chainer-mnist-gru.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h2>RNN for image classification</h2>\n",
    "\n",
    "Trained on mnist\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import chainer\n",
    "import chainer.links as L\n",
    "import chainer.functions as F\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import scipy.ndimage\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# get mnist dataset\n",
    "train, test = chainer.datasets.get_mnist()\n",
    "train_x, train_t = [np.array(i) for i in zip(*train)]\n",
    "test_x, test_t = [np.array(i) for i in zip(*test)]\n",
    "\n",
    "train_x = train_x.reshape((-1, 28, 28))\n",
    "test_x = test_x.reshape((-1, 28, 28))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# GRU aggregating loss over all time points\n",
    "\n",
    "Calculate GRU output per timepoint (1 to 28, width) using features (1 to 28, height). Take output per timepoint and map each to a softmax layer. Sum loss over all 28 time points and backpropagate. To calculate cross-entropy loss the true label is repeated 28 times"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "class NstepGRU_all(chainer.ChainList):\n",
    "    def __init__(self, depth, n_input, n_hidden, n_output):\n",
    "        links = []\n",
    "        links.append(L.NStepGRU(n_layers=depth, in_size=n_input, out_size=n_hidden, dropout=0.1, use_cudnn=False))\n",
    "        links.append(L.Linear(None, n_output))\n",
    "        super(NstepGRU_all, self).__init__(*links)\n",
    "\n",
    "    def __call__(self, x, train):\n",
    "        hy, ys = self[0](hx=None, xs=x, train=train)\n",
    "        y = [self[1](iy) for iy in ys]\n",
    "        return y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "net = NstepGRU_all(depth=2, n_input=28, n_hidden=50, n_output=10)\n",
    "optim = chainer.optimizers.Adam(alpha=1e-03)\n",
    "optim.setup(net)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 0: train accuracy = 0.682\n",
      "Epoch 1: train accuracy = 0.679\n",
      "Epoch 2: train accuracy = 0.6829\n",
      "Epoch 3: train accuracy = 0.6741\n",
      "Epoch 4: train accuracy = 0.6843\n",
      "Epoch 5: train accuracy = 0.6802\n",
      "Epoch 6: train accuracy = 0.6863\n",
      "Epoch 7: train accuracy = 0.6759\n",
      "Epoch 8: train accuracy = 0.6739\n",
      "Epoch 9: train accuracy = 0.6823\n",
      "Epoch 10: train accuracy = 0.6853\n",
      "Epoch 11: train accuracy = 0.6871\n",
      "Epoch 12: train accuracy = 0.6793\n",
      "Epoch 13: train accuracy = 0.6747\n",
      "Epoch 14: train accuracy = 0.6796\n",
      "Epoch 15: train accuracy = 0.6845\n",
      "Epoch 16: train accuracy = 0.6772\n",
      "Epoch 17: train accuracy = 0.6803\n",
      "Epoch 18: train accuracy = 0.6765\n",
      "Epoch 19: train accuracy = 0.6789\n",
      "Epoch 20: train accuracy = 0.6927\n",
      "Epoch 21: train accuracy = 0.6719\n",
      "Epoch 22: train accuracy = 0.6866\n",
      "Epoch 23: train accuracy = 0.6984\n",
      "Epoch 24: train accuracy = 0.678\n",
      "Epoch 25: train accuracy = 0.6843\n",
      "Epoch 26: train accuracy = 0.6808\n",
      "Epoch 27: train accuracy = 0.7026\n",
      "Epoch 28: train accuracy = 0.6842\n",
      "Epoch 29: train accuracy = 0.7005\n",
      "Epoch 30: train accuracy = 0.6751\n",
      "Epoch 31: train accuracy = 0.6871\n",
      "Epoch 32: train accuracy = 0.6842\n",
      "Epoch 33: train accuracy = 0.6757\n",
      "Epoch 34: train accuracy = 0.7035\n",
      "Epoch 35: train accuracy = 0.702\n",
      "Epoch 36: train accuracy = 0.6951\n",
      "Epoch 37: train accuracy = 0.6717\n",
      "Epoch 38: train accuracy = 0.6969\n",
      "Epoch 39: train accuracy = 0.6947\n",
      "Epoch 40: train accuracy = 0.7006\n",
      "Epoch 41: train accuracy = 0.6813\n",
      "Epoch 42: train accuracy = 0.6853\n",
      "Epoch 43: train accuracy = 0.6809\n",
      "Epoch 44: train accuracy = 0.6932\n",
      "Epoch 45: train accuracy = 0.6888\n",
      "Epoch 46: train accuracy = 0.6704\n",
      "Epoch 47: train accuracy = 0.6792\n",
      "Epoch 48: train accuracy = 0.7028\n",
      "Epoch 49: train accuracy = 0.7078\n",
      "Epoch 50: train accuracy = 0.681\n",
      "Epoch 51: train accuracy = 0.6888\n",
      "Epoch 52: train accuracy = 0.6884\n",
      "Epoch 53: train accuracy = 0.6791\n",
      "Epoch 54: train accuracy = 0.6804\n",
      "Epoch 55: train accuracy = 0.6935\n",
      "Epoch 56: train accuracy = 0.6937\n",
      "Epoch 57: train accuracy = 0.6822\n",
      "Epoch 58: train accuracy = 0.708\n",
      "Epoch 59: train accuracy = 0.687\n",
      "Epoch 60: train accuracy = 0.6871\n",
      "Epoch 61: train accuracy = 0.6883\n",
      "Epoch 62: train accuracy = 0.6775\n",
      "Epoch 63: train accuracy = 0.7006\n",
      "Epoch 64: train accuracy = 0.7047\n",
      "Epoch 65: train accuracy = 0.6761\n",
      "Epoch 66: train accuracy = 0.6947\n",
      "Epoch 67: train accuracy = 0.7004\n",
      "Epoch 68: train accuracy = 0.6911\n",
      "Epoch 69: train accuracy = 0.7026\n",
      "Epoch 70: train accuracy = 0.6913\n",
      "Epoch 71: train accuracy = 0.7045\n",
      "Epoch 72: train accuracy = 0.6952\n",
      "Epoch 73: train accuracy = 0.6796\n",
      "Epoch 74: train accuracy = 0.6971\n",
      "Epoch 75: train accuracy = 0.6866\n",
      "Epoch 76: train accuracy = 0.7113\n",
      "Epoch 77: train accuracy = 0.7129\n",
      "Epoch 78: train accuracy = 0.6967\n",
      "Epoch 79: train accuracy = 0.6933\n",
      "Epoch 80: train accuracy = 0.6886\n",
      "Epoch 81: train accuracy = 0.7154\n",
      "Epoch 82: train accuracy = 0.7019\n",
      "Epoch 83: train accuracy = 0.6943\n",
      "Epoch 84: train accuracy = 0.7077\n",
      "Epoch 85: train accuracy = 0.7046\n",
      "Epoch 86: train accuracy = 0.6908\n",
      "Epoch 87: train accuracy = 0.7073\n",
      "Epoch 88: train accuracy = 0.6865\n",
      "Epoch 89: train accuracy = 0.7043\n",
      "Epoch 90: train accuracy = 0.7113\n",
      "Epoch 91: train accuracy = 0.6968\n",
      "Epoch 92: train accuracy = 0.7006\n",
      "Epoch 93: train accuracy = 0.6962\n",
      "Epoch 94: train accuracy = 0.7024\n",
      "Epoch 95: train accuracy = 0.7018\n",
      "Epoch 96: train accuracy = 0.6888\n",
      "Epoch 97: train accuracy = 0.6994\n",
      "Epoch 98: train accuracy = 0.6907\n",
      "Epoch 99: train accuracy = 0.6906\n",
      "Epoch 100: train accuracy = 0.689\n",
      "Epoch 101: train accuracy = 0.7071\n",
      "Epoch 102: train accuracy = 0.6971\n",
      "Epoch 103: train accuracy = 0.691\n",
      "Epoch 104: train accuracy = 0.696\n",
      "Epoch 105: train accuracy = 0.6914\n",
      "Epoch 106: train accuracy = 0.7077\n",
      "Epoch 107: train accuracy = 0.6979\n",
      "Epoch 108: train accuracy = 0.6875\n",
      "Epoch 109: train accuracy = 0.6954\n",
      "Epoch 110: train accuracy = 0.7028\n",
      "Epoch 111: train accuracy = 0.7052\n",
      "Epoch 112: train accuracy = 0.7012\n",
      "Epoch 113: train accuracy = 0.7022\n",
      "Epoch 114: train accuracy = 0.7153\n",
      "Epoch 115: train accuracy = 0.7014\n",
      "Epoch 116: train accuracy = 0.6887\n",
      "Epoch 117: train accuracy = 0.7031\n",
      "Epoch 118: train accuracy = 0.6885\n",
      "Epoch 119: train accuracy = 0.6953\n",
      "Epoch 120: train accuracy = 0.7112\n",
      "Epoch 121: train accuracy = 0.6995\n",
      "Epoch 122: train accuracy = 0.7138\n",
      "Epoch 123: train accuracy = 0.7113\n",
      "Epoch 124: train accuracy = 0.7031\n",
      "Epoch 125: train accuracy = 0.7172\n",
      "Epoch 126: train accuracy = 0.7169\n",
      "Epoch 127: train accuracy = 0.6935\n",
      "Epoch 128: train accuracy = 0.691\n",
      "Epoch 129: train accuracy = 0.6974\n",
      "Epoch 130: train accuracy = 0.6985\n",
      "Epoch 131: train accuracy = 0.7055\n",
      "Epoch 132: train accuracy = 0.7067\n",
      "Epoch 133: train accuracy = 0.7065\n",
      "Epoch 134: train accuracy = 0.6924\n",
      "Epoch 135: train accuracy = 0.7105\n",
      "Epoch 136: train accuracy = 0.721\n",
      "Epoch 137: train accuracy = 0.7144\n",
      "Epoch 138: train accuracy = 0.7235\n",
      "Epoch 139: train accuracy = 0.7054\n",
      "Epoch 140: train accuracy = 0.7\n",
      "Epoch 141: train accuracy = 0.6894\n",
      "Epoch 142: train accuracy = 0.7115\n",
      "Epoch 143: train accuracy = 0.7108\n",
      "Epoch 144: train accuracy = 0.702\n",
      "Epoch 145: train accuracy = 0.7166\n",
      "Epoch 146: train accuracy = 0.7055\n",
      "Epoch 147: train accuracy = 0.7015\n",
      "Epoch 148: train accuracy = 0.7121\n",
      "Epoch 149: train accuracy = 0.6895\n",
      "Epoch 150: train accuracy = 0.6989\n",
      "Epoch 151: train accuracy = 0.7275\n",
      "Epoch 152: train accuracy = 0.7159\n",
      "Epoch 153: train accuracy = 0.7136\n",
      "Epoch 154: train accuracy = 0.7121\n",
      "Epoch 155: train accuracy = 0.709\n",
      "Epoch 156: train accuracy = 0.7098\n",
      "Epoch 157: train accuracy = 0.7087\n",
      "Epoch 158: train accuracy = 0.6935\n",
      "Epoch 159: train accuracy = 0.7137\n",
      "Epoch 160: train accuracy = 0.7154\n",
      "Epoch 161: train accuracy = 0.7131\n",
      "Epoch 162: train accuracy = 0.703\n",
      "Epoch 163: train accuracy = 0.6958\n",
      "Epoch 164: train accuracy = 0.712\n",
      "Epoch 165: train accuracy = 0.7235\n",
      "Epoch 166: train accuracy = 0.7059\n",
      "Epoch 167: train accuracy = 0.6864\n",
      "Epoch 168: train accuracy = 0.7029\n",
      "Epoch 169: train accuracy = 0.6901\n",
      "Epoch 170: train accuracy = 0.7113\n",
      "Epoch 171: train accuracy = 0.7137\n",
      "Epoch 172: train accuracy = 0.7173\n",
      "Epoch 173: train accuracy = 0.711\n",
      "Epoch 174: train accuracy = 0.702\n",
      "Epoch 175: train accuracy = 0.7157\n",
      "Epoch 176: train accuracy = 0.7052\n",
      "Epoch 177: train accuracy = 0.6879\n",
      "Epoch 178: train accuracy = 0.7196\n",
      "Epoch 179: train accuracy = 0.6971\n",
      "Epoch 180: train accuracy = 0.7099\n",
      "Epoch 181: train accuracy = 0.7002\n",
      "Epoch 182: train accuracy = 0.7066\n",
      "Epoch 183: train accuracy = 0.7079\n",
      "Epoch 184: train accuracy = 0.7074\n",
      "Epoch 185: train accuracy = 0.7209\n",
      "Epoch 186: train accuracy = 0.7188\n",
      "Epoch 187: train accuracy = 0.6979\n",
      "Epoch 188: train accuracy = 0.7102\n",
