Sarah Volkamer

I came to the conclusion that it wouldn't work at all since we don't know which motif will be found first. Therefor we aren't able to tell the function that...

Today I came across this paper: https://core.ac.uk/download/pdf/287941767.pdf In the paper the authors talk about the implications of z-normalization in the matrix profile. They claim that, when comparing subsequences that are...

Yes, that makes sense! So we could first finish the tutorial with an easier concept and I could dive into the other problem afterwards.

Hey! Since we were discussing this topic in detail [here](https://github.com/TDAmeritrade/stumpy/discussions/494) and have already generalized the multidimensional motif discovery to finding the top k motifs and top k nearest neighbors, I...

Should I just use the REFIT dataset for the tutorial? If so, do you think it would suffice to just use the dataset locally and refer to the "download instructions"...

All right, thank you! How do you want to visualize the motif pair? Should we plot it inside the (currently very colorful) graph or should I make a zoom-in of...

Our time series has the length 37,440. But the authors claim that their length is 17,000. If I understand everything correctly, you were converting that to minutes. I'm trying to...

I don't think so since the similarity still holds in the other dimensions. I think that they will still be chosen and the additional one ignored.

Puhh, difficult to tell, but we could try. If it were a one-dimensional problem I would have tried it with annotation vectors. But I don't know if this is possible...

yeah I could try, but I think that may eliminate too much. I think its better to set only zero values to `np.nan`. But I'm also not sure if this...