Loss 26.719702\n",
      "Epoch 3631, Loss 26.718895\n",
      "Epoch 3632, Loss 26.718084\n",
      "Epoch 3633, Loss 26.717278\n",
      "Epoch 3634, Loss 26.716469\n",
      "Epoch 3635, Loss 26.715666\n",
      "Epoch 3636, Loss 26.714855\n",
      "Epoch 3637, Loss 26.714045\n",
      "Epoch 3638, Loss 26.713240\n",
      "Epoch 3639, Loss 26.712431\n",
      "Epoch 3640, Loss 26.711620\n",
      "Epoch 3641, Loss 26.710814\n",
      "Epoch 3642, Loss 26.710005\n",
      "Epoch 3643, Loss 26.709196\n",
      "Epoch 3644, Loss 26.708391\n",
      "Epoch 3645, Loss 26.707579\n",
      "Epoch 3646, Loss 26.706774\n",
      "Epoch 3647, Loss 26.705969\n",
      "Epoch 3648, Loss 26.705162\n",
      "Epoch 3649, Loss 26.704351\n",
      "Epoch 3650, Loss 26.703541\n",
      "Epoch 3651, Loss 26.702736\n",
      "Epoch 3652, Loss 26.701927\n",
      "Epoch 3653, Loss 26.701117\n",
      "Epoch 3654, Loss 26.700312\n",
      "Epoch 3655, Loss 26.699505\n",
      "Epoch 3656, Loss 26.698698\n",
      "Epoch 3657, Loss 26.697887\n",
      "Epoch 3658, Loss 26.697081\n",
      "Epoch 3659, Loss 26.696276\n",
      "Epoch 3660, Loss 26.695467\n",
      "Epoch 3661, Loss 26.694658\n",
      "Epoch 3662, Loss 26.693853\n",
      "Epoch 3663, Loss 26.693043\n",
      "Epoch 3664, Loss 26.692238\n",
      "Epoch 3665, Loss 26.691429\n",
      "Epoch 3666, Loss 26.690624\n",
      "Epoch 3667, Loss 26.689812\n",
      "Epoch 3668, Loss 26.689009\n",
      "Epoch 3669, Loss 26.688200\n",
      "Epoch 3670, Loss 26.687389\n",
      "Epoch 3671, Loss 26.686586\n",
      "Epoch 3672, Loss 26.685780\n",
      "Epoch 3673, Loss 26.684973\n",
      "Epoch 3674, Loss 26.684166\n",
      "Epoch 3675, Loss 26.683355\n",
      "Epoch 3676, Loss 26.682554\n",
      "Epoch 3677, Loss 26.681744\n",
      "Epoch 3678, Loss 26.680937\n",
      "Epoch 3679, Loss 26.680128\n",
      "Epoch 3680, Loss 26.679319\n",
      "Epoch 3681, Loss 26.678516\n",
      "Epoch 3682, Loss 26.677706\n",
      "Epoch 3683, Loss 26.676899\n",
      "Epoch 3684, Loss 26.676098\n",
      "Epoch 3685, Loss 26.675287\n",
      "Epoch 3686, Loss 26.674482\n",
      "Epoch 3687, Loss 26.673674\n",
      "Epoch 3688, Loss 26.672863\n",
      "Epoch 3689, Loss 26.672058\n",
      "Epoch 3690, Loss 26.671257\n",
      "Epoch 3691, Loss 26.670446\n",
      "Epoch 3692, Loss 26.669640\n",
      "Epoch 3693, Loss 26.668835\n",
      "Epoch 3694, Loss 26.668028\n",
      "Epoch 3695, Loss 26.667225\n",
      "Epoch 3696, Loss 26.666414\n",
      "Epoch 3697, Loss 26.665611\n",
      "Epoch 3698, Loss 26.664803\n",
      "Epoch 3699, Loss 26.663996\n",
      "Epoch 3700, Loss 26.663191\n",
      "Epoch 3701, Loss 26.662390\n",
      "Epoch 3702, Loss 26.661573\n",
      "Epoch 3703, Loss 26.660772\n",
      "Epoch 3704, Loss 26.659962\n",
      "Epoch 3705, Loss 26.659157\n",
      "Epoch 3706, Loss 26.658354\n",
      "Epoch 3707, Loss 26.657545\n",
      "Epoch 3708, Loss 26.656740\n",
      "Epoch 3709, Loss 26.655933\n",
      "Epoch 3710, Loss 26.655128\n",
      "Epoch 3711, Loss 26.654325\n",
      "Epoch 3712, Loss 26.653515\n",
      "Epoch 3713, Loss 26.652710\n",
      "Epoch 3714, Loss 26.651903\n",
      "Epoch 3715, Loss 26.651094\n",
      "Epoch 3716, Loss 26.650288\n",
      "Epoch 3717, Loss 26.649487\n",
      "Epoch 3718, Loss 26.648682\n",
      "Epoch 3719, Loss 26.647875\n",
      "Epoch 3720, Loss 26.647066\n",
      "Epoch 3721, Loss 26.646263\n",
      "Epoch 3722, Loss 26.645458\n",
      "Epoch 3723, Loss 26.644646\n",
      "Epoch 3724, Loss 26.643843\n",
      "Epoch 3725, Loss 26.643042\n",
      "Epoch 3726, Loss 26.642235\n",
      "Epoch 3727, Loss 26.641426\n",
      "Epoch 3728, Loss 26.640623\n",
      "Epoch 3729, Loss 26.639818\n",
      "Epoch 3730, Loss 26.639013\n",
      "Epoch 3731, Loss 26.638206\n",
      "Epoch 3732, Loss 26.637402\n",
      "Epoch 3733, Loss 26.636597\n",
      "Epoch 3734, Loss 26.635786\n",
      "Epoch 3735, Loss 26.634981\n",
      "Epoch 3736, Loss 26.634172\n",
      "Epoch 3737, Loss 26.633373\n",
      "Epoch 3738, Loss 26.632566\n",
      "Epoch 3739, Loss 26.631758\n",
      "Epoch 3740, Loss 26.630949\n",
      "Epoch 3741, Loss 26.630150\n",
      "Epoch 3742, Loss 26.629341\n",
      "Epoch 3743, Loss 26.628540\n",
      "Epoch 3744, Loss 26.627733\n",
      "Epoch 3745, Loss 26.626930\n",
      "Epoch 3746, Loss 26.626122\n",
      "Epoch 3747, Loss 26.625322\n",
      "Epoch 3748, Loss 26.624514\n",
      "Epoch 3749, Loss 26.623711\n",
      "Epoch 3750, Loss 26.622902\n",
      "Epoch 3751, Loss 26.622095\n",
      "Epoch 3752, Loss 26.621294\n",
      "Epoch 3753, Loss 26.620485\n",
      "Epoch 3754, Loss 26.619684\n",
      "Epoch 3755, Loss 26.618879\n",
      "Epoch 3756, Loss 26.618074\n",
      "Epoch 3757, Loss 26.617268\n",
      "Epoch 3758, Loss 26.616461\n",
      "Epoch 3759, Loss 26.615660\n",
      "Epoch 3760, Loss 26.614855\n",
      "Epoch 3761, Loss 26.614052\n",
      "Epoch 3762, Loss 26.613245\n",
      "Epoch 3763, Loss 26.612440\n",
      "Epoch 3764, Loss 26.611639\n",
      "Epoch 3765, Loss 26.610828\n",
      "Epoch 3766, Loss 26.610029\n",
      "Epoch 3767, Loss 26.609224\n",
      "Epoch 3768, Loss 26.608421\n",
      "Epoch 3769, Loss 26.607613\n",
      "Epoch 3770, Loss 26.606812\n",
      "Epoch 3771, Loss 26.606007\n",
      "Epoch 3772, Loss 26.605202\n",
      "Epoch 3773, Loss 26.604399\n",
      "Epoch 3774, Loss 26.603588\n",
      "Epoch 3775, Loss 26.602789\n",
      "Epoch 3776, Loss 26.601982\n",
      "Epoch 3777, Loss 26.601179\n",
      "Epoch 3778, Loss 26.600374\n",
      "Epoch 3779, Loss 26.599573\n",
      "Epoch 3780, Loss 26.598766\n",
      "Epoch 3781, Loss 26.597963\n",
      "Epoch 3782, Loss 26.597160\n",
      "Epoch 3783, Loss 26.596352\n",
      "Epoch 3784, Loss 26.595551\n",
      "Epoch 3785, Loss 26.594744\n",
      "Epoch 3786, Loss 26.593939\n",
      "Epoch 3787, Loss 26.593138\n",
      "Epoch 3788, Loss 26.592329\n",
      "Epoch 3789, Loss 26.591530\n",
      "Epoch 3790, Loss 26.590727\n",
      "Epoch 3791, Loss 26.589918\n",
      "Epoch 3792, Loss 26.589123\n",
      "Epoch 3793, Loss 26.588316\n",
      "Epoch 3794, Loss 26.587507\n",
      "Epoch 3795, Loss 26.586706\n",
      "Epoch 3796, Loss 26.585901\n",
      "Epoch 3797, Loss 26.585100\n",
      "Epoch 3798, Loss 26.584295\n",
      "Epoch 3799, Loss 26.583490\n",
      "Epoch 3800, Loss 26.582689\n",
      "Epoch 3801, Loss 26.581881\n",
      "Epoch 3802, Loss 26.581078\n",
      "Epoch 3803, Loss 26.580278\n",
      "Epoch 3804, Loss 26.579470\n",
