From 73534bd85c680f56d8223efe8bdd7b137d3aa671 Mon Sep 17 00:00:00 2001 From: Pavithra Devi Murugesan Date: Thu, 7 Oct 2021 22:00:16 +0530 Subject: [PATCH] added face emotion recognition in vedio stream --- FaceEmotionRecognition/README.md | 8 + .../predict-emotion-using-webcam.ipynb | 245 +++++ FaceEmotionRecognition/train.ipynb | 934 ++++++++++++++++++ 3 files changed, 1187 insertions(+) create mode 100644 FaceEmotionRecognition/README.md create mode 100644 FaceEmotionRecognition/predict-emotion-using-webcam.ipynb create mode 100644 FaceEmotionRecognition/train.ipynb diff --git a/FaceEmotionRecognition/README.md b/FaceEmotionRecognition/README.md new file mode 100644 index 00000000..3dc3ed81 --- /dev/null +++ b/FaceEmotionRecognition/README.md @@ -0,0 +1,8 @@ +# Face Emotion Recognition using python + +- Model trainig can be found here --> train.ipynb +- You can get the trained model here - [Trained Model](https://drive.google.com/file/d/1-6O9vR9qvU18h7WEToyqMHXO2U_zleri/view?usp=sharing) +- You can take the model and try how this model work through webcam using this code ---> predict-emotion-using-webcam.ipynb + +**Dataset Used**: +- fer2013 (collected from kaggle) \ No newline at end of file diff --git a/FaceEmotionRecognition/predict-emotion-using-webcam.ipynb b/FaceEmotionRecognition/predict-emotion-using-webcam.ipynb new file mode 100644 index 00000000..4a7aa83b --- /dev/null +++ b/FaceEmotionRecognition/predict-emotion-using-webcam.ipynb @@ -0,0 +1,245 @@ +{ + "cells": [ + { + "cell_type": "code", + "execution_count": 5, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Collecting opencv-python\n", + " Downloading https://files.pythonhosted.org/packages/70/a8/e52a82936be6d5696fb06c78450707c26dc13df91bb6bf49583bb9abbaa0/opencv_python-4.5.1.48-cp37-cp37m-win_amd64.whl (34.9MB)\n", + "Requirement already satisfied: numpy>=1.14.5 in d:\\anaconda\\installed_files\\lib\\site-packages (from opencv-python) (1.16.5)\n", + "Installing collected packages: opencv-python\n", + "Successfully installed opencv-python-4.5.1.48\n" + ] + } + ], + "source": [ + "! pip install opencv-python" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "metadata": {}, + "outputs": [], + "source": [ + "import pandas as pd\n", + "import numpy as np\n", + "import matplotlib.pyplot as plt\n", + "import cv2\n", + "#tensorflow packages\n", + "from tensorflow.keras.models import load_model\n", + "from tensorflow.keras.preprocessing import image" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "metadata": {}, + "outputs": [], + "source": [ + "# Face Emotion Recognition\n", + "#Here i am using my trained model, that is trained and saved as a h5 file\n", + "faceDetection_model = 'D:\\pavi\\DeepLearningProjects\\Face_Emosion_Recognition\\pretrained_model\\Face_Detection_TrainedModel\\haarcascade_frontalface_default.xml'" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Model: \"sequential_1\"\n", + "_________________________________________________________________\n", + "Layer (type) Output Shape Param # \n", + "=================================================================\n", + "conv2d_6 (Conv2D) (None, 48, 48, 64) 1664 \n", + "_________________________________________________________________\n", + "conv2d_7 (Conv2D) (None, 48, 48, 64) 102464 \n", + "_________________________________________________________________\n", + "batch_normalization_3 (Batch (None, 48, 48, 64) 256 \n", + "_________________________________________________________________\n", + "max_pooling2d_3 (MaxPooling2 (None, 24, 24, 64) 0 \n", + "_________________________________________________________________\n", + "conv2d_8 (Conv2D) (None, 24, 24, 128) 73856 \n", + "_________________________________________________________________\n", + "conv2d_9 (Conv2D) (None, 24, 24, 128) 147584 \n", + "_________________________________________________________________\n", + "batch_normalization_4 (Batch (None, 24, 24, 128) 512 \n", + "_________________________________________________________________\n", + "max_pooling2d_4 (MaxPooling2 (None, 12, 12, 128) 0 \n", + "_________________________________________________________________\n", + "conv2d_10 (Conv2D) (None, 12, 12, 256) 295168 \n", + "_________________________________________________________________\n", + "conv2d_11 (Conv2D) (None, 12, 12, 256) 590080 \n", + "_________________________________________________________________\n", + "batch_normalization_5 (Batch (None, 12, 12, 256) 1024 \n", + "_________________________________________________________________\n", + "max_pooling2d_5 (MaxPooling2 (None, 6, 6, 256) 0 \n", + "_________________________________________________________________\n", + "flatten_1 (Flatten) (None, 9216) 0 \n", + "_________________________________________________________________\n", + "dense_3 (Dense) (None, 1024) 9438208 \n", + "_________________________________________________________________\n", + "dropout_2 (Dropout) (None, 1024) 0 \n", + "_________________________________________________________________\n", + "dense_4 (Dense) (None, 