I know for regression problems we have MSELossFlat(). I can’t find any fastai documentation for the loss options. Is L1 loss an option as well, or would I have to build it as a custom function? My data is positively skewed, so I’m thinking L1 will be a better loss function than MSE.
2 Likes