iaf-mnist-implementation-Copy1.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/hang/anaconda3/lib/python3.6/site-packages/h5py/__init__.py:34: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`.\n",
      "  from ._conv import register_converters as _register_converters\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import math\n",
    "import tensorflow as tf\n",
    "import matplotlib.pyplot as plt\n",
    "import scipy\n",
    "from sklearn import preprocessing\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From /home/hang/anaconda3/lib/python3.6/site-packages/tensorflow/contrib/learn/python/learn/datasets/base.py:198: retry (from tensorflow.contrib.learn.python.learn.datasets.base) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use the retry module or similar alternatives.\n",
      "WARNING:tensorflow:From <ipython-input-2-ed54aa3a7fe8>:2: read_data_sets (from tensorflow.contrib.learn.python.learn.datasets.mnist) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Please use alternatives such as official/mnist/dataset.py from tensorflow/models.\n",
      "WARNING:tensorflow:From /home/hang/anaconda3/lib/python3.6/site-packages/tensorflow/contrib/learn/python/learn/datasets/mnist.py:260: maybe_download (from tensorflow.contrib.learn.python.learn.datasets.base) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Please write your own downloading logic.\n",
      "WARNING:tensorflow:From /home/hang/anaconda3/lib/python3.6/site-packages/tensorflow/contrib/learn/python/learn/datasets/mnist.py:262: extract_images (from tensorflow.contrib.learn.python.learn.datasets.mnist) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Please use tf.data to implement this functionality.\n",
      "Extracting MNIST_data/train-images-idx3-ubyte.gz\n",
      "WARNING:tensorflow:From /home/hang/anaconda3/lib/python3.6/site-packages/tensorflow/contrib/learn/python/learn/datasets/mnist.py:267: extract_labels (from tensorflow.contrib.learn.python.learn.datasets.mnist) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Please use tf.data to implement this functionality.\n",
      "Extracting MNIST_data/train-labels-idx1-ubyte.gz\n",
      "WARNING:tensorflow:From /home/hang/anaconda3/lib/python3.6/site-packages/tensorflow/contrib/learn/python/learn/datasets/mnist.py:110: dense_to_one_hot (from tensorflow.contrib.learn.python.learn.datasets.mnist) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Please use tf.one_hot on tensors.\n",
      "Extracting MNIST_data/t10k-images-idx3-ubyte.gz\n",
      "Extracting MNIST_data/t10k-labels-idx1-ubyte.gz\n",
      "WARNING:tensorflow:From /home/hang/anaconda3/lib/python3.6/site-packages/tensorflow/contrib/learn/python/learn/datasets/mnist.py:290: DataSet.__init__ (from tensorflow.contrib.learn.python.learn.datasets.mnist) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Please use alternatives such as official/mnist/dataset.py from tensorflow/models.\n",
      "Number of samples 55000 Shape of y[(55000, 10)] Shape of X[(55000, 784)]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAP8AAAD8CAYAAAC4nHJkAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n+j8PZtfAf3LXewAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f47a195af28>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from tensorflow.examples.tutorials.mnist import input_data\n",
    "mnist = input_data.read_data_sets('MNIST_data', one_hot=True)\n",
    "n_samples = mnist.train.num_examples\n",
    "print(\"Number of samples {} Shape of y[{}] Shape of X[{}]\"\n",
    "      .format(n_samples, mnist.train.labels.shape, mnist.train.images.shape))\n",
    "plt.imshow(np.reshape(-mnist.train.images[4242], (28, 28)), interpolation='none',cmap=plt.get_cmap('gray'))\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# encoder neural network - take as input observation x, output mu and sigma of latent variable z and content h\n",
    "class Encoder(object):\n",
    "    def __init__(self,inputs,hidden_dim,z_dim,h_dim):\n",
    "        self.X=inputs\n",
    "        self.layer_dims=[inputs.get_shape()[1], hidden_dim,hidden_dim, z_dim * 2 + h_dim] # output of Encoder (z_mu, z_sigma, h)\n",
    "        self.nz=z_dim\n",
    "        self.nh=h_dim\n",
    "        with tf.variable_scope('encoder'):\n",
    "            self.parameters = self._initialize_parameters()\n",
    "            \n",
    "    def _initialize_parameters(self):\n",
    "        L=len(self.layer_dims)\n",
    "        parameters={}\n",
    "        for l in range(1,L-1):            \n",
    "            parameters['W' + str(l)] = tf.get_variable('W'+str(l),[self.layer_dims[l],self.layer_dims[l-1]],initializer=tf.contrib.layers.variance_scaling_initializer(factor=0.002))\n",
    "            parameters['b' + str(l)] = tf.get_variable('b'+str(l),self.layer_dims[l],initializer=tf.zeros_initializer())\n",
    "        \n",
    "        w_s = tf.get_variable('w_s',[self.nz,self.layer_dims[L-2]],initializer=tf.contrib.layers.variance_scaling_initializer(factor=0.002))\n",
    "        w_o = tf.get_variable('w_o',[self.nz+self.nh,self.layer_dims[L-2]],initializer=tf.contrib.layers.variance_scaling_initializer())\n",
    "        parameters['W'+str(L-1)] = tf.concat([w_s,w_o],axis = 0,name='W'+str(L-1))\n",
    "        parameters['b'+str(L-1)] = tf.get_variable('b'+str(L-1),self.layer_dims[L-1],initializer=tf.constant_initializer(-6))\n",
    "        return parameters\n",
    "    \n",
    "    def encoding(self):\n",
    "        L=len(self.layer_dims)\n",
    "        A=self.X\n",
    "        for l in range(1,L):\n",
    "            Z = tf.add(tf.matmul(A,tf.transpose(self.parameters['W'+str(l)])),self.parameters['b'+str(l)])\n",
    "            if l < (L-1):  # activation for hidden layer\n",
    "                A = tf.nn.relu(Z)\n",
    "            else:  # activation for output layer\n",
    "                A1 = Z[:,:self.nz] #sigma when initialize, set weights values very small using factor=0.002\n",
    "                A2 = Z[:,self.nz:2 * self.nz] #mu\n",
    "                A3 = Z[:,2 * self.nz:] #h\n",
    "        return A1, A2, A3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# decoder neural network - take as input latent variable z, output reconstructed x_hat\n",
    "class Decoder(object):\n",
    "    def __init__(self,inputs,hidden_dim):\n",
    "        self.X=inputs\n",
    "        self.nz = inputs.get_shape()[1]\n",
    "        self.layer_dims = [self.nz, hidden_dim,hidden_dim, 784]\n",
    "        with tf.variable_scope('decoder'):\n",
    "            self.parameters = self._initialize_parameters()\n",
    "            \n",
    "    def _initialize_parameters(self):\n",
    "        L=len(self.layer_dims)\n",
    "        parameters={}\n",
    "        for l in range(1,L):            \n",
    "            parameters['W' + str(l)] = tf.get_variable('W'+str(l),[self.layer_dims[l],self.layer_dims[l-1]],initializer=tf.contrib.layers.variance_scaling_initializer())\n",
    "            parameters['b' + str(l)] = tf.get_variable('b'+str(l),self.layer_dims[l],initializer=tf.zeros_initializer())\n",
    "        return parameters\n",
    "    \n",
    "    def decoding(self):\n",
    "        L=len(self.layer_dims)    \n",
    "        A=self.X\n",
    "        for l in range(1,L):\n",
    "            Z = tf.add(tf.matmul(A,tf.transpose(self.parameters['W'+str(l)])),self.parameters['b'+str(l)])\n",
    "            if l<L-1:\n",
    "                A = tf.nn.relu(Z)\n",
    "            else:\n",
    "                A = tf.nn.sigmoid(Z)   \n",
    "\n",
    "    \n",
    "        return A "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# autoregressive neural network - take as input z_(t-1) and h, output mean and sigma of z_(t)\n",
    "class AutoregNN(object):\n",
    "    def __init__(self,input_z,input_h,layer_dims,idx,verbose = False):\n",
    "        self.Z = input_z\n",
    "        self.h = input_h\n",
    "        self.z_dim = input_z.get_shape().as_list()[1]\n",
    "        self.h_dim = input_h.get_shape().as_list()[1]\n",
    "        self.D= self.z_dim\n",
    "        self.layer_dims=layer_dims # z_dim + h_dim, arbitrary, arbitrary, z_dim + z_dim\n",
    "        assert self.layer_dims[0] == self.z_dim + self.h_dim\n",
    "        assert self.layer_dims[-1] == self.z_dim * 2\n",
    "        self.verbose = verbose\n",
    "        \n",
    "        with tf.variable_scope('autoregressive_nn'+str(idx)):\n",
    "            self.parameters=self._initialize_parameters()\n",
    "        with tf.name_scope('autoregressive_nn'+str(idx)):\n",
    "            self.mask=self.build_mask() \n",
    "        \n",
    "    #initialize parameters\n",
    "    def _initialize_parameters(self):\n",
    "        L = len(self.layer_dims)\n",
    "        parameters={}\n",
    "        for l in range(1,L-1):\n",
    "            \n",
    "            parameters['W' + str(l)] = tf.get_variable('W'+str(l),[self.layer_dims[l],self.layer_dims[l-1]],initializer=tf.contrib.layers.variance_scaling_initializer(factor=0.2))\n",
    "            parameters['b' + str(l)] = tf.get_variable('b'+str(l),self.layer_dims[l],initializer=tf.zeros_initializer()) \n",
    "        \n",
