avaliacao_de_resultado.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "gpuType": "V28",
      "authorship_tag": "ABX9TyPiUiL9Jnwu5ln7Mdqx1OOW",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "TPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/gyselle-marques/CalculoDeMetricas-DesafioDIO/blob/main/avaliacao_de_resultado.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Importando bibliotecas necessárias\n",
        "\n",
        "import numpy as np\n",
        "import itertools\n",
        "import sklearn\n",
        "\n",
        "import matplotlib\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "# Importando módulos e funções específicas do Scikit-learn\n",
        "from sklearn.datasets import make_classification\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.metrics import confusion_matrix\n",
        "\n",
        "# Instalando o XGBoost\n",
        "!pip install xgboost\n",
        "\n",
        "# Importando o XGBoost Classifier para treinamento\n",
        "from xgboost import XGBClassifier"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "uBb_oBNLYKrZ",
        "outputId": "b20cd8ca-7996-4024-dc3e-7491fa3b1a8e"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Collecting xgboost\n",
            "  Downloading xgboost-2.1.3-py3-none-manylinux_2_28_x86_64.whl.metadata (2.1 kB)\n",
            "Requirement already satisfied: numpy in /usr/local/lib/python3.11/dist-packages (from xgboost) (1.26.4)\n",
            "Collecting nvidia-nccl-cu12 (from xgboost)\n",
            "  Downloading nvidia_nccl_cu12-2.24.3-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl.metadata (1.8 kB)\n",
            "Requirement already satisfied: scipy in /usr/local/lib/python3.11/dist-packages (from xgboost) (1.13.1)\n",
            "Downloading xgboost-2.1.3-py3-none-manylinux_2_28_x86_64.whl (153.9 MB)\n",
            "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m153.9/153.9 MB\u001b[0m \u001b[31m7.0 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hDownloading nvidia_nccl_cu12-2.24.3-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (201.3 MB)\n",
            "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m201.3/201.3 MB\u001b[0m \u001b[31m5.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hInstalling collected packages: nvidia-nccl-cu12, xgboost\n",
            "Successfully installed nvidia-nccl-cu12-2.24.3 xgboost-2.1.3\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Generando um dataset sintético para realizar a avaliação de resultado\n",
        "X, y = make_classification(\n",
        "    n_samples=10000,\n",
        "    n_features=25,\n",
        "    n_informative=10,\n",
        "    n_redundant=0,\n",
        "    n_classes=5\n",
        ")\n",
        "\n",
        "# Definindo o nome das classes\n",
        "class_names = ['class-1', 'class-2', 'class-3', 'class-4', 'class-5']\n",
        "\n",
        "# Dividindo os dados em conjuntos de treinamento (67%) e teste (33%)\n",
        "X_train, X_test, y_train, y_test = train_test_split(\n",
        "    X, y,\n",
        "    test_size=0.33  # Proporção de dados usados ​​para teste\n",
        ")\n",
        "\n",
        "# Initializando o XGBoost Classifier\n",
        "clf = XGBClassifier()\n",
        "\n",
        "# Treinamento do Classifier\n",
        "clf.fit(X_train, y_train)\n",
        "\n",
        "# Fazendo previsões sobre os dados de teste\n",
        "y_pred = clf.predict(X_test)"
      ],
      "metadata": {
        "id": "K0A__truZg5o"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Gera a matriz de confusão\n",
        "matrix = confusion_matrix(y_test, y_pred)\n",
        "\n",
        "plt.clf()\n",
        "\n",
        "plt.gca().xaxis.tick_top()\n",
        "plt.gca().xaxis.set_label_position('top')\n",
        "\n",
        "# Plota a matriz de confusão com um mapa de cores azul\n",
        "plt.imshow(matrix, interpolation='nearest', cmap=plt.cm.Blues)\n",
        "\n",
        "# Adiciona uma barra colorida para indicar a intensidade dos valores\n",
        "plt.colorbar()\n",
        "\n",
        "# Define o formato dos valores da matriz (formato integer)\n",