1024) 1049600 \n", + "_________________________________________________________________\n", + "dropout_3 (Dropout) (None, 1024) 0 \n", + "_________________________________________________________________\n", + "dense_5 (Dense) (None, 7) 7175 \n", + "=================================================================\n", + "Total params: 11,707,591\n", + "Trainable params: 11,706,695\n", + "Non-trainable params: 896\n", + "_________________________________________________________________\n" + ] + } + ], + "source": [ + "Emotion_Detction_model = 'D:\\pavi\\DeepLearningProjects\\Face_Emosion_Recognition\\pretrained_model\\Face_Emotion_model\\FER_vggnet.h5'\n", + "vggnet = load_model(Emotion_Detction_model)\n", + "vggnet.summary()" + ] + }, + { + "cell_type": "code", + "execution_count": 13, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "prediction ['Sad']\n", + "Sad\n" + ] + }, + { + "ename": "TypeError", + "evalue": "only size-1 arrays can be converted to Python scalars", + "output_type": "error", + "traceback": [ + "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m", + "\u001b[1;31mTypeError\u001b[0m Traceback (most recent call last)", + "\u001b[1;32m\u001b[0m in \u001b[0;36m\u001b[1;34m\u001b[0m\n\u001b[0;32m 42\u001b[0m \u001b[0mfontScale\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;36m0.6\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m 43\u001b[0m \u001b[0mthickness\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;36m3\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 44\u001b[1;33m \u001b[0mcv2\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mputText\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mframe\u001b[0m \u001b[1;33m,\u001b[0m \u001b[0mlabel\u001b[0m \u001b[1;33m,\u001b[0m \u001b[0mlabel_position\u001b[0m \u001b[1;33m,\u001b[0m \u001b[0mcv2\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mFONT_HERSHEY_SIMPLEX\u001b[0m \u001b[1;33m,\u001b[0m \u001b[0mfontScale\u001b[0m \u001b[1;33m,\u001b[0m \u001b[1;33m(\u001b[0m\u001b[1;36m0\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;36m255\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;36m0\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;33m,\u001b[0m \u001b[0mthickness\u001b[0m \u001b[1;33m,\u001b[0m \u001b[0mcv2\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mLINE_AA\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m 45\u001b[0m \u001b[1;32melse\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m 46\u001b[0m \u001b[0mcv2\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mputText\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mframe\u001b[0m \u001b[1;33m,\u001b[0m \u001b[1;34m'No Face Detection'\u001b[0m \u001b[1;33m,\u001b[0m \u001b[0mlabel_position\u001b[0m \u001b[1;33m,\u001b[0m \u001b[0mcv2\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mFONT_HERSHEY_SIMPLEX\u001b[0m \u001b[1;33m,\u001b[0m \u001b[1;36m0.6\u001b[0m \u001b[1;33m,\u001b[0m \u001b[1;33m(\u001b[0m\u001b[1;36m0\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;36m255\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;36m0\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;33m,\u001b[0m \u001b[1;36m3\u001b[0m \u001b[1;33m,\u001b[0m\u001b[0mcv2\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mLINE_AA\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n", + "\u001b[1;31mTypeError\u001b[0m: only size-1 arrays can be converted to Python scalars" + ] + } + ], + "source": [ + "#defining the emotion classes for classification\n", + "classes = np.array((\"Angry\", \"Disgust\", \"Fear\", \"Happy\", \"Sad\", \"Surprise\", \"Neutral\"))\n", + "\n", + "#video capturing and classifing\n", + "\n", + "faceCascade = cv2.CascadeClassifier(faceDetection_model)\n", + "video_capture = cv2.VideoCapture(0)\n", + "\n", + "while True:\n", + " ret,frame = video_capture.read()\n", + " \n", + " cv2.imshow('Original Video' , frame)\n", + " \n", + " gray = cv2.cvtColor(frame , cv2.COLOR_BGR2GRAY)\n", + " \n", + " face = faceCascade.detectMultiScale(gray ,scaleFactor=1.1 , minNeighbors=5,)\n", + " \n", + " #draw rectangle around the face and cut the face only\n", + " for (x,y,w,h) in face:\n", + " \n", + " cv2.rectangle( frame , (x,y) , (x+w , y+h) , (0,255,255) , 2)\n", + " face_img = gray[ y:(y+h) , x:(x+w)]\n", + " x = cv2.resize(face_img, (48,48) , interpolation = cv2.INTER_AREA)\n", + " \n", + " if np.sum([x])!