ANN examples

Author: ugik

Simple_Neural_Network.ipynb

@@ -2,7 +2,7 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": 8,
    "metadata": {
     "collapsed": false
    },
@@ -14,23 +14,35 @@
       "Random starting synaptic weights: \n",
       "[[-0.16595599]\n",
       " [ 0.44064899]\n",
-      " [-0.99977125]]\n",
-      "error after 0 iterations: 0.578374046722\n",
-      "error after 1000 iterations: 0.0353771814512\n",
-      "error after 2000 iterations: 0.024323319584\n",
-      "error after 3000 iterations: 0.0196075022358\n",
-      "error after 4000 iterations: 0.016850233908\n",
-      "error after 5000 iterations: 0.014991814044\n",
-      "error after 6000 iterations: 0.0136320935305\n",
-      "error after 7000 iterations: 0.01258242301\n",
-      "error after 8000 iterations: 0.0117408289409\n",
-      "error after 9000 iterations: 0.0110467781322\n",
+      " [-0.99977125]\n",
+      " [-0.39533485]]\n",
+      "error after 0 iterations: 0.449903018833\n",
+      "error after 1000 iterations: 0.013750440233\n",
+      "error after 2000 iterations: 0.00965974377145\n",
+      "error after 3000 iterations: 0.00786218913086\n",
+      "error after 4000 iterations: 0.00679524113652\n",
+      "error after 5000 iterations: 0.00606919069443\n",
+      "error after 6000 iterations: 0.00553435906148\n",
+      "error after 7000 iterations: 0.00511936227674\n",
+      "error after 8000 iterations: 0.00478528276951\n",
+      "error after 9000 iterations: 0.00450886340585\n",
+      "error after 10000 iterations: 0.00427523967482\n",
+      "error after 11000 iterations: 0.00407440954071\n",
+      "error after 12000 iterations: 0.00389936317766\n",
+      "error after 13000 iterations: 0.00374502274318\n",
+      "error after 14000 iterations: 0.00360760779755\n",
+      "error after 15000 iterations: 0.00348423814505\n",
+      "error after 16000 iterations: 0.00337267580097\n",
+      "error after 17000 iterations: 0.00327115193553\n",
+      "error after 18000 iterations: 0.00317824759008\n",
+      "error after 19000 iterations: 0.00309280947588\n",
       "New synaptic weights after training: \n",
-      "[[ 12.79547496]\n",
-      " [ -4.2162058 ]\n",
-      " [ -4.21608782]]\n",
-      "Considering new situation [1, 0, 0] -> ?: \n",
-      "[ 0.99999723]\n"
+      "[[ 3.16744718]\n",
+      " [-0.24954816]\n",
+      " [-5.53795575]\n",
+      " [ 2.93806832]]\n",
+      "Considering new situation [1, 0, 1, 1] -> ?: \n",
+      "[ 0.63819991]\n"
      ]
     }
    ],
@@ -47,7 +59,7 @@
     "        # We model a single neuron, with 3 input connections and 1 output connection.\n",
     "        # We assign random weights to a 3 x 1 matrix, with values in the range -1 to 1\n",
     "        # and mean 0.\n",
-    "        self.synaptic_weights = 2 * np.random.random((3, 1)) - 1\n",
+    "        self.synaptic_weights = 2 * np.random.random((4, 1)) - 1\n",
     "\n",
     "    # The Sigmoid function, which describes an S shaped curve.\n",
     "    # We pass the weighted sum of the inputs through this function to\n",
@@ -98,25 +110,25 @@
     "\n",
     "    # The training set. We have 4 examples, each consisting of 3 input values\n",
     "    # and 1 output value.\n",
-    "    training_set_inputs = np.array([[0, 0, 1], [1, 1, 1], [1, 0, 1], [0, 1, 0]])\n",
+    "    training_set_inputs = np.array([[0, 0, 1, 0], [1, 1, 0, 1], [1, 0, 0, 1], [0, 1, 1, 0]])\n",
     "    training_set_outputs = np.array([[0, 1, 1, 0]]).T\n",
     "\n",
     "    # Train the neural network using a training set.\n",
     "    # Do it 10,000 times and make small adjustments each time.\n",
-    "    neural_network.train(training_set_inputs, training_set_outputs, 10000)\n",
+    "    neural_network.train(training_set_inputs, training_set_outputs, 20000)\n",
     "\n",
     "    print (\"New synaptic weights after training: \")\n",
     "    print (neural_network.synaptic_weights)\n",
     "\n",
     "    # Test the neural network with a new pattern\n",
-    "    test = [1, 0, 0]\n",
+    "    test = [1, 0, 1, 1]\n",
     "    print (\"Considering new situation %s -> ?: \" % test )\n",
     "    print (neural_network.think(np.array(test)))\n"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 14,
+   "execution_count": 6,
    "metadata": {
     "collapsed": false
    },
@@ -185,21 +197,21 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python 3",
+   "display_name": "Python 2",
    "language": "python",
-   "name": "python3"
+   "name": "python2"
   },
   "language_info": {
    "codemirror_mode": {
     "name": "ipython",
-    "version": 3
+    "version": 2
    },
    "file_extension": ".py",
    "mimetype": "text/x-python",
    "name": "python",
    "nbconvert_exporter": "python",
-   "pygments_lexer": "ipython3",
-   "version": "3.5.2"
+   "pygments_lexer": "ipython2",
+   "version": "2.7.12"
   }
  },
  "nbformat": 4,

