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

S-rep hypothesis testing work #291

Open
bpaniagua opened this issue Dec 14, 2022 · 6 comments
Open

S-rep hypothesis testing work #291

bpaniagua opened this issue Dec 14, 2022 · 6 comments
Assignees

Comments

@bpaniagua
Copy link
Contributor

  1. Generate a module that does batch fits of s-reps
  2. Read the documentation sent by Mohsen incl. paper describing the method and R-codebase
  3. Generate a module for hypothesis testing in S-reps
@Connor-Bowley
Copy link
Collaborator

The new pipeline creator is a good candidate for this. The (unmerged) new pipeline creator plus the work in this commit can make a module that does initial-fit + refinement. The batch processing part of the new pipeline infrastructure has not been started however.

@Connor-Bowley Connor-Bowley removed their assignment Mar 9, 2023
@vicory
Copy link
Member

vicory commented Jun 8, 2023

  1. Generate a module that does batch fits of s-reps

@allemangD @HarryDC

Going to use this issue to expand on this part. Batch fitting needs to run two modules:

  1. SRepCreator: Given an input surface and some parameters generates an initial s-rep
  2. SRepRefiner: Given an input surface and the initial s-rep generated by 1), improves the fit

The initial stitching of those two together should be in the commit that Connor linked above, the missing piece is just the batch processing.

@HarryDC
Copy link
Collaborator

HarryDC commented Jun 16, 2023

Yea I looked at the two pipelines, i think those are good to go, I have the creator in a state too where it can potentially store multiple output paramters/nodes. And we are ready to create a pipeline from the two parts.

I can put up a SlicerSalt branch that has the connors branch in SRep and my pipelines working branch tied together, i'll set up a combined pipeline so that all the parameters are exposed in the Creator/Refiner pipeline, you should then be able to create a .csv with all the input nodes and input parameters and run that through the Creator/Refiner

@HarryDC HarryDC self-assigned this Jun 21, 2023
@HarryDC
Copy link
Collaborator

HarryDC commented Jun 21, 2023

@vicory Where should the created combined module live ? Checked into SREP as a module or as something under SlicerSALT ?

@bpaniagua
Copy link
Contributor Author

Hi Harry, The new module should live under s-reps.

@HarryDC
Copy link
Collaborator

HarryDC commented Jun 21, 2023

Quick update, it took a bit of hand editing (the annotations that limit the inputs and confer the defaults aren't copied by the pipeline creator) but this is the Combined Module.

image

The list of parameters is long as I exposed all the parameters of both the generation and the refinement pipelines

I'll push updates Slicer/SRep/Pipenline branches so you can test this

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

No branches or pull requests

4 participants