notebook-code/TN3_Ensemble_ferplus_randomoversampling.ipynb

{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"name":"TN3_Ensemble_ferplus_randomoversampling.ipynb","provenance":[],"collapsed_sections":[]},"kernelspec":{"name":"python3","display_name":"Python 3"},"accelerator":"GPU"},"cells":[{"cell_type":"code","metadata":{"id":"PL21Mz5-OeDg","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1610855627944,"user_tz":-420,"elapsed":598,"user":{"displayName":"Dũng Nguyễn Minh","photoUrl":"","userId":"05826558414739051995"}},"outputId":"55972c34-1727-474f-c054-629e1f93c594"},"source":["# Mount \"My Drive\" into /content/drive\r\n","from google.colab import drive\r\n","\r\n","google_drive_dir = \"Final_Project_DL\"  # @param\r\n","#bml-notebooks/\r\n","drive.mount('/content/drive')\r\n","\r\n","mount_point = \"/content/drive/MyDrive/{}\".format(google_drive_dir)\r\n","\r\n","# Change the root directory to your mount_point\r\n","%cd '$mount_point'"],"execution_count":null,"outputs":[{"output_type":"stream","text":["Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n","/content/drive/.shortcut-targets-by-id/1BOvFtzoVhfYNe70lBGxYzhPlDElSrazo/Final_Project_DL\n"],"name":"stdout"}]},{"cell_type":"markdown","metadata":{"papermill":{"duration":0.044808,"end_time":"2020-12-18T22:02:11.416442","exception":false,"start_time":"2020-12-18T22:02:11.371634","status":"completed"},"tags":[],"id":"eI1PdIKtLrpv"},"source":["# <span style=\"color: royalblue;\">Load Data</span>\n","Load the image data with labels."]},{"cell_type":"code","metadata":{"id":"HM1rN5gqd5aG"},"source":["import numpy as np\r\n","from src.loadData import loadFerPlus\r\n","from keras.utils import to_categorical\r\n"],"execution_count":null,"outputs":[]},{"cell_type":"code","metadata":{"id":"Nt9oKq3mo60M","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1610855630316,"user_tz":-420,"elapsed":518,"user":{"displayName":"Dũng Nguyễn Minh","photoUrl":"","userId":"05826558414739051995"}},"outputId":"faba1b81-2a60-43ce-9835-2fea8d252a1b"},"source":["emotions = {0: 'Neutral', 1:'Happiness', 2:'Surprise', 3: 'Sadness', 4: 'Anger', 5: 'Disgust', 6: 'Fear', 7:'Contempt'}\r\n","class_names = list(emotions.values())\r\n","class_names"],"execution_count":null,"outputs":[{"output_type":"execute_result","data":{"text/plain":["['Neutral',\n"," 'Happiness',\n"," 'Surprise',\n"," 'Sadness',\n"," 'Anger',\n"," 'Disgust',\n"," 'Fear',\n"," 'Contempt']"]},"metadata":{"tags":[]},"execution_count":8}]},{"cell_type":"code","metadata":{"id":"ct5pobN6eBgW"},"source":["X_train,X_val,X_test,y_train,y_val,y_test = loadFerPlus('fer2013/ferplus_data.csv')\r\n","\r\n","y_train_onehot = to_categorical(y_train)\r\n","y_val_onehot = to_categorical(y_val)\r\n","y_test_onehot = to_categorical(y_test)"],"execution_count":null,"outputs":[]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"icEw_PExckCo","executionInfo":{"status":"ok","timestamp":1610855639138,"user_tz":-420,"elapsed":5636,"user":{"displayName":"Dũng Nguyễn Minh","photoUrl":"","userId":"05826558414739051995"}},"outputId":"50ca7463-147b-430c-ff6c-d8d04fc038b2"},"source":["X_train.shape"],"execution_count":null,"outputs":[{"output_type":"execute_result","data":{"text/plain":["(28389, 48, 