Hyper_Parameter_Tuning.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "source": [
        "#Objective"
      ],
      "metadata": {
        "id": "G2LprI3o2fAl"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "The Objective of this Hyperparamter tuning is to know what machine learning model is fit for Prediction Model"
      ],
      "metadata": {
        "id": "0dw9NKcW2iN_"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "#Import Libraries"
      ],
      "metadata": {
        "id": "UnFhNrZh1A-k"
      }
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "id": "Spb7hw5y0eWe"
      },
      "outputs": [],
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.preprocessing import StandardScaler\n",
        "from tensorflow.keras.models import Sequential\n",
        "from tensorflow.keras.layers import Dense, Dropout\n",
        "from tensorflow.keras.optimizers import Adam\n",
        "from tensorflow.keras.callbacks import EarlyStopping\n",
        "import matplotlib.pyplot as plt\n",
        "from tensorflow.keras.layers import Input"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#Import Datasets"
      ],
      "metadata": {
        "id": "CbEbBhun1uwq"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "sheet_url = 'https://docs.google.com/spreadsheets/d/1haVIkVa4DHsXbiCG4Yy37wbNgwK59b3H4xUeYcZyg7U/edit#gid=1499193781'\n",
        "sheet_url_trf = sheet_url.replace('/edit#gid=', '/export?format=csv&gid=')\n",
        "df = pd.read_csv(sheet_url_trf)"
      ],
      "metadata": {
        "id": "IzkEkucK1wUK"
      },
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Exploring the Dataframe"
      ],
      "metadata": {
        "id": "O10vRtjH1fPU"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Separate features and target\n",
        "X = df.drop(['RowNumber', 'CustomerId', 'Surname', 'Exited'], axis=1)\n",
        "y = df['Exited']"
      ],
      "metadata": {
        "id": "O0t1RuhP1bWK"
      },
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Perform one-hot encoding for categorical variables\n",
        "X = pd.get_dummies(X, columns=['Geography', 'Gender'])"
      ],
      "metadata": {
        "id": "OyUMqZqi25kr"
      },
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Split the data into training and testing sets\n",
        "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)"
      ],
      "metadata": {
        "id": "sedbmZu728Iw"
      },
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Scale the features\n",
        "scaler = StandardScaler()\n",
        "X_train_scaled = scaler.fit_transform(X_train)\n",
        "X_test_scaled = scaler.transform(X_test)"
      ],
      "metadata": {
        "id": "7IW9b4GM29tk"
      },
      "execution_count": 7,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Get the number of features\n",
        "n_features = X_train_scaled.shape[1]"
      ],
      "metadata": {
        "id": "XMcHJZY82_fa"
      },
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Create the model\n",
        "model = Sequential([\n",
        "    Input(shape=(n_features,)),\n",
        "    Dense(128, activation='relu'),\n",
        "    Dropout(0.3),\n",
        "    Dense(64, activation='relu'),\n",
        "    Dropout(0.3),\n",
        "    Dense(32, activation='relu'),\n",
        "    Dense(1, activation='sigmoid')\n",
        "])"
      ],
      "metadata": {
        "id": "0Id81f8h3CAQ"
      },
      "execution_count": 9,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Compile the model\n",
        "model.compile(optimizer=Adam(learning_rate=0.001),\n",
        "              loss='binary_crossentropy',\n",
        "              metrics=['accuracy'])"
      ],
      "metadata": {
        "id": "uIhAaY0l3EKC"
      },
      "execution_count": 10,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Define early stopping\n",
        "early_stopping = EarlyStopping(monitor='val_loss', patience=10, restore_best_weights=True)"
      ],
      "metadata": {
        "id": "6U9TvqIr3GMP"
      },
      "execution_count": 11,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Train the model\n",
        "history = model.fit(X_train_scaled, y_train,\n",
        "                    epochs=100,\n",
        "                    batch_size=32,\n",
        "                    validation_split=0.2,\n",
        "                    callbacks=[early_stopping],\n",
        "                    verbose=1)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "HwDE5i923HsY",
        "outputId": "ff24ab5c-724c-4a61-fce9-3f012a295ff8"
      },
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/100\n",
