INNER CODE UNIT · Python

evaluate

ShannonAI/glyce · glyce/models/latticeLSTM/main.py:159

def evaluate(data, model, name):
    if name == "train":
        instances = data.train_Ids
    elif name == "dev":
        instances = data.dev_Ids
    elif name == 'test':
        instances = data.test_Ids
    elif name == 'raw':
        instances = data.raw_Ids
    else:
        logger.info("Error: wrong evaluate name," + name)
    pred_results = []
    gold_results = []
    model.eval()
    start_time = time.time()
    train_num = len(instances)
    total_batch = train_num//data.HP_batch_size+1

View source record →