DOAT_Slices_deeplab-ResNet50.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# DOTA 使用deeplab做15类的异常检测, ResNet50对error_map分类\n",
    "\n",
    "## 输出3*15通道 [x0,x1,...,x14], 查看wx 与  [x0,x1,...,x14]之间的距离"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "import torchvision\n",
    "from torchvision import datasets, transforms\n",
    "from torch.autograd import Variable\n",
    "import os\n",
    "import matplotlib.pyplot as plt\n",
    "import datetime\n",
    "import time\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 载入数据"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "#data_dir = r\"D:/datasets/ShipVSField\"\n",
    "data_dir = r\"D:\\datasets\\dota\\Slices\"\n",
    "data_transform = {x:transforms.Compose([transforms.Resize([256,256]),\n",
    "                                        transforms.ToTensor(),\n",
    "                                        transforms.Normalize(mean=[0.5]*3, std=[0.5]*3)])\n",
    "                                      for x in [\"train\"] } #, \"valid\"\n",
    "\n",
    "# 删除隐藏文件夹 .ipynb_checkpoints\n",
    "for x in [\"train\", \"valid\"]:\n",
    "    if os.path.exists( os.path.join(data_dir,x,\".ipynb_checkpoints\") ):\n",
    "        print(u\"Delete \".format(os.path.join(data_dir,x,\".ipynb_checkpoints\") ))\n",
    "        os.removedirs(os.path.join(data_dir,x,\".ipynb_checkpoints\"))\n",
    "    \n",
    "image_datasets = {x:datasets.ImageFolder(root = os.path.join(data_dir,x),\n",
    "                                        transform = data_transform[x]) for x in [\"train\"]} #, \"valid\"\n",
    "\n",
    "dataloader = {x:torch.utils.data.DataLoader(dataset = image_datasets[x],\n",
    "                                           batch_size = 8, #16,\n",
    "                                           shuffle = True) for x in [\"train\"]}#, \"valid\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'baseball-diamond': 0, 'basketball-court': 1, 'bridge': 2, 'ground-track-field': 3, 'harbor': 4, 'helicopter': 5, 'large-vehicle': 6, 'plane': 7, 'roundabout': 8, 'ship': 9, 'small-vehicle': 10, 'soccer-ball-field': 11, 'storage-tank': 12, 'swimming-pool': 13, 'tennis-court': 14}\n"
     ]
    }
   ],
   "source": [
    "X_example,y_example = next(iter( dataloader[\"train\"] ) )\n",
    "example_classes = image_datasets[\"train\"].classes\n",
    "index_classes = image_datasets[\"train\"].class_to_idx\n",
    "cls_num = len(index_classes)\n",
    "\n",
    "print(index_classes)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 配置模型"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# deeplabv3_resnet50\n",
    "num_classes = 3 * 15 # 15类\n",
    "#model_fcn = torchvision.models.segmentation.fcn_resnet50(pretrained=False, progress=True, num_classes=num_classes, aux_loss=None)\n",
    "model_deeplab = torchvision.models.segmentation.deeplabv3_resnet50(pretrained=False, progress=True, \n",
    "                                                                   num_classes=num_classes, aux_loss=None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Resnet50 分类\n",
    "model_res = torchvision.models.resnet50(pretrained= False)\n",
    "state_dict = torch.load(\"resnet50-19c8e357.pth\")\n",
    "model_res.load_state_dict(state_dict)\n",
    "\n",
    "# 修改输入输出通道数\n",
    "model_res.fc = torch.nn.Linear(2048, 15)\n",
    "model_res.conv1 = torch.nn.Conv2d(3*15,64, kernel_size=7,padding=3,bias=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<All keys matched successfully>"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#加载模型参数\n",
    "model_res.load_state_dict(torch.load('DOAT_Slices_deeplab-ResNet50_model_last.pth')['model_res'])\n",
    "model_deeplab.load_state_dict(torch.load('DOAT_Slices_deeplab-ResNet50_model_last.pth')['model_deeplab'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([8, 45, 256, 256])\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(tensor(192299.3906, grad_fn=<SumBackward0>),)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_fcn = model_deeplab(X_example )['out'] # * masked_gaussian\n",
    "print(y_fcn.shape)\n",
    "\n",
    "cls_num = len(index_classes)\n",
    "\n",
    "Classes = np.random.randint(low=0, high=cls_num, size=len(X_example))\n",
    "y_class = torch.cat( [y_fcn[i:i+1,np.array(range(0,3))+3*Classes[i],:,:] for i in range(len(Classes))], axis=0)\n",
    "\n",
    "torch.pow(X_example - y_class, 2).sum(), \n",
    "\n",
    "#torch.tensor(X_example.shape).prod(dim=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([8, 45, 256, 256])\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "torch.Size([8, 15])"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "error_map = torch.pow(torch.cat([X_example]*cls_num, axis=1) - y_fcn, 2)\n",
    "print(error_map.shape)\n",
    "y_res = model_res(error_map)\n",
    "y_res.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 损失和优化函数"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "class SquaredErrorLoss(torch.nn.Module):\n",
    "    def __init__(self):\n",
    "        super(SquaredErrorLoss, self).__init__()\n",
    "        \n",
    "    def forward(self, X, Y, Classes):\n",
    "        input_shape = torch.tensor(X.size())\n",
    "        n = X.shape[1]\n",
    "        index = torch.tensor((range(0,n))).to(Classes.device)\n",
    "        Y_class = torch.cat( [Y[i:i+1,index+n*Classes[i],:,:] for i in range(len(Classes))], axis=0)\n",
    "        Error = torch.pow(X - Y_class, 2).sum() / input_shape.prod()\n",
    "        return Error\n",
    "    \n",
    "# FCN 损失和优化函数\n",
    "loss_fcn = SquaredErrorLoss()\n",
    "\n",
    "# ResNet损失和优化函数\n",
    "loss_res = torch.nn.CrossEntropyLoss()\n",
    "\n",
    "import itertools\n",
    "optimizer = torch.optim.Adam(itertools.chain(model_deeplab.parameters(),model_res.parameters()), lr=0.00001)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "Classes = torch.tensor( np.random.randint(low=0, high=cls_num, size=len(X_example)) )\n",
    "print(Classes)\n",
    "\n",
    "loss_fcn(X_example,y_fcn, Classes)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 配置GPU，开始进行训练"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "import collections\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "import torchvision\n",
    "from torchvision import datasets, transforms\n",
    "from torch.autograd import Variable\n",
    "import torch.nn.functional as F\n",
    "import time\n",
    "import pdb"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "try:\n",
    "    del X_example\n",
    "    del y\n",
    "    del y_class\n",
    "    del error_map\n",
    "except:\n",
    "    1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "torch.cuda.empty_cache()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "_CudaDeviceProperties(name='NVIDIA GeForce GTX 1080', major=6, minor=1, total_memory=8192MB, multi_processor_count=20)\n",
      "Let's use 2 GPUs!\n"
     ]
    }
   ],
   "source": [
    "# 配置GPU，开始进行训练\n",
    "Use_gpu = torch.cuda.is_available()\n",
    "\n",
    "if Use_gpu:\n",
    "    print( torch.cuda.get_device_properties(device = 0) )\n",
    "    model_deeplab = model_deeplab.cuda()\n",
    "    model_res = model_res.cuda()\n",
    "    \n",
    "# 配置多GPU\n",
    "if torch.cuda.device_count() > 1:\n",
    "    os.environ['CUDA_VISIBLE_DEVICES'] = '0,1'\n",
    "    print(\"Let's use\", torch.cuda.device_count(), \"GPUs!\")\n",
    "    model_deeplab = torch.nn.DataParallel(model_deeplab)\n",
    "    model_res = torch.nn.DataParallel(model_res)\n",
    "    \n",
    "masked_gaussian = masked_gaussian.to(next(model_deeplab.parameters()).device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 20/99\n",
      "----------\n",
      "Training...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\torch\\cuda\\nccl.py:15: UserWarning: PyTorch is not compiled with NCCL support\n",
      "  warnings.warn('PyTorch is not compiled with NCCL support')\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Batch 500, Train Loss:0.0306, Train ACC:99.7750\n",
      "Current time 8.21 min\n",
      "Batch 1000, Train Loss:0.0297, Train ACC:99.7750\n",
      "Current time 16.32 min\n",
      "Batch 1500, Train Loss:0.0310, Train ACC:99.7500\n",
      "Current time 24.42 min\n",
      "Batch 2000, Train Loss:0.0314, Train ACC:99.7125\n",
      "Current time 32.53 min\n",
      "Batch 2500, Train Loss:0.0307, Train ACC:99.7400\n",
      "Current time 40.65 min\n",
      "Batch 3000, Train Loss:0.0307, Train ACC:99.7292\n",
      "Current time 48.74 min\n",
      "Batch 3500, Train Loss:0.0304, Train ACC:99.7429\n",
      "Current time 56.88 min\n",
      "Batch 4000, Train Loss:0.0311, Train ACC:99.7250\n",
      "Current time 64.98 min\n",
      "Batch 4500, Train Loss:0.0312, Train ACC:99.7361\n",
      "Current time 73.07 min\n",
      "Batch 5000, Train Loss:0.0317, Train ACC:99.7350\n",
      "Current time 81.17 min\n",
      "Batch 5500, Train Loss:0.0318, Train ACC:99.7386\n",
      "Current time 89.27 min\n",
      "Batch 6000, Train Loss:0.0321, Train ACC:99.7313\n",
      "Current time 97.37 min\n",
      "Batch 6500, Train Loss:0.0321, Train ACC:99.7385\n",
      "Current time 105.49 min\n",
      "Batch 7000, Train Loss:0.0319, Train ACC:99.7429\n",
      "Current time 113.58 min\n",
      "Batch 7500, Train Loss:0.0321, Train ACC:99.7383\n",
      "Current time 121.67 min\n",
      "Batch 8000, Train Loss:0.0319, Train ACC:99.7438\n",
      "Current time 129.78 min\n",
      "Batch 8500, Train Loss:0.0319, Train ACC:99.7382\n",
      "Current time 137.88 min\n",
      "Batch 9000, Train Loss:0.0320, Train ACC:99.7361\n",
      "Current time 145.98 min\n",
      "Batch 9500, Train Loss:0.0321, Train ACC:99.7316\n",
      "Current time 154.05 min\n",
      "Batch 10000, Train Loss:0.0320, Train ACC:99.7313\n",
      "Current time 162.14 min\n",
      "Batch 10500, Train Loss:0.0321, Train ACC:99.7286\n",
      "Current time 170.24 min\n",
      "Batch 11000, Train Loss:0.0320, Train ACC:99.7307\n",
      "Current time 178.35 min\n",
      "Batch 11500, Train Loss:0.0320, Train ACC:99.7304\n",
      "Current time 186.45 min\n",
      "train Loss:0.0639  Acc:99.7296%\n",
      "Finish time 193.41 min\n",
      "Epoch 21/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0289, Train ACC:99.8000\n",
      "Current time 201.25 min\n",
      "Batch 1000, Train Loss:0.0283, Train ACC:99.8125\n",
      "Current time 209.08 min\n",
      "Batch 1500, Train Loss:0.0298, Train ACC:99.7500\n",
      "Current time 216.92 min\n",
      "Batch 2000, Train Loss:0.0306, Train ACC:99.7250\n",
      "Current time 224.76 min\n",
      "Batch 2500, Train Loss:0.0309, Train ACC:99.7350\n",
      "Current time 232.64 min\n",
      "Batch 3000, Train Loss:0.0304, Train ACC:99.7583\n",
      "Current time 240.63 min\n",
      "Batch 3500, Train Loss:0.0301, Train ACC:99.7643\n",
      "Current time 248.81 min\n",
      "Batch 4000, Train Loss:0.0318, Train ACC:99.7219\n",
      "Current time 257.08 min\n",
      "Batch 4500, Train Loss:0.0317, Train ACC:99.7250\n",
      "Current time 265.38 min\n",
      "Batch 5000, Train Loss:0.0317, Train ACC:99.7350\n",
      "Current time 273.69 min\n",
      "Batch 5500, Train Loss:0.0320, Train ACC:99.7341\n",
      "Current time 281.99 min\n",
      "Batch 6000, Train Loss:0.0323, Train ACC:99.7208\n",
      "Current time 290.27 min\n",
      "Batch 6500, Train Loss:0.0320, Train ACC:99.7269\n",
      "Current time 298.56 min\n",
      "Batch 7000, Train Loss:0.0317, Train ACC:99.7375\n",
      "Current time 306.85 min\n",
      "Batch 7500, Train Loss:0.0324, Train ACC:99.7167\n",
      "Current time 315.14 min\n",
      "Batch 8000, Train Loss:0.0325, Train ACC:99.7156\n",
      "Current time 323.41 min\n",
      "Batch 8500, Train Loss:0.0323, Train ACC:99.7191\n",
      "Current time 331.68 min\n",
      "Batch 9000, Train Loss:0.0323, Train ACC:99.7181\n",
      "Current time 339.96 min\n",
      "Batch 9500, Train Loss:0.0323, Train ACC:99.7158\n",
      "Current time 348.25 min\n",
      "Batch 10000, Train Loss:0.0324, Train ACC:99.7138\n",
      "Current time 356.56 min\n",
      "Batch 10500, Train Loss:0.0323, Train ACC:99.7143\n",
      "Current time 364.84 min\n",
      "Batch 11000, Train Loss:0.0321, Train ACC:99.7193\n",
      "Current time 373.13 min\n",
      "Batch 11500, Train Loss:0.0320, Train ACC:99.7217\n",
      "Current time 381.45 min\n",
      "train Loss:0.0647  Acc:99.7129%\n",
      "Finish time 388.61 min\n",
      "Epoch 22/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0306, Train ACC:99.8500\n",
      "Current time 396.78 min\n",
      "Batch 1000, Train Loss:0.0310, Train ACC:99.8250\n",
      "Current time 404.98 min\n",
      "Batch 1500, Train Loss:0.0299, Train ACC:99.8333\n",
      "Current time 413.17 min\n",
      "Batch 2000, Train Loss:0.0304, Train ACC:99.7938\n",
      "Current time 421.35 min\n",
      "Batch 2500, Train Loss:0.0301, Train ACC:99.7750\n",
      "Current time 429.53 min\n",
      "Batch 3000, Train Loss:0.0293, Train ACC:99.8000\n",
      "Current time 437.72 min\n",
      "Batch 3500, Train Loss:0.0295, Train ACC:99.7929\n",
      "Current time 445.92 min\n",
      "Batch 4000, Train Loss:0.0295, Train ACC:99.7906\n",
      "Current time 454.11 min\n",
      "Batch 4500, Train Loss:0.0297, Train ACC:99.7944\n",
      "Current time 462.31 min\n",
      "Batch 5000, Train Loss:0.0296, Train ACC:99.7975\n",
      "Current time 470.53 min\n",
      "Batch 5500, Train Loss:0.0301, Train ACC:99.7841\n",
      "Current time 478.74 min\n",
      "Batch 6000, Train Loss:0.0300, Train ACC:99.7771\n",
      "Current time 486.93 min\n",
      "Batch 6500, Train Loss:0.0299, Train ACC:99.7827\n",
      "Current time 495.15 min\n",
      "Batch 7000, Train Loss:0.0303, Train ACC:99.7679\n",
      "Current time 503.34 min\n",
      "Batch 7500, Train Loss:0.0309, Train ACC:99.7533\n",
      "Current time 511.54 min\n",
      "Batch 8000, Train Loss:0.0310, Train ACC:99.7516\n",
      "Current time 519.76 min\n",
      "Batch 8500, Train Loss:0.0308, Train ACC:99.7529\n",
      "Current time 527.96 min\n",
      "Batch 9000, Train Loss:0.0308, Train ACC:99.7486\n",
      "Current time 536.16 min\n",
      "Batch 9500, Train Loss:0.0308, Train ACC:99.7487\n",
      "Current time 544.37 min\n",
      "Batch 10000, Train Loss:0.0307, Train ACC:99.7500\n",
      "Current time 552.59 min\n",
      "Batch 10500, Train Loss:0.0308, Train ACC:99.7464\n",
      "Current time 560.80 min\n",
      "Batch 11000, Train Loss:0.0309, Train ACC:99.7500\n",
      "Current time 569.07 min\n",
      "Batch 11500, Train Loss:0.0314, Train ACC:99.7380\n",
      "Current time 577.29 min\n",
      "train Loss:0.0631  Acc:99.7349%\n",
      "Finish time 584.38 min\n",
      "Epoch 23/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0338, Train ACC:99.7000\n",
      "Current time 592.50 min\n",
      "Batch 1000, Train Loss:0.0324, Train ACC:99.7375\n",
      "Current time 600.66 min\n",
      "Batch 1500, Train Loss:0.0301, Train ACC:99.7667\n",
      "Current time 608.80 min\n",
      "Batch 2000, Train Loss:0.0304, Train ACC:99.7562\n",
      "Current time 617.01 min\n",
      "Batch 2500, Train Loss:0.0299, Train ACC:99.7800\n",
      "Current time 625.14 min\n",
      "Batch 3000, Train Loss:0.0296, Train ACC:99.7958\n",
      "Current time 633.30 min\n",
      "Batch 3500, Train Loss:0.0291, Train ACC:99.7964\n",
      "Current time 641.43 min\n",
      "Batch 4000, Train Loss:0.0291, Train ACC:99.7906\n",
      "Current time 649.61 min\n",
      "Batch 4500, Train Loss:0.0294, Train ACC:99.7861\n",
      "Current time 657.76 min\n",
      "Batch 5000, Train Loss:0.0305, Train ACC:99.7550\n",
      "Current time 665.91 min\n",
      "Batch 5500, Train Loss:0.0305, Train ACC:99.7523\n",
      "Current time 674.08 min\n",
      "Batch 6000, Train Loss:0.0303, Train ACC:99.7625\n",
      "Current time 682.27 min\n",
      "Batch 6500, Train Loss:0.0302, Train ACC:99.7654\n",
      "Current time 690.43 min\n",
      "Batch 7000, Train Loss:0.0305, Train ACC:99.7679\n",
      "Current time 698.61 min\n",
      "Batch 7500, Train Loss:0.0305, Train ACC:99.7683\n",
      "Current time 706.74 min\n",
      "Batch 8000, Train Loss:0.0308, Train ACC:99.7547\n",
      "Current time 714.92 min\n",
      "Batch 8500, Train Loss:0.0305, Train ACC:99.7603\n",
      "Current time 723.11 min\n",
      "Batch 9000, Train Loss:0.0303, Train ACC:99.7653\n",
      "Current time 731.26 min\n",
      "Batch 9500, Train Loss:0.0302, Train ACC:99.7658\n",
      "Current time 739.39 min\n",
      "Batch 10000, Train Loss:0.0301, Train ACC:99.7687\n",
      "Current time 747.49 min\n",
      "Batch 10500, Train Loss:0.0301, Train ACC:99.7619\n",
      "Current time 755.60 min\n",
      "Batch 11000, Train Loss:0.0305, Train ACC:99.7545\n",
      "Current time 763.72 min\n",
      "Batch 11500, Train Loss:0.0304, Train ACC:99.7533\n",
      "Current time 771.83 min\n",
      "train Loss:0.0606  Acc:99.7569%\n",
      "Finish time 778.79 min\n",
      "Epoch 24/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0277, Train ACC:99.8500\n",
      "Current time 786.90 min\n",
      "Batch 1000, Train Loss:0.0297, Train ACC:99.7625\n",
      "Current time 795.00 min\n",
      "Batch 1500, Train Loss:0.0308, Train ACC:99.7417\n",
      "Current time 803.12 min\n",
      "Batch 2000, Train Loss:0.0312, Train ACC:99.7375\n",
      "Current time 811.22 min\n",
      "Batch 2500, Train Loss:0.0306, Train ACC:99.7450\n",
      "Current time 819.35 min\n",
      "Batch 3000, Train Loss:0.0322, Train ACC:99.7125\n",
      "Current time 827.47 min\n",
      "Batch 3500, Train Loss:0.0319, Train ACC:99.7071\n",
      "Current time 835.59 min\n",
      "Batch 4000, Train Loss:0.0316, Train ACC:99.7125\n",
      "Current time 843.70 min\n",
      "Batch 4500, Train Loss:0.0316, Train ACC:99.7250\n",
      "Current time 851.82 min\n",
      "Batch 5000, Train Loss:0.0314, Train ACC:99.7225\n",
      "Current time 859.95 min\n",
      "Batch 5500, Train Loss:0.0317, Train ACC:99.7182\n",
      "Current time 868.04 min\n",
      "Batch 6000, Train Loss:0.0313, Train ACC:99.7333\n",
      "Current time 876.16 min\n",
      "Batch 6500, Train Loss:0.0312, Train ACC:99.7404\n",
      "Current time 884.28 min\n",
      "Batch 7000, Train Loss:0.0312, Train ACC:99.7411\n",
      "Current time 892.36 min\n",
      "Batch 7500, Train Loss:0.0312, Train ACC:99.7400\n",
      "Current time 900.45 min\n",
      "Batch 8000, Train Loss:0.0316, Train ACC:99.7281\n",
      "Current time 908.57 min\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Batch 8500, Train Loss:0.0314, Train ACC:99.7324\n",
      "Current time 916.68 min\n",
      "Batch 9000, Train Loss:0.0314, Train ACC:99.7306\n",
      "Current time 924.79 min\n",
      "Batch 9500, Train Loss:0.0312, Train ACC:99.7329\n",
      "Current time 932.91 min\n",
      "Batch 10000, Train Loss:0.0310, Train ACC:99.7362\n",
      "Current time 941.04 min\n",
      "Batch 10500, Train Loss:0.0310, Train ACC:99.7369\n",
      "Current time 949.16 min\n",
      "Batch 11000, Train Loss:0.0309, Train ACC:99.7375\n",
      "Current time 957.29 min\n",
      "Batch 11500, Train Loss:0.0308, Train ACC:99.7370\n",
      "Current time 965.40 min\n",
      "train Loss:0.0617  Acc:99.7317%\n",
      "Finish time 972.37 min\n",
      "Epoch 25/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0255, Train ACC:99.9000\n",
      "Current time 980.51 min\n",
      "Batch 1000, Train Loss:0.0262, Train ACC:99.8625\n",
      "Current time 988.62 min\n",
      "Batch 1500, Train Loss:0.0276, Train ACC:99.8250\n",
      "Current time 996.74 min\n",
      "Batch 2000, Train Loss:0.0286, Train ACC:99.8063\n",
      "Current time 1004.90 min\n",
      "Batch 2500, Train Loss:0.0286, Train ACC:99.8050\n",
      "Current time 1013.01 min\n",
      "Batch 3000, Train Loss:0.0303, Train ACC:99.7625\n",
      "Current time 1021.17 min\n",
      "Batch 3500, Train Loss:0.0298, Train ACC:99.7750\n",
      "Current time 1029.28 min\n",
      "Batch 4000, Train Loss:0.0294, Train ACC:99.7781\n",
      "Current time 1037.38 min\n",
      "Batch 4500, Train Loss:0.0294, Train ACC:99.7750\n",
      "Current time 1045.49 min\n",
      "Batch 5000, Train Loss:0.0297, Train ACC:99.7725\n",
      "Current time 1053.59 min\n",
      "Batch 5500, Train Loss:0.0298, Train ACC:99.7636\n",
      "Current time 1061.72 min\n",
      "Batch 6000, Train Loss:0.0296, Train ACC:99.7542\n",
      "Current time 1069.85 min\n",
      "Batch 6500, Train Loss:0.0296, Train ACC:99.7519\n",
      "Current time 1077.99 min\n",
      "Batch 7000, Train Loss:0.0296, Train ACC:99.7571\n",
      "Current time 1086.11 min\n",
      "Batch 7500, Train Loss:0.0294, Train ACC:99.7650\n",
      "Current time 1094.28 min\n",
      "Batch 8000, Train Loss:0.0295, Train ACC:99.7641\n",
      "Current time 1102.42 min\n",
      "Batch 8500, Train Loss:0.0291, Train ACC:99.7765\n",
      "Current time 1110.57 min\n",
      "Batch 9000, Train Loss:0.0289, Train ACC:99.7861\n",
      "Current time 1118.67 min\n",
      "Batch 9500, Train Loss:0.0287, Train ACC:99.7908\n",
      "Current time 1126.79 min\n",
      "Batch 10000, Train Loss:0.0290, Train ACC:99.7887\n",
      "Current time 1134.91 min\n",
      "Batch 10500, Train Loss:0.0292, Train ACC:99.7845\n",
      "Current time 1143.06 min\n",
      "Batch 11000, Train Loss:0.0295, Train ACC:99.7727\n",
      "Current time 1151.20 min\n",
      "Batch 11500, Train Loss:0.0295, Train ACC:99.7739\n",
      "Current time 1159.33 min\n",
      "train Loss:0.0590  Acc:99.7737%\n",
      "Finish time 1166.31 min\n",
      "Epoch 26/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0291, Train ACC:99.8000\n",
      "Current time 1174.45 min\n",
      "Batch 1000, Train Loss:0.0281, Train ACC:99.7875\n",
      "Current time 1182.59 min\n",
      "Batch 1500, Train Loss:0.0299, Train ACC:99.7667\n",
      "Current time 1190.73 min\n",
      "Batch 2000, Train Loss:0.0302, Train ACC:99.7438\n",
      "Current time 1198.88 min\n",
      "Batch 2500, Train Loss:0.0296, Train ACC:99.7600\n",
      "Current time 1207.03 min\n",
      "Batch 3000, Train Loss:0.0286, Train ACC:99.7875\n",
      "Current time 1215.16 min\n",
      "Batch 3500, Train Loss:0.0287, Train ACC:99.7893\n",
      "Current time 1223.30 min\n",
      "Batch 4000, Train Loss:0.0287, Train ACC:99.7938\n",
      "Current time 1231.51 min\n",
      "Batch 4500, Train Loss:0.0285, Train ACC:99.7972\n",
      "Current time 1239.63 min\n",
      "Batch 5000, Train Loss:0.0282, Train ACC:99.8100\n",
      "Current time 1247.77 min\n",
      "Batch 5500, Train Loss:0.0283, Train ACC:99.8023\n",
      "Current time 1255.88 min\n",
      "Batch 6000, Train Loss:0.0286, Train ACC:99.7896\n",
      "Current time 1264.02 min\n",
      "Batch 6500, Train Loss:0.0291, Train ACC:99.7788\n",
      "Current time 1272.12 min\n",
      "Batch 7000, Train Loss:0.0292, Train ACC:99.7786\n",
      "Current time 1280.24 min\n",
      "Batch 7500, Train Loss:0.0289, Train ACC:99.7883\n",
      "Current time 1288.34 min\n",
      "Batch 8000, Train Loss:0.0290, Train ACC:99.7875\n",
      "Current time 1296.52 min\n",
      "Batch 8500, Train Loss:0.0290, Train ACC:99.7838\n",
      "Current time 1304.68 min\n",
      "Batch 9000, Train Loss:0.0289, Train ACC:99.7889\n",
      "Current time 1312.83 min\n",
      "Batch 9500, Train Loss:0.0288, Train ACC:99.7961\n",
      "Current time 1320.95 min\n",
      "Batch 10000, Train Loss:0.0287, Train ACC:99.7963\n",
      "Current time 1329.09 min\n",
      "Batch 10500, Train Loss:0.0286, Train ACC:99.8024\n",
      "Current time 1337.24 min\n",
      "Batch 11000, Train Loss:0.0286, Train ACC:99.8011\n",
      "Current time 1345.37 min\n",
      "Batch 11500, Train Loss:0.0286, Train ACC:99.7989\n",
      "Current time 1353.52 min\n",
      "train Loss:0.0572  Acc:99.7999%\n",
      "Finish time 1360.48 min\n",
      "Epoch 27/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0260, Train ACC:99.8250\n",
      "Current time 1368.62 min\n",
      "Batch 1000, Train Loss:0.0259, Train ACC:99.8625\n",
      "Current time 1376.75 min\n",
      "Batch 1500, Train Loss:0.0262, Train ACC:99.8250\n",
      "Current time 1384.90 min\n",
      "Batch 2000, Train Loss:0.0281, Train ACC:99.7313\n",
      "Current time 1393.01 min\n",
      "Batch 2500, Train Loss:0.0286, Train ACC:99.7500\n",
      "Current time 1401.16 min\n",
      "Batch 3000, Train Loss:0.0294, Train ACC:99.7250\n",
      "Current time 1409.32 min\n",
      "Batch 3500, Train Loss:0.0290, Train ACC:99.7393\n",
      "Current time 1417.50 min\n",
      "Batch 4000, Train Loss:0.0284, Train ACC:99.7594\n",
      "Current time 1425.66 min\n",
      "Batch 4500, Train Loss:0.0287, Train ACC:99.7500\n",
      "Current time 1433.79 min\n",
      "Batch 5000, Train Loss:0.0286, Train ACC:99.7525\n",
      "Current time 1441.93 min\n",
      "Batch 5500, Train Loss:0.0285, Train ACC:99.7523\n",
      "Current time 1450.07 min\n",
      "Batch 6000, Train Loss:0.0283, Train ACC:99.7604\n",
      "Current time 1458.22 min\n",
      "Batch 6500, Train Loss:0.0283, Train ACC:99.7615\n",
      "Current time 1466.32 min\n",
      "Batch 7000, Train Loss:0.0284, Train ACC:99.7554\n",
      "Current time 1474.40 min\n",
      "Batch 7500, Train Loss:0.0286, Train ACC:99.7533\n",
      "Current time 1482.45 min\n",
      "Batch 8000, Train Loss:0.0286, Train ACC:99.7547\n",
      "Current time 1490.54 min\n",
      "Batch 8500, Train Loss:0.0288, Train ACC:99.7500\n",
      "Current time 1498.63 min\n",
      "Batch 9000, Train Loss:0.0290, Train ACC:99.7486\n",
      "Current time 1506.68 min\n",
