실행 시간을 단축하고자 num_of_episodes와 timesteps_per_episode의 값을 10으로 설정했지만, 훈련을 좀 더 정교하게 하려면 100 이상의 값을 설정해 주어야 합니다.
다음은 앞서 생성한 네트워크를 출력한 결과입니다.
Model: "sequential"
________________________________________________________________
Layer (type) Output Shape Param #
================================================================
embedding (Embedding) (None, 1, 10) 5000
________________________________________________________________
reshape (Reshape) (None, 10) 0
________________________________________________________________
dense (Dense) (None, 50) 550
________________________________________________________________
dense_1 (Dense) (None, 50) 2550
________________________________________________________________
dense_2 (Dense) (None, 50) 2550
________________________________________________________________
dense_3 (Dense) (None, 6) 306
================================================================
Total params: 10,956
Trainable params: 10,956
Non-trainable params: 0
________________________________________________________________