Week4KerasNeuralNetworkCode.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Week4KerasNeuralNetworkCode.ipynb",
      "provenance": [],
      "authorship_tag": "ABX9TyOF3uYbIVMLJ39vpW+jmNRi",
      "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/SiP-AI-ML/MD/blob/master/Week4KerasNeuralNetworkCode.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "sIL-VPRUPetm",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from numpy import loadtxt\n",
        "from keras.models import Sequential\n",
        "from keras.layers import Dense\n",
        "import matplotlib.pyplot as plt"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zKfiwmUvjZGl",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "dataset = loadtxt('https://raw.githubusercontent.com/SiP-AI-ML/LessonMaterials/master/pima-indians-diabetes.csv', delimiter = ',')\n",
        "X = dataset[:,0:8]\n",
        "y = dataset[:,8]"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Ob7Aa4TnkBSB",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 255
        },
        "outputId": "5786f293-afc9-4312-9e7e-658699eca0ef"
      },
      "source": [
        "model = Sequential()\n",
        "model.add(Dense(12, input_dim=8, activation='relu'))\n",
        "model.add(Dense(8, activation='relu'))\n",
        "model.add(Dense(1, activation='sigmoid'))\n",
        "model.summary()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential_9\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "dense_25 (Dense)             (None, 12)                108       \n",
            "_________________________________________________________________\n",
            "dense_26 (Dense)             (None, 8)                 104       \n",
            "_________________________________________________________________\n",
            "dense_27 (Dense)             (None, 1)                 9         \n",
            "=================================================================\n",
            "Total params: 221\n",
            "Trainable params: 221\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Wwspjt0rrW6w",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DO1QJ69hrpRi",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "bdea91bb-7b4e-4585-ad24-4a844885f2d2"
      },
      "source": [
        "training = model.fit(X, y, epochs=150, batch_size=10, verbose = 2)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/150\n",
            " - 0s - loss: 1.7774 - accuracy: 0.4831\n",
            "Epoch 2/150\n",
            " - 0s - loss: 0.8048 - accuracy: 0.5924\n",
            "Epoch 3/150\n",
            " - 0s - loss: 0.7051 - accuracy: 0.6497\n",
            "Epoch 4/150\n",
            " - 0s - loss: 0.6806 - accuracy: 0.6562\n",
            "Epoch 5/150\n",
            " - 0s - loss: 0.6501 - accuracy: 0.6745\n",
            "Epoch 6/150\n",
            " - 0s - loss: 0.6859 - accuracy: 0.6706\n",
            "Epoch 7/150\n",
            " - 0s - loss: 0.6434 - accuracy: 0.6745\n",
            "Epoch 8/150\n",
            " - 0s - loss: 0.6163 - accuracy: 0.7148\n",
            "Epoch 9/150\n",
            " - 0s - loss: 0.6114 - accuracy: 0.6940\n",
            "Epoch 10/150\n",
            " - 0s - loss: 0.6115 - accuracy: 0.6940\n",
            "Epoch 11/150\n",
            " - 0s - loss: 0.6204 - accuracy: 0.7044\n",
            "Epoch 12/150\n",
