Assignment_1_Babak_Ansari_Deep_Learning_using_Keras.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Assignment 1 - Babak Ansari - Deep Learning using Keras ",
      "provenance": [],
      "collapsed_sections": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/babakansari/DeepLearning/blob/master/Assignment_1_Babak_Ansari_Deep_Learning_using_Keras.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "1ILo-suNOWXP",
        "colab_type": "text"
      },
      "source": [
        "# Assignment 1 - Babak Ansari-Jaberi"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FqjAAJ0-7LEt",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xY-ZjajW4I9U",
        "colab_type": "text"
      },
      "source": [
        "  The general conclusion from the experiments in this assignment is that the Model is prone to overfitting for too many Epochs with more neurons and layers.\n",
        "\n",
        "\n",
        "  Generally lower Epochs with more layers and Neurons performs better.\n",
        "  \n",
        "  \n",
        "  Both tanh and MSE seems perform better with lower Epochs"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8w3oM-fvRpFX",
        "colab_type": "text"
      },
      "source": [
        " ## Data Prepration and review"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "aMq-R-cDNbLk",
        "colab_type": "code",
        "outputId": "eef6290b-21a3-4645-cfca-02f94e7a3526",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "from keras.datasets import imdb\n",
        "\n",
        "(train_data, train_labels), (test_data, test_labels) = imdb.load_data(num_words=10000)"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Using TensorFlow backend.\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3aIvbgKINqX_",
        "colab_type": "text"
      },
      "source": [
        "View some of the decoded texts:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ai34NRECNoiF",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# word_index is a dictionary mapping words to an integer index\n",
        "word_index = imdb.get_word_index()\n",
        "# We reverse it, mapping integer indices to words\n",
        "reverse_word_index = dict([(value, key) for (key, value) in word_index.items()])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tBnin22ONzBO",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Function to show decoded text in a row\n",
        "def decode_sentence(index):\n",
        "  return ' '.join([reverse_word_index.get(i - 3, '?') for i in train_data[index]])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "G8FEhExZOAkK",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 124
        },
        "outputId": "3cbef419-46a8-420c-de5e-33c8f155acd4"
      },
      "source": [
        "# Display 5 of the decoded rows\n",
        "for i in range(5):\n",
        "  print(decode_sentence(i))"
      ],
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "? this film was just brilliant casting location scenery story direction everyone's really suited the part they played and you could just imagine being there robert ? is an amazing actor and now the same being director ? father came from the same scottish island as myself so i loved the fact there was a real connection with this film the witty remarks throughout the film were great it was just brilliant so much that i bought the film as soon as it was released for ? and would recommend it to everyone to watch and the fly fishing was amazing really cried at the end it was so sad and you know what they say if you cry at a film it must have been good and this definitely was also ? to the two little boy's that played the ? of norman and paul they were just brilliant children are often left out of the ? list i think because the stars that play them all grown up are such a big profile for the whole film but these children are amazing and should be praised for what they have done don't you think the whole story was so lovely because it was true and was someone's life after all that was shared with us all\n",
