age-and-sex-prediction.ipynb

{"cells":[{"cell_type":"code","execution_count":13,"metadata":{"_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","trusted":true},"outputs":[{"name":"stdout","output_type":"stream","text":["['UTKFace', 'crop_part1']\n"]}],"source":["import os\n","import cv2\n","import numpy as np\n","import pandas as pd\n","from PIL import Image\n","import tensorflow as tf\n","os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'\n","from tensorflow.keras import layers\n","from sklearn.model_selection import train_test_split\n","print(os.listdir(\"./data/utkface_aligned_cropped/\"))"]},{"cell_type":"markdown","metadata":{},"source":["## Load and Visualize data"]},{"cell_type":"code","execution_count":14,"metadata":{"_cell_guid":"79c7e3d0-c299-4dcb-8224-4455121ee9b0","_uuid":"d629ff2d2480ee46fbb7e2d37f6b5fab8052498a","trusted":true},"outputs":[{"name":"stdout","output_type":"stream","text":["(23708, 2, 1)\n","[['42']\n"," ['0']]\n"]}],"source":["files = os.listdir(\"./data/utkface_aligned_cropped/UTKFace\")\n","y = np.array([[[i.split('_')[0]],[i.split('_')[1]]] for i in files])\n","# y = np.array([[i.split('_')[1] for i in onlyfiles]]).T\n","print(y.shape)\n","print(y[0])"]},{"cell_type":"code","execution_count":5,"metadata":{"trusted":true},"outputs":[{"data":{"text/plain":["(23708, 32, 32, 3)"]},"execution_count":5,"metadata":{},"output_type":"execute_result"}],"source":["X_data =[]\n","for file in files:\n","    face = cv2.imread(\"./data/utkface_aligned_cropped/UTKFace/\"+file,cv2.IMREAD_COLOR)\n","    face = cv2.cvtColor(face, cv2.COLOR_BGR2RGB)\n","    face =cv2.resize(face, (32,32) )\n","    X_data.append(face)\n","X_data=np.array(X_data)\n","X_data.shape"]},{"cell_type":"code","execution_count":6,"metadata":{"trusted":true},"outputs":[{"data":{"image/png":"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","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{},"output_type":"display_data"},{"name":"stdout","output_type":"stream","text":["[['27']\n"," ['0']]\n"]}],"source":["X = np.squeeze(X_data)\n","imshow(X[1])\n","print(y[1])"]},{"cell_type":"code","execution_count":7,"metadata":{"trusted":true},"outputs":[],"source":["X_train, X_valid, y_train, y_valid = train_test_split(X, y, test_size=0.33)\n","y_train=[y_train[:,1],y_train[:,0]]\n","y_valid=[y_valid[:,1],y_valid[:,0]]"]},{"cell_type":"markdown","metadata":{},"source":["## Build Model"]},{"cell_type":"code","execution_count":2,"metadata":{"trusted":true},"outputs":[{"name":"stderr","output_type":"stream","text":["2021-11-01 18:29:23.818997: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:923] could not open file to read NUMA node: /sys/bus/pci/devices/0000:01:00.0/numa_node\n","Your kernel may have been built without NUMA support.\n","2021-11-01 18:29:23.819178: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcudart.so.11.0'; dlerror: libcudart.so.11.0: cannot open shared object file: No such file or directory\n","2021-11-01 18:29:23.819221: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcublas.so.11'; dlerror: libcublas.so.11: cannot open shared object file: No such file or directory\n","2021-11-01 18:29:23.819256: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcublasLt.so.11'; dlerror: libcublasLt.so.11: cannot open shared object file: No such file or directory\n","2021-11-01 18:29:23.819292: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcufft.so.10'; dlerror: libcufft.so.10: cannot open shared object file: No such file or directory\n","2021-11-01 18:29:23.819326: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcurand.so.10'; dlerror: libcurand.so.10: cannot open shared object file: No such file or directory\n","2021-11-01 18:29:23.819359: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcusolver.so.11'; dlerror: libcusolver.so.11: cannot open shared object file: No such file or directory\n","2021-11-01 18:29:23.819393: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcusparse.so.11'; dlerror: libcusparse.so.11: cannot open shared object file: No such file or directory\n","2021-11-01 18:29:23.819427: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcudnn.so.8'; dlerror: libcudnn.so.8: cannot open shared object file: No such file or directory\n","2021-11-01 18:29:23.819436: W tensorflow/core/common_runtime/gpu/gpu_device.cc:1835] Cannot dlopen some GPU libraries. Please make sure the missing libraries mentioned above are installed properly if you would like to use GPU. Follow the guide at https://www.tensorflow.org/install/gpu for how to download and setup the required libraries for your platform.\n","Skipping registering GPU devices...\n","2021-11-01 18:29:23.820561: I tensorflow/core/platform/cpu_feature_guard.cc:142] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  AVX2 AVX512F FMA\n","To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.\n"]},{"data":{"image/png":"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","text/plain":["<PIL.PngImagePlugin.PngImageFile image mode=RGB size=807x1512 at 0x7F781AB15FA0>"]},"execution_count":2,"metadata":{},"output_type":"execute_result"}],"source":["\n","def gen_model():\n","    inputs = tf.keras.layers.Input(shape=(32, 32, 3))\n","    x = inputs\n","    x = layers.Conv2D(32,3,activation='relu')(x)\n","    x = layers.Conv2D(32,3,activation='relu')(x)\n","    x = layers.MaxPool2D(2)(x)\n","    x = layers.Dropout(0.3)(x)\n","    x = layers.Conv2D(64,3,activation='relu')(x)\n","    x = layers.Conv2D(64,3,activation='relu')(x)\n","    x = layers.MaxPool2D(2)(x)\n","    x = layers.Dropout(0.3)(x)\n","    x = layers.Conv2D(84,3,activation='relu')(x)\n","    x = layers.Dropout(0.3)(x)\n","    x = layers.Flatten()(x)\n","    x1 = layers.Dense(64,activation='relu')(x)\n","    x2 = layers.Dense(64,activation='relu')(x)\n","    x1 = layers.Dense(1,activation='sigmoid',name='sex_out')(x1)\n","    x2 = layers.Dense(1,activation='relu',name='age_out')(x2)\n","    model = tf.keras.models.Model(inputs=inputs, outputs=[x1, x2])\n","    model.compile(optimizer='Adam', loss=['binary_crossentropy','mae'])\n","    tf.keras.utils.plot_model(model, 'model.png',show_shapes=True)  \n","    return model\n","model=gen_model()\n","\n","Image.open('model.png')"]},{"cell_type":"markdown","metadata":{},"source":["## Train Model"]},{"cell_type":"code","execution_count":9,"metadata":{"trusted":true},"outputs":[{"name":"stdout","output_type":"stream","text":["__________________________________________________________________________________________________\n","Layer (type)                    Output Shape         Param #     Connected to                     \n","==================================================================================================\n","input_1 (InputLayer)            (None, 32, 32, 3)    0                                            \n","__________________________________________________________________________________________________\n","conv2d (Conv2D)                 (None, 30, 30, 32)   896         input_1[0][0]                    \n","__________________________________________________________________________________________________\n","conv2d_1 (Conv2D)               (None, 28, 28, 32)   9248        conv2d[0][0]                     \n","__________________________________________________________________________________________________\n","max_pooling2d (MaxPooling2D)    (None, 14, 14, 32)   0           conv2d_1[0][0]                   \n","__________________________________________________________________________________________________\n","dropout (Dropout)               (None, 14, 14, 32)   0           max_pooling2d[0][0]              \n","__________________________________________________________________________________________________\n","conv2d_2 (Conv2D)               (None, 12, 12, 64)   18496       dropout[0][0]                    \n","__________________________________________________________________________________________________\n","conv2d_3 (Conv2D)               (None, 10, 10, 64)   36928       conv2d_2[0][0]                   \n","__________________________________________________________________________________________________\n","max_pooling2d_1 (MaxPooling2D)  (None, 5, 5, 64)     0           conv2d_3[0][0]                   \n","__________________________________________________________________________________________________\n","dropout_1 (Dropout)             (None, 5, 5, 64)     0           max_pooling2d_1[0][0]            \n","__________________________________________________________________________________________________\n","conv2d_4 (Conv2D)               (None, 3, 3, 84)     48468       dropout_1[0][0]                  \n","__________________________________________________________________________________________________\n","dropout_2 (Dropout)             (None, 3, 3, 84)     0           conv2d_4[0][0]                   \n","__________________________________________________________________________________________________\n","flatten (Flatten)               (None, 756)          0           dropout_2[0][0]                  \n","__________________________________________________________________________________________________\n","dense (Dense)                   (None, 64)           48448       flatten[0][0]                    \n","__________________________________________________________________________________________________\n","dense_1 (Dense)                 (None, 64)           48448       flatten[0][0]                    \n","__________________________________________________________________________________________________\n","sex_out (Dense)                 (None, 1)            65          dense[0][0]                      \n","__________________________________________________________________________________________________\n","age_out (Dense)                 (None, 1)            65          dense_1[0][0]                    \n","==================================================================================================\n","Total params: 211,062\n","Trainable params: 211,062\n","Non-trainable params: 0\n","__________________________________________________________________________________________________\n","Train on 15884 samples, validate on 7824 samples\n","Epoch 1/200\n","15884/15884 [==============================] - 5s 343us/sample - loss: 20.1352 - sex_out_loss: 1.3369 - age_out_loss: 18.7983 - val_loss: 18.1432 - val_sex_out_loss: 0.6765 - val_age_out_loss: 17.4667\n","Epoch 2/200\n","15884/15884 [==============================] - 1s 74us/sample - loss: 15.0300 - sex_out_loss: 0.6538 - age_out_loss: 14.3762 - val_loss: 14.1098 - val_sex_out_loss: 0.5856 - val_age_out_loss: 13.5242\n","Epoch 3/200\n","15884/15884 [==============================] - 1s 70us/sample - loss: 14.0605 - sex_out_loss: 0.6004 - age_out_loss: 13.4601 - val_loss: 13.9105 - val_sex_out_loss: 0.5442 - val_age_out_loss: 13.3663\n","Epoch 4/200\n","15884/15884 [==============================] - 1s 67us/sample - loss: 12.8940 - sex_out_loss: 0.5634 - age_out_loss: 12.3306 - val_loss: 12.4436 - val_sex_out_loss: 0.5005 - val_age_out_loss: 11.9431\n","Epoch 5/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 12.4806 - sex_out_loss: 0.5393 - age_out_loss: 11.9413 - val_loss: 12.2279 - val_sex_out_loss: 0.4798 - val_age_out_loss: 11.7481\n","Epoch 6/200\n","15884/15884 [==============================] - 1s 69us/sample - loss: 11.5448 - sex_out_loss: 0.5220 - age_out_loss: 11.0228 - val_loss: 14.9710 - val_sex_out_loss: 0.4792 - val_age_out_loss: 14.4918\n","Epoch 7/200\n","15884/15884 [==============================] - 1s 69us/sample - loss: 11.0985 - sex_out_loss: 0.5045 - age_out_loss: 10.5940 - val_loss: 10.2500 - val_sex_out_loss: 0.4463 - val_age_out_loss: 9.8037\n","Epoch 8/200\n","15884/15884 [==============================] - 1s 75us/sample - loss: 10.6759 - sex_out_loss: 0.4868 - age_out_loss: 10.1891 - val_loss: 9.8628 - val_sex_out_loss: 0.4298 - val_age_out_loss: 9.4330\n","Epoch 9/200\n","15884/15884 [==============================] - 1s 77us/sample - loss: 10.3572 - sex_out_loss: 0.4848 - age_out_loss: 9.8724 - val_loss: 9.5360 - val_sex_out_loss: 0.4290 - val_age_out_loss: 9.1069\n","Epoch 10/200\n","15884/15884 [==============================] - 1s 70us/sample - loss: 10.2395 - sex_out_loss: 0.4716 - age_out_loss: 9.7679 - val_loss: 8.9616 - val_sex_out_loss: 0.4309 - val_age_out_loss: 8.5307\n","Epoch 11/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 10.0904 - sex_out_loss: 0.4701 - age_out_loss: 9.6203 - val_loss: 10.9089 - val_sex_out_loss: 0.4164 - val_age_out_loss: 10.4925\n","Epoch 12/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 9.7824 - sex_out_loss: 0.4577 - age_out_loss: 9.3248 - val_loss: 9.4953 - val_sex_out_loss: 0.4039 - val_age_out_loss: 9.0914\n","Epoch 13/200\n","15884/15884 [==============================] - 1s 67us/sample - loss: 9.5763 - sex_out_loss: 0.4565 - age_out_loss: 9.1198 - val_loss: 11.5833 - val_sex_out_loss: 0.4057 - val_age_out_loss: 11.1776\n","Epoch 14/200\n","15884/15884 [==============================] - 1s 67us/sample - loss: 9.4367 - sex_out_loss: 0.4406 - age_out_loss: 8.9961 - val_loss: 10.9122 - val_sex_out_loss: 0.4034 - val_age_out_loss: 10.5087\n","Epoch 15/200\n","15884/15884 [==============================] - 1s 69us/sample - loss: 9.1426 - sex_out_loss: 0.4458 - age_out_loss: 8.6969 - val_loss: 8.5305 - val_sex_out_loss: 0.3878 - val_age_out_loss: 8.1427\n","Epoch 16/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 9.1498 - sex_out_loss: 0.4369 - age_out_loss: 8.7129 - val_loss: 8.4546 - val_sex_out_loss: 0.3769 - val_age_out_loss: 8.0777\n","Epoch 17/200\n","15884/15884 [==============================] - 1s 67us/sample - loss: 8.8935 - sex_out_loss: 0.4326 - age_out_loss: 8.4609 - val_loss: 