2
$\begingroup$

I'm trying to solve timeseries classification problem. That's my model:

model = Sequential()
model.add(Bidirectional(LSTM(32, activation='tanh', return_sequences=True), input_shape=(1000, 1), merge_mode='concat'))
model.add(TimeDistributed(Dense(4, activation='softmax')))

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

I've got strange plots of my training. Training is unstable either for training or validation. Is it wrong?enter image description here

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.