소스코드 원본 링크 : https://pytorch.org/tutorials/intermediate/seq2seq_translation_tutorial.html
Preparing Training Data
모델을 학습시키기 위해서 각 pair 마다 입력 텐서와 타겟 텐서가 필요하다.
def indexesFromSentence(lang, sentence):
return [lang.word2index[word] for word in sentence.split(' ')]
def tensorFromSentence(lang, sentence):
indexes = indexesFromSentence(lang, sentence)
indexes.append(EOS_token)
return torch.tensor(indexes, dtype=torch.long, device=device).view(-1, 1)
def tensorsFromPair(pair):
input_tensor = tensorFromSentence(input_lang, pair[0])
target_tensor = tensorFromSentence(output_lang, pair[1])
return (input_tensor, target_tensor)
Training the Model
모델을 학습시키기 위해서 인코더를 통해 입력 문장을 넣고 모든 출력과 마지막 hidden state를 저장한다. 그 다음 디코더가 첫번째 입력과 hidden state를 받을 때, 입력은 SOS 토큰을 받고 hidden state는 인코더의 마지막 hidden state를 받는다.
“Teacher forcing”은 매번 다음 입력으로 디코더의 예측을 사용하는 것이 아닌 실제 타겟 출력을 사용하는 방식이다. teacher forcing은 빠르게 수렴할 수 있지만 네트워크 학습이 잘 이뤄지지 않았을 때 불안정성을 보인다. teacher forcing을 랜덤 변수와 teacher_forcing_ratio를 통해 무작위로 사용한다.
디코더의 다음 입력을 타겟으로 feed 할 것인지 또는 디코더 자체의 예측값을 feed 할 것인지는 teacher_forcing에 따라 결정된다.
teacher_forcing_ratio = 0.5
def train(input_tensor, target_tensor, encoder, decoder, encoder_optimizer, decoder_optimizer, criterion, max_length=MAX_LENGTH):
encoder_hidden = encoder.initHidden()
encoder_optimizer.zero_grad()
decoder_optimizer.zero_grad()
input_length = input_tensor.size(0)
target_length = target_tensor.size(0)
encoder_outputs = torch.zeros(max_length, encoder.hidden_size, device=device)
loss = 0
for ei in range(input_length):
encoder_output, encoder_hidden = encoder(
input_tensor[ei], encoder_hidden)
encoder_outputs[ei] = encoder_output[0, 0]
decoder_input = torch.tensor([[SOS_token]], device=device)
decoder_hidden = encoder_hidden
use_teacher_forcing = True if random.random() < teacher_forcing_ratio else False
if use_teacher_forcing:
# Teacher forcing: Feed the target as the next input
for di in range(target_length):
decoder_output, decoder_hidden, decoder_attention = decoder(
decoder_input, decoder_hidden, encoder_outputs)
loss += criterion(decoder_output, target_tensor[di])
decoder_input = target_tensor[di] # Teacher forcing
else:
# Without teacher forcing: use its own predictions as the next input
for di in range(target_length):
decoder_output, decoder_hidden, decoder_attention = decoder(
decoder_input, decoder_hidden, encoder_outputs)
topv, topi = decoder_output.topk(1)
decoder_input = topi.squeeze().detach() # detach from history as input
loss += criterion(decoder_output, target_tensor[di])
if decoder_input.item() == EOS_token:
break
loss.backward()
encoder_optimizer.step()
decoder_optimizer.step()
return loss.item() / target_length
Helper Function
import time
import math
def asMinutes(s):
m = math.floor(s / 60)
s -= m * 60
return '%dm %ds' % (m, s)
def timeSince(since, percent):
now = time.time()
s = now - since
es = s / (percent)
rs = es - s
return '%s (- %s)' % (asMinutes(s), asMinutes(rs))
학습 프로세스
- Start a timer
- Initialize optimizers and criterion
- Create set of training pairs
- Start empty losses array for plotting
Then we call train many times and occasionally print the progress (% of examples, time so far, estimated time) and average loss.
SGD optimizer 와 NLLLoss를 사용하였다.
