Tukukkk

I try to make a LSTM model continue were its last run left off. All compiles fine till I try to fit the network. Then it gives an error:

ValueError: Error when checking target: expected dense_29 to have 3 dimensions, but got array with shape (672, 1)

I checked various articles such as this and this but I don't see what is wrong in my code.

from keras import Sequential

from keras.preprocessing.sequence import pad_sequences

from sklearn.model_selection import train_test_split

from keras.models import Sequential,Model

from keras.layers import LSTM, Dense, Bidirectional, Input,Dropout,BatchNormalization

from keras import backend as K

from keras.engine.topology import Layer

from keras import initializers, regularizers, constraints



from keras.callbacks import ModelCheckpoint

from keras.models import load_model

import os.path

import os

filepath="Train-weights.best.hdf5"

act = 'relu'



model = Sequential()

model.add(BatchNormalization(input_shape=(10, 128)))

model.add(Bidirectional(LSTM(128, dropout=0.5, activation=act, return_sequences=True)))

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



if (os.path.exists(filepath)):

   print("extending training of previous run")

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

   with open('model_architecture.json', 'r') as f:

      model = model_from_json(f.read())

   model.load_weights(filepath)

else:

   print("First run")      

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

   model.fit(x_train, y_train, validation_data=(x_val, y_val), epochs=100, batch_size=32, callbacks=callbacks_list, verbose=2)

   model.save_weights(filepath)

   with open('model_architecture.json', 'w') as f:

       f.write(model.to_json())



 checkpoint = ModelCheckpoint(filepath, monitor='val_acc', verbose=1, save_best_only=True, mode='max')

 callbacks_list = [checkpoint]



 model.fit(x_train, y_train, validation_data=(x_val, y_val), epochs=100, batch_size=32, callbacks=callbacks_list, verbose=0)

Try model.summary() and you would see that the output shape of last layer (i.e. the Dense layer) in the network is (None, 10, 1). Therefore, the labels you provide to the model (i.e. y_train) must also have a shape of (num_samples, 10, 1).

If the output shape (None, 10, 1) is not what you wanted (e.g. you want (None, 1) as the output shape of your model) then you need to modify your model definition. One simple modification to achieve that is removing the return_sequences=True argument from the LSTM layer.