9.2926 - val_sex_out_loss: 0.3823 - val_age_out_loss: 8.9103\n","Epoch 18/200\n","15884/15884 [==============================] - 1s 72us/sample - loss: 8.8730 - sex_out_loss: 0.4292 - age_out_loss: 8.4439 - val_loss: 12.8921 - val_sex_out_loss: 0.3881 - val_age_out_loss: 12.5040\n","Epoch 19/200\n","15884/15884 [==============================] - 1s 72us/sample - loss: 9.2552 - sex_out_loss: 0.4228 - age_out_loss: 8.8325 - val_loss: 9.5977 - val_sex_out_loss: 0.3846 - val_age_out_loss: 9.2131\n","Epoch 20/200\n","15884/15884 [==============================] - 1s 80us/sample - loss: 8.5678 - sex_out_loss: 0.4144 - age_out_loss: 8.1535 - val_loss: 8.0909 - val_sex_out_loss: 0.3782 - val_age_out_loss: 7.7127\n","Epoch 21/200\n","15884/15884 [==============================] - 1s 71us/sample - loss: 8.6392 - sex_out_loss: 0.4159 - age_out_loss: 8.2233 - val_loss: 8.4743 - val_sex_out_loss: 0.3650 - val_age_out_loss: 8.1093\n","Epoch 22/200\n"]},{"name":"stdout","output_type":"stream","text":["15884/15884 [==============================] - 1s 71us/sample - loss: 8.5555 - sex_out_loss: 0.4161 - age_out_loss: 8.1394 - val_loss: 8.9015 - val_sex_out_loss: 0.3744 - val_age_out_loss: 8.5271\n","Epoch 23/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 8.5484 - sex_out_loss: 0.4115 - age_out_loss: 8.1370 - val_loss: 9.3168 - val_sex_out_loss: 0.3624 - val_age_out_loss: 8.9544\n","Epoch 24/200\n","15884/15884 [==============================] - 1s 69us/sample - loss: 8.4799 - sex_out_loss: 0.4046 - age_out_loss: 8.0753 - val_loss: 10.6214 - val_sex_out_loss: 0.3641 - val_age_out_loss: 10.2573\n","Epoch 25/200\n","15884/15884 [==============================] - 1s 79us/sample - loss: 8.2865 - sex_out_loss: 0.4010 - age_out_loss: 7.8855 - val_loss: 10.3150 - val_sex_out_loss: 0.3536 - val_age_out_loss: 9.9614\n","Epoch 26/200\n","15884/15884 [==============================] - 1s 69us/sample - loss: 8.2774 - sex_out_loss: 0.3972 - age_out_loss: 7.8802 - val_loss: 8.9000 - val_sex_out_loss: 0.3488 - val_age_out_loss: 8.5512\n","Epoch 27/200\n","15884/15884 [==============================] - 1s 71us/sample - loss: 8.1444 - sex_out_loss: 0.3980 - age_out_loss: 7.7463 - val_loss: 9.3798 - val_sex_out_loss: 0.3630 - val_age_out_loss: 9.0168\n","Epoch 28/200\n","15884/15884 [==============================] - 1s 81us/sample - loss: 8.0504 - sex_out_loss: 0.3951 - age_out_loss: 7.6553 - val_loss: 8.3641 - val_sex_out_loss: 0.3432 - val_age_out_loss: 8.0209\n","Epoch 29/200\n","15884/15884 [==============================] - 1s 73us/sample - loss: 7.9929 - sex_out_loss: 0.3955 - age_out_loss: 7.5974 - val_loss: 9.3312 - val_sex_out_loss: 0.3474 - val_age_out_loss: 8.9838\n","Epoch 30/200\n","15884/15884 [==============================] - 1s 74us/sample - loss: 7.9898 - sex_out_loss: 0.3902 - age_out_loss: 7.5996 - val_loss: 7.8321 - val_sex_out_loss: 0.3390 - val_age_out_loss: 7.4931\n","Epoch 31/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 7.8594 - sex_out_loss: 0.3922 - age_out_loss: 7.4671 - val_loss: 8.9444 - val_sex_out_loss: 0.3436 - val_age_out_loss: 8.6008\n","Epoch 32/200\n","15884/15884 [==============================] - 1s 69us/sample - loss: 7.8567 - sex_out_loss: 0.3856 - age_out_loss: 7.4711 - val_loss: 7.5929 - val_sex_out_loss: 0.3340 - val_age_out_loss: 7.2589\n","Epoch 33/200\n","15884/15884 [==============================] - 1s 71us/sample - loss: 8.0169 - sex_out_loss: 0.3832 - age_out_loss: 7.6337 - val_loss: 10.0690 - val_sex_out_loss: 0.3546 - val_age_out_loss: 9.7144\n","Epoch 34/200\n","15884/15884 [==============================] - 1s 69us/sample - loss: 7.9444 - sex_out_loss: 0.3875 - age_out_loss: 7.5569 - val_loss: 9.0076 - val_sex_out_loss: 0.3340 - val_age_out_loss: 8.6736\n","Epoch 35/200\n","15884/15884 [==============================] - 1s 69us/sample - loss: 7.8949 - sex_out_loss: 0.3834 - age_out_loss: 7.5115 - val_loss: 7.9032 - val_sex_out_loss: 0.3359 - val_age_out_loss: 7.5673\n","Epoch 36/200\n","15884/15884 [==============================] - 1s 69us/sample - loss: 7.5980 - sex_out_loss: 0.3779 - age_out_loss: 7.2202 - val_loss: 8.4003 - val_sex_out_loss: 0.3504 - val_age_out_loss: 8.0499\n","Epoch 37/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 7.4959 - sex_out_loss: 0.3748 - age_out_loss: 7.1211 - val_loss: 7.3328 - val_sex_out_loss: 0.3281 - val_age_out_loss: 7.0048\n","Epoch 38/200\n","15884/15884 [==============================] - 1s 78us/sample - loss: 7.5416 - sex_out_loss: 0.3734 - age_out_loss: 7.1683 - val_loss: 8.8389 - val_sex_out_loss: 0.3301 - val_age_out_loss: 8.5088\n","Epoch 39/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 7.4221 - sex_out_loss: 0.3764 - age_out_loss: 7.0457 - val_loss: 7.8612 - val_sex_out_loss: 0.3257 - val_age_out_loss: 7.5355\n","Epoch 40/200\n","15884/15884 [==============================] - 1s 67us/sample - loss: 7.5151 - sex_out_loss: 0.3742 - age_out_loss: 7.1409 - val_loss: 7.3235 - val_sex_out_loss: 0.3239 - val_age_out_loss: 6.9995\n","Epoch 41/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 7.5859 - sex_out_loss: 0.3649 - age_out_loss: 7.2209 - val_loss: 7.4734 - val_sex_out_loss: 0.3284 - val_age_out_loss: 7.1450\n","Epoch 42/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 7.4874 - sex_out_loss: 0.3691 - age_out_loss: 7.1183 - val_loss: 7.9868 - val_sex_out_loss: 0.3245 - val_age_out_loss: 7.6623\n","Epoch 43/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 7.3944 - sex_out_loss: 0.3690 - age_out_loss: 7.0254 - val_loss: 7.8527 - val_sex_out_loss: 0.3284 - val_age_out_loss: 7.5242\n","Epoch 44/200\n","15884/15884 [==============================] - 1s 67us/sample - loss: 7.3632 - sex_out_loss: 0.3743 - age_out_loss: 6.9889 - val_loss: 8.9406 - val_sex_out_loss: 0.3386 - val_age_out_loss: 8.6020\n","Epoch 45/200\n","15884/15884 [==============================] - 1s 67us/sample - loss: 7.2799 - sex_out_loss: 0.3676 - age_out_loss: 6.9122 - val_loss: 7.3837 - val_sex_out_loss: 0.3226 - val_age_out_loss: 7.0611\n","Epoch 46/200\n","15884/15884 [==============================] - 1s 67us/sample - loss: 7.2500 - sex_out_loss: 0.3652 - age_out_loss: 6.8848 - val_loss: 7.4845 - val_sex_out_loss: 0.3192 - val_age_out_loss: 7.1653\n","Epoch 47/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 7.2595 - sex_out_loss: 0.3649 - age_out_loss: 6.8946 - val_loss: 8.3388 - val_sex_out_loss: 0.3223 - val_age_out_loss: 8.0165\n","Epoch 48/200\n","15884/15884 [==============================] - 1s 75us/sample - loss: 7.3048 - sex_out_loss: 0.3582 - age_out_loss: 6.9467 - val_loss: 7.5819 - val_sex_out_loss: 0.3148 - val_age_out_loss: 7.2672\n","Epoch 49/200\n","15884/15884 [==============================] - 1s 67us/sample - loss: 7.1656 - sex_out_loss: 0.3564 - age_out_loss: 6.8092 - val_loss: 9.5111 - val_sex_out_loss: 0.3171 - val_age_out_loss: 9.1940\n","Epoch 50/200\n","15884/15884 [==============================] - 1s 67us/sample - loss: 7.1650 - sex_out_loss: 0.3661 - age_out_loss: 6.7989 - val_loss: 8.5420 - val_sex_out_loss: 0.3178 - val_age_out_loss: 8.2243\n","Epoch 51/200\n","15884/15884 [==============================] - 1s 69us/sample - loss: 7.1359 - sex_out_loss: 0.3622 - age_out_loss: 6.7737 - val_loss: 7.6493 - val_sex_out_loss: 0.3286 - val_age_out_loss: 7.3207\n","Epoch 52/200\n","15884/15884 [==============================] - 1s 67us/sample - loss: 7.0275 - sex_out_loss: 0.3531 - age_out_loss: 6.6744 - val_loss: 7.4650 - val_sex_out_loss: 0.3157 - val_age_out_loss: 7.1493\n","Epoch 53/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 6.8848 - sex_out_loss: 0.3535 - age_out_loss: 6.5313 - val_loss: 8.5717 - val_sex_out_loss: 0.3301 - val_age_out_loss: 8.2416\n","Epoch 54/200\n","15884/15884 [==============================] - 1s 67us/sample - loss: 7.0925 - sex_out_loss: 0.3585 - age_out_loss: 6.7340 - val_loss: 7.7798 - val_sex_out_loss: 0.3139 - val_age_out_loss: 7.4659\n","Epoch 55/200\n","15884/15884 [==============================] - 1s 67us/sample - loss: 6.9874 - sex_out_loss: 0.3592 - age_out_loss: 6.6282 - val_loss: 7.6692 - val_sex_out_loss: 0.3156 - val_age_out_loss: 7.3535\n","Epoch 56/200\n","15884/15884 [==============================] - 1s 67us/sample - loss: 7.0565 - sex_out_loss: 0.3634 - age_out_loss: 6.6932 - val_loss: 7.6276 - val_sex_out_loss: 0.3141 - val_age_out_loss: 7.3135\n","Epoch 57/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 6.7982 - sex_out_loss: 0.3527 - age_out_loss: 6.4455 - val_loss: 8.5989 - val_sex_out_loss: 0.3131 - val_age_out_loss: 8.2858\n","Epoch 58/200\n","15884/15884 [==============================] - 1s 86us/sample - loss: 6.9841 - sex_out_loss: 0.3492 - age_out_loss: 6.6349 - val_loss: 7.8830 - val_sex_out_loss: 0.3120 - val_age_out_loss: 7.5711\n","Epoch 59/200\n","15884/15884 [==============================] - 1s 69us/sample - loss: 6.8444 - sex_out_loss: 0.3452 - age_out_loss: 6.4992 - val_loss: 7.2035 - val_sex_out_loss: 0.3087 - val_age_out_loss: 6.8948\n","Epoch 60/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 6.9350 - sex_out_loss: 0.3538 - age_out_loss: 6.5812 - val_loss: 7.4012 - val_sex_out_loss: 0.3137 - val_age_out_loss: 7.0875\n","Epoch 61/200\n"]},{"name":"stdout","output_type":"stream","text":["15884/15884 [==============================] - 1s 67us/sample - loss: 6.8448 - sex_out_loss: 0.3479 - age_out_loss: 6.4970 - val_loss: 7.5870 - val_sex_out_loss: 0.3070 - val_age_out_loss: 7.2800\n","Epoch 62/200\n","15884/15884 [==============================] - 1s 67us/sample - loss: 6.8080 - sex_out_loss: 0.3516 - age_out_loss: 6.4564 - val_loss: 7.5578 - val_sex_out_loss: 0.3153 - val_age_out_loss: 7.2425\n","Epoch 63/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 6.7323 - sex_out_loss: 0.3468 - age_out_loss: 6.3855 - val_loss: 7.0846 - val_sex_out_loss: 0.3132 - val_age_out_loss: 6.7714\n","Epoch 64/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 6.7692 - sex_out_loss: 0.3429 - age_out_loss: 6.4263 - val_loss: 7.6551 - val_sex_out_loss: 0.3079 - val_age_out_loss: 7.3473\n","Epoch 65/200\n","15884/15884 [==============================] - 1s 67us/sample - loss: 6.7172 - sex_out_loss: 0.3435 - age_out_loss: 6.3737 - val_loss: 7.2490 - val_sex_out_loss: 0.3047 - val_age_out_loss: 6.9444\n","Epoch 66/200\n","15884/15884 [==============================] - 1s 67us/sample - loss: 6.7033 - sex_out_loss: 0.3554 - age_out_loss: 6.3478 - val_loss: 7.1217 - val_sex_out_loss: 0.3075 - val_age_out_loss: 6.8142\n","Epoch 67/200\n","15884/15884 [==============================] - 1s 67us/sample - loss: 6.8777 - sex_out_loss: 0.3412 - age_out_loss: 6.5365 - val_loss: 8.2572 - val_sex_out_loss: 0.3116 - val_age_out_loss: 7.9456\n","Epoch 68/200\n","15884/15884 [==============================] - 1s 78us/sample - loss: 6.6997 - sex_out_loss: 0.3451 - age_out_loss: 6.3546 - val_loss: 7.4087 - val_sex_out_loss: 0.3141 - val_age_out_loss: 7.0945\n","Epoch 69/200\n","15884/15884 [==============================] - 1s 69us/sample - loss: 6.6830 - sex_out_loss: 0.3534 - age_out_loss: 6.3296 - val_loss: 6.9254 - val_sex_out_loss: 0.3151 - val_age_out_loss: 6.6104\n","Epoch 70/200\n","15884/15884 [==============================] - 1s 67us/sample - loss: 6.6365 - sex_out_loss: 0.3441 - age_out_loss: 6.2924 - val_loss: 7.2816 - val_sex_out_loss: 0.3048 - val_age_out_loss: 6.9768\n","Epoch 71/200\n","15884/15884 [==============================] - 1s 67us/sample - loss: 6.5210 - sex_out_loss: 0.3388 - age_out_loss: 6.1822 - val_loss: 6.9758 - val_sex_out_loss: 0.3048 - val_age_out_loss: 6.6710\n","Epoch 72/200\n","15884/15884 [==============================] - 1s 67us/sample - loss: 6.6119 - sex_out_loss: 0.3399 - age_out_loss: 6.2720 - val_loss: 6.9865 - val_sex_out_loss: 0.3030 - val_age_out_loss: 6.6834\n","Epoch 73/200\n","15884/15884 [==============================] - 1s 67us/sample - loss: 6.5442 - sex_out_loss: 0.3388 - age_out_loss: 6.2054 - val_loss: 7.0952 - val_sex_out_loss: 0.3042 - val_age_out_loss: 6.7910\n","Epoch 74/200\n","15884/15884 [==============================] - 1s 82us/sample - loss: 6.6852 - sex_out_loss: 0.3397 - age_out_loss: 6.3455 - val_loss: 8.0494 - val_sex_out_loss: 0.3045 - val_age_out_loss: 7.7450\n","Epoch 75/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 6.6041 - sex_out_loss: 0.3403 - age_out_loss: 6.2638 - val_loss: 7.2689 - val_sex_out_loss: 0.3011 - val_age_out_loss: 6.9678\n","Epoch 76/200\n","15884/15884 [==============================] - 1s 70us/sample - loss: 6.5735 - sex_out_loss: 0.3370 - age_out_loss: 6.2365 - val_loss: 7.2474 - val_sex_out_loss: 0.3038 - val_age_out_loss: 6.9436\n","Epoch 77/200\n","15884/15884 [==============================] - 1s 71us/sample - loss: 6.4829 - sex_out_loss: 0.3359 - age_out_loss: 6.1470 - val_loss: 7.7817 - val_sex_out_loss: 0.3052 - val_age_out_loss: 7.4764\n","Epoch 78/200\n","15884/15884 [==============================] - 1s 83us/sample - loss: 6.6138 - sex_out_loss: 0.3325 - age_out_loss: 6.2813 - val_loss: 7.4002 - val_sex_out_loss: 0.3067 - val_age_out_loss: 7.0935\n","Epoch 79/200\n","15884/15884 [==============================] - 1s 73us/sample - loss: 6.5401 - sex_out_loss: 0.3378 - age_out_loss: 6.2023 - val_loss: 6.9236 - val_sex_out_loss: 0.3048 - val_age_out_loss: 6.6187\n","Epoch 80/200\n","15884/15884 [==============================] - 1s 70us/sample - loss: 6.4133 - sex_out_loss: 0.3366 - age_out_loss: 6.0767 - val_loss: 7.0717 - val_sex_out_loss: 0.3078 - val_age_out_loss: 6.7638\n","Epoch 81/200\n","15884/15884 [==============================] - 1s 70us/sample - loss: 6.5599 - sex_out_loss: 0.3411 - age_out_loss: 6.2187 - val_loss: 7.0080 - val_sex_out_loss: 0.3036 - val_age_out_loss: 6.7043\n","Epoch 82/200\n","15884/15884 [==============================] - 1s 70us/sample - loss: 6.4614 - sex_out_loss: 