            "\u001b[1m200/200\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 3ms/step - accuracy: 0.7731 - loss: 0.5164 - val_accuracy: 0.8306 - val_loss: 0.4095\n",
            "Epoch 2/100\n",
            "\u001b[1m200/200\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.8278 - loss: 0.4192 - val_accuracy: 0.8500 - val_loss: 0.3764\n",
            "Epoch 3/100\n",
            "\u001b[1m200/200\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.8402 - loss: 0.3821 - val_accuracy: 0.8544 - val_loss: 0.3586\n",
            "Epoch 4/100\n",
            "\u001b[1m200/200\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.8433 - loss: 0.3805 - val_accuracy: 0.8581 - val_loss: 0.3513\n",
            "Epoch 5/100\n",
            "\u001b[1m200/200\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8472 - loss: 0.3723 - val_accuracy: 0.8594 - val_loss: 0.3521\n",
            "Epoch 6/100\n",
            "\u001b[1m200/200\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.8648 - loss: 0.3434 - val_accuracy: 0.8556 - val_loss: 0.3483\n",
            "Epoch 7/100\n",
            "\u001b[1m200/200\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.8548 - loss: 0.3598 - val_accuracy: 0.8525 - val_loss: 0.3482\n",
            "Epoch 8/100\n",
            "\u001b[1m200/200\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.8532 - loss: 0.3590 - val_accuracy: 0.8606 - val_loss: 0.3456\n",
            "Epoch 9/100\n",
            "\u001b[1m200/200\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.8592 - loss: 0.3462 - val_accuracy: 0.8581 - val_loss: 0.3406\n",
            "Epoch 10/100\n",
            "\u001b[1m200/200\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.8505 - loss: 0.3442 - val_accuracy: 0.8556 - val_loss: 0.3420\n",
            "Epoch 11/100\n",
            "\u001b[1m200/200\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.8589 - loss: 0.3419 - val_accuracy: 0.8556 - val_loss: 0.3430\n",
            "Epoch 12/100\n",
            "\u001b[1m200/200\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8596 - loss: 0.3475 - val_accuracy: 0.8569 - val_loss: 0.3416\n",
            "Epoch 13/100\n",
            "\u001b[1m200/200\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.8667 - loss: 0.3444 - val_accuracy: 0.8625 - val_loss: 0.3407\n",
            "Epoch 14/100\n",
            "\u001b[1m200/200\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - accuracy: 0.8519 - loss: 0.3491 - val_accuracy: 0.8612 - val_loss: 0.3384\n",
            "Epoch 15/100\n",
            "\u001b[1m200/200\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - accuracy: 0.8616 - loss: 0.3404 - val_accuracy: 0.8600 - val_loss: 0.3413\n",
            "Epoch 16/100\n",
            "\u001b[1m200/200\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.8625 - loss: 0.3435 - val_accuracy: 0.8575 - val_loss: 0.3418\n",
            "Epoch 17/100\n",
            "\u001b[1m200/200\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.8539 - loss: 0.3497 - val_accuracy: 0.8600 - val_loss: 0.3451\n",
            "Epoch 18/100\n",
            "\u001b[1m200/200\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.8664 - loss: 0.3310 - val_accuracy: 0.8556 - val_loss: 0.3420\n",
            "Epoch 19/100\n",
            "\u001b[1m200/200\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.8691 - loss: 0.3260 - val_accuracy: 0.8569 - val_loss: 0.3391\n",
            "Epoch 20/100\n",
            "\u001b[1m200/200\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.8685 - loss: 0.3271 - val_accuracy: 0.8606 - val_loss: 0.3401\n",
            "Epoch 21/100\n",
            "\u001b[1m200/200\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.8588 - loss: 0.3411 - val_accuracy: 0.8569 - val_loss: 0.3399\n",
            "Epoch 22/100\n",
            "\u001b[1m200/200\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.8666 - loss: 0.3194 - val_accuracy: 0.8612 - val_loss: 0.3403\n",
            "Epoch 23/100\n",
            "\u001b[1m200/200\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.8706 - loss: 0.3227 - val_accuracy: 0.8569 - val_loss: 0.3402\n",
            "Epoch 24/100\n",
            "\u001b[1m200/200\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.8632 - loss: 0.3305 - val_accuracy: 0.8587 - val_loss: 0.3380\n",
            "Epoch 25/100\n",
            "\u001b[1m200/200\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.8692 - loss: 0.3291 - val_accuracy: 0.8600 - val_loss: 0.3413\n",
            "Epoch 26/100\n",
            "\u001b[1m200/200\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8678 - loss: 0.3212 - val_accuracy: 0.8594 - val_loss: 0.3413\n",
            "Epoch 27/100\n",