      "Batch 9500, Train Loss:0.0290, Train ACC:99.7500\n",
      "Current time 1514.72 min\n",
      "Batch 10000, Train Loss:0.0290, Train ACC:99.7463\n",
      "Current time 1522.75 min\n",
      "Batch 10500, Train Loss:0.0291, Train ACC:99.7440\n",
      "Current time 1530.81 min\n",
      "Batch 11000, Train Loss:0.0290, Train ACC:99.7466\n",
      "Current time 1538.86 min\n",
      "Batch 11500, Train Loss:0.0289, Train ACC:99.7500\n",
      "Current time 1546.90 min\n",
      "train Loss:0.0578  Acc:99.7496%\n",
      "Finish time 1553.81 min\n",
      "Epoch 28/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0256, Train ACC:99.8250\n",
      "Current time 1561.86 min\n",
      "Batch 1000, Train Loss:0.0266, Train ACC:99.7625\n",
      "Current time 1569.91 min\n",
      "Batch 1500, Train Loss:0.0263, Train ACC:99.7750\n",
      "Current time 1578.00 min\n",
      "Batch 2000, Train Loss:0.0261, Train ACC:99.7938\n",
      "Current time 1586.11 min\n",
      "Batch 2500, Train Loss:0.0266, Train ACC:99.8000\n",
      "Current time 1594.18 min\n",
      "Batch 3000, Train Loss:0.0273, Train ACC:99.7875\n",
      "Current time 1602.27 min\n",
      "Batch 3500, Train Loss:0.0290, Train ACC:99.7571\n",
      "Current time 1610.39 min\n",
      "Batch 4000, Train Loss:0.0287, Train ACC:99.7719\n",
      "Current time 1618.52 min\n",
      "Batch 4500, Train Loss:0.0289, Train ACC:99.7556\n",
      "Current time 1626.62 min\n",
      "Batch 5000, Train Loss:0.0287, Train ACC:99.7750\n",
      "Current time 1634.75 min\n",
      "Batch 5500, Train Loss:0.0284, Train ACC:99.7750\n",
      "Current time 1642.88 min\n",
      "Batch 6000, Train Loss:0.0282, Train ACC:99.7833\n",
      "Current time 1651.01 min\n",
      "Batch 6500, Train Loss:0.0278, Train ACC:99.7942\n",
      "Current time 1659.14 min\n",
      "Batch 7000, Train Loss:0.0278, Train ACC:99.7946\n",
      "Current time 1667.28 min\n",
      "Batch 7500, Train Loss:0.0279, Train ACC:99.7883\n",
      "Current time 1675.42 min\n",
      "Batch 8000, Train Loss:0.0280, Train ACC:99.7906\n",
      "Current time 1683.56 min\n",
      "Batch 8500, Train Loss:0.0282, Train ACC:99.7882\n",
      "Current time 1691.69 min\n",
      "Batch 9000, Train Loss:0.0282, Train ACC:99.7889\n",
      "Current time 1700.06 min\n",
      "Batch 9500, Train Loss:0.0281, Train ACC:99.7934\n",
      "Current time 1708.46 min\n",
      "Batch 10000, Train Loss:0.0279, Train ACC:99.7963\n",
      "Current time 1716.85 min\n",
      "Batch 10500, Train Loss:0.0276, Train ACC:99.8012\n",
      "Current time 1725.20 min\n",
      "Batch 11000, Train Loss:0.0277, Train ACC:99.7989\n",
      "Current time 1733.59 min\n",
      "Batch 11500, Train Loss:0.0279, Train ACC:99.7935\n",
      "Current time 1741.97 min\n",
      "train Loss:0.0558  Acc:99.7936%\n",
      "Finish time 1749.20 min\n",
      "Epoch 29/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0262, Train ACC:99.8500\n",
      "Current time 1757.52 min\n",
      "Batch 1000, Train Loss:0.0284, Train ACC:99.8000\n",
      "Current time 1765.83 min\n",
      "Batch 1500, Train Loss:0.0290, Train ACC:99.7917\n",
      "Current time 1774.14 min\n",
      "Batch 2000, Train Loss:0.0271, Train ACC:99.8375\n",
      "Current time 1782.46 min\n",
      "Batch 2500, Train Loss:0.0268, Train ACC:99.8300\n",
      "Current time 1790.77 min\n",
      "Batch 3000, Train Loss:0.0271, Train ACC:99.8292\n",
      "Current time 1799.08 min\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Batch 3500, Train Loss:0.0273, Train ACC:99.8036\n",
      "Current time 1807.38 min\n",
      "Batch 4000, Train Loss:0.0276, Train ACC:99.7906\n",
      "Current time 1815.69 min\n",
      "Batch 4500, Train Loss:0.0281, Train ACC:99.7639\n",
      "Current time 1824.02 min\n",
      "Batch 5000, Train Loss:0.0277, Train ACC:99.7825\n",
      "Current time 1832.36 min\n",
      "Batch 5500, Train Loss:0.0276, Train ACC:99.7841\n",
      "Current time 1840.64 min\n",
      "Batch 6000, Train Loss:0.0276, Train ACC:99.7812\n",
      "Current time 1848.96 min\n",
      "Batch 6500, Train Loss:0.0275, Train ACC:99.7885\n",
      "Current time 1857.26 min\n",
      "Batch 7000, Train Loss:0.0274, Train ACC:99.7875\n",
      "Current time 1865.60 min\n",
      "Batch 7500, Train Loss:0.0273, Train ACC:99.7900\n",
      "Current time 1873.92 min\n",
      "Batch 8000, Train Loss:0.0278, Train ACC:99.7781\n",
      "Current time 1882.21 min\n",
      "Batch 8500, Train Loss:0.0280, Train ACC:99.7765\n",
      "Current time 1890.54 min\n",
      "Batch 9000, Train Loss:0.0278, Train ACC:99.7792\n",
      "Current time 1898.84 min\n",
      "Batch 9500, Train Loss:0.0276, Train ACC:99.7868\n",
      "Current time 1907.14 min\n",
      "Batch 10000, Train Loss:0.0275, Train ACC:99.7862\n",
      "Current time 1915.44 min\n",
      "Batch 10500, Train Loss:0.0275, Train ACC:99.7893\n",
      "Current time 1923.74 min\n",
      "Batch 11000, Train Loss:0.0277, Train ACC:99.7852\n",
      "Current time 1932.04 min\n",
      "Batch 11500, Train Loss:0.0281, Train ACC:99.7728\n",
      "Current time 1940.37 min\n",
      "train Loss:0.0567  Acc:99.7674%\n",
      "Finish time 1947.52 min\n",
      "Epoch 30/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0246, Train ACC:99.8500\n",
      "Current time 1955.66 min\n",
      "Batch 1000, Train Loss:0.0249, Train ACC:99.8375\n",
      "Current time 1963.78 min\n",
      "Batch 1500, Train Loss:0.0255, Train ACC:99.8333\n",
      "Current time 1971.91 min\n",
      "Batch 2000, Train Loss:0.0269, Train ACC:99.8125\n",
      "Current time 1980.04 min\n",
      "Batch 2500, Train Loss:0.0269, Train ACC:99.8150\n",
      "Current time 1988.16 min\n",
      "Batch 3000, Train Loss:0.0268, Train ACC:99.8167\n",
      "Current time 1996.32 min\n",
      "Batch 3500, Train Loss:0.0275, Train ACC:99.7964\n",
      "Current time 2004.45 min\n",
      "Batch 4000, Train Loss:0.0278, Train ACC:99.7938\n",
      "Current time 2012.58 min\n",
      "Batch 4500, Train Loss:0.0283, Train ACC:99.7778\n",
      "Current time 2020.73 min\n",
      "Batch 5000, Train Loss:0.0283, Train ACC:99.7775\n",
      "Current time 2028.85 min\n",
      "Batch 5500, Train Loss:0.0280, Train ACC:99.7818\n",
      "Current time 2036.97 min\n",
      "Batch 6000, Train Loss:0.0279, Train ACC:99.7854\n",
      "Current time 2045.11 min\n",
      "Batch 6500, Train Loss:0.0281, Train ACC:99.7827\n",
      "Current time 2053.23 min\n",
      "Batch 7000, Train Loss:0.0279, Train ACC:99.7875\n",
      "Current time 2061.35 min\n",
      "Batch 7500, Train Loss:0.0278, Train ACC:99.7900\n",
      "Current time 2069.51 min\n",
      "Batch 8000, Train Loss:0.0276, Train ACC:99.7953\n",
      "Current time 2077.64 min\n",
      "Batch 8500, Train Loss:0.0276, Train ACC:99.7941\n",
      "Current time 2085.79 min\n",
      "Batch 9000, Train Loss:0.0276, Train ACC:99.7903\n",
      "Current time 2093.92 min\n",
      "Batch 9500, Train Loss:0.0278, Train ACC:99.7829\n",
      "Current time 2102.07 min\n",
      "Batch 10000, Train Loss:0.0278, Train ACC:99.7837\n",
      "Current time 2110.22 min\n",
      "Batch 10500, Train Loss:0.0275, Train ACC:99.7905\n",
      "Current time 2118.34 min\n",
      "Batch 11000, Train Loss:0.0275, Train ACC:99.7943\n",
      "Current time 2126.46 min\n",
      "Batch 11500, Train Loss:0.0277, Train ACC:99.7902\n",
      "Current time 2134.60 min\n",
      "train Loss:0.0552  Acc:99.7915%\n",
      "Finish time 2141.60 min\n",
      "Epoch 31/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0250, Train ACC:99.8500\n",
      "Current time 2149.72 min\n",
      "Batch 1000, Train Loss:0.0247, Train ACC:99.8375\n",
      "Current time 2157.85 min\n",
      "Batch 1500, Train Loss:0.0241, Train ACC:99.8500\n",
      "Current time 2166.02 min\n",
      "Batch 2000, Train Loss:0.0240, Train ACC:99.8812\n",
      "Current time 2174.20 min\n",
      "Batch 2500, Train Loss:0.0235, Train ACC:99.9000\n",
      "Current time 2182.39 min\n",
      "Batch 3000, Train Loss:0.0240, Train ACC:99.8750\n",
      "Current time 2190.51 min\n",
      "Batch 3500, Train Loss:0.0239, Train ACC:99.8857\n",
      "Current time 2198.67 min\n",
      "Batch 4000, Train Loss:0.0252, Train ACC:99.8469\n",
      "Current time 2206.82 min\n",
      "Batch 4500, Train Loss:0.0250, Train ACC:99.8528\n",
      "Current time 2214.90 min\n",
      "Batch 5000, Train Loss:0.0248, Train ACC:99.8625\n",
      "Current time 2223.05 min\n",
      "Batch 5500, Train Loss:0.0252, Train ACC:99.8545\n",
      "Current time 2231.16 min\n",
      "Batch 6000, Train Loss:0.0251, Train ACC:99.8500\n",
      "Current time 2239.29 min\n",
      "Batch 6500, Train Loss:0.0252, Train ACC:99.8500\n",
      "Current time 2247.41 min\n",
      "Batch 7000, Train Loss:0.0254, Train ACC:99.8464\n",
      "Current time 2255.56 min\n",
      "Batch 7500, Train Loss:0.0252, Train ACC:99.8517\n",
      "Current time 2263.73 min\n",
      "Batch 8000, Train Loss:0.0257, Train ACC:99.8422\n",
      "Current time 2271.88 min\n",
      "Batch 8500, Train Loss:0.0258, Train ACC:99.8397\n",
      "Current time 2280.01 min\n",
      "Batch 9000, Train Loss:0.0261, Train ACC:99.8292\n",
      "Current time 2288.11 min\n",
      "Batch 9500, Train Loss:0.0265, Train ACC:99.8197\n",
      "Current time 2296.26 min\n",
      "Batch 10000, Train Loss:0.0266, Train ACC:99.8200\n",
      "Current time 2304.43 min\n",
      "Batch 10500, Train Loss:0.0264, Train ACC:99.8274\n",
      "Current time 2312.56 min\n",
      "Batch 11000, Train Loss:0.0266, Train ACC:99.8239\n",
      "Current time 2320.69 min\n",
      "Batch 11500, Train Loss:0.0266, Train ACC:99.8228\n",
      "Current time 2328.84 min\n",
      "train Loss:0.0536  Acc:99.8219%\n",
      "Finish time 2335.81 min\n",
      "Epoch 32/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0296, Train ACC:99.6750\n",
      "Current time 2343.94 min\n",
      "Batch 1000, Train Loss:0.0283, Train ACC:99.7625\n",
      "Current time 2352.31 min\n",
      "Batch 1500, Train Loss:0.0278, Train ACC:99.7750\n",
      "Current time 2360.68 min\n",
      "Batch 2000, Train Loss:0.0278, Train ACC:99.7750\n",
      "Current time 2369.06 min\n",
      "Batch 2500, Train Loss:0.0284, Train ACC:99.7650\n",
      "Current time 2377.44 min\n",
      "Batch 3000, Train Loss:0.0282, Train ACC:99.7750\n",
      "Current time 2385.82 min\n",
      "Batch 3500, Train Loss:0.0275, Train ACC:99.7893\n",
      "Current time 2394.18 min\n",
      "Batch 4000, Train Loss:0.0273, Train ACC:99.7906\n",
      "Current time 2402.54 min\n",
      "Batch 4500, Train Loss:0.0269, Train ACC:99.8028\n",
      "Current time 2410.96 min\n",
      "Batch 5000, Train Loss:0.0270, Train ACC:99.8050\n",
      "Current time 2419.35 min\n",
      "Batch 5500, Train Loss:0.0267, Train ACC:99.8114\n",
      "Current time 2427.73 min\n",
      "Batch 6000, Train Loss:0.0274, Train ACC:99.7958\n",
      "Current time 2436.11 min\n",
      "Batch 6500, Train Loss:0.0273, Train ACC:99.7962\n",
      "Current time 2444.47 min\n",
      "Batch 7000, Train Loss:0.0271, Train ACC:99.8054\n",
      "Current time 2452.85 min\n",
      "Batch 7500, Train Loss:0.0270, Train ACC:99.8017\n",
      "Current time 2461.24 min\n",
      "Batch 8000, Train Loss:0.0268, Train ACC:99.8094\n",
      "Current time 2469.65 min\n",
      "Batch 8500, Train Loss:0.0266, Train ACC:99.8176\n",
      "Current time 2478.04 min\n",
      "Batch 9000, Train Loss:0.0267, Train ACC:99.8125\n",
      "Current time 2486.42 min\n",
      "Batch 9500, Train Loss:0.0267, Train ACC:99.8132\n",
      "Current time 2494.80 min\n",
      "Batch 10000, Train Loss:0.0270, Train ACC:99.8063\n",
      "Current time 2503.18 min\n",
      "Batch 10500, Train Loss:0.0268, Train ACC:99.8095\n",
      "Current time 2511.58 min\n",
      "Batch 11000, Train Loss:0.0269, Train ACC:99.8000\n",
      "Current time 2519.95 min\n",
      "Batch 11500, Train Loss:0.0270, Train ACC:99.7946\n",
      "Current time 2528.32 min\n",
      "train Loss:0.0539  Acc:99.7978%\n",
      "Finish time 2535.50 min\n",
      "Epoch 33/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0261, Train ACC:99.8250\n",
      "Current time 2543.65 min\n",
      "Batch 1000, Train Loss:0.0250, Train ACC:99.8625\n",
      "Current time 2551.79 min\n",
      "Batch 1500, Train Loss:0.0243, Train ACC:99.8833\n",
      "Current time 2559.92 min\n",
      "Batch 2000, Train Loss:0.0235, Train ACC:99.9125\n",
      "Current time 2568.09 min\n",
      "Batch 2500, Train Loss:0.0250, Train ACC:99.8750\n",
      "Current time 2576.25 min\n",
      "Batch 3000, Train Loss:0.0260, Train ACC:99.8333\n",
      "Current time 2584.41 min\n",
      "Batch 3500, Train Loss:0.0271, Train ACC:99.8214\n",
      "Current time 2592.59 min\n",
      "Batch 4000, Train Loss:0.0265, Train ACC:99.8375\n",
      "Current time 2600.74 min\n",
      "Batch 4500, Train Loss:0.0265, Train ACC:99.8389\n",
      "Current time 2608.94 min\n",
      "Batch 5000, Train Loss:0.0266, Train ACC:99.8425\n",
      "Current time 2617.08 min\n",
      "Batch 5500, Train Loss:0.0264, Train ACC:99.8477\n",
      "Current time 2625.27 min\n",
      "Batch 6000, Train Loss:0.0269, Train ACC:99.8354\n",
      "Current time 2633.41 min\n",
      "Batch 6500, Train Loss:0.0271, Train ACC:99.8250\n",
      "Current time 2641.58 min\n",
      "Batch 7000, Train Loss:0.0273, Train ACC:99.8179\n",
      "Current time 2649.72 min\n",
      "Batch 7500, Train Loss:0.0272, Train ACC:99.8250\n",
      "Current time 2657.85 min\n",
      "Batch 8000, Train Loss:0.0270, Train ACC:99.8281\n",
      "Current time 2666.02 min\n",
      "Batch 8500, Train Loss:0.0267, Train ACC:99.8353\n",
      "Current time 2674.18 min\n",
      "Batch 9000, Train Loss:0.0265, Train ACC:99.8347\n",
      "Current time 2682.32 min\n",
      "Batch 9500, Train Loss:0.0267, Train ACC:99.8303\n",
      "Current time 2690.48 min\n",
      "Batch 10000, Train Loss:0.0265, Train ACC:99.8325\n",
      "Current time 2698.62 min\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Batch 10500, Train Loss:0.0264, Train ACC:99.8298\n",
      "Current time 2706.77 min\n",
      "Batch 11000, Train Loss:0.0268, Train ACC:99.8239\n",
      "Current time 2714.92 min\n",
      "Batch 11500, Train Loss:0.0272, Train ACC:99.8054\n",
      "Current time 2723.03 min\n",
      "train Loss:0.0543  Acc:99.8051%\n",
      "Finish time 2730.07 min\n",
      "Epoch 34/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0264, Train ACC:99.8500\n",
      "Current time 2738.23 min\n",
      "Batch 1000, Train Loss:0.0267, Train ACC:99.8250\n",
      "Current time 2746.33 min\n",
      "Batch 1500, Train Loss:0.0260, Train ACC:99.8333\n",
      "Current time 2754.44 min\n",
      "Batch 2000, Train Loss:0.0248, Train ACC:99.8625\n",
      "Current time 2762.54 min\n",
      "Batch 2500, Train Loss:0.0249, Train ACC:99.8550\n",
      "Current time 2770.64 min\n",
      "Batch 3000, Train Loss:0.0255, Train ACC:99.8375\n",
      "Current time 2778.74 min\n",
      "Batch 3500, Train Loss:0.0260, Train ACC:99.8286\n",
      "Current time 2786.94 min\n",
      "Batch 4000, Train Loss:0.0257, Train ACC:99.8312\n",
      "Current time 2795.32 min\n",
      "Batch 4500, Train Loss:0.0252, Train ACC:99.8417\n",
      "Current time 2803.69 min\n",
      "Batch 5000, Train Loss:0.0252, Train ACC:99.8450\n",
      "Current time 2812.06 min\n",
      "Batch 5500, Train Loss:0.0252, Train ACC:99.8477\n",
      "Current time 2820.40 min\n",
      "Batch 6000, Train Loss:0.0254, Train ACC:99.8396\n",
      "Current time 2828.80 min\n",
      "Batch 6500, Train Loss:0.0255, Train ACC:99.8327\n",
      "Current time 2837.18 min\n",
      "Batch 7000, Train Loss:0.0259, Train ACC:99.8214\n",
      "Current time 2845.55 min\n",
      "Batch 7500, Train Loss:0.0260, Train ACC:99.8217\n",
      "Current time 2853.99 min\n",
      "Batch 8000, Train Loss:0.0260, Train ACC:99.8250\n",
      "Current time 2862.68 min\n",
      "Batch 8500, Train Loss:0.0260, Train ACC:99.8235\n",
      "Current time 2871.38 min\n",
      "Batch 9000, Train Loss:0.0259, Train ACC:99.8236\n",
      "Current time 2880.09 min\n",
      "Batch 9500, Train Loss:0.0261, Train ACC:99.8211\n",
      "Current time 2888.73 min\n",
      "Batch 10000, Train Loss:0.0260, Train ACC:99.8225\n",
      "Current time 2897.37 min\n",
      "Batch 10500, Train Loss:0.0261, Train ACC:99.8214\n",
      "Current time 2905.74 min\n",
      "Batch 11000, Train Loss:0.0261, Train ACC:99.8193\n",
      "Current time 2914.05 min\n",
      "Batch 11500, Train Loss:0.0260, Train ACC:99.8217\n",
      "Current time 2922.35 min\n",
      "train Loss:0.0520  Acc:99.8187%\n",
      "Finish time 2929.37 min\n",
      "Epoch 35/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0326, Train ACC:99.6250\n",
      "Current time 2937.35 min\n",
      "Batch 1000, Train Loss:0.0282, Train ACC:99.7375\n",
      "Current time 2945.41 min\n",
      "Batch 1500, Train Loss:0.0268, Train ACC:99.8000\n",
      "Current time 2953.43 min\n",
      "Batch 2000, Train Loss:0.0263, Train ACC:99.8187\n",
      "Current time 2961.43 min\n",
      "Batch 2500, Train Loss:0.0252, Train ACC:99.8500\n",
      "Current time 2969.48 min\n",
      "Batch 3000, Train Loss:0.0264, Train ACC:99.8208\n",
      "Current time 2977.59 min\n",
      "Batch 3500, Train Loss:0.0259, Train ACC:99.8286\n",
      "Current time 2985.76 min\n",
      "Batch 4000, Train Loss:0.0256, Train ACC:99.8312\n",
      "Current time 2994.00 min\n",
      "Batch 4500, Train Loss:0.0258, Train ACC:99.8167\n",
      "Current time 3002.10 min\n",
      "Batch 5000, Train Loss:0.0263, Train ACC:99.8025\n",
      "Current time 3010.13 min\n",
      "Batch 5500, Train Loss:0.0259, Train ACC:99.8182\n",
      "Current time 3018.14 min\n",
      "Batch 6000, Train Loss:0.0260, Train ACC:99.8167\n",
      "Current time 3026.11 min\n",
      "Batch 6500, Train Loss:0.0261, Train ACC:99.8096\n",
      "Current time 3034.08 min\n",
      "Batch 7000, Train Loss:0.0260, Train ACC:99.8125\n",
      "Current time 3042.04 min\n",
      "Batch 7500, Train Loss:0.0259, Train ACC:99.8117\n",
      "Current time 3050.04 min\n",
      "Batch 8000, Train Loss:0.0261, Train ACC:99.8078\n",
      "Current time 3058.01 min\n",
      "Batch 8500, Train Loss:0.0260, Train ACC:99.8088\n",
      "Current time 3066.00 min\n",
      "Batch 9000, Train Loss:0.0263, Train ACC:99.8014\n",
      "Current time 3073.97 min\n",
      "Batch 9500, Train Loss:0.0263, Train ACC:99.8000\n",
      "Current time 3081.92 min\n",
      "Batch 10000, Train Loss:0.0263, Train ACC:99.8000\n",
      "Current time 3089.88 min\n",
      "Batch 10500, Train Loss:0.0262, Train ACC:99.8036\n",
      "Current time 3097.84 min\n",
      "Batch 11000, Train Loss:0.0261, Train ACC:99.8080\n",
      "Current time 3105.79 min\n",
      "Batch 11500, Train Loss:0.0260, Train ACC:99.8098\n",
      "Current time 3113.75 min\n",
      "train Loss:0.0523  Acc:99.8093%\n",
      "Finish time 3120.58 min\n",
      "Epoch 36/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0248, Train ACC:99.8000\n",
      "Current time 3128.48 min\n",
      "Batch 1000, Train Loss:0.0231, Train ACC:99.8875\n",
      "Current time 3136.35 min\n",
      "Batch 1500, Train Loss:0.0228, Train ACC:99.8833\n",
      "Current time 3144.26 min\n",
      "Batch 2000, Train Loss:0.0254, Train ACC:99.8312\n",
      "Current time 3152.14 min\n",
      "Batch 2500, Train Loss:0.0257, Train ACC:99.8250\n",
      "Current time 3160.03 min\n",
      "Batch 3000, Train Loss:0.0252, Train ACC:99.8333\n",
      "Current time 3167.92 min\n",
      "Batch 3500, Train Loss:0.0251, Train ACC:99.8321\n",
      "Current time 3175.83 min\n",
      "Batch 4000, Train Loss:0.0256, Train ACC:99.8156\n",
      "Current time 3183.71 min\n",
      "Batch 4500, Train Loss:0.0254, Train ACC:99.8194\n",
      "Current time 3191.59 min\n",
      "Batch 5000, Train Loss:0.0260, Train ACC:99.8125\n",
      "Current time 3199.63 min\n",
      "Batch 5500, Train Loss:0.0259, Train ACC:99.8114\n",
      "Current time 3207.84 min\n",
      "Batch 6000, Train Loss:0.0258, Train ACC:99.8167\n",
      "Current time 3215.96 min\n",
      "Batch 6500, Train Loss:0.0256, Train ACC:99.8231\n",
      "Current time 3224.12 min\n",
      "Batch 7000, Train Loss:0.0254, Train ACC:99.8286\n",
      "Current time 3232.27 min\n",
      "Batch 7500, Train Loss:0.0255, Train ACC:99.8250\n",
      "Current time 3240.47 min\n",
      "Batch 8000, Train Loss:0.0258, Train ACC:99.8125\n",
      "Current time 3248.63 min\n",
      "Batch 8500, Train Loss:0.0256, Train ACC:99.8191\n",
      "Current time 3256.77 min\n",
      "Batch 9000, Train Loss:0.0257, Train ACC:99.8167\n",
      "Current time 3264.97 min\n",
      "Batch 9500, Train Loss:0.0258, Train ACC:99.8171\n",
      "Current time 3273.11 min\n",
      "Batch 10000, Train Loss:0.0257, Train ACC:99.8225\n",
      "Current time 3281.30 min\n",
      "Batch 10500, Train Loss:0.0257, Train ACC:99.8238\n",
      "Current time 3289.48 min\n",
      "Batch 11000, Train Loss:0.0259, Train ACC:99.8216\n",
      "Current time 3297.64 min\n",
      "Batch 11500, Train Loss:0.0259, Train ACC:99.8174\n",
      "Current time 3305.78 min\n",
      "train Loss:0.0516  Acc:99.8198%\n",
      "Finish time 3312.78 min\n",
      "Epoch 37/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0271, Train ACC:99.8000\n",
      "Current time 3320.79 min\n",
      "Batch 1000, Train Loss:0.0261, Train ACC:99.8125\n",
      "Current time 3328.77 min\n",
      "Batch 1500, Train Loss:0.0280, Train ACC:99.7750\n",
      "Current time 3336.77 min\n",
      "Batch 2000, Train Loss:0.0268, Train ACC:99.8125\n",
      "Current time 3344.75 min\n",
      "Batch 2500, Train Loss:0.0263, Train ACC:99.8250\n",
      "Current time 3352.74 min\n",
      "Batch 3000, Train Loss:0.0259, Train ACC:99.8375\n",
      "Current time 3360.72 min\n",
      "Batch 3500, Train Loss:0.0261, Train ACC:99.8393\n",
      "Current time 3368.70 min\n",
      "Batch 4000, Train Loss:0.0263, Train ACC:99.8281\n",
      "Current time 3376.71 min\n",
      "Batch 4500, Train Loss:0.0261, Train ACC:99.8306\n",
      "Current time 3384.70 min\n",
      "Batch 5000, Train Loss:0.0259, Train ACC:99.8300\n",
      "Current time 3392.69 min\n",
      "Batch 5500, Train Loss:0.0261, Train ACC:99.8341\n",
      "Current time 3400.70 min\n",
      "Batch 6000, Train Loss:0.0264, Train ACC:99.8312\n",
      "Current time 3408.71 min\n",
      "Batch 6500, Train Loss:0.0265, Train ACC:99.8346\n",
      "Current time 3416.69 min\n",
      "Batch 7000, Train Loss:0.0265, Train ACC:99.8321\n",
      "Current time 3424.68 min\n",
      "Batch 7500, Train Loss:0.0263, Train ACC:99.8383\n",
      "Current time 3432.69 min\n",
      "Batch 8000, Train Loss:0.0262, Train ACC:99.8406\n",
      "Current time 3440.66 min\n",
      "Batch 8500, Train Loss:0.0261, Train ACC:99.8441\n",
      "Current time 3448.66 min\n",
      "Batch 9000, Train Loss:0.0259, Train ACC:99.8500\n",
      "Current time 3456.64 min\n",
      "Batch 9500, Train Loss:0.0259, Train ACC:99.8513\n",
      "Current time 3464.63 min\n",
      "Batch 10000, Train Loss:0.0258, Train ACC:99.8525\n",
      "Current time 3472.60 min\n",
      "Batch 10500, Train Loss:0.0257, Train ACC:99.8524\n",
      "Current time 3480.59 min\n",
      "Batch 11000, Train Loss:0.0261, Train ACC:99.8443\n",
      "Current time 3488.61 min\n",
      "Batch 11500, Train Loss:0.0260, Train ACC:99.8446\n",
      "Current time 3496.60 min\n",
      "train Loss:0.0525  Acc:99.8397%\n",
      "Finish time 3503.46 min\n",
      "Epoch 38/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0218, Train ACC:99.9750\n",
      "Current time 3511.36 min\n",
      "Batch 1000, Train Loss:0.0232, Train ACC:99.9250\n",
      "Current time 3519.24 min\n",
      "Batch 1500, Train Loss:0.0242, Train ACC:99.9000\n",
      "Current time 3527.14 min\n",
      "Batch 2000, Train Loss:0.0244, Train ACC:99.8750\n",
      "Current time 3535.05 min\n",
      "Batch 2500, Train Loss:0.0240, Train ACC:99.8800\n",
      "Current time 3542.94 min\n",
      "Batch 3000, Train Loss:0.0242, Train ACC:99.8750\n",
      "Current time 3550.82 min\n",
      "Batch 3500, Train Loss:0.0240, Train ACC:99.8750\n",
      "Current time 3558.72 min\n",
      "Batch 4000, Train Loss:0.0241, Train ACC:99.8750\n",
      "Current time 3566.60 min\n",
      "Batch 4500, Train Loss:0.0242, Train ACC:99.8750\n",
      "Current time 3574.47 min\n",
      "Batch 5000, Train Loss:0.0241, Train ACC:99.8725\n",
      "Current time 3582.35 min\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Batch 5500, Train Loss:0.0241, Train ACC:99.8705\n",
      "Current time 3590.22 min\n",
      "Batch 6000, Train Loss:0.0251, Train ACC:99.8417\n",
      "Current time 3598.10 min\n",
      "Batch 6500, Train Loss:0.0253, Train ACC:99.8346\n",
      "Current time 3605.97 min\n",
      "Batch 7000, Train Loss:0.0254, Train ACC:99.8375\n",
      "Current time 3613.87 min\n",
      "Batch 7500, Train Loss:0.0254, Train ACC:99.8367\n",
      "Current time 3622.04 min\n",
      "Batch 8000, Train Loss:0.0251, Train ACC:99.8422\n",
      "Current time 3630.18 min\n",
      "Batch 8500, Train Loss:0.0251, Train ACC:99.8426\n",
      "Current time 3638.32 min\n",
      "Batch 9000, Train Loss:0.0252, Train ACC:99.8403\n",
      "Current time 3646.45 min\n",
      "Batch 9500, Train Loss:0.0251, Train ACC:99.8408\n",
      "Current time 3654.60 min\n",
      "Batch 10000, Train Loss:0.0251, Train ACC:99.8425\n",
      "Current time 3662.75 min\n",
