Ameya Daigavane
Ameya Daigavane
This should be a lot clearer in the code now. To answer the above questions, suppose you have `n` samples `x1` from a distribution `p(x)`, and `n` samples `x2` from...
I haven't experimented with very large datasets. Let me know if you can get anything useful out of densratio_py. One point you should ensure is that you pick the same...
I believe that if you take a sufficient number of samples from both distributions - not necessarily the entire training/test set - you might still get a good result with...
I think I might be able to make some progress on this.
I did make progress on my fork. I also fixed a bug in SCRIMP++ where the matrix profile indices were taking invalid values. I also added comments in some places,...
There was another bug where the exclusion zones were not applied to the distance profiles. This has been fixed - SCRIMP++ now gives me the same output as STOMP.
I want to make a pull request with all of my contributions and bug fixes, but I think I've added quite a few things (exclusion_zone_fraction) and modified the names of...
It's not about the standards, it's about the implementation here: both STOMP and SCRIMP++ (when run to completion) compute the exact matrix profile. There is no reason for the outputs...
Correction on the exclusion zones - there shouldn't be any need for them in PreSCRIMP. There seems to be some issue with the noise correction I've implemented.
Sorry, I've done so now!