      "Epoch 189: train accuracy = 0.6969\n",
      "Epoch 190: train accuracy = 0.6982\n",
      "Epoch 191: train accuracy = 0.7219\n",
      "Epoch 192: train accuracy = 0.7185\n",
      "Epoch 193: train accuracy = 0.6891\n",
      "Epoch 194: train accuracy = 0.7149\n",
      "Epoch 195: train accuracy = 0.7135\n",
      "Epoch 196: train accuracy = 0.7238\n",
      "Epoch 197: train accuracy = 0.7245\n",
      "Epoch 198: train accuracy = 0.6896\n",
      "Epoch 199: train accuracy = 0.7212\n"
     ]
    }
   ],
   "source": [
    "loss = 0\n",
    "batch_size = 500\n",
    "for epoch in range(200):\n",
    "    i_ = np.random.choice(len(train_x), batch_size, replace=False)\n",
    "    l = [chainer.Variable(train_x[i, :, :]) for i in i_]\n",
    "    y = net(l, train=True)\n",
    "    loss = sum(F.softmax_cross_entropy(iy, np.repeat(it, 28, 0)) for iy, it in zip(y, list(train_t[i_])))\n",
    "    acc = np.mean([F.accuracy(iy, np.repeat(it, 28, 0)).data for iy, it in zip(y, list(train_t[i_]))])\n",
    "    net.cleargrads()\n",
    "    loss.backward()\n",
    "    optim.update()\n",
    "    print('Epoch ' + str(epoch) + ': train accuracy = ' + str(round(acc, 4)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# GRU using output at last time point\n",
    "\n",
    "Calculate GRU output per time point (1 to 28, width) using features (1 to 28, height). Take the last time point and map to softmax layer. Get loss for that last time point, backpropagate."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 225,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "class NstepGRU_last(chainer.ChainList):\n",
    "    def __init__(self, depth, n_input, n_hidden, n_output):\n",
    "        links = []\n",
    "        links.append(L.NStepGRU(n_layers=depth, in_size=n_input, out_size=n_hidden, dropout=0.1, use_cudnn=False))\n",
    "        links.append(L.Linear(None, n_output))\n",
    "        super(NstepGRU_last, self).__init__(*links)\n",
    "\n",
    "    def __call__(self, x, train):\n",
    "        hy, ys = self[0](hx=None, xs=x, train=train)\n",
    "        z = F.get_item(hy, -1)\n",
    "        y = self[1](z)\n",
    "        return y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 314,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# initialize CNN and optimizer\n",
    "net = NstepGRU_last(depth=2, n_input=28, n_hidden=50, n_output=10)\n",
    "optim = chainer.optimizers.Adam(alpha=1e-03)\n",
    "optim.setup(net)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 315,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 0: train accuracy = 0.08\n",
      "Epoch 1: train accuracy = 0.12\n",
      "Epoch 2: train accuracy = 0.165\n",
      "Epoch 3: train accuracy = 0.23\n",
      "Epoch 4: train accuracy = 0.165\n",
      "Epoch 5: train accuracy = 0.27\n",
      "Epoch 6: train accuracy = 0.255\n",
      "Epoch 7: train accuracy = 0.275\n",
      "Epoch 8: train accuracy = 0.175\n",
      "Epoch 9: train accuracy = 0.305\n",
      "Epoch 10: train accuracy = 0.305\n",
      "Epoch 11: train accuracy = 0.285\n",
      "Epoch 12: train accuracy = 0.335\n",
      "Epoch 13: train accuracy = 0.32\n",
      "Epoch 14: train accuracy = 0.33\n",
      "Epoch 15: train accuracy = 0.32\n",
      "Epoch 16: train accuracy = 0.295\n",
      "Epoch 17: train accuracy = 0.33\n",
      "Epoch 18: train accuracy = 0.37\n",
      "Epoch 19: train accuracy = 0.365\n",
      "Epoch 20: train accuracy = 0.315\n",
      "Epoch 21: train accuracy = 0.37\n",
      "Epoch 22: train accuracy = 0.31\n",
      "Epoch 23: train accuracy = 0.32\n",
      "Epoch 24: train accuracy = 0.28\n",
      "Epoch 25: train accuracy = 0.275\n",
      "Epoch 26: train accuracy = 0.32\n",
      "Epoch 27: train accuracy = 0.36\n",
      "Epoch 28: train accuracy = 0.415\n",
      "Epoch 29: train accuracy = 0.4\n",
      "Epoch 30: train accuracy = 0.36\n",
      "Epoch 31: train accuracy = 0.365\n",
      "Epoch 32: train accuracy = 0.465\n",
      "Epoch 33: train accuracy = 0.43\n",
      "Epoch 34: train accuracy = 0.42\n",
      "Epoch 35: train accuracy = 0.52\n",
      "Epoch 36: train accuracy = 0.44\n",
      "Epoch 37: train accuracy = 0.515\n",
      "Epoch 38: train accuracy = 0.44\n",
      "Epoch 39: train accuracy = 0.515\n",
      "Epoch 40: train accuracy = 0.44\n",
      "Epoch 41: train accuracy = 0.445\n",
      "Epoch 42: train accuracy = 0.505\n",
      "Epoch 43: train accuracy = 0.495\n",
      "Epoch 44: train accuracy = 0.54\n",
      "Epoch 45: train accuracy = 0.445\n",
      "Epoch 46: train accuracy = 0.5\n",
      "Epoch 47: train accuracy = 0.5\n",
      "Epoch 48: train accuracy = 0.455\n",
      "Epoch 49: train accuracy = 0.415\n",
      "Epoch 50: train accuracy = 0.52\n",
      "Epoch 51: train accuracy = 0.66\n",
      "Epoch 52: train accuracy = 0.52\n",
      "Epoch 53: train accuracy = 0.545\n",
      "Epoch 54: train accuracy = 0.505\n",
      "Epoch 55: train accuracy = 0.525\n",
      "Epoch 56: train accuracy = 0.525\n",
      "Epoch 57: train accuracy = 0.64\n",
      "Epoch 58: train accuracy = 0.57\n",
      "Epoch 59: train accuracy = 0.585\n",
      "Epoch 60: train accuracy = 0.57\n",
      "Epoch 61: train accuracy = 0.6\n",
      "Epoch 62: train accuracy = 0.56\n",
      "Epoch 63: train accuracy = 0.555\n",
      "Epoch 64: train accuracy = 0.58\n",
      "Epoch 65: train accuracy = 0.615\n",
      "Epoch 66: train accuracy = 0.65\n",
      "Epoch 67: train accuracy = 0.61\n",
      "Epoch 68: train accuracy = 0.615\n",
      "Epoch 69: train accuracy = 0.615\n",
      "Epoch 70: train accuracy = 0.595\n",
      "Epoch 71: train accuracy = 0.585\n",
      "Epoch 72: train accuracy = 0.645\n",
      "Epoch 73: train accuracy = 0.65\n",
      "Epoch 74: train accuracy = 0.645\n",
      "Epoch 75: train accuracy = 0.665\n",
      "Epoch 76: train accuracy = 0.68\n",
      "Epoch 77: train accuracy = 0.68\n",
      "Epoch 78: train accuracy = 0.685\n",
      "Epoch 79: train accuracy = 0.695\n",
      "Epoch 80: train accuracy = 0.645\n",
      "Epoch 81: train accuracy = 0.685\n",
      "Epoch 82: train accuracy = 0.67\n",
      "Epoch 83: train accuracy = 0.65\n",
      "Epoch 84: train accuracy = 0.67\n",
      "Epoch 85: train accuracy = 0.705\n",
      "Epoch 86: train accuracy = 0.665\n",
      "Epoch 87: train accuracy = 0.675\n",
      "Epoch 88: train accuracy = 0.71\n",
      "Epoch 89: train accuracy = 0.68\n",
      "Epoch 90: train accuracy = 0.65\n",
      "Epoch 91: train accuracy = 0.67\n",
      "Epoch 92: train accuracy = 0.76\n",
      "Epoch 93: train accuracy = 0.785\n",
      "Epoch 94: train accuracy = 0.645\n",
      "Epoch 95: train accuracy = 0.71\n",
      "Epoch 96: train accuracy = 0.73\n",
      "Epoch 97: train accuracy = 0.7\n",
      "Epoch 98: train accuracy = 0.705\n",
      "Epoch 99: train accuracy = 0.745\n",
      "Epoch 100: train accuracy = 0.78\n",
      "Epoch 101: train accuracy = 0.73\n",
      "Epoch 102: train accuracy = 0.685\n",
      "Epoch 103: train accuracy = 0.7\n",
      "Epoch 104: train accuracy = 0.745\n",
      "Epoch 105: train accuracy = 0.765\n",
      "Epoch 106: train accuracy = 0.75\n",
      "Epoch 107: train accuracy = 0.765\n",
      "Epoch 108: train accuracy = 0.74\n",
      "Epoch 109: train accuracy = 0.76\n",
      "Epoch 110: train accuracy = 0.75\n",
      "Epoch 111: train accuracy = 0.775\n",
      "Epoch 112: train accuracy = 0.71\n",
      "Epoch 113: train accuracy = 0.7\n",
      "Epoch 114: train accuracy = 0.78\n",
      "Epoch 115: train accuracy = 0.8\n",
      "Epoch 116: train accuracy = 0.81\n",
      "Epoch 117: train accuracy = 0.765\n",
      "Epoch 118: train accuracy = 0.725\n",
      "Epoch 119: train accuracy = 0.75\n",
      "Epoch 120: train accuracy = 0.79\n",
      "Epoch 121: train accuracy = 0.805\n",
      "Epoch 122: train accuracy = 0.825\n",
      "Epoch 123: train accuracy = 0.8\n",
      "Epoch 124: train accuracy = 0.79\n",
      "Epoch 125: train accuracy = 0.81\n",
      "Epoch 126: train accuracy = 0.785\n",
      "Epoch 127: train accuracy = 0.765\n",
      "Epoch 128: train accuracy = 0.76\n",
      "Epoch 129: train accuracy = 0.765\n",
      "Epoch 130: train accuracy = 0.855\n",
      "Epoch 131: train accuracy = 0.805\n",
      "Epoch 132: train accuracy = 0.845\n",
      "Epoch 133: train accuracy = 0.85\n",
      "Epoch 134: train accuracy = 0.735\n",
      "Epoch 135: train accuracy = 0.82\n",
      "Epoch 136: train accuracy = 0.8\n",
      "Epoch 137: train accuracy = 0.83\n",
      "Epoch 138: train accuracy = 0.805\n",
      "Epoch 139: train accuracy = 0.86\n",
      "Epoch 140: train accuracy = 0.85\n",
      "Epoch 141: train accuracy = 0.8\n",
      "Epoch 142: train accuracy = 0.865\n",
      "Epoch 143: train accuracy = 0.845\n",
      "Epoch 144: train accuracy = 0.84\n",
      "Epoch 145: train accuracy = 0.825\n",
      "Epoch 146: train accuracy = 0.835\n",
      "Epoch 147: train accuracy = 0.825\n",
      "Epoch 148: train accuracy = 0.805\n",
      "Epoch 149: train accuracy = 0.875\n",
      "Epoch 150: train accuracy = 0.85\n",
      "Epoch 151: train accuracy = 0.845\n",
      "Epoch 152: train accuracy = 0.865\n",
      "Epoch 153: train accuracy = 0.875\n",
      "Epoch 154: train accuracy = 0.82\n",
      "Epoch 155: train accuracy = 0.81\n",
      "Epoch 156: train accuracy = 0.8\n",
      "Epoch 157: train accuracy = 0.855\n",
      "Epoch 158: train accuracy = 0.89\n",
      "Epoch 159: train accuracy = 0.9\n",
      "Epoch 160: train accuracy = 0.88\n",
      "Epoch 161: train accuracy = 0.83\n",
      "Epoch 162: train accuracy = 0.875\n",
      "Epoch 163: train accuracy = 0.86\n",
      "Epoch 164: train accuracy = 0.78\n",
      "Epoch 165: train accuracy = 0.87\n",
      "Epoch 166: train accuracy = 0.86\n",
      "Epoch 167: train accuracy = 0.865\n",
      "Epoch 168: train accuracy = 0.88\n",
      "Epoch 169: train accuracy = 0.855\n",
      "Epoch 170: train accuracy = 0.845\n",
      "Epoch 171: train accuracy = 0.815\n",
      "Epoch 172: train accuracy = 0.865\n",
      "Epoch 173: train accuracy = 0.85\n",
      "Epoch 174: train accuracy = 0.835\n",
      "Epoch 175: train accuracy = 0.845\n",
      "Epoch 176: train accuracy = 0.88\n",
      "Epoch 177: train accuracy = 0.885\n",
      "Epoch 178: train accuracy = 0.865\n",
      "Epoch 179: train accuracy = 0.87\n",
      "Epoch 180: train accuracy = 0.835\n",
      "Epoch 181: train accuracy = 0.88\n",