            "? big hair big boobs bad music and a giant safety pin these are the words to best describe this terrible movie i love cheesy horror movies and i've seen hundreds but this had got to be on of the worst ever made the plot is paper thin and ridiculous the acting is an abomination the script is completely laughable the best is the end showdown with the cop and how he worked out who the killer is it's just so damn terribly written the clothes are sickening and funny in equal ? the hair is big lots of boobs ? men wear those cut ? shirts that show off their ? sickening that men actually wore them and the music is just ? trash that plays over and over again in almost every scene there is trashy music boobs and ? taking away bodies and the gym still doesn't close for ? all joking aside this is a truly bad film whose only charm is to look back on the disaster that was the 80's and have a good old laugh at how bad everything was back then\n",
            "? this has to be one of the worst films of the 1990s when my friends i were watching this film being the target audience it was aimed at we just sat watched the first half an hour with our jaws touching the floor at how bad it really was the rest of the time everyone else in the theatre just started talking to each other leaving or generally crying into their popcorn that they actually paid money they had ? working to watch this feeble excuse for a film it must have looked like a great idea on paper but on film it looks like no one in the film has a clue what is going on crap acting crap costumes i can't get across how ? this is to watch save yourself an hour a bit of your life\n",
            "? the ? ? at storytelling the traditional sort many years after the event i can still see in my ? eye an elderly lady my friend's mother retelling the battle of ? she makes the characters come alive her passion is that of an eye witness one to the events on the ? heath a mile or so from where she lives br br of course it happened many years before she was born but you wouldn't guess from the way she tells it the same story is told in bars the length and ? of scotland as i discussed it with a friend one night in ? a local cut in to give his version the discussion continued to closing time br br stories passed down like this become part of our being who doesn't remember the stories our parents told us when we were children they become our invisible world and as we grow older they maybe still serve as inspiration or as an emotional ? fact and fiction blend with ? role models warning stories ? magic and mystery br br my name is ? like my grandfather and his grandfather before him our protagonist introduces himself to us and also introduces the story that stretches back through generations it produces stories within stories stories that evoke the ? wonder of scotland its rugged mountains ? in ? the stuff of legend yet ? is ? in reality this is what gives it its special charm it has a rough beauty and authenticity ? with some of the finest ? singing you will ever hear br br ? ? visits his grandfather in hospital shortly before his death he burns with frustration part of him ? to be in the twenty first century to hang out in ? but he is raised on the western ? among a ? speaking community br br yet there is a deeper conflict within him he ? to know the truth the truth behind his ? ancient stories where does fiction end and he wants to know the truth behind the death of his parents br br he is pulled to make a last ? journey to the ? of one of ? most ? mountains can the truth be told or is it all in stories br br in this story about stories we ? bloody battles ? lovers the ? of old and the sometimes more ? ? of accepted truth in doing so we each connect with ? as he lives the story of his own life br br ? the ? ? is probably the most honest ? and genuinely beautiful film of scotland ever made like ? i got slightly annoyed with the ? of hanging stories on more stories but also like ? i ? this once i saw the ? picture ' forget the box office ? of braveheart and its like you might even ? the ? famous ? of the wicker man to see a film that is true to scotland this one is probably unique if you maybe ? on it deeply enough you might even re ? the power of storytelling and the age old question of whether there are some truths that cannot be told but only experienced\n",