            "\u001b[1m200/200\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.8618 - loss: 0.3294 - val_accuracy: 0.8587 - val_loss: 0.3398\n",
            "Epoch 28/100\n",
            "\u001b[1m200/200\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.8709 - loss: 0.3213 - val_accuracy: 0.8619 - val_loss: 0.3369\n",
            "Epoch 29/100\n",
            "\u001b[1m200/200\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.8701 - loss: 0.3260 - val_accuracy: 0.8575 - val_loss: 0.3436\n",
            "Epoch 30/100\n",
            "\u001b[1m200/200\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.8682 - loss: 0.3216 - val_accuracy: 0.8575 - val_loss: 0.3417\n",
            "Epoch 31/100\n",
            "\u001b[1m200/200\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.8691 - loss: 0.3091 - val_accuracy: 0.8562 - val_loss: 0.3449\n",
            "Epoch 32/100\n",
            "\u001b[1m200/200\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.8700 - loss: 0.3175 - val_accuracy: 0.8587 - val_loss: 0.3418\n",
            "Epoch 33/100\n",
            "\u001b[1m200/200\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - accuracy: 0.8755 - loss: 0.3204 - val_accuracy: 0.8562 - val_loss: 0.3409\n",
            "Epoch 34/100\n",
            "\u001b[1m200/200\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - accuracy: 0.8666 - loss: 0.3108 - val_accuracy: 0.8619 - val_loss: 0.3412\n",
            "Epoch 35/100\n",
            "\u001b[1m200/200\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - accuracy: 0.8628 - loss: 0.3301 - val_accuracy: 0.8612 - val_loss: 0.3404\n",
            "Epoch 36/100\n",
            "\u001b[1m200/200\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.8726 - loss: 0.3154 - val_accuracy: 0.8600 - val_loss: 0.3442\n",
            "Epoch 37/100\n",
            "\u001b[1m200/200\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8727 - loss: 0.3104 - val_accuracy: 0.8569 - val_loss: 0.3442\n",
            "Epoch 38/100\n",
            "\u001b[1m200/200\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.8704 - loss: 0.3016 - val_accuracy: 0.8556 - val_loss: 0.3437\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Plot learning curves\n",
        "plt.figure(figsize=(12, 4))\n",
        "\n",
        "plt.subplot(1, 2, 1)\n",
        "plt.plot(history.history['loss'], label='Training Loss')\n",
        "plt.plot(history.history['val_loss'], label='Validation Loss')\n",
        "plt.title('Model Loss')\n",
        "plt.xlabel('Epoch')\n",
        "plt.ylabel('Loss')\n",
        "plt.legend()\n",
        "\n",
        "plt.subplot(1, 2, 2)\n",
        "plt.plot(history.history['accuracy'], label='Training Accuracy')\n",
        "plt.plot(history.history['val_accuracy'], label='Validation Accuracy')\n",
        "plt.title('Model Accuracy')\n",
        "plt.xlabel('Epoch')\n",
        "plt.ylabel('Accuracy')\n",
        "plt.legend()\n",
        "\n",
        "plt.tight_layout()\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 407
        },
        "id": "LA_0tLSc3QFa",
        "outputId": "41aacfbb-422e-44d2-85ba-46003c0a0128"
      },
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x400 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Make predictions\n",
        "y_pred = model.predict(X_test_scaled)\n",
        "y_pred_classes = (y_pred > 0.5).astype(int)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "o-1LPI5M3Zov",
        "outputId": "1f2e9559-afd2-413b-e3b1-3124a58a41ef"
      },
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 6ms/step\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Calculate additional metrics\n",
        "from sklearn.metrics import classification_report, confusion_matrix\n",
        "\n",
        "print(\"\\nClassification Report:\")\n",
        "print(classification_report(y_test, y_pred_classes))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "cddU3jPM3bkb",
        "outputId": "1dab1257-0bd0-4b4f-ee86-332b58425f0b"
      },
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "Classification Report:\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.88      0.95      0.92      1607\n",
            "           1       0.71      0.48      0.57       393\n",
            "\n",
            "    accuracy                           0.86      2000\n",
            "   macro avg       0.80      0.72      0.74      2000\n",
            "weighted avg       0.85      0.86      0.85      2000\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#Hyper Parameter Tuning"
      ],
      "metadata": {
        "id": "G693QyHD3fPz"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from tensorflow.keras.models import Sequential\n",
