default=True,
help='using masked data or not')
parser.add_argument('--use_glove',
type=str2bool,
default=True,
help='Using Glove embedding or not')
parser.add_argument('--use_extra_feature',
type=str2bool,
default=True,
help='Using extra feature, e.g. ' + 'NER, POS, ')
args = parser.parse_args()
file_name = args.file_name
is_masked = args.is_masked
use_glove = args.use_glove
use_extra_feature = args.use_extra_feature
options_file = 'data/elmo_2x4096_512_2048cnn_2xhighway_options.json'
weight_file = 'data/elmo_2x4096_512_2048cnn_2xhighway_weights.hdf5'
glove_file = 'data/glove.840B.300d.txt'
logger = config_logger('Preprocess')
preprocesser = Preprocesser(file_name,
logger,
is_masked=is_masked,
use_glove=use_glove,
use_extra_feature=use_extra_feature,
options_file=options_file,
weight_file=weight_file,
glove_file=glove_file)
preprocesser.preprocess()
hops = args.hop_num
epochs = args.epochs
batch_size = args.batch_size
training_info_interval = args.info_interval
dropout = args.dropout
encoding_size = args.encoding_size
pos_emb_size = args.pos_emb_size
ner_emb_size = args.ner_emb_size
options_file = 'data/elmo_2x4096_512_2048cnn_2xhighway_options.json'
weight_file = 'data/elmo_2x4096_512_2048cnn_2xhighway_weights'
encoding_type_map = {'lstm': 'lstm', 'linear': 'linear'}
model_name = 'BAG'
if evaluation_mode:
logger = config_logger('evaluation/' + model_name)
else:
logger = config_logger('BAG')
model_path = os.getcwd() + '/models/' + model_name + '/'
if not os.path.exists(model_path):
os.makedirs(model_path)
tokenize = TweetTokenizer().tokenize
logger.info('Hop number is %s', hops)
logger.info('Learning rate is %s', learning_rate)
logger.info('Training epoch is %s', epochs)
logger.info('Batch size is %s', batch_size)
logger.info('Dropout rate is %f', dropout)
logger.info('Encoding size for nodes and query feature is %s',
encoding_size)
parser.add_argument('--patience',
type=int,
default=5,
help='Epoch early stopping patience')
args = parser.parse_args()
device = torch.device(
'cuda') if torch.cuda.is_available() else torch.device('cpu')
encoding_type_map = {'lstm': 'lstm', 'linear': 'linear'}
best_res = {'acc': 0.0}
pred_res = []
model_name = 'BAG-pytorch'
if args.evaluation_mode:
logger = config_logger('evaluation-pytorch/' + model_name)
else:
logger = config_logger('BAG-pytorch')
model_path = os.getcwd() + '/models-pytorch/' + model_name + '/'
if not os.path.exists(model_path):
os.makedirs(model_path)
model_path = os.path.join(model_path, 'best_model.bin')
for item in vars(args).items():
logger.info('%s : %s', item[0], str(item[1]))
'''Check whether preprocessed files are existed'''
train_file_name_prefix, fileExist = checkPreprocessFile(
args.in_file, args.add_query_node)
if not fileExist:
logger.info('Cannot find preprocess data %s, program will shut down.',
hops = args.hop_num
epochs = args.epochs
batch_size = args.batch_size
training_info_interval = args.info_interval
dropout = args.dropout
encoding_size = args.encoding_size
pos_emb_size = args.pos_emb_size
ner_emb_size = args.ner_emb_size
options_file = 'data/elmo_2x4096_512_2048cnn_2xhighway_options.json'
weight_file = 'data/elmo_2x4096_512_2048cnn_2xhighway_weights'
encoding_type_map = {'lstm': 'lstm', 'linear': 'linear'}
model_name = 'CQ-GCN'
if evaluation_mode:
logger = config_logger('evaluation/' + model_name)
else:
logger = config_logger('CQ_GCN')
model_path = os.getcwd() + '/best_models/' + model_name + '/'
if not os.path.exists(model_path):
os.makedirs(model_path)
tokenize = TweetTokenizer().tokenize
logger.info('Hop number is %s', hops)
logger.info('Learning rate is %s', learning_rate)
logger.info('Training epoch is %s', epochs)
logger.info('Batch size is %s', batch_size)
logger.info('Dropout rate is %f', dropout)
logger.info('Encoding size for nodes and query feature is %s',
encoding_size)
hops = args.hop_num
epochs = args.epochs
batch_size = args.batch_size
training_info_interval = args.info_interval
dropout = args.dropout
encoding_size = args.encoding_size
pos_emb_size = args.pos_emb_size
ner_emb_size = args.ner_emb_size
options_file = 'data/elmo_2x4096_512_2048cnn_2xhighway_options.json'
weight_file = 'data/elmo_2x4096_512_2048cnn_2xhighway_weights'
encoding_type_map = {'lstm': 'lstm', 'linear': 'linear'}
model_name = 'PathBasedGCN'
if evaluation_mode:
logger = config_logger('evaluation/' + model_name)
else:
logger = config_logger('PathBasedGCN')
model_path = os.getcwd() + '/models/' + model_name + '/'
if not os.path.exists(model_path):
os.makedirs(model_path)
tokenize = TweetTokenizer().tokenize
logger.info('Hop number is %s', hops)
logger.info('Learning rate is %s', learning_rate)
logger.info('Training epoch is %s', epochs)
logger.info('Batch size is %s', batch_size)
logger.info('Dropout rate is %f', dropout)
logger.info('Encoding size for nodes and query feature is %s',
encoding_size)