def validate(model,
dev_data,
vocab_src,
vocab_tgt,
epoch,
config,
direction=None):
model.eval()
device = torch.device(
"cpu") if config["device"] == "cpu" else torch.device("cuda:0")
with torch.no_grad():
model_hypotheses = []
references = []
val_dl = DataLoader(dev_data,
batch_size=config["batch_size_eval"],
shuffle=False,
num_workers=4)
val_dl = BucketingParallelDataLoader(val_dl)
for sentences_x, sentences_y in val_dl:
if direction == None or direction == "xy":
x_in, _, x_mask, x_len = create_batch(sentences_x, vocab_src,
device)
x_mask = x_mask.unsqueeze(1)
else:
x_in, _, x_mask, x_len = create_batch(sentences_y, vocab_src,
device)
x_mask = x_mask.unsqueeze(1)
enc_output, enc_hidden = model.encode(x_in, x_len)
dec_hidden = model.init_decoder(enc_output, enc_hidden)
raw_hypothesis = beam_search(model.decoder, model.emb_tgt,
model.generate_tm, enc_output,
dec_hidden, x_mask, vocab_tgt.size(),
vocab_tgt[SOS_TOKEN],
vocab_tgt[EOS_TOKEN],
vocab_tgt[PAD_TOKEN], config)
hypothesis = batch_to_sentences(raw_hypothesis, vocab_tgt)
model_hypotheses += hypothesis.tolist()
if direction == None or direction == "xy":
references += sentences_y.tolist()
else:
references += sentences_x.tolist()
save_hypotheses(model_hypotheses, epoch, config)
model_hypotheses, references = clean_sentences(model_hypotheses,
references, config)
bleu = compute_bleu(model_hypotheses, references, epoch, config,
direction)
return bleu
(
'https://master.dl.sourceforge.net/project/autshumato/Corpora/ENG-NSO.Release.zip', # url
'lcontent.DACB.DataVirVrystellingOpWeb.(eng-nso).nso.1.0.0.CAM.2010-09-23.txt', # file_name
'sepedi/sepedi.txt', # output_name
5301, # lines_to_remove (constitution with poor formatting)
)
]
for url, file_name, output_name, lines_to_remove in datasets:
print('processing:', url)
r = requests.get(url)
zip = zipfile.ZipFile(BytesIO(r.content))
corpus = zip.open(file_name)
corpus = corpus.read()
corpus = corpus.decode('utf-8')
sentences = corpus.splitlines()
sentences = utils.clean_sentences(sentences, min_length=3, lines_to_remove=lines_to_remove)
output_file_name = os.path.join(args.output_dir, output_name)
with open(output_file_name, 'w', encoding='utf-8') as f:
f.write('\n'.join(sentences))
print('total sentences in {}:'.format(output_name), len(sentences))
print('Autshumato datasets provided under Creative Commons Attribution Non-Commercial ShareAlike, '
'CTexT (Centre for Text Technology, North-West University), South Africa; '
'Department of Arts and Culture, South Africa. '
'http://autshumato.sourceforge.net/ and http://www.nwu.ac.za/ctext')
article = re.sub(r'\(\d*\)', '', article)
# remove extra whitespace
article = re.sub('\\s+', ' ', article)
# replace strange quote character
article = utils.normalize_quote_characters(article)
# discard articles with imbalanced quotes
if article.count('"') % 2 != 0:
continue
# split article into array of sentences
# regex help from https://stackoverflow.com/questions/11502598/how-to-match-something-with-regex-that-is-not-between-two-special-characters
sentences = re.split('(?<=\.|\!|\?) (?=(?:[^"]*"[^"]*")*[^"]*\Z)',
article)
corpus = corpus + sentences
corpus = utils.clean_sentences(corpus,
min_length=16,
illegal_substrings=['@', '%2'])
print('total sentences:', len(corpus))
with open(os.path.join(args.output_dir + '/isizulu/',
os.path.basename("isizulu.txt")),
'w',
encoding='utf-8') as f:
f.write('\n'.join(corpus) + '\n')
def validate(model,
dev_data,