48)"]},"metadata":{"tags":[]},"execution_count":10}]},{"cell_type":"markdown","metadata":{"id":"-9KsqSFYycZW"},"source":["#Preprocessing"]},{"cell_type":"code","metadata":{"id":"lIAx61wiye5z"},"source":["def preprocessX(X):\r\n","    return np.expand_dims(X,axis=-1)\r\n","\r\n","X_processed_val = preprocessX(X_val)\r\n","X_processed_test = preprocessX(X_test)\r\n"],"execution_count":null,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"-DkOqeRecp0l"},"source":["#Training"]},{"cell_type":"code","metadata":{"id":"WfXhrNLKi0sy"},"source":["import keras\r\n","from keras.preprocessing import image\r\n","from keras import applications as keras_applications"],"execution_count":null,"outputs":[]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"zncRqtUejaMI","executionInfo":{"status":"ok","timestamp":1610855663938,"user_tz":-420,"elapsed":8321,"user":{"displayName":"Dũng Nguyễn Minh","photoUrl":"","userId":"05826558414739051995"}},"outputId":"099e4f10-94bc-421a-e988-536707c7b327"},"source":["!pip install keras_applications\r\n","!pip install git+https://github.com/rcmalli/keras-vggface.git"],"execution_count":null,"outputs":[{"output_type":"stream","text":["Collecting keras_applications\n","\u001b[?25l  Downloading https://files.pythonhosted.org/packages/71/e3/19762fdfc62877ae9102edf6342d71b28fbfd9dea3d2f96a882ce099b03f/Keras_Applications-1.0.8-py3-none-any.whl (50kB)\n","\r\u001b[K     |██████▌                         | 10kB 20.6MB/s eta 0:00:01\r\u001b[K     |█████████████                   | 20kB 25.0MB/s eta 0:00:01\r\u001b[K     |███████████████████▍            | 30kB 22.5MB/s eta 0:00:01\r\u001b[K     |█████████████████████████▉      | 40kB 21.2MB/s eta 0:00:01\r\u001b[K     |████████████████████████████████| 51kB 6.9MB/s \n","\u001b[?25hRequirement already satisfied: h5py in /usr/local/lib/python3.6/dist-packages (from keras_applications) (2.10.0)\n","Requirement already satisfied: numpy>=1.9.1 in /usr/local/lib/python3.6/dist-packages (from keras_applications) (1.19.5)\n","Requirement already satisfied: six in /usr/local/lib/python3.6/dist-packages (from h5py->keras_applications) (1.15.0)\n","Installing collected packages: keras-applications\n","Successfully installed keras-applications-1.0.8\n","Collecting git+https://github.com/rcmalli/keras-vggface.git\n","  Cloning https://github.com/rcmalli/keras-vggface.git to /tmp/pip-req-build-hmemj31l\n","  Running command git clone -q https://github.com/rcmalli/keras-vggface.git /tmp/pip-req-build-hmemj31l\n","Requirement already satisfied: numpy>=1.9.1 in /usr/local/lib/python3.6/dist-packages (from keras-vggface==0.6) (1.19.5)\n","Requirement already satisfied: scipy>=0.14 in /usr/local/lib/python3.6/dist-packages (from keras-vggface==0.6) (1.4.1)\n","Requirement already satisfied: h5py in /usr/local/lib/python3.6/dist-packages (from keras-vggface==0.6) (2.10.0)\n","Requirement already satisfied: pillow in /usr/local/lib/python3.6/dist-packages (from keras-vggface==0.6) (7.0.0)\n","Requirement already satisfied: keras in /usr/local/lib/python3.6/dist-packages (from keras-vggface==0.6) (2.4.3)\n","Requirement already satisfied: six>=1.9.0 in /usr/local/lib/python3.6/dist-packages (from keras-vggface==0.6) (1.15.0)\n","Requirement already satisfied: pyyaml in /usr/local/lib/python3.6/dist-packages (from keras-vggface==0.6) (3.13)\n","Building