def trainIters(encoder, decoder, n_iters, print_every=1000, plot_every=100, learning_rate=0.01):
start = time.time()
plot_losses = []
print_loss_total = 0 # Reset every print_every
plot_loss_total = 0 # Reset every plot_every
encoder_optimizer = optim.SGD(encoder.parameters(), lr=learning_rate)
decoder_optimizer = optim.SGD(decoder.parameters(), lr=learning_rate)
training_pairs = [tensorsFromPair(random.choice(pairs))
for i in range(n_iters)]
criterion = nn.NLLLoss()
for iter in range(1, n_iters + 1):
training_pair = training_pairs[iter - 1]
input_tensor = training_pair[0]
target_tensor = training_pair[1]
loss = train(input_tensor, target_tensor, encoder,
decoder, encoder_optimizer, decoder_optimizer, criterion)
print_loss_total += loss
plot_loss_total += loss
if iter % print_every == 0:
print_loss_avg = print_loss_total / print_every
print_loss_total = 0
print('%s (%d %d%%) %.4f' % (timeSince(start, iter / n_iters),
iter, iter / n_iters * 100, print_loss_avg))
if iter % plot_every == 0:
plot_loss_avg = plot_loss_total / plot_every
plot_losses.append(plot_loss_avg)
plot_loss_total = 0
showPlot(plot_losses)
Plotting Results
import matplotlib.pyplot as plt
plt.switch_backend('agg')
import matplotlib.ticker as ticker
import numpy as np
def showPlot(points):
plt.figure()
fig, ax = plt.subplots()
# this locator puts ticks at regular intervals
loc = ticker.MultipleLocator(base=0.2)
ax.yaxis.set_major_locator(loc)
plt.plot(points)
Evaluation
평가 (Evaluation) 는 학습 (training) 과 비슷하다. 하지만 평가 단계는 타겟이 없어서 매 단계마다 디코더의 예측을 feed 한다. 매번 단어를 예측하고 출력 string에 그것을 더한다. 그러다가 EOS 토큰을 예측하게 되면 반복을 멈춘다. 그리고 나중에 display를 위해서 디코더의 어텐션 output들도 저장한다.
def evaluate(encoder, decoder, sentence, max_length=MAX_LENGTH):
with torch.no_grad():
input_tensor = tensorFromSentence(input_lang, sentence)
input_length = input_tensor.size()[0]
encoder_hidden = encoder.initHidden()
encoder_outputs = torch.zeros(max_length, encoder.hidden_size, device=device)
for ei in range(input_length):
encoder_output, encoder_hidden = encoder(input_tensor[ei],
encoder_hidden)
encoder_outputs[ei] += encoder_output[0, 0]
decoder_input = torch.tensor([[SOS_token]], device=device) # SOS
decoder_hidden = encoder_hidden
decoded_words = []
decoder_attentions = torch.zeros(max_length, max_length)
for di in range(max_length):
decoder_output, decoder_hidden, decoder_attention = decoder(
decoder_input, decoder_hidden, encoder_outputs)
decoder_attentions[di] = decoder_attention.data
topv, topi = decoder_output.data.topk(1)
if topi.item() == EOS_token:
decoded_words.append('<EOS>')
break
else:
decoded_words.append(output_lang.index2word[topi.item()])
decoder_input = topi.squeeze().detach()
return decoded_words, decoder_attentions[:di + 1]
학습 set으로 부터 랜덤 문장들을 평가하고 입력,타겟 그리고 output 들을 print 해준다.
def evaluateRandomly(encoder, decoder, n=10):
for i in range(n):
pair = random.choice(pairs)
print('>', pair[0])
print('=', pair[1])
output_words, attentions = evaluate(encoder, decoder, pair[0])
output_sentence = ' '.join(output_words)
print('<', output_sentence)
print('')
Training And Evaluating
hidden_size = 256
encoder1 = EncoderRNN(input_lang.n_words, hidden_size).to(device)
attn_decoder1 = AttnDecoderRNN(hidden_size, output_lang.n_words, dropout_p=0.1).to(device)
trainIters(encoder1, attn_decoder1, 75000, print_every=5000)
Visualizing Attention
attention output을 시각화하여 네트워크가 매 단계에서 어느 부분에 ‘focus’ 중 인지 확인할 수 있다.
output_words, attentions = evaluate(
encoder1, attn_decoder1, "je suis trop froid .")
plt.matshow(attentions.numpy())
비슷한 방식으로 여러 입력 문장에 대해 attention output을 시각화하고 확인할 수 있다.
def showAttention(input_sentence, output_words, attentions):
# Set up figure with colorbar
fig = plt.figure()
ax = fig.add_subplot(111)
cax = ax.matshow(attentions.numpy(), cmap='bone')
fig.colorbar(cax)
# Set up axes
ax.set_xticklabels([''] + input_sentence.split(' ') +
['<EOS>'], rotation=90)
ax.set_yticklabels([''] + output_words)
# Show label at every tick
ax.xaxis.set_major_locator(ticker.MultipleLocator(1))
ax.yaxis.set_major_locator(ticker.MultipleLocator(1))
plt.show()
def evaluateAndShowAttention(input_sentence):
output_words, attentions = evaluate(
encoder1, attn_decoder1, input_sentence)
print('input =', input_sentence)
print('output =', ' '.join(output_words))
showAttention(input_sentence, output_words, attentions)
evaluateAndShowAttention("elle a cinq ans de moins que moi .")
evaluateAndShowAttention("elle est trop petit .")
evaluateAndShowAttention("je ne crains pas de mourir .")
evaluateAndShowAttention("c est un jeune directeur plein de talent .")
