SHE++_2_2022.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "authorship_tag": "ABX9TyPraQiWKeB62V30mgKKf8mc",
      "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/FatimaQuiEsc/SHEplusplus2019-22/blob/main/SHE%2B%2B_2_2022.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "xzUol-GQIbfz"
      },
      "outputs": [],
      "source": [
        "import glob as glob\n",
        "import numpy as np\n",
        "import copy\n",
        "import matplotlib.pyplot as plt\n",
        "import h5py\n",
        "import scipy\n",
        "from PIL import Image\n",
        "from scipy import ndimage\n",
        "\n",
        "%matplotlib inline"
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "eTQpoWoNIoKJ"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "wUpohzNwKS2D",
        "outputId": "a5bf9393-4baf-4e56-8d09-681def3523d9"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mounted at /content/drive\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#Los datos corresponden a un dataset cats vs dogs de https://www.kaggle.com/c/dogs-vs-cats\n",
        "files=glob.glob('/content/drive/MyDrive/SHE++ 2022/train/*')"
      ],
      "metadata": {
        "id": "xa6b-6v5Kquh"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "len(files)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "RFwB__kpLv9o",
        "outputId": "94136570-3201-429c-dc13-6b3c1b522c20"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "4340"
            ]
          },
          "metadata": {},
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#numero de pixeles para nuestras imágenes. \n",
        "num_px=64"
      ],
      "metadata": {
        "id": "Vr-PTGkFXiZY"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_x=[]\n",
        "train_y=[]\n",
        "for file in files[0:100]:\n",
        "    img=Image.open(file)\n",
        "    img=img.resize((num_px,num_px)) #se genera un arreglo de num_px de lado\n",
        "    data=np.asarray(img)\n",
        "    train_x.append(data)\n",
        "    img.close()\n",
        "    if 'dog' in file:\n",
        "        train_y.append(1)\n",
        "    if 'cat' in file: \n",
        "        train_y.append(0)\n",
        "train_y=np.asarray(train_y)\n",
        "train_x=np.asarray(train_x)\n",
        "#Se asigana un valor al arreglo"
      ],
      "metadata": {
        "id": "Zzz1TVzqVifH"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "test_x=[]\n",
        "test_y=[]\n",
        "for file in files[4300:]:\n",
        "    #print(file)\n",
        "    img=Image.open(file)\n",
        "    img=img.resize((num_px,num_px))\n",
        "    data=np.asarray(img)\n",
        "    test_x.append(data)\n",
        "    img.close()\n",
        "    if 'dog' in file:\n",
        "        test_y.append(1)\n",
        "    if 'cat' in file: \n",
        "        test_y.append(0)\n",
        "test_y=np.asarray(test_y)\n",
        "test_x=np.asarray(test_x)"
      ],
      "metadata": {
        "id": "SrRXnBlda3fP"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plt.imshow(train_x[0])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 286
        },
        "id": "PbeSlNG6Xsk8",
        "outputId": "4ace8645-e933-4d43-de63-f8411a20ff1a"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.image.AxesImage at 0x7fe2216d0e90>"
            ]
          },
          "metadata": {},
          "execution_count": 24
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "np.shape(train_x)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "TYBu-JP3ZhBM",
        "outputId": "940ccfb7-b017-4c99-864a-7d7beeca0b1b"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(100, 64, 64, 3)"
            ]
          },
          "metadata": {},
          "execution_count": 20
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.imshow(test_x[1])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 286
        },
        "id": "aS3gAuYxZ1WY",
        "outputId": "111b2f5a-2b73-45bf-e0d5-c55a52d11cd4"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.image.AxesImage at 0x7fe221648650>"
            ]
          },
          "metadata": {},
          "execution_count": 25
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Hacemos una transformación de los arreglos que contienen la imágen para que sean arreglos de una dimensión. "
      ],
      "metadata": {
