Backup.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1046,
   "outputs": [],
   "source": [
    "import torch\n",
    "from torch import nn\n",
    "from torch.utils.data import DataLoader, TensorDataset\n",
    "from torchvision import datasets\n",
    "from torchvision.transforms import ToTensor, Lambda, Compose\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 1047,
   "outputs": [
    {
     "data": {
      "text/plain": "     PassengerId HomePlanet CryoSleep     Cabin    Destination   Age    VIP  \\\n0        0001_01     Europa     False     B/0/P    TRAPPIST-1e  39.0  False   \n1        0002_01      Earth     False     F/0/S    TRAPPIST-1e  24.0  False   \n2        0003_01     Europa     False     A/0/S    TRAPPIST-1e  58.0   True   \n3        0003_02     Europa     False     A/0/S    TRAPPIST-1e  33.0  False   \n4        0004_01      Earth     False     F/1/S    TRAPPIST-1e  16.0  False   \n...          ...        ...       ...       ...            ...   ...    ...   \n8688     9276_01     Europa     False    A/98/P    55 Cancri e  41.0   True   \n8689     9278_01      Earth      True  G/1499/S  PSO J318.5-22  18.0  False   \n8690     9279_01      Earth     False  G/1500/S    TRAPPIST-1e  26.0  False   \n8691     9280_01     Europa     False   E/608/S    55 Cancri e  32.0  False   \n8692     9280_02     Europa     False   E/608/S    TRAPPIST-1e  44.0  False   \n\n      RoomService  FoodCourt  ShoppingMall     Spa  VRDeck               Name  \\\n0             0.0        0.0           0.0     0.0     0.0    Maham Ofracculy   \n1           109.0        9.0          25.0   549.0    44.0       Juanna Vines   \n2            43.0     3576.0           0.0  6715.0    49.0      Altark Susent   \n3             0.0     1283.0         371.0  3329.0   193.0       Solam Susent   \n4           303.0       70.0         151.0   565.0     2.0  Willy Santantines   \n...           ...        ...           ...     ...     ...                ...   \n8688          0.0     6819.0           0.0  1643.0    74.0  Gravior Noxnuther   \n8689          0.0        0.0           0.0     0.0     0.0    Kurta Mondalley   \n8690          0.0        0.0        1872.0     1.0     0.0       Fayey Connon   \n8691          0.0     1049.0           0.0   353.0  3235.0   Celeon Hontichre   \n8692        126.0     4688.0           0.0     0.0    12.0   Propsh Hontichre   \n\n      Transported  \n0           False  \n1            True  \n2           False  \n3           False  \n4            True  \n...           ...  \n8688        False  \n8689        False  \n8690         True  \n8691        False  \n8692         True  \n\n[8693 rows x 14 columns]",
      "text/html": "<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>PassengerId</th>\n      <th>HomePlanet</th>\n      <th>CryoSleep</th>\n      <th>Cabin</th>\n      <th>Destination</th>\n      <th>Age</th>\n      <th>VIP</th>\n      <th>RoomService</th>\n      <th>FoodCourt</th>\n      <th>ShoppingMall</th>\n      <th>Spa</th>\n      <th>VRDeck</th>\n      <th>Name</th>\n      <th>Transported</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>0001_01</td>\n      <td>Europa</td>\n      <td>False</td>\n      <td>B/0/P</td>\n      <td>TRAPPIST-1e</td>\n      <td>39.0</td>\n      <td>False</td>\n      <td>0.0</td>\n      <td>0.0</td>\n      <td>0.0</td>\n      <td>0.0</td>\n      <td>0.0</td>\n      <td>Maham Ofracculy</td>\n      <td>False</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>0002_01</td>\n      <td>Earth</td>\n      <td>False</td>\n      <td>F/0/S</td>\n      <td>TRAPPIST-1e</td>\n      <td>24.0</td>\n      <td>False</td>\n      <td>109.0</td>\n      <td>9.0</td>\n      <td>25.0</td>\n      <td>549.0</td>\n      <td>44.0</td>\n      <td>Juanna Vines</td>\n      <td>True</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>0003_01</td>\n      <td>Europa</td>\n      <td>False</td>\n      <td>A/0/S</td>\n      <td>TRAPPIST-1e</td>\n      <td>58.0</td>\n      <td>True</td>\n      <td>43.0</td>\n      <td>3576.0</td>\n      <td>0.0</td>\n      <td>6715.0</td>\n      <td>49.0</td>\n      <td>Altark Susent</td>\n      <td>False</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>0003_02</td>\n      <td>Europa</td>\n      <td>False</td>\n      <td>A/0/S</td>\n      <td>TRAPPIST-1e</td>\n      <td>33.0</td>\n      <td>False</td>\n      <td>0.0</td>\n      <td>1283.0</td>\n      <td>371.0</td>\n      <td>3329.0</td>\n      <td>193.0</td>\n      <td>Solam Susent</td>\n      <td>False</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>0004_01</td>\n      <td>Earth</td>\n      <td>False</td>\n      <td>F/1/S</td>\n      <td>TRAPPIST-1e</td>\n      <td>16.0</td>\n      <td>False</td>\n      <td>303.0</td>\n      <td>70.0</td>\n      <td>151.0</td>\n      <td>565.0</td>\n      <td>2.0</td>\n      <td>Willy Santantines</td>\n      <td>True</td>\n    </tr>\n    <tr>\n      <th>...</th>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n    </tr>\n    <tr>\n      <th>8688</th>\n      <td>9276_01</td>\n      <td>Europa</td>\n      <td>False</td>\n      <td>A/98/P</td>\n      <td>55 Cancri e</td>\n      <td>41.0</td>\n      <td>True</td>\n      <td>0.0</td>\n      <td>6819.0</td>\n      <td>0.0</td>\n      <td>1643.0</td>\n      <td>74.0</td>\n      <td>Gravior Noxnuther</td>\n      <td>False</td>\n    </tr>\n    <tr>\n      <th>8689</th>\n      <td>9278_01</td>\n      <td>Earth</td>\n      <td>True</td>\n      <td>G/1499/S</td>\n      <td>PSO J318.5-22</td>\n      <td>18.0</td>\n      <td>False</td>\n      <td>0.0</td>\n      <td>0.0</td>\n      <td>0.0</td>\n      <td>0.0</td>\n      <td>0.0</td>\n      <td>Kurta Mondalley</td>\n      <td>False</td>\n    </tr>\n    <tr>\n      <th>8690</th>\n      <td>9279_01</td>\n      <td>Earth</td>\n      <td>False</td>\n      <td>G/1500/S</td>\n      <td>TRAPPIST-1e</td>\n      <td>26.0</td>\n      <td>False</td>\n      <td>0.0</td>\n      <td>0.0</td>\n      <td>1872.0</td>\n      <td>1.0</td>\n      <td>0.0</td>\n      <td>Fayey Connon</td>\n      <td>True</td>\n    </tr>\n    <tr>\n      <th>8691</th>\n      <td>9280_01</td>\n      <td>Europa</td>\n      <td>False</td>\n      <td>E/608/S</td>\n      <td>55 Cancri e</td>\n      <td>32.0</td>\n      <td>False</td>\n      <td>0.0</td>\n      <td>1049.0</td>\n      <td>0.0</td>\n      <td>353.0</td>\n      <td>3235.0</td>\n      <td>Celeon Hontichre</td>\n      <td>False</td>\n    </tr>\n    <tr>\n      <th>8692</th>\n      <td>9280_02</td>\n      <td>Europa</td>\n      <td>False</td>\n      <td>E/608/S</td>\n      <td>TRAPPIST-1e</td>\n      <td>44.0</td>\n      <td>False</td>\n      <td>126.0</td>\n      <td>4688.0</td>\n      <td>0.0</td>\n      <td>0.0</td>\n      <td>12.0</td>\n      <td>Propsh Hontichre</td>\n      <td>True</td>\n    </tr>\n  </tbody>\n</table>\n<p>8693 rows × 14 columns</p>\n</div>"
     },
     "execution_count": 1047,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "training_data=pd.read_csv(\"data/test.csv\")\n",
    "training_data"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 1048,
   "outputs": [
    {
     "data": {
      "text/plain": "     PassengerId HomePlanet CryoSleep     Cabin    Destination   Age    VIP  \\\n0        0001_01     Europa     False     B/0/P    TRAPPIST-1e  39.0  False   \n1        0002_01      Earth     False     F/0/S    TRAPPIST-1e  24.0  False   \n2        0003_01     Europa     False     A/0/S    TRAPPIST-1e  58.0   True   \n3        0003_02     Europa     False     A/0/S    TRAPPIST-1e  33.0  False   \n4        0004_01      Earth     False     F/1/S    TRAPPIST-1e  16.0  False   \n...          ...        ...       ...       ...            ...   ...    ...   \n8688     9276_01     Europa     False    A/98/P    55 Cancri e  41.0   True   \n8689     9278_01      Earth      True  G/1499/S  PSO J318.5-22  18.0  False   \n8690     9279_01      Earth     False  G/1500/S    TRAPPIST-1e  26.0  False   \n8691     9280_01     Europa     False   E/608/S    55 Cancri e  32.0  False   \n8692     9280_02     Europa     False   E/608/S    TRAPPIST-1e  44.0  False   \n\n      RoomService  FoodCourt  ShoppingMall     Spa  VRDeck               Name  \\\n0             0.0        0.0           0.0     0.0     0.0    Maham Ofracculy   \n1           109.0        9.0          25.0   549.0    44.0       Juanna Vines   \n2            43.0     3576.0           0.0  6715.0    49.0      Altark Susent   \n3             0.0     1283.0         371.0  3329.0   193.0       Solam Susent   \n4           303.0       70.0         151.0   565.0     2.0  Willy Santantines   \n...           ...        ...           ...     ...     ...                ...   \n8688          0.0     6819.0           0.0  1643.0    74.0  Gravior Noxnuther   \n8689          0.0        0.0           0.0     0.0     0.0    Kurta Mondalley   \n8690          0.0        0.0        1872.0     1.0     0.0       Fayey Connon   \n8691          0.0     1049.0           0.0   353.0  3235.0   Celeon Hontichre   \n8692        126.0     4688.0           0.0     0.0    12.0   Propsh Hontichre   \n\n      Transported  \n0           False  \n1            True  \n2           False  \n3           False  \n4            True  \n...           ...  \n8688        False  \n8689        False  \n8690         True  \n8691        False  \n8692         True  \n\n[8693 rows x 14 columns]",
      "text/html": "<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>PassengerId</th>\n      <th>HomePlanet</th>\n      <th>CryoSleep</th>\n      <th>Cabin</th>\n      <th>Destination</th>\n      <th>Age</th>\n      <th>VIP</th>\n      <th>RoomService</th>\n      <th>FoodCourt</th>\n      <th>ShoppingMall</th>\n      <th>Spa</th>\n      <th>VRDeck</th>\n      <th>Name</th>\n      <th>Transported</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>0001_01</td>\n      <td>Europa</td>\n      <td>False</td>\n      <td>B/0/P</td>\n      <td>TRAPPIST-1e</td>\n      <td>39.0</td>\n      <td>False</td>\n      <td>0.0</td>\n      <td>0.0</td>\n      <td>0.0</td>\n      <td>0.0</td>\n      <td>0.0</td>\n      <td>Maham Ofracculy</td>\n      <td>False</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>0002_01</td>\n      <td>Earth</td>\n      <td>False</td>\n      <td>F/0/S</td>\n      <td>TRAPPIST-1e</td>\n      <td>24.0</td>\n      <td>False</td>\n      <td>109.0</td>\n      <td>9.0</td>\n      <td>25.0</td>\n      <td>549.0</td>\n      <td>44.0</td>\n      <td>Juanna Vines</td>\n      <td>True</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>0003_01</td>\n      <td>Europa</td>\n      <td>False</td>\n      <td>A/0/S</td>\n      <td>TRAPPIST-1e</td>\n      <td>58.0</td>\n      <td>True</td>\n      <td>43.0</td>\n      <td>3576.0</td>\n      <td>0.0</td>\n      <td>6715.0</td>\n      <td>49.0</td>\n      <td>Altark Susent</td>\n      <td>False</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>0003_02</td>\n      <td>Europa</td>\n      <td>False</td>\n      <td>A/0/S</td>\n      <td>TRAPPIST-1e</td>\n      <td>33.0</td>\n      <td>False</td>\n      <td>0.0</td>\n      <td>1283.0</td>\n      <td>371.0</td>\n      <td>3329.0</td>\n      <td>193.0</td>\n      <td>Solam Susent</td>\n      <td>False</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>0004_01</td>\n      <td>Earth</td>\n      <td>False</td>\n      <td>F/1/S</td>\n      <td>TRAPPIST-1e</td>\n      <td>16.0</td>\n      <td>False</td>\n      <td>303.0</td>\n      <td>70.0</td>\n      <td>151.0</td>\n      <td>565.0</td>\n      <td>2.0</td>\n      <td>Willy Santantines</td>\n      <td>True</td>\n    </tr>\n    <tr>\n      <th>...</th>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n    </tr>\n    <tr>\n      <th>8688</th>\n      <td>9276_01</td>\n      <td>Europa</td>\n      <td>False</td>\n      <td>A/98/P</td>\n      <td>55 Cancri e</td>\n      <td>41.0</td>\n      <td>True</td>\n      <td>0.0</td>\n      <td>6819.0</td>\n      <td>0.0</td>\n      <td>1643.0</td>\n      <td>74.0</td>\n      <td>Gravior Noxnuther</td>\n      <td>False</td>\n    </tr>\n    <tr>\n      <th>8689</th>\n      <td>9278_01</td>\n      <td>Earth</td>\n      <td>True</td>\n      <td>G/1499/S</td>\n      <td>PSO J318.5-22</td>\n      <td>18.0</td>\n      <td>False</td>\n      <td>0.0</td>\n      <td>0.0</td>\n      <td>0.0</td>\n      <td>0.0</td>\n      <td>0.0</td>\n      <td>Kurta Mondalley</td>\n      <td>False</td>\n    </tr>\n    <tr>\n      <th>8690</th>\n      <td>9279_01</td>\n      <td>Earth</td>\n      <td>False</td>\n      <td>G/1500/S</td>\n      <td>TRAPPIST-1e</td>\n      <td>26.0</td>\n      <td>False</td>\n      <td>0.0</td>\n      <td>0.0</td>\n      <td>1872.0</td>\n      <td>1.0</td>\n      <td>0.0</td>\n      <td>Fayey Connon</td>\n      <td>True</td>\n    </tr>\n    <tr>\n      <th>8691</th>\n      <td>9280_01</td>\n      <td>Europa</td>\n      <td>False</td>\n      <td>E/608/S</td>\n      <td>55 Cancri e</td>\n      <td>32.0</td>\n      <td>False</td>\n      <td>0.0</td>\n      <td>1049.0</td>\n      <td>0.0</td>\n      <td>353.0</td>\n      <td>3235.0</td>\n      <td>Celeon Hontichre</td>\n      <td>False</td>\n    </tr>\n    <tr>\n      <th>8692</th>\n      <td>9280_02</td>\n      <td>Europa</td>\n      <td>False</td>\n      <td>E/608/S</td>\n      <td>TRAPPIST-1e</td>\n      <td>44.0</td>\n      <td>False</td>\n      <td>126.0</td>\n      <td>4688.0</td>\n      <td>0.0</td>\n      <td>0.0</td>\n      <td>12.0</td>\n      <td>Propsh Hontichre</td>\n      <td>True</td>\n    </tr>\n  </tbody>\n</table>\n<p>8693 rows × 14 columns</p>\n</div>"
     },
     "execution_count": 1048,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "training_data.fillna(0,inplace=True)\n",
    "training_data"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 1049,
   "outputs": [
    {
     "data": {
      "text/plain": "array([['0001_01', 'Europa', False, ..., 0.0, 'Maham Ofracculy', False],\n       ['0002_01', 'Earth', False, ..., 44.0, 'Juanna Vines', True],\n       ['0003_01', 'Europa', False, ..., 49.0, 'Altark Susent', False],\n       ...,\n       ['9279_01', 'Earth', False, ..., 0.0, 'Fayey Connon', True],\n       ['9280_01', 'Europa', False, ..., 3235.0, 'Celeon Hontichre',\n        False],\n       ['9280_02', 'Europa', False, ..., 12.0, 'Propsh Hontichre', True]],\n      dtype=object)"
     },
     "execution_count": 1049,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "training_data=training_data.to_numpy()\n",
    "training_data"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 1050,
   "outputs": [
    {
     "data": {
      "text/plain": "['B', '0', 'P']"
     },
     "execution_count": 1050,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "training_data[:,3][0].split('/')"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 1051,
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\shaow\\AppData\\Local\\Temp\\ipykernel_17632\\657575255.py:7: FutureWarning: arrays to stack must be passed as a \"sequence\" type such as list or tuple. Support for non-sequence iterables such as generators is deprecated as of NumPy 1.16 and will raise an error in the future.\n",
      "  new_seperated_column=np.hstack(map(split_3rd, training_data[:,3])).reshape(training_data.shape[0],3)\n"
     ]
    }
   ],
   "source": [
    "def split_3rd(x):\n",
    "    if x==0:\n",
    "        return [0,0,0]\n",
    "    else:\n",
    "        a,b,c=x.split('/')\n",
    "        return a,int(b),c\n",
    "new_seperated_column=np.hstack(map(split_3rd, training_data[:,3])).reshape(training_data.shape[0],3)"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 1052,
   "outputs": [],
   "source": [
    "training_data=np.delete(training_data,3,axis=1)"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 1053,
   "outputs": [
    {
     "data": {
      "text/plain": "array([['0001_01', 'Europa', False, ..., 'B', '0', 'P'],\n       ['0002_01', 'Earth', False, ..., 'F', '0', 'S'],\n       ['0003_01', 'Europa', False, ..., 'A', '0', 'S'],\n       ...,\n       ['9279_01', 'Earth', False, ..., 'G', '1500', 'S'],\n       ['9280_01', 'Europa', False, ..., 'E', '608', 'S'],\n       ['9280_02', 'Europa', False, ..., 'E', '608', 'S']], dtype=object)"
     },
     "execution_count": 1053,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "training_data=np.hstack((training_data,new_seperated_column))\n",
    "training_data"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 1054,
   "outputs": [],
   "source": [
    "def class_mapping(column_number):\n",
    "    classes={}\n",
    "    index=0\n",
    "    for id in training_data[:,column_number]:\n",
    "        if id not in classes:\n",
    "            classes[id]=index\n",
    "            index+=1\n",
    "    return classes"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 1055,
   "outputs": [],
   "source": [
    "map_planet=class_mapping(1)"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 1056,
   "outputs": [
    {
     "data": {
      "text/plain": "array([['0001_01', 0, False, ..., 'B', '0', 'P'],\n       ['0002_01', 1, False, ..., 'F', '0', 'S'],\n       ['0003_01', 0, False, ..., 'A', '0', 'S'],\n       ...,\n       ['9279_01', 1, False, ..., 'G', '1500', 'S'],\n       ['9280_01', 0, False, ..., 'E', '608', 'S'],\n       ['9280_02', 0, False, ..., 'E', '608', 'S']], dtype=object)"
     },
     "execution_count": 1056,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "training_data[:,1]=np.vectorize(map_planet.get)(training_data[:,1])\n",
    "training_data"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 1057,
   "outputs": [],
   "source": [
    "def replace_with_mapped_value(column_number,map_name):\n",
    "    training_data[:,column_number]=np.vectorize(map_name.get)(training_data[:,column_number])"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 1058,
   "outputs": [
    {
     "data": {
      "text/plain": "{False: 0, True: 1}"
     },
     "execution_count": 1058,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "TF_map=class_mapping(2)\n",
    "TF_map"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 1059,
   "outputs": [
    {
     "data": {
      "text/plain": "{'TRAPPIST-1e': 0, 'PSO J318.5-22': 1, '55 Cancri e': 2, 0: 3}"
     },
     "execution_count": 1059,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "destination_map=class_mapping(3)\n",
    "destination_map"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 1060,
   "outputs": [
    {
     "data": {
      "text/plain": "{'B': 0, 'F': 1, 'A': 2, 'G': 3, '0': 4, 'E': 5, 'D': 6, 'C': 7, 'T': 8}"
     },
     "execution_count": 1060,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cabin0=class_mapping(13)\n",
    "cabin0"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 1061,
   "outputs": [
    {
     "data": {
      "text/plain": "{'P': 0, 'S': 1, '0': 2}"
     },
     "execution_count": 1061,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cabin2=class_mapping(15)\n",
    "cabin2"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 1062,
   "outputs": [],
   "source": [
    "replace_with_mapped_value(2,TF_map)\n",
    "replace_with_mapped_value(3,destination_map)\n",
    "replace_with_mapped_value(5,TF_map)\n",
    "replace_with_mapped_value(12,cabin0)\n",
    "replace_with_mapped_value(14,cabin2)"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 1063,
   "outputs": [
    {
     "data": {
      "text/plain": "array([['0001_01', 0, 0, ..., 0, '0', 0],\n       ['0002_01', 1, 0, ..., 1, '0', 1],\n       ['0003_01', 0, 0, ..., 2, '0', 1],\n       ...,\n       ['9279_01', 1, 0, ..., 3, '1500', 1],\n       ['9280_01', 0, 0, ..., 5, '608', 1],\n       ['9280_02', 0, 0, ..., 5, '608', 1]], dtype=object)"
     },
     "execution_count": 1063,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "training_data"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 1064,
   "outputs": [],
   "source": [
    "np.random.shuffle(training_data)\n",
    "training_data,test_data=training_data[:int(len(training_data)*0.8)],training_data[int(len(training_data)*0.8):]\n",
    "# x=np.delete(training_data,[0,1,2],1).astype(float)\n",
    "# y=training_data[:,[1]].astype(float)\n",
    "# tensor_x=torch.Tensor(x)\n",
    "# tensor_y=torch.Tensor(y)\n",
    "# training_data=TensorDataset(tensor_x,tensor_y)\n",
    "def to_tensor(np_array,removed_columns,target_column):\n",
    "    removed_columns.append(target_column)\n",
    "    x=np.delete(np_array,removed_columns,axis=1).astype(float)\n",
    "    y=np_array[:,target_column].astype(int)\n",
    "    y_with_zeros=[]\n",
    "    for class_element in y:\n",
    "        y_with_zeros.append(np.zeros(len(TF_map)))\n",
    "        y_with_zeros[-1][TF_map[class_element]]=1\n",
    "    y=y_with_zeros\n",
    "    tensor_x=torch.Tensor(x)\n",
    "    tensor_y=torch.Tensor(y)\n",
    "    tensor=TensorDataset(tensor_x,tensor_y)\n",
    "    return tensor"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 1065,
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 13])\n",
      "torch.Size([128, 2]) torch.float32\n"
     ]
    }
   ],
   "source": [
    "batch_size = 128\n",
    "\n",
    "# Create data loaders.\n",
    "train_dataloader=DataLoader(to_tensor(training_data,[0,11],12),batch_size=batch_size,shuffle=True)\n",
    "test_dataloader = DataLoader(to_tensor(test_data,[0,11],12), batch_size=batch_size)\n",
    "\n",
    "for X, y in train_dataloader:\n",
    "    print(X.shape)\n",
    "    print(y.shape, y.dtype)\n",
    "    break\n",
    "\n",
    "# # Display sample data\n",
    "# figure = plt.figure(figsize=(10, 8))\n",
    "# cols, rows = 5, 5\n",
    "# for i in range(1, cols * rows + 1):\n",
    "#     idx = torch.randint(len(test_data), size=(1,)).item()\n",
    "#     img, label = test_data[idx]\n",
    "#     figure.add_subplot(rows, cols, i)\n",
    "#     plt.title(label)\n",
    "#     plt.axis(\"off\")\n",
    "#     plt.imshow(img.squeeze(), cmap=\"gray\")\n",
    "# plt.show()"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 1066,
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Using cuda device\n",
      "NeuralNetwork(\n",
      "  (linear_relu_stack): Sequential(\n",
      "    (0): Linear(in_features=13, out_features=512, bias=True)\n",
      "    (1): ReLU()\n",
      "    (2): Linear(in_features=512, out_features=512, bias=True)\n",
      "    (3): ReLU()\n",
      "    (4): Linear(in_features=512, out_features=512, bias=True)\n",
      "    (5): ReLU()\n",
      "    (6): Linear(in_features=512, out_features=512, bias=True)\n",
      "    (7): ReLU()\n",
      "    (8): Linear(in_features=512, out_features=512, bias=True)\n",
      "    (9): ReLU()\n",
      "    (10): Linear(in_features=512, out_features=512, bias=True)\n",
      "    (11): Sigmoid()\n",
      "    (12): Linear(in_features=512, out_features=2, bias=True)\n",
      "    (13): Sigmoid()\n",
      "  )\n",
      ")\n"
     ]
    }
   ],
   "source": [
    "# Get cpu or gpu device for training.\n",
    "device = \"cuda\" if torch.cuda.is_available() else \"cpu\"\n",
    "print(\"Using {} device\".format(device))\n",
    "layer_size=512\n",
    "input_size=13\n",
    "\n",
    "# Define model\n",
    "class NeuralNetwork(nn.Module):\n",
    "    def __init__(self):\n",
    "        super(NeuralNetwork, self).__init__()\n",
    "        self.linear_relu_stack = nn.Sequential(\n",
    "            nn.Linear(input_size, layer_size),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(layer_size, layer_size),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(layer_size, layer_size),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(layer_size, layer_size),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(layer_size, layer_size),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(layer_size, layer_size),\n",
    "            nn.Sigmoid(),\n",
    "            nn.Linear(layer_size, len(TF_map)),\n",
    "            nn.Sigmoid(),\n",
    "        )\n",
    "\n",
    "    def forward(self, x):\n",
    "        logits = self.linear_relu_stack(x)\n",
    "        return logits\n",
    "\n",
    "model = NeuralNetwork().to(device)\n",
    "print(model)"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 1067,
   "outputs": [],
   "source": [
    "# loss_fn = nn.L1Loss()\n",
    "loss_fn = nn.CrossEntropyLoss()\n",
    "learning_rate = 1e-4\n",
    "optimizer = torch.optim.SGD(model.parameters(), lr=learning_rate)"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 1068,
   "outputs": [],
   "source": [
    "def train(dataloader, model, loss_fn, optimizer):\n",
    "    size = len(dataloader.dataset)\n",
    "    for batch, (X, y) in enumerate(dataloader):\n",
    "        # y=y.type(torch.float)\n",
    "        X, y = X.to(device), y.to(device)\n",
    "\n",
    "        # Compute prediction error\n",
    "        pred = model(X)\n",
    "        loss = loss_fn(pred, y)\n",
    "\n",
    "        # Backpropagation\n",
    "        optimizer.zero_grad()\n",
    "        loss.backward()\n",
    "        optimizer.step()\n",
    "\n",
    "        if batch % 8 == 0:\n",
    "            loss, current = loss.item(), batch * len(X)\n",
    "            print(f\"loss: {loss:>7f}  [{current:>5d}/{size:>5d}]\")\n"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 1069,
   "outputs": [],
   "source": [
    "def test(dataloader, model):\n",
    "    size = len(dataloader.dataset)\n",
    "    model.eval()\n",
    "    test_loss, correct = 0, 0\n",
    "    with torch.no_grad():\n",
    "        for X, y in dataloader:\n",
    "            X, y = X.to(device), y.to(device)\n",
    "            pred = model(X)\n",
    "            # print(pred.argmax(1).shape,y.shape)\n",
    "            test_loss += loss_fn(pred, y).item()\n",
    "            correct += (pred.argmax(1) == y.argmax(1)).type(torch.float).sum().item()\n",
    "    test_loss /= size\n",
    "    correct /= size\n",
    "    print(f\"Test Error: \\n Accuracy: {(100*correct):>0.1f}%, Avg loss: {test_loss:>8f} \\n\")"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1\n",
      "-------------------------------\n",
      "loss: 0.693936  [    0/ 6954]\n",
      "loss: 0.693217  [ 1024/ 6954]\n",
      "loss: 0.693723  [ 2048/ 6954]\n",
      "loss: 0.692265  [ 3072/ 6954]\n",
      "loss: 0.690589  [ 4096/ 6954]\n",
      "loss: 0.693615  [ 5120/ 6954]\n",
      "loss: 0.693757  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 57.4%, Avg loss: 0.005572 \n",
      "\n",
      "Epoch 2\n",
      "-------------------------------\n",
      "loss: 0.691493  [    0/ 6954]\n",
      "loss: 0.691856  [ 1024/ 6954]\n",
      "loss: 0.691150  [ 2048/ 6954]\n",
      "loss: 0.690648  [ 3072/ 6954]\n",
      "loss: 0.689436  [ 4096/ 6954]\n",
      "loss: 0.690220  [ 5120/ 6954]\n",
      "loss: 0.693103  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 59.0%, Avg loss: 0.005559 \n",
      "\n",
      "Epoch 3\n",
      "-------------------------------\n",
      "loss: 0.689610  [    0/ 6954]\n",
      "loss: 0.690604  [ 1024/ 6954]\n",
      "loss: 0.687947  [ 2048/ 6954]\n",
      "loss: 0.691219  [ 3072/ 6954]\n",
      "loss: 0.690070  [ 4096/ 6954]\n",
      "loss: 0.689188  [ 5120/ 6954]\n",
      "loss: 0.688398  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 60.8%, Avg loss: 0.005546 \n",
      "\n",
      "Epoch 4\n",
      "-------------------------------\n",
      "loss: 0.690001  [    0/ 6954]\n",
      "loss: 0.689899  [ 1024/ 6954]\n",
      "loss: 0.691415  [ 2048/ 6954]\n",
      "loss: 0.688717  [ 3072/ 6954]\n",
      "loss: 0.687453  [ 4096/ 6954]\n",
      "loss: 0.687946  [ 5120/ 6954]\n",
      "loss: 0.687649  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 63.5%, Avg loss: 0.005534 \n",
      "\n",
      "Epoch 5\n",
      "-------------------------------\n",
      "loss: 0.693556  [    0/ 6954]\n",
      "loss: 0.683537  [ 1024/ 6954]\n",
      "loss: 0.688343  [ 2048/ 6954]\n",
      "loss: 0.687498  [ 3072/ 6954]\n",
      "loss: 0.683156  [ 4096/ 6954]\n",
      "loss: 0.684852  [ 5120/ 6954]\n",
      "loss: 0.687732  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 67.2%, Avg loss: 0.005522 \n",
      "\n",
      "Epoch 6\n",
      "-------------------------------\n",
      "loss: 0.687016  [    0/ 6954]\n",
      "loss: 0.684534  [ 1024/ 6954]\n",
      "loss: 0.685723  [ 2048/ 6954]\n",
      "loss: 0.685748  [ 3072/ 6954]\n",
      "loss: 0.686637  [ 4096/ 6954]\n",
      "loss: 0.684824  [ 5120/ 6954]\n",
      "loss: 0.685557  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 70.4%, Avg loss: 0.005510 \n",
      "\n",
      "Epoch 7\n",
      "-------------------------------\n",
      "loss: 0.685512  [    0/ 6954]\n",
      "loss: 0.683612  [ 1024/ 6954]\n",
      "loss: 0.683104  [ 2048/ 6954]\n",
      "loss: 0.684880  [ 3072/ 6954]\n",
      "loss: 0.683076  [ 4096/ 6954]\n",
      "loss: 0.685322  [ 5120/ 6954]\n",
      "loss: 0.683933  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 73.4%, Avg loss: 0.005499 \n",
      "\n",
      "Epoch 8\n",
      "-------------------------------\n",
      "loss: 0.681172  [    0/ 6954]\n",
      "loss: 0.687288  [ 1024/ 6954]\n",
      "loss: 0.683397  [ 2048/ 6954]\n",
      "loss: 0.684505  [ 3072/ 6954]\n",
      "loss: 0.682345  [ 4096/ 6954]\n",
      "loss: 0.682046  [ 5120/ 6954]\n",
      "loss: 0.681208  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 74.5%, Avg loss: 0.005489 \n",
      "\n",
      "Epoch 9\n",
      "-------------------------------\n",
      "loss: 0.685131  [    0/ 6954]\n",
      "loss: 0.681270  [ 1024/ 6954]\n",
      "loss: 0.686254  [ 2048/ 6954]\n",
      "loss: 0.682080  [ 3072/ 6954]\n",
      "loss: 0.678233  [ 4096/ 6954]\n",
      "loss: 0.682256  [ 5120/ 6954]\n",
      "loss: 0.684214  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 74.9%, Avg loss: 0.005479 \n",
      "\n",
      "Epoch 10\n",
      "-------------------------------\n",
      "loss: 0.679842  [    0/ 6954]\n",
      "loss: 0.682231  [ 1024/ 6954]\n",
      "loss: 0.681211  [ 2048/ 6954]\n",
      "loss: 0.680194  [ 3072/ 6954]\n",
      "loss: 0.680992  [ 4096/ 6954]\n",
      "loss: 0.677121  [ 5120/ 6954]\n",
      "loss: 0.679927  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 75.3%, Avg loss: 0.005469 \n",
      "\n",
      "Epoch 11\n",
      "-------------------------------\n",
      "loss: 0.681479  [    0/ 6954]\n",
      "loss: 0.679224  [ 1024/ 6954]\n",
      "loss: 0.680575  [ 2048/ 6954]\n",
      "loss: 0.675498  [ 3072/ 6954]\n",
      "loss: 0.681410  [ 4096/ 6954]\n",
      "loss: 0.681633  [ 5120/ 6954]\n",
      "loss: 0.676311  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 75.2%, Avg loss: 0.005459 \n",
      "\n",
      "Epoch 12\n",
      "-------------------------------\n",
      "loss: 0.679124  [    0/ 6954]\n",
      "loss: 0.677119  [ 1024/ 6954]\n",
      "loss: 0.683464  [ 2048/ 6954]\n",
      "loss: 0.682254  [ 3072/ 6954]\n",
      "loss: 0.682210  [ 4096/ 6954]\n",
      "loss: 0.679895  [ 5120/ 6954]\n",
      "loss: 0.675703  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 75.2%, Avg loss: 0.005450 \n",
      "\n",