      "Epoch 3805, Loss 26.578669\n",
      "Epoch 3806, Loss 26.577868\n",
      "Epoch 3807, Loss 26.577057\n",
      "Epoch 3808, Loss 26.576258\n",
      "Epoch 3809, Loss 26.575453\n",
      "Epoch 3810, Loss 26.574646\n",
      "Epoch 3811, Loss 26.573847\n",
      "Epoch 3812, Loss 26.573042\n",
      "Epoch 3813, Loss 26.572237\n",
      "Epoch 3814, Loss 26.571442\n",
      "Epoch 3815, Loss 26.570635\n",
      "Epoch 3816, Loss 26.569830\n",
      "Epoch 3817, Loss 26.569031\n",
      "Epoch 3818, Loss 26.568226\n",
      "Epoch 3819, Loss 26.567419\n",
      "Epoch 3820, Loss 26.566614\n",
      "Epoch 3821, Loss 26.565813\n",
      "Epoch 3822, Loss 26.565010\n",
      "Epoch 3823, Loss 26.564203\n",
      "Epoch 3824, Loss 26.563404\n",
      "Epoch 3825, Loss 26.562603\n",
      "Epoch 3826, Loss 26.561800\n",
      "Epoch 3827, Loss 26.560997\n",
      "Epoch 3828, Loss 26.560192\n",
      "Epoch 3829, Loss 26.559387\n",
      "Epoch 3830, Loss 26.558586\n",
      "Epoch 3831, Loss 26.557787\n",
      "Epoch 3832, Loss 26.556982\n",
      "Epoch 3833, Loss 26.556181\n",
      "Epoch 3834, Loss 26.555376\n",
      "Epoch 3835, Loss 26.554573\n",
      "Epoch 3836, Loss 26.553772\n",
      "Epoch 3837, Loss 26.552965\n",
      "Epoch 3838, Loss 26.552168\n",
      "Epoch 3839, Loss 26.551363\n",
      "Epoch 3840, Loss 26.550562\n",
      "Epoch 3841, Loss 26.549761\n",
      "Epoch 3842, Loss 26.548956\n",
      "Epoch 3843, Loss 26.548151\n",
      "Epoch 3844, Loss 26.547350\n",
      "Epoch 3845, Loss 26.546547\n",
      "Epoch 3846, Loss 26.545746\n",
      "Epoch 3847, Loss 26.544945\n",
      "Epoch 3848, Loss 26.544140\n",
      "Epoch 3849, Loss 26.543337\n",
      "Epoch 3850, Loss 26.542536\n",
      "Epoch 3851, Loss 26.541731\n",
      "Epoch 3852, Loss 26.540932\n",
      "Epoch 3853, Loss 26.540129\n",
      "Epoch 3854, Loss 26.539330\n",
      "Epoch 3855, Loss 26.538525\n",
      "Epoch 3856, Loss 26.537720\n",
      "Epoch 3857, Loss 26.536915\n",
      "Epoch 3858, Loss 26.536116\n",
      "Epoch 3859, Loss 26.535311\n",
      "Epoch 3860, Loss 26.534513\n",
      "Epoch 3861, Loss 26.533709\n",
      "Epoch 3862, Loss 26.532909\n",
      "Epoch 3863, Loss 26.532110\n",
      "Epoch 3864, Loss 26.531305\n",
      "Epoch 3865, Loss 26.530500\n",
      "Epoch 3866, Loss 26.529699\n",
      "Epoch 3867, Loss 26.528898\n",
      "Epoch 3868, Loss 26.528099\n",
      "Epoch 3869, Loss 26.527294\n",
      "Epoch 3870, Loss 26.526495\n",
      "Epoch 3871, Loss 26.525690\n",
      "Epoch 3872, Loss 26.524889\n",
      "Epoch 3873, Loss 26.524086\n",
      "Epoch 3874, Loss 26.523287\n",
      "Epoch 3875, Loss 26.522486\n",
      "Epoch 3876, Loss 26.521681\n",
      "Epoch 3877, Loss 26.520880\n",
      "Epoch 3878, Loss 26.520081\n",
      "Epoch 3879, Loss 26.519281\n",
      "Epoch 3880, Loss 26.518476\n",
      "Epoch 3881, Loss 26.517675\n",
      "Epoch 3882, Loss 26.516874\n",
      "Epoch 3883, Loss 26.516069\n",
      "Epoch 3884, Loss 26.515270\n",
      "Epoch 3885, Loss 26.514467\n",
      "Epoch 3886, Loss 26.513670\n",
      "Epoch 3887, Loss 26.512867\n",
      "Epoch 3888, Loss 26.512066\n",
      "Epoch 3889, Loss 26.511263\n",
      "Epoch 3890, Loss 26.510460\n",
      "Epoch 3891, Loss 26.509663\n",
      "Epoch 3892, Loss 26.508858\n",
      "Epoch 3893, Loss 26.508057\n",
      "Epoch 3894, Loss 26.507257\n",
      "Epoch 3895, Loss 26.506453\n",
      "Epoch 3896, Loss 26.505653\n",
      "Epoch 3897, Loss 26.504850\n",
      "Epoch 3898, Loss 26.504057\n",
      "Epoch 3899, Loss 26.503254\n",
      "Epoch 3900, Loss 26.502453\n",
      "Epoch 3901, Loss 26.501648\n",
      "Epoch 3902, Loss 26.500847\n",
      "Epoch 3903, Loss 26.500044\n",
      "Epoch 3904, Loss 26.499245\n",
      "Epoch 3905, Loss 26.498446\n",
      "Epoch 3906, Loss 26.497643\n",
      "Epoch 3907, Loss 26.496849\n",
      "Epoch 3908, Loss 26.496044\n",
      "Epoch 3909, Loss 26.495245\n",
      "Epoch 3910, Loss 26.494444\n",
      "Epoch 3911, Loss 26.493641\n",
      "Epoch 3912, Loss 26.492842\n",
      "Epoch 3913, Loss 26.492043\n",
      "Epoch 3914, Loss 26.491240\n",
      "Epoch 3915, Loss 26.490440\n",
      "Epoch 3916, Loss 26.489637\n",
      "Epoch 3917, Loss 26.488838\n",
      "Epoch 3918, Loss 26.488043\n",
      "Epoch 3919, Loss 26.487242\n",
      "Epoch 3920, Loss 26.486439\n",
      "Epoch 3921, Loss 26.485640\n",
      "Epoch 3922, Loss 26.484835\n",
      "Epoch 3923, Loss 26.484037\n",
      "Epoch 3924, Loss 26.483234\n",
      "Epoch 3925, Loss 26.482433\n",
      "Epoch 3926, Loss 26.481640\n",
      "Epoch 3927, Loss 26.480837\n",
      "Epoch 3928, Loss 26.480036\n",
      "Epoch 3929, Loss 26.479237\n",
      "Epoch 3930, Loss 26.478432\n",
      "Epoch 3931, Loss 26.477633\n",
      "Epoch 3932, Loss 26.476835\n",
      "Epoch 3933, Loss 26.476032\n",
      "Epoch 3934, Loss 26.475231\n",
      "Epoch 3935, Loss 26.474438\n",
      "Epoch 3936, Loss 26.473633\n",
      "Epoch 3937, Loss 26.472834\n",
      "Epoch 3938, Loss 26.472034\n",
      "Epoch 3939, Loss 26.471235\n",
      "Epoch 3940, Loss 26.470440\n",
      "Epoch 3941, Loss 26.469635\n",
      "Epoch 3942, Loss 26.468840\n",
      "Epoch 3943, Loss 26.468035\n",
      "Epoch 3944, Loss 26.467236\n",
      "Epoch 3945, Loss 26.466438\n",
      "Epoch 3946, Loss 26.465643\n",
      "Epoch 3947, Loss 26.464842\n",
      "Epoch 3948, Loss 26.464039\n",
      "Epoch 3949, Loss 26.463240\n",
      "Epoch 3950, Loss 26.462439\n",
      "Epoch 3951, Loss 26.461639\n",
      "Epoch 3952, Loss 26.460844\n",
      "Epoch 3953, Loss 26.460041\n",
      "Epoch 3954, Loss 26.459246\n",
      "Epoch 3955, Loss 26.458443\n",
      "Epoch 3956, Loss 26.457642\n",
      "Epoch 3957, Loss 26.456841\n",
      "Epoch 3958, Loss 26.456041\n",
      "Epoch 3959, Loss 26.455242\n",
      "Epoch 3960, Loss 26.454445\n",
      "Epoch 3961, Loss 26.453642\n",
      "Epoch 3962, Loss 26.452845\n",
      "Epoch 3963, Loss 26.452045\n",
      "Epoch 3964, Loss 26.451246\n",
      "Epoch 3965, Loss 26.450447\n",
      "Epoch 3966, Loss 26.449652\n",
      "Epoch 3967, Loss 26.448853\n",
      "Epoch 3968, Loss 26.448053\n",
      "Epoch 3969, Loss 26.447248\n",
      "Epoch 3970, Loss 26.446453\n",
      "Epoch 3971, Loss 26.445650\n",
      "Epoch 3972, Loss 26.444857\n",
      "Epoch 3973, Loss 26.444056\n",
      "Epoch 3974, Loss 26.443260\n",
      "Epoch 3975, Loss 26.442461\n",
      "Epoch 3976, Loss 26.441662\n",
      "Epoch 3977, Loss 26.440857\n",
      "Epoch 3978, Loss 26.440058\n",
      "Epoch 3979, Loss 26.439262\n",