        "from tensorflow.keras.layers import BatchNormalization\n",
        "from tensorflow.keras.optimizers import RMSprop\n",
        "from tensorflow.keras.regularizers import l2"
      ],
      "metadata": {
        "id": "yUVbP9ml3nco"
      },
      "execution_count": 16,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def build_model(neurons1, neurons2, neurons3, dropout_rate, learning_rate):\n",
        "    model = Sequential([\n",
        "        Input(shape=(X_train_scaled.shape[1],)),\n",
        "        Dense(neurons1, activation='relu', kernel_regularizer=l2(0.01)),\n",
        "        BatchNormalization(),\n",
        "        Dropout(dropout_rate),\n",
        "        Dense(neurons2, activation='relu', kernel_regularizer=l2(0.01)),\n",
        "        BatchNormalization(),\n",
        "        Dropout(dropout_rate),\n",
        "        Dense(neurons3, activation='relu', kernel_regularizer=l2(0.01)),\n",
        "        Dense(1, activation='sigmoid')\n",
        "    ])\n",
        "    model.compile(optimizer=RMSprop(learning_rate=learning_rate),\n",
        "                  loss='binary_crossentropy',\n",
        "                  metrics=['accuracy'])\n",
        "    return model"
      ],
      "metadata": {
        "id": "9RfdvsYL3pWr"
      },
      "execution_count": 17,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Define hyperparameters to search\n",
        "param_grid = {\n",
        "    'neurons1': [32, 64, 128],\n",
        "    'neurons2': [16, 32, 64],\n",
        "    'neurons3': [8, 16, 32],\n",
        "    'dropout_rate': [0.2, 0.3, 0.4],\n",
        "    'learning_rate': [0.0001, 0.001, 0.01],\n",
        "    'batch_size': [16, 32, 64],\n",
        "    'epochs': [50, 100, 150]\n",
        "}\n",
        "\n",
        "# Number of random combinations to try\n",
        "n_iter = 20\n",
        "\n",
        "best_score = 0\n",
        "best_params = {}\n",
        "best_model = None\n",
        "best_history = None"
      ],
      "metadata": {
        "id": "EPwzxJcl3tBK"
      },
      "execution_count": 18,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "for i in range(n_iter):\n",
        "    print(f\"Iteration {i+1}/{n_iter}\")\n",
        "\n",
        "    # Randomly sample hyperparameters\n",
        "    neurons1 = np.random.choice(param_grid['neurons1'])\n",
        "    neurons2 = np.random.choice(param_grid['neurons2'])\n",
        "    neurons3 = np.random.choice(param_grid['neurons3'])\n",
        "    dropout_rate = np.random.choice(param_grid['dropout_rate'])\n",
        "    learning_rate = np.random.choice(param_grid['learning_rate'])\n",
        "    batch_size = np.random.choice(param_grid['batch_size'])\n",
        "    epochs = np.random.choice(param_grid['epochs'])\n",
        "\n",
        "    # Build and train the model\n",
        "    model = build_model(neurons1, neurons2, neurons3, dropout_rate, learning_rate)\n",
        "    early_stopping = EarlyStopping(monitor='val_loss', patience=10, restore_best_weights=True)\n",
        "    history = model.fit(X_train_scaled, y_train,\n",
        "                        epochs=epochs,\n",
        "                        batch_size=batch_size,\n",
        "                        validation_split=0.2,\n",
        "                        callbacks=[early_stopping],\n",
        "                        verbose=0)\n",
        "\n",
        "    # Evaluate the model\n",
        "    score = model.evaluate(X_test_scaled, y_test, verbose=0)[1]  # Get accuracy\n",
        "\n",
        "    print(f\"Score: {score}\")\n",
        "\n",
        "    if score > best_score:\n",
        "        best_score = score\n",
        "        best_params = {\n",
        "            'neurons1': neurons1,\n",
        "            'neurons2': neurons2,\n",
        "            'neurons3': neurons3,\n",
        "            'dropout_rate': dropout_rate,\n",
        "            'learning_rate': learning_rate,\n",
        "            'batch_size': batch_size,\n",
        "            'epochs': epochs\n",
        "        }\n",
        "        best_model = model\n",
        "        best_history = history"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "QH1JkwZP3wfm",
        "outputId": "2fa84ac4-b49a-4c02-8c04-00bbe020dc80"
      },
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Iteration 1/20\n",
            "Score: 0.862500011920929\n",
            "Iteration 2/20\n",
            "Score: 0.8565000295639038\n",
            "Iteration 3/20\n",
            "Score: 0.8414999842643738\n",
            "Iteration 4/20\n",
