Marco Z

As of today, `data-default` has 563 reverse dependencies (http://packdeps.haskellers.com/reverse/data-default). It may be that a number of them are stale, but there are also a number of prominent packages on that...

On this note, I don't think the gradient of the softmax function defined in backprop-learn (https://github.com/mstksg/backprop-learn/blob/master/src/Backprop/Learn/Model/Function.hs#L122) does the right thing.

@mstksg Sorry to bother you but I could really use your input on this, it's been a head scratcher for a few days.

Thank you @mstksg for the explanation (and apologies for responding with a mere 7 months of difference! Though I did read this back in January)

I've only recently started using backprop, but pretty much like ad it assumes that the computation to be differentiated is pure. I can't quite follow the internals of the library...

@jh14778 @ivanovs-4 I just recalled : you could also wrap your effects inside `ABP` , which will provide the Backprop instance for you : https://hackage.haskell.org/package/backprop-0.2.6.4/docs/Numeric-Backprop.html#t:ABP

Since this error is only due to the tests using Either, and we cannot avoid the MonadFail constraint since it is due to `makeRegexM`, I suggest we switch to Nothing.

For posterity, this shows up when compiling against GHC 8.8.1 e.g. see these CI logs : https://github.com/ocramz/aeson-schema/runs/6031979441?check_suite_focus=true#step:4:1082

I've addressed the required changes in #13

Glad @kozross stepped up for maintainership, I just have a couple patches to contribute for now. @Fuuzetsu I checked out `aeson-schemas` from LeapYear and while it is a very cool...