ExaScience / smurff

Bayesian Factorization with Side Information in C++ with Python wrapper
MIT License
70 stars 14 forks source link

Remove code duplication in train_rmse implementations #59

Closed motoharu-yano closed 6 years ago

motoharu-yano commented 6 years ago

part of train_rmse does the same thing as sumsq. This is because due to the square operation we dont care about the sign of the result when subtracting (pred - value) or (value - pred). We can use sumsq as part of train_rmse.