How To Use Keras Predict_proba To Output 2 Columns Of Probability?

I use this code to predict the probability of 0 and 1 in x_test, but the result is only one column of probability. I really don’t know whether the probability of this column is t

Solution 1:

First, you need to convert y_train to one-hot encoding by

from sklearn.preprocessingimportLabelEncoderfrom keras.utilsimport np_utils

encoder = LabelEncoder()
encoder.fit(y_train)
encoded_y = encoder.transform(y_train)
y_train = np_utils.to_categorical(encoded_y)

running this code, y_train will become

array([[1., 0.],
       [1., 0.],
       [1., 0.],
       [1., 0.],
       [1., 0.],
       [1., 0.],
       [1., 0.],
       [0., 1.],
       [0., 1.],
       [0., 1.],
       [0., 1.]], dtype=float32)

Secondly, you need to change the output layer to

model.add(Dense(2, activation='softmax'))

with these two modifications, you will get the desired output.