vocab_src,
vocab_tgt,
epoch,
config,
direction=None):
model.eval()
device = torch.device(
"cpu") if config["device"] == "cpu" else torch.device("cuda:0")
with torch.no_grad():
model_hypotheses = []
references = []
val_dl = DataLoader(dev_data,
batch_size=config["batch_size_eval"],
shuffle=False,
num_workers=2)
val_dl = BucketingParallelDataLoader(val_dl)
val_kl = 0
for sentences_x, sentences_y in val_dl:
if direction == None or direction == "xy":
x_in, _, x_mask, x_len = create_batch(sentences_x, vocab_src,
device)
x_mask = x_mask.unsqueeze(1)
else:
x_in, _, x_mask, x_len = create_batch(sentences_y, vocab_src,
device)
x_mask = x_mask.unsqueeze(1)
qz = model.inference(x_in, x_mask, x_len)
z = qz.mean
pz = torch.distributions.Normal(loc=model.prior_loc,
scale=model.prior_scale).expand(
qz.mean.size())
kl_loss = torch.distributions.kl.kl_divergence(qz, pz)
kl_loss = kl_loss.sum(dim=1)
val_kl += kl_loss.sum(dim=0)
enc_output, enc_hidden = model.encode(x_in, x_len, z)
dec_hidden = model.init_decoder(enc_output, enc_hidden, z)
raw_hypothesis = beam_search(model.decoder, model.emb_tgt,
model.generate_tm, enc_output,
dec_hidden, x_mask, vocab_tgt.size(),
vocab_tgt[SOS_TOKEN],
vocab_tgt[EOS_TOKEN],
vocab_tgt[PAD_TOKEN], config, z)
hypothesis = batch_to_sentences(raw_hypothesis, vocab_tgt)
model_hypotheses += hypothesis.tolist()
if direction == None or direction == "xy":
references += sentences_y.tolist()
else:
references += sentences_x.tolist()
val_kl /= len(dev_data)
save_hypotheses(model_hypotheses, epoch, config, direction)
model_hypotheses, references = clean_sentences(model_hypotheses,
references, config)
bleu = compute_bleu(model_hypotheses,
references,
epoch,
config,
direction,
kl=val_kl)
return bleu
def main():
config = setup_config()
config["dev_prefix"] = "comparable"
vocab_src, vocab_tgt = load_vocabularies(config)
_, dev_data, _ = load_data(config,
vocab_src=vocab_src,
vocab_tgt=vocab_tgt)
model, _, validate_fn = create_model(vocab_src, vocab_tgt, config)
model.to(torch.device(config["device"]))
checkpoint_path = "{}/cond_nmt_de-en_run_7/checkpoints/cond_nmt_de-en_run_7".format(
config["out_dir"])
state = torch.load(checkpoint_path)
model.load_state_dict(state['state_dict'])
model.eval()
device = torch.device(
"cpu") if config["device"] == "cpu" else torch.device("cuda:0")
with torch.no_grad():
model_hypotheses = []
references = []
val_dl = DataLoader(dev_data,
batch_size=config["batch_size_eval"],
shuffle=False,
num_workers=4)
# val_dl = BucketingParallelDataLoader(val_dl)
for sentences_x, sentences_y in tqdm(val_dl):
sentences_x = np.array(sentences_x)
seq_len = np.array([len(s.split()) for s in sentences_x])
sort_keys = np.argsort(-seq_len)
sentences_x = sentences_x[sort_keys]
# #
sentences_y = np.array(sentences_y)
x_in, _, x_mask, x_len = create_batch(sentences_x, vocab_src,
device)
x_mask = x_mask.unsqueeze(1)
if config["model_type"] == "aevnmt":
qz = model.inference(x_in, x_mask, x_len)
z = qz.mean
enc_output, enc_hidden = model.encode(x_in, x_len, z)
dec_hidden = model.init_decoder(enc_output, enc_hidden, z)
raw_hypothesis = beam_search(model.decoder, model.emb_tgt,
model.generate_tm, enc_output,
dec_hidden, x_mask,
vocab_tgt.size(),
vocab_tgt[SOS_TOKEN],
vocab_tgt[EOS_TOKEN],
vocab_tgt[PAD_TOKEN], config)
else:
enc_output, enc_hidden = model.encode(x_in, x_len)
dec_hidden = model.decoder.initialize(enc_output, enc_hidden)
raw_hypothesis = beam_search(model.decoder, model.emb_tgt,
