graphs_tensorflow.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# TensorFlow Graphs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/eblancoh/anaconda3/envs/universe/lib/python3.5/site-packages/h5py/__init__.py:34: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`.\n",
      "  from ._conv import register_converters as _register_converters\n"
     ]
    }
   ],
   "source": [
    "# Importamos la librería TensorFlow\n",
    "import tensorflow as tf"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Generación de un grafo sencillo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Creamos un par de nodos definidos como dos constantes:\n",
    "num_1 = tf.constant(1)\n",
    "num_2 = tf.constant(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Se define un tercer nodo de operación donde se suman las dos constantes anteriores.\n",
    "num_3 = num_1 + num_2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3\n"
     ]
    }
   ],
   "source": [
    "# Como ya sabemos, abrimos sesión para poder evaluar el resultado de nuestra operación:\n",
    "with tf.Session() as sess:\n",
    "    resultado = sess.run(num_3)\n",
    "print(resultado)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Tensor(\"add:0\", shape=(), dtype=int32)\n"
     ]
    }
   ],
   "source": [
    "print (num_3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<tensorflow.python.framework.ops.Graph object at 0x7f07b61e7cc0>\n"
     ]
    }
   ],
   "source": [
    "# Si ejecutamos el siguiente comando, se nos devolverá el objeto graph por defecto que tenemos ahora \n",
    "# mismo almacenado en memoria\n",
    "print(tf.get_default_graph())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[<tf.Operation 'Const' type=Const>, <tf.Operation 'Const_1' type=Const>, <tf.Operation 'add' type=Add>]\n"
     ]
    }
   ],
   "source": [
    "# TensorFlow también nos permite visualizar todas las operaciones soportadas en el grafo por defecto.\n",
    "g = tf.get_default_graph()\n",
    "print(g.get_operations())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "El grafo está definido por tres nodos de operaciones: dos de constantes (num_1 y num_2) y uno de suma (num_3)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Podemos definir otro grafo mediante la siguiente expresión.\n",
    "graph = tf.Graph()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<tensorflow.python.framework.ops.Graph object at 0x7f07b61e0c50>\n"
     ]
    }
   ],
   "source": [
    "# Que es claramente distinto al otro grafo que tenemos definido por defecto en tf.get_default_graph()\n",
    "print(graph)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Generemos un grafo y convirtámoslo en el que se debe considerar por defecto:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "graph_one = tf.get_default_graph()\n",
    "graph_two = tf.Graph()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Podemos ver que graph_one es el grafo por defecto\n",
    "graph_one is tf.get_default_graph()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "graph_two is tf.get_default_graph()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "True\n"
     ]
    }
   ],
   "source": [
    "with graph_two.as_default():\n",
    "    print(graph_two is tf.get_default_graph())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "graph_two is tf.get_default_graph()"
   ]
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python [default]",
   "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"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}