14_recurrent_neural_networks.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "**Chapter 14 – Recurrent Neural Networks**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "_This notebook contains all the sample code and solutions to the exercices in chapter 14._"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "# Setup"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "First, let's make sure this notebook works well in both python 2 and 3, import a few common modules, ensure MatplotLib plots figures inline and prepare a function to save the figures:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# To support both python 2 and python 3\n",
    "from __future__ import division, print_function, unicode_literals\n",
    "\n",
    "# Common imports\n",
    "import numpy as np\n",
    "import numpy.random as rnd\n",
    "import os\n",
    "\n",
    "# to make this notebook's output stable across runs\n",
    "rnd.seed(42)\n",
    "\n",
    "# To plot pretty figures\n",
    "%matplotlib inline\n",
    "import matplotlib\n",
    "import matplotlib.pyplot as plt\n",
    "plt.rcParams['axes.labelsize'] = 14\n",
    "plt.rcParams['xtick.labelsize'] = 12\n",
    "plt.rcParams['ytick.labelsize'] = 12\n",
    "\n",
    "# Where to save the figures\n",
    "PROJECT_ROOT_DIR = \".\"\n",
    "CHAPTER_ID = \"rnn\"\n",
    "\n",
    "def save_fig(fig_id, tight_layout=True):\n",
    "    path = os.path.join(PROJECT_ROOT_DIR, \"images\", CHAPTER_ID, fig_id + \".png\")\n",
    "    print(\"Saving figure\", fig_id)\n",
    "    if tight_layout:\n",
    "        plt.tight_layout()\n",
    "    plt.savefig(path, format='png', dpi=300)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "Then of course we will need TensorFlow:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "import tensorflow as tf"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "# Basic RNNs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Manual RNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "tf.reset_default_graph()\n",
    "\n",
    "n_inputs = 3\n",
    "n_neurons = 5\n",
    "\n",
    "X0 = tf.placeholder(tf.float32, [None, n_inputs])\n",
    "X1 = tf.placeholder(tf.float32, [None, n_inputs])\n",
    "\n",
    "Wx = tf.Variable(tf.random_normal(shape=[n_inputs, n_neurons], dtype=tf.float32))\n",
    "Wy = tf.Variable(tf.random_normal(shape=[n_neurons, n_neurons], dtype=tf.float32))\n",
    "b = tf.Variable(tf.zeros([1, n_neurons], dtype=tf.float32))\n",
    "\n",
    "Y0 = tf.tanh(tf.matmul(X0, Wx) + b)\n",
    "Y1 = tf.tanh(tf.matmul(Y0, Wy) + tf.matmul(X1, Wx) + b)\n",
    "\n",
    "init = tf.global_variables_initializer()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "X0_batch = np.array([[0, 1, 2], [3, 4, 5], [6, 7, 8], [9, 0, 1]]) # t = 0\n",
    "X1_batch = np.array([[9, 8, 7], [0, 0, 0], [6, 5, 4], [3, 2, 1]]) # t = 1\n",
    "\n",
    "with tf.Session() as sess:\n",
    "    init.run()\n",
    "    Y0_val, Y1_val = sess.run([Y0, Y1], feed_dict={X0: X0_batch, X1: X1_batch})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[-0.82053208  0.99431098 -0.99946314  0.9865402  -0.89121401]\n",
      " [ 0.10393736  1.         -1.          0.99997586 -0.99995434]\n",
      " [ 0.87802321  1.         -1.          0.99999994 -1.        ]\n",
      " [ 0.99999923 -0.07158625 -0.55363363 -0.99990493 -0.55132049]]\n"
     ]
    }
   ],
   "source": [
    "print(Y0_val)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[ 0.99977911  1.         -1.          0.99999851 -1.        ]\n",
      " [-0.49182287 -0.63029516  0.96565855 -0.02148083  0.21879137]\n",
      " [ 0.9952867   1.         -0.9999947   0.99873829 -0.99993926]\n",
      " [ 0.96052271  0.99886537 -0.7533018   0.06383554 -0.85634291]]\n"
     ]
    }
   ],
   "source": [
    "print(Y1_val)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Using `rnn()`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "tf.reset_default_graph()\n",
    "\n",
    "n_inputs = 3\n",
    "n_neurons = 5\n",
    "\n",
    "X0 = tf.placeholder(tf.float32, [None, n_inputs])\n",
    "X1 = tf.placeholder(tf.float32, [None, n_inputs])\n",
    "\n",
    "basic_cell = tf.contrib.rnn.BasicRNNCell(num_units=n_neurons)\n",
    "output_seqs, states = tf.contrib.rnn.static_rnn(basic_cell, [X0, X1], dtype=tf.float32)\n",
    "Y0, Y1 = output_seqs\n",
    "\n",
    "init = tf.global_variables_initializer()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "X0_batch = np.array([[0, 1, 2], [3, 4, 5], [6, 7, 8], [9, 0, 1]])\n",
    "X1_batch = np.array([[9, 8, 7], [0, 0, 0], [6, 5, 4], [3, 2, 1]])\n",
    "\n",
    "with tf.Session() as sess:\n",
    "    init.run()\n",
    "    Y0_val, Y1_val = sess.run([Y0, Y1], feed_dict={X0: X0_batch, X1: X1_batch})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 0.68331969, -0.78789645, -0.91668612, -0.94697189, -0.24038485],\n",
       "       [ 0.99775273, -0.9506833 , -0.99979204, -0.99975032, -0.53510815],\n",
       "       [ 0.99998641, -0.98928219, -0.99999952, -0.99999881, -0.73949718],\n",
       "       [ 0.99957275,  0.99909806,  0.32584012,  0.93206716,  0.84370828]], dtype=float32)"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Y0_val"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 0.99999291, -0.36171386, -0.99999785, -0.99996597, -0.00428927],\n",
       "       [-0.41839153,  0.75096804,  0.606565  ,  0.67534345,  0.59465897],\n",
       "       [ 0.99926353,  0.54629689, -0.99900544, -0.9924041 ,  0.1139629 ],\n",
       "       [ 0.89085555, -0.8939994 , -0.89453429, -0.93726897, -0.63662177]], dtype=float32)"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Y1_val"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "from IPython.display import clear_output, Image, display, HTML\n",
    "\n",
    "def strip_consts(graph_def, max_const_size=32):\n",
    "    \"\"\"Strip large constant values from graph_def.\"\"\"\n",
    "    strip_def = tf.GraphDef()\n",
    "    for n0 in graph_def.node:\n",
    "        n = strip_def.node.add() \n",
    "        n.MergeFrom(n0)\n",
    "        if n.op == 'Const':\n",
    "            tensor = n.attr['value'].tensor\n",
    "            size = len(tensor.tensor_content)\n",
    "            if size > max_const_size:\n",
    "                tensor.tensor_content = \"b<stripped %d bytes>\"%size\n",
    "    return strip_def\n",
    "\n",
    "def show_graph(graph_def, max_const_size=32):\n",
    "    \"\"\"Visualize TensorFlow graph.\"\"\"\n",
    "    if hasattr(graph_def, 'as_graph_def'):\n",
    "        graph_def = graph_def.as_graph_def()\n",
    "    strip_def = strip_consts(graph_def, max_const_size=max_const_size)\n",
    "    code = \"\"\"\n",
    "        <script>\n",
    "          function load() {{\n",
    "            document.getElementById(\"{id}\").pbtxt = {data};\n",
    "          }}\n",
    "        </script>\n",
    "        <link rel=\"import\" href=\"https://tensorboard.appspot.com/tf-graph-basic.build.html\" onload=load()>\n",
    "        <div style=\"height:600px\">\n",
    "          <tf-graph-basic id=\"{id}\"></tf-graph-basic>\n",
    "        </div>\n",
    "    \"\"\".format(data=repr(str(strip_def)), id='graph'+str(np.random.rand()))\n",
    "\n",
    "    iframe = \"\"\"\n",
    "        <iframe seamless style=\"width:1200px;height:620px;border:0\" srcdoc=\"{}\"></iframe>\n",
    "    \"\"\".format(code.replace('\"', '&quot;'))\n",
    "    display(HTML(iframe))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "        <iframe seamless style=\"width:1200px;height:620px;border:0\" srcdoc=\"\n",
       "        <script>\n",
       "          function load() {\n",
       "            document.getElementById(&quot;graph0.3745401188473625&quot;).pbtxt = 'node {\\n  name: &quot;Placeholder&quot;\\n  op: &quot;Placeholder&quot;\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;shape&quot;\\n    value {\\n      shape {\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;Placeholder_1&quot;\\n  op: &quot;Placeholder&quot;\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;shape&quot;\\n    value {\\n      shape {\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/Shape&quot;\\n  op: &quot;Shape&quot;\\n  input: &quot;Placeholder&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;out_type&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/strided_slice/stack&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_INT32\\n        tensor_shape {\\n          dim {\\n            size: 1\\n          }\\n        }\\n        int_val: 0\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/strided_slice/stack_1&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_INT32\\n        tensor_shape {\\n          dim {\\n            size: 1\\n          }\\n        }\\n        int_val: 1\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/strided_slice/stack_2&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_INT32\\n        tensor_shape {\\n          dim {\\n            size: 1\\n          }\\n        }\\n        int_val: 1\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/strided_slice&quot;\\n  op: &quot;StridedSlice&quot;\\n  input: &quot;rnn/Shape&quot;\\n  input: &quot;rnn/strided_slice/stack&quot;\\n  input: &quot;rnn/strided_slice/stack_1&quot;\\n  input: &quot;rnn/strided_slice/stack_2&quot;\\n  attr {\\n    key: &quot;Index&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;begin_mask&quot;\\n    value {\\n      i: 0\\n    }\\n  }\\n  attr {\\n    key: &quot;ellipsis_mask&quot;\\n    value {\\n      i: 0\\n    }\\n  }\\n  attr {\\n    key: &quot;end_mask&quot;\\n    value {\\n      i: 0\\n    }\\n  }\\n  attr {\\n    key: &quot;new_axis_mask&quot;\\n    value {\\n      i: 0\\n    }\\n  }\\n  attr {\\n    key: &quot;shrink_axis_mask&quot;\\n    value {\\n      i: 1\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/stack/1&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_INT32\\n        tensor_shape {\\n        }\\n        int_val: 5\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/stack&quot;\\n  op: &quot;Pack&quot;\\n  input: &quot;rnn/strided_slice&quot;\\n  input: &quot;rnn/stack/1&quot;\\n  attr {\\n    key: &quot;N&quot;\\n    value {\\n      i: 2\\n    }\\n  }\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;axis&quot;\\n    value {\\n      i: 0\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/zeros/Const&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_FLOAT\\n        tensor_shape {\\n        }\\n        float_val: 0.0\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/zeros&quot;\\n  op: &quot;Fill&quot;\\n  input: &quot;rnn/stack&quot;\\n  input: &quot;rnn/zeros/Const&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/basic_rnn_cell/weights/Initializer/random_uniform/shape&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;_class&quot;\\n    value {\\n      list {\\n        s: &quot;loc:@rnn/basic_rnn_cell/weights&quot;\\n      }\\n    }\\n  }\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_INT32\\n        tensor_shape {\\n          dim {\\n            size: 2\\n          }\\n        }\\n        tensor_content: &quot;\\\\010\\\\000\\\\000\\\\000\\\\005\\\\000\\\\000\\\\000&quot;\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/basic_rnn_cell/weights/Initializer/random_uniform/min&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;_class&quot;\\n    value {\\n      list {\\n        s: &quot;loc:@rnn/basic_rnn_cell/weights&quot;\\n      }\\n    }\\n  }\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_FLOAT\\n        tensor_shape {\\n        }\\n        float_val: -0.6793662309646606\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/basic_rnn_cell/weights/Initializer/random_uniform/max&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;_class&quot;\\n    value {\\n      list {\\n        s: &quot;loc:@rnn/basic_rnn_cell/weights&quot;\\n      }\\n    }\\n  }\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_FLOAT\\n        tensor_shape {\\n        }\\n        float_val: 0.6793662309646606\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/basic_rnn_cell/weights/Initializer/random_uniform/RandomUniform&quot;\\n  op: &quot;RandomUniform&quot;\\n  input: &quot;rnn/basic_rnn_cell/weights/Initializer/random_uniform/shape&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;_class&quot;\\n    value {\\n      list {\\n        s: &quot;loc:@rnn/basic_rnn_cell/weights&quot;\\n      }\\n    }\\n  }\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;seed&quot;\\n    value {\\n      i: 0\\n    }\\n  }\\n  attr {\\n    key: &quot;seed2&quot;\\n    value {\\n      i: 0\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/basic_rnn_cell/weights/Initializer/random_uniform/sub&quot;\\n  op: &quot;Sub&quot;\\n  input: &quot;rnn/basic_rnn_cell/weights/Initializer/random_uniform/max&quot;\\n  input: &quot;rnn/basic_rnn_cell/weights/Initializer/random_uniform/min&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;_class&quot;\\n    value {\\n      list {\\n        s: &quot;loc:@rnn/basic_rnn_cell/weights&quot;\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/basic_rnn_cell/weights/Initializer/random_uniform/mul&quot;\\n  op: &quot;Mul&quot;\\n  input: &quot;rnn/basic_rnn_cell/weights/Initializer/random_uniform/RandomUniform&quot;\\n  input: &quot;rnn/basic_rnn_cell/weights/Initializer/random_uniform/sub&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;_class&quot;\\n    value {\\n      list {\\n        s: &quot;loc:@rnn/basic_rnn_cell/weights&quot;\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/basic_rnn_cell/weights/Initializer/random_uniform&quot;\\n  op: &quot;Add&quot;\\n  input: &quot;rnn/basic_rnn_cell/weights/Initializer/random_uniform/mul&quot;\\n  input: &quot;rnn/basic_rnn_cell/weights/Initializer/random_uniform/min&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;_class&quot;\\n    value {\\n      list {\\n        s: &quot;loc:@rnn/basic_rnn_cell/weights&quot;\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/basic_rnn_cell/weights&quot;\\n  op: &quot;VariableV2&quot;\\n  attr {\\n    key: &quot;_class&quot;\\n    value {\\n      list {\\n        s: &quot;loc:@rnn/basic_rnn_cell/weights&quot;\\n      }\\n    }\\n  }\\n  attr {\\n    key: &quot;container&quot;\\n    value {\\n      s: &quot;&quot;\\n    }\\n  }\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;shape&quot;\\n    value {\\n      shape {\\n        dim {\\n          size: 8\\n        }\\n        dim {\\n          size: 5\\n        }\\n      }\\n    }\\n  }\\n  attr {\\n    key: &quot;shared_name&quot;\\n    value {\\n      s: &quot;&quot;\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/basic_rnn_cell/weights/Assign&quot;\\n  op: &quot;Assign&quot;\\n  input: &quot;rnn/basic_rnn_cell/weights&quot;\\n  input: &quot;rnn/basic_rnn_cell/weights/Initializer/random_uniform&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;_class&quot;\\n    value {\\n      list {\\n        s: &quot;loc:@rnn/basic_rnn_cell/weights&quot;\\n      }\\n    }\\n  }\\n  attr {\\n    key: &quot;use_locking&quot;\\n    value {\\n      b: true\\n    }\\n  }\\n  attr {\\n    key: &quot;validate_shape&quot;\\n    value {\\n      b: true\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/basic_rnn_cell/weights/read&quot;\\n  op: &quot;Identity&quot;\\n  input: &quot;rnn/basic_rnn_cell/weights&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/basic_rnn_cell/basic_rnn_cell/concat/axis&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_INT32\\n        tensor_shape {\\n        }\\n        int_val: 1\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/basic_rnn_cell/basic_rnn_cell/concat&quot;\\n  op: &quot;ConcatV2&quot;\\n  input: &quot;Placeholder&quot;\\n  input: &quot;rnn/zeros&quot;\\n  input: &quot;rnn/basic_rnn_cell/basic_rnn_cell/concat/axis&quot;\\n  attr {\\n    key: &quot;N&quot;\\n    value {\\n      i: 2\\n    }\\n  }\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;Tidx&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/basic_rnn_cell/basic_rnn_cell/MatMul&quot;\\n  op: &quot;MatMul&quot;\\n  input: &quot;rnn/basic_rnn_cell/basic_rnn_cell/concat&quot;\\n  input: &quot;rnn/basic_rnn_cell/weights/read&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;transpose_a&quot;\\n    value {\\n      b: false\\n    }\\n  }\\n  attr {\\n    key: &quot;transpose_b&quot;\\n    value {\\n      b: false\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/basic_rnn_cell/biases/Initializer/Const&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;_class&quot;\\n    value {\\n      list {\\n        s: &quot;loc:@rnn/basic_rnn_cell/biases&quot;\\n      }\\n    }\\n  }\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_FLOAT\\n        tensor_shape {\\n          dim {\\n            size: 5\\n          }\\n        }\\n        float_val: 0.0\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/basic_rnn_cell/biases&quot;\\n  op: &quot;VariableV2&quot;\\n  attr {\\n    key: &quot;_class&quot;\\n    