      "Batch 10500, Train Loss:0.0252, Train ACC:99.8429\n",
      "Current time 3670.93 min\n",
      "Batch 11000, Train Loss:0.0258, Train ACC:99.8273\n",
      "Current time 3679.10 min\n",
      "Batch 11500, Train Loss:0.0259, Train ACC:99.8261\n",
      "Current time 3687.26 min\n",
      "train Loss:0.0518  Acc:99.8271%\n",
      "Finish time 3694.24 min\n",
      "Epoch 39/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0237, Train ACC:99.9000\n",
      "Current time 3702.28 min\n",
      "Batch 1000, Train Loss:0.0230, Train ACC:99.9125\n",
      "Current time 3710.30 min\n",
      "Batch 1500, Train Loss:0.0228, Train ACC:99.9167\n",
      "Current time 3718.37 min\n",
      "Batch 2000, Train Loss:0.0233, Train ACC:99.9000\n",
      "Current time 3726.42 min\n",
      "Batch 2500, Train Loss:0.0233, Train ACC:99.9050\n",
      "Current time 3734.48 min\n",
      "Batch 3000, Train Loss:0.0230, Train ACC:99.9125\n",
      "Current time 3742.50 min\n",
      "Batch 3500, Train Loss:0.0233, Train ACC:99.8893\n",
      "Current time 3750.54 min\n",
      "Batch 4000, Train Loss:0.0233, Train ACC:99.8875\n",
      "Current time 3758.60 min\n",
      "Batch 4500, Train Loss:0.0230, Train ACC:99.8972\n",
      "Current time 3766.64 min\n",
      "Batch 5000, Train Loss:0.0232, Train ACC:99.8925\n",
      "Current time 3774.66 min\n",
      "Batch 5500, Train Loss:0.0232, Train ACC:99.8955\n",
      "Current time 3782.71 min\n",
      "Batch 6000, Train Loss:0.0232, Train ACC:99.9000\n",
      "Current time 3790.75 min\n",
      "Batch 6500, Train Loss:0.0232, Train ACC:99.9000\n",
      "Current time 3798.81 min\n",
      "Batch 7000, Train Loss:0.0230, Train ACC:99.9036\n",
      "Current time 3806.84 min\n",
      "Batch 7500, Train Loss:0.0231, Train ACC:99.8950\n",
      "Current time 3814.88 min\n",
      "Batch 8000, Train Loss:0.0236, Train ACC:99.8828\n",
      "Current time 3822.92 min\n",
      "Batch 8500, Train Loss:0.0236, Train ACC:99.8838\n",
      "Current time 3830.97 min\n",
      "Batch 9000, Train Loss:0.0239, Train ACC:99.8764\n",
      "Current time 3839.01 min\n",
      "Batch 9500, Train Loss:0.0242, Train ACC:99.8711\n",
      "Current time 3847.11 min\n",
      "Batch 10000, Train Loss:0.0244, Train ACC:99.8650\n",
      "Current time 3855.17 min\n",
      "Batch 10500, Train Loss:0.0245, Train ACC:99.8595\n",
      "Current time 3863.21 min\n",
      "Batch 11000, Train Loss:0.0248, Train ACC:99.8568\n",
      "Current time 3871.23 min\n",
      "Batch 11500, Train Loss:0.0249, Train ACC:99.8522\n",
      "Current time 3879.27 min\n",
      "train Loss:0.0498  Acc:99.8533%\n",
      "Finish time 3886.16 min\n",
      "Epoch 40/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0264, Train ACC:99.7250\n",
      "Current time 3894.05 min\n",
      "Batch 1000, Train Loss:0.0242, Train ACC:99.8250\n",
      "Current time 3901.93 min\n",
      "Batch 1500, Train Loss:0.0235, Train ACC:99.8583\n",
      "Current time 3909.82 min\n",
      "Batch 2000, Train Loss:0.0236, Train ACC:99.8438\n",
      "Current time 3917.71 min\n",
      "Batch 2500, Train Loss:0.0242, Train ACC:99.8100\n",
      "Current time 3925.64 min\n",
      "Batch 3000, Train Loss:0.0243, Train ACC:99.8125\n",
      "Current time 3933.55 min\n",
      "Batch 3500, Train Loss:0.0240, Train ACC:99.8286\n",
      "Current time 3941.45 min\n",
      "Batch 4000, Train Loss:0.0243, Train ACC:99.8187\n",
      "Current time 3949.37 min\n",
      "Batch 4500, Train Loss:0.0240, Train ACC:99.8278\n",
      "Current time 3957.27 min\n",
      "Batch 5000, Train Loss:0.0239, Train ACC:99.8325\n",
      "Current time 3965.17 min\n",
      "Batch 5500, Train Loss:0.0239, Train ACC:99.8318\n",
      "Current time 3973.11 min\n",
      "Batch 6000, Train Loss:0.0248, Train ACC:99.8208\n",
      "Current time 3980.99 min\n",
      "Batch 6500, Train Loss:0.0248, Train ACC:99.8269\n",
      "Current time 3988.89 min\n",
      "Batch 7000, Train Loss:0.0247, Train ACC:99.8286\n",
      "Current time 3996.76 min\n",
      "Batch 7500, Train Loss:0.0246, Train ACC:99.8267\n",
      "Current time 4004.67 min\n",
      "Batch 8000, Train Loss:0.0250, Train ACC:99.8172\n",
      "Current time 4012.59 min\n",
      "Batch 8500, Train Loss:0.0249, Train ACC:99.8206\n",
      "Current time 4020.52 min\n",
      "Batch 9000, Train Loss:0.0250, Train ACC:99.8208\n",
      "Current time 4028.42 min\n",
      "Batch 9500, Train Loss:0.0252, Train ACC:99.8197\n",
      "Current time 4036.31 min\n",
      "Batch 10000, Train Loss:0.0252, Train ACC:99.8213\n",
      "Current time 4044.19 min\n",
      "Batch 10500, Train Loss:0.0253, Train ACC:99.8190\n",
      "Current time 4052.08 min\n",
      "Batch 11000, Train Loss:0.0252, Train ACC:99.8239\n",
      "Current time 4060.01 min\n",
      "Batch 11500, Train Loss:0.0251, Train ACC:99.8272\n",
      "Current time 4067.91 min\n",
      "train Loss:0.0503  Acc:99.8281%\n",
      "Finish time 4074.66 min\n",
      "Epoch 41/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0257, Train ACC:99.8000\n",
      "Current time 4082.57 min\n",
      "Batch 1000, Train Loss:0.0238, Train ACC:99.8625\n",
      "Current time 4090.47 min\n",
      "Batch 1500, Train Loss:0.0234, Train ACC:99.8750\n",
      "Current time 4098.37 min\n",
      "Batch 2000, Train Loss:0.0239, Train ACC:99.8688\n",
      "Current time 4106.26 min\n",
      "Batch 2500, Train Loss:0.0236, Train ACC:99.8850\n",
      "Current time 4114.15 min\n",
      "Batch 3000, Train Loss:0.0233, Train ACC:99.8917\n",
      "Current time 4122.06 min\n",
      "Batch 3500, Train Loss:0.0233, Train ACC:99.8893\n",
      "Current time 4129.99 min\n",
      "Batch 4000, Train Loss:0.0239, Train ACC:99.8688\n",
      "Current time 4137.91 min\n",
      "Batch 4500, Train Loss:0.0240, Train ACC:99.8694\n",
      "Current time 4145.81 min\n",
      "Batch 5000, Train Loss:0.0239, Train ACC:99.8725\n",
      "Current time 4153.70 min\n",
      "Batch 5500, Train Loss:0.0238, Train ACC:99.8705\n",
      "Current time 4161.63 min\n",
      "Batch 6000, Train Loss:0.0238, Train ACC:99.8708\n",
      "Current time 4169.55 min\n",
      "Batch 6500, Train Loss:0.0239, Train ACC:99.8712\n",
      "Current time 4177.45 min\n",
      "Batch 7000, Train Loss:0.0238, Train ACC:99.8750\n",
      "Current time 4185.40 min\n",
      "Batch 7500, Train Loss:0.0238, Train ACC:99.8750\n",
      "Current time 4193.30 min\n",
      "Batch 8000, Train Loss:0.0237, Train ACC:99.8797\n",
      "Current time 4201.21 min\n",
      "Batch 8500, Train Loss:0.0234, Train ACC:99.8853\n",
      "Current time 4209.14 min\n",
      "Batch 9000, Train Loss:0.0237, Train ACC:99.8847\n",
      "Current time 4217.08 min\n",
      "Batch 9500, Train Loss:0.0236, Train ACC:99.8855\n",
      "Current time 4225.26 min\n",
      "Batch 10000, Train Loss:0.0236, Train ACC:99.8875\n",
      "Current time 4233.44 min\n",
      "Batch 10500, Train Loss:0.0237, Train ACC:99.8810\n",
      "Current time 4241.61 min\n",
      "Batch 11000, Train Loss:0.0239, Train ACC:99.8795\n",
      "Current time 4249.80 min\n",
      "Batch 11500, Train Loss:0.0238, Train ACC:99.8826\n",
      "Current time 4257.96 min\n",
      "train Loss:0.0474  Acc:99.8805%\n",
      "Finish time 4264.94 min\n",
      "Epoch 42/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0225, Train ACC:99.8750\n",
      "Current time 4273.02 min\n",
      "Batch 1000, Train Loss:0.0229, Train ACC:99.8750\n",
      "Current time 4281.09 min\n",
      "Batch 1500, Train Loss:0.0248, Train ACC:99.8167\n",
      "Current time 4289.13 min\n",
      "Batch 2000, Train Loss:0.0239, Train ACC:99.8438\n",
      "Current time 4297.17 min\n",
      "Batch 2500, Train Loss:0.0241, Train ACC:99.8500\n",
      "Current time 4305.19 min\n",
      "Batch 3000, Train Loss:0.0235, Train ACC:99.8667\n",
      "Current time 4313.25 min\n",
      "Batch 3500, Train Loss:0.0233, Train ACC:99.8786\n",
      "Current time 4321.30 min\n",
      "Batch 4000, Train Loss:0.0235, Train ACC:99.8719\n",
      "Current time 4329.36 min\n",
      "Batch 4500, Train Loss:0.0245, Train ACC:99.8528\n",
      "Current time 4337.42 min\n",
      "Batch 5000, Train Loss:0.0251, Train ACC:99.8425\n",
      "Current time 4345.46 min\n",
      "Batch 5500, Train Loss:0.0261, Train ACC:99.8250\n",
      "Current time 4353.52 min\n",
      "Batch 6000, Train Loss:0.0260, Train ACC:99.8333\n",
      "Current time 4361.59 min\n",
      "Batch 6500, Train Loss:0.0256, Train ACC:99.8423\n",
      "Current time 4369.64 min\n",
      "Batch 7000, Train Loss:0.0254, Train ACC:99.8446\n",
      "Current time 4377.69 min\n",
      "Batch 7500, Train Loss:0.0252, Train ACC:99.8467\n",
      "Current time 4385.74 min\n",
      "Batch 8000, Train Loss:0.0254, Train ACC:99.8453\n",
      "Current time 4393.79 min\n",
      "Batch 8500, Train Loss:0.0251, Train ACC:99.8515\n",
      "Current time 4401.84 min\n",
      "Batch 9000, Train Loss:0.0250, Train ACC:99.8528\n",
      "Current time 4409.87 min\n",
      "Batch 9500, Train Loss:0.0249, Train ACC:99.8553\n",
      "Current time 4417.91 min\n",
      "Batch 10000, Train Loss:0.0252, Train ACC:99.8475\n",
      "Current time 4425.99 min\n",
      "Batch 10500, Train Loss:0.0255, Train ACC:99.8405\n",
      "Current time 4434.04 min\n",
      "Batch 11000, Train Loss:0.0253, Train ACC:99.8455\n",
      "Current time 4442.11 min\n",
      "Batch 11500, Train Loss:0.0253, Train ACC:99.8446\n",
      "Current time 4450.19 min\n",
      "train Loss:0.0507  Acc:99.8428%\n",
      "Finish time 4457.12 min\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 43/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0250, Train ACC:99.8750\n",
      "Current time 4464.99 min\n",
      "Batch 1000, Train Loss:0.0230, Train ACC:99.9125\n",
      "Current time 4472.88 min\n",
      "Batch 1500, Train Loss:0.0217, Train ACC:99.9333\n",
      "Current time 4480.75 min\n",
      "Batch 2000, Train Loss:0.0216, Train ACC:99.9313\n",
      "Current time 4488.60 min\n",
      "Batch 2500, Train Loss:0.0212, Train ACC:99.9350\n",
      "Current time 4496.50 min\n",
      "Batch 3000, Train Loss:0.0222, Train ACC:99.9083\n",
      "Current time 4504.37 min\n",
      "Batch 3500, Train Loss:0.0225, Train ACC:99.9000\n",
      "Current time 4512.43 min\n",
      "Batch 4000, Train Loss:0.0226, Train ACC:99.9031\n",
      "Current time 4520.49 min\n",
      "Batch 4500, Train Loss:0.0230, Train ACC:99.8917\n",
      "Current time 4528.56 min\n",
      "Batch 5000, Train Loss:0.0231, Train ACC:99.8950\n",
      "Current time 4536.56 min\n",
      "Batch 5500, Train Loss:0.0231, Train ACC:99.8955\n",
      "Current time 4544.50 min\n",
      "Batch 6000, Train Loss:0.0233, Train ACC:99.8875\n",
      "Current time 4552.42 min\n",
      "Batch 6500, Train Loss:0.0232, Train ACC:99.8865\n",
      "Current time 4560.35 min\n",
      "Batch 7000, Train Loss:0.0235, Train ACC:99.8839\n",
      "Current time 4568.25 min\n",
      "Batch 7500, Train Loss:0.0234, Train ACC:99.8867\n",
      "Current time 4576.26 min\n",
      "Batch 8000, Train Loss:0.0232, Train ACC:99.8891\n",
      "Current time 4584.43 min\n",
      "Batch 8500, Train Loss:0.0232, Train ACC:99.8824\n",
      "Current time 4592.55 min\n",
      "Batch 9000, Train Loss:0.0234, Train ACC:99.8750\n",
      "Current time 4600.67 min\n",
      "Batch 9500, Train Loss:0.0233, Train ACC:99.8763\n",
      "Current time 4608.81 min\n",
      "Batch 10000, Train Loss:0.0237, Train ACC:99.8713\n",
      "Current time 4616.94 min\n",
      "Batch 10500, Train Loss:0.0236, Train ACC:99.8750\n",
      "Current time 4625.07 min\n",
      "Batch 11000, Train Loss:0.0237, Train ACC:99.8693\n",
      "Current time 4633.18 min\n",
      "Batch 11500, Train Loss:0.0240, Train ACC:99.8609\n",
      "Current time 4641.32 min\n",
      "train Loss:0.0479  Acc:99.8617%\n",
      "Finish time 4648.27 min\n",
      "Epoch 44/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0215, Train ACC:99.9250\n",
      "Current time 4656.31 min\n",
      "Batch 1000, Train Loss:0.0253, Train ACC:99.8375\n",
      "Current time 4664.31 min\n",
      "Batch 1500, Train Loss:0.0246, Train ACC:99.8417\n",
      "Current time 4672.32 min\n",
      "Batch 2000, Train Loss:0.0242, Train ACC:99.8500\n",
      "Current time 4680.31 min\n",
      "Batch 2500, Train Loss:0.0234, Train ACC:99.8750\n",
      "Current time 4688.32 min\n",
      "Batch 3000, Train Loss:0.0231, Train ACC:99.8833\n",
      "Current time 4696.36 min\n",
      "Batch 3500, Train Loss:0.0227, Train ACC:99.8964\n",
      "Current time 4704.37 min\n",
      "Batch 4000, Train Loss:0.0224, Train ACC:99.8937\n",
      "Current time 4712.36 min\n",
      "Batch 4500, Train Loss:0.0225, Train ACC:99.8861\n",
      "Current time 4720.37 min\n",
      "Batch 5000, Train Loss:0.0227, Train ACC:99.8850\n",
      "Current time 4728.37 min\n",
      "Batch 5500, Train Loss:0.0232, Train ACC:99.8659\n",
      "Current time 4736.40 min\n",
      "Batch 6000, Train Loss:0.0231, Train ACC:99.8708\n",
      "Current time 4744.41 min\n",
      "Batch 6500, Train Loss:0.0233, Train ACC:99.8615\n",
      "Current time 4752.42 min\n",
      "Batch 7000, Train Loss:0.0238, Train ACC:99.8554\n",
      "Current time 4760.45 min\n",
      "Batch 7500, Train Loss:0.0238, Train ACC:99.8517\n",
      "Current time 4768.48 min\n",
      "Batch 8000, Train Loss:0.0241, Train ACC:99.8484\n",
      "Current time 4776.51 min\n",
      "Batch 8500, Train Loss:0.0239, Train ACC:99.8515\n",
      "Current time 4784.57 min\n",
      "Batch 9000, Train Loss:0.0238, Train ACC:99.8569\n",
      "Current time 4792.62 min\n",
      "Batch 9500, Train Loss:0.0238, Train ACC:99.8592\n",
      "Current time 4800.65 min\n",
      "Batch 10000, Train Loss:0.0237, Train ACC:99.8575\n",
      "Current time 4808.68 min\n",
      "Batch 10500, Train Loss:0.0239, Train ACC:99.8583\n",
      "Current time 4816.72 min\n",
      "Batch 11000, Train Loss:0.0241, Train ACC:99.8534\n",
      "Current time 4824.78 min\n",
      "Batch 11500, Train Loss:0.0240, Train ACC:99.8543\n",
      "Current time 4832.78 min\n",
      "train Loss:0.0478  Acc:99.8543%\n",
      "Finish time 4839.67 min\n",
      "Epoch 45/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0217, Train ACC:99.9250\n",
      "Current time 4847.57 min\n",
      "Batch 1000, Train Loss:0.0212, Train ACC:99.9250\n",
      "Current time 4855.43 min\n",
      "Batch 1500, Train Loss:0.0207, Train ACC:99.9333\n",
      "Current time 4863.33 min\n",
      "Batch 2000, Train Loss:0.0204, Train ACC:99.9437\n",
      "Current time 4871.22 min\n",
      "Batch 2500, Train Loss:0.0206, Train ACC:99.9450\n",
      "Current time 4879.10 min\n",
      "Batch 3000, Train Loss:0.0205, Train ACC:99.9500\n",
      "Current time 4887.01 min\n",
      "Batch 3500, Train Loss:0.0209, Train ACC:99.9393\n",
      "Current time 4894.91 min\n",
      "Batch 4000, Train Loss:0.0219, Train ACC:99.9219\n",
      "Current time 4902.76 min\n",
      "Batch 4500, Train Loss:0.0223, Train ACC:99.9111\n",
      "Current time 4910.63 min\n",
      "Batch 5000, Train Loss:0.0221, Train ACC:99.9075\n",
      "Current time 4918.52 min\n",
      "Batch 5500, Train Loss:0.0223, Train ACC:99.9045\n",
      "Current time 4926.39 min\n",
      "Batch 6000, Train Loss:0.0227, Train ACC:99.8896\n",
      "Current time 4934.29 min\n",
      "Batch 6500, Train Loss:0.0227, Train ACC:99.8904\n",
      "Current time 4942.17 min\n",
      "Batch 7000, Train Loss:0.0227, Train ACC:99.8929\n",
      "Current time 4950.07 min\n",
      "Batch 7500, Train Loss:0.0226, Train ACC:99.8917\n",
      "Current time 4957.97 min\n",
      "Batch 8000, Train Loss:0.0228, Train ACC:99.8859\n",
      "Current time 4965.83 min\n",
      "Batch 8500, Train Loss:0.0229, Train ACC:99.8838\n",
      "Current time 4973.70 min\n",
      "Batch 9000, Train Loss:0.0231, Train ACC:99.8833\n",
      "Current time 4981.59 min\n",
      "Batch 9500, Train Loss:0.0230, Train ACC:99.8842\n",
      "Current time 4989.49 min\n",
      "Batch 10000, Train Loss:0.0229, Train ACC:99.8863\n",
      "Current time 4997.39 min\n",
      "Batch 10500, Train Loss:0.0230, Train ACC:99.8857\n",
      "Current time 5005.28 min\n",
      "Batch 11000, Train Loss:0.0229, Train ACC:99.8875\n",
      "Current time 5013.16 min\n",
      "Batch 11500, Train Loss:0.0228, Train ACC:99.8913\n",
      "Current time 5021.07 min\n",
      "train Loss:0.0457  Acc:99.8910%\n",
      "Finish time 5027.85 min\n",
      "Epoch 46/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0278, Train ACC:99.6250\n",
      "Current time 5035.71 min\n",
      "Batch 1000, Train Loss:0.0253, Train ACC:99.7625\n",
      "Current time 5043.62 min\n",
      "Batch 1500, Train Loss:0.0244, Train ACC:99.8167\n",
      "Current time 5051.53 min\n",
      "Batch 2000, Train Loss:0.0236, Train ACC:99.8563\n",
      "Current time 5059.42 min\n",
      "Batch 2500, Train Loss:0.0241, Train ACC:99.8500\n",
      "Current time 5067.31 min\n",
      "Batch 3000, Train Loss:0.0235, Train ACC:99.8667\n",
      "Current time 5075.23 min\n",
      "Batch 3500, Train Loss:0.0234, Train ACC:99.8679\n",
      "Current time 5083.12 min\n",
      "Batch 4000, Train Loss:0.0233, Train ACC:99.8656\n",
      "Current time 5090.99 min\n",
      "Batch 4500, Train Loss:0.0232, Train ACC:99.8667\n",
      "Current time 5098.86 min\n",
      "Batch 5000, Train Loss:0.0236, Train ACC:99.8550\n",
      "Current time 5106.74 min\n",
      "Batch 5500, Train Loss:0.0233, Train ACC:99.8636\n",
      "Current time 5114.64 min\n",
      "Batch 6000, Train Loss:0.0233, Train ACC:99.8604\n",
      "Current time 5122.52 min\n",
      "Batch 6500, Train Loss:0.0232, Train ACC:99.8615\n",
      "Current time 5130.40 min\n",
      "Batch 7000, Train Loss:0.0231, Train ACC:99.8643\n",
      "Current time 5138.25 min\n",
      "Batch 7500, Train Loss:0.0231, Train ACC:99.8600\n",
      "Current time 5146.16 min\n",
      "Batch 8000, Train Loss:0.0235, Train ACC:99.8469\n",
      "Current time 5154.02 min\n",
      "Batch 8500, Train Loss:0.0235, Train ACC:99.8471\n",
      "Current time 5161.90 min\n",
      "Batch 9000, Train Loss:0.0239, Train ACC:99.8431\n",
      "Current time 5169.77 min\n",
      "Batch 9500, Train Loss:0.0240, Train ACC:99.8382\n",
      "Current time 5177.64 min\n",
      "Batch 10000, Train Loss:0.0240, Train ACC:99.8400\n",
      "Current time 5185.54 min\n",
      "Batch 10500, Train Loss:0.0240, Train ACC:99.8417\n",
      "Current time 5193.39 min\n",
      "Batch 11000, Train Loss:0.0240, Train ACC:99.8455\n",
      "Current time 5201.25 min\n",
      "Batch 11500, Train Loss:0.0239, Train ACC:99.8467\n",
      "Current time 5209.14 min\n",
      "train Loss:0.0478  Acc:99.8481%\n",
      "Finish time 5215.92 min\n",
      "Epoch 47/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0237, Train ACC:99.8500\n",
      "Current time 5223.83 min\n",
      "Batch 1000, Train Loss:0.0226, Train ACC:99.8875\n",
      "Current time 5231.71 min\n",
      "Batch 1500, Train Loss:0.0235, Train ACC:99.8833\n",
      "Current time 5239.59 min\n",
      "Batch 2000, Train Loss:0.0225, Train ACC:99.9000\n",
      "Current time 5247.48 min\n",
      "Batch 2500, Train Loss:0.0228, Train ACC:99.9000\n",
      "Current time 5255.39 min\n",
      "Batch 3000, Train Loss:0.0231, Train ACC:99.8917\n",
      "Current time 5263.28 min\n",
      "Batch 3500, Train Loss:0.0238, Train ACC:99.8679\n",
      "Current time 5271.18 min\n",
      "Batch 4000, Train Loss:0.0234, Train ACC:99.8781\n",
      "Current time 5279.12 min\n",
      "Batch 4500, Train Loss:0.0230, Train ACC:99.8889\n",
      "Current time 5287.01 min\n",
      "Batch 5000, Train Loss:0.0227, Train ACC:99.8925\n",
      "Current time 5294.95 min\n",
      "Batch 5500, Train Loss:0.0229, Train ACC:99.8886\n",
      "Current time 5302.85 min\n",
      "Batch 6000, Train Loss:0.0229, Train ACC:99.8854\n",
      "Current time 5310.72 min\n",
      "Batch 6500, Train Loss:0.0231, Train ACC:99.8788\n",
      "Current time 5318.60 min\n",
      "Batch 7000, Train Loss:0.0231, Train ACC:99.8750\n",
      "Current time 5326.48 min\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Batch 7500, Train Loss:0.0230, Train ACC:99.8750\n",
      "Current time 5334.38 min\n",
      "Batch 8000, Train Loss:0.0232, Train ACC:99.8703\n",
      "Current time 5342.27 min\n",
      "Batch 8500, Train Loss:0.0231, Train ACC:99.8706\n",
      "Current time 5350.19 min\n",
      "Batch 9000, Train Loss:0.0231, Train ACC:99.8694\n",
      "Current time 5358.10 min\n",
      "Batch 9500, Train Loss:0.0232, Train ACC:99.8605\n",
      "Current time 5366.00 min\n",
      "Batch 10000, Train Loss:0.0231, Train ACC:99.8650\n",
      "Current time 5373.92 min\n",
      "Batch 10500, Train Loss:0.0230, Train ACC:99.8679\n",
      "Current time 5381.84 min\n",
      "Batch 11000, Train Loss:0.0229, Train ACC:99.8705\n",
      "Current time 5389.73 min\n",
      "Batch 11500, Train Loss:0.0229, Train ACC:99.8696\n",
      "Current time 5397.66 min\n",
      "train Loss:0.0459  Acc:99.8701%\n",
      "Finish time 5404.44 min\n",
      "Epoch 48/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0196, Train ACC:99.9500\n",
      "Current time 5412.36 min\n",
      "Batch 1000, Train Loss:0.0211, Train ACC:99.9000\n",
      "Current time 5420.29 min\n",
      "Batch 1500, Train Loss:0.0209, Train ACC:99.9167\n",
      "Current time 5428.19 min\n",
      "Batch 2000, Train Loss:0.0223, Train ACC:99.8625\n",
      "Current time 5436.12 min\n",
      "Batch 2500, Train Loss:0.0219, Train ACC:99.8800\n",
      "Current time 5444.04 min\n",
      "Batch 3000, Train Loss:0.0218, Train ACC:99.8875\n",
      "Current time 5451.95 min\n",
      "Batch 3500, Train Loss:0.0225, Train ACC:99.8750\n",
      "Current time 5459.87 min\n",
      "Batch 4000, Train Loss:0.0225, Train ACC:99.8781\n",
      "Current time 5467.79 min\n",
      "Batch 4500, Train Loss:0.0226, Train ACC:99.8750\n",
      "Current time 5475.70 min\n",
      "Batch 5000, Train Loss:0.0224, Train ACC:99.8750\n",
      "Current time 5483.61 min\n",
      "Batch 5500, Train Loss:0.0224, Train ACC:99.8750\n",
      "Current time 5491.54 min\n",
      "Batch 6000, Train Loss:0.0225, Train ACC:99.8750\n",
      "Current time 5499.43 min\n",
      "Batch 6500, Train Loss:0.0227, Train ACC:99.8673\n",
      "Current time 5507.37 min\n",
      "Batch 7000, Train Loss:0.0228, Train ACC:99.8643\n",
      "Current time 5515.26 min\n",
      "Batch 7500, Train Loss:0.0231, Train ACC:99.8567\n",
      "Current time 5523.18 min\n",
      "Batch 8000, Train Loss:0.0231, Train ACC:99.8578\n",
      "Current time 5531.11 min\n",
      "Batch 8500, Train Loss:0.0232, Train ACC:99.8603\n",
      "Current time 5539.01 min\n",
      "Batch 9000, Train Loss:0.0230, Train ACC:99.8639\n",
      "Current time 5546.96 min\n",
      "Batch 9500, Train Loss:0.0230, Train ACC:99.8658\n",
      "Current time 5554.89 min\n",
      "Batch 10000, Train Loss:0.0229, Train ACC:99.8688\n",
      "Current time 5562.82 min\n",
      "Batch 10500, Train Loss:0.0229, Train ACC:99.8714\n",
      "Current time 5570.93 min\n",
      "Batch 11000, Train Loss:0.0228, Train ACC:99.8739\n",
      "Current time 5579.13 min\n",
      "Batch 11500, Train Loss:0.0230, Train ACC:99.8717\n",
      "Current time 5587.31 min\n",
      "train Loss:0.0466  Acc:99.8627%\n",
      "Finish time 5594.31 min\n",
      "Epoch 49/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0264, Train ACC:99.7250\n",
      "Current time 5602.48 min\n",
      "Batch 1000, Train Loss:0.0240, Train ACC:99.8125\n",
      "Current time 5610.64 min\n",
      "Batch 1500, Train Loss:0.0262, Train ACC:99.8000\n",
      "Current time 5618.79 min\n",
      "Batch 2000, Train Loss:0.0257, Train ACC:99.8187\n",
      "Current time 5626.93 min\n",
      "Batch 2500, Train Loss:0.0253, Train ACC:99.8250\n",
      "Current time 5635.06 min\n",
      "Batch 3000, Train Loss:0.0245, Train ACC:99.8250\n",
      "Current time 5643.20 min\n",
      "Batch 3500, Train Loss:0.0237, Train ACC:99.8429\n",
      "Current time 5651.35 min\n",
      "Batch 4000, Train Loss:0.0234, Train ACC:99.8500\n",
      "Current time 5659.56 min\n",
      "Batch 4500, Train Loss:0.0233, Train ACC:99.8528\n",
      "Current time 5667.77 min\n",
      "Batch 5000, Train Loss:0.0231, Train ACC:99.8550\n",
      "Current time 5676.00 min\n",
      "Batch 5500, Train Loss:0.0233, Train ACC:99.8545\n",
      "Current time 5684.23 min\n",
      "Batch 6000, Train Loss:0.0234, Train ACC:99.8479\n",
      "Current time 5692.43 min\n",
      "Batch 6500, Train Loss:0.0234, Train ACC:99.8442\n",
      "Current time 5700.58 min\n",
      "Batch 7000, Train Loss:0.0233, Train ACC:99.8482\n",
      "Current time 5708.77 min\n",
      "Batch 7500, Train Loss:0.0232, Train ACC:99.8517\n",
      "Current time 5716.93 min\n",
      "Batch 8000, Train Loss:0.0232, Train ACC:99.8500\n",
      "Current time 5725.09 min\n",
      "Batch 8500, Train Loss:0.0230, Train ACC:99.8544\n",
      "Current time 5733.26 min\n",
      "Batch 9000, Train Loss:0.0228, Train ACC:99.8611\n",
      "Current time 5741.42 min\n",
      "Batch 9500, Train Loss:0.0226, Train ACC:99.8671\n",
      "Current time 5749.59 min\n",
      "Batch 10000, Train Loss:0.0228, Train ACC:99.8637\n",
      "Current time 5757.79 min\n",
      "Batch 10500, Train Loss:0.0228, Train ACC:99.8619\n",
      "Current time 5765.96 min\n",
      "Batch 11000, Train Loss:0.0226, Train ACC:99.8648\n",
      "Current time 5774.14 min\n",
      "Batch 11500, Train Loss:0.0225, Train ACC:99.8663\n",
