train_BA-2motif.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import sys\n",
    "sys.path.append('./codes/forgraph')\n",
    "from config import args\n",
    "import tensorflow as tf\n",
    "import time\n",
    "from models import GCN\n",
    "from metrics import *\n",
    "import pickle as pkl\n",
    "from matplotlib import pyplot as plt\n",
    "import networkx as nx\n",
    "import numpy as np\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open('./dataset/BA-2motif.pkl','rb') as fin:\n",
    "    (adjs, feas, labels) = pkl.load(fin)\n",
    "    \n",
    "args.early_stop =500\n",
    "\n",
    "def vis(adj):\n",
    "    G = nx.from_numpy_matrix(adj)\n",
    "    pos = nx.kamada_kawai_layout(G)\n",
    "    nx.draw_networkx_nodes(G, pos)\n",
    "    nx.draw_networkx_edges(G, pos)\n",
    "\n",
    "    plt.axis('off')\n",
    "    plt.show()\n",
    "    plt.clf()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 432x288 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "vis(adjs[0])\n",
    "vis(adjs[500])\n",
    "\n",
    "order = np.arange(adjs.shape[0])\n",
    "shuffle_adjs = adjs[order]\n",
    "shuffle_feas = feas[order]\n",
    "shuffle_labels = labels[order]\n",
    "\n",
    "train_split = int(adjs.shape[0] * 0.8)\n",
    "val_split = int(adjs.shape[0] * 0.9)\n",
    "\n",
    "train_adjs = shuffle_adjs[:train_split]\n",
    "train_feas = shuffle_feas[:train_split]\n",
    "train_labels = shuffle_labels[:train_split]\n",
    "\n",
    "val_adjs = shuffle_adjs[train_split:val_split]\n",
    "val_feas = shuffle_feas[train_split:val_split]\n",
    "val_labels = shuffle_labels[train_split:val_split]\n",
    "\n",
    "test_adjs = shuffle_adjs[val_split:]\n",
    "test_feas = shuffle_feas[val_split:]\n",
    "test_labels = shuffle_labels[val_split:]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 0001 train_loss= 0.27996 train_acc= 0.37500 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0002 train_loss= 1.05782 train_acc= 0.06500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0003 train_loss= 1.30047 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0004 train_loss= 1.18415 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0005 train_loss= 1.05200 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0006 train_loss= 1.01376 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0007 train_loss= 1.02491 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0008 train_loss= 1.03581 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0009 train_loss= 1.03472 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0010 train_loss= 1.03092 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0011 train_loss= 1.02917 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0012 train_loss= 1.02821 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0013 train_loss= 1.02689 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0014 train_loss= 1.02595 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0015 train_loss= 1.02423 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0016 train_loss= 1.02115 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0017 train_loss= 1.01778 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0018 train_loss= 1.01385 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0019 train_loss= 1.01010 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0020 train_loss= 1.00628 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0021 train_loss= 1.00237 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0022 train_loss= 0.99902 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0023 train_loss= 0.99552 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0024 train_loss= 0.99195 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0025 train_loss= 0.98861 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0026 train_loss= 0.98546 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0027 train_loss= 0.98247 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0028 train_loss= 0.97951 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0029 train_loss= 0.97695 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0030 train_loss= 0.97487 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0031 train_loss= 0.97301 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0032 train_loss= 0.97115 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0033 train_loss= 0.96921 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0034 train_loss= 0.96728 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0035 train_loss= 0.96547 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0036 train_loss= 0.96386 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0037 train_loss= 0.96227 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0038 train_loss= 0.96104 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0039 train_loss= 0.95945 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0040 train_loss= 0.95814 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0041 train_loss= 0.95692 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0042 train_loss= 0.95558 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0043 train_loss= 0.95460 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0044 train_loss= 0.94414 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0045 train_loss= 0.95781 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0046 train_loss= 0.95697 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0047 train_loss= 0.95400 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0048 train_loss= 0.95168 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0049 train_loss= 0.95016 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0050 train_loss= 0.94921 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0051 train_loss= 0.94837 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0052 train_loss= 0.94767 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0053 train_loss= 0.94691 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0054 train_loss= 0.94571 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0055 train_loss= 0.94533 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0056 train_loss= 0.94459 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0057 train_loss= 0.94358 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0058 train_loss= 0.94250 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0059 train_loss= 0.94178 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0060 train_loss= 0.94108 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0061 train_loss= 0.94040 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0062 train_loss= 0.93977 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0063 train_loss= 0.93918 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0064 train_loss= 0.93862 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0065 train_loss= 0.93807 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0066 train_loss= 0.93734 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0067 train_loss= 0.93667 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0068 train_loss= 0.93623 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0069 train_loss= 0.93577 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0070 train_loss= 0.93525 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0071 train_loss= 0.93474 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0072 train_loss= 0.93427 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0073 train_loss= 0.93382 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0074 train_loss= 0.93335 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0075 train_loss= 0.93289 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0076 train_loss= 0.93243 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 0077 train_loss= 0.93200 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0078 train_loss= 0.93160 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0079 train_loss= 0.93116 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0080 train_loss= 0.93071 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0081 train_loss= 0.93029 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0082 train_loss= 0.92990 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0083 train_loss= 0.92954 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0084 train_loss= 0.92917 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0085 train_loss= 0.92878 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0086 train_loss= 0.92843 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0087 train_loss= 0.92807 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0088 train_loss= 0.92761 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0089 train_loss= 0.92729 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0090 train_loss= 0.92688 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0091 train_loss= 0.92659 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0092 train_loss= 0.92637 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0093 train_loss= 0.92604 