value {\\n      list {\\n        s: &quot;loc:@rnn/basic_rnn_cell/biases&quot;\\n      }\\n    }\\n  }\\n  attr {\\n    key: &quot;container&quot;\\n    value {\\n      s: &quot;&quot;\\n    }\\n  }\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;shape&quot;\\n    value {\\n      shape {\\n        dim {\\n          size: 5\\n        }\\n      }\\n    }\\n  }\\n  attr {\\n    key: &quot;shared_name&quot;\\n    value {\\n      s: &quot;&quot;\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/basic_rnn_cell/biases/Assign&quot;\\n  op: &quot;Assign&quot;\\n  input: &quot;rnn/basic_rnn_cell/biases&quot;\\n  input: &quot;rnn/basic_rnn_cell/biases/Initializer/Const&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;_class&quot;\\n    value {\\n      list {\\n        s: &quot;loc:@rnn/basic_rnn_cell/biases&quot;\\n      }\\n    }\\n  }\\n  attr {\\n    key: &quot;use_locking&quot;\\n    value {\\n      b: true\\n    }\\n  }\\n  attr {\\n    key: &quot;validate_shape&quot;\\n    value {\\n      b: true\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/basic_rnn_cell/biases/read&quot;\\n  op: &quot;Identity&quot;\\n  input: &quot;rnn/basic_rnn_cell/biases&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/basic_rnn_cell/BiasAdd&quot;\\n  op: &quot;BiasAdd&quot;\\n  input: &quot;rnn/basic_rnn_cell/basic_rnn_cell/MatMul&quot;\\n  input: &quot;rnn/basic_rnn_cell/biases/read&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;data_format&quot;\\n    value {\\n      s: &quot;NHWC&quot;\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/basic_rnn_cell/Tanh&quot;\\n  op: &quot;Tanh&quot;\\n  input: &quot;rnn/basic_rnn_cell/BiasAdd&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/basic_rnn_cell_1/basic_rnn_cell/concat/axis&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_INT32\\n        tensor_shape {\\n        }\\n        int_val: 1\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/basic_rnn_cell_1/basic_rnn_cell/concat&quot;\\n  op: &quot;ConcatV2&quot;\\n  input: &quot;Placeholder_1&quot;\\n  input: &quot;rnn/basic_rnn_cell/Tanh&quot;\\n  input: &quot;rnn/basic_rnn_cell_1/basic_rnn_cell/concat/axis&quot;\\n  attr {\\n    key: &quot;N&quot;\\n    value {\\n      i: 2\\n    }\\n  }\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;Tidx&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/basic_rnn_cell_1/basic_rnn_cell/MatMul&quot;\\n  op: &quot;MatMul&quot;\\n  input: &quot;rnn/basic_rnn_cell_1/basic_rnn_cell/concat&quot;\\n  input: &quot;rnn/basic_rnn_cell/weights/read&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;transpose_a&quot;\\n    value {\\n      b: false\\n    }\\n  }\\n  attr {\\n    key: &quot;transpose_b&quot;\\n    value {\\n      b: false\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/basic_rnn_cell_1/BiasAdd&quot;\\n  op: &quot;BiasAdd&quot;\\n  input: &quot;rnn/basic_rnn_cell_1/basic_rnn_cell/MatMul&quot;\\n  input: &quot;rnn/basic_rnn_cell/biases/read&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;data_format&quot;\\n    value {\\n      s: &quot;NHWC&quot;\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/basic_rnn_cell_1/Tanh&quot;\\n  op: &quot;Tanh&quot;\\n  input: &quot;rnn/basic_rnn_cell_1/BiasAdd&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;init&quot;\\n  op: &quot;NoOp&quot;\\n  input: &quot;^rnn/basic_rnn_cell/weights/Assign&quot;\\n  input: &quot;^rnn/basic_rnn_cell/biases/Assign&quot;\\n}\\n';\n",
       "          }\n",
       "        </script>\n",
       "        <link rel=&quot;import&quot; href=&quot;https://tensorboard.appspot.com/tf-graph-basic.build.html&quot; onload=load()>\n",
       "        <div style=&quot;height:600px&quot;>\n",
       "          <tf-graph-basic id=&quot;graph0.3745401188473625&quot;></tf-graph-basic>\n",
       "        </div>\n",
       "    \"></iframe>\n",
       "    "
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "show_graph(tf.get_default_graph())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Packing sequences"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "tf.reset_default_graph()\n",
    "\n",
    "n_steps = 2\n",
    "n_inputs = 3\n",
    "n_neurons = 5\n",
    "\n",
    "X = tf.placeholder(tf.float32, [None, n_steps, n_inputs])\n",
    "X_seqs = tf.unstack(tf.transpose(X, perm=[1, 0, 2]))\n",
    "\n",
    "basic_cell = tf.contrib.rnn.BasicRNNCell(num_units=n_neurons)\n",
    "output_seqs, states = tf.contrib.rnn.static_rnn(basic_cell, X_seqs, dtype=tf.float32)\n",
    "outputs = tf.transpose(tf.stack(output_seqs), perm=[1, 0, 2])\n",
    "\n",
    "init = tf.global_variables_initializer()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "X_batch = np.array([\n",
    "        # t = 0      t = 1 \n",
    "        [[0, 1, 2], [9, 8, 7]], # instance 1\n",
    "        [[3, 4, 5], [0, 0, 0]], # instance 2\n",
    "        [[6, 7, 8], [6, 5, 4]], # instance 3\n",
    "        [[9, 0, 1], [3, 2, 1]], # instance 4\n",
    "    ])\n",
    "\n",
    "with tf.Session() as sess:\n",
    "    init.run()\n",
    "    outputs_val = outputs.eval(feed_dict={X: X_batch})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[ 0.99999988  1.         -0.92499006  0.99976337  0.99848008]\n",
      " [-0.35472792  0.71584183 -0.71373397  0.07423807  0.82909638]\n",
      " [ 0.9999674   0.99999803 -0.88509911  0.99682868  0.99839389]\n",
      " [ 0.99848074  0.99611902  0.36314142  0.99215239  0.95797563]]\n"
     ]
    }
   ],
   "source": [
    "print(np.transpose(outputs_val, axes=[1, 0, 2])[1])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Using `dynamic_rnn()`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "tf.reset_default_graph()\n",
    "\n",
    "n_steps = 2\n",
    "n_inputs = 3\n",
    "n_neurons = 5\n",
    "\n",
    "X = tf.placeholder(tf.float32, [None, n_steps, n_inputs])\n",
    "\n",
    "basic_cell = tf.contrib.rnn.BasicRNNCell(num_units=n_neurons)\n",
    "outputs, states = tf.nn.dynamic_rnn(basic_cell, X, dtype=tf.float32)\n",
    "\n",
    "init = tf.global_variables_initializer()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "outputs = [[[ 0.62797421  0.30141702 -0.44933137  0.26236424 -0.51391608]\n",
      "  [-0.9057309  -0.99141216  0.85657239 -0.99995953 -0.9993754 ]]\n",
      "\n",
      " [[ 0.29503232 -0.67023116 -0.30630386 -0.88455033 -0.97637779]\n",
      "  [-0.74294037  0.60441327 -0.08187051  0.76437837 -0.79939926]]\n",
      "\n",
      " [[-0.12919907 -0.95900756 -0.14796644 -0.99562371 -0.99911046]\n",
      "  [-0.9955582  -0.97180027  0.50988579 -0.9849022  -0.99946135]]\n",
      "\n",
      " [[-0.9996137  -0.90426421  0.99914223 -0.99596131 -0.99996263]\n",
      "  [-0.99554646 -0.97491413 -0.05848801 -0.36693698 -0.99495977]]]\n"
     ]
    }
   ],
   "source": [
    "X_batch = np.array([\n",
    "        [[0, 1, 2], [9, 8, 7]], # instance 1\n",
    "        [[3, 4, 5], [0, 0, 0]], # instance 2\n",
    "        [[6, 7, 8], [6, 5, 4]], # instance 3\n",
    "        [[9, 0, 1], [3, 2, 1]], # instance 4\n",
    "    ])\n",
    "\n",
    "with tf.Session() as sess:\n",
    "    init.run()\n",
    "    print(\"outputs =\", outputs.eval(feed_dict={X: X_batch}))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "        <iframe seamless style=\"width:1200px;height:620px;border:0\" srcdoc=\"\n",
       "        <script>\n",
       "          function load() {\n",
       "            document.getElementById(&quot;graph0.9507143064099162&quot;).pbtxt = 'node {\\n  name: &quot;Placeholder&quot;\\n  op: &quot;Placeholder&quot;\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;shape&quot;\\n    value {\\n      shape {\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;transpose/perm&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_INT32\\n        tensor_shape {\\n          dim {\\n            size: 3\\n          }\\n        }\\n        tensor_content: &quot;\\\\001\\\\000\\\\000\\\\000\\\\000\\\\000\\\\000\\\\000\\\\002\\\\000\\\\000\\\\000&quot;\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;transpose&quot;\\n  op: &quot;Transpose&quot;\\n  input: &quot;Placeholder&quot;\\n  input: &quot;transpose/perm&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;Tperm&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/Shape&quot;\\n  op: &quot;Shape&quot;\\n  input: &quot;transpose&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;out_type&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/strided_slice/stack&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_INT32\\n        tensor_shape {\\n          dim {\\n            size: 1\\n          }\\n        }\\n        int_val: 1\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/strided_slice/stack_1&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_INT32\\n        tensor_shape {\\n          dim {\\n            size: 1\\n          }\\n        }\\n        int_val: 2\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/strided_slice/stack_2&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_INT32\\n        tensor_shape {\\n          dim {\\n            size: 1\\n          }\\n        }\\n        int_val: 1\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/strided_slice&quot;\\n  op: &quot;StridedSlice&quot;\\n  input: &quot;rnn/Shape&quot;\\n  input: &quot;rnn/strided_slice/stack&quot;\\n  input: &quot;rnn/strided_slice/stack_1&quot;\\n  input: &quot;rnn/strided_slice/stack_2&quot;\\n  attr {\\n    key: &quot;Index&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;begin_mask&quot;\\n    value {\\n      i: 0\\n    }\\n  }\\n  attr {\\n    key: &quot;ellipsis_mask&quot;\\n    value {\\n      i: 0\\n    }\\n  }\\n  attr {\\n    key: &quot;end_mask&quot;\\n    value {\\n      i: 0\\n    }\\n  }\\n  attr {\\n    key: &quot;new_axis_mask&quot;\\n    value {\\n      i: 0\\n    }\\n  }\\n  attr {\\n    key: &quot;shrink_axis_mask&quot;\\n    value {\\n      i: 1\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/strided_slice_1/stack&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_INT32\\n        tensor_shape {\\n          dim {\\n            size: 1\\n          }\\n        }\\n        int_val: 1\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/strided_slice_1/stack_1&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_INT32\\n        tensor_shape {\\n          dim {\\n            size: 1\\n          }\\n        }\\n        int_val: 2\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/strided_slice_1/stack_2&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_INT32\\n        tensor_shape {\\n          dim {\\n            size: 1\\n          }\\n        }\\n        int_val: 1\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/strided_slice_1&quot;\\n  op: &quot;StridedSlice&quot;\\n  input: &quot;rnn/Shape&quot;\\n  input: &quot;rnn/strided_slice_1/stack&quot;\\n  input: &quot;rnn/strided_slice_1/stack_1&quot;\\n  input: &quot;rnn/strided_slice_1/stack_2&quot;\\n  attr {\\n    key: &quot;Index&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;begin_mask&quot;\\n    value {\\n      i: 0\\n    }\\n  }\\n  attr {\\n    key: &quot;ellipsis_mask&quot;\\n    value {\\n      i: 0\\n    }\\n  }\\n  attr {\\n    key: &quot;end_mask&quot;\\n    value {\\n      i: 0\\n    }\\n  }\\n  attr {\\n    key: &quot;new_axis_mask&quot;\\n    value {\\n      i: 0\\n    }\\n  }\\n  attr {\\n    key: &quot;shrink_axis_mask&quot;\\n    value {\\n      i: 1\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/stack/1&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_INT32\\n        tensor_shape {\\n        }\\n        int_val: 5\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/stack&quot;\\n  op: &quot;Pack&quot;\\n  input: &quot;rnn/strided_slice&quot;\\n  input: &quot;rnn/stack/1&quot;\\n  attr {\\n    key: &quot;N&quot;\\n    value {\\n      i: 2\\n    }\\n  }\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;axis&quot;\\n    value {\\n      i: 0\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/zeros/Const&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_FLOAT\\n        tensor_shape {\\n        }\\n        float_val: 0.0\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/zeros&quot;\\n  op: &quot;Fill&quot;\\n  input: &quot;rnn/stack&quot;\\n  input: &quot;rnn/zeros/Const&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/Shape_1&quot;\\n  op: &quot;Shape&quot;\\n  input: &quot;transpose&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;out_type&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/strided_slice_2/stack&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_INT32\\n        tensor_shape {\\n          dim {\\n            size: 1\\n          }\\n        }\\n        int_val: 0\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/strided_slice_2/stack_1&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_INT32\\n        tensor_shape {\\n          dim {\\n            size: 1\\n          }\\n        }\\n        int_val: 1\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/strided_slice_2/stack_2&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_INT32\\n        tensor_shape {\\n          dim {\\n            size: 1\\n          }\\n        }\\n        int_val: 1\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/strided_slice_2&quot;\\n  op: &quot;StridedSlice&quot;\\n  input: &quot;rnn/Shape_1&quot;\\n  input: &quot;rnn/strided_slice_2/stack&quot;\\n  input: &quot;rnn/strided_slice_2/stack_1&quot;\\n  input: &quot;rnn/strided_slice_2/stack_2&quot;\\n  attr {\\n    key: &quot;Index&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;begin_mask&quot;\\n    value {\\n      i: 0\\n    }\\n  }\\n  attr {\\n    key: &quot;ellipsis_mask&quot;\\n    value {\\n      i: 0\\n    }\\n  }\\n  attr {\\n    key: &quot;end_mask&quot;\\n    value {\\n      i: 0\\n    }\\n  }\\n  attr {\\n    key: &quot;new_axis_mask&quot;\\n    value {\\n      i: 0\\n    }\\n  }\\n  attr {\\n    key: &quot;shrink_axis_mask&quot;\\n    value {\\n      i: 1\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/strided_slice_3/stack&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_INT32\\n        tensor_shape {\\n          dim {\\n            size: 1\\n          }\\n        }\\n        int_val: 1\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/strided_slice_3/stack_1&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_INT32\\n        tensor_shape {\\n          dim {\\n            size: 1\\n          }\\n        }\\n        int_val: 2\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/strided_slice_3/stack_2&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_INT32\\n        tensor_shape {\\n          dim {\\n            size: 1\\n          }\\n        }\\n        int_val: 1\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/strided_slice_3&quot;\\n  op: &quot;StridedSlice&quot;\\n  input: &quot;rnn/Shape_1&quot;\\n  input: &quot;rnn/strided_slice_3/stack&quot;\\n  input: &quot;rnn/strided_slice_3/stack_1&quot;\\n  input: &quot;rnn/strided_slice_3/stack_2&quot;\\n  attr {\\n    key: &quot;Index&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;begin_mask&quot;\\n    value {\\n      i: 0\\n    }\\n  }\\n  attr {\\n    key: &quot;ellipsis_mask&quot;\\n    value {\\n      i: 0\\n    }\\n  }\\n  attr {\\n    key: &quot;end_mask&quot;\\n    value {\\n      i: 0\\n    }\\n  }\\n  attr {\\n    key: &quot;new_axis_mask&quot;\\n    value {\\n      i: 0\\n    }\\n  }\\n  attr {\\n    key: &quot;shrink_axis_mask&quot;\\n    value {\\n      i: 1\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/stack_1/1&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_INT32\\n        tensor_shape {\\n        }\\n        int_val: 5\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/stack_1&quot;\\n  op: &quot;Pack&quot;\\n  input: &quot;rnn/strided_slice_3&quot;\\n  input: &quot;rnn/stack_1/1&quot;\\n  attr {\\n    key: &quot;N&quot;\\n    value {\\n      i: 2\\n    }\\n  }\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;axis&quot;\\n    value {\\n      i: 0\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/zeros_1/Const&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_FLOAT\\n        tensor_shape {\\n        }\\n        float_val: 0.0\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/zeros_1&quot;\\n  op: &quot;Fill&quot;\\n  input: &quot;rnn/stack_1&quot;\\n  input: &quot;rnn/zeros_1/Const&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/time&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_INT32\\n        tensor_shape {\\n        }\\n        int_val: 0\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/TensorArray&quot;\\n  op: &quot;TensorArrayV3&quot;\\n  input: &quot;rnn/strided_slice_2&quot;\\n  attr {\\n    key: &quot;clear_after_read&quot;\\n    value {\\n      b: true\\n    }\\n  }\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;dynamic_size&quot;\\n    value {\\n      b: false\\n    }\\n  }\\n  attr {\\n    key: &quot;element_shape&quot;\\n    value {\\n      shape {\\n        unknown_rank: true\\n      }\\n    }\\n  }\\n  attr {\\n    key: &quot;tensor_array_name&quot;\\n    value {\\n      s: &quot;rnn/dynamic_rnn/output_0&quot;\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/TensorArray_1&quot;\\n  op: &quot;TensorArrayV3&quot;\\n  input: &quot;rnn/strided_slice_2&quot;\\n  attr {\\n    key: &quot;clear_after_read&quot;\\n    value {\\n      b: true\\n    }\\n  }\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;dynamic_size&quot;\\n    value {\\n      b: false\\n    }\\n  }\\n  attr {\\n    key: &quot;element_shape&quot;\\n    value {\\n      shape {\\n        unknown_rank: true\\n      }\\n    }\\n  }\\n  attr {\\n    key: &quot;tensor_array_name&quot;\\n    value {\\n      s: &quot;rnn/dynamic_rnn/input_0&quot;\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/TensorArrayUnstack/Shape&quot;\\n  op: &quot;Shape&quot;\\n  input: &quot;transpose&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;out_type&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/TensorArrayUnstack/strided_slice/stack&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_INT32\\n        tensor_shape {\\n          dim {\\n            size: 1\\n          }\\n        }\\n        int_val: 0\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/TensorArrayUnstack/strided_slice/stack_1&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_INT32\\n        tensor_shape {\\n          dim {\\n            size: 1\\n          }\\n        }\\n        int_val: 1\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/TensorArrayUnstack/strided_slice/stack_2&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_INT32\\n        tensor_shape {\\n          dim {\\n            size: 1\\n          }\\n        }\\n        int_val: 1\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/TensorArrayUnstack/strided_slice&quot;\\n  op: &quot;StridedSlice&quot;\\n  input: &quot;rnn/TensorArrayUnstack/Shape&quot;\\n  input: &quot;rnn/TensorArrayUnstack/strided_slice/stack&quot;\\n  input: &quot;rnn/TensorArrayUnstack/strided_slice/stack_1&quot;\\n  input: &quot;rnn/TensorArrayUnstack/strided_slice/stack_2&quot;\\n  attr {\\n    key: &quot;Index&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;begin_mask&quot;\\n    value {\\n      i: 0\\n    }\\n  }\\n  attr {\\n    key: &quot;ellipsis_mask&quot;\\n    value {\\n      i: 0\\n    }\\n  }\\n  attr {\\n    key: &quot;end_mask&quot;\\n    value {\\n      i: 0\\n    }\\n  }\\n  attr {\\n    key: &quot;new_axis_mask&quot;\\n    value {\\n      i: 0\\n    }\\n  }\\n  attr {\\n    key: &quot;shrink_axis_mask&quot;\\n    value {\\n      i: 1\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/TensorArrayUnstack/range/start&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_INT32\\n        tensor_shape {\\n        }\\n        int_val: 0\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/TensorArrayUnstack/range/delta&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_INT32\\n        tensor_shape {\\n        }\\n        int_val: 1\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/TensorArrayUnstack/range&quot;\\n  op: &quot;Range&quot;\\n  input: &quot;rnn/TensorArrayUnstack/range/start&quot;\\n  input: &quot;rnn/TensorArrayUnstack/strided_slice&quot;\\n  input: &quot;rnn/TensorArrayUnstack/range/delta&quot;\\n  attr {\\n    key: &quot;Tidx&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/TensorArrayUnstack/TensorArrayScatter/TensorArrayScatterV3&quot;\\n  op: &quot;TensorArrayScatterV3&quot;\\n  input: &quot;rnn/TensorArray_1&quot;\\n  input: &quot;rnn/TensorArrayUnstack/range&quot;\\n  input: &quot;transpose&quot;\\n  input: &quot;rnn/TensorArray_1:1&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;_class&quot;\\n    value {\\n      list {\\n        s: &quot;loc:@rnn/TensorArray_1&quot;\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/while/Enter&quot;\\n  op: &quot;Enter&quot;\\n  input: &quot;rnn/time&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;frame_name&quot;\\n    value {\\n      s: &quot;rnn/while/rnn/while/&quot;\\n    }\\n  }\\n  attr {\\n    key: &quot;is_constant&quot;\\n    value {\\n      b: false\\n    }\\n  }\\n  attr {\\n    key: &quot;parallel_iterations&quot;\\n    value {\\n      i: 32\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/while/Enter_1&quot;\\n  op: &quot;Enter&quot;\\n  input: &quot;rnn/TensorArray:1&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;frame_name&quot;\\n    value {\\n      s: &quot;rnn/while/rnn/while/&quot;\\n    }\\n  }\\n  attr {\\n    key: &quot;is_constant&quot;\\n    value {\\n      b: false\\n    }\\n  }\\n  attr {\\n    key: &quot;parallel_iterations&quot;\\n    value {\\n      i: 32\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/while/Enter_2&quot;\\n  op: &quot;Enter&quot;\\n  input: &quot;rnn/zeros&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;frame_name&quot;\\n    value {\\n      s: &quot;rnn/while/rnn/while/&quot;\\n    }\\n  }\\n  attr {\\n    key: &quot;is_constant&quot;\\n    value {\\n      b: false\\n    }\\n  }\\n  attr {\\n    key: &quot;parallel_iterations&quot;\\n    value {\\n      i: 32\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/while/Merge&quot;\\n  op: &quot;Merge&quot;\\n  input: &quot;rnn/while/Enter&quot;\\n  input: &quot;rnn/while/NextIteration&quot;\\n  attr {\\n    key: &quot;N&quot;\\n    value {\\n      i: 2\\n    }\\n  }\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/while/Merge_1&quot;\\n  op: &quot;Merge&quot;\\n  input: &quot;rnn/while/Enter_1&quot;\\n  input: &quot;rnn/while/NextIteration_1&quot;\\n  attr {\\n    key: &quot;N&quot;\\n    value {\\n      i: 2\\n    }\\n  }\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/while/Merge_2&quot;\\n  op: &quot;Merge&quot;\\n  input: &quot;rnn/while/Enter_2&quot;\\n  input: &quot;rnn/while/NextIteration_2&quot;\\n  attr {\\n    key: &quot;N&quot;\\n    value {\\n      i: 2\\n    }\\n  }\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/while/Less/Enter&quot;\\n  op: &quot;Enter&quot;\\n  input: &quot;rnn/strided_slice_2&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;frame_name&quot;\\n    value {\\n      s: &quot;rnn/while/rnn/while/&quot;\\n    }\\n  }\\n  attr {\\n    key: &quot;is_constant&quot;\\n    value {\\n      b: true\\n    }\\n  }\\n  attr {\\n    key: &quot;parallel_iterations&quot;\\n    value {\\n      i: 32\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/while/Less&quot;\\n  op: &quot;Less&quot;\\n  input: &quot;rnn/while/Merge&quot;\\n  input: &quot;rnn/while/Less/Enter&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/while/LoopCond&quot;\\n  op: &quot;LoopCond&quot;\\n  input: &quot;rnn/while/Less&quot;\\n}\\nnode {\\n  name: &quot;rnn/while/Switch&quot;\\n  op: &quot;Switch&quot;\\n  input: &quot;rnn/while/Merge&quot;\\n  input: &quot;rnn/while/LoopCond&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;_class&quot;\\n    value {\\n      list {\\n        s: &quot;loc:@rnn/while/Merge&quot;\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/while/Switch_1&quot;\\n  op: &quot;Switch&quot;\\n  input: &quot;rnn/while/Merge_1&quot;\\n  input: &quot;rnn/while/LoopCond&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;_class&quot;\\n    value {\\n      list {\\n        s: &quot;loc:@rnn/while/Merge_1&quot;\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/while/Switch_2&quot;\\n  op: &quot;Switch&quot;\\n  input: &quot;rnn/while/Merge_2&quot;\\n  input: &quot;rnn/while/LoopCond&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;_class&quot;\\n    value {\\n      list {\\n        s: &quot;loc:@rnn/while/Merge_2&quot;\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/while/Identity&quot;\\n  op: &quot;Identity&quot;\\n  input: &quot;rnn/while/Switch:1&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/while/Identity_1&quot;\\n  op: &quot;Identity&quot;\\n  input: &quot;rnn/while/Switch_1:1&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/while/Identity_2&quot;\\n  op: &quot;Identity&quot;\\n  input: &quot;rnn/while/Switch_2:1&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/while/TensorArrayReadV3/Enter&quot;\\n  op: &quot;Enter&quot;\\n  input: &quot;rnn/TensorArray_1&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_RESOURCE\\n    }\\n  }\\n  attr {\\n    key: &quot;_class&quot;\\n    value {\\n      list {\\n        s: &quot;loc:@rnn/TensorArray_1&quot;\\n      }\\n    }\\n  }\\n  attr {\\n    key: &quot;frame_name&quot;\\n    value {\\n      s: &quot;rnn/while/rnn/while/&quot;\\n    }\\n  }\\n  attr {\\n    key: &quot;is_constant&quot;\\n    value {\\n      b: true\\n    }\\n  }\\n  attr {\\n    key: &quot;parallel_iterations&quot;\\n    value {\\n      i: 32\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/while/TensorArrayReadV3/Enter_1&quot;\\n  op: &quot;Enter&quot;\\n  input: &quot;rnn/TensorArrayUnstack/TensorArrayScatter/TensorArrayScatterV3&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;_class&quot;\\n    value {\\n      list {\\n        s: &quot;loc:@rnn/TensorArray_1&quot;\\n      }\\n    }\\n  }\\n  attr {\\n    key: &quot;frame_name&quot;\\n    value {\\n      s: &quot;rnn/while/rnn/while/&quot;\\n    }\\n  }\\n  attr {\\n    key: &quot;is_constant&quot;\\n    value {\\n      b: true\\n    }\\n  }\\n  attr {\\n    key: &quot;parallel_iterations&quot;\\n    value {\\n      i: 32\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/while/TensorArrayReadV3&quot;\\n  op: &quot;TensorArrayReadV3&quot;\\n  input: &quot;rnn/while/TensorArrayReadV3/Enter&quot;\\n  input: &quot;rnn/while/Identity&quot;\\n  input: &quot;rnn/while/TensorArrayReadV3/Enter_1&quot;\\n  attr {\\n    key: &quot;_class&quot;\\n    value {\\n      list {\\n        s: &quot;loc:@rnn/TensorArray_1&quot;\\n      }\\n    }\\n  }\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/basic_rnn_cell/weights/Initializer/random_uniform/shape&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;_class&quot;\\n    value {\\n      list {\\n        s: &quot;loc:@rnn/basic_rnn_cell/weights&quot;\\n      }\\n    }\\n  }\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_INT32\\n        tensor_shape {\\n          dim {\\n            size: 2\\n          }\\n        }\\n        tensor_content: &quot;\\\\010\\\\000\\\\000\\\\000\\\\005\\\\000\\\\000\\\\000&quot;\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/basic_rnn_cell/weights/Initializer/random_uniform/min&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;_class&quot;\\n    value {\\n      list {\\n        s: &quot;loc:@rnn/basic_rnn_cell/weights&quot;\\n      }\\n    }\\n  }\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_FLOAT\\n        tensor_shape {\\n        }\\n        float_val: -0.6793662309646606\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/basic_rnn_cell/weights/Initializer/random_uniform/max&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;_class&quot;\\n    value {\\n      list {\\n        s: &quot;loc:@rnn/basic_rnn_cell/weights&quot;\\n      }\\n    }\\n  }\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_FLOAT\\n        tensor_shape {\\n        }\\n        float_val: 0.6793662309646606\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/basic_rnn_cell/weights/Initializer/random_uniform/RandomUniform&quot;\\n  op: &quot;RandomUniform&quot;\\n  input: &quot;rnn/basic_rnn_cell/weights/Initializer/random_uniform/shape&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;_class&quot;\\n    value {\\n      list {\\n        s: &quot;loc:@rnn/basic_rnn_cell/weights&quot;\\n      }\\n    }\\n  }\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;seed&quot;\\n    value {\\n      i: 0\\n    }\\n  }\\n  attr {\\n    key: &quot;seed2&quot;\\n    value {\\n      i: 0\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/basic_rnn_cell/weights/Initializer/random_uniform/sub&quot;\\n  op: &quot;Sub&quot;\\n  input: &quot;rnn/basic_rnn_cell/weights/Initializer/random_uniform/max&quot;\\n  input: &quot;rnn/basic_rnn_cell/weights/Initializer/random_uniform/min&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;_class&quot;\\n    value {\\n      list {\\n        s: &quot;loc:@rnn/basic_rnn_cell/weights&quot;\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/basic_rnn_cell/weights/Initializer/random_uniform/mul&quot;\\n  op: &quot;Mul&quot;\\n  input: &quot;rnn/basic_rnn_cell/weights/Initializer/random_uniform/RandomUniform&quot;\\n  input: &quot;rnn/basic_rnn_cell/weights/Initializer/random_uniform/sub&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;_class&quot;\\n    value {\\n      list {\\n        s: &quot;loc:@rnn/basic_rnn_cell/weights&quot;\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/basic_rnn_cell/weights/Initializer/random_uniform&quot;\\n  op: &quot;Add&quot;\\n  input: &quot;rnn/basic_rnn_cell/weights/Initializer/random_uniform/mul&quot;\\n  input: &quot;rnn/basic_rnn_cell/weights/Initializer/random_uniform/min&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;_class&quot;\\n    value {\\n      list {\\n        s: &quot;loc:@rnn/basic_rnn_cell/weights&quot;\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/basic_rnn_cell/weights&quot;\\n  op: &quot;VariableV2&quot;\\n  attr {\\n    key: &quot;_class&quot;\\n    value {\\n      list {\\n        s: &quot;loc:@rnn/basic_rnn_cell/weights&quot;\\n      }\\n    }\\n  }\\n  attr {\\n    key: &quot;container&quot;\\n    value {\\n      s: &quot;&quot;\\n    }\\n  }\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;shape&quot;\\n    value {\\n      shape {\\n        dim {\\n          size: 8\\n        }\\n        dim {\\n          size: 5\\n        }\\n      }\\n    }\\n  }\\n  attr {\\n    key: &quot;shared_name&quot;\\n    value {\\n      s: &quot;&quot;\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/basic_rnn_cell/weights/Assign&quot;\\n  op: &quot;Assign&quot;\\n  input: &quot;rnn/basic_rnn_cell/weights&quot;\\n  input: &quot;rnn/basic_rnn_cell/weights/Initializer/random_uniform&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;_class&quot;\\n    value {\\n      list {\\n        s: &quot;loc:@rnn/basic_rnn_cell/weights&quot;\\n      }\\n    }\\n  }\\n  attr {\\n    key: &quot;use_locking&quot;\\n    value {\\n      b: true\\n    }\\n  }\\n  attr {\\n    key: &quot;validate_shape&quot;\\n    value {\\n      b: true\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/basic_rnn_cell/weights/read&quot;\\n  op: &quot;Identity&quot;\\n  input: &quot;rnn/basic_rnn_cell/weights&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/while/basic_rnn_cell/basic_rnn_cell/concat/axis&quot;\\n  op: &quot;Const&quot;\\n  input: &quot;^rnn/while/Identity&quot;\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_INT32\\n        tensor_shape {\\n        }\\n        int_val: 1\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/while/basic_rnn_cell/basic_rnn_cell/concat&quot;\\n  op: &quot;ConcatV2&quot;\\n  input: &quot;rnn/while/TensorArrayReadV3&quot;\\n  input: &quot;rnn/while/Identity_2&quot;\\n  input: &quot;rnn/while/basic_rnn_cell/basic_rnn_cell/concat/axis&quot;\\n  attr {\\n    key: &quot;N&quot;\\n    value {\\n      i: 2\\n    }\\n  }\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;Tidx&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/while/basic_rnn_cell/basic_rnn_cell/MatMul/Enter&quot;\\n  op: &quot;Enter&quot;\\n  input: &quot;rnn/basic_rnn_cell/weights/read&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;frame_name&quot;\\n    value {\\n      s: &quot;rnn/while/rnn/while/&quot;\\n    }\\n  }\\n  attr {\\n    key: &quot;is_constant&quot;\\n    value {\\n      b: true\\n    }\\n  }\\n  attr {\\n    key: &quot;parallel_iterations&quot;\\n    value {\\n      i: 32\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/while/basic_rnn_cell/basic_rnn_cell/MatMul&quot;\\n  op: &quot;MatMul&quot;\\n  input: &quot;rnn/while/basic_rnn_cell/basic_rnn_cell/concat&quot;\\n  input: &quot;rnn/while/basic_rnn_cell/basic_rnn_cell/MatMul/Enter&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;transpose_a&quot;\\n    value {\\n      b: false\\n    }\\n  }\\n  attr {\\n    key: &quot;transpose_b&quot;\\n    value {\\n      b: false\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/basic_rnn_cell/biases/Initializer/Const&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;_class&quot;\\n    value {\\n      list {\\n        s: &quot;loc:@rnn/basic_rnn_cell/biases&quot;\\n      }\\n    }\\n  }\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_FLOAT\\n        tensor_shape {\\n          dim {\\n            size: 5\\n          }\\n        }\\n        float_val: 0.0\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/basic_rnn_cell/biases&quot;\\n  op: &quot;VariableV2&quot;\\n  attr {\\n    key: &quot;_class&quot;\\n    value {\\n      list {\\n        s: &quot;loc:@rnn/basic_rnn_cell/biases&quot;\\n      }\\n    }\\n  }\\n  attr {\\n    key: &quot;container&quot;\\n    value {\\n      s: &quot;&quot;\\n    }\\n  }\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;shape&quot;\\n    value {\\n      shape {\\n        dim {\\n          size: 5\\n        }\\n      }\\n    }\\n  }\\n  attr {\\n    key: &quot;shared_name&quot;\\n    value {\\n      s: &quot;&quot;\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/basic_rnn_cell/biases/Assign&quot;\\n  op: &quot;Assign&quot;\\n  input: &quot;rnn/basic_rnn_cell/biases&quot;\\n  input: &quot;rnn/basic_rnn_cell/biases/Initializer/Const&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;_class&quot;\\n    value {\\n      list {\\n        s: &quot;loc:@rnn/basic_rnn_cell/biases&quot;\\n      }\\n    }\\n  }\\n  attr {\\n    key: &quot;use_locking&quot;\\n    value {\\n      b: true\\n    }\\n  }\\n  attr {\\n    key: &quot;validate_shape&quot;\\n    value {\\n      b: true\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/basic_rnn_cell/biases/read&quot;\\n  op: &quot;Identity&quot;\\n  input: &quot;rnn/basic_rnn_cell/biases&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/while/basic_rnn_cell/BiasAdd/Enter&quot;\\n  op: &quot;Enter&quot;\\n  input: &quot;rnn/basic_rnn_cell/biases/read&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;frame_name&quot;\\n    value {\\n      s: &quot;rnn/while/rnn/while/&quot;\\n    }\\n  }\\n  attr {\\n    key: &quot;is_constant&quot;\\n    value {\\n      b: true\\n    }\\n  }\\n  attr {\\n    key: &quot;parallel_iterations&quot;\\n    value {\\n      i: 32\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/while/basic_rnn_cell/BiasAdd&quot;\\n  op: &quot;BiasAdd&quot;\\n  input: &quot;rnn/while/basic_rnn_cell/basic_rnn_cell/MatMul&quot;\\n  input: &quot;rnn/while/basic_rnn_cell/BiasAdd/Enter&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;data_format&quot;\\n    value {\\n      s: &quot;NHWC&quot;\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/while/basic_rnn_cell/Tanh&quot;\\n  op: &quot;Tanh&quot;\\n  input: &quot;rnn/while/basic_rnn_cell/BiasAdd&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/while/TensorArrayWrite/TensorArrayWriteV3/Enter&quot;\\n  op: &quot;Enter&quot;\\n  input: &quot;rnn/TensorArray&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_RESOURCE\\n    }\\n  }\\n  attr {\\n    key: &quot;_class&quot;\\n    value {\\n      list {\\n        s: &quot;loc:@rnn/TensorArray&quot;\\n      }\\n    }\\n  }\\n  attr {\\n    key: &quot;frame_name&quot;\\n    value {\\n      s: &quot;rnn/while/rnn/while/&quot;\\n    }\\n  }\\n  attr {\\n    key: &quot;is_constant&quot;\\n    value {\\n      b: true\\n    }\\n  }\\n  attr {\\n    key: &quot;parallel_iterations&quot;\\n    value {\\n      i: 32\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/while/TensorArrayWrite/TensorArrayWriteV3&quot;\\n  op: &quot;TensorArrayWriteV3&quot;\\n  input: &quot;rnn/while/TensorArrayWrite/TensorArrayWriteV3/Enter&quot;\\n  input: &quot;rnn/while/Identity&quot;\\n  input: &quot;rnn/while/basic_rnn_cell/Tanh&quot;\\n  input: &quot;rnn/while/Identity_1&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;_class&quot;\\n    value {\\n      list {\\n        s: &quot;loc:@rnn/TensorArray&quot;\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/while/add/y&quot;\\n  op: &quot;Const&quot;\\n  input: &quot;^rnn/while/Identity&quot;\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_INT32\\n        tensor_shape {\\n        }\\n        int_val: 1\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/while/add&quot;\\n  op: &quot;Add&quot;\\n  input: &quot;rnn/while/Identity&quot;\\n  input: &quot;rnn/while/add/y&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/while/NextIteration&quot;\\n  op: &quot;NextIteration&quot;\\n  input: &quot;rnn/while/add&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/while/NextIteration_1&quot;\\n  op: &quot;NextIteration&quot;\\n  input: &quot;rnn/while/TensorArrayWrite/TensorArrayWriteV3&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/while/NextIteration_2&quot;\\n  op: &quot;NextIteration&quot;\\n  input: &quot;rnn/while/basic_rnn_cell/Tanh&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/while/Exit&quot;\\n  op: &quot;Exit&quot;\\n  input: &quot;rnn/while/Switch&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/while/Exit_1&quot;\\n  op: &quot;Exit&quot;\\n  input: &quot;rnn/while/Switch_1&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/while/Exit_2&quot;\\n  op: &quot;Exit&quot;\\n  input: &quot;rnn/while/Switch_2&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/TensorArrayStack/TensorArraySizeV3&quot;\\n  op: &quot;TensorArraySizeV3&quot;\\n  input: &quot;rnn/TensorArray&quot;\\n  input: &quot;rnn/while/Exit_1&quot;\\n  attr {\\n    key: &quot;_class&quot;\\n    value {\\n      list {\\n        s: &quot;loc:@rnn/TensorArray&quot;\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/TensorArrayStack/range/start&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;_class&quot;\\n    value {\\n      list {\\n        s: &quot;loc:@rnn/TensorArray&quot;\\n      }\\n    }\\n  }\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_INT32\\n        tensor_shape {\\n        }\\n        int_val: 0\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/TensorArrayStack/range/delta&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;_class&quot;\\n    value {\\n      list {\\n        s: &quot;loc:@rnn/TensorArray&quot;\\n      }\\n    }\\n  }\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_INT32\\n        tensor_shape {\\n        }\\n        int_val: 1\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/TensorArrayStack/range&quot;\\n  op: &quot;Range&quot;\\n  input: &quot;rnn/TensorArrayStack/range/start&quot;\\n  input: &quot;rnn/TensorArrayStack/TensorArraySizeV3&quot;\\n  input: &quot;rnn/TensorArrayStack/range/delta&quot;\\n  attr {\\n    key: &quot;Tidx&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;_class&quot;\\n    value {\\n      list {\\n        s: &quot;loc:@rnn/TensorArray&quot;\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/TensorArrayStack/TensorArrayGatherV3&quot;\\n  op: &quot;TensorArrayGatherV3&quot;\\n  input: &quot;rnn/TensorArray&quot;\\n  input: &quot;rnn/TensorArrayStack/range&quot;\\n  input: &quot;rnn/while/Exit_1&quot;\\n  attr {\\n    key: &quot;_class&quot;\\n    value {\\n      list {\\n        s: &quot;loc:@rnn/TensorArray&quot;\\n      }\\n    }\\n  }\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;element_shape&quot;\\n    value {\\n      shape {\\n        dim {\\n          size: -1\\n        }\\n        dim {\\n          size: 5\\n        }\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/transpose/perm&quot;\\n  op: &quot;Const&quot;\\n  attr {\\n    key: &quot;dtype&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n  attr {\\n    key: &quot;value&quot;\\n    value {\\n      tensor {\\n        dtype: DT_INT32\\n        tensor_shape {\\n          dim {\\n            size: 3\\n          }\\n        }\\n        tensor_content: &quot;\\\\001\\\\000\\\\000\\\\000\\\\000\\\\000\\\\000\\\\000\\\\002\\\\000\\\\000\\\\000&quot;\\n      }\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;rnn/transpose&quot;\\n  op: &quot;Transpose&quot;\\n  input: &quot;rnn/TensorArrayStack/TensorArrayGatherV3&quot;\\n  input: &quot;rnn/transpose/perm&quot;\\n  attr {\\n    key: &quot;T&quot;\\n    value {\\n      type: DT_FLOAT\\n    }\\n  }\\n  attr {\\n    key: &quot;Tperm&quot;\\n    value {\\n      type: DT_INT32\\n    }\\n  }\\n}\\nnode {\\n  name: &quot;init&quot;\\n  op: &quot;NoOp&quot;\\n  input: &quot;^rnn/basic_rnn_cell/weights/Assign&quot;\\n  input: &quot;^rnn/basic_rnn_cell/biases/Assign&quot;\\n}\\n';\n",
       "          }\n",
       "        </script>\n",
       "        <link rel=&quot;import&quot; href=&quot;https://tensorboard.appspot.com/tf-graph-basic.build.html&quot; onload=load()>\n",
       "        <div style=&quot;height:600px&quot;>\n",
       "          <tf-graph-basic id=&quot;graph0.9507143064099162&quot;></tf-graph-basic>\n",
       "        </div>\n",
       "    \"></iframe>\n",
       "    "
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "show_graph(tf.get_default_graph())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Setting the sequence lengths"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "tf.reset_default_graph()\n",
    "\n",
    "n_steps = 2\n",
    "n_inputs = 3\n",
    "n_neurons = 5\n",
    "\n",
    "X = tf.placeholder(tf.float32, [None, n_steps, n_inputs])\n",
    "seq_length = tf.placeholder(tf.int32, [None])\n",
    "\n",
    "basic_cell = tf.contrib.rnn.BasicRNNCell(num_units=n_neurons)\n",
    "outputs, states = tf.nn.dynamic_rnn(basic_cell, X, sequence_length=seq_length, dtype=tf.float32)\n",
    "\n",
    "init = tf.global_variables_initializer()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "X_batch = np.array([\n",
    "        # step 0     step 1\n",
    "        [[0, 1, 2], [9, 8, 7]], # instance 1\n",
    "        [[3, 4, 5], [0, 0, 0]], # instance 2 (padded with zero vectors)\n",
    "        [[6, 7, 8], [6, 5, 4]], # instance 3\n",
    "        [[9, 0, 1], [3, 2, 1]], # instance 4\n",
    "    ])\n",
    "seq_length_batch = np.array([2, 1, 2, 2])\n",
    "\n",
    "with tf.Session() as sess:\n",
    "    init.run()\n",
    "    outputs_val, states_val = sess.run(\n",
    "        [outputs, states], feed_dict={X: X_batch, seq_length: seq_length_batch})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[[-0.6111083   0.49218178  0.8701691  -0.04231749  0.54580343]\n",
      "  [-0.99996614 -0.99676019  0.99999988 -0.99485856  0.6408028 ]]\n",
      "\n",
      " [[-0.9907124  -0.30499423  0.99963558 -0.75243163  0.79337811]\n",
      "  [ 0.          0.          0.          0.          0.        ]]\n",
      "\n",
      " [[-0.9998197  -0.82394147  0.99999905 -0.95748097  0.91355109]\n",
      "  [-0.9996509  -0.99599117  0.99998784 -0.99218011  0.57273889]]\n",
      "\n",
      " [[-0.99922258 -0.99998683  0.99476534 -0.01298772  0.37090829]\n",
      "  [-0.89854306 -0.97039658  0.99201375 -0.83898473  0.58962703]]]\n"
     ]
    }
   ],
   "source": [
    "print(outputs_val)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[-0.99996614 -0.99676019  0.99999988 -0.99485856  0.6408028 ]\n",
      " [-0.9907124  -0.30499423  0.99963558 -0.75243163  0.79337811]\n",
      " [-0.9996509  -0.99599117  0.99998784 -0.99218011  0.57273889]\n",
      " [-0.89854306 -0.97039658  0.99201375 -0.83898473  0.58962703]]\n"
     ]
    }
   ],
   "source": [
    "print(states_val)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Training a sequence classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "tf.reset_default_graph()\n",
    "\n",
    "from tensorflow.contrib.layers import fully_connected\n",
    "\n",
    "n_steps = 28\n",
    "n_inputs = 28\n",
    "n_neurons = 150\n",
    "n_outputs = 10\n",
    "\n",
    "learning_rate = 0.001\n",
    "\n",
    "X = tf.placeholder(tf.float32, [None, n_steps, n_inputs])\n",
    "y = tf.placeholder(tf.int32, [None])\n",
    "\n",
    "with tf.variable_scope(\"rnn\", initializer=tf.contrib.layers.variance_scaling_initializer()):\n",
    "    basic_cell = tf.contrib.rnn.BasicRNNCell(num_units=n_neurons, activation=tf.nn.relu)\n",
    "    outputs, states = tf.nn.dynamic_rnn(basic_cell, X, dtype=tf.float32)\n",
    "\n",
    "logits = fully_connected(states, n_outputs, activation_fn=None)\n",
    "xentropy = tf.nn.sparse_softmax_cross_entropy_with_logits(labels=y, logits=logits)\n",
    "loss = tf.reduce_mean(xentropy)\n",
    "optimizer = tf.train.AdamOptimizer(learning_rate=learning_rate)\n",
    "training_op = optimizer.minimize(loss)\n",
    "correct = tf.nn.in_top_k(logits, y, 1)\n",
    "accuracy = tf.reduce_mean(tf.cast(correct, tf.float32))\n",
    "\n",
    "init = tf.global_variables_initializer()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Extracting /tmp/data/train-images-idx3-ubyte.gz\n",
      "Extracting /tmp/data/train-labels-idx1-ubyte.gz\n",
      "Extracting /tmp/data/t10k-images-idx3-ubyte.gz\n",
      "Extracting /tmp/data/t10k-labels-idx1-ubyte.gz\n"
     ]
    }
   ],
   "source": [
    "from tensorflow.examples.tutorials.mnist import input_data\n",
    "mnist = input_data.read_data_sets(\"/tmp/data/\")\n",
    "X_test = mnist.test.images.reshape((-1, n_steps, n_inputs))\n",
    "y_test = mnist.test.labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0 Train accuracy: 0.973333 Test accuracy: 0.8671\n",
      "1 Train accuracy: 0.946667 Test accuracy: 0.9438\n",
      "2 Train accuracy: 0.94 Test accuracy: 0.944\n",
      "3 Train accuracy: 0.953333 Test accuracy: 0.9541\n",
      "4 Train accuracy: 0.98 Test accuracy: 0.9585\n",
      "5 Train accuracy: 0.966667 Test accuracy: 0.9643\n",
      "6 Train accuracy: 0.966667 Test accuracy: 0.9639\n",
      "7 Train accuracy: 0.973333 Test accuracy: 0.9648\n",
      "8 Train accuracy: 0.973333 Test accuracy: 0.9634\n",
      "9 Train accuracy: 0.98 Test accuracy: 0.9637\n",
      "10 Train accuracy: 0.98 Test accuracy: 0.9622\n",
      "11 Train accuracy: 0.96 Test accuracy: 0.9658\n",
      "12 Train accuracy: 0.986667 Test accuracy: 0.9704\n",
      "13 Train accuracy: 1.0 Test accuracy: 0.9685\n",
      "14 Train accuracy: 0.966667 Test accuracy: 0.969\n",
      "15 Train accuracy: 0.993333 Test accuracy: 0.9724\n",
      "16 Train accuracy: 0.993333 Test accuracy: 0.9746\n",
      "17 Train accuracy: 0.986667 Test accuracy: 0.9657\n",
      "18 Train accuracy: 0.986667 Test accuracy: 0.9734\n",
      "19 Train accuracy: 0.986667 Test accuracy: 0.9739\n",
      "20 Train accuracy: 0.98 Test accuracy: 0.9741\n",
      "21 Train accuracy: 0.986667 Test accuracy: 0.9762\n",
      "22 Train accuracy: 0.986667 Test accuracy: 0.9757\n",
      "23 Train accuracy: 0.973333 Test accuracy: 0.9674\n",
      "24 Train accuracy: 0.993333 Test accuracy: 0.9746\n",
      "25 Train accuracy: 0.96 Test accuracy: 0.9704\n",
      "26 Train accuracy: 0.98 Test accuracy: 0.9721\n",
      "27 Train accuracy: 0.98 Test accuracy: 0.9717\n",
      "28 Train accuracy: 1.0 Test accuracy: 0.9744\n",
      "29 Train accuracy: 0.986667 Test accuracy: 0.9731\n",
      "30 Train accuracy: 0.986667 Test accuracy: 0.9779\n",
      "31 Train accuracy: 1.0 Test accuracy: 0.9768\n",
      "32 Train accuracy: 0.993333 Test accuracy: 0.9748\n",
      "33 Train accuracy: 1.0 Test accuracy: 0.9719\n",
      "34 Train accuracy: 0.986667 Test accuracy: 0.9773\n",
      "35 Train accuracy: 0.993333 Test accuracy: 0.9755\n",
      "36 Train accuracy: 0.993333 Test accuracy: 0.9792\n",
      "37 Train accuracy: 0.993333 Test accuracy: 0.974\n",
      "38 Train accuracy: 0.986667 Test accuracy: 0.9771\n",
      "39 Train accuracy: 0.986667 Test accuracy: 0.9788\n",
      "40 Train accuracy: 0.993333 Test accuracy: 0.9756\n",
      "41 Train accuracy: 0.986667 Test accuracy: 0.9723\n",
      "42 Train accuracy: 1.0 Test accuracy: 0.9769\n",
      "43 Train accuracy: 0.993333 Test accuracy: 0.979\n",
      "44 Train accuracy: 0.98 Test accuracy: 0.9745\n",
      "45 Train accuracy: 0.986667 Test accuracy: 0.9777\n",
      "46 Train accuracy: 0.993333 Test accuracy: 0.9761\n",
      "47 Train accuracy: 0.993333 Test accuracy: 0.9759\n",
      "48 Train accuracy: 0.986667 Test accuracy: 0.9776\n",
      "49 Train accuracy: 0.993333 Test accuracy: 0.9778\n",
      "50 Train accuracy: 0.986667 Test accuracy: 0.977\n",
      "51 Train accuracy: 0.986667 Test accuracy: 0.9785\n",
      "52 Train accuracy: 0.986667 Test accuracy: 0.9751\n",
      "53 Train accuracy: 1.0 Test accuracy: 0.9799\n",
      "54 Train accuracy: 1.0 Test accuracy: 0.9756\n",
      "55 Train accuracy: 0.986667 Test accuracy: 0.9762\n",
      "56 Train accuracy: 0.993333 Test accuracy: 0.9804\n",
      "57 Train accuracy: 1.0 Test accuracy: 0.9796\n",
      "58 Train accuracy: 0.986667 Test accuracy: 0.9799\n",
      "59 Train accuracy: 0.993333 Test accuracy: 0.9807\n",
      "60 Train accuracy: 0.98 Test accuracy: 0.9777\n",
      "61 Train accuracy: 0.986667 Test accuracy: 0.9751\n",
      "62 Train accuracy: 1.0 Test accuracy: 0.9784\n",
      "63 Train accuracy: 1.0 Test accuracy: 0.9778\n",
      "64 Train accuracy: 0.993333 Test accuracy: 0.9793\n",
      "65 Train accuracy: 0.993333 Test accuracy: 0.9794\n",
      "66 Train accuracy: 0.993333 Test accuracy: 0.9797\n",
      "67 Train accuracy: 1.0 Test accuracy: 0.981\n",
      "68 Train accuracy: 1.0 Test accuracy: 0.9786\n",
      "69 Train accuracy: 0.993333 Test accuracy: 0.972\n",
      "70 Train accuracy: 0.993333 Test accuracy: 0.9811\n",
      "71 Train accuracy: 0.993333 Test accuracy: 0.9778\n",
      "72 Train accuracy: 1.0 Test accuracy: 0.9803\n",
      "73 Train accuracy: 1.0 Test accuracy: 0.9783\n",
      "74 Train accuracy: 1.0 Test accuracy: 0.9797\n",
      "75 Train accuracy: 0.986667 Test accuracy: 0.979\n",
      "76 Train accuracy: 0.993333 Test accuracy: 0.9793\n",
      "77 Train accuracy: 1.0 Test accuracy: 0.9784\n",
