my_optimizer.ipynb

{
  "cells": [
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "v0giJZ3Q0fBK"
      },
      "outputs": [],
      "source": [
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "from tensorflow.keras.models import Sequential\n",
        "from tensorflow.keras.layers import Dense\n",
        "from tensorflow.keras.optimizers import SGD, Adagrad"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "id": "KukLYt_60fBM"
      },
      "outputs": [],
      "source": [
        "# dataset\n",
        "X = np.array([[0, 0], [0, 1], [1, 0], [1, 1]])\n",
        "y = np.array([[1, 0], [0, 1], [0, 1], [1, 0]])\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "id": "khWLeWtU0fBN"
      },
      "outputs": [],
      "source": [
        "# Create, Compile and Train the model with SGD optimizer\n",
        "model_sgd = Sequential([\n",
        "    Dense(4, activation='relu', input_shape=(2,)),\n",
        "    Dense(2, activation='softmax')\n",
        "])\n",
        "\n",
        "sgd_optimizer = SGD(learning_rate=0.01, momentum=0.9)\n",
        "model_sgd.compile(optimizer=sgd_optimizer, loss='categorical_crossentropy', metrics=['accuracy'])\n",
        "\n",
        "history_sgd = model_sgd.fit(X, y, epochs=1000, batch_size=4, verbose=0)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "id": "ybaS0AzX0fBN"
      },
      "outputs": [],
      "source": [
        "# Create, Compile and Train the model with Adagrad optimizer\n",
        "model_adagrad = Sequential([\n",
        "    Dense(4, activation='relu', input_shape=(2,)),\n",
        "    Dense(2, activation='softmax')\n",
        "])\n",
        "\n",
        "adagrad_optimizer = Adagrad(learning_rate=0.01)\n",
        "model_adagrad.compile(optimizer=adagrad_optimizer, loss='categorical_crossentropy', metrics=['accuracy'])\n",
        "\n",
        "history_adagrad = model_adagrad.fit(X, y, epochs=1000, batch_size=4, verbose=0)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "tmJSpBxt0fBO",
        "outputId": "dd1391b1-c8f1-4182-8ff3-b4bc43a4a1c6"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "1/1 [==============================] - 0s 82ms/step\n",
            "1/1 [==============================] - 0s 57ms/step\n"
          ]
        }
      ],
      "source": [
        "# Predictions with SGD and Adagrad optimizer\n",
        "predictions_sgd = np.argmax(model_sgd.predict(X), axis=1)\n",
        "\n",
        "predictions_adagrad = np.argmax(model_adagrad.predict(X), axis=1)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5AZ4sF_D0fBO",
        "outputId": "20f6cfcd-f138-4f27-a21a-55af5d32122c"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "1/1 [==============================] - 0s 57ms/step\n"
          ]
        }
      ],
      "source": [
        "from tensorflow.keras.optimizers import RMSprop\n",
        "\n",
        "#  Create, Compile, Train and Predict the model with RMSprop optimizer\n",
        "tahir_model = Sequential([\n",
        "    Dense(4, activation='relu', input_shape=(2,)),\n",
        "    Dense(2, activation='softmax')\n",
        "])\n",
        "\n",
        "tahir_optimizer = RMSprop(learning_rate=0.001,) and  Adagrad(learning_rate=0.01)\n",
        "tahir_model.compile(optimizer=tahir_optimizer, loss='categorical_crossentropy', metrics=['accuracy'])\n",
        "\n",
        "tahir_history = tahir_model.fit(X, y, epochs=1000, batch_size=4, verbose=0)\n",
        "\n",
        "tahir_predictions = np.argmax(tahir_model.predict(X), axis=1)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 472
        },
        "id": "zga1cWmJ0fBP",
        "outputId": "2ad00a85-346e-4509-f0d3-973410e42856"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "#plt.figure(figsize=(10, 6))\n",
        "plt.plot(history_sgd.history['accuracy'], label='SGD')\n",
        "plt.plot(history_adagrad.history['accuracy'], label='Adagrad')\n",
        "plt.plot(tahir_history.history['accuracy'], label='Tahir Model')\n",
        "plt.title('Model Accuracy Comparison')\n",
        "plt.xlabel('Epoch')\n",
        "plt.ylabel('Accuracy')\n",
        "plt.legend()\n",
        "plt.grid()\n",
        "plt.show()"
      ]
    }
  ],
  "metadata": {
    "kernelspec": {
      "display_name": "base",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.11.5"
    },
    "colab": {
      "provenance": []
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}