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

Dev #22

Merged
merged 22 commits into from
Mar 30, 2021
Merged

Dev #22

merged 22 commits into from
Mar 30, 2021

Conversation

Hu-JIN
Copy link
Collaborator

@Hu-JIN Hu-JIN commented Mar 30, 2021

No description provided.

Hu-JIN and others added 22 commits March 8, 2021 12:29
Embed COSMIC catalogs inside the package.
add refitting with sparsity and reassign in Denovo class
Validation for denovo and refitting calculations
… DenovoSig

1. Parallelized wrappedMVNMF and make that parallelization compatible with the parallelization in DenovoSig. Performed tests showing expected gain in running time.
2. In DenovoSig, added attributes to save the raw W's and H's from each replicate run, also added attributes to save the selected lambda_tilde for mvNMF.
Some of the catalog data files are not in the correct order for the 96 features. Those are corrected.
First remove bad solutions. Then do clustering. This gives better results overall.
When the solutions from all replicates have the same reconstruction error, previous code will throw an error. Now that is taken care of by replacing < by <=.
1. We should perform filtering at the beginning. It is possible that after filtering, there is only one solution left. So we move filtering to before "if len(Ws) == 1".
2. We save the number of solutions after filtering to self.n_replicates_after_filtering_all. Comparing this number to self.n_replicates gives an idea of how many solutions are considered bad and thus filtered out. If too many solutions are filtered out, then it might be a problem.
@Hu-JIN Hu-JIN merged commit b5b2a2b into main Mar 30, 2021
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.

2 participants