      "78 Train accuracy: 0.993333 Test accuracy: 0.9751\n",
      "79 Train accuracy: 0.993333 Test accuracy: 0.9793\n",
      "80 Train accuracy: 0.993333 Test accuracy: 0.977\n",
      "81 Train accuracy: 1.0 Test accuracy: 0.9789\n",
      "82 Train accuracy: 0.993333 Test accuracy: 0.9779\n",
      "83 Train accuracy: 1.0 Test accuracy: 0.9799\n",
      "84 Train accuracy: 0.986667 Test accuracy: 0.9769\n",
      "85 Train accuracy: 0.986667 Test accuracy: 0.9793\n",
      "86 Train accuracy: 1.0 Test accuracy: 0.9772\n",
      "87 Train accuracy: 0.98 Test accuracy: 0.9795\n",
      "88 Train accuracy: 1.0 Test accuracy: 0.981\n",
      "89 Train accuracy: 0.993333 Test accuracy: 0.9785\n",
      "90 Train accuracy: 0.986667 Test accuracy: 0.9788\n",
      "91 Train accuracy: 0.98 Test accuracy: 0.9747\n",
      "92 Train accuracy: 1.0 Test accuracy: 0.9797\n",
      "93 Train accuracy: 1.0 Test accuracy: 0.9801\n",
      "94 Train accuracy: 1.0 Test accuracy: 0.9807\n",
      "95 Train accuracy: 1.0 Test accuracy: 0.979\n",
      "96 Train accuracy: 0.993333 Test accuracy: 0.9809\n",
      "97 Train accuracy: 1.0 Test accuracy: 0.9775\n",
      "98 Train accuracy: 0.986667 Test accuracy: 0.9795\n",
      "99 Train accuracy: 1.0 Test accuracy: 0.9809\n"
     ]
    }
   ],
   "source": [
    "n_epochs = 100\n",
    "batch_size = 150\n",
    "\n",
    "with tf.Session() as sess:\n",
    "    init.run()\n",
    "    for epoch in range(n_epochs):\n",
    "        for iteration in range(mnist.train.num_examples // batch_size):\n",
    "            X_batch, y_batch = mnist.train.next_batch(batch_size)\n",
    "            X_batch = X_batch.reshape((-1, n_steps, n_inputs))\n",
    "            sess.run(training_op, feed_dict={X: X_batch, y: y_batch})\n",
    "        acc_train = accuracy.eval(feed_dict={X: X_batch, y: y_batch})\n",
    "        acc_test = accuracy.eval(feed_dict={X: X_test, y: y_test})\n",
    "        print(epoch, \"Train accuracy:\", acc_train, \"Test accuracy:\", acc_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "# Multi-layer RNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "tf.reset_default_graph()\n",
    "\n",
    "from tensorflow.contrib.layers import fully_connected\n",
    "\n",
    "n_steps = 28\n",
    "n_inputs = 28\n",
    "n_neurons1 = 150\n",
    "n_neurons2 = 100\n",
    "n_outputs = 10\n",
    "\n",
    "learning_rate = 0.001\n",
    "\n",
    "X = tf.placeholder(tf.float32, [None, n_steps, n_inputs])\n",
    "y = tf.placeholder(tf.int32, [None])\n",
    "\n",
    "hidden1 = tf.contrib.rnn.BasicRNNCell(num_units=n_neurons1, activation=tf.nn.relu)\n",
    "hidden2 = tf.contrib.rnn.BasicRNNCell(num_units=n_neurons2, activation=tf.nn.relu)\n",
    "multi_layer_cell = tf.contrib.rnn.MultiRNNCell([hidden1, hidden2])\n",
    "outputs, states_tuple = tf.nn.dynamic_rnn(multi_layer_cell, X, dtype=tf.float32)\n",
    "states = tf.concat(axis=1, values=states_tuple)\n",
    "logits = fully_connected(states, n_outputs, activation_fn=None)\n",
    "xentropy = tf.nn.sparse_softmax_cross_entropy_with_logits(labels=y, logits=logits)\n",
    "loss = tf.reduce_mean(xentropy)\n",
    "optimizer = tf.train.AdamOptimizer(learning_rate=learning_rate)\n",
    "training_op = optimizer.minimize(loss)\n",
    "correct = tf.nn.in_top_k(logits, y, 1)\n",
    "accuracy = tf.reduce_mean(tf.cast(correct, tf.float32))\n",
    "\n",
    "init = tf.global_variables_initializer()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0 Train accuracy: 0.94 Test accuracy: 0.9272\n",
      "1 Train accuracy: 0.986667 Test accuracy: 0.9504\n",
      "2 Train accuracy: 0.98 Test accuracy: 0.9667\n",
      "3 Train accuracy: 0.98 Test accuracy: 0.9672\n",
      "4 Train accuracy: 0.98 Test accuracy: 0.9706\n",
      "5 Train accuracy: 0.973333 Test accuracy: 0.9737\n",
      "6 Train accuracy: 0.973333 Test accuracy: 0.9725\n",
      "7 Train accuracy: 0.993333 Test accuracy: 0.9774\n",
      "8 Train accuracy: 0.993333 Test accuracy: 0.9795\n",
      "9 Train accuracy: 0.98 Test accuracy: 0.9748\n",
      "10 Train accuracy: 0.986667 Test accuracy: 0.9799\n",
      "11 Train accuracy: 0.98 Test accuracy: 0.9782\n",
      "12 Train accuracy: 0.973333 Test accuracy: 0.9823\n",
      "13 Train accuracy: 0.98 Test accuracy: 0.9795\n",
      "14 Train accuracy: 0.986667 Test accuracy: 0.9814\n",
      "15 Train accuracy: 0.993333 Test accuracy: 0.9787\n",
      "16 Train accuracy: 0.993333 Test accuracy: 0.9757\n",
      "17 Train accuracy: 0.98 Test accuracy: 0.9801\n",
      "18 Train accuracy: 1.0 Test accuracy: 0.9813\n",
      "19 Train accuracy: 0.993333 Test accuracy: 0.9838\n",
      "20 Train accuracy: 0.993333 Test accuracy: 0.9823\n",
      "21 Train accuracy: 0.993333 Test accuracy: 0.9849\n",
      "22 Train accuracy: 0.986667 Test accuracy: 0.9828\n",
      "23 Train accuracy: 0.993333 Test accuracy: 0.9795\n",
      "24 Train accuracy: 0.993333 Test accuracy: 0.9823\n",
      "25 Train accuracy: 0.993333 Test accuracy: 0.982\n",
      "26 Train accuracy: 1.0 Test accuracy: 0.9825\n",
      "27 Train accuracy: 1.0 Test accuracy: 0.9869\n",
      "28 Train accuracy: 1.0 Test accuracy: 0.9829\n",
      "29 Train accuracy: 1.0 Test accuracy: 0.9848\n",
      "30 Train accuracy: 0.986667 Test accuracy: 0.9842\n",
      "31 Train accuracy: 0.993333 Test accuracy: 0.9841\n",
      "32 Train accuracy: 1.0 Test accuracy: 0.9847\n",
      "33 Train accuracy: 0.993333 Test accuracy: 0.9861\n",
      "34 Train accuracy: 0.98 Test accuracy: 0.9843\n",
      "35 Train accuracy: 1.0 Test accuracy: 0.9849\n",
      "36 Train accuracy: 1.0 Test accuracy: 0.9856\n",
      "37 Train accuracy: 1.0 Test accuracy: 0.9872\n",
      "38 Train accuracy: 0.993333 Test accuracy: 0.9862\n",
      "39 Train accuracy: 0.986667 Test accuracy: 0.9841\n",
      "40 Train accuracy: 0.993333 Test accuracy: 0.9856\n",
      "41 Train accuracy: 1.0 Test accuracy: 0.9864\n",
      "42 Train accuracy: 1.0 Test accuracy: 0.9844\n",
      "43 Train accuracy: 1.0 Test accuracy: 0.9842\n",
      "44 Train accuracy: 0.986667 Test accuracy: 0.9863\n",
      "45 Train accuracy: 0.993333 Test accuracy: 0.9835\n",
      "46 Train accuracy: 1.0 Test accuracy: 0.986\n",
      "47 Train accuracy: 0.986667 Test accuracy: 0.9834\n",
      "48 Train accuracy: 0.993333 Test accuracy: 0.9832\n",
      "49 Train accuracy: 1.0 Test accuracy: 0.9882\n",
      "50 Train accuracy: 0.993333 Test accuracy: 0.9865\n",
      "51 Train accuracy: 1.0 Test accuracy: 0.9879\n",
      "52 Train accuracy: 1.0 Test accuracy: 0.9883\n",
      "53 Train accuracy: 0.993333 Test accuracy: 0.9873\n",
      "54 Train accuracy: 1.0 Test accuracy: 0.9849\n",
      "55 Train accuracy: 1.0 Test accuracy: 0.9873\n",
      "56 Train accuracy: 1.0 Test accuracy: 0.9857\n",
      "57 Train accuracy: 1.0 Test accuracy: 0.9878\n",
      "58 Train accuracy: 1.0 Test accuracy: 0.987\n",
      "59 Train accuracy: 0.993333 Test accuracy: 0.9872\n",
      "60 Train accuracy: 0.986667 Test accuracy: 0.9873\n",
      "61 Train accuracy: 0.993333 Test accuracy: 0.9873\n",
      "62 Train accuracy: 1.0 Test accuracy: 0.9885\n",
      "63 Train accuracy: 0.986667 Test accuracy: 0.9849\n",
      "64 Train accuracy: 1.0 Test accuracy: 0.9882\n",
      "65 Train accuracy: 1.0 Test accuracy: 0.9887\n",
      "66 Train accuracy: 0.986667 Test accuracy: 0.9829\n",
      "67 Train accuracy: 1.0 Test accuracy: 0.9877\n",
      "68 Train accuracy: 1.0 Test accuracy: 0.9865\n",
      "69 Train accuracy: 1.0 Test accuracy: 0.9886\n",
      "70 Train accuracy: 1.0 Test accuracy: 0.9881\n",
      "71 Train accuracy: 1.0 Test accuracy: 0.9884\n",
      "72 Train accuracy: 1.0 Test accuracy: 0.9836\n",
      "73 Train accuracy: 1.0 Test accuracy: 0.9897\n",
      "74 Train accuracy: 0.993333 Test accuracy: 0.9838\n",
      "75 Train accuracy: 1.0 Test accuracy: 0.9884\n",
      "76 Train accuracy: 0.986667 Test accuracy: 0.9857\n",
      "77 Train accuracy: 0.986667 Test accuracy: 0.9854\n",
      "78 Train accuracy: 0.993333 Test accuracy: 0.9815\n",
      "79 Train accuracy: 1.0 Test accuracy: 0.987\n",
      "80 Train accuracy: 0.98 Test accuracy: 0.9856\n",
      "81 Train accuracy: 1.0 Test accuracy: 0.9857\n",
      "82 Train accuracy: 1.0 Test accuracy: 0.9884\n",
      "83 Train accuracy: 1.0 Test accuracy: 0.9801\n",
      "84 Train accuracy: 1.0 Test accuracy: 0.9868\n",
      "85 Train accuracy: 1.0 Test accuracy: 0.9881\n",
      "86 Train accuracy: 1.0 Test accuracy: 0.9886\n",
      "87 Train accuracy: 0.98 Test accuracy: 0.987\n",
      "88 Train accuracy: 1.0 Test accuracy: 0.9866\n",
      "89 Train accuracy: 0.993333 Test accuracy: 0.9871\n",
      "90 Train accuracy: 0.993333 Test accuracy: 0.9845\n",
      "91 Train accuracy: 1.0 Test accuracy: 0.9856\n",
      "92 Train accuracy: 1.0 Test accuracy: 0.9865\n",
      "93 Train accuracy: 1.0 Test accuracy: 0.9833\n",
      "94 Train accuracy: 0.993333 Test accuracy: 0.985\n",
      "95 Train accuracy: 1.0 Test accuracy: 0.988\n",
      "96 Train accuracy: 0.993333 Test accuracy: 0.9858\n",
      "97 Train accuracy: 1.0 Test accuracy: 0.9882\n",
      "98 Train accuracy: 1.0 Test accuracy: 0.989\n",
      "99 Train accuracy: 1.0 Test accuracy: 0.9893\n"
     ]
    }
   ],
   "source": [
    "n_epochs = 100\n",
    "batch_size = 150\n",
    "\n",
    "with tf.Session() as sess:\n",
    "    init.run()\n",
    "    for epoch in range(n_epochs):\n",
    "        for iteration in range(mnist.train.num_examples // batch_size):\n",
    "            X_batch, y_batch = mnist.train.next_batch(batch_size)\n",
    "            X_batch = X_batch.reshape((-1, n_steps, n_inputs))\n",
    "            sess.run(training_op, feed_dict={X: X_batch, y: y_batch})\n",
    "        acc_train = accuracy.eval(feed_dict={X: X_batch, y: y_batch})\n",
    "        acc_test = accuracy.eval(feed_dict={X: X_test, y: y_test})\n",
    "        print(epoch, \"Train accuracy:\", acc_train, \"Test accuracy:\", acc_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "# Time series"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "t_min, t_max = 0, 30\n",
    "resolution = 0.1\n",
    "\n",
    "def time_series(t):\n",
    "    return t * np.sin(t) / 3 + 2 * np.sin(t*5)\n",
    "\n",
    "def next_batch(batch_size, n_steps):\n",
    "    t0 = np.random.rand(batch_size, 1) * (t_max - t_min - n_steps * resolution)\n",
    "    Ts = t0 + np.arange(0., n_steps + 1) * resolution\n",
    "    ys = time_series(Ts)\n",
    "    return ys[:, :-1].reshape(-1, n_steps, 1), ys[:, 1:].reshape(-1, n_steps, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saving figure time_series_plot\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ageron/dev/py/envs/mltst/lib/python3.5/site-packages/ipykernel/__main__.py:1: DeprecationWarning: object of type <class 'float'> cannot be safely interpreted as an integer.\n",
      "  if __name__ == '__main__':\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAxAAAAEYCAYAAADMNRC5AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fa7d3a77da0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "t = np.linspace(t_min, t_max, (t_max - t_min) // resolution)\n",
    "\n",
    "n_steps = 20\n",
    "t_instance = np.linspace(12.2, 12.2 + resolution * (n_steps + 1), n_steps + 1)\n",
    "\n",
    "plt.figure(figsize=(11,4))\n",
    "plt.subplot(121)\n",
    "plt.title(\"A time series (generated)\", fontsize=14)\n",
    "plt.plot(t, time_series(t), label=r\"$t . \\sin(t) / 3 + 2 . \\sin(5t)$\")\n",
    "plt.plot(t_instance[:-1], time_series(t_instance[:-1]), \"b-\", linewidth=3, label=\"A training instance\")\n",
    "plt.legend(loc=\"lower left\", fontsize=14)\n",
    "plt.axis([0, 30, -17, 13])\n",
    "plt.xlabel(\"Time\")\n",
    "plt.ylabel(\"Value\")\n",
    "\n",
    "plt.subplot(122)\n",
    "plt.title(\"A training instance\", fontsize=14)\n",
    "plt.plot(t_instance[:-1], time_series(t_instance[:-1]), \"bo\", markersize=10, label=\"instance\")\n",
    "plt.plot(t_instance[1:], time_series(t_instance[1:]), \"w*\", markersize=10, label=\"target\")\n",
    "plt.legend(loc=\"upper left\")\n",
    "plt.xlabel(\"Time\")\n",
    "\n",
    "\n",
    "save_fig(\"time_series_plot\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "X_batch, y_batch = next_batch(1, n_steps)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 0.3266291 , -0.97081587],\n",
       "       [-0.97081587, -2.09298955],\n",
       "       [-2.09298955, -2.84017944],\n",
       "       [-2.84017944, -3.10464787],\n",
       "       [-3.10464787, -2.89622372],\n",
       "       [-2.89622372, -2.33910087],\n",
       "       [-2.33910087, -1.64063575],\n",
       "       [-1.64063575, -1.03979198],\n",
       "       [-1.03979198, -0.74786335],\n",
       "       [-0.74786335, -0.89599678],\n",
       "       [-0.89599678, -1.50237198],\n",
       "       [-1.50237198, -2.46708161],\n",
       "       [-2.46708161, -3.59597343],\n",
       "       [-3.59597343, -4.64762444],\n",
       "       [-4.64762444, -5.39195331],\n",
       "       [-5.39195331, -5.66612695],\n",
       "       [-5.66612695, -5.41407933],\n",
       "       [-5.41407933, -4.69997176],\n",
       "       [-4.69997176, -3.69230347],\n",
       "       [-3.69230347, -2.6225661 ]])"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.c_[X_batch[0], y_batch[0]]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Using an `OuputProjectionWrapper`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "tf.reset_default_graph()\n",
    "\n",
    "from tensorflow.contrib.layers import fully_connected\n",
    "\n",
    "n_steps = 20\n",
    "n_inputs = 1\n",
    "n_neurons = 100\n",
    "n_outputs = 1\n",
    "\n",
    "X = tf.placeholder(tf.float32, [None, n_steps, n_inputs])\n",
    "y = tf.placeholder(tf.float32, [None, n_steps, n_outputs])\n",
    "\n",
    "cell = tf.contrib.rnn.OutputProjectionWrapper(\n",
    "    tf.contrib.rnn.BasicRNNCell(num_units=n_neurons, activation=tf.nn.relu),\n",
    "    output_size=n_outputs)\n",
    "outputs, states = tf.nn.dynamic_rnn(cell, X, dtype=tf.float32)\n",
    "\n",
    "n_outputs = 1\n",
    "learning_rate = 0.001\n",
    "\n",
    "loss = tf.reduce_sum(tf.square(outputs - y))\n",
    "optimizer = tf.train.AdamOptimizer(learning_rate=learning_rate)\n",
    "training_op = optimizer.minimize(loss)\n",
    "\n",
    "init = tf.global_variables_initializer()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0 \tMSE: 22584.6\n",
      "100 \tMSE: 786.52\n",
      "200 \tMSE: 427.682\n",
      "300 \tMSE: 213.502\n",
      "400 \tMSE: 154.842\n",
      "500 \tMSE: 83.3785\n",
      "600 \tMSE: 56.379\n",
      "700 \tMSE: 57.2104\n",
      "800 \tMSE: 64.23\n",
      "900 \tMSE: 62.5516\n",
      "[[[-3.51760769]\n",
      "  [-2.56235027]\n",
      "  [-1.24227786]\n",
      "  [ 0.61775202]\n",
      "  [ 2.14323711]\n",
      "  [ 2.98960495]\n",
      "  [ 3.46584606]\n",
      "  [ 3.40469408]\n",
      "  [ 2.77786183]\n",
      "  [ 2.13859129]\n",
      "  [ 1.71483374]\n",
      "  [ 1.59727895]\n",
      "  [ 1.89304471]\n",
      "  [ 2.69040084]\n",
      "  [ 3.86570907]\n",
      "  [ 5.14488888]\n",
      "  [ 6.14800119]\n",
      "  [ 6.62051868]\n",
      "  [ 6.62806034]\n",
      "  [ 6.06975651]]]\n"
     ]
    }
   ],
   "source": [
    "n_iterations = 1000\n",
    "batch_size = 50\n",
    "\n",
    "with tf.Session() as sess:\n",
    "    init.run()\n",
    "    for iteration in range(n_iterations):\n",
    "        X_batch, y_batch = next_batch(batch_size, n_steps)\n",
    "        sess.run(training_op, feed_dict={X: X_batch, y: y_batch})\n",
    "        if iteration % 100 == 0:\n",
    "            mse = loss.eval(feed_dict={X: X_batch, y: y_batch})\n",
    "            print(iteration, \"\\tMSE:\", mse)\n",
    "    \n",
    "    X_new = time_series(np.array(t_instance[:-1].reshape(-1, n_steps, n_inputs)))\n",
    "    y_pred = sess.run(outputs, feed_dict={X: X_new})\n",
    "    print(y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saving figure time_series_pred_plot\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAagAAAEYCAYAAAAJeGK1AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fa7d8133048>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.title(\"Testing the model\", fontsize=14)\n",
    "plt.plot(t_instance[:-1], time_series(t_instance[:-1]), \"bo\", markersize=10, label=\"instance\")\n",
    "plt.plot(t_instance[1:], time_series(t_instance[1:]), \"w*\", markersize=10, label=\"target\")\n",
    "plt.plot(t_instance[1:], y_pred[0,:,0], \"r.\", markersize=10, label=\"prediction\")\n",
    "plt.legend(loc=\"upper left\")\n",
    "plt.xlabel(\"Time\")\n",
    "\n",
    "save_fig(\"time_series_pred_plot\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Without using an `OutputProjectionWrapper`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "tf.reset_default_graph()\n",
    "\n",