      "Current time 5782.32 min\n",
      "train Loss:0.0451  Acc:99.8680%\n",
      "Finish time 5789.33 min\n",
      "Epoch 50/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0247, Train ACC:99.8500\n",
      "Current time 5797.36 min\n",
      "Batch 1000, Train Loss:0.0224, Train ACC:99.8875\n",
      "Current time 5805.40 min\n",
      "Batch 1500, Train Loss:0.0225, Train ACC:99.8833\n",
      "Current time 5813.42 min\n",
      "Batch 2000, Train Loss:0.0221, Train ACC:99.8937\n",
      "Current time 5821.41 min\n",
      "Batch 2500, Train Loss:0.0226, Train ACC:99.8700\n",
      "Current time 5829.43 min\n",
      "Batch 3000, Train Loss:0.0221, Train ACC:99.8875\n",
      "Current time 5837.44 min\n",
      "Batch 3500, Train Loss:0.0217, Train ACC:99.9000\n",
      "Current time 5845.46 min\n",
      "Batch 4000, Train Loss:0.0213, Train ACC:99.9062\n",
      "Current time 5853.46 min\n",
      "Batch 4500, Train Loss:0.0212, Train ACC:99.9111\n",
      "Current time 5861.47 min\n",
      "Batch 5000, Train Loss:0.0211, Train ACC:99.9175\n",
      "Current time 5869.46 min\n",
      "Batch 5500, Train Loss:0.0213, Train ACC:99.9136\n",
      "Current time 5877.47 min\n",
      "Batch 6000, Train Loss:0.0216, Train ACC:99.9062\n",
      "Current time 5885.48 min\n",
      "Batch 6500, Train Loss:0.0221, Train ACC:99.8942\n",
      "Current time 5893.51 min\n",
      "Batch 7000, Train Loss:0.0221, Train ACC:99.8929\n",
      "Current time 5901.50 min\n",
      "Batch 7500, Train Loss:0.0226, Train ACC:99.8833\n",
      "Current time 5909.52 min\n",
      "Batch 8000, Train Loss:0.0225, Train ACC:99.8844\n",
      "Current time 5917.50 min\n",
      "Batch 8500, Train Loss:0.0223, Train ACC:99.8868\n",
      "Current time 5925.50 min\n",
      "Batch 9000, Train Loss:0.0224, Train ACC:99.8875\n",
      "Current time 5933.48 min\n",
      "Batch 9500, Train Loss:0.0223, Train ACC:99.8895\n",
      "Current time 5941.48 min\n",
      "Batch 10000, Train Loss:0.0223, Train ACC:99.8888\n",
      "Current time 5949.46 min\n",
      "Batch 10500, Train Loss:0.0224, Train ACC:99.8845\n",
      "Current time 5957.48 min\n",
      "Batch 11000, Train Loss:0.0225, Train ACC:99.8830\n",
      "Current time 5965.52 min\n",
      "Batch 11500, Train Loss:0.0224, Train ACC:99.8859\n",
      "Current time 5973.60 min\n",
      "train Loss:0.0447  Acc:99.8858%\n",
      "Finish time 5980.50 min\n",
      "Epoch 51/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0225, Train ACC:99.9250\n",
      "Current time 5988.43 min\n",
      "Batch 1000, Train Loss:0.0231, Train ACC:99.9125\n",
      "Current time 5996.40 min\n",
      "Batch 1500, Train Loss:0.0232, Train ACC:99.9000\n",
      "Current time 6004.41 min\n",
      "Batch 2000, Train Loss:0.0223, Train ACC:99.9125\n",
      "Current time 6012.50 min\n",
      "Batch 2500, Train Loss:0.0226, Train ACC:99.9050\n",
      "Current time 6020.53 min\n",
      "Batch 3000, Train Loss:0.0228, Train ACC:99.8917\n",
      "Current time 6028.49 min\n",
      "Batch 3500, Train Loss:0.0234, Train ACC:99.8750\n",
      "Current time 6036.65 min\n",
      "Batch 4000, Train Loss:0.0230, Train ACC:99.8781\n",
      "Current time 6044.88 min\n",
      "Batch 4500, Train Loss:0.0228, Train ACC:99.8806\n",
      "Current time 6053.08 min\n",
      "Batch 5000, Train Loss:0.0231, Train ACC:99.8750\n",
      "Current time 6061.28 min\n",
      "Batch 5500, Train Loss:0.0233, Train ACC:99.8705\n",
      "Current time 6069.48 min\n",
      "Batch 6000, Train Loss:0.0231, Train ACC:99.8771\n",
      "Current time 6077.67 min\n",
      "Batch 6500, Train Loss:0.0230, Train ACC:99.8769\n",
      "Current time 6085.86 min\n",
      "Batch 7000, Train Loss:0.0229, Train ACC:99.8804\n",
      "Current time 6094.03 min\n",
      "Batch 7500, Train Loss:0.0229, Train ACC:99.8767\n",
      "Current time 6102.20 min\n",
      "Batch 8000, Train Loss:0.0228, Train ACC:99.8781\n",
      "Current time 6110.40 min\n",
      "Batch 8500, Train Loss:0.0229, Train ACC:99.8735\n",
      "Current time 6118.57 min\n",
      "Batch 9000, Train Loss:0.0231, Train ACC:99.8722\n",
      "Current time 6126.76 min\n",
      "Batch 9500, Train Loss:0.0231, Train ACC:99.8697\n",
      "Current time 6134.95 min\n",
      "Batch 10000, Train Loss:0.0230, Train ACC:99.8738\n",
      "Current time 6143.14 min\n",
      "Batch 10500, Train Loss:0.0230, Train ACC:99.8750\n",
      "Current time 6151.32 min\n",
      "Batch 11000, Train Loss:0.0228, Train ACC:99.8761\n",
      "Current time 6159.50 min\n",
      "Batch 11500, Train Loss:0.0229, Train ACC:99.8783\n",
      "Current time 6167.69 min\n",
      "train Loss:0.0455  Acc:99.8805%\n",
      "Finish time 6174.74 min\n",
      "Epoch 52/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0195, Train ACC:99.9250\n",
      "Current time 6182.74 min\n",
      "Batch 1000, Train Loss:0.0252, Train ACC:99.8250\n",
      "Current time 6190.75 min\n",
      "Batch 1500, Train Loss:0.0242, Train ACC:99.8417\n",
      "Current time 6198.76 min\n",
      "Batch 2000, Train Loss:0.0240, Train ACC:99.8312\n",
      "Current time 6206.79 min\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Batch 2500, Train Loss:0.0247, Train ACC:99.8200\n",
      "Current time 6214.79 min\n",
      "Batch 3000, Train Loss:0.0239, Train ACC:99.8458\n",
      "Current time 6222.79 min\n",
      "Batch 3500, Train Loss:0.0237, Train ACC:99.8536\n",
      "Current time 6230.82 min\n",
      "Batch 4000, Train Loss:0.0237, Train ACC:99.8625\n",
      "Current time 6238.83 min\n",
      "Batch 4500, Train Loss:0.0240, Train ACC:99.8444\n",
      "Current time 6246.83 min\n",
      "Batch 5000, Train Loss:0.0237, Train ACC:99.8525\n",
      "Current time 6254.85 min\n",
      "Batch 5500, Train Loss:0.0236, Train ACC:99.8523\n",
      "Current time 6262.84 min\n",
      "Batch 6000, Train Loss:0.0241, Train ACC:99.8375\n",
      "Current time 6270.83 min\n",
      "Batch 6500, Train Loss:0.0240, Train ACC:99.8365\n",
      "Current time 6278.83 min\n",
      "Batch 7000, Train Loss:0.0237, Train ACC:99.8446\n",
      "Current time 6286.84 min\n",
      "Batch 7500, Train Loss:0.0235, Train ACC:99.8467\n",
      "Current time 6294.83 min\n",
      "Batch 8000, Train Loss:0.0233, Train ACC:99.8531\n",
      "Current time 6302.84 min\n",
      "Batch 8500, Train Loss:0.0235, Train ACC:99.8529\n",
      "Current time 6310.84 min\n",
      "Batch 9000, Train Loss:0.0235, Train ACC:99.8528\n",
      "Current time 6318.86 min\n",
      "Batch 9500, Train Loss:0.0239, Train ACC:99.8500\n",
      "Current time 6326.85 min\n",
      "Batch 10000, Train Loss:0.0238, Train ACC:99.8525\n",
      "Current time 6334.88 min\n",
      "Batch 10500, Train Loss:0.0238, Train ACC:99.8512\n",
      "Current time 6342.92 min\n",
      "Batch 11000, Train Loss:0.0237, Train ACC:99.8557\n",
      "Current time 6350.91 min\n",
      "Batch 11500, Train Loss:0.0235, Train ACC:99.8598\n",
      "Current time 6358.92 min\n",
      "train Loss:0.0469  Acc:99.8596%\n",
      "Finish time 6365.80 min\n",
      "Epoch 53/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0216, Train ACC:99.9250\n",
      "Current time 6373.75 min\n",
      "Batch 1000, Train Loss:0.0220, Train ACC:99.8875\n",
      "Current time 6381.66 min\n",
      "Batch 1500, Train Loss:0.0224, Train ACC:99.8917\n",
      "Current time 6389.56 min\n",
      "Batch 2000, Train Loss:0.0220, Train ACC:99.8937\n",
      "Current time 6397.53 min\n",
      "Batch 2500, Train Loss:0.0216, Train ACC:99.9000\n",
      "Current time 6405.47 min\n",
      "Batch 3000, Train Loss:0.0213, Train ACC:99.9125\n",
      "Current time 6413.41 min\n",
      "Batch 3500, Train Loss:0.0213, Train ACC:99.9071\n",
      "Current time 6421.34 min\n",
      "Batch 4000, Train Loss:0.0219, Train ACC:99.8875\n",
      "Current time 6429.30 min\n",
      "Batch 4500, Train Loss:0.0221, Train ACC:99.8778\n",
      "Current time 6437.22 min\n",
      "Batch 5000, Train Loss:0.0220, Train ACC:99.8850\n",
      "Current time 6445.14 min\n",
      "Batch 5500, Train Loss:0.0218, Train ACC:99.8932\n",
      "Current time 6453.05 min\n",
      "Batch 6000, Train Loss:0.0216, Train ACC:99.8937\n",
      "Current time 6460.99 min\n",
      "Batch 6500, Train Loss:0.0217, Train ACC:99.8846\n",
      "Current time 6468.95 min\n",
      "Batch 7000, Train Loss:0.0218, Train ACC:99.8839\n",
      "Current time 6476.86 min\n",
      "Batch 7500, Train Loss:0.0218, Train ACC:99.8883\n",
      "Current time 6484.79 min\n",
      "Batch 8000, Train Loss:0.0218, Train ACC:99.8922\n",
      "Current time 6492.72 min\n",
      "Batch 8500, Train Loss:0.0223, Train ACC:99.8853\n",
      "Current time 6500.65 min\n",
      "Batch 9000, Train Loss:0.0222, Train ACC:99.8875\n",
      "Current time 6508.58 min\n",
      "Batch 9500, Train Loss:0.0222, Train ACC:99.8868\n",
      "Current time 6516.52 min\n",
      "Batch 10000, Train Loss:0.0221, Train ACC:99.8900\n",
      "Current time 6524.44 min\n",
      "Batch 10500, Train Loss:0.0220, Train ACC:99.8929\n",
      "Current time 6532.41 min\n",
      "Batch 11000, Train Loss:0.0219, Train ACC:99.8920\n",
      "Current time 6540.35 min\n",
      "Batch 11500, Train Loss:0.0220, Train ACC:99.8902\n",
      "Current time 6548.29 min\n",
      "train Loss:0.0439  Acc:99.8910%\n",
      "Finish time 6555.14 min\n",
      "Epoch 54/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0187, Train ACC:99.9750\n",
      "Current time 6563.07 min\n",
      "Batch 1000, Train Loss:0.0217, Train ACC:99.9125\n",
      "Current time 6571.02 min\n",
      "Batch 1500, Train Loss:0.0217, Train ACC:99.9250\n",
      "Current time 6578.98 min\n",
      "Batch 2000, Train Loss:0.0212, Train ACC:99.9250\n",
      "Current time 6586.92 min\n",
      "Batch 2500, Train Loss:0.0208, Train ACC:99.9350\n",
      "Current time 6594.86 min\n",
      "Batch 3000, Train Loss:0.0203, Train ACC:99.9458\n",
      "Current time 6602.77 min\n",
      "Batch 3500, Train Loss:0.0205, Train ACC:99.9464\n",
      "Current time 6610.73 min\n",
      "Batch 4000, Train Loss:0.0208, Train ACC:99.9406\n",
      "Current time 6618.67 min\n",
      "Batch 4500, Train Loss:0.0210, Train ACC:99.9361\n",
      "Current time 6626.61 min\n",
      "Batch 5000, Train Loss:0.0210, Train ACC:99.9350\n",
      "Current time 6634.54 min\n",
      "Batch 5500, Train Loss:0.0213, Train ACC:99.9273\n",
      "Current time 6642.47 min\n",
      "Batch 6000, Train Loss:0.0212, Train ACC:99.9250\n",
      "Current time 6650.42 min\n",
      "Batch 6500, Train Loss:0.0216, Train ACC:99.9115\n",
      "Current time 6658.35 min\n",
      "Batch 7000, Train Loss:0.0223, Train ACC:99.9054\n",
      "Current time 6666.24 min\n",
      "Batch 7500, Train Loss:0.0225, Train ACC:99.8950\n",
      "Current time 6674.11 min\n",
      "Batch 8000, Train Loss:0.0225, Train ACC:99.8969\n",
      "Current time 6681.99 min\n",
      "Batch 8500, Train Loss:0.0227, Train ACC:99.8941\n",
      "Current time 6689.86 min\n",
      "Batch 9000, Train Loss:0.0225, Train ACC:99.8958\n",
      "Current time 6697.72 min\n",
      "Batch 9500, Train Loss:0.0224, Train ACC:99.8987\n",
      "Current time 6705.58 min\n",
      "Batch 10000, Train Loss:0.0224, Train ACC:99.8950\n",
      "Current time 6713.43 min\n",
      "Batch 10500, Train Loss:0.0226, Train ACC:99.8940\n",
      "Current time 6721.26 min\n",
      "Batch 11000, Train Loss:0.0224, Train ACC:99.8966\n",
      "Current time 6729.14 min\n",
      "Batch 11500, Train Loss:0.0224, Train ACC:99.8967\n",
      "Current time 6736.98 min\n",
      "train Loss:0.0450  Acc:99.8931%\n",
      "Finish time 6743.72 min\n",
      "Epoch 55/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0201, Train ACC:99.9500\n",
      "Current time 6751.65 min\n",
      "Batch 1000, Train Loss:0.0230, Train ACC:99.8625\n",
      "Current time 6759.56 min\n",
      "Batch 1500, Train Loss:0.0229, Train ACC:99.8833\n",
      "Current time 6767.48 min\n",
      "Batch 2000, Train Loss:0.0219, Train ACC:99.9062\n",
      "Current time 6775.42 min\n",
      "Batch 2500, Train Loss:0.0214, Train ACC:99.9200\n",
      "Current time 6783.38 min\n",
      "Batch 3000, Train Loss:0.0221, Train ACC:99.9042\n",
      "Current time 6791.31 min\n",
      "Batch 3500, Train Loss:0.0219, Train ACC:99.9036\n",
      "Current time 6799.27 min\n",
      "Batch 4000, Train Loss:0.0217, Train ACC:99.9062\n",
      "Current time 6807.20 min\n",
      "Batch 4500, Train Loss:0.0218, Train ACC:99.8972\n",
      "Current time 6815.14 min\n",
      "Batch 5000, Train Loss:0.0220, Train ACC:99.8825\n",
      "Current time 6823.07 min\n",
      "Batch 5500, Train Loss:0.0220, Train ACC:99.8795\n",
      "Current time 6830.98 min\n",
      "Batch 6000, Train Loss:0.0219, Train ACC:99.8833\n",
      "Current time 6838.93 min\n",
      "Batch 6500, Train Loss:0.0218, Train ACC:99.8846\n",
      "Current time 6846.88 min\n",
      "Batch 7000, Train Loss:0.0218, Train ACC:99.8875\n",
      "Current time 6854.82 min\n",
      "Batch 7500, Train Loss:0.0220, Train ACC:99.8783\n",
      "Current time 6862.75 min\n",
      "Batch 8000, Train Loss:0.0219, Train ACC:99.8828\n",
      "Current time 6870.72 min\n",
      "Batch 8500, Train Loss:0.0217, Train ACC:99.8897\n",
      "Current time 6878.68 min\n",
      "Batch 9000, Train Loss:0.0223, Train ACC:99.8764\n",
      "Current time 6886.67 min\n",
      "Batch 9500, Train Loss:0.0224, Train ACC:99.8789\n",
      "Current time 6894.63 min\n",
      "Batch 10000, Train Loss:0.0223, Train ACC:99.8812\n",
      "Current time 6902.54 min\n",
      "Batch 10500, Train Loss:0.0224, Train ACC:99.8821\n",
      "Current time 6910.51 min\n",
      "Batch 11000, Train Loss:0.0225, Train ACC:99.8795\n",
      "Current time 6918.49 min\n",
      "Batch 11500, Train Loss:0.0225, Train ACC:99.8793\n",
      "Current time 6926.62 min\n",
      "train Loss:0.0448  Acc:99.8826%\n",
      "Finish time 6933.65 min\n",
      "Epoch 56/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0246, Train ACC:99.7500\n",
      "Current time 6941.87 min\n",
      "Batch 1000, Train Loss:0.0224, Train ACC:99.8250\n",
      "Current time 6950.07 min\n",
      "Batch 1500, Train Loss:0.0214, Train ACC:99.8667\n",
      "Current time 6958.27 min\n",
      "Batch 2000, Train Loss:0.0232, Train ACC:99.8500\n",
      "Current time 6966.48 min\n",
      "Batch 2500, Train Loss:0.0229, Train ACC:99.8550\n",
      "Current time 6974.66 min\n",
      "Batch 3000, Train Loss:0.0232, Train ACC:99.8542\n",
      "Current time 6982.84 min\n",
      "Batch 3500, Train Loss:0.0228, Train ACC:99.8643\n",
      "Current time 6991.02 min\n",
      "Batch 4000, Train Loss:0.0232, Train ACC:99.8469\n",
      "Current time 6999.23 min\n",
      "Batch 4500, Train Loss:0.0229, Train ACC:99.8528\n",
      "Current time 7007.43 min\n",
      "Batch 5000, Train Loss:0.0228, Train ACC:99.8550\n",
      "Current time 7015.62 min\n",
      "Batch 5500, Train Loss:0.0227, Train ACC:99.8591\n",
      "Current time 7023.82 min\n",
      "Batch 6000, Train Loss:0.0225, Train ACC:99.8646\n",
      "Current time 7032.02 min\n",
      "Batch 6500, Train Loss:0.0228, Train ACC:99.8596\n",
      "Current time 7040.19 min\n",
      "Batch 7000, Train Loss:0.0231, Train ACC:99.8554\n",
      "Current time 7048.38 min\n",
      "Batch 7500, Train Loss:0.0231, Train ACC:99.8583\n",
      "Current time 7056.58 min\n",
      "Batch 8000, Train Loss:0.0228, Train ACC:99.8672\n",
      "Current time 7064.79 min\n",
      "Batch 8500, Train Loss:0.0227, Train ACC:99.8676\n",
      "Current time 7072.98 min\n",
      "Batch 9000, Train Loss:0.0226, Train ACC:99.8722\n",
      "Current time 7081.17 min\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Batch 9500, Train Loss:0.0228, Train ACC:99.8658\n",
      "Current time 7089.38 min\n",
      "Batch 10000, Train Loss:0.0227, Train ACC:99.8688\n",
      "Current time 7097.56 min\n",
      "Batch 10500, Train Loss:0.0226, Train ACC:99.8714\n",
      "Current time 7105.79 min\n",
      "Batch 11000, Train Loss:0.0225, Train ACC:99.8727\n",
      "Current time 7114.00 min\n",
      "Batch 11500, Train Loss:0.0223, Train ACC:99.8750\n",
      "Current time 7122.19 min\n",
      "train Loss:0.0445  Acc:99.8743%\n",
      "Finish time 7129.24 min\n",
      "Epoch 57/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0247, Train ACC:99.8750\n",
      "Current time 7137.40 min\n",
      "Batch 1000, Train Loss:0.0228, Train ACC:99.9125\n",
      "Current time 7145.58 min\n",
      "Batch 1500, Train Loss:0.0220, Train ACC:99.9333\n",
      "Current time 7153.75 min\n",
      "Batch 2000, Train Loss:0.0217, Train ACC:99.9375\n",
      "Current time 7161.94 min\n",
      "Batch 2500, Train Loss:0.0217, Train ACC:99.9250\n",
      "Current time 7170.13 min\n",
      "Batch 3000, Train Loss:0.0219, Train ACC:99.9250\n",
      "Current time 7178.31 min\n",
      "Batch 3500, Train Loss:0.0218, Train ACC:99.9286\n",
      "Current time 7186.51 min\n",
      "Batch 4000, Train Loss:0.0222, Train ACC:99.9125\n",
      "Current time 7194.63 min\n",
      "Batch 4500, Train Loss:0.0219, Train ACC:99.9139\n",
      "Current time 7202.83 min\n",
      "Batch 5000, Train Loss:0.0221, Train ACC:99.9100\n",
      "Current time 7211.00 min\n",
      "Batch 5500, Train Loss:0.0222, Train ACC:99.9023\n",
      "Current time 7219.16 min\n",
      "Batch 6000, Train Loss:0.0222, Train ACC:99.9000\n",
      "Current time 7227.34 min\n",
      "Batch 6500, Train Loss:0.0222, Train ACC:99.9019\n",
      "Current time 7235.52 min\n",
      "Batch 7000, Train Loss:0.0221, Train ACC:99.9036\n",
      "Current time 7243.68 min\n",
      "Batch 7500, Train Loss:0.0222, Train ACC:99.8983\n",
      "Current time 7251.84 min\n",
      "Batch 8000, Train Loss:0.0222, Train ACC:99.8969\n",
      "Current time 7260.05 min\n",
      "Batch 8500, Train Loss:0.0223, Train ACC:99.8985\n",
      "Current time 7268.21 min\n",
      "Batch 9000, Train Loss:0.0221, Train ACC:99.8986\n",
      "Current time 7276.40 min\n",
      "Batch 9500, Train Loss:0.0222, Train ACC:99.8921\n",
      "Current time 7284.57 min\n",
      "Batch 10000, Train Loss:0.0222, Train ACC:99.8912\n",
      "Current time 7292.75 min\n",
      "Batch 10500, Train Loss:0.0223, Train ACC:99.8893\n",
      "Current time 7300.96 min\n",
      "Batch 11000, Train Loss:0.0222, Train ACC:99.8886\n",
      "Current time 7309.09 min\n",
      "Batch 11500, Train Loss:0.0223, Train ACC:99.8848\n",
      "Current time 7317.25 min\n",
      "train Loss:0.0447  Acc:99.8826%\n",
      "Finish time 7324.26 min\n",
      "Epoch 58/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0195, Train ACC:99.9500\n",
      "Current time 7332.20 min\n",
      "Batch 1000, Train Loss:0.0212, Train ACC:99.9125\n",
      "Current time 7340.16 min\n",
      "Batch 1500, Train Loss:0.0222, Train ACC:99.8833\n",
      "Current time 7348.08 min\n",
      "Batch 2000, Train Loss:0.0233, Train ACC:99.8625\n",
      "Current time 7355.99 min\n",
      "Batch 2500, Train Loss:0.0225, Train ACC:99.8900\n",
      "Current time 7363.90 min\n",
      "Batch 3000, Train Loss:0.0226, Train ACC:99.8792\n",
      "Current time 7371.85 min\n",
      "Batch 3500, Train Loss:0.0223, Train ACC:99.8821\n",
      "Current time 7379.79 min\n",
      "Batch 4000, Train Loss:0.0222, Train ACC:99.8844\n",
      "Current time 7387.72 min\n",
      "Batch 4500, Train Loss:0.0222, Train ACC:99.8833\n",
      "Current time 7395.63 min\n",
      "Batch 5000, Train Loss:0.0223, Train ACC:99.8775\n",
      "Current time 7403.55 min\n",
      "Batch 5500, Train Loss:0.0221, Train ACC:99.8795\n",
      "Current time 7411.51 min\n",
      "Batch 6000, Train Loss:0.0218, Train ACC:99.8875\n",
      "Current time 7419.45 min\n",
      "Batch 6500, Train Loss:0.0216, Train ACC:99.8962\n",
      "Current time 7427.37 min\n",
      "Batch 7000, Train Loss:0.0214, Train ACC:99.9000\n",
      "Current time 7435.30 min\n",
      "Batch 7500, Train Loss:0.0215, Train ACC:99.8950\n",
      "Current time 7443.26 min\n",
      "Batch 8000, Train Loss:0.0214, Train ACC:99.8953\n",
      "Current time 7451.21 min\n",
      "Batch 8500, Train Loss:0.0216, Train ACC:99.8868\n",
      "Current time 7459.36 min\n",
      "Batch 9000, Train Loss:0.0218, Train ACC:99.8819\n",
      "Current time 7467.55 min\n",
      "Batch 9500, Train Loss:0.0216, Train ACC:99.8855\n",
      "Current time 7475.75 min\n",
      "Batch 10000, Train Loss:0.0216, Train ACC:99.8863\n",
      "Current time 7483.95 min\n",
      "Batch 10500, Train Loss:0.0218, Train ACC:99.8857\n",
      "Current time 7492.15 min\n",
      "Batch 11000, Train Loss:0.0217, Train ACC:99.8875\n",
      "Current time 7500.32 min\n",
      "Batch 11500, Train Loss:0.0216, Train ACC:99.8891\n",
      "Current time 7508.52 min\n",
      "train Loss:0.0432  Acc:99.8868%\n",
      "Finish time 7515.56 min\n",
      "Epoch 59/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0197, Train ACC:99.9500\n",
      "Current time 7523.65 min\n",
      "Batch 1000, Train Loss:0.0190, Train ACC:99.9500\n",
      "Current time 7531.77 min\n",
      "Batch 1500, Train Loss:0.0187, Train ACC:99.9667\n",
      "Current time 7539.89 min\n",
      "Batch 2000, Train Loss:0.0199, Train ACC:99.9375\n",
      "Current time 7548.02 min\n",
      "Batch 2500, Train Loss:0.0204, Train ACC:99.9300\n",
      "Current time 7556.11 min\n",
      "Batch 3000, Train Loss:0.0209, Train ACC:99.9125\n",
      "Current time 7564.23 min\n",
      "Batch 3500, Train Loss:0.0208, Train ACC:99.9143\n",
      "Current time 7572.35 min\n",
      "Batch 4000, Train Loss:0.0207, Train ACC:99.9125\n",
      "Current time 7580.48 min\n",
      "Batch 4500, Train Loss:0.0211, Train ACC:99.9000\n",
      "Current time 7588.60 min\n",
      "Batch 5000, Train Loss:0.0210, Train ACC:99.9025\n",
      "Current time 7596.70 min\n",
      "Batch 5500, Train Loss:0.0213, Train ACC:99.9000\n",
      "Current time 7604.80 min\n",
      "Batch 6000, Train Loss:0.0216, Train ACC:99.8937\n",
      "Current time 7612.90 min\n",
      "Batch 6500, Train Loss:0.0215, Train ACC:99.8962\n",
      "Current time 7621.00 min\n",
      "Batch 7000, Train Loss:0.0213, Train ACC:99.9018\n",
      "Current time 7629.13 min\n",
      "Batch 7500, Train Loss:0.0214, Train ACC:99.9000\n",
      "Current time 7637.23 min\n",
      "Batch 8000, Train Loss:0.0216, Train ACC:99.8922\n",
      "Current time 7645.33 min\n",
      "Batch 8500, Train Loss:0.0215, Train ACC:99.8956\n",
      "Current time 7653.43 min\n",
      "Batch 9000, Train Loss:0.0213, Train ACC:99.9000\n",
      "Current time 7661.52 min\n",
      "Batch 9500, Train Loss:0.0214, Train ACC:99.9000\n",
      "Current time 7669.64 min\n",
      "Batch 10000, Train Loss:0.0214, Train ACC:99.8975\n",
      "Current time 7677.75 min\n",
      "Batch 10500, Train Loss:0.0215, Train ACC:99.8964\n",
      "Current time 7685.89 min\n",
      "Batch 11000, Train Loss:0.0215, Train ACC:99.8955\n",
      "Current time 7694.02 min\n",
      "Batch 11500, Train Loss:0.0218, Train ACC:99.8902\n",
      "Current time 7702.12 min\n",
      "train Loss:0.0435  Acc:99.8900%\n",
      "Finish time 7709.07 min\n",
      "Epoch 60/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0187, Train ACC:99.9750\n",
      "Current time 7717.01 min\n",
      "Batch 1000, Train Loss:0.0191, Train ACC:99.9625\n",
      "Current time 7724.95 min\n",
      "Batch 1500, Train Loss:0.0194, Train ACC:99.9667\n",
      "Current time 7732.89 min\n",
      "Batch 2000, Train Loss:0.0190, Train ACC:99.9688\n",
      "Current time 7740.81 min\n",
      "Batch 2500, Train Loss:0.0198, Train ACC:99.9450\n",
      "Current time 7748.76 min\n",
      "Batch 3000, Train Loss:0.0198, Train ACC:99.9417\n",
      "Current time 7756.73 min\n",
      "Batch 3500, Train Loss:0.0207, Train ACC:99.9214\n",
      "Current time 7764.66 min\n",
      "Batch 4000, Train Loss:0.0206, Train ACC:99.9250\n",
      "Current time 7772.59 min\n",
      "Batch 4500, Train Loss:0.0212, Train ACC:99.9167\n",
      "Current time 7780.50 min\n",
      "Batch 5000, Train Loss:0.0211, Train ACC:99.9150\n",
      "Current time 7788.47 min\n",
      "Batch 5500, Train Loss:0.0212, Train ACC:99.9068\n",
      "Current time 7796.39 min\n",
      "Batch 6000, Train Loss:0.0212, Train ACC:99.9062\n",
      "Current time 7804.31 min\n",
      "Batch 6500, Train Loss:0.0211, Train ACC:99.9058\n",
      "Current time 7812.26 min\n",
      "Batch 7000, Train Loss:0.0210, Train ACC:99.9107\n",
      "Current time 7820.22 min\n",
      "Batch 7500, Train Loss:0.0209, Train ACC:99.9133\n",
      "Current time 7828.16 min\n",
      "Batch 8000, Train Loss:0.0207, Train ACC:99.9188\n",
      "Current time 7836.11 min\n",
      "Batch 8500, Train Loss:0.0212, Train ACC:99.9118\n",
      "Current time 7844.05 min\n",
      "Batch 9000, Train Loss:0.0212, Train ACC:99.9125\n",
      "Current time 7852.02 min\n",
      "Batch 9500, Train Loss:0.0216, Train ACC:99.9079\n",
      "Current time 7859.97 min\n",
      "Batch 10000, Train Loss:0.0215, Train ACC:99.9112\n",
      "Current time 7867.89 min\n",
      "Batch 10500, Train Loss:0.0214, Train ACC:99.9107\n",
      "Current time 7875.88 min\n",
      "Batch 11000, Train Loss:0.0215, Train ACC:99.9091\n",
      "Current time 7883.80 min\n",
      "Batch 11500, Train Loss:0.0217, Train ACC:99.9043\n",
      "Current time 7891.69 min\n",
      "train Loss:0.0432  Acc:99.9057%\n",
      "Finish time 7898.47 min\n",
      "Epoch 61/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0215, Train ACC:99.9000\n",
      "Current time 7906.40 min\n",
