Mendefinisikan decoder

Penerjemahan Mesin dengan Keras

Thushan Ganegedara

Data Scientist and Author

Model encoder-decoder

Model encoder-decoder

Model encoder-decoder

Ulangi context vector dari encoder sebanyak N kali

Repeat vector pada model encoder-decoder

Layer RepeatVector:

Satu argumen untuk panjang urutan output
Input berbentuk (batch_size, ukuran input) (mis. input ukuran 2 x 3)
Output berbentuk (batch_size, panjang urutan, ukuran input) (mis. output ukuran 2 x 3 x 3)

Fungsionalitas RepeatVector

from tensorflow.keras.layers import RepeatVector
rep = RepeatVector(5)

r_inp = Input(shape=(3,))
r_out = rep(r_inp)

repeat_model = Model(inputs=r_inp, outputs=r_out)

rep = RepeatVector(5)
r_out = rep(r_inp)

r_out = RepeatVector(5)(r_inp)

x = np.array([[0,1,2],[3,4,5]])
y = repeat_model.predict(x)
print('x.shape = ',x.shape,'\ny.shape = ',y.shape)

x.shape =  (2, 3) 
y.shape =  (2, 5, 3)

Mendefinisikan decoder

de_inputs = RepeatVector(fr_len)(en_state)

decoder_gru = GRU(hsize, return_sequences=True)

Mengatur state awal decoder

gru_outputs = decoder_gru(de_inputs, initial_state=en_state)

enc_dec = Model(inputs=en_inputs, outputs=gru_outputs)

Penerjemahan Mesin dengan Keras