    "from tensorflow.contrib.layers import fully_connected\n",
    "\n",
    "n_steps = 20\n",
    "n_inputs = 1\n",
    "n_neurons = 100\n",
    "\n",
    "X = tf.placeholder(tf.float32, [None, n_steps, n_inputs])\n",
    "y = tf.placeholder(tf.float32, [None, n_steps, n_outputs])\n",
    "\n",
    "basic_cell = tf.contrib.rnn.BasicRNNCell(num_units=n_neurons, activation=tf.nn.relu)\n",
    "rnn_outputs, states = tf.nn.dynamic_rnn(basic_cell, X, dtype=tf.float32)\n",
    "\n",
    "n_outputs = 1\n",
    "learning_rate = 0.001\n",
    "\n",
    "stacked_rnn_outputs = tf.reshape(rnn_outputs, [-1, n_neurons])\n",
    "stacked_outputs = fully_connected(stacked_rnn_outputs, n_outputs, activation_fn=None)\n",
    "outputs = tf.reshape(stacked_outputs, [-1, n_steps, n_outputs])\n",
    "\n",
    "loss = tf.reduce_sum(tf.square(outputs - y))\n",
    "optimizer = tf.train.AdamOptimizer(learning_rate=learning_rate)\n",
    "training_op = optimizer.minimize(loss)\n",
    "\n",
    "init = tf.global_variables_initializer()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0 \tMSE: 17691.8\n",
      "100 \tMSE: 687.35\n",
      "200 \tMSE: 267.795\n",
      "300 \tMSE: 125.488\n",
      "400 \tMSE: 71.624\n",
      "500 \tMSE: 50.981\n",
      "600 \tMSE: 52.506\n",
      "700 \tMSE: 49.3854\n",
      "800 \tMSE: 51.2834\n",
      "900 \tMSE: 43.3292\n",
      "[[[-3.49691629]\n",
      "  [-2.50379086]\n",
      "  [-1.15322065]\n",
      "  [ 0.6370607 ]\n",
      "  [ 2.08771992]\n",
      "  [ 2.97146773]\n",
      "  [ 3.42202783]\n",
      "  [ 3.31741285]\n",
      "  [ 2.79457593]\n",
      "  [ 2.14914632]\n",
      "  [ 1.71075094]\n",
      "  [ 1.60508287]\n",
      "  [ 1.92197335]\n",
      "  [ 2.71487164]\n",
      "  [ 3.87283397]\n",
      "  [ 5.07521629]\n",
      "  [ 6.01868391]\n",
      "  [ 6.52164173]\n",
      "  [ 6.515872  ]\n",
      "  [ 5.958179  ]]]\n"
     ]
    }
   ],
   "source": [
    "n_iterations = 1000\n",
    "batch_size = 50\n",
    "\n",
    "with tf.Session() as sess:\n",
    "    init.run()\n",
    "    for iteration in range(n_iterations):\n",
    "        X_batch, y_batch = next_batch(batch_size, n_steps)\n",
    "        sess.run(training_op, feed_dict={X: X_batch, y: y_batch})\n",
    "        if iteration % 100 == 0:\n",
    "            mse = loss.eval(feed_dict={X: X_batch, y: y_batch})\n",
    "            print(iteration, \"\\tMSE:\", mse)\n",
    "    \n",
    "    X_new = time_series(np.array(t_instance[:-1].reshape(-1, n_steps, n_inputs)))\n",
    "    y_pred = sess.run(outputs, feed_dict={X: X_new})\n",
    "    print(y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYAAAAEhCAYAAAB2h3f0AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nBO7a0QAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fa7d812c400>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.title(\"Testing the model\", fontsize=14)\n",
    "plt.plot(t_instance[:-1], time_series(t_instance[:-1]), \"bo\", markersize=10, label=\"instance\")\n",
    "plt.plot(t_instance[1:], time_series(t_instance[1:]), \"w*\", markersize=10, label=\"target\")\n",
    "plt.plot(t_instance[1:], y_pred[0,:,0], \"r.\", markersize=10, label=\"prediction\")\n",
    "plt.legend(loc=\"upper left\")\n",
    "plt.xlabel(\"Time\")\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Generating a creative new sequence"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0 \tMSE: 14167.3\n",
      "100 \tMSE: 480.774\n",
      "200 \tMSE: 194.57\n",
      "300 \tMSE: 66.9419\n",
      "400 \tMSE: 79.4116\n",
      "500 \tMSE: 46.7239\n",
      "600 \tMSE: 48.8176\n",
      "700 \tMSE: 45.1877\n",
      "800 \tMSE: 46.8532\n",
      "900 \tMSE: 39.5473\n",
      "1000 \tMSE: 46.5509\n",
      "1100 \tMSE: 45.1024\n",
      "1200 \tMSE: 48.5357\n",
      "1300 \tMSE: 38.5474\n",
      "1400 \tMSE: 49.4831\n",
      "1500 \tMSE: 45.0398\n",
      "1600 \tMSE: 50.1059\n",
      "1700 \tMSE: 44.9415\n",
      "1800 \tMSE: 42.2207\n",
      "1900 \tMSE: 49.906\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAq8AAAEYCAYAAAByRh/NAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fa7d80f1588>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "n_iterations = 2000\n",
    "batch_size = 50\n",
    "with tf.Session() as sess:\n",
    "    init.run()\n",
    "    for iteration in range(n_iterations):\n",
    "        X_batch, y_batch = next_batch(batch_size, n_steps)\n",
    "        sess.run(training_op, feed_dict={X: X_batch, y: y_batch})\n",
    "        if iteration % 100 == 0:\n",
    "            mse = loss.eval(feed_dict={X: X_batch, y: y_batch})\n",
    "            print(iteration, \"\\tMSE:\", mse)\n",
    "\n",
    "    sequence1 = [0. for i in range(n_steps)]\n",
    "    for iteration in range(len(t) - n_steps):\n",
    "        X_batch = np.array(sequence1[-n_steps:]).reshape(1, n_steps, 1)\n",
    "        y_pred = sess.run(outputs, feed_dict={X: X_batch})\n",
    "        sequence1.append(y_pred[0, -1, 0])\n",
    "\n",
    "    sequence2 = [time_series(i * resolution + t_min + (t_max-t_min/3)) for i in range(n_steps)]\n",
    "    for iteration in range(len(t) - n_steps):\n",
    "        X_batch = np.array(sequence2[-n_steps:]).reshape(1, n_steps, 1)\n",
    "        y_pred = sess.run(outputs, feed_dict={X: X_batch})\n",
    "        sequence2.append(y_pred[0, -1, 0])\n",
    "\n",
    "plt.figure(figsize=(11,4))\n",
    "plt.subplot(121)\n",
    "plt.plot(t, sequence1, \"b-\")\n",
    "plt.plot(t[:n_steps], sequence1[:n_steps], \"b-\", linewidth=3)\n",
    "plt.xlabel(\"Time\")\n",
    "plt.ylabel(\"Value\")\n",
    "\n",
    "plt.subplot(122)\n",
    "plt.plot(t, sequence2, \"b-\")\n",
    "plt.plot(t[:n_steps], sequence2[:n_steps], \"b-\", linewidth=3)\n",
    "plt.xlabel(\"Time\")\n",
    "#save_fig(\"creative_sequence_plot\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "# Deep RNN"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## MultiRNNCell"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "tf.reset_default_graph()\n",
    "\n",
    "n_inputs = 2\n",
    "n_neurons = 100\n",
    "n_layers = 3\n",
    "n_steps = 5\n",
    "keep_prob = 0.5\n",
    "\n",
    "X = tf.placeholder(tf.float32, [None, n_steps, n_inputs])\n",
    "basic_cell = tf.contrib.rnn.BasicRNNCell(num_units=n_neurons)\n",
    "multi_layer_cell = tf.contrib.rnn.MultiRNNCell([basic_cell] * n_layers)\n",
    "outputs, states = tf.nn.dynamic_rnn(multi_layer_cell, X, dtype=tf.float32)\n",
    "\n",
    "init = tf.global_variables_initializer()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "X_batch = rnd.rand(2, n_steps, n_inputs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "with tf.Session() as sess:\n",
    "    init.run()\n",
    "    outputs_val, states_val = sess.run([outputs, states], feed_dict={X: X_batch})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2, 5, 100)"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "outputs_val.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Dropout"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "tf.reset_default_graph()\n",
    "from tensorflow.contrib.layers import fully_connected\n",
    "\n",
    "n_inputs = 1\n",
    "n_neurons = 100\n",
    "n_layers = 3\n",
    "n_steps = 20\n",
    "n_outputs = 1\n",
    "\n",
    "keep_prob = 0.5\n",
    "learning_rate = 0.001\n",
    "\n",
    "is_training = True\n",
    "\n",
    "def deep_rnn_with_dropout(X, y, is_training):\n",
    "    cell = tf.contrib.rnn.BasicRNNCell(num_units=n_neurons)\n",
    "    if is_training:\n",
    "        cell = tf.contrib.rnn.DropoutWrapper(cell, input_keep_prob=keep_prob)\n",
    "    multi_layer_cell = tf.contrib.rnn.MultiRNNCell([cell] * n_layers)\n",
    "    rnn_outputs, states = tf.nn.dynamic_rnn(multi_layer_cell, X, dtype=tf.float32)\n",
    "\n",
    "    stacked_rnn_outputs = tf.reshape(rnn_outputs, [-1, n_neurons])\n",
    "    stacked_outputs = fully_connected(stacked_rnn_outputs, n_outputs, activation_fn=None)\n",
    "    outputs = tf.reshape(stacked_outputs, [-1, n_steps, n_outputs])\n",
    "\n",
    "    loss = tf.reduce_sum(tf.square(outputs - y))\n",
    "    optimizer = tf.train.AdamOptimizer(learning_rate=learning_rate)\n",
    "    training_op = optimizer.minimize(loss)\n",
    "\n",
    "    return outputs, loss, training_op\n",
    "\n",
    "X = tf.placeholder(tf.float32, [None, n_steps, n_inputs])\n",
    "y = tf.placeholder(tf.float32, [None, n_steps, n_outputs])\n",
    "outputs, loss, training_op = deep_rnn_with_dropout(X, y, is_training)\n",
    "init = tf.global_variables_initializer()\n",
    "saver = tf.train.Saver()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0 \tMSE: 13446.5\n",
      "100 \tMSE: 6108.62\n",
      "200 \tMSE: 4628.27\n",
      "300 \tMSE: 4554.2\n",
      "400 \tMSE: 3862.41\n",
      "500 \tMSE: 3427.27\n",
      "600 \tMSE: 2879.59\n",
      "700 \tMSE: 3460.99\n",
      "800 \tMSE: 2547.23\n",
      "900 \tMSE: 2595.12\n",
      "1000 \tMSE: 2987.26\n",
      "1100 \tMSE: 2284.86\n",
      "1200 \tMSE: 2123.55\n",
      "1300 \tMSE: 2566.21\n",
      "1400 \tMSE: 2494.94\n",
      "1500 \tMSE: 2439.69\n",
      "1600 \tMSE: 2611.02\n",
      "1700 \tMSE: 2650.31\n",
      "1800 \tMSE: 2597.95\n",
      "1900 \tMSE: 2258.47\n"
     ]
    }
   ],
   "source": [
    "n_iterations = 2000\n",
    "batch_size = 50\n",
    "\n",
    "with tf.Session() as sess:\n",
    "    if is_training:\n",
    "        init.run()\n",
    "        for iteration in range(n_iterations):\n",
    "            X_batch, y_batch = next_batch(batch_size, n_steps)\n",
    "            sess.run(training_op, feed_dict={X: X_batch, y: y_batch})\n",
    "            if iteration % 100 == 0:\n",
    "                mse = loss.eval(feed_dict={X: X_batch, y: y_batch})\n",
    "                print(iteration, \"\\tMSE:\", mse)\n",
    "        save_path = saver.save(sess, \"/tmp/my_model.ckpt\")\n",
    "    else:\n",
    "        saver.restore(sess, \"/tmp/my_model.ckpt\")\n",
    "        X_new = time_series(np.array(t_instance[:-1].reshape(-1, n_steps, n_inputs)))\n",
    "        y_pred = sess.run(outputs, feed_dict={X: X_new})\n",
    "        \n",
    "        plt.title(\"Testing the model\", fontsize=14)\n",
    "        plt.plot(t_instance[:-1], time_series(t_instance[:-1]), \"bo\", markersize=10, label=\"instance\")\n",
    "        plt.plot(t_instance[1:], time_series(t_instance[1:]), \"w*\", markersize=10, label=\"target\")\n",
    "        plt.plot(t_instance[1:], y_pred[0,:,0], \"r.\", markersize=10, label=\"prediction\")\n",
    "        plt.legend(loc=\"upper left\")\n",
    "        plt.xlabel(\"Time\")\n",
    "        plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "# LSTM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "tf.reset_default_graph()\n",
    "\n",
    "from tensorflow.contrib.layers import fully_connected\n",
    "\n",
    "n_steps = 28\n",
    "n_inputs = 28\n",
    "n_neurons = 150\n",
    "n_outputs = 10\n",
    "\n",
    "learning_rate = 0.001\n",
    "\n",
    "X = tf.placeholder(tf.float32, [None, n_steps, n_inputs])\n",
    "y = tf.placeholder(tf.int32, [None])\n",
    "\n",
    "lstm_cell = tf.contrib.rnn.BasicLSTMCell(num_units=n_neurons)\n",
    "multi_cell = tf.contrib.rnn.MultiRNNCell([lstm_cell]*3)\n",
    "outputs, states = tf.nn.dynamic_rnn(multi_cell, X, dtype=tf.float32)\n",
    "top_layer_h_state = states[-1][1]\n",
    "logits = fully_connected(top_layer_h_state, n_outputs, activation_fn=None, scope=\"softmax\")\n",
    "xentropy = tf.nn.sparse_softmax_cross_entropy_with_logits(labels=y, logits=logits)\n",
    "loss = tf.reduce_mean(xentropy, name=\"loss\")\n",
    "optimizer = tf.train.AdamOptimizer(learning_rate=learning_rate)\n",
    "training_op = optimizer.minimize(loss)\n",
    "correct = tf.nn.in_top_k(logits, y, 1)\n",
    "accuracy = tf.reduce_mean(tf.cast(correct, tf.float32))\n",
    "    \n",
    "init = tf.global_variables_initializer()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(LSTMStateTuple(c=<tf.Tensor 'rnn/while/Exit_2:0' shape=(?, 150) dtype=float32>, h=<tf.Tensor 'rnn/while/Exit_3:0' shape=(?, 150) dtype=float32>),\n",
       " LSTMStateTuple(c=<tf.Tensor 'rnn/while/Exit_4:0' shape=(?, 150) dtype=float32>, h=<tf.Tensor 'rnn/while/Exit_5:0' shape=(?, 150) dtype=float32>),\n",
       " LSTMStateTuple(c=<tf.Tensor 'rnn/while/Exit_6:0' shape=(?, 150) dtype=float32>, h=<tf.Tensor 'rnn/while/Exit_7:0' shape=(?, 150) dtype=float32>))"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "states"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<tf.Tensor 'rnn/while/Exit_7:0' shape=(?, 150) dtype=float32>"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "top_layer_h_state"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 0 Train accuracy = 0.966667 Test accuracy = 0.9486\n",
      "Epoch 1 Train accuracy = 0.993333 Test accuracy = 0.9685\n",
      "Epoch 2 Train accuracy = 0.946667 Test accuracy = 0.9766\n",
      "Epoch 3 Train accuracy = 0.986667 Test accuracy = 0.9768\n",
      "Epoch 4 Train accuracy = 0.98 Test accuracy = 0.9758\n",
      "Epoch 5 Train accuracy = 1.0 Test accuracy = 0.9835\n",
      "Epoch 6 Train accuracy = 0.986667 Test accuracy = 0.9876\n",
      "Epoch 7 Train accuracy = 1.0 Test accuracy = 0.9851\n",
      "Epoch 8 Train accuracy = 1.0 Test accuracy = 0.9865\n",
      "Epoch 9 Train accuracy = 1.0 Test accuracy = 0.9882\n"
     ]
    }
   ],
   "source": [
    "n_epochs = 10\n",
    "batch_size = 150\n",
    "\n",
    "with tf.Session() as sess:\n",
    "    init.run()\n",
    "    for epoch in range(n_epochs):\n",
    "        for iteration in range(mnist.train.num_examples // batch_size):\n",
    "            X_batch, y_batch = mnist.train.next_batch(batch_size)\n",
    "            X_batch = X_batch.reshape((batch_size, n_steps, n_inputs))\n",
    "            sess.run(training_op, feed_dict={X: X_batch, y: y_batch})\n",
    "        acc_train = accuracy.eval(feed_dict={X: X_batch, y: y_batch})\n",
    "        acc_test = accuracy.eval(feed_dict={X: X_test, y: y_test})\n",
    "        print(\"Epoch\", epoch, \"Train accuracy =\", acc_train, \"Test accuracy =\", acc_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "# Distributing layers across devices"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "\n",
    "class DeviceCellWrapper(tf.contrib.rnn.RNNCell):\n",
    "  def __init__(self, device, cell):\n",
    "    self._cell = cell\n",
    "    self._device = device\n",
    "\n",
    "  @property\n",
    "  def state_size(self):\n",
    "    return self._cell.state_size\n",
    "\n",
    "  @property\n",
    "  def output_size(self):\n",
    "    return self._cell.output_size\n",
    "\n",
    "  def __call__(self, inputs, state, scope=None):\n",
    "    with tf.device(self._device):\n",
    "        return self._cell(inputs, state, scope)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "tf.reset_default_graph()\n",
    "\n",
    "n_inputs = 5\n",
    "n_neurons = 100\n",
    "devices = [\"/cpu:0\"]*5\n",
    "n_steps = 20\n",
    "X = tf.placeholder(tf.float32, shape=[None, n_steps, n_inputs])\n",
    "lstm_cells = [DeviceCellWrapper(device, tf.contrib.rnn.BasicRNNCell(num_units=n_neurons))\n",
    "              for device in devices]\n",
    "multi_layer_cell = tf.contrib.rnn.MultiRNNCell(lstm_cells)\n",
    "outputs, states = tf.nn.dynamic_rnn(multi_layer_cell, X, dtype=tf.float32)\n",
    "init = tf.global_variables_initializer()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[[-0.01708752 -0.00816594 -0.08531895 ...,  0.01293303 -0.03598548\n",
      "    0.00492087]\n",
      "  [-0.09519478 -0.00846841 -0.10969562 ...,  0.05941417  0.04731981\n",
      "    0.09693822]\n",
      "  [-0.14912453 -0.10635467 -0.12964553 ..., -0.01562085 -0.04393777\n",
      "   -0.05746868]\n",
      "  ..., \n",
      "  [-0.24811372 -0.02024984 -0.22714953 ...,  0.02804995 -0.0127665\n",
      "   -0.09021875]\n",
      "  [-0.24003756 -0.16685337 -0.15991297 ..., -0.28325534 -0.12053813\n",
      "    0.02008178]\n",
      "  [-0.26741144 -0.00421038 -0.57989764 ..., -0.11607802  0.15951051\n",
      "    0.13105012]]\n",
      "\n",
      " [[-0.03169292  0.01983191 -0.08063509 ...,  0.01142524 -0.07150294\n",
      "   -0.01657837]\n",
      "  [-0.23104219  0.06165724 -0.11865821 ...,  0.01108671  0.02383714\n",
      "    0.09903613]\n",
      "  [-0.16916461 -0.13944981 -0.19668107 ..., -0.01075597 -0.09825428\n",
      "   -0.10360008]\n",
      "  ..., \n",
      "  [-0.10707516  0.15896417 -0.06263893 ..., -0.05023656 -0.06619275\n",
      "   -0.1176419 ]\n",
      "  [-0.2273972   0.0458602  -0.34042427 ..., -0.3957144   0.02994739\n",
      "    0.17840534]\n",
      "  [-0.28331926 -0.08697947 -0.37225938 ..., -0.44854224  0.0617586\n",
      "   -0.03557429]]]\n"
     ]
    }
   ],
   "source": [
    "with tf.Session() as sess:\n",
    "    init.run()\n",