    "        for l in [L-1]:\n",
    "            parameters['W' + str(l)] = tf.get_variable('W'+str(l),[self.layer_dims[l],self.layer_dims[l-1]],initializer=tf.contrib.layers.variance_scaling_initializer())\n",
    "            parameters['b' + str(l)] = tf.get_variable('b'+str(l),self.layer_dims[l],initializer=tf.constant_initializer(2.)) \n",
    "\n",
    "        \n",
    "        #for direct connection\n",
    "        parameters['WA']=tf.get_variable('WA',[self.layer_dims[L-1],self.z_dim],initializer=tf.contrib.layers.xavier_initializer())\n",
    "        parameters['bA']=tf.get_variable('bA',self.layer_dims[L-1],initializer=tf.constant_initializer(1.))\n",
    "        \n",
    "        return parameters \n",
    "\n",
    "    # build mask    \n",
    "    def single_mask(self, m_h, l, Isoutput=False):\n",
    "        if l==0:\n",
    "            raise ValueError('No mask for W0')        \n",
    "        if Isoutput==False:\n",
    "            M=np.array([[m_h[l][k]>=m_h[l-1][d] for d in range(self.layer_dims[l-1])] for k in range(self.layer_dims[l])],dtype=int)\n",
    "        if Isoutput==True:\n",
    "            M=np.array([[m_h[l][k]>m_h[l-1][d] for d in range(self.layer_dims[l-1])] for k in range(self.layer_dims[l])],dtype=int)\n",
    "        if self.verbose:\n",
    "            print('M%d:'%l + '{}'.format(M))\n",
    "        return M    \n",
    "    \n",
    "    def build_mask(self):\n",
    "        L=len(self.layer_dims)\n",
    "        mask={}        \n",
    "        #create mask number of each unit\n",
    "        input_sel_z = np.arange(0,self.z_dim)\n",
    "        input_sel_h = np.zeros(self.h_dim)-1 # a little trick to let input h be fully connected to first hidden layer \n",
    "                                    # it shouldn't have mask, but we set the mask to be True = set input_sel_h smaller than any mask number\n",
    "        input_sel = np.concatenate((input_sel_z,input_sel_h))\n",
    "        output_sel = np.concatenate((input_sel_z, input_sel_z))\n",
    "        m_h=[input_sel]\n",
    "        #np.random.seed(0)\n",
    "        for l in range(1,L-1):\n",
    "            m_h.append(np.random.randint(self.D-1,size=self.layer_dims[l]))\n",
    "        m_h.append(output_sel)\n",
    "        if self.verbose:\n",
    "            print('m_h: {}'.format(m_h))\n",
    "        \n",
    "        for l in range(1,L-1):\n",
    "            M=self.single_mask(m_h,l,Isoutput=False)\n",
    "            mask['M'+str(l)]=tf.constant(M,dtype=tf.float32,name='M'+str(l))\n",
    "        for l in range(L-1,L):           \n",
    "            M=self.single_mask(m_h,l,Isoutput=True)\n",
    "            mask['M'+str(l)]=tf.constant(M,dtype=tf.float32,name='M'+str(l))\n",
    "    \n",
    "        # for direct connection\n",
    "        input_sel_direct = input_sel_z\n",
    "        output_sel_direct = np.concatenate((input_sel_z, input_sel_z))\n",
    "        M=np.array([[i<j for i in input_sel_direct] for j in output_sel_direct],dtype=int)\n",
    "        if self.verbose:\n",
    "            print('MA: {}'.format(M))\n",
    "        mask['MA']=tf.constant(M,dtype=tf.float32,name='MA')                      \n",
    "        \n",
    "        return mask\n",
    "    \n",
    "    # forward compute\n",
    "    def iaf(self):\n",
    "        L=len(self.layer_dims)    \n",
    "        A_in=tf.concat((self.Z,self.h),axis = 1)\n",
    "        A = A_in\n",
    "        for l in range(1,L):\n",
    "            masked_W=np.dot(self.parameters['W'+str(l)], self.mask['M'+str(l)])\n",
    "            b=self.parameters['b'+str(l)]\n",
    "            Z=tf.add(tf.matmul(A,tf.transpose(masked_W)),b)\n",
    "            if l < (L-1):  # activation for hidden layer\n",
    "                A=tf.nn.relu(Z)\n",
    "            else:  # activation for output layer\n",
    "                A=Z \n",
    "                \n",
    "        #for direct conection\n",
    "        masked_W= np.dot(self.parameters['WA'], self.mask['MA'])\n",
    "        A_direct=tf.add(tf.matmul(self.Z,tf.transpose(masked_W)),self.parameters['bA'])\n",
    "        \n",
    "        A = tf.add(A,A_direct)\n",
    "        A1 = A[:,self.z_dim:] #s\n",
    "        A2 = A[:,:self.z_dim] #m\n",
    "\n",
    "        \n",
    "        return A1,A2,self.parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# built model\n",
    "tf.reset_default_graph()\n",
    "with tf.name_scope('inputs'):\n",
    "    X = tf.placeholder(\"float\", shape=[None, 784]) \n",
    "\n",
    "with tf.name_scope('encoding_stage'):\n",
    "    encoder=Encoder(inputs=X,hidden_dim=500,z_dim=2,h_dim=10)\n",
    "    z_log_sigma_0,z_mu_0,h=encoder.encoding()\n",
    "    epsilon = tf.random_normal(tf.shape(z_mu_0),0,1,dtype=tf.float32)\n",
    "    Z = np.dot(tf.exp(z_log_sigma_0/2) , epsilon) + z_mu_0\n",
    "\n",
    "with tf.name_scope('iaf_stage'):\n",
    "    z_sigmas=[]\n",
    "    reg_loss = 0.\n",
    "    for t in range(4):\n",
    "        reverse = t>0\n",
    "        if reverse:\n",
    "            Z = Z[:,::-1]\n",
    "        IAF = AutoregNN(input_z=Z,input_h=h,layer_dims=[12,100,100,4],idx=t)\n",
    "        m,s,iparam = IAF.iaf()\n",
    "        z_sigma = tf.sigmoid(s)\n",
    "        Z = np.dot(z_sigma , Z) + np.dot(1 - z_sigma ,  m)\n",
    "        z_sigmas.append(z_sigma)\n",
    "        reg_loss += tf.nn.l2_loss(iparam[\"W1\"])+tf.nn.l2_loss(iparam[\"W2\"])+tf.nn.l2_loss(iparam[\"W3\"])+tf.nn.l2_loss(iparam[\"WA\"])\n",
    "with tf.name_scope('decoding_stage'):\n",
    "    decoder=Decoder(inputs=Z,hidden_dim=500)\n",
    "    x_mu_hat = decoder.decoding()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# vae - iaf negative evidence lower bound - based on monte carlo sampling\n",
    "def neg_elbo(X,x_mu_hat,Z):\n",
    "    # log p(x|z_t)\n",
    "    log_px = -tf.reduce_sum(X * tf.log(1e-10 + x_mu_hat) + (1-X) * tf.log(1e-10 + 1 - x_mu_hat), axis=-1)   \n",
    "    \n",
    "    # log p(z_t)\n",
    "    log_pz = -tf.reduce_sum(0.5*Z**2, axis=-1)\n",
    "  \n",
    "    # log q(z_t|x)\n",
    "    det=np.zeros_like(z_sigmas[0])\n",
    "    for item in z_sigmas:\n",
    "        det+=tf.log(item)    \n",
    "    log_qz = -tf.reduce_sum(0.5*epsilon**2 + 0.5*z_log_sigma_0+det,axis=-1) \n",
    "    \n",
    "    return tf.reduce_mean(log_px+log_qz-log_pz) + 0.01*reg_loss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# training\n",
    "def training(num_epochs = 1000, batch_size = 32, log_every=10,learning_rate = 0.001, verbose = False):\n",
    "\n",
    "    with tf.name_scope('training_loss'):\n",
    "        loss = neg_elbo(X,x_mu_hat,Z)\n",
    "    optimizer = tf.train.AdamOptimizer(learning_rate=learning_rate).minimize(loss)\n",
    "    \n",
    "    # training\n",
    "    saver = tf.train.Saver()\n",
    "    with tf.Session() as sess:\n",
    "        sess.run(tf.global_variables_initializer())\n",
    "\n",
    "        for epoch in range(num_epochs):\n",
    "            avg_cost = 0.\n",
    "            num_batch = int(n_samples / batch_size)\n",
    "            \n",
    "            for i in range(num_batch):\n",
    "                mini_batch, mini_label = mnist.train.next_batch(batch_size)\n",
    "                _,temp_cost,z_values = sess.run([optimizer, loss,z_log_sigma_0], feed_dict={X: mini_batch})\n",
    "                avg_cost += temp_cost / n_samples * batch_size\n",
    "            \n",
    "            if verbose and epoch % log_every ==0:\n",
    "                save_path = saver.save(sess, \"iafmnist/iafmnist.ckpt\")\n",
    "                print('%d-th train ' % (epoch) + 'cost:{} {}'.format(avg_cost,np.mean(z_values)))\n",
    "    return None"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0-th train cost:218.00868685635675 -3.148116111755371\n",
      "1-th train cost:177.21460438121457 -3.7423784732818604\n",
      "2-th train cost:170.39865417480456 -4.151371955871582\n",
      "3-th train cost:166.68361849698152 -4.4144110679626465\n",
      "4-th train cost:164.490971291282 -4.456427574157715\n",
      "5-th train cost:162.7261896306818 -4.5876641273498535\n",
      "6-th train cost:161.5760364879261 -4.645597457885742\n",
      "7-th train cost:160.05920731977986 -4.8163957595825195\n",
      "8-th train cost:159.24323042436086 -5.0451459884643555\n",
      "9-th train cost:158.35958901145227 -4.995668411254883\n",
      "10-th train cost:157.77385620117175 -5.127269268035889\n",
      "11-th train cost:157.1939127419212 -5.183229923248291\n",
      "12-th train cost:156.6668263383345 -5.231800556182861\n",
      "13-th train cost:156.00657986727632 -5.251016616821289\n",
      "14-th train cost:155.6352054110441 -5.339222431182861\n",
      "15-th train cost:155.12962063876066 -5.339994430541992\n",
      "16-th train cost:154.79751253995028 -5.251043796539307\n",
      "17-th train cost:154.63506841486156 -5.331535816192627\n",
      "18-th train cost:154.29300365101216 -5.437460899353027\n",
      "19-th train cost:153.77996848366482 -5.394901752471924\n",
      "20-th train cost:153.99494257146657 -5.34739875793457\n",
      "21-th train cost:153.37770457874652 -5.516284942626953\n",