        "fmt = 'd'\n",
        "\n",
        "# Adiciona o número de previsões a cada célula da matriz\n",
        "thresh = matrix.max() / 2.  # Limite para decidir a cor do texto\n",
        "for i, j in itertools.product(range(matrix.shape[0]), range(matrix.shape[1])):\n",
        "    # Sobrepõe o valor em cada célula com a cor de texto apropriada\n",
        "    plt.text(j, i, format(matrix[i, j], fmt),\n",
        "             horizontalalignment=\"center\",\n",
        "             color=\"white\" if matrix[i, j] > thresh else \"black\")\n",
        "\n",
        "# Personalize marcas de escala e rótulos para os eixos x e y\n",
        "tick_marks = np.arange(len(class_names))\n",
        "plt.xticks(tick_marks, class_names, rotation=45)  # Gira os rótulos do eixo x para facilitar a leitura\n",
        "plt.yticks(tick_marks, class_names)\n",
        "\n",
        "# Ajusta o layout do gráfico para melhor espaçamento\n",
        "plt.tight_layout()\n",
        "\n",
        "# Adiciona rótulos de eixo\n",
        "plt.ylabel('True label', size=14)\n",
        "plt.xlabel('Predicted label', size=14)\n",
        "\n",
        "# Exibi a matriz de confusão\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 507
        },
        "id": "bdLvjqfXanen",
        "outputId": "958e5bf1-83af-4db8-a870-f46b874e4edd"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Obtêm os elementos da matriz de confusão para uma classe específica\n",
        "def elements(class_idx):\n",
        "    \"\"\"\n",
        "    Calculate VP, FN, FP, VN for a specific class in the existing confusion matrix.\n",
        "\n",
        "    Parameters:\n",
        "        class_idx (int): The index of the class.\n",
        "\n",
        "    Returns:\n",
        "        dict: A dictionary with VP, FN, FP, VN.\n",
        "    \"\"\"\n",
        "    # Verdadeiro Positivo (VP):\n",
        "    VP = matrix[class_idx, class_idx]\n",
        "\n",
        "    # Falso Negativo (FN):\n",
        "    FN = np.sum(matrix[class_idx, :]) - VP\n",
        "\n",
        "    # Falso Positivo (FP):\n",
        "    FP = np.sum(matrix[:, class_idx]) - VP  # Sum of the column minus VP\n",
        "\n",
        "    # Verdadeiro Negativo (VN):\n",
        "    VN = np.sum(matrix) - (VP + FN + FP)\n",
        "\n",
        "    # Retorna os elementos da matriz de confusão como um dicionário\n",
        "    return {'VP': VP, 'FN': FN, 'FP': FP, 'VN': VN}"
      ],
      "metadata": {
        "id": "pjVBI4d2c6JG"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "results_class1 = elements(0)\n",
        "results_class2 = elements(1)\n",
        "results_class3 = elements(2)\n",
        "results_class4 = elements(3)\n",
        "results_class5 = elements(4)\n",
        "\n",
        "print(results_class1)\n",
        "print(results_class2)\n",
        "print(results_class3)\n",
        "print(results_class4)\n",
        "print(results_class5)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "z2knWZjEe4YP",
        "outputId": "ae4869ca-f684-4da5-94d5-fd600114177d"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "{'VP': 544, 'FN': 82, 'FP': 120, 'VN': 2554}\n",
            "{'VP': 490, 'FN': 143, 'FP': 133, 'VN': 2534}\n",
            "{'VP': 598, 'FN': 99, 'FP': 103, 'VN': 2500}\n",
            "{'VP': 587, 'FN': 107, 'FP': 76, 'VN': 2530}\n",
            "{'VP': 555, 'FN': 95, 'FP': 94, 'VN': 2556}\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Função de acurácia: calcular a acurácia geral do modelo\n",
        "def accuracy(y_test, y_pred):\n",
        "    \"\"\"\n",
        "    Calculate the accuracy of the model.\n",
        "\n",
        "    Parameters:\n",
        "        y_test (array-like): The true labels.\n",
        "        y_pred (array-like): The predicted labels.\n",
        "\n",
        "    Returns:\n",
        "        float: The accuracy.\n",
        "    \"\"\"\n",
        "    # Previsões Corretas: Soma dos principais elementos diagonais da matriz de confusão\n",
        "    correct_predictions = np.trace(matrix)  # Verdadeiros Positivos para todas as classes combinadas\n",
