Copy_of_artificial_neural_network.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Copy of artificial_neural_network.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true,
      "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/sanit4u/Cookbook/blob/master/Copy_of_artificial_neural_network.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "lP6JLo1tGNBg",
        "colab_type": "text"
      },
      "source": [
        "# Artificial Neural Network"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gWZyYmS_UE_L",
        "colab_type": "text"
      },
      "source": [
        "### Importing the libraries"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MxkJoQBkUIHC",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "import tensorflow as tf"
      ],
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZaTwK7ojXr2F",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        },
        "outputId": "14cf78df-7f1e-4481-ac52-3192336d30ef"
      },
      "source": [
        "tf.__version__"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            },
            "text/plain": [
              "'2.2.0'"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 2
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "1E0Q3aoKUCRX",
        "colab_type": "text"
      },
      "source": [
        "## Part 1 - Data Preprocessing"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "cKWAkFVGUU0Z",
        "colab_type": "text"
      },
      "source": [
        "### Importing the dataset"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jRnJNgX94O1i",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "dataset = pd.read_csv(\"Churn_Modelling.csv\")\n",
        "x = dataset.iloc[:, 3:-1].values\n",
        "y = dataset.iloc[:, -1].values"
      ],
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jy-uqo895syT",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 136
        },
        "outputId": "2526374b-707e-4b8e-cca6-60dc3fc1e8b5"
      },
      "source": [
        "print(x)"
      ],
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[[619 'France' 'Female' ... 1 1 101348.88]\n",
            " [608 'Spain' 'Female' ... 0 1 112542.58]\n",
            " [502 'France' 'Female' ... 1 0 113931.57]\n",
            " ...\n",
            " [709 'France' 'Female' ... 0 1 42085.58]\n",
            " [772 'Germany' 'Male' ... 1 0 92888.52]\n",
            " [792 'France' 'Female' ... 1 0 38190.78]]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wFQeYyKz5u8z",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "b3dc6083-a5e2-4230-bb9b-de0199efb8f3"
      },
      "source": [
        "print(y)"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[1 0 1 ... 1 1 0]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "N6bQ0UgSU-NJ",
        "colab_type": "text"
      },
      "source": [
        "### Encoding categorical data"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "le5MJreAbW52",
        "colab_type": "text"
      },
      "source": [
        "Label Encoding the \"Gender\" column"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "RgUaof8A6WeW",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from sklearn.preprocessing import LabelEncoder\n",
        "le = LabelEncoder()\n",
        "x[:, 2] = le.fit_transform(x[:, 2])"
      ],
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yB_g0epG7GZW",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 136
        },
        "outputId": "325cc68d-2c38-43d6-a2ea-f8fc4da1c4c1"
      },
      "source": [
        "print(x)"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[[619 'France' 0 ... 1 1 101348.88]\n",
            " [608 'Spain' 0 ... 0 1 112542.58]\n",
            " [502 'France' 0 ... 1 0 113931.57]\n",
            " ...\n",
            " [709 'France' 0 ... 0 1 42085.58]\n",
            " [772 'Germany' 1 ... 1 0 92888.52]\n",
            " [792 'France' 0 ... 1 0 38190.78]]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "CUxGZezpbMcb",
        "colab_type": "text"
      },
      "source": [
        "One Hot Encoding the \"Geography\" column"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "R5f3H1Mi9cCC",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from sklearn.compose import ColumnTransformer\n",
        "from sklearn.preprocessing import OneHotEncoder\n",
        "ct = ColumnTransformer(transformers=[('encoder', OneHotEncoder(), [1])], remainder='passthrough')\n",
        "x = np.array(ct.fit_transform(x))"
      ],
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3KLlTcbu-t8m",
        "colab_type": "text"
      },
      "source": [
        "The dummy column which has been applied with onehotencoder, is shifted to first column."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Hse2uEJB-Tx6",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 136
        },
        "outputId": "03bc0090-2df9-4311-b75f-0ea4d0543dac"
      },
      "source": [
        "print(x)"
      ],
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[[1.0 0.0 0.0 ... 1 1 101348.88]\n",
            " [0.0 0.0 1.0 ... 0 1 112542.58]\n",
            " [1.0 0.0 0.0 ... 1 0 113931.57]\n",
            " ...\n",
            " [1.0 0.0 0.0 ... 0 1 42085.58]\n",
            " [0.0 1.0 0.0 ... 1 0 92888.52]\n",
            " [1.0 0.0 0.0 ... 1 0 38190.78]]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vHol938cW8zd",
        "colab_type": "text"
      },
      "source": [
        "### Splitting the dataset into the Training set and Test set"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "saFuIf_K-_jh",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "X_train, X_test, y_train, y_test = train_test_split(x, y, test_size = 0.25, random_state = 0 )"
      ],
      "execution_count": 10,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "RE_FcHyfV3TQ",
        "colab_type": "text"
      },
      "source": [
        "### Feature Scaling"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "MrvMxBT3_NFx",
        "colab_type": "text"
      },
      "source": [
        "*  Feature scaling is absolutely necessay for deep learning. That is fundamental.\n",
        "*  Here, we are apply the scaler object and transform to trainning datasets. and same scaler to test for transform "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QALQlDIy_bWh",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from sklearn.preprocessing import StandardScaler\n",
        "sc = StandardScaler()\n",
        "X_train= sc.fit_transform(X_train) \n",
        "X_test = sc.transform(X_test)"
      ],
      "execution_count": 11,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-zfEzkRVXIwF",
        "colab_type": "text"
      },
      "source": [
        "## Part 2 - Building the ANN"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "KvdeScabXtlB",
        "colab_type": "text"
      },
      "source": [
        "### Initializing the ANN"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ca1fFgjUC4MD",
        "colab_type": "text"
      },
      "source": [
        "* sequential is in the keras.\n",
        "Now keras is integrated with tensorFlow.\n",