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0094 train_loss= 0.92576 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0095 train_loss= 0.92571 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0096 train_loss= 0.92559 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0097 train_loss= 0.92544 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0098 train_loss= 0.92534 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0099 train_loss= 0.92539 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0100 train_loss= 0.92517 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0101 train_loss= 0.92503 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0102 train_loss= 0.92495 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0103 train_loss= 0.92490 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0104 train_loss= 0.92487 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0105 train_loss= 0.92484 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0106 train_loss= 0.92467 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0107 train_loss= 0.92461 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0108 train_loss= 0.92448 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0109 train_loss= 0.92446 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0110 train_loss= 0.92437 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0111 train_loss= 0.92432 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0112 train_loss= 0.92430 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0113 train_loss= 0.92417 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0114 train_loss= 0.92411 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0115 train_loss= 0.92396 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0116 train_loss= 0.92394 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0117 train_loss= 0.92398 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0118 train_loss= 0.92383 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0119 train_loss= 0.92378 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0120 train_loss= 0.92380 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0121 train_loss= 0.92369 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0122 train_loss= 0.92350 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0123 train_loss= 0.92352 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0124 train_loss= 0.92361 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0125 train_loss= 0.92351 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0126 train_loss= 0.92324 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0127 train_loss= 0.92322 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0128 train_loss= 0.92309 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0129 train_loss= 0.92315 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0130 train_loss= 0.92324 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0131 train_loss= 0.92312 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0132 train_loss= 0.92304 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0133 train_loss= 0.92301 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0134 train_loss= 0.92276 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0135 train_loss= 0.92274 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0136 train_loss= 0.92277 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0137 train_loss= 0.92278 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0138 train_loss= 0.92269 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0139 train_loss= 0.92237 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0140 train_loss= 0.92181 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0141 train_loss= 0.92171 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0142 train_loss= 0.92184 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0143 train_loss= 0.92416 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0144 train_loss= 0.92752 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0145 train_loss= 0.92719 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0146 train_loss= 0.92525 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0147 train_loss= 0.92310 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0148 train_loss= 0.92155 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0149 train_loss= 0.92095 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0150 train_loss= 0.92051 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0151 train_loss= 0.92069 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0152 train_loss= 0.92047 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 0153 train_loss= 0.92009 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0154 train_loss= 0.92034 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0155 train_loss= 0.92014 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0156 train_loss= 0.92006 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0157 train_loss= 0.92008 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0158 train_loss= 0.91998 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0159 train_loss= 0.92011 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0160 train_loss= 0.91975 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0161 train_loss= 0.91987 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0162 train_loss= 0.91985 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0163 train_loss= 0.91972 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0164 train_loss= 0.91981 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0165 train_loss= 0.91969 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0166 train_loss= 0.91950 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0167 train_loss= 0.91960 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0168 train_loss= 0.91947 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0169 train_loss= 0.91949 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0170 train_loss= 0.91935 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0171 train_loss= 0.91936 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0172 train_loss= 0.91916 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0173 train_loss= 0.91925 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0174 train_loss= 0.91908 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0175 train_loss= 0.91916 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0176 train_loss= 0.91897 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0177 train_loss= 0.91900 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0178 train_loss= 0.91880 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0179 train_loss= 0.91889 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0180 train_loss= 0.91900 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0181 train_loss= 0.91875 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0182 train_loss= 0.91953 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0183 train_loss= 0.91953 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0184 train_loss= 0.91994 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0185 train_loss= 0.91912 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0186 train_loss= 0.91778 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0187 train_loss= 0.91708 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0188 train_loss= 0.91734 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0189 train_loss= 0.91699 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0190 train_loss= 0.91683 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0191 train_loss= 0.91710 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0192 train_loss= 0.91677 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0193 train_loss= 0.91665 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0194 train_loss= 0.91688 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0195 train_loss= 0.91660 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0196 train_loss= 0.91672 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0197 train_loss= 0.91672 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0198 train_loss= 0.91662 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0199 train_loss= 0.91652 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0200 train_loss= 0.91665 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0201 train_loss= 0.91594 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0202 train_loss= 0.91620 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0203 train_loss= 0.91633 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0204 train_loss= 0.91613 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0205 train_loss= 0.91621 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0206 train_loss= 0.91600 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0207 train_loss= 0.91605 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0208 train_loss= 0.91579 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0209 train_loss= 0.91576 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0210 train_loss= 