      "22-th train cost:153.42689608487214 -5.386690616607666\n",
      "23-th train cost:153.01361966219815 -5.425744533538818\n",
      "24-th train cost:153.0178480668501 -5.4503936767578125\n",
      "25-th train cost:152.75864329944955 -5.487807750701904\n",
      "26-th train cost:152.71899097789415 -5.502547740936279\n",
      "27-th train cost:152.70799588290132 -5.4847259521484375\n",
      "28-th train cost:152.14240994540134 -5.453840732574463\n",
      "29-th train cost:152.49985101873227 -5.366496562957764\n",
      "30-th train cost:151.94742370605474 -5.484867095947266\n",
      "31-th train cost:152.07214760520245 -5.507652759552002\n",
      "32-th train cost:151.91379516601555 -5.4517436027526855\n",
      "33-th train cost:151.49132263183597 -5.436618804931641\n",
      "34-th train cost:151.18110340465208 -5.499475002288818\n",
      "35-th train cost:151.69848266601565 -5.437005519866943\n",
      "36-th train cost:151.22468544699925 -5.580052375793457\n",
      "37-th train cost:151.2774519486859 -5.377370834350586\n",
      "38-th train cost:151.13408269708802 -5.490818500518799\n",
      "39-th train cost:150.92049360795457 -5.413520336151123\n",
      "40-th train cost:151.19689636230467 -5.588696479797363\n",
      "41-th train cost:150.89022444291552 -5.401366710662842\n",
      "42-th train cost:150.5031813188034 -5.525547981262207\n",
      "43-th train cost:150.65826948686077 -5.490979194641113\n",
      "44-th train cost:150.79020652077418 -5.443892955780029\n",
      "45-th train cost:150.5262726384943 -5.320644378662109\n",
      "46-th train cost:150.63042147549717 -5.433286190032959\n",
      "47-th train cost:150.49196183638134 -5.431879997253418\n",
      "48-th train cost:150.33475430575274 -5.423576831817627\n",
      "49-th train cost:150.4704508278588 -5.351459980010986\n",
      "50-th train cost:150.38993996360085 -5.40809440612793\n",
      "51-th train cost:150.15818481445308 -5.409157752990723\n",
      "52-th train cost:150.07431352095168 -5.403265476226807\n",
      "53-th train cost:149.97513305664063 -5.398151874542236\n",
      "54-th train cost:150.0950326260653 -5.357182502746582\n",
      "55-th train cost:149.9895684259588 -5.373350143432617\n",
      "56-th train cost:149.77352905273446 -5.306253433227539\n",
      "57-th train cost:149.79822104714154 -5.358694076538086\n",
      "58-th train cost:150.54006963556463 -5.324352264404297\n",
      "59-th train cost:150.39377774325294 -5.342655181884766\n",
      "60-th train cost:150.0738071511008 -5.2726006507873535\n",
      "61-th train cost:149.57082330877142 -5.296346664428711\n",
      "62-th train cost:150.47425603693176 -5.391776084899902\n",
      "63-th train cost:150.1461385276103 -5.2581400871276855\n",
      "64-th train cost:150.8389983021129 -5.386761665344238\n",
      "65-th train cost:151.16815063476568 -5.204340934753418\n",
      "66-th train cost:150.32417402787644 -5.322531223297119\n",
      "67-th train cost:149.94955344460212 -5.254387378692627\n",
      "68-th train cost:149.76514015891323 -5.221478462219238\n",
      "69-th train cost:150.29854087136007 -5.411254405975342\n",
      "70-th train cost:149.7192113148085 -5.346751689910889\n",
      "71-th train cost:149.44174072265633 -5.374781608581543\n",
      "72-th train cost:149.94559753417968 -5.199875354766846\n",
      "73-th train cost:149.465121681907 -5.330209732055664\n",
      "74-th train cost:149.63794877485788 -5.259869575500488\n",
      "75-th train cost:149.86442915482945 -5.260986328125\n",
      "76-th train cost:149.46865656072433 -5.220213413238525\n",
      "77-th train cost:149.4932753129439 -5.298486232757568\n",
      "78-th train cost:149.14706259987568 -5.299930572509766\n",
      "79-th train cost:149.15507651589132 -5.257238388061523\n",
      "80-th train cost:149.02199157714844 -5.284868240356445\n",
      "81-th train cost:149.40133128773078 -5.1348876953125\n",
      "82-th train cost:149.3755933726918 -5.142019271850586\n",
      "83-th train cost:149.20410483620387 -5.18563985824585\n",
      "84-th train cost:149.3147068647905 -5.264022350311279\n",
      "85-th train cost:148.849286887429 -5.227761268615723\n",
      "86-th train cost:149.59405184659096 -5.189188003540039\n",
      "87-th train cost:149.21876675692468 -5.084738731384277\n",
      "88-th train cost:148.96467545942838 -5.11662483215332\n",
      "89-th train cost:149.0113619162819 -5.094388961791992\n",
      "90-th train cost:148.8122929798473 -5.191395282745361\n",
      "91-th train cost:150.64968838778412 -5.1755523681640625\n",
      "92-th train cost:150.0482855779475 -5.1599273681640625\n",
      "93-th train cost:150.2085540216619 -5.301529407501221\n",
      "94-th train cost:150.2383895596591 -5.082477569580078\n",
      "95-th train cost:149.77007673783731 -5.073733329772949\n",
      "96-th train cost:148.71324462890632 -5.000961780548096\n",
      "97-th train cost:149.13094843084176 -5.066928386688232\n",
      "98-th train cost:149.47067099831318 -4.964327335357666\n",
      "99-th train cost:148.97431690562854 -4.994226455688477\n",
      "100-th train cost:148.88064380992537 -5.125198841094971\n",
      "101-th train cost:148.59165027965207 -5.119596004486084\n",
      "102-th train cost:149.09271007191055 -5.056618690490723\n",
      "103-th train cost:148.94250016645958 -4.866429328918457\n",
      "104-th train cost:148.82522477583453 -4.9097819328308105\n",
      "105-th train cost:148.88070356889205 -4.960266590118408\n",
      "106-th train cost:149.0469512939452 -4.984337329864502\n",
      "107-th train cost:148.31326804421164 -4.963067531585693\n",
      "108-th train cost:149.0629521595349 -5.032447814941406\n",
      "109-th train cost:148.86487199263146 -4.784565448760986\n",
      "110-th train cost:149.03024763627488 -5.04065465927124\n",
      "111-th train cost:149.60824679287998 -4.969454288482666\n",
      "112-th train cost:148.87415932395248 -4.896158218383789\n",
      "113-th train cost:148.33932739257804 -4.916331768035889\n",
      "114-th train cost:148.65714194557876 -4.970778942108154\n",
      "115-th train cost:149.11350946599785 -4.777448654174805\n",
      "116-th train cost:149.06110129616482 -4.824072360992432\n",
      "117-th train cost:148.64318093039785 -4.984994888305664\n",
      "118-th train cost:149.36125893332743 -4.900351047515869\n",
      "119-th train cost:148.60176591352982 -4.89777946472168\n",
      "120-th train cost:148.42704673073507 -4.877147197723389\n",
      "121-th train cost:148.67414966930042 -4.619448184967041\n",
      "122-th train cost:148.70664495294747 -4.81134557723999\n",
      "123-th train cost:148.43473827015262 -4.937173843383789\n",
      "124-th train cost:148.31617220791912 -4.862160682678223\n",
      "125-th train cost:148.13473743785508 -4.945007801055908\n",
      "126-th train cost:148.54650312943886 -4.751020908355713\n",
      "127-th train cost:148.0456228360263 -4.814948081970215\n",
      "128-th train cost:148.59420659845526 -4.8047099113464355\n",
      "129-th train cost:148.27643232865765 -4.7900590896606445\n",
      "130-th train cost:148.0466166548296 -4.86168098449707\n",
      "131-th train cost:148.2147424316406 -4.768293857574463\n",
      "132-th train cost:148.12525446111508 -4.552240371704102\n",
      "133-th train cost:148.2561925714666 -4.713444232940674\n",
      "134-th train cost:147.98688393332736 -4.714982986450195\n",
      "135-th train cost:148.35683066628192 -4.834035873413086\n",
      "136-th train cost:148.00548633922236 -4.678591728210449\n",
      "137-th train cost:148.05735212846236 -4.811198711395264\n",
      "138-th train cost:148.31451748934663 -4.817802906036377\n",
      "139-th train cost:148.91297707297585 -4.734835624694824\n",
      "140-th train cost:149.4780621337891 -4.730052471160889\n",
      "141-th train cost:149.91196311257107 -4.7744269371032715\n",
      "142-th train cost:148.9783256392045 -4.744291305541992\n",
      "143-th train cost:148.46533425071016 -4.565881252288818\n",
      "144-th train cost:148.78719665527336 -4.492016792297363\n",
      "145-th train cost:148.5031078546698 -4.726830959320068\n",
      "146-th train cost:147.98058532714842 -4.7892537117004395\n",
      "147-th train cost:148.33087463378908 -4.727622032165527\n",
      "148-th train cost:147.66814686168328 -4.759475231170654\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "149-th train cost:148.2402505215732 -4.5843987464904785\n",
      "150-th train cost:147.6915802001953 -4.708986282348633\n",
      "151-th train cost:148.010614790483 -4.470135688781738\n",
      "152-th train cost:147.7234923761541 -4.610180854797363\n",
      "153-th train cost:148.49032792524878 -4.685436248779297\n",
      "154-th train cost:148.28733098810372 -4.493285655975342\n",
      "155-th train cost:148.2677935791015 -4.549884796142578\n",
      "156-th train cost:147.96915399724796 -4.558027267456055\n",
      "157-th train cost:148.59637867320666 -4.663832187652588\n",
      "158-th train cost:148.18778592196375 -4.629491806030273\n",
      "159-th train cost:148.53308116566043 -4.473679065704346\n",
      "160-th train cost:148.37379161487934 -4.489471435546875\n",