        "id": "JTrOe-91RjgN"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "train_x_flatten = train_x.reshape(train_x.shape[0], -1).T\n",
        "\n",
        "test_x_flatten = test_x.reshape(test_x.shape[0], -1).T\n",
        "\n",
        "train_set_y=train_y.reshape((1,train_y.shape[0]))\n",
        "test_set_y=test_y.reshape((1,test_y.shape[0]))"
      ],
      "metadata": {
        "id": "vPiJd5HVMIWF"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_set_x= train_x_flatten/255\n",
        "test_set_x= test_x_flatten/255"
      ],
      "metadata": {
        "id": "SbT7nhYIj0Zh"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#Función de activación. Sigmpide \n",
        "def sigmoid(z):\n",
        "  s=1./(1+np.exp(-z))\n",
        "  return s"
      ],
      "metadata": {
        "id": "T8HsQ-gtZgpY"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "z=np.linspace(-10,10,10)\n",
        "plt.plot(z, sigmoid(z))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 283
        },
        "id": "QtqXlUf4bwiQ",
        "outputId": "7cea49e2-003c-4520-8333-98295af6d0fb"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[<matplotlib.lines.Line2D at 0x7fe221049350>]"
            ]
          },
          "metadata": {},
          "execution_count": 48
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#Inicializamos los pesos y los sesgos a cero. \n",
        "def initialize_with_zeros(dim):\n",
        "    w=np.zeros((dim,1))\n",
        "    b=0.\n",
        "    return w, b\n"
      ],
      "metadata": {
        "id": "ZOf1nLYjXahK"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#Calculamos nuestra función de activación, función de costo y las derivadas, dw y db\n",
        "def propagate(w, b, X, Y):\n",
        "    m = X.shape[1]\n",
        "    A = sigmoid(np.dot(w.T,X)+b)\n",
        "    cost = -(1./m)*np.sum(Y*np.log(A)+(1-Y)*np.log(1-A))\n",
        "    #Quita la estructura para dejarlo como un numero\n",
        "    dw = (1/m)* np.dot(X,(A-Y).T)\n",
        "    db = (1/m)* np.sum(A-Y)\n",
        "    cost = np.squeeze(np.array(cost))\n",
        "    #\n",
        "    grads = {\"dw\": dw,\n",
        "             \"db\": db}\n",
        "    return grads, cost"
      ],
      "metadata": {
        "id": "yoFLDo2peGi_"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#Realizamos la optimización de nuestra red. \n",
        "\n",
        "def optimize(w, b, X, Y, num_iterations=100, learning_rate=0.009, print_cost=False):\n",
        "\n",
        "    \n",
        "    w = copy.deepcopy(w)\n",
        "    b = copy.deepcopy(b)\n",
        "    \n",
        "    costs = []\n",
        "    \n",
        "    #Calculo de gradientes \n",
        "    for i in range(num_iterations):\n",
        "        grads, cost = propagate(w, b, X, Y)\n",
        "        dw = grads[\"dw\"]\n",
        "        db = grads[\"db\"]\n",
        "        w=w-learning_rate*dw\n",
        "        b=b-learning_rate*db\n",
        "        \n",
        "        #Modulo, division entera \n",
        "        if i % 100 == 0:\n",
        "            costs.append(cost)\n",
        "            if print_cost:\n",
        "                print (\"Cost after iteration %i: %f\" %(i, cost))\n",
        "    \n",
        "    params = {\"w\": w,\n",
        "              \"b\": b}\n",
        "    \n",
        "    grads = {\"dw\": dw,\n",
        "             \"db\": db}\n",
        "    \n",
        "    return params, grads, costs"
      ],
      "metadata": {
        "id": "ltnYgSQmedpa"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#Calculamos la predicción que hace nuestra función\n",
        "def predict(w, b, X):\n",
        "    m = X.shape[1]\n",
        "    Y_prediction = np.zeros((1, m))\n",
        "    w = w.reshape(X.shape[0], 1)\n",
        "    \n",
        "    A = sigmoid(np.dot(w.T,X)+b)\n",
        "     \n",
        "    for i in range(A.shape[1]):\n",
        "        if A[0,i]>0.5:\n",
        "            Y_prediction[0,i] = int(1)\n",
        "        else:\n",
        "            Y_prediction[0,i] = int(0)\n",
        "    \n",
        "    return Y_prediction"
      ],
      "metadata": {
        "id": "qNFzAKm4gWY8"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#Ponemos todo junto en nuestro modelo \n",
        "# GRADED FUNCTION: model\n",
        "\n",