0.91548 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0211 train_loss= 0.91537 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0212 train_loss= 0.91528 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0213 train_loss= 0.91522 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0214 train_loss= 0.91516 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0215 train_loss= 0.91510 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0216 train_loss= 0.91501 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0217 train_loss= 0.91499 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0218 train_loss= 0.91408 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0219 train_loss= 0.91487 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0220 train_loss= 0.91559 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0221 train_loss= 0.91553 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0222 train_loss= 0.91567 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0223 train_loss= 0.91495 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0224 train_loss= 0.91481 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0225 train_loss= 0.91484 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0226 train_loss= 0.91509 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0227 train_loss= 0.91548 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0228 train_loss= 0.91515 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0229 train_loss= 0.91554 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 0230 train_loss= 0.91537 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0231 train_loss= 0.91452 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0232 train_loss= 0.91356 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0233 train_loss= 0.91277 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0234 train_loss= 0.91262 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0235 train_loss= 0.91224 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0236 train_loss= 0.91165 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0237 train_loss= 0.91150 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0238 train_loss= 0.91061 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0239 train_loss= 0.91069 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0240 train_loss= 0.91047 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0241 train_loss= 0.91030 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0242 train_loss= 0.91010 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0243 train_loss= 0.90968 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0244 train_loss= 0.90983 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0245 train_loss= 0.90915 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0246 train_loss= 0.90987 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0247 train_loss= 0.90975 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0248 train_loss= 0.90891 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0249 train_loss= 0.90890 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0250 train_loss= 0.90852 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0251 train_loss= 0.90837 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0252 train_loss= 0.90873 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0253 train_loss= 0.90783 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0254 train_loss= 0.90789 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0255 train_loss= 0.90776 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0256 train_loss= 0.90659 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0257 train_loss= 0.90864 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0258 train_loss= 0.90710 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0259 train_loss= 0.90759 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0260 train_loss= 0.90624 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0261 train_loss= 0.90673 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0262 train_loss= 0.90609 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0263 train_loss= 0.90604 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0264 train_loss= 0.90537 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0265 train_loss= 0.90363 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0266 train_loss= 0.90431 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0267 train_loss= 0.90248 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0268 train_loss= 0.90301 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0269 train_loss= 0.89983 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0270 train_loss= 0.90442 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0271 train_loss= 0.90106 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0272 train_loss= 0.90125 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0273 train_loss= 0.90028 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0274 train_loss= 0.90136 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0275 train_loss= 0.89646 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0276 train_loss= 0.89901 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0277 train_loss= 0.89749 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0278 train_loss= 0.89760 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0279 train_loss= 0.89371 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0280 train_loss= 0.89459 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0281 train_loss= 0.89173 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0282 train_loss= 0.89655 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0283 train_loss= 0.88947 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0284 train_loss= 0.89325 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0285 train_loss= 0.88902 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0286 train_loss= 0.89222 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0287 train_loss= 0.88617 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0288 train_loss= 0.88793 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0289 train_loss= 0.88302 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0290 train_loss= 0.88751 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0291 train_loss= 0.86305 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0292 train_loss= 0.89909 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0293 train_loss= 0.88414 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0294 train_loss= 0.88511 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0295 train_loss= 0.87842 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0296 train_loss= 0.88093 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0297 train_loss= 0.86991 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0298 train_loss= 0.88210 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0299 train_loss= 0.85435 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0300 train_loss= 0.87661 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0301 train_loss= 0.85164 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0302 train_loss= 0.87073 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0303 train_loss= 0.83845 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0304 train_loss= 0.86584 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0305 train_loss= 0.81979 train_acc= 0.62500 val_acc= 0.03000 test_acc= 0.01000 test_best_acc= 1.00000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 0306 train_loss= 0.85374 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0307 train_loss= 0.75799 train_acc= 0.62625 val_acc= 0.57000 test_acc= 0.45000 test_best_acc= 1.00000\n",
      "Epoch: 0308 train_loss= 0.91055 train_acc= 0.60750 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0309 train_loss= 0.80612 train_acc= 0.62500 val_acc= 0.15000 test_acc= 0.09000 test_best_acc= 1.00000\n",
      "Epoch: 0310 train_loss= 0.83987 train_acc= 0.62500 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0311 train_loss= 0.77961 train_acc= 0.62500 val_acc= 0.38000 test_acc= 0.30000 test_best_acc= 1.00000\n",
      "Epoch: 0312 train_loss= 0.82640 train_acc= 0.62375 val_acc= 0.07000 test_acc= 0.06000 test_best_acc= 1.00000\n",
      "Epoch: 0313 train_loss= 0.75006 train_acc= 0.63750 val_acc= 0.57000 test_acc= 0.53000 test_best_acc= 1.00000\n",
      "Epoch: 0314 train_loss= 0.81962 train_acc= 0.62375 val_acc= 0.22000 test_acc= 0.13000 test_best_acc= 1.00000\n",
      "Epoch: 0315 train_loss= 0.67452 train_acc= 0.69750 val_acc= 0.95000 test_acc= 0.93000 test_best_acc= 1.00000\n",
      "Epoch: 0316 train_loss= 0.79333 train_acc= 0.60875 val_acc= 0.46000 test_acc= 0.38000 test_best_acc= 1.00000\n",
      "Epoch: 0317 train_loss= 0.55032 train_acc= 0.84375 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0318 train_loss= 0.82699 train_acc= 0.55125 val_acc= 0.56000 test_acc= 0.46000 test_best_acc= 1.00000\n",
      "Epoch: 0319 train_loss= 0.29587 train_acc= 0.96750 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0320 train_loss= 1.40886 train_acc= 0.34625 val_acc= 0.00000 test_acc= 0.00000 test_best_acc= 1.00000\n",