wheels for collected packages: keras-vggface\n","  Building wheel for keras-vggface (setup.py) ... \u001b[?25l\u001b[?25hdone\n","  Created wheel for keras-vggface: filename=keras_vggface-0.6-cp36-none-any.whl size=8310 sha256=62669ec9dcef8f2ef8d3359bfc460bcede25c3488b63722b05384e0ea0ce94af\n","  Stored in directory: /tmp/pip-ephem-wheel-cache-xncmjc_4/wheels/36/07/46/06c25ce8e9cd396dabe151ea1d8a2bc28dafcb11321c1f3a6d\n","Successfully built keras-vggface\n","Installing collected packages: keras-vggface\n","Successfully installed keras-vggface-0.6\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"cFzE7T-qjdO1"},"source":["from keras_vggface.vggface import VGGFace\r\n","from keras_vggface import utils"],"execution_count":null,"outputs":[]},{"cell_type":"code","metadata":{"id":"c2b8HuspiHVx"},"source":["from copy import deepcopy as copy\r\n","\r\n","def smooth_labels(y, smooth_factor):\r\n","    '''Convert a matrix of one-hot row-vector labels into smoothed versions.\r\n","\r\n","    # Arguments\r\n","        y: matrix of one-hot row-vector labels to be smoothed\r\n","        smooth_factor: label smoothing factor (between 0 and 1)\r\n","\r\n","    # Returns\r\n","        A matrix of smoothed labels.\r\n","    '''\r\n","    assert len(y.shape) == 2, 'input should be a batch of one-hot-encoded data'\r\n","    y2 = copy(y)\r\n","    if 0 <= smooth_factor <= 1:\r\n","        # label smoothing ref: https://www.robots.ox.ac.uk/~vgg/rg/papers/reinception.pdf\r\n","        y2 *= 1 - smooth_factor\r\n","        y2 += smooth_factor / y.shape[1]\r\n","    else:\r\n","        raise Exception(\r\n","            'Invalid label smoothing factor: ' + str(smooth_factor))\r\n","    return y2"],"execution_count":null,"outputs":[]},{"cell_type":"code","metadata":{"id":"CHi8FiFwiKuV"},"source":["from keras.utils import Sequence\r\n","from keras.utils import to_categorical\r\n","import cv2\r\n","from math import floor\r\n","\r\n","class data_sequence(Sequence):\r\n","    '''\r\n","      yield sequence of data\r\n","      features -- list of features\r\n","      labels -- list of labels\r\n","      target_channels {int} -- 1 (gray) or 3(RGB)\r\n","    '''\r\n","    def __init__(self, features, labels, batch_size=128, target_dim=(224,224), \r\n","                 n_classes=7, shuffle=True, smooth=0.0, ver=2):\r\n","        'Initialization'\r\n","        assert len(features)==len(labels), 'number of feature and labels not consistent'\r\n","        self.features = features\r\n","        self.labels = labels\r\n","        self.batch_size = batch_size\r\n","        self.target_dim = target_dim\r\n","        self.target_channels = 3\r\n","        self.n_classes = n_classes\r\n","        self.shuffle = shuffle\r\n","        self.smooth = smooth\r\n","        self.sample_count = len(labels)\r\n","        self.indexes = np.arange(self.sample_count)\r\n","        self.on_epoch_end()\r\n","        self.ver = ver\r\n","#         self.verbose = verbose\r\n","\r\n","    def __len__(self):\r\n","        'Denotes the number of batches per epoch'\r\n","        return floor(self.sample_count / self.batch_size)\r\n","\r\n","    def __gray2RGB__(self,x):\r\n","      if len(x.shape)==2:\r\n","        return np.stack((x,x,x),-1)\r\n","      else:\r\n","        