        "\n",
        "    # Previsões totais: soma de todos os elementos da matriz de confusão\n",
        "    total_predictions = np.sum(matrix)  # Número total de instâncias\n",
        "\n",
        "    # Calcule a acurácia como a proporção de previsões corretas\n",
        "    acc = correct_predictions / total_predictions\n",
        "\n",
        "    return acc  # Retorne a precisão como um valor decimal\n",
        "\n",
        "# Imprime o valor bruto da acurácia\n",
        "print(accuracy(y_test, y_pred))\n",
        "\n",
        "# Imprime a acurácia como uma porcentagem\n",
        "print(f\"accuracy: {round(accuracy(y_test, y_pred)*100)}%\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3ivr0G2pfAii",
        "outputId": "537a5d2a-0cf8-4d95-890f-6235dd23e389"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0.8406060606060606\n",
            "accuracy: 84%\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Função de precisão\n",
        "def precision(class_idx):\n",
        "    \"\"\"\n",
        "    Calculate the precision for a specific class.\n",
        "\n",
        "    Parameters:\n",
        "        class_idx (int): The index of the class.\n",
        "\n",
        "    Returns:\n",
        "        float: The precision.\n",
        "    \"\"\"\n",
        "    # Obtêm os elementos VP, FP, etc. da função elements()\n",
        "    metrics = elements(class_idx)\n",
        "\n",
        "    # Acessa e extrai VP e FP do dicionário retornado da função elements()\n",
        "    VP = metrics['VP']\n",
        "    FP = metrics['FP']\n",
        "\n",
        "    pr = VP/(VP+FP)\n",
        "    return pr\n",
        "\n",
        "print(f\"Precision: {round(precision(2)*100)}%\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "gAci8_kJgNoW",
        "outputId": "ceddff60-6c87-4f59-b2b8-5d9de0160fee"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Precision: 85%\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Função de sensibilidade (recall)\n",
        "def recall(class_idx):\n",
        "    \"\"\"\n",
        "    Calculate the recall for a specific class.\n",
        "\n",
        "    Parameters:\n",
        "        class_idx (int): The index of the class.\n",
        "\n",
        "    Returns:\n",
        "        float: The recall.\n",
        "    \"\"\"\n",
        "    metrics = elements(class_idx)\n",
        "\n",
        "    VP = metrics['VP']\n",
        "    FN = metrics['FN']\n",
        "\n",
        "    # Calcula o recall\n",
        "    re = VP / (VP+FN)\n",
        "    return re\n",
        "\n",
        "print(f\"Recall: {round(recall(2)*100)}%\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "AmK8II5JgyG0",
        "outputId": "028eb3a2-938a-4c6c-90fd-e2f1ca5d6e50"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Recall: 86%\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Função de especificidade\n",
        "def specificity(class_idx):\n",
        "    \"\"\"\n",
        "    Calculate the specificity for a specific class.\n",
        "\n",
        "    Parameters:\n",
        "        class_idx (int): The index of the class.\n",
        "\n",
        "    Returns:\n",
        "        float: The specificity.\n",
        "    \"\"\"\n",
        "    metrics = elements(class_idx)\n",
        "\n",
        "    FP = metrics['FP']\n",
        "    VN = metrics['VN']\n",
        "\n",
        "    # Calcula a especificidade\n",
        "    spec = VN / (FP + VN)\n",
        "    return spec\n",
        "\n",
        "print(f\"Specificity: {round(specificity(2)*100)}%\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "mu2iO7qVhG9D",
        "outputId": "2312c4ea-719a-45f7-ef5d-48c23fbd882e"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Specificity: 96%\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Função F-Score\n",
        "# 2*(P*R)/(P+R)\n",
        "def f_score(class_idx):\n",
        "    \"\"\"\n",
        "    Calculate the F-score for a specific class.\n",
        "\n",
        "    Parameters:\n",
        "        class_idx (int): The index of the class.\n",