      "Batch 1000, Train Loss:0.0215, Train ACC:99.8875\n",
      "Current time 7914.33 min\n",
      "Batch 1500, Train Loss:0.0208, Train ACC:99.9083\n",
      "Current time 7922.28 min\n",
      "Batch 2000, Train Loss:0.0206, Train ACC:99.9188\n",
      "Current time 7930.20 min\n",
      "Batch 2500, Train Loss:0.0204, Train ACC:99.9250\n",
      "Current time 7938.12 min\n",
      "Batch 3000, Train Loss:0.0206, Train ACC:99.9208\n",
      "Current time 7946.03 min\n",
      "Batch 3500, Train Loss:0.0216, Train ACC:99.8929\n",
      "Current time 7953.96 min\n",
      "Batch 4000, Train Loss:0.0213, Train ACC:99.9000\n",
      "Current time 7961.90 min\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Batch 4500, Train Loss:0.0209, Train ACC:99.9083\n",
      "Current time 7969.83 min\n",
      "Batch 5000, Train Loss:0.0218, Train ACC:99.8925\n",
      "Current time 7977.74 min\n",
      "Batch 5500, Train Loss:0.0216, Train ACC:99.9000\n",
      "Current time 7985.67 min\n",
      "Batch 6000, Train Loss:0.0220, Train ACC:99.8896\n",
      "Current time 7993.61 min\n",
      "Batch 6500, Train Loss:0.0219, Train ACC:99.8962\n",
      "Current time 8001.56 min\n",
      "Batch 7000, Train Loss:0.0220, Train ACC:99.8946\n",
      "Current time 8009.49 min\n",
      "Batch 7500, Train Loss:0.0220, Train ACC:99.8883\n",
      "Current time 8017.44 min\n",
      "Batch 8000, Train Loss:0.0220, Train ACC:99.8875\n",
      "Current time 8025.39 min\n",
      "Batch 8500, Train Loss:0.0219, Train ACC:99.8882\n",
      "Current time 8033.29 min\n",
      "Batch 9000, Train Loss:0.0220, Train ACC:99.8861\n",
      "Current time 8041.20 min\n",
      "Batch 9500, Train Loss:0.0220, Train ACC:99.8842\n",
      "Current time 8049.13 min\n",
      "Batch 10000, Train Loss:0.0220, Train ACC:99.8838\n",
      "Current time 8057.06 min\n",
      "Batch 10500, Train Loss:0.0220, Train ACC:99.8821\n",
      "Current time 8064.98 min\n",
      "Batch 11000, Train Loss:0.0219, Train ACC:99.8841\n",
      "Current time 8072.89 min\n",
      "Batch 11500, Train Loss:0.0220, Train ACC:99.8848\n",
      "Current time 8080.82 min\n",
      "train Loss:0.0439  Acc:99.8847%\n",
      "Finish time 8087.60 min\n",
      "Epoch 62/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0229, Train ACC:99.8750\n",
      "Current time 8095.56 min\n",
      "Batch 1000, Train Loss:0.0219, Train ACC:99.8875\n",
      "Current time 8103.47 min\n",
      "Batch 1500, Train Loss:0.0216, Train ACC:99.9083\n",
      "Current time 8111.39 min\n",
      "Batch 2000, Train Loss:0.0221, Train ACC:99.8875\n",
      "Current time 8119.35 min\n",
      "Batch 2500, Train Loss:0.0217, Train ACC:99.8950\n",
      "Current time 8127.28 min\n",
      "Batch 3000, Train Loss:0.0213, Train ACC:99.9042\n",
      "Current time 8135.20 min\n",
      "Batch 3500, Train Loss:0.0216, Train ACC:99.8929\n",
      "Current time 8143.14 min\n",
      "Batch 4000, Train Loss:0.0211, Train ACC:99.9062\n",
      "Current time 8151.08 min\n",
      "Batch 4500, Train Loss:0.0208, Train ACC:99.9167\n",
      "Current time 8159.05 min\n",
      "Batch 5000, Train Loss:0.0206, Train ACC:99.9150\n",
      "Current time 8166.99 min\n",
      "Batch 5500, Train Loss:0.0213, Train ACC:99.9045\n",
      "Current time 8174.93 min\n",
      "Batch 6000, Train Loss:0.0210, Train ACC:99.9104\n",
      "Current time 8182.87 min\n",
      "Batch 6500, Train Loss:0.0210, Train ACC:99.9038\n",
      "Current time 8190.79 min\n",
      "Batch 7000, Train Loss:0.0212, Train ACC:99.8964\n",
      "Current time 8198.72 min\n",
      "Batch 7500, Train Loss:0.0213, Train ACC:99.8933\n",
      "Current time 8206.63 min\n",
      "Batch 8000, Train Loss:0.0212, Train ACC:99.8969\n",
      "Current time 8214.57 min\n",
      "Batch 8500, Train Loss:0.0213, Train ACC:99.8956\n",
      "Current time 8222.50 min\n",
      "Batch 9000, Train Loss:0.0214, Train ACC:99.8944\n",
      "Current time 8230.49 min\n",
      "Batch 9500, Train Loss:0.0213, Train ACC:99.8974\n",
      "Current time 8238.47 min\n",
      "Batch 10000, Train Loss:0.0212, Train ACC:99.9013\n",
      "Current time 8246.38 min\n",
      "Batch 10500, Train Loss:0.0211, Train ACC:99.9036\n",
      "Current time 8254.36 min\n",
      "Batch 11000, Train Loss:0.0210, Train ACC:99.9057\n",
      "Current time 8262.32 min\n",
      "Batch 11500, Train Loss:0.0211, Train ACC:99.9033\n",
      "Current time 8270.27 min\n",
      "train Loss:0.0422  Acc:99.9057%\n",
      "Finish time 8277.09 min\n",
      "Epoch 63/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0254, Train ACC:99.8750\n",
      "Current time 8285.03 min\n",
      "Batch 1000, Train Loss:0.0237, Train ACC:99.8500\n",
      "Current time 8292.96 min\n",
      "Batch 1500, Train Loss:0.0233, Train ACC:99.8750\n",
      "Current time 8300.93 min\n",
      "Batch 2000, Train Loss:0.0230, Train ACC:99.8688\n",
      "Current time 8309.12 min\n",
      "Batch 2500, Train Loss:0.0222, Train ACC:99.8950\n",
      "Current time 8317.32 min\n",
      "Batch 3000, Train Loss:0.0221, Train ACC:99.8958\n",
      "Current time 8325.50 min\n",
      "Batch 3500, Train Loss:0.0216, Train ACC:99.9107\n",
      "Current time 8333.70 min\n",
      "Batch 4000, Train Loss:0.0216, Train ACC:99.9062\n",
      "Current time 8341.93 min\n",
      "Batch 4500, Train Loss:0.0221, Train ACC:99.8889\n",
      "Current time 8350.13 min\n",
      "Batch 5000, Train Loss:0.0223, Train ACC:99.8875\n",
      "Current time 8358.30 min\n",
      "Batch 5500, Train Loss:0.0222, Train ACC:99.8909\n",
      "Current time 8366.48 min\n",
      "Batch 6000, Train Loss:0.0221, Train ACC:99.8875\n",
      "Current time 8374.68 min\n",
      "Batch 6500, Train Loss:0.0219, Train ACC:99.8865\n",
      "Current time 8382.87 min\n",
      "Batch 7000, Train Loss:0.0218, Train ACC:99.8875\n",
      "Current time 8391.07 min\n",
      "Batch 7500, Train Loss:0.0220, Train ACC:99.8817\n",
      "Current time 8399.28 min\n",
      "Batch 8000, Train Loss:0.0219, Train ACC:99.8812\n",
      "Current time 8407.47 min\n",
      "Batch 8500, Train Loss:0.0222, Train ACC:99.8765\n",
      "Current time 8415.68 min\n",
      "Batch 9000, Train Loss:0.0221, Train ACC:99.8778\n",
      "Current time 8423.90 min\n",
      "Batch 9500, Train Loss:0.0220, Train ACC:99.8816\n",
      "Current time 8432.10 min\n",
      "Batch 10000, Train Loss:0.0219, Train ACC:99.8863\n",
      "Current time 8440.28 min\n",
      "Batch 10500, Train Loss:0.0219, Train ACC:99.8821\n",
      "Current time 8448.49 min\n",
      "Batch 11000, Train Loss:0.0218, Train ACC:99.8852\n",
      "Current time 8456.70 min\n",
      "Batch 11500, Train Loss:0.0218, Train ACC:99.8859\n",
      "Current time 8464.89 min\n",
      "train Loss:0.0435  Acc:99.8868%\n",
      "Finish time 8471.98 min\n",
      "Epoch 64/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0185, Train ACC:99.9750\n",
      "Current time 8479.97 min\n",
      "Batch 1000, Train Loss:0.0181, Train ACC:99.9875\n",
      "Current time 8487.95 min\n",
      "Batch 1500, Train Loss:0.0184, Train ACC:99.9667\n",
      "Current time 8495.93 min\n",
      "Batch 2000, Train Loss:0.0194, Train ACC:99.9437\n",
      "Current time 8503.90 min\n",
      "Batch 2500, Train Loss:0.0196, Train ACC:99.9250\n",
      "Current time 8511.87 min\n",
      "Batch 3000, Train Loss:0.0197, Train ACC:99.9083\n",
      "Current time 8519.84 min\n",
      "Batch 3500, Train Loss:0.0210, Train ACC:99.8821\n",
      "Current time 8527.82 min\n",
      "Batch 4000, Train Loss:0.0209, Train ACC:99.8844\n",
      "Current time 8535.80 min\n",
      "Batch 4500, Train Loss:0.0206, Train ACC:99.8944\n",
      "Current time 8543.76 min\n",
      "Batch 5000, Train Loss:0.0205, Train ACC:99.8950\n",
      "Current time 8551.84 min\n",
      "Batch 5500, Train Loss:0.0205, Train ACC:99.8955\n",
      "Current time 8560.04 min\n",
      "Batch 6000, Train Loss:0.0206, Train ACC:99.8937\n",
      "Current time 8568.21 min\n",
      "Batch 6500, Train Loss:0.0206, Train ACC:99.8942\n",
      "Current time 8576.42 min\n",
      "Batch 7000, Train Loss:0.0207, Train ACC:99.8946\n",
      "Current time 8584.67 min\n",
      "Batch 7500, Train Loss:0.0207, Train ACC:99.8883\n",
      "Current time 8592.89 min\n",
      "Batch 8000, Train Loss:0.0208, Train ACC:99.8875\n",
      "Current time 8601.08 min\n",
      "Batch 8500, Train Loss:0.0207, Train ACC:99.8897\n",
      "Current time 8609.26 min\n",
      "Batch 9000, Train Loss:0.0212, Train ACC:99.8792\n",
      "Current time 8617.46 min\n",
      "Batch 9500, Train Loss:0.0210, Train ACC:99.8842\n",
      "Current time 8625.67 min\n",
      "Batch 10000, Train Loss:0.0212, Train ACC:99.8838\n",
      "Current time 8633.85 min\n",
      "Batch 10500, Train Loss:0.0211, Train ACC:99.8821\n",
      "Current time 8642.07 min\n",
      "Batch 11000, Train Loss:0.0212, Train ACC:99.8795\n",
      "Current time 8650.26 min\n",
      "Batch 11500, Train Loss:0.0213, Train ACC:99.8783\n",
      "Current time 8658.49 min\n",
      "train Loss:0.0426  Acc:99.8784%\n",
      "Finish time 8665.53 min\n",
      "Epoch 65/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0217, Train ACC:99.8500\n",
      "Current time 8673.58 min\n",
      "Batch 1000, Train Loss:0.0204, Train ACC:99.9000\n",
      "Current time 8681.63 min\n",
      "Batch 1500, Train Loss:0.0211, Train ACC:99.9083\n",
      "Current time 8689.70 min\n",
      "Batch 2000, Train Loss:0.0205, Train ACC:99.9250\n",
      "Current time 8697.75 min\n",
      "Batch 2500, Train Loss:0.0211, Train ACC:99.9150\n",
      "Current time 8705.79 min\n",
      "Batch 3000, Train Loss:0.0211, Train ACC:99.9000\n",
      "Current time 8713.81 min\n",
      "Batch 3500, Train Loss:0.0209, Train ACC:99.9036\n",
      "Current time 8721.86 min\n",
      "Batch 4000, Train Loss:0.0206, Train ACC:99.9094\n",
      "Current time 8729.91 min\n",
      "Batch 4500, Train Loss:0.0203, Train ACC:99.9167\n",
      "Current time 8737.92 min\n",
      "Batch 5000, Train Loss:0.0206, Train ACC:99.9075\n",
      "Current time 8745.93 min\n",
      "Batch 5500, Train Loss:0.0210, Train ACC:99.8932\n",
      "Current time 8753.99 min\n",
      "Batch 6000, Train Loss:0.0212, Train ACC:99.8917\n",
      "Current time 8762.03 min\n",
      "Batch 6500, Train Loss:0.0210, Train ACC:99.8962\n",
      "Current time 8770.06 min\n",
      "Batch 7000, Train Loss:0.0209, Train ACC:99.8982\n",
      "Current time 8778.07 min\n",
      "Batch 7500, Train Loss:0.0208, Train ACC:99.9017\n",
      "Current time 8786.12 min\n",
      "Batch 8000, Train Loss:0.0206, Train ACC:99.9016\n",
      "Current time 8794.15 min\n",
      "Batch 8500, Train Loss:0.0208, Train ACC:99.8985\n",
      "Current time 8802.19 min\n",
      "Batch 9000, Train Loss:0.0209, Train ACC:99.8958\n",
      "Current time 8810.21 min\n",
      "Batch 9500, Train Loss:0.0210, Train ACC:99.8961\n",
      "Current time 8818.23 min\n",
      "Batch 10000, Train Loss:0.0210, Train ACC:99.9000\n",
      "Current time 8826.27 min\n",
      "Batch 10500, Train Loss:0.0208, Train ACC:99.9036\n",
      "Current time 8834.32 min\n",
      "Batch 11000, Train Loss:0.0210, Train ACC:99.9000\n",
      "Current time 8842.37 min\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Batch 11500, Train Loss:0.0209, Train ACC:99.9000\n",
      "Current time 8850.41 min\n",
      "train Loss:0.0420  Acc:99.8984%\n",
      "Finish time 8857.32 min\n",
      "Epoch 66/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0186, Train ACC:99.9000\n",
      "Current time 8865.24 min\n",
      "Batch 1000, Train Loss:0.0191, Train ACC:99.9375\n",
      "Current time 8873.14 min\n",
      "Batch 1500, Train Loss:0.0191, Train ACC:99.9417\n",
      "Current time 8881.06 min\n",
      "Batch 2000, Train Loss:0.0187, Train ACC:99.9562\n",
      "Current time 8889.02 min\n",
      "Batch 2500, Train Loss:0.0186, Train ACC:99.9500\n",
      "Current time 8897.08 min\n",
      "Batch 3000, Train Loss:0.0194, Train ACC:99.9417\n",
      "Current time 8905.24 min\n",
      "Batch 3500, Train Loss:0.0199, Train ACC:99.9357\n",
      "Current time 8913.42 min\n",
      "Batch 4000, Train Loss:0.0201, Train ACC:99.9281\n",
      "Current time 8921.63 min\n",
      "Batch 4500, Train Loss:0.0201, Train ACC:99.9306\n",
      "Current time 8929.79 min\n",
      "Batch 5000, Train Loss:0.0201, Train ACC:99.9275\n",
      "Current time 8938.01 min\n",
      "Batch 5500, Train Loss:0.0203, Train ACC:99.9182\n",
      "Current time 8946.21 min\n",
      "Batch 6000, Train Loss:0.0202, Train ACC:99.9208\n",
      "Current time 8954.41 min\n",
      "Batch 6500, Train Loss:0.0204, Train ACC:99.9154\n",
      "Current time 8962.58 min\n",
      "Batch 7000, Train Loss:0.0203, Train ACC:99.9161\n",
      "Current time 8970.75 min\n",
      "Batch 7500, Train Loss:0.0206, Train ACC:99.9117\n",
      "Current time 8978.95 min\n",
      "Batch 8000, Train Loss:0.0208, Train ACC:99.9078\n",
      "Current time 8987.13 min\n",
      "Batch 8500, Train Loss:0.0208, Train ACC:99.9088\n",
      "Current time 8995.33 min\n",
      "Batch 9000, Train Loss:0.0206, Train ACC:99.9139\n",
      "Current time 9003.51 min\n",
      "Batch 9500, Train Loss:0.0205, Train ACC:99.9171\n",
      "Current time 9011.69 min\n",
      "Batch 10000, Train Loss:0.0205, Train ACC:99.9150\n",
      "Current time 9019.89 min\n",
      "Batch 10500, Train Loss:0.0205, Train ACC:99.9143\n",
      "Current time 9028.09 min\n",
      "Batch 11000, Train Loss:0.0205, Train ACC:99.9136\n",
      "Current time 9036.28 min\n",
      "Batch 11500, Train Loss:0.0205, Train ACC:99.9141\n",
      "Current time 9044.46 min\n",
      "train Loss:0.0412  Acc:99.9120%\n",
      "Finish time 9051.51 min\n",
      "Epoch 67/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0214, Train ACC:99.8750\n",
      "Current time 9059.50 min\n",
      "Batch 1000, Train Loss:0.0244, Train ACC:99.8625\n",
      "Current time 9067.46 min\n",
      "Batch 1500, Train Loss:0.0242, Train ACC:99.8167\n",
      "Current time 9075.43 min\n",
      "Batch 2000, Train Loss:0.0229, Train ACC:99.8500\n",
      "Current time 9083.41 min\n",
      "Batch 2500, Train Loss:0.0229, Train ACC:99.8500\n",
      "Current time 9091.36 min\n",
      "Batch 3000, Train Loss:0.0222, Train ACC:99.8667\n",
      "Current time 9099.35 min\n",
      "Batch 3500, Train Loss:0.0230, Train ACC:99.8607\n",
      "Current time 9107.33 min\n",
      "Batch 4000, Train Loss:0.0230, Train ACC:99.8594\n",
      "Current time 9115.32 min\n",
      "Batch 4500, Train Loss:0.0224, Train ACC:99.8750\n",
      "Current time 9123.28 min\n",
      "Batch 5000, Train Loss:0.0222, Train ACC:99.8775\n",
      "Current time 9131.21 min\n",
      "Batch 5500, Train Loss:0.0218, Train ACC:99.8841\n",
      "Current time 9139.19 min\n",
      "Batch 6000, Train Loss:0.0215, Train ACC:99.8896\n",
      "Current time 9147.18 min\n",
      "Batch 6500, Train Loss:0.0213, Train ACC:99.8942\n",
      "Current time 9155.16 min\n",
      "Batch 7000, Train Loss:0.0212, Train ACC:99.8946\n",
      "Current time 9163.13 min\n",
      "Batch 7500, Train Loss:0.0210, Train ACC:99.9000\n",
      "Current time 9171.11 min\n",
      "Batch 8000, Train Loss:0.0208, Train ACC:99.9047\n",
      "Current time 9179.08 min\n",
      "Batch 8500, Train Loss:0.0208, Train ACC:99.9000\n",
      "Current time 9187.03 min\n",
      "Batch 9000, Train Loss:0.0206, Train ACC:99.9042\n",
      "Current time 9194.99 min\n",
      "Batch 9500, Train Loss:0.0206, Train ACC:99.9013\n",
      "Current time 9202.95 min\n",
      "Batch 10000, Train Loss:0.0206, Train ACC:99.9000\n",
      "Current time 9210.93 min\n",
      "Batch 10500, Train Loss:0.0206, Train ACC:99.8976\n",
      "Current time 9218.89 min\n",
      "Batch 11000, Train Loss:0.0207, Train ACC:99.8932\n",
      "Current time 9226.84 min\n",
      "Batch 11500, Train Loss:0.0209, Train ACC:99.8913\n",
      "Current time 9234.81 min\n",
      "train Loss:0.0418  Acc:99.8921%\n",
      "Finish time 9241.64 min\n",
      "Epoch 68/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0211, Train ACC:99.8500\n",
      "Current time 9249.58 min\n",
      "Batch 1000, Train Loss:0.0193, Train ACC:99.9250\n",
      "Current time 9257.51 min\n",
      "Batch 1500, Train Loss:0.0194, Train ACC:99.9250\n",
      "Current time 9265.45 min\n",
      "Batch 2000, Train Loss:0.0197, Train ACC:99.9125\n",
      "Current time 9273.39 min\n",
      "Batch 2500, Train Loss:0.0200, Train ACC:99.9100\n",
      "Current time 9281.32 min\n",
      "Batch 3000, Train Loss:0.0203, Train ACC:99.9083\n",
      "Current time 9289.27 min\n",
      "Batch 3500, Train Loss:0.0205, Train ACC:99.9000\n",
      "Current time 9297.18 min\n",
      "Batch 4000, Train Loss:0.0214, Train ACC:99.8812\n",
      "Current time 9305.08 min\n",
      "Batch 4500, Train Loss:0.0213, Train ACC:99.8944\n",
      "Current time 9312.99 min\n",
      "Batch 5000, Train Loss:0.0212, Train ACC:99.8925\n",
      "Current time 9320.91 min\n",
      "Batch 5500, Train Loss:0.0213, Train ACC:99.8886\n",
      "Current time 9328.84 min\n",
      "Batch 6000, Train Loss:0.0213, Train ACC:99.8896\n",
      "Current time 9336.76 min\n",
      "Batch 6500, Train Loss:0.0211, Train ACC:99.8923\n",
      "Current time 9344.66 min\n",
      "Batch 7000, Train Loss:0.0210, Train ACC:99.8946\n",
      "Current time 9352.62 min\n",
      "Batch 7500, Train Loss:0.0208, Train ACC:99.8983\n",
      "Current time 9360.53 min\n",
      "Batch 8000, Train Loss:0.0210, Train ACC:99.8937\n",
      "Current time 9368.47 min\n",
      "Batch 8500, Train Loss:0.0208, Train ACC:99.8985\n",
      "Current time 9376.39 min\n",
      "Batch 9000, Train Loss:0.0207, Train ACC:99.9042\n",
      "Current time 9384.32 min\n",
      "Batch 9500, Train Loss:0.0205, Train ACC:99.9079\n",
      "Current time 9392.25 min\n",
      "Batch 10000, Train Loss:0.0206, Train ACC:99.9062\n",
      "Current time 9400.16 min\n",
      "Batch 10500, Train Loss:0.0207, Train ACC:99.9024\n",
      "Current time 9408.10 min\n",
      "Batch 11000, Train Loss:0.0205, Train ACC:99.9068\n",
      "Current time 9416.02 min\n",
      "Batch 11500, Train Loss:0.0209, Train ACC:99.8978\n",
      "Current time 9423.95 min\n",
      "train Loss:0.0416  Acc:99.9005%\n",
      "Finish time 9430.71 min\n",
      "Epoch 69/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0210, Train ACC:99.9000\n",
      "Current time 9438.64 min\n",
      "Batch 1000, Train Loss:0.0193, Train ACC:99.9375\n",
      "Current time 9446.56 min\n",
      "Batch 1500, Train Loss:0.0192, Train ACC:99.9333\n",
      "Current time 9454.50 min\n",
      "Batch 2000, Train Loss:0.0189, Train ACC:99.9375\n",
      "Current time 9462.43 min\n",
      "Batch 2500, Train Loss:0.0190, Train ACC:99.9350\n",
      "Current time 9470.34 min\n",
      "Batch 3000, Train Loss:0.0190, Train ACC:99.9375\n",
      "Current time 9478.24 min\n",
      "Batch 3500, Train Loss:0.0199, Train ACC:99.9143\n",
      "Current time 9486.14 min\n",
      "Batch 4000, Train Loss:0.0204, Train ACC:99.9000\n",
      "Current time 9494.04 min\n",
      "Batch 4500, Train Loss:0.0204, Train ACC:99.9056\n",
      "Current time 9502.00 min\n",
      "Batch 5000, Train Loss:0.0203, Train ACC:99.9075\n",
      "Current time 9509.93 min\n",
      "Batch 5500, Train Loss:0.0208, Train ACC:99.8977\n",
      "Current time 9517.83 min\n",
      "Batch 6000, Train Loss:0.0207, Train ACC:99.9042\n",
      "Current time 9525.73 min\n",
      "Batch 6500, Train Loss:0.0216, Train ACC:99.8885\n",
      "Current time 9533.61 min\n",
      "Batch 7000, Train Loss:0.0217, Train ACC:99.8821\n",
      "Current time 9541.52 min\n",
      "Batch 7500, Train Loss:0.0217, Train ACC:99.8800\n",
      "Current time 9549.43 min\n",
      "Batch 8000, Train Loss:0.0215, Train ACC:99.8859\n",
      "Current time 9557.35 min\n",
      "Batch 8500, Train Loss:0.0214, Train ACC:99.8882\n",
      "Current time 9565.27 min\n",
      "Batch 9000, Train Loss:0.0212, Train ACC:99.8944\n",
      "Current time 9573.15 min\n",
      "Batch 9500, Train Loss:0.0209, Train ACC:99.9000\n",
      "Current time 9581.08 min\n",
      "Batch 10000, Train Loss:0.0210, Train ACC:99.8988\n",
      "Current time 9588.97 min\n",
      "Batch 10500, Train Loss:0.0210, Train ACC:99.8988\n",
      "Current time 9596.89 min\n",
      "Batch 11000, Train Loss:0.0209, Train ACC:99.9011\n",
      "Current time 9604.78 min\n",
      "Batch 11500, Train Loss:0.0209, Train ACC:99.9000\n",
      "Current time 9612.70 min\n",
      "train Loss:0.0417  Acc:99.8994%\n",
      "Finish time 9619.50 min\n",
      "Epoch 70/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0208, Train ACC:99.9000\n",
      "Current time 9627.41 min\n",
      "Batch 1000, Train Loss:0.0215, Train ACC:99.9000\n",
      "Current time 9635.30 min\n",
      "Batch 1500, Train Loss:0.0233, Train ACC:99.8583\n",
      "Current time 9643.21 min\n",
      "Batch 2000, Train Loss:0.0220, Train ACC:99.8812\n",
      "Current time 9651.37 min\n",
      "Batch 2500, Train Loss:0.0216, Train ACC:99.8950\n",
      "Current time 9659.56 min\n",
      "Batch 3000, Train Loss:0.0211, Train ACC:99.9042\n",
      "Current time 9667.74 min\n",
      "Batch 3500, Train Loss:0.0212, Train ACC:99.9000\n",
      "Current time 9675.89 min\n",
      "Batch 4000, Train Loss:0.0215, Train ACC:99.8844\n",
      "Current time 9684.08 min\n",
      "Batch 4500, Train Loss:0.0214, Train ACC:99.8889\n",
      "Current time 9692.29 min\n",
      "Batch 5000, Train Loss:0.0211, Train ACC:99.9000\n",
      "Current time 9700.46 min\n",
      "Batch 5500, Train Loss:0.0211, Train ACC:99.9000\n",
      "Current time 9708.61 min\n",
      "Batch 6000, Train Loss:0.0212, Train ACC:99.8979\n",
      "Current time 9716.79 min\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Batch 6500, Train Loss:0.0213, Train ACC:99.8904\n",
      "Current time 9724.97 min\n",
      "Batch 7000, Train Loss:0.0212, Train ACC:99.8946\n",
      "Current time 9733.15 min\n",
      "Batch 7500, Train Loss:0.0212, Train ACC:99.8883\n",
      "Current time 9741.34 min\n",
      "Batch 8000, Train Loss:0.0212, Train ACC:99.8875\n",
      "Current time 9749.53 min\n",
      "Batch 8500, Train Loss:0.0212, Train ACC:99.8897\n",
      "Current time 9757.71 min\n",
      "Batch 9000, Train Loss:0.0212, Train ACC:99.8889\n",
      "Current time 9765.88 min\n",
      "Batch 9500, Train Loss:0.0211, Train ACC:99.8921\n",
      "Current time 9774.10 min\n",
      "Batch 10000, Train Loss:0.0210, Train ACC:99.8925\n",
      "Current time 9782.28 min\n",
      "Batch 10500, Train Loss:0.0210, Train ACC:99.8940\n",
      "Current time 9790.47 min\n",
      "Batch 11000, Train Loss:0.0209, Train ACC:99.8955\n",
      "Current time 9798.69 min\n",
      "Batch 11500, Train Loss:0.0208, Train ACC:99.8989\n",
      "Current time 9806.86 min\n",
      "train Loss:0.0416  Acc:99.9015%\n",
      "Finish time 9813.86 min\n",
      "Epoch 71/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0174, Train ACC:99.9500\n",
      "Current time 9821.82 min\n",
      "Batch 1000, Train Loss:0.0176, Train ACC:99.9750\n",
      "Current time 9829.80 min\n",
      "Batch 1500, Train Loss:0.0178, Train ACC:99.9750\n",
      "Current time 9837.76 min\n",
      "Batch 2000, Train Loss:0.0183, Train ACC:99.9562\n",
      "Current time 9845.71 min\n",
      "Batch 2500, Train Loss:0.0191, Train ACC:99.9250\n",
      "Current time 9853.68 min\n",
      "Batch 3000, Train Loss:0.0189, Train ACC:99.9333\n",
      "Current time 9861.64 min\n",
      "Batch 3500, Train Loss:0.0189, Train ACC:99.9393\n",
      "Current time 9869.61 min\n",
      "Batch 4000, Train Loss:0.0188, Train ACC:99.9406\n",
      "Current time 9877.55 min\n",
      "Batch 4500, Train Loss:0.0188, Train ACC:99.9361\n",
      "Current time 9885.52 min\n",
      "Batch 5000, Train Loss:0.0189, Train ACC:99.9325\n",
      "Current time 9893.49 min\n",
      "Batch 5500, Train Loss:0.0193, Train ACC:99.9250\n",
      "Current time 9901.46 min\n",
      "Batch 6000, Train Loss:0.0192, Train ACC:99.9292\n",
      "Current time 9909.37 min\n",
      "Batch 6500, Train Loss:0.0192, Train ACC:99.9308\n",
      "Current time 9917.35 min\n",
      "Batch 7000, Train Loss:0.0192, Train ACC:99.9286\n",
      "Current time 9925.31 min\n",
      "Batch 7500, Train Loss:0.0193, Train ACC:99.9217\n",
      "Current time 9933.26 min\n",
      "Batch 8000, Train Loss:0.0193, Train ACC:99.9219\n",
      "Current time 9941.24 min\n",
      "Batch 8500, Train Loss:0.0196, Train ACC:99.9191\n",
      "Current time 9949.18 min\n",
      "Batch 9000, Train Loss:0.0198, Train ACC:99.9167\n",
      "Current time 9957.14 min\n",
      "Batch 9500, Train Loss:0.0197, Train ACC:99.9197\n",
      "Current time 9965.07 min\n",
      "Batch 10000, Train Loss:0.0197, Train ACC:99.9225\n",
      "Current time 9973.01 min\n",
      "Batch 10500, Train Loss:0.0198, Train ACC:99.9190\n",
      "Current time 9980.96 min\n",
      "Batch 11000, Train Loss:0.0200, Train ACC:99.9159\n",
      "Current time 9989.07 min\n",
      "Batch 11500, Train Loss:0.0200, Train ACC:99.9163\n",
      "Current time 9997.24 min\n",
      "train Loss:0.0402  Acc:99.9130%\n",
      "Finish time 10004.26 min\n",
      "Epoch 72/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0241, Train ACC:99.8000\n",
      "Current time 10012.40 min\n",
      "Batch 1000, Train Loss:0.0226, Train ACC:99.8250\n",
      "Current time 10020.54 min\n",
      "Batch 1500, Train Loss:0.0210, Train ACC:99.8833\n",
      "Current time 10028.67 min\n",
