正确保存/加载MatrixFactorizationModel

Question

我有MatrixFactorizationModel对象。如果我试图在通过ALS.train(...)构建模型后立即向单个用户推荐产品然后它需要300ms (对于我的数据和硬件)。但如果我将模型保存到磁盘并加载回来，那么推荐几乎需要2000ms。斯帕克还警告说：

15/07/17 11:05:47 WARN MatrixFactorizationModel: User factor does not have a partitioner. Prediction on individual records could be slow.
15/07/17 11:05:47 WARN MatrixFactorizationModel: User factor is not cached. Prediction could be slow.
15/07/17 11:05:47 WARN MatrixFactorizationModel: Product factor does not have a partitioner. Prediction on individual records could be slow.
15/07/17 11:05:47 WARN MatrixFactorizationModel: Product factor is not cached. Prediction could be slow.

如何在加载模型后创建/设置分区程序并缓存用户和产品因子？以下方法无济于事：

model.userFeatures().cache();
model.productFeatures().cache();

此外，我还试图对这些rdds进行重新分区，并从重新分区的版本中创建新模型，但这也没有帮助。

Answer 1

你不必使用括号，userFeatures是(Int，ArrayDouble)的RDD，它不带参数。

这将对您有所帮助：

model.userFeatures.cache
model.productFeatures.cache