            "? worst mistake of my life br br i picked this movie up at target for 5 because i figured hey it's sandler i can get some cheap laughs i was wrong completely wrong mid way through the film all three of my friends were asleep and i was still suffering worst plot worst script worst movie i have ever seen i wanted to hit my head up against a wall for an hour then i'd stop and you know why because it felt damn good upon bashing my head in i stuck that damn movie in the ? and watched it burn and that felt better than anything else i've ever done it took american psycho army of darkness and kill bill just to get over that crap i hate you sandler for actually going through with this and ruining a whole day of my life\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gv-r2R-SPAjD",
        "colab_type": "text"
      },
      "source": [
        "Vectorizing the data:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wbdnLkO8PETJ",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import numpy as np\n",
        "\n",
        "def vectorize_sequences(sequences, dimension=10000):\n",
        "    # Create an all-zero matrix of shape (len(sequences), dimension)\n",
        "    results = np.zeros((len(sequences), dimension))\n",
        "    for i, sequence in enumerate(sequences):\n",
        "        results[i, sequence] = 1.  # set specific indices of results[i] to 1s\n",
        "    return results\n",
        "\n",
        "# Our vectorized training data\n",
        "x_train = vectorize_sequences(train_data)\n",
        "# Our vectorized test data\n",
        "x_test = vectorize_sequences(test_data)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "YGEls-UjRUJA",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Our vectorized labels\n",
        "y_train = np.asarray(train_labels).astype('float32')\n",
        "y_test = np.asarray(test_labels).astype('float32')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "PswVnMugRfWn",
        "colab_type": "text"
      },
      "source": [
        "## Building neural network"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gMczn3MFqkef",
        "colab_type": "text"
      },
      "source": [
        "Prepare keras and parameters:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DMh1g9qOjjNv",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from keras import models\n",
        "from keras import layers\n",
        "from keras import losses\n",
        "from keras import metrics\n",
        "from keras import optimizers\n",
        "\n",
        "import os\n",
        "os.environ['TF_CPP_MIN_LOG_LEVEL'] = '3'\n",
        "x_val = x_train[:10000]\n",
        "partial_x_train = x_train[10000:]\n",
        "\n",
        "y_val = y_train[:10000]\n",
        "partial_y_train = y_train[10000:]"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "IoXUKSm_qtTc",
        "colab_type": "text"
      },
      "source": [
        "Create a function to plot the \"Training and validation loss\" and \"Training and validation accuracy\""
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lylN43nOqtyc",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "\n",
        "import matplotlib.pyplot as plt\n",
        "# dict_keys(['val_loss', 'val_binary_accuracy', 'loss', 'binary_accuracy'])\n",
        "def validation_plot(history):\n",
        "  \n",
        "  history_dict = history.history\n",
        "  history_dict.keys()\n",
        "  binary_accuracy = history.history['binary_accuracy']\n",
        "  \n",
        "  val_binary_accuracy = history.history['val_binary_accuracy']\n",
        "\n",
        "  loss = history.history['loss']\n",
        "  val_loss = history.history['val_loss']\n",
        "\n",
        "  epochs = range(1, len(binary_accuracy) + 1)\n",
        "\n",
        "  # \"bo\" is for \"blue dot\"\n",
        "  plt.plot(epochs, loss, 'bo', label='Training loss')\n",
        "  # b is for \"solid blue line\"\n",
        "  plt.plot(epochs, val_loss, 'b', label='Validation loss')\n",
        "  plt.title('Training and validation loss')\n",
        "  plt.xlabel('Epochs')\n",
        "  plt.ylabel('Loss')\n",
        "  plt.legend()\n",
        "\n",
        "  plt.show()\n",
        "  \n",
        "  plt.clf()   # clear figure\n",
        "  # acc_values = history_dict['acc']\n",
        "  binary_accuracy = history_dict['binary_accuracy']\n",
        "\n",
        "  # val_acc_values = history_dict['val_acc']\n",
        "  val_binary_accuracy = history_dict['val_binary_accuracy']\n",
        "\n",
        "  plt.plot(epochs, binary_accuracy, 'bo', label='Training binary_accuracy')\n",
        "  # plt.plot(epochs, val_acc, 'b', label='Validation binary_accuracy')\n",
