-
assigned issue to
findThreshold is slow
A user reports that findThreshold is very slow (hours). Can we improve this?
Comments (6)
-
-
We are working on a new methodology to resolve this issue. Very close to the end, but needs more investigations...
-
@namita1025, can you update your threshold method with a parameter on # sequences to downsample the data. The default value should be something reasonable so that the method runs in reasonable time.
-
This parameter exists already but I didn't want to make a default value to subsample in case people didn't notice the parameter.
-
Documentation notes that subsampling to 15000 is robust, but that still takes time. I haven't tested subsampling further down.
-
reporter - changed status to resolved
Doesn't seem to be a fix for the density fit method, but the new method has been integrated and it's faster and set default.
- Log in to comment