Show HN: Dreaming up dresses via WGAN

Show HN: Dreaming up dresses via WGAN(sanealytics.com)

1 points by ignorant 9 years ago | 3 comments

minimaxir 9 years ago |

> This gives an RMSE (don’t worry about what this means, higher is better) of 0.28.

You want a lower RMSE; the E stands for "error"

gwern 9 years ago | |

It's also odd that the RMSE increases when adding a nonlinear term. The in-sample RMSE should only decrease when adding additional terms and nonlinearities (that's where overfitting comes from).

minimaxir 9 years ago | | |

He didn't add the log-transform, he replaced the linear relationship with the log-transform, which would cause RMSE to increase if there were a lot of high-skewing response values, which appears to be the case from the diagram.(log-transforming the response might help)