        "  plt.plot(epochs, val_binary_accuracy, 'b', label='Validation binary_accuracy')\n",
        "\n",
        "  plt.title('Training and validation accuracy')\n",
        "  plt.xlabel('Epochs')\n",
        "  plt.ylabel('Loss')\n",
        "  plt.legend()\n",
        "\n",
        "  plt.show()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "C-19LnyYSsK_",
        "colab_type": "text"
      },
      "source": [
        "### **Experiment 1:** Using 1 hidden layers"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8BBNkz_7RtA1",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 263
        },
        "outputId": "ab066786-4698-4a23-f28f-22583bf1381f"
      },
      "source": [
        "model1 = models.Sequential()\n",
        "model1.add(layers.Dense(16, activation='relu', input_shape=(10000,)))\n",
        "# model1.add(layers.Dense(16, activation='relu')) # Make it just one hiden layer\n",
        "model1.add(layers.Dense(1, activation='sigmoid'))\n",
        "model1.compile(optimizer=optimizers.RMSprop(lr=0.001),\n",
        "              loss=losses.binary_crossentropy,\n",
        "              metrics=[metrics.binary_accuracy])"
      ],
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:66: The name tf.get_default_graph is deprecated. Please use tf.compat.v1.get_default_graph instead.\n",
            "\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:541: The name tf.placeholder is deprecated. Please use tf.compat.v1.placeholder instead.\n",
            "\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:4432: The name tf.random_uniform is deprecated. Please use tf.random.uniform instead.\n",
            "\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/optimizers.py:793: The name tf.train.Optimizer is deprecated. Please use tf.compat.v1.train.Optimizer instead.\n",
            "\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:3657: The name tf.log is deprecated. Please use tf.math.log instead.\n",
            "\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow/python/ops/nn_impl.py:180: add_dispatch_support.<locals>.wrapper (from tensorflow.python.ops.array_ops) is deprecated and will be removed in a future version.\n",
            "Instructions for updating:\n",
            "Use tf.where in 2.0, which has the same broadcast rule as np.where\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "lfmZkcToS2ZE",
        "colab_type": "text"
      },
      "source": [
        "**Evaluation Measure :** Code for Validation and Testing Accuracy (one layer)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "TOGQlDH1S7LT",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 228
        },
        "outputId": "945849aa-e73a-4461-9fdf-42c81cb5f2ba"
      },
      "source": [
        "\n",
        "h1 = model1.fit(partial_x_train,\n",
        "                    partial_y_train,\n",
        "                    epochs=4,\n",
        "                    batch_size=512,\n",
        "                    validation_data=(x_val, y_val))"
      ],
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:1033: The name tf.assign_add is deprecated. Please use tf.compat.v1.assign_add instead.\n",
            "\n",
            "Train on 15000 samples, validate on 10000 samples\n",
            "Epoch 1/4\n",
            "15000/15000 [==============================] - 2s 146us/step - loss: 0.5001 - binary_accuracy: 0.8008 - val_loss: 0.3943 - val_binary_accuracy: 0.8674\n",
            "Epoch 2/4\n",
            "15000/15000 [==============================] - 2s 128us/step - loss: 0.3229 - binary_accuracy: 0.9025 - val_loss: 0.3257 - val_binary_accuracy: 0.8821\n",
            "Epoch 3/4\n",
            "15000/15000 [==============================] - 2s 134us/step - loss: 0.2519 - binary_accuracy: 0.9225 - val_loss: 0.2958 - val_binary_accuracy: 0.8863\n",
            "Epoch 4/4\n",
            "15000/15000 [==============================] - 2s 135us/step - loss: 0.2072 - binary_accuracy: 0.9352 - val_loss: 0.2788 - val_binary_accuracy: 0.8899\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "2TRELMcgkgTP",
        "colab_type": "text"
      },
      "source": [
        "### Experiment 1: Using 3 hidden layers"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "1RvCTFsQlWXB",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "model2 = models.Sequential()\n",
