OCT2017_Xception_metricas.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "OCT2017_Xception_metricas.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "authorship_tag": "ABX9TyOizqaGQ7RFngQYnjZo7E10",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/rrfsantos/Projeto-Redes-Neurais-OCT-Images/blob/main/OCT2017_Xception_metricas.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "HW4Gn3yU-nc1"
      },
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "from glob import glob\n",
        "import warnings\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "from matplotlib.pyplot import figure\n",
        "import matplotlib.ticker as mtick\n",
        "import itertools\n",
        "%matplotlib inline\n",
        "\n",
        "warnings.filterwarnings(\"ignore\")"
      ],
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qGykqJdu_cQp",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "4a3d9afd-f7fd-4bc5-ec70-d98e19c4a350"
      },
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive/')"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mounted at /content/drive/\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xcJ9TQY_JjFN"
      },
      "source": [
        "reports = glob('/content/drive/MyDrive/TCC/oct_kfolds/kfold*_classification_report.csv')\n",
        "reports =  sorted(reports)"
      ],
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GkW7gUH2Jv1b"
      },
      "source": [
        "df = pd.DataFrame()\n",
        "acc = []\n",
        "for i in range(len(reports)):\n",
        "    k = i + 1\n",
        "    d = pd.read_csv(reports[i]).T\n",
        "    new_header = d.iloc[0]\n",
        "    d = d[1:]\n",
        "    d.columns = new_header\n",
        "    acc.append(list(np.unique(d.accuracy.astype(float))))\n",
        "    d.set_axis(['precision', 'recall', 'f1-score', 'support'], axis=0, inplace=True)\n",
        "    df = df.append(d) \n",
        "df['metrica'] = df.index    "
      ],
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "e-z-nYOpToik"
      },
      "source": [
        "### Acurária para cada Kfold"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4HAfX75vMFEF"
      },
      "source": [
        "acc = list(itertools.chain.from_iterable(acc))\n",
        "kfolds = [1,2,3,4,5]\n",
        "df_accuracy = pd.DataFrame({'kfolds': kfolds,'accuracy %': acc })"
      ],
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "UA52OpcbI6ZV",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 545
        },
        "outputId": "66b75bed-3a49-4817-c183-fc8665438d0b"
      },
      "source": [
        "figure(figsize=(12, 8))\n",
        "ax = plt.axes()\n",
        "ax.plot(kfolds, acc, label='Accuracy', color='g')\n",
        "for x,y in zip(kfolds,acc):\n",
        "    label = \"{:.2%}\".format(y)\n",
        "    plt.annotate(label, (x,y), textcoords=\"offset points\", xytext=(0,10), ha='center', fontsize=12)\n",
        "\n",
        "plt.xticks(kfolds)\n",
        "plt.ylim(0.5,1)\n",
        "ax.yaxis.set_major_formatter(mtick.PercentFormatter(xmax=1, decimals=None, symbol='%', is_latex=False))\n",
        "\n",
        "plt.legend(loc='lower right', fontsize = 12)\n",
        "plt.xlabel('Kfold', fontsize = 16)\n",
        "plt.ylabel('Accuracy', fontsize = 16)\n",
        "plt.title('Accuracy\\n', fontsize = 20)\n",
        "plt.show();"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 864x576 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "F9yaJDFOVd5g"
      },
      "source": [
        "### Precisão para cada Kfold"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hFB32c1tYwjB"
      },
      "source": [
        "df_precision = df[df['metrica']=='precision']"
      ],
      "execution_count": 7,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "P0OnTeNwuToa",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 587
        },
        "outputId": "a1cce1ed-703a-4543-af0a-6d0a5bc7d50b"
      },
      "source": [
        "cnv = df_precision['CNV']\n",
        "dme = df_precision['DME']\n",
        "drusen = df_precision['DRUSEN']\n",
        "normal = df_precision['NORMAL']\n",
        "\n",
        "figure(figsize=(12, 8))\n",