        "\n",
        "    Returns:\n",
        "        float: The F-score.\n",
        "    \"\"\"\n",
        "    score = 2 * ((precision(class_idx) * recall(class_idx)) / (precision(class_idx) + recall(class_idx)))\n",
        "    return score\n",
        "\n",
        "print(f\"F-Score: {round(f_score(2)*100)}%\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "xwd0Aiswhd_h",
        "outputId": "872b5714-ee01-4ba8-c056-353fd040f97f"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "F-Score: 86%\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Compara os resultados com o sklearn\n",
        "from sklearn.metrics import classification_report\n",
        "print(classification_report(y_test, y_pred, target_names=class_names))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4089dGB6hvQm",
        "outputId": "1456d04a-e736-4ce3-c79a-4c999678fb50"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "     class-1       0.82      0.87      0.84       626\n",
            "     class-2       0.79      0.77      0.78       633\n",
            "     class-3       0.85      0.86      0.86       697\n",
            "     class-4       0.89      0.85      0.87       694\n",
            "     class-5       0.86      0.85      0.85       650\n",
            "\n",
            "    accuracy                           0.84      3300\n",
            "   macro avg       0.84      0.84      0.84      3300\n",
            "weighted avg       0.84      0.84      0.84      3300\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(f\"accuracy: \\n {accuracy(y_test,y_pred):.2f}\")\n",
        "print(f\"precision: \\n class-1: {precision(0):.2f} \\n class-2: {precision(1):.2f}\\n class-3: {precision(2):.2f}\\n class-4: {precision(3):.2f}\\n class-5: {precision(4):.2f}\")\n",
        "print(f\"recall:\\n class-1: {recall(0):.2f}\\n class-2: {recall(1):.2f}\\n class-3: {recall(2):.2f}\\n class-4: {recall(3):.2f} \\n class-5: {recall(4):.2f}\")\n",
        "print(f\"specificity: \\n class-1: {specificity(0):.2f}\\n class-2: {specificity(1):.2f}\\n class-3: {specificity(2):.2f}\\n class-4: {specificity(3):.2f} \\n class-5: {specificity(4):.2f}\")\n",
        "print(f\"f_score: \\n class-1: {f_score(0):.2f}\\n class-2: {f_score(1):.2f}\\n class-3: {f_score(2):.2f}\\n class-4: {f_score(3):.2f} \\n class-5: {f_score(4):.2f}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "J7yCZ2Zgh7Mc",
        "outputId": "feeefa44-30b9-4c9a-baa7-cfdcf08ad8a9"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "accuracy: \n",
            " 0.84\n",
            "precision: \n",
            " class-1: 0.82 \n",
            " class-2: 0.79\n",
            " class-3: 0.85\n",
            " class-4: 0.89\n",
            " class-5: 0.86\n",
            "recall:\n",
            " class-1: 0.87\n",
            " class-2: 0.77\n",
            " class-3: 0.86\n",
            " class-4: 0.85 \n",
            " class-5: 0.85\n",
            "specificity: \n",
            " class-1: 0.96\n",
            " class-2: 0.95\n",
            " class-3: 0.96\n",
            " class-4: 0.97 \n",
            " class-5: 0.96\n",
            "f_score: \n",
            " class-1: 0.84\n",
            " class-2: 0.78\n",
            " class-3: 0.86\n",
            " class-4: 0.87 \n",
            " class-5: 0.85\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Calcula métricas manuais para cada classe usando funções personalizadas\n",
        "manual_data = {\n",
        "    'Precision': [precision(0), precision(1), precision(2), precision(3), precision(4)],\n",
        "    'Recall': [recall(0), recall(1), recall(2), recall(3), recall(4)],\n",
        "    'F1-Score': [f_score(0), f_score(1), f_score(2), f_score(3), f_score(4)],\n",
        "}\n",
        "\n",
        "# Importa funções necessárias do sklearn.metrics\n",
        "from sklearn.metrics import precision_score, recall_score, f1_score\n",
        "\n",
        "# Calcula métricas para cada classe usando funções integradas do sklearn\n",
        "sklearn_data = {\n",
        "    'Precision': precision_score(y_test, y_pred, average=None),\n",