      "161-th train cost:147.67511596679694 -4.420216083526611\n",
      "162-th train cost:147.89278015136722 -4.463669300079346\n",
      "163-th train cost:147.92621626420453 -4.512628078460693\n",
      "164-th train cost:147.58342862215912 -4.593786716461182\n",
      "165-th train cost:147.55684093128548 -4.506802558898926\n",
      "166-th train cost:147.57333257501773 -4.455812454223633\n",
      "167-th train cost:148.15131819291543 -4.392423152923584\n",
      "168-th train cost:147.88833435058584 -4.387714385986328\n",
      "169-th train cost:148.27745716441765 -4.481260776519775\n",
      "170-th train cost:147.8607307572798 -4.563187122344971\n",
      "171-th train cost:147.61767589222302 -4.430404186248779\n",
      "172-th train cost:147.95582630504262 -4.431612014770508\n",
      "173-th train cost:147.82004538796173 -4.368283271789551\n",
      "174-th train cost:147.75499339710586 -4.329284191131592\n",
      "175-th train cost:147.4350632546165 -4.484237194061279\n",
      "176-th train cost:147.78340393066398 -4.494898319244385\n",
      "177-th train cost:148.09717934348367 -4.401096343994141\n",
      "178-th train cost:147.93553777521305 -4.346109390258789\n",
      "179-th train cost:147.24770208185367 -4.427379131317139\n",
      "180-th train cost:147.66093627929703 -4.404469966888428\n",
      "181-th train cost:147.3503101695668 -4.4295430183410645\n",
      "182-th train cost:147.59531438654108 -4.384237766265869\n",
      "183-th train cost:149.09708490545094 -4.311746597290039\n",
      "184-th train cost:148.4479565429687 -4.274155139923096\n",
      "185-th train cost:147.6205442671342 -4.294784069061279\n",
      "186-th train cost:147.33783380681814 -4.239249229431152\n",
      "187-th train cost:147.54391163219103 -4.175917148590088\n",
      "188-th train cost:147.80834555886008 -4.439558029174805\n",
      "189-th train cost:147.7362599320845 -4.174889087677002\n",
      "190-th train cost:147.7679172585227 -4.308653354644775\n",
      "191-th train cost:147.69724775834527 -4.15174674987793\n",
      "192-th train cost:147.8594330388851 -4.192262649536133\n",
      "193-th train cost:147.94096463290117 -4.254464626312256\n",
      "194-th train cost:147.5654527143998 -4.369325160980225\n",
      "195-th train cost:147.16470126065343 -4.19225549697876\n",
      "196-th train cost:147.4538246848366 -4.128523349761963\n",
      "197-th train cost:147.30056768243966 -4.246490001678467\n",
      "198-th train cost:148.1076438765093 -4.042891502380371\n",
      "199-th train cost:148.4520670387961 -4.150082111358643\n",
      "200-th train cost:147.71198547363286 -4.083271503448486\n",
      "201-th train cost:147.55283020019536 -3.9864606857299805\n",
      "202-th train cost:147.44792519309297 -4.147150993347168\n",
      "203-th train cost:147.15345958362934 -4.096502780914307\n",
      "204-th train cost:147.9212267511541 -4.014430046081543\n",
      "205-th train cost:147.9532735928622 -4.023240566253662\n",
      "206-th train cost:148.62319069602262 -3.942615270614624\n",
      "207-th train cost:147.64442255193543 -4.0149359703063965\n",
      "208-th train cost:147.192226229581 -4.142016410827637\n",
      "209-th train cost:147.17427495783028 -4.041334629058838\n",
      "210-th train cost:147.07565557306458 -4.0866594314575195\n",
      "211-th train cost:147.21082375266343 -3.953023910522461\n",
      "212-th train cost:146.93706681685018 -4.005169868469238\n",
      "213-th train cost:147.25410533558247 -4.073240280151367\n",
      "214-th train cost:147.02770696466627 -4.025442600250244\n",
      "215-th train cost:147.1567099276457 -4.050473690032959\n",
      "216-th train cost:147.12900473854768 -4.012850761413574\n",
      "217-th train cost:147.0966890092329 -4.1098952293396\n",
      "218-th train cost:146.92874689275558 -3.9375247955322266\n",
      "219-th train cost:146.8290947931463 -3.929891586303711\n",
      "220-th train cost:147.23732921253557 -4.015653133392334\n",
      "221-th train cost:147.09337552157317 -3.86971378326416\n",
      "222-th train cost:146.795815929066 -4.007357597351074\n",
      "223-th train cost:146.77703535600153 -4.180009841918945\n",
      "224-th train cost:146.88927090731528 -3.721996545791626\n",
      "225-th train cost:147.15683455033727 -3.9295668601989746\n",
      "226-th train cost:147.04993408203134 -3.947141170501709\n",
      "227-th train cost:146.8781231134587 -3.8450396060943604\n",
      "228-th train cost:146.88190224387432 -3.940875291824341\n",
      "229-th train cost:147.11161066228692 -3.9102861881256104\n",
      "230-th train cost:146.63517783425064 -3.8126142024993896\n",
      "231-th train cost:146.96200062144877 -3.9783380031585693\n",
      "232-th train cost:146.7512339643999 -3.6600029468536377\n",
      "233-th train cost:146.66994013006047 -3.9707446098327637\n",
      "234-th train cost:146.69324879039414 -3.806675434112549\n",
      "235-th train cost:147.12182056773798 -3.67610239982605\n",
      "236-th train cost:147.68653969504612 -3.6920037269592285\n",
      "237-th train cost:146.65127291592685 -3.8254408836364746\n",
      "238-th train cost:146.84832691539415 -3.8949451446533203\n",
      "239-th train cost:146.80791293057538 -3.86361026763916\n",
      "240-th train cost:147.1884684059836 -3.835300922393799\n",
      "241-th train cost:147.49018987482245 -3.896439790725708\n",
      "242-th train cost:148.22977666681467 -3.920867919921875\n",
      "243-th train cost:147.86706204501058 -3.815214157104492\n",
      "244-th train cost:147.06392922141342 -3.8546972274780273\n",
      "245-th train cost:146.82020940607248 -3.895333766937256\n",
      "246-th train cost:146.87012295809663 -3.773083448410034\n",
      "247-th train cost:147.76181207830246 -3.8176679611206055\n",
      "248-th train cost:147.60092085404824 -3.590891122817993\n",
      "249-th train cost:146.98086803089498 -3.81540584564209\n",
      "250-th train cost:146.79464483087713 -3.758906364440918\n",
      "251-th train cost:147.01505748402005 -3.714761972427368\n",
      "252-th train cost:146.81546852805405 -3.7432985305786133\n",
      "253-th train cost:146.47168240633886 -3.678366184234619\n",
      "254-th train cost:146.67382202148434 -3.4756481647491455\n",
      "255-th train cost:146.8731574596058 -3.619494915008545\n",
      "256-th train cost:146.41205011541186 -3.6751461029052734\n",
      "257-th train cost:147.1883703058417 -3.708811044692993\n",
      "258-th train cost:146.3356307705966 -3.7296924591064453\n",
      "259-th train cost:146.54602228338075 -3.6786184310913086\n",
      "260-th train cost:146.98036282626057 -3.7330081462860107\n",
      "261-th train cost:146.8679877263849 -3.685955762863159\n",
      "262-th train cost:146.76423672762786 -3.7014238834381104\n",
      "263-th train cost:146.7925415039064 -3.687054395675659\n",
      "264-th train cost:146.73564891468402 -3.647324323654175\n",
      "265-th train cost:146.7809366677024 -3.5755491256713867\n",
      "266-th train cost:146.50382057883513 -3.6805059909820557\n",
      "267-th train cost:146.37811706542965 -3.663994073867798\n",
      "268-th train cost:146.80440479625358 -3.6412172317504883\n",
      "269-th train cost:147.5827069091798 -3.7145421504974365\n",
      "270-th train cost:146.83016507235436 -3.678553581237793\n",
      "271-th train cost:146.86964227849788 -3.464345693588257\n",
      "272-th train cost:146.86178250399504 -3.666792631149292\n",
      "273-th train cost:146.8244189453126 -3.7026562690734863\n",
      "274-th train cost:147.36189713911577 -3.701727867126465\n",
      "275-th train cost:146.7142833362926 -3.5000836849212646\n",
      "276-th train cost:146.38865855823857 -3.4725730419158936\n",
      "277-th train cost:146.8134109219639 -3.517531156539917\n",
      "278-th train cost:146.64365728204908 -3.6356797218322754\n",
      "279-th train cost:146.6457299249823 -3.708362102508545\n",
      "280-th train cost:146.54852084073156 -3.4943785667419434\n",
      "281-th train cost:146.03116283069963 -3.6716692447662354\n",
      "282-th train cost:146.2058098255505 -3.548905372619629\n",
      "283-th train cost:146.33939075816755 -3.514937162399292\n",
      "284-th train cost:146.59731584028762 -3.5667362213134766\n",
      "285-th train cost:146.51563970392394 -3.4901821613311768\n",
      "286-th train cost:146.24756658380684 -3.509000778198242\n",
      "287-th train cost:146.5343397105824 -3.433548927307129\n",
      "288-th train cost:146.4689368785511 -3.5312283039093018\n",
      "289-th train cost:146.13817194158372 -3.565077304840088\n",
      "290-th train cost:147.18412753018472 -3.4753611087799072\n",
      "291-th train cost:147.44434342817829 -3.4831578731536865\n",
      "292-th train cost:146.963628096147 -3.531254768371582\n",
      "293-th train cost:146.56765475186435 -3.640779972076416\n",
      "294-th train cost:146.76913196910508 -3.493467330932617\n",
      "295-th train cost:146.6870432350853 -3.4147539138793945\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "296-th train cost:147.30108908913343 -3.5591986179351807\n",
      "297-th train cost:146.8197939231179 -3.470412254333496\n",
      "298-th train cost:146.8741386274857 -3.379929780960083\n",
