Julien Herzen
Julien Herzen
@nzw0301 would be happy to have a look if I find a bit of time. Did you narrow down to some preferred way of solving the problem?
Hi, could you please post a minimal self-contained snippet showing the case where the same trained deterministic model produces different predictions? Thanks.
Hi, could you send a self-contained code snippet, including everything required to reproduce (incl. a way to generate/import a toy dataset on which the issue occurs)?
See also: https://github.com/unit8co/darts/issues/672
Thanks @rijkvandermeulen. It looks like a good way to do it. We'll try to review it sometime soon (we are a bit behind these days, apologies). We also will have...
Hi, thanks for the suggestion. Could you share example use-cases? I'm wondering if/how this should fit into Darts.
good idea, adding to the backlog :) (and contributions are welcome!)
Hi @Beerstabr, after checking, I think it should already be possible to do something like this: ```python my_model = RegressionModel(..., lags=n_lags) my_model.fit(..., sample_weight=[1. / (in_len - i) for i in...
Hi @Beerstabr , first off, I'm sorry because I realised I made a mistake in my previous message - the `sample_weight` are (obviously) per-sample weights and not per-dimension weights, as...
Hi, this is intended, if there is at least one past covariates, then `past_covariates[0]` should be defined and have a `width`. Could you share a code snippet to reproduce the...