def main():
opt.dataset_file = dataset_files[opt.dataset]
# set random seed
setup_seed(opt.seed)
if not os.path.exists('log'):
os.mkdir('log')
log_file = '{}-{}-{}.log'.format(
opt.model_name, opt.dataset, strftime("%Y-%m-%d_%H:%M:%S",
localtime()))
logger.addHandler(logging.FileHandler("%s/%s" % ('./log', log_file)))
start_time = time.time()
ins = Instructor(opt.dataset_file['test_query'],
opt.dataset_file['test_reply'])
ins.run(opt.dataset_file['train_query'], opt.dataset_file['train_reply'])
time_dif = get_time_dif(start_time)
logger.info("Time usage: {}".format(time_dif))
if args.embedding == 'random':
embedding = 'random'
model_name = args.model # 'TextRCNN' # TextCNN
x = import_module('models.' +
model_name) #一个函数运行需要根据不同项目的配置，动态导入对应的配置文件运行。
config = x.Config(dataset) #进入到对应模型的__init__方法进行参数初始化
start_time = time.time()
print("Loading data...")
train_data, dev_data, test_data, train_sentences, test_sentences, dev_sentences, word_to_id, id_to_word, tag_to_id, id_to_tag = load_model_dataset(
config)
config.n_vocab = len(word_to_id)
time_dif = data_utils.get_time_dif(start_time)
print("Time usage:", time_dif)
embedding_pretrained = data_utils.load_word2vec(config, id_to_word)
train_X, train_Y = data_utils.get_X_and_Y_data(train_data, config.max_len,
len(tag_to_id))
dev_X, dev_Y = data_utils.get_X_and_Y_data(dev_data, config.max_len,
len(tag_to_id))
test_X, test_Y = data_utils.get_X_and_Y_data(test_data, config.max_len,
len(tag_to_id))
train_dataset = tf.data.Dataset.from_tensor_slices((train_X, train_Y))
train_dataset = train_dataset.shuffle(len(train_X)).batch(