      "Epoch: 0321 train_loss= 0.81895 train_acc= 0.62625 val_acc= 0.54000 test_acc= 0.45000 test_best_acc= 1.00000\n",
      "Epoch: 0322 train_loss= 0.53406 train_acc= 0.92000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0323 train_loss= 0.66391 train_acc= 0.70500 val_acc= 0.96000 test_acc= 0.93000 test_best_acc= 1.00000\n",
      "Epoch: 0324 train_loss= 0.46888 train_acc= 0.87375 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0325 train_loss= 0.50656 train_acc= 0.75125 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0326 train_loss= 0.30463 train_acc= 0.72375 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0327 train_loss= 0.36846 train_acc= 0.86500 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0328 train_loss= 0.31011 train_acc= 0.94500 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0329 train_loss= 0.22866 train_acc= 0.94500 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0330 train_loss= 0.18268 train_acc= 0.91875 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0331 train_loss= 0.15878 train_acc= 0.92375 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0332 train_loss= 0.12788 train_acc= 0.96125 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0333 train_loss= 0.10663 train_acc= 0.96625 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0334 train_loss= 0.09283 train_acc= 0.97250 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0335 train_loss= 0.08578 train_acc= 0.97250 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0336 train_loss= 0.08028 train_acc= 0.98375 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0337 train_loss= 0.06896 train_acc= 0.99000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0338 train_loss= 0.05583 train_acc= 0.99625 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0339 train_loss= 0.05439 train_acc= 0.99750 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0340 train_loss= 0.05143 train_acc= 0.99750 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0341 train_loss= 0.04666 train_acc= 0.99875 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0342 train_loss= 0.04167 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0343 train_loss= 0.03939 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0344 train_loss= 0.03682 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0345 train_loss= 0.03334 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0346 train_loss= 0.03195 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0347 train_loss= 0.03063 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0348 train_loss= 0.02842 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0349 train_loss= 0.02751 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0350 train_loss= 0.02575 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0351 train_loss= 0.02449 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0352 train_loss= 0.02355 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0353 train_loss= 0.02248 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0354 train_loss= 0.02139 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0355 train_loss= 0.02070 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0356 train_loss= 0.01975 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0357 train_loss= 0.01906 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0358 train_loss= 0.01822 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0359 train_loss= 0.01764 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0360 train_loss= 0.01720 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0361 train_loss= 0.01609 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0362 train_loss= 0.01589 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0363 train_loss= 0.01531 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0364 train_loss= 0.01467 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0365 train_loss= 0.01429 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0366 train_loss= 0.01360 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0367 train_loss= 0.01347 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0368 train_loss= 0.01280 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0369 train_loss= 0.01230 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0370 train_loss= 0.01212 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0371 train_loss= 0.01156 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0372 train_loss= 0.01125 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0373 train_loss= 0.01097 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0374 train_loss= 0.01048 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0375 train_loss= 0.01034 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0376 train_loss= 0.00989 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0377 train_loss= 0.00956 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0378 train_loss= 0.00941 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0379 train_loss= 0.00912 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0380 train_loss= 0.00879 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0381 train_loss= 0.00865 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0382 train_loss= 0.00824 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 0383 train_loss= 0.00820 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0384 train_loss= 0.00788 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0385 train_loss= 0.00769 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0386 train_loss= 0.00754 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0387 train_loss= 0.00721 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0388 train_loss= 0.00714 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0389 train_loss= 0.00693 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0390 train_loss= 0.00672 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0391 train_loss= 0.00662 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0392 train_loss= 0.00643 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0393 train_loss= 0.00622 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0394 train_loss= 0.00617 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0395 train_loss= 0.00594 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0396 train_loss= 0.00589 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0397 train_loss= 0.00578 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0398 train_loss= 0.00556 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0399 train_loss= 0.00534 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0400 train_loss= 0.00502 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0401 train_loss= 0.00481 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0402 train_loss= 0.00459 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0403 train_loss= 0.00446 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0404 train_loss= 0.00429 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0405 train_loss= 0.00411 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0406 train_loss= 0.00390 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0407 train_loss= 0.00369 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0408 train_loss= 0.00354 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0409 train_loss= 0.00338 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0410 train_loss= 0.00325 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0411 train_loss= 0.00314 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0412 train_loss= 0.00302 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0413 train_loss= 0.00291 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0414 train_loss= 0.00279 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0415 train_loss= 0.00270 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0416 train_loss= 0.00263 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0417 train_loss= 0.00253 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0418 train_loss= 0.00246 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0419 train_loss= 0.00238 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0420 train_loss= 0.00231 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0421 train_loss= 0.00224 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0422 train_loss= 0.00217 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0423 train_loss= 0.00212 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0424 train_loss= 0.00204 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0425 train_loss= 0.00200 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0426 train_loss= 0.00193 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0427 train_loss= 0.00188 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0428 train_loss= 0.00183 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0429 train_loss= 0.00177 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0430 train_loss= 0.00174 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0431 