Error when checking target: expected dense_2 to have 2 dimensions, but got array with shape (1, 1226, 2)

Here is the code that I am trying to run:

y = Df[['label']]

y_train = np.column_stack((y_train['label']))

y_test = np.column_stack((y_test['label']))

data_dim = 18
timesteps = 1
num_classes = 2

model = Sequential()

model.add(LSTM(19, return_sequences=True,
               input_shape=(timesteps, data_dim)))  
model.add(LSTM(19, return_sequences=True))  
model.add(LSTM(19))  # return a single vector of dimension 30
model.add(Dense(1, activation='softmax'))

model.compile(loss='binary_crossentropy',
              optimizer='rmsprop',
              metrics=['accuracy'])

model.summary()

model.fit(X_train, y_train, batch_size = 400, epochs = 20, verbose = 

accuracy_train = accuracy_score(y_train, model.predict(X_train))
accuracy_test = accuracy_score(y_test, model.predict(X_test))
print('\nTrain Accuracy:{: .2f}%'.format(accuracy_train*100))
print('Test Accuracy:{: .2f}%'.format(accuracy_test*100))

I am getting the following error:

Error when checking target: expected dense_1 to have shape (1,) but got array with shape (1226,)

Model Summary:

_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
lstm_37 (LSTM)               (None, 1, 19)             2888      
_________________________________________________________________
lstm_38 (LSTM)               (None, 1, 19)             2964      
_________________________________________________________________
lstm_39 (LSTM)               (None, 19)                2964      
_________________________________________________________________
dense_13 (Dense)             (None, 1)                 20        
=================================================================
Total params: 8,836
Trainable params: 8,836
Non-trainable params: 0

X_train:

[[[-0.02444971  0.02444971  1.          1.          1.
    1.         -1.          2.          3.          4.
    3.          1.          0.          0.06938705 -0.04329106
    0.02458854  0.06025883  0.01439807]]]
[[[ 0.00733477  0.02033006 -1.          1.          1.
    1.         -1.          0.          1.          2.
    1.          0.          0.          0.03837079 -0.00829683
   -0.00734757  0.00985466 -0.04543226]]]

y_train:

[[ 1  1 -1 ...  1  1 -1]]

Solution 1:

It seems you are doing binary classification. Therefore, there are a few things that need to be fixed:

1. The y_train should consists of zeros and ones. The following would convert all the -1 labels to zeros:

   y_train = (y_train == 1).astype('float32')
   

2. Reshape the labels to have a shape of (n_samples, 1):

   y_train = y_train.reshape(-1, 1)
   

3. Use 'sigmoid' as the activation of last layer:

   model.add(Dense(1, activation='sigmoid'))