rbm result is worse than average model

[zhegeliang2]

i found the rmse of rbm is worse than average model，can you show me better parameters，now the dataset is movielens ml-100k, and my parameters is ： rec.recommender.class=rbm rec.iterator.maximum=20 data.input.path=movielens/ml-100k/ratings.txt rec.factor.number=500 rec.epsilonw=0.01 rec.epsilonvb=0.01 rec.epsilonhb=0.01 rec.tstep=1 rec.momentum=0.1 rec.lamtaw=0.01 rec.lamtab=0.0 the result of rbm is: 17/10/11 20:12:15 INFO TextDataConvertor: Dataset: .../data/movielens/ml-100k/ratings.txt 17/10/11 20:12:15 INFO TextDataConvertor: All dataset files [../data/movielens/ml-100k/ratings.txt] 17/10/11 20:12:15 INFO TextDataConvertor: All dataset files size 2079173 17/10/11 20:12:15 INFO TextDataConvertor: Now loading dataset file ratings 17/10/11 20:12:16 INFO TextDataModel: Transform data to Convertor successfully! 17/10/11 20:12:16 INFO TextDataModel: Split data to train Set and test Set successfully! 17/10/11 20:12:16 INFO TextDataModel: Data size of training is 80019 17/10/11 20:12:16 INFO TextDataModel: Data size of testing is 19981 17/10/11 20:12:17 INFO RBMRecommender: Job Setup completed. 17/10/11 20:13:32 INFO RBMRecommender: Job Train completed. 17/10/11 20:13:44 INFO RBMRecommender: Job End. 17/10/11 20:13:45 INFO RecommenderJob: Evaluator value:RMSE is 1.4432797642638389 17/10/11 20:13:45 INFO RecommenderJob: Evaluator value:MAE is 1.1286155077841684 17/10/11 20:13:45 INFO RecommenderJob: Evaluator value:MSE is 2.0830564779334826 17/10/11 20:13:45 INFO RecommenderJob: Evaluator value:MPE is 0.9190230719183224

the result of average model: 17/10/11 20:17:15 INFO TextDataConvertor: Dataset: .../data/movielens/ml-100k/ratings.txt 17/10/11 20:17:15 INFO TextDataConvertor: All dataset files [../data/movielens/ml-100k/ratings.txt] 17/10/11 20:17:15 INFO TextDataConvertor: All dataset files size 2079173 17/10/11 20:17:15 INFO TextDataConvertor: Now loading dataset file ratings 17/10/11 20:17:16 INFO TextDataModel: Transform data to Convertor successfully! 17/10/11 20:17:16 INFO TextDataModel: Split data to train Set and test Set successfully! 17/10/11 20:17:16 INFO TextDataModel: Data size of training is 80019 17/10/11 20:17:16 INFO TextDataModel: Data size of testing is 19981 17/10/11 20:17:16 INFO GlobalAverageRecommender: Job Setup completed. 17/10/11 20:17:16 INFO GlobalAverageRecommender: Job Train completed. 17/10/11 20:17:16 INFO GlobalAverageRecommender: Job End. 17/10/11 20:17:16 INFO RecommenderJob: Evaluator value:MSE is 1.2660027571902484 17/10/11 20:17:16 INFO RecommenderJob: Evaluator value:MPE is 1.0 17/10/11 20:17:16 INFO RecommenderJob: Evaluator value:MAE is 0.943605900855436 17/10/11 20:17:16 INFO RecommenderJob: Evaluator value:RMSE is 1.1251678795585343 17/10/11 20:17:16 INFO RecommenderJob: Result path is ../result/movielens/ml-100k/ratings.txt-globalaverage-output/globalaverage