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

Add "safety margin" argument for ICE #83

Closed
mimakaev opened this issue Sep 22, 2017 · 3 comments
Closed

Add "safety margin" argument for ICE #83

mimakaev opened this issue Sep 22, 2017 · 3 comments

Comments

@mimakaev
Copy link
Contributor

basically:
cooler.ICE(..., nnz_safety_margin = 0) # suggested value ~3-4
if 2 * cooler.info["nnz"] / cooler.info["nbins"] < nnz_safery_margin * min_nnz:
do_nothing_and_dont_do_ICE

if nnz_safery_margin = 1, and cooler.ICE is performed, then exactly at the margin about 1/2 of the bins will be filtered out, which is bad. If nnz_safety_margin=3, then maybe only 5-10% of the bins will be filtered out, and if the dataset is more sparce, ICE will be aborted.

@gfudenberg
Copy link
Member

relatedly, should the actual value that the mad_max filter thresholds at be reported in the balance.py stats? e.g. 'mad_max_cutoff':cutoff
this could be useful for debugging balancing.

@gfudenberg
Copy link
Member

bumping this-- for balancing, it would be useful to add an easy way to see what thresholds in terms of counts a given mad-max value would return for a given cooler. it would also be useful to save the #counts as metadata.

@nvictus nvictus mentioned this issue Jun 30, 2020
5 tasks
@nvictus
Copy link
Member

nvictus commented Jun 30, 2020

See #210

@nvictus nvictus closed this as completed Jun 30, 2020
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

3 participants