Tensorflow chat-bot model.ipynb

@@ -2,19 +2,11 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": 2,
    "metadata": {
     "collapsed": false
    },
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "hdf5 is not supported on this machine (please install/reinstall h5py for optimal experience)\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "# things we need for NLP\n",
     "import nltk\n",
@@ -30,7 +22,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": 3,
    "metadata": {
     "collapsed": false
    },
@@ -44,7 +36,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 4,
    "metadata": {
     "collapsed": false
    },
@@ -91,7 +83,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 5,
    "metadata": {
     "collapsed": false
    },
@@ -132,7 +124,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 6,
    "metadata": {
     "collapsed": false,
     "scrolled": true
@@ -142,10 +134,10 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Training Step: 3999  | total loss: \u001b[1m\u001b[32m0.79875\u001b[0m\u001b[0m | time: 0.009s\n",
-      "| Adam | epoch: 1000 | loss: 0.79875 - acc: 0.9100 -- iter: 24/27\n",
-      "Training Step: 4000  | total loss: \u001b[1m\u001b[32m0.72213\u001b[0m\u001b[0m | time: 0.012s\n",
-      "| Adam | epoch: 1000 | loss: 0.72213 - acc: 0.9190 -- iter: 27/27\n",
+      "Training Step: 3999  | total loss: \u001b[1m\u001b[32m1.00123\u001b[0m\u001b[0m | time: 0.009s\n",
+      "| Adam | epoch: 1000 | loss: 1.00123 - acc: 0.9098 -- iter: 24/27\n",
+      "Training Step: 4000  | total loss: \u001b[1m\u001b[32m0.91795\u001b[0m\u001b[0m | time: 0.013s\n",
+      "| Adam | epoch: 1000 | loss: 0.91795 - acc: 0.9188 -- iter: 27/27\n",
       "--\n",
       "INFO:tensorflow:/home/gk/gensim/notebooks/model.tflearn is not in all_model_checkpoint_paths. Manually adding it.\n"
      ]
@@ -170,7 +162,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": 7,
    "metadata": {
     "collapsed": true
    },
@@ -201,7 +193,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": 8,
    "metadata": {
     "collapsed": false
    },
@@ -224,7 +216,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": 9,
    "metadata": {
     "collapsed": false
    },
@@ -233,7 +225,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "[[0.0006383281433954835, 1.1489937890019064e-08, 0.005845240317285061, 2.1674446543329395e-05, 0.9596946239471436, 1.4315231666728323e-08, 0.0018385507864877582, 1.7271321439693565e-06, 0.031959764659404755]]\n"
+      "[[0.01907932758331299, 1.0800805583244255e-08, 0.0016339969588443637, 1.445684461032215e-09, 0.6044806838035583, 1.0841611128853401e-06, 2.067307436348642e-09, 6.506108093162766e-06, 0.37479835748672485]]\n"
      ]
     }
    ],
@@ -243,7 +235,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": 10,
    "metadata": {
     "collapsed": false
    },

Tensorflow chat-bot response.ipynb

@@ -264,6 +264,25 @@
     "response('what kind of mopeds do you rent?')"
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": 26,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Bye! Come back again soon.\n"
+     ]
+    }
+   ],
+   "source": [
+    "response('Goodbye, see you later')"
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": 12,