def minibatch_ready(self, model, items_ready, num_items):
if self.training_prog is None:
self.training_prog = lib.ProgressBar(num_items, 5)
self.training_prog.inc_value()
from framework import config
lib.create_dir(config.results_dir + '/imageimportance')
for dataset_name in ['flickr8k', 'flickr30k', 'mscoco']:
print(dataset_name)
datasources = data.load_datasources(dataset_name)
datasources['test'].tokenize_sents()
image_keywords = [{
token
for sent in sent_group
for (token, tag) in nltk.pos_tag(sent, tagset='universal')
if tag == 'NOUN'
} for sent_group in datasources['test'].get_text_sent_groups()]
prog = lib.ProgressBar(len(image_keywords), 5)
with open(config.results_dir + '/imageimportance/foils_' + dataset_name +
'.txt',
'w',
encoding='utf-8') as f:
for (i, (curr_img, curr_keywords)) in enumerate(
zip(datasources['test'].images, image_keywords)):
index = min(range(len(image_keywords)),
key=lambda j:
(image_keywords[j] & curr_keywords, -distance.cosine(
datasources['test'].images[j], curr_img)))
print(index, file=f)
prog.update_value(i + 1)
print()
print()
(sents_logprobs,
tokens_logprobs) = model.get_sents_logprobs(
max_batch_size=config.val_batch_size,
index_sents=dataset.test.index_sents)
langmod_prob_stats = evaluation.get_probability_stats(
sents_logprobs, dataset.test.index_sents.lens)
with open(config.results_dir + '/langmodtrans/' + corpus +
'/' + dir_name + '/1_probs.txt',
'w',
encoding='utf-8') as f:
for logprobs in tokens_logprobs:
print(*logprobs, sep='\t', file=f)
print('Generating sentences...')
prog = lib.ProgressBar(capgen_test.num_groups, 5)
(index_sents, logprobs) = model.generate_sents_sample(
max_batch_size=config.val_batch_size,
images=[None] * capgen_test.num_groups,
lower_bound_len=config.lower_bound_len,
upper_bound_len=config.upper_bound_len,
temperature=config.temperature,
listener=lambda num_ready: prog.inc_value())
text_sents = index_sents.decompile_sents(
langmod_vocab).sents
with open(config.results_dir + '/langmodtrans/' + corpus +
'/' + dir_name + '/1_sents.txt',
'w',
encoding='utf-8') as f:
for sent in text_sents:
print(*sent, sep=' ', file=f)