0.3346 - age_out_loss: 6.1269 - val_loss: 8.0120 - val_sex_out_loss: 0.3052 - val_age_out_loss: 7.7068\n","Epoch 83/200\n","15884/15884 [==============================] - 1s 70us/sample - loss: 6.5514 - sex_out_loss: 0.3367 - age_out_loss: 6.2147 - val_loss: 7.0620 - val_sex_out_loss: 0.3087 - val_age_out_loss: 6.7533\n","Epoch 84/200\n","15884/15884 [==============================] - 1s 67us/sample - loss: 6.5582 - sex_out_loss: 0.3354 - age_out_loss: 6.2228 - val_loss: 7.8868 - val_sex_out_loss: 0.3121 - val_age_out_loss: 7.5747\n","Epoch 85/200\n","15884/15884 [==============================] - 1s 69us/sample - loss: 6.4014 - sex_out_loss: 0.3350 - age_out_loss: 6.0664 - val_loss: 7.1654 - val_sex_out_loss: 0.3036 - val_age_out_loss: 6.8618\n","Epoch 86/200\n","15884/15884 [==============================] - 1s 75us/sample - loss: 6.3882 - sex_out_loss: 0.3336 - age_out_loss: 6.0546 - val_loss: 7.8609 - val_sex_out_loss: 0.3087 - val_age_out_loss: 7.5522\n","Epoch 87/200\n","15884/15884 [==============================] - 1s 87us/sample - loss: 6.4437 - sex_out_loss: 0.3345 - age_out_loss: 6.1092 - val_loss: 7.0774 - val_sex_out_loss: 0.3036 - val_age_out_loss: 6.7739\n","Epoch 88/200\n","15884/15884 [==============================] - 1s 69us/sample - loss: 6.3004 - sex_out_loss: 0.3310 - age_out_loss: 5.9694 - val_loss: 7.5358 - val_sex_out_loss: 0.2971 - val_age_out_loss: 7.2386\n","Epoch 89/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 6.3239 - sex_out_loss: 0.3266 - age_out_loss: 5.9973 - val_loss: 6.9391 - val_sex_out_loss: 0.2973 - val_age_out_loss: 6.6418\n","Epoch 90/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 6.3798 - sex_out_loss: 0.3314 - age_out_loss: 6.0484 - val_loss: 6.8031 - val_sex_out_loss: 0.2991 - val_age_out_loss: 6.5039\n","Epoch 91/200\n","15884/15884 [==============================] - 1s 67us/sample - loss: 6.3861 - sex_out_loss: 0.3319 - age_out_loss: 6.0542 - val_loss: 6.7465 - val_sex_out_loss: 0.3068 - val_age_out_loss: 6.4397\n","Epoch 92/200\n","15884/15884 [==============================] - 1s 70us/sample - loss: 6.3433 - sex_out_loss: 0.3288 - age_out_loss: 6.0145 - val_loss: 7.3655 - val_sex_out_loss: 0.2979 - val_age_out_loss: 7.0677\n","Epoch 93/200\n","15884/15884 [==============================] - 1s 67us/sample - loss: 6.2950 - sex_out_loss: 0.3257 - age_out_loss: 5.9693 - val_loss: 6.8278 - val_sex_out_loss: 0.2979 - val_age_out_loss: 6.5299\n","Epoch 94/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 6.2383 - sex_out_loss: 0.3313 - age_out_loss: 5.9070 - val_loss: 7.0591 - val_sex_out_loss: 0.3038 - val_age_out_loss: 6.7553\n","Epoch 95/200\n","15884/15884 [==============================] - 1s 69us/sample - loss: 6.3799 - sex_out_loss: 0.3250 - age_out_loss: 6.0548 - val_loss: 7.1770 - val_sex_out_loss: 0.3015 - val_age_out_loss: 6.8755\n","Epoch 96/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 6.1776 - sex_out_loss: 0.3244 - age_out_loss: 5.8533 - val_loss: 6.8260 - val_sex_out_loss: 0.2971 - val_age_out_loss: 6.5288\n","Epoch 97/200\n","15884/15884 [==============================] - 1s 79us/sample - loss: 6.2634 - sex_out_loss: 0.3252 - age_out_loss: 5.9382 - val_loss: 6.9898 - val_sex_out_loss: 0.2963 - val_age_out_loss: 6.6934\n","Epoch 98/200\n","15884/15884 [==============================] - 1s 70us/sample - loss: 6.2692 - sex_out_loss: 0.3295 - age_out_loss: 5.9396 - val_loss: 6.7850 - val_sex_out_loss: 0.2965 - val_age_out_loss: 6.4885\n","Epoch 99/200\n","15884/15884 [==============================] - 1s 71us/sample - loss: 6.2915 - sex_out_loss: 0.3305 - age_out_loss: 5.9610 - val_loss: 7.7015 - val_sex_out_loss: 0.2993 - val_age_out_loss: 7.4023\n","Epoch 100/200\n"]},{"name":"stdout","output_type":"stream","text":["15884/15884 [==============================] - 1s 71us/sample - loss: 6.2696 - sex_out_loss: 0.3231 - age_out_loss: 5.9465 - val_loss: 7.0993 - val_sex_out_loss: 0.2969 - val_age_out_loss: 6.8024\n","Epoch 101/200\n","15884/15884 [==============================] - 1s 70us/sample - loss: 6.1783 - sex_out_loss: 0.3217 - age_out_loss: 5.8567 - val_loss: 6.6926 - val_sex_out_loss: 0.3003 - val_age_out_loss: 6.3923\n","Epoch 102/200\n","15884/15884 [==============================] - 1s 71us/sample - loss: 6.2913 - sex_out_loss: 0.3277 - age_out_loss: 5.9636 - val_loss: 6.7772 - val_sex_out_loss: 0.2996 - val_age_out_loss: 6.4776\n","Epoch 103/200\n","15884/15884 [==============================] - 1s 70us/sample - loss: 6.1569 - sex_out_loss: 0.3190 - age_out_loss: 5.8380 - val_loss: 6.6913 - val_sex_out_loss: 0.2944 - val_age_out_loss: 6.3968\n","Epoch 104/200\n","15884/15884 [==============================] - 1s 72us/sample - loss: 6.2198 - sex_out_loss: 0.3263 - age_out_loss: 5.8935 - val_loss: 7.5988 - val_sex_out_loss: 0.3080 - val_age_out_loss: 7.2908\n","Epoch 105/200\n","15884/15884 [==============================] - 1s 70us/sample - loss: 6.1820 - sex_out_loss: 0.3306 - age_out_loss: 5.8514 - val_loss: 6.8951 - val_sex_out_loss: 0.2969 - val_age_out_loss: 6.5982\n","Epoch 106/200\n","15884/15884 [==============================] - 1s 72us/sample - loss: 6.1975 - sex_out_loss: 0.3266 - age_out_loss: 5.8709 - val_loss: 6.6246 - val_sex_out_loss: 0.2970 - val_age_out_loss: 6.3276\n","Epoch 107/200\n","15884/15884 [==============================] - 1s 77us/sample - loss: 6.0087 - sex_out_loss: 0.3175 - age_out_loss: 5.6911 - val_loss: 7.1565 - val_sex_out_loss: 0.2956 - val_age_out_loss: 6.8609\n","Epoch 108/200\n","15884/15884 [==============================] - 1s 71us/sample - loss: 6.1042 - sex_out_loss: 0.3158 - age_out_loss: 5.7884 - val_loss: 6.8128 - val_sex_out_loss: 0.2884 - val_age_out_loss: 6.5244\n","Epoch 109/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 6.0402 - sex_out_loss: 0.3184 - age_out_loss: 5.7218 - val_loss: 6.7541 - val_sex_out_loss: 0.2960 - val_age_out_loss: 6.4581\n","Epoch 110/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 6.0871 - sex_out_loss: 0.3192 - age_out_loss: 5.7679 - val_loss: 6.9057 - val_sex_out_loss: 0.2921 - val_age_out_loss: 6.6136\n","Epoch 111/200\n","15884/15884 [==============================] - 1s 69us/sample - loss: 6.0735 - sex_out_loss: 0.3145 - age_out_loss: 5.7591 - val_loss: 6.8155 - val_sex_out_loss: 0.3005 - val_age_out_loss: 6.5150\n","Epoch 112/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 6.0344 - sex_out_loss: 0.3138 - age_out_loss: 5.7206 - val_loss: 7.3311 - val_sex_out_loss: 0.2912 - val_age_out_loss: 7.0399\n","Epoch 113/200\n","15884/15884 [==============================] - 1s 70us/sample - loss: 6.1075 - sex_out_loss: 0.3153 - age_out_loss: 5.7922 - val_loss: 7.0297 - val_sex_out_loss: 0.2955 - val_age_out_loss: 6.7342\n","Epoch 114/200\n","15884/15884 [==============================] - 1s 67us/sample - loss: 6.1954 - sex_out_loss: 0.3244 - age_out_loss: 5.8710 - val_loss: 6.6914 - val_sex_out_loss: 