Nima Sarajpoor

@alvii147 @seanlaw Okay...I got something... And, I hope you can work on it to see if we can make it better / cleaner:  And,  --- ``` # stump_uint_v3...

@alvii147 > @NimaSarajpoor wow that looks promising! I didn't think using np.arange would be slower than using range and then converting to np.uint64. Yeah...I had no idea as well... I...

@alvii147 I tried to make the changes again and I got the following results on my pc: ``` # n: length of time series # m: window size ``` |...

@seanlaw > However, can we simply see/understand how much of this improvement is coming from the use of `range` by itself? I think what @alvii147 is trying to do here...

> this is probably the document that got me to spend more time with the matrix profile > https://www.cs.ucr.edu/~eamonn/100_Time_Series_Data_Mining_Questions__with_Answers.pdf > selling me solutions, not a new shiny methodology Related issue:...

Cool. This swiss army knife would be a great addition. I was wondering if it would be okay to provide my feedbacks on this PR. I think this can help...

> Do we need to update `aamp_snippets.py` as well? I think we are good in the non-normalized version.

> > I think we are good in the non-normalized version. > > Why wouldn't we need to also do: > > ``` > M_T[start : start + m], >...

@seanlaw btw, for `k=2`, we got error (PROBABLY due to imprecision). I changed it to `k=1` as my goal was to fix this issue for now. I think you may...

I think we also need to revise this line of code in both normalized and non-normalized version: ``` # last for-loop in function snippets / aampdist_snippets total_min = total_min -...