techwithshadab

# splitting path of all images into male and female
train_male_image = []
train_female_image = []
for each in train_image_name:
    if each.split('/')[7] in train_male_data['Filename'].values:
        train_male_image.append(each)
    else:
        train_female_image.append(each)

techwithshadab

optimizer = RMSprop(lr=1e-4)
objective = 'binary_crossentropy'


def malefemale():
    
    model = Sequential()

    model.add(Convolution2D(32, 3, 3, border_mode='same', input_shape=(3, ROWS, COLS), activation='relu'))
    model.add(Convolution2D(32, 3, 3, border_mode='same', activation='relu'))

techwithshadab

nb_epoch = 10
batch_size = 16
labs = train_data.iloc[:,1].values.tolist()

## Callback for loss logging per epoch
class LossHistory(Callback):
    def on_train_begin(self, logs={}):
        self.losses = []
        self.val_losses = []
        

techwithshadab

model.fit(train, labs, batch_size=batch_size, epochs=nb_epoch,
              validation_split=0.25, verbose=0, shuffle=True, callbacks=[history, early_stopping])

techwithshadab

predictions = model.predict(test, verbose=0)
predictions

techwithshadab

loss = history.losses
val_loss = history.val_losses

plt.xlabel('Epochs')
plt.ylabel('Loss')
plt.title('VGG-16 Loss Trend')
plt.plot(loss, 'blue', label='Training Loss')
plt.plot(val_loss, 'green', label='Validation Loss')
plt.xticks(range(0,nb_epoch)[0::2])
plt.legend()

techwithshadab

for i in range(0,6):
    if predictions[i, 0] >= 0.5: 
        print('I am {:.2%} sure this is a Female'.format(predictions[i][0]))
    else: 
        print('I am {:.2%} sure this is a Male'.format(1-predictions[i][0]))
        
    plt.imshow(test[i].T)
    plt.show()

techwithshadab

model.summary()

techwithshadab

sudo apt update
sudo apt upgrade
# Install pip
sudo apt install python-pip
# Install mlflow
pip install mlflow
# Check version of mlflow installed
mlflow --version

techwithshadab

curl -I http://172.18.x.x:5000