      "Epoch 13\n",
      "-------------------------------\n",
      "loss: 0.672946  [    0/ 6954]\n",
      "loss: 0.672842  [ 1024/ 6954]\n",
      "loss: 0.674914  [ 2048/ 6954]\n",
      "loss: 0.672596  [ 3072/ 6954]\n",
      "loss: 0.674557  [ 4096/ 6954]\n",
      "loss: 0.678757  [ 5120/ 6954]\n",
      "loss: 0.681418  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 75.8%, Avg loss: 0.005441 \n",
      "\n",
      "Epoch 14\n",
      "-------------------------------\n",
      "loss: 0.675888  [    0/ 6954]\n",
      "loss: 0.677775  [ 1024/ 6954]\n",
      "loss: 0.675389  [ 2048/ 6954]\n",
      "loss: 0.677548  [ 3072/ 6954]\n",
      "loss: 0.678594  [ 4096/ 6954]\n",
      "loss: 0.673560  [ 5120/ 6954]\n",
      "loss: 0.676281  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 76.1%, Avg loss: 0.005432 \n",
      "\n",
      "Epoch 15\n",
      "-------------------------------\n",
      "loss: 0.678515  [    0/ 6954]\n",
      "loss: 0.675865  [ 1024/ 6954]\n",
      "loss: 0.678061  [ 2048/ 6954]\n",
      "loss: 0.676427  [ 3072/ 6954]\n",
      "loss: 0.677670  [ 4096/ 6954]\n",
      "loss: 0.676992  [ 5120/ 6954]\n",
      "loss: 0.667772  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 76.1%, Avg loss: 0.005423 \n",
      "\n",
      "Epoch 16\n",
      "-------------------------------\n",
      "loss: 0.676387  [    0/ 6954]\n",
      "loss: 0.676394  [ 1024/ 6954]\n",
      "loss: 0.671219  [ 2048/ 6954]\n",
      "loss: 0.674717  [ 3072/ 6954]\n",
      "loss: 0.672479  [ 4096/ 6954]\n",
      "loss: 0.676129  [ 5120/ 6954]\n",
      "loss: 0.674701  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 76.0%, Avg loss: 0.005415 \n",
      "\n",
      "Epoch 17\n",
      "-------------------------------\n",
      "loss: 0.670341  [    0/ 6954]\n",
      "loss: 0.680188  [ 1024/ 6954]\n",
      "loss: 0.673705  [ 2048/ 6954]\n",
      "loss: 0.669899  [ 3072/ 6954]\n",
      "loss: 0.674234  [ 4096/ 6954]\n",
      "loss: 0.671695  [ 5120/ 6954]\n",
      "loss: 0.664812  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 76.0%, Avg loss: 0.005407 \n",
      "\n",
      "Epoch 18\n",
      "-------------------------------\n",
      "loss: 0.674044  [    0/ 6954]\n",
      "loss: 0.674926  [ 1024/ 6954]\n",
      "loss: 0.668959  [ 2048/ 6954]\n",
      "loss: 0.668824  [ 3072/ 6954]\n",
      "loss: 0.673727  [ 4096/ 6954]\n",
      "loss: 0.670537  [ 5120/ 6954]\n",
      "loss: 0.673805  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 75.9%, Avg loss: 0.005399 \n",
      "\n",
      "Epoch 19\n",
      "-------------------------------\n",
      "loss: 0.673586  [    0/ 6954]\n",
      "loss: 0.665975  [ 1024/ 6954]\n",
      "loss: 0.669753  [ 2048/ 6954]\n",
      "loss: 0.670736  [ 3072/ 6954]\n",
      "loss: 0.669342  [ 4096/ 6954]\n",
      "loss: 0.667868  [ 5120/ 6954]\n",
      "loss: 0.671069  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 75.9%, Avg loss: 0.005391 \n",
      "\n",
      "Epoch 20\n",
      "-------------------------------\n",
      "loss: 0.668882  [    0/ 6954]\n",
      "loss: 0.666772  [ 1024/ 6954]\n",
      "loss: 0.673666  [ 2048/ 6954]\n",
      "loss: 0.672975  [ 3072/ 6954]\n",
      "loss: 0.664919  [ 4096/ 6954]\n",
      "loss: 0.665600  [ 5120/ 6954]\n",
      "loss: 0.668992  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 75.9%, Avg loss: 0.005384 \n",
      "\n",
      "Epoch 21\n",
      "-------------------------------\n",
      "loss: 0.666645  [    0/ 6954]\n",
      "loss: 0.677614  [ 1024/ 6954]\n",
      "loss: 0.671254  [ 2048/ 6954]\n",
      "loss: 0.668852  [ 3072/ 6954]\n",
      "loss: 0.677523  [ 4096/ 6954]\n",
      "loss: 0.669382  [ 5120/ 6954]\n",
      "loss: 0.665588  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 76.0%, Avg loss: 0.005376 \n",
      "\n",
      "Epoch 22\n",
      "-------------------------------\n",
      "loss: 0.666969  [    0/ 6954]\n",
      "loss: 0.673273  [ 1024/ 6954]\n",
      "loss: 0.662354  [ 2048/ 6954]\n",
      "loss: 0.665473  [ 3072/ 6954]\n",
      "loss: 0.670532  [ 4096/ 6954]\n",
      "loss: 0.669507  [ 5120/ 6954]\n",
      "loss: 0.673649  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 76.0%, Avg loss: 0.005369 \n",
      "\n",
      "Epoch 23\n",
      "-------------------------------\n",
      "loss: 0.668311  [    0/ 6954]\n",
      "loss: 0.665214  [ 1024/ 6954]\n",
      "loss: 0.667419  [ 2048/ 6954]\n",
      "loss: 0.668333  [ 3072/ 6954]\n",
      "loss: 0.671964  [ 4096/ 6954]\n",
      "loss: 0.669767  [ 5120/ 6954]\n",
      "loss: 0.658794  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 76.0%, Avg loss: 0.005362 \n",
      "\n",
      "Epoch 24\n",
      "-------------------------------\n",
      "loss: 0.663128  [    0/ 6954]\n",
      "loss: 0.671737  [ 1024/ 6954]\n",
      "loss: 0.669660  [ 2048/ 6954]\n",
      "loss: 0.669004  [ 3072/ 6954]\n",
      "loss: 0.670466  [ 4096/ 6954]\n",
      "loss: 0.666591  [ 5120/ 6954]\n",
      "loss: 0.664326  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 76.1%, Avg loss: 0.005355 \n",
      "\n",
      "Epoch 25\n",
      "-------------------------------\n",
      "loss: 0.668464  [    0/ 6954]\n",
      "loss: 0.670623  [ 1024/ 6954]\n",
      "loss: 0.667464  [ 2048/ 6954]\n",
      "loss: 0.660584  [ 3072/ 6954]\n",
      "loss: 0.664448  [ 4096/ 6954]\n",
      "loss: 0.664339  [ 5120/ 6954]\n",
      "loss: 0.661615  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 76.1%, Avg loss: 0.005349 \n",
      "\n",
      "Epoch 26\n",
      "-------------------------------\n",
      "loss: 0.667653  [    0/ 6954]\n",
      "loss: 0.658968  [ 1024/ 6954]\n",
      "loss: 0.670952  [ 2048/ 6954]\n",
      "loss: 0.668699  [ 3072/ 6954]\n",
      "loss: 0.669914  [ 4096/ 6954]\n",
      "loss: 0.666249  [ 5120/ 6954]\n",
      "loss: 0.655708  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 76.1%, Avg loss: 0.005342 \n",
      "\n",
      "Epoch 27\n",
      "-------------------------------\n",
      "loss: 0.664521  [    0/ 6954]\n",
      "loss: 0.666113  [ 1024/ 6954]\n",
      "loss: 0.660538  [ 2048/ 6954]\n",
      "loss: 0.665891  [ 3072/ 6954]\n",
      "loss: 0.664630  [ 4096/ 6954]\n",
      "loss: 0.672951  [ 5120/ 6954]\n",
      "loss: 0.664827  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 76.2%, Avg loss: 0.005335 \n",
      "\n",
      "Epoch 28\n",
      "-------------------------------\n",
      "loss: 0.666990  [    0/ 6954]\n",
      "loss: 0.659759  [ 1024/ 6954]\n",
      "loss: 0.660311  [ 2048/ 6954]\n",
      "loss: 0.669366  [ 3072/ 6954]\n",
      "loss: 0.657726  [ 4096/ 6954]\n",
      "loss: 0.673419  [ 5120/ 6954]\n",
      "loss: 0.666771  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 76.1%, Avg loss: 0.005329 \n",
      "\n",
      "Epoch 29\n",
      "-------------------------------\n",
      "loss: 0.658336  [    0/ 6954]\n",
      "loss: 0.668289  [ 1024/ 6954]\n",
      "loss: 0.665706  [ 2048/ 6954]\n",
      "loss: 0.662014  [ 3072/ 6954]\n",
      "loss: 0.664598  [ 4096/ 6954]\n",
      "loss: 0.663516  [ 5120/ 6954]\n",
      "loss: 0.662521  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 76.1%, Avg loss: 0.005322 \n",
      "\n",
      "Epoch 30\n",
      "-------------------------------\n",
      "loss: 0.664710  [    0/ 6954]\n",
      "loss: 0.666667  [ 1024/ 6954]\n",
      "loss: 0.668456  [ 2048/ 6954]\n",
      "loss: 0.654294  [ 3072/ 6954]\n",
      "loss: 0.668293  [ 4096/ 6954]\n",
      "loss: 0.672735  [ 5120/ 6954]\n",
      "loss: 0.648107  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 76.1%, Avg loss: 0.005316 \n",
      "\n",
      "Epoch 31\n",
      "-------------------------------\n",
      "loss: 0.670000  [    0/ 6954]\n",
      "loss: 0.662291  [ 1024/ 6954]\n",
      "loss: 0.661703  [ 2048/ 6954]\n",
      "loss: 0.660552  [ 3072/ 6954]\n",
      "loss: 0.656468  [ 4096/ 6954]\n",
      "loss: 0.662830  [ 5120/ 6954]\n",
      "loss: 0.659881  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 76.1%, Avg loss: 0.005310 \n",
      "\n",
      "Epoch 32\n",
      "-------------------------------\n",
      "loss: 0.660115  [    0/ 6954]\n",
      "loss: 0.658983  [ 1024/ 6954]\n",
      "loss: 0.661472  [ 2048/ 6954]\n",
      "loss: 0.660613  [ 3072/ 6954]\n",
      "loss: 0.659646  [ 4096/ 6954]\n",
      "loss: 0.660843  [ 5120/ 6954]\n",
      "loss: 0.670846  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 76.2%, Avg loss: 0.005303 \n",
      "\n",
      "Epoch 33\n",
      "-------------------------------\n",
      "loss: 0.658026  [    0/ 6954]\n",
      "loss: 0.658407  [ 1024/ 6954]\n",
      "loss: 0.657806  [ 2048/ 6954]\n",
      "loss: 0.657374  [ 3072/ 6954]\n",
      "loss: 0.660593  [ 4096/ 6954]\n",
      "loss: 0.656989  [ 5120/ 6954]\n",
      "loss: 0.658406  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 76.3%, Avg loss: 0.005297 \n",
      "\n",
      "Epoch 34\n",
      "-------------------------------\n",
      "loss: 0.658003  [    0/ 6954]\n",
      "loss: 0.656208  [ 1024/ 6954]\n",
      "loss: 0.661113  [ 2048/ 6954]\n",
      "loss: 0.661319  [ 3072/ 6954]\n",
      "loss: 0.661487  [ 4096/ 6954]\n",
      "loss: 0.652287  [ 5120/ 6954]\n",
      "loss: 0.660655  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 76.3%, Avg loss: 0.005291 \n",
      "\n",
      "Epoch 35\n",
      "-------------------------------\n",
      "loss: 0.668951  [    0/ 6954]\n",
      "loss: 0.659705  [ 1024/ 6954]\n",
      "loss: 0.662975  [ 2048/ 6954]\n",
      "loss: 0.656413  [ 3072/ 6954]\n",
      "loss: 0.667296  [ 4096/ 6954]\n",
      "loss: 0.658834  [ 5120/ 6954]\n",
      "loss: 0.656467  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 76.4%, Avg loss: 0.005285 \n",
      "\n",
      "Epoch 36\n",
      "-------------------------------\n",
      "loss: 0.655556  [    0/ 6954]\n",
      "loss: 0.652288  [ 1024/ 6954]\n",
      "loss: 0.658689  [ 2048/ 6954]\n",
      "loss: 0.653542  [ 3072/ 6954]\n",
      "loss: 0.663851  [ 4096/ 6954]\n",
      "loss: 0.654329  [ 5120/ 6954]\n",
      "loss: 0.661597  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 76.4%, Avg loss: 0.005279 \n",
      "\n",
      "Epoch 37\n",
      "-------------------------------\n",
      "loss: 0.650715  [    0/ 6954]\n",
      "loss: 0.657711  [ 1024/ 6954]\n",
      "loss: 0.661914  [ 2048/ 6954]\n",
      "loss: 0.660867  [ 3072/ 6954]\n",
      "loss: 0.652229  [ 4096/ 6954]\n",
      "loss: 0.654526  [ 5120/ 6954]\n",
      "loss: 0.653867  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 76.4%, Avg loss: 0.005273 \n",
      "\n",
      "Epoch 38\n",
      "-------------------------------\n",
      "loss: 0.651693  [    0/ 6954]\n",
      "loss: 0.656449  [ 1024/ 6954]\n",
      "loss: 0.667554  [ 2048/ 6954]\n",
      "loss: 0.660449  [ 3072/ 6954]\n",
      "loss: 0.649890  [ 4096/ 6954]\n",
      "loss: 0.661699  [ 5120/ 6954]\n",
      "loss: 0.648619  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 76.4%, Avg loss: 0.005267 \n",
      "\n",
      "Epoch 39\n",
      "-------------------------------\n",
      "loss: 0.659375  [    0/ 6954]\n",
      "loss: 0.655619  [ 1024/ 6954]\n",
      "loss: 0.652637  [ 2048/ 6954]\n",
      "loss: 0.655663  [ 3072/ 6954]\n",
      "loss: 0.654086  [ 4096/ 6954]\n",
      "loss: 0.654822  [ 5120/ 6954]\n",
      "loss: 0.661079  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 76.4%, Avg loss: 0.005262 \n",
      "\n",
      "Epoch 40\n",
      "-------------------------------\n",
      "loss: 0.663372  [    0/ 6954]\n",
      "loss: 0.646753  [ 1024/ 6954]\n",
      "loss: 0.654496  [ 2048/ 6954]\n",
      "loss: 0.647022  [ 3072/ 6954]\n",
      "loss: 0.647881  [ 4096/ 6954]\n",
      "loss: 0.661524  [ 5120/ 6954]\n",
      "loss: 0.667776  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 76.4%, Avg loss: 0.005256 \n",
      "\n",
      "Epoch 41\n",
      "-------------------------------\n",
      "loss: 0.662060  [    0/ 6954]\n",
      "loss: 0.659586  [ 1024/ 6954]\n",
      "loss: 0.644480  [ 2048/ 6954]\n",
      "loss: 0.657387  [ 3072/ 6954]\n",
      "loss: 0.638123  [ 4096/ 6954]\n",
      "loss: 0.652699  [ 5120/ 6954]\n",
      "loss: 0.663269  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 76.4%, Avg loss: 0.005250 \n",
      "\n",
      "Epoch 42\n",
      "-------------------------------\n",
      "loss: 0.656980  [    0/ 6954]\n",
      "loss: 0.649676  [ 1024/ 6954]\n",
      "loss: 0.662698  [ 2048/ 6954]\n",
      "loss: 0.649393  [ 3072/ 6954]\n",
      "loss: 0.654171  [ 4096/ 6954]\n",
      "loss: 0.657313  [ 5120/ 6954]\n",
      "loss: 0.656473  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 76.3%, Avg loss: 0.005245 \n",
      "\n",
      "Epoch 43\n",
      "-------------------------------\n",
      "loss: 0.643661  [    0/ 6954]\n",
      "loss: 0.658988  [ 1024/ 6954]\n",
      "loss: 0.648413  [ 2048/ 6954]\n",
      "loss: 0.655363  [ 3072/ 6954]\n",
      "loss: 0.650737  [ 4096/ 6954]\n",
      "loss: 0.658346  [ 5120/ 6954]\n",
      "loss: 0.655144  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 76.4%, Avg loss: 0.005239 \n",
      "\n",
      "Epoch 44\n",
      "-------------------------------\n",
      "loss: 0.653217  [    0/ 6954]\n",
      "loss: 0.646651  [ 1024/ 6954]\n",
      "loss: 0.656797  [ 2048/ 6954]\n",
      "loss: 0.659116  [ 3072/ 6954]\n",
      "loss: 0.653618  [ 4096/ 6954]\n",
      "loss: 0.654492  [ 5120/ 6954]\n",
      "loss: 0.649791  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 76.5%, Avg loss: 0.005233 \n",
      "\n",
      "Epoch 45\n",
      "-------------------------------\n",
      "loss: 0.651878  [    0/ 6954]\n",
      "loss: 0.649462  [ 1024/ 6954]\n",
      "loss: 0.650847  [ 2048/ 6954]\n",
      "loss: 0.663810  [ 3072/ 6954]\n",
      "loss: 0.652054  [ 4096/ 6954]\n",
      "loss: 0.652559  [ 5120/ 6954]\n",
      "loss: 0.655324  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 76.5%, Avg loss: 0.005228 \n",
      "\n",
      "Epoch 46\n",
      "-------------------------------\n",
      "loss: 0.668802  [    0/ 6954]\n",
      "loss: 0.651414  [ 1024/ 6954]\n",
      "loss: 0.659968  [ 2048/ 6954]\n",
      "loss: 0.652690  [ 3072/ 6954]\n",
      "loss: 0.658250  [ 4096/ 6954]\n",
      "loss: 0.660497  [ 5120/ 6954]\n",
      "loss: 0.649915  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 76.5%, Avg loss: 0.005222 \n",
      "\n",
      "Epoch 47\n",
      "-------------------------------\n",
      "loss: 0.655995  [    0/ 6954]\n",
      "loss: 0.655687  [ 1024/ 6954]\n",
      "loss: 0.651634  [ 2048/ 6954]\n",
      "loss: 0.647346  [ 3072/ 6954]\n",
      "loss: 0.651593  [ 4096/ 6954]\n",
      "loss: 0.649622  [ 5120/ 6954]\n",
      "loss: 0.647146  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 76.7%, Avg loss: 0.005217 \n",
      "\n",
      "Epoch 48\n",
      "-------------------------------\n",
      "loss: 0.644102  [    0/ 6954]\n",
      "loss: 0.654003  [ 1024/ 6954]\n",
      "loss: 0.647932  [ 2048/ 6954]\n",
      "loss: 0.646592  [ 3072/ 6954]\n",
      "loss: 0.660583  [ 4096/ 6954]\n",
      "loss: 0.649966  [ 5120/ 6954]\n",
      "loss: 0.665759  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 76.8%, Avg loss: 0.005212 \n",
      "\n",
      "Epoch 49\n",
      "-------------------------------\n",
      "loss: 0.642468  [    0/ 6954]\n",
      "loss: 0.646937  [ 1024/ 6954]\n",
      "loss: 0.646084  [ 2048/ 6954]\n",
      "loss: 0.656913  [ 3072/ 6954]\n",
      "loss: 0.653693  [ 4096/ 6954]\n",
      "loss: 0.638390  [ 5120/ 6954]\n",
      "loss: 0.648977  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 76.9%, Avg loss: 0.005207 \n",
      "\n",
      "Epoch 50\n",
      "-------------------------------\n",
      "loss: 0.650107  [    0/ 6954]\n",
      "loss: 0.650992  [ 1024/ 6954]\n",
      "loss: 0.646243  [ 2048/ 6954]\n",
      "loss: 0.648013  [ 3072/ 6954]\n",
      "loss: 0.644698  [ 4096/ 6954]\n",
      "loss: 0.652870  [ 5120/ 6954]\n",
      "loss: 0.649355  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 77.0%, Avg loss: 0.005201 \n",
      "\n",
      "Epoch 51\n",
      "-------------------------------\n",
      "loss: 0.652834  [    0/ 6954]\n",
      "loss: 0.644070  [ 1024/ 6954]\n",
      "loss: 0.649156  [ 2048/ 6954]\n",
      "loss: 0.638016  [ 3072/ 6954]\n",
      "loss: 0.638957  [ 4096/ 6954]\n",
      "loss: 0.656785  [ 5120/ 6954]\n",
      "loss: 0.646439  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 77.0%, Avg loss: 0.005196 \n",
      "\n",
      "Epoch 52\n",
      "-------------------------------\n",
      "loss: 0.651672  [    0/ 6954]\n",
      "loss: 0.648173  [ 1024/ 6954]\n",
      "loss: 0.646632  [ 2048/ 6954]\n",
      "loss: 0.646714  [ 3072/ 6954]\n",
      "loss: 0.656784  [ 4096/ 6954]\n",
      "loss: 0.640507  [ 5120/ 6954]\n",
      "loss: 0.644804  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 77.0%, Avg loss: 0.005191 \n",
      "\n",
      "Epoch 53\n",
      "-------------------------------\n",
      "loss: 0.642085  [    0/ 6954]\n",
      "loss: 0.658787  [ 1024/ 6954]\n",
      "loss: 0.645013  [ 2048/ 6954]\n",
      "loss: 0.662250  [ 3072/ 6954]\n",
      "loss: 0.650870  [ 4096/ 6954]\n",
      "loss: 0.656671  [ 5120/ 6954]\n",
      "loss: 0.646611  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 77.1%, Avg loss: 0.005186 \n",
      "\n",
      "Epoch 54\n",
      "-------------------------------\n",
      "loss: 0.647379  [    0/ 6954]\n",
      "loss: 0.637491  [ 1024/ 6954]\n",
      "loss: 0.656738  [ 2048/ 6954]\n",
      "loss: 0.656128  [ 3072/ 6954]\n",
      "loss: 0.644035  [ 4096/ 6954]\n",
      "loss: 0.651680  [ 5120/ 6954]\n",
      "loss: 0.640006  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 77.1%, Avg loss: 0.005181 \n",
      "\n",
      "Epoch 55\n",
      "-------------------------------\n",
      "loss: 0.647857  [    0/ 6954]\n",
      "loss: 0.652711  [ 1024/ 6954]\n",
      "loss: 0.654023  [ 2048/ 6954]\n",
      "loss: 0.633476  [ 3072/ 6954]\n",
      "loss: 0.621637  [ 4096/ 6954]\n",
      "loss: 0.647147  [ 5120/ 6954]\n",
      "loss: 0.646338  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 77.1%, Avg loss: 0.005175 \n",
      "\n",
      "Epoch 56\n",
      "-------------------------------\n",
      "loss: 0.647141  [    0/ 6954]\n",
      "loss: 0.647927  [ 1024/ 6954]\n",
      "loss: 0.635665  [ 2048/ 6954]\n",
      "loss: 0.645193  [ 3072/ 6954]\n",
      "loss: 0.637045  [ 4096/ 6954]\n",
      "loss: 0.645281  [ 5120/ 6954]\n",
      "loss: 0.645650  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 77.2%, Avg loss: 0.005170 \n",
      "\n",
      "Epoch 57\n",
      "-------------------------------\n",
      "loss: 0.646817  [    0/ 6954]\n",
      "loss: 0.639518  [ 1024/ 6954]\n",
      "loss: 0.655962  [ 2048/ 6954]\n",
      "loss: 0.644612  [ 3072/ 6954]\n",
      "loss: 0.647609  [ 4096/ 6954]\n",
      "loss: 0.639595  [ 5120/ 6954]\n",
      "loss: 0.637345  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 77.2%, Avg loss: 0.005165 \n",
      "\n",
      "Epoch 58\n",
      "-------------------------------\n",
      "loss: 0.649992  [    0/ 6954]\n",
      "loss: 0.641897  [ 1024/ 6954]\n",
      "loss: 0.654785  [ 2048/ 6954]\n",
      "loss: 0.654482  [ 3072/ 6954]\n",
      "loss: 0.643132  [ 4096/ 6954]\n",
      "loss: 0.632299  [ 5120/ 6954]\n",
      "loss: 0.625058  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 77.3%, Avg loss: 0.005160 \n",
      "\n",
      "Epoch 59\n",
      "-------------------------------\n",
      "loss: 0.646965  [    0/ 6954]\n",
      "loss: 0.634244  [ 1024/ 6954]\n",
      "loss: 0.626685  [ 2048/ 6954]\n",
      "loss: 0.640488  [ 3072/ 6954]\n",
      "loss: 0.652694  [ 4096/ 6954]\n",
      "loss: 0.635134  [ 5120/ 6954]\n",
      "loss: 0.638493  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 77.4%, Avg loss: 0.005155 \n",
      "\n",
      "Epoch 60\n",
      "-------------------------------\n",
      "loss: 0.647409  [    0/ 6954]\n",
      "loss: 0.659733  [ 1024/ 6954]\n",
      "loss: 0.647544  [ 2048/ 6954]\n",
      "loss: 0.644824  [ 3072/ 6954]\n",
      "loss: 0.646634  [ 4096/ 6954]\n",
      "loss: 0.647151  [ 5120/ 6954]\n",
      "loss: 0.644753  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 77.5%, Avg loss: 0.005150 \n",
      "\n",
      "Epoch 61\n",
      "-------------------------------\n",
      "loss: 0.641275  [    0/ 6954]\n",
      "loss: 0.634075  [ 1024/ 6954]\n",
      "loss: 0.643058  [ 2048/ 6954]\n",
      "loss: 0.640815  [ 3072/ 6954]\n",
      "loss: 0.648284  [ 4096/ 6954]\n",
      "loss: 0.635400  [ 5120/ 6954]\n",
      "loss: 0.643036  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 77.6%, Avg loss: 0.005145 \n",
      "\n",
      "Epoch 62\n",
      "-------------------------------\n",
      "loss: 0.640356  [    0/ 6954]\n",
      "loss: 0.634913  [ 1024/ 6954]\n",
      "loss: 0.650316  [ 2048/ 6954]\n",
      "loss: 0.642491  [ 3072/ 6954]\n",
      "loss: 0.649128  [ 4096/ 6954]\n",
      "loss: 0.636924  [ 5120/ 6954]\n",
      "loss: 0.641301  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 77.6%, Avg loss: 0.005140 \n",
      "\n",
      "Epoch 63\n",
      "-------------------------------\n",
      "loss: 0.643985  [    0/ 6954]\n",
      "loss: 0.650717  [ 1024/ 6954]\n",
      "loss: 0.637985  [ 2048/ 6954]\n",
      "loss: 0.639370  [ 3072/ 6954]\n",
      "loss: 0.640490  [ 4096/ 6954]\n",
      "loss: 0.623179  [ 5120/ 6954]\n",
      "loss: 0.641757  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 77.5%, Avg loss: 0.005135 \n",
      "\n",
      "Epoch 64\n",
      "-------------------------------\n",
      "loss: 0.640826  [    0/ 6954]\n",
      "loss: 0.659720  [ 1024/ 6954]\n",
      "loss: 0.658192  [ 2048/ 6954]\n",
      "loss: 0.638162  [ 3072/ 6954]\n",
      "loss: 0.642971  [ 4096/ 6954]\n",
      "loss: 0.651347  [ 5120/ 6954]\n",
      "loss: 0.638406  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 77.3%, Avg loss: 0.005130 \n",
      "\n",
      "Epoch 65\n",
      "-------------------------------\n",
      "loss: 0.666121  [    0/ 6954]\n",
      "loss: 0.646434  [ 1024/ 6954]\n",
      "loss: 0.643743  [ 2048/ 6954]\n",
      "loss: 0.645618  [ 3072/ 6954]\n",
      "loss: 0.628104  [ 4096/ 6954]\n",
      "loss: 0.637337  [ 5120/ 6954]\n",
      "loss: 0.635861  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 77.3%, Avg loss: 0.005125 \n",
      "\n",
      "Epoch 66\n",
      "-------------------------------\n",
      "loss: 0.640292  [    0/ 6954]\n",
      "loss: 0.638745  [ 1024/ 6954]\n",
      "loss: 0.647111  [ 2048/ 6954]\n",
      "loss: 0.631522  [ 3072/ 6954]\n",
      "loss: 0.647967  [ 4096/ 6954]\n",
      "loss: 0.639954  [ 5120/ 6954]\n",
      "loss: 0.641505  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 77.4%, Avg loss: 0.005120 \n",
      "\n",
      "Epoch 67\n",
      "-------------------------------\n",
      "loss: 0.633441  [    0/ 6954]\n",
      "loss: 0.635358  [ 1024/ 6954]\n",
      "loss: 0.631424  [ 2048/ 6954]\n",
      "loss: 0.650522  [ 3072/ 6954]\n",
      "loss: 0.650046  [ 4096/ 6954]\n",
      "loss: 0.650506  [ 5120/ 6954]\n",
      "loss: 0.626157  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 77.5%, Avg loss: 0.005115 \n",
      "\n",
      "Epoch 68\n",
      "-------------------------------\n",
      "loss: 0.645532  [    0/ 6954]\n",
      "loss: 0.639539  [ 1024/ 6954]\n",
      "loss: 0.630360  [ 2048/ 6954]\n",
      "loss: 0.643155  [ 3072/ 6954]\n",
      "loss: 0.643025  [ 4096/ 6954]\n",
      "loss: 0.657229  [ 5120/ 6954]\n",
      "loss: 0.623079  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 77.5%, Avg loss: 0.005110 \n",
      "\n",
      "Epoch 69\n",
      "-------------------------------\n",
      "loss: 0.638191  [    0/ 6954]\n",
      "loss: 0.625644  [ 1024/ 6954]\n",
      "loss: 0.630209  [ 2048/ 6954]\n",
      "loss: 0.640736  [ 3072/ 6954]\n",
      "loss: 0.630449  [ 4096/ 6954]\n",
      "loss: 0.641300  [ 5120/ 6954]\n",
      "loss: 0.642117  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 77.6%, Avg loss: 0.005106 \n",
      "\n",
      "Epoch 70\n",
      "-------------------------------\n",
      "loss: 0.653845  [    0/ 6954]\n",
      "loss: 0.638847  [ 1024/ 6954]\n",
      "loss: 0.628436  [ 2048/ 6954]\n",
      "loss: 0.633101  [ 3072/ 6954]\n",
      "loss: 0.639607  [ 4096/ 6954]\n",
      "loss: 0.641284  [ 5120/ 6954]\n",
      "loss: 0.648162  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 77.7%, Avg loss: 0.005101 \n",
      "\n",
      "Epoch 71\n",
      "-------------------------------\n",
      "loss: 0.649406  [    0/ 6954]\n",
      "loss: 0.633209  [ 1024/ 6954]\n",
      "loss: 0.635237  [ 2048/ 6954]\n",
      "loss: 0.641601  [ 3072/ 6954]\n",
      "loss: 0.638263  [ 4096/ 6954]\n",
      "loss: 0.629488  [ 5120/ 6954]\n",
      "loss: 0.645884  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 77.7%, Avg loss: 0.005096 \n",
      "\n",
      "Epoch 72\n",
      "-------------------------------\n",
      "loss: 0.648501  [    0/ 6954]\n",
      "loss: 0.623571  [ 1024/ 6954]\n",
      "loss: 0.634771  [ 2048/ 6954]\n",
      "loss: 0.641905  [ 3072/ 6954]\n",
      "loss: 0.641311  [ 4096/ 6954]\n",
      "loss: 0.635261  [ 5120/ 6954]\n",
      "loss: 0.631460  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 77.9%, Avg loss: 0.005091 \n",
      "\n",
      "Epoch 73\n",
      "-------------------------------\n",
      "loss: 0.628474  [    0/ 6954]\n",
      "loss: 0.626973  [ 1024/ 6954]\n",
      "loss: 0.637892  [ 2048/ 6954]\n",
      "loss: 0.632809  [ 3072/ 6954]\n",
      "loss: 0.637168  [ 4096/ 6954]\n",
      "loss: 0.619017  [ 5120/ 6954]\n",
      "loss: 0.635370  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 77.9%, Avg loss: 0.005086 \n",
      "\n",
      "Epoch 74\n",
      "-------------------------------\n",
      "loss: 0.648132  [    0/ 6954]\n",
      "loss: 0.649654  [ 1024/ 6954]\n",
      "loss: 0.638339  [ 2048/ 6954]\n",
      "loss: 0.631300  [ 3072/ 6954]\n",
      "loss: 0.643773  [ 4096/ 6954]\n",
      "loss: 0.637244  [ 5120/ 6954]\n",
      "loss: 0.633189  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 77.9%, Avg loss: 0.005082 \n",
      "\n",
      "Epoch 75\n",
      "-------------------------------\n",
      "loss: 0.644646  [    0/ 6954]\n",
      "loss: 0.632374  [ 1024/ 6954]\n",
      "loss: 0.621588  [ 2048/ 6954]\n",
      "loss: 0.635039  [ 3072/ 6954]\n",
      "loss: 0.638376  [ 4096/ 6954]\n",
      "loss: 0.650015  [ 5120/ 6954]\n",
      "loss: 0.630248  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 77.9%, Avg loss: 0.005077 \n",
      "\n",
      "Epoch 76\n",
      "-------------------------------\n",
      "loss: 0.625403  [    0/ 6954]\n",
      "loss: 0.645020  [ 1024/ 6954]\n",
      "loss: 0.627652  [ 2048/ 6954]\n",
      "loss: 0.635092  [ 3072/ 6954]\n",
      "loss: 0.634074  [ 4096/ 6954]\n",
      "loss: 0.621899  [ 5120/ 6954]\n",
      "loss: 0.635295  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 77.9%, Avg loss: 0.005072 \n",
      "\n",
      "Epoch 77\n",
      "-------------------------------\n",
      "loss: 0.637212  [    0/ 6954]\n",
      "loss: 0.647237  [ 1024/ 6954]\n",
      "loss: 0.632766  [ 2048/ 6954]\n",
      "loss: 0.644922  [ 3072/ 6954]\n",
      "loss: 0.631982  [ 4096/ 6954]\n",
      "loss: 0.630910  [ 5120/ 6954]\n",
      "loss: 0.640356  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 78.0%, Avg loss: 0.005067 \n",
      "\n",
      "Epoch 78\n",
      "-------------------------------\n",
      "loss: 0.634066  [    0/ 6954]\n",
      "loss: 0.637952  [ 1024/ 6954]\n",
      "loss: 0.618576  [ 2048/ 6954]\n",
      "loss: 0.645381  [ 3072/ 6954]\n",
      "loss: 0.632397  [ 4096/ 6954]\n",
      "loss: 0.647851  [ 5120/ 6954]\n",
      "loss: 0.630429  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 78.0%, Avg loss: 0.005062 \n",
      "\n",
      "Epoch 79\n",
      "-------------------------------\n",
      "loss: 0.614661  [    0/ 6954]\n",
      "loss: 0.614307  [ 1024/ 6954]\n",
      "loss: 0.641801  [ 2048/ 6954]\n",
      "loss: 0.625561  [ 3072/ 6954]\n",
      "loss: 0.644461  [ 4096/ 6954]\n",
      "loss: 0.630580  [ 5120/ 6954]\n",
      "loss: 0.641663  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 78.0%, Avg loss: 0.005058 \n",
      "\n",
      "Epoch 80\n",
      "-------------------------------\n",
      "loss: 0.629615  [    0/ 6954]\n",
      "loss: 0.622594  [ 1024/ 6954]\n",
      "loss: 0.628706  [ 2048/ 6954]\n",
      "loss: 0.635691  [ 3072/ 6954]\n",
      "loss: 0.644253  [ 4096/ 6954]\n",
      "loss: 0.646173  [ 5120/ 6954]\n",
      "loss: 0.629050  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 78.1%, Avg loss: 0.005053 \n",
      "\n",
      "Epoch 81\n",
      "-------------------------------\n",
      "loss: 0.635758  [    0/ 6954]\n",
      "loss: 0.612368  [ 1024/ 6954]\n",
      "loss: 0.648236  [ 2048/ 6954]\n",
      "loss: 0.632367  [ 3072/ 6954]\n",
      "loss: 0.618585  [ 4096/ 6954]\n",
      "loss: 0.639389  [ 5120/ 6954]\n",
      "loss: 0.633157  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 78.1%, Avg loss: 0.005048 \n",
      "\n",
      "Epoch 82\n",
      "-------------------------------\n",
      "loss: 0.630937  [    0/ 6954]\n",
      "loss: 0.648893  [ 1024/ 6954]\n",
      "loss: 0.648106  [ 2048/ 6954]\n",
      "loss: 0.615293  [ 3072/ 6954]\n",
      "loss: 0.641362  [ 4096/ 6954]\n",
      "loss: 0.628854  [ 5120/ 6954]\n",
      "loss: 0.643204  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 78.2%, Avg loss: 0.005044 \n",
      "\n",
      "Epoch 83\n",
      "-------------------------------\n",
      "loss: 0.624395  [    0/ 6954]\n",
      "loss: 0.614295  [ 1024/ 6954]\n",
      "loss: 0.629331  [ 2048/ 6954]\n",
      "loss: 0.637657  [ 3072/ 6954]\n",
      "loss: 0.621710  [ 4096/ 6954]\n",
      "loss: 0.627639  [ 5120/ 6954]\n",
      "loss: 0.618262  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 78.1%, Avg loss: 0.005039 \n",
      "\n",
      "Epoch 84\n",
      "-------------------------------\n",
      "loss: 0.626433  [    0/ 6954]\n",
      "loss: 0.641962  [ 1024/ 6954]\n",
      "loss: 0.645317  [ 2048/ 6954]\n",
      "loss: 0.647976  [ 3072/ 6954]\n",
      "loss: 0.627821  [ 4096/ 6954]\n",
      "loss: 0.621100  [ 5120/ 6954]\n",
      "loss: 0.627530  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 78.2%, Avg loss: 0.005035 \n",
      "\n",
      "Epoch 85\n",
      "-------------------------------\n",
      "loss: 0.638872  [    0/ 6954]\n",
      "loss: 0.636169  [ 1024/ 6954]\n",
      "loss: 0.635205  [ 2048/ 6954]\n",
      "loss: 0.643333  [ 3072/ 6954]\n",
      "loss: 0.638999  [ 4096/ 6954]\n",
      "loss: 0.626295  [ 5120/ 6954]\n",
