Sigmoid function output

Question

I have the following neural network model.

nn_classifier = Sequential()
nn_classifier.add(Dense(output_dim = 16 ,activation='relu',input_dim = 13))
nn_classifier.add(Dense(output_dim = 16,activation='relu'))
nn_classifier.add(Dense(output_dim = 1, activation = 'sigmoid'))

nn_classifier.compile(optimizer = 'sgd', loss = 'binary_crossentropy', metrics=[tf.keras.metrics.BinaryAccuracy(threshold=0.5)])
model=nn_classifier.fit(X_train, Y_train ,validation_split=0.33, batch_size = 10, nb_epoch = 100)
Y_pred = nn_classifier.predict(X_test)

As I have used the sigmoid function in my output layer, I was expecting the predicted values (Y_pred) are either 0 or 1. But I get some decimal values. Is my understanding wrong?

score 0 · Accepted Answer · answered Jan 20 '21 at 17:42

0

Sigmoid always gives a value in [0,1] you need to round the value that means fix a threshold if it is higher than threshold then 1 else 0.

answered Jan 20 '21 at 17:42

AKSHAY KUMAR RAY

388
1
9

Got it. I thought sigmoid will give either 0 or 1. So its giving between 0 to1. Is there way to setup a threshold value in the model to give 0 if the predicted value is less than 0.5 and 1 otherwise. – user6771624 Jan 20 '21 at 17:46
you can use round function the threshold will be 0.5 if less than that then 0 else 1. – AKSHAY KUMAR RAY Jan 20 '21 at 17:48
Can you show me a piece of code where round function is used? – user6771624 Jan 20 '21 at 17:56
np.argmax(Y_pred,axis=1) this should work. – AKSHAY KUMAR RAY Jan 20 '21 at 18:10

Sigmoid function output

1 Answers1