      "Batch 2000, Train Loss:0.0210, Train ACC:99.8750\n",
      "Current time 10036.82 min\n",
      "Batch 2500, Train Loss:0.0210, Train ACC:99.8750\n",
      "Current time 10044.95 min\n",
      "Batch 3000, Train Loss:0.0207, Train ACC:99.8833\n",
      "Current time 10053.10 min\n",
      "Batch 3500, Train Loss:0.0203, Train ACC:99.9000\n",
      "Current time 10061.25 min\n",
      "Batch 4000, Train Loss:0.0199, Train ACC:99.9062\n",
      "Current time 10069.37 min\n",
      "Batch 4500, Train Loss:0.0198, Train ACC:99.9056\n",
      "Current time 10077.51 min\n",
      "Batch 5000, Train Loss:0.0198, Train ACC:99.9100\n",
      "Current time 10085.65 min\n",
      "Batch 5500, Train Loss:0.0199, Train ACC:99.9023\n",
      "Current time 10093.77 min\n",
      "Batch 6000, Train Loss:0.0202, Train ACC:99.8937\n",
      "Current time 10101.91 min\n",
      "Batch 6500, Train Loss:0.0203, Train ACC:99.8885\n",
      "Current time 10110.07 min\n",
      "Batch 7000, Train Loss:0.0202, Train ACC:99.8946\n",
      "Current time 10118.21 min\n",
      "Batch 7500, Train Loss:0.0202, Train ACC:99.8967\n",
      "Current time 10126.36 min\n",
      "Batch 8000, Train Loss:0.0204, Train ACC:99.8953\n",
      "Current time 10134.50 min\n",
      "Batch 8500, Train Loss:0.0204, Train ACC:99.8985\n",
      "Current time 10142.63 min\n",
      "Batch 9000, Train Loss:0.0204, Train ACC:99.8958\n",
      "Current time 10150.78 min\n",
      "Batch 9500, Train Loss:0.0205, Train ACC:99.8947\n",
      "Current time 10158.93 min\n",
      "Batch 10000, Train Loss:0.0204, Train ACC:99.8975\n",
      "Current time 10167.06 min\n",
      "Batch 10500, Train Loss:0.0203, Train ACC:99.9012\n",
      "Current time 10175.19 min\n",
      "Batch 11000, Train Loss:0.0203, Train ACC:99.9034\n",
      "Current time 10183.33 min\n",
      "Batch 11500, Train Loss:0.0205, Train ACC:99.9011\n",
      "Current time 10191.48 min\n",
      "train Loss:0.0408  Acc:99.9036%\n",
      "Finish time 10198.47 min\n",
      "Epoch 73/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0196, Train ACC:99.9500\n",
      "Current time 10206.37 min\n",
      "Batch 1000, Train Loss:0.0193, Train ACC:99.9375\n",
      "Current time 10214.28 min\n",
      "Batch 1500, Train Loss:0.0194, Train ACC:99.9417\n",
      "Current time 10222.21 min\n",
      "Batch 2000, Train Loss:0.0198, Train ACC:99.9437\n",
      "Current time 10230.15 min\n",
      "Batch 2500, Train Loss:0.0196, Train ACC:99.9450\n",
      "Current time 10238.06 min\n",
      "Batch 3000, Train Loss:0.0209, Train ACC:99.9167\n",
      "Current time 10245.97 min\n",
      "Batch 3500, Train Loss:0.0211, Train ACC:99.9107\n",
      "Current time 10253.88 min\n",
      "Batch 4000, Train Loss:0.0213, Train ACC:99.9031\n",
      "Current time 10261.79 min\n",
      "Batch 4500, Train Loss:0.0209, Train ACC:99.9111\n",
      "Current time 10269.71 min\n",
      "Batch 5000, Train Loss:0.0209, Train ACC:99.9125\n",
      "Current time 10277.61 min\n",
      "Batch 5500, Train Loss:0.0206, Train ACC:99.9159\n",
      "Current time 10285.52 min\n",
      "Batch 6000, Train Loss:0.0207, Train ACC:99.9146\n",
      "Current time 10293.43 min\n",
      "Batch 6500, Train Loss:0.0209, Train ACC:99.9135\n",
      "Current time 10301.33 min\n",
      "Batch 7000, Train Loss:0.0210, Train ACC:99.9054\n",
      "Current time 10309.23 min\n",
      "Batch 7500, Train Loss:0.0210, Train ACC:99.9050\n",
      "Current time 10317.13 min\n",
      "Batch 8000, Train Loss:0.0209, Train ACC:99.9078\n",
      "Current time 10325.05 min\n",
      "Batch 8500, Train Loss:0.0210, Train ACC:99.9059\n",
      "Current time 10332.96 min\n",
      "Batch 9000, Train Loss:0.0209, Train ACC:99.9069\n",
      "Current time 10341.11 min\n",
      "Batch 9500, Train Loss:0.0208, Train ACC:99.9105\n",
      "Current time 10349.28 min\n",
      "Batch 10000, Train Loss:0.0208, Train ACC:99.9112\n",
      "Current time 10357.43 min\n",
      "Batch 10500, Train Loss:0.0208, Train ACC:99.9107\n",
      "Current time 10365.61 min\n",
      "Batch 11000, Train Loss:0.0209, Train ACC:99.9068\n",
      "Current time 10373.79 min\n",
      "Batch 11500, Train Loss:0.0210, Train ACC:99.9043\n",
      "Current time 10381.97 min\n",
      "train Loss:0.0418  Acc:99.9067%\n",
      "Finish time 10388.94 min\n",
      "Epoch 74/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0194, Train ACC:99.9500\n",
      "Current time 10397.02 min\n",
      "Batch 1000, Train Loss:0.0200, Train ACC:99.9125\n",
      "Current time 10405.11 min\n",
      "Batch 1500, Train Loss:0.0213, Train ACC:99.8750\n",
      "Current time 10413.18 min\n",
      "Batch 2000, Train Loss:0.0209, Train ACC:99.9000\n",
      "Current time 10421.25 min\n",
      "Batch 2500, Train Loss:0.0201, Train ACC:99.9150\n",
      "Current time 10429.35 min\n",
      "Batch 3000, Train Loss:0.0197, Train ACC:99.9292\n",
      "Current time 10437.43 min\n",
      "Batch 3500, Train Loss:0.0201, Train ACC:99.9107\n",
      "Current time 10445.50 min\n",
      "Batch 4000, Train Loss:0.0199, Train ACC:99.9156\n",
      "Current time 10453.60 min\n",
      "Batch 4500, Train Loss:0.0200, Train ACC:99.9139\n",
      "Current time 10461.69 min\n",
      "Batch 5000, Train Loss:0.0199, Train ACC:99.9150\n",
      "Current time 10469.79 min\n",
      "Batch 5500, Train Loss:0.0198, Train ACC:99.9136\n",
      "Current time 10477.88 min\n",
      "Batch 6000, Train Loss:0.0202, Train ACC:99.9042\n",
      "Current time 10485.96 min\n",
      "Batch 6500, Train Loss:0.0203, Train ACC:99.8981\n",
      "Current time 10494.04 min\n",
      "Batch 7000, Train Loss:0.0202, Train ACC:99.9036\n",
      "Current time 10502.11 min\n",
      "Batch 7500, Train Loss:0.0203, Train ACC:99.9033\n",
      "Current time 10510.19 min\n",
      "Batch 8000, Train Loss:0.0203, Train ACC:99.9062\n",
      "Current time 10518.29 min\n",
      "Batch 8500, Train Loss:0.0203, Train ACC:99.9088\n",
      "Current time 10526.36 min\n",
      "Batch 9000, Train Loss:0.0202, Train ACC:99.9097\n",
      "Current time 10534.44 min\n",
      "Batch 9500, Train Loss:0.0202, Train ACC:99.9092\n",
      "Current time 10542.52 min\n",
      "Batch 10000, Train Loss:0.0202, Train ACC:99.9087\n",
      "Current time 10550.59 min\n",
      "Batch 10500, Train Loss:0.0202, Train ACC:99.9071\n",
      "Current time 10558.69 min\n",
      "Batch 11000, Train Loss:0.0202, Train ACC:99.9068\n",
      "Current time 10566.77 min\n",
      "Batch 11500, Train Loss:0.0202, Train ACC:99.9065\n",
      "Current time 10574.88 min\n",
      "train Loss:0.0404  Acc:99.9067%\n",
      "Finish time 10581.80 min\n",
      "Epoch 75/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0182, Train ACC:99.9750\n",
      "Current time 10589.71 min\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Batch 1000, Train Loss:0.0184, Train ACC:99.9625\n",
      "Current time 10597.61 min\n",
      "Batch 1500, Train Loss:0.0191, Train ACC:99.9250\n",
      "Current time 10605.49 min\n",
      "Batch 2000, Train Loss:0.0190, Train ACC:99.9375\n",
      "Current time 10613.39 min\n",
      "Batch 2500, Train Loss:0.0188, Train ACC:99.9400\n",
      "Current time 10621.31 min\n",
      "Batch 3000, Train Loss:0.0191, Train ACC:99.9375\n",
      "Current time 10629.25 min\n",
      "Batch 3500, Train Loss:0.0193, Train ACC:99.9250\n",
      "Current time 10637.17 min\n",
      "Batch 4000, Train Loss:0.0201, Train ACC:99.9062\n",
      "Current time 10645.09 min\n",
      "Batch 4500, Train Loss:0.0200, Train ACC:99.9111\n",
      "Current time 10652.97 min\n",
      "Batch 5000, Train Loss:0.0199, Train ACC:99.9175\n",
      "Current time 10660.86 min\n",
      "Batch 5500, Train Loss:0.0198, Train ACC:99.9227\n",
      "Current time 10668.75 min\n",
      "Batch 6000, Train Loss:0.0198, Train ACC:99.9188\n",
      "Current time 10676.67 min\n",
      "Batch 6500, Train Loss:0.0197, Train ACC:99.9212\n",
      "Current time 10684.60 min\n",
      "Batch 7000, Train Loss:0.0199, Train ACC:99.9214\n",
      "Current time 10692.53 min\n",
      "Batch 7500, Train Loss:0.0198, Train ACC:99.9233\n",
      "Current time 10700.42 min\n",
      "Batch 8000, Train Loss:0.0198, Train ACC:99.9250\n",
      "Current time 10708.32 min\n",
      "Batch 8500, Train Loss:0.0202, Train ACC:99.9191\n",
      "Current time 10716.22 min\n",
      "Batch 9000, Train Loss:0.0201, Train ACC:99.9208\n",
      "Current time 10724.13 min\n",
      "Batch 9500, Train Loss:0.0201, Train ACC:99.9184\n",
      "Current time 10731.99 min\n",
      "Batch 10000, Train Loss:0.0201, Train ACC:99.9188\n",
      "Current time 10739.89 min\n",
      "Batch 10500, Train Loss:0.0201, Train ACC:99.9167\n",
      "Current time 10747.79 min\n",
      "Batch 11000, Train Loss:0.0201, Train ACC:99.9170\n",
      "Current time 10755.69 min\n",
      "Batch 11500, Train Loss:0.0201, Train ACC:99.9174\n",
      "Current time 10763.60 min\n",
      "train Loss:0.0401  Acc:99.9183%\n",
      "Finish time 10770.37 min\n",
      "Epoch 76/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0199, Train ACC:99.9000\n",
      "Current time 10778.27 min\n",
      "Batch 1000, Train Loss:0.0190, Train ACC:99.9500\n",
      "Current time 10786.14 min\n",
      "Batch 1500, Train Loss:0.0183, Train ACC:99.9667\n",
      "Current time 10794.02 min\n",
      "Batch 2000, Train Loss:0.0186, Train ACC:99.9562\n",
      "Current time 10801.92 min\n",
      "Batch 2500, Train Loss:0.0193, Train ACC:99.9400\n",
      "Current time 10809.82 min\n",
      "Batch 3000, Train Loss:0.0202, Train ACC:99.9125\n",
      "Current time 10817.73 min\n",
      "Batch 3500, Train Loss:0.0201, Train ACC:99.9143\n",
      "Current time 10825.65 min\n",
      "Batch 4000, Train Loss:0.0201, Train ACC:99.9188\n",
      "Current time 10833.54 min\n",
      "Batch 4500, Train Loss:0.0205, Train ACC:99.9083\n",
      "Current time 10841.43 min\n",
      "Batch 5000, Train Loss:0.0203, Train ACC:99.9150\n",
      "Current time 10849.30 min\n",
      "Batch 5500, Train Loss:0.0201, Train ACC:99.9182\n",
      "Current time 10857.21 min\n",
      "Batch 6000, Train Loss:0.0201, Train ACC:99.9167\n",
      "Current time 10865.10 min\n",
      "Batch 6500, Train Loss:0.0202, Train ACC:99.9135\n",
      "Current time 10873.00 min\n",
      "Batch 7000, Train Loss:0.0201, Train ACC:99.9143\n",
      "Current time 10880.90 min\n",
      "Batch 7500, Train Loss:0.0200, Train ACC:99.9183\n",
      "Current time 10888.81 min\n",
      "Batch 8000, Train Loss:0.0198, Train ACC:99.9234\n",
      "Current time 10896.69 min\n",
      "Batch 8500, Train Loss:0.0196, Train ACC:99.9279\n",
      "Current time 10904.57 min\n",
      "Batch 9000, Train Loss:0.0198, Train ACC:99.9278\n",
      "Current time 10912.44 min\n",
      "Batch 9500, Train Loss:0.0200, Train ACC:99.9184\n",
      "Current time 10920.31 min\n",
      "Batch 10000, Train Loss:0.0201, Train ACC:99.9175\n",
      "Current time 10928.22 min\n",
      "Batch 10500, Train Loss:0.0200, Train ACC:99.9202\n",
      "Current time 10936.12 min\n",
      "Batch 11000, Train Loss:0.0200, Train ACC:99.9182\n",
      "Current time 10944.00 min\n",
      "Batch 11500, Train Loss:0.0201, Train ACC:99.9163\n",
      "Current time 10951.93 min\n",
      "train Loss:0.0402  Acc:99.9151%\n",
      "Finish time 10958.70 min\n",
      "Epoch 77/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0176, Train ACC:99.9750\n",
      "Current time 10966.62 min\n",
      "Batch 1000, Train Loss:0.0179, Train ACC:99.9500\n",
      "Current time 10974.51 min\n",
      "Batch 1500, Train Loss:0.0183, Train ACC:99.9417\n",
      "Current time 10982.41 min\n",
      "Batch 2000, Train Loss:0.0199, Train ACC:99.9000\n",
      "Current time 10990.35 min\n",
      "Batch 2500, Train Loss:0.0216, Train ACC:99.8800\n",
      "Current time 10998.48 min\n",
      "Batch 3000, Train Loss:0.0213, Train ACC:99.8833\n",
      "Current time 11006.69 min\n",
      "Batch 3500, Train Loss:0.0211, Train ACC:99.8893\n",
      "Current time 11014.88 min\n",
      "Batch 4000, Train Loss:0.0207, Train ACC:99.9000\n",
      "Current time 11023.06 min\n",
      "Batch 4500, Train Loss:0.0204, Train ACC:99.9083\n",
      "Current time 11031.23 min\n",
      "Batch 5000, Train Loss:0.0201, Train ACC:99.9125\n",
      "Current time 11039.41 min\n",
      "Batch 5500, Train Loss:0.0203, Train ACC:99.9023\n",
      "Current time 11047.59 min\n",
      "Batch 6000, Train Loss:0.0206, Train ACC:99.8979\n",
      "Current time 11055.77 min\n",
      "Batch 6500, Train Loss:0.0204, Train ACC:99.9000\n",
      "Current time 11063.96 min\n",
      "Batch 7000, Train Loss:0.0202, Train ACC:99.9071\n",
      "Current time 11072.13 min\n",
      "Batch 7500, Train Loss:0.0203, Train ACC:99.9083\n",
      "Current time 11080.29 min\n",
      "Batch 8000, Train Loss:0.0203, Train ACC:99.9047\n",
      "Current time 11088.45 min\n",
      "Batch 8500, Train Loss:0.0202, Train ACC:99.9044\n",
      "Current time 11096.61 min\n",
      "Batch 9000, Train Loss:0.0201, Train ACC:99.9097\n",
      "Current time 11104.78 min\n",
      "Batch 9500, Train Loss:0.0201, Train ACC:99.9092\n",
      "Current time 11112.97 min\n",
      "Batch 10000, Train Loss:0.0206, Train ACC:99.9013\n",
      "Current time 11121.16 min\n",
      "Batch 10500, Train Loss:0.0207, Train ACC:99.9012\n",
      "Current time 11129.32 min\n",
      "Batch 11000, Train Loss:0.0206, Train ACC:99.8989\n",
      "Current time 11137.52 min\n",
      "Batch 11500, Train Loss:0.0205, Train ACC:99.9011\n",
      "Current time 11145.72 min\n",
      "train Loss:0.0413  Acc:99.8984%\n",
      "Finish time 11152.75 min\n",
      "Epoch 78/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0198, Train ACC:99.8750\n",
      "Current time 11160.75 min\n",
      "Batch 1000, Train Loss:0.0189, Train ACC:99.9125\n",
      "Current time 11168.75 min\n",
      "Batch 1500, Train Loss:0.0189, Train ACC:99.9333\n",
      "Current time 11176.76 min\n",
      "Batch 2000, Train Loss:0.0186, Train ACC:99.9500\n",
      "Current time 11184.74 min\n",
      "Batch 2500, Train Loss:0.0183, Train ACC:99.9550\n",
      "Current time 11192.74 min\n",
      "Batch 3000, Train Loss:0.0182, Train ACC:99.9625\n",
      "Current time 11200.73 min\n",
      "Batch 3500, Train Loss:0.0186, Train ACC:99.9607\n",
      "Current time 11208.67 min\n",
      "Batch 4000, Train Loss:0.0189, Train ACC:99.9500\n",
      "Current time 11216.66 min\n",
      "Batch 4500, Train Loss:0.0193, Train ACC:99.9389\n",
      "Current time 11224.64 min\n",
      "Batch 5000, Train Loss:0.0196, Train ACC:99.9325\n",
      "Current time 11232.61 min\n",
      "Batch 5500, Train Loss:0.0197, Train ACC:99.9341\n",
      "Current time 11240.59 min\n",
      "Batch 6000, Train Loss:0.0196, Train ACC:99.9354\n",
      "Current time 11248.60 min\n",
      "Batch 6500, Train Loss:0.0196, Train ACC:99.9327\n",
      "Current time 11256.58 min\n",
      "Batch 7000, Train Loss:0.0195, Train ACC:99.9339\n",
      "Current time 11264.52 min\n",
      "Batch 7500, Train Loss:0.0193, Train ACC:99.9367\n",
      "Current time 11272.52 min\n",
      "Batch 8000, Train Loss:0.0194, Train ACC:99.9359\n",
      "Current time 11280.49 min\n",
      "Batch 8500, Train Loss:0.0194, Train ACC:99.9368\n",
      "Current time 11288.46 min\n",
      "Batch 9000, Train Loss:0.0196, Train ACC:99.9292\n",
      "Current time 11296.44 min\n",
      "Batch 9500, Train Loss:0.0195, Train ACC:99.9316\n",
      "Current time 11304.38 min\n",
      "Batch 10000, Train Loss:0.0195, Train ACC:99.9313\n",
      "Current time 11312.33 min\n",
      "Batch 10500, Train Loss:0.0195, Train ACC:99.9321\n",
      "Current time 11320.30 min\n",
      "Batch 11000, Train Loss:0.0196, Train ACC:99.9273\n",
      "Current time 11328.28 min\n",
      "Batch 11500, Train Loss:0.0197, Train ACC:99.9239\n",
      "Current time 11336.26 min\n",
      "train Loss:0.0394  Acc:99.9235%\n",
      "Finish time 11343.10 min\n",
      "Epoch 79/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0176, Train ACC:99.9750\n",
      "Current time 11351.02 min\n",
      "Batch 1000, Train Loss:0.0184, Train ACC:99.9500\n",
      "Current time 11358.96 min\n",
      "Batch 1500, Train Loss:0.0194, Train ACC:99.9417\n",
      "Current time 11366.87 min\n",
      "Batch 2000, Train Loss:0.0194, Train ACC:99.9250\n",
      "Current time 11374.76 min\n",
      "Batch 2500, Train Loss:0.0190, Train ACC:99.9400\n",
      "Current time 11382.73 min\n",
      "Batch 3000, Train Loss:0.0189, Train ACC:99.9375\n",
      "Current time 11390.67 min\n",
      "Batch 3500, Train Loss:0.0195, Train ACC:99.9286\n",
      "Current time 11398.56 min\n",
      "Batch 4000, Train Loss:0.0202, Train ACC:99.9188\n",
      "Current time 11406.50 min\n",
      "Batch 4500, Train Loss:0.0201, Train ACC:99.9194\n",
      "Current time 11414.40 min\n",
      "Batch 5000, Train Loss:0.0202, Train ACC:99.9175\n",
      "Current time 11422.36 min\n",
      "Batch 5500, Train Loss:0.0201, Train ACC:99.9114\n",
      "Current time 11430.51 min\n",
      "Batch 6000, Train Loss:0.0202, Train ACC:99.9042\n",
      "Current time 11438.67 min\n",
      "Batch 6500, Train Loss:0.0203, Train ACC:99.9038\n",
      "Current time 11446.85 min\n",
      "Batch 7000, Train Loss:0.0202, Train ACC:99.9089\n",
      "Current time 11455.04 min\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Batch 7500, Train Loss:0.0201, Train ACC:99.9100\n",
      "Current time 11463.21 min\n",
      "Batch 8000, Train Loss:0.0200, Train ACC:99.9141\n",
      "Current time 11471.40 min\n",
      "Batch 8500, Train Loss:0.0199, Train ACC:99.9176\n",
      "Current time 11479.59 min\n",
      "Batch 9000, Train Loss:0.0200, Train ACC:99.9139\n",
      "Current time 11487.79 min\n",
      "Batch 9500, Train Loss:0.0200, Train ACC:99.9118\n",
      "Current time 11495.98 min\n",
      "Batch 10000, Train Loss:0.0199, Train ACC:99.9150\n",
      "Current time 11504.16 min\n",
      "Batch 10500, Train Loss:0.0199, Train ACC:99.9155\n",
      "Current time 11512.36 min\n",
      "Batch 11000, Train Loss:0.0198, Train ACC:99.9159\n",
      "Current time 11520.53 min\n",
      "Batch 11500, Train Loss:0.0198, Train ACC:99.9163\n",
      "Current time 11528.70 min\n",
      "train Loss:0.0399  Acc:99.9130%\n",
      "Finish time 11535.72 min\n",
      "Epoch 80/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0195, Train ACC:99.9500\n",
      "Current time 11543.77 min\n",
      "Batch 1000, Train Loss:0.0186, Train ACC:99.9750\n",
      "Current time 11551.85 min\n",
      "Batch 1500, Train Loss:0.0181, Train ACC:99.9833\n",
      "Current time 11559.89 min\n",
      "Batch 2000, Train Loss:0.0189, Train ACC:99.9500\n",
      "Current time 11567.92 min\n",
      "Batch 2500, Train Loss:0.0186, Train ACC:99.9600\n",
      "Current time 11575.96 min\n",
      "Batch 3000, Train Loss:0.0195, Train ACC:99.9417\n",
      "Current time 11583.98 min\n",
      "Batch 3500, Train Loss:0.0195, Train ACC:99.9393\n",
      "Current time 11592.04 min\n",
      "Batch 4000, Train Loss:0.0196, Train ACC:99.9344\n",
      "Current time 11600.07 min\n",
      "Batch 4500, Train Loss:0.0199, Train ACC:99.9167\n",
      "Current time 11608.08 min\n",
      "Batch 5000, Train Loss:0.0198, Train ACC:99.9150\n",
      "Current time 11616.09 min\n",
      "Batch 5500, Train Loss:0.0196, Train ACC:99.9205\n",
      "Current time 11624.14 min\n",
      "Batch 6000, Train Loss:0.0195, Train ACC:99.9229\n",
      "Current time 11632.18 min\n",
      "Batch 6500, Train Loss:0.0194, Train ACC:99.9250\n",
      "Current time 11640.19 min\n",
      "Batch 7000, Train Loss:0.0195, Train ACC:99.9250\n",
      "Current time 11648.21 min\n",
      "Batch 7500, Train Loss:0.0193, Train ACC:99.9300\n",
      "Current time 11656.26 min\n",
      "Batch 8000, Train Loss:0.0191, Train ACC:99.9344\n",
      "Current time 11664.30 min\n",
      "Batch 8500, Train Loss:0.0193, Train ACC:99.9338\n",
      "Current time 11672.32 min\n",
      "Batch 9000, Train Loss:0.0195, Train ACC:99.9278\n",
      "Current time 11680.33 min\n",
      "Batch 9500, Train Loss:0.0195, Train ACC:99.9237\n",
      "Current time 11688.36 min\n",
      "Batch 10000, Train Loss:0.0197, Train ACC:99.9200\n",
      "Current time 11696.39 min\n",
      "Batch 10500, Train Loss:0.0198, Train ACC:99.9179\n",
      "Current time 11704.44 min\n",
      "Batch 11000, Train Loss:0.0197, Train ACC:99.9182\n",
      "Current time 11712.47 min\n",
      "Batch 11500, Train Loss:0.0196, Train ACC:99.9207\n",
      "Current time 11720.52 min\n",
      "train Loss:0.0392  Acc:99.9225%\n",
      "Finish time 11727.42 min\n",
      "Epoch 81/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0169, Train ACC:99.9750\n",
      "Current time 11735.36 min\n",
      "Batch 1000, Train Loss:0.0169, Train ACC:99.9875\n",
      "Current time 11743.23 min\n",
      "Batch 1500, Train Loss:0.0196, Train ACC:99.8917\n",
      "Current time 11751.14 min\n",
      "Batch 2000, Train Loss:0.0194, Train ACC:99.9000\n",
      "Current time 11758.99 min\n",
      "Batch 2500, Train Loss:0.0204, Train ACC:99.8950\n",
      "Current time 11766.85 min\n",
      "Batch 3000, Train Loss:0.0205, Train ACC:99.8958\n",
      "Current time 11774.78 min\n",
      "Batch 3500, Train Loss:0.0207, Train ACC:99.8964\n",
      "Current time 11782.91 min\n",
      "Batch 4000, Train Loss:0.0210, Train ACC:99.8812\n",
      "Current time 11791.04 min\n",
      "Batch 4500, Train Loss:0.0208, Train ACC:99.8861\n",
      "Current time 11799.14 min\n",
      "Batch 5000, Train Loss:0.0206, Train ACC:99.8925\n",
      "Current time 11807.25 min\n",
      "Batch 5500, Train Loss:0.0203, Train ACC:99.9000\n",
      "Current time 11815.32 min\n",
      "Batch 6000, Train Loss:0.0201, Train ACC:99.9062\n",
      "Current time 11823.44 min\n",
      "Batch 6500, Train Loss:0.0200, Train ACC:99.9077\n",
      "Current time 11831.54 min\n",
      "Batch 7000, Train Loss:0.0201, Train ACC:99.9054\n",
      "Current time 11839.66 min\n",
      "Batch 7500, Train Loss:0.0205, Train ACC:99.9017\n",
      "Current time 11847.73 min\n",
      "Batch 8000, Train Loss:0.0204, Train ACC:99.9016\n",
      "Current time 11855.81 min\n",
      "Batch 8500, Train Loss:0.0204, Train ACC:99.9059\n",
      "Current time 11863.92 min\n",
      "Batch 9000, Train Loss:0.0203, Train ACC:99.9083\n",
      "Current time 11872.00 min\n",
      "Batch 9500, Train Loss:0.0204, Train ACC:99.9092\n",
      "Current time 11880.11 min\n",
      "Batch 10000, Train Loss:0.0203, Train ACC:99.9112\n",
      "Current time 11888.22 min\n",
      "Batch 10500, Train Loss:0.0204, Train ACC:99.9095\n",
      "Current time 11896.37 min\n",
      "Batch 11000, Train Loss:0.0203, Train ACC:99.9091\n",
      "Current time 11904.48 min\n",
      "Batch 11500, Train Loss:0.0203, Train ACC:99.9087\n",
      "Current time 11912.63 min\n",
      "train Loss:0.0405  Acc:99.9109%\n",
      "Finish time 11919.63 min\n",
      "Epoch 82/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0171, Train ACC:100.0000\n",
      "Current time 11927.56 min\n",
      "Batch 1000, Train Loss:0.0174, Train ACC:99.9875\n",
      "Current time 11935.48 min\n",
      "Batch 1500, Train Loss:0.0185, Train ACC:99.9667\n",
      "Current time 11943.44 min\n",
      "Batch 2000, Train Loss:0.0182, Train ACC:99.9688\n",
      "Current time 11951.39 min\n",
      "Batch 2500, Train Loss:0.0181, Train ACC:99.9650\n",
      "Current time 11959.31 min\n",
      "Batch 3000, Train Loss:0.0188, Train ACC:99.9500\n",
      "Current time 11967.27 min\n",
      "Batch 3500, Train Loss:0.0188, Train ACC:99.9536\n",
      "Current time 11975.20 min\n",
      "Batch 4000, Train Loss:0.0186, Train ACC:99.9594\n",
      "Current time 11983.16 min\n",
      "Batch 4500, Train Loss:0.0184, Train ACC:99.9639\n",
      "Current time 11991.11 min\n",
      "Batch 5000, Train Loss:0.0192, Train ACC:99.9425\n",
      "Current time 11999.06 min\n",
      "Batch 5500, Train Loss:0.0192, Train ACC:99.9432\n",
      "Current time 12007.00 min\n",
      "Batch 6000, Train Loss:0.0198, Train ACC:99.9250\n",
      "Current time 12014.95 min\n",
      "Batch 6500, Train Loss:0.0197, Train ACC:99.9250\n",
      "Current time 12022.94 min\n",
      "Batch 7000, Train Loss:0.0197, Train ACC:99.9232\n",
      "Current time 12030.89 min\n",
      "Batch 7500, Train Loss:0.0196, Train ACC:99.9267\n",
      "Current time 12038.85 min\n",
      "Batch 8000, Train Loss:0.0195, Train ACC:99.9281\n",
      "Current time 12046.83 min\n",
      "Batch 8500, Train Loss:0.0195, Train ACC:99.9265\n",
      "Current time 12054.76 min\n",
      "Batch 9000, Train Loss:0.0196, Train ACC:99.9292\n",
      "Current time 12062.70 min\n",
      "Batch 9500, Train Loss:0.0195, Train ACC:99.9303\n",
      "Current time 12070.63 min\n",
      "Batch 10000, Train Loss:0.0195, Train ACC:99.9300\n",
      "Current time 12078.58 min\n",
      "Batch 10500, Train Loss:0.0194, Train ACC:99.9321\n",
      "Current time 12086.53 min\n",
      "Batch 11000, Train Loss:0.0194, Train ACC:99.9341\n",
      "Current time 12094.48 min\n",
      "Batch 11500, Train Loss:0.0196, Train ACC:99.9315\n",
      "Current time 12102.45 min\n",
      "train Loss:0.0391  Acc:99.9319%\n",
      "Finish time 12109.27 min\n",
      "Epoch 83/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0214, Train ACC:99.8250\n",
      "Current time 12117.21 min\n",
      "Batch 1000, Train Loss:0.0198, Train ACC:99.8750\n",
      "Current time 12125.11 min\n",
      "Batch 1500, Train Loss:0.0188, Train ACC:99.9167\n",
      "Current time 12133.00 min\n",