=0:\n", + " #preprocessing\n", + " x = x.astype('float')/255.0 \n", + " x = image.img_to_array(x)\n", + " x = np.expand_dims(x , axis = 0)\n", + " \n", + " \n", + " #face_img = face_img.reshape(48,48)\n", + " \n", + " # prediction\n", + " p = vggnet.predict(x)\n", + " a = np.argmax(p,axis=1)\n", + " print('prediction',classes[a])\n", + " label = str(classes[a][0])\n", + " print(label)\n", + " label_position = (x-10,y-10)\n", + " \n", + " fontScale = 0.6\n", + " thickness = 3\n", + " cv2.putText(frame , label , label_position , cv2.FONT_HERSHEY_SIMPLEX , fontScale , (0,255,0) , thickness , cv2.LINE_AA)\n", + " else:\n", + " cv2.putText(frame , 'No Face Detection' , label_position , cv2.FONT_HERSHEY_SIMPLEX , 0.6 , (0,255,0) , 3 ,cv2.LINE_AA)\n", + " \n", + " #cv2.imshow('croped image' , face_img)\n", + " #display the resulting frame \n", + " \n", + " cv2.imshow('Face Detected Video' , frame)\n", + " \n", + " #break the capturing\n", + " if cv2.waitKey(1) & 0xFF == ord('q'):\n", + " break\n", + " \n", + "video_capture.release()\n", + "cv2.destroyAllWindows()\n" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "'sad'" + ] + }, + "execution_count": 1, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "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": "3.7.4" + } + }, + "nbformat": 4, + "nbformat_minor": 2 +} diff --git a/FaceEmotionRecognition/train.ipynb b/FaceEmotionRecognition/train.ipynb new file mode 100644 index 00000000..e8971893 --- /dev/null +++ b/FaceEmotionRecognition/train.ipynb @@ -0,0 +1,934 @@ +{ + "nbformat": 4, + "nbformat_minor": 0, + "metadata": { + "colab": { + "name": "FER_VGG", + "provenance": [], + "collapsed_sections": [], + "mount_file_id": "1h99nhsz-bcCCR3mFDHOGSxjF-yoVaw_V", + "authorship_tag": "ABX9TyNiZlTWpGjZ2ZFhGZ2PVCwF", + "include_colab_link": true + }, + "kernelspec": { + "name": "python3", + "display_name": "Python 3" + }, + "accelerator": "GPU" + }, + "cells": [ + { + "cell_type": "markdown", + "metadata": { + "id": "view-in-github", + "colab_type": "text" + }, + "source": [ + "\"Open" + ] + }, + { + "cell_type": "code", + "metadata": { + "id": "jqyF8qGKCiMw" + }, + "source": [ + "#import the libraries\r\n", + "import pandas as pd\r\n", + "import numpy as np\r\n", + "import matplotlib.pyplot as plt\r\n", + "import seaborn as sns\r\n", + "import itertools\r\n", + "import cv2\r\n", + "from PIL import Image\r\n", + "#keras \r\n", + "from keras.utils.np_utils import to_categorical\r\n", + "from keras.callbacks import EarlyStopping,ReduceLROnPlateau\r\n", + "from keras.models import load_model\r\n", + "from keras.preprocessing import image\r\n", + "from sklearn.metrics import confusion_matrix\r\n", + "\r\n", + "#keras layers\r\n", + "from keras.models import Sequential \r\n", + "from keras.layers import Conv2D,MaxPooling2D,BatchNormalization,AveragePooling2D\r\n", + "from keras.layers import Flatten,Dropout,Dense" + ], + "execution_count": 27, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 195 + }, + "id": "OxACXptuC1Cg", + "outputId": "bb24c466-44d4-471f-d708-8f7c7c5236e8" + }, + "source": [ + "#read the dataset\r\n", + "data = pd.read_csv('/content/drive/MyDrive/Colab Notebooks/dataFiles/Face_emotion_detection/fer2013.csv')\r\n", + "\r\n", + "print(\"the classs labels\",data['emotion'].unique()) #7 classes\r\n", + "print(\"The shape of the dataset\",data.shape)\r\n", + "\r\n", + "data.head(4)" + ], + "execution_count": 28, + "outputs": [ + { + "output_type": "stream", + "text": [ + "the classs labels [0 2 4 6 3 5 1]\n", + "The shape of the dataset (35887, 3)\n" + ], + "name": "stdout" + }, + { + "output_type": "execute_result", + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
emotionpixelsUsage
0070 80 82 72 58 58 60 63 54 58 60 48 89 115 121...Training
10151 150 147 155 148 133 111 140 170 174 182 15...Training
22231 212 156 164 174 138 161 173 182 200 106 38...Training
3424 32 36 30 32 23 19 20 30 41 21 22 32 34 21 1...Training
\n", + "
" + ], + "text/plain": [ + " emotion pixels Usage\n", + "0 0 70 80 82 72 58 58 60 63 54 58 60 48 89 115 121... Training\n", + "1 0 151 150 147 155 148 133 111 140 170 174 182 15... Training\n", + "2 2 231 212 156 164 174 138 161 173 182 200 106 38... Training\n", + "3 4 24 32 36 30 32 23 19 20 30 41 21 22 32 34 21 1... Training" + ] + }, + "metadata": { + "tags": [] + }, + "execution_count": 28 + } + ] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "B8J5NdMKEszy", + "outputId": "b2af4d14-3c9b-45de-974e-015cfc0afaec" + }, + "source": [ + "data['Usage'].value_counts()" + ], + "execution_count": 29, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "Training 28709\n", + "PublicTest 3589\n", + "PrivateTest 3589\n", + "Name: Usage, dtype: int64" + ] + }, + "metadata": { + "tags": [] + }, + "execution_count": 29 + } + ] + }, + { + "cell_type": "code", + "metadata": { + "id": "S-PpUQcKDOYa" + }, + "source": [ + "#pixcels are in str formate so need to read and convert it to numpy array\r\n", + "\r\n", + "def generate_dataset(data):\r\n", + " X_train,y_train,X_test,y_test,X_val,y_val = [],[],[],[],[],[]\r\n", + "\r\n", + " print(\"Collecting all data...................