Siniara

Alright reproducible code! (Hopefully ) ```python from pyts.bag_of_words import BagOfWords bow = BagOfWords(window_size=20, word_size=3, n_bins=3, threshold_std=0.01) X = np.array([[-2. , -2.29999995, -2.29999995, -2.4000001 , -2.5 , -2.5 , -2.5 ,...

Yes, thank you, this seems to work for me as well :) What would then be the syntax to use the multivariate transformer with SAX-VSM? I tried the following, which...

What I do presently is almost what the BOW transformation returns From `pyts` BOW:  and I have it like this:  so I do the feature counts for each...

Thanks again for your help! > So you need to merge the bag of words for all the samples that belong to the same class (or equivalently sum the histograms,...

I think I'm close to getting my grid search working; the main issue I'm having is that when using the RF classifier each split needs to have the same amount...

This time a bit of a delay before my next enormous thank you! :) I spent some time playing around with this, and have some comments, observations, questions: - The...

Ahh your input about dimensions and array sizes make sense, (indeed my background is not computer science :) ) I also happen to be dealing with an imbalanced classification problem,...

I took a small brake, and I am now in the final leg of this project. :) I wanted to ask you about the symbol selection for constant subsequences. Is...

Right indeed you are! This makes sense, I remember now :) Lastly, I wanted to visualise the relevant patterns I find with `pyts` on my original timeseries. I've seen the...

I actually managed to use your example code to achieve what I want, I had misunderstood the example. If I provided the feature of interest as input to your code...