High performance, easy-to-use, and scalable machine learning (ML) package, including linear model (LR), factorization machines (FM), and field-aware factorization machines (FFM) for Python and CLI interface.
I train a FM model on my own data, and the final test auc is 71.4%.
Then, I use the linear weight dumps from the trained FM model to check the importance of my handcraft feature. But, I found that some feature about the items id which are not very "hot" on online statistic show very big weight.
Can you give me some advice about this problem?
I train a FM model on my own data, and the final test auc is 71.4%. Then, I use the linear weight dumps from the trained FM model to check the importance of my handcraft feature. But, I found that some feature about the items id which are not very "hot" on online statistic show very big weight. Can you give me some advice about this problem?