How to run test？

[supercoderhawk]

How to run test？

[supercoderhawk]

1 默认就能够使用GPU训练和预测，并已在1080Ti和2080Ti上成功运行多次，如不能使用，请检查显卡驱动，CUDA版本与torch版本是否对应。
2 调用copyrnn\predict.py

# Your model path
model_path = 'data/models/model.model'
# your vocab path
vocab_path = 'data/vocab_kp20k.txt'
keyword_generator = CopyRnnPredictor({'model': model_path},
                                     vocab_info=vocab_path,
                                     beam_size=50,
                                     max_target_len=5,
                                     max_src_length=800)

# test some single cases, or use as component in online service
tokens = ['numerous', 'studies', 'have', 'demonstrated', 'that', 'h2o2-induced', 'apoptosis', 'is', 'mediated', 'by', 'activation', 'of', 'mapks']
keyword_generator.predict([tokens], delimiter=' ', tokenized=True)

# evaluate file
from munch import Munch
src_filename = 'data/nus.jsonl'
dest_filename = 'data/nus_pred.jsonl'
config = read_json('data/models/model.json')
keyword_generator.eval_predict(src_filename, dest_filename,args=Munch(config))