      "299-th train cost:147.67090387517763 -3.465153694152832\n",
      "300-th train cost:148.37411132812497 -3.4332849979400635\n",
      "301-th train cost:147.29272133567108 -3.5252137184143066\n",
      "302-th train cost:147.3609674627131 -3.5417914390563965\n",
      "303-th train cost:147.28299926757802 -3.422255754470825\n",
      "304-th train cost:146.9980691805753 -3.4581713676452637\n",
      "305-th train cost:146.7411431329901 -3.4046173095703125\n",
      "306-th train cost:147.6541283069957 -3.390043258666992\n",
      "307-th train cost:147.03337907270947 -3.403937578201294\n",
      "308-th train cost:147.0003455699574 -3.438666343688965\n",
      "309-th train cost:147.61058455033742 -3.2912676334381104\n",
      "310-th train cost:147.10892017711285 -3.3697845935821533\n",
      "311-th train cost:147.09236722079189 -3.2829370498657227\n",
      "312-th train cost:146.77244262695308 -3.3464438915252686\n",
      "313-th train cost:146.1830881014736 -3.291743278503418\n",
      "314-th train cost:146.92778608842326 -3.3149971961975098\n",
      "315-th train cost:146.783978548917 -3.409710645675659\n",
      "316-th train cost:146.45784806684998 -3.3144288063049316\n",
      "317-th train cost:146.73873268821023 -3.3132131099700928\n",
      "318-th train cost:146.90772660688924 -3.4515650272369385\n",
      "319-th train cost:146.66228132768103 -3.3646087646484375\n",
      "320-th train cost:146.7400364546341 -3.32224702835083\n",
      "321-th train cost:147.19812327991832 -3.277648687362671\n",
      "322-th train cost:146.56390275435007 -3.3117682933807373\n",
      "323-th train cost:146.31675270774153 -3.4218802452087402\n",
      "324-th train cost:147.06288252397022 -3.407665967941284\n",
      "325-th train cost:147.26429798473015 -3.1026384830474854\n",
      "326-th train cost:146.90877718838777 -3.4000349044799805\n",
      "327-th train cost:146.7334697931462 -3.2799336910247803\n",
      "328-th train cost:146.33783730246785 -3.1560451984405518\n",
      "329-th train cost:146.78036454634233 -3.257458448410034\n",
      "330-th train cost:146.64562117143115 -3.3202831745147705\n",
      "331-th train cost:147.35734391645948 -3.2187652587890625\n",
      "332-th train cost:147.52078740900217 -3.1489901542663574\n",
      "333-th train cost:146.9448684969815 -3.279378890991211\n",
      "334-th train cost:147.2011333118785 -3.0842084884643555\n",
      "335-th train cost:147.23398870294756 -3.2375457286834717\n",
      "336-th train cost:146.8006357088956 -3.320279836654663\n",
      "337-th train cost:146.37533058860092 -3.1276116371154785\n",
      "338-th train cost:146.69301874334164 -3.1764941215515137\n",
      "339-th train cost:146.33294694380334 -3.149444580078125\n",
      "340-th train cost:146.10410466974426 -3.3090293407440186\n",
      "341-th train cost:146.03679493297227 -3.210446834564209\n",
      "342-th train cost:146.30053217107601 -3.075063467025757\n",
      "343-th train cost:146.24379893909807 -3.196303606033325\n",
      "344-th train cost:146.1230955366656 -3.0842764377593994\n",
      "345-th train cost:146.08552173961294 -3.216035842895508\n",
      "346-th train cost:146.18321805086993 -3.183683395385742\n",
      "347-th train cost:146.38102688876077 -3.1915807723999023\n",
      "348-th train cost:146.46075711337002 -3.1013312339782715\n",
      "349-th train cost:145.9372364390981 -3.230823278427124\n",
      "350-th train cost:146.09460160688923 -2.9640872478485107\n",
      "351-th train cost:146.47547141335224 -3.1407997608184814\n",
      "352-th train cost:146.0774046186968 -3.268022060394287\n",
      "353-th train cost:146.24756192294035 -3.238213539123535\n",
      "354-th train cost:146.2160458096591 -2.9382476806640625\n",
      "355-th train cost:146.12950456099068 -3.019073486328125\n",
      "356-th train cost:145.88159079811794 -3.0271904468536377\n",
      "357-th train cost:145.59278198242183 -3.2426276206970215\n",
      "358-th train cost:145.73875571511002 -3.100546360015869\n",
      "359-th train cost:146.63921719637779 -2.901637554168701\n",
      "360-th train cost:146.01397227894176 -2.980027198791504\n",
      "361-th train cost:146.27550548206662 -2.926295757293701\n",
      "362-th train cost:145.880782248757 -2.9779741764068604\n",
      "363-th train cost:145.54066644841973 -2.901257276535034\n",
      "364-th train cost:145.57625127618954 -3.0901148319244385\n",
      "365-th train cost:145.784628129439 -3.0352563858032227\n",
      "366-th train cost:145.84719227183945 -2.9211459159851074\n",
      "367-th train cost:146.1800494384766 -3.0913596153259277\n",
      "368-th train cost:145.71540338689633 -2.9829185009002686\n",
      "369-th train cost:146.06292119806466 -3.0992119312286377\n",
      "370-th train cost:145.8020444003019 -2.908076763153076\n",
      "371-th train cost:145.61062228116123 -3.032010793685913\n",
      "372-th train cost:146.32222339976914 -2.9658682346343994\n",
      "373-th train cost:145.9446853915127 -3.1368725299835205\n",
      "374-th train cost:145.53247713955977 -3.0291552543640137\n",
      "375-th train cost:145.592929132635 -3.0401313304901123\n",
      "376-th train cost:145.72234996448856 -3.0538275241851807\n",
      "377-th train cost:145.55412420099424 -2.8867602348327637\n",
      "378-th train cost:146.4496502685546 -2.521254062652588\n",
      "379-th train cost:145.7663806152343 -2.934776544570923\n",
      "380-th train cost:145.63407487349076 -2.889242172241211\n",
      "381-th train cost:145.7443262828479 -2.819981575012207\n",
      "382-th train cost:146.00979797363283 -3.1294450759887695\n",
      "383-th train cost:145.32896467729051 -2.9511115550994873\n",
      "384-th train cost:145.36131536310364 -3.05900502204895\n",
      "385-th train cost:145.5526003750888 -2.898855686187744\n",
      "386-th train cost:145.77928477894162 -2.960671901702881\n",
      "387-th train cost:145.5503127774325 -2.9165074825286865\n",
      "388-th train cost:145.37652393687853 -3.0468108654022217\n",
      "389-th train cost:145.75906749378544 -2.7871642112731934\n",
      "390-th train cost:145.55375216397368 -2.8680830001831055\n",
      "391-th train cost:145.50743508078835 -2.8998663425445557\n",
      "392-th train cost:145.2049740323154 -2.8445959091186523\n",
      "393-th train cost:145.6626298939098 -2.858105182647705\n",
      "394-th train cost:145.32606606223374 -3.0647189617156982\n",
      "395-th train cost:145.53507901278417 -2.9214839935302734\n",
      "396-th train cost:145.49306868119666 -2.9193506240844727\n",
      "397-th train cost:145.71752507990064 -2.8860013484954834\n",
      "398-th train cost:145.5236062344637 -2.954592227935791\n",
      "399-th train cost:145.61192626953135 -2.88150691986084\n",
      "400-th train cost:145.36883900035517 -2.7839930057525635\n",
      "401-th train cost:145.433653009588 -2.9649646282196045\n",
      "402-th train cost:145.71775346235796 -2.8392410278320312\n",
      "403-th train cost:145.6155838844992 -2.675882577896118\n",
      "404-th train cost:145.79096085981885 -2.7920279502868652\n",
      "405-th train cost:145.50702131791553 -2.918649673461914\n",
      "406-th train cost:146.11327808726918 -2.9573733806610107\n",
      "407-th train cost:145.7348322642933 -2.819671630859375\n",
      "408-th train cost:145.3174339155717 -2.650296926498413\n",
      "409-th train cost:145.3448516845703 -2.9921786785125732\n",
      "410-th train cost:145.45798356489703 -2.703822612762451\n",
      "411-th train cost:145.32980368874289 -2.8441531658172607\n",
      "412-th train cost:145.29072515314272 -2.716520309448242\n",
      "413-th train cost:145.3623473011364 -2.8449816703796387\n",
      "414-th train cost:145.3025902210583 -2.750237464904785\n",
      "415-th train cost:145.41054565429684 -2.616959810256958\n",
      "416-th train cost:145.32687389026984 -2.7228057384490967\n",
      "417-th train cost:145.62109164151275 -2.8489608764648438\n",
      "418-th train cost:145.63069563432168 -2.712810754776001\n",
      "419-th train cost:145.06904873934658 -2.835129976272583\n",
      "420-th train cost:145.16056141246455 -2.675443649291992\n",
      "421-th train cost:145.4159904896128 -2.777998685836792\n",
      "422-th train cost:145.44810164018105 -2.824066400527954\n",
      "423-th train cost:145.08054914994685 -2.704993963241577\n",
      "424-th train cost:145.4327781538529 -2.7306528091430664\n",
      "425-th train cost:145.1519113436611 -2.5597991943359375\n",
      "426-th train cost:145.38847867098713 -2.8944830894470215\n",
      "427-th train cost:145.0189615145596 -2.911829948425293\n",
      "428-th train cost:144.98404590953476 -2.73187255859375\n",
      "429-th train cost:145.06315895774154 -2.5639281272888184\n",
      "430-th train cost:145.50332120028406 -2.7376341819763184\n",
      "431-th train cost:145.03477594549 -2.6292591094970703\n",
      "432-th train cost:145.26908230868264 -2.7906558513641357\n",
      "433-th train cost:145.21785350452785 -2.636946439743042\n",
      "434-th train cost:145.51224376331672 -2.631124258041382\n",
      "435-th train cost:144.99327009721253 -2.565261125564575\n",