        "model2.add(layers.Dense(16, activation='relu', input_shape=(10000,)))\n",
        "model2.add(layers.Dense(16, activation='relu')) \n",
        "model2.add(layers.Dense(16, activation='relu')) # Add one more layer to make it three\n",
        "model2.add(layers.Dense(1, activation='sigmoid'))\n",
        "model2.compile(optimizer=optimizers.RMSprop(lr=0.001),\n",
        "              loss=losses.binary_crossentropy,\n",
        "              metrics=[metrics.binary_accuracy])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Y2Nyi3UgodjU",
        "colab_type": "text"
      },
      "source": [
        "**Evaluation Measure :** Code for Validation and Testing Accuracy (three layer)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "TkonOgQIluIV",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 173
        },
        "outputId": "925f36ad-d762-4f1d-8c4d-0023be9b0210"
      },
      "source": [
        "\n",
        "h2 = model2.fit(partial_x_train,\n",
        "                    partial_y_train,\n",
        "                    epochs=4,\n",
        "                    batch_size=512,\n",
        "                    validation_data=(x_val, y_val))"
      ],
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Train on 15000 samples, validate on 10000 samples\n",
            "Epoch 1/4\n",
            "15000/15000 [==============================] - 2s 156us/step - loss: 0.5263 - binary_accuracy: 0.7700 - val_loss: 0.3802 - val_binary_accuracy: 0.8665\n",
            "Epoch 2/4\n",
            "15000/15000 [==============================] - 2s 135us/step - loss: 0.2969 - binary_accuracy: 0.8975 - val_loss: 0.2940 - val_binary_accuracy: 0.8876\n",
            "Epoch 3/4\n",
            "15000/15000 [==============================] - 2s 127us/step - loss: 0.2038 - binary_accuracy: 0.9311 - val_loss: 0.3563 - val_binary_accuracy: 0.8585\n",
            "Epoch 4/4\n",
            "15000/15000 [==============================] - 2s 132us/step - loss: 0.1554 - binary_accuracy: 0.9463 - val_loss: 0.3024 - val_binary_accuracy: 0.8800\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "n24yvp6Rx0zp",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "8fa1d535-e9da-45c4-cc6e-ed902836f669"
      },
      "source": [
        "print(\"One Layer validation\")\n",
        "validation_plot(h1)\n",
        "\n",
        "print(\"Three Layers validation\")\n",
        "validation_plot(h2)\n",
        "\n"
      ],
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "One Layer validation\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYsAAAEWCAYAAACXGLsWAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nEYlLYSEiInEpLEREJK7/D4iSfZGGiuMaAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYsAAAEWCAYAAACXGLsWAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Three Layers validation\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYsAAAEWCAYAAACXGLsWAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZIAAAEWCAYAAABMoxE0AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "mM8-ERwDnEGO",
        "colab_type": "text"
      },
      "source": [
        "**Conclusion #1:**\n",
        "\n",
        "Following is the result from my executions: \n",
        "\n",
        "*20 Epochs*\n",
        "\n",
        "\n",
        "  The model seems overfitting when number of layes are increased.\n",
        "  \n",
        "  The training loss is decreasing while validation loss is increasing in three layers. Also, training accuracy is almost not changing while validation accuracy is decreasing.\n",
        "  \n",
        "\n",
        "*   (1 Layer) loss: 0.0236 - binary_accuracy: 0.9976 - val_loss: 0.4639 - val_binary_accuracy: 0.8712\n",
        "*   (3 Layers) loss: 0.0092 - binary_accuracy: 0.9971 - val_loss: 0.7938 - val_binary_accuracy: 0.8587\n",
        "\n",
        "\n",
        "*4 Epochs*\n",
        "\n",
        "\n",
        "  The model performs better when number of layers are increased.\n",
        "\n",
        "\n",
        "*   (1 Layer) loss: 0.2020 - binary_accuracy: 0.9389 - val_loss: 0.2874 - val_binary_accuracy: 0.8856\n",
        "*   (3 Layers) loss: 0.1828 - binary_accuracy: 0.9452 - val_loss: 0.2813 - val_binary_accuracy: 0.8876\n",
        "\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "bpQH6tGhOf1i",
        "colab_type": "text"