        "def model(X_train, Y_train, X_test, Y_test, num_iterations=2000, learning_rate=0.5, print_cost=False):  \n",
        "    w,b=initialize_with_zeros(X_train.shape[0])\n",
        "    params, grads, costs = optimize(w, b, X_train, Y_train, num_iterations=num_iterations, learning_rate=learning_rate, print_cost=print_cost)\n",
        "    w=params[\"w\"]\n",
        "    b=params[\"b\"]\n",
        "    \n",
        "    Y_prediction_test = predict(w,b,X_test)\n",
        "    Y_prediction_train = predict(w,b,X_train)\n",
        "    if print_cost:\n",
        "        print(\"train accuracy: {} %\".format(100 - np.mean(np.abs(Y_prediction_train - Y_train)) * 100))\n",
        "        print(\"test accuracy: {} %\".format(100 - np.mean(np.abs(Y_prediction_test - Y_test)) * 100))\n",
        "        \n",
        "    d = {\"costs\": costs,\n",
        "         \"Y_prediction_test\": Y_prediction_test, \n",
        "         \"Y_prediction_train\" : Y_prediction_train, \n",
        "         \"w\" : w, \n",
        "         \"b\" : b,\n",
        "         \"learning_rate\" : learning_rate,\n",
        "         \"num_iterations\": num_iterations}\n",
        "    \n",
        "    return d"
      ],
      "metadata": {
        "id": "TkUNoDKCgqNa"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "logistic_regression_model = model(train_set_x, train_set_y, test_set_x, test_set_y, num_iterations=4000, learning_rate=0.01, print_cost=True)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "TzaH2ToJh2BO",
        "outputId": "5f6e9f39-9985-4481-ca8a-cbbb7a69482d"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Cost after iteration 0: 0.693147\n",
            "Cost after iteration 100: 1.427346\n",
            "Cost after iteration 200: 0.987371\n",
            "Cost after iteration 300: 0.160595\n",
            "Cost after iteration 400: 0.091825\n",
            "Cost after iteration 500: 0.076458\n",
            "Cost after iteration 600: 0.066472\n",
            "Cost after iteration 700: 0.059175\n",
            "Cost after iteration 800: 0.053490\n",
            "Cost after iteration 900: 0.048882\n",
            "Cost after iteration 1000: 0.045046\n",
            "Cost after iteration 1100: 0.041788\n",
            "Cost after iteration 1200: 0.038982\n",
            "Cost after iteration 1300: 0.036535\n",
            "Cost after iteration 1400: 0.034380\n",
            "Cost after iteration 1500: 0.032467\n",
            "Cost after iteration 1600: 0.030757\n",
            "Cost after iteration 1700: 0.029218\n",
            "Cost after iteration 1800: 0.027826\n",
            "Cost after iteration 1900: 0.026560\n",
            "Cost after iteration 2000: 0.025404\n",
            "Cost after iteration 2100: 0.024345\n",
            "Cost after iteration 2200: 0.023370\n",
            "Cost after iteration 2300: 0.022469\n",
            "Cost after iteration 2400: 0.021636\n",
            "Cost after iteration 2500: 0.020861\n",
            "Cost after iteration 2600: 0.020140\n",
            "Cost after iteration 2700: 0.019467\n",
            "Cost after iteration 2800: 0.018837\n",
            "Cost after iteration 2900: 0.018247\n",
            "Cost after iteration 3000: 0.017692\n",
            "Cost after iteration 3100: 0.017169\n",
            "Cost after iteration 3200: 0.016677\n",
            "Cost after iteration 3300: 0.016212\n",
            "Cost after iteration 3400: 0.015772\n",
            "Cost after iteration 3500: 0.015355\n",
            "Cost after iteration 3600: 0.014959\n",
            "Cost after iteration 3700: 0.014583\n",
            "Cost after iteration 3800: 0.014226\n",
            "Cost after iteration 3900: 0.013885\n",
            "train accuracy: 100.0 %\n",
            "test accuracy: 57.5 %\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "index= 3\n",
        "plt.imshow(test_set_x[:, index].reshape((num_px, num_px, 3)))\n",
        "print (\"y = \" + str(test_set_y[0,index]) + \", you predicted that it is a \\\"\" + str(logistic_regression_model['Y_prediction_test'][0,index]))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 286
        },
        "id": "-6GnF6P-lr73",
        "outputId": "6cf2dfa9-bfe8-4705-9b88-c8c3615c99c7"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "y = 1, you predicted that it is a \"1.0\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    }
  ]
}