hugstone

连续特征的embedding问题

3

hi，看了代码实现，有个疑问，连续特征乘以k维需要训练的embedding向量（1个数分解为多个数），这样对连续特征也带来了信息损失吧？

Q1. compare with the https://github.com/Intel-bigdata/imllib-spark，the auc performance of this spark-fm is very bad in my ctr prediction task, i have try mini-batch SGD、L-BFGS、parallel SGD and parallel FTRL, all auc <...

self.hidden_size is constant，divide self.hidden_size is not necessary