train_loss= 0.00168 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0432 train_loss= 0.00165 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0433 train_loss= 0.00160 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0434 train_loss= 0.00157 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0435 train_loss= 0.00151 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0436 train_loss= 0.00150 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0437 train_loss= 0.00145 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0438 train_loss= 0.00142 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0439 train_loss= 0.00139 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0440 train_loss= 0.00135 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0441 train_loss= 0.00132 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0442 train_loss= 0.00129 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0443 train_loss= 0.00127 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0444 train_loss= 0.00123 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0445 train_loss= 0.00121 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0446 train_loss= 0.00118 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0447 train_loss= 0.00116 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0448 train_loss= 0.00113 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0449 train_loss= 0.00111 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0450 train_loss= 0.00108 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0451 train_loss= 0.00107 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0452 train_loss= 0.00104 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0453 train_loss= 0.00103 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0454 train_loss= 0.00101 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0455 train_loss= 0.00099 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0456 train_loss= 0.00097 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0457 train_loss= 0.00095 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0458 train_loss= 0.00094 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 0459 train_loss= 0.00092 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0460 train_loss= 0.00090 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0461 train_loss= 0.00088 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0462 train_loss= 0.00087 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0463 train_loss= 0.00086 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0464 train_loss= 0.00084 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0465 train_loss= 0.00083 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0466 train_loss= 0.00082 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0467 train_loss= 0.00080 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0468 train_loss= 0.00079 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0469 train_loss= 0.00077 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0470 train_loss= 0.00077 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0471 train_loss= 0.00075 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0472 train_loss= 0.00074 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0473 train_loss= 0.00073 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0474 train_loss= 0.00072 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0475 train_loss= 0.00071 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0476 train_loss= 0.00069 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0477 train_loss= 0.00069 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0478 train_loss= 0.00067 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0479 train_loss= 0.00066 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0480 train_loss= 0.00065 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0481 train_loss= 0.00064 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0482 train_loss= 0.00063 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0483 train_loss= 0.00063 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0484 train_loss= 0.00061 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0485 train_loss= 0.00061 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0486 train_loss= 0.00059 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0487 train_loss= 0.00059 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0488 train_loss= 0.00058 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0489 train_loss= 0.00057 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0490 train_loss= 0.00056 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0491 train_loss= 0.00056 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0492 train_loss= 0.00055 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0493 train_loss= 0.00054 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0494 train_loss= 0.00053 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0495 train_loss= 0.00053 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0496 train_loss= 0.00052 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0497 train_loss= 0.00051 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0498 train_loss= 0.00050 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0499 train_loss= 0.00050 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0500 train_loss= 0.00049 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0501 train_loss= 0.00048 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0502 train_loss= 0.00048 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0503 train_loss= 0.00047 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0504 train_loss= 0.00047 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0505 train_loss= 0.00046 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0506 train_loss= 0.00046 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0507 train_loss= 0.00045 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0508 train_loss= 0.00044 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0509 train_loss= 0.00044 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0510 train_loss= 0.00043 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0511 train_loss= 0.00043 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0512 train_loss= 0.00042 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0513 train_loss= 0.00042 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0514 train_loss= 0.00041 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0515 train_loss= 0.00041 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0516 train_loss= 0.00040 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0517 train_loss= 0.00040 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0518 train_loss= 0.00039 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0519 train_loss= 0.00039 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0520 train_loss= 0.00038 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0521 train_loss= 0.00038 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0522 train_loss= 0.00037 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0523 train_loss= 0.00037 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0524 train_loss= 0.00037 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0525 train_loss= 0.00036 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0526 train_loss= 0.00036 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0527 train_loss= 0.00035 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0528 train_loss= 0.00035 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0529 train_loss= 0.00035 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0530 train_loss= 0.00034 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0531 train_loss= 0.00033 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0532 train_loss= 0.00033 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0533 train_loss= 0.00033 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0534 train_loss= 0.00032 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0535 train_loss= 0.00032 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 0536 train_loss= 0.00032 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0537 train_loss= 0.00031 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0538 train_loss= 0.00031 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0539 train_loss= 0.00031 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0540 train_loss= 0.00030 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0541 train_loss= 0.00030 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0542 train_loss= 0.00030 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0543 train_loss= 0.00029 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0544 train_loss= 0.00029 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0545 train_loss= 0.00029 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0546 train_loss= 0.00028 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0547 train_loss= 0.00028 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0548 