      },
      "source": [
        "\n",
        "\n",
        "---\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "pmrrwUmU9bWt",
        "colab_type": "text"
      },
      "source": [
        "**Experiment 2:** Using more Hidden Units from 16 to 32 in one hidden layer"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wJETOHOSRO9D",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "model3 = models.Sequential()\n",
        "model3.add(layers.Dense(64, activation='relu', input_shape=(10000,)))\n",
        "model3.add(layers.Dense(1, activation='sigmoid'))\n",
        "model3.compile(optimizer=optimizers.RMSprop(lr=0.001),\n",
        "              loss=losses.binary_crossentropy,\n",
        "              metrics=[metrics.binary_accuracy])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "lV85EUFMBU6n",
        "colab_type": "text"
      },
      "source": [
        "**Evaluation Measure :** Code for Validation and Testing Accuracy (one layer 32 units)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "RdBYsfXcRPKn",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 173
        },
        "outputId": "308a63df-9420-426c-b06e-03e38ec6bb89"
      },
      "source": [
        "\n",
        "h3 = model3.fit(partial_x_train,\n",
        "                    partial_y_train,\n",
        "                    epochs=4,\n",
        "                    batch_size=512,\n",
        "                    validation_data=(x_val, y_val))"
      ],
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Train on 15000 samples, validate on 10000 samples\n",
            "Epoch 1/4\n",
            "15000/15000 [==============================] - 3s 185us/step - loss: 0.4632 - binary_accuracy: 0.7938 - val_loss: 0.3395 - val_binary_accuracy: 0.8781\n",
            "Epoch 2/4\n",
            "15000/15000 [==============================] - 2s 160us/step - loss: 0.2704 - binary_accuracy: 0.9085 - val_loss: 0.2991 - val_binary_accuracy: 0.8802\n",
            "Epoch 3/4\n",
            "15000/15000 [==============================] - 2s 163us/step - loss: 0.2010 - binary_accuracy: 0.9345 - val_loss: 0.2888 - val_binary_accuracy: 0.8837\n",
            "Epoch 4/4\n",
            "15000/15000 [==============================] - 2s 161us/step - loss: 0.1608 - binary_accuracy: 0.9472 - val_loss: 0.2765 - val_binary_accuracy: 0.8911\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lWAgQ-0JBlc6",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 590
        },
        "outputId": "c65a9d24-9424-4cbc-dee7-75a20eaa7669"
      },
      "source": [
        "print(\"One Layer validation with 32 units\")\n",
        "validation_plot(h3)\n",
        "\n"
      ],
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "One Layer validation with 32 units\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYsAAAEWCAYAAACXGLsWAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n237k3YSsAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYsAAAEWCAYAAACXGLsWAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nLo8XtJyn/G4AAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "yzAnHoNsCc4c",
        "colab_type": "text"
      },
      "source": [
        "**Conclusion #2:**\n",
        "\n",
        " Following is the result from my executions:\n",
        "\n",
        "*20 Epochs:*\n",
        "\n",
        "  The model seems overfitting when number of units are increased.\n",
        "\n",
        "  The training loss is decreasing while validation loss is increasing in more layers. Also, training accuracy is increasing while validation accuracy is decreasing.\n",
        "\n",
        "\n",
        "\n",
        "*   (16 units) loss: 0.0236 - binary_accuracy: 0.9976 - val_loss: 0.4639 - val_binary_accuracy: 0.8712\n",
        "*   (32 units) loss: 0.0132 - binary_accuracy: 0.9992 - val_loss: 0.5562 - val_binary_accuracy: 0.8677\n",
        "\n",
        "\n",
        "*4 Epochs:*\n",
        "\n",
        "\n",
        "  The model perform better with more neurons. Both loss get lower and accuracy get higher with higher number of hiden units.\n",
        "\n",
        "*   (16 units) loss: 0.2020 - binary_accuracy: 0.9389 - val_loss: 0.2874 - val_binary_accuracy: 0.8856\n",
        "*   (32 units) loss: 0.1598 - binary_accuracy: 0.9471 - val_loss: 0.2989 - val_binary_accuracy: 0.8820\n",
        "\n",