            "Score: 0.8600000143051147\n",
            "Iteration 5/20\n",
            "Score: 0.8610000014305115\n",
            "Iteration 6/20\n",
            "Score: 0.8289999961853027\n",
            "Iteration 7/20\n",
            "Score: 0.8550000190734863\n",
            "Iteration 8/20\n",
            "Score: 0.847000002861023\n",
            "Iteration 9/20\n",
            "Score: 0.8615000247955322\n",
            "Iteration 10/20\n",
            "Score: 0.8475000262260437\n",
            "Iteration 11/20\n",
            "Score: 0.8539999723434448\n",
            "Iteration 12/20\n",
            "Score: 0.8634999990463257\n",
            "Iteration 13/20\n",
            "Score: 0.8600000143051147\n",
            "Iteration 14/20\n",
            "Score: 0.8629999756813049\n",
            "Iteration 15/20\n",
            "Score: 0.8644999861717224\n",
            "Iteration 16/20\n",
            "Score: 0.8429999947547913\n",
            "Iteration 17/20\n",
            "Score: 0.8610000014305115\n",
            "Iteration 18/20\n",
            "Score: 0.8654999732971191\n",
            "Iteration 19/20\n",
            "Score: 0.8529999852180481\n",
            "Iteration 20/20\n",
            "Score: 0.8215000033378601\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(\"Best parameters:\", best_params)\n",
        "print(\"Best score:\", best_score)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "a9jjNpPC7tWw",
        "outputId": "d146e12d-6121-4b83-de53-9084b76bceaa"
      },
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Best parameters: {'neurons1': 32, 'neurons2': 16, 'neurons3': 8, 'dropout_rate': 0.4, 'learning_rate': 0.001, 'batch_size': 32, 'epochs': 100}\n",
            "Best score: 0.8654999732971191\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Plot learning curves for the best model\n",
        "plt.figure(figsize=(12, 4))\n",
        "\n",
        "plt.subplot(1, 2, 1)\n",
        "plt.plot(best_history.history['loss'], label='Training Loss')\n",
        "plt.plot(best_history.history['val_loss'], label='Validation Loss')\n",
        "plt.title('Model Loss')\n",
        "plt.xlabel('Epoch')\n",
        "plt.ylabel('Loss')\n",
        "plt.legend()\n",
        "\n",
        "plt.subplot(1, 2, 2)\n",
        "plt.plot(best_history.history['accuracy'], label='Training Accuracy')\n",
        "plt.plot(best_history.history['val_accuracy'], label='Validation Accuracy')\n",
        "plt.title('Model Accuracy')\n",
        "plt.xlabel('Epoch')\n",
        "plt.ylabel('Accuracy')\n",
        "plt.legend()\n",
        "\n",
        "plt.tight_layout()\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 407
        },
        "id": "cTGM6iRn7vIJ",
        "outputId": "80279ea5-ba3f-4b98-f679-e0e7062cd3b1"
      },
      "execution_count": 21,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x400 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Make predictions with the best model\n",
        "y_pred = best_model.predict(X_test_scaled)\n",
        "y_pred_classes = (y_pred > 0.5).astype(int)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "OMNvRAG27xgR",
        "outputId": "2538af8a-2fa7-4e4f-da9b-1c71b689a78a"
      },
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 7ms/step\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Print classification report and confusion matrix\n",
        "from sklearn.metrics import classification_report, confusion_matrix\n",
        "\n",
        "print(\"\\nClassification Report:\")\n",
        "print(classification_report(y_test, y_pred_classes))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "7HPp5n4p7zKC",
        "outputId": "dd666ac9-5804-459f-a09a-513c6e9f0683"
      },
      "execution_count": 23,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "Classification Report:\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.88      0.97      0.92      1607\n",
            "           1       0.77      0.45      0.57       393\n",
            "\n",
            "    accuracy                           0.87      2000\n",
            "   macro avg       0.82      0.71      0.74      2000\n",
            "weighted avg       0.86      0.87      0.85      2000\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(\"\\nConfusion Matrix:\")\n",
        "print(confusion_matrix(y_test, y_pred_classes))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "lYHuvDEP71P0",
        "outputId": "8c7ab4af-ac68-4b72-cc37-a139fd18840b"
      },
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "Confusion Matrix:\n",
            "[[1553   54]\n",
            " [ 215  178]]\n"
          ]
        }
      ]
    }
  ]
}