    "    print(sess.run(outputs, feed_dict={X: rnd.rand(2, n_steps, n_inputs)}))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "# Embeddings"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "This section is based on TensorFlow's [Word2Vec tutorial](https://www.tensorflow.org/versions/r0.11/tutorials/word2vec/index.html)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Fetch the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "from six.moves import urllib\n",
    "\n",
    "import errno\n",
    "import os\n",
    "import zipfile\n",
    "\n",
    "WORDS_PATH = \"datasets/words\"\n",
    "WORDS_URL = 'http://mattmahoney.net/dc/text8.zip'\n",
    "\n",
    "def mkdir_p(path):\n",
    "    \"\"\"Create directories, ok if they already exist.\n",
    "    \n",
    "    This is for python 2 support. In python >=3.2, simply use:\n",
    "    >>> os.makedirs(path, exist_ok=True)\n",
    "    \"\"\"\n",
    "    try:\n",
    "        os.makedirs(path)\n",
    "    except OSError as exc:\n",
    "        if exc.errno == errno.EEXIST and os.path.isdir(path):\n",
    "            pass\n",
    "        else:\n",
    "            raise\n",
    "\n",
    "def fetch_words_data(words_url=WORDS_URL, words_path=WORDS_PATH):\n",
    "    os.makedirs(words_path, exist_ok=True)\n",
    "    zip_path = os.path.join(words_path, \"words.zip\")\n",
    "    if not os.path.exists(zip_path):\n",
    "        urllib.request.urlretrieve(words_url, zip_path)\n",
    "    with zipfile.ZipFile(zip_path) as f:\n",
    "        data = f.read(f.namelist()[0])\n",
    "    return data.decode(\"ascii\").split()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "words = fetch_words_data()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['anarchism', 'originated', 'as', 'a', 'term']"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "words[:5]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Build the dictionary"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "from collections import Counter\n",
    "\n",
    "vocabulary_size = 50000\n",
    "\n",
    "vocabulary = [(\"UNK\", None)] + Counter(words).most_common(vocabulary_size - 1)\n",
    "vocabulary = np.array([word for word, _ in vocabulary])\n",
    "dictionary = {word: code for code, word in enumerate(vocabulary)}\n",
    "data = np.array([dictionary.get(word, 0) for word in words])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "('anarchism originated as a term of abuse first used',\n",
       " array([5236, 3081,   12,    6,  195,    2, 3135,   46,   59]))"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\" \".join(words[:9]), data[:9]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'guided originated as a term of extinct first used'"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\" \".join([vocabulary[word_index] for word_index in [5241, 3081, 12, 6, 195, 2, 3134, 46, 59]])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "('culottes', 0)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "words[24], data[24]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Generate batches"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "import random\n",
    "from collections import deque\n",
    "\n",
    "def generate_batch(batch_size, num_skips, skip_window):\n",
    "    global data_index\n",
    "    assert batch_size % num_skips == 0\n",
    "    assert num_skips <= 2 * skip_window\n",
    "    batch = np.ndarray(shape=(batch_size), dtype=np.int32)\n",
    "    labels = np.ndarray(shape=(batch_size, 1), dtype=np.int32)\n",
    "    span = 2 * skip_window + 1 # [ skip_window target skip_window ]\n",
    "    buffer = deque(maxlen=span)\n",
    "    for _ in range(span):\n",
    "        buffer.append(data[data_index])\n",
    "        data_index = (data_index + 1) % len(data)\n",
    "    for i in range(batch_size // num_skips):\n",
    "        target = skip_window  # target label at the center of the buffer\n",
    "        targets_to_avoid = [ skip_window ]\n",
    "        for j in range(num_skips):\n",
    "            while target in targets_to_avoid:\n",
    "                target = random.randint(0, span - 1)\n",
    "            targets_to_avoid.append(target)\n",
    "            batch[i * num_skips + j] = buffer[skip_window]\n",
    "            labels[i * num_skips + j, 0] = buffer[target]\n",
    "        buffer.append(data[data_index])\n",
    "        data_index = (data_index + 1) % len(data)\n",
    "    return batch, labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "data_index=0\n",
    "batch, labels = generate_batch(8, 2, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([3081, 3081,   12,   12,    6,    6,  195,  195], dtype=int32),\n",
       " ['originated', 'originated', 'as', 'as', 'a', 'a', 'term', 'term'])"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "batch, [vocabulary[word] for word in batch]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([[  12],\n",
       "        [5236],\n",
       "        [   6],\n",
       "        [3081],\n",
       "        [ 195],\n",
       "        [  12],\n",
       "        [   2],\n",
       "        [   6]], dtype=int32),\n",
       " ['as', 'anarchism', 'a', 'originated', 'term', 'as', 'of', 'a'])"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "labels, [vocabulary[word] for word in labels[:, 0]]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Build the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "batch_size = 128\n",
    "embedding_size = 128  # Dimension of the embedding vector.\n",
    "skip_window = 1       # How many words to consider left and right.\n",
    "num_skips = 2         # How many times to reuse an input to generate a label.\n",
    "\n",
    "# We pick a random validation set to sample nearest neighbors. Here we limit the\n",
    "# validation samples to the words that have a low numeric ID, which by\n",
    "# construction are also the most frequent.\n",
    "valid_size = 16     # Random set of words to evaluate similarity on.\n",
    "valid_window = 100  # Only pick dev samples in the head of the distribution.\n",
    "valid_examples = rnd.choice(valid_window, valid_size, replace=False)\n",
    "num_sampled = 64    # Number of negative examples to sample.\n",
    "\n",
    "learning_rate = 0.01"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "tf.reset_default_graph()\n",
    "\n",
    "# Input data.\n",
    "train_inputs = tf.placeholder(tf.int32, shape=[batch_size])\n",
    "train_labels = tf.placeholder(tf.int32, shape=[batch_size, 1])\n",
    "valid_dataset = tf.constant(valid_examples, dtype=tf.int32)\n",
    "\n",
    "# Look up embeddings for inputs.\n",
    "init_embeddings = tf.random_uniform([vocabulary_size, embedding_size], -1.0, 1.0)\n",
    "embeddings = tf.Variable(init_embeddings)\n",
    "embed = tf.nn.embedding_lookup(embeddings, train_inputs)\n",
    "\n",
    "# Construct the variables for the NCE loss\n",
    "nce_weights = tf.Variable(\n",
    "    tf.truncated_normal([vocabulary_size, embedding_size],\n",
    "                        stddev=1.0 / np.sqrt(embedding_size)))\n",
    "nce_biases = tf.Variable(tf.zeros([vocabulary_size]))\n",
    "\n",
    "# Compute the average NCE loss for the batch.\n",
    "# tf.nce_loss automatically draws a new sample of the negative labels each\n",
    "# time we evaluate the loss.\n",
    "loss = tf.reduce_mean(\n",
    "    tf.nn.nce_loss(nce_weights, nce_biases, train_labels, embed,\n",
    "                   num_sampled, vocabulary_size))\n",
    "\n",
    "# Construct the Adam optimizer\n",
    "optimizer = tf.train.AdamOptimizer(learning_rate)\n",
    "training_op = optimizer.minimize(loss)\n",
    "\n",
    "# Compute the cosine similarity between minibatch examples and all embeddings.\n",
    "norm = tf.sqrt(tf.reduce_sum(tf.square(embeddings), axis=1, keep_dims=True))\n",
    "normalized_embeddings = embeddings / norm\n",
    "valid_embeddings = tf.nn.embedding_lookup(normalized_embeddings, valid_dataset)\n",
    "similarity = tf.matmul(valid_embeddings, normalized_embeddings, transpose_b=True)\n",
    "\n",
    "# Add variable initializer.\n",
    "init = tf.global_variables_initializer()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Train the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Iteration: 0\tAverage loss at step  0 :  251.248123169\n",
      "Nearest to people: anchors, tests, satisfactorily, erg, brasilia, administered, assistants, quays,\n",
      "Nearest to can: jenny, proliferate, elusive, utility, dieting, toms, affecting, scrambled,\n",
      "Nearest to than: troika, lane, shrink, medi, narrator, rin, rides, alces,\n",
      "Nearest to its: ritual, columbia, precludes, besieged, hunmin, vis, mandan, microsystems,\n",
      "Nearest to their: antioxidants, syme, conformity, trousers, stature, rapier, personhood, rinks,\n",
      "Nearest to have: hiragana, abstained, grid, conti, gender, corner, agade, consequent,\n",
      "Nearest to six: expansions, millionaires, checker, colonized, profound, hydrogenation, thermoregulation, bootleg,\n",
      "Nearest to over: hapless, draugr, mark, dre, surgically, forgery, maximise, cyril,\n",
      "Nearest to two: anagram, evenings, eee, brutus, buyer, addons, bam, guattari,\n",
      "Nearest to UNK: cornwell, selina, ericaceae, browed, blake, biostatistics, cages, penetrates,\n",
      "Nearest to was: strains, currents, caligula, fet, digestion, gohan, fibrous, thumb,\n",
      "Nearest to his: disestablishment, jute, nicopolis, acquaintance, zn, nisan, executive, deconstructing,\n",
      "Nearest to b: tierra, corroborate, eases, emigrate, thames, exploring, galileo, royalties,\n",
      "Nearest to this: assign, thracian, probing, illness, tuple, ht, words, dysfunction,\n",
      "Nearest to use: scrupulously, reprinting, grenade, dw, masai, chests, verifies, elapsed,\n",
      "Nearest to one: wgn, palma, nlcs, petrograd, haughey, graal, tel, purpurea,\n",
      "Iteration: 2000\tAverage loss at step  2000 :  132.988740146\n",
      "Iteration: 4000\tAverage loss at step  4000 :  63.6886414816\n",
      "Iteration: 6000\tAverage loss at step  6000 :  42.0200037532\n",
      "Iteration: 8000\tAverage loss at step  8000 :  31.0967397873\n",
      "Iteration: 10000\tAverage loss at step  10000 :  25.9374142761\n",
      "Nearest to people: dehydration, chromatic, up, unarable, reelection, associating, aardvarks, schools,\n",
      "Nearest to can: to, condom, dish, is, correspondences, devastated, satisfies, emphasized,\n",
      "Nearest to than: worships, montagu, zuid, intercollegiate, incredibly, expedition, infants, denounced,\n",
      "Nearest to its: the, alp, fallacy, blore, pulmonary, breastfeeding, sucker, chung,\n",
      "Nearest to their: the, argon, looming, separate, gide, milner, incredibly, griffith,\n",
      "Nearest to have: be, had, fragile, ankh, altaic, ssp, refuted, topalov,\n",
      "Nearest to six: two, seven, eight, nine, one, three, four, zero,\n",
      "Nearest to over: conformation, spin, ankh, snowball, abbots, magyar, wheel, subatomic,\n",
      "Nearest to two: three, six, one, zero, four, seven, nine, five,\n",
      "Nearest to UNK: and, astatine, the, two, three, hephaestion, antigua, one,\n",
      "Nearest to was: is, and, floppy, by, in, with, carrel, ales,\n",
      "Nearest to his: the, s, pashtun, quotient, affidavit, deuterostomes, counting, cooler,\n",
      "Nearest to b: seo, and, pleas, nine, one, culinary, conformation, murchison,\n",
      "Nearest to this: the, abscess, a, aquarius, faq, nationalist, antenna, altaic,\n",
      "Nearest to use: gates, reign, concerts, omotic, abbreviations, automorphism, desalination, achill,\n",
      "Nearest to one: seven, nine, three, five, six, two, eight, four,\n",
      "Iteration: 12000\tAverage loss at step  12000 :  22.2112427891\n",
      "Iteration: 14000\tAverage loss at step  14000 :  20.5654449351\n",
      "Iteration: 16000\tAverage loss at step  16000 :  18.5331052436\n",
      "Iteration: 18000\tAverage loss at step  18000 :  17.0019776267\n",
      "Iteration: 20000\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\tAverage loss at step  20000 :  17.2390571074\n",
      "Nearest to people: topalov, arianism, aardvarks, outrage, feeble, morpork, witch, brownian,\n",
      "Nearest to can: twinned, achill, rigor, starship, spooner, elves, got, tel,\n",
      "Nearest to than: vortigern, more, gide, steals, extract, zwickau, argon, aidan,\n",
      "Nearest to its: renormalization, haliotis, a, bars, phonology, spawning, zimbabwe, impiety,\n",
      "Nearest to their: the, and, audiogalaxy, deuterostomes, rijndael, vilayet, taoiseach, bins,\n",
      "Nearest to have: has, are, halfdan, cin, tarkovsky, elves, by, sprinkled,\n",
      "Nearest to six: one, seven, nine, eight, five, three, zero, two,\n",
      "Nearest to over: responsibilities, ment, tetrapods, homes, archaeological, acronyms, ankh, tattooing,\n",
      "Nearest to two: five, one, three, zero, four, seven, nine, six,\n",
      "Nearest to UNK: and, one, nine, amitabh, the, altaic, seven, five,\n",
      "Nearest to was: and, to, magdeburg, in, is, currie, by, amnon,\n",
      "Nearest to his: the, afc, goto, apologists, some, anacreon, seville, halfdan,\n",
      "Nearest to b: seven, eight, nine, and, aryans, hadrian, morphisms, american,\n",
      "Nearest to this: the, anacreon, patricians, a, theoretic, it, shari, albertosaurus,\n",
      "Nearest to use: condoms, acronyms, abandoning, annihilated, order, germania, subsistence, tarleton,\n",
      "Nearest to one: five, nine, eight, six, two, seven, three, zero,\n",
      "Iteration: 22000\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\tAverage loss at step  22000 :  16.2764769258\n",
      "Iteration: 24000\tAverage loss at step  24000 :  15.6715580997\n",
      "Iteration: 26000\tAverage loss at step  26000 :  15.2251547008\n",
      "Iteration: 28000\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\tAverage loss at step  28000 :  14.8846785804\n",
      "Iteration: 30000\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\tAverage loss at step  30000 :  14.8717326654\n",
      "Nearest to people: topalov, hochschule, grasses, thrusters, suebi, ftl, replicants, crocodiles,\n",
      "Nearest to can: to, regnal, suu, reinstate, landau, impresario, is, mobility,\n",
      "Nearest to than: bakersfield, willard, amaranth, nodules, notification, xp, jacobi, shamans,\n",
      "Nearest to its: the, abiword, apeiron, adret, spawning, rosso, cuyp, octavius,\n",
      "Nearest to their: and, the, some, in, rationalistic, ankh, herbrand, deuterostomes,\n",
      "Nearest to have: been, anchovies, and, are, has, be, pornography, also,\n",
      "Nearest to six: nine, one, eight, seven, three, four, five, zero,\n",
      "Nearest to over: soto, setter, potatoes, dime, first, indelible, ghanaian, agade,\n",
      "Nearest to two: zero, one, four, three, five, eight, nine, six,\n",
      "Nearest to UNK: and, the, zero, nine, two, occultism, cpc, to,\n",
      "Nearest to was: is, dnia, were, seven, celt, chadic, are, nine,\n",
      "Nearest to his: hazel, the, droughts, acquis, unalaska, mishnayot, iarc, soroban,\n",
      "Nearest to b: nine, seo, manar, lambar, polypeptide, yoshinkan, eight, one,\n",
      "Nearest to this: quantifying, a, some, the, and, it, seo, that,\n",
      "Nearest to use: annoyance, tashfin, fuzes, staggering, benzene, pygmalion, ornamentation, glucocorticoids,\n",
      "Nearest to one: nine, seven, four, two, six, eight, three, zero,\n",
      "Iteration: 32000\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\tAverage loss at step  32000 :  14.1600210441\n",
      "Iteration: 34000\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\tAverage loss at step  34000 :  13.8883676932\n",
      "Iteration: 36000\tAverage loss at step  36000 :  13.4337159057\n",
      "Iteration: 38000\t\t\t\t\t\t\t\t\tAverage loss at step  38000 :  12.1775156796\n",
      "Iteration: 40000\tAverage loss at step  40000 :  12.6818081163\n",
      "Nearest to people: aon, strictness, unarable, savoy, kongo, rt, workbench, ftl,\n",
      "Nearest to can: shiv, noumenon, arising, peaking, ik, fremont, is, minorca,\n",
      "Nearest to than: annulled, nimzowitsch, callisto, alda, sarai, maarten, caribs, depriving,\n",
      "Nearest to its: taro, the, astrometric, topalov, ballantine, uat, mut, uml,\n",
      "Nearest to their: the, cherokees, a, billions, pity, harpoon, peaking, caapi,\n",
      "Nearest to have: but, halfdan, are, manar, gunships, silurian, invokes, has,\n",
      "Nearest to six: five, one, four, nine, eight, seven, two, three,\n",
      "Nearest to over: woz, antiderivative, that, calamine, tetrapods, lifeson, publ, carnot,\n",
      "Nearest to two: zero, one, five, four, eight, three, nine, six,\n",
      "Nearest to UNK: and, one, the, unarable, is, harpoon, lair, glucocorticoids,\n",
      "Nearest to was: were, in, is, by, and, has, equalled, bitnet,\n",
      "Nearest to his: the, and, he, difranco, calder, lexington, jitter, seo,\n",