\")\r\n", + " for i in range(data.shape[0]):\r\n", + " d = data.iloc[i,:]\r\n", + " value = d['pixels'].split(' ')\r\n", + " if (d['Usage'] == 'Training'):\r\n", + " X_train.append(np.array(value,'float32'))\r\n", + " y_train.append(d['emotion'])\r\n", + " elif d['Usage'] =='PrivateTest':\r\n", + " X_val.append(np.array(value,'float32'))\r\n", + " y_val.append(d['emotion'])\r\n", + " else:\r\n", + " X_test.append(np.array(value,'float32'))\r\n", + " y_test.append(d['emotion'])\r\n", + "\r\n", + "\r\n", + " print(\"Converting to numpy array>>>>>>>>>>>>>>>>>>\")\r\n", + " #convert list to numpy array\r\n", + " X_train = np.array(X_train,'float32') \r\n", + " y_train = np.array(y_train,'float32') \r\n", + " X_test = np.array(X_test,'float32') \r\n", + " y_test = np.array(y_test,'float32')\r\n", + " X_val = np.array(X_val,'float32') \r\n", + " y_val = np.array(y_val,'float32')\r\n", + "\r\n", + " print(\"Normalizing the data>>>>>>>>>>>>>>>>>>>>>>\")\r\n", + " #normalize the data\r\n", + " X_train = X_train/255.0\r\n", + " X_test = X_test/255.0\r\n", + " X_val = X_val/255.0\r\n", + "\r\n", + " print(\"Converting target to one hot encoded values>>>>>>>>>>>>>>>>>>>>>>\")\r\n", + " #convert to numerical values to 0,1\r\n", + " y_train = to_categorical(y_train,num_classes=7)\r\n", + " y_test = to_categorical(y_test,num_classes=7)\r\n", + " y_val = to_categorical(y_val,num_classes=7)\r\n", + "\r\n", + " print(\"reshaping the data>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\")\r\n", + " #reshape the train X data to 48 x 48 picxels\r\n", + " #the len of the given picxels is 2304 ------>(48*48)\r\n", + " X_train = X_train.reshape(X_train.shape[0] , 48 ,48 , 1)\r\n", + " X_test = X_test.reshape(X_test.shape[0] , 48 ,48 , 1)\r\n", + " X_val = X_val.reshape(X_val.shape[0] , 48 ,48 , 1)\r\n", + "\r\n", + " print(\"Preprocessing completed!!!!!!!!!! stay happy :)\")\r\n", + " return X_train,y_train,X_test,y_test,X_val,y_val" + ], + "execution_count": 30, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "0pH1i-FWFf9g", + "outputId": "46e7869b-6def-4f91-ff8a-21c9b078db51" + }, + "source": [ + "X_train,y_train,X_test,y_test,X_val,y_val = generate_dataset(data)" + ], + "execution_count": 31, + "outputs": [ + { + "output_type": "stream", + "text": [ + "Collecting all data...................\n", + "Converting to numpy array>>>>>>>>>>>>>>>>>>\n", + "Normalizing the data>>>>>>>>>>>>>>>>>>>>>>\n", + "Converting target to one hot encoded values>>>>>>>>>>>>>>>>>>>>>>\n", + "reshaping the data>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\n", + "Preprocessing completed!!!!!!!!!! stay happy :)\n" + ], + "name": "stdout" + } + ] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "tMa9PBjXFwxv", + "outputId": "005c86ad-8cb1-4a9a-ff2f-a8018b82136f" + }, + "source": [ + "print(\"The size of the train data-------------------->\",X_train.shape)\r\n", + "print(\"The size of the train target data------------->\",y_train.shape)\r\n", + "print()\r\n", + "print(\"The size of the test data--------------------->\",X_test.shape)\r\n", + "print(\"The size of the test target data-------------->\",y_test.shape)\r\n", + "print()\r\n", + "print(\"The size of the validation data--------------->\",X_val.shape)\r\n", + "print(\"The size of the validation target data-------->\",y_val.shape)" + ], + "execution_count": 32, + "outputs": [ + { + "output_type": "stream", + "text": [ + "The size of the train data--------------------> (28709, 48, 48, 1)\n", + "The size of the train target data-------------> (28709, 7)\n", + "\n", + "The size of the test data---------------------> (3589, 48, 48, 1)\n", + "The size of the test target data--------------> (3589, 7)\n", + "\n", + "The size of the validation data---------------> (3589, 48, 48, 1)\n", + "The size of the validation target data--------> (3589, 7)\n" + ], + "name": "stdout" + } + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "8Q1Ja_M93dt4" + }, + "source": [ + "# plot functions" + ] + }, + { + "cell_type": "code", + "metadata": { + "id": "FvHpBaWQ3cxL" + }, + "source": [ + "def plot_confueion_matrix(y_test , y_pred , title = \"Confusion Matrix\"):\r\n", + " classes = np.array((\"Angry\", \"Disgust\", \"Fear\", \"Happy\", \"Sad\", \"Surprise\", \"Neutral\"))\r\n", + "\r\n", + " cmap = plt.cm.Blues\r\n", + " cm = confusion_matrix(y_test , y_pred)\r\n", + " \r\n", + " #plot the cm\r\n", + " plt.figure(figsize=(7,7))\r\n", + " plt.imshow(cm , interpolation='nearest' , cmap= cmap)\r\n", + " plt.colorbar()\r\n", + " thresh = cm.min() + (cm.max() - cm.min()) / 2.\r\n", + " for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):\r\n", + " plt.text(j,i, cm[i, j],\r\n", + " horizontalalignment=\"center\",\r\n", + " color=\"white\" if cm[i, j] > thresh else \"black\")\r\n", + "\r\n", + " tick_marks = np.arange(len(classes))\r\n", + " plt.xticks(tick_marks, classes, rotation=45,fontsize=10)\r\n", + " plt.yticks(tick_marks, classes,fontsize = 10)\r\n", + "\r\n", + " plt.title(title)\r\n", + " plt.xlabel('Predicted value')\r\n", + " plt.ylabel('True values')\r\n", + "\r\n", + " plt.tight_layout()\r\n", + " plt.show()\r\n", + " #plt.save('path.ghb.png')\r\n", + "\r\n", + "\r\n", + "def plot_accuracy_loss_graph(result):\r\n", + " plt.figure(figsize=(20,8))\r\n", + " plt.subplot(1,2,1)\r\n", + "\r\n", + " plt.plot(result.history['accuracy'])\r\n", + " plt.plot(result.history['val_accuracy'])\r\n", + " plt.title('VGG Model Accuracy')\r\n", + " plt.ylabel('Accuracy')\r\n", + " plt.xlabel('Epoch')\r\n", + " plt.legend(['Train', 'Test'], loc='upper