'------------- Training epoch {} --------------'.format(i + 1))
train_accuracy = trainer.train(train_set, args.batch_size,
args.debug_single_batch)
if args.debug_single_batch: sys.exit(0)
dev_scores = trainer.test(dev_set, args.batch_size)
dev_map, dev_mrr = compute_map_mrr(args.dataset_folder, dev_set,
dev_scores)
logger.info("------- MAP {}, MRR {}".format(dev_map, dev_mrr))
if dev_map - best_map > 1e-3: # new map is better than best map
best_model = i
best_map = dev_map
QAModel.save(net, args.model_outfile)
logger.info('Achieved better dev_map ... saved model')
if args.test_on_each_epoch:
test_scores = trainer.test(test_set, args.batch_size)
map, mrr = compute_map_mrr(args.dataset_folder, test_set,
test_scores)
logger.info("------- MAP {}, MRR {}".format(map, mrr))
if (i - best_model) >= args.patience:
logger.warning('No improvement since the last {} epochs. Stopping training'\
.format(i - best_model))
break
logger.info(' ------------ Training epochs completed! ------------')
logger.info('Best dev MAP in training phase = {:.4f}'.format(best_map))
torch.manual_seed(1234)
np.random.seed(1234)
train_set, dev_set, test_set = 'train', 'clean-dev', 'clean-test'
if args.train_all:
train_set, dev_set, test_set = 'train-all', 'raw-dev', 'raw-test'
# cache word embeddings
cache_file = os.path.splitext(args.word_vectors_file)[0] + '.cache'
utils.cache_word_embeddings(args.word_vectors_file, cache_file)
vocab_size, vec_dim = utils.load_embedding_dimensions(cache_file)
# instantiate model
net = QAModel(vec_dim, args.filter_width, args.num_conv_filters, args.no_ext_feats) #filter width is 5
QAModel.save(net, args.dataset_folder, args.model_fname)
torch.set_num_threads(args.num_threads)
trainer = Trainer(net, args.eta, args.mom, args.no_loss_reg, vec_dim)
logger.info("Loading input data...")
trainer.load_input_data(args.dataset_folder, cache_file, train_set, dev_set, test_set)
best_map = 0.0
best_model = 0
for i in range(args.epochs):
logger.info('------------- Training epoch {} --------------'.format(i+1))
train_accuracy = trainer.train(train_set, args.batch_size, args.debugSingleBatch)
if args.debugSingleBatch: sys.exit(0)