            " - 0s - loss: 0.6210 - accuracy: 0.6745\n",
            "Epoch 13/150\n",
            " - 0s - loss: 0.6058 - accuracy: 0.6966\n",
            "Epoch 14/150\n",
            " - 0s - loss: 0.6074 - accuracy: 0.7083\n",
            "Epoch 15/150\n",
            " - 0s - loss: 0.5963 - accuracy: 0.7174\n",
            "Epoch 16/150\n",
            " - 0s - loss: 0.5894 - accuracy: 0.7122\n",
            "Epoch 17/150\n",
            " - 0s - loss: 0.5669 - accuracy: 0.7201\n",
            "Epoch 18/150\n",
            " - 0s - loss: 0.5689 - accuracy: 0.7214\n",
            "Epoch 19/150\n",
            " - 0s - loss: 0.6047 - accuracy: 0.6992\n",
            "Epoch 20/150\n",
            " - 0s - loss: 0.5766 - accuracy: 0.7318\n",
            "Epoch 21/150\n",
            " - 0s - loss: 0.5749 - accuracy: 0.7370\n",
            "Epoch 22/150\n",
            " - 0s - loss: 0.5697 - accuracy: 0.7109\n",
            "Epoch 23/150\n",
            " - 0s - loss: 0.5740 - accuracy: 0.7044\n",
            "Epoch 24/150\n",
            " - 0s - loss: 0.5662 - accuracy: 0.7305\n",
            "Epoch 25/150\n",
            " - 0s - loss: 0.5726 - accuracy: 0.7031\n",
            "Epoch 26/150\n",
            " - 0s - loss: 0.5528 - accuracy: 0.7331\n",
            "Epoch 27/150\n",
            " - 0s - loss: 0.5705 - accuracy: 0.7188\n",
            "Epoch 28/150\n",
            " - 0s - loss: 0.5532 - accuracy: 0.7305\n",
            "Epoch 29/150\n",
            " - 0s - loss: 0.5510 - accuracy: 0.7253\n",
            "Epoch 30/150\n",
            " - 0s - loss: 0.5659 - accuracy: 0.7253\n",
            "Epoch 31/150\n",
            " - 0s - loss: 0.5377 - accuracy: 0.7331\n",
            "Epoch 32/150\n",
            " - 0s - loss: 0.5801 - accuracy: 0.7318\n",
            "Epoch 33/150\n",
            " - 0s - loss: 0.5531 - accuracy: 0.7370\n",
            "Epoch 34/150\n",
            " - 0s - loss: 0.5426 - accuracy: 0.7448\n",
            "Epoch 35/150\n",
            " - 0s - loss: 0.5581 - accuracy: 0.7279\n",
            "Epoch 36/150\n",
            " - 0s - loss: 0.5403 - accuracy: 0.7435\n",
            "Epoch 37/150\n",
            " - 0s - loss: 0.5431 - accuracy: 0.7331\n",
            "Epoch 38/150\n",
            " - 0s - loss: 0.5365 - accuracy: 0.7474\n",
            "Epoch 39/150\n",
            " - 0s - loss: 0.5478 - accuracy: 0.7174\n",
            "Epoch 40/150\n",
            " - 0s - loss: 0.5362 - accuracy: 0.7422\n",
            "Epoch 41/150\n",
            " - 0s - loss: 0.5478 - accuracy: 0.7370\n",
            "Epoch 42/150\n",
            " - 0s - loss: 0.5505 - accuracy: 0.7357\n",
            "Epoch 43/150\n",
            " - 0s - loss: 0.5393 - accuracy: 0.7357\n",
            "Epoch 44/150\n",
            " - 0s - loss: 0.5450 - accuracy: 0.7370\n",
            "Epoch 45/150\n",
            " - 0s - loss: 0.5318 - accuracy: 0.7448\n",
            "Epoch 46/150\n",
            " - 0s - loss: 0.5395 - accuracy: 0.7396\n",
            "Epoch 47/150\n",
            " - 0s - loss: 0.5459 - accuracy: 0.7396\n",
            "Epoch 48/150\n",
            " - 0s - loss: 0.5597 - accuracy: 0.7422\n",
            "Epoch 49/150\n",
            " - 0s - loss: 0.5335 - accuracy: 0.7383\n",
            "Epoch 50/150\n",
            " - 0s - loss: 0.5298 - accuracy: 0.7396\n",
            "Epoch 51/150\n",