        "\n",
        "    \n",
        "    \n",
        "    "
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "LvEMXAGOGvNX",
        "colab_type": "text"
      },
      "source": [
        "--------"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "2HfibBOoG0zF",
        "colab_type": "text"
      },
      "source": [
        "**Experiment 3:** Using MSE instead of binary_crossentropy with one layer 16 hidden units"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "e5q4wgQYHN7W",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "model4 = models.Sequential()\n",
        "model4.add(layers.Dense(16, activation='relu', input_shape=(10000,)))\n",
        "model4.add(layers.Dense(1, activation='sigmoid'))\n",
        "model4.compile(optimizer=optimizers.RMSprop(lr=0.001),\n",
        "              loss=losses.MSE,\n",
        "              metrics=[metrics.binary_accuracy])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "XgRH7hC4Gz92",
        "colab_type": "text"
      },
      "source": [
        "**Evaluation Measure :** Code for Validation and Testing Accuracy (one layer 16 units - MSE)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pWl5HL3FH3Kk",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 173
        },
        "outputId": "1b457475-09ae-4c36-d788-7fddb7ffe22a"
      },
      "source": [
        "h4 = model4.fit(partial_x_train,\n",
        "                    partial_y_train,\n",
        "                    epochs=4,\n",
        "                    batch_size=512,\n",
        "                    validation_data=(x_val, y_val))"
      ],
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Train on 15000 samples, validate on 10000 samples\n",
            "Epoch 1/4\n",
            "15000/15000 [==============================] - 2s 143us/step - loss: 0.1660 - binary_accuracy: 0.7867 - val_loss: 0.1194 - val_binary_accuracy: 0.8702\n",
            "Epoch 2/4\n",
            "15000/15000 [==============================] - 2s 122us/step - loss: 0.0936 - binary_accuracy: 0.9063 - val_loss: 0.1003 - val_binary_accuracy: 0.8789\n",
            "Epoch 3/4\n",
            "15000/15000 [==============================] - 2s 123us/step - loss: 0.0701 - binary_accuracy: 0.9293 - val_loss: 0.0937 - val_binary_accuracy: 0.8787\n",
            "Epoch 4/4\n",
            "15000/15000 [==============================] - 2s 131us/step - loss: 0.0564 - binary_accuracy: 0.9449 - val_loss: 0.0854 - val_binary_accuracy: 0.8891\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "02XPDq4kH4u-",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 590
        },
        "outputId": "d6f4f71b-b0fa-47b1-fa75-8bfb74eaef5c"
      },
      "source": [
        "print(\"One Layer validation with 16 units and MSE\")\n",
        "validation_plot(h4)"
      ],
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "One Layer validation with 16 units and MSE\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYsAAAEWCAYAAACXGLsWAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQYUAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYsAAAEWCAYAAACXGLsWAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nDy0AAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "MpEZjs-ibCmT",
        "colab_type": "text"
      },
      "source": [
        "**Conclusion #3:**\n",
        "\n",
        "Following is the result from my executions:\n",
        "\n",
        "*20 Epochs*\n",
        "\n",
        "  The model seems perform better on MSE.\n",
        "\n",
        "  The training loss and validation both are decreasing on MSE. But, training accuracy and validation accuracy are almost the same.\n",
        "\n",
        "    \n",
        "    (binary_crossentropy) loss: 0.0236 - binary_accuracy: 0.9976 - val_loss: 0.4639 - val_binary_accuracy: 0.8712\n",
        "    (MSE) loss: 0.0070 - binary_accuracy: 0.9957 - val_loss: 0.1017 - val_binary_accuracy: 0.8699\n",
        "    \n",
        "    \n",
        "*4 Epochs*\n",
        "\n",
        "\n",
        "  The model performs better for MSE.\n",
        "\n",
        "\n",
        "*   (binary_crossentropy) loss: 0.2020 - binary_accuracy: 0.9389 - val_loss: 0.2874 - val_binary_accuracy: 0.8856\n",
        "*   (MSE) loss: 0.0573 - binary_accuracy: 0.9425 - val_loss: 0.0859 - val_binary_accuracy: 0.8889\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "i0X3e-J-bDE_",