0.2924 - val_age_out_loss: 6.3990\n","Epoch 115/200\n","15884/15884 [==============================] - 1s 75us/sample - loss: 6.0202 - sex_out_loss: 0.3151 - age_out_loss: 5.7051 - val_loss: 6.7303 - val_sex_out_loss: 0.2898 - val_age_out_loss: 6.4405\n","Epoch 116/200\n","15884/15884 [==============================] - 1s 74us/sample - loss: 6.0770 - sex_out_loss: 0.3275 - age_out_loss: 5.7495 - val_loss: 6.6497 - val_sex_out_loss: 0.2915 - val_age_out_loss: 6.3582\n","Epoch 117/200\n","15884/15884 [==============================] - 1s 72us/sample - loss: 6.1526 - sex_out_loss: 0.3294 - age_out_loss: 5.8232 - val_loss: 7.1466 - val_sex_out_loss: 0.2980 - val_age_out_loss: 6.8486\n","Epoch 118/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 6.1001 - sex_out_loss: 0.3275 - age_out_loss: 5.7726 - val_loss: 8.4500 - val_sex_out_loss: 0.2942 - val_age_out_loss: 8.1558\n","Epoch 119/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 5.9816 - sex_out_loss: 0.3147 - age_out_loss: 5.6669 - val_loss: 7.4351 - val_sex_out_loss: 0.2975 - val_age_out_loss: 7.1376\n","Epoch 120/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 6.0468 - sex_out_loss: 0.3205 - age_out_loss: 5.7263 - val_loss: 6.8330 - val_sex_out_loss: 0.2922 - val_age_out_loss: 6.5408\n","Epoch 121/200\n","15884/15884 [==============================] - 1s 69us/sample - loss: 6.0628 - sex_out_loss: 0.3208 - age_out_loss: 5.7419 - val_loss: 6.7341 - val_sex_out_loss: 0.3060 - val_age_out_loss: 6.4282\n","Epoch 122/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 6.1162 - sex_out_loss: 0.3221 - age_out_loss: 5.7941 - val_loss: 6.7238 - val_sex_out_loss: 0.2963 - val_age_out_loss: 6.4276\n","Epoch 123/200\n","15884/15884 [==============================] - 1s 67us/sample - loss: 6.0523 - sex_out_loss: 0.3188 - age_out_loss: 5.7335 - val_loss: 7.4663 - val_sex_out_loss: 0.2979 - val_age_out_loss: 7.1684\n","Epoch 124/200\n","15884/15884 [==============================] - 1s 67us/sample - loss: 5.9870 - sex_out_loss: 0.3173 - age_out_loss: 5.6697 - val_loss: 7.1544 - val_sex_out_loss: 0.2996 - val_age_out_loss: 6.8548\n","Epoch 125/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 5.9862 - sex_out_loss: 0.3163 - age_out_loss: 5.6698 - val_loss: 6.9800 - val_sex_out_loss: 0.2966 - val_age_out_loss: 6.6834\n","Epoch 126/200\n","15884/15884 [==============================] - 1s 75us/sample - loss: 6.0102 - sex_out_loss: 0.3125 - age_out_loss: 5.6977 - val_loss: 7.1265 - val_sex_out_loss: 0.2908 - val_age_out_loss: 6.8357\n","Epoch 127/200\n","15884/15884 [==============================] - 1s 87us/sample - loss: 6.0039 - sex_out_loss: 0.3183 - age_out_loss: 5.6856 - val_loss: 7.4628 - val_sex_out_loss: 0.2898 - val_age_out_loss: 7.1730\n","Epoch 128/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 5.8735 - sex_out_loss: 0.3129 - age_out_loss: 5.5605 - val_loss: 6.8130 - val_sex_out_loss: 0.2980 - val_age_out_loss: 6.5150\n","Epoch 129/200\n","15884/15884 [==============================] - 1s 69us/sample - loss: 5.9714 - sex_out_loss: 0.3158 - age_out_loss: 5.6555 - val_loss: 6.6515 - val_sex_out_loss: 0.2905 - val_age_out_loss: 6.3610\n","Epoch 130/200\n","15884/15884 [==============================] - 1s 72us/sample - loss: 6.0217 - sex_out_loss: 0.3186 - age_out_loss: 5.7031 - val_loss: 6.7675 - val_sex_out_loss: 0.2923 - val_age_out_loss: 6.4752\n","Epoch 131/200\n","15884/15884 [==============================] - 1s 72us/sample - loss: 5.9452 - sex_out_loss: 0.3142 - age_out_loss: 5.6310 - val_loss: 6.7210 - val_sex_out_loss: 0.2944 - val_age_out_loss: 6.4266\n","Epoch 132/200\n","15884/15884 [==============================] - 1s 76us/sample - loss: 5.8963 - sex_out_loss: 0.3158 - age_out_loss: 5.5806 - val_loss: 6.6950 - val_sex_out_loss: 0.2874 - val_age_out_loss: 6.4076\n","Epoch 133/200\n","15884/15884 [==============================] - 1s 70us/sample - loss: 5.9635 - sex_out_loss: 0.3059 - age_out_loss: 5.6576 - val_loss: 6.9140 - val_sex_out_loss: 0.2904 - val_age_out_loss: 6.6236\n","Epoch 134/200\n","15884/15884 [==============================] - 1s 71us/sample - loss: 6.0783 - sex_out_loss: 0.3121 - age_out_loss: 5.7663 - val_loss: 6.7101 - val_sex_out_loss: 0.2950 - val_age_out_loss: 6.4151\n","Epoch 135/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 5.9983 - sex_out_loss: 0.3186 - age_out_loss: 5.6797 - val_loss: 7.8679 - val_sex_out_loss: 0.2875 - val_age_out_loss: 7.5803\n","Epoch 136/200\n","15884/15884 [==============================] - 1s 80us/sample - loss: 5.9160 - sex_out_loss: 0.3110 - age_out_loss: 5.6049 - val_loss: 6.7793 - val_sex_out_loss: 0.2912 - val_age_out_loss: 6.4880\n","Epoch 137/200\n","15884/15884 [==============================] - 1s 67us/sample - loss: 5.8365 - sex_out_loss: 0.3113 - age_out_loss: 5.5252 - val_loss: 6.6562 - val_sex_out_loss: 0.2881 - val_age_out_loss: 6.3681\n","Epoch 138/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 5.8593 - sex_out_loss: 0.3123 - age_out_loss: 5.5470 - val_loss: 6.8498 - val_sex_out_loss: 0.2954 - val_age_out_loss: 6.5544\n","Epoch 139/200\n"]},{"name":"stdout","output_type":"stream","text":["15884/15884 [==============================] - 1s 68us/sample - loss: 5.8364 - sex_out_loss: 0.3047 - age_out_loss: 5.5317 - val_loss: 7.0722 - val_sex_out_loss: 0.2827 - val_age_out_loss: 6.7895\n","Epoch 140/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 5.8005 - sex_out_loss: 0.3057 - age_out_loss: 5.4948 - val_loss: 6.5007 - val_sex_out_loss: 0.2851 - val_age_out_loss: 6.2156\n","Epoch 141/200\n","15884/15884 [==============================] - 1s 69us/sample - loss: 5.8974 - sex_out_loss: 0.3110 - age_out_loss: 5.5863 - val_loss: 6.8533 - val_sex_out_loss: 0.2922 - val_age_out_loss: 6.5612\n","Epoch 142/200\n","15884/15884 [==============================] - 1s 74us/sample - loss: 5.8565 - sex_out_loss: 0.3116 - age_out_loss: 5.5449 - val_loss: 6.5035 - val_sex_out_loss: 0.2869 - val_age_out_loss: 6.2166\n","Epoch 143/200\n","15884/15884 [==============================] - 1s 71us/sample - loss: 5.7221 - sex_out_loss: 0.3093 - age_out_loss: 5.4128 - val_loss: 6.9542 - val_sex_out_loss: 0.2914 - val_age_out_loss: 6.6628\n","Epoch 144/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 5.8743 - sex_out_loss: 0.3116 - age_out_loss: 5.5627 - val_loss: 6.5057 - val_sex_out_loss: 0.2883 - val_age_out_loss: 6.2174\n","Epoch 145/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 5.8120 - sex_out_loss: 0.3093 - age_out_loss: 5.5027 - val_loss: 6.7308 - val_sex_out_loss: 0.2879 - val_age_out_loss: 6.4429\n","Epoch 146/200\n","15884/15884 [==============================] - 1s 76us/sample - loss: 5.7841 - sex_out_loss: 0.3100 - age_out_loss: 5.4741 - val_loss: 6.6089 - val_sex_out_loss: 0.2841 - val_age_out_loss: 6.3249\n","Epoch 147/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 5.6955 - sex_out_loss: 0.3084 - age_out_loss: 5.3870 - val_loss: 6.7028 - val_sex_out_loss: 0.2789 - val_age_out_loss: 6.4239\n","Epoch 148/200\n","15884/15884 [==============================] - 1s 69us/sample - loss: 5.8841 - sex_out_loss: 0.3129 - age_out_loss: 5.5711 - val_loss: 6.8946 - val_sex_out_loss: 0.2896 - val_age_out_loss: 6.6050\n","Epoch 149/200\n","15884/15884 [==============================] - 1s 67us/sample - loss: 5.7727 - sex_out_loss: 0.3108 - age_out_loss: 5.4620 - val_loss: 6.4519 - val_sex_out_loss: 0.2791 - val_age_out_loss: 6.1728\n","Epoch 150/200\n","15884/15884 [==============================] - 1s 67us/sample - loss: 5.7678 - sex_out_loss: 0.3060 - age_out_loss: 5.4618 - val_loss: 6.7018 - val_sex_out_loss: 0.2835 - val_age_out_loss: 6.4183\n","Epoch 151/200\n","15884/15884 [==============================] - 1s 67us/sample - loss: 5.7223 - sex_out_loss: 0.3063 - age_out_loss: 5.4160 - val_loss: 6.5794 - val_sex_out_loss: 0.2851 - val_age_out_loss: 6.2943\n","Epoch 152/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 5.7507 - sex_out_loss: 0.3119 - age_out_loss: 5.4388 - val_loss: 6.6915 - val_sex_out_loss: 0.2779 - val_age_out_loss: 6.4136\n","Epoch 153/200\n","15884/15884 [==============================] - 1s 67us/sample - loss: 5.7818 - sex_out_loss: 0.3108 - age_out_loss: 5.4711 - val_loss: 6.7060 - val_sex_out_loss: 0.2841 - val_age_out_loss: 6.4220\n","Epoch 154/200\n","15884/15884 [==============================] - 1s 67us/sample - loss: 5.7913 - sex_out_loss: 0.3012 - age_out_loss: 5.4901 - val_loss: 6.6363 - val_sex_out_loss: 0.2903 - val_age_out_loss: 6.3460\n","Epoch 155/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 5.7254 - sex_out_loss: 0.3106 - age_out_loss: 5.4148 - val_loss: 6.4701 - val_sex_out_loss: 0.2852 - val_age_out_loss: 6.1849\n","Epoch 156/200\n","15884/15884 [==============================] - 1s 77us/sample - loss: 5.7464 - sex_out_loss: 0.3064 - age_out_loss: 5.4400 - val_loss: 6.5757 - val_sex_out_loss: 0.2878 - val_age_out_loss: 6.2879\n","Epoch 157/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 5.7730 - sex_out_loss: 0.3160 - age_out_loss: 5.4570 - val_loss: 6.5214 - val_sex_out_loss: 0.2833 - val_age_out_loss: 6.2381\n","Epoch 158/200\n","15884/15884 [==============================] - 1s 67us/sample - loss: 5.8329 - sex_out_loss: 0.3102 - age_out_loss: 5.5227 - val_loss: 6.6869 - val_sex_out_loss: 0.2826 - val_age_out_loss: 6.4043\n","Epoch 159/200\n","15884/15884 [==============================] - 1s 67us/sample - loss: 5.7120 - sex_out_loss: 0.3093 - age_out_loss: 5.4026 - val_loss: 6.8534 - val_sex_out_loss: 0.2806 - val_age_out_loss: 6.5727\n","Epoch 160/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 5.8674 - sex_out_loss: 0.3112 - age_out_loss: 5.5562 - val_loss: 7.1051 - val_sex_out_loss: 0.2777 - val_age_out_loss: 6.8274\n","Epoch 161/200\n","15884/15884 [==============================] - 1s 69us/sample - loss: 5.7581 - sex_out_loss: 0.2994 - age_out_loss: 5.4587 - val_loss: 6.6399 - val_sex_out_loss: 0.2805 - val_age_out_loss: 6.3593\n","Epoch 162/200\n","15884/15884 [==============================] - 1s 67us/sample - loss: 5.6980 - sex_out_loss: 0.3101 - age_out_loss: 5.3880 - val_loss: 6.6515 - val_sex_out_loss: 0.2856 - val_age_out_loss: 6.3659\n","Epoch 163/200\n","15884/15884 [==============================] - 1s 67us/sample - loss: 5.7365 - sex_out_loss: 0.3054 - age_out_loss: 5.4311 - val_loss: 6.6367 - val_sex_out_loss: 0.2867 - val_age_out_loss: 6.3500\n","Epoch 164/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 5.7270 - sex_out_loss: 0.3065 - age_out_loss: 5.4205 - val_loss: 6.7252 - val_sex_out_loss: 0.2814 - val_age_out_loss: 6.4438\n","Epoch 165/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 5.7244 - sex_out_loss: 0.3057 - age_out_loss: 5.4188 - val_loss: 6.4935 - val_sex_out_loss: 0.2827 - val_age_out_loss: 6.2108\n","Epoch 166/200\n","15884/15884 [==============================] - 1s 81us/sample - loss: 5.6422 - sex_out_loss: 0.2974 - age_out_loss: 5.3447 - val_loss: 6.4834 - val_sex_out_loss: 0.2816 - val_age_out_loss: 6.2019\n","Epoch 167/200\n","15884/15884 [==============================] - 1s 69us/sample - loss: 5.7653 - sex_out_loss: 0.3103 - age_out_loss: 5.4550 - val_loss: 6.5243 - val_sex_out_loss: 0.2825 - val_age_out_loss: 6.2418\n","Epoch 168/200\n","15884/15884 [==============================] - 1s 70us/sample - loss: 5.7414 - sex_out_loss: 0.3077 - age_out_loss: 5.4337 - val_loss: 6.8434 - val_sex_out_loss: 0.2839 - val_age_out_loss: 6.5595\n","Epoch 169/200\n","15884/15884 [==============================] - 1s 70us/sample - loss: 5.7172 - sex_out_loss: 0.3039 - age_out_loss: 5.4133 - val_loss: 6.7327 - val_sex_out_loss: 0.2916 - val_age_out_loss: 6.4411\n","Epoch 170/200\n","15884/15884 [==============================] - 1s 69us/sample - loss: 5.6827 - sex_out_loss: 0.3096 - age_out_loss: 5.3731 - val_loss: 6.7474 - val_sex_out_loss: 0.2787 - val_age_out_loss: 6.4687\n","Epoch 171/200\n","15884/15884 [==============================] - 1s 77us/sample - loss: 5.6061 - sex_out_loss: 0.3004 - age_out_loss: 5.3057 - val_loss: 6.5060 - val_sex_out_loss: 0.2883 - val_age_out_loss: 6.2177\n","Epoch 172/200\n","15884/15884 [==============================] - 1s 69us/sample - loss: 5.5926 - sex_out_loss: 0.3027 - age_out_loss: 5.2899 - val_loss: 6.5091 - val_sex_out_loss: 0.2814 - val_age_out_loss: 6.2278\n","Epoch 173/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 5.6167 - sex_out_loss: 0.3037 - age_out_loss: 5.3130 - val_loss: 6.8441 - val_sex_out_loss: 0.2808 - val_age_out_loss: 6.5633\n","Epoch 174/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 5.6580 - sex_out_loss: 0.3053 - age_out_loss: 5.3527 - val_loss: 6.6767 - val_sex_out_loss: 0.2806 - val_age_out_loss: 6.3961\n","Epoch 175/200\n","15884/15884 [==============================] - 1s 78us/sample - loss: 5.6145 - sex_out_loss: 0.3009 - age_out_loss: 5.3136 - val_loss: 6.5098 - val_sex_out_loss: 0.2778 - val_age_out_loss: 6.2320\n","Epoch 176/200\n","15884/15884 [==============================] - 1s 75us/sample - loss: 5.6156 - sex_out_loss: 0.3021 - age_out_loss: 5.3135 - val_loss: 6.6447 - val_sex_out_loss: 0.2801 - val_age_out_loss: 6.3646\n","Epoch 177/200\n","15884/15884 [==============================] - 1s 69us/sample - loss: 5.5608 - sex_out_loss: 0.2978 - age_out_loss: 5.2630 - val_loss: 7.0326 - val_sex_out_loss: 0.2857 - val_age_out_loss: 6.7469\n","Epoch 178/200\n"]},{"name":"stdout","output_type":"stream","text":["15884/15884 [==============================] - 1s 68us/sample - loss: 5.6133 - sex_out_loss: 0.2967 - age_out_loss: 5.3166 - val_loss: 6.7804 - val_sex_out_loss: 0.2840 - val_age_out_loss: 6.4964\n","Epoch 179/200\n","15884/15884 [==============================] - 1s 69us/sample - loss: 5.6865 - sex_out_loss: 0.3032 - age_out_loss: 5.3833 - val_loss: 6.5397 - val_sex_out_loss: 0.2832 - val_age_out_loss: 6.2565\n","Epoch 