        "\n",
        "ax = plt.axes()\n",
        "\n",
        "ax.plot(kfolds, cnv, label='CNV', color='g')\n",
        "for x,y in zip(kfolds,cnv):\n",
        "    label = \"{:.2%}\".format(y)\n",
        "    plt.annotate(label, (x,y), textcoords=\"offset points\", xytext=(0,10), ha='center', fontsize=12) \n",
        "\n",
        "ax.plot(kfolds, dme, label='DME', color='r')\n",
        "for x,y in zip(kfolds,dme):\n",
        "    label = \"{:.2%}\".format(y)\n",
        "    plt.annotate(label, (x,y), textcoords=\"offset points\", xytext=(0,10), ha='center', fontsize=12)\n",
        "\n",
        "ax.plot(kfolds, drusen, label='DRUSEN', color='b')\n",
        "for x,y in zip(kfolds,drusen):\n",
        "    label = \"{:.2%}\".format(y)\n",
        "    plt.annotate(label, (x,y), textcoords=\"offset points\", xytext=(0,10), ha='center', fontsize=12)\n",
        "\n",
        "ax.plot(kfolds, normal, label='NORMAL', color='y')\n",
        "for x,y in zip(kfolds,normal):\n",
        "    label = \"{:.2%}\".format(y)\n",
        "    plt.annotate(label, (x,y), textcoords=\"offset points\", xytext=(0,10), ha='center', fontsize=12)\n",
        "    \n",
        "plt.xticks(kfolds)\n",
        "plt.ylim(0.5,1)\n",
        "ax.yaxis.set_major_formatter(mtick.PercentFormatter(xmax=1, decimals=None, symbol='%', is_latex=False))\n",
        "\n",
        "plt.legend(loc='lower right', bbox_to_anchor=(1.15, 1), fontsize=12)\n",
        "plt.xlabel('Kfold', fontsize = 16)\n",
        "plt.ylabel('Precision', fontsize = 16)\n",
        "plt.title('Precision\\n', fontsize = 20)\n",
        "plt.show();"
      ],
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 864x576 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "wPy_BE2k-BBe"
      },
      "source": [
        "### Recall para cada Kfold"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PMD7p5HZ5UYr"
      },
      "source": [
        "df_recall = df[df['metrica']=='recall']"
      ],
      "execution_count": 9,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mtpDs_Zr5cdN",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 587
        },
        "outputId": "41915911-d188-460a-80b4-35c8fb74a04a"
      },
      "source": [
        "cnv = df_recall['CNV']\n",
        "dme = df_recall['DME']\n",
        "drusen = df_recall['DRUSEN']\n",
        "normal = df_recall['NORMAL']\n",
        "\n",
        "figure(figsize=(12, 8))\n",
        "\n",
        "ax = plt.axes()\n",
        "\n",
        "ax.plot(kfolds, cnv, label='CNV', color='g')\n",
        "for x,y in zip(kfolds,cnv):\n",
        "    label = \"{:.2%}\".format(y)\n",
        "    plt.annotate(label, (x,y), textcoords=\"offset points\", xytext=(0,10), ha='center', fontsize=12) \n",
        "\n",
        "ax.plot(kfolds, dme, label='DME', color='r')\n",
        "for x,y in zip(kfolds,dme):\n",
        "    label = \"{:.2%}\".format(y)\n",
        "    plt.annotate(label, (x,y), textcoords=\"offset points\", xytext=(0,10), ha='center', fontsize=12)\n",
        "\n",
        "ax.plot(kfolds, drusen, label='DRUSEN', color='b')\n",
        "for x,y in zip(kfolds,drusen):\n",
        "    label = \"{:.2%}\".format(y)\n",
        "    plt.annotate(label, (x,y), textcoords=\"offset points\", xytext=(0,10), ha='center', fontsize=12)\n",
        "\n",
        "ax.plot(kfolds, normal, label='NORMAL', color='y')\n",
        "for x,y in zip(kfolds,normal):\n",
        "    label = \"{:.2%}\".format(y)\n",
        "    plt.annotate(label, (x,y), textcoords=\"offset points\", xytext=(0,10), ha='center', fontsize=12)\n",
        "    \n",
        "plt.xticks(kfolds)\n",
        "plt.ylim(0.5,1)\n",
        "ax.yaxis.set_major_formatter(mtick.PercentFormatter(xmax=1, decimals=None, symbol='%', is_latex=False))\n",
        "\n",
        "plt.legend(loc='lower right', bbox_to_anchor=(1.15, 1), fontsize=12)\n",
        "plt.xlabel('Kfold', fontsize = 16)\n",
        "plt.ylabel('Recall', fontsize = 16)\n",
        "plt.title('Recall\\n', fontsize = 20)\n",
        "plt.show();"
      ],
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 864x576 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    }
  ]
}