[160520]CoNLL X English

Branch: master

Command: spark-submit --driver-memory 20g --executor-memory 20g --master 'local[*]' glm_parser.py -s 8 -i 1 -p /ugrad/2/jca303/data/CoNLL-X/English/data --train="conll08st_train.conll" --test="conll08st_test.conll" --learner=perceptron --fgen=english_1st_fgen --parser=ceisner --config=/ugrad/2/jca303/glm-parser/src/config/conllx.config

Traceback (most recent call last): File "/ugrad/2/jca303/glm-parser/src/glm_parser.py", line 408, in gp.evaluate(training_time) File "/ugrad/2/jca303/glm-parser/src/glm_parser.py", line 104, in evaluate self.evaluator.evaluate(test_data_pool, self.parser, self.w_vector, training_time) File "/ugrad/2/jca303/glm-parser/src/evaluate/evaluator.py", line 63, in evaluate logging.info("Unlabeled accuracy: %.12f (%d, %d)" % (self.get_acc_unlabeled_accuracy(), self.unlabeled_correct_num, self.unlabeled_gold_set_size)) File "/ugrad/2/jca303/glm-parser/src/evaluate/evaluator.py", line 102, in get_acc_unlabeled_accuracy return self.unlabeled_correct_num / self.unlabeled_gold_set_size ZeroDivisionError: division by zero