Lorenzo Pellegrini

This may be a good starting point: https://github.com/yaoyao-liu/mini-imagenet-tools They also have a link to a drive with all the processed images! However, there doesn't seem to be an official static...

Currently, metrics are logged: - after train epoch - after eval exp - after eval We might consider adding a pass-through flag to print incoming metrics directly in `log_single_metric` here:...

FFCV support has now been added in the master branch (it is NOT included in the latest pip release, v0.4.0). See the examples for more details.

Both ClassBalanced and ParametricBuffer now convert each target to int explicitly, so this is no longer an issue. It may be possible that some benchmarks still return targets as tensors...

Added in #1463

@AntonioCarta The PR is almost completed and I'd like to add more unit tests to be sure everything works fine. The only real part that I still need to figure...

I'm moving the development status to #1315. This PR, in its current state, is very difficult to merge in a single step as it is too big and based on...

We already have a nightly Docker image which is also updated using the CI. It would be a waste not to add a reference to it in the GitBook doc....

Alas, it seems quite complex to add new parameters to an existing optimizer. Maybe the best solution is to allow the user to customize the behavior of `make_optimizer` by overriding...

Was this fixed in the meantime?