with Giovanni

build

def build_model(vocab_size, embedding_dim, rnn_units):

model=tf.keras.Sequential([

tf.keras.layers.Embedding(vocab_size, embedding_dim),

tf.keras.layers.LSTM(rnn_units, return_sequences=True),

tf.keras.layers.Dense(vocab_size)])

return model

charset_size=len(char_array)

embedding_dim=256

rnn_units=512

tf.random.set_seed(1)

model=build_model(vocab_size=charset_size, embedding_dim=embedding_dim, rnn_units=rnn_units)

model.summary()

Model: "sequential"

_________________________________________________________________

Layer (type) Output Shape Param #

=================================================================

embedding (Embedding) (None, None, 256) 20480

_________________________________________________________________

lstm (LSTM) (None, None, 512) 1574912

_________________________________________________________________

dense (Dense) (None, None, 80) 41040

=================================================================

Total params: 1,636,432

Trainable params: 1,636,432

Non-trainable params: 0

_________________________________________________________________

LSTM은 랭크 3 크기의 (None, None, 512)를 출력한다

첫번째 차원은 batch, 두 번째 차원은 출력 시퀀스 길이(40), 세번째 차원은 은닉 유닛의 개수에 해 당

최종 출력 또한 (None, None, 80)으로 랭크 3의 텐서가 출력된다.

compile & fit

model.compile(

optimizer='adam',

loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True))

model.fit(ds, epochs=20)

Language modeling 2_RNN model