train_loss= 0.00028 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0549 train_loss= 0.00028 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0550 train_loss= 0.00027 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0551 train_loss= 0.00027 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0552 train_loss= 0.00027 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0553 train_loss= 0.00026 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0554 train_loss= 0.00026 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0555 train_loss= 0.00026 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0556 train_loss= 0.00026 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0557 train_loss= 0.00025 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0558 train_loss= 0.00025 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0559 train_loss= 0.00025 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0560 train_loss= 0.00024 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0561 train_loss= 0.00024 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0562 train_loss= 0.00024 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0563 train_loss= 0.00024 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0564 train_loss= 0.00023 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0565 train_loss= 0.00023 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0566 train_loss= 0.00023 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0567 train_loss= 0.00023 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0568 train_loss= 0.00022 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0569 train_loss= 0.00022 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0570 train_loss= 0.00022 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0571 train_loss= 0.00022 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0572 train_loss= 0.00022 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0573 train_loss= 0.00021 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0574 train_loss= 0.00021 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0575 train_loss= 0.00021 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0576 train_loss= 0.00021 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0577 train_loss= 0.00020 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0578 train_loss= 0.00020 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0579 train_loss= 0.00020 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0580 train_loss= 0.00020 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0581 train_loss= 0.00020 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0582 train_loss= 0.00019 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0583 train_loss= 0.00019 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0584 train_loss= 0.00019 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0585 train_loss= 0.00019 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0586 train_loss= 0.00019 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0587 train_loss= 0.00019 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0588 train_loss= 0.00018 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0589 train_loss= 0.00018 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0590 train_loss= 0.00018 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0591 train_loss= 0.00018 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0592 train_loss= 0.00018 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0593 train_loss= 0.00017 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0594 train_loss= 0.00017 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0595 train_loss= 0.00017 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0596 train_loss= 0.00017 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0597 train_loss= 0.00017 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0598 train_loss= 0.00017 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0599 train_loss= 0.00016 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0600 train_loss= 0.00016 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0601 train_loss= 0.00016 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0602 train_loss= 0.00016 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0603 train_loss= 0.00016 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0604 train_loss= 0.00016 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0605 train_loss= 0.00016 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0606 train_loss= 0.00015 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0607 train_loss= 0.00015 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0608 train_loss= 0.00015 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0609 train_loss= 0.00015 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0610 train_loss= 0.00015 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0611 train_loss= 0.00015 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0612 train_loss= 0.00015 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0613 train_loss= 0.00014 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 0614 train_loss= 0.00014 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0615 train_loss= 0.00014 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0616 train_loss= 0.00014 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0617 train_loss= 0.00014 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0618 train_loss= 0.00014 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0619 train_loss= 0.00014 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0620 train_loss= 0.00014 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0621 train_loss= 0.00013 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0622 train_loss= 0.00013 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0623 train_loss= 0.00013 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0624 train_loss= 0.00013 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0625 train_loss= 0.00013 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0626 train_loss= 0.00013 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0627 train_loss= 0.00013 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0628 train_loss= 0.00012 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0629 train_loss= 0.00012 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0630 train_loss= 0.00012 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0631 train_loss= 0.00012 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0632 train_loss= 0.00012 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0633 train_loss= 0.00012 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0634 train_loss= 0.00012 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0635 train_loss= 0.00012 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0636 train_loss= 0.00012 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0637 train_loss= 0.00012 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0638 train_loss= 0.00011 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0639 train_loss= 0.00011 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0640 train_loss= 0.00011 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0641 train_loss= 0.00011 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0642 train_loss= 0.00011 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0643 train_loss= 0.00011 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0644 train_loss= 0.00011 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0645 train_loss= 0.00011 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0646 train_loss= 0.00011 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0647 train_loss= 0.00011 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0648 train_loss= 0.00011 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0649 train_loss= 0.00011 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0650 train_loss= 0.00010 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0651 train_loss= 0.00010 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0652 train_loss= 0.00010 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0653 train_loss= 0.00010 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0654 train_loss= 0.00010 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0655 train_loss= 0.00010 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0656 train_loss= 0.00010 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0657 train_loss= 0.00010 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0658 train_loss= 0.00010 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0659 train_loss= 0.00010 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0660 train_loss= 0.00010 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0661 train_loss= 0.00010 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0662 train_loss= 0.00009 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0663 train_loss= 0.00009 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0664 train_loss= 0.00009 