      "loss: 0.635374  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 78.1%, Avg loss: 0.005030 \n",
      "\n",
      "Epoch 86\n",
      "-------------------------------\n",
      "loss: 0.638581  [    0/ 6954]\n",
      "loss: 0.634994  [ 1024/ 6954]\n",
      "loss: 0.627768  [ 2048/ 6954]\n",
      "loss: 0.639261  [ 3072/ 6954]\n",
      "loss: 0.636361  [ 4096/ 6954]\n",
      "loss: 0.614507  [ 5120/ 6954]\n",
      "loss: 0.604924  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 78.1%, Avg loss: 0.005025 \n",
      "\n",
      "Epoch 87\n",
      "-------------------------------\n",
      "loss: 0.638147  [    0/ 6954]\n",
      "loss: 0.631485  [ 1024/ 6954]\n",
      "loss: 0.641578  [ 2048/ 6954]\n",
      "loss: 0.622526  [ 3072/ 6954]\n",
      "loss: 0.629709  [ 4096/ 6954]\n",
      "loss: 0.614211  [ 5120/ 6954]\n",
      "loss: 0.637417  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 78.2%, Avg loss: 0.005021 \n",
      "\n",
      "Epoch 88\n",
      "-------------------------------\n",
      "loss: 0.623925  [    0/ 6954]\n",
      "loss: 0.622184  [ 1024/ 6954]\n",
      "loss: 0.628581  [ 2048/ 6954]\n",
      "loss: 0.625324  [ 3072/ 6954]\n",
      "loss: 0.633104  [ 4096/ 6954]\n",
      "loss: 0.627878  [ 5120/ 6954]\n",
      "loss: 0.617491  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 78.5%, Avg loss: 0.005016 \n",
      "\n",
      "Epoch 89\n",
      "-------------------------------\n",
      "loss: 0.620464  [    0/ 6954]\n",
      "loss: 0.632465  [ 1024/ 6954]\n",
      "loss: 0.629055  [ 2048/ 6954]\n",
      "loss: 0.641155  [ 3072/ 6954]\n",
      "loss: 0.634100  [ 4096/ 6954]\n",
      "loss: 0.645496  [ 5120/ 6954]\n",
      "loss: 0.622010  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 78.5%, Avg loss: 0.005012 \n",
      "\n",
      "Epoch 90\n",
      "-------------------------------\n",
      "loss: 0.634317  [    0/ 6954]\n",
      "loss: 0.634797  [ 1024/ 6954]\n",
      "loss: 0.625991  [ 2048/ 6954]\n",
      "loss: 0.630826  [ 3072/ 6954]\n",
      "loss: 0.628920  [ 4096/ 6954]\n",
      "loss: 0.616822  [ 5120/ 6954]\n",
      "loss: 0.627921  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 78.6%, Avg loss: 0.005007 \n",
      "\n",
      "Epoch 91\n",
      "-------------------------------\n",
      "loss: 0.624290  [    0/ 6954]\n",
      "loss: 0.623066  [ 1024/ 6954]\n",
      "loss: 0.634873  [ 2048/ 6954]\n",
      "loss: 0.631035  [ 3072/ 6954]\n",
      "loss: 0.627842  [ 4096/ 6954]\n",
      "loss: 0.625354  [ 5120/ 6954]\n",
      "loss: 0.629627  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 78.8%, Avg loss: 0.005002 \n",
      "\n",
      "Epoch 92\n",
      "-------------------------------\n",
      "loss: 0.619061  [    0/ 6954]\n",
      "loss: 0.613437  [ 1024/ 6954]\n",
      "loss: 0.625300  [ 2048/ 6954]\n",
      "loss: 0.629702  [ 3072/ 6954]\n",
      "loss: 0.612718  [ 4096/ 6954]\n",
      "loss: 0.619365  [ 5120/ 6954]\n",
      "loss: 0.608149  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 78.7%, Avg loss: 0.004998 \n",
      "\n",
      "Epoch 93\n",
      "-------------------------------\n",
      "loss: 0.638298  [    0/ 6954]\n",
      "loss: 0.632430  [ 1024/ 6954]\n",
      "loss: 0.619715  [ 2048/ 6954]\n",
      "loss: 0.635276  [ 3072/ 6954]\n",
      "loss: 0.630512  [ 4096/ 6954]\n",
      "loss: 0.634674  [ 5120/ 6954]\n",
      "loss: 0.618966  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 78.7%, Avg loss: 0.004993 \n",
      "\n",
      "Epoch 94\n",
      "-------------------------------\n",
      "loss: 0.620605  [    0/ 6954]\n",
      "loss: 0.624458  [ 1024/ 6954]\n",
      "loss: 0.628065  [ 2048/ 6954]\n",
      "loss: 0.628625  [ 3072/ 6954]\n",
      "loss: 0.618435  [ 4096/ 6954]\n",
      "loss: 0.627331  [ 5120/ 6954]\n",
      "loss: 0.621577  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 78.7%, Avg loss: 0.004989 \n",
      "\n",
      "Epoch 95\n",
      "-------------------------------\n",
      "loss: 0.623020  [    0/ 6954]\n",
      "loss: 0.608261  [ 1024/ 6954]\n",
      "loss: 0.619843  [ 2048/ 6954]\n",
      "loss: 0.638778  [ 3072/ 6954]\n",
      "loss: 0.638207  [ 4096/ 6954]\n",
      "loss: 0.638961  [ 5120/ 6954]\n",
      "loss: 0.640843  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 78.7%, Avg loss: 0.004984 \n",
      "\n",
      "Epoch 96\n",
      "-------------------------------\n",
      "loss: 0.630616  [    0/ 6954]\n",
      "loss: 0.627875  [ 1024/ 6954]\n",
      "loss: 0.621817  [ 2048/ 6954]\n",
      "loss: 0.639773  [ 3072/ 6954]\n",
      "loss: 0.617389  [ 4096/ 6954]\n",
      "loss: 0.621094  [ 5120/ 6954]\n",
      "loss: 0.629501  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 78.7%, Avg loss: 0.004980 \n",
      "\n",
      "Epoch 97\n",
      "-------------------------------\n",
      "loss: 0.619887  [    0/ 6954]\n",
      "loss: 0.635482  [ 1024/ 6954]\n",
      "loss: 0.643695  [ 2048/ 6954]\n",
      "loss: 0.624261  [ 3072/ 6954]\n",
      "loss: 0.618312  [ 4096/ 6954]\n",
      "loss: 0.612811  [ 5120/ 6954]\n",
      "loss: 0.629148  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 78.7%, Avg loss: 0.004975 \n",
      "\n",
      "Epoch 98\n",
      "-------------------------------\n",
      "loss: 0.628958  [    0/ 6954]\n",
      "loss: 0.619769  [ 1024/ 6954]\n",
      "loss: 0.620986  [ 2048/ 6954]\n",
      "loss: 0.619920  [ 3072/ 6954]\n",
      "loss: 0.625369  [ 4096/ 6954]\n",
      "loss: 0.623356  [ 5120/ 6954]\n",
      "loss: 0.601119  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 78.6%, Avg loss: 0.004971 \n",
      "\n",
      "Epoch 99\n",
      "-------------------------------\n",
      "loss: 0.622070  [    0/ 6954]\n",
      "loss: 0.602666  [ 1024/ 6954]\n",
      "loss: 0.625104  [ 2048/ 6954]\n",
      "loss: 0.614964  [ 3072/ 6954]\n",
      "loss: 0.622689  [ 4096/ 6954]\n",
      "loss: 0.615910  [ 5120/ 6954]\n",
      "loss: 0.624809  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 78.6%, Avg loss: 0.004966 \n",
      "\n",
      "Epoch 100\n",
      "-------------------------------\n",
      "loss: 0.616058  [    0/ 6954]\n",
      "loss: 0.619846  [ 1024/ 6954]\n",
      "loss: 0.617851  [ 2048/ 6954]\n",
      "loss: 0.610928  [ 3072/ 6954]\n",
      "loss: 0.619692  [ 4096/ 6954]\n",
      "loss: 0.625237  [ 5120/ 6954]\n",
      "loss: 0.621325  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 78.7%, Avg loss: 0.004962 \n",
      "\n",
      "Epoch 101\n",
      "-------------------------------\n",
      "loss: 0.618151  [    0/ 6954]\n",
      "loss: 0.616302  [ 1024/ 6954]\n",
      "loss: 0.613320  [ 2048/ 6954]\n",
      "loss: 0.628822  [ 3072/ 6954]\n",
      "loss: 0.610007  [ 4096/ 6954]\n",
      "loss: 0.630237  [ 5120/ 6954]\n",
      "loss: 0.642082  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 78.8%, Avg loss: 0.004957 \n",
      "\n",
      "Epoch 102\n",
      "-------------------------------\n",
      "loss: 0.627816  [    0/ 6954]\n",
      "loss: 0.611064  [ 1024/ 6954]\n",
      "loss: 0.620078  [ 2048/ 6954]\n",
      "loss: 0.628515  [ 3072/ 6954]\n",
      "loss: 0.608306  [ 4096/ 6954]\n",
      "loss: 0.618585  [ 5120/ 6954]\n",
      "loss: 0.635461  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 78.8%, Avg loss: 0.004952 \n",
      "\n",
      "Epoch 103\n",
      "-------------------------------\n",
      "loss: 0.615528  [    0/ 6954]\n",
      "loss: 0.611871  [ 1024/ 6954]\n",
      "loss: 0.632227  [ 2048/ 6954]\n",
      "loss: 0.616908  [ 3072/ 6954]\n",
      "loss: 0.620082  [ 4096/ 6954]\n",
      "loss: 0.619078  [ 5120/ 6954]\n",
      "loss: 0.623767  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 78.8%, Avg loss: 0.004948 \n",
      "\n",
      "Epoch 104\n",
      "-------------------------------\n",
      "loss: 0.605997  [    0/ 6954]\n",
      "loss: 0.630579  [ 1024/ 6954]\n",
      "loss: 0.614772  [ 2048/ 6954]\n",
      "loss: 0.619690  [ 3072/ 6954]\n",
      "loss: 0.613541  [ 4096/ 6954]\n",
      "loss: 0.610956  [ 5120/ 6954]\n",
      "loss: 0.621802  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 78.8%, Avg loss: 0.004944 \n",
      "\n",
      "Epoch 105\n",
      "-------------------------------\n",
      "loss: 0.607015  [    0/ 6954]\n",
      "loss: 0.631010  [ 1024/ 6954]\n",
      "loss: 0.629594  [ 2048/ 6954]\n",
      "loss: 0.620664  [ 3072/ 6954]\n",
      "loss: 0.618204  [ 4096/ 6954]\n",
      "loss: 0.625509  [ 5120/ 6954]\n",
      "loss: 0.606686  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 78.8%, Avg loss: 0.004939 \n",
      "\n",
      "Epoch 106\n",
      "-------------------------------\n",
      "loss: 0.627035  [    0/ 6954]\n",
      "loss: 0.623322  [ 1024/ 6954]\n",
      "loss: 0.624789  [ 2048/ 6954]\n",
      "loss: 0.613007  [ 3072/ 6954]\n",
      "loss: 0.604784  [ 4096/ 6954]\n",
      "loss: 0.624711  [ 5120/ 6954]\n",
      "loss: 0.618278  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 78.8%, Avg loss: 0.004935 \n",
      "\n",
      "Epoch 107\n",
      "-------------------------------\n",
      "loss: 0.618204  [    0/ 6954]\n",
      "loss: 0.617574  [ 1024/ 6954]\n",
      "loss: 0.604875  [ 2048/ 6954]\n",
      "loss: 0.631549  [ 3072/ 6954]\n",
      "loss: 0.641211  [ 4096/ 6954]\n",
      "loss: 0.618951  [ 5120/ 6954]\n",
      "loss: 0.633075  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 78.8%, Avg loss: 0.004930 \n",
      "\n",
      "Epoch 108\n",
      "-------------------------------\n",
      "loss: 0.615838  [    0/ 6954]\n",
      "loss: 0.630810  [ 1024/ 6954]\n",
      "loss: 0.591800  [ 2048/ 6954]\n",
      "loss: 0.629466  [ 3072/ 6954]\n",
      "loss: 0.613447  [ 4096/ 6954]\n",
      "loss: 0.621041  [ 5120/ 6954]\n",
      "loss: 0.616859  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 78.9%, Avg loss: 0.004926 \n",
      "\n",
      "Epoch 109\n",
      "-------------------------------\n",
      "loss: 0.612630  [    0/ 6954]\n",
      "loss: 0.607912  [ 1024/ 6954]\n",
      "loss: 0.619495  [ 2048/ 6954]\n",
      "loss: 0.619974  [ 3072/ 6954]\n",
      "loss: 0.644328  [ 4096/ 6954]\n",
      "loss: 0.623729  [ 5120/ 6954]\n",
      "loss: 0.610019  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 79.0%, Avg loss: 0.004922 \n",
      "\n",
      "Epoch 110\n",
      "-------------------------------\n",
      "loss: 0.604778  [    0/ 6954]\n",
      "loss: 0.618104  [ 1024/ 6954]\n",
      "loss: 0.598952  [ 2048/ 6954]\n",
      "loss: 0.630924  [ 3072/ 6954]\n",
      "loss: 0.602978  [ 4096/ 6954]\n",
      "loss: 0.633566  [ 5120/ 6954]\n",
      "loss: 0.627149  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 79.0%, Avg loss: 0.004917 \n",
      "\n",
      "Epoch 111\n",
      "-------------------------------\n",
      "loss: 0.620693  [    0/ 6954]\n",
      "loss: 0.619990  [ 1024/ 6954]\n",
      "loss: 0.613132  [ 2048/ 6954]\n",
      "loss: 0.600375  [ 3072/ 6954]\n",
      "loss: 0.622656  [ 4096/ 6954]\n",
      "loss: 0.610073  [ 5120/ 6954]\n",
      "loss: 0.616202  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 79.0%, Avg loss: 0.004913 \n",
      "\n",
      "Epoch 112\n",
      "-------------------------------\n",
      "loss: 0.618196  [    0/ 6954]\n",
      "loss: 0.615820  [ 1024/ 6954]\n",
      "loss: 0.610314  [ 2048/ 6954]\n",
      "loss: 0.615054  [ 3072/ 6954]\n",
      "loss: 0.607397  [ 4096/ 6954]\n",
      "loss: 0.599726  [ 5120/ 6954]\n",
      "loss: 0.598218  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 79.0%, Avg loss: 0.004909 \n",
      "\n",
      "Epoch 113\n",
      "-------------------------------\n",
      "loss: 0.603204  [    0/ 6954]\n",
      "loss: 0.599520  [ 1024/ 6954]\n",
      "loss: 0.612777  [ 2048/ 6954]\n",
      "loss: 0.624886  [ 3072/ 6954]\n",
      "loss: 0.617622  [ 4096/ 6954]\n",
      "loss: 0.597472  [ 5120/ 6954]\n",
      "loss: 0.619085  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 79.1%, Avg loss: 0.004905 \n",
      "\n",
      "Epoch 114\n",
      "-------------------------------\n",
      "loss: 0.624414  [    0/ 6954]\n",
      "loss: 0.619148  [ 1024/ 6954]\n",
      "loss: 0.631915  [ 2048/ 6954]\n",
      "loss: 0.613164  [ 3072/ 6954]\n",
      "loss: 0.625879  [ 4096/ 6954]\n",
      "loss: 0.603886  [ 5120/ 6954]\n",
      "loss: 0.621161  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 79.4%, Avg loss: 0.004900 \n",
      "\n",
      "Epoch 115\n",
      "-------------------------------\n",
      "loss: 0.616754  [    0/ 6954]\n",
      "loss: 0.618616  [ 1024/ 6954]\n",
      "loss: 0.614549  [ 2048/ 6954]\n",
      "loss: 0.622276  [ 3072/ 6954]\n",
      "loss: 0.613890  [ 4096/ 6954]\n",
      "loss: 0.594706  [ 5120/ 6954]\n",
      "loss: 0.629987  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 79.4%, Avg loss: 0.004896 \n",
      "\n",
      "Epoch 116\n",
      "-------------------------------\n",
      "loss: 0.604522  [    0/ 6954]\n",
      "loss: 0.621493  [ 1024/ 6954]\n",
      "loss: 0.612201  [ 2048/ 6954]\n",
      "loss: 0.618705  [ 3072/ 6954]\n",
      "loss: 0.597899  [ 4096/ 6954]\n",
      "loss: 0.614314  [ 5120/ 6954]\n",
      "loss: 0.595221  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 79.5%, Avg loss: 0.004892 \n",
      "\n",
      "Epoch 117\n",
      "-------------------------------\n",
      "loss: 0.600153  [    0/ 6954]\n",
      "loss: 0.620628  [ 1024/ 6954]\n",
      "loss: 0.594628  [ 2048/ 6954]\n",
      "loss: 0.611659  [ 3072/ 6954]\n",
      "loss: 0.603585  [ 4096/ 6954]\n",
      "loss: 0.615999  [ 5120/ 6954]\n",
      "loss: 0.615568  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 79.5%, Avg loss: 0.004888 \n",
      "\n",
      "Epoch 118\n",
      "-------------------------------\n",
      "loss: 0.611247  [    0/ 6954]\n",
      "loss: 0.609724  [ 1024/ 6954]\n",
      "loss: 0.623733  [ 2048/ 6954]\n",
      "loss: 0.611944  [ 3072/ 6954]\n",
      "loss: 0.623565  [ 4096/ 6954]\n",
      "loss: 0.618088  [ 5120/ 6954]\n",
      "loss: 0.601926  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 79.5%, Avg loss: 0.004883 \n",
      "\n",
      "Epoch 119\n",
      "-------------------------------\n",
      "loss: 0.607275  [    0/ 6954]\n",
      "loss: 0.625096  [ 1024/ 6954]\n",
      "loss: 0.620760  [ 2048/ 6954]\n",
      "loss: 0.622567  [ 3072/ 6954]\n",
      "loss: 0.616852  [ 4096/ 6954]\n",
      "loss: 0.591481  [ 5120/ 6954]\n",
      "loss: 0.591362  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 79.6%, Avg loss: 0.004879 \n",
      "\n",
      "Epoch 120\n",
      "-------------------------------\n",
      "loss: 0.605793  [    0/ 6954]\n",
      "loss: 0.630760  [ 1024/ 6954]\n",
      "loss: 0.620565  [ 2048/ 6954]\n",
      "loss: 0.607357  [ 3072/ 6954]\n",
      "loss: 0.621727  [ 4096/ 6954]\n",
      "loss: 0.615984  [ 5120/ 6954]\n",
      "loss: 0.606149  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 79.8%, Avg loss: 0.004875 \n",
      "\n",
      "Epoch 121\n",
      "-------------------------------\n",
      "loss: 0.632836  [    0/ 6954]\n",
      "loss: 0.604297  [ 1024/ 6954]\n",
      "loss: 0.598712  [ 2048/ 6954]\n",
      "loss: 0.602898  [ 3072/ 6954]\n",
      "loss: 0.610355  [ 4096/ 6954]\n",
      "loss: 0.623567  [ 5120/ 6954]\n",
      "loss: 0.612305  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 79.8%, Avg loss: 0.004871 \n",
      "\n",
      "Epoch 122\n",
      "-------------------------------\n",
      "loss: 0.635742  [    0/ 6954]\n",
      "loss: 0.615839  [ 1024/ 6954]\n",
      "loss: 0.597251  [ 2048/ 6954]\n",
      "loss: 0.602564  [ 3072/ 6954]\n",
      "loss: 0.606307  [ 4096/ 6954]\n",
      "loss: 0.592439  [ 5120/ 6954]\n",
      "loss: 0.603319  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 79.8%, Avg loss: 0.004867 \n",
      "\n",
      "Epoch 123\n",
      "-------------------------------\n",
      "loss: 0.606704  [    0/ 6954]\n",
      "loss: 0.632648  [ 1024/ 6954]\n",
      "loss: 0.607652  [ 2048/ 6954]\n",
      "loss: 0.630455  [ 3072/ 6954]\n",
      "loss: 0.603966  [ 4096/ 6954]\n",
      "loss: 0.608623  [ 5120/ 6954]\n",
      "loss: 0.596511  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 79.9%, Avg loss: 0.004863 \n",
      "\n",
      "Epoch 124\n",
      "-------------------------------\n",
      "loss: 0.615883  [    0/ 6954]\n",
      "loss: 0.614257  [ 1024/ 6954]\n",
      "loss: 0.606144  [ 2048/ 6954]\n",
      "loss: 0.625853  [ 3072/ 6954]\n",
      "loss: 0.621173  [ 4096/ 6954]\n",
      "loss: 0.628904  [ 5120/ 6954]\n",
      "loss: 0.615097  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.0%, Avg loss: 0.004858 \n",
      "\n",
      "Epoch 125\n",
      "-------------------------------\n",
      "loss: 0.614588  [    0/ 6954]\n",
      "loss: 0.599953  [ 1024/ 6954]\n",
      "loss: 0.602659  [ 2048/ 6954]\n",
      "loss: 0.604721  [ 3072/ 6954]\n",
      "loss: 0.607911  [ 4096/ 6954]\n",
      "loss: 0.594952  [ 5120/ 6954]\n",
      "loss: 0.614937  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.0%, Avg loss: 0.004854 \n",
      "\n",
      "Epoch 126\n",
      "-------------------------------\n",
      "loss: 0.624259  [    0/ 6954]\n",
      "loss: 0.596770  [ 1024/ 6954]\n",
      "loss: 0.612135  [ 2048/ 6954]\n",
      "loss: 0.609767  [ 3072/ 6954]\n",
      "loss: 0.607465  [ 4096/ 6954]\n",
      "loss: 0.623997  [ 5120/ 6954]\n",
      "loss: 0.603875  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.0%, Avg loss: 0.004850 \n",
      "\n",
      "Epoch 127\n",
      "-------------------------------\n",
      "loss: 0.593778  [    0/ 6954]\n",
      "loss: 0.607420  [ 1024/ 6954]\n",
      "loss: 0.623488  [ 2048/ 6954]\n",
      "loss: 0.605351  [ 3072/ 6954]\n",
      "loss: 0.592107  [ 4096/ 6954]\n",
      "loss: 0.607222  [ 5120/ 6954]\n",
      "loss: 0.606766  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.0%, Avg loss: 0.004846 \n",
      "\n",
      "Epoch 128\n",
      "-------------------------------\n",
      "loss: 0.600497  [    0/ 6954]\n",
      "loss: 0.610779  [ 1024/ 6954]\n",
      "loss: 0.618075  [ 2048/ 6954]\n",
      "loss: 0.589267  [ 3072/ 6954]\n",
      "loss: 0.634051  [ 4096/ 6954]\n",
      "loss: 0.606700  [ 5120/ 6954]\n",
      "loss: 0.597539  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.0%, Avg loss: 0.004842 \n",
      "\n",
      "Epoch 129\n",
      "-------------------------------\n",
      "loss: 0.613633  [    0/ 6954]\n",
      "loss: 0.596116  [ 1024/ 6954]\n",
      "loss: 0.595328  [ 2048/ 6954]\n",
      "loss: 0.609041  [ 3072/ 6954]\n",
      "loss: 0.606360  [ 4096/ 6954]\n",
      "loss: 0.608473  [ 5120/ 6954]\n",
      "loss: 0.615623  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.0%, Avg loss: 0.004838 \n",
      "\n",
      "Epoch 130\n",
      "-------------------------------\n",
      "loss: 0.632816  [    0/ 6954]\n",
      "loss: 0.592983  [ 1024/ 6954]\n",
      "loss: 0.602055  [ 2048/ 6954]\n",
      "loss: 0.597167  [ 3072/ 6954]\n",
      "loss: 0.597470  [ 4096/ 6954]\n",
      "loss: 0.632121  [ 5120/ 6954]\n",
      "loss: 0.602089  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.1%, Avg loss: 0.004834 \n",
      "\n",
      "Epoch 131\n",
      "-------------------------------\n",
      "loss: 0.610037  [    0/ 6954]\n",
      "loss: 0.600967  [ 1024/ 6954]\n",
      "loss: 0.621865  [ 2048/ 6954]\n",
      "loss: 0.603629  [ 3072/ 6954]\n",
      "loss: 0.601987  [ 4096/ 6954]\n",
      "loss: 0.601947  [ 5120/ 6954]\n",
      "loss: 0.611793  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.1%, Avg loss: 0.004831 \n",
      "\n",
      "Epoch 132\n",
      "-------------------------------\n",
      "loss: 0.603837  [    0/ 6954]\n",
      "loss: 0.592662  [ 1024/ 6954]\n",
      "loss: 0.592636  [ 2048/ 6954]\n",
      "loss: 0.614505  [ 3072/ 6954]\n",
      "loss: 0.618693  [ 4096/ 6954]\n",
      "loss: 0.607137  [ 5120/ 6954]\n",
      "loss: 0.612016  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.1%, Avg loss: 0.004827 \n",
      "\n",
      "Epoch 133\n",
      "-------------------------------\n",
      "loss: 0.577813  [    0/ 6954]\n",
      "loss: 0.603370  [ 1024/ 6954]\n",
      "loss: 0.597421  [ 2048/ 6954]\n",
      "loss: 0.621277  [ 3072/ 6954]\n",
      "loss: 0.604357  [ 4096/ 6954]\n",
      "loss: 0.620386  [ 5120/ 6954]\n",
      "loss: 0.604191  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.1%, Avg loss: 0.004823 \n",
      "\n",
      "Epoch 134\n",
      "-------------------------------\n",
      "loss: 0.583203  [    0/ 6954]\n",
      "loss: 0.612790  [ 1024/ 6954]\n",
      "loss: 0.581267  [ 2048/ 6954]\n",
      "loss: 0.605259  [ 3072/ 6954]\n",
      "loss: 0.589772  [ 4096/ 6954]\n",
      "loss: 0.593616  [ 5120/ 6954]\n",
      "loss: 0.600659  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.0%, Avg loss: 0.004819 \n",
      "\n",
      "Epoch 135\n",
      "-------------------------------\n",
      "loss: 0.612743  [    0/ 6954]\n",
      "loss: 0.630531  [ 1024/ 6954]\n",
      "loss: 0.609869  [ 2048/ 6954]\n",
      "loss: 0.618164  [ 3072/ 6954]\n",
      "loss: 0.603975  [ 4096/ 6954]\n",
      "loss: 0.601062  [ 5120/ 6954]\n",
      "loss: 0.593811  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.0%, Avg loss: 0.004815 \n",
      "\n",
      "Epoch 136\n",
      "-------------------------------\n",
      "loss: 0.619800  [    0/ 6954]\n",
      "loss: 0.614448  [ 1024/ 6954]\n",
      "loss: 0.628334  [ 2048/ 6954]\n",
      "loss: 0.601308  [ 3072/ 6954]\n",
      "loss: 0.609131  [ 4096/ 6954]\n",
      "loss: 0.651805  [ 5120/ 6954]\n",
      "loss: 0.599949  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.2%, Avg loss: 0.004811 \n",
      "\n",
      "Epoch 137\n",
      "-------------------------------\n",
      "loss: 0.620946  [    0/ 6954]\n",
      "loss: 0.603201  [ 1024/ 6954]\n",
      "loss: 0.584811  [ 2048/ 6954]\n",
      "loss: 0.606878  [ 3072/ 6954]\n",
      "loss: 0.607677  [ 4096/ 6954]\n",
      "loss: 0.611829  [ 5120/ 6954]\n",
      "loss: 0.602480  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.2%, Avg loss: 0.004807 \n",
      "\n",
      "Epoch 138\n",
      "-------------------------------\n",
      "loss: 0.617785  [    0/ 6954]\n",
      "loss: 0.603535  [ 1024/ 6954]\n",
      "loss: 0.606799  [ 2048/ 6954]\n",
      "loss: 0.578649  [ 3072/ 6954]\n",
      "loss: 0.596403  [ 4096/ 6954]\n",
      "loss: 0.626489  [ 5120/ 6954]\n",
      "loss: 0.589681  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.1%, Avg loss: 0.004804 \n",
      "\n",
      "Epoch 139\n",
      "-------------------------------\n",
      "loss: 0.625359  [    0/ 6954]\n",
      "loss: 0.598124  [ 1024/ 6954]\n",
      "loss: 0.592686  [ 2048/ 6954]\n",
      "loss: 0.630656  [ 3072/ 6954]\n",
      "loss: 0.610677  [ 4096/ 6954]\n",
      "loss: 0.592104  [ 5120/ 6954]\n",
      "loss: 0.614841  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.1%, Avg loss: 0.004800 \n",
      "\n",
      "Epoch 140\n",
      "-------------------------------\n",
      "loss: 0.586779  [    0/ 6954]\n",
      "loss: 0.600269  [ 1024/ 6954]\n",
      "loss: 0.592943  [ 2048/ 6954]\n",
      "loss: 0.591871  [ 3072/ 6954]\n",
      "loss: 0.619003  [ 4096/ 6954]\n",
      "loss: 0.609632  [ 5120/ 6954]\n",
      "loss: 0.593123  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.1%, Avg loss: 0.004796 \n",
      "\n",
      "Epoch 141\n",
      "-------------------------------\n",
      "loss: 0.600012  [    0/ 6954]\n",
      "loss: 0.596411  [ 1024/ 6954]\n",
      "loss: 0.594274  [ 2048/ 6954]\n",
      "loss: 0.594690  [ 3072/ 6954]\n",
      "loss: 0.592267  [ 4096/ 6954]\n",
      "loss: 0.604088  [ 5120/ 6954]\n",
      "loss: 0.597861  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.0%, Avg loss: 0.004792 \n",
      "\n",
      "Epoch 142\n",
      "-------------------------------\n",
      "loss: 0.608196  [    0/ 6954]\n",
      "loss: 0.598022  [ 1024/ 6954]\n",
      "loss: 0.596752  [ 2048/ 6954]\n",
      "loss: 0.609329  [ 3072/ 6954]\n",
      "loss: 0.602776  [ 4096/ 6954]\n",
      "loss: 0.618770  [ 5120/ 6954]\n",
      "loss: 0.607510  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.0%, Avg loss: 0.004789 \n",
      "\n",
      "Epoch 143\n",
      "-------------------------------\n",
      "loss: 0.604000  [    0/ 6954]\n",
      "loss: 0.615402  [ 1024/ 6954]\n",
      "loss: 0.628935  [ 2048/ 6954]\n",
      "loss: 0.615221  [ 3072/ 6954]\n",
      "loss: 0.616971  [ 4096/ 6954]\n",
      "loss: 0.592374  [ 5120/ 6954]\n",
      "loss: 0.587236  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.1%, Avg loss: 0.004785 \n",
      "\n",
      "Epoch 144\n",
      "-------------------------------\n",
      "loss: 0.591775  [    0/ 6954]\n",
      "loss: 0.599333  [ 1024/ 6954]\n",
      "loss: 0.592962  [ 2048/ 6954]\n",
      "loss: 0.599803  [ 3072/ 6954]\n",
      "loss: 0.595295  [ 4096/ 6954]\n",
      "loss: 0.604957  [ 5120/ 6954]\n",
      "loss: 0.626625  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.1%, Avg loss: 0.004781 \n",
      "\n",
      "Epoch 145\n",
      "-------------------------------\n",
      "loss: 0.611811  [    0/ 6954]\n",
      "loss: 0.611351  [ 1024/ 6954]\n",
      "loss: 0.610383  [ 2048/ 6954]\n",
      "loss: 0.601387  [ 3072/ 6954]\n",
      "loss: 0.586768  [ 4096/ 6954]\n",
      "loss: 0.574589  [ 5120/ 6954]\n",
      "loss: 0.605856  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.0%, Avg loss: 0.004778 \n",
      "\n",
      "Epoch 146\n",
      "-------------------------------\n",
      "loss: 0.596339  [    0/ 6954]\n",
      "loss: 0.598984  [ 1024/ 6954]\n",
      "loss: 0.585198  [ 2048/ 6954]\n",
      "loss: 0.610548  [ 3072/ 6954]\n",
      "loss: 0.589154  [ 4096/ 6954]\n",
      "loss: 0.590368  [ 5120/ 6954]\n",
      "loss: 0.582374  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.1%, Avg loss: 0.004774 \n",
      "\n",
      "Epoch 147\n",
      "-------------------------------\n",
      "loss: 0.595648  [    0/ 6954]\n",
      "loss: 0.607227  [ 1024/ 6954]\n",
      "loss: 0.601793  [ 2048/ 6954]\n",
      "loss: 0.611681  [ 3072/ 6954]\n",
      "loss: 0.605572  [ 4096/ 6954]\n",
      "loss: 0.605633  [ 5120/ 6954]\n",
      "loss: 0.581305  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 79.9%, Avg loss: 0.004770 \n",
      "\n",
      "Epoch 148\n",
      "-------------------------------\n",
      "loss: 0.596164  [    0/ 6954]\n",
      "loss: 0.583560  [ 1024/ 6954]\n",
      "loss: 0.602088  [ 2048/ 6954]\n",
      "loss: 0.622632  [ 3072/ 6954]\n",
      "loss: 0.607986  [ 4096/ 6954]\n",
      "loss: 0.600329  [ 5120/ 6954]\n",
      "loss: 0.584105  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 79.9%, Avg loss: 0.004767 \n",
      "\n",
      "Epoch 149\n",
      "-------------------------------\n",
      "loss: 0.594236  [    0/ 6954]\n",
      "loss: 0.583059  [ 1024/ 6954]\n",
      "loss: 0.615250  [ 2048/ 6954]\n",
      "loss: 0.589887  [ 3072/ 6954]\n",
      "loss: 0.606613  [ 4096/ 6954]\n",
      "loss: 0.583768  [ 5120/ 6954]\n",
      "loss: 0.600657  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.0%, Avg loss: 0.004763 \n",
      "\n",
      "Epoch 150\n",
      "-------------------------------\n",
      "loss: 0.623856  [    0/ 6954]\n",
      "loss: 0.596373  [ 1024/ 6954]\n",
      "loss: 0.594766  [ 2048/ 6954]\n",
      "loss: 0.611308  [ 3072/ 6954]\n",
      "loss: 0.599983  [ 4096/ 6954]\n",
      "loss: 0.600819  [ 5120/ 6954]\n",
      "loss: 0.600024  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.0%, Avg loss: 0.004759 \n",
      "\n",
      "Epoch 151\n",
      "-------------------------------\n",
      "loss: 0.597684  [    0/ 6954]\n",
      "loss: 0.597198  [ 1024/ 6954]\n",
      "loss: 0.586252  [ 2048/ 6954]\n",
      "loss: 0.602038  [ 3072/ 6954]\n",
      "loss: 0.586118  [ 4096/ 6954]\n",
      "loss: 0.613078  [ 5120/ 6954]\n",
      "loss: 0.584977  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.0%, Avg loss: 0.004756 \n",
      "\n",
      "Epoch 152\n",
      "-------------------------------\n",
      "loss: 0.582681  [    0/ 6954]\n",
      "loss: 0.603553  [ 1024/ 6954]\n",
      "loss: 0.604625  [ 2048/ 6954]\n",
      "loss: 0.592073  [ 3072/ 6954]\n",
      "loss: 0.607433  [ 4096/ 6954]\n",
      "loss: 0.613667  [ 5120/ 6954]\n",
      "loss: 0.585715  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.1%, Avg loss: 0.004752 \n",
      "\n",
      "Epoch 153\n",
      "-------------------------------\n",
      "loss: 0.579603  [    0/ 6954]\n",
      "loss: 0.594113  [ 1024/ 6954]\n",
      "loss: 0.602347  [ 2048/ 6954]\n",
      "loss: 0.590623  [ 3072/ 6954]\n",
      "loss: 0.582561  [ 4096/ 6954]\n",
      "loss: 0.604421  [ 5120/ 6954]\n",
      "loss: 0.594416  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.1%, Avg loss: 0.004749 \n",
      "\n",
      "Epoch 154\n",
      "-------------------------------\n",
      "loss: 0.563963  [    0/ 6954]\n",
      "loss: 0.578964  [ 1024/ 6954]\n",
      "loss: 0.593462  [ 2048/ 6954]\n",
      "loss: 0.599126  [ 3072/ 6954]\n",
      "loss: 0.574811  [ 4096/ 6954]\n",
      "loss: 0.589613  [ 5120/ 6954]\n",
      "loss: 0.593778  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.0%, Avg loss: 0.004745 \n",
      "\n",
      "Epoch 155\n",
      "-------------------------------\n",
      "loss: 0.620968  [    0/ 6954]\n",
      "loss: 0.593884  [ 1024/ 6954]\n",
      "loss: 0.591803  [ 2048/ 6954]\n",
      "loss: 0.579335  [ 3072/ 6954]\n",
      "loss: 0.587496  [ 4096/ 6954]\n",
      "loss: 0.622121  [ 5120/ 6954]\n",
      "loss: 0.578641  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.0%, Avg loss: 0.004742 \n",
      "\n",
      "Epoch 156\n",
      "-------------------------------\n",
      "loss: 0.588721  [    0/ 6954]\n",
      "loss: 0.589144  [ 1024/ 6954]\n",
      "loss: 0.585674  [ 2048/ 6954]\n",
      "loss: 0.599299  [ 3072/ 6954]\n",
      "loss: 0.585378  [ 4096/ 6954]\n",
      "loss: 0.609161  [ 5120/ 6954]\n",
      "loss: 0.592031  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.0%, Avg loss: 0.004739 \n",
      "\n",
      "Epoch 157\n",
      "-------------------------------\n",
      "loss: 0.593584  [    0/ 6954]\n",
      "loss: 0.609663  [ 1024/ 6954]\n",
      "loss: 0.591834  [ 2048/ 6954]\n",
      "loss: 0.590808  [ 3072/ 6954]\n",
      "loss: 0.592779  [ 4096/ 6954]\n",
      "loss: 0.599038  [ 5120/ 6954]\n",
      "loss: 0.618412  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.0%, Avg loss: 0.004735 \n",
      "\n",
      "Epoch 158\n",
      "-------------------------------\n",
      "loss: 0.613784  [    0/ 6954]\n",
      "loss: 0.593398  [ 1024/ 6954]\n",
      "loss: 0.628505  [ 2048/ 6954]\n",
      "loss: 0.600605  [ 3072/ 6954]\n",
      "loss: 0.583596  [ 4096/ 6954]\n",
