• PyTorch-Forecasting version: 1.0.0
• PyTorch version: 2.1.0
• Python version: 3.10.10
• Operating System: OS X

There is currently no example about how to set up any model to predict multiple targets (auto-regressively).

targets = ['col1', 'col2', ...] dataset_ts = TimeSeriesDataSet( df, time_idx="time_idx", target=targets, group_ids=["group"], time_varying_unknown_reals=targets, target_normalizer=MultiNormalizer([TorchNormalizer() for _ in targets]), )

My targets are also the input to the model, as it is an auto-regressive use case.

I've update the code, see my comments below. Now it will work with multi-targets! :)