      "Epoch 182: train accuracy = 0.825\n",
      "Epoch 183: train accuracy = 0.89\n",
      "Epoch 184: train accuracy = 0.87\n",
      "Epoch 185: train accuracy = 0.91\n",
      "Epoch 186: train accuracy = 0.85\n",
      "Epoch 187: train accuracy = 0.855\n",
      "Epoch 188: train accuracy = 0.895\n",
      "Epoch 189: train accuracy = 0.92\n",
      "Epoch 190: train accuracy = 0.9\n",
      "Epoch 191: train accuracy = 0.925\n",
      "Epoch 192: train accuracy = 0.885\n",
      "Epoch 193: train accuracy = 0.905\n",
      "Epoch 194: train accuracy = 0.89\n",
      "Epoch 195: train accuracy = 0.875\n",
      "Epoch 196: train accuracy = 0.875\n",
      "Epoch 197: train accuracy = 0.92\n",
      "Epoch 198: train accuracy = 0.865\n",
      "Epoch 199: train accuracy = 0.88\n",
      "Epoch 200: train accuracy = 0.87\n",
      "Epoch 201: train accuracy = 0.885\n",
      "Epoch 202: train accuracy = 0.895\n",
      "Epoch 203: train accuracy = 0.89\n",
      "Epoch 204: train accuracy = 0.905\n",
      "Epoch 205: train accuracy = 0.92\n",
      "Epoch 206: train accuracy = 0.865\n",
      "Epoch 207: train accuracy = 0.875\n",
      "Epoch 208: train accuracy = 0.9\n",
      "Epoch 209: train accuracy = 0.89\n",
      "Epoch 210: train accuracy = 0.885\n",
      "Epoch 211: train accuracy = 0.91\n",
      "Epoch 212: train accuracy = 0.895\n",
      "Epoch 213: train accuracy = 0.91\n",
      "Epoch 214: train accuracy = 0.86\n",
      "Epoch 215: train accuracy = 0.9\n",
      "Epoch 216: train accuracy = 0.915\n",
      "Epoch 217: train accuracy = 0.885\n",
      "Epoch 218: train accuracy = 0.885\n",
      "Epoch 219: train accuracy = 0.88\n",
      "Epoch 220: train accuracy = 0.905\n",
      "Epoch 221: train accuracy = 0.89\n",
      "Epoch 222: train accuracy = 0.91\n",
      "Epoch 223: train accuracy = 0.88\n",
      "Epoch 224: train accuracy = 0.91\n",
      "Epoch 225: train accuracy = 0.91\n",
      "Epoch 226: train accuracy = 0.865\n",
      "Epoch 227: train accuracy = 0.9\n",
      "Epoch 228: train accuracy = 0.905\n",
      "Epoch 229: train accuracy = 0.875\n",
      "Epoch 230: train accuracy = 0.905\n",
      "Epoch 231: train accuracy = 0.915\n",
      "Epoch 232: train accuracy = 0.93\n",
      "Epoch 233: train accuracy = 0.87\n",
      "Epoch 234: train accuracy = 0.865\n",
      "Epoch 235: train accuracy = 0.945\n",
      "Epoch 236: train accuracy = 0.925\n",
      "Epoch 237: train accuracy = 0.94\n",
      "Epoch 238: train accuracy = 0.905\n",
      "Epoch 239: train accuracy = 0.88\n",
      "Epoch 240: train accuracy = 0.875\n",
      "Epoch 241: train accuracy = 0.895\n",
      "Epoch 242: train accuracy = 0.91\n",
      "Epoch 243: train accuracy = 0.915\n",
      "Epoch 244: train accuracy = 0.95\n",
      "Epoch 245: train accuracy = 0.895\n",
      "Epoch 246: train accuracy = 0.92\n",
      "Epoch 247: train accuracy = 0.89\n",
      "Epoch 248: train accuracy = 0.925\n",
      "Epoch 249: train accuracy = 0.89\n",
      "Epoch 250: train accuracy = 0.93\n",
      "Epoch 251: train accuracy = 0.9\n",
      "Epoch 252: train accuracy = 0.945\n",
      "Epoch 253: train accuracy = 0.955\n",
      "Epoch 254: train accuracy = 0.905\n",
      "Epoch 255: train accuracy = 0.95\n",
      "Epoch 256: train accuracy = 0.93\n",
      "Epoch 257: train accuracy = 0.925\n",
      "Epoch 258: train accuracy = 0.93\n",
      "Epoch 259: train accuracy = 0.86\n",
      "Epoch 260: train accuracy = 0.92\n",
      "Epoch 261: train accuracy = 0.87\n",
      "Epoch 262: train accuracy = 0.88\n",
      "Epoch 263: train accuracy = 0.905\n",
      "Epoch 264: train accuracy = 0.925\n",
      "Epoch 265: train accuracy = 0.91\n",
      "Epoch 266: train accuracy = 0.9\n",
      "Epoch 267: train accuracy = 0.955\n",
      "Epoch 268: train accuracy = 0.915\n",
      "Epoch 269: train accuracy = 0.94\n",
      "Epoch 270: train accuracy = 0.92\n",
      "Epoch 271: train accuracy = 0.9\n",
      "Epoch 272: train accuracy = 0.94\n",
      "Epoch 273: train accuracy = 0.935\n",
      "Epoch 274: train accuracy = 0.91\n",
      "Epoch 275: train accuracy = 0.92\n",
      "Epoch 276: train accuracy = 0.94\n",
      "Epoch 277: train accuracy = 0.89\n",
      "Epoch 278: train accuracy = 0.915\n",
      "Epoch 279: train accuracy = 0.94\n",
      "Epoch 280: train accuracy = 0.93\n",
      "Epoch 281: train accuracy = 0.895\n",
      "Epoch 282: train accuracy = 0.945\n",
      "Epoch 283: train accuracy = 0.92\n",
      "Epoch 284: train accuracy = 0.945\n",
      "Epoch 285: train accuracy = 0.935\n",
      "Epoch 286: train accuracy = 0.93\n",
      "Epoch 287: train accuracy = 0.895\n",
      "Epoch 288: train accuracy = 0.97\n",
      "Epoch 289: train accuracy = 0.9\n",
      "Epoch 290: train accuracy = 0.915\n",
      "Epoch 291: train accuracy = 0.92\n",
      "Epoch 292: train accuracy = 0.915\n",
      "Epoch 293: train accuracy = 0.95\n",
      "Epoch 294: train accuracy = 0.905\n",
      "Epoch 295: train accuracy = 0.92\n",
      "Epoch 296: train accuracy = 0.955\n",
      "Epoch 297: train accuracy = 0.92\n",
      "Epoch 298: train accuracy = 0.94\n",
      "Epoch 299: train accuracy = 0.905\n",
      "Epoch 300: train accuracy = 0.905\n",
      "Epoch 301: train accuracy = 0.935\n",
      "Epoch 302: train accuracy = 0.92\n",
      "Epoch 303: train accuracy = 0.945\n",
      "Epoch 304: train accuracy = 0.945\n",
      "Epoch 305: train accuracy = 0.91\n",
      "Epoch 306: train accuracy = 0.91\n",
      "Epoch 307: train accuracy = 0.925\n",
      "Epoch 308: train accuracy = 0.935\n",
      "Epoch 309: train accuracy = 0.885\n",
      "Epoch 310: train accuracy = 0.94\n",
      "Epoch 311: train accuracy = 0.945\n",
      "Epoch 312: train accuracy = 0.91\n",
      "Epoch 313: train accuracy = 0.915\n",
      "Epoch 314: train accuracy = 0.885\n",
      "Epoch 315: train accuracy = 0.945\n",
      "Epoch 316: train accuracy = 0.94\n",
      "Epoch 317: train accuracy = 0.945\n",
      "Epoch 318: train accuracy = 0.955\n",
      "Epoch 319: train accuracy = 0.96\n",
      "Epoch 320: train accuracy = 0.925\n",
      "Epoch 321: train accuracy = 0.92\n",
      "Epoch 322: train accuracy = 0.905\n",
      "Epoch 323: train accuracy = 0.925\n",
      "Epoch 324: train accuracy = 0.92\n",
      "Epoch 325: train accuracy = 0.96\n",
      "Epoch 326: train accuracy = 0.945\n",
      "Epoch 327: train accuracy = 0.93\n",
      "Epoch 328: train accuracy = 0.93\n",
      "Epoch 329: train accuracy = 0.935\n",
      "Epoch 330: train accuracy = 0.955\n",
      "Epoch 331: train accuracy = 0.95\n",
      "Epoch 332: train accuracy = 0.935\n",
      "Epoch 333: train accuracy = 0.93\n",
      "Epoch 334: train accuracy = 0.955\n",
      "Epoch 335: train accuracy = 0.955\n",
      "Epoch 336: train accuracy = 0.94\n",
      "Epoch 337: train accuracy = 0.94\n",
      "Epoch 338: train accuracy = 0.955\n",
      "Epoch 339: train accuracy = 0.97\n",
      "Epoch 340: train accuracy = 0.96\n",
      "Epoch 341: train accuracy = 0.94\n",
      "Epoch 342: train accuracy = 0.935\n",
      "Epoch 343: train accuracy = 0.94\n",
      "Epoch 344: train accuracy = 0.925\n",
      "Epoch 345: train accuracy = 0.96\n",
      "Epoch 346: train accuracy = 0.94\n",
      "Epoch 347: train accuracy = 0.94\n",
      "Epoch 348: train accuracy = 0.93\n",
      "Epoch 349: train accuracy = 0.93\n",
      "Epoch 350: train accuracy = 0.945\n",
      "Epoch 351: train accuracy = 0.93\n",
      "Epoch 352: train accuracy = 0.93\n",
      "Epoch 353: train accuracy = 0.955\n",
      "Epoch 354: train accuracy = 0.96\n",
      "Epoch 355: train accuracy = 0.965\n",
      "Epoch 356: train accuracy = 0.93\n",
      "Epoch 357: train accuracy = 0.94\n",
      "Epoch 358: train accuracy = 0.96\n",
      "Epoch 359: train accuracy = 0.935\n",
      "Epoch 360: train accuracy = 0.925\n",
      "Epoch 361: train accuracy = 0.92\n",
      "Epoch 362: train accuracy = 0.94\n",
      "Epoch 363: train accuracy = 0.905\n",
      "Epoch 364: train accuracy = 0.945\n",
      "Epoch 365: train accuracy = 0.955\n",
      "Epoch 366: train accuracy = 0.96\n",
      "Epoch 367: train accuracy = 0.945\n",
      "Epoch 368: train accuracy = 0.92\n",
      "Epoch 369: train accuracy = 0.94\n",
      "Epoch 370: train accuracy = 0.95\n",
      "Epoch 371: train accuracy = 0.955\n",
      "Epoch 372: train accuracy = 0.955\n",
      "Epoch 373: train accuracy = 0.935\n",
      "Epoch 374: train accuracy = 0.95\n",
      "Epoch 375: train accuracy = 0.905\n",
      "Epoch 376: train accuracy = 0.95\n",
      "Epoch 377: train accuracy = 0.93\n",
      "Epoch 378: train accuracy = 0.92\n",
      "Epoch 379: train accuracy = 0.935\n",
      "Epoch 380: train accuracy = 0.935\n",
      "Epoch 381: train accuracy = 0.96\n",
      "Epoch 382: train accuracy = 0.94\n",
      "Epoch 383: train accuracy = 0.955\n",
      "Epoch 384: train accuracy = 0.965\n",
      "Epoch 385: train accuracy = 0.955\n",
      "Epoch 386: train accuracy = 0.935\n",
      "Epoch 387: train accuracy = 0.93\n",
      "Epoch 388: train accuracy = 0.935\n",
      "Epoch 389: train accuracy = 0.91\n",
      "Epoch 390: train accuracy = 0.96\n",
      "Epoch 391: train accuracy = 0.935\n",
      "Epoch 392: train accuracy = 0.95\n",
      "Epoch 393: train accuracy = 0.925\n",
      "Epoch 394: train accuracy = 0.94\n",
      "Epoch 395: train accuracy = 0.925\n",
      "Epoch 396: train accuracy = 0.96\n",
      "Epoch 397: train accuracy = 0.925\n",
      "Epoch 398: train accuracy = 0.925\n",
      "Epoch 399: train accuracy = 0.975\n",
      "Epoch 400: train accuracy = 0.92\n",
      "Epoch 401: train accuracy = 0.96\n",
      "Epoch 402: train accuracy = 0.935\n",
      "Epoch 403: train accuracy = 0.94\n",
      "Epoch 404: train accuracy = 0.93\n",
      "Epoch 405: train accuracy = 0.96\n",
      "Epoch 406: train accuracy = 0.975\n",
      "Epoch 407: train accuracy = 0.935\n",
      "Epoch 408: train accuracy = 0.95\n",
      "Epoch 409: train accuracy = 0.975\n",
      "Epoch 410: train accuracy = 0.96\n",
      "Epoch 411: train accuracy = 0.92\n",
      "Epoch 412: train accuracy = 0.955\n",
      "Epoch 413: train accuracy = 0.955\n",
      "Epoch 414: train accuracy = 0.93\n",
      "Epoch 415: train accuracy = 0.97\n",
      "Epoch 416: train accuracy = 0.96\n",
      "Epoch 417: train accuracy = 0.955\n",
      "Epoch 418: train accuracy = 0.95\n",