      "loss: 0.596960  [ 5120/ 6954]\n",
      "loss: 0.577438  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.1%, Avg loss: 0.004732 \n",
      "\n",
      "Epoch 159\n",
      "-------------------------------\n",
      "loss: 0.586245  [    0/ 6954]\n",
      "loss: 0.589773  [ 1024/ 6954]\n",
      "loss: 0.589036  [ 2048/ 6954]\n",
      "loss: 0.603067  [ 3072/ 6954]\n",
      "loss: 0.571804  [ 4096/ 6954]\n",
      "loss: 0.579169  [ 5120/ 6954]\n",
      "loss: 0.591444  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.1%, Avg loss: 0.004728 \n",
      "\n",
      "Epoch 160\n",
      "-------------------------------\n",
      "loss: 0.584772  [    0/ 6954]\n",
      "loss: 0.588891  [ 1024/ 6954]\n",
      "loss: 0.580740  [ 2048/ 6954]\n",
      "loss: 0.608519  [ 3072/ 6954]\n",
      "loss: 0.608418  [ 4096/ 6954]\n",
      "loss: 0.581334  [ 5120/ 6954]\n",
      "loss: 0.601931  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.1%, Avg loss: 0.004725 \n",
      "\n",
      "Epoch 161\n",
      "-------------------------------\n",
      "loss: 0.575483  [    0/ 6954]\n",
      "loss: 0.592969  [ 1024/ 6954]\n",
      "loss: 0.589034  [ 2048/ 6954]\n",
      "loss: 0.565193  [ 3072/ 6954]\n",
      "loss: 0.606574  [ 4096/ 6954]\n",
      "loss: 0.556253  [ 5120/ 6954]\n",
      "loss: 0.585924  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.1%, Avg loss: 0.004722 \n",
      "\n",
      "Epoch 162\n",
      "-------------------------------\n",
      "loss: 0.588790  [    0/ 6954]\n",
      "loss: 0.606960  [ 1024/ 6954]\n",
      "loss: 0.603546  [ 2048/ 6954]\n",
      "loss: 0.582267  [ 3072/ 6954]\n",
      "loss: 0.590224  [ 4096/ 6954]\n",
      "loss: 0.573990  [ 5120/ 6954]\n",
      "loss: 0.593986  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.2%, Avg loss: 0.004718 \n",
      "\n",
      "Epoch 163\n",
      "-------------------------------\n",
      "loss: 0.573538  [    0/ 6954]\n",
      "loss: 0.590922  [ 1024/ 6954]\n",
      "loss: 0.607867  [ 2048/ 6954]\n",
      "loss: 0.592732  [ 3072/ 6954]\n",
      "loss: 0.585326  [ 4096/ 6954]\n",
      "loss: 0.619145  [ 5120/ 6954]\n",
      "loss: 0.583667  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.2%, Avg loss: 0.004715 \n",
      "\n",
      "Epoch 164\n",
      "-------------------------------\n",
      "loss: 0.566631  [    0/ 6954]\n",
      "loss: 0.597720  [ 1024/ 6954]\n",
      "loss: 0.610910  [ 2048/ 6954]\n",
      "loss: 0.568700  [ 3072/ 6954]\n",
      "loss: 0.606339  [ 4096/ 6954]\n",
      "loss: 0.591416  [ 5120/ 6954]\n",
      "loss: 0.602258  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.1%, Avg loss: 0.004712 \n",
      "\n",
      "Epoch 165\n",
      "-------------------------------\n",
      "loss: 0.598116  [    0/ 6954]\n",
      "loss: 0.581694  [ 1024/ 6954]\n",
      "loss: 0.585700  [ 2048/ 6954]\n",
      "loss: 0.592587  [ 3072/ 6954]\n",
      "loss: 0.611882  [ 4096/ 6954]\n",
      "loss: 0.598374  [ 5120/ 6954]\n",
      "loss: 0.587864  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.1%, Avg loss: 0.004708 \n",
      "\n",
      "Epoch 166\n",
      "-------------------------------\n",
      "loss: 0.568501  [    0/ 6954]\n",
      "loss: 0.593089  [ 1024/ 6954]\n",
      "loss: 0.606223  [ 2048/ 6954]\n",
      "loss: 0.599792  [ 3072/ 6954]\n",
      "loss: 0.602331  [ 4096/ 6954]\n",
      "loss: 0.618450  [ 5120/ 6954]\n",
      "loss: 0.597529  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.1%, Avg loss: 0.004705 \n",
      "\n",
      "Epoch 167\n",
      "-------------------------------\n",
      "loss: 0.633470  [    0/ 6954]\n",
      "loss: 0.589346  [ 1024/ 6954]\n",
      "loss: 0.585577  [ 2048/ 6954]\n",
      "loss: 0.557524  [ 3072/ 6954]\n",
      "loss: 0.619344  [ 4096/ 6954]\n",
      "loss: 0.599522  [ 5120/ 6954]\n",
      "loss: 0.596133  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.2%, Avg loss: 0.004702 \n",
      "\n",
      "Epoch 168\n",
      "-------------------------------\n",
      "loss: 0.587659  [    0/ 6954]\n",
      "loss: 0.613813  [ 1024/ 6954]\n",
      "loss: 0.587091  [ 2048/ 6954]\n",
      "loss: 0.589111  [ 3072/ 6954]\n",
      "loss: 0.606115  [ 4096/ 6954]\n",
      "loss: 0.607487  [ 5120/ 6954]\n",
      "loss: 0.621673  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.2%, Avg loss: 0.004699 \n",
      "\n",
      "Epoch 169\n",
      "-------------------------------\n",
      "loss: 0.567847  [    0/ 6954]\n",
      "loss: 0.601757  [ 1024/ 6954]\n",
      "loss: 0.596146  [ 2048/ 6954]\n",
      "loss: 0.587639  [ 3072/ 6954]\n",
      "loss: 0.599701  [ 4096/ 6954]\n",
      "loss: 0.602038  [ 5120/ 6954]\n",
      "loss: 0.605828  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.2%, Avg loss: 0.004695 \n",
      "\n",
      "Epoch 170\n",
      "-------------------------------\n",
      "loss: 0.572433  [    0/ 6954]\n",
      "loss: 0.603210  [ 1024/ 6954]\n",
      "loss: 0.573849  [ 2048/ 6954]\n",
      "loss: 0.583273  [ 3072/ 6954]\n",
      "loss: 0.611542  [ 4096/ 6954]\n",
      "loss: 0.581443  [ 5120/ 6954]\n",
      "loss: 0.617525  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.2%, Avg loss: 0.004692 \n",
      "\n",
      "Epoch 171\n",
      "-------------------------------\n",
      "loss: 0.578191  [    0/ 6954]\n",
      "loss: 0.569078  [ 1024/ 6954]\n",
      "loss: 0.615169  [ 2048/ 6954]\n",
      "loss: 0.607149  [ 3072/ 6954]\n",
      "loss: 0.602874  [ 4096/ 6954]\n",
      "loss: 0.575050  [ 5120/ 6954]\n",
      "loss: 0.575823  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.2%, Avg loss: 0.004689 \n",
      "\n",
      "Epoch 172\n",
      "-------------------------------\n",
      "loss: 0.588772  [    0/ 6954]\n",
      "loss: 0.598209  [ 1024/ 6954]\n",
      "loss: 0.589460  [ 2048/ 6954]\n",
      "loss: 0.588881  [ 3072/ 6954]\n",
      "loss: 0.592078  [ 4096/ 6954]\n",
      "loss: 0.602449  [ 5120/ 6954]\n",
      "loss: 0.587755  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.2%, Avg loss: 0.004686 \n",
      "\n",
      "Epoch 173\n",
      "-------------------------------\n",
      "loss: 0.600935  [    0/ 6954]\n",
      "loss: 0.594858  [ 1024/ 6954]\n",
      "loss: 0.600409  [ 2048/ 6954]\n",
      "loss: 0.567632  [ 3072/ 6954]\n",
      "loss: 0.572460  [ 4096/ 6954]\n",
      "loss: 0.582485  [ 5120/ 6954]\n",
      "loss: 0.587379  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.2%, Avg loss: 0.004683 \n",
      "\n",
      "Epoch 174\n",
      "-------------------------------\n",
      "loss: 0.615398  [    0/ 6954]\n",
      "loss: 0.580271  [ 1024/ 6954]\n",
      "loss: 0.594213  [ 2048/ 6954]\n",
      "loss: 0.574306  [ 3072/ 6954]\n",
      "loss: 0.582435  [ 4096/ 6954]\n",
      "loss: 0.577924  [ 5120/ 6954]\n",
      "loss: 0.618963  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.3%, Avg loss: 0.004680 \n",
      "\n",
      "Epoch 175\n",
      "-------------------------------\n",
      "loss: 0.609356  [    0/ 6954]\n",
      "loss: 0.607144  [ 1024/ 6954]\n",
      "loss: 0.589110  [ 2048/ 6954]\n",
      "loss: 0.577618  [ 3072/ 6954]\n",
      "loss: 0.581508  [ 4096/ 6954]\n",
      "loss: 0.600197  [ 5120/ 6954]\n",
      "loss: 0.578969  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.4%, Avg loss: 0.004677 \n",
      "\n",
      "Epoch 176\n",
      "-------------------------------\n",
      "loss: 0.608022  [    0/ 6954]\n",
      "loss: 0.588204  [ 1024/ 6954]\n",
      "loss: 0.601859  [ 2048/ 6954]\n",
      "loss: 0.596473  [ 3072/ 6954]\n",
      "loss: 0.593445  [ 4096/ 6954]\n",
      "loss: 0.599609  [ 5120/ 6954]\n",
      "loss: 0.587167  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.4%, Avg loss: 0.004673 \n",
      "\n",
      "Epoch 177\n",
      "-------------------------------\n",
      "loss: 0.595137  [    0/ 6954]\n",
      "loss: 0.580490  [ 1024/ 6954]\n",
      "loss: 0.622484  [ 2048/ 6954]\n",
      "loss: 0.580283  [ 3072/ 6954]\n",
      "loss: 0.583694  [ 4096/ 6954]\n",
      "loss: 0.620695  [ 5120/ 6954]\n",
      "loss: 0.598211  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.4%, Avg loss: 0.004670 \n",
      "\n",
      "Epoch 178\n",
      "-------------------------------\n",
      "loss: 0.591074  [    0/ 6954]\n",
      "loss: 0.559047  [ 1024/ 6954]\n",
      "loss: 0.588150  [ 2048/ 6954]\n",
      "loss: 0.588262  [ 3072/ 6954]\n",
      "loss: 0.599180  [ 4096/ 6954]\n",
      "loss: 0.570583  [ 5120/ 6954]\n",
      "loss: 0.582978  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.4%, Avg loss: 0.004667 \n",
      "\n",
      "Epoch 179\n",
      "-------------------------------\n",
      "loss: 0.599584  [    0/ 6954]\n",
      "loss: 0.580094  [ 1024/ 6954]\n",
      "loss: 0.610536  [ 2048/ 6954]\n",
      "loss: 0.578159  [ 3072/ 6954]\n",
      "loss: 0.598804  [ 4096/ 6954]\n",
      "loss: 0.562127  [ 5120/ 6954]\n",
      "loss: 0.623241  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.4%, Avg loss: 0.004664 \n",
      "\n",
      "Epoch 180\n",
      "-------------------------------\n",
      "loss: 0.604946  [    0/ 6954]\n",
      "loss: 0.602517  [ 1024/ 6954]\n",
      "loss: 0.582181  [ 2048/ 6954]\n",
      "loss: 0.573140  [ 3072/ 6954]\n",
      "loss: 0.595266  [ 4096/ 6954]\n",
      "loss: 0.589501  [ 5120/ 6954]\n",
      "loss: 0.576516  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.4%, Avg loss: 0.004661 \n",
      "\n",
      "Epoch 181\n",
      "-------------------------------\n",
      "loss: 0.580352  [    0/ 6954]\n",
      "loss: 0.588362  [ 1024/ 6954]\n",
      "loss: 0.588485  [ 2048/ 6954]\n",
      "loss: 0.586583  [ 3072/ 6954]\n",
      "loss: 0.576855  [ 4096/ 6954]\n",
      "loss: 0.596538  [ 5120/ 6954]\n",
      "loss: 0.614699  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.4%, Avg loss: 0.004658 \n",
      "\n",
      "Epoch 182\n",
      "-------------------------------\n",
      "loss: 0.595747  [    0/ 6954]\n",
      "loss: 0.578775  [ 1024/ 6954]\n",
      "loss: 0.605707  [ 2048/ 6954]\n",
      "loss: 0.568059  [ 3072/ 6954]\n",
      "loss: 0.585344  [ 4096/ 6954]\n",
      "loss: 0.591278  [ 5120/ 6954]\n",
      "loss: 0.599399  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.4%, Avg loss: 0.004655 \n",
      "\n",
      "Epoch 183\n",
      "-------------------------------\n",
      "loss: 0.583283  [    0/ 6954]\n",
      "loss: 0.587200  [ 1024/ 6954]\n",
      "loss: 0.579920  [ 2048/ 6954]\n",
      "loss: 0.556422  [ 3072/ 6954]\n",
      "loss: 0.601680  [ 4096/ 6954]\n",
      "loss: 0.581765  [ 5120/ 6954]\n",
      "loss: 0.570074  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.4%, Avg loss: 0.004652 \n",
      "\n",
      "Epoch 184\n",
      "-------------------------------\n",
      "loss: 0.598716  [    0/ 6954]\n",
      "loss: 0.614673  [ 1024/ 6954]\n",
      "loss: 0.588562  [ 2048/ 6954]\n",
      "loss: 0.577573  [ 3072/ 6954]\n",
      "loss: 0.592376  [ 4096/ 6954]\n",
      "loss: 0.563545  [ 5120/ 6954]\n",
      "loss: 0.598149  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.4%, Avg loss: 0.004649 \n",
      "\n",
      "Epoch 185\n",
      "-------------------------------\n",
      "loss: 0.561567  [    0/ 6954]\n",
      "loss: 0.602296  [ 1024/ 6954]\n",
      "loss: 0.589967  [ 2048/ 6954]\n",
      "loss: 0.610269  [ 3072/ 6954]\n",
      "loss: 0.578150  [ 4096/ 6954]\n",
      "loss: 0.568504  [ 5120/ 6954]\n",
      "loss: 0.617054  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.4%, Avg loss: 0.004646 \n",
      "\n",
      "Epoch 186\n",
      "-------------------------------\n",
      "loss: 0.607809  [    0/ 6954]\n",
      "loss: 0.584882  [ 1024/ 6954]\n",
      "loss: 0.568176  [ 2048/ 6954]\n",
      "loss: 0.557339  [ 3072/ 6954]\n",
      "loss: 0.611085  [ 4096/ 6954]\n",
      "loss: 0.602102  [ 5120/ 6954]\n",
      "loss: 0.579259  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.4%, Avg loss: 0.004643 \n",
      "\n",
      "Epoch 187\n",
      "-------------------------------\n",
      "loss: 0.590568  [    0/ 6954]\n",
      "loss: 0.595592  [ 1024/ 6954]\n",
      "loss: 0.587618  [ 2048/ 6954]\n",
      "loss: 0.572333  [ 3072/ 6954]\n",
      "loss: 0.601400  [ 4096/ 6954]\n",
      "loss: 0.596886  [ 5120/ 6954]\n",
      "loss: 0.603480  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.4%, Avg loss: 0.004640 \n",
      "\n",
      "Epoch 188\n",
      "-------------------------------\n",
      "loss: 0.582228  [    0/ 6954]\n",
      "loss: 0.589399  [ 1024/ 6954]\n",
      "loss: 0.610986  [ 2048/ 6954]\n",
      "loss: 0.594656  [ 3072/ 6954]\n",
      "loss: 0.594172  [ 4096/ 6954]\n",
      "loss: 0.574573  [ 5120/ 6954]\n",
      "loss: 0.562334  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.4%, Avg loss: 0.004637 \n",
      "\n",
      "Epoch 189\n",
      "-------------------------------\n",
      "loss: 0.604589  [    0/ 6954]\n",
      "loss: 0.600366  [ 1024/ 6954]\n",
      "loss: 0.606873  [ 2048/ 6954]\n",
      "loss: 0.593555  [ 3072/ 6954]\n",
      "loss: 0.602490  [ 4096/ 6954]\n",
      "loss: 0.582448  [ 5120/ 6954]\n",
      "loss: 0.583567  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.4%, Avg loss: 0.004634 \n",
      "\n",
      "Epoch 190\n",
      "-------------------------------\n",
      "loss: 0.600705  [    0/ 6954]\n",
      "loss: 0.589650  [ 1024/ 6954]\n",
      "loss: 0.584424  [ 2048/ 6954]\n",
      "loss: 0.583071  [ 3072/ 6954]\n",
      "loss: 0.571121  [ 4096/ 6954]\n",
      "loss: 0.579495  [ 5120/ 6954]\n",
      "loss: 0.590962  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.4%, Avg loss: 0.004631 \n",
      "\n",
      "Epoch 191\n",
      "-------------------------------\n",
      "loss: 0.597965  [    0/ 6954]\n",
      "loss: 0.570246  [ 1024/ 6954]\n",
      "loss: 0.590151  [ 2048/ 6954]\n",
      "loss: 0.608797  [ 3072/ 6954]\n",
      "loss: 0.587136  [ 4096/ 6954]\n",
      "loss: 0.584017  [ 5120/ 6954]\n",
      "loss: 0.569541  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.4%, Avg loss: 0.004628 \n",
      "\n",
      "Epoch 192\n",
      "-------------------------------\n",
      "loss: 0.570752  [    0/ 6954]\n",
      "loss: 0.589191  [ 1024/ 6954]\n",
      "loss: 0.571049  [ 2048/ 6954]\n",
      "loss: 0.577270  [ 3072/ 6954]\n",
      "loss: 0.597216  [ 4096/ 6954]\n",
      "loss: 0.581636  [ 5120/ 6954]\n",
      "loss: 0.604352  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.4%, Avg loss: 0.004625 \n",
      "\n",
      "Epoch 193\n",
      "-------------------------------\n",
      "loss: 0.573166  [    0/ 6954]\n",
      "loss: 0.606277  [ 1024/ 6954]\n",
      "loss: 0.563611  [ 2048/ 6954]\n",
      "loss: 0.585914  [ 3072/ 6954]\n",
      "loss: 0.599844  [ 4096/ 6954]\n",
      "loss: 0.603185  [ 5120/ 6954]\n",
      "loss: 0.564890  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.4%, Avg loss: 0.004623 \n",
      "\n",
      "Epoch 194\n",
      "-------------------------------\n",
      "loss: 0.609682  [    0/ 6954]\n",
      "loss: 0.583114  [ 1024/ 6954]\n",
      "loss: 0.569727  [ 2048/ 6954]\n",
      "loss: 0.571398  [ 3072/ 6954]\n",
      "loss: 0.573379  [ 4096/ 6954]\n",
      "loss: 0.586885  [ 5120/ 6954]\n",
      "loss: 0.588835  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.4%, Avg loss: 0.004620 \n",
      "\n",
      "Epoch 195\n",
      "-------------------------------\n",
      "loss: 0.557528  [    0/ 6954]\n",
      "loss: 0.592209  [ 1024/ 6954]\n",
      "loss: 0.582806  [ 2048/ 6954]\n",
      "loss: 0.558252  [ 3072/ 6954]\n",
      "loss: 0.565287  [ 4096/ 6954]\n",
      "loss: 0.584284  [ 5120/ 6954]\n",
      "loss: 0.564245  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.5%, Avg loss: 0.004617 \n",
      "\n",
      "Epoch 196\n",
      "-------------------------------\n",
      "loss: 0.584899  [    0/ 6954]\n",
      "loss: 0.551910  [ 1024/ 6954]\n",
      "loss: 0.585030  [ 2048/ 6954]\n",
      "loss: 0.579991  [ 3072/ 6954]\n",
      "loss: 0.579278  [ 4096/ 6954]\n",
      "loss: 0.558571  [ 5120/ 6954]\n",
      "loss: 0.569726  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.6%, Avg loss: 0.004614 \n",
      "\n",
      "Epoch 197\n",
      "-------------------------------\n",
      "loss: 0.577094  [    0/ 6954]\n",
      "loss: 0.573147  [ 1024/ 6954]\n",
      "loss: 0.573175  [ 2048/ 6954]\n",
      "loss: 0.573626  [ 3072/ 6954]\n",
      "loss: 0.588547  [ 4096/ 6954]\n",
      "loss: 0.570989  [ 5120/ 6954]\n",
      "loss: 0.587619  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.4%, Avg loss: 0.004611 \n",
      "\n",
      "Epoch 198\n",
      "-------------------------------\n",
      "loss: 0.571168  [    0/ 6954]\n",
      "loss: 0.596978  [ 1024/ 6954]\n",
      "loss: 0.575693  [ 2048/ 6954]\n",
      "loss: 0.562549  [ 3072/ 6954]\n",
      "loss: 0.569308  [ 4096/ 6954]\n",
      "loss: 0.563889  [ 5120/ 6954]\n",
      "loss: 0.583944  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.4%, Avg loss: 0.004609 \n",
      "\n",
      "Epoch 199\n",
      "-------------------------------\n",
      "loss: 0.555573  [    0/ 6954]\n",
      "loss: 0.587141  [ 1024/ 6954]\n",
      "loss: 0.585484  [ 2048/ 6954]\n",
      "loss: 0.580208  [ 3072/ 6954]\n",
      "loss: 0.599602  [ 4096/ 6954]\n",
      "loss: 0.555035  [ 5120/ 6954]\n",
      "loss: 0.583230  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.4%, Avg loss: 0.004606 \n",
      "\n",
      "Epoch 200\n",
      "-------------------------------\n",
      "loss: 0.576990  [    0/ 6954]\n",
      "loss: 0.589135  [ 1024/ 6954]\n",
      "loss: 0.577693  [ 2048/ 6954]\n",
      "loss: 0.560012  [ 3072/ 6954]\n",
      "loss: 0.585707  [ 4096/ 6954]\n",
      "loss: 0.568459  [ 5120/ 6954]\n",
      "loss: 0.561540  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.4%, Avg loss: 0.004603 \n",
      "\n",
      "Epoch 201\n",
      "-------------------------------\n",
      "loss: 0.566324  [    0/ 6954]\n",
      "loss: 0.601654  [ 1024/ 6954]\n",
      "loss: 0.578130  [ 2048/ 6954]\n",
      "loss: 0.601472  [ 3072/ 6954]\n",
      "loss: 0.572264  [ 4096/ 6954]\n",
      "loss: 0.563397  [ 5120/ 6954]\n",
      "loss: 0.578817  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.5%, Avg loss: 0.004601 \n",
      "\n",
      "Epoch 202\n",
      "-------------------------------\n",
      "loss: 0.586363  [    0/ 6954]\n",
      "loss: 0.588941  [ 1024/ 6954]\n",
      "loss: 0.583079  [ 2048/ 6954]\n",
      "loss: 0.597064  [ 3072/ 6954]\n",
      "loss: 0.594514  [ 4096/ 6954]\n",
      "loss: 0.589400  [ 5120/ 6954]\n",
      "loss: 0.578086  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.6%, Avg loss: 0.004598 \n",
      "\n",
      "Epoch 203\n",
      "-------------------------------\n",
      "loss: 0.595228  [    0/ 6954]\n",
      "loss: 0.546773  [ 1024/ 6954]\n",
      "loss: 0.573453  [ 2048/ 6954]\n",
      "loss: 0.586877  [ 3072/ 6954]\n",
      "loss: 0.607766  [ 4096/ 6954]\n",
      "loss: 0.556664  [ 5120/ 6954]\n",
      "loss: 0.589561  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.6%, Avg loss: 0.004595 \n",
      "\n",
      "Epoch 204\n",
      "-------------------------------\n",
      "loss: 0.562870  [    0/ 6954]\n",
      "loss: 0.567977  [ 1024/ 6954]\n",
      "loss: 0.568396  [ 2048/ 6954]\n",
      "loss: 0.563050  [ 3072/ 6954]\n",
      "loss: 0.592018  [ 4096/ 6954]\n",
      "loss: 0.573106  [ 5120/ 6954]\n",
      "loss: 0.564834  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.6%, Avg loss: 0.004593 \n",
      "\n",
      "Epoch 205\n",
      "-------------------------------\n",
      "loss: 0.570245  [    0/ 6954]\n",
      "loss: 0.584689  [ 1024/ 6954]\n",
      "loss: 0.587632  [ 2048/ 6954]\n",
      "loss: 0.563002  [ 3072/ 6954]\n",
      "loss: 0.568023  [ 4096/ 6954]\n",
      "loss: 0.590551  [ 5120/ 6954]\n",
      "loss: 0.555376  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.6%, Avg loss: 0.004590 \n",
      "\n",
      "Epoch 206\n",
      "-------------------------------\n",
      "loss: 0.559243  [    0/ 6954]\n",
      "loss: 0.580585  [ 1024/ 6954]\n",
      "loss: 0.599177  [ 2048/ 6954]\n",
      "loss: 0.569079  [ 3072/ 6954]\n",
      "loss: 0.596403  [ 4096/ 6954]\n",
      "loss: 0.584576  [ 5120/ 6954]\n",
      "loss: 0.583400  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.6%, Avg loss: 0.004587 \n",
      "\n",
      "Epoch 207\n",
      "-------------------------------\n",
      "loss: 0.602506  [    0/ 6954]\n",
      "loss: 0.595657  [ 1024/ 6954]\n",
      "loss: 0.617264  [ 2048/ 6954]\n",
      "loss: 0.579461  [ 3072/ 6954]\n",
      "loss: 0.567618  [ 4096/ 6954]\n",
      "loss: 0.572991  [ 5120/ 6954]\n",
      "loss: 0.554838  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.6%, Avg loss: 0.004585 \n",
      "\n",
      "Epoch 208\n",
      "-------------------------------\n",
      "loss: 0.580674  [    0/ 6954]\n",
      "loss: 0.577553  [ 1024/ 6954]\n",
      "loss: 0.581970  [ 2048/ 6954]\n",
      "loss: 0.588198  [ 3072/ 6954]\n",
      "loss: 0.568690  [ 4096/ 6954]\n",
      "loss: 0.588432  [ 5120/ 6954]\n",
      "loss: 0.592476  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.6%, Avg loss: 0.004582 \n",
      "\n",
      "Epoch 209\n",
      "-------------------------------\n",
      "loss: 0.562622  [    0/ 6954]\n",
      "loss: 0.577064  [ 1024/ 6954]\n",
      "loss: 0.576134  [ 2048/ 6954]\n",
      "loss: 0.586223  [ 3072/ 6954]\n",
      "loss: 0.592032  [ 4096/ 6954]\n",
      "loss: 0.573596  [ 5120/ 6954]\n",
      "loss: 0.575021  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.6%, Avg loss: 0.004580 \n",
      "\n",
      "Epoch 210\n",
      "-------------------------------\n",
      "loss: 0.562146  [    0/ 6954]\n",
      "loss: 0.577059  [ 1024/ 6954]\n",
      "loss: 0.549983  [ 2048/ 6954]\n",
      "loss: 0.575409  [ 3072/ 6954]\n",
      "loss: 0.592495  [ 4096/ 6954]\n",
      "loss: 0.589224  [ 5120/ 6954]\n",
      "loss: 0.573317  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.6%, Avg loss: 0.004577 \n",
      "\n",
      "Epoch 211\n",
      "-------------------------------\n",
      "loss: 0.571902  [    0/ 6954]\n",
      "loss: 0.596232  [ 1024/ 6954]\n",
      "loss: 0.587763  [ 2048/ 6954]\n",
      "loss: 0.570716  [ 3072/ 6954]\n",
      "loss: 0.610823  [ 4096/ 6954]\n",
      "loss: 0.575777  [ 5120/ 6954]\n",
      "loss: 0.558169  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.6%, Avg loss: 0.004574 \n",
      "\n",
      "Epoch 212\n",
      "-------------------------------\n",
      "loss: 0.582013  [    0/ 6954]\n",
      "loss: 0.570986  [ 1024/ 6954]\n",
      "loss: 0.565430  [ 2048/ 6954]\n",
      "loss: 0.616336  [ 3072/ 6954]\n",
      "loss: 0.569317  [ 4096/ 6954]\n",
      "loss: 0.581221  [ 5120/ 6954]\n",
      "loss: 0.572845  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.6%, Avg loss: 0.004572 \n",
      "\n",
      "Epoch 213\n",
      "-------------------------------\n",
      "loss: 0.554893  [    0/ 6954]\n",
      "loss: 0.578242  [ 1024/ 6954]\n",
      "loss: 0.557863  [ 2048/ 6954]\n",
      "loss: 0.585361  [ 3072/ 6954]\n",
      "loss: 0.579281  [ 4096/ 6954]\n",
      "loss: 0.572803  [ 5120/ 6954]\n",
      "loss: 0.563315  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.6%, Avg loss: 0.004569 \n",
      "\n",
      "Epoch 214\n",
      "-------------------------------\n",
      "loss: 0.566344  [    0/ 6954]\n",
      "loss: 0.570350  [ 1024/ 6954]\n",
      "loss: 0.587121  [ 2048/ 6954]\n",
      "loss: 0.568104  [ 3072/ 6954]\n",
      "loss: 0.553549  [ 4096/ 6954]\n",
      "loss: 0.580980  [ 5120/ 6954]\n",
      "loss: 0.582665  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.6%, Avg loss: 0.004567 \n",
      "\n",
      "Epoch 215\n",
      "-------------------------------\n",
      "loss: 0.586939  [    0/ 6954]\n",
      "loss: 0.553990  [ 1024/ 6954]\n",
      "loss: 0.565676  [ 2048/ 6954]\n",
      "loss: 0.580567  [ 3072/ 6954]\n",
      "loss: 0.558357  [ 4096/ 6954]\n",
      "loss: 0.583233  [ 5120/ 6954]\n",
      "loss: 0.543290  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.6%, Avg loss: 0.004564 \n",
      "\n",
      "Epoch 216\n",
      "-------------------------------\n",
      "loss: 0.539583  [    0/ 6954]\n",
      "loss: 0.552720  [ 1024/ 6954]\n",
      "loss: 0.599005  [ 2048/ 6954]\n",
      "loss: 0.606613  [ 3072/ 6954]\n",
      "loss: 0.557580  [ 4096/ 6954]\n",
      "loss: 0.569300  [ 5120/ 6954]\n",
      "loss: 0.592241  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.6%, Avg loss: 0.004562 \n",
      "\n",
      "Epoch 217\n",
      "-------------------------------\n",
      "loss: 0.577334  [    0/ 6954]\n",
      "loss: 0.568909  [ 1024/ 6954]\n",
      "loss: 0.559270  [ 2048/ 6954]\n",
      "loss: 0.571592  [ 3072/ 6954]\n",
      "loss: 0.574179  [ 4096/ 6954]\n",
      "loss: 0.584037  [ 5120/ 6954]\n",
      "loss: 0.575123  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.6%, Avg loss: 0.004559 \n",
      "\n",
      "Epoch 218\n",
      "-------------------------------\n",
      "loss: 0.556224  [    0/ 6954]\n",
      "loss: 0.584448  [ 1024/ 6954]\n",
      "loss: 0.568327  [ 2048/ 6954]\n",
      "loss: 0.582090  [ 3072/ 6954]\n",
      "loss: 0.577840  [ 4096/ 6954]\n",
      "loss: 0.572260  [ 5120/ 6954]\n",
      "loss: 0.548965  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.6%, Avg loss: 0.004557 \n",
      "\n",
      "Epoch 219\n",
      "-------------------------------\n",
      "loss: 0.561688  [    0/ 6954]\n",
      "loss: 0.593449  [ 1024/ 6954]\n",
      "loss: 0.555495  [ 2048/ 6954]\n",
      "loss: 0.582894  [ 3072/ 6954]\n",
      "loss: 0.582138  [ 4096/ 6954]\n",
      "loss: 0.577541  [ 5120/ 6954]\n",
      "loss: 0.580482  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.6%, Avg loss: 0.004554 \n",
      "\n",
      "Epoch 220\n",
      "-------------------------------\n",
      "loss: 0.601658  [    0/ 6954]\n",
      "loss: 0.583133  [ 1024/ 6954]\n",
      "loss: 0.590867  [ 2048/ 6954]\n",
      "loss: 0.553744  [ 3072/ 6954]\n",
      "loss: 0.578405  [ 4096/ 6954]\n",
      "loss: 0.557329  [ 5120/ 6954]\n",
      "loss: 0.572871  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004552 \n",
      "\n",
      "Epoch 221\n",
      "-------------------------------\n",
      "loss: 0.564124  [    0/ 6954]\n",
      "loss: 0.585308  [ 1024/ 6954]\n",
      "loss: 0.570081  [ 2048/ 6954]\n",
      "loss: 0.590009  [ 3072/ 6954]\n",
      "loss: 0.536463  [ 4096/ 6954]\n",
      "loss: 0.606816  [ 5120/ 6954]\n",
      "loss: 0.580010  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004550 \n",
      "\n",
      "Epoch 222\n",
      "-------------------------------\n",
      "loss: 0.569728  [    0/ 6954]\n",
      "loss: 0.560589  [ 1024/ 6954]\n",
      "loss: 0.568387  [ 2048/ 6954]\n",
      "loss: 0.589734  [ 3072/ 6954]\n",
      "loss: 0.571711  [ 4096/ 6954]\n",
      "loss: 0.565972  [ 5120/ 6954]\n",
      "loss: 0.567055  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.6%, Avg loss: 0.004547 \n",
      "\n",
      "Epoch 223\n",
      "-------------------------------\n",
      "loss: 0.582310  [    0/ 6954]\n",
      "loss: 0.553952  [ 1024/ 6954]\n",
      "loss: 0.572107  [ 2048/ 6954]\n",
      "loss: 0.578340  [ 3072/ 6954]\n",
      "loss: 0.580617  [ 4096/ 6954]\n",
      "loss: 0.586565  [ 5120/ 6954]\n",
      "loss: 0.539535  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.6%, Avg loss: 0.004545 \n",
      "\n",
      "Epoch 224\n",
      "-------------------------------\n",
      "loss: 0.580257  [    0/ 6954]\n",
      "loss: 0.562061  [ 1024/ 6954]\n",
      "loss: 0.576403  [ 2048/ 6954]\n",
      "loss: 0.553408  [ 3072/ 6954]\n",
      "loss: 0.571823  [ 4096/ 6954]\n",
      "loss: 0.549092  [ 5120/ 6954]\n",
      "loss: 0.584884  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.6%, Avg loss: 0.004542 \n",
      "\n",
      "Epoch 225\n",
      "-------------------------------\n",
      "loss: 0.592112  [    0/ 6954]\n",
      "loss: 0.623180  [ 1024/ 6954]\n",
      "loss: 0.549289  [ 2048/ 6954]\n",
      "loss: 0.562205  [ 3072/ 6954]\n",
      "loss: 0.574793  [ 4096/ 6954]\n",
      "loss: 0.601594  [ 5120/ 6954]\n",
      "loss: 0.550704  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.6%, Avg loss: 0.004540 \n",
      "\n",
      "Epoch 226\n",
      "-------------------------------\n",
      "loss: 0.583716  [    0/ 6954]\n",
      "loss: 0.602725  [ 1024/ 6954]\n",
      "loss: 0.583854  [ 2048/ 6954]\n",
      "loss: 0.594536  [ 3072/ 6954]\n",
      "loss: 0.565096  [ 4096/ 6954]\n",
      "loss: 0.551486  [ 5120/ 6954]\n",
      "loss: 0.604712  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.6%, Avg loss: 0.004538 \n",
      "\n",
      "Epoch 227\n",
      "-------------------------------\n",
      "loss: 0.588272  [    0/ 6954]\n",
      "loss: 0.571214  [ 1024/ 6954]\n",
      "loss: 0.580102  [ 2048/ 6954]\n",
      "loss: 0.572885  [ 3072/ 6954]\n",
      "loss: 0.544198  [ 4096/ 6954]\n",
      "loss: 0.542070  [ 5120/ 6954]\n",
      "loss: 0.580564  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.6%, Avg loss: 0.004536 \n",
      "\n",
      "Epoch 228\n",
      "-------------------------------\n",
      "loss: 0.583873  [    0/ 6954]\n",
      "loss: 0.589004  [ 1024/ 6954]\n",
      "loss: 0.576061  [ 2048/ 6954]\n",
      "loss: 0.561535  [ 3072/ 6954]\n",
      "loss: 0.553286  [ 4096/ 6954]\n",
      "loss: 0.556853  [ 5120/ 6954]\n",
      "loss: 0.581326  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.6%, Avg loss: 0.004533 \n",
      "\n",
      "Epoch 229\n",
      "-------------------------------\n",
      "loss: 0.559532  [    0/ 6954]\n",
      "loss: 0.593424  [ 1024/ 6954]\n",
      "loss: 0.590665  [ 2048/ 6954]\n",
      "loss: 0.559988  [ 3072/ 6954]\n",
      "loss: 0.566880  [ 4096/ 6954]\n",
      "loss: 0.593181  [ 5120/ 6954]\n",
      "loss: 0.601248  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.6%, Avg loss: 0.004531 \n",
      "\n",
      "Epoch 230\n",
      "-------------------------------\n",
      "loss: 0.577746  [    0/ 6954]\n",
      "loss: 0.577393  [ 1024/ 6954]\n",
      "loss: 0.568584  [ 2048/ 6954]\n",
      "loss: 0.593279  [ 3072/ 6954]\n",
      "loss: 0.551578  [ 4096/ 6954]\n",
      "loss: 0.570856  [ 5120/ 6954]\n",
      "loss: 0.589293  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.6%, Avg loss: 0.004529 \n",
      "\n",
      "Epoch 231\n",
      "-------------------------------\n",
      "loss: 0.582609  [    0/ 6954]\n",
      "loss: 0.560466  [ 1024/ 6954]\n",
      "loss: 0.610402  [ 2048/ 6954]\n",
      "loss: 0.539695  [ 3072/ 6954]\n",
      "loss: 0.546150  [ 4096/ 6954]\n",
      "loss: 0.572755  [ 5120/ 6954]\n",
      "loss: 0.567770  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.6%, Avg loss: 0.004527 \n",