        "colab_type": "text"
      },
      "source": [
        "----"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "YtYu8w8CcCiY",
        "colab_type": "text"
      },
      "source": [
        "**Experiment 4:** Using tanh activation instead of relu with one layer 16 hidden units"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AWo3dBxPbD3w",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "model5 = models.Sequential()\n",
        "model5.add(layers.Dense(16, activation='tanh', input_shape=(10000,)))\n",
        "model5.add(layers.Dense(1, activation='sigmoid'))\n",
        "model5.compile(optimizer=optimizers.RMSprop(lr=0.001),\n",
        "              loss=losses.MSE,\n",
        "              metrics=[metrics.binary_accuracy])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "YZyZr1mQcIpY",
        "colab_type": "text"
      },
      "source": [
        "**Evaluation Measure :** Code for Validation and Testing Accuracy (one layer 16 units - tanh)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3DXSfe1zcRFI",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 173
        },
        "outputId": "045fe0c5-9a5b-40d9-e762-07a207df2aee"
      },
      "source": [
        "h5 = model4.fit(partial_x_train,\n",
        "                    partial_y_train,\n",
        "                    epochs=4,\n",
        "                    batch_size=512,\n",
        "                    validation_data=(x_val, y_val))"
      ],
      "execution_count": 21,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Train on 15000 samples, validate on 10000 samples\n",
            "Epoch 1/4\n",
            "15000/15000 [==============================] - 2s 143us/step - loss: 0.0474 - binary_accuracy: 0.9531 - val_loss: 0.0850 - val_binary_accuracy: 0.8838\n",
            "Epoch 2/4\n",
            "15000/15000 [==============================] - 2s 128us/step - loss: 0.0403 - binary_accuracy: 0.9613 - val_loss: 0.0841 - val_binary_accuracy: 0.8851\n",
            "Epoch 3/4\n",
            "15000/15000 [==============================] - 2s 131us/step - loss: 0.0346 - binary_accuracy: 0.9688 - val_loss: 0.0839 - val_binary_accuracy: 0.8861\n",
            "Epoch 4/4\n",
            "15000/15000 [==============================] - 2s 130us/step - loss: 0.0300 - binary_accuracy: 0.9745 - val_loss: 0.0844 - val_binary_accuracy: 0.8833\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "q_iKFEbpds-F",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 590
        },
        "outputId": "253d2d84-5bfe-48f1-d294-fdfa1b683e80"
      },
      "source": [
        "print(\"One Layer validation with 16 units and MSE\")\n",
        "validation_plot(h4)"
      ],
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "One Layer validation with 16 units and MSE\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYsAAAEWCAYAAACXGLsWAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQYUAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYsAAAEWCAYAAACXGLsWAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nDy0AAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "S-zg_W4HeA6d",
        "colab_type": "text"
      },
      "source": [
        "**Conclusion #4:**\n",
        " \n",
        " \n",
        " Following is the result from my executions:\n",
        " \n",
        "*20 Epochs*\n",
        " \n",
        "  The model seems berform better on tanh.\n",
        "\n",
        "  The training loss and validation both are decreasing on MSE. But, training accuracy accuracy are almost the same, validation accuracy is decreased.\n",
        "\n",
        "    \n",
        "    (relu) loss: 0.0236 - binary_accuracy: 0.9976 - val_loss: 0.4639 - val_binary_accuracy: 0.8712\n",
        "    (tanh) loss: 0.0017 - binary_accuracy: 0.9984 - val_loss: 0.1214 - val_binary_accuracy: 0.8579\n",
        "    \n",
        "    \n",
        "*4 Epochs*\n",
        "\n",
        "\n",
        "  The model performs better for tanh.\n",
        "\n",
        "\n",
        "*   (relu) loss: 0.2020 - binary_accuracy: 0.9389 - val_loss: 0.2874 - val_binary_accuracy: 0.8856\n",
        "*   (tanh) loss: 0.0303 - binary_accuracy: 0.9739 - val_loss: 0.0857 - val_binary_accuracy: 0.8831"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Yih2HY_Ld73_",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}