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0665 train_loss= 0.00009 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0666 train_loss= 0.00009 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0667 train_loss= 0.00009 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0668 train_loss= 0.00009 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0669 train_loss= 0.00009 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0670 train_loss= 0.00009 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0671 train_loss= 0.00009 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0672 train_loss= 0.00009 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0673 train_loss= 0.00009 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0674 train_loss= 0.00009 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0675 train_loss= 0.00009 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0676 train_loss= 0.00009 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0677 train_loss= 0.00008 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0678 train_loss= 0.00008 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0679 train_loss= 0.00008 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0680 train_loss= 0.00008 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0681 train_loss= 0.00008 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0682 train_loss= 0.00008 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0683 train_loss= 0.00008 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0684 train_loss= 0.00008 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0685 train_loss= 0.00008 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0686 train_loss= 0.00008 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0687 train_loss= 0.00008 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0688 train_loss= 0.00008 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0689 train_loss= 0.00008 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0690 train_loss= 0.00008 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0691 train_loss= 0.00008 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 0692 train_loss= 0.00007 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0693 train_loss= 0.00007 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0694 train_loss= 0.00007 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0695 train_loss= 0.00007 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0696 train_loss= 0.00007 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0697 train_loss= 0.00007 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0698 train_loss= 0.00007 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0699 train_loss= 0.00007 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0700 train_loss= 0.00007 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0701 train_loss= 0.00007 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0702 train_loss= 0.00007 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0703 train_loss= 0.00007 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0704 train_loss= 0.00007 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0705 train_loss= 0.00007 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0706 train_loss= 0.00007 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0707 train_loss= 0.00007 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0708 train_loss= 0.00007 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0709 train_loss= 0.00007 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0710 train_loss= 0.00007 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0711 train_loss= 0.00007 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0712 train_loss= 0.00007 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0713 train_loss= 0.00006 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0714 train_loss= 0.00006 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0715 train_loss= 0.00006 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0716 train_loss= 0.00006 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0717 train_loss= 0.00006 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0718 train_loss= 0.00006 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0719 train_loss= 0.00006 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0720 train_loss= 0.00006 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0721 train_loss= 0.00006 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0722 train_loss= 0.00006 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0723 train_loss= 0.00006 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0724 train_loss= 0.00006 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0725 train_loss= 0.00006 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0726 train_loss= 0.00006 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0727 train_loss= 0.00006 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0728 train_loss= 0.00006 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0729 train_loss= 0.00006 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0730 train_loss= 0.00006 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0731 train_loss= 0.00006 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0732 train_loss= 0.00006 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0733 train_loss= 0.00006 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0734 train_loss= 0.00006 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0735 train_loss= 0.00005 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0736 train_loss= 0.00005 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0737 train_loss= 0.00005 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0738 train_loss= 0.00005 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0739 train_loss= 0.00005 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0740 train_loss= 0.00005 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0741 train_loss= 0.00005 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0742 train_loss= 0.00005 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0743 train_loss= 0.00005 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0744 train_loss= 0.00005 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0745 train_loss= 0.00005 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0746 train_loss= 0.00005 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0747 train_loss= 0.00005 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0748 train_loss= 0.00005 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0749 train_loss= 0.00005 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0750 train_loss= 0.00005 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0751 train_loss= 0.00005 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0752 train_loss= 0.00005 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0753 train_loss= 0.00005 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0754 train_loss= 0.00005 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0755 train_loss= 0.00005 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0756 train_loss= 0.00005 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0757 train_loss= 0.00005 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0758 train_loss= 0.00005 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0759 train_loss= 0.00005 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0760 train_loss= 0.00005 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0761 train_loss= 0.00005 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0762 train_loss= 0.00005 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0763 train_loss= 0.00004 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0764 train_loss= 0.00004 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0765 train_loss= 0.00004 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0766 train_loss= 0.00004 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0767 train_loss= 0.00004 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0768 train_loss= 0.00004 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0769 train_loss= 0.00004 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 0770 train_loss= 0.00004 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0771 train_loss= 0.00004 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0772 train_loss= 0.00004 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0773 train_loss= 0.00004 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0774 train_loss= 0.00004 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0775 train_loss= 0.00004 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0776 train_loss= 0.00004 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0777 train_loss= 0.00004 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0778 train_loss= 0.00004 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0779 train_loss= 0.00004 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0780 train_loss= 0.00004 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0781 train_loss= 0.00004 