left')\r\n", + " \r\n", + " #plt.savefig('ResNet Model Loss.png')\r\n", + "\r\n", + " plt.subplot(1,2,2)\r\n", + "\r\n", + " plt.plot(result.history['loss'])\r\n", + " plt.plot(result.history['val_loss'])\r\n", + " plt.title('VGG Model Loss')\r\n", + " plt.ylabel('Loss')\r\n", + " plt.xlabel('Epoch')\r\n", + " plt.legend(['Train', 'Test'], loc='upper left')\r\n", + " plt.show()\r\n", + " #plt.savefig('ResNet Model Loss.png')\r\n" + ], + "execution_count": 33, + "outputs": [] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "DVNnDNrzyERQ" + }, + "source": [ + "# building the model--->ALEXNET Architechture" + ] + }, + { + "cell_type": "code", + "metadata": { + "id": "0HJwNf_jGMdH" + }, + "source": [ + "#building the model\r\n", + "def VGGnet(input_shape , num_classes , ):\r\n", + " model = Sequential()\r\n", + " model.add(Conv2D(64, (5, 5), activation='relu', padding='same', input_shape=input_shape)) \r\n", + " model.add(Conv2D(64, (5, 5), activation='relu', padding='same', ))\r\n", + " model.add(BatchNormalization())\r\n", + " model.add(MaxPooling2D(pool_size=(2,2)))\r\n", + "\r\n", + " #model.add(Conv2D(64, (5, 5), activation='relu', padding='same', ))\r\n", + " #model.add(BatchNormalization())\r\n", + " #model.add(MaxPooling2D(pool_size=(2,2)))\r\n", + "\r\n", + "\r\n", + " model.add(Conv2D(128, (3, 3), activation='relu', padding='same', ))\r\n", + " model.add(Conv2D(128, (3, 3), activation='relu', padding='same',))\r\n", + " model.add(BatchNormalization())\r\n", + " model.add(MaxPooling2D(pool_size=(2,2)))\r\n", + "\r\n", + "\r\n", + " model.add(Conv2D(256, (3, 3), activation='relu', padding='same',))\r\n", + " model.add(Conv2D(256, (3, 3), activation='relu', padding='same',))\r\n", + " model.add(BatchNormalization())\r\n", + " model.add(MaxPooling2D(pool_size=(2,2)))\r\n", + "\r\n", + "\r\n", + " model.add(Flatten())\r\n", + "\r\n", + " model.add(Dense(1024, activation='relu',))\r\n", + " model.add(Dropout(0.2))\r\n", + " model.add(Dense(1024, activation='relu', ))\r\n", + " model.add(Dropout(0.2))\r\n", + " model.add(Dense(num_classes, activation='sigmoid'))\r\n", + "\r\n", + "\r\n", + " # compile model\r\n", + " \r\n", + " model.compile(optimizer='adam', loss='categorical_crossentropy', metrics=['accuracy'])\r\n", + "\r\n", + " #model summary\r\n", + " print(model.summary())\r\n", + "\r\n", + " return model" + ], + "execution_count": 34, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "F9-dSgCoz0tt", + "outputId": "de172708-d5bb-4b12-f599-c00fdc589bc3" + }, + "source": [ + "model = VGGnet(input_shape = (48,48,1), num_classes = 7)" + ], + "execution_count": 35, + "outputs": [ + { + "output_type": "stream", + "text": [ + "Model: \"sequential\"\n", + "_________________________________________________________________\n", + "Layer (type) Output Shape Param # \n", + "=================================================================\n", + "conv2d (Conv2D) (None, 48, 48, 64) 1664 \n", + "_________________________________________________________________\n", + "conv2d_1 (Conv2D) (None, 48, 48, 64) 102464 \n", + "_________________________________________________________________\n", + "batch_normalization (BatchNo (None, 48, 48, 64) 256 \n", + "_________________________________________________________________\n", + "max_pooling2d (MaxPooling2D) (None, 24, 24, 64) 0 \n", + "_________________________________________________________________\n", + "conv2d_2 (Conv2D) (None, 24, 24, 128) 73856 \n", + "_________________________________________________________________\n", + "conv2d_3 (Conv2D) (None, 24, 24, 128) 147584 \n", + "_________________________________________________________________\n", + "batch_normalization_1 (Batch (None, 24, 24, 128) 512 \n", + "_________________________________________________________________\n", + "max_pooling2d_1 (MaxPooling2 (None, 12, 12, 128) 0 \n", + "_________________________________________________________________\n", + "conv2d_4 (Conv2D) (None, 12, 12, 256) 295168 \n", + "_________________________________________________________________\n", + "conv2d_5 (Conv2D) (None, 12, 12, 256) 590080 \n", + "_________________________________________________________________\n", + "batch_normalization_2 (Batch (None, 12, 12, 256) 1024 \n", + "_________________________________________________________________\n", + "max_pooling2d_2 (MaxPooling2 (None, 6, 6, 256) 0 \n", + "_________________________________________________________________\n", + "flatten (Flatten) (None, 9216) 0 \n", + "_________________________________________________________________\n", + "dense (Dense) (None, 1024) 9438208 \n", + "_________________________________________________________________\n", + "dropout (Dropout) (None, 1024) 0 \n", + "_________________________________________________________________\n", + "dense_1 (Dense) (None, 1024) 1049600 \n", + "_________________________________________________________________\n", + "dropout_1 (Dropout) (None, 1024) 0 \n", + "_________________________________________________________________\n", + "dense_2 (Dense) (None, 7) 7175 \n", + "=================================================================\n", + "Total params: 11,707,591\n", + "Trainable params: 11,706,695\n", + "Non-trainable params: 896\n", + "_________________________________________________________________\n", + "None\n" + ], + "name": "stdout" + } + ] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "3yWZaJSmz7OL", + "outputId": "d1b5a148-113c-4dc1-b32f-ee3b9dac7229" + }, + "source": [ + "# set callbacks\r\n", + "early_stoppping = EarlyStopping(monitor = 'val_loss',\r\n", + " min_delta = 0.001,\r\n", + " patience = 10,\r\n", + " restore_best_weights=True)\r\n", + "\r\n", + "#set the global values\r\n", + "epoches = 40\r\n", + "batch_size = 64\r\n", + "\r\n", + "#fit the model\r\n", + "history = model.fit(X_train, y_train, \r\n", + " batch_size=batch_size, \r\n", + " epochs=epoches, \r\n", + " verbose=1, \r\n", + " validation_data=(X_val, y_val), \r\n", + " shuffle=True) #final accuracy --> at epoch 70 -->449/449 [==============================] - 16s 36ms/step - loss: 0.0594 - accuracy: 0.9826 - val_loss: 2.5800 - val_accuracy: 0.5938" + ], + "execution_count": 36, + "outputs": [ + { + "output_type": "stream", + "text": [ + "Epoch 1/40\n", + "449/449 [==============================] - 18s 38ms/step - loss: 2.7455 - accuracy: 0.2460 - val_loss: 1.7373 - val_accuracy: 0.2967\n", + "Epoch 2/40\n", + "449/449 [==============================] - 17s 37ms/step - loss: 1.6862 - accuracy: 0.3258 - val_loss: 1.6626 - val_accuracy: 0.3483\n", + "Epoch 3/40\n", + "449/449 [==============================] - 16s 36ms/step - loss: 1.5759 - accuracy: 0.3778 - val_loss: 1.6093 - val_accuracy: 0.3566\n", + "Epoch 4/40\n", + "449/449 [==============================] - 16s 36ms/step - loss: 1.4688 - accuracy: 0.4213 - val_loss: 1.7651 - val_accuracy: 0.2689\n", + "Epoch 5/40\n", + "449/449 [==============================] - 16s 36ms/step - loss: 1.3601 - accuracy: 0.4661 - val_loss: 1.3815 - val_accuracy: 0.4575\n", + "Epoch 6/40\n", + "449/449 [==============================] - 16s 36ms/step - loss: 1.2621 - accuracy: 0.5154 - val_loss: 1.3051 - val_accuracy: 0.5127\n", + "Epoch 7/40\n", + "449/449 [==============================] - 16s 36ms/step - loss: 1.1779 - accuracy: 0.5555 - val_loss: 1.2888 - val_accuracy: 0.5152\n", + "Epoch 8/40\n", + "449/449 [==============================] - 16s 36ms/step - loss: 1.1024 - accuracy: 0.5842 - val_loss: 1.3117 - val_accuracy: 0.5143\n", + "Epoch 9/40\n", + "449/449 [==============================] - 16s 36ms/step - loss: 1.0088 - accuracy: 0.6214 - val_loss: 1.1823 - val_accuracy: 0.5430\n", + "Epoch 10/40\n", + "449/449 [==============================] - 16s 36ms/step - loss: 0.9046 - accuracy: 0.6624 - val_loss: 1.1508 - val_accuracy: 0.5690\n", + "Epoch 11/40\n", + "449/449 [==============================] - 16s 36ms/step - loss: 0.8052 - accuracy: 0.7021 - val_loss: 1.2172 - val_accuracy: 0.5717\n", + "Epoch 12/40\n", + "449/449 [==============================] - 16s 36ms/step - loss: 0.7025 - accuracy: 0.7421 - val_loss: 1.2273 - val_accuracy: 0.5667\n", + "Epoch 13/40\n", + "449/449 [==============================] - 16s 36ms/step - loss: 0.6045 - accuracy: 0.7825 - val_loss: 1.1777 - val_accuracy: 0.5837\n", + "Epoch 14/40\n", + "449/449 [==============================] - 16s 36ms/step - loss: 0.5160 - accuracy: 0.8132 - val_loss: 1.4756 - val_accuracy: 0.5244\n", + "Epoch 15/40\n", + "449/449 [==============================] - 16s 36ms/step - loss: 0.4385 - accuracy: 0.8444 - val_loss: 1.3496 - val_accuracy: 0.5921\n", + "Epoch 16/40\n", + "449/449 [==============================] - 16s 36ms/step - loss: 0.4251 - accuracy: 0.8501 - val_loss: 1.3901 - val_accuracy: 0.5887\n", + "Epoch 17/40\n", + "449/449 [==============================] - 16s 36ms/step - loss: 0.3718 - accuracy: 0.8697 - val_loss: 1.3958 - val_accuracy: 0.5798\n", + "Epoch 18/40\n", + "449/449 [==============================] - 16s 36ms/step - loss: 0.3304 - accuracy: 0.8874 - val_loss: 1.5547 - val_accuracy: 0.5868\n", + "Epoch 19/40\n", + "449/449 [==============================] - 16s 36ms/step - loss: 0.3145 - accuracy: 0.8966 - val_loss: 1.4460 - val_accuracy: 0.5921\n", + "Epoch 20/40\n", + "449/449 [==============================] - 16s 36ms/step - loss: 0.2735 - accuracy: 0.9080 - val_loss: 1.5531 - val_accuracy: 0.5899\n", + "Epoch 21/40\n", + "449/449 [==============================] - 16s 36ms/step - loss: 0.2602 - accuracy: 0.9112 - val_loss: 1.4395 - val_accuracy: 0.5874\n", + "Epoch 22/40\n", + "449/449 [==============================] - 16s 36ms/step - loss: 0.2424 - accuracy: 0.9199 - val_loss: 1.5788 - val_accuracy: 0.5952\n", + "Epoch 23/40\n", + "449/449 [==============================] - 16s 36ms/step - loss: 0.2182 - accuracy: 0.9268 - val_loss: 1.6415 - val_accuracy: 0.5692\n", + "Epoch 24/40\n", + "449/449 [==============================] - 16s 36ms/step - loss: 0.2138 - accuracy: 0.9322 - val_loss: 1.7128 - val_accuracy: 0.5876\n", + "Epoch 