      "Batch 2000, Train Loss:0.0184, Train ACC:99.9375\n",
      "Current time 12140.92 min\n",
      "Batch 2500, Train Loss:0.0183, Train ACC:99.9450\n",
      "Current time 12148.83 min\n",
      "Batch 3000, Train Loss:0.0197, Train ACC:99.9208\n",
      "Current time 12156.74 min\n",
      "Batch 3500, Train Loss:0.0194, Train ACC:99.9321\n",
      "Current time 12164.65 min\n",
      "Batch 4000, Train Loss:0.0196, Train ACC:99.9250\n",
      "Current time 12172.55 min\n",
      "Batch 4500, Train Loss:0.0198, Train ACC:99.9194\n",
      "Current time 12180.43 min\n",
      "Batch 5000, Train Loss:0.0195, Train ACC:99.9275\n",
      "Current time 12188.33 min\n",
      "Batch 5500, Train Loss:0.0194, Train ACC:99.9295\n",
      "Current time 12196.22 min\n",
      "Batch 6000, Train Loss:0.0193, Train ACC:99.9313\n",
      "Current time 12204.12 min\n",
      "Batch 6500, Train Loss:0.0194, Train ACC:99.9327\n",
      "Current time 12212.01 min\n",
      "Batch 7000, Train Loss:0.0194, Train ACC:99.9339\n",
      "Current time 12219.94 min\n",
      "Batch 7500, Train Loss:0.0198, Train ACC:99.9250\n",
      "Current time 12227.84 min\n",
      "Batch 8000, Train Loss:0.0196, Train ACC:99.9297\n",
      "Current time 12235.75 min\n",
      "Batch 8500, Train Loss:0.0199, Train ACC:99.9265\n",
      "Current time 12243.64 min\n",
      "Batch 9000, Train Loss:0.0200, Train ACC:99.9222\n",
      "Current time 12251.56 min\n",
      "Batch 9500, Train Loss:0.0200, Train ACC:99.9197\n",
      "Current time 12259.47 min\n",
      "Batch 10000, Train Loss:0.0199, Train ACC:99.9213\n",
      "Current time 12267.36 min\n",
      "Batch 10500, Train Loss:0.0201, Train ACC:99.9155\n",
      "Current time 12275.30 min\n",
      "Batch 11000, Train Loss:0.0200, Train ACC:99.9193\n",
      "Current time 12283.20 min\n",
      "Batch 11500, Train Loss:0.0200, Train ACC:99.9207\n",
      "Current time 12291.10 min\n",
      "train Loss:0.0398  Acc:99.9225%\n",
      "Finish time 12297.86 min\n",
      "Epoch 84/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0202, Train ACC:99.8750\n",
      "Current time 12305.75 min\n",
      "Batch 1000, Train Loss:0.0186, Train ACC:99.9375\n",
      "Current time 12313.65 min\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Batch 1500, Train Loss:0.0191, Train ACC:99.9167\n",
      "Current time 12321.59 min\n",
      "Batch 2000, Train Loss:0.0201, Train ACC:99.9125\n",
      "Current time 12329.74 min\n",
      "Batch 2500, Train Loss:0.0198, Train ACC:99.9150\n",
      "Current time 12337.91 min\n",
      "Batch 3000, Train Loss:0.0193, Train ACC:99.9292\n",
      "Current time 12346.10 min\n",
      "Batch 3500, Train Loss:0.0196, Train ACC:99.9214\n",
      "Current time 12354.26 min\n",
      "Batch 4000, Train Loss:0.0203, Train ACC:99.8969\n",
      "Current time 12362.43 min\n",
      "Batch 4500, Train Loss:0.0203, Train ACC:99.8889\n",
      "Current time 12370.60 min\n",
      "Batch 5000, Train Loss:0.0200, Train ACC:99.8975\n",
      "Current time 12378.79 min\n",
      "Batch 5500, Train Loss:0.0198, Train ACC:99.9023\n",
      "Current time 12386.94 min\n",
      "Batch 6000, Train Loss:0.0200, Train ACC:99.9021\n",
      "Current time 12395.11 min\n",
      "Batch 6500, Train Loss:0.0204, Train ACC:99.8904\n",
      "Current time 12403.27 min\n",
      "Batch 7000, Train Loss:0.0205, Train ACC:99.8893\n",
      "Current time 12411.47 min\n",
      "Batch 7500, Train Loss:0.0207, Train ACC:99.8867\n",
      "Current time 12419.66 min\n",
      "Batch 8000, Train Loss:0.0205, Train ACC:99.8906\n",
      "Current time 12427.80 min\n",
      "Batch 8500, Train Loss:0.0204, Train ACC:99.8941\n",
      "Current time 12435.99 min\n",
      "Batch 9000, Train Loss:0.0202, Train ACC:99.8972\n",
      "Current time 12444.16 min\n",
      "Batch 9500, Train Loss:0.0201, Train ACC:99.9013\n",
      "Current time 12452.33 min\n",
      "Batch 10000, Train Loss:0.0200, Train ACC:99.9013\n",
      "Current time 12460.48 min\n",
      "Batch 10500, Train Loss:0.0200, Train ACC:99.9012\n",
      "Current time 12468.65 min\n",
      "Batch 11000, Train Loss:0.0201, Train ACC:99.9023\n",
      "Current time 12476.84 min\n",
      "Batch 11500, Train Loss:0.0202, Train ACC:99.9000\n",
      "Current time 12484.99 min\n",
      "train Loss:0.0405  Acc:99.8994%\n",
      "Finish time 12491.97 min\n",
      "Epoch 85/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0168, Train ACC:100.0000\n",
      "Current time 12499.89 min\n",
      "Batch 1000, Train Loss:0.0169, Train ACC:100.0000\n",
      "Current time 12507.80 min\n",
      "Batch 1500, Train Loss:0.0193, Train ACC:99.9500\n",
      "Current time 12515.74 min\n",
      "Batch 2000, Train Loss:0.0187, Train ACC:99.9625\n",
      "Current time 12523.66 min\n",
      "Batch 2500, Train Loss:0.0191, Train ACC:99.9600\n",
      "Current time 12531.59 min\n",
      "Batch 3000, Train Loss:0.0188, Train ACC:99.9667\n",
      "Current time 12539.54 min\n",
      "Batch 3500, Train Loss:0.0186, Train ACC:99.9643\n",
      "Current time 12547.46 min\n",
      "Batch 4000, Train Loss:0.0187, Train ACC:99.9531\n",
      "Current time 12555.40 min\n",
      "Batch 4500, Train Loss:0.0187, Train ACC:99.9528\n",
      "Current time 12563.38 min\n",
      "Batch 5000, Train Loss:0.0188, Train ACC:99.9475\n",
      "Current time 12571.35 min\n",
      "Batch 5500, Train Loss:0.0192, Train ACC:99.9409\n",
      "Current time 12579.31 min\n",
      "Batch 6000, Train Loss:0.0191, Train ACC:99.9417\n",
      "Current time 12587.27 min\n",
      "Batch 6500, Train Loss:0.0194, Train ACC:99.9365\n",
      "Current time 12595.20 min\n",
      "Batch 7000, Train Loss:0.0196, Train ACC:99.9250\n",
      "Current time 12603.16 min\n",
      "Batch 7500, Train Loss:0.0196, Train ACC:99.9233\n",
      "Current time 12611.11 min\n",
      "Batch 8000, Train Loss:0.0195, Train ACC:99.9250\n",
      "Current time 12619.02 min\n",
      "Batch 8500, Train Loss:0.0195, Train ACC:99.9235\n",
      "Current time 12626.91 min\n",
      "Batch 9000, Train Loss:0.0196, Train ACC:99.9236\n",
      "Current time 12634.83 min\n",
      "Batch 9500, Train Loss:0.0196, Train ACC:99.9237\n",
      "Current time 12642.73 min\n",
      "Batch 10000, Train Loss:0.0195, Train ACC:99.9262\n",
      "Current time 12650.61 min\n",
      "Batch 10500, Train Loss:0.0194, Train ACC:99.9286\n",
      "Current time 12658.48 min\n",
      "Batch 11000, Train Loss:0.0193, Train ACC:99.9307\n",
      "Current time 12666.37 min\n",
      "Batch 11500, Train Loss:0.0196, Train ACC:99.9250\n",
      "Current time 12674.26 min\n",
      "train Loss:0.0391  Acc:99.9235%\n",
      "Finish time 12680.99 min\n",
      "Epoch 86/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0201, Train ACC:99.9500\n",
      "Current time 12688.85 min\n",
      "Batch 1000, Train Loss:0.0187, Train ACC:99.9750\n",
      "Current time 12696.71 min\n",
      "Batch 1500, Train Loss:0.0181, Train ACC:99.9833\n",
      "Current time 12704.57 min\n",
      "Batch 2000, Train Loss:0.0177, Train ACC:99.9875\n",
      "Current time 12712.43 min\n",
      "Batch 2500, Train Loss:0.0185, Train ACC:99.9500\n",
      "Current time 12720.29 min\n",
      "Batch 3000, Train Loss:0.0193, Train ACC:99.9208\n",
      "Current time 12728.12 min\n",
      "Batch 3500, Train Loss:0.0200, Train ACC:99.9107\n",
      "Current time 12736.00 min\n",
      "Batch 4000, Train Loss:0.0196, Train ACC:99.9219\n",
      "Current time 12743.85 min\n",
      "Batch 4500, Train Loss:0.0194, Train ACC:99.9306\n",
      "Current time 12751.70 min\n",
      "Batch 5000, Train Loss:0.0196, Train ACC:99.9300\n",
      "Current time 12759.56 min\n",
      "Batch 5500, Train Loss:0.0194, Train ACC:99.9318\n",
      "Current time 12767.41 min\n",
      "Batch 6000, Train Loss:0.0194, Train ACC:99.9313\n",
      "Current time 12775.25 min\n",
      "Batch 6500, Train Loss:0.0198, Train ACC:99.9231\n",
      "Current time 12783.11 min\n",
      "Batch 7000, Train Loss:0.0199, Train ACC:99.9179\n",
      "Current time 12790.97 min\n",
      "Batch 7500, Train Loss:0.0199, Train ACC:99.9200\n",
      "Current time 12798.82 min\n",
      "Batch 8000, Train Loss:0.0198, Train ACC:99.9234\n",
      "Current time 12806.67 min\n",
      "Batch 8500, Train Loss:0.0197, Train ACC:99.9221\n",
      "Current time 12814.49 min\n",
      "Batch 9000, Train Loss:0.0196, Train ACC:99.9250\n",
      "Current time 12822.35 min\n",
      "Batch 9500, Train Loss:0.0196, Train ACC:99.9250\n",
      "Current time 12830.17 min\n",
      "Batch 10000, Train Loss:0.0196, Train ACC:99.9250\n",
      "Current time 12838.04 min\n",
      "Batch 10500, Train Loss:0.0197, Train ACC:99.9262\n",
      "Current time 12845.88 min\n",
      "Batch 11000, Train Loss:0.0195, Train ACC:99.9295\n",
      "Current time 12853.73 min\n",
      "Batch 11500, Train Loss:0.0196, Train ACC:99.9283\n",
      "Current time 12861.59 min\n",
      "train Loss:0.0390  Acc:99.9308%\n",
      "Finish time 12868.47 min\n",
      "Epoch 87/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0191, Train ACC:99.9250\n",
      "Current time 12876.62 min\n",
      "Batch 1000, Train Loss:0.0181, Train ACC:99.9500\n",
      "Current time 12884.72 min\n",
      "Batch 1500, Train Loss:0.0195, Train ACC:99.9000\n",
      "Current time 12892.82 min\n",
      "Batch 2000, Train Loss:0.0203, Train ACC:99.8812\n",
      "Current time 12900.94 min\n",
      "Batch 2500, Train Loss:0.0209, Train ACC:99.8850\n",
      "Current time 12909.09 min\n",
      "Batch 3000, Train Loss:0.0204, Train ACC:99.8958\n",
      "Current time 12917.21 min\n",
      "Batch 3500, Train Loss:0.0199, Train ACC:99.9071\n",
      "Current time 12925.34 min\n",
      "Batch 4000, Train Loss:0.0197, Train ACC:99.9125\n",
      "Current time 12933.47 min\n",
      "Batch 4500, Train Loss:0.0198, Train ACC:99.9111\n",
      "Current time 12941.63 min\n",
      "Batch 5000, Train Loss:0.0200, Train ACC:99.9100\n",
      "Current time 12949.81 min\n",
      "Batch 5500, Train Loss:0.0197, Train ACC:99.9182\n",
      "Current time 12957.95 min\n",
      "Batch 6000, Train Loss:0.0196, Train ACC:99.9208\n",
      "Current time 12966.09 min\n",
      "Batch 6500, Train Loss:0.0197, Train ACC:99.9154\n",
      "Current time 12974.24 min\n",
      "Batch 7000, Train Loss:0.0196, Train ACC:99.9161\n",
      "Current time 12982.41 min\n",
      "Batch 7500, Train Loss:0.0195, Train ACC:99.9183\n",
      "Current time 12990.55 min\n",
      "Batch 8000, Train Loss:0.0198, Train ACC:99.9125\n",
      "Current time 12998.69 min\n",
      "Batch 8500, Train Loss:0.0197, Train ACC:99.9176\n",
      "Current time 13006.84 min\n",
      "Batch 9000, Train Loss:0.0198, Train ACC:99.9125\n",
      "Current time 13015.02 min\n",
      "Batch 9500, Train Loss:0.0197, Train ACC:99.9132\n",
      "Current time 13023.20 min\n",
      "Batch 10000, Train Loss:0.0196, Train ACC:99.9163\n",
      "Current time 13031.36 min\n",
      "Batch 10500, Train Loss:0.0195, Train ACC:99.9190\n",
      "Current time 13039.50 min\n",
      "Batch 11000, Train Loss:0.0194, Train ACC:99.9227\n",
      "Current time 13047.64 min\n",
      "Batch 11500, Train Loss:0.0194, Train ACC:99.9239\n",
      "Current time 13055.81 min\n",
      "train Loss:0.0388  Acc:99.9225%\n",
      "Finish time 13062.81 min\n",
      "Epoch 88/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0189, Train ACC:99.9000\n",
      "Current time 13070.70 min\n",
      "Batch 1000, Train Loss:0.0201, Train ACC:99.9000\n",
      "Current time 13078.62 min\n",
      "Batch 1500, Train Loss:0.0198, Train ACC:99.9000\n",
      "Current time 13086.54 min\n",
      "Batch 2000, Train Loss:0.0197, Train ACC:99.9000\n",
      "Current time 13094.44 min\n",
      "Batch 2500, Train Loss:0.0197, Train ACC:99.8900\n",
      "Current time 13102.35 min\n",
      "Batch 3000, Train Loss:0.0194, Train ACC:99.9042\n",
      "Current time 13110.23 min\n",
      "Batch 3500, Train Loss:0.0198, Train ACC:99.8893\n",
      "Current time 13118.13 min\n",
      "Batch 4000, Train Loss:0.0199, Train ACC:99.8937\n",
      "Current time 13126.02 min\n",
      "Batch 4500, Train Loss:0.0201, Train ACC:99.8944\n",
      "Current time 13133.96 min\n",
      "Batch 5000, Train Loss:0.0198, Train ACC:99.9025\n",
      "Current time 13141.81 min\n",
      "Batch 5500, Train Loss:0.0196, Train ACC:99.9114\n",
      "Current time 13149.67 min\n",
      "Batch 6000, Train Loss:0.0196, Train ACC:99.9125\n",
      "Current time 13157.50 min\n",
      "Batch 6500, Train Loss:0.0197, Train ACC:99.9096\n",
      "Current time 13165.35 min\n",
      "Batch 7000, Train Loss:0.0196, Train ACC:99.9107\n",
      "Current time 13173.19 min\n",
      "Batch 7500, Train Loss:0.0195, Train ACC:99.9150\n",
      "Current time 13181.05 min\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Batch 8000, Train Loss:0.0193, Train ACC:99.9203\n",
      "Current time 13188.88 min\n",
      "Batch 8500, Train Loss:0.0192, Train ACC:99.9221\n",
      "Current time 13196.73 min\n",
      "Batch 9000, Train Loss:0.0191, Train ACC:99.9236\n",
      "Current time 13204.58 min\n",
      "Batch 9500, Train Loss:0.0194, Train ACC:99.9184\n",
      "Current time 13212.47 min\n",
      "Batch 10000, Train Loss:0.0197, Train ACC:99.9112\n",
      "Current time 13220.58 min\n",
      "Batch 10500, Train Loss:0.0198, Train ACC:99.9107\n",
      "Current time 13228.73 min\n",
      "Batch 11000, Train Loss:0.0200, Train ACC:99.9068\n",
      "Current time 13236.85 min\n",
      "Batch 11500, Train Loss:0.0199, Train ACC:99.9087\n",
      "Current time 13244.99 min\n",
      "train Loss:0.0398  Acc:99.9109%\n",
      "Finish time 13251.97 min\n",
      "Epoch 89/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0202, Train ACC:99.9750\n",
      "Current time 13260.09 min\n",
      "Batch 1000, Train Loss:0.0188, Train ACC:99.9625\n",
      "Current time 13268.21 min\n",
      "Batch 1500, Train Loss:0.0179, Train ACC:99.9750\n",
      "Current time 13276.25 min\n",
      "Batch 2000, Train Loss:0.0177, Train ACC:99.9750\n",
      "Current time 13284.34 min\n",
      "Batch 2500, Train Loss:0.0185, Train ACC:99.9550\n",
      "Current time 13292.41 min\n",
      "Batch 3000, Train Loss:0.0185, Train ACC:99.9500\n",
      "Current time 13300.48 min\n",
      "Batch 3500, Train Loss:0.0184, Train ACC:99.9571\n",
      "Current time 13308.57 min\n",
      "Batch 4000, Train Loss:0.0184, Train ACC:99.9562\n",
      "Current time 13316.67 min\n",
      "Batch 4500, Train Loss:0.0183, Train ACC:99.9528\n",
      "Current time 13324.73 min\n",
      "Batch 5000, Train Loss:0.0185, Train ACC:99.9475\n",
      "Current time 13332.82 min\n",
      "Batch 5500, Train Loss:0.0184, Train ACC:99.9500\n",
      "Current time 13340.99 min\n",
      "Batch 6000, Train Loss:0.0184, Train ACC:99.9479\n",
      "Current time 13349.14 min\n",
      "Batch 6500, Train Loss:0.0184, Train ACC:99.9481\n",
      "Current time 13357.24 min\n",
      "Batch 7000, Train Loss:0.0184, Train ACC:99.9446\n",
      "Current time 13365.38 min\n",
      "Batch 7500, Train Loss:0.0186, Train ACC:99.9417\n",
      "Current time 13373.56 min\n",
      "Batch 8000, Train Loss:0.0187, Train ACC:99.9391\n",
      "Current time 13381.70 min\n",
      "Batch 8500, Train Loss:0.0186, Train ACC:99.9397\n",
      "Current time 13389.85 min\n",
      "Batch 9000, Train Loss:0.0188, Train ACC:99.9375\n",
      "Current time 13397.98 min\n",
      "Batch 9500, Train Loss:0.0188, Train ACC:99.9382\n",
      "Current time 13406.15 min\n",
      "Batch 10000, Train Loss:0.0190, Train ACC:99.9325\n",
      "Current time 13414.31 min\n",
      "Batch 10500, Train Loss:0.0190, Train ACC:99.9321\n",
      "Current time 13422.45 min\n",
      "Batch 11000, Train Loss:0.0189, Train ACC:99.9341\n",
      "Current time 13430.60 min\n",
      "Batch 11500, Train Loss:0.0189, Train ACC:99.9359\n",
      "Current time 13438.78 min\n",
      "train Loss:0.0383  Acc:99.9277%\n",
      "Finish time 13445.80 min\n",
      "Epoch 90/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0177, Train ACC:99.9750\n",
      "Current time 13453.83 min\n",
      "Batch 1000, Train Loss:0.0181, Train ACC:99.9625\n",
      "Current time 13461.87 min\n",
      "Batch 1500, Train Loss:0.0176, Train ACC:99.9750\n",
      "Current time 13469.89 min\n",
      "Batch 2000, Train Loss:0.0179, Train ACC:99.9625\n",
      "Current time 13477.91 min\n",
      "Batch 2500, Train Loss:0.0186, Train ACC:99.9400\n",
      "Current time 13485.93 min\n",
      "Batch 3000, Train Loss:0.0194, Train ACC:99.9208\n",
      "Current time 13493.94 min\n",
      "Batch 3500, Train Loss:0.0193, Train ACC:99.9179\n",
      "Current time 13502.01 min\n",
      "Batch 4000, Train Loss:0.0190, Train ACC:99.9281\n",
      "Current time 13510.01 min\n",
      "Batch 4500, Train Loss:0.0188, Train ACC:99.9361\n",
      "Current time 13518.02 min\n",
      "Batch 5000, Train Loss:0.0186, Train ACC:99.9425\n",
      "Current time 13526.04 min\n",
      "Batch 5500, Train Loss:0.0187, Train ACC:99.9409\n",
      "Current time 13534.08 min\n",
      "Batch 6000, Train Loss:0.0189, Train ACC:99.9333\n",
      "Current time 13542.11 min\n",
      "Batch 6500, Train Loss:0.0191, Train ACC:99.9327\n",
      "Current time 13550.13 min\n",
      "Batch 7000, Train Loss:0.0194, Train ACC:99.9250\n",
      "Current time 13558.11 min\n",
      "Batch 7500, Train Loss:0.0194, Train ACC:99.9233\n",
      "Current time 13566.12 min\n",
      "Batch 8000, Train Loss:0.0195, Train ACC:99.9203\n",
      "Current time 13574.14 min\n",
      "Batch 8500, Train Loss:0.0195, Train ACC:99.9221\n",
      "Current time 13582.17 min\n",
      "Batch 9000, Train Loss:0.0197, Train ACC:99.9181\n",
      "Current time 13590.16 min\n",
      "Batch 9500, Train Loss:0.0196, Train ACC:99.9224\n",
      "Current time 13598.20 min\n",
      "Batch 10000, Train Loss:0.0196, Train ACC:99.9213\n",
      "Current time 13606.22 min\n",
      "Batch 10500, Train Loss:0.0195, Train ACC:99.9238\n",
      "Current time 13614.24 min\n",
      "Batch 11000, Train Loss:0.0194, Train ACC:99.9261\n",
      "Current time 13622.26 min\n",
      "Batch 11500, Train Loss:0.0193, Train ACC:99.9283\n",
      "Current time 13630.31 min\n",
      "train Loss:0.0385  Acc:99.9298%\n",
      "Finish time 13637.17 min\n",
      "Epoch 91/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0186, Train ACC:99.9500\n",
      "Current time 13645.12 min\n",
      "Batch 1000, Train Loss:0.0196, Train ACC:99.8875\n",
      "Current time 13653.04 min\n",
      "Batch 1500, Train Loss:0.0188, Train ACC:99.9167\n",
      "Current time 13660.95 min\n",
      "Batch 2000, Train Loss:0.0186, Train ACC:99.9250\n",
      "Current time 13668.87 min\n",
      "Batch 2500, Train Loss:0.0193, Train ACC:99.9100\n",
      "Current time 13676.83 min\n",
      "Batch 3000, Train Loss:0.0191, Train ACC:99.9208\n",
      "Current time 13684.74 min\n",
      "Batch 3500, Train Loss:0.0188, Train ACC:99.9321\n",
      "Current time 13692.65 min\n",
      "Batch 4000, Train Loss:0.0186, Train ACC:99.9406\n",
      "Current time 13700.66 min\n",
      "Batch 4500, Train Loss:0.0184, Train ACC:99.9472\n",
      "Current time 13708.88 min\n",
      "Batch 5000, Train Loss:0.0187, Train ACC:99.9450\n",
      "Current time 13717.06 min\n",
      "Batch 5500, Train Loss:0.0186, Train ACC:99.9477\n",
      "Current time 13725.21 min\n",
      "Batch 6000, Train Loss:0.0186, Train ACC:99.9458\n",
      "Current time 13733.37 min\n",
      "Batch 6500, Train Loss:0.0188, Train ACC:99.9404\n",
      "Current time 13741.55 min\n",
      "Batch 7000, Train Loss:0.0189, Train ACC:99.9339\n",
      "Current time 13749.72 min\n",
      "Batch 7500, Train Loss:0.0191, Train ACC:99.9267\n",
      "Current time 13757.88 min\n",
      "Batch 8000, Train Loss:0.0191, Train ACC:99.9266\n",
      "Current time 13766.05 min\n",
      "Batch 8500, Train Loss:0.0190, Train ACC:99.9279\n",
      "Current time 13774.24 min\n",
      "Batch 9000, Train Loss:0.0191, Train ACC:99.9264\n",
      "Current time 13782.40 min\n",
      "Batch 9500, Train Loss:0.0191, Train ACC:99.9276\n",
      "Current time 13790.55 min\n",
      "Batch 10000, Train Loss:0.0190, Train ACC:99.9300\n",
      "Current time 13798.70 min\n",
      "Batch 10500, Train Loss:0.0190, Train ACC:99.9321\n",
      "Current time 13806.87 min\n",
      "Batch 11000, Train Loss:0.0189, Train ACC:99.9352\n",
      "Current time 13815.04 min\n",
      "Batch 11500, Train Loss:0.0188, Train ACC:99.9380\n",
      "Current time 13823.21 min\n",
      "train Loss:0.0378  Acc:99.9382%\n",
      "Finish time 13830.21 min\n",
      "Epoch 92/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0180, Train ACC:99.9750\n",
      "Current time 13838.19 min\n",
      "Batch 1000, Train Loss:0.0176, Train ACC:99.9625\n",
      "Current time 13846.17 min\n",
      "Batch 1500, Train Loss:0.0180, Train ACC:99.9417\n",
      "Current time 13854.15 min\n",
      "Batch 2000, Train Loss:0.0193, Train ACC:99.9188\n",
      "Current time 13862.12 min\n",
      "Batch 2500, Train Loss:0.0198, Train ACC:99.9150\n",
      "Current time 13870.14 min\n",
      "Batch 3000, Train Loss:0.0201, Train ACC:99.9083\n",
      "Current time 13878.13 min\n",
      "Batch 3500, Train Loss:0.0196, Train ACC:99.9214\n",
      "Current time 13886.11 min\n",
      "Batch 4000, Train Loss:0.0193, Train ACC:99.9313\n",
      "Current time 13894.12 min\n",
      "Batch 4500, Train Loss:0.0195, Train ACC:99.9278\n",
      "Current time 13902.12 min\n",
      "Batch 5000, Train Loss:0.0194, Train ACC:99.9300\n",
      "Current time 13910.12 min\n",
      "Batch 5500, Train Loss:0.0194, Train ACC:99.9273\n",
      "Current time 13918.11 min\n",
      "Batch 6000, Train Loss:0.0196, Train ACC:99.9229\n",
      "Current time 13926.11 min\n",
      "Batch 6500, Train Loss:0.0199, Train ACC:99.9154\n",
      "Current time 13934.08 min\n",
      "Batch 7000, Train Loss:0.0197, Train ACC:99.9214\n",
      "Current time 13942.09 min\n",
      "Batch 7500, Train Loss:0.0197, Train ACC:99.9183\n",
      "Current time 13950.07 min\n",
      "Batch 8000, Train Loss:0.0199, Train ACC:99.9172\n",
      "Current time 13958.05 min\n",
      "Batch 8500, Train Loss:0.0198, Train ACC:99.9176\n",
      "Current time 13966.01 min\n",
      "Batch 9000, Train Loss:0.0196, Train ACC:99.9222\n",
      "Current time 13973.97 min\n",
      "Batch 9500, Train Loss:0.0197, Train ACC:99.9224\n",
      "Current time 13981.96 min\n",
      "Batch 10000, Train Loss:0.0196, Train ACC:99.9237\n",
      "Current time 13989.93 min\n",
      "Batch 10500, Train Loss:0.0196, Train ACC:99.9226\n",
      "Current time 13997.89 min\n",
      "Batch 11000, Train Loss:0.0195, Train ACC:99.9250\n",
      "Current time 14005.86 min\n",
      "Batch 11500, Train Loss:0.0194, Train ACC:99.9261\n",
      "Current time 14013.85 min\n",
      "train Loss:0.0389  Acc:99.9235%\n",
      "Finish time 14020.68 min\n",
      "Epoch 93/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0196, Train ACC:99.9000\n",
      "Current time 14028.61 min\n",
      "Batch 1000, Train Loss:0.0183, Train ACC:99.9500\n",
      "Current time 14036.53 min\n",
      "Batch 1500, Train Loss:0.0178, Train ACC:99.9667\n",
      "Current time 14044.42 min\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Batch 2000, Train Loss:0.0187, Train ACC:99.9188\n",
      "Current time 14052.32 min\n",
      "Batch 2500, Train Loss:0.0186, Train ACC:99.9250\n",
      "Current time 14060.22 min\n",
      "Batch 3000, Train Loss:0.0183, Train ACC:99.9375\n",
      "Current time 14068.12 min\n",
      "Batch 3500, Train Loss:0.0182, Train ACC:99.9393\n",
      "Current time 14076.04 min\n",
      "Batch 4000, Train Loss:0.0182, Train ACC:99.9375\n",
      "Current time 14083.93 min\n",
      "Batch 4500, Train Loss:0.0182, Train ACC:99.9361\n",
      "Current time 14091.80 min\n",
      "Batch 5000, Train Loss:0.0184, Train ACC:99.9250\n",
      "Current time 14099.69 min\n",
      "Batch 5500, Train Loss:0.0184, Train ACC:99.9227\n",
      "Current time 14107.61 min\n",
      "Batch 6000, Train Loss:0.0188, Train ACC:99.9188\n",
      "Current time 14115.49 min\n",
      "Batch 6500, Train Loss:0.0187, Train ACC:99.9192\n",
      "Current time 14123.37 min\n",
      "Batch 7000, Train Loss:0.0189, Train ACC:99.9125\n",
      "Current time 14131.30 min\n",
      "Batch 7500, Train Loss:0.0191, Train ACC:99.9083\n",
      "Current time 14139.18 min\n",
      "Batch 8000, Train Loss:0.0192, Train ACC:99.9078\n",
      "Current time 14147.10 min\n",
      "Batch 8500, Train Loss:0.0192, Train ACC:99.9088\n",
      "Current time 14154.99 min\n",
      "Batch 9000, Train Loss:0.0193, Train ACC:99.9125\n",
      "Current time 14162.89 min\n",
      "Batch 9500, Train Loss:0.0193, Train ACC:99.9105\n",
      "Current time 14170.79 min\n",
      "Batch 10000, Train Loss:0.0192, Train ACC:99.9150\n",
      "Current time 14178.72 min\n",
      "Batch 10500, Train Loss:0.0192, Train ACC:99.9167\n",
      "Current time 14186.62 min\n",
      "Batch 11000, Train Loss:0.0192, Train ACC:99.9182\n",
      "Current time 14194.53 min\n",
      "Batch 11500, Train Loss:0.0192, Train ACC:99.9174\n",
      "Current time 14202.43 min\n",
      "train Loss:0.0382  Acc:99.9193%\n",
      "Finish time 14209.19 min\n",
      "Epoch 94/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0170, Train ACC:100.0000\n",
      "Current time 14217.12 min\n",
      "Batch 1000, Train Loss:0.0170, Train ACC:99.9750\n",
      "Current time 14224.99 min\n",
      "Batch 1500, Train Loss:0.0170, Train ACC:99.9667\n",