      "Epoch 419: train accuracy = 0.955\n",
      "Epoch 420: train accuracy = 0.94\n",
      "Epoch 421: train accuracy = 0.945\n",
      "Epoch 422: train accuracy = 0.975\n",
      "Epoch 423: train accuracy = 0.92\n",
      "Epoch 424: train accuracy = 0.95\n",
      "Epoch 425: train accuracy = 0.93\n",
      "Epoch 426: train accuracy = 0.97\n",
      "Epoch 427: train accuracy = 0.93\n",
      "Epoch 428: train accuracy = 0.96\n",
      "Epoch 429: train accuracy = 0.935\n",
      "Epoch 430: train accuracy = 0.925\n",
      "Epoch 431: train accuracy = 0.94\n",
      "Epoch 432: train accuracy = 0.945\n",
      "Epoch 433: train accuracy = 0.97\n",
      "Epoch 434: train accuracy = 0.955\n",
      "Epoch 435: train accuracy = 0.94\n",
      "Epoch 436: train accuracy = 0.97\n",
      "Epoch 437: train accuracy = 0.975\n",
      "Epoch 438: train accuracy = 0.975\n",
      "Epoch 439: train accuracy = 0.94\n",
      "Epoch 440: train accuracy = 0.925\n",
      "Epoch 441: train accuracy = 0.955\n",
      "Epoch 442: train accuracy = 0.96\n",
      "Epoch 443: train accuracy = 0.945\n",
      "Epoch 444: train accuracy = 0.92\n",
      "Epoch 445: train accuracy = 0.95\n",
      "Epoch 446: train accuracy = 0.945\n",
      "Epoch 447: train accuracy = 0.935\n",
      "Epoch 448: train accuracy = 0.955\n",
      "Epoch 449: train accuracy = 0.94\n",
      "Epoch 450: train accuracy = 0.95\n",
      "Epoch 451: train accuracy = 0.975\n",
      "Epoch 452: train accuracy = 0.955\n",
      "Epoch 453: train accuracy = 0.945\n",
      "Epoch 454: train accuracy = 0.935\n",
      "Epoch 455: train accuracy = 0.96\n",
      "Epoch 456: train accuracy = 0.94\n",
      "Epoch 457: train accuracy = 0.98\n",
      "Epoch 458: train accuracy = 0.93\n",
      "Epoch 459: train accuracy = 0.945\n",
      "Epoch 460: train accuracy = 0.925\n",
      "Epoch 461: train accuracy = 0.945\n",
      "Epoch 462: train accuracy = 0.97\n",
      "Epoch 463: train accuracy = 0.95\n",
      "Epoch 464: train accuracy = 0.94\n",
      "Epoch 465: train accuracy = 0.94\n",
      "Epoch 466: train accuracy = 0.97\n",
      "Epoch 467: train accuracy = 0.935\n",
      "Epoch 468: train accuracy = 0.945\n",
      "Epoch 469: train accuracy = 0.94\n",
      "Epoch 470: train accuracy = 0.95\n",
      "Epoch 471: train accuracy = 0.935\n",
      "Epoch 472: train accuracy = 0.955\n",
      "Epoch 473: train accuracy = 0.925\n",
      "Epoch 474: train accuracy = 0.965\n",
      "Epoch 475: train accuracy = 0.93\n",
      "Epoch 476: train accuracy = 0.955\n",
      "Epoch 477: train accuracy = 0.96\n",
      "Epoch 478: train accuracy = 0.955\n",
      "Epoch 479: train accuracy = 0.95\n",
      "Epoch 480: train accuracy = 0.965\n",
      "Epoch 481: train accuracy = 0.935\n",
      "Epoch 482: train accuracy = 0.955\n",
      "Epoch 483: train accuracy = 0.945\n",
      "Epoch 484: train accuracy = 0.96\n",
      "Epoch 485: train accuracy = 0.94\n",
      "Epoch 486: train accuracy = 0.96\n",
      "Epoch 487: train accuracy = 0.965\n",
      "Epoch 488: train accuracy = 0.95\n",
      "Epoch 489: train accuracy = 0.965\n",
      "Epoch 490: train accuracy = 0.925\n",
      "Epoch 491: train accuracy = 0.97\n",
      "Epoch 492: train accuracy = 0.945\n",
      "Epoch 493: train accuracy = 0.945\n",
      "Epoch 494: train accuracy = 0.95\n",
      "Epoch 495: train accuracy = 0.99\n",
      "Epoch 496: train accuracy = 0.97\n",
      "Epoch 497: train accuracy = 0.975\n",
      "Epoch 498: train accuracy = 0.955\n",
      "Epoch 499: train accuracy = 0.95\n",
      "Epoch 500: train accuracy = 0.965\n",
      "Epoch 501: train accuracy = 0.945\n",
      "Epoch 502: train accuracy = 0.95\n",
      "Epoch 503: train accuracy = 0.955\n",
      "Epoch 504: train accuracy = 0.95\n",
      "Epoch 505: train accuracy = 0.95\n",
      "Epoch 506: train accuracy = 0.97\n",
      "Epoch 507: train accuracy = 0.945\n",
      "Epoch 508: train accuracy = 0.965\n",
      "Epoch 509: train accuracy = 0.945\n",
      "Epoch 510: train accuracy = 0.96\n",
      "Epoch 511: train accuracy = 0.945\n",
      "Epoch 512: train accuracy = 0.95\n",
      "Epoch 513: train accuracy = 0.955\n",
      "Epoch 514: train accuracy = 0.95\n",
      "Epoch 515: train accuracy = 0.96\n",
      "Epoch 516: train accuracy = 0.985\n",
      "Epoch 517: train accuracy = 0.95\n",
      "Epoch 518: train accuracy = 0.925\n",
      "Epoch 519: train accuracy = 0.965\n",
      "Epoch 520: train accuracy = 0.945\n",
      "Epoch 521: train accuracy = 0.965\n",
      "Epoch 522: train accuracy = 0.965\n",
      "Epoch 523: train accuracy = 0.97\n",
      "Epoch 524: train accuracy = 0.96\n",
      "Epoch 525: train accuracy = 0.935\n",
      "Epoch 526: train accuracy = 0.965\n",
      "Epoch 527: train accuracy = 0.935\n",
      "Epoch 528: train accuracy = 0.95\n",
      "Epoch 529: train accuracy = 0.96\n",
      "Epoch 530: train accuracy = 0.96\n",
      "Epoch 531: train accuracy = 0.99\n",
      "Epoch 532: train accuracy = 0.96\n",
      "Epoch 533: train accuracy = 0.96\n",
      "Epoch 534: train accuracy = 0.92\n",
      "Epoch 535: train accuracy = 0.97\n",
      "Epoch 536: train accuracy = 0.945\n",
      "Epoch 537: train accuracy = 0.95\n",
      "Epoch 538: train accuracy = 0.95\n",
      "Epoch 539: train accuracy = 0.955\n",
      "Epoch 540: train accuracy = 0.965\n",
      "Epoch 541: train accuracy = 0.94\n",
      "Epoch 542: train accuracy = 0.945\n",
      "Epoch 543: train accuracy = 0.93\n",
      "Epoch 544: train accuracy = 0.97\n",
      "Epoch 545: train accuracy = 0.955\n",
      "Epoch 546: train accuracy = 0.985\n",
      "Epoch 547: train accuracy = 0.95\n",
      "Epoch 548: train accuracy = 0.955\n",
      "Epoch 549: train accuracy = 0.94\n",
      "Epoch 550: train accuracy = 0.95\n",
      "Epoch 551: train accuracy = 0.985\n",
      "Epoch 552: train accuracy = 0.975\n",
      "Epoch 553: train accuracy = 0.955\n",
      "Epoch 554: train accuracy = 0.965\n",
      "Epoch 555: train accuracy = 0.97\n",
      "Epoch 556: train accuracy = 0.945\n",
      "Epoch 557: train accuracy = 0.965\n",
      "Epoch 558: train accuracy = 0.94\n",
      "Epoch 559: train accuracy = 0.935\n",
      "Epoch 560: train accuracy = 0.96\n",
      "Epoch 561: train accuracy = 0.96\n",
      "Epoch 562: train accuracy = 0.95\n",
      "Epoch 563: train accuracy = 0.935\n",
      "Epoch 564: train accuracy = 0.97\n",
      "Epoch 565: train accuracy = 0.975\n",
      "Epoch 566: train accuracy = 0.97\n",
      "Epoch 567: train accuracy = 0.94\n",
      "Epoch 568: train accuracy = 0.955\n",
      "Epoch 569: train accuracy = 0.965\n",
      "Epoch 570: train accuracy = 0.955\n",
      "Epoch 571: train accuracy = 0.955\n",
      "Epoch 572: train accuracy = 0.96\n",
      "Epoch 573: train accuracy = 0.97\n",
      "Epoch 574: train accuracy = 0.995\n",
      "Epoch 575: train accuracy = 0.96\n",
      "Epoch 576: train accuracy = 0.97\n",
      "Epoch 577: train accuracy = 0.97\n",
      "Epoch 578: train accuracy = 0.965\n",
      "Epoch 579: train accuracy = 0.955\n",
      "Epoch 580: train accuracy = 0.97\n",
      "Epoch 581: train accuracy = 0.965\n",
      "Epoch 582: train accuracy = 0.98\n",
      "Epoch 583: train accuracy = 0.965\n",
      "Epoch 584: train accuracy = 0.955\n",
      "Epoch 585: train accuracy = 0.97\n",
      "Epoch 586: train accuracy = 0.95\n",
      "Epoch 587: train accuracy = 0.945\n",
      "Epoch 588: train accuracy = 0.955\n",
      "Epoch 589: train accuracy = 0.95\n",
      "Epoch 590: train accuracy = 0.985\n",
      "Epoch 591: train accuracy = 0.97\n",
      "Epoch 592: train accuracy = 0.965\n",
      "Epoch 593: train accuracy = 0.97\n",
      "Epoch 594: train accuracy = 0.965\n",
      "Epoch 595: train accuracy = 0.99\n",
      "Epoch 596: train accuracy = 0.925\n",
      "Epoch 597: train accuracy = 0.955\n",
      "Epoch 598: train accuracy = 0.96\n",
      "Epoch 599: train accuracy = 0.955\n",
      "Epoch 600: train accuracy = 0.975\n",
      "Epoch 601: train accuracy = 0.97\n",
      "Epoch 602: train accuracy = 0.97\n",
      "Epoch 603: train accuracy = 0.965\n",
      "Epoch 604: train accuracy = 0.985\n",
      "Epoch 605: train accuracy = 0.935\n",
      "Epoch 606: train accuracy = 0.94\n",
      "Epoch 607: train accuracy = 0.96\n",
      "Epoch 608: train accuracy = 0.925\n",
      "Epoch 609: train accuracy = 0.955\n",
      "Epoch 610: train accuracy = 0.95\n",
      "Epoch 611: train accuracy = 0.975\n",
      "Epoch 612: train accuracy = 0.955\n",
      "Epoch 613: train accuracy = 0.935\n",
      "Epoch 614: train accuracy = 0.955\n",
      "Epoch 615: train accuracy = 0.925\n",
      "Epoch 616: train accuracy = 0.935\n",
      "Epoch 617: train accuracy = 0.945\n",
      "Epoch 618: train accuracy = 0.97\n",
      "Epoch 619: train accuracy = 0.945\n",
      "Epoch 620: train accuracy = 0.965\n",
      "Epoch 621: train accuracy = 0.97\n",
      "Epoch 622: train accuracy = 0.97\n",
      "Epoch 623: train accuracy = 0.95\n",
      "Epoch 624: train accuracy = 0.955\n",
      "Epoch 625: train accuracy = 0.955\n",
      "Epoch 626: train accuracy = 0.975\n",
      "Epoch 627: train accuracy = 0.94\n",
      "Epoch 628: train accuracy = 0.975\n",
      "Epoch 629: train accuracy = 0.965\n",
      "Epoch 630: train accuracy = 0.975\n",
      "Epoch 631: train accuracy = 0.97\n",
      "Epoch 632: train accuracy = 0.96\n",
      "Epoch 633: train accuracy = 0.945\n",
      "Epoch 634: train accuracy = 0.94\n",
      "Epoch 635: train accuracy = 0.95\n",
      "Epoch 636: train accuracy = 0.96\n",
      "Epoch 637: train accuracy = 0.96\n",
      "Epoch 638: train accuracy = 0.955\n",
      "Epoch 639: train accuracy = 0.965\n",
      "Epoch 640: train accuracy = 0.94\n",
      "Epoch 641: train accuracy = 0.935\n",
      "Epoch 642: train accuracy = 0.94\n",
      "Epoch 643: train accuracy = 0.96\n",
      "Epoch 644: train accuracy = 0.96\n",
      "Epoch 645: train accuracy = 0.935\n",
      "Epoch 646: train accuracy = 0.96\n",
      "Epoch 647: train accuracy = 0.93\n",
      "Epoch 648: train accuracy = 0.935\n",
      "Epoch 649: train accuracy = 0.95\n",
      "Epoch 650: train accuracy = 0.95\n",
      "Epoch 651: train accuracy = 0.985\n",
      "Epoch 652: train accuracy = 0.965\n",
      "Epoch 653: train accuracy = 0.96\n",
      "Epoch 654: train accuracy = 0.95\n",