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0782 train_loss= 0.00004 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0783 train_loss= 0.00004 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0784 train_loss= 0.00004 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0785 train_loss= 0.00004 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0786 train_loss= 0.00004 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0787 train_loss= 0.00004 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0788 train_loss= 0.00004 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0789 train_loss= 0.00004 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0790 train_loss= 0.00004 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0791 train_loss= 0.00004 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0792 train_loss= 0.00004 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0793 train_loss= 0.00004 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0794 train_loss= 0.00004 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0795 train_loss= 0.00004 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0796 train_loss= 0.00004 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0797 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0798 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0799 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0800 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0801 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0802 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0803 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0804 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0805 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0806 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0807 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0808 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0809 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0810 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0811 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0812 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0813 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0814 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0815 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0816 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0817 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0818 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0819 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0820 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0821 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0822 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0823 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0824 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0825 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0826 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0827 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0828 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0829 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0830 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0831 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0832 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0833 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0834 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0835 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0836 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0837 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0838 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0839 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0840 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0841 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Epoch: 0842 train_loss= 0.00003 train_acc= 1.00000 val_acc= 1.00000 test_acc= 1.00000 test_best_acc= 1.00000\n",
      "Early stopping...\n"
     ]
    }
   ],
   "source": [
    "\n",
    "model = GCN(input_dim=train_feas.shape[-1], output_dim=train_labels.shape[1])\n",
    "\n",
    "optimizer = tf.keras.optimizers.Adam(learning_rate=args.lr)\n",
    "\n",
    "train_adjs_tensor = tf.convert_to_tensor(train_adjs,dtype=tf.float32)\n",
    "train_features_tensor = tf.convert_to_tensor(train_feas,dtype=tf.float32)\n",
    "train_labels_tensor = tf.convert_to_tensor(train_labels,dtype=tf.float32)\n",
    "\n",
    "val_adjs_tensor = tf.convert_to_tensor(val_adjs,dtype=tf.float32)\n",
    "val_features_tensor = tf.convert_to_tensor(val_feas,dtype=tf.float32)\n",
    "val_labels_tensor = tf.convert_to_tensor(val_labels,dtype=tf.float32)\n",
    "\n",
    "test_adjs_tensor = tf.convert_to_tensor(test_adjs,dtype=tf.float32)\n",
    "test_features_tensor = tf.convert_to_tensor(test_feas,dtype=tf.float32)\n",
    "test_labels_tensor = tf.convert_to_tensor(test_labels,dtype=tf.float32)\n",
    "\n",
    "best_test_acc = 0\n",
    "best_val_acc = 0\n",
    "best_val_loss = 10000\n",
    "clip_value_min = -2.0\n",
    "clip_value_max = 2.0\n",
    "best_train_acc = 0\n",
    "\n",
    "curr_step = 0\n",
    "for epoch in range(args.epochs):\n",
    "    if args.batch:\n",
    "        begin = 0\n",
    "        batch_size= 64\n",
    "        end = batch_size\n",
    "        trainsize = train_adjs.shape[0]\n",
    "        outputs = []\n",
    "        while begin<trainsize:\n",
    "            batch_train_adjs_tensor = tf.convert_to_tensor(train_adjs[begin:end], dtype=tf.float32)\n",
    "            batch_train_features_tensor = tf.convert_to_tensor(train_feas[begin:end], dtype=tf.float32)\n",
    "            batch_train_labels_tensor = tf.convert_to_tensor(train_labels[begin:end], dtype=tf.float32)\n",
    "\n",
    "            with tf.GradientTape() as tape:\n",
    "                output = model.call((batch_train_features_tensor,batch_train_adjs_tensor),training=True)\n",
    "                cross_loss = softmax_cross_entropy(output, batch_train_labels_tensor)\n",
    "                lossL2 = tf.add_n([tf.nn.l2_loss(v) for v in model.trainable_variables])\n",
    "                loss = cross_loss + args.weight_decay*lossL2\n",
    "                grads = tape.gradient(loss, model.trainable_variables)\n",
    "                cliped_grads = [tf.clip_by_value(t, clip_value_min, clip_value_max) for t in grads]\n",
    "            optimizer.apply_gradients(zip(cliped_grads, model.trainable_variables))\n",
    "            begin = end\n",
    "            end = min(end+batch_size,trainsize)\n",
    "            outputs.append(output)\n",
    "        output = tf.concat(outputs,axis=0)\n",
    "    else:\n",
    "        with tf.GradientTape() as tape:\n",
    "            output = model.call((train_feas_tensor,train_adjs_tensor),training=True)\n",
    "            cross_loss = softmax_cross_entropy(output, train_labels_tensor)\n",
    "            lossL2 = tf.add_n([tf.nn.l2_loss(v) for v in model.trainable_variables])\n",
    "            loss = cross_loss + args.weight_decay*lossL2\n",
    "            grads = tape.gradient(loss, model.trainable_variables)\n",
    "            cliped_grads = [tf.clip_by_value(t, clip_value_min, clip_value_max) for t in grads]\n",
    "        optimizer.apply_gradients(zip(cliped_grads, model.trainable_variables))\n",
    "\n",
    "    train_acc = accuracy(output, train_labels_tensor)\n",
    "    val_output = model.call((val_features_tensor, val_adjs_tensor), training=False)\n",
    "    val_acc  = accuracy(val_output, val_labels_tensor)\n",
    "    val_loss = softmax_cross_entropy(val_output, val_labels_tensor)\n",
    "    \n",
    "    \n",
    "\n",
    "    test_output = model.call((test_features_tensor, test_adjs_tensor), training=False)\n",
    "    test_acc  = accuracy(test_output, test_labels_tensor)\n",
    "    test_loss = softmax_cross_entropy(test_output, test_labels_tensor)\n",
    "\n",
    "    if val_acc > best_val_acc or (train_acc > best_train_acc and best_val_acc==val_acc):\n",
    "        curr_step = 0\n",
    "        best_test_acc = test_acc\n",
    "        best_val_acc = val_acc\n",
    "        best_val_loss= val_loss\n",
    "        if args.save_model:\n",
    "            model.save_weights(args.save_path+args.dataset)\n",
    "    else:\n",
    "        curr_step +=1\n",
    "    if curr_step > args.early_stop:\n",
    "        print(\"Early stopping...\")\n",
    "        break\n",
    "    if train_acc > best_train_acc:\n",
    "        best_train_acc = train_acc\n",
    "            \n",
    "    \n",
    "\n",
    "    print(\"Epoch:\", '%04d' % (epoch + 1), \"train_loss=\", \"{:.5f}\".format(cross_loss), \"train_acc=\",\n",
    "          \"{:.5f}\".format(train_acc), \"val_acc=\", \"{:.5f}\".format(val_acc),\n",
    "          \"test_acc=\", \"{:.5f}\".format(test_acc),\n",
    "          \"test_best_acc=\", \"{:.5f}\".format(best_test_acc)\n",
    "         )\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "tf2",
   "language": "python",
   "name": "tf2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}