      "\n",
      "Epoch 232\n",
      "-------------------------------\n",
      "loss: 0.556367  [    0/ 6954]\n",
      "loss: 0.552748  [ 1024/ 6954]\n",
      "loss: 0.568540  [ 2048/ 6954]\n",
      "loss: 0.576305  [ 3072/ 6954]\n",
      "loss: 0.558874  [ 4096/ 6954]\n",
      "loss: 0.560687  [ 5120/ 6954]\n",
      "loss: 0.587206  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.6%, Avg loss: 0.004524 \n",
      "\n",
      "Epoch 233\n",
      "-------------------------------\n",
      "loss: 0.576800  [    0/ 6954]\n",
      "loss: 0.559718  [ 1024/ 6954]\n",
      "loss: 0.595081  [ 2048/ 6954]\n",
      "loss: 0.540109  [ 3072/ 6954]\n",
      "loss: 0.509417  [ 4096/ 6954]\n",
      "loss: 0.560732  [ 5120/ 6954]\n",
      "loss: 0.577142  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.6%, Avg loss: 0.004522 \n",
      "\n",
      "Epoch 234\n",
      "-------------------------------\n",
      "loss: 0.567872  [    0/ 6954]\n",
      "loss: 0.564276  [ 1024/ 6954]\n",
      "loss: 0.538365  [ 2048/ 6954]\n",
      "loss: 0.535816  [ 3072/ 6954]\n",
      "loss: 0.601415  [ 4096/ 6954]\n",
      "loss: 0.570979  [ 5120/ 6954]\n",
      "loss: 0.598926  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004520 \n",
      "\n",
      "Epoch 235\n",
      "-------------------------------\n",
      "loss: 0.554964  [    0/ 6954]\n",
      "loss: 0.575552  [ 1024/ 6954]\n",
      "loss: 0.606189  [ 2048/ 6954]\n",
      "loss: 0.595742  [ 3072/ 6954]\n",
      "loss: 0.517570  [ 4096/ 6954]\n",
      "loss: 0.591667  [ 5120/ 6954]\n",
      "loss: 0.558518  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.6%, Avg loss: 0.004518 \n",
      "\n",
      "Epoch 236\n",
      "-------------------------------\n",
      "loss: 0.566765  [    0/ 6954]\n",
      "loss: 0.567758  [ 1024/ 6954]\n",
      "loss: 0.576516  [ 2048/ 6954]\n",
      "loss: 0.605100  [ 3072/ 6954]\n",
      "loss: 0.580299  [ 4096/ 6954]\n",
      "loss: 0.588182  [ 5120/ 6954]\n",
      "loss: 0.591162  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.6%, Avg loss: 0.004516 \n",
      "\n",
      "Epoch 237\n",
      "-------------------------------\n",
      "loss: 0.559901  [    0/ 6954]\n",
      "loss: 0.555319  [ 1024/ 6954]\n",
      "loss: 0.589310  [ 2048/ 6954]\n",
      "loss: 0.563386  [ 3072/ 6954]\n",
      "loss: 0.584542  [ 4096/ 6954]\n",
      "loss: 0.594689  [ 5120/ 6954]\n",
      "loss: 0.544005  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.6%, Avg loss: 0.004514 \n",
      "\n",
      "Epoch 238\n",
      "-------------------------------\n",
      "loss: 0.556423  [    0/ 6954]\n",
      "loss: 0.573974  [ 1024/ 6954]\n",
      "loss: 0.582673  [ 2048/ 6954]\n",
      "loss: 0.563275  [ 3072/ 6954]\n",
      "loss: 0.592640  [ 4096/ 6954]\n",
      "loss: 0.594415  [ 5120/ 6954]\n",
      "loss: 0.588412  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004511 \n",
      "\n",
      "Epoch 239\n",
      "-------------------------------\n",
      "loss: 0.573082  [    0/ 6954]\n",
      "loss: 0.591912  [ 1024/ 6954]\n",
      "loss: 0.570927  [ 2048/ 6954]\n",
      "loss: 0.570493  [ 3072/ 6954]\n",
      "loss: 0.585840  [ 4096/ 6954]\n",
      "loss: 0.579700  [ 5120/ 6954]\n",
      "loss: 0.589137  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004509 \n",
      "\n",
      "Epoch 240\n",
      "-------------------------------\n",
      "loss: 0.590581  [    0/ 6954]\n",
      "loss: 0.540142  [ 1024/ 6954]\n",
      "loss: 0.568999  [ 2048/ 6954]\n",
      "loss: 0.573247  [ 3072/ 6954]\n",
      "loss: 0.566235  [ 4096/ 6954]\n",
      "loss: 0.551712  [ 5120/ 6954]\n",
      "loss: 0.555743  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004507 \n",
      "\n",
      "Epoch 241\n",
      "-------------------------------\n",
      "loss: 0.566702  [    0/ 6954]\n",
      "loss: 0.551835  [ 1024/ 6954]\n",
      "loss: 0.569119  [ 2048/ 6954]\n",
      "loss: 0.578530  [ 3072/ 6954]\n",
      "loss: 0.565890  [ 4096/ 6954]\n",
      "loss: 0.576589  [ 5120/ 6954]\n",
      "loss: 0.572496  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004505 \n",
      "\n",
      "Epoch 242\n",
      "-------------------------------\n",
      "loss: 0.596112  [    0/ 6954]\n",
      "loss: 0.592002  [ 1024/ 6954]\n",
      "loss: 0.579327  [ 2048/ 6954]\n",
      "loss: 0.571912  [ 3072/ 6954]\n",
      "loss: 0.595298  [ 4096/ 6954]\n",
      "loss: 0.571497  [ 5120/ 6954]\n",
      "loss: 0.558136  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.6%, Avg loss: 0.004503 \n",
      "\n",
      "Epoch 243\n",
      "-------------------------------\n",
      "loss: 0.592172  [    0/ 6954]\n",
      "loss: 0.550919  [ 1024/ 6954]\n",
      "loss: 0.564764  [ 2048/ 6954]\n",
      "loss: 0.582888  [ 3072/ 6954]\n",
      "loss: 0.575901  [ 4096/ 6954]\n",
      "loss: 0.559786  [ 5120/ 6954]\n",
      "loss: 0.578801  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.6%, Avg loss: 0.004501 \n",
      "\n",
      "Epoch 244\n",
      "-------------------------------\n",
      "loss: 0.553389  [    0/ 6954]\n",
      "loss: 0.575454  [ 1024/ 6954]\n",
      "loss: 0.565637  [ 2048/ 6954]\n",
      "loss: 0.575283  [ 3072/ 6954]\n",
      "loss: 0.581985  [ 4096/ 6954]\n",
      "loss: 0.544794  [ 5120/ 6954]\n",
      "loss: 0.565038  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004499 \n",
      "\n",
      "Epoch 245\n",
      "-------------------------------\n",
      "loss: 0.545819  [    0/ 6954]\n",
      "loss: 0.553392  [ 1024/ 6954]\n",
      "loss: 0.570469  [ 2048/ 6954]\n",
      "loss: 0.553558  [ 3072/ 6954]\n",
      "loss: 0.612082  [ 4096/ 6954]\n",
      "loss: 0.558856  [ 5120/ 6954]\n",
      "loss: 0.567689  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004497 \n",
      "\n",
      "Epoch 246\n",
      "-------------------------------\n",
      "loss: 0.567503  [    0/ 6954]\n",
      "loss: 0.586670  [ 1024/ 6954]\n",
      "loss: 0.573160  [ 2048/ 6954]\n",
      "loss: 0.575599  [ 3072/ 6954]\n",
      "loss: 0.587329  [ 4096/ 6954]\n",
      "loss: 0.563861  [ 5120/ 6954]\n",
      "loss: 0.560261  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004495 \n",
      "\n",
      "Epoch 247\n",
      "-------------------------------\n",
      "loss: 0.549901  [    0/ 6954]\n",
      "loss: 0.554730  [ 1024/ 6954]\n",
      "loss: 0.547566  [ 2048/ 6954]\n",
      "loss: 0.577382  [ 3072/ 6954]\n",
      "loss: 0.564222  [ 4096/ 6954]\n",
      "loss: 0.575239  [ 5120/ 6954]\n",
      "loss: 0.554867  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004493 \n",
      "\n",
      "Epoch 248\n",
      "-------------------------------\n",
      "loss: 0.552221  [    0/ 6954]\n",
      "loss: 0.575507  [ 1024/ 6954]\n",
      "loss: 0.589404  [ 2048/ 6954]\n",
      "loss: 0.552258  [ 3072/ 6954]\n",
      "loss: 0.531150  [ 4096/ 6954]\n",
      "loss: 0.569795  [ 5120/ 6954]\n",
      "loss: 0.556794  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004491 \n",
      "\n",
      "Epoch 249\n",
      "-------------------------------\n",
      "loss: 0.580939  [    0/ 6954]\n",
      "loss: 0.546854  [ 1024/ 6954]\n",
      "loss: 0.574594  [ 2048/ 6954]\n",
      "loss: 0.574774  [ 3072/ 6954]\n",
      "loss: 0.528612  [ 4096/ 6954]\n",
      "loss: 0.548990  [ 5120/ 6954]\n",
      "loss: 0.588550  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004489 \n",
      "\n",
      "Epoch 250\n",
      "-------------------------------\n",
      "loss: 0.562565  [    0/ 6954]\n",
      "loss: 0.594696  [ 1024/ 6954]\n",
      "loss: 0.534286  [ 2048/ 6954]\n",
      "loss: 0.590620  [ 3072/ 6954]\n",
      "loss: 0.556050  [ 4096/ 6954]\n",
      "loss: 0.551641  [ 5120/ 6954]\n",
      "loss: 0.612630  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004487 \n",
      "\n",
      "Epoch 251\n",
      "-------------------------------\n",
      "loss: 0.568476  [    0/ 6954]\n",
      "loss: 0.557481  [ 1024/ 6954]\n",
      "loss: 0.599840  [ 2048/ 6954]\n",
      "loss: 0.529659  [ 3072/ 6954]\n",
      "loss: 0.577507  [ 4096/ 6954]\n",
      "loss: 0.598285  [ 5120/ 6954]\n",
      "loss: 0.578443  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004485 \n",
      "\n",
      "Epoch 252\n",
      "-------------------------------\n",
      "loss: 0.617461  [    0/ 6954]\n",
      "loss: 0.567608  [ 1024/ 6954]\n",
      "loss: 0.570584  [ 2048/ 6954]\n",
      "loss: 0.587080  [ 3072/ 6954]\n",
      "loss: 0.555860  [ 4096/ 6954]\n",
      "loss: 0.566098  [ 5120/ 6954]\n",
      "loss: 0.569652  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004483 \n",
      "\n",
      "Epoch 253\n",
      "-------------------------------\n",
      "loss: 0.548002  [    0/ 6954]\n",
      "loss: 0.560521  [ 1024/ 6954]\n",
      "loss: 0.557202  [ 2048/ 6954]\n",
      "loss: 0.562034  [ 3072/ 6954]\n",
      "loss: 0.555499  [ 4096/ 6954]\n",
      "loss: 0.572746  [ 5120/ 6954]\n",
      "loss: 0.598259  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004481 \n",
      "\n",
      "Epoch 254\n",
      "-------------------------------\n",
      "loss: 0.573159  [    0/ 6954]\n",
      "loss: 0.563446  [ 1024/ 6954]\n",
      "loss: 0.555208  [ 2048/ 6954]\n",
      "loss: 0.566559  [ 3072/ 6954]\n",
      "loss: 0.548655  [ 4096/ 6954]\n",
      "loss: 0.541463  [ 5120/ 6954]\n",
      "loss: 0.549114  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004480 \n",
      "\n",
      "Epoch 255\n",
      "-------------------------------\n",
      "loss: 0.595018  [    0/ 6954]\n",
      "loss: 0.576289  [ 1024/ 6954]\n",
      "loss: 0.584677  [ 2048/ 6954]\n",
      "loss: 0.595029  [ 3072/ 6954]\n",
      "loss: 0.577182  [ 4096/ 6954]\n",
      "loss: 0.582665  [ 5120/ 6954]\n",
      "loss: 0.591293  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004478 \n",
      "\n",
      "Epoch 256\n",
      "-------------------------------\n",
      "loss: 0.569559  [    0/ 6954]\n",
      "loss: 0.528698  [ 1024/ 6954]\n",
      "loss: 0.558953  [ 2048/ 6954]\n",
      "loss: 0.539887  [ 3072/ 6954]\n",
      "loss: 0.583575  [ 4096/ 6954]\n",
      "loss: 0.608793  [ 5120/ 6954]\n",
      "loss: 0.547917  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004476 \n",
      "\n",
      "Epoch 257\n",
      "-------------------------------\n",
      "loss: 0.559186  [    0/ 6954]\n",
      "loss: 0.591594  [ 1024/ 6954]\n",
      "loss: 0.581304  [ 2048/ 6954]\n",
      "loss: 0.573342  [ 3072/ 6954]\n",
      "loss: 0.584619  [ 4096/ 6954]\n",
      "loss: 0.557532  [ 5120/ 6954]\n",
      "loss: 0.570819  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004474 \n",
      "\n",
      "Epoch 258\n",
      "-------------------------------\n",
      "loss: 0.558589  [    0/ 6954]\n",
      "loss: 0.580414  [ 1024/ 6954]\n",
      "loss: 0.553829  [ 2048/ 6954]\n",
      "loss: 0.573573  [ 3072/ 6954]\n",
      "loss: 0.575776  [ 4096/ 6954]\n",
      "loss: 0.575547  [ 5120/ 6954]\n",
      "loss: 0.586676  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004472 \n",
      "\n",
      "Epoch 259\n",
      "-------------------------------\n",
      "loss: 0.573135  [    0/ 6954]\n",
      "loss: 0.558618  [ 1024/ 6954]\n",
      "loss: 0.535383  [ 2048/ 6954]\n",
      "loss: 0.573619  [ 3072/ 6954]\n",
      "loss: 0.534346  [ 4096/ 6954]\n",
      "loss: 0.598263  [ 5120/ 6954]\n",
      "loss: 0.561573  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004470 \n",
      "\n",
      "Epoch 260\n",
      "-------------------------------\n",
      "loss: 0.534269  [    0/ 6954]\n",
      "loss: 0.541391  [ 1024/ 6954]\n",
      "loss: 0.599191  [ 2048/ 6954]\n",
      "loss: 0.517309  [ 3072/ 6954]\n",
      "loss: 0.588618  [ 4096/ 6954]\n",
      "loss: 0.585017  [ 5120/ 6954]\n",
      "loss: 0.562428  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004468 \n",
      "\n",
      "Epoch 261\n",
      "-------------------------------\n",
      "loss: 0.584023  [    0/ 6954]\n",
      "loss: 0.530793  [ 1024/ 6954]\n",
      "loss: 0.556175  [ 2048/ 6954]\n",
      "loss: 0.570849  [ 3072/ 6954]\n",
      "loss: 0.584923  [ 4096/ 6954]\n",
      "loss: 0.576507  [ 5120/ 6954]\n",
      "loss: 0.559120  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004467 \n",
      "\n",
      "Epoch 262\n",
      "-------------------------------\n",
      "loss: 0.556570  [    0/ 6954]\n",
      "loss: 0.566568  [ 1024/ 6954]\n",
      "loss: 0.571525  [ 2048/ 6954]\n",
      "loss: 0.584319  [ 3072/ 6954]\n",
      "loss: 0.567477  [ 4096/ 6954]\n",
      "loss: 0.544507  [ 5120/ 6954]\n",
      "loss: 0.526854  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004465 \n",
      "\n",
      "Epoch 263\n",
      "-------------------------------\n",
      "loss: 0.588144  [    0/ 6954]\n",
      "loss: 0.559686  [ 1024/ 6954]\n",
      "loss: 0.550332  [ 2048/ 6954]\n",
      "loss: 0.542569  [ 3072/ 6954]\n",
      "loss: 0.557392  [ 4096/ 6954]\n",
      "loss: 0.585390  [ 5120/ 6954]\n",
      "loss: 0.574045  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004463 \n",
      "\n",
      "Epoch 264\n",
      "-------------------------------\n",
      "loss: 0.571572  [    0/ 6954]\n",
      "loss: 0.563275  [ 1024/ 6954]\n",
      "loss: 0.564640  [ 2048/ 6954]\n",
      "loss: 0.520721  [ 3072/ 6954]\n",
      "loss: 0.565137  [ 4096/ 6954]\n",
      "loss: 0.553012  [ 5120/ 6954]\n",
      "loss: 0.578621  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004461 \n",
      "\n",
      "Epoch 265\n",
      "-------------------------------\n",
      "loss: 0.567439  [    0/ 6954]\n",
      "loss: 0.616870  [ 1024/ 6954]\n",
      "loss: 0.572003  [ 2048/ 6954]\n",
      "loss: 0.579103  [ 3072/ 6954]\n",
      "loss: 0.552896  [ 4096/ 6954]\n",
      "loss: 0.580172  [ 5120/ 6954]\n",
      "loss: 0.574252  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004460 \n",
      "\n",
      "Epoch 266\n",
      "-------------------------------\n",
      "loss: 0.580264  [    0/ 6954]\n",
      "loss: 0.574064  [ 1024/ 6954]\n",
      "loss: 0.598572  [ 2048/ 6954]\n",
      "loss: 0.537810  [ 3072/ 6954]\n",
      "loss: 0.550718  [ 4096/ 6954]\n",
      "loss: 0.548350  [ 5120/ 6954]\n",
      "loss: 0.549814  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004458 \n",
      "\n",
      "Epoch 267\n",
      "-------------------------------\n",
      "loss: 0.576478  [    0/ 6954]\n",
      "loss: 0.550586  [ 1024/ 6954]\n",
      "loss: 0.549277  [ 2048/ 6954]\n",
      "loss: 0.567524  [ 3072/ 6954]\n",
      "loss: 0.591838  [ 4096/ 6954]\n",
      "loss: 0.539013  [ 5120/ 6954]\n",
      "loss: 0.598511  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004456 \n",
      "\n",
      "Epoch 268\n",
      "-------------------------------\n",
      "loss: 0.566800  [    0/ 6954]\n",
      "loss: 0.563989  [ 1024/ 6954]\n",
      "loss: 0.585710  [ 2048/ 6954]\n",
      "loss: 0.589331  [ 3072/ 6954]\n",
      "loss: 0.576946  [ 4096/ 6954]\n",
      "loss: 0.591843  [ 5120/ 6954]\n",
      "loss: 0.538838  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004454 \n",
      "\n",
      "Epoch 269\n",
      "-------------------------------\n",
      "loss: 0.558897  [    0/ 6954]\n",
      "loss: 0.571572  [ 1024/ 6954]\n",
      "loss: 0.527895  [ 2048/ 6954]\n",
      "loss: 0.579050  [ 3072/ 6954]\n",
      "loss: 0.572681  [ 4096/ 6954]\n",
      "loss: 0.560609  [ 5120/ 6954]\n",
      "loss: 0.573740  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004453 \n",
      "\n",
      "Epoch 270\n",
      "-------------------------------\n",
      "loss: 0.547538  [    0/ 6954]\n",
      "loss: 0.576204  [ 1024/ 6954]\n",
      "loss: 0.558233  [ 2048/ 6954]\n",
      "loss: 0.577813  [ 3072/ 6954]\n",
      "loss: 0.532544  [ 4096/ 6954]\n",
      "loss: 0.589406  [ 5120/ 6954]\n",
      "loss: 0.560299  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004451 \n",
      "\n",
      "Epoch 271\n",
      "-------------------------------\n",
      "loss: 0.580965  [    0/ 6954]\n",
      "loss: 0.589625  [ 1024/ 6954]\n",
      "loss: 0.578569  [ 2048/ 6954]\n",
      "loss: 0.559833  [ 3072/ 6954]\n",
      "loss: 0.558827  [ 4096/ 6954]\n",
      "loss: 0.558452  [ 5120/ 6954]\n",
      "loss: 0.566343  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004449 \n",
      "\n",
      "Epoch 272\n",
      "-------------------------------\n",
      "loss: 0.546982  [    0/ 6954]\n",
      "loss: 0.559895  [ 1024/ 6954]\n",
      "loss: 0.553266  [ 2048/ 6954]\n",
      "loss: 0.563874  [ 3072/ 6954]\n",
      "loss: 0.559315  [ 4096/ 6954]\n",
      "loss: 0.588335  [ 5120/ 6954]\n",
      "loss: 0.542955  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004448 \n",
      "\n",
      "Epoch 273\n",
      "-------------------------------\n",
      "loss: 0.549802  [    0/ 6954]\n",
      "loss: 0.562413  [ 1024/ 6954]\n",
      "loss: 0.579814  [ 2048/ 6954]\n",
      "loss: 0.586402  [ 3072/ 6954]\n",
      "loss: 0.600378  [ 4096/ 6954]\n",
      "loss: 0.559736  [ 5120/ 6954]\n",
      "loss: 0.559884  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004446 \n",
      "\n",
      "Epoch 274\n",
      "-------------------------------\n",
      "loss: 0.571081  [    0/ 6954]\n",
      "loss: 0.575354  [ 1024/ 6954]\n",
      "loss: 0.545384  [ 2048/ 6954]\n",
      "loss: 0.554184  [ 3072/ 6954]\n",
      "loss: 0.563926  [ 4096/ 6954]\n",
      "loss: 0.553143  [ 5120/ 6954]\n",
      "loss: 0.571437  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004444 \n",
      "\n",
      "Epoch 275\n",
      "-------------------------------\n",
      "loss: 0.568915  [    0/ 6954]\n",
      "loss: 0.597707  [ 1024/ 6954]\n",
      "loss: 0.547308  [ 2048/ 6954]\n",
      "loss: 0.578284  [ 3072/ 6954]\n",
      "loss: 0.533452  [ 4096/ 6954]\n",
      "loss: 0.585804  [ 5120/ 6954]\n",
      "loss: 0.565627  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004443 \n",
      "\n",
      "Epoch 276\n",
      "-------------------------------\n",
      "loss: 0.548234  [    0/ 6954]\n",
      "loss: 0.585342  [ 1024/ 6954]\n",
      "loss: 0.558608  [ 2048/ 6954]\n",
      "loss: 0.607691  [ 3072/ 6954]\n",
      "loss: 0.539470  [ 4096/ 6954]\n",
      "loss: 0.520665  [ 5120/ 6954]\n",
      "loss: 0.582785  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004441 \n",
      "\n",
      "Epoch 277\n",
      "-------------------------------\n",
      "loss: 0.510830  [    0/ 6954]\n",
      "loss: 0.554185  [ 1024/ 6954]\n",
      "loss: 0.546687  [ 2048/ 6954]\n",
      "loss: 0.611561  [ 3072/ 6954]\n",
      "loss: 0.568318  [ 4096/ 6954]\n",
      "loss: 0.571272  [ 5120/ 6954]\n",
      "loss: 0.564945  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004440 \n",
      "\n",
      "Epoch 278\n",
      "-------------------------------\n",
      "loss: 0.559644  [    0/ 6954]\n",
      "loss: 0.543735  [ 1024/ 6954]\n",
      "loss: 0.555486  [ 2048/ 6954]\n",
      "loss: 0.600930  [ 3072/ 6954]\n",
      "loss: 0.563395  [ 4096/ 6954]\n",
      "loss: 0.551666  [ 5120/ 6954]\n",
      "loss: 0.580366  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004438 \n",
      "\n",
      "Epoch 279\n",
      "-------------------------------\n",
      "loss: 0.559607  [    0/ 6954]\n",
      "loss: 0.534498  [ 1024/ 6954]\n",
      "loss: 0.568789  [ 2048/ 6954]\n",
      "loss: 0.528402  [ 3072/ 6954]\n",
      "loss: 0.530897  [ 4096/ 6954]\n",
      "loss: 0.569962  [ 5120/ 6954]\n",
      "loss: 0.557466  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004436 \n",
      "\n",
      "Epoch 280\n",
      "-------------------------------\n",
      "loss: 0.575168  [    0/ 6954]\n",
      "loss: 0.555617  [ 1024/ 6954]\n",
      "loss: 0.559385  [ 2048/ 6954]\n",
      "loss: 0.565860  [ 3072/ 6954]\n",
      "loss: 0.560874  [ 4096/ 6954]\n",
      "loss: 0.606049  [ 5120/ 6954]\n",
      "loss: 0.581048  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004435 \n",
      "\n",
      "Epoch 281\n",
      "-------------------------------\n",
      "loss: 0.545465  [    0/ 6954]\n",
      "loss: 0.548676  [ 1024/ 6954]\n",
      "loss: 0.552171  [ 2048/ 6954]\n",
      "loss: 0.539805  [ 3072/ 6954]\n",
      "loss: 0.598852  [ 4096/ 6954]\n",
      "loss: 0.595447  [ 5120/ 6954]\n",
      "loss: 0.573658  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004433 \n",
      "\n",
      "Epoch 282\n",
      "-------------------------------\n",
      "loss: 0.577206  [    0/ 6954]\n",
      "loss: 0.573540  [ 1024/ 6954]\n",
      "loss: 0.566554  [ 2048/ 6954]\n",
      "loss: 0.561361  [ 3072/ 6954]\n",
      "loss: 0.571244  [ 4096/ 6954]\n",
      "loss: 0.549347  [ 5120/ 6954]\n",
      "loss: 0.532871  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004432 \n",
      "\n",
      "Epoch 283\n",
      "-------------------------------\n",
      "loss: 0.534769  [    0/ 6954]\n",
      "loss: 0.571546  [ 1024/ 6954]\n",
      "loss: 0.585362  [ 2048/ 6954]\n",
      "loss: 0.555169  [ 3072/ 6954]\n",
      "loss: 0.580449  [ 4096/ 6954]\n",
      "loss: 0.565741  [ 5120/ 6954]\n",
      "loss: 0.591024  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004430 \n",
      "\n",
      "Epoch 284\n",
      "-------------------------------\n",
      "loss: 0.574821  [    0/ 6954]\n",
      "loss: 0.549637  [ 1024/ 6954]\n",
      "loss: 0.560064  [ 2048/ 6954]\n",
      "loss: 0.586646  [ 3072/ 6954]\n",
      "loss: 0.553838  [ 4096/ 6954]\n",
      "loss: 0.572085  [ 5120/ 6954]\n",
      "loss: 0.536468  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004429 \n",
      "\n",
      "Epoch 285\n",
      "-------------------------------\n",
      "loss: 0.582402  [    0/ 6954]\n",
      "loss: 0.539204  [ 1024/ 6954]\n",
      "loss: 0.546265  [ 2048/ 6954]\n",
      "loss: 0.573651  [ 3072/ 6954]\n",
      "loss: 0.573831  [ 4096/ 6954]\n",
      "loss: 0.553458  [ 5120/ 6954]\n",
      "loss: 0.521936  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004427 \n",
      "\n",
      "Epoch 286\n",
      "-------------------------------\n",
      "loss: 0.549278  [    0/ 6954]\n",
      "loss: 0.586011  [ 1024/ 6954]\n",
      "loss: 0.554715  [ 2048/ 6954]\n",
      "loss: 0.567197  [ 3072/ 6954]\n",
      "loss: 0.515993  [ 4096/ 6954]\n",
      "loss: 0.585397  [ 5120/ 6954]\n",
      "loss: 0.561847  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004425 \n",
      "\n",
      "Epoch 287\n",
      "-------------------------------\n",
      "loss: 0.550767  [    0/ 6954]\n",
      "loss: 0.594242  [ 1024/ 6954]\n",
      "loss: 0.581140  [ 2048/ 6954]\n",
      "loss: 0.579201  [ 3072/ 6954]\n",
      "loss: 0.560582  [ 4096/ 6954]\n",
      "loss: 0.557624  [ 5120/ 6954]\n",
      "loss: 0.572891  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004424 \n",
      "\n",
      "Epoch 288\n",
      "-------------------------------\n",
      "loss: 0.558301  [    0/ 6954]\n",
      "loss: 0.545060  [ 1024/ 6954]\n",
      "loss: 0.552790  [ 2048/ 6954]\n",
      "loss: 0.581393  [ 3072/ 6954]\n",
      "loss: 0.587525  [ 4096/ 6954]\n",
      "loss: 0.588710  [ 5120/ 6954]\n",
      "loss: 0.538073  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004423 \n",
      "\n",
      "Epoch 289\n",
      "-------------------------------\n",
      "loss: 0.524831  [    0/ 6954]\n",
      "loss: 0.590732  [ 1024/ 6954]\n",
      "loss: 0.544470  [ 2048/ 6954]\n",
      "loss: 0.524731  [ 3072/ 6954]\n",
      "loss: 0.535671  [ 4096/ 6954]\n",
      "loss: 0.538483  [ 5120/ 6954]\n",
      "loss: 0.581862  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004421 \n",
      "\n",
      "Epoch 290\n",
      "-------------------------------\n",
      "loss: 0.558288  [    0/ 6954]\n",
      "loss: 0.567324  [ 1024/ 6954]\n",
      "loss: 0.550096  [ 2048/ 6954]\n",
      "loss: 0.562109  [ 3072/ 6954]\n",
      "loss: 0.581189  [ 4096/ 6954]\n",
      "loss: 0.541960  [ 5120/ 6954]\n",
      "loss: 0.557814  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004419 \n",
      "\n",
      "Epoch 291\n",
      "-------------------------------\n",
      "loss: 0.575404  [    0/ 6954]\n",
      "loss: 0.562266  [ 1024/ 6954]\n",
      "loss: 0.556062  [ 2048/ 6954]\n",
      "loss: 0.531109  [ 3072/ 6954]\n",
      "loss: 0.556077  [ 4096/ 6954]\n",
      "loss: 0.591931  [ 5120/ 6954]\n",
      "loss: 0.574984  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004418 \n",
      "\n",
      "Epoch 292\n",
      "-------------------------------\n",
      "loss: 0.583251  [    0/ 6954]\n",
      "loss: 0.595012  [ 1024/ 6954]\n",
      "loss: 0.566217  [ 2048/ 6954]\n",
      "loss: 0.553324  [ 3072/ 6954]\n",
      "loss: 0.554078  [ 4096/ 6954]\n",
      "loss: 0.569540  [ 5120/ 6954]\n",
      "loss: 0.544433  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004417 \n",
      "\n",
      "Epoch 293\n",
      "-------------------------------\n",
      "loss: 0.565078  [    0/ 6954]\n",
      "loss: 0.560732  [ 1024/ 6954]\n",
      "loss: 0.521295  [ 2048/ 6954]\n",
      "loss: 0.593417  [ 3072/ 6954]\n",
      "loss: 0.528295  [ 4096/ 6954]\n",
      "loss: 0.587907  [ 5120/ 6954]\n",
      "loss: 0.544757  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004415 \n",
      "\n",
      "Epoch 294\n",
      "-------------------------------\n",
      "loss: 0.596736  [    0/ 6954]\n",
      "loss: 0.566530  [ 1024/ 6954]\n",
      "loss: 0.569658  [ 2048/ 6954]\n",
      "loss: 0.554114  [ 3072/ 6954]\n",
      "loss: 0.540408  [ 4096/ 6954]\n",
      "loss: 0.552898  [ 5120/ 6954]\n",
      "loss: 0.548424  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004414 \n",
      "\n",
      "Epoch 295\n",
      "-------------------------------\n",
      "loss: 0.555013  [    0/ 6954]\n",
      "loss: 0.554940  [ 1024/ 6954]\n",
      "loss: 0.532328  [ 2048/ 6954]\n",
      "loss: 0.567451  [ 3072/ 6954]\n",
      "loss: 0.574060  [ 4096/ 6954]\n",
      "loss: 0.575704  [ 5120/ 6954]\n",
      "loss: 0.538664  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004412 \n",
      "\n",
      "Epoch 296\n",
      "-------------------------------\n",
      "loss: 0.563124  [    0/ 6954]\n",
      "loss: 0.540765  [ 1024/ 6954]\n",
      "loss: 0.518765  [ 2048/ 6954]\n",
      "loss: 0.561013  [ 3072/ 6954]\n",
      "loss: 0.553761  [ 4096/ 6954]\n",
      "loss: 0.552019  [ 5120/ 6954]\n",
      "loss: 0.558530  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004411 \n",
      "\n",
      "Epoch 297\n",
      "-------------------------------\n",
      "loss: 0.567696  [    0/ 6954]\n",
      "loss: 0.560626  [ 1024/ 6954]\n",
      "loss: 0.578200  [ 2048/ 6954]\n",
      "loss: 0.520810  [ 3072/ 6954]\n",
      "loss: 0.562112  [ 4096/ 6954]\n",
      "loss: 0.579217  [ 5120/ 6954]\n",
      "loss: 0.561143  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004409 \n",
      "\n",
      "Epoch 298\n",
      "-------------------------------\n",
      "loss: 0.597391  [    0/ 6954]\n",
      "loss: 0.551143  [ 1024/ 6954]\n",
      "loss: 0.556258  [ 2048/ 6954]\n",
      "loss: 0.569595  [ 3072/ 6954]\n",
      "loss: 0.536731  [ 4096/ 6954]\n",
      "loss: 0.543986  [ 5120/ 6954]\n",
      "loss: 0.547640  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004408 \n",
      "\n",
      "Epoch 299\n",
      "-------------------------------\n",
      "loss: 0.550747  [    0/ 6954]\n",
      "loss: 0.556060  [ 1024/ 6954]\n",
      "loss: 0.537007  [ 2048/ 6954]\n",
      "loss: 0.580747  [ 3072/ 6954]\n",
      "loss: 0.556972  [ 4096/ 6954]\n",
      "loss: 0.546287  [ 5120/ 6954]\n",
      "loss: 0.550236  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004407 \n",
      "\n",
      "Epoch 300\n",
      "-------------------------------\n",
      "loss: 0.564896  [    0/ 6954]\n",
      "loss: 0.564004  [ 1024/ 6954]\n",
      "loss: 0.550204  [ 2048/ 6954]\n",
      "loss: 0.541165  [ 3072/ 6954]\n",
      "loss: 0.570622  [ 4096/ 6954]\n",
      "loss: 0.571563  [ 5120/ 6954]\n",
      "loss: 0.544709  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004405 \n",
      "\n",
      "Epoch 301\n",
      "-------------------------------\n",
      "loss: 0.544871  [    0/ 6954]\n",
      "loss: 0.561950  [ 1024/ 6954]\n",
      "loss: 0.583613  [ 2048/ 6954]\n",
      "loss: 0.546334  [ 3072/ 6954]\n",
      "loss: 0.596917  [ 4096/ 6954]\n",
      "loss: 0.572803  [ 5120/ 6954]\n",
      "loss: 0.547799  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004404 \n",
      "\n",
      "Epoch 302\n",
      "-------------------------------\n",
      "loss: 0.587314  [    0/ 6954]\n",
      "loss: 0.578447  [ 1024/ 6954]\n",
      "loss: 0.552249  [ 2048/ 6954]\n",
      "loss: 0.569848  [ 3072/ 6954]\n",
      "loss: 0.552700  [ 4096/ 6954]\n",
      "loss: 0.570815  [ 5120/ 6954]\n",
      "loss: 0.534038  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004402 \n",
      "\n",
      "Epoch 303\n",
      "-------------------------------\n",
      "loss: 0.552766  [    0/ 6954]\n",
      "loss: 0.561956  [ 1024/ 6954]\n",
      "loss: 0.574001  [ 2048/ 6954]\n",
      "loss: 0.549214  [ 3072/ 6954]\n",
      "loss: 0.578293  [ 4096/ 6954]\n",
      "loss: 0.565319  [ 5120/ 6954]\n",
      "loss: 0.547727  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004401 \n",
      "\n",
      "Epoch 304\n",
      "-------------------------------\n",
      "loss: 0.587094  [    0/ 6954]\n",
      "loss: 0.534133  [ 1024/ 6954]\n",
      "loss: 0.557758  [ 2048/ 6954]\n",
      "loss: 0.562935  [ 3072/ 6954]\n",
      "loss: 0.548457  [ 4096/ 6954]\n",
      "loss: 0.537234  [ 5120/ 6954]\n",
      "loss: 0.552099  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004400 \n",
      "\n",
      "Epoch 305\n",
      "-------------------------------\n",
      "loss: 0.536461  [    0/ 6954]\n",
      "loss: 0.540182  [ 1024/ 6954]\n",
      "loss: 0.578257  [ 2048/ 6954]\n",
      "loss: 0.551482  [ 3072/ 6954]\n",
      "loss: 0.563448  [ 4096/ 6954]\n",
      "loss: 0.563867  [ 5120/ 6954]\n",
      "loss: 0.541569  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004398 \n",
      "\n",
      "Epoch 306\n",
      "-------------------------------\n",
      "loss: 0.536365  [    0/ 6954]\n",
      "loss: 0.583816  [ 1024/ 6954]\n",
      "loss: 0.578202  [ 2048/ 6954]\n",
      "loss: 0.577623  [ 3072/ 6954]\n",
      "loss: 0.580140  [ 4096/ 6954]\n",
      "loss: 0.583487  [ 5120/ 6954]\n",
      "loss: 0.519628  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004397 \n",
      "\n",
      "Epoch 307\n",
      "-------------------------------\n",
      "loss: 0.570558  [    0/ 6954]\n",
      "loss: 0.537807  [ 1024/ 6954]\n",
      "loss: 0.504987  [ 2048/ 6954]\n",
      "loss: 0.564384  [ 3072/ 6954]\n",
      "loss: 0.575632  [ 4096/ 6954]\n",