      "Epoch 655: train accuracy = 0.965\n",
      "Epoch 656: train accuracy = 0.95\n",
      "Epoch 657: train accuracy = 0.955\n",
      "Epoch 658: train accuracy = 0.975\n",
      "Epoch 659: train accuracy = 0.975\n",
      "Epoch 660: train accuracy = 0.965\n",
      "Epoch 661: train accuracy = 0.965\n",
      "Epoch 662: train accuracy = 0.955\n",
      "Epoch 663: train accuracy = 0.955\n",
      "Epoch 664: train accuracy = 0.935\n",
      "Epoch 665: train accuracy = 0.96\n",
      "Epoch 666: train accuracy = 0.945\n",
      "Epoch 667: train accuracy = 0.965\n",
      "Epoch 668: train accuracy = 0.955\n",
      "Epoch 669: train accuracy = 0.955\n",
      "Epoch 670: train accuracy = 0.95\n",
      "Epoch 671: train accuracy = 0.96\n",
      "Epoch 672: train accuracy = 0.965\n",
      "Epoch 673: train accuracy = 0.98\n",
      "Epoch 674: train accuracy = 0.96\n",
      "Epoch 675: train accuracy = 0.96\n",
      "Epoch 676: train accuracy = 0.96\n",
      "Epoch 677: train accuracy = 0.965\n",
      "Epoch 678: train accuracy = 0.945\n",
      "Epoch 679: train accuracy = 0.97\n",
      "Epoch 680: train accuracy = 0.955\n",
      "Epoch 681: train accuracy = 0.955\n",
      "Epoch 682: train accuracy = 0.955\n",
      "Epoch 683: train accuracy = 0.965\n",
      "Epoch 684: train accuracy = 0.96\n",
      "Epoch 685: train accuracy = 0.975\n",
      "Epoch 686: train accuracy = 0.97\n",
      "Epoch 687: train accuracy = 0.99\n",
      "Epoch 688: train accuracy = 0.98\n",
      "Epoch 689: train accuracy = 0.975\n",
      "Epoch 690: train accuracy = 0.97\n",
      "Epoch 691: train accuracy = 0.96\n",
      "Epoch 692: train accuracy = 0.96\n",
      "Epoch 693: train accuracy = 0.985\n",
      "Epoch 694: train accuracy = 0.975\n",
      "Epoch 695: train accuracy = 0.96\n",
      "Epoch 696: train accuracy = 0.975\n",
      "Epoch 697: train accuracy = 0.98\n",
      "Epoch 698: train accuracy = 0.98\n",
      "Epoch 699: train accuracy = 0.95\n",
      "Epoch 700: train accuracy = 0.96\n",
      "Epoch 701: train accuracy = 0.96\n",
      "Epoch 702: train accuracy = 0.98\n",
      "Epoch 703: train accuracy = 0.965\n",
      "Epoch 704: train accuracy = 0.98\n",
      "Epoch 705: train accuracy = 0.985\n",
      "Epoch 706: train accuracy = 0.97\n",
      "Epoch 707: train accuracy = 0.94\n",
      "Epoch 708: train accuracy = 0.97\n",
      "Epoch 709: train accuracy = 0.98\n",
      "Epoch 710: train accuracy = 0.965\n",
      "Epoch 711: train accuracy = 0.96\n",
      "Epoch 712: train accuracy = 0.965\n",
      "Epoch 713: train accuracy = 0.94\n",
      "Epoch 714: train accuracy = 0.965\n",
      "Epoch 715: train accuracy = 0.965\n",
      "Epoch 716: train accuracy = 0.96\n",
      "Epoch 717: train accuracy = 0.96\n",
      "Epoch 718: train accuracy = 0.94\n",
      "Epoch 719: train accuracy = 0.985\n",
      "Epoch 720: train accuracy = 0.97\n",
      "Epoch 721: train accuracy = 0.97\n",
      "Epoch 722: train accuracy = 0.955\n",
      "Epoch 723: train accuracy = 0.955\n",
      "Epoch 724: train accuracy = 0.935\n",
      "Epoch 725: train accuracy = 0.985\n",
      "Epoch 726: train accuracy = 0.98\n",
      "Epoch 727: train accuracy = 0.935\n",
      "Epoch 728: train accuracy = 0.98\n",
      "Epoch 729: train accuracy = 0.935\n",
      "Epoch 730: train accuracy = 0.99\n",
      "Epoch 731: train accuracy = 0.955\n",
      "Epoch 732: train accuracy = 0.985\n",
      "Epoch 733: train accuracy = 0.975\n",
      "Epoch 734: train accuracy = 0.96\n",
      "Epoch 735: train accuracy = 0.935\n",
      "Epoch 736: train accuracy = 0.97\n",
      "Epoch 737: train accuracy = 0.96\n",
      "Epoch 738: train accuracy = 0.955\n",
      "Epoch 739: train accuracy = 0.965\n",
      "Epoch 740: train accuracy = 0.95\n",
      "Epoch 741: train accuracy = 0.97\n",
      "Epoch 742: train accuracy = 0.98\n",
      "Epoch 743: train accuracy = 0.955\n",
      "Epoch 744: train accuracy = 0.965\n",
      "Epoch 745: train accuracy = 0.975\n",
      "Epoch 746: train accuracy = 0.96\n",
      "Epoch 747: train accuracy = 0.97\n",
      "Epoch 748: train accuracy = 0.97\n",
      "Epoch 749: train accuracy = 0.975\n",
      "Epoch 750: train accuracy = 0.96\n",
      "Epoch 751: train accuracy = 0.98\n",
      "Epoch 752: train accuracy = 0.975\n",
      "Epoch 753: train accuracy = 0.99\n",
      "Epoch 754: train accuracy = 0.985\n",
      "Epoch 755: train accuracy = 0.97\n",
      "Epoch 756: train accuracy = 0.975\n",
      "Epoch 757: train accuracy = 0.98\n",
      "Epoch 758: train accuracy = 0.985\n",
      "Epoch 759: train accuracy = 0.965\n",
      "Epoch 760: train accuracy = 0.975\n",
      "Epoch 761: train accuracy = 0.96\n",
      "Epoch 762: train accuracy = 0.96\n",
      "Epoch 763: train accuracy = 0.97\n",
      "Epoch 764: train accuracy = 0.965\n",
      "Epoch 765: train accuracy = 0.975\n",
      "Epoch 766: train accuracy = 0.955\n",
      "Epoch 767: train accuracy = 0.975\n",
      "Epoch 768: train accuracy = 0.94\n",
      "Epoch 769: train accuracy = 0.965\n",
      "Epoch 770: train accuracy = 0.95\n",
      "Epoch 771: train accuracy = 0.96\n",
      "Epoch 772: train accuracy = 0.96\n",
      "Epoch 773: train accuracy = 0.96\n",
      "Epoch 774: train accuracy = 0.98\n",
      "Epoch 775: train accuracy = 0.945\n",
      "Epoch 776: train accuracy = 0.96\n",
      "Epoch 777: train accuracy = 0.97\n",
      "Epoch 778: train accuracy = 0.975\n",
      "Epoch 779: train accuracy = 0.965\n",
      "Epoch 780: train accuracy = 0.95\n",
      "Epoch 781: train accuracy = 0.97\n",
      "Epoch 782: train accuracy = 0.945\n",
      "Epoch 783: train accuracy = 0.965\n",
      "Epoch 784: train accuracy = 0.97\n",
      "Epoch 785: train accuracy = 0.985\n",
      "Epoch 786: train accuracy = 0.97\n",
      "Epoch 787: train accuracy = 0.975\n",
      "Epoch 788: train accuracy = 0.98\n",
      "Epoch 789: train accuracy = 0.975\n",
      "Epoch 790: train accuracy = 0.955\n",
      "Epoch 791: train accuracy = 0.97\n",
      "Epoch 792: train accuracy = 0.96\n",
      "Epoch 793: train accuracy = 0.96\n",
      "Epoch 794: train accuracy = 0.94\n",
      "Epoch 795: train accuracy = 0.97\n",
      "Epoch 796: train accuracy = 0.965\n",
      "Epoch 797: train accuracy = 0.95\n",
      "Epoch 798: train accuracy = 0.95\n",
      "Epoch 799: train accuracy = 0.955\n",
      "Epoch 800: train accuracy = 0.96\n",
      "Epoch 801: train accuracy = 0.97\n",
      "Epoch 802: train accuracy = 0.945\n",
      "Epoch 803: train accuracy = 0.98\n",
      "Epoch 804: train accuracy = 0.96\n",
      "Epoch 805: train accuracy = 0.955\n",
      "Epoch 806: train accuracy = 0.97\n",
      "Epoch 807: train accuracy = 0.965\n",
      "Epoch 808: train accuracy = 0.945\n",
      "Epoch 809: train accuracy = 0.965\n",
      "Epoch 810: train accuracy = 0.965\n",
      "Epoch 811: train accuracy = 0.975\n",
      "Epoch 812: train accuracy = 0.97\n",
      "Epoch 813: train accuracy = 0.97\n",
      "Epoch 814: train accuracy = 0.975\n",
      "Epoch 815: train accuracy = 0.98\n",
      "Epoch 816: train accuracy = 0.96\n",
      "Epoch 817: train accuracy = 0.98\n",
      "Epoch 818: train accuracy = 0.965\n",
      "Epoch 819: train accuracy = 0.95\n",
      "Epoch 820: train accuracy = 0.99\n",
      "Epoch 821: train accuracy = 0.98\n",
      "Epoch 822: train accuracy = 0.97\n",
      "Epoch 823: train accuracy = 0.985\n",
      "Epoch 824: train accuracy = 0.97\n",
      "Epoch 825: train accuracy = 0.965\n",
      "Epoch 826: train accuracy = 0.95\n",
      "Epoch 827: train accuracy = 0.95\n",
      "Epoch 828: train accuracy = 0.945\n",
      "Epoch 829: train accuracy = 0.955\n",
      "Epoch 830: train accuracy = 0.965\n",
      "Epoch 831: train accuracy = 0.97\n",
      "Epoch 832: train accuracy = 0.975\n",
      "Epoch 833: train accuracy = 0.98\n",
      "Epoch 834: train accuracy = 0.96\n",
      "Epoch 835: train accuracy = 0.955\n",
      "Epoch 836: train accuracy = 0.975\n",
      "Epoch 837: train accuracy = 0.97\n",
      "Epoch 838: train accuracy = 0.95\n",
      "Epoch 839: train accuracy = 0.975\n",
      "Epoch 840: train accuracy = 0.985\n",
      "Epoch 841: train accuracy = 0.965\n",
      "Epoch 842: train accuracy = 0.945\n",
      "Epoch 843: train accuracy = 0.975\n",
      "Epoch 844: train accuracy = 0.985\n",
      "Epoch 845: train accuracy = 0.965\n",
      "Epoch 846: train accuracy = 0.99\n",
      "Epoch 847: train accuracy = 0.975\n",
      "Epoch 848: train accuracy = 0.98\n",
      "Epoch 849: train accuracy = 0.97\n",
      "Epoch 850: train accuracy = 0.98\n",
      "Epoch 851: train accuracy = 0.945\n",
      "Epoch 852: train accuracy = 0.98\n",
      "Epoch 853: train accuracy = 0.995\n",
      "Epoch 854: train accuracy = 0.98\n",
      "Epoch 855: train accuracy = 0.98\n",
      "Epoch 856: train accuracy = 0.985\n",
      "Epoch 857: train accuracy = 0.965\n",
      "Epoch 858: train accuracy = 0.965\n",
      "Epoch 859: train accuracy = 0.99\n",
      "Epoch 860: train accuracy = 0.955\n",
      "Epoch 861: train accuracy = 0.98\n",
      "Epoch 862: train accuracy = 0.975\n",
      "Epoch 863: train accuracy = 0.965\n",
      "Epoch 864: train accuracy = 0.97\n",
      "Epoch 865: train accuracy = 0.97\n",
      "Epoch 866: train accuracy = 0.98\n",
      "Epoch 867: train accuracy = 0.97\n",
      "Epoch 868: train accuracy = 0.965\n",
      "Epoch 869: train accuracy = 0.975\n",
      "Epoch 870: train accuracy = 0.975\n",
      "Epoch 871: train accuracy = 0.955\n",
      "Epoch 872: train accuracy = 0.975\n",
      "Epoch 873: train accuracy = 0.945\n",
      "Epoch 874: train accuracy = 0.955\n",
      "Epoch 875: train accuracy = 0.96\n",
      "Epoch 876: train accuracy = 0.96\n",
      "Epoch 877: train accuracy = 0.96\n",
      "Epoch 878: train accuracy = 0.98\n",
      "Epoch 879: train accuracy = 0.98\n",
      "Epoch 880: train accuracy = 0.965\n",
      "Epoch 881: train accuracy = 0.97\n",
      "Epoch 882: train accuracy = 0.965\n",
      "Epoch 883: train accuracy = 0.98\n",
      "Epoch 884: train accuracy = 0.99\n",
      "Epoch 885: train accuracy = 0.97\n",
      "Epoch 886: train accuracy = 0.96\n",
      "Epoch 887: train accuracy = 0.96\n",
      "Epoch 888: train accuracy = 0.985\n",
      "Epoch 889: train accuracy = 0.98\n",
      "Epoch 890: train accuracy = 0.98\n",
      "Epoch 891: train accuracy = 0.975\n",