25/40\n", + "449/449 [==============================] - 16s 36ms/step - loss: 0.2088 - accuracy: 0.9309 - val_loss: 1.7268 - val_accuracy: 0.6027\n", + "Epoch 26/40\n", + "449/449 [==============================] - 16s 36ms/step - loss: 0.1952 - accuracy: 0.9401 - val_loss: 1.5523 - val_accuracy: 0.5979\n", + "Epoch 27/40\n", + "449/449 [==============================] - 16s 36ms/step - loss: 0.1787 - accuracy: 0.9454 - val_loss: 1.6237 - val_accuracy: 0.5929\n", + "Epoch 28/40\n", + "449/449 [==============================] - 16s 36ms/step - loss: 0.1829 - accuracy: 0.9418 - val_loss: 1.6136 - val_accuracy: 0.5924\n", + "Epoch 29/40\n", + "449/449 [==============================] - 16s 36ms/step - loss: 0.1936 - accuracy: 0.9438 - val_loss: 1.8842 - val_accuracy: 0.5887\n", + "Epoch 30/40\n", + "449/449 [==============================] - 16s 36ms/step - loss: 0.1624 - accuracy: 0.9488 - val_loss: 1.6560 - val_accuracy: 0.5991\n", + "Epoch 31/40\n", + "449/449 [==============================] - 16s 36ms/step - loss: 0.1502 - accuracy: 0.9536 - val_loss: 1.7298 - val_accuracy: 0.6027\n", + "Epoch 32/40\n", + "449/449 [==============================] - 16s 36ms/step - loss: 0.1442 - accuracy: 0.9548 - val_loss: 1.5711 - val_accuracy: 0.5726\n", + "Epoch 33/40\n", + "449/449 [==============================] - 16s 36ms/step - loss: 0.1534 - accuracy: 0.9534 - val_loss: 1.9092 - val_accuracy: 0.5907\n", + "Epoch 34/40\n", + "449/449 [==============================] - 16s 36ms/step - loss: 0.1482 - accuracy: 0.9547 - val_loss: 1.8262 - val_accuracy: 0.5999\n", + "Epoch 35/40\n", + "449/449 [==============================] - 16s 36ms/step - loss: 0.1421 - accuracy: 0.9581 - val_loss: 1.8526 - val_accuracy: 0.5977\n", + "Epoch 36/40\n", + "449/449 [==============================] - 16s 36ms/step - loss: 0.1280 - accuracy: 0.9629 - val_loss: 1.8528 - val_accuracy: 0.5946\n", + "Epoch 37/40\n", + "449/449 [==============================] - 16s 36ms/step - loss: 0.1265 - accuracy: 0.9615 - val_loss: 1.8285 - val_accuracy: 0.5932\n", + "Epoch 38/40\n", + "449/449 [==============================] - 16s 36ms/step - loss: 0.1215 - accuracy: 0.9646 - val_loss: 1.6476 - val_accuracy: 0.5879\n", + "Epoch 39/40\n", + "449/449 [==============================] - 16s 36ms/step - loss: 0.1185 - accuracy: 0.9638 - val_loss: 1.7336 - val_accuracy: 0.5974\n", + "Epoch 40/40\n", + "449/449 [==============================] - 16s 36ms/step - loss: 0.1055 - accuracy: 0.9686 - val_loss: 1.9449 - val_accuracy: 0.6035\n" + ], + "name": "stdout" + } + ] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 35 + }, + "id": "owuVkM9B5iHi", + "outputId": "873b0dc3-15b0-4746-c3e4-cd4ac216c67f" + }, + "source": [ + "\"\"\"# save model and architecture to single file\r\n", + "model.save(\"/content/drive/MyDrive/Colab Notebooks/dataFiles/FER_vggnet.h5\")\r\n", + "print(\"Saved model to disk\")\"\"\"" + ], + "execution_count": 37, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "application/vnd.google.colaboratory.intrinsic+json": { + "type": "string" + }, + "text/plain": [ + "'# save model and architecture to single file\\nmodel.save(\"/content/drive/MyDrive/Colab Notebooks/dataFiles/FER_vggnet.h5\")\\nprint(\"Saved model to disk\")'" + ] + }, + "metadata": { + "tags": [] + }, + "execution_count": 37 + } + ] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "LXvaiLyYorxZ", + "outputId": "f66e3832-aad8-44fa-dc60-b467af2d3c3d" + }, + "source": [ + "# load the model from the disk\r\n", + "\r\n", + "fer_vggnet = load_model('/content/drive/MyDrive/Colab Notebooks/dataFiles/Face_emotion_detection/FER_vggnet.h5')\r\n", + "fer_vggnet.summary()" + ], + "execution_count": 38, + "outputs": [ + { + "output_type": "stream", + "text": [ + "Model: \"sequential_1\"\n", + "_________________________________________________________________\n", + "Layer (type) Output Shape Param # \n", + "=================================================================\n", + "conv2d_6 (Conv2D) (None, 48, 48, 64) 1664 \n", + "_________________________________________________________________\n", + "conv2d_7 (Conv2D) (None, 48, 48, 64) 102464 \n", + "_________________________________________________________________\n", + "batch_normalization_3 (Batch (None, 48, 48, 64) 256 \n", + "_________________________________________________________________\n", + "max_pooling2d_3 (MaxPooling2 (None, 24, 24, 64) 0 \n", + "_________________________________________________________________\n", + "conv2d_8 (Conv2D) (None, 24, 24, 128) 73856 \n", + "_________________________________________________________________\n", + "conv2d_9 (Conv2D) (None, 24, 24, 128) 147584 \n", + "_________________________________________________________________\n", + "batch_normalization_4 (Batch (None, 24, 24, 128) 512 \n", + "_________________________________________________________________\n", + "max_pooling2d_4 (MaxPooling2 (None, 12, 12, 128) 0 \n", + "_________________________________________________________________\n", + "conv2d_10 (Conv2D) (None, 12, 