Teacher Forcing으로 모델 학습

Keras로 배우는 Machine Translation

Thushan Ganegedara

Data Scientist and Author

모델 학습 자세히 보기

모델 학습에는 다음이 필요합니다:
- 손실 함수(예: categorical crossentropy)
- 옵티마이저(예: Adam)

모델 학습 자세히 보기

손실 계산에 필요한 항목:
- 입력으로 생성된 확률 예측([batch_size, seq_len, vocab_size])
  - 예: [[0.11,...,0.81,0.04], [0.05,...,0.01, 0.93], ..., [0.78,..., 0.03,0.01]]
- 실제 원-핫 인코딩된 프랑스어 타깃([batch_size, seq_len, vocab_size])
  - 예: [[0, ..., 1, 0], [0, ..., 0, 1],..., [0, ..., 1, 0]]
- 크로스엔트로피: 타깃과 예측 단어의 차이
손실은 옵티마이저로 전달되어, 손실을 최소화하도록 모델 파라미터를 갱신합니다

Teacher Forcing으로 모델 학습

n_epochs, bsize = 3, 250
for ei in range(n_epochs):
  for i in range(0,data_size,bsize):

    # Encoder inputs, decoder inputs and outputs
    en_x = sents2seqs('source', en_text[i:i+bsize], onehot=True, reverse=True)
    de_xy = sents2seqs('target', fr_text[i:i+bsize], onehot=True)

    # Separating decoder inputs and outputs
    de_x = de_xy[:,:-1,:]    
    de_y = de_xy[:,1:,:]

    # Training and evaulating on a single batch
    nmt_tf.train_on_batch([en_x,de_x], de_y)    
    res = nmt_tf.evaluate([en_x,de_x], de_y, batch_size=bsize, verbose=0)
    print("{} => Train Loss:{}, Train Acc: {}".format(ei+1,res[0], res[1]*100.0))

배열 슬라이싱 자세히 보기

    de_x = de_xy[:,:-1,:]    
    de_y = de_xy[:,1:,:]

배열 슬라이싱

학습/검증 데이터 생성

train_size, valid_size = 800, 200
# Creating data indices
inds = np.arange(len(en_text))
np.random.shuffle(inds)


# Separating train and valid indices
train_inds = inds[:train_size]
valid_inds = inds[train_size:train_size+valid_size]


# Extracting train and valid data
tr_en = [en_text[ti] for ti in train_inds]
tr_fr = [fr_text[ti] for ti in train_inds]

v_en = [en_text[vi] for vi in valid_inds]
v_fr = [fr_text[vi] for vi in valid_inds]

print('Training (EN):\n', tr_en[:2], '\nTraining (FR):\n', tr_fr[:2])
print('\nValid (EN):\n', tr_en[:2], '\nValid (FR):\n', tr_fr[:2])

Valid (EN): ['she saw that rusty blue truck .', 'france is usually snowy during autumn , ...'] 
Valid (FR): ['elle a vu ce camion bleu rouillé .', 'la france est généralement enneigée en automne ...']

검증을 포함한 학습

for ei in range(n_epochs):
  for i in range(0,train_size,bsize):    
    en_x = sents2seqs('source', tr_en[i:i+bsize], onehot=True, reverse=True)
    de_xy = sents2seqs('target', tr_fr[i:i+bsize], onehot=True)
    de_x, de_y = de_xy[:,:-1,:], de_xy[:,1:,:]
    nmt_tf.train_on_batch([en_x, de_x], de_y)

  v_en_x = sents2seqs('source', v_en, onehot=True, reverse=True)
  v_de_xy = sents2seqs('target', v_fr, onehot=True)
  v_de_x, v_de_y = v_de_xy[:,:-1,:], v_de_xy[:,1:,:]

  res = nmt_tf.evaluate([v_en_x, v_de_x], v_de_y, batch_size=valid_size, verbose=0)
  print("Epoch {} => Loss:{}, Val Acc: {}".format(ei+1,res[0], res[1]*100.0))

Epoch 1 => Loss:4.784221172332764, Val Acc: 1.4999999664723873
Epoch 2 => Loss:4.716882228851318, Val Acc: 44.458332657814026
Epoch 3 => Loss:4.63267183303833, Val Acc: 47.333332896232605

학습해 봅시다!

Keras로 배우는 Machine Translation