search

nsh87 / receptormarker

Source for 'receptormarker' package for R: antibody receptor and phenotypic marker analysis
http://receptormarker.com
BSD 2-Clause "Simplified" License
4 stars 9 forks source link

Make NbClust faster #75

Open catterbu opened 8 years ago

catterbu commented 8 years ago

This PR goes through all of the code in the NbClust function and optimizes it for better speed and fixes all errors related to it.

catterbu commented 8 years ago

@nsh87 I went through all of the errors that I could find testing the code that I changed. I rebased in the changes from your dev branch. My hope is that there will not be any errors, but we will see.

nsh87 commented 8 years ago

Nice! I think it's a little tight to merge this in before the review in a week, but after would work.

On July 25, 2016 10:48:19 PM catterbu notifications@github.com wrote:

@nsh87 I went through all of the errors that I could find testing the code that I changed. I rebased in the changes from your dev branch. My hope is that there will not be any errors, but we will see.

You are receiving this because you were mentioned. Reply to this email directly or view it on GitHub: https://github.com/nsh87/receptormarker/pull/75#issuecomment-235169138

catterbu commented 8 years ago

@nsh87 well that makes me feel good! The only reason the first build failed was due to me using the .Internal function to speed up mean. My thought was that adding hclust to the multi_clust function should be my next feature. Is there something else that I am forgetting?

nsh87 commented 8 years ago

@catterbu what is .Internal?? I think you'd want to add a faceted PCA graph (one PCA plot for each # of clusters, k); this was one of the first comments we got about how to make the PCA plot more useful. We also need good documentation at the bottom of all the pages. I'd focus on those two first.

catterbu commented 8 years ago

.Internal makes the mean function a lot faster. I can do the faceted PCA for sure. When you write "good documentation at the bottom of the page," how would I do that? I am unclear on what that would physically require me to do.

On Aug 3, 2016, at 5:42 AM, Nikhil Haas notifications@github.com wrote:

@catterbu what is .Internal?? I think you'd want to add a faceted PCA graph (one PCA plot for each # of clusters, k); this was one of the first comments we got about how to make the PCA plot more useful. We also need good documentation at the bottom of all the pages. I'd focus on those two first.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or mute the thread.

nsh87 commented 8 years ago

Just start writing, we can put it in the right medium later. Start with Mark down in a text file.

Nikhil

On August 3, 2016 6:30:01 PM catterbu notifications@github.com wrote:

.Internal makes the mean function a lot faster. I can do the faceted PCA for sure. When you write "good documentation at the bottom of the page," how would I do that? I am unclear on what that would physically require me to do.

On Aug 3, 2016, at 5:42 AM, Nikhil Haas notifications@github.com wrote:

@catterbu what is .Internal?? I think you'd want to add a faceted PCA graph (one PCA plot for each # of clusters, k); this was one of the first comments we got about how to make the PCA plot more useful. We also need good documentation at the bottom of all the pages. I'd focus on those two first.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or mute the thread.

You are receiving this because you were mentioned. Reply to this email directly or view it on GitHub: https://github.com/nsh87/receptormarker/pull/75#issuecomment-237293967