        "    'Recall': recall_score(y_test, y_pred, average=None),\n",
        "    'F1-Score': f1_score(y_test, y_pred, average=None),\n",
        "}\n",
        "\n",
        "# Importa a biblioteca do pandas\n",
        "import pandas as pd\n",
        "\n",
        "# Cria um DataFrame para comparar métricas manuais e sklearn para todas as classes\n",
        "comparison_df = pd.DataFrame({\n",
        "    'Class': ['class-1', 'class-2', 'class-3', 'class-4', 'class-5'],\n",
        "    'Manual Precision': manual_data['Precision'],\n",
        "    'Sklearn Precision': sklearn_data['Precision'],\n",
        "    'Manual Recall': manual_data['Recall'],\n",
        "    'Sklearn Recall': sklearn_data['Recall'],\n",
        "    'Manual F1-Score': manual_data['F1-Score'],\n",
        "    'Sklearn F1-Score': sklearn_data['F1-Score'],\n",
        "})\n",
        "\n",
        "\n",
        "print(comparison_df)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "fyvL2UZKiCyn",
        "outputId": "34df7604-8205-432a-eb1d-9849d0cbd753"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "     Class  Manual Precision  Sklearn Precision  Manual Recall  \\\n",
            "0  class-1          0.819277           0.819277       0.869010   \n",
            "1  class-2          0.786517           0.786517       0.774092   \n",
            "2  class-3          0.853067           0.853067       0.857963   \n",
            "3  class-4          0.885370           0.885370       0.845821   \n",
            "4  class-5          0.855162           0.855162       0.853846   \n",
            "\n",
            "   Sklearn Recall  Manual F1-Score  Sklearn F1-Score  \n",
            "0        0.869010         0.843411          0.843411  \n",
            "1        0.774092         0.780255          0.780255  \n",
            "2        0.857963         0.855508          0.855508  \n",
            "3        0.845821         0.865144          0.865144  \n",
            "4        0.853846         0.854503          0.854503  \n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Configurar dados do gráfico\n",
        "classes = comparison_df['Class']\n",
        "x = np.arange(len(classes))  # Class positions\n",
        "bar_width = 0.35  # Bar width\n",
        "\n",
        "# Criar plotagem\n",
        "fig, ax = plt.subplots(3, 1, figsize=(5, 6))\n",
        "\n",
        "# Precisão\n",
        "ax[0].bar(x - bar_width / 2, comparison_df['Manual Precision'], bar_width, label='Manual Precision', color='blue')\n",
        "ax[0].bar(x + bar_width / 2, comparison_df['Sklearn Precision'], bar_width, label='Sklearn Precision', color='orange')\n",
        "ax[0].set_title('Precision')\n",
        "ax[0].set_xticks(x)\n",
        "ax[0].set_xticklabels(classes)\n",
        "ax[0].set_ylim(0, 1)\n",
        "ax[0].legend()\n",
        "\n",
        "# Sensibilidade/Recall\n",
        "ax[1].bar(x - bar_width / 2, comparison_df['Manual Recall'], bar_width, label='Manual Recall', color='green')\n",
        "ax[1].bar(x + bar_width / 2, comparison_df['Sklearn Recall'], bar_width, label='Sklearn Recall', color='red')\n",
        "ax[1].set_title('Recall')\n",
        "ax[1].set_xticks(x)\n",
        "ax[1].set_xticklabels(classes)\n",
        "ax[1].set_ylim(0, 1)\n",
        "ax[1].legend()\n",
        "\n",
        "# F1-Score\n",
        "ax[2].bar(x - bar_width / 2, comparison_df['Manual F1-Score'], bar_width, label='Manual F1-Score', color='purple')\n",
        "ax[2].bar(x + bar_width / 2, comparison_df['Sklearn F1-Score'], bar_width, label='Sklearn F1-Score', color='pink')\n",
        "ax[2].set_title('F1-Score')\n",
        "ax[2].set_xticks(x)\n",
        "ax[2].set_xticklabels(classes)\n",
        "ax[2].set_ylim(0, 1)\n",
        "ax[2].legend()\n",
        "\n",
        "plt.tight_layout()\n",
        "plt.show()\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 607
        },
        "id": "Z1dwSZ9kjZ90",
        "outputId": "eaac22a3-87c5-4d22-bf87-cebf3c69eb0a"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 500x600 with 3 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    }
  ]
}