      "436-th train cost:145.27222112482247 -2.6965978145599365\n",
      "437-th train cost:145.55048789284444 -2.6802752017974854\n",
      "438-th train cost:145.10105391068905 -2.6463887691497803\n",
      "439-th train cost:144.78761041814627 -2.6604843139648438\n",
      "440-th train cost:145.38941134366132 -2.7789270877838135\n",
      "441-th train cost:145.0872240656073 -2.620683193206787\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "442-th train cost:145.2643988037109 -2.656325578689575\n",
      "443-th train cost:145.37442577015273 -2.6215126514434814\n",
      "444-th train cost:145.57739468661228 -2.677982807159424\n",
      "445-th train cost:145.300122347745 -2.493199586868286\n",
      "446-th train cost:144.90382690429703 -2.6004421710968018\n",
      "447-th train cost:144.98497381036935 -2.680288791656494\n",
      "448-th train cost:144.781675415039 -2.6644115447998047\n",
      "449-th train cost:144.97189641779119 -2.5348474979400635\n",
      "450-th train cost:145.06567238547595 -2.7773194313049316\n",
      "451-th train cost:145.07361477938557 -2.671238899230957\n",
      "452-th train cost:145.39682861328134 -2.7243573665618896\n",
      "453-th train cost:145.2332433527167 -2.7101776599884033\n",
      "454-th train cost:145.80351839932527 -2.522465944290161\n",
      "455-th train cost:145.9169194446911 -2.639166831970215\n",
      "456-th train cost:146.34249983354044 -2.617385149002075\n",
      "457-th train cost:147.60428089488636 -2.698066473007202\n",
      "458-th train cost:147.04690923517396 -2.778419256210327\n",
      "459-th train cost:146.74687045010646 -2.6345250606536865\n",
      "460-th train cost:146.13940163352268 -2.571657419204712\n",
      "461-th train cost:146.3214087468928 -2.616488456726074\n",
      "462-th train cost:145.88542780095887 -2.440964698791504\n",
      "463-th train cost:146.44330455433237 -2.585463762283325\n",
      "464-th train cost:145.7951727294923 -2.590433359146118\n",
      "465-th train cost:145.61712452281606 -2.4768829345703125\n",
      "466-th train cost:146.27163707386367 -2.4559106826782227\n",
      "467-th train cost:145.68886369185014 -2.597724199295044\n",
      "468-th train cost:145.08940734863273 -2.5977678298950195\n",
      "469-th train cost:145.10543196244672 -2.5123534202575684\n",
      "470-th train cost:145.29355857155534 -2.5773303508758545\n",
      "471-th train cost:145.28334006569605 -2.6968445777893066\n",
      "472-th train cost:145.21921325683599 -2.555037260055542\n",
      "473-th train cost:145.2606122381037 -2.588869094848633\n",
      "474-th train cost:145.25641285289413 -2.429561138153076\n",
      "475-th train cost:145.14681335449228 -2.5854499340057373\n",
      "476-th train cost:145.32853493430386 -2.513233184814453\n",
      "477-th train cost:145.4644034645775 -2.4790966510772705\n",
      "478-th train cost:144.95424050071023 -2.571258306503296\n",
      "479-th train cost:145.0332318115234 -2.3378429412841797\n",
      "480-th train cost:144.80559686834167 -2.5066945552825928\n",
      "481-th train cost:145.07155345569947 -2.516700029373169\n",
      "482-th train cost:144.7050790682706 -2.503411293029785\n",
      "483-th train cost:144.97128029563217 -2.498825788497925\n",
      "484-th train cost:144.76130809437146 -2.477091073989868\n",
      "485-th train cost:144.93653481223367 -2.3941869735717773\n",
      "486-th train cost:144.84844121759588 -2.377285957336426\n",
      "487-th train cost:144.75019891912294 -2.361992597579956\n",
      "488-th train cost:144.81941567160868 -2.4091811180114746\n",
      "489-th train cost:145.01962002840898 -2.3663582801818848\n",
      "490-th train cost:144.70943753329175 -2.5043909549713135\n",
      "491-th train cost:144.96953008478351 -2.337432622909546\n",
      "492-th train cost:144.58403891823505 -2.468348741531372\n",
      "493-th train cost:145.52907298694953 -2.232729911804199\n",
      "494-th train cost:145.02152870871808 -2.4140658378601074\n",
      "495-th train cost:145.1312370161577 -2.310673236846924\n",
      "496-th train cost:144.86936889648436 -2.5348732471466064\n",
      "497-th train cost:144.78533458362926 -2.511216163635254\n",
      "498-th train cost:145.26999395197095 -2.5359926223754883\n",
      "499-th train cost:145.10993075284085 -2.3712663650512695\n"
     ]
    }
   ],
   "source": [
    "training(num_epochs=500, batch_size = 200, log_every = 1, verbose=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Restoring parameters from iafmnist/iafmnist.ckpt\n"
     ]
    },
    {
     "ename": "NotFoundError",
     "evalue": "Key autoregressive_nn0/W1 not found in checkpoint\n\t [[Node: save/RestoreV2 = RestoreV2[dtypes=[DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, ..., DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT], _device=\"/job:localhost/replica:0/task:0/device:CPU:0\"](_arg_save/Const_0_0, save/RestoreV2/tensor_names, save/RestoreV2/shape_and_slices)]]\n\nCaused by op 'save/RestoreV2', defined at:\n  File \"/home/hang/anaconda3/lib/python3.6/runpy.py\", line 193, in _run_module_as_main\n    \"__main__\", mod_spec)\n  File \"/home/hang/anaconda3/lib/python3.6/runpy.py\", line 85, in _run_code\n    exec(code, run_globals)\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py\", line 16, in <module>\n    app.launch_new_instance()\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/traitlets/config/application.py\", line 658, in launch_instance\n    app.start()\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/ipykernel/kernelapp.py\", line 477, in start\n    ioloop.IOLoop.instance().start()\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/zmq/eventloop/ioloop.py\", line 177, in start\n    super(ZMQIOLoop, self).start()\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/tornado/ioloop.py\", line 888, in start\n    handler_func(fd_obj, events)\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/tornado/stack_context.py\", line 277, in null_wrapper\n    return fn(*args, **kwargs)\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/zmq/eventloop/zmqstream.py\", line 440, in _handle_events\n    self._handle_recv()\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/zmq/eventloop/zmqstream.py\", line 472, in _handle_recv\n    self._run_callback(callback, msg)\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/zmq/eventloop/zmqstream.py\", line 414, in _run_callback\n    callback(*args, **kwargs)\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/tornado/stack_context.py\", line 277, in null_wrapper\n    return fn(*args, **kwargs)\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/ipykernel/kernelbase.py\", line 283, in dispatcher\n    return self.dispatch_shell(stream, msg)\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/ipykernel/kernelbase.py\", line 235, in dispatch_shell\n    handler(stream, idents, msg)\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/ipykernel/kernelbase.py\", line 399, in execute_request\n    user_expressions, allow_stdin)\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/ipykernel/ipkernel.py\", line 196, in do_execute\n    res = shell.run_cell(code, store_history=store_history, silent=silent)\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/ipykernel/zmqshell.py\", line 533, in run_cell\n    return super(ZMQInteractiveShell, self).run_cell(*args, **kwargs)\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/IPython/core/interactiveshell.py\", line 2717, in run_cell\n    interactivity=interactivity, compiler=compiler, result=result)\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/IPython/core/interactiveshell.py\", line 2821, in run_ast_nodes\n    if self.run_code(code, result):\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/IPython/core/interactiveshell.py\", line 2881, in run_code\n    exec(code_obj, self.user_global_ns, self.user_ns)\n  File \"<ipython-input-9-2ae783dc9a70>\", line 3, in <module>\n    saver = tf.train.Saver()\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/tensorflow/python/training/saver.py\", line 1311, in __init__\n    self.build()\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/tensorflow/python/training/saver.py\", line 1320, in build\n    self._build(self._filename, build_save=True, build_restore=True)\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/tensorflow/python/training/saver.py\", line 1357, in _build\n    build_save=build_save, build_restore=build_restore)\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/tensorflow/python/training/saver.py\", line 809, in _build_internal\n    restore_sequentially, reshape)\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/tensorflow/python/training/saver.py\", line 448, in _AddRestoreOps\n    restore_sequentially)\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/tensorflow/python/training/saver.py\", line 860, in bulk_restore\n    return io_ops.restore_v2(filename_tensor, names, slices, dtypes)\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/tensorflow/python/ops/gen_io_ops.py\", line 1458, in restore_v2\n    shape_and_slices=shape_and_slices, dtypes=dtypes, name=name)\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/tensorflow/python/framework/op_def_library.py\", line 787, in _apply_op_helper\n    op_def=op_def)\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/tensorflow/python/framework/ops.py\", line 3290, in create_op\n    op_def=op_def)\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/tensorflow/python/framework/ops.py\", line 1654, in __init__\n    self._traceback = self._graph._extract_stack()  # pylint: disable=protected-access\n\nNotFoundError (see above for traceback): Key autoregressive_nn0/W1 not found in checkpoint\n\t [[Node: save/RestoreV2 = RestoreV2[dtypes=[DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, ..., DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT], _device=\"/job:localhost/replica:0/task:0/device:CPU:0\"](_arg_save/Const_0_0, save/RestoreV2/tensor_names, save/RestoreV2/shape_and_slices)]]\n",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNotFoundError\u001b[0m                             Traceback (most recent call last)",