      "Epoch 892: train accuracy = 0.95\n",
      "Epoch 893: train accuracy = 0.985\n",
      "Epoch 894: train accuracy = 0.965\n",
      "Epoch 895: train accuracy = 0.985\n",
      "Epoch 896: train accuracy = 0.97\n",
      "Epoch 897: train accuracy = 0.98\n",
      "Epoch 898: train accuracy = 0.95\n",
      "Epoch 899: train accuracy = 0.96\n",
      "Epoch 900: train accuracy = 0.975\n",
      "Epoch 901: train accuracy = 0.985\n",
      "Epoch 902: train accuracy = 0.97\n",
      "Epoch 903: train accuracy = 0.955\n",
      "Epoch 904: train accuracy = 0.985\n",
      "Epoch 905: train accuracy = 0.975\n",
      "Epoch 906: train accuracy = 0.98\n",
      "Epoch 907: train accuracy = 0.97\n",
      "Epoch 908: train accuracy = 0.965\n",
      "Epoch 909: train accuracy = 0.965\n",
      "Epoch 910: train accuracy = 0.98\n",
      "Epoch 911: train accuracy = 0.965\n",
      "Epoch 912: train accuracy = 0.97\n",
      "Epoch 913: train accuracy = 0.97\n",
      "Epoch 914: train accuracy = 0.965\n",
      "Epoch 915: train accuracy = 0.97\n",
      "Epoch 916: train accuracy = 0.965\n",
      "Epoch 917: train accuracy = 0.97\n",
      "Epoch 918: train accuracy = 0.98\n",
      "Epoch 919: train accuracy = 0.98\n",
      "Epoch 920: train accuracy = 0.95\n",
      "Epoch 921: train accuracy = 0.965\n",
      "Epoch 922: train accuracy = 0.975\n",
      "Epoch 923: train accuracy = 0.99\n",
      "Epoch 924: train accuracy = 0.97\n",
      "Epoch 925: train accuracy = 0.975\n",
      "Epoch 926: train accuracy = 0.975\n",
      "Epoch 927: train accuracy = 0.965\n",
      "Epoch 928: train accuracy = 0.985\n",
      "Epoch 929: train accuracy = 0.985\n",
      "Epoch 930: train accuracy = 0.98\n",
      "Epoch 931: train accuracy = 0.985\n",
      "Epoch 932: train accuracy = 0.97\n",
      "Epoch 933: train accuracy = 0.985\n",
      "Epoch 934: train accuracy = 0.97\n",
      "Epoch 935: train accuracy = 0.965\n",
      "Epoch 936: train accuracy = 0.985\n",
      "Epoch 937: train accuracy = 0.96\n",
      "Epoch 938: train accuracy = 0.985\n",
      "Epoch 939: train accuracy = 0.985\n",
      "Epoch 940: train accuracy = 0.99\n",
      "Epoch 941: train accuracy = 1.0\n",
      "Epoch 942: train accuracy = 0.935\n",
      "Epoch 943: train accuracy = 0.965\n",
      "Epoch 944: train accuracy = 0.965\n",
      "Epoch 945: train accuracy = 0.975\n",
      "Epoch 946: train accuracy = 0.99\n",
      "Epoch 947: train accuracy = 0.965\n",
      "Epoch 948: train accuracy = 0.955\n",
      "Epoch 949: train accuracy = 0.975\n",
      "Epoch 950: train accuracy = 0.98\n",
      "Epoch 951: train accuracy = 0.965\n",
      "Epoch 952: train accuracy = 0.99\n",
      "Epoch 953: train accuracy = 0.965\n",
      "Epoch 954: train accuracy = 0.96\n",
      "Epoch 955: train accuracy = 0.985\n",
      "Epoch 956: train accuracy = 0.98\n",
      "Epoch 957: train accuracy = 0.945\n",
      "Epoch 958: train accuracy = 0.955\n",
      "Epoch 959: train accuracy = 0.985\n",
      "Epoch 960: train accuracy = 0.98\n",
      "Epoch 961: train accuracy = 0.975\n",
      "Epoch 962: train accuracy = 0.98\n",
      "Epoch 963: train accuracy = 0.98\n",
      "Epoch 964: train accuracy = 0.98\n",
      "Epoch 965: train accuracy = 0.98\n",
      "Epoch 966: train accuracy = 0.98\n",
      "Epoch 967: train accuracy = 0.95\n",
      "Epoch 968: train accuracy = 0.99\n",
      "Epoch 969: train accuracy = 0.985\n",
      "Epoch 970: train accuracy = 0.95\n",
      "Epoch 971: train accuracy = 0.97\n",
      "Epoch 972: train accuracy = 0.985\n",
      "Epoch 973: train accuracy = 0.975\n",
      "Epoch 974: train accuracy = 0.95\n",
      "Epoch 975: train accuracy = 0.955\n",
      "Epoch 976: train accuracy = 0.95\n",
      "Epoch 977: train accuracy = 0.985\n",
      "Epoch 978: train accuracy = 0.98\n",
      "Epoch 979: train accuracy = 0.985\n",
      "Epoch 980: train accuracy = 0.98\n",
      "Epoch 981: train accuracy = 1.0\n",
      "Epoch 982: train accuracy = 0.99\n",
      "Epoch 983: train accuracy = 0.98\n",
      "Epoch 984: train accuracy = 0.985\n",
      "Epoch 985: train accuracy = 0.965\n",
      "Epoch 986: train accuracy = 0.97\n",
      "Epoch 987: train accuracy = 0.985\n",
      "Epoch 988: train accuracy = 0.975\n",
      "Epoch 989: train accuracy = 0.975\n",
      "Epoch 990: train accuracy = 0.99\n",
      "Epoch 991: train accuracy = 0.97\n",
      "Epoch 992: train accuracy = 0.965\n",
      "Epoch 993: train accuracy = 0.98\n",
      "Epoch 994: train accuracy = 0.975\n",
      "Epoch 995: train accuracy = 0.97\n",
      "Epoch 996: train accuracy = 0.96\n",
      "Epoch 997: train accuracy = 0.985\n",
      "Epoch 998: train accuracy = 0.965\n",
      "Epoch 999: train accuracy = 0.965\n"
     ]
    }
   ],
   "source": [
    "loss = 0\n",
    "batch_size = 200\n",
    "for epoch in range(1000):\n",
    "    i_ = np.random.choice(len(train_x), batch_size, replace=False)\n",
    "    l = [chainer.Variable(train_x[i, :, :]) for i in i_]\n",
    "    y = net(l, train=True)\n",
    "    loss = F.softmax_cross_entropy(y, train_t[i_])\n",
    "    net.cleargrads()\n",
    "    loss.backward()\n",
    "    optim.update()\n",
    "    print('Epoch ' + str(epoch) + ': train accuracy = ' + str(round(F.accuracy(y, train_t[i_]).data, 4)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 316,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test accuracy: 0.961\n"
     ]
    }
   ],
   "source": [
    "# calculate test accuracy\n",
    "l = [chainer.Variable(test_x[i, :, :]) for i in range(1000)]\n",
    "y = net(l, train=False)\n",
    "print('Test accuracy: ' + str(round(F.accuracy(y, test_t[:1000]).data, 4)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 317,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0xb0af9cc0>"
      ]
     },
     "execution_count": 317,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfoAAAFhCAYAAACYtGjJAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n8Qrj/wNUz6OrzSurKwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x8724c668>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "im = np.random.randint(0, 5000)\n",
    "hy, ys = net[0](hx=None, xs=[chainer.Variable(test_x[im,:,:])], train=False)\n",
    "plt.subplot(121)\n",
    "plt.imshow(test_x[im], aspect='auto')\n",
    "plt.subplot(122)\n",
    "plt.imshow(ys[0].data.T, aspect='auto')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# GRU from StatefulGRU in Chainer\n",
    "\n",
    "Do the same as in previous section but using a different link. Manually select the last time point and map to top softmax layer."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "class SimpleGRU(chainer.ChainList):\n",
    "    def __init__(self, n_input, n_hidden, n_output):\n",
    "        links = []\n",
    "        links.append(L.StatefulGRU(in_size=n_input, out_size=n_hidden))\n",
    "        links.append(L.Linear(None, n_output))\n",
    "        super(SimpleGRU, self).__init__(*links)\n",
    "        \n",
    "    def reset_state(self):\n",
    "        self[0].reset_state()\n",
    "\n",
    "    def __call__(self, x):\n",
    "        z = []     \n",
    "        \n",
    "        for t in range(x.shape[1]):\n",
    "            z.append(self[0](x[:,t,:]))\n",
    "            \n",
    "        y = self[1](z[-1])\n",
    "        return y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# initialize CNN and optimizer\n",
    "net2 = SimpleGRU(n_input=28, n_hidden=50, n_output=10)\n",
    "optim = chainer.optimizers.Adam(alpha=1e-03)\n",
    "optim.setup(net2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 0: train accuracy = 0.066\n",
      "Epoch 1: train accuracy = 0.134\n",
      "Epoch 2: train accuracy = 0.132\n",
      "Epoch 3: train accuracy = 0.19\n",
      "Epoch 4: train accuracy = 0.162\n",
      "Epoch 5: train accuracy = 0.228\n",
      "Epoch 6: train accuracy = 0.192\n",
      "Epoch 7: train accuracy = 0.174\n",
      "Epoch 8: train accuracy = 0.216\n",
      "Epoch 9: train accuracy = 0.246\n",
      "Epoch 10: train accuracy = 0.188\n",
      "Epoch 11: train accuracy = 0.236\n",
      "Epoch 12: train accuracy = 0.232\n",
      "Epoch 13: train accuracy = 0.236\n",
      "Epoch 14: train accuracy = 0.256\n",
      "Epoch 15: train accuracy = 0.236\n",
      "Epoch 16: train accuracy = 0.3\n",
      "Epoch 17: train accuracy = 0.262\n",
      "Epoch 18: train accuracy = 0.292\n",
      "Epoch 19: train accuracy = 0.298\n",
      "Epoch 20: train accuracy = 0.274\n",
      "Epoch 21: train accuracy = 0.288\n",
      "Epoch 22: train accuracy = 0.292\n",
      "Epoch 23: train accuracy = 0.356\n",
      "Epoch 24: train accuracy = 0.336\n",
      "Epoch 25: train accuracy = 0.366\n",
      "Epoch 26: train accuracy = 0.356\n",
      "Epoch 27: train accuracy = 0.35\n",
      "Epoch 28: train accuracy = 0.39\n",
      "Epoch 29: train accuracy = 0.36\n",
      "Epoch 30: train accuracy = 0.364\n",
      "Epoch 31: train accuracy = 0.346\n",
      "Epoch 32: train accuracy = 0.388\n",
      "Epoch 33: train accuracy = 0.382\n",
      "Epoch 34: train accuracy = 0.402\n",
      "Epoch 35: train accuracy = 0.4\n",
      "Epoch 36: train accuracy = 0.398\n",
      "Epoch 37: train accuracy = 0.424\n",
      "Epoch 38: train accuracy = 0.41\n",
      "Epoch 39: train accuracy = 0.426\n",
      "Epoch 40: train accuracy = 0.386\n",
      "Epoch 41: train accuracy = 0.428\n",
      "Epoch 42: train accuracy = 0.438\n",
      "Epoch 43: train accuracy = 0.418\n",
      "Epoch 44: train accuracy = 0.424\n",
      "Epoch 45: train accuracy = 0.43\n",
      "Epoch 46: train accuracy = 0.45\n",
      "Epoch 47: train accuracy = 0.468\n",
      "Epoch 48: train accuracy = 0.494\n",
      "Epoch 49: train accuracy = 0.452\n",
      "Epoch 50: train accuracy = 0.512\n",
      "Epoch 51: train accuracy = 0.45\n",
      "Epoch 52: train accuracy = 0.494\n",
      "Epoch 53: train accuracy = 0.518\n",
      "Epoch 54: train accuracy = 0.536\n",
      "Epoch 55: train accuracy = 0.53\n",
      "Epoch 56: train accuracy = 0.514\n",
      "Epoch 57: train accuracy = 0.5\n",
      "Epoch 58: train accuracy = 0.496\n",
      "Epoch 59: train accuracy = 0.494\n",
      "Epoch 60: train accuracy = 0.496\n",
      "Epoch 61: train accuracy = 0.552\n",
      "Epoch 62: train accuracy = 0.564\n",
      "Epoch 63: train accuracy = 0.516\n",
      "Epoch 64: train accuracy = 0.53\n",
      "Epoch 65: train accuracy = 0.568\n",