12, 256) 295168 \n", + "_________________________________________________________________\n", + "conv2d_11 (Conv2D) (None, 12, 12, 256) 590080 \n", + "_________________________________________________________________\n", + "batch_normalization_5 (Batch (None, 12, 12, 256) 1024 \n", + "_________________________________________________________________\n", + "max_pooling2d_5 (MaxPooling2 (None, 6, 6, 256) 0 \n", + "_________________________________________________________________\n", + "flatten_1 (Flatten) (None, 9216) 0 \n", + "_________________________________________________________________\n", + "dense_3 (Dense) (None, 1024) 9438208 \n", + "_________________________________________________________________\n", + "dropout_2 (Dropout) (None, 1024) 0 \n", + "_________________________________________________________________\n", + "dense_4 (Dense) (None, 1024) 1049600 \n", + "_________________________________________________________________\n", + "dropout_3 (Dropout) (None, 1024) 0 \n", + "_________________________________________________________________\n", + "dense_5 (Dense) (None, 7) 7175 \n", + "=================================================================\n", + "Total params: 11,707,591\n", + "Trainable params: 11,706,695\n", + "Non-trainable params: 896\n", + "_________________________________________________________________\n" + ], + "name": "stdout" + } + ] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 512 + }, + "id": "0GBiL_8TqI-v", + "outputId": "a6cbc580-57cd-44ac-956a-d95eb110e3b8" + }, + "source": [ + "# plot the metrics\r\n", + "\r\n", + "y_pred = fer_vggnet.predict(X_test , verbose=1)\r\n", + "y_pred = np.argmax(y_pred , axis = 1)\r\n", + "\r\n", + "#change the test value to numerical\r\n", + "y_test = np.argmax(y_test , axis = 1)\r\n", + "\r\n", + "plot_confueion_matrix(y_test = y_test , y_pred = y_pred)\r\n" + ], + "execution_count": 39, + "outputs": [ + { + "output_type": "stream", + "text": [ + "113/113 [==============================] - 1s 6ms/step\n" + ], + "name": "stdout" + }, + { + "output_type": "display_data", + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [], + "needs_background": "light" + } + } + ] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 512 + }, + "id": "7RXxq9tFza5X", + "outputId": "65ea5c80-60fc-4afa-b6ba-7634131f848e" + }, + "source": [ + "plot_accuracy_loss_graph(history)" + ], + "execution_count": 40, + "outputs": [ + { + "output_type": "display_data", + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [], + "needs_background": "light" + } + } + ] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 283 + }, + "id": "y4h4zP531WEC", + "outputId": "e7a26ed6-4439-403d-ad75-9497e3b37b25" + }, + "source": [ + "#testing with custom imgaes\r\n", + "classes = np.array((\"Angry\", \"Disgust\", \"Fear\", \"Happy\", \"Sad\", \"Surprise\", \"Neutral\"))\r\n", + "\r\n", + "img = X_test[1223]\r\n", + "img_pixels = np.expand_dims(img, axis = 0) #any image reshpe to (1,48,48,1)\r\n", + "imre = img.reshape(48,48) #reshape to show by plt (48,48)\r\n", + "plt.imshow(imre)\r\n", + "\r\n", + "p = fer_vggnet.predict(img_pixels)\r\n", + "a = np.argmax(p,axis=1)\r\n", + "classes[a]" + ], + "execution_count": 41, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "array(['Happy'], dtype='" + ] + }, + "metadata": { + "tags": [], + "needs_background": "light" + } + } + ] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 350 + }, + "id": "fxxloPNNCC-v", + "outputId": "4ada94e3-1011-4a0e-8a01-bb12a529ff21" + }, + "source": [ + "\r\n", + "img = image.load_img('/content/drive/MyDrive/Colab Notebooks/dataFiles/Face_emotion_detection/sadboy.jpg' , target_size = (48,48) , grayscale =True)\r\n", + "shoe_img = image.load_img('/content/drive/MyDrive/Colab Notebooks/dataFiles/Face_emotion_detection/sadboy.jpg' , target_size = (200,200) , grayscale =True)\r\n", + "x = image.img_to_array(img)\r\n", + "x = np.expand_dims(x , axis = 0)\r\n", + "x =x/255.0\r\n", + "print(x.shape)\r\n", + "#predict the image\r\n", + "pred = fer_vggnet.predict(x)\r\n", + "a = np.argmax( pred, axis=1)\r\n", + "#show the image\r\n", + "plt.imshow(shoe_img)\r\n", + "print()\r\n", + "print(\"The predicted class of the Image ---->\",*classes[a])" + ], + "execution_count": 42, + "outputs": [ + { + "output_type": "stream", + "text": [ + "/usr/local/lib/python3.6/dist-packages/keras_preprocessing/image/utils.py:107: UserWarning: grayscale is deprecated. Please use color_mode = \"grayscale\"\n", + " warnings.warn('grayscale is deprecated. Please use '\n" + ], + "name": "stderr" + }, + { + "output_type": "stream", + "text": [ + "(1, 48, 48, 1)\n", + "\n", + "The predicted class of the Image ----> Sad\n" + ], + "name": "stdout" + }, + { + "output_type": "display_data", + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [], + "needs_background": "light" + } + } + ] + }, + { + "cell_type": "code", + "metadata": { + "id": "_ujH8wm4mt2-" + }, + "source": [ + "" + ], + "execution_count": 43, + "outputs": [] + } + ] +} \ No newline at end of file