            " - 0s - loss: 0.5318 - accuracy: 0.7487\n",
            "Epoch 52/150\n",
            " - 0s - loss: 0.5336 - accuracy: 0.7565\n",
            "Epoch 53/150\n",
            " - 0s - loss: 0.5300 - accuracy: 0.7578\n",
            "Epoch 54/150\n",
            " - 0s - loss: 0.5670 - accuracy: 0.7305\n",
            "Epoch 55/150\n",
            " - 0s - loss: 0.5317 - accuracy: 0.7370\n",
            "Epoch 56/150\n",
            " - 0s - loss: 0.5217 - accuracy: 0.7539\n",
            "Epoch 57/150\n",
            " - 0s - loss: 0.5240 - accuracy: 0.7396\n",
            "Epoch 58/150\n",
            " - 0s - loss: 0.5162 - accuracy: 0.7526\n",
            "Epoch 59/150\n",
            " - 0s - loss: 0.5175 - accuracy: 0.7448\n",
            "Epoch 60/150\n",
            " - 0s - loss: 0.5474 - accuracy: 0.7344\n",
            "Epoch 61/150\n",
            " - 0s - loss: 0.5351 - accuracy: 0.7435\n",
            "Epoch 62/150\n",
            " - 0s - loss: 0.5181 - accuracy: 0.7474\n",
            "Epoch 63/150\n",
            " - 0s - loss: 0.5231 - accuracy: 0.7344\n",
            "Epoch 64/150\n",
            " - 0s - loss: 0.5192 - accuracy: 0.7487\n",
            "Epoch 65/150\n",
            " - 0s - loss: 0.5193 - accuracy: 0.7487\n",
            "Epoch 66/150\n",
            " - 0s - loss: 0.5175 - accuracy: 0.7487\n",
            "Epoch 67/150\n",
            " - 0s - loss: 0.5096 - accuracy: 0.7435\n",
            "Epoch 68/150\n",
            " - 0s - loss: 0.5183 - accuracy: 0.7422\n",
            "Epoch 69/150\n",
            " - 0s - loss: 0.5140 - accuracy: 0.7461\n",
            "Epoch 70/150\n",
            " - 0s - loss: 0.5266 - accuracy: 0.7487\n",
            "Epoch 71/150\n",
            " - 0s - loss: 0.5124 - accuracy: 0.7500\n",
            "Epoch 72/150\n",
            " - 0s - loss: 0.4981 - accuracy: 0.7474\n",
            "Epoch 73/150\n",
            " - 0s - loss: 0.5083 - accuracy: 0.7526\n",
            "Epoch 74/150\n",
            " - 0s - loss: 0.5253 - accuracy: 0.7474\n",
            "Epoch 75/150\n",
            " - 0s - loss: 0.5024 - accuracy: 0.7578\n",
            "Epoch 76/150\n",
            " - 0s - loss: 0.4999 - accuracy: 0.7526\n",
            "Epoch 77/150\n",
            " - 0s - loss: 0.5033 - accuracy: 0.7591\n",
            "Epoch 78/150\n",
            " - 0s - loss: 0.5091 - accuracy: 0.7617\n",
            "Epoch 79/150\n",
            " - 0s - loss: 0.5005 - accuracy: 0.7539\n",
            "Epoch 80/150\n",
            " - 0s - loss: 0.4959 - accuracy: 0.7656\n",
            "Epoch 81/150\n",
            " - 0s - loss: 0.5144 - accuracy: 0.7552\n",
            "Epoch 82/150\n",
            " - 0s - loss: 0.4981 - accuracy: 0.7578\n",
            "Epoch 83/150\n",
            " - 0s - loss: 0.5045 - accuracy: 0.7591\n",
            "Epoch 84/150\n",
            " - 0s - loss: 0.5138 - accuracy: 0.7617\n",
            "Epoch 85/150\n",
            " - 0s - loss: 0.4939 - accuracy: 0.7786\n",
            "Epoch 86/150\n",
            " - 0s - loss: 0.4862 - accuracy: 0.7708\n",
            "Epoch 87/150\n",
            " - 0s - loss: 0.5044 - accuracy: 0.7487\n",
            "Epoch 88/150\n",
            " - 0s - loss: 0.5181 - accuracy: 0.7461\n",
            "Epoch 89/150\n",
            " - 0s - loss: 0.5116 - accuracy: 0.7513\n",
            "Epoch 90/150\n",