      "loss: 0.543498  [ 5120/ 6954]\n",
      "loss: 0.574334  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004396 \n",
      "\n",
      "Epoch 308\n",
      "-------------------------------\n",
      "loss: 0.546950  [    0/ 6954]\n",
      "loss: 0.536720  [ 1024/ 6954]\n",
      "loss: 0.534307  [ 2048/ 6954]\n",
      "loss: 0.563614  [ 3072/ 6954]\n",
      "loss: 0.564899  [ 4096/ 6954]\n",
      "loss: 0.542522  [ 5120/ 6954]\n",
      "loss: 0.571890  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004394 \n",
      "\n",
      "Epoch 309\n",
      "-------------------------------\n",
      "loss: 0.546365  [    0/ 6954]\n",
      "loss: 0.573581  [ 1024/ 6954]\n",
      "loss: 0.537742  [ 2048/ 6954]\n",
      "loss: 0.600544  [ 3072/ 6954]\n",
      "loss: 0.522481  [ 4096/ 6954]\n",
      "loss: 0.554829  [ 5120/ 6954]\n",
      "loss: 0.528549  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004393 \n",
      "\n",
      "Epoch 310\n",
      "-------------------------------\n",
      "loss: 0.573523  [    0/ 6954]\n",
      "loss: 0.533681  [ 1024/ 6954]\n",
      "loss: 0.530505  [ 2048/ 6954]\n",
      "loss: 0.542590  [ 3072/ 6954]\n",
      "loss: 0.533154  [ 4096/ 6954]\n",
      "loss: 0.605694  [ 5120/ 6954]\n",
      "loss: 0.565649  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004392 \n",
      "\n",
      "Epoch 311\n",
      "-------------------------------\n",
      "loss: 0.593007  [    0/ 6954]\n",
      "loss: 0.561739  [ 1024/ 6954]\n",
      "loss: 0.563043  [ 2048/ 6954]\n",
      "loss: 0.574374  [ 3072/ 6954]\n",
      "loss: 0.546013  [ 4096/ 6954]\n",
      "loss: 0.552906  [ 5120/ 6954]\n",
      "loss: 0.542861  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004390 \n",
      "\n",
      "Epoch 312\n",
      "-------------------------------\n",
      "loss: 0.568631  [    0/ 6954]\n",
      "loss: 0.548745  [ 1024/ 6954]\n",
      "loss: 0.541670  [ 2048/ 6954]\n",
      "loss: 0.561921  [ 3072/ 6954]\n",
      "loss: 0.539487  [ 4096/ 6954]\n",
      "loss: 0.529375  [ 5120/ 6954]\n",
      "loss: 0.530171  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004389 \n",
      "\n",
      "Epoch 313\n",
      "-------------------------------\n",
      "loss: 0.567318  [    0/ 6954]\n",
      "loss: 0.569408  [ 1024/ 6954]\n",
      "loss: 0.571966  [ 2048/ 6954]\n",
      "loss: 0.546588  [ 3072/ 6954]\n",
      "loss: 0.570041  [ 4096/ 6954]\n",
      "loss: 0.513652  [ 5120/ 6954]\n",
      "loss: 0.566355  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004388 \n",
      "\n",
      "Epoch 314\n",
      "-------------------------------\n",
      "loss: 0.594423  [    0/ 6954]\n",
      "loss: 0.561154  [ 1024/ 6954]\n",
      "loss: 0.562706  [ 2048/ 6954]\n",
      "loss: 0.540394  [ 3072/ 6954]\n",
      "loss: 0.535980  [ 4096/ 6954]\n",
      "loss: 0.548872  [ 5120/ 6954]\n",
      "loss: 0.582889  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004387 \n",
      "\n",
      "Epoch 315\n",
      "-------------------------------\n",
      "loss: 0.529776  [    0/ 6954]\n",
      "loss: 0.550163  [ 1024/ 6954]\n",
      "loss: 0.551159  [ 2048/ 6954]\n",
      "loss: 0.529811  [ 3072/ 6954]\n",
      "loss: 0.566460  [ 4096/ 6954]\n",
      "loss: 0.567200  [ 5120/ 6954]\n",
      "loss: 0.562128  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004385 \n",
      "\n",
      "Epoch 316\n",
      "-------------------------------\n",
      "loss: 0.501845  [    0/ 6954]\n",
      "loss: 0.542994  [ 1024/ 6954]\n",
      "loss: 0.555908  [ 2048/ 6954]\n",
      "loss: 0.577904  [ 3072/ 6954]\n",
      "loss: 0.586258  [ 4096/ 6954]\n",
      "loss: 0.534155  [ 5120/ 6954]\n",
      "loss: 0.577143  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004384 \n",
      "\n",
      "Epoch 317\n",
      "-------------------------------\n",
      "loss: 0.547511  [    0/ 6954]\n",
      "loss: 0.588802  [ 1024/ 6954]\n",
      "loss: 0.551045  [ 2048/ 6954]\n",
      "loss: 0.544568  [ 3072/ 6954]\n",
      "loss: 0.546420  [ 4096/ 6954]\n",
      "loss: 0.561353  [ 5120/ 6954]\n",
      "loss: 0.536162  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004383 \n",
      "\n",
      "Epoch 318\n",
      "-------------------------------\n",
      "loss: 0.559014  [    0/ 6954]\n",
      "loss: 0.572706  [ 1024/ 6954]\n",
      "loss: 0.575655  [ 2048/ 6954]\n",
      "loss: 0.570008  [ 3072/ 6954]\n",
      "loss: 0.581636  [ 4096/ 6954]\n",
      "loss: 0.537095  [ 5120/ 6954]\n",
      "loss: 0.584130  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004381 \n",
      "\n",
      "Epoch 319\n",
      "-------------------------------\n",
      "loss: 0.569630  [    0/ 6954]\n",
      "loss: 0.551990  [ 1024/ 6954]\n",
      "loss: 0.540181  [ 2048/ 6954]\n",
      "loss: 0.585503  [ 3072/ 6954]\n",
      "loss: 0.578437  [ 4096/ 6954]\n",
      "loss: 0.514354  [ 5120/ 6954]\n",
      "loss: 0.564373  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004380 \n",
      "\n",
      "Epoch 320\n",
      "-------------------------------\n",
      "loss: 0.560183  [    0/ 6954]\n",
      "loss: 0.567935  [ 1024/ 6954]\n",
      "loss: 0.577612  [ 2048/ 6954]\n",
      "loss: 0.570887  [ 3072/ 6954]\n",
      "loss: 0.544538  [ 4096/ 6954]\n",
      "loss: 0.571087  [ 5120/ 6954]\n",
      "loss: 0.552376  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004379 \n",
      "\n",
      "Epoch 321\n",
      "-------------------------------\n",
      "loss: 0.543657  [    0/ 6954]\n",
      "loss: 0.523413  [ 1024/ 6954]\n",
      "loss: 0.562721  [ 2048/ 6954]\n",
      "loss: 0.577241  [ 3072/ 6954]\n",
      "loss: 0.534726  [ 4096/ 6954]\n",
      "loss: 0.559451  [ 5120/ 6954]\n",
      "loss: 0.554083  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004378 \n",
      "\n",
      "Epoch 322\n",
      "-------------------------------\n",
      "loss: 0.551778  [    0/ 6954]\n",
      "loss: 0.562495  [ 1024/ 6954]\n",
      "loss: 0.584581  [ 2048/ 6954]\n",
      "loss: 0.552901  [ 3072/ 6954]\n",
      "loss: 0.543343  [ 4096/ 6954]\n",
      "loss: 0.557450  [ 5120/ 6954]\n",
      "loss: 0.547805  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004377 \n",
      "\n",
      "Epoch 323\n",
      "-------------------------------\n",
      "loss: 0.583261  [    0/ 6954]\n",
      "loss: 0.563301  [ 1024/ 6954]\n",
      "loss: 0.551345  [ 2048/ 6954]\n",
      "loss: 0.526024  [ 3072/ 6954]\n",
      "loss: 0.528497  [ 4096/ 6954]\n",
      "loss: 0.548304  [ 5120/ 6954]\n",
      "loss: 0.553894  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004375 \n",
      "\n",
      "Epoch 324\n",
      "-------------------------------\n",
      "loss: 0.560623  [    0/ 6954]\n",
      "loss: 0.553802  [ 1024/ 6954]\n",
      "loss: 0.528258  [ 2048/ 6954]\n",
      "loss: 0.547350  [ 3072/ 6954]\n",
      "loss: 0.585412  [ 4096/ 6954]\n",
      "loss: 0.569370  [ 5120/ 6954]\n",
      "loss: 0.590859  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004374 \n",
      "\n",
      "Epoch 325\n",
      "-------------------------------\n",
      "loss: 0.549865  [    0/ 6954]\n",
      "loss: 0.540958  [ 1024/ 6954]\n",
      "loss: 0.541131  [ 2048/ 6954]\n",
      "loss: 0.545058  [ 3072/ 6954]\n",
      "loss: 0.551855  [ 4096/ 6954]\n",
      "loss: 0.580726  [ 5120/ 6954]\n",
      "loss: 0.560955  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004373 \n",
      "\n",
      "Epoch 326\n",
      "-------------------------------\n",
      "loss: 0.533439  [    0/ 6954]\n",
      "loss: 0.541508  [ 1024/ 6954]\n",
      "loss: 0.569129  [ 2048/ 6954]\n",
      "loss: 0.529958  [ 3072/ 6954]\n",
      "loss: 0.529400  [ 4096/ 6954]\n",
      "loss: 0.553979  [ 5120/ 6954]\n",
      "loss: 0.567732  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004372 \n",
      "\n",
      "Epoch 327\n",
      "-------------------------------\n",
      "loss: 0.526535  [    0/ 6954]\n",
      "loss: 0.574262  [ 1024/ 6954]\n",
      "loss: 0.568043  [ 2048/ 6954]\n",
      "loss: 0.527238  [ 3072/ 6954]\n",
      "loss: 0.548513  [ 4096/ 6954]\n",
      "loss: 0.578009  [ 5120/ 6954]\n",
      "loss: 0.588284  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004371 \n",
      "\n",
      "Epoch 328\n",
      "-------------------------------\n",
      "loss: 0.549552  [    0/ 6954]\n",
      "loss: 0.580579  [ 1024/ 6954]\n",
      "loss: 0.549304  [ 2048/ 6954]\n",
      "loss: 0.565795  [ 3072/ 6954]\n",
      "loss: 0.576682  [ 4096/ 6954]\n",
      "loss: 0.559745  [ 5120/ 6954]\n",
      "loss: 0.574559  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004369 \n",
      "\n",
      "Epoch 329\n",
      "-------------------------------\n",
      "loss: 0.502657  [    0/ 6954]\n",
      "loss: 0.550736  [ 1024/ 6954]\n",
      "loss: 0.564179  [ 2048/ 6954]\n",
      "loss: 0.565341  [ 3072/ 6954]\n",
      "loss: 0.566093  [ 4096/ 6954]\n",
      "loss: 0.529871  [ 5120/ 6954]\n",
      "loss: 0.553220  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004368 \n",
      "\n",
      "Epoch 330\n",
      "-------------------------------\n",
      "loss: 0.592653  [    0/ 6954]\n",
      "loss: 0.523434  [ 1024/ 6954]\n",
      "loss: 0.528049  [ 2048/ 6954]\n",
      "loss: 0.558493  [ 3072/ 6954]\n",
      "loss: 0.542414  [ 4096/ 6954]\n",
      "loss: 0.558330  [ 5120/ 6954]\n",
      "loss: 0.564551  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004367 \n",
      "\n",
      "Epoch 331\n",
      "-------------------------------\n",
      "loss: 0.522321  [    0/ 6954]\n",
      "loss: 0.570308  [ 1024/ 6954]\n",
      "loss: 0.509780  [ 2048/ 6954]\n",
      "loss: 0.578057  [ 3072/ 6954]\n",
      "loss: 0.544384  [ 4096/ 6954]\n",
      "loss: 0.530945  [ 5120/ 6954]\n",
      "loss: 0.544342  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004366 \n",
      "\n",
      "Epoch 332\n",
      "-------------------------------\n",
      "loss: 0.607695  [    0/ 6954]\n",
      "loss: 0.550392  [ 1024/ 6954]\n",
      "loss: 0.542381  [ 2048/ 6954]\n",
      "loss: 0.560971  [ 3072/ 6954]\n",
      "loss: 0.525695  [ 4096/ 6954]\n",
      "loss: 0.574555  [ 5120/ 6954]\n",
      "loss: 0.540005  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004365 \n",
      "\n",
      "Epoch 333\n",
      "-------------------------------\n",
      "loss: 0.576488  [    0/ 6954]\n",
      "loss: 0.584406  [ 1024/ 6954]\n",
      "loss: 0.604049  [ 2048/ 6954]\n",
      "loss: 0.531875  [ 3072/ 6954]\n",
      "loss: 0.531437  [ 4096/ 6954]\n",
      "loss: 0.567148  [ 5120/ 6954]\n",
      "loss: 0.602629  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004364 \n",
      "\n",
      "Epoch 334\n",
      "-------------------------------\n",
      "loss: 0.546044  [    0/ 6954]\n",
      "loss: 0.563759  [ 1024/ 6954]\n",
      "loss: 0.548755  [ 2048/ 6954]\n",
      "loss: 0.534106  [ 3072/ 6954]\n",
      "loss: 0.570689  [ 4096/ 6954]\n",
      "loss: 0.538758  [ 5120/ 6954]\n",
      "loss: 0.564475  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004363 \n",
      "\n",
      "Epoch 335\n",
      "-------------------------------\n",
      "loss: 0.543078  [    0/ 6954]\n",
      "loss: 0.557090  [ 1024/ 6954]\n",
      "loss: 0.545568  [ 2048/ 6954]\n",
      "loss: 0.555801  [ 3072/ 6954]\n",
      "loss: 0.534372  [ 4096/ 6954]\n",
      "loss: 0.565615  [ 5120/ 6954]\n",
      "loss: 0.549405  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004361 \n",
      "\n",
      "Epoch 336\n",
      "-------------------------------\n",
      "loss: 0.540500  [    0/ 6954]\n",
      "loss: 0.557821  [ 1024/ 6954]\n",
      "loss: 0.595998  [ 2048/ 6954]\n",
      "loss: 0.560973  [ 3072/ 6954]\n",
      "loss: 0.543201  [ 4096/ 6954]\n",
      "loss: 0.539469  [ 5120/ 6954]\n",
      "loss: 0.536928  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004360 \n",
      "\n",
      "Epoch 337\n",
      "-------------------------------\n",
      "loss: 0.591349  [    0/ 6954]\n",
      "loss: 0.548357  [ 1024/ 6954]\n",
      "loss: 0.583946  [ 2048/ 6954]\n",
      "loss: 0.533333  [ 3072/ 6954]\n",
      "loss: 0.516944  [ 4096/ 6954]\n",
      "loss: 0.551311  [ 5120/ 6954]\n",
      "loss: 0.553877  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004359 \n",
      "\n",
      "Epoch 338\n",
      "-------------------------------\n",
      "loss: 0.528808  [    0/ 6954]\n",
      "loss: 0.543396  [ 1024/ 6954]\n",
      "loss: 0.522824  [ 2048/ 6954]\n",
      "loss: 0.550607  [ 3072/ 6954]\n",
      "loss: 0.551181  [ 4096/ 6954]\n",
      "loss: 0.519686  [ 5120/ 6954]\n",
      "loss: 0.565474  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004358 \n",
      "\n",
      "Epoch 339\n",
      "-------------------------------\n",
      "loss: 0.519849  [    0/ 6954]\n",
      "loss: 0.567757  [ 1024/ 6954]\n",
      "loss: 0.603204  [ 2048/ 6954]\n",
      "loss: 0.566387  [ 3072/ 6954]\n",
      "loss: 0.543615  [ 4096/ 6954]\n",
      "loss: 0.570018  [ 5120/ 6954]\n",
      "loss: 0.578254  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004357 \n",
      "\n",
      "Epoch 340\n",
      "-------------------------------\n",
      "loss: 0.578240  [    0/ 6954]\n",
      "loss: 0.583963  [ 1024/ 6954]\n",
      "loss: 0.572993  [ 2048/ 6954]\n",
      "loss: 0.566853  [ 3072/ 6954]\n",
      "loss: 0.513164  [ 4096/ 6954]\n",
      "loss: 0.544325  [ 5120/ 6954]\n",
      "loss: 0.546694  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004356 \n",
      "\n",
      "Epoch 341\n",
      "-------------------------------\n",
      "loss: 0.528951  [    0/ 6954]\n",
      "loss: 0.599075  [ 1024/ 6954]\n",
      "loss: 0.528991  [ 2048/ 6954]\n",
      "loss: 0.558127  [ 3072/ 6954]\n",
      "loss: 0.553958  [ 4096/ 6954]\n",
      "loss: 0.559056  [ 5120/ 6954]\n",
      "loss: 0.526564  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004355 \n",
      "\n",
      "Epoch 342\n",
      "-------------------------------\n",
      "loss: 0.559801  [    0/ 6954]\n",
      "loss: 0.539314  [ 1024/ 6954]\n",
      "loss: 0.544012  [ 2048/ 6954]\n",
      "loss: 0.555720  [ 3072/ 6954]\n",
      "loss: 0.545766  [ 4096/ 6954]\n",
      "loss: 0.573181  [ 5120/ 6954]\n",
      "loss: 0.572359  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004354 \n",
      "\n",
      "Epoch 343\n",
      "-------------------------------\n",
      "loss: 0.602259  [    0/ 6954]\n",
      "loss: 0.546322  [ 1024/ 6954]\n",
      "loss: 0.541903  [ 2048/ 6954]\n",
      "loss: 0.524327  [ 3072/ 6954]\n",
      "loss: 0.516616  [ 4096/ 6954]\n",
      "loss: 0.592436  [ 5120/ 6954]\n",
      "loss: 0.529592  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004353 \n",
      "\n",
      "Epoch 344\n",
      "-------------------------------\n",
      "loss: 0.538961  [    0/ 6954]\n",
      "loss: 0.609502  [ 1024/ 6954]\n",
      "loss: 0.586447  [ 2048/ 6954]\n",
      "loss: 0.526833  [ 3072/ 6954]\n",
      "loss: 0.556747  [ 4096/ 6954]\n",
      "loss: 0.554240  [ 5120/ 6954]\n",
      "loss: 0.554771  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004352 \n",
      "\n",
      "Epoch 345\n",
      "-------------------------------\n",
      "loss: 0.584318  [    0/ 6954]\n",
      "loss: 0.557111  [ 1024/ 6954]\n",
      "loss: 0.514828  [ 2048/ 6954]\n",
      "loss: 0.575373  [ 3072/ 6954]\n",
      "loss: 0.578993  [ 4096/ 6954]\n",
      "loss: 0.559414  [ 5120/ 6954]\n",
      "loss: 0.560861  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004351 \n",
      "\n",
      "Epoch 346\n",
      "-------------------------------\n",
      "loss: 0.512605  [    0/ 6954]\n",
      "loss: 0.566798  [ 1024/ 6954]\n",
      "loss: 0.554656  [ 2048/ 6954]\n",
      "loss: 0.551232  [ 3072/ 6954]\n",
      "loss: 0.523384  [ 4096/ 6954]\n",
      "loss: 0.576173  [ 5120/ 6954]\n",
      "loss: 0.578572  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004350 \n",
      "\n",
      "Epoch 347\n",
      "-------------------------------\n",
      "loss: 0.504970  [    0/ 6954]\n",
      "loss: 0.557274  [ 1024/ 6954]\n",
      "loss: 0.546635  [ 2048/ 6954]\n",
      "loss: 0.545777  [ 3072/ 6954]\n",
      "loss: 0.571734  [ 4096/ 6954]\n",
      "loss: 0.548121  [ 5120/ 6954]\n",
      "loss: 0.564110  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004348 \n",
      "\n",
      "Epoch 348\n",
      "-------------------------------\n",
      "loss: 0.532174  [    0/ 6954]\n",
      "loss: 0.575547  [ 1024/ 6954]\n",
      "loss: 0.557076  [ 2048/ 6954]\n",
      "loss: 0.583509  [ 3072/ 6954]\n",
      "loss: 0.539359  [ 4096/ 6954]\n",
      "loss: 0.592343  [ 5120/ 6954]\n",
      "loss: 0.561218  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004347 \n",
      "\n",
      "Epoch 349\n",
      "-------------------------------\n",
      "loss: 0.556413  [    0/ 6954]\n",
      "loss: 0.570312  [ 1024/ 6954]\n",
      "loss: 0.561853  [ 2048/ 6954]\n",
      "loss: 0.511717  [ 3072/ 6954]\n",
      "loss: 0.531755  [ 4096/ 6954]\n",
      "loss: 0.565396  [ 5120/ 6954]\n",
      "loss: 0.579018  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004346 \n",
      "\n",
      "Epoch 350\n",
      "-------------------------------\n",
      "loss: 0.555097  [    0/ 6954]\n",
      "loss: 0.530550  [ 1024/ 6954]\n",
      "loss: 0.588805  [ 2048/ 6954]\n",
      "loss: 0.548091  [ 3072/ 6954]\n",
      "loss: 0.538521  [ 4096/ 6954]\n",
      "loss: 0.533245  [ 5120/ 6954]\n",
      "loss: 0.555512  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004345 \n",
      "\n",
      "Epoch 351\n",
      "-------------------------------\n",
      "loss: 0.549892  [    0/ 6954]\n",
      "loss: 0.549740  [ 1024/ 6954]\n",
      "loss: 0.542256  [ 2048/ 6954]\n",
      "loss: 0.564197  [ 3072/ 6954]\n",
      "loss: 0.526928  [ 4096/ 6954]\n",
      "loss: 0.576717  [ 5120/ 6954]\n",
      "loss: 0.545192  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004344 \n",
      "\n",
      "Epoch 352\n",
      "-------------------------------\n",
      "loss: 0.512664  [    0/ 6954]\n",
      "loss: 0.609675  [ 1024/ 6954]\n",
      "loss: 0.562816  [ 2048/ 6954]\n",
      "loss: 0.566604  [ 3072/ 6954]\n",
      "loss: 0.558203  [ 4096/ 6954]\n",
      "loss: 0.606061  [ 5120/ 6954]\n",
      "loss: 0.566839  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004343 \n",
      "\n",
      "Epoch 353\n",
      "-------------------------------\n",
      "loss: 0.559865  [    0/ 6954]\n",
      "loss: 0.588917  [ 1024/ 6954]\n",
      "loss: 0.562574  [ 2048/ 6954]\n",
      "loss: 0.571979  [ 3072/ 6954]\n",
      "loss: 0.546866  [ 4096/ 6954]\n",
      "loss: 0.593224  [ 5120/ 6954]\n",
      "loss: 0.556160  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004342 \n",
      "\n",
      "Epoch 354\n",
      "-------------------------------\n",
      "loss: 0.553478  [    0/ 6954]\n",
      "loss: 0.538061  [ 1024/ 6954]\n",
      "loss: 0.546432  [ 2048/ 6954]\n",
      "loss: 0.576429  [ 3072/ 6954]\n",
      "loss: 0.556483  [ 4096/ 6954]\n",
      "loss: 0.549337  [ 5120/ 6954]\n",
      "loss: 0.590013  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004341 \n",
      "\n",
      "Epoch 355\n",
      "-------------------------------\n",
      "loss: 0.572171  [    0/ 6954]\n",
      "loss: 0.559136  [ 1024/ 6954]\n",
      "loss: 0.525115  [ 2048/ 6954]\n",
      "loss: 0.572036  [ 3072/ 6954]\n",
      "loss: 0.585083  [ 4096/ 6954]\n",
      "loss: 0.551901  [ 5120/ 6954]\n",
      "loss: 0.579361  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004340 \n",
      "\n",
      "Epoch 356\n",
      "-------------------------------\n",
      "loss: 0.517459  [    0/ 6954]\n",
      "loss: 0.547662  [ 1024/ 6954]\n",
      "loss: 0.511029  [ 2048/ 6954]\n",
      "loss: 0.571888  [ 3072/ 6954]\n",
      "loss: 0.537043  [ 4096/ 6954]\n",
      "loss: 0.579860  [ 5120/ 6954]\n",
      "loss: 0.552925  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004339 \n",
      "\n",
      "Epoch 357\n",
      "-------------------------------\n",
      "loss: 0.525123  [    0/ 6954]\n",
      "loss: 0.529877  [ 1024/ 6954]\n",
      "loss: 0.559289  [ 2048/ 6954]\n",
      "loss: 0.508898  [ 3072/ 6954]\n",
      "loss: 0.524532  [ 4096/ 6954]\n",
      "loss: 0.533471  [ 5120/ 6954]\n",
      "loss: 0.536431  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004338 \n",
      "\n",
      "Epoch 358\n",
      "-------------------------------\n",
      "loss: 0.527833  [    0/ 6954]\n",
      "loss: 0.574376  [ 1024/ 6954]\n",
      "loss: 0.531767  [ 2048/ 6954]\n",
      "loss: 0.498586  [ 3072/ 6954]\n",
      "loss: 0.561541  [ 4096/ 6954]\n",
      "loss: 0.569982  [ 5120/ 6954]\n",
      "loss: 0.557882  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004337 \n",
      "\n",
      "Epoch 359\n",
      "-------------------------------\n",
      "loss: 0.574148  [    0/ 6954]\n",
      "loss: 0.544757  [ 1024/ 6954]\n",
      "loss: 0.552617  [ 2048/ 6954]\n",
      "loss: 0.547300  [ 3072/ 6954]\n",
      "loss: 0.549731  [ 4096/ 6954]\n",
      "loss: 0.562631  [ 5120/ 6954]\n",
      "loss: 0.576363  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004337 \n",
      "\n",
      "Epoch 360\n",
      "-------------------------------\n",
      "loss: 0.536758  [    0/ 6954]\n",
      "loss: 0.550408  [ 1024/ 6954]\n",
      "loss: 0.529850  [ 2048/ 6954]\n",
      "loss: 0.508149  [ 3072/ 6954]\n",
      "loss: 0.552949  [ 4096/ 6954]\n",
      "loss: 0.560987  [ 5120/ 6954]\n",
      "loss: 0.511809  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.7%, Avg loss: 0.004335 \n",
      "\n",
      "Epoch 361\n",
      "-------------------------------\n",
      "loss: 0.583398  [    0/ 6954]\n",
      "loss: 0.541473  [ 1024/ 6954]\n",
      "loss: 0.539859  [ 2048/ 6954]\n",
      "loss: 0.527837  [ 3072/ 6954]\n",
      "loss: 0.551008  [ 4096/ 6954]\n",
      "loss: 0.555611  [ 5120/ 6954]\n",
      "loss: 0.544008  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004335 \n",
      "\n",
      "Epoch 362\n",
      "-------------------------------\n",
      "loss: 0.530826  [    0/ 6954]\n",
      "loss: 0.553322  [ 1024/ 6954]\n",
      "loss: 0.587436  [ 2048/ 6954]\n",
      "loss: 0.537400  [ 3072/ 6954]\n",
      "loss: 0.521052  [ 4096/ 6954]\n",
      "loss: 0.540045  [ 5120/ 6954]\n",
      "loss: 0.591098  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004334 \n",
      "\n",
      "Epoch 363\n",
      "-------------------------------\n",
      "loss: 0.567889  [    0/ 6954]\n",
      "loss: 0.563773  [ 1024/ 6954]\n",
      "loss: 0.567900  [ 2048/ 6954]\n",
      "loss: 0.576807  [ 3072/ 6954]\n",
      "loss: 0.540187  [ 4096/ 6954]\n",
      "loss: 0.541881  [ 5120/ 6954]\n",
      "loss: 0.554466  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004333 \n",
      "\n",
      "Epoch 364\n",
      "-------------------------------\n",
      "loss: 0.544999  [    0/ 6954]\n",
      "loss: 0.531644  [ 1024/ 6954]\n",
      "loss: 0.560991  [ 2048/ 6954]\n",
      "loss: 0.583914  [ 3072/ 6954]\n",
      "loss: 0.564672  [ 4096/ 6954]\n",
      "loss: 0.540786  [ 5120/ 6954]\n",
      "loss: 0.555130  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004332 \n",
      "\n",
      "Epoch 365\n",
      "-------------------------------\n",
      "loss: 0.544735  [    0/ 6954]\n",
      "loss: 0.565531  [ 1024/ 6954]\n",
      "loss: 0.547035  [ 2048/ 6954]\n",
      "loss: 0.536582  [ 3072/ 6954]\n",
      "loss: 0.523914  [ 4096/ 6954]\n",
      "loss: 0.514898  [ 5120/ 6954]\n",
      "loss: 0.528581  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004331 \n",
      "\n",
      "Epoch 366\n",
      "-------------------------------\n",
      "loss: 0.563729  [    0/ 6954]\n",
      "loss: 0.570655  [ 1024/ 6954]\n",
      "loss: 0.575417  [ 2048/ 6954]\n",
      "loss: 0.517302  [ 3072/ 6954]\n",
      "loss: 0.560639  [ 4096/ 6954]\n",
      "loss: 0.552757  [ 5120/ 6954]\n",
      "loss: 0.546770  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004330 \n",
      "\n",
      "Epoch 367\n",
      "-------------------------------\n",
      "loss: 0.516428  [    0/ 6954]\n",
      "loss: 0.557489  [ 1024/ 6954]\n",
      "loss: 0.549986  [ 2048/ 6954]\n",
      "loss: 0.554730  [ 3072/ 6954]\n",
      "loss: 0.547931  [ 4096/ 6954]\n",
      "loss: 0.546371  [ 5120/ 6954]\n",
      "loss: 0.532713  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004329 \n",
      "\n",
      "Epoch 368\n",
      "-------------------------------\n",
      "loss: 0.543275  [    0/ 6954]\n",
      "loss: 0.565145  [ 1024/ 6954]\n",
      "loss: 0.540595  [ 2048/ 6954]\n",
      "loss: 0.560151  [ 3072/ 6954]\n",
      "loss: 0.563817  [ 4096/ 6954]\n",
      "loss: 0.570415  [ 5120/ 6954]\n",
      "loss: 0.530160  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004328 \n",
      "\n",
      "Epoch 369\n",
      "-------------------------------\n",
      "loss: 0.527525  [    0/ 6954]\n",
      "loss: 0.588316  [ 1024/ 6954]\n",
      "loss: 0.519231  [ 2048/ 6954]\n",
      "loss: 0.531543  [ 3072/ 6954]\n",
      "loss: 0.559191  [ 4096/ 6954]\n",
      "loss: 0.593810  [ 5120/ 6954]\n",
      "loss: 0.553606  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004327 \n",
      "\n",
      "Epoch 370\n",
      "-------------------------------\n",
      "loss: 0.555643  [    0/ 6954]\n",
      "loss: 0.582876  [ 1024/ 6954]\n",
      "loss: 0.561863  [ 2048/ 6954]\n",
      "loss: 0.544416  [ 3072/ 6954]\n",
      "loss: 0.592798  [ 4096/ 6954]\n",
      "loss: 0.558956  [ 5120/ 6954]\n",
      "loss: 0.530187  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004326 \n",
      "\n",
      "Epoch 371\n",
      "-------------------------------\n",
      "loss: 0.574947  [    0/ 6954]\n",
      "loss: 0.546001  [ 1024/ 6954]\n",
      "loss: 0.559831  [ 2048/ 6954]\n",
      "loss: 0.518888  [ 3072/ 6954]\n",
      "loss: 0.565835  [ 4096/ 6954]\n",
      "loss: 0.583732  [ 5120/ 6954]\n",
      "loss: 0.528992  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004325 \n",
      "\n",
      "Epoch 372\n",
      "-------------------------------\n",
      "loss: 0.552902  [    0/ 6954]\n",
      "loss: 0.556900  [ 1024/ 6954]\n",
      "loss: 0.520684  [ 2048/ 6954]\n",
      "loss: 0.504053  [ 3072/ 6954]\n",
      "loss: 0.594349  [ 4096/ 6954]\n",
      "loss: 0.566197  [ 5120/ 6954]\n",
      "loss: 0.551549  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004324 \n",
      "\n",
      "Epoch 373\n",
      "-------------------------------\n",
      "loss: 0.560259  [    0/ 6954]\n",
      "loss: 0.572533  [ 1024/ 6954]\n",
      "loss: 0.556514  [ 2048/ 6954]\n",
      "loss: 0.552498  [ 3072/ 6954]\n",
      "loss: 0.584018  [ 4096/ 6954]\n",
      "loss: 0.517635  [ 5120/ 6954]\n",
      "loss: 0.544499  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.8%, Avg loss: 0.004323 \n",
      "\n",
      "Epoch 374\n",
      "-------------------------------\n",
      "loss: 0.556709  [    0/ 6954]\n",
      "loss: 0.524257  [ 1024/ 6954]\n",
      "loss: 0.534149  [ 2048/ 6954]\n",
      "loss: 0.566609  [ 3072/ 6954]\n",
      "loss: 0.547505  [ 4096/ 6954]\n",
      "loss: 0.571263  [ 5120/ 6954]\n",
      "loss: 0.528269  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004323 \n",
      "\n",
      "Epoch 375\n",
      "-------------------------------\n",
      "loss: 0.522607  [    0/ 6954]\n",
      "loss: 0.538651  [ 1024/ 6954]\n",
      "loss: 0.512188  [ 2048/ 6954]\n",
      "loss: 0.586681  [ 3072/ 6954]\n",
      "loss: 0.540028  [ 4096/ 6954]\n",
      "loss: 0.581434  [ 5120/ 6954]\n",
      "loss: 0.540097  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004322 \n",
      "\n",
      "Epoch 376\n",
      "-------------------------------\n",
      "loss: 0.555152  [    0/ 6954]\n",
      "loss: 0.550497  [ 1024/ 6954]\n",
      "loss: 0.593135  [ 2048/ 6954]\n",
      "loss: 0.549883  [ 3072/ 6954]\n",
      "loss: 0.542736  [ 4096/ 6954]\n",
      "loss: 0.550054  [ 5120/ 6954]\n",
      "loss: 0.514347  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004321 \n",
      "\n",
      "Epoch 377\n",
      "-------------------------------\n",
      "loss: 0.518712  [    0/ 6954]\n",
      "loss: 0.529734  [ 1024/ 6954]\n",
      "loss: 0.558970  [ 2048/ 6954]\n",
      "loss: 0.568269  [ 3072/ 6954]\n",
      "loss: 0.557965  [ 4096/ 6954]\n",
      "loss: 0.519860  [ 5120/ 6954]\n",
      "loss: 0.544316  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004320 \n",
      "\n",
      "Epoch 378\n",
      "-------------------------------\n",
      "loss: 0.573725  [    0/ 6954]\n",
      "loss: 0.529354  [ 1024/ 6954]\n",
      "loss: 0.600338  [ 2048/ 6954]\n",
      "loss: 0.566565  [ 3072/ 6954]\n",
      "loss: 0.495506  [ 4096/ 6954]\n",
      "loss: 0.524362  [ 5120/ 6954]\n",
      "loss: 0.543401  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004319 \n",
      "\n",
      "Epoch 379\n",
      "-------------------------------\n",
      "loss: 0.559231  [    0/ 6954]\n",
      "loss: 0.587005  [ 1024/ 6954]\n",
      "loss: 0.557027  [ 2048/ 6954]\n",
      "loss: 0.559397  [ 3072/ 6954]\n",
      "loss: 0.514830  [ 4096/ 6954]\n",
      "loss: 0.574486  [ 5120/ 6954]\n",
      "loss: 0.518611  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004318 \n",
      "\n",
      "Epoch 380\n",
      "-------------------------------\n",
      "loss: 0.517341  [    0/ 6954]\n",
      "loss: 0.554122  [ 1024/ 6954]\n",
      "loss: 0.571930  [ 2048/ 6954]\n",
      "loss: 0.562090  [ 3072/ 6954]\n",
      "loss: 0.521651  [ 4096/ 6954]\n",
      "loss: 0.513362  [ 5120/ 6954]\n",
      "loss: 0.593201  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004317 \n",
      "\n",
      "Epoch 381\n",
      "-------------------------------\n",
      "loss: 0.546585  [    0/ 6954]\n",
      "loss: 0.580766  [ 1024/ 6954]\n",
      "loss: 0.550888  [ 2048/ 6954]\n",
      "loss: 0.550692  [ 3072/ 6954]\n",
      "loss: 0.561353  [ 4096/ 6954]\n",
      "loss: 0.555270  [ 5120/ 6954]\n",
      "loss: 0.543630  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004317 \n",
      "\n",
      "Epoch 382\n",
      "-------------------------------\n",
      "loss: 0.503312  [    0/ 6954]\n",
      "loss: 0.573567  [ 1024/ 6954]\n",
      "loss: 0.567433  [ 2048/ 6954]\n",
      "loss: 0.491200  [ 3072/ 6954]\n",
      "loss: 0.547069  [ 4096/ 6954]\n",
      "loss: 0.552246  [ 5120/ 6954]\n",
      "loss: 0.539741  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004316 \n",
      "\n",
      "Epoch 383\n",
      "-------------------------------\n",
      "loss: 0.506068  [    0/ 6954]\n",
      "loss: 0.526276  [ 1024/ 6954]\n",
      "loss: 0.550449  [ 2048/ 6954]\n",
      "loss: 0.515328  [ 3072/ 6954]\n",
      "loss: 0.484690  [ 4096/ 6954]\n",
      "loss: 0.530356  [ 5120/ 6954]\n",