model.generate_tm, enc_output,
dec_hidden, x_mask,
vocab_tgt.size(),
vocab_tgt[SOS_TOKEN],
vocab_tgt[EOS_TOKEN],
vocab_tgt[PAD_TOKEN], config)
hypothesis = batch_to_sentences(raw_hypothesis, vocab_tgt)
inverse_sort_keys = np.argsort(sort_keys)
model_hypotheses += hypothesis[inverse_sort_keys].tolist()
references += sentences_y.tolist()
save_hypotheses(model_hypotheses, 0, config, None)
model_hypotheses, references = clean_sentences(model_hypotheses,
references, config)
bleu = sacrebleu.raw_corpus_bleu(model_hypotheses, [references]).score
print(bleu)
]
for url, file_name, output_name, lines_to_remove in datasets:
print('processing:', url)
r = requests.get(url)
zip = zipfile.ZipFile(BytesIO(r.content))
corpus = zip.open(file_name).read().decode('utf-8').strip()
# remove tags containing article filenames
corpus = re.sub(r'<fn>.*</fn>', '', corpus)
# put each sentence on a new line
corpus = corpus.replace('. ', '.\n')
# remove empty lines from corpus
sentences = corpus.splitlines()
sentences = utils.clean_sentences(
sentences,
illegal_substrings=['\ufeff', '='],
lines_to_remove=lines_to_remove,
)
# write article to file (with each sentence on a new line)
output_file_name = os.path.join(args.output_dir, output_name)
with open(output_file_name, 'w', encoding='utf-8') as f:
f.write('\n'.join(sentences))
print('total sentences in {}:'.format(output_name), corpus.count('\n'))
def evaluate(model, dev_data, vocab_src, vocab_tgt, config, direction=None):
model.eval()
with torch.no_grad():
model_hypotheses = []
references = []
device = torch.device(
"cpu") if config["device"] == "cpu" else torch.device("cuda:0")
val_dl = DataLoader(dev_data,
batch_size=config["batch_size_eval"],
shuffle=False,
num_workers=4)
# val_dl = BucketingParallelDataLoader(val_dl)
for sentences_x, sentences_y in tqdm(val_dl):
if direction == None or direction == "xy":
sentences_x, sentences_y, sort_keys = sort_sentences(
sentences_x, sentences_y)
x_in, _, x_mask, x_len = create_batch(sentences_x, vocab_src,
device)
x_mask = x_mask.unsqueeze(1)
else:
sentences_y, sentences_x, sort_keys = sort_sentences(
sentences_y, sentences_x)
x_in, _, x_mask, x_len = create_batch(sentences_y, vocab_src,
device)
x_mask = x_mask.unsqueeze(1)
if config["model_type"] == "coaevnmt":
qz = model.inference(x_in, x_mask, x_len)
z = qz.mean
enc_output, enc_hidden = model.encode(x_in, x_len, z)
dec_hidden = model.init_decoder(enc_output, enc_hidden, z)
raw_hypothesis = beam_search(model.decoder,
model.emb_tgt,
model.generate_tm,
enc_output,
dec_hidden,
x_mask,
vocab_tgt.size(),
vocab_tgt[SOS_TOKEN],
vocab_tgt[EOS_TOKEN],
vocab_tgt[PAD_TOKEN],
config,
z=z)
elif config["model_type"] == "conmt":
enc_output, enc_hidden = model.encode(x_in, x_len)
dec_hidden = model.decoder.initialize(enc_output, enc_hidden)
raw_hypothesis = beam_search(model.decoder, model.emb_tgt,
model.generate_tm, enc_output,
dec_hidden, x_mask,
vocab_tgt.size(),
vocab_tgt[SOS_TOKEN],
vocab_tgt[EOS_TOKEN],
vocab_tgt[PAD_TOKEN], config)
hypothesis = batch_to_sentences(raw_hypothesis, vocab_tgt)
inverse_sort_keys = np.argsort(sort_keys)
model_hypotheses += hypothesis[inverse_sort_keys].tolist()
if direction == None or direction == "xy":
references += sentences_y.tolist()
else:
references += sentences_x.tolist()
model_hypotheses, references = clean_sentences(model_hypotheses,
references, config)
bleu = sacrebleu.raw_corpus_bleu(model_hypotheses, [references]).score
print(bleu)