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

Cannot do ImageStack.{transform,apply} after ImageStack.{sel,isel} #1268

Closed
ttung opened this issue Apr 25, 2019 · 4 comments
Closed

Cannot do ImageStack.{transform,apply} after ImageStack.{sel,isel} #1268

ttung opened this issue Apr 25, 2019 · 4 comments
Assignees
Labels
bug An issue with an existing feature high priority
Milestone

Comments

@ttung
Copy link
Collaborator

ttung commented Apr 25, 2019

The resulting ImageStack takes a reference to the backing MPDataArray. When we pass the MPDataArray to the child processes, we reshape the raw byte buffer to the shape of the resulting ImageStack. However, the MPDataArray is sized to reflect the original ImageStack's size, and so the reshape operation fails.

@ttung ttung self-assigned this Apr 25, 2019
@shanaxel42 shanaxel42 added bug An issue with an existing feature and removed spacejam labels May 7, 2019
@ambrosejcarr
Copy link
Member

ambrosejcarr commented May 9, 2019

attempts to do in-place operations using starfish.image.Filter also fail after .sel and .isel

@ttung
Copy link
Collaborator Author

ttung commented May 9, 2019

That does not surprise me.

@ambrosejcarr
Copy link
Member

ambrosejcarr commented Aug 27, 2019

@ttung you mentioned that if we swap starfish to multi-threading you think this issue goes away. What is your impression of the impacts on:

  1. Starfish's execution speed
  2. The complexity of the ImageStack object. Can we do away with the extra MPDataArray container and operate directly on the numpy array?

@ttung
Copy link
Collaborator Author

ttung commented Aug 27, 2019

  1. Starfish will likely be slower. Not clear how much slower it will be. Most numpy operations will not be affected. Non-numpy operations may be.
  2. Yes.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug An issue with an existing feature high priority
Projects
None yet
Development

No branches or pull requests

4 participants