      "Epoch 3980, Loss 26.438463\n",
      "Epoch 3981, Loss 26.437666\n",
      "Epoch 3982, Loss 26.436871\n",
      "Epoch 3983, Loss 26.436071\n",
      "Epoch 3984, Loss 26.435270\n",
      "Epoch 3985, Loss 26.434471\n",
      "Epoch 3986, Loss 26.433672\n",
      "Epoch 3987, Loss 26.432878\n",
      "Epoch 3988, Loss 26.432074\n",
      "Epoch 3989, Loss 26.431274\n",
      "Epoch 3990, Loss 26.430475\n",
      "Epoch 3991, Loss 26.429682\n",
      "Epoch 3992, Loss 26.428881\n",
      "Epoch 3993, Loss 26.428078\n",
      "Epoch 3994, Loss 26.427284\n",
      "Epoch 3995, Loss 26.426485\n",
      "Epoch 3996, Loss 26.425690\n",
      "Epoch 3997, Loss 26.424885\n",
      "Epoch 3998, Loss 26.424089\n",
      "Epoch 3999, Loss 26.423290\n",
      "Epoch 4000, Loss 26.422497\n",
      "Epoch 4001, Loss 26.421698\n",
      "Epoch 4002, Loss 26.420900\n",
      "Epoch 4003, Loss 26.420103\n",
      "Epoch 4004, Loss 26.419306\n",
      "Epoch 4005, Loss 26.418505\n",
      "Epoch 4006, Loss 26.417706\n",
      "Epoch 4007, Loss 26.416906\n",
      "Epoch 4008, Loss 26.416107\n",
      "Epoch 4009, Loss 26.415314\n",
      "Epoch 4010, Loss 26.414518\n",
      "Epoch 4011, Loss 26.413715\n",
      "Epoch 4012, Loss 26.412916\n",
      "Epoch 4013, Loss 26.412121\n",
      "Epoch 4014, Loss 26.411322\n",
      "Epoch 4015, Loss 26.410522\n",
      "Epoch 4016, Loss 26.409727\n",
      "Epoch 4017, Loss 26.408928\n",
      "Epoch 4018, Loss 26.408131\n",
      "Epoch 4019, Loss 26.407335\n",
      "Epoch 4020, Loss 26.406536\n",
      "Epoch 4021, Loss 26.405743\n",
      "Epoch 4022, Loss 26.404940\n",
      "Epoch 4023, Loss 26.404142\n",
      "Epoch 4024, Loss 26.403349\n",
      "Epoch 4025, Loss 26.402550\n",
      "Epoch 4026, Loss 26.401751\n",
      "Epoch 4027, Loss 26.400951\n",
      "Epoch 4028, Loss 26.400158\n",
      "Epoch 4029, Loss 26.399359\n",
      "Epoch 4030, Loss 26.398561\n",
      "Epoch 4031, Loss 26.397766\n",
      "Epoch 4032, Loss 26.396961\n",
      "Epoch 4033, Loss 26.396168\n",
      "Epoch 4034, Loss 26.395369\n",
      "Epoch 4035, Loss 26.394575\n",
      "Epoch 4036, Loss 26.393780\n",
      "Epoch 4037, Loss 26.392975\n",
      "Epoch 4038, Loss 26.392181\n",
      "Epoch 4039, Loss 26.391386\n",
      "Epoch 4040, Loss 26.390589\n",
      "Epoch 4041, Loss 26.389793\n",
      "Epoch 4042, Loss 26.388994\n",
      "Epoch 4043, Loss 26.388195\n",
      "Epoch 4044, Loss 26.387402\n",
      "Epoch 4045, Loss 26.386599\n",
      "Epoch 4046, Loss 26.385801\n",
      "Epoch 4047, Loss 26.385010\n",
      "Epoch 4048, Loss 26.384211\n",
      "Epoch 4049, Loss 26.383415\n",
      "Epoch 4050, Loss 26.382618\n",
      "Epoch 4051, Loss 26.381823\n",
      "Epoch 4052, Loss 26.381029\n",
      "Epoch 4053, Loss 26.380226\n",
      "Epoch 4054, Loss 26.379427\n",
      "Epoch 4055, Loss 26.378632\n",
      "Epoch 4056, Loss 26.377840\n",
      "Epoch 4057, Loss 26.377043\n",
      "Epoch 4058, Loss 26.376244\n",
      "Epoch 4059, Loss 26.375450\n",
      "Epoch 4060, Loss 26.374653\n",
      "Epoch 4061, Loss 26.373852\n",
      "Epoch 4062, Loss 26.373058\n",
      "Epoch 4063, Loss 26.372259\n",
      "Epoch 4064, Loss 26.371460\n",
      "Epoch 4065, Loss 26.370668\n",
      "Epoch 4066, Loss 26.369875\n",
      "Epoch 4067, Loss 26.369080\n",
      "Epoch 4068, Loss 26.368280\n",
      "Epoch 4069, Loss 26.367479\n",
      "Epoch 4070, Loss 26.366688\n",
      "Epoch 4071, Loss 26.365887\n",
      "Epoch 4072, Loss 26.365093\n",
      "Epoch 4073, Loss 26.364296\n",
      "Epoch 4074, Loss 26.363495\n",
      "Epoch 4075, Loss 26.362703\n",
      "Epoch 4076, Loss 26.361910\n",
      "Epoch 4077, Loss 26.361107\n",
      "Epoch 4078, Loss 26.360319\n",
      "Epoch 4079, Loss 26.359522\n",
      "Epoch 4080, Loss 26.358721\n",
      "Epoch 4081, Loss 26.357927\n",
      "Epoch 4082, Loss 26.357134\n",
      "Epoch 4083, Loss 26.356331\n",
      "Epoch 4084, Loss 26.355541\n",
      "Epoch 4085, Loss 26.354744\n",
      "Epoch 4086, Loss 26.353945\n",
      "Epoch 4087, Loss 26.353151\n",
      "Epoch 4088, Loss 26.352356\n",
      "Epoch 4089, Loss 26.351557\n",
      "Epoch 4090, Loss 26.350763\n",
      "Epoch 4091, Loss 26.349970\n",
      "Epoch 4092, Loss 26.349171\n",
      "Epoch 4093, Loss 26.348377\n",
      "Epoch 4094, Loss 26.347584\n",
      "Epoch 4095, Loss 26.346785\n",
      "Epoch 4096, Loss 26.345995\n",
      "Epoch 4097, Loss 26.345192\n",
      "Epoch 4098, Loss 26.344398\n",
      "Epoch 4099, Loss 26.343605\n",
      "Epoch 4100, Loss 26.342804\n",
      "Epoch 4101, Loss 26.342010\n",
      "Epoch 4102, Loss 26.341219\n",
      "Epoch 4103, Loss 26.340422\n",
      "Epoch 4104, Loss 26.339628\n",
      "Epoch 4105, Loss 26.338835\n",
      "Epoch 4106, Loss 26.338036\n",
      "Epoch 4107, Loss 26.337242\n",
      "Epoch 4108, Loss 26.336443\n",
      "Epoch 4109, Loss 26.335649\n",
      "Epoch 4110, Loss 26.334852\n",
      "Epoch 4111, Loss 26.334057\n",
      "Epoch 4112, Loss 26.333265\n",
      "Epoch 4113, Loss 26.332470\n",
      "Epoch 4114, Loss 26.331676\n",
      "Epoch 4115, Loss 26.330877\n",
      "Epoch 4116, Loss 26.330080\n",
      "Epoch 4117, Loss 26.329285\n",
      "Epoch 4118, Loss 26.328491\n",
      "Epoch 4119, Loss 26.327698\n",
      "Epoch 4120, Loss 26.326899\n",
      "Epoch 4121, Loss 26.326109\n",
      "Epoch 4122, Loss 26.325312\n",
      "Epoch 4123, Loss 26.324516\n",
      "Epoch 4124, Loss 26.323719\n",
      "Epoch 4125, Loss 26.322926\n",
      "Epoch 4126, Loss 26.322132\n",
      "Epoch 4127, Loss 26.321339\n",
      "Epoch 4128, Loss 26.320545\n",
      "Epoch 4129, Loss 26.319746\n",
      "Epoch 4130, Loss 26.318947\n",
      "Epoch 4131, Loss 26.318159\n",
      "Epoch 4132, Loss 26.317366\n",
      "Epoch 4133, Loss 26.316566\n",
      "Epoch 4134, Loss 26.315773\n",
      "Epoch 4135, Loss 26.314980\n",
      "Epoch 4136, Loss 26.314184\n",
      "Epoch 4137, Loss 26.313391\n",
      "Epoch 4138, Loss 26.312593\n",
      "Epoch 4139, Loss 26.311804\n",
      "Epoch 4140, Loss 26.311007\n",
      "Epoch 4141, Loss 26.310217\n",
      "Epoch 4142, Loss 26.309418\n",
      "Epoch 4143, Loss 26.308626\n",
      "Epoch 4144, Loss 26.307829\n",
      "Epoch 4145, Loss 26.307035\n",
      "Epoch 4146, Loss 26.306242\n",
      "Epoch 4147, Loss 26.305449\n",
      "Epoch 4148, Loss 26.304649\n",
      "Epoch 4149, Loss 26.303862\n",
      "Epoch 4150, Loss 26.303064\n",
      "Epoch 4151, Loss 26.302271\n",
      "Epoch 4152, Loss 26.301476\n",
      "Epoch 4153, Loss 26.300682\n",
      "Epoch 4154, Loss 26.299883\n",
      "Epoch 4155, Loss 26.299095\n",
      "Epoch 4156, Loss 26.298296\n",
      "Epoch 4157, Loss 26.297503\n",
      "Epoch 4158, Loss 26.296709\n",
      "Epoch 4159, Loss 26.295921\n",
      "Epoch 4160, Loss 26.295124\n",
      "Epoch 4161, Loss 26.294331\n",
      "Epoch 4162, Loss 26.293535\n",
      "Epoch 4163, Loss 26.292742\n",
      "Epoch 4164, Loss 26.291945\n",
      "Epoch 4165, Loss 26.291145\n",
      "Epoch 4166, Loss 26.290356\n",
      "Epoch 4167, Loss 26.289562\n",
      "Epoch 4168, Loss 26.288776\n",
      "Epoch 4169, Loss 26.287975\n",
      "Epoch 4170, Loss 26.287188\n",
      "Epoch 4171, Loss 26.286388\n",
      "Epoch 4172, Loss 26.285597\n",
      "Epoch 4173, Loss 26.284803\n",
      "Epoch 4174, Loss 26.284008\n",
      "Epoch 4175, Loss 26.283215\n",
      "Epoch 4176, Loss 26.282421\n",
      "Epoch 4177, Loss 26.281633\n",
      "Epoch 4178, Loss 26.280836\n",
      "Epoch 4179, Loss 26.280046\n",
      "Epoch 4180, Loss 26.279249\n",
      "Epoch 4181, Loss 26.278460\n",
      "Epoch 4182, Loss 26.277660\n",
      "Epoch 4183, Loss 26.276867\n",
      "Epoch 4184, Loss 26.276073\n",
      "Epoch 4185, Loss 26.275280\n",
      "Epoch 4186, Loss 26.274492\n",
      "Epoch 4187, Loss 26.273695\n",
      "Epoch 4188, Loss 26.272900\n",
      "Epoch 4189, Loss 26.272112\n",
      "Epoch 4190, Loss 26.271313\n",
      "Epoch 4191, Loss 26.270525\n",
      "Epoch 4192, Loss 26.269728\n",
      "Epoch 4193, Loss 26.268932\n",
      "Epoch 4194, Loss 26.268147\n",
      "Epoch 4195, Loss 26.267351\n",
      "Epoch 4196, Loss 26.266560\n",
      "Epoch 4197, Loss 26.265764\n",
      "Epoch 4198, Loss 26.264971\n",
      "Epoch 4199, Loss 26.264177\n",
      "Epoch 4200, Loss 26.263386\n",
      "Epoch 4201, Loss 26.262592\n",
      "Epoch 4202, Loss 26.261803\n",
      "Epoch 4203, Loss 26.261009\n",
      "Epoch 4204, Loss 26.260218\n",
      "Epoch 4205, Loss 26.259422\n",
      "Epoch 4206, Loss 26.258629\n",
      "Epoch 4207, Loss 26.257837\n",
      "Epoch 4208, Loss 26.257048\n",
      "Epoch 4209, Loss 26.256250\n",
      "Epoch 4210, Loss 26.255461\n",
      "Epoch 4211, Loss 26.254667\n",
      "Epoch 4212, Loss 26.253874\n",
      "Epoch 4213, Loss 26.253082\n",
      "Epoch 4214, Loss 26.252289\n",
      "Epoch 4215, Loss 26.251493\n",
      "Epoch 4216, Loss 26.250708\n",
      "Epoch 4217, Loss 26.249916\n",
      "Epoch 4218, Loss 26.249117\n",
      "Epoch 4219, Loss 26.248333\n",
      "Epoch 4220, Loss 26.247536\n",
      "Epoch 4221, Loss 26.246742\n",
      "Epoch 4222, Loss 26.245949\n",
      "Epoch 4223, Loss 26.245155\n",
      "Epoch 4224, Loss 26.244366\n",
      "Epoch 4225, Loss 26.243578\n",
      "Epoch 4226, Loss 26.242785\n",
      "Epoch 4227, Loss 26.241991\n",
      "Epoch 4228, Loss 26.241203\n",
      "Epoch 4229, Loss 26.240404\n",
      "Epoch 4230, Loss 26.239613\n",
      "Epoch 4231, Loss 26.238825\n",
      "Epoch 4232, Loss 26.238031\n",
      "Epoch 4233, Loss 26.237242\n",
      "Epoch 4234, Loss 26.236444\n",
      "Epoch 4235, Loss 26.235657\n",
      "Epoch 4236, Loss 26.234863\n",
      "Epoch 4237, Loss 26.234072\n",
      "Epoch 4238, Loss 26.233276\n",
      "Epoch 4239, Loss 26.232489\n",
      "Epoch 4240, Loss 26.231697\n",
      "Epoch 4241, Loss 26.230902\n",
      "Epoch 4242, Loss 26.230114\n",
      "Epoch 4243, Loss 26.229321\n",
      "Epoch 4244, Loss 26.228527\n",
      "Epoch 4245, Loss 26.227736\n",
      "Epoch 4246, Loss 26.226946\n",
      "Epoch 4247, Loss 26.226158\n",
      "Epoch 4248, Loss 26.225365\n",
      "Epoch 4249, Loss 26.224571\n",
      "Epoch 4250, Loss 26.223783\n",
      "Epoch 4251, Loss 26.222990\n",
      "Epoch 4252, Loss 26.222198\n",
      "Epoch 4253, Loss 26.221403\n",
      "Epoch 4254, Loss 26.220617\n",
      "Epoch 4255, Loss 26.219826\n",
      "Epoch 4256, Loss 26.219032\n",
      "Epoch 4257, Loss 26.218239\n",
      "Epoch 4258, Loss 26.217455\n",
      "Epoch 4259, Loss 26.216658\n",
      "Epoch 4260, Loss 26.215868\n",
      "Epoch 4261, Loss 26.215075\n",
      "Epoch 4262, Loss 26.214287\n",
      "Epoch 4263, Loss 26.213495\n",
      "Epoch 4264, Loss 26.212702\n",
      "Epoch 4265, Loss 26.211914\n",
      "Epoch 4266, Loss 26.211121\n",
      "Epoch 4267, Loss 26.210329\n",
      "Epoch 4268, Loss 26.209539\n",
      "Epoch 4269, Loss 26.208748\n",
      "Epoch 4270, Loss 26.207958\n",
      "Epoch 4271, Loss 26.207165\n",
      "Epoch 4272, Loss 26.206377\n",
      "Epoch 4273, Loss 26.205589\n",
      "Epoch 4274, Loss 26.204796\n",
      "Epoch 4275, Loss 26.204002\n",
      "Epoch 4276, Loss 26.203215\n",
      "Epoch 4277, Loss 26.202415\n",
      "Epoch 4278, Loss 26.201635\n",
      "Epoch 4279, Loss 26.200844\n",
      "Epoch 4280, Loss 26.200050\n",
      "Epoch 4281, Loss 26.199261\n",
      "Epoch 4282, Loss 26.198463\n",
      "Epoch 4283, Loss 26.197676\n",
      "Epoch 4284, Loss 26.196888\n",
      "Epoch 4285, Loss 26.196098\n",
      "Epoch 4286, Loss 26.195305\n",
      "Epoch 4287, Loss 26.194513\n",
      "Epoch 4288, Loss 26.193724\n",
      "Epoch 4289, Loss 26.192938\n",
      "Epoch 4290, Loss 26.192144\n",
      "Epoch 4291, Loss 26.191353\n",
      "Epoch 4292, Loss 26.190563\n",
      "Epoch 4293, Loss 26.189772\n",
      "Epoch 4294, Loss 26.188984\n",
      "Epoch 4295, Loss 26.188194\n",
      "Epoch 4296, Loss 26.187401\n",
      "Epoch 4297, Loss 26.186615\n",
      "Epoch 4298, Loss 26.185823\n",
      "Epoch 4299, Loss 26.185034\n",
      "Epoch 4300, Loss 26.184242\n",
      "Epoch 4301, Loss 26.183453\n",
      "Epoch 4302, Loss 26.182661\n",
      "Epoch 4303, Loss 26.181873\n",
      "Epoch 4304, Loss 26.181078\n",
      "Epoch 4305, Loss 26.180292\n",
      "Epoch 4306, Loss 26.179504\n",
      "Epoch 4307, Loss 26.178713\n",
      "Epoch 4308, Loss 26.177923\n",
      "Epoch 4309, Loss 26.177130\n",
      "Epoch 4310, Loss 26.176342\n",
      "Epoch 4311, Loss 26.175549\n",
      "Epoch 4312, Loss 26.174761\n",
      "Epoch 4313, Loss 26.173975\n",
      "Epoch 4314, Loss 26.173185\n",
      "Epoch 4315, Loss 26.172394\n",
      "Epoch 4316, Loss 26.171608\n",
      "Epoch 4317, Loss 26.170815\n",
      "Epoch 4318, Loss 26.170027\n",
      "Epoch 4319, Loss 26.169233\n",
      "Epoch 4320, Loss 26.168446\n",
      "Epoch 4321, Loss 26.167656\n",
      "Epoch 4322, Loss 26.166864\n",
      "Epoch 4323, Loss 26.166077\n",
      "Epoch 4324, Loss 26.165285\n",
      "Epoch 4325, Loss 26.164497\n",
      "Epoch 4326, Loss 26.163704\n",
      "Epoch 4327, Loss 26.162914\n",
      "Epoch 4328, Loss 26.162127\n",
      "Epoch 4329, Loss 26.161339\n",
      "Epoch 4330, Loss 26.160553\n",
      "Epoch 4331, Loss 26.159765\n",
      "Epoch 4332, Loss 26.158974\n",
      "Epoch 4333, Loss 26.158184\n",
      "Epoch 4334, Loss 26.157393\n",
      "Epoch 4335, Loss 26.156605\n",
      "Epoch 4336, Loss 26.155815\n",
      "Epoch 4337, Loss 26.155024\n",
      "Epoch 4338, Loss 26.154238\n",
      "Epoch 4339, Loss 26.153448\n",
      "Epoch 4340, Loss 26.152660\n",
      "Epoch 4341, Loss 26.151873\n",
      "Epoch 4342, Loss 26.151079\n",
      "Epoch 4343, Loss 26.150293\n",
      "Epoch 4344, Loss 26.149504\n",
      "Epoch 4345, Loss 26.148712\n",
      "Epoch 4346, Loss 26.147924\n",
      "Epoch 4347, Loss 26.147137\n",
      "Epoch 4348, Loss 26.146345\n",
      "Epoch 4349, Loss 26.145557\n",
      "Epoch 4350, Loss 26.144770\n",
      "Epoch 4351, Loss 26.143980\n",
      "Epoch 4352, Loss 26.143190\n",
      "Epoch 4353, Loss 26.142401\n",
      "Epoch 4354, Loss 26.141619\n",
      "Epoch 4355, Loss 26.140827\n",
      "Epoch 4356, Loss 26.140039\n",
      "Epoch 4357, Loss 26.139250\n",
      "Epoch 4358, Loss 26.138460\n",
      "Epoch 4359, Loss 26.137671\n",
      "Epoch 4360, Loss 26.136883\n",
      "Epoch 4361, Loss 26.136095\n",
      "Epoch 4362, Loss 26.135307\n",
      "Epoch 4363, Loss 26.134521\n",
      "Epoch 4364, Loss 26.133728\n",
      "Epoch 4365, Loss 26.132940\n",
      "Epoch 4366, Loss 26.132153\n",
      "Epoch 4367, Loss 26.131365\n",
      "Epoch 4368, Loss 26.130573\n",
      "Epoch 4369, Loss 26.129789\n",
      "Epoch 4370, Loss 26.129000\n",
      "Epoch 4371, Loss 26.128212\n",
      "Epoch 4372, Loss 26.127424\n",
      "Epoch 4373, Loss 26.126631\n",
      "Epoch 4374, Loss 26.125843\n",
      "Epoch 4375, Loss 26.125055\n",
      "Epoch 4376, Loss 26.124268\n",
      "Epoch 4377, Loss 26.123480\n",
      "Epoch 4378, Loss 26.122694\n",
      "Epoch 4379, Loss 26.121904\n",
      "Epoch 4380, Loss 26.121117\n",
      "Epoch 4381, Loss 26.120329\n",
      "Epoch 4382, Loss 26.119537\n",
      "Epoch 4383, Loss 26.118752\n",
      "Epoch 4384, Loss 26.117964\n",
      "Epoch 4385, Loss 26.117180\n",
      "Epoch 4386, Loss 26.116394\n",
      "Epoch 4387, Loss 26.115606\n",
      "Epoch 4388, Loss 26.114809\n",
      "Epoch 4389, Loss 26.114021\n",
      "Epoch 4390, Loss 26.113237\n",
      "Epoch 4391, Loss 26.112450\n",
      "Epoch 4392, Loss 26.111666\n",
      "Epoch 4393, Loss 26.110878\n",
      "Epoch 4394, Loss 26.110090\n",
      "Epoch 4395, Loss 26.109303\n",
      "Epoch 4396, Loss 26.108515\n",
      "Epoch 4397, Loss 26.107721\n",
      "Epoch 4398, Loss 26.106936\n",
      "Epoch 4399, Loss 26.106148\n",
      "Epoch 4400, Loss 26.105364\n",
      "Epoch 4401, Loss 26.104576\n",
      "Epoch 4402, Loss 26.103788\n",
      "Epoch 4403, Loss 26.102999\n",
      "Epoch 4404, Loss 26.102211\n",
      "Epoch 4405, Loss 26.101421\n",
      "Epoch 4406, Loss 26.100636\n",
      "Epoch 4407, Loss 26.099852\n",
      "Epoch 4408, Loss 26.099062\n",
      "Epoch 4409, Loss 26.098274\n",
      "Epoch 4410, Loss 26.097490\n",
      "Epoch 4411, Loss 26.096703\n",
      "Epoch 4412, Loss 26.095915\n",
      "Epoch 4413, Loss 26.095123\n",
      "Epoch 4414, Loss 26.094336\n",
      "Epoch 4415, Loss 26.093550\n",
      "Epoch 4416, Loss 26.092768\n",
      "Epoch 4417, Loss 26.091980\n",
      "Epoch 4418, Loss 26.091192\n",
      "Epoch 4419, Loss 26.090401\n",
      "Epoch 4420, Loss 26.089617\n",
      "Epoch 4421, Loss 26.088831\n",
      "Epoch 4422, Loss 26.088045\n",
      "Epoch 4423, Loss 26.087263\n",
      "Epoch 4424, Loss 26.086473\n",
      "Epoch 4425, Loss 26.085682\n",
      "Epoch 4426, Loss 26.084898\n",
      "Epoch 4427, Loss 26.084108\n",
      "Epoch 4428, Loss 26.083324\n",
      "Epoch 4429, Loss 26.082539\n",
      "Epoch 4430, Loss 26.081749\n",
      "Epoch 4431, Loss 26.080965\n",
      "Epoch 4432, Loss 26.080172\n",
      "Epoch 4433, Loss 26.079390\n",
      "Epoch 4434, Loss 26.078608\n",
      "Epoch 4435, Loss 26.077814\n",
      "Epoch 4436, Loss 26.077026\n",
      "Epoch 4437, Loss 26.076241\n",
      "Epoch 4438, Loss 26.075459\n",
      "Epoch 4439, Loss 26.074675\n",
      "Epoch 4440, Loss 26.073883\n",
      "Epoch 4441, Loss 26.073095\n",
      "Epoch 4442, Loss 26.072309\n",
      "Epoch 4443, Loss 26.071522\n",
      "Epoch 4444, Loss 26.070740\n",
      "Epoch 4445, Loss 26.069952\n",
      "Epoch 4446, Loss 26.069166\n",
      "Epoch 4447, Loss 26.068382\n",
      "Epoch 4448, Loss 26.067596\n",
      "Epoch 4449, Loss 26.066805\n",
      "Epoch 4450, Loss 26.066021\n",
      "Epoch 4451, Loss 26.065233\n",
      "Epoch 4452, Loss 26.064451\n",
      "Epoch 4453, Loss 26.063660\n",
      "Epoch 4454, Loss 26.062874\n",
      "Epoch 4455, Loss 26.062092\n",
      "Epoch 4456, Loss 26.061308\n",
      "Epoch 4457, Loss 26.060516\n",
      "Epoch 4458, Loss 26.059731\n",
      "Epoch 4459, Loss 26.058947\n",
      "Epoch 4460, Loss 26.058165\n",
      "Epoch 4461, Loss 26.057373\n",
      "Epoch 4462, Loss 26.056591\n",
      "Epoch 4463, Loss 26.055803\n",
      "Epoch 4464, Loss 26.055016\n",
      "Epoch 4465, Loss 26.054228\n",
      "Epoch 4466, Loss 26.053444\n",
      "Epoch 4467, Loss 26.052660\n",
      "Epoch 4468, Loss 26.051874\n",
      "Epoch 4469, Loss 26.051086\n",
      "Epoch 4470, Loss 26.050301\n",
      "Epoch 4471, Loss 26.049517\n",
      "Epoch 4472, Loss 26.048729\n",
      "Epoch 4473, Loss 26.047941\n",
      "Epoch 4474, Loss 26.047157\n",
      "Epoch 4475, Loss 26.046375\n",
      "Epoch 4476, Loss 26.045591\n",
      "Epoch 4477, Loss 26.044807\n",
      "Epoch 4478, Loss 26.044018\n",
      "Epoch 4479, Loss 26.043232\n",
      "Epoch 4480, Loss 26.042448\n",
      "Epoch 4481, Loss 26.041658\n",
      "Epoch 4482, Loss 26.040882\n",
      "Epoch 4483, Loss 26.040092\n",
      "Epoch 4484, Loss 26.039305\n",
      "Epoch 4485, Loss 26.038521\n",
      "Epoch 4486, Loss 26.037737\n",
      "Epoch 4487, Loss 26.036955\n",
      "Epoch 4488, Loss 26.036165\n",
      "Epoch 4489, Loss 26.035378\n",
      "Epoch 4490, Loss 26.034599\n",
      "Epoch 4491, Loss 26.033808\n",
      "Epoch 4492, Loss 26.033026\n",
      "Epoch 4493, Loss 26.032244\n",
      "Epoch 4494, Loss 26.031456\n",
      "Epoch 4495, Loss 26.030670\n",
      "Epoch 4496, Loss 26.029884\n",
      "Epoch 4497, Loss 26.029100\n",
      "Epoch 4498, Loss 26.028315\n",
      "Epoch 4499, Loss 26.027529\n",
      "Epoch 4500, Loss 26.026747\n",
      "Epoch 4501, Loss 26.025969\n",
      "Epoch 4502, Loss 26.025177\n",
      "Epoch 4503, Loss 26.024395\n",
      "Epoch 4504, Loss 26.023603\n",
      "Epoch 4505, Loss 26.022821\n",
      "Epoch 4506, Loss 26.022039\n",
      "Epoch 4507, Loss 26.021252\n",
      "Epoch 4508, Loss 26.020470\n",
      "Epoch 4509, Loss 26.019686\n",
      "Epoch 4510, Loss 26.018898\n",
      "Epoch 4511, Loss 26.018116\n",
      "Epoch 4512, Loss 26.017328\n",
      "Epoch 4513, Loss 26.016546\n",
      "Epoch 4514, Loss 26.015764\n",
      "Epoch 4515, Loss 26.014978\n",
      "Epoch 4516, Loss 26.014193\n",
      "Epoch 4517, Loss 26.013411\n",
      "Epoch 4518, Loss 26.012623\n",
      "Epoch 4519, Loss 26.011841\n",
      "Epoch 4520, Loss 26.011059\n",
      "Epoch 4521, Loss 26.010277\n",
      "Epoch 4522, Loss 26.009485\n",
      "Epoch 4523, Loss 26.008705\n",
      "Epoch 4524, Loss 26.007921\n",
      "Epoch 4525, Loss 26.007139\n",
      "Epoch 4526, Loss 26.006353\n",
      "Epoch 4527, Loss 26.005568\n",
      "Epoch 4528, Loss 26.004786\n",
      "Epoch 4529, Loss 26.003998\n",
      "Epoch 4530, Loss 26.003218\n",
      "Epoch 4531, Loss 26.002436\n",
      "Epoch 4532, Loss 26.001648\n",
      "Epoch 4533, Loss 26.000868\n",
      "Epoch 4534, Loss 26.000080\n",
      "Epoch 4535, Loss 25.999298\n",
      "Epoch 4536, Loss 25.998510\n",
      "Epoch 4537, Loss 25.997730\n",
      "Epoch 4538, Loss 25.996941\n",
      "Epoch 4539, Loss 25.996161\n",
      "Epoch 4540, Loss 25.995380\n",
      "Epoch 4541, Loss 25.994596\n",
      "Epoch 4542, Loss 25.993813\n",
      "Epoch 4543, Loss 25.993025\n",
      "Epoch 4544, Loss 25.992243\n",
      "Epoch 4545, Loss 25.991455\n",
      "Epoch 4546, Loss 25.990679\n",
      "Epoch 4547, Loss 25.989893\n",
      "Epoch 4548, Loss 25.989107\n",
      "Epoch 4549, Loss 25.988325\n",
      "Epoch 4550, Loss 25.987543\n",
      "Epoch 4551, Loss 25.986757\n",
      "Epoch 4552, Loss 25.985979\n",
      "Epoch 4553, Loss 25.985191\n",
      "Epoch 4554, Loss 25.984413\n",
      "Epoch 4555, Loss 25.983625\n",
      "Epoch 4556, Loss 25.982843\n",
      "Epoch 4557, Loss 25.982056\n",
      "Epoch 4558, Loss 25.981276\n",
      "Epoch 4559, Loss 25.980494\n",
      "Epoch 4560, Loss 25.979712\n",
      "Epoch 4561, Loss 25.978926\n",
      "Epoch 4562, Loss 25.978144\n",
      "Epoch 4563, Loss 25.977358\n",
      "Epoch 4564, Loss 25.976576\n",
      "Epoch 4565, Loss 25.975792\n",
      "Epoch 4566, Loss 25.975012\n",
      "Epoch 4567, Loss 25.974226\n",
      "Epoch 4568, Loss 25.973450\n",
      "Epoch 4569, Loss 25.972662\n",
      "Epoch 4570, Loss 25.971876\n",
      "Epoch 4571, Loss 25.971098\n",
      "Epoch 4572, Loss 25.970308\n",
      "Epoch 4573, Loss 25.969532\n",
      "Epoch 4574, Loss 25.968744\n",
      "Epoch 4575, Loss 25.967968\n",
      "Epoch 4576, Loss 25.967182\n",
      "Epoch 4577, Loss 25.966402\n",
      "Epoch 4578, Loss 25.965620\n",
      "Epoch 4579, Loss 25.964832\n",
      "Epoch 4580, Loss 25.964052\n",
      "Epoch 4581, Loss 25.963264\n",
      "Epoch 4582, Loss 25.962486\n",
      "Epoch 4583, Loss 25.961702\n",
      "Epoch 4584, Loss 25.960920\n",
      "Epoch 4585, Loss 25.960142\n",
      "Epoch 4586, Loss 25.959352\n",
      "Epoch 4587, Loss 25.958576\n",
      "Epoch 4588, Loss 25.957787\n",
      "Epoch 4589, Loss 25.957008\n",
      "Epoch 4590, Loss 25.956224\n",
      "Epoch 4591, Loss 25.955444\n",
      "Epoch 4592, Loss 25.954662\n",
      "Epoch 4593, Loss 25.953876\n",
      "Epoch 4594, Loss 25.953096\n",
      "Epoch 4595, Loss 25.952314\n",
      "Epoch 4596, Loss 25.951536\n",
      "Epoch 4597, Loss 25.950750\n",
      "Epoch 4598, Loss 25.949968\n",
      "Epoch 4599, Loss 25.949190\n",
      "Epoch 4600, Loss 25.948406\n",
      "Epoch 4601, Loss 25.947624\n",
      "Epoch 4602, Loss 25.946838\n",
      "Epoch 4603, Loss 25.946056\n",
      "Epoch 4604, Loss 25.945280\n",
      "Epoch 4605, Loss 25.944494\n",
      "Epoch 4606, Loss 25.943712\n",
      "Epoch 4607, Loss 25.942926\n",
      "Epoch 4608, Loss 25.942146\n",
      "Epoch 4609, Loss 25.941368\n",
      "Epoch 4610, Loss 25.940584\n",
      "Epoch 4611, Loss 25.939800\n",
      "Epoch 4612, Loss 25.939024\n",
      "Epoch 4613, Loss 25.938240\n",
      "Epoch 4614, Loss 25.937456\n",
      "Epoch 4615, Loss 25.936674\n",
      "Epoch 4616, Loss 25.935894\n",
      "Epoch 4617, Loss 25.935112\n",
      "Epoch 4618, Loss 25.934332\n",
      "Epoch 4619, Loss 25.933546\n",
      "Epoch 4620, Loss 25.932768\n",
      "Epoch 4621, Loss 25.931984\n",
      "Epoch 4622, Loss 25.931202\n",
      "Epoch 4623, Loss 25.930422\n",
      "Epoch 4624, Loss 25.929644\n",
      "Epoch 4625, Loss 25.928862\n",
      "Epoch 4626, Loss 25.928076\n",
      "Epoch 4627, Loss 25.927299\n",
      "Epoch 4628, Loss 25.926514\n",
      "Epoch 4629, Loss 25.925732\n",
      "Epoch 4630, Loss 25.924957\n",
      "Epoch 4631, Loss 25.924171\n",
      "Epoch 4632, Loss 25.923389\n",
      "Epoch 4633, Loss 25.922607\n",
      "Epoch 4634, Loss 25.921825\n",
      "Epoch 4635, Loss 25.921043\n",
      "Epoch 4636, Loss 25.920265\n",
      "Epoch 4637, Loss 25.919483\n",
      "Epoch 4638, Loss 25.918707\n",
      "Epoch 4639, Loss 25.917919\n",
      "Epoch 4640, Loss 25.917143\n",
      "Epoch 4641, Loss 25.916359\n",
      "Epoch 4642, Loss 25.915575\n",
      "Epoch 4643, Loss 25.914801\n",
      "Epoch 4644, Loss 25.914019\n",
      "Epoch 4645, Loss 25.913233\n",
      "Epoch 4646, Loss 25.912451\n",
      "Epoch 4647, Loss 25.911676\n",
      "Epoch 4648, Loss 25.910889\n",
      "Epoch 4649, Loss 25.910112\n",
      "Epoch 4650, Loss 25.909327\n",
      "Epoch 4651, Loss 25.908552\n",
      "Epoch 4652, Loss 25.907770\n",
      "Epoch 4653, Loss 25.906988\n",
      "Epoch 4654, Loss 25.906206\n",
      "Epoch 4655, Loss 25.905428\n",
      "Epoch 4656, Loss 25.904650\n",
      "Epoch 4657, Loss 25.903868\n",
      "Epoch 4658, Loss 25.903084\n",
      "Epoch 4659, Loss 25.902306\n",
      "Epoch 4660, Loss 25.901522\n",
      "Epoch 4661, Loss 25.900745\n",
      "Epoch 4662, Loss 25.899963\n",
      "Epoch 4663, Loss 25.899187\n",
      "Epoch 4664, Loss 25.898401\n",
      "Epoch 4665, Loss 25.897619\n",
      "Epoch 4666, Loss 25.896837\n",
      "Epoch 4667, Loss 25.896063\n",
      "Epoch 4668, Loss 25.895281\n",
      "Epoch 4669, Loss 25.894503\n",
      "Epoch 4670, Loss 25.893715\n",
      "Epoch 4671, Loss 25.892942\n",
      "Epoch 4672, Loss 25.892162\n",
      "Epoch 4673, Loss 25.891376\n",
      "Epoch 4674, Loss 25.890596\n",
      "Epoch 4675, Loss 25.889820\n",
      "Epoch 4676, Loss 25.889038\n",
      "Epoch 4677, Loss 25.888262\n",
      "Epoch 4678, Loss 25.887480\n",
      "Epoch 4679, Loss 25.886700\n",
      "Epoch 4680, Loss 25.885918\n",
      "Epoch 4681, Loss 25.885138\n",
      "Epoch 4682, Loss 25.884363\n",
      "Epoch 4683, Loss 25.883577\n",
      "Epoch 4684, Loss 25.882801\n",
      "Epoch 4685, Loss 25.882019\n",
      "Epoch 4686, Loss 25.881237\n",
      "Epoch 4687, Loss 25.880461\n",
      "Epoch 4688, Loss 25.879683\n",
      "Epoch 4689, Loss 25.878901\n",
      "Epoch 4690, Loss 25.878119\n",
      "Epoch 4691, Loss 25.877338\n",
      "Epoch 4692, Loss 25.876564\n",
      "Epoch 4693, Loss 25.875780\n",
      "Epoch 4694, Loss 25.875002\n",
      "Epoch 4695, Loss 25.874220\n",
      "Epoch 4696, Loss 25.873436\n",
      "Epoch 4697, Loss 25.872662\n",
      "Epoch 4698, Loss 25.871880\n",
      "Epoch 4699, Loss 25.871107\n",
      "Epoch 4700, Loss 25.870323\n",
      "Epoch 4701, Loss 25.869545\n",
      "Epoch 4702, Loss 25.868767\n",
      "Epoch 4703, Loss 25.867987\n",
      "Epoch 4704, Loss 25.867205\n",
      "Epoch 4705, Loss 25.866426\n",
      "Epoch 4706, Loss 25.865643\n",
      "Epoch 4707, Loss 25.864870\n",
      "Epoch 4708, Loss 25.864092\n",
      "Epoch 4709, Loss 25.863310\n",
      "Epoch 4710, Loss 25.862526\n",
      "Epoch 4711, Loss 25.861753\n",
      "Epoch 4712, Loss 25.860970\n",
      "Epoch 4713, Loss 25.860191\n",
      "Epoch 4714, Loss 25.859411\n",
      "Epoch 4715, Loss 25.858637\n",
      "Epoch 4716, Loss 25.857857\n",
      "Epoch 4717, Loss 25.857075\n",
      "Epoch 4718, Loss 25.856293\n",
      "Epoch 4719, Loss 25.855516\n",
      "Epoch 4720, Loss 25.854736\n",
      "Epoch 4721, Loss 25.853956\n",
      "Epoch 4722, Loss 25.853178\n",
      "Epoch 4723, Loss 25.852398\n",
      "Epoch 4724, Loss 25.851622\n",
      "Epoch 4725, Loss 25.850847\n",
      "Epoch 4726, Loss 25.850065\n",
      "Epoch 4727, Loss 25.849287\n",
      "Epoch 4728, Loss 25.848511\n",
      "Epoch 4729, Loss 25.847729\n",
      "Epoch 4730, Loss 25.846952\n",
      "Epoch 4731, Loss 25.846169\n",
      "Epoch 4732, Loss 25.845392\n",
      "Epoch 4733, Loss 25.844616\n",
      "Epoch 4734, Loss 25.843836\n",
      "Epoch 4735, Loss 25.843056\n",
      "Epoch 4736, Loss 25.842278\n",
      "Epoch 4737, Loss 25.841497\n",
      "Epoch 4738, Loss 25.840715\n",
      "Epoch 4739, Loss 25.839937\n",
      "Epoch 4740, Loss 25.839165\n",
      "Epoch 4741, Loss 25.838390\n",
      "Epoch 4742, Loss 25.837610\n",
      "Epoch 4743, Loss 25.836828\n",
      "Epoch 4744, Loss 25.836052\n",
      "Epoch 4745, Loss 25.835268\n",
      "Epoch 4746, Loss 25.834496\n",
      "Epoch 4747, Loss 25.833717\n",
      "Epoch 4748, Loss 25.832935\n",
      "Epoch 4749, Loss 25.832159\n",
      "Epoch 4750, Loss 25.831377\n",
      "Epoch 4751, Loss 25.830603\n",
      "Epoch 4752, Loss 25.829826\n",
      "Epoch 4753, Loss 25.829046\n",
      "Epoch 4754, Loss 25.828270\n",
      "Epoch 4755, Loss 25.827486\n",
      "Epoch 4756, Loss 25.826710\n",
      "Epoch 4757, Loss 25.825933\n",
      "Epoch 4758, Loss 25.825153\n",
      "Epoch 4759, Loss 25.824379\n",
      "Epoch 4760, Loss 25.823597\n",
      "Epoch 4761, Loss 25.822821\n",
      "Epoch 4762, Loss 25.822044\n",
      "Epoch 4763, Loss 25.821264\n",
      "Epoch 4764, Loss 25.820490\n",
      "Epoch 4765, Loss 25.819710\n",
      "Epoch 4766, Loss 25.818933\n",
      "Epoch 4767, Loss 25.818153\n",
      "Epoch 4768, Loss 25.817375\n",
      "Epoch 4769, Loss 25.816601\n",
      "Epoch 4770, Loss 25.815825\n",
      "Epoch 4771, Loss 25.815046\n",
      "Epoch 4772, Loss 25.814264\n",
      "Epoch 4773, Loss 25.813488\n",
      "Epoch 4774, Loss 25.812712\n",
      "Epoch 4775, Loss 25.811934\n",
      "Epoch 4776, Loss 25.811163\n",
      "Epoch 4777, Loss 25.810387\n",
      "Epoch 4778, Loss 25.809603\n",
      "Epoch 4779, Loss 25.808821\n",
      "Epoch 4780, Loss 25.808044\n",
      "Epoch 4781, Loss 25.807268\n",
      "Epoch 4782, Loss 25.806494\n",
      "Epoch 4783, Loss 25.805717\n",
      "Epoch 4784, Loss 25.804939\n",
      "Epoch 4785, Loss 25.804161\n",
      "Epoch 4786, Loss 25.803383\n",
      "Epoch 4787, Loss 25.802610\n",
      "Epoch 4788, Loss 25.801830\n",
      "Epoch 4789, Loss 25.801054\n",
      "Epoch 4790, Loss 25.800274\n",
      "Epoch 4791, Loss 25.799500\n",
      "Epoch 4792, Loss 25.798723\n",
      "Epoch 4793, Loss 25.797945\n",
      "Epoch 4794, Loss 25.797173\n",
      "Epoch 4795, Loss 25.796387\n",
      "Epoch 4796, Loss 25.795607\n",
      "Epoch 4797, Loss 25.794830\n",
      "Epoch 4798, Loss 25.794060\n",
      "Epoch 4799, Loss 25.793282\n",
      "Epoch 4800, Loss 25.792505\n",
      "Epoch 4801, Loss 25.791729\n",
      "Epoch 4802, Loss 25.790955\n",
      "Epoch 4803, Loss 25.790178\n",
      "Epoch 4804, Loss 25.789400\n",
      "Epoch 4805, Loss 25.788624\n",
      "Epoch 4806, Loss 25.787848\n",
      "Epoch 4807, Loss 25.787071\n",
      "Epoch 4808, Loss 25.786293\n",
      "Epoch 4809, Loss 25.785517\n",
      "Epoch 4810, Loss 25.784742\n",
      "Epoch 4811, Loss 25.783966\n",
      "Epoch 4812, Loss 25.783186\n",
      "Epoch 4813, Loss 25.782410\n",
      "Epoch 4814, Loss 25.781633\n",
      "Epoch 4815, Loss 25.780859\n",
      "Epoch 4816, Loss 25.780079\n",
      "Epoch 4817, Loss 25.779305\n",
      "Epoch 4818, Loss 25.778532\n",
      "Epoch 4819, Loss 25.777754\n",
      "Epoch 4820, Loss 25.776978\n",
      "Epoch 4821, Loss 25.776197\n",
      "Epoch 4822, Loss 25.775423\n",
      "Epoch 4823, Loss 25.774647\n",
      "Epoch 4824, Loss 25.773870\n",
      "Epoch 4825, Loss 25.773094\n",
      "Epoch 4826, Loss 25.772326\n",
      "Epoch 4827, Loss 25.771540\n",
      "Epoch 4828, Loss 25.770769\n",
      "Epoch 4829, Loss 25.769991\n",
      "Epoch 4830, Loss 25.769215\n",
      "Epoch 4831, Loss 25.768440\n",
      "Epoch 4832, Loss 25.767662\n",
      "Epoch 4833, Loss 25.766882\n",
      "Epoch 4834, Loss 25.766111\n",
      "Epoch 4835, Loss 25.765337\n",
      "Epoch 4836, Loss 25.764559\n",
      "Epoch 4837, Loss 25.763783\n",
      "Epoch 4838, Loss 25.763006\n",
      "Epoch 4839, Loss 25.762232\n",
      "Epoch 4840, Loss 25.761456\n",
      "Epoch 4841, Loss 25.760679\n",
      "Epoch 4842, Loss 25.759907\n",
      "Epoch 4843, Loss 25.759127\n",
      "Epoch 4844, Loss 25.758354\n",
      "Epoch 4845, Loss 25.757576\n",
      "Epoch 4846, Loss 25.756804\n",
      "Epoch 4847, Loss 25.756025\n",
      "Epoch 4848, Loss 25.755247\n",
      "Epoch 4849, Loss 25.754475\n",
      "Epoch 4850, Loss 25.753698\n",
      "Epoch 4851, Loss 25.752928\n",
      "Epoch 4852, Loss 25.752144\n",
      "Epoch 4853, Loss 25.751371\n",