      "loss: 0.498282  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004315 \n",
      "\n",
      "Epoch 384\n",
      "-------------------------------\n",
      "loss: 0.552876  [    0/ 6954]\n",
      "loss: 0.520241  [ 1024/ 6954]\n",
      "loss: 0.542702  [ 2048/ 6954]\n",
      "loss: 0.548489  [ 3072/ 6954]\n",
      "loss: 0.551021  [ 4096/ 6954]\n",
      "loss: 0.544115  [ 5120/ 6954]\n",
      "loss: 0.524338  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004314 \n",
      "\n",
      "Epoch 385\n",
      "-------------------------------\n",
      "loss: 0.584240  [    0/ 6954]\n",
      "loss: 0.558597  [ 1024/ 6954]\n",
      "loss: 0.588236  [ 2048/ 6954]\n",
      "loss: 0.581352  [ 3072/ 6954]\n",
      "loss: 0.533573  [ 4096/ 6954]\n",
      "loss: 0.523752  [ 5120/ 6954]\n",
      "loss: 0.561097  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004313 \n",
      "\n",
      "Epoch 386\n",
      "-------------------------------\n",
      "loss: 0.536451  [    0/ 6954]\n",
      "loss: 0.541760  [ 1024/ 6954]\n",
      "loss: 0.524517  [ 2048/ 6954]\n",
      "loss: 0.527584  [ 3072/ 6954]\n",
      "loss: 0.574062  [ 4096/ 6954]\n",
      "loss: 0.565479  [ 5120/ 6954]\n",
      "loss: 0.551980  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004312 \n",
      "\n",
      "Epoch 387\n",
      "-------------------------------\n",
      "loss: 0.516446  [    0/ 6954]\n",
      "loss: 0.553976  [ 1024/ 6954]\n",
      "loss: 0.529665  [ 2048/ 6954]\n",
      "loss: 0.558728  [ 3072/ 6954]\n",
      "loss: 0.542306  [ 4096/ 6954]\n",
      "loss: 0.493970  [ 5120/ 6954]\n",
      "loss: 0.553479  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004311 \n",
      "\n",
      "Epoch 388\n",
      "-------------------------------\n",
      "loss: 0.564662  [    0/ 6954]\n",
      "loss: 0.544370  [ 1024/ 6954]\n",
      "loss: 0.527768  [ 2048/ 6954]\n",
      "loss: 0.551151  [ 3072/ 6954]\n",
      "loss: 0.550704  [ 4096/ 6954]\n",
      "loss: 0.546059  [ 5120/ 6954]\n",
      "loss: 0.567090  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004311 \n",
      "\n",
      "Epoch 389\n",
      "-------------------------------\n",
      "loss: 0.497332  [    0/ 6954]\n",
      "loss: 0.544977  [ 1024/ 6954]\n",
      "loss: 0.546577  [ 2048/ 6954]\n",
      "loss: 0.585094  [ 3072/ 6954]\n",
      "loss: 0.551582  [ 4096/ 6954]\n",
      "loss: 0.536796  [ 5120/ 6954]\n",
      "loss: 0.562068  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004310 \n",
      "\n",
      "Epoch 390\n",
      "-------------------------------\n",
      "loss: 0.554929  [    0/ 6954]\n",
      "loss: 0.569179  [ 1024/ 6954]\n",
      "loss: 0.516601  [ 2048/ 6954]\n",
      "loss: 0.554642  [ 3072/ 6954]\n",
      "loss: 0.549448  [ 4096/ 6954]\n",
      "loss: 0.556895  [ 5120/ 6954]\n",
      "loss: 0.565149  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004309 \n",
      "\n",
      "Epoch 391\n",
      "-------------------------------\n",
      "loss: 0.529331  [    0/ 6954]\n",
      "loss: 0.587404  [ 1024/ 6954]\n",
      "loss: 0.539083  [ 2048/ 6954]\n",
      "loss: 0.513824  [ 3072/ 6954]\n",
      "loss: 0.524927  [ 4096/ 6954]\n",
      "loss: 0.534345  [ 5120/ 6954]\n",
      "loss: 0.575427  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004308 \n",
      "\n",
      "Epoch 392\n",
      "-------------------------------\n",
      "loss: 0.526195  [    0/ 6954]\n",
      "loss: 0.590411  [ 1024/ 6954]\n",
      "loss: 0.539409  [ 2048/ 6954]\n",
      "loss: 0.537086  [ 3072/ 6954]\n",
      "loss: 0.503066  [ 4096/ 6954]\n",
      "loss: 0.519693  [ 5120/ 6954]\n",
      "loss: 0.568046  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004307 \n",
      "\n",
      "Epoch 393\n",
      "-------------------------------\n",
      "loss: 0.555556  [    0/ 6954]\n",
      "loss: 0.525810  [ 1024/ 6954]\n",
      "loss: 0.524294  [ 2048/ 6954]\n",
      "loss: 0.549763  [ 3072/ 6954]\n",
      "loss: 0.514780  [ 4096/ 6954]\n",
      "loss: 0.575639  [ 5120/ 6954]\n",
      "loss: 0.566899  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004306 \n",
      "\n",
      "Epoch 394\n",
      "-------------------------------\n",
      "loss: 0.534677  [    0/ 6954]\n",
      "loss: 0.543302  [ 1024/ 6954]\n",
      "loss: 0.544896  [ 2048/ 6954]\n",
      "loss: 0.522403  [ 3072/ 6954]\n",
      "loss: 0.584963  [ 4096/ 6954]\n",
      "loss: 0.563120  [ 5120/ 6954]\n",
      "loss: 0.514758  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004306 \n",
      "\n",
      "Epoch 395\n",
      "-------------------------------\n",
      "loss: 0.534967  [    0/ 6954]\n",
      "loss: 0.560352  [ 1024/ 6954]\n",
      "loss: 0.530539  [ 2048/ 6954]\n",
      "loss: 0.512266  [ 3072/ 6954]\n",
      "loss: 0.576614  [ 4096/ 6954]\n",
      "loss: 0.590844  [ 5120/ 6954]\n",
      "loss: 0.553017  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004305 \n",
      "\n",
      "Epoch 396\n",
      "-------------------------------\n",
      "loss: 0.507082  [    0/ 6954]\n",
      "loss: 0.569191  [ 1024/ 6954]\n",
      "loss: 0.518423  [ 2048/ 6954]\n",
      "loss: 0.560534  [ 3072/ 6954]\n",
      "loss: 0.516924  [ 4096/ 6954]\n",
      "loss: 0.556829  [ 5120/ 6954]\n",
      "loss: 0.548939  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004304 \n",
      "\n",
      "Epoch 397\n",
      "-------------------------------\n",
      "loss: 0.581136  [    0/ 6954]\n",
      "loss: 0.515704  [ 1024/ 6954]\n",
      "loss: 0.507193  [ 2048/ 6954]\n",
      "loss: 0.542176  [ 3072/ 6954]\n",
      "loss: 0.593356  [ 4096/ 6954]\n",
      "loss: 0.589478  [ 5120/ 6954]\n",
      "loss: 0.559938  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004304 \n",
      "\n",
      "Epoch 398\n",
      "-------------------------------\n",
      "loss: 0.536023  [    0/ 6954]\n",
      "loss: 0.538173  [ 1024/ 6954]\n",
      "loss: 0.567558  [ 2048/ 6954]\n",
      "loss: 0.575098  [ 3072/ 6954]\n",
      "loss: 0.511049  [ 4096/ 6954]\n",
      "loss: 0.525168  [ 5120/ 6954]\n",
      "loss: 0.545424  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004303 \n",
      "\n",
      "Epoch 399\n",
      "-------------------------------\n",
      "loss: 0.542770  [    0/ 6954]\n",
      "loss: 0.569132  [ 1024/ 6954]\n",
      "loss: 0.508373  [ 2048/ 6954]\n",
      "loss: 0.575966  [ 3072/ 6954]\n",
      "loss: 0.524524  [ 4096/ 6954]\n",
      "loss: 0.572775  [ 5120/ 6954]\n",
      "loss: 0.549294  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004302 \n",
      "\n",
      "Epoch 400\n",
      "-------------------------------\n",
      "loss: 0.529043  [    0/ 6954]\n",
      "loss: 0.591168  [ 1024/ 6954]\n",
      "loss: 0.543519  [ 2048/ 6954]\n",
      "loss: 0.587626  [ 3072/ 6954]\n",
      "loss: 0.581648  [ 4096/ 6954]\n",
      "loss: 0.523754  [ 5120/ 6954]\n",
      "loss: 0.536295  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004301 \n",
      "\n",
      "Epoch 401\n",
      "-------------------------------\n",
      "loss: 0.585157  [    0/ 6954]\n",
      "loss: 0.577775  [ 1024/ 6954]\n",
      "loss: 0.566191  [ 2048/ 6954]\n",
      "loss: 0.568338  [ 3072/ 6954]\n",
      "loss: 0.489926  [ 4096/ 6954]\n",
      "loss: 0.546181  [ 5120/ 6954]\n",
      "loss: 0.523110  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004301 \n",
      "\n",
      "Epoch 402\n",
      "-------------------------------\n",
      "loss: 0.497559  [    0/ 6954]\n",
      "loss: 0.532400  [ 1024/ 6954]\n",
      "loss: 0.558388  [ 2048/ 6954]\n",
      "loss: 0.578110  [ 3072/ 6954]\n",
      "loss: 0.544367  [ 4096/ 6954]\n",
      "loss: 0.543995  [ 5120/ 6954]\n",
      "loss: 0.591465  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004300 \n",
      "\n",
      "Epoch 403\n",
      "-------------------------------\n",
      "loss: 0.528106  [    0/ 6954]\n",
      "loss: 0.553633  [ 1024/ 6954]\n",
      "loss: 0.551683  [ 2048/ 6954]\n",
      "loss: 0.557718  [ 3072/ 6954]\n",
      "loss: 0.542915  [ 4096/ 6954]\n",
      "loss: 0.550115  [ 5120/ 6954]\n",
      "loss: 0.507098  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004299 \n",
      "\n",
      "Epoch 404\n",
      "-------------------------------\n",
      "loss: 0.510604  [    0/ 6954]\n",
      "loss: 0.507794  [ 1024/ 6954]\n",
      "loss: 0.491673  [ 2048/ 6954]\n",
      "loss: 0.536433  [ 3072/ 6954]\n",
      "loss: 0.511988  [ 4096/ 6954]\n",
      "loss: 0.599428  [ 5120/ 6954]\n",
      "loss: 0.527179  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004298 \n",
      "\n",
      "Epoch 405\n",
      "-------------------------------\n",
      "loss: 0.492268  [    0/ 6954]\n",
      "loss: 0.534364  [ 1024/ 6954]\n",
      "loss: 0.528083  [ 2048/ 6954]\n",
      "loss: 0.534520  [ 3072/ 6954]\n",
      "loss: 0.590191  [ 4096/ 6954]\n",
      "loss: 0.544487  [ 5120/ 6954]\n",
      "loss: 0.526056  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004297 \n",
      "\n",
      "Epoch 406\n",
      "-------------------------------\n",
      "loss: 0.556655  [    0/ 6954]\n",
      "loss: 0.553259  [ 1024/ 6954]\n",
      "loss: 0.523681  [ 2048/ 6954]\n",
      "loss: 0.582098  [ 3072/ 6954]\n",
      "loss: 0.515346  [ 4096/ 6954]\n",
      "loss: 0.536753  [ 5120/ 6954]\n",
      "loss: 0.564533  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004297 \n",
      "\n",
      "Epoch 407\n",
      "-------------------------------\n",
      "loss: 0.512540  [    0/ 6954]\n",
      "loss: 0.578331  [ 1024/ 6954]\n",
      "loss: 0.555300  [ 2048/ 6954]\n",
      "loss: 0.542169  [ 3072/ 6954]\n",
      "loss: 0.519786  [ 4096/ 6954]\n",
      "loss: 0.544479  [ 5120/ 6954]\n",
      "loss: 0.540904  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004296 \n",
      "\n",
      "Epoch 408\n",
      "-------------------------------\n",
      "loss: 0.551867  [    0/ 6954]\n",
      "loss: 0.557123  [ 1024/ 6954]\n",
      "loss: 0.575276  [ 2048/ 6954]\n",
      "loss: 0.551759  [ 3072/ 6954]\n",
      "loss: 0.511494  [ 4096/ 6954]\n",
      "loss: 0.539580  [ 5120/ 6954]\n",
      "loss: 0.553564  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004295 \n",
      "\n",
      "Epoch 409\n",
      "-------------------------------\n",
      "loss: 0.546126  [    0/ 6954]\n",
      "loss: 0.505533  [ 1024/ 6954]\n",
      "loss: 0.538372  [ 2048/ 6954]\n",
      "loss: 0.530698  [ 3072/ 6954]\n",
      "loss: 0.551563  [ 4096/ 6954]\n",
      "loss: 0.557315  [ 5120/ 6954]\n",
      "loss: 0.529021  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004294 \n",
      "\n",
      "Epoch 410\n",
      "-------------------------------\n",
      "loss: 0.550768  [    0/ 6954]\n",
      "loss: 0.544377  [ 1024/ 6954]\n",
      "loss: 0.604725  [ 2048/ 6954]\n",
      "loss: 0.544037  [ 3072/ 6954]\n",
      "loss: 0.553475  [ 4096/ 6954]\n",
      "loss: 0.564651  [ 5120/ 6954]\n",
      "loss: 0.583622  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004294 \n",
      "\n",
      "Epoch 411\n",
      "-------------------------------\n",
      "loss: 0.535614  [    0/ 6954]\n",
      "loss: 0.575831  [ 1024/ 6954]\n",
      "loss: 0.572446  [ 2048/ 6954]\n",
      "loss: 0.578447  [ 3072/ 6954]\n",
      "loss: 0.518791  [ 4096/ 6954]\n",
      "loss: 0.563115  [ 5120/ 6954]\n",
      "loss: 0.560834  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004293 \n",
      "\n",
      "Epoch 412\n",
      "-------------------------------\n",
      "loss: 0.596287  [    0/ 6954]\n",
      "loss: 0.547697  [ 1024/ 6954]\n",
      "loss: 0.559313  [ 2048/ 6954]\n",
      "loss: 0.574379  [ 3072/ 6954]\n",
      "loss: 0.559974  [ 4096/ 6954]\n",
      "loss: 0.552650  [ 5120/ 6954]\n",
      "loss: 0.514347  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004292 \n",
      "\n",
      "Epoch 413\n",
      "-------------------------------\n",
      "loss: 0.541755  [    0/ 6954]\n",
      "loss: 0.554166  [ 1024/ 6954]\n",
      "loss: 0.539982  [ 2048/ 6954]\n",
      "loss: 0.558779  [ 3072/ 6954]\n",
      "loss: 0.570898  [ 4096/ 6954]\n",
      "loss: 0.506112  [ 5120/ 6954]\n",
      "loss: 0.544126  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004292 \n",
      "\n",
      "Epoch 414\n",
      "-------------------------------\n",
      "loss: 0.527526  [    0/ 6954]\n",
      "loss: 0.496321  [ 1024/ 6954]\n",
      "loss: 0.534637  [ 2048/ 6954]\n",
      "loss: 0.519006  [ 3072/ 6954]\n",
      "loss: 0.553986  [ 4096/ 6954]\n",
      "loss: 0.521274  [ 5120/ 6954]\n",
      "loss: 0.577289  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004291 \n",
      "\n",
      "Epoch 415\n",
      "-------------------------------\n",
      "loss: 0.598692  [    0/ 6954]\n",
      "loss: 0.543892  [ 1024/ 6954]\n",
      "loss: 0.506455  [ 2048/ 6954]\n",
      "loss: 0.532895  [ 3072/ 6954]\n",
      "loss: 0.561936  [ 4096/ 6954]\n",
      "loss: 0.526587  [ 5120/ 6954]\n",
      "loss: 0.543278  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004290 \n",
      "\n",
      "Epoch 416\n",
      "-------------------------------\n",
      "loss: 0.541060  [    0/ 6954]\n",
      "loss: 0.566678  [ 1024/ 6954]\n",
      "loss: 0.561688  [ 2048/ 6954]\n",
      "loss: 0.539533  [ 3072/ 6954]\n",
      "loss: 0.528388  [ 4096/ 6954]\n",
      "loss: 0.550957  [ 5120/ 6954]\n",
      "loss: 0.532755  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004289 \n",
      "\n",
      "Epoch 417\n",
      "-------------------------------\n",
      "loss: 0.567135  [    0/ 6954]\n",
      "loss: 0.561714  [ 1024/ 6954]\n",
      "loss: 0.542542  [ 2048/ 6954]\n",
      "loss: 0.579863  [ 3072/ 6954]\n",
      "loss: 0.562761  [ 4096/ 6954]\n",
      "loss: 0.548423  [ 5120/ 6954]\n",
      "loss: 0.581232  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004289 \n",
      "\n",
      "Epoch 418\n",
      "-------------------------------\n",
      "loss: 0.512521  [    0/ 6954]\n",
      "loss: 0.553957  [ 1024/ 6954]\n",
      "loss: 0.575353  [ 2048/ 6954]\n",
      "loss: 0.546839  [ 3072/ 6954]\n",
      "loss: 0.570188  [ 4096/ 6954]\n",
      "loss: 0.534523  [ 5120/ 6954]\n",
      "loss: 0.515223  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004288 \n",
      "\n",
      "Epoch 419\n",
      "-------------------------------\n",
      "loss: 0.509596  [    0/ 6954]\n",
      "loss: 0.524668  [ 1024/ 6954]\n",
      "loss: 0.555314  [ 2048/ 6954]\n",
      "loss: 0.550414  [ 3072/ 6954]\n",
      "loss: 0.578822  [ 4096/ 6954]\n",
      "loss: 0.536332  [ 5120/ 6954]\n",
      "loss: 0.521426  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004287 \n",
      "\n",
      "Epoch 420\n",
      "-------------------------------\n",
      "loss: 0.572435  [    0/ 6954]\n",
      "loss: 0.519112  [ 1024/ 6954]\n",
      "loss: 0.555129  [ 2048/ 6954]\n",
      "loss: 0.558528  [ 3072/ 6954]\n",
      "loss: 0.553785  [ 4096/ 6954]\n",
      "loss: 0.540567  [ 5120/ 6954]\n",
      "loss: 0.544652  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004287 \n",
      "\n",
      "Epoch 421\n",
      "-------------------------------\n",
      "loss: 0.516021  [    0/ 6954]\n",
      "loss: 0.539852  [ 1024/ 6954]\n",
      "loss: 0.588078  [ 2048/ 6954]\n",
      "loss: 0.562525  [ 3072/ 6954]\n",
      "loss: 0.561178  [ 4096/ 6954]\n",
      "loss: 0.523496  [ 5120/ 6954]\n",
      "loss: 0.562072  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004286 \n",
      "\n",
      "Epoch 422\n",
      "-------------------------------\n",
      "loss: 0.525423  [    0/ 6954]\n",
      "loss: 0.526637  [ 1024/ 6954]\n",
      "loss: 0.550515  [ 2048/ 6954]\n",
      "loss: 0.559068  [ 3072/ 6954]\n",
      "loss: 0.529081  [ 4096/ 6954]\n",
      "loss: 0.570800  [ 5120/ 6954]\n",
      "loss: 0.512692  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004285 \n",
      "\n",
      "Epoch 423\n",
      "-------------------------------\n",
      "loss: 0.532967  [    0/ 6954]\n",
      "loss: 0.541311  [ 1024/ 6954]\n",
      "loss: 0.540067  [ 2048/ 6954]\n",
      "loss: 0.530898  [ 3072/ 6954]\n",
      "loss: 0.564405  [ 4096/ 6954]\n",
      "loss: 0.557207  [ 5120/ 6954]\n",
      "loss: 0.549703  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004284 \n",
      "\n",
      "Epoch 424\n",
      "-------------------------------\n",
      "loss: 0.533764  [    0/ 6954]\n",
      "loss: 0.584521  [ 1024/ 6954]\n",
      "loss: 0.542479  [ 2048/ 6954]\n",
      "loss: 0.546616  [ 3072/ 6954]\n",
      "loss: 0.561223  [ 4096/ 6954]\n",
      "loss: 0.542307  [ 5120/ 6954]\n",
      "loss: 0.523513  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004284 \n",
      "\n",
      "Epoch 425\n",
      "-------------------------------\n",
      "loss: 0.582065  [    0/ 6954]\n",
      "loss: 0.533607  [ 1024/ 6954]\n",
      "loss: 0.528090  [ 2048/ 6954]\n",
      "loss: 0.513337  [ 3072/ 6954]\n",
      "loss: 0.523730  [ 4096/ 6954]\n",
      "loss: 0.559744  [ 5120/ 6954]\n",
      "loss: 0.523387  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 80.9%, Avg loss: 0.004283 \n",
      "\n",
      "Epoch 426\n",
      "-------------------------------\n",
      "loss: 0.561902  [    0/ 6954]\n",
      "loss: 0.543049  [ 1024/ 6954]\n",
      "loss: 0.575456  [ 2048/ 6954]\n",
      "loss: 0.554061  [ 3072/ 6954]\n",
      "loss: 0.536435  [ 4096/ 6954]\n",
      "loss: 0.526980  [ 5120/ 6954]\n",
      "loss: 0.590671  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004282 \n",
      "\n",
      "Epoch 427\n",
      "-------------------------------\n",
      "loss: 0.547260  [    0/ 6954]\n",
      "loss: 0.581603  [ 1024/ 6954]\n",
      "loss: 0.573402  [ 2048/ 6954]\n",
      "loss: 0.533399  [ 3072/ 6954]\n",
      "loss: 0.577169  [ 4096/ 6954]\n",
      "loss: 0.542530  [ 5120/ 6954]\n",
      "loss: 0.550462  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004282 \n",
      "\n",
      "Epoch 428\n",
      "-------------------------------\n",
      "loss: 0.565068  [    0/ 6954]\n",
      "loss: 0.573357  [ 1024/ 6954]\n",
      "loss: 0.586051  [ 2048/ 6954]\n",
      "loss: 0.512639  [ 3072/ 6954]\n",
      "loss: 0.508989  [ 4096/ 6954]\n",
      "loss: 0.554557  [ 5120/ 6954]\n",
      "loss: 0.536920  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004281 \n",
      "\n",
      "Epoch 429\n",
      "-------------------------------\n",
      "loss: 0.516433  [    0/ 6954]\n",
      "loss: 0.540053  [ 1024/ 6954]\n",
      "loss: 0.539702  [ 2048/ 6954]\n",
      "loss: 0.537838  [ 3072/ 6954]\n",
      "loss: 0.601300  [ 4096/ 6954]\n",
      "loss: 0.571820  [ 5120/ 6954]\n",
      "loss: 0.534251  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004280 \n",
      "\n",
      "Epoch 430\n",
      "-------------------------------\n",
      "loss: 0.611359  [    0/ 6954]\n",
      "loss: 0.544695  [ 1024/ 6954]\n",
      "loss: 0.553756  [ 2048/ 6954]\n",
      "loss: 0.562078  [ 3072/ 6954]\n",
      "loss: 0.540526  [ 4096/ 6954]\n",
      "loss: 0.518868  [ 5120/ 6954]\n",
      "loss: 0.566383  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004280 \n",
      "\n",
      "Epoch 431\n",
      "-------------------------------\n",
      "loss: 0.525844  [    0/ 6954]\n",
      "loss: 0.550030  [ 1024/ 6954]\n",
      "loss: 0.521576  [ 2048/ 6954]\n",
      "loss: 0.540599  [ 3072/ 6954]\n",
      "loss: 0.548481  [ 4096/ 6954]\n",
      "loss: 0.538616  [ 5120/ 6954]\n",
      "loss: 0.558133  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004279 \n",
      "\n",
      "Epoch 432\n",
      "-------------------------------\n",
      "loss: 0.516234  [    0/ 6954]\n",
      "loss: 0.597182  [ 1024/ 6954]\n",
      "loss: 0.580751  [ 2048/ 6954]\n",
      "loss: 0.568983  [ 3072/ 6954]\n",
      "loss: 0.535614  [ 4096/ 6954]\n",
      "loss: 0.521306  [ 5120/ 6954]\n",
      "loss: 0.586863  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004278 \n",
      "\n",
      "Epoch 433\n",
      "-------------------------------\n",
      "loss: 0.529126  [    0/ 6954]\n",
      "loss: 0.492845  [ 1024/ 6954]\n",
      "loss: 0.540325  [ 2048/ 6954]\n",
      "loss: 0.560576  [ 3072/ 6954]\n",
      "loss: 0.502643  [ 4096/ 6954]\n",
      "loss: 0.530654  [ 5120/ 6954]\n",
      "loss: 0.541582  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004278 \n",
      "\n",
      "Epoch 434\n",
      "-------------------------------\n",
      "loss: 0.566732  [    0/ 6954]\n",
      "loss: 0.515481  [ 1024/ 6954]\n",
      "loss: 0.550087  [ 2048/ 6954]\n",
      "loss: 0.565271  [ 3072/ 6954]\n",
      "loss: 0.554654  [ 4096/ 6954]\n",
      "loss: 0.514580  [ 5120/ 6954]\n",
      "loss: 0.550423  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004277 \n",
      "\n",
      "Epoch 435\n",
      "-------------------------------\n",
      "loss: 0.529707  [    0/ 6954]\n",
      "loss: 0.580457  [ 1024/ 6954]\n",
      "loss: 0.542048  [ 2048/ 6954]\n",
      "loss: 0.538004  [ 3072/ 6954]\n",
      "loss: 0.561927  [ 4096/ 6954]\n",
      "loss: 0.534851  [ 5120/ 6954]\n",
      "loss: 0.568995  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004276 \n",
      "\n",
      "Epoch 436\n",
      "-------------------------------\n",
      "loss: 0.536376  [    0/ 6954]\n",
      "loss: 0.574598  [ 1024/ 6954]\n",
      "loss: 0.575867  [ 2048/ 6954]\n",
      "loss: 0.550274  [ 3072/ 6954]\n",
      "loss: 0.567401  [ 4096/ 6954]\n",
      "loss: 0.577639  [ 5120/ 6954]\n",
      "loss: 0.563267  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004276 \n",
      "\n",
      "Epoch 437\n",
      "-------------------------------\n",
      "loss: 0.576635  [    0/ 6954]\n",
      "loss: 0.554704  [ 1024/ 6954]\n",
      "loss: 0.579698  [ 2048/ 6954]\n",
      "loss: 0.514116  [ 3072/ 6954]\n",
      "loss: 0.568014  [ 4096/ 6954]\n",
      "loss: 0.523584  [ 5120/ 6954]\n",
      "loss: 0.551980  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004275 \n",
      "\n",
      "Epoch 438\n",
      "-------------------------------\n",
      "loss: 0.544933  [    0/ 6954]\n",
      "loss: 0.526619  [ 1024/ 6954]\n",
      "loss: 0.566194  [ 2048/ 6954]\n",
      "loss: 0.571807  [ 3072/ 6954]\n",
      "loss: 0.526882  [ 4096/ 6954]\n",
      "loss: 0.561174  [ 5120/ 6954]\n",
      "loss: 0.497207  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004274 \n",
      "\n",
      "Epoch 439\n",
      "-------------------------------\n",
      "loss: 0.501636  [    0/ 6954]\n",
      "loss: 0.540027  [ 1024/ 6954]\n",
      "loss: 0.578170  [ 2048/ 6954]\n",
      "loss: 0.562038  [ 3072/ 6954]\n",
      "loss: 0.529739  [ 4096/ 6954]\n",
      "loss: 0.570610  [ 5120/ 6954]\n",
      "loss: 0.587398  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004274 \n",
      "\n",
      "Epoch 440\n",
      "-------------------------------\n",
      "loss: 0.515869  [    0/ 6954]\n",
      "loss: 0.544982  [ 1024/ 6954]\n",
      "loss: 0.565197  [ 2048/ 6954]\n",
      "loss: 0.549195  [ 3072/ 6954]\n",
      "loss: 0.514948  [ 4096/ 6954]\n",
      "loss: 0.553638  [ 5120/ 6954]\n",
      "loss: 0.525511  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004273 \n",
      "\n",
      "Epoch 441\n",
      "-------------------------------\n",
      "loss: 0.544417  [    0/ 6954]\n",
      "loss: 0.543864  [ 1024/ 6954]\n",
      "loss: 0.554604  [ 2048/ 6954]\n",
      "loss: 0.515545  [ 3072/ 6954]\n",
      "loss: 0.535973  [ 4096/ 6954]\n",
      "loss: 0.547477  [ 5120/ 6954]\n",
      "loss: 0.545727  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004273 \n",
      "\n",
      "Epoch 442\n",
      "-------------------------------\n",
      "loss: 0.571302  [    0/ 6954]\n",
      "loss: 0.568573  [ 1024/ 6954]\n",
      "loss: 0.537647  [ 2048/ 6954]\n",
      "loss: 0.549341  [ 3072/ 6954]\n",
      "loss: 0.571852  [ 4096/ 6954]\n",
      "loss: 0.520482  [ 5120/ 6954]\n",
      "loss: 0.524957  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004272 \n",
      "\n",
      "Epoch 443\n",
      "-------------------------------\n",
      "loss: 0.532613  [    0/ 6954]\n",
      "loss: 0.519824  [ 1024/ 6954]\n",
      "loss: 0.545951  [ 2048/ 6954]\n",
      "loss: 0.552767  [ 3072/ 6954]\n",
      "loss: 0.575037  [ 4096/ 6954]\n",
      "loss: 0.540621  [ 5120/ 6954]\n",
      "loss: 0.509024  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004271 \n",
      "\n",
      "Epoch 444\n",
      "-------------------------------\n",
      "loss: 0.538725  [    0/ 6954]\n",
      "loss: 0.548087  [ 1024/ 6954]\n",
      "loss: 0.557482  [ 2048/ 6954]\n",
      "loss: 0.534627  [ 3072/ 6954]\n",
      "loss: 0.553399  [ 4096/ 6954]\n",
      "loss: 0.550010  [ 5120/ 6954]\n",
      "loss: 0.558804  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004271 \n",
      "\n",
      "Epoch 445\n",
      "-------------------------------\n",
      "loss: 0.554553  [    0/ 6954]\n",
      "loss: 0.556579  [ 1024/ 6954]\n",
      "loss: 0.518920  [ 2048/ 6954]\n",
      "loss: 0.551873  [ 3072/ 6954]\n",
      "loss: 0.567057  [ 4096/ 6954]\n",
      "loss: 0.581373  [ 5120/ 6954]\n",
      "loss: 0.538865  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004270 \n",
      "\n",
      "Epoch 446\n",
      "-------------------------------\n",
      "loss: 0.531127  [    0/ 6954]\n",
      "loss: 0.539563  [ 1024/ 6954]\n",
      "loss: 0.547003  [ 2048/ 6954]\n",
      "loss: 0.578953  [ 3072/ 6954]\n",
      "loss: 0.520104  [ 4096/ 6954]\n",
      "loss: 0.538727  [ 5120/ 6954]\n",
      "loss: 0.571316  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004270 \n",
      "\n",
      "Epoch 447\n",
      "-------------------------------\n",
      "loss: 0.581447  [    0/ 6954]\n",
      "loss: 0.542126  [ 1024/ 6954]\n",
      "loss: 0.560030  [ 2048/ 6954]\n",
      "loss: 0.516678  [ 3072/ 6954]\n",
      "loss: 0.586290  [ 4096/ 6954]\n",
      "loss: 0.535209  [ 5120/ 6954]\n",
      "loss: 0.541822  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004269 \n",
      "\n",
      "Epoch 448\n",
      "-------------------------------\n",
      "loss: 0.522528  [    0/ 6954]\n",
      "loss: 0.522653  [ 1024/ 6954]\n",
      "loss: 0.531833  [ 2048/ 6954]\n",
      "loss: 0.581911  [ 3072/ 6954]\n",
      "loss: 0.572834  [ 4096/ 6954]\n",
      "loss: 0.562951  [ 5120/ 6954]\n",
      "loss: 0.556240  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004268 \n",
      "\n",
      "Epoch 449\n",
      "-------------------------------\n",
      "loss: 0.541095  [    0/ 6954]\n",
      "loss: 0.553592  [ 1024/ 6954]\n",
      "loss: 0.552259  [ 2048/ 6954]\n",
      "loss: 0.533903  [ 3072/ 6954]\n",
      "loss: 0.534055  [ 4096/ 6954]\n",
      "loss: 0.582585  [ 5120/ 6954]\n",
      "loss: 0.583230  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004268 \n",
      "\n",
      "Epoch 450\n",
      "-------------------------------\n",
      "loss: 0.533075  [    0/ 6954]\n",
      "loss: 0.536506  [ 1024/ 6954]\n",
      "loss: 0.514542  [ 2048/ 6954]\n",
      "loss: 0.556148  [ 3072/ 6954]\n",
      "loss: 0.533605  [ 4096/ 6954]\n",
      "loss: 0.580474  [ 5120/ 6954]\n",
      "loss: 0.547614  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004267 \n",
      "\n",
      "Epoch 451\n",
      "-------------------------------\n",
      "loss: 0.571427  [    0/ 6954]\n",
      "loss: 0.536110  [ 1024/ 6954]\n",
      "loss: 0.555066  [ 2048/ 6954]\n",
      "loss: 0.513399  [ 3072/ 6954]\n",
      "loss: 0.521064  [ 4096/ 6954]\n",
      "loss: 0.548437  [ 5120/ 6954]\n",
      "loss: 0.516231  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004267 \n",
      "\n",
      "Epoch 452\n",
      "-------------------------------\n",
      "loss: 0.536707  [    0/ 6954]\n",
      "loss: 0.562831  [ 1024/ 6954]\n",
      "loss: 0.576616  [ 2048/ 6954]\n",
      "loss: 0.527221  [ 3072/ 6954]\n",
      "loss: 0.498557  [ 4096/ 6954]\n",
      "loss: 0.543497  [ 5120/ 6954]\n",
      "loss: 0.546924  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004266 \n",
      "\n",
      "Epoch 453\n",
      "-------------------------------\n",
      "loss: 0.553722  [    0/ 6954]\n",
      "loss: 0.580692  [ 1024/ 6954]\n",
      "loss: 0.540290  [ 2048/ 6954]\n",
      "loss: 0.568402  [ 3072/ 6954]\n",
      "loss: 0.565455  [ 4096/ 6954]\n",
      "loss: 0.478139  [ 5120/ 6954]\n",
      "loss: 0.564708  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004265 \n",
      "\n",
      "Epoch 454\n",
      "-------------------------------\n",
      "loss: 0.569262  [    0/ 6954]\n",
      "loss: 0.515751  [ 1024/ 6954]\n",
      "loss: 0.537225  [ 2048/ 6954]\n",
      "loss: 0.559341  [ 3072/ 6954]\n",
      "loss: 0.513118  [ 4096/ 6954]\n",
      "loss: 0.577080  [ 5120/ 6954]\n",
      "loss: 0.547853  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004265 \n",
      "\n",
      "Epoch 455\n",
      "-------------------------------\n",
      "loss: 0.563046  [    0/ 6954]\n",
      "loss: 0.513397  [ 1024/ 6954]\n",
      "loss: 0.534552  [ 2048/ 6954]\n",
      "loss: 0.561698  [ 3072/ 6954]\n",
      "loss: 0.542572  [ 4096/ 6954]\n",
      "loss: 0.507951  [ 5120/ 6954]\n",
      "loss: 0.553856  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004264 \n",
      "\n",
      "Epoch 456\n",
      "-------------------------------\n",
      "loss: 0.541326  [    0/ 6954]\n",
      "loss: 0.540898  [ 1024/ 6954]\n",
      "loss: 0.566511  [ 2048/ 6954]\n",
      "loss: 0.545038  [ 3072/ 6954]\n",
      "loss: 0.526582  [ 4096/ 6954]\n",
      "loss: 0.560807  [ 5120/ 6954]\n",
      "loss: 0.542121  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004264 \n",
      "\n",
      "Epoch 457\n",
      "-------------------------------\n",
      "loss: 0.548289  [    0/ 6954]\n",
      "loss: 0.559655  [ 1024/ 6954]\n",
      "loss: 0.573728  [ 2048/ 6954]\n",
      "loss: 0.486941  [ 3072/ 6954]\n",
      "loss: 0.507020  [ 4096/ 6954]\n",
      "loss: 0.529733  [ 5120/ 6954]\n",
      "loss: 0.506551  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004263 \n",
      "\n",
      "Epoch 458\n",
      "-------------------------------\n",
      "loss: 0.509307  [    0/ 6954]\n",
      "loss: 0.519878  [ 1024/ 6954]\n",
      "loss: 0.545300  [ 2048/ 6954]\n",
      "loss: 0.567131  [ 3072/ 6954]\n",
      "loss: 0.553006  [ 4096/ 6954]\n",
      "loss: 0.548579  [ 5120/ 6954]\n",
      "loss: 0.570118  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004262 \n",
      "\n",
      "Epoch 459\n",
      "-------------------------------\n",
      "loss: 0.553028  [    0/ 6954]\n",
      "loss: 0.523244  [ 1024/ 6954]\n",
      "loss: 0.504066  [ 2048/ 6954]\n",
      "loss: 0.542438  [ 3072/ 6954]\n",
      "loss: 0.538616  [ 4096/ 6954]\n",
      "loss: 0.572294  [ 5120/ 6954]\n",
      "loss: 0.548003  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004262 \n",
