thunlp / JointNRE

Joint Neural Relation Extraction with Text and KGs
MIT License
187 stars 36 forks source link

Some fixes related to flags' type in test.py #7

Closed yzho0907 closed 5 years ago

yzho0907 commented 5 years ago

Some of flags below are not necessarily float! I have changed them to integers.

tf.app.flags.DEFINE_integer('nbatch_kg',100,'entity numbers used each training time')

tf.app.flags.DEFINE_integer('ent_total',lib.getEntityTotal(),'total of entities') tf.app.flags.DEFINE_integer('rel_total',lib.getRelationTotal(),'total of relations') tf.app.flags.DEFINE_integer('tri_total',lib.getTripleTotal(),'total of triples') tf.app.flags.DEFINE_integer('katt_flag', 1, '1 for katt, 0 for att')

tf.app.flags.DEFINE_integer('max_length',config['fixlen'],'maximum of number of words in one sentence') tf.app.flags.DEFINE_integer('pos_num', config['maxlen'] * 2 + 1,'number of position embedding vectors') tf.app.flags.DEFINE_integer('num_classes', config['textual_rel_total'],'maximum of relations')

tf.app.flags.DEFINE_integer('hidden_size',230,'hidden feature size') tf.app.flags.DEFINE_integer('pos_size',5,'position embedding size')

tf.app.flags.DEFINE_integer('max_epoch',30,'maximum of training epochs') tf.app.flags.DEFINE_integer('batch_size',131*2,'entity numbers used each training time')

tf.app.flags.DEFINE_integer('test_batch_size',131*2,'entity numbers used each test time')

THUCSTHanxu13 commented 5 years ago

I have fix flag's type in test.py