assert len(x.shape)==3\r\n","        if len(x[0,0,:]) == 1:\r\n","          return np.stack((x[:,:,0],x[:,:,0],x[:,:,0]),-1)\r\n","        else:\r\n","          assert len(x[0,0,:])==self.target_channels\r\n","      return x\r\n","\r\n","\r\n","    def __getitem__(self, idx):\r\n","        'Generate one batch of data'\r\n","        # Generate indexes of the batch\r\n","        indexes = self.indexes[idx * self.batch_size : (idx + 1) * self.batch_size]\r\n","        X = np.empty((self.batch_size, *self.target_dim, self.target_channels))\r\n","        Y = np.empty((self.batch_size, self.n_classes))\r\n","        for i,ind in enumerate(indexes):\r\n","          x = self.features[ind]\r\n","          # resize image to the target size \r\n","          x = cv2.resize(x,self.target_dim,interpolation=cv2.INTER_CUBIC)\r\n","          x = self.__gray2RGB__(x)\r\n","          X[i] = utils.preprocess_input(x, version=self.ver) # or version=2 for VGGFace2 ResNet50  \r\n","          y = self.labels[ind]\r\n","          if isinstance(y,int):\r\n","            Y[i]=to_categorical(y,7)\r\n","          else:\r\n","            assert len(y)==self.n_classes\r\n","            Y[i]=y\r\n","        X = np.array(X)\r\n","        Y = np.array(Y)\r\n","        if self.smooth > 0.0:\r\n","          smooth_labels(Y, self.smooth)\r\n","        return X,Y\r\n","\r\n","    def on_epoch_end(self):\r\n","        'Updates indexes after each epoch'\r\n","        if self.shuffle == True:\r\n","            np.random.shuffle(self.indexes)"],"execution_count":null,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"SQARMLYepeAF"},"source":["#Model"]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"3CsCUETVetdt","executionInfo":{"status":"ok","timestamp":1610855680422,"user_tz":-420,"elapsed":744,"user":{"displayName":"Dũng Nguyễn Minh","photoUrl":"","userId":"05826558414739051995"}},"outputId":"cda90270-f5b9-4394-ff65-b3286f8f2b3e"},"source":["ls Final_model/"],"execution_count":null,"outputs":[{"output_type":"stream","text":["vggface1_ferplus.h5  vggface2_ferplus.h5    VGGFACE2_V1_RandomOverSampling.h5\n","VGGFACE1.h5          VGGFACE2_V1_adasyn.h5\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"WDSeaJn_cjq9"},"source":["model = keras.models.load_model('Final_model/vggface2_ferplus.h5')"],"execution_count":null,"outputs":[]},{"cell_type":"code","metadata":{"id":"kHDz2FECd3Nx"},"source":["model_vggface1 = keras.models.load_model('Final_model/vggface1_ferplus.h5')"],"execution_count":null,"outputs":[]},{"cell_type":"code","metadata":{"id":"nd4dL2rGut6n"},"source":["val_sequence  = data_sequence(X_processed_val, y_val_onehot, batch_size=19,target_dim=(224,224),n_classes=8,shuffle=False,smooth=0.0,ver=2)"],"execution_count":null,"outputs":[]},{"cell_type":"code","metadata":{"id":"Ry1_ACAIibmg"},"source":["test_sequence  = data_sequence(X_processed_test, y_test_onehot, batch_size=18,target_dim=(224,224),n_classes=8,shuffle=False,smooth=0.0,ver=2)\r\n"],"execution_count":null,"outputs":[]},{"cell_type":"code","metadata":{"id":"SOLRJR7SiWF-"},"source":["y_pred_v2 = model.predict(val_sequence)"],"execution_count":null,"outputs":[]},{"cell_type":"code","metadata":{"id":"ZVM-3AC4jCXx"},"source":["y_pred_v1 = model_vggface1.predict(val_sequence)\r\n"],"execution_count":null,"outputs":[]},{"cell_type":"code","metadata":{"id":"fL3IhNRNi13e"},"source":["from sklearn.metrics import accuracy_score\r\n"],"execution_count":null,"outputs":[]},{"cell_type":"code","metadata":{"id":"EXkpk1wVjTx3"},"source":["y_pred = (y_pred_v1+y_pred_v2)/2\r\n","y_pred = np.argmax(y_pred, axis=-1)"],"execution_count":null,"outputs":[]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"TcDdTaLOj3xk","executionInfo":{"status":"ok","timestamp":1610855831764,"user_tz":-420,"elapsed":616,"user":{"displayName":"Dũng Nguyễn Minh","photoUrl":"","userId":"05826558414739051995"}},"outputId":"5b249448-e9cd-41bf-99bf-8ebce3e59929"},"source":["print(accuracy_score(y_pred,y_val))\r\n"],"execution_count":null,"outputs":[{"output_type":"stream","text":["0.8570222347312131\n"],"name":"stdout"}]},{"cell_type":"markdown","metadata":{"id":"oUP8yxbMNfzS"},"source":["#test data"]},{"cell_type":"code","metadata":{"id":"kyeHv4hXe4_R"},"source":["y_pred_v2 = model.predict(test_sequence)"],"execution_count":null,"outputs":[]},{"cell_type":"code","metadata":{"id":"R6aaMWv3e6mn"},"source":["y_pred_v1 = model_vggface1.predict(test_sequence)"],"execution_count":null,"outputs":[]},{"cell_type":"code","metadata":{"id":"-QOoC7s0e_MN"},"source":["y_pred = (y_pred_v1+y_pred_v2)/2\n","y_pred = np.argmax(y_pred, axis=-1)"],"execution_count":null,"outputs":[]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"8-S1oqFzfD4B","executionInfo":{"status":"ok","timestamp":1610855896052,"user_tz":-420,"elapsed":642,"user":{"displayName":"Dũng Nguyễn Minh","photoUrl":"","userId":"05826558414739051995"}},"outputId":"7d457bea-657b-490f-a439-a129a2c2e58f"},"source":["print(accuracy_score(y_pred,y_test))\n"],"execution_count":null,"outputs":[{"output_type":"stream","text":["0.8432036097010717\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"R7ZwvC--fFSa"},"source":["import pandas as pd\n","import tensorflow as tf\n","\n","confusion_matrix = tf.math.confusion_matrix(labels=y_test, predictions=y_pred).numpy()\n","confusion_matrix = np.around(confusion_matrix.astype('float') / confusion_matrix.sum(axis=1)[:, np.newaxis], decimals=2)\n","\n","confusion_matrix = pd.DataFrame(confusion_matrix,\n","                                index = class_names, \n","                                columns = class_names)"],"execution_count":null,"outputs":[]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":599},"id":"ItkQl8AdfWM1","executionInfo":{"status":"ok","timestamp":1610855899649,"user_tz":-420,"elapsed":1097,"user":{"displayName":"Dũng Nguyễn Minh","photoUrl":"","userId":"05826558414739051995"}},"outputId":"15e7cb74-f107-4cc3-b461-cb6bae0c3abf"},"source":["import seaborn as sns\n","import matplotlib.pyplot as plt\n","\n","figure = plt.figure(figsize=(8, 8))\n","sns.heatmap(confusion_matrix, annot=True,cmap=plt.cm.Blues)\n","plt.tight_layout()\n","plt.ylabel('True label')\n","plt.xlabel('Predicted label')\n","plt.show()"],"execution_count":null,"outputs":[{"output_type":"display_data","data":{"image/png":"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\n","text/plain":["<Figure size 576x576 with 2 Axes>"]},"metadata":{"tags":[],"needs_background":"light"}}]},{"cell_type":"code","metadata":{"id":"q4SRwoWIfdPf"},"source":[""],"execution_count":null,"outputs":[]}]}