      "\n",
      "Epoch 460\n",
      "-------------------------------\n",
      "loss: 0.567297  [    0/ 6954]\n",
      "loss: 0.518164  [ 1024/ 6954]\n",
      "loss: 0.539977  [ 2048/ 6954]\n",
      "loss: 0.539858  [ 3072/ 6954]\n",
      "loss: 0.534905  [ 4096/ 6954]\n",
      "loss: 0.576133  [ 5120/ 6954]\n",
      "loss: 0.553785  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004261 \n",
      "\n",
      "Epoch 461\n",
      "-------------------------------\n",
      "loss: 0.526921  [    0/ 6954]\n",
      "loss: 0.578558  [ 1024/ 6954]\n",
      "loss: 0.507885  [ 2048/ 6954]\n",
      "loss: 0.527873  [ 3072/ 6954]\n",
      "loss: 0.528018  [ 4096/ 6954]\n",
      "loss: 0.515699  [ 5120/ 6954]\n",
      "loss: 0.542971  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004261 \n",
      "\n",
      "Epoch 462\n",
      "-------------------------------\n",
      "loss: 0.473528  [    0/ 6954]\n",
      "loss: 0.504721  [ 1024/ 6954]\n",
      "loss: 0.575627  [ 2048/ 6954]\n",
      "loss: 0.519942  [ 3072/ 6954]\n",
      "loss: 0.499044  [ 4096/ 6954]\n",
      "loss: 0.553024  [ 5120/ 6954]\n",
      "loss: 0.529450  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004260 \n",
      "\n",
      "Epoch 463\n",
      "-------------------------------\n",
      "loss: 0.594422  [    0/ 6954]\n",
      "loss: 0.550315  [ 1024/ 6954]\n",
      "loss: 0.514685  [ 2048/ 6954]\n",
      "loss: 0.539044  [ 3072/ 6954]\n",
      "loss: 0.558237  [ 4096/ 6954]\n",
      "loss: 0.514722  [ 5120/ 6954]\n",
      "loss: 0.506797  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004260 \n",
      "\n",
      "Epoch 464\n",
      "-------------------------------\n",
      "loss: 0.536370  [    0/ 6954]\n",
      "loss: 0.529510  [ 1024/ 6954]\n",
      "loss: 0.490393  [ 2048/ 6954]\n",
      "loss: 0.522142  [ 3072/ 6954]\n",
      "loss: 0.531110  [ 4096/ 6954]\n",
      "loss: 0.545266  [ 5120/ 6954]\n",
      "loss: 0.517887  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004259 \n",
      "\n",
      "Epoch 465\n",
      "-------------------------------\n",
      "loss: 0.525877  [    0/ 6954]\n",
      "loss: 0.541467  [ 1024/ 6954]\n",
      "loss: 0.538071  [ 2048/ 6954]\n",
      "loss: 0.536909  [ 3072/ 6954]\n",
      "loss: 0.531586  [ 4096/ 6954]\n",
      "loss: 0.547069  [ 5120/ 6954]\n",
      "loss: 0.500440  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004259 \n",
      "\n",
      "Epoch 466\n",
      "-------------------------------\n",
      "loss: 0.593216  [    0/ 6954]\n",
      "loss: 0.491209  [ 1024/ 6954]\n",
      "loss: 0.487973  [ 2048/ 6954]\n",
      "loss: 0.575844  [ 3072/ 6954]\n",
      "loss: 0.533928  [ 4096/ 6954]\n",
      "loss: 0.529843  [ 5120/ 6954]\n",
      "loss: 0.574695  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004258 \n",
      "\n",
      "Epoch 467\n",
      "-------------------------------\n",
      "loss: 0.561519  [    0/ 6954]\n",
      "loss: 0.593727  [ 1024/ 6954]\n",
      "loss: 0.524849  [ 2048/ 6954]\n",
      "loss: 0.524481  [ 3072/ 6954]\n",
      "loss: 0.544453  [ 4096/ 6954]\n",
      "loss: 0.564060  [ 5120/ 6954]\n",
      "loss: 0.540296  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004257 \n",
      "\n",
      "Epoch 468\n",
      "-------------------------------\n",
      "loss: 0.560750  [    0/ 6954]\n",
      "loss: 0.574373  [ 1024/ 6954]\n",
      "loss: 0.522608  [ 2048/ 6954]\n",
      "loss: 0.515110  [ 3072/ 6954]\n",
      "loss: 0.548040  [ 4096/ 6954]\n",
      "loss: 0.523324  [ 5120/ 6954]\n",
      "loss: 0.532205  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004257 \n",
      "\n",
      "Epoch 469\n",
      "-------------------------------\n",
      "loss: 0.532430  [    0/ 6954]\n",
      "loss: 0.513544  [ 1024/ 6954]\n",
      "loss: 0.578797  [ 2048/ 6954]\n",
      "loss: 0.538321  [ 3072/ 6954]\n",
      "loss: 0.581814  [ 4096/ 6954]\n",
      "loss: 0.583022  [ 5120/ 6954]\n",
      "loss: 0.529286  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004256 \n",
      "\n",
      "Epoch 470\n",
      "-------------------------------\n",
      "loss: 0.519079  [    0/ 6954]\n",
      "loss: 0.560519  [ 1024/ 6954]\n",
      "loss: 0.556934  [ 2048/ 6954]\n",
      "loss: 0.546212  [ 3072/ 6954]\n",
      "loss: 0.556044  [ 4096/ 6954]\n",
      "loss: 0.552604  [ 5120/ 6954]\n",
      "loss: 0.558277  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004256 \n",
      "\n",
      "Epoch 471\n",
      "-------------------------------\n",
      "loss: 0.543328  [    0/ 6954]\n",
      "loss: 0.571819  [ 1024/ 6954]\n",
      "loss: 0.550337  [ 2048/ 6954]\n",
      "loss: 0.525723  [ 3072/ 6954]\n",
      "loss: 0.525834  [ 4096/ 6954]\n",
      "loss: 0.567677  [ 5120/ 6954]\n",
      "loss: 0.496076  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004255 \n",
      "\n",
      "Epoch 472\n",
      "-------------------------------\n",
      "loss: 0.568172  [    0/ 6954]\n",
      "loss: 0.520626  [ 1024/ 6954]\n",
      "loss: 0.545171  [ 2048/ 6954]\n",
      "loss: 0.528984  [ 3072/ 6954]\n",
      "loss: 0.546179  [ 4096/ 6954]\n",
      "loss: 0.500481  [ 5120/ 6954]\n",
      "loss: 0.544407  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004255 \n",
      "\n",
      "Epoch 473\n",
      "-------------------------------\n",
      "loss: 0.550837  [    0/ 6954]\n",
      "loss: 0.516562  [ 1024/ 6954]\n",
      "loss: 0.552409  [ 2048/ 6954]\n",
      "loss: 0.555580  [ 3072/ 6954]\n",
      "loss: 0.531425  [ 4096/ 6954]\n",
      "loss: 0.540335  [ 5120/ 6954]\n",
      "loss: 0.497487  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004254 \n",
      "\n",
      "Epoch 474\n",
      "-------------------------------\n",
      "loss: 0.523924  [    0/ 6954]\n",
      "loss: 0.567674  [ 1024/ 6954]\n",
      "loss: 0.548989  [ 2048/ 6954]\n",
      "loss: 0.539230  [ 3072/ 6954]\n",
      "loss: 0.549465  [ 4096/ 6954]\n",
      "loss: 0.485049  [ 5120/ 6954]\n",
      "loss: 0.586253  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.0%, Avg loss: 0.004254 \n",
      "\n",
      "Epoch 475\n",
      "-------------------------------\n",
      "loss: 0.555380  [    0/ 6954]\n",
      "loss: 0.542665  [ 1024/ 6954]\n",
      "loss: 0.565558  [ 2048/ 6954]\n",
      "loss: 0.535435  [ 3072/ 6954]\n",
      "loss: 0.515042  [ 4096/ 6954]\n",
      "loss: 0.542159  [ 5120/ 6954]\n",
      "loss: 0.527821  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004253 \n",
      "\n",
      "Epoch 476\n",
      "-------------------------------\n",
      "loss: 0.490782  [    0/ 6954]\n",
      "loss: 0.576507  [ 1024/ 6954]\n",
      "loss: 0.547204  [ 2048/ 6954]\n",
      "loss: 0.522295  [ 3072/ 6954]\n",
      "loss: 0.509163  [ 4096/ 6954]\n",
      "loss: 0.559778  [ 5120/ 6954]\n",
      "loss: 0.530516  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004253 \n",
      "\n",
      "Epoch 477\n",
      "-------------------------------\n",
      "loss: 0.538433  [    0/ 6954]\n",
      "loss: 0.517889  [ 1024/ 6954]\n",
      "loss: 0.579174  [ 2048/ 6954]\n",
      "loss: 0.502699  [ 3072/ 6954]\n",
      "loss: 0.535052  [ 4096/ 6954]\n",
      "loss: 0.522055  [ 5120/ 6954]\n",
      "loss: 0.543073  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004252 \n",
      "\n",
      "Epoch 478\n",
      "-------------------------------\n",
      "loss: 0.515978  [    0/ 6954]\n",
      "loss: 0.553112  [ 1024/ 6954]\n",
      "loss: 0.549677  [ 2048/ 6954]\n",
      "loss: 0.593345  [ 3072/ 6954]\n",
      "loss: 0.600540  [ 4096/ 6954]\n",
      "loss: 0.499519  [ 5120/ 6954]\n",
      "loss: 0.536449  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004252 \n",
      "\n",
      "Epoch 479\n",
      "-------------------------------\n",
      "loss: 0.510715  [    0/ 6954]\n",
      "loss: 0.522077  [ 1024/ 6954]\n",
      "loss: 0.552803  [ 2048/ 6954]\n",
      "loss: 0.551162  [ 3072/ 6954]\n",
      "loss: 0.554272  [ 4096/ 6954]\n",
      "loss: 0.543003  [ 5120/ 6954]\n",
      "loss: 0.533981  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004251 \n",
      "\n",
      "Epoch 480\n",
      "-------------------------------\n",
      "loss: 0.552109  [    0/ 6954]\n",
      "loss: 0.596240  [ 1024/ 6954]\n",
      "loss: 0.574675  [ 2048/ 6954]\n",
      "loss: 0.538671  [ 3072/ 6954]\n",
      "loss: 0.524308  [ 4096/ 6954]\n",
      "loss: 0.563472  [ 5120/ 6954]\n",
      "loss: 0.558459  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004251 \n",
      "\n",
      "Epoch 481\n",
      "-------------------------------\n",
      "loss: 0.536291  [    0/ 6954]\n",
      "loss: 0.517006  [ 1024/ 6954]\n",
      "loss: 0.518304  [ 2048/ 6954]\n",
      "loss: 0.543154  [ 3072/ 6954]\n",
      "loss: 0.533940  [ 4096/ 6954]\n",
      "loss: 0.549762  [ 5120/ 6954]\n",
      "loss: 0.535949  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004250 \n",
      "\n",
      "Epoch 482\n",
      "-------------------------------\n",
      "loss: 0.563419  [    0/ 6954]\n",
      "loss: 0.494293  [ 1024/ 6954]\n",
      "loss: 0.520024  [ 2048/ 6954]\n",
      "loss: 0.512268  [ 3072/ 6954]\n",
      "loss: 0.535685  [ 4096/ 6954]\n",
      "loss: 0.577795  [ 5120/ 6954]\n",
      "loss: 0.513953  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004249 \n",
      "\n",
      "Epoch 483\n",
      "-------------------------------\n",
      "loss: 0.573500  [    0/ 6954]\n",
      "loss: 0.555521  [ 1024/ 6954]\n",
      "loss: 0.553014  [ 2048/ 6954]\n",
      "loss: 0.525630  [ 3072/ 6954]\n",
      "loss: 0.565852  [ 4096/ 6954]\n",
      "loss: 0.511920  [ 5120/ 6954]\n",
      "loss: 0.536546  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004249 \n",
      "\n",
      "Epoch 484\n",
      "-------------------------------\n",
      "loss: 0.519149  [    0/ 6954]\n",
      "loss: 0.607455  [ 1024/ 6954]\n",
      "loss: 0.564285  [ 2048/ 6954]\n",
      "loss: 0.521664  [ 3072/ 6954]\n",
      "loss: 0.539012  [ 4096/ 6954]\n",
      "loss: 0.531185  [ 5120/ 6954]\n",
      "loss: 0.518320  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004248 \n",
      "\n",
      "Epoch 485\n",
      "-------------------------------\n",
      "loss: 0.570807  [    0/ 6954]\n",
      "loss: 0.542412  [ 1024/ 6954]\n",
      "loss: 0.534909  [ 2048/ 6954]\n",
      "loss: 0.545915  [ 3072/ 6954]\n",
      "loss: 0.528852  [ 4096/ 6954]\n",
      "loss: 0.549786  [ 5120/ 6954]\n",
      "loss: 0.502911  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004248 \n",
      "\n",
      "Epoch 486\n",
      "-------------------------------\n",
      "loss: 0.556264  [    0/ 6954]\n",
      "loss: 0.542602  [ 1024/ 6954]\n",
      "loss: 0.501070  [ 2048/ 6954]\n",
      "loss: 0.505905  [ 3072/ 6954]\n",
      "loss: 0.484703  [ 4096/ 6954]\n",
      "loss: 0.516911  [ 5120/ 6954]\n",
      "loss: 0.584598  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004247 \n",
      "\n",
      "Epoch 487\n",
      "-------------------------------\n",
      "loss: 0.540766  [    0/ 6954]\n",
      "loss: 0.574749  [ 1024/ 6954]\n",
      "loss: 0.517224  [ 2048/ 6954]\n",
      "loss: 0.504014  [ 3072/ 6954]\n",
      "loss: 0.533646  [ 4096/ 6954]\n",
      "loss: 0.491123  [ 5120/ 6954]\n",
      "loss: 0.574997  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004247 \n",
      "\n",
      "Epoch 488\n",
      "-------------------------------\n",
      "loss: 0.538597  [    0/ 6954]\n",
      "loss: 0.536660  [ 1024/ 6954]\n",
      "loss: 0.531512  [ 2048/ 6954]\n",
      "loss: 0.544052  [ 3072/ 6954]\n",
      "loss: 0.598208  [ 4096/ 6954]\n",
      "loss: 0.542975  [ 5120/ 6954]\n",
      "loss: 0.529193  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004246 \n",
      "\n",
      "Epoch 489\n",
      "-------------------------------\n",
      "loss: 0.518038  [    0/ 6954]\n",
      "loss: 0.552665  [ 1024/ 6954]\n",
      "loss: 0.572645  [ 2048/ 6954]\n",
      "loss: 0.557410  [ 3072/ 6954]\n",
      "loss: 0.506374  [ 4096/ 6954]\n",
      "loss: 0.538774  [ 5120/ 6954]\n",
      "loss: 0.505995  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004246 \n",
      "\n",
      "Epoch 490\n",
      "-------------------------------\n",
      "loss: 0.495697  [    0/ 6954]\n",
      "loss: 0.549570  [ 1024/ 6954]\n",
      "loss: 0.529334  [ 2048/ 6954]\n",
      "loss: 0.505776  [ 3072/ 6954]\n",
      "loss: 0.525663  [ 4096/ 6954]\n",
      "loss: 0.544961  [ 5120/ 6954]\n",
      "loss: 0.492533  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004245 \n",
      "\n",
      "Epoch 491\n",
      "-------------------------------\n",
      "loss: 0.520735  [    0/ 6954]\n",
      "loss: 0.557443  [ 1024/ 6954]\n",
      "loss: 0.540663  [ 2048/ 6954]\n",
      "loss: 0.562305  [ 3072/ 6954]\n",
      "loss: 0.537360  [ 4096/ 6954]\n",
      "loss: 0.526524  [ 5120/ 6954]\n",
      "loss: 0.514987  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004245 \n",
      "\n",
      "Epoch 492\n",
      "-------------------------------\n",
      "loss: 0.552896  [    0/ 6954]\n",
      "loss: 0.525465  [ 1024/ 6954]\n",
      "loss: 0.560382  [ 2048/ 6954]\n",
      "loss: 0.531618  [ 3072/ 6954]\n",
      "loss: 0.561775  [ 4096/ 6954]\n",
      "loss: 0.560712  [ 5120/ 6954]\n",
      "loss: 0.512187  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004244 \n",
      "\n",
      "Epoch 493\n",
      "-------------------------------\n",
      "loss: 0.494463  [    0/ 6954]\n",
      "loss: 0.572170  [ 1024/ 6954]\n",
      "loss: 0.578904  [ 2048/ 6954]\n",
      "loss: 0.553885  [ 3072/ 6954]\n",
      "loss: 0.575525  [ 4096/ 6954]\n",
      "loss: 0.533923  [ 5120/ 6954]\n",
      "loss: 0.516178  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004244 \n",
      "\n",
      "Epoch 494\n",
      "-------------------------------\n",
      "loss: 0.550397  [    0/ 6954]\n",
      "loss: 0.511327  [ 1024/ 6954]\n",
      "loss: 0.513949  [ 2048/ 6954]\n",
      "loss: 0.530826  [ 3072/ 6954]\n",
      "loss: 0.524146  [ 4096/ 6954]\n",
      "loss: 0.537552  [ 5120/ 6954]\n",
      "loss: 0.525651  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004243 \n",
      "\n",
      "Epoch 495\n",
      "-------------------------------\n",
      "loss: 0.601765  [    0/ 6954]\n",
      "loss: 0.614137  [ 1024/ 6954]\n",
      "loss: 0.548735  [ 2048/ 6954]\n",
      "loss: 0.579609  [ 3072/ 6954]\n",
      "loss: 0.543142  [ 4096/ 6954]\n",
      "loss: 0.575034  [ 5120/ 6954]\n",
      "loss: 0.554994  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004243 \n",
      "\n",
      "Epoch 496\n",
      "-------------------------------\n",
      "loss: 0.570105  [    0/ 6954]\n",
      "loss: 0.544297  [ 1024/ 6954]\n",
      "loss: 0.536781  [ 2048/ 6954]\n",
      "loss: 0.481706  [ 3072/ 6954]\n",
      "loss: 0.565229  [ 4096/ 6954]\n",
      "loss: 0.538457  [ 5120/ 6954]\n",
      "loss: 0.504970  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004243 \n",
      "\n",
      "Epoch 497\n",
      "-------------------------------\n",
      "loss: 0.525136  [    0/ 6954]\n",
      "loss: 0.527923  [ 1024/ 6954]\n",
      "loss: 0.560396  [ 2048/ 6954]\n",
      "loss: 0.542464  [ 3072/ 6954]\n",
      "loss: 0.539736  [ 4096/ 6954]\n",
      "loss: 0.530825  [ 5120/ 6954]\n",
      "loss: 0.567811  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004242 \n",
      "\n",
      "Epoch 498\n",
      "-------------------------------\n",
      "loss: 0.567840  [    0/ 6954]\n",
      "loss: 0.601182  [ 1024/ 6954]\n",
      "loss: 0.531639  [ 2048/ 6954]\n",
      "loss: 0.541309  [ 3072/ 6954]\n",
      "loss: 0.500257  [ 4096/ 6954]\n",
      "loss: 0.511716  [ 5120/ 6954]\n",
      "loss: 0.555258  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004242 \n",
      "\n",
      "Epoch 499\n",
      "-------------------------------\n",
      "loss: 0.552496  [    0/ 6954]\n",
      "loss: 0.493035  [ 1024/ 6954]\n",
      "loss: 0.505163  [ 2048/ 6954]\n",
      "loss: 0.542855  [ 3072/ 6954]\n",
      "loss: 0.573604  [ 4096/ 6954]\n",
      "loss: 0.555087  [ 5120/ 6954]\n",
      "loss: 0.529485  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004241 \n",
      "\n",
      "Epoch 500\n",
      "-------------------------------\n",
      "loss: 0.531836  [    0/ 6954]\n",
      "loss: 0.523113  [ 1024/ 6954]\n",
      "loss: 0.551459  [ 2048/ 6954]\n",
      "loss: 0.553440  [ 3072/ 6954]\n",
      "loss: 0.512269  [ 4096/ 6954]\n",
      "loss: 0.552335  [ 5120/ 6954]\n",
      "loss: 0.521416  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004241 \n",
      "\n",
      "Epoch 501\n",
      "-------------------------------\n",
      "loss: 0.512213  [    0/ 6954]\n",
      "loss: 0.565588  [ 1024/ 6954]\n",
      "loss: 0.514162  [ 2048/ 6954]\n",
      "loss: 0.583202  [ 3072/ 6954]\n",
      "loss: 0.525387  [ 4096/ 6954]\n",
      "loss: 0.546052  [ 5120/ 6954]\n",
      "loss: 0.523271  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004240 \n",
      "\n",
      "Epoch 502\n",
      "-------------------------------\n",
      "loss: 0.507479  [    0/ 6954]\n",
      "loss: 0.487901  [ 1024/ 6954]\n",
      "loss: 0.488832  [ 2048/ 6954]\n",
      "loss: 0.519206  [ 3072/ 6954]\n",
      "loss: 0.578995  [ 4096/ 6954]\n",
      "loss: 0.511102  [ 5120/ 6954]\n",
      "loss: 0.568235  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004240 \n",
      "\n",
      "Epoch 503\n",
      "-------------------------------\n",
      "loss: 0.574736  [    0/ 6954]\n",
      "loss: 0.546819  [ 1024/ 6954]\n",
      "loss: 0.572214  [ 2048/ 6954]\n",
      "loss: 0.527092  [ 3072/ 6954]\n",
      "loss: 0.557121  [ 4096/ 6954]\n",
      "loss: 0.513024  [ 5120/ 6954]\n",
      "loss: 0.587593  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004239 \n",
      "\n",
      "Epoch 504\n",
      "-------------------------------\n",
      "loss: 0.533666  [    0/ 6954]\n",
      "loss: 0.509350  [ 1024/ 6954]\n",
      "loss: 0.519696  [ 2048/ 6954]\n",
      "loss: 0.516302  [ 3072/ 6954]\n",
      "loss: 0.521582  [ 4096/ 6954]\n",
      "loss: 0.594129  [ 5120/ 6954]\n",
      "loss: 0.597974  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004239 \n",
      "\n",
      "Epoch 505\n",
      "-------------------------------\n",
      "loss: 0.517392  [    0/ 6954]\n",
      "loss: 0.543787  [ 1024/ 6954]\n",
      "loss: 0.532353  [ 2048/ 6954]\n",
      "loss: 0.533350  [ 3072/ 6954]\n",
      "loss: 0.507745  [ 4096/ 6954]\n",
      "loss: 0.551153  [ 5120/ 6954]\n",
      "loss: 0.581562  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004238 \n",
      "\n",
      "Epoch 506\n",
      "-------------------------------\n",
      "loss: 0.576672  [    0/ 6954]\n",
      "loss: 0.543048  [ 1024/ 6954]\n",
      "loss: 0.524875  [ 2048/ 6954]\n",
      "loss: 0.534801  [ 3072/ 6954]\n",
      "loss: 0.565040  [ 4096/ 6954]\n",
      "loss: 0.532097  [ 5120/ 6954]\n",
      "loss: 0.549569  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004238 \n",
      "\n",
      "Epoch 507\n",
      "-------------------------------\n",
      "loss: 0.550551  [    0/ 6954]\n",
      "loss: 0.556505  [ 1024/ 6954]\n",
      "loss: 0.502686  [ 2048/ 6954]\n",
      "loss: 0.575326  [ 3072/ 6954]\n",
      "loss: 0.528273  [ 4096/ 6954]\n",
      "loss: 0.545025  [ 5120/ 6954]\n",
      "loss: 0.519540  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004237 \n",
      "\n",
      "Epoch 508\n",
      "-------------------------------\n",
      "loss: 0.579280  [    0/ 6954]\n",
      "loss: 0.526602  [ 1024/ 6954]\n",
      "loss: 0.552894  [ 2048/ 6954]\n",
      "loss: 0.585518  [ 3072/ 6954]\n",
      "loss: 0.528489  [ 4096/ 6954]\n",
      "loss: 0.480202  [ 5120/ 6954]\n",
      "loss: 0.472357  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004237 \n",
      "\n",
      "Epoch 509\n",
      "-------------------------------\n",
      "loss: 0.517060  [    0/ 6954]\n",
      "loss: 0.512921  [ 1024/ 6954]\n",
      "loss: 0.518735  [ 2048/ 6954]\n",
      "loss: 0.558291  [ 3072/ 6954]\n",
      "loss: 0.552163  [ 4096/ 6954]\n",
      "loss: 0.558203  [ 5120/ 6954]\n",
      "loss: 0.520573  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004236 \n",
      "\n",
      "Epoch 510\n",
      "-------------------------------\n",
      "loss: 0.502836  [    0/ 6954]\n",
      "loss: 0.523573  [ 1024/ 6954]\n",
      "loss: 0.528265  [ 2048/ 6954]\n",
      "loss: 0.558701  [ 3072/ 6954]\n",
      "loss: 0.542682  [ 4096/ 6954]\n",
      "loss: 0.568958  [ 5120/ 6954]\n",
      "loss: 0.511225  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004236 \n",
      "\n",
      "Epoch 511\n",
      "-------------------------------\n",
      "loss: 0.575820  [    0/ 6954]\n",
      "loss: 0.477269  [ 1024/ 6954]\n",
      "loss: 0.501137  [ 2048/ 6954]\n",
      "loss: 0.545465  [ 3072/ 6954]\n",
      "loss: 0.552009  [ 4096/ 6954]\n",
      "loss: 0.516355  [ 5120/ 6954]\n",
      "loss: 0.508185  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004235 \n",
      "\n",
      "Epoch 512\n",
      "-------------------------------\n",
      "loss: 0.524680  [    0/ 6954]\n",
      "loss: 0.572272  [ 1024/ 6954]\n",
      "loss: 0.541166  [ 2048/ 6954]\n",
      "loss: 0.522915  [ 3072/ 6954]\n",
      "loss: 0.525721  [ 4096/ 6954]\n",
      "loss: 0.556940  [ 5120/ 6954]\n",
      "loss: 0.526050  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004235 \n",
      "\n",
      "Epoch 513\n",
      "-------------------------------\n",
      "loss: 0.516946  [    0/ 6954]\n",
      "loss: 0.539776  [ 1024/ 6954]\n",
      "loss: 0.503509  [ 2048/ 6954]\n",
      "loss: 0.558619  [ 3072/ 6954]\n",
      "loss: 0.498520  [ 4096/ 6954]\n",
      "loss: 0.517576  [ 5120/ 6954]\n",
      "loss: 0.592866  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004235 \n",
      "\n",
      "Epoch 514\n",
      "-------------------------------\n",
      "loss: 0.574879  [    0/ 6954]\n",
      "loss: 0.537071  [ 1024/ 6954]\n",
      "loss: 0.513927  [ 2048/ 6954]\n",
      "loss: 0.554904  [ 3072/ 6954]\n",
      "loss: 0.551923  [ 4096/ 6954]\n",
      "loss: 0.547686  [ 5120/ 6954]\n",
      "loss: 0.512368  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004234 \n",
      "\n",
      "Epoch 515\n",
      "-------------------------------\n",
      "loss: 0.538545  [    0/ 6954]\n",
      "loss: 0.538864  [ 1024/ 6954]\n",
      "loss: 0.586318  [ 2048/ 6954]\n",
      "loss: 0.559833  [ 3072/ 6954]\n",
      "loss: 0.543799  [ 4096/ 6954]\n",
      "loss: 0.536294  [ 5120/ 6954]\n",
      "loss: 0.494796  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004234 \n",
      "\n",
      "Epoch 516\n",
      "-------------------------------\n",
      "loss: 0.563715  [    0/ 6954]\n",
      "loss: 0.506123  [ 1024/ 6954]\n",
      "loss: 0.559234  [ 2048/ 6954]\n",
      "loss: 0.518148  [ 3072/ 6954]\n",
      "loss: 0.546321  [ 4096/ 6954]\n",
      "loss: 0.565750  [ 5120/ 6954]\n",
      "loss: 0.541247  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004233 \n",
      "\n",
      "Epoch 517\n",
      "-------------------------------\n",
      "loss: 0.518644  [    0/ 6954]\n",
      "loss: 0.542048  [ 1024/ 6954]\n",
      "loss: 0.497842  [ 2048/ 6954]\n",
      "loss: 0.481139  [ 3072/ 6954]\n",
      "loss: 0.533316  [ 4096/ 6954]\n",
      "loss: 0.488736  [ 5120/ 6954]\n",
      "loss: 0.557478  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004233 \n",
      "\n",
      "Epoch 518\n",
      "-------------------------------\n",
      "loss: 0.554411  [    0/ 6954]\n",
      "loss: 0.532842  [ 1024/ 6954]\n",
      "loss: 0.547647  [ 2048/ 6954]\n",
      "loss: 0.538200  [ 3072/ 6954]\n",
      "loss: 0.496540  [ 4096/ 6954]\n",
      "loss: 0.539850  [ 5120/ 6954]\n",
      "loss: 0.515263  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004232 \n",
      "\n",
      "Epoch 519\n",
      "-------------------------------\n",
      "loss: 0.523537  [    0/ 6954]\n",
      "loss: 0.565273  [ 1024/ 6954]\n",
      "loss: 0.513575  [ 2048/ 6954]\n",
      "loss: 0.558703  [ 3072/ 6954]\n",
      "loss: 0.547212  [ 4096/ 6954]\n",
      "loss: 0.568518  [ 5120/ 6954]\n",
      "loss: 0.518517  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004232 \n",
      "\n",
      "Epoch 520\n",
      "-------------------------------\n",
      "loss: 0.525605  [    0/ 6954]\n",
      "loss: 0.610029  [ 1024/ 6954]\n",
      "loss: 0.548352  [ 2048/ 6954]\n",
      "loss: 0.553104  [ 3072/ 6954]\n",
      "loss: 0.495287  [ 4096/ 6954]\n",
      "loss: 0.540829  [ 5120/ 6954]\n",
      "loss: 0.539379  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004232 \n",
      "\n",
      "Epoch 521\n",
      "-------------------------------\n",
      "loss: 0.523488  [    0/ 6954]\n",
      "loss: 0.555802  [ 1024/ 6954]\n",
      "loss: 0.505795  [ 2048/ 6954]\n",
      "loss: 0.546533  [ 3072/ 6954]\n",
      "loss: 0.620048  [ 4096/ 6954]\n",
      "loss: 0.530125  [ 5120/ 6954]\n",
      "loss: 0.528749  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004231 \n",
      "\n",
      "Epoch 522\n",
      "-------------------------------\n",
      "loss: 0.565524  [    0/ 6954]\n",
      "loss: 0.548177  [ 1024/ 6954]\n",
      "loss: 0.533154  [ 2048/ 6954]\n",
      "loss: 0.541722  [ 3072/ 6954]\n",
      "loss: 0.526828  [ 4096/ 6954]\n",
      "loss: 0.555771  [ 5120/ 6954]\n",
      "loss: 0.501774  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004231 \n",
      "\n",
      "Epoch 523\n",
      "-------------------------------\n",
      "loss: 0.547821  [    0/ 6954]\n",
      "loss: 0.552114  [ 1024/ 6954]\n",
      "loss: 0.503551  [ 2048/ 6954]\n",
      "loss: 0.519441  [ 3072/ 6954]\n",
      "loss: 0.540997  [ 4096/ 6954]\n",
      "loss: 0.591419  [ 5120/ 6954]\n",
      "loss: 0.489534  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004230 \n",
      "\n",
      "Epoch 524\n",
      "-------------------------------\n",
      "loss: 0.530637  [    0/ 6954]\n",
      "loss: 0.490758  [ 1024/ 6954]\n",
      "loss: 0.522100  [ 2048/ 6954]\n",
      "loss: 0.512522  [ 3072/ 6954]\n",
      "loss: 0.525131  [ 4096/ 6954]\n",
      "loss: 0.503592  [ 5120/ 6954]\n",
      "loss: 0.577762  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004230 \n",
      "\n",
      "Epoch 525\n",
      "-------------------------------\n",
      "loss: 0.515275  [    0/ 6954]\n",
      "loss: 0.524974  [ 1024/ 6954]\n",
      "loss: 0.553033  [ 2048/ 6954]\n",
      "loss: 0.526993  [ 3072/ 6954]\n",
      "loss: 0.529338  [ 4096/ 6954]\n",
      "loss: 0.547019  [ 5120/ 6954]\n",
      "loss: 0.478788  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004230 \n",
      "\n",
      "Epoch 526\n",
      "-------------------------------\n",
      "loss: 0.560754  [    0/ 6954]\n",
      "loss: 0.574929  [ 1024/ 6954]\n",
      "loss: 0.534547  [ 2048/ 6954]\n",
      "loss: 0.558783  [ 3072/ 6954]\n",
      "loss: 0.484416  [ 4096/ 6954]\n",
      "loss: 0.531527  [ 5120/ 6954]\n",
      "loss: 0.525028  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004229 \n",
      "\n",
      "Epoch 527\n",
      "-------------------------------\n",
      "loss: 0.526695  [    0/ 6954]\n",
      "loss: 0.564448  [ 1024/ 6954]\n",
      "loss: 0.531694  [ 2048/ 6954]\n",
      "loss: 0.536658  [ 3072/ 6954]\n",
      "loss: 0.499027  [ 4096/ 6954]\n",
      "loss: 0.552119  [ 5120/ 6954]\n",
      "loss: 0.522653  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004229 \n",
      "\n",
      "Epoch 528\n",
      "-------------------------------\n",
      "loss: 0.553318  [    0/ 6954]\n",
      "loss: 0.533261  [ 1024/ 6954]\n",
      "loss: 0.564624  [ 2048/ 6954]\n",
      "loss: 0.537102  [ 3072/ 6954]\n",
      "loss: 0.541188  [ 4096/ 6954]\n",
      "loss: 0.550618  [ 5120/ 6954]\n",
      "loss: 0.568933  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004228 \n",
      "\n",
      "Epoch 529\n",
      "-------------------------------\n",
      "loss: 0.518525  [    0/ 6954]\n",
      "loss: 0.552850  [ 1024/ 6954]\n",
      "loss: 0.547035  [ 2048/ 6954]\n",
      "loss: 0.550002  [ 3072/ 6954]\n",
      "loss: 0.552038  [ 4096/ 6954]\n",
      "loss: 0.527723  [ 5120/ 6954]\n",
      "loss: 0.515034  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004228 \n",
      "\n",
      "Epoch 530\n",
      "-------------------------------\n",
      "loss: 0.542699  [    0/ 6954]\n",
      "loss: 0.558036  [ 1024/ 6954]\n",
      "loss: 0.536749  [ 2048/ 6954]\n",
      "loss: 0.531523  [ 3072/ 6954]\n",
      "loss: 0.499046  [ 4096/ 6954]\n",
      "loss: 0.566490  [ 5120/ 6954]\n",
      "loss: 0.567749  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004228 \n",
      "\n",
      "Epoch 531\n",
      "-------------------------------\n",
      "loss: 0.571683  [    0/ 6954]\n",
      "loss: 0.526705  [ 1024/ 6954]\n",
      "loss: 0.516225  [ 2048/ 6954]\n",
      "loss: 0.542624  [ 3072/ 6954]\n",
      "loss: 0.504024  [ 4096/ 6954]\n",
      "loss: 0.545391  [ 5120/ 6954]\n",
      "loss: 0.578269  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004227 \n",
      "\n",
      "Epoch 532\n",
      "-------------------------------\n",
      "loss: 0.558252  [    0/ 6954]\n",
      "loss: 0.509782  [ 1024/ 6954]\n",
      "loss: 0.506378  [ 2048/ 6954]\n",
      "loss: 0.488417  [ 3072/ 6954]\n",
      "loss: 0.534408  [ 4096/ 6954]\n",
      "loss: 0.573703  [ 5120/ 6954]\n",
      "loss: 0.490524  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004227 \n",
      "\n",
      "Epoch 533\n",
      "-------------------------------\n",
      "loss: 0.610767  [    0/ 6954]\n",
      "loss: 0.545732  [ 1024/ 6954]\n",
      "loss: 0.521032  [ 2048/ 6954]\n",
      "loss: 0.575810  [ 3072/ 6954]\n",
      "loss: 0.555031  [ 4096/ 6954]\n",
      "loss: 0.528887  [ 5120/ 6954]\n",
      "loss: 0.554050  [ 6144/ 6954]\n",
      "Test Error: \n",
      " Accuracy: 81.1%, Avg loss: 0.004227 \n",
      "\n",
      "Epoch 534\n",
      "-------------------------------\n",
      "loss: 0.534358  [    0/ 6954]\n",
      "loss: 0.535381  [ 1024/ 6954]\n"
     ]
    }
   ],
   "source": [
    "epochs = 1000\n",
    "for t in range(epochs):\n",
    "    print(f\"Epoch {t+1}\\n-------------------------------\")\n",
    "    train(train_dataloader, model, loss_fn, optimizer)\n",
    "    test(test_dataloader, model)\n",
    "print(\"Done!\")"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n",
     "is_executing": true
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "outputs": [],
   "source": [
    "torch.save(model,'model/m0')"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n",
     "is_executing": true
    }
   }
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}