Ignacio Oguiza

Hi @ivanzvonkov , I fully agree with you. It's something I've always had in mind, and I'll start doing it from now onwards. If you look at the [RNN documentation](https://timeseriesai.github.io/tsai/models.rnn.html#converting-a-model-to-torchscript)...

There isn't a way to test all models, but I'll add some functionality soon as it makes sense.

Issue closed due to lack of response.

Once Pytorch and fastai fully support the new apple chips we'll make tsai compatible as well. But I'm afraid there's not much we can do for now.

Hi @firmai, Sorry for the late reply. You can actually use multiple columns as unique_id_cols. All of them are used for sorting. In addition to that, you may add other...

Closed due to lack of response.

There was a bug in the code that has already been fixed. If you upgrade `tsai` to the latest version you shouldn't get this issue.

Issue closed due to lack of response.

I'd recommend you to take a look at this nb: https://github.com/timeseriesAI/tsai/blob/main/nbs/008_data.metadatasets.ipynb You'll need to adapt the loop to work with files and load the data. But the purpose of metadatasets...

Hi @hailthedawn, That's strange. I haven't seen that type of variation in the score. Are you using something similar to this? ```python from tsai.models.MINIROCKET import MiniRocketClassifier # Univariate classification with...