        "* Sequential creates a sequential layer."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6sWDso86C0UY",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "ann = tf.keras.models.Sequential()"
      ],
      "execution_count": 12,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "rP6urV6SX7kS",
        "colab_type": "text"
      },
      "source": [
        "### Adding the input layer and the first hidden layer"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xJoIsWMPEQzJ",
        "colab_type": "text"
      },
      "source": [
        "Dense(Units , activation, ..) \n",
        "* Units: neurons . There is no rule for selection of number of neurons. its experimental. though it depends on the independent variables.\n",
        "* Activation : relu is code name for rectifier activation fucntion."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-3hhinaWDISl",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "ann.add(tf.keras.layers.Dense(units=6,activation='relu'))"
      ],
      "execution_count": 13,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "BELWAc_8YJze",
        "colab_type": "text"
      },
      "source": [
        "### Adding the second hidden layer"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JTXYHIJ4FWvm",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "ann.add(tf.keras.layers.Dense(units=6,activation='relu'))"
      ],
      "execution_count": 14,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "OyNEe6RXYcU4",
        "colab_type": "text"
      },
      "source": [
        "### Adding the output layer"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5OLu6yY3Fija",
        "colab_type": "text"
      },
      "source": [
        "Output Layer \n",
        "*  it depends on the dimension of output. Here in this example dimension is 1.\n",
        "* if there are multiple dimensions, then we need that many neurons to onehotencode those output dimensions.\n",
        "* In the output layer, when we have the binary, we need to use sigmoid function as activation function for the output layer.  "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4rcZSI1TFdUk",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "ann.add(tf.keras.layers.Dense(units=1,activation='sigmoid'))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "JT4u2S1_Y4WG",
        "colab_type": "text"
      },
      "source": [
        "## Part 3 - Training the ANN"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8GWlJChhY_ZI",
        "colab_type": "text"
      },
      "source": [
        "### Compiling the ANN"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0QR_G5u7ZLSM",
        "colab_type": "text"
      },
      "source": [
        "### Training the ANN on the Training set"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "tJj5k2MxZga3",
        "colab_type": "text"
      },
      "source": [
        "## Part 4 - Making the predictions and evaluating the model"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "84QFoqGYeXHL",
        "colab_type": "text"
      },
      "source": [
        "### Predicting the result of a single observation"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "CGRo3eacgDdC",
        "colab_type": "text"
      },
      "source": [
        "**Homework**\n",
        "\n",
        "Use our ANN model to predict if the customer with the following informations will leave the bank: \n",
        "\n",
        "Geography: France\n",
        "\n",
        "Credit Score: 600\n",
        "\n",
        "Gender: Male\n",
        "\n",
        "Age: 40 years old\n",
        "\n",
        "Tenure: 3 years\n",
        "\n",
        "Balance: \\$ 60000\n",
        "\n",
        "Number of Products: 2\n",
        "\n",
        "Does this customer have a credit card? Yes\n",
        "\n",
        "Is this customer an Active Member: Yes\n",
        "\n",
        "Estimated Salary: \\$ 50000\n",
        "\n",
        "So, should we say goodbye to that customer?"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ZhU1LTgPg-kH",
        "colab_type": "text"
      },
      "source": [
        "**Solution**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2d8IoCCkeWGL",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "957f3970-e197-4c3b-a150-7f69dc567f5d"
      },
      "source": [
        "print(ann.predict(sc.transform([[1, 0, 0, 600, 1, 40, 3, 60000, 2, 1, 1, 50000]])) > 0.5)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[[False]]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "wGjx94g2n7OV",
        "colab_type": "text"
      },
      "source": [
        "Therefore, our ANN model predicts that this customer stays in the bank!\n",
        "\n",
        "**Important note 1:** Notice that the values of the features were all input in a double pair of square brackets. That's because the \"predict\" method always expects a 2D array as the format of its inputs. And putting our values into a double pair of square brackets makes the input exactly a 2D array.\n",
        "\n",
        "**Important note 2:** Notice also that the \"France\" country was not input as a string in the last column but as \"1, 0, 0\" in the first three columns. That's because of course the predict method expects the one-hot-encoded values of the state, and as we see in the first row of the matrix of features X, \"France\" was encoded as \"1, 0, 0\". And be careful to include these values in the first three columns, because the dummy variables are always created in the first columns."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "u7yx47jPZt11",
        "colab_type": "text"
      },
      "source": [
        "### Predicting the Test set results"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nIyEeQdRZwgs",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 137
        },
        "outputId": "82330ba8-9bdc-4fd1-d3cf-b6d78ee7c2a3"
      },
      "source": [
        "y_pred = ann.predict(X_test)\n",
        "y_pred = (y_pred > 0.5)\n",
        "print(np.concatenate((y_pred.reshape(len(y_pred),1), y_test.reshape(len(y_test),1)),1))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[[0 0]\n",
            " [0 1]\n",
            " [0 0]\n",
            " ...\n",
            " [0 0]\n",
            " [0 0]\n",
            " [0 0]]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "o0oyfLWoaEGw",
        "colab_type": "text"
      },
      "source": [
        "### Making the Confusion Matrix"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ci6K_r6LaF6P",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 68
        },
        "outputId": "4d854e9e-22d5-432f-f6e5-a102fe3ae0bd"
      },
      "source": [
        "from sklearn.metrics import confusion_matrix, accuracy_score\n",
        "cm = confusion_matrix(y_test, y_pred)\n",
        "print(cm)\n",
        "accuracy_score(y_test, y_pred)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[[1516   79]\n",
            " [ 200  205]]\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.8605"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 20
        }
      ]
    }
  ]
}