            " - 0s - loss: 0.5049 - accuracy: 0.7630\n",
            "Epoch 91/150\n",
            " - 0s - loss: 0.4995 - accuracy: 0.7591\n",
            "Epoch 92/150\n",
            " - 0s - loss: 0.4878 - accuracy: 0.7552\n",
            "Epoch 93/150\n",
            " - 0s - loss: 0.5007 - accuracy: 0.7578\n",
            "Epoch 94/150\n",
            " - 0s - loss: 0.4897 - accuracy: 0.7682\n",
            "Epoch 95/150\n",
            " - 0s - loss: 0.4947 - accuracy: 0.7630\n",
            "Epoch 96/150\n",
            " - 0s - loss: 0.4985 - accuracy: 0.7604\n",
            "Epoch 97/150\n",
            " - 0s - loss: 0.4972 - accuracy: 0.7552\n",
            "Epoch 98/150\n",
            " - 0s - loss: 0.5176 - accuracy: 0.7474\n",
            "Epoch 99/150\n",
            " - 0s - loss: 0.4814 - accuracy: 0.7734\n",
            "Epoch 100/150\n",
            " - 0s - loss: 0.4909 - accuracy: 0.7656\n",
            "Epoch 101/150\n",
            " - 0s - loss: 0.4968 - accuracy: 0.7552\n",
            "Epoch 102/150\n",
            " - 0s - loss: 0.4808 - accuracy: 0.7695\n",
            "Epoch 103/150\n",
            " - 0s - loss: 0.4851 - accuracy: 0.7565\n",
            "Epoch 104/150\n",
            " - 0s - loss: 0.4906 - accuracy: 0.7669\n",
            "Epoch 105/150\n",
            " - 0s - loss: 0.4995 - accuracy: 0.7539\n",
            "Epoch 106/150\n",
            " - 0s - loss: 0.4929 - accuracy: 0.7669\n",
            "Epoch 107/150\n",
            " - 0s - loss: 0.4879 - accuracy: 0.7695\n",
            "Epoch 108/150\n",
            " - 0s - loss: 0.4920 - accuracy: 0.7526\n",
            "Epoch 109/150\n",
            " - 0s - loss: 0.4881 - accuracy: 0.7734\n",
            "Epoch 110/150\n",
            " - 0s - loss: 0.4779 - accuracy: 0.7695\n",
            "Epoch 111/150\n",
            " - 0s - loss: 0.4980 - accuracy: 0.7513\n",
            "Epoch 112/150\n",
            " - 0s - loss: 0.4944 - accuracy: 0.7695\n",
            "Epoch 113/150\n",
            " - 0s - loss: 0.4786 - accuracy: 0.7695\n",
            "Epoch 114/150\n",
            " - 0s - loss: 0.4853 - accuracy: 0.7643\n",
            "Epoch 115/150\n",
            " - 0s - loss: 0.4774 - accuracy: 0.7604\n",
            "Epoch 116/150\n",
            " - 0s - loss: 0.4950 - accuracy: 0.7591\n",
            "Epoch 117/150\n",
            " - 0s - loss: 0.4784 - accuracy: 0.7656\n",
            "Epoch 118/150\n",
            " - 0s - loss: 0.4800 - accuracy: 0.7630\n",
            "Epoch 119/150\n",
            " - 0s - loss: 0.4875 - accuracy: 0.7734\n",
            "Epoch 120/150\n",
            " - 0s - loss: 0.4820 - accuracy: 0.7812\n",
            "Epoch 121/150\n",
            " - 0s - loss: 0.4780 - accuracy: 0.7630\n",
            "Epoch 122/150\n",
            " - 0s - loss: 0.5029 - accuracy: 0.7500\n",
            "Epoch 123/150\n",
            " - 0s - loss: 0.4817 - accuracy: 0.7617\n",
            "Epoch 124/150\n",
            " - 0s - loss: 0.4822 - accuracy: 0.7669\n",
            "Epoch 125/150\n",
            " - 0s - loss: 0.4817 - accuracy: 0.7643\n",
            "Epoch 126/150\n",
            " - 0s - loss: 0.4863 - accuracy: 0.7682\n",
            "Epoch 127/150\n",
            " - 0s - loss: 0.4803 - accuracy: 0.7734\n",
            "Epoch 128/150\n",
            " - 0s - loss: 0.4719 - accuracy: 0.7826\n",