      "\u001b[0;32m/home/hang/anaconda3/lib/python3.6/site-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36m_do_call\u001b[0;34m(self, fn, *args)\u001b[0m\n\u001b[1;32m   1326\u001b[0m     \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1327\u001b[0;31m       \u001b[0;32mreturn\u001b[0m \u001b[0mfn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1328\u001b[0m     \u001b[0;32mexcept\u001b[0m \u001b[0merrors\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mOpError\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/home/hang/anaconda3/lib/python3.6/site-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36m_run_fn\u001b[0;34m(feed_dict, fetch_list, target_list, options, run_metadata)\u001b[0m\n\u001b[1;32m   1311\u001b[0m       return self._call_tf_sessionrun(\n\u001b[0;32m-> 1312\u001b[0;31m           options, feed_dict, fetch_list, target_list, run_metadata)\n\u001b[0m\u001b[1;32m   1313\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/home/hang/anaconda3/lib/python3.6/site-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36m_call_tf_sessionrun\u001b[0;34m(self, options, feed_dict, fetch_list, target_list, run_metadata)\u001b[0m\n\u001b[1;32m   1419\u001b[0m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_session\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0moptions\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfeed_dict\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfetch_list\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtarget_list\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1420\u001b[0;31m             status, run_metadata)\n\u001b[0m\u001b[1;32m   1421\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/home/hang/anaconda3/lib/python3.6/site-packages/tensorflow/python/framework/errors_impl.py\u001b[0m in \u001b[0;36m__exit__\u001b[0;34m(self, type_arg, value_arg, traceback_arg)\u001b[0m\n\u001b[1;32m    515\u001b[0m             \u001b[0mcompat\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mas_text\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mc_api\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mTF_Message\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstatus\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstatus\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 516\u001b[0;31m             c_api.TF_GetCode(self.status.status))\n\u001b[0m\u001b[1;32m    517\u001b[0m     \u001b[0;31m# Delete the underlying status object from memory otherwise it stays alive\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mNotFoundError\u001b[0m: Key autoregressive_nn0/W1 not found in checkpoint\n\t [[Node: save/RestoreV2 = RestoreV2[dtypes=[DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, ..., DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT], _device=\"/job:localhost/replica:0/task:0/device:CPU:0\"](_arg_save/Const_0_0, save/RestoreV2/tensor_names, save/RestoreV2/shape_and_slices)]]",
      "\nDuring handling of the above exception, another exception occurred:\n",
      "\u001b[0;31mNotFoundError\u001b[0m                             Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-9-2ae783dc9a70>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      3\u001b[0m \u001b[0msaver\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mSaver\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[0;32mwith\u001b[0m \u001b[0mtf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mSession\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0msess\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 5\u001b[0;31m     \u001b[0msaver\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrestore\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msess\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m\"iafmnist/iafmnist.ckpt\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      6\u001b[0m     \u001b[0mx_sample\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmnist\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtest\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnext_batch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m64\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      7\u001b[0m     \u001b[0mx_reconstruct\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mz_vals\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mz_mean_val\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mz_log_sigma_0_val\u001b[0m  \u001b[0;34m=\u001b[0m \u001b[0msess\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrun\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx_mu_hat\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mZ\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mz_mu_0\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mz_log_sigma_0\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfeed_dict\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m{\u001b[0m\u001b[0mX\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mx_sample\u001b[0m\u001b[0;34m}\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/home/hang/anaconda3/lib/python3.6/site-packages/tensorflow/python/training/saver.py\u001b[0m in \u001b[0;36mrestore\u001b[0;34m(self, sess, save_path)\u001b[0m\n\u001b[1;32m   1773\u001b[0m     \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1774\u001b[0m       sess.run(self.saver_def.restore_op_name,\n\u001b[0;32m-> 1775\u001b[0;31m                {self.saver_def.filename_tensor_name: save_path})\n\u001b[0m\u001b[1;32m   1776\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1777\u001b[0m   \u001b[0;34m@\u001b[0m\u001b[0mstaticmethod\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/home/hang/anaconda3/lib/python3.6/site-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36mrun\u001b[0;34m(self, fetches, feed_dict, options, run_metadata)\u001b[0m\n\u001b[1;32m    903\u001b[0m     \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    904\u001b[0m       result = self._run(None, fetches, feed_dict, options_ptr,\n\u001b[0;32m--> 905\u001b[0;31m                          run_metadata_ptr)\n\u001b[0m\u001b[1;32m    906\u001b[0m       \u001b[0;32mif\u001b[0m \u001b[0mrun_metadata\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    907\u001b[0m         \u001b[0mproto_data\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtf_session\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mTF_GetBuffer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mrun_metadata_ptr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/home/hang/anaconda3/lib/python3.6/site-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36m_run\u001b[0;34m(self, handle, fetches, feed_dict, options, run_metadata)\u001b[0m\n\u001b[1;32m   1138\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mfinal_fetches\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0mfinal_targets\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mhandle\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0mfeed_dict_tensor\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1139\u001b[0m       results = self._do_run(handle, final_targets, final_fetches,\n\u001b[0;32m-> 1140\u001b[0;31m                              feed_dict_tensor, options, run_metadata)\n\u001b[0m\u001b[1;32m   1141\u001b[0m     \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1142\u001b[0m       \u001b[0mresults\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/home/hang/anaconda3/lib/python3.6/site-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36m_do_run\u001b[0;34m(self, handle, target_list, fetch_list, feed_dict, options, run_metadata)\u001b[0m\n\u001b[1;32m   1319\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mhandle\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1320\u001b[0m       return self._do_call(_run_fn, feeds, fetches, targets, options,\n\u001b[0;32m-> 1321\u001b[0;31m                            run_metadata)\n\u001b[0m\u001b[1;32m   1322\u001b[0m     \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1323\u001b[0m       \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_do_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0m_prun_fn\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mhandle\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfeeds\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfetches\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/home/hang/anaconda3/lib/python3.6/site-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36m_do_call\u001b[0;34m(self, fn, *args)\u001b[0m\n\u001b[1;32m   