      "Epoch 66: train accuracy = 0.508\n",
      "Epoch 67: train accuracy = 0.572\n",
      "Epoch 68: train accuracy = 0.59\n",
      "Epoch 69: train accuracy = 0.59\n",
      "Epoch 70: train accuracy = 0.582\n",
      "Epoch 71: train accuracy = 0.624\n",
      "Epoch 72: train accuracy = 0.608\n",
      "Epoch 73: train accuracy = 0.578\n",
      "Epoch 74: train accuracy = 0.62\n",
      "Epoch 75: train accuracy = 0.62\n",
      "Epoch 76: train accuracy = 0.552\n",
      "Epoch 77: train accuracy = 0.602\n",
      "Epoch 78: train accuracy = 0.614\n",
      "Epoch 79: train accuracy = 0.648\n",
      "Epoch 80: train accuracy = 0.664\n",
      "Epoch 81: train accuracy = 0.61\n",
      "Epoch 82: train accuracy = 0.674\n",
      "Epoch 83: train accuracy = 0.586\n",
      "Epoch 84: train accuracy = 0.646\n",
      "Epoch 85: train accuracy = 0.662\n",
      "Epoch 86: train accuracy = 0.654\n",
      "Epoch 87: train accuracy = 0.634\n",
      "Epoch 88: train accuracy = 0.656\n",
      "Epoch 89: train accuracy = 0.644\n",
      "Epoch 90: train accuracy = 0.68\n",
      "Epoch 91: train accuracy = 0.638\n",
      "Epoch 92: train accuracy = 0.674\n",
      "Epoch 93: train accuracy = 0.69\n",
      "Epoch 94: train accuracy = 0.708\n",
      "Epoch 95: train accuracy = 0.678\n",
      "Epoch 96: train accuracy = 0.652\n",
      "Epoch 97: train accuracy = 0.656\n",
      "Epoch 98: train accuracy = 0.682\n",
      "Epoch 99: train accuracy = 0.688\n",
      "Epoch 100: train accuracy = 0.66\n",
      "Epoch 101: train accuracy = 0.73\n",
      "Epoch 102: train accuracy = 0.71\n",
      "Epoch 103: train accuracy = 0.722\n",
      "Epoch 104: train accuracy = 0.73\n",
      "Epoch 105: train accuracy = 0.73\n",
      "Epoch 106: train accuracy = 0.694\n",
      "Epoch 107: train accuracy = 0.732\n",
      "Epoch 108: train accuracy = 0.674\n",
      "Epoch 109: train accuracy = 0.708\n",
      "Epoch 110: train accuracy = 0.68\n",
      "Epoch 111: train accuracy = 0.712\n",
      "Epoch 112: train accuracy = 0.73\n",
      "Epoch 113: train accuracy = 0.7\n",
      "Epoch 114: train accuracy = 0.734\n",
      "Epoch 115: train accuracy = 0.726\n",
      "Epoch 116: train accuracy = 0.712\n",
      "Epoch 117: train accuracy = 0.73\n",
      "Epoch 118: train accuracy = 0.758\n",
      "Epoch 119: train accuracy = 0.726\n",
      "Epoch 120: train accuracy = 0.73\n",
      "Epoch 121: train accuracy = 0.722\n",
      "Epoch 122: train accuracy = 0.732\n",
      "Epoch 123: train accuracy = 0.746\n",
      "Epoch 124: train accuracy = 0.754\n",
      "Epoch 125: train accuracy = 0.772\n",
      "Epoch 126: train accuracy = 0.758\n",
      "Epoch 127: train accuracy = 0.77\n",
      "Epoch 128: train accuracy = 0.764\n",
      "Epoch 129: train accuracy = 0.788\n",
      "Epoch 130: train accuracy = 0.766\n",
      "Epoch 131: train accuracy = 0.772\n",
      "Epoch 132: train accuracy = 0.81\n",
      "Epoch 133: train accuracy = 0.78\n",
      "Epoch 134: train accuracy = 0.758\n",
      "Epoch 135: train accuracy = 0.742\n",
      "Epoch 136: train accuracy = 0.806\n",
      "Epoch 137: train accuracy = 0.786\n",
      "Epoch 138: train accuracy = 0.784\n",
      "Epoch 139: train accuracy = 0.752\n",
      "Epoch 140: train accuracy = 0.76\n",
      "Epoch 141: train accuracy = 0.788\n",
      "Epoch 142: train accuracy = 0.79\n",
      "Epoch 143: train accuracy = 0.826\n",
      "Epoch 144: train accuracy = 0.796\n",
      "Epoch 145: train accuracy = 0.818\n",
      "Epoch 146: train accuracy = 0.784\n",
      "Epoch 147: train accuracy = 0.812\n",
      "Epoch 148: train accuracy = 0.812\n",
      "Epoch 149: train accuracy = 0.794\n",
      "Epoch 150: train accuracy = 0.776\n",
      "Epoch 151: train accuracy = 0.82\n",
      "Epoch 152: train accuracy = 0.79\n",
      "Epoch 153: train accuracy = 0.818\n",
      "Epoch 154: train accuracy = 0.82\n",
      "Epoch 155: train accuracy = 0.79\n",
      "Epoch 156: train accuracy = 0.838\n",
      "Epoch 157: train accuracy = 0.824\n",
      "Epoch 158: train accuracy = 0.808\n",
      "Epoch 159: train accuracy = 0.806\n",
      "Epoch 160: train accuracy = 0.804\n",
      "Epoch 161: train accuracy = 0.83\n",
      "Epoch 162: train accuracy = 0.848\n",
      "Epoch 163: train accuracy = 0.812\n",
      "Epoch 164: train accuracy = 0.86\n",
      "Epoch 165: train accuracy = 0.828\n",
      "Epoch 166: train accuracy = 0.786\n",
      "Epoch 167: train accuracy = 0.83\n",
      "Epoch 168: train accuracy = 0.83\n",
      "Epoch 169: train accuracy = 0.83\n",
      "Epoch 170: train accuracy = 0.834\n",
      "Epoch 171: train accuracy = 0.836\n",
      "Epoch 172: train accuracy = 0.852\n",
      "Epoch 173: train accuracy = 0.82\n",
      "Epoch 174: train accuracy = 0.836\n",
      "Epoch 175: train accuracy = 0.824\n",
      "Epoch 176: train accuracy = 0.844\n",
      "Epoch 177: train accuracy = 0.816\n",
      "Epoch 178: train accuracy = 0.866\n",
      "Epoch 179: train accuracy = 0.866\n",
      "Epoch 180: train accuracy = 0.844\n",
      "Epoch 181: train accuracy = 0.832\n",
      "Epoch 182: train accuracy = 0.852\n",
      "Epoch 183: train accuracy = 0.844\n",
      "Epoch 184: train accuracy = 0.856\n",
      "Epoch 185: train accuracy = 0.852\n",
      "Epoch 186: train accuracy = 0.852\n",
      "Epoch 187: train accuracy = 0.854\n",
      "Epoch 188: train accuracy = 0.848\n",
      "Epoch 189: train accuracy = 0.874\n",
      "Epoch 190: train accuracy = 0.862\n",
      "Epoch 191: train accuracy = 0.856\n",
      "Epoch 192: train accuracy = 0.86\n",
      "Epoch 193: train accuracy = 0.836\n",
      "Epoch 194: train accuracy = 0.87\n",
      "Epoch 195: train accuracy = 0.878\n",
      "Epoch 196: train accuracy = 0.854\n",
      "Epoch 197: train accuracy = 0.832\n",
      "Epoch 198: train accuracy = 0.862\n",
      "Epoch 199: train accuracy = 0.874\n",
      "Epoch 200: train accuracy = 0.854\n",
      "Epoch 201: train accuracy = 0.838\n",
      "Epoch 202: train accuracy = 0.882\n",
      "Epoch 203: train accuracy = 0.882\n",
      "Epoch 204: train accuracy = 0.86\n",
      "Epoch 205: train accuracy = 0.882\n",
      "Epoch 206: train accuracy = 0.89\n",
      "Epoch 207: train accuracy = 0.876\n",
      "Epoch 208: train accuracy = 0.85\n",
      "Epoch 209: train accuracy = 0.874\n",
      "Epoch 210: train accuracy = 0.882\n",
      "Epoch 211: train accuracy = 0.874\n",
      "Epoch 212: train accuracy = 0.87\n",
      "Epoch 213: train accuracy = 0.848\n",
      "Epoch 214: train accuracy = 0.876\n",
      "Epoch 215: train accuracy = 0.868\n",
      "Epoch 216: train accuracy = 0.866\n",
      "Epoch 217: train accuracy = 0.866\n",
      "Epoch 218: train accuracy = 0.898\n",
      "Epoch 219: train accuracy = 0.864\n",
      "Epoch 220: train accuracy = 0.854\n",
      "Epoch 221: train accuracy = 0.882\n",
      "Epoch 222: train accuracy = 0.88\n",
      "Epoch 223: train accuracy = 0.87\n",
      "Epoch 224: train accuracy = 0.878\n",
      "Epoch 225: train accuracy = 0.85\n",
      "Epoch 226: train accuracy = 0.892\n",
      "Epoch 227: train accuracy = 0.89\n",
      "Epoch 228: train accuracy = 0.894\n",
      "Epoch 229: train accuracy = 0.904\n",
      "Epoch 230: train accuracy = 0.896\n",
      "Epoch 231: train accuracy = 0.9\n",
      "Epoch 232: train accuracy = 0.898\n",
      "Epoch 233: train accuracy = 0.898\n",
      "Epoch 234: train accuracy = 0.882\n",
      "Epoch 235: train accuracy = 0.9\n",
      "Epoch 236: train accuracy = 0.884\n",
      "Epoch 237: train accuracy = 0.912\n",
      "Epoch 238: train accuracy = 0.916\n",
      "Epoch 239: train accuracy = 0.892\n",
      "Epoch 240: train accuracy = 0.904\n",
      "Epoch 241: train accuracy = 0.882\n",
      "Epoch 242: train accuracy = 0.89\n",
      "Epoch 243: train accuracy = 0.908\n",
      "Epoch 244: train accuracy = 0.886\n",
      "Epoch 245: train accuracy = 0.92\n",
      "Epoch 246: train accuracy = 0.89\n",
      "Epoch 247: train accuracy = 0.908\n",
      "Epoch 248: train accuracy = 0.878\n",
      "Epoch 249: train accuracy = 0.912\n",
      "Epoch 250: train accuracy = 0.906\n",
      "Epoch 251: train accuracy = 0.912\n",
      "Epoch 252: train accuracy = 0.912\n",
      "Epoch 253: train accuracy = 0.896\n",
      "Epoch 254: train accuracy = 0.88\n",
      "Epoch 255: train accuracy = 0.874\n",
      "Epoch 256: train accuracy = 0.902\n",
      "Epoch 257: train accuracy = 0.898\n",
      "Epoch 258: train accuracy = 0.89\n",
      "Epoch 259: train accuracy = 0.89\n",
      "Epoch 260: train accuracy = 0.896\n",
      "Epoch 261: train accuracy = 0.924\n",
      "Epoch 262: train accuracy = 0.912\n",
      "Epoch 263: train accuracy = 0.91\n",
      "Epoch 264: train accuracy = 0.918\n",
      "Epoch 265: train accuracy = 0.908\n",
      "Epoch 266: train accuracy = 0.888\n",
      "Epoch 267: train accuracy = 0.92\n",
      "Epoch 268: train accuracy = 0.912\n",
      "Epoch 269: train accuracy = 0.912\n",
      "Epoch 270: train accuracy = 0.91\n",
      "Epoch 271: train accuracy = 0.91\n",
      "Epoch 272: train accuracy = 0.932\n",
      "Epoch 273: train accuracy = 0.918\n",
      "Epoch 274: train accuracy = 0.882\n",
      "Epoch 275: train accuracy = 0.908\n",
      "Epoch 276: train accuracy = 0.914\n",
      "Epoch 277: train accuracy = 0.914\n",
      "Epoch 278: train accuracy = 0.916\n",
      "Epoch 279: train accuracy = 0.902\n",
      "Epoch 280: train accuracy = 0.928\n",
      "Epoch 281: train accuracy = 0.934\n",
      "Epoch 282: train accuracy = 0.928\n",
      "Epoch 283: train accuracy = 0.906\n",
      "Epoch 284: train accuracy = 0.898\n",
      "Epoch 285: train accuracy = 0.932\n",
      "Epoch 286: train accuracy = 0.896\n",
      "Epoch 287: train accuracy = 0.914\n",
      "Epoch 288: train accuracy = 0.918\n",
      "Epoch 289: train accuracy = 0.896\n",
      "Epoch 290: train accuracy = 0.92\n",
      "Epoch 291: train accuracy = 0.926\n",
      "Epoch 292: train accuracy = 0.934\n",
      "Epoch 293: train accuracy = 0.928\n",
      "Epoch 294: train accuracy = 0.938\n",
      "Epoch 295: train accuracy = 0.912\n",
      "Epoch 296: train accuracy = 0.922\n",
      "Epoch 297: train accuracy = 0.938\n",
      "Epoch 298: train accuracy = 0.924\n",
      "Epoch 299: train accuracy = 0.892\n",
      "Epoch 300: train accuracy = 0.942\n",
      "Epoch 301: train accuracy = 0.922\n",