      "Current time 14232.88 min\n",
      "Batch 2000, Train Loss:0.0176, Train ACC:99.9562\n",
      "Current time 14240.81 min\n",
      "Batch 2500, Train Loss:0.0180, Train ACC:99.9500\n",
      "Current time 14248.68 min\n",
      "Batch 3000, Train Loss:0.0182, Train ACC:99.9417\n",
      "Current time 14256.59 min\n",
      "Batch 3500, Train Loss:0.0181, Train ACC:99.9429\n",
      "Current time 14264.53 min\n",
      "Batch 4000, Train Loss:0.0180, Train ACC:99.9437\n",
      "Current time 14272.43 min\n",
      "Batch 4500, Train Loss:0.0182, Train ACC:99.9444\n",
      "Current time 14280.33 min\n",
      "Batch 5000, Train Loss:0.0184, Train ACC:99.9450\n",
      "Current time 14288.21 min\n",
      "Batch 5500, Train Loss:0.0189, Train ACC:99.9364\n",
      "Current time 14296.12 min\n",
      "Batch 6000, Train Loss:0.0187, Train ACC:99.9396\n",
      "Current time 14304.07 min\n",
      "Batch 6500, Train Loss:0.0188, Train ACC:99.9365\n",
      "Current time 14312.23 min\n",
      "Batch 7000, Train Loss:0.0188, Train ACC:99.9339\n",
      "Current time 14320.38 min\n",
      "Batch 7500, Train Loss:0.0187, Train ACC:99.9367\n",
      "Current time 14328.53 min\n",
      "Batch 8000, Train Loss:0.0187, Train ACC:99.9328\n",
      "Current time 14336.71 min\n",
      "Batch 8500, Train Loss:0.0186, Train ACC:99.9353\n",
      "Current time 14344.85 min\n",
      "Batch 9000, Train Loss:0.0186, Train ACC:99.9375\n",
      "Current time 14353.01 min\n",
      "Batch 9500, Train Loss:0.0186, Train ACC:99.9382\n",
      "Current time 14361.15 min\n",
      "Batch 10000, Train Loss:0.0185, Train ACC:99.9412\n",
      "Current time 14369.31 min\n",
      "Batch 10500, Train Loss:0.0185, Train ACC:99.9429\n",
      "Current time 14377.47 min\n",
      "Batch 11000, Train Loss:0.0187, Train ACC:99.9341\n",
      "Current time 14385.61 min\n",
      "Batch 11500, Train Loss:0.0188, Train ACC:99.9315\n",
      "Current time 14393.74 min\n",
      "train Loss:0.0376  Acc:99.9329%\n",
      "Finish time 14400.74 min\n",
      "Epoch 95/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0185, Train ACC:99.9250\n",
      "Current time 14408.78 min\n",
      "Batch 1000, Train Loss:0.0175, Train ACC:99.9625\n",
      "Current time 14416.83 min\n",
      "Batch 1500, Train Loss:0.0176, Train ACC:99.9500\n",
      "Current time 14424.89 min\n",
      "Batch 2000, Train Loss:0.0172, Train ACC:99.9625\n",
      "Current time 14432.96 min\n",
      "Batch 2500, Train Loss:0.0184, Train ACC:99.9500\n",
      "Current time 14441.03 min\n",
      "Batch 3000, Train Loss:0.0190, Train ACC:99.9292\n",
      "Current time 14449.09 min\n",
      "Batch 3500, Train Loss:0.0188, Train ACC:99.9321\n",
      "Current time 14457.15 min\n",
      "Batch 4000, Train Loss:0.0187, Train ACC:99.9281\n",
      "Current time 14465.18 min\n",
      "Batch 4500, Train Loss:0.0186, Train ACC:99.9306\n",
      "Current time 14473.24 min\n",
      "Batch 5000, Train Loss:0.0185, Train ACC:99.9325\n",
      "Current time 14481.30 min\n",
      "Batch 5500, Train Loss:0.0190, Train ACC:99.9205\n",
      "Current time 14489.34 min\n",
      "Batch 6000, Train Loss:0.0189, Train ACC:99.9271\n",
      "Current time 14497.37 min\n",
      "Batch 6500, Train Loss:0.0187, Train ACC:99.9327\n",
      "Current time 14505.42 min\n",
      "Batch 7000, Train Loss:0.0187, Train ACC:99.9321\n",
      "Current time 14513.47 min\n",
      "Batch 7500, Train Loss:0.0189, Train ACC:99.9300\n",
      "Current time 14521.53 min\n",
      "Batch 8000, Train Loss:0.0190, Train ACC:99.9297\n",
      "Current time 14529.57 min\n",
      "Batch 8500, Train Loss:0.0189, Train ACC:99.9324\n",
      "Current time 14537.61 min\n",
      "Batch 9000, Train Loss:0.0190, Train ACC:99.9278\n",
      "Current time 14545.64 min\n",
      "Batch 9500, Train Loss:0.0191, Train ACC:99.9263\n",
      "Current time 14553.70 min\n",
      "Batch 10000, Train Loss:0.0190, Train ACC:99.9287\n",
      "Current time 14561.72 min\n",
      "Batch 10500, Train Loss:0.0190, Train ACC:99.9298\n",
      "Current time 14569.73 min\n",
      "Batch 11000, Train Loss:0.0190, Train ACC:99.9284\n",
      "Current time 14577.79 min\n",
      "Batch 11500, Train Loss:0.0191, Train ACC:99.9261\n",
      "Current time 14585.84 min\n",
      "train Loss:0.0381  Acc:99.9266%\n",
      "Finish time 14592.73 min\n",
      "Epoch 96/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0203, Train ACC:99.9500\n",
      "Current time 14600.67 min\n",
      "Batch 1000, Train Loss:0.0196, Train ACC:99.9625\n",
      "Current time 14608.58 min\n",
      "Batch 1500, Train Loss:0.0188, Train ACC:99.9750\n",
      "Current time 14616.45 min\n",
      "Batch 2000, Train Loss:0.0194, Train ACC:99.9562\n",
      "Current time 14624.36 min\n",
      "Batch 2500, Train Loss:0.0190, Train ACC:99.9650\n",
      "Current time 14632.26 min\n",
      "Batch 3000, Train Loss:0.0192, Train ACC:99.9417\n",
      "Current time 14640.13 min\n",
      "Batch 3500, Train Loss:0.0196, Train ACC:99.9286\n",
      "Current time 14647.98 min\n",
      "Batch 4000, Train Loss:0.0193, Train ACC:99.9375\n",
      "Current time 14655.93 min\n",
      "Batch 4500, Train Loss:0.0192, Train ACC:99.9389\n",
      "Current time 14664.02 min\n",
      "Batch 5000, Train Loss:0.0191, Train ACC:99.9425\n",
      "Current time 14672.12 min\n",
      "Batch 5500, Train Loss:0.0191, Train ACC:99.9409\n",
      "Current time 14680.22 min\n",
      "Batch 6000, Train Loss:0.0191, Train ACC:99.9354\n",
      "Current time 14688.34 min\n",
      "Batch 6500, Train Loss:0.0190, Train ACC:99.9365\n",
      "Current time 14696.45 min\n",
      "Batch 7000, Train Loss:0.0188, Train ACC:99.9357\n",
      "Current time 14704.55 min\n",
      "Batch 7500, Train Loss:0.0190, Train ACC:99.9317\n",
      "Current time 14712.64 min\n",
      "Batch 8000, Train Loss:0.0190, Train ACC:99.9344\n",
      "Current time 14720.77 min\n",
      "Batch 8500, Train Loss:0.0189, Train ACC:99.9368\n",
      "Current time 14728.84 min\n",
      "Batch 9000, Train Loss:0.0189, Train ACC:99.9333\n",
      "Current time 14736.88 min\n",
      "Batch 9500, Train Loss:0.0191, Train ACC:99.9303\n",
      "Current time 14744.97 min\n",
      "Batch 10000, Train Loss:0.0190, Train ACC:99.9325\n",
      "Current time 14753.04 min\n",
      "Batch 10500, Train Loss:0.0189, Train ACC:99.9345\n",
      "Current time 14761.08 min\n",
      "Batch 11000, Train Loss:0.0188, Train ACC:99.9330\n",
      "Current time 14769.14 min\n",
      "Batch 11500, Train Loss:0.0188, Train ACC:99.9337\n",
      "Current time 14777.18 min\n",
      "train Loss:0.0375  Acc:99.9350%\n",
      "Finish time 14784.10 min\n",
      "Epoch 97/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0185, Train ACC:99.9500\n",
      "Current time 14791.98 min\n",
      "Batch 1000, Train Loss:0.0184, Train ACC:99.9375\n",
      "Current time 14799.86 min\n",
      "Batch 1500, Train Loss:0.0186, Train ACC:99.9333\n",
      "Current time 14807.73 min\n",
      "Batch 2000, Train Loss:0.0200, Train ACC:99.9125\n",
      "Current time 14815.61 min\n",
      "Batch 2500, Train Loss:0.0199, Train ACC:99.9200\n",
      "Current time 14823.48 min\n",
      "Batch 3000, Train Loss:0.0194, Train ACC:99.9333\n",
      "Current time 14831.37 min\n",
      "Batch 3500, Train Loss:0.0193, Train ACC:99.9393\n",
      "Current time 14839.25 min\n",
      "Batch 4000, Train Loss:0.0189, Train ACC:99.9469\n",
      "Current time 14847.13 min\n",
      "Batch 4500, Train Loss:0.0187, Train ACC:99.9500\n",
      "Current time 14855.03 min\n",
      "Batch 5000, Train Loss:0.0185, Train ACC:99.9550\n",
      "Current time 14862.93 min\n",
      "Batch 5500, Train Loss:0.0185, Train ACC:99.9523\n",
      "Current time 14870.83 min\n",
      "Batch 6000, Train Loss:0.0186, Train ACC:99.9458\n",
      "Current time 14878.72 min\n",
      "Batch 6500, Train Loss:0.0186, Train ACC:99.9462\n",
      "Current time 14886.60 min\n",
      "Batch 7000, Train Loss:0.0185, Train ACC:99.9482\n",
      "Current time 14894.48 min\n",
      "Batch 7500, Train Loss:0.0184, Train ACC:99.9517\n",
      "Current time 14902.34 min\n",
      "Batch 8000, Train Loss:0.0189, Train ACC:99.9437\n",
      "Current time 14910.21 min\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Batch 8500, Train Loss:0.0189, Train ACC:99.9412\n",
      "Current time 14918.07 min\n",
      "Batch 9000, Train Loss:0.0189, Train ACC:99.9403\n",
      "Current time 14925.94 min\n",
      "Batch 9500, Train Loss:0.0190, Train ACC:99.9408\n",
      "Current time 14933.80 min\n",
      "Batch 10000, Train Loss:0.0193, Train ACC:99.9313\n",
      "Current time 14941.69 min\n",
      "Batch 10500, Train Loss:0.0192, Train ACC:99.9333\n",
      "Current time 14949.53 min\n",
      "Batch 11000, Train Loss:0.0194, Train ACC:99.9295\n",
      "Current time 14957.45 min\n",
      "Batch 11500, Train Loss:0.0194, Train ACC:99.9315\n",
      "Current time 14965.32 min\n",
      "train Loss:0.0387  Acc:99.9340%\n",
      "Finish time 14972.08 min\n",
      "Epoch 98/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0208, Train ACC:99.8750\n",
      "Current time 14979.87 min\n",
      "Batch 1000, Train Loss:0.0193, Train ACC:99.9000\n",
      "Current time 14987.65 min\n",
      "Batch 1500, Train Loss:0.0184, Train ACC:99.9250\n",
      "Current time 14995.43 min\n",
      "Batch 2000, Train Loss:0.0202, Train ACC:99.8937\n",
      "Current time 15003.22 min\n",
      "Batch 2500, Train Loss:0.0197, Train ACC:99.9100\n",
      "Current time 15011.02 min\n",
      "Batch 3000, Train Loss:0.0202, Train ACC:99.9125\n",
      "Current time 15018.81 min\n",
      "Batch 3500, Train Loss:0.0201, Train ACC:99.9179\n",
      "Current time 15026.57 min\n",
      "Batch 4000, Train Loss:0.0205, Train ACC:99.9125\n",
      "Current time 15034.37 min\n",
      "Batch 4500, Train Loss:0.0204, Train ACC:99.9111\n",
      "Current time 15042.16 min\n",
      "Batch 5000, Train Loss:0.0201, Train ACC:99.9200\n",
      "Current time 15049.94 min\n",
      "Batch 5500, Train Loss:0.0198, Train ACC:99.9227\n",
      "Current time 15057.77 min\n",
      "Batch 6000, Train Loss:0.0196, Train ACC:99.9292\n",
      "Current time 15065.53 min\n",
      "Batch 6500, Train Loss:0.0194, Train ACC:99.9288\n",
      "Current time 15073.31 min\n",
      "Batch 7000, Train Loss:0.0195, Train ACC:99.9286\n",
      "Current time 15081.23 min\n",
      "Batch 7500, Train Loss:0.0196, Train ACC:99.9233\n",
      "Current time 15089.26 min\n",
      "Batch 8000, Train Loss:0.0196, Train ACC:99.9250\n",
      "Current time 15097.33 min\n",
      "Batch 8500, Train Loss:0.0194, Train ACC:99.9279\n",
      "Current time 15105.36 min\n",
      "Batch 9000, Train Loss:0.0193, Train ACC:99.9292\n",
      "Current time 15113.42 min\n",
      "Batch 9500, Train Loss:0.0193, Train ACC:99.9289\n",
      "Current time 15121.52 min\n",
      "Batch 10000, Train Loss:0.0192, Train ACC:99.9325\n",
      "Current time 15129.61 min\n",
      "Batch 10500, Train Loss:0.0190, Train ACC:99.9357\n",
      "Current time 15137.72 min\n",
      "Batch 11000, Train Loss:0.0189, Train ACC:99.9386\n",
      "Current time 15145.80 min\n",
      "Batch 11500, Train Loss:0.0189, Train ACC:99.9380\n",
      "Current time 15153.89 min\n",
      "train Loss:0.0375  Acc:99.9403%\n",
      "Finish time 15160.82 min\n",
      "Epoch 99/99\n",
      "----------\n",
      "Training...\n",
      "Batch 500, Train Loss:0.0163, Train ACC:100.0000\n",
      "Current time 15168.82 min\n",
      "Batch 1000, Train Loss:0.0187, Train ACC:99.9500\n",
      "Current time 15176.82 min\n",
      "Batch 1500, Train Loss:0.0184, Train ACC:99.9417\n",
      "Current time 15184.82 min\n",
      "Batch 2000, Train Loss:0.0195, Train ACC:99.9188\n",
      "Current time 15192.82 min\n",
      "Batch 2500, Train Loss:0.0190, Train ACC:99.9250\n",
      "Current time 15200.81 min\n",
      "Batch 3000, Train Loss:0.0188, Train ACC:99.9292\n",
      "Current time 15208.83 min\n",
      "Batch 3500, Train Loss:0.0185, Train ACC:99.9357\n",
      "Current time 15216.86 min\n",
      "Batch 4000, Train Loss:0.0195, Train ACC:99.9281\n",
      "Current time 15224.89 min\n",
      "Batch 4500, Train Loss:0.0193, Train ACC:99.9333\n",
      "Current time 15232.90 min\n",
      "Batch 5000, Train Loss:0.0198, Train ACC:99.9250\n",
      "Current time 15240.92 min\n",
      "Batch 5500, Train Loss:0.0197, Train ACC:99.9273\n",
      "Current time 15248.92 min\n",
      "Batch 6000, Train Loss:0.0196, Train ACC:99.9313\n",
      "Current time 15256.94 min\n",
      "Batch 6500, Train Loss:0.0195, Train ACC:99.9269\n",
      "Current time 15264.94 min\n",
      "Batch 7000, Train Loss:0.0194, Train ACC:99.9250\n",
      "Current time 15272.94 min\n",
      "Batch 7500, Train Loss:0.0200, Train ACC:99.9100\n",
      "Current time 15280.98 min\n",
      "Batch 8000, Train Loss:0.0199, Train ACC:99.9109\n",
      "Current time 15289.01 min\n",
      "Batch 8500, Train Loss:0.0199, Train ACC:99.9103\n",
      "Current time 15297.02 min\n",
      "Batch 9000, Train Loss:0.0197, Train ACC:99.9139\n",
      "Current time 15305.03 min\n",
      "Batch 9500, Train Loss:0.0197, Train ACC:99.9145\n",
      "Current time 15313.06 min\n",
      "Batch 10000, Train Loss:0.0195, Train ACC:99.9163\n",
      "Current time 15321.11 min\n",
      "Batch 10500, Train Loss:0.0195, Train ACC:99.9167\n",
      "Current time 15329.11 min\n",
      "Batch 11000, Train Loss:0.0194, Train ACC:99.9193\n",
      "Current time 15337.11 min\n",
      "Batch 11500, Train Loss:0.0195, Train ACC:99.9174\n",
      "Current time 15345.13 min\n",
      "train Loss:0.0388  Acc:99.9183%\n",
      "Finish time 15352.02 min\n"
     ]
    }
   ],
   "source": [
    "epoch_n = 100\n",
    "time_open = time.time()\n",
    "f = open('Loss_log.txt', 'a+')\n",
    "f.write(\"\\nDOAT_Slices_deeplabv3 begin\\n\")\n",
    "\n",
    "for epoch in range(20, epoch_n):  ##### 20\n",
    "    print(\"Epoch {}/{}\".format(epoch, epoch_n - 1))\n",
    "    print(\"-\"*10)\n",
    "    \n",
    "    for phase in [\"train\"]: # \"valid\"\n",
    "        if phase == \"train\":\n",
    "            print(\"Training...\")\n",
    "            model_deeplab.train(True)\n",
    "            model_res.train(True)\n",
    "        else:\n",
    "            print(\"Valid...\")\n",
    "            model_deeplab.train(False)\n",
    "            model_res.train(False) # 把BatchNormalization和DropOut固定住，不会取平均，而是用训练好的值\n",
    "            \n",
    "        running_loss = 0.0\n",
    "        running_corrects = 0.0\n",
    "        \n",
    "        for batch, data in enumerate(dataloader[phase], start = 1): #组合为一个索引序列\n",
    "            X, y = data\n",
    "            if Use_gpu:\n",
    "                X, y = Variable(X.cuda()), Variable(y.cuda())\n",
    "            else:\n",
    "                X, y = Variable(X), Variable(y)\n",
    "            \n",
    "            # 添加高斯掩膜\n",
    "            y_fcn = model_deeplab.forward(X * masked_gaussian.to(X.device))['out']\n",
    "            \n",
    "            error_map = torch.pow(torch.cat([X]*cls_num, axis=1) - y_fcn, 2)\n",
    "            y_pred = model_res.forward(error_map)\n",
    "            \n",
    "            \n",
    "            _, pred = torch.max(y_pred.data, 1)\n",
    "            \n",
    "            loss = loss_fcn(X, y_fcn, y) + loss_res(y_pred, y)\n",
    "            \n",
    "            # 对参数梯度的归零\n",
    "            optimizer.zero_grad()\n",
    "            if phase == \"train\":\n",
    "                # 计算反向传播梯度值 P167\n",
    "                loss.backward() \n",
    "                # 对节点的参数进行梯度更新 P169\n",
    "                optimizer.step()\n",
    "                \n",
    "            running_loss += float(loss.data)\n",
    "            running_corrects += float(torch.sum(pred == y.data))\n",
    "            \n",
    "            if batch%500 == 0 and phase == \"train\":\n",
    "                print(\"Batch {}, Train Loss:{:.4f}, Train ACC:{:.4f}\"\n",
    "                      .format(batch, running_loss/batch, 100*running_corrects/batch/len(pred)))\n",
    "                print(\"Current time {:.2f} min\".format((time.time() - time_open)/60))\n",
    "            \n",
    "        epoch_loss = running_loss*16/len(image_datasets[phase])\n",
    "        epoch_acc = 100*running_corrects/len(image_datasets[phase])\n",
    "        \n",
    "        print(\"{} Loss:{:.4f}  Acc:{:.4f}%\".format(phase, epoch_loss, epoch_acc))\n",
    "        time_end = time.time() - time_open\n",
    "        print(\"Finish time {:.2f} min\".format(time_end/60))\n",
    "        \n",
    "        if torch.cuda.device_count() > 1:\n",
    "            torch.save({'model_res':model_res.module.state_dict(),\n",
    "                        'model_deeplab':model_deeplab.state_dict()}, # module.\n",
    "                        'DOAT_Slices_deeplab-ResNet50_epoch_{}.pth'.format(epoch))\n",
    "        else:\n",
    "            torch.save({'model_res':model_res.state_dict(), \n",
    "                        'model_deeplab':model_deeplab.state_dict()},\n",
    "                        'DOAT_Slices_deeplab-ResNet50_epoch_{}.pth'.format(epoch))\n",
    "    \n",
    "        \n",
    "# 清除显卡缓存\n",
    "torch.cuda.empty_cache()\n",
    "f.close()\n",
    "if torch.cuda.device_count() > 1:\n",
    "    torch.save({'model_res':model_res.module.state_dict(),\n",
    "                'model_deeplab':model_deeplab.state_dict()}, #.module\n",
    "                'DOAT_Slices_deeplab-ResNet50_model_last.pth')\n",
    "else:\n",
    "    torch.save({'model_res':model_res.state_dict(), \n",
    "                'model_deeplab':model_deeplab.state_dict()},\n",
    "                'DOAT_Slices_deeplab-ResNet50_model_last.pth')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "if torch.cuda.device_count() > 1:\n",
    "    torch.save({'model_res':model_res.module.state_dict(),\n",
    "                'model_deeplab':model_deeplab.state_dict()}, #.module\n",
    "                'DOAT_Slices_deeplab-ResNet50_model_last.pth')\n",
    "else:\n",
    "    torch.save({'model_res':model_res.state_dict(), \n",
    "                'model_deeplab':model_deeplab.state_dict()},\n",
    "                'DOAT_Slices_deeplab-ResNet50_model_last.pth')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "休眠 2022-07-26 17:52:10.801712\n",
      "休眠 2022-07-27 11:04:16.536572\n"
     ]
    }
   ],
   "source": [
    "print(\"休眠\",datetime.datetime.now())\n",
    "os.system(\"rundll32.exe powrprof.dll,SetSuspendState 0,1,0\")\n",
    "os.system(\"rundll32.exe powrprof.dll,SetSuspendState\")\n",
    "print(\"休眠\",datetime.datetime.now())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Valid"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "_CudaDeviceProperties(name='NVIDIA GeForce GTX 1080', major=6, minor=1, total_memory=8192MB, multi_processor_count=20)\n",
      "Let's use 2 GPUs!\n"
     ]
    }
   ],
   "source": [
    "# 配置GPU，开始进行训练\n",
    "Use_gpu = torch.cuda.is_available()\n",
    "\n",
    "if Use_gpu:\n",
    "    print( torch.cuda.get_device_properties(device = 0) )\n",
    "    model_deeplab = model_deeplab.cuda()\n",
    "    model_res = model_res.cuda()\n",
    "    \n",
    "# 配置多GPU\n",
    "if torch.cuda.device_count() > 1:\n",
    "    os.environ['CUDA_VISIBLE_DEVICES'] = '0,1'  # \n",
    "    print(\"Let's use\", torch.cuda.device_count(), \"GPUs!\")\n",
    "    model_deeplab = torch.nn.DataParallel(model_deeplab)\n",
    "    model_res = torch.nn.DataParallel(model_res)\n",
    "    \n",
    "masked_gaussian = masked_gaussian.to(next(model_deeplab.parameters()).device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Valid...\n",
      "Batch 500, Train Loss:0.0168, Train ACC:99.9500\n",
      "Current time 2.46 min\n",
      "Batch 1000, Train Loss:0.0162, Train ACC:99.9750\n",
      "Current time 4.84 min\n",
      "Batch 1500, Train Loss:0.0160, Train ACC:99.9750\n",
      "Current time 7.23 min\n",
      "Batch 2000, Train Loss:0.0159, Train ACC:99.9750\n",
      "Current time 9.64 min\n",
      "Batch 2500, Train Loss:0.0164, Train ACC:99.9600\n",
      "Current time 12.04 min\n",
      "Batch 3000, Train Loss:0.0165, Train ACC:99.9500\n",
      "Current time 14.44 min\n",
      "Batch 3500, Train Loss:0.0175, Train ACC:99.9357\n",
      "Current time 16.85 min\n",
      "Batch 4000, Train Loss:0.0172, Train ACC:99.9437\n",
      "Current time 19.26 min\n",
      "Batch 4500, Train Loss:0.0172, Train ACC:99.9389\n",
      "Current time 21.67 min\n",
      "Batch 5000, Train Loss:0.0171, Train ACC:99.9425\n",
      "Current time 24.09 min\n",
      "Batch 5500, Train Loss:0.0169, Train ACC:99.9432\n",
      "Current time 26.52 min\n",
      "Batch 6000, Train Loss:0.0169, Train ACC:99.9437\n",
      "Current time 28.93 min\n",
      "Batch 6500, Train Loss:0.0168, Train ACC:99.9481\n",
      "Current time 31.35 min\n",
      "Batch 7000, Train Loss:0.0170, Train ACC:99.9446\n",
      "Current time 33.78 min\n",
      "Batch 7500, Train Loss:0.0170, Train ACC:99.9467\n",
      "Current time 36.21 min\n",
      "Batch 8000, Train Loss:0.0172, Train ACC:99.9437\n",
      "Current time 38.65 min\n",
      "Batch 8500, Train Loss:0.0172, Train ACC:99.9412\n",
      "Current time 41.08 min\n",
      "Batch 9000, Train Loss:0.0173, Train ACC:99.9389\n",
      "Current time 43.50 min\n",
      "Batch 9500, Train Loss:0.0175, Train ACC:99.9368\n",
      "Current time 45.92 min\n",
      "Batch 10000, Train Loss:0.0175, Train ACC:99.9350\n",
      "Current time 48.36 min\n",
      "Batch 10500, Train Loss:0.0175, Train ACC:99.9369\n",
      "Current time 50.79 min\n",
      "Batch 11000, Train Loss:0.0175, Train ACC:99.9352\n",
      "Current time 53.22 min\n",
      "Batch 11500, Train Loss:0.0174, Train ACC:99.9359\n",
      "Current time 55.63 min\n",
      "train Loss:0.0347  Acc:99.9382%\n",
      "Finish time 57.72 min\n"
     ]
    }
   ],
   "source": [
    "time_open = time.time()\n",
    "print(\"Valid...\")\n",
    "\n",
    "phase = \"train\"\n",
    "\n",
    "model_deeplab.train(False) # 把BatchNormalization和DropOut固定住，不会取平均，而是用训练好的值\n",
    "model_res.train(False)         \n",
    "    \n",
    "running_loss = 0.0\n",
    "running_corrects = 0.0\n",
    "predict = [[] for i in range(len(example_classes))]\n",
    "\n",
    "for batch, data in enumerate(dataloader[phase], start = 1): #组合为一个索引序列\n",
    "    X, y = data\n",
    "    if Use_gpu:\n",
    "        X, y = Variable(X.cuda()), Variable(y.cuda())\n",
    "    else:\n",
    "        X, y = Variable(X), Variable(y)\n",
    "\n",
    "    # 添加高斯掩膜\n",
    "    y_fcn = model_deeplab.forward(X * masked_gaussian.to(X.device))['out']\n",
    "\n",
    "    error_map = torch.pow(torch.cat([X]*cls_num, axis=1) - y_fcn, 2)\n",
    "    y_pred = model_res.forward(error_map)\n",
    "    \n",
    "    _, pred = torch.max(y_pred.data, 1)\n",
    "\n",
    "    loss = loss_fcn(X, y_fcn, y) + loss_res(y_pred, y)\n",
    "    \n",
    "    # 记录预测结果\n",
    "    for i in range(len(y)):\n",
    "        predict[y[i]].append( y_pred[i].cpu().detach().numpy() )  #.argmax(dim=1)\n",
    "    \n",
    "    running_loss += float(loss.data)\n",
    "    running_corrects += float(torch.sum(pred == y.data))\n",
    "\n",
    "    if batch%500 == 0:# and phase == \"train\":\n",
    "        print(\"Batch {}, Train Loss:{:.4f}, Train ACC:{:.4f}\"\n",
    "              .format(batch, running_loss/batch, 100*running_corrects/batch/len(pred)))\n",
    "        print(\"Current time {:.2f} min\".format((time.time() - time_open)/60))\n",
    "\n",
    "epoch_loss = running_loss*16/len(image_datasets[phase])\n",
    "epoch_acc = 100*running_corrects/len(image_datasets[phase])\n",
    "\n",
    "print(\"{} Loss:{:.4f}  Acc:{:.4f}%\".format(phase, epoch_loss, epoch_acc))\n",
    "time_end = time.time() - time_open\n",
    "print(\"Finish time {:.2f} min\".format(time_end/60))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "baseball-diamond : 100.0%, Num : 415\n",
      "basketball-court : 100.0%, Num : 515\n",
      "bridge : 99.7340425531915%, Num : 1880\n",
      "ground-track-field : 100.0%, Num : 325\n",
      "harbor : 99.83255190890824%, Num : 5972\n",
      "helicopter : 100.0%, Num : 629\n",
      "large-vehicle : 99.9939893009557%, Num : 16637\n",
      "plane : 99.98743718592965%, Num : 7960\n",
      "roundabout : 100.0%, Num : 398\n",
      "ship : 99.98143541380463%, Num : 26933\n",
      "small-vehicle : 99.89543955602028%, Num : 24866\n",
      "soccer-ball-field : 99.68847352024922%, Num : 321\n",
      "storage-tank : 99.86628036550033%, Num : 4487\n",
      "swimming-pool : 100.0%, Num : 1731\n",
      "tennis-court : 99.83058026260059%, Num : 2361\n"
     ]
    }
   ],
   "source": [
    "pred_vec = [np.array([]) for i in range(len(example_classes))]\n",
    "Correct,Num = 0,0\n",
    "for cat in range(len(predict)):\n",
    "    catname = list(index_classes.keys())[list(index_classes.values()).index(cat)]\n",
    "    rate = np.sum(np.array([p.argmax() for p in predict[cat]]) == cat)/ len(predict[cat])\n",
    "    print(\"{} : {}%, Num : {}\".format(catname, rate*100, len(predict[cat])))\n",
    "    \n",
    "    Correct =  Correct + np.sum(np.array([p.argmax() for p in predict[cat]]) == cat)\n",
    "    Num = Num + len(predict[cat])\n",
    "    \n",
    "    pred_vec[cat] = np.sum( predict[cat], axis=0 ) / len(predict[cat])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "pd.DataFrame(pred_vec).to_csv(\"pred_vec_FCN.csv\")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}