gender-model.py

input = Input(shape=(224, 224, 3))
cnn1 = Conv2D(36, kernel_size=3, activation='relu')(input)
cnn1 = MaxPool2D(pool_size=3, strides=2)(cnn1)
cnn2 = Conv2D(64, kernel_size=3, activation='relu')(cnn1)
cnn2 = MaxPool2D(pool_size=3, strides=2)(cnn2)
cnn3 = Conv2D(128, kernel_size=3, activation='relu')(cnn2)
cnn3 = MaxPool2D(pool_size=3, strides=2)(cnn3)
cnn4 = Conv2D(256, kernel_size=3, activation='relu')(cnn3)
cnn4 = MaxPool2D(pool_size=3, strides=2)(cnn4)
cnn5 = Conv2D(512, kernel_size=3, activation='relu')(cnn4)
cnn5 = MaxPool2D(pool_size=3, strides=2)(cnn5)
dense = Flatten()(cnn5)
dense = Dropout(0.2)(dense)
dense = Dense(512, activation='relu')(dense)
dense = Dense(512, activation='relu')(dense)
output = Dense(1, activation='sigmoid', name='gender')(dense)
sex_model = Model(input, output)
sex_model.compile(optimizer=Adam(learning_rate=0.0001), loss='binary_crossentropy', metrics=['accuracy'])