1338\u001b[0m         \u001b[0;32mexcept\u001b[0m \u001b[0mKeyError\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1339\u001b[0m           \u001b[0;32mpass\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1340\u001b[0;31m       \u001b[0;32mraise\u001b[0m \u001b[0mtype\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0me\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnode_def\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mop\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmessage\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1341\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1342\u001b[0m   \u001b[0;32mdef\u001b[0m \u001b[0m_extend_graph\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mNotFoundError\u001b[0m: Key autoregressive_nn0/W1 not found in checkpoint\n\t [[Node: save/RestoreV2 = RestoreV2[dtypes=[DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, ..., DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT], _device=\"/job:localhost/replica:0/task:0/device:CPU:0\"](_arg_save/Const_0_0, save/RestoreV2/tensor_names, save/RestoreV2/shape_and_slices)]]\n\nCaused by op 'save/RestoreV2', defined at:\n  File \"/home/hang/anaconda3/lib/python3.6/runpy.py\", line 193, in _run_module_as_main\n    \"__main__\", mod_spec)\n  File \"/home/hang/anaconda3/lib/python3.6/runpy.py\", line 85, in _run_code\n    exec(code, run_globals)\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py\", line 16, in <module>\n    app.launch_new_instance()\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/traitlets/config/application.py\", line 658, in launch_instance\n    app.start()\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/ipykernel/kernelapp.py\", line 477, in start\n    ioloop.IOLoop.instance().start()\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/zmq/eventloop/ioloop.py\", line 177, in start\n    super(ZMQIOLoop, self).start()\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/tornado/ioloop.py\", line 888, in start\n    handler_func(fd_obj, events)\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/tornado/stack_context.py\", line 277, in null_wrapper\n    return fn(*args, **kwargs)\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/zmq/eventloop/zmqstream.py\", line 440, in _handle_events\n    self._handle_recv()\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/zmq/eventloop/zmqstream.py\", line 472, in _handle_recv\n    self._run_callback(callback, msg)\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/zmq/eventloop/zmqstream.py\", line 414, in _run_callback\n    callback(*args, **kwargs)\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/tornado/stack_context.py\", line 277, in null_wrapper\n    return fn(*args, **kwargs)\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/ipykernel/kernelbase.py\", line 283, in dispatcher\n    return self.dispatch_shell(stream, msg)\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/ipykernel/kernelbase.py\", line 235, in dispatch_shell\n    handler(stream, idents, msg)\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/ipykernel/kernelbase.py\", line 399, in execute_request\n    user_expressions, allow_stdin)\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/ipykernel/ipkernel.py\", line 196, in do_execute\n    res = shell.run_cell(code, store_history=store_history, silent=silent)\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/ipykernel/zmqshell.py\", line 533, in run_cell\n    return super(ZMQInteractiveShell, self).run_cell(*args, **kwargs)\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/IPython/core/interactiveshell.py\", line 2717, in run_cell\n    interactivity=interactivity, compiler=compiler, result=result)\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/IPython/core/interactiveshell.py\", line 2821, in run_ast_nodes\n    if self.run_code(code, result):\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/IPython/core/interactiveshell.py\", line 2881, in run_code\n    exec(code_obj, self.user_global_ns, self.user_ns)\n  File \"<ipython-input-9-2ae783dc9a70>\", line 3, in <module>\n    saver = tf.train.Saver()\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/tensorflow/python/training/saver.py\", line 1311, in __init__\n    self.build()\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/tensorflow/python/training/saver.py\", line 1320, in build\n    self._build(self._filename, build_save=True, build_restore=True)\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/tensorflow/python/training/saver.py\", line 1357, in _build\n    build_save=build_save, build_restore=build_restore)\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/tensorflow/python/training/saver.py\", line 809, in _build_internal\n    restore_sequentially, reshape)\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/tensorflow/python/training/saver.py\", line 448, in _AddRestoreOps\n    restore_sequentially)\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/tensorflow/python/training/saver.py\", line 860, in bulk_restore\n    return io_ops.restore_v2(filename_tensor, names, slices, dtypes)\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/tensorflow/python/ops/gen_io_ops.py\", line 1458, in restore_v2\n    shape_and_slices=shape_and_slices, dtypes=dtypes, name=name)\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/tensorflow/python/framework/op_def_library.py\", line 787, in _apply_op_helper\n    op_def=op_def)\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/tensorflow/python/framework/ops.py\", line 3290, in create_op\n    op_def=op_def)\n  File \"/home/hang/anaconda3/lib/python3.6/site-packages/tensorflow/python/framework/ops.py\", line 1654, in __init__\n    self._traceback = self._graph._extract_stack()  # pylint: disable=protected-access\n\nNotFoundError (see above for traceback): Key autoregressive_nn0/W1 not found in checkpoint\n\t [[Node: save/RestoreV2 = RestoreV2[dtypes=[DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, ..., DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT], _device=\"/job:localhost/replica:0/task:0/device:CPU:0\"](_arg_save/Const_0_0, save/RestoreV2/tensor_names, save/RestoreV2/shape_and_slices)]]\n"
     ]
    }
   ],
   "source": [
    "#plot results\n",
    "#copy from https://github.com/oduerr/dl_tutorial/blob/master/tensorflow/vae/vae_demo.ipynb\n",
    "saver = tf.train.Saver()\n",
    "with tf.Session() as sess:\n",
    "    saver.restore(sess, \"iafmnist/iafmnist.ckpt\")\n",
    "    x_sample = mnist.test.next_batch(64)[0]\n",
    "    x_reconstruct,z_vals,z_mean_val,z_log_sigma_0_val  = sess.run((x_mu_hat,Z, z_mu_0, z_log_sigma_0), feed_dict={X: x_sample})\n",
    "\n",
    "    plt.figure(figsize=(8, 12))\n",
    "    for i in range(5):\n",
    "        plt.subplot(5, 3, 3*i + 1)\n",
    "        plt.imshow(x_sample[i].reshape(28, 28), vmin=0, vmax=1,  interpolation='none',cmap=plt.get_cmap('gray'))\n",
    "        plt.title(\"Test input\")\n",
    "        \n",
    "        #plt.colorbar()\n",
    "        plt.subplot(5, 3, 3*i + 2)\n",
    "        plt.scatter(z_vals[:,0],z_vals[:,1], c='gray', alpha=0.5)\n",
    "        plt.scatter(z_mean_val[i,0],z_mean_val[i,1], c='green', s=64, alpha=0.5)\n",
    "        plt.scatter(z_vals[i,0],z_vals[i,1], c='blue', s=16, alpha=0.5)\n",
    "               \n",
    "        plt.xlim((-3,3))\n",
    "        plt.ylim((-3,3))\n",
    "        plt.title(\"Latent Space\")\n",
    "        \n",
    "        plt.subplot(5, 3, 3*i + 3)\n",
    "        plt.imshow(x_reconstruct[i].reshape(28, 28), vmin=0, vmax=1, interpolation='none',cmap=plt.get_cmap('gray'))\n",
    "        plt.title(\"Reconstruction\")\n",
    "        #plt.colorbar()\n",
    "    plt.tight_layout()\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Restoring parameters from iafmnist/iafmnist.ckpt\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f64063152b0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "nx = ny = 20\n",
    "batch_size = 64\n",
    "x_values = np.linspace(-3, 3, nx)\n",
    "y_values = np.linspace(-3, 3, ny)\n",
    "canvas = np.empty((28*ny, 28*nx))\n",
    "saver = tf.train.Saver()\n",
    "with tf.Session() as sess:\n",
    "    saver.restore(sess, \"iafmnist/iafmnist.ckpt\")\n",
    "    d = np.zeros([batch_size,2],dtype='float32')\n",
    "    for i, yi in enumerate(x_values):\n",
    "        for j, xi in enumerate(y_values):\n",
    "            z_mu = np.array([[xi, yi]])\n",
    "            d[0] = z_mu\n",
    "            x_mean = sess.run(x_mu_hat, feed_dict={Z: d})\n",
    "            canvas[(nx-i-1)*28:(nx-i)*28, j*28:(j+1)*28] = x_mean[0].reshape(28, 28)\n",
    "\n",
    "plt.figure(figsize=(8, 10))        \n",
    "Xi, Yi = np.meshgrid(x_values, y_values)\n",
    "plt.imshow(canvas, origin=\"upper\", vmin=0, vmax=1,interpolation='none',cmap=plt.get_cmap('gray'))\n",
    "plt.tight_layout()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.6.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}