      "Epoch 302: train accuracy = 0.924\n",
      "Epoch 303: train accuracy = 0.9\n",
      "Epoch 304: train accuracy = 0.91\n",
      "Epoch 305: train accuracy = 0.904\n",
      "Epoch 306: train accuracy = 0.93\n",
      "Epoch 307: train accuracy = 0.922\n",
      "Epoch 308: train accuracy = 0.896\n",
      "Epoch 309: train accuracy = 0.918\n",
      "Epoch 310: train accuracy = 0.934\n",
      "Epoch 311: train accuracy = 0.902\n",
      "Epoch 312: train accuracy = 0.922\n",
      "Epoch 313: train accuracy = 0.926\n",
      "Epoch 314: train accuracy = 0.92\n",
      "Epoch 315: train accuracy = 0.918\n",
      "Epoch 316: train accuracy = 0.938\n",
      "Epoch 317: train accuracy = 0.924\n",
      "Epoch 318: train accuracy = 0.928\n",
      "Epoch 319: train accuracy = 0.934\n",
      "Epoch 320: train accuracy = 0.912\n",
      "Epoch 321: train accuracy = 0.922\n",
      "Epoch 322: train accuracy = 0.916\n",
      "Epoch 323: train accuracy = 0.912\n",
      "Epoch 324: train accuracy = 0.936\n",
      "Epoch 325: train accuracy = 0.914\n",
      "Epoch 326: train accuracy = 0.944\n",
      "Epoch 327: train accuracy = 0.908\n",
      "Epoch 328: train accuracy = 0.942\n",
      "Epoch 329: train accuracy = 0.924\n",
      "Epoch 330: train accuracy = 0.922\n",
      "Epoch 331: train accuracy = 0.92\n",
      "Epoch 332: train accuracy = 0.956\n",
      "Epoch 333: train accuracy = 0.93\n",
      "Epoch 334: train accuracy = 0.928\n",
      "Epoch 335: train accuracy = 0.93\n",
      "Epoch 336: train accuracy = 0.93\n",
      "Epoch 337: train accuracy = 0.918\n",
      "Epoch 338: train accuracy = 0.924\n",
      "Epoch 339: train accuracy = 0.942\n",
      "Epoch 340: train accuracy = 0.932\n",
      "Epoch 341: train accuracy = 0.946\n",
      "Epoch 342: train accuracy = 0.92\n",
      "Epoch 343: train accuracy = 0.916\n",
      "Epoch 344: train accuracy = 0.934\n",
      "Epoch 345: train accuracy = 0.916\n",
      "Epoch 346: train accuracy = 0.922\n",
      "Epoch 347: train accuracy = 0.914\n",
      "Epoch 348: train accuracy = 0.918\n",
      "Epoch 349: train accuracy = 0.94\n",
      "Epoch 350: train accuracy = 0.924\n",
      "Epoch 351: train accuracy = 0.926\n",
      "Epoch 352: train accuracy = 0.918\n",
      "Epoch 353: train accuracy = 0.926\n",
      "Epoch 354: train accuracy = 0.94\n",
      "Epoch 355: train accuracy = 0.924\n",
      "Epoch 356: train accuracy = 0.92\n",
      "Epoch 357: train accuracy = 0.93\n",
      "Epoch 358: train accuracy = 0.944\n",
      "Epoch 359: train accuracy = 0.944\n",
      "Epoch 360: train accuracy = 0.92\n",
      "Epoch 361: train accuracy = 0.94\n",
      "Epoch 362: train accuracy = 0.942\n",
      "Epoch 363: train accuracy = 0.936\n",
      "Epoch 364: train accuracy = 0.928\n",
      "Epoch 365: train accuracy = 0.906\n",
      "Epoch 366: train accuracy = 0.924\n",
      "Epoch 367: train accuracy = 0.938\n",
      "Epoch 368: train accuracy = 0.932\n",
      "Epoch 369: train accuracy = 0.956\n",
      "Epoch 370: train accuracy = 0.934\n",
      "Epoch 371: train accuracy = 0.938\n",
      "Epoch 372: train accuracy = 0.942\n",
      "Epoch 373: train accuracy = 0.954\n",
      "Epoch 374: train accuracy = 0.916\n",
      "Epoch 375: train accuracy = 0.926\n",
      "Epoch 376: train accuracy = 0.928\n",
      "Epoch 377: train accuracy = 0.944\n",
      "Epoch 378: train accuracy = 0.94\n",
      "Epoch 379: train accuracy = 0.92\n",
      "Epoch 380: train accuracy = 0.952\n",
      "Epoch 381: train accuracy = 0.934\n",
      "Epoch 382: train accuracy = 0.928\n",
      "Epoch 383: train accuracy = 0.936\n",
      "Epoch 384: train accuracy = 0.938\n",
      "Epoch 385: train accuracy = 0.906\n",
      "Epoch 386: train accuracy = 0.936\n",
      "Epoch 387: train accuracy = 0.952\n",
      "Epoch 388: train accuracy = 0.918\n",
      "Epoch 389: train accuracy = 0.952\n",
      "Epoch 390: train accuracy = 0.94\n",
      "Epoch 391: train accuracy = 0.938\n",
      "Epoch 392: train accuracy = 0.948\n",
      "Epoch 393: train accuracy = 0.926\n",
      "Epoch 394: train accuracy = 0.934\n",
      "Epoch 395: train accuracy = 0.934\n",
      "Epoch 396: train accuracy = 0.938\n",
      "Epoch 397: train accuracy = 0.948\n",
      "Epoch 398: train accuracy = 0.928\n",
      "Epoch 399: train accuracy = 0.944\n",
      "Epoch 400: train accuracy = 0.942\n",
      "Epoch 401: train accuracy = 0.944\n",
      "Epoch 402: train accuracy = 0.946\n",
      "Epoch 403: train accuracy = 0.922\n",
      "Epoch 404: train accuracy = 0.954\n",
      "Epoch 405: train accuracy = 0.924\n",
      "Epoch 406: train accuracy = 0.952\n",
      "Epoch 407: train accuracy = 0.926\n",
      "Epoch 408: train accuracy = 0.946\n",
      "Epoch 409: train accuracy = 0.926\n",
      "Epoch 410: train accuracy = 0.928\n",
      "Epoch 411: train accuracy = 0.944\n",
      "Epoch 412: train accuracy = 0.948\n",
      "Epoch 413: train accuracy = 0.948\n",
      "Epoch 414: train accuracy = 0.924\n",
      "Epoch 415: train accuracy = 0.96\n",
      "Epoch 416: train accuracy = 0.956\n",
      "Epoch 417: train accuracy = 0.938\n",
      "Epoch 418: train accuracy = 0.93\n",
      "Epoch 419: train accuracy = 0.938\n",
      "Epoch 420: train accuracy = 0.942\n",
      "Epoch 421: train accuracy = 0.946\n",
      "Epoch 422: train accuracy = 0.936\n",
      "Epoch 423: train accuracy = 0.936\n",
      "Epoch 424: train accuracy = 0.932\n",
      "Epoch 425: train accuracy = 0.928\n",
      "Epoch 426: train accuracy = 0.936\n",
      "Epoch 427: train accuracy = 0.938\n",
      "Epoch 428: train accuracy = 0.924\n",
      "Epoch 429: train accuracy = 0.952\n",
      "Epoch 430: train accuracy = 0.952\n",
      "Epoch 431: train accuracy = 0.95\n",
      "Epoch 432: train accuracy = 0.958\n",
      "Epoch 433: train accuracy = 0.942\n",
      "Epoch 434: train accuracy = 0.934\n",
      "Epoch 435: train accuracy = 0.938\n",
      "Epoch 436: train accuracy = 0.954\n",
      "Epoch 437: train accuracy = 0.928\n",
      "Epoch 438: train accuracy = 0.924\n",
      "Epoch 439: train accuracy = 0.946\n",
      "Epoch 440: train accuracy = 0.946\n",
      "Epoch 441: train accuracy = 0.952\n",
      "Epoch 442: train accuracy = 0.952\n",
      "Epoch 443: train accuracy = 0.95\n",
      "Epoch 444: train accuracy = 0.95\n",
      "Epoch 445: train accuracy = 0.96\n",
      "Epoch 446: train accuracy = 0.93\n",
      "Epoch 447: train accuracy = 0.922\n",
      "Epoch 448: train accuracy = 0.956\n",
      "Epoch 449: train accuracy = 0.952\n",
      "Epoch 450: train accuracy = 0.938\n",
      "Epoch 451: train accuracy = 0.932\n",
      "Epoch 452: train accuracy = 0.932\n",
      "Epoch 453: train accuracy = 0.932\n",
      "Epoch 454: train accuracy = 0.928\n",
      "Epoch 455: train accuracy = 0.948\n",
      "Epoch 456: train accuracy = 0.956\n",
      "Epoch 457: train accuracy = 0.94\n",
      "Epoch 458: train accuracy = 0.934\n",
      "Epoch 459: train accuracy = 0.958\n",
      "Epoch 460: train accuracy = 0.946\n",
      "Epoch 461: train accuracy = 0.952\n",
      "Epoch 462: train accuracy = 0.94\n",
      "Epoch 463: train accuracy = 0.934\n",
      "Epoch 464: train accuracy = 0.936\n",
      "Epoch 465: train accuracy = 0.936\n",
      "Epoch 466: train accuracy = 0.928\n",
      "Epoch 467: train accuracy = 0.942\n",
      "Epoch 468: train accuracy = 0.974\n",
      "Epoch 469: train accuracy = 0.946\n",
      "Epoch 470: train accuracy = 0.944\n",
      "Epoch 471: train accuracy = 0.942\n",
      "Epoch 472: train accuracy = 0.942\n",
      "Epoch 473: train accuracy = 0.934\n",
      "Epoch 474: train accuracy = 0.932\n",
      "Epoch 475: train accuracy = 0.932\n",
      "Epoch 476: train accuracy = 0.94\n",
      "Epoch 477: train accuracy = 0.94\n",
      "Epoch 478: train accuracy = 0.962\n",
      "Epoch 479: train accuracy = 0.944\n",
      "Epoch 480: train accuracy = 0.94\n",
      "Epoch 481: train accuracy = 0.948\n",
      "Epoch 482: train accuracy = 0.964\n",
      "Epoch 483: train accuracy = 0.944\n",
      "Epoch 484: train accuracy = 0.934\n",
      "Epoch 485: train accuracy = 0.948\n",
      "Epoch 486: train accuracy = 0.92\n",
      "Epoch 487: train accuracy = 0.944\n",
      "Epoch 488: train accuracy = 0.944\n",
      "Epoch 489: train accuracy = 0.952\n",
      "Epoch 490: train accuracy = 0.95\n",
      "Epoch 491: train accuracy = 0.942\n",
      "Epoch 492: train accuracy = 0.958\n",
      "Epoch 493: train accuracy = 0.944\n",
      "Epoch 494: train accuracy = 0.948\n",
      "Epoch 495: train accuracy = 0.934\n",
      "Epoch 496: train accuracy = 0.936\n",
      "Epoch 497: train accuracy = 0.946\n",
      "Epoch 498: train accuracy = 0.94\n",
      "Epoch 499: train accuracy = 0.942\n"
     ]
    }
   ],
   "source": [
    "loss = 0\n",
    "batch_size = 500\n",
    "for epoch in range(500):\n",
    "    net2.reset_state()\n",
    "    i_ = np.random.choice(len(train_x), batch_size, replace=False)\n",
    "    y = net2(train_x[i_])\n",
    "    loss = F.softmax_cross_entropy(y, train_t[i_])\n",
    "    net2.cleargrads()\n",
    "    loss.backward()\n",
    "    optim.update()\n",
    "    print('Epoch ' + str(epoch) + ': train accuracy = ' + str(round(F.accuracy(y, train_t[i_]).data, 4)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test accuracy: 0.932\n"
     ]
    }
   ],
   "source": [
    "# calculate test accuracy\n",
    "net2.reset_state()\n",
    "y = net2(test_x[:1000, :, :])\n",
    "print('Test accuracy: ' + str(round(F.accuracy(y, test_t[:1000]).data, 4)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x3bbc4438>"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfoAAAFhCAYAAACYtGjJAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nmP8H4tJQhdBjg+0AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1622b390>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "im = np.random.randint(0, 5000)\n",
    "net2.reset_state()\n",
    "h = []\n",
    "for t in range(28):\n",
    "    h.append(net2[0](test_x[im:im+1,t,:]).data)\n",
    "    \n",
    "h = np.array(h).squeeze()\n",
    "plt.subplot(121)\n",
    "plt.imshow(test_x[im], aspect='auto')\n",
    "plt.subplot(122)\n",
    "plt.imshow(h.T, aspect='auto')"
   ]
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python [conda root]",
   "language": "python",
   "name": "conda-root-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}