Spark and Surprise MAE和RMSE为0.0

Question

我使用以下代码在Spark环境中训练我的惊喜库推荐器模型

def fit(data):
    trainset = data["train"]
    testset = data["test"]

    sim_options = { "name": "cosine", "user_based": False }
    model = KNNBasic(sim_options=sim_options)
    model.fit(trainset)

    predictions = model.test(testset)
    return accuracy.rmse(predictions)

# KNN Basic with cosine
print("Item-based with cosine")
data = [dm.loadData(spark)]

predictions = spark.sparkContext \
 .parallelize(data, 1) \
 .map(fit) \
 .collect()

print(predictions)

我返回RMSE，该值为0.0。当我在本地计算机上运行此火车代码时，它是4.5。有人可以帮我解决这个问题吗？