            "Epoch 129/150\n",
            " - 0s - loss: 0.4911 - accuracy: 0.7734\n",
            "Epoch 130/150\n",
            " - 0s - loss: 0.4744 - accuracy: 0.7630\n",
            "Epoch 131/150\n",
            " - 0s - loss: 0.4684 - accuracy: 0.7799\n",
            "Epoch 132/150\n",
            " - 0s - loss: 0.4733 - accuracy: 0.7747\n",
            "Epoch 133/150\n",
            " - 0s - loss: 0.4796 - accuracy: 0.7734\n",
            "Epoch 134/150\n",
            " - 0s - loss: 0.4587 - accuracy: 0.7786\n",
            "Epoch 135/150\n",
            " - 0s - loss: 0.4809 - accuracy: 0.7565\n",
            "Epoch 136/150\n",
            " - 0s - loss: 0.4786 - accuracy: 0.7656\n",
            "Epoch 137/150\n",
            " - 0s - loss: 0.4867 - accuracy: 0.7617\n",
            "Epoch 138/150\n",
            " - 0s - loss: 0.4864 - accuracy: 0.7591\n",
            "Epoch 139/150\n",
            " - 0s - loss: 0.4714 - accuracy: 0.7786\n",
            "Epoch 140/150\n",
            " - 0s - loss: 0.4643 - accuracy: 0.7839\n",
            "Epoch 141/150\n",
            " - 0s - loss: 0.4959 - accuracy: 0.7604\n",
            "Epoch 142/150\n",
            " - 0s - loss: 0.4736 - accuracy: 0.7826\n",
            "Epoch 143/150\n",
            " - 0s - loss: 0.4939 - accuracy: 0.7734\n",
            "Epoch 144/150\n",
            " - 0s - loss: 0.4776 - accuracy: 0.7682\n",
            "Epoch 145/150\n",
            " - 0s - loss: 0.4777 - accuracy: 0.7630\n",
            "Epoch 146/150\n",
            " - 0s - loss: 0.4627 - accuracy: 0.7721\n",
            "Epoch 147/150\n",
            " - 0s - loss: 0.4643 - accuracy: 0.7773\n",
            "Epoch 148/150\n",
            " - 0s - loss: 0.4637 - accuracy: 0.7786\n",
            "Epoch 149/150\n",
            " - 0s - loss: 0.4791 - accuracy: 0.7695\n",
            "Epoch 150/150\n",
            " - 0s - loss: 0.4616 - accuracy: 0.7734\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Xac8oP1xsHyO",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "outputId": "b7b6707f-5c97-4a7f-ca86-ca437875d32f"
      },
      "source": [
        "plt.plot(training.history['accuracy'])\n",
        "plt.title('Model Accuracy')\n",
        "plt.ylabel('Accuracy')\n",
        "plt.xlabel('Epochs')\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SSH7K-6as994",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "outputId": "214ecd27-5b05-4c0d-e76f-a3feec0cecbf"
      },
      "source": [
        "plt.plot(training.history['loss'])\n",
        "plt.title('Model Loss')\n",
        "plt.ylabel('Loss')\n",
        "plt.xlabel('Epochs')\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4P7cQLmktbQE",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 102
        },
        "outputId": "c9595026-9b8d-4afb-b240-a460d06a6eb7"
      },
      "source": [
        "_, accuracy = model.evaluate(X, y)\n",
        "print(model.evaluate(X,y))\n",
        "print(accuracy * 100)\n",
        "print('Accuracy: %.2f' % (accuracy*100))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "768/768 [==============================] - 0s 39us/step\n",
            "768/768 [==============================] - 0s 19us/step\n",
            "[0.4498697618643443, 0.7890625]\n",
            "78.90625\n",
            "Accuracy: 78.91\n"
          ],
          "name": "stdout"
        }
      ]
    }
  ]
}