      "Epoch 4854, Loss 25.750597\n",
      "Epoch 4855, Loss 25.749823\n",
      "Epoch 4856, Loss 25.749043\n",
      "Epoch 4857, Loss 25.748268\n",
      "Epoch 4858, Loss 25.747496\n",
      "Epoch 4859, Loss 25.746721\n",
      "Epoch 4860, Loss 25.745947\n",
      "Epoch 4861, Loss 25.745171\n",
      "Epoch 4862, Loss 25.744396\n",
      "Epoch 4863, Loss 25.743620\n",
      "Epoch 4864, Loss 25.742842\n",
      "Epoch 4865, Loss 25.742071\n",
      "Epoch 4866, Loss 25.741295\n",
      "Epoch 4867, Loss 25.740522\n",
      "Epoch 4868, Loss 25.739740\n",
      "Epoch 4869, Loss 25.738970\n",
      "Epoch 4870, Loss 25.738192\n",
      "Epoch 4871, Loss 25.737425\n",
      "Epoch 4872, Loss 25.736645\n",
      "Epoch 4873, Loss 25.735868\n",
      "Epoch 4874, Loss 25.735090\n",
      "Epoch 4875, Loss 25.734320\n",
      "Epoch 4876, Loss 25.733543\n",
      "Epoch 4877, Loss 25.732771\n",
      "Epoch 4878, Loss 25.731997\n",
      "Epoch 4879, Loss 25.731220\n",
      "Epoch 4880, Loss 25.730446\n",
      "Epoch 4881, Loss 25.729670\n",
      "Epoch 4882, Loss 25.728895\n",
      "Epoch 4883, Loss 25.728125\n",
      "Epoch 4884, Loss 25.727348\n",
      "Epoch 4885, Loss 25.726570\n",
      "Epoch 4886, Loss 25.725798\n",
      "Epoch 4887, Loss 25.725019\n",
      "Epoch 4888, Loss 25.724249\n",
      "Epoch 4889, Loss 25.723478\n",
      "Epoch 4890, Loss 25.722700\n",
      "Epoch 4891, Loss 25.721930\n",
      "Epoch 4892, Loss 25.721149\n",
      "Epoch 4893, Loss 25.720381\n",
      "Epoch 4894, Loss 25.719610\n",
      "Epoch 4895, Loss 25.718828\n",
      "Epoch 4896, Loss 25.718056\n",
      "Epoch 4897, Loss 25.717279\n",
      "Epoch 4898, Loss 25.716503\n",
      "Epoch 4899, Loss 25.715734\n",
      "Epoch 4900, Loss 25.714960\n",
      "Epoch 4901, Loss 25.714184\n",
      "Epoch 4902, Loss 25.713408\n",
      "Epoch 4903, Loss 25.712635\n",
      "Epoch 4904, Loss 25.711859\n",
      "Epoch 4905, Loss 25.711090\n",
      "Epoch 4906, Loss 25.710310\n",
      "Epoch 4907, Loss 25.709539\n",
      "Epoch 4908, Loss 25.708765\n",
      "Epoch 4909, Loss 25.707989\n",
      "Epoch 4910, Loss 25.707216\n",
      "Epoch 4911, Loss 25.706450\n",
      "Epoch 4912, Loss 25.705666\n",
      "Epoch 4913, Loss 25.704895\n",
      "Epoch 4914, Loss 25.704119\n",
      "Epoch 4915, Loss 25.703346\n",
      "Epoch 4916, Loss 25.702578\n",
      "Epoch 4917, Loss 25.701799\n",
      "Epoch 4918, Loss 25.701025\n",
      "Epoch 4919, Loss 25.700254\n",
      "Epoch 4920, Loss 25.699480\n",
      "Epoch 4921, Loss 25.698708\n",
      "Epoch 4922, Loss 25.697931\n",
      "Epoch 4923, Loss 25.697161\n",
      "Epoch 4924, Loss 25.696386\n",
      "Epoch 4925, Loss 25.695606\n",
      "Epoch 4926, Loss 25.694836\n",
      "Epoch 4927, Loss 25.694061\n",
      "Epoch 4928, Loss 25.693293\n",
      "Epoch 4929, Loss 25.692516\n",
      "Epoch 4930, Loss 25.691740\n",
      "Epoch 4931, Loss 25.690973\n",
      "Epoch 4932, Loss 25.690197\n",
      "Epoch 4933, Loss 25.689425\n",
      "Epoch 4934, Loss 25.688648\n",
      "Epoch 4935, Loss 25.687878\n",
      "Epoch 4936, Loss 25.687101\n",
      "Epoch 4937, Loss 25.686329\n",
      "Epoch 4938, Loss 25.685560\n",
      "Epoch 4939, Loss 25.684786\n",
      "Epoch 4940, Loss 25.684013\n",
      "Epoch 4941, Loss 25.683235\n",
      "Epoch 4942, Loss 25.682465\n",
      "Epoch 4943, Loss 25.681688\n",
      "Epoch 4944, Loss 25.680922\n",
      "Epoch 4945, Loss 25.680143\n",
      "Epoch 4946, Loss 25.679375\n",
      "Epoch 4947, Loss 25.678600\n",
      "Epoch 4948, Loss 25.677826\n",
      "Epoch 4949, Loss 25.677057\n",
      "Epoch 4950, Loss 25.676281\n",
      "Epoch 4951, Loss 25.675508\n",
      "Epoch 4952, Loss 25.674732\n",
      "Epoch 4953, Loss 25.673962\n",
      "Epoch 4954, Loss 25.673187\n",
      "Epoch 4955, Loss 25.672422\n",
      "Epoch 4956, Loss 25.671642\n",
      "Epoch 4957, Loss 25.670870\n",
      "Epoch 4958, Loss 25.670097\n",
      "Epoch 4959, Loss 25.669329\n",
      "Epoch 4960, Loss 25.668550\n",
      "Epoch 4961, Loss 25.667784\n",
      "Epoch 4962, Loss 25.667006\n",
      "Epoch 4963, Loss 25.666237\n",
      "Epoch 4964, Loss 25.665461\n",
      "Epoch 4965, Loss 25.664690\n",
      "Epoch 4966, Loss 25.663918\n",
      "Epoch 4967, Loss 25.663147\n",
      "Epoch 4968, Loss 25.662373\n",
      "Epoch 4969, Loss 25.661602\n",
      "Epoch 4970, Loss 25.660826\n",
      "Epoch 4971, Loss 25.660053\n",
      "Epoch 4972, Loss 25.659285\n",
      "Epoch 4973, Loss 25.658514\n",
      "Epoch 4974, Loss 25.657734\n",
      "Epoch 4975, Loss 25.656969\n",
      "Epoch 4976, Loss 25.656191\n",
      "Epoch 4977, Loss 25.655426\n",
      "Epoch 4978, Loss 25.654650\n",
      "Epoch 4979, Loss 25.653881\n",
      "Epoch 4980, Loss 25.653105\n",
      "Epoch 4981, Loss 25.652330\n",
      "Epoch 4982, Loss 25.651562\n",
      "Epoch 4983, Loss 25.650789\n",
      "Epoch 4984, Loss 25.650023\n",
      "Epoch 4985, Loss 25.649242\n",
      "Epoch 4986, Loss 25.648470\n",
      "Epoch 4987, Loss 25.647699\n",
      "Epoch 4988, Loss 25.646931\n",
      "Epoch 4989, Loss 25.646160\n",
      "Epoch 4990, Loss 25.645386\n",
      "Epoch 4991, Loss 25.644615\n",
      "Epoch 4992, Loss 25.643841\n",
      "Epoch 4993, Loss 25.643068\n",
      "Epoch 4994, Loss 25.642302\n",
      "Epoch 4995, Loss 25.641529\n",
      "Epoch 4996, Loss 25.640755\n",
      "Epoch 4997, Loss 25.639982\n",
      "Epoch 4998, Loss 25.639210\n",
      "Epoch 4999, Loss 25.638439\n",
      "Epoch 5000, Loss 25.637672\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "tensor([ 2.5233, -1.2029])"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "params = training_loop(n_epochs = 5000, learning_rate = 1e-4, params = torch.tensor([1.0, 0.0]), t_u = t_un, t_c= t_c)\n",
    "params"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "922c3c40",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x1b3cb5f1a00>]"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 3840x2880 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "from matplotlib import pyplot as plt\n",
    "\n",
    "t_p = model(t_un, *params)\n",
    "\n",
    "fig = plt.figure(dpi=600)\n",
    "plt.xlabel(\"Temperature (°Fahrenheit)\")\n",
    "plt.ylabel(\"Temperature (°Celsius)\")\n",
    "plt.plot(t_u.numpy(), t_p.detach().numpy())\n",
    "plt.plot(t_u.numpy(), t_c.numpy(), 'o')"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "DL_Pytorch",
   "language": "python",
   "name": "dl_pytorch"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.15"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}