      "Nearest to b: difranco, w, sanctity, agp, scarp, tua, manar, scoreboard,\n",
      "Nearest to this: deuterostomes, the, postumus, astrometric, phillippe, aaas, bulbs, a,\n",
      "Nearest to use: gide, rutskoy, renormalization, outgrowth, redefinition, way, livery, imparts,\n",
      "Nearest to one: nine, eight, six, five, two, four, seven, zero,\n",
      "Iteration: 42000\t\t\t\t\tAverage loss at step  42000 :  12.5647115765\n",
      "Iteration: 44000\t\t\t\t\tAverage loss at step  44000 :  12.5112489903\n",
      "Iteration: 46000\t\t\t\tAverage loss at step  46000 :  12.1351359482\n",
      "Iteration: 48000\tAverage loss at step  48000 :  11.5563955258\n",
      "Iteration: 50000\tAverage loss at step  50000 :  11.9526369007\n",
      "Nearest to people: roasting, ftl, topalov, barium, spooner, laetrile, inverses, swaps,\n",
      "Nearest to can: abiword, to, but, antiprotons, unn, are, caapi, not,\n",
      "Nearest to than: are, philology, is, fulgencio, alexei, alboin, arbitrage, to,\n",
      "Nearest to its: purchasers, ankh, childbirth, shorten, whist, thc, the, accede,\n",
      "Nearest to their: the, and, difranco, tetrapods, to, beagles, asbestos, angilbert,\n",
      "Nearest to have: has, anac, beagles, regulator, are, been, also, aidan,\n",
      "Nearest to six: five, one, two, nine, three, four, zero, eight,\n",
      "Nearest to over: blacki, gdp, rationalize, neko, faustina, tetrapods, angilbert, with,\n",
      "Nearest to two: zero, one, nine, three, four, six, five, the,\n",
      "Nearest to UNK: and, the, of, warr, one, in, a, aleksander,\n",
      "Nearest to was: by, were, bhfiann, is, has, frazee, gide, fachhochschule,\n",
      "Nearest to his: the, and, him, snowball, s, akimbo, kenji, naismith,\n",
      "Nearest to b: c, series, two, sutra, fistful, haber, donoghue, apprehension,\n",
      "Nearest to this: the, it, a, which, trs, anoa, autocorrelation, antiprotons,\n",
      "Nearest to use: appease, widdecombe, gide, sharpen, scourge, island, alyattes, barium,\n",
      "Nearest to one: five, three, two, nine, four, seven, six, eight,\n",
      "Iteration: 52000\tAverage loss at step  52000 :  11.8388420435\n",
      "Iteration: 54000\tAverage loss at step  54000 :  11.935847087\n",
      "Iteration: 56000\tAverage loss at step  56000 :  11.5109407967\n",
      "Iteration: 58000\t\tAverage loss at step  58000 :  11.6105584927\n",
      "Iteration: 60000\tAverage loss at step  60000 :  11.1421181628\n",
      "Nearest to people: przewodnicz, ambients, hasmonaean, gamal, beagles, convicts, swnts, crematoria,\n",
      "Nearest to can: sfg, but, aclu, of, darkened, albinism, may, anacreon,\n",
      "Nearest to than: much, mnd, is, less, tanaka, cae, arbitrage, peloponnese,\n",
      "Nearest to its: the, thrusters, abiword, foreknowledge, parrot, and, UNK, permeability,\n",
      "Nearest to their: and, affidavit, the, crematoria, medea, UNK, kat, cca,\n",
      "Nearest to have: be, are, in, not, been, and, on, chanter,\n",
      "Nearest to six: one, four, nine, zero, three, five, eight, two,\n",
      "Nearest to over: linearity, on, alboin, pasternak, expositions, frazee, gysin, manipulative,\n",
      "Nearest to two: zero, one, four, five, three, nine, six, eight,\n",
      "Nearest to UNK: and, of, in, the, danios, airshows, dme, renouf,\n",
      "Nearest to was: UNK, been, beagles, is, when, jehoash, antiprotons, adware,\n",
      "Nearest to his: he, randle, aruban, lifeson, bryozoans, manar, renormalization, aidan,\n",
      "Nearest to b: eight, knack, unarable, phillippe, geffen, vikernes, vcr, mexicanus,\n",
      "Nearest to this: a, the, assemblers, is, bhangra, to, which, tezuka,\n",
      "Nearest to use: addition, jubilees, capp, renovate, dialectics, ehret, memex, tashfin,\n",
      "Nearest to one: nine, four, six, three, two, zero, eight, five,\n",
      "Iteration: 62000\tAverage loss at step  62000 :  10.5445741624\n",
      "Iteration: 64000\tAverage loss at step  64000 :  10.0993400878\n",
      "Iteration: 66000\tAverage loss at step  66000 :  10.922524215\n",
      "Iteration: 68000\tAverage loss at step  68000 :  10.6693151739\n",
      "Iteration: 70000\tAverage loss at step  70000 :  10.8248517294\n",
      "Nearest to people: donny, those, ifn, heenan, lomborg, woz, form, peake,\n",
      "Nearest to can: and, to, are, in, bavarian, a, is, the,\n",
      "Nearest to than: astrometric, therus, fraternities, for, aaas, squeak, erythrocytes, nabonidus,\n",
      "Nearest to its: the, medulla, tarkovsky, oddie, kyrenia, shuja, lilies, agade,\n",
      "Nearest to their: a, the, akimbo, orser, overtraining, xry, polskiej, and,\n",
      "Nearest to have: cancelling, are, halfdan, perturbed, has, islamists, californium, that,\n",
      "Nearest to six: four, eight, five, three, one, nine, two, zero,\n",
      "Nearest to over: ankh, bundaberg, cerium, manar, time, persians, hippolyte, lifeson,\n",
      "Nearest to two: zero, one, three, five, six, nine, four, eight,\n",
      "Nearest to UNK: and, the, in, of, ankh, is, or, a,\n",
      "Nearest to was: is, berenson, cladogram, antiprotons, amway, has, monologues, his,\n",
      "Nearest to his: the, was, UNK, mileva, in, kournikova, a, curule,\n",
      "Nearest to b: chevalier, chondrites, cruz, palatalized, ncia, d, massimo, william,\n",
      "Nearest to this: the, a, and, in, of, abhidharma, that, astrometric,\n",
      "Nearest to use: aalen, hanyu, cebit, arif, furtherance, orientalist, turn, tentacles,\n",
      "Nearest to one: two, five, three, nine, six, zero, eight, seven,\n",
      "Iteration: 72000\tAverage loss at step  72000 :  10.4764668052\n",
      "Iteration: 74000\tAverage loss at step  74000 :  10.519153582\n",
      "Iteration: 76000\tAverage loss at step  76000 :  10.546341171\n",
      "Iteration: 78000\tAverage loss at step  78000 :  10.3008789356\n",
      "Iteration: 80000\tAverage loss at step  80000 :  10.5960232033\n",
      "Nearest to people: beagles, literals, bragi, aino, asymptotes, frisians, inches, yourselves,\n",
      "Nearest to can: habr, cyg, trs, would, to, not, tupper, childe,\n",
      "Nearest to than: jannaeus, crespigny, for, mmo, canova, vanzetti, much, more,\n",
      "Nearest to its: the, a, an, thabit, and, brainfuck, all, mut,\n",
      "Nearest to their: the, swnts, antiprotons, flutie, kilgore, dysrhythmias, vortigern, albertosaurus,\n",
      "Nearest to have: has, unarable, apoptotic, pops, lycus, are, harmonizing, redfern,\n",
      "Nearest to six: nine, five, one, zero, eight, seven, three, four,\n",
      "Nearest to over: langevin, and, one, of, in, alhazred, vortigern, applesoft,\n",
      "Nearest to two: zero, four, one, three, five, nine, seven, eight,\n",
      "Nearest to UNK: and, beadwork, chandragupta, this, boney, the, ustawa, benhavn,\n",
      "Nearest to was: is, lambs, adware, in, by, and, diomedea, appellant,\n",
      "Nearest to his: the, fasciculata, he, melilla, chojn, s, couplet, rhea,\n",
      "Nearest to b: federko, danios, boitano, vertebra, is, prophylaxis, encyclopedists, balalaika,\n",
      "Nearest to this: the, and, a, boney, UNK, that, which, anac,\n",
      "Nearest to use: bofh, akimbo, adt, asteraceae, ttt, sittings, or, fraternities,\n",
      "Nearest to one: nine, seven, eight, three, two, five, six, zero,\n",
      "Iteration: 82000\tAverage loss at step  82000 :  10.4969443651\n",
      "Iteration: 84000\tAverage loss at step  84000 :  10.4385484533\n",
      "Iteration: 86000\tAverage loss at step  86000 :  10.3878999927\n",
      "Iteration: 88000\tAverage loss at step  88000 :  9.89178018618\n",
      "Iteration: 90000\tAverage loss at step  90000 :  10.0904062961\n",
      "Nearest to people: lawns, botham, dme, ciboney, ehret, moser, californium, alnus,\n",
      "Nearest to can: to, is, glucocorticoids, linea, should, this, heenan, maguey,\n",
      "Nearest to than: is, stakes, bundaberg, levitt, alhazred, xxxii, cheddar, embouchure,\n",
      "Nearest to its: clu, shuttlecock, contravention, the, their, ystem, alcott, diomedea,\n",
      "Nearest to their: maguey, of, catus, the, yoannis, eucalyptus, cebit, benchers,\n",
      "Nearest to have: are, had, jehoash, has, been, lw, rzeczypospolitej, also,\n",
      "Nearest to six: one, four, nine, two, three, eight, seven, five,\n",
      "Nearest to over: for, beagles, manar, boomerangs, refectory, anacreon, amway, with,\n",
      "Nearest to two: zero, three, one, eight, six, nine, four, seven,\n",
      "Nearest to UNK: and, in, danios, the, is, one, wt, cheerleading,\n",
      "Nearest to was: chianti, is, in, antiprotons, for, when, beagles, also,\n",
      "Nearest to his: susima, chiuchow, dysrhythmias, californium, agade, the, triage, frazee,\n",
      "Nearest to b: sip, erm, parallel, magma, schedule, preferred, erie, chogm,\n",
      "Nearest to this: the, it, mut, amphisbaena, maguey, of, astatine, cira,\n",
      "Nearest to use: disjunction, non, chievo, chicanos, rearrangement, is, amount, anoa,\n",
      "Nearest to one: eight, seven, nine, three, four, six, two, five,\n",
      "Iteration: 92000\tAverage loss at step  92000 :  10.2444101722\n",
      "Iteration: 94000\tAverage loss at step  94000 :  9.86116436267\n",
      "Iteration: 96000\tAverage loss at step  96000 :  9.87077867138\n",
      "Iteration: 98000\tAverage loss at step  98000 :  9.76025918794\n",
      "Iteration: 100000\tAverage loss at step  100000 :  9.85069271541\n",
      "Nearest to people: audiogalaxy, criswell, vla, screech, of, in, harrier, fraktion,\n",
      "Nearest to can: to, chickenpox, cantigas, handled, halpin, benchers, shuttlecock, clu,\n",
      "Nearest to than: more, much, sayyaf, to, corruptions, mmo, be, appease,\n",
      "Nearest to its: a, the, arbitrators, cebit, agade, governorates, danios, boomerangs,\n",
      "Nearest to their: the, and, provers, s, octavius, of, aargau, several,\n",
      "Nearest to have: jehoash, nepeta, has, and, are, aon, glucocorticoids, abaye,\n",
      "Nearest to six: one, five, four, nine, eight, two, seven, three,\n",
      "Nearest to over: aardvarks, mahdi, dactylic, limestones, rations, subshell, brainfuck, incitement,\n",
      "Nearest to two: one, five, zero, four, eight, seven, six, three,\n",
      "Nearest to UNK: and, the, in, by, of, albinism, uat, s,\n",
      "Nearest to was: and, is, gand, UNK, in, oliveira, that, but,\n",
      "Nearest to his: and, s, the, bocks, colloquialism, swnt, jarry, phallus,\n",
      "Nearest to b: seven, nine, one, pisan, morales, in, continuo, eight,\n",
      "Nearest to this: the, and, a, that, audiogalaxy, it, affidavit, afghana,\n",
      "Nearest to use: home, form, cheese, choose, bandy, evidence, bicycles, words,\n",
      "Nearest to one: two, six, eight, five, three, four, nine, seven,\n"
     ]
    }
   ],
   "source": [
    "num_steps = 100001\n",
    "\n",
    "with tf.Session() as session:\n",
    "    init.run()\n",
    "\n",
    "    average_loss = 0\n",
    "    for step in range(num_steps):\n",
    "        print(\"\\rIteration: {}\".format(step), end=\"\\t\")\n",
    "        batch_inputs, batch_labels = generate_batch(batch_size, num_skips, skip_window)\n",
    "        feed_dict = {train_inputs : batch_inputs, train_labels : batch_labels}\n",
    "\n",
    "        # We perform one update step by evaluating the training op (including it\n",
    "        # in the list of returned values for session.run()\n",
    "        _, loss_val = session.run([training_op, loss], feed_dict=feed_dict)\n",
    "        average_loss += loss_val\n",
    "\n",
    "        if step % 2000 == 0:\n",
    "            if step > 0:\n",
    "                average_loss /= 2000\n",
    "            # The average loss is an estimate of the loss over the last 2000 batches.\n",
    "            print(\"Average loss at step \", step, \": \", average_loss)\n",
    "            average_loss = 0\n",
    "\n",
    "        # Note that this is expensive (~20% slowdown if computed every 500 steps)\n",
    "        if step % 10000 == 0:\n",
    "            sim = similarity.eval()\n",
    "            for i in range(valid_size):\n",
    "                valid_word = vocabulary[valid_examples[i]]\n",
    "                top_k = 8 # number of nearest neighbors\n",
    "                nearest = (-sim[i, :]).argsort()[1:top_k+1]\n",
    "                log_str = \"Nearest to %s:\" % valid_word\n",
    "                for k in range(top_k):\n",
    "                    close_word = vocabulary[nearest[k]]\n",
    "                    log_str = \"%s %s,\" % (log_str, close_word)\n",
    "                print(log_str)\n",
    "\n",
    "    final_embeddings = normalized_embeddings.eval()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "Let's save the final embeddings (of course you can use a TensorFlow `Saver` if you prefer):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "np.save(\"./my_final_embeddings.npy\", final_embeddings)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Plot the embeddings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "def plot_with_labels(low_dim_embs, labels):\n",
    "    assert low_dim_embs.shape[0] >= len(labels), \"More labels than embeddings\"\n",
    "    plt.figure(figsize=(18, 18))  #in inches\n",
    "    for i, label in enumerate(labels):\n",
    "        x, y = low_dim_embs[i,:]\n",
    "        plt.scatter(x, y)\n",
    "        plt.annotate(label,\n",
    "                     xy=(x, y),\n",
    "                     xytext=(5, 2),\n",
    "                     textcoords='offset points',\n",
    "                     ha='right',\n",
    "                     va='bottom')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABB8AAAQSCAYAAADnplJ2AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n+AAAAAAAAGz1/75YHw/rwIP2AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f9c07273470>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.manifold import TSNE\n",
    "\n",
    "tsne = TSNE(perplexity=30, n_components=2, init='pca', n_iter=5000)\n",
    "plot_only = 500\n",
    "low_dim_embs = tsne.fit_transform(final_embeddings[:plot_only,:])\n",
    "labels = [vocabulary[i] for i in range(plot_only)]\n",
    "plot_with_labels(low_dim_embs, labels)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "# Machine Translation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "The `basic_rnn_seq2seq()` function creates a simple Encoder/Decoder model: it first runs an RNN to encode `encoder_inputs` into a state vector, then runs a decoder initialized with the last encoder state on `decoder_inputs`. Encoder and decoder use the same RNN cell type but they don't share parameters."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "tf.reset_default_graph()\n",
    "\n",
    "n_steps = 50\n",
    "n_neurons = 200\n",
    "n_layers = 3\n",
    "num_encoder_symbols = 20000\n",
    "num_decoder_symbols = 20000\n",
    "embedding_size = 150\n",
    "learning_rate = 0.01\n",
    "\n",
    "X = tf.placeholder(tf.int32, [None, n_steps]) # English sentences\n",
    "Y = tf.placeholder(tf.int32, [None, n_steps]) # French translations\n",
    "W = tf.placeholder(tf.float32, [None, n_steps - 1, 1])\n",
    "Y_input = Y[:, :-1]\n",
    "Y_target = Y[:, 1:]\n",
    "\n",
    "encoder_inputs = tf.unstack(tf.transpose(X)) # list of 1D tensors\n",
    "decoder_inputs = tf.unstack(tf.transpose(Y_input)) # list of 1D tensors\n",
    "\n",
    "lstm_cell = tf.contrib.rnn.BasicLSTMCell(num_units=n_neurons)\n",
    "cell = tf.contrib.rnn.MultiRNNCell([lstm_cell] * n_layers)\n",
    "\n",
    "output_seqs, states = tf.contrib.legacy_seq2seq.embedding_rnn_seq2seq(\n",
    "    encoder_inputs,\n",
    "    decoder_inputs,\n",
    "    cell,\n",
    "    num_encoder_symbols,\n",
    "    num_decoder_symbols,\n",
    "    embedding_size)\n",
    "\n",
    "logits = tf.transpose(tf.unstack(output_seqs), perm=[1, 0, 2])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "logits_flat = tf.reshape(logits, [-1, num_decoder_symbols])\n",
    "Y_target_flat = tf.reshape(Y_target, [-1])\n",
    "W_flat = tf.reshape(W, [-1])\n",
    "xentropy = W_flat * tf.nn.sparse_softmax_cross_entropy_with_logits(labels=Y_target_flat, logits=logits_flat)\n",
    "loss = tf.reduce_mean(xentropy)\n",
    "optimizer = tf.train.AdamOptimizer(learning_rate=learning_rate)\n",
    "training_op = optimizer.minimize(loss)\n",
    "\n",
    "init = tf.global_variables_initializer()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "source": [
    "# Exercise solutions"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "**Coming soon**"
   ]
  }
 ],
 "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": "3.5.2+"
  },
  "nav_menu": {},
  "toc": {
   "navigate_menu": true,
   "number_sections": true,
   "sideBar": true,
   "threshold": 6,
   "toc_cell": false,
   "toc_section_display": "block",
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}