Max Cohen

Hello, for the dataset used in these examples, please see #2 . The expected structure of the input data is described in the [Transformer's documentation](https://timeseriestransformer.readthedocs.io/en/latest/transformer.html#transformer.Transformer.forward); you can implement your own...

`K` is the length of the time series. In your example `K=j`, each batch of data should consist of inputs with shape `(batch_size, j, 2)` and outputs with shape `(batch_size,...

Yes exactly !

Hi, I don't see directly where a NaN could come from, I encourage you to debug during the validation loss computation in order to see what tensor or function is...

Hi, if you're talking about the default value of `pe_period`, I would like to keep it as is. It's well documented too. Do tell me if there is something else...

I see, I'll let you work on it, and see if I can find a alternative later this week.

Hi, yes it can. It actually for this kind of application this model was designed.

Same here on Ubuntu **System Configuration** - OS: Ubuntu 20.04 - Firefox: 84.0 - Vim-Vixen: 0.32

Hi, thanks for reporting this issue, it should be fixed with 997b08e18fc06452a505d14c39f33294b4d9e65e.

I haven't gotten around to code this setting, fell free to send a PR if you've got the time to write it yourself. The function is defined here: https://github.com/maxjcohen/transformer/blob/997b08e18fc06452a505d14c39f33294b4d9e65e/tst/utils.py#L7-L29