Mark Colley

Internally, for example [CrossEntropy](https://github.com/jdb78/pytorch-forecasting/blob/master/pytorch_forecasting/metrics/point.py) uses _F.cross_entropy_. This allows to **weight** the loss (see also [here](https://github.com/pytorch/pytorch/blob/main/torch/nn/modules/loss.py)). A feature request would be to expose this also in pytorch-forecasting as this is quiet...

While the documentation states that for CrossEntropy, the reduction used is "mean" (https://pytorch-forecasting.readthedocs.io/en/v1.0.0/api/pytorch_forecasting.metrics.point.CrossEntropy.html), the implementation defaults to "none" (see [here](https://github.com/jdb78/pytorch-forecasting/blob/master/pytorch_forecasting/metrics/point.py)). Is this to be expected? For me, this was unintuitive....

Question: Upgrade to PyTorch version >2

2

Hey there, thank you very much for your work. I was wondering whether there are any plans and how they might look like to update the models to Pytorch version...

Hey there, for a CC BY license, it would be great to already include the CC license icon logo and write in the copyright line "This work is licensed under...

[Feature request] Nested Folders and [Fix] Wordcount

2

Hello, I often find myself needing folders to, for example, group projects for a conference. These occur per year. Today, I have a folder for every year on the top-layer....