180/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 5.6064 - sex_out_loss: 0.3039 - age_out_loss: 5.3025 - val_loss: 6.5596 - val_sex_out_loss: 0.2845 - val_age_out_loss: 6.2751\n","Epoch 181/200\n","15884/15884 [==============================] - 1s 86us/sample - loss: 5.5104 - sex_out_loss: 0.3043 - age_out_loss: 5.2061 - val_loss: 6.6120 - val_sex_out_loss: 0.2760 - val_age_out_loss: 6.3360\n","Epoch 182/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 5.6483 - sex_out_loss: 0.3008 - age_out_loss: 5.3475 - val_loss: 6.5117 - val_sex_out_loss: 0.2829 - val_age_out_loss: 6.2287\n","Epoch 183/200\n","15884/15884 [==============================] - 1s 71us/sample - loss: 5.5565 - sex_out_loss: 0.2972 - age_out_loss: 5.2593 - val_loss: 7.0550 - val_sex_out_loss: 0.2790 - val_age_out_loss: 6.7760\n","Epoch 184/200\n","15884/15884 [==============================] - 1s 69us/sample - loss: 5.5588 - sex_out_loss: 0.2969 - age_out_loss: 5.2619 - val_loss: 6.5991 - val_sex_out_loss: 0.2903 - val_age_out_loss: 6.3087\n","Epoch 185/200\n","15884/15884 [==============================] - 2s 95us/sample - loss: 5.5853 - sex_out_loss: 0.2947 - age_out_loss: 5.2907 - val_loss: 6.6381 - val_sex_out_loss: 0.2787 - val_age_out_loss: 6.3593\n","Epoch 186/200\n","15884/15884 [==============================] - 1s 73us/sample - loss: 5.7557 - sex_out_loss: 0.3000 - age_out_loss: 5.4556 - val_loss: 6.8502 - val_sex_out_loss: 0.2789 - val_age_out_loss: 6.5713\n","Epoch 187/200\n","15884/15884 [==============================] - 1s 73us/sample - loss: 5.5758 - sex_out_loss: 0.2944 - age_out_loss: 5.2814 - val_loss: 7.1475 - val_sex_out_loss: 0.2808 - val_age_out_loss: 6.8667\n","Epoch 188/200\n","15884/15884 [==============================] - 1s 70us/sample - loss: 5.5643 - sex_out_loss: 0.3087 - age_out_loss: 5.2556 - val_loss: 6.9116 - val_sex_out_loss: 0.2816 - val_age_out_loss: 6.6300\n","Epoch 189/200\n","15884/15884 [==============================] - 1s 70us/sample - loss: 5.5716 - sex_out_loss: 0.2974 - age_out_loss: 5.2742 - val_loss: 6.5503 - val_sex_out_loss: 0.2756 - val_age_out_loss: 6.2747\n","Epoch 190/200\n","15884/15884 [==============================] - 1s 69us/sample - loss: 5.5096 - sex_out_loss: 0.3017 - age_out_loss: 5.2078 - val_loss: 6.5620 - val_sex_out_loss: 0.2839 - val_age_out_loss: 6.2781\n","Epoch 191/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 5.8086 - sex_out_loss: 0.2988 - age_out_loss: 5.5097 - val_loss: 6.6342 - val_sex_out_loss: 0.2804 - val_age_out_loss: 6.3539\n","Epoch 192/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 5.5110 - sex_out_loss: 0.2992 - age_out_loss: 5.2117 - val_loss: 7.0300 - val_sex_out_loss: 0.2845 - val_age_out_loss: 6.7455\n","Epoch 193/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 5.6249 - sex_out_loss: 0.3011 - age_out_loss: 5.3238 - val_loss: 7.0730 - val_sex_out_loss: 0.2919 - val_age_out_loss: 6.7811\n","Epoch 194/200\n","15884/15884 [==============================] - 1s 72us/sample - loss: 5.4952 - sex_out_loss: 0.3004 - age_out_loss: 5.1948 - val_loss: 6.5297 - val_sex_out_loss: 0.2803 - val_age_out_loss: 6.2494\n","Epoch 195/200\n","15884/15884 [==============================] - 1s 74us/sample - loss: 5.5088 - sex_out_loss: 0.3034 - age_out_loss: 5.2054 - val_loss: 6.6871 - val_sex_out_loss: 0.2851 - val_age_out_loss: 6.4020\n","Epoch 196/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 5.5050 - sex_out_loss: 0.2997 - age_out_loss: 5.2053 - val_loss: 6.5933 - val_sex_out_loss: 0.2849 - val_age_out_loss: 6.3084\n","Epoch 197/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 5.4654 - sex_out_loss: 0.2944 - age_out_loss: 5.1710 - val_loss: 6.6028 - val_sex_out_loss: 0.2830 - val_age_out_loss: 6.3198\n","Epoch 198/200\n","15884/15884 [==============================] - 1s 74us/sample - loss: 5.5306 - sex_out_loss: 0.3078 - age_out_loss: 5.2227 - val_loss: 6.4570 - val_sex_out_loss: 0.2820 - val_age_out_loss: 6.1750\n","Epoch 199/200\n","15884/15884 [==============================] - 1s 69us/sample - loss: 5.5794 - sex_out_loss: 0.2925 - age_out_loss: 5.2869 - val_loss: 6.5044 - val_sex_out_loss: 0.2756 - val_age_out_loss: 6.2289\n","Epoch 200/200\n","15884/15884 [==============================] - 1s 68us/sample - loss: 5.5496 - sex_out_loss: 0.2990 - age_out_loss: 5.2507 - val_loss: 6.6433 - val_sex_out_loss: 0.2866 - val_age_out_loss: 6.3566\n"]},{"data":{"text/plain":["<tensorflow.python.keras.callbacks.History at 0x7f8815162e48>"]},"execution_count":9,"metadata":{},"output_type":"execute_result"}],"source":["import random\n","random_id=random.random()\n","model.summary()\n","callbacks = [\n","    tf.keras.callbacks.EarlyStopping(patience=75, monitor='val_loss',restore_best_weights=True),\n","    tf.keras.callbacks.TensorBoard(log_dir='./logs/'+str(random_id))\n","]\n","model.fit(X_train, y_train, epochs=200,batch_size=240,validation_data=(X_valid,y_valid),callbacks=callbacks, shuffle=True)\n"]},{"cell_type":"code","execution_count":null,"metadata":{"trusted":true},"outputs":[],"source":["model_json = model.to_json()\n","with open('age_gender_model.json', 'w') as json_file:\n","    json_file.write(model_json)\n","model.save_weights('age_gender.h5')"]},{"cell_type":"markdown","metadata":{},"source":["## Evaluate Model"]},{"cell_type":"code","execution_count":null,"metadata":{"trusted":true},"outputs":[],"source":["model.evaluate(X_valid,y_valid)"]},{"cell_type":"code","execution_count":3,"metadata":{},"outputs":[],"source":["model.load_weights(os.path.join('saved_models', 'gender_model',\"age_gender.h5\"))"]},{"cell_type":"code","execution_count":5,"metadata":{},"outputs":[],"source":["face = cv2.imread(\"./data/utkface_aligned_cropped/brajesh.png\",cv2.IMREAD_COLOR)\n","face = cv2.cvtColor(face, cv2.COLOR_BGR2RGB)\n","face =cv2.resize(face, (32,32) )"]},{"cell_type":"code","execution_count":8,"metadata":{},"outputs":[],"source":["face = np.expand_dims(face, axis=0)"]},{"cell_type":"code","execution_count":9,"metadata":{},"outputs":[{"data":{"text/plain":["(1, 32, 32, 3)"]},"execution_count":9,"metadata":{},"output_type":"execute_result"}],"source":["face.shape"]},{"cell_type":"code","execution_count":10,"metadata":{},"outputs":[{"data":{"text/plain":["[array([[0.48005745]], dtype=float32), array([[21.94534]], dtype=float32)]"]},"execution_count":10,"metadata":{},"output_type":"execute_result"}],"source":["model.predict(face)"]},{"cell_type":"code","execution_count":null,"metadata":{"trusted":true},"outputs":[],"source":["p_id=2\n","imshow(X_valid[p_id])\n","print(y_valid[0][p_id],y_valid[1][p_id])\n","print(model.predict([[X_valid[p_id]]]))"]}],"metadata":{"interpreter":{"hash":"29c338b1dda19cb1dbd2d91f2577f6bdb48de9ab64d115f18be145e310c610bf"},"kernelspec":{"display_name":"Python 3.8.10 64-bit ('venv': venv)","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":"3.8.10"}},"nbformat":4,"nbformat_minor":1}