Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fixed critical bugs for samples larger than n=300; Improved useability #8

Merged
merged 15 commits into from
Jan 20, 2015

Conversation

nicholst
Copy link
Owner

Changes to make software run with up to 2000 subjects.

@bathybicBee : Can you you download and do a final test? Once you say it's good, I'll make "release" and send it to Basel and advertise it on the webpage.

nicholst and others added 15 commits January 17, 2015 17:35
1) No longer allocated device memory for Phi (used for logistic regression) and for the residuals. This saves quite a bit of memory.
2) Increase the number of subjects allowed to 2000.  (I will change this so it is no longer constant later).
This has been tested with 956 subjects 15 covariates, requiring 1565 MB of device (GPU) memory (out of K20c's 4799 MB); 2000 subjects should be feasible.

Note max # covariates is 20.
Max # subjects now 2000—which should take up less than 3300MB of device Memory w/15 covariates.
nicholst added a commit that referenced this pull request Jan 20, 2015
Fixed critical bugs for samples larger than n=300; Improved useability
@nicholst nicholst merged commit 8703512